Resolution of racemic 2-chlorophenyl glycine with immobilized penicillin G acylase

Article abstract of DOI:10.1016/j.tetasy.2008.10.015

Racemic 2-chlorophenyl glycine has been resolved to obtain (S)-α-amino-(2-chlorophenyl)acetic acid with >99% enantiomeric purity via enantioselective hydrolysis of its N-phenylacetyl derivative with penicillin G acylase immobilized on Eupergit C. The reso

Full text of DOI:10.1016/j.tetasy.2008.10.015

Tetrahedron: Asymmetry 19 (2008) 2363–2366
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Resolution of racemic 2-chlorophenyl glycine with immobilized
penicillin G acylase
*
Nitin W. Fadnavis , Avala Vedamayee Devi, Lakshmi Swarnalatha Jasti
Biotransformations Laboratory, Indian Institute of Chemical Technology, Uppal Road, Habsiguda, Hyderabad 500 007, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Racemic 2-chlorophenyl glycine has been resolved to obtain (S)-a-amino-(2-chlorophenyl)acetic acid
Received 5 August 2008
Accepted 15 October 2008
Available online 19 November 2008
with >99% enantiomeric purity via enantioselective hydrolysis of its N-phenylacetyl derivative with pen-
icillin G acylase immobilized on Eupergit C. The resolution can be conveniently performed in water at
substrate concentration of 10% in 5 h with an enzyme:substrate ratio of 1:5 (w/w), and the enzyme
can be recycled >25 times without any loss of activity. The unwanted (R)-enantiomer can be easily race-
mized for recycling.
Ó 2008 Published by Elsevier Ltd.
1. Introduction
and necessity of fermenting a specific bacterial strain limit the
application of this process on a large scale.
(S)-2-Amino-(2-chlorophenyl)ethanoic acid [(S)-2-chlorophenyl
glycine] 1 is an important intermediate in the synthesis of Clopido-
grel, a potent oral antiplatelet agent often used in the treatment of
coronary artery disease, peripheral vascular disease, and cerebro-
vascular disease.1 The drug is marketed by Bristol-Myers Squibb
and Sanofi-Aventis under the trade names Plavix and Iscover, and
by Sun Pharmaceuticals under the trade name Clopilet. It is also
used, along with aspirin, for the prevention of thromboembolism
after placement of intracoronary stent.
Although the synthesis of clopidogrel can be carried out via sev-
eral routes, most routes utilize either the 2-chloromandalate or 2-
chlorophenyl glycine derivatives as starting materials. The final
product is usually made in racemic form and resolved via fractional
crystallization with a resolving agent such as camphor sulfonic
acid.1–7 Economically, the use of an enantiomerically pure reagent
at the start of a synthetic sequence is more cost effective and less
polluting. Therefore recently, strategies for the preparation of
enantiomerically pure 2-chloromandalate using enzymatic routes,
such as kinetic resolution using hydrolases8,9 or enantioselective
reduction of a prochiral ketoacid,10 have been successfully demon-
strated. The resolution of racemic 2-chlorophenyl glycine has been
Herein, we report a simple methodology for obtaining enantio-
merically pure (S)- -amino-(2-chlorophenyl)acetic acid on a mul-
a
tigram scale, which is based on the principle of enantioselective
hydrolysis of an N-phenylacetyl derivative of a racemic amino acid
with the commercially available enzyme penicillin G acylase
(E.C.3.5.1.11) immobilized on Eupergit C.15 The enzyme selectively
hydrolyzes only the (S)-enantiomer to give the (S)-amino acid with
ee >99%. It is possible to carry out the reaction at a substrate con-
centration as high as 20% and recycle the enzyme >25 times with-
out loss of activity (Scheme 1).
2. Results and discussion
Penicillin G acylase (penicillin amidohydrolase, EC 3.5.1.11)
cleaves the acyl chain of penicillins to yield 6-amino penicillanic
acid (6-APA) and the corresponding organic acid, and is used
industrially, mainly for the production of 6-APA and semi-syn-
thetic antibiotics.16–18 It is also useful in peptide synthesis19 and
in resolution of secondary alcohols,20 amino alcohols,21,22 amino
acids,23–25 b-hydroxy-
etc. The stereochemical preference of penicillin G acylase toward
the -enantiomer of amino acids such as 2-phenyl glycine and 4-
a
-amino acids,26 aryloxycarboxylic acids,27
carried out by a classical method using D-camphor sulfonic acid in
L
rather low yield (42% of theoretical value).11 Alternatively, the
methyl ester of 2-chlorophenyl glycine can be resolved with (+)-
tartaric acid.12 Although enantioselective hydrolysis of 2-chloro-
phenyl glycine amide by microbes possessing amidase activity
has been reported,13,14 the low solubility of the amide in water
hydroxyphenyl glycine is well known.28 However, hydrolases gen-
erally do not show high selectivities towards ortho-substituted
substrates due to steric hindrance,9,29–31 and it is necessary to
screen several enzymes and manipulate the reaction conditions
and substrate structure in order to achieve the required enantiose-
lectivity. It was gratifying to discover that penicillin G acylase rec-
ognized only the (S)-enantiomer of N-phenylacetyl derivative of 2-
chlorophenyl glycine for hydrolysis.
* Corresponding author. Tel.: +91 40 27191631; fax: +91 40 27160512.
E-mail addresses: fadnavis@iict.res.in, fadnavisnw@yahoo.com (N.W. Fadnavis).
0957-4166/$ - see front matter Ó 2008 Published by Elsevier Ltd.
doi:10.1016/j.tetasy.2008.10.015

2364
N. W. Fadnavis et al. / Tetrahedron: Asymmetry 19 (2008) 2363–2366
NHCOCH2Ph
COOH
NHCOCH2Ph
COOH
NH2
COOH
pH 8.0, 27oC
PhCH2COOH
+
+
Immobilized Pen G Acylase
Cl
Cl
Cl
(S)-1: (46%, e.e.>99%)
2
(RS)-
(R)-2: (50%, e.e.>99%)
Racemization by heating at 170oC, 1 5min
Scheme 1.
2.1. Effect of substrate concentration on reaction rate and
enantioselectivity
20
18
16
14
12
10
8
In our earlier studies on the resolution of 2-aminobutanol,21 we
observed that the enantiomeric purity of the product was depen-
dent on the substrate concentration and conversion due to high
apparent Km value (Kmapp) for the substrate and moderate E value
of 44. Thus, the effect of substrate concentration on the rate and
enantioselectivity of the reaction was studied. To determine the
E-value, experiments were conducted with enantiomerically pure
(R)- and (S)-amides, separately, and it was confirmed that the en-
zyme did not accept the (R)-enantiomer (E ꢀ 200).
6
4
In the second experiment, the enzyme concentration was held
constant at 50 mg in a reaction volume of 5 mL. Substrate concen-
trations were varied from 0.5% (w/v) to 20% (w/v). Reactions were
conducted in glass vials with magnetic stirring. It was observed that
the enantioselectivity of the reaction did not change with substrate
concentration and the hydrolytic reaction stopped at 48–50% con-
version, even after prolonged contact (24 h) with the enzyme. A typ-
ical reaction curve of conversion versus time is shown in Figure 1.
2
0
-2
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
[substrate], M
Figure 2. Effect of substrate concentration on the initial rate of hydrolysis of 2 by
penicillin G acylase immobilized on Eupergit C in aqueous solution at pH 8.0, 27 °C.
50
40
30
20
10
0
about 0.3 M (9% w/v) (Fig. 2). Thus, the preparative resolution
was carried out at a substrate concentration of 10%. With a sub-
strate to enzyme ratio of 5:1 (w/w), the reaction typically reached
about 48% conversion in 5 h. The enzyme was filtered, and the
aqueous solution was cooled, acidified to pH 2, and extracted with
dichloromethane. The resolved (S)-amino acid was recovered by
evaporating the aqueous layer to dryness and purified (yield 46%,
92% of theoretical value; ee >99%). The dichloromethane layer
was evaporated to obtain unreacted (R)-2 along with phenylacetic
acid. This was extracted several times with boiling cyclohexane to
remove the phenylacetic acid. The residual amide was finally
recrystallized to obtain pure (R)-2. For a sample of the amide with
ee >99%, the enzymatic reaction was allowed to proceed overnight.
2.3. Enzyme recycle
-1
0
1
2
3
4
5
6
7
8
Reaction Time, h
Enzyme recycle studies were carried out in an Amicon stirred
cell as a representative filter reactor equipped with a mechanical
stirrer. The total reaction volume was 50 mL, and reactions were
carried out using a 10% substrate solution and 1 g of immobilized
enzyme. The reactants were stirred at 80–100 rpm. The reaction
was followed by HPLC and stopped after 5 h when the rate of reac-
tion became considerably slow. The reactants were then filtered
out, and next batch was carried out with a fresh substrate solution
using the same enzyme. Studies with 25 recycles showed no loss of
enzyme activity or enantioselectivity.
Figure 1. Hydrolysis of 2-(2-chlorophenyl)-2-[(2-phenylacetyl)amino]acetic acid 2
by penicillin G acylase immobilized on Eupergit C in aqueous solution at pH 8.0,
27 °C. [(R,S)-2] = 0.33 M, 10% (w/v). Immobilized enzyme, 50 mg, in 5 mL reaction
volume.
The reaction followed the usual Michaelis–Menten kinetics
(Fig. 2). Using nonlinear regression analysis of the initial rates,
the apparent Vmax (Vmax,app) and apparent Km (Km,app) were found
to be 19 mM/min/g and 74 mM, respectively (correlation coeffi-
cient r = 0.988).
2.2. Preparative scale enzymatic resolution
2.4. Racemization of (R)-2
From the kinetic data, it was observed that the reaction velocity
reached a plateau of 0.8 mM/min at a substrate concentration of
Racemization of the unwanted (R)-enantiomer was carried out
by the method reported by Grabley.32 A mixture of phenylacetic

N. W. Fadnavis et al. / Tetrahedron: Asymmetry 19 (2008) 2363–2366
2365
acid and (R)-2 recovered after enzymatic reaction was heated in an
residue gave (R)-phenylacetylamino acid
2
(2.5 g, 50%; mp
oil bath at 170 °C. The melt was held at this temperature for
15 min, and then cooled. Measurement of the specific rotation
indicated complete racemization. No decomposition products were
found in HPLC analysis. The racemization was confirmed by HPLC
analysis on a chiral stationary column.
210 °C. Ee >99%; ½a D25
¼ ꢃ96 (c 1, chloroform).
ꢂ
The amino acid (S)-1 was recovered by evaporating the aqueous
solution on a rotavaporator after adjusting the pH to 7.0, redissolv-
ing in hot isopropanol to remove salts, and evaporating the isopro-
panol layer (1.4 g, 92% of theoretical value). The enantiomeric
purity of the product was determined by chiral HPLC analysis after
converting it to its phenylacetyl derivative. ½a D25
¼ þ89 (c 1, 1 M
ꢂ
HCl),33 (lit.11 +115.6 (c 1, 1 M HCl). Ee >99%. 1H NMR (D2O,
200 MHz : d ppm 4.68 (br s, 2H, NH2), 4.94 (br s, 1H, COOH),
5.20 (s, 1H, CHCOOH), 7.40–7.64 (m, 4H, aromatic).
3. Conclusion
The present methodology provides an excellent alternative to
the existing route for the resolution of 2-chlorophenyl glycine with
almost quantitative yield and consistent enantiomeric purity of
>99%. The enzyme penicillin G acylase is commercially available
in large quantities at a reasonable price and can be recycled several
times. Although we have carried out reactions at 10% substrate
concentration, it is possible to use a substrate concentration as
high as 20%. Considering the observed Vmax of 19 mM/min/g, a
judicious design of a bioreactor can provide an excellent process
on an industrial scale.
4.3. Reverse phase HPLC analysis
The hydrolysis of phenylacetyl derivative 2 was followed by re-
verse phase HPLC. Column C-8 (250 ꢁ 5 mm), Chrompack, The
Netherlands. Mobile phase, 50% acetonitrile–water containing
0.1% perchloric acid. Flow rate, 0.7 ml/min. Detection wavelength,
220 nm. Retention times: 1: 4.14; phenylacetic acid 7.44; 2:
9.55 min.
4.3.1. HPLC analysis with chiral stationary phase
4. Experimental
Enantiomeric purity was determined by HPLC analysis on Chi-
ralcel AD-H column (250 ꢁ 5 mm), Daicel Chemical Industries, Ja-
pan. Mobile phase, 6% 2-propanol in hexane containing 0.1%
trifluoroacetic acid. Flow rate, 1 ml/min. Detection wavelength,
220 nm. Retention times (S)-2: 44. 38; (R)-2: 48.06 min.
2-Chlorophenyl glycine was purchased from Aldrich. Immobi-
lized Penicillin G acylase was a gift from M/s KDL Biotech Ltd, Savr-
oli village, India. HPLC analyses were carried out on Hewlett
Packard HP1090 unit with diode array detector and HP CHEM STATION
software. Curve fittings were performed with Graph Pad Prism ver-
sion 5, GRAPH PAD Software, San Diego, California, USA.
Acknowledgment
We thank the Council of Scientific and Industrial Research, New
Delhi for financial support.
4.1. 2-(2-Chlorophenyl)-2-[(2-phenylacetyl)amino]ethanoic
acid 2
References
The racemic amino acid 1 (18.56 g, 0.1 mol) dissolved in 4 M
NaOH (50 mL) was stirred in an ice-bath. Phenylacetyl chloride
(17 g, 0.11 mol) was added dropwise with 4 M NaOH (50 mL) with
vigorous stirring. After complete addition, the reactants were stir-
red overnight and extracted with dichloromethane. The aqueous
layer was then cooled in ice and acidified with 6 M HCl. The precip-
itated phenylacetyl derivative was filtered, washed with cold
water, dried, and finally recrystallized from ethylacetate–hexane
to obtain 2 (28 g, 92%). Mp 208–209 °C. 1H NMR (CDCl3 + DMSO-
d6, 200 MHz): d ppm 3.50 (s, 2H, COCH2Ph), 5.71 (d, 1H, J = 7.3 Hz
CHCOOH), 7.14–7.42 (m, 9H, aromatic), 8.00 (d, 1H, J = 7.3 Hz, –
NH). 13C NMR (CDCl3 + DMSO-d6, 200 MHz): d ppm 176.1 (Ph-CO,
175.2 (COOH), 140.3, 138.3, 134.3, 134.0, 133.0, 131.8, 131.2 aro-
matic), 58.7 (CH–NH), 44.0 (PhCH2).
1. Badorc, A; Frehel, D. U.S. Patent 4,847,265, 1989.
2. Daniel, A.; Ferrand, C.; Maffrand, J. P. U.S. Patent 4,529, 596, 1985.
3. Bakonyi, M.; Csatari nee Nagy, M.; Molnar, L.; Gajary, A.; Alattyani, E. PCT Int.
Appl. WO 1998/051689, 1998.
4. Bousquet, A.; Musolino, A. PCT Int. Appl. WO/1999/018110, 1999.
5. Bakonyi, M.; Csatari nee Nagy, M. Molnar nee Bako, E. Gajary, A.; Alattyani, E.
US Patent 6,180,793, 2001.
6. Mukarram, M. S. J.; Merwade, Y. A.; Khan, R. A. U.S. Patent 7,291, 735 B2, 2007.
7. Castaldi, G.; Barreca, G.; Bologna, A. U.S. Patent 7,329,751 B2, 2008.
8. Wang, P.-Y.; Chen, T.-L.; Tsai, S.-W. Enzyme Microbiol. Technol. 2006, 39, 930–
935.
9. Uhm, K.-N.; Lee, S.-J.; Kim, H.-K.; Kang, H.-Y.; Lee, Y. J. Mol. Catal. B: Enzym.
2007, 45, 34–38.
10. Tadashi, E.; Nobuyasu, O.; Sayaka, I.; Takashi, S. Org. Biomol. Chem. 2007, 5,
1175–1176.
11. Lin, S. S.-S.; Chen, C.-C.; U.S. Patent 2004/0176637A1, 2004.
12. Reddy, B. S. PCT Int. Appl. WO 2006/003671 A1, 2006.
13. Katoh, O.; Uragaki, T.; Nakamura, T. PCT Int. Appl. WO 0187819, 2001.
14. Asano, Y. European Patent EP1770166, 2007.
15. The immobilized enzyme has an activity of 190 units/g wet at 37 °C. One unit is
defined as one micromole of penicillin G hydrolyzed per minute.
16. Wegman, M.; Janssen, M.; van Rantwijk, F.; Sheldon, R. Adv. Synth. Catal. 2001,
343, 559–576.
4.2. Enzymatic reaction
Substrate 2 (5 g, 16.5 mmoles) was suspended in water (20 mL)
and dissolved with the addition of 5% ammonia solution. The pH
was adjusted to 8.0 with 2 M HCl and the volume was finally made
up to 50 mL with water. The enzyme (1 g wet) was added and the
reaction mixture was stirred at room temperature (27 °C) in an
Amicon stirred ultrafiltration cell, model 8200 at 80–100 rpm.
The reaction was followed by reverse phase HPLC. Typically, the
reaction stopped at approximately 48% conversion in 5 h. The reac-
tants were filtered out, and the enzyme was washed with water
(3 ꢁ 5 mL) and the combined aqueous solution was cooled in ice
and acidified with 6 M HCl to pH 2. The precipitated mixture of
(R)-phenylacetylamino acid 2 and phenylacetic acid 3 was ex-
tracted with dichloromethane (3 ꢁ 10 mL) and the solvent was
evaporated. Phenylacetic acid was recovered from the mixture by
extracting repeatedly with boiling cyclohexane (1.0 g, 47.5%). The
´
17. Arroyo, M.; de la Mata, I.; Acebal, C.; Castillon, M. Appl. Microbiol. Biotechnol.
2003, 60, 507–514.
18. Chandel, A. K.; Venkateswar Rao, L.; Lakshmi Narasu, M.; Singh, O. V. Enzyme
Microbiol. Technol. 2008, 42, 199–207.
19. van Langen, L. M.; van Rantwijk, F.; Svedas, V. K.; Sheldon, R. A. Tetrahedron:
Asymmetry 2000, 11, 1077–1083.
20. Fuganti, C.; Grasselli, P.; Seneci, P. F.; Servi, S.; Casati, P. Tetrahedron Lett. 1986,
27, 2061–2062.
21. Fadnavis, N. W.; Sharfuddin, M.; Vadivel, S. K. Tetrahedron: Asymmetry 1999, 10,
4495–4500.
22. Fadnavis, N. W.; Radhika, K. R.; Vedamayee Devi, A. Tetrahedron: Asymmetry
2006, 17, 240–244.
23. Cardillo, G.; Tolomelli, A.; Tomasini, C. J. Org. Chem. 1996, 61, 8651–8654.
24. Fadnavis, N. W.; Sharfuddin, M.; Vadivel, S. K.; Bhalerao, U. T. J. Chem. Soc.,
Perkin Trans. 1 1997, 3577–3578.
25. Landis, B. H.; Ng, J. S.; Topgi, R.; Yonan, E. E.; Wang, P. T. U.S. Patent 6,214,609,
2001.

2366
N. W. Fadnavis et al. / Tetrahedron: Asymmetry 19 (2008) 2363–2366
26. Bossi, A.; Cretich, M.; Righetti, P. G. Biotechnol. Bioeng. 1998, 60, 454–461.
27. Calleri, E.; Massolini, G.; Loiodice, F.; Fracchiolla, G.; Temporini, C.; Félix, G.;
Tortorella, P.; Caccialanza, G. J. Chromatogr., A 2002, 958, 131–140.
28. Basso, A.; Braiuca, P.; De Martin, L.; Ebert, C.; Gardossi, L.; Linda, P. Tetrahedron:
Asymmetry 2000, 11, 1789–1796.
31. Sharma, S. K.; Husain, M.; Kumar, R.; Samuelson, L. A.; Kumar, J.; Watterson, A.
C.; Parmar, V. S. Pure Appl. Chem. 2005, 7, 209–226.
32. Grabley, S. U.S. Patent 4,638,086, 1987.
33. The reported specific rotation of +115.6 in Ref. 11 may be due to contamination
with (+)-camphor sulfonic acid. We have consistently obtained the product
29. Kawasaki, M.; Nakamura, K.; Kawabata, S. J. Mol. Catal. B: Enzym. 1999, 6, 447–
451.
with
a specific rotation of +89, and chiral HPLC analysis indicated an
enantiomeric purity of >99%.
30. Wimmer, Z.; Skouridou, V.; Zarevúcka, M.; Šaman, D.; Kolisis, F. N. Tetrahedron:
Asymmetry 2004, 15, 3911–3917.