Lipase catalyzed regio- and stereospecific hydrolysis: Chemoenzymatic synthesis of both (R)- and (S)-enantiomers of α-lipoic acid

Article abstract of DOI:10.1016/S0957-4166(98)00447-9

Native lipase of Candida rugosa (EC 3.1.1.3) enantioselectively and regiospecifically hydrolyses the n-butyl ester of 2,4-dithioacetyl butanoic acid either at the carboxylic acid terminus or at the α-thioacetate to provide enantiomerically pure (R)-2,4-dithioacetyl butyric acid and (S)- butyl 2-thio-4-thioacetyl butyrate (ee >98%) while the lipase modified by treatment with diethyl p-nitrophenyl phosphate attacks only the α- thioacetate giving enantiomerically pure (S)-butyl 2-thio-4-thioacetyl butyrate. These enantiomerically pure intermediates can be used as chiral building blocks to obtain both(S)- and (R)-enantiomers of α-lipoic acid and their analogues.

Full text of DOI:10.1016/S0957-4166(98)00447-9

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 4109–4112
Lipase catalyzed regio- and stereospecific hydrolysis:
chemoenzymatic synthesis of both (R)- and (S)-enantiomers of
α-lipoic acid †
N. W. Fadnavis,∗ Ravi Luke Babu, S. Kumara Vadivel, Ashlesha A. Deshpande and
U. T. Bhalerao
Biotransformations Laboratory, Organic Division-I, Indian Institute of Chemical Technology, Hyderabad 500 007, India
Received 24 September 1998; accepted 30 October 1998
Abstract
Native lipase of Candida rugosa (EC 3.1.1.3) enantioselectively and regiospecifically hydrolyses the n-butyl
ester of 2,4-dithioacetyl butanoic acid either at the carboxylic acid terminus or at the α-thioacetate to provide
enantiomerically pure (R)-2,4-dithioacetyl butyric acid and (S)-butyl 2-thio-4-thioacetyl butyrate (ee >98%) while
the lipase modified by treatment with diethyl p-nitrophenyl phosphate attacks only the α-thioacetate giving
enantiomerically pure (S)-butyl 2-thio-4-thioacetyl butyrate. These enantiomerically pure intermediates can be
used as chiral building blocks to obtain both (S)- and (R)-enantiomers of α-lipoic acid and their analogues. © 1998
Elsevier Science Ltd. All rights reserved.
Lipases are now routinely being used in resolution of racemic alcohols and carboxylic acids.1 These
versatile reagents are not only enantioselective but also display diastereoselectivity2 and regiospecificity3
in several cases. Butyl 2,4-dithioacetyl butyrate 1 possesses three hydrolysable groups, namely an ester
of α-carboxylic acid, an acetate of a thiol group attached to a stereogenic carbon atom and an acetate of
thiol group placed two carbon atoms away from the stereogenic centre. It was interesting to determine
whether the enzyme binding site discriminates between the carboxylic acid and the thiol functionality,
and whether this leads to enantiomerically pure products that can be converted to enantiomerically pure
α-lipoic acid, an important growth factor for a variety of micro-organisms, and also a co-factor for the
multi-enzyme complex that catalyses oxidative decarboxylation of α-keto acids.4
Thus 1 (2 mmol), prepared by the method of Claeson,5 was hydrolysed in 0.05 M phosphate buffer (100
ml) at pH 7.5 in the presence of lipase from Candida rugosa (Sigma, USA; 50 mg) at room temperature.
The reaction stopped at 50% conversion after 3 h and the recovered products were identified as 2,4-
dithioacetyl butyric acid 2 and butyl 2-thio-4-thioacetyl butyrate 3. The enantiomeric purity of 2 and
∗
†
Corresponding author. E-mail: root@csiict.ren.nic.in
IICT Communication no. 4100.
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00447-9

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
4110
N. W. Fadnavis et al. / Tetrahedron: Asymmetry 9 (1998) 4109–4112
3 was determined by chiral HPLC and both the products were found to have ee >98% with opposite
configuration6 (Scheme 1).
Scheme 1.
1
The presence of an intact ester group, the disappearance of a singlet in the H NMR spectra of 3
at δ 2.2 ppm and its failure to form a thiolactone when treated with NaH in THF confirmed that the
primary thioacetate was not hydrolysed. The stereochemistry of 2 and 3 was established as (R)-2 and
(S)-3 by conversion to the corresponding enantiomer of α-lipoic acid (Scheme 2). Enantiomerically pure
2,4-dithioacetyl butyric acid 2 was reduced with BH3·DMS to the alcohol which was oxidised with PCC
to the aldehyde 4. Wittig reaction of 4 with the triphenylphosphonium salt of ethyl 4-bromobutyrate at
−78°C in THF and subsequent hydrogenation of the olefin with Wilkinson’s catalyst gave the ethyl ester
5. The ethyl and thioacetyl groups were removed by hydrolysis with non-enantiospecific enzyme wheat
germ lipase in phosphate buffer. Treatment of the hydrolysed product with oxidative enzyme mushroom
tyrosinase7 (EC 1.14.18.1) in the same pot gave α-lipoic acid 6 which was isolated after acidification
and extraction with n-hexane (Scheme 2). Comparison of optical rotation8 and the retention time from
chiral HPLC of the final product with the literature9 showed that the (S)-lipoic acid was obtained from
2. (R)-Lipoic acid was also obtained in a similar fashion, starting with 3, after acetylation. Hence the
absolute configuration of 2 was assigned as (R) and that of 3 as (S).10
Scheme 2. (i) BH3·DMS, 0°C; (ii) PCC; (iii) Br−PPh3+(CH2)3COOEt, NaHMDS, −78°C; (iv) (PPh3)3RhCl, H2; (v) wheatgerm
lipase, pH 7.0; (vi) tyrosinase
Lipase catalysed esterification of (S)-3-(acetylthio)-2-methyl propanoic acid with n-butanol in organic
solvent leads to the formation of the butyl ester in the first step and deacylation of the thioacetate in the
second step to give the butyl ester of (S)-2-methyl-3-thio-propanoic acid.11,12 By analogy, hydrolysis of 1
could lead to 2 (reaction stopping at ester hydrolysis stage), but formation of 3 with an intact ester group
is surprising since the reaction is being carried out in an aqueous buffer. The formation of 3 suggested
the possibility of the presence of a contaminating esterase or protease which selectively deacetylates the
thiol function but does not attack the ester function.
SDS-PAGE experiments with commercial lipase showed a major enzyme band at 55–60 kD and some
low molecular weight (15–20 kD) protein fractions. The enzyme was purified by cross-flow membrane
filtration using a membrane with a molecular weight cut-off of 30 kD. This effectively removed the
impurities present in the enzyme. This purified enzyme preparation was again used to hydrolyze the
racemic 1. It was observed that both the product distribution and optical purities were unchanged,
indicating that the formation of 3 is not due to a contaminating enzyme, otherwise a change in product
distribution ratio would have been observed. When the purified lipase of Candida rugosa was treated with
diethyl p-nitrophenyl phosphate by the method described for pancreatic lipase,13,14 the lipase preparation
so obtained (DPNP-lipase) was found to be active towards 1 in phosphate buffer at pH 7.5. Although
formation of 3 was observed in optical and chemical yields comparable to that obtained with the native
lipase, 2 was not formed. The unreacted (R)-enantiomer of 1 (ee >99%) was recovered instead of 2.

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
N. W. Fadnavis et al. / Tetrahedron: Asymmetry 9 (1998) 4109–4112
4111
Similar results were obtained when a methyl ester instead of butyl ester of 1 was used, indicating that
small changes in hydrophobicity of the substrate were not important. Interestingly, DPNP-lipase was
found to be inactive towards not only the hydrophobic substrates such as olive oil and tributyrin, but also
towards water soluble p-nitrophenyl acetate, indicating that chemical treatment of the enzyme brings
about significant changes not only in regiospecificity but also in its substrate specificity.
A change in enzyme enantioselectivity has been observed on treatment of Candida rugosa lipase with
isopropanol.15 Changes in the enzyme conformation were suggested to be responsible for this. However,
neither the regiospecificity nor the enantioselectivity was affected when a lipase preparation after similar
isopropanol treatment was used for hydrolysis of 1.
Crystal structure studies16,17 have shown the presence of a ‘lid’ which can adopt an ‘open’ or ‘closed’
conformation and oxyanion-stabilizing residues which orient correctly in the active site. The substrate
binds in a narrow hydrophobic tunnel and must adopt a tuning fork conformation. It appears to us that
in the present case, two enantiomers of 1 can bind in different ways and the molecular recognition for
the substrate is such that the carboxylic group of only the (R) substrate and the thioacetyl group of
only the (S) substrate are recognised by the enzyme. Chemical modification of the amino acid residues
which are responsible for opening or closing of the lid and stabilising the oxyanion by treatment
with diethyl p-nitrophenyl phosphate can affect substrate binding and cause significant changes in
regiospecificity/substrate selectivity of the enzyme which is reflected in the results obtained in the present
case. Further work is in progress to study this aspect in detail.
Acknowledgements
We are thankful to CSIR, New Delhi for a grant of Senior Research Fellowship to SKV and for financial
support.
References
1. Wong, C.-H.; Whitesides, G. M. Enzymes in Synthetic Organic Chemistry; Baldwin, J. E.; Magnes, P. D., Eds.; Tetrahedron
Organic Chemistry Series; Pergamon: Oxford, 1994; Vol. 12.
2. Schoffers, E.; Golebiowshi, A.; Johnson, C. R. Tetrahedron 1996, 52, 3769.
3. Bhalerao, U. T.; Dasarathi, L.; Neelakantan, P.; Fadnavis, N. W. J. Chem. Soc., Chem. Commun. 1991, 1197.
4. Yadav, J. S.; Mysorekar, S. V.; Garyali, K. J. Sci. Ind. Res. 1990, 49, 400 and references cited therein.
5. Claeson, G. Acta. Chimica. Scand. 1955, 9, 178.
6. HPLC conditions: Column Chiralcel OJTM, Daicel, Japan; 5×250 mm, λ 232 nm; 1: solvent 10% isopropanol in hexane;
20
flow rate 0.5 ml/min; retention times (S)-1: 26.9 and (R)-1: 29.1 min; 2: [α]D −117.4 (c 1, methanol); ee 98.5%
20
determined after converting into corresponding butyl ester (R)-1. 3: [α]D −35.2 (c 1, methanol); ee 99.2% determined
after acetylation to (S)-1. Optical purity of 3 can also be measured directly by chiral HPLC: solvent 12% isopropanol in
hexane; flow rate 0.6 ml/min; retention times (S)-3: 23.3 and (R)-3: 24.3 min; ee 99.2%.
7. Sridhar, M.; Vadivel, S. K.; Bhalerao, U. T. Synth. Commun. 1998, 28, 1499.
8. Merck Index, pp. 1591, 9462.
9. Fadnavis, N. W.; Koteshwar, K. Tetrahedron: Asymmetry 1997, 8, 337.
10. HPLC conditions 6: Chiralcel ODTM, Daicel, Japan; 5×250 mm, λ 330 nm; 1% isopropanol in hexane; flow rate 0.6
20
20
ml/min; retention times: (R)-6: 17.8 and (S)-6: 18.3 min; ee 98.3%. [α]D −103.0 (c 1.9, benzene), lit.12 [α]D −113 (c
1.88, benzene). Selected 1 H NMR (CDCl3 200 MHz) δ 1: 0.9 (t, 3H, J=7.5 Hz), 1.3 (m, 2H), 1.6 (m, 2H), 1.9–2.3 (m,
2H), 2.2 (s, 3H), 2.3 (s, 3H), 2.8 (t, 2H, 6.1 Hz), 4.0 (t, 2H, J=5.6 Hz), 4.1 (t, 1H, 7.0 Hz). 2: 1.85–2.2 (m, 2H), 2.21 (s,
3H), 2.26 (s, 3H), 2.9 (m, 2H), 4.3 (t, 1H, 7.1 Hz). 3: 1.0 (t, 3H, J=7.2 Hz), 1.3 (m, 2H), 1.6 (m, 2H), 1.9–2.3 (m, 2H), 2.3
(s, 3H), 2.9 (t, 2H, J=5.8 Hz), 4.0 (t, 2H, J=5.9 Hz), 4.3 (t, 1H, 7.2 Hz). 4: 1.5–2.0 (m, 6H), 2.4 (t, 2H, J=7.1 Hz), 2.5 (m,
2H), 3.15 (m, 2H), 3.6 (m, 1H).

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
4112
N. W. Fadnavis et al. / Tetrahedron: Asymmetry 9 (1998) 4109–4112
11. Bianchi, D.; Cesti, P. J. Org. Chem, 1990, 55, 5657.
12. Sproull, K. C.; Bowman, G. T.; Carta, G.; Gainer, J. L. Biotechnol. Prog. 1997, 13, 71.
13. Antonov, V. K.; Dyakov, V. L.; Mishin, A. A.; Rotanov, T. V. Biochimie 1988, 1235.
14. Chapus, C.; Semeriva, M. Biochemistry 1976, 15, 4988.
15. Colton, I. J.; Ahmed, S. N.; Kazlaukskas, R. J. J. Org. Chem. 1995, 60, 212.
16. Grochulski, P.; Schrag, J. D.; Cygler, M. Prot. Sci. 1994, 3, 82.
17. Grochulski, P.; Bouthillier, F.; Kazlaukskas, R.; Senreqi, A. N.; Schrag, J. D.; Ziomek, E.; Cygler, M. Biochemistry 1994,
33, 3494.