Comparative homology modeling-inspired protein engineering for ...

Biotechnol Lett (2009) 31:1617–1624 DOI 10.1007/s10529-009-0055-9

ORIGINAL RESEARCH PAPER

Comparative homology modeling-inspired protein engineering for improvement of catalytic activity of Mugil cephalus epoxide hydrolase Sung Hee Choi Æ Hee Sook Kim Æ Eun Yeol Lee

Received: 29 April 2009 / Revised: 2 June 2009 / Accepted: 2 June 2009 / Published online: 23 June 2009 Ó Springer Science+Business Media B.V. 2009

Abstract The epoxide hydrolase (EH) of a marine fish, Mugil cephalus, was engineered to improve the catalytic activity based on comparative homology modeling. The 3-D crystal structure of the EH from Aspergillus niger was used as a template. A triple point mutant, F193Y for spatial orientation of the nucleophile (D199), W200L for removing electron density overlap between W200 and Y348, and E378D for good charge relay in the active site, was developed. The initial hydrolysis rate, the reaction time to reach 98 %ee, and yield were enhanced up to 35-fold, 26-fold and 32%, respectively, by homology modeling-inspired site-directed mutagenesis of M. cephalus EH. Keywords Epoxide hydrolase Homology modeling Mugil cephalus Site-directed mutagenesis

S. H. Choi H. S. Kim (&) Department of Food Science and Biotechnology, Kyungsung University, Busan 608-736, Korea e-mail: [email protected] E. Y. Lee (&) Department of Chemical Engineering, Kyung Hee University and Green Energy Center, Gyeonggi-do 446-701, Korea e-mail: [email protected]

Introduction Epoxide hydrolase (EH) catalyzes the enantioselective hydrolysis of racemic epoxides for preparing enantiopure epoxides and vicinal diols (Lee and Shuler 2007; Lee 2008; Strauss et al. 1999). Enantiopure epoxides and diols are useful starting materials for preparing more complex enantiopure pharmaceuticals (Archelas and Furstoss 2001). Various biocatalytic approaches for the production of enantiopure epoxides and diols have been investigated based on epoxide hydrolase-catalyzed kinetic resolution and enantioconvergent process (Hwang et al. 2008a, b; Karboune et al. 2006; Lee 2007). Enhancements of catalytic activity and/or enantioselectivity of enzymes by protein engineering and directed evolution are prerequisite for a commercialization of enzyme-catalyzed bioprocesses. Recently, the EH from Aspergillus niger has been engineered to improve its enantioselectivity (Cedrone et al. 2003; Reetz et al. 2004). The EH gene was mutated by error-prone polymerase chain reaction (epPCR), and the mutants were screened by using a high-throughput assay based on electron spray ionization-mass spectrometer analysis (Reetz 2006; Schrader et al. 2002). The enantioselectivity was enhanced up to E-value of 10.8. The enantioselectivity of Agrobacterium radiobacter AD1 EH increased by epPCR and DNA shuffling (Rink et al. 1999; van Loo et al. 2004). The E-value of the mutant with F108I, P205H,

123

1618

Y215H and E271V was enhanced by 10-fold. The EH of Ag. radiobacter AD1 was also engineered by DNA shuffling and saturation mutagenesis. The mutant with L190F showed a 4.8-fold enhanced catalytic activity (Cao et al. 2006; Rui et al. 2004, 2005). All these results clearly showed the potential of protein engineering and directed evolutions to improve catalytic properties of microbial EHs. EHs can be classified into the a/b-hydrolase-fold family (Barth et al 2004). The proteins in the same family share the same basic fold in spite of low level of sequence identity. Comparative homology modeling takes advantages of the structural similarities within the same family. When 3-D crystal structures of the relevant proteins are available, homology modeling has predictive potential for gene products by comparison of the template structures. The structures of proteins of interest can be modeled based on template protein whose structure is determined by X-ray crystallography. Homology modeling can offer a possibility for the identification of target amino acid residues for protein engineering. EHs are ubiquitous and found in animals, plants, yeasts, filamentous fungi, and soil bacteria. Recently, a marine microbial EH has been cloned and characterized (Hwang et al. 2008a, b). We have cloned and characterized a marine fish EH from Mugil cephalus (GenBank nucleotide sequence number: FJ911552) (Lee et al. 2007). The M. cephalus EH has a broad range of substrate spectrum. It possesses enantioselective hydrolysis activities towards a variety of aromatic and aliphatic epoxides and exhibits low product inhibition, which is an advantage as the biocatalyst for the production of enantiopure epoxides and diols. On the contrary, the catalytic activity of M. cephalus EH was rather low compared to those of microorganisms (Kim et al. 2008). The objective of this paper is to enhance the catalytic activity of M. cephalus EH by site-directed mutagenesis approach based on homology modeling. The EH gene was manipulated to change its 3-D structure, as deduced from analysis of the active site and loop structures by homology modeling with A. niger EH as the template. The mutated genes were expressed in Escherichia coli, and the catalytic activities were compared to that of wild-type M. cephalus EH.

123

Biotechnol Lett (2009) 31:1617–1624

Materials and methods Protein sequence homology analysis and homology modeling of M. cephalus EH The conserved features of Mugil cephalus EH were compared with other EHs by multiple sequence alignment by using BioEdit program. A homology model for M. cephalus EH was constructed based on X-ray crystallographic structure of A. niger EH (pdb code: 1qo7) and Ag. radiobacter EH (pdb code: 1ehy). Homology modeling was carried out using SwissModel, protein modeling server, accessible via the EXPASY (http://www.expasy.org/). Superimposition of M. cephalus EH model on that of A. niger EH was constructed by using RasMol (http://www.umass.edu/ microbio/rasmol/) and Deep-View program (http:// spdbv.vital-it.ch/). Strains, plasmids, site-directed mutagenesis and PCR The site-specific mutagenesis of M. cephalus EH gene was generated by using Quikchange sitedirected mutagenesis kit (Stratagene Co, USA) with various primers shown in Table 1. The recombinant pET-21b(?)/mMcEH plasmid was used as the template for site-directed mutagenesis (Lee et al. 2007). The mutagenic PCR operation conditions were as follows: 30 s at 95°C, followed by 12 cycles of 30 s at 95°C, 1 min at 55°C, 7 min 20 s at 68°C, ending with 2 min on ice to cool the reaction to 37°C. DpnI restriction enzyme was added directly to the reaction tube and the reaction was immediately incubated at 37°C for 1 h to degrade the methylated template. The purified PCR product was transformed into E. coli BL21 (DE3) and cultured on the LB agar plate containing 50 lg ampicillin/ml. To express the recombinant proteins, the recombinant E. coli was cultivated at 37°C for 2 h to be an OD600 of 0.4–0.6 with shaking at 190 rpm. The cells were incubated at 15°C for 30 min and cultivated at 15°C for 24 h to express the mutated EH genes by addition of 1 mM IPTG. To analyze the expressed gene products, the suspension cell with or without IPTG induction was held at 95°C for 10 min with Laemmli buffer, separated on 12% (v/v) SDS-polyacrylamide gel,


1619

Table 1 Oligonucleotides used in site-directed mutation Primer

Target

Gene sequence (50 ? 30 )

5MCF-F193Y

F193Y

GGGGTTCTCAAAGTACTATCTGCAGGGAGG

3MCR-F193Y

F193Y

CCTCCCTGCAGATAGTACTTTGAGAACCCC

5MCF-E378D 3MCR-E378D

E378D E378D

GCTGCCTTCCCTGGAGACCTGATGCATTGCCCTAAATC GATTTAGGGCAATGCATCAGGTCTCCAGGGAAGGCAGC

5MCF-W200L

W200L

GCAGGGAGGAGACTTGGGCTCGCCCATCAC

3MCR-W200L

W200L

GTGATGGGCGAGCCCAAGTCTCCTCCCTGC

Underlined letters indicate the changed nucleotide for site-directed mutagenesis

and blotted onto a nitrocellulose membrane. The membrane was incubated with polyclonal antibody against hexahistidine (H-15, Santa Cruz Biotechnology Inc., USA) and peroxidase-conjugated anti-rabbit IgG (Jackson Immunoresearch, USA), and then visualized with CN/DAB(4-chloronaphthol/3,30 -diaminobenzidine) Substrate Kit (Pierce, USA). Kinetic resolution of racemic styrene oxide by the recombinant E. coli possessing the mutated M. cephalus EH gene Kinetic resolution of racemic styrene oxides was conducted in 1 ml 100 mM KH2PO4 buffer in 10 ml screw-cap bottles sealed with a rubber septum. The recombinant E. coli cells expressing the mutated M. cephalus EH genes were used as the biocatalyst. Kinetic resolution started with the addition of 20 mM racemic styrene oxide at 30°C and shaking at 230 rpm. The reaction was stopped by extraction with equal volume of cyclohexane. The progression of hydrolysis reaction was followed by the analysis of samples withdrawn periodically from the reaction mixture. Analysis Cell concentration was calculated from the OD600. Enantiomeric excess (ee = 100 9 (S - R)/(S ? R)) and yield for enantiopure styrene oxide were determined by GC analysis. The reaction mixture was extracted with equal volume of cyclohexane, and 1 ll of the organic layer was analyzed by using GC with a fused silica capillary beta-DEX 120 column (0.25 mm ID 9 30 m, 0.25 lm film thickness, Supelco, USA) and a FID detector. The temperatures

of the column, injector, and detector were 100, 220, and 220°C, respectively.

Results and discussion Homology modeling of M. cephalus EH The M. cephalus EH has a broad range of substrate spectrum and reaction temperature compared to those of microbial EHs. On the contrary, the hydrolytic activity was rather low (Lee et al. 2007; Kim et al. 2008). Hence, we carried out homology modelinginspired protein engineering to enhance the activity of M. cephalus EH. Because the sequence similarity of M. cephalus EH with A. niger EH using PredictProtein server (http://www.predictprotein.org/) is about 30%, the active site of M. cephalus EH was successfully modeled on the basis of A. niger EH (Kim et al. 2006; Nardini et al. 1999; Zou et al. 2000). The modeled 3-D structure of the active site of M. cephalus EH is shown in Fig. 1. The catalytic triad consisting of D199, E378 and H405 is highly conserved. Two tyrosines (Y273 and Y348) that assist the activation of the epoxide ring of substrates by forming hydrogen bond are also highly conserved in the active site of M. cephalus EH. Correction of spatial orientation of nucleophile for effective catalysis Even though similar topologies of the active sites have been obtained for M. cephalus EH and A. niger EH when active sites are superimposed, the 3-D structure of active site of M. cephalus EH is somewhat different from that of A. niger EH. While

123

1620

Fig. 1 The active sites of A. niger EH (AnEH, 1qo7.pdb, cyan) and wild-type M. cephalus EH (McEH, modeled, yellow). The cyan letters indicate the amino acids of A. niger EH and black letter for the amino acids of M. cephalus EH. The red circle indicates that the spatial orientation of the nucleophilic aspartate of wild-type M. cephalus EH is different from that of A. niger EH

the positions of two tyrosine residues (Y273 and Y348) and histidine residue (H405) are the same, the spatial orientation of nucleophile (D199) of M. cephalus EH is different from that of A. niger EH. We decided to modify a certain amino acid near D199 to hold the aspartate in a proper orientation to epoxide substrate for more effective catalysis. In the first step, the loop structure near D199 was analyzed (Fig. 2). The helix 5, one of the basic scaffolds of a/b-hydrolase fold enzymes, of M. cephalus EH and A. niger EH are similar in their structure. However, as shown in Fig. 2b, there is difference in loop structures between M. cephalus EH and A. niger EH, and this difference might cause difference in aspartate orientation. Hence, we replaced phenylalanine (F193) and tyrosine (Y194) residues with various amino acids, and then modeled the mutants in silico by using homology modeling. When tyrosine instead of F193 or methionine instead of Y194 was replaced, the loop structure of M. cephalus EH changed to fit with that of A. niger EH, and then the aspartate of the mutants has a good fitting with that of A. niger EH. Therefore, F193 or Y194 can be a reasonable candidate as the target amino acid for site-directed mutagenesis. The single point mutants (F193Y and Y194M) were developed by site-directed mutagenesis based

123


on comparison of tertiary structure of active sites. To check whether the mutants have a better activity, we carried out kinetic resolution of racemic styrene oxide by using wild-type M. cephalus EH and the single point mutants. We compared the recombinant E. coli cells expressing the wild-type M. cephalus EH and mutated EHs because the wild-type M. cephalus EH is microsomal protein and microsomal EH loose its activity during protein purification procedure (Kronenburg and de Bont 2001; Lee et al. 2007). The expression level of wild-type and mutated EH genes were the almost same (Fig. 3). As shown in Fig. 4a, the reaction time to reach 98 %ee decreased from 260 min to 180 min for the F193Y mutant, which clearly indicates that the catalytic activity was enhanced by repositioning of aspartate. In the case of Y194M, however, the catalytic activity of the mutant lost its hydrolytic activity. Replacement of glutamate by aspartate in charge-relay system In general, bacterial EH such as A. radiobacter EH has aspartate in its charge-relay system. On the contrary, M. cephalus EH has glutamic acid (E378). Based on this feature, we can expect that the hydrolysis rate can be increased by changing the E378 to aspartate because the pKa value of aspartic acid is smaller than that of glutamic acid (Arand et al. 1999). A double point mutant with F193Y and E378D was developed to investigate whether there is a cumulative effect of two beneficial mutations. When kinetic resolutions of racemic styrene oxide were conducted, the reaction time required for the completion of kinetic resolution decreased from 260 min down to 55 min. Therefore, double point mutations have a cumulative positive effect on the catalytic activity of M. cephalus EH (Fig. 4b). Development of triple point mutant We analyzed the distance between catalytic triad and electron density map in the active site of the double mutant. The catalytic aspartate residue is properly positioned as expected after changing F193 to tyrosine. The carboxyl functional group of aspartate was corrected to orient toward the epoxide substrate like in that of A. niger EH. The distance between


1621

Fig. 2 Multiple sequence alignments of helix 5 and loop region of M. cephalus EH (McEH), A. niger EH (AnEH), R. glutinis EH (RgEH) and Ag. radiobacter EH (ArEH) (a), and overlap of backbone structures from 172 to 194 of A. niger EH, and from 179 to 201 of wild-type and mutated M. cephalus EHs based on homology modeling (b). The backbone model with blue color represents the wild-type M. cephalus EH and yellow one represents A. niger EH. The mutated amino acid residues and nucleophilic Asp residue are indicated by arrow. The red (F193Y) and green (Y194M) lines represent the mutant’s loops

hydroxyl groups of tyrosine (Y273) in the right-hand side and carboxylate group of the nucleophilic ˚ , very similar to that in aspartate is about 5.69 A A. niger EH (Fig. 5). The mutated aspartate in charge relay system is separated to amine group of histidine ˚ , which is also similar to A. niger EH. by 2.99–2.94 A All these simulated results can indicate that the features of the catalytic triad of the double point mutant of M. cephalus EH were changed to be fit with those of A. niger EH. There is, however, one striking difference between the active sites of the double point mutant of M. cephalus EH and A. niger EH. As shown in Fig. 5, there is some overlap of electron density map between W200 and Y348 of the double point mutant of M. cephalus EH. When tryptophan at 200 is changed to leucine, the overlap of electron density can be removed since leucine possessing iso-butyl group as a side chain is less bulky than tryptophan side chain. A triple point mutant, F193Y for spatial orientation of D199, W200L for removing electron density overlap between W200 and Y348, and E378D for good charge relay in the active site, was developed by using site-directed mutagenesis.

We compared the triple point mutant to the wildtype M. cephalus EH. To check whether the triple point mutant has a better activity, we carried out kinetic resolutions of racemic styrene oxide by using wild-type M. cephalus EH and the triple point mutant. All the experimental conditions were exactly same except biocatalysts. As shown in Fig. 4c, the reaction time was further decreased down to 10 min from 260 min. It appears that three amino acids positions (F193, W200 and E378) were good targets for improving the catalytic properties of M. cephalus EH. By introducing triple point mutations, the initial hydrolysis rate was enhanced from 0.01 for wild-type M. cephalus EH to 0.35 lmol/minmg of cell for the triple point mutant. The yield was increased up to 32%. The reaction time required for the completion of kinetic resolution was decreased up to 26-fold by homology modeling-inspired triple point mutations. Compared to microbial EH, the hydrolytic activity of triple mutants was all most similar to that of the recombinant E. coli cells expressing the microbial EH gene of Rhodotorula glutinis (Kim et al. 2006). When the enantioselective hydrolysis reaction was carried out in the presence of phenyl-1,2-ethanediol up to

123

1622

Biotechnol Lett (2009) 31:1617–1624 10

(a) Concentration (mM)

8

6

4

2

0 0

50

100

150

200

250

Time (min) 10

(b) Concentration (mM)

8

6

4

2

0 0

200 mM, no or little inhibition was observed for the triple mutant, indicating that the mutants retained the original advantageous properties such as low sensitivity to diol product (data not shown). Even though there are still gaps in our knowledge about the relationship between modeled 3-D structure and function, we could enhance the catalytic activity of a marine fish EH by site-directed mutagenesis based on homology modeling in an efficient way.

Conclusion The catalytic activity of a marine fish EH was enhanced by homology modeling-inspired protein engineering. A triple point mutant of M. cephalus EH, F193Y for spatial orientation of nucleophile,

123

100

150

200

250

Time (min) 10

(c) 8

Concentration (mM)

Fig. 3 SDS-PAGE (a) and immunoblotting (b) analysis of the expressed M. cephalus EH and mutant EHs in recombinant E. coli. M Marker, lane 1, 2 pET-21b(?) vector, lane 3, 4 wild type M. cephalus EH, lane 5, 6 F193Y, lane 7, 8 F193YE378D, lane 9, 10 F193Y-W200L-E378D. Lane 1, 3, 5, 7, 9 without induction (X), lane 2, 4, 6, 8, 10 with induction (O)

50

6

4

2

0 0

50

100

150

200

250

Time (min)

W200L for removing electron density overlap between W200 and Y348, and E378D for good charge relay in the active site, was developed by sitedirected mutagenesis based on homology modeling. The initial reaction rate reaction was enhanced 35-fold and the reaction time required for the production of 98 %ee (S)-styrene oxide was 26-fold decreased by

Biotechnol Lett (2009) 31:1617–1624 b Fig. 4 Batch kinetic resolutions of racemic styrene oxides by

using the recombinant E. coli expressing wild type M. cephalus EH gene and various point mutated genes. a Single point mutation (F193Y) (inverted filled triangle: (S)-styrene oxide, filled circle: (R)-styrene oxide for the wild-type EH, inverted open triangle: (S)-styrene oxide; open circle: (R)-styrene oxide for the Phe193Tyr mutant). b Double point mutation (F193YE378D) (inverted filled triangle: (S)-styrene oxide, filled circle: (R)-styrene oxide for the wild-type EH; inverted open triangle: (S)-styrene oxide, open circle: (R)-styrene oxide for the F193Y-E378D mutant). c Triple point mutation (F193YW200L-E378D) (inverted filled triangle: (S)-styrene oxide, filled circle: (R)-styrene oxide for the wild-type EH; inverted open triangle: (S)-styrene oxide, open circle: (R)-styrene oxide for the F193Y-W200L-E378D mutant)

1623

homology modeling-inspired site-directed mutagenesis of M. cephalus EH. As homology modeling and protein structure prediction method are improving in near future, the target amino sites for enhancing catalytic activity of enzymes will be readily available for protein engineering to enhance the desired catalytic properties. Acknowledgments This work was supported by the Marine and Extreme Genome Research Center Program, Ministry of Land, Transportation and Maritime Affairs, Republic of Korea. The stipend for S. H. Choi was partially supported by the Ministry of Knowledge Economy (MKE) and Korea Industrial Technology Foundation (KOTEF) through the Human Resource Training Project for Strategic Technology.

References

Fig. 5 Electron density maps of A. niger EH (a) and the F193Y-E378D double point mutant of M. cephalus EH (b)

Arand M, Muller F, Mecky A, Hinz W, Urban P, Pompon D, Kellner R, Oesch F (1999) Catalytic triad of microsomal epoxide hydrolase: replacement of Glu404 with Asp leads to a strongly increased turnover rate. Biochem J 337: 37–43 Archelas A, Furstoss R (2001) Synthetic applications of epoxide hydrolases. Curr Opin Chem Biol 5:112–119 Barth S, Fischer M, Schmid RD, Pleiss J (2004) The database of epoxide hydrolase and haloalkane dehalogenases: one structure, many functions. Bioinformatics 20:2845–2847 Cao L, Lee J, Chen JW, Wood TK (2006) Enantioconvergent production of (R)-1-phenyl-1,2-ethanediol from styrene oxide by combining the Solanum tuberosum and an evolved Agrobacterium radiobacter AD1 epoxide hydrolases. Biotechnol Bioeng 94:522–529 Cedrone F, Niel S, Ait-abdelkader N, Torre C, Krumm H, Roca S, Bhatnagar T, Maichele A, Reetz MT, Baratti JC (2003) Directed evolution of the epoxide hydrolase from Aspergillus niger. Biocatal Biotransformation 21:357–364 Hwang S, Choi CY, Lee EY (2008a) One-pot biotransformation of racemic styrene oxide into (R)-1,2-phenylethandiol by two recombinant microbial epoxide hydrolases. Biotechnol Bioprocess Eng 13:274–278 Hwang Y-O, Kang SG, Woo J-H, Kwon KK, Sato T, Lee EY, Han MS, Kim S-J (2008b) Screening enantioselective epoxide hydrolase activities from marine microorganisms: detection of activities in Erythrobacter spp. Mar Biotechnol 10:366–373 Karboune S, Archelas A, Baratti J (2006) Properties of epoxide hydrolase from Aspergillus niger for the hydrolytic kinetic resolution of epoxides in pure organic media. Enzyme Microb Technol 39:318–324 Kim HS, Lee SJ, Lee EY (2006) Development and characterization of recombinant whole-cell biocatalysts expressing epoxide hydrolase from Rhodotorula glutinis for enantioselective resolution of racemic epoxides. J Mol Catal B 43:2–8 Kim HS, Lee OK, Hwang S, Kim BJ, Lee EY (2008) Biosynthesis of (R)-phenyl-1,2-ethanediol from racemic

123

1624 styrene oxide by using bacterial and marine fish epoxide hydrolases. Biotechnol Lett 30:127–133 Kronenburg NAE, de Bont JAM (2001) Effects of detergents on specific activity and enantioselectivity of the epoxide hydrolase from Rhodotorula glutinis. Enzyme Microb Technol 28:210–217 Lee EY (2007) Enantioselective hydrolysis of epichlorohydrin in organic solvents using recombinant epoxide hydrolase. J Ind Eng Chem 13:159–162 Lee EY (2008) Epoxide hydrolase-mediated enantioconvergent bioconversions to prepare chiral epoxides and alcohols. Biotechnol Lett 30:1509–1514 Lee EY, Shuler ML (2007) Molecular engineering of epoxide hydrolase and its application to asymmetric and enantioconvergent hydrolysis. Biotechnol Bioeng 98:318–327 Lee SJ, Kim HS, Kim SJ, Park S, Kim BJ, Shuler ML, Lee EY (2007) Cloning, expression and enantioselective hydrolytic catalysis of a microsomal epoxide hydrolase from a marine fish, Mugil cephalus. Biotechnol Lett 29:237–246 Nardini M, Ridder IS, Rozeboom HJ, Kalk KH, Rink R, Janssen DB, Dijkstra BW (1999) The X-ray structure of epoxide hydrolase from Agrobacterium radiobacter AD1. J Biol Chem 274:14579–14596 Reetz MT (2006) Directed evolution of enantioselective enzymes as catalysts for organic synthesis. Adv Catal 49:1–69 Reetz MT, Torre C, Eipper A, Lohmer R, Hermes M, Brunner B, Maichele A, Bocola M, Arand M, Cronin A, Genzel Y, Archelas A, Furstoss R (2004) Enhancing the enantioselectivity of an epoxide hydrolase by directed evolution. Org Lett 6:177–180 Rink R, Spelberg JHL, Pieters RJ, Kingma J, Nardini M, Kellogg RM, Dijkstra BW, Janssen DB (1999) Mutation

123

Biotechnol Lett (2009) 31:1617–1624 of tyrosine residues involved in the alkylation half reaction of epoxide hydrolase from Agrobacterium radiobacter AD1 results in improved enantioselectivity. J Am Chem Soc 121:7417–7418 Rui L, Cao L, Chen W, Reardon KF, Wood TK (2004) Active site engineering of the epoxide hydrolase from Agrobacterium radiobacter AD1 to enhance aerobic mineralization of cis-1,2-dichloroethylene in cells expressing an evolved toluene ortho-monooxygenase. J Biol Chem 279:46810– 46817 Rui L, Cao L, Chen W, Reardon KF, Wood TK (2005) Protein engineering of epoxide hydrolase from Agrobacterium radiobacter AD1 for enhanced activity and enantioselective production of (R)-1-phenylethane-1,2-diol. Appl Environ Microbiol 71:3995–4003 Schrader W, Eipper A, Pugh Dj, Reetz MT (2002) Secondgeneration MS-based high-throughput screening system for enantioselective catalysis and biocatalysis. Can J Chem 80:626–632 Strauss UT, Felfer U, Faber K (1999) Biocatalytic transformation of racemates into chiral building blocks in 100% chemical yield and 100% enantiomeric excess. Tetrahedron 10:107–117 van Loo B, Spelberg JHL, Kingma J, Sonke T, Wubbolts MG, Janssen DB (2004) Directed evolution of epoxide hydrolase from A. radiobacter toward higher enantioselectivity by error-prone PCR and DNA shuffling. Chem Biol 11:981–990 Zou J, Hallberg BM, Bergfors T, Oesch F, Arand M, Mowbray SL, Jones TA (2000) Structure of Aspergillus niger ˚ resolution: implication for the epoxide hydrolase at 1.8 A structure and function of the mammalian microsomal class of epoxide hydrolases. Structure 8:111–122