Enzymatic resolution of syn-2-azido-1,3,4-trihydroxybutane catalysed by lipases in the transesterification mode

Article abstract of DOI:10.1016/S0957-4166(00)00045-8

The enzymatic resolution of both syn-2-azido-1,3,4-trihydroxybutane 1 and syn-2-azido-1,4-diacetoxy-3-hydroxybutane 2 have been undertaken with different lipases as catalysts and vinyl acetate as acylating agent. Lipases Amano PS and Amano AK proved to be the superior catalysts for this resolution. Indeed, both enantiomers of 1 are easily available in good yields and very good e.e.s (up to >99%). The use of chiral HPLC with a Chiralcel OD- H column allowed the determination of e.e.s of both diacetate 2 and triacetate 3 (syn-2-azido-1,3,4-triacetoxybutane) in a single analysis and thus facilitated the precise control of the reaction. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00045-8

Tetrahedron: Asymmetry 11 (2000) 1313±1321
Enzymatic resolution of syn-2-azido-1,3,4-trihydroxybutane
catalysed by lipases in the transesteri®cation mode
G. Iacazio,a,* D. Martini,a S. Sancheza and B. Faureb
a
Â
 Ã
Laboratoire de Microbiologie, CNRS ERS 6100, Faculte des Sciences de St Jerome, Avenue Escadrille Normandie-Niemen,
13397 Marseille Cedex 20, France
b
Â
Laboratoire de Chimie de Coordination, Universite de Provence, Avenue Escadrille Normandie-Niemen,
13397 Marseille Cedex 20, France
Received 20 January 2000; accepted 26 January 2000
Abstract
The enzymatic resolution of both syn-2-azido-1,3,4-trihydroxybutane 1 and syn-2-azido-1,4-diacetoxy-3-
hydroxybutane 2 have been undertaken with di€erent lipases as catalysts and vinyl acetate as acylating
agent. Lipases Amano PS and Amano AK proved to be the superior catalysts for this resolution. Indeed,
both enantiomers of 1 are easily available in good yields and very good e.e.s (up to >99%). The use of
chiral HPLC with a Chiralcel OD-H column allowed the determination of e.e.s of both diacetate 2 and
triacetate 3 (syn-2-azido-1,3,4-triacetoxybutane) in a single analysis and thus facilitated the precise control
of the reaction. # 2000 Elsevier Science Ltd. All rights reserved.
1. Introduction
Until recently, the `chiral pool' was the main and simplest source of non-racemic materials
available to chemists to perform asymmetric synthesis or asymmetric catalysis. In recent years the
extraordinary development of both asymmetric catalysis1 and bioconversion2 has given chemists
the opportunity to access new chiral synthons out of the `chiral pool'. In this connection we were
interested in the resolution of syn-2-azido-1,3,4-trihydroxybutane 1 by bioconversion, using lipases
in the transesteri®cation mode. Indeed, 1 is a multifunctional small molecule, readily available in
racemic form through simple chemistry, starting from commercial cis-2,3-butene-1,4-diol
(Scheme 1). Enantiomerically enriched derivatives of 1 have already been described: they have
been obtained, for example, from transformations of ascorbic acid3 or monoprotected optically
active cis-1,4-dihydroxy-2,3-butene oxide.4 The latter could be generated through Sharpless
asymmetric epoxidation4 or by bioconversion.5,6 In the ®rst case the transformation of ascorbic
acid is tedious and, in the second case, a mixture of isomers is formed when opening the epoxide
* Corresponding author. Tel: 33-4-91-28-87-68; fax: 33-4-91-28-80-30; e-mail: gilles.iacazio@microbio.u-3mrs.fr
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00045-8

1314
G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
Scheme 1. Synthesis of syn-2-azido-1,3,4-trihydroxy-butane 1
functionality with NaN3 unless some regiocontrol could be exerted in the azidation process. It
should also be noted that the enzymatic resolution of related 2,3-azido alcohols has been reported
previously.7 We believed that if a method could be developed to access both enantiomers of 1,
such a molecule would be a valuable chiral building block in the synthesis, for example, of
sphingosines and aza-sugars.
2. Results and discussion
2.1. Synthesis of 1
As mentioned in the introduction, 1 is easily available from cis-2,3-butene-1,4-diol through
simple chemistry and in good yield (84%), as depicted in Scheme 1.
2.2. Enzymatic access to racemic syn-2-azido-1,4-diacetoxy-3-hydroxy-butane 2
In a ®rst attempt to realise the enzymatic resolution of 1, we made the observation that porcine
pancreatic lipase (PPL) was able to convert (+/^)-1 to the corresponding racemic 1,4-diacetate 2,
with vinyl acetate as both solvent and acyl donor, in very good yield (Scheme 2). This reaction
proved to be superior to its chemical counterpart in terms of regioselectivity since in our hands
the chemical acetylation of (+/^)-1 led to a mixture of mono-, di- and triacetates of 1.
Scheme 2. Enzymatic synthesis of (+/^)-2 with PPL
No triacetate 3 could be detected, indicating that 2 is not a substrate of PPL. This result
prompted us to investigate the resolution of both (+/^)-1 and enzymatically formed (+/^)-2.
2.3. Enzymatic resolution of diacetate 2
Three lipases were tested, i.e. lipase Amano AK, lipase Amano PS and lipase Amano AY,
respectively, from bacteria Pseudomonas sp. (both AK and PS) and from yeast Candida rugosa
(formerly C. cylindracea). In the three cases vinyl acetate was used both as solvent and acyl donor
and the in¯uence of added co-solvents such as benzene and THF was also investigated. The

G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
1315
reactions were followed by chiral HPLC using a Chiralcel OD-H column. Both diacetate 2 and
triacetate 3 are simultaneously resolved by this column using a mixture of hexane/isopropanol
(95/5, vol./vol.) as eluent, delivered at a ¯ow rate of 0.8 mL/min.8 The results obtained are
grouped in Table 1 (entries 1 to 5).
Table 1
Enzymatic resolution of diacetate 2 and triol 1
It can be seen from Table 1 (entries 1 to 3) that lipase Amano AK is the most successful enzyme
in the resolution of (+/^)-2 (E=26). In order to enhance the selectivity of lipase Amano AK, we
have investigated the possible role of benzene and THF as co-solvent3 in the reaction. From
Table 2 (entries 4 and 5) it could be concluded that the use of THF as co-solvent has a negative
impact on the selectivity of lipase Amano AK (E=15). With benzene a slightly better enantio-
meric factor is obtained (E=29) and the reaction is twice as fast.
2.4. Enzymatic resolution of triol 1
Following the resolution of diacetate 2, we are interested in that of the parent compound, triol
1. For this study, only lipases Amano PS and AK were selected and the reactions run either with
only vinyl acetate as solvent or in a 1/1 mixture with benzene. The results are grouped in Table 1
(entries 6 to 9). It is worth mentioning that in all cases, at the end of the reaction, only diacetate 2
and triacetate 3 are detected either by TLC or chiral HPLC. Also, the reaction times are higher

1316
G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
Table 2
Enzymatic preparative resolution of triol 1 and diacetate 2
when 1 is used as substrate (probably due to lower solubility); improved selectivities are obtained
for reactions involving lipase Amano AK without co-solvent and for lipase Amano PS with
added benzene (entries 6 and 9).
2.5. Enzymatic preparative resolutions of 2 and 1
Following the results described in Table 1, we have undertaken the enzymatic preparative
resolution of diacetate 2 and triol 1, according to the best procedures developed previously. The
reactions have been done on a 1 gram scale, using lipases Amano PS or AK in a 1/1 mixture of
vinyl acetate and benzene and followed by chiral HPLC chromatography. When the desired e.e.s
were reached, puri®cation of both 2 and 3 has been undertaken and speci®c rotations determined,
as well as the absolute con®gurations of enantiomerically enriched 2 and 3. The results are
collected in Table 2.
The results depicted in Table 2 indicate that (2S,3R)-2 is easily available in high e.e.s (up to
>99%) and satisfactory yields, using either Amano AK or PS lipases as catalysts and triol 1 or
diacetate 2 as substrates. On the other hand, if (2R,3S)-3 is to be obtained with good e.e. and
good yield, lipase Amano AK is the catalyst of choice with triol 1 as substrate.
2.6. Determination of absolute con®gurations
In 1994, Fuji et al.9 have described, in a work devoted to the synthesis of enantiomerically pure
2-aziridinemethanols, the preparation and absolute con®guration of (2S,3R)-2-azido-1-benzyl-
oxy-3,4-diacetoxybutane 6 (Scheme 3). Compound 6 could easily be obtained from 2 or 3 by the
reaction sequence depicted in Scheme 3.
In order to determine the absolute con®gurations of enzymatically formed 2 and 3, we have
20
20
D
®rst transformed (+)-2 (e.e. 57%, ‰ꢀŠ +7.8 (c=1.50, CHCl3)) into (^)-3 (e.e. 57%, ‰ꢀŠ ^7.2
D
(c=2.89, CHCl3)) by acetylation with acetic anhydride and then, following the sequence depicted

G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
1317
Scheme 3. Conversion of diacetate 2 or triacetate 3 to compound 6
20
D
in Scheme 3, into (+)-6 (e.e. 59%,10 ‰ꢀŠ +6.7 (c=2.81, CHCl3)). The positive sign of the speci®c
rotation thus obtained for compound 6 allowed the assignment of its absolute con®guration as
(2S,3R)-2-azido-1-benzyloxy-3,4-diacetoxybutane9 and hence that of starting (+)-2, that is also
(2S,3R). This means that this enantiomer of 2 is the ®rst eluted one on Chiralcel OD-H column.8
By the way, the ®rst eluted enantiomer of 3,8 having a negative optical rotation sign, has the same
(2S,3R) absolute con®guration.
3. Conclusion
By using very simple chemistry followed by lipase resolution we were able to access, on a
preparative scale, both enantiomers of triol 1, with very high e.e.s (up to>99%) and good yields.
We believe that such a small multifunctional chiral synthon could have a lot of synthetic applications,
some of which have already been described.3±5 Indeed, as shown in Scheme 3, the vic-diol func-
tionality is easily di€erentiated from the third alcohol functionality using 3-pentanone. After
suitable protection of the remaining alcohol function, the acetal could be released, leaving one
primary alcohol and one secondary alcohol, that are thus easily di€erentiated chemically.
4. Experimental
4.1. General methods
Porcine pancreatic lipase was from Sigma. Lipases AK, AY and PS were a gift from Amano.
cis-2-Buten-1,4-diol, m-CPBA and Amberlyst 15 were from Fluka. Vinyl acetate, p-toluene-
sulfonic acid, 3-pentanone, sodium methoxide (0.5 M in methanol), sodium hydride, benzyl
bromide and acetic anhydride were from Aldrich. NMR spectra were run on a Bruker AC 200
spectrometer with TMS as internal standard. Chemical shifts are given in ppm down®eld of TMS
(s=singlet, d=doublet, dd=doublet of doublets, dt=doublet of triplets, t=triplet, q=quadruplet,

1318
G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
m=multiplet). Optical rotations were measured on a Perkin±Elmer 341 polarimeter. Chiral
HPLC analyses were run with a Chiralcel OD-H column, using a 95/5 (hexane/2-propanol) mix-
ture as eluent, delivered at 0.8 mL/min. UV detection occurred at l=214 nm. Enzymatic reac-
tions were carried out in screw cap tubes. The tubes were incubated at 30ꢀC, on a reciprocal
shaker, working at 60 oscillations per minute.
4.2. Synthesis of 1,4-dihydroxy-cis-2,3-butene oxide
In a 0.5 L round-bottomed ¯ask, 2.2 g (25 mmol) of 1,4-dihydroxy-cis-2,3-butene were dis-
solved in 50 mL of CH2Cl2. Then 9.5 g (27.5 mmol, 1.1 equiv.) of m-CPBA dissolved in 100 mL
of CH2Cl2 were added slowly. At the end of the addition, the reaction mixture was vigorously
stirred and brought to re¯ux. The progress of the reaction was followed by TLC with a 7/3
CH2Cl2/MeOH eluent. Consumption of 1,4-dihydroxy-cis-2,3-butene was complete after 4 h. The
reaction mixture was then poured on a silica column equilibrated with Et2O. Unreacted m-CPBA
and formed m-CBA were ®rst eluted with 2% acetic acid in Et2O. The product was then eluted
with an 8/2 mixture of CH2Cl2/MeOH a€ording 2.33 g (23 mmol) of 1,4-dihydroxy-cis-2,3-butene
ꢀ
1
oxide as a white amorphous solid in 92% yield, mp 38±40 C. H NMR (200 MHz, CD3OD) ꢁ
3.11±3.19 (2H, m), 3.56±3.80 (4H, m), 4.73 (2H, br. s). 13C NMR (50 MHz, CD3OD) ꢁ 57.7 (CH),
61.1 (CH2).
4.3. Synthesis of syn-2-azido-1,3,4-trihydroxy-butane 1
In a 0.25 L round-bottomed ¯ask, 2.25 g (22 mmol) of 1,4-dihydroxy-cis-2,3-butene oxide were
dissolved in 90 mL of an 8/1 methanol/water mixture. Then 2.8 g (44 mmol, 2 equiv.) of NaN3
and 5.8 g (110 mmol, 5 equiv.) of NH4Cl were added and the reaction brought to re¯ux. The
progress of the reaction was followed by TLC (CH2Cl2/MeOH, 7/3). After 24 h, the starting
epoxide was totally consumed. The salts were then ®ltered and the product ®rst eluted on a silica
column using methanol as eluent. The fractions containing 1 were collected, evaporated and
taken to dryness with P2O5 under vacuum. This crude mixture was then puri®ed by ¯ash chro-
matography on silica gel with a CH2Cl2/MeOH mixture (7/3) a€ording 2.9 g (20 mmol) of 1 as a
1
colourless oil (91% yield). H NMR (200 MHz, CD3OD) ꢁ 3.45 (1H, br. s), 3.52 (1H, m), 3.62
(1H, m), 3.70 (2H, m), 4.36 (2H, br. s). 13C NMR (50 MHz, CD3OD) ꢁ 62.8 (CH), 64.0 (CH),
65.4 (CH2), 72.4 (CH2).
4.4. Enzymatic synthesis of racemic diacetate 2
In a screw cap tube were placed 1 g of 1 (6.8 mmol), 10 mL of dry THF and 20 mL of vinyl
acetate. Then 1 g of PPL was added and the tube was incubated at 30ꢀC, on a reciprocal shaker,
working at 60 oscillations per minute. The reaction progress was followed by TLC (CH2Cl2/
MeOH, 95/5). After two days of reaction, 500 mg of PPL were further added. After one week of
reaction the lipase was ®ltered, the solvents evaporated and the product ®nally puri®ed over
silica (CH2Cl2/Et2O, 80/20) a€ording 1.48 g (6.4 mmol) of (+/^)-2 (94% yield) as a white solid,
mp 44±45ꢀC. 1H NMR (200 MHz, CDCl3) ꢁ 2.12 (6H, s), 2.83 (1H, br. s), 3.73 (1H, ddd, J=7.8,
4.9, 3.6 Hz), 3.97 (1H, m), 4.20 (2H, m), 4.30 (1H, dd, J=11.6, 7.8 Hz), 4.40 (1H, dd, J=11.6, 4.9
Hz). 13C NMR (50 MHz, CDCl3) ꢁ 20.7 (CH3), 20.8 (CH3), 61.4 (CH), 63.8 (CH2), 65.3 (CH2),
69.1 (CH), 170.8 (CO), 171.1 (CO).

G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
1319
4.5. Preparative enzymatic resolution of (+/^)-1 and (+/^)-2
In a screw cap tube were placed 1 g (6.8 mmol) of 1 or 1 g (4.3 mmol) of 2, 20 mL of vinyl
acetate and 20 mL of benzene (1) or 40 ml of vinyl acetate (2). One gram of lipase was then added
and the tube incubated at 30ꢀC, on a reciprocal shaker, working at 60 oscillations per minute.
The reaction progress was followed by chiral HPLC. When the desired e.e.s were obtained, the
enzyme was ®ltered and diacetate 2 and triacetate 3 were puri®ed by ¯ash column chromato-
graphy over silica, using a 9/1 to 3/7 gradient of petroleum ether (low bp) and diethyl ether. See
Table 2 for more details upon yields, e.e.s and speci®c rotations. Compound 3 (colourless oil): 1H
NMR (200 MHz, CDCl3) ꢁ 2.08 (3H, s), 2.11 (3H, s), 2.13 (3H, s), 3.87 (1H, ddd, J=7.2, 4.7, 4.7
Hz), 4.15 (2H, m), 4.31 (1H, dd, J=7.2, 4.7 Hz), 4.34 (1H, dd, J=7.2, 4.7 Hz), 5.22 (1H, ddd,
J=6.0, 4.7, 4.7 Hz). 13C NMR (50 MHz, CDCl3) ꢁ 20.7 (CH3), 59.6 (CH), 62.3 (CH2), 63.0
(CH2), 70.0 (CH), 169.9 (CO), 170.4 (CO).
4.6. Chemical acetylation of diacetate (+)-2
In a 50 mL two necked round-bottomed ¯ask equipped with a re¯ux condenser were placed
20
D
494 mg (2 mmol) of enzymatically generated (+)-2 (e.e. 57%, ‰ꢀŠ +7.8 (c=1.50, CHCl3)). Then
224 mg of acetic anhydride (2.2 mmol, 1.1 equiv.), 1 mL of triethylamine and 20 mL of anhy-
drous dichloromethane were added and the reaction mixture was brought to re¯ux under mag-
netic agitation. After overnight re¯uxing the solvent and triethylamine were ®rst evaporated
under reduced pressure and the crude residual mixture puri®ed over silica using a 7/3 mixture of
petroleum ether (low bp)/diethyl ether as eluent. 519 mg of triacetate 3 were thus recovered (1.9
20
D
standing. Same spectral characteristics as reported above.
mmol, 95% yield, ‰ꢀŠ ^7.2 (c=2.89, CHCl3)) as a colourless oil which partially solidi®ed upon
4.7. Chemical deacetylation of triacetate (^)-3
In a two necked 50 mL round-bottomed ¯ask equipped with a magnetic stirrer were placed 450 mg
(1.65 mmol) of triacetate (^)-3. Dry methanol (20 mL) was added and the ¯ask placed under inert
atmosphere (N2). Then 3.6 mL of a 0.5 M NaOMe solution in methanol (1.8 mmol) were added and
the reaction placed under magnetic agitation at 20ꢀC. After 1 h of reaction, methanol was
removed under reduced pressure and the crude residual mixture puri®ed over silica using a 7/3
mixture of dichloromethane/methanol as eluent a€ording 219 mg of (+)-1 (1.49 mmol, 90% yield,
20
D
‰ꢀŠ +12.4 (c=1.31, MeOH)) as a colourless oil. Same spectral characteristics as reported above.
4.8. Synthesis of syn-2-azido-1-hydroxy-3,4-(pentane-3,3-diyloxy)-butane 4
In a 50 mL round-bottomed ¯ask were placed 200 mg (1.36 mmol) of (+)-1, 20 mL of freshly
distillated 3-pentanone, 10 mg of p-toluenesulfonic acid and 1 g of molecular sieves. The reaction
was then warmed at 80ꢀC under magnetic agitation for 3 days. The reaction mixture was then
®ltered, 3-pentanone evaporated under reduced pressure and the crude residue puri®ed over
silica using a 9/1 to 7/3 CH2Cl2/MeOH gradient of elution a€ording 216 mg of acetal (+)-4 (1.0
20
D
mmol, 76% yield, ‰ꢀŠ +9.1 (c=1.05, CHCl3)) as a colourless oil. 1H NMR (200 MHz, CDCl3) ꢁ
0.90 (3H, t, J=7.4 Hz), 0.93 (3H, t, J=7.3 Hz), 1.64 (2H, q, J=7.3 Hz), 1.70 (2H, q, J=7.4 Hz),
2.86 (1H, br. s), 3.47 (1H, dd, J=11.0, 5.8 Hz), 3.6±3.8 (3H, m), 4.07 (1H, dd, J=8.1, 6.4 Hz),

1320
G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
4.22 (1H, ddd, J=8.3, 6.3, 6.3 Hz). 13C NMR (50 MHz, CDCl3) ꢁ 8.1 (CH3), 28.9 (CH2), 29.5
(CH2), 32.5 (CH2), 64.6 (CH2), 63.0 (CH), 66.7 (CH2), 76.4 (CH), 113.9 (C).
4.9. Synthesis of syn-2-azido-1-benzyloxy-3,4-(pentane-3,3-diyloxy)-butane 5
In a two necked 50 ml round-bottomed ¯ask were placed under nitrogen 190 mg (0.88 mmol)
of (+)-4, 10 ml of dry DMF, 23 mg of NaH (0.97 mmol, 1.1 equiv.) and 165 mg of benzyl bromide
(0.97 mmol, 1.1 equiv.). The reaction mixture was left overnight under magnetic agitation at
20ꢀC. The reaction was then diluted with 100 mL of water and extracted three times with 100 mL
of diethyl ether. The mixed organic phases were washed twice with water (100 mL), dried over
sodium sulfate, ®ltered and evaporated under reduced pressure. The crude mixture thus obtained
was puri®ed over silica using a 9/1 to 7/3 gradient of petroleum ether (low bp)/diethyl ether
20
a€ording 235 mg of (+)-5 (0.77 mmol, 87% yield, ‰ꢀŠ +9.7 (c=0.85, CHCl3)) as a colourless oil.
D
1H NMR (200 MHz, CDCl3) ꢁ 0.87 (3H, t, J=7.4 Hz), 0.94 (3H, t, J=7.3 Hz), 1.64 (2H, q,
J=7.4 Hz), 1.69 (2H, q, J=7.3 Hz), 3.50±3.61 (3H, m), 3.75 (1H, dd, J=7.9, 7.9 Hz), 4.01 (1H,
dd, J=8.1, 6.4 Hz), 4.16 (1H, ddd, J=7.5, 6.0, 6.0 Hz), 4.54 (2H, s), 7.33 (5H, s). 13C NMR (50
MHz, CDCl3) ꢁ 8.1 (CH3), 29.1 (CH2), 29.6 (CH2), 62.5 (CH), 66.8 (CH2), 69.8 (CH2), 73.6
(CH2), 76.3 (CH), 113.6 (C), 127.7 (CH), 127.9 (CH), 128.5 (CH), 137.5 (C).
4.10. Synthesis of syn-2-azido-1-benzyloxy-3,4-diacetoxy-butane 6
In a 50 mL round-bottomed ¯ask equipped with a re¯ux condenser were placed 200 mg of (+)-5
(0.85 mmol), 20 mL of MeOH and 200 mg of Amberlyst 15. The reaction was then brought to re¯ux
for 3 days. After total deprotection of the acetal (TLC analysis) the reaction mixture was ®ltered and
evaporated to dryness under reduced pressure. The crude residue thus obtained was then acetylated
overnight with acetic anhydride (191 mg, 1.87 mmol, 2.2 equiv.) in 10 mL of dry dichloromethane
containing 0.5 mL of triethylamine under re¯uxing conditions. Following completion of the reaction,
the mixture was evaporated to dryness under reduced pressure and the crude residue puri®ed over
silica using a 1/1 mixture of petroleum ether (low bp)/diethyl ether a€ording 247 mg of (+)-6 (0.77
20
D
mmol, 90% yield, two steps, ‰ꢀŠ +6.7 (c=2.81, CHCl3)) as a colourless oil. 1H NMR (200 MHz,
CDCl3) ꢁ 2.03 (3H, s), 2.05 (3H, s), 3.60 (2H, m), 3.70 (1H, ddd, J=6.8, 4.4, 4.4 Hz), 4.12 (1H,
dd, J=11.9, 6.2 Hz), 4.32 (1H, dd, J=11.9, 4.4 Hz), 4.54 (2H, s), 5.24 (1H, ddd, J=6.1, 4.5, 4.5 Hz),
7.27±7.36 (5H, m). 13C NMR (50 MHz, CDCl3) ꢁ 20.7 (CH3), 60.5 (CH), 62.6 (CH2), 69.3 (CH2),
70.3 (CH), 73.6 (CH2), 127.8 (CH), 128.0 (CH), 128.5 (CH), 137.3 (C), 169.9 (CO), 170.4 (CO).
Acknowledgements
We are indebted to Dr J.-M. Brunel for helpful discussions and to Amano Ltd for the kind
supply of various lipases.
References
1. For general reviews on asymmetric catalysis, see: (a) Dawson, G. J.; Bower, J. F.; Williams, J. M. J. Contemp. Org.
Synth. 1996, 277. (b) Regan, A. C. J. Chem. Soc., Perkin Trans. 1 1998, 1151. (c) Wills, M. J. Chem. Soc., Perkin
Trans. 1 1998, 3101.

G. Iacazio et al. / Tetrahedron: Asymmetry 11 (2000) 1313±1321
1321
2. For general reviews on bioconversion, see: (a) Jones, J. B. Tetrahedron 1986, 42, 3403. (b) Santaniello, E.;
Ferraboshi, P.; Grisenti, P.; Manzocchi, A. Chem. Rev. 1992, 92, 1071. (c) Azerad, R. Bull. Soc. Chim. Fr. 1995,
132, 17. (d) Roberts, S. M. J. Chem. Soc., Perkin Trans. 1 1998, 157.
3. Kim, K. S.; Cho, I. H.; Ahn, Y. H.; Park, J. I. J. Chem. Soc., Perkin Trans. 1 1995, 1783.
4. Holt, K. E.; Leeper, F. J.; Handa, S. J. Chem. Soc., Perkin Trans. 1 1994, 231.
5. Grandjean, D.; Pale, P.; Chuche, J. Tetrahedron Lett. 1991, 32, 3043.
6. Moseley, J. D.; Staunton, J. Tetrahedron: Asymmetry 1998, 9, 3619.
7. Gruber-Khadjawi, M.; Honig, H.; Weber, H. Chirality 1992, 4, 103.
8. Chiral HPLC chromatogram trace of (+/^)-2 and (+/^)-3. Column Chiralcel OD-H, eluent 95/5 hexane/
isopropanol delivered at 0.8 mL/min, UV detection at 214 nm. Retention times: (2S,3R)-(^)-3 20.4 min, (2R,3S)-
(+)-3 23.4 min, (2S,3R)-(+)-2 31.5 min, (2R,3S)-(^)-2 36.0 min.
9. Fuji, N.; Nakai, K.; Habashita, H.; Hotta, Y.; Tamamura, H.; Otaka, A.; Ibuka, T. Chem. Pharm. Bull. 1994, 42,
2241.
10. The e.e. of the chemically synthesised 6 was determined based on the comparison of its speci®c rotation with that
20
of the same compound having an e.e. of 88%, ‰ꢀŠ +9.73 (c=0.781, CHCl3).4
D