Figure 3. Synthesis of papayanol (3) and its tosylate (4).
Figure 2. 70 eV mass spectrum of papayanol (3).
had to be assigned based on mass spectrometric data and
microreactions. The mass spectrum (Figure 2) showed a base
peak at m/z 139 and a general fragmentation pattern
resembling that of a terpene. The highest visible signal at
m/z 155 indicated a possible molecular mass of M ) 170:
the signal at m/z 139 would then represent (M+-hydroxym-
ethyl) or (M+-methoxy), while the one at m/z 155 would be
formed upon loss of a methyl group from the molecular ion.
The presence of a hydroxymethyl group was supported
by the fact that the product could be trifluoroacetylated.3
Furthermore, microhydrogenation or reaction with lithium
aluminum tetrahydride3 left the molecule unchanged, which
excluded the presence of a C-C double bond or a carbonyl
group. The high intensity of m/z 139 indicated the formation
of an oxonium ion,4,5 which suggested the presence of a
second oxygen in the molecule. Therefore, a dioxygenated
monoterpene with a molecular formula of M ) C10H18O2
was the most logical, and because of the two degrees of
unsaturation, the compound was postulated to be bicyclic.
Taking into account that the two earlier identified compo-
nents of the pheromone bouquet were grandisal and grand-
isol,2 it was reasonable to believe the target compound to
be structurally related to the same terpene skeleton. Conse-
quently, a reasonable candidate seemed to be 2-hydroxym-
ethyl-2,6-dimethyl-3-oxabicyclo[4.2.0]octane, which would
be formed after epoxidation of the 1-methylethenyl group
of grandisol followed by opening of the epoxide intermediate
upon intramolecular attack of the hydroxyl group. Epoxides
are well-known biogenetic intermediates in the formation of
pheromones, and the process outlined here strongly resembles
that postulated for the formation of pityol from sulcatol.6
In a kind of biomimetic approach, racemic grandisol was
reacted with m-chloroperbenzoic acid (Figure 3). The
expected epoxide could not be isolated, as the target com-
pound had already been spontaneously formed from the
intermediate under the employed reaction conditions.
Figure 4. Enantioselective separation of papayanol (3).
The reaction yielded two (racemic) diastereomers, which
were well separated by gas chromatography on a conven-
tional column. The mixture was strongly biased by the later
eluting racemate, and the natural extract showed the same
relations: a very minor peak and a highly dominating one.
Both natural products showed the same mass spectra and
the same retention times (coinjection) as the synthetic
compounds.
Upon enantioselective gas chromatography, using a modi-
fied ꢀ-cyclodextrin as the stationary phase,7 the two pairs
of synthetic enantiomers, obtained from racemic grandisol,
were well separated. Using (1R,2S)-(+)-grandisol as the
starting material8 revealed the earlier eluting main compound
and its later eluting minor stereoisomer to show (1R,6R)-
configuration. Under the same conditions, the major pair of
enantiomers was baseline separated, showing an R-value
tr(1S,6S):tr(1R,6R) ) 1.19. Comparison of the retention times of
synthetic and natural products (coinjection) revealed the latter
to show (1R,6R)-configuration (Figure 4).
Finally, the absolute configuration at C2 of the bicyclus
was assigned according to results of NMR experiments using
(3) Attygalle, A. B. In Methods in Chemical Ecology, 1; Millar, J. G.,
Haynes, K. E., Eds.; Kluwer-Academic Publishers: Norwell, 1998; pp
207-294.
(4) Francke, W.; Schro¨der, W. Curr. Org. Chem. 1999, 3, 407
(5) Francke, W.; Kitching, W. Curr. Org. Chem. 2001, 5, 233
.
.
(7) Heptakis-(2,6-di-O-methyl-3-O-pentyl)-ꢀ-cyclodextrin: Ko¨nig, W. A.;
Gehrcke, B.; Icheln, D.; Evers, P.; Do¨nnecke, J.; Wang, W. J. High. Res.
Chromatogr. 1992, 15, 367.
(6) Francke, W.; Schulz, S. Pheromones In ComprehensiVe Natural
Products Chemistry; Barton, D., Nakanishi, K., Eds.; Elsevier: Amsterdam,
1999; Vol. 8, pp 197-261.
(8) Mori, K.; Miyake, M. Tetrahedron 1987, 43, 2229.
2448
Org. Lett., Vol. 12, No. 11, 2010