Stereoselective synthesis of optically active perhydrofuro[2,3-b]furan derivatives

Article abstract of DOI:10.1016/S0040-4039(01)00813-9

(1R,5S)-2,8-Dioxabicyclo[3.3.0]octan-3-one and its derivatives, important subunits in various biologically active natural products, have been synthesized based on a new approach using the asymmetric oxyselenenylation of 2,3-dihydrofuran as the key step.

Full text of DOI:10.1016/S0040-4039(01)00813-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 4653–4656
Stereoselective synthesis of optically active
perhydrofuro[2,3-b]furan derivatives
Masahiko Uchiyama,* Manabu Hirai, Momoko Nagata, Rui Katoh, Risa Ogawa and Akihiro Ohta
School of Pharmacy, Tokyo University of Pharmacy and Life Science, 1432-1 Horinouchi, Hachioji, Tokyo 192-0392, Japan
Received 3 April 2001; revised 19 April 2001; accepted 11 May 2001
Abstract—(1R,5S)-2,8-Dioxabicyclo[3.3.0]octan-3-one and its derivatives, important subunits in various biologically active natural
products, have been synthesized based on a new approach using the asymmetric oxyselenenylation of 2,3-dihydrofuran as the key
step. © 2001 Elsevier Science Ltd. All rights reserved.
of enantiomerically pure 1 using the asymmetric oxy-
selenenylation of 2 as the key step. Asymmetric synthe-
sis of chiral acetals have been generally based on con-
trolling the stereochemistry at acetal carbon by the
other stereocenter(s) in the molecule because direct
method of constructing chiral acetal center has not
been known to date and acetals easily isomerize under
acidic conditions. In contrast with the methods
reported before, the absolute configuration at the acetal
carbon is directly produced by asymmetric reaction in
our approach outlined in Scheme 1. We thought that
the asymmetric oxyselenenylation of 2 could introduce
not only the asymmetry at the acetal carbon but also a
Furo[2,3-b]furan structure is encountered in various
biologically active natural products, such as aflatoxins,1
dihydroclerodin,2,3 and norrisolide,4 Therefore, the syn-
thesis of such furofurans have attracted considerable
attention from organic chemists, and many synthetic
methods have been reported during the last decade5.
However, the method for preparation of enantiomeri-
cally pure furo[2,3-b]furan derivatives is not sufficient
yet. For instance, the synthesis of enantiomerically
enriched (1S,5R)-2,8-dioxabicyclo[3.3.0]octan-3-one (1),
which is considered to be a chiral building block for the
synthesis of dihydroclerodin, has been reported by only
two groups up to now.6
O
H
OAc
O
O
O
O
H
H
H
H
H
H
H
H
O
H
H
O
O
O
O
O
H
O
H
MeO
O
OAc
OAc
O
(1S ,5R)-1
aflatoxin B1
norrisolide
dihydroclerodin
seleno group available for subsequent cyclization and
the use of methyl glycolate as an oxygen nucleophile
gave an adduct suitable for subsequent transforma-
tions. In this paper, we describe the stereoselective
synthesis of enantiomerically pure 1 by a new approach
based on the asymmetric oxyselenenylation of 2.
Very recently, we have demonstrated that the asymmet-
ric methoxyselenenylation of alkyl vinyl ethers affords
the corresponding acetals or ketals with moderate to
good diastereoselectivity.7 For example, the reaction of
2,3-dihydrofuran (2) gave the anti-adduct in 87% yield
as a 74:26 diastereomeric mixture. This observation
prompted us to undertake the stereoselective synthesis
At first, we examined the asymmetric oxyselenenylation
of 2 with methyl glycolate according to the procedure
reported in our previous paper7,8 (Scheme 2). When the
reaction was carried out at −78°C, the desired anti-
adduct 49 was obtained in low yield as an inseparable
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)00813-9

4654
M. Uchiyama et al. / Tetrahedron Letters 42 (2001) 4653–4656
OH
H
H
SeAr*
CO2Me
H
H
H
H
O
O
O
O
O
O
O
O
O
(1S ,5R)-1
Ar*SeX
HOCH2CO2Me
+
O
2
Scheme 1.
78:22 diastereomeric mixture along with a substantial
amount of lactol 6.9,10 The absolute stereochemistry of
4 could not been determined at this stage. Because the
poor yield of 4 was probably due to the low reactivity
of methyl glycolate toward the seleniranium cation
intermediate which was formed by the addition of
Ar*SeBF4 to 2, in order to improve the yield of 4, we
tried to raise the reaction temperature as shown in
Table 1. Raising the reaction temperature from −78°C
to −50°C gave no improvement of the yield of 4 (entry
1), however, raising to −20°C fairly improved the yield
without the deterioration of the diastereomeric ratio
(dr) (entry 2). When the temperature was further raised
to 0°C, a slight decrease of the yield of 4 and the
formation of syn-adduct 59 was observed. The forma-
tion of 5 is thought to result from the nucleophilic
attack of methyl glycolate to an open oxocarbonium
cation rather than a bridged seleniranium cation. We
imagined that an unreacted seleniranium cation opened
to the oxocarbonium cation through an equilibrium
under the reaction conditions in entry 3. This assump-
tion was supported by the result in entry 4, and also by
the results of another experiments, namely, that trans-
adduct 4 did not isomerize to cis-adduct 5 under the
same reaction conditions as used in entry 3 and 4.
Although sufficient diastereoselectivity was not attained
and 4 was obtained as an inseparable mixture of two
diastereomers, we next set about the transformation of
anti-adduct 4 to target compound 1 as shown in
Scheme 3.9
The anti-adduct 4 (dr=78:22) was half-reduced with
DIBAL-H to give the corresponding aldehyde, which
was subsequently subjected to a radical cyclization by
treatment with Bu3SnH and AIBN to afford exo alco-
hol 7 and endo alcohol 8. exo Alcohol 7 could be
converted to the desired 8 in two steps. Optical resolu-
tion of 8 was carried out by use of (R)-1-(1-naph-
thyl)ethyl isocyanate11 as a resolving agent; 8 was
heated with the isocyanate in benzene at 60°C to afford
the corresponding carbamates 9 and 10, which were
separated by preparative TLC. The major diastereomer
NMe2
3) HOCH2CO2Me
-78oC
SeAr*
CO2Me
SeAr*
CO2Me
SeAr*
OH
1) SO2Cl2, CH2Cl2, rt
2) AgBF4, CH2Cl2, 0oC
Ar*SeBF4
+
+
, ToC
O
O
O
O
O
Se)2
≡ (Ar*Se)2
3
4)
O
4
5
6
2
Scheme 2.
Table 1. Asymmetric oxyselenenylation of 2,3-dihydrofuran using methyl glycilatea
Entry
Temp. (°C)
4 (anti-adducts)b
5 (syn-adducts)b
6c
Yield (%)d
dre,f,g
Yield (%)d
dre,f,h
Yield (%)d
1
2
−78“−50
−78“−20
−78“0
0
43
73
69
31
78:22
78:22
78:22
69:31
0
0
5
–
–
52
14
15
3
3
76:24
68:32
4i
21
a All reactions were performed using 6 equiv. of methyl glycolate to diselenide 3. A solution of AgBF4 in acetonitrile was used.
b Obtained as an inseparable mixture of two diastereomers.
c Obtained as an inseparable mixture of four diastereomers.
d Isolated yield.
e Diastereomeric ratio.
f Determined by 1H NMR integration.
g The absolute stereostructure of the major diastereomer was determined to be 4 depicted in Scheme 2 by subsequent transformation to 8.
h The absolute stereostructure of the major diastereomer was supposed to be 5 depicted in Scheme 2 in consideration for the mechanism of the
reaction.
i Methyl glycolate and 2,3-dihydrofuran were successively added at 0°C.

M. Uchiyama et al. / Tetrahedron Letters 42 (2001) 4653–4656
4655
Scheme 3. Reagents and conditions: (a) DIBAL-H, toluene, −78°C; (b) Bu3SnH, AIBN, benzene, reflux, 7: 13% from 4, 8: 44%
from 4; (c) PCC, CH2Cl2, rt; (d) NaBH4, EtOH, −15°C, 69% from 7; (e) (R)-1-(1-naphthyl)ethyl isocyanate, N,N-dimethyl-
aminoethanol, benzene, 60°C, 9: 73%, 10: 22%; (f) LiAlH4, THF, rt, (−)-8: 86% from 9, (+)-8: 78% from 10; (g) Ph3P, imidazole,
,
I2, benzene, 60°C, 87%; (h) DBU, rt ; (i) THF, 40% H2SO4, rt, 72% from 11; (e) PCC, NaOAc, MS 4 A, CH2Cl2, rt, 77%.
of dihydroclerodin are currently underway in our labor-
atory.
9 and minor 10 were separately treated with LiAlH4 to
provide optically pure (−)-812 and (+)-8,12 respectively.
Thus, the absolute stereochemistry of the major
diastereomer obtained from the reaction in Scheme 2
was confirmed at this stage. Unfortunately, the major
product (−)-8 corresponded with (1R,5S) antipode
instead of our target (1S,5R)-1; however, we proceeded
in the transformation of (−)-8 into (1R,5S)-1 in order
to demonstrate the synthetic route leading to optically
active 1.
References
1. For a review, see: Schuda, P. F. Top. Curr. Chem. 1980,
91, 75–111.
2. For reviews on clerodanes, see: (a) van Beeck, T. A.; de
Groot, Ae. Recl. Trav. Chim. Pays-Bas 1986, 105, 513–
527; (b) Merritt, A. T.; Ley, S. V. Nat. Prod. Rep. 1992,
243–287.
(−)-8 was converted to exo iodide 11, the iodine atom
of which was in the suitable orientation for controlling
the regiochemistry in the next elimination step. Treat-
ment of 11 with DBU afforded the 2,3-dihydrofuran
intermediate, which was hydrated without isolation
because of its volatility to provide lactol 12 as an
inseparable mixture of epimers. Oxidation of 12 with
PCC gave (1R,5S)-1,6a,13 the enantiomeric purity of
which was determined to be >98% ee by 1H NMR
analysis upon comparison with racemic 114 using (R)-
(−)-2,2,2-trifluoro-1-(anthryl)ethanol as the chiral sol-
vating reagent.15 It should be also mentioned that the
optical purity of our synthetic (1R,5S)-1 is higher than
those reported in the literature.6a,13
3. (a) Barton, D. H. R.; Cheung, H. T.; Cross, A. D.;
Jackman, L. M.; Martin-Smith, M. J. Chem. Soc. 1961,
5061–5073; (b) Hosozawa, S.; Kato, N.; Munakata, K.
Phytochemistry 1973, 12, 1833–1834.
4. (a) Hochlowski, J.; Faulkner, D. J.; Matsumoto, G.;
Clardy, J. J. Org. Chem. 1983, 48, 1141–1142; (b) Tak-
izawa, P. A.; Yucel, J. K.; Veit, B.; Faulkner, D. J.;
Deerinck, T.; Soto, G.; Ellisman, M.; Malhotra, V. Cell
1993, 73, 1079–1090.
5. Vader, J.; Sengers, H.; de Groot, A. Tetrahedron 1989,
45, 2131–2142 and references cited therein.
6. (a) Petit, F.; Furstoss, R. Synthesis 1995, 1517–1520; (b)
Enders, D.; Va´zquez, J.; Raabe, G. J. Chem. Soc., Chem.
Commun. 1999, 701–702.
7. Uchiyama, M.; Satoh, S.; Ohta, A. Tetrahedron Lett.
In conclusion, we have accomplished the synthesis of
enantiomerically pure (1R,5S)-1 and presented herein a
new approach to optically active perhydrofuro[2,3-b]-
furan derivatives using the asymmetric oxyselenenyla-
tion of 2,3-dihydrofuran (2). Although the diastereo-
selectivity of the key reaction shown in Scheme 2 is not
so high and the obtained (1R,5S)-1 is the antipode of
our original target, it is supposed that these problems
could be solved by optimization of the starting di-
selenide. Further studies directed towards an improve-
ment of the diastereoselectivity and the total synthesis
2001, 42, 1559–1562.
8. Uchiyama, M.; Oka, M.; Harai, S.; Ohta, A. Tetrahedron
Lett. 2001, 42, 1931–1934.
9. All new compounds were characterized by 1H, 13C, IR,
and mass spectrometric analysis.
10. Presumably, unreacted seleniranium cation and/or oxo-
carbonium cation reacted with H2O, which was added as
an aqueous solution of NaHCO3 in quenching the reac-
tion, to give lactol 6.
11. Pirkle, W. H.; Hoekstra, M. S. J. Org. Chem. 1974, 39,
3904–3906.

4656
M. Uchiyama et al. / Tetrahedron Letters 42 (2001) 4653–4656
12. (a) Ghosh, A. K.; Thompson, W. J.; Fitzgerald, P. M.
D.; Culberson, J. C.; Axel, M. G.; McKee, S. P.; Huff,
J. R.; Anderson, P. S. J. Med. Chem. 1994, 37, 2506–
2508; (b) Ghosh, A. K.; Chen, Y. Tetrahedron Lett.
1995, 36, 505–508; (c) Chen, X.; Li, L.; Kempf, D. J.;
Sham, H.; Wideburg, N. E.; Saldivar, A.; Vasa-
vanonda, S.; Marsh, K. C.; McDonald, E.; Norbeck,
D. W. Bioorg. Med. Chem. Lett. 1996, 6, 2847–
2852.
13. Petit, F.; Furstoss, R. Tetrahedron: Asymmetry 1993, 4,
1341–1352.
14. Racemic
1 was prepared according to the method
shown in Scheme 3 using PhSeCl as the starting mate-
rial. Also see: Jalali, M.; Boussac, G.; Lallemand, J.-Y.
Tetrahedron Lett. 1983, 24, 4307–4310.
15. (a) Pirkle, W. H.; Sikkenga, D. L.; Pavlin, M. S. J.
Org. Chem. 1977, 42, 384–387; (b) Pirkle, W. H.;
Sikkenga, D. L. J. Org. Chem. 1977, 42, 1370–1374.
.