A-ring hydroxymethyl 19-nor analogs of the natural hormone 1α,25-dihydroxyvitamin D3: Synthesis and preliminary biological evaluation

Article abstract of DOI:10.1016/j.bmc.2005.04.010

A series 5-8 of 1- and 3-CH₂OH 19-nor analogs of the hormone calcitriol (1) has been prepared. Surprisingly, 19-nor 1α-CH₂OH analog 5a is more antiproliferative at 100 nM concentration than the corresponding regioisomeric analog 6a w

Full text of DOI:10.1016/j.bmc.2005.04.010

Bioorganic & Medicinal Chemistry 13 (2005) 3964–3976
A-ring hydroxymethyl 19-nor analogs of the natural
hormone 1a,25-dihydroxyvitamin D3: synthesis and
preliminary biological evaluation
Mark A. Hatcher,a Sara Peleg,b Patrick Dolan,c Thomas W. Kensler,c
Amy Sarjeanta and Gary H. Posnera,*
aDepartment of Chemistry, The Johns Hopkins University, Baltimore, MD 21218, USA
bDepartment of Medical Specialties, The University of Texas, M. D. Anderson Cancer Center, Houston, TX 77030, USA
cDivision of Environmental Health Sciences, Bloomberg School of Public Health, Johns Hopkins University,
Baltimore, MD 21205, USA
Received 17 February 2005; revised 1 April 2005; accepted 5 April 2005
Available online 4 May 2005
Abstract—A series 5–8 of 1- and 3-CH2OH 19-nor analogs of the hormone calcitriol (1) has been prepared. Surprisingly, 19-nor 1a-
CH2OH analog 5a is more antiproliferative at 100 nM concentration than the corresponding regioisomeric analog 6a with the nat-
ural 1a-OH group, and 1a-CH2OH hybrid analog 7a is similar in antiproliferative potency to calcitriol (1) even at low nanomolar
concentrations.
ꢀ 2005 Elsevier Ltd. All rights reserved.
1. Introduction
The natural hormone 1a,25-dihydroxyvitamin D3 (1,
calcitriol) regulates not only calcium and phosphorus
levels in our bodies, but it regulates also proliferation
and differentiation of human cells.1,2 Medicinal organic
chemists have prepared a series of calcitriol analogs that
are relatively nontoxic (low calcemic) and that are effica-
cious in chemotherapy of such human diseases as
secondary hyperparathyroidism, psoriasis, and osteopo-
rosis.3,4 More than six analogs of calcitriol are now
clinically used designer drugs promoting healthier
living.4
Challenging the accepted wisdom in the vitamin D field
at the time, we prepared a homolog of the natural hor-
mone containing a 1-hydroxymethyl group in place of
the natural 1-OH group that had been considered essen-
tial for biological activity; homolog 2a had measurable
but somewhat lower in vitro antiproliferative activity
and desirably much lower calcemic activity than the nat-
ural hormone calcitriol (1).5 Combining this 1-hydroxy-
Keywords: Calcitriol; 19-Nor; Hydroxymethyl A-ring; Novel analogs.
*
methyl replacement with a potentiating side chain led
to hybrid analog 3a that incorporated the calcemia
Corresponding author. Tel.: +1 410 516 4670; fax: +1 410 516
8420; e-mail: ghp@jhu.edu
0968-0896/$ - see front matter ꢀ 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2005.04.010

M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
3965
ablating 1-hydroxymethyl group, a potentiating 16,17-
double bond, and a catabolism-blocking 24,24-difluoro
unit.6 We have reported that this hybrid analog 3a is
80–100 times less calcemic than calcitriol,7 inhibits mul-
tistage skin tumorigenesis when administered topically
to mice,8 and reduces the size of xenograft tumors of hu-
man neuroblastoma cells when administered systemi-
cally (IP) to mice.9 Hybrid analog 3a is significantly
more antiproliferative than the natural hormone.6 Hy-
brid analog 4a having a 1-CH2OH group and a KH-
1060 potentiating side chain is as antiproliferative in vi-
tro as the natural hormone 1 even at low nanomolar
concentrations and is substantially less calcemic than
calcitriol (1).10 Now we report synthesis and preliminary
biological evaluation of a new series 5–8 of calcitriol
analogs lacking the 19-methylene group (i.e., 19-nor
analogs)11 but containing a hydroxymethyl group either
at the 1-position (analogs 5 and 7) or at the 3-position
(analogs 6 and 8).
the mono-silyl protected allylic alcohol ( )-16 in 71%
yield.16
This allylic alcohol was oxidized to enone ( )-17, which
was then selectively epoxidized with hydrogen peroxide
to give epoxy ketone ( )-18 in predominantly the
trans-orientation (>95%).19,20 Wittig reaction using tri-
ethyl phosphonoacetate and epoxy ketone ( )-18 gave
a 1.5:1 ratio of ( )-19-Z to ( )-19-E in overall 95% yield.
These two geometrical isomers could be cleanly sepa-
rated by standard flash column chromatography and ta-
ken on to their respective A-ring phosphine oxides
(Scheme 2). The olefin geometry of ( )-19-Z was easily
1
assigned by H NMR in which the downfield doublet
(4.62 ppm) correlates to the proton attached to the
epoxide at the c-position with respect to the ethyl ester.
This doublet proton is much more downfield with re-
spect to the proton of ( )-19-E (3.36 ppm) due to the
deshielding effect of the ethyl ester carbonyl group in
2. Results
( )-19-Z.21,22 As seen in Eqs. 1 and 2, c-epoxy-a,b-
unsaturated esters ( )-19-E and ( )-19-Z were subjected
separately to a palladium catalyzed reductive opening,
which unfortunately gave inseparable mixtures of E
and Z products.19,20,23
2.1. Chemistry
Preparation of the 19-nor hybrid analogs 5–8 can be ret-
rosynthetically envisioned through the disconnections
shown in Scheme 1. Analogs 5a and 5b can be prepared
through Horner–Wadsworth–Emmons coupling of
C,D-ring ketone (+)-912,13 and 1-hydroxymethyl-A-ring
phosphine oxide ( )-10. Likewise, analogs 6a and 6b can
be prepared through coupling of C,D-ring ketone (+)-9
and 3-hydroxymethyl-A-ring phosphine oxide ( )-11.
The two isomeric A-ring phosphine oxides ( )-10 and
( )-11 can be further disconnected back to readily avail-
able 3-cyclohexene-1-carboxylic acid ( )-12.14–18
ð1Þ
ð2Þ
As shown in Scheme 2, carboxylic acid ( )-12 was re-
acted under standard iodolactonization conditions to
give exclusively the trans-iodolactone ( )-13 in excellent
yield.16–18 After elimination of the iodide with fresh
DBU and reductive opening of lactone ( )-14, diol
( )-15 was selectively protected at the primary alcohol
position with tert-butyl dimethylsilyl chloride to give

3966
M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
Scheme 1.
Scheme 2.
This setback steered us in a new direction for the synthe-
sis of the A-ring phosphine oxides using an in situ regio-
selective bis-reduction of both the epoxide and ester
moiety (Scheme 3). Epoxy ester ( )-19-E was reacted
with DIBAL-H to open the epoxide as well as to reduce
the ester to the allylic alcohol. The hydride was selec-
tively delivered to the c carbon as expected from good
literature analogy.24

M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
3967
Scheme 3.
The resulting bis-alcohol was then selectively protected
at the primary allylic alcohol to give the pivaloyl ester
( )-20.25 After protection of the secondary alcohol as
the tert-butyl dimethylsilyl ether and removal of the
pivaloyl ester by lithium aluminum hydride (LAH)
reduction, allylic alcohol ( )-21 was tosylated, reacted
with lithium diphenylphosphide, and finally oxidized
with hydrogen peroxide to give the desired 19-nor-1-
hydroxymethyl-A-ring phosphine oxide ( )-10 in a total
of 14 steps and 9% overall yield from ( )-12. The 19-
nor-3-hydroxymethyl-A-ring phosphine oxide ( )-11
was prepared in a similar fashion in 14 steps and 16%
overall yield (Scheme 3). Coupling of the natural calci-
triol C,D-ring (+)-9 with the 1-hydroxymethyl-A-ring
( )-10 and subsequent deprotection with TBAF gave
the desired 19-nor-1-hydroxymethyl hybrid analogs 5a
and 5b as 1:2 ratio of diastereomers in 57% yield
(Scheme 4).26 Separation of the diastereomers by HPLC
gave enantiomerically pure analogs (+)-5a and (+)-5b.
merically pure (+)-6a and (+)-6b. The more potentiating
KH-1060 C,D-ring side chain (ꢀ)-2410,27,28 was coupled
with the 1-hydroxymethyl-A-ring (+)-10 with subse-
quent deprotection using TBAF to give analogs 7a
and 7b as a 1:2 ratio of diastereomers in 50% yield,
which then gave enantiomerically pure (ꢀ)-7a and (ꢀ)-
7b after HPLC separation (Scheme 5).
Coupling of the KH-1060 C,D-ring side chain (ꢀ)-24
with enantiomerically pure (1a,3b)-3-hydromethyl A-
ring (+)-11 (99.5% ee by Chiral HPLC) followed by
TBAF deprotection gave enantiomerically pure analog
(+)-8 in 16% yield.
The stereochemical assignment of these 19-nor-hybrid
analogs is based on three grounds: first, enantiomeri-
cally enriched 1-hydroxymethyl-A-ring (ꢀ)-10 and 3-
hydroxymethyl-A-ring (ꢀ)-11 were synthesized using
enantiomerically enriched (R)-3-cyclohexene-1-carbox-
ylic acid (+)-12, prepared by recrystallization of the cor-
responding brucine salt (Scheme 6).17
Following the same procedure, 3-hydroxymethyl-A-ring
( )-11 was coupled with C,D-ring ketone (+)-9 to give
19-nor-3-hydroxymethyl hybrid analogs 6a and 6b in
74% yield, which upon HPLC separation gave enantio-
With both enantiomerically enriched A-rings in hand,
they were both coupled once again with C,D-ring ketone

3968
M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
Scheme 4.
Scheme 5.
(+)-9 to give analogs (+)-5b and (+)-6b, respectively.
These enantiomerically pure analogs were characterized
previously prepared analogs. Second, a sufficiently re-
solved X-ray crystal structure29 was obtained of analog
(+)-6b.
1
by both H NMR and by co-injection on HPLC with

M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
3969
Scheme 6.
As depicted in Figure 1, analog (+)-6b has the 1-OH in
the b-position and the 3-hydroxymethyl in the a-posi-
tion. This X-ray evidence of the 1b,3a stereochemistry
in analog (+)-6b was then correlated to show that analog
(+)-5b indeed has the 1b,3a stereochemistry as well. Fi-
nally, all the analogs containing a 1-hydroxy-A-ring, 19-
methylene or 19-nor, share similar characteristic 1H
NMR signals as well as optical rotation trends. As seen
in Table 1, the analogs containing the 1a,3b stereochem-
1
istry have a more upfield C-18 methyl signal in the H
NMR spectrum and a more positive optical rotation.30
2.2. Biology
Our standard murine keratinocyte assay for in vitro
antiproliferative activity6 showed that 1-a-CH2OH ana-
log 5a is as expected, considerably more potent than its
diastereomeric 1-b-CH2OH analog 5b (Fig. 2). Surpris-
ingly, however, 19-nor 1-a-CH2OH analog 5a is more
antiproliferative at 100 nM concentration than the cor-
responding regioisomeric analog 6a containing the natu-
ral 1-a-OH group (Fig. 2); this is the first report of any
1-a-CH2OH analog with the natural hormonal C,D-ring
side chain being more antiproliferative than the corre-
sponding 1-a-OH analog. This unusually high antipro-
liferative activity of the 1-a-CH2OH analog 5a is
precedented to some degree by the exceptionally high
antiproliferative activity of 1-a-CH2OH hybrid analog
3a;6 the 1-a-CH2OH group (being larger than a 1-a-
OH group) may alter the A-ring conformation so as to
raise antiproliferative potency. A similar but less pro-
nounced trend was seen between hybrid KH-1060-like
analogs 7a and 8 (Fig. 3). It is noteworthy also that 1-
a-CH2OH hybrid analog 7a is comparable in antiprolif-
erative potency to calcitriol 1 even at low nanomolar
concentrations (Fig. 3).
Figure 1. X-ray structure of 19-nor VD3 analog 6b.

3970
M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
Table 1. Characteristic 1H NMR C-18 methyl signals (ppm) and optical rotations of hybrid analogs 2–5 and 7
Analog 2a (1a,3b) 2b (1b,3a) 3a (1a,3b) 3b (1b,3a) 4a (1a,3b) 4b (1b,3a) 5a (1a,3b) 5b (1b,3a) 7a (1a,3b) 7b (1b,3a)
C-18
25
½aꢁD
0.50
+24.0
0.54
ꢀ64.0
0.66
+78.0
0.68
ꢀ1.3
0.52
+4.0
0.55
ꢀ140.0
0.54
+40.3
0.56
+24.1
0.55
ꢀ8.0
0.57
ꢀ36.7
sponding regioisomeric analog (6a) having the natural
1a-OH group. Incorporating in one new analog a 1a-
CH2OH group, the absence of a 19-methylene group,
and a KH-1060 potentiating side chain makes analog
7a comparable to the natural hormone calcitriol (1) in
antiproliferative potency even at physiologically rele-
vant low nanomolar concentrations. These SAR gener-
alizations should facilitate design of even more potent
and safe new vitamin D analogs.
4. Experimental
Unless otherwise noted, all reactions were performed in
oven-dried glassware stirred under an atmosphere of ul-
tra-high-purity argon. THF was distilled from Na/ben-
zophenone ketyl and CH2Cl2 distilled from CaH2
immediately prior to use. Organolithiums were titrated
prior to use following known methods. All other re-
agents were used as received from commercial suppliers.
Analytical TLC analysis was conducted on precoated
glass-backed silica gel plates (Merck Kieselgel 60
F254, 250 mm thickness) and visualized with p-anisalde-
hyde or KMnO4 stains. Column chromatography was
performed using flash silica gel (particle size 230–
400 mesh). Preparative-plate chromatography was per-
formed using silica-gel-coated glass preparative plates
(500–1000 lm) from Analtech and analyzed by UV.
HPLC was carried out using a Rainin HPLXTM system
equipped with two 25 mL/min preparative pump heads
Figure 2.
using (1)
a
Chiral Technologies Chiralcelꢁ OJ
10 mm · 250 mm (semipreparative) column packed with
cellulose tris(4-methylbenzoate) on a 10 lm silica-gel
substrate or (2) a Chiral Technologies Chiralcelꢁ OD
10 mm · 250 mm (semipreparative) column and a Rai-
nin DynamaxTM UV-C dual-beam variable-wavelength
detector set at 254 nm. Yields are reported for pure
products (>95% based on their chromatographic and
spectroscopic homogeneity) and are unoptimized. Melt-
ing points were determined in open capillaries using a
Mel-Temp metal-block apparatus and are uncorrected.
Optical rotations were measured at the Na line using a
Perkin–Elmer 141 Polarimeter. NMR spectra were ob-
tained on a Varian XL-400 spectrometer operating at
400 MHz for 1H, 376 MHz for 19F, and 100 MHz for
13C and a Bruker 300 AMX spectrometer operating at
Figure 3.
Our standard protocol31 for VDR-mediated in vitro
transcriptional activity uses CV-1 cells co-transfected
with the recombinant human VDR plasmid and the
osteocalcin VDRE attached to the thymidine kinase
promoter-growth hormone reporter. This assay showed
1-a-hydroxymethyl analog 5a to be active but signifi-
cantly less potent (ED50 = 300 nM) than calcitriol (1,
ED50 = 1.0 nM, data not shown). Thus, as we have ob-
served before, transcriptional potency does not always
correlate directly with in vitro antiproliferative
potency.30
1
300 MHz for H. Chemical shifts are reported in ppm
1
(d) and are referenced to CDCl3 (7.26 ppm for H and
77.0 ppm for 13C), tetramethylsilane (TMS, 0.00 ppm
1
for H), and CFCl3 (0.00 ppm for 19F). IR spectra were
obtained using a Perkin–Elmer 1600 Series FT-IR
instrument. HRMS were obtained at the mass spectro-
metry facility at the Ohio State University on a Micro-
mass QTOF Electrospray mass spectrometer. Elemental
analyses were performed by Atlantic Microlab Inc.,
Norcross, GA.
3. Conclusion
We present here the first example in which a 1a-CH2OH
analog (5a) with the natural hormonal C,D-ring side
chain is more antiproliferative in vitro than the corre-

M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
3971
4.1. Iodolactone ( )-13
quenched with ice cold water (20 mL) and then extracted
with diethyl ether. The organics were combined, dried
over MgSO4, concentrated under reduced pressure and
purified by column chromatography (80% hexanes/
20% ethyl acetate) to give 1.967 g (71%) of diol monosi-
lyl ether 16 as an oil. 1H NMR (CDCl3): d 5.76–5.59 (m,
2H), 4.29 (m, 1H), 3.55–3.44 (m, 2H), 2.15–2.02 (m, 2H),
1.98–1.70 (m, 3H), 1.22–1.18 (m, 1H), 0.88 (s, 9H), 0.04
(s, 6H); 13C NMR (CDCl3): d 131.10, 128.30, 67.53,
67.45, 35.73, 28.18, 25.88, 18.29, ꢀ5.41; identical in all
respects to literature reports.16
A 250 mL flask was charged with carboxylic acid ( )-12
(1.770 g, 14.04 mmol) and a solution of NaHCO3 (3.54 g,
42.12 mmol) in water (35 mL) and stirred until homoge-
neous. To this mixture was added a solution of iodine
(3.74 g, 14.75 mmol) and KI (13.98 g, 84.24 mmol) in
water (35 mL) and then immediately protected from light
by wrapping aluminum foil around the flask. The reac-
tion was stirred for 24 h at room temperature followed
by extraction with CHCl3 (3 · 30 mL). The organics
were combined, washed with Na2S2O3 (100 mL, 10% in
water) and NaHCO3 (100 mL, 10% in water), dried over
MgSO4, and then removed under reduced pressure to
give 3.397 g (96%) of iodolactone 13 as a yellow solid.
4.5. Enone ( )-17
Diol monosilyl ether ( )-16 (1.658 g, 6.818 mmol) was
dissolved in methylene chloride (25 mL, anhydrous)
˚
and then 4 A molecular sieves (7 g) was added. To this
Mp 132–134 ꢂC; 1H NMR (CDCl3):
J = 5.6 Hz, 1H), 4.5 (t, J = 4.4 Hz, 1H), 2.79 (d,
J = 12.4 Hz, 1H), 2.68–2.65 (m, 1H), 2.48–2.35 (m,
3H), 2.10 (dd, J = 16.8, 5.2 Hz, 1H), 1.94–1.79 (m, 2H);
identical in all respects to literature reports.16,17
d
4.82 (t,
mixture was added N-methylmorpholine oxide (1.597 g,
13.63 mmol) and then tetrapropyl ammonium perruthe-
nate (0.120 g, 0.341 mmol). The reaction was stirred at
room temperature for 1.5 h. The reaction was then
passed over a plug of silica and the organics were com-
bined, dried over MgSO4, concentrated under reduced
pressure, and purified by column chromatography
(80% hexanes/20% ethyl acetate) to give 1.460 g (89%)
4.2. Olefinic lactone ( )-14
Iodolactone ( )-13 (3.394 g, 13.46 mmol) was dissolved in
toluene (40 mL, anhydrous) and fresh 1,3-diazabicyclo-
[5.4.0]undec-7-ene (DBU, 3.06 mL, 20.19 mmol). This
was heated at reflux for 6 h and then cooled down to
room temperature. The reaction was filtered, concen-
trated under reduced pressure, and purified by silica
gel chromatography (60% hexanes/40% ethyl acetate)
to give 1.593 g (95%) of olefinic lactone 14 as a pale-yel-
low oil. 1H NMR (CDCl3): d 6.20 (m, 1H), 5.83–5.79 (m,
1H), 4.72 (m, 1H), 2.87 (m, 1H), 2.52–2.38 (m, 3H), 2.06
(d, J = 11.2 Hz, 1H); identical in all respects to literature
reports.16,17
1
of enone 17 as a yellow oil. H NMR (CDCl3): d 6.97
(ddd, J = 10.0, 6.0, 2.8 Hz, 1H), 6.01 (d, J = 9.2 Hz,
1H), 3.59–3.51 (m, 2H), 2.50–2.33 (m, 2H), 2.32–2.20
(m, 3H), 0.88 (s, 9H), 0.037 (s, 6H); 13C NMR (CDCl3):
d 199.78, 149.71, 129.63, 66.05, 40.82, 37.69, 28.57, 25.83,
18.25, ꢀ5.47, ꢀ5.49; IR (neat, cmꢀ1) 2929, 2857, 1682,
1471, 1386, 1113, 837; HRMS (CI) m/z (M+Na) calcd
263.143775 for C13H24O2SiNa+, found 263.144255.
4.6. Epoxy ketone ( )-18
4.3. Diol ( )-15
Enone ( )-17 (1.437 g, 5.977 mmol) was dissolved in
methanol (100 mL) and then hydrogen peroxide (30%,
6.78 mL, 59.77 mmol) was added and the mixture was
cooled to ꢀ20 ꢂC. To this mixture was added an aqueous
solution of sodium hydroxide (3 M, 300 lL,
0.897 mmol). The reaction was stirred at ꢀ20 ꢂC for
5 h and upon disappearance of starting material (TLC)
saturated ammonium chloride was added (5 mL). The
reaction was extracted with diethyl ether (3 · 25 mL)
and the organics were combined, dried over MgSO4, con-
centrated under reduced pressure and then purified by
column chromatography (90% hexanes/10% ethyl ace-
tate) to give 1.400 g (91%) of epoxy ketone 18 as a color-
less oil. 1H NMR (CDCl3): d 3.59 (m, 1H), 3.51–3.44 (m,
2H), 3.23 (d, J = 4 Hz, 1H), 2.49 (dd, J = 18.0, 4.8 Hz,
1H), 2.32–2.20 (m, 2H), 2.01 (dd, J = 18.4, 10.8 Hz,
1H), 1.83 (ddd, J = 14.4, 10.0, 0.8 Hz, 1H), 0.87 (s,
9H), 0.03 (s, 3H), 0.02 (s, 3H); 13C NMR (CDCl3): d
205.55, 65.95, 55.02, 54.68, 39.41, 30.34, 26.09, 25.81,
18.21, ꢀ5.55; IR (neat, cmꢀ1) 2930, 2857, 1715, 1471,
1252, 1106, 837; HRMS (CI) m/z (M+Na) calcd
279.138690 for C13H24O3SiNa+, found 279.138261.
Lactone ( )-14 (1.54 g, 13.46 mmol) was dissolved in
THF (4 mL, anhydrous) and then added (cannula) to
a flask containing a solution of LiAlH4 (0.470 g,
12.40 mmol, 1 M/ether) in THF (25 mL) at 0 ꢂC. The
reaction was stirred for 2 h and then quenched with
water (1 mL), 15% NaOH (1 mL) and the more water
(2 mL) and let stir vigorously for 1.5 h at room temper-
ature. The mixture was dried over MgSO4, concentrated
under reduced pressure, and then purified by silica gel
chromatography (60% hexanes/40% ethyl acetate) to
give 1.557 g (98%) of diol 15 as an oil. 1H NMR
(CDCl3): d 5.81–5.76 (m, 1H), 5.73–5.69 (m, 1H), 4.32
(m, 1H), 3.58 (d, J = 6 Hz, 2H), 2.16–2.09 (m, 2H),
1.98–1.88 (m, 1H), 1.84–1.76 (m, 1H), 1.35–1.26 (m,
1H); identical in all respects to literature reports.16,17
4.4. Diol monosilyl ether ( )-16
Diol ( )-15 (1.45 g, 11.32 mmol) was dissolved in DMF
(15 mL, anhydrous) and then imidazole (1.542 g,
22.65 mmol) was added. The reaction was cooled to
0 ꢂC and then tert-butyldimethylsilyl chloride (1.70 g,
11.32 mmol) was added. The reaction was stirred at
0 ꢂC for 1 h and then slowly warmed to room tempera-
ture and stirred for another 5 h. The reaction was
4.7. Enoate esters ( )-19-E and ( )-19-Z
To a 25 mL flask was added NaH (0.103 g, 4.08 mmol)
and THF (5 mL, anhydrous). The slurry was cooled to

3972
M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
0 ꢂC and then triethyl phosphonoacetate (0.809 lL,
4.08 mmol) was added via syringe. This was then
warmed to room temperature and stirred for 30 min.
The reaction was then cooled back to 0 ꢂC and then a
3 mL THF solution of epoxy ketone ( )-18 (0.523 g,
2.04 mmol) was added via cannula. The reaction was
stirred at 0 ꢂC until complete disappearance of starting
material (TLC, ꢂ3 h). The reaction was quenched with
satd NH4Cl and extracted with diethyl ether
(3 · 25 mL). The organics were combined, dried over
MgSO4, concentrated under reduced pressure, and then
purified by column chromatography (90% hexanes/10%
ethyl acetate) to give 0.248 g (37%) of enoate ester ( )-
19-E and 0.388 g (58%) of enoate ester ( )-19-Z (95%
steps) as a colorless oil. 1H NMR (CDCl3): 5.40 (t,
J = 6.8 Hz, 1H), 4.58 (d, J = 7.2 Hz, 2H), 4.10 (m,
1H), 3.50–3.43 (m, 2H), 2.55 (dd, J = 13.2, 3.6 Hz,
1H), 2.35 (d, J = 13.6 Hz, 1H), 2.22 (dd, J = 13.2,
4.0 Hz, 1H), 1.96–1.86 (m, 1H), 1.81–1.68 (m, 4H),
1.48 (ddd, J = 13.6, 10.8, 2.8 Hz, 1H), 1.17 (s, 9H),
0.88 (s, 9H), 0.03 (s, 6H); 13C NMR (CDCl3): d
178.50, 140.52, 120.32, 67.39, 67.14, 60.37, 43.88,
38.68, 35.87, 35.74, 31.32, 27.16, 25.87, 18.26, ꢀ5.41;
IR (neat, cmꢀ1) 3502, 2929, 1724, 1459, 1280, 1155;
HRMS (CI) m/z (M+Na) calcd 393.243155 for
C20H38O4SiNa+, found 393.24329.
4.9. Allylic alcohol ( )-21
1
total) as colorless oils. Compound ( )-19-E: H NMR
(CDCl3): d 6.03 (m, 1H), 4.16 (q, J = 7.2 Hz, 2H),
3.52–3.46 (m, 2H), 3.42 (dd, J = 9.6, 5.6 Hz, 1H), 3.36
(d, J = 4 Hz, 1H), 3.10 (dd, J = 18, 4 Hz, 1H), 2.26–
2.13 (m, 2H), 1.95–1.84 (m, 1H), 1.64 (ddd, J = 14.4,
12, 1.2 Hz, 1H), 1.27 (t, J = 7.2 Hz, 3H), 0.87 (s, 9H),
0.03 (s, 3H); 13C NMR (CDCl3): d 165.46, 153.31,
121.98, 66.74, 59.89, 55.04, 54.23, 30.18, 28.60, 26.80,
25.87, 18.26, 14.24, ꢀ5.46; IR (neat, cmꢀ1) 2930, 2857,
1717, 1646, 1471, 1384, 1255, 1106, 837; HRMS (CI)
m/z (M+Na) calcd 349.180555 for C17H30O4SiNa+,
found 349.182443. Compound ( )-19-Z: 1H NMR
(CDCl3): d 5.96 (m, 1H), 4.62 (d, J = 3.6 Hz, 1H), 4.18
(dq, J = 7.2, 1.6 Hz, 2H), 3.45–3.36 (m, 3H), 2.37 (dd,
J = 14.8, 3.2 Hz, 1H), 2.18 (dd, J = 14.8, 4.8 Hz, 1H),
1.97 (ddd, J = 15.2, 10.0, 2.0 Hz, 1H), 1.89–1.81 (m,
1H), 1.68–1.61 (m, 1H), 1.28 (t, J = 7.2 Hz, 3H), 0.87
(s, 9H), 0.01 (s, 3H); 13C NMR (CDCl3): d 165.69,
152.75, 122.77, 66.31, 60.08, 54.38, 49.54, 33.77, 31.82,
To a 10 mL flask was added pivaloate ester ( )-20
(0.068 g, 0.183 mmol) and DMF (3.0 mL). To this mix-
ture was added imidazole (0.050 g, 0.734 mmol) and
tert-butyldimethylsilyl chloride (0.083 g, 0.550 mmol)
and then the reaction was heated to 60 ꢂC for 3 h. After
full consumption of the starting material (TLC), the
reaction was poured into ice water and then extracted
with diethyl ether (3 · 25 mL). The organics were com-
bined, dried over MgSO4, and concentrated under re-
duced pressure to give the crude bis-silyl ether
product, which was carried forward without further
purification.
In a 25 mL flame-dried flask was placed lithium alumi-
num hydride (0.004 g, 0.092 mmol) and THF (2 mL)
that was cooled to 0 ꢂC. In a separate 10 mL pear
shaped flask was placed the bis-silyl ether (0.183 mmol)
and THF (1.5 mL), which was cannulated over into the
reaction flask. The reaction was then stirred at 0 ꢂC for
1.5 h and then quenched by adding Na2SO4Æ10H2O.
This was stirred together for 30 min and then the salts
were filtered off and washed with diethyl ether. The
organics were concentrated under reduced pressure
and then purified by silica gel chromatography (80%
hexanes/20% ethyl acetate) to give 0.061 g of allylic alco-
hol 21 (83%, two steps) as a colorless oil. 1H NMR
(CDCl3): d 5.49 (t, J = 7.6 Hz, 1H), 4.16 (dd, J = 12.0,
7.6 Hz, 1H), 4.03 (dd, J = 12.4, 7.2 Hz, 1H), 3.94 (m,
1H), 3.50–3.42 (m, 2H), 2.33–2.24 (m, 2H), 2.11–1.98
(m, 3H), 1.68–1.62 (m, 1H), 1.52–1.46 (m, 1H), 0.89 (s,
9H), 0.85 (s, 9H), 0.04 (d, J = 1.2 Hz, 6H), 0.02 (d,
J = 4.8 Hz, 6H);13C NMR (CDCl3): d 139.36, 124.46,
68.09, 66.26, 58.24, 45.32, 36.99, 36.11, 30.02, 25.86,
25.78, 18.24, 18.09, ꢀ4.73, ꢀ4.79, ꢀ5.26, ꢀ5.36; IR
(neat, cmꢀ1) 3354, 2928, 2856, 1471, 1254, 1115, 1073,
836; HRMS (CI) m/z (M+Na) calcd 423.272117 for
C21H44O3Si2Na+, found 423.27055.
26.69, 25.82, 18.21, 14.23, ꢀ5.50; IR (neat, cmꢀ1
)
2930, 2857, 1717, 1646, 1471, 1384, 1255, 1106, 837;
HRMS (CI) m/z (M+Na) calcd 349.180555 for
C17H30O4SiNa+, found 349.180774.
4.8. Pivaloate ester ( )-20
To a 25 mL flask was added enoate ester ( )-19-E
(0.085 g, 0.260 mmol) and hexanes (1.5 mL). This mix-
ture was cooled to 0 ꢂC and then DIBAL-H (1.04 mL,
1.04 mmol, 1.0 M in hexanes) was added slowly via syr-
inge. The reaction was stirred for an additional 1.5 h
and then quenched with a saturated Na+, K+ tartrate
solution (2 mL), and stirred at room temperature for an-
other 30 min. The reaction was extracted with ethyl ace-
tate (3 · 25 mL) and the organics were combined, dried
over MgSO4 and concentrated under reduced pressure
to give the crude diol (0.077 g), which was carried for-
ward without further purification.
The crude diol (0.260 mmol) was placed in a 25 mL flask
and dissolved in pyridine (2 mL, anhydrous). DMAP
(0.035 g, 0.286 mmol) was added and the reaction was
cooled to 0 ꢂC. To this mixture was added trimethylace-
tyl chloride (35 lL, 0.286 mmol) dropwise via syringe.
The reaction was then stirred at 0 ꢂC for 3 h and then
quenched with water (2 mL), extracted with diethyl
ether (3 · 25 mL), dried over MgSO4, and purified by
silica gel chromatography (80% hexanes/20% ethyl ace-
tate) to give 0.068 g of pivaloate ester 20 (72%, two
4.10. Allylic phosphine oxide ( )-10
Allylic alcohol ( )-21 (0.113 g, 0.282 mmol) was placed
in a 25 mL flask with THF (1.5 mL) and cooled to
0 ꢂC. To this was added n-BuLi (194 lL, 0.310 mmol,
1.6 M in hexanes) dropwise via syringe. The reaction
was stirred for 10 min and then a THF solution
(1.5 mL) of p-toluenesulfonyl chloride (0.059 g,
0.310 mmol) was added via cannula and stirred for 3 h
at 0 ꢂC. In a separate flask was added diphenyl phos-

M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
3973
phine (245 lL, 1.409 mmol) and THF (3 mL). To this
was slowly added n-BuLi (0.881 mL, 1.409 mmol) via
syringe and then stirred for 10 min. This lithium diphe-
nyl phosphine solution was then cannulated into the
previous reaction flask until a dark orange color per-
sisted for 2 min. This was stirred at 0 ꢂC for another
1 h, after which water was added (2 mL) and the organ-
ics were brought to dryness under reduced pressure. To
this oily water mixture was added methylene chloride
(3 mL) and hydrogen peroxide (2 mL, 30%) and stirred
vigorously overnight. The reaction was then extracted
with methylene chloride (3 · 25 mL), the organics were
concentrated under reduced pressure and then purified
by silica gel chromatography (60% hexanes/40% ethyl
acetate) to give 0.117 g of allylic phosphine oxide 10
(71%) as a colorless oil. 1H NMR (CDCl3): d 7.75–
7.69 (m, 4H), 7.52–7.40 (m, 6H), 5.22 (m, 1H), 3.95
(m, 1H), 3.37–3.30 (m, 2H), 3.20–3.05 (m, 2H), 2.26–
2.21 (m, 1H), 2.16–2.04 (m, 2H), 1.90–1.80 (m, 2H),
1.63–1.59 (m, 1H), 1.49–1.44 (m, 1H), 1.37–1.30 (m,
1H), 0.87 (s, 9H), 0.81 (s, 9H), 0.01 (s, 6H), ꢀ0.03 (d,
J = 5.2 Hz, 6H); 13C NMR (CDCl3): d 140.16, 140.04,
131.63, 131.60, 131.13, 131.04, 130.95, 128.50, 128.46,
128.39, 128.34, 112.25, 112.17, 67.79, 67.10, 44.70,
36.72, 35.22, 30.90, 30.65, 29.94, 25.87, 25.79, 18.20,
25.91, 18.32, ꢀ5.41; IR (neat, cmꢀ1) 3504, 2928, 1725,
1459, 1283, 1155; HRMS (CI) m/z (M+Na) calcd
393.243155 for C20H38O4SiNa+, found 393.24347.
4.12. Allylic alcohol ( )-23
To a 10 mL flask was added pivaloate ester ( )-22
(0.085 g, 0.229 mmol) and DMF (2.0 mL). To this mix-
ture was added imidazole (0.047 g, 0.689 mmol) and
tert-butyldimethylsilyl chloride (0.070 g, 0.459 mmol)
and then the reaction was heated to 60 ꢂC for 3 h. After
full consumption of the starting material (TLC), the
reaction was poured into ice water and then extracted
with diethyl ether (3 · 25 mL). The organics were com-
bined, dried over MgSO4, and concentrated under re-
duced pressure to give the crude bis-silyl ether
(0.115 g), which was carried forward without further
purification.
In a 25 mL flame-dried flask was placed lithium alumi-
num hydride (0.004 g, 0.115 mmol) and THF (2 mL)
that was cooled to 0 ꢂC. In a separate 10 mL pear
shaped flask was placed the bis-silyl ether (0.229 mmol)
and THF (1.5 mL), which was cannulated over into the
reaction flask. The reaction was then stirred at 0 ꢂC for
1.5 h and then quenched by adding Na2SO4Æ10H2O.
This was stirred together for 30 min and then the salts
were filtered off and washed with diethyl ether. The
organics were concentrated under reduced pressure
and then purified by silica gel chromatography (80%
hexanes/20% ethyl acetate) to give 0.067 g of allylic alco-
hol 23 (81%, two steps) as a colorless oil. 1H NMR
(CDCl3): d 5.66 (t, J = 7.2 Hz, 1H), 4.19 (m, 1H), 4.12
(dd, J = 11.6, 7.2 Hz, 1H), 3.96 (dd, J = 11.6, 7.2 Hz,
1H), 3.47–3.39 (m, 2H), 2.61–2.56 (m, 1H), 2.25–2.21
(m, 1H), 2.05–1.94 (m, 2H), 1.82–1.73 (m, 2H), 1.35
(dt, J = 10.8, 2.8 Hz, 1H), 0.88 (s, 9H), 0.86 (s, 9H),
0.05 (d, J = 4.0 Hz, 6H), 0.03 (s, 6H); 13C NMR
(CDCl3): d 140.69, 123.82, 68.14, 67.59, 58.02, 39.33,
36.72, 36.67, 36.60, 25.90, 25.85, 18.29, 18.13, ꢀ4.75,
ꢀ4.83, ꢀ5.37, ꢀ5.40; IR (neat, cmꢀ1) 3358, 2928,
2856, 1471, 1360, 1253, 1116, 1073, 839; HRMS (CI)
m/z (M+Na) calcd 423.272117 for C21H44O3Si2Na+,
found 423.27122.
18.08, ꢀ4.79, ꢀ4.87, ꢀ5.35, ꢀ5.38; IR (neat, cmꢀ1
)
2926, 2854, 1470, 1252, 1117, 836; HRMS (CI) m/z
(M+Na) calcd 607.316304 for C33H53O3PSi2 Na+, found
607.322159.
4.11. Pivaloate ( )-22
To a 25 mL flask was added enoate ester ( )-19-Z
(0.075 g, 0.229 mmol) and hexanes (1.5 mL). This mix-
ture was cooled to 0 ꢂC and then DIBAL-H
(0.918 mL, 0.918 mmol, 1.0 M in hexanes) was added
slowly via syringe. The reaction was stirred for an addi-
tional 1.5 h and then quenched with a saturated Na+,
K+ tartrate solution (2 mL), and stirred at room temper-
ature for another 30 min. The reaction was extracted
with ethyl acetate (3 · 25 mL) and the organics were
combined, dried over MgSO4 and concentrated under
reduced pressure to give the crude diol (0.067 g), which
was carried forward without further purification.
The crude diol (0.229 mmol) was placed in a 25 mL flask
and dissolved in pyridine (2 mL, anhydrous). DMAP
(0.031 g, 0.253 mmol) was added, and the reaction was
cooled to 0 ꢂC. To this mixture was added trimethylace-
tyl chloride (31 lL, 0.253 mmol) dropwise via syringe.
The reaction was then stirred at 0 ꢂC for 3 h and then
quenched with water (2 mL), extracted with diethyl
ether (3 · 25 mL), dried over MgSO4, and purified by
silica gel chromatography (80% hexanes/20% ethyl
acetate) to give 0.076 g of pivaloate ester 22 (90%, two
4.13. Allylic phosphine oxide ( )-11
Allylic alcohol ( )-23 (0.180 g, 0.449 mmol) was placed
in a 25 mL flask with THF (3 mL) and cooled to 0 ꢂC.
To this was added n-BuLi (309 lL, 0.494 mmol, 1.6 M
in hexanes) dropwise via syringe. The reaction was stir-
red for 10 min and then a THF solution (2 mL) of p-tolu-
enesulfonyl chloride (0.094 g, 0.494 mmol) was added
via cannula and stirred for 3 h at 0 ꢂC. In a separate
flask was added diphenyl phosphine (391 lL, 2.24
mmol) and THF (4 mL). To this was slowly added n-
BuLi (1.40 mL, 2.24 mmol) via syringe and then stirred
for 10 min. This lithium diphenyl phosphine solution
was then cannulated into the previous reaction flask un-
til a dark orange color persisted for 2 min. This was stir-
red at 0 ꢂC for another 1 h, after which water was added
(2 mL) and the organics were brought to dryness under
reduced pressure. To this oily water mixture was added
1
steps) as a colorless oil. H NMR (CDCl3): d 5.46 (t,
J = 6.8 Hz, 1H), 4.77 (dd, J = 12.0, 8.8 Hz, 1H), 4.37
(dd, J = 12.4, 6.0 Hz, 1H), 4.16 (m, 1H), 3.44 (ddd,
J = 10.0, 9.2, 5.2 Hz, 2H), 2.80–2.76 (m, 1H), 2.28–
2.24 (m, 1H), 2.12–1.78 (m, 5H), 1.39 (dt, J = 13.6,
2.4 Hz, 1H), 1.17 (s, 9H), 0.87 (s, 9H), 0.02 (s, 6H);
13C NMR (CDCl3): d 179.07, 141.70, 119.83, 67.42,
66.96, 60.63, 39.22, 38.76, 36.28, 36.03, 35.59, 27.10,

3974
M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
methylene chloride (5 mL) and hydrogen peroxide
(3 mL, 30%) and stirred vigorously overnight. The reac-
tion was then extracted with methylene chloride
(3 · 25 mL), the organics were concentrated under re-
duced pressure and then purified by silica gel chroma-
tography (60% hexanes/40% ethyl acetate) to give
0.184 g of allylic phosphine oxide 11 (71%) as a colorless
36.02, 35.93, 31.15, 29.69, 29.35, 29.19, 28.93, 27.65,
23.52, 22.27, 20.77, 18.79, 12.00; IR (neat, cmꢀ1) 3358,
2930, 2871, 1465, 1374, 1214; UV (EtOH) kmax 251 (e
22,738); HRMS m/z (M+) calcd 441.333913 for
C27H46O3Na+, found 441.33351. Compound (+)-5b
25
(1b,3a): (Ret. time = 24.6 min); ½aꢁ +24.1 (c 0.06,
D
CHCl3); 1H NMR (400 MHz, CDCl3): d 6.21 (d,
J = 11.2 Hz, 1H), 5.85 (d, J = 11.2 Hz, 1H), 4.10 (m,
1H), 3.56 (m, 2H), 2.82–2.72 (m, 2H), 2.43 (d,
J = 13.6 Hz, 1H), 2.27 (dd, J = 14.0, 4.4 Hz, 1H), 2.01–
1.74 (m, 7H), 1.69–1.25 (m, 21H), 1.21 (s, 6H), 0.94
(d, J = 6.4 Hz, 3H), 0.56 (s, 3H); 13C NMR (CDCl3): d
142.28, 133.02, 122.28, 115.28, 71.10, 67.55, 67.48,
56.48, 56.26, 45.72, 44.71, 44.38, 40.47, 36.36, 36.09,
35.78, 30.91, 29.69, 29.35, 29.19, 28.80, 27.64, 23.41,
22.31, 20.78, 18.79, 12.13; IR (neat, cmꢀ1) 3357, 2926,
2871, 1466, 1377, 1214; UV (EtOH) kmax 251 (e
31,195); HRMS m/z (M+) calcd 441.333913 for
C27H46O3Na+, found 441.334625.
1
oil. H NMR (CDCl3) 7.75–7.66 (m, 4H), 7.54–7.41 (m,
6H), 5.27 (m, 1H), 3.94 (m, 1H), 3.37–3.29 (m, 2H),
3.23–3.13 (m, 1H), 3.05–2.96 (m, 1H), 2.19–2.12 (m,
2H), 2.01–1.86 (m, 2H), 1.78–1.74 (m, 2H), 1.64–1.59
(m, 1H), 1.35 (dt, J = 10.0, 2.8 Hz, 1H), 0.85 (s, 9H),
0.84 (s, 9H), 0.01 (d, J = 3.2 Hz, 6H), ꢀ0.01 (s, 6H);
13C NMR (CDCl3): d 140.87, 140.75, 131.70, 131.64,
131.16, 131.08, 131.03, 130.94, 128.51, 128.40, 112.02,
111.94, 67.60, 67.05, 38.93, 36.96, 36.66, 36.05, 30.31,
29.62, 25.88, 25.74, 18.24, 18.00, ꢀ4.71, ꢀ4.90, ꢀ5.36,
ꢀ5.40; IR (neat, cmꢀ1) 2928, 2855, 1470, 1252, 1253,
1118; HRMS (CI) m/z (M+Na) calcd 607.316304 for
C33H53O3PSi2Na+, found 607.31423.
4.15. Analogs (+)-6a and (+)-6b
4.14. Analogs (+)-5a and (+)-5b
Prior to reaction, phosphine oxide ( )-11 and C,D-ring
ketone (+)-9 were azeotrophically dried with benzene
and left under vacuum for 48 h. A solution of n-BuLi
in hexanes (80 lL, 0.128 mmol, 1.6 M) was added drop-
wise to a cold (ꢀ78 ꢂC) solution of phosphine oxide ( )-
11 (0.075 g, 0.128 mmol) in THF (1.0 mL) under dry ar-
gon. The resulting deep red solution was stirred for
40 min, at which time a cold (ꢀ78 ꢂC) solution of
C,D-ring ketone (+)-9 (0.029 g, 0.081 mmol) in THF
(1.5 mL) was added dropwise via cannula. The resulting
solution was stirred at ꢀ78 ꢂC in the dark for approxi-
mately 2 h, after which the dark red color had faded
to a light orange color. The reaction mixture was
quenched with pH 7.0 phosphate buffer (1 mL), warmed
to rt, extracted with Et2O (3 · 20 mL), and washed with
brine. The organics were dried over MgSO4, filtered,
concentrated, and purified by silica gel column chroma-
tography (95% hexanes/5% ethyl acetate) to afford the
coupled products as a clear oil (0.047 g, 80%). This oil
was immediately dissolved in THF (1.0 mL) and treated
with triethylamine (54 lL, 0.392 mmol) followed by
TBAF (392 lL, 0.392 mmol, 1.0 M in THF) and stirred
in the dark for ꢂ16 h. The reaction mixture was
quenched with H2O (1 mL), extracted with EtOAc
(3 · 25 mL), dried over MgSO4, filtered, concentrated,
and purified by silica gel column chromatography
(80% ethyl acetate/20% hexanes) to afford analogs 6a/b
(0.025 g, 93%) as a 1.5/1 mixture of diastereomers. This
diastereomeric mixture was separated by HPLC (Chiral-
cel-OD, 10% i-PrOH/hexanes, 2.5 mL/min) giving enan-
tiomerically pure, vitamin-D3 analogs 6a and 6b.
Prior to reaction, phosphine oxide ( )-10 and CD-ring
ketone (+)-9 were azeotrophically dried with benzene
and left under vacuum for 48 h. A solution of n-BuLi
in hexanes (81 lL, 0.121 mmol, 1.5 M) was added drop-
wise to a cold (ꢀ78 ꢂC) solution of phosphine oxide ( )-
10 (0.071 g, 0.121 mmol) in THF (1.0 mL) under dry
argon. The resulting deep red solution was stirred for
40 min, at which time a cold (ꢀ78 ꢂC) solution of
C,D-ring ketone (+)-9 (0.025 g, 0.072 mmol) in THF
(1.5 mL) was added dropwise via cannula. The resulting
solution was stirred at ꢀ78 ꢂC in the dark for approxi-
mately 3 h, after which the dark red color had faded
to a light orange color. The reaction mixture was
quenched with pH 7.0 phosphate buffer (1 mL), warmed
to rt, extracted with Et2O (3 · 20 mL), and washed with
brine. The organics were dried over MgSO4, filtered,
concentrated, and purified by silica gel column chroma-
tography (90% hexanes, 10% ethyl acetate) to afford the
coupled products as a clear oil (0.031 g, 60%). This oil
was immediately dissolved in THF (1.0 mL) and treated
with triethylamine (30 lL, 0.215 mmol) followed by
TBAF (215 lL, 0.215 mmol, 1.0 M in THF) and stirred
in the dark for 16 h. The reaction mixture was quenched
with H2O (1 mL), extracted with EtOAc (3 · 25 mL),
dried over MgSO4, filtered, concentrated, and purified
by silica gel column chromatography (80% ethyl acetate,
20% hexanes) to afford analogs 5a/b (0.017 g, 95%) as a
1/2 mixture of diastereomers. This diastereomeric mix-
ture was separated by HPLC (Chiralcel-OD, 9%
i-PrOH/hexanes, 2.5 mL/min) giving enantiomerically
pure, vitamin-D3 analogs (+)-5a and (+)-5b. Compound
Compound (+)-6a (1a,3b): (Ret. time = 21.3 min);
25
25
½aꢁ +82.3 (c 0.095, CHCl3); 1H NMR (400 MHz,
(+)-5a (1a,3b): (Ret. time = 34.8 min); ½aꢁ +40.3 (c
D
D
0.046, CHCl3); 1H NMR (400 MHz, CDCl3): d 6.22
(d, J = 11.2 Hz, 1H), 5.87 (d, J = 11.2 Hz, 1H), 4.13
(m, 1H), 3.56 (m, 2H), 2.81–2.78 (d, J = 12.0 Hz 2H),
2.43 (d, J = 14.4 Hz, 1H), 2.30–2.26 (m, 1H), 2.04–1.84
(m, 5H), 1.76–1.25 (m, 23H), 1.21 (s, 6H), 0.94 (d,
J = 6.8 Hz, 3H), 0.54 (s, 3H); 13C NMR (CDCl3): d
142.41, 133.20, 122.29, 115.25, 71.09, 67.64, 67.50,
56.47, 56.26, 45.74, 44.68, 44.38, 40.45, 36.36, 36.09,
CDCl3): d 6.35 (d, J = 11.6 Hz, 1H), 5.83 (d,
J = 10.8 Hz, 1H), 4.11 (m, 1H), 3.55–3.48 (m, 2H),
2.75 (ddd, J = 36.0, 12.4, 4.4 Hz, 2H), 2.33 (dd,
J = 35.6, 10.4 Hz, 2H), 2.01–1.82 (m, 6H), 1.72–1.22
(m, 22H), 1.21 (s, 6H), 0.93 (d, J = 6.4 Hz, 3H), 0.54
(s, 3H); 13C NMR (CDCl3): d 142.49, 133.12, 122.75,
115.24, 71.10, 67.44, 67.26, 56.49, 56.28, 45.74, 44.39,
40.48, 39.34, 36.71, 36.40, 36.37, 36.07, 29.36, 29.17,

M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
3975
28.86, 27.64, 23.45, 22.24, 20.78, 18.79, 12.07; IR (neat,
cmꢀ1) 3356, 2942, 2871, 1371, 1031, 756; UV (EtOH)
kmax 251 (e 27,340); HRMS m/z (M+) calcd 441.333913
142.38, 133.20, 122.27, 115.21, 78.26, 74.10, 68.79,
67.64, 67.51, 56.78, 55.76, 45.68, 44.71, 40.30, 36.05,
35.95, 35.65, 31.16, 31.12, 30.83, 28.99, 25.12, 24.22,
for C27H46O3Na+, found 441.333921. Compound (+)-
23.47, 22.47, 18.25, 12.65, 7.86, 7.82; IR (neat, cmꢀ1
)
25
6b (1b,3a): (Ret. time = 26.9 min); ½aꢁ +47.6 (c 0.35,
3370, 2940, 2874, 1454, 1371, 1214, 1102, 949, 755;
UV (EtOH) kmax 251 (e 38,978); HRMS m/z (M+) calcd
485.360128 for C29H50O4Na+, found 485.36118. Com-
D
MeOH); 1H NMR (400 MHz, CDCl3): d 6.37 (d,
J = 11.2 Hz, 1H), 5.83 (d, J = 11.2 Hz, 1H), 4.16 (m,
1H), 3.58–3.48 (m, 2H), 2.83–2.76 (m, 2H), 2.38 (d,
J = 10.4 Hz, 1H), 2.20 (d, J = 13.6 Hz, 1H), 2.01–1.84
(m, 6H), 1.72–1.23 (m, 23H), 1.21 (s, 6H), 0.94 (d,
J = 6.4 Hz, 3H), 0.54 (s, 3H); 13C NMR (CD3OD): d
141.56, 135.42, 122.43, 117.22, 71.47, 68.05, 67.07,
57.96, 57.53, 46.72, 45.28, 41.92, 40.19, 37.76, 37.59,
37.49, 37.24, 37.13, 29.76, 29.25, 29.13, 28.79, 24.53,
23.31, 21.90, 19.38, 12.39; IR (neat, cmꢀ1) 3309, 2921,
2855, 1450, 1377, 1141; UV (EtOH) kmax 251 (e
33,152); HRMS m/z (M+) calcd 441.333913 for
C27H46O3Na+, found 441.333592.
pound (ꢀ)-7b (1b,3a): (Ret. time = 34.1 min);
25
½aꢁ ꢀ36.7 (c 0.28, MeOH); 1H NMR (400 MHz,
D
CDCl3):
d 6.21 (d, J = 11.2 Hz, 1H), 5.83 (d,
J = 11.2 Hz, 1H), 4.10 (m, 1H), 3.56–3.49 (m, 2H),
3.30–3.19 (m, 2H), 2.82–2.72 (m, 2H), 2.43 (d,
J = 14.0 Hz, 1H), 2.27 (d, J = 14.0 Hz, 1H), 2.15 (d,
J = 12.8 Hz, 1H), 2.03–1.17 (m, 23H), 1.09 (d,
J = 6.4 Hz, 3H), 0.91–0.78 (m, 6H), 0.57 (s, 3H); 13C
NMR (CDCl3): d 142.22, 133.05, 122.26, 115.27,
78.26, 74.13, 68.80, 67.56, 67.44, 56.81, 55.76, 45.65,
44.75, 40.33, 36.14, 35.80, 35.66, 31.13, 30.94, 30.83,
28.86, 25.10, 24.23, 23.37, 22.51, 18.26, 12.79, 7.85,
7.83; IR (neat, cmꢀ1) 3371, 2940, 2874, 1454, 1371,
1215, 1103, 754; UV (EtOH) kmax 251 (e 36,203); HRMS
m/z (M+) calcd 485.360128 for C29H50O4Na+, found
485.35879.
4.16. Analogs (ꢀ)-7a and (ꢀ)-7b
Prior to reaction, phosphine oxide ( )-10 and C,D-ring
ketone (ꢀ)-24 were azeotrophically dried with benzene
and left under vacuum for 48 h. A solution of n-BuLi
in hexanes (44 lL, 0.071 mmol, 1.6 M) was added
dropwise to a cold (ꢀ78 ꢂC) solution of phosphine
oxide ( )-10 (0.041 g, 0.071 mmol) in THF (1.0 mL)
under dry argon. The resulting deep red solution was
stirred for 40 min, at which time a cold (ꢀ78 ꢂC) solu-
tion of C,D-ring ketone (ꢀ)-24 (0.014 g, 0.035 mmol) in
THF (1.5 mL) was added dropwise via cannula. The
resulting solution was stirred at ꢀ78 ꢂC in the dark
for approximately 3 h, after which the dark red color
had faded to a light orange color. The reaction mixture
was quenched with pH 7.0 phosphate buffer (1 mL),
warmed to rt, extracted with Et2O (3 · 20 mL), and
washed with brine. The organics were dried over
MgSO4, filtered, concentrated, and purified by silica
gel column chromatography (90% hexanes/10% ethyl
acetate) to afford the coupled products as a clear oil
(0.015 g, 55%).
4.17. Analog (+)-8
Prior to reaction, phosphine oxide (+)-11 and C,D-ring
ketone (ꢀ)-24 were azeotrophically dried with benzene
and left under vacuum for 48 h. A solution of n-BuLi
in hexanes (49 lL, 0.078 mmol, 1.6 M) was added drop-
wise to a cold (ꢀ78 ꢂC) solution of phosphine oxide (+)-
11 (0.046 g, 0.078 mmol) in THF (1.0 mL) under dry ar-
gon. The resulting deep red solution was stirred for
40 min, at which time a cold (ꢀ78 ꢂC) solution of
C,D-ring ketone (ꢀ)-24 (0.021 g, 0.053 mmol) in THF
(1.5 mL) was added dropwise via cannula. The resulting
solution was stirred at ꢀ78 ꢂC in the dark for approxi-
mately 1 h, after which the dark red color had faded
to a light orange color. The reaction mixture was
quenched with pH 7.0 phosphate buffer (1 mL), warmed
to rt, extracted with Et2O (3 · 20 mL), and washed with
brine. The organics were dried over MgSO4, filtered,
concentrated, and purified by silica gel column chroma-
tography (95% hexanes/5% ethyl acetate) to afford the
coupled product as a clear oil (0.009 g, 23%). This oil
was immediately dissolved in THF (1.0 mL) and treated
with triethylamine (25 lL, 0.177 mmol) followed by
TBAF (118 lL, 0.118 mmol, 1.0 M in THF) and stirred
in the dark for ꢂ16 h. The reaction mixture was
quenched with H2O (1 mL), extracted with EtOAc
(3 · 25 mL), dried over MgSO4, filtered, concentrated,
and purified by silica gel column chromatography
(80% ethyl acetate/20% hexanes) to afford analog
(+)-8 (0.005 g) in 93% yield. This analog was purified
by HPLC (Chiralcel-OD, 9% EtOH/hexanes, 2.5 mL/
The coupled product (0.034 g, 0.045 mmol) was dis-
solved in THF (1.0 mL) and treated with triethylamine
(94 lL, 0.677 mmol) followed by TBAF (452 lL,
0.452 mmol, 1.0 M in THF) and stirred in the dark for
ꢂ16 h. The reaction mixture was quenched with H2O
(1 mL), extracted with EtOAc (3 · 25 mL), dried over
MgSO4, filtered, concentrated, and purified by silica
gel column chromatography (80% ethyl acetate/20%
hexanes) to afford analogs 7a/b (0.019 g, 90%) as a 1/2
mixture of diastereomers. This diastereomeric mixture
was separated by HPLC (Chiralcel-OD, 5% EtOH/hex-
anes, 2.5 mL/min) giving enantiomerically pure, vita-
min-D3 analogs (ꢀ)-7a and (ꢀ)-7b. Compound (ꢀ)-7a
25
(1a,3b): (Ret. time = 41.2 min); ½aꢁ ꢀ8.02 (c 0.32,
min). Compound (+)-8 (1a,3b): (Ret. time = 31.0 min);
D
25
D
MeOH); 1H NMR (400 MHz, CDCl3): d 6.21 (d,
J = 11.2 Hz, 1H), 5.85 (d, J = 11.2 Hz, 1H), 4.12 (m,
1H), 3.61–3.51 (m, 2H), 3.30–3.18 (m, 2H), 2.83–2.76
(m, 2H), 2.43 (d, J = 13.6 Hz, 1H), 2.27 (d,
J = 14.0 Hz, 1H), 2.15 (d, J = 12.4 Hz, 1H), 2.07–1.18
(m, 23H), 1.08 (d, J = 6.0 Hz, 3H), 0.85 (dt, J = 7.2,
1.6 Hz, 6H), 0.55 (s, 3H); 13C NMR (CDCl3): d
½aꢁ +8.5 (c 0.25, MeOH); 1H NMR (400 MHz, CDCl3):
d 6.35 (d, J = 11.2 Hz, 1H), 5.81 (d, J = 11.2 Hz, 1H),
4.12 (m, 1H), 3.58–3.48 (m, 2H), 3.30–3.19 (m, 2H),
2.75 (ddd, J = 38.0, 12.4, 4.4 Hz, 2H), 2.37 (d,
J = 9.6 Hz, 1H), 2.27 (d, J = 10.0 Hz, 1H), 2.15 (d,
J = 12.4 Hz, 1H), 2.03–1.18 (m, 23H), 1.09 (d,
J = 6.0 Hz, 3H), 0.85 (dt, J = 7.6, 2.0 Hz, 6H), 0.55 (s,

3976
M. A. Hatcher et al. / Bioorg. Med. Chem. 13 (2005) 3964–3976
3H); 13C NMR (CDCl3): d 142.49, 133.09, 122.78,
115.19, 78.25, 74.08, 68.81, 67.48, 67.28, 56.80, 55.78,
45.68, 40.32, 39.36, 36.75, 36.41, 36.12, 35.67, 31.12,
30.82, 28.92, 25.11, 24.22, 23.41, 22.44, 18.25, 12.79,
7.87, 7.84; IR (neat, cmꢀ1) 3374, 2925, 2874, 1453,
1370, 1103; UV (EtOH) kmax 251 (e 26,975); HRMS
m/z (M+) calcd 485.360128 for C29H50O4Na+, found
485.36278.
13. Baggiolini, E. G.; Iacobelli, J. A.; Hennessy, B. M.;
Batcho, A. D.; Sereno, J. F.; Uskokovic, M. R. J. Org.
Chem. 1986, 51, 3098–3108.
14. Ceder, O.; Hansson, B. Acta Chem. Scand. 1970, 24, 2693.
15. Ireland, R. E.; Highsmith, T. K.; Gegnas, L. D.; Gleason,
J. L. J. Org. Chem. 1992, 57, 5071–5073.
16. Linde, R. G.; Egbertson, M.; Coleman, R. S.; Jones, A. B.;
Danishefsky, S. J. J. Org. Chem. 1990, 55, 2771–2776.
17. Martin, S. F.; Dappen, M. S.; Dupre, B.; Murphy, C. J.;
Colapret, J. A. J. Org. Chem. 1989, 54, 2209–2216.
18. Marshall, J. A.; Xie, S. P. J. Org. Chem. 1995, 60, 7230–
7237.
Acknowledgements
19. Hanazawa, T.; Wada, T.; Masuda, T.; Okamoto, S.; Sato,
F. Org. Lett. 2001, 3, 3975–3977.
20. Hanazawa, T.; Inamori, H.; Masuda, T.; Okamoto, S.;
Sato, F. Org. Lett. 2001, 3, 2205–2207.
We thank the NIH for financial support (CA 93547).
21. Bensel, N.; Marschall, H.; Weyerstahl, P. Tetrahedron
Lett. 1976, 2293–2296.
References and notes
22. Strekowski, L.; Visnick, M.; Battiste, M. A. Tetrahedron
Lett. 1984, 25, 5603–5606.
23. Oshima, M.; Yamazaki, H.; Shimizu, I.; Nisar, M.; Tsuji,
J. J. Am. Chem. Soc. 1989, 111, 6280–6287.
1. Holick, M. F. Vitamin D: Molecular Biology, Physiology,
and Clinical Applications; Humana: Totowa, NJ, 1999.
2. Feldman, D.; Glorieux, F.; Pike, J. W. Vitamin D;
Academic: San Diego, CA, 1997.
3. Bouillon, R.; Okamura, W. H.; Norman, A. W. Endocri-
nol. Rev. 1995, 16, 200–257.
4. Posner, G. H.; Kahraman, M. Eur. J. Org. Chem. 2003,
3889–3895.
5. Posner, G. H.; Nelson, T. D.; Guyton, K. Z.; Kensler, T.
W. J. Med. Chem. 1992, 35, 3280–3287.
24. Spencer, T. A.; Clark, D. S.; Johnson, G. A.; Erickson, S.
K.; Curtiss, L. K. Bioorg. Med. Chem. 1997, 5, 873–882.
25. Kutner, A.; Zhao, H.; Fitak, H.; Wilson, S. R. Bioorg.
Chem. 1995, 23, 22–32.
26. Posner, G. H.; Crawford, K. R.; Yang, H. W.; Kahraman,
M.; Jeon, H. B.; Li, H. B.; Lee, J. K.; Suh, B. C.; Hatcher,
M. A.; Labonte, T.; Usera, A.; Dolan, P. M.; Kensler, T.
W.; Peleg, S.; Jones, G.; Zhang, A. Q.; Korczak, L.; Saha,
U.; Chuang, S. S. J. Steroid Biochem. Mol. Biol. 2004, 89–
90, 5–12.
6. Posner, G. H.; Lee, J. K.; Wang, Q.; Peleg, S.; Burke, M.;
Brem, H.; Dolan, P.; Kensler, T. W. J. Med. Chem. 1998,
41, 3008–3014.
7. Posner, G. H.; Jeon, H. B.; Sarjeant, A.; Riccio, E. S.;
Doppalapudi, R. S.; Kapetanovic, I. M.; Saha, U.; Dolan,
P.; Kensler, T. W. Steroids 2004, 69, 757–762.
8. Kensler, T. W.; Dolan, P. M.; Gange, S. J.; Lee, J. K.;
Wang, Q.; Posner, G. H. Carcinogenesis 2000, 21, 1341–
1345.
9. Reddy, C. D.; Patti, R.; Guttapalli, A.; Yanamandra, N.;
Sutton, L. N.; Phillips, P. C.; Posner, G. H. J. Cell.
Biochem., submitted for publication.
27. Wilson, S. R.; Hong, Z.; Dewan, J. Bioorg. Med. Chem.
Lett. 1993, 3, 341–344.
28. Posner, G. H.; Li, Z. G.; White, M. C.; Vinader, V.;
Takeuchi, K.; Guggino, S. E.; Dolan, P.; Kensler, T. W. J.
Med. Chem. 1995, 38, 4529–4537.
29. Crystallographic data (excluding structure factors) for the
structures in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplemen-
tary publication no. CCDC 263794. Copies of the data can
be obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge, CB2 1EZ, UK (fax: +44 (0)1223
336033 or deposit@ccdc.cam.ac.uk).
10. Posner, G. H.; Lee, J. K.; White, M. C.; Hutchings, R. H.;
Dai, H. Y.; Kachinski, J. L.; Dolan, P.; Kensler, T. W. J.
Org. Chem. 1997, 62, 3299–3314.
11. Perlman, K. L.; Sicinski, R. R.; Schnoes, H. K.; Deluca,
H. F. Tetrahedron Lett. 1990, 31, 1823–1824.
12. Nagasawa, K.; Matsuda, N.; Noguchi, Y.; Yamanashi,
M.; Zako, Y.; Shimizu, I. J. Org. Chem. 1993, 58, 1483–
1490.
30. The (1a,3b) stereochemistry for analogs 2a, 3a, and 4a has
been revised from the previously reported (1b,3a)
stereochemistry.7
31. Posner, G. H.; Halford, B. A.; Peleg, S.; Dolan, P.;
Kensler, T. W. J. Med. Chem. 2002, 45, 1723–1730.