Sequential deracemization of sulfoxides via whole-cell resolution and heterogeneous oxidation

Article abstract of DOI:10.1016/j.apcata.2012.06.050

A new concept for the synthesis of asymmetric sulfoxides exploiting sequential deracemization of racemic (rac) sulfoxides using a two-stage protocol had been developed. Enantio-pure sulfoxide was obtained using Escherichia coli (E. coli) cells to catalyze the reduction of (R)-sulfoxide to the thioether, and then using a heterogeneous Ta₂O₅-SiO₂ catalyst for catalyzing the oxidation of the ethers to the rac-sulfoxide. The performance of E. coli depended on the strain and the conditions under which the bacteria growth was carried out (e.g. source of carbon, presence of vitamins, and concentration of dimethyl sulfoxide, DMSO). Under optimized conditions, the E. coli cells performed the rac-sulfoxide resolution for different sufoxides leading to the best enantiomeric excess (ee) of around 62% (S)-MTSO (methyl tolyl sulfoxide) with 49% conversion of the rac-MTSO. The sequential deracemization process applied for rac-MTSO was cyclically performed leading to 97.5% ee of (S)-sulfoxide and a 56% yield in (S)-sulfoxide after three deracemization cycles of rac-MTSO.

Full text of DOI:10.1016/j.apcata.2012.06.050

Applied Catalysis A: General 441–442 (2012) 42–46
Contents lists available at SciVerse ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Sequential deracemization of sulfoxides via whole-cell resolution and
heterogeneous oxidation
Madalina Tudorachea, Simona Nicab, Emerich Barthab, Iulia Lupanc, Vasile I. Parvulescua,∗
a University of Bucharest, Department of Organic Chemistry, Biochemistry and Catalysis, Bd. Regina Elisabeta 4-12, Bucharest 030016, Romania
b Center of Organic Chemistry “C.D. Nenitzescu”, Bucharest, Romania
c Institute for Interdisciplinarity Experimental Research “Babes-Bolyai” University, Molecular Biology Center, Cluj, Romania
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 May 2012
Received in revised form 29 June 2012
Accepted 30 June 2012
Available online 21 July 2012
A new concept for the synthesis of asymmetric sulfoxides exploiting sequential deracemization of racemic
(rac) sulfoxides using a two-stage protocol had been developed. Enantio-pure sulfoxide was obtained
using Escherichia coli (E. coli) cells to catalyze the reduction of (R)-sulfoxide to the thioether, and then
using a heterogeneous Ta2O5–SiO2 catalyst for catalyzing the oxidation of the ethers to the rac-sulfoxide.
The performance of E. coli depended on the strain and the conditions under which the bacteria growth was
carried out (e.g. source of carbon, presence of vitamins, and concentration of dimethyl sulfoxide, DMSO).
Under optimized conditions, the E. coli cells performed the rac-sulfoxide resolution for different sufoxides
leading to the best enantiomeric excess (ee) of around 62% (S)-MTSO (methyl tolyl sulfoxide) with 49%
conversion of the rac-MTSO. The sequential deracemization process applied for rac-MTSO was cyclically
performed leading to 97.5% ee of (S)-sulfoxide and a 56% yield in (S)-sulfoxide after three deracemization
cycles of rac-MTSO.
Keywords:
Sequential deracemization
Racemic sulfoxides
Escherichia coli
Enantioselective reduction
Selective heterogeneous oxidation
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
During the last two decades, several chemical approaches have
been proposed for the synthesis of enantiopure sulfoxides [7–9].
Chiral sulfoxides belong to the class of chiral organosulfur
compounds widely used as important intermediates in organic syn-
thesis (e.g. asymmetric synthesis) [1] and as bioactive ingredients in
the pharmaceutical industry [2]. Their application in synthesis has
now become a well-established and reliable strategy (i.e. the chiral
groups are easy to introduce and remove, and give high asymmetric
induction in many reactions). To date chiral sulfoxides have been
investigated in a wide range of reactions such as the reduction of
␤-ketosulfoxides, the Michael addition to ␣,␤-unsaturated sulfox-
ides, the extension of carbon chains with new C C bonds, or the
Diels–Alder reaction of vinyl derivatives [2]. Optically active sulfox-
ides also have many industrial applications in the pharmaceutical
field, e.g. the synthesis of NSAID (an efficient non-steroidal anti-
inflammatory drug) [3] and Nexium (esomeprazole – one of the
most effective anti-ulcer drugs) [4]. Usually, only one enantiomer
(the eutomer) exhibits the desired biological activity [5]. For exam-
ple, the (S)-enantiomer of the ␤-receptor antagonist propanolol has
100-fold higher activity than the (R)-enantiomer [6].
Optical resolution, asymmetric oxidation of prochiral sulfides, and
nucleophilic addition of alkyl or aryl ligands to diastereoisomeri-
cally pure chiral sulfinates are representative examples of available
methods [10,11]. Biocatalytic approaches for enantio-enriched sul-
foxides can be also performed involving the oxidation of prochiral
sulfide precursors with isolated enzyme or whole-cell system. Such
processes have been developed with high enantioselectivity under
environmentally benign reaction conditions [8,9,12]. However they
could not be applied to every substrate and there are demands for
new systems with better performance.
Asymmetric reduction of racemic sulfoxides by bacterial
enzymes is another biological route used for the synthesis of
enantio-pure sulfoxides [4,13]. However, this option has received
much less attention until recently. Hanlon et al. reported an investi-
gation of three different microbial cells (e.g. Rhodobacter capsulatus,
Escherichia coli and Proteus vulgaris) [13] indicating that bacte-
rial (S)-oxide reductases, found in anaerobic bacteria, can exhibit
activity for the reduction of sulfoxides. The catalytic activity of
this reductase enzyme was evaluated both as whole-cell bacteria
and in the isolated form. Holt and co-workers [4] demonstrated
that microbial deoxygenating systems such as Citrobacter braakii,
Klebsiella sp. and Serratia sp. represent alternatives for the syn-
thesis of enantiopure sulfoxides. Also, Pseudomonas putida UV4
bacterium demonstrated stereo-differentiating reductase-catalytic
∗
Corresponding author. Tel.: +40 21 4100241; fax: +40 21 4100241.
E-mail addresses: v parvulescu@yahoo.com, vasile.parvulescu@g.unibuc.ro
(V.I. Parvulescu).
0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2012.06.050

M. Tudorache et al. / Applied Catalysis A: General 441–442 (2012) 42–46
43
MM (minimal medium) were prepared according to Hanlon et al.
[13] with the following composition: 48 mM Na2HPO4 × 2H2O,
22 mM KH2PO4, 8.5 mM NaCl and 2 g/L yeast. After autoclaving, the
trace element solution was added to the medium (1:50 dilution
of trace element solution). The stock trace elements solution con-
tained: 64 mg/L H3BO3, 18 mg/L CoCl2·6H2O, 4 mg/L CuCl2·2H2O,
340 mg/L ZnCl2, 2 mM MgSO4, 0.1 mM CaCl2 and 1 mg/L Na2MoO4.
MM solution was enriched with 20 ␮g/L vitamin B1 and 40 ␮g/L
vitamin B6. The carbon source for growth was either glucose or
glycerol (2 or 10 g/L, respectively) and the nitrogen source was
NH4Cl (1 g/L). The MM also contained DMSO (dimethyl sulfoxide)
at different concentrations (e.g. 0.1, 0.2, 0.4, 0.5, 1, 2, 4 and 8 g/L) as
the respiratory electron acceptor of E. coli bacteria under anaerobic
conditions.
Scheme 1. DKR (A) versus deracemization (B) strategies for the synthesis of chiral
molecules (e.g. alcohols, amines and sulfoxides).
activity for deoxygenation of rac-sulfoxides [14]. In spite of the
excellent results, these reports suggest that the use of bacterial
enzymes for deoxygenation of sulfoxides in a one-step process
allows only partial enantio-purification of the optically-active sul-
foxide [4,13].
Stock solution of phosphate saline buffer (PBS) of 100 mM was
prepared by mixing 80 g NaCl, 2 g KCl, 14.3 g Na2HPO4·2H2O and
3.43 g KH2PO4 in 1 L distilled water. The stock solution was diluted
to 25 mM PBS with distilled water and the pH value was adjusted
to 7.4 with either NaOH or HCl.
Dynamic kinetic resolution (DKR) and also deracemization con-
cepts offer a versatile alternative for the enantioselective synthesis
of asymmetric compounds (Scheme 1A). These methodologies only
been reported in connection with the synthesis of chiral alcohols,
amines and amino acids [15–19]. In principle, both concepts (e.g.
DKR and deracemization) require two steps: racemization followed
by a subsequent asymmetric transformation. For such purposes, a
tandem application of biocatalytic resolution and inorganic cata-
lyst racemization was reported by several groups as being effective
[20–24]. However, despite the chemo-, regio- and enantioselectiv-
ity of enzymes, the use of a single (purified) enzyme has several
disadvantages such as the high cost of the process (due to the
enzyme purification step), poor enzyme stability in the reaction
medium, or requirement of additional reagents (e.g. coenzymes and
activators). In order to circumvent these disadvantages, a whole-
cell approach in which the cells express a certain enzyme could be
an alternative.
2.2. rac-Sulfoxide synthesis and characterization
Methyl phenyl sulfoxide (MPSO) was used as received from
Acros Organics. Absolute configuration of methyl phenyl sulfoxide
was assigned by comparison of the sign of specific rotations with
pure enantiomers that were isolated by HPLC semi-preparative
chromatography. The optical rotation [˛] was recorded on a JASCO
OR–1590 polarimeter at 589 nm and 25 ◦C and compared to litera-
ture data [26].
Methyl p-tolyl sulfoxide (MTSO) and methyl mesityl sulfoxide
(MMSO) were prepared via the oxidation of the corresponding
thioether (e.g. methyl p-tolyl sulfide and methyl mesityl sulfide).
Methyl p-tolyl sulfide was purchased from Sigma–Aldrich and
mesityl methyl sulfide was prepared by chemical synthesis in-
house.
In the present study, we investigated a new concept for the
synthesis of asymmetric sulfoxides exploiting sequential deracem-
ization of racemic (rac) sulfoxides using a two-step protocol
(Scheme 1B). This involves the enantioselective deoxygenation of
rac-sulfoxides using E. coli bacteria combined with the selective
heterogeneous-catalytic oxidation of the resulting sulfide to the
racemate sulfoxide by a Ta2O5–SiO2 catalyst. Dimethyl sulfoxide
(DMSO) reductase contained in the E. coli bacteria is responsible
for deracemization of rac-sulfoxide [13]. A whole-cell system (E.
coli) is used because DMSO reductase, a membrane-bound enzyme,
often loses catalytic capacity when it is isolated from the cell [25].
E. coli resolution is efficient in terms of selectivity, but suffers from
being limited by a maximum theoretical conversion of 50% (kinetic
resolution). The combination of selective heterogeneous oxidation
with an enantioselective reduction is a strategy that can increase
not only the sulfoxides enantiopurity but also the efficiency of the
reaction. To the best of our knowledge this is the first example
of a deracemization process using a whole-cell biocatalyst com-
bined with a heterogeneous catalyst. The deracemization protocol
described in Scheme 1 involves a transformation process leading
back to the sulfide. This differs from the already well-known DKR
methodology (Scheme 1B).
2.2.1. Methyl mesityl sulfide
2,4,6-Trimethylbenzenethiol (10 mmol, 1.55 g) was added to
a suspension of NaH (20 mmol, 0.48 g) in benzene (50 mL). The
reaction mixture was stirred until no hydrogen evolution could
be observed. Then, the benzene was distilled off and the residue
re-dissolved in DMSO (15 mL) and treated with methyl iodide
(40 mmol, 5.68 g). The resulting solution was stirred overnight at
room temperature, then decomposed with water and extracted
with diethyl ether. The organic extracts were washed with brine,
NaHCO3 and dried over Na2SO4. Purification by silica gel chro-
matography with a mixture of petroleum ether/diethyl ether (3/1,
v/v) afforded 1.62 g (97.6%) of a viscous compound. 1H NMR
(CDCl3): 6.95 (s, 2H, ar-H), 2.53 (s, 6H, o-CH3), 2.27 (s, 3H, p-CH3),
2.22 (s, 3H, SCH3). Anal. Calcd. C, 72.23; H, 8.49; S, 19.28; Found: C,
72.25; H, 8.46; S, 19.34.
2.2.2. Oxidation of sulfides
A solution of H2O2 in acetic acid (65.8 mg/mL H2O2) was pre-
pared. One equivalent of H2O2 solution was added to a cooled
(0–5 ◦C) solution of sulfide in acetic acid (10 mM sulfide) using
a syringe. The resulting mixture was stirred for 20 min and then
stored in a refrigerator overnight. Then, the mixture was extracted
with diethyl ether (3 × 50 mL). The combined organic phases were
washed with brine, dried over Na2SO4, filtered, and evaporated
under reduced pressure. Crude extract was separated on a silica
gel column with a mixture of petroleum ether:diethyl ether (from
25% to 100% diethyl ether) as eluent. Pure sulfoxides were obtained.
All the prepared sulfoxides were characterized based on 1H NMR
spectra recorded on a Varian Gemini 300BB NMR spectrometer
(TMS internal standard). Chemical shifts (ı) are expressed in ppm,
2. Experimental
2.1. Chemicals and solutions
E. coli ATTC11303, glycerol, glucose and all of the salts used to
prepare the working solutions were purchased from Sigma-Aldrich.
E. coli top 10, dalfa5HL and Mac1 were purchased from Invitrogen.

44
M. Tudorache et al. / Applied Catalysis A: General 441–442 (2012) 42–46
Fig. 1. Influence of DMSO concentration in the culture medium on the E. coli growth (A) and sulfoxide enantio-reduction processes (S-ee (%)). Conditions: (A) E. coli (ACC11303
strain) growing conditions – 2 g/L glucose, 20 ␮g/L vitamin B1 and 40 ␮g/L vitamin B6 in MM, and (B) deracemization (1.0 g/L rac-MPSO and 2 g/L glucose in 25 mM PBS (pH
7.4)).
and J values are given in Hz. Elemental analysis was performed
on a Perkin Elmer CHN 240B instrument using a silica gel column
(70–230 mesh ASTM).
19.9 min and 25.7 min for (R)- and (S)-methyl p-tolyl sulfoxide
respectively; 25.1 min and 30.3 min for (R)- and (S)-methyl mesityl
sulfoxide respectively; 5.6 min for methyl phenyl sulfide; 5.9 min
for methyl p-tolyl sulfide; and 6.2 min for methyl mesityl sulfide.
It has been noticed that the HPLC chromatograms contained only
three peaks, e.g. (R) and (S) enantiomer of rac-sulfoxide, and the
corresponding sulfide.
2.2.3. Methyl p-tolyl sulfoxide
This compound was prepared based on the reaction of methyl
p-tolyl sulfide (10 mmol, 1.38 g) and H2O2 in acetic acid (5.16 mL of
65.8 mg/mL solution). After reaction, 1.50 g (97.4% purity) methyl
p-tolyl sulfoxide was isolated. 1H NMR (CDCl3): 7.54 (dAB, 2H,
J = 9.6 Hz, ar-H), 7.34 (dAB, 2H, J = 9.5 Hz, ar-H), 2.71 (s, 3H, SOCH3),
2.42 (s, 3H, p-CH3). Anal. Calcd. C, 62.30; H, 6.54; O, 10.37; S, 20.79;
Found: C, 62.28; H, 6.55; S, 20.74.
3. Results and discussion
3.1. Whole-cell resolution of rac-sulfoxide
Bio-resolution of the rac-sulfoxide by whole-cell (E. coli) reduc-
tion has been systematically optimized with respect to the source
and growth conditions of the biocatalyst. Thus, different strains of
E. coli (e.g. E. coli ATTC11303, top 10, dalfa5HL and Mac1) were
tested for the resolution of rac-sulfoxides. The reactions focused on
the separation of the (S) enantiomer to the detriment of the (R) one.
Methyl phenyl sulfoxide (MPS) racemate was the first substrate
investigated in this context.
2.2.4. Methyl mesityl sulfoxide
This compound was prepared from methyl mesityl sulfide
(10 mmol, 1.66 g) and H2O2 in acetic acid (5.16 mL of 65.8 mg/mL
solution). 1.62 g (89.0% purity) methyl mesityl sulfoxide was iso-
lated after reaction. 1H NMR (CDCl3): 6.85 (s, 2H, ar-H), 2.83 (s, 3H,
SOCH3), 2.55 (s, 6H, o-CH3), 2.27 (s, 3H, p-CH3). Anal. Calcd. C, 65.89;
H, 7.74; S, 17.59; Found: C, 65.91; H, 7.72; S, 17.51.
Deracemization of rac-sulfoxides to sulfides was achieved by
dimethyl sulfoxide (DMSO) reductase contained in the E. coli bacte-
ria. DMSO reductase was expressed during the anaerobic bacterial
growth in MM (minimal medium) enriched with DMSO as an
enzyme substrate, and also with additional carbon (glycerol or glu-
cose) and energy (B1 and B6 vitamins) sources.
The experimental conditions for bacteria growth were opti-
mized taking into account the composition of the growth medium.
No growth of the cultures was detected when only DMSO enriched
the MM (e.g. a maximum absorbance of only 0.4 for culture sus-
pensions was reached after 24 h – Table 1). Also, no significant
differences in growth rate were obtained changing the carbon
source (0.77 and 0.60 absorbance for glycerol and glucose, respec-
tively – Table 1). However, a consistent enhancement of the growth
rate was observed when B1 and B6 vitamins were added, for both
carbon sources equally (Table 1).
2.3. E. coli growth
E. coli cells were grown anaerobically in MM at 37 ◦C under
a gentle stirring. The cultures were grown under inert nitrogen
atmosphere in glass bottles equipped for gas bubbling. Nitrogen
atmosphere ensured anaerobic conditions for the bacterial growth.
The cell harvesting was performed at stationary phase followed by
centrifugation, washing and dispersing the E. coli cells in 25 mM
PBS (pH 7.4).
The active bacterial growth was followed using spectrophoto-
metric analysis of the liquid medium. Samples from the growing
medium were taken at different incubation times and analyzed
with a UV–Vis detector set up at 610 nm (the absorbance signal
increases with bacterial cell population).
Fig. 1 shows the effect of the DMSO concentration in the cul-
ture medium on the E. coli (ATTC11303 strain) growth process
(Fig. 1A) and the correlation of growth rate with DMSO reductase
2.4. Sequential deracemization of rac-sulfoxide
Deracemization was performed in a glass vessel containing
10 mL PBS (25 mM, pH 7.4), 1.8 mg glucose and 10 mg sulfoxide.
The vessel was sealed with a rubber stopper and nitrogen was
flushed inside. The reaction mixture was incubated at 37 ◦C without
agitation for 24 h. After reaction, the solution was extracted with
diethyl ether (∼2 mL) and evaporated to dryness. The residue was
re-dissolved in hexane/iso-propanol (8/2).
Table 1
Optimization of culture medium content for E. coli growth expressing DMSO reduc-
tase enzyme. Conditions: E. coli ATTC11303 strain, 0.5 g/L DMSO and 24 h incubation
time.
DMSO (g/L)
Glycerol (g/L)
Glucose (g/L)
Vitamins
Abs.a
The enantiomeric excess (ee) of the separated sulfoxides was
determined by HPLC analysis (AGILENT 1100/1200 with a S,S-
Whelk-O chiral column (4.6 × 250 mm) and a mobile phase of
hexane:iso-propanol (8/2) with a flow rate of 1 mL/min). The detec-
tion of the sulfoxides was performed at 254 4 nm. The elution
times for sulfoxides and corresponding thioethers were 18.6 min
and 23.3 min for (R)- and (S)-methyl phenyl sulfoxide respectively;
B1 (␮g/L)
B6 (␮g/L)
0.5
0.5
0.5
0.5
0.5
–
10
–
10
–
–
–
2
–
2
–
–
–
20
20
–
–
–
40
40
0.40
0.77
0.60
1.06
1.03
a
Abs. – optical absorbance of the culture medium.

M. Tudorache et al. / Applied Catalysis A: General 441–442 (2012) 42–46
45
Table 2
Screening of E. coli in rac-MPSO deoxygenation.
ATTC11303
Mac1
dH5alfa
Top10
Abs.a
0.90
25.60
1.19
23.60
1.04
2.56
1.37
3.54
ee (%)b
a
E. coli (ATTC11303 strain) (growing conditions – 2 g/L glucose, 20 ␮g/L vitamin
B1 and 40 ␮g/L vitamin B6 in MM; incubation time 8 h).
b
rac-MPSO enantio-reduction (1.0 g/L rac-MPSO and 2 g/L glucose in 25 mM PBS
(pH 7.4); ee (%) of (S)-MPSO).
activity (Fig. 1B). These dependences demonstrate that the E. coli
(ATTC11303 strain) growth process was influenced by the presence
of DMSO in the growth medium. The culture density measured after
a standard time increased slowly with DMSO concentration in the
range of 0.2–2.0 g/L (Fig. 1A). A higher DMSO concentration (e.g. in
the range 4–8 g/L DMSO) led to poisoning of the bacteria and thus
to an inhibition of the growth process (Fig. 1A). Deracemization of
rac-MPSO was found to depend on the DMSO concentration in the
growth medium (Fig. 1B). The maximum conversion of rac-MPSO
was achieved when bacteria were treated in a growth medium with
0.5 g/L DMSO (26% ee of (S)-MPSO). Thus, 0.5 g/L DMSO was cho-
sen as an optimum DMSO concentration in the growth medium
for further experiments. In summary, E. coli (ATTC11303 strain)
bacteria were cultivated in a culture medium with an optimum
composition: 2 g/L glucose, 0.5 g/L DMSO, 20 ␮g/L B1 and 40 ␮g/L
B6 vitamins dissolved in MM [13].
Fig. 2. Testing the catalytic capacity of E. coli (ATTC11303 strain) bacteria on
enantio-reduction of rac-sulfoxides. Conditions: (A) E. coli (ATTC11303 strain)
(growing conditions – 0.5 g/L DMSO, 2 g/L glucose, 20 ␮g/L vitamin B1 and 40 ␮g/L
vitamin B6 in MM), and (B) deracemization (1.0 g/L rac-sulfoxide and 2 g/L glucose
in 25 mM PBS (pH 7.4); ee (%) of (S)-MPSO).
Collections of different strains of E. coli (e.g. ATTC11303, Mac1,
top10 and dalfa5HL) were investigated in terms of DMSO reduc-
tase capacity of rac-sulfoxides. Rac-MPSO was selected again as a
model for testing the catalytic capacity of E. coli strains. The result-
ing DMSO reductases showed catalytic activity which varied for
the different bacteria strains (Table 2). Thus, the E. coli top 10 strain
exhibited the highest rate of rate of growth but had a low der-
acemization activity (1.37 absorbance of E. coli culture and 4% ee in
(S)-MPSO). In contrast, E. coli ATTC11303 and Mac1 exhibited high
preference for the R-MPSO enantiomer leading to high enantiose-
lectivities in (S)-MPSO (26 and 24% ee, respectively).
and tungsten-based catalysts tend to generate leaching but with an
extent correlated with the nature of the support [28,29].
However, the Ta2O5–SiO2 catalyst is characterized by high selec-
tivity and easy recovery from the reaction mixture [27], which
are important properties for our oxidation system. The reason we
tolerate such low conversion efficiencies is to preserve the high
selectivity. Obviously increased conversions, led to further oxida-
tion of the sulfoxides to sulfones, which is easily achieved [9,30,31].
3.3. Sequential deracemization of sulfoxides
Next, the catalytic activity of DMSO reductase of E. coli
(ATTC11303) was investigated with respect to different rac-
sulfoxides. Methyl phenyl sulfoxide (MPSO), methyl p-tolyl
sulfoxide (MTSO) and methyl mesityl sulfoxide (MMSO) were
examined as substrates (Fig. 2). The whole-cell bacteria catalyzed
the deracemization of rac-sulfoxide converting the (R)-enantiomer
to the thioether, except MMSO that was not recognized as a
DMSO reductase substrate (1% ee (S)-MMSO). The investigated
strain developed 62% ee of (S)-MTSO (E = 4) with 38% conversion
of rac-MTSO in one reduction step of rac-MTSO. The ability of the
developed cells to performed enantio-conversion of sulfoxide to
thioether is confirmed by the data already reported in the literature
(e.g. 60% ee (S)-MTSO for conversion lower than 50%) [4].
The main drawback of the bio-resolution of rac-sulfoxide is the
chemical conversion of 49.3% (limited to a maximum of 50%). This
limitation was overcome by combining E. coli resolution with achi-
ral oxidation of thioether product back to the rac-sulfoxide.
Deracemization steps (e.g. sulfoxide desoxygenation and
thioether oxidation) were performed in different vials due to
the incompatibility of the experimental parameters for these two
chemical processes (i.e. sulfide oxidation requires oxygen precur-
sor, while sulfoxide desoxygenation is perturbed by the presence of
oxygen). First, the desoxygenation step of rac-MTSO allowed us to
reach 62% ee of (S)-MTSO (E = 4) with 38% conversion of rac-MTSO
in one reduction step of rac-MTSO. The conversion of rac-MTSO
involved the (R)-MTSO deoxyreduction to methyl p-tolyl sulfide. No
secondary products were detected in the reaction mixture extract
(Section 2.4).
3.2. Heterogeneous oxidation of thioethers
Oxidation of the resulting thioethers (sulfides) to sulfoxides
was carried out with a silica-incorporated Ta catalyst (15 wt%
Ta2O5) prepared following a procedure reported elsewhere [27].
The sulfoxidation of thioether was carried out in an ionic liquid
([bmim][NTf2]) using an aqueous solution of 30 wt% hydrogen per-
oxide, as the oxidant, at 40 ◦C, for 15 min with 5 mg catalyst. The
reaction occurred with a selectivity of 92% in MPSO for a conver-
sion of 20%, 93% MTSO for a conversion of 22%, and 94% MDMTSO
for a conversion of 19%. The difference in the selectivity from 100%
was explained by the presence of the corresponding sulfones. The
selection of the Ta-based catalyst was made due to the fact that it
generates no leaching under the investigated conditions. Titanium
The enantioselective reduction was repeated three times with
fresh E. coli leading to an ee enrichment of 97.3% (S)-MTSO (E = 73)
for 49.3% conversion of rac-sulfoxide. Then, the resulting sulfide
was purified by TLC technique (alumina-coated plate, 60% EtAc
in hexane) [14]. The purified sulfide was subjected to oxidation
with 22% conversion of sulfide. Working with rac-sulfoxides led to
the same selectivities as those obtained using pure substrates. The
resulting rac-sulfoxide was exposed to a new deracemization cycle.
After three cycles, purified (S)-MTSO was obtained with 56% yield
(98% ee of (S)-MTSO). Coupling the conventional catalytic oxidation
process with the bio-reduction of rac-sulfoxide led a 20% enhance-
ment of the process yield (the yield of the bio-reduction step was

46
M. Tudorache et al. / Applied Catalysis A: General 441–442 (2012) 42–46
36%, while for the sequential deracemization process the yield was
56%).
[3] A. Korte, J. Legros, C. Bolm, Synlett (2004) 2397–2399.
[4] H.R. Luckarift, H. Dalton, N.D. Sharma, D.R. Boyd, R.A. Holt, Appl. Microbiol.
Biotechnol. 65 (2004) 678–685.
[5] I. Fernandez, N. Khiar, Chem. Rev. 103 (2003) 3651–3705.
[6] R.J. Butlin, I.D. Linney, M.F. Mahon, H. Tye, M. Wills, J. Chem. Soc. Perkins Trans.
1 (1996) 95–105.
4. Conclusions
[7] G. Solladie, Synthesis 3 (1981) 185–196.
[8] A.-T. Li, H.-L. Yu, J. Pan, J.-D. Zhang, J.-H. Xu, G.-Q. Lin, Bioresour. Technol. 102
(2011) 1537–1542.
[9] E. Wojaczyn´ ska, J. Wojaczyn´ ski, Chem. Rev. 110 (2010) 4303–4356.
[10] J. Legros, J.R. Dehli, C. Bolm, Adv. Synth. Catal. 347 (2005) 19–31.
[11] Q.Y. Huang, J. Zhu, J.G. Deng, Chin. J. Org. Chem. 25 (2005) 496–506.
[12] S. Colonna, N. Gaggero, A. Manfredi, L. Casella, M. Gullotti, G. Carrea, P. Pasta,
Biochemistry 29 (1990) 10465–10468.
In conclusion, optically pure sulfoxides ((S)-enantiomer) were
obtained based on cyclic deracemization with a sequential proto-
col. Practically, E. coli cell catalyzed the reduction of (R)-sulfoxide to
the thioether and a heterogeneous Ta2O5–SiO2 catalyst catalyzed
the oxidation of the ethers to the rac-sulfoxide. The resulting rac-
sulfoxide was then exposed to a new deracemization cycle. This
process was repeated until the (R)-enantiomer was eliminated
from the reaction phase. The performances of E. coli depended on
the strain and the conditions in which bacteria growth occurred
(e.g. source of carbon, presence of vitamins, and concentration of
DMSO). Under optimized conditions, the E. coli cells performed the
rac-MTSO resolution leading to an enantiomeric excess of around
97% (S)-MTSO for 49% conversion of the rac-MTSO. The oxidation
of the separated thioether to the sulfoxide occurred with a high
selectivity for conversion of 22%. Finally, 98% ee of (S)-sulfoxide
with 56% yield in (S)-sulfoxide was reached after three deracemiza-
tion cycles for rac-MTSO. The sequential deracemization approach
is an efficient way to enantio-separate rac-sulfoxides allowing to
gain an additional 20% yield. A scale-up of the proposed sequential
deracemization process will be carried out in the next future.
[13] S.P. Hanlon, D.L. Graham, P.J. Hogan, R.A. Holt, C.D. Reeve, A.L. Shaw, A.G. McE-
wan, Microbiology 144 (1998) 2247–2253.
[14] D.R. Boyd, N.D. Sharma, A.W.T. King, S.D. Shepherd, C.C.R. Allen, R.A. Holt, H.R.
Luckarift, H. Dalton, Org. Biomol. Chem. 2 (2004) 554–561.
[15] F.F. Huerta, A.B.E. Minidis, J.-E. Backvall, Chem. Soc. Rev. 30 (2001) 321–331.
[16] A. Parvulescu, D. de Vos, P. Jacobs, Chem. Commun. (2005) 5307–5309.
[17] B. Martin-Matute, J.-E. Backvall, Curr. Opin. Chem. Biol. 11 (2007) 226–232.
[18] J. Paetzold, J.-E. Backvall, J. Am. Chem. Soc. 127 (2005) 17620–17621.
[19] J.M. Foulkes, K.J. Malone, V.S. Coker, N.J. Turner, J.R. Lloyd, ACS Catal. 1 (2011)
1589–1594.
[20] S. Wuyts, K. de Temmerman, D. de Vos, P. Jacobs, Chem. Commun. (2003)
1928–1929.
[21] P. Lozano, E. Garcia-Verdugo, N. Karbass, K. Montague, T. de Diego, M.I. Bur-
guete, S.V. Luis, Green Chem. 12 (2010) 1803–1810.
[22] O. Langvik, T. Saloranta, A. Kirilin, A. Liljeblad, P. Maki-Arvela, L.T. Kanerva, D.Y.
Mursin, R. Leino, ChemCatChem 2 (2010) 1615–1621.
[23] C.V. Voss, C.C. Gruber, W. Kroutil, Angew. Chem. Int. Ed. 47 (2008) 741–745.
[24] N.J. Turner, Curr. Opin. Chem. Biol. 14 (2010) 115–121.
[25] W.A. Duetz, J.B. Van Beilen, B. Witholt, Curr. Opin. Biotechnol. 12 (2001)
419–425.
[26] J.M. Brunel, P. Diter, M. Duetsch, H.B. Kagan, J. Org. Chem. 60 (1995) 80–86.
[27] V. Cimpeanu, V. Parvulescu, V.I. Parvulescu, M. Capron, P. Granger, J.M. Thomp-
son, C. Hardacre, J. Catal. 235 (2005) 184–194.
[28] D.C. Radu, V.I. Pârvulescu, V. Câmpeanu, E. Bartha, A. Jonas, P. Grange, Appl.
Catal. A: Gen. 242 (2003) 77–84.
[29] A. Crosman, G. Gelbard, G. Poncelet, V.I. Parvulescu, Appl. Catal. A: Gen. 264
(2004) 23–32.
[30] A. Lattanzi, P. Iannece, A. Scettri, Tetrahedron: Asymmetry 15 (2004)
1779–1785.
Acknowledgments
This work was financially supported by the PN II HR YT-96 pro-
gram, contract no. 91/2010 from CNCSIS.
References
[31] J. Legros, C. Bolm, Chem. Eur. J. 11 (2005) 1086–1092.
[1] S. Nakamura, Y. Watanable, T. Toru, J. Org. Chem. 65 (2000) 1758–1766.
[2] H. Pellissier, Tetrahedron 62 (2006) 5559–5601.