First total synthesis of xestobergsterol A and active structural analogues of the xestobergsterols.

Article abstract of DOI:10.1021/ol991057x

[formula: see text] A novel pentacyclic polyhydroxylated sterol, xestobergsterol A (1a), has been synthesized in 24 steps and in good overall yield from stigmasterol 17. The key steps of the synthesis are the Breslow remote functionalization of the polyox

Full text of DOI:10.1021/ol991057x

ORGANIC
LETTERS
1999
Vol. 1, No. 10
1671-1674
First Total Synthesis of Xestobergsterol
A and Active Structural Analogues of
the Xestobergsterols1
Michael E. Jung* and Ted W. Johnson2
Department of Chemistry and Biochemistry, UniVersity of California,
Los Angeles, California 90095-1569
jung@chem.ucla.edu
Received September 16, 1999
ABSTRACT
A novel pentacyclic polyhydroxylated sterol, xestobergsterol A (1a), has been synthesized in 24 steps and in good overall yield from stigmasterol
17. The key steps of the synthesis are the Breslow remote functionalization of the polyoxygenated steroid derived from 25 and the base-
catalyzed epimerization−aldol condensation of the dione derived from 27.
The xestobergsterols are a class of novel pentacyclic poly-
hydroxylated 14â 15-keto steroids which are strong inhibitors
of the release of histamines from rat mast cells.3 Three such
compounds have been isolated, xestobergsterols A-C (1a-
c), and their structures determined4 (Figure 1). Three other
steroidal compounds, contignasterol (2),5 haliclostanone (3),6
and 15-dehydro-14â-anosmagenin,7 have also been isolated,
all of which have similar structures but are missing the
additional carbocyclic E ring. Of these, only contignasterol
has been shown to have histamine release inhibitory proper-
ties, being 16 times less active than xestobergsterol A (IC50:
1a, 50 nM; 1b, 100 nM; 2, 800 nM).3,8 Recently we published
the first total synthesis9 of a member of this class, namely
7-deoxyxestobergsterol A (1d), by a route that used an
application of Breslow’s remote functionalization process10
(to produce the 14,15-alkene) and a novel epimerization-
aldol condensation process to form the DE ring system with
the correct stereochemistry at C14, C16, and C23. Since our
work, the Krafft group11 has reported an approach to
analogues of 1 using the same final aldol condensation
sequence. We report herein the first total synthesis of the
most potent member of this structural class, xestobergsterol
A (1a), as well as two analogues and the biological activity
of these compounds.
Before beginning the total synthesis itself, we first tested
our route on a simpler model having the unfunctionalized
cholesterol side chain (Scheme 1). Of the possible methods12
for the oxidation of cholesteryl acetate 4 to the enone 5 we
(8) (a) Takei, M.; Umeyama, A.; Shoji, N.; Arihara, S.; Endo, K.
Experientia 1993, 49, 145. (b) Takei, M.; Burgoyne, D. L.; Andersen, R. J.
J. Pharm. Sci. 1994, 83, 1234. (c) Bramley, A. M.; Langlands, J. M.; Jones,
A. K.; Burgoyne, D. L.; Li, Y.; Andersen, R. J.; Salari, H. Brit. J.
Pharmacol. 1995, 115, 1433. (d) These compounds are significantly more
active inhibitors than is the antiallergy drug disodium cromoglycate, which
has an IC50 of 262 µM for histamine release.
(9) Jung, M. E.; Johnson, T. W. J. Am. Chem. Soc. 1997, 119, 12412.
(10) (a) Breslow, R. Chemtracts: Org. Chem. 1988, 1, 333. (b) Breslow,
R.; Baldwin, S.; Flechtner, T.; Kalicky, P.; Liu, S.; Washburn, W. J. Am.
Chem. Soc. 1973, 95, 3251.
(1) Presented at the 76th Japan Chemical Society meeting, Tokyo, March
1999.
(2) Saul Winstein Fellow; UCLA Graduate Division Fellow, 1998-1999.
(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.
(11) (a) Krafft, M. E.; Dasse, O. A.; Fu, Z. J. Org. Chem. 1999, 64,
2475. For earlier work by this group, see: b) Krafft, M. E.; Dasse, O. A.;
Shao, B. Tetrahedron 1998, 54, 7033. (c) Krafft, M. E.; Chirico, X.
Tetrahedron Lett. 1994, 35, 4511.
(6) Crews, P.; Sperry, S. J. Org. Chem. 1997, 60, 229.
(7) Gonzalez, A. G.; Freire-barreira, R.; Garcia-Francisco, C.; Salazar-
Rocio, J. A.; Suarez-Lopez, E. An. Quim. 1974, 70, 250.
10.1021/ol991057x CCC: $18.00 © 1999 American Chemical Society
Published on Web 10/10/1999

Figure 1.
chose a catalytic ruthenium-based oxidation which gave 5
in 63% yield. Although other oxidants, e.g. chromium, gave
higher yields, this proved the best method for efficient
preparation of 5. Luche reduction gave the equatorial allylic
alcohol 6, which was acetylated to give the diacetate 7 in
good yield. Hydroboration-oxidation of 7 followed by
peracetylation afforded the triacetate 9 in 79% yield from 7.
Selective hydrolysis of the least hindered 3R-acetate gave
in excellent yield the monoalcohol, which was coupled with
the acid 1013 using Mitsunobu conditions to give the
photolysis substrate 11. Photolysis and cleavage of all of
the esters furnished the desired ∆14-alkene triol 12 in good
yield. Thus, the additional acetate at C7 causes no problems
with the Breslow remote functionalization process. The
completion of the synthesis of the xestobergsterol A model
16 (Scheme 2) required protection of the triol, which proved
to be very difficult, since the 7â-alcohol is quite hindered.14
We solved this problem by treating the triol 12 with methylal
14R-H 15-ketone 14 in good yield. We carried out molecular
mechanics (MM2) strain energy calculations on 14 and its
14-epimer 15 which predict the energy difference to be about
1.6 kcal/mol in favor of 15. This is in direct contrast to the
7-deoxy series, where the trans CD ring juncture was
significantly more stable than the corresponding cis CD ring
juncture.9 The difference is presumably due to the steric
interaction of the 7-alkoxy group with the C14-C15 bond
since they are in a pseudo syn-pentane arrangement. In
agreement with the calculations, treatment of 14 with base
under mild conditions gave completely the isomeric ketone
15.17 Acidic hydrolysis of the acetal protecting groups in 15
gave the xestobergsterol A analogue 16 having the simple
cholesterol side chain, which is available from 4 in 14 steps
and 6% overall yield.
The successful completion of the synthesis of the tetra-
cyclic analogue of xestobergsterol A 16 led us to complete
the synthesis of the natural product. Commercially available
stigmasterol 17 was converted into the alcohol 18 in four
steps, as described previously (Scheme 3).9 Acetylation and
opening of the cyclopropylcarbinyl ether with acetic acid
15
and P2O5 to give the MOM ether cyclic methylene acetal
13 in good yield. Hydroboration-oxidation of 13 gave 75%
of the 14R-H 15R-alcohol,16 which was oxidized to the
Scheme 1
1672
Org. Lett., Vol. 1, No. 10, 1999

Scheme 2
gave the diacetate 19 in excellent yield. Allylic oxidation of
in situ).18 The final stages of the synthesis were modeled on
our earlier synthesis of the 7-deoxy analogue.9 Desilylation
of 26 followed by oxidation and addition of isobutyl Grignard
reagent afforded the alcohols 27 as a 1:1 diastereomeric
mixture in good yield. Hydroboration-oxidation of the
alkene gave mainly the 14â-H 15â-alcohol, which was
oxidized under Dess-Martin conditions19 to give the 15,-
23-dione; this compound was then epimerized and cyclized
under basic conditions to afford the desired protected
xestobergsterol A 28 in 70% overall yield from 27.20 The
final step is the acidic hydrolysis of the MOM ethers to give
xestobergsterol A (1a) in 80% yield, thus ending a relatively
short synthesis of 1a from stigmasterol 17.21 The structure
was assigned by comparison (spectral data, TLC, optical
rotation)22 to those reported for xestobergsterol A.4,23
19 using catalytic ruthenium furnished the enone 20 which
was reduced to the 7â-alcohol 21 and then acetylated to give
the triacetate 22. Hydroboration-oxidation of 22 afforded
the tetrol 23, which was converted via the tetraacetate 24
into the diol diacetate 25 in good yield. Selective protection
of the primary alcohol in the presence of the secondary
alcohol gave the TIPS ether, which was converted into the
required benzophenone ester by a Mitsunobu coupling with
the acid 10 to give the 3R-ester. Photolysis as before and
PCC oxidation (to remove the 1:1 mixture of diastereomers
at the new benzhydryl center to facilitate purification) gave
the desired 14-alkene in 63% yield. Hydride reduction gave
the triol, which was converted into the tris-MOM ether 26
by reaction with MOM iodide (prepared from the chloride
Scheme 3
Org. Lett., Vol. 1, No. 10, 1999
1673

We have had two analogues of xestobergsterol A tested
for inhibition of histamine release from rat mast cells, namely
7-deoxyxestobergsterol A (1d)9 and the tetracyclic analogue
with the simple cholesterol side chain, 16. Both compounds
were very effective inhibitors in this assay (IC50: 1d, 500
nM; 16, 750 nM),24 indicating that both the E ring and the
oxidized side chain are not required for potent activity.
(12) (a) Parish, E. J.; Wei, T.-Y.; Livant, P. Lipids 1987, 22, 760. (b)
Miller, R. A.; Li, W. W.; Humphrey, G. R. Tetrahedron Lett. 1996, 37,
3429. (c) Selinsky, B. S.; Jones, S. R.; Kinney, W. A.; Rao, M. N.; Zhang,
X.; Tham, F. S. J. Org. Chem. 1998, 63, 3786. (d) Pearson, A. J.; Chen,
Y.-S.; Han, G. R.; Hsu, S.-Y.; Ray, T. J. Chem. Soc., Perkin Trans. 1 1985,
269. (e) Salvador, J. A. R.; Melo, M. L. S.; Neves, A. S. C. Tetrahedron
Lett. 1997, 38, 119.
In summary, we have achieved the first total synthesis of
xestobergsterol A by an efficient route, which utilizes the
Breslow remote functionalization process. Two analogues
prepared by this route have good activities in the standard
histamine release assay and may prove to be useful leads in
the search for new compounds for the treatment of allergic
diseases. Further work on the synthesis of the other members
of this class is in progress.
(13) Zderic, J. A.; Kubitschek, M. J.; Bonner, W. A. J. Org. Chem. 1961,
26, 1635.
(14) For example, treatment of the triol 12 with various silyl protecting
groups afforded mainly the disilyl ether. The triacetate could be prepared
but was unstable to the basic oxidation conditions of the next step.
(15) Fuji, K.; Nakano, S.; Fujita, E. Synthesis 1975, 276.
(16) In this reaction, we also isolated 13% of the 14â-H 15â-alcohol
which was oxidized to the ketone 15.
(17) The stereochemistry of the 14-hydrogen was easily established by
proton NMR, since the methyl group at C-13 resonates at higher field (δ
∼0.7-1.0) in the trans CD ring compounds and at lower field (δ ∼1.1-
1.3) in the cis CD ring compounds, e.g., at δ 0.78 in 14 and at δ 1.16 in
15. For additional examples, see: Johnson, T. W. Ph.D. Thesis, UCLA,
1999.
(18) Narasaka, K.; Sakakura, T.; Uchimaru, T.; Guedin-Vuong, D. J.
Am. Chem. Soc. 1984, 106, 2954.
(19) Martin, J. C.; Dess, D. B. J. Org. Chem. 1983, 48, 4156.
(20) Since the epimerization of 14 to give 15 is quite fast under these
conditions, we believe that epimerization precedes the aldol condensation
in this system.
Acknowledgment. We thank the National Institutes of
Health and the Agricultural Research Division of American
Cyanamid Co. for generous financial support. We thank Dr.
Akemi Umeyama (Tokushima Bunri University) for a sample
of natural xestobergsterol A and Dr. Maseo Takei (Japan
National Cancer Research Institute) for the biological assays.
We also thank Dr. Charles McCombs of Eastman Chemical
Co. for a gift of stigmasterol.
Supporting Information Available: Spectroscopic data
(proton and carbon NMR and IR) for all new compounds
reported. This material is available free of charge via the
Internet at http://pubs.acs.org.
(21) 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. This is especially true of the only very low yielding step,
namely the hydroboration-oxidation of 22, which usually proceeds in much
higher yield, e.g., 7 f 9 in 79% yield.
(22) The optical rotations[R]D25 ) -37.7°, c ) 2.8 (CHCl3)sis identical
with that reported for xestobergsterol A.23
OL991057X
(23) We thank Dr. Akemi Umeyama for providing us with a sample of
natural xestobergsterol A and Dr. Maseo Takei for informing us of the
optical rotation.
(24) We thank Dr. Takei for performing these assays.
1674
Org. Lett., Vol. 1, No. 10, 1999