A mixed population of antagonist and agonist binding conformers in a single crystal explains partial agonism against vitamin D receptor: Active vitamin D analogues with 22R-alkyl group

Article abstract of DOI:10.1021/jm500392t

We are continuing to study the structural basis of vitamin D receptor (VDR) agonism and antagonism by using 22S-alkyl vitamin D analogues. Here we report the synthesis and biological evaluation of 22R-alkyl analogues and the X-ray crystallographic analysis of vitamin D receptor ligand-binding domain (VDR-LBD) complexed with a 22R-analogue. VDR-LBD complexed with the partial agonist 8a showed that 8a binds to VDR-LBD with two conformations, one of which is the antagonist/VDR-LBD complex structure and the other is the agonist/VDR-LBD complex structure. The results indicate that the partial agonist activity of 8a depends on the sum of antagonistic and agonistic activities caused by the antagonist and agonist binding conformers, respectively. The structural basis observed here must be applicable to the partial agonism of other ligand-dependent nuclear receptors. This is the first report describing the trapping of a conformational subset of the ligand and the nuclear receptor in a single crystal.

Full text of DOI:10.1021/jm500392t

Article
pubs.acs.org/jmc
A Mixed Population of Antagonist and Agonist Binding Conformers
in a Single Crystal Explains Partial Agonism against Vitamin D
Receptor: Active Vitamin D Analogues with 22R‑Alkyl Group
Yasuaki Anami, Toshimasa Itoh, Daichi Egawa, Nobuko Yoshimoto, and Keiko Yamamoto*
Laboratory of Drug Design and Medicinal Chemistry, Showa Pharmaceutical University, 3-3165 Higashi-Tamagawagakuen, Machida,
Tokyo 194-8543, Japan
ABSTRACT: We are continuing to study the structural basis
of vitamin D receptor (VDR) agonism and antagonism by
using 22S-alkyl vitamin D analogues. Here we report the
synthesis and biological evaluation of 22R-alkyl analogues and
the X-ray crystallographic analysis of vitamin D receptor
ligand-binding domain (VDR-LBD) complexed with a 22R-
analogue. VDR-LBD complexed with the partial agonist 8a
showed that 8a binds to VDR-LBD with two conformations,
one of which is the antagonist/VDR-LBD complex structure
and the other is the agonist/VDR-LBD complex structure. The
results indicate that the partial agonist activity of 8a depends on the sum of antagonistic and agonistic activities caused by the
antagonist and agonist binding conformers, respectively. The structural basis observed here must be applicable to the partial
agonism of other ligand-dependent nuclear receptors. This is the first report describing the trapping of a conformational subset of
the ligand and the nuclear receptor in a single crystal.
Chart 1
INTRODUCTION
■
1α,25-Dihydroxyvitamin D3, 1, is an active form of vitamin D
and is recognized as a hormone that plays a pivotal role in
calcium homeostasis, cell differentiation and proliferation, and
immunomodulation.1 It exerts actions through gene tran-
scription by binding to vitamin D receptor (VDR) belonging to
the nuclear receptor superfamily.2 VDR changes conformation
from an inactive form to an active form by binding 1 as well as
various agonists. Upon binding, VDR forms a heterodimer with
retinoid X receptor (RXR). The RXR−VDR heterodimer on
response element of the target gene recruits coactivator and
transcripts the target gene.3
Various vitamin D analogues have been developed by many
groups, since hormone 1 is pharmaceutically important. In fact,
VDR agonists are clinically used to treat metabolic bone
diseases and skin diseases such as psoriasis.4,5 All analogues
clinically used are VDR agonists. VDR antagonists are also
interesting in terms of the mechanism of VDR antagonism in
addition to their potential use for treating VDR hyperfunction.
At present, VDR antagonists are classified structurally into
three types. The first type is composed of analogues with a
bulky side chain, such as carboxylic ester (ZK series)
compounds6 (Chart 1) and adamantane compounds.7 The
second type is composed of analogues without a bulky side
chain, such as (23S)-25-dehydro-1α-hydroxyvitamin D3-26,23-
lactone (TEI9647)8 (Chart 1) and its derivatives.8−12 The third
type is composed of analogues with a 22S-alkyl group, such as
22S-butyl analogue 2 (Figure 1).
agonism and antagonism by using 22S-alkyl analogues. First, we
synthesized 22S-alkylated analogues systematically and eval-
uated their biological activity. We found that these analogues
show strong antagonist, partial agonist, or full agonist activity,
depending on the side chain structure.15 Next, we conducted X-
ray crystallographic analysis of the ligand-binding domain
(LBD) of VDR complexed with 22S-analogue and found that
the 22S-butyl group induces an extra cavity.13,16 Figure 1 shows
the Connolly channel surface of the ligand-binding pocket
(LBP) and clearly indicates the generation of an extra cavity
(butyl pocket) to accommodate the butyl group (Figure 1b and
Figure 1c). These crystal structures provided important
evidence for understanding the structural basis of agonism
and antagonism. To date, we have confirmed the follow-
We identified 22S-butyl analogue 2 as a VDR antagonist13,14
and have been continuing to study the structural basis of VDR
Received: March 13, 2014
Published: April 17, 2014
© 2014 American Chemical Society
4351
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Figure 1. Connolly channel surface of the ligand-binding pocket observed in the crystal structure of the VDR-LBD/ligand complex and the ligand
structure: (a) hormone 1/VDR-LBD complex,20 (b) antagonist 2/VDR-LBD complex,13 (c) agonist 3/VDR-LBD complex,16 (d) GEMINI 4/VDR-
LBD complex.23 (b−d) 22S-Butyl group of 2 and 3 or the second side chain of 4 induce butyl pocket formation. Of these, the ligand that poorly
interacts with the C-terminus (helix 11/12) of VDR shows antagonistic activity like 2, while the ligand that intimately interacts with C-terminus
(helix 11/12) via hydrogen bond with His393 shows agonistic activity like 3 and 4.
Figure 2. Structures of vitamin D analogues.
ing.13−16 (1) Analogues with a 22S-butyl group induce butyl
pocket formation to accommodate the butyl group (Figure
1b,c). (2) Among ligands that induce butyl pocket formation, a
ligand acts as an antagonist when it interacts poorly with the C-
4352
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Scheme 1. Synthesis of Compounds 5a,b, 6a,b, and 9a,b
terminus (helix 11/12) of VDR (Figure 1b), whereas a ligand
acts as an agonist when it interacts intimately with the C-
terminus via a hydrogen bond with His393 (Figure 1c,d). (3)
Ligands that do not induce an extra cavity like a butyl pocket
exhibit increased agonistic activity when the ligand increases
interactions with the C-terminus of VDR (Figure 1a). Thus,
butyl pocket formation in VDR strongly affects the agonistic or
antagonistic behavior of ligands.
We were interested in determining whether the analogues
with a 22R-alkyl group show antagonistic activity and how the
side chain of the analogue is accommodated in the LBP. Here
we report the synthesis and biological evaluation of 22R-alkyl
analogues as a counterpart of 22S-alkyl compounds and the X-
ray crystallographic analysis of VDR-LBD complexed with a
22R-analogue.
prepared by the procedure reported previously.15,17 Treatment
of E-enoate 10 with Bu2CuLi in the presence of HMPA and
TMSCl afforded diastereoselective conjugate addition product
11a (81%) with 93% diastereoselectivity.17 Compound 11a was
reduced with DIBAL to give alcohol 12a, which was then
deprotected with CSA to afford desired compound 5a in 85%
yield from 11a. Compound 11a was treated with MeLi to give
24,24-dimethyl compound 13a, which was then deprotected
with CSA to afford target compound 6a in 75% yield from 11a.
Target compound 9a was obtained by a similar procedure using
EtMgBr from 11a in 83% yield.
22R-Ethyl compounds 5b, 6b, and 9b were synthesized by
the same synthetic method as shown in Scheme 1. Introduction
of the 22R-ethyl group into 10 was performed with 95%
diastereoselectivity.
22R-Butyl analogues with the hydroxyl group at the 25-
position, 7a and 8a, were synthesized as shown in Scheme 2.
Tosylate 15a derived from 12a was treated with KCN to give
cyano compound 16a, which was then reduced with DIBAL to
give aldehyde 17a. The aldehyde 17a was reduced with NaBH4
to alcohol 18a and then deprotected with CSA to afford the
target compound 7a. Oxidation of aldehyde 17a to ester 19a
was performed by treating with iodine and KOH in methanol.
This oxidation reaction was developed by us previously.18 Ester
19a was treated with MeLi and then CSA to afford compound
SYNTHESIS
■
We designed and synthesized new analogues with a 22R-alkyl
substituent, 5a−9a and 5b−9b, to clarify the effects of
stereochemistry at the 22-position on biological activity and
VDR-LBD conformation (Figure 2). Synthesis was performed
using the method reported previously for the synthesis of 22S-
alkyl analogues.15,16
As shown in Scheme 1, 22R-butyl vitamin D derivatives 5a,
6a, and 9a were synthesized from E-enoate 10, which was
4353
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Scheme 2. Synthesis of Compounds 7a,b and 8a,b
a
Table 1. VDR Binding Affinity of Synthetic Analogues 5−9, IC50 (nM)
a
Competitive binding of 1 and synthetic compounds 5−9 to the human vitamin D receptor. The experiments were carried out in duplicate. The IC50
values are derived from dose−response curves and represent the compound concentration required for 50% displacement of radiolabeled 1α,25-
dihydroxyvitamin D3 from the receptor protein.
8a. 22R-Ethyl compounds 7b and 8b were obtained by the
same synthetic method (Scheme 2).
analogues (5a−8a) showed stronger affinity than the
corresponding ethyl analogues (5b−8b) except for 9a and
9b. Analogues with a dimethyl alcohol at the side chain
terminus (6a,b and 8a,b) showed more potent VDR affinity
compared to the corresponding analogues with a primary
alcohol (5a,b and 7a,b).
The transactivation ability of synthetic compounds 5−9 was
examined using the mouse osteopontin luciferase reporter gene
assay system in Cos7 cells and HEK293 cells. The results are
shown in Figure 3. In Cos7 cells, 6a, 7a, and 5b−7b showed
little transactivation, while they concentration-dependently
inhibited the transactivation induced by the hormone 1.
BIOLOGICAL ACTIVITIES
■
Binding affinity for VDR was evaluated by a competitive
binding assay using tritiated hormone 1. Recombinant human
VDR-LBD, which was expressed as a C-terminus GST-tagged
protein using the pGEX-VDR vector15,19 in E. coli BL21, was
used as the VDR. The results are summarized in Table 1. All
compounds showed significant and specific affinity for the
VDR, indicating that they are VDR ligands. 22R-Butyl
4354
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Figure 3. Transactivation of compounds 5−9 in Cos7 cells (a−d) and in HEK293 cells (e−h). Transcriptional activity was evaluated by the dual
luciferase assay using a full-length human VDR expression plasmid (pCMX-hVDR), a reporter plasmid containing three copies of mouse osteopontin
VDRE (SPPx3-TK-Luc), and an internal control plasmid containing sea pansy luciferase expression constructs (pRL-CMV) as described
previously.13 In Cos7 cells, the luciferase activity of 10−8 M of 1 was defined as 1 (a−d), and in HEK293 cells the activity of 10−9 M of 1 was defined
as 1 (e−h). Inhibitory effect on the transactivation induced by 1,25-(OH)2D3 (1) was also evaluated (c, d, g, h).
Table 2. Summary of Data Collection Statistics and Refinement
ligand
8a (crystal A)
KEK-PF BL-5A
8a (crystal B)
KEK-PF BL-5A
9a
KEK-PF BL-5A
X-ray source
wavelength (Å)
space group
1.000 00
1.000 00
1.000 00
C2
C2
C2
unit cell dimensions
bond (Å)
a = 127.84, b = 44.81, c = 46.06
α = 90.00, β = 93.57, γ = 90.00
13.52−1.90 (1.94−1.90)
66 487
a = 127.05, b = 44.77, c = 45.11
α = 90.00, β = 94.20, γ = 90.00
44.99−2.00 (2.06−2.00)
63 521
a = 128.21, b = 44.31, c = 45.81
α = 90.00, β = 93.56, γ = 90.00
63.98−2.40 (2.51−2.40)
18 809
angle (deg)
a
resolution range (Å)
total no. of reflections
no. of unique reflections
20 411
17 061
8562
a
% completeness
98.6 (99.7)
98.6 (98.9)
84.3 (84.9)
ab
,
Rmerge
Refinement Statistics
0.061 (0.336)
0.066 (0.310)
0.058 (0.210)
a
resolution range (Å)
13.49−1.90 (1.94−1.90)
0.1885/0.1625
45.03−2.00 (2.06−2.00)
0.2240/0.1989
45.76−2.40 (2.51−2.40)
0.1944/0.1762
ac
,
R factor (Rfree/Rwork
)
a
b
Values in parentheses are for the highest-resolution shell. Rmerge = ∑|(Ihkl − ⟨Ihkl⟩)|/(∑Ihkl), where ⟨Ihkl⟩ is the mean intensity of all reflections
c
equivalent to reflection hkl. Rwork (Rfree) = ∑||Fobs| − |Fcalc||/∑|Fobs|, where 5% of randomly selected data were used for Rfree
.
4355
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Figure 4. Binding of 22R-butyl vitamin D analogues to VDR-LBD. Electron density maps with refined compound 8a bound to the VDR-LBD and
Connolly channel surface with 8a in crystal A (a−d) and in crystal B (e−h) are shown. Electron density map and Connolly channel surface with
compound 9a (i, j) are also depicted. (c, g) Superimposition of major and minor conformers.
Therefore, they are VDR antagonists. Compounds 5a, 8a, 8b,
and 9b showed weak transactivation and inhibited hormone-
induced activation, indicating that they are partial agonists for
VDR. Only 9a showed full agonistic activity. In HEK293 cells,
7a and 5b−7b showed antagonistic activity; 5a, 6a, and 8a
showed partial agonistic activity; and 9a, 8b, and 9b showed full
agonistic activity.
PEG4000, and 5% ethylene glycol. Distinct crystals of the
VDR-LBD/8a complex were also obtained when the reservoir
contained 0.1 M MOPS-Na (pH 7.0), 0.075 M sodium formate,
19% (w/v) PEG4000, and 5% ethylene glycol. We termed the
former crystals “crystal A” and the latter crystals “crystal B”.
Crystals of the VDR-LBD/9a complex were obtained when the
reservoir contained 0.1 M MOPS-Na (pH 7.0), 0.125 M
sodium formate, 16% (w/v) PEG4000, and 5% ethylene glycol.
Each structure of the protein was elucidated by molecular
replacement using the structure of the VDR-LBD/1 complex
(PDB code 2ZLC).20 Table 2 summarizes the data collection
statistics and refinement for the three structures. All three
proteins had a space group of C2 with approximate unit cells of
a = 128 Å, b = 44 Å, and c = 46 Å. Structures of crystals A and
B were determined at 1.9 and 2.0 Å resolution, respectively.
The crystal structure of VDR-LBD complexed with 9a was
determined at 2.4 Å resolution.
X-RAY CRYSTAL STRUCTURE
■
To understand the influence of the 22R-alkyl substituent on
biological activity and protein conformation, we attempted
cocrystallization of rat VDR-LBD and 22R-alkyl vitamin D
derivatives 5−9. In the absence of coactivator peptide, we could
not obtain crystals. However, in the presence of coactivator
peptide DRIP205 we could obtain crystals of VDR-LBD
accommodating partial agonist 8a and crystals accommodating
full agonist 9a. Good quality crystals of the VDR-LBD/8a
complex were obtained when the reservoir contained 0.1 M
MOPS-Na (pH 7.0), 0.1 M sodium formate, 20% (w/v)
All three crystals showed the formation of a butyl pocket to
accommodate the 22-butyl group or the original side chain
4356
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Figure 5. Interactions between a ligand and VDR. (a, d) In the major conformer, the 1α- and 3β-hydroxyl groups of 8a form pincer-type hydrogen
bonds with Ser233 and Arg270 and with Tyr143 and Ser274, respectively, while only His301 is used to form the hydrogen bond with the 25-
hydroxyl group via water. (b, e) In the minor conformer, the 1α- and 3β-hydroxyl groups of 8a form pincer-type hydrogen bonds, as is the case with
the major conformer. The 25-hydroxyl group of 8a forms a hydrogen bond with both His301 and His393 via water. (c) The benzene ring of Phe418
in the major conformer is distorted compared with the canonical Phe, but the benzene ring in the minor conformer is not. (f) Compound 9a
interacts with VDR-LBD in a fashion similar to the 22S-epimer 3, including hydrogen bond formation and hydrophobic interactions.
bearing the hydroxyl group of the ligand (Figure 4). In
addition, the main-chain structures of all three crystals were
similar to that of the VDR-LBD/2 complex (data not shown).
The 2Fo − Fc map of crystal A indicated that partial agonist
8a binds to VDR with two conformations (Figure 4a−c). The
ratio was 7:3. In the major conformer (70%), the original side
chain bearing the 25-hydroxyl group of 8a occupied the butyl
pocket and the 22R-butyl group occupied the canonical pocket,
as expected (Figure 4a,d). In contrast, in the minor conformer
(30%), the original side chain occupied the canonical pocket of
VDR and the 22R-butyl group occupied the butyl pocket
(Figure 4b,d). Figure 4c shows a superimposition of Figure 4a
and Figure 4b: the 2Fo − Fc map clearly indicates the existence
of two conformers. Interestingly, the Connolly surface of the
LBP indicated the generation of a small channel opening to the
outside at the top of the butyl pocket (Figure 4d).
The 1α- and 3β-hydroxyl groups of 8a were found to form
pincer-type hydrogen bonds with Ser233 and Arg270 and with
Tyr143 and Ser274, respectively (Figure 5a,b), as is the case
with most vitamin D analogues.20,21 In the minor conformer,
the 25-hydroxyl group at the side chain of 8a forms a hydrogen
bond with both His301 and His393 via water (Figure 5b), while
in the major conformer, only His301 is used to form a
hydrogen bond with the 25-hydroxyl group via water (Figure
5a).
From the 2Fo − Fc map, the side chains of His301, His393
and Phe418 lining the LBP of the VDR-LBD were found to
adopt two conformations, with the occupancy ratio correspond-
ing to the major conformer (70%) and the minor (30%)
(Figure 5a−c). Thus, the protein conformation was coupled to
the ligand conformation. The imidazole ring of His301 in the
minor conformer was flipped over compared with the canonical
His, whereas its orientation in the major conformer was the
same as that of the canonical His (Figure 5a,b).20,21 The
benzene ring of Phe418 in the major conformer was distorted
compared with the canonical Phe because of steric repulsion
with the 22R-butyl group of 8a, whereas in the minor
conformer, it was not (Figure 5c). Thus, we were successful
in trapping a conformational subset of the ligand and the
protein in a single crystal. We assigned the minor conformer as
the agonist binding form of the VDR-LBD/8a complex because
it has a pincer-type hydrogen bond between the 25-hydroxyl
group and both His301 and His393 via water and exhibits
intimate interactions between the ligand and helix 11/12. In
contrast, the major conformer reflects an antagonist binding
form of the complex because of the lack of the pincer-type
hydrogen bond and the intimate interactions between the
ligand and helix 11/12. The fractional populations of the major
and minor conformers must therefore contribute to the
antagonistic and agonistic activities, respectively.
4357
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Figure 6. Ligand-specific repositioning of Leu305 triggers butyl pocket formation. (a) Interactions of VDR with hormone 1 (green), 22S-butyl
analogue 2 (cyan), and 22R-butyl analogue 9a (yellow). Ligand-specific repositioning of the side chain of Leu305 was observed. (b) Interactions of
VDR with hormone 1 (green), GEMINI 4 (orange), and 22R-butyl analogue 8a (blue). The original side chain of 8a and 4 caused drastic
repositioning of not only the side chain of Leu305 but also the main chain of Leu305. (c) Hydrogen bond between the carbonyl of Leu305 and the
NH of Leu309 to form the α-helix. The color of each molecule is same as in (a). (d) Disappearance of the hydrogen bond between the carbonyl of
Leu305 and the NH of Leu309 for the α-helix formation. The color of each molecule is the same as in (b). (e) New channel generated in the 8a/
VDR-LBD complex. The channel is generated by the repositioning of the main chain of Glu392, which is caused by the repositioning of the side
chain of Leu305.
The overall structure of crystal B is quite similar to that of
crystal A. The 2Fo − Fc map of crystal B also indicated that
partial agonist 8a binds to VDR with two conformations, with
the ratio 7:3 (Figure 4e−g). The Connolly surface of the LBP
indicated that the channel opening to the outside is slightly
larger than in crystal A (Figure 4h). Unlike in crystal A, the
electron density for His393 was concentrated, indicating that
His393 adopts one conformation (Figure 5d,e), whereas
electron density for Leu305 was diffuse, indicating that
Leu305 adopts two conformations (data not shown).
As shown in Figure 4i, the 2Fo − Fc map indicated that the
full agonist 9a/VDR-LBD complex was crystallized in only one
conformation. In this complex, surprisingly, the 22R-butyl
group of 9a occupies the butyl pocket and the side chain with
the hydroxyl group occupies the canonical pocket of VDR
(Figure 4i,j), similar to the minor conformer of the 8a/VDR-
LBD complex. This observation was unexpected, as we had
observed that the 22S-butyl group of 3, which is the 22-epimer
of 9a, occupies the butyl pocket,16 so we had expected that the
22R-butyl group of 9a would occupy the canonical pocket of
VDR. Interestingly, 9a interacts with VDR-LBD in a fashion
similar to the 22S-epimer 3,16 including hydrogen bond
formation and hydrophobic interactions (Figure 5f). The
binding mode of 9a is thus surprising but coincides well with
the agonistic activity of 9a.
DISCUSSION
■
Previously, we found that the 22S-alkyl group of vitamin D
analogues 2 and 3 was docked into the butyl pocket (Figure 1b
and Figure 1c).13,16 Therefore, we assumed that the 22R-alkyl
group of the analogues synthesized here would be docked into
the canonical pocket of VDR-LBD and, consequently, that the
22R-alkyl analogues would be unable to interact intimately with
the C-terminus (helix 11/12) via a hydrogen bond with His393.
For this reason, we predicted that the 22R-alkyl analogues
would act as a VDR antagonist. The majority of the analogues
worked as an antagonist, as expected, but 5a, 8a,b, and 9a,b
acted as either a partial agonist or a full agonist.
Cocrystallization of VDR-LBD and individual analogues was
attempted but was only successful with the partial agonist 8a
and the full agonist 9a. Failure to crystallize the VDR-LBD/
antagonist complex is likely due to the lability of the complex.
The crystal structures of the 8a/VDR-LBD and 9a/VDR-LBD
complexes gave us a structural basis for understanding why 8a
and 9a act as a partial and full agonist, respectively.
In crystals of the VDR-LBD/8a complex, 8a binds to VDR
with two conformations (Figure 4a−c). As mentioned in our
previous paper, no crystal structure of the VDR-LBD/
antagonist complex, which would clearly explain the structural
basis of the antagonism, has been reported. Indeed, although
compound 2 is an antagonist, in the crystal structure of the 2/
VDR-LBD complex, the ligand is bound to the active
conformation of VDR.13 Considering the action of compound
2 as a VDR antagonist, the active conformation of the VDR-
4358
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
Figure 7. B-factor of the crystal structures of VDR-LBD complexed with 1, 8a, and 9a. Green and yellow lines show the B-factor of the protein
accommodating agonist, 1 and 9a, respectively. Blue and black lines show B-factor of the protein accommodating the partial agonist 8a in crystals A
and B, respectively. Relative value of B-factor indicates a value when the B-factor of K236 is defined as 1.00.
LBD/2/peptide complex seems to be a temporary and less
preferred structure but was selected during the crystallization
process. The same interpretation, i.e., a temporary and less
preferred structure selected during crystallization, is also
applicable to the major conformer of the 8a/VDR-LBD
complex, since there is no active conformation network
between the side chain of 8a and the C-terminus (helix 11/
12) (Figure 4d and Figure 5a). Therefore, we are terming the
major conformer the antagonist binding conformation. The
minor conformer of the VDR-LBD/8a complex is the agonist
binding form of VDR, in which intimate interactions between
the ligand and the protein terminus are observed via a hydrogen
bond with His393 (Figure 4d, Figure 5b). The major and minor
conformers must contribute to the antagonistic and agonistic
activities, respectively. We concluded that 22R-butyl analogue
8a in living cells shows partial agonist activity reflecting the sum
of antagonistic and agonistic activities caused by the major and
minor conformers, respectively. Similar structural elucidations
regarding partial agonism against nuclear receptor have been
reported where distinct crystal structures with different binding
modes were observed in separate crystals but not in a single
crystal.22 Pochetti et al. reported different binding modes
between PPARγ and the partial agonistic ligand observed in
separate crystals, which were generated using different
crystallization procedures.22a Nettles et al. reported that
coupling of receptor conformation and ligand orientation
determines the partial activity by investigating estrogen
receptor (ER) and a partial agonist.22b They found distinct
ER conformations and distinct ligand orientation in different
crystals of wild type and mutated ER. We emphasize that the
different binding modes reported by Pochetti et al. and Nettles
et al. were observed in different (separate) crystals, whereas in
our study different binding modes (two conformations) were
detected in a single crystal. Thus, this is the first report showing
the trapping of a conformational subset of a ligand and the
nuclear receptor in a single crystal.
group, which was observed in VDR-LBD complexed with 2, 3,
and 9a (Figure 1b, Figure 1c, and Figure 4j). This pocket is
generated by repulsion between the butyl group and the side
chain of Leu305 at the beginning of helix 7 (Figure 6a). The
second type of pocket is induced by the original side chain of
the ligand, as observed in the complex with GEMINI 4 which is
an analogue with two identical side chains23 (Figure 1d) and 8a
(Figure 4d and Figure 4h). The second type of pocket is also
generated by repulsion of the original side chain of the ligand
(Figure 6b). These demonstrated that ligand-specific reposi-
tioning of the side chain of Leu305 triggers the formation of the
butyl pocket. It is reasonable that the second type of pocket is
larger than the first type because the original side chain of
GEMINI and 8a is larger than the butyl group. Interestingly, in
the VDR-LBD/ligand complex that forms the second type of
pocket, a helix break occurs at the beginning of helix 7 because
of the disappearance of the hydrogen bond between the
carbonyl of Leu305 and the NH of Leu309 to form the α-helix
(Figure 6d). Figure 6b clearly indicates that the disappearance
of the hydrogen bond causes the obvious shift of the protein
main chain. On the other hand, in the complex forming the first
type of pocket, no helix break was observed (Figure 6c). By an
investigation into crystal structures of other nuclear receptors,
we found similar pocket formation at the same position in
PPARγ24 and ERα.25 Taken together, it appears that the region
surrounded by the N-terminal of helix 7 and helix 11 in nuclear
receptors exhibits some flexibility without loss of binding
affinity for the ligand.
The formation of a new channel at the top of the butyl
pocket was observed in the VDR-LBD/8a complex but not in
the VDR-LBD/GEMINI complex (Figure 4d, Figure 4h, Figure
1d). The appearance of this new channel was elucidated from
the main-chain shift of Glu392, whose shift is caused by the
ligand-specific and drastic repositioning of the side chain of
Leu305 as shown in Figure 6e.
Figure 7 shows the relative values of B-factors for the crystal
structure of VDR-LBD accommodating hormone 1, partial
agonist 8a, and full agonist 9a. VDR-LBD complexed with full
agonist 1 and 9a demonstrated a relatively stable structure,
while the VDR-LBD/8a complex showed a rather unstable
structure, especially after helix 10. These results indicate that
helix 11 of the VDR-LBD/antagonist complex, even in a crystal,
is unstable because of the lack of intimate interactions between
the ligand and the C-terminus (helix 11/12) of the protein via
the hydrogen bond with His393. This observation corroborates
a report of the solution structure of VDR-LBD determined
using NMR by Markley, in which the residues between Lys395
The VDR-LBD/9a complex is the agonist binding form of
VDR, in which intimate interactions between the protein
terminus and the ligand via a hydrogen bond with His393 are
observed (Figure 4j, Figure 5f). Interestingly, the whole
molecule of 22R-butyl analogue 9a occupies the same spatial
region as that of 22S-butyl compound 3; this is accomplished
by shifting the C-20, -21, and -22 to the C-terminus of the
protein compared to 3. This structure coincides well with the
strong agonistic activity of 9a.
In our study, two types of butyl pocket were observed. The
first type of pocket is formed by an induced fit with the butyl
4359
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
(3R)-3-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-
heptan-1-ol (12a). Ester 11a (837.9 mg, 1.19 mmol) in dry THF (15
mL) was treated with DIBAL (1 M in toluene solution, 3.58 mL, 3.58
mmol) at 0 °C for 1.5 h. The reaction was quenched with 1 N HCl
and was extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (35 g) with 5% AcOEt/hexane to
and Glu421 of VDR-LBD complexed with antagonist were
demonstrated to be disordered.26
CONCLUSIONS
■
We observed a mixed population of antagonist and agonist
binding conformers in a single crystal. This observation
indicates that the partial agonist activity of 22R-butyl analogue
8a depends on the sum of antagonistic and agonistic activities
caused by antagonist and agonist binding conformers,
respectively. Combined with our previous study, we draw the
following conclusions regarding VDR-agonism, partial agonism,
and antagonism. Intimate interactions between the ligand and
the C-terminus (helix 11/12) of VDR via a hydrogen bond with
His393 overcome the structural strain in VDR caused by the
formation of butyl pocket, resulting in the ligand acting as a
VDR agonist. However, the ligand acts as a VDR antagonist
when it interacts weakly with the C-terminus (helix 11/12) of
VDR. Partial agonistic activity is dependent on the sum of
agonistic behavior and antagonistic behavior of the ligand and
VDR. The structural basis of VDR partial agonism observed
here must be applicable to the partial agonism of other ligand-
dependent nuclear receptors, such as steroid hormone
receptors, retinoid hormone receptors, and so on.
1
afford 12a (749.1 mg, 95%). H NMR δ 0.03, 0.05, 0.07, 0.08 (each
3H, s, SiMe), 0.56 (3H, s, H-18), 0.80 (3H, d, J = 6.3 Hz, H-21), 0.87,
0.90 (each 9H, s, t-Bu), 2.83 (1H, m, H-9), 3.67 (2H, m, H-24), 4.42
(2H, m, H-3, 1), 4.93 (1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84
(1H, d, J = 11.1 Hz, H-7), 6.22 (1H, d, J = 11.2 Hz, H-6). 13C NMR δ
−5.1, −4.92, −4.86 (2 carbons), 12.0, 13.0, 14.1, 18.2, 18.3, 22.1, 23.1,
23.4, 25.78 (3 carbons), 25.84 (3 carbons), 27.7, 28.8, 30.1, 32.1, 32.5,
36.2, 37.2, 38.6, 40.7, 45.7, 47.6, 53.8, 56.4, 62.4, 71.7, 72.5, 106.2,
116.1, 122.4, 132.8, 141.2, 153.0. MS (ESI) m/z 681.6 [(M + Na)+,
35]. HRMS (ESI) calcd for C40H74NaO3Si2 (M + Na)+ 681.5074,
found 681.5118. IR (neat) 3298, 2955, 2928, 2893, 2858, 1472, 1462,
1256, 1101, 1070, 935, 897, 835, 775, 667 cm−1. UV (hexane) λmax
246, 255, 265 nm.
22R-Butyl-2-methylidene-19,25,26,27-tetranor-1α,24-dihy-
droxyvitamin D3 (5a). To a solution of 12a (2.9 mg, 0.0047 mmol)
in MeOH (200 μL) was added camphor sulfonic acid (4.7 mg, 0.0202
mmol) at 0 °C, and the mixture was stirred at room temperature for 2
h. The reaction was quenched with saturated NaHCO3 aqueous
solution and extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (5 g) with 60% AcOEt/hexane to
afford 5a (1.7 mg, 89%). 1H NMR δ 0.57 (3H, s, H-18), 0.79 (3H, d, J
= 6.2 Hz, H-21), 0.90 (3H, t, J = 6.6, 6.8 Hz, -(CH2)3-CH3), 3.60 (1H,
m, H-24), 3.73 (1H, m, H-24), 4.48 (2H, m, H-1, 3), 5.09, 5.11 (each
1H, s, -CCH2), 5.88 (1H, d, J = 11.3 Hz, H-7), 6.36 (1H, d, J = 11.3
Hz, H-6). MS (EI) m/z 430.4 (M+, 15), 315.3 (11), 297.3 (12), 269.3
(9), 251.2 (10), 185.2 (11), 69.1 (100). HRMS (EI) calcd for
C28H46O3 (M+) 430.3447, found 430.3436. IR (neat) 3325, 2949,
2926, 2858, 1659, 1456, 1072, 1049 cm−1. UV (EtOH) λmax 245, 254,
263 nm.
EXPERIMENTAL SECTION
All reagents were purchased from commercial sources. Unless
otherwise stated, NMR spectra were recorded at 300 MHz for H
■
1
NMR and 75 MHz for 13C NMR in CDCl3 solution with TMS as an
internal standard and the chemical shifts are given in δ values. High
and low resolution mass spectra were obtained with a JEOL JMS D-
300, JEOL AccuTOF LC-plus JMS-T100LP, and JEOL JMS-HX110A
spectrometers. Relative intensities are given in parentheses in low
mass. IR spectra were recorded on a Shimadzu FTIR-8400S
spectrophotometer, and data are given in cm−1. UV spectra were
recorded on a Beckman DU7500 spectrophotometer. All air and
moisture sensitive reactions were carried out under argon or nitrogen
atmosphere. Purity was determined by HPLC [Pegasil silica SP100, 4.6
mm × 150 mm, hexane/CHCl3/MeOH (100:25:8), flow rate 1.0 mL/
min] and was >95% for all compounds tested.
(4R)-4-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-2-
methyloctan-2-ol (13a). To a solution of ester 11a (57.5 mg, 0.082
mmol) in dry THF (1 mL) at −78 °C was added MeLi (452 μL, 0.49
mmol), and the mixture was heated to −50 °C for 1 h. The reaction
was quenched with saturated NH4Cl aqueous solution, and the
mixture was extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (5 g) with 5% AcOEt/hexane to afford
Ethyl (3R)-3-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)-
silyl]oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-
heptanoate (11a). A suspension of CuBr/Me2S (2.44 g, 11.86
mmol) in dry THF (10 mL) was cooled to −30 °C, and to this
solution was added n-BuLi (1.62 M in hexanes, 14.64 mL, 23.71
mmol), and the mixture was stirred for 15 min. To this solution was
added TMSCl (1.90 mL, 14.82 mmol), HMPA (2.58 mL, 14.82
mmol), and a solution of enoate 10 (953 mg, 1.48 mmol) in dry THF
(10 mL) in this order. The mixture was stirred at −30 °C for 1.5 h,
and the reaction was quenched with 1 N HCl. The mixture was
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (200 g, 40−50 μm) with 0.5% AcOEt/hexane to afford 11a
1
13a (44.1 mg, 78%). H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s,
SiMe), 0.58 (3H, s, H-18), 0.79 (3H, d, J = 6.6 Hz, H-21), 0.87, 0.89
(each 9H, s, t-Bu), 1.23 (6H, s, H-25, 26), 2.82 (1H, m, H-9), 4.43 (H,
m, H-1, 3), 4.92, 4.97 (each 1H, s,-CCH2), 5.84 (1H, d, J = 11.1 Hz,
H-7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ −5.1, −4.9, −4.88,
−4.85, 12.1, 12.9, 14.2, 18.2, 18.2, 22.3, 23.1, 23.4, 25.79 (3 carbons),
25.84 (3 carbons), 28.1, 28.8, 30.2, 30.3, 30.5, 33.4, 36.0, 37.6, 38.6,
40.6, 43.0, 45.7, 47.6, 53.5, 56.4, 71.7 (2 carbons), 72.5, 106.2, 116.2,
122.4, 132.8, 141.2, 153.0. MS (EI) m/z 686.5 (M+, 2), 554.4 (18),
536.3 (13), 366.1 (24), 73.0 (100). HRMS (EI) calcd for C42H78O3Si2
(M+) 686.5490, found 686.5483. IR (neat) 2955, 2928, 2856, 1462,
1373, 1256, 1101, 1072, 935, 896, 835, 775 cm−1. UV (hexane) λmax
246, 255, 264 nm.
22R-Butyl-2-methylidene-19,24-dinor-1α,25-dihydroxyvita-
min D3 (6a). To a solution of 13a (21.6 mg, 0.0315 mmol) in MeOH
(200 μL) was added camphor sulfonic acid (24.6 mg, 0.106 mmol) at
0 °C, and the mixture was stirred at room temperature for 1 h. The
reaction was quenched with saturated NaHCO3 aqueous solution and
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (10 g) with 40% AcOEt/hexane to afford 6a (13.8 mg, 96%).
1H NMR δ 0.58 (3H, s, H-18), 0.78 (3H, d, J = 6.6 Hz, H-21), 1.24
(845 mg, 81%) and its 22S-isomer15 (60.0 mg, 6%). 11a: H NMR δ
1
0.03, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.56 (3H, s, H-18), 0.79 (3H,
d, J = 6.7 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu), 1.26 (3H, t, J = 7.1
Hz, -OCH2CH3), 2.83 (1H, m, H-9), 4.13 (2H, q, J = 7.2,
-OCH2CH3), 4.42 (2H, m, H-3, 1), 4.93 (1H, s, CCH2), 4.97
(1H, s, CCH2), 5.84 (1H, d, J = 11.1 Hz, H-7), 6.22 (1H, d, J = 11.1
Hz, H-6). 13C NMR δ −5.1, −4.92, −4.87 (2 carbons), 12.0, 13.1,
14.1, 14.3, 18.2, 18.3, 22.1, 22.9, 23.4, 25.78 (3 carbons), 25.84 (3
carbons), 27.5, 28.7, 29.6, 32.7, 34.9, 37.1, 37.3, 38.6, 40.7, 45.7, 47.6,
54.1, 56.3, 60.1, 71.6, 72.5, 106.3, 116.2, 122.4, 132.9, 141.0, 153.0,
174.5. MS (FAB+) m/z 701.7 [(M + H)+, 8], 569.5 (26). HRMS
(FAB+) calcd for C42H77O4Si2 (M + H)+ 701.5360, found 701.5358.
IR (neat) 2955, 2928, 2893, 2856, 1738, 1472, 1462, 1377, 1326, 1256,
1166, 1101, 1070, 935, 896, 835, 775 cm−1. UV (hexane) λmax 245,
253, 263 nm.
(6H, s, H-25, 26), 4.48 (2H, m, H-1, 3), 5.08, 5.10 (each 1H, s, -C
4360
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
CH2), 5.88 (1H, d, J = 11.2 Hz, H-7), 6.35 (1H, d, J = 11.2 Hz, H-6).
13C NMR δ 12.1, 12.9, 14.2, 22.4, 23.1, 23.5, 28.1, 29.0, 30.2, 30.3,
23.4, 25.78 (3 carbons), 25.83 (3 carbons), 27.5, 28.6, 28.7, 29.8, 31.5,
35.9, 36.8, 38.6, 40.6, 45.6, 47.6, 53.7, 56.2, 70.0, 71.7, 72.4, 106.3,
116.2, 122.4, 127.9 (2 carbons), 129.8 (2 carbons), 132.9, 133.4, 140.1,
144.6, 153.0. MS (ESI) m/z 835.5 [(M + Na)+, 95]. HRMS (ESI)
calcd for C47H80NaO5SSi2 (M + Na)+ 835.5163, found 835.5173. IR
(neat) 2955, 2928, 2893, 2856, 1472, 1462, 1362, 1252, 1188, 1178,
1099, 1070, 937, 897, 835, 775, 663 cm−1. UV (hexane) λmax 246, 255,
265 nm.
30.4, 33.4, 36.0, 37.5, 38.2, 40.5, 43.0, 45.8, 45.9, 53.5, 56.5, 70.7, 71.7,
71.9, 107.7, 115.3, 124.3, 130.4, 143.4, 152.0. MS (EI) m/z 458.6 (M+,
9), 368.5 (10), 285.3 (15), 161.3 (22), 147.3 (21), 135.3 (39), 69.2
(100). HRMS (EI) calcd for C30H50O3 (M+) 458.3760, found
458.3762. IR (neat) 3356, 2955, 2928, 2868, 1468, 1377, 1074,
1045, 758 cm−1. UV (EtOH) λmax 246, 254, 263 nm.
(4R)-4-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-
octanenitrile (16a). A solution of tosylate 15a (773 mg, 0.95 mmol)
and KCN (173.3 mg, 2.66 mmol) in dry DMSO (24 mL) was stirred
at 70 °C for 1 h. The reaction was quenched with ice−water, and the
mixture was extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (35 g) with 5% AcOEt/hexane to
(5R)-4-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-3-
ethylnonan-3-ol (14a). To a solution of ester 11a (46.9 mg, 0.067
mmol) in dry THF (1.2 mL) was added EtMgBr (1.0 M in THF
solution, 201 μL, 0.20 mmol), and the mixture was stirred at room
temperature. After 2.5 h, to the mixture was added an additional
EtMgBr (1.0 M in THF solution, 134 μL, 0.13 mmol), and the mixture
was stirred. In addition, after 2 h, to the mixture was added an
additional EtMgBr (1.0 M in THF solution, 134 μL, 0.13 mmol), and
the mixture was stirred for 1.5 h. The reaction was quenched with
saturated NH4Cl aqueous solution, and the mixture was extracted with
AcOEt. The organic layer was washed with brine, dried over MgSO4,
and evaporated. The residue was chromatographed on silica gel (5 g)
1
afford 16a (591.2 mg, 93%). H NMR δ 0.03, 0.05, 0.07, 0.08 (each
3H, s, SiMe), 0.56 (3H, s, H-18), 0.79 (3H, d, J = 6.7 Hz, H-21), 0.87,
0.90 (each 9H, s, t-Bu), 2.83 (1H, m, H-9), 4.43 (2H, m, H-3, 1), 4.93
(1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84 (1H, d, J = 11.1 Hz, H-
7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ −5.1, −4.92, −4.87,
−4.86, 12.0, 12.9, 14.1, 16.0, 18.2, 18.3, 22.1, 23.0, 23.4, 25.3, 25.78 (3
carbons), 25.84 (3 carbons), 27.7, 28.7, 29.8, 30.8, 36.9, 38.6, 39.2,
40.7, 45.7, 47.6, 53.8, 56.3, 71.7, 72.4, 106.3, 116.3, 120.2, 122.4, 132.9,
140.9, 153.0. MS (ESI) m/z 690.5 [(M + Na)+, 100]. HRMS (ESI)
calcd for C41H73NNaO2Si2 (M + Na)+ 690.5078, found 690.5071. IR
(neat) 2955, 2929, 2893, 2856, 2245, 1472, 1462, 1252, 1101, 1072,
935, 897, 835, 775 cm−1. UV (hexane) λmax 246, 255, 264 nm.
(4R)-4-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-
octanal (17a). Cyanide 16a (36.2 mg, 0.054 mmol) in dry THF (1
mL) was treated with DIBAL (1 M in toluene solution, 70 μL, 0.070
mmol) at −20 °C. After 1.5 h, to the mixture was added an additional
DIBAL (1 M in toluene solution, 54 μL, 0.054 mmol), and the mixture
was was stirred at 0 °C for 1.5 h. The reaction was quenched with 1 N
HCl and extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (4 g) with 2% AcOEt/hexane to afford
1
with 5% AcOEt/hexane to afford 14a (41.4 mg, 86%). H NMR δ
0.04, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.53 (3H, s, H-18), 0.79 (3H,
d, J = 6.7 Hz, H-21), 0.87, 0.89 (each 9H, s, t-Bu), 2.82 (1H, m, H-9),
4.42 (2H, m, H-3, 1), 4.93 (1H, s, CCH2), 4.97 (1H, s, CCH2),
5.83 (1H, d, J = 11.1 Hz, H-7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C
NMR δ −5.1, −4.92, −4.89, −4.86, 8.0, 8.1, 12.0, 13.0, 14.2, 18.18,
18.24, 22.3, 23.4, 25.78 (3 carbons), 25.83 (3 carbons), 28.2, 28.8,
30.3, 31.0, 31.5, 33.2, 34.9, 37.7, 38.0, 38.6, 40.6, 45.7, 47.5, 53.6, 56.4,
71.7, 72.4, 75.3, 106.2, 116.1, 122.4, 132.7, 141.2, 153.0. MS (FAB+)
m/z 715.6 [(M + H)+, 3)], 583.5 (14). HRMS (FAB+) calcd for
C44H83O3Si2 (M + H)+ 715.5881, found 715.5893. IR (neat) 3501,
2955, 2928, 2856, 1732, 1661, 1620, 1462, 1254, 1101, 1072, 835, 775,
667 cm−1. UV (hexane) λmax 246, 255, 265 nm.
22R-Butyl-2-methylidene-26,27-dimethyl-19,24-dinor-
1α,25-dihydroxyvitamin D3 (9a). A solution of 14a (29 mg, 0.041
mmol) in dry MeOH (0.725 mL) was added camphor sulfonic acid
(28.3 mg, 0.122 mmol) at 0 °C, and the mixture was stirred at room
temperature for 1 h. The reaction was quenched with saturated
NaHCO3 aqueous solution and extracted with AcOEt. The organic
layer was washed with brine, dried over MgSO4, and evaporated. The
residue was chromatographed on silica gel (10 g) with 50% AcOEt/
1
17a (22.4 mg, 62%). H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s,
SiMe), 0.56 (3H, s, H-18), 0.81 (3H, d, J = 6.1 Hz, H-21), 0.87, 0.90
(each 9H, s, t-Bu), 2.83 (1H, m, H-9), 4.42 (2H, m, H-3, 1), 4.93 (1H,
s, CCH2), 4.97 (1H, s, CCH2), 5.84 (1H, d, J = 11.1 Hz, H-7),
6.22 (1H, d, J = 11.1 Hz, H-6), 9.78 (1H, t, J = 1.9 Hz, −CHO). 13C
NMR δ −5.1, −4.92, −4.88, −4.86, 12.0, 12.9, 14.1, 18.17, 18.24, 21.3,
22.1, 23.0, 23.4, 25.78 (3 carbons), 25.83 (3 carbons), 27.7, 28.8, 30.0,
31.2, 37.1, 38.6, 39.5, 40.7, 43.0, 45.7, 47.6, 53.8, 56.3, 71.7, 72.5,
106.2, 116.2, 122.4, 132.8, 141.1, 153.0, 203.2. MS (ESI) m/z 693.5
[(M + Na)+, 55]. HRMS (ESI) calcd for C41H74O3Si2 (M+) 670.5177,
found 670.5190. IR (neat) 2955, 2928, 2893, 2856, 2710, 1728, 1472,
1462, 1252, 1101, 1070, 935, 897, 835, 775 cm−1. UV (hexane) λmax
246, 255, 265 nm.
(4R)-4-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}octan-
1-ol (18a). To a solution of aldehyde 17a (14.3 mg, 0.021 mmol) in
dry MeOH/dry CH2Cl2 (1:1, 2 mL) was added NaBH4 (12.89 mg,
0.34 mmol) at 0 °C, and the mixture was stirred at room temperature
for 1.5 h. To the solution was added ice−water, and the mixture was
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (4 g) with 15% AcOEt/hexane to afford 18a (14.5 mg,
quant). 1H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.56 (3H,
s, H-18), 0.78 (3H, d, J = 5.9 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu),
2.83 (1H, m, H-9), 3.64 (2H, t, J = 6.6 Hz, H-25), 4.43 (2H, m, H-3,
1), 4.93 (1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84 (1H, d, J =
11.1 Hz, H-7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ −5.1,
−4.92, −4.88, −4.86, 12.0, 13.0, 14.2, 18.2, 18.3, 22.2, 23.1, 23.5, 25.0,
25.78 (3 carbons), 25.84 (3 carbons), 27.7, 28.8, 30.2, 31.7, 31.9, 37.2,
38.6, 39.8, 40.7, 45.7, 47.6, 53.9, 56.3, 63.6, 71.7, 72.5, 106.2, 116.1,
122.4, 132.8, 141.2, 153.0. MS (FAB+) m/z 673.6 [(M + H)+, 1],
541.6 (4). HRMS (FAB+) calcd for C41H77O3Si2 (M + H)+ 673.5411,
1
hexane to afford 9a (19.1 mg, 97%). H NMR δ 0.58 (3H, s, H-18),
0.79 (3H, d, J = 6.7 Hz, H-21), 0.84−0.92 (9H, m, CH2CH3 × 3), 2.83
(1H, m, H-9), 4.47 (2H, m, H-3, 1), 5.09 (1H, s, CCH2), 5.11 (1H,
s, CCH2), 5.89 (1H, d, J = 11.2 Hz, H-7), 6.36 (1H, d, J = 11.2 Hz,
H-6). 13C NMR δ 8.0, 8.1, 12.1, 13.0,14.2, 22.4, 23.1, 23.5, 28.1, 29.0,
30.3, 31.1, 31.4, 33.2, 34.9, 37.6, 38.0, 38.1, 40.5, 45.75, 45.85, 53.5,
56.4, 70.6, 71.8, 75.3, 107.7, 115.3, 124.2, 130.4, 143.4, 152.0. MS
(ESI) m/z 509.4 [(M + Na)+, 100]. HRMS (ESI) calcd for
C32H54NaO3 (M + Na)+ 509.3971, found 509.3981. IR (neat) 3346,
2959, 2932, 2874, 1717, 1651, 1614, 1456, 1377, 1078, 1041, 914, 758,
667 cm−1. UV (EtOH) λmax 246, 254, 264 nm.
(3R)-3-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}heptyl
4-Methylbenzenesulfonate (15a). To a solution of alcohol 11a
(47.9 mg, 0.073 mmol) in dry pyridine (1 mL) was added p-
toluenesulfonyl chloride (20.78 mg, 0.11 mmol) at 0 °C, and the
mixture was stirred for 15 h. To this solution was added ice−water and
1 N HCl, and the mixture was extracted with AcOEt. The organic layer
was washed with saturated NaHCO3 aqueous solution and brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (5 g) with 5% AcOEt/hexane to afford 15a (55.1 mg, 93%).
1H NMR δ 0.04, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.53 (3H, s, H-
18), 0.72 (3H, d, J = 6.7 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu), 2.45
(3H, s, Ph-Me), 2.83 (1H, m, H-9), 4.07 (2H, m, H-24), 4.43 (2H, m,
H-3, 1), 4.93 (1H, s, CCH2), 4.97 (1H, s, CCH2), 5.83 (1H, d, J
= 11.1 Hz, H-7), 6.21 (1H, d, J = 11.1 Hz, H-6), 7.35 (2H, d, J = 8.5
Hz, arom-H), 7.80 (2H, d, J = 8.3 Hz, arom-H). 13C NMR δ −5.1,
−4.92, −4.88, −4.86, 12.0, 12.8, 14.1, 18.17, 18.24, 21.6, 22.1, 22.9,
4361
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
found 673.5424. IR (neat) 3331, 2953, 2928, 2895, 2856, 1655, 1618,
1472, 1462, 1256, 1101, 1070, 935, 897, 835, 775, 667 cm−1. UV
(hexane) λmax 246, 255, 265 nm.
0 °C, and the mixture was stirred at room temperature for 1 h. The
reaction was quenched with saturated NaHCO3 aqueous solution and
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (10 g) with 50% AcOEt/hexane to afford 8a (13.0 mg, 97%).
1H NMR δ 0.56 (3H, s, H-18), 0.79 (3H, d, J = 6.0 Hz, H-21), 0.90
(3H, m, CH2CH3), 1.22, 1.23 (each 3H, s, H-26, 27), 2.58 (1H, dd, J =
13.2, 3.8 Hz, H-14), 2.84 (2H, m, H-9), 4.48 (2H, m, H-3, 1), 5.09
(1H, s, CCH2), 5.11 (1H, s, CCH2), 5.89 (1H, d, J = 11.3 Hz, H-
7), 6.37 (1H, d, J = 11.2 Hz, H-6). 13C NMR δ 12.1, 13.2, 14.2, 22.2,
23.1, 23.3, 23.5, 27.7, 29.0, 29.1, 29.5, 30.2, 31.6, 37.1, 38.2, 40.4, 40.5,
43.1, 45.8, 45.9, 54.0, 56.4, 70.7, 71.3, 71.9, 107.7, 115.3, 124.3, 130.4,
143.5, 152.0. MS (ESI) m/z 495.4 [(M + Na)+, 100]. HRMS (ESI)
calcd for C31H52NaO3 (M + Na)+ 495.3814, found 495.3820. IR
(neat) 3379, 2957, 2926, 2856, 1456, 1379, 1074, 1047, 667 cm−1. UV
(EtOH) λmax 246, 254, 264 nm.
22R-Butyl-2-methylidene-19,26,27-trinor-1α,25-dihydroxy-
vitamin D3 (7a). To a solution of 18a (29.0 mg, 0.043 mmol) in dry
MeOH (0.5 mL) was added camphor sulfonic acid (30.0 mg, 0.13
mmol) at 0 °C, and the mixture was stirred at room temperature for 1
h. The reaction was quenched with saturated NaHCO3 aqueous
solution and extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (10 g) with 50% AcOEt/hexane to
afford 7a (18.3 mg, 96%). 1H NMR δ 0.56 (3H, s, H-18), 0.78 (3H, d,
J = 5.9 Hz, H-21), 0.90 (3H, m, CH2CH3), 2.57 (1H, dd, J = 13.3, 3.7
Hz, H-14), 2.84 (2H, m, H-9), 3.64 (2H, t, J = 6.6 Hz, H-25), 4.48
(2H, m, H-3, 1), 5.09 (1H, s, CCH2), 5.11 (1H, s, CCH2), 5.89
(1H, d, J = 11.3 Hz, H-7), 6.36 (1H, d, J = 11.2 Hz, H-6). 13C NMR δ
12.0 13.0, 14.2, 22.2, 23.1, 23.5, 25.0, 27.7, 29.0, 30.2, 31.6, 31.8, 37.1,
38.2, 39.7, 40.5, 45.77, 45.84, 53.9, 56.4, 63.6, 70.7, 71.8, 107.7, 115.3,
124.2, 130.4, 143.4, 152.0. MS (ESI) m/z 467.4 [(M + Na)+, 65].
HRMS (ESI) calcd for C29H48NaO3 (M + Na)+ 467.3501, found
467.3510. IR (neat) 3379, 2928, 2866, 1462, 1379, 1074, 1047, 667
cm−1. UV (EtOH) λmax 246, 254, 264 nm.
Methyl (4R)-4-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl-
(dimethyl)silyl]oxy}-2-methylidene-9,10-secoestra-5,7-dien-
17-yl]ethyl}octanoate (19a). To a solution of aldehyde 17a (14.3
mg, 0.021 mmol) in dry MeOH (10 mL) were added KOH (7.2 mg,
0.11 mmol) and then iodine (19.6 mg, 0.055 mmol) at 0 °C, and the
mixture was stirred for 1 h. The reaction was quenched with 10%
Na2SO3, and the mixture was extracted with AcOEt. The organic layer
was washed with brine, dried over MgSO4, and evaporated. The
residue was chromatographed on silica gel (10 g) with 5% AcOEt/
hexane to afford 19a (10.5 mg, 70%). 1H NMR δ 0.03, 0.05, 0.07, 0.08
(each 3H, s, SiMe), 0.55 (3H, s, H-18), 0.79 (3H, d, J = 6.0 Hz, H-21),
0.87, 0.90 (each 9H, s, t-Bu), 2.83 (1H, m, H-9), 3.68 (3H, s, -OCH3),
4.42 (2H, m, H-3, 1), 4.93 (1H, s, CCH2), 4.97 (1H, s, CCH2),
5.83 (1H, d, J = 11.1 Hz, H-7), 6.21 (1H, d, J = 11.1 Hz, H-6). 13C
NMR δ −5.1, −4.92, −4.89, −4.86, 11.9, 12.9, 14.1, 18.17, 18.24, 22.1,
23.0, 23.4, 24.5, 25.78 (3 carbons), 25.83 (3 carbons), 27.7, 28.8, 29.7,
30.0, 31.2, 33.1, 37.2, 38.6, 39.5, 40.6, 45.7, 47.6, 51.5, 53.7, 56.3, 71.7,
72.5, 106.2, 116.1, 122.4, 132.8, 141.2, 153.0, 174.6. MS (FAB+) m/z
701.6 [(M + H)+, 6], 569.5 (24). HRMS (FAB+) calcd for
C42H77O4Si2 (M + H)+ 701.5360, found 701.5377. IR (neat) 2953,
2928, 2895, 2856, 1742, 1466, 1458, 1256, 1101, 1070, 935, 897, 835,
775, 667 cm−1. UV (hexane) λmax 246, 255, 265 nm.
Ethyl (3R)-4-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)-
silyl]oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]-3-eth-
ylpentanoate (11b). A suspension of CuBr/Me2S (444.1 mg, 2.12
mmol) in dry THF (4 mL) was cooled to −30 °C, and to this solution
was added EtLi (8.50 mL, 4.25 mmol, 0.5 M solution benzene/
cyclohexane = 90/10). The mixture was stirred for 15 min. To this
solution were added TMSCl (338 μL, 2.65 mmol), HMPA (463 μL,
2.65 mmol), and a solution of enoate 10 (170.7 mg, 0.27 mmol) in dry
THF (2 mL) in this order. The mixture was stirred at −30 °C for 1.5
h, and the reaction was quenched with 1 N HCl. The mixture was
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (60 g, 40−50 μm) with 0.3% AcOEt/hexane to afford 11b
(130 mg, 73%) and its 22S-isomer15 (7 mg, 4%). H NMR δ 0.03,
1
0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.57 (3H, s, H-18), 0.79 (3H, d, J
= 6.7 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu), 1.26 (6H, t, J = 7.1 Hz,
-CH2CH3 × 2), 2.83 (1H, m, H-9), 4.13 (2H, m, -OCH2CH3), 4.43
(2H, m, H-3, 1), 4.92 (1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84
(1H, d, J = 11.1 Hz, H-7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ
−5.1, −4.9, −4.8 (2 carbons), 12.0, 12.1, 13.0, 14.3, 18.2, 18.3, 22.1,
23.4, 25.7, 25.77 (3 carbons), 25.84 (3 carbons), 27.6, 28.7, 34.7, 36.6,
38.6, 39.2, 40.7, 45.7, 47.6, 54.0, 56.4, 60.1, 71.6, 72.5, 106.3, 116.2,
122.4, 132.9, 141.0, 153.0, 174.5. MS (FAB+) m/z 673.5 [(M + H)+,
4], 541.5 (30). HRMS (FAB+) calcd for C40H73O4Si2 (M + H)+
673.5047, found 673.5033. IR (neat) 2955, 2928, 2895, 2856, 1738,
1655, 1618, 1462, 1256, 1101, 1072, 935, 897, 835, 775, 667 cm−1. UV
(hexane) λmax 246, 255, 265 nm.
(3R)-4-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]oxy}-
2-methylidene-9,10-secoestra-5,7-dien-17-yl]-3-ethylpentan-
1-ol (12b). Ester 11b (183.6 mg, 0.27 mmol) in dry THF (2 mL) was
treated with DIBAL (1 M in toluene solution, 409 μL, 0.41 mmol) at 0
°C. After 1 h, to the mixture was added an additional DIBAL (1 M in
toluene solution, 409 μL, 0.41 mmol), and the mixture was stirred at 0
°C for 30 min. The reaction was quenched with 1 N HCl and was
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (5 g) with 10% AcOEt/hexane to afford 12b (160.0 mg,
93%). 1H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.57 (3H,
s, H-18), 0.79 (3H, d, J = 6.5 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu),
2.83 (1H, m, H-9), 3.67 (2H, m, H-24), 4.42 (2H, m, H-3, 1), 4.92
(1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84 (1H, d, J = 11.2 Hz, H-
7), 6.22 (1H, d, J = 11.2 Hz, H-6). 13C NMR δ −5.1, −4.92, −4.86 (2
carbons), 12.0, 12.5, 12.9, 18.2, 18.3, 22.1, 23.4, 24.9, 25.78 (3
carbons), 25.84 (3 carbons), 27.8, 28.8, 32.4, 36.4, 38.2, 38.6, 40.7,
45.7, 47.6, 53.7, 56.4, 62.4, 71.6, 72.5, 106.2, 116.2, 122.4, 132.8, 141.1,
153.0. MS (FAB+) m/z 631.5 [(M + H)+, 2], 499.4 (22). HRMS
(FAB+) calcd for C38H71O3Si2 (M + H)+ 631.4942, found 631.4931.
IR (neat) 3306, 2955, 2930, 2895, 2856, 1732, 1661, 1620, 1472, 1256,
1101, 1070, 1005, 935, 897, 835, 775, 667 cm−1. UV (hexane) λmax
246, 255, 265 nm.
(5R)-5-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]-
oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-2-
methylnonan-2-ol (20a). To a solution of ester 19a (29.3 mg, 0.042
mmol) in dry THF (1 mL) at −78 °C was added MeLi (1.04 M in
Et2O solution, 241 μL, 0.25 mmol), and the mixture was stirred. The
mixture was heated to −10 °C for 2 h and then stirred for 2 h. The
reaction was quenched with saturated NH4Cl aqueous solution, and
the mixture was extracted with AcOEt. The organic layer was washed
with brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (10 g) with 10% AcOEt/hexane to
afford 20a (25.3 mg, 86%). 1H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H,
s, SiMe), 0.55 (3H, s, H-18), 0.79 (3H, d, J = 5.9 Hz, H-21), 0.86, 0.90
(each 9H, s, t-Bu), 1.22, 1.23 (each 3H, s, H-26, 27), 2.83 (1H, m, H-
9), 4.42 (2H, m, H-3, 1), 4.93 (1H, s, CCH2), 4.97 (1H, s, C
CH2), 5.84 (1H, d, J = 11.2 Hz, H-7), 6.22 (1H, d, J = 11.1 Hz, H-6).
13C NMR δ −5.1, −4.91, −4.88, −4.85, 12.0, 13.2, 14.2, 18.2, 18.3,
22.2, 23.1, 23.3, 23.5, 25.78 (3 carbons), 25.84 (3 carbons), 27.7, 28.8,
29.1, 29.5, 30.2, 31.7, 37.2, 38.6, 40.5, 40.7, 43.1, 45.7, 47.6, 54.0, 56.3,
71.3, 71.7, 72.5, 106.2, 116.1, 122.4, 132.7, 141.3, 153.0. MS (ESI) m/z
701.6 [(M + H)+, 4]. HRMS (ESI) calcd for C43H81O3Si2 (M + H)+
701.5724, found 701.5725. IR (neat) 3375, 2955, 2928, 2895, 2856,
1472, 1464, 1256, 1101, 1072, 935, 908, 897, 835, 775, 667 cm−1. UV
(hexane) λmax 246, 255, 265 nm.
22R-Ethyl-2-methylidene-19,25,26,27-tetranor-1α,24-dihy-
droxyvitamin D3 (5b). To a solution of 12b (12.9 mg, 0.021 mmol)
in dry MeOH (0.37 mL) was added camphor sulfonic acid (14.2 mg,
0.061 mmol) at 0 °C, and the mixture was stirred at room temperature
22R-Butyl-2-methylidene-19-nor-1α,25-dihydroxyvitamin
D3 (8a). To a solution of 20a (20.0 mg, 0.029 mmol) in dry MeOH
(0.5 mL) was added camphor sulfonic acid (19.9 mg, 0.086 mmol) at
4362
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
for 1.5 h. The reaction was quenched with saturated NaHCO3 aqueous
solution and extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (10 g) with 40% AcOEt/hexane to
afford 5b (8.2 mg, quant). 1H NMR δ 0.57 (3H, s, H-18), 0.79 (3H, d,
J = 6.5 Hz, H-21), 0.87 (3H, m, CH2CH3), 2.83 (1H, m, H-9), 3.67
(2H, m, H-24), 4.48 (2H, m, H-1,3), 5.09, 5.11 (each 1H, s, -C
CH2), 5.89 (1H, d, J = 11.3 Hz, H-7), 6.36 (1H, d, J = 11.3 Hz, H-6).
13C NMR δ 12.0, 12.4, 12.9, 22.2, 23.5, 24.8, 27.7, 29.0, 32.4, 36.4, 38.1
(2 carbons), 40.5, 45.78, 45.83, 53.7, 56.4, 62.3, 70.7, 71.8, 107.7,
115.3, 124.2, 130.4, 143.4, 152.0. MS (ESI) m/z 425.3 [(M + Na)+,
100]. HRMS (ESI) calcd for C28H42NaO3 (M + Na)+ 425.3032, found
425.3069. IR (neat) 3352, 2957, 2872, 1722, 1657, 1616, 1456, 1380,
1072, 1047, 912, 866, 756, 667 cm−1. UV (EtOH) λmax 246, 254, 264
nm.
0.03, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.58 (3H, s, H-18), 0.79 (3H,
d, J = 6.7 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu), 2.82 (1H, m, H-9),
4.42 (2H, m, H-3, 1), 4.92 (1H, s, CCH2), 4.97 (1H, s, CCH2),
5.84 (1H, d, J = 11.2 Hz, H-7), 6.22 (1H, d, J = 11.2 Hz, H-6). 13C
NMR δ −5.1, −4.92, −4.88, −4.86, 12.0, 12.5, 12.9, 18.2, 18.2, 22.3,
23.4, 25.78 (3 carbons), 25.83 (3 carbons), 25.9, 28.2, 28.8, 31.0, 31.5,
36.87, 36.94, 38.0, 38.6, 40.6, 45.7, 47.6, 53.5, 56.4, 71.7, 72.5, 75.2,
106.2, 116.2, 122.4, 132.7, 141.2, 153.0. MS (FAB+) m/z 687.6 [(M +
H)+, 4], 555.5 (20). HRMS (FAB+) calcd for C42H79O3Si2 (M + H)+
687.5568, found 687.5562. IR (neat) 3497, 2955, 2930, 2883, 2856,
1730, 1655, 1618, 1472, 1462, 1256, 1101, 1072, 935, 897, 775, 667
cm−1. UV (hexane) λmax 246, 255, 265 nm.
22R-Ethyl-2-methylidene-26,27-dimethyl-19,24-dinor-
1α,25-dihydroxyvitamin D3 (9b). To a solution of 14b (33.3 mg,
0.049 mmol) in dry MeOH (1 mL) was added camphor sulfonic acid
(33.8 mg, 0.145 mmol) at 0 °C, and the mixture was stirred at room
temperature for 1 h. The reaction was quenched with saturated
NaHCO3 aqueous solution and extracted with AcOEt. The organic
layer was washed with brine, dried over MgSO4, and evaporated. The
residue was chromatographed on silica gel (10 g) with 40% AcOEt/
hexane to afford 9b (22.8 mg, quant). 1H NMR δ 0.59 (3H, s, H-18),
0.78 (3H, d, J = 6.7 Hz, H-21), 0.87 (9H, t, J = 7.6 Hz, CH2CH3 × 3),
2.57 (1H, dd, J = 13.3, 3.8 Hz), 2.83 (2H, m, H-9), 4.48 (2H, m, H-3,
1), 5.09 (1H, s, CCH2), 5.11 (1H, s, CCH2), 5.88 (1H, d, J =
11.3 Hz, H-7), 6.36 (1H, d, J = 11.3 Hz, H-6) .13C NMR δ 8.0, 8.1,
12.0, 12.5, 12.9, 22.4, 23.5, 25.9, 28.2, 29.0, 31.1, 31.4, 36.8, 36.9, 38.0,
38.1, 40.5, 45.8, 45.9, 53.4, 56.5, 70.6, 71.8, 75.3, 107.7, 115.3, 124.2,
130.4, 143.4, 152.0. MS (ESI) m/z 481.4 [(M + Na)+, 75]. HRMS
(ESI) calcd for C30H50NaO3 (M + Na)+ 481.3658, found 481.3617. IR
(neat) 3340, 2962, 2876, 1715, 1651, 1456, 1381, 1074, 1043, 976,
914, 758, 667 cm−1. UV (EtOH) λmax 246, 254, 264 nm.
(4R)-5-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]oxy}-
2-methylidene-9,10-secoestra-5,7-dien-17-yl]-4-ethyl-2-meth-
ylhexan-2-ol (13b). To a solution of ester 11b (28.0 mg, 0.042
mmol) in dry THF (0.5 mL) at −78 °C was added MeLi (1.04 M in
Et2O solution, 200 μL, 0.21 mmol), and the mixture was stirred for 1
h. The reaction was quenched with saturated NH4Cl aqueous solution,
and the mixture was extracted with AcOEt. The organic layer was
washed with brine, dried over MgSO4, and evaporated. The residue
was chromatographed on silica gel (20 g) with 2% AcOEt/hexane to
afford 13b (21.5 mg, 79%). 1H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H,
s, SiMe), 0.58 (3H, s, H-18), 0.78 (3H, J = 6.6 Hz, d, H-21), 0.87, 0.90
(each 9H, s, t-Bu), 1.24 (6H, s, diMe), 2.82 (1H, m, H-9), 4.42 (2H,
m, H-1,3), 4.92, 4.97 (each 1H, s, -CCH2), 5.84 (1H, d, J = 11.2 Hz,
H-7), 6.22 (1H, d, J = 11.2 Hz, H-6). 13C NMR δ −5.1, −4.92, −4.88,
−4.86, 12.1, 12.5, 12.9, 18.17, 18.24, 22.3, 23.4, 25.78 (3 carbons),
25.83 (3 carbons), 26.1, 28.2, 28.8, 30.1, 30.5, 36.8, 38.0, 38.6, 40.6,
43.0, 45.7, 47.6, 53.4, 56.4, 71.6, 71.7, 72.5, 106.2, 116.2, 122.4, 132.8,
141.2, 153.0. MS (FAB+) m/z 659.6 [(M + H)+, 2], 527.5 (11).
HRMS (FAB+) calcd for C40H75O3Si2 (M + H)+ 659.5255, found
659.5258. IR (neat) 3377, 2955, 2928, 2893, 2856, 1730, 1657, 1618,
1472, 1464, 1381, 1256, 1101, 1072, 935, 897, 835, 777 cm−1. UV
(hexane) λmax 246, 255, 265 nm.
22R-Ethyl-2-methylidene-19,24-dinor-1α,25-dihydroxyvita-
min D3 (6b). To a solution of 13b (16.1 mg, 0.025 mmol) in dry
MeOH (0.45 mL) was added camphor sulfonic acid (17.1 mg, 0.074
mmol) at 0 °C, and the mixture was stirred at room temperature for
1.5 h. The reaction was quenched with saturated NaHCO3 aqueous
solution and extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (15 g) with 40% AcOEt/hexane to
afford 6b (10.9 mg, quant). 1H NMR δ 0.59 (3H, s, H-18), 0.78 (3H, J
= 6.7 Hz, d, H-21), 0.87 (3H, t, J = 7.3 Hz, CH2CH3), 1.24 (6H, s,
diMe), 2.84 (2H, m, H-9), 4.48 (2H, m, H-1,3), 5.09, 5.11 (each 1H, s,
-CCH2), 5.89 (1H, d, J = 11.3 Hz, H-7), 6.36 (1H, d, J = 11.3 Hz,
H-6). 13C NMR δ 12.1, 12.5, 12.9, 22.4, 23.5, 26.1, 28.1, 29.0, 30.2,
30.4, 36.8, 37.9, 38.2, 40.4, 42.9, 45.8, 45.9, 53.4, 56.5, 70.6, 71.7, 71.8,
107.7, 115.3, 124.2, 130.4, 143.4, 152.0. MS (ESI) m/z 453.3 [(M +
Na)+, 100]. HRMS (ESI) calcd for C28H46NaO3 (M + Na)+ 453.3345,
found 453.3337. IR (neat) 3371, 2961, 2928, 2872, 1718, 1655, 1616,
1464, 1381, 1074, 1045, 903, 756, 665 cm−1. UV (EtOH) λmax 246,
254, 264 nm.
(3R)-4-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]oxy}-
2-methylidene-9,10-secoestra-5,7-dien-17-yl]-3-ethylpentyl 4-
Methylbenzenesulfonate (15b). To a solution of alcohol 12b (160
mg, 0.25 mmol) in dry pyridine (3 mL) was added p-toluenesulfonyl
chloride (72.5 mg, 0.38 mmol) at 0 °C, and the mixture was stirred for
37 h. To this solution was added ice−water and 1 N HCl, and the
mixture was extracted with AcOEt. The organic layer was washed with
saturated NaHCO3 aqueous solution and brine, dried over MgSO4,
and evaporated. The residue was chromatographed on silica gel (10 g)
1
with 5% AcOEt/hexane to afford 15b (164.9 mg, 83%). H NMR δ
0.03, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.53 (3H, s, H-18), 0.72 (3H,
d, J = 6.7 Hz, H-21), 0.78 (3H, t, J = 7.3 Hz, CH2CH3), 0.87, 0.90
(each 9H, s, t-Bu), 2.45 (3H, s, Ph-Me), 2.83 (1H, m, H-9), 3.99−4.17
(2H, m, H-24), 4.43 (2H, m, H-3, 1), 4.93 (1H, s, CCH2), 4.97
(1H, s, CCH2), 5.83 (1H, d, J = 11.0 Hz, H-7), 6.21 (1H, d, J = 11.0
Hz, H-6), 7.35 (2H, d, J = 8.0 Hz, arom-H), 7.80 (2H, d, J = 8.3 Hz,
arom-H). 13C NMR δ −5.1, −4.92, −4.86 (2 carbons), 11.9, 12.2, 12.7,
18.2, 18.3, 21.6, 22.1, 23.4, 24.4, 25.78 (3 carbons), 25.83 (3 carbons),
27.6, 28.6, 28.7, 36.2, 37.8, 38.6, 40.7, 45.6, 47.6, 53.6, 56.3, 70.0, 71.7,
72.5, 106.3, 116.2, 122.4, 127.9 (2 carbons), 129.8 (2 carbons), 132.9,
133.4, 140.1, 144.6, 153.0. MS (APCI) m/z 807.5 [(M + Na)+, 100].
HRMS (APCI) calcd for C45H76NaO5SSi2 (M + Na)+ 807.4850, found
807.4827. IR (neat) 2955, 2928, 2893, 2856, 1472, 1462, 1362, 1252,
1188, 1178, 1099, 1070, 935, 897, 835, 775, 667 cm−1. UV (hexane)
λmax 246, 255, 265 nm.
(4R)-5-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]oxy}-
2-methylidene-9,10-secoestra-5,7-dien-17-yl]-4-ethylhexane-
nitrile (16b). A solution of tosylate 15b (160.2 mg, 0.20 mmol) and
KCN (37.2 mg, 0.57 mmol) in dry DMSO (5 mL) was stirred at 70
°C for 1.5 h. The reaction was quenched with ice−water, and the
mixture was extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (10 g) with 2% AcOEt/hexane to
(5R,6R)-6-{1-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)-
silyl]oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]ethyl}-
3,5-diethylheptan-3-ol (14b). To a solution of ester 11b (56.0 mg,
0.083 mmol) in dry THF (1.5 mL) was added EtMgBr (1.0 M in THF
solution, 250 μL, 0.26 mmol), and the mixture was stirred at room
temperature. After 1.5 h, to the mixture was added an additional
EtMgBr (1.0 M in THF solution, 250 μL, 0.26 mmol), and the mixture
was stirred. In addition, after 1.5 h, to the mixture was added an
additional EtMgBr (1.0 M in THF solution, 250 μL, 0.26 mmol), and
the mixture was stirred for 1.5 h. The reaction was quenched with
saturated NH4Cl aqueous solution, and the mixture was extracted with
AcOEt. The organic layer was washed with brine, dried over MgSO4,
and evaporated. The residue was chromatographed on silica gel (10 g)
1
afford 16b (116.3 mg, 89%). H NMR δ 0.03, 0.05, 0.07, 0.08 (each
3H, s, SiMe), 0.57 (3H, s, H-18), 0.79 (3H, d, J = 6.8 Hz, H-21), 0.87,
0.90 (each 9H, s, t-Bu), 2.83 (1H, m, H-9), 4.43 (2H, m, H-3, 1), 4.93
(1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84 (1H, d, J = 11.1 Hz, H-
7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ −5.1, −4.92, −4.87 (2
carbons), 12.0, 12.2, 12.8, 16.0, 18.2, 18.3, 22.1, 23.4, 23.6, 25.2, 25.78
1
with 3% AcOEt/hexane to afford 14b (47.1 mg, 82%). H NMR δ
4363
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
(3 carbons), 25.83 (3 carbons), 27.7, 28.7, 36.1, 38.6, 40.7, 41.1, 45.7,
47.6, 53.7, 56.3, 71.7, 72.5, 106.3, 116.3, 120.2, 122.4, 132.9, 140.9,
153.0. MS (APCI) m/z 662.5 [(M + Na)+, 15]. HRMS (APCI) calcd
for C39H69NNaO2Si2 (M + Na)+ 662.4765, found 662.4751. IR (neat)
2955, 2930, 2885, 2856, 2245, 1472, 1462, 1252, 1101, 1070, 935, 897,
835, 775 cm−1. UV (hexane) λmax 246, 255, 264 nm.
(4R)-5-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]oxy}-
2-methylidene-9,10-secoestra-5,7-dien-17-yl]-4-ethylhexanal
(17b). Cyanide 16b (10.9 mg, 0.017 mmol) in dry CH2Cl2 (1 mL)
was treated with DIBAL (1 M in toluene solution, 34 μL, 0.034 mmol)
at −20 °C for 1 h. The reaction was quenched with 1 N HCl and
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (5 g) with 2% AcOEt/hexane to afford 17b (9.1 mg, 84%).
1H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.56 (3H, s, H-
18), 0.80 (3H, d, J = 6.4 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu), 2.83
(1H, m, H-9), 4.43 (2H, m, H-3, 1), 4.92 (1H, s, CCH2), 4.97 (1H,
s, CCH2), 5.84 (1H, d, J = 11.2 Hz, H-7), 6.22 (1H, d, J = 11.2 Hz,
H-6), 9.78 (1H, t, J = 1.9 Hz, -CHO). 13C NMR δ −5.1, −4.92, −4.86
(2 carbons), 11.9, 12.4, 12.8, 18.2, 18.3, 21.2, 22.1, 23.4, 24.0, 25.78 (3
carbons), 25.84 (3 carbons), 27.7, 28.8, 29.7, 36.4, 38.6, 40.7, 41.5,
43.0, 45.7, 47.6, 53.7, 56.3, 71.6, 72.5, 106.2, 116.2, 122.4, 132.8, 141.1,
153.0, 203.2. MS (FAB+) m/z 643.5 [(M + H)+, 1], 511.4 (9). HRMS
(FAB+) calcd for C39H71O3Si2 (M + H)+ 643.4942, found 643.4927.
IR (neat) 2955, 2928, 2895, 2856, 2709, 1728, 1472, 1463, 1256, 1101,
1070, 935, 897, 835, 775 cm−1. UV (hexane) λmax 246, 255, 265 nm.
(4R)-5-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]oxy}-
2-methylidene-9,10-secoestra-5,7-dien-17-yl]-4-ethylhexan-1-
ol (18b). To a solution of aldehyde 17b (21.2 mg, 0.033 mmol) in dry
MeOH/dry CH2Cl2 (1:1, 4 mL) was added NaBH4 (20.0 mg, 0.53
mmol) at 0 °C, and the mixture was stirred at room temperature for
1.5 h. To the solution was added ice−water, and the mixture was
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (5 g) with 10% AcOEt/hexane to afford 18b (20.0 mg, 94%).
1H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s, SiMe), 0.56 (3H, s, H-
18), 0.77 (3H, d, J = 6.3 Hz, H-21), 0.87, 0.90 (each 9H, s, t-Bu), 2.83
(1H, m, H-9), 3.64 (2H, t, J = 6.5 Hz, H-25), 4.43 (2H, m, H-3, 1),
4.92 (1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84 (1H, d, J = 11.1
Hz, H-7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ −5.1, −4.92,
−4.86 (2 carbons), 12.0, 12.6, 13.0, 18.2, 18.3, 22.2, 23.5, 24.4, 25.0,
25.78 (3 carbons), 25.84 (3 carbons), 27.8, 28.8, 31.9, 36.5, 38.6, 40.7,
41.7, 45.7, 47.6, 53.8, 56.4, 63.6, 71.6, 72.5, 106.2, 116.1, 122.4, 132.8,
141.2, 153.0. MS (APCI) m/z 667.5 [(M + Na)+, 7]. HRMS (APCI)
calcd for C39H72NaO3Si2 (M + Na)+ 667.4918, found 667.4952. IR
(neat) 3331, 2955, 2928, 2894, 2858, 1655, 1618, 1472, 1462, 1256,
1101, 1070, 935, 908, 897, 835, 775 cm−1. UV (hexane) λmax 246, 255,
265 nm.
mmol) in dry MeOH (20 mL) was added KOH (24.1 mg, 0.43 mmol)
and then iodine (54.5 mg, 0.21 mmol) at 0 °C, and the mixture was
stirred for 1.5 h. The reaction was quenched with 10% Na2SO3, and
the mixture was extracted with AcOEt. The organic layer was washed
with brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (5 g) with 2% AcOEt/hexane to afford
1
19b (51.0 mg, 92%). H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s,
SiMe), 0.56 (3H, s, H-18), 0.79 (3H, d, J = 6.2 Hz, H-21), 0.87, 0.90
(each 9H, s, t-Bu), 2.83 (1H, m, H-9), 3.68 (3H, s, -OCHH3), 4.43
(2H, m, H-3, 1), 4.93 (1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84
(1H, d, J = 11.1 Hz, H-7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ
−5.1, −4.92, −4.88 (2 carbons), 11.9, 12.4, 12.8, 18.2, 18.3, 22.1, 23.4,
24.0, 24.4, 25.77 (3 carbons), 25.84 (3 carbons), 27.7, 28.8, 33.1, 36.4,
38.6, 40.6, 41.4, 45.7, 47.6, 51.5, 53.6, 56.3, 71.6, 72.5, 106.2, 116.1,
122.4, 132.8, 141.2, 153.0, 174.7. MS (APCI) m/z 695.5 [(M + Na)+,
45]. HRMS (APCI) calcd for C40H72NaO4Si2 (M + Na)+ 695.4867,
found 695.4863. IR (neat) 2955, 2930, 2893, 2856, 1742, 1472, 1462,
1256, 1101, 1070, 935, 897, 835, 775 cm−1. UV (hexane) λmax 246,
255, 265 nm.
(5R)-6-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)silyl]oxy}-
2-methylidene-9,10-secoestra-5,7-dien-17-yl]-5-ethyl-2-meth-
ylheptan-2-ol (20b). To a solution of ester 19b (46.7 mg, 0.069
mmol) in dry THF (2 mL) at −78 °C was added MeLi (1.04 M in
Et2O solution, 400 μL, 0.42 mmol), and the mixture was stirred. The
mixture was heated to −10 °C for 2 h and then stirred for 2 h. The
reaction was quenched with saturated NH4Cl aqueous solution, and
the mixture was extracted with AcOEt. The organic layer was washed
with brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (10 g) with 7% AcOEt/hexane to afford
1
20b (40 mg, 86%). H NMR δ 0.03, 0.05, 0.07, 0.08 (each 3H, s,
SiMe), 0.56 (3H, s, H-18), 0.79 (3H, d, J = 6.2 Hz, H-21), 0.87, 0.90
(each 9H, s, t-Bu), 1.22, 1.23 (each 3H, s, H-26, 27), 2.33 (1H, dd, J =
13.3, 3.2 Hz, H-14), 2.83 (1H, m, H-9), 4.42 (2H, m, H-3, 1), 4.92
(1H, s, CCH2), 4.97 (1H, s, CCH2), 5.84 (1H, d, J = 11.1 Hz, H-
7), 6.22 (1H, d, J = 11.1 Hz, H-6). 13C NMR δ −5.1, −4.92, −4.88 (2
carbons), 12.0, 12.6, 13.1, 18.2, 18.3, 22.1, 23.2, 23.5, 24.5, 25.77 (3
carbons), 25.84 (3 carbons), 27.8, 28.8, 29.1, 29.5, 36.5, 38.6, 40.7,
42.4, 43.1, 45.7, 47.6, 53.9, 56.3, 71.3, 71.6, 72.5, 106.2, 116.1, 122.4,
132.7, 141.3, 153.0. MS (APCI) m/z 673.6 [(M + H)+, 5], 657.6 (5),
541.5 (45), 523.5 (90), 409.4 (40), 391.4 (100). HRMS (APCI) calcd
for C41H77O3Si2 (M + H)+ 673.5411, found 673.5380. IR (neat) 3387,
2957, 2930, 2895, 2856, 1659, 1620, 1472, 1464, 1256, 1101, 1072,
935, 908, 897, 835, 775 cm−1. UV (hexane) λmax 246, 255, 265 nm.
22R-Ethyl-2-methylidene-19-nor-1α,25-dihydroxyvitamin
D3 (8b). A solution of 20b (30.1 mg, 0.045 mmol) in dry MeOH (1
mL) was added camphor sulfonic acid (31.15 mg, 0.13 mmol) at 0 °C,
and the mixture was stirred at room temperature for 1.5 h. The
reaction was quenched with saturated NaHCO3 aqueous solution and
extracted with AcOEt. The organic layer was washed with brine, dried
over MgSO4, and evaporated. The residue was chromatographed on
silica gel (10 g) with 40% AcOEt/hexane to afford 8b (20.5 mg,
quant). 1H NMR δ 0.57 (3H, s, H-18), 0.79 (3H, d, J = 6.2 Hz, H-21),
0.86 (3H, m, CH2CH3), 1.217, 1.224 (each 3H, s, H-26, 27), 2.57
(1H, dd, J = 13.4, 3.9 Hz, H-14), 2.84 (2H, m, H-9), 4.47 (2H, m, H-3,
1), 5.09 (1H, s, CCH2), 5.11 (1H, s, CCH2), 5.89 (1H, d, J =
11.2 Hz, H-7), 6.36 (1H, d, J = 11.2 Hz, H-6). 13C NMR δ 12.0, 12.6,
13.1, 22.2, 23.2, 23.5, 24.4, 27.7, 29.0 (2 carbons), 29.5, 36.4, 38.1,
40.5, 42.4, 43.1, 45.8, 45.9, 53.9, 56.4, 70.6, 71.3, 71.8, 107.7, 115.3,
124.2, 130.4, 143.4, 152.0. MS (APCI) m/z 467.4 [(M + Na)+, 15].
HRMS (APCI) calcd for C29H48NaO3 (M + Na)+ 467.3501, found
467.3502. IR (neat) 3364, 2959, 2928, 2870, 1456, 1379, 1151, 1045,
667 cm−1. UV (EtOH) λmax 246, 254, 264 nm.
22R-Ethyl-2-methylidene-19,26,27-trinor-1α,25-dihydroxy-
vitamin D3 (7b). A solution of 18b (14.4 mg, 0.022 mmol) in dry
MeOH (0.5 mL) was added camphor sulfonic acid (15.5 mg, 0.067
mmol) at 0 °C, and the mixture was stirred at room temperature for 1
h. The reaction was quenched with saturated NaHCO3 aqueous
solution and extracted with AcOEt. The organic layer was washed with
brine, dried over MgSO4, and evaporated. The residue was
chromatographed on silica gel (10 g) with 40% AcOEt/hexane to
afford 7b (8.2 mg, 88%). 1H NMR δ 0.57 (3H, s, H-18), 0.77 (3H, d, J
= 6.3 Hz, H-21), 0.87 (3H, m, CH2CH3), 2.58 (1H, dd, J = 13.2, 3.8
Hz, H-14), 2.84 (2H, m, H-9), 3.64 (2H, t, J = 6.6 Hz, H-25), 4.48
(2H, m, H-3, 1), 5.09 (1H, s, CCH2), 5.11 (1H, s, CCH2), 5.89
(1H, d, J = 11.3 Hz, H-7), 6.36 (1H, d, J = 11.3 Hz, H-6). 13C NMR δ
12.0, 12.5, 12.9, 22.2, 23.5, 24.4, 25.0, 27.7, 29.0, 31.8, 36.4, 38.2, 40.5,
41.7, 45.8, 45.9, 53.8, 56.4, 63.6, 70.7, 71.8, 107.7, 115.3, 124.2, 130.4,
143.4, 152.0. MS (APCI) m/z 439.3 [(M + Na)+, 17]. HRMS (APCI)
calcd for C27H44NaO3 (M + Na)+ 439.3188, found 439.3187. IR
(neat) 3352, 2924, 2854, 1379, 1074, 1045, 756, 667 cm−1. UV
(EtOH) λmax 246, 254, 264 nm.
Competitive Binding Assay, Human VDR. The human
recombinant VDR ligand-binding domain (LBD) was expressed as
an N-terminal GST-tagged protein in E. coli Rosetta2 (DE3) pLysS.19
The cells were lysed by sonication. The supernatants were diluted
approximately 500 times in 50 mM Tris buffer (100 mM KCl, 5 mM
DTT, 0.5% CHAPS, pH 7.5) containing bovine serum albumin (100
μg/mL). Binding to GST-hVDR-LBD was evaluated according to the
procedure reported.27The receptor solution (570 μL) in an assay tube
Methyl (4R)-5-[(1R,3R,7E,17β)-1,3-Bis{[tert-butyl(dimethyl)-
silyl]oxy}-2-methylidene-9,10-secoestra-5,7-dien-17-yl]-4-eth-
ylhexanoate (19b). To a solution of aldehyde 17b (53.1 mg, 0.083
4364
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
was incubated with [3H]-1α,25-dihydroxyvitamin D3 (specific activity,
5.85 TBq/mmol, 2000 cpm) together with graded amounts of each
vitamin D analogue (0.001−100 nM) or vehicle for 16 h at 4 °C. The
bound and free [3H]-1α,25-dihydroxyvitamin D3 molecules were
separated by treating with dextran-coated charcoal for 30 min at 4 °C.
The assay tubes were centrifuged at 1000g for 10 min. The
radioactivity of the supernatant was counted. Nonspecific binding
was subtracted. These experiments were done in duplicate.
with the software Phaser30 in the CCP4 program31 using rat VDR-
LBD coordinates (PDB code 2ZLC), and finalized sets of atomic
coordinates were obtained after iterative rounds of model modification
with the program Coot32 and refinement with CNS and refmac5.33−37
The coordinate data for the structures were deposited in Protein Data
Bank with accession numbers 3WT5 (VDR-LBD/8a complex, crystal
A), 3WT6 (VDR-LBD/8a complex, crystal B), 3WT7 (VDR-LBD/9a
complex).
Transfection and Transactivation Assay. COS-7 cells were
cultured in Dulbecco’s modified Eagle medium (DMEM) supple-
mented with 5% fetal bovine serum (FBS). Cells were seeded on 24-
well plates at a density of 2 × 104 per well. After 24 h, the cells were
transfected with a reporter plasmid containing three copies of the
mouse osteopontin VDRE (5′-GGTTCAcgaGGTTCA, SPPx3-TK-
Luc), a wild-type or mutant hVDR expression plasmid (pCMX-
hVDR), and the internal control plasmid containing sea pansy
luciferase expression constructs (pRL-CMV) by the lipofection
method as described previously.13 After an 8 h incubation, the cells
were treated with either the ligand or ethanol vehicle and cultured for
16 h. Cells in each well were harvested with a cell lysis buffer, and the
luciferase activity was measured with a luciferase assay kit (Promega,
WI, U.S.). Transactivation measured by the luciferase activity was
normalized with the internal control. All experiments were done in
triplicate.
Protein Expression and Purification. Expression of rat VDR-
LBD (residues 116−423, Δ165−211) and the following cell-lysis and
centrifugation were done by the procedure reported previously.16 The
supernatant was applied to cOmplete His-tag purification resin (Roche
Diagnostics GmbH, Mannheim, Germany), and the resin was
thoroughly washed in wash buffer (50 mM Na/K phosphate, pH
7.0, 20 mM imdazole, 500 mM NaCl, 5% glycerol, 1% Tween 20, 1
mM TCEP). The rat VDR-LBD was eluted with elution buffer (250
mM imidazole, 50 mM Na/K phosphate, pH 7.0, 500 mM NaCl, 5%
glycerol, 1% Tween 20, 1 mM TCEP). The protein was dialyzed
overnight, dialysis with 500 mL of buffer A (20 mM Na/K phosphate,
pH 7.0, 5% glycerol, 1 mM EDTA, 0.5 mM DTT), and then loaded
onto a HiTrap SP HP (5 mL) column (GE Healthcare) equilibrated
with buffer A. The elution was performed by NaCl gradient buffer
from 0 to 1.0 M. His-tag of the protein in elution mixture (8 mL) was
cleaved by addition of 70 units of thrombin and subsequent incubation
at 4 °C for 18 h. Then NaCl (1.0 g) was added to the digested mixture
(8 mL), and the resulting mixture was passed through a HiTrap
benzamidine FF (1 mL) columun (GE healthcare) with buffer (10
mM Tris-HCl, pH 7.0, 2 M NaCl, 0.1 mM DTT). The flow-through
was further purified by Superdex S75 gel filtration (25 mL) column
(GE Healthcare) with a buffer (100 mM NaCl, Tris-HCl, pH 7.0).
Purified rat VDR-LBD was concentrated in buffer (10 mM Tris-HCl,
pH 7.0, 2 mM NaN3, 10 mM DTT) to 7.0 mg/mL, which was
estimated by UV absorbance at 280 nm.
Crystallization. A mixture of rVDR-LBD and a ligand (5 equiv)
was incubated at room temperature for 1 h. Then coactivator peptide
(H2N-KNHPMLMNLLKDN-CONH2) derived from DRIP205 in
buffer (25 mM Tris-HCl, pH 8.0; 50 mM NaCl; 10 mM DTT; 2
mM NaN3) was added. The mixture of VDR-LBD/ligand/peptide was
allowed to crystallize by the vapor diffusion method using a series of
precipitant solutions containing 0.1 M MOPS-Na (pH 7.0), 0.075−0.2
M sodium formate, 12−22% (w/v) PEG4000, and 5% ethylene glycol.
Droplets for crystallization were prepared by mixing 1 μL of complex
solution and 1 μL of precipitant solution, and droplets were
equilibrated against 300 μL of precipitant solution at 20 °C. The
mixture was stored at 20 °C, and crystals appeared after a few days.
X-ray Crystallographic Analysis. Prior to diffraction data
collection, crystals were soaked in a cryoprotectant solution containing
0.1 M MOPS-Na, pH 7.0, 0.075−0.2 M sodium formate, 16−20% (w/
v) PEG4000, and 18−24% ethylene glycol. Diffraction data sets were
collected at 100 K in a stream of nitrogen gas at beamline BL-5A of
KEK-PF (Tsukuba, Japan). Reflections were recorded with an
oscillation range per image of 1.0°. Diffraction data were indexed,
integrated, and scaled using the program iMOSFLM.28,29 The
structures of ternary complex were solved by molecular replacement
AUTHOR INFORMATION
Corresponding Author
*E-mail: yamamoto@ac.shoyaku.ac.jp. Phone: +81 42 721
1580. Fax: +81 42 721 1580.
Notes
■
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by Platform for Drug Discovery,
Informatics, and Structural Life Science and by a Grant-in-Aid
for Scientific Research (Grant 23590135) from the Ministry of
Education, Culture, Sports, Science and Technology, Japan. We
also thank the Takeda Science Foundation, Japan, for financial
support. Synchrotron-radiation experiments were performed at
the Photon Factory (Proposal 2011G685), and we are grateful
for the assistance provided by the beamline scientists at the
Photon Factory.
ABBREVIATIONS USED
■
VDR, vitamin D receptor; LBD, ligand binding domain; LBP,
ligand binding pocket; 1,25-(OH)2D3, 1α,25-dihydroxyvitamin
D3; RXR, retinoid X receptor; ER, estrogen receptor
REFERENCES
■
(1) Pike, J. W.; Meyer, M. B.; Bishop, K. A. Regulation of target gene
expression by the vitamin D receptoran update on mechanisms. Rev.
Endocr. Metab. Disord. 2012, 13, 45−55.
(2) Jurutka, P. W.; Whitfield, G. K.; Hsieh, J. C.; Thompson, P. D.;
Haussler, C. A.; Haussler, M. R. Molecular nature of the vitamin D
receptor and its role in regulation of gene expression. Rev. Endocr.
Metab. Disord. 2001, 2, 203−216.
(3) Orlov, I.; Rochel, N.; Moras, D.; Klaholz, B. P. Structure of the
full human RXR/VDR nuclear receptor heterodimer complex with its
DR3 target DNA. EMBO J. 2012, 31, 291−300.
(4) Kubodera, N. Pharmaceutical studies on vitamin D derivatives
and practical syntheses of six commercially available vitamin D
derivatives that contribute to current clinical practice. Heterocycles
2010, 80, 83−98.
(5) Binderup, L.; Binderup, E. Godtfredsen, W. O. Development of
New Vitamin D Analogs. In Vitamin D; Feldman, D., Pike, J. W.,
Glorieux, F. H., Eds.; Elsevier Academic Press: San Diego, CA, 1997;
1027−1043.
(6) Bury, Y.; Steinmeyer, A.; Carlberg, C. Structure activity
relationship of carboxylic ester antagonists of the vitamin D3 receptor.
Mol. Pharmacol. 2000, 58, 1067−1074.
(7) Igarashi, M.; Yoshimoto, N.; Yamamoto, K.; Shimizu, 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.
(8) Miura, D.; Manabe, K.; Ozono, K.; Saito, M.; Gao, Q.; Norman,
A. W.; Ishizuka, S. Antagonistic action of novel 1α,25-dihydrox-
yvitamin D3-26,23-lactone analogs on differentiation of human
leukemia cells (HL-60) induced by 1α,25-dihydroxyvitamin D3. J.
Biol. Chem. 1999, 274, 16392−16399.
(9) Ishizuka, S.; Kurihara, N.; Miura, D.; Takenouchi, K.; Cornish, J.;
Cundy, T.; Reddy, S. V.; Roodman, G. D. Vitamin D antagonist, TEI-
4365
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
9647, inhibits osteoclast formation induced by 1α,25-dihydroxyvitamin
D3 from pagetic bone marrow cells. J. Steroid Biochem. Mol. Biol. 2004,
89−90, 331−334.
(24) (a) Xu, H. E.; Lambert, M. H.; Montana, V. G.; Plunket, K. D.;
Moore, L. B.; Collins, J. L.; Oplinger, J. A.; Kliewer, S. A.; Gampe, R.
T., Jr.; McKee, D. D.; Moore, J. T.; Willson, T. M. Structural
determinants of ligand binding selectivity between the peroxisome
proliferator-activated receptors. Proc. Natl. Acad. Sci. U.S.A. 2001, 98,
13919−13924. (b) Mahindroo, N.; Huang, C. F.; Peng, Y. H.; Wang,
C. C.; Liao, C. C.; Lien, T. W.; Chittimalla, S. K.; Huang, W. J.; Chai,
C. H.; Prakash, E.; Chen, C. P.; Hsu, T. A.; Peng, C. H.; Lu, I. L.; Lee,
L. H.; Chang, Y. W.; Chen, W. C.; Chou, Y. C.; Chen, C. T.; Goparaju,
C. M.; Chen, Y. S.; Lan, S. J.; Yu, M. C.; Chen, X.; Chao, Y. S.; Wu, S.
Y.; Hsieh, H. P. Novel indole-based peroxisome proliferator-activated
receptor agonists: design, SAR, structural biology, and biological
activities. J. Med. Chem. 2005, 48, 8194−8208. (c) Burgermeister, E.;
Schnoebelen, A.; Flament, A.; Benz, J.; Stihle, M.; Gsell, B.; Rufer, A.;
(10) Roodman, G. D.; Windle, J. J. Paget disease of bone. J. Clin.
Invest. 2005, 115, 200−208.
(11) Saito, N.; Saito, H.; Anzai, M.; Yoshida, A.; Fujishima, T.;
Takenouchi, K.; Miura, D.; Ishizuka, S.; Takayama, H.; Kittaka, A.
Dramatic enhancement of antagonistic activity on vitamin D receptor:
a double functionalization of 1α-hydroxyvitamin D3 26,23-lactones.
Org. Lett. 2003, 5, 4859−4862.
(12) Yoshimoto, N.; Inaba, Y.; Yamada, S.; Makishima, M.; Shimizu,
M.; Yamamoto, K. 2-Methylene 19-nor-25-dehydro-1α-hydroxyvita-
min D3 26,23-lactones: synthesis, biological activities and molecular
basis of passive antagonism. Bioorg. Med. Chem. 2008, 16, 457−473.
(13) Inaba, Y.; Yoshimoto, N.; Sakamaki, Y.; Nakabayashi, M.; Ikura,
T.; Tamamura, H.; Ito, N.; Shimizu, M.; Yamamoto, K. A new class of
vitamin D analogues that induce structural rearrangement of the
ligand-binding pocket of the receptor. J. Med. Chem. 2009, 52, 1438−
1449.
Ruf, A.; Kuhn, B.; Marki, H. P.; Mizrahi, J.; Sebokova, E.; Niesor, E.;
̈
Meyer, M. A novel partial agonist of peroxisome proliferator-activated
receptor-γ (PPARγ) recruits PPARγ-coactivator-1α, prevents trigly-
ceride accumulation, and potentiates insulin signaling in vitro. Mol.
Endocrinol. 2006, 20, 809−830.
(14) Inaba, Y.; Nakabayashi, M.; Itoh, T.; Yoshimoto, N.; Ikura, T.;
Ito, N.; Shimizu, M.; Yamamoto, K. 22S-Butyl-1α,24R-dihydroxyvita-
min D3: recovery of vitamin D receptor agonistic activity. J. Steroid
Biochem. Mol. Biol. 2010, 121, 146−150.
(25) (a) Brzozowski, A. M.; Pike, A. C.; Dauter, Z.; Hubbard, R. E.;
Bonn, T.; Engstrom, O.; Ohman, L.; Greene, G. L.; Gustafsson, J. A.;
̈
Carlquist, M. Molecular basis of agonism and antagonism in the
oestrogen receptor. Nature 1997, 389, 753−758. (b) 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.
(15) Sakamaki, Y.; Inaba, Y.; Yoshimoto, N.; Yamamoto, K. Potent
antagonist for the vitamin D receptor: vitamin D analogues with
simple side chain structure. J. Med. Chem. 2010, 53, 5813−5826.
(16) Yoshimoto, N.; Sakamaki, Y.; Haeta, M.; Kato, A.; Inaba, Y.;
Itoh, T.; Nakabayashi, M.; Ito, N.; Yamamoto, K. Butyl pocket
formation in the vitamin D receptor strongly affects the agonistic or
antagonistic behavior of ligands. J. Med. Chem. 2012, 55, 4373−4381.
(17) Yamamoto, K.; Ogura, H.; Jukuta, J.; Inoue, H.; Hamada, K.;
Sugiyama, Y.; Yamada, S. Stereochemical and mechanistic studies on
conjugate addition of organocuprates to acyclic enones and enoates:
simple rule for diastereofacial selectivity. J. Org. Chem. 1998, 63,
4449−4458.
(18) Yamada, S.; Morizono, D.; Yamamoto, K. Mild oxidation of
aldehydes to the corresponding carboxylic acids esters: alkaline iodine
oxidation revised. Tetrahedron Lett. 1992, 33, 4329−4332.
(19) 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.
(20) 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-1α,25-dihydroxy-26,27-di-
methyl-19-norvita min D3 analogs: synthesis, biological evaluation,
and crystal structure. Bioorg. Med. Chem. 2008, 16, 6949−6964.
(21) 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.
(22) (a) Pochetti, G.; Godio, C.; Mitro, N.; Caruso, D.; Galmozzi, A.;
Scurati, S.; Loiodice, F.; Fracchiolla, G.; Tortorella, P.; Laghezza, A.;
Lavecchia, A.; Novellino, E.; Mazza, F.; Crestani, M. Insights into the
mechanism of partial agonism: crystal structures of the peroxisome
proliferator-activated receptor gamma ligand-binding domain in the
complex with two enantiomeric ligands. J. Biol. Chem. 2007, 282,
17314−17324. (b) Bruning, J. B.; Parent, A. A.; Gil, G.; Zhao, M.;
Nowak, J.; Pace, M. C.; Smith, C. L.; Afonine, P. V.; Adams, P. D.;
Katzenellenbogen, J. A.; Nettles, K. W. Coupling of receptor
conformation and ligand orientation determine graded activity. Nat.
Chem. Biol. 2010, 6, 837−843.
(23) (a) Ciesielski, F.; Rochel, N.; Mitschler, A.; Kouzmenko, A.;
Moras, D. Structural investigation of the ligand binding domain of the
zebrafish VDR in complexes with 1α,25(OH)2D3 and Gemini:
purification, crystallization and preliminary X-ray diffraction analysis.
J. Steroid Biochem. Mol. Biol. 2004, 89−90, 55−59. (b) 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.
(26) Singarapu, K. K.; Zhu, J.; Tonelli, M.; Rao, H.; Assadi-Porter, F.
M.; DeLuca, H. F.; Markley, J. L. Ligand-specific structural changes in
the vitamin D receptor in solution. Biochemistry 2011, 50, 11025−
11033.
(27) Yamamoto, K.; Sun, W.-Y.; Ohta, M.; Hamada, K.; DeLuca, H.
F.; Yamada, S. Conformationally restricted analogs of 1α,25-
dihydroxyvitamin D3 and its 20-epimer: compounds for study of the
three-dimensional structure of vitamin D responsible for binding to
the receptor. J. Med. Chem. 1996, 39, 2727−2737.
(28) Leslie, A. G. W.; Powell, H. R. In Evolving Methods for
Macromolecular Crystallography; Read, R. J., Sussman, J. L., Eds.;
NATO Science Series II, Vol. 245; Springer Verlag: Dordrecht, The
Netherlands, 2007; pp 41−51.
(29) Battye, T. G.; Kontogiannis, L.; Johnson, O.; Powell, H. R.;
Leslie, A. G. iMOSFLM: a new graphical interface for diffraction-image
processing with MOSFLM. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2011, 67, 271−281.
(30) McCoy, A. J.; Grosse-Kunstleve, R. W.; Adams, P. D.; Winn, M.
D.; Storoni, L. C.; Read, R. J. Phaser crystallographic software. J. Appl.
Crystallogr. 2007, 40, 658−674.
(31) Collaborative Computational Project Number 4. The CCP4
Suite: Programs for Protein Crystallography. Acta Crystallogr., Sect. D:
Biol. Crystallogr. 1994, 50, 760−763
(32) Emsley, P.; Lohkamp, B.; Scott, W. G.; Cowtan, K. Features and
development of Coot. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2010,
66, 486−501.
(33) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement of
Macromolecular Structures by the Maximum-Likelihood method. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53, 240−255.
(34) Pannu, N. S.; Murshudov, G. N.; Dodson, E. J.; Read, R. J.
Incorporation of Prior Phase Information Strengthens Maximum-
Likelihood Structure Refinement. Acta Crystallogr., Sect. D: Biol.
Crystallogr. 1998, 54, 1285−1294.
(35) Winn, M. D.; Isupov, M. N.; Murshudov, G. N. Use of TLS
parameters to model anisotropic displacements in macromolecular
refinement. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2001, 57, 122−
133.
(36) Steiner, R. A.; Lebedev, A. A.; Murshudov, G. N. Fisher’s
information in maximum-likelihood macromolecular crystallographic
refinement. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2003, 59, 2114−
2124.
4366
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367

Journal of Medicinal Chemistry
Article
́
(37) Murshudov, G. N.; Skubak, P.; Lebedev, A. A.; Pannu, N. S.;
Steiner, R. A.; Nicholls, R. A.; Winn, M. D.; Long, F.; Vagin, A. A.
REFMAC5 for the refinement of macromolecular crystal structures.
Acta Crystallogr., Sect. D: Biol. Crystallogr. 2011, 67, 355−367.
4367
dx.doi.org/10.1021/jm500392t | J. Med. Chem. 2014, 57, 4351−4367