Dynamic kinetic resolution: Synthesis of optically active α-amino acid derivatives

Article abstract of DOI:10.1016/S0957-4166(00)00108-7

Candida antarctica lipase B (Novozyme) is an effective catalyst for the dynamic kinetic resolution of racemic 2-benzyl-4-substituted-5(4H)- oxazolones, in the presence of an alcohol, yielding optically active N- benzoyl amino acid esters. The reaction has been optimised with respect to the solvent and the effect of a catalytic amount of organic base is investigated. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(00)00108-7

Tetrahedron: Asymmetry 11 (2000) 1687±1690
Dynamic kinetic resolution: synthesis of optically active
a-amino acid derivatives
Stuart A. Brown,a Marie-Claire Parkerb and Nicholas J. Turnera,*
aEdinburgh Centre for Protein Technology, Department of Chemistry, The University of Edinburgh, King's Buildings,
West Mains Road, Edinburgh EH9 3JJ, UK
bDepartment of Chemistry, Joseph Black Building, University of Glasgow, Glasgow G12 8QQ, UK
Received 22 February 2000; accepted 20 March 2000
Abstract
Candida antarctica lipase B (Novozyme) is an e€ective catalyst for the dynamic kinetic resolution of
racemic 2-benzyl-4-substituted-5(4H)-oxazolones, in the presence of an alcohol, yielding optically active
N-benzoyl amino acid esters. The reaction has been optimised with respect to the solvent and the e€ect of a
catalytic amount of organic base is investigated. # 2000 Elsevier Science Ltd. All rights reserved.
Despite the success of enzyme-catalysed kinetic resolutions for the generation of a wide range
of chiral building blocks,1 there is an increasing desire to develop transformations that are not
limited by a maximum yield of only 50% of the desired product. Increasing attention has been
given in recent years to the discovery of dynamic kinetic resolution (DKR) processes in which the
unreactive enantiomer equilibrates, in situ under the reaction conditions, with the reactive
antipode.2 DKR reactions can thus result in quantitative yields with enantiomeric excesses (e.e.s)
approaching 100%. Previously, we3 and others4 have reported the lipase-catalysed DKR of
5(4H)-oxazolones 1 as an entry into enantiomerically pure amino acids. Oxazolones are excellent
substrates for DKR reactions on account of (i) the low pKa of the C-4 proton;5 and (ii) their
inherent reactivity towards lipase-catalysed alcoholysis.
In our initial studies we described an ecient method for the synthesis of l-(S)-tert-leucine
using the oxazolone approach based upon Rhizomucor miehei lipase (Lipozyme) as the catalyst.5
* Corresponding author. E-mail: n.j.turner@ed.ac.uk
0957-4166/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(00)00108-7

1688
S. A. Brown et al. / Tetrahedron: Asymmetry 11 (2000) 1687±1690
Although this lipase gives excellent results for tert-leucine (94% yield, 99.5% e.e.), less sterically
demanding substrates result in a dramatic drop in the e.e. and additionally the reaction displays a
pronounced dependence on the chain length of the alcohol.6 Herein we report that these limita-
tions can be overcome by employing Candida antarctica lipase B (Novozyme) as the catalyst,
enabling a wider range of substrates to be transformed with high yield and e.e.
Initially Lipozyme and Novozyme were compared as catalysts using 2-phenyl-4-benzyl-5(4H)-
oxazolone 2 as the substrate in toluene with n-butanol as the nucleophile (Table 1).7 The results in
Table 1 illustrate clearly that the use of Novozyme results in a higher yield and e.e. compared to
Lipozyme. The data also con®rm our previous observation that addition of Et3N (0.25 equiv.)
results in a signi®cant increase in the e.e. of the product as deduced by comparing entries 1±2 and
3±4, respectively. The initial moderate yield obtained for the Novozyme reaction (entry 4) was
due to the competing hydrolysis reaction which could be substantially reduced by pre-drying the
crushed lipase over phosphorus pentoxide to remove excess water. An increase in the yield from
51% to 81% (entry 5) was observed with no detrimental e€ect on the e.e.
Table 1
Dynamic kinetic resolution of oxazolone 1 using Lipozyme and Novozyme
The alkyl chain length of the alcohol nucleophile was also varied and found to have little e€ect
on the e.e. of the product (with the exception of n-pentanol) in the Novozyme mediated
biotransformation, which is in direct contrast to the results found with Lipozyme (Table 2).
Table 2
E€ect of nucleophile alkyl chain length

S. A. Brown et al. / Tetrahedron: Asymmetry 11 (2000) 1687±1690
1689
In an attempt to increase the e.e. of the products in the Novozyme reaction the solvent used
was varied in the presence and absence of Et3N (Table 3). Chlorinated solvents gave poorer yields
and e.e.s. while ethers gave comparable results to those obtained in toluene. In general the e€ect
of the Et3N was to enhance the e.e. of the product, in some cases in a dramatic fashion (i.e.
tBuOMe, DIPE). However, the use of tetrahydrofuran and acetonitrile as solvent produced the
reverse e€ect. In both cases the e.e. remained high and the yield in fact increased in the absence of
Et3N. Although the initial reason for adding Et3N to these DKR reactions was to accelerate the
rate of substrate racemisation it appears that it has a more complex role. Kinetic experiments8
have shown that, in non-polar solvents such as hexane and toluene, small amounts of carboxylic
acid by-product (e.g. N-benzoyl phenylalanine), resulting from hydrolysis of the ozaxolone
substrate, cause a marked lowering of the e.e. of the product. Addition of triethylamine reverses
this e€ect, presumably due to the formation of a soluble ion-pair.9 In the absence of the base, the
carboxylic acid generated may remain bound to the lipase surface thus altering the ionogenic
state of the enzyme and leading to a change in conformation and hence enantioselectivity.
In polar solvents, such as acetonitrile and tetrahydrofuran, the acid by-product is more soluble
and therefore less likely to interact with the lipase, resulting in no reduction of the observed e.e.
In terms of overall simplicity the use of acetonitrile as solvent is to be preferred.
Table 3
Dynamic kinetic resolution of oxazolone 2 using di€erent solvents in the presence/absence of Et3N
Finally we investigated the substrate speci®city of Novozyme. A series of C-4 substituted
oxazolones were examined under conditions of either CH3CN or toluene/Et3N as solvent. The results
are shown in Table 4. It can be seen that for a range of di€erent side chains (e.g. R=CH2Ph,
CH2CH(CH3)2, CH(CH3)2) the yields and e.e.s of the products are generally good to excellent
and that the reaction conditions have little e€ect. For entries 4 and 5, acetonitrile was found to be
inferior to toluene/Et3N. Perhaps expectedly, when the R group is either sterically demanding,
e.g. tert-butyl or the small methyl group, the enantioselectivity is observed to be poor.
In summary we have demonstrated that Novozyme is a versatile catalyst for the synthesis of
optically active a-amino acid derivatives based upon DKR of 5(4H)-oxazolone substrates.

1690
S. A. Brown et al. / Tetrahedron: Asymmetry 11 (2000) 1687±1690
Table 4
Dynamic kinetic resolution of 4-substituted oxazolones
Acknowledgements
We are grateful to the BBSRC for a studentship (S.A.B.) and a David Phillips Fellowship
(M.C.P.). The samples of Novozyme and Lipozyme were generously supplied by Boehringer
Mannheim, Germany. E. Alan Fisher is acknowledged for some preliminary experiments.
References
1. Faber, K. Lipases. In Biotransformations in Organic Chemistry; Springer-Verlag, 1995; pp. 80±105.
2. For recent reviews on DKR, see: Noyori, R.; Tokungaga, M.; Kitamura, M. Bull. Chem. Soc. Jpn. 1995, 68, 36±56;
Ward, R. S. Tetrahedron: Asymmetry 1995, 6, 1475±1490; Caddick, S.; Jenkins, K. Chem. Soc. Rev. 1996, 447±456.
3. Turner, N. J.; Winterman, J. R.; McCague, R.; Parratt, J. S.; Taylor, S. J. C. Tetrahedron Lett. 1995, 36, 1113±1116.
4. Bevinakatti, H. S.; Newadkar, R. V.; Banerji, A. A. J. Chem. Soc., Chem. Commun. 1990, 1091; Bevinakatti, H. S.;
Banerji, A. A.; Newadkar, R. V.; Mokashi, A. Tetrahedron: Asymmetry 1992, 3, 1505; Gu, R.-L.; Lee, I. S.; Sih,
C. J. Tetrahedron Lett., 1992, 33, 1953; Crich, J.; Brieva, R.; Marquart, P.; Gu, R.-L.; Flemming, S.; Sih, C. J. J.
Org. Chem. 1993, 58, 3252.
5. de Jersey, J.; Zerner, B. Biochemistry 1969, 8, 1967.
6. Winterman, J. R. PhD thesis; Chemo-Enzymic Methods for the Synthesis of Optically Active a-Amino Acids;
University of Exeter, 1996.
7. To a solution of oxazolone (100 mg) dissolved in organic solvent (8 mL) were added alcohol (2.0 equiv.), Et3N
(0.25 equiv., optional) and lipase (100 mg). The ¯ask was stoppered and placed in an orbital incubator at 37ꢀC and
the reaction monitored by TLC. Upon consumption of the starting material the lipase was ®ltered o€ and washed
with solvent (2Â10 mL). The combined organic fractions were evaporated under reduced pressure and the crude
product puri®ed by column chromatography to give the desired product as a colourless solid. In all cases the (S)-
isomer of the product predominated.
8. Brown, S. A.; Parker, M.-C.; Robertson, L.; Turner, N. J. J. Chem. Soc., Chem. Commun. 1998, 2247.
9. Maugard, T.; Remaud-Simeon, M.; Petre, D.; Monsan, P. Tetrahedron 1997, 53, 7587±7594.