Synthesis of (+)- and (-)-phaseolinic acid by combination of enzymatic hydrolysis and chemical transformations with revision of the absolute configuration of the natural product

Full text of DOI:10.1021/jo972032j

J . Org. Chem. 1998, 63, 2385-2388
2385
Syn th esis of (+)- a n d (-)-P h a seolin ic Acid
Ch a r t 1
by Com bin a tion of En zym a tic Hyd r olysis
a n d Ch em ica l Tr a n sfor m a tion s w ith
Revision of th e Absolu te Con figu r a tion of
th e Na tu r a l P r od u ct
Sara Drioli, Fulvia Felluga, Cristina Forzato,
Patrizia Nitti, Giuliana Pitacco, and Ennio Valentin*
Dipartimento di Scienze Chimiche, Universit a` , via Giorgieri
1
, 34127 Trieste, Italy
Sch em e 1
Received November 5, 1997
In tr od u ction
(
-)-Phaseolinic acid (1),1 a metabolite of a fungus,
Macrophomina phaseolina, and (-)-methylenolactocin
2), an antitumoral antibiotic from Penicillum sp. (Chart
2
(
1
), have recently received considerable attention, and a
few formal and total syntheses of these compounds have
been published.3
In the course of our studies aimed at the synthesis of
4
enantiomerically pure γ-butyrolactones we focused our
attention on the synthesis of both enantiomers of phaseo-
linic acid (1) and on the determination of their absolute
configuration by means of chemical and spectroscopic
correlations. The strategy was to correlate (-)-phaseo-
linic acid (1) with (-)-methylenolactocin (2) through the
butenolide (-)-3 (Chart 1).
Sch em e 2
Resu lts a n d Discu ssion
(
-)-Phaseolinic acid (1) and (-)-methylenolactocin (2)
were prepared starting from two diastereomeric lactones,
and 6, respectively (Scheme 1). These latter compounds
5
5
were obtained from the keto diester 4, as already
proposed by J ohnson and co-workers for the synthesis
6
of racemic protolichesterinic acid. The diastereomeric
ratio 5:6 was 4:1 in favor of the less stable isomer 5 when
the reduction of the keto carbonyl group was performed
at -15 °C.
The enantiomers of phaseolinic acid were synthesized
as shown in Scheme 2. The ethoxycarbonyl group of the
racemic cis-lactone 5 was chemically hydrolyzed to the
acid lactone 7, which was transformed into racemic
(
1) Mahato, S. B.; Siddiqui, K. A. I.; Bhattacharya, G.; Ghosal, T.;
Miyahara, K.; Sholichin, M.; Kawasaki, T. J . Nat. Prod. 1987, 50, 245.
2) Park, B. K.; Nakagawa, M.; Hirota, A.; Nakayama, M. J .
Antibiotics 1988, 41, 751.
3) (a) Zhang, Z.; Lu, X. Tetrahedron: Asymmetry, 1996, 7, 1923.
b) J acobi, P. A.; Herradura, P. Tetrahedron Lett., 1996, 37, 8297. (c)
Martin, T.; Rodriguez, C. M.; Martin, V. S. J . Org. Chem. 1996, 61,
450. (d) Sorkar, S.; Ghosh, S. Tetrahedron Lett. 1996, 37, 4809. (e)
Sibi, M. P.; Deshpande, P. K.; La Loggia, A. J . Synlett 1996, 343. (f)
Saicic, R. N.; Zard, S. Z. Chem. Commun. 1996, 1631. (g) Mawson, S.
D.; Weavers, R. T. Tetrahedron 1995, 51, 11257. (h) Takahata, H.;
Uchida, Y.; Momose, T. J . Org. Chem. 1995, 60, 5628. (i) Vaupel, A.;
Knochel, P. Tetrahedron Lett., 1995, 26, 231. (j) Zhu, G., Lu, X. J . Org.
Chem., 1995, 60, 1087. (k) de Azevedo, M. B. M.; Murta, M. M.; Greene,
A. E. J . Org. Chem. 1992, 57, 4567.
(
(
(
phaseolinic acid 1 by stereocontrolled R-methylation.7,8
6
This latter compound was then esterified to 8 with ethyl
9
iodide in the presence of DBU at room temperature.
The resulting racemic lactone 8 was then kinetically
resolved by means of pig liver esterase (PLE) at room
(
4) (a) Forzato, C.; Nitti, P.; Pitacco, G.; Valentin, E. Gazz. Chim.
(7) Grieco, P. A.; Miyashita, M. J . Org. Chem. 1974, 39, 120.
7
Ital. 1995, 125, 223. (b) Drioli, S.; Felluga, F.; Forzato, C.; Nitti P.;
Pitacco, G. Chem. Commun. 1996, 1289.
(8) Differently from the conditions used by Grieco, NaN(SiMe
3
)
2
was
used instead of LDA and methyl iodide was added in large excess in
order to improve yields.
(5) Patrick, T. M. J r. J . Org. Chem. 1952, 17, 1009.
(
6) Martin, J .; Watts, P. C.; J ohnson, F. J . Org. Chem. 1974, 39,
(9) Ono, N.; Yamada, T.; Saito, T.; Tanaka, K.; Kaji, A. Bull. Chem.
Soc. J pn. 1978, 51, 2401.
1
676.
S0022-3263(97)02032-X CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/06/1998

2
386 J . Org. Chem., Vol. 63, No. 7, 1998
Notes
Ch a r t 2
Sch em e 3
temperature. After 2 h (20% conversion), the acid (+)-1
was obtained in 15% yield and 94% ee,10 while the
unreacted ester (-)-8 (59%) was found to have 32% ee.
When the reaction time was prolonged for 10 h, the
unreacted ester (-)-8 was obtained in 35% yield and 96%
ee, while the remaining acid (+)-1 (35% yield) was found
to have an ee of 47%.10 Hydrolysis of (-)-8, carried out
in refluxing dioxane under acidic conditions for 30 min
furnished pure phaseolinic acid (-)-(1) with 99% ee,10
after crystallization. In this manner both enantiomers
of phaseolinic acid (1) were available in good enantio-
meric purity.
Sch em e 4
Other enzymes such as Candida cylindracea lipase
(CRL) and porcine pancreatic lipase (PPL) were uneffec-
tive, whereas porcine liver acetone powder (PLAP) af-
forded the acid (+)-1 in 10% yield and 88% ee, at 25%
conversion, and the ester (-)-8 in 25% yield and 97% ee,
at 75% conversion.
(-)-Phaseolinic acid (1) thus obtained was converted
into the corresponding butenolide (-)-3 (Scheme 2). The
sequence of reactions involved R-phenylselenenylation in
the presence of LDA as a base with HMPA added,
oxidation, and syn elimination of the phenylselenyloxide
intermediate, in accordance with the procedure by Grieco
from the methyl ester of the unreacted lactone (-)-1315
and Miyashita.7
,11
Since the reaction did not go to
(35% yield, 93% ee).
completion, the crude mixture was eventually treated
with diazomethane to esterify the carboxy groups present
and to allow separation of the products. Besides the
butenolide (-)-3 (99% ee), in fact, also the methyl ester
The butenolide (-)-3 isolated from lactone (-)-11 was
the same enantiomer as that previously obtained from
phaseolinic acid (-)-(1). Since its C-2 does not enter in
any of the above reactions, it seems reasonable to assign
the S configuration also to C-2 of phaseolinic acid (-)-1.
Further experiments were performed to settle the abso-
lute configuration of phaseolinic acid (-)-1, as shown in
1
2
of the unreacted phaseolinic acid (-)-9 and the all-cis-
lactone (-)-1013 were isolated by chromatography (Chart
2
). The three lactones (-)-3, (-)-9, and (-)-10 were in
the ratio 75:15:10.
Scheme 4.
The same butenolide (-)-3 was arrived at starting from
The ester lactone (-)-5 having 92% ee16 isomerized to
-)-6 by treatment with DBU in refluxing dichlo-
the lactone (-)-11 (Scheme 3), whose configuration is
(
known3
k,4b
to be 2S,3R, being a direct precursor of (-)-
romethane. Since it is unlikely that inversion of config-
uration occurs at C-2 and because lactone (-)-6 is the
ethyl ester of the lactone (-)-11, a direct precursor of (-)-
methylenolactocin (2) (Scheme 3), the configuration of C-2
in (-)-6 and hence in (-)-5 must be S.
The ethoxycarbonyl group in (-)-5 was then chemically
hydrolyzed and the resulting acid lactone (-)-7 (62% ee)
was R-methylated with methyl iodide in excess and NaN-
methylenolactocin (2).
The acid lactone (-)-11,4b having 93% ee, was R-meth-
ylated under the stereocontrolled conditions by Grieco.7
The resulting lactone (-)-12 (88% yield) was transformed
into the butenolide (-)-3 through the same sequence of
reactions reported above for the conversion of (-)-1 into
,8
-)-314 (Scheme 3). The butenolide (-)-3 was isolated in
(
(
3 2
SiMe ) as the base, furnishing phaseolinic acid (-)-1,
1
8% yield and 94% ee after chromatographic separation
with 63% ee (Scheme 4). Therefore, the configuration at
C-2 must be the same both in (-)-phaseolinic acid (1) and
(
(
10) Determined by chiral HRGC of either its methyl or ethyl ester.
11) The reaction yield was satisfactory only when phenylselenyl
(
-)-methylenolactocin (2). A further support to this
chloride was used in large excess (twelvefold).
12) The geometry of (-)-9 was confirmed by means of DIFNOE
conclusion came from the analysis of the CD spectra of
phaseolinic acid (-)-1 and the butenolide (-)-3.
(
measurements. Irradiation of H-2 doublet of double doublet at 4.65
ppm enhanced H-3 doublet of doublets (10%), while irradiating the
methyl doublet at 1.30 ppm, the signals relative to H-2 (2%), H-3 (3%)
and H-4 (4%) were enhanced.
Phaseolinic acid (-)-1 exhibited a positive Cotton effect
1
(
∆ꢀ 216 ) +2.2, dioxane, lit. ∆ꢀ ) +1.76). Interestingly,
(
13) The geometry of (-)-10 was established by means of DIFNOE
measurements. Irradiation of H-2 doublet triplet at 4.34 ppm enhanced
both H-3 doublet of doublets (6%) and H-4 double quartet (4%), while
irradiating H-3 at 3.25 ppm, the signals of H-2 (6%) and H-4 (3%) were
enhanced. Finally, an enhancement of H-2 (3%) and H-3 (4%) was
observed by irradiating H-4 signal at 2.83 ppm.
(15) The geometry of (-)-13 was established by means of DIFNOE
measurements. Irradiation of H-2 produced enhancement of H-4
doublet quartet (4%), while irradiation of H-3 at 2.64 ppm enhanced
the methyl doublet at 1.31 ppm (4%)
(16) Drioli, S.; Felluga, F.; Forzato, C.; Nitti, P.; Pitacco, G.; Valentin,
E. J . Mol Catal. Part B: Enzymatic 1997, 3, 203.
3 2
(14) NaN(SiMe ) was used in place of LDA, which gave no results.

Notes
J . Org. Chem., Vol. 63, No. 7, 1998 2387
Ch a r t 3
h. Then the mixture was allowed to warm until -20 °C, 2 N
HCl (8 mL) was added, the mixture was extracted three times
with ether, and the organic phases were washed with brine and
2 4
dried on Na SO . After removal of the solvent, racemic phaseo-
linic acid (()-1 (0.152 g, 95% yield) was isolated in mixture with
the unreacted lactone (()-7 (5%). Separation of the two lactones
was accomplished by transformation into the corresponding ethyl
esters (()-5 and (()-8 with ethyl iodide and 1,8-diazabicyclo-
9
[
5.4.0]undec-7-ene (DBU) and separation by flash chromatog-
Ch a r t 4
raphy (eluent, ethyl acetate-light petroleum, gradient from 0%
to 20%). Phaseolinic acid (()-1 was then recovered by acidic
hydrolysis in dioxane of compound (()-8.
(
2R *,3R *,4R *)-4-Me t h yl-5-oxo-2-p e n t yl-t e t r a h yd r o-3-
fu r an car boxylic Acid (1): mp 150-1 °C (from light petroleum-
ethyl acetate): 1H NMR (400 MHz) δ 5.8 (1 H, bs, OH), 4.64 (1
H, dt, J ) 8.3, 8.3, 5.4, H-2), 3.25 (1 H, dd, J ) 9.3, 8.3, H-3),
pertusarinic acid (-)-1417 (Chart 3), possessing the same
S,3S,4S configuration, although a different aliphatic
chain at C-2, showed a positive Cotton effect at 221 nm
∆ꢀ221 ) +1.38, CH CN) due to the n f π* transition.
The butenolide (-)-3 exhibited in its CD spectrum a
weak negative n f π* Cotton effect at 256 nm (∆ꢀ256
0.6) and a strong positive π f π* Cotton effect at 229
nm (∆ꢀ229 ) +2.5), which are consistent with the S
2
.90 (1 H, dq, J ) 9.3, 6.8, H-4), 1.54 (2 H, m, 2 H-1′), 1.44 (1 H,
2
m, H-2′), 1.34 (1 H, m, H-2′), 1.25 (4 H, m, 2 H-3′, 2 H-4′), 1.16
(3H, d, J ) 6.8, Me at C-3), 0.81 (3 H, bt, Me of the chain); 13
C
(
3
NMR (100.4 MHz) δ 178.4 (s), 172.1 (s), 78.5 (d, C-2), 52.4 (d,
C-3), 37.8 (d, C-4), 32.6 (t, C-3′), 32.3 (t, C-1′), 26.7 (t, C-2′), 23.6
(
t, C-4′), 15.0 (q, Me at C-4), 14.7 (q, Me of the chain). (Found:
C, 61.80, H, 8.42. C11 requires: C, 61.66, H, 8.47.)
2R*,3R*,4R*)-4-Meth yl-5-oxo-2-pen tyltetr ah ydr o-3-fu r an -
ca r boxylic Acid Eth yl Ester (8). The ester (()-8 was isolated
as an oil: IR (neat) 1782, 1736 (CdO), 1200 (O-CO); H NMR
(400 MHz) δ 4.66 (1 H, ddd, J ) 10.0, 8.3, 3.4, H-2), 4.23 (2 H,
)
18 4
H O
-
(
1
8
configuration of its stereocenter. This attribution fol-
1
1
8a
lows from the studies initially made by Uchida
and
more recently by Gawronski1 and Kirby on a series
8b
18c
q, J ) 7.3, OCH CH ), 3.17 (1 H, dd, J ) 9.8, 8.3, H-3), 3.05 (1
2
3
H, dq, J ) 9.8, 7.3, H-4), 1.52 (2 H, m, H-1′, H-2′), 1.42 (1 H, m,
H-1′), 1.40-1.20 (11 H, d, t, m, 2 Me, H-2′, 2 H-3′, 2 H-4′), 1.30
of natural and synthetic substituted R-butenolides. They
found that the sign of the Cotton effect of both transitions
are correlated with the absolute configuration of the
γ-carbon atom of the R,â-unsaturated γ-lactone.
(
t, J ) 7.3, OCH CH ), 1.29 (d, J ) 7.3, Me at C-4), 0.88 (3 H,
2 3
13
bt, Me of the chain); C NMR (100.4 MHz) δ 177.6 (s), 169.5
(s), 77.4 (d, C-2), 61.2 (t, OCH CH ), 51.6 (d, C-3), 36.2 (d, C-4),
2
3
Since all our results are nonconflicting, the conclusion
can be drawn that the configuration of natural phaseo-
linic acid having negative specific rotation is 2S,3S,4S
and not 2R,3R,4R, as reported in the literaure.1
31.2 (t, C-3′), 30.9 (t, C-1′), 25.1 (t, C-2′), 22.2 (t, C-4′), 14.3 (q,
Me at C-4), 14.0 (q, OCH CH ), 13.7 (q, Me of the chain); mass
spectrum (EI) m/z 243 (MH , 0.8), 242 (M , 0.5), 214 ([M -
2
3
+
+.
+
2 5
C H ] , 0.8), 196 (14), 195 (13), 183 (13), 171 (77), 169 (40), 143
,3a,3b
(100), 142 (27), 127 (18), 115 (60), 99 (34), 97 (79), 95 (20), 87
55), 83 (23), 81 (19), 71 (23), 69 (98), 55 (70).
(
Exp er im en ta l Section
Kin etic Resolu tion of P h a seolin ic Acid Eth yl Ester (()-
8
. To a solution of the lactone (()-8 (0.460 g, 1.9 mmol), in a
Tetrahydrofuran was distilled from sodium benzophenone
ketyl. Lipase from porcine pancreas (type II, crude, no. L-3126)
and from Candida cylindracea (no. L-1754), esterase from pig
2 4 2 4
buffer solution (0.1 M KH PO /Na HPO , 5.0 mL), was added
PLE (170 units/mg, Sigma, 0.10 mL). The pH value was
mantained at 7.5 by continuous addition of 1 N NaOH. After 2
h (20% conversion), the crude reaction mixture was extracted
with diethyl ether. After the usual workup, the lactone (-)-8
0.271 g, 59% yield) in 32% ee was obtained. The aqueous phase
was acidified to pH 2 with 1 N HCl and extracted several times
4 2 4
liver in 3.2 M (NH ) SO suspension (no. E-2884), and porcine
liver acetone powder (no. L-8251) were supplied from Sigma
Chemicals Co. Thin-layer chromatography (TLC) was performed
on Merck 60F254 glass-backed silica gel plates with visualization
by UV light (254 nm) or iodine. Flash chromatography (FC)
purifications were carried out on silica gel (Merck 60, 230-400
(
with diethyl ether. The usual workup furnished the acid (+)-1
(
0.061 g, 15% yield): mp 141-2 °C (from light petroleum-ethyl
mesh). NMR spectra were recorded in CDCl
residual CHCl at 7.26 ppm ( H) and 77.0 ppm ( C). Mass
3
3
and referenced to
25
acetate); [R]
4% ee.
D
) +111.4 (c 0.22, CHCl ); ∆ꢀ220 ) -2.0, dioxane;
3
1
13
9
spectra were obtained at 70 eV. Chiral HRGC analyses were
obtained using a CHIRALDEX type G-TA, trifluoroacetyl-γ-
cyclodextrin, 40 m × 0.25 mm, column.
The kinetic resolution of (()-8 (0.460 g, 1.9 mmol), carried
out under the same conditions as above for 10 h, furnished the
acid (+)-1 (0.142 g, 35% yield) with 47% ee and the ester (-)-8
All reactions leading to enantiomerically pure compounds
were also repeated on racemic mixtures, to evaluate the enan-
tiomeric excess by chiral HRGC.
25
(
∆
+
0.161 g, 35% yield) with 96% ee: [R]
D
) -90.4 (c 0.23, CHCl
3
),
25
ꢀ
219 ) +2.3, dioxane; [R]
D
) -98.3 (c 0.24, CH
3
CN), ∆ꢀ218 )
1.8, CH CN.
3
Numbering of butanolide and butenolide is given in Chart 4.
Syn th esis of Ra cem ic P h a seolin ic Acid (()-(1). Lactone
Hydrolysis of (-)-8 (0.100 g, 0.40 mmol) was carried out in
refluxing dioxane (4.5 mL) in the presence of 6 N HCl (2.2 mL)
for 30 min. After elimination of dioxane the mother liquors were
treated with a solution of NaHCO and extracted with ether.
3
The basic solution was treated with 6 N HCl until pH 2 and
extracted with ether. After the usual workup, evaporation of
(
()-5 (0.6 g, 2.6 mmol) was refluxed in 50 mL of a 2:1 mixture
of dioxane and 6 N HCl for 30 min. Evaporation of the solvent
left the acid (()-7 (0.52 g, 98% yield, after crystallization from
9
:1 light petroleum-ethyl acetate). A solution of (()-7 (0.150
g, 0.75 mmol) in 1.4 mL of anhydrous THF was added to sodium
bis(trimethylsilyl)amide (1.65 mL, 1.65 mmol) (1.0 M solution
the solvent left a solid (0.080 g, 91% yield) which was crystallized
25
with light petroleum-ethyl acetate, (-)-1: mp 140-1 °C; [R]
D
1
7
in THF) at -78 °C under Ar over 30 min. The mixture was
)
-112.3 (c 0.26, CHCl
3
); ∆ꢀ219 ) +2.2, dioxane; 99% ee [lit.
stirred at -78 °C for 1 h, then methyl iodide (0.45 mL, 7.2 mmol)
3a
mp 139-40 °C, [R]
D
) -150 (c 0.2, CHCl
) -142 (c 0.22, CHCl
3
); lit. mp 138-40 °C,
was slowly added and the mixture was stirred for additional 2
25
3b
26
[
(
R]
c 0.37, CHCl
Con ver sion of (-)-P h a seolin ic Acid (1) to th e Bu ten o-
D
3
); lit. mp 137-8 °C, [R]
D
) -147
3
)].
(
17) (a) Huneck, S.; Tønsberg, T.; Bohlmann F. Phytochemistry,
986, 25, 453. (b) Shimada, S.; Hashimoto, Y.; Saigo, K. J . Org. Chem.
993, 58, 5226.
1
1
lid e (-)-3. To a stirred solution of LDA (1.8 mmol) (1.5 M
solution in cyclohexane) in anhydrous THF (2 mL) at -78 °C
under Ar was added (-)-phaseolinic acid (1) (0.15 g, 0.7 mmol)
in THF (1.6 mL) slowly. The mixture was stirred at -78 °C for
3 h, then a solution of HMPA (1.5 mL) and phenylselenyl
chloride (1.60 g, 8.3 mmol) in THF (27 mL) was added and the
(
18) (a) Uchida, I.; Kuriyama, K. Tetrahedron Lett. 1974, 3761. (b)
Gawronski J . K.; van Oeveren, A.; van der Deen, H.; Leung, C. W.;
Feringa, B. L. J . Org. Chem. 1996, 61, 1513. (c) Cain, R. B.; Kelly, S.
M.; Kirby, G. W.; McLenaghan, H. J . S.; Marshall, A. J .; Price, N. C.;
Rao G. V.; Schmidt S. J . Chem. Research (S), 1996, 526.

2
388 J . Org. Chem., Vol. 63, No. 7, 1998
Notes
reaction was set aside for 2 h. The mixture was stirred for
further 1.5 h at -78 °C, then the temperature was allowed to
raise to -20 °C and 2 N HCl was added. The solution was
extracted four times with ether. The organic phases were
washed with brine and extracted four times with a saturated
H-4), 1.70 (1 H, m, H-1′), 1.50 (2 H, m, H-1′, H-2′), 1.35 (1 H, m,
H-2′), 1.25 (4 H, m, 2 H-3′, 2 H-4′), 1.18 (3 H, d, J ) 7.3, Me at
C-4), 0.83 (3 H, bd, Me of the chain); 13C NMR (100.4 MHz) δ
177.0 (s), 170.0 (s), 78.1 (d, C-2), 51.7 (q, OMe), 50.7 (d, C-3),
39.1 (d, C-4), 31.4 (t, C-3′), 30.9 (t, C-1′), 25.5 (t, C-2′), 22.4 (t,
2
5
solution of NaHCO
3
. The basic solution was acidified with 6 N
C-4′), 13.9 (q, Me of the chain), 10.3 (q, Me at C-4); [R] ) -45.3
D
HCl and extracted four times with diethyl ether. Evaporation
of the solvent left a semisolid crude reaction mixture (0.140 g)
which was dissolved in THF and stirred at 0 °C. Glacial acetic
acid (4 drops) and 30% hydrogen peroxide (2.0 mL) were added.
After stirring at 0 °C for 1 h, 2 N HCl was added and the mixture
extracted three times with ether. Esterification of the resulting
reaction mixture with diazomethane gave an oil (0.140 g) which
was a mixture of the butenolide (-)-3 (75%), the methyl ester
of the unreacted lactone (-)-9 (15%), and a small amount of the
all-cis-lactone (-)-10 (10%). Separation by flash chromatogra-
phy (eluent, ethyl acetate-light petroleum gradient from 0% to
(c 0.23, CH CN); ∆ꢀ ) -0.93; 99% ee.
3
219
Con ver sion of th e tr a n s-La cton e (-)-11 to th e Bu ten o-
4b
lid e (-)-3. The lactone (-)-11 (0.100 g, 0.5 mmol) was
methylated following the procedure described for the methyla-
tion7 of the cis-lactone 7. The crude reaction mixture obtained
(0.096 g, 96% yield) contained the unreacted lactone (-)-11 (8%)
and the methyl lactone (-)-12 (92%). Since they could not be
separated, the subsequent reactions were carried out on the
mixture of (-)-11 and (-)-12. This mixture (0.096 g) in
anhydrous THF (1.0 mL) was added to a stirred solution of
sodium bis(trimethylsilyl)amide (0.8 mL, 0.8 mmol) (1.0 M
solution in THF) at -78 °C under Ar. After 1 h, phenylselenyl
chloride (0.70 g, 3.6 mmol) in THF (12 mL) was added and the
reaction set aside for 2 h. The mixture was stirred for further
1
0%) afforded the butenolide (-)-3 (0.060 g, 37% yield), the
methyl ester of the parent lactone (-)-9, and the all-cis-lactone
-)-10.
(
(
2S)-(-)-2,5-Dih yd r o-4-m et h yl-5-oxo-2-p en t yl-3-fu r a n -
1
.5 h at -78 °C, then the temperature was allowed to raise to
ca r boxylic Acid Meth yl Ester (3). The ester was isolated as
an oil. IR (neat) 1767, 1722, 1664, 1228; H NMR (400 MHz) δ
5
1
-20 °C and 2 N HCl was added. The solution was extracted
four times with ether. The organic phases were washed with
brine and extracted four times with a saturated solution of
.07 (1 H, ddq, J ) 2.2, 2.2, 7.8, H-2), 3.86 (3 H, s, OMe), 2.16
(
3 H, d, J ) 2.2, Me at C-4), 2.05 (1 H, m, H-1′), 1.53 (1 H, m,
3
NaHCO . The basic solution was acidified to pH 2 and extracted
H-1′), 1.37 (2 H, m, 2 H-2′), 1.23 (4 H, m, 2 H-3′, 2 H-4′), 0.85 (3
1
3
four times with diethyl ether. The solvent was evaporated, and
to the crude reaction mixture (0.095 g) dissolved in THF (5 mL)
were added glacial acetic acid (2 drops) and 30% hydrogen
peroxide (1.0 mL). Esterification of the resulting mixture with
diazomethane gave an oil (0.060 g) which was a 1:2 mixture of
H, bt, Me of the chain); C NMR (100.4 MHz) δ 172.8 (s), 162.6
s), 147.5 (s), 137.3 (s), 81.3 (d, C-2), 52.2 (q, OMe), 32.69 (t C-1′),
(
3
1
1.3 (t, C-3′), 24.3 (t, C-2′), 22.3 (t, C-4′), 13.8 (q, Me of the chain),
+
0.7 (q, Me at C-4); mass spectrum (EI) m/z 227 (MH , 0.4), 226
+
.
+
+.
(
(
(
)
(
M , 0.5), 197 ([M - C
2
H
5
] , 60), 156 ([M-C
5
H10] , 100), 128
(
-)-3 and (-)-13, while only traces of the parent lactone (-)-11
45), 127 (37), 124 (52), 123 (31), 99 (36), 71 (31), 67 (43), 55
1
2
5
were found in the H NMR spectrum of the crude reaction
mixture. Separation by flash chromatography (eluent, ethyl
acetate-light petroleum gradient from 0% to 6%) afforded the
saturated lactone (-)-13 (0.040 g, 35% yield) and the unsatur-
ated lactone (-)-3 (0.020 g, 18% yield).
23), 43 (62); [R]
+2.5, CH
D
) -47.5 (c 0.12, CH CN); ∆ꢀ256 ) -0.6; ∆ꢀ225
3
3
CN; 94% ee. The retention time was 33.55 min for
-)-3 and 30.13 min for (+)-3 (120 °C for 2 min, 3 °C/min up to
50 °C within 40 min).
1
(
2S,3S,4S)-(-)-4-Meth yl-5-oxo-2-pen tyltetr ah ydr o-3-fu r an -
ca r boxylic Acid Meth yl Ester (9). The ester (-)-9 was
(2S ,3R ,4R )-(-)-4-Me t h yl-5-oxo-2-p e n t ylt e t r a h yd r o-3-
fu r a n ca r boxylic Acid Meth yl Ester (13). The ester (-)-13
isolated as an oil. IR (neat) 1780, 1735 (CdO), 1205 (O-CO);
1
1
H NMR (400 MHz) δ 4.65 (1 H, ddd, J ) 10.5, 8.3, 3.4, H-2),
was isolated as an oil. IR (neat) 1782, 1736; H NMR (400 MHz)
3
.68 (3H, s, OMe), 3.17 (1 H, dd, J ) 9.8, 8.3, H-3), 3.04 (1 H,
δ 4.35 (1H, ddd, J ) 3.9, 8.3, 9.3 Hz, H-2), 3.77 (3 H, s, OMe),
dq, J ) 9.8, 7.3, H-4), 1.45 (2 H, m, H-1′, H-2′), 1.32 (2 H, m,
2
.94 (1 H, dq, J ) 6.8, 11.2, H-4), 2.64 (1H, dd, J ) 9.3, 11.2,
H-1′, H-2′), 1.25-1.15 (7 H, m and d, Me, 2 H-3′, 2 H-4′), 1.30
H-3), 1.70 (1 H, m, H-1′), 1.61 (1 H, m, H-1′), 1.52 (1H, m, H-2′),
(
(
5
d, J ) 7.3, Me at C-4), 0.90 (3 H, bt, Me of the chain); 13C NMR
100.4 MHz) δ 177.6 (s), 170.2 (s), 77.5 (d, C-2), 52.3 (q, OMe),
1
.38 (1H, m, H-2′), 1.33-1.17 (7 H, m and d, 2 H-3′, 2 H-4′, Me
at C-4), 1.31 (d, J ) 6.8, Me at C-4), 0.88 (3 H, bd, Me of the
1.7 (d, C-3), 36.3 (d, C-4), 31.3 (t, C-3′), 31.2 (t, C-1′), 25.2 (t,
13
chain); C NMR (100.4 MHz) δ 176.8 (s), 171.2 (s), 79.6 (d, C-2),
C-2′), 22.4 (t, C-4′), 14.4 (q, Me at C-4), 13.9 (q, Me of the chain);
5
4.2 (d, C-3), 52.6 (q, OMe), 39.9 (d, C-4), 34.8 (t, C-1′), 31.3 (t,
2
5
[
R]
D
3
) -81.7 (c 0.30, CH CN).
C-3′), 24.9 (t, C-2′), 22.4 (t, C-4′), 14.4 (q, Me at C-4), 13.9 (q, Me
(
2S,3S,4R)-(-)-4-Meth yl-5-oxo-2-pen tyltetr ah ydr o-3-fu r an -
+
of the chain); mass spectrum (EI) m/z 229 (MH , 0.7), 197 (6),
ca r boxylic Acid Meth yl Ester (10). The ester (-)-10 was
25
1
0
57 (45), 129 (77), 102 (100), 97 (45), 69 (90); [R]
D
) -36.7 (c
1
isolated as an oil. IR (neat) 1780, 1730; H NMR (400 MHz) δ
.15, CH CN); ∆ꢀ220 ) -1.1, CH CN; 94% ee.
3
3
4
(
.34 (1H, dt, J ) 5.4, 5.4, 7.8 Hz, H-2), 3.68 (3 H, s, OMe), 3.25
1H, dd, J ) 5.4, 7.8, H-3), 2.83 (1 H, apparent quintet, J ) 7.3,
J O972032J