Candida antarctica lipase A in the dynamic resolution of novel furylbenzotiazol-based cyanohydrin acetates

Article abstract of DOI:10.1016/S0957-4166(03)00025-9

A series of novel (R)-furylbenzotiazol-based cyanohydrin acetates were prepared in over 90% isolated yields from the corresponding furancarbaldehydes. The one-pot method combines a basic resin to produce hydrogen cyanide from acetone cyanohydrin, an equil

Full text of DOI:10.1016/S0957-4166(03)00025-9

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 619–627
Candida antarctica lipase A in the dynamic resolution of novel
furylbenzotiazol-based cyanohydrin acetates
Csaba Paizs,a,b Monica Tos¸a,b Cornelia Majdik,b Petri Ta¨htinen,a Florin Dan Irimieb and
Liisa T. Kanervaa,*
aLaboratory of Synthetic Drug Chemistry and Department of Chemistry, University of Turku, Lemminka¨isenkatu 2,
FIN-20520 Turku, Finland
bDepartment of Biochemistry and Biochemical Engineering, Babes¸-Bolyai University, Arany Ja´nos 11,
RO-3400 Cluj-Napoca, Romania
Received 4 December 2002; revised 20 December 2002; accepted 31 December 2002
Abstract—A series of novel (R)-furylbenzotiazol-based cyanohydrin acetates were prepared in over 90% isolated yields from the
corresponding furancarbaldehydes. The one-pot method combines a basic resin to produce hydrogen cyanide from acetone
cyanohydrin, an equilibrium between the formation and decomposition of furylbenzotiazol-based cyanohydrins and the unique
enantioselectivity of Candida antarctica lipase A, allowing the acylation of (R)-cyanohydrins in the presence of vinyl acetate in
anhydrous acetonitrile. © 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
highly enantiopure, contain reactive functional groups
through which a building block can be attached to
other structural components of a final drug molecule
and allow an easy transformation of the intermediate to
other types of compound. Optically active cyanohy-
drins can fulfil all these criteria, the criterium for high
enantiopurity being most challenging. We earlier
worked with the preparation of highly enantiopure
Heterocyclic ring systems can be commonly found in
the active components of many medicines. This is not
surprising since these systems present a diverse array of
pharmacophores, acting as hydrogen bond donors and
acceptors and also providing positive ions. Substitution
of the parent heterocyclic ring with an aromatic or
heteroaromatic ring system can further increase phar-
macophoric diversity. Accordingly, there is a need for
general heterocyclic building blocks and intermediates
which fulfil one or more of the following criteria: are
(R)-5-phenylfuran-2-ylcyanomethyl
butanoates
1
through Pseudomonas cepacia lipase (lipase PS)
catalysed dynamic kinetic resolution (Scheme 1).1 Still
based on the furyl ring as the parent heterocyclic ring,
Scheme 1.
* Corresponding author. E-mail: lkanerva@utu.fi
0957-4166/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0957-4166(03)00025-9

620
C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
the focus herein is on the preparation of highly enan-
tiopure furylbenzotiazol-based cyanohydrin acetates 2.
cyanohydrins and the stability of cyanohydrin esters.
Traditional kinetic resolution provides the enantioselec-
tivity basis for successful dynamic kinetic resolution.
Accordingly, the lipase-catalysed acylation of novel
cyanohydrins 4a–d was first studied (Scheme 3).
Currently, the lipase-catalysed kinetic resolution of
racemic compounds with various functionalities has
become an important method for the preparation of
enantiopure compounds.2 When the method is success-
ful both enantiomers of a racemate at ca. 50% theoret-
ical yields are simultaneously available for drug
development. However, it can occur that the other
enantiomer exerts a different, even harmful, physiologi-
cal effect or shows lowered or no activity. Accordingly,
there is a final bias toward the preparation of only one
of the enantiomers, leaving behind a question about
what to do with the undesired enantiomer of the kinetic
resolution. To solve this problem, the principle of
dynamic kinetic resolution has been previously
exploited in an efficient one-pot synthesis of optically
active cyanohydrin esters in the presence of a basic
resin and a lipase.1,3–5 Another elegant way has been
the acylation of the free cyanohydrin enantiomer in a
resolved mixture (one enantiomer as a cyanohydrin and
the other as a cyanohydrin ester) under the Mitsunobu
conditions, leading to the inversion of configuration
and accordingly to the desired ester enantiomer with
100% theoretical yield when calculated according to the
racemic starting material.6 In the case of labile
cyanohydrin substrates, however, a decrease in enan-
tiopurity can result under the Mitsunobu conditions.
2. Results and discussion
2.1. Kinetic resolution by acylation
A large number of commercially available lipases
exhibit a high degree of substrate selectivity, providing
products with various levels of enantioselectivity.
Potential lipases were screened for the enantioselective
acylation of racemic cyanohydrin 4a as a model sub-
strate in vinyl acetate. Most of the lipases, including
lipase PS (previously shown to have the most
potential1,3,4), were catalytically inactive. CAL-B with
minor reactivity (time needed to reach a certain conver-
sion) and enantioselectivity (Table 1; entry 1) was also
useless although successfully used for the dynamic reso-
lution of mandelonitrile.5 Only CAL-A was highly
active and catalysed the reaction enantioselectively (E=
23 0.4; entry 4). It is interesting to recognize that also
in previous work this enzyme provided an excellent,
very fast and environmentally benign method for the
preparation of racemic cyanohydrin esters 1.1 The
enzyme has only rarely been applied to enantioselective
reactions.7 There are, however, some interesting appli-
cations with sterically hindered substrates, including the
highly enantioselective acylations of b-amino esters,8–10
methyl pipecolinate11 and 1-phenylethan-1,2-diol.12 Evi-
dently, the size of compounds 2 and/or the heteroaro-
matic structure of the benzotiazol ring enhance
enzymatic enantioselectivity over that of compounds 1
as substrates. For good catalytic activity and enantiose-
lectivity of CAL-A, it has been essential to use the
enzyme adsorbed on Celite in the presence of sucrose
(enzyme preparation).13 In the present work, the reac-
tion of 4a with vinyl acetate in acetonitrile yielded an
Herein we now report the preparation of novel (R)-
cyanohydrin acetates 2a–d from the corresponding
aldehydes 3a–d and acetone cyanohydrin (source of
HCN) through effective lipase-catalysed dynamic
kinetic resolution of cyanohydrins 4a–d with vinyl ace-
tate (achiral acyl donor) in the presence of Amberlite
IRA-904 basic resin in organic solvents (Scheme 2).
Vinyl acetate, one of the most commonly used acyl
donors was successfully used, although it had been
previously emphasized that only isopropenyl acetate
was usable.4 In this work, use is made of the lability of
Scheme 2. One-pot synthesis of (R)-(+)-2a–d by dynamic resolution.

C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
621
Scheme 3. Kinetic resolution of racemic cyanohydrins 4a–d.
Table 1. Enzymatic acylation of racemic cyanohydrin 4a (13 mM)
Entry
Enzyme (mg ml−1
)
Solvent
Acyl donor (mM)
Time (h)
Conv. (%)
E
1
2
3
4
5
6
7
8
CAL-B (50)
CAL-B (50)
CAL-B (50)
CAL-A (10)
CAL-A (10)
CAL-A (10)
CAL-A (10)
CAL-A (10)
CAL-A (5)
CAL-A (20)
CAL-A (50)
Vinyl acetate
Ethyl acetate
Acetonitrile
Vinyl acetate
Ethyl acetate
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
Acetonitrile
Vinyl acetate
Ethyl acetate
Vinyl acetate (26)
Vinyl acetate
Ethyl acetate
Vinyl acetate (26)
Vinyl acetate (39)
Vinyl acetate (52)
Vinyl acetate (26)
Vinyl acetate (26)
Vinyl acetate (26)
168
168
168
12
168
24
24
24
48
12
8
0
390.1
–
23
48
12
44
47
48
45
43
51
590.1
2390.4
890.1
6292
6191
6291
6492
2490.7
990.3
9
10
11
5
almost racemic product at the initial rate of wo=0.07
mmol mg−1 h−1 in the presence of the native enzyme,
whereas wo=0.4 mmol mg−1 h−1 resulted when the
enzyme preparation was used.
For the optimization of enantioselectivity, attention is
often paid to the structure of an achiral acyl donor. In
the terms of the two-step reaction mechanism (typical
to serine hydrolases) proceeding through an acyl-
enzyme intermediate (RCO-enzyme), enantiodiscrimi-
nation occurs when the intermediate reacts with a
cyanohydrin as a nucleophile. Accordingly, the nature
of the intermediate is changed when group R is varied.
The previous results for the acylation of the amino
group in amino esters indicated enhanced reactivity and
enantioselectivity with increasing carbon chain R in the
achiral acyl donor.8 On this basis, higher E values can
be expected for the CAL-A-catalysed acylation of 4a–d
if vinyl butanoate is used in the place of the acetate as
an acyl donor. However, vinyl acetate has been used
throughout the present work as an acyl donor due to
analytical problems in detecting the enantiomers of
butanoylated (or higher analogues) cyanohydrins by the
used HPLC-method.
Due to the very low solubility of racemic cyanohydrin
4a in most organic solvents, only reactions in acetoni-
trile and ethyl acetate were studied for further opti-
mization. Acetonitrile proved to be most useful for
CAL-A-mediated acylation with vinyl acetate (Table 1;
entries 6–8), while reduced reactivity and selectivity was
evident for the reaction in ethyl acetate (entry 5). The
results for the acylation of 4a in acetonitrile indicate
that the concentration of vinyl acetate had no effect on
reactivity or enantioselectivity (entries 6–8). On the
other hand, an unexpected decrease in enantioselectivity
is clear with increasing the amount of CAL-A prepara-
tion from 10 to 50 mg ml−1 (entries 6, 10 and 11). The
explanation for this is unclear especially since reactivity
increases with increasing enzyme content, excluding the
possibility to mass transfer effects. When racemic
cyanohydrin acetate 2a was subjected under the enzy-
matic reaction conditions without an added nucleophile
in the presence of 5 and 50 mg ml−1 of the enzyme
preparation, the hydrolysis of the cyanohydrin acetate
2a to the cyanohydrin 4a by water coming from the
seemingly dry enzyme preparation was not observed.
According to the above results, the reactions of 4a–d
with vinyl acetate in the presence of CAL-A prepara-
tion (10 mg ml−1) in acetonitrile were studied. The
reactions proceeded smoothly with enantioselectivities
dependent on the nature of the substituents at the
benzene ring (Table 2). A substituent at the position 6%

622
C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
of the benzotiazol ring lowers enantioselectivity (entries
2 and 3, Table 2) compared to the unsubstituted com-
pound (entry 1) while the substituent at the position 4%
has no effect (entry 4).
kinetic resolution (Scheme 3), the resolved mixture of
one of the pairs (S)-4a–d and (R)-2a–d against time was
treated with the basic resin (5 mg ml−1; 0.65 mmol
ml−1). The free (S)-cyanohydrins 4a–d were racemized
within less than 2 h, whereas the enantiomeric composi-
tion of (R)-cyanohydrin esters 2a–d remained unaltered
even after 7 days.
Table 2. Acylation of racemic cyanohydrins 4a–d (13 mM)
with vinyl acetate (40 mM) in acetonitrile mediated by
CAL-A preparation (10 mg/ml)
The amount of basic resin is critical for the dynamic
resolution as previously shown for the case of mande-
lonitrile.5 In the present work, acetic acid which is
liberating from vinyl acetate through the reaction with
the water present in the enzyme preparation can neu-
tralize the resin, and accordingly lead to normal kinetic
rather than dynamic kinetic resolution. In addition,
some hydrogen cyanide is also present in spite of the
fast formation of cyanohydrins from the aldehydes.
That the basic resin is still active at the end of the
dynamic kinetic resolution was shown by adding a new
portion of acetone cyanohydrin after the dynamic reso-
lution of 4a. The dark brown colour appeared immedi-
ately, indicating that the base-catalysed decomposition
of acetone cyanohydrin produced hydrogen cyanide
followed by the base-catalysed polymerisation.14 The
dark brown color was also observed when 10 mg ml−1
(1.3 mmol ml−1) of the resin was used instead of the
normal 5 mg ml−1, but in this case the enzymatic
acylation was also prevented.
Entry
Resolution
products
E
Time (h)
Conversion
(%)
1
2
3
4
(S)-4a, (R)-2a 6292
(S)-4b, (R)-2b 1490.3
(S)-4c, (R)-2c 4892
(S)-4d, (R)-2d 6492
24
24
24
24
44
39
46
48
2.2. Dynamic kinetic resolution and preparation of (R)-
cyanohydrin acetates
In an efficient enzymatic dynamic kinetic resolution, the
less reactive enantiomer should be rapidly racemized
under the same conditions where the product of the
enzyme-catalysed reaction remains stable. Moreover,
the enzyme should keep its activity throughout the
reaction. Under such conditions, the starting material is
always practically racemic, allowing the maximal enan-
tiopurity (as determined by the E value of the corre-
sponding kinetic resolution) to be reached at the
theoretical zero conversion for the more reactive enan-
tiomer throughout the reaction. The use of appropriate
conditions is extremely important, because ee of the
product enantiomer in normal enzymatic kinetic resolu-
tion tends to decrease with increasing conversion.
In the one-pot synthesis of (R)-cyanohydrin acetates
2a–d by dynamic kinetic resolution, CAL-A provided
reactivities that were comparable to those found in
normal kinetic resolution (Table 3 compared to Table
2). More importantly, the observed ee values for the
acetates 2a–d were in accordance with the theoretical
values that can be calculated from the E values in Table
2 by extrapolating to zero conversion using the equa-
tion of Chen et al.15 The reactions were practically at
100% conversion after 48 or 72 h in the presence of 10
mg ml−1 of the enzyme preparation. At lower enzyme
content longer reaction times were needed (entry 2,
Table 3). These results indicated that the present
dynamic kinetic resolution do not suffer from water
effects which were previously described for the CAL-B-
catalysed dynamic kinetic resolution of mandelonitrile.5
In accordance with the high enzyme content of 50
mg/ml the very fast reaction of aldehyde 3a gave
product 2a at ee=57% compared to the theoretical
value of 80% (entry 3), indicating that the racemization
In the present dynamic kinetic resolution method for
the production of (R)-cyanohydrin acetates 2a–d
(Scheme 2), the fast equilibrium between a cyanohydrin
and the corresponding aldehyde (and acetone) and
hydrogen cyanide in the presence of Amberlite IRA-904
basic resin is essential. In order to demonstrate that
under the reaction conditions cyanohydrins can be
smoothly produced in reasonable amounts, a mixture
of aldehyde 3a (65 mol) and acetone cyanohydrin (200
mol) was equilibrated in acetonitrile in the presence of
the basic resin (5 mg ml−1; 0.65 mmol ml−1). The
equilibrium value of 8 for [4a]/[3a] in less than 2 h
clearly fulfilled the requirement. To verify the enan-
tiomeric stability of the products of the above-described
Table 3. Dynamic kinetic resolution of racemic cyanohydrins 4a–d (13 mM) with vinyl acetate (40 mM) in acetonitrile
mediated by CAL-A preparation (10 mg ml−1
)
Entry
Product
Conv. (%)
Isolated yield (%)
Time (h)
ee (%)
Theoretical ee (%)
[h]2D5
1
2
3
4
5
6
(R)-2a
(R)-2aa
(R)-2ab
(R)-2b
(R)-2c
(R)-2d
\99
\99
\99
\99
\99
\99
91
–
–
92
93
92
48
120
12
72
48
96
97
97
80
86
95
97
+17.9
–
92a
57b
84
–
+14.2
+18.2
+8.1
95
96
48
a 5 mg ml−1 of CAL-A preparation.
b 50 mg ml−1 of CAL-A preparation.

C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
623
4. Experimental
4.1. Materials and methods
of 4a was not fast enough compared to the fast enzy-
matic reaction as shown in Table 1 (entry 11).
2.3. Absolute configuration
Anilines, 2-furoyl chloride, vinyl acetate, trimetylsilyl
cyanide, P2S5, PClO3 and ZnI2 were products of Aldrich
or Fluka. Amberlite IRA 904 was purchased from Acros
and it was conditioned as previously described.4 All
solvents were purified and dried by standard methods as
required. Lipase A (CAL-A) and lipase B (CAL-B,
Chirazyme L2) from Candida antarctica were purchased
from Boehringer-Mannheim. Before use CAL-A was
adsorbed on Celite (4.0 g; product from Aldrich high
purity analytical grade) by dissolving the enzyme and
sucrose (0.24 g) in Tris–HCl buffer (pH 7.9) and the
solution was thereafter left to dry at room temperature
as described previously.13 The final lipase content in the
enzyme preparation was 10% (w/w). All enzymatic reac-
tions were performed at room temperature (23–24°C).
The R absolute configurations of the enzymatically
produced enantiomers of cyanohydrin esters 1 were
previously reported for the lipase PS-catalysed kinetic
resolution of the corresponding cyanohydrins by
acylation in accordance with the R enantioselectivity
of the lipase PS-mediated kinetic resolution of furan-
2-ylhydroxyacetonitrile and with the S enantioselectiv-
ity of the lipase PS-catalysed acylation of
mandelonitrile and related compounds (the change
from R to S is due to the change in the priority
order of the substituents from furanyl to phenyl-
based cyanohydrins).1,3,4 The R enantiopreference can
also be addressed to the CAL-A-mediated acylation
of furan-2-ylhydroxyacetonitrile with vinyl acetate by
comparing the peak positions in the HPLC chro-
matograms to those observed in the case of lipase PS
catalysis. As a further support, the reaction yielded
the corresponding enantiomerically enriched (+)-acetic
esters with [h]2D5=+18.4 (c 1, CHCl3) (ee 67%) in
accordance with the literature value of [h]2D5=+24.3 (c
1.6, CHCl3) (ee=98%) for the (R)-(+)-acetic ester.16
The 1H and 13C nuclear magnetic resonance (NMR)
spectra were recorded on a Jeol Lambda 400 spectrom-
eter operating at 399.78 and 100.54 MHz, respectively.
1H spectra were referenced internally to the solvent signal
(d6-DMSO, 2.49 ppm); 13C spectra to the solvent signal
(d6-DMSO, 39.50 ppm). The correct assignment of the
chemical shifts was confirmed by a combination of
standard CHSHF (f1-decoupled, optimized to 145 Hz
1JCH coupling) and COLOC (optimized to nJCH couplings
of 4, 6 and 8 Hz) 13C, 1H correlation measurements. Mass
spectra (MS) were taken on a VG 7070E mass spectrom-
eter. Thin layer chromatography (TLC) was carried out
using Merck Kieselgel 60F254 sheets. Spots were visual-
ized by treatment with 5% ethanolic phosphomolybdic
acid solution and heating. Preparative chromatographic
separations were performed using column chromatogra-
phy on Merck Kieselgel 60 (0.063–0.200 mm). Melting
points were determined by hot plate method and are
uncorrected. Optical rotations were measured in acetoni-
trile (c 1) on a Jasco DIP-360 polarimeter and [h]D values
are given in units of 10−1 deg cm2 g−1. HPLC analyses
were conducted with a HP 1090 instrument using a
CHIRACEL OD column (0.46×25 cm) and a mixture of
hexane and isopropanol (80:20) as eluent at the flow rate
of 1 ml min−1. For good baseline separation, the unre-
acted cyanohydrin in the sample was derivatized with
chloroacetylchloride in the presence of pyridine contain-
ing 1% 4-N,N-dimethylaminopyridine (DMAP) before
injections. Retention time for 5-(benzothiazol-2-yl)-
furan-2-ylcyanomethyl acetates (R,S-2a–d) and for
3. Conclusions
The present work describes the usefulness of
a
dynamic kinetic resolution method for the preparation
of novel cyanohydrin acetates (R)-2a–d. This method
exploits the reversible nature of the base-catalysed
cyanohydrin formation from the corresponding alde-
hydes 3a–d and hydrogen cyanide and the base-
catalysed cyanohydrin decomposition to the aldehyde
and hydrogen cyanide, leading to the effective racem-
ization of cyanohydrins 4a–d at the same time when
the R enantiomer is selectively acylated in the CAL-
A-catalysed acylation (Scheme 2). The method
exploits acetone cyanohydrin as an in situ source of
hydrogen cyanide. Everything takes place in one pot,
allowing at least 99% of the original aldehyde to be
transformed to the corresponding cyanohydrin ester
(R)-3a–d with high enantiopurity (ee 84–96%). The
possibility to avoid the separate preparation and
purification of the relatively labile cyanohydrins 4a–d
and the handling of hydrogen cyanide are great
advantages of this method.
5-(benzothiazol-2-yl)-furan-2-ylcyanomethyl
chloro-
acetates (R,S-2%a–d) are presented in Table 4. The deter-
Table 4. Retention time for acetates (R,S-2a–d) and
chloroacetates (R,S-2%a–d)
An important aspect of the present work is that a
readily available lipase can be used for the produc-
tion of enantiopure cyanohydrin intermediates
because of the high stability, good availability and
wide substrate selectivity of the enzymes. It has been
possible to show the unique nature of the CAL-A
enzyme to work with sterically hindered substrates
under conditions where more common lipases, such
as lipase PS and CAL-B, fail.
Rt (min)
Compound
(R)-2
(S)-2
(R)-2%
(S)-2%
a
b
c
46.4
27.4
17.2
23.3
9.3
14.6
6.8
53.4
31.3
22.4
25.1
11.5
18.3
14.5
11.5
d
10.6

624
C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
mination of E was based on equation E=ln[(1−c)(1−
eeS)]/ln[(1−c)(1+eeS)].15 Using linear regression E is
achieved as the slope of a line. In the case of the
dynamic kinetic resolution, conversion was determined
(100) [M+], 228 (15), 227 (2), 202 (1), 201 (6), 200 (7),
174 (3), 173 (13), 172 (37), 147 (1), 146 (5), 145 (8), 140
(2), 128 (5); H NMR: l=7.53 (ddd, J=1.2 Hz, J=7.1
1
Hz, J=8.1 Hz, 1H, (C(5%)-H), 7.57 (d, J=3.9 Hz, 1H,
C(4)-H), 7.60 (ddd, J=1.3 Hz, J=7.1 Hz, J=8.2 Hz,
1H, C(6%)-H), 7.74 (d, J=3.9 Hz, 1H, C(3)-H), 8.10
(ddd, J=0.6 Hz, J=1.2 Hz, J=8.2 Hz, 1H, C(7%)-H),
8.21 (ddd, J=0.6 Hz, J=1.3 Hz, J=8.1 Hz, 1H, C(4%)-
1
using the H NMR spectra of the reaction mixture by
integrating characteristic proton signals for all the
chemical species involved (aldehydic proton for 3a–d,
proton from cyanohydrin group for 4a–d and protons
from acetate for the esters 2a–d).
H), 9.73 (s, 1H, CH6
O); 13C NMR: l=113.4 (C(4)),
122.7 (C(4%)), 123.3 (C(7%)), 124.3 (C(3)), 126.3 (C(5%)),
127.2 (C(6%)), 134.3 (C(3a%)), 151.6 (C(5)), 152.9 (C(2)),
153.2 (C(7a%)), 155.6 (C(2%)), 178.8 (CHO).
4.2. Preparation of racemic cyanohydrins and their
esters
The synthesis of racemic cyanohydrins 4a–d and esters
2a–d was straightforward, as shown in Scheme 4. 2-
Furan-2-ylbenzothiazoles were first prepared as previ-
ously described starting from furan-2-carboxylic acid
phenylamides.17,18 These products were further trans-
formed into carbaldehydes 3a–d using the Vilsmeier–
Haack formylation method.19 The aldehydes yielded
novel racemic cyanohydrins 4a–d via a reaction with
trimethylsilylcyanide and ZnI2, followed by desilylation
with 3 M HCl.20 Esterification for analytical purposes
was performed with acetic anhydride in the presence of
DMAP and triethyl amine.
4.2.1.2. 5-(6-Chlorobenzothiazol-2-yl)furan-2-carbalde-
hyde 3b. Yield: 69%. Mp 213°C; HRMS M+ found (M+
calculated for C12H6ClNO2S): 262.98001 (262.98078);
MS: m/z (relative intensity)=267 (2), 266 (5), 265 (37),
264 (18), 263 (100) [M+], 262 (12), 237 (2), 236 (2), 235
(6), 234 (5), 209 (4), 208 (14), 207 (12), 206 (35), 182 (1),
181 (3), 180 (4), 179 (6), 174 (1), 172 (2), 171 (3), 144
(9), 142 (4); 1H NMR: l=7.60 (d, J=3.9 Hz, 1H,
C(4)-H), 7.63 (dd, J=2.2 Hz, J=8.8 Hz, 1H, C(5%)-H),
7.75 (d, J=3.9 Hz, 1H, C(3)-H), 8.10 (d, J=8.8 Hz,
1H, C(4%)-H), 8.38 (d, J=2.2 Hz, 1H, C(7%)-H), 9.74 (s,
1H, CH6
O); 13C NMR: l=113.9 (C(4)), 122.4 (C(7%)),
124.2 (C(3)), 124.4 (C(4%)), 127.7 (C(5%)), 130.8 (C(6%)),
135.8 (C(3a%)), 151.1 (C(5)), 152.0 (C(7a%)), 153.0 (C(2)),
156.6 (C(2%)), 178.9 (CHO).
4.2.1. Preparation of aldehydes 3a–d. Phosphorus oxy-
chloride (10 ml) was added dropwise into ice-cooled
dimethylformamide (30 ml), followed by addition of
one of the 2-furan-2-yl benzothiazoles (5a–d) (100
mmol). The resulting mixture was stirred at 100°C for 2
h and then poured into ice (500 g). The pH of the
reaction mass was set to 6 by addition of saturated
aqueous sodium acetate solution. After 3 h the crude
product was filtered, and dried. Chromatography on
silica gel using dichloromethane as an eluent followed
by recrystallization from ethyl acetate resulted in the
formyl derivatives.
4.2.1.3.
5-(6-Methylbenzothiazol-2-yl)furan-2-carb-
aldehyde 3c. Yield: 72%. Mp 168°C, (lit.19 169–170°C);
HRMS M+ found (M+ calculated for C13H9NO2S):
243.03428 (243.03540); MS: m/z (relative intensity)=
246 (1), 245 (6), 244 (16), 243 (100) [M+], 242 (14), 241
(2), 216 (1), 215 (3), 214 (10), 188 (1), 187 (4), 186 (17),
1
185 (3), 184 (1), 154 (4), 122 (4), 121 (15); H NMR:
l=2.70 (s, 3H, CH6 3), 7.39 (m, 1H, C(5%)-H), 7.40 (m,
1H, C(7%)-H), 7.53 (d, J=3.9 Hz, 1H, C(4)-H), 7.73 (d,
J=3.9 Hz, 1H, C(3)-H), 7.99 (m, Hz, 1H, C(4%)-H),
4.2.1.1. 5-(Benzothiazol-2-yl)furan-2-carbaldehyde 3a.
Yield: 74%. Mp 176°C, (lit.19 176°C); HRMS M+ found
(M+ calculated for C12H7NO2S): 229.02068 (229.01975);
MS: m/z (relative intensity)=231 (6), 230 (15), 229
9.72 (s, 1H, CH6 6 H3), 113.4
O); 13C NMR: l=18.0 (C
(C(4)), 119.9 (C(4%)), 124.3 (C(3)), 126.3 (C(7%)), 127.4
(C(5%)), 133.0 (C(6%)), 134.2 (C(3a%)), 151.6 (C(5)), 152.6
(C(7a%)), 152.8 (C(2)), 154.4 (C(2%)), 178.7 (CHO).
Scheme 4. Reagents and conditions: (i) P2S5/Py; (ii) Baker’s yeast; (iii) POCl3/DMF; (iv) (CH3)3SiCN/CH2Cl2; (v) Ac2O
(DMAP/Py)/CH2Cl2.

C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
625
4.2.1.4. 5-(4-Chlorobenzothiazol-2-yl)furan-2-carbalde-
hyde 3d. Yield: 71%. Mp 207°C; HRMS M+ found (M+
calculated for C12H6ClNO2S): 262.98111 (262.98078);
MS: m/z (relative intensity)=267 (2), 266 (5), 265 (37),
264 (20), 263 (100) [M+], 262 (16), 237 (2), 236 (3), 235
(4), 234 (6), 209 (4), 208 (10), 207 (9), 206 (25), 182 (1),
181 (2), 180 (3), 179 (4), 174 (1), 172 (2), 171 (3), 144
1H, C(2)-CH(CN)OH
6
), 7.58 (dd, J=2.2 Hz, J=8.8 Hz,
1H, C(5%)-H), 8.02 (d, J=8.8 Hz, 1H, C(4%)-H), 8.30 (d,
J=2.2 Hz, 1H, C(7%)-H); 13C NMR: l=55.9
(C
(CH(C
6
H(CN)OH), 112.0 (C(3)), 113.3 (C(4)), 118.1
N)OH), 122.1 (C(7%)), 123.9 (C(4%)), 127.3 (C(5%)),
6
130.1 (C(6%)), 135.3 (C(3a%)), 148.1 (C(2)), 152.0 (C(5)),
152.0 (C(7a%)), 157.0 (C(2%)).
1
(5), 142 (2); H NMR: l=7.51 (dd, J=8.0 Hz, J=8.1
Hz, 1H, (C(6%)-H), 7.63 (d, J=3.9 Hz, 1H, C(4)-H),
7.69 (dd, J=1.0 Hz, J=8.0 Hz, 1H, C(5%)-H), 7.75 (d,
J=3.9 Hz, 1H, C(3)-H), 8.19 (dd, J=1.0, J=8.1 Hz,
4.2.2.3. 5-(6-Methylbenzothiazol-2-y)furan-2-ylhydr-
oxyacetonitrile rac-4c. Yield: 87%. Mp 160°C; HRMS
M+ found (M+ calculated for C14H10N2O2S): 270.04421
(270.04630); MS: m/z (relative intensity)=270 (1) [M+],
253 (1), 246 (1), 245 (6), 244 (16), 243 (100), 242 (14),
241 (2), 216 (1), 215 (3), 214 (10), 188 (1), 187 (4), 186
(17), 185 (3), 184 (1), 154 (4), 122 (4), 121 (15); 1H
Hz, 1H, C(7%)-H), 9.75 (s, 1H, CH6
O); 13C NMR: l=
114.2 (C(4)), 121.8 (C(7%)), 124.1 (C(3)), 127.0 (C(4%)),
127.1 (C(6%)), 127.2 (C(5%)), 136.0 (C(3a%)), 149.9
(C(7a%)), 151.0 (C(5)), 153.1 (C(2)), 156.8 (C(2%)), 178.9
(CHO).
NMR: l=2.69 (s, 3H, CH
6 3), 6.04 (d, J=7.0 Hz, 1H,
C(2)-CH(CN)OH), 6.84 (dd, J=0.5 Hz, J=3.6 Hz, 1H,
6
4.2.2. Synthesis of racemic cyanohydrins 4a–d. To a
stirred solution of one of the aldehydes (3a–d, 1 mmol)
in dry dichloromethane (10 ml) a catalytic amount of
ZnI2 (3.2 mg, 10 mmol) and trimethylsilyl cyanide (119
mg, 150 ml, 1.2 mmol) were added dropwise at room
temperature and the resulting mixture was stirred fur-
ther at room temperature, until the entire quantity of
the aldehyde was transformed. The dichloromethane
was evaporated and the crude product was redissolved
in 5 ml acetonitrile. The formed trimethylsilyl cyanohy-
drin was decomposed by adding 3 M HCl (100 ml). The
solvent was evaporated and the resulted crude cyanohy-
drins were purified by recrystallization from ethyl
acetate.
C(3)-H), 7.34 (d, J=3.6 Hz, 1H, C(4)-H), 7.35 (m, 1H,
C(7%)-H), 7.36 (m, 1H, C(5%)-H), 7.42 (d, J=7.0 Hz, 1H,
C(2)-CH(CN)OH
l=18.0 (CH3), 55.9 (C
(C(4)), 118.2 (CH(CN)OH), 119.7 (C(4%)), 125.7 (C(7%)),
6
), 7.94 (m, 1H, C(4%)-H); 13C NMR:
6
6 H(CN)OH), 111.9 (C(3)), 112.6
6
127.2 (C(5%)), 132.5 (C(6%)), 133.6 (C(3a%)), 148.6 (C(2)),
151.6 (C(5)), 152.5 (C(7a%)), 155.0 (C(2%)).
4.2.2.4. 5-(4-Chlorobenzothiazol-2-yl)furan-2-ylhydr-
oxyacetonitrile rac-4d. Yield: 85%. Mp 191°C; HRMS
M+ found (M+ calculated for C13H7ClN2O2S):
289.98334 (289.99167); MS: m/z (relative intensity)=
292 (1), 290 (3) [M+], 273 (1), 267 (2), 266 (5), 265 (37),
264 (20), 263 (100), 262 (16), 237 (2), 236 (3), 235 (4),
234 (6), 209 (4), 208 (10), 207 (9), 206 (25), 182 (1), 181
(2), 180 (3), 179 (4), 174 (1), 172 (2), 171 (3), 144 (5),
4.2.2.1. 5-(Benzothiazol-2-yl)furan-2-ylhydroxyaceto-
nitrile rac-4a. Yield: 86%. Mp 139°C; HRMS M+
found (M+ calculated for C13H8N2O2S): 256.031143
(256.03065); MS: m/z (relative intensity)=256 (2) [M+],
239 (2), 231 (6), 230 (15), 229 (100), 228 (15), 227 (2),
202 (1), 201 (6), 200 (7), 174 (3), 173 (13), 172 (37), 147
1
142 (2); H NMR: l=6.06 (d, J=7.1 Hz, 1H, C(2)-
CH6 (CN)OH), 6.87 (dd, J=0.7 Hz, J=3.7 Hz, 1H,
C(3)-H)), 7.43 (d, J=3.7 Hz, 1H, C(4)-H), 7.45 (dd,
J=7.9 Hz, J=8.1 Hz, 1H, C(6%)-H), 7.46 (d, J=7.1 Hz,
1H C(2)-CH(CN)OH
1H, C(5%)-H), 8.12 (dd, J=1.1 Hz, J=8.1 Hz, 1H,
C(7%)-H); 13C NMR: l=55.9 (C
H(CN)OH), 112.0
(C(3)), 113.7 (C(4)), 118.1 (CH(CN)OH), 121.5 (C(7%)),
6 ), 7.64 (dd, J=1.1 Hz, J=7.9 Hz,
1
(1), 146 (5), 145 (8), 140 (2), 128 (5); H NMR: l=6.03
(d, J=6.5 Hz, 1H, C(2)-CH
Hz, J=3.7 Hz, 1H, C(3)-H)), 7.37 (d, J=3.7 Hz, 1H,
C(4)-H), 7.43 (d, J=6.5 Hz, 1H, C(2)-CH(CN)OH),
6
(CN)OH), 6.85 (dd, J=0.5
6
6
6
126.5 (C(6%)), 126.6 (C(4%)), 126.9 (C(5%)), 135.4 (C(3a%)),
148.0 (C(2)), 149.9 (C(7a%)), 152.2 (C(5)), 157.2 (C(2%)).
7.47 (ddd, J=1.2 Hz, J=7.2 Hz, J=8.2 Hz, 1H, C(5%)-
H), 7.55 (ddd, J=1.2 Hz, J=7.2 Hz, J=8.2 Hz, 1H,
C(6%)-H), 8.04 (ddd, J=0.6 Hz, J=1.2 Hz, J=8.2 Hz,
1H, C(7%)-H), 8.15 (ddd, J=0.6 Hz, J=1.2 Hz, J=8.2
4.2.3. Chemical acylation of rac-4a–d with acetic anhy-
dride. Acetic anhydride (79 mg, 81.8 ml, 0.5 mmol), a
catalytic amount of 4-N,N-dimethylaminopyridine in
pyridine (5 ml; 1% solution) and triethylamine (50.5 mg,
69.2 ml, 0.5 mmol) were added into the solution of one
of the racemic cyanohydrins (rac-4a–d, 0.5 mmol) in
dichloromethane (5 ml). After stirring for 15 min at
room temperature the solvent was evaporated in vac-
uum and the crude product was purified by column
chromatography using dichloromethane as an eluent.
Finally the cyanohydrin esters rac-2a–d were recrystal-
lized from diisopropyl ether.
Hz, 1H, C(4%)-H); 13C NMR: l=55.9 (C
6
H(CN)OH),
111.9 (C(3)), 112.8 (C(4)), 118.1 (CH(C
6 N)OH), 122.5
(C(4%)), 122.8 (C(7%)), 125.7 (C(5%)), 126.9 (C(6%)), 133.8
(C(3a%)), 148.5 (C(2)), 151.7 (C(5)), 153.2 (C(7a%)), 156.1
(C(2%)).
4.2.2.2. 5-(6-Chlorobenzothiazol-2-yl)furan-2-ylhydr-
oxyacetonitrile rac-4b. Yield: 87%. Mp 197°C; HRMS
M+ found (M+ calculated for C13H7ClN2O2S):
290.01025 (289.99167); MS: m/z (relative intensity)=
292 (1), 290 (3) [M+], 273 (1), 267 (2), 266 (5), 265 (37),
264 (18), 263 (100), 262 (12), 237 (2), 236 (2), 235 (6),
234 (5), 209 (4), 208 (14), 207 (12), 206 (35), 182 (1), 181
(3), 180 (4), 179 (6), 174 (1), 172 (2), 171 (3), 144 (9),
4.2.3.1. 5-(Benzothiazol-2-yl)-furan-2-ylcyanomethyl
acetate rac-2a. Yield: 67%. Mp 109°C; HRMS M+
found (M+ calculated for C15H10N2O3S): 298.041260
(298.041218); MS: m/z (relative intensity)=300 (3), 299
(8), 298 (46) [M+], 258 (5), 256 (87), 255 (62), 254 (2),
241 (6), 240 (18), 239 (100), 238 (8), 237 (10), 230 (3),
1
142 (4); H NMR: l=6.03 (d, J=6.8 Hz, 1H, C(2)-
CH
6 (CN)OH), 6.86 (d, J=0.5 Hz, J=3.6 Hz, 1H, C(3)-
H), 7.39 (d, J=3.6 Hz, 1H, C(4)-H), 7.43 (d, J=6.8 Hz,

626
C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
229 (5), 228 (10), 212 (6), 211 (19), 210 (29), 202 (2), 201
(4), 174 (4), 173 (6), 172 (12), 141 (1), 140 (6), 136 (6),
135 (6), 134 (11); 1H NMR: l=2.20 (s, 3H, C(2)-
(11), 247 (5), 246 (5), 245 (13), 244 (9), 238 (8), 237 (40),
210 (6), 209 (9), 208 (8), 207 (6), 206 (13); H NMR: l
1
=2.20 (s, 3H, C(2)-CH(CN)OCOCH3
6 ), 7.01 (s, 1H,
CH(CN)OCOCH3
6
), 6.99 (s, 1H, C(2)-CH
6
(CN)OCO-
C(2)-CH(CN)OCOCH3), 7.09 (dd, J=0.4 Hz, J=3.5
6
CH3), 7.07 (d, J=0.5 Hz, J=3.7 Hz, 1H, C(3)-H), 7.43
(d, J=3.7 Hz, 1H, C(4)-H), 7.49 (ddd, J=1.2 Hz,
J=7.3 Hz, J=8.1 Hz, 1H, C(5%)-H), 7.58 (ddd, J=1.2
Hz, J=7.3 Hz, J=8.3 Hz, 1H, C(6%)-H), 8.05 (ddd,
J=0.6 Hz, J=1.2 Hz, J=8.3 Hz, 1H, C(7%)-H), 8.17
(ddd, J=0.6 Hz, J=1.2 Hz, J=8.1 Hz, 1H, C(4%)-H);
Hz, 1H, C(3)-H), 7.48 (dd, J=8.0 Hz, J=8.1 Hz, 1H,
C(6%)-H), 7.50 (d, J=3.5 Hz, 1H, C(4)-H), 7.67 (dd,
J=1.1 Hz, J=8.0 Hz, 1H, C(5%)-H), 8.15 (dd, J=1.1
Hz, J=8.1 Hz, 1H, C(7%)-H); 13C NMR: l=20.1
(CH(CN)OCOC
(C(4)), 114.6 (CH(C
(C(7%)), 126.7 (C(4%)), 126.8 (C(6%)), 127.0 (C(5%)), 135.5
(C(3a%)), 146.7 (C(2)), 149.1 (C(5)), 149.9 (C(7a%)), 156.9
6
H3), 56.1 (C
6 H(CN)OCOCH3), 113.8
6
N)OCOCH3), 115.6 (C(3)), 121.6
13C NMR: l=20.0 (CH(CN)OCOC
6
H3), 56.1
(C6
H(CN)OCOCH3), 112.9 (C(4)), 114.6 (CH(CN)-
6
OCOCH3), 115.4 (C(3)), 122.5 (C(4%)), 123.0 (C(7%)),
125.9 (C(5%)), 127.0 (C(6%)), 133.9 (C(3a%)), 146.3 (C(2)),
149.5 (C(5)), 153.1 (C(7a%)), 155.7 (C(2%)), 168.7
(C(2%)), 168.8 (CH(CN)OC6 OCH3).
(CH(CN)OC
6
OCH3).
4.3. Kinetic resolution of racemic cyanohydrins rac-4a–d
4.2.3.2. 5-(6-Chlorobenzothiazol-2-yl)furan-2-ylcyano-
methyl acetate rac-2b. Yield: 68%. Mp 151°C; HRMS
M+ found (M+ calculated for C15H9ClN2O3S):
332.00296 (332.00224); MS: m/z (relative intensity)=
335 (2), 334 (16), 333 (7), 332 (43) [M+], 294 (2), 293 (6),
292 (35), 291 (32), 290 (95), 289 (47), 288 (2), 276 (7),
275 (38), 274 (22), 273 (100), 272 (9), 264 (7), 263 (6),
262 (12), 247 (6), 246 (10), 245 (16), 244 (17), 238 (6),
In a typical small scale experiment, one of the cyanohy-
drins 4a–d (13 mol) and vinyl acetate (3.4 mg, 3.6 l, 40
mol) were dissolved in dry organic solvent (1 ml).
CAL-A preparation (10 mg, corresponding to 1 mg of
the enzyme) was added. Samples (20 l) were taken after
2, 4, 6, 8, 12, 24, 48, 72, 96, 120, 144, 168 h and diluted
with the same solvent (80 l) as the mobile phase for
HPLC (Table 2).
1
237 (26), 210 (7), 209 (7), 208 (9), 207 (6), 206 (15); H
NMR: l=2.20 (s, 3H, C(2)-CH(CN)OCOCH3
6 ), 6.99 (s,
1H, C(2)-CH(CN)OCOCH3), 7.08 (d, J=3.6 Hz, 1H,
6
4.4. One-pot synthesis of cyanohydrin esters (R)-(+)-
3a–d
C(3)-H), 7.45 (d, J=3.6 Hz, 1H, C(4)-H), 7.59 (dd,
J=2.1 Hz, J=8.8 Hz, 1H, C(5%)-H), 8.05 (d, J=8.8 Hz,
1H, C(4%)-H), 8.32 (d, J=2.1 Hz, 1H, C(7%)-H); 13C
One of the aldehydes 3a–d (65 mol), acetone cyanohy-
drin (17 mg, 18 l, 200 mol) and vinyl acetate (17 mg, 18
l, 200 mol) were added in dry acetonitrile (5 ml). To
this solution Amberlite IRA 904 (−OH form, 5 mg
NMR: l=20.0 (CH(CN)OCOC
OCH3), 113.4 (C(4)), 114.6 (CH(C
(C(3)), 122.2 (C(7%)), 124.1 (C(4%)), 127.4 (C(5%)), 130.3
(C(6%)), 135.4 (C(3a%)), 146.5 (C(2)), 149.1 (C(5)), 151.9
6
H3), 56.1 (C
6 H(CN)OC-
6
N)OCOCH3), 115.5
−
(0.0065 mmol OH) ml−1) and immobilised CAL-A (10
(C(7a%)), 156.6 (C(2%)), 168.7 (CH(CN)OC
6
OCH3).
mg ml−1) were added. The reaction mixture was stirred
at room temperature for 48–72 h. The enzyme and the
resin were filtered off and washed with acetonitrile (2×5
ml). Solvents were distilled off from the filtrate and the
residue was purified by column chromatography on
silica gel with dichloromethane yielding (R)-(+)-2a–d
(Table 3).
4.2.3.3. 5-(6-Methylbenzothiazol-2-yl)furan-2-ylcyano-
methyl acetate rac-2c. Yield: 67%. Mp 125–126°C;
HRMS M+ found (M+ calculated for C16H12N2O3S):
312.05621 (312.05686); MS: m/z (relative intensity)=
314 (3), 313 (8), 312 (44) [M+], 272 (4), 271 (13), 270
(64), 269 (61), 255 (6), 254 (19), 253 (100), 252 (11), 251
(18), 226 (8), 225 (9), 224 (8), 223 (11), 188 (3), 187 (4),
1
186 (10), 185 (4), 150 (3), 149 (3), 148 (7); H NMR:
l=2.20 (s, 3H, C(2)-CH(CN)OCOCH3
6
), 2.69 (s, 3H,
Acknowledgements
C(6%)-CH3), 7.00 (s, 1H, C(2)-CH(CN)OCOCH3), 7.07
6
6
(dd, J=0.4 Hz, J=3.6 Hz, 1H, C(3)-H), 7.36 (m, 1H,
C(5%)-H), 7.37 (m, 1H, C(7%)-H), 7.40 (d, J=3.6 Hz, 1H,
C(4)-H), 7.96 (m, 1H, C(4%)-H); 13C NMR: l=18.0
Cs.P. is grateful for a grant from CIMO in Finland
(The Center of International Mobility). Financial sup-
port from the Arany Ja´nos Foundation is also grate-
fully acknowledged.
(C(6%)-C
6
H3), 20.1 (CH(CN)OCOC
6
H3), 56.1 (C
6 H(CN)-
OCOCH3), 112.6 (C(4)), 114.7 (CH(C
6
N)OCOCH3),
115.5 (C(3)), 119.8 (C(4%)), 125.9 (C(7%)), 127.2 (C(5%)),
132.7 (C(6%)), 133.7 (C(3a%)), 146.1 (C(2)), 149.7 (C(5)),
152.5 (C(7a%)), 154.6 (C(2%)), 168.8 (CH(CN)OC6 OCH3).
References
4.2.3.4. 5-(4-Chlorobenzothiazol-2-yl)-furan-2-ylcyano-
methyl acetate rac-2d. Yield: 70%. Mp 129°C; HRMS
M+ found (M+ calculated for C15H9ClN2O3S):
332.00247 (332.00224); MS: m/z (relative intensity)=
335 (2), 334 (13), 333 (6), 332 (34) [M+], 294 (2), 293 (6),
292 (37), 291 (32), 290 (100), 289 (45), 288 (2), 276 (5),
275 (27), 274 (13), 273 (71), 272 (2), 264 (7), 263 (5), 262
1. Paizs, Cs.; Ta¨htinen, P.; Lundell, K.; Poppe, L.; Irimie,
F. D.; Kanerva, L. T. Eur. J. Org. Chem., submitted.
2. Drauz, K.; Waldmann, H. Enzyme Catalysis in Organic
Synthesis; Wiley-VCH: New York, 2002.
3. Kanerva, L. T.; Rahiala, K.; Sundholm, O. Biocatalysis
1994, 10, 169–180.

C. Paizs et al. / Tetrahedron: Asymmetry 14 (2003) 619–627
627
4. Inagaki, M.; Hiratake, J.; Nishioka, T.; Oda, J. J. Org.
Chem. 1992, 57, 5643–5649.
Tetrahedron: Asymmetry 2001, 12, 2447–2455.
13. Kanerva, L. T.; Sundholm, O. J. Chem. Soc., Perkin
Trans. 1 1993, 2407–2410.
5. Li, Y.; Straathof, J. J.; Hanefeld, U. Tetrahedron: Asym-
14. Wadsten, T.; Andersson, S. Acta Chem. Scand. 1959, 13,
1069–1074.
15. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J.
Am. Chem. Soc. 1982, 104, 7294–7299.
metry 2002, 13, 739–743.
6. Va¨nttinen, E.; Kanerva, L. T. Tetrahedron: Asymmetry
1995, 6, 1779–1786.
7. Kirk, O.; Christensen, M. W. Org. Proc. Res. Dev. 2002,
6, 446–451.
16. Schmidt, M.; Herve´, S.; Kleimpier, N.; Griengl, H. Tetra-
hedron 1996, 52, 7833–7840.
8. Kanerva, L. T.; Csomo´s, P.; Sundholm, O.; Berna´th, G.;
Fu¨lo¨p, F. Tetrahedron: Asymmetry 1996, 7, 1705–1716.
9. Gedey, Sz.; Liljeblad, A.; La´za´r, L.; Fu¨lo¨p, F.; Kanerva,
L. T. Tetrahedron: Asymmetry 2001, 12, 105–110.
10. Solyma´r, M.; Fu¨lo¨p, F.; Kanerva, L. T. Tetrahedron:
Asymmetry 2002, 13, 2383–2388.
11. Liljeblad, A.; Lindborg, J.; Kanerva, A.; Katajisto, J.;
Kanerva, L. T. Tetrahedron Lett. 2002, 43, 2471–2474.
12. Virsu, P.; Liljeblad, A.; Kanerva, A.; Kanerva, L. T.
17. Irimie, F. D.; Paizs, Cs.; Chender, C.; Joo, F.; Silaghi-
Dumitrescu, R.; Majdik, C.; Tosa, M. Chemia 1998, 43,
109–114.
18. Paizs, Cs.; Tosa, M.; Majdik, C.; Misca, R.; Irimie, F.
Roum. Biotechnol. Lett. 2001, 6, 325–331.
19. Farcasan, V.; Paiu, F. Chemia 1971, 16, 111–115.
20. Gassman, P. G.; Talley, J. J. Tetrahedron Lett. 1978, 40,
3773–3776.