Synthesis and biological evaluation of 1α,25-dihydroxyvitamin D 3 analogues with aromatic side chains attached at C-17

Article abstract of DOI:10.1016/j.ejmech.2014.08.031

Two new analogues of the steroid hormone 1α,25-dihydroxyvitamin D₃ with aromatic side chains attached at C-17 were designed to investigate their effects on VDR, HL-60 cell differentiation and tumor cell proliferation. These analogues were prepared by the classical photochemical ring opening approach. After the protection of both the 1α- and 3β-hydroxyl in 1α-hydroxydehydroepiandrosterone with TBS groups, followed by bromination with NBS and debromination in the presence of γ-collidine, the diene intermediate was obtained. Hydrazone formation followed by iodine oxidation gave a vinyl iodide. The aromatic side chain at C-17 was introduced via the Negishi coupling of the resulting intermediate with an in situ generated zinc reagent with the substituted aryl bromide (CD-side chain) in the presence of catalytic amount of Pd(PPh₃)₄. After the removal of the TBDMS and MOM protective groups, followed by UV irradiation and the subsequent thermal reaction, the 1α,25-(OH) ₂-D₃ analogues with a substituted phenyl ring attached at C-17 to replace the C-20 and C-21 were prepared. In the VDR competitive binding assay, compounds 2 and 3 almost lost their binding ability, and were only 0.01% and 0.015% as potent as the 1α,25-dihydroxyvitamin D₃. However, compounds 2 and 3 were as potent as 1α,25-(OH)₂-D₃ in inducing HL-60 cell differentiation at concentrations of 30, 100, 300, 1000 nM, respectively. Moreover, compounds 2 and 3 exhibited similar or better antiproliferative potency against MCF-7 human breast cancer cells, the IC ₅₀ values for analogues 2, 3 and the natural hormone were 7.08, 7.56, and 12.5 μM, respectively.

Full text of DOI:10.1016/j.ejmech.2014.08.031

European Journal of Medicinal Chemistry 85 (2014) 569e575
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Synthesis and biological evaluation of 1a,25-dihydroxyvitamin D3
analogues with aromatic side chains attached at C-17
Chao Liu a, Guo-Dong Zhao a, Xinliang Mao b, Tsutomu Suenaga c, Toshie Fujishima c,
,
*
Cheng-Mei Zhang a, Zhao-Peng Liu a
a Institute of Medicinal Chemistry, Key Laboratory of Chemical Biology (Ministry of Education), School of Pharmaceutical Sciences, Shandong University,
Jinan 250012, PR China
b Cyrus Tang Hematology Center, Soochow University, Suzhou, PR China
c Faculty of Pharmaceutical Sciences at Kagawa Campus, Tokushima Bunri University, Shido, Sanuki, Kagawa 769-2193, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Two new analogues of the steroid hormone 1a,25-dihydroxyvitamin D3 with aromatic side chains
attached at C-17 were designed to investigate their effects on VDR, HL-60 cell differentiation and tumor
cell proliferation. These analogues were prepared by the classical photochemical ring opening approach.
Received 26 February 2014
Received in revised form
8 August 2014
Accepted 8 August 2014
Available online 9 August 2014
After the protection of both the 1
a
- and 3
b
-hydroxyl in 1
a-hydroxydehydroepiandrosterone with TBS
groups, followed by bromination with NBS and debromination in the presence of
g
-collidine, the diene
intermediate was obtained. Hydrazone formation followed by iodine oxidation gave a vinyl iodide. The
aromatic side chain at C-17 was introduced via the Negishi coupling of the resulting intermediate with an
in situ generated zinc reagent with the substituted aryl bromide (CD-side chain) in the presence of
catalytic amount of Pd(PPh3)4. After the removal of the TBDMS and MOM protective groups, followed by
UV irradiation and the subsequent thermal reaction, the 1a,25-(OH)2-D3 analogues with a substituted
phenyl ring attached at C-17 to replace the C-20 and C-21 were prepared. In the VDR competitive binding
Keywords:
Vitamin D
Vitamin D analogues
Vitamin D receptor
Antiproliferative
HL-60 cell differentiation
assay, compounds 2 and 3 almost lost their binding ability, and were only 0.01% and 0.015% as potent as
the 1a,25-dihydroxyvitamin D3. However, compounds 2 and 3 were as potent as 1a,25-(OH)2-D3 in
inducing HL-60 cell differentiation at concentrations of 30, 100, 300, 1000 nM, respectively. Moreover,
compounds 2 and 3 exhibited similar or better antiproliferative potency against MCF-7 human breast
cancer cells, the IC50 values for analogues 2, 3 and the natural hormone were 7.08, 7.56, and 12.5 mM,
respectively.
© 2014 Elsevier Masson SAS. All rights reserved.
1. Introduction
rickets. A number of 1a,25-(OH)2-D3 analogues also show
extremely promising prospects for the treatment of cancer, AIDS,
rheumatoid arthritis, inflammatory bowel diseases, type 1 diabetes,
and Alzheimer's disease [4e6]. However, the use of calcitriol as a
drug to treat hyperproliferative disorders is limited by its undesired
hypercalcemic side effects [7]. Therefore, extensive efforts have
The hormonal metabolite of vitamin D, 1a,25-dihydroxyvitamin
D3 [1a,25-(OH)2-D3 or 1,25D] or calcitriol (1) (Fig. 1), initiates
numerous biological functions required for human health. Besides
its typical role in calcium and phosphorus homeostasis, calcitriol
also regulates cell growth, differentiation, apoptosis, and adaptive/
innate immune responses through the rapid activation of signal
transduction pathways as well as the classic transcriptional acti-
vation pathways that require the nuclear vitamin D receptor (VDR)
[1e3]. 1a,25-(OH)2-D3 and its analogues have been used in the
clinic for the treatment of psoriasis, renal osteodystrophy, osteo-
porosis, secondary hyperparathyroidism, and vitamin D-resistant
been made in the design of novel 1a,25-(OH)2-D3 analogues that
could dissociate between antiproliferative and/or prodifferentiat-
ing action and calcemic effects. Modifications on the A and/or CD
rings or the aliphatic chain of the natural ligand have led to some
potent vitamin D analogues that mediate transcriptional activity
with a magnitude at least 10-fold higher than the natural ligand
with identical or lower calcemic properties [8e14]. For example,
the inversion of the configuration at C-20 usually resulted in
increased activity both in vivo and in vitro [15]. The replacement of
C-21 methyl group with a cyclopropyl ring at C-20 of 1
D3 reduced the binding affinity for VDR and DBP, slightly improved
a,25-(OH)2-
* Corresponding author.
E-mail addresses: liuzhaop@sdu.edu.cn, zhaopengliu@hotmail.com (Z.-P. Liu).
http://dx.doi.org/10.1016/j.ejmech.2014.08.031
0223-5234/© 2014 Elsevier Masson SAS. All rights reserved.

570
C. Liu et al. / European Journal of Medicinal Chemistry 85 (2014) 569e575
the resulting zinc derivative was allowed to react with a preformed
mixture of iodide 11 and Pd(PPh3)4 to give compounds 12a and 12b,
which were treated sequentially with TBAF (tetrabutylammonium
fluoride) and TsOH (p-toluenesulfonic acid) to remove the TBDMS
and MOM protective groups to give the provitamins 13a and 13b in
31% and 30% yields (from 10), respectively. Finally, these pro-
vitamins were converted to the desired vitamin D analogues 2 and
3 by the biomimetic sequence of photoirradiation with a high-
pressure mercury lamp, followed by thermal isomerization and
HPLC purification, in 16% and 14% yields, respectively.
Fig. 1. 1a,25-(OH)2-D3 (1) and its analogues 2 and 3.
3. Pharmacology
the transactivational activity, and lowered the calcemic effects by
10-fold [16,17]. ,25(OH)2-16-ene-20-cyclopropylvitamin D3
proved to be several fold more potent than the natural hormone
,25-(OH)2-D3 as an anti-inflammatory agent [18]. Okamura's
group reported a series of 1 ,25-(OH)2-D3 analogues incorporating
a phenyl ring at the aliphatic side chains, and found some ana-
logues were just as potent as 1 ,25-(OH)2-D3 in inducing HL-60
leukemic cell differentiation or inhibiting its proliferation, but
were no toxic to the proliferation of normal human myeloid stem
cells [19]. Many other side chain modifications and/or 16-ene
3.1. Competitive VDR binding assay
1
a
To examine the binding affinity of compounds 2 and 3 to the
VDR, competitive VDR binding assay using bovine thymus was
carried out by the standard method [30]. The relative potency of the
analogues was calculated from the concentration required to
1
a
a
displace 50% of the [3H]-1
ceptor compared with the activity of 1
a
,25-dihydroxyvitamin D3 from the re-
,25-dihydroxyvitamin D3,
a
a
which was assigned as 100 by definition.
introduction also resulted in a number of 1a,25-(OH)2-D3 ana-
3.2. Cell proliferation assay
logues that exhibited more potent pro-differentiation, anti-
proliferative effects and transactivation potency, and some of them
have the potential as anticancer agents [20e26]. Here we first
report the synthesis and biological evaluations of two new vitamin
D analogues 2 and 3 (Fig. 1) with a substituted phenyl ring attached
at C-17 to replace the C-20 and C-21. Compared with the natural
The ability of compounds 2 and 3 to inhibit the proliferation of
human MCF-7 breast cancer cells was examined by the conven-
tional MTT assay. MCF-7 cells were incubated with compounds 2
and 3 at the indicated concentrations for 72 h, and their inhibiting
rates were determined. 1
reference drug.
a,25-Dihydroxyvitamin D3 was used as a
1a,25-(OH)2-D3, this modification leads to the loss of two chiral
centers at C-17 and C-20 in molecules 2 and 3.
3.3. HL-60 cell differentiation assay
2. Chemistry
To evaluate the effects of compounds 2 and 3 on cell differen-
tiation, HL-60 cells were incubated and treated with compound 2, 3
or calcitriol (1) for 48 h at the indicated concentrations. 0.1% DMSO
was used as vehicle. After the treatment, cells were transferred to
slides with cytospin and applied for Wright'seGiemsa staining [31].
Forty microscopic view fields were analyzed and photographed.
The classical strategy, including photo-reaction of provitamin D3
into previtamin D3 and the subsequent thermal conversion of
previtamin D3 into vitamin D3, was applied for the synthesis of
analogues 2 and 3. To achieve this purpose, we first built the aro-
matic side chains through a new synthetic route. As shown in
Scheme 1, the commercial available alcohol, 4a or 4b, was first
converted to an iodide, which was reacted with the sodium salt of
diethyl malonate to give a diester. Upon hydrolysis, compounds 5a
and 5b were obtained in 63% and 65% yields, respectively. Decar-
boxylation and subsequent esterification produced the ethyl esters
6a and 6b in excellent yields. Coupling reaction of 6a or 6b with
methylmagnesium bromide gave the tertiary alcohol, which was
treated with chloromethyl methyl ether (MOMCl) to furnish the
target aromatic side chains 7a or 7b in good yield.
4. Results and discussion
1a,25-Dihydroxyvitamin D3 functions as the ligand for VDR,
with the hormoneereceptor complex inducing calcemic and
phosphatemic effects that result in normal bone mineralization and
remodeling. In the VDR binding assay, compounds 2 and 3
exhibited very low VDR affinity (see Supplementary data). They
were comparable in potency to that of 25-OH-D3, but were only
1a
-Hydroxydehydroepiandrosterone 8 (Scheme 2) was pre-
0.01% and 0.015% as potent as the parent 1a,25-(OH)2-D3 (Table 1).
pared from commercial available dehydroepiandrosterone by a
modified procedure developed by our group, applying recoverable
Pd/C catalyst mediated dehydrogenation of sterols as a key step
[27]. After the TBDMS ether formation to protect the diols, diene 9
It is apparent that introducing a phenyl ring at the C-17 has much
influence on the binding ability to the VDR.
X-ray crystal structure of Moras VDR(LBD)-1,25D complex was
used for molecular docking studies by Sybylex1.3 software (PDB
was obtained via sequential treatment with NBS and
g-collidine
code: 1DB1) [32]. Docking analysis of compounds 2 and 3
[28]. Diene 9 was then reacted with hydrazine hydrate in the
presence of hydrazine sulfate to give hydrazone 10 in 87% yield.
Oxidation of hydrazone 10 with iodine in the presence of a hin-
dered guanidine base yielded the vinyl iodide 11. The iodide 11 was
instable when exposed to air, so it was used for the next reaction
without further purification. Attaching the aromatic side chains to
the C-17 position was successfully achieved through the Negishi
coupling approach [29]. Bromide 7a or 7b was treated with n-
butyllithium at ꢀ78 ꢁC and then with zinc chloride in THF solution,
demonstrated that the A, seco-B, C, and D rings present confor-
mations that are similar to those observed in the presence of the
natural ligand (Fig. 2). The A-ring hydroxyl groups make the same
hydrogen bonds as the hVDR LBD bound to 1a,25-(OH)2-D3 com-
plex, 1-OH with Ser237 and Arg274, 3-OH with Tyr143 and Ser278.
However, the side chain tertiary hydroxyl group in compounds 2
and 3 form hydrogen bonding with His397, while the natural 25-
OH binds to His305. In addition, the phenyl ring at the C-17
makes the side chain longer than that of natural hormone. All these

C. Liu et al. / European Journal of Medicinal Chemistry 85 (2014) 569e575
571
Scheme 1. Synthesis of the aromatic side chains 7a and 7b.
Scheme 2. Synthesis of 1a,25-(OH)2-D3 analogues 2 and 3.
factors might play important roles in reducing the binding affinity
of compounds 2 and 3 to VDR.
chromatin, deep stained cytoplasm. However, compounds 2, 3 or
,25-(OH)2-D3 treatment led to maturing morphological changes,
1a
Despite their low VDR binding affinity, compounds 2 and 3 were
including decreased ratio of nuclear to cytoplasmic, reduced or
disappeared nucleolus, condensed chromatin as well as lobed
nuclei.
also as potent as the native 1
1.25, 2.5, 5.0 M in the inhibition of the proliferation of human
MCF-7 breast cancer cells, but were slightly more potent than
,25-(OH)2-D3 at the relative high concentrations of 10, 20, 40
(Fig. 3). In addition, the IC50 values for compounds 2, 3 and 1 ,25-
(OH)2-D3 were 7.08, 7.56 and 12.5 M, respectively (Table 1).
In the HL-60 cell differentiation inducing assay, compounds 2
and 3 were as potent as 1 ,25-(OH)2-D3 in inducing HL-60 cell
a,25-(OH)2-D3 at the concentrations of
m
1
a
mM
a
5. Conclusion
m
In summary, incorporating a phenyl ring at C-17 to replace the
a
C-20 and C-21 of the native 1a,25-(OH)2-D3 made analogues 2 and
differentiation at concentrations of 30, 100, 300, 1000 nM, respec-
tively. As shown in Fig. 4, the vehicle-treated HL-60 cells showed
morphologic features of blast cells, including high nuclear cyto-
plasmic ratio, one or two nucleoli, non-condensed nuclear
3 almost lose their VDR binding ability, but maintain their anti-
proliferative and prodifferentiating potency, indicating that these
compounds may have therapeutic potential for the treatment of
different hyperproliferative disorders.

572
C. Liu et al. / European Journal of Medicinal Chemistry 85 (2014) 569e575
Table 1
recorded on an API 4000 spectrometer. Column chromatography
was performed on silica gel (300e400 mesh; Anhui Liangchen
Silicon Material Co. Ltd., Anhui, PR China). All reactions involving
oxygen- or moisture-sensitive compounds were carried out under a
dry nitrogen atmosphere. Unless otherwise noted, reagents were
added by syringe. THF was distilled from sodium/benzophenone
immediately prior to use. All tested compounds are >98.5% pure by
HPLC analysis, performed on a Shimadzu LC-20AT instrument, us-
ing the ProntoSIL Eurobond C18 (250 mm ꢂ 4.6 mm) or Agilent
Eclipse XDB-C18 (150 ꢂ 4.6 mm) column.
Results of VDR binding and MCF-7 cell proliferation assay.
Compds
VDR
0.01
0.015
100
0.01
MCF-7 (IC50: mM)
2
3
7.08 0.25
7.56 0.28
12.5 0.19
e
1
a
,25-(OH)2-D3
25-OH-D3
6.1.1. 2-[2-(2-Bromophenyl)ethyl]malonic acid (5a)
To a solution of alcohol 4a (15.0 g, 74.4 mmol) in dichloro-
methane (225 mL) was added triphenylphosphine (24.9 g,
94.9 mmol) and imidazole (6.45 g, 94.7 mmol). After stirring at
room temperature for 20 min, iodine (22.5 g, 88.7 mmol) was
added and the resulting dark brown mixture was stirred for over-
night at 0 ꢁC. A saturated solution of sodium thiosulfate (50 mL)
was added and the mixture was allowed to warm to room tem-
perature. The reaction mixture was extracted with dichloro-
methane, washed with brine, dried over Na2SO4, filtered, and
concentrated under reduce pressure. Hexane was added to the
residue and the white precipitate was filtered off. The filtrate was
concentrated to give the crude 2-bromophenethyl iodide (23.0 g),
which was used without further purification.
To a solution of ethyl malonate (295 mL) was slowly added NaH
(12.0 g, 30.0 mmol) at 0 ꢁC. After stirring for 30 min, a solution of
the crude 2-bromophenethyl iodide (23.0 g) in THF (100 mL) was
added. The reaction mixture was stirred at room temperature for
overnight. After the completion of the reaction indicated by TLC,
aqueous NH4Cl (50 mL) was added. The resulting mixture was
extracted with EtOAc, washed with brine, dried over Na2SO4,
filtered, and concentrated. The crude product was purified by col-
umn chromatography on silica gel (hexane/AcOEt ¼ 50:1) to give
the substituted ethyl malonate (17.0 g) as a colorless oil, which was
dissolved in 40 mL of 50% aqueous ethanol. To this solution, 9.30 g
(165.8 mmol) of potassium hydroxide was added. After refluxing
for 6 h, the reaction mixture was diluted with water and acidified
with 15% hydrochloric acid solution. A colorless precipitate was
collected to afford 5a (13.5 g) in a total yield of 63% from 5a. Mp:
157e158 ꢁC. This data is in agreement with the reported value
(156e158 ꢁC) [33].
Fig. 2. Predicted conformations and binding modes of analogues 2 (yellow), 3 (blue)
and 1 ,25-(OH)2-D3 (red) in the VDR LBP. (For interpretation of the references to color
in this figure legend, the reader is referred to the web version of this article.)
a
6.1.2. 2-[2-(3-Bromophenyl)ethyl]malonic acid (5b)
This compound was prepared from 4b by the same procedure as
described for compound 5a. Yield: 65%; Mp: 161e163 ꢁC. 1H NMR
(600 MHz, CDCl3)
2H), 3.37 (t, 1H, J ¼ 7.2 Hz), 1.67e2.70 (m, 2H), 2.33e2.26 (m, 2H);
13C NMR (150 MHz, DMSO-d6)
170.7, 144.0, 131.1, 130.6, 129.0,
d 7.36 (s, 1H), 7.32e7.34 (m, 1H), 7.13e7.17 (m,
d
127.5, 121.7, 51.0, 32.3, 30.0; HRMS (ESI): Calcd for C11H11BrO4
[MþNa]þ 308.9738 and 310.9718, found [MþNa]þ 308.9752 and
310.9730.
Fig. 3. The inhibition of MCF-7 human breast cancer cells at different concentrations.
Error bars represent standard deviation (SD). *P < 0.05, **P < 0.01, ***P < 0.001.
6.1.3. Ethyl 4-(2-bromophenyl)butanoate (6a)
6. Experimental section
Acid 5a (18.0 g, 62.7 mmol) was heated at 170 ꢁC for 30 min until
the evolution of carbon dioxide ceased. Ethanol (152 mL) was
added, followed by dropwise addition of thionyl chloride (4.9 mL,
67.5 mmol). After stirring at room temperature for 3 h, the solvent
was evaporated in vacuo. The residue was extracted with EtOAc,
washed with brine, dried over Na2SO4, filtered, and concentrated.
The crude product was purified by column chromatography on
silica gel (hexane/AcOEt ¼ 10:1) to give 6a (16.0 g) as a colorless oil.
6.1. Chemistry
Melting points (uncorrected) were determined on an X-6 micro-
melting point apparatus (Beijing Tech. Co., Ltd.) and were uncor-
rected. UV data were recorded on a UV-2550 spectrophotometer
(Shimadzu, Japan). NMR spectra were recorded on Bruker Avance
DRX-600 or Varian INOVA 600 spectrometers operating at 600 (1H)
and 150 or 100 (13C) MHz with TMS as internal standard. HRESIMS
were carried out on an LTQ-Orbitrap XL. ESI-MS spectra were
Yield: 94%. 1H NMR (600 MHz, CDCl3)
d
7.52 (dd, 1H, J ¼ 7.8 Hz,
2.4 Hz), 7.21e7.24 (m, 2H), 7.04e7.07 (m, 1H), 4.11e4.15 (m, 2H),
2.76e2.79 (m, 2H), 2.34e2.37(m, 2H), 1.93e1.99 (m, 2H), 1.25e1.27

C. Liu et al. / European Journal of Medicinal Chemistry 85 (2014) 569e575
573
Fig. 4. Induction of HL-60 cell differentiation by analogues 2, 3 and 1a,25-(OH)2-D3.
(m, 3H); 13C NMR (100 MHz, CDCl3)
d
173.3, 140.8, 132.9, 130.5,
(600 MHz, CDCl3)
d
7.34 (s, 1H), 7.32 (d, 1H, J ¼ 6 Hz), 7.14 (t, 1H,
127.8, 127.4, 124.5, 60.3, 35.3, 33.7, 25.0, 14.3; HRMS (ESI): Calcd for
J ¼ 6 Hz), 7.11 (d, 1H, J ¼ 6 Hz), 4.68 (s, 2H), 3.36 (s, 3H), 2.57 (t, 2H,
C
12H15BrO2 [MþH]þ 271.0334 and 273.0313, found [MþH]þ
J ¼ 8 Hz), 1.64e1.70 (m, 2H), 1.51e1.54 (m, 2H), 1.21 (s, 6H); 13C NMR
271.0342 and 273.0321.
(150 MHz, CDCl3) d 144.9, 131.5, 129.9, 128.9, 127.1, 122.4, 91.0, 76.1,
55.2, 41.5, 36.0, 26.3, 26.3, 25.8; HRMS (ESI): Calcd for C14H21BrO2
[MþNa]þ 323.0623 and 325.0602, found [MþNa]þ 323.0631 and
325.0610.
6.1.4. Ethyl 4-(3-bromophenyl)butanoate (6b)
This compound was prepared from 5b by the same procedure as
described for compound 6a. Colorless oil; yield: 93%. 1H NMR
(600 MHz, CDCl3)
d
7.33 (d, 2H, J ¼ 9 Hz), 7.15 (t, 1H, J ¼ 7.8 Hz), 7.11
6.1.7. (1a,3b)-1,3-Bis(tert-butyldimethylsilyloxy)androsta-5,7-
dien-17-hydrazone (10)
(d, 1H, J ¼ 7.8 Hz), 4.13 (q, 2H, J ¼ 7.2 Hz), 2.62 (t, 2H, J ¼ 7.2 Hz), 2.31
(t, 2H, J ¼ 7.8 Hz), 1.95 (m, 2H), 1.26 (t, 3H, J ¼ 7.2 Hz); 13C NMR
To a solution of ketone 9 (0.56 g, 1.05 mmol) in ethanol (5 mL)
was added hydrazine hydrate (0.22 mL, 4.51 mmol) and catalytic
hydrazine sulfate (1 mg) in water (0.10 mL). After stirring at room
temperature for 18 h, the mixture was poured into cold water, and
the precipitate was collected to give hydrazone 10 (0.50 g, 87%) as a
(100 MHz, CDCl3) d 173.3,143.8,131.5,130.0,129.1,127.2,122.5, 60.4,
34.8, 33.5, 26.3, 14.3; HRMS (ESI): Calcd for C12H15BrO2 [MþH]þ
271.0334 and 273.0313, found [MþH]þ 271.0338 and 273.0317.
white solid. Mp: 147e150 ꢁC. 1H NMR (600 MHz, CDCl3)
d 5.60 (d,
6.1.5. 1-Bromo-2-(4-(methoxymethoxy)-4-methylpentyl)benzene
1H, J ¼ 4.8 Hz), 5.41 (d, 1H, J ¼ 4.8 Hz), 4.79 (brs, 2H), 3.97e4.07 (m,
1H), 3.71 (s, 1H), 2.85 (t, 1H, J ¼ 6 Hz), 2.25e2.40 (m, 5H), 1.90e2.05
(m, 4H), 1.52e1.75 (m, 6H), 1.38e1.46 (m, 1H), 0.93 (s, 3H), 0.88 (s,
9H), 0.87 (s, 9H), 0.81 (s, 3H), 0.10 (s, 3H), 0.06 (s, 3H), 0.05 (s, 6H);
(7a)
To a solution of 6a (8.0 g, 29.5 mmol) in THF (60 mL) at ꢀ10 ꢁ
C
was slowly added methylmagnesium bromide (24.6 mL, 3.0 M in
THF, 73.8 mmol). The mixture was allowed to warm to room tem-
perature and stirred for 8 h. The reaction mixture was quenched
with aqueous NH4Cl, extracted with EtOAc, washed with brine,
dried over Na2SO4, filtered, and concentrated. The obtained crude
tertiary alcohol was dissolved in 70 mL CH2Cl2, followed by addition
of diisopropylethylamine (14.9 mL, 90.1 mmol) and MOMCl (6.5 mL,
84.6 mmol). After stirring at room temperature for 3 h, the reaction
mixture was extracted with CH2Cl2, washed with brine, dried over
Na2SO4, filtered, and concentrated. The residue was purified by
column chromatography on silica (hexane/AcOEt ¼ 50:1) to give 7a
13C NMR (100 MHz, CDCl3)
d 165.6, 138.5, 138.3, 119.8, 115.4, 73.9,
66.3, 52.1, 44.3, 42.9, 40.9, 39.0, 38.0, 33.9, 26.0, 25.9, 24.2, 22.3,
20.4, 18.2, 18.0, 17.0, 16.9, -3.8, -4.3, -4.5, -5.0; HRMS (ESI): Calcd for
C
31H56N2O2Si2 [MþH]þ 545.3959, found [MþH]þ 545.3962.
6.1.8. (1a,3b)-1,3-Bis(tert-butyldimethylsilyloxy)-17-iodoandrosta-
5,7,16-triene (11)
To a stirred solution of iodine (0.50 g, 1.97 mmol) in anhydrous
THF (14 mL) and Et2O (7 mL) at 0 ꢁC was added 1,1,3,3-
tetramethylguanidine (TMG, 0.65 mL, 5.18 mmol). A THF solution
(8 mL) of 10 (0.50 g, 0.92 mmol) was added dropwise into the
iodine solution over 2 h to maintain the reaction temperature at
0 ꢁC. After stirring for 1 h, saturated solution of sodium sulfite was
added, and the reaction mixture was extracted with AcOEt, washed
with brine, dried over Na2SO4, filtered, and concentrated. The crude
iodide 11 (0.59 g) was obtained without further purification.
(7.15 g, 94%) as a colorless oil. 1H NMR (600 MHz, CDCl3)
d 7.52 (d,
1H, J ¼ 7.8 Hz), 7.22 (d, 2H, J ¼ 4.2 Hz), 7.03e7.06 (m, 1H), 4.69 (s,
2H), 3.36 (s, 3H), 2.72 (t, 2H, J ¼ 7.8 Hz), 1.66e1.71 (m, 2H),
1.58e1.61 (m, 2H), 1.21 (s, 6H); 13C NMR (150 MHz, CDCl3)
d 141.8,
132.8, 130.3, 127.5, 127.4, 124.5, 91.0, 76.2, 55.2, 41.4, 36.6, 26.4, 26.3,
24.5; HRMS (ESI): Calcd for C14H21BrO2 [MþNa]þ 323.0623 and
325.0602, found [MþNa]þ 323.0625 and 325.0604.
6.1.6. 1-Bromo-3-(4-(methoxymethoxy)-4-methylpentyl)benzene
(7b)
6.1.9. (1a,3b)-17-[2-(4-Hydroxy-4-methylpentyl)phenyl]androsta-
5,7,16-trien-1,3-diol (13a)
This compound was prepared from 6b by the same procedure as
To a stirred solution of compound 7a (2.07 g, 6.87 mmol) in THF
described for compound 7a. Colorless oil; yield: 92%. 1H NMR
(23 mL) at ꢀ78 ꢁC was added n-butyllithium (2.3 M solution in

574
C. Liu et al. / European Journal of Medicinal Chemistry 85 (2014) 569e575
hexanes, 2.50 mL, 5.73 mmol). After stirring for 7 min, zinc chloride
(0.5 M solution in THF, 15.7 mL, 7.85 mmol) was added. The mixture
was stirred for another 10 min, and then the cooling bath was
removed and the stirring was continued for 30 min to form the zinc
derivative solution. In parallel, the crude iodide 11 (0.59 g) was
dissolved in anhydrous THF (15 mL) and Pd(PPh3)4 (140 mg) was
added. The resulting solution was immediately transferred to the
zinc derivative solution via a double-end needle under nitrogen
atmosphere. The reaction mixture was stirred for 18 h at room
temperature. Aqueous NH4Cl was added, and the mixture was
extracted with AcOEt, washed with brine, dried over Na2SO4,
filtered, and concentrated. The residue was chromatographed on
silica gel (hexane/AcOEt ¼ 50:1) to afford crude 12a (324 mg).
A solution of crude silyl ether 12a and TBAF (680 mg, 2.16 mmol)
in THF (10 mL) was refluxed for 4 h. The solvent was evaporated,
and the residue was extracted with AcOEt, washed with water,
dried over Na2SO4, filtered, and concentrated. The residue was
dissolved in MeOH (7 mL) and TsOH (157 mg, 0.91 mmol) was
added. The reaction mixture was stirred at room temperature for
10 min. Et3N (0.5 mL) was added, and the solvent was evaporated.
The residue was chromatographed on silica gel (hexane/
AcOEt ¼ 1:3) to afford 13a (137 mg) as a white solid in 31% yield
J ¼ 10.8 Hz), 5.56 (s, 1H), 5.37 (s, 1H), 5.06 (s, 1H), 4.48 (t, 1H,
J ¼ 9.6 Hz), 4.25 (t, 1H, J ¼ 9.6 Hz), 2.84 (dd, 1H, J ¼ 3.6, 13.2 Hz),
2.60e2.65 (m, 4H), 2.45 (dt,1H, J ¼ 2.4,11.4 Hz), 2.35 (dd,1H, J ¼ 6.6,
13.2 Hz), 2.21e2.33 (m, 1H), 2.03e2.08 (m, 1H), 1.91e1.96 (m, 1H),
1.81e1.87 (m, 1H), 1.70e1.76 (m, 1H), 1.44e1.68 (m, 8H), 1.21 (s, 6H),
0.86 (s, 3H); 13C NMR (150 MHz, CDCl3)
d 153.2, 147.7, 142.2, 140.8,
137.0, 133.3, 128.9, 128.8, 127.2, 126.8, 124.9, 124.9, 117.2, 111.8, 71.0,
70.8, 66.9, 58.3, 52.2, 45.2, 43.9, 42.88, 35.1, 33.8, 30.4, 29.3, 29.2,
28.6, 27.0, 23.4, 17.1; HRMS (ESI): Calcd for C31H42O3 [MþH]þ,
463.3207, found [MþH]þ 463.3218; UV (EtOH) lmax 262 nm
(ε ¼ 17860).
6.1.12. (1a,3b)-17-[3-(4-Hydroxy-4-methylpentyl)phenyl]-9,10-
secoandrosta-5,7,10(19),16-tetraen-1,3-diol (3)
this compound was prepared from 13b by the same procedure
as described for compound 2 in 14% yield. H NMR (600 MHz, CDCl3)
d
7.18e7.23 (m, 3H), 7.06 (d, 1H, J ¼ 7.2 Hz), 6.40 (d, 1H, J ¼ 11.4 Hz),
6.17 (d, 1H, J ¼ 11.4 Hz), 5.89 (s, 1H), 5.36 (s, 1H), 5.05 (s, 1H), 4.47
(brs, 1H), 4.26 (brs, 1H), 2.84 (dd, 1H, J ¼ 4.2, 12.6 Hz), 2.58e2.63 (m,
4H), 2.33e2.42 (m, 2H), 2.14e2.18 (m, 1H), 2.04e2.07 (m, 2H),
1.91e1.95 (m, 1H), 1.80e1.86 (m, 2H), 1.67e1.76 (m, 3H), 1.51e1.58
from compound 10. 1H NMR (600 MHz, CDCl3)
d
7.25 (d, 1H,
(m, 6H), 1.21 (s, 6H), 0.97 (s, 3H); 13C NMR (150 MHz, CDCl3)
d 154.5,
J ¼ 7.2 Hz), 7.20 (t, 1H, J ¼ 7.2 Hz), 7.13 (t, 1H, J ¼ 7.2 Hz), 7.08 (d, 1H,
J ¼ 7.2 Hz), 5.78 (d, 1H, J ¼ 3 Hz), 5.60 (s, 1H), 5.53 (s, 1H), 4.02e4.13
(m, 1H), 3.77 (s, 1H), 2.82e2.87 (m, 1H), 2.55e2.62 (m, 4H),
2.32e2.42 (m, 3H), 2.26e2.29 (m, 1H), 1.73e1.78 (m, 2H), 1.58e1.66
147.7, 142.2, 142.1, 137.2, 133.3,128.1, 126.9, 126.8, 126.3, 124.9, 124.1,
117.4, 111.8, 71.0, 70.8, 66.9, 58.6, 50.0, 45.2, 43.6, 42.9, 36.4, 35.5,
30.0, 29.3, 29.2, 28.6, 26.3, 23.6, 17.5; HRMS (ESI): Calcd for
C
31H42O3 [MþNH4]þ 480.3466, found 480.3472; UV (EtOH) lmax
(m, 8H), 1.47e1.54 (m, 2H), 1.20 (s, 6H), 0.96 (s, 3H), 0.98 (s, 3H); 13
C
260 nm (ε ¼ 17880).
NMR (100 MHz, CDCl3)
d 158.0, 145.5, 144.5, 142.9, 140.7, 140.3,
136.4, 128.8, 127.8, 124.9, 122.0, 115.0, 72.6, 71.0, 65.4, 56.0, 50.0,
44.0, 42.5, 40.2, 34.5, 33.8, 30.8, 30.6, 29.8, 29.3, 26.9, 21.0, 20.9,
17.5, 16.5; HRMS (ESI): Calcd for C31H42O3 [MþH]þ, 463.3207, found
[MþH]þ 463.3210.
6.2. VDR binding assay
The bovine thymus VDR receptor was obtained from Yamasa
Biochemical (Chiba, Japan) and dissolved in 0.05 M phosphate
buffer (pH 7.4) containing 0.3 M KCl and 5 mM dithiothreitol
immediately prior to use. The receptor solution (500 mL) was pre-
6.1.10. (1a,3b)-17-[3-(4-Hydroxy-4-methylpentyl)phenyl]androsta-
5,7,16-trien-1,3-diol (13b)
This compound was prepared from 10 by the same procedure as
mixed with 50 mL of an ethanol solution of
dihydroxyvitamin D3 or an analogue at various concentrations for
60 min at 25 ꢁC, prior to the before addition of [3H]-1
,25-
dihydroxyvitamin D3 (50 mL). The receptor mixture was then left
to stand overnight with 0.1 nM [3H]-1
,25-dihydroxyvitamin D3 at
ꢁC. The bound and free [3H]-1
,25-dihydroxyvitamin D3 were
1a,25-
described for compound 13a in 30% yield. 1H NMR (600 MHz,
CDCl3)
d
7.19e7.24 (m, 3H), 7.07 (d, 1H, J ¼ 7.2 Hz), 5.94 (t, 1H,
a
J ¼ 2.4 Hz), 5.78 (dt, 1H, J ¼ 6.0, 2.4 Hz), 5.53 (dt, 1H, J ¼ 6.0, 2.4 Hz),
4.09 (ddd, 1H, J ¼ 4.8, 6.0, 10.2 Hz), 3.81 (s, 1H), 2.82 (t, 1H,
J ¼ 8.4 Hz), 2.62 (t, 2H, J ¼ 7.8 Hz), 2.55e2.59 (m, 1H), 2.52 (t, 1H,
J ¼ 7.2 Hz), 2.34e2.40 (m, 2H), 2.14e2.18 (m, 2H), 1.88 (ddd, 1H,
J ¼ 5.4,12.0,17.4 Hz),1.78 (dt,1H, J ¼ 1.8,13.2 Hz),1.67e1.73 (m, 3H),
1.51e1.59 (m, 7H), 1.21 (s, 6H), 1.08 (s, 3H), 1.02 (s, 3H); 13C NMR
a
4
a
separated by treatment with a dextran-coated charcoal (Norit SX-II)
suspension (200 mL) for 30 min at 4 ꢁC, followed by centrifugation
at 3000 rpm for 10 min. The supernatant (500 mL) was mixed with
Insta-Gel® Plus (9.5 mL) (PerkinElmer, USA) and the radioactivity
was counted. The relative potency of the analogues was calculated
(100 MHz, CDCl3)
d 154.6, 142.2, 140.0, 137.2, 136.4, 128.1, 127.0,
126.7, 126.6, 124.1, 122.0, 115.1, 72.7, 71.0, 65.4, 56.3, 47.8, 43.6, 42.5,
40.1, 38.6, 37.6, 36.4, 35.0, 30.5, 29.3, 26.2, 21.1, 21.0, 17.5, 16.2;
HRMS (ESI): Calcd for C31H42O3 [MþH]þ 463.3207, found [MþH]þ
463.3208.
from the concentration required to displace 50% of the [3H]-1
a
,25-
dihydroxyvitamin D3 from the receptor compared with the activity
of 1 ,25-dihydroxyvitamin D3, which was assigned as 100 by
a
definition.
6.1.11. (1a,3b)-17-[2-(4-Hydroxy-4-methylpentyl)phenyl]-9,10-
secoandrosta-5,7,10(19),16-tetraen-1,3-diol (2)
The solution of provitamin D 13a (60.0 mg, 0.13 mmol) in THF
(150 mL) was bubbled with nitrogen at 0 ꢁC for 10 min and then
irradiated with a 500 W high pressure mercury lamp until most of
the starting material was consumed. The resulting solution was
then refluxed for 2 h under the atmosphere of nitrogen. After the
removal of the solvent, the residue was purified by column chro-
matography on silica gel (hexane/AcOEt ¼ 1:3) followed by a
reversed-phase HPLC [20 mm ꢂ 25 cm YMC-Pack ODS-AQ column,
8 mL/min, MeOH/H2O (80:20)] to give the pure compound 2
6.3. Cell proliferation assay
Human breast cancer MCF-7 cell line was obtained from
American Type Culture Collection (Rockville, MD, USA). Cells were
maintained in RPMI-1640 medium (Hyclone) supplemented with
10% fetal bovine serum (Invitrogen, CA), penicillin (100 units/mL)
and streptomycin (100
conditions.
m
g/mL), and were grown at 37 ꢁC and 5% CO2
MCF-7 cell was plated into 96-well plates at a density of
4 ꢂ 103 cells/well and treated with compound 2, 3 or calcitriol (1) at
indicated concentrations for 72 h, followed by MTT assay as
described previously [34].
(10 mg, 16%) as a white solid. 1H NMR (600 MHz, CDCl3)
d 7.24 (d,
1H, J ¼ 7.8 Hz), 7.19 (dt, 1H, J ¼ 1.2, 7.8 Hz), 7.12 (dt, 1H, J ¼ 1.2,
7.8 Hz), 7.08 (d, 1H, J ¼ 7.8 Hz), 6.40 (d, 1H, J ¼ 10.8 Hz), 6.18 (d, 1H,

C. Liu et al. / European Journal of Medicinal Chemistry 85 (2014) 569e575
[12] S. Yamada, M. Shimizu, K. Yamamoto, Med. Res. Rev. 23 (2003) 89e115.
575
6.4. HL-60 cell differentiation assay
[13] C.-F. Zhang, Z.-P. Liu, Curr. Med. Chem. 20 (2013) 2080e2095.
[14] C.-F. Zhang, R.-Z. Wan, Z.-P. Liu, ChemMedChem 8 (2013) 1249e1260.
[15] L. Binderup, S. Latini, E. Binderup, C. Bretting, M. Calverley, K. Hansen, Bio-
chem. Pharmacol. 42 (1991) 1569e1575.
[16] C.J. Olivera, C.M. Bula, J.E. Bishop, L. Adorini, P. Manchand, M.R. Uskokovic,
A.W. Norman, J. Steroid Biochem. Mol. Biol. 89e90 (2004) 99e106.
[17] M.R. Uskokovic, P. Manchand, S. Marczak, H. Maehr, P. Jankowski, L. Adorini,
G.S. Reddy, Curr. Top. Med. Chem. 6 (2006) 1289e1296.
Human acute promyelocytic leukemia cell line HL-60 was ob-
tained from American Type Culture Collection (Rockville, MD, USA).
Cells were maintained in RPMI-1640 medium (Hyclone) supple-
mented with 10% fetal bovine serum (Invitrogen, CA), penicillin
(100 units/mL) and streptomycin (100
mg/mL), and were grown at
37 ꢁC and 5% CO2 conditions.
[18] G. Laverny, G. Penna, M. Uskokovic, S. Marczak, H. Maehr, P. Jankowski,
C. Ceailles, P. Vouros, B. Smith, M. Robinson, G.S. Reddy, L. Adorini, J. Med.
Chem. 52 (2009) 2204e2213.
HL-60 Cells were treated with compound 2, 3, calcitriol (1) for
48 h at 30e1000 nM. After the treatment, cells were transferred to
slides with cytospin and applied for Wright'seGiemsa staining.
Forty microscopic view fields were analyzed and photographed.
ꢁ
[19] B. Figadere, A.W. Norman, H.L. Henry, H.P. Koeffler, J.Y. Zhou, W.H. Okamura,
J. Med. Chem. 34 (1991) 2452e2463.
ꢀ
ꢀ
[20] E. Goceka, M. Kiełbinski, P. Wyłob, A. Kutner, E. Marcinkowska, Steroids 73
(2008) 1359e1366.
ꢀꢂ
ꢀ
ꢂ
[21] R. Fraga, F. Zacconi, F. Sussman, P. Ordonz-Moran, A. Munoz, T. Huet,
ꢀ
ꢂ
F. Molnar, D. Moras, N. Rochel, M. Maestro, A. Mourino, Chem. Eur. J. 18 (2012)
603e612.
Acknowledgments
[22] R.K. Jain, D.L. Trump, M.J. Egorin, M. Fernandez, C.S. Johnson, R.K. Ramanathan,
Invest. New Drugs 29 (2011) 1420e1425.
We are grateful to Shandong Provincial Natural Science Foun-
[23] G. Kumagai, M. Takano, K. Shindo, D. Sawada, N. Saito, H. Saito, S. Kakuda, K.-
I. Takagi, M. Takimoto-Kamimura, K. Takenouchi, T.C. Chen, A. Kittaka, Anti-
cancer Res. 32 (2012) 311e318.
dation (key project, ZR2011CZ001) for financial support.
Appendix A. Supplementary data
ꢀ
[24] M. Milczarek, M. Chodynski, B. Filip-Psurska, A. Martowicz, M. Krupa,
K. Krajewski, A. Kutner, J. Wietrzyk, Cancers 5 (2013) 1355e1378.
ꢀ
ꢀ
[25] H. Baurska, M. Kiełbinski, P. Biecek, O. Haus, B. Jazwiec, A. Kutner,
E. Marcinkowska, Leuk. Res. 38 (2014) 638e647.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2014.08.031.
[26] S.Y. Rhieu, A.j. Annalora, E. LaPorta, J. Welsh, T. Itoh, K. Yamamoto, T. Sakaki,
T.C. Chen, M.R. Uskokovic, G.S. Reddy, J. Cell. Biochem. 115 (2014) 1392e1402.
[27] Y.-Z. Yin, C. Liu, L.-Q. Tang, Z.-P. Liu, Steroids 77 (2012) 1419e1422.
[28] H. Shimizu, K. Shimizu, N. Kubodera, T. Mikami, K. Tsuzaki, H. Suwa,
K. Harada, A. Hiraide, M. Shimizu, K. Koyama, Y. Ichikawa, D. Hirasawa, Y. Kito,
M. Kobayashi, M. Kigawa, M. Kato, T. Kozono, H. Tanaka, M. Tanabe, M. Iguchi,
M. Yoshida, Org. Proc. Res. Dev. 9 (2005) 278e287.
References
[1] L.A. Plum, H.F. DeLuca, Nat. Rev. Drug Discov. 9 (2010) 941e955.
[2] J.W. Pike, M.B. Meyer, K.A. Bishop, Rev. Endocr. Metab. Disord. 13 (2012)
45e55.
[3] C. Carlberg, F. Molnar, Curr. Top. Med. Chem. 12 (2012) 528e547.
[4] F.S.G. Cheung, F.J. Lovicu, J.K.V. Reichardt, Expert Rev. Anticancer Ther. 12
(2012) 811e837.
[5] K. Noboru, Bitamin 87 (2013) 95e103.
[6] C.-Y. Yang, P.S.C. Leung, I.E. Adamopoulos, M.E. Gershwin, Clin. Rev. Allergy
Immunol. 45 (2013) 217e226.
[29] A. Przezdziecka, A. Kurek-Tyrlik, J. Wicha, Collect. Czech. Chem. Commun. 67
(2002) 1658e1668.
[30] Y. Imae, A. Manaka, N. Yoshida, Y. Ishimi, T. Shinki, E. Abe, T. Suda, K. Konno,
H. Takayama, S. Yamada, Biochim. Biophys. Acta 1213 (1994) 302e308.
[31] H. Jiang, W. Gao, D.M. Sze, H. Xiong, J. Hou, Int. J. Hematol. 86 (2007) 429e437.
[32] N. Rochel, J.M. Wurtz, A. Mitschler, B. Klaholz, D. Moras, Mol. Cell. 5 (2000)
173e179.
[33] M. Adamczyk, D.S. Watt, D.A. Netzel, J. Org. Chem. 49 (1984) 4226e4237.
[34] X. Mao, B. Cao, T.E. Wood, R. Hurren, J. Tong, X. Wang, W. Wang, J. Li, Y. Jin,
W. Sun, P.A. Spagnuolo, N. MacLean, M.F. Moran, A. Datti, J. Wrana, R.A. Batey,
A.D. Schimmer, Blood 117 (2011) 1986e1997.
[7] G. Eelen, C. Gysemans, L. Verlinden, E. Vanoirbeek, P. De Clercq, D. Van Haver,
C. Mathieu, R. Bouillon, A. Verstuyf, Curr. Med. Chem. 14 (2007) 1893e1910.
[8] A.J. Brown, E. Slatopolsky, Mol. Asp. Med. 29 (2008) 433e452.
ꢀ
[9] C. Carlberg, F. Molnar, Curr. Top. Med. Chem. 12 (2012) 528e547.
[10] M.S. Stein, J.D. Wark, Expert Opin. Investig. Drugs 12 (2003) 825e840.
[11] S. Nagpal, S. Na, R. Rathnachalam, Endocr. Rev. 26 (2005) 662e687.