Bacterial monooxygenase mediated preparation of nonracemic chiral oxiranes: Study of the effects of substituent nature and position

Article abstract of DOI:10.1016/j.tetasy.2004.04.005

Monooxygenation of styrene derivatives using recombinant E. coli biocatalyst is an efficient way to prepare the corresponding oxiranes. The electronic and geometric effects of the ring substituents are described and show the relaxed specificity of the enzyme and its high stereoselectivity.

Full text of DOI:10.1016/j.tetasy.2004.04.005

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 1603–1606
Bacterial monooxygenase mediated preparation of nonracemic
chiral oxiranes: study of the effects of substituent nature
and position
Silvana Bernasconi,a Fulvia Orsini,a Guido Selloa,* and Patrizia Di Gennarob
aDipartimento di Chimica Organica e Industriale, Universitꢀa degli Studi di Milano, via Venenzian 21, 20133 Milano, Italy
bDipartimento di Scienze dell’ Ambiente e del Territorio, Universitꢀa di Milano-Bicocca, P.za della Scienza 1, 20126 Milano, Italy
Received 8 March 2004; accepted 1 April 2004
Available online 27 April 2004
Abstract—Monooxygenation of styrene derivatives using recombinant E. coli biocatalyst is an efficient way to prepare the corre-
sponding oxiranes. The electronic and geometric effects of the ring substituents are described and show the relaxed specificity of the
enzyme and its high stereoselectivity.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
study and with the support of the results of other groups
we have reached various conclusions concerning the
bioconversions. The first, and most important, factor
affecting the reaction is definitely the need to perform
the experiment under two-phase conditions. In fact, the
substrates are often volatile and toxic to the cell, the
products are unstable, volatile, and toxic. Consequently,
effort has been spent on the search for good (nontoxic,
stable, water immiscible, etc.) organic phases8 and for
optimal reaction conditions (phase ratio, substrate
concentration, stir speed, etc.).
The importance of homochiral oxiranes as building
block in the preparation of synthons or auxiliaries is
well known.1 Recently, the biocatalytic approach has
gained more and more importance as an alternative to
chemical catalysis due to its environmental compatibil-
ity and low cost.2 There are two main biocatalytic
approaches to homochiral oxirane production: racemate
resolution by, mainly, hydrolases,3 and direct double
bond oxygenation by monooxygenases.4–6 The enzymes
of this last class are difficult to isolate and purify, and
the cofactor [NAD(P)H] regeneration is difficult to
optimize. Nevertheless, at least one group has developed
sound processes for oxirane production at good scale.2
In contrast, not much work has been devoted to the
potential of enzyme recognition. This is in part due to
the previously cited difficulty of the enzyme isolation; in
part due to the need to work with reactive compounds
(oxiranes), whose purification and characterization can
present problems.
Herein we describe our new results concerning enzyme
recognition towards some substituted styrenes.
2. Results
In the course of our research, we have used a
recombinant E. coli JM109 (pTAB19)9 biocatalyst that
was realized by inserting the styrene monooxygenase
from P. fluorescens ST, a strain selected from the soil for
its ability to grow on styrene.10 This biocatalyst was
used for the transformation of a preliminary set of sty-
rene derivatives and showed a good recognition selec-
tivity.11 The initial compound set was randomly chosen
to test different substrate classes: ring substituted sty-
renes, chain substituted styrenes, cyclic compounds.
Herein we have analyzed four complete sets of ring
monosubstituted styrenes to study their electronic and
Our group has been involved in the use of whole cell
bioconversions for many years and, despite the
encountered difficulties, we have studied the biotrans-
formation of several compounds.7 In the course of this
* Corresponding author: Tel.: +39-02-2663469; fax: +39-02-2664874;
e-mail: guido.sello@unimi.it
0957-4166/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.04.005

1604
S. Bernasconi et al. / Tetrahedron: Asymmetry 15 (2004) 1603–1606
position effects, which were chosen as representative of
alkylated, halogenated, electron donating, and electron
withdrawing groups. The reactions were performed
under two principal reaction conditions: in the first the
substrates are dissolved into isooctane, in the second
they are dissolved into isooctane–isopropyl ether mix-
tures; in all cases the organic phases contained either
10 g/L of substrate, or a saturated lower concentration,
in the perspective of providing a sufficient but nontoxic
amount of substrate. This new experimental approach
produced two interesting effects: first, we noted an
increase in the efficiency of the bioconversion also for
the already studied compounds (styrene, 4-Me styrene);
second, we can transform more substrates. All the
experiments were performed with a comparison bio-
conversion of styrene; in this way, we wanted to render
as much as possible the results independent of the bio-
catalyst current activity. As a consequence, the enzyme
recognition is expressed as a percentage of the perfor-
mance on styrene. The results are presented in Table 1.
It is clear that the recognition is influenced by both the
substituent position and nature.
The first very interesting remark concerns the unex-
pected monotonic increase of the yield along the series
ortho, meta, para. In contrast with the common elec-
Table 1. Substrates, products, and relative activity, of the bioconversion of styrene derivatives
Substrate
Product
Relative
activitya (%)
Eeb
Substrate
Product
Relative
activitya (%)
Eeb
O
46
56
138 25%
diolc
>95
O
O
O
O
H
H
1a
3b
O
101 (142)
>95
222
n.d.c
O
O
OH
H
H
1b
OH
3c
130 (153)
>95
––
0
CN
1c
4a
O
60
>95
>95
55
>95
O
NC
NC
NC
Cl
H
H
Cl
4b
4c
2a
107 (100)
––
0
O
O
Cl
Cl
Cl
Cl
H
H
2b
H
O
H
145
55
>95
43 (184)
>95
5
2c
n.d.c
OH
O
OH
O
3a
a Relative activity ¼ [specific activity of substrate (i)]/[specific activity of styrene] · 100; Specific activity of (i) ¼ mmoles of products formed in 1 h by
1 g of cells (dry cell weight) in 1 L of broth, as calculated from the correlation line of the first conversion hours. Number in parentheses refer to the
results reported by Hollmann et al.12
b Enantiomeric excess is measured by chiral GLC.
c n.d. ¼ not determined; the racemic diols derive from the chemical hydrolysis of the corresponding epoxides and are not representative of their
enantiomeric excess.

S. Bernasconi et al. / Tetrahedron: Asymmetry 15 (2004) 1603–1606
1605
tronic effects ortho-substituted compounds do not show
an influence greater than the corresponding meta-
substituted compounds. This result is logically corre-
lated to the simultaneous geometry effect, that seems to
be stronger in the ortho position, lowering the corre-
sponding yields. The yield ratio is roughly 1:2 compar-
ing ortho- and meta-substituted compounds, whilst
meta- and para-substituted substrates show a smaller
ratio (ꢀ1:1.3). The position effect is so strong that when
considering the expected electronic effect (OMe > Me ꢀ
Cl > CN) in the ortho-substituted compounds we can
note a quasi-null influence; the only exception being the
CN group that is deactivating enough to completely
stop the transformation.
addition, we detected the production of small amounts of
reduced compounds (2- and4-OMe 2-phenyl ethanol),
due to a reductive activity of the cells.
The third point concerns the enantiomeric excesses of
the products. As in most of our previous experiments we
expected to isolate nearly enantiopure oxiranes; this was
the case for all compounds except o-methyl styrene. The
corresponding oxirane shows a 75:25 enantiomer ratio,
which represents the lowest excess we have ever deter-
mined: We do not have a clear rationalization for this.
Excluding the possibility of a nonenzymatic racemiza-
tion of the compound, we can think only of a relatively
low enzyme facial recognition. In order to get a further
piece of data concerning the problem of o-substituted
substrates, we performed the bioconversion using 1,2-
dihydronaphthalene as a substrate representative of
alkyl ortho substituents with low steric hindrance, but
with completely removed rotation freedom. The result
confirms the lower transformation rate, in agreement
with all the previous data, but, in contrast to the methyl
substituent, the ee was high, as usual.
Looking at the meta-substituted compounds we observe
a more expected result; the electronic effect is not very
strong and all the compounds are transformed in similar
amounts. Again the CN group depresses the reactivity,
even if, in this case, we can isolate the bioconversion
product. Finally, the para-substituted compounds show
the expected influence of the group electronics, with a
clear preference for the most electron donating group
(OMe) (ratio 1.5–1.7). We can also include in this topic
the surprisingly higher yield presented by the para-
substituted compounds in comparison to styrene. In
many cases, the ‘native’ substrate is transformed by
enzymes at a higher rate than the other substrates; this is
not true in our case, all the para compounds, CN-
substituted excluded, give very good conversion rates
(from 1.3 to 2.2 higher than styrene). Even meta-
substituted compounds show rates comparable to sty-
rene. This fact is in disagreement with both our previous
experiments11 and the results reported in the literature;2
nevertheless, all the present experiments give this result.
It is interesting to note that the obtained activity ratios
are in partial agreement with the initial rate ratio re-
ported by Hollmann et al.12 If we consider that they
were using an isolated environment (enzyme and metal
recycling of FAD cofactor) we can speculate that the
reaction conditions that we are presently using are
independent of the transport mechanism; of course, this
fact requires an experimental proof.
3. Conclusion
Using whole cell transformations of styrenes by the
cloned SMO (styrene mono oxygenase) from P. fluo-
rescens ST we could prepare several chiral epoxides
with different enantiomeric excess, most of them prac-
tically enantiopure. In addition, we could partly eluci-
date the enzyme recognition capability, obtaining
interesting and unexpected results. Work is in progress
to determine the correct conditions to prepare also
those oxiranes that showed too high reactivity towards
hydrolysis.
4. Experimental
4.1. Analytical methods
Reaction progress was followed by gas chromatography
(DANI 86.10 gas chromatograph with FID) on a
Chrompack Cp-Sil 8CB column (T ¼ 100 ꢁ 150 ꢁC at
15 ꢁC min, splitless injection) with the appropriate
internal standard (dodecane or hexadecane). At inter-
vals, 2 mL samples were taken from the reaction emul-
sion. The two phases were separated by filtration. The
water phase was followed by HPLC on a Merck/Hitachi
(L-6200) system connected to a UV-detector set at
230 nm on a (C18 Hibar Lichrosorb 50334, 5 lm, 25 cm)
column with 50:50 CH3CN/water. The absolute config-
uration of biocatalytically prepared (S)-styrene oxide
could be proven via comparison with commercially
available, enantiopure (S)-styrene oxide (Aldrich).
Based on the analogous chromatographic behavior of
racemic mixtures, we assume the (S)-configuration for
all epoxides. Enantiomeric excesses were measured using
a Chrompack ChiralDex-CB column by comparison to
synthetic racemic mixture. Optical rotations were mea-
sured with a Perkin Elmer 341 polarimeter. NMR
spectra were recorded on a Bruker AC 200 (1H NMR at
A second interesting point concerns the isolated products
in the case of activated aromatic rings (MeO-substituted
on the ortho and para position). Here, the only isolated
product is the 1,2-diol that clearly derives from the
spontaneous hydrolysis of the corresponding epoxide.
This is confirmed by the complete lack of enantiomeric
excess, by a control experiment performed on the iso-
lated oxiranes in the absence of the biocatalyst, and by
the great difficulty that we experienced also in the iso-
lation of the chemically prepared epoxides (reaction that
is performed in the presence of water). In the performed
experiments we operated using an isooctane/isopropyl
ether mixture that could solubilize good substrate
amounts and that, we hoped, could extract most of the
product from the water phase. This unfortunately did
not happen and the product hydrolyzed. Nevertheless,
the production rate is still representative. Having learnt
of this possibility, we always checked also the water
phase looking for the existence of possible diols. In

1606
S. Bernasconi et al. / Tetrahedron: Asymmetry 15 (2004) 1603–1606
200 MHz). All signals are expressed as ppm down field
from tetramethylsilane. All the compounds show spectra
in agreement with the literature data.
Acknowledgements
We acknowledge partial funding by MIUR under con-
tract (Cofin 2002, protocol no.: 200258141-007):
‘Composti ossigenati ottenuti per biotrasformazione:
ottimizzazione delle condizioni di bioconversione, iso-
lamento, caratterizzazione e applicazioni sintetiche’.
4.2. Biocatalyst preparation and bioconversion procedure
Biocatalyst E. coli JM109 (pTAB19) was prepared by
adding 1 mL of an overnight LB culture in 100 mL M9
medium containing: glucose 10 mM; thiamine 0.05 mM;
References and notes
kanamycin 50 lg/mL; IPTG (isopropyl-b-D-thiogalacto-
pyranoside) 1 mM as inducer; incubated overnight on a
shaker at 30 ꢁC. After the growth, OD 1.2–2.0 (k
600 nm), the cells were separated by centrifugation
(10,000 rpm, 4 ꢁC) and added to 70 mL M9 medium
containing glucose 10 mM on a shaker at 30 ꢁC; the
bioconversion was started by adding the substrate
(concentration of 10 g/L in the organic phase) dissolved
into 30 mL of isooctane (or isooctane/isopropyl ether 9:1
mixture when appropriate). The transformation was
carried out at 30 ꢁC.
1. Kroutil, W.; Faber, K. In Stereoselective Biocatalysis;
Patel, R. N., Ed.; Marcel Dekker: New York, 2000; pp
205–237.
2. Schmid, A.; Hofstetter, K.; Feiten, H.-J.; Hollmann, F.;
Witholt, B. Adv. Synth. Catal. 2001, 343, 732–737, and
references cited therein.
3. Mateo, C.; Archelas, A.; Fernandez-Lafuente, R.; Guisan,
J. M.; Furstoss, R. Org. Biomol. Chem. 2003, 1, 2739–2743.
4. Abraham, W.-R. In Stereoselective Biocatalysis; Patel, R.
N., Ed.; Marcel Dekker: New York, 2000; pp 181–203.
5. Flitsch, S.; Grogan, G.; Ashcroft, D. In Enzyme Catalysis
in Organic Synthesis; Drauz, K., Waldmann, H., Eds.; 2nd
ed.; Wiley-VCH GmbH: Weinheim, Germany, 2002; pp
1065–1108.
4.3. General procedure for chemical preparation of
alkenyl substrates
6. (a) Archelas, A.; Furstoss, R. Ann. Rev. Microbiol. 1997,
51, 491–525; (b) Archelas, A.; Furstoss, R. Top. Curr.
Chem. 1999, 200, 159–191.
7. Sello, G.; Orsini, F. Mini-Rev. Org. Chem. 2004, 1, 77–92
(Chem. Abstr.: 2004-115913).
Methyl-triphenylphosphonium
iodide
(20 mmol),
potassium carbonate (25 mmol), the aldehyde
(20 mmol), were added to 20 mL of dioxane containing a
small amount of water; the solution has been refluxed
under agitation for 5–7 h, or until the substrate disap-
pears. The mixture was then filtered to eliminate the
salts, dried on Na2SO4 and evaporated at reduced
pressure very cautiously to avoid material loss. The
products were purified by silica gel chromatography to
eliminate triphenylphosphine oxide and used in the
bioconversions.
8. (a) Deziel, E.; Comeau, Y.; Villemur, R. Biodegradation
1999, 10, 219–233; (b) Guieysse, B.; Cirne, M. d. D. T. G.;
Mattiasson, B. Appl. Microbiol. Biotechnol. 2001, 56, 796–
802; (c) Daugulis, A. J. Curr. Opinion Biotechnol. 1997, 8,
169–174; (d) Favre-Bulle, O.; Witholt, B. Enzyme Microb.
Technol. 1992, 14, 931–937; (e) Birman, I.; Alexander, M.
Appl. Microbiol. Biotechnol. 1996, 45, 267–272; (f) Collins,
L. D.; Daugulis, A. J. Biotechnol. Bioengineer. 1997, 55,
155–162; (g) Collins, L. D.; Daugulis, A. J. Appl.
Microbiol. Biotechnol. 1997, 48, 18–22; (h) Ortega-Calvo,
J.-J.; Martin, A. Appl. Environ. Microbiol. 1994, 60, 2643–
2646; (i) Osswald, P.; Baveye, P.; Block, J. C. Biodegra-
dation 1996, 7, 297–302; (j) Panke, S.; Held, M.; Wubbolts,
M. G.; Witholt, B.; Schmid, A. Biotechnol. Bioengineer.
2002, 80, 33–41; (k) Quintana, M. G.; Dalton, H. Enzyme
Microb. Technol. 1999, 24, 232–236.
9. The biocatalyst has been prepared in our laboratories (see
Ref. Di Gennaro, P.; Colmegna, A.; Galli, E.; Sello, G.;
Pelizzoni, F.; Bestetti, G. Appl. Environ. Microbiol. 1999,
65, 2794) and is available upon request from the authors.
10. Beltrametti, F.; Marconi, A. M.; Bestetti, G.; Colombo,
C.; Galli, E.; Ruzzi, M.; Zennaro, E. Appl. Environ.
Microbiol. 1997, 63, 2232–2239.
4.4. General procedure for chemical preparation of
racemic epoxides
The olefin (2 mmol) in CH2Cl2 (10 mL) was mixed with
an equal amount of water containing NaHCO3 (1 g); to
this solution was cautiously added 3-chloroperbenzoic
acid (2.2 mmol). The reaction mixture was stirred at rt
for 2.5 h, or until the substrate disappears. Afterwards it
was washed with aqueous Na2SO3 (1.3 g in 10 mL) for
20 min; the aqueous phase was then extracted with
CH2Cl2 (2 · 10 mL). The organic phases were washed
with NaHCO3 (2 · 25 mL) and with water. The CH2Cl2
phase was dried over anhydrous MgSO4 and evaporated
at reduced pressure.
11. Bernasconi, S.; Orsini, F.; Sello, G.; Colmegna, A.; Galli,
E.; Bestetti, G. Tetrahedron Lett. 2000, 41, 9157–9161.
12. Hollmann, F.; Lin, P.-C.; Witholt, B.; Schmid, A. J. Am.
Chem. Soc. 2003, 125, 8209–8217.