Combination of triple bond and adamantane ring on the vitamin D side chain produced partial agonists for vitamin D receptor

Article abstract of DOI:10.1021/jm401989c

Vitamin D receptor (VDR) ligands are therapeutic agents that are used for the treatment of psoriasis, osteoporosis, and secondary hyperparathyroidism and have immense potential as therapeutic agents for autoimmune diseases, cancers, and cardiovascular diseases. However, the major side effect of VDR ligands, the development of hypercalcemia, limits their expanded use. To develop tissue-selective VDR modulators, we have designed vitamin D analogues with an adamantane ring at the side chain terminal, which would interfere with helix 12, the activation function 2, and modulate the VDR potency. Here we report 25- or 26-adamantyl-23,23,24,24-tetradehydro-19-norvitamin D derivatives (ADTK1-4, 4b,a and 5a,b). These compounds showed high VDR affinities (90% at maximum), partial agonistic activities (EC₅₀ 10^-9-10^-8 M with 40-80% efficacy) in transactivation, and tissue-selective activity in target gene expressions. We investigate the structure-activity relationships of these compounds on the basis of their X-ray crystal structures.

Full text of DOI:10.1021/jm401989c

Article
pubs.acs.org/jmc
Combination of Triple Bond and Adamantane Ring on the Vitamin D
Side Chain Produced Partial Agonists for Vitamin D Receptor
Takeru Kudo,† Michiyasu Ishizawa,§,⊥ Kazuki Maekawa,†,⊥ Makoto Nakabayashi,∥,§ Yusuke Watarai,†
Hikaru Uchida,§ Hiroaki Tokiwa,‡,† Teikichi Ikura,# Nobutoshi Ito,# Makoto Makishima,*,§
and Sachiko Yamada*,§
‡
†Department of Chemistry, Research Center for Smart Molecules, Faculty of Science, Rikkyo University, Toshima-ku, Tokyo
171-8501, Japan
§Department of Biomedical Sciences, Nihon University School of Medicine, Itabashi-ku, Tokyo 173-8610, Japan
∥Graduate School of Biomedical Science and #Medical Research Institute, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo
113-8510, Japan
S
* Supporting Information
ABSTRACT: Vitamin D receptor (VDR) ligands are therapeutic
agents that are used for the treatment of psoriasis, osteoporosis,
and secondary hyperparathyroidism and have immense potential as
therapeutic agents for autoimmune diseases, cancers, and
cardiovascular diseases. However, the major side effect of VDR
ligands, the development of hypercalcemia, limits their expanded
use. To develop tissue-selective VDR modulators, we have
designed vitamin D analogues with an adamantane ring at the
side chain terminal, which would interfere with helix 12, the
activation function 2, and modulate the VDR potency. Here we
report 25- or 26-adamantyl-23,23,24,24-tetradehydro-19-norvita-
min D derivatives (ADTK1−4, 4b,a and 5a,b). These compounds
showed high VDR affinities (90% at maximum), partial agonistic
activities (EC50 10−9−10−8 M with 40−80% efficacy) in trans-
activation, and tissue-selective activity in target gene expressions. We investigate the structure−activity relationships of these
compounds on the basis of their X-ray crystal structures.
including bisphosphonates, estrogen, and selective estrogen
receptor modulators (SERMs), prevent further bone loss, they
do not build bone once it has been lost. The US Food and Drug
Administration (FDA) has approved a recombinant human
parathyroid hormone, also known as teriparatide, as an anabolic
bone-building agent for the treatment of osteoporosis.5 The FDA
also recently approved the anti-RANKL antibody (denosumab)6
for the treatment of osteoporosis.
INTRODUCTION
■
The fundamental actions of the steroid hormone, 1α,25-
dihydroxyvitamin D3 [1,25(OH)2D3, 1], are to maintain calcium
and phosphorus homeostasis in vertebrate organisms. This
activity is initiated by direct binding of the hormone to the
vitamin D receptor (VDR), a member of the nuclear receptor
superfamily, in the intestine, kidney, and bone. In the intestine
and kidney, transepithelial transport of calcium is known to
involve the apical calcium ion channels TRPV5 and TRPV6,1 In
contrast, activity in the skeleton is driven primarily by RANKL, a
TNF-like factor produced by stromal cells and osteoblasts, which
are both necessary and sufficient for the formation, activation,
and survival of bone-resorbing osteoclasts.2 Perhaps most
importantly, the primary regulator of TRPV5, TRPV6, and
RANKL expression is 1,25(OH)2D3.
Bone degenerative disease such as osteoporosis occurs in a
substantial proportion of the elderly population.3 Osteoporosis
encompasses a heterogeneous group of disorders that represents
a major risk for bone fractures and a substantial burden on the
health care system. More than 15 billion dollars are spent
annually in the United State on medical care for the treatment of
osteoporosis.4 Although a number of antiresorpative agents,
Active vitamin D derivatives have bone anabolic activity7 and
are naturally derived agents for the treatment of osteoporosis.
However, the use of active vitamin D derivatives for the
treatment of osteoporosis is difficult because of concerns
regarding hypercalcemia and hypercalciuria. Active vitamin D
analogues have therefore not yet been approved as osteoporosis
agents in the United States and European countries. However,
active vitamin D analogues (alfacalcitol and eldecalcitol) have
been successfully used in the treatment of osteoporosis in Japan.8
Selective VDR modulators can open up possibilities for VDR
ligands. 2-Methylene-20-epi-19-norvitamin D (2MD) was first
Received: December 27, 2013
Published: April 28, 2014
© 2014 American Chemical Society
4073
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
Chart 1. Structures of Compounds Discussed in This Article
reported to be a bone-selective anabolic ligand in rats,9 however,
it was recently shown to increase bone turnover, but not mineral
density, in women with osteopenia.10 Nonsecosteroidal, non-
calcemic, and tissue selective ligands have been reported but still
not been proved to be potential therapeutic agents.11
tetradehydro-19-norvitamin D derivatives (4a,b and 5a,b)
(Chart 1). These vitamin D derivatives have significant VDR
affinities, partial agonistic activities, and selectivities in the
expression of genes in various cell types. The X-ray crystal
structures of rVDR-ligand binding domain (LBD) complexed
with 4b, 5a, and 5b revealed in part their selective activities.
Vitamin D analogues for use as therapeutic agents should not
necessarily be super agonist but should have selective activity. We
thought that vitamin D compounds that can change the
conformation of helix (H) 12 could have antagonist/partial
agonist characteristics and may have selective activities. This idea
is similar to that for other nuclear receptors such as SERMs12 and
selective progesterone receptor modulators.13 The side chain
terminal 26-methyl groups of 1,25(OH)2D3 interacts with the
residues Phe422 and Val418 on H12, and these interactions are
thought to be important for its agonistic activity.14 We have
synthesized compounds with a double bond and an adamantane
ring on the side chain of vitamin D (2 and 3) (Chart 1).14 The
terminal adamantane ring was expected to clash with the residues
on H12, changing the H12 conformation, and the double bond at
position 22 was expected to increase the side chain rigidity. The
2-methylene-19-nor A-ring system was selected because it is
much more stable to acids, oxidation, irradiation, and heat than
the natural triene system of vitamin D is, and it can be
synthesized much more readily than normal vitamin D
compounds. The 2-methylene-19-nor A-ring system was
developed by DeLuca’s group and is found in super agonistic
compounds such as 2MD.9 Our compounds (2 and 3) had
significant VDR affinities (2−100% that of 1) and selective VDR
modulator activities.14 However, their efficacies of transcriptional
activities were low (<15%): i.e., these compounds 2 and 3 act as
antagonists. The need for analogues with higher transactivation
efficacies prompted us to synthesize further analogues with more
rigidity, i.e., 25- and 26-adamantyl-2-methylene-23,23,24,24-
RESULTS
■
Synthesis of 25- and 26-Adamantyl-23-yne-19-norvi-
tamin D Compounds ADTK1−4 (4b,a and 5a,b). We
synthesized four new 2-methylene-19-norvitamin D derivatives
(4a,b and 5a,b) starting from 22-tosylate 6, which was
synthesized14 from D-(−)-quinic acid as an A-ring precursor
and vitamin D2 as a CD-ring plus side chain precursor (Scheme
1). The 22-tosylate 6 was treated with TMS-acetylene (MeLi in
dioxane, 105 °C, 80%) and then with K2CO3 (THF/MeOH,
95%) to remove the C-TMS group to give acetylene compound
7b. To compound 7b was added nBuLi/THF at 0 °C, and after
several minutes the solution was treated with 1-adamantylfor-
maldehyde (n = 0) or 1-adamantylacetaldehyde (n = 1), giving
the adamantyl alcohols 8 and 9 in 93% and 80.5%, respectively, as
a 1:1 mixture of epimers at C(25). The diastereomeric mixture 8
was separated by HPLC to give less polar 8a and more polar 8b.
The diastereomeric mixture 9, which could not be separated by
common HPLC columns, was converted to (R)- and (S)-α-
methoxy-α-(trifluoromethyl)-phenyl acetic acid esters [(R)- and
(S)-MTPA esters, 11a,b and 11c,d, respectively] by treatment
with (S)- and (R)-MTPA-Cl (Et3N, dimethylaminopyridine
DMP, CH2Cl2, 81% and 41%), respectively, which were readily
separated by HPLC. Deprotection (camphor sulfonic acid CSA,
MeOH, room temperature) of 8a and 8b yielded the target
compounds 4a (ADTK2) (94%) and 4b (ADTK1) (91%),
respectively. Similarly, deprotection of 11a and 11d ((1) CSA,
4074
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
Scheme 1. Resolution and Stereoselective Synthesis of 25- and 26-Adamantyl-2-methylene-22,22,23,23-tetradehydro-19-
norvitamin D Derivatives (4a, 4b, 5a, and 5b)
MeOH; (2) K2CO3, MeOH) yielded compound 5a (ADTK3)
(56%) and 11b and 11c gave 5b (ADTK4, 67%).
R- and S-catalysts therefore showed the same stereoselectivities
as previously reported.15 Catalysts with the R-configuration gave
the 25R-isomer 8a, and with S-configuration yielded 25S-isomer
8b with high selectivities. Selective reduction of 12b with (R)-
and (S)-CBS catalysts yielded 9a and 9b, respectively, with high
selectivity and in good yields (>95% de 62% yield and >95% de
62% yield, respectively).
Stereoselective Synthesis of ADTK1−4 (4b,a and 5a,b).
The stereoselective syntheses of 4a,b and 5a,b were achieved by
reduction of the 25-keto compounds 12a and 12b with a chiral
oxazaborolidine catalyst.15 The 25-hydroxyl compounds 8 and 9
were oxidized (Dess−Martin periodinane, DMP) to ketones 12a
(84%) and 12b (68%), respectively, and selective reduction of
the ketones was examined. Among several asymmetric catalysts,
B-methyl-4,5,5-triphenyl-1,3,2-oxazaborolidine (BMTO)15d and
Corey−Bakshi−Shibata catalyst (CBS),15a−c worked excellently.
Reduction of 12a with the (R)-BMTO catalyst in the presence of
BH3−SMe2 (THF, 0 °C) gave the 25R-epimer 8a in 70% yield
with 78% de. Reduction of 12a with the (R)-CBS catalyst (BH3−
SMe2, THF, 0 °C) gave 8a more selectively (91% de) in 86%
yield. Reduction of 12a with the (S)-CBS catalyst proceeded
similarly to yield the 25S-epimer 8b (87% de) in 75% yield. The
Determination of C(25) Stereochemistry of ADTK1−4
(4b,a and 5a,b) by 1H NMR Spectroscopy. We determined
the stereochemistry at C(25) of 4a, 4b, 5a, and 5b by using the
new Mosher method.16 Epimer 8a of the 25-hydroxy compound
was treated with (S)- and (R)-MTPA-chloride to give to (R)- and
(S)-MTPA esters 10a and 10b, respectively. We also obtained
pairs of (R)- and (S)-MTPA esters of the 26-adamantyl
compound (11a,b and 11c,d, respectively), as described above.
We used one- and two-dimensional 1H NMR spectra to identify
the signals from protons in the side chain. The Δδ values (δS −
4075
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
δR) of the protons on the side chains of three pairs of MTPA
esters are shown in Chart 2C,D,E. The Δδ values of H-20 and
Chart 2. Determination of Stereochemistry at C(25) of
Adamanty-19-norvitamin D Derivatives (4a,b and 5a,b) Using
a
the New Mosher Method
Figure 1. Competitive binding assays of adamantyl-19-norvitamin D
derivatives 4a, 4b, 5a, and 5b compared with 1,25(OH)2D3 (1). hVDR-
LBD expressed as glutathione S-transferase fusion protein was incubated
with [3H]-1,25(OH)2D3 in the presence of nonradioactive 1,25-
(OH)2D3 (1) (red tilted square), 4a (blue square), 4b (cyan triangle),
5a (green square), and 5b (light green triangle) at a range of
concentrations. All values represent means six standard deviations of
triplicate assays.
that of 1,25(OH)2D3. The IC50s of 25R-adamantyl (4a), 25R-
(5a), and 25S-adamantyl (5b) longer side-chain compounds
were 12, 13, and 12 nM, respectively.
Transcriptional Activity. VDR transactivation by ADTK1−4
(4b,a and 5a,b) was evaluated using a luciferase reporter assay in
human kidney cell lines (HEK293) transfected with mouse
osteopontin vitamin D response elements (VDRE, SPP × 3-tk-
LUC) and hVDR (pCMX hVDR) (Figure 2A).14d Compound
4b had the highest activity (EC50 0.07 nM, efficacy 81% that of
the natural hormone) of the four compounds. The other
compounds 4a, 5a, and 5b had similar activities: EC50 (efficacy)
4a 6.6 nM (38%), 5a 1.5 nM (70%), and 5b 1.1 nM (63%). It is
worth noting that all the synthetic analogues have partial agonist
activities in the transcriptional assay in HEK293 cells (Figure
2A). Their activation efficacies do not exceed that of 1,25-
(OH)2D3 (1), even at doses 100 times higher. These compounds
4a, 5a, and 5b, except for 4b, inhibited weakly the transactivation
induced by the natural hormone 1 (10 nM) (Figure 2B), while
antagonist 3a inhibited dose dependently the action of
1,25(OH)2D3.
Mammalian Two-Hybrid Assays. The effects of the
adamantyl vitamin D compounds on the binding of VDR to
various cofactors were evaluated by mammalian two-hybrid
assays in HEK293 cells.18 In these experiments, the effects of the
ligands on VDR-retinoid X receptor α (RXRα) heterodimeriza-
tion,19,20 as well as VDR-coregulator binding including the
steroid receptor coactivator 1 (SRC-1; or nuclear receptor
coactivator 1 NCoA1),21 nuclear receptor corepressor 1 (N-
CoR),22 and silencing mediator of retinoic acid and thyroid
hormone receptor (SMRT or nuclear receptor corepressor 2, N-
CoR2),23 were evaluated by a luciferase reporter assay.18 The
results are shown in Figure 3.
a
(A,B) Determination of the C(25) stereochemistries of adamantyl
Δ22-vitamin D compounds 3b and 3c as reported.14a Determination of
the C(25) stereochemistries of 11a and 11d derived from ADTK3
(5a) (C), 11b and 11c derived from ADTK4 (5b) (D), and 10a and
10b derived from ADTK2 (4a) (E).
−21 (both −0.06) of 10a and 10b were correlated to the 25R-
configuration, although the H-22 signals cannot be well assigned.
Because the C(25) stereochemistry of 4b (ADTK1), which was
obtained by deprotection of 8b, was confirmed to be S by X-ray
crystallographic analysis of the rVDR complex (described
below), the stereochemistries of 4a and 4b were confirmed to
be R and S, respectively.
As shown above, we confirmed that 11a (R-MTPA-ester) and
11d (S-MTPA-ester) were converted to 5a, and 11b (R-MTPA-
ester) and 11c (S-MTPA-ester) to 5b. The Δδ values δ(11d) −
δ(11a) were −0.02, −0.02 (H22), −0.03 (H21), and +0.03
(H26), and accorded with the 25R stereochemistry (Chart 2C).
The Δδ values δ(11c) − δ (11b) were +0.03 (H22), 0.0 (H21),
and −0.06 (H26) and accorded with the 25S stereochemistry
(Chart 2D). The stereochemistries of 5a and 5b were
determined to be R and S, respectively, by X-ray crystal structural
analysis of their rVDR complexes, as described below. All the
stereochemistries at C(25) of the synthetic compounds 4a,b and
5a,b were therefore successfully determined using the new
Mosher method. Chart 2 shows the present results compared
with the corresponding results for our double-bond compounds
(3b and 3c).14a
Biological Activities of Adamantyl Vitamin D Com-
pounds ADTK1−4 (4b,a and 5a,b). VDR Affinity. The VDR
affinities of these vitamin D derivatives (4a, 4b, 5a, and 5b) were
determined based on competitive binding between [3H]-1,25-
(OH)2D3 and the substrate using recombinant hVDR-LBD.17
The results are shown in Figure 1. The 25S-adamantyl
compound 4b had the highest activity, IC50 0.5 nM, about 90%
Binding of VDR to RXRα. Ligand binding promotes
heterodimerization of VDR with RXRα, which is essential for
the VDR to recognize VDREs in the promoters of the target
genes.19,20 Adamantyl vitamin D 4b activates (EC50 0.6 nM) the
VDR to bind to RXRα similarly to 1,25(OH)2D3 (EC50 0.7 nM)
(Figure 3A). The other compounds 4a (EC50 4.4 nM), 5a (EC50
2.3 nM), and 5b (EC50 0.8 nM) were a little less active than 4b
4076
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
Figure 2. Transactivation of hVDR by adamantyl-19-norvitamin D
derivatives 4a, 4b, 5a, and 5b compared with 1,25(OH)2D3 (1). (A)
HEK293 cells were cotransfected with TK-Spp × 3-LUC reporter
plasmid and pCMX-VDR, and 8 h after transfection the cells were
treated with several concentrations of 4a, 4b, 5a, 5b, and 1,25(OH)2D3
(1). After 24 h, the cells were harvested for assaying luciferase and β-
galactosidase activity using a luminometer (Molecular Devices,
Sunnyvale, CA). (B) Antagonistic effect of vitamin D derivatives, 4a,
4b, 5a, 5b, and ADTT 3a, on hVDR activated by 1,25(OH)2D3 (1).
HEK293 cells were cotransfected as in A and were treated with vitamin
D derivatives (4a, 4b, 5a, 5b, and 3a) at a range of concentrations in the
presence of 1,25(OH)2D3 (1) (10 nM).
(Figure 3A). These results show that all analogues induce VDR-
RXR binding.
Binding of VDR to SRC-1. SRC-121 directly binds to the
activation function 2 (AF2) surface of nuclear receptors and
stimulates the transcriptional activities in a hormone-dependent
manner. This coactivator plays a central role in creating
multisubunit coactivator complexes that act by participating in
both chromatin remodeling and recruitment of general tran-
scription factors.21 The (25S)-adamantyl compound 4b activated
the binding of VDR to SRC-1 more strongly (EC50 1.5 nM) than
the natural hormone 1 did (EC50 1.9 nM) (Figure 3B). The other
compounds 4a (EC50 15.0 nM), 5a (EC50 6.5 nM), and 5b (EC50
13.5 nM) activated the VDR less potently. These results indicate
that SRC-1 binds more strongly to the AF-2 surface of the VDR
bound 4b than that bound the natural hormone 1.
Figure 3. Effect of vitamin D derivatives 4a, 4b, 5a, and 5b on
coactivator recruitment in a mammalian two-hybrid assay. (A) Effects of
vitamin D derivatives on interaction between VDR and RXRα. HEK293
cells were cotransfected with CMX-GAL4-RXRα, CMX-VP16-VDR,
and MH100(UAS) × 4-tk-LUC and treated with vitamin D compounds
4a, 4b, 5a, 5b, or 1,25(OH)2D3 (1) at a range of concentrations. (B)
Effects on interactions between VDR and SRC-1. HEK293 cells were
cotransfected with CMX-GAL4-SRC-1, CMX-VP16-VDR, and
MH100(UAS) × 4-tk-LUC and were treated with vitamin D
compounds 4a, 4b, 5a, 5b, or 1,25(OH)2D3 (1) at a range of
concentrations. Effects on interactions of VDR with NCoR-1 (C) or
SMRT (D). HEK293 cells were cotransfected with CMX-GAL4-N-CoR
or CMX-GAL4-SMRT in combination with CMX-VP16-VDR and
MH100(UAS) × 4-tk-LUC and were treated with vitamin D derivatives
4a, 4b, 5a, 5b, or 1,25(OH)2D3 (1) at a range of concentrations.
Binding of VDR to N-CoR. N-CoR22 mediates transcriptional
repression and, as part of a complex, promotes histone
deacetylation and the formation of repressive chromatin
structures which may impede the access of basal transcription
factors. Thus, 4b dose-dependently inhibits binding of the VDR
to N-CoR (EC50 0.4 nM) slightly more potently than 1,25(OH)
2D3 does (EC50 0.5 nM) (Figure 3C). The other analogues, 4a
(EC50 4.0 nM), 5a (EC50 3.7 nM), and 5b (EC50 1.8 nM),
inhibited binding of the VDR to N-CoR less potently than 4b.
The ability of adamantyl compounds to inhibit the binding of the
two proteins is inversely proportional to the ability to activate
binding of the VDR to SRC-1.
Binding of VDR to SMRT (NCoR2). The effect of 4b (EC50 0.5
nM) in inhibiting binding of the VDR to SMRT23 was slightly
less potent than that of 1,25(OH)2D3 (EC50 0.4 nM). Other
analogues inhibited the binding of SMRT to the VDR less
4077
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
Figure 4. Effects of vitamin D derivatives 4a, 4b, 5a, and 5b on expression of CYP24A1 gene in kidney derived HEK293 (A), intestinal SW480 (B),
keratinocyte HaCaT cells (C), monocyte-derived U937, (D), monocyte-derived THP-1 (E), and bone-derived MG63 (F) cells. Effects on the
expression of CAMP gene in HaCaT (G), U937 (H), and THP-1 cells (I), and of TRPV6 gene in SW480 cells (J). Cells were treated with each sample
(100 nM) for 24 h. *, p < 0.05; **, p < 0.01; ***, p < 0.001.
potently than 4b: 4a (EC50 4.1 nM), 5a (EC50 2.4 nM), and 5b
(EC50 2.8 nM) (Figure 3D).
CYP24A1. CYP24A1 is an enzyme24 that inactivates 1,25-
(OH)2D3 and its precursor 25-hydroxyvitamin D3 (25-OHD3)
by hydroxylating their 24-position to yield the corresponding 24-
hydroxylated metabolites. Binding of the ligands to the VDR
induces enzyme expression in all the target tissues. The 25S-
adamantyl compound 4b induced CYP24A1 mRNA expression
in HEK293 cells similarly to the natural hormone (Figure 4A).
The other compounds, 4a, 5a, and 5b, increased Cyp24A1
mRNA expression by 15%, 45%, and 30% of the activity of the
natural hormone.
Interestingly, cell-type-selective gene induction was observed.
In intestinal SW480 and in bone MG63 cells, 4b activated the
CYP24A1 gene with about 50% of the activity of the natural
hormone (Figure 4B,F). In skin HaCaT cell, mRNA expression
of CYP24A1 was 30% (Figure 3C) and in blood U937 and THP-
Effect of Adamantyl Vitamin D Analogues on
Endogenous Gene Expression in Various Cells. To examine
the tissue-selective action of analogues 4a,b and 5a,b, we
evaluated the expression of CYP24A124 gene in kidney-
epithelium-derived HEK293, intestinal-mucosa-derived SW480,
osteoblast-derived MG63, myeloid-derived THP-1 and U937,
and skin-keratinocyte-derived HaCaT cells. We also examined
the effect of the analogues on the expression of other genes, such
as cathelicidin antimicrobial peptides (CAMP)25 in THP-1,
U937, and HaCaT cells, and transient receptor potential
vanilloid 6 (TRPV6)26 in SW480 cells. In all the experiments,
we used equal concentrations (100 nM) of the ligands (1, 3a,b,
and 4a,b).
4078
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
Table 1. Data Collection and Refinement Statistics
ligand
ADTK1 (4b)
3VTB
ADTK3 (5a)
3VTC
ADTK4 (5b)
3VTD
PDB ID
X-ray source
space group
cell dimensions
KEK-PF BL-5A
KEK-PF BL-5A
KEK-PFAR NW12A
C2
C2
C2
a, b, c (Å)
α, β, γ (deg)
153.56, 43.20, 42.49
90.00, 95.56, 90.00
50.00−2.00
(2.07−2.00)
66943
126.85, 45.73, 46.64
90.00, 93.87, 90.00
50.00−1.50
(1.55−1.50)
147172
153.18, 43.90, 42.53
90.00, 95.73, 90.00
50.00−2.70
(2.80−2.70)
29041
resolution range (Å)
(outer shell)
no. of reflections
unique reflections
completeness (%)
(outer shell)
19 104
43 106
7900
97.3
96.8
98.6
(99.9)
(83.3)
(99.0)
Rmerge
0.042
0.041
0.061
(outer shell)
(0.249)
(0.182)
(0.311)
Refinement Statistics
resolution range (Å)
(outer shell)
50.00−2.00
(2.07−2.00)
0.286/0.246
(0.506/0.521)
50.00−1.50
(1.55−1.50)
0.209/0.184
(0.237/0.206)
50.00−2.70
(2.80−2.70)
0.278/0.213
(0.400/0.325)
R factor Rfree/Rwork
(outer shell)
1 cells, 70−90% that of the natural hormone (Figures 3D,E). The
25R-isomer 4a showed a low but clearer cell type dependent
activity differences: in U937 and THP-1 cells 1%, MG63 5%,
HEK293 14%, HaCaT cells 22%, and SW480 32%. The longer
side chain compounds 5a and 5b showed about 50% activity in
bone (MG63) and monocyte (U937) cells but showed lower
activities (10−30%) in monocyte (THP1), intestine (SW480),
and skin (HaCaT) cells (Figure 4). These adamantyl compounds
(4a,b and 5a,b) therefore all have different tissue selectivities.
Cathelicidine Antimicrobial Peptide (CAMP). CAMP25 is an
innate antimicrobial peptide and its induction has been observed,
for example, in myeloid cells, keratinocyte, and intestinal cell
lines. The expressions of the CAMP gene by 4b in the monocytic
cell lines U937 and THP-1 and in keratinocyte HaCaT cells were
increased compared with those of CYP24A1 (85%, 70%, and
30%, respectively) to 125%, 105%, and 60%, respectively (Figure
4H,I,G). The 25-epimer 4a showed lower activities in U937,
THP-1, and HaCaT cells, 30%, 20%, and 15%, respectively. The
longer homologues 5a and 5b had moderate activities in U937
cells, 75% and 65%, respectively, and in THP-1 cells 74% and
55%, respectively (Figure 4H,I). However, they showed much
lower activities in HaCaT cells, 20% and 15%. Thus, 5a and 5b
showed 3.5−4-fold higher activities in monocytic cells than in
skin cells (Figure 4G).
resolutions of 2.0, 1.5, and 2.7 Å, respectively (Table 1). The
complex with 4a (ADTK2) was also crystallized, but it was
unstable to achieve good resolution (Supporting Information
Figure S1, Table S1, and crystal data of rVDR-LBD/4a/DRIP).
All of the complexes belonged to the space group C2. The root-
mean-square deviations (rmsd) of the Cα atoms of the rVDR-
LBDs complexed with 4b, 5a, and 5b compared with the
1,25(OH)2D3 complex (2ZLC)28a were 0.32, 0.56, and 0.38 Å,
respectively. The complex with 5a had the largest rmsd and was
found to have somewhat a different unit cell structure
(Supporting Information Figure S2) from the others, such as
complexes with 4b and 1 (2ZLC). Why does the complex with 5a
have a different unit cell structure? First, this might be a result of
the positioning of the 25-hydroxyl group of 5a. As shown in the
overlay of the rVDR LBDs complexed with 5a and 1 (2ZLC)
(Figure 5A), the 25-hydroxyl group of 5a takes a different
position, which overlaps with the His393 of the complexes with
1. Then the torsion angle (N−Cα−Cβ−Cγ) of His393 was
changed significantly, from 170.7° of the complex with 1 to
−72.1° of the complex with 5a, which then caused the Cα
positional changes of Thr302 (1.4 Å), Leu303 (2.5 Å), Glu304
(3.1 Å), and Leu305 (2.4 Å), as shown in Figure 5B.
Interestingly, in 5b complex, the positional shifts of Thr302
(1.1), Leu303 (0.7 Å), Glu304 (0.9 Å), and Leu305 (1.0 Å) are
not as large as those of 5a complex, and Leu303 and Glu304 are
situated similarly to those of the complex with 1 (Figure 5C). In
the VDR/4b complex, Leu303 and Glu304 interact with Ile134
(3.7 Å) and His130 (2.7 Å), respectively, of another VDR in the
neighboring crystal unit (Supporting Information Figure S3A);
this would stabilize the crystal packing. However, in the complex
with 5a, similar intercrystal-unit interactions are impossible
because of the significant positional shifts of these residues
(Supporting Information Figure S3B). In the complex with 5b,
the two residues, Leu303 and Glu304, are positioned similarly to
4b and 1 rather than 5a (Supporting Information Figure S3B).
The rVDR complexes of the ligands 4b, 5a, and 5b all adopted
the canonical active conformation (Supporting Information
Figure S4). The side chain of the all ligands in the VDR adopted
similar 16,17,20,22 dihedral angles directed toward the 21-
methyl group of the natural hormone (Figure 5A,C): the
Transient Receptor Potential Vanilloid 6 (TRPV6). TRPV626
is a membrane calcium channel that is responsible for the first
step in calcium absorption in the intestine. Its expression by 4b in
intestinal SW480 cells (Figure 4J) was similar (65%) to the
expression of the CYP24A1 gene. The longer homologues 5a and
5b induced TRPV6 expression less strongly (40% and 35%,
respectively) but twice as strongly as the CYP24A1 gene. The
activities of 4a were similar (30%) for TRPV6 and CYP24A1.
X-ray Crystal Structural Analysis of rVDR-LBD Com-
plexed with ADTK1, ADTK3, and ADTK4 (4b, 5a, and 5b).
X-ray crystallographic structural analyses are essential for
investigating the structure−activity relationships of the target
compounds (4a,b and 5a,b). The rVDR-LBD bound to 4b, 5a,
and 5b were crystallized as ternary complexes with a peptide
containing LXXLL motif derived from VDR-interacting protein
complex component DRIP205 (or MED1)27 and analyzed to
4079
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
Figure 5. X-ray crystal structures of rVDR-LBD complexed with vitamin D compounds 4b, 5a, and 5b. (A) Overlay of rVDR-LBD complexes with 5a
(magenta) and 1 (atom type). Residues His393 and His301-Leu305 are shown with narrow sticks. This view is shown focusing on the ligand structures.
(B) Different view of the structures shown in (A) focusing on the residues His393 and His301-Leu305. (C) Overlay of rVDR-LBD complexes with 5a
(magenta), 5b (yellow), and 1 (atom type). Residues His393 and His301-Leu305 are shown with the same color as the ligand. (D) van der Waals
interactions (magenta line) of 4b (atom type) with VDR residues (atom type) within 4 Å distance from 4b. (E) Overlay of rVDR-LBD complexes with
4b (cyan) and 1 (atom type). Hydrogen bonding interactions (red dotted lines) and van der Waals interactions with H12 residues (magenta) are shown.
(F) Overlay of rVDR-LBD complexes with 5a (magenta) and 1 (atom type). Hydrogen bonding interactions (red dotted line) and van der Waals
interactions (magenta line) with H12 residues are shown. (G) Overlay of rVDR-LBD complexes with 5b (yellow) and 1 (atom type). Hydrogen
bonding interactions and van der Waals interactions with H12 residues are shown.
16,17,20,22-torsion angles of 4b, 5a, and 5b were 54.5, 53.0, and
48.6°, respectively, whereas that of 1,25(OH)2D3 was −44.3°
and 16,17,20,21 dihedral angle was 77.2°. In these conforma-
tions, the terminal adamantane rings of 4b, 5a, and 5b are placed
around the 26,27-dimethyl group of 1, and the other parts of the
side chains occupy different positions from those in the natural
hormone (Figure 5A,B,C).
The adamantane ring of 4b interacts with as many as 10
residues within 4 Å, i.e., Leu223, Leu226, Ala227, and Val230
(H3), His301 (loop 6−7), Tyr397 and Leu400 (H11), Leu410
4080
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
(loop11−12), and Val414 and Phe418 (H12) (Figure 5D). Most
importantly, Phe418 and Vl414 of 4b have closer contacts with
the adamantane ring (both 3.8 Å) than the 26-methyl group of
1,25(OH)2D3 (1) in all known VDR-LBD complexes: rat (both
4.3 Å), human (4.0 and 4.3 Å, respectively), and zebra fish (4.4
and 4.1 Å, respectively). Ligand 4b in the VDR forms hydrogen
bonds (2.7−2.9 Å) with the same six residues as the natural
hormone (Figure 5E), but His301 and 393 adopt different
conformations compared with the 1,25(OH)2D3 complex
(Figure 5E), although the Cα positional shifts are not large
(0.6 and 0.4 Å, respectively). The methylene group at C(2) of 4b
interacts with Tyr143 (3.8 Å), Arg270 (3.5 Å), and Phe150 (4.0
Å) as does 2MD (Figure 5D). Phe418 in the complex of 4b forms
CH−π bonds with His393. All these data accord with 4b having a
high VDR affinity.
compounds faces H12 (Figure 6A). We analyzed the interactions
between the residues Phe418 and Val414 on H12 and the
The complex of C(25) epimer 4a forms normal hydrogen
bonds with Arg270 (2.8 Å), Ser274 (2.8 Å), and His393 (2.5 Å),
weak hydrogen bonds with Trp143 (3.1 Å) and Ser233 (3.2 Å),
and no hydrogen bond with His301 (4.6 Å) (Supporting
Information Figure S1). The VDR therefore anchors 4a more
weakly than 4b.
The complex with 5a has the largest positional shifts at 301−
306 (rmsd 1.8 Å) compared with the 1,25(OH)2D3 complex
(2ZLC) (Figure 5B). It was also noted that the positioning of
DRIP peptide differs significantly (at the LXXLL part, rmsd 0.8
Å). The two A-ring hydroxyl groups form hydrogen bonds with
Ser233, Arg270, Tyr143, and Ser274 (2.7−2.9 Å) (Figure 5F).
The 25-hydroxyl group does not form hydrogen bond with the
imidazole nitrogen of His393 but does with its main chain
carbonyl group (2.8 Å); it forms no hydrogen bond with His301.
The methylene group at C(2) interacts with the same residues
Tyr143 (3.9 Å), Arg270 (3.5 Å), and Phe150 (4.0 Å) as 4b does.
In the complex with rVDR-LBD, 5b interacts similarly to 5a
except around C25. The 25-hydroxyl group forms weak
hydrogen bonds with the imidazole nitrogen of His301 (3.1 Å)
and the main chain carbonyl of His393 (3.0 Å) (Figure 5G). The
C(2) methylene group of 5b interacts differently from that of 5a
and 2MD: in addition to Tyr143 (3.9 Å), Arg270 (3.5 Å), and
Phe150 (4.0 Å), Ser233 (3.9 Å) interacts with the CH2 at a closer
distance than that of 5a complex (4.2 Å). Here, the difference in
the side chain structure affects the conformations around the A
ring too.
Figure 6. Interactions of adamantane ring with H12 residues in the
rVDR-LBD complexes. (A) Overlay of rVDR complexes of 4b (cyan),
5a (magenta), and 5b (yellow). The protein of 4b complex is shown
with pink ribbon with H12 magenta. Interaction of 4b with the H12
residues, Phe418 and Val414, are shown with black dotted lines. The
interactions of other ligands with H12 residues are shown in Table 2. (B)
Overlay of the complexes with partial agonist 4b (cyan), antagonist 3a
(green), and natural hormone 1 (atom type). The protein of 4b complex
is shown with a pink ribbon with H12 in magenta and three residues,
Fe418, Val414 and Leu410, interacting with the adamantane ring of the
ligands are shown.
DISCUSSION
■
We have previously determined the crystal structures of the
complex of rVDR bound to an antagonist/partial agonist 3a
(ADTT) [VDR affinity 80% of the natural hormone (1),
transactivation EC50 10−9 M (15% efficacy), inhibition of the
transactivation of 1 (IC50 3 × 10−9 M)] and the DRIP peptide.
However, the VDR/3a complex adopts the active conforma-
tion.14c The rVDR complex with the one-carbon-longer analogue
3e (ADMI4) [1/20 relative VDR affinity, transactivation EC50
2.4 × 10−8 M (less than 10% efficacy), inhibition of the
transactivation of 1 IC50 3 × 10−8 M] also adopts the active
form.14c We thus confirmed that rVDR complexed with antagonists
can adopt the active conformation in the crystal structure in the
presence of the DRIP peptide. We assumed that the DRIP peptide
trapped the active conformation that exists as a minor
conformation in solution (assuming 10−15% on the basis of
the transactivation efficacy) by binding to its AF2 surface.
It is therefore not strange that the crystal structures of the
rVDR-LBDs bound to the partial agonists 4b, 5a, and 5b adopt
the active conformations. The adamantane ring in all these
adamantane ring of the ligands (4a, 5a, and 5b) in detail and
compared them with those of other known VDR ligands (Table
2). It seems that H12 binds more tightly to the adamantine ring
of 4b than the 26-methyl group of 1,25(OH)2D3, as shown by the
distance between the two residues and side chain terminals of the
ligands (Table 2). The distances between Phe418(422) and
Val414(418) and 1,25(OH)2D3 are equal to or longer than 4 Å in
humans,28c rats,28a,b and zebrafish.28d The side chain terminal of
the super agonist KH1060,28e which has a long side chain, is also
in a similar region (Table 2, entry 10; Chart 1). However, the
adamantine rings of ADTK1 (both 3.8 Å) and antagonist ADMI4
(3.5 and 4.0 Å) are at closer positions (Table 2). According to the
equilibrium theory between active and inactive conforma-
tions,14c,29 we assume that in solution these complexes would
shift to an inactive form. It was reported for the zVDR complexes
of 26,27-hexafluoro analogue CD578 (Chart 1; entry 13, Table
2) that the terminal fluorine atoms of the ligand are very close to
Phe448 (3.6 Å), Val444 (3.5 Å), and Leu440 (3.3 Å), and these
close interactions are correlated with the high potency of the
4081
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
compounds showed significant cell-type selectivity. We are
investigating further activities in in vivo expressions of various
genes and also the effects on the elevation of calcium
concentration and bone mineralization.
Table 2. Distances (Å) between H12 Residues with the
Ligands in VDR
a
VDR/ligand complexes
(PDB code no.)
Phe418 (r), 422
Val414 (r),
418 (h), 444 (z)
b
c
entry
(h), 448, (z)
1
2
3
4
rVDR ADTK1 (4b) (3VTB)
rVDR ADTK3 (5a) (3VTC)
rVDR ADTK4 (5b) (3VTD)
3.8, 3.9
4.0
3.8
EXPERIMENTAL SECTION
4.60
4.8
■
4.2, 4.3
4.5
Chemistry. All nonaqueous reactions were carried out under argon
atmosphere in freshly distilled anhydrous solvents. We conducted high-
pressure liquid chromatography (HPLC) by using Jasco PU-980
intelligent pumps equipped with an 801-SC solvent programmer and a
Jasco UV-970 detector. All samples for biological assays were purified by
HPLC and shown to have a purity of >95%. Columns used are YMC-
Pack ODS-AM SH-342-5 (10 × 250 mm) or CHIRALPACK IE (4.6
mm × 250 mm). Nuclear magnetic resonance spectra were recorded in
CDCl3 solution on a Bruker Ultra Shield 400 MHz spectrometer (400
MHz for 1H NMR and 100 MHz for 13C NMR). Coupling constants are
reported in hertz (Hz). Abbreviations used are singlet (s), doublet (d),
triplet (t), quartet (q), and multiplet (m). Low-resolution mass spectra
(MS) were obtained by electronic ionization on a Shimazu GCMSQP-
2010NC PLUS 100 spectrometer at 70 eV, and m/z values are given
with relative intensities in parentheses. High resolution mass spectra
(HRMS) were obtained by JEOL JMS-T100LP with DART (direct
analysis in real time). Ultraviolet spectra were recorded on a Hitachi U-
3300 spectrometer.
rVDR ADTT (3a)
4.3
(2ZMI)14c
5
rVDR ADMI4 (3e)
3.5
4.0
(2ZMJ)14c
6
rVDR ADNY (3f)
4.1, 4.2
4.3
4.4
(2ZMH)14c
7
rVDR-LBD/1,25(OH)2D3
4.3
(2ZLC)28a
8
hVDR-LBD/1,25(OH)2D3
(1DB1)28c
4.3
4.0
9
zVDR-LBD/1,25(OH)2D3
4.4
4.1
(2HC4)28d
10
hVDR-LBD/KH1060
(1IE8)28e
rVDR-LBD/2MD (1RJK)28b
4.4
4.4, 4.0, 4.3
11
12
4.1, 4.2
3.5
4.5
3.6
zVDR-LBD/CD578
(3DR1)28f
a
b
c
Rat. Human. Zebrafish.
1α-Hydroxy-2-methylidene-24-trimethylsilyl-23,23,24,24-
tetradehydro-19,25,26,27-tetranorvitamin D3 1,3-Bis(tert-bu-
tyldimethylsilyl) Ether (7a). To a solution of trimethylsilylacetylene
(282 μL, 2.03 mmol, 3 equiv) in anhydrous dioxane (3.0 mL) at 0 °C
was added a 3.0 M hexane solution of MeLi (678 μL, 2.03 mmol, 3
equiv). The mixture was stirred at 0 °C for 30 min. To this mixture was
added a dioxane (3.0 mL) solution of tosylate 6 (494.1 mg, 0.678
mmol), and then the mixture was heated in a sealed tube at 105 °C for 15
h. The reaction mixture was cooled to room temperature, a saturated
NH4Cl solution was added, and the mixture was extracted with ethyl
acetate. The organic layer was washed with saline, dried over MgSO4,
and evaporated. The residue was chromatographed on silica gel (54 g)
and eluted with 3% ethyl acetate/hexane to give 7a (356.6 mg, 80%). 7a:
1H NMR (CDCl3) d δ 0.02, 0.05, 0.06, 0.08 (each 3 H, s, OSiMe), 0.16
compound. However, these fluorine atoms interact directly with
Phe422, with no intervening hydrogen atom. The F−C distance
should be compared with (C)H−C distance, which is at least 1 Å
shorter (3.0−3.3 Å) than the C(H)−C distance (4.0−4.3 Å).
The positioning of the adamantane ring of 4b is different from
that of the double bond analogue 3a, an antagonist, as shown in
the VDR complexes (Figure 6B). The adamantine ring of 3a in
the VDR complex moves to the bottom of the protein to avoid
crowding with Phe418 and Vl414 on H12. Furthermore, in the
VDR, the vitamin D derivatives (4 and 5) with an adamantane
ring and a triple bond have different side chain structures from
those in the natural hormone (Figures 5A,B,C,E,F,G and 6B);
this might be one reason why these compounds show different
selectivities compared with the natural hormone.
(9 H, s, CSiMe3), 0.56 (3 H, s, 18-Me), 0.86, 0.90 (each 9H, s, SiBut),
1.09 (3H, d, J = 6.4 Hz, 21-Me), 4.41−4.45 (2H, m, H-1 and −3), 4.92,
4.97 (each 1H, s, CCH2), 5.84 (1H, d, J = 11.2 Hz, H-7), 6.21 (1H, d, J
= 11.2 Hz, H-6). MS (EI) m/z (%): 654 (M+, 2), 522 (30), 450 (10),
366 (10), 234 (10), 73 (100).
CONCLUSIONS
1α-Hydroxy-2-methylidene-23,23,24,24-tetradehydro-
19,25,26,27-tetranorvitamin D3 1,3-Bis(tert-butyldimethylsilyl)
ether (7b). To a solution of 7a (27.0 mg, 0.041 mmol) in THF/MeOH
(1:0.7, 1 mL) was added K2CO3 (28.5 mg, 0.21 mmol, 5 equiv), and the
mixture was stirred at room temperature for 24 h. After addition of a
saturated solution of NH4Cl, the mixture was extracted with ethyl
acetate. The extract was washed with saline, dried over MgSO4, and
evaporated. The residue was chromatographed on silica gel (6.8 g) and
eluted with 1% ethyl acetate/hexane to give 7b (22.8 mg, 95%). 7b: 1H
NMR (CDCl3) δ 0.02, 0.06, 0.07, 0.08 (each 3 H, s, SiMe), 0.56 (3 H, s,
18-Me), 0.86, 0.90 (each 9 H, s, t-Bu), 1.09 (3 H, d, J = 6.4 Hz, 21-Me),
1.95 (1H, t, J = 2.4 Hz, terminal acetylene H), 4.41−4.45 (2 H, m, H-1
and -3), 4.92, 4.97 (each 1 H, s, CCH2), 5.84 (1 H, d, J = 11.2 Hz, H-
7), 6.21 (1 H, d, J = 11.2 Hz, H-6). 13C NMR (CDCl3) δ −4.90, −4.86,
12.17, 18.16, 18.24, 19.09, 22.18, 23.36, 25.60, 25.78, 25.84, 27.53, 28.69,
30.90, 35.65, 38.58, 40.35, 45.58, 45.62, 47.62, 55.33, 56.16, 69.17, 71.63,
72.52, 83.36, 106.28, 116.25, 122.35, 132.91, 140.87, 152.96. MS (EI)
m/z (%): 582 (M+, 2), 450 (65), 366 (18), 351 (10), 234 (18), 73 (100).
HRMS (DART) m/z calcd for C36H63O3Si2 (M+ + OH) 599.432, found
599.427.
■
Newly synthesized adamantyl vitamin D analogues (4a,b and
5a,b) with a triple bond on the side chain are shown to be partial
agonists. The adamantane rings of all the compounds face H12
and interact with Phe418 and Val414; furthermore, the side chain
positioning differs significantly from that of the natural hormone
(Figures 5A−C). The biological potencies of all the analogues
are significantly high in terms of VDR affinity and transactivation
(Figures 1 and 2). These compounds show significant selectivity
in gene expressions in various cell types (Figure 4); for example,
the activities of 4b (ADTK1) in CYP24A1 gene expression in
kidney, intestine, and bone were 1, 1/2, and 1/3, respectively,
compared with the natural hormone. Similarly, CYP24A1 gene
expressions of 5a (ADTK3) in kidney, bone, and monocyte
(U937) were significant for about 50% of the natural hormone
but weak in skin (13%) and intestinal (19%) cells. ADTK2 (4a)
showed selectivity in the expression of different genes; it showed
no activity in the expression of CYP24A1 in monocyte U937 and
THP-1 cells, but showed significant activities, 33% and 22%,
respectively, in the expression of the CAMP gene in the same
cells. The side chain structural differences between the 23-yne-
adamantyl vitamin D compounds (4b and 5a,b) and the natural
hormone (Figures 5A−C) may be a reason why the former
25-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,26,27-trinorvitamin D3 1,3-Bis-
(tert-butyldimethylsilyl) Ether (8). A 1.65 M hexane solution of n-
BuLi (86 μL, 0.14 mmol, 2 equiv) was added to a solution of acetylene
compound 7b (41.0 mg, 0.07 mmol) at 0 °C, and 7 min later a solution
of 1-adamantylformaldehyde (34.7 mg, 0.21 mmol, 3 equiv ) in THF
4082
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
(420 μL) was added. The mixture was stirred at 0 °C for 2 h, and then
saturated NH4Cl solution was added. The mixture was extracted with
ethyl acetate, washed with saline, dried over MgSO4, and evaporated.
The residue was chromatographed on silica gel (4.7 g) and eluted with
5% ethyl acetate/hexane to give 8 (48.8 mg, 93%) as a mixture of
diastereomers at C(25). The mixture was separated by HPLC [Hibar
RT LiChrosorb Si 60 (7 μm), 10 mm × 250 mm, CH2Cl2/hexane 2/3,
4.0 mL/min] to give less polar 8a and more polar 8b. 8a (more polar):
1H NMR (CDCl3) δ 0.02, 0.05, 0.06, 0.08 (each 3H, s, SiMe), 0.56 (3H,
the above experiments. After work-up and chromatography on silica gel
(5 g) with 2% ethyl acetate/hexane, (R)-MTPA ester 11 (5.4 mg, 81%)
was obtained as a 1:1 mixture of epimers at C-25. The mixture was
separated by HPLC [Hiber RT LiChrosorb Si 60 (7 μm), 250 mm × 10
mm, CH2Cl2/hexane 2/3 4.0 mL/min] to give less polar 11a and more
polar 11b. 11a (less polar): 1H NMR (CDCl3) δ 0.025, 0.05, 0.065, 0.08
(each 3 H, s, SiMe), 0.54 (3 H, s, 18-Me), 0.86, 0.90 (each 9 H, s, SiBut),
1.04 (3 H, d, J = 6.4 Hz, 21-Me), 1.56−1.71 (2 H, overlapping, H-26),
3.53 (3 H, s, OMe), 4.41−4.45 (2H, m, H-1 and -3), 4.92, 4.97 (each 1
H, s, CCH2), 5.60 (1 H, t, J = 6.4 Hz, H-25), 5.83 (1 H, d, J = 11.2 Hz,
H-7), 6.21 (1 H, d, J = 11.2 Hz, H-6), 7.37−7.56 (5 H, m, phenyl). 11b
(more polar): 1H NMR (CCDCl3) δ 0.025, 0.05, 0.065, 0.08 (each 3H,
s, SiMe), 0.54 (3 H, s, 18-Me), 0.86, 0.90 (each 9 H, s, SiBut), 1.03 (3 H,
d, J = 6.4 Hz, 21-Me), 1.45−1.88 (2 H, overlapping, H-26), 3.53 (3 H, s,
OMe), 4.41−4.45 (2 H, m, H-1 and −3), 4.92, 4.97 (each 1 H, s, C
CH2), 5.60 (1 H, t, J = 6.4 Hz, H-25), 5.83 (1 H, d, J = 11.2 Hz, H-7),
6.21 (1 H, d, J = 11.2 Hz, H-6), 7.37−7.56 (5 H, m, phenyl).
26-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,27-dinorvitamin D3 1,3-Bis(tert-
butyldimethylsilyl) Ether 25-[(S)-α-Methoxy-α-(trifluorometh-
yl)]- phenylacetate (11c and 11d). 26-Adamantyl-25-hydroxyl
compound 9 (5.6 mg, 7.4 μmol) was allowed to react with (R)-
(−)-MTPACl (9.6 mg, 41 μmol, 5.6 equiv) in the presence of Et3N (10
μL, 74 μmol, 10 equiv) and DMAP (5 mg, 41 μmol, 5.6 equiv) similarly
to the above experiments. After work-up and chromatography on silica
gel (5 g) with 2% ethyl acetate/hexane, (S)-MTPA ester 11 (3 mg, 41%)
was obtained as a 1:1 mixture of epimers at C-25. The mixture was
separated by HPLC [Hiber RT LiChrosorb Si 60 (7 μm), 250 mm × 10
mm, CH2Cl2/hexane 2/3 4.0 mL/min] to give less polar 11c and more
polar 11d. 11c (less polar): 1H NMR (CDCl3) δ 0.025, 0.05, 0.065, 0.08
(each 3 H, s, SiMe), 0.54 (3 H, s, 18-Me), 0.86, 0.90 (each 9 H, s, SiBut),
1.04 (3 H, d, J = 6.4 Hz, 21-Me), 3.53 (3 H, s, OMe), 4.41−4.45 (2 H, m,
H-1,3), 4.92, 4.97 (each 1 H, s, CCH2), 5.60 (1H, t, J = 6.4 Hz, H-25),
5.83 (1H, d, J = 11.2 Hz, H-7), 6.21 (1H, d, J = 11.2 Hz, H-6), 7.36−7.59
(5 H, m, phenyl). 11d (more polar): 1H NMR (CDCl3) δ 0.025, 0.05,
0.065, 0.08 (each 3 H, s), 0.54 (3 H, s), 0.86, 0.90 (each 9 H, s), 1.03 (3
H, d, J = 6.4 Hz, 21-Me), 3.53 (3 H, s, OMe), 4.41−4.45 (2 H, m, H-1,
3), 4.92, 4.97 (each 1 H, s, CCH2), 5.60 (1 H, t, J = 6.4 Hz, H-25),
5.83 (1 H, d, J = 11.2 Hz, H-7), 6.21 (1H, d, J = 11.2 Hz, H-6), 7.38−7.56
(5 H, m, phenyl).
s, 18-Me), 0.86, 0.90 (each 9H, s, SiBut), 1.11 (3H, d, J = 6.4 Hz, 21-Me),
3.86 (1 H, s, H-25), 4.41−4.45 (2H, m, H-1 and −3), 4.92, 4.97 (each
1H, s, CCH2), 5.84 (1H, d, J = 11.2 Hz, H-7), 6.21 (1H, d, J = 11.2 Hz,
H-6). MS m/z (%): 746 (M+, 2), 614 (18), 596 (20), 366 (20), 234 (12),
135 (100), 73 (80). 8b (less polar): 1H NMR (CDCl3) δ 0.02, 0.05, 0.07,
0.08 (each 3H, s, SiMe), 0.56 (3H, s, 18-Me), 0.86, 0.90 (each 9H, s,
SiBut), 1.10 (3H, d, J = 2.4 Hz, 21-Me), 3.86 (1 H, s, H-25), 4.41−4.45
(2H, m, H-1 and −3), 4.92, 4.97 (each 1H, s, CCH2), 5.84 (1H, d, J =
11.2 Hz, H-7), 6.21 (1H, d, J = 11.2 Hz, H-6). MS m/z (%): 746 (M+, 2),
614 (18), 596 (20), 366 (20), 234 (12), 135 (100), 73 (80).
26-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19, 27-dinorvitamin D3 1,3-Bis(tert-
butyldimethylsilyl) Ether (9). Similarly, one-carbon longer homo-
logue 9 was synthesized from 7b (6.4 mg, 0.011 mmol) and 1-
adamantylacetaldehyde (5.88 mg, 0.033 mmol, 3 equiv). After
chromatography on silica gel (1.1 g) with 1% ethyl acetate/hexane, 9
(5.3 mg, 80.4%) was obtained as a 1:1 mixture of C-25 epimers. 9: 1H
NMR (CDCl3) δ 0.02, 0.05, 0.06, 0.08 (each 3 H, s, SiMe), 0.55 (3 H, s,
18-Me), 0.86, 0.90 (each 9 H, s, SiBut), 1.06 (3 H, d, J = 8.0 Hz, 21-Me),
4.41−4.45 (2 H, m, H-1 and −3), 4.51 (1 H, m, 25-H), 4.92, 4.97 (each 1
H, s, CCH2), 5.84 (1 H, d, J = 11.2 Hz, H-7), 6.21 (1 H, d, J = 11.2 Hz,
H-6). MS m/z (%): 760 (M+, 2), 610 (16), 475 (18), 366 (20), 234
(12), 135 (100), 73 (75).
25R-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,26,27-trinorvitamin D3 1,3-Bis-
(tert-butyldimethylsilyl) Ether 25-[(R)-α-Methoxy-α-(trifluoro-
methyl)]-phenylacetate (10a). To a solution of 25-hydroxyl
compound 8a (2.3 mg, 3.1 μmol) in CH2Cl2 (400 μL) were added
Et3N (4.3 μL, 0.031 mmol, 10 equiv) and DMAP (2.1 mg, 0.017 mmol,
5.6 equiv). To this solution was added at 0 °C a solution of (S)-(+)-α-
methoxy-α-(trifluoromethyl)]-phenylacetyl chloride (MTPA-Cl, 3.8
mg, 0.016 mmol, 5 equiv) in CH2Cl2 (250 μL) and stirred at the
same temperature for 10 min and then at room temperature for 45 min.
DMAP (2.8 mg, 0.023 mmol, 7.5 equiv) was added to the mixture and
stirred further at room temperature for 2 h. Ice water was added to the
reaction, and the mixture was extracted with ethyl acetate, and the
extracts were washed with saline, dried over MgSO4, and evaporated.
The residue was chromatographed on silica gel to give 10a (200 μg,
(25R)-25-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,26,27-trinorvitamin D3 (4a). The
less polar 8a (1.35 mg, 1.8 μmol) was treated with CSA (1.68 mg, 7.2
μmol, 4 equiv) in MeOH (500 μL) at room temperature for 3 h.
Saturated NaHCO3 solution was added to the reaction at 0 °C, the
mixture was extracted with ethyl acetate, the extract was washed with
saline, dried over MgSO4, and evaporated. The residue was chromato-
graphed on Sephadex LH-20 (1 g) and eluted with CHCl3/hexane/
1
6.6%). 10a: H NMR (CDCl3) δ 0.02, 0.05, 0.07, 0.08 (each 3 H, s,
SiMe), 0.56 (3 H, s, 18-Me), 0.86, 0.90 (each 9 H, s, SiBut)), 1.10 (3 H,
d, J = 6.4 Hz, 21-Me), 3.59 (3 H, s, OMe), 4.41−4.45 (2 H, m, H-1 and
−3), 4.92, 4.97 (each 1 H, s, CCH2), 5.10 (1 H, s, H-25), 5.83 (1 H, d,
J = 11.2 Hz, H-7), 6.21 (1 H, d, J = 11.2 Hz, H-6), 7.36−7.59 (5 H, m,
phenyl).
1
MeOH 70/30/1 to give 4a 880 μg, 94%). 4a (ADTK2): H NMR
(CDCl3) δ 0.57 (3 H, s, 18-Me), 1.10 (3 H, d, J = 6.4 Hz, 21-Me), 3.86 (1
H, s, H-25), 4.44−4.51 (2H, m, H-1, 3), 5.10, 5.11 (each 1H, s, C
CH2), 5.89 (1H, d, J = 12 Hz, H-7), 6.36 (1H, d, J = 12 Hz, H-6). 13
C
NMR (CDCl3) δ 12.23, 19.41, 22.25, 23.45, 25.99, 27.47, 28.29, 28.94,
29.70, 35.73, 37.15, 37.50, 37.79, 38.15, 40.29, 45.70, 45.79, 55.32, 56.32,
70.71, 71.81, 71.93, 77.22, 80.26, 85.13, 107.77, 115.43, 124.20, 130.58,
143.13, 151.93. MS m/z (%): 518 (M+, 10), 365 (10), 347 (10), 295
(10), 135 (100), 93 (25), 79 (25). HRMS (DART) m/z calcd for
C35H50O3 (M+) 518.376, found 518.368.
(25S)-25-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,26,27-trinorvitamin D3 (4b). The
more polar 8b (1.07 mg, 1.4 μmol) was similarly treated with CSA (1.31
mg, 5.6 μmol, 4 equiv) in MeOH (700 μL) at room temperature for 3 h.
After similar work-up, the residue was chromatographed on Sephadex
LH-20 (1 g) and eluted with CHCl3/hexane/MeOH 70/30/1 to give 4b
(683 μg, 91%). 4b (ADTK1): 1H NMR (CDCl3) δ: 0.57 (3H, s, 18-Me),
1.10 (3H, d, J = 6.4 Hz, 21-Me), 3.86 (1H, d, J = 4 Hz, H-25), 4.46−4.50
(2H, m, H-1, 3), 5.10, 5.11 (each 1H, s, CCH2), 5.89 (1H, d, J = 12
Hz, H-7), 6.36 (1H, d, J = 12 Hz, H-6). 13C NMR (CDCl3) δ 12.23,
19.41, 22.25, 23.45, 25.98, 27.46, 28.30, 28.94, 29.70, 35.71, 37.15, 37.49,
37.80, 38.15, 40.29, 45.70, 45.79, 55.32, 56.32, 70.71, 71.81, 71.92, 77.21,
(25R)-25-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,26,27-trinorvitamin D3 1,3-Bis-
(tert-butyldimethylsilyl) Ether 25-[(S)-α-Methoxy-α-(trifluoro-
methyl)]- phenylacetate (10b). 25-Hydroxyl compound 8a (3.0
mg, 4 μmol) was treated with (R)-(−)-MTPA-Cl to give (S)-MTPA
ester (10b) (500 mg, 13%) similarly as above. 10b: 1H NMR (CDCl3) δ
0.02, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.56 (3 H, s, 18-Me), 0.86, 0.90
(each 9 H, s, SiBut), 1.05 (3 H, d, J = 6.4 Hz, 21-Me), 3.55 (3 H, s, OMe),
4.41−4.45 (2 H, m, H-1 and −3), 4.92, 4.97 (each 1 H, s, CCH2), 5.06
(1 H, s, H-25), 5.83 (1 H, d, J = 11.2 Hz, H-7), 6.21 (1 H, d, J = 11.2 Hz,
H-6), 7.36−7.59 (5H, m, phenyl).
26-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,27-dinorvitamin D3 1,3-Bis(tert-
butyldimethylsilyl) Ether 25-[(R)-α-Methoxy-α-(trifluorometh-
yl)]- phenylacetate (11a and 11b). 26-Adamantyl-25-hydroxyl
compound 9 (5.2 mg, 6.8 μmol) was allowed to react with (S)−(+)−
MTPA-Cl (8.9 mg, 38 μmol, 5.6 equiv) in the presence of Et3N (9.5 μL,
68 μmol, 10 equiv) and DMAP (5 mg, 38 μmol, 5.6 equiv) similarly to
4083
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
80.26, 85.13, 107.77, 115.43, 124.20, 130.58, 143.13, 151.94. MS m/z
(%): 518 (M+, 10), 365 (10), 347 (10), 295 (10), 135 (100), 93 (25), 79
(25). HRMS (DART) m/z calcd for C35H50O3 (M+) 518.376, found
518.366.
4.40−4.46 (2 H, m, H-1 and -3), 4.92, 4.98 (each 1 H, s, CCH2), 5.84
(1 H, d, J = 11.2 Hz, H-7), 6.21 (1 H, d, J = 11.2 Hz, H-6). MS m/z (%):
759 (M+, 2), 701 (3), 626 (40), 383 (10), 366 (30), 243 (30), 135 (100),
73 (50). HRMS (DART) m/z calcd for C48H79O4Si2 (M+ + OH)
775.552, found 775.588.
(25R)-26-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,27-dinorvitamin D3 (5a). To a
solution of the less polar isomer of R-MTPA ester 11a (1.03 mg, 1
μmol) in MeOH (500 μL)was added a solution of CSA (1.22 mg, 5.2
μmol, 5 equiv) in MeOH (200 μL) at 0 °C, and the mixture was stirred
at room temperature for 3 h. Saturated NaHCO3 solution was added at 0
°C, the mixture was extracted with ethyl acetate, and the extract was
washed with saline, dried over MgSO4, and evaporated. The residue was
dissolved in MeOH (600 μL), K2CO3 (130 mg) was added, and the
mixture was stirred at room temperature for 24 h. Saturated NH4Cl
solution was added to the reaction mixture at 0 °C, and the mixture was
extracted with ethyl acetate. The extract was washed with saline, dried,
and evaporated. The residue was chromatographed on Sephadex LH-20
(1 g) and eluted with CHCl3/hexane/MeOH 70/30/1 to give 5a (314
μg, 56%). 5a (ADTK3): 1H NMR (CDCl3) δ 0.56 (3 H, s, 18-Me), 1.07
(3 H, d, J = 6.4 Hz, 21-Me), 4.45−4.53 (3 H, m, H-1,-3, and 25), 5.10,
5.11 (each 1 H, s, CCH2), 5.88 (1H, d, J = 11.2 Hz, H-7), 6.36 (1H, d,
J = 11.2 Hz, H-6). MS m/z (%): 532 (M+, 10), 429 (10), 361 (10), 309
(10), 135 (100), 93 (40), 79 (40). HRMS (DART) m/z calcd for
C36H52O3 (M+) 532.392, found 532.373.
Stereo Selective Reduction of 12a with (4R)-2-Methyl-4,5,5-
triphenyl-1,3,2-oxazaborolidine. To a solution of 25-keto com-
pound 12a (6.1 mg, 8.2 μmol) in THF (10 μL) was added a solution of
(R)-BMTO (4.0 mg, 13 μmol, 1.6 equiv) and borane−dimethyl sulfide
complex (BMS, 10.0−10.2 M, 1.4 μL, 14 μmol, 1.7 equiv) in THF (25
μL) at 0 °C, and the mixture was stirred for 1 h. MeOH was added at 0
°C, and the mixture was evaporated. The residue was chromatographed
on silica gel (4.7 g) and eluted with 5% ethyl acetate/hexane to give 8a
(4.3 mg, 5.8 μmol 70%, 78% de). 8a: 1H NMR (CDCl3) δ 0.02, 0.05,
0.06, 0.08 (each 3 H, s, SiMe), 0.56 (3 H, s, 18-Me), 0.86, 0.90 (each 9 H,
s, SiBut), 1.11 (3 H, d, J = 6.4 Hz, 21-Me), 3.86 (1 H, s, H-25), 4.41−4.45
(2 H, m, H-1 and −3), 4.92, 4.97 (each 1 H, s, CCH2), 5.84 (1 H, d, J
= 11.2 Hz, H-7), 6.21 (1 H, d, J = 11.2 Hz, H-6). MS m/z (%): 746 (M+,
2), 614 (18), 596 (20), 366 (20), 234 (12), 135 (100), 73 (80).
Stereo Selective Reduction of 12a with (R)-3,3-Diphenyl-1-
methyltetrahydro-1H,3H-pyrrolo[1,2-c][1,3,2]oxazaborole. To a
solution of 25-keto compound 12a (4.4 mg, 5.9 μmol) in THF (20 μL)
was added a solution of (R)-CBS (2.1 mg, 7.6 μmol, 1.3 equiv) and BMS
(10.0−10.2 M, 0.8 μL, 8.0 μmol, 1.3 equiv) in THF (14 μL) at 0 °C, and
the mixture was stirred for 50 min. MeOH was added at 0 °C, and the
mixture was evaporated. The residue was chromatographed on silica gel
(5.5 g) and eluted with 5% ethyl acetate/hexane to give 8a (3.8 mg, 86%,
and 91% de).
Stereo Selective Reduction of 12a with (S)-3,3-Diphenyl-1-
methyltetrahydro-1H,3H-pyrrolo[1,2-c][1,3,2]oxazaborole. To a
solution of 25-keto compound 12a (23.6 mg, 32 μmol) in THF (100
μL) was added a solution of (S)-CBS (12.5 mg, 45 μmol, 1.4 equiv) and
BMS (10.0−10.2 M, 4.5 μL, 45 μmol, 1.4 equiv) in THF (80 μL) at 0 °C,
and the mixture was stirred for 50 min. After similar work-up, the residue
was chromatographed on silica gel to give 8b (17.8 mg, 75%, 87% de)
and recovered 12a (2.2 mg, 9.3%). 8b: 1H NMR (CDCl3) δ 0.02, 0.05,
0.07, 0.08 (each 3 H, s, SiMe), 0.56 (3 H, s, 18-Me), 0.86, 0.90 (each 9 H,
s, SiBut), 1.10 (3 H, d, J = 2.4 Hz, 21-Me), 3.86 (1 H, s, H-25), 4.41−4.45
(2 H, m, H-1 and −3), 4.92, 4.97 (each 1 H, s, CCH2), 5.84 (1 H, d, J
= 11.2 Hz, H-7), 6.21 (1 H, d, J = 11.2 Hz, H-6). MS m/z (%): 746 (M+,
2), 614 (18), 596 (20), 366 (20), 234 (12), 135 (100), 73 (80).
Stereo Selective Reduction of 12b with a (R)-3,3-Diphenyl-1-
methyltetrahydro-1H,3H-pyrrolo-[1,2-c][1,3,2]oxazaborole.
Similarly, one-carbon longer homologue 9a (25R) was synthesized from
12b (2.1 mg, 2.7 μmol) by reduction with (R)-CBS (2.2 mg, 7.9 μmol,
2.9 equiv) and BMS (10.0−10.2 M, 0.92 μL, 9.2 μmol, 3.4 equiv) in
THF (180 μL). After chromatography on silica gel (6.0 g) with 5% ethyl
acetate/hexane, 9a (1.4 mg, 1.9 μmol, 68%, >95%de) was obtained. 9a:
1H NMR (CDCl3) δ 0.02, 0.05, 0.06, 0.08 (each 3 H, s, SiMe), 0.55 (3 H,
The more polar isomer of S-MTPA ester 11d was similarly
hydrolyzed to give 5a.
(25S)-26-(1-Adamantyl)-1α,25-dihydroxy-2-methylidene-
23,23,24,24-tetradehydro-19,27-dinorvitamin D3 (5b). The more
polar isomer of longer side chain (R)-Mosher ester 11b (1.05 mg, 1.1
mmol) was deprotected similarly to the above experiment. After
purification by Sephadex LH-20 (1 g) column chromatography, 5b (383
μg, 67%) was obtained. 5b (ADTK4): 1H NMR (CDCl3) δ 0.56 (3 H, s,
18-Me), 1.07 (3 H, d, J = 6.4 Hz, 21-Me), 4.45−4.53 (3 H, m, H-1, -3,
and -25), 5.10, 5.11 (each 1 H, s, CCH2), 5.88 (1H, d, J = 11.2 Hz, H-
7), 6.36 (1H, d, J = 11.2 Hz, H-6). MS m/z (%): 532 (M+, 10), 429 (10),
361 (10), 309 (10), 135 (100), 93 (40), 79 (40). HRMS (DART) m/z
calcd for C36H52O3 (M+) 532.392, found 532.368.
The less polar isomer of S-MTPA ester 11c was similarly hydrolyzed
to give 5b.
25-(1-Adamantyl)-1α-hydroxy-2-methylidene-25-oxo-
23,23,24,24-tetradehydro-19,26,27-trinorvitamin D3 1,3-Bis-
(tert-butyldimethylsilyl) Ether (12a). To a solution of an epimeric
mixture at C(25) of 8 (39.0 mg, 52 μmol) in CH2Cl2 (1.4 mL) was
added Dess−Martin periodinane (DMP, 85.9 mg, 202 μmol, 4 equiv)
and stirred at room temperature for 4 h. Saturated Na2SO3 solution was
added to the reaction, and the mixture was extracted with CH2Cl2, and
the extracts were washed with saline, dried over MgSO4, and evaporated.
The residue was chromatographed on silica gel to give 12a (32.8 mg,
1
84%). 12a: H NMR (CDCl3) δ 0.03, 0.05, 0.07, 0.08 (each 3 H, s,
s, 18-Me), 0.86, 0.90 (each 9 H, s, SiBut), 1.06 (3 H, d, J = 8.0 Hz, 21-
Me), 4.41−4.45 (2 H, m, H-1 and -3), 4.51 (1 H, m, 25-H), 4.92, 4.97
(each 1 H, s, CCH2), 5.84 (1 H, d, J = 11.2 Hz, H-7), 6.21 (1 H, d, J =
11.2 Hz, H-6). MS m/z (%): 760 (M+, 2), 610 (16), 475 (18), 366 (20),
234 (12), 135 (100), 73 (75).
Stereo Selective Reduction of 12b with a (S)-3,3-Diphenyl-1-
methyltetrahydro-1H,3H-pyrrolo[1,2-c][1,3,2]oxazaborole.
Similarly, one-carbon longer homologue 9b (25S) was synthesized from
12b (2.1 mg, 2.7 μmol) by reduction with (S)-CBS (2.2 mg, 7.8 μmol,
2.8 equiv) and BMS (10.0−10.2 M, 0.92 μL, 9.2 μmol, 3.4 equiv) in
THF (180 μL). After chromatography on silica gel (6.0 g) with 5% ethyl
acetate/hexane, 9b (1.3 mg, 1.7 μmol, 62%, >95% de) was obtained. 9b:
1H NMR (CDCl3) δ 0.02, 0.05, 0.06, 0.08 (each 3 H, s, SiMe), 0.55 (3 H,
SiMe), 0.57 (3 H, s, 18-Me), 0.86, 0.90 (each 9 H, s, SiBut), 1.14 (3 H, d,
J = 6.8 Hz, 21-Me), 4.40−4.45 (2 H, m, H-1 and -3), 4.93, 4.97 (each 1
H, s, CCH2), 5.85 (1 H, d, J = 11.2 Hz, H-7), 6.21 (1 H, d, J = 11.2 Hz,
H-6). 13C NMR (CDCl3) δ −4.90, −4.85, 0.22, 12.17, 18.16, 18.24,
19.55, 22.17, 23.33, 25.78, 25.84, 26.40, 27.69, 27.91, 28.67, 35.60, 36.56,
38.12, 38.59, 40.42, 45.63, 46.73, 47.62, 55.49, 56.23, 71.63, 72.51, 77.20,
80.12, 94.57, 106.31, 116.40, 122.28, 133.11, 140.59, 152.93, 194.06. MS
m/z (%): 745 (M+, 1), 612 (36), 383 (10), 366 (20), 229 (100), 135
(30), 73 (50). HRMS (DART) m/z calcd for C47H77O4Si2 (M+ + OH)
761.536, found 761.528.
26-(1-Adamantyl)-1α-hydroxy-2-methylidene-25-oxo-
23,23,24,24-tetradehydro-19,27-dinorvitamin D3 1,3-Bis(tert-
butyldimethylsilyl) Ether (12b). To a solution of 9 (29.7 mg, 39
μmol) in CH2Cl2 (1.4 mL) was added DMP (46.7b mg, 110 μmol, 2.8
equiv), and the mixture was stirred at room temperature for 2.5 h.
Saturated Na2SO3 solution was added to the reaction, and the mixture
was extracted with CH2Cl2, and the extracts were washed with saline,
dried over MgSO4, and evaporated. The residue was chromatographed
on silica gel to give 12b (20.1 mg, 68%). 12b: 1H NMR (CDCl3) δ 0.03,
0.05, 0.06, 0.08 (each 3 H, s, SiMe), 0.56 (3 H, s, 18-Me), 0.86, 0.90
(each 9 H, s, SiBut), 1.12 (3 H, d, J = 6.8 Hz, 21-Me), 2.15 (2 H, s, H-26),
s, 18-Me), 0.86, 0.90 (each 9 H, s, SiBut), 1.06 (3 H, d, J = 8.0 Hz, 21-
Me), 4.41−4.45 (2 H, m, H-1 and -3), 4.51 (1 H, m, 25-H), 4.92, 4.97
(each 1 H, s, CCH2), 5.84 (1 H, d, J = 11.2 Hz, H-7), 6.21 (1 H, d, J =
11.2 Hz, H-6). MS m/z (%): 760 (M+, 2), 610 (16), 475 (18), 366 (20),
234 (12), 135 (100), 73 (75).
Cell Lines and Cell Cultures. Human kidney HEK293 cells
(RIKEN Cell Bank, Tsukuba, Japan) were cultured in Dulbecco’s
Modified Eagle’s Medium (DMEM) containing 5% inactivated fetal
4084
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
bovine serum (FBS), 100 U/mL penicillin, and 0.1 mg/mL
streptomycin (Nacalai Tesque, Kyoto, Japan). Human colon carcinoma
SW480 cells (American Type Culture Collection, Manassas, VA) and
immortalized keratinocyte HaCaT cells (kindly provided by Dr. Tadashi
Terui, Department of Dermatology, Nihon University School of
Medicine) were cultured in DMEM containing 10% FBS, 100 U/mL
penicillin, and 0.1 mg/mL streptomycin. Human ostesarcoma MG63
cells (RIKEN Cell Bank) were cultured in minimum essential medium
containing 10% FBS, 100 U/mL penicillin, and 0.1 mg/mL
streptomycin. Human myeloid leukemia THP-1 cells and U937 cells
(RIKEN Cell Bank) were cultured in RPMI1640 medium containing
10% FBS, 100 U/mL penicillin, and 0.1 mg/mL streptomycin. The
condition of cell culturing was kept at 37 °C in a humidified atmosphere
containing 5% CO2.
Plasmids. The expression vectors pCMX-VDR, pCMX-RXRα,
pCMX-VP16-VDR, pCMX-GAL4-RXRα, pCMX-GAL4-SRC-1,
pCMX-GAL4-N-CoR, and pCMX-GAL4-SMART were reported
previously.13b The nuclear receptor-interacting domains of SRC-1
(amino acids 595−771; GenBank accession no. U90661), N-CoR
(amino acids 1990−2416; GenBank accession no. U35312), and SMRT
(amino acids 2003−2517; GenBank accession no. AF113003) were
inserted into the pCMX-GAL4 vector to make pCMX-GAL4-SRC-1,
pCMX-GAL4-N-CoR, and pCMX-GAL4-SMRT, respectively. VDR
responsive Spp × 3-tk-LUC and GAL4-responsive MH100(UAS) × 4-
tk-LUC reporter vectors were previously reported.14d pGEX vector (GE
Healthcare, Chalfont St. Giles, Buckinghamshire, UK) was used to
generate glutathione transferase (GST) fusions.30 The human
recombinant VDR ligand-binding domain (LBD) (amino acids 140−
427) was inserted into pGEX vector to generate pGEX-VDR.
Vitamin D Receptor-Binding Assay. The pGEX or pGEX-hVDR
was expressed as a GST fusion protein in Escherichia coli BL21 (EMD
Millipore). The cells were lysed by sonication in sonication buffer (50
mM Tris-HCl pH 8.0, 50 mM NaCl, 1 mM EDTA, and 1 mM DTT).
Then 1 μg of the supernatants was diluted in binding buffer (25 mM Tris
pH 7.5, 100 mM KCl, 25 mM DTT, 4 mM CHAPS, pH 7.5) containing
bovine serum albumin (100 μg/mL). A solution containing an
increasing amount of 1,25(OH)2D3 (1) or synthetic analogues in 15
μL of EtOH was added to 570 μL of the receptor solution in each tube,
and the mixture was vortexed 2−3 times. The mixture was incubated for
30 min at room temperature. [26,27-Methyl-3H] 1,25(OH)2D3
(PerkinElmer) in 15 μL of EtOH was added, vortexed 2−3 times, and
the whole mixture was then allowed to stand at 4 °C for 20 h. At the end
of the second incubation, 400 μL of dextran-coated charcoal solution
(Sigma) was added to bind any free ligands (or to remove free ligands)
and the sample was vortexed. After 30 min at room temperature, bound
and free [3H]-1,25(OH)2D3 were separated by centrifugation at 3000
rpm for 10 min at 0 °C. Aliquots (800 μL) of the supernatant were mixed
with 9.2 mL of Bio Fluor (PerkinElmer) and submitted for radioactivity
counting. Each assay was performed at least twice in triplicate.
Transcription Assays. HEK293 cells were cultured in Dulbecco’s
Modified Eagle’s Medium containing 5% fetal bovine serum and
antibiotic−antimycotics (Nacalai) at 37 °C in a humidified atmosphere
of 5% CO2 in air. Transfections of 50 ng of TK-Spp × 3-LUC reporter
plasmid, 10 ng of pCMX-β-galactosidase, 15 ng of pCMX-VDR, and 15
ng of pCMX-RXRα for each well of a 96-well plate were performed by
the calcium phosphate coprecipitation method as described previous-
ly.14d Then 18 h after transfection, test compounds were added. Cells
were harvested after 16−24 h for luciferase and β-galactosidase activity
using a luminometer (Molecular Devices, Sunnyvale, CA).
chloroform method.31 cDNAs were synthesized using the ImProm-II
reverse transcription system (Promega, Madison, WI).14d Real-time
PCR was performed on the ABI PRISM 7000 sequence detection
system (Applied Biosystems, Foster City, CA) using Power SYBR Green
PCR Master Mix (Applied Biosystems). Primer sequences were as
follows: CYP24A1:5′-TGAACGTTGGCTTCAGGAGAA-3′ and 5′-
A G G G T G C C T G A G T G T A G C A T C T - 3 ′ , T R P V 6 : 5 ′ -
TGAACGTTGGCTTCAGGAGAA-3′ and 5′-AGGGTGCCT-
G A G T G T A G C A T C T - 3 ′ , C A M P 5 ′ - G C T A A C C T C -
TACCGCCTCCT-3′ and 5′-GGTCACTGTCCCCATACACC-3′,
and ACTIN: 5′-GACAGGATGCAGAAGGAGAT-3′ and 5′-GAAG-
CATTTGCGGTGGACGAT-3′. mRNA values were normalized to an
amount of ACTIN mRNA.
Protein Expression and Purification. The rat VDR LBD (residues
116−423, Δ165−211) was cloned as an N-terminal His6-tagged fusion
protein into the pET14b expression vector and was overexpressed in E.
coli C41. The cells were grown at 37 °C in LB medium (including 100
mg/L ampicilin) and were subsequently induced for 6 h with 15 μM
isopropyl-β-D-thiogalactopyranoside (IPTG) at 23 °C. The purification
procedure included affinity chromatography on a Ni-NTA column,
followed by dialysis and ion-exchange chromatography (SP-Sepharose).
After tag removal by thrombin digestion, the protease was removed by
filtration through a HiTrap benzamidine column, and the protein was
further purified by gel filtration on a Superdex 200 column. The purity
and homogeneity of the rVDR LBP were assessed by SDS-PAGE.
Crystallization. Purified rVDR-LBD solution was concentrated to
about 0.75 mg/mL by ultrafiltration. To an aliquot (800 μL) of the
protein solution was added a ligand (4a, 4b, 5a or 5b, ca. 10 equiv); the
solution was further concentrated to about 1/8, and then a solution (25
mM Tris-HCl, pH 8.0; 50 mM NaCl; 10 mM DTT; 0.02% NaN3) of
coactivator peptide (H2N-KNHPMLMNLLKDN-CONH2) derived
from DRIP205 was added. This solution of VDR/ligand/peptide was
allowed to crystallize by the vapor-diffusion method that used a series of
precipitant solutions containing 0.1 M MOPS-NaOH (pH 7.0), 0.1−0.4
M sodium formate, 12−22% (w/v) PEG4000, and 5% (v/v) ethylene
glycol. Droplets for crystallization were prepared by mixing 2 μL of
complex solution and 1 μL of precipitant solution, and droplets were
equilibrated against 500 μL of precipitant solution at 20 °C. It took 1−2
days to obtain crystals of X-ray diffraction quality for VDR complexes
with 4a, 4b, 5a, or 5b as a ligand.
Diffraction Experiment and Structure Analysis. Prior to the
diffraction data collection, crystals were soaked in a cryoprotectant
solution that contained 0.1 M MOPS-NaOH (pH 7.0), 0.1−0.4 M
sodium formate, 15−20% PEG4000, and 17−20% ethylene glycol.
Diffraction data sets were collected at 100 K in a stream of nitrogen gas
at beamlines BL-6A of KEK-PF and NW12A of PF-AR (Tsukuba,
Japan). Reflections were recorded with an oscillation range per image of
1.0°. Diffraction data were indexed, integrated, and scaled using the
program HKL2000.32 The structures of the complex were solved by
molecular replacement with the program CNS,33 and finalized sets of
atomic coordinates were obtained after iterative rounds of model
modification with the program XtalView34 and after refinement with
CNS by rigid body refinement, simulated annealing, positional
minimization, water molecule identification, and individual isotropic
B-value refinement.
ASSOCIATED CONTENT
■
S
* Supporting Information
The X-ray crystal structure of rVDR-LBD complexed with
ADTK2 (4a); unit cell structures of the ternary rVDR-LBD
complexes with 4b (ADTK1), 5a (ADTK3 or 1,25(OH)2D3 (1),
and DRIP205; inter unit-cell interactions in the rVDR-LBD
complex with 4b; ternary rVDR-LBD complexes with ligands
(4b, 5a, and 5b) and DRIP205 peptide in canonical active
conformations; summary of data collection statistics and
refinement; X-ray crystal data of compound 4a (ADTK2). This
material is available free of charge via the Internet at http://pubs.
acs.org.
Mammalian two-hybrid assay for cofactor interaction to VDR was
used 50 ng of TK-MH100(UAS) × 4-LUC reporter plasmid, 10 ng of
pCMX-β-galactosidase, 15 ng of pCMX-GAL4-RXRα, pCMX-GAL4-
SRC1, pCMX-GAL4-NCoR, or pCMX-GAL4-SMRT and 15 ng of
pCMX-VP16-VDR for each well of a 96-well plate. Luciferase data were
normalized to the internal β-galactosidase control.
Reverse Transcription and Quantitative Real-Time PCR
Analysis. For gene expression analysis, 1 × 104 cells per well were
plated in 24 well plate. After 24 h, cells were treated with ethanol control
or 100 nM of 1,25(OH)2D3 or its analogues for 24 h. Total RNAs from
samples were prepared by the acid guanidine thiocyanate−phenol/
4085
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
Kutilek, S.; Adami, S.; Zanchetta, J.; Libanati, C.; Siddhanti, S.;
Christiansen, C. Denosumab for prevention of fractures in post-
menopausal women with osteoporosis. N. Engl. J. Med. 2009, 361, 756−
765.
Accession Codes
PDB codes for 4b, 5a, and 5b are 3VTB, 3VTC and 3VTD,
respectively.
(7) Xue, Y.; Karaplis, A. C.; Hendy, G. N.; Goltzman, D.; Miao, D.
Exogenous 1,25-dihydroxyvitamin D3 exerts a skeletal anabolic effect
and improves mineral ion homeostasis in mice that are homozygous for
both the 1alpha-hydroxylase and parathyroid hormone null alleles.
Endocrinology 2006, 147, 4801−4810.
AUTHOR INFORMATION
Corresponding Authors
*For S.Y.: Phone, +81 3 3972 8111 ext 2363; E-mail, yamada.
vd@image.ocn.ne.jp.
*For M.M.: Phone, +81 3 3972 8111 ext 2243; E-mail,
makishima.makoto@nihon-u.ac.jp.
■
(8) Kubodera, N. D-Hormone derivatives for the treatment of
osteoporosis: from alfacalcidol to eldecalcitol. Mini-Rev. Med. Chem.
2009, 9, 1416−1422.
Author Contributions
⊥These authors contributed equally.
(9) Shevde, N. K.; Plum, L. A.; Clagett-Dame, M.; Yamamoto, H.; Pike,
J. W.; DeLuca, H. F. A potent analog of 1alpha,25-dihydroxyvitamin D3
selectively induces bone formation. Proc. Natl. Acad. Sci. U. S. A. 2002,
99, 13487−13491. Plum, L. A.; Fitzpatrick, L. A.; Ma, X.; Binkley, N. C.;
Zella, J. B.; Clagett-Dame, M.; DeLuca, H. F. 2MD, a new anabolic agent
for osteoporosis treatment. Osteoporosis Int. 2006, 17, 704−715.
(10) DeLuca, H. F.; Bedale, W.; Binkley, N.; Gallagher, J. C.;
Bolognese, M.; Peacock, M.; Aloia, J.; Clagett-Dame, M.; Plum, L. The
vitamin D analogue 2MD increases bone turnover but not BMD in
postmenopausal women with osteopenia: results of a 1-year phase 2
double-blind, placebo-controlled, randomized clinical trial. J. Bone
Miner. Res. 2011, 26, 538−545.
(11) Ma, Y.; Khalifa, B.; Yee, Y. K.; Lu, J.; Memezawa, A.; Savkur, R. S.;
Yamamoto, Y.; Chintalacharuvu, S. R.; Yamaoka, K.; Stayrook, K. R.;
Bramlett, K. S.; Zeng, Q. Q.; Chandrasekhar, S.; Yu, X. P.; Linebarger, J.
H.; Iturria, S. J.; Burris, T. P.; Kato, S.; Chin, W. W.; Nagpal, S.
Identification and characterization of noncalcemic, tissue-selective,
nonsecosteroidal vitamin D receptor modulators. J. Clin. Invest. 2006,
116, 892−904.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Part of this work was supported by an Adaptable and Seamless
Technology Transfer Program Feasibility Study
(AS232Z02361G) from Japan Science and Technology Agency.
This work was also supported in part by Grants-in-Aid for
Scientific Research (grant no. 25460394) from JSPS and Grants-
in-Aid for Scientific Research (C) (grant no. 25460157) from
MEXT. We acknowledge MEXT Supported Program for the
Strategic Research Foundation at Private Universities, 2013−
2018. The synchrotron-radiation experiment was performed at
the Photon Factory. We acknowledge the help provided by the
beamlines scientists at the Photon Factory. We thank professor
Hector F. DeLuca at the Department of Biochemistry, University
of WisconsinMadison, for providing an expression plasmid of
the rVDR-LBD.
(12) Shiau, A. K.; Barstad, D.; Loria, P. M.; Cheng, L.; Kushner, P. J.;
Agard, D. A.; Greene, G. L. The structural basis of estrogen receptor/
coactivator recognition and the antagonism of this interaction by
tamoxifen. Cell 1998, 95, 927−937.
ABBREVIATIONS USED
■
(13) (a) Lusher, S. J.; Raaijmakers, H. C.; Vu-Pham, D.; Kazemier, B.;
Bosch, R.; McGuire, R.; Azevedo, R.; Hamersma, H.; Dechering, K.;
Oubrie, A.; van Duin, M.; de Vlieg, J. X-ray structures of progesterone
receptor ligand binding domain in its agonist state reveal differing
mechanisms for mixed profiles of 11β-substituted steroids. J. Biol. Chem.
2012, 287, 20333−20343. (b) Lusher, S. J.; Raaijmakers, H. C.; Vu-
Pham, D.; Dechering, K.; Lam, T. W.; Brown, A. R.; Hamilton, N. M.;
Nimz, O.; Bosch, R.; McGuire, R.; Oubrie, A.; de Vlieg, J. Structural basis
for agonism and antagonism for a set of chemically related progesterone
receptor modulators. J. Biol. Chem. 2011, 286, 35079−35086.
(14) (a) Igarashi, M.; Yoshimoto, N.; Yamamoto, K.; Shimizu, M.;
Ishizawa, M.; Makishima, M.; DeLuca, H. F.; Yamada, S. Identification
of a highly potent vitamin D receptor antagonist: (25S)-26-adamantyl-
25-hydroxy-2-methylene-22,23-didehydro-19,27-dinor-20-epi-vitamin
D3 (ADMI3). Arch. Biochem. Biophys. 2007, 460, 240−253. (b) Inaba,
Y.; Yamamoto, K.; Yoshimoto, N.; Matsunawa, M.; Uno, S.; Yamada, S.;
Makishima, M. Vitamin D3 derivatives with adamantane or lactone ring
side chains are cell type-selective vitamin D receptor modulators. Mol.
Pharmacol. 2007, 71, 1298−1311. (c) Nakabayashi, M.; Yamada, S.;
Yoshimoto, N.; Tanaka, T.; Igarashi, M.; Ikura, T.; Ito, N.; Makishima,
M.; Tokiwa, H.; DeLuca, H. F.; Shimizu, M. Crystal structures of rat
vitamin D receptor bound to adamantyl vitamin D analogs: structural
basis for vitamin D receptor antagonism and partial agonism. J. Med.
Chem. 2008, 51, 5320−5329. (d) Inaba, Y.; Yamamoto, K.; Yoshimoto,
N.; Matsunawa, M.; Uno, S.; Yamada, S.; Makishima, M. Vitamin D3
derivatives with adamantane or lactone ring side chains are cell type-
selective vitamin D receptor modulators. Mol. Pharmacol. 2007, 71,
1298−1311.
DMP, dimethylaminopyridine; CSA, camphor sulfonic acid;
DMP, Dess−Martin periodinane; CBS, Corey−Bakshi−Shibata;
RXRα, retinoid X receptor α; SRC-1, steroid receptor coactivator
1; NCoA1, nuclear receptor coactivator 1; N-CoR, nuclear
receptor corepressor 1; SMRT, silencing mediator of retinoic
acid and thyroid hormone receptor; N-CoR2, nuclear receptor
corepressor 2; AF2, activation function 2; CAMP, cathelicidin
antimicrobial peptides; TRPV6, transient receptor potential
vanilloid 6; MTPA, α-methoxy-α-(trifluoromethyl)]-phenyl-
acetyl; BMTO, B-methyl-4,5,5-triphenyl-1,3,2-oxazaborolidine;
BMS, borane−dimethyl sulfide complex
REFERENCES
■
(1) Pike, J. W.; Zella, L. A.; Meyer, M. B.; Fretz, J. A.; Kim, S. Molecular
actions of 1,25-dihydroxyvitamin D3 on genes involved in calcium
homeostasis. J. Bone Miner. Res. 2007, 22 (Suppl 2), V16−V19.
(2) Lacey, D. L.; Timms, E.; Tan, H. L.; Kelley, M. J.; Dunstan, C. R.;
Burgess, T.; Elliott, R.; Colombero, A.; Elliott, G.; Scully, S.; Hsu, H.;
Sullivan, J.; Hawkins, N.; Davy, E.; Capparelli, C.; Eli, A.; Qian, Y. X.;
Kaufman, S.; Sarosi, I.; Shalhoub, V.; Senaldi, G.; Guo, J.; Delaney, J.;
Boyle, W. J. Osteoprotegerin ligand is a cytokine that regulates
osteoclast differentiation and activation. Cell 1998, 93, 165−176.
(3) Yasothan, U.; Kar, S. Osteoporosis: overview and pipeline. Nature
Rev. Drug Discovery 2008, 7, 725−726.
(4) Kawai, M.; Modder, U. I.; Khosla, S.; Rosen, C. J. Emerging
̈
therapeutic opportunities for skeletal restoration. Nature Rev. Drug
(15) (a) Itsuno, S.; Ito, K.; Hirao, A.; Nakahama, S. Asymmetric
reduction of aromatic ketones with the reagent prepared from (S)-
(−)-2-amino-3-methyl-1,1-diphenylbutan-1-ol and borane. J. Chem.
Soc.: Chem. Commun. 1983, 8, 469. (b) Corey, E. J.; Bakshi, R. K.;
Shibata, S. Highly enantioselective borane reduction of ketones
catalyzed by chiral oxazaborolidines. Mechanism and synthetic
Discovery 2011, 10, 141−156.
(5) Chitre, M.; Shechter, D.; Grauer, A. Denosumab for treatment of
postmenopausal osteoporosis. Am. J. Health-Syst. Pharm. 2011, 68,
1409−1418.
(6) Cummings, S. R.; San Martin, J.; McClung, M. R.; Siris, E. S.;
Eastell, R.; Reid, I. R.; Delmas, P.; Zoog, H. B.; Austin, M.; Wang, A.;
4086
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087

Journal of Medicinal Chemistry
Article
implications. J. Am. Chem. Soc. 1987, 109, 5551−5553. (c) Corey, E. J.;
Helal, C. J. Reduction of Carbonyl Compounds with Chiral Oxaza-
borolidine Catalysts: A New Paradigm for Enantioselective Catalysis
and a Powerful New Synthetic Method. Angew. Chem., Int. Ed. 1998, 37,
1986−2012. (d) Bach, J.; Berenguer, R.; Garcia, J.; Loscertales, T.;
Vilarrasa, J. Highly Enantioenriched Propargylic Alcohols by Oxazabor-
olidine-Mediated Reduction of Acetylenic Ketones. J. Org. Chem. 1996,
61, 9021−9025.
(16) (a) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H.; High-field,
F. T. NMR application of Mosher’s method. The absolute
configurations of marine terpenoids. J. Am. Chem. Soc. 1991, 113,
4092−4096. (b) Kusumi, T.; Fujita, Y.; Ohtani, I.; Kakisawa, K. Anomaly
in the modified Mosher’s method: absolute configurations of some
marine cembranolides. Tetrahedron Lett. 1991, 32, 2923−2926.
(17) Kobayashi, E.; Shimazaki, M.; Miyamoto, Y.; Masuno, H.;
Yamamoto, K.; DeLuca, H. F.; Yamada, S.; Shimizu, M. Structure−
activity relationships of 19-norvitamin D analogs having a fluoroethy-
lidene group at the C-2 position. Bioorg. Med. Chem. 2007, 15, 1475−
1482.
(18) Ishizawa, M.; Matsunawa, M.; Adachi, R.; Uno, S.; Ikeda, K.;
Masuno, H.; Shimizu, M.; Iwasaki, K.; Yamada, S.; Makishima, M.
Lithocholic acid derivatives act as selective vitamin D receptor
modulators without inducing hypercalcemia. J. Lipid Res. 2008, 49,
763−772.
(19) MacDonald, P. N.; Dowd, D. R.; Nakajima, S.; Galligan, M. A.;
Reeder, M. C.; Haussler, C. A.; Ozato, K.; Haussler, M. R. Retinoid X
receptors stimulate and 9-cis-retinoic acid inhibits 1,25-dihydroxyvita-
min D3-activated expression of the rat osteocalcin gene. Mol. Cell. Biol.
1993, 13, 5907−5917.
(20) Choi, M.; Yamada, S.; Makishima, M. Dynamic and ligand-
selective interactions of vitamin D receptor with retinoid X receptor and
cofactors in living cells. Mol. Pharmacol. 2011, 80, 1147−1155.
(21) Onate, S. A.; Boonyaratanakornkit, V.; Spencer, T. E.; Tsai, S. Y.;
Tsai, M. J.; Edwards, D. P.; O’Malley, B. W. The steroid receptor
coactivator-1 contains multiple receptor interacting and activation
domains that cooperatively enhance the activation function 1 (AF1) and
AF2 domains of steroid receptors. J. Biol. Chem. 1998, 273, 12101−
12108.
(22) Underhill, C.; Qutob, M. S.; Yee, S. P.; Torchia, J. A novel nuclear
receptor corepressor complex, N-CoR, contains components of the
mammalian SWI/SNF complex and the corepressor KAP-1. J. Biol.
Chem. 2000, 275, 40463−40470.
and crystal structure. Bioorg. Med. Chem. 2008, 16, 6949−6964.
(b) Vanhooke, J. L.; Benning, M. M.; Bauer, C. B.; Pike, J. W.;
DeLuca, H. F. Molecular structure of the rat vitamin D receptor ligand
binding domain complexed with 2-carbon-substituted vitamin D3
hormone analogues and a LXXLL-containing coactivator peptide.
Biochemistry 2004, 43, 4101−4110. (c) Rochel, N.; Wurtz, J. M.;
Mitschler, A.; Klaholz, B.; Moras, D. The crystal structure of the nuclear
receptor for vitamin D bound to its natural ligand. Mol. Cell 2000, 5,
173−179. (d) Ciesielski, F.; Rochel, N.; Moras, D. Adaptability of the
Vitamin D nuclear receptor to the synthetic ligand Gemini: remodelling
the LBP with one side chain rotation. J. Steroid Biochem. Mol. Biol. 2007,
103, 235−242. (e) Tocchini-Valentini, G.; Rochel, N.; Wurtz, J. M.;
Mitschler, A.; Moras, D. Crystal structures of the vitamin D receptor
complexed to superagonist 20-epi ligands. Proc. Natl. Acad. Sci. U. S. A.
2001, 98, 5491−5496. (f) Eelen, G.; Valle, N.; Sato, Y.; Rochel, N.;
Verlinden, L.; De Clercq, P.; Moras, D.; Bouillon, R.; Munoz, A.;
̃
Verstuyf, A. Superagonistic fluorinated vitamin D3 analogs stabilize helix
12 of the vitamin D receptor. Chem. Biol. 2008, 15, 1029−1034.
(29) Smith, C. L.; O’Malley, B. W. Coregulator function: a key to
understanding tissue specificity of selective receptor modulators. Endocr.
Rev. 2004, 25, 45−71.
(30) Kaneko, E.; Matsuda, M.; Yamada, Y.; Tachibana, Y.; Shimomura,
I.; Makishima, M. Induction of Intestinal ATP-binding Cassette
Transporters by a Phytosterol-Derived Liver X Receptor Agonist. J.
Biol. Chem. 2003, 278, 36091−36098.
(31) Tavangar, K.; Hoffman, A. R.; Kraemer, F. B. A micromethod for
the isolation of total RNA from adipose tissue. Anal. Biochem. 1990, 186,
60−63.
(32) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data
collected in oscillation mode. Methods Enzymol. 1997, 276, 307−326.
(33) Brunger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.; Gros,
P.; Grosse-Kunstleve, R. W.; Jiang, J. S.; Kuszewski, J.; Nilges, M.;
Pannu, N. S.; Read, R. J.; Rice, L. M.; Simonson, T.; Warren, G. L.
Crystallography & NMR system: a new software suite for macro-
molecular structure determination. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 1998, 54 (Pt 5), 905−921.
(34) McRee, D. E. XtalView Xfita versatile program for
manipulating atomic coordinates and electron density. J. Struct. Biol.
1999, 125 (2−3), 156−165.
(23) Park, E. J.; Schroen, D. J.; Yang, M.; Li, H.; Li, L.; Chen, J. D.
SMRTe, a silencing mediator for retinoid and thyroid hormone
receptors-extended isoform that is more related to the nuclear receptor
corepressor. Proc. Natl. Acad. Sci. U. S. A. 1999, 96, 3519−3524.
(24) Chen, K. S.; Prahl, J. M.; DeLuca, H. F. Isolation and expression of
human 1,25-dihydroxyvitamin D3 24-hydroxylase cDNA. Proc. Natl.
Acad. Sci. U. S. A. 1993, 90, 4543−4547.
(25) Liu, P. T.; Stenger, S.; Li, H.; Wenzel, L.; Tan, B. H.; Krutzik, S. R.;
Ochoa, M. T.; Schauber, J.; Wu, K.; Meinken, C.; Kamen, D. L.; Wagner,
M.; Bals, R.; Steinmeyer, A.; Zugel, U.; Gallo, R. L.; Eisenberg, D.;
̈
Hewison, M.; Hollis, B. W.; Adams, J. S.; Bloom, B. R.; Modlin, R. L.
Toll-like receptor triggering of a vitamin D-mediated human
antimicrobial response. Science 2006, 311, 1770−1773.
(26) Peng, J. B.; Chen, X. Z.; Berger, U. V.; Weremowicz, S.; Morton,
C. C.; Vassilev, P. M.; Brown, E. M.; Hediger, M. A. Human calcium
transport protein CaT1. Biochem. Biophys. Res. Commun. 2000, 278,
326−32. Wood, R. J.; Tchack, L.; Taparia, S. 1,25-Dihydroxyvitamin D3
increases the expression of the CaT1 epithelial calcium channel in the
Caco-2 human intestinal cell line. BMC Physiol. 2001, 1, 11.
(27) Wu, Q.; Burghardt, R.; Safe, S. Vitamin D-interacting protein 205
(DRIP205) coactivation of estrogen receptor alpha (ERalpha) involves
multiple domains of both proteins. J. Biol. Chem. 2004, 279, 53602−
53612.
(28) (a) Shimizu, M.; Miyamoto, Y.; Takaku, H.; Matsuo, M.;
Nakabayashi, M.; Masuno, H.; Udagawa, N.; DeLuca, H. F.; Ikura, T.;
Ito, N. 2-Substituted-16-ene-22-thia-1alpha,25-dihydroxy-26,27-di-
methyl-19-norvitamin D3 analogs: Synthesis, biological evaluation,
4087
dx.doi.org/10.1021/jm401989c | J. Med. Chem. 2014, 57, 4073−4087