bazole was accomplished via oxidative cleavage of the
double bond (via the diol) followed by treatment of the
resulting dicarbonyl with acid to afford 14 in 95% yield over
the three steps. Treatment of tetrahydrocarbazole 14 with
triphenylphosphonium isopropyl ylide produced the desired
prenyl moiety and alkene 15 in 73% yield. Subjecting alkene
15 to Sharpless asymmetric dihydroxylation conditions8 using
(DHQD)2PHAL resulted in diol formation in 98% yield and
45% ee.9 While the enantioselectivity for the dihydroxylation
was less then optimal, it was deemed sufficient to obtain
optical rotations and provide insight into the stereochemistry
of the natural products and was therefore carried through
the synthesis as the scalemic mixture. With the diol in hand
treatment with 2-methoxypropene under acidic conditions
provided acetonide 16 in 98% yield. Subjection of 16 to
carbonylative conditions yielded the corresponding ester in
59% yield along with unreacted starting material which could
be resubjected to the reaction conditions to give the desired
ester in 71% (96% BRSM) after two cycles.10
Detosylation was then carried out with TBAF to yield
clausamine B in 84% and 46% ee. Again the hydroxyl group
was relinquished via treatment of clausamine B with BBr3
allowing access to clausamine A in 75% yield and 35% ee.13
The stereochemistry of the diol and natural products were
assigned by using the Sharpless mnemonic8 (Scheme 5) in
Scheme 5. Assignment of Stereochemistry
With the phenoxy ester in hand treatment with DDQ
afforded the fully aromatized carbazole 17 in 94% yield.
Simple deprotection of the acetonide with TsOH in ethylene
glycol allowed for the formation of desired lactone 18 in
nearly quantitative yield. Finally, removal of the tosyl group
with TBAF11 produced clausamine C in 82% yield and 47%
ee. Treatment of clausamine C with BBr3 then revealed the
hydroxyl group, thus affording clausevatine D in 72% yield
and 49% ee.
With the completion of clausamine C and clausevatine D,
we turned our attention to the syntheses of clausamines A
and B (see Scheme 4). Our initial attempts to form the
which dihydroxylation occurs from the top face of the alkene
leading to the formation of the (R)-alcohol. With the
successful completion of clausamines A-C and clausevatine
D the optical rotations were measured. Interestingly, contrary
to what was expected, all compounds gave strong positive
rotations of 62.8°, 33.2°, 50.0°, and 50.7° for clausamines
A-C and clausevatine D, respectively, despite their modest
enantiomeric excess. Even at the low concentrations reported
in the literature for clausamines A-C a rotation was
observed. Furthermore, the optical rotation observed for
clausevatine D was an order of magnitude larger than that
which was reported (note also, opposite in sign). Although
a more detailed comparison of the rotations of the synthetic
samples to the natural products is warranted, we were unable
to obtain natural samples toward this end.
While these results may raise more questions than they
answer, several possible explanations exist. It is clear that
these molecules do not have inherently small rotations since
our samples with modest enantiomeric purities gave signifi-
cant rotations. It is possible that the natural products then,
are racemic (or nearly so). Assuming that an enzymatic
process produced enantiomerically distinct products, race-
mization may have occurred post-biosynthesis, either in
nature or during laboratory handling in the isolation process.
What is known, however, is that, based on the generally
Scheme 4. Synthesis of Clausamines A and B
required isopropenyl moiety from alcohol 18 using typical
dehydration conditions (POCl3, pyridine; SOCl2, pyridine;
MsCl, NEt3 or NEt(iPr)2) resulted in significant formation
of the undesired endo tetrasubstituted double bond being
formed. Fortunately, this could be circumvented by treatment
of the alcohol with Martin’s sulfurane12 which formed the
desired exo methylene almost exclusively in 77% yield.
(8) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J.; Jeong, K.-S.; Kwong, H.-L.; Morikawa, K.; Wang, Z.-M.; Xu,
D.; Zhang, X. L. J. Org. Chem. 1992, 57, 2768.
(9) A modest increase to 55% ee could be obtained by cooling reaction
to 0 °C but required significantly longer reaction times.
(10) Initial attempts to carry out a carbonylative lactonization on the
unprotected diol leading directly to the lactone ring were unsucessful,
forming instead a benzodihydrofuran.
(11) Yasuhara, A.; Sakamoto, T. Tetrahedron Lett. 1998, 39, 595.
(12) Arhart, R. J.; Martin, J. C. J. Am. Chem. Soc. 1972, 94, 5003.
(13) The slight erosion in ee which was observed is likely due to Lewis
acid-mediated epimerization via an allylic carbocation.
Org. Lett., Vol. 10, No. 5, 2008
999