A novel approach to enantiopure cyclopropane compounds from biotransformation of nitriles

Article abstract of DOI:10.1039/b200110a

Rhodococcus sp. AJ270, a powerful and versatile nitrile hydratase/amidase containing microbial whole-cell system, catalyzed the enantioselective hydrolysis of both racemic trans- and cis-2-arylcyclopropanecarbonitriles to afford the corresponding amides and acids with enantiomeric excesses as high as >99%. The reaction rate and enantioselectivity observed for both nitrile hydratase and amidase were also strongly dependent upon the nature of the substituent and substitution pattern on the benzene ring of the substrates. The application of and the advantages of biotransformation of nitriles were demonstrated by preparing (1S,2R)-2-phenylcyclopropylamine and (1R,2R)-2-phenylcyclopropylmethylamine through facile and straightforward chemical manipulations of (1S,2S)-2-phenylcyclopropanecarboxylic acid and (1R,2R)-2-phenylcyclopropanecarboxamide, respectively.

Full text of DOI:10.1039/b200110a

View Article Online / Journal Homepage / Table of Contents for this issue
A novel approach to enantiopure cyclopropane compounds from
biotransformation of nitrilesy
Mei-Xiang Wang* and Guo-Qiang Feng
Laboratory of Chemical Biology, Center for Molecular Science, Institute of Chemistry,
Chinese Academy of Sciences, Beijing 100080, China. E-mail: mxwang@infoc3.icas.ac.cn
Received (in Montpellier, France) 3rd January 2002, Accepted 3rd September 2002
First published as an Advance Article on the web 18th October 2002
Rhodococcus sp. AJ270, a powerful and versatile nitrile hydratase/amidase containing microbial whole-cell
system, catalyzed the enantioselective hydrolysis of both racemic trans- and cis-2-arylcyclopropanecarbonitriles
to afford the corresponding amides and acids with enantiomeric excesses as high as > 99%. The reaction rate
and enantioselectivity observed for both nitrile hydratase and amidase were also strongly dependent upon the
nature of the substituent and substitution pattern on the benzene ring of the substrates. The application of and
the advantages of biotransformation of nitriles were demonstrated by preparing (1S,2R)-2-
phenylcyclopropylamine and (1R,2R)-2-phenylcyclopropylmethylamine through facile and straightforward
chemical manipulations of (1S,2S)-2-phenylcyclopropanecarboxylic acid and (1R,2R)-2-
phenylcyclopropanecarboxamide, respectively.
Although the past two decades have seen tremendous develop-
ment of enzyme-catalyzed reactions,1,2 nitrile-hydrolyzing
enzymes in organic synthesis remained largely unexplored until
recently,3,4 despite the fact that nitriles are very important
compounds5 and the bioconversion of nitriles into the corre-
sponding carboxylic acids has been known for decades.6
Enzyme-catalyzed hydrolyses of nitriles have been shown to
proceed by either direct transformation to carboxylic acid with
a nitrilase7 or by the nitrile-hydratase-catalyzed formation of
amides that are further hydrolyzed to the acids by an amidase.8
So far, a range of microorganisms containing a nitrilase or
nitrile hydratase/amidase, or both, has been isolated9 and
some nitrilases, nitrile hydratases and amidases have also been
purified.10 Mechanisms for the enzymatic hydrolysis of nitriles
using nitrilase and nitrile hydratase have been proposed, and
for the latter the hydration step is believed to involve com-
plexation of the nitrile function to a transition metal such as
iron or cobalt.11 It is particularly worth noting that the micro-
bial hydration of acrylonitrile is currently being applied in
Japan to produce tens of thousands of tonnes of acrylamide
per year.12
against a-substituted phenylacetamides while the nitrile hydra-
tase displayed a low R-selectivity against nitriles.16 To further
explore its potential in organic synthesis, and also to gain a
deeper insight into the mechanisms of enzymatic hydrolyses
of nitrile and amide, we have undertaken a systematic study
of Rhodococcus sp. AJ270 whole-cell catalyzed hydrolysis of
nitriles and amides. We report herein the enantioselective
biotransformation of 2-arylcyclopropanecarbonitriles and
-carboxamides, a novel enzymatic method for the preparation
of optically pure cyclopropane derivatives.18
Syntheses19 of enantiopure cyclopropane derivatives have
attracted much attention in recent years because such com-
pounds occur widely in natural products and in synthetic phar-
maceuticals and agrochemicals and their enantiomers often
exhibit different biological activities. For example, (+)-
(1S,2R)-2-phenylcyclopropylamine has been shown to display
fivefold more potent monoamine oxidase inhibitor activity
than its enantiomer,20 whereas (ꢀ)-(1R,2S)-2-(2-hydroxyphe-
nyl)-N,N-di-n-propylcyclopropylamine is a strong and selec-
tive 5-hydroxytryptamine receptor agonist.21 Among the
various studies constructing chiral cyclopropanes, the progress
of catalytic asymmetric cyclopropanation has been most
noticeable due to the pioneering works of Nozaki,22 Aratani,23
Pfaltz,24 Masamune,25 Evans26 and Doyle.19a,b Though high
enantiocontrol has been achieved for catalytic asymmetric
intermolecular cyclopropanation reactions between diazoace-
tates and alkenes, it should be noted that few examples19,27
have been reported to effect both high enantiocontrol and dia-
stereocontrol. In other words, highly diastereoselectivity for
either the trans or, particularly, cis isomer with high enantio-
meric excess is difficult to accomplish. The lipase catalysis
has been reported28 to be able to catalyze a kinetic resolution
of trans-2-phenylcyclopropanecarboxylic acid ester while, in
Lonza, the amidase has been successfully utilized to resolve
2,2-dimethylcyclopropanecarboxamide.29
In previous studies,9,10 we have demonstrated that Rhodo-
coccus sp. AJ270, a novel isolate from a soil sample,13 is a
powerful and robust nitrile hydratase/amidase containing
microorganism. Compared with other microorganisms
reported recently, it has a broad activity against almost all
types of nitriles, and both amides and acids can be produced
in high yields from appropriate nitriles.9 Excellent regioselec-
tivity has also been observed with this microorganism when
it catalyzed the hydrolysis of aromatic dinitriles and a variety
of aliphatic dinitriles.10 Very recently, we have found that
Rhodococcus sp. AJ270 can enantioselectively transform
racemic14–16 and prochiral nitriles.17 The amidase involved in
Rhodococcus sp. AJ270 exhibited a high S-enantiocontrol
y Electronic supplementary information (ESI) available: preparation
of racemic nitrile, amides and acids; spectroscopic data of racemic
nitriles; biotransformation of racemic amides; chiral HPLC analyses
of nitriles, amides, acids and amines. See http://www.rsc.org/supp-
data/nj/b2/b200110a/
Results and discussion
In order to examine the scope and limitations of the biotrans-
formations catalyzed by Rhodococcus sp. AJ270 and also to
DOI: 10.1039/b200110a
New J. Chem., 2002, 26, 1575–1583
1575
This journal is # The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2002

View Article Online
Table 1 Biotransformations of racemic trans-2-arylcyclopropanecar-
bonitriles 1
probe the active sites of the enzymes involved in the microbial
cells, we prepared 2-arylcyclopropanecarbonitriles from the
reaction between trimethylsulfoxonium iodide and substituted
cinnamonitriles.30 Both trans and cis isomers were readily
separated using column chromatography.
2
3
Time/
h
Entry Nitrile 1a Ar
Yield/%b ee/%c Yield/%b ee/%c
1
2
1a
1a
1a
1a
1b
1b
1b
1c
1d
1d
1e
1e
1e
1f
C6H5
0.5d 37
78 37
> 99 39
> 99 61
73 (81)e
C6H5
C6H5
C6H5
1
50
25
–
55
19
3
3
Biotransformations of racemic trans-2-
arylcyclopropanecarbonitriles
4
7
–
85
> 99 43
> 99 62
> 99 77
> 99 70
72 45
0
5
4-Cl–C6H4
4-Cl–C6H4
4-Cl–C6H4
3-Cl–C6H4
2-Cl–C6H4
2-Cl–C6H4
4-F–C6H4
4-F–C6H4
4-F–C6H4
4-Me–C6H4
4-Me–C6H4
4-MeO–C6H4
4d
25
5.5 29
52 (78)e
41
25
6
To begin our study, we first examined the hydrolysis of 2-phe-
nylcyclopropanecarbonitrile, 1a, catalyzed by Rhodococcus sp.
AJ270 cells in aqueous phosphate buffer with pH 7.0 at 30 ꢁC.
As illustrated in Scheme 1, the reaction proceeded efficiently to
produce optically active (1R,2R)-2-phenylcyclopropancarbox-
amide, 2a, and (1S,2S)-2-phenylcyclopropanecarboxylic acid,
3a. The results summarized in Table 1 indicates that, with
the progression of the reaction, the enantiomeric excess of 2a
increased while the enantiomeric excess of 3a decreased
(entries 1–3), and optically inactive acid 3a was obtained after
the complete biocatalytic hydrolysis of nitrile 1a and amide 2a
(entry 4). The nitrile 1a recovered from the reaction (entry 1)
was found to be almost optically inactive with an enantiomeric
excess lower than 5%.
7
12
16
25
48
21
44
33
16
46
32
31
39
8
144
30
43
2
31
9
72
11
75 (91)e
30
25
10
11
12
13
14
15
16
17
> 99 70
89 48
5
> 99 66
> 99 73
69 32
7
2d
75
1f
5
7d
> 99 53
44 36
21 41
71 (82)e
17
12
1g
1g
4-MeO–C6H4 10d
a
b
c
Nitrile (1.73 mmol) was used. Isolated yield. Determined by chiral HPLC
analysis. d Nitrile (ca. 20%, ee < 5%) was recovered. e Enantiomeric excess after
recrystallization.
A similar change in enantiomeric excesses of both (1R,2R)-
2-arylcyclopropancarboxamides 2b–g and (1S,2S)-2-arylcyclo-
propanecarboxylic acids 3b–g against the conversion of nitriles
and amides was observed when the substrate was extended to a
variety of 2-arylcyclopropanecarbonitrile analogs 1b–g. The
introduction of a substituent onto the benzene ring, however,
led to an intriguing change in the reaction rate and enantio-
meric selectivity. While the parent 2-phenylcyclopropanecar-
bonitrile 1a and its para-substituted phenyl analogs 1b and
1e–g were efficient substrates and their conversion to the
corresponding amides and acids finished in several hours,
the reaction rate of the ortho- and particularly the meta-
substituted substrates decreased remarkably. For example,
the hydrolysis of 2-(4-chlorophenyl)cyclopropanecarbonitrile,
1b, and 2-(2-chlorophenyl)cyclopropanecarbonitrile, 1d, took
place in hours and in several tens of hours, respectively, the
effective conversion of 2-(3-chlorophenyl)cyclopropanecarbo-
nitrile 1c, was observed to take 6 days. Although the substitu-
tion pattern affected the hydrolysis rate, it did not drastically
influence the enantioselectivity, as exemplified by the observa-
tion of comparable enantiomeric excesses obtained among
amides 2b–d and among acids 3b–d from hydrolysis of chlorine-
substituted phenylcyclopropanecarbonitriles 1b–d (entries
5–10 in Table 1). Instead, it is the nature of the substituent that
plays an important role in determining the enantiocontrol of
the reaction. Good to excellent enantioselectivity was obtained
from the biotransformations of 2-phenylcyclopropanecarboni-
trile, 1a, and its analogs 1b–f bearing a methyl, chloro, or
fluoro substituent, and the enantiomeric purity of (1S,2S)-2-
arylcyclopropanecarboxylic acids 3 could be further improved
by recrystallization. In contrast, the reaction of 2-(4-methoxy-
phenyl)cyclopropanecarbonitrile, 1g, led to low enantioselec-
tivity, giving the corresponding amide 2g and acid 3g in only
21–44% and 12–17% ee, respectively. It should be noted that
the loss of mass balance from hydrolysis of nitriles is probably
due to the degradation of organic molecules mediated by other
enzymes present in Rhodococcus sp. AJ270 cells.
The outcomes obtained suggest that the formation of
(1R,2R)-2-arylcyclopropancarboxamides 2 and (1S,2S)-2-aryl-
cyclopropanecarboxylic acids 3 results from the combined
effect of tandem kinetic resolutions of nitrile hydration and
amide hydrolysis that are catalyzed, respectively, by the nitrile
hydratase and the amidase in the cells. The nitrile hydratase
involved in Rhodococcus sp. AJ270 shows very low stereoselec-
tivity against the trans-2-arylcyclopropanecarbonitriles exam-
ined, which is exemplified by the fact that the recovered
nitriles 1 had extremely low enantiomeric excesses (entries 1,
5, 14, 16 and 17 in Table 1). The amidase involved in the cell,
however, exhibited 1S-selectivity against the trans-2-arylcyclo-
propanecarboxamides 2 generated from the hydration of the
nitriles 1. With the only exception of 4-methoxyphenyl-substi-
tuted compounds 2g and 3g, the formation of highly enantio-
merically enriched amides 2 and acids 3, irrespective of the
nature of aryl group of the substrate, suggests that the amidase
has a relatively broad spectrum against amides and its chiral
recognition relies mainly on the primary structure of trans-2-
phenylcyclopropanecarboxamide rather than on the substitu-
ent on the aromatic ring. It should be pointed out that the
precise reason for the poor enantioselectivity of the biotrans-
formation of trans-2-(4-methoxphenyl)cyclopropanecarboni-
trile, 1g, is not clear, although it is probably due to the
electronic nature or/and most likely the steric effect of the
methoxy group as other sterically smaller groups, either elec-
tron-withdrawing or electron-donating, caused no dramatic
influence on the enantioselectivity of the reaction.
Kinetic resolution of racemic trans-2-
arylcyclopropanecarboxamides
To shed further light on the stereoselective biotransformations
of nitriles, some racemic trans-2-arylcyclopropanecarboxa-
mides 2 were synthesized31 and subjected to the biohydrolysis.
Under identical conditions, Rhodococcus sp. AJ270 cells cata-
lyzed the hydrolysis of amides effectively (Scheme 2), and half
of the amides was transformed within several hours into (ꢀ)-
(1R,2R)-2-arylcyclopropanecarboxamides 2a, 2b and 2f and
(+)-(1S,2S)-2-arylcyclopropanecarboxylic acids 3a, 3b and 3f
(Table 2). Although the enantioselectivity of this kinetic reso-
lution was only moderate (E ¼ 2.1–6.8), it confirmed the pre-
ferential interaction of the amidase towards (1S,2S)-2-aryl-
cyclopropanecarboxamide isomers and the combined effect of
Scheme 1
1576
New J. Chem., 2002, 26, 1575–1583

View Article Online
Scheme 3
Scheme 2
the substituent, the slower the reaction. For instance, cis-2-
(4-fluorophenyl)cyclopropanecarbonitrile, 4e, gave the corre-
sponding optically pure amide 5e and acid 6e in 38% and
46% yields, respectively, whereas the methyl and methoxy sub-
stituted nitrile analogs 4f and 4g produced only small amounts
of the amide ( < 20% yield) and no acid at all was formed, even
with the use of twice the amount of biomass or b-cyclodextrin
(b-CD) as an additive in the latter case (entries 9–14 in Table 3).
Changing of the chloro substitution from the para (4b) to ortho
(4d) positions of the benzene ring seemed to slightly facilitate
biotransformations of nitrile and amide, which led to higher
conversion of 4d and to higher yields of enantiopure 2-(2-
chlorophenyl)cyclopropanecarboxylic acid, 6d (entries 7 and 8).
It has been generally believed that the nitrile hydratase in
various microbial organisms exhibits only low to moderate
enantioselectivity against a variety of racemic nitriles.5,16,32
This is also true for the hydration of trans-2-arylcyclopropane-
carbonitriles 1 as aforementioned. In the case of hydration of some
cis-2-arylcyclopropanecarbonitriles such as 4a and 4e, however,
the nitrile hydratase of Rhodococcus sp. AJ270 displays good
1S-enantioselectivity, as the nitrile recovered after more than
50% conversion showed an enantiomeric excess value as high as
99%. More surprisingly, enantioselection of the nitrile hydra-
tase inverted from 1S to 1R when the substituent on the benzene
ring varied from others to the para-methoxy group. As indi-
cated by the formation of enantiopure (1S,2R)-2-arylcyclo-
propanecarboxylic acids 6a, 6b, 6d and 6e, the amidase, which
is very sensitive to the presence of the substituent on the
benzene ring, has a very strict 1S-enentiocontrol against amides.
It has been proposed that the nitrile hydratase is an iron or
cobalt-containing metelloenzyme while amidase is a normal
hydrolytic enzyme with nucleophilic group such as esterases
and proteases. To account for the enantioselectivity of bio-
transformations of trans- and cis-2-arylcyclopropanenitriles,
we propose that a readily reachable reactive site be embedded
within the larger pocket of the 1S-enantioselective nitrile
hydratase while the amidase comprises a relatively deep-buried
and size-limited 1S-enantioselective active site. The efficiency
of catalysis and enantioselection, however, results from effective
and selective binding of the substrates to the surface of the
active site of the enzymes, which in turn is strongly governed
by the structure of the substrates. In other words, easy and
rapid binding of nitrile or amide substrates to the active sites
of nitrile hydratase or amidase, respectively, in the absence
of a steric interaction between the substrate and enzyme during
the chiral recognition process, would result in efficient reaction
but with low enantioselectivity. As a consequence, a sterically
less crowded cyano functional group such as in both enantio-
mers of trans-2-arylcyclopropanecarbonitrile 1 could fit com-
fortably and efficiently in the spacious active site of nitrile
hydratase, leading to rapid hydration of nitrile 1 with almost
no enantioselectivity. In contrast, bearing an adjacent cis-aryl
group, the heavily sterically hindered cyano group of 4 could
not be bound freely and non-selectively by the nitrile hydra-
tase. Only slow and selective hydration can occur in a lengthy
incubation with 1S,2R-2-phenylcyclopropanecarbonitrile, 4a,
and its 2-(4-fluorophenyl)-substituted analog 4e, yielding
recovered (1R,2S)-2-arylcyclopropanecarbonitriles in excellent
enantiomeric excesses. Although it remains a mystery, the
inversion of the enantioselectivity observed for cis-2-(4-meth-
oxyphenyl)cyclopropanecarbonitrile, 4g, could also probably
stereoselective amidase and nitrile hydratase towards the
hydrolysis of nitriles. The lower enantioselectivity observed
in these kinetic resolutions is probably attributed to the low
solubility of amides 2 in aqueous buffer, which leads to only
a limited amount of both enantiomers of amide in solution
available to the amidase. We did find a similar phenomenon
that when a very small amount of nitrile or amide was fed to
the organism under the same conditions; the conversion to acid
was rapid and usually went to completion with the formation
of racemic acid.16 In the case of trans-2-(4-methoxyphenyl)cy-
clopropanecarboxamide, 2g, the same low enantioselectivity
(E ¼ 1.3) as that of nitrile hydrolysis was obtained (entry 6
in Table 2), suggesting again that the methoxy group causes
the low efficiency of enantio-differentiation of the amidase.
In practical synthesis, it is apparently advantageous to apply
the biotransformation of nitriles rather than amides to prepare
enantiopure (1R,2R)-2-arylcyclopropanecarboxamides 2 and
optically active (1S,2S)-2-arylcyclopropanecarboxylic acids 3.
Biotransformations of racemic cis-2-
arylcyclopropanecarbonitriles
An extraordinarily high enantioselectivity and a sluggish reac-
tion were observed when racemic cis-2-arylcyclopropanecarbo-
nitrile substrates 430 were incubated with Rhodococcus sp.
AJ270 (Table 3). Compared to the rapid and efficient biocon-
version of trans isomer 1a, 7 days’ interaction of cis isomer 4a
with biocatalyst under the identical conditions did not even
furnish complete hydration, and more than half the starting
nitrile were recovered (entry 1 in Table 3). The use of a co-
solvent such as acetone (entry 3 in Table 3) and the employ-
ment of a biphasic system of aqueous phosphate buffer and
n-hexane (entry 4 in Table 3) did not improve the conversion
of nitriles. Only when the substrate concentration decreased
to 4 mM, did the complete transformation of nitrile 1a take
place in a week to afford enantiopure (1R,2S)-2-phenylcyclo-
propanecarboxamide, 5a, and (1S,2R)-2-phenylcyclopropane-
carboxylic acid, 6a, in excellent yields (entry 6 in Table 3
and Scheme 3). It is also noteworthy that the reaction outcome
was strongly governed by both the nature of the substituent
and the substitution pattern of the benzene ring as well. The
presence of a substituent at the para position of the aromatic
ring of 4 inhibited the hydration of the nitrile and, more
greatly, the hydrolysis of the amide. The sterically bulkier
Table 2 Kinetic resolution of racemic trans-2-arylcyclopropanecar-
boxamides 2
2
3
Time/ Yield/ ee/ Yield/ ee/
b
c
b
c
Entry (+/ꢀ)-2a Ar
h
%
%
%
%
E
1
2
3
4
5
6
2a
2a
2b
2b
2f
C6H5
2
4
4
6
6
8
47
22
52
37
54
55
51
77
58
68
57
2
53
77
47
59
42
32
33
8
3.2
2.1
4.5
2.7
6.8
1.3
C6H5
4-Cl–C6H4
4-Cl–C6H4
4-Me–C6H4
4-MeO–C6H4
44
20
59
13
2g
a
b
Amide (1.73 mmol) was used. Isolated yield. Determined by chiral HPLC
c
analysis.
New J. Chem., 2002, 26, 1575–1583
1577

View Article Online
Table 3 Biotransformations of racemic cis-2-arylcyclopropanecarbonitriles 4
Recovered 4
(1R,2S)-5
(1S,2R)-6
Entry
Nitrile 4
Ar
Conc/mmol
Time/days
Yield/%a
ee/%b
Config.
Yield/%a
ee/%b
Yield/%a
ee/%b
1
2
4a
4a
4a
4a
4a
4a
4b
4d
4e
4f
C6H5
C6H5
1.73
1
7
7
7
7
2
7
7
7
7
7
7
7
7
7
52
15
27
90
49
–
90
> 99
> 99
9
1R,2S
1R,2S
1R,2S
1R,2S
1R,2S
–
7
32
24
–
79
> 99
> 99
–
30
49
43
9
> 99
> 99
> 99
90
3
C6H5
C6H5
1c
4
1d
0.5
5
C6H5
C6H5
> 99
–
6
66
43
48
8
> 99
> 99
> 99
> 99
> 99
–
6
0.2
41
18
15
38
4
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
> 99
7
4-Cl–C6H4
2-Cl–C6H4
4-F–C6H4
4-Me–C6H4
4-Me–C6H4
4-MeO–C6H4
4-MeO–C6H4
4-MeO–C6H4
0.25
0.6
67
45
11
85
80
77
71
62
6
49
Nde
8
1R,2S
1R,2S
Nde
Nde
33
46
–
9
0.6
0.6
> 99
2
10
11
12
13
14
4f
0.25
0.25
0.25f
0.25g
8
17
11
18
17
15
–
–
4g
4g
4g
1S,2R
1S,2R
1S,2R
–
–
22
27
–
–
–
–
a
Isolated yield. b Determined by chiral HPLC analysis. c A biphasic system of phosphate buffer (25 ml) and n-hexane (25 ml) was used. d Acetone (3 ml) was added as
e
f
g
a co-solvent. Configuration was not determined. Four grams of wet cell were used. b-CD (100 mg) was added.
be due to the steric effect. Such an ever increased aryl group as
4-methoxyphenyl might cause a mismatch in the recognition
between the 1S,2R-enantiomer of 4g and the chiral nitrile
hydratase, and instead, the 1R,2S-enantiospecific but ineffi-
cient catalysis took place, which gave enantiopure (1S,2R)-2-
(4-methoxyphenyl)cyclopropanecarboxamide, 5g, in 18%
yield.
Being different from the nitrile hydratase, 1S-enantio-
selective amidase appears to be much more sensitive to the
structure of substrates and has always showed differentiation
between a pair of enantiomers of both trans- and cis-2-arylcy-
clopropanecarboxamides, albeit in different degrees. In the
case of less sterically hindered trans-2-arylcyclopropanecar-
boxamides 2, an efficient and stereoselective interaction
between the amidase and substrate is allowed, which gives rise
to the rapid formation of (1S,2S)-2-arylcyclopropanecar-
boxylic acids 3 in moderate to good enantiomeric excesses.
When cis-2-arylcyclopropanecarboxamides 5a–e, which were
generated from the hydration of nitriles 4a–e, were available
to amidase, enantiospecific hydrolytic reaction proceeded
slowly to give (1S,2R)-2-arylcyclopropanecarboxylic acids
6a–e as the sole product in good yield. The proposed steric-
sensitive amidase model is also supported by the fact that
the migration of chlorine from the 4- to 2-position of the phe-
nyl accelerated the conversion of amide into the acid, as the
accessibility of the amide group to the surface of amidase
increases from 5b to 5d. The observation of no biotransforma-
tion of amides 5f and 5g, which bear even larger substituents
such as methyl and methoxy, respectively, further supports
the model of size-limited amidase.
Scheme 4
of esters, that is not only the optically active carboxylic acid
but also the optically active amides, important organic nitro-
gen compounds, are readily produced.
Conclusion
Rhodococcus sp. AJ270 is a powerful and versatile nitrile
hydratase/amidase containing biocatalytic system able to cat-
alyze the enantioselective hydrolysis of both racemic trans- and
cis-2-arylcyclopropanecarbonitriles to afford the optically
active trans- and cis-2-arylcyclopropanecarboxylic acids and
their derivatives with high enantiomeric excesses. High enan-
tioselectivity of the biotransformations of nitriles results from
the combined effect of the nitrile hydratase and amidase. The
reaction rate and enantioselectivity of both nitrile hydratase
and amidase present in Rhodococcus sp. AJ270 microbial cells
are strongly dependent upon the configuration of the sub-
strates, and the nature of the substituents and their substitu-
tion pattern on the benzene ring of the substrates. On the
basis of the outcomes of this study, it is proposed that a readily
reachable reactive site be embedded within the spacious pocket
of the 1S-enantioselective nitrile hydratase while the amidase
might comprise a relatively deep-buried and size-limited 1S-
enantioselective active site. We have also demonstrated the
advantages of biotransformation of nitriles over that of esters
by converting (1S,2S)-2-phenylcyclopropanecarboxylic acid,
3a, and (1R,2R)-2-phenylcyclopropanecarboxamide, 2a, into
(1S,2R)-2-phenylcyclopropylamine, 7, and (1R,2R)-2-phenyl-
cyclopropylmethylamine, 8, respectively, through facile and
straightforward chemical manipulations.
Preparation of chiral amines containing the cyclopropane moiety
Both optically active 2-arylcyclopropanecarboxylic acids and
their amide derivatives obtained from this study are useful
precursors of cyclopropane-containing amines of therapeutic
significance. Antidepressant tranylcypromine, (+)-(1S,2R)-
2-phenylcyclopropylamine, 7, for example, was readily pre-
pared from (1S,2S)-2-phenylcyclopropanecarboxylic acid, 3a,
through a modified Curtius rearrangement.28,33 In addition,
a convenient and straightforward reduction of (1R,2R)-2-
phenylcyclopropanecarboxamide, 2a, using LiAlH4 afforded
high yields of enantiopure (1R,2R)-2-phenylcyclopropylmethy-
lamine, a potential candidate as an antihypertensive agent34
(Scheme 4). These chemo-enzymatic examples reflect one of
the extra advantages of biotransformations of nitriles over that
1578
New J. Chem., 2002, 26, 1575–1583

View Article Online
d 7.12–7.34 (m, 5H, Ar-H), 6.25 (br s, 2H, NH2), 2.56–2.63
(m, 1H, H2), 1.67–1.77 (m, 2H, H1 and H3b), 1.37–1.44 (m,
1H, H3a); IR (KBr) 3353, 3174 (NH2), 1659, 1628; MS (EI)
m/z 161 (M+, 23%), 144 (18), 117 (100), 115 (59). Anal. calcd.
for C10H11NO: C, 74.51; H, 6.88; N, 8.69. Found: C, 74.17; H,
6.70; N, 8.59.
Experimental
General
Both melting points, which were determined using a Reichert
Kofler hot-stage apparatus, and boiling points are uncor-
rected. NMR spectra were recorded in CDCl3 solution on a
Bruker AM 300 spectrometer. Chemical shifts are reported
in ppm and coupling constants are given in Hertz. IR spectra
were obtained on a Bruker IMX20 instrument as liquid films
or KBr discs. Mass spectra were measured on an AEI MS-50
mass spectrameter and microanalyses were carried out by the
Analytical Laboratory of the Institute. Polarimetry was carried
out using an Optical Activity AA-10R polarimeter and the
measurements were made at the sodium D-line with a 0.5 dm
pathlength cell. Concentrations (c) are given in g per 100 ml.
All HPLC were run using a Shimadzu SCL-10AVP HPLC
system with a UV detector set at 204 nm. The enantiomeric
excess (ee) values of all products were obtained by means of
chiral HPLC analyses using authentic racemic samples as
references.
All regents were obtained from commercial suppliers and
used without further purification. Column chromatography
was performed using 200–300 mesh silica gel. Racemic trans-
and cis-2-arylcyclopropanecarbonitriles 1 and 4 were prepared
following the literature method30 and their configurations were
determined by the 1H NMR spectrum, which gave distinct
coupling patterns of 2-H (Fig. 1).35 Racemic amides31 2 and
5 and racemic acids36 3 and 6 were obtained from chemical
hydrolysis of nitriles 1 and 4 according to the literature.
(+)-(1S,2S)-2-Phenylcyclopropanecarboxylic acid (3a). 0.5 h
(37%); ee 73% (ee 81% after recrystallization, chiral HPLC);
mp 57–59 ꢁC (ee 81%) [lit.28 46–47 ꢁC, (+)-(1S, 2S) enantio-
mer, ee > 99%]; [a]D ¼ + 253.3 (c ¼ 4.7, CHCl3) [lit.28
25
[a]D ¼ + 377.1 (c ¼ 1.04, CHCl3), ee > 99%]; 1H NMR d
20
7.70 (br s, 1H, COOH), 7.10–7.32 (m, 5H, Ar–H), 2.58–2.64
(m, 1H, H2), 1.87–1.94 (m, 1H, H1), 1.64–1.70 (m, 1H, H3b),
1.39–1.45 (m, 1H, H3a); IR (KBr) 2530–3031 (COOH), 1697
(C=O), 1603, 1453; MS (EI) m/z 162 (M+, 34%), 144 (17),
117 (100), 116 (30). Anal. calcd for C10H10O2 : C, 74.06; H,
6.21. Found: C, 74.11; H 6.24.
(+)-(1S,2S)-2-Phenylcyclopropanecarbonitrile (1a). 0.5
h
(24%); ee 3% (chiral HPLC); mp 50–52 ꢁC (DL) [lit. 51–52 ꢁC
25
(DL)37; 78–80 ꢁC (+) (1S, 2S) enantiomer42]; [a]D ¼ + 8.75
25
(c ¼ 1.6, CHCl3) [lit.42 [a]D ¼ + 337.5 (c ¼ 0.2400, EtOH)];
1H NMR d 7.11–7.35 (m, 5H, Ar–H), 2.61–2.66 (m, 1H, H2),
1.57–1.64 (m, 1H, H1), 1.51–1.56 (m, 1H, H3b) 1.42–1.47 (m,
1H, H3a); IR(KBr), 2223 (CN); MS (EI) m/z 143 (M+,
100%), 116 (72), 115 (76).
Enzymatic hydrolysis of (ꢂ)-trans-2-(4-chlorophenyl)cyclo-
propanecarbonitrile (1b). (ꢀ)-(1R,2R)-2-(4-Chlorophenyl)-
cyclopropanecarboxamide (2b). 5.5 h (29%); ee > 99% (chiral
HPLC); mp 170–172 ꢁC [lit.20a 183–184.5 ꢁC (DL)]; [a]D
¼
25
ꢀ302 (c ¼ 0.9, CHCl3); 1H NMR d 7.25 (d, J 8.4, 2H,
Ar–H), 7.04 (d, J 8.4, 2H, Ar–H), 6.40 (br s, 1H, NHH),
5.99 (br s, 1H, NHH), 2.52–2.58 (m, 1H, H2), 1.64–1.72 (m,
2H, H1 and H3b), 1.30–1.38 (m, 1H, H3a); IR (KBr) 3366,
3202 (NH2), 1649, 1625; MS (EI) m/z 197 (M+ + 2, 12), 195
(M+, 40%), 180 (10), 178 (27), 153 (15), 151 (44), 117 (49),
116 (51), 115 (100). Anal. calcd for C10H10ClNO: C, 61.39;
H, 5.15; N, 7.16. Found: C, 61.26; H 5.21; N, 7.12.
General procedure for the biotransformations of nitriles and
amides
To an Erlenmeyer flask (250 ml) with a screw cap was added
Rhodococcus sp. AJ270 cells (2 g wet weight) and potassium
phosphate buffer (0.1 M, pH 7.0, 50 ml) and the resting cells
were activated at 30 ꢁC for 0.5 h with orbital shaking. Nitriles
or amides (see Tables 2–4) were added in one portion to the
flask and the mixture was incubated at 30 ꢁC using an orbital
shaker (200 rpm). The reaction, monitored by TLC, was
quenched after a period of time (see Tables 2–4) by removing
the biomass through Celite pad filtration. The resulting aqu-
eous solution was basified to pH 12 with aqueous NaOH
(2 M). Extraction with CH2Cl2 or ethyl acetate gave, after dry-
ing (MgSO4) and concentration, the amide and unconverted
nitrile. Separation of amide and nitrile was effected by column
chromatography. The aqueous solution was then acidified
using aqueous HCl (2 M) to pH 2 and extracted with CH2Cl2
or ethyl acetate. Acid was obtained after removal of the sol-
vent. All products were characterized by their spectral data
and comparison of the melting points and optical rotary power
with the known compounds, which are listed below, or by full
characterization.
(+)-(1S,2S)-2-(4-Chlorophenyl)cyclopropanecarboxlic acid
(3b). 4 h (43%); ee 52% (ee 78% after recrystallization, chiral
HPLC); mp 94–95 ꢁC (ee 78%) [lit.38 115–116 ꢁC (DL)];
25
[a]D ¼ +178.8 (c ¼ 3.2, CHCl3); 1H NMR d 7.95 (br s,
1H, COOH), 7.24 (d, J 8.3, 2H, Ar–H), 7.03 (d, J 8.3, 2H,
Ar–H), 2.54–2.60 (m, 1H, H2), 1.84–1.90 (m, 1H, H1), 1.63–
1.70 (m, 1H, H3b), 1.34–1.41 (m, 1H, H3a); IR (KBr) 2533–
3025 (COOH), 1691 (C=O), 1434, 1242; MS (EI) m/z 198
(M+ + 2, 10), 196 (M+, 33%), 153 (18), 151 (55), 141 (17),
116 (54), 115 (100). Anal. calcd for C10H9ClO2 : C, 61.08; H,
4.61. Found: C, 61.01; H, 4.62.
(+)-(1S,2S)-2-(4-Chlorophenyl)cyclopropanecarbonitrile (1b).
4 h (29%); ee 3% (chiral HPLC); mp 84–85 ꢁC (DL);
25
[a]D ¼ +17 (c ¼ 3.75, CHCl3); 1H NMR d 7.28 (d, J 8.4,
2H, Ar–H), 7.05 (d, J 8.4, 2H, Ar–H), 2.60–2.63 (m, 1H,
H2), 1.61–1.68 (m, 1H, H1), 1.50–1.56 (m, 1H, H3b), 1.39–
1.46 (m, 1H, H3a); IR (KBr) 2232 (CN); MS (EI) m/z 179
(M+ + 2, 12), 177 (M+, 38%), 142 (71), 115 (100). Anal. calcd
for C10H8ClN: C, 67.64; H, 4.51; N, 7.89. Found: C, 67.91; H
4.64; N, 8.04.
Enzymatic hydrolysis of (ꢂ)-trans-2-phenylcyclopropane-
carbonitrile (1a). (ꢀ)-(1R,2R)-2-Phenylcyclopropanecarboxa-
mide (2a). 1 h (50%); ee > 99% (chiral HPLC); mp 138–
25
140 ꢁC [lit.37 189–190 ꢁC (DL)]; [a]D ¼ ꢀ312 (c ¼ 1.2,
CH3OH) [lit.37 [a]D ¼ ꢀ312 (c ¼ 1.35, CHCl3)]; 1H NMR
25
Enzymatic hydrolysis of (ꢂ)-trans-2-(3-chlorophenyl)cyclo-
propanecarbonitrile (1c). (ꢀ)-(1R,2R)-2-(3-Chlorophenyl)-
cyclopropanecarboxamide (2c). 144 h (25%); ee > 99% (chiral
25
1
HPLC); mp 133–134 ꢁC; [a]D ¼ ꢀ250 (c ¼ 0.7, CHCl3); H
NMR d 7.02–7.38 (m, 4H, Ar–H), 6.71 (br s, 1H, NHH),
4.23 (br s, 1H, NHH), 2.59–2.65 (m, 1H, H2), 1.74–1.83 (m,
1H, H1), 1.67–1.73 (m, 1H, H3b), 1.41–1.49 (m, 1H, H3a); IR
(KBr) 3354, 3173 (NH2), 1665, 1629, 1611, 1434, 1283, 1043;
MS (EI) m/z 197 (M+ + 2, 30), 195 (M+, 88%), 153 (25), 151
(76), 117 (83), 116 (51), 115 (100). Anal. calcd for
Fig. 1
New J. Chem., 2002, 26, 1575–1583
1579

View Article Online
HPLC); mp 93–95 ꢁC (ee 82%) [lit.38 119–120 ꢁC (DL)];
[a]D ¼ +258.2 (c ¼ 2.2, CHCl3); H NMR d 7.12 (d, J 7.9,
2H, Ar–H), 7.03 (d, J 7.9, 2H, Ar–H), 6.50 (br s, 1H, COOH),
2.57–2.64 (m, 1H, H2), 2.35 (s, 3H, CH3), 1.87–1.93 (m, 1H,
H1), 1.64–1.70 (m, 1H, H3b) 1.38–1.45 (m, 1H, H3a); IR
(KBr) 2505–3000 (COOH), 1692 (C=O), 1427, 1287; MS (EI)
m/z 176 (M+, 38%), 158 (18), 131 (100). Anal. calcd. for
C11H12O2 : C, 74.98; H, 6.86. Found C, 74.90; H, 6.74.
C10H10ClNO: C, 61.39; H, 5.15; N, 7.16. Found: C, 61.29; H
5.15; N, 7.16.
25
1
(+)-(1S,2S)-2-(3-Chlorophenyl)cyclopropanecarboxlic acid
(3c). 144 h (70%); ee 31% (chiral HPLC); mp 56–58 ꢁC [lit.38
108.0–108.5 ꢁC]; [a]D ¼ +122.9 (c ¼ 3.05, CHCl3); 1H
25
NMR d 7.95 (br s, 1H, COOH), 7.01–7.26 (m, 4H, Ar–H),
2.54–2.60 (m, 1H, H2), 1.84–1.90 (m, 1H, H1), 1.63–1.70 (m,
1H, H3b), 1.34–1.41 (m, 1H, H3a); IR (KBr) 2535–3053
(COOH), 1691 (C=O), 1434, 1242; MS (EI) m/z 198
(M+ + 2, 10), 196 (M+, 33%), 153 (22), 151 (55), 141 (17),
117 (23), 116 (54), 115 (100). Anal. calcd for C10H9ClO2 : C,
61.08; H, 4.61. Found: C, 60.90; H, 4.46.
(+)-(1S,2S)-2-(4-Methylphenyl)cyclopropanecarbonitrile (1f).
2 h (22%); ee 4% (chiral HPLC); mp 85–86 ꢁC (DL);
[a]D ¼ +16 (c ¼ 2.4, CHCl3); 1H NMR d 7.13 (d, J 8.0,
25
2H, Ar–H), 7.01 (d, J 8.0, 2H, Ar–H), 2.58–2.65 (m, 1H,
H2), 2.34 (s, 3H, CH3), 1.58–1.64 (m, 1H, H1), 1.49–1.55 (m,
1H, H3b), 1.40–1.17 (m, 1H, H3a); IR (KBr) 2229 (CN); MS
(EI) m/z 157 (M+, 100%), 156 (89), 142 (68), 129 (39), 115
(73). Anal. calcd for C11H11N: C, 84.08; H 7.01; N, 8.91.
Found: C, 84.08; H, 6.96; N, 9.10.
Enzymatic hydrolysis of (ꢂ)-trans-2-(2-chlorophenyl)cyclo-
propanecarbonitrile (1d). (ꢀ)-(1R,2R)-2-(2-Chlorophenyl)-
cyclopropanecarboxamide (2d). 43 h (21%); ee > 99% (chiral
25
HPLC); mp 101–102 ꢁC; [a]D ¼ ꢀ94.4% (c ¼ 1.25, CHCl3);
1H NMR d 6.99–7.40 (m, 4H, Ar–H), 6.56 (br s, 1H, NHH),
5.20 (br s, 1H, NHH), 2.72–2.82 (m, 1H, H2), 1.83–1.88 (m,
1H, H1), 1.68–1.77 (m, 1H, H3b), 1.44–1.50 (m, 1H, H3a); IR
(KBr) 3345, 3184 (NH2), 1654, 1624, 1436; MS (EI) m/z 197
(M+ + 2, 6), 195 (M+, 18%), 160 (100), 153 (10), 151 (30),
117 (71), 116 (50), 115 (78). Anal. calcd for C10H10ClNO: C,
61.39; H, 5.15; N, 7.16. Found: C, 61.76; H 5.22; N, 7.03.
(+)-(1S,2S)-2-(2-Chlorophenyl)cyclopropanecarboxlic acid
(3d). 30 h (45%); ee 72% (chiral HPLC); mp 64–65 ꢁC;
Enzymatic hydrolysis of (ꢂ)-trans-2-(4-methoxylphenyl)-
cyclopropanecarbonitrile (1g). (ꢀ)-(1R,2R)-2-(4-Methoxyl-
phenyl)cyclopropanecarboxamide (2g). 7 h (31%); ee 44%
25
(chiral HPLC); mp 155–157 ꢁC; [a]D ¼ ꢀ92 (c ¼ 0.5,
CH3OH); 1H NMR d 7.04 (d, J 8.6, 2H, Ar–H), 6.85 (d, J
8.6, 2H, Ar–H), 5.90 (br s, 1H, NHH), 3.80 (s, 3H, CH3O),
3.20 (br s, 1H, NHH), 2.51–2.57 (m, 1H, H2), 1.58–1.66 (m,
2H, H1 and H3b), 1.25–1.32 (m, 1H, H3a); IR (KBr) 3353,
3178 (NH2), 1663, 1624, 1246, 1023; MS (EI) m/z 191 (M+,
75%), 174 (75), 159 (20), 147 (100), 131 (37), 115 (36), 103
(30). Anal. calcd for C11H13NO2 : C, 69.09; H, 6.85; N, 7.32.
Found: C, 68.93; H, 6.79; N 7.30.
25
[a]D ¼ +112.6 (c ¼ 5.4, CHCl3); 1H NMR d 9.20 (br s,
1H, COOH), 7.02–7.40 (m, 4H, Ar–H), 2.80–2.85 (m, 1H,
H2), 1.80–1.85 (m, 1H, H1), 1.67–1.73 (m, 1H, H3b), 1.40–
1.43 (m, 1H, H3a); IR (KBr) 2536–3018 (COOH), 1693
(C=O), 1450, 1240; MS (EI) m/z 198 (M+ + 2, 12), 196 (M+,
37%), 161 (22), 153 (15), 151 (47), 141 (26), 117 (22), 116
(53), 115 (100). Anal. calcd for C10H9ClO2 : C, 61.08; H,
4.61. Found: C, 61.19; H, 4.57.
(+)-(1S,2S)-2-(4-Methoxylphenyl)cyclopropanecarboxylic
acid (3g). 7 h (36%), ee 17% (chiral HPLC); mp 104–106 ꢁC
[lit.38 112–113 ꢁC (DL)]; [a]D ¼ +49.4 (c ¼ 3.4, CH3OH);
25
1H NMR d 10.75 (br, 1H, COOH), 7.05 (d, J 7.9, 2H, Ar–
H), 6.84 (d, J 7.9, 2H, Ar–H), 3.82 (s, 3H, CH3O), 2.56–2.63
(m, 1H, H2), 1.82–1.88 (m, 1H, H1), 1.61–1.68 (m, 1H, H3b),
1.35–1.41 (m, 1H, H3a); IR (KBr) 2508–3003 (COOH), 1702
(C=O), 1612, 1242, 1027; MS (EI) m/z 192 (M+, 39%), 147
(100). Anal. calcd for C11H12O3 : C, 68.74; H, 6.29. Found:
C, 68.60; H, 6.19.
Enzymatic hydrolysis of (ꢂ)-trans-2-(4-fluorophenyl)cyclo-
propanecarbonitrile (1e). (ꢀ)-(1R,2R)-2-(4-Fluorophenyl)-
cyclopropanecarboxamide (2e). 5 h (33%); ee > 99% (chiral
25
HPLC); mp 153–155 ꢁC; [a]D ¼ ꢀ292.3 (c ¼ 0.65, CHCl3);
1H NMR d 6.96–7.12 (m, 4H, Ar–H), 5.74 (br s, 2H, NH2),
2.53–2.59 (m, 1H, H2), 1.61–1.69 (m, 2H, H1 and H3b), 1.24–
1.33 (m, 1H, H3a); IR (KBr) 3398, 3212 (NH2), 1652, 1616;
MS (EI) m/z 179 (M+, 54%), 162 (38), 136 (27), 135 (100),
134 (39), 133 (69). Anal. calcd for C10H10FNO: C, 67.03; H,
5.62; N, 7.82. Found: C, 66.79; H, 5.67; N, 7.70.
trans-2-(4-Methoylphenyl)cyclopropanecarbonitrile (1g). 7 h
25
1
(20%); mp 76–78 ꢁC (DL); [a]D ¼ 0; H NMR d 7.06 (d, J
8.6, 2H, Ar–H), 6.87 (d, J 8.6, 2H, Ar–H), 3.82 (s, 3H,
CH3O), 2.59–2.65 (m, 1H, H2), 1.57–1.63 (m, 1H, H1), 1.45–
1.54 (m, 1H, H3b), 1.39–1.48 (m, 1H, H3a); IR (KBr) 2241
(CN); MS (EI) m/z 173 (M+, 100%), 158 (23), 130 (36). Anal.
calcd for C11H11NO: C, 76.31; H, 6.36; N, 8.09. Found: C,
76.29; H, 6.21; N, 8.08.
(+)-(1S,2S)-2-(4-Fluorophenyl)cyclopropanecarboxylic acid
(3e). 2 h (48%); ee 75% (ee 91% after recrystallization, chiral
HPLC); mp 74–76 ꢁC (ee 91%) [lit.33b 110–112 ꢁC (DL)],
25
[a]D ¼ + 261 (c ¼ 1.7, CHCl3)]; 1H NMR d 6.94–7.29 (m,
4H, Ar–H), 6.88 (br s, 1H, COOH), 2.56–2.63 (m, 1H, H2),
1.83–1.88 (m, 1H, H1), 1.62–1.69 (m, 1H, H3b), 1.34–1.41 (m,
1H, H3a); IR (KBr) 2535–3025 (COOH), 1702 (C=O), 1227;
MS (EI) m/z 180 (M+, 37%), 162 (19), 135 (100), 134 (27),
133 (49). Anal. calcd for C10H9FO2 : C, 66.66; H, 5.03. Found:
C, 66.51; H. 5.05.
Enzymatic hydrolysis of (ꢂ)-cis-2-phenylcyclopropanecarbo-
nitrile (4a). (+)-(1R,2S)-2-Phenylcyclopropanecarboxamide
(5a). 41% yield from 0.2 mmol of racemic nitrile 4a in 7 days;
25
ee > 99% (chiral HPLC); mp 86–88 ꢁC; [a]D ¼ +47.3
(c ¼ 0.55, CHCl3); 1H NMR d 7.18–7.31 (m, 5H, Ar–H),
5.58 (br s, 1H, NHH), 3.60 (br s, 1H, NHH), 2.54 (q, J 8.7,
1H, H2), 1.92–2.00 (m, 1H, H1), 1.66–1.73 (m, 1H, H3a),
1.30–1.38 (m, 1H, H3b); IR (KBr) 3398, 3223 (NH2), 1633,
1604; MS (EI) m/z 161 (M+, 40%), 144 (29), 118 (23), 117
(100), 116 (25), 115 (49). Anal. calcd for C10H11NO: C,
74.51; H, 6.88; N, 8.69. Found: C, 74.76; H, 6.84; N, 8.62.
(+)-(1S,2R)-2-Phenylcyclopropanecarboxylic acid (6a). 49%
yield from 1 mmol of racemic nitrile 4a in 7 days; ee > 99%
(chiral HPLC); mp 72–74 ꢁC [lit.40 105.0–105.6 ꢁC (DL)];
Enzymatic hydrolysis of (ꢂ)-trans-2-(4-methylphenyl)cyclo-
propanecarbonitrile (1f ). (ꢀ)-(1R,2R)-2-(4-Methylphenyl)-
cyclopropanecarboxamide (2f). 5 h (32%); ee > 99% (chiral
HPLC); mp 185–187 ꢁC [lit.39 208 ꢁC (DL)]; [a]D ¼ ꢀ333.3
25
(c ¼ 0.75, CHCl3); 1H NMR d 7.10 (d, J 7.79, 2H, Ar–H),
7.01 (d, J 7.81, 2H, Ar–H), 6.05 (br s, 2H, NH2), 2.51–2.57
(m, 1H, H2), 2.33 (s, 3H, CH3), 1.61–1.71 (m, 2H, H1 and
H
3b), 1.29–1.35 (m, 1H, H3a); IR (KBr) 3383, 3210 (NH2),
25
25
1627, 1616; MS (EI) m/z 175 (M+, 56%), 158 (38), 132 (20),
131 (100), 117 (35), 116 (33), 115 (52). Anal. calcd for
C11H13NO: C, 75.40; H, 7.48; N, 7.99. Found: C, 74.98; H,
7.29; N, 7.92.
[a]D ¼ + 26.2 (c ¼ 2.06, CHCl3) [lit.41 [a]D ¼ +15
1
(c ¼ 2.0, CHCl3); H NMR d 7.60 (br s, 1H, COOH), 7.18–
7.28 (m, 5H, Ar–H), 2.62 (q, J 8.7, 1H, H2), 1.99–2.07 (m,
1H, H1), 1.62–1.68 (m, 1H, H3a), 1.33–1.40 (m, 1H, H3b); IR
(KBr) 2568–3463 (COOH), 1690 (C=O), 1437, 1295, 1224;
MS (EI) m/z 162 (M+, 41%), 144 (19), 117 (100), 116 (30),
(+)-(1S,2S)-2-(4-Methylphenyl)cyclopropanecarboxylic acid
(3f). 5 h (53%); ee 71% (ee 82% after recrystallization, chiral
1580
New J. Chem., 2002, 26, 1575–1583

View Article Online
115 (56). Anal. calcd for C10H10O2 : C, 74.06; H 6.21. Found:
C, 74.21; H 6.26.
179 (M+ + 2, 17), 177 (M+, 49%), 142 (83), 115 (100). Anal.
calcd for C10H8ClN: C, 67.64; H, 4.51; N, 7.89. Found: C,
67.28; H, 4.54; N, 7.78.
(ꢀ)-(1R,2S)-2-Phenylcyclopropanecarbonitrile (4a). 49%
yield recovered from 0.5 mmol of racemic nitrile 4a in 2 days;
ee > 99% (chiral HPLC); mp 44–45 ꢁC [lit. 37 ꢁC (DL)30b; 59–
Enzymatic hydrolysis of (ꢂ)-cis-2-(4-fluorophenyl)cyclopro-
panecarbonitrile (4e). (+)-(1R,2S)-2-(4-Fluorophenyl)cyclo-
propanecarboxamide (5e). 7 days (38%); ee > 99% (chiral
25
60 ꢁC (-) (1R, 2S) enantiomer42]; [a]D ¼ ꢀ23.6 (c ¼ 1.1,
CHCl3) [lit.42 [a]D ¼ ꢀ21.8 (c ¼ 0.3584, CH3OH)]; 1H
25
25
NMR d 7.26–7.40 (m, 5H, Ar–H), 2.55 (q, J 8.2, 1H, H2),
1.80–1.88 (m, 1H, H1), 1.49–1.59 (m, 2H, CH2); IR (KBr)
2231 (CN); MS (EI) m/z 143 (M+, 100%), 142 (23), 116 (87),
115 (93). Anal. calcd for C10H9N: C, 83.88; H, 6.34; N, 9.78.
Found: C, 83.83; H, 6.27; N, 9.75.
HPLC); mp 64–66 ꢁC; [a]D ¼ + 55 (c ¼ 0.80, CHCl3); 1H
NMR d 6.93–7.24 (m, 4H, Ar–H), 6.15 (br s, 1H, NHH);
4.80 (br s, 1H, NHH), 2.55 (q, J 8.4, 1H, H2), 1.96–2.04 (m,
1H, H1), 1.66–1.72 (m, 1H, H3a), 1.34–1.42 (m, 1H, H3b); IR
(KBr ) 3415, 3215 (NH2), 1652, 1617; MS (EI) m/z 179 (M+,
60%), 162 (34), 135 (100), 133 (52). Anal. calcd for
C10H10FNO: C, 67.03; H; 5.62; N, 7.82. Found: C, 66.96; H,
5.57; N, 7.85.
Enzymatic hydrolysis of (ꢂ)-cis-2-(4-chlorophenyl)cyclo-
propanecarbonitrile (4b). (+)-(1R,2S)-2-(4-Chlorophenyl)-
cyclopropanecarboxamide (5b). 7 days (18%); ee > 99% (chiral
(+)-(1S,2R)-2-(4-Fluorophenyl)cyclopropanecarboxylic acid
(6e). 7 days (46%); ee > 99% (chiral HPLC); mp 97–99 ꢁC;
25
HPLC); mp 120–122 ꢁC; [a]D ¼ + 67.7 (c ¼ 0.65, CHCl3);
25
1
[a]D ¼ + 26.6 (c ¼ 1.5, CHCl3); H NMR d 8.30 (br s, 1H,
COOH), 6.89–7.19 (m, 4H, Ar–H), 2.57 (q, J 8.6, 1H, H2),
1.97–2.04 (m, 1H, H1), 1.57–1.63 (m, 1H, H3a), 1.32–1.39 (m,
1H, H3b); IR (KBr) 2578–3505 (COOH), 1671 (C=O), 1515,
1424, 1372, 1231; MS (EI) m/z 180 (M+, 36%), 162 (18), 135
(100), 134 (27), 133 (50). Anal. calcd for C10H9FO2 : C,
66.66; H, 5.03. Found: C, 66.62; H, 4.98.
1H NMR d 7.27 (d, J 8.4, 2H, Ar–H), 7.22 (d, J 8.4, 2H,
Ar–H), 6.26 (br s, 1H, NHH), 3.75 (br s, 1H, NHH), 2.57
(q, J 8.4, 1H, H2), 2.98–2.05 (m, 1H, H1), 1.72–1.79 (m, 1H,
H3a), 1.40–1.46 (m, 1H, H3b); IR (KBr) 3410, 3222 (NH2),
1618; MS (EI) m/z 197 (M+ + 2, 14), 195 (M+, 36%), 178
(29), 151 (48), 117 (53), 116 (51), 115 (100). Anal. calcd for
C10H10ClNO: C, 61.39; H, 5.15; N, 7.16. Found: C, 61.49;
H, 5.12; N, 7.09.
(ꢀ)-(1R,2S)-2-(4-Fluorophenyl)cyclopropanecarbonitrile (4e).
25
7
days (11%); ee > 99% (chiral HPLC); oil; [a]D
¼
(+)-(1R,2S)-2-(4-Chlorophenyl)cyclopropanecarboxylic acid
(6b). 7 days (8%); ee > 99% (chiral HPLC); mp 88–90 ꢁC
ꢀ29.6 (c ¼ 0.475, CHCl3); 1H NMR d 7.03–7.28 (m, 4H,
Ar–H), 2.53 (q, J 8.0, 1H, H2), 1.80–1.86 (m, 1H, H1), 1.47–
1.57 (m, 2H, CH2); IR (KBr) 2238 (CN); MS (EI) m/z 161
(M+, 100%), 134 (67), 133 (83). Anal. calcd for C10H8FN: C,
74.52; H, 5.00; N, 8.69. Found: C, 74.55; H, 5.38; N, 8.75.
[lit.40 128.1–129.1 ꢁC (DL)]; [a]D ¼ + 16 (c ¼ 0.45, CHCl3);
25
1H NMR d 7.22 (d, J 8.5, 2H, Ar–H), 7.17 (d, J 8.5, 2H,
Ar–H), 3.80 (br s, 1H, COOH), 2.59 (q, J 8.6, 1H, H2),
2.03–2.10 (m, 1H, H1), 1.61–1.68 (m, 1H, H3a), 1.36–1.43 (m,
1H, H3b); IR (KBr) 2555–3463 (COOH), 1696 (C=O), 1493,
1438, 1230; MS (EI) m/z 198 (M+ + 2, 12), 196 (M+, 36%),
178 (15), 153 (22), 151 (65), 116 (55), 115 (100). Anal. calcd
for C10H9ClO2 : C, 61.08; H, 4.61. Found: C, 60.96; H, 4.70.
cis-2-(4-Chlorophenyl)cyclopropanecarbonitrile (4b). 7 days
Enzymatic hydrolysis of (ꢂ)-cis-2-(4-methylphenyl)cyclopro-
panecarbonitrile (4f). (+)-(1R,2S)-2-(4-Methylphenyl)cyclo-
propanecarboxamide (5f). 11% yield from 0.25 mmol of
racemic nitrile 4f in 7 days; ee > 99% (chiral HPLC); mp
25
1
122–124 ꢁC; [a]D ¼ + 36 (c ¼ 0.5, CHCl3); H NMR d 7.30
(br s, 1H, NHH), 7.12 (d, J 8.0, 2H, Ar–H), 7.07 (d, J 8.0,
2H, Ar–H), 6.20 (br s, 1H, NHH), 2.60 (q, J 8.5, 1H, H2),
2.29 (s, 3H, CH3), 2.04–2.12 (m, 1H, H1), 1.68–1.74(m, 1H,
H3a), 1.34–1.41 (m, 1H, H3b); IR (KBr) 3416, 3221 (NH2),
1634; MS (EI) m/z 175 (M+, 48%), 158 (36), 131 (100), 117
(30), 116 (30), 115 (45). Anal. calcd for C11H13NO: C, 75.40;
H, 7.48; N, 7.99. Found: C, 75.68; H, 7.51; N, 8.09.
25
(67%); ee 6% (chiral HPLC); mp 49–50 ꢁC (DL); [a]D ꢃ 0;
1H NMR d 7.32 (d, J 8.4, 2H, Ar–H), 7.20 (d, J 8.3, 2H,
Ar–H), 2.51 (q, J 8.0, 1H, H2), 1.81–1.89 (m, 1H, H1), 1.50–
1.58 (m, 2H, CH2); IR (KBr) 2235 (CN); MS (EI) m/z 179
(M+ + 2, 10), 177 (M+, 31%), 142 (80), 115 (100). Anal. calcd
for C10H8ClN: C, 67.64; H, 4.51; N, 7.89. Found: C, 67.43; H,
4.74; N, 7.80.
cis-2-(4-Methylphenyl)cyclopropanecarbonitrile (4f). 80%
yield recovered from 0.25 mmol of racemic nitrile 4f in 7 days;
Enzymatic hydrolysis of (ꢂ)-cis-2-(2-chlorophenyl)cyclo-
propanecarbonitrile (4d). (ꢀ)-(1R,2S)-2-(2-Chlorophenyl)-
cyclopropanecarboxamide (5d). 7 days (15%); ee > 99% (chiral
25
ee 8% (chiral HPLC); mp 43–44 ꢁC (DL); [a]D ꢃ 0; 1H NMR
d 7.155 (s, 5H, Ar–H), 2.49 (q, J 8.2, 1H, H2), 2.33 (s, 3H,
CH3), 1.75–1.82 (m, 1H, H1), 1.44–1.54 (m, 2H, CH2); IR
(KBr) 2230 (CN); MS (EI) m/z 157 (M+, 100%), 156 (90),
142 (73), 129 (46), 115 (89). Anal. calcd for C11H11N: C,
84.08; H, 7.01; N, 8.91. Found: C, 84.14; H, 7.06; N, 8.81.
25
1
HPLC); oil; [a]D ¼ ꢀ70 (c ¼ 1.5, CHCl3); H NMR d 7.16–
7.34 (m, 4H, Ar–H), 5.60 (br s, 1H, NHH), 5.28 (br s, 1H,
NHH), 2.55 (q, J 8.4, 1H, H2), 2.02–2.13 (m, 1H, H1), 1.75–
1.81 (m, 1H, H3a), 1.29–1.38 (m, 1H, H3b); IR (KBr) 3470,
3404 (NH2), 1663, 1612; MS (EI) m/z 197 (M+ + 2, 7), 195
(M+, 21%), 160 (100), 151 (23), 117 (54), 116 (38), 115 (63).
ESI-FTMS calcd for C10H10ClNO: 196.0524 [M+ + 1]; found
196.052 [M+ + 1].
Enzymatic hydrolysis of (ꢂ)-cis-2-(4-methoylphenyl)cyclopro-
panecarbonitrile (4g). (+)-(1R,2S)-2-(4-Methoylphenyl)cyclo-
propanecarboxamide (5g). 18% yield from 0.25 mmol of
racemic nitrile 4g in 7 days; ee > 99% (chiral HPLC); mp
(+)-(1S,2R)-2-(2-Chlorophenyl)cyclopropanecarboxylic acid
(6d). 7 days (33%); ee > 99% (chiral HPLC); mp 84–85 ꢁC;
25
1
125–127 ꢁC; [a]D ¼ + 66 (c ¼ 4.5, CHCl3); H NMR d 7.18
(d, J 8.6, 2H, Ar–H), 6.86 (d, J 8.6, 2H, Ar–H), 6.30 (br s,
2H, NH2), 3.80 (s, 3H, CH3O), 2.63 (q, J 8.5, 1H, H2), 2.04–
2.12 (m, 1H, H1), 1.66–1.73 (m, 1H, H3a), 1.39–1.46 m, 1H,
25
1
[a]D ¼ + 96.6 (c ¼ 2.75, CHCl3); H NMR d 7.19–7.38 (m,
4H, Ar–H), 3.04 (br s, 1H, COOH), 2.63–2.67 (m, 1H, H2),
2.20–2.27 (m, 1H, H1), 1.67–1.73 (m, 1H, H3a), 1.46–1.54 (m,
1H, H3b); IR (KBr) 2566–3500 (COOH), 1696 (C=O), 1447,
1325, 1233; MS (EI) m/z 198 (M+ + 2, 21), 196 (M+, 62%),
153 (18), 151 (55), 117 (24), 116 (52), 115 (100). Anal. calcd
for C10H9ClO2 : C, 61.08; H, 4.61. Found: C, 60.87; H, 4.55.
(ꢀ)-(1R,2S)-2-(2-Chlorophenyl)cyclopropanecarbonitrile (4d).
H
3b); IR (KBr) 3463, 3357 (NH2), 1652; MS (EI) m/z 191
(M+, 60%), 174 (52), 147 (100). Anal. calcd for C11H13NO2 :
C, 69.09; H, 6.85; N, 7.32. Found: C, 69.09; H, 6.69; N, 7.17.
(+)-(1S,2R)-2-(4-Methoylphenyl)cyclopropanecarbonitrile
(4g). 77% yield recovered from 0.25 mmol of racemic nitrile 4g
25
7
days (45%); ee 49% (chiral HPLC); mp 43–45 ꢁC;
in 7 days; ee 17% (chiral HPLC); oil; [a]D ¼ + 4.8 (c ¼ 1.45,
25
1
[a]D ¼ ꢀ76 (c ¼ 1.5, CHCl3); 1H NMR d 7.21–7.48 (m,
4H, Ar–H), 2.69 (q, J 8.2, 1H, H2), 1.96–2.04 (m, 1H, H1),
1.56–1.66 (m, 2H, CH2); IR (KBr) 2231 (CN); MS (EI) m/z
CHCl3); H NMR d 7.21 (d, J 8.6, 2H, Ar–H), 6.90 (d, J 8.6,
2H, Ar–H), 3.81 (s, 3H, CH3O), 2.52 (q, J 8.0, 1H, H2), 1.77–
1.84 (m, 1H, H1), 1.48–1.53 (m, 2H, CH2); IR (KBr) 2237
New J. Chem., 2002, 26, 1575–1583
1581

View Article Online
(CN); MS (EI) m/z 173 (M+, 100%), 172 (29), 158 (26), 130
(39), 121 (25). Anal. calcd for C11H11NO: C, 76.31; H, 6.36;
N, 8.09. Found: C, 76.40; H, 6.32; N, 8.17.
References
1
(a) C.-H. Wong and G. M. Whitesides, Enzymes in Synthetic
Organic Chemistry, Pergamon Press, Oxford, 1994; (b) K. Drauz
and H. Waldmann, Enzyme Catalysis in Organic Chemistry,
VCH, Weinheim, 1995; (c) K. Faber, Pure Appl. Chem., 1997,
69, 1613; (d ) H.-J. Rehm and G. Reed, Biotechnology, Wiley-
VCH, Weinheim, 1998, vol. 8a; (e) W.-D. Fessner, Biocatalysis:
From Discovery to Application, Springer Verlag, Berlin and
Heidelberg, 1999.
(a) J. B. Jones, Aldrichim. Acta, 1993, 26, 105; (b) E. Schoffers,
A. Golebiowski and C. R. Johnson, Tetrahedron, 1996, 52, 3769.
(a) T. Sugai, T. Yamazaki, M. Yokoyama and H. Ohta, Biosci.,
Biotechnol., Biochem., 1997, 61, 1419; (b) J. Crosby, J. Moiliet,
J. S. Parratt and N. J. Turner, J. Chem. Soc., Perkin Trans. 1,
1994, 1679.
Recent examples of biotransformation of nitriles (a) J. E.
Gavagan, S. K. Fager, R. D. Fallon, P. W. Folsom, F. E.
Herkes, A. Eisenberg, E. C. Hann and R. DiCosmo, J. Org.
Chem., 1998, 63, 4792; (b) K. Matoishi, A. Sano, N. Imai, T.
Yamazaki, M. Yokoyama, T. Sugai and H. Ohta, Tetrahedron:
Asymmetry, 1998, 9, 1097; (c) F. Effenberger and B. W. Graef,
J. Biotechnol., 1998, 60, 165; (d ) M. A. Wegman, U. Heinemann,
A. Stolz, F. van Rantijk and R. A. Sheldon, Org. Process Res.
Dev., 2000, 4, 318; (e) F. B. Cooling, S. K. Fager, R. D. Fallon,
P. W. Folsom, F. G. Gallagher, J. E. Gavagan, E. C. Hann,
F. E. Herkers, R. L. Phillips, A. Sigmund, I. N. Wanger,
W. Wu and R. DiCosmo, J. Mol. Catal. B: Enzym., 2001, 11, 295.
Z. Rappoport, The Chemistry of the Cyano Group, Wiley Inter-
science, New York, 1970.
(a) D. Evgred and S. Harnett, Cyanide Compounds in Biology
(Ciba Foundation Symposium 140), Wiley, Chichester, 1988;
(b) J.-C. Jallegeas, A. Arnaud and P. Galzy, Adv. Biochem.
Eng., 1980, 12, 1; (c) J. L. Legras, G. Chuzel, A. Arnaud and P.
Galzy, World J. Microbiol. Biotechnol., 1990, 6, 83.
(a) D. B. Harper, Biochem. Soc. Trans., 1976, 4, 502; (b) D. B.
Harper, Biochem. J., 1977, 165, 309; (c) M. Kobayashi and S.
Shimizu, FEMS Microbiol. Lett., 1994, 120, 217.
(a) Y. Asano, Y. Tani and H. Yamada, Agric. Biol. Chem., 1980,
44, 2251; (b) Y. Asano, K. Fujishiro, Y. Tani and H. Yamada,
Agric. Biol. Chem., 1982, 46, 1165; (c) Y. Asano, M. Tachibana,
Y. Tani and H. Yamada, Agric. Biol. Chem., 1982, 46, 1175.
For a useful overview, see: O. Meth-Cohn and M.-X. Wang,
J. Chem. Soc., Perkin Trans. 1, 1997, 1099 and references cited
therein.
Preparation of (+)-(1S,2R)-trans-2-phenylcyclopropylamine (7)
To a mixture of (+)-(1S,2S)-2-phenylcyclopropanecarboxylic
acid (3a, 70 mg, 1.43 mmol, ee 82%) and triethylamine
(61 mg, 0.604 mmol) in dry acetone (3 ml) at ꢀ10 ꢁC was added
ethyl chloroformate (71 mg, 0.656 mmol). After 2 h stirring, a
solution of NaN3(49 mg) in water (400 ml) was added and the
mixture was stirred for another 1 h. The resulting suspension
was poured into cold water (10 ml) and was extracted with
ether (4 ꢄ 10 ml). The organic layers were combined and dried
with MgSO4 . After filtration and removal of the solvent in
vacuum, the residue was then dissolved in anhydrous toluene
(10 ml) and heated on an oil bath (90 ꢁC) until no more N2
was evolved (about 3 h). The toluene was then removed in
vacuum and the resulting residue was dissolved again in dry
tBuOH (10 ml) and refluxed for 23 h. Concentration of the
mixture followed by column chromatography using silica gave
pure tert-butyl carbamate as a white solid. The solid was mixed
with THF (1 ml) and aqueous hydrochloric acid (33%, 1 ml)
and was warmed to 40 ꢁC for 30 min. After cooling to room
temperature, the mixture was basified to pH 13–14 by slowly
adding aqueous NaOH (20%), extracted with ether (4 ꢄ 10
ml) and dried with anhydrous MgSO4 . Evaporation of the sol-
vent, after filtration through a short pad of basic aluminum
oxide, afforded 7 as a colorless oil (25 mg, 68% yield);
2
3
4
5
6
25
[a]D ¼ + 93.8 (c ¼ 1.3, CHCl3) [lit.28 [a]D20 + 134.9
(c ¼ 0.85, CHCl3), ee > 99%]. Treatment of 7 with hydrogen
chloride in dry ether gave (+)-(1S, 2R)-trans-2-phenylcyclo-
propylamine hydrochloride as a white solid (27 mg): ee 91%
(chiral HPLC); mp 141–143 ꢁC; [a]D25 + 47 (c ¼ 1.1, CH3OH);
1H NMR (7ꢅHCl, 300 MHz, D2O): d 7.00–7.21 (m, 5H, Ar–H),
4.67 (s, 3H, NH3+), 2.66–2.72 (m, 1H, PhCH), 2.21–2.27 (m,
1H, CHN), 1.19–1.28 (m, 1H, CHH), 1.12–1.19 (m, 1H,
CHH); IR (KBr, 7ꢅHCl) 2095 (br, s), 1604, 1498; MS (EI)
m/z 133 (78%), 132 (100), 115(52), 103(15), 91(19), 77(23).
7
8
9
10 For a useful overview, see: O. Meth-Cohn and M.-X. Wang,
J. Chem. Soc., Perkin Trans. 1, 1997, 3197 and references cited
therein.
11 M. Kobayashi and S. Shimizu, Curr. Opin. Chem. Biol., 2000,
4, 95.
Preparation of (ꢀ)-(1R,2R)-trans-1-aminomethyl-2-
phenylcyclopropane (8)
(ꢀ)-(1R,2R)-2-phenylcyclopropanecarboxamide (2a, 84 mg,
0.52 mmol, ee > 99%) and LiAlH4(85 mg) were added to
20 ml of freshly distilled anhydrous THF under N2 protection.
The mixture was heated to reflux for 24 h and then allowed to
cool down to room temperature. After the reaction was care-
fully quenched with a small amount of ice water, the mixture
was filtered through a short pad of basic aluminum oxide
and the filter cake was thoroughly washed with THF. The fil-
trate was concentrated in vacuum to give 8 as a colorless oil
12 (a) T. Nagasawa, S. Schimizu and H. Yamada, Appl. Microbiol.
Biotechnol., 1993, 40, 189; (b) M. Kobayashi, T. Nagasawa and
H. Yamada, Trends Biotechnol., 1992, 40, 189.
13 A. J. Blakey, J. Colby, E. Williams and C. O’Reilly, FEMS Micro-
biol. Lett., 1995, 129, 57.
14 M.-X. Wang, G. Lu, G.-J. Ji, Z.-T. Huang, O. Meth-Cohn and
J. Colby, Tetrahedron: Asymmetry, 2000, 11, 1123.
15 M.-X. Wang, J.-J. Li, G.-J. Ji and J. S. Li, J. Mol. Catal.
B: Enzym., 2001, 14, 77.
16 M.-X. Wang and S.-J. Lin, Tetrahedron Lett., 2001, 42, 6925.
17 (a) M.-X. Wang, C.-S. Liu, J.-S. Li and O. Meth-Cohn, Tetra-
hedron Lett., 2000, 41, 8549; (b) M.-X. Wang, C.-S. Liu and
J.-S. Li, Tetrahedron: Asymmetry, 2001, 12, 3367.
18 Part of this work has appeared as a preliminary communication:
M.-X. Wang and G.-Q. Feng, Tetrahedron Lett., 2000, 41, 6501.
19 (a) M. P. Doyle and D. C. Forbes, Chem. Rev., 1998, 98, 911; (b)
M. P. Doyle and M. N. Protopopova, Tetrahedron, 1998, 54,
7919; (c) V. K. Singh, A. D. Gupta and G. Sekar, Synthesis,
1997, 137; (d ) M. Lautens, W. Klute and W. Tam, Chem. Rev.,
1996, 96, 49.
(66 mg, 85%); ee > 99% (chiral HPLC analysis of the corre-
25
sponding hydrochloride salt); [a]D
ꢀ43.75 (c ¼ 1.6,
CH3OH); 1H NMR d 7.02–7.27 (m, 5H, Ar–H), 2.66–2.78
(m, 2H, CH2NH2), 2.19 (br s, 2H, NH2), 1.72–1.78 (m, 1H,
PhCH), 1.26–1.32 (m, 1H, CHCH2NH2), 0.83–0.95 (m, 2H,
CH2); MS (EI) m/z 147 (M+, 4%), 130 (12), 129 (18), 115
(25), 106 (92). FAB-MS calcd. for C10H13N (M+ + 1):
148.11121; found (M+ + 1) 148.1121.
20 (a) C. Kaiser, B. M. Lester, C. L. Zirkle, A. Burger, C. S. Davis,
T. J. Della and L. Zirngibl, J. Med. Pharm. Chem., 1962, 5, 1243;
(b) C. L. Zirkle, C. Kaiser, D. H. Tedeschi, R. E. Tedeschi and
A. Burger, J. Med. Pharm. Chem., 1962, 5, 1265.
21 L.-E. Arvidsson, J. L. G. Johansson, K. Svensson, S. Hjorth, T.
Magnusson, A. Carlsson, P. Lindberg, B. Andersson, D. Sanchez,
H. Wikstrom and S. Sundell, J. Med. Chem., 1988, 31, 92.
22 H. Nozaki, S. Moriuti, H. Takaya and R. Noyori, Tetrahedron
Lett., 1966, 5239.
Acknowledgement
The financial support from the Major Basic Research Develop-
ment Program (No. 2000077506), the National Natural
Science Foundation of China and the Chinese Academy of
Sciences is gratefully acknowledged. M.-X. Wang thanks O.
Meth-Cohn and J. Colby at Sunderland University for helpful
discussions.
23 T. Aratani, Pure Appl. Chem., 1985, 57, 1839.
1582
New J. Chem., 2002, 26, 1575–1583

View Article Online
24 (a) D. Muller, G. Umbricht, B. Weber and A. Pfaltz, Helv. Chim.
¨
Turner, Tetrahedron: Asymmetry, 1992, 3, 1543; (d ) T. Beard,
M. A. Cohen, J. S. Parratt, N. J. Turner, J. Crosby and J.
Moilliet, Tetrahedron: Asymmetry, 1993, 4, 1085.
33 (a) U. M. Teotino, D. D. Bella, A. Gandini and G. Benelli,
J. Med. Chem., 1967, 10, 1091; (b) J. Vallgarda, U. Appelberg,
L.-E. Arvidsson, S. Hjorth, B. E. Svensson and U. Hacksell,
J. Med. Chem., 1996, 39, 1485.
Acta, 1991, 74, 232; (b) U. Leutenegger, G. Umbricht, C. Fahrni,
P. von Matt and A. Pfaltz, Tetrahedron, 1992, 48, 2143; (c) A.
Pfaltz, Acc. Chem. Res., 1993, 26, 339.
25 (a) R. E. Lowenthal, A. Abiko and S. Masamune, Tetrahedron
Lett., 1990, 31, 6005; (b) R. E. Lowenthal and S. Masamune, Tet-
rahedron Lett., 1991, 32, 7373.
26 (a) D. A. Evans, K. A. Woerpel, M. M. Hinman and M. M. Faul,
J. Am. Chem. Soc., 1991, 113, 726; (b) D. A. Evans, K. A.
Woerpel and M. J. Scott, Angew. Chem., Int. Ed. Engl., 1992,
31, 430.
34 R. F. Borne, M. L. Forrester and I. W. Waters, J. Med. Chem.,
1977, 20, 771.
35 A. Solladie-Cavallo and T. Isarno, Tetrahedron Lett., 1999, 40,
1579.
27 (a) T. Uchida, R. Irie and T. Katsuki, Tetrahedron, 2000, 56, 3501;
(b) T. Niimi, T. Uchida, R. Irie and T. Katsuki, Tetrahedron Lett.,
2000, 41, 3647.
28 R. Csuk, M. J. Schabel and Y. von Scholz, Tetrahedron: Asymme-
try, 1996, 7, 3505.
29 S. C. Stinton, Chem. Eng. News, 1994, 72, 38.
30 (a) C. Kaiser, B. M. Trost, J. Beeson and J. Weinstock, J. Org.
Chem., 1965, 30, 3972; (b) C. R. Johnson and P. Rogers, J. Org.
Chem., 1973, 38, 1793.
36 G. A. Pinna, M. Loruga, G. Paglietti and G. Cignarella, Farmaco,
1980, 35, 684.
37 S. C. Yasui and T. A. Keiderling, J. Am. Chem. Soc., 1987, 109,
2311.
38 T. A. Wittstruck and E. N. Trachtenberg, J. Am. Chem. Soc.,
1967, 89, 3810.
39 T. J. W. Jorden and G. W. Perold, Chem. Ber., 1953, 86, 991.
40 R. Fuchs, C. A. Kaplan, J. J. Bloomfield and L. F. Hatch, J. Org.
Chem., 1962, 27, 733.
31 J. H. Hall and M. Gisler, J. Org. Chem., 1976, 41, 3769.
32 (a) H. Kakeya, N. Sakai, T. Sugai and H. Ohta, Tetrahedron Lett.,
1991, 32, 1343; (b) F. Effenberger and J. Bo¨hme, Bioorg. Med.
Chem., 1994, 2, 715; (c) M. A. Cohen, J. S. Parratt and N. J.
41 T. Aratani, Y. Nakanisi and H. Nozaki, Tetrahedron, 1970, 26,
1675.
42 W. E. Doering and E. A. Barsa, Tetrahedron Lett., 1978,
2495.
New J. Chem., 2002, 26, 1575–1583
1583