(R2 ) n-Pr) to give tetrahydroindenofuranone 4c (50% yield)
(Table 1, entry 3).
The relative configuration of (3aR*,8R*,8aS*)-4a and
(3aR*,8R*,8aS*)-4c was deduced after transformation in known
compounds. Specifically, hydrolysis of the ester moiety with
LiOH/DME and subsequent decarboxylation performed in
toluene at 100 °C with a catalytic amount of AcOH gave,
respectively, the tetrahydroindenofuranones (3aR*,8S*,8aR*)-
6a and (3aR*,8R*,8aR*)-6c whose spectral data fitted quite well
those reported by Muraoka and Momose (Scheme 2).10 By
To extend the scope of the above domino process, we
briefly investigated the substitution effect in the starting
oxirane. Monosubstituted oxirane 1b-Li, bearing a lithiated
electron-rich aryl moiety (Table 1, entry 4), reacts with 5b
affording the expected product 4d in moderate yield (50%).
In contrast, trans-disubstituted oxiranes 1c (R1 ) Me) and
1d (R1 ) Ph) react with aromatic and aliphatic malonates
5a and 5c providing, after stirring the reaction mixture
overnight at room temperature, the uncyclized Michael
adducts 2e-g (Scheme 1). The substitution of oxiranes 1c,d
could be likely responsible, for steric reasons, for the
unsuccessful intramolecular cyclization.
Scheme 2
With the Michael adducts 2e-g in hand, we checked the
suitable experimental conditions for their cyclization to the
corresponding tetrahydroindenofuranones. To this end, we
first investigated the role of the cation in the ring-opening
reaction: a solution of 2f in THF was treated at room
temperature with t-BuOK, but after 12 h stirring, only starting
material was recovered. Similarly, treatment of 2f with NaH
in DMF did not furnish any expected product. After several
attempts, we found that using EtOLi, EtONa, or t-BuOK in
EtOH or t-BuOH at room temperature, tetrahydroindeno-
furanone 4f was obtained as the sole diastereomer in a good
yield (70% yield in the case of the t-BuOK/t-BuOH system,
Table 1, entry 6).7 The same methodology was used to obtain
products 4e,g from the corresponding oxiranes 2e,g.
We also checked the possibility of making the tetrahy-
droindenofuranone 4f in one-pot simply by adding, to the
ethereal reaction medium after completion of the Michael
addition, a polar protic solvent such as EtOH. After stirring
the reaction mixture for 3 days at room temperature, we
obtained, after workup and purification by column chroma-
tography on silica gel, the desired tetrahydroindenofuranone
4f in 50% yield (dr > 98/2).8
performing the hydrolysis-decarboxylation sequence on the
second diastereomer (3aR*,8S*,8aS*)-4a, formed in the reaction
of 1a-Li with 5a, we obtained the tetrahydroindenofuranone
(3aR*,8R*,8aR*)-6a whose relative stereochemistry was as-
signed on the basis of the coupling constant value between the
benzylic proton Ha and the vicinal proton Hb in the 1H NMR,
3
the JHa-Hb (9.9 Hz) being diagnostic of a dihedral angle
Moreover, it is worth noting that the stereoselectivity of
the conjugate addition of ortho-lithiated trans-1,2-disubsti-
tuted oxiranes 1c,d to malonates was strongly improved in
the case of adduct 4f obtained as the sole diastereomer (Table
1, entry 6).9
H8-C8-C8a-H8a near zero.11,12
This observation is in agreement with the nearly absent
coupling (3JHa-Hb ≈ 0 Hz) in the epimeric compound
(3aR*,8S*,8aR*)-6a whose dihedral angle H8-C8-C8a-H8a
value is nearly 90°.
The relative configuration of tetrahydroindenofuranones
(6) The Michael addition of aryllithiums to R,ꢀ-unsaturated malonates
was reported by Kende and coworkers, as a key step in the racemic synthesis
of an epipodophyllotoxin derivative. Such a methodology was employed
in the construction of the tetrahydronaphthalene core starting from an ortho-
bromobenzylether: Kende, A. S.; King, M. L.; Curran, D. P. J. Org. Chem.
1981, 46, 2826–2828.
1
4b,d-f was achieved by H NMR analysis on the basis of
the dibenzylic H8 chemical shift analogy with the reference
product 4a. Indeed, we noted that in the diastereoisomers
bearing the ethoxycarbonyl group and the H8 on the same
side, the resonance of the latter was downfield shifted of
about 0.4-0.5 ppm with respect to the corresponding
epimers, most probably because of the anisotropic effect of
the carbonyl ester group. To support this assumption, we
noted that both epimers of tetrahydroindenofuranone 6a, in
which the carbonyl anisotropic effect is missing, show a
(7) After treating 2e-g with t-BuOK, compounds 4e-g were isolated
as a mixture of ester and the corresponding carboxylic acid, because of a
partial hydrolysis of the lactone and/or ethoxycarbonyl groups. Therefore,
it was necessary to treat the crude with EtOH/H2SO4(cat) in order to obtain
exclusively the desired indenofuranones 4e-g (for details, see Supporting
Information).
(8) The one-pot procedure was also employed for the synthesis of 4g
obtained in 50% yield (dr 80/20) after stirring the reaction mixture for one
day.
(9) Starting from cis-1d, we thought that it would be possible to make
epimeric tetrahydroindenofuranone diast-4f with the stereogenic center C3
“R”sinstead of “S”sconfigured, as previously synthesized starting from
trans-1d. All the attempts carried out by using ortho-lithiated cis-1d resulted
in a very complex mixture of products containing only traces of a 60/40
diastereomeric mixture of diast-4f. After column cromatography, it was
possible to isolate and characterize only the major diastereoisomer (for
details, see Supporting Information).
(10) Muraoka, O.; Tanabe, G.; Kyohko, S.; Minematsu, T.; Momose,
T. J. Chem. Soc., Perkin Trans. 1 1994, 1833–1845.
(11) For all tetrahydroindenofuranones 4a-g and 6a-c, it is reasonable
to assume that, for thermodynamic reasons, the γ-lactone ring may be fused
to the indane system only in a cis fashion
(12) Semiempirical calculations on the substituted indenofuranone
systems confirm such values of dihedral angles
.
.
Org. Lett., Vol. 10, No. 10, 2008
1949