Highly enantioselective biotransformations of 2-aryl-4-pentenenitriles, a novel chemoenzymatic approach to (R)-(-)-baclofen

Article abstract of DOI:10.1016/S0040-4039(02)01449-1

Catalyzed by Rhodococcus sp. AJ270 microbial cells under mild conditions, a range of racemic 2-aryl-4-pentenenitriles 1 underwent effective hydrolysis to afford excellent yields of enantiomerically pure (R)-(-)-2-aryl-4-pentenamides 2 and (S)-(+)-2-aryl-4-pentenoic acids 3 in most cases. The application of this biotransformation has been shown by a two-step synthesis of (R)-(-)-baclofen.

Full text of DOI:10.1016/S0040-4039(02)01449-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 6617–6620
Highly enantioselective biotransformations of
2-aryl-4-pentenenitriles, a novel chemoenzymatic approach to
(R)-(−)-baclofen
Mei-Xiang Wang* and Sheng-Min Zhao
Center for Molecular Science, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100080, China
Received 22 May 2002; revised 2 July 2002; accepted 12 July 2002
Abstract—Catalyzed by Rhodococcus sp. AJ270 microbial cells under mild conditions, a range of racemic 2-aryl-4-pentenenitriles
1 underwent effective hydrolysis to afford excellent yields of enantiomerically pure (R)-(−)-2-aryl-4-pentenamides 2 and
(S)-(+)-2-aryl-4-pentenoic acids 3 in most cases. The application of this biotransformation has been shown by a two-step synthesis
of (R)-(−)-baclofen. © 2002 Elsevier Science Ltd. All rights reserved.
Biotransformations of nitriles, either through a direct
conversion from a nitrile to a carboxylic acid catalyzed
by a nitrilase1 or through the nitrile hydratase-catalyzed
hydration of a nitrile followed by amide hydrolysis
catalyzed by an amidase,2 are effective and environmen-
tally benign methods for the production of carboxylic
acids and their amide derivatives. Microbial hydration
of acrylonitrile to acrylamide, for instance, has become
one of the largest industrial biotransformations in the
world.3 Recent endeavors by others4 and us5 have
demonstrated that biotransformations of nitriles are
also a unique complementary addition to the existing
asymmetric chemical and enzymatic synthesis of car-
boxylic acids and their derivatives. One of the distinct
features of enzymatic transformation of nitriles is not
only the formation of enantiopure carboxylic acids, but
also the straightforward generation of enantiopure
amides, valuable organonitrogen compounds in organic
synthesis.6
hydratase/amidase-containing biocatalyst. Compared
with other microbes reported, it displays broad enzy-
matic activity against almost all types of nitrile includ-
ing aromatic, heterocyclic and aliphatic ones, and both
amides and acids can be obtained in high yields from
appropriate nitriles.8 It shows excellent regioselectivity
in hydrolyzing aromatic dinitriles and a variety of
aliphatic dinitriles bearing a suitably placed second
chelating moiety.9 It has been demonstrated recently
that Rhodococcus sp. AJ270 is an efficient enantioselec-
tive biocatalytic system able to transform some racemic
nitriles such as a-alkyl5a,b and a-amino5c substituted
arylacetonitriles,
and
2-arylcyclopropanecarbo-
nitriles5d,e into the corresponding amides and acids in
enantiomerically pure form. Our interests in under-
standing the mechanism of enantioselective biotransfor-
mations of nitriles and in exploring this methodology to
create unique and versatile chiral intermediates led us
to undertake the current study. We wish to report
herein a direct and convenient microbial production of
2-aryl-4-pentenoic acid derivatives and a new chemo-
enzymatic synthesis of (R)-(−)-baclofen.
Rhodococcus sp. AJ270, a novel isolate from a soil
sample,7 appears to be a robust and useful nitrile
Scheme 1. Biotransformations of racemic nitriles 1.
Keywords: biotransformations; nitrile hydratase; amidase; chemoenzymatic synthesis; (R)-(−)-baclofen.
* Corresponding author. Tel.: +8610-62554628; fax: +8610-62569564; e-mail: mxwang@infoc3.icas.ac.cn
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)01449-1

6618
M.-X. Wang, S.-M. Zhao / Tetrahedron Letters 43 (2002) 6617–6620
To begin our study, we first examined the reaction of
2-phenyl-4-pentenenitrile 1a. Catalyzed by Rhodococcus
sp. AJ270 cells under very mild conditions,10 1a under-
went an efficient and a complete hydration reaction to
form amide 2a in a few hours. Highly optically pure
(R)-(−)-2-phenyl-4-pentenamide 2a11 and (S)-2-phenyl-
4-pentenoic acid 3a11 were obtained in almost quantita-
tive yield after about 3 days incubation (Scheme 1). As
illustrated in Table 1, an increase of and a decrease of
the enantiomeric excesses of the amide 2a and of the
acid 3a, respectively, during the progress of amide
hydrolysis indicated a kinetic resolution of the amide
effected by the S-enantioselective amidase involved in
Rhodococcus sp. AJ270 cells.
rate observed for these nitriles revealed that the nitrile
hydratase is very sensitive to the steric effect of sub-
stituents, even remote from the cyano function. On the
other hand, in most cases, the transformation of the
amide into the acid was effective, except for 2g which
remained almost intact after 7 days reaction (entry 12).
The sluggish hydrolysis of 2g was most probably due to
the crowding in the molecule caused by the ortho-
methylphenyl group, which inhibited amidase action. It
is interesting to note that substitution of the benzene
ring with a fluorine or a methoxy group facilitated the
conversion of amide into the acid while the presence of
chlorine or methyl groups slowed down the amide
hydrolysis.
To test the generality of this biotransformation and
also to gain a better understanding of the enantioselec-
tivity of nitrile hydratase and amidase enzymes, a num-
ber of 2-aryl-4-pentenenitrile analogs 1b–g were
synthesized13 and subjected to biocatalytic hydrolysis.
The low to moderate enantiomeric excess values
obtained for the recovered nitriles (R)-(+)-1c–e11
demonstrated a less effective enantioselectivity of the
S-nitrile hydratase (entries 5, 7 and 9). In contrast, the
amidase showed a very high S-enantioselection, which
was exemplified by the excellent enantiomeric excess
achieved for most of the (S)-acids. Only for (S)-(+)-
acids 3d and 3g, did slightly lower optical yields result.
It is worth noting that the overall enantioselectivity of
the reaction is the result of the combined actions of
S-nitrile hydratase and S-amidase. In other words, the
double selections of two S-enantioselective enzymes
account for the high enantiomeric excesses of the
products.
The results summarized in Table 1 show a dramatic
effect of the substituent attached to the benzene ring on
both the reaction rate and enantioselectivity. When a
para-fluorine group was introduced into the molecule,
no detrimental influence was observed, as the hydrolysis
of 1b proceeded similarly and as efficiently to that of
the parent substrate 1a to give enantiopure (R)-amide
2b and (S)-acid 3b in excellent yield. Introduction of
other substituents such as chlorine, methoxy and
methyl at different positions on the benzene ring, how-
ever, resulted in slow conversions and, in some cases,
low optical yield under identical reaction conditions.
For example, after a week of exposure to the biocata-
lyst, the reaction of nitriles 1c, 1d and 1e still gave a
considerable amount of the starting materials (entries 5,
7 and 9). Only when the concentration of the substrate
was halved and the incubation continued for a longer
period, did the hydration reaction go to completion,
with higher yields of the corresponding amide and acid
being obtained (entries 6, 8 and 10). The slow hydration
The biotransformation of 2-aryl-4-pentenenitriles pro-
vides a simple and straightforward approach to enan-
tiopure 2-aryl-4-pentenoic acids and their amide
derivatives. The former compounds were only available
from a somewhat tedious resolution14 using a chiral
base or from the less efficient multi-step asymmetric
synthesis.15 To demonstrate the synthetic potential of
enantioselective biotransformations of nitriles which
afford straightforwardly the enantiopure organonitro-
gen compounds, we carried out a synthesis of (R)-(−)-
baclofen.
Table 1. Biotransformations of racemic 2-aryl-pentenenitriles 1
Entry
Substrate
Ar
Conditionsa
Nitrile 1
Amide 2
Acid 3
Yield (%)b
Ee (%)c
Yield (%)b
Ee (%)c
Yield (%)b
Ee (%)c
1
2
3
4
5
6
7
8
1a
1a
1a
1b
1c
1c
1d
1d
1e
1e
1f
C6H5
C6H5
C6H5
1 mmol, 58 h
1 mmol, 69 h
1 mmol, 83 h
1 mmol, 60 h
1 mmol, 7 d
0.5 mmol, 6 d
1 mmol, 7 d
0.5 mmol, 73 h
1 mmol, 8 d
0.5 mmol, 7 d
0.5 mmol, 6 d
0.5 mmol, 7 d
–
–
–
–
–
–
–
–
59.1
–
7.1
–
47.0
–
–
–
62
49
39
50
20
44
37
47
28
49
49
93
70.8
99.2
\99.5
\99.5
99.2
38
49
61
50
39
50
49
51
26
49
49
4
\99.5
96.8
80.1
99.3
99.2
\99.5
89.7
87.4
\99.5
94.3
\99.5
78.5
4-F-C6H4
4-Cl-C6H4
4-Cl-C6H4
4-MeO-C6H4
4-MeO-C6H4
4-Me-C6H4
4-Me-C6H4
3-Me-C6H4
2-Me-C6H4
36
Trace
11
–
41
99.3
\99.5
\99.5
3.7
\99.5
\99.5
3.2
9
10
11
12
–
–
1g
Trace
a Rhodococcus sp. AJ270 cells (2 g wet weight) in phosphate buffer (0.1 M. pH 7.0, 50 ml) were used. The reaction conditions were not optimized.
b Isolated yield.
c Determined by chiral HPLC.12

M.-X. Wang, S.-M. Zhao / Tetrahedron Letters 43 (2002) 6617–6620
6619
Kobayashi, M.; Shimizu, S. FEMS Microbiol. Lett. 1994,
120, 217–224.
2. (a) Asano, Y.; Tani, Y.; Yamada, H. Agric. Biol. Chem.
1980, 44, 2251–2252; (b) Asano, Y.; Fujishiro, K.; Tani,
Y.; Yamada, H. Agric. Biol. Chem. 1982, 46, 1165–1174;
(c) Asano, Y.; Tachibana, M.; Tani, Y.; Yanada, H.
Agric. Biol. Chem. 1982, 46, 1175–1181.
3. (a) Nagasawa, T.; Schimizu, H.; Yamada, H. Appl.
Microbiol. Biotechnol. 1993, 40, 189–195; (b) Kobayashi,
M.; Nagasawa, T.; Yamada, H. Trends Biotechnol. 1992,
10, 402–408.
4. For a recent review, see: Sugai, T.; Yamazaki, T.;
Yokoyama, M.; Ohta, H. Biosci. Biotech. Biochem. 1997,
61, 1419–1427.
Scheme 2. Synthesis of (R)-(−)-baclofen.
5. (a) Wang, M.-X.; Lu, G.; Ji, G.-J.; Huang, Z.-T.; Meth-
Cohn, O.; Colby, J. Tetrahedron: Asymmetry 2000, 11,
1123–1135; (b) Wang, M.-X.; Li, J.-J.; Ji, G.-J.; Li, J.-S.
J. Mol. Cat. B—Enzym. 2001, 14, 77–83; (c) Wang,
M.-X.; Lin, S.-J. Tetrahedron Lett. 2001, 42, 6925–6927;
(d) Wang, M.-X.; Feng, G.-Q. Tetrahedron Lett. 2000,
41, 6501–6505; (e) Feng, G.-Q.; Wang, M.-X. Chinese J.
Chem. 2001, 19, 113–115.
(R)-(−)-baclofen, as a specific GABAB (g-aminobutyric
acid) receptor agonist and a potent antispastic agent,16
has attracted much attention in recent years. Syntheses
of optically active (R)-(−)-baclofen have been reported
either through the chymotrypsin mediated hydrolysis of
dimethyl 3-(4-chlorophenyl)glutarate followed by sev-
eral chemical transformations,17 multi-step asymmetric
reactions employing the Evans chiral enolate proto-
col,18 or multi-step stereoselective transformations from
(S)-glutamic acid19 and (S)-3-(4-chlorophenyl)pyrroli-
dine.20 With (R)-2-(4-chlorophenyl)-4-pentenamide 2c
in hand, we performed a direct and convenient two-step
synthesis of (R)-(−)-baclofen utilizing routine reduction
and oxidation reactions. Thus, in the presence of
LiAlH4, (R)-(−)-amide 2c was reduced to (R)-2-(4-
chlorophenyl)-4-pentenylamine 421 in 77% yield. Direct
oxidation of 4 using OsO4 and Jones’ agent led to the
formation of (R)-(−)-baclofen 522 in 53% yield (Scheme
2). To our knowledge, this is the shortest synthetic
route to (R)-(−)-baclofen known so far.
6. The Chemistry of Amides; Patai, S., Ed.; John Wiley and
Sons, 1975.
7. Blakey, A. J.; Colby, J.; Williams, E.; O’Reilly, C. FEMS
Microbiol. Lett. 1995, 129, 57–62.
8. Meth-Cohn, O.; Wang, M.-X. J. Chem. Soc., Perkin
Trans. 1 1997, 1099–1104.
9. Meth-Cohn, O.; Wang, M.-X. J. Chem. Soc., Perkin
Trans. 1 1997, 3197–3204.
10. General procedure for the biotransformation reaction: To
an Erlenmeyer flask (150 ml) with a screw cap was added
Rhodococcus sp. AJ270 cells (2 g wet weight) and potas-
sium 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. Nitrile 1 was 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 the specified period of time by removing
the biomass by filtration through a Celite pad. The
resulting aqueous solution was basified to pH 12 with
aqueous NaOH (2 M). Extraction with ethyl acetate
gave, after drying (MgSO4), removal of solvent, and
column chromatography, the unconverted nitrile 1 and
amide 2. The aqueous solution was then acidified using
aqueous HCl (2 M) to pH 2 and extracted with ethyl
acetate to give acid 3.
In summary, we have shown a highly enantioselective
synthesis of 2-aryl-4-pentenoic acids and their amides
from the biotransformations of easily prepared13 2-aryl-
4-pentenenitriles utilizing Rhodococcus sp. AJ270 cells
under very mild conditions. The potential applications
of this reaction have been demonstrated by the prepara-
tion of (R)-(−)-baclofen by a very short route. Further
synthetic applications of enantiopure (S)-2-aryl-4-pen-
tenoic acids and their (R)-amides are being actively
investigated in this laboratory and will be published in
due course.
11. The absolute configurations of (S)-2-aryl-4-pentenoic
acids 3 were determined by measuring and comparing
their specific rotations with that of authentic (S)-2-
phenyl-4-pentenoic acid reported in the literature (see:
Refs. 14 and 15). The absolute configurations of (R)-2-
aryl-4-pentenamides 2 and of (R)-2-aryl-4-pentenenitriles
amides 1 were obtained by converting them to the acids
and then measuring the specific rotation. Selected data:
Compound 2c: solid (43%); Mp 99–100.3°C; [h]2D5 −71.0
(c 2.3, CHCl3); ee 99.3%; IR (KBr) wmax 3417, 3200
Acknowledgements
We thank the Major Basic Research Development Pro-
gram (No. G2000077506), the National Natural Science
Foundation of China and the Chinese Academy of
Sciences for their financial support. M.-X.W. also
thanks O. Meth-Cohn and J. Colby for discussion.
1
(CONH2), 1653 (CꢀO), 1620 cm−1 (CꢀC); H NMR (300
MHz, CDCl3) l 7.30 (d, J=8.3 Hz, 2H, ArH), 7.24 (d,
J=8.3 Hz, 2H, ArH), 5.99 (s, br., 1H, CONHH), 5.73–
5.53 (m, 1H), 5.53 (s, br., 1H, CONHH), 5.05 (d, J=12.3
Hz, 1H), 5.00 (d, J=10.9 Hz, 1H), 3.44 (t, J=7.5 Hz,
1H), 2.90–2.80 (m, 1H), 2.53–2.46 (m, 1H); 13C NMR
(CDCl3) l 175.1, 137.7, 135.2, 133.3, 129.4, 129.0, 117.3,
References
1. (a) Harper, D. B. Biochem. Soc. Trans. 1976, 4, 502–504;
(b) Harper, D. B. Biochem. J. 1977, 165, 309–319; (c)

6620
M.-X. Wang, S.-M. Zhao / Tetrahedron Letters 43 (2002) 6617–6620
51.9, 37.3; MS (EI) m/z (%) 211 (M++2, 5), 209 (M+, 14),
17. Cheˆnevert, R.; Desjardins, M. Tetrahedron Lett. 1991, 32,
4249–4250.
18. Schoenfelder, A.; Mann, A.; Coz, A. L. Synlett 1993,
63–64.
19. Herdeis, C.; Hubmann, H. P. Tetrahedron: Asymmetry
1992, 9, 1213–1221.
168 (23), 167 (36), 166 (39), 165 (100), 153 (12), 150 (30),
142 (10), 140 (30), 131 (32), 130 (50), 129 (43), 128 (19),
127 (33), 125 (64). Found: C, 62.84; H, 5.71; N, 6.48.
C11H12NOCl requires: C, 63.01; H, 5.77; N, 6.68. Com-
pound 3c: oil (50%); [h]D25 +49.8 (c 2.75, CHCl3); ee>99.5%;
IR (KBr) wmax 3400–2800 (COOH), 1708 (CꢀO), 1644 cm−1
(CꢀC); 1H NMR (300 MHz, CDCl3) l 11.4 (s, br., COOH),
7.37 (d, J=8.5 Hz, 2H, ArH), 7.32 (d, J=8.5 Hz, 2H, ArH),
5.80–5.69 (m, 1H), 5.14 (d, J=18.2 Hz, 1H), 5.09 (d, J=10.8
Hz, 1H), 3.70 (t, J=7.6 Hz, 1H), 2.92–2.82 (m, 1H),
2.73–2.53 (m, 1H); 13C NMR (CDCl3) l 179.3, 136.2, 134.4,
133.5, 129.4, 128.9, 117.6, 50.7, 37.0; MS (EI) m/z (%) 212
(M++2, 4), 210 (M+, 11), 171 (35), 169 (100), 167 (26), 165
(78), 143 (12), 141 (40), 129 (70), 128 (45), 77 (35). HRMS:
209.0372 (M−H). C11H10O2Cl requires 209.0374.
20. Yoshifuji, S.; Kaname, M. Chem. Pharm. Bull. 1995, 43,
1302–1306.
21. Selected data for compound 4: oil (77.4%); [h]2D5 −12 (c 1,
1
CHCl3); IR (KBr) wmax 1640 (CꢀC), 1595 cm−1; H NMR
(300 MHz, CDCl3) l 7.31 (d, J=8.3 Hz, 2H, ArH), 7.13
(d, J=8.3 Hz, 2H, ArH), 5.74–5.60 (m, 1H), 5.00 (d,
J=14.9 Hz, 1H), 4.96 (d, J=8.6 Hz, 1H), 2.95–2.99 (m,
1H), 2.81–2.88 (m, 1H), 2.67–2.70 (m, 1H), 2.30–2.44 (m,
2H); 13C NMR (CDCl3) l 141.4, 136.1, 132.2, 129.3,
128.7, 116.5, 48.6, 47.2, 38.2; MS (EI) m/z (%) 195 (M+,
2), 30 (100). HRMS: 196.0889 (M+H). C11H15NCl
requires: 196.0887.
12. Enantiomeric excess values were obtained from HPLC
analysis using a Chiracel OD, OJ or OB column with a
mixture of n-hexane and 2-propanol (9:1) as the mobile
phase.
13. Makosza, M.; Serafin, B. Rocz. Chem. 1965, 39, 1401–
1409. Chem. Abs. 1966, 64, 17474g.
22. Selected data for compound 5: solid (53.2%); Mp 161°C
(dec.) (lit.20 mp. 215°C for (R)-baclofen hydrochloride);
[h]2D5 −2.0 (c 0.2, H2O) [lit.20 [h]D25 −1.3 (c 0.2, H2O)]; ee
91% (HPLC analysis using a CR(+) chiral column); IR
14. Veldstra, H.; Van De Westeringh, C. Recl. Trav. Chim.
(KBr) wmax 3400–2800 (br.), 1627 (CꢀO), 1572, 1540 cm−1
;
1951, 70, 1113–1226.
15. Zaponakis, G. S.; Katerinopoulos, H. E. Tetrahedron
Lett. 2001, 42, 6393–6396.
16. Olpe, H. R.; Demie´ville, H.; Baltzer, V.; Bencze, W. L.;
Koella, W. P.; Wolf, P.; Haas, H. L. Eur. J. Pharmacol.
1978, 52, 133–136.
1H NMR (300 MHz, D2O) l 7.33 (d, J=7.7 Hz, 2H,
ArH), 7.22 (d, J=7.5 Hz, 2H, ArH), 3.22 (d, J=11.3 Hz,
2H), 3.06–3.14 (m, 1H), 2.55–2.37 (m, 2H); 13C
NMR(D2O) l 179.4, 138.1, 132.9, 129.3, 129.1, 44.0, 42.1,
40.7.