Multienzymatic preparation of (-)-[3-(oxiran-2-yl)phenyl]methanol and (-)-3-(oxiran-2-yl)benzoic acid

Article abstract of DOI:10.1016/j.tetasy.2009.03.023

The cascade use of enzymatic activities allows for the preparation of enantiomerically pure epoxides. In particular, using whole-cell biocatalysts we can prepare both (-)-[3-(oxiran-2-yl)phenyl]methanol and (-)-3-(oxiran-2-yl)benzoic acid in one-pot, two

Full text of DOI:10.1016/j.tetasy.2009.03.023

Tetrahedron: Asymmetry 20 (2009) 563–565
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Multienzymatic preparation of (ꢀ)-[3-(oxiran-2-yl)phenyl]methanol
and (ꢀ)-3-(oxiran-2-yl)benzoic acid
a
a
b
Guido Sello a, , Silvana Bernasconi , Fulvia Orsini , Patrizia Di Gennaro
*
a Dipartimento di Chimica Organica e Industriale, Universita’ degli Studi di Milano, via Venezian 21, 20133 Milano, Italy
b Dipartimento di Scienze dell’Ambiente e del Territorio, Universita’ di Milano-Bicocca, p.za della Scienza 1, 20126 Milano, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The cascade use of enzymatic activities allows for the preparation of enantiomerically pure epoxides. In
particular, using whole-cell biocatalysts we can prepare both (ꢀ)-[3-(oxiran-2-yl)phenyl]methanol and
(ꢀ)-3-(oxiran-2-yl)benzoic acid in one-pot, two or three steps procedure. The yield is quantitative and
enantiomeric purity greater than 95%. The selected biocatalysts contain a styrene monoxygenase from
Pseudomonas fluorescens ST and a naphthalene dihydrodiol dehydrogenase from P. fluorescens N3, cloned
and expressed in Escherichia coli.
Received 5 November 2008
Revised 16 March 2009
Accepted 20 March 2009
Available online 22 April 2009
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Another enzymatic activity is present in our biocatalysts that
introduce a new factor of interest. As already described10 Esche-
In the framework of our ongoing interest in biocatalysis we
have studied the possibility of preparing enantiomerically pure
epoxides by multienzymatic synthesis.1–7 The enzyme activities
were selected from our biocatalyst pool to perform the transfor-
mation of a styrene derivative into interesting new compounds.
The substrate chosen was 3-vinyl benzaldehyde and is commer-
cially available. Our intention was the preparation of 3-(oxiran-
2-yl)benzoic acid 4, in good yield and high enantiomeric purity.
Moreover, we wanted to prepare this compound using only
enzymatic activities. Both enzymes have been the subject of
previous studies and had demonstrated high selectivity and
good performance.8,9 However, we knew that the styrene mon-
oxygenase (SMO) cannot transform acid derivatives; therefore,
the reaction sequence should be: epoxidation and oxidation
(Scheme 1).
richia coli, our host organism, expresses a reductive activity that
chemoselectively transforms aldehydes into the corresponding
alcohols (Scheme 2).
E. coli
1
2
O
HO
Scheme 2. Reduction of 3-vinylbenzaldehyde into 3-(vinyl)phenylmethanol 2.
This activity is highly efficient and it is impossible to prevent its
action on any type of aldehyde that we tested. The consequence is
O
O
NDDH
SMO
1
4
O
O
OH
O
Scheme 1. Synthesis of (ꢀ)-3-(oxiran-2-yl)benzoic acid 4.
that the designed transformation can proceed in three successive
steps as reported in Scheme 3.
* Corresponding author. Tel.: +39 0250314107; fax: +39 0250314106.
E-mail address: guido.sello@unimi.it (G. Sello).
0957-4166/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2009.03.023

564
G. Sello et al. / Tetrahedron: Asymmetry 20 (2009) 563–565
O
O
E. coli
NDDH
SMO
1
2
3
4
O
HO
HO
OH
O
Scheme 3. Three-step preparation of (ꢀ)-3-(oxiran-2-yl)benzoic acid 4.
The reaction sequence will work only if the last transformation
occurs when the other two have finished.
2.5
2
1.5
1
2. Results and discussion
0.5
0
Initially, we separately tested all of the transformations, to as-
sess their performance on the chosen substrate. The results were
very good demonstrating that the enzymes correctly transform
each substrate into the corresponding product. Then, we checked
the reaction sequence using two different approaches whose fi-
nal result was identical. The first approach uses three biocata-
0
1
2
3
4
time (h)
vbal
epoxal
Figure 2. Synthesis of (ꢀ)-[3-(oxiran-2-yl)phenyl]methanol 3 (epoxal, j) and 3-
(vinyl)phenylmethanol 2 increase and decrease (vbal, ꢀ).
lysts in
a cascade reaction: E. coli JM109, E. coli JM109
(pTAB19), expressing monoxygenase activity, and E. coli JM109
(pVL2028), expressing alcohol dehydrogenase activity. The sec-
ond uses the last two biocatalysts, only. The first approach was
tested to demonstrate that it is possible to operate three succes-
sive transformations without a visible difference in comparison
to the two-step sequence.
2.5
2
The procedure developed allows for the preparation of both
1.5
1
(ꢀ)-[3-(oxiran-2-yl)phenyl]methanol
3
and (ꢀ)-3-(oxiran-2-
yl)benzoic acid 4, simply adding the third step if the acid is
the desired product. Both products are the sole compounds
present at the end of the respective reaction sequences; they
are easily recovered from the cultures and they both are enan-
tiomerically pure.
0.5
0
0
1
2
3
4
time (h)
In Figure 1 the transformation of 2-vinylbenzaldehyde 1 to 3-
(vinyl)phenylmethanol 2 by E. coli JM109 is shown. The reaction
is very fast and goes to completion in less than 30 min. As can be
observed from Figure 2, the same transformation occurs when
E. coli JM109 (pTAB19) is used. However, using this biocatalyst
the 3-(vinyl)phenylmethanol 2 is further transformed into the cor-
responding epoxide by the SMO activity. This reaction is slower
and needs 3 h to reach completion. Using compound 2 directly
with E. coli JM109 (pTAB19) we can observe the transformation
to the epoxide at a similar rate (Fig. 3).
In Figure 4 the HPLC analysis of the three-step transformation of
compound 1 into compound 4 is reported. In the figure it is possi-
ble to observe the initial fast transformation of 1 into 2 (30 min),
followed by the transformation of 2 into 3 (3 h). At this time the
second biocatalyst is added to the mixture and compound 3 begins
its transformation into compound 4. This last conversion is quite
vbal
epoxal
Figure 3. Synthesis of (ꢀ)-[3-(oxiran-2-yl)phenyl]methanol 3 (epoxal, j) and 3-
(vinyl)phenylmethanol 2 decrease (vbal, ꢀ).
2.5
2
1.5
1
0.5
0
10
0
2
4
6
8
time (h)
vba
vbal
epoxal
epoxac
Figure 4. Synthesis of (ꢀ)-3-(oxiran-2-yl)benzoic acid 4; (ꢀ)-[3-(oxiran-2-yl)
phenyl]methanol (epoxal, N) and 3-(vinyl)phenylmethanol increase and
2.5
2
1.5
1
3
2
decrease (vbal, j), 3-vinylbenzaldehyde 1 (vba, ꢀ) decrease and (ꢀ)-3-(oxiran-2-
yl)benzoic acid 4 (epoxac, d) increase.
0.5
0
slow; however, the observed rate is in complete agreement with
our previous experiments. NDDH is not a standard alcohol dehy-
drogenase and its activity towards primary and secondary alcohol
highly depends on the substrate structure. Nevertheless, at the end
of the transformation the final product is recovered. The complete
sequence can and should be optimised to improve the rate and the
time yield.
0
0.2
0.4
0.6
0.8
time (h)
vbal
Figure 1. Synthesis of 3-(vinyl)phenylmethanol 2 (vbal, ꢀ).

G. Sello et al. / Tetrahedron: Asymmetry 20 (2009) 563–565
565
dd, J = 5.5, 4.1 Hz), 3.9 (1H, dd, J = 4.1, 2.7 Hz), 4.7 (2H, s), 7.2–7.4
(4H, m); dC (75.5 MHz, CDCl3) 51.4 (t), 52.5 (d), 65.4 (t), 124 (d),
125.3 (d), 126.9 (d), 129 (d), 138.2 (s), 141.5 (s); m/z (EI) 150
(M+, 12), 149 (12), 119 (100), 91 (83); HRMS (EI): M+, found
3. Conclusion
This preliminary study demonstrates the power of multienzy-
matic synthesis, indicating that it should be possible to manage
many independent transformations without the need of intermedi-
ate isolation and purification. We are currently developing both
new sequences and newly regulated biocatalysts that can express
different activities at the desired time.
150.06660. C9H10O2 requires 150.06808; ½aꢁD ¼ ꢀ22:25 (c 4.8 mg,
CHCl3). Retention times in chiral GLC (t0 = 110 °C for 5 min,
tf = 180 °C, 3 °C/min, PHe = 0.8 atm) of both enantiomers: 41.0 and
41.8 min.
4.2.3. Preparation of 3-(oxiran-2-yl)benzoic acid 4
4. Experimental
At first, 2.27 mM 3-vinylbenzaldehyde 1, directly suspended in
the medium, was allowed to react with E. coli JM109 (pTAB19) (1 g/
L CDW) in 50 mL of M9 medium, containing glucose (0.2% w/v), at
30 °C. After 3 h an equal amount of E. coli JM109 (pVL2028) (1 g/L
CDW) was added and the reaction continued, at 30 °C. The reaction
was allowed to react for 24 h, then the cells were separated by cen-
trifugation (10,000 rpm, 4 °C); the surnatant was acidified using
HCl 1 M and extracted using AcOEt (three 30 mL portions). The or-
ganic phases were collected, washed with 2 ꢂ 50 mL portions of
the NaHCO3 water solution and with water to neutral pH, dried
over Na2SO4 and evaporated at reduced pressure. The crude prod-
uct (14.5 mg, 79%) only contained 3-(oxiran-2-yl)benzoic acid 4.
Oil; dH (200 MHz, CDCl3) 2.85 (1H, dd, J = 5.4, 2.6 Hz), 3.2 (1H, dd,
J = 5.2, 4.1 Hz), 3.9 (1H, dd, J = 4.1, 2.6 Hz), 7.5–7.6 (2H, m), 8.1–
8.2 (2H, m); dC (75.5 MHz, CDCl3) 51.4 (t), 52.0 (d), 127.6 (d),
128.6 (d), 130 (d), 130.8 (d), 134.7 (s), 138.5 (s), 171.3 (s); m/z
(EI) 164 (M+, 20), 163 (30), 148 (40), 119 (100), 105 (55), 91
(50); HRMS (EI): M+, found 164.04321. C9H10O2 requires
4.1. Analytical methods
Substrates and products were monitored by analysing the water
phase with HPLC, Hitachi-Merck, UV–visible detector at 220 nm,
reverse phase column C18 (Hibar LICHROSORB 50334, 10 lm,
25 cm), H2O/CH3CN 1:1 eluent, 1 mL/min flow, Hitachi D2500
integrator.
The absolute (S)-configuration of biocatalytically prepared (S)-
styrene oxide (proven via comparison with commercially available,
enantiopure (S)-styrene oxide (Aldrich)) was used as the reference
for all epoxides and the configuration accordingly presumed. Enan-
tiomeric excesses were measured using a Chrompack ChiralDex-CB
column.
lH NMR and 13C NMR spectra were obtained in CDCl3 (Merck)
using Bruker AC-200 instrument. All signals are expressed as
ppm down field from tetramethylsilane.
HRMS spectra were obtained using Autospec 246M VG Fisons
instrument.
164.04734; ½aꢁD ¼ ꢀ12:6 (c 4.4 mg, CH3OH). Retention times in chi-
ral GLC (t0 = 110 °C for 5 min, tf = 180 °C, 3 °C/min, PHe = 0.8 atm) of
Optical rotation was obtained in CHCl3 or CH3OH using JASCO P-
1030 polarimeter.
both enantiomers: 72.6–73.8 min.
Acknowledgements
4.2. Preparation procedures
We acknowledge partial funding by MIUR and Università degli
Studi di Milano (PRIN2007: Study of regulatory and catalytic sys-
tems for the development of bioconversion and biodegradation
processes).
4.2.1. Preparation of 3-(vinyl)phenylmethanol 2
At first, 2.27 mM 3-vinylbenzaldehyde 1, directly suspended in
the medium, was allowed to react with E. coli JM109 (1 g/L CDW) in
50 mL of M9 medium, containing glucose (0.2% w/v), at 30 °C. After
30 min the cells were separated by centrifugation (10,000 rpm,
4 °C); the surnatant was extracted using AcOEt (three 30 mL por-
tions), the organic phases were collected, dried over Na2SO4 and
evaporated at reduced pressure. The crude product (12 mg, 79%)
only contained 3-(vinyl)phenylmethanol 2. Oil; dH (200 MHz,
CDCl3) 4.7 (2H, s), 5.3 (1H, d, J = 10.8 Hz), 5.8 (1H, d, J = 17.5 Hz),
6.7 (1H, dd, J = 10.8, 17.5 Hz), 7.3–7.45 (4H, m).
References
1. Chen, G.; Fournier, R. L.; Varanasi, S. Enzyme Microbiol. Technol. 1997, 21, 491–
495.
2. Kihumbu, D.; Stillger, T.; Hummel, W.; Liese, A. Tetrahedron: Asymmetry 2002,
13, 1069–1072.
3. (a) Glueck, S. M.; Fabian, W. M. F.; Faber, K.; Mayer, S. F. Chem. Eur. J. 2004, 10,
3467–3478; (b) Larissegger-Schnell, B.; Kroutil, W.; Faber, K. Synlett 2005,
1936–1938.
4. Hecquet, L.; Bolte, J.; Demuynck, C. Tetrahedron 1996, 52, 8223–8232.
5. Guerard, C.; Alphand, V.; Archelas, A.; Demuynck, C.; Hecquet, L.; Furstoss, R.;
Bolte, J. Eur. J. Org. Chem 1999, 3399–3402.
6. Zimmermann, F. T.; Schneider, A.; Schoerken, U.; Sprenger, G. A.; Fessner, W.-D.
Tetrahedron: Asymmetry 1999, 10, 1643–1646.
7. Bestetti, G.; Di Gennaro, P.; Galli, E.; Leoni, B.; Pelizzoni, F.; Sello, G.; Bianchi, D.
Appl. Microbiol. Biotechnol. 1994, 40, 791–793.
8. Bernasconi, S.; Orsini, F.; Sello, G.; Di Gennaro, P. Tetrahedron: Asymmetry 2004,
15, 1603–1606.
9. Sello, G.; Bernasconi, S.; Orsini, F.; Mattavelli, P.; Di Gennaro, P.; Bestetti, G. J.
Mol. Catal. B: Enzym. 2008, 52–53, 67–73.
4.2.2. Preparation of [3-(oxiran-2-yl)phenyl]methanol 3
At first, 2.24 mM 3-(vinyl)phenylmethanol 2, directly sus-
pended in the medium, was allowed to react with E. coli JM109
(pTAB19) (1 g/L CDW) in 50 mL of M9 medium, containing glucose
(0.2% w/v), at 30 °C. After 3 h the cells were separated by centrifu-
gation (10,000 rpm, 4 °C); the surnatant was extracted using AcOEt
(three 30 mL portions), the organic phases were collected, dried
over Na2SO4 and evaporated at reduced pressure. The crude prod-
uct (14 mg, 83%) only contained [3-(oxiran-2-yl)phenyl]methanol
3. Oil; dH (200 MHz, CDCl3) 2.8 (1H, dd, J = 5.5, 2.7 Hz), 3.2 (1H,
10. Sello, G.; Orsini, F.; Bernasconi, S.; Di Gennaro, P. Molecules 2006, 11, 365–369.