differentiation from the malonate permitted selective acid
formation at C40 in the triester 3. Ultimately, reduction of
the derived C40 acid could take place concurrently with the
C30 ester in 4, maximizing overall step efficiency.
Mechanistic studies for the [3+2] cycloaddition have
provided evidence for a stereospecific nucleophilic substitution
mechanism wherein the aldehyde acts as a nucleophile toward
a configurationally stable intimate ion pair in a process that
proceeds with inversion at the cyclopropane donor site;13
however, the conversion of 1 - 3 in the present study
proceeds with retention at C3 of the cyclopropane. How
then does 3 arise? To answer this question, we examined
the chemistry of tetrahydrofuran 5 which arises from a
moderate-yielding [3+2] cycloaddition of 1 and 2 using SnCl4
as the Lewis acid. The substituents at C40 and C50 of this
diastereomer possess the cis relationship that would be
expected from the nucleophilic substitution mechanism. Upon
exposure to AlCl3 in CH2Cl2, tetrahydrofuran 5 was converted
to 3 (C40/C50-trans) along with piperonal and cyclopropane-
derived decomposition products (Scheme 2). If 5 is formed
initially in the AlCl3-catalyzed cycloaddition it is conceivable
that further association with the AlCl3 isomerizes the product
to the observed diastereomer. A possible mechanism involves
AlCl3 binding with the ether of the tetrahydrofuran ring
causing reversible ring cleavage at both C20 and C50. The
resultant carbenium ions 6 and 7 are stabilized by the strongly
electron releasing aryl groups (Scheme 2) and isomerization of
this type is precedented.9,13,16–19 The appearance of piperonal
(2) but not veratraldehyde in the reaction of 5 and AlCl3
may be understood in terms of the relative facility of retro-
aldolization of intermediates 6 and 7. It is apparent that C40 is
the only non-epimerizable stereocenter and acts as the
keystone that regulates the C20/C50 stereochemical outcome
via what is apparently thermodynamic control. Our data do
not exclude pre-cycloaddition isomerization of trans-1 to cis-1
and subsequent stereospecific cycloaddition via the established
nucleophilic substitution pathway; however, efforts to induce
such an isomerization have yielded no evidence of cis-1.
The stability of the C40 stereocenter was confirmed in a
subsequent asymmetric synthesis of (+)-virgatusin. The
Scheme 3 (+)-Virgatusin synthesis.
requisite preparation of enantioenriched cyclopropane 1 is
detailed in Scheme 3. An organocatalytic Michael addition–
intramolecular alkylation20 between enal 8 and bromomalonate
9 gave the formyl cyclopropane 10 (er 90 : 10). Oxidation20
and nucleophilic esterification gave (2R,3S)-1. Cycloaddition
with 2 was performed as with the racemate and a single
recrystallization gave (2S,4S,5S)-3 with an er of 99 : 1.
The synthesis was completed as detailed for the racemate in
Scheme 1 to give (+)-virgatusin.
In conclusion, we have shown that the [3+2] cyclopropane–
aldehyde cycloaddition previously developed in our laboratory
can be used to synthesize more complex tetrahydrofuran
derivatives such as virgatusin. This synthesis is straight-
forward and should be amenable to other members of the
furanolignan family of natural products.
Funding for this work was provided by the NSF (CHE-
0749691) and Novartis (Early Career Award to J.S.J.). S.D.S.
acknowledges
a National Physical Science Consortium
Fellowship. The authors thank Professor Stephen Marsden
for providing an authentic synthetic sample of (+)-virgatusin.
Notes and references
1 C.-C. Chen, Y-L Huang, F-L Hsu and C-F Chen, J. Nat. Prod.,
1996, 59, 520–521.
2 K. Akiyama, S. Yamauchi, T. Nakato, M. Maruyama,
T. Sugahara and T. Kishida, Biosci., Biotechnol., Biochem., 2007,
71, 1028–1035.
3 M. Maruyama, S. Yamauchi, K. Akiyama, T. Sugahara,
T. Kishida and Y. Koba, Biosci., Biotechnol., Biochem., 2007, 71,
677–680.
4 T. Nakato, R. Tago, K. Akiyama, M. Maruyama, T. Sugahara,
T. Kishida and S. Yamauchi, Biosci., Biotechnol., Biochem., 2008,
72, 197–203.
5 S. Yamauchi, T. Nakato, M. Tsuchiya, K. Akiyama,
M. Maruyama, T. Sugahara and T. Kishida, Biosci., Biotechnol.,
Biochem., 2007, 71, 2248–2255.
6 H. Yoda, M. Mizutani and K. Takabe, Tetrahedron Lett., 1999, 40,
4701–4702.
7 S. Yamauchi, M. Okazaki, K. Akiyama, T. Sugahara, T. Kishida
and T. Kashiwagi, Org. Biomol. Chem., 2005, 3, 1670–1675.
8 K. Matcha and S. Ghosh, Tetrahedron Lett., 2008, 49, 3433–3436.
9 T. Akindele, S. P. Marsden and J. G. Cumming, Org. Lett., 2005,
7, 3685–3688.
´
10 S. Martinet, A. Meou and P. Brun, Eur. J. Org. Chem., 2009,
2306–2311.
11 P. D. Pohlhaus and J. S. Johnson, J. Org. Chem., 2005, 70,
1057–1059.
12 P. D. Pohlhaus and J. S. Johnson, J. Am. Chem. Soc., 2005, 127,
16014–16015.
Scheme 2 Isomerization mechanism.
ꢂc
This journal is The Royal Society of Chemistry 2009
5136 | Chem. Commun., 2009, 5135–5137