Synthesis of (S)-1-(4-hydroxyphenyl)alcohols by eugenol dehydrogenase from Pseudomonas fluorescens E118

Article abstract of DOI:10.1016/S0957-4166(99)00164-0

(S)-1-(4-Hydroxyphenyl)ethanol and (S)-1-(4-hydroxyphenyl)propanol were synthesized with enantiomeric excesses of 96.6% and 95.2%, respectively, from the corresponding 4-alkylphenols by eugenol dehydrogenase from Pseudomonas fluorescens E118. The enantios

Full text of DOI:10.1016/S0957-4166(99)00164-0

Pergamon
Tetrahedron: Asymmetry 10 (1999) 1627–1630
Synthesis of (S)-1-(4-hydroxyphenyl)alcohols by eugenol
dehydrogenase from Pseudomonas fluorescens E118
Marco Wieser,a Hirotaka Furukawa,b Hiroshi Morita,b Toyokazu Yoshida a and
Toru Nagasawa a,∗
aDepartment of Biomolecular Science, Faculty of Engineering, Gifu University, Gifu 501-1193, Japan
bChisso Corporation, Yokohama Research Center, Kanazawa-ku, Yokohama 236-8605, Japan
Received 5 March 1999; accepted 27 April 1999
Abstract
(S)-1-(4-Hydroxyphenyl)ethanol and (S)-1-(4-hydroxyphenyl)propanol were synthesized with enantiomeric ex-
cesses of 96.6% and 95.2%, respectively, from the corresponding 4-alkylphenols by eugenol dehydrogenase from
Pseudomonas fluorescens E118. The enantioselectivity of the enzyme was shown to be pH-dependent. © 1999
Elsevier Science Ltd. All rights reserved.
1. Introduction
(S)-1-(4-Hydroxyphenyl)alcohols [(S)-4-(1-hydroxyalkyl)phenols] 2 are employed in the synthesis of
anti-ulcer1 and anti-inflammatory2 agents and liquid crystal materials.3 The chemical access to enan-
tiomerically enriched forms of 1-(4-hydroxyphenyl)alkanols is based on either a classical preferential
crystallization of the phthalate ester salts after protection of the phenolic moiety as a formaldehyde
acetal,4 or the treatment of the O-protected 4-hydroxyphenyl alkyl ketones with the enantiomers of
chlorodiisopinocampheylborane followed by deprotection.5 However, due to the necessity of chiral
additives and auxiliaries, and a protection and deprotection step, enzymatic one-step preparations under
mild reaction conditions have attracted increased interest. So far, two enzymatic syntheses of chiral 1-
(4-hydroxyphenyl)alcohols6–8 using 4-alkylphenol methylhydroxylases from two Pseudomonas putida
species have been described, however, these enzymes have the tendency to dehydrogenate the alcohol
further to the corresponding ketones, and in one case additionally 4-vinylphenol was formed as a
by-product,6 leading to the requirement of additional separation steps during product isolation. We
describe here a new enzymatic method for the synthesis of (S)-1-(4-hydroxyphenyl)alcohols using
eugenol dehydrogenase from Pseudomonas fluorescens E118, which has been previously purified and
∗
Corresponding author. Tel: (+81) 58 293-2647; fax: (+81) 58 293-2647; e-mail: tonagasa@apchem.gifu-u.ac.jp
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00164-0

1628
M. Wieser et al. / Tetrahedron: Asymmetry 10 (1999) 1627–1630
characterized.9 This enzyme has been shown to catalyze the dehydrogenation of a wide range of 4-
hydroxybenzylic structures including the conversion of 4-alkylphenols (1) to the corresponding 1-(4-
hydroxyphenyl)alcohols after addition of an artificial electron acceptor such as phenazine methosulfate,
K3Fe(CN)6, 2,6-dichlorophenol-indophenol or cytochrome c as the enzyme cofactor. Since the hydroxyl-
ation of prochiral 4-alkylphenols to 1-(4-hydroxyphenyl)alcohols introduces an asymmetric carbon in the
alkyl side chain, we have examined the (pro-)stereoselectivity of this enzyme.
2. Results and discussion
Eugenol dehydrogenase from Pseudomonas fluorescens E118 converted a number of 4-alkylphenols
to give products whose absolute configurations were (S), as shown by HPLC using a Shimadzu LC-6A
with a Chiracel OB column (4.6×250 mm, Daicel, Tokyo, Japan) and n-hexane:n-propanol (9:1, v/v)
as eluent at a flow rate of 0.2 ml/min and 30°C, monitored at 280 nm. The retention times of 1-(4-
hydroxyphenyl)ethanol were 23.7 min for the (S)-enantiomer and 27.0 min for the (R)-enantiomer, and
the (S)- and (R)-forms of 1-(4-hydroxyphenyl)propanol eluted after 41.5 min and 45.9 min, respectively.
The substrate specificity was determined in a photometric activity test in a quartz cuvette containing
1 µg of the enzyme purified as described,9 75 µM 2,6-dichlorophenol indophenol, 150 µM phenazine
methosulfate and 10 mM of different 4-alkylphenols in 1 ml 100 mM potassium phosphate, pH 7.0.
The reaction was followed by measuring the reduction of 2,6-dichlorophenol indophenol at 600 nm
using a Shimadzu UV 1200 spectrophotometer. Since the enzyme activity decreased with the increa-
sing length of the alkyl side chain or the introduction of branched alkyl chains in the substrates (4-
ethylphenol 100% relative activity, 4-n-propylphenol 96.6%, 4-n-butylphenol 84.6%, 4-n-hexylphenol
48.7%, 4-n-octylphenol 17.9%, 4-isopropylphenol 7.7%, 4-sec-butylphenol 15.4%), we have confined
our bioconversion studies here to the conversion of 4-ethylphenol and 4-n-propylphenol.
For each of the bioconversions of 40 mM and 75 mM 4-ethylphenol and 60 mM 4-propylphenol, 5
mg of the purified enzyme was used in a 3 h reaction in 20 ml 100 mM potassium phosphate buffer,
pH 7.0, containing 10 mM phenazine methosulfate. Bioconversions were monitored by HPLC using an
ODS C18 column (4.6×150, MS Instruments, Tokyo, Japan) and a methanol:H2O:acetic acid (45:52:3,
v/v/v) eluent at 1 ml/min, detected at 280 nm, and the enantiomeric excesses (e.e.) were determined
from the ratio of the peak areas obtained by the above-mentioned chiral stationary phase. 4-Ethylphenol
(40 mM) was converted quantitatively in a batch reaction to 40 mM (S)-1-(4-hydroxyphenyl)ethanol
(Fig. 1A) with an enantiomeric excess of 96.6%. Due to a relatively high toxicity and inhibition effect
on the enzyme, 4-propylphenol was added in three portionwise feeding steps of 20 mM each after 0, 1
and 2 h, finally leading to 60 mM (S)-1-(4-hydroxyphenyl)propanol (Fig. 1 B) with 95.2% e.e. Using a
similar fed-batch approach for 4-ethylphenol added in three steps of 25 mM each, finally 74 mM (S)-1-(4-
hydroxyphenyl)ethanol were accumulated with 96.6% e.e. In all of these bioconversions, neither ketones
nor 4-vinylphenol, the latter of which could have only been formed in the reaction of 4-ethylphenol, were
detected as by-products.
1-(4-Hydroxyphenyl)ethanol and 1-(4-hydroxyphenyl)propanol were isolated by ethyl acetate ex-
traction, evaporation and silica gel chromatography (Wakogel C300, Wako, Osaka, Japan) using ben-

M. Wieser et al. / Tetrahedron: Asymmetry 10 (1999) 1627–1630
1629
Figure 1. Time course of the conversion of 4-ethylphenol (A) and 4-propylphenol (B) by eugenol dehydrogenase from
Pseudomonas fluorescens E118. The reactions were carried out at 30°C in 20 ml 100 mM potassium phosphate buffer, pH
7.0, containing 10 mM phenazine methosulfate and 5 mg of the purified enzyme, and followed by HPLC. 4-Ethylphenol was
converted in a batch reaction, and 4-propylphenol in a fed-batch reaction
zene:methanol 95:5 (v/v) as eluent with yields of 63% and 66%, respectively. The product identity was
1
confirmed by GC–MS and H NMR. GC–MS spectra were recorded in a Trio-1 mass spectrometer
(Raleigh, USA) connected with a 5890 Hewlett–Packard gas chromatograph (Palo Alto, USA) plus
DB-1 capillary column (J&W Scientific, Tokyo, Japan) using helium as carrier gas and a temperature
program of 1 min at 50°C and 50–250°C at 15°C/min. NMR spectra were obtained from a Bruker WM-
360 high field NMR spectrometer (Billerica, USA) with methanol-d4 as solvent and tetramethylsilane as
internal standard. The mass spectrum of the ethanol product showed peaks at m/z 138 (M+, 22.2% relative
abundance), 123 (84.4%), 121 (16.7%), 120 (100%), 95 (55.6%), 94 (15.5%), 91 (71.1%), 77 (48.9%),
65 (37.8%) and 63 (17.8%) consistent with the MS pattern of authentic 1-(4-hydroxyphenyl)ethanol.6
The m/z values of the (S)-propanol were 152 (M+, 8.9%), 134 (33.3%), 133 (24.4%), 123 (100%), 107
(13.3%), 94 (44.4%), 91 (11.1%) and 77 (48.9%). The GC–MS data indicated the incorporation of an
1
oxygen in the aliphatic carbon chain at the carbon adjacent to the aromatic ring. H NMR spectra of
both products indicated a methine proton adjacent to the aromatic ring and methylene protons at the
α position of the aliphatic carbon chain. The chemical shifts δ for the (S)-ethanol product were 1.39
(d, 3, CH3), 4.72 (q, 1, C–H), 6.74 and 7.18 ppm (2 d, 4, Ar–H), and the protons of the product of
4-propylphenol appeared at chemical shifts δ of 0.85 (d, 3, CH3), 1.76 and 1.81 (2 d, 2, CH2), 4.41
(q, 1, C–H), 6.72 and 7.13 ppm (2 d, 4, Ar–H). Besides identical molecule ion masses, the MS and
1H NMR spectra of reference compounds bearing the alcohol group at the distal aliphatic carbon, 2-(4-
hydroxyphenyl)-1-ethanol and 3-(4-hydroxyphenyl)-1-propanol, clearly differed from our bioconversion
products, excluding the possibility of the hydroxylation of the distal aliphatic carbon. The specific
20
rotation [α]D of the isolated (S)-ethanol product, measured in a Jasco DIP-1000 digital polarimeter
(Easton, USA), was −44.8 (c 1.74, ethanol) in agreement with the value of the almost enantiomerically
20
20
pure (S)-form [α]D −47.5 (c 4.98, ethanol).6 The specific rotation [α]D of the isolated (S)-propanol
was −36.7 (c 1.35, ethanol).
The enantiomeric excess of the products depended on the pH of the bioconversion mixture with the
highest e.e. values reached at pH 7.0 (Fig. 2). This is likely due to the pH-dependent enantioselectivity
of the enzyme, because the enantiomeric excesses of the products were not affected by acid and base
treatment. A pH-dependent stereoselectivity has not been reported so far for similar 4-alkylphenol
methylhydroxylases.6–8 The enantioselectivities of eugenol dehydrogenase were in the range of the

1630
M. Wieser et al. / Tetrahedron: Asymmetry 10 (1999) 1627–1630
Figure 2. Effect of the pH on the enantioselectivity of eugenol dehydrogenase from Pseudomonas fluorescens E118. The reaction
mixture was composed as described in the legend of Fig. 1 except for using potassium phosphate buffer of different pH. The
reaction was stopped after 3 h, and e.e. values were determined by chiral HPLC
analogous bioconversions catalyzed by 4-alkylphenol methylhydroxylases, which resulted in 31–94%
e.e. for the (S)-alcohol6,7 and 92–98% e.e. for the (R)-alcohol,8 depending on the kind of alkyl rest. On
the other hand, the 4-alkylphenol methylhydroxylases were only enantioselective with cytochrome c, but
almost not with phenazine methosulfate, leading to a nearly racemic mixture of the alcohol products
when phenazine methosulfate was used. This contrasts with eugenol dehydrogenase, which was highly
enantioselective with phenazine methosulfate, and the enantioselectivity of which did not depend on
the kind of electron acceptor as tested with phenazine methosulfate, K3Fe(CN)6, 2,6-dichlorophenol-
indophenol and cytochrome c. The asymmetric bioconversion described here has advantages compared
to previous enzymatic preparations with regard to productivity, yield and absence of by-products.
Acknowledgements
M. W. was supported by the Japanese Society for the Promotion of Science.
References
1. Munson, H. R.; Boswell, R. F. US Patent, 5,171,753, 1992.
2. Batt, D. G.; Wright, S. W. PTC International Patent, WO 91/11,445, 1991.
3. Azumai, T.; Kurimoto, I.; Toda, S.; Minamii, M. Japan Kokai Patent, 3/264,566, 1991.
4. Corey, E. J.; Gras, J. L.; Ulrich, P. Tetrahedron Lett. 1976, 11, 809–812.
5. Everhart, E. T.; Craig, J. C. J. Chem. Soc., Perkin Trans. 1 1991, 1701–1707.
6. McIntire, W. S.; Hopper, D. J.; Craig, J. C.; Everhart, E. T.; Webster, R. V.; Causer, M. J.; Singer, T. P. Biochem. J. 1984,
224, 617–621.
7. McIntire, W. S.; Bohmont, M. J. In Flavins and Flavoproteins; Edmondson, D. E.; McCormick, D. B., Eds; Unusual
properties of the flavocytochrome p-cresol methylhydroxylase. Walter de Gruyter: Berlin, 1987; pp. 677–689.
8. Reeve, C. D.; Carver, M. A.; Hopper, D. J. Biochem. J. 1990, 269, 815–819.
9. Furukawa, H.; Wieser, M.; Morita, H.; Sugio, T.; Nagasawa, T. Arch. Microbiol. 1998, 171, 37–43.