A Short Synthesis of (±)-Epiasarinin

Article abstract of DOI:10.1021/ol025569e

matrix presented Epiasarinin, an endo-endo furofuran, has been synthesized from piperonal via a five-step route with good stereocontrol. The sequence involves Darzens condensation, alkenyl epoxide-dihydrofuran rearrangement, and a Lewis acid mediated cyclization.

Full text of DOI:10.1021/ol025569e

ORGANIC
LETTERS
2
002
Vol. 4, No. 7
159-1162
A Short Synthesis of (±)-Epiasarinin
1
†
‡
,‡
David J. Aldous, Anne J. Dalen c¸ on, and Patrick G. Steel*
Department of Chemistry, UniVersity of Durham, South Road, Durham,
DH1 3LE, UK, and AVentis Pharma, Inc., Route 202-206, Bridgewater,
P.O. Box 6800, New Jersey 08807
p.g.steel@durham.ac.uk
Received January 16, 2002
ABSTRACT
Epiasarinin, an endo-endo furofuran, has been synthesized from piperonal via a five-step route with good stereocontrol. The sequence involves
Darzens condensation, alkenyl epoxide−dihydrofuran rearrangement, and a Lewis acid mediated cyclization.
Compounds containing the 2,6-diaryl-3,7-dioxabicyclo[3,3,0]-
octane skeleton (furofurans) belong to the lignan family of
natural products. There is considerable structural variation
in this series, in both the nature and the stereochemistry of
the aryl substituents, e.g., Epiasarinin, Praderin, or Kobusin,
In our model studies, we had generated the alkenyl epoxide
8 through a Wadsworth-Emmons reaction of epoxide 7.
1
Figure 1.
Due to their large range of biological properties, which
include antioxidant, anticancer, antiviral, and immunosup-
pressant activities, these natural products represent very
attractive synthetic targets. Since activity is dependent on
stereochemistry, synthetic routes that can provide controlled
but tuneable access to one particular class are very attractive.
In this respect, we have been interested in developing
efficient ways to synthesize the furofuran skeleton. Recently,
we have reported a sequence based on the thermal rear-
rangement of alkenyl epoxides to dihydrofurans and the
subsequent Lewis acid catalyzed cyclization to generate the
2
furofuran nucleus with good stereocontrol, Scheme 1. While
many syntheses of Sesamin (exo-exo furofuran 3) have been
published, there have been no accounts of approaches to the
less stable endo-endo furofuran diastereoisomer.3 In this
paper, we report an efficient stereoselective approach to a
-6
7
natural endo-endo furofuran represented by Epiasarinin.
†
Aventis Pharma, Inc.
University of Durham.
‡
(
(
(
1) Ward, R. S. Nat. Prod. Rep. 1999, 16, 75-96.
2) Aldous, D. J.; Dutton, W. M.; Steel, P. G. Synlett 1999, 474-476.
3) Pelter, A.; Ward, R. S.; Watson, D. J.; Jack, I. R. J. Chem. Soc.,
Perkin Trans. 1 1982, 183-190.
4) Suginome, H.; Orito, K.; Yorita, K.; Ishikawa, M.; Shimoyama, N.;
Sasaki, T. J. Org. Chem. 1995, 60, 3052-3064.
(
Figure 1. Typical natural furofuran lignans.
1
0.1021/ol025569e CCC: $22.00 © 2002 American Chemical Society
Published on Web 03/14/2002

developed. The optimized conditions involved addition of
Scheme 1. Model Studiesa
LDA at -20 °C to a solution of bromocrotonate (1 equiv)
and piperonal (4 equiv). After a mild acidic quench (NH -
4
Cl), the excess of piperonal was scavenged as the bisulfite
addition product and the resulting oil was purified by flash
chromatography (1:3 ether/petrol). The desired alkenyl
epoxide was obtained in 70% yield as a single alkene isomer
(
trans) and a 3:4 syn/anti mixture of epoxides. Since the
stereochemical information is lost in the alkenyl epoxide-
dihydrofuran rearrangement, this mixture was used directly
in the next step with no further purification.
In our preliminary studies, alkenyl epoxide-dihydrofuran
rearrangements were usually performed in an autoclave or a
sealed tube, involving high temperatures for prolonged times
(
e.g., 180 °C for 20 h for the phenyl equivalent 8). However,
with ester 13, only traces of the desired piperonyl analogue
6 could be detected in the intractable crude product and
1
a
t
Conditions: (i) BuOOH, NaOH, 73%; (ii) (EtO)
2
P(O)CH
2
CO
2
Et,
we suspected either the product or the starting material to
be thermally unstable under these conditions. Therefore, the
rearrangement was carried out using flash vacuum pyrolysis
NaH, PhMe, 68%; (iii) sealed tube, PhMe, 180 °C, 70%; (iv)
LiAlH , Et O, 0 °C, 100%; (v) ArCH(OMe) , TMSOTf, DCM, -20
C, 81%.
4
2
2
°
(FVP) at 500 °C and 0.04 mbar. This method afforded the
cis:trans dihydrofuryl ester 16 as a mixture of isomers (8:
1). This ratio was independent of the stereochemistry of the
starting epoxide and represents preferential disrotary ring
closure of the less hindered ylid intermediate to afford the
However, attempts to generate the corresponding piperonyl-
derived epoxide 15 were not successful.
This problem was attributed to the electron-donating effect
of the substituents on the aryl group promoting epoxide ring
opening reactions. Since, despite considerable experimenta-
tion, we could not circumvent this difficulty, we sought an
alternative route to the vinyl epoxide 13. Recognizing that
the Darzens condensation of bromocrotonate 11 was a
possible method for preparing the vinyl epoxide 13, we
9
,10
desired cis isomer, Scheme 3.
Following this protocol,
Scheme 3. Alkenyl Epoxide-Dihydrofuran Rearrangement
8
undertook a study of this transformation, Scheme 2.
Scheme 2. Darzens Condensation
we isolated pure cis dihydrofuryl ester 16c in 66% yield after
flash column chromatography.
The reduction of the cis dihydrofuryl ester 16c into the
alcohol was performed with lithium aluminum hydride at
Although poor conversions and extensive decomposition
complicated our initial efforts, an efficient method was
-40 °C. The dihydrofuryl alcohol 17 could not be stored
for more than a couple of hours at low temperatures under
an inert atmosphere and, consequently, was used in the next
step with no further purification (purity > 90% by NMR).
The alcohol 17 was then combined with piperonal dimethyl
acetal 18a in the presence of a Lewis acid (trimethylsilyl
triflate) to generate the oxonium ion 19, Scheme 4. Cycliza-
(
5) Hull, H. M.; Jones, R. G.; Knight, D. W. J. Chem. Soc., Perkin Trans.
1998, 1779-1787.
6) Rana, K. K.; Guin, C.; Roy, S. C. Tetrahedron Lett. 2000, 41, 9337-
338.
7) Hearon, W. M.; MacGregor, W. S. Chem. ReV. 1955, 55, 991-1001.
1
(
9
(
A recent report (ref 14) has outlined the first selective synthesis of an endo-
exo furofuran. In addition, a synthetic route to diaxial (endo-endo) 2,4-
diarylfurofurans has been reported; see: Pelter, A.; Ward, R. S.; Collins,
P.; Venkateswarlu, R.; Kamakshi, C. Tetrahedron Lett. 1992, 33, 4361-
(9) Eberbach, W.; Burchard, B. Chem. Ber. 1978, 111, 3665-3698.
(10) Eberbach, W.; Seiler, W.; Fritz, H. Chem. Ber. 1980, 113, 875-
901.
4
364.
8) Koppel, G. A. Tetrahedron Lett. 1972, 15, 1507-1509.
(
1160
Org. Lett., Vol. 4, No. 7, 2002

Scheme 4. Lewis Acid Catalyzed Cyclization
Table 1. Ratio of Reduction Products Acetal 21a:Epiasarinin
1
:Asarinin 2:Sesamin 3a
temp
time
(h)
(°C)
21a
1
-
-
-
-
0
78
40
40
78 to rt
to rt
4
4
15
4
100
90
10
0
0
10
67
0
48
0b
0b
a
Percentage determined by H NMR. b Decomposition.
1
exo 3), being detected in different ratios, Table 1. The
epimerization can be rationalized as Lewis acid promoted
ring fragmentation and leads to the most stable exo-exo
furofuran, Scheme 5.
Scheme 5 Reduction of the Methyl Acetal
tion then occurred to generate the second oxonium ion 20,
which was quenched, by a molecule of methanol derived
from the initial acetal, to provide the methyl furofuryl acetal
2
1a. Attempts to use piperonal directly in this process failed,
and while the more stable and convenient dioxolane acetal
8b was a viable substrate, the resultant hydroxyethyl
1
glycoside 21b was relatively unstable and we were not able
to isolate it in sufficient purity or yield. As in our preliminary
report, the endo isomer was exclusively generated at low
temperatures (kinetic product), whereas the exo isomer
After some experimentation, the conditions used to reduce
the furofuryl acetal 21a with a good conversion and a small
rate of epimerization were Et
1.1 equiv) at -40 °C for 15 h. This produced a mixture of
starting acetal 21a, Epiasarinin 1, and Asarinin 2 in a 10:
7:23 ratio, Scheme 5. After a difficult separation by flash
3
SiH (10 equiv) and BF
3
‚OEt
2
(
(thermodynamic) was favored when the reaction mixture was
allowed to warm to room temperature for 48 h. To have
access to the Epiasarinin skeleton (endo), the optimized
conditions involved reaction at -40 °C overnight with, to
avoid a mixture of furofuryl acetals, a methanol quench.
Following this procedure, we obtained the endo methyl
furofuryl acetal 21a after purification by flash chromatog-
raphy (4:1 petrol/ether) in 55% yield as a single diastereo-
isomer. To complete the synthesis of Epiasarinin, reduction
of the methyl furofuryl acetal was required. Related trans-
formations have been described in the literature.11,12 How-
6
chromatography (silica gel, 3:1:0.1 petrol/ether/triethy-
lamine), pure samples of Epiasarinin 1 and Asarinin 2 were
obtained. The structure of Epiasarinin was confirmed by
detailed spectroscopic analysis (see Supporting Information).
In particular, the NMR data reflected a symmetrical com-
pound, that was clearly different from that reported for
4,5
Sesamin 3. The sample of Asarinin was identical in all
13,14
respects to that reported in the literature.
Attempts to
enhance this reduction using different Lewis acids, TiCl
Sc(OTf) , or more powerful hydride sources, e.g., Cl SiH/
BF ‚OEt or DIBAL, proved not to be viable.
4
,
ever, using these conditions (Et
3
SiH, BF
3
2
‚OEt , DCM, 48
3
3
h), extensive decomposition resulted.
3
2
Following a detailed study of these conditions, it was found
that the stereochemistry of the diaryl furofuran product
depended on the temperature, with Epiasarinin and its two
diastereoisomers, Asarinin (exo-endo 2) and Sesamin (exo-
In conclusion, a very concise stereoselective synthesis of
the endo-endo furofuran Epiasaranin has been achieved for
the first time (five steps, 20% overall yield). The route is
amenable to considerable variation, and the preparation of
(
11) Pelter, A.; Ward, R. S.; Collins, P.; Venkateswarlu, R.; Kay, I. T.
J. Chem. Soc., Perkin Trans. 1 1985, 587-594.
12) Pelter, A.; Ward, R. S.; Venkateswarlu, R.; Kamakshi, C. Tetra-
(13) Pelter, A.; Ward, R. S. Tetrahedron 1976, 32, 2783-2788.
(14) Brown, R. C. D.; Bataille, C. J.; Bataille, C. J. R.; Bruton, G.; Hinks,
J. D.; Swain, N. A. J. Org. Chem. 2001, 66, 6719-6728.
(
hedron 1992, 48, 7209-7220.
Org. Lett., Vol. 4, No. 7, 2002
1161

structural and stereochemical analogues is in progress and
will be reported in due course.
Supporting Information Available: Experimental sec-
tion containing procedures and characterization of all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
Acknowledgment. We thank the EPSRC and Aventis for
financial support of this work (CASE award to A.J.D.), Dr.
A. M. Kenwright for assistance with NMR experiments, and
Dr. M. Jones for mass spectra.
OL025569E
1162
Org. Lett., Vol. 4, No. 7, 2002