Papers
vacuo. The residue was purified by preparative TLC (5% EtOAC/
hexane) to give (18.45 mg, 72% yield) of 10E-19-(2-
bromoacetoxy)methyl-6R-methoxy-3,5-cyclovitamin D3 (13). 1H
NMR: ␦ 0.50 (3H, s, C18-H), 0.81 (6H, d, C26,27-H), 0.89 (3H, d,
C21-H), 1.03–2.01 (m), 2.54 (m), 3.21 (3H, s, OMe), 3.82 (2H, s,
CH2Br), 4.18 (2H, d, C6-H), 4.59–4.81 (2H, m, OCH2), 4.95 (1H,
d, C7-H), 5.73 (1H, t, C19-H). In a similar manner, 10E-25-
hydroxy-19-(2-bromoacetoxy)methyl-6R-methoxy-3,5-
cyclovitamin D3 (14). (7.12 mg, 70% yield) was obtained from
10E-25-hydroxy-19-(hydroxymethyl)-6R-meth-
oxy-3,5-cyclovitamin D3 (12) (8 mg, 0.02 mmol), BrCH2CO2H
(7.49 mg, 0.05 mmol, 3 eq.), DCC (18.56 mg, 0.09 mmol, 5 eq.),
and DMAP (3.297 mg, 0.027 mmol, 1.5 eq.).
(1H, d, C6-H), 6.70 (1H, d, C5-phenyl H), 7.14 (1H, d, C6-phenyl
H), 7.98 (1H, d, C3-phenyl H), 8.34 (1H, t, NH).
Competitive radioligand binding assays of 25-OH-D3, (17) and
(18) with human serum DBP. These assays were carried out
according to published procedure.13 In general, samples of DBP
(200 ng), 3H-25-OH-D3 (3,500 cpm in 10 L of ethanol) and
various concentrations of either 25-OH-D3 or (17) or (18) (1
pmol–63.8 pmol) were incubated in a buffer (50 mM Tris.HCl, 150
mM NaCl, 1.5 mM EDTA, 0.1% Triton X-100, pH 8.3, total
volume 0.5 mL) at 4°C for 20 h, followed by the incubation with
of ice-cold slurry of Dextran-coated charcoal,13 centrifugation and
radioactive counting.
5E[19-(2-Bromoacetoxy)methyl]vitamin D3 (16). A mixture of
(13) (10 mg, 0.018 mmol) and p-TsOH.H2O (1.38 mg, 0.07 mmol,
0.4 eq.) in 1 mL of a 3:1 mixture 1,4-dioxane:H2O was heated for
15 min at 55°C. After cooling to room temperature, 5 mL of
EtOAC was added and the mixture washed with 2 mL of saturated
aqueous NaHCO3 solution. The organic phase was separated and
dried with anhydrous MgSO4 and evaporated in vacuo. Preparative
TLC (25% EtOAC/hexane) of the residue gave (4.59 mg, 48.70%
yield) of compound (16).
In a similar manner, 5E-[19-(2-bromoacetoxy)methyl] 25-
hydroxyvitamin D3 (17) (2.13 mg, 45% yield) was obtained from
10E-25-hydroxy-19-(2-bromoacetoxy)methyl-6R-methoxy-3,5-
cyclovitamin D3 (14) (5 mg, 0.008 mmol) and p-TsOH.H2O (0.67
mg, 0.004 mmol, 0.40 eq.). 1H NMR of (17) (400 MHz, in CDCl3:
␦ 0.56 (3H, s, C18-H), 0.95 and 0.96 (6H, two singlets, C26,27-Hs),
1.01–2.82 (m), 3.81 (2H, s, CH2Br), 3.99 (1H, m, C3-H), 4.90 (2H,
d, CH2OCO), 5.43 (1H, t, C19-H), 5.84 (1H, d, C7-H), 6.22 (1H, d,
C6-H).
Results and discussion
Development of affinity/photoaffinity analogs require, as a
general rule, synthesis of corresponding radiolabeled analog
in a few short steps using a commercially available radio-
active precursor of high specific activity.26 Faced with these
restrictions, synthesis of C19-modified affinity/photoaffinity
reagents by the procedure of Yamada et al., involving
base-catalyzed functionalization of the sulfur-diooxide ad-
duct of vitamin D3,27 deemed completely impractical.
In 1983, Paaren et al. reported the oxidation of C10–19
methylene of 25-hydroxy 3,5-cyclovitamin D3 to its C19-
nor-C19-keto derivative.25 This information provided us
with a starting point to elaborate the C19-position of vitamin
D molecule. We first set out to do model studies using
vitamin D3 due to the prohibitive cost of 25-OH-D3. Cyclo-
vitamin D3 Ketone (5) was prepared in an improved yield of
85% from 6R-methoxy-3,5-cyclovitamin D3 (3) by a mod-
ification of the of the published procedure.23 We examined
the reaction of this ketone using a variety of Wittig or
Horner–Emmons reagents (Figure 1). In our hands, only
diethylcyanomethylphosphonate [(EtO)2POCH2CN] gave a
fruitful Horner–Emmons reaction with this ketone using
DME as the solvent, to give a 10:1 mixture of the E and
Z-isomers (7a) and (7b) separable by preparative TLC in a
combined yield of 80%.24 The following Wittig or Horner–
Emmons reagents did not give any fruitful reaction with
ketone (5): (EtO)2POCH2COOEt, [Ph3Pϩ-CH2CH2CH-
(OCH2CH2O) BrϪ, Ph3Pϩ-CH2CH-(OCH2CH2O) BrϪ,
(EtO)2POCN, (EtO)2POCH2ACH2CO2Et. Also, the reac-
tion with (EtO)2POCH2CN does not occur in THF or tolu-
ene. The structure of the Horner–Emmons adducts were
confirmed by their 1H-NMR spectra showing a single C19-H
at ␦5.6 for the E-isomer (7a) and ␦5.8 for Z-isomer (7b).
The C19-H of (7a) was expected to be more shielded by the
7,8-olefinic electrons than that of (7b). An NOE experiment
was conducted on each isomer to establish their stereochem-
istry. In the E-isomer (7a), a positive NOE was observed in
10E-6R-methoxy-25-hydroxy-3,5-cyclovitamin D3-19-methyl-
[(4-azido-2-nitro)phenyl]glycinate (15). A mixture of (12) (10
mg, 0.02 mmol), 4-azido-2-nitrophenylglycine16 (15.93 mg, 0.067
mmol, 3 eq.), DCC (23 mg, 0.112 mmol, 5 eq.), and DMAP (4.09
mg, 0.033 mmol, 1.5 eq.) was stirred in 1 mL of anhydrous CH2Cl2
at room temperature for 30 min in the dark. The reaction was
quenched with 5 mL of H2O, and the mixture was evaporated with
EtOAC (3 ϫ 5 mL), filtered, dried with anhydrous MgSO4, and
concentrated in vacuo. The residue was purified by preparative
TLC (25% EtOAC/hexane) to give 6.5 mg (79% yield) of 10E-
6R-methoxy-25-hydroxy-3,5-cyclovitamin D3-19-methyl[(4-azi-
do-2-nitro)phenyl]glycinate (15). 1H NMR: ␦ 0.48 (3H, s, C18-H),
0.91–0.92 (6H, two singlets, C26,27-H), 0.88 (3H, d, C21-H), 1.01–
2.03 (m), 3.22 (3H, s, OMe), 3.81 (2H, s, CH2Br), 4.16 (3H, m,
C6-H and -NCH2), 4.61 and 4.62 (2H, dd, OCH2), 4.95 (1H, d,
C7-H), 5.71 (1H, t, C19-H), 6.71 (1H, d, C5-phenyl), 7.12 (1H, d,
C6-phenyl), 7.89 (1H, C3-phenyl), and 8.34 (1H, t, NH).
25-hydroxyvitamin D3-19-methyl[(4-azido-2-nitro)phenyl]gly-
cinate (18). To a solution of 10E-6R-methoxy-25-hydroxy-3,5-
cyclovitamin D3-19-methyl[(4-azido-2-nitro)phenyl]glycinate (15)
(5 mg, 0.008 mmol) in 1 mL of 3:1 mixture of 1,4-dioxane-H2O,
was added p-TsOH.H2O (0.57 mg, 0.003 mmol, 0.4 eq.), and the
solution was stirred at room temperature for 30 min in the dark.
After cooling to room temperature, EtOAc (3 ϫ 5 mL) was added
to extract the organic layer. The organic layer was separated from
the aqueous layer, filtered, dried with anhydrous MgSO4, and
evaporated in vacuo. Preparative TLC (40% EtOAC/Hexane) of
1
the H resonance signal corresponding to the C19-H single
proton when the C7-H was irradiated. No NOE effect was
1
observed in the H resonance signal of the corresponding
hydrogen of the Z-isomer (7b). Reduction of the major,
E-isomer (7a) with DIBAL followed by acid treatment gave
the aldehyde (9) (60% yield), which, on reduction with
NaBH4, gave the alcohol (11) (55% yield). The affinity
label (16) was obtained by treatment of (11) with
BrCH2COOH in the presence of DCC and DMAP (72%
1
the crude extract gave 2.34 mg, 48% yield of (18). H NMR: ␦
0.51 (3H, s, C18-H), 0.89 and 0.91 (6H, two singlets, C26,27-H),
1.02–2.81 (m), 3.98 (1H, m, C3-H), 4.07 (2H, d, CH2O2C), 4.84
(2H, d, NHCH2CO2), 5.45 (1H, t, C19-H), 5.89 (1H, d, C7-H), 6.23
220 Steroids, 1998, vol. 63, April