Synthesis of the aglycone of the shark repellent pavoninin-4 using remote functionalization

Article abstract of DOI:10.1021/ol060079y

The aglycone of shark repellent pavoninin-4, (25R)-5α-cholestan- 3α,15α,26-triol 26-acetate 1a, was synthesized from (25R)-cholest-5-en-3β,-26-diol 4 (26-hydroxycholesterol) in eight steps in 18% overall yield. Breslow's remote functionalization strategy

Full text of DOI:10.1021/ol060079y

ORGANIC
LETTERS
2006
Vol. 8, No. 11
2253-2255
Synthesis of the Aglycone of the Shark
Repellent Pavoninin-4 Using Remote
Functionalization
Hua Gong and John R. Williams*
Department of Chemistry, Temple UniVersity, Philadelphia, PennsylVania 19122-2585
john.r.williams@temple.edu
Received January 11, 2006
ABSTRACT
The aglycone of shark repellent pavoninin-4, (25R)-5r-cholestan-3r,15r,26-triol 26-acetate 1a, was synthesized from (25R)-cholest-5-en-3â,-
26-diol 4 (26-hydroxycholesterol) in eight steps in 18% overall yield. Breslow’s remote functionalization strategy was used as a key step to
introduce the C-15
r
alcohol on a steroid D ring. An efficient synthesis of the 26-hydroxycholesterol from the 16â hydroxyl steroid, (25R)-
cholest-5-ene-3 ,16
â
â,26-triol (3a), is also reported.
During the latter half of the 20th century, chemists have
sought to discover nonmicrobial methods for the function-
alization of unactivated carbon atoms of various molecules.1
Breslow and co-workers were pioneers in this area, whose
works involved the generation of a tertiary carbon radical
on the R-face of the steroid nucleus by an attached template
through a radical relay reaction (Breslow reaction) and
benzophenone irradiation.1 In 1973, Breslow et al. reported
that the reaction of a series of esters derived from 3R-
hydroxy-5R-cholestane and p-benzoylphenylalkanoic acids,
on irradiation in benzene, underwent intramolecular attack
by the attached benzophenone on the steroid hydrogens and
gave alkenes in moderate yield.2 The remote functionalization
of the C-14 of steroids by photolysis of benzophenone esters
R-linked to the steroid backbone at C-3 can be also used to
introduce the C-15 keto2 and C-15R hydroxy3 groups in
steroids. Recently, the remote functionalization technique has
been found to be useful in the synthesis of natural steroids.4
In our approaches to the synthesis of the shark repellents,
pavoninin-4, 1b, and -5, 2b5,6 (Figure 1), diosgenin was
Figure 1. Shark repellents pavoninin-4, 1b, and pavoninin-5, 2b.
chosen as the starting material because it has an appropriately
placed oxygen functionality. Reduction of diosgenin resulted
in the preparation of the (25R)-cholest-5-en-3â,16â,26-triol
(3a). Transposition of the C-16â hydroxyl of the triol 3a to
(1) For reviews on remote functionalization reactions in steroids, see:
(a) Breslow, R. Acc. Chem. Res. 1980, 13, 170. (b) Reese, P. B. Steroids
2001, 66, 481.
(2) Breslow, R.; Baldwin, S.; Flechtner, T.; Kalicky, P.; Liu, S.;
Washburn, W. J. Am. Chem. Soc. 1973, 95, 3251.
(3) For a recent reference, see: Yang, J.; Gabriele, B.; Belvedere, S.;
Huang, Y.; Breslow, R. J. Org. Chem. 2002, 67, 5057.
(4) For example, the first total synthesis of xestobergsterol A: (a) Jung,
M. E.; John, T. W. Org. Lett. 1999, 1, 1671. (b) Jung, M. E.; John, T. W.
J. Am. Chem. Soc. 1997, 119, 12412.
10.1021/ol060079y CCC: $33.50
© 2006 American Chemical Society
Published on Web 05/02/2006

the 15R position afforded a method for the synthesis of the
aglycone of 26-O-deacetyl pavoninin-5, 2a.7 Recently, we
reported a more efficient method for transposition of a C-16â
hydroxyl to the 15R position, via the unexpected â-reduction
of a C-15 ketone.8 On the basis of this 1,2-transposition
methodology, the first synthesis of pavoninin-4, 1b, from
diosgenin has been achieved.9 In this paper, we report an
alternative synthesis of the aglycone of pavoninin-4 (1a) from
(25R)-26-hydroxycholesterol 4, an intermediate in the meta-
bolic pathway of cholesterol, using Breslow’s remote func-
tionalization strategy to biomimetically introduce a C-15R
hydroxyl on a steroid D ring. A very efficient synthesis of
the starting (25R)-26-hydroxycholesterol 4, from diosgenin,
is also reported.
The synthesis of the aglycone of pavoninin-4, 1a, started
from (25R)-26-hydroxycholesterol 4 because it is suitably
substituted with alcohols at C-3 and C-26. The syntheses of
(25R)-26-hydroxycholesterol 4 have been reported mainly
from two readily available natural products, kryptogenin and
diosgenin, and by addition of a side-chain building block to
the steroid backbone.10 The best yield achieved so far was
58% in four steps from diosgenin, via a modified Clem-
mensen reduction followed by a Barton deoxygenation
reaction.10 To remove the C-16â hydroxy of the (25R)-
cholest-5-en-3â,16â,26-triol (3a), selective protection of the
C-3â and C-26 hydroxyl groups is needed. This can be
achieved by chemoselective reaction of the cholest-5-en-3â,-
16â,26-triol (3a) with tert-butyldimethylsilyl chloride10 or
benzoyl chloride11 under basic conditions into the corre-
sponding 3,26-disilyl ethers or 3,26-dibenzoyl esters, re-
spectively. However, normally flash column chromatography
is required to separate the desired disubstituted alcohol from
other di- and trisubstituted byproducts. Actually, the benzoyl
group is not bulky enough for this chemoselective protection
because the expected selective product (25R)-cholest-5-en-
3â,26-dibenzoyloxy-16-ol was formed in only 36% yield.11
We found that trimethylacetyl chloride was a very good
alternative to benzoyl chloride for the chemoselective
conversion of the C-3 and C-26 alcohols of the triol 3a to
the corresponding C-3 and C-26 esters (Scheme 1).
Scheme 1. Synthesis of (25R)-26-Hydroxycholesterol, 4
product could be easily purified by a simple recrystallization
from ethyl acetate to yield pure 3â,26-dipivaloate 3b.
Mesylation of the free C-16 alcohol of 3b and subsequent
LiAlH4 reduction of the C-3 and C-26 diprotected 16â-
mesylate under reflux conditions proceeded smoothly to
reduce off both dipivaloates and at the same time displace
the C-16 mesylate. Recrystallization afforded (25R)-26-
hydroxycholesterol 4 in 90% yield and 97% purity based
on a GC-MS analysis.13 Therefore, this method provides an
efficient method for the synthesis of (25R)-26-hydroxycho-
lesterol (4) from the triol 3 in 83% yield over three steps.
This represents a considerable improvement over the Barton
deoxygenation described in our previous paper.10 More
importantly, this synthesis can be easily scaled up, because
all purifications were achieved by recrystallization, without
requiring column chromatography.
Starting from (25R)-26-hydroxycholesterol 4, the aglycone
of pavoninin-4 (1a) was synthesized using Breslow’s remote
functionalization strategy, as shown in Scheme 2.
Catalytic reduction of (25R)-26-hydroxycholesterol 4 af-
forded the saturated diol 5a, which was chemoselectively
protected14 at the primary C-26 hydroxyl by 3-pivaloyl-1,3-
thiazolidine and sodium hydride as the 3â,26-diol 26-
pivaloate 5b in 80% yield. Treatment of the 3â alcohol 5b
under Mitsunobu conditions with the known acid15 7 yielded
the inverted ester 8 in 87% yield, with the required 3R
stereochemistry for pavoninin-4, 1b. Photolysis of 8 using a
450-W medium-pressure Hanovia lamp with a Pyrex filter
and subsequent hydrolysis of the C-3 and C-26 esters with
potassium hydroxide in reflux conditions, followed by
conversion of the resulting 3R,26-diol into a diacetate,
provided a mixture of the desired (25R)-5R-cholestan-14-
en-3R,26-diol 3R,26-diacetate 9a and the unreacted saturated
(25R)-5R-cholestan-3R,26-diol 3R,26-diacetate 9b in a 3:2
ratio in 65% combined yield. Subsequent hydroboration and
oxidation16 of the unsaturated and saturated mixture 9a,b,
accompanied by a simultaneous hydrolysis of bisprotected
acetyl groups at C-3 and C-26 hydroxyls, afforded an easy
separation of the more polar (25R)-5R-cholestan-3R,15R,-
26-triol, 10. The chemical shift for the C-15â hydrogen in
10 is a doublet of triplets (J ) 9.1, 3.0 Hz) at 3.80 ppm.
These values are consistent with (25R)-5R-cholestan-3â,-
15R,26-triol, the 15R configuration of which was confirmed
by an X-ray structure.8
Trimethylacetyl chloride has been used as an excellent
reagent for selective acylation of a primary alcohol over a
secondary one.12 The selectively diprotected 3â,26-dipiv-
aloate 3b was formed almost quantitatively so that the
(5) (a) Tachibana, K.; Sakaitani, M.; Nakanishi, K. Science 1984, 226,
703. (b) Tachibana, K.; Sakaitani, M.; Nakanishi, K. Tetrahedron 1985,
41, 1027.
(6) For a recent review, see: Williams, J. R.; Gong, H. Lipids 2004, 39,
795.
(7) Williams, J. R.; Chai, D.; Bloxton , J. D., II; Gong, H.; Solvibile,
W. R. Tetrahedron 2003, 59, 3183.
(8) Williams, J. R.; Gong, H.; Hoff, N.; Olubodun, O. I.; Carroll, P. J.
Org. Lett. 2004, 6, 269.
(9) Williams, J. R.; Gong, H.; Hoff, N.; Olubodun, O. I. J. Org. Chem.
2005, 70, 10732.
(10) Williams, J. R.; Chai, D.; Wright, D. Steroids 2002, 67, 1041 and
references therein.
(11) Noam, M.; Tamir, I.; Breuer, E.; Mechoulam, R. Tetrahedron 1981,
37, 597.
(12) (a) Nicolaou, K. C.; Webber, S. E. Synthesis 1986, 453. (b) Boschelli,
D.; Takemasa, Y.; Nishitani, Y.; Masamune, S. Tetrahedron Lett. 1985,
26, 5239. (c) Nagaoka, H.; Rutsch, W.; Schmid, G.; Ilio, H.; Johnson, M.
R.; Kishi, Y. J. Am. Chem. Soc. 1989, 102, 7962.
(13) GC-MS analysis was provided by Dr. J. Goodman at the University
of Kentucky.
(14) Yamada, S. J. Org. Chem. 1992, 57, 1591.
(15) Zderic, J. A.; Kubitschek, M. J.; Bonner, W. A. J. Org. Chem. 1961,
26, 1635.
(16) Taylor, E. J.; Djerassi, C. J. Org. Chem. 1977, 42, 3571.
2254
Org. Lett., Vol. 8, No. 11, 2006

Scheme 2. Synthesis of the Aglycone of the Shark Repellent Pavoninin-4, 1a
yield over eight steps, using Breslow’s remote functional-
ization strategy as a key step.
Finally, selective acetylation of the primary C-26 hydroxyl
by Otera’s transesterification strategy,17 using vinyl acetate
with distannoxane as a catalyst, afforded the aglycone of
pavoninin-4 (1a) in 80% yield. This represents a new formal
synthesis of the shark repellent pavoninin-4.9
Acknowledgment. Financial support for this research was
provided by grants from the Temple University Research
Incentive Fund, GlaxoSmithKline, Pfizer, Bristol-Myers
Squibb, Incyte, and Merck. We thank Dr. Jeffrey Honovich
of Drexel University for the high resolution mass spectra.
In conclusion, we report a highly efficient method for the
synthesis of the (25R)-26-hydroxycholesterol 4 in 83% yield
over three steps, from the (25R)-cholest-5-en-3â,16â,26-triol
3a. From this useful intermediate 4, the synthesis of the
aglycone of pavoninin-4, 1a, was achieved in 18% overall
Supporting Information Available: General experimen-
tal procedures and 1H and 13C NMR spectra for all
synthesized compounds 3b, 4, 5a,b, 6-8, 10, and pavoni-
nin-4 aglycone 1a. This material is available free of charge
via the Internet at http://pubs.acs.org.
(17) (a) Orita, A.; Mistutome, A.; Otera, J. J. Org. Chem. 1998, 63, 2420.
(b) Orita, A.; Sakamoto, K.; Hamada, K.; Mistutome, A.; Otera, J.
Tetrahedron 1999, 55, 2899.
OL060079Y
Org. Lett., Vol. 8, No. 11, 2006
2255