Synthesis of 7-Deoxyxestobergsterol A, a Novel Pentacyclic Steroid of the Xestobergsterol Class

Full text of DOI:10.1021/ja9733189

12412
J. Am. Chem. Soc. 1997, 119, 12412-12413
Scheme 1
Synthesis of 7-Deoxyxestobergsterol A, a Novel
Pentacyclic Steroid of the Xestobergsterol Class1
Michael E. Jung*,2 and Ted W. Johnson
Department of Chemistry and Biochemistry
UniVersity of California at Los Angeles
Los Angeles, California 90095-1569
ReceiVed September 22, 1997
In 1992, Umeyama and co-workers reported the isolation of
xestobergsterols A and B, 1ab, novel pentacyclic steroids with
a cis C/D ring junction which are powerful inhibitors of
histamine release.3 Three years later, another member of this
novel class, xestobergsterol C, 1c, was isolated and the original
stereochemistry proposed for C23 was corrected.4 Another
inhibitor of histamine release is contignasterol, 2, which has a
similar structure but is missing the additional E ring.5 To date,
class, the nonnatural analogue 7-deoxyxestobergsterol A (1d)
by a route which should be applicable to the natural products.
Breslow has reported a very useful technique for “remote
functionalization” of steroids that involves the abstraction of a
tertiary hydrogen atom by a radical species generated by
photolysis of an appropriate precursor which is attached to the
steroid backbone so that it sits preferentially under one of the
several abstractable tertiary hydrogens in a steroid.10 We first
tested this process for the synthesis of the steroid skeleton
necessary for the xestobergsterols (Scheme 1). The diol 3,
prepared in 78% yield by hydroboration-oxidation of choles-
teryl acetate, was treated with the known acid 411 and excess
DEAD and triphenylphosphine to produce the monoester 5 in
60% yield. This selective Mitsunobu reaction proceeds well
due to the severe steric hindrance in the inversion of the 6R-
alcohol. Photolysis and cleavage with hydroxide followed by
protection gave the desired ∆14-alkene diacetate 6 in 45%
isolated yield along with 35% of the saturated 3R,6R-diacetate,
which could be recycled. Conversion of the alkene to the
desired 14â-H 15-ketone 7 was quite difficult. Simple hy-
droboration-oxidation gave mainly the 14R-H 15R-alcohol12
which could be oxidized to the 14R-H 15-ketone, but this
compound was not epimerized to the desired epimer. Therefore
a stereoselective route was developed. Epoxidation13 of 6 gave
the R-epoxide which was opened with hydrogen over platinum
oxide in acetic acid to give the 14â-H 15R-alcohol which was
then oxidized with PCC to the desired ketone 7. Treatment of
7 with either strong base or strong acid caused epimerization
to the more stable 14R-H 15-ketone.
very little has been published on routes to these molecules.6
All of these compounds strongly inhibited histamine release
from rat mast cells induced by anti-IgE in a dose-dependent
manner (1a IC50 0.05 µM; 1b 0.10 µM; and 2 0.8 µM).3,7 Since
the well-known antiallergy drug disodium cromoglycate has an
IC50 of 262 µM, these compounds are much more potent
inhibitors of histamine release, being up to 5000 times more
active inhibitors. Contignasterol is also very potent in both in
vivo and in vitro models of allergen-induced bronchoconstriction
and airway smooth muscle contraction and therefore is extremely
effective as an antiasthma agent.8 Recently it has been shown
that the mechanism of action of xestobergsterol A is through
strong inhibition of phosphatidylinositol phospholipase C (PI-
PLC).9 We report herein the first total synthesis of this structural
(1) Presented at the 213th National American Chemical Society meeting,
San Francisco, CA, April 1997, ORGN 118.
(2) American Chemical Society Arthur C. Cope Scholar, 1995.
(3) Shoji, N.; Umeyama, A.; Shin, K.; Takeda, K.; Arihara, S.; Kobayashi,
J.; Takei, M. J. Org. Chem. 1992, 57, 2996.
(4) Kobayashi, J.; Shinonaga, H.; Shigemori, H.; Umeyama, A.; Shoji,
N.; Arihara, S. J. Nat. Prod. 1995, 58, 312.
(5) Burgoyne, D. L.; Andersen, R. J.; Allen, T. M. J. Org. Chem. 1992,
57, 525.
(6) Krafft, M. E.; Chirico, X. Tetrahedron Lett. 1994, 35, 4511.
(7) Bramley, A. M.; Langlands, J. M.; Jones, A. K.; Burgoyne, D. L.;
Li, Y.; Andersen, R. J.; Salari, H. Br. J. Pharmacol. 1995, 115, 1433.
(8) Takei, M.; Burgoyne, D. L.; Andersen, R. J. J. Pharm. Sci. 1994,
83, 1234.
(9) Takei, M.; Umeyama, A.; Shoji, N.; Arihara, S.; Endo, K. Experientia
1993, 49, 145.
To determine the strategy necessary for final production of
the cis C/D ring junction as well as the final E ring, we carried
out a series of molecular mechanics calculations on a tricyclic
(10) (a) Breslow, R. Chemtracts 1988, 1, 333. (b) Breslow, R.; Baldwin,
S.; Flechtner, T.; Kalicky, P.; Liu, S.; Washburn, W. J. Am. Chem. Soc.
1973, 95, 3251.
(11) Zderic, J. A.; Kubitschek, M. J.; Bonner, W. A. J. Org. Chem. 1961,
26, 1635.
(12) In addition to the major product, a small amount of the desired 14â-H
15â-alcohol was also isolated (an approximate 10:1 ratio), which could be
taken on to 7 by simple oxidation.
(13) Lardon, A.; Sigg, H. P.; Reichstein, T. HelV. Chim. Acta 1959, 42,
1457.
S0002-7863(97)03318-0 CCC: $14.00 © 1997 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 119, No. 50, 1997 12413
Scheme 2
Hydroboration-oxidation of the alkene gave the 15R,23-diol16
which on oxidation with PCC gave the diketone 15. Acidic
removal of the TBS groups and final aldol condensation-
epimerization afforded 7-deoxyxestobergsterol A (1d) in nearly
quantitative yield.17 Thus the calculations in the model system
were borne out in the full steroidal skeleton. The structure was
assigned by comparison of the chemical shifts and coupling
constants to those reported for xestobergsterol A4 and by a
NOESY experiment which showed correlation between the
angular 16-H and the hydrogens at C24 and C25, thus
establishing the stereochemistry at C23.18
model system which indicated that the desired final aldol
product, A, was much more stable than any of the other possible
aldols, A′, B, or B′.14 Thus we theorized that one could carry
out the desired aldol on the more easily accessible 14R-H 15-
ketone to generate the desired 14â-H 15-ketone structure.
In summary, we have achieved the first total synthesis of a
pentacyclic steroid of the xestobergsterol class by an efficient
route, which utilizes the Breslow remote functionalization
process. In this way, 7-deoxyxestobergsterol A (1d) is available
in 17 steps and 10% overall yield from the known alcohol 9
(22 steps and 7% overall yield from stigmasterol (8)).19 Further
work on the synthesis of the other members of this class is in
progress.
Acknowledgment. We thank the National Institutes of Health and
the Agricultural Research Division of American Cyanamid Company
for generous financial support. We also thank Dr. Charles McCombs
of Eastman Chemical Company for a gift of stigmasterol.
Supporting Information Available: Experimental procedures for
compounds 12 and 13 and characterization data for compounds 10-
15 and 1d (3 pages). See any current masthead page for ordering and
Internet access instructions.
The synthesis of 7-deoxyxestobergsterol A (1d) began with
the known alcohol 9, prepared in 5 steps and 69% yield from
stigmasterol (8)15 (Scheme 2). Protection as the pivalate,
opening of the cyclopropylmethyl methyl ether, and protection
of the alcohol gave the 3â-benzyl ether 10 in excellent yield.
Hydroboration-oxidation of the alkene, followed by acetylation,
and debenzylation gave the 6R-acetoxy 3â-alcohol 11. Mit-
sunobu reaction with the acid 4 afforded the diester 12 which
was photolyzed, treated with base as above, and then protected
to give the desired ∆14-alkene trisilyl ether 13 in 48% yield
along with 32% of the saturated triether which could be recycled.
Monodeprotection of the primary silyl ether with aqueous acetic
acid gave the alcohol 14. Oxidation to the aldehyde and addition
of isobutylmagnesium bromide gave the secondary alcohol.
JA9733189
(16) Again a small amount of the 14â-H 15â-alcohol was also isolated
(an approximate 10:1 ratio) in this hydroboration-oxidation, which could
be oxidized to the 14â-H epimer of 15.
(17) No intermediates are observed in this reaction so we have no
evidence for whether aldol condensation precedes epimerization at C-14
or vice versa.
(18) The stereochemistry of the 14-hydrogen is easily established by the
chemical shift of the 7-hydrogens. In the ketone 15, the 7â-H resonates at
low field (δ 2.87) as a double of triplets (J ) 12.6, 3.2 Hz), whereas in
compound 7 or 1d, the 7R-H resonates at low field as a pseudo quartet (7,
δ 2.55, J ) 12.8 Hz; 1d, δ 2.51, J ) 11.7 Hz). The low-field resonance is
due to the deshielding effect of the 15-ketone which has a pseudo 1,3-
diaxial interaction with the 7R- or 7â-H, depending on the C-14 stereo-
chemistry.
(14) The eight additional possible aldol products, all bridged systems,
had energies much higher than any of these four.
(15) Cho, J.-H.; Djerassi, C. J. Org. Chem. 1987, 52, 4517.
(19) No serious attempts have been made to optimize the yields at each
step, so it is likely that the overall yield and efficiency of this process can
be increased.