Synthesis and biological activities of 2α-chloro-1-epicalcitriol and 1- epicalcitriol

Article abstract of DOI:10.1016/S0039-128X(97)00106-2

Anomalous diequatorial epoxide ring opening of 1β, 2β-oxido-cholesta- 5,7-diene-3β,25-diol 1 produces the 1β-hydroxy-2α-chloro-provitamin 2 and its corresponding 1β-hydroxy-provitamin 3. The provitamins 2 and 3 are transformed by irradiation and thermal isomerization to 2α-chloro-1- epicalcitriol NS3 (4) and 1-epicalcitriol NS8 (5), respectively. These two A- ring derivatives were tested for their in vitro biological activity in the mesenchymal, murine cell line C3H10T 1/4 , and their effects were compared with those of the native vitamin D₃ derivatives 25(OH)D₃ and 1,25(OH)₂D₃. NS3 and NS8 showed marked differences in their affinity for the vitamin D binding protein (DBP) and in their ability to inhibit cell proliferation. NS8 has the ability to bind to a high-affinity DBP-binding site for which 25(OH)D₃ has none affinity. The 2α-chloro-substitution (NS3) prevents binding to the postulated noncompetitive, NS8-specific DBP-binding site and diminishes the affinity to the vitamin D receptor (VDR) and therefore diminishing NS3 's biological abilities. The elucidation of the structure-function relationships at the DBP-binding-sites could have major impact on the development of new vitamin D₃ derivatives with extended serum half-life.

Full text of DOI:10.1016/S0039-128X(97)00106-2

Synthesis and biological activities of
2
1
␣-chloro-1-epicalcitriol and
-epicalcitriol
Bruno Sch o¨ necker,* Manfred Reichenb a¨ cher,* Sabine Gliesing,*
Manuela Gonschior,* Sirid Griebenow,* Dietmar Scheddin,† and Hubert Mayer†
*
†
Department of Chemistry, The Friedrich Schiller University Jena, Jena, Germany; and
Department of Gene Regulation and Differentiation, Association for Biotechnological Research,
Braunschweig, Germany
Anomalous diequatorial epoxide ring opening of 1␤,2␤-oxido-cholesta-5,7-diene-3␤,25-diol 1 produces the
1␤-hydroxy-2␣-chloro-provitamin 2 and its corresponding 1␤-hydroxy-provitamin 3. The provitamins 2 and 3
are transformed by irradiation and thermal isomerization to 2␣-chloro-1-epicalcitriol NS3 (4) and 1-epicalcitriol
NS8 (5), respectively. These two A-ring derivatives were tested for their in vitro biological activity in the
1
mesenchymal, murine cell line C3H10T ⁄
2
, and their effects were compared with those of the native vitamin D3
derivatives 25(OH)D and 1,25(OH) D . NS3 and NS8 showed marked differences in their affinity for the vitamin
3
2 3
D binding protein (DBP) and in their ability to inhibit cell proliferation. NS8 has the ability to bind to a
high-affinity DBP-binding site for which 25(OH)D has none affinity. The 2␣-chloro-substitution (NS3) prevents
3
binding to the postulated noncompetitive, NS8-specific DBP-binding site and diminishes the affinity to the
vitamin D receptor (VDR) and therefore diminishing NS3’s biological abilities. The elucidation of the structure-
function relationships at the DBP-binding-sites could have major impact on the development of new vitamin D3
derivatives with extended serum half-life. (Steroids 63:28–36, 1998) © 1998 by Elsevier Science Inc.
Keywords: vitamin D; A-ring derivatives; diequatorial 1␤,2␤-epoxide cleavage; vitamin D responsive element; vitamin D
binding protein; affinity studies
Introduction
NS8 (1-epicalcitriol) has been previously characterized as a
potent antagonist of 1,25(OH) D -induced intestinal cal-
2
8
3
The biologically active vitamin D metabolite, 1,25(OH) D
is of pivotal importance in mediating calcium and phospho-
rous homeostasis as well as in controlling the differentiation
and proliferation of several cell types.
mented the efforts to develop vitamin D derivatives with
3
2
3
cium transport (transcaltachia). 2␣-Chloro-1-epicalcitriol
NS3) is a new derivative and to our knowledge the first
(
member of the 2␣-substituted 1-epicalcitriols. Recently, 2␤-
1
–4
This has aug-
substituted 1-epicalcitriols with interesting biological activ-
3
9
ities have been described.
highly specific effects in contrast to the pluripotent activity
of 1,25(OH) D . The interest is centered on the develop-
For the cell culture experiments, we used the murine
2
3
1
fibroblastic cell line C3H10T ⁄
2
. After stable transfection
ment of vitamin D derivatives with low calcemic activity
3
with the human bone morphogenetic-4 (BMP-4) gene, this
cell line has the ability to differentiate into chondrocytes,
adipocytes, and cells of the osteogenic lineage by addition
and with the ability to inhibit proliferation and to modulate
5
–7
differentiation. Most of these substances were side chain
derivatives of 1,25(OH) D with unchanged A-ring struc-
2
3
10
of ascorbate and ␤-glycerophosphate. It is possible to
ture and configuration. In this study, we report the synthesis
and the biological capabilities of 1-epicalcitriol and 2␣-
chloro-1-epicalcitriol, both derived from 1␤,2␤-oxido-
cholesta-5,7-dien-3␤,25-diol 1. The 1␤-hydroxy derivative
cultivate this cell line for up to four weeks in the absence of
fetal calf serum (FCS) so that the in vitro cellular effects of
vitamin D derivatives can be separated from their DBP-
3
mediated actions (FCS contains significant amounts of
DBP). We used this highly vitamin D-sensitive cell line to
Address reprint requests to Bruno Sch o¨ necker, Institute of Organic and
Macromolecular Chemistry, The Friedrich Schiller University Jena,
Humboldtstra␤e 10, D-07743 Jena, Germany, or Hubert Mayer, Depart-
ment of Gene Regulation and Differentiation, Association for Biotechno-
logical Research, Mascheroder Weg 1, D-38124 Braunschweig, Germany.
Received April 28, 1997; accepted September 10, 1997.
test the ability of vitamin D derivatives to inhibit cell
3
1
1
proliferation and modulate adipocyte differentiation.
1
C3H10T ⁄ -cells were also stably transfected with a con-
2
struct containing the vitamin D responsive element (VDRE)
of the human osteocalcin gene coupled to a chlorampheni-
colacetyltransferase (CAT)-reporter gene in order to evalu-
Dedicated to Prof. Dr. P. Welzel on the occasion of his 60th birthday.
Steroids 63:28–36, 1998
©
1998 by Elsevier Science Inc. All rights reserved.
0039-128X/98/$19.00
6
55 Avenue of the Americas, New York, NY 10010
PII S0039-128X(97)00106-2

Synthesis and biological activities of 1-epicalcitriols: Sch o¨ necker et al.
ate the ability of the vitamin D derivatives to activate gene
transcription at the VDRE.
2␣-Chloro-cholesta-5,7-dien-1␤,3␤,25-triol 2
3
Epoxide 1 (1.25 g; 3.1 mmol) was dissolved in N,N-dimethyl
formamide (80 mL) in absence of UV light. Conc. hydrochloric
acid (3.3 mL; 4.2 mmol) in N,N-dimethyl formamide (20 mL) was
added while stirring after 1 to 2 h, (TLC: ethyl acetate/isohexane
Experimental
4
0/60 v/v) the reaction mixture was poured into chilled water. The
Chemistry
mixture was extracted with diethyl ether, and the organic layer was
dried over Na SO . After evaporation in vacuo, a white solid
(1.23 g) was obtained, which was recrystallized from ethyl acetate/
isohexane (75/25 v/v) giving white crystals of 2 (0.92 g). Flash
chromatography of the mother liquor further gave crystalline 2
(0.13 g). Yield: 1.05 g (76%). m.p. 198–203°C. UV (CH3OH)
␭max ϭ 284 nm (⑀ ϭ 11300). HNMR (400 MHz, CDCl ) ␦: 0.63
(s, 3H, 18-H); 0.97 (d,
1
␤,2␤-oxido-cholesta-5,7-dien-3␤,25-diol 1 was synthesized from
2
4
1
2
(
8
20S)-20-(p-toluene-sulfonyloxymethyl)-pregna-1,5-dien-3␤-ol
in
steps, as previously described.13,14 The 1␤,2␤-epoxy group of
compound 1 reacts with hydrochloric acid in N,N-dimethyl-
formamide via an anomalous trans-diequatorial intermediate, giv-
ing the 1␤-hydroxy-2␣-chloro compound 2 ( H NMR data). By
reductive cleavage of the 1␤,2␤-epoxy group of 1 with lithium
aluminum hydride, the hitherto unknown 1␤-hydroxy compound 3
1
1
3
3
J ϭ 6.5Hz, 3H, 21-H); 1.10 (s, 3H, 19-H);
1.23 (s, 6H, 26-H and 27-H); 3.62 (m, 2H, 1
␣
-H and 3␣-H); 3.96
(t, J ϭ 9.9Hz, 1H, 2␤-H); 5.33 (d, J ϭ 5.9Hz, 1H, 7-H); 5.68 (dd,
3
3
(provitamin of 1-epicalcitriol) was formed, in contrast to the 2␤-
1
3
hydroxy compound we previously specified.
J ϭ 5.8Hz, 1H, 6-H). Addition of trichloroacetyl isocyanate: ␦:
3
For the transformation of the provitamins into the previtamins,
simultaneous irradiation at ϽϪ40°C with light in the range of 285
to 300 nm and Ͼ330 nm was carried out (filter solution), as
described for the synthesis of precalcitriol.15 Irradiation was ter-
minated after nearly 50% of the provitamins was consumed in
order to avoid the production of undesired side products. HPLC
and flash chromatography furnished the desired previtamins,
starting materials, and mixtures of the other photoisomers. The
starting materials (provitamins) and the other photoisomers were
recycled to improve the yields of previtamins.
After chromatographic purification, thermal isomerization of
the previtamins gave vitamins NS3 and NS8. The lower yields in
comparison to the synthesis of calcitriol are due to the lack of
optimized reaction conditions (irradiation and isomerization). In
this way, 1-epicalcitriol (NS8) was synthesized for the first time by
UV irradiation and thermal isomerization.
The trans-diequatorial epoxide opening of compound 1 is sur-
prising. Normally, 1␤,2␤-epoxides with a saturated A-ring give the
1.60 (s, 6H, 26-H and 27-H); 4.27 (t, J ϭ 10.5Hz, 1H, 2␤-H); 4.90
(m, 1H, 3
8
3
␣
-H); 5.25 (d, J ϭ 10.4Hz, 1H, 1␣-H); 8.18 (s, 1H, NH);
.37 (s, 2H, NH). HRMS m/z found 450.2921 (calculated for
C H O Cl: 450.2889)
2
7 43 3
1
6
(5Z,7E)-9,10-Seco-2␣-chloro-cholesta-5,7,10(19)-
trien-1␤,3␤,25-triol NS3 4
2
␣-Chloro-provitamin 2 (150 mg; 0.33 mmol) was dissolved in
1
5
CH OH (100 mL) and tert-butyl methyl ether (350 mL), trans-
3
ferred to a photoreactor, and flushed with argon for 45 min. After
cooling to Ϫ40°C, irradiation was carried out using a filter solution
of 2,7-dimethyl-3,6-diaza-cyclohepta-1,6-diene tetrafluoroborate
and biphenyl in ethanol and a high pressure mercury lamp TQ 150
Z1. After 140 min, the irradiation was stopped (HPLC control:
LiChrospher DIOL, i-propanol/i-hexane (8/92 v/v, detection ␭ ϭ
2
The residue was dissolved in ethyl acetate/i-hexane (20 mL, 70/30
v/v) and crystallized at Ϫ18°C (12 h). Unreacted provitamin 2 (55
mg; 37%) was obtained. The mother liqour was evaporated in
vacuo, the residue was dissolved in i-propanol/i-hexane (5 mL,
1
5
1
60 nm, ␭ ϭ 280 nm), and the solvents were evaporated in vacuo.
2
1
7–19
trans-diaxial 1␣-substituted 2␤-hydroxy compounds.
The rea-
sons for the anomalous epoxide cleavage of compound 1, which
has an unsaturated B-ring, are under investigation. Through this
anomalous cleavage, a route was developed to the production of a
new class of provitamins and vitamins possessing a 1␤-hydroxy
group and a substituent in the 2␣-position.
1
2
␭1
0/90 v/v) and separated by HPLC (LiChrospher DIOL 7 ␮m
50 ϫ 25 mm, i-propanol/i-hexane 8/92 v/v, 15 mL/min detection
ϭ 260 nm, ␭ ϭ 280 nm) to give the expected previtamin (20
2
mg, 13%) as a colorless foam, further starting provitamin 2 (15
mg, 10%), and a mixture of 2 and other photoproducts (30 mg,
General remarks
2
0%). After recycling of 2, a total amount of 29 mg (19%)
Solvents were distilled and dried before use according to conven-
tional methods. N,N-Dimethyl formamide was dried over molec-
ular sieves 4Å and distilled in vacuo. All reactions were conducted
under an atmosphere of dry argon and under exclusion of UV light.
TLC: silicagel 60 F254 (MERCK) plates, layer thickness 0.2 mm,
detection by UV (␭ ϭ 254 nm; ␭ ϭ 366 nm) and spraying with
previtamin was obtained.
This previtamin (UV ␭max ϭ 259 nm) was immediately
dissolved in i-propanol/i-hexane (10 mL, 10/90 v/v) in the dark
under argon and heated to 40 to 50°C for 5 h (HPLC control:
LiChrospher DIOL, i-propanol/i-hexane 8/92 v/v, detection ␭ ϭ
260 nm). The solution was evaporated in vacuo, the residue was
dissolved in i-propanol/i-hexane (2 mL, 10/90 v/v) and separated
by HPLC (LiChrosorb DIOL 7 ␮m 250 ϫ 25 mm, i-propanol/i-
hexane 10/90 v/v, 17 mL/min, detection ␭ ϭ 260 nm, ␭ ϭ 230
nm). After evaporation and drying in vacuo, the following prod-
ucts were obtained: 1.) 15 mg previtamin (53%) 2.) 9 mg 2␣-
chlorovitamin 4 (31%). The unreacted previtamin was thermically
isomerized once more giving additional vitamin 4 as white nee-
1
2
a solution consisting of 80 mL conc. sulfuric acid and 20 mL
methanol and by heating at 120°C. Flash column chromatogra-
1
6
phy: LiChroprep (MERCK, 25–40 ␮m) [eluents (v/v) are given
in parentheses]. HPLC: SHIMADZU LC-8A equipped with a
UV/VIS spectrophotometric detector SPD-10AV, stationary phas-
es: LiChrosorb DIOL (MERCK, 150 ϫ 3 mm, 5 ␮m) or LiChro-
sorb DIOL (MERCK, 250 ϫ 25 mm, 7 ␮m) [eluents (v/v) are
given in parentheses]. Melting points: Boetius micromelting point
apparatus, uncorrected values. High-resolution mass spectra
1
2
dles. m.p. 173–179°C (CH OH). UV (CH OH) ␭ ϭ 264 nm
3
3
max
1
(⑀ ϭ 15300). H NMR (400MHz, CDCl ) ␦: 0.50 (s, 3H, 18-H);
3
3
(
HRMS): AMD 402 intectra with electron impact ionization at 70
eV. Nuclear magnetic resonance (NMR) spectra: BRUKER DRX
00 [chemical shifts (␦) are reported in ppm, relative to tetra-
0.91 (d, J ϭ 6.3Hz, 3H, 21-H); 1.19 (s, 6H, 26-H and 27-H); 2.25
3
(m, 1H, 4␤-H); 2.54 (t, J ϭ 9.7Hz, 4␣-H); 2.71 (m, 1H, 9␤-H);
4
3.81–3.83 (m, 2H, 2␤-H and 3␣-H); 4.12 (m, 1H, 1␣-H); 5.11 and
4
3
methylsilane as the internal standard] Signals are assigned by
TOCSY, selective TOCSY. UV spectra: CARL ZEISS JENA
5.56 (2 ϫ t, J ϭ 2Hz, 2H, 19-H); 5.94 and 6.38 (2 ϫ d, J ϭ
11,3Hz, 2H, 7- and 6-H). Addition of trichloroacetyl isocyanate: ␦:
Ϫ1
3
Specord M 500 [extinction coefficients (⑀) are given in 1 mol
cm in parentheses].
1.49 (s, 6H, 26-H and 27-H); 4.13 (t, J ϭ 7.1Hz, 1H, 2␤-H); 5.22
Ϫ1
3
(m, 1H, 3␣-H); 5.51 (d, J ϭ 7.3Hz 1H, 1␣-H); 8.17 (s, 1H, NH);
Steroids, 1998, vol. 63, January 29

Papers
Scheme a: HCI/DMF, r.t.; b:
i photo-
isomerization: tert-butyl methyl ether/
CH OH (80/20 v/v); photoreactor, Hg high
3
pressure lamp (TQ 150 Z1), filter solu-
tion (2,7-dimethyl-3,6-diaza-cyclohepta-1,6-
diene-tetrafluoroborate/biphenyl), 140 min;
ii HPLC: diol gel 7 ␮m, propan-2-ol/i-hexane
(
8/92 v/v); iii thermal isomerization: propan-
-ol/i-hexane (10/90 v/v); 40–50°C, 5 h; iv
HPLC: diol gel 7 ␮m, propan-2-ol/i-hexane
2
(
10/90 v/v); c: LiAIH /ether, r.t.; d: i photo-
4
isomerization: tert-butyl methyl ether/
CH OH (90/10 v/v); photoreactor, Hg high
3
pressure lamp (TQ 150 Z1), filter solu-
tion (2,7-dimethyl-3,6-diaza-cyclohepta-1,6-
diene-tetrafluoroborate/biphenyl), 45 min; ii
ϩ
flash chromatography: Ag -modified silica
gel 25–40 ␮m, ethyl acetate/hexane (70/30
v/v); iii thermal isomerization: ethyl acetate/
hexane (75/25 v/v); 55–60°C, 5 h; iv flash
ϩ
chromatography: Ag -modified silica gel
25–40 ␮m, ethyl acetate/hexane (75/25 v/v).
8
.58 (s, 2H, NH). HRMS m/z found 450.2917 (calculated for
modified silica gel, ethyl acetate/hexane 75/25 v/v). The following
photoisomers were eluted in this order: previtamin (44 mg, 29.3%,
colourless foam, UV (Ethanol): ␭max ϭ 259 nm), tachysterol (11
mg, 7.3%, colorless foam, UV (Ethanol): ␭max ϭ 284 nm), and
unreacted 3 (8 mg, 5%). After irradiation recycling of 3 (56 mg),
a total amount of 61 mg (40.7%) previtamin was obtained.
The corresponding tachysterol (10 mg in 40 mL tert-butyl
methyl ether) was converted to the previtamin by sensitized pho-
C H O Cl: 450.2889).
2
7 43 3
1
␤,3␤,25-Trihydroxy-cholesta-5,7-dien 3
Epoxide 1 (250 mg; 0.6 mmol) was dissolved in THF (20 mL). A
solution of LiAIH (76 mg; 2 mmol) in THF (25 mL) was added
at 0°C with stirring, and the mixture was refluxed (about 30 min)
until starting material was no longer detectable by TLC (TLC:
ethyl acetate/hexane 75/25 v/v; R : 1: 0.45; 3: 0.28). Diethylether
4
2
0
toisomerization using fluorenone (16 mg) as the sensitizer. TLC
and flash-chromomatographic separation of the reaction mixture as
described above gave further previtamin (11 mg, 68.7% from
tachysterol). Previtamin (70 mg) was dissolved in ethylacetate/
hexane (50 mL, 75/25 v/v) under argon and was heated to 60°C in
f
(100 mL) was added, and the reaction mixture was poured into
chilled water. The organic layer was separated, and the A1(OH)3
residue was extracted several times with methanol (50 mL). The
combined extracts were dried over Na SO , and the solvent was
ϩ
the dark for 5 h (TLC on Ag -modified plates: ethyl acetate/
2
4
removed under reduced pressure. The yellowish crude product was
recrystallized from methanol giving white needles of 3 (165 mg,
hexane 75/25 v/v). The solvent was evaporated under reduced
pressure, and the mixture was separated by flash-chromatography
(Ag -modified silica gel, ethyl acetate/hexane 75/25 v/v). After
ϩ
6
5.7%), and the mother liquor was purified by flash chromatog-
raphy (ethyl acetate/hexane 75/25 v/v) giving more 3 (48 mg,
recycling of the unreacted previtamin (21 mg), a total amount of 56
1
1
(
9.1%). UV (CH OH) ␭
ϭ 284 nm (⑀ ϭ 11400). H NMR
max
mg 5 (37, 3% from 3) was obtained as white needles. m.p.
3
3
1
400MHz, CDCl ) ␦: 0.61 (s, 3H, 18-H); 0.94 (d, J Ϸ 6.3Hz, 3H,
210–12°C. UV (C
NMR (400MHz, CDCl
2
H
5
OH) ␭max ϭ 265 nm (⑀ ϭ 16 600). - H
3
3
2
2
1-H); 1.03 (s, 3H, 19-H); 1.22 (s, 6H, 26-H and 27-H); 3.70 (m,
H, 1␣-H and 3␣-H); 5.25 (m, 1H, 7-H); 5.64 (dd, J Ϸ 6.5Hz,
3
): ␦: 0.53 (s, 3H, 18-H); 0.94 (d, J Ϸ
3
6.4Hz, 3H, 21-H); 1.20 (s, 6H, 26-H and 27-H); 4.09 (m, 1H,
4
4
J Ϸ 1.5Hz, 1H, 6-H). Addition of trichloroacetyl isocyanate: ␦:
3␣-H), 4.34 (t, 1H, 1␣-H), 4.99 (d, J Ϸ 1.9Hz, 1H, 19-H); 5.27 (s,
3
3
0
.60 (s, 3H, 18-H); 0.92 (d, J Ϸ 6.1 Hz, 3H, 21-H); 1.16 (s, 3H,
1H, 19-H), 6.03 and 6.45 (2 ϫ d, AB pattern, J Ϸ 11.3Hz, 2H,
3
4
19-H); 1.53 (s, 6H, 26-H and 27-H); 5.39 (dd, J Ϸ 6.5Hz, J Ϸ
1.5Hz, 1H, 7-H); 5.73 (m, 1H, 6-H); 8.20, 8.34 and 8.40 (3 ϫ s,
3 ϫ 1H, 3 ϫ NH) HRMS m/z found 416.3279 (calculated for
6-H and 7-H). Addition of trichloroacetyl isocyanate: ␦: 0.50 (s,
3
3H, 18-H); 0,90 (d, J Ϸ 6.4Hz, 3H, 21-H); 1.51 (s, 6H, 26-H and
27-H); 5.19 (s, 1H, 19-H); 5.20 (m, 1H, 3␣-H); 5.48 (t, 1H, 1␣-H);
3
C H O ; 416.3290).
5.51 (s, 1H, 19-H); 5.87 and 6.37 (d, J Ϸ 10.8Hz, 2H, 6-H and
2
7 44 3
7
-H); 8.12, 8.59 and 8.69 (3 ϫ s, 3 ϫ 1H, 3 ϫ NH). These physical
2
1
data are in agreement with data in the literature.
(
5Z,7E)-9,10-Seco-cholesta-5,7,10(19)-trien-
1
␤,3␤,25-triol NS8 5
Biological methods and materials
Provitamin 3 (150 mg, 0.36 mmol) was dissolved in methanol (100
mL), and tert-butyl methyl ether (350 mL) was added. The solu-
tion was transferred into a 450 mL photoreactor, and the photore-
action was carried out as described above. After 45 min of irradi-
Cell culture and transfection experiments
C3H10T1⁄
cells were obtained from the European Collection of
2
ϩ
ation time, about 60% of 3 was consumed (TLC on Ag -modified
Animal Cell Cultures (Porton Down) and cultured in Dulbecco’s
modified Eagle’s medium (DMEM) supplemented with 10% FCS,
2mM glutamine, 60 mg/L penicillin, and 100 mg/L streptomycin at
37°C under a humidified atmosphere (95% humidity) with 5%
plates: ethyl acetate/hexane 75/25 v/v), the irradiation was
stopped, and the solvent was evaporated under reduced pressure.
The residue (yellowish oil) was dissolved in ethyl acetate/hexane
and allowed to crystallize the unreacted 3 at Ϫ18°C (48 mg, 32%).
CO . All cell culture flasks were precoated prior to cell culture for
2
ϩ
The mother liqour was separated by flash chromatography (Ag -
1 h at 37°C with a solution of gelatin (10 mg/mL). The cell culture
3
0
Steroids, 1998, vol. 63, January

Synthesis and biological activities of 1-epicalcitriols: Sch o¨ necker et al.
1
medium of the stably transfected cell lines C3H10T ⁄
2
(BMP-4) and
Following incubation for 2 h at 4°C, the radioactivity was har-
vested on glass fiber filters (XTalScint, Beckman) with a semi-
automatic cell harvester (Bibby Dunn, 2800 Series). The radioac-
tivity on each filter was counted in a Beckman LS 6000 IC
scintillation counter. Six parallel experiments were conducted with
each vitamin D derivative concentration.
1
C3H10T ⁄
2
(4xVDRE-CAT) were additionally supplemented with
␮g/mL puromycin during the entire cultivation period.
C3H10T ⁄ -cells stably transfected with the eukaryotic expres-
sion vector pMBC-2T-f1, containing the entire coding sequence of
the human version of BMP-4, were used for the DNA-synthesis
and the adipogenesis inhibition experiments. To create a vitamin
5
1
2
D -specific cellular gene activation test system, four direct repeats
3
of the 32-base-pair DNA-fragment from the human osteocalcin-
vitamin D responsive element (osteocalcin-VDRE) with the
sequence 5Ј-CTTGGTGACTCACCGGGTGAACGGGGGCA-
TTA-3Ј were cloned into the BamH1 site of the chlorampheni-
colacetyltransferase(CAT)-expression vector pBLCAT2. The
Determination of CAT-activity
C3H10T1⁄
(4ϫVDRE)-cells were precultured to near-confluence,
2
diluted to 30,000 cells/mL, and transferred into petri dishes (96
mm diameter) in DMEM containing 10% FCS. The cells were
subsequently washed with PBS and further cultured in DMEM
without FCS and in the presence of various amounts of vitamin D3
derivative or solvent. The solvent concentration in the medium
never exceeded 0.1%. After an incubation period of 60 h, the cells
were washed three times with PBS. The washed cells were lysed,
and the CAT-concentration of each sample was determined using
a CAT-enzyme linked immunosorbent assay (ELISA) kit (Boeh-
ringer Mannheim).
resulting expression vector pBLCAT2(4ϫVDRE) was stably
1
transfected into C3H10T ⁄
2
-cells. Stable transfections were pro-
duced using the calcium-phosphate precipition method. Puro-
mycin resistance of the stable transfectants was mediated by
cotransfection with the plasmid pB Spac⌬p, which codes for the
puromycin-acetyltransferase-gene.
Measurement of DNA synthesis
C3H10T1⁄
(BMP-4)-cells were precultured to near-confluence, di-
2
Determination of the affinity of vitamin D3
derivatives for the vitamin D receptor (VDR)
luted to 30,000 cells/mL, and transferred into 96-well plates. After
cultivation to confluence, the culture medium was substituted with
DMEM without FCS supplementation, and the cells were treated
A
1,25(OH) D -assay system (Amersham-Buchler, Braun-
2 3
Ϫ13
Ϫ5
with 10
M to 10
M vitamin D3 derivatives or solvent
schweig, Germany) was modified to determine the affinity of
(
ethanol). Five days later, the cells were labeled with 1 ␮Ci/mL
vitamin D derivatives for the VDR. Lyophilized intestinal VDR
3
3
[methyl- H]-thymidine (47 Ci/mmol). After 4.5 h of incubation
from rachitic chicken was reconstituted with 31 mL of 0.05 M
potassium phosphate buffer, pH 7.4, containing 0.05 M potassium
chloride, 1 mM dithiothreitol, and 0.1% (w/v) bovine ␥ globulin.
and washing with PBS, 50 ␮L of 0.4 M NaOH were added to each
well, and each plate was sealed with a plate sealer sheet. The cells
were lysed for 30 min at 80°C, and after the plates were cooled,
3
VDR solution (500 ␮L), 1,25-dihydroxy[26,27-methyl- H]vitamin
2
00 ␮L of 2% perchloro acetic acid were added to each well.
D (25 ␮L, 0.5 ng/mL, 0.24 ␮Ci/mL), and a sample of vitamin D
3
3
Figure 1 Inhibition of cell proliferation of C3H10T1/2(BMP-4)-cells by 1,25(OH) D , 25(OH)D , NS3 and NS8. C3H10T1/2(BMP-4)-cells
2
3
3
were grown in 96-well plates and treated in serum-free medium with various concentrations of vitamin D3 derivatives or solvent
3
(
ethanol). Five days later the cells were labeled with 1 ␮Ci/well [Methyl- H]-thymidine (47 Ci/mmol) for 4.5 h, washed with PBS and
3
lysed and [ H]-Thymidine incorporation was determined. †, 1,25(OH) D -induced cell lysis. Each point of the dose-response curve is the
average of six wells. The results shown are representative of two to three experiments.
2
3
Steroids, 1998, vol. 63, January 31

Papers
derivative (50 ␮L) were combined and incubated for 1 h at 22°C.
Immediately after this incubation period, free and bound vitamin
to 1,25(OH) D , NS3 and 25(OH)D showed only a signif-
2 3 3
Ϫ8
icant inhibitory effect at concentrations above 10 M. NS8
D were separated by adding 200 ␮L of a dextran-coated charcoal
3
was the best proliferation-inhibitor of the A-ring derivatives
suspension (0.5% w/v charcoal, 0.05% w/v dextran). Displacement
of 1,25(OH) D -tracer from VDR by the vitamin D derivatives
Ϫ10
of 1,25(OH) D with 30% inhibition at 10
M.
2
3
2
3
3
The most significant differences between 1,25(OH) D
2
3
was determined by assessing the radioactivity with a liquid scin-
tillation counter.
and the synthetic vitamin D derivatives were their gene
3
activating properties at the VDRE-CAT-construct.
1
,25(OH) D induces the CAT-gene seventeen-fold at a
2 3
Ϫ11
Determination of the affinity of vitamin D3
derivatives for vitamin D binding protein (DBP)
concentration as low as 10
activating activities of NS3 and 25(OH)D were nearly
M (Figure 2). The gene
3
1
0,000 times weaker. Both derivatives had a similar effect
A 25(OH)D -assay system (Amersham-Buchler, Braunschweig,
3
at all tested concentrations. NS8 was slightly more active
than NS3 and 25(OH)D with nearly three times higher
CAT-activity at 10 M.
Germany) was modified to determine the affinity of vitamin D3
derivatives for serum DBP. Lyophilized DBP from sheep serum
was dissolved in 0.05 M potassium phosphate buffer, pH 7.4
3
Ϫ8
containing 0.01% (w/v) bovine immunoglobulin G and 0.01%
The 1␤-derivatives displayed a weak affinity for the
3
(
w/v) thimerosol. Fifty ␮L 25(OH)[26,27-methyl- H]vitamin D3
VDR similar to that of the 1,25(OH) D -precursor
2
VDR affinity of 1,25(OH) D (Figure 3). As shown by the
2
3
(0.5 ng/mL, 0.19 ␮Ci/mL) and 50 ␮L of the vitamin D derivative
5(OH)D , which is about 20,000 times weaker than the
3
sample were added to 500 ␮L DPB solution, mixed, and incubated
for 2 h at 4°C. Bound and free vitamin D3 were separated by
addition of 500 ␮L dextran-coated charcoal suspension (0.275%
w/v charcoal, 0.0275% w/v dextran) immediately after the incu-
2
3
Ϫ7
5
0% displacement of NS8 at a concentration of 9 ϫ 10
M, this metabolite has nearly two times higher VDR affinity
than 25(OH)D . The affinity of the 2␣-chlor-derivative of
NS8, NS3, was nearly ten times lower.
bation period. Displacement of the 25(OH)D -tracer was measured
3
3
with a liquid scintillation counter.
NS3 and NS8 showed major differences in DBP affinity.
Ϫ7
At a concentration of 2 ϫ 10 M, NS3 displaced 50% of
the 25(OH)D -tracer (Figure 4). NS8 had a 50% displace-
ment concentration of 9 ϫ 10 M, which is nearly 22 times
lower than that of NS3 and only nearly six times higher than
Determination of adipocyte differentiation and
cell number
3
Ϫ9
C3H10T1⁄
(BMP-4)-cells were precultured to near-confluence, di-
2
luted to 30,000 cells/mL, and transferred into 12-well plates. After
culturing to confluence, the culture medium was replaced with
that of 25(OH)D , the main transportable form of vitamin
3
Ϫ8
D in the blood. At concentrations of NS8 below 10 M,
3
DMEM supplemented with 10% FCS, 50 ␮g/mL ascorbate, and 10
the DBP binding curve in the competition experiment dif-
fered significantly from the ideal sigmoidal form. At these
low concentrations, NS8 showed an even higher affinity for
DBP than 25(OH)D3.
Ϫ13
mM ␤-glycerophosphate, and the cells were treated with 10
M
Ϫ5
to 10 M vitamin D derivatives or solvent. Experiments were
3
carried out in triplicate. Every 4–5 days, half of the media was
replaced by fresh medium including the vitamin D derivatives.
3
In contrast to all other tested substances, 1,25(OH) D
After a culture period of 19 days, the cells were washed two times
with PBS and 2 mL PBS per well were subsequently added. To
each well, 20 ␮L of a solution of 100 ␮g/mL nile red in acetone
were added. After an incubation period of 20 min at room tem-
perature, the supernatant was discarded and 2 mL of PBS were
added. The plates were scanned in a Cytofluor 2350 fluorescence
photometer (excitation 485 nm/emission 530 nm) to determine the
neutral lipid content of the cells. Subsequently, the cells were
washed three times with PBS and fixed for 30 min with methanol,
pH 8.5. Five hundred ␮L of a 0.1% (w/v) methylene blue solution
in 10 mM borate buffer, pH 8.5, were added to each well. After 20
min, the cells were washed three times with 10 mM borate buffer,
pH 8.5. For destaining, 4 mL of 0.1 M HCl containing 20% ethanol
were added. After 20 min, the absorbance of the destaining solu-
tion was measured in a Multiskan photometer at 650 nm and
correlated with a standard curve.
2 3
displayed a strong inhibitory effect on adipocyte differen-
tiation. At all but the lowest concentration tested,
1,25(OH) D diminished the neutral lipid concentration of
2
3
1
C3H10T ⁄
centration of 10
2
(BMP-4)-cells to 20% of control. Even at a con-
Ϫ13
M, the fat concentration barely ex-
ceeded the 60% mark. The microscopic control of cell
morphology confirmed the fluorescence measurement be-
cause of the virtual absence of preadipocyte-like, fat
droplets-containing, rounded cells at all concentrations with
low fluorescence. The main reason for the inhibition of
adipocyte differentiation is the proliferation-inhibitory ac-
tivity of 1,25(OH) D . Its ID50 concentration was about ten
2 3
times higher compared to its inhibition of proliferation in
serum-free cell culture. The other three tested derivatives
were weak inhibitors of adipocyte differentiation and cell
Ϫ7
proliferation at concentrations above 10 M. The signifi-
Results
cantly stronger proliferation-inhibitory effect of NS8 on
1
The actions of 1,25(OH) D , 25(OH)D , NS3, and NS8 on
C3H10T ⁄
2
-cells in serum-free culture compared to both
2
3
3
1
the murine fibroblastic cell line C3H10T ⁄
over a concentration range of 10 –10
2
were examined
25(OH)D and NS3 was not seen in the presence of FCS.
3
Ϫ5
Ϫ13
M. The biologi-
Inhibition of adipogenesis and inhibition of cell prolifera-
tion of these three derivatives seems to be tightly coupled
under these conditions.
cally active vitamin D metabolite 1,25(OH) D was the
3
2 3
(BMP-4)-cells
1
best inhibitor of proliferation of C3H10T ⁄
2
of all tested vitamin D derivatives. Even at the low dose of
3
Ϫ11
1
0
M inhibition of proliferation exceeds 50% (Figure 1).
Discussion
All other examined vitamin D derivatives were signifi-
3
cantly weaker. NS3 and 25(OH)D inhibited proliferation of
We examined the in vitro biological activities of
3
1
C3H10T ⁄
2
(BMP-4)-cells in quite a similar way. In contrast
1,25(OH) D , 25(OH)D3, and two 1␤-derivatives on
2
3
3
2
Steroids, 1998, vol. 63, January

Synthesis and biological activities of 1-epicalcitriols: Sch o¨ necker et al.
Figure 2 Dependence of CAT-production in C3H10T1/2(4xVDRE-CAT) cells on 1,25(OH) D , 25(OH)D , NS3 and NS8. Stable transfected
2
3
3
C3H10T1/2 cells with chloramphenicol-acetyltransferase expression vector pBLCAT2 containing four direct repeats of the human
osteocalcin VDRE were grown in 96 well plates and treated in serum-free medium with various concentrations of vitamin D derivatives
3
or solvent (ethanol) for 60 h. After washing the cells were lysed and the CAT concentration was determined using a CAT enzyme linked
immunosorbent assay kit. †, 1,25(OH) D -induced cell lysis. The experiments were carried out in triplicates. The results shown are
2
3
representative of two to three experiments.
C3H10T1⁄
2
-cells and their binding affinities for VDR and
putative independency of adipocyte differentiation and in-
hibition of proliferation induced by 1,25(OH) D .
DBP. We tested the inhibition of cell proliferation and the
gene-activating activity of these seco-steroids in the absence
of FCS. Therefore, it was possible to separate DBP-coupled
2
3
The two examined 1␤-derivatives of 1,25(OH) D , NS3
2
3
and NS8, both displayed a low affinity for VDR in the range
processes from their direct cellular activities. The
of the weakly active 1,25(OH) D -precursor 25(OH)D .
2
3
3
1
C3H10T ⁄
2
(BMP-4)-cells are highly sensitive for the native,
NS8 showed a slightly higher VDR-affinity than 25(OH)D3,
whereas NS3 bound with the lowest affinity of all tested
substances. This explains the low proliferation-inhibitory
and CAT-activating activities of NS3. In both experiments,
biologically active vitamin D3 metabolite 1,25(OH) D :
2
3
proliferation-inhibitory activity was clearly detectable at
Ϫ11
concentrations as low as 10
M. The same concentration
Ϫ8
was also sufficient to activate a VDRE-CAT-construct in
a concentration of more than 10 M was needed to see a
1
C3H10T ⁄
2
(4ϫVDRE-CAT)-cells nearly 18-fold relative to
significant effect. The slightly higher VDR-affinity of NS8
compared to that of NS3 led to a somewhat higher CAT-
activity. However, the greater than hundred fold higher
inhibition of cell proliferation is not fully explained by the
VDR-affinity of NS8. Since the VDR-affinity of NS8 is
congruent with its CAT-activity and because gene-
activating and cell proliferation experiments were both car-
the control. However, in the presence of FCS, a nearly ten
times higher concentration of 1,25(OH) D was needed to
2
3
reach half-maximal inhibition of cell proliferation due to the
presence of DBP and lipoprotein in FCS to which binds a
significant amount of 1,25(OH) D . The effect of
2
3
1
,25(OH) D on cell proliferation in the absence as well as
2 3
1
22
in the presence of FCS was strictly dose dependent. Al-
though mainly dependent on the strong proliferation-
inhibitory activity of 1,25(OH) D , the inhibitory effect of
ried out in C3H10T ⁄
2
-cells,
different catabolic or
“steroid-sorting” processes by membrane transporters dur-
ing inhibition of cell proliferation are unlikely. One possi-
bility is that the VDR needs a different dimerization partner
for inhibition of cell proliferation than for VDRE-CAT
activation. Since ligand binding and heterodimerization of
2
3
1
,25(OH) D on adipocyte differentiation seems to be reg-
2 3
ulated by an “on or off” mechanism. Further experiments
with a lower concentration range are needed to elucidate the
Steroids, 1998, vol. 63, January 33

Papers
3
Figure
3
2
Displacement of [ H]-
1
,25(OH) D3 from chicken intes-
tine vitamin D receptor by increas-
ing concentrations of 1,25(OH) D ,
2
3
2
1
5(OH)D3, NS3 and NS8. An
,25(OH) D assay system was used
2
3
to determine the affinity of vitamin
D3 derivatives to VDR. Intestinal
VDR from rachitic chicken was incu-
bated with 1,25-dihydroxy[26,27-
3
methyl- H] vitamin D3 and increas-
ing concentrations of vitamin D3
derivatives for 1 h at 22°C. Bound
3
and free [ H]-1,25(OH) D were sep-
2
3
arated by the addition of dextran
coated charcoal suspension. Dis-
3
placement of [ H]1,25(OH) D was
2
3
measured in a liquid scintillation
counter. Experiments were carried
out in duplicates.
members of the steroid hormone receptor superfamily are
tightly coupled processes, the binding of NS8 possibly
favors the heterodimerization and subsequent transactiva-
tion events needed for effective inhibition of cell prolifera-
tion. NS8 could be a vitamin D derivative which shows a
3
dissociation of the multiple vitamin D effects at a tran-
3
scriptional level.
All three derivatives display only weak effects on adi-
pocyte differentiation (Figure 5) because of the high DBP-
affinities and low VDR-affinities of 25(OH)D3 and NS8,
and the extremely weak VDR-affinity of NS3.
The dose-response relationships of DBP-binding of the
1
␤-derivatives NS3 and NS8 differ significantly from each
other. Bishop et al. have previously found that NS8 has a
2
3
higher affinity for DBP than 1,25(OH) D . In Figure 4,
2
3
NS8 does not show the sigmoidal binding curve typical for
equilibrium binding experiments. The higher affinity of
NS8 compared to that of 25(OH)D for DBP in the low
3
Ϫ10
Ϫ9
concentration range between 10
M and 10 M could
only be explained by the existence of more than one pop-
2
4
ulation of binding sites on DBP. At least one population of
binding sites seems to have a higher affinity for NS8 than
for 25(OH)D . A 2␣-chloro substituent led back to a sig-
3
moidal monophasic binding curve as shown by NS3. This
derivative seemed to have lost the ability to bind to high
affinity binding site population(s) typical for NS8 binding
3
Figure 4 Displacement of [ H]-25(OH)D from sheep serum vi-
3
tamin
D binding protein by increasing concentrations of
1
,25(OH) D , 25(OH)D , NS3 and NS8. A 25 (OH)D assay system
2 3 3 3
and had a lower affinity for the 25(OH)D -specific binding
3
was used to determine the affinity of vitamin D
derivatives to
3
site than 25(OH)D . A possible explanation for this surpris-
serum DBP. DBP from sheep serum were incubated with
3
3
2
5(OH)[26,27-methyl- H] vitamin D3 and increasing concentra-
ingly different behavior could be differences in the A-ring
conformation of these two A-ring derivatives of vitamin D3.
The 2␣-chloro substituent and the 1␤- and 3␤-hydroxy
functions in NS3 stabilize a conformation with the three
substituents in the equatorial position. Therefore, the whole
tions of vitamin D3 derivatives for 2 h at 4°C. Bound and free
3
[
H]-25(OH)D were separated by the addition of dextran coated
3
3
charcoal suspension. Displacement of [ H]-25(OH)D3 was mea-
sured in a liquid scintillation counter. Experiments were carried
out in duplicates.
3
4
Steroids, 1998, vol. 63, January

Synthesis and biological activities of 1-epicalcitriols: Sch o¨ necker et al.
Figure 5 Interrelation of adipocyte differentiation and inhibition of cell proliferation of C3H10T1/2(BMP-4)-cells by 1,25(OH) D ,
2
3
2
5(OH)D , NS3 and NS8. C3H10T1/2(BMP-4)-cells were cultured to confluence in 12-well plates. At confluence, the medium was
3
replaced with DMEM supplemented with 10% FCS, 50 ␮g/mL ascorbate, and 10 mM ␤-glycerophosphate and treated with vitamin D3
derivatives or solvent. Every 4–5 days, halve of the medium was replaced with fresh medium including the vitamin D derivatives. After
3
a culture period of 19 days, the cells were washed and after addition of nile red and washing, the neutral lipid content was determined
in a Cytofluor 2350 fluorescence photometer (excitation 485 nm/emission 530 nm). Subsequently the cells were fixed in methanol
treated with methylene blue solution for 20 min. After washing and lysis of the cells the absorbance was measured in a multiscan
photometer at 650 nm and correlated with a standard curve. †, 1,25(OH) D -induced cell lysis. Experiments were carried out in
2
3
triplicates. The results shown are representative of two to three experiments.
A-ring is fixed in one of the two chair conformations. This
seems to be unfavorable for at least one population of
binding sites for which NS8 has higher affinity than
is gratefully acknowledged. We thank G. Gross for provid-
ing the BMP cells.
2
5(OH)D . Both chair conformations in NS8 are, in contrast
3
to NS3, energetically possible. However, it is unclear which
specific A-ring conformation is favored for binding to the
postulated population of high and low affinity binding sites
and which other structural elements of the ligand are im-
portant for optimal binding.
The in vitro biological effects of the examined 1␤-
References
1
.
Itin PH, Pittelkow MR, Kumar R (1994). Effects of vitamin D
metabolites on proliferation and differentiation of cultured human
epidermal keratinocytes grown in serum-free or defined culture
medium. Endocrinology 135:1793–1798.
2
5
2
.
Franceschi RT, James WM, Zerlauth
G
(1985). 1␣,25-
dihydroxyvitamin D3 regulation of growth, morphology, and fi-
bronectin in a human osteosarcoma cell line. J Cell Phys 123:401–
409.
vitamin D derivatives are, as expected from the VDR
binding data, weak compared to those of 1,25(OH) D .
3
2
3
However, a closer examination of the heterodimerization
properties of the VDR in the presence of NS8 and the
unusually high affinity binding of NS8 to new binding
site(s) on the DBP, especially at low metabolite concentra-
tions, could be interesting for the modulation of specific
transcriptional activities and of pharmacokinetical charac-
teristics of other highly biologically active vitamin D3
derivatives.
3.
Clohisy DR, Bar-Shavit Z, Chappel JC, Teitelbaum SL (1987).
1
,25-dihydroxy-vitamin D3 modulates bone marrow macrophage
precursor proliferation and differentiation. J Biol Chem 262:15922–
5929.
Thavarajah M, Evans DB, Kanis JA (1991). 1,25(OH) D induces
differentiation of osteoclast-like cells from human bone marrow
cultures. Biochem Biophys Res Commun 176:1189–1195.
1
4
.
.
2
3
5
Mathieu C, Waer M, Casteels K, Laureys J, Bouillon R (1995).
Prevention of Type 1 diabetes in NOD mice by nonhypercalcemic
doses of a new structural analog of 1,25-dihydroxyvitamin D , KH
3
1
060. Endocrinology 136:866–872.
6
7
.
.
Okamura WH, Palenzuela JA, Plumet J, Midland MM (1992).
Vitamin D: structure-function analyses and the design of analogs.
J Cell Biochem 49:10–18.
Binderup L, Bramm E (1988). Effects of a novel vitamin D analogue
MC 903 on cell proliferation and differentiation in vitro and on
calcium metabolism in vivo. Biochem Pharm 37:889–895.
Acknowledgments
Financial support of this work by the Deutsche Forschungs-
gemeinschaft (Scho 488/1-1, Scho 488/1-2, Ma 852/6-1,
Scho 488/1-3, Ma 852/6-2) and by the Schering AG (B.S.)
Steroids, 1998, vol. 63, January 35

Papers
8
.
Norman AW, Nemere I, Muralidharan KR, Okamura WH (1992).
␤,25(OH) -vitamin D stimulated transcaltachia (the rapid hor-
monal stimulation of intestinal calcium transport). Biochem Biophys
Res Commun 189:1450–1456.
Posner GH, Cho C-G, Anjeh TEN, Johnson N, Horst RL, Kobayashi
T, Okano T, Tsugawa N (1995). 2-Fluoroalkyl A-ring analogs of
Dihydroxy-cholecalciferol (Calcitriol) aus 1␣,3␤,25-Trihydroxy-
cholesta-5,7-dien (Procalcitriol). J Prakt Chem 338:634–641.
Reichenb a¨ cher M, Gliesing S, Hopf G (1991). Eine verbesserte
Apparatur f u¨ r die Flash-Chromatographie. J Prakt Chem 333:779–
784.
Ponsold K (1963). Stickstoffhaltige Steroide, IV: Darstellung von
Azidohydrinen und vicinalen Aminoalkoholen aus Epoxysterinen.
Chem Ber 96:1855–1864.
Glotter E, Krinsky P (1978). Neighbouring group effects in the
acid-catalysed opening of steroidal epoxides. J Chem Soc Perkin
Trans 1 413–419.
Oshima E, Sai H, Takatsuto S, Ikekawa N, Kobayashi Y, Tanaka Y,
DeLuca HF (1984). Synthesis and biological activity of 1␣-fluoro-
1
2
3
16.
17.
9
.
1
,25-dihydroxyvitamin D . Stereocontrolled total synthesis via in-
3
tramolecular and intermolecular Diels–Alder cycloadditions. Pre-
liminary biological testing. J Org Chem 60:4617–4628.
Ahrens M, Ankenbauer T, Schr o¨ der D, Hollnagel A, Mayer H,
1
8.
1
0.
Gross
G (1993). Expression of human bone morphogenetic
1
19.
proteins-2 or -4 in murine mesenchymal progenitor C3H10T ⁄
2
cells
induced differentiation into distinct mesenchymal cell lineages.
DNA Cell Biol 12:871–880.
Scheddin D, Mayer H, Wittmann S, Sch o¨ necker B, Gliesing S,
Reichenb a¨ cher M (1996). Synthesis and biological activities of
2
5-hydroxyvitamin D . Chem Pharm Bull (Tokyo) 32:3525–3531.
3
2
2
0.
1.
Eyley SC, Williams DH (1975). Photolytic productions of vitamin
D. The preparative value of a photo-sensitizer. J Chem Soc, Chem
Commun 858.
Muralidharan KR, de Lera AR, Isaeff SD, Norman AW, Okamura
WH (1993). Studies on the A-ring diastereomers of 1␣,25-
1
1.
2.
2
6-hydroxy-27-nor-derivatives of 1␣,25-dihydroxyvitamin D . Ste-
3
roids 61:598–608.
1
Wittmann S, Krieg R, Prousa R, Sch o¨ necker B, Droescher P (1996).
Transformation of (20S)-20-(hydroxymethyl)pregna-1,4-dien-3-one
into (20S)-20-(p-toluenesulfonyloxymethyl)pregna-1,4-dien-3␤-ol
and -3␣-ol, intermediates of vitamin D derivatives. J Prakt Chem
dihydroxyvitamin D . J Org Chem 58:1895–1899.
3
2
2
2.
3.
Thompson EB (1995). Membrane transporters of steroid hormones.
Curr Biol 5:730–732.
Bishop JE, Collins DE, Okamura WH, Norman AW (1994). Profile
of ligand specificity of the vitamin D binding protein for 1␣,25-
3
38:214–219.
1
3.
4.
Sch o¨ necker B, Prousa R, Reichenb a¨ cher M, Gliesing S, Kosan H,
Droescher P, Hauschild U, Thieroff-Ekerdt R (1993). Synthesis of
dihydroxyvitamin D and its analogs. J Bone Min Res 9:1277–1288.
3
2
2
4.
5.
Hulme EC, Birdsall NJM (1992). Strategy and tactics in receptor-
binding studies. In: Hulme EC (ed), Receptor-Ligand Interactions:
A Practical Approach. IRL Press, Oxford, New York, Tokyo, pp.
3␤-acetoxy-1␤,2␤-epoxy-25-hydroxy-cholesta-5,7-diene and 2␤,25-
dihydroxyvitamin D . Bioorg Med Chem Lett 3:1849–1854.
3
1
Sch o¨ necker B, Droescher P, Adam G, Marquardt V, Walther D,
Fischer R, Prousa R (1992). Ausgangsverbindungen zur Herstellung
von Calcitriol sowie dessen Abk o¨ mmlingen, Verfahren zur Herstel-
lung dieser Ausgangsverbindungen sowie Zwischenprodukte f u¨ r
dieses Verfahren. PCT/DE 92/00128. Chem Abstr (1993)
6
3–176.
Okamura WH, Midland MM, Hammond MW, Rahman NA, Dor-
manen MC, Nemere I, Norman AW (1994). Conformation and
related topological features of Vitamin D: structure-function rela-
tionships In: Norman AW, Bouillon R, Thomasset M (eds), Vitamin
D, a Pluripotent Steroid Hormone: Structural, Molecular Endocri-
nology and Clinical Applications. Walter de Gruyter, Berlin, New
York, pp. 12–19.
1
18:124879.
1
5.
Reichenb a¨ cher M, Gliesing S, Lange C, Gonschior M, Sch o¨ necker
B (1996). Ein effektives Verfahren zur Herstellung von 1␣,25-
3
6
Steroids, 1998, vol. 63, January