- Email: [email protected]
Enantioselectivity of a recombinant esterase from Pseudomonas fluorescens towards alcohols and carboxylic acids
Journal of Biotechnology 60 (1998) 105 – 111
Enantioselectivity of a recombinant esterase from Pseudomonas fluorescens towards alcohols and carboxylic acids Niels Krebsfa¨nger, Kerstin Schierholz, Uwe T. Bornscheuer * Uni6ersita¨t Stuttgart, Institut fu¨r Technische Biochemie, Allmandring 31, 70569 Stuttgart, Germany Received 28 July 1997; received in revised form 3 December 1997; accepted 5 December 1997
Abstract A recombinant esterase from Pseudomonas fluorescens (PFE) was produced from E. coli cultures and the enantioselectivity towards a series of racemic substrates was investigated. PFE exhibited high rate and enantioselectivity in the acylation of a-phenyl ethanol with vinyl acetate in toluene (E \ 100) and the hydrolysis of the corresponding acetate in phosphate buffer (E =58). In sharp contrast, extremely low enantioselectivity (E from 1.1 to 7) was found for the acylation of a series of 1,2-O-protected glycerol derivatives and the hydrolysis of 3-phenylbutyric acid methylester. Almost no reaction occured with a-phenyl propanol and its acetate and 2-phenylbutyric acid ethylester. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Esterase; Enantioselectivity; Pseudomonas fluorescens
1. Introduction Esterases belong to the group of hydrolases (Carboxylester hydrolases E.C. 3.1.1.1). They catalyze the formation or cleavage of ester bonds of water soluble substrates. Esterases are stable in a wide range of pH and temperature, and are also active in organic solvents (Wong and Whitesides,
* Corresponding author. Tel.: + 49 711 6854523; fax: + 49 711 6853196; e-mail: [email protected]
1994). Most of the commercially available esterases are of mammalian origin, and esterase from pig liver was often used in organic synthesis (Toone et al., 1990; Zhu and Tedford, 1990). However, most of the esterases are not available in recombinant form, thus properties—namely stability, activity and enantioselectivity —may vary significantly from batch to batch. Most importantly, modification of enzyme properties by, e.g. site-directed mutagenesis is not possible. Some microbial esterases have been cloned (Choi et al., 1990; Nishizawa et al., 1995), but only a
0168-1656/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 8 - 1 6 5 6 ( 9 7 ) 0 0 1 9 2 - 2
106
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
Fig. 1. Plasmid map of pJOE2792 bearing the gene encoding esterase from P. fluorescens (estF) and the rhamnose inducible promotor (rhaP).
few of them have been used in the resolution of chiral compounds. For instance, an esterase from Arthrobacter globiformis was successfully employed in the resolution of ethyl chrysanthemate derivatives (Nishizawa et al., 1993) and carboxylesterase NP was used for resolving phenoxypropionyl esters (Azzolina et al., 1995). In this paper, we describe the cultivation and isolation of a recombinant esterase from P. fluorescens (PFE; Pelletier and Altenbuchner, 1995). The enantioselectivity of this enzyme towards several racemic alcohols as well as two related racemic carboxylic acids was investigated with the intention to determine the substrate profile of this new esterase. The substrates used here are applied in their optical pure form as chiral auxiliaries or building blocks (Kogure and Eliel, 1984; Ley et al., 1990; Zadel et al., 1991).
2. Materials and methods
2.1. Chemicals All chemicals were purchased from Fluka, Buchs, Switzerland and Sigma, Steinheim, Germany, at the highest purity available.
2.2. Construction of plasmid pJOE2792 The plasmid pJOE2792 (Fig. 1) is a derivative of the rhamnose-inducible expression vector pJOE2702 (Volff et al., 1996). Six his-codons (His-Tag) were inserted into the vector pJOE2702 by the two complementary oligonucleotides S695 (5%-GATCCCATCATCATCATCATCATTGACTGCA-3%) and S696 (5%-TGCAATGCAGTCAATGATGATGATGATGATGG-3%) between the BamHI and HindIII site (plasmid
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
107
pJOE2775). The esterase gene was amplified by PCR from the plasmid pUE1251 (Choi et al., 1990) using the oligonucleotides S919 (5%-AAAACATATGAGCACATTTGTTGCAAAAG-3%) and S920 (5%-AAAAGGATCCGCGTTTCAAGAACGCCAACAG-3%), the PCR fragment cleaved with NdeI and BamHI and inserted into the vector pJOE2775, which was cut with the same enzymes to get pJOE2792.
was added at 37°C. Liberated acetic acid was titrated automatically in a pH-stat (Metrohm, Herisau, Switzerland) with 0.01 N NaOH in order to maintain the pH constant at pH 7.5. One unit of esterase activity was defined as the amount of enzyme, which liberates 1 mmol acetic acid per min under assay conditions. Protein content was determined with the bicinchoninic acid kit (Pierce, Rockford, IL).
2.3. Production of esterase
2.5. Electrophoretic methods
E. coli JM109 harboring the rhamnose inducible plasmid pJOE2792 were grown at 37°C in Luria-Bertani broth supplemented with ampicillin (100 mg ml − 1) until the early exponential phase (OD600 0.5, approx. 3 h). Gene expression was then induced by adding rhamnose (final concentration 0.2% w/v) to the culture, followed by further incubation for 3.5 h at 37°C yielding a wet cell weight of 1.4 g. Cells were collected by centrifugation (5000 rpm, 10 min, 4°C) and washed twice with sodium phosphate buffer (50 mM, pH 7.5, 4°C). After resuspension in the same buffer, cells were disrupted by sonication and cell debris was removed by centrifugation (5000 rpm, 15 min, 4°C). No formation of inclusion bodies was observed. Samples from this cell extract were directly used in hydrolysis experiments or lyophilized and immobilized on Celite 545 (see below) for transesterification reactions in organic solvent.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed with a PhastSystem and ready-to-use gels (Pharmacia, Freiburg, Germany) as described previously (Bornscheuer et al., 1994). As reference proteins the low-molecular-weight standard mixture from Pharmacia was used. After electrophoresis for 67 Vh the gels were stained with Coomassie Brilliant Blue. Activity staining (zymogram) was performed with a-naphthyl acetate and Fast Red (Sigma), leading upon esterase activity to a red complex (Bornscheuer et al., 1994).
2.4. Esterase acti6ity and protein determination Esterase activity during cultivation and of the lyophilized enzyme was measured photometrically in sodium phosphate buffer (50 mM, pH 7.5) using p-nitrophenol acetate (10 mM dissolved in DMSO) as substrate and the amount of p-nitrophenol was determined at 410 nm (e =15*103 M − 1 cm − 1) and 25°C. One unit (U) of activity was defined as the amount of enzyme releasing 1 mmol p-nitrophenol per min under assay conditions. Esterase activity for biotransformations was based on a pH-stat assay: To 20 ml of an emulsion containing 5% (w/v) ethyl acetate and 2% (w/v) gum arabic, a known amount of esterase
2.6. Immobilization on Celite 545 One hundred milligrams lyophilized cell extract were dissolved in 1.5 ml sodium phosphate buffer (50 mM, pH 7.5) and mixed for 20 min with 1 g Celite 545. The slurry was filtrated through a Bu¨chner-funnel and washed with 20 ml chilled acetone followed by drying at room temperature. Esterase activity was then determined by pH-stat as described above.
2.7. Chemical synthesis NMR spectra were recorded on a Bruker spectrometer (400 MHz). Optical rotations were determined with a polarimeter 241 from Perkin-Elmer. 1a and 2a: 40 mmol acetic acid chloride were dissolved in 20 ml pyridine at 4°C and the mixture was stirred thoroughly. Forty millimoles 1 or 2 were added dropwise and the mixture was stirred for 20 h at RT. The mixture was extracted twice with ether, twice with saturated sodium hydrogencarbonate solution and the combined organic
108
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
layer was dried over anhydrous Na2SO4 followed by evaporation of excess solvent. The acetates 1a or 2a were purified by silica gel chromatography (petrol ether:diethyl ether (PE:E)=2:1). 1a, 75% y, colorless oil; 1H-NMR (CDCl3): d (ppm) = 1.52 –1.55 (d, 3H, CH3 – CHO, 3J =6.6 Hz); 2.07 (s, 3H, CH3COO); 5.84 – 5.92 (q, 1H, CH, 3J =6.6 Hz); 7.28–7.36 (m, 5H, phenyl). 13C-NMR (CDCl3): d (ppm)= 21.35 and 22.21; 72.31; 76.53 and 77.55; 126.10; 127.87; 128.50; 141.69; 170.32. 2a: 35% y, colorless oil; 1H-NMR (CDCl3): d (ppm) = 0.84–0.9 (t, 3H, CH3 – CH2, 3J =7.4 Hz); 1.77 –1.95 (dq, 2H, CH2, 3J =7.4 Hz, 3J =6.9 Hz); 2.06 (s, 3H, CH3COO); 5.63 – 5.69 (t, 1H, CH, 3J= 6.9 Hz); 7.26 – 7.34 (m, 5H, phenyl). 13 C-NMR (CDCl3): d (ppm) = 9.91; 21.24; 29.30; 77.59 and 76.57; 126.57; 127.90; 128.38; 140.56; 170.38. 4–8: 1,2-O-protected glycerol derivatives 4 – 8 were synthesized from glycerol (130 mmol) and the corresponding ketone (100 mmol) in toluene (100 ml) using p-toluene sulfonic acid (1 mmol) as catalyst. The mixture was refluxed and water was removed in a Dean-Stark trap until the formation of water stopped. The mixture was washed with water, saturated sodium hydrogencarbonate solution and again with water. The organic layer was dried over anhydrous Na2SO4 followed by evaporation of excess solvent. 4 – 8 were isolated as colorless oils in 47 – 67% yields. All compounds are known from literature and structures were confirmed by 1H- and 13C-NMR (data not shown). 9: Thirty millimole 3-phenyl butyric acid and 66 mmol methanol were dissolved in 100 ml toluene and 1 ml conc. H2SO4 was added. The mixture was refluxed and water was removed in a DeanStark trap until the formation of water stopped. The mixture was washed with water, saturated sodium hydrogencarbonate solution and again with water. The organic layer was dried over anhydrous Na2SO4 followed by evaporation of excess solvent. The methyl ester 9 was purified by silica gel chromatography (PE:E= 2:1). 9, 62% y, colorless oil; 1H-NMR (CDCl3): d (ppm) = 1.28 – 1.30 (d, CH3, J=7 Hz); 2.52 – 2.61 (m, CH2COO, J = 6 Hz); 3.23–3.29 (m, CH, J= 7 Hz); 3.61 (s, OCH3); 7.19–7.32 (m, 5 aryl-H). 13C-NMR (CDCl3): d (ppm)=21.77; 36.43; 42.72; 51.46;
126.41; 126.71; 128.51; 145.70; 172.80. 10: Synthesis was performed similarly to 9, but ethanol was used instead, 25% y, colorless oil; 1H-NMR (CDCl3): d (ppm)= 0.87–0.90 (t, CH3-CH2, J= 7 Hz); 1.18–1.20 (d, CH3CH2, J=7 Hz); 1.74–1.83 and 2.05–2.13 (m, CH3CH2, J =7 Hz); 3.41–3.44 (t, CH, J= 7 Hz); 4.05–4.17 (m, CH3CH2, J= 7 Hz; 7.21–7.32 (m, 5 aryl-H). 13C-NMR (CDCl3): d (ppm)= 12.17; 14.15; 26.81; 53.56; 60.58; 127.09; 127.95; 128.51; 139.28; 174.05.
2.8. Biotransformations 2.8.1. Hydrolysis Hydrolysis of 9 was performed in a pH-stat system by adding 1 mmol of methylester to 20 ml water (pH 7.5) at 40°C and 500 rpm followed by addition of 400 U lyophilized PFE. After consumption of 0.1 N NaOH corresponded to the desired conversion, the mixture was extracted twice with chloroform. The organic phase was dried over anhydrous Na2SO4 followed by evaporation of solvent. Substrate and product were separated by silical gel column chromatography (PE:E= 2:1) and the optical rotation was determined to [a]20 D :15.4° (c= 1, CHCl3) at 60% conv. and [a]20 :9.4° (c =1, CHCl3) at 40% conv. EnanD tiomeric excess and enantioselectivity given in Table 1 were calculated from the ratio of the optical rotation of the sample to the literature value of the pure compound, because separation of enantiomers by GC or HPLC was not possible. Hydrolysis of 10 gave almost no conversion after 24 h and substrate and product were not isolated. Hydrolysis of 1a (0.5 mmol) and 2a (0.5 mmol) was performed in round bottom flasks in 3 ml sodium phosphate buffer (50 mM, pH 7.5) and 2 ml toluene at 40°C using 200 U PFE. Enantiomeric excess was determined by gas chromatography. 2.8.2. Acylation Racemic alcohols: 0.5 mmol 1–8 were dissolved in 3 ml toluene at 40°C (dried over molecular sieve) containing 1 mmol vinyl acetate and 200 U Celite-esterase were added. Samples from the reaction mixture were centrifugated and the supernatant was analyzed by GC.
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
109
2.9. GC analysis Gas chromatographic analysis (HRGC Mega 2 series, Autosampler A 200S, software package Chrom Card for Windows, Fisons Instruments, Mainz-Kastel, Germany) was performed using a chiral stationary phase (FS-cyclodex beta-I/P = heptakis-(2,3,6-tri-O-methyl)-b-cyclodextrin, 50 m, i.d. 0.25 mm, CS-Chromatographie Service GmbH, Langerwehe, Germany), a flame ionization detector (FID) and hydrogen as carrier gas (60 kPa, split 1:100). For all enantiomers a base line separation was achieved without derivatization and the enantiomeric excess was calculated from the corresponding peak areas. Conversion and enantioselectivity were calculated using the following equations: c =eeS/(eeS +eeP) and E = Table 1 Results of esterase-catalyzed biotransformations of compounds 1-10 Compound
1 1a 2 2a 3
Enantiomeric excess
[%eeS]
[%eeP]
99 60 4 B1 6
99 94 74 27 3
(S) (R) (S) (R) (R)
Conversiona [%]
Ea
[ln {(1 − c)(1− eeS)}]/[ln {(1 − c)(1+eeS)}] (Chen et al., 1982). Configurations of 1–3 were assigned by comparison with pure standards, for 4–8 a similar elution order in gas chromatography as found for 3 was assumed, and in the case of 9 the configuration was assigned from the sense of optical rotation based on literature values.
3. Results and Discussion
(R) (S) (R) (S) (S)
50 39 5 B1 67
\100 58 7 2 1.1
4
45 (R)
38 (S)
54
3.5
5
83 (R)
35 (S)
70
4.9
6
33 (R)
39 (S)
46
3.0
7
80 (R)
30 (S)
72
4.2
8
18 (R)
14 (S)
55
1.5
9
51 (S)
34 (R)
60b
3.4
9
31 (S)
47 (R)
40b
3.7
10
n.d.
n.d.
n.d.
n.d .
a
Fig. 2. SDS-PAGE indicating production (Coomassie brilliant blue staining, lanes 1 – 6) and activity (a-naphthyl acetate/Fast red staining, lanes 7 – 10) of esterase from P. fluorescens (PFE). Lanes 1 and 6: low molecular weight standard; lanes 2 and 7: before induction; lanes 3 and 8: 1 h after induction; lanes 4 and 9: 3.5 h after induction; lanes 5 and 10: crude cell extract.
Calculated from the enantiomeric excess according to Chen et al. (1982). b As determined from NaOH consumption.
3.1. Production and isolation of P. fluorescens esterase The esterase was produced by cultivation of E. coli harboring the esterase gene in 250 ml LB-media. Before induction, no esterase activity could be detected. After induction with rhamnose expression took place, which was confirmed by SDS-PAGE and activity staining with a-naphthyl acetate and Fast Red (Fig. 2). No formation of inclusion bodies was observed. The esterase was isolated as described in Materials and methods and the cell extract had a specific activity of 90 U ml − 1 as determined with p-nitrophenyl acetate (pNPA). After lyophilization, a specific activity of 86 U mg − 1 protein (pNPA) and 28 U mg − 1 protein using ethyl acetate as substrate were determined. For reactions in organic solvents, the esterase was immobilized on Celite 545 resulting in an activity of 650 U g − 1 carrier. Crude PFE was purified by immobilized zinc ion affinity chro-
110
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
Fig. 3. Chiral substrates used in the esterase-catalyzed biotransformations. Ac = Acetyl-, Me =Methyl-, Et =Ethyl-, Pr= Propyl-, Ph =C6H5-.
matography yielding a homogeneous esterase with a specific activity of 120 U mg − 1, but the pure enzyme had no activity in organic solvents, even after immobilization on Celite 545. PFE is active in a wide range of pH (5 – 10) and temperature (30 – 70°C), but is rather unstable at temperatures above 50°C. Crude and purified PFE exhibited similar rates and enantioselectivity in hydrolysis (data not shown).
3.2. Kinetic resolution of racemic alcohols 1 – 8 and esters 1a– 2a and 9 – 10 The biotransformations of compounds 1 – 10 (Fig. 3) resulted in a quite diverse substrate and enantioselectivity pattern of PFE (Table 1). The acylation of a-phenyl ethanol 1 with vinyl acetate in toluene (E \ 100) and the hydrolysis of the acetate 1a (E = 58) proceeded at good reaction rates and allowed the preparation of optical pure substrate and product. In sharp contrast, reactions with the closely related a-phenyl propanol 2 or its acetate 2a gave almost no conversion (even after 200 h) and also the enantioselectivity (E) dropped to values between 2 and 7. In the case of glycerol derivatives 3–8 reactivity was not a problem (almost quantitative conversions were achieved after 24 h), but again the enantioselectivity was never satisfying. Interestingly, the variation of the protecting group led to an altered enantioselectivity, which led to an increase of E from 1.1 for compound 3 to 4.9 for compound 5. Also, the activity of PFE towards chiral carboxylic acid esters 9 – 10 was quite diverse. Whereas the hydrolysis of 9 proceeded with acceptable conversion (40% after 8 h) and moderate E (3.5), almost no reaction was observed with substrate 10.
4. Conclusion In summary, PFE is easy to produce, active and stable at elevated temperatures and in organic solvents. PFE accepts a rather wide range of substrates, but reaction rate and enantioselectivity differ significantly. a-Phenyl ethanol as well as the series of glycerol derivatives and 3-phenyl butyric acid methyl ester were converted at acceptable rates. However, the enantioselectivity was excellent only in the case of a-phenyl ethanol or its acetate and moderate for compounds 5, 7 and 9. Compared to literature, a-phenyl propanol could be much more efficiently resolved with, e.g. porcine pancreatic lipase (E \ 100) (Morgan et al., 1992). Small changes in the structure of the substrate led to a drastic change in the reaction rate as found in the case of a-phenyl propanol and 2-phenylbutyric acid ethyl ester. Previously, we investigated the resolution of 1,2-O-protected glycerol derivatives using lipase from Pseudomonas cepacia (Amano PS) and also observed low enantioselectivity and a similar pattern of changes in enantioselectivity as found here for compounds 3–8 (Gaziola et al., 1996). In accordance with the investigations of Weissfloch and Kazlauskas (1995) lipases (and PFE) seem to lack a sufficient chiral recognition for primary alcohols, especially when an oxygen is attached to the chiral center. PFE has approx. 55% homology to bromoperoxidase A2, of which the structure has been resolved (Hecht et al., 1994). On the basis of the data presented here, we are now using computer modeling to predict mutations to alter the substrate and enantioselectivity of this esterase.
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
Acknowledgements The authors would like to thank Prof. O.J. Yoo, KAIST, Taejon, Korea and Dr J. Altenbuchner, Institute for Industrial Genetics, University of Stuttgart, Germany for providing us with the esterase gene, and Prof. R.D. Schmid, Institute for Technical Biochemistry, University of Stuttgart, Germany, for useful discussions.
References Azzolina, O., Collina, S., Vercesi, D., 1995. Stereoselective hydrolysis by esterase: A strategy for resolving 2-(R,R%phenoxy)propionyl ester racemates. Il Farmaco 50, 725– 733. Bornscheuer, U., Reif, O.-W., Lausch, R., Freitag, R., Scheper, T., Kolisis, F.N., Menge, U., 1994. Lipase of Pseudomonas cepacia for biotechnological purposes: purification, crystallization and characterization. Biochim. Biophys. Acta 1201, 55–60. Chen, C.S., Fujimoto, Y., Sih, C.J., 1982. Quantitative analyses of biochemical kinetic resolutions of enantiomers. J. Am. Chem. Soc. 104, 7294–7299. Choi, K.D., Jeohn, G.H., Rhee, J.S., Yoo, O.J., 1990. Cloning and nucleotide sequence of an esterase gene from Pseudomonas fluorescens and expression of the gene in Escherichia coli. Agric. Biol. Chem. 54, 2039–2045. Gaziola, L., Bornscheuer, U., Schmid, R.D., 1996. A rapid and effective separation of enantiomers of glycerol derivatives by gas chromatography and their lipase-catalyzed biotransformation. Enantiomer 1, 49–54. Hecht, H.J., Sobek, H., Haag, T., Pfeifer, O., van Pe´e, K.H., 1994. The metal-ion-free oxidoreductase from Streptomyces aureofaciens has an a/b hydrolase fold. Nat. Struct. Biol. 1, 532–537. Kogure, T., Eliel, E.L., 1984. A convergent asymmetric synthesis of ( −)-malyngolide and its three stereoisomers. J. Org. Chem. 49, 576 –578.
.
111
Ley, S.V., Parra, M., Redgrave, A.J., Sternfeld, F., 1990. Microbial oxidation in synthesis: preparation of myoinositol phosphates and related cyclitol derivatives from benzene. Tetrahedron 46, 4995 – 5026. Morgan, B., Oehlschlager, A.C., Stokes, T.M., 1992. Enzyme reactions in apolar solvent. 5. The effect of adjacent unsaturation on the PPL-catalyzed kinetic resolution of secondary alcohols. J. Org. Chem. 57, 3231 – 3236. Nishizawa, M., Shimizu, M., Ohkawa, O., Kanaoka, M., 1995. Stereoselective production of ( + )-trans-chrysanthemic acid by a microbial esterase: Cloning, nucleotide sequence, and overexpression of the esterase gene of Arthrobacter globiformis in Escherichia coli. Appl. Environm. Microbiol. 61, 3208 – 3215. Nishizawa, M., Gomi, H., Kishimoto, F., 1993. Purification and some properties of carboxylesterase from Arthrobacter globiformis; Stereoselective hydrolysis of ethyl chrysanthemate. Biosci. Biotech. Biochem. 57, 594 – 598. Pelletier, I., Altenbuchner, J., 1995. A bacterial esterase is homologous with non-haem haloperoxidases and displays brominating activity. Microbiology 141, 459 – 468. Toone, E.J., Werth, M.J., Jones, J.B., 1990. Active-site model for interpreting and predicting the specificity of pig liver esterase. J. Am. Chem. Soc. 112, 4946 – 4952. Volff, J.N., Eichenseer, C., Viell, P., Piendl, W., Altenbuchner, J., 1996. Nucleotide sequence and role in DNA amplification of the direct repeats composing the amplifiable element AUD1 of Streptomyces li6idans 66. Mol. Microbiol. 21, 1037 – 1047. Weissfloch, A.N.E., Kazlauskas, R.J., 1995. Enantiopreference of lipase from Pseudomonas cepacia toward primary alcohols. J. Org. Chem. 60, 6959 – 6969. Wong, C.H., Whitesides, G.M., 1994. Enzymes in Synthetic Organic Chemistry. Pergamon Press, Oxford. Zadel, G., Rieger, G., Breitmaier, E., 1991. Enantioselective terpene syntheses by Diels-Alder reaction of l-(1-arylalkoxy)-2-methyl-1,3-butadiene with isoprene. Liebigs Ann. Chem. 1343 – 1346. Zhu, L.M., Tedford, M.C., 1990. Applications of pig liver esterases (PLE) in asymmetric synthesis. Tetrahedron 46, 6587 – 6611.
.
Enantioselectivity of a recombinant esterase from Pseudomonas fluorescens towards alcohols and carboxylic acids Niels Krebsfa¨nger, Kerstin Schierholz, Uwe T. Bornscheuer * Uni6ersita¨t Stuttgart, Institut fu¨r Technische Biochemie, Allmandring 31, 70569 Stuttgart, Germany Received 28 July 1997; received in revised form 3 December 1997; accepted 5 December 1997
Abstract A recombinant esterase from Pseudomonas fluorescens (PFE) was produced from E. coli cultures and the enantioselectivity towards a series of racemic substrates was investigated. PFE exhibited high rate and enantioselectivity in the acylation of a-phenyl ethanol with vinyl acetate in toluene (E \ 100) and the hydrolysis of the corresponding acetate in phosphate buffer (E =58). In sharp contrast, extremely low enantioselectivity (E from 1.1 to 7) was found for the acylation of a series of 1,2-O-protected glycerol derivatives and the hydrolysis of 3-phenylbutyric acid methylester. Almost no reaction occured with a-phenyl propanol and its acetate and 2-phenylbutyric acid ethylester. © 1998 Elsevier Science B.V. All rights reserved. Keywords: Esterase; Enantioselectivity; Pseudomonas fluorescens
1. Introduction Esterases belong to the group of hydrolases (Carboxylester hydrolases E.C. 3.1.1.1). They catalyze the formation or cleavage of ester bonds of water soluble substrates. Esterases are stable in a wide range of pH and temperature, and are also active in organic solvents (Wong and Whitesides,
* Corresponding author. Tel.: + 49 711 6854523; fax: + 49 711 6853196; e-mail: [email protected]
1994). Most of the commercially available esterases are of mammalian origin, and esterase from pig liver was often used in organic synthesis (Toone et al., 1990; Zhu and Tedford, 1990). However, most of the esterases are not available in recombinant form, thus properties—namely stability, activity and enantioselectivity —may vary significantly from batch to batch. Most importantly, modification of enzyme properties by, e.g. site-directed mutagenesis is not possible. Some microbial esterases have been cloned (Choi et al., 1990; Nishizawa et al., 1995), but only a
0168-1656/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 8 - 1 6 5 6 ( 9 7 ) 0 0 1 9 2 - 2
106
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
Fig. 1. Plasmid map of pJOE2792 bearing the gene encoding esterase from P. fluorescens (estF) and the rhamnose inducible promotor (rhaP).
few of them have been used in the resolution of chiral compounds. For instance, an esterase from Arthrobacter globiformis was successfully employed in the resolution of ethyl chrysanthemate derivatives (Nishizawa et al., 1993) and carboxylesterase NP was used for resolving phenoxypropionyl esters (Azzolina et al., 1995). In this paper, we describe the cultivation and isolation of a recombinant esterase from P. fluorescens (PFE; Pelletier and Altenbuchner, 1995). The enantioselectivity of this enzyme towards several racemic alcohols as well as two related racemic carboxylic acids was investigated with the intention to determine the substrate profile of this new esterase. The substrates used here are applied in their optical pure form as chiral auxiliaries or building blocks (Kogure and Eliel, 1984; Ley et al., 1990; Zadel et al., 1991).
2. Materials and methods
2.1. Chemicals All chemicals were purchased from Fluka, Buchs, Switzerland and Sigma, Steinheim, Germany, at the highest purity available.
2.2. Construction of plasmid pJOE2792 The plasmid pJOE2792 (Fig. 1) is a derivative of the rhamnose-inducible expression vector pJOE2702 (Volff et al., 1996). Six his-codons (His-Tag) were inserted into the vector pJOE2702 by the two complementary oligonucleotides S695 (5%-GATCCCATCATCATCATCATCATTGACTGCA-3%) and S696 (5%-TGCAATGCAGTCAATGATGATGATGATGATGG-3%) between the BamHI and HindIII site (plasmid
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
107
pJOE2775). The esterase gene was amplified by PCR from the plasmid pUE1251 (Choi et al., 1990) using the oligonucleotides S919 (5%-AAAACATATGAGCACATTTGTTGCAAAAG-3%) and S920 (5%-AAAAGGATCCGCGTTTCAAGAACGCCAACAG-3%), the PCR fragment cleaved with NdeI and BamHI and inserted into the vector pJOE2775, which was cut with the same enzymes to get pJOE2792.
was added at 37°C. Liberated acetic acid was titrated automatically in a pH-stat (Metrohm, Herisau, Switzerland) with 0.01 N NaOH in order to maintain the pH constant at pH 7.5. One unit of esterase activity was defined as the amount of enzyme, which liberates 1 mmol acetic acid per min under assay conditions. Protein content was determined with the bicinchoninic acid kit (Pierce, Rockford, IL).
2.3. Production of esterase
2.5. Electrophoretic methods
E. coli JM109 harboring the rhamnose inducible plasmid pJOE2792 were grown at 37°C in Luria-Bertani broth supplemented with ampicillin (100 mg ml − 1) until the early exponential phase (OD600 0.5, approx. 3 h). Gene expression was then induced by adding rhamnose (final concentration 0.2% w/v) to the culture, followed by further incubation for 3.5 h at 37°C yielding a wet cell weight of 1.4 g. Cells were collected by centrifugation (5000 rpm, 10 min, 4°C) and washed twice with sodium phosphate buffer (50 mM, pH 7.5, 4°C). After resuspension in the same buffer, cells were disrupted by sonication and cell debris was removed by centrifugation (5000 rpm, 15 min, 4°C). No formation of inclusion bodies was observed. Samples from this cell extract were directly used in hydrolysis experiments or lyophilized and immobilized on Celite 545 (see below) for transesterification reactions in organic solvent.
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed with a PhastSystem and ready-to-use gels (Pharmacia, Freiburg, Germany) as described previously (Bornscheuer et al., 1994). As reference proteins the low-molecular-weight standard mixture from Pharmacia was used. After electrophoresis for 67 Vh the gels were stained with Coomassie Brilliant Blue. Activity staining (zymogram) was performed with a-naphthyl acetate and Fast Red (Sigma), leading upon esterase activity to a red complex (Bornscheuer et al., 1994).
2.4. Esterase acti6ity and protein determination Esterase activity during cultivation and of the lyophilized enzyme was measured photometrically in sodium phosphate buffer (50 mM, pH 7.5) using p-nitrophenol acetate (10 mM dissolved in DMSO) as substrate and the amount of p-nitrophenol was determined at 410 nm (e =15*103 M − 1 cm − 1) and 25°C. One unit (U) of activity was defined as the amount of enzyme releasing 1 mmol p-nitrophenol per min under assay conditions. Esterase activity for biotransformations was based on a pH-stat assay: To 20 ml of an emulsion containing 5% (w/v) ethyl acetate and 2% (w/v) gum arabic, a known amount of esterase
2.6. Immobilization on Celite 545 One hundred milligrams lyophilized cell extract were dissolved in 1.5 ml sodium phosphate buffer (50 mM, pH 7.5) and mixed for 20 min with 1 g Celite 545. The slurry was filtrated through a Bu¨chner-funnel and washed with 20 ml chilled acetone followed by drying at room temperature. Esterase activity was then determined by pH-stat as described above.
2.7. Chemical synthesis NMR spectra were recorded on a Bruker spectrometer (400 MHz). Optical rotations were determined with a polarimeter 241 from Perkin-Elmer. 1a and 2a: 40 mmol acetic acid chloride were dissolved in 20 ml pyridine at 4°C and the mixture was stirred thoroughly. Forty millimoles 1 or 2 were added dropwise and the mixture was stirred for 20 h at RT. The mixture was extracted twice with ether, twice with saturated sodium hydrogencarbonate solution and the combined organic
108
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
layer was dried over anhydrous Na2SO4 followed by evaporation of excess solvent. The acetates 1a or 2a were purified by silica gel chromatography (petrol ether:diethyl ether (PE:E)=2:1). 1a, 75% y, colorless oil; 1H-NMR (CDCl3): d (ppm) = 1.52 –1.55 (d, 3H, CH3 – CHO, 3J =6.6 Hz); 2.07 (s, 3H, CH3COO); 5.84 – 5.92 (q, 1H, CH, 3J =6.6 Hz); 7.28–7.36 (m, 5H, phenyl). 13C-NMR (CDCl3): d (ppm)= 21.35 and 22.21; 72.31; 76.53 and 77.55; 126.10; 127.87; 128.50; 141.69; 170.32. 2a: 35% y, colorless oil; 1H-NMR (CDCl3): d (ppm) = 0.84–0.9 (t, 3H, CH3 – CH2, 3J =7.4 Hz); 1.77 –1.95 (dq, 2H, CH2, 3J =7.4 Hz, 3J =6.9 Hz); 2.06 (s, 3H, CH3COO); 5.63 – 5.69 (t, 1H, CH, 3J= 6.9 Hz); 7.26 – 7.34 (m, 5H, phenyl). 13 C-NMR (CDCl3): d (ppm) = 9.91; 21.24; 29.30; 77.59 and 76.57; 126.57; 127.90; 128.38; 140.56; 170.38. 4–8: 1,2-O-protected glycerol derivatives 4 – 8 were synthesized from glycerol (130 mmol) and the corresponding ketone (100 mmol) in toluene (100 ml) using p-toluene sulfonic acid (1 mmol) as catalyst. The mixture was refluxed and water was removed in a Dean-Stark trap until the formation of water stopped. The mixture was washed with water, saturated sodium hydrogencarbonate solution and again with water. The organic layer was dried over anhydrous Na2SO4 followed by evaporation of excess solvent. 4 – 8 were isolated as colorless oils in 47 – 67% yields. All compounds are known from literature and structures were confirmed by 1H- and 13C-NMR (data not shown). 9: Thirty millimole 3-phenyl butyric acid and 66 mmol methanol were dissolved in 100 ml toluene and 1 ml conc. H2SO4 was added. The mixture was refluxed and water was removed in a DeanStark trap until the formation of water stopped. The mixture was washed with water, saturated sodium hydrogencarbonate solution and again with water. The organic layer was dried over anhydrous Na2SO4 followed by evaporation of excess solvent. The methyl ester 9 was purified by silica gel chromatography (PE:E= 2:1). 9, 62% y, colorless oil; 1H-NMR (CDCl3): d (ppm) = 1.28 – 1.30 (d, CH3, J=7 Hz); 2.52 – 2.61 (m, CH2COO, J = 6 Hz); 3.23–3.29 (m, CH, J= 7 Hz); 3.61 (s, OCH3); 7.19–7.32 (m, 5 aryl-H). 13C-NMR (CDCl3): d (ppm)=21.77; 36.43; 42.72; 51.46;
126.41; 126.71; 128.51; 145.70; 172.80. 10: Synthesis was performed similarly to 9, but ethanol was used instead, 25% y, colorless oil; 1H-NMR (CDCl3): d (ppm)= 0.87–0.90 (t, CH3-CH2, J= 7 Hz); 1.18–1.20 (d, CH3CH2, J=7 Hz); 1.74–1.83 and 2.05–2.13 (m, CH3CH2, J =7 Hz); 3.41–3.44 (t, CH, J= 7 Hz); 4.05–4.17 (m, CH3CH2, J= 7 Hz; 7.21–7.32 (m, 5 aryl-H). 13C-NMR (CDCl3): d (ppm)= 12.17; 14.15; 26.81; 53.56; 60.58; 127.09; 127.95; 128.51; 139.28; 174.05.
2.8. Biotransformations 2.8.1. Hydrolysis Hydrolysis of 9 was performed in a pH-stat system by adding 1 mmol of methylester to 20 ml water (pH 7.5) at 40°C and 500 rpm followed by addition of 400 U lyophilized PFE. After consumption of 0.1 N NaOH corresponded to the desired conversion, the mixture was extracted twice with chloroform. The organic phase was dried over anhydrous Na2SO4 followed by evaporation of solvent. Substrate and product were separated by silical gel column chromatography (PE:E= 2:1) and the optical rotation was determined to [a]20 D :15.4° (c= 1, CHCl3) at 60% conv. and [a]20 :9.4° (c =1, CHCl3) at 40% conv. EnanD tiomeric excess and enantioselectivity given in Table 1 were calculated from the ratio of the optical rotation of the sample to the literature value of the pure compound, because separation of enantiomers by GC or HPLC was not possible. Hydrolysis of 10 gave almost no conversion after 24 h and substrate and product were not isolated. Hydrolysis of 1a (0.5 mmol) and 2a (0.5 mmol) was performed in round bottom flasks in 3 ml sodium phosphate buffer (50 mM, pH 7.5) and 2 ml toluene at 40°C using 200 U PFE. Enantiomeric excess was determined by gas chromatography. 2.8.2. Acylation Racemic alcohols: 0.5 mmol 1–8 were dissolved in 3 ml toluene at 40°C (dried over molecular sieve) containing 1 mmol vinyl acetate and 200 U Celite-esterase were added. Samples from the reaction mixture were centrifugated and the supernatant was analyzed by GC.
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
109
2.9. GC analysis Gas chromatographic analysis (HRGC Mega 2 series, Autosampler A 200S, software package Chrom Card for Windows, Fisons Instruments, Mainz-Kastel, Germany) was performed using a chiral stationary phase (FS-cyclodex beta-I/P = heptakis-(2,3,6-tri-O-methyl)-b-cyclodextrin, 50 m, i.d. 0.25 mm, CS-Chromatographie Service GmbH, Langerwehe, Germany), a flame ionization detector (FID) and hydrogen as carrier gas (60 kPa, split 1:100). For all enantiomers a base line separation was achieved without derivatization and the enantiomeric excess was calculated from the corresponding peak areas. Conversion and enantioselectivity were calculated using the following equations: c =eeS/(eeS +eeP) and E = Table 1 Results of esterase-catalyzed biotransformations of compounds 1-10 Compound
1 1a 2 2a 3
Enantiomeric excess
[%eeS]
[%eeP]
99 60 4 B1 6
99 94 74 27 3
(S) (R) (S) (R) (R)
Conversiona [%]
Ea
[ln {(1 − c)(1− eeS)}]/[ln {(1 − c)(1+eeS)}] (Chen et al., 1982). Configurations of 1–3 were assigned by comparison with pure standards, for 4–8 a similar elution order in gas chromatography as found for 3 was assumed, and in the case of 9 the configuration was assigned from the sense of optical rotation based on literature values.
3. Results and Discussion
(R) (S) (R) (S) (S)
50 39 5 B1 67
\100 58 7 2 1.1
4
45 (R)
38 (S)
54
3.5
5
83 (R)
35 (S)
70
4.9
6
33 (R)
39 (S)
46
3.0
7
80 (R)
30 (S)
72
4.2
8
18 (R)
14 (S)
55
1.5
9
51 (S)
34 (R)
60b
3.4
9
31 (S)
47 (R)
40b
3.7
10
n.d.
n.d.
n.d.
n.d .
a
Fig. 2. SDS-PAGE indicating production (Coomassie brilliant blue staining, lanes 1 – 6) and activity (a-naphthyl acetate/Fast red staining, lanes 7 – 10) of esterase from P. fluorescens (PFE). Lanes 1 and 6: low molecular weight standard; lanes 2 and 7: before induction; lanes 3 and 8: 1 h after induction; lanes 4 and 9: 3.5 h after induction; lanes 5 and 10: crude cell extract.
Calculated from the enantiomeric excess according to Chen et al. (1982). b As determined from NaOH consumption.
3.1. Production and isolation of P. fluorescens esterase The esterase was produced by cultivation of E. coli harboring the esterase gene in 250 ml LB-media. Before induction, no esterase activity could be detected. After induction with rhamnose expression took place, which was confirmed by SDS-PAGE and activity staining with a-naphthyl acetate and Fast Red (Fig. 2). No formation of inclusion bodies was observed. The esterase was isolated as described in Materials and methods and the cell extract had a specific activity of 90 U ml − 1 as determined with p-nitrophenyl acetate (pNPA). After lyophilization, a specific activity of 86 U mg − 1 protein (pNPA) and 28 U mg − 1 protein using ethyl acetate as substrate were determined. For reactions in organic solvents, the esterase was immobilized on Celite 545 resulting in an activity of 650 U g − 1 carrier. Crude PFE was purified by immobilized zinc ion affinity chro-
110
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
Fig. 3. Chiral substrates used in the esterase-catalyzed biotransformations. Ac = Acetyl-, Me =Methyl-, Et =Ethyl-, Pr= Propyl-, Ph =C6H5-.
matography yielding a homogeneous esterase with a specific activity of 120 U mg − 1, but the pure enzyme had no activity in organic solvents, even after immobilization on Celite 545. PFE is active in a wide range of pH (5 – 10) and temperature (30 – 70°C), but is rather unstable at temperatures above 50°C. Crude and purified PFE exhibited similar rates and enantioselectivity in hydrolysis (data not shown).
3.2. Kinetic resolution of racemic alcohols 1 – 8 and esters 1a– 2a and 9 – 10 The biotransformations of compounds 1 – 10 (Fig. 3) resulted in a quite diverse substrate and enantioselectivity pattern of PFE (Table 1). The acylation of a-phenyl ethanol 1 with vinyl acetate in toluene (E \ 100) and the hydrolysis of the acetate 1a (E = 58) proceeded at good reaction rates and allowed the preparation of optical pure substrate and product. In sharp contrast, reactions with the closely related a-phenyl propanol 2 or its acetate 2a gave almost no conversion (even after 200 h) and also the enantioselectivity (E) dropped to values between 2 and 7. In the case of glycerol derivatives 3–8 reactivity was not a problem (almost quantitative conversions were achieved after 24 h), but again the enantioselectivity was never satisfying. Interestingly, the variation of the protecting group led to an altered enantioselectivity, which led to an increase of E from 1.1 for compound 3 to 4.9 for compound 5. Also, the activity of PFE towards chiral carboxylic acid esters 9 – 10 was quite diverse. Whereas the hydrolysis of 9 proceeded with acceptable conversion (40% after 8 h) and moderate E (3.5), almost no reaction was observed with substrate 10.
4. Conclusion In summary, PFE is easy to produce, active and stable at elevated temperatures and in organic solvents. PFE accepts a rather wide range of substrates, but reaction rate and enantioselectivity differ significantly. a-Phenyl ethanol as well as the series of glycerol derivatives and 3-phenyl butyric acid methyl ester were converted at acceptable rates. However, the enantioselectivity was excellent only in the case of a-phenyl ethanol or its acetate and moderate for compounds 5, 7 and 9. Compared to literature, a-phenyl propanol could be much more efficiently resolved with, e.g. porcine pancreatic lipase (E \ 100) (Morgan et al., 1992). Small changes in the structure of the substrate led to a drastic change in the reaction rate as found in the case of a-phenyl propanol and 2-phenylbutyric acid ethyl ester. Previously, we investigated the resolution of 1,2-O-protected glycerol derivatives using lipase from Pseudomonas cepacia (Amano PS) and also observed low enantioselectivity and a similar pattern of changes in enantioselectivity as found here for compounds 3–8 (Gaziola et al., 1996). In accordance with the investigations of Weissfloch and Kazlauskas (1995) lipases (and PFE) seem to lack a sufficient chiral recognition for primary alcohols, especially when an oxygen is attached to the chiral center. PFE has approx. 55% homology to bromoperoxidase A2, of which the structure has been resolved (Hecht et al., 1994). On the basis of the data presented here, we are now using computer modeling to predict mutations to alter the substrate and enantioselectivity of this esterase.
N. Krebsfa¨nger et al. / Journal of Biotechnology 60 (1998) 105–111
Acknowledgements The authors would like to thank Prof. O.J. Yoo, KAIST, Taejon, Korea and Dr J. Altenbuchner, Institute for Industrial Genetics, University of Stuttgart, Germany for providing us with the esterase gene, and Prof. R.D. Schmid, Institute for Technical Biochemistry, University of Stuttgart, Germany, for useful discussions.
References Azzolina, O., Collina, S., Vercesi, D., 1995. Stereoselective hydrolysis by esterase: A strategy for resolving 2-(R,R%phenoxy)propionyl ester racemates. Il Farmaco 50, 725– 733. Bornscheuer, U., Reif, O.-W., Lausch, R., Freitag, R., Scheper, T., Kolisis, F.N., Menge, U., 1994. Lipase of Pseudomonas cepacia for biotechnological purposes: purification, crystallization and characterization. Biochim. Biophys. Acta 1201, 55–60. Chen, C.S., Fujimoto, Y., Sih, C.J., 1982. Quantitative analyses of biochemical kinetic resolutions of enantiomers. J. Am. Chem. Soc. 104, 7294–7299. Choi, K.D., Jeohn, G.H., Rhee, J.S., Yoo, O.J., 1990. Cloning and nucleotide sequence of an esterase gene from Pseudomonas fluorescens and expression of the gene in Escherichia coli. Agric. Biol. Chem. 54, 2039–2045. Gaziola, L., Bornscheuer, U., Schmid, R.D., 1996. A rapid and effective separation of enantiomers of glycerol derivatives by gas chromatography and their lipase-catalyzed biotransformation. Enantiomer 1, 49–54. Hecht, H.J., Sobek, H., Haag, T., Pfeifer, O., van Pe´e, K.H., 1994. The metal-ion-free oxidoreductase from Streptomyces aureofaciens has an a/b hydrolase fold. Nat. Struct. Biol. 1, 532–537. Kogure, T., Eliel, E.L., 1984. A convergent asymmetric synthesis of ( −)-malyngolide and its three stereoisomers. J. Org. Chem. 49, 576 –578.
.
111
Ley, S.V., Parra, M., Redgrave, A.J., Sternfeld, F., 1990. Microbial oxidation in synthesis: preparation of myoinositol phosphates and related cyclitol derivatives from benzene. Tetrahedron 46, 4995 – 5026. Morgan, B., Oehlschlager, A.C., Stokes, T.M., 1992. Enzyme reactions in apolar solvent. 5. The effect of adjacent unsaturation on the PPL-catalyzed kinetic resolution of secondary alcohols. J. Org. Chem. 57, 3231 – 3236. Nishizawa, M., Shimizu, M., Ohkawa, O., Kanaoka, M., 1995. Stereoselective production of ( + )-trans-chrysanthemic acid by a microbial esterase: Cloning, nucleotide sequence, and overexpression of the esterase gene of Arthrobacter globiformis in Escherichia coli. Appl. Environm. Microbiol. 61, 3208 – 3215. Nishizawa, M., Gomi, H., Kishimoto, F., 1993. Purification and some properties of carboxylesterase from Arthrobacter globiformis; Stereoselective hydrolysis of ethyl chrysanthemate. Biosci. Biotech. Biochem. 57, 594 – 598. Pelletier, I., Altenbuchner, J., 1995. A bacterial esterase is homologous with non-haem haloperoxidases and displays brominating activity. Microbiology 141, 459 – 468. Toone, E.J., Werth, M.J., Jones, J.B., 1990. Active-site model for interpreting and predicting the specificity of pig liver esterase. J. Am. Chem. Soc. 112, 4946 – 4952. Volff, J.N., Eichenseer, C., Viell, P., Piendl, W., Altenbuchner, J., 1996. Nucleotide sequence and role in DNA amplification of the direct repeats composing the amplifiable element AUD1 of Streptomyces li6idans 66. Mol. Microbiol. 21, 1037 – 1047. Weissfloch, A.N.E., Kazlauskas, R.J., 1995. Enantiopreference of lipase from Pseudomonas cepacia toward primary alcohols. J. Org. Chem. 60, 6959 – 6969. Wong, C.H., Whitesides, G.M., 1994. Enzymes in Synthetic Organic Chemistry. Pergamon Press, Oxford. Zadel, G., Rieger, G., Breitmaier, E., 1991. Enantioselective terpene syntheses by Diels-Alder reaction of l-(1-arylalkoxy)-2-methyl-1,3-butadiene with isoprene. Liebigs Ann. Chem. 1343 – 1346. Zhu, L.M., Tedford, M.C., 1990. Applications of pig liver esterases (PLE) in asymmetric synthesis. Tetrahedron 46, 6587 – 6611.
.