Stereoselective reductase-catalysed deoxygenation of sulfoxides in aerobic and anaerobic bacteria.

Article abstract of DOI:10.1039/b313714g

Direct and indirect evidence, of unexpected stereoselective reductase-catalysed deoxygenations of sulfoxides, was found. The deoxygenations proceeded simultaneously, with the expected dioxygenase-catalysed asymmetric sulfoxidation of sulfides, during some biotransformations with the aerobic bacterium Pseudomonas putida UV4. Stereoselective reductase-catalysed asymmetric deoxygenation of racemic alkylaryl, dialkyl and phenolic sulfoxides was observed, without evidence of the reverse sulfoxidation reaction, using anaerobic bacterial strains. A purified dimethyl sulfoxide reductase, obtained from the intact cells of the anaerobic bacterium Citrobacter braakii DMSO 11, yielded, from the corresponding racemates, enantiopure alkylaryl sulfoxide and thiosulfinate samples.

Full text of DOI:10.1039/b313714g

View Article Online / Journal Homepage / Table of Contents for this issue
Stereoselective reductase-catalysed deoxygenation of sulfoxides in
aerobic and anaerobic bacteria
Derek R. Boyd,*a Narain D. Sharma,a Alistair W. T. King,a Steven D. Shepherd,a
Christopher C. R. Allen,b Robert A. Holt,c Heather R. Luckariftd and Howard Daltond
a School of Chemistry, The Queen’s University of Belfast, Belfast, UK BT9 5AG
b The QUESTOR Centre, The Queen’s University of Belfast, Belfast, UK BT9 5AG
c Avecia Pharmaceuticals, Billingham, Cleveland, UK TS23 1YN
d Department of Biological Sciences, The University of Warwick, Coventry, UK CV4 7AL
Received 29th October 2003, Accepted 1st December 2003
First published as an Advance Article on the web 16th January 2004
Direct and indirect evidence, of unexpected stereoselective reductase-catalysed deoxygenations of sulfoxides,
was found. The deoxygenations proceeded simultaneously, with the expected dioxygenase-catalysed asymmetric
sulfoxidation of sulfides, during some biotransformations with the aerobic bacterium Pseudomonas putida UV4.
Stereoselective reductase-catalysed asymmetric deoxygenation of racemic alkylaryl, dialkyl and phenolic sulfoxides
was observed, without evidence of the reverse sulfoxidation reaction, using anaerobic bacterial strains. A purified
dimethyl sulfoxide reductase, obtained from the intact cells of the anaerobic bacterium Citrobacter braakii DMSO
11, yielded, from the corresponding racemates, enantiopure alkylaryl sulfoxide and thiosulfinate samples.
Based on the sulfoxidation results, obtained using whole cell
Introduction
P. putida mutant strains containing dioxygenases,9,10,17 it could
be assumed that the remarkably high stereoselectivity and yield
often observed was entirely due to a process of asymmetric
Heteroatom oxidation of sulfides to yield sulfoxides is among
the most common and facile metabolic steps in eukaryotic and
prokaryotic organisms.1–22 This reaction has been catalysed by a
range of enzyme systems including peroxidases,14,15 monooxy-
genases,1–8,11,12 and dioxygenases.8–10,17–22 Particular emphasis
has recently been placed on the application of dioxygenases,
e.g. toluene dioxygenase (TDO) and naphthalene dioxygenase
(NDO), as biocatalysts in chiral sulfoxidation.8,9,17–22 The di-
oxygenases, present in different strains of the soil bacterium
Pseudomonas putida, e.g. UV4 (a source of TDO), NCIMB
8859 (a source of NDO) and 9816/11 strains (a source of
NDO), have been found to catalyse the stereoselective sulfoxid-
ation of a wide range of alkylaryl and diaryl sulfides.8–10,17–22 To
date, > 40 sulfoxides have been isolated with high enantiomeric
excess values (> 90% ee), using TDO and NDO enzymes. The
dioxygenases are, thus, among the best enzymes currently
available for the production of enantiopure sulfoxides. Typical
examples of enantiopure sulfoxides, obtained using the
P. putida UV4 mutant strain, are shown in Scheme 1.
oxidation (A
[ϩ]-B or [Ϫ]-B, Scheme 2). Preliminary
studies,17 with whole cell cultures of P. putida UV4, however,
indicate that kinetic resolution could also be responsible for the
residual enantioenriched (R)-alkylaryl sulfoxides recovered
from some racemic precursors without corresponding bioprod-
ucts being isolated. The kinetic resolutions could be explained
by several mechanistic pathways including: (i) a non-stereo-
selective slow deoxygenation process (B
tant fast stereoselective sulfoxidation step to yield the (R)
enantiomer (A B, Scheme 2) or (ii) the stereoselective
A) and a concomi-
removal of the (S) enantiomer, e.g. by cis-dihydroxylation,
resulting in the formation of very water-soluble metabolites
that could not be isolated by solvent extraction (ethyl acetate).
Precedence for the production of an enantioenriched alkyl-
aryl sulfoxide, where enzyme-catalysed oxygenation and deoxy-
genation mechanisms both occur, is found in an earlier report23
of a fungal biotransformation of sulfide A (R = 4-H2N.C6H4,
RЈ = Me) to yield the corresponding sulfoxide B; this was
accompanied by a simultaneous, but slower, stereoselective
enzyme-catalysed deoxygenation of sulfoxide B (Scheme 2) to
give sulfide A.23 Under these circumstances, where a fast sulf-
oxidation step (and possibly other enzyme-catalysed oxid-
ations) happens simultaneously with a relatively slow deoxy-
genation process, it is very difficult to detect the sulfide and thus
obtain direct evidence of the reduction reaction. The present
study is concerned with: (i) the quest for both direct and
indirect evidence of a stereoselective sulfoxide deoxygenation
mechanism during aerobic and anaerobic bacterial biotrans-
formations, and (ii) the potential application of the enzymatic
deoxygenation process in the kinetic resolution of racemic sulf-
oxides which are difficult to obtain as single enantiomers by
enzymatic asymmetric oxidation methods.
Scheme 1
Asymmetric sulfoxidation of methylphenyl sulfide 1 (R =
Me) substrate was of particular interest since it yielded either
the (R)-sulfoxide enantiomer 2 (R = Me, > 98% ee; 90% yield),17
using TDO or the (S) enantiomer (> 98% ee; 98% yield) using
NDO.9,10 Further noteworthy examples, of enantiocomplemen-
tary biocatalysis using TDO and NDO, have been found during
the sulfoxidation of bicyclic bis-sulfides where the resulting
enantiopure sulfoxides proved to be of value in synthesis.22
Although the stereoselective TDO-catalysed sulfoxidation, of
alkylaryl and diaryl sulfides in P. putida UV4, has been very
successful, sulfoxidation of dialkyl sulfides has proved to be
much more difficult, with this biocatalyst.18
Several biotransformation pathways, involving enzyme-
catalysed deoxygenation, which could account for the kinetic
resolution of racemic sulfoxides, are shown in Scheme 2. These
include: (i) sulfoxide deoxygenation to yield a sulfide (B
(ii) sulfoxide deoxygenation–monooxygenation to yield a sul-
fide alcohol (B E), (iii) sulfoxide deoxygenation–
A),
A
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
554

View Article Online
Scheme 2
monooxygenation to yield a new sulfoxide diastereoisomer
(B F) and (iv) sulfoxide dioxygenation–deoxygenation
to yield a sulfide cis-dihydrodiol (B D) (Scheme 2).
studies of the racemic alkylaryl sulfoxides.17 In our current
study, the biotransformations of alkylaryl sulfoxides have been
re-examined in light of this one example of a deoxygenation
A
C
Employing P. putida UV4 whole cells, and a series of racemic
alkylaryl sulfoxide substrates, evidence for each of these mech-
anisms is presented in the results and discussion section.
Bacterial dimethyl sulfoxide (DMSO) reductase enzymes
play an important role in the environment, by regulating the
level of DMSO and dimethyl sulfide (DMS) in the global sulfur
cycle. Thus, the deoxygenation of DMSO, formed by photo-
oxidation of volatile DMS in the atmosphere, is catalysed by
DMSO reductases found mainly in ocean-dwelling bacteria.
The second part of this study is focused, exclusively, on poten-
tial application of such facultative anaerobic bacteria which
have been shown to contain sulfoxide reductase enzymes. To
date, few reports are available on the use of sulfoxide reductase
enzymes in the kinetic resolution of racemic sulfoxides, and
none on the application of these enzymes in the production of
single enantiomer thiosulfinates. These biocatalysts have now
been found to deoxygenate, stereoselectively, a new range of
sulfoxides, including a thiosulfinate, without evidence of the
reverse sulfoxidation reaction occurring.
(5
6)21 and the recent improvements in down-stream process-
ing of bioproducts from P. putida UV4.18 The latter process
involved removal of most of the water under vacuum, from the
centrifuged culture medium containing the metabolites, prior to
solvent extraction. With this method, previously undetected
metabolites, including water-soluble sulfoxide cis-dihydrodiols,
have been isolated.18
When racemic methylphenyl sulfoxide 2 (R = Me, Scheme 1)
was earlier added to whole-cell cultures of P. putida UV4, the
normal extraction procedure was adopted for the isolation
of metabolites (saturating the aqueous culture medium with
common salt followed by its repeated extractions with
EtOAc).17 Under these conditions, only a low yield (7%) of
enantioenriched residual sulfoxide (2R, 85% ee) was recovered,
without any trace of methylphenyl sulfide 1 or other metabol-
ites.17 The biotransformation of racemic sulfoxide 2 was
repeated during the present study, in the hope of finding other
metabolites, which would help in understanding the mechanism
of this kinetic resolution. The new protocol involved: (i) an
extended period of biotransformation and (ii) the improved
work-up procedure.18 Using this modified technique, no resi-
dual sulfoxide 2 was recovered; the resulting crude mixture of
metabolites, on 1H NMR spectral analysis, showed the presence
of cis-diol sulfoxides (7a, 34%),7b (26%), cis-diol sulfide 8
(30%) and cis-diol sulfone 9 (10%) (Scheme 4). The mixture of
cis-diol sulfoxide diastereoisomers 7a/7b could not be separated
by PLC, but pure samples of the sulfide diol 8 and sulfone diol
9 were obtained by this PLC. Compounds 7a, 8 and 9 were
reported earlier17,19,24 only as minor cis-dihydrodiol metabolites
or derivatives.
Results and discussion
(A) Enzyme-catalysed deoxygenation of racemic sulfoxides,
using the UV4 mutant strain of the aerobic bacterium P. putida
The biotransformation, of thiophene 3 in P. putida UV4, has
recently been found to yield the corresponding achiral sulfoxide
intermediate 4.20,21 This unstable thiophene oxide 4 was found
to dimerise to yield the racemic bis-sulfoxide 5 (Scheme 3).
Stereoselective and regioselective deoxygenation of one sulf-
oxide group of the dimer was observed as a further step (5
6)
during the biotransformation of thiophene 3, and also when
racemic bis-sulfoxide 5 was added as a substrate, to yield an
excess (77% ee) of one enantiomer of monosulfoxide 6 (Scheme
3).21
Scheme 3
Scheme 4
This direct evidence of a dioxygenase-catalysed sulfoxidation
(3
4), and an enzyme-catalysed deoxygenation (5
6) oper-
Application of similar biotransformation and work-up con-
ditions, and methylphenyl sulfide 1 as substrate, yielded cis-diol
sulfide 8 as the sole isolable metabolite. These observations are
ating in tandem, was assumed to be due to the very slow rate of
the reverse TDO-catalysed sulfoxidation of a dialkyl sulfide
(6
5). While stereoselective dioxygenase-catalysed (P. putida
consistent with either a single-step mechanism (sulfide 1
diol sulfide 8) or a three-step mechanism (sulfide 1 sulfoxide
cis-diol sulfoxide 7a/7b cis-diol sulfide 8, Scheme 4). In
cis-
UV4) sulfoxidation reactions, of a wide range of alkylaryl sul-
fides, have been reported,8,9,17,18,22 no other direct evidence, of
the reverse deoxygenation reaction, had been obtained in earlier
2
order to elucidate the mechanism, a further experiment with
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
555

View Article Online
Scheme 5
Scheme 6
sulfide 1 was conducted using a shorter time of biotransform-
ation. On this occasion, the enantiopure cis-diol sulfoxide
intermediate 7a was intercepted as the only bioproduct. When
sulfoxide diol 7a was, in turn, added as substrate, both sulfide
diol 8 and sulfone diol 9 were formed. These results clearly
minor metabolites, 3-hydroxy-2,3-dihydrobenzo[b]-thiophene
12 (5% yield) and cis-dihydrodiols 14 (3% yield) and 15 (2%
yield) (Scheme 5). The enantioenriched (83% ee) metabolite
(ϩ)-(3S)-12 was assumed to have an identical absolute con-
figuration (allowing for Sequence Rule priorities) to that found
for alcohols obtained earlier by TDO-catalysed (P. putida UV4)
benzylic hydroxylations of benzocycloalkanes, benzocyclo-
heteroalkanes, and benzocycloalkenes.25–34 The minor cis-
dihydrodiol metabolites 14 and 15 had earlier been isolated as
bioproducts from benzo[b]thiophene 13; compound 13 was
obtained by dehydration of 3-hydroxy-2,3-dihydrobenzo[b]-
thiophene 12.35,36 The structures and stereochemical assign-
ments of cis-dihydrodiols 14 and 15 (which spontaneously
isomerise to the corresponding trans-isomers)36 will be dis-
cussed elsewhere. Although the sulfide bioproduct 11, derived
from deoxygenation of sulfoxide substrate 10, was not observed
directly, the formation of mono- (12) and dihydroxylation
products (14 and 15) provides indirect evidence of a slow
reductase-catalysed deoxygenation of the alkylaryl sulfoxide 10
indicate that an unprecedented three-step sequence (1
2
7a
8, Scheme 4), including an enzyme-catalysed deoxygen-
ation step, is preferred to the single-step (1
8) mechanism.
The observation also supports the premise that dioxygenase-
catalysed sulfoxidation of alkylaryl sulfides generally occurs
much faster than dioxygenase-catalysed arene-cis-dihydroxyl-
ation. Conversely, cis-dihydroxylation of dialkyl sulfides, e.g.
methylbenzyl sulfide,18 proceeds at a much faster rate than
sulfoxidation.
It is, thus, possible that the kinetic resolution earlier observed
for sulfoxide 2 (R = Me)17 was mainly due to cis-dihydroxyl-
ation to yield the water-soluble cis-diol sulfoxide metabolites
7a/7b; the water-soluble metabolites were not isolated due to
the less efficient extraction procedure. While dioxygenase–
catalysed sulfoxidations8–10,17–22 occur readily, sulfones e.g.
metabolite 9, have rarely been observed as bioproducts from
these biotransformations. A cis-diol sulfone metabolite, of
similar structure and configuration to compound 9, was how-
ever recently obtained when ethylphenyl sulfide was used as
substrate.19 The enzyme, responsible for sulfone formation in
P. putida UV 4, has not yet been identified.
(B
A
E, Scheme 2).
Biotransformation of 1,3-disulfide 16 had been found to
yield the corresponding (1S,2R)-(17cis) and (1S,2S)-(17trans) sulf-
oxide enantiomers (> 98% ee, Scheme 6).8 During the present
study, the racemic sulfoxides 17cis and 17trans were, separately,
added as substrates to P. putida UV4 (Scheme 6); no direct
evidence for reductase-catalysed deoxygenation of racemic sulf-
oxides 17cis and 17trans was obtained. However, isolation of resi-
dual (1S,2R)-17cis (40% recovered yield, 47% ee), from racemic
substrate, is indicative of a kinetic resolution process in oper-
ation. Furthermore, isolation of diastereoisomer (1S,2S)-17trans
(> 98% ee), as a metabolite of racemic substrate 17cis, is consist-
ent with its deoxygenation to form sulfide 16 followed by TDO-
It was noted that neither of the direct deoxygenation reac-
tions, found using P. putida UV4 (5
6 and 7a
8), involved
an alkylaryl sulfoxide substrate. While cis-dihydroxylation may
play a major role in the kinetic resolution of the racemic
alkylaryl sulfoxide 2, the possibility of a competing pathway,
i.e. slow deoxygenation of compound 2 followed by a fast
sulfoxidation of the resulting sulfide 1, could not at this stage be
discounted.
The first evidence, of a slow deoxygenation process in alkyl-
aryl sulfoxides, was obtained when the racemic bicyclic com-
pounds 10, 17cis, 17trans and 18 were used as substrates (Schemes
5–7). Biotransformation of 2,3-dihydrobenzothiophene 11 had
catalysed asymmetric sulfoxidation (B
A
F, Scheme 2).8
Similar indirect evidence, of reductase-catalysed stereoselective
deoxygenation, was observed with racemic substrate 17trans; the
residual (1S,2S)-sulfoxide 17trans was found to be enantioen-
riched (18% ee) and the diastereoisomeric metabolite (1S,2R)-
17cis was enantiopure (> 98% ee), as expected8 from TDO-
catalysed oxidation of sulfide 16. The very low yields (1–2%) of
diastereoisomeric sulfoxide metabolites 17trans and 17cis, derived
from racemic samples of 17cis and 17trans, respectively, again
suggests that the deoxygenation step is slow.
been found to yield the sulfoxide (1R)-10 (26% ee, 11
10,
Scheme 5);22 it was also accompanied by 3-hydroxy-2,3-di-
hydrobenzo[b]-thiophene 12, as a result of TDO-catalysed ben-
zylic hydroxylation (11
12). When racemic sulfoxide 10 was
added to P. putida UV4, during the present study, the residual
substrate (1R)-10 (19% yield, 3% ee) was isolated along with the
The bicyclic alkylaryl sulfide 19 was exceptional in not being
oxidized to the corresponding sulfoxide 18 by whole cells of
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
556

View Article Online
reduce several sulfoxides through the action of a periplasmic
DMSO reductase that had already been purified and fully char-
acterized by X-ray crystallography.39,40 The other bacterial
strains that showed most promise, in terms of stereoselective
deoxygenation of sulfoxide 20 (ee values 32 to > 98%), are
shown in Table 1. Although attempts were made in each case to
follow the course of biotransformations by HPLC analysis, and
to terminate the biotransformation after ∼50% conversion of
substrate, it proved difficult, in some cases, due to the different
growth characteristics, solubilities and biotransformation rates
associated with the range of bacteria and substrates. Thus, the
enantiopurity figures shown in Table 1 merely represent the
values observed, upon termination of the biotransformation,
which were not always optimal.
A comparison of results, obtained using whole cells, and
pure DMSO reductases from R. capsulatus and the newly iso-
lated bacterium C. braakii, on biotransformations of sulfoxide
20, showed that the progress of reaction with pure enzymes
could be more easily monitored, and generally gave higher ee
values. The DMSO reductase enzymes, in R. capsulatus and
R. sphaeroides, had each been purified and were found to select-
ively deoxygenate the (S) enantiomer of methyl p-tolyl sulf-
oxide 20 yielding, exclusively, the (R) enantiomer.37–39 With
whole cells of either R. capsulatus or R. sphaeroides, and ensur-
ing that the biotransformation was stopped at ∼50% comple-
tion, it was also possible to obtain the (R)-sulfoxide 20 in
enantiopure form (>98% ee). Other strains, examined for com-
parison purposes, included E. coli HB 101 and P. vulgaris
NCIMB 67, that had earlier been found to preferentially
metabolize the (R) enantiomer of sulfoxide 20 yielding
enantiopure samples of the (S) sulfoxide.39 During the study,
the E. coli strain gave (S)-20 of lower enantiopurity (60% ee
after a 42% substrate conversion), while P. vulgaris showed
>98% ee (after 50% substrate conversion) for the residual
(S) sulfoxide, by chiral stationary phase HPLC (CSPHPLC)
analysis.
Scheme 7
P. putida UV4.22 However, biotransformation of racemic sulf-
oxide 18 followed by isolation of residual sulfoxide (ϩ)-(1R)-18
(72% ee), showed that kinetic resolution had occurred. In con-
trast with the other alkylaryl sulfoxides studied (10, 17cis and
17trans), only the biotransformation of sulfoxide 18 yielded an
isolable sample of the corresponding sulfide metabolite (19).
This is the first direct evidence of a reductase-catalysed
deoxygenation of an alkylaryl sulfoxide using P. putida UV4 (B
A, Scheme 2). The absolute configuration of (Ϫ)-sulfoxide
18 had been determined as (1S) by X-ray crystallography.22
The evidence, presented herein, confirms the recent prelimin-
ary conclusion that P. putida UV4 contains a reductase
enzyme.21 While this strain can deoxygenate both dialkyl and
alkylaryl sulfoxides, it also contains a dioxygenase enzyme that
can catalyse the reverse reaction on alkylaryl sulfides. The
reductase enzyme also showed a similar stereopreference
toward one enantiomer of the racemic bicyclic sulfoxides. Thus,
the residual sulfoxides 10 (3% ee, 1R), 17cis (47% ee, 1S), 17trans
(25% ee, 1S) and 18 (72% ee, 1R) had an enantiomeric excess
favouring the same absolute configuration (allowing for
Sequence Rule priorities). Although P. putida UV4 undoubt-
edly contains a reductase enzyme, the relatively slow rate of
deoxygenation, compared with the faster reverse process, would
limit its use in the kinetic resolution of racemic sulfoxides. In
view of this limitation, alternative strains of bacteria were
sought with minimal dioxygenase-catalysed sulfoxidation
activity and maximal sulfoxide reductase activity.
Three new bacterial strains, isolated from soil and marine
environments,41 were also found to deoxygenate racemic sulf-
oxide 20 in a stereoselective manner. These bacteria were
provisionally identified as C. braakii DMSO 11 (from canal
sediment, Coventry), Klebsiella sp. DMSO 7 (from the North
Sea), and Serratia sp. DMSO 10 (from the North Sea). The
Klebsiella DMSO 7 strain gave an excess of the residual
(R) sulfoxide 20 (54% ee after a 39% substrate conversion)
and thus a similar stereochemical preference to R. capsulatus
and R. sphaeroides. By contrast, an excess of the residual
(S) sulfoxide 20 was found after biotransformations with the
Serratia sp. DMSO 10 (>98% ee) and C. braakii DMSO 11
(32% ee).
(B) Enzyme-catalysed deoxygenation of racemic sulfoxides
using the anaerobic bacteria Rhodobacter capsulatus,
Escherichia coli, Proteus vulgaris, Citrobacter braakii,
Klebsiella sp. and Serratia sp.
Despite the well-established role of DMSO reductases in the
environment, relatively few studies have been carried out to
assess their value in the kinetic resolution of racemic sulf-
oxides.37–39 One study had shown that chiral sulfoxides, includ-
ing methyl p-tolyl sulfoxide 20, could be stereoselectively
deoxygenated to yield sulfide 21 using Rhodobacter sphaeroides
f.sp. denitrificans IL 106 (Scheme 8).37,38 Similar stereochemical
results from one of our laboratories were also reported with
R. capulatus DSM 938 and sulfoxide 20 as substrate.39 E. coli
ATCC 33694, and P. vulgaris NCIMB 67 strains were again
found to reduce sulfoxide 20 but with opposite enantio-
selectivity. Several enantiopure sulfoxides were thus obtained
by this kinetic resolution approach involving DMSO reductase-
catalysed deoxygenation.39 In our study, the use of reported,
and new, sulfoxide reductases in kinetic resolutions of sulf-
oxides has been evaluated with the help of a wider range of
anaerobic bacteria and substrates.
The DMSO reductase enzymes, present in R. capsulatus38,39
and R. sphaeroides,37 were also found to deoxygenate chiral di-
alkyl sulfoxides in a stereoselective manner. Thus, methylbenzyl
sulfoxide 22 was recovered with a 50% excess of the (R)
enantiomer (R. sphaeroides)37 and methylthiomethylmethyl sul-
foxide (MeSCH2SOMe) was recovered as a single enantiomer
of unknown configuration (R. capsulatus).39 When racemic
methylbenzyl sulfoxide 22 was used as substrate with five bac-
terial strains (Table 1), the residual dialkyl sulfoxide 22 was
consistently found to have the (S) configuration (26–77% ee) –
an opposite configuration to that found using R. sphaeroides.
It appears, from the results in Table 1, that the P. vulgaris
strain is among the most stereoselective strains for the
The commercially available racemic and enantiopure forms
of methyl p-tolyl sulfoxide 20 were, initially, selected as
model substrates for a range of facultative anaerobic bacteria.
R. capulatus DSM 938 was studied first, due to its capability to
Scheme 8
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
557

View Article Online
Table 1 Enantiopurity (% ee) and absolute configuration (Ab. con.) of residual sulfoxides 20 and 22 after stereoselective DMSO reductase-catalysed
deoxygenation to yield the corresponding sulfides 21 and 23
Sulfoxide 20
Sulfoxide 22
% ee
Ab. con.
% ee
Ab. con.
b
b
Rhodobacter capsulatus 938
Escherichia coli HB 101
Proteus vulgaris NCIMB 67
Klebsiella sp. DMSO 7
Serratia sp. DMSO 10
Citrobacter braakii DMSO 11
58 (>98)a,
60 (>98)a
>98 (>98)a
54
R (R)ab,
S (S)a
S (S)a
R
—
68
77
44
37
26
—
S
S
S
S
>98
S
32 (>95)b
S (S)b
S
a From earlier whole cell time course studies.39 b By using pure DMSO reductase.
Scheme 9
Scheme 10
Scheme 11
deoxygenation of both alkylaryl and dialkyl sulfoxides. This
strain was also found to yield enantiopure (CSPHPLC
analysis) ethyl-2-pyridyl sulfoxide 24 of undetermined absolute
configuration.39 Biotransformation (P. vulgaris) of racemic
ethyl-2-pyridyl sulfoxide 24 to yield sulfide 25 was repeated
on a larger scale (2.0 g); the residual sulfoxide 24 (45% recovery)
was found to have the (Ϫ)-(R) configuration (>98% ee) by
comparison of its optical rotation with the literature value
(Scheme 9).42
The new C. braakii DMSO 11 strain, appeared to deoxygen-
ate a wider range of racemic sulfoxides, and in a more stereo-
selective manner, than the other strains listed in Table 1. An
enantiopure sample of 2-methylsulfinyl phenol 26 was required
as a chiral ligand for a separate study of asymmetric alkylation
reactions of aldehydes. Initial attempts to obtain enantiopure
sulfoxide 26, via TDO- or NDO-catalysed sulfoxidation of the
parent sulfide, were unsuccessful. DMSO reductase enzymes,
present in P. vulgaris NCIMB 67 and R. capsulatus DSM
938, were used to partially resolve sulfoxide 26 (32% and 48%
ee respectively) via deoxygenation. However, whole cells of
C. braakii DMSO 11 deoxygenated, selectively, the (R) enan-
tiomer yielding enantiopure (Ϫ)-(S)-2-methylsulfinyl phenol 26
(Scheme 10). The configuration (S) was determined by methyl-
ation (CH2N2) and comparison of the CD spectrum of the
resulting 2-methoxy analogue with that of (Ϫ)-(S)-methyl-
phenyl sulfoxide 2.17
showed promise in the stereoselective deoxygenation of 1,4-di-
hydrobenzo-2,3-dithian-2-oxide 28 (Scheme 11). The residual
(ϩ)-thiosulfinate 28 was found to have a modest excess of the
(S) enantiomer (18% ee, 46% yield). The (ϩ)-(S) absolute
configuration for sulfoxide 28 had been determined by X-ray
crystallography of an enantiopure sample separated by semi-
preparative CSPHPLC.19
Fortunately the enzyme responsible for sulfoxide deoxygen-
ation reactions in C. braakii DMSO 11 was isolated, identified
and characterised as a periplasmic DMSO reductase.41 When
the purified DMSO reductase enzyme was used with racemic
methyl p-tolyl sulfoxide 20 and 1,4-dihydrobenzo-2,3-dithian-2-
oxide 28, as substrates, it was found that essentially only single
enantiomers of the residual sulfoxide (Ϫ)-(S)-20 (> 98% ee)
and thiosulfinate (ϩ)-(S)-28 (> 95% ee) remained and the
corresponding sulfides 20 and 29 were formed. This appears
to be the first example of the formation of an enantiopure
thiosulfinate via an enzyme-catalysed kinetic resolution of a
racemate.
Conclusion
Methods for the detection of a stereoselective reductase-
catalysed deoxygenation of sulfoxides, while operating simul-
taneously with an oxygenase-catalysed sulfoxidation of sulfides,
have been developed. Evidence has been gathered of enzyme-
catalysed: (i) sulfoxide deoxygenation to yield an isolable
sulfide, (ii) sulfoxide deoxygenation and benzylic monooxy-
genation to form a sulfide alcohol, (iii) sulfoxide deoxygenation
and sulfoxidation to yield a diastereoisomeric sulfoxide, and
(iv) sulfoxide dihydroxylation and deoxygenation to furnish a
cis-dihydrodiol sulfide. A remarkable enzyme-catalysed sulf-
oxidation-cis-dihydroxylation–deoxygenation sequence has
been observed during the biotransformation of an alkylaryl
sulfide to yield the corresponding cis-dihydrodiol.
Attempts were made to obtain enantiopure sulfoxide
(thiosulfinate) 28, as part of an earlier investigation of the
enzyme-catalysed asymmetric oxidation of 1,2-disulfides, using
monooxygenases, dioxygenases and peroxidases.19 Unfortu-
nately, these efforts were only partially successful in the produc-
tion of enantioenriched thiosulfinate 28 (11% ee with NDO and
58% ee using chloroperoxidase).19 As an alternative approach,
the racemic thiosulfinate 28 was added as substrate to the
anaerobic bacteria shown in Table 1; only C. braakii DMSO 11
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
558

View Article Online
The stereoselective reductase-catalysed deoxygenation
method, utilizing both established and new bacterial strains and
purified enzymes, has been used in kinetic resolution studies to
yield functionalised sulfoxide and thiosulfinate enantiomers
that are not readily available by the enzyme-catalysed asym-
metric sulfoxidation method.
o-xylene-α,αЈ-dithiol (4.2 g, 24.7 mmol) in acetone (250 cm3).
After stirring for 2 h, the mixture was filtered, concentrated and
purified by column chromatography (30% EtOAc in hexane) to
afford 1,4-dihydrobenzo-2,3-dithian 29 as a white crystalline
solid (3.7 g, 98%). Rf 0.68 (10% ethyl acetate in hexane); mp 70–
71 ЊC (CHCl3/hexane) (lit.,46 77–78 ЊC); (Found: C, 57.0; H, 4.7.
C8H8S2 requires C, 57.1; H, 4.8%); m/z (EI) 168 (Mϩ, 99%), 134
(34), 104 (100), 91 (14), 78 (51), 64 (24), 51 (23); δH (500 MHz,
CDCl3) 4.07 (4 H, s, (SCH2)2), 7.06–7.10 (2 H, m, 2 × Ar–H),
7.15–7.18 (2 H, m, 2 × Ar–H); δC (125 MHz, CDCl3) 34.57,
126.80, 130.14, 132.86.
Experimental
1NMR spectra of compounds were recorded on Bruker Avance
DPX-300 and DPX-500 instruments. Column chromatography
and PLC were performed on Merck Kieselgel type 60 (250–400
mesh) and PF254/366 respectively. Merck Kieselgel 60F254 analyti-
cal plates were used for TLC. Optical rotation ([α]D) measure-
ments were carried out with a Perkin-Elmer 214 polarimeter at
ambient temperature (ca. 20 ЊC) and are given in units of 10Ϫ1
deg cm2 gϪ1. CD spectra were recorded using a JASCO J-720
instrument and spectroscopic grade methanol as solvent.
Unless mentioned otherwise the enantiopurity of sulfoxides
was determined by CSPHPLC using specified Chiralcel
columns and 10% 2-propanol in hexane as eluent.
A solution of sodium periodate (0.12 g, 0.55 mmol) in water
(10 cm3) was added dropwise to a stirred solution of 1,4-di-
hydrobenzo-2,3-dithian 29 (0.084 g, 0.50 mmol) in methanol
(30 cm3) at 0 ЊC. The mixture was allowed to stir overnight and
then filtered. The filtrate was concentrated and extracted
(CHCl3); the extracts were dried (Na2SO4) and concentrated to
yield the crude product. Purification by column chromato-
graphy (30% ethyl acetate in hexane) gave 1,4-dihydrobenzo-
2,3-dithian-2-oxide 28 as a solid (2.47 g, 75%); Rf 0.18 (30%
ethyl acetate in hexane); mp 130–132 ЊC (CHCl3/hexane) (lit.,47
128–129 ЊC); (Found: C, 51.7; H, 4.3. C8H8OS2 requires C, 52.1;
Authentic samples of sulfides 1 (R = Me), 11, 16, 19, 21, 23,
25, 27 and 29 and racemic sulfoxides 2, 10, 17cis, 17trans, 18, 20,
22, 24, 26, 28 were available. Some were obtained commercially
(1, 2, 20, 21, 23, 27), others from earlier synthetic studies in
H, 4.4%);νmax/cmϪ1 1072 (S᎐O); m/z (EI) 184 (Mϩ, 9%), 168 (4),
᎐
136 (63), 135 (100), 104 (87), 91 (25), 78 (37); δH (500 MHz,
CDCl3) 3.96 (1 H, d, JA,B 13.5, CHAHBS(O)), 4.23 (1 H, d,
JA,B 13.1, CHAHBS), 4.32 (1 H, d, JB,A 13.1, CHAHBS), 4.37
(1 H, d, JB,A 13.5, CHAHBS(O)), 7.37–7.43 (4 H, m, 4 × Ar–H);
δC (125 MHz, CDCl3) 33.04, 59.69, 127.73, 127.86, 128.47,
129.34, 131.76, 135.56.
these laboratories (10,22 11,22 16,22 17cis,22 17trans 22 18,22 19,22 22,22
,
24,39 25,39) and the rest were synthesized by literature methods
(26–29). All sulfoxide substrates showed a characteristic strong
S᎐O group absorption in the IR spectra (νmax/cmϪ1 1023–1094).
᎐
Synthesis of 2-(methylsulfinyl)phenol 26
Typical biotransformations of racemic sulfoxides using aerobic
and anaerobic bacteria
A stirred solution of 2-(methylsulfanyl)phenol 27 (5 g, 35.7
mmol), in acetone (50 cm3) maintained at 0 ЊC, was oxidised
by the dropwise addition of an acetone solution of dimethyl-
dioxirane (0.08 M) prepared in accordance with the literature
procedure.43 The progress of the oxidation reaction was fol-
lowed by TLC analysis (60% EtOAc in hexane) and upon com-
pletion, acetone was evaporated off, the residue dried (vaccum
pump) and crystallised, to furnish sulfoxide 26 (5.30 g, 95%) as
a colourless crystalline solid; mp 123 ЊC (from MeOH) (lit.44
127–128 ЊC); Rf 0.23 (5% MeOH in CHCl3); δH (500 MHz,
CDCl3) 2.96 (3 H, s, Me), 6.93 (1 H, m, Ar), 6.97 (1 H, d, J 7.7,
Ar), 7.07 (1 H, d, J 7.8, Ar), 7.37 (1 H, m, Ar); δC (125 MHz,
CDCl3) 41.9, 119.71, 119.86, 123.07, 124.83, 133.05; m/z (EI)
156 (Mϩ, 100%), 141 (95%).
(i) Aerobic bacteria: P. putida UV4. Biotransformations
with P. putida UV4 were carried out using the shake-flask
conditions described earlier for the sulfoxidation of both
acyclic and cyclic sulfides.8,17,18,22 The biotransformation times
involved were in some cases modified as described in the
text. The improved work-up procedure involved removal of
most of the water, from the aqueous centrifuged culture
medium, under reduced pressure and relatively low temper-
atures (≤ 40 ЊC) and repeated extraction of the residual viscous
concentrate with EtOAc. The combined extracts were dried
(Na2SO4) and the solvent evaporated to yield the crude mixture
of bioproducts.
Biotransformation of racemic methylphenyl sulfoxide 2 (R =
Me) using P. putida UV4. Biotransformation of sulfoxide 2
(R = Me, 0.2 g, 1.43 mmol) followed by concentration, extrac-
tion (EtOAc), and separation by PLC on silica gel (8% MeOH
in CHCl3; two elutions) gave cis-dihydrodiol sulfide 8, cis-
dihydrodiol sulfone 9 and an inseparable mixture of the corre-
sponding cis-dihydrodiol sulfoxides 7a/7b.
(1S,2S)-1,2-Dihydroxy-3-methylsulfanylcyclohexa-3,5-diene
8. Light yellow coloured solid (0.01 g, 4%); Rf 0.10 (5% MeOH
in CHCl3); mp 58–60 ЊC (lit.,17 mp 57–61 ЊC); [α]D ϩ63 (c 1.29,
MeOH); (Found: Mϩ, 158.0399. C7H10O2S requires 158.0402);
m/z (EI) 158 (Mϩ, 20%), 140 (100), 125 (41), 97 (66), 53 (42),
39 (60), 29 (47); δH (500 MHz, CDCl3) 1.94 (1 H, br s, OH), 2.33
(3 H, s, Me), 2.47 (1 H, br s, OH), 4.18–4.26 (1 H, dd, JOH,2 8.3,
J2,1 5.9, 2-H), 4.29 (1 H, m, 1-H), 5.52 (1 H, d, J4,5 5.7, 4-H),
5.81–5.84 (1 H, dd, J6,5 9.5, J6,1 4.1, 6-H), 6.00–6.04 (1H, dd,
J5,6 9.4, J5,4 5.7, 5-H); δC (125 MHz, CDCl3) 14.08, 68.20, 71.50,
113.43, 124.59, 125.36, 142.30.
(1S,2S)-1,2-Dihydroxy-3-methylsulfonylcyclohexa-3,5-diene
9. Colourless crystals (0.020 g, 6.0%); Rf 0.3 (5% MeOH in
CHCl3); mp 126–128 ЊC (from CHCl3/(CH3)2CO) (Found: C,
44.1; H, 5.4. C7H10O4S requires C, 44.20; H, 5.30%); [α]D Ϫ5.3
(c 0.73, MeOH); (Found: Mϩ, 190.0291. C7H10O4S requires
190.0300); δH (500 MHz, CDCl3) 2.75–2.95 (2 H, m, OH), 3.08
(3 H, s, Me), 4.58 (1 H, dd, J1,2 6.3, J1,6 2.4,1-H), 4.64 (1 H, d,
Synthesis of 1,4-dihydrobenzo-2,3-dithian 29 and racemic
1,4-dihydrobenzo-2,3-dithian-2-oxide 28
α,αЈ-Dibromo-o-xylene (6.6 g, 25.0 mmol) and thiourea (4.75 g,
62.4 mmol) were dissolved in ethanol (62.5 cm3) and the solu-
tion refluxed for 6 h. The ethanol was removed under reduced
pressure and the remaining solid was dissolved in water (125
cm3). A solution of NaOH (4 g) dissolved in water (37.5 cm3)
was added and the reaction mixture refluxed for a further 6 h.
After cooling and addition of sulfuric acid (2 M) until neutral
pH, the aqueous phase was extracted (CH2Cl2). The organic
extracts were dried (Na2SO4) and concentrated under reduced
pressure to give crude o-xylene-α,αЈ-dithiol as a malodorous
low melting solid (4.2 g, 100%). It was identified spectrally and
used directly without further purification. Rf 0.10 (10% EtOAc
in hexane); mp 33–35 ЊC (lit.,45 44 ЊC); (Found: Mϩ, 170.0228.
C8H10S2 requires 170.0224) m/z (EI) 170 (Mϩ, 40%), 137 (100),
123 (80), 104 (97), 90 (47), 76 (47), 63 (82), 51 (89), 45 (97), 39
(85); δH (500 MHz, CDCl3), 1.85 (2 H, t, J1,2 7.1, 2 × SH), 3.86
(4 H, d, J2,1 7.1, 2 × CH2SH), 7.20–7.23 (2 H, m, 2 × Ar–H),
7.27–7.29 (2 H, m, 2 × Ar–H); δC (125 MHz, CDCl3) 25.94,
127.92, 129.75, 138.71.
Potassium permanganate/copper sulfate pentahydrate
catalyst (10.5 g, 1 : 1) was added to a stirred solution of
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
559

View Article Online
J2,1 6.3, 2-H), 6.17 (1 H, m, J5,4 5.5, J5,6 9.5 5-H), 6.25 (1 H, dd,
J6,5 9.5, J6,1 2.4, 6-H), 7.05 (1H, d, J4,5 5.5, 4-H).
Biotransformation of racemic trans-2-methyl-1,3-benzo-
dithiole-1-oxide 17trans by P. putida UV4. Racemic 2-methyl-
1,3-benzodithiole-1-oxide 17trans (0.20 g, 1.09 mmol) yielded
(1S,2S)-1,2-Dihydroxy-3-(R and S)-methylsulfinyl-cyclo-
hexa-3,5-diene (7a and 7b). Colourless viscous oil (0.1 g, 40%,
56 : 44); Rf 0.35 (15% MeOH in CHCl3); δH (300 MHz, CDCl3)
2.81 (3 H, s, Me7a), 2.83 (3 H, s, Me7b), 3.65 (1 H, br s, OH), 3.85
(1 H, br s, OH), 4.03 (2 H, br s, 2 × OH), 4.34 (1 H, dd, J1,2 6.0,
J1,6 2.8, 1-H7a), 4.43 (1 H, d, J1,2 5.9, 1-H7b), 4.53 (1 H, d, J2,1 5.9,
2-H7b), 4.60 (1 H, d, J2,1 6.0, 2-H7a), 6.09–6.19 (4 H, m, 5-H7a,
5-H7b, 6-H7a, 6-H7b), 6.51 (1 H, m, 4-H7a), 6.60 (1 H, m,
4-H7b).Biotransformation, of methyl phenyl sulfide 1 (R = Me,
0.2 g, 1.6 mmol), was repeated but using a higher cell density
and a longer biotransformation time (20 h) compared with the
earlier experiment.17 Using the improved work-up procedure,
(1S,2S)-1,2-dihydroxy-3-methylsulfanyl-cyclohexa-3,5-diene 8
(0.06 g, 24%) was isolated as the sole metabolite. When the
latter biotransformation was repeated again but using a shorter
(8 h) biotransformation period, a pure sample of (1S,2S)-1,2-
dihydroxy-3-(R)-methylsulfinyl-cyclohexa-3,5-diene 7a was
intercepted (80% yield), [α]D ϩ200 (c 1.36, MeOH); (Found:
recovered
(Ϫ)-(1S,2S)-2-methyl-1,3-benzodithiole-1-oxide
17trans (0.069 g, 35%), (Ϫ)-(1S,2R)-cis-2-methyl-1,3-benzo-
dithiole-1-oxide 17cis (0.002 g, 1%) after PLC separation
(diethyl ether).
(Ϫ)-(1S,2S)-2-Methyl-1,3-benzodithiole-1-oxide 17trans
. Rf
0.36 (diethyl ether); [α]D Ϫ32.2 (c 1.41, CHCl3) (lit.,48 Ϫ28 in
EtOH, 25% ee); ee 18%, Chiralcel OD column (α = 1.1, late
[1S,2S]).
(Ϫ)-(1S,2R)-2-Methyl-1,3-benzodithiole-1-oxide 17cis. Rf
0.46 (diethyl ether); ee > 98%, Chiralcel OD column (α = 1.2,
late [1S,2R]).
Biotransformationofracemic1,2,3,4-tetrahydro-2H-1,5-benzo-
dithiepine-1-oxide 18 by P. putida UV4. Racemic 1,2,3,4-
tetrahydro-2H-1,5-benzodithiepine-1-oxide 18 (0.131 g, 0.66
mmol)
yielded
(ϩ)-(R)-1,2,3,4-tetrahydro-2H-1,5-benzo-
dithiepine-1-oxide 18R (0.004 g, 6%) and 3,4-dihydro-2H-1,5-
benzodithiepine 19 (0.002 g, 2%), after separation by column
chromatography (CHCl3).
Mϩ,174.0355 C7H10OS requires 174.0351); νmax/cmϪ1 1098 (S᎐
᎐
O); m/z: 174 (Mϩ, 38), 157 (22), 111(90); CD data (MeOH)
λ/nm 214 (∆ε Ϫ2.50) and 278 (∆ε 2.56). Addition of the
(1S,2S)-cis-diol-3-(R)-sulfoxide 7a (0.25 g, 1.44 mmol) as sub-
strate to P. putida UV4 yielded a mixture of cis-diol sulfide 8
and cis-diol sulfone 9 which proved to be identical to the
samples obtained when sulfoxide 2 was used as the substrate.
Biotransformation of racemic 2,3-dihydrobenzo[b]thiophene-
1-oxide 10 using P. putida UV4. Racemic 2,3-dihydrobenzo[b]-
thiophene-1-oxide 10 (0.20 g, 1.32 mmol) yielded unreacted
(Ϫ)-(R)-2,3-dihydrobenzo[b]thiophene-1-oxide 10 (0.038 g,
19%), (ϩ)-3-(S)-hydroxy-2,3-dihydrobenzo[b]thiophene 12
(0.009 g, 5%) which were separated, by column chromato-
graphy (CHCl3), from a mixture (0.01 g) of dihydrodiols, 2,3-
dihydro-2,3-dihydroxybenzo[b]thiophene 15cis, 2,3-dihydro-2,3-
dihydroxybenzo[b]thiophene 15trans and (4R,5S)-4,5-dihydro-
(ϩ)-(R)-1,2,3,4-Tetrahydro-2H-1,5-benzodithiepine-1-oxide
18R. [α]D ϩ52.0 (c 0.28, CHCl3) (lit.,22 Ϫ82, 93% ee); ee 72%,
Chiralcel OB column (30% 2-propanol in hexane, α = 1.3, late
R).
3,4-Dihydro-2H-1,5-benzodithiepine 19. Metabolite 19 was
spectrally identical to an authentic sample.
(ii) Biotransformations using the anaerobic bacteria:
R. capsulatus 938, E. coli HB 101, P. vulgaris NCIMB 67,
Klebsiella sp. DMSO 7, Serratia sp. DMSO 10 and C. braakii
DMSO 11. Typical procedure. The bacterium was grown anaer-
obically in a supplemented basal salts medium. DMSO
(40 mmol) was added as the electron donor and glucose (5 g
L
Ϫ1) as the carbon source for growth. Bacteria were grown
in Suba-sealed flasks (250 cm3) containing growth medium
(200 cm3). The cultures were flushed with nitrogen and incu-
bated at 30 ЊC on a rotary shaker (250 rev minϪ1) for 16 h.
Following growth, bacteria were harvested by centrifugation
(5000 rpm for 15 min) and washed in potassium phosphate
buffer (25 mmol, pH 7.4). The cell pellets were then re-
suspended in fresh phosphate buffer to give a 10-fold concen-
tration of the initial cell density. The cell suspension (20 cm3
from a 200 cm3 culture) was transferred to a Suba-sealed flask
(50 cm3 growth medium) containing racemic sulfoxide (1 mg
cmϪ3) and glucose (10 mmol) under nitrogen. The reaction
flasks were incubated at 30 ЊC on a rotary shaker (250 rev
minϪ1) until about 50% of the sulfoxide substrate had been
deoxygenated. After centrifugation (5000 rpm, 15 min), the
metabolites from the remaining aqueous supernatant, were
4,5-dihydroxybenzo[b]thiophene 14. Metabolites 15cis, 15trans
,
1
and 14, although not separated, were identified by H NMR
spectral comparisons with samples isolated earlier from the
biotransformation of benzo[b]thiophene.36
(Ϫ)-(R)-2,3-Dihydrobenzo[b]thiophene-1-oxide 10R. Ee 3%,
Chiralcel OB column (30% 2-propanol in hexane, α = 1.8, early
R).
(ϩ)-3-(S)-Hydroxy-2,3-dihydrobenzo[b]thiophene 12. Col-
ourless viscous oil; Rf 0.41 (2% MeOH in CHCl3); [α]D ϩ18
(c 0.61, CHCl3); (Found: Mϩ, 152.0293. C8H8OS requires
152.0296); m/z (EI) 152 (Mϩ, 100%), 135 (56), 122 (26), 121
(37), 91 (46), 77 (41), 57 (88), 45 (51); δH (500 MHz, CDCl3)
2.08 (1H, d, JOH,3 7.4, OH), 3.31 (1 H, dd, J2A,2B 12.0, J2A,3 3.8,
CH2), 3.60 (1 H, dd, J2B,2A 12.0, J2B,3 6.2, CH2), 5.36 (1 H, m,
3-H), 7.10–7.13 (1 H, ddd, J 7.8, J 4.8, J 3.3, 7-H), 7.25–
7.26 (2 H, m, 5 and 6-H), 7.38–7.40 (1 H, d, J 7.6, 4-H);
δC (125 MHz, CDCl3) 41.53, 70.58, 122.81, 124.75, 125.31,
129.72, 140.85, 140.93; ee 83%, Chiralcel OD column (α = 1.1,
early S).
Biotransformation of racemic cis-2-methyl-1,3-benzodithiole-
1-oxide 17cis by P. putida UV4. Racemic cis-2-methyl-1,3-benzo-
dithiole-1-oxide 17cis (0.113 g, 0.62 mmol) gave trans-2-methyl-
1,3-benzodithiole-1-oxide 17trans (0.003 g, 2%), and residual
(Ϫ)-(1S,2R)-cis-2-methyl-1,3-benzodithiole-1-oxide 17cis (0.045
g, 40%) after separation by PLC (diethyl ether). Sulfoxides 17cis
and 17trans were spectrally indistinguishable from authentic
samples obtained earlier.
extracted in
a similar manner to the P. putida UV4
biotransformations.
Biotransformation of racemic 2-(methylsulfinyl) phenol 26
by C. braakii DMSO 11. Sulfoxide 26 (0.5 g), was used as a
substrate and progress of the biotransformation was monitored
by RP-HPLC (30% H2O in MeOH). The biotransformation
was terminated after ∼50% of the substrate had been metabol-
ised to the corresponding sulfide 27. The crude metabolite
was purified by PLC (5% MeOH in CHCl3) to give (Ϫ)-(S)
sulfoxide 26 (0.135 g, 27%); [α]D Ϫ189 (c 0.5, CHCl3)
(lit.49 ϩ 188, CHCl3); ee > 98%, Chiralcel OB column (α = 1.60,
early S).
Biotransformation of racemic 1,4-dihydrobenzo-2,3-dithian-2-
oxide 28 by C. braakii DMSO 11. Biotransformation of 1,4-di-
hydrobenzo-2,3-dithian-2-oxide 28 (0.25 g, 1.36 mmol) fol-
lowed by ethyl acetate extraction of the centrifuged medium
yielded residual (ϩ)-(S)-1,4-dihydro-2,3-benzodithian-2-oxide
28S (0.058 g, 46%) after purification by PLC (3% MeOH in
CHCl3); [α]D ϩ43.1 (c 0.56, CHCl3), ee 18% by CSPHPLC,
Whelk-01 column (t-BuOMe, α = 2.8, early S).
(Ϫ)-(1S,2R)-2-Methyl-1,3-benzodithiole-1-oxide 17cis. Rf
0.46 (diethyl ether); [α]D Ϫ165 (c 3.0, CHCl3) (lit.,48 Ϫ69, in
EtOH, 25% ee); ee 47%, Chiralcel OD column (α = 1.2, late
[1S,2R])
(Ϫ)-(1S,2S)-2-Methyl-1,3-benzodithiole-1-oxide 17trans
. Rf
0.36 (diethyl ether); ee > 98%, Chiralcel OD column (α = 1.2,
early[1S,2S])
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
560

View Article Online
20 D. R. Boyd, N. D. Sharma, S. A. Haughey, J. F. Malone,
The 1,4-dihydrobenzo-2,3-dithian-2-oxide 28 (0.005 g, 0.027
mmol) was used as substrate with purified DMSO reductase
from C. braakii DMSO 11.41 Ethyl acetate extraction, of the
centrifuged medium, yielded unreacted (ϩ)-(S)-1,4-dihydro-
2,3-benzodithian-2-oxide 28S and 1,4-dihydrobenzo-2,3-dithian
29. Separation and purification by PLC (3% MeOH in CHCl3)
gave thiosulfinate 28S (0.002 g, 40%, > 95% ee) and 1,2-disulfide
29 (0.0005 g, 10%).
Biotransformation of racemic ethyl-2-pyridyl sulfoxide 24 by
P. vulgaris NCIMB 67. The aqueous mixture of bioproducts,
obtained when racemic ethyl-2-pyridyl sulfoxide 24 (2.0 g) was
metabolised with P. vulgaris NCIMB 67, was extracted with
EtOAc. PLC of the crude product yielded (Ϫ)-(R)-ethyl-2-
pyridyl sulfoxide 24 (0.90 g, 45%); [α]D Ϫ164 (c 1.6, MeOH)
(lit.42 Ϫ168, MeOH); ee > 98%, Chiralcel OB column (α = 1.95).
B. T. McMurray, G. N. Sheldrake, C. C. R. Allen and H. Dalton,
J. Chem. Soc., Chem. Commun., 1996, 2363.
21 D. R. Boyd, N. D. Sharma, N. Gunaratne, S. A. Haughey,
M. A. Kennedy, J. F. Malone, C. C. R. Allen and H. Dalton, Org.
Biomol. Chem., 2003, 1, 984.
22 D. R. Boyd, N. D. Sharma, S. Haughey, M. A. Kennedy,
J. F. Malone, S. Shepherd, C. C. R. Allen and Howard Dalton,
Tetrahedron, 2004, 60, 549.
23 H. Kexel and H. C. Schmidt, Biochem. Pharmacol., 1972, 212, 1009.
24 M. V. Hand, S. Kelly, R. A. More O’Ferrall, S. N. Rao,
N. D. Sharma, G. N. Sheldrake and H. Dalton, J. Chem. Soc., Chem.
Commun., 1994, 313.
25 D. R. Boyd, R. A. S. McMordie, N. D. Sharma, H. Dalton, R. O.
Jenkins and P. Williams, J. Chem. Soc. Chem. Commun., 1989, 339.
26 D. R. Boyd, M. R. J. Dorrity, J. F. Malone, R. A. S. McMordie,
N. D. Sharma, H. Dalton and P. Williams, J. Chem. Soc.,
Perkin Trans. 1, 1990, 489.
27 R. Agarwal, D. R. Boyd, R. A. S. McMordie, G. A. O’Kane,
P. Porter, N. D. Sharma, H. Dalton and D. J. Gray, J. Chem. Soc.,
Chem. Commun., 1990, 1711.
Acknowledgements
28 D. R. Boyd, N. D. Sharma, P. J. Stevenson, J. Chima, D. J. Gray and
H. Dalton, Tetrahedron Lett., 1991, 32, 3887.
29 D. R. Boyd, N. D. Sharma, R. Boyle, J. F. Malone, J. Chima and
H. Dalton, Tetrahedron: Asymmetry, 1993, 4, 1307.
30 R. Agarwal, D. R. Boyd, N. D. Sharma, H. Dalton, N. A. Kerley,
R. A. S. McMordie, G. N. Sheldrake and P. Williams, J. Chem. Soc.,
Perkin Trans., 1, 1996, 67.
We thank the BBSRC, DTI, Avecia Pharmaceuticals, Astra-
Zeneca and Oxford Asymmetry for support under the LINK
Applied Biocatalysis Scheme (NDS, HL) and DENI for post-
graduate studentships (SDS, AWTK).
31 D. R. Boyd, N. D. Sharma, N. I. Bowers, P. A. Goodrich,
M. R. Groocock, A. J. Blacker, D. A. Clarke, T. Howard and
H. Dalton, Tetrahedron: Asymmetry, 1996, 7, 1559.
32 N. I. Bowers, D. R. Boyd, N. D. Sharma, P. A. Goodrich, M. R.
Groocock, A. J. Blacker, D. A. Clarke, P. Goode and H. Dalton,
J. Chem. Soc., Perkin Trans. 1, 1999, 1453.
33 D. R. Boyd, N. D. Sharma, N. I. Bowers, R. Boyle, J. Duffy,
J. S. Harrison and H. Dalton, J. Chem. Soc., Perkin Trans. 1, 2000,
1345.
References
1 I. Fernandez and N. Khiar, Chem. Rev., 2003, 103, 3651.
2 B. J. Auret, D. R. Boyd, H. B. Henbest and S. Ross, J. Chem. Soc. C,
1968, 2371.
3 E. Abushanab, D. Reed, F. Suzuki and G. J. Sih, Tetrahedron Lett.,
1978, 19, 3415.
4 H. L. Holland, C. G. Rand, P. Viski and F. M. Brown, Can. J. Chem.,
1991, 69, 1989.
5 H. L. Holland, F. M. Brown and B. G. Larsen, Tetrahedron:
Asymmetry, 1994, 5, 1241.
6 H. Ohta, Y. Okamoto and G. Tsuchihashi, Chem. Lett., 1984,
205.
7 Y. Okamoto, H. Ohta and G. Tsuchihashi, Chem. Lett., 1986,
2049.
8 J. R. Cashman, L. D. Olsen, D. R. Boyd, R. A. S. McMordie,
R. Dunlop and H. Dalton, J. Am. Chem. Soc., 1992, 114,
8772.
9 C. C. R. Allen, D. R. Boyd, H. Dalton, N. D. Sharma, S. A.
Haughey, R. A. S. McMordie, B. T. McMurray, G. N. Sheldrake and
K. Sproule, J. Chem. Soc., Chem. Commun., 1995, 119.
10 K. Lee, J. M. Brand and D. T. Gibson, Biochem. Biophys. Res.
Commun., 1995, 212, 9.
11 S. Colonna, N. Gaggero, G. Carrea and P. Pasta, Tetrahedron:
Asymmetry, 1996, 7, 565.
34 D. R. Boyd, N. D. Sharma, N. I. Bowers, R. Boyle, J. S. Harrison,
K. Lee and D. T. Gibson, Org. Biomol. Chem., 2003, 1, 1298.
35 D. R. Boyd, N. D. Sharma, R. Boyle, B. T. McMurray, T. A. Evans,
J. F. Malone, H. Dalton, J. Chima and G. N. Sheldrake, J. Chem.
Soc., Chem. Commun., 1993, 49.
36 D. R. Boyd, N. D. Sharma, I. N. Brannigan, D. A. Clarke,
H. Dalton, S. A. Haughey and J. F. Malone, J. Chem. Soc., Chem.
Commun., 1996, 2361.
37 M. Abo, M. Tachibana, A. Okubo and S. Yamazaki, Bioorg. Med.
Chem., 1995, 3, 109.
38 M. Abo, A. Okubo and S. Yamazaki, Tetrahedron: Asymmetry,
1997, 8, 345.
39 S. P. Hanlon, D. L. Graham, P. J. Hogan, R. A. Holt, C. D. Reeve
and A. L. McEwan, Microbiology, 1998, 144, 2247.
40 A. S. McAlpine, A. G. McEwan and S. Bailey, J. Mol. Biol., 1998,
275, 613.
12 D. R. Kelly, C. J. Knowles, J. G. Mahdi, I. N. Taylor and
M. A. Wright, Tetrahedron: Asymmetry, 1996, 7, 365.
13 V. Alphand, N. Gaggero, S. Colonna, P. Pasta and R. Furstoss,
Tetrahedron, 1997, 53, 9695.
14 M. Andersson, A. Willetts and S. Allenmark, J. Org. Chem., 1997,
62, 8455.
41 H. R. Luckarift, H. Dalton, D. R. Boyd, N. D. Sharma and
R. A. Holt, manuscript in preparation.
42 J. Zhu, Y. Qin, Z-He, F. Fu, Z. Zhou, J. Deng, Y. Jiang and T. Chau,
Tetrahedron: Asymmetry, 1997, 15, 2505.
43 W. Adam, J. Bialis and L. Hadjiarapoglou, Chem. Ber., 1991, 124,
2377.
15 M. Andersson and S. Allenmark, Tetrahedron, 1998, 54, 15293.
16 A. Kerridge, A. Willetts and H. Holland, J. Mol. Catal., B: Enzym.,
1999, 6, 59.
17 D. R. Boyd, N. D. Sharma, S. A. Haughey, M. A. Kennedy,
B. T. McMurray, G. N. Sheldrake, C. C. R. Allen, H. Dalton and
K. Sproule, J. Chem. Soc., Perkin Trans. 1, 1998, 1, 1929.
18 D. R. Boyd, N. D. Sharma, S. A. Haughey, J. F. Malone, A. King,
B. T. McMurray, R. Holt and H. Dalton, J. Chem. Soc.,
Perkin Trans. 1, 2001, 3288.
44 A. E. Sopchik and C. A. Kingsbury, J. Chem. Soc., Perkin Trans. 2,
1979, 1058.
45 J. J. Mayerle, S. E. Denmark, B. V. DePamphilis, J. A. Ibers and
R. H. Holm, J. Am. Chem. Soc., 1975, 97, 1032.
46 B. Milligan and J. M. Swan, J. Chem. Soc., 1965, 2901.
47 P. K. Singh, L. Field and B. J. Sweetman, Phosphorus Sulfur Relat.
Elem., 1988, 39, 61.
48 D. R. Boyd, N. D. Sharma, J. H. Dorman, R. Dunlop, J. F. Malone,
R. A. S. McMordie and A. F. Drake, J. Chem. Soc., Perkin Trans. 1,
1992, 1105.
19 D. R. Boyd, N. D. Sharma, B. E. Byrne, S. D. Shepherd, V. Ljubez,
C. C. R. Allen, L. A. Kulakov, M. L. Larkin and H. Dalton,
Chem. Commun., 2002, 1914.
49 G. Wagner and S. Boehme, Arch. Pharm. Ber. Deutsch., Pharm. Ges.,
1964, 297, 257.
O r g . B i o m o l . C h e m . , 2 0 0 4 , 2, 5 5 4 – 5 6 1
561