In an effort to access the bicyclo[3.2.1]octane skeleton by
an intramolecular Prins reaction, aldehyde 18, obtained from
15 in two steps ((1) LiAlH4; (2) Dess-Martin periodinane)
in 95% overall yield, was treated with a catalytic amount of
Et2AlCl in CH2Cl2 at room temperature (Figure 2). While
Scheme 3. Synthesis of Tricyclic Ketone 23a
a Reagents and conditions: (a) MCPBA (2.3 mol equiv)/CH2Cl2,
rt, 4 h (97%); (b) 50% aq NaOH (8/1), PhSH (4 mol equiv), reflux,
6 h (99%); (c) PhOP(dO)Cl2 (2 mol equiv), Et3N (3 mol equiv),
0 °C, 1 h, then PhSeH (3 mol equiv)/Et3N (5 mol equiv), 0 °C, 0.5
h (70%); (d) (Bu3Sn)2 (2 mol equiv)/PhH, sun lamp, reflux, 2 h;
(e) Raney nickel/MeOH, -20 °C, 5 min; (f) H2 (55 psi), PtO2 (cat)/
EtOAc, 2.5 h.
Figure 2. Prins Reaction of Aldehyde 18.
the reaction produced the bicyclo[2.2.2]octane 19 as a 9:1
regioisomeric alkene mixture in 90% yield within 30 min,
favoring the endocyclic alkene, the formation of the desired
tricyclic alcohol with the bicyclo[3.2.1]octane skeleton, 20,
was not observed. In view of the similar SEs of these two
bicyclooctane systems,8 exclusive formation of 19 may be
attributable, at least partially, to the relative carbocation-
stabilizing abilities of the methyl and bromine groups in
intermediates i and ii, respectively.18
Our efforts were then directed to the use of a nonreductive
acyl radical cyclization to construct the bicyclo[3.2.1]octane
skeleton, 23. To obviate the complications resulting from
the direct interaction between the acyl radical and the S atom
introduced as a nucleophile during the solvolysis reaction
of 13, the 4-methoxyphenyl sulfide in 14 was first oxidized
to the corresponding sulfone (Scheme 3). In addition, it was
envisaged that, in view of the electrophilic nature of an acyl
radical, the presence of a bromine atom should make the
radical cyclization less favorable, and a means of replacing
the Br with an electron-donating group was sought. This
group would also have to be readily reductively removable.
Fortuitously, during an attempt to hydrolyze the methyl ester
of 14 with the 4-methoxyphenyl sulfone group (MeOH, aq
NaOH, reflux), it was observed that the bromine atom was
replaced with a methoxyl group in addition to hydrolysis of
the methyl ester. This presumably involved deprotonation
of the R-H to the sulfonyl group followed by replacement
of the Br with MeO in the vinyl sulfone intermediate.
Treatment of the sulfone obtained from sulfide 14 [Nu )
4-(MeO)PhS-] with PhSH under basic conditions resulted
in quantitative formation of the hydrolysis product with the
Br cleanly replaced by the PhS group (Scheme 3). Interest-
ingly, as in the case of using MeO- as the nucleophile, the
corresponding conjugated sulfone 24 was not observed. As
expected, the acyl radical generated from seleno ester 21
under nonreductive conditions underwent regioselective
cyclization to provide tricyclic ketone 22 with a bicyclo-
[3.2.1]octane framework. This was subsequently transformed
into tricyclic ketone 23.
The elaboration of the tricyclic ketone 23 into acanthodoral
(1) was achieved by the Wolff rearrangement reaction of
R-diazoketone 26 (Scheme 4), which was prepared from
R-oxime ketone 25. Treatment of ketone 23 with isoamyl
nitrite in the presence of KOBut resulted in the formation of
a mixture of syn- and anti-oxime derivatives 25 (δ 13.19
(12) (a) Takai, K.; Hotta, Y.; Oshima, K.; Nozaki, H. Tetrahedron Lett.
1978, 2417-2420. (b) Takai, K.; Hotta, Y.; Oshima, K.; Nozaki, H. Bull.
Chem. Soc. Jpn. 1980, 53, 1698-1702. (c) Lombardo, L. Tetrahedron Lett.
1982, 23, 4293-4296. See also: (d) Lombardo, L. Organic Syntheses;
Wiley: New York, 1993; Collect. Vol. 8, pp 386-390.
(13) Oppolzer, W.; Bedoya-Zurita, M.; Switzer, C. Y. Tetrahedron Lett.
1988, 29, 6433-6436.
1
and 8.80 ppm, respectively, for the oxime OH H peaks).
The ratio of these two isomers was dependent upon the
amount of base used. If the reaction was quenched with acetic
acid at -78 °C, the formation of syn isomer was favored by
5:1. In contrast, when a large excess (6 mol equiv) of the
base was used and the reaction mixture was allowed to warm
to room temperature over 1 h, the anti isomer was predomi-
nant (2:1 and 88% combined yield). This ratio favoring
formation of the anti isomer of 25 could not be improved.
However, the isolated syn isomer could be equilibrated to a
2:1 anti/syn mixture under photochemical conditions in
MeOH (rt, 1 h), providng the desired anti isomer in 75%
(14) Review: Oppolzer, W. Angew. Chem., Int. Ed. Engl. 1989, 28, 35-
52.
(15) For another example of a palladium-ene reaction that produces an
endocyclic alkene product, see: Zair, T.; Santellirouvier, C.; Santelli, M.
Tetrahedron 1993, 49, 3313-3324.
(16) Schadt, F. L.; v. R. Schleyer, P.; Bentley, T. W. Tetrahedron Lett.
1974, 2335-2338.
(17) (a) Danheiser, R. L.; Morin, J. M.; Yu, M.; Basak, A. Tetrahedron
Lett. 1981, 22, 4205-4208. (b) Banwell, M. G.; Harvey, J. E.; Hockless,
D. C. R.; Wu, A. W. J. Org. Chem. 2000, 65, 4241-4250.
(18) Olah, G. A.; Mo, Y. K. In Carbonium Ions; Olah, G. A., v. R.
Schleyer, P., Eds.; Wiley-Interscience: New York, 1976; Vol. 5, pp 2135-
2262.
Org. Lett., Vol. 6, No. 4, 2004
539