Rational design, synthesis, and biological activity of novel conformationally restricted vitamin D analogues, (22R)- and (22S)-22-ethyl-1,25-dihydroxy-23,24-didehydro-24a,24b-dihomo-20-epivitamin D3

Article abstract of DOI:10.1021/jm0105631

Two new vitamin D analogues, (22R)- and (22S)-22-ethyl-1,25-dihydroxy-23,24-didehydro-24a,24b-dihomo-20-epivitamin D₃ (3 and 4), were rationally designed on the basis of the active space group concept previously proposed by us. The 22R ethyl group of 3 restricts the mobility of the side chain to active space regions, whereas the 22S ethyl group of 4 confines the side chain to an inactive region. The double bond at C(23) further restricts the side chain flexibility. These compounds (3 and 4) were synthesized using ortho ester Claisen rearrangement as the key step. As expected, the 22R isomer 3 has nearly 100 times higher efficacy than 1,25-dihydroxyvitamin D₃ (1) in cell differentiation, although its affinity for the vitamin D receptor (VDR) was one-seventh of that of 1. The 22S isomer 4 has significantly lower efficacy than 3. A docking study in combination with site-directed mutation analysis revealed that two carbon elongated side chain analogue 3 could be fitted in the ligand binding pocket of the VDR by adopting a stable conformation.

Full text of DOI:10.1021/jm0105631

J . Med. Chem. 2002, 45, 1825-1834
1825
Ra tion a l Design , Syn th esis, a n d Biologica l Activity of Novel Con for m a tion a lly
Restr icted Vita m in D An a logu es, (22R)- a n d
(22S)-22-Eth yl-1,25-d ih yd r oxy-23,24-d id eh yd r o-24a ,24b-d ih om o-20-ep ivita m in D3
Hiroyuki Masuno,† Keiko Yamamoto,*,† Xinxiang Wang,† Mihwa Choi,† Hiroshi Ooizumi,‡
Toshimasa Shinki,§ and Sachiko Yamada*,†
Institute of Biomaterials and Bioengineering, Tokyo Medical and Dental University, 2-3-10, Kanda-Surugadai, Chiyoda-ku,
Tokyo 101-0062, J apan, Department of Genetics and Molecular Biology, Kyoto University, Kyoto 606-8507, J apan, and
School of Dentistry, Showa University, Tokyo 142-8555, J apan
Received December 11, 2001
Two new vitamin D analogues, (22R)- and (22S)-22-ethyl-1,25-dihydroxy-23,24-didehydro-24a,-
24b-dihomo-20-epivitamin D3 (3 and 4), were rationally designed on the basis of the active
space group concept previously proposed by us. The 22R ethyl group of 3 restricts the mobility
of the side chain to active space regions, whereas the 22S ethyl group of 4 confines the side
chain to an inactive region. The double bond at C(23) further restricts the side chain flexibility.
These compounds (3 and 4) were synthesized using ortho ester Claisen rearrangement as the
key step. As expected, the 22R isomer 3 has nearly 100 times higher efficacy than 1,25-
dihydroxyvitamin D3 (1) in cell differentiation, although its affinity for the vitamin D receptor
(VDR) was one-seventh of that of 1. The 22S isomer 4 has significantly lower efficacy than 3.
A docking study in combination with site-directed mutation analysis revealed that two carbon
elongated side chain analogue 3 could be fitted in the ligand binding pocket of the VDR by
adopting a stable conformation.
Ch a r t 1
In tr od u ction
1R,25-Dihydroxyvitamin D3 [1,25-(OH)2D3, 1] (Chart
1), which was originally recognized as a specific hor-
mone for calcium homeostasis, is now known to be a
multifunctional hormone, functioning in inducing cell
differentiation, in suppressing cell proliferation, and in
regulating the immune response.1 These biological
activities are mediated by the vitamin D receptor
(VDR),2 which is a member of the nuclear receptor (NR)
superfamily.3 Because of their potential value in clinical
medicine, nearly a thousand vitamin D analogues have
been synthesized and their biological activities have
been evaluated.4 Many of them have already been
developed, and some are under development, as clinical
agents for treating metabolic bone diseases, skin dis-
eases such as psoriasis, immune disorders, or malignant
tumors.5
We have been studying the structure-function rela-
tionship of vitamin D, focusing on side chain structure
and conformation6 because most active vitamin D
analogues are modified in the side chain and we
consider that side chain modification is the key to
developing potential analogues of vitamin D. On the
basis of conformational analysis of the side chain of 1
and its 20-epimer (2) and studies using conformationally
restricted synthetic vitamin D analogues,6 we have
developed the “active space region” concept of the
vitamin D side chain. In this concept, the spatial region
that the vitamin D side chain can occupy is divided into
* To whom correspondence should be addressed. Tel.: +81-3-
5280-8036. Fax: 03-5280-8005. E-mail: yamada.mr@tmd.ac.jp and
yamamoto.mr@tmd.ac.jp.
† Tokyo Medical and Dental University.
‡ Kyoto University.
§ Showa University.
10.1021/jm0105631 CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/27/2002

1826 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Masuno et al.
F igu r e 1. Spatial regions occupied by the side chains of 1, 2, 22-oxa-1,25-(OH)2D3 (OCT), and 22-Et-23-ene-24-dihomo-20-epi-
1,25-(OH)2D3 (3 and 4). Dots show the regions where the 25-oxygen of each compound can occupy. Black, 1; cyan, 2; red, OCT;
green, 3; magenta, 4. (a) Dot map of 1, 2, and OCT. (b) Dot map of 1-4.
10
four regions (A, G, EA, and EG); the front A and EA
regions were suggested to be important for high potency
of vitamin D (Figure 1a). Furthermore, by applying this
theory to most of the known active vitamin D analogues,
we found a new active region F occupied by the side
chains of vitamin D analogues having 22-oxa, 22,23-
didehydro, 18-nor, or 16-ene modification (Figure 1a, red
dots).7 Altogether, in terms of the space region, the
orders of activity were found as follows: affinity for the
VDR, EA > A > F > G > EG; affinity for vitamin D
binding protein, A . G, EA, EG; target gene transac-
tivation, EA > F > A > EG g G; cell differentiation,
EA > F > A > EG g G; bone calcium mobilization, EA
> G g A > F g EG; and intestinal calcium absorption,
EA ) A g G . EG.8
From these studies, we learned that introduction of
an alkyl group, specifically a methyl or ethyl group, at
C(22) is effective in restricting the side chain conforma-
tion and inducing high activity. In this paper, to
examine further the effect of the double bond at C(23)
and the length of the side chain, we designed and
synthesized two new 22-substituted vitamin D ana-
logues, (22R)- and (22S)-22-ethyl-1,25-dihydroxy-23,24-
didehydro-24-dihomo-20-epivitamin D3 [22-Et-23-ene-
24-dihomo-20-epi-1,25-(OH)2D3, 3 and 4] and evaluated
their biological activities. As predicted, the 22R isomer
3 has higher biological potency than 1, although its
affinity for the VDR is lower (one-seventh) than that of
1.
D3 followed by force field energy minimization. The
dihedral angles at C(16,17,20,22), C(17,20,22,23), C(20,-
22,23,24), and C(23,24,24a,24b) of the global minimum
conformation are as follows: -175.6, 173.0, -88.0, and
112.9°, respectively, for 3, and 178.5, -53.6, 136.2, and
102.4°, respectively, for 4. In these compounds with a
23,24-double bond, the side chain direction is deter-
mined by the configuration at C(22): the double bond
points to the same direction as the hydrogen at C(22).
The results of the conformational analysis are shown
as a dot map (Figure 1b). In the dot map, the global
minimum conformations of 3 and 4 are drawn with a
stick model, with the side chain parts omitted for clarity,
and superimposed. The positions of the 25-oxygen atom
of all possible side chain conformers of 3 (green) and 4
(magenta) were plotted as dots and overlaid with the
dot map of 1 (black) and 20-epi-1,25-(OH)2D3 (2) (cyan).6
Compound 3 occupies areas EA to F. The double bond
at C(23) acts to direct the side chain of 3 to the front
region. Compound 4 occupies the area behind the EG
region (Figure 1b). This analysis predicts compound 3
to be highly potent but 4 to have low potency.
Ster eoselective Syn th esis of 3 a n d 4. The key step
in the synthesis of compounds 3 and 4 is the introduc-
tion of an ethyl group at C(22). We have reported a
highly stereoselective method for introducing an alkyl
group at C(22): stereoselective conjugate addition of
organocuprates to steroidal 22-en-24-ones and 22-en-
24-oates in the presence of TMSCl and hexamethyl-
phosphoramide.11 Using the method, we successfully
synthesized four diastereomers at C(20) and C(22) of
22-methyl-1,25-(OH)2D3.6 A drawback of this method is
that organocuprates do not efficiently add to 20-epi-22Z-
en-24-one; therefore, the stereoselectivity is not high.
In searching for a better method, we examined ortho
ester Claisen rearrangement of 20-epi-22-en-24-ols (10).
By Claisen rearrangement of the allylic alcohols (10),
the 22-substituent and 23,24-double bond can be intro-
duced stereoselectively at the same time. Because the
Claisen rearrangement is known to proceed stereose-
lectively,12 the stereochemistry at C(22) of the 22-alkyl
derivatives can be readily deduced if the stereochemis-
try at C(24) of the starting allylic alcohols (10) is known.
The chirality at a carbon atom bearing a secondary
hydroxyl group can be readily determined by a method
Resu lts
Design a n d Con for m a tion a l An a lysis of 3 a n d 4.
The two side chain analogues of vitamin D, 3 and 4,
were designed with regard to the following character-
istics: (i) the ethyl group at C(22) directs the side chain
to the EA (3) or EG (4) regions; (ii) the double bond at
C(23) further restricts the flexibility of the side chain
and has some beneficial effect on potency; (iii) the two
methylene elongation increases the appropriate hydro-
phobic interaction with the receptor and has some effect
in the separation of activities.9 We analyzed the side
chain conformation of the two compounds 3 and 4 by
systematic search in SYBYL.6b The initial conformations
(structures having global minimum energy) of 3 and 4
for the conformational analysis were obtained by modi-
fication of the crystal structure of 25-hydroxyvitamin
using H nuclear magnetic resonance (NMR).13 Thus,
1

Conformationally Restricted Vitamin D Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1827
Sch em e 1
we planned the synthesis of the two compounds (3 and
4) as shown in Scheme 1. The allylic alcohol (10) can be
derived from 22-en-24-one (9), which is, in turn, a
product of aldol condensation of the C(22)-aldehyde 7
with a side chain fragment, 5-methyl-5-hydroxy-2-
hexanone methoxymethyl (MOM) ether (8).
ketone 8 in the presence of LDA in tetrahydrofuran
(THF; -78 °C, then room temperature) (Scheme 2).
Aldol condensation and subsequent dehydration oc-
curred in one pot to give the 22-en-24-one 9 (61%).
Reduction of the enone 9 with NaBH4 in combination
with CeCl3 (MeOH/pyridine) gave the 24R- and 24S-
alcohols 10a ,b in a 28:72 ratio (88%). In trying to
improve the yield of the 24R isomer (10a ), which is
The side chain fragment 8 was readily synthesized
from γ-valerolactone. 22-Aldehyde 7 was treated with
Sch em e 2

1828 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Masuno et al.
Ta ble 1. Biological Activities of Compounds 3 and 4a
compds
VDRb transcriptionc differentiationc,d calcemice
1,25(OH)2D3 100
100
2500
33
100
9800
77
100
140
38
3 (22R)
4 (22S)
5 (22R)f
6 (22S)f
14
0.2
1100
0.4
10 000
3
12 000
8
220
5
a
b
Activities are presented as percent effect of 1. The relative
activities were calculated at 50% displacement. c The relative
d
activities were calculated at ED50
.
HL-60 cell differentiation was
determined by a NBT assay. e Calcemic activity was measured by
the rise in serum calcium. f Biological activities have been reported
previously.8a
transformed to iodide 13a (75%), which was reduced by
NaBH4 in dimethyl sulfoxide (DMSO) to give 14a (71%).
The alcohol 12b was converted to tosylate 13b (55%),
and the tosylate (13b) was reduced by LAH in THF to
afford 14b (41%). The MOM protecting group was
removed by treatment with p-toluenesulfonic acid, and
the resulting provitamins (15a ,b) were converted to
vitamin D (3 and 4) by photolysis followed by thermal
isomerization.
Biologica l Eva lu a tion . The biological activities of
compounds 3 and 4 were evaluated as follows: affinity
for the VDR, transactivation of a target gene,15 induc-
tion of HL-60 cell differentiation,16 and elevation of
serum calcium in vitamin D and Ca deficient rats.17 The
results are summarized in Table 1 and compared with
the activities of the 22-epimeric pair of 22-methyl-20-
epi-1,25-(OH)2D3 (5 and 6)8a and of 1. Isomer 3, with
the 22R configuration, showed much higher potency
than isomer 4, with the 22S configuration, in all
activities examined. The results are in good agreement
with our prediction based on the active space region
concept (Figure 1b) and previous results.7 Particularly,
it is interesting that compound 3 is nearly 100 times
more potent than 1 in inducing cell differentiation,
although its affinity for the VDR is one-seventh of that
of the natural hormone (1). The 22R isomer 3 had lower
affinity for the VDR than did the natural hormone,
probably because its bulky and flexible side chain made
it difficult to enter the hydrophobic pocket of the VDR.
However, it is likely that this compound has a stronger
hydrophobic interaction with the amino acid residues
lining the ligand binding pocket (LBP) once it enters
the LBP. This may explain why compound 3 has higher
biological activity than 1, regardless of its lower affinity
for the VDR. The affinity of a ligand for the VDR and
its transcriptional and cell differentiating activities are
not always parallel. We can understand this discrepancy
by considering the mechanism of transactivation by NRs
(Scheme 3).
F igu r e 2. Determination of stereochemistry at C-24 of 10a ,b
by modified Mosher’s method. The difference of chemical shift
values between R- and S-MTPA esters is shown in Hertz units.
expected to give the final product (3) with higher
biological potency, we examined Zn(BH4)2 reduction.
However, this reduction also gave the two isomers in a
similar ratio (10a :10b, 43:57, 74%).
The stereochemistry of the epimeric 24-alcohols was
determined by modified Mosher’s method.13 A portion
of the epimeric alcohols 10a ,b was separated,14 and each
isomer was converted to its MTPA ester. Chemical shift
differences in the 1H NMR spectra of the two MTPA
esters were determined, and the results are summarized
in Figure 2. From these results, we assigned the major
isomer 10b as 24S and the minor isomer 10a as 24R.
A mixture of the epimeric 24-alcohols (10a ,b) was
subjected to ortho ester Claisen rearrangement12 by
treatment with triethyl orthoacetate in the presence of
a catalytic amount of propionic acid in refluxing toluene
to give the desired rearranged products (11a ,b) in good
yield (74%). In the Claisen rearrangement of the pure
epimers (10a ,b), the reaction proceeded stereoselec-
tively, 10a giving exclusively 11a and 10b giving
exclusively 11b. The configuration at C(22) of 11a ,b was
deduced on the basis of the accepted mechanism of
Claisen rearrangement: the 3,3-sigmatropic rearrange-
ment proceeds stereoselectively via a six-membered ring
transition state. We assumed the reaction of 10a to
proceed via I, giving 22R-11a , and that of 10b via II,
giving 22S-11b (Scheme 2).
On the basis of generally accepted mechanism of
transactivation regulated by NRs,18 we postulate a
schematic transactivation pathway shown in Scheme 3.
(i) In ligand free apo NR, the protein is in dynamic
equilibrium between the two major conformations,
inactive conformation I and active conformation without
ligand II; (ii) the ligand enters the LBP of the inactive
conformation I to form III; (iii) binding of the ligand
significantly stabilizes the active conformation IV; (iv)
the LXXLL motif of a coactivator binds to the AF-2
(activation function 2) surface of the stabilized complex
IV to form an NR/ligand/coactivator ternary complex V.
Reduction of 11 with DIBAL gave alcohol 12 in which
the methoxycarbonyl protecting groups of the A-ring
hydroxyl groups have been removed. The 22-epimers
12a ,b were easily separated by simple column chroma-
tography. The hydroxyethyl group was converted to an
ethyl group. The primary alcohol 12a was selectively

Conformationally Restricted Vitamin D Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1829
Sch em e 3. Mechanism of Transactivation by the NRa
Dock in g of An a logu es 3 a n d 4 t o VDR -LBD.
Crystal structures of three VDR-LBD deletion mutant
(∆165-215)/ligand complexes have been reported as
follows: complexes with 1,19 2,20 and 26,27-dimethyl-
24-homo-20-epi-22-oxa-1,25-(OH)2D3 (KH1060).20 The
amino acid residues lining the LBP change their con-
formation only slightly in forming complexes with these
ligands with distinct side chain structures. The only
differences observed between the complex with natural
hormone (1) and those with the 20-epi analogues (2 and
KH1060) are ø1 and ø2 (the torsional angles of the side
chain) of I271 and L313. The long and bulky side chain
of KH1060 is packed tightly in the same space of the
VDR-LBP adopting a near-eclipsed conformation (24.8°)
at C(20)-O(22)-C(23)-C(24).
We examined docking of 3 and 4 in the LBP of the
crystal structure of VDR-LBD (∆165-215). We docked
these compounds manually into the hydrophobic pocket
of VDR as follows. (i) We overlaid the A to CD ring part
of the ligands (3 and 4) with that of 1 in the crystal
structure of the VDR. (ii) The side chain conformation
of the ligand was changed, avoiding collision between
the ligand and the amino acid residues lining the LBP
and placing the 25-OH group of the ligand within a
hydrogen-bonding distance from H305 and H397, which
have been shown to form hydrogen bonds with the 25-
OH of the natural ligand. The ligand in the docking
models thus constructed was energy-minimized by
keeping the protein part fixed. The resulting docking
models are shown in Figure 3. In this model, compound
3 adopts a compact conformation with near local mini-
mum energy (Figure 3a); the dihedral angles are C(16,-
17,20,22), -169.8°; C(17,20,22,23), 157.4°; C(20,22,23,-
24), -151.7°; C(23,24,24a,24b), 91.7°; C(24,24a,24b,25),
-66.9°; C(24a,24b,25,O), 61.9°; and C(20,22,22a,22b),
170.5°.
a
The ligand is depicted as a space-filling model.
Binding of the coactivator is thought to be essential for
the commitment of gene transcription. Affinity of ligands
for the VDR is determined by competitive binding
between radioactive 1 and a substrate for the receptor
in the equilibrium state (II / I / III / IV); therefore,
KD ) ([I] + [II])[ligand]/([III] + [IV]). However, in
transactivation, kinetics from IV to V should be impor-
tant. Then, the potency may be proportional to the
product of the concentrations of IV and CoA; therefore,
potency [IV][CoA]. This difference explains why the
transactivation potency of the substrate does not neces-
sarily parallel its binding potency for the VDR.
As described below in the docking model, the elon-
gated side chain of compound 3 can be packed tightly
inside the LBP. The 22S isomer 4 still has significant
cell differentiating and calcemic activities, although its
affinity for the VDR is rather low, 1/500 of that of 1.
The activity ratios of the 22R and 22S isomer pair (3
and 4) (for example, 70 in VDR affinity and 127 in cell
differentiation) are much lower than those found for the
22-epimeric pair of 22-methylated compounds 5 and 6
(for example, 2750 in VDR affinity and 1500 in cell
differentiation). The higher efficacy of 4 relative to 6
may also be explained by tight hydrophobic interaction
of the bulky and long side chain with the hydrophobic
residues inside the LBP.
It is interesting to note that regardless of the two
carbon elongation, the 25-OH group of 3 can form a
hydrogen bond with H397 as does the 25-OH group of
the natural hormone (1). We confirmed this by evaluat-
ing the transactivation potency of one point mutants,
H397A and H305A. We have previously shown that of
the two residues H397 and H305 within hydrogen-
bonding distance from the 25-OH, H397 plays a major
role whereas H305 has a supporting role in anchoring
the ligand to the LBP, because one point mutant H397A
completely abolished the transactivation potency of VDR
F igu r e 3. Complex models of the VDR-LBD with (a) 22R-Et-23-ene-24-dihomo-20-epi-1,25-(OH)2D3 3 and (b) the 22S isomer 4
(stereoview).

1830 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Masuno et al.
VDR-LBD. Our model including the docking properties
of the natural ligand coincided well with the crystal
structure of the VDR-LBD deletion mutant (∆165-215)
reported at nearly the same time.19
In this study, we successfully designed a biologically
potent vitamin D side chain analogue 3 on the basis of
the active space group concept. From the biological
activities and VDR docking studies of 3 and 4, the
following conclusions were drawn. Introduction of R and
S ethyl groups at C(22) has inverse effects on potency,
in accordance with our theory. However, as compared
with a similar compound without the 22-ethyl and
23-ene modifications [24-homo-20-epi-1,25-(OH)2D3,
EB1231],23 the affinity of 3 for the VDR is considerably
reduced (to approximately one-eighth). Regardless of its
reduced affinity for the VDR, 22R isomer 3 is 100 times
more potent than the natural hormone in cell dif-
ferentiating activity. We explained this discrepancy
using schematically presented NR transactivation mech-
anism (Scheme 3). We postulated that the increased
hydrophobic interaction between the amino acid resi-
dues lining the LBP and the elongated side chain have
a role to stabilize the active conformation IV harboring
the ligand and in turn to drive the complex further to
ternary coactivator complex V. Our postulation was
supported by ligand docking study and mutational
analysis. The compound 3 is docked in the LBD with a
compact form where the ethyl group and the elongated
side chain have tight contacts with the LBP amino acid
residues. We assume that the double bond at the side
chain enables the compact packing of 3. Furthermore,
we demonstrated that the 25-OH group of 3 still can
form a hydrogen bond with H397, as the 25-OH of the
natural hormone does. Altogether, we conclude that two
carbon elongation is a suitable modification to harbor
vitamin D ligand tightly in the LBD. The activity
spectrum of 3 is similar to that of the highly potent
analogue KH1060. Detailed biological studies of this
potential analogue 3 are now under way.
F igu r e 4. Transcriptional activity of WT and mutant (H305A
and H397A) hVDRs. COS-7 cells were cotransfected with WT
or mutant hVDR expression vectors, SPP × 3-TK-Luc as a
reporter plasmid and pRL-CMV vector as an internal control.
Before harvesting, cells were treated with 10-8 M 1, 10-10
M
3, or ethanol vehicle for 18 h. Transactivation was determined
by luciferase activity and normalized to the internal control.
induced by 1 but mutant H305A did not significantly
reduce the potency (80%, Figure 4a).21 Similar results
were obtained with the dihomoanalogue 3: H397A
abolished the transactivation potency of VDR induced
by 3, whereas H305A reduced that potency to only about
70% (Figure 4b). In the docking model of 3, tight
hydrophobic interactions are observed at C(23), C(24),
and C(24a) with V234, at the 22-ethyl group with V300
and H305, and at C(21) with I268. These lines of
evidence explain the high potency of compound 3. On
the other hand, the 22S isomer 4 (Figure 3b) is forced
to adopt a considerably unstable conformation in the
docking model shown.
Discu ssion
It is generally known in the vitamin D field that
analogues with the unnatural 20S configuration have
a much higher potency than the corresponding com-
pounds with the natural 20R configuration. However,
as we reported, 22S methylation of 2 drastically reduces
potency, whereas 22R methylation of the same com-
pound significantly elevates potency.6b,8a In terms of
affinity for the VDR, 22R-methyl-20-epi-1,25-(OH)2D3
(5) has 11-fold higher potency, and the 22S methyl
analogue (6) has 1/250 lower potency than 1.8a This
comprises the base of our active space region theory
described above: the 25-oxygen of the active 22R methyl
analogue 5 occupies the EA region, whereas that of the
inactive 22S analogue 6 occupies the EG region. The
active space group concept in our structure-function
theory of vitamin D is based exclusively on the ligand
structure and is not correlated with the structure of the
VDR-LBP. We recently modeled the VDR-LBD harbor-
ing 1 and were able to explain this active space group
region concept on the basis of the ligand-docking
properties of the VDR.22 Compound 1 is placed in the
VDR-LBP with the side chain directed toward the front
region, and the biologically important 25-OH group is
situated at the place where the A, F, and EA regions
overlap (Figure 1). The conformation of the ligand in
which the side chain is directed toward the G and EG
regions suffers steric congestion with helix 11 of the
Exp er im en ta l Section
Computational analysis and modeling were performed using
the program SYBYL with ADVANCED COMPUTATION and
1
BIOPOLYMER (Tripos inc.). H NMR spectra were recorded
in CDCl3 at 400 MHz, and chemical shifts are reported as δ
units relative to tetramethylsilane or solvent signal as an
internal standard. 13C NMR spectra were recorded in CDCl3
at 100 MHz, and chemical shifts are reported as δ units
relative to solvent signal as an internal standard. High- and
low-resolution mass spectra were obtained on a J EOL J MS-
AX505HA spectrometer at 70 eV. Relative intensities are given
in parentheses in low mass. IR spectra were recorded on a
J ASCO J anssen FTIR spectrometer. UV spectra were recorded
on a BECKMAN DU7500 spectrophotometer. All air and
moisture sensitive reactions were carried out under an argon
atmosphere. The phrase “dried and evaporated” indicates
drying over MgSO4 followed by evaporation of the solvents
under house vacuum.
(22E)-(1S,3R,20S)-1,3-Bis[(m et h oxyca r b on yl)oxy]-25-
[(m et h oxym et h yl)oxy]-24a ,24b -d ih om o-5,7,22-ch olest a -
tr ien -24-on e (9). To a solution of diisopropylamine (122 µL,
0.87 mmol) in dry THF (1 mL) was added dropwise n-BuLi
(1.6 M in hexane, 410 µL, 0.65 mmol) at -78 °C, and the
solution was stirred at that temperature for 15 min. To this
LDA solution was added a solution of the ketone 8 (114 mg,
0.65 mmol) in dry THF (1 mL) at -78 °C, and the mixture
was stirred for 15 min. To a solution of 22-aldehyde 7 (104
mg, 0.22 mmol) in dry THF (1 mL) was added the resulting

Conformationally Restricted Vitamin D Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1831
enolate solution at -78 °C. Then, the mixture was stirred at
-78 °C for 1 h, at -20 °C for 1 h, and at 0 °C for 2.5 h. The
reaction mixture was quenched with aqueous NH4Cl and
extracted with AcOEt. The organic layer was washed with
water, dried, and evaporated. The residue was chromato-
graphed on silica gel (30 g) with 5-30% AcOEt-hexane to give
enone 9 (85 mg, 61%). 1 H NMR: δ 0.58 (s, 3 H, H-18), 0.98 (s,
3 H, H-19), 1.01 (d, 3 H, J ) 6.6 Hz, H-21), 1.24 (s, 6 H, H-26
and -27), 3.36 (s, 3 H, CH3OCH2), 3.77 and 3.79 (s, each 3 H,
CH3OCO), 4.69 (s, 2 H, CH3OCH2), 4.80 (m, 1 H, H-1), 4.90
(m, 1 H, H-3), 5.37 and 5.67 (m, each 1 H, H-6 and -7), 6.07
(d, 1 H, J ) 15.8 Hz, H-23), 6.74 (dd, 1 H, J ) 15.8, 9.8 Hz,
H-22). IR (neat): 2957, 1746, 1443, 1254 cm-1. MS m/z: 584
(M+ - MeOH, 7), 554 (62), 478 (72), 402(90), 368 (21), 278 (40),
256 (57), 178 (84), 125 (59), 69 (100). HRMS calcd for C34H48O8
(M+ - MeOH), 584.3349; found, 584.3373.
(22E)-(1S,3R,20S)-24a,24b-Dih om o-5,7,22-ch olestatr ien e-
1,3,24,25-tetr ol 1,3-bis(m eth yl ca r bon a te) 25-Meth oxy-
m eth yl Eth er (10). Red u ction w ith Na BH4-CeCl3. To a
solution of enone 9 (454 mg, 0.73 mmol) in pyridine (0.6 mL)
and methanol (2.4 mL) were added CeCl3‚7H2O (412 mg, 1.10
mmol) and NaBH4 (42 mg, 1.10 mmol) at 0 °C, and the mixture
was stirred for 1 h. Then, CeCl3‚7H2O (276 mg, 0.74 mmol)
and NaBH4 (28 mg, 0.74 mmol) were added again, and the
mixture was stirred for 3 h. The reaction mixture was diluted
with AcOEt, washed with aqueous HCl and water, dried, and
evaporated. The residue was chromatographed on silica gel
(15 g) with 5-20% AcOEt-benzene to yield alcohol 10 (400
mg, 88%, 10a :10b ) 28:72) as a mixture of 24-epimers.
Red u ction w ith Zn (BH4)2. A solution of ZnCl2 (0.5 g, 3.7
mmol) in ether (6 mL) was refluxed until all ZnCl2 was
dissolved. The solution was cooled to room temperature.
NaBH4 (0.35 g, 9.2 mmol) was added, and the mixture was
stirred overnight. A supernatant of the reaction mixture was
collected and used as a Zn(BH4)2 solution at the following
reaction. To a solution of enone 9 (505 mg, 0.82 mmol) in ether
(0.5 mL) was added Zn(BH4)2 solution in ether (6 mL) at 0 °C,
and the mixture was stirred at that temperature for 1.5 h.
The reaction mixture was poured into water and extracted
with AcOEt. The aqueous layer was extracted again with
AcOEt after acidification. The combined organic layer was
washed with brine, dried, and evaporated. The residue was
chromatographed on silica gel (20 g) with 20% AcOEt-hexane
to yield 24-alcohol 10 (376 mg, 74%, 10a :10b ) 43:57) as a
mixture of 24-epimers. A part of the mixture was separated
by column chromatography (10% AcOEt-hexane) to give pure
epimers 10a ,b.
was diluted with benzene, washed with aqueous NaHCO3 and
water, dried, and evaporated. The residue was chromato-
graphed on silica gel (30 g) with 2-30% AcOEt-benzene to
give the Claisen rearrangement product (11) (416 mg, 74%,
11a :11b ) 37:63) as a mixture of 22-isomers.
Rea ction of 24-Ep im er 10a . 24R-alcohol 10a (52 mg, 0.084
mmol), triethyl orthoacetate (460 µL, 2.5 mmol), and propionic
acid (0.3 µL, 4.0 µmol) in toluene (0.5 mL) were treated
according to the procedure described above to yield rearrange-
ment product 11 (31 mg, 54%, 11a :11b ) 94:6). 11a : 1H
NMR: δ 0.65 (s, 3 H, H-18), 0.77 (d, 3 H, J ) 6.8 Hz, H-21),
1.02 (s, 3 H, H-19), 1.21 (s, 6 H, H-26 and -27), 1.23 (t, 3 H,
J ) 7.1 Hz, CH3CH2), 3.36 (s, 3 H, -OCH3), 3.77 and 3.79 (s,
each 3 H, -COOCH3), 4.10 (q, 2 H, J ) 7.1 Hz, CH3CH2), 4.69
(s, 2 H, OCH2O), 4.83 (m, 1 H, H-1), 4.90 (m, 1 H, H-3), 5.39
(m, 2 H, H-23 and -24), 5.37 and 5.68 (m, each 1 H, H-6 and
-7). 13C NMR: δ 12.0, 13.9, 14.3, 16.0, 20.4, 22.9, 26.3, 27.2,
27.3, 31.7, 34.0, 35.6, 37.7, 38.0, 39.4, 41.2, 41.6, 41.7, 43.0,
52.3, 54.4, 54.6, 54.8, 55.1, 60.1, 72.2, 76.0, 78.4, 91.0, 115.5,
122.1, 131.0, 131.8, 133.9, 141.0, 155.0, 173.3. MS m/z: 656
(M+ - MeOH, 3), 626 (5), 550 (36), 474 (100), 279 (48), 249
(45), 209 (32). HRMS calcd for C38H56O9 (M+ - MeOH),
656.3924; found, 656.3950.
Rea ction of 24-Ep im er 10b. 24S-alcohol 10b (41 mg, 0.066
mmol), triethyl orthoacetate (364 µL, 1.99 mmol), and propi-
onic acid (0.25 µL, 3.3 µmol) in toluene were treated according
to the procedure described above for Claisen rearrangement
to yield 11 (26 mg, 57%, 11a :11b ) 2:98). 11b: 1H NMR: δ
0.63 (s, 3 H, H-18), 0.76 (d, 3 H, J ) 6.7 Hz, H-21), 1.01 (s, 3
H, H-19), 1.22 (s, 6 H, H-26 and -27), 1.24 (t, 3 H, J ) 7.0 Hz,
CH3CH2-), 3.36 (s, 3 H, CH2OCH3), 3.77 and 3.80 (s, each 3 H,
OCOCH3), 4.11 (m, 2 H, CH3CH2O), 4.70 (s, 2 H, OCH2O), 4.84
(m, 1 H, H-1), 4.90 (m, 1 H, H-3), 5.39 (m, 2 H, H-22 and -23),
5.38 and 5.68 (m, each 1 H, H-6 and -7). IR (neat): 2961, 2878,
1753, 1443 cm-1. MS m/z: 656 (M+ - MeOH, 3), 626 (5), 550
(29), 474 (100). HRMS calcd for C38H56O9 (M+ - MeOH),
656.3924; found, 656.3918.
(23E)-(1S,3R,20S)-22-(2-Hydr oxyeth yl)-24a,24b-dih om o-
5,7,23-ch olestatr ien e-1,3,25-tr iol 25-Meth oxym eth yl Eth er
(12). To a solution of ester 11 (118 mg, 0.17 mmol, 11a :11b )
27:73) in dry CH2Cl2 (1 mL) was added DIBAL (1 M in hexane,
2.0 mL, 2.0 mmol) at -20 °C, and the mixture was stirred at
that temperature for 1 h. The reaction mixture was quenched
with aqueous NH4Cl and stirred at 0 °C for 20 min. The
mixture was extracted with AcOEt. The extracts were washed
with brine, dried, and evaporated. The residue was chromato-
graphed on silica gel (6 g) with 5% EtOH-CH2Cl2 to give 22S
isomer 12b (60 mg, 66%) and 22R isomer 12a (23 mg, 26%) in
this order.
12b. 1H NMR: δ 0.63 (s, 3 H, H-18), 0.80 (d, 3 H, J ) 5.8
Hz, H-21), 0.92 (s, 3 H, H-19), 1.22 (s, 6 H, H-26 and -27),
3.37 (s, 3 H, CH3OCH2), 3.63 (m, 2 H, CH2OH), 3.78 (m, 1 H,
H-1), 4.06 (m, 1 H, H-3), 4.70 (s, 2 H, OCH2O), 5.33-5.44 (s,
3 H, H-7, -22 and -23) 5.73 (m, 1 H, H-6). 13C NMR: δ 12.6,
13.4, 16.2, 20.9, 22.7, 26.2, 26.9, 27.6, 36,7, 37.5, 38.0, 38.5,
39.1, 39.4, 40.0, 41.2, 42.0, 42.3, 43.4, 52.8, 54.3, 55.1, 61.6,
65.4, 72.8, 76.1, 91.0, 115.2, 122.1, 130.2, 132.3, 135.9, 141.3.
IR (neat): 3378, 2934, 2878, 1450, 1367 cm-1. MS m/z: 530
(M+, 3), 498 (12), 468 (48), 450 (72), 432 (46), 267 (100). HRMS
calcd for C32H50O4 (M+ - MeOH), 498.7309; found, 498.3696.
10a . 1H NMR: δ 0.59 (s, 3 H, H-18), 0.95 (d, 3 H, J ) 6.5
Hz, H-21), 0.98 (s, 3 H, H-19), 1.23 (s, 6 H, H-26 and -27),
3.37 (s, 3 H, -OCH3), 3.77 and 3.78 (s, each 3 H, -COOCH3),
4.04 (m 1 H, H-24), 4.71 (s, 2 H, OCH2O), 4.81 (m 1 H, H-1),
4.89 (m, 1 H, H-3), 5.40 (dd, 1 H, J ) 15.3, 6.9 Hz, H-23), 5.54
(dd, 1 H, J ) 15.3, 9.1 Hz, H-22), 5.37 and 5.68 (m, each 1 H,
H-6 and -7). IR (neat): 3478, 2955, 1746, 1443 cm-1. MS
m/z: 618 (M+, 0.2), 586 (0.6), 556 (3), 510 (5), 480 (24), 404
(100), 249 (40), 125 (75). HRMS calcd for C33H48O7 (M+
MOMOH), 556.3400; found, 556.3409.
-
10b. 1H NMR: δ 0.60 (s, 3 H, H-18), 0.94 (d, 3 H, J ) 6.5
Hz, H-21), 0.99 (s, 3 H, H-19), 1.23 (s, 6 H, H-26 and -27),
3.37 (s, 3 H, -OCH3), 3.77 and 3.78 (s, each 3 H, -COOCH3),
4.04 (m, 1 H, H-24), 4.71 (s, 2 H, OCH2O), 4.81 (m, 1 H, H-1),
4.89 (m, 1 H, H-3), 5.39 (dd, 1 H, J ) 15.4, 6.3 Hz, H -23),
5.53 (dd, 1 H, J ) 15.4, 9.0 Hz, H-22), 5.36 and 5.68 (m, each
12a . 1H NMR: δ 0.63 (s, 3 H, H-18), 0.80 (d, 3 H, J ) 5.9
Hz, H-21), 0.92 (s, 3 H, H-19), 1.22 (s, 6 H, H-26 and -27),
3.37 (s, 3 H, CH3OCH2), 3.50 and 3.62 (m, each 1 H, CH2OH),
3.78 (m, 1 H, H-1), 4.20 (m, 1 H, H-3), 4.70 (s, 2 H, OCH2O),
5.40 (m, 3 H, H-7, -22 and -23), 5.61 (m, 1 H, H-6). 13C
NMR: δ 11.9, 13.7, 16.2, 20.8, 22.6, 26.3, 27.2, 37.7, 38.3, 41.0,
41.4, 41.8, 42.6, 52.3, 55.1, 73.7, 76.0, 91.0, 114.9, 122.0, 130.6,
1 H, H-6 and -7). IR (neat): 3457, 2957, 1746, 1443 cm-1
.
MS m/z: 618 (M+, 0.4), 586 (1), 556 (3), 510 (6), 480 (26), 404
(100), 386 (31), 249 (45), 125 (70). HRMS calcd for C34H50O8
(M+ - MeOH), 586.3506; found, 586.3510.
133.6, 136.3, 141.3. IR (neat): 3360, 2932, 1460, 1368 cm-1
.
(23E)-(1S,3R,20S)-22-(Eth oxyca r bon yl)m eth yl-24a ,24b-
d ih om o-5,7,23-ch olesta tr ien e-1,3,25-tr iol 1,3-Bis(m eth yl
ca r bon a te) 25-Meth oxym eth yl Eth er (11). To a solution
of 24-alcohol (10) (506 mg, 0.82 mmol, 10a :10b ) 42:58) in
toluene (4 mL) were added triethyl orthoacetate (4.5 mL, 24.5
mmol) and 1% propionic acid-toluene (300 µL, 0.041 mmol),
and the mixture was refluxed for 3 h. The reaction mixture
MS m/z: 530 (M+, 3), 498 (10), 468 (43), 450 (48), 432 (8), 394
(15), 267 (100).
(23E)-(1S,3R,20S,22R)-22-(2-Iodoeth yl)-24a,24b-dih om o-
5,7,23-ch olestatr ien e-1,3,25-tr iol 25-Meth oxym eth yl Eth er
(13a ). To a solution of 22R isomer 12a (20 mg, 0.037 mmol),
triphenylphosphine (20 mg, 0.076 mmol), and imidazole (7 mg,

1832 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Masuno et al.
0.10 mmol) in THF (0.4 mL) was added iodine (14 mg, 0.055
mmol) at 0 °C and stirred at the temperature for 1 h. The
reaction mixture was quenched with aqueous Na2S2O5 and
extracted with AcOEt. The extracts were washed with water,
dried, and evaporated. The residue was chromatographed on
silica gel (3 g) with 2% EtOH/CH2Cl2 to give iodide 13a (18
mg, 75%) and starting compound 12a (2.6 mg, 13%). 13a : 1H
NMR: δ 0.63 (s, 3 H, H-18), 0.80 (d, 3 H, J ) 6.8 Hz, H-21),
0.95 (s, 3 H, H-19), 1.22 (s, 6 H, H-26 and -27), 2.95 and 3.22
(m, each 1 H, CH2I), 3.37 (s, 3 H, -OCH3), 3.78 (m, 1 H, H-1),
4.07 (m, 1 H, H-3), 4.71 (s, 2 H, OCH2O), 5.26 (dd, 1 H, J )
15.2, 8.8 Hz, H-23), 5.49 (dt, 1 H, J ) 15.2, 6.6 Hz, H-24), 5.38
and 5.73 (m, each 1 H, H-6 and -7). 13C NMR: δ 5.6, 12.2,
14.2, 16.3, 20.8, 22.9, 26.3, 26.8, 27.3, 33.0, 38.0, 38.6, 38.7,
39.7, 40.0, 41.8, 42.3, 43.3, 47.0, 52.1, 54.4, 55.1, 65.5, 72.9,
76.0, 91.0, 115.3, 122.0, 131.6, 132.2, 136.0, 141.0. MS m/z:
640 (M+, 6), 578 (59), 560 (100), 504 (41), 267 (64).
(23E)-(1S,3R,20S,22S)-22-[2-(4-Meth ylben zen e)su lfon yl-
oxy]eth yl-24a,24b-dih om o-5,7,23-ch olestatr ien e-1,3,25-tr iol
25-Meth oxym eth yl Eth er (13b). To a solution of 22S isomer
12b (13 mg, 0.024 mmol) in pyridine (0.25 mL) was added
p-toluenesulfonyl chloride (7 mg, 0.037 mmol) at 0 °C, and the
mixture was stirred at that temperature for 2 h. p-Toluene-
sulfonyl chloride (14 mg, 0.073 mmol) was added again, and
the mixture was stirred for 24 h. The reaction mixture was
quenched with water and extracted with AcOEt. The extracts
were washed with water, dried, and evaporated. The residue
was chromatographed on silica gel (2.5 g) with 5% EtOH-
CH2Cl2 to give tosylate 13b (9.3 mg, 55%). 1H NMR: δ 0.57
(s, 3 H, H-18), 0.73 (d, 3 H, J ) 6.7 Hz, H-21), 0.96 (s, 3 H,
H-19), 1.22 (s, 6 H, H-26 and -27), 2.45 (s, 3 H, CH3-Ph),
3.37 (s, 3 H, CH3O), 3.79 (m, 1 H, H-1), 3.98 and 4.06 (m, each
1 H, CH2OTs), 4.06 (m, 1 H, H-3), 4.71 (s, 2 H, OCH2O), 5.19
(m, 2 H, H-22 and -23), 5.37 and 5.73 (m, each 1 H, H-6 and
-7), 7.34 and 7.78 (d, each 2 H, J ) 8.2 Hz, H-aromatic).
(23E)-(1S,3R,20S,22R)-22-Eth yl-24a ,24b-d ih om o-5,7,23-
ch olestatr ien e-1,3,25-tr iol 25-Meth oxym eth yl Eth er (14a).
To a solution of iodide 13a (21 mg, 0.032 mmol) in DMSO (0.4
mL) was added NaBH4 (6 mg, 0.16 mmol), and the mixture
was stirred at room temperature for 3 h. The reaction mixture
was quenched with water and extracted with AcOEt. The
extracts were washed with brine, dried, and evaporated. The
residue was chromatographed on silica gel (3 g) with 2% EtOH/
CH2Cl2 to yield 22-ethyl derivative 14a (12 mg, 71%). 14a : 1H
NMR: δ 0.64 (s, 3 H, H-18), 0.77 (d, 3 H, J ) 6.3 Hz, H-21),
0.80 (t, 3 H, J ) 7.3 Hz, CH3CH2), 0.95 (s, 3 H, H-19), 1.23 (s,
6 H, H-26 and -27), 3.37 (s, 3 H, CH3O), 3.77 (m, 1 H, H-1),
4.07 (m, 1 H, H-3), 4.71 (s, 2 H, OCH2O), 5.30-5.45 (m, 3 H,
H-7, -23 and -24), 5.73 (dd, 1 H, J ) 5.6, 2.1 Hz, H-6). MS
m/z: 514 (M+, 4), 482 (14), 465 (12), 452 (74), 434 (100).
(23E)-(1S,3R,20S,22S)-22-Eth yl-24a ,24b-d ih om o-5,7,23-
ch olestatr ien e-1,3,25-tr iol 25-Meth oxym eth yl Eth er (14b).
A suspension of LAH (7 mg, 0.18 mmol) in dry THF (0.3 mL)
was refluxed for 15 min and cooled to room temperature. To
this suspension was added a solution of tosylate 13b (6.3 mg,
0.009 mmol) in dry THF (0.1 mL), and the mixture was
refluxed for 20 min. The mixture was cooled and quenched
with THF-H2O (1:1). The mixture was diluted with aqueous
HCl and extracted with AcOEt. The extracts were washed with
water, dried, and evaporated. The residue was chromato-
graphed on silica gel (2.5 mg) with 5% EtOH-CH2Cl2 to yield
22-ethyl derivative 14b (2.1 mg, 44%). 14b: 1H NMR: δ 0.62
(s, 3 H, H-18), 0.75 (d, 3H, J ) 6.9 Hz, H-21), 0.84 (t, 3 H, J
) 7.4 Hz, CH3CH2), 0.95 (s, 3 H, H-19), 1.23 (s, 6 H, H-26 and
-27), 3.37 (s, 3 H, CH3O), 3.78 (m, 1 H, H-1), 4.07 (m, 1 H,
H-3), 4.72 (s, 2 H, OCH2O), 5.25-5.39 (m, 3 H, H-7, -22 and
-23), 5.74 (m, 1 H, H-6). IR (neat): 3380, 2932, 1723, 1451,
1379, 1210, 1146, 1090, 1042, 970, 916 cm-1. MS m/z: 514 (M+,
2), 496 (5), 482 (8), 464 (12), 452 (72), 434 (100). HRMS calcd
for C33H52O3 (M+ - H2O), 496.3916; found, 496.3937.
for 15 min. The reaction mixture was diluted with Et3N and
then evaporated. The residue was diluted with AcOEt, washed
with brine, dried, and evaporated. The residue was chromato-
graphed on silica gel (3 g) with 3% EtOH-CH2Cl2 to yield
provitamin D 15a (6.6 mg, 90%). UV (95% EtOH): λmax 293,
1
282, 271 nm. H NMR: δ 0.64 (s, 3 H, H-18), 0.77 (d, 3 H, J )
6.3 Hz, H-21), 0.80 (t, 3 H, J ) 7.3 Hz, CH2CH3), 0.95 (s, 3 H,
H-19), 1.23 (s, 6 H, H-26 and -27), 3.77 (m, 1 H, H-1), 4.07
(m, 1 H, H-3), 5.31-5.45 (m, 3 H, H-7, -23 and -24), 5.73
(dd, 1 H, J ) 5.6, 2.3 Hz, H-6). 13C NMR: δ 12.1, 12.3, 13.9,
16.2, 20.8, 20.9, 22.9, 27.3, 27.8, 29.3, 38.1, 38.5, 38.6, 40.0,
40.5, 42.3, 43.2, 43.6, 47.0, 52.2, 54.6, 65.5, 72.9, 115.2, 122.1,
130.5, 134.4, 135.8, 141.3. MS m/z: 470 (M+, 11), 452, (100),
434 (32). HRMS calcd for C31H50O3, 470.3760; found, 470.3765.
(23E)-(1S,3R,20S,22S)-22-Eth yl-24a ,24b-d ih om o-5,7,23-
ch olesta tr ien e-1,3,25-tr iol (15b). MOM ether 14b (5.9 mg,
0.011 mmol) was treated with TsOH‚H2O (6.7 mg, 0.035 mmol)
according to the procedure described for 22R isomer 14a to
yield provitamin D 15b (3.4 mg, 63%). UV (95% EtOH): λmax
1
294, 282, 271 nm. H NMR: δ 0.62 (s, 3 H, H-18), 0.75 (d, 3
H, J ) 6.8 Hz, H-21), 0.84 (t, 3 H, J ) 7.4 Hz, CH3CH2), 0.95
(s, 3 H, H-19), 1.23 (s, 6 H, H-26 and -27), 3.77 (m, 1 H, H-1),
4.06 (m, 1 H, H-3), 5.28-5.42 (m, 3 H, H-7, -22 and -23),
5.73 (m, 1 H, H-6). MS m/z: 470 (M+, 19), 452 (100), 434 (42),
297 (26), 279 (36), 227 (26). HRMS calcd for C31H50O3,
470.3760; found, 470.3782.
(5Z,7E,23E)-(1S,3R,20S,22R)-22-Eth yl-24a ,24b-d ih om o-
9,10-seco-5,7,10(19),23-ch olesta tetr a en e-1,3,25-tr iol (3). A
solution of the provitamin D 15a (16.8 mg, 0.036 mmol) in
benzene/EtOH (150:20, 170 mL) was flushed with argon for
20 min and then irradiated at 0 °C under argon with a 100 W
high-pressure mercury lamp (Shigemi Standard, Tokyo) through
a Vycor filter until most of the starting material 15a was
consumed. After the solvent was removed, the residue was
chromatographed on Sephadex LH-20 (20 g) with CHCl3/
hexane/MeOH (70:30:0.7) to give previtamin D (7.31 mg, 44%).
The previtamin D in 95% EtOH (5 mL) was stored in the dark
at room temperature. After 14 days, the solution was evapo-
rated, and the residue was chromatographed on Sephadex LH-
20 (20 g) with CHCl3/hexane/MeOH (70:30:1) to give vitamin
D 3 (3.97 mg, 54%). The purity of 3 was proved to be about
100% by two HPLC systems: 1, YMC-Pack ODS-AM, 4.6 mm
× 150 mm; H2O/MeOH (15:85), 1.0 mL/min, retention time,
11.3 min. 2, LiChrospher Si 60, 4.0 mm × 250 mm; hexane/
CHCl3/MeOH (100:25:10), 1.0 mL/min, retention time, 16.5
1
min. 3: UV (95% EtOH): λmax 264 nm. H NMR: δ 0.54 (s, 3
H, H-18), 0.76 (d, 3 H, J ) 6.4 Hz, H-21), 0.81 (t, 3 H, J ) 7.3
Hz, CH3CH2), 1.22 (s, 6 H, H-26 and -27), 4.23 (m, 1 H, H-3),
4.43 (m, 1 H, H-1), 5.00 and 5.33 (s, each 1 H, H-19), 5.28-
5.44 (m, 2 H, H-23 and -24), 6.02 and 6.38 (d, each 1 H, J )
11.2 Hz, H-7 and -6). 13C NMR: δ 12.2, 12.4, 14.1, 21.1, 22.2,
23.6, 26.6, 27.8, 29.1, 29.3, 39.6, 40.2, 42.8, 43.6, 45.3, 46.1,
47.3, 52.8, 56.2, 66.8, 70.8, 71.1, 111.8, 117.0, 125.0, 130.5,
132.9, 134.3, 143.1, 147.6. IR (neat): 3387, 2959, 2930, 2874
cm-1. MS m/z: 470 (M+, 15), 452 (100), 434 (84), 416 (44), 297
(34), 279 (38), 134 (59). HRMS calcd for C31H50O3, 470.3760;
found, 470.3751.
(5Z,7E,23E)-(1S,3R,20S,22S)-22-Eth yl-24a ,24b-d ih om o-
9,10-seco-5,7,10(19),23-ch olest a t et r a en e-1,3,25-t r iol (4).
Provitamin D 15b (3.17 mg, 6.74 × 10-3 mmol) was converted
to target vitamin D 4 (0.84 mg, 26%) according to the procedure
for 3. The purity of 4 was proved to be about 100% by two
HPLC systems: 1, YMC-Pack ODS-AM, 4.6 mm × 150 mm;
H2O/MeOH (15:85), 1.0 mL/min, retention time, 10.3 min. 2,
LiChrospher Si 60, 4.0 mm × 250 mm; hexane/CHCl3/MeOH
(100:25:10), 1.0 mL/min, retention time, 16.2 min. 4: UV (95%
1
EtOH): λmax 263 nm. H NMR: δ 0.53 (s, 3 H, H-18), 0.74 (d,
3 H, J ) 6.8 Hz, H-21), 0.83 (t, 3 H, J ) 7.4 Hz, CH3CH2),
1.24 (s, 6 H, H-26 and -27), 4.23 (m, 1 H, H-3), 4.43 (m, 1 H,
H-1), 5.00 and 5.33 (s, each 1 H, H-19), 5.28-5.42 (m, 2 H,
H-23 and -24), 6.02 and 6.38 (d, each 1 H, J ) 11.3 Hz, H-7
and -6). IR (neat): 3382, 2959, 2928, 2874 cm-1. MS m/z: 470
(M+, 1), 452 (12), 434 (41), 416 (100), 297 (12), 279 (45). HRMS
calcd for C31H50O3, 470.3760; found, 470.3768.
(23E)-(1S,3R,20S,22R)-22-Eth yl-24a ,24b-d ih om o-5,7,23-
ch olesta tr ien e-1,3,25-tr iol (15a ). To a solution of MOM
ether 14a (8 mg, 0.015 mmol) in 95% EtOH (0.3 mL) was added
TsOH‚H2O (6.8 mg, 0.036 mmol), and the mixture was refluxed

Conformationally Restricted Vitamin D Analogues
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9 1833
(22E)-(1S,3R,20S,24R)-24a ,24b-Dih om o-5,7,22-ch olesta -
tr ien e-1,3,24,25-tetr ol 1,3-Bis(m eth yl car bon ate) 25-Meth -
oxym eth yl Eth er 24-[(R)-Meth oxy(tr iflu or om eth yl)p h en -
yla ceta te] (16a ). To a solution of 24-alcohol 10a (9.2 mg, 0.015
mmol) in dichloromethane (0.2 mL) was added triethylamine
(3.1 µL, 0.022 mmol), 4-(dimethylamino)pyridine (10.3 mg, 0.84
mmol), and (S)-methoxy(trifluoromethyl)phenylacetyl chloride
(5.6 µL, 0.030 mmol) at 0 °C. The solution was stirred at that
temperature for 20 min. The reaction mixture was poured into
ice and water and extracted with dichloromethane. The organic
layer was washed with water, dried, and evaporated. The
residue was chromatographed on silica gel (2.5 g) with 20%
55.1 and 55.3 (CH3O-), 55.6 (C-17), 72.2 (C-3), 75.4 (C-25), 78.3
(C-24), 90.9 (OCH2O), 115.4 (C-7), 122.0 (C-6), 125.0 (C-23),
127.4, 128.3, 129.5 and 132.4 (Phenyl), 133.8 and 141.0 (C-5
and -8), 142.9 (C-22), 154.8 and 154.9 (OCOO-), 165.8 (car-
bonyl of MTPA). 19F NMR: δ -71.9.
Com p et it ive Bin d in g Assa y, Bovin e Th ym u s VDR .
Binding to bovine thymus VDR was evaluated according to
the procedure reported.6b,16 Bovine thymus VDR was pur-
chased from Yamasa Biochemical (Choshi, Chiba, J apan) and
dissolved in 0.05 M phosphate buffer (pH 7.4) containing 0.3
M KCl and 5 mM dithiothreitol just before use. The receptor
solution (500 µL) in an assay tube was incubated with 0.1 nM
[3H]-1,25-(OH)2D3 together with graded amounts of each
vitamin D analogue for 19 h at 4 °C. The bound and free [3H]-
1,25-(OH)2D3 were separated by treating with dextran-coated
charcoal for 20 min at 4 °C. The assay tubes were centrifuged
at 1000g for 10 min. The radioactivity of the supernatant was
counted. These experiments were done in duplicate.
Tr a n scr ip tion a l Activity. 10T1/2 cells were grown in
Dulbecco’s modified Eagle’s medium supplemented with 10%
resin-charcoal-stripped fetal bovine serum at 37 °C in 5% CO2.
Transfection was performed via the calcium-phosphate pre-
cipitation method as described15 with 200 ng of reporter
plasmid (SPP × 3-TK-LUC), 350 ng of pCMX-â-GAL as
internal control, and 200 ng of carrier plasmid pUC18. The
cells were transfected for 12 h, and after the DNA precipitates
were washed, they were incubated for an additional 48 h with
the ligand (10-7, 10-8, 10-9, 10-10, and 10-11 M). Cells were
harvested and were assayed for luciferase and â-galactosidase
activities. All points were duplicated.
1
AcOEt-hexane to give R-MTPA ester 16a (9.7 mg, 78%). H
NMR: δ 0.57 (s, 3 H, H-18), 0.93 (d, 3 H, J ) 6.9 Hz, H-21),
0.97 (s, 3 H, H-19), 1.14 and 1.16 (s, each 3 H, H-26 and -27),
3.32 (s, 3 H, CH3OCH2-), 3.54 (s, 3 H, CH3O-), 3.77 and 3.79
(s, each 3 H, CH3OCO-), 4.62 (s, 2 H, OCH2O), 4.79 (m, 1 H,
H-1), 4.89 (m, 1 H, H-3), 5.36-5.48 (m, 3 H, H-7, -23 and
-24), 5.67 (dd, 1 H, J ) 2.2, 5.7 Hz. H-6), 5.79 (dd, 1 H, J )
14.9, 9.4 Hz, H-22), 7.34-7.53 (m, 5 H, Ph-H). 19F NMR: δ
-71.6.
(22E)-(1S,3R,20S,24R)-24a ,24b-Dih om o-5,7,22-ch olesta -
tr ien e-1,3,24,25-tetr ol 1,3-Bis(m eth yl car bon ate) 25-Meth -
oxym eth yl Eth er 24-[(S)-Meth oxy(tr iflu or om eth yl)p h en -
yla ceta te] (16b). 10a (10.9 mg, 0.018 mmol) was treated with
(R)-methoxy(trifluoromethyl)phenylacetyl chloride according
to the procedure described above to give S-MTPA ester 16b
1
(4.5 mg, 31%). H NMR: δ 0.54 (s, 3 H, H-18), 0.92 (d, 3 H, J
) 6.6 Hz, H-21), 0.97 (s, 3 H, H-19), 1.19 (s, 6 H, H-26 and
-27), 3.33 (s, 3 H, CH3OCH2-), 3.55 (s, 3 H, CH3O-), 3.77 and
3.79 (s, each 3 H, CH3OCO-), 4.66 (s, 2 H, OCH2O), 4.79 (m, 1
H, H-1), 4.89 (m, 1 H, H-3), 5.26 (dd, 1 H, J ) 15.3, 7.9 Hz,
H-23), 5.36-5.45 (m, 2 H, H-7 and -24), 5.68 (m, 1 H, H-6),
5.74 (dd, 1 H, J ) 15.3, 9.4 Hz, H-22), 7.34-7.52 (m, 5 H, Ph-
H).
Differ en tia tion of HL-60 Cells. Human promyelocytic
leukemia cells (HL-60) were cultured in RPMI 1640 medium
(GIBCO) supplemented with 10% heat-inactivated fetal calf
serum in a humidified atmosphere of 5% CO2 in air at 37 °C.
For the evaluation of cell differentiation, HL-60 cells (5 × 106)
were incubated with each vitamin D compound (10-7, 10-8
,
(22E)-(1S,3R,20S,24S)-24a ,24b-Dih om o-5,7,22-ch olesta -
tr ien e-1,3,24,25-tetr ol 1,3-Bis(m eth yl car bon ate) 25-Meth -
oxym eth yl Eth er 24-[(R)-Meth oxy(tr iflu or om eth yl)p h en -
yla ceta te] (17a ). 24S-alcohol 10b (20.2 mg, 0.033 mmol) was
treated with (S)-methoxy(trifluoromethyl)phenylacetyl chloride
according to the procedure described above to give 17a (25.1
mg, 92%). 1H NMR: δ 0.56 (s, 3 H, H-18), 0.92 (d, 2 H, J ) 6.6
Hz, H-21), 0.94 (s, 3 H, H-19), 1.19 (s, 6 H, H-26 and -27),
3.33 (s, 3 H, CH3OCH2), 3.54 (s, 3 H, CH3O), 3.77 and 3.78 (s,
each 3 H, CH3OCOO), 4.66 (s, 2 H, OCH2O), 4.76 (m, 1 H, H-1),
4.89 (m, 1 H, H-3), 5.28-5.40 (m, 3 H, H-7, -23 and -24),
5.62-5.68 (m, 2 H, H-6 and -22), 7.35-7.43 (m, 3 H, Ph-H),
7.50-7.54 (m, 2 H, Ph-H). 13C NMR: δ 12.0 (C-18), 15.9 (C-
19), 20.0 (C-11), 21.2 (C-21), 22.8 (C-15), 26.2 and 26.3 (C-26
and -27), 27.6 (C-12), 29.0 (C-24a), 31.7 (C-2), 35.6 (C-4), 37.2
(C-24b), 37.7 (C-9), 37.8 (C-16), 40.4 (C-20), 41.2 (C-10), 42.8
(C-13), 54.2 (C-14), 54.6, 54.8, 55.1 and 55.3 (CH3O-), 55.7 (C-
17), 72.2 (C-3), 75.5 (C-25), 78.3 (C-24), 78.4 (C-1), 91.0
(OCH2O), 115.4 (C-7), 122.0 (C-6), 124.9 (C-23), 127.5, 128.3,
129.5 and 132.4 (Phenyl), 133.8 and 141.0 (C-5 and -8), 142.1
(C-22), 154.9 and 155.0 (OCOO-), 165.7 (carbonyl of MTPA).
19F NMR: δ -71.8.
10-9, 10-10, 10-11, and 10-12 M) for 3 days. The differentiation
activity was determined by an NBT reduction assay method
as previously reported.16
In Vivo Bon e Ca lciu m Mobiliza tion .8a,16,17 Weanling
male rats were purchased from Shizuoka Laboratories Animal
Center and were fed a vitamin D-deficient low calcium diet
(Teklad, Madison, WI; 0.03% calcium and 0.6% phosphorus)
for 3 weeks. At the end of the 3 week feeding period, they
received a single intravenous dose of either vehicle (ethanol,
50 µL) or 1 µg of active vitamin D analogue dissolved in the
vehicle. Serum calcium was measured at 0, 6, 12, 24, and 48
h after dosing. Serum calcium concentration was determined
by colorimetric assay using o-cresolphthalein complexone
(Calcium C-test Wako, Wako Pure Chemical, Osaka, J apan)
according to the manufacturer’s instructions.
Tr a n sa ctiva tion Assa y of Mu ta n t VDR. Transactivation
potency of mutant VDRs was tested according to the procedure
previously reported.21 COS-7 cells transfected with a reporter
plasmid containing three copies of the mouse osteopontin
VDRE (SPP × 3-TK-Luc), a wild-type (WT) or mutant hVDR
expression plasmid (pCMX-hVDR), and the internal control
plasmid containing sea pansy luciferase expression constructs
(pRL-CMV) were treated with either 10-8 M 1, 10-10 M 3, or
ethanol vehicle and cultured for 18 h. Cells were harvested,
and luciferase activity was measured with a luciferase assay
kit (Toyo Ink, Inc., J apan) as reported.21 All experiments were
done in triplicate.
(22E)-(1S,3R,20S,24S)-24a ,24b-Dih om o-5,7,22-ch olesta -
tr ien e-1,3,24,25-tetr ol 1,3-Bis(m eth yl car bon ate) 25-Meth -
oxym eth yl Eth er 24-[(S)-Meth oxy(tr iflu or om eth yl)p h en -
yla ceta te] (17b). 24S-alcohol 10b (17.5 mg, 0.028 mmol) was
treated with (R)-methoxy(trifluoromethyl)phenylacetyl chlo-
ride according to the procedure described above to give MTPA
1
ester 17b (18.7 mg, 79%). H NMR: δ 0.54 (s, 3 H, H-18), 0.90
(s, 3 H, H-19), 0.94 (d, J ) 6.6 Hz, H-21), 1.13 and 1.14 (s,
each 3 H, H-26 and -27), 3.32 (s, 3 H, CH3OCH2), 3.54 (s, 3
H, CH3O), 3.76 and 3.77 (s, each 3 H, CH3OCOO), 4.61 (s, 2
H, OCH2O), 4.69 (m, 1 H, H-1), 4.87 (m, 1 H, H-3), 5.34-5.43
(m, 3 H, H-7, -23 and -24), 5.67 (m, 1 H, H-6), 5.74 (m, 1 H,
H-22), 7.34-7.40 (m, 3 H, Ph-H), 7.51-7.55 (m, 2 H, Ph-H).
13C NMR: δ 12.10 (C-18), 16.0 (C-19), 20.0 (C-11), 21.2 (C-21),
22.8 (C-15), 26.1 and 26.4 (C-26 and 27), 27.6 (C-12), 28.9 (C-
24a), 31.7 (C-2), 35.6 (C-4), 37.0 (C-24b), 37.7 (C-9), 37.8 (C-
16), 40.4 (C-20), 41.1 (C-10), 42.8 (C-13), 54.2 (C-14), 54.6, 54.8,
Su p p or tin g In for m a tion Ava ila ble: Experimental syn-
thesis of compound 8 and experimental details in Table 1. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Refer en ces
(1) Feldman, D., Glorieux, F. H., Pike, J . W., Eds. Vitamin D.
Academic Press: San Diego, CA, 1997.

1834 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 9
Masuno et al.
(2) (a) Pike, J . W. Vitamin D3 receptors: structure and function in
transcription. Annu. Rev. Nutr. 1991, 11, 189-216. (b) DeLuca,
H. F. Mechanism of action of 1,25-Dihydroxyvitamin D3: 1990
version. J . Bone Miner. Metab. 1990, 8, 1-9. (c) DeLuca, H. F.;
Krisinger, J .; Darwish, H. The vitamin D system: 1990. Kidney
Int. (Suppl.) 1990, 38, S2-S8. (d) Minghetti, P. P.; Norman, A.
W. 1,25(OH)2-vitamin D3 receptors: gene regulation and genetic
circuitry. FASEB J . 1988, 2, 3043-3053.
studies on conjugate addition of organocuprates to acyclic enones
and enolates: simple rule for diastereofacial selectivity. J . Org.
Chem. 1998, 63, 4449-4458.
(12) (a) White, J . D.; Amedio, J . C., J r. Total synthesis of geodiamolide
A, a novel cyclodepsipeptide of marine organ. J . Org. Chem.
1989, 54, 736-738. (b) Daub, G. W.; Shanklin, P. L.; Tata, C.
Stereoselective acid-catalyzed Claisen rearrangements. J . Org.
Chem. 1986, 51, 3402-3405.
(3) Mangelsdorf, D. J .; Thummel, C.; Beato, M.; Herrlich, P.; Schu¨tz,
G.; Umesono, K.; Blumberg, B.; Kastner, P.; Mark, M.; Chambon,
P.; Evans, R. M. The nuclear Receptor superfamily: The second
decade. Cell 1995, 83, 835-839.
(13) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. High-field
FT NMR application of Mosher’s method. The absolute configu-
rations of marine terpenoids. J . Am. Chem. Soc. 1991, 113,
4092-4096.
(14) Because the 24-epimers 10a and 10b were difficult to separate,
the majority of the products were subjected to the next Claisen
rearrangement as an epimeric mixture and separated later at
the 22-hydroxymethyl derivative (12).
(15) Umesono, K.; Murakami, K. K.; Thompson, C. C.; Evans, R. M.
Direct repeats as selective response elements for the thyroid
hormone, retinoic acid, and vitamin D3 receptors. Cell 1991, 65,
1255-1266.
(16) Imae, Y.; Manaka, A.; Yoshida, N.; Ishimi, Y.; Shinki, T.; Abe,
E.; Suda, T.; Konno, K.; Takayama, H.; Yamada, S. Biological
Activities of 24-Fluoro-1R,25-dihydroxyvitamin D2 and its 24-
Epimer. Biochim. Biophys. Acta 1994, 1213, 302-308.
(17) Sicinski, R. R.; Perlman, K. L.; Prahl, J .; Smith, C.; DeLuca, H.
F. Synthesis and biological activity of 1R,25-dihydroxy-18-
norvitamin D3 and 1R,25-dihydroxy-18, 19-dinorvitamin D3. J .
Med. Chem. 1996, 39, 4497-4506.
(18) Freedman, L. P. Increasing the complexity of coactivation in
nuclear receptor signaling. Cell 1999, 97, 5-8.
(19) 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.
(20) Tocchini-Valentini, G.; Rochel, N.; Wurtz, J . M.; Mitschler, A.;
Moras, D. Crystal structures of the vitamin D receptor com-
plexed to superagonist 20-epi ligands. Proc. Natl. Acad. Sci.
U.S.A. 2001, 98, 5491-5496.
(21) Choi, M.; Yamamoto, K.; Masuno, H.; Nakashima, K.; Taga, T.;
Yamada, S. Ligand Recognition by the Vitamin D Receptor.
Bioorg. Med. Chem. 2001, 9, 1721-1730.
(22) Yamamoto, K.; Masuno, H.; Choi, M.; Nakashima, K.; Taga, T.;
Ooizumi, H.; Umesono, K.; Sicinska, W.; VanHooke, J .; DeLuca,
H. F.; Yamada, S. Three-dimensional modeling of and ligand
docking to vitamin D receptor ligand binding domain. Proc. Natl.
Acad. Sci. U.S.A. 2000, 97, 1467-1472.
(23) Binderup, L.; Latini, S.; Binderup, E.; Bretting, C.; Calverley,
M.; Hansen, K. 20-Epi-vitamin D3 analogues: A novel class of
potent regulators of cell growth and immune responses. Biochem.
Pharmacol. 1991, 42, 1569-1575.
(4) Bouillon, R.; Okamura, W. H.; Norman, A. W. Structure-
function relationships in the vitamin
D endocrine system.
Endocrine Rev. 1995, 16, 200-257.
(5) Verstuyf, A.; Segaert, S.; Verlinden, L.; Bouillon, R.; Mathieu,
C. Recent developments in the use of vitamin D analogues. Exp.
Opin. Invest. Drugs 2000, 9, 443-455.
(6) (a) Yamamoto, K.; Takahashi, J .; Hamano, K.; Yamada, S.
Stereoselective synthesis of (22R)- and (22S)-methyl-1R,25-
dihydroxyvitamin D3: Active vitamin D3 analogues with re-
stricted side chain conformation. J . Org. Chem. 1993, 58, 2530-
2537. (b) Yamamoto, K.; Sun, W.-Y.; Ohta, M.; Hamada, K.;
DeLuca, H. F.; Yamada, S. Conformationally restricted ana-
logues of 1R,25-dihydroxyvitamin D3 and its 20-epimer: com-
pounds for study of the three-dimensional structure of vitamin
D responsible for binding to the receptor. J . Med. Chem. 1996,
39, 2727-2737.
(7) Yamada, S.; Yamamoto, K.; Masuno, H.; Ohta, M. Conformation-
function relationship of vitamin D: conformational analysis
predicts potential side-chain structure. J . Med. Chem. 1998, 41,
1467-1475.
(8) (a) Yamamoto, K.; Ooizumi, H.; Umesono, K.; Verstuyf, A.;
Bouillon, R.; DeLuca, H. F.; Shinki, T.; Suda, T.; Yamada, S.
Three-dimensional structure-function relationship of vitamin
D: side chain location and various activities. Bioorg. Med. Chem.
Lett. 1999, 9, 1041-1046. (b) Yamada, S.; Yamamoto, K.;
Masuno, H. Structure-function analysis of vitamin D and VDR
model. Curr. Pharm. Des. 2000, 6, 733-748. (c) Yamada, S.;
Yamamoto, K.; Masuno, H.; Choi, M. Three-dimensional struc-
ture-function relationship of vitamin D and vitamin D receptor
model. Steroids 2001, 66, 177-187.
(9) Perlman, K.; Kutner, A.; Prahl, J .; Smith, C.; Inaba, M.; Schnoes,
H. K.; DeLuca, H. F. 24-Homologated 1,25-dihydroxyvitamin D3
compounds: separation of calcium and cell differentiation activi-
ties. Biochemistry 1990, 29, 190-196.
(10) Toan, T.; Ryan, R. C.; Simon, G. L.; Calabrese, J . C.; Dahl, L.
F.; DeLuca, H. F. Crystal structure of 25-hydroxy-vitamin D3
monohydrate. J . Chem. Soc., Perkin Trans. 2 1977, 393-401.
(11) Yamamoto, K.; Ogura, H.; J ukuta, J .; Inoue, H.; Hamada, K.;
Sugiyama, Y.; Yamada, S. Stereochemical and mechanistic
J M0105631