Altering the Substrate Specificity of Cephalosporin Acylase by Directed Evolution of the -Subunit* Received for publication, August 14, 2002 Published, JBC Papers in Press, August 26, 2002, DOI 10.1074/jbc.M208317200 Linda G. Otten, Charles F. Sio, Johanna Vrielink, Robbert H. Cool, and Wim J. Quax‡ From the Department of Pharmaceutical Biology, University Centre for Pharmacy, University of Groningen, Antonius Deusinglaan 1, 9713 AV Groningen, The Netherlands Using directed evolution, we have selected an adipyl acylase enzyme that can be used for a one-step biocon- version of adipyl-7-aminodesacetoxycephalosporanic acid (adipyl-7-ADCA) to 7-ADCA, an important com- pound for the synthesis of semisynthetic cephalospor- ins. The starting point for the directed evolution was the glutaryl acylase from Pseudomonas SY-77. The gene fragment encoding the -subunit was divided into five overlapping parts that were mutagenized separately us- ing error-prone PCR. Mutants were selected in a leucine-deficient host using adipyl-leucine as the sole leucine source. In total, 24 out of 41 plate-selected mu- tants were found to have a significantly improved ratio of adipyl-7-ADCA versus glutaryl-7-ACA hydrolysis. Sev- eral mutations around the substrate-binding site were isolated, especially in two hot spot positions: residues Phe-375 and Asn-266. Five mutants were further charac- terized by determination of their Michaelis-Menten pa- rameters. Strikingly, mutant SY-77 N266H shows a nearly 10-fold improved catalytic efficiency (k cat /K m ) on adipyl- 7-ADCA, resulting from a 50% increase in k cat and a 6-fold decrease in K m , without decreasing the catalytic efficiency on glutaryl-7-ACA. In contrast, the improved adipyl/glutaryl activity ratio of mutant SY-77 F375L mainly is a consequence of a decreased catalytic efficiency toward glutaryl-7-ACA. These results are dis- cussed in the light of a structural model of SY-77 glutaryl acylase. Semisynthetic cephalosporins and penicillins are the most widely used antibiotics. All clinically important semisynthetic cephalosporins are made from 7-aminocephalosporanic acid (7- ACA) 1 or 7-aminodesacetoxycephalosporanic acid (7-ADCA). 7-ACA is derived from cephalosporin C (aminoadipyl-7-ACA), which is obtained by fermentation of the fungus Cephalospo- rium acremonium. Deacylation is performed either chemically or by a two-step enzymatic process using a D-amino acid oxi- dase and a glutaryl acylase. The latter enzyme can be found in several Pseudomonas and Acinetobacter species (1–7) as well as in some Gram-positive bacteria (8, 9). 7-ADCA is produced from penicillin G made by Penicillium chrysogenum involving sev- eral polluting chemical steps followed by enzymatic deacylation by penicillin acylase (10). A first step toward the introduction of a simplified, more environmentally friendly production of 7-ADCA was the development of a genetically modified P. chry- sogenum that produces adipyl-7-ADCA (AD-7-ADCA) (11). For the deacylation of this novel -lactam, an adipyl acylase is needed. Since the presently identified acylases show little or no activity toward AD-7-ADCA, it is of interest to investigate whether a glutaryl acylase can be converted into an adipyl acylase. In the past few years, directed evolution has been success- fully implemented in changing the substrate specificity of sev- eral enzymes (12, 13), resulting in biocatalysts with novel ac- tivities. It has become clear that the success of a directed evolution experiment greatly depends on the availability of a good selective substrate, which unfortunately is absent for most bioconversions (12). Artificial substrates that mimic one of the desired catalytic steps may be used for selection; how- ever, it is not clear to what extent the resulting mutants will have lost activity on their natural substrate. Here we describe a strategy to evolve the glutaryl acylase of Pseudomonas SY-77 into an adipyl acylase with an improved activity toward AD-7-ADCA. The glutaryl acylase from Pseudo- monas SY-77 has proven to be particularly suitable for devel- oping an industrial process for deacylation (14). The natural action of the enzyme seems to be directed at hydrolyzing di- amino acids with a glutaryl side chain as judged from its high activity on glutaryl-7-A(D)CA. It appears that the enzyme also has a low activity on AD-7-ADCA but no activity on cephalo- sporin C (2). Since deacylation of -lactam compounds cannot be used for a growth selection, we took advantage of the diamino hydro- lyzing capability of the acylase by replacing the -lactam moi- ety with leucine, a compound that can be selected for in a leucine-deficient Escherichia coli host strain (15). In this way, only enzymatic hydrolysis of adipyl-leucine allows for growth on minimal medium. To obtain the desired acylase variant, we have constructed five libraries of overlapping gene fragments of the -subunit of SY-77 by error-prone PCR (epPCR) and used these libraries separately in the selection procedure. Transfor- mants that prevailed in growth were further characterized and tested for their activity on GL-7-ACA and AD-7-ADCA. Mu- tants with an improved growth capability on the selection substrate also showed an improved activity toward the -lac- tam substrate. Mutations were found to accumulate in the proximity of the substrate binding pocket. Frequent mutations were identified at positions Asn-266 and Phe-375. Crystallo- graphic models have pointed at the role of Phe-375 in the determination of substrate specificity, whereas Asn-266 was * This work was sponsored by contract GBI.4707 from STW, which is part of the Dutch Organization for Science. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ To whom correspondence should be addressed. Tel.: 31-50-3632558; Fax: 31-50-3633000; E-mail: w.j.quax@farm.rug.nl. 1 The abbreviations used are: 7-ACA, 7-aminocephalosporanic acid; 7-ADCA, 7-aminodesacetoxycephalosporanic acid; AD-7-ADCA, adipyl- 7-ADCA; GL-7-ACA, glutaryl-7-ACA; epPCR, error-prone PCR; WT, wild type; AD-Leu, minimal agar supplemented with adipyl-leucine; GL-Leu, minimal agar supplemented with glutaryl-leucine; MLeu, min- imal agar supplemented with L-leucine; Min, minimal agar without leucine source; AD/GL, hydrolysis rate of AD-7-ADCA/hydrolysis rate of GL-7-ACA. THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 277, No. 44, Issue of November 1, pp. 42121–42127, 2002 © 2002 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A. This paper is available on line at http://www.jbc.org 42121 by guest on January 24, 2016 http://www.jbc.org/ Downloaded from