Total synthesis of the anticancer natural product OSW-1

Article abstract of DOI:10.1021/ja012119t

The highly potent anticancer natural saponin OSW-1 has been successfully synthesized from commercially available 5-androsten-3β-ol-17-one 79 in 10 operations with 28% overall yield. The key steps in the total synthesis included a highly regio- and stereoselective selenium dioxide-mediated allylic oxidation of 80 and a highly stereoselective 1,4-addition of α-alkoxy vinyl cuprates 68 to steroid 17(20)-en-16-one 12E to introduce the steroid side chain. This total synthesis demonstrated once again the versatile synthetic applications of α-halo vinyl ether chemistry developed in our laboratories.

Full text of DOI:10.1021/ja012119t

Published on Web 05/17/2002
Total Synthesis of the Anticancer Natural Product OSW-1
Wensheng Yu and Zhendong Jin*
Contribution from the DiVision of Medicinal and Natural Products Chemistry,
College of Pharmacy, The UniVersity of Iowa, Iowa City, Iowa 52242
Received September 5, 2001
Abstract: The highly potent anticancer natural saponin OSW-1 has been successfully synthesized from
commercially available 5-androsten-3â-ol-17-one 79 in 10 operations with 28% overall yield. The key steps
in the total synthesis included a highly regio- and stereoselective selenium dioxide-mediated allylic oxidation
of 80 and a highly stereoselective 1,4-addition of R-alkoxy vinyl cuprates 68 to steroid 17(20)-en-16-one
12E to introduce the steroid side chain. This total synthesis demonstrated once again the versatile synthetic
applications of R-halo vinyl ether chemistry developed in our laboratories.
Introduction
OSW-1 (1), a highly potent anticancer natural product, and
its four natural analogues (2-5) have been isolated from the
bulbs of Ornithogalum saundersiae, a perennial grown in
southern Africa where it is cultivated as a cut flower and garden
plant.1 These natural products are members of the cholestane
glycosides. Their absolute structures have been determined by
extensive application of spectroscopic methods.1 The structural
novelty of compounds 1-5 is characterized by the attachment
of a disaccharide to the C-16 position of the steroid aglycone,
whereas compounds 4 and 5 have another glycosyl sugar
associated with the C-3 alcohol position of the steroid (Figure
1).
Compounds 1-5 exhibited extremely potent cytostatic activ-
ity against human promyelocytic leukemia HL-60 cells, showing
IC50 values ranging between 0.1 and 0.3 nM. The activity of
OSW-1 (1) in this assay is much more potent than that of
clinically used anticancer agents such as etoposide, adriamycin,
and methotrexate.2 OSW-1 (1), the main constituent of the bulbs,
exhibited exceptionally potent cytostatic activities against vari-
ous human malignant tumor cells.2 Its cytostatic activities are
from 10- to 100-fold more potent than some well-known
anticancer agents in clinical use, such as mitomycin C, adria-
mycin, cisplatin, camptothecin, and even taxol, but it has
significantly lower toxicity (IC50 1500 nM) to normal human
pulmonary cells.2 The surprising similarity of the cytotoxicity
profile of OSW-1 to that of cephalostatins,3 one of the most
active anticancer agents tested by NIH, with correlation coef-
ficient of 0.60-0.83, suggests they might have the same
Figure 1.
mechanism of action.4 It has been speculated by Fuchs that the
C22-oxonium ions might be the active intermediate for the
potent anticancer activity of OSW-1 (1) and cephalostatins.5
This suggests that OSW-1 (1) might represent a new class of
anticancer agents with a new mechanism of action. All of these
factors make OSW-1 (1) a very attractive synthetic target.4,6
As part of our program studying the chemistry and biology of
anticancer natural products, we recently initiated a project
directed toward the total synthesis of OSW-1 (1). We report
herein our full account of our studies toward the total synthesis
of this highly promising anticancer natural product.
Results and Discussion
Retrosynthetic Analysis. The C-20 carbon of OSW-1 has
the “normal” 20S configuration. Molecular mechanics calcula-
tions (MM2) have shown that compound 1 is about 3.1 kcal/
mol more stable than its 20R epimer 6, whereas 7 is about 2.4
kcal/mol more stable than 8 (Figure 2).7 Therefore, we thought
that it was not necessary to control the stereochemistry at C-20
during the synthesis and anticipated that compound 6 would
eventually epimerize to the thermodynamically more stable 1
at the end of the synthesis.
Figure 3 outlines our retrosynthetic analysis of OSW-1 (1).
Disconnection at the glycoside bond reveals the protected
* To whom correspondence should be addressed. E-mail:
zhendong-jin@uiowa.edu.
(1) Kubo, S.; Mimaki, Y.; Terao M.; Sashida, Y.; Nikaido, T.; Ohmoto, T.
Phytochemitstry 1992, 31, 3969.
(2) Mimaki, Y.; Kuroda, M.; Kameyama, A.; Sashida, Y.; Hirano, T.; Oka,
K.; Maekawa, R.; Wada, T.; Sugita, K.; Beutler, J. A. Bioorg. Med. Chem.
Lett. 1997, 7, 633.
(3) (a) Pettit, G. R.; Inoue, M.; Kamano, Y.; Herald, D. L.; Arm, C.; Dufresne,
C.; Christie, N. D.; Schmidt, J. M.; Doubek, D. L.; Krupa, T. S. J. Am.
Chem. Soc. 1988, 110, 2006. (b) LaCour, T. G.; Guo, C.; Bhandaru, S.;
Boyd, M. R.; Fuchs, P. L. J. Am. Chem. Soc. 1998, 120, 692.
(4) Guo, C.; Fuchs, P. L. Tetrahedron Lett. 1998, 39, 1099.
(5) Guo, C.; LaCour, T. G.; Fuchs, P. L. Bioorg. Med. Chem. Lett. 1999, 9,
419.
(6) (a) Deng, S.; Yu, B.; Lou, Y.; Hui, Y. J. Org. Chem. 1999, 64, 202. (b)
Morzycki, J. W.; Gryszkiewicz, A.; Jastrzebska, I. Tetrahedron Lett. 2000,
41, 3751. (c) Yu, W.; Jin, Z. J. Am. Chem. Soc. 2001, 123, 3369.
(7) MM2 calculation was performed using CS Chem Ultra program on a
powerMac. The calculation results were consistent with experimental results,
see ref 10.
9
6576
J. AM. CHEM. SOC. 2002, 124, 6576-6583
10.1021/ja012119t CCC: $22.00 © 2002 American Chemical Society

Synthesis of the Anticancer Natural Product OSW-1
A R T I C L E S
the commercially available steroid 14. Further disconnection
at the glycoside bond of the disaccharide fragment 10 shows
two monosaccharide units 15 and 16 which could be derived
from L-arabinose and D-xylose, respectively.
Synthesis of the Disaccharide 10. The first monosaccharide
15 was prepared from tetraacetyl-L-arabinose 17 as illustrated
in Figure 4. Thioglycoside 18 was prepared according to the
standard methods8 followed by deacetylation to give compound
19 in excellent yield. Regioselective protection of the cis diol
of 19 followed by protection of the C-2 hydroxyl group gave
20 in 90% yield. Deprotection of the acetonide afforded diol
21. It is well-known that the equatorial C-3 hydroxyl group in
many sugars is more reactive than C-4 axial hydroxyl group.
To our surprise, high selectivity at C-4 hydroxyl group was
observed when 21 was treated with TESOTf and lutidine at low
temperature affording the desired product 15 in 90% yield.
Figure 2.
The second monosaccharide 16 was prepared from tetraacetyl-
D-xylose 22. The thio ortho ester 24 was prepared via the
glycoside bromide 23 according to the literature procedures
(Figure 5).8 Protecting-group manipulations followed by zinc
chloride-promoted intramolecular ring opening of the thio ortho
ester 26 gave thioglycoside 27 in excellent yield. After deacety-
aglycone 9 and the disaccharide 10 as the potential key
fragments for the construction of the target molecule. Compound
9 was envisioned to be formed via a triply convergent strategy
which would involve 1,4-addition of acyl anion equivalent 11
to enone 12 followed by in situ stereoselective oxidation of the
resulting enolate. Enone 12 was envisaged to be prepared from
Figure 3.
Figure 4.
9
J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002 6577

A R T I C L E S
Yu and Jin
Figure 5.
Figure 6.
Figure 7.
Dess-Martin oxidation11 of the allyl alcohol afforded 96% yield
of enone 12 as a mixture of Z- and E-stereoisomers with a ratio
of 2:1.12
lation, the p-methoxy benzoyl group was introduced at the C-2
position to afford 29, which was subsequently converted to 16
in 95% yield.9
1,4-Addition of an acyl anion synthon to enone 12 was the
key reaction to install the side chain of the aglycon in our
strategy. Our studies on the addition of various R-thioacetal
anions to enone 12 are summarized in Figure 8.
The reaction between 1,3-dithiane anion 33 and enone 12
gave exclusively 1,2-addition product 35 even in the presence
of HMPA and at room temperature (Figure 8).13 The softer anion
36 reacted with enone 12 in 1,2-fashion at -78 °C and then, in
the presence of HMPA, rearranged to the 1,4-addition product
Glycosylation of 15 with 16 in the presence of BF3‚Et2O
afforded the â-disaccharide 30 in 93% yield (Figure 6).
Disaccharide 30 was then converted to the trichloroacetimidate
10, which was then ready to couple with the protected steroid
aglycone.
The Attempted Synthesis of the Protected Steroid Agly-
cone. The commercially available 5-pregnen-16,17-epoxy-3â-
ol-20-one 14 was protected by a TBS group to give compound
31 (Figure 7). Reduction of the R,â-epoxy ketone 31 by
hydrazine hydrate gave the allyl alcohol 32 in 73% yield.10
(10) (a) Kessar, S. V.; Rampal, A. L. Tetrahedron, 1968, 24, 887. (b) Kessar,
S. V.; Gupta, Y. P.; Mahajan, R. K.; Rampal, A. L. Tetrahedron 1968, 24,
893.
(8) Nicolaou, K. C.; Trujillo, J. I.; Chibale, K. Tetrahedron 1997, 53, 8751.
(9) Nicolaou, K. C.; Ohshima, T.; Hosokawa, S.; Delft, F. L. van; Vourloumis,
D.; et al. J. Am. Chem. Soc. 1998, 120, 8674.
(11) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277.
(12) The stereochemistry of the double bond was determined by NOESY
spectrum.
9
6578 J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002

Synthesis of the Anticancer Natural Product OSW-1
A R T I C L E S
spectra. The desired products 47 and 48 which have âC-16
hydroxyl groups were obtained in a combined yield of 86%.
The metalation of both 47 and 48 proved to be quite difficult.
After screening a few strong bases with or without additives
such as HMPA or TMEDA, R-thioacetal anions 50 and 51 were
successfully generated from 47 and 48 by treatment with super
base (n-BuLi/t-BuOK) (Figure 10).15 The formation of these
anions was confirmed by deuterium incorporation after the
reaction was quenched with D2O at -78 °C. Unfortunately, the
attempt to quench them with electrophiles such as methyl iodide
or allyl bromide resulted in quick decomposition of the anions
50 and 51. Although it is still not clear exactly what happened,
we speculate that the addition of electrophiles might accelerate
the R-elimination of the highly bulky tertiary anions 50 and
51. This speculation is supported by the fact that the reaction
mixture smelled like thiophenol in the metalation step, and the
odor of the thiophenol intensified immediately after the addition
of an electrophile.
Figure 8.
The unexpected difficulty in the alkylation of R-thioacetal
anions 50 and 51, coupled with the difficulties in the 1,4-addition
of the sterically hindered tertiary R-thioacetal anions, led us to
modify our original approach.
The Attempted Synthesis of OSW-1. An R-alkoxy vinyl
anion, such as anion 56, is another kind of acyl anion equivalent,
which is more reactive and smaller compared to R-thioacetal
anion 42 (Figure 11). This suggests a new approach in which
R-alkoxy vinyl anion will be employed as the acyl anion
equivalent.
In this new approach, we needed to prepare the â-isobutyl
substituted R-methoxy vinyl cuprate 58 (Figure 12). However,
there was no literature procedure for the quantitative generation
of the requisite â-isobutyl substituted R-methoxy vinyl anion.
To solve this problem, we developed a new methodology for
the regio- and stereoselective synthesis of R-halo vinyl ether
that could serve as the precursor of the R-alkoxy vinyl anion.16
The acetylenic ether 59 was prepared according to a literature
procedure.17 The R-bromovinyl ether 60 and the required
38 upon warming up the reaction mixture. However, the yield
was quite low, and 20% of the starting material enone was
recovered due to the equilibrium. Anion 39 appeared to be the
best choice, as it gave 65% yield of compound 41. Unfortu-
nately, due to the steric hindrance of the tertiary anion 42, the
reaction was too slow and the yield was quite low. Therefore
we decided to introduce the side chain in two steps via the
addition of anion 39.
1,4-Addition of anion 39 to enone 12 afforded enolate
intermediate 44, which was easily oxidized by dibenzyl per-
oxydicarbonate 1314 at -78 °C to give compound 45 in 63%
yield (Figure 9). It appeared that the oxidation of the thioacetal
by dibenzyl peroxydicarbonate 13 was much slower than the
oxidation of the enolate at -78 °C, and no sulfoxide product
was isolated. Hydrolysis followed by stereoselective reduction
by LiAlH4 afforded three diastereoisomers 47, 48, and 49. The
stereochemistry of the C-21 methyl group and the C-16 hydroxyl
group were determined by analysis of the corresponding ROESY
Figure 9.
9
J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002 6579

A R T I C L E S
Yu and Jin
Figure 10.
Figure 11.
Figure 13.
Figure 14.
Figure 12.
R-methoxy vinyl cuprate 58 was prepared according to our
newly developed methodology.16
We deliberately used compound 12Z (the major isomer of
the enone mixture 12) to examine the proposed 1,4-addition
reaction (Figure 13). It was anticipated that the 12Z would be
less reactive than the 12E isomer. Although we had several
successful model studies on the 1,4-addition of cuprate 58 to
various simple R,â-unsaturated ketones,18 to our surprise, the
reaction between cuprate 58 and enone 12Z did not lead to any
desired product. Both low-order and high-order cuprates 61 and
58, respectively, were carefully examined,19 but no desired
Figure 15.
product was obtained even with TMSCl activation.20,21 Enone
12Z was recovered nearly quantitatively each time.
From the reaction mixture, a UV-active side product was
isolated and it was found to be compound 64, which had
obviously been formed via the Wu¨rtz coupling of the R-methoxy
(13) (a) Reich, H. J.; Sikorski, W. H. J. Org. Chem. 1999, 64, 14. (b) Brown,
C. A.; Yamaichi, A. Chem. Commun. 1979, 100.
(14) Gore, M. P.; Vederas, J. C. J. Org. Chem. 1986, 51, 3700.
(15) Schlosser, M.; Strunk, S. Tetrahedron Lett. 1984, 25, 741.
(16) Yu, W.; Jin, Z. J. Am. Chem. Soc. 2000, 122, 9840.
(17) Moyano, A.; Charbonnier, F.; Greene, A. E. J. Org. Chem. 1987, 52, 2,
2919.
(19) (a) Lipshutz, B. H.; Wilhelm, R. S.; Kozlowski, J. A. Tetrahedron 1984,
40, 5005 and references cited therein. (b) Lipshutz, B. H. Synthesis 1987,
87, 325 and references therein.
(20) Corey, E. J.; Boaz, N. W. Tetrahedron Lett. 1985, 26, 6019.
(21) Alexakis, A.; Berlan, J.; Besace, Y. Tetrahedron Lett. 1986, 27, 1047.
(18) Yu, W.; Jin, Z. Unpublished results.
9
6580 J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002

Synthesis of the Anticancer Natural Product OSW-1
A R T I C L E S
vinyl cuprate 68 was prepared. The size of the R-alkoxy group
was increased from methoxy group to cyclohexyloxy group
(Figure 15).
As expected, cuprate 68 underwent smooth 1,4-addition to
enone 12Z in the presence of TMSCl to afford the desired silyl
enol ether 69 in 92% yield (Figure 16). However, 3 equiv of
cuprate 68 was needed to drive the reaction to completion.
With the silyl enol ether 69 in hand, we needed to generate
the enolate 70 and then oxidize the enolate 70 in situ to introduce
the C-17 hydroxyl group (Figure 17). The literature procedure
using MeLi to cleave the silyl enol ether 69 was found to be
extremely slow.24 Some dry fluoride reagents were also
employed, but none of them gave any satisfactory results. To
solve this problem, a new methodology for the generation of
enolates from silyl enol ethers by using potassium ethoxide was
developed.25 Employing our new methodology, silyl enol ether
69 was cleaved in 5 min at 0 °C to give the potassium enolate
71 in quantitative yield.
Figure 16.
vinyl cuprate 58 (Figure 14). The isolation of compound 64
suggested that the Wu¨rtz coupling was much faster than the
desired 1,4-addition. Oxygen is normally considered to be the
reason for the Wu¨rtz-coupling side reaction of organocuprates.22
However, careful degassing of the reaction solvent and careful
removal of possible traces of oxygen in argon by installing a
Pyrogallol filter23 still failed to stop the Wu¨rtz coupling.
The two neighboring methoxy groups in the Wu¨rtz-coupling
product 64 are close to each other. We speculated that increasing
the size of these two alkoxy groups might suppress the formation
of the Wu¨rtz-coupling product. However, the alkoxy group
should not be too bulky, otherwise the 1,4-addition would also
be difficult. On the basis of the above analysis, R-cyclohexyloxy
Efforts to oxidize enolate 71 with Davis reagent,26 molecular
oxygen,27 or dibenzyl peroxydicarbonate14 were unsuccessful.
Figure 17.
9
J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002 6581

A R T I C L E S
Yu and Jin
Figure 18.
Figure 19.
Figure 20.
This problem is probably due to the presence of another labile
enol ether moiety on the steroid side chain which is also prone
to various oxidative reaction conditions. Thus, silyl enol ether
69 was converted to enol acetate 73, which enabled us to
regiospecifically convert the enol ether functionality at C-22 to
cycloketal 74. Either EtOK or t-BuOK28 was used to generate
the enolate from enol acetate 74, and the enolate was then
oxidized in situ by Davis reagent to give the R-hydroxyl ketone
75 in 78% yield. Stereospecific reduction of the C-16 ketone
by LiAlH4 at -78 °C afforded the trans diol 76 in 98% yield.
The stereospecificity of the LiAlH4 reduction was presumably
due to the directing effort of C-17 hydroxy group.29
various basic conditions (pyridine, DBU, phosphazene base P2-
t-Bu,32 etc.) and acidic conditions were investigated, the
epimerization of C-20 methyl group to the S-configuration was
not observed.
The reason for this stereochemistry problem at C-20 was
directly related to the enone 12, a mixture of stereoisomers in
favor of the undesired Z-isomer (Figure 7). Therefore, a new
approach was needed to synthesize enone 12E stereoselectively
with the correct stereochemistry at C-20.
Total Synthesis of OSW-1. A new approach for the
stereospecific synthesis of enone 12E was developed (Figure
18). Compound 80 with the requisite Z-configuration was
prepared according to a literature procedure from commercially
available 5-androsten-3â-ol-17-one 79.33 Selenium dioxide-
mediated allylic oxidation provided 32E with complete chemo-,
regio-, and stereoselectivity.34 Swern oxidation of 32E afforded
enone 12E in nearly quantitative yield.
Glycosylation of the diol 76 with the disaccharide 10 in the
presence of TMSOTf provided â-glycoside 77 in 88% yield.30
All the protecting groups, including two PMB, one TBS, one
TES, and one cycloketal, were removed by sequential treatment
31
with DDQ and Pd(CN)2Cl2 in a single operation to give 78
(C-20 epimer of OSW-1) in 81% yield.
With enone 12E in hand, TMSCl-activated 1,4-addition of
R-alkoxy vinyl cuprate 68 to enone 12E went smoothly to give
silyl enol ether intermediate 81, which was further converted
to enol acetate 82 without isolation of 81 (Figure 19). Compound
82 was then converted to compound 83 in excellent yield.
Generation of the enolate from 83 by potassium ethoxide or
t-BuOK35 followed by in situ stereoselective oxidation by Davis
reagent36 gave R-hydroxyl ketone 84 in 76% yield. Stereose-
To complete the total synthesis of OSW-1 (1), the stereo-
chemistry of the C-20 methyl group needed to be epimerized
to the requisite S-configuration. Unfortunately, our effort to
epimerize the C-20 methyl group was not successful. Although
(22) Blanchot-Courtois, V.; Hanna, I. Tetrahedron Lett. 1992, 33, 8087.
(23) Kim, S.; Sutton, S. C.; Guo, C.; LaCour, T. G.; Fuchs, P. L. J. Am. Chem.
Soc. 1999, 121, 2056.
(24) Stork, G.; Hudrlik, P. F. J. Am. Chem. Soc. 1968, 90, 4464.
(25) Yu, W.; Jin, Z. Tetrahedron Lett. 2001, 42, 369.
(26) Davis, F. A.; Sheppard, A. C. Tetrahedron 1989, 45, 5703.
(27) Corey, E. J.; Ensley, H. E. J. Am. Chem. Soc. 1975, 97, 6908.
(28) (a) Duhamel, P.; Cahard, D.; Poirier, J. M. J. Chem. Soc., Perkin Trans. 1
1993, 21, 2509. (b) Quesnel, Y.; Bidois-Sery, L.; Poirier, J.-M.; Duhamel,
L. Synlett 1998, 413.
(31) Lipshutz, B. H.; Pollart, D.; Monforte, J.; Kotsuki, H. Tetrahedron Lett.
1985, 26, 705.
(32) Schwesinger, R. Angew. Chem., Int. Ed. Engl. 1987, 26, 1164.
(33) Schmuff, N. R.; Trost, B. M. J. Org. Chem. 1983, 48, 1404.
(34) Snider, B. B.; Shi, B.; Tetrahedron 1999, 55, 14823.
(35) Duhamel, P.; Cahard, D.; Poirier, J. M. J. Chem. Soc., Perkin Trans. 1
1993, 21, 2509.
(29) We found that LiAlH4 reduction of an analogue of compound 75 without
C-17 hydroxy group was very slow at low temperature. Yu, W.; Jin, Z.
Unpublished results.
(30) Jiang, Z. H.; Schmidt, R. R. Liebigs Ann. Chem. 1992, 975.
(36) Davis, F. A.; Sheppard, A. C. Tetrahedron 1989, 45, 5703.
9
6582 J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002

Synthesis of the Anticancer Natural Product OSW-1
A R T I C L E S
lective reduction of 84 by LiAlH4 at -78 °C provided the
requisite trans-16â,17R-diol 85 in 97% yield. The stereochem-
istry at C16 and C17 of compound 85 was determined by
NOESY spectra. Thus, the protected aglycon of OSW-1 (1) was
synthesized with eight operations in 48% overall yield.
Coupling of disaccharide 10 with the steroid aglycone 85
under the standard conditions30 gave â-glycoside 86 in 71%
yield. Removal of all the protecting groups by sequential
treatment of compound 86 with DDQ and bis(acetonitrile)-
dichloropalladium(II) in one operation afforded OSW-1 (1) in
81% yield (Figure 20). The physical data of synthetic OSW-1
(1) are identical to those reported by Sashida.1
a steroid side chain. This total synthesis demonstrated once again
the versatile synthetic applications of R-halo vinyl ether
chemistry developed in our laboratories. Currently, synthesis
of designed analogues of OSW-1 (1) and invesitgation of its
structure-activity relationship are on the way in our laboratories
and will be reported in due course.
Acknowledgment. This work was supported by a Research
Project Grant RPG-00-030-01-CDD from the American Cancer
Society, the Central Investment Fund for Research Enhancement
at The University of Iowa. Thanks are due to the Center for
Biocatalysis and Bioprocessing at The University of Iowa for
providing a fellowship to W. Yu. We also express our apprecia-
tion to Ms. Johanne Ohannessian and Mr. Zhenmao Hua for
their help in some of the experiments.
Conclusions
In conclusion, the highly potent anticancer natural product
OSW-1 (1) has been successfully synthesized in only 10 linear
operations from the commercially available starting material
5-androsten-3â-ol-17-one 79 in 28% overall yield. The highly
convergent and stereoselective construction of the protected
aglycon has verified that 1,4-addition of an acyl anion equivalent
to 17(20)-en-16-one steroids is an attractive strategy to install
Supporting Information Available: Complete experimental
procedures and spectroscopic and analytical data including
copies of NMR spectra (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
JA012119T
9
J. AM. CHEM. SOC. VOL. 124, NO. 23, 2002 6583