Enantioselective reduction and deracemisation using the non-conventional yeast Pichia glucozyma in water/organic solvent biphasic systems: Preparation of (S)-1,2-diaryl-2-hydroxyethanones (benzoins)

Article abstract of DOI:10.1016/j.tet.2011.11.014

Water/organic solvent two-liquid-phase systems have been successfully applied in the synthesis of enantiomerically pure (S)-benzoin through two different methodologies catalysed by whole cells from the non-conventional yeast Pichia glucozyma: the stereoselective monoreduction of benzil and the deracemisation of benzoin. The presence of the organic solvent influences the redox systems implied in the reactions, avoiding the formation of the corresponding diols, increasing the enantioselectivity and allowing the easy isolation of the products in high yields and excellent enantiomeric excesses. The use of both strategies has been extended to the preparation of different chiral benzoin derivatives.

Full text of DOI:10.1016/j.tet.2011.11.014

Tetrahedron 68 (2012) 523e528
Contents lists available at SciVerse ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Enantioselective reduction and deracemisation using the non-conventional
yeast Pichia glucozyma in water/organic solvent biphasic systems: preparation
of (S)-1,2-diaryl-2-hydroxyethanones (benzoins)
,
b
b
a
c
b,
Maria Caterina Fragnelli a , Pilar Hoyos , Diego Romano , Raffaela Gandolfi , Andres R. Alcantara
*
,
ꢀ
ꢀ
,
Francesco Molinari a
*
a
ꢁ
Dipartimento di Scienze e Tecnologie Alimentari e Microbiologiche, Universita degli Studi di Milano, Via Mangiagalli 25, Milano 20133, Italy
Departamento de Química Organica y Farmaceutica, Universidad Complutense de Madrid, Plaza de Ramon y Cajal s/n, Madrid 28040, Spain
Dipartimento di Scienze Molecolari Applicate ai Biosistemi, Universita degli Studi di Milano, Via Venezian 21, Milan 20133, Italy
b
ꢀ
ꢀ
ꢀ
c
ꢁ
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 July 2011
Received in revised form 17 October 2011
Accepted 7 November 2011
Available online 13 November 2011
Water/organic solvent two-liquid-phase systems have been successfully applied in the synthesis of
enantiomerically pure (S)-benzoin through two different methodologies catalysed by whole cells from
the non-conventional yeast Pichia glucozyma: the stereoselective monoreduction of benzil and the de-
racemisation of benzoin. The presence of the organic solvent influences the redox systems implied in the
reactions, avoiding the formation of the corresponding diols, increasing the enantioselectivity and al-
lowing the easy isolation of the products in high yields and excellent enantiomeric excesses. The use of
both strategies has been extended to the preparation of different chiral benzoin derivatives.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Benzoin
Deracemisation
Pichia glucozyma
Benzil
Dehydrogenase
Stereoselective reduction
Two-liquid-phase system
1. Introduction
the preparation of 2-amino-1,2-diphenylethanol which, in its op-
tically pure form, has recently received widespread attention due to
Homochiral benzoins (1,2-diaryl-2-hydroxyethanone struc-
tures) can be chemically obtained through benzoin condensation,
one of the most traditional CeC bond forming reactions, catalysed
by chiral thiazolium and triazolium salts1,2 or chiral metal-
lophosphites.3 Alternatively, biocatalytic methods can offer an at-
tractive platform by exploiting different enzymatic reactions,4 such
as benzoin condensation catalysed by thiamine-diphosphate-
dependent enzymes,5e8 dynamic kinetic resolution of racemic
the use of these amino alcohols as chiral auxiliaries in asymmetric
synthesis,21 chiral stationary phases for HPLC applications22 and as
ligands in asymmetric catalysis.23
Microbial dehydrogenases bound to whole cells have been used
for the stereoselective reductions of aromatic
a-diketones, with
formation of the desired chiral -hydroxyketone as major product,
a
and in some cases to a full reduction to the corresponding diol,
most likely because the whole cell contains other competing
ketoreductases. In fact, the reduction of benzil (1) was firstly re-
ported with the most classical bio-reduction biocatalyst (baker’s
yeast, whole cells from Saccharomyces cerevisiae), although the
reaction proceeded with no enantioselectivity.24 More recent
studies with this same biocatalyst show the production of (R)-
benzoin ((R)-2) with 50% enantiomeric excess, while the use of
organic co-solvents led to higher enantiopurity values.13 Other
microorganisms have also been described for this same process
with more success,11,12,14,15 but often these biotransformations are
hampered by the low rates reported. The monoreduction of dif-
ferent 1,2-diarylethanediones (including benzil) with lyophilised
benzoins,9,10 enantioselective reduction of
a
-diketones,11e16 or
deracemisation of racemic benzoins using redox processes cata-
lysed by whole cells.17
Optically pure benzoin is considered a useful building block in
the synthesis of different drugs and fine chemicals due to its
functional nature.4,18e20 More concretely, benzoin can be used for
* Corresponding authors. Tel.: þ39 (0)2 5031 9148; fax: þ39 (0)2 5031 6694
(F.M.); tel.: þ34 91 394 1820; fax: þ34 91 394 1822 (A.R.A.); e-mail addresses:
ꢀ
andresrf@farm.ucm.es (A.R. Alcantara), francesco.molinari@unimi.it (F. Molinari).
0040-4020/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2011.11.014

524
M.C. Fragnelli et al. / Tetrahedron 68 (2012) 523e528
whole cells from Pichia glucozyma CBS 5766 proved to be fast and
highly regio- and enantioselective in most of the cases; (S)-benzoin
(S)-2 was obtained with total conversion within 3 h, but with
a moderate enantioselectivity (ee 75%).16 It should be noted that
this yeast has shown interesting performances in the stereo-
selective reduction of different aromatic ketones.25,26
Another biocatalytic approach for obtaining homochiral benzoin
is the deracemisation of racemic benzoin. In this way, Rhizopus
oryzae was able to catalyse the chirality inversion of benzoins, to
afford both enantiomers depending on the pH of the medium.17 The
formation of both enantiomers using different reaction conditions
suggests the occurrence of ketoreductases acting on benzoin/benzil
with different enantioselectivity, displaying the highest activity
under different pH conditions.
1
2
3
e.e. (S)-2
100
80
60
40
20
0
100
80
60
40
20
0
0
2
4
6
8
10
Time (h)
Fig. 1. Reaction progress of the reduction of 1a catalysed by untreated whole cells from
P. glucozyma CBS 5766 in phosphate buffer.
As a part of our ongoing research focused on the biocatalytic
synthesis of
a
-hydroxyketones,16,9 in this work we have studied
two biocatalytical methods for producing (S)-benzoin using a new
organic/aqueous biphasic system: the first one is the enantiose-
lective bio-reduction of benzil, while the second one is the dera-
cemisation of racemic benzoin. It has been previously described
that the use of two-liquid-phase systems increases the solubility of
organic substrates, allowing the achievement of higher conver-
sions, as well as an easier isolation of the products;27,28 moreover, it
is known that the use of organic co-solvents in microbial reductions
considerably reduces the rate of undesired reactions, including
reduction catalysed by enzymes with different stereo-
selectivity.27e30 Both biotransformations are catalysed by enzy-
matic redox systems occurring in the non-conventional yeast P.
glucozyma.
R-enantiomer. The reduction to 3a was completed after 10 h and
the isolated diol resulted a 35:65 mixture of (R,R)-3a and (R,S)-3a.
These results suggest that different dehydrogenases active on 1a
and 2a are present in the whole cells. The overall activity of these
enzymes can be influenced by the process of lyophilisation, which
inactivates the enzymes active on 2a, thus allowing its accumula-
tion, as previously shown.16,31 Other treatments at low temperature
(such as maintenance at 4 ꢀC or ꢁ20 ꢀC for one month) gave similar
effects, providing cells with lowered reducing activity towards 2a;
Table 1 summarizes the results obtained in the reduction of 1a
using cells after different treatments.
Table 1
2. Results and discussion
2.1. Benzil reduction
Reduction of benzil (1a) catalysed by whole cells from P. glucozyma after different
treatments
Cells
2a (%)
ee (S)-2a (%)
3a (%)
t (h)
Untreated
8
99
85
95
99
75
88
86
92
d
15
5
10
3
3
Lyophilised16
We firstly carried out the reduction of benzil (1a) catalysed by
whole untreated cells of P. glucozyma. Thus, cells from growth
cultures were only centrifuged and suspended in phosphate buffer;
the reaction was started by the addition of 1a previously dissolved
in DMSO as described in Experimental section. Compound 1a was
reduced producing hydrobenzoin (3a) with transient formation of
(S)-benzoin ((S)-2a) (Scheme 1).
Maintained at 4 ꢀC
Maintained at ꢁ20 ꢀC
3
Cells maintained at ꢁ20 ꢀC gave total conversion of the sub-
strate; (S)-2a was formed in high yield (95%) and 86% ee, with
minor amounts of 3a (5%) observed after 3 h. Consequently,
cells maintained at ꢁ20 ꢀC were used in all the further experiments
aimed at the synthesis of (S)-2a with high yields and
enantioselectivity.
The biotransformation of 1a was performed in a two-liquid-
phase system composed of water and n-heptane (1:1); n-heptane
(log P¼4.0) was chosen since it allows high partition of the sub-
strate into the organic phase and it is known to be often compatible
with enzymatic activities.28 Bio-reduction of 1a using P. glucozyma
in a two-liquid-phase system composed of water and n-heptane
yielded (S)-2a (90% conversion, 99% ee) after 48 h.
This two-liquid-phase system was also tested in the
reduction of other benzil derivatives (1,2-diaryl-1,2-ethanodione
structures) (Table 2). In all cases only the corresponding
a-
Scheme 1. Benzil reduction catalysed by untreated whole cells from P. glucozyma CBS
5766 in phosphate buffer.
hydroxyketones were obtained in high yields and excellent enan-
tiomeric excesses, and no diols could be detected. All substrates
and products were identified by comparison of HPLC retention
times with racemic and optically active standards previously pre-
pared following described methodologies.9 In all cases the ster-
eorecognition was similar and the absolute configurations of the
products were assigned to be (S) except for those compounds
containing a thienyl moiety (entries 3 and 4, Table 2), whose ab-
solute configurations were (R) because of the preference of the
sulfur atom.
This benzil reduction reaction was monitored, in order to study
the catalytic behaviour of the possible dehydrogenases contained in
the cells (Fig. 1). Compound (S)-2a was initially produced with low
enantiomeric excess (85% molar conversion and 45% ee after 1 h);
the enantiomeric excess was gradually increased as (S)-2a was
being further reduced to 3a, indicating that the dehydrogenase
acting on 2 has a strong stereopreference for the reduction of the

M.C. Fragnelli et al. / Tetrahedron 68 (2012) 523e528
525
Table 2
in Scheme 2. Moreover, the presence of the organic solvent com-
pletely prevents the formation of the diol.
Reduction of different benzoins catalysed by whole cells from P. glucozyma in a two-
liquid-phase systema
O
O
Pichia glucozyma
Ar
Ar
OH
Ar
Ar
Phosphate buffer 0.1 M,pH 7.0:
O
n-Heptane (1:1), 28 ºC
2
1
1a, 2a: Ar = Phenyl
1b, 2b: Ar = 2-Furyl
1c, 2c: Ar = 2-Thienyl
1d, 2d: Ar = 3-Thienyl
Entry
Substrate
Product
t (h)
Conv. (%)
eep (%)
1
2
3
4
1a
1b
1c
1d
(S)-2a
(S)-2b
(R)-2c
(R)-2d
48
48
48
48
90
95
97
90
99
99
98
Scheme 2. Proposed mechanism for 2a deracemisation catalysed by whole cells from
P. glucozyma maintained at ꢁ20 ꢀC in an aqueous/organic biphasic system.
85
a
Reaction conditions: cells from P. glucozyma (maintained at ꢁ20 ꢀC, 20 g/L) were
suspended in a 10 mL biphasic system phosphate buffer/n-heptane 1:1containing
5% glucose and they were incubated for 45 min at 28 ꢀC. Then 1 (0.02 mmol) solved
in DMSO (0.1 mL) was added. The mixture was shaken at 28 ꢀC.
The deracemisation of rac-2a was also studied using organic
solvents with different polarity values (Table 3). 2-MeTHF has
emerged in last years as a more sustainable substitute of traditional
THF and its chemical properties permit its use in many organic and
biocatalytical processes.33,34,10 It is less water miscible than THF,
allowing an easier isolation of products. For these reasons we de-
cided to test a biphasic system composed by this greener solvent.
However, its polarity seemed to be too high, as no conversion was
detected (entry 1, Table 3). Similar results were obtained when
employing diisopropylether and toluene (entries 2 and 3, re-
spectively, Table 3), which completely switched off the redox sys-
tem of P. glucozyma active on (R,S)-2a.
The biphasic system allowed not only the synthesis of opti-
cally active benzoins, without the formation of the corresponding
diols, but also an easier isolation of the products. It was not
necessary to carry out different extractions to isolate the prod-
ucts. The organic phase containing the final chiral compounds
could be easily separated from water and products could be
simply recovered for their purification and characterization.
2.2. Benzoin deracemisation
Table 3
Results obtained in the deracemisation of (R,S)-2a catalysed by whole cells from P.
glucozyma in a biphasic system composed by phosphate buffer and different organic
solventsa
Biotransformation of 2 was first studied using untreated cells
from P. glucozyma. The reduction of rac-2a in aqueous medium
(phosphate buffer 0.1 M, pH 7) furnished 3a with total conversion
in 8 h. Isolated 3a was a 50:50 mixture of (R,R)- and meso-stereo-
isomers. From the stereochemical outcome of this bio-
transformation, it can be assumed that the carbonyl group of 2a
should be asymmetrically reduced, forming a new stereocenter
having the (R)-configuration. To confirm this hypothesis, the
reduction of commercially available (R)-2a was carried out under
the same reaction conditions. After 8 h the biotransformation
reached 90% conversion: isolated 3a showed a stereochemical
composition of 75:25 of (R,R)-/meso-stereoisomers. This result is in
apparent contradiction with the previous hypothesis and could be
explained by considering a stereoinversion of the stereogenic
centre of (R)-2a. It is known that stereoinversion of secondary al-
cohols can occur by concurrent tandem biocatalytic oxidation and
reduction, whereby one enantiomer is oxidized to a carbonyl and
then reduced to the opposite enantiomer.32
These observations led us to investigate the possibility of car-
rying out deracemisation of racemic 2a. The biotransformation of
rac-2a was performed in water and in the two-liquid-phase system
composed of water and n-heptane using cells maintained at
ꢁ20 ꢀC. The biotransformation in water did not give any results,
while the one performed in the two-liquid-phase system showed
deracemisation with presence of (S)-2a (85%) as the only enantio-
mer. Compound 1a was detected (10e15%) in the organic solvent
during the progress of the reaction, pointing out the occurrence of
an oxidative activity. In fact, highly hydrophobic 1a partitions
mostly in the organic phase, therefore allowing for its transient
accumulation. The process of stereoinversion might be caused
by the activity of different dehydrogenases: thus, one de-
hydrogenase would catalyse the oxidation of (R)-2a to 1a, which is
subsequently stereoselectively reduced to (S)-2a by another (S)-
selective dehydrogenase with high affinity towards 1a, as shown
Entries
Solvent
log P
2a (%)
(S)-2a ee (%)
1
2
3
4
5
6
2-MeTHF
Diisopropylether
Toluene
Hexane
Heptane
0.72
1.9
2.5
3.5
4.0
4.7
>99
90
92
83
85
d
<5
<5
>98
>98
>98
Isooctane
91
a
Reaction conditions: cells from P. glucozyma (maintained at ꢁ20 ꢀC, 20 g/L) were
suspended in a 10 mL biphasic system phosphate buffer/organic solvent 1:1con-
taining 5% glucose and they were incubated for 45 min at 28 ꢀC. Then (R,S)-2a
(0.02 mmol) solved in DMSO (0.1 mL) was added. The mixture was shaken at 28 ꢀC
for 48 h.
Only highly hydrophobic solvents (log P>3.5) allowed for
deracemisation, obtaining best results with isooctane. The
enantiomeric excess reached in the case of a biphasic system con-
taining isooctane was higher than that obtained in n-heptane
(entries 5 and 6, Table 2). This result is in accordance with other
previously described, showing a similar correlation between the
log P of the organic solvent and ADHs activities (the higher the
better).35
Same reaction conditions were applied in the deracemisation of
other benzoins (1,2-diaryl-2-hydroxyethanone structures). Results
are shown in Table 4.
In most cases complete deracemisation was achieved, and
enantiomerically pure benzoins were accumulated in the reaction
medium for 48 h. To the best of our knowledge, this is the shortest
benzoin deracemisation reaction, which leads to optically active
products, easily isolated because of the use of a water/organic
solvent biphasic system. Just in the case of substrate (R,S)-2f
(entry 5, Table 4) low enantiomeric excess was achieved, probably
because the bulkier substituents on the aryl moieties, which could
interfere with the redox system action.

526
M.C. Fragnelli et al. / Tetrahedron 68 (2012) 523e528
Table 4
Molar conversions were determined by HPLC analysis using
a Hypersil ODS column (250 mmꢂ4.6 mm), a diode-array detector
and a mixture of water/acetonitrile (6:4) as eluent with a flow rate
of 0.8 mL/min.
Deracemisation of different benzoins catalysed by whole cells from P. glucozyma in
a biphasic system phosphate buffer/isooctane, 1:1a
O
O
Pichia glucozyma
Ar
Ar
OH
Conversions and enantiomeric excesses of were routinely
determined by HPLC using a ChiralcelÒ OD-H (Daicel Chemical
Industries) analytical column (mobile phase of n-hexane/2-
propanol, 90:10 at a flow rate of 1 mL/min, 25 ꢀC). HPLC analy-
ses: retention times of standards: (R)-2a, t¼20.3 min, (S)-2a,
t¼14.6 min; (R)-2b, t¼19.8 min, (S)-2b, t¼16.6 min; (R)-2c,
t¼27.2 min, (S)-2c, t¼21.8 min; (R)-2d, t¼22.7 min, (S)-2d,
t¼17.9 min; (R)-2e, t¼23.9 min, (S)-2e, t¼21.3 min; (R)-2f,
t¼17.2 min, (S)-2f, t¼16.5 min.
Ar
(R,S)-2
Ar
Phosphate buffer 0.1 M,pH 7.0:
isooctane (1:1), 28 ºC
OH
(S)-2
Entry
Substrate
Ar
Product
ee (%)
1
2
3
4
5
(R,S)-2a
(R,S)-2b
(R,S)-2d
(R,S)-2e
(R,S)-2f
Phenyl
2-Furyl
3-Thienyl
4-Methoxyphenyl
4-Ethoxyphenyl
(S)-2a
(S)-2b
(R)-2d
(S)-2e
(S)-2f
95
99
99
91
58
The stereochemical composition of hydrobenzoin was de-
termined by HPLC using a ChiralcelÒ OJ column (mobile phase of n-
hexane/2-propanol, 90:10 at a flow rate of 0.8 mL/min, 25 ꢀC); the
absolute configuration was assigned on the basis of the chro-
matographic behaviour of commercially available optically pure
standards: (R,R)-3: 12.4 min, meso-3: 16.7 min.
a
Reaction conditions: cells from P. glucozyma (maintained at ꢁ20 ꢀC, 20 g/L) were
suspended in a 10 mL biphasic system phosphate buffer/isooctane 1:1containing 5%
glucose and they were incubated for 45 min at 28 ꢀC. Then (R,S)-2 (0.02 mmol)
solved in DMSO (0.1 mL) was added. The mixture was shaken at 28 ꢀC for 48 h.
Enantiomerically pure products were identified by NMR spec-
troscopic analysis and HPLC data comparison with previously
prepared chiral standards.9 Optical rotations were measured and
the absolute configurations were assigned to be S according to
a correlation of the positive optical rotation values of these com-
pounds with data from literature.5,9 As well as in the reduction
reaction, the deracemisation of the different benzoins leads to the
formation of the (S) enantiomer, except for 3-thenoin, identified as
the (R) enantiomer.
4.2. Microorganisms, media and culture conditions
P. glucozyma CBS 5766 was cultured in 3.0 L fermenters with
1.0 L of malt broth pH 6.0 for 48 h, at 27 ꢀC and agitation speed
100 rpm furnishing 8.15 g of dry weight biomass. The cells from
submerged cultures were recovered by centrifugation and washed
with 0.1 M phosphate buffer, pH 7.0 before further centrifugation.
Whole wet cells were stored at 4 ꢀC or at ꢁ20 ꢀC.
3. Conclusion
P. glucozyma CBS 5766 expresses different redox enzymes with
activity on benzil and benzoin. Most of the activity towards
benzoin is inactivated when cells are maintained at ꢁ20 ꢀC,
therefore the reduction of benzil occurs with accumulation of
benzoin. The enantioselectivity of the reaction is increased when
it is carried out in a two-liquid-phase system composed of water
and a hydrophobic organic solvent (i.e., the use of n-heptane yiel-
ded (S)-2a with 90% conversion and 99% ee).
The presence of a hydrophobic solvent (i.e., heptane or iso-
octane) can be also exploited for the deracemisation of racemic
benzoins. The simultaneous occurrence of oxidative activities to-
wards racemic benzoin and enantioselective reduction of benzil
allows for one-pot deracemisation.
4.3. Synthesis of 1,2-diarylethanediones employed as
substrates of the reduction reactions. Synthesis of 1,2-
di(furan-2-yl)ethane-1,2-dione (1b)
The synthesis of 1,2-diarylethanediones employed as sub-
strates of the reduction reactions was carried out following the
methodology previously described:16 pyridinium chlorochromate
(565 mg, 2.62 mmol) was suspended in CH2Cl2 (3.5 mL) in a 50 mL
round bottom flask fitted with a reflux condenser. 2-Furoin ((R,S)-
2b) (355 mg, 1.85 mmol) in CH2Cl2 (300 mL) was added and the
mixture was stirred. After 6 h, 5 mL of ether was added and the
supernatant decanted from the black residue. This solid was
washed with ether (3ꢂ5 mL) and the combined organic solution
was evaporated under vacuum. The resulting oil was purified by
silica gel column chromatography (CH2Cl2), collecting 310 mg
(1.63 mmol) of a yellow solid (88% yield): mp 161e162 ꢀC (lit.36 mp
160e162 ꢀC).
In conclusion, this work shows how to achieve highly regio- and
enantioselective reduction of benzil to (S)-benzoin and total dera-
cemisation of racemic benzoin, as well as different benzoin de-
rivatives, using the non-conventional yeast P. glucozyma.
Anal. Calcd for C10H6O4: C, 63.16; H, 3.18. Found: C, 62.80; H,
4. Experimental section
4.1. General
3.00. 1H NMR (250 MHz, CDCl3)
d
(ppm): 6.58 (2H, dd, J¼3.7,1.7 Hz),
7.59 (2H, dd, J¼3.7, 0.7 Hz), 7.72 (2H, dd, J¼1.7, 0.7 Hz). 13C NMR
(63 MHz, CDCl3)
d (ppm): 113.5, 125.2, 149.8, 149.9, 177.2.
Benzil (1a), rac-benzoin ((R,S)-2a), (R)-benzoin ((R)-2a), the
three stereoisomers of hydrobenzoin ((R,R)-3a, (R,S)-3a, (S,S)-3a),
2-furoin ((R,S)-2b) and 4,40-dimethoxybenzoin ((R,S)-2e) were
purchased from Sigma Aldrich and used as-received. P. glucozyma
CBS 5766 was obtained from Centraal Bureau voor Schimmelcul-
tures (Utrecht, The Netherlands).
4.3.1. Synthesis of 1,2-di(thiophene-2-yl)ethane-1,2-dione
(1c). Yellow solid, 348 mg (1.55 mmol) (84% yield): mp 80e81 ꢀC
(lit.37 81e82 ꢀC). Anal. Calcd for C10H6O2S2: C, 54.03; H, 2.72; S,
28.85. Found: C, 53.78; H, 2.60; S, 28.50. 1H NMR (250 MHz, CDCl3)
d
(ppm): 7.24 (2H, dd, J¼4.8, 3.9 Hz), 7.88 (2H, dd, J¼4.8,1.0 Hz), 8.10
(2H, dd, J¼3.9, 1.0 Hz). 13C NMR (63 MHz, CDCl3)
d (ppm): 129.1,
NMR spectra were recorded on a Bruker AC-250. Chemical
137.7, 137.9, 139.0, 182.8.
shifts (
(1H:
7.27 ppm) and CDCl3
Column chromatography purifications were conducted on silica
gel 60 (40e63 m). TLC was carried out on aluminium sheets pre-
coated with silica gel; the spots were visualized under UV light
¼254 nm).
d
) are reported in parts per million (ppm) relative to CHCl3
d
(
13C:
d
77.0 ppm).
4.3.2. Synthesis of 1,2-di(thiophene-3-yl)ethane-1,2-dione
(1d). Yellow solid, 370 mg (1.66 mmol) (89% yield): mp 79e80 ꢀC
(lit.38 79e80 ꢀC). Anal. Calcd for C10H6O2S2: C, 54.03; H, 2.72; S,
28.85. Found: C, 53.70; H, 2.50; S, 28.50. 1H NMR (250 MHz, CDCl3)
m
(l
d
(ppm): 7.24 (2H, dd, J¼4.9, 4.0 Hz), 7.88 (2H, dd, J¼4.9, 1.2 Hz),

M.C. Fragnelli et al. / Tetrahedron 68 (2012) 523e528
527
8.00 (2H, dd, J¼4.0, 1.2 Hz). 13C NMR (63 MHz, CDCl3)
d
(ppm):
70 mL of AcOEt. The organic phases were collected and dried over
Na2SO4 and the solvent was evaporated. The crude residue was
purified with flash chromatography: benzoin was recovered using
hexane/AcOEt 9:1, while for hydrobenzoin hexane/AcOEt 7:3 was
used.
129.1, 137.7, 137.9, 139.0, 182.8.
4.4. Synthesis of 1,2-diaryl-2-hydroxyethanones employed as
substrates of the deracemisation reaction. Synthesis of 2-
thenoin [2-hydroxy-1,2-di(thiophen-2-yl)ethanone] ((R,S)-2c)
4.6. Biotransformation conditions (biphasic system)
The synthesis of the 1,2-diaryl-2-hydroxyethanones employed
as standards and as substrates of the deracemisation reactions was
carried out following the methodology previously described:9
Thiamine hydrochloride (1.686 mg, 5 mmol) was dissolved in ab-
solute ethanol (30 mL) and triethylamine (4.2 mL, 30 mmol) and
2-thiophenecarboxaldehyde (8.9 mL, 100 mmol) were added. The
mixture was stirred at room temperature under argon atmosphere.
After 24 h the product started to precipitate; the mixture was fil-
tered and the solid was collected and washed with cold ethanol,
yielding 10.75 g (48 mmol) of a white solid (48% yield): mp
107e108 ꢀC (lit.39 107e108 ꢀC).
Whole cells stored at ꢁ20 ꢀC were suspended in 10 mL of a bi-
phasic system composed by 0.1 M phosphate buffer pH 7.0 (5 mL)
and the organic solvent (5 mL), containing 5% glucose to reach the
desired biomass concentration (20 g/L dry weight). After 45 min of
incubation at 28 ꢀC, the substrate 1 (reduction reaction) or 2
(deracemisation reaction) (0.02 mmol), dissolved in 0.1 mL of
dimethylsulfoxide, was added. The flasks were shaken on a re-
ciprocal shaker (100 spm) at 28 ꢀC. Samples from the organic phase
(0.5 mL) were taken at intervals, dried over Na2SO4 and evaporated.
Products were dissolved in a mixture hexane/2-propanol (90:10)
and analysed by HPLC.
Anal. Calcd for C10H8O2S2: C, 53.53; H, 3.59; S, 28.59. Found: C,
53.58; H, 3.62; S, 28.56. 1H NMR (250 MHz, CDCl3)
d (ppm): 4.41
Reactions were scaled to 20 mL volume in order to isolate and
purify the enantiomerically pure products. NMR data of isolated
chiral products were in accordance with those described for race-
mic benzoins. Enantiomeric excesses were determined by HPLC
(1H, s), 6.07 (1H, s), 7.01 (1H, dd, J¼5.09, 3.50 Hz), 7.13 (1H, dd,
J¼3.50, 1.21 Hz), 7.15 (1H, dd, J¼4.91, 3.80 Hz), 7.34 (1H, dd, J¼5.09,
1.21 Hz), 7.75 (1H, dd, J¼4.91, 1.10 Hz), 7.79 (1H, dd, J¼3.80, 1.10 Hz).
13C NMR (63 MHz, CDCl3)
d (ppm): 71.3, 127.2, 127.3, 127.6, 128.8,
analysis and optical rotations were measured:
20
134.7, 135.8, 139.6, 142.4, 190.3.
Reduction products: (S)-2a, [
a
]
þ30.4 (c 0.33, CHCl3); (S)-2b,
D
20
20
[a
]
þ37.8 (c 0.36, CHCl3); (R)-2c, [
a
]
þ31.4 (c 0.36, CHCl3); (R)-
D
2d, [
D
20
4.4.1. Synthesis of 3-thenoin [2-hydroxy-1,2-di(thiophen-3-yl)etha-
none] ((R,S)-2d). White solid, 9.18 g (41 mmol, 41% yield): mp
106e107 ꢀC (lit.40 106e107 ꢀC). Anal. Calcd for C10H8O2S2: C, 53.53;
H, 3.59; S, 28.59. Found: C, 53.52; H, 3.92; S, 27.36. 1H NMR
a
]
þ34.6 (c 0.33, CHCl3).
D
Deracemisation products: (S)-2a, [
a
]
20 þ16.9 (c 0.26, CHCl3); (S)-
D
a
20
20
2b, [
a]
þ55.2 (c 0.46, CHCl3); (R)-2d, [
]
þ52.9 (c 0.46, CHCl3);
D
D
20
(S)-2e, [
a
]
20 þ34.0 (c 0.4, CHCl3); (S)-2f, [
a
]
þ17.2 (c 0.4, CHCl3).
D
D
(250 MHz, CDCl3)
d
(ppm): 4.34 (1H, d, J¼6.03 Hz), 5.84 (1H, d,
J¼6.03 Hz), 6.99 (1H, dd, J¼4.99, 1.21 Hz), 7.28 (1H, dd, J¼4.99,
2.83 Hz), 7.29 (1H, dd, J¼2.91, 0.69 Hz), 7.33 (1H, dd, J¼2.83,
1.21 Hz), 7.51 (1H, dd, J¼5.13, 0.69 Hz), 8.04 (1H, dd, J¼2.13,
Acknowledgements
The financial support by the project CTQ2009-11801 from
MICINN (Spanish Ministry of Science and Innovation) is gratefully
acknowledged.
2.91 Hz). 13C NMR (63 MHz, CDCl3):
127.2, 134.2, 138.5, 140.2, 192.4.
d 72.4, 124.2,124.3, 125.3, 126.5,
4.4.2. Synthesis of 1,2-bis(4-ethoxyphenyl)-2-hydroxyethanone
((R,S)-2f). White solid, 960 mg (3.2 mmol, 53% yield): mp
89e90 ꢀC (lit.41 86 ꢀC). Anal. Calcd for C18H20O4: C, 71.98; H, 6.71.
References and notes
1. Enders, D.; Kallfass, U. Angew. Chem., Int. Ed. 2002, 41, 1743e1745.
2. Enders, D.; Han, J. Tetrahedron: Asymmetry 2008, 19, 1367e1371.
3. Linghu, X.; Potnick, J. R.; Johnson, J. S. J. Am. Chem. Soc. 2004, 126, 3070e3071.
4. Hoyos, P.; Sinisterra, J. V.; Molinari, F.; Alcantara, A. R.; De Maria, P. D. Acc. Chem.
Res. 2010, 43, 288e299.
Found: C, 71.79; H, 6.70. 1H NMR (250 MHz, CDCl3)
d (ppm): 1.36
(3H, t, J¼7.00 Hz), 1.39 (3H, t, J¼7.10 Hz), 3.96 (2H, q, J¼7.00 Hz),
4.03 (2H, q, J¼7.10 Hz), 4.57 (1H, d, J¼6.02 Hz), 5.83 (1H, d,
J¼6.02 Hz), 6.72e6.81 (2H, m), 6.82e6.93 (2H, m), 7.05e7.19 (1H,
m), 7.20e7.38 (1H, m), 7.70e7.80 (1H, m), 7.83e7.95 (1H, m). 13C
€
5. Demir, A. S.; S¸ es¸ enoglu, O.; Eren, E.; Hosrik, B.; Pohl, M.; Janzen, E.; Kolter, D.;
€
€
Feldmann, R.; Dunkelmann, P.; Muller, M. Adv. Synth. Catal. 2002, 344, 96e103.
€
6. Dunkelmann, P.; Kolter-Jung, D.; Nitsche, A.; Demir, A. S.; Siegert, P.; Lingen, B.;
NMR (63 MHz, CDCl3)
d (ppm): 14.6, 14.8, 63.4, 63.8, 75.2, 114.3,
€
Baumann, M.; Pohl, M.; Muller, M. J. Am. Chem. Soc. 2002, 124, 12084e12085.
7. Demir, A. S.; Ayhan, P.; Sopaci, S. B. Clean–Soil, Air, Water. 2007, 35, 406e412.
8. Muller, M.; Gocke, D.; Pohl, M. FEBS J. 2009, 276, 2894e2904.
114.9, 128.9, 131.5, 136.1, 158.5, 163.4, 197.3.
€
9. Hoyos, P.; Fernandez, M.; Sinisterra, J. V.; Alcantara, A. R. J. Org. Chem. 2006, 71,
7632e7637.
10. Hoyos, P.; Quezada, M. A.; Sinisterra, J. V.; Alcantara, A. R. J. Mol. Catal. B: Enzym.
4.5. Biotransformation conditions (phosphate buffer
medium)
ꢀ
2011, 72, 20e24.
11. Maruyama, R.; Nishizawa, M.; Itoi, Y.; Ito, S.; Inoue, M. J. Biotechnol. 2002, 94,
157e169.
Whole cells were suspended in 10 mL of 0.1 M phosphate buffer
pH 7.0 containing 5% glucose to reach the desired biomass con-
centration (20 g/L dry weight). After 45 min of incubation at 28 ꢀC,
the substrate 1 (reduction reaction) or 2 (deracemisation reaction)
(0.02 mmol), dissolved in 0.1 mL of dimethylsulfoxide, was added.
In experiments with two-liquid-phase systems, different solvents
were added to reach the desired volumes before adding the sub-
strate. The flasks were shaken on a reciprocal shaker (100 rpm) at
28 ꢀC. Samples (0.5 mL) were taken at intervals, added with an
equal volume of AcOEt containing 1-phenylethanol as an internal
standard and centrifuged. The organic phase was dried over an-
hydrous Na2SO4 and analysed. Preparative biotransformations were
carried out on 100 mL scale; when the reaction was over, 100 mL of
AcOEt was added and the resulting mixture was shaken and
centrifuged; the aqueous phase was extracted twice more with
12. Saito, T.; Maruyama, R.; Ono, S.; Yasukawa,, N.; Kodaira, K.; Nishizawa, M.; Ito,
S.; Inoue, M. Appl. Biochem. Biotechnol. 2003, 111, 185e190.
13. Mahmoodi, N. O.; Mohammadi, H. G. Monatsh. Chem. 2003, 134, 1283e1288.
14. Demir, A. S.; Hamamci, H.; Ayhan, P.; Duygu, A. N.; Igdir, A. C.; Capanoglu, D.
Tetrahedron: Asymmetry 2004, 15, 2579e2582.
15. Demir, A. S.; Ayhan, P.; Demirtas, U.; ErkIlIc¸ , U. J. Mol. Catal. B: Enzym. 2008, 55,
164e168.
16. Hoyos, P.; Sansottera, G.; Fernandez, M.; Molinari, F.; Sinisterra, J. V.; Alcantara,
A. R. Tetrahedron 2008, 64, 7929e7936.
17. Demir, A. S.; Hamamci, H.; Sesenoglu, O.; Neslihanoglu, R.; Asikoglu, B.; Capa-
noglu, D. Tetrahedron Lett. 2002, 43, 6447e6449.
18. Okano, K. Tetrahedron 2011, 67, 2483e2512.
19. Tanaka, T.; Kawase, M.; Tani, S. Bioorg. Med. Chem. 2004, 12, 501e505.
20. Aoyagi, Y.; Agata, N.; Shibata, N.; Horiguchi, M.; Williams, R. M. Tetrahedron Lett.
2000, 41, 10159e10162.
21. Williams, R. M. Aldrichimica Acta 1992, 25, 11e25.
22. Yuki, Y.; Saigo, K.; Tachibana, K.; Hasegawa, M. Chem. Lett. 1986, 1347e1350.
23. Brown, B.; Hegedus, L. S. J. Org. Chem. 1998, 63, 8012e8018.

528
24. Chevenert, R.; Thiboutot, S. Chem. Lett. 1988, 1191e1192.
25. Molinari, F.; Gandolfi, R.; Aragozzini, F.; Leon, R.; Prazeres, D. M. F. Enzyme
Microb. Technol. 1999, 25, 729e735.
M.C. Fragnelli et al. / Tetrahedron 68 (2012) 523e528
ˇ
33. Aycock, D. F. Org. Process Res. Dev. 2007, 11, 156e159.
34. Shanmuganathan, S.; Natalia, D.; van den Wittenboer, A.; Kohlmann, C.;
Greiner, L.; Dominguez de Maria, P. Green Chem. 2010, 12, 2240e2245.
35. de Gonzalo, G.; Lavandera, I.; Faber, K.; Kroutil, W. Org. Lett. 2007, 9,
2163e2166.
€
€
26. Husain, S. M.; Stillger, T.; Dunkelmann, P.; Lodige, M.; Walter, L.; Breitling, E.;
€
€
Pohl, M.; Burchner, M.; Krossing, I.; Muller, M.; Romano, D.; Molinari, F. Adv.
Synth. Catal. 2011, 353, 2359e2362.
36. Wang, X.; Zhang, Y. Tetrahedron 2003, 59, 4201e4207.
ꢂ
ꢀ
ꢂ
27. Mahmoodi, N.; Navrood, M. N. Arkivoc 2007, iii, 37e45.
28. Cabral, J. M. S.; Aires-Barros, M. R.; Pinheiro, H.; Prazeres, D. M. F. J. Biotechnol.
1997, 59, 133e143.
37. Lukes, V. r.; Breza, M.; Vegh, D.; Hrdlovic, P.; Laurinc, V. Synth. Met. 2003, 138,
399e408.
38. Hyatt, J. L.; Stacy, V.; Wadkins, R. M.; Yoon, K. J. P.; Wierdl, M.; Edwards, C. C.;
Zeller, M.; Hunter, A. D.; Danks, M. K.; Crundwell, G.; Potter, P. M. J. Med. Chem.
2005, 48, 5543e5550.
29. D’Arrigo, P.; Fuganti, C.; Pedrocchi Fantoni, G.; Servi, S. Tetrahedron 1998, 54,
15017e15026.
30. Molinari, F.; Occhiato, E. G.; Aragozzini, F.; Guarna, A. Tetrahedron: Asymmetry
1998, 9, 1389e1394.
31. Carballeira, J. D.; Quezada, M. A.; Hoyos, P.; Simeo, Y.; Hernaiz, M. J.; Alcantara,
39. Bag, S.; Vaze, V. V.; Degani, M. S. J. Chem. Res. 2006, 2006, 267e269.
40. Roberts-Bleming, S. J.; Davies, G. L.; Kalaji, M.; Murphy, P. J.; Celli, A. M.; Donati,
D.; Ponticelli, F. J. Org. Chem. 2003, 68, 7115e7118.
ꢀ
A. R.; Sinisterra, J. V. Biotechnol. Adv. 2009, 27, 686e714.
41. Weissberger, A.; Strasser, E.; Mainz, H.; Schwarze, W. Justus Liebigs Ann. Chem.
32. Voss, C. V.; Gruber, C. C.; Kroutil, W. Angew. Chem., Int. Ed. 2008, 47, 741e745.
1930, 478, 112e128.