Simple circular dichroic method for the determination of absolute configuration of 5-substituted 2(5H)-furanones

Full text of DOI:10.1021/jo951400l

J . Org. Chem. 1996, 61, 1513-1515
1513
Sim p le Cir cu la r Dich r oic Meth od for th e
Deter m in a tion of Absolu te Con figu r a tion
of 5-Su bstitu ted 2(5H)-F u r a n on es
while 5-(acyloxy)-2(5H)-furanones were resolved via li-
pase-catalyzed transesterification reactions.10
Optically active butenolides can also be prepared by
diastereoselective addition of a chiral tin(II) enolate to
γ-hydroxybutenolides,11 and γ-C-glycosylated butenolides
are available from aldehydosugars through condensation
with 2-(trimethylsiloxy)furan.12 The 2-(trimethylsiloxy)-
J acek K. Gawronski,*,† Arjan van Oeveren,
Hanncke van der Deen, Chiu W. Leung, and
Ben L. Feringa*
furan-butenolide pathway was employed in the synthe-
Department of Chemistry, A. Mickiewicz University,
13
sis of 6-deoxy-6,6,6-trifluorosugars.
Optically active
6
0-780 Pozna n˜ , Poland, and Department of Organic and
Molecular Inorganic Chemistry, Groningen Centre for
Catalysis and Synthesis, University of Groningen,
Nijenborgh 4, 9747 AG Groningen, the Netherlands
butenolides are furthermore available from sugar lac-
tones14 and from chiral propargylic alcohols via ruthenium-
catalyzed Alder ene reactions.15
γ-Alkyl- and γ-alkoxybutenolide structures are fre-
Received J uly 28, 1995
quently found in several biologically active compounds,
e.g., strigol16 and marine natural products.
17,18
In addi-
Chiral 2(5H)-furanones (butenolides) have emerged as
important synthetic intermediates for which several
general synthetic protocols have been devised.1 In these
compounds (A), the stereogenic center is most frequently
located at C(5).
tion, chiral 5-methyl-substituted 2(5H)-furanone rings
,2
are structural components of numerous natural products,
1
9
20
examples of which are acetogenins, muconolactones,
and leptosphaerin.21 Compounds of these classes are
frequently unsuitable for X-ray crystallographic studies
due to their waxy nature; hence, in such instances the
absolute configuration of the butenolide fragment may
not be known.
The need for a rapid and universal method for the
determination of the absolute configuration of synthetic
butenolides is obvious because of their frequent use as
intermediates in asymmetric synthesis.22 So far, chemi-
cal correlations have been necessary to determine the
absolute configuration.23
Hanessian3 extensively used (S)-5-(hydroxymethyl)-
(5H)-furanone (A, R ) CH OH) derived from (S)-
2
glutamic acid in the synthesis of polypropionates and
polyols and applied it in cycloadditions and conjugate
addition reactions. Alternative routes to optially active
butenolides involve D-ribonolactone, D-mannitol, or levo-
glucosenone as starting materials.4 Optically active
2
Despite the importance of chiral butenolides in syn-
thesis and in natural products chemistry the chiroptical
24
data for these compounds are scarce and their relation
5
-alkyl-substituted butenolides have been explored by
to the structure has been discussed mainly in the case
5
several groups. A chromium carbene route to optically
active 5-alkoxy-5-alkyl-2(5H)-furanones was recently de-
(9) Toda, F.; Tanaka, K.; Leung, C. W.; Meetsma, A.; Feringa, B. L.
J . Chem. Soc., Chem. Commun. 1994, 2371.
scribed by Hegedus and co-workers.6
(10) Van der Deen, H.; Hof, R. P.; Van Oeveren, A.; Feringa, B. L.;
The chemistry of γ-alkoxybutenolides has been ex-
plored in our laboratory.7 Diastereomerically pure buteno-
lides have been prepared with the aid of a chiral
auxiliary, (+)- or (-)-menthol (A, R ) menthyloxy).2
Racemic 5-alkoxy-2(5H)-furanones were kinetically re-
solved by the addition of thiols catalyzed by cinchona
alkaloids.8 Efficient resolution of γ-alkoxybutenolides
through inclusion complexation with chiral host com-
pounds derived from tartaric acid has been reported,9
†
A. Mickiewicz University.
(
1) (a) Hanessian, S.; Murray, P. J . Tetrahedron 1987, 43, 5055. (b)
Knight, D. W. Contemp. Org. Synth. 1994, 1, 287.
2) (a) Feringa, B. L.; de Lange, B.; J ansen, J . F. G. A.; de J ong, J .
C.; Lubben, M.; Faber, W. S.; Schudde, E. P. Pure Appl. Chem. 1992,
(
6
1
1
4, 1865. (b) Feringa, B. L.; de J ong, J . C. Bull. Soc. Chim. Belg. 1992,
01, 627. (c) Feringa, B. L.; de Lange, B.; de J ong, J . C. J . Org. Chem.
989, 54, 2471.
(3) See ref 1 and (a) Hanessian, S. The Total Synthesis of Natural
Products, the Chiron Approach; Pergamon: New York, 1993. (b)
Hanessian, S.; Sahoo, S. P.; Botta, M. Tetrahedron Lett. 1987, 28, 1147.
(4) (a) Camps, P.; Goudellach, J .; Font, J .; Ortuno, R. M.; Ponsati,
O. Tetrahedron 1982, 38, 2395. (b) Takano, S.; Goto, E.; Hirama, M.;
Ogasawara, K. Heterocycles 1981, 16, 951. (c) Koseki, K.; Ebata, T.;
Kawakami, H.; Matsushita, H.; Naoi, Y.; Itoh, K. Heterocycles 1990,
3
1, 423.
(22) 5-Alkyl-2(5H)-butenolides synthesized from optically active
propargylic alcohols have not been originally assigned absolute con-
figuration: Vigneron, J . P.; Blanchard, J . M. Tetrahedron Lett. 1980,
21, 1739.
(23) (a) Beckmann, M.; Hildebrandt, H.; Winterfeldt, E. Tetrahe-
dron: Asymmetry 1990, 1, 335. (b) Trost, B. M.; M u¨ ller, T. J . J . J . Am.
Chem. Soc. 1994, 116, 4985. (c) Ishigami, K.; Kitahara, T. Tetrahedron
1995, 51, 6431.
(5) (a) Tomioka, K.; Ishiguro, T.; Koga, K. Tetrahedron Lett. 1980,
2
1, 2973. (b) J urczak, J .; Tkacz, M. Synthesis 1979, 42. (c) Ortuno, R.
M.; Bigorra, J .; Font, J . Tetrahedron 1987, 43, 2199.
6) (a) Hegedus, L. S.; Bates, R. W.; Soderberg, B. C. J . Am. Chem.
Soc. 1991, 113, 923. (b) Reed, A. D.; Hegedus, L. S. J . Org. Chem. 1995,
0, 3787.
7) See ref 2 and Van Oeveren, A.; J ansen, J . F. G. A.; Feringa, B.
L. J . Org. Chem. 1994, 59, 5999.
8) Faber, W. S.; Kok, J .; de Lange, B.; Feringa, B. L. Tetrahedron
994, 50, 4775.
(
6
(
(24) Frischmuth, K.; Wagner, U.; Samson, E.; Weigelt, D.; Koll, P.;
Meuer, H.; Sheldrick, W. S.; Welzel, P. Tetrahedron: Asymmetry 1993,
4, 351.
(
1
0
022-3263/96/1961-1513$12.00/0 © 1996 American Chemical Society

1
514 J . Org. Chem., Vol. 61, No. 4, 1996
Notes
Ch a r t 1
of bi- and polycyclic 2(5H)-furanones.25 Beecham dis-
cussed the effect of ring nonplanarity in polycyclic
butenolides on the n-π* transition rotatory strength,2
and Richardson examined this effect by the INDO-MO
calculations.2 Uchida and Kuriyama studied the π-π*
Cotton effect of mono- and polycyclic R,â-unsaturated
γ-lactones and found that the sign of the Cotton effect
correlates with the configuration of the more polarizable
bond at γ carbon atom.2 Our analysis of the CD data
of 5-substituted 2(5H)-furanones takes advantage of the
well-established planarity of the butenolide ring.2 The
R,â-unsaturated lactone chromophore in the butenolide
ring is achiral, and it becomes optically active in the
presence of a perturber at the stereogenic center at C(5).
The observed Cotton effects due to the n-π* and π-π*
transitions of the R,â-unsaturated lactone chromophore
may be correlated directly to the absolute configuration
of the stereogenic center. This approach has been
successfully applied to planar 2-cyclopentenones bearing
The 5-(menthyloxy)- (7-9, 17, and 18), 5-methoxy- (6),
and 5-isopropoxybutenolides (16)9 were prepared as
previously described. Optically active 5-acetoxy-2(5H)-
furanone (15)10 was obtained in 31% yield via enzymatic
kinetic resolution of racemic 15. When lipase PS im-
mobilized on Hyflo Super Cell in n-hexane:n-butanol (3:
2
5a
5b
1
) was used, a transesterification occurred and the
starting material recovered after 4 h was enantiomeri-
cally pure (ee > 99% as determined by chiral GC). The
absolute configurations of all the butenolides except for
5c
,26
1
5 and 16 are known from chemical correlation and the
absolute configuration of 5-(hydroxymethyl)-2(5H)-fura-
none and d- or l-menthol. The CD data of all butenolides
are given in Table 1.
From the data of Table 1 it can be readily seen that
absolute configuration at C(5) is correlated to the sign of
the Cotton effects of the n-π* (235-250 nm) and π-π*
200-220 nm) transitions according to Figure 1. Thus,
right-handed (P) helicity of the R-C(5)-CdC bond
(
2
7
an oxygen substituent at C(4).
Here we report the results of our CD studies of a series
of 5-alkyl- and 5-alkoxy-substituted butenolides 1-18
(Chart 1).
(25) (a) Beecham, A. F. Tetrahedron 1972, 28, 5543. (b) Kreigh, H.;
Richardson, F. S. J . Chem. Soc., Perkin Trans. 2 1976, 1674. (c) Uchida,
I.; Kuriyama, K. Tetrahedron Lett. 1974, 3761.
(
26) Examples of X-ray structure determination of butenolides: (a)
Coggon, P.; Luhan, P. A.; McPhail, A. T. J . Chem. Soc. Perkin Trans.
1973, 465. (b) Ru zˇ i c´ -Toro sˇ , Zˇ .; Koji c´ -Prodi c´ , B. Acta Crystallogr. 1976,
2
B32, 2333. (c) Ru zˇ i c´ -Toro sˇ , Zˇ .; Lazarini, F. Ibid. 1978, B34, 854. (d)
Ru zˇ i c´ -Toro sˇ , Zˇ .; Leban, I. Ibid. 1978, B34, 1226. (e) Ru zˇ i c´ -Toro sˇ , Zˇ .;
Koji c´ -Prodi c´ , B. Ibid. 1982, B38, 1664. (f) Parvez, M.; Fajardo, V.;
Shamma, M. ibid. 1988, C44, 553. (g) Font, J .; Sanchez-Ferrando, F.;
Segura, C.; Piniella, J . F.; J effrey, G. A.; Ruble, J . R. J . Heterocycl.
Chem. 1990, 27, 183. (h) Ammon, H. L.; DeShong, P.; Lessen, T. A.
Acta Crystallogr. 1991, C47, 2488. (i) Hynes, R. C. Ibid. 1992, C48,
F igu r e 1. Correlation of the butenolide Cotton effects with
absolute configuration.
(27) Gawronski, J . K. In Houben-Weyl Methods of Organic Chem-
istry, 4th ed.; Helmchen, G., Hoffmann, R. W., Mulzer, J ., Schaumann,
E., Eds.; Thieme: Stuttgart, 1995; Vol. E21a, pp 506-508.
3
95. (j) Linden, A. Ibid. 1992, C48, 1514. (k) Parvez, M.; Yu, S.; Keay,
B. A. Ibid. 1993, C49, 992.

Notes
Ta ble 1. CD a n d UV Da ta for Ch ir a l 2(5H)-F u r a n on es
J . Org. Chem., Vol. 61, No. 4, 1996 1515
The configurational rule can be particularly useful for
determining the absolute configuration of the butenolide
moiety in acetogenins, previously possible only through
synthesis.29 Thus, acetogenins bearing the 5-methyl-
(5H)-furanone group of (S)-configuration (B) as the sole
chromophore display a negative n-π* Cotton effect at
235-240 nm and a positive π-π* Cotton effect at 205-
CD, ∆ꢀ (nm)
UV, ꢀ (nm)
π-π*
compd
n-π*
π-π*
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-0.1 (239)
-0.2 (239)
-0.3 (239)
-0.3 (242)
-0.4 (242)
-2.3 (249)
-1.9 (248)
-3.7 (242)
-2.4 (245)
+6.7 (203)
+5.5 (202)
+7.0 (201)
+6.4 (205)
+10.9 (206)
+1.9 (203)
+5.6 (201)
+7.5 (214)
+4.7 (212)
-13.0 (209)
-13.5 (207)
-10.0 (207)
-5.8 (220)
-8.0 (205)
-13.6 (198)
-2.6 (201)
-5.7 (201)
-6.0 (213)
-3.4 (218)
+6.0 (222)
6300 (207)
6800 (205)
8700 (205)
7200 (204)
7900 (204)
6500 (201)
6300 (201)
9500 (207)
7400 (203)
7200 (206)
7100 (206)
11 600 (208)
5300 sh (220)
7600 (204)
6900 (201)
6400 (201)
6300 (201)
2
a
30,31
210 nm.
The configurational rule also seems to be applicable
to 2(5H)-furanones having an allylic hydroxy group in a
ring fused to the butenolide ring. The two diastereoiso-
meric lactarorufins32 LRA (19) and epi-LRA (20) display
Cotton effects of opposite signs, and in each case the signs
of the n-π* and π-π* transition Cotton effects follow
the helicity rule for the HO-C-CdC bond system.
1
1
1
1
1
1
1
1
1
1
2
+0.6 (236)
+0.2 (243)
+2.3 (250)
+0.75 (248)
+1.9 (249)
+2.7 (247)
+1.8 (241)
-0.3 (250)
Exp er im en ta l Section
UV and CD spectra were recorded with a Shimadzu 160
spectrophotometer and a J obin-Yvon III dichrograph, respec-
b
-
3
tively, in methanol solutions (10 M).
9
2c
Butenolides 6, 16, 7-9, 17, and 18 were prepared following
reported procedures.
a
De 88%. b Data from ref 24. Solvent: acetonitrile.
(
R,S)-5-Acetoxy-2(5H)-fu r a n on e ((R,S)-15). To a mixture
system (where R is an alkyl or alkoxy group) gives rise
to a negative n-π* and a positive π-π* Cotton effect.
Opposite sign pattern is observed for left-handed (M)
helicity of the bond system mentioned above.28
of 5-hydroxy-2(5H)-furanone (22.8 g, 227 mmol) and acetic
anhydride (24.75 g, 242 mmol) was added a catalytic amount of
p-toluenesulfonic acid. The solution was heated to 50 °C for 2.5
h, and the product was distilled to provide 28.99 g (89%) of (R,S)-
1
1
5 as a pale yellow oil: bp 102-104 °C (0.05 mmHg); H NMR
The following characteristic features are observed: (i)
the rule works regardless of an additional chiral center
in the substituent R (compounds 4, 5, 7-9, 13, 14, 17,
(CDCl
3
, 200 MHz) δ 7.32 (dd, 1H, J ) 5.6, 1.3 Hz), 6.95 (d, 1H,
J ) 1.2 Hz), 6.30 (dd, 1H, J ) 5.6, 1.3 Hz), 2.13 (s, 3H);
13
C
NMR (50 MHz) δ 168.90, 169.50, 149.80, 125.09, 93.76, 20.56.
Tr a n sester ifica tion of (R,S)-5-Acetoxy-2(5H)-fu r a n on e.
To a solution of racemic 15 (8.00 g, 56.3 mmol) in 1.5 L of a
mixture of n-hexane:n-butanol (3:1) was added 3.0 g (30 m/m
1
8); (ii) the rule can also be applied to butenolides bearing
additional methyl groups attached to the CdC bond
compounds 8, 9, 12, and 18); (iii) Cotton effects associ-
ated with the n-π* transition are weak except for
-alkoxy-substituted 2(5H)-furanones. For example, the
(
%) of lipase PS immobilized on Hyflo Super Cell. The mixture
was vigorously stirred at room temperature, and the progress
of the reaction and the enantiomeric excess were monitored by
chiral GC (capillary column coated with CP cyclodextrin-b-2,3,6-
M-19). After 4 h, an ee > 99% was found and the stirring was
stopped. After precipitation of the solid material (10 min) the
solution was decanted from the enzyme slurry and filtered
through Celite. The solvents were removed by distillation under
vacuum at room temperature, and the remaining yellow oil was
purified by chromatography (silica gel, hexane:EtOAc 3:1). Pure
5
n-π* Cotton effect is not observed for 10, 11 and 13 (in
methanolic solution). Therefore it is generally advisable
to correlate the absolute configuration with the sign of
the π-π* Cotton effect.
As a corollary it seems that the sign of rotation of the
butenolide22 at the sodium D-line can be related to its
absolute configuration. In the absence of a chiral center
other than at C(5) the rotation is determined primarily
by the sign of the strong Cotton effect of the π-π*
transition; i.e., it is positive for P-helicity and negative
for M-helicity of the R-C-CdC bond system (Figure 1).
The relation between the absolute configuration and the
rotation has been previously observed on a purely
experimental basis for γ-C-glycosylated butenolides,1
and it holds for many other butenolides having chiral
substituents at C(5) such as the menthyloxy group (7-
15 was obtained as an oil (2.55 g, 31.8%): [R]
D
) +25.4 (c 1.00,
CHCl ).
3
Ack n ow led gm en t. We gratefully acknowledge the
following groups that provided us with the samples for
2
0b
CD study: Prof. A. Sutherland (1-3), Prof. T. Yamaza-
1
3
33
34
ki (4, 5, 14), Prof. J . Font (12), Prof. P. Garner (13),
3
2
and Prof. Daniewski (19, 20).
J .K.G. thanks the
2a
Committee for Scientific Research (KBN) for partial
support. This work was sponsored in part (A.v.O., H.v/d
D., B.L.F.) by the Netherlands Foundation for Scientific
Research (NWO/SON) and (C.W.L.) by the EC project
Stereoselective Organic Synthesis (Human Capital and
Mobility Network).
9
, 17, and 18); see, however, ref 11. It needs to be
emphasized that such a relationship should be treated
with care, and analysis of the CD spectra provides a far
more reliable way of determining the absolute configu-
ration of butenolides.
On the basis of the rules presented here, the 5(S)
configuration could now be assigned to (-)-5-acetoxy-
J O951400L
(
29) See, for example: Hoye, T. R.; Humpal, P. E.; J im e´ nez, J . I.;
Mayer, M. J .; Tan, L., Ye, Z. Tetrahedron Lett. 1994, 35, 7517.
30) Gawronski, J . K.; Wu, Y.-C. Unpublished results.
(
2
(5H)-furanone (15) and (-)-5-isopropoxy-2(5H)-furanone
(31) Published CD data on acetogenins. Reticulatacin: Saad, J . M.;
Hui, Y.-H.; Rupprecht, J . K.; Anderson, J . E.; Kozlowski, J . F.; Zhao,
G.; Wood, K. V.; McLaughlin, J . L. Tetrahedron 1991, 47, 2751.
Bullatalicin: Hui, Y.-H.; Rupprecht, J . K.; Anderson, J . E.; Liv, Y.-
M.; Smith, D. L.; Chang, Ch.-J .; McLaughlin, J . L. Tetrahedron 1989,
45, 6941. Bullatacin, asimicin, rolliniastatin: Hui, Y.-H.; Rupprecht,
J . K.; Liu, Y.-M.; Anderson, J . E.; Smith, D. L.; Chang, C.-J .;
McLaughlin, J . L. J . Nat. Prod. 1989, 52, 463. On the basis of the CD
data the absolute configuration at C(36) of bullatacin as shown in this
paper should be reversed.
(16) from their CD spectra.
(28) It is apparent that of the two allylic bonds at C(5), C-alkyl (or
C-alkoxy), and CH, each having opposite helicity relation to the CdC
bond, the former, more polarizable, has a dominating effect on the sign
of the π-π* Cotton effect; see earlier work on allylic axial chirality
effect: (a) Beecham, A. F.; Mathieson, A. McL.; J ohns, S. R.; Lamber-
ton, J . A.; Sioumis, A. A.; Batterham, T. J .; Young, I. G. Tetrahedron
1
971, 27, 3725. (b) Beecham, A. F. Ibid. 1971, 27, 5207. (c) Scott, A. I.;
(32) Daniewski, W. M.; Koc o´ r, M.; Kr o´ l, J . Roczniki Chem. 1976,
50, 2019.
(33) Alibe’s, R.; Bourdelande, J . L.; Font, J . Tetrahedron Lett. 1993,
34, 7455.
Wrixon, A. D. Ibid. 1971, 27, 4787. (d) Totty, R. N.; Hudec, J . J . Chem.
Soc., Chem. Commun. 1971, 785. (e) Burgstahler, A. W.; Barkhurst,
R. C.; Gawronski, J . K. In Modern Methods of Steroid Analysis;
Heftmann, E., Ed.; Academic Press: New York, 1973; pp 349-379.
(34) Garner, P.; Park, J . M. J . Org. Chem. 1990, 55, 3772.