Chemoenzymatic synthesis of optically active 4-methyl-tetrahydro-5-oxo- 2-furancarboxylic acids and esters

Article abstract of DOI:10.1016/S0957-4166(00)00057-4

Enantiomerically pure 4-methyl-tetrahydro-5-oxo-2-furancarboxylic acids and esters are prepared by enzymatic resolution of the chiral racemic esters. Their stereochemistry as well as their absolute configurations have been established by chemical correlation. The influence of the alkoxycarbonyl group at C-2 and that of the methyl group at C-4 on the sign of the Cotton effect in their CD spectra have been investigated. Formation of enantiomerically pure hydroxydiesters, precursors of the above-mentioned γ- lactones, by baker's yeast reduction of the corresponding ketodiesters was unsatisfactory. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00057-4

Tetrahedron: Asymmetry 11 (2000) 1353±1366
Chemoenzymatic synthesis of optically active 4-methyl-
tetrahydro-5-oxo-2-furancarboxylic acids and esters
Sara Drioli, Cristina Forzato, Patrizia Nitti, Giuliana Pitacco and Ennio Valentin*
Á
Dipartimento di Scienze Chimiche, Universita di Trieste, via Licio Giorgieri 1, I-34127, Trieste, Italy
Received 21 January 2000; accepted 7 February 2000
Abstract
Enantiomerically pure 4-methyl-tetrahydro-5-oxo-2-furancarboxylic acids and esters are prepared by
enzymatic resolution of the chiral racemic esters. Their stereochemistry as well as their absolute con®gurations
have been established by chemical correlation. The in¯uence of the alkoxycarbonyl group at C-2 and that of
the methyl group at C-4 on the sign of the Cotton e€ect in their CD spectra have been investigated. Formation
of enantiomerically pure hydroxydiesters, precursors of the above-mentioned g-lactones, by baker's yeast
reduction of the corresponding ketodiesters was unsatisfactory. # 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction
The synthesis of enantiomerically pure g-lactones bearing either a carboxy group or an alkoxy-
carbonyl group on the ring can be accomplished in many ways,1 among which the cyclization
reactions of the appropriate optically active functionalized hydroxydiesters and the enzymatic
kinetic resolution of the desired chiral racemic lactonic esters should be mentioned1b,d (Scheme 1).
Scheme 1.
In particular, enantiomerically pure tetrahydro-5-oxo-2-furancarboxylic esters were obtained
by enzymatic resolution of their racemates.1b Their a-methyl derivatives, namely the trans- and
cis-4-methyl-tetrahydro-5-oxo-2-furancarboxylic esters 1 and 2 (Fig. 1), are precursors of (+)- and
(^)-marmelo lactones A and B,2 two monoterpenes isolated by Tsuneya et al.3 from the fruit of a
quince (Cydonia oblonga MILL, marmelo) as the main ¯avour components.
* Corresponding author. Tel: +39-040-6763917; fax: +39-040-6763903; e-mail: valentin@dsch.univ.trieste.it
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00057-4

1354
S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
Figure 1. Structures of marmelo lactones A and B
In principle, the trans- and cis-lactones 1 and 2 could be prepared by direct a-methylation of
the commercially available enantiomerically pure tetrahydro-5-oxo-2-furancarboxylic acids. However,
this reaction is known to proceed predominantly in the anti mode, at least on the g-alkoxymethyl
analogues,4 furnishing the trans-diastereomer preferentially. In the present paper, we propose
the synthesis of a few derivatives of 1 and 2 with good enantiomeric excesses, by the use of
enzymes in the enantiodi€erentiating step. An alternative approach was also explored, namely the
bioreduction with raw baker's yeast of the ketodiester precursor of the desired lactones.
2. Results and discussion
2.1. Synthesis of the substrates
The condensation of diethyl methylbutanedioate 3 with ethyl oxalate, acid hydrolysis and
decarboxylation of the triester 4, followed by esteri®cation with ethanol of the resulting diacid 55
(Scheme 2), furnished diethyl 4-methyl-2-oxopentanedioate 6.6,7 Its chemical reduction led to the
corresponding hydroxydiesters 7 and 8 as a 1:1 mixture of syn- and anti-diastereomers. This
assignment was made after cyclization of 7 and 8 into the respective g-lactones 9 and 10, which
Scheme 2. (a) EtO^Na+, EtOH, diethyloxalate; (b) 4N HCl, 100ꢀC; (c) EtOH, C6H6, PTSA, 80ꢀC; (d) H2, 5% Pd(C),
EtOAc; (e) C6H6, PTSA, 80ꢀC

S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
1355
were correctly attributed the trans and cis stereochemistries, respectively, as will be shown later.
The reduction of the ketodiester 6 was carried out under heterogeneous catalysis because the use
of sodium borohydride produced mixtures of products, as a consequence of the simultaneous
partial reduction of the ester groups.1n
2.2. Determination of the geometry of lactones 9 and 10
In order to assign the correct stereochemistry to lactones 9 and 10 and hence to their precursors
7 and 8, NOE di€erence measurements were performed on both diastereomers, as well as on their
acid derivatives 11 and 12. Unfortunately, the only signi®cant result (Fig. 2) was not considered
conclusive proof, owing to the low value of the NOE e€ect observed.
Figure 2. DIFNOE measurements on lactone (^)-11
To produce further evidence, the commercially available lactone (R)-(^)-tetrahydro-5-oxo-2-
furancarboxylic acid 13 was a-methylated following the procedure of Grieco and Miyashita.8 The
resulting lactonic acid (^)-11 was isolated as a single diastereomer and puri®ed as its methyl
ester (^)-14 (Scheme 3). Although the yield of the product (^)-11 was very poor (10%), no trace of
1
its cis-diastereomer 12 was detected of the H NMR spectrum of the crude reaction mixture.
Since the a-methylation of g-lactones is known to be stereocontrolled, at least for the g-alkoxy-
methyl g-lactones,4 the reaction product (^)-11 and hence its ethyl ester (^)-9 were assigned the
trans con®guration.
Scheme 3. (a) CH3I, LDA, HMPA, THF, ^78ꢀC; (b) CH2N2
2.3. Enzymatic resolution of the lactonic esters (þ)-9 and (þ)-10
Of the various enzymes checked, the best result for the enzymatic resolution of the trans-dia-
stereomer (þ)-9 was obtained with a-chymotrypsin (a-CT), which showed the highest enantio-
meric ratio E9 (Table 1). At low conversion value (20%) the hydrolysis product (^)-11 was
isolated with 82% e.e. and 12% yield, while at high conversion (78%) the lactonic ester (+)-9 was
obtained with 98% e.e. and 21% yield. While porcine pancreatic lipase (PPL) and proteinase from
Bacillus subtilis showed the same enantiopreference as a-CT, lipase PS (Pseudomonas ¯uorescens)
and Pseudomonas ¯uorescens lipase (PFL) showed opposite enantiopreferences. However, in all
cases the enantioselectivity was very poor. Other enzymes such as porcine lipase acetone powder
(PLAP), lipase AY (Candida rugosa), lipase AP12 (Aspergillus niger) and Mucor miehei lipase
(MML) were also e€ective in hydrolyzing the substrate, although with no enantioselectivity.

1356
S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
Table 1
Hydrolysis of ethyl trans-4-methyl-tetrahydro-5-oxo-2-furancarboxylate (þ)-9
A particular reactivity was observed for horse liver esterase (HLAP), which hydrolyzed both
the ethoxycarbonyl group and the lactone group, only the latter enantioselectively. In fact, at about
20% conversion, extraction of the reaction mixture (pH 7.4) a€orded the lactonic ester (^)-9 with
19% e.e. Acidi®cation of the mother liquors to pH 1, followed by extraction, a€orded about a 1:1
mixture of the racemic lactonic acid (þ)-11, analyzed as its methyl ester by chiral HRGC, and the
lactonic ethyl ester (+)-9 having 78% e.e. Therefore, it is evident that this latter compound
was formed by an enantioselective enzymatic ring opening of (þ)-9 and subsequent cyclization of
the hydroxyacid 15, not isolated, under acidic conditions (Scheme 4).

S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
1357
Scheme 4.
The most signi®cant results of the enzymatic resolution of (þ)-10 are listed in Table 2. Using
PPL, the lactonic acid (+)-12 was isolated with 92% e.e. and 15% yield at low conversion values,
while at high conversion values the unreacted lactonic ester (^)-10 was obtained with 99% e.e.
and 30% yield. The same enantiopreference was observed for a-CT and for proteinase from
Bacillus subtilis, although with lower enantiomeric excesses. Other enzymes such as lipase AY,
lipase PS, PFL and MML were also able to bring about the hydrolysis of the ethoxycarbonyl
group, although with no enantioselectivity.
Table 2
Hydrolysis of ethyl cis-4-methyl-tetrahydro-5-oxo-2-furancarboxylate (þ)-10

1358
S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
2.4. Determination of the absolute con®guration of the g-lactones (+)-9 and (^)-10
The absolute con®guration of the lactones (+)-9 and (^)-10 was determined by reduction of
their ethoxycarbonyl group with sodium borohydride at room temperature to give the corre-
sponding lactonic alcohols (+)-16 and (+)-17, respectively, whose absolute con®gurations are
known (Fig. 3).4a As a consequence, the absolute con®guration of the dextrorotatory ethyl trans-
4-methyl-tetrahydro-5-oxo-2-furancarboxylate (+)-9 is 2S,4R and that of the laevorotatory cis-
diastereomer (^)-10 is 2S,4S.
Figure 3. Structures of lactones (+)-16 and (+)-17
Incidentally, in this manner the assignment of the trans±cis relative stereochemistries was also
con®rmed.
2.5. Baker's yeast reduction of the ketodiester (þ)-6
Reduction of the ketodiester (þ)-6 with baker's yeast was performed in water, after a thermal
pretreatment of the yeast at 50ꢀC for some period, as indicated in Table 3.
Table 3
Reduction of the ketodiester (þ)-6 with raw baker's yeast in watera
In fact, without this preincubation, the only product formed was the hydroxyester 18. This was
formed by hydrolysis and decarboxylation of the ethoxycarbonyl group close to the carbonyl group,
followed by reduction of the resulting formyl group, as already found for diethyl 2-oxopentanedioate1b

S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
1359
(Scheme 5). Formation of the hydroxyester 18, which was not enantioselective, was minimized
but not prevented when baker's yeast was preincubated at 50ꢀC.
Scheme 5.
As shown in Table 3, baker's yeast reduction was by no means enantioselective or diastereo-
selective, as mixtures of syn- and anti-alcohols 7 and 8, having low enantiomeric excesses, were
always found. Interestingly, prolonged reaction times favoured the inversion of con®guration of
both stereocentres in the syn-hydroxydiester 7. From an analysis of the variation of the relative
percentages of the four stereoisomers and that of the racemic decomposition product it can be
deduced that the oxidative process which initiates the decomposition of the hydroxydiesters
occurred preferentially on the (2S,4R) and (2S,4S) diastereomers. However, since the enantio-
meric excess of the anti-hydroxydiester 8 remained constant, the oxidative process must be
counterbalanced by the reductive process occurring on the resulting ketodiester 6, owing to the
fact that reductases were still active.
Figure 4. CD spectra of lactones (+)-9, (^)-10 and (+)-19

1360
S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
2.6. Analysis of the CD spectra
The CD spectra of the g-lactones (+)-9 and (^)-10, bearing the ethoxycarbonyl group at the
g-position, are shown in Fig. 4, while those of the g-lactones bearing the hydroxymethyl group at
the same position, (+)-16 and (+)-17, are shown in Fig. 5. Both ®gures also include, for a com-
parison, the respective lactones bearing no substituent at the a-position, namely ethyl (S)-(+)-
tetrahydro-5-oxo-2-furancarboxylate 19 in Fig. 4 and (S)-(+)-5-hydroxymethyl-4,5-dihydro-
2(3H)-furanone 20 in Fig. 5.
Figure 5. CD spectra of lactones (+)-16, (+)-17 and (+)-20
The synthesis of (S)-(+)-191b from (S)-(+)-13 was achieved by the procedure used for the race-
mization-free peptide coupling,10 to avoid undesired side reactions occurring by direct ester-
i®cation.1n By treatment with sodium borohydride (S)-(+)-19 a€orded the corresponding
hydroxymethyl-substituted g-lactone (S)-(+)-20.
The g-ethoxycarbonyl g-lactones (+)-9, (^)-10 and (+)-19, having the same absolute con®g-
uration of the g-carbon atom, showed a positive Cotton e€ect associated with the n!p* transi-
tion of about the same magnitude. This e€ect is ascribed to the sole ethoxycarbonyl group at the
g-position, as already underlined by Matsumoto et al.,11 whereas the con®guration of the a-carbon
atom has only minor, if any, e€ect on the sign of the Cotton e€ect. In fact, when the ethoxy-
carbonyl group was reduced to the hydroxymethyl group, as in (+)-16, (+)-17 and (+)-20, the
Cotton e€ect depends only on the con®guration of C-a, i.e. on the orientation of the methyl
group, in accordance with Okuda's rule.12,13 In fact the hydroxymethyl-substituted g-lactone 20
showed a weak negative Cotton e€ect, which was enhanced by the consignate e€ect of the methyl

S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
1361
group in (+)-16, whereas it was overwhelmed by the strong opposite e€ect of the same substituent
in (+)-17.
3. Conclusions
It should be underlined that of the two diastereomeric 4-methyl-tetrahydro-5-oxo-2-furan-
carboxylic acids 11 and 12 only the trans-isomer could be prepared enantiomerically pure from
the commercially available (R)-(+)- and (S)-(^)-tetrahydro-5-oxo-2-furancarboxylic acid. The
chemoenzymatic procedure proposed in the present paper allows the synthesis of both enantio-
mers of both diastereomers with good enantiomeric excesses and satisfactory yields. Their CD
spectra showed the relevance of the presence of a carboxy group at the g-position of a g-lactone
ring.
4. Experimental
4.1. General
IR spectra were recorded in CHCl3, unless otherwise stated, on a Jasco FT/IR-200 spectro-
1
meter. H and 13C NMR spectra were run on a Jeol EX-400 spectrometer (400 MHz for proton
and 100.4 MHz for carbon) using deuteriochloroform as a solvent and tetramethylsilane as an
internal standard. Coupling constants are given in hertz. Optical rotations were determined on a
Perkin±Elmer Model 241 polarimeter. CD spectra were obtained on a Jasco J-710 spectro-
polarimeter for methanol solutions. GLC analyses were performed on a Carlo Erba GC 8000 and
on a Shimadzu GC-14B instrument, the capillary columns being EC-WAX, 30 mÂ0.32 mm,
OV1701, 25 mÂ0.32 mm (carrier gas He 40 kPa, split 1:50, 10 min at 100ꢀC, 3ꢀC/min, 200ꢀC),
ChiraldexTM type G-TA, tri¯uoroacetyl g-cyclodextrin 40 mÂ0.25 mm (carrier gas He 180 kPa,
split 1:100, 10 min at 100ꢀC, 3ꢀC/min, 150ꢀC) or DMePe b-cyclodextrin 25 mÂ0.26 mm (carrier
gas He 110 kPa, split 1:50, 10 min at 100ꢀC, 3ꢀC/min, 150ꢀC). Mass spectra were run in the
electron impact mode (20 eV and 70 eV) on a VG 7070 spectrometer. Enzymatic hydrolyses were
performed using a pH-stat controller PHM290 Radiometer, Copenhagen. TLCs were performed
on Whatman K6F silica gel plates (eluant: light petroleum:ethyl acetate, 7:3). Flash chromato-
graphy was run on a silica gel 230±400 mesh ASTM (Kieselgel 60, Merck). Light petroleum refers
to the fraction with b.p. 40±70ꢀC and ether to diethyl ether.
a-Chymotrypsin (a-CT, 51.8 U/mg), Pseudomonas ¯uorescens lipase (PFL, 42.5 U/mg), proteinase
from Bacillus subtilis (13.1 U/mg) and Mucor Miehei lipase (Lipozyme) were purchased from Fluka;
porcine pancreatic lipase Type II (PPL) and pig liver acetone powder (PLAP) were purchased
from Sigma; lipase AY (Candida rugosa), lipase PS (Pseudomonas ¯uorescens) and lipase AP12
(Aspergillus niger) were purchased from Amano.
4.2. Synthesis of the substrates
4.2.1. Diethyl 3-ethoxycarbonyl-4-methyl-2-oxopentanedioate 46
The reaction between diethyl methylbutanedioate 3 and ethyl oxalate was carried out in accord-
ance with the literature.6 The product 4, obtained as a 2:3 mixture of syn- and anti-diastereomers,

1362
S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
was puri®ed by distillation, b.p. 125±128ꢀC (0.7 torr); 28% yield; IR (®lm), ꢀ, cm^1: 1800 (sh),
1739 (COO), 1250 (C±O); 1H NMR, ꢁ, ppm: 4.44 (0.4H, d, J=7.3 Hz, H-3), 4.37 (0.6H, d, J=8.4
Hz, H-3), 4.27 (2H, 2 q, OCH2CH3), 4.11 (4H, m, 2 OCH2CH3), 3.17 (1H, 2 dq, CHCH3), 1.30,
1.29 (3H, 2 t, OCH2CH3), 1.18 (9H, m, 2 OCH2CH3, CH3); 13C NMR, ꢁ, ppm, for the syn-isomer:
187.4 (s), 173.1 (s), 167.2 (s), 159.8 (s), 61.6 (2 t), 60.8 (t), 55.5 (d), 38.3 (d), 14.0 (q), 13.66 (q),
13.63 (2 q); 13C NMR, ꢁ, ppm, for the anti-isomer: 187.9 (s), 173.4 (s), 167.0 (s), 160.1 (s), 62.6 (2
t), 60.9 (t), 55.9 (d), 38.2 (d), 14.5 (q), 13.7 (q), 13.63 (2 q).
4.2.2. Diethyl 4-methyl-2-oxopentanedioate 67
Acidic treatment of the triester 4 under re¯ux furnished the diacid 5,5 which was esteri®ed with
ethanol under re¯ux in benzene with p-toluene sulfonic acid (PTSA), to give the ketodiester 6. Its
1
IR, H NMR and MS spectra were in accordance with those reported in the literature.7 13
C
NMR, ꢁ, ppm: 192.1 (s), 174.6 (s), 160.2 (s), 62.1 (t), 60.4 (t), 42.0 (t), 34.4 (d), 16.6 (q), 13.7 (q),
13.6 (q).
4.3. Synthesis of the hydroxydiesters 7 and 8
The ketodiester 6 (2.0 g, 9.3 mmol) in ethyl acetate (100 ml) was hydrogenated at room
temperature and atmospheric pressure (H2/5% Pd/C, 0.4 g). After 20 h the suspension was
®ltered on Celite and a 1:1 mixture of compounds 7 and 8 was obtained in quantitative yield.
Attempts to separate the diastereomeric mixture by ¯ash chromatography resulted in the isolation
of mixtures of 7 and 8 rich in each component. The spectroscopic data for the two diastereomers
are given separately.
4.3.1. Diethyl (2R*,4S*)-2-hydroxy-4-methylpentanedioate 7
1
IR (®lm), ꢀ, cm^1: 3500 (OH), 1740 (COO), 1250, 1180 (C±O); H NMR, ꢁ, ppm: 4.25 (2H, q,
OCH2CH3), 4.22 (1H, m, H-2), 4.16 (2H, dq, OCH2CH3), 2.95 (1H, d, J=5.5 Hz, OH), 2.75 (1H,
m, H-4), 2.23 (1H, ddd, J=13.7, 9.7, 3.8 Hz, H-3), 1.66 (1H, ddd, J=13.7, 9.1, 4.4 Hz, H-3), 1.31,
1.27 (6H, 2 t, OCH2CH3), 1.21 (3H, d, J=7.0 Hz, CH3); 13C NMR, ꢁ, ppm: 176.2 (s), 174.9 (s),
68.8 (d), 61.8 (t), 60.5 (t), 38.0 (t), 35.7 (d), 17.8 (q), 14.2 (2 q); MS, m/z: 99 (100).
4.3.2. Diethyl (2R*,4R*)-2-hydroxy-4-methylpentanedioate 8
IR (®lm), ꢀ, cm^1: 3500 (OH), 1740 (COO), 1250, 1180 (C±O); 1H NMR, ꢁ, ppm: 4.24 (2H, dq,
OCH2CH3), 4.20 (1H, m, H-2), 4.13 (2H, q, OCH2CH3), 2.85 (1H, d, J=6.2 Hz, OH), 2.75 (1H,
m, H-4), 2.04 (1H, ddd, J=13.9, 9.5, 6.6 Hz, H-3), 1.85 (1H, ddd, J=13.9, 7.7, 3.8 Hz, H-3), 1.31,
1.25 (6H, 2 t, OCH2CH3), 1.23 (3H, d, J=7.3 Hz, CH3); 13C NMR, ꢁ, ppm: 176.3 (s), 174.9 (s),
68.6 (d), 61.8 (t), 60.5 (t), 37.7 (t), 35.9 (d), 16.7 (q), 14.1 (2 q); MS, m/z: 99 (100).
4.4. Synthesis of lactones 9 and 10
The mixture of 7 and 8 (2.0 g, 9.2 mmol) was re¯uxed in benzene in the presence of PTSA for 2
h. The solution was washed with a saturated solution of NaHCO3 and the organic phase dried on
anhydrous Na2SO4. After evaporation of the solvent, the lactones 9 and 10 were separated by
¯ash chromatography (eluant: light petroleum:ethyl acetate, in gradients from 100:0 to 80:20).
The trans-lactone 9 was obtained in 31% yield (0.50 g, 2.9 mmol) and the cis-lactone 10 was
obtained in 28% yield (0.44 g, 2.6 mmol).

S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
1363
4.4.1. Ethyl trans-4-methyl-tetrahydro-5-oxo-2-furancarboxylate 9
1
Oil; IR (®lm), ꢀ, cm^1: 1770 (COO), 1736 (COOEt); H NMR, ꢁ, ppm: 4.88 (1H, dd, J1=2.2,
J2=9.2, H-2), 4.26 (2H, q, J=7.1, OCH2CH3), 2.74 (1H, m, H-4), 2.55 (1H, ddd, J1=2.2,
J2=8.8, J3=13.2, H-3), 2.21 (1H, ddd, J1=9.2, J2=11.2, J3=13.2, H-3), 1.32 (3H, d, J=7.0,
CH3), 1.30 (3H, t, J=7.1, CH2CH3); 13C NMR, ꢁ, ppm: 178.8 (s), 169.5 (s), 73.7 (d), 62.0 (t), 34.4
(t), 34.2 (d), 15.2 (q), 14.0 (q); MS, m/z: 172 (M+Á, 1), 99 (100), 71 (39), 43 (52).
4.4.2. Ethyl cis-4-methyl-tetrahydro-5-oxo-2-furancarboxylate 10
1
Oil; IR (®lm), ꢀ, cm^1: 1770 (COO), 1736 (COOEt); H NMR, ꢁ, ppm: 4.80 (1H, dd, J1=7.1,
J2=9.3, H-2), 4.28 (2H, q, J=7.1, OCH2CH3), 2.75 (2H, m, H-3, H-4), 1.95 (1H, ddd,
J1=J2=9.2, J3=12.5, H-3), 1.32 (3H, t, J=7.1, CH2CH3), 1.32 (3H, d, J=7.0, CH3); 13C NMR,
ꢁ, ppm: 178.0 (s), 169.5 (s), 74.0 (d), 61.9 (t), 34.4 (d), 34.0 (t), 15.4 (q), 14.0 (q); MS, m/z: 172
(M+Á, 1), 99 (100), 71 (38), 43 (38).
4.5. General procedure for enzymatic hydrolyses of lactones 9 and 10
The lactonic ester was reacted with the enzyme in 0.1 M phosphate bu€er (pH 7.4) under
stirring. The pH value was continuously adjusted with 1N NaOH by the use of a pH-stat.
At the desired conversion value, the unreacted lactonic ester was extracted from the aqueous
phase with diethyl ether (four times). From the mother liquors, acidi®ed to pH 1 with 2N HCl,
the lactonic acid was extracted with diethyl ether (four times). When the enzyme opened the
heterocyclic ring to the corresponding hydroxy hemiester, this organic phase also contained the
lactonic ester formed by cyclization of this latter compound. The organic phase was then dried on
anhydrous Na2SO4.
The enantiomeric excesses of the lactonic esters and acids were determined by chiral HRGC
after esteri®cation of the lactonic acids with CH2N2.
4.5.1. Ethyl (2S,4R)-(+)-trans-4-methyl-tetrahydro-5-oxo-2-furancarboxylate 9
25
D
The lactonic ester was isolated with 98% e.e., ‰ꢂŠ +41.0 (c 0.10, MeOH); CD: Á"211 +1.3
(MeOH).
4.5.2. Ethyl (2S,4S)-(^)-cis-4-methyl-tetrahydro-5-oxo-2-furancarboxylate 10
25
D
The lactonic ester was isolated with 99% e.e., ‰ꢂŠ ^3.9 (c 0.18, MeOH); CD: Á"216 +1.4
(MeOH).
4.5.3. (2R,4S)-(^)-trans-4-Methyl-tetrahydro-5-oxo-2-furancarboxylic acid 112b
IR, ꢀ, cm^1: 3680, 3610, 3450 (OH), 1770 (COO), 1740 (COOH); 1H NMR, ꢁ, ppm: 9.00 (1H, s,
COOH), 4.94 (1H, dd, J1=2.3, J2=9.3, H-2), 2.80 (1H, m, H-4), 2.62 (1H, ddd, J1=2.3, J2=8.7,
J3=13.0, H-3), 2.25 (1H, ddd, J1=9.3, J2=10.6, J3=13.0, H-3), 1.29 (3H, d, J=6.9, CH3); 13C
NMR, ꢁ, ppm: 179.7 (s), 173.9 (s), 73.6 (d), 34.0 (t), 32.6 (d), 15.0 (q); MS, m/z: 145 (MH+, 35),
25
D
99 (100), 71 (62); 82% e.e., ‰ꢂŠ ±18.9 (c 0.65, MeOH); CD: Á"209 ^0.7 (MeOH).
4.5.4. (2R,4R)-(+)-cis-4-Methyl-tetrahydro-5-oxo-2-furancarboxylic acid 122b
IR, ꢀ, cm^1: 3680, 3610, 3450 (OH), 1770 (COO), 1740 (COOH); 1H NMR, ꢁ, ppm: 9.12 (1H, s,
COOH), 4.87 (1H, dd, J1=7.1, J2=9.2, H-2), 2.81 (2H, m, H-3, H-4), 2.00 (1H, ddd, J1=J2=9.2,
J3=11.7, H-3), 1.32 (3H, d, J=6.8, CH3); 13C NMR, ꢁ, ppm: 178.8 (s), 173.1 (s), 73.7 (d), 34.6

1364
S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
25
(d), 33.8 (t), 15.2 (q); MS, m/z: 145 (MH+, 100), 99 (63), 71 (69); 92% e.e., ‰ꢂŠ +3.2 (c 0.95,
D
MeOH); CD: Á"215 ^1.1 (MeOH).
4.6. Methylation of (R)-(^)-tetrahydro-5-oxo-2-furancarboxylic acid 13
The lactone of (R)-(^)-13 (0.8 g, 6.2 mmol) was methylated according to the literature.8 The
crude reaction mixture was treated with CH2N2 and puri®ed by ¯ash chromatography by which
the lactone (^)-14 (0.1 g, 0.6 mmol, 10% yield) was obtained.
4.6.1. Methyl (2R,4S)-(^)-trans-4-methyl-tetrahydro-5-oxo-2-furancarboxylate 14
IR, ꢀ, cm^1: 1770 (COO), 1736 (COO); 1H NMR, ꢁ, ppm: 4.92 (1H, dd, J1=2.2, J2=9.2, H-2),
3.81 (3H, s, OCH3), 2.74 (1H, m, H-4), 2.56 (1H, ddd, J1=2.2, J2=8.8, J3=13.2, H-3), 2.22 (1H,
ddd, J1=9.2, J2=11.0, J3=13.2, H-3), 1.29 (3H, d, J=7.0, CH3); 13C NMR, ꢁ, ppm: 178.7 (s),
170.5 (s), 73.5 (d), 52.7 (q), 34.1 (t), 32.4 (d), 15.1 (q); >99% e.e., ‰ꢂŠ25 ^40.0 (c 0.13, MeOH); CD:
D
Á"210 ^1.3 (MeOH).
4.7. Reduction of the g-lactones 9 and 10 with NaBH4
4.7.1. (3R,5S)-(+)-trans-5-Hydroxymethyl-3-methyl-4,5-dihydro-2(3H)-furanone 16
The trans-lactonic ester (+)-9 (0.150 g, 0.9 mmol) was reduced with NaBH4 following the
procedure reported in the literature1n to give the corresponding hydroxymethyl lactone 16. Its
1
IR, H NMR and 13C NMR spectra were in accordance with those reported in the literature.4a
25
D
25
D
Its enantiomeric excess was 98%; ‰ꢂŠ +29.7 (c 0.38, EtOH) [lit.4a ‰ꢂŠ +36.2 (c 0.3, EtOH)]; CD:
Á"216 ^0.9 (EtOH), [Â]216 ^3.0Â103 [lit.14 [Â]218 ^2.2Â103].
4.7.2. (3S,5S)-(+)-cis-5-Hydroxymethyl-3-methyl-4,5-dihydro-2(3H)-furanone 17
The cis-lactonic ester (^)-10 (0.150 g, 0.9 mmol) was reduced with NaBH4 according to the
1
literature.1n Its IR and H NMR spectra were in accordance with those reported in the litera-
25
D
ture.4a 13C NMR, ꢁ, ppm: 179.8 (s), 78.8 (d), 63.4 (t), 35.5 (d), 31.5 (t), 15.0 (q); 94% e.e., ‰ꢂŠ
25
+17.2 (c 0.9, EtOH) [lit.4a ‰ꢂŠ +22.5 (c 0.3, EtOH)]; CD: Á"220 +0.3 (EtOH), [Â]220 +1.0Â103
D
[lit.14 [Â]222 +0.2Â103].
4.8. Reductions of the ketodiester 6 with baker's yeast
A suspension of baker's yeast (46.0 g) in tap water (92 ml) was preincubated for di€erent periods
(Table 3). The ketodiester 6 (1.0 g, 4.6 mmol) was then added and the mixture stirred at room
temperature for 72 h.
The course of the reduction was monitored by HRGC and 1H NMR. At the end of the reaction,
brine was added and the broth was extracted with ethyl ether. The organic phase was dried and
evaporated and the products were separated by ¯ash chromatography (eluant: light petroleum:ethyl
acetate, in gradients from 95:5 to 80:20) to yield a mixture of the hydroxydiesters 7 and 8 (0.1 g, 0.46
mmol, 10% yield), and ethyl 2-hydroxy-4-methylbutanedioate 18 (0.08 g, 0.5 mmol, 12% yield).15
The mixture of the hydroxydiesters 7 and 8 was re¯uxed in benzene and PTSA for 30 min, then
the solution was washed with a saturated solution of NaHCO3 and dried on anhydrous Na2SO4.
Evaporation of the solvent gave the two lactones 9 and 10, respectively, which were separated by
¯ash chromatography and analyzed by chiral HRGC.

S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
1365
4.8.1. Ethyl 4-hydroxy-2-methylbutanedioate 1815
IR (®lm), ꢀ, cm^1: 3420 (OH), 1740, 1720 (COO), 1180 (C±O); H NMR, ꢁ, ppm: 4.14 (2H, q,
OCH2CH3), 3.69 (2H, m, 2 H-4), 2.62 (1H, ddq, J=8.2, 7.1, 5.8 Hz, H-2), 1.93 (1H, m, H-3), 1.70
(1H, m, H-3), 1.65 (1H, bs, OH), 1.26 (3H, t, OCH2CH3), 1.20 (3H, d, J=7.1 Hz, CH3); 13C
NMR, ꢁ, ppm: 176.8 (s), 60.7 (t), 60.4 (t), 36.5 (d), 36.3 (t), 17.2 (q), 14.2 (q).
1
4.9. Synthesis of model compounds 19 and 20 for the analysis of CD spectra
4.9.1. Ethyl (S)-(+)-tetrahydro-5-oxo-2-furancarboxylate 19
Esteri®cation of compound (S)-(+)-13 was carried out following the procedure used for peptide
coupling.10 Compound (S)-(+)-13 (Aldrich) (0.13 g, 1.0 mmol) was dissolved in CH2Cl2 (2 ml).
1-Hydroxybenzotriazole (0.135 g, 1.0 mmol) and N-ethyl-N0-(3-(dimethylamino)propyl)carbo-
diimide hydrochloride (EDC) (0.23 g, 1.2 mmol) were added under magnetic stirring. After a few
minutes, abs. EtOH (0.1 ml) and Et3N (0.17 ml, 1.2 mmol) were added. After 15 h the solvent was
evaporated, then the crude reaction mixture was dissolved in 100 ml of ethyl acetate and washed
twice with 40 ml of a solution of 0.5 M citric acid, H2O, 5% NaHCO3, H2O and dried on
Na2SO4. Evaporation of the solvent gave 0.1 g (0.63 mmol, 63% yield) of (S)-(+)-19. All spec-
25
25
D
troscopic data were in accordance with the literature;1b,n,16 ‰ꢂŠ +13.6 (c 0.56, EtOH) [lit.1b ‰ꢂŠ
D
25
D
25
D
+13.3 (c 0.56, EtOH), lit.1n ‰ꢂŠ +15.1 (c 0.6, EtOH), lit.16 ‰ꢂŠ +11.3 (c 10, EtOH)]; CD: Á"213
+1.7 (MeOH).
4.9.2. (S)-(+)-5-Hydroxymethyl-4,5-dihydro-2(3H)-furanone 20
Lactone 20 was prepared according to the literature.1n All spectroscopic data were in accor-
25
D
20
D
dance with the literature;1n ‰ꢂŠ +28.0 (c 0.3, EtOH) [lit.1n ‰ꢂŠ +29.6 (c 0.4, EtOH)]; CD: Á"211
^0.1 (EtOH), [Â]211 ±0.5Â103 [lit.14 [Â]210 ^0.7Â103].
Acknowledgements
The authors thank the MURST, CNR (Rome) and the University of Trieste for ®nancial support.
References
1. (a) Donohoe, T. J.; Stevenson, C. A.; Helliwell, M.; Irshad, R.; Ladduwahetty, T. Tetrahedron: Asymmetry 1999,
10, 1315±1322; (b) Drioli, S.; Nitti, P.; Pitacco, G.; Tossut, L.; Valentin, E. Tetrahedron: Asymmetry 1999, 10,
2713±2728; (c) Chandrasekharam, M.; Rai-Shung, L. J. Org. Chem. 1998, 63, 9122±9124; (d) Drioli, S.; Felluga,
F.; Forzato, C.; Nitti, P.; Pitacco, G.; Valentin, E. J. Org. Chem. 1998, 63, 2385±2388; (e) Ghattak, A.; Sarkar, S.;
Ghosh, S. Tetrahedron 1997, 53, 17 335±17 342; (f) Vaupel, A.; Knochel, P. J. Org. Chem. 1996, 61, 5743±5753; (g)
Sibi, M. P.; Deshpande, P. K.; La Loggia, A. J. Synlett. 1996, 343±345; (h) Maiti, G.; Roy, S. C. J. Chem. Soc.,
Perkin Trans. 1 1996, 403±404; (i) Zhu, G.; Lu, X. Tetrahedron: Asymmetry 1995, 6, 885±892; (j) Mawson, S. D.;
Weavers, R. T. Tetrahedron 1995, 51, 11 257±11 270; (k) Honda, T.; Kimura, N. J. Chem. Soc., Chem. Commun.
1994, 77±78; (l) de Azavedo, M. B. M.; Murta, M. M.; Greene, A. E. J. Org. Chem. 1992, 57, 4567±4569; (m)
Mori, K.; Yamane, K. Tetrahedron 1982, 38, 2919±2921; (n) Ravid, U.; Silverstein, R. M.; Smith, L. R.
Tetrahedron 1978, 34, 1449±1452; (o) Hussain, S. A. M. T.; Ollis, W. D.; Smith, C.; Stoddard, J. F. J. Chem. Soc.,
Perkin Trans. 1 1975, 1480±1492.
2. (a) Nishida, Y.; Ohrui, H.; Meguro, H. Agric. Biol. Chem. 1983, 47, 2123±2125; (b) Nishida, Y.; Ohrui, H.;
Meguro, H. Agric. Biol. Chem. 1984, 48, 1211±1216.

1366
S. Drioli et al. / Tetrahedron: Asymmetry 11 (2000) 1353±1366
3. Tsuneya, T.; Ishihara, M.; Shiota, H.; Shiga, M. Agric. Biol. Chem. 1980, 44, 957±958.
4. (a) Herdeis, C.; Lutsch, K. Tetrahedron: Asymmetry 1993, 4, 121±131; (b) Wilson, L. J.; Liotta, D. C. J. Org.
Chem. 1992, 57, 1948±1950; (c) Tomioka, K.; Cho, Y.-S.; Sato, F.; Koga, K. J. Org. Chem. 1988, 53, 4094±4098;
(d) Hanessian, S.; Murray, P. J.; Sahoo, S. P. Tetrahedron Lett. 1985, 26, 5623±5626.
5. Echalier, F.; Constant, O.; Bolte, J. J. Org. Chem. 1993, 58, 2747±2750.
6. Blaise, E. E.; Gault, H. Bull. Soc. Chim. Fr. 1911, 458±464.
7. Dowd, P.; Choi, S.-C.; Duah, F. Tetrahedron 1988, 44, 2137±2148.
8. Grieco, P. A.; Miyashita, M. J. Org. Chem. 1974, 39, 120±122.
9. Chen, C.-S.; Fujimoto, Y.; Girdaukauas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104, 7294±7299.
10. (a) Carpino, L. A. J. Am. Chem. Soc., 1993, 115, 4397±4398; (b) Konig, W.; Geiger, R. Chem. Ber. 1970 103, 788±
798.
11. Matsumoto, T.; Takahashi, M.; Kashihara, Y. Bull. Chem. Soc. Jpn. 1979, 52, 3329±3336.
12. Okuda, T.; Harigaya, S.; Kiyomoto, A. Chem. Pharm. Bull. 1964, 12, 504±506.
13. Forzato, C.; Nitti, P.; Pitacco, G. Tetrahedron: Asymmetry 1997, 8, 4101±4110.
14. Nishida, Y.; Ohrui, H.; Meguro, H. Croatica Chim. Acta 1989, 62(2B), 377±390.
15. (a) Kochinashvili, M. V.; Kuramshin, E. M.; Kotlyar, S. A.; Zlotskii, S. S.; Rakhmankulov, D. L. Zh. Vses. Khim.
O-va. im. D. I. Mendeleeva 1986, 31, 596±597; Chem. Abstr. 1987, 107, 216892t; (b) Sakykh-Zade, S. I.; Rzaeva,
A. S. Azerb. Khim. Zh. 1968, 6, 97±102; Chem. Abstr. 1969, 71, 90750n; (c) Boguslavskaya, L. S.; Razuvaev, G. A.
Zh. Obshch. Khim. 1963, 33, 2021±2027; Chem. Abstr. 1966, 64, 586a.
16. Doolittle, R. E.; Tumlinson, J. H.; Proveaux, A. T.; Heat, R. R. J. Chem. Ecol. 1980, 6, 473±485.