Tetrahedron Letters, Vol. 36, No. 17, pp. 2963-2966, 1995
Elsevier Science Ltd Printed in Great Britain 0040-4039/95 $9.50+0.00
Enhancement of the Enantioselectivity of Penicillin G Acylase from E. coil by "Substrate Tuning" Torsten Pohl and Herbert Waldmann* Universit~t Karlsruhe, Institut f'dr Organische Chemie, Richard-WiUs~tter-Allee 2, D-76128 Karlsmhe
The efficiency of penicillin G acylase catalyzed transformations is enhanced significantly with respect to reaction velocity, and in particular, stereoselectivity by appropriate "substrate tuning", i.e. by the introduction of a nitrogen atom into the part of the substrates which is recognized by the enzyme.
The application of biocatalysts in organic synthesis has offered numerous advantageous methods for the synthesis of enantiomerically pure compounds. 1 A particularly valuable enzyme which has been used for this purpose on a laboratory as well as an industrial scale is penicillin G acylase from E. coli. 2 This commercially available and fairly stable biocatalyst selectively hydrolyzes esters and amides of phenylacetic acid. 3 Whereas it tolerates only minor changes in the structure of the acyl group of its substrates, 3,4 their amine and alcohol part may vary within a wide range. Thus, penicillin acylase has successfully been used for the kinetic resolution of alcohols 5 and amines, 6 for the removal of protecting groups from amino acids, 7 nucleosides 8 and carbohydrates,7b and in industry its ability to cleave phenylacetic acid amides is exploited for the production of semisynthetic penicillins on a large scale. 2 However, in the enantioselective saponification of racemic phenylacetic acid esters and amides, this biocatalyst unfortunately often displays only a moderate stereodiscrimination. In addition, due to the unpolar character of phenylacetic acid, the substrates of penicillin acylase often are only sparingly soluble in the aqueous reaction media employed, resulting in low conversion rates.7, 8 This disadvantage may be overcome by the addition of solubility-enhancing organic cosolvents, in particular DMF, DMSO and N-methylpyrrolidinone (NMPD). In their presence, however, the enzyme is denatured to a major extent so that fairly large amounts of biocatalyst have to be used to effect a preparatively useful transformation. In the light of the wide applicability of penicillin G acylase (vide supra) and its industrial relevance, the development of strategies which serve to overcome the abovementioned problems caused by the properties of the enzyme itself and of its substrates is of great interest. For this purpose the modification of the protein by genetic engineering ("protein engineering "9) is interesting and promising. But, it requires the application of techniques which usually are not available in an organic chemistry laboratory. Also, a variation of the reaction medium offers opportunities ("medium engineering"t°). However, the fundamental relations between protein structure and solvent properties currently are only poorly understood so that this field still is in its infancy. An advantageous alternative to these strategies is provided by the variation of the structure of the substrates itself in terms of the introduction of steric demand and/or polarity. 11 Thus, this approach calls for a fine tuning of the substrate structure by means of chemical synthesis. Therefore we would like to call it "substrate tuning". It does not require additional technologies and conceptually it can draw from data provided by mapping the substrate tolerance of a given enzyme. 2963
m ,l,inOacy,. OR
phosphate buffer, pH=8
Table 1. Enzyme mediated hydrolysis of pyridyl acetates 1-8 by penicillin G acylase substlate
N~ ~ ~ O
a 100 = hydrolysisof penicillinG at pH=8 In this paper we would like to report that by application of "substrate tuning" the enantioselectivity displayed by penicillin G acylase can be enhanced to a considerable extent. Various experiments on the substrate tolerance of penicillin acylase revealed that the enzyme recognizes and is fairly specific for aryl acetic acid derivatives.3,4 It only tolerates the introduction of small substituents into this group. On the basis of this information we investigated pyridyl acetic acid esters as possible substrates for the acylase. The incorporation of a nitrogen atom into the group which is recognized by the enzyme should give rise to alternative polar interactions within the active site which are not operative for phenyl acetates, and which might lead to a different binding. As a result thereof the enantioselectivity might be enhanced. The possibility to place the nitrogen in ortho-, recta- orparaposition to the alkyl substituent and to introduce a charge by N-alkylation or N-protonation should give a further degree of freedom for a fine tuning of the substrate structure. In addition, the enhanced solubility of pyridine derivatives in aqueous solutions should have a beneficial effect on the velocity of the enzymatic transformation. In a first series of experiments the pyridyl acetates 1-8 which are readily obtained from the parent acids by acidcatalyzed estedfication, were investigated as possible substrates for penicillin G acylase. The esters 1-8 are fully soluble in a mixture of 0.07 M phosphate buffer (pH 8) and methanol (9:1 v/v) and are rapidly hydrolyzed by the enzyme (Scheme 1, Table 1). The relative velocities of the ensuing enzymatic transformations reveal that the enzyme activity is strongly depending on the position of the nitrogen atom in the aromatic ring. In general, 4-
pyridyl acetates are saponified faster than 3-pyridyl acetates which are better substrates than 2-pyridyl acetates (Table 1, compare for instance entries 1 and 2 with entries 5 and 6 and with entry 8). The methyl ester I actually nearly reaches the natural substrate penicillin G, the velocity of the enzymatic reaction declines, however, with increasing steric demand of the alcohol part of the substrates (Table 1, compare entries 1 to 4). These results already demonstrate that the principle of "substrate tuning" can be applied successfully to penicillin G acylasemediated transformations in terms of enhancing the physical properties of the substrates (i.e. solubility) and their acceptance by the enzyme. After these encouraging results in a second series of experiments the influence of the introduction of a nitrogen atom on the enantioselectivity of the enzyme was investigated. To this end, the racemic pyridyl acetates 9-14, which are synthesized in a straightforward way by activation of the respective carboxylic acids with a carbodiimide, 12 were subjected to the enzymatic hydrolysis under the conditions mentioned above. The results of these experiments confu'm the already observed trend that 4-pyridyl acetates are attacked faster than 3-pyridyl acetates which in turn are better substrates than 2-pyridyl acetates (Table 2, compare entries 1, 5 and 6).
Table2. Kinetic resolution of pyridyl acetates by hydrolysis with penicillin G acylase entry
conversion ee [%] [%]b
N~ ~ 0
N~ ~ 0 0 ~
~ ~ O
a 100 = hydrolysisof penicillinG at pH=8 b enantiomericexcessand absoluteconfigurationof the liberatedalcohol
However, of much higher interest is the observation that the introduction of the nitrogen atom into the aryl acetic acid moiety results in a significant increase of the enantioselectivity displayed by penicillin G acylase. Thus, from the ester 9 (R)-l-phenylethanol was obtained with 73% ee at 40% conversion, whereas the corresponding phenyl acetate delivered this alcohol only with 28% ee under identical conditions. Also, from the pyridyl acetate 10 (R)-l-phenylpropanol was liberated with 98% ee at 40% conversion, whereas the enzymatic saponification
of the respective phenyl acetate resulted in an ee value of 90% at 35% conversion (at only 12% conversion 94% ee is achieved5e). Similarly, from the esters 11 and 12 the (S)- and (R)-configured enantiomers of the chiral alcohols were obtained with 64% and 54% ee, respectively, whereas the corresponding phenyl acetates delivered these stereoisomers with 56% and 36% ee under similar conditions. 5c Furthermore, it is noteworthy that the enantioselectivity is influenced by the position of the nitrogen in the aryl acetic acid moiety, too. Thus, similar to the situation recorded for the reaction velocity, 4-pyridyl acetates are attacked with higher stereoselectivity than 3-pyridyl acetates which again are superior to 2-pyridyl acetates (Table 2, compare entries 1, 5 and 6). Still, however the 3-pyridyl derivative gives a better result than the analogous phenyl acetate (65% ee vs. 28% ee, vide supra), and only the 2-pyridyl acetic acid ester is comparable to the phenyl acetic acid derivative. In addition to these experiments it was investigated if the introduction of a positive charge into the aryl acetic acid residue would influence the stereoselectivity of the enzymatic transformations. However, neither a protonation of the nitrogen by conducting the hydrolyses at acidic pH, nor the use of N-methyl-pyridinium derivatives resulted in a better stereodiscrimination. In conclusion, we have demonstrated that the efficiency of penicillin G acylase catalyzed reactions can be enhanced significantly in terms of reaction velocity and, in particular, stereoselectivity by appropriate "substrate tuning". The application of this principle to different problems, i.e. the enzymatic manipulation of protecting groups,l, 13 should allow the development of new techniques and give rise to solutions for prevailing problems in biotransformations. Acknowledgement: This research was supported by the Fonds der Chemischen Industrie and the Deutsche Forschungsgemeinschaft. T. Pohl is grateful to the Hermann-Schlosser-Stiftung for a fellowship. References and Notes 1. Enzymes in Organic Synthesis - A Comprehensive Handbook (K. Drauz, H. Waldmann, eds.), VCH, Weinheim 1995. 2. Review: J. G. Shewale, B. S. Deshpande, V. K. Sudhakaran, S. S. Ambedkar, Process Biochem. Int. 1990, 97. 3. a) M. Cole, Biochem. J. 1969, 115, 741; b) M. Cole, Biochem. J. 1969, 115, 747. 4. a) A. Guy, A. Dumant, P. Sziraky, Bioorg. Med. Chem. Lett. 1993, 3, 1041; b) D. Rossi, A. Calcagni, A. Romeo, J. Org. Chem. 1979, 44, 2222; c) C. Fuganti, C. M. Rosell, S. Servi, A. Tagliani, M. Terreni, Tetrahedron Asymmetry 1992, 3, 383; d) I. B. Stoineva, B. P. Galunsky, V. S. Lozanov, I. P. Ivanov, D. D. Petkov, Tetrahedron 1992, 48, 1115. 5. a) C. Fuganti, P. Grasselli, P. F. Seneci, S. Servi, P. Casati, Tetrahedron Lett. 1986, 27, 2061; b) C. Fuganti, P. Grasselli, S. Servi, A. Lazzarini, P. Casati, Tetrahedron 1988, 44, 2575; c) E. Baldaro, P. D'Arrigo, G. Pedrocchi-Fantoni, C. M. Rosell, S. Servi, A. Tagliani, M. Terreni, Tetrahedron Asymmetry 1993, 4, 1031; d) H. Waldmann, Tetrahedron Lett. 1989, 30, 3057; e) E. Baldaro, D. Faiardi, C. Fuganti, P. Grasselli, A. Lazzarini, Tetrahedron Lett. 1988, 36, 4623. 6. a) A. Romeo, G. Lucente, D. Rossi, G. Zanotti, Tetrahedron Lett. 1971, 21, 1799; b) D. Rossi, A. Romeo, G. Lucente, J. Org. Chem. 1978, 43, 2576; c) M. J. Zmijewski Jr., B. S. Briggs, A. R. Thompson, I. G. Wright, Tetrahedron Lett. 1991, 13, 1621; d) V. A. Solodenko, T. N. Kasheva, V. P. Kukhar, E. V. Kozlova, D. A. Mironenko, V. K. Svedas, Tetrahedron 1991, 24, 3989; e) V. A. Soloshonok, V. K. Svedas, V. P. Kukhar, A. G. Kirilenko, A. V. Rybakova, V. A. Solodenko, N. A. Fokina, O. V. Kogut, I. Y. Galaev, E. V. Kozlova, I. P. Shishkina, S. V. Galushko, Synlett 1993, 339; f) V. A. Solodenko, M. Y. Belik, S.V. Galushko, V. P. Kukhar, E. V. Kozlova, D. A. Mironenko, V. K. Svedas, Tetrahedron Asymmetry 1993, 9, 1965. 7. a) C. Fuganti, P. Grasselli, P. Casati, Tetrahedron Lett. 1986, 27, 3191; b) H. Waldmann, Liebigs Ann. Chem. 1988, 1175; c) H. Waldmann, Tetrahedron Lett. 1988, 29, 1131; d) R. Didziapetris, B. Drabnig, V. ScheUenberger, H.-D. Jakubke, V. Svedas, FEBS Lett. 1991, 287, 31. 8. H. Waldmann, A. Heuser, A. Reidel, Synlett 1994, 65. 9. a) Chemical Reagents for Protein Modification, R. L. Lundblad, 2nd Edn., CRC Press, London 1991; b) The Proteins, A. N. Glazer, H. Neurath, R. L. Hill (eds), Vol 2, Academic Press, London 1976. 10. a) M. Gupta, Eur. J. Biochem. 1992, 203, 25; b) P. Aldercreutz, B. Mattiasson, Biocatalysis 1987, 1, 99. 11. Biotransformations in Organic Chemistry, K. Faber, Springer-Verlag, Berlin Heidelberg 1992. 12. B. Neises, W. Steglich, Angew. Chem. 1978, 90, 556, Int. Ed. 1978, 90, 522. 13. Review: H. Waldmann, D. Sebastian, Chem. Rev. 1994, 94, 911.
(Received in Germany 10 February 1995; accepted 2 March 1995)