Enantioselective reduction of flavanone and oxidation of cis- and trans-flavan-4-ol by selected yeast cultures

Article abstract of DOI:10.1016/j.molcatb.2014.08.006

This research investigated stereochemistry of reduction of racemic flavanone and a concurrent competitive process of oxidation, taking place in cultures of live yeast strains. The results obtained gave us information about capability of tested biocatalysts for enantioselective (with respect to both substrate and product) reduction of flavanone and for enantioselective oxidation of the resulting cis- and trans-flavan-4-ols. As a result of our experiments we obtained (2S,4S)-cis-flavan-4-ol with 43% of conversion and 96% of enantiomeric excess, and (2R,4S)-trans-flavan-4-ol with 41% of conversion and ee > 99% in the culture of Rhodotorula rubra; (2S,4S)-cis-flavan-4-ol (43%, ee = 96%) along with (2R,4R)-cis-flavan-4-ol (44%, ee = 61%) in the culture of Zygosaccharomyces bailii KCh 907. Additionally, some of the tested strains demonstrated an excellent capability for enantioselective oxidation of (±)-cis-flavan-4-ol and (±)-trans-flavan-4-ol, obtained by chemical synthesis. A one-day biotransformation in the culture of Candida parapsilosis KCh 909 afforded (S)-flavanone (ee = 93%) as 49% of the reaction mixture and 49% of unreacted (2R,4R)-cis-flavan-4-ol with ee = 97%. Racemic trans-flavan-4-ol was effectively oxidized in the culture of Yarrowia lipolytica KCh 71 - after a three-day biotransformation the reaction mixture contained 52% of (R)-flavanone (ee = 85%) and 48% of (2R,4S)-trans-flavan-4-ol with a high enantiomeric excess (ee = 93%).

Full text of DOI:10.1016/j.molcatb.2014.08.006

Journal of Molecular Catalysis B: Enzymatic 109 (2014) 47–52
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Enantioselective reduction of flavanone and oxidation of cis- and
trans-flavan-4-ol by selected yeast cultures
Tomasz Janeczko∗, Monika Dymarska, Monika Siepka, Radosław Gniłka,
Agnieszka Les´niak, Jarosław Popłon´ ski, Edyta Kostrzewa-Susłow
Department of Chemistry, Wrocław University of Environmental and Life Sciences, Norwida 25, 50-375 Wrocław, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 26 March 2014
Received in revised form 8 August 2014
Accepted 11 August 2014
Available online 20 August 2014
This research investigated stereochemistry of reduction of racemic flavanone and a concurrent com-
petitive process of oxidation, taking place in cultures of live yeast strains. The results obtained gave us
information about capability of tested biocatalysts for enantioselective (with respect to both substrate
and product) reduction of flavanone and for enantioselective oxidation of the resulting cis- and trans-
flavan-4-ols. As a result of our experiments we obtained (2S,4S)-cis-flavan-4-ol with 43% of conversion
and 96% of enantiomeric excess, and (2R,4S)-trans-flavan-4-ol with 41% of conversion and ee > 99% in
the culture of Rhodotorula rubra; (2S,4S)-cis-flavan-4-ol (43%, ee = 96%) along with (2R,4R)-cis-flavan-
4-ol (44%, ee = 61%) in the culture of Zygosaccharomyces bailii KCh 907. Additionally, some of the tested
strains demonstrated an excellent capability for enantioselective oxidation of ( )-cis-flavan-4-ol and ( )-
trans-flavan-4-ol, obtained by chemical synthesis. A one-day biotransformation in the culture of Candida
parapsilosis KCh 909 afforded (S)-flavanone (ee = 93%) as 49% of the reaction mixture and 49% of unreacted
(2R,4R)-cis-flavan-4-ol with ee = 97%. Racemic trans-flavan-4-ol was effectively oxidized in the culture of
Yarrowia lipolytica KCh 71 – after a three-day biotransformation the reaction mixture contained 52% of
(R)-flavanone (ee = 85%) and 48% of (2R,4S)-trans-flavan-4-ol with a high enantiomeric excess (ee = 93%).
Keywords:
Flavanone
Flavan-4-ol
Biotransformation
Yeast
Enantioselective oxidation
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
A, and Eruberin B [14,15], means that the hydroxyl group is in a
pseudo axial position, and the established absolute configuration
Flavonoids are natural compounds widely present in the plant
kingdom, which are biosynthesized by plants from phenylala-
nine [1,2]. They are naturally found in fruits, vegetables, seeds
and plant products, therefore they are an integral part of human
diet. Flavonoids proved to be safe for human organism and their
daily intake has been estimated to be about 1 g per day [3–5].
Natural flavan-4-ones are homochiral and show levorotation. In
isolated from plants flavanones: pinocembrin, pinostrobin, hes-
peretin and naringenin at the C-2 stereogenic centre there has been
(S)-configuration assigned [6–9]. Natural flavan-4-ols have also S-
configuration at C-2, whereas the hydroxyl group at C-4 is situated
trans to the phenyl group [10,11]. Such an orientation of the sub-
stituents at C-2 and C-4, which has been described in literature for
luteoforol, apiforol [12,13] and glycosides: abacopterin I, Triphyllin
at C-4 is R [12,15,16].
It has been reported in literature that racemic flavanone (4)
can be enantiospecifically reduced to (2R,4R)-cis-flavan-4-ol (5)
with the help of chiral ruthenium complexes [17]. Except for
the respective flavan-4-ol (5) this method afforded also unre-
acted (S)-flavanone (4) with a high enantiomeric excess [17].
(S)-flavanone (4) and (2S,4S)-cis-flavan-4-ol (5) were also obtained
by means of an enantioselective Ru/NHC-catalyzed hydrogena-
tion of flavone [18]. (S)-flavanone (4) was prepared in six steps,
starting from commercially available 3,5-dimethoxyphenol and
(+)-ethyl (R)-3-hydroxy-3-phenylpropanoate [19]. Optically active
(R)- and (S)-flavanone (4) were prepared by the enantioselective
enzymatic hydrolysis of ( )-flavanone O-acyl oxime using lipase
[20]. Optically active cis-flavan-4-ols (5) were obtained in two
ways: by enzymatic esterification of respective alcohols [21], and
by enantioselective hydrolysis of respective esters [22]. Incubation
of ( )-flavanone (4) in the culture of bakers’ yeast gave (+)-(2S,4S)-
cis-4-hydroxyflavan (5) in 32% yield (83% ee) and optically active
(+)-(R)-flavanone (4) in 51% yield (20% ee) [22].
∗
Corresponding author. Tel.: +48 71 320 52 52; fax: +48 71 3284124.
E-mail addresses: janeczko13@interia.pl, tomasz.janeczko@up.wroc.pl
(T. Janeczko).
http://dx.doi.org/10.1016/j.molcatb.2014.08.006
1381-1177/© 2014 Elsevier B.V. All rights reserved.

48
T. Janeczko et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 47–52
In our previous papers we described biotransformations of
)-flavanone in the cultures of Aspergillus and Penicillium strains
and trans-flavan-4-ol (7) were prepared according to the procedure
described above. NMR spectra were recorded on a DRX 600 MHz
Bruker spectrometer and measured in CDCl3. Optical rotations were
measured with an Autopol IV automatic polarimeter (Rudolph).
Absolute configurations of the products were determined by com-
parison of their optical rotation values with literature data.
(
[23–25]. Except from achiral flavones we obtained mainly the
products of hydroxylation of flavanone at C-6 and C-4ꢀ (with the
enantiomeric excesses not exceeding 36%) [23]. The obtained prod-
ucts with reduced carbonyl group were racemic [25]. In our article
we report an enantiospecific reduction of racemic flavanone (4)
and an enantioselective oxidation of both ( )-cis-flavan-4-ol (5)
and ( )-trans-flavan-4-ol (7) by means of selected yeast cultures.
The objective of this study was to develop a convenient method of
obtaining optically active flavanones and cis- and trans-flavan-4-ols
(5,7).
2.3. Screening procedure
Erlenmeyer flasks (300 ml), each containing 100 ml of the ster-
ile medium consisting of 3 g glucose and 1 g aminobac dissolved in
water, were inoculated with a suspension of microorganisms and
then incubated for 3–7 days at 25 ◦C on a rotary shaker (190 rpm).
After full growth of the culture 20 mg of a substrate dissolved in 1 ml
of acetone was added. After 12 h and 1, 3, 6, and 9 days of incubation
under the above conditions, portions of 10 ml of the transforma-
tion mixture were taken out and extracted with CHCl3 (3× 10 ml).
The extracts were dried over MgSO4, concentrated in vacuo and
analyzed by GC. All the experiments were repeated three times.
2. Experimental
2.1. Materials
The substrates: Racemic cis-flavan-4-ol (5) was prepared by
reducing ( )-flavanone (4) with sodium borohydride in methanol.
Flavanone was obtained from 2ꢀ-hydroxychalcone (3) according to
the previously described method [26]. 10 g of the substrate was
dissolved in ethanol (75 ml) with addition of sodium acetate (8 g)
and the reaction mixture was refluxed for 48 h. Crystallization from
ethanol afforded pure ( )-flavanone. 2ꢀ-Hydroxychalcone (3) was
obtained from 2ꢀ-hydroxyacetophenone (1) and benzaldehyde (2)
following the method by Yadav et al. [27]; trans-flavan-4-ol (7)
was prepared by transforming cis-flavan-4-ol (5) into the the tosy-
late (6), which was then hydrolysed under basic conditions. The
microorganisms (twelve strains of yeast) were obtained as fol-
lows: Candida pelliculosa ZP22 was obtained from the Department
of Biotechnology and Food Microbiology of Wrocław University
of Environmental and Life Sciences [28], Yarrowia lipolytica KCh
71, Candida parapsilosis KCh 909, Candida wiswanathi KCh 120,
Rhodotorula rubra KCh 4, R. rubra KCh 82, Rhodotorula glutinis KCh
242, Saccharomyces cerevisiae KCh 464, Saccharomyces brasiliensis
KCh 905, Saccharomyces pastorianus KCh 906, Zygosaccharomyces
bailii KCh 907, and Candida sake KCh 908 were obtained from the
Department of Chemistry of Wrocław University of Environmental
and Life Sciences. All the strains were cultivated on a Sabouraud
agar consisting of aminobac (5 g), peptone K (5 g), glucose (40 g)
and agar (15 g) dissolved in 1 l of distilled water, at 28 ◦C and pH
5.5 and stored in a refrigerator at 4 ◦C.
2.4. Preparative biotransformation
The same transformations were performed on the preparative
scale in 2000 ml flasks, each containing 500 ml of the sterile cul-
tivation medium. The cultures were incubated under the same
conditions and then 100 mg of substrates dissolved in 10 ml of
acetone were added to the grown cultures. After incubation the
mixtures were extracted with CHCl3 (3× 300 ml), dried (MgSO4)
and concentrated in vacuo. The transformation products were sep-
arated by column chromatography and analyzed (TLC, GC, 1H NMR
and also confirmed by 13C NMR and correlation spectroscopy).
2.5. Spectral data of isolated metabolites
2.5.1. (2R,4R)-cis-flavan-4-ol (5)
One-day transformation of ( )-cis-flavan-4-ol (5) (100 mg) in
the culture C. parapsilosis KCh 909 yielded 46 mg of (2R,4R)-cis-
20
flavan-4-ol (5) (colorless crystals);
˛
[ ]D
= −64.4◦ (c = 1.1, CHCl3)
25
(97% ee) (lit. ˛
= +65.7◦ (c = 1.49, CHCl3), 92% ee zmiez˙ one dla
[ ]D
(2S,4S)-cis-flavan-4-ol (5) [21]. 1H NMR (600 MHz) (CDCl3) ı (ppm):
1.73 (d, 1H, J = 8.6 Hz, OH); 2.13 (dt, 1H, J = 13.0; 11.3 Hz, H-3a);
2.53 (ddd, 1H, J = 13.0, 6.3, 1.5 Hz, H-3e); 5.11 (ddd, 1H, J = 11.3,
8.6, 6.3 Hz, H-4a); 5.18 (dd, 1H, J = 11.3, 1.5 Hz, H-2a); 6.90 (d, 1H,
J = 8.2 Hz, H-8); 6.99 (t, 1H, J = 7.4 Hz, H-6); 7.21 (t, 1H, J = 8.0 Hz, H-
7); 7.34 (t, 1H, J = 7.3 Hz, H-4ꢀ); 7.40 (t, 2H, J = 7.6 Hz, H-3ꢀ and H-5ꢀ);
7.45 (d, 2H, J = 7.4 Hz, H-2ꢀ and H-6ꢀ); and 7.52 (d, 1H, J = 7.7 Hz, H-5).
13C NMR (151 MHz, CDCl3) ı = 40.1 (C-3), 65.8 (C-4), 76.8 (C-2),
116.8 (C-8), 121.0 (C-6), 125.7 (C-4a), 126.1 (C-2ꢀ and C-6ꢀ), 127.0
(C-5), 128.2 (C-4ꢀ), 128.7 (C-3ꢀ and C-5ꢀ), 129.2 (C-7), 140.5 (C-1ꢀ),
and 154.5 (C-8a).
2.2. Analytical methods
The course of biotransformation was controlled by means of
TLC. Analytical TLC was carried out on silica gel G 60 F254 plates
(Merck). Chromatograms were developed using hexane/acetone
mixture (4:1, v/v) as the eluent. Compounds were detected by
spraying the plates with 1% Ce(SO4)2 and 2% H3[P(Mo3O10)4] in
10% H2SO4. The products were separated by column chromatogra-
phy using silica gel (SiO2, Kieselgel 60, 230-400 mesh, 40–63 ␮m,
Merck) and hexane/acetone mixture (4:1, v/v) as the developing
system. Composition of biotransformation mixtures was estab-
lished by gas chromatography (GC) on Agilent Technologies 7890
A GC instrument, fitted with a flame ionization detector (FID) and
a chiral column CYCLOSIL-B 30 m × 0.25 mm × 0.25 ␮m. To deter-
minate the composition and enantiomeric excesses of the ketone
and alcohol mixtures the following temperature program was
used: 120 ◦C/0 min, gradient 1 ◦C min−1 to 194 ◦C/0 min, gradient
10 ◦C min−1 to 250 ◦C/5 min; the retention times: (S)-flavanone (4)
– 55.9 min; (R)-flavanone (4) – 56.5 min, (2S,4R)-trans-flavan-4-ol
(7) – 61.6 min; (2R,4S)-trans-flavan-4-ol (7) – 62.3 min, (2S,4S)-cis-
flavan-4-ol (5) – 67.1 min; (2R,4R)-cis-flavan-4-ol (5) – 67.9 min.
Reference samples of the racemic flavanone (4), cis-flavan-4-ol (5)
In the same biotransformation we isolated also (S)-flavanone (4)
20
25
(42 mg) ˛
[ ]D
= −46.1◦ (c = 1.0, CHCl ) (93% ee) (lit. ˛
= −56.2◦
[ ]D
3
(c = 0.50, CHCl3), 92% ee [22]. Oxidation of (2R,4R)-cis-flavan-4-ol
(5) with Jones reagent was carried out in the same way as described
20
previously [29], to give (R)-flavanone (4) (31 mg)
˛
[ ]D
= −61.6◦
(c = 0.5, CHCl3) (97% ee). 1H NMR (600 MHz) (CDCl3) ı (ppm): 2.91
(dd, 1H, J = 16.9; 3.0 Hz, H-3e); 3.11 (dd, 1H, J = 16.9, 13.2 Hz, H-3a);
5.49 (dd, 1H, J = 13.2; 3.0 Hz, H-2a); 7.05–7.08 (m, 2H, H-6 and H-8),
7.34–7.59 (m, 6H, H-7, H-2ꢀ, H-3ꢀ, H-4ꢀ, H-5ꢀand H-6ꢀ), and 7.95 (dd,
1H, J = 8.1; 1.8 Hz, H-5).
13C NMR (151 MHz, CDCl3) ı = 44.7 (C-3), 75.6 (C-2), 118.1 (C-8),
120.0 (C-4a), 121.6 (C-6), 126.1 (C-2ꢀ and C-6ꢀ), 127.1 (C-5), 128.7
(C-4ꢀ), 128.9 (C-3ꢀ and C-5ꢀ), 136.1 (C-7), 138.8 (C-1ꢀ), 161.6 (C-8a),
and 191.8 (C-4).

T. Janeczko et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 47–52
49
2.5.2. trans-flavan-4-ol tosylate (6)
with respect to the S- or R-enantiomer of 4 preference and spatial
position of the hydroxyl group formed by reduction of the car-
bonyl group of the substrate (Table 1). In the cultures of most of
the strains (R. rubra KCh 4 and KCh 82, R. glutinis KCh 242, Z. bailii
KCh 907, C. sake KCh 908 and S. pastorianus KCh 906) reduction of
(R)-flavanone was significantly faster than of the S-enantiomer. As
opposed to this, the strain C. pelliculosa ZP22 demonstrated enan-
tioselectivity towards (S)-flavanone, whereas, in the culture of the
strain C. wiswanathi KCh 120 at first (S)-flavanone was more effec-
tively reduced, but after 3 days we observed lower concentration
of the R-enantiomer compared to the S-one.
In all the cultures of the genus Rhodotorula, i.e. R. rubra KCh 4,
R. rubra KCh 82, and R. glutinis KCh 242 a very high S-selectivity
of the reduction of substrate 4 was observed. After one day of the
transformation (2R,4S)-trans-flavan-4-ol (7) was identified in the
reaction mixtures (41% for R. rubra KCh 4 and R. rubra KCh 82; 13%
for R. glutinis KCh 242) with the enantiomeric excess of over 99%.
At the first stage the reduction proceeded also with a very high
diastereoselectivity – after 12 h de = 85% for the strain R. rubra KCh
4; 81% for R. rubra KCh 82 and 94% for R. glutinis KCh 242. How-
ever, between the first and the third day of the transformation we
observed a significant increase in concentration of the cis-alcohol.
Determination of the absolute configuration of both stereogenic
centres of this product as (2S,4S)-cis-flavan-4-ol (5) showed that it
was formed by the reduction of (S)-flavanone (4).
Certain similarities were noticed for biotransformation of
racemic flavanone (4) by the strains C. wiswanathi KCh 120 and C.
pelliculosa ZP22. Both more effectively reduced (S)-flavanone than
the R-one. The reduction, like in the case of the strains of the genus
Rhodotorula, proceeded with a high S-specificity to give the respec-
tive (2S,4S)-cis-flavan-4-ol (5) (43% with ee = 96% after 1 day in the
culture of C. wiswanathi KCh 120 and 33% with ee = 75% after 3 days
in the culture of C. pelliculosa ZP22). In the culture of the strain C.
wiswanathi KCh 120 an increase in amount of the trans-alcohol with
the reaction time was also observed – 51% of (2R,4S)-trans-flavan-
4-ol (7) with ee = 92% after 3 days of the biotransformation. After
six days we noticed a considerable increase in concentration of fla-
vanone (4). This might be due to increase of oxidative properties of
this strain during the biotransformation.
Colorless crystals; 1H NMR (600 MHz) (CDCl3) ı (ppm): 2.29 (s,
3H, CH3 from tosyl group), 2.61 (d, 1H, J = 15.5 Hz, H-3e), 2.72 (dd,
1H, J = 15.5, 11.9, 4.7 Hz, H-3a), 4.86 (d, 1H, J = 11.9 Hz, H-2a), 6.66
(s, 1H, H-4e), 6.99 (t, 1H, J = 7.4 Hz, H-6), 7.05 (d, 2H, J = 7.5 Hz, H-3ꢀꢀ
and H-5ꢀꢀ from tosyl group), 7.07 (d, 1H, J = 7.5 Hz, H-8), 7.21 (d, 1H,
J = 7.7 Hz, H-5), 7.29 (t, 1H, J = 7.6 Hz, H-4ꢀ), 7.39–7.31 (m, 4H, H-2ꢀ,
H-3ꢀ, H-5ꢀ and H-6ꢀ), 7.41 (t, 1H, J = 7.4 Hz, H-7), and 7.69 (d, 2H,
J = 7.5 Hz, H-2ꢀꢀ and H-6ꢀꢀ from tosyl group).
13C NMR (151 MHz, CDCl3) ı = 21.26 ( CH3 from tosyl group),
37.21 (C-3), 65.18 (C-4), 71.96 (C-2), 113.29 (C-4a), 118.39 (C-8),
122.43 (C-6), 128.89 (C-2ꢀꢀ and C-6ꢀꢀ from tosyl group), 126.40 (C-
2ꢀ and C-6ꢀ), 128.63 (C-3ꢀꢀ and C-5ꢀꢀ from tosyl group), 128.73 (C-3ꢀ
and C-5ꢀ), 128.74 (C-4ꢀ), 131.19 (C-5), 132.57 (C-7), 138.32 (C-4ꢀꢀ
from tosyl group), 139.23 (C-1ꢀꢀ from tosyl group), 143.79 (C-1ꢀ),
and 156.45 (C-8a).
2.5.3. (2S,4R)-trans-flavan-4-ol (7)
Three-day transformation of ( )-trans-flavan-4-ol (7) (100 mg)
in the culture of Y. lipolytica KCh 71 yielded 41 mg of (2S,4R)-trans-
20
flavan-4-ol (7) (colorless crystals);
˛
= +12.4◦ (c = 0.7, CHCl3)
[ ]D
(94% ee). 1H NMR (600 MHz) (CDCl3) ı (ppm): 2.12 (ddd, 1H, J = 14.4;
12.2; 3.3 Hz, H-3a); 2.28 (ddd, 1H, J = 14.4, 2.5; 1.7 Hz, H-3e); 4.85
(dd, 1H, J = 3.3; 2.5 Hz, H-4e); 5.28 (dd, 1H, J = 12.2; 1.7 Hz, H-2a);
6.98 (t, 1H, J = 7.8 Hz, H-6), 6.99 (d, 1H, J = 8.1 Hz, H-8), 7.27 (t, 1H,
J = 7.6 Hz, H-7), 7.37 (t, 1H, J = 7.6 Hz, H-4ꢀ), 7.38 (d, 1H, J = 7.6 Hz, H-
5), 7.43 (t, 2H, J = 7.3 Hz, H-3ꢀ and H-5ꢀ), and 7.50 (d, 2H, J = 7.7 Hz,
H-2ꢀ and H-6ꢀ).
13C NMR (151 MHz, CDCl3) ı = 38.34 (C-3), 63.73 (C-4), 73.11
(C-2), 117.45 (C-8), 120.80 (C-6), 123.67 (C-4a), 126.29 (C-2ꢀ and
C-6ꢀ), 128.04 (C-4ꢀ), 128.60 (C-3ꢀ and C-5ꢀ), 129.93 (C-7), 130.14 (C-
5), 141.04 (C-1ꢀ), and 154.90 (C-8a). In the same biotransformation
20
we isolated also (R)-flavanone (4) (38 mg)
˛
= +54.8◦ (c = 0.8,
[ ]D
25
CHCl ) (85% ee) (lit.
˛
[ ]D
= +66.5◦ (c = 0.48, CHCl3), 98% ee [22].
3
Oxidation of (2S,4R)-trans-flavan-4-ol (7) with Jones reagent was
carried out in the same way as described previously [29], to give
20
(S)-flavanone (4) (28 mg) ˛
= −62.0◦ (c = 0.5, CHCl3) (94% ee).
[ ]D
Preparative biotransformation of ( )-flavanone (4) in the cul-
ture of C. wiswanathi KCh 120 was also carried out. The three-day
In the culture of the strain S. brasiliensis KCh 905 we
observed a drop in (S)-flavanone (4) concentration during the
biotransformation, however we identified the product of R-
specific dehydrogenase activity – (2S,4R)-trans-flavan-4-ol (7) with
ee > 90%. The R-selectivity of the reduction was also observed for
the strain S. pastorianus KCh 906. We identified both cis- (5) and
trans-alcohol (7) with the R-configuration at C-4. In the culture of
C. sake KCh 908 we observed a very high selectivity of the reduction
towards R-flavanone, but low enantioselectivity with respect to the
product. Both products: cis- and trans-flavan-4-ols were formed by
the reduction of the R-ketone. The strain Z. bailii KCh 907 demon-
strated high enantioselectivity towards both the substrate and the
product. After three days we identified in the reaction mixture 42%
of (2S,4R)-trans-flavan-4-ol (7) with ee = 62% and 44% of (2R,4R)-
cis-flavan-4-ol (5) with ee = 61% and de = 53%.
Because of the observed changes in enantiomeric excesses of
compounds in biotransformation mixtures and the increase in
amount of the substrate in some of the cultures we subjected to the
biotransformation racemic mixtures of both cis- and trans-flavan-
4-ol (5,7) (Tables 2 and 3).
In Table 2 we did not present the results for four strains: S.
brasiliensis KCh 905, S. pastorianus KCh 906, S. cerevisiae KCh 464
and Z. bailii KCh 907, for which substrate conversion was less
than 5%, even after ten days of incubation. All the strains omit-
ted in Table 2 belong to the genus Saccharomyces, including the
strain of S. cerevisiae, which is commonly used for biocatalysis.
In the cultures of the genus Rhodotorula we observed a drop in
biotransformation using 100 mg of the substrate afforded 45 mg of
20
(2S,4S)-cis-flavan-4-ol (5)
and 32 mg of (2R,4S)-trans-flavan-4-ol (7)
CHCl3) (92% ee).
˛
= +59.8◦ (c = 1.2, CHCl3) (95% ee)
[ ]D
20
˛
= −14.8◦ (c = 0.9,
[ ]D
3. Results and discussion
We performed biotransformations of racemic flavanone (4) and
cis- and trans-flavan-4-ols (5,7), which were prepared in a few step
synthesis (Scheme 1). For the biotransformations we used cultures
of twelve yeast strains. The selected microorganism comprised:
yeast of the genus Rhodotorula: R. rubra KCh 4, R. rubra KCh 82, and
R. glutinis KCh 242; of the genus Candida: C. wiswanathi KCh 120,
C. parapsilosis KCh 909, C. sake KCh 908, and C. pelliculosa ZP22;
of the genus Saccharomyces: S. cerevisiae KCh 464, S. brasiliensis
KCh 905, and S. pastorianus KCh 906 and two other yeast strains:
Y. lipolytica KCh 71 and Z. bailii KCh 907. The biocatalysts were
selected based on their biocatalytic properties, such as fast growth
rate, capability for effective enantioselective reduction of aliphatic-
aromatic ketones and oxidation of the respective alcohols [30–33].
Moreover, unlike in the case of filamentous fungi, in the cultures of
yeasts competitive reactions such as hydroxylation, dehydratation
and degradation rarely occur [34,35].
In the case of biotransformation of flavanone (4) we observed
that depending on the yeast species used as a biocatalyst, there
were considerable differences in enantioselectivity of the process

50
T. Janeczko et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 47–52
OH
OH
+
O
O
(2S,4S)-5
(2R,4S)-7
Candida wiswanathi
O
O
O
O
CH3COONa, EtOH
reflux for 48h
H
NaOH, MeOH
reflux for 2h
+
O
OH
OH
1
2
3
4
NaBH4 MeOH
1h Rt
OTs
OH
OH
NaOH, MeOH, H2O
reflux for 1h
TsCl, Py
o
24h, 0 C
O
O
O
6
7
5
Candida parapsilosis
Yarrowia lipolytica
O
O
OH
OH
O
+
+
O
O
O
(R)-4
(2S,4R)-7
(S)-4
(2R,4R)-5
Scheme 1. Chemoenzymatic synthesis of optically active flavanone (4) and cis- and trans-flavan-4-ols (5,7).
concentration of (2R,4R)-cis-flavan-4-ol (5) with the reaction
time, along with an increase in (2R,4S)-trans-flavan-4-ol (7) con-
centration. This was due to the enantioselective oxidation of
(2R,4R)-cis-flavan-4-ol (5) to (R)-flavanone (4), followed by the
stereoselective reduction of this enatiomer of 4 to (2R,4S)-trans-
flavan-4-ol (7). The process was the most effective in the culture
of R. glutinis KCh 242. After ten days the reaction mixture con-
tained 68% of (2R,4S)-trans-flavan-4-ol (7) with ee = 52% and 27%
of (R)-flavanone (4) with ee = 94%. In the cultures of the strains C.
wiswanathi KCh 120 and Candida sake KCh 908 an enantioselec-
tive oxidation of (2S,4S)-cis-flavan-4-ol (5) took place. For these
strains substrate conversion was low (11 and 19% of (S)-flavanone
(4), respectively, was observed in the reaction mixture), but the
enantioselectivity was high (Table 2). Surprisingly, C. parapsilosis
KCh 909 proved to be the most effective biocatalyst in this part of
our study. After one day of incubation of ( )-cis-flavan-4-ol (5) in
Table 1
Microbial transformation of racemic flavanone (4).
Microorganism
t [days]
Flavanone (4)
trans-flavan-4-ol (7)
cis-flavan-4-ol (5)
%
ee
Config.
%
ee
Config.
%
ee
Config.
R. rubra KCh 4
R. rubra KCh 82
1
1
3
3
6
6
3
6
6
6
6
3
51
50
28
11
73
87
55
90
81
78
100
42
53
59
56
54
32
0
33
0
6
S
S
S
S
S
–
R
41
>99
>99
98
92
93
15
40
12
91
84
–
2R,4S
2R,4S
2R,4S
2R,4S
2R,4S
2S,4R
2R,4S
2R,4S
2S,4R
2S,4R
–
8
46
40
91
95
75
53
75
27
22
66
–
2S,4S
2S,4S
2S,4S
2S,4S
2R,4R
2R,4R
2S,4S
2S,4S
2R,4R
2R,4R
–
41
42
51
17
9
12
6
12
6
9
29
38
11
4
33
4
7
R. glutinis KCh 242
C. wiswanathi KCh 120
C. sake KCh 908
C. parapsilosis KCh 909
C. pelliculosa ZP22
S. cerevisiae KCh 464
S. brasiliensis KCh 905
S. pastorianus KCh 906
Y. lipolytica KCh 71
Z. bailii KCh 907
R
S
–
S
7
0
62
16
0
44
0
14
36
2R,4S
61
2R,4R

T. Janeczko et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 47–52
51
Table 2
Microbial transformation of cis-flavan-4-ol (5).
Microorganism
t [days]
cis-flavan-4-ol (5)
Flavanone (4)
ee
trans-flavan-4-ol (7)
%
ee
Config.
%
Config.
%
ee
Config.
R. rubra KCh 4
R. rubra KCh 82
6
6
6
6
6
1
6
3
74
86
66
83
76
49
94
75
22
15
40
17
17
97
8
2S,4S
2S,4S
2S,4S
2R,4R
2R,4R
2R,4R
2R,4R
2S,4S
7
52
42
59
70
75
93
52
35
S
S
S
S
S
S
S
R
19
10
23
6
5
2
93
86
88
21
7
56
28
27
2R,4S
2R,4S
2R,4S
2S,4R
2R,4S
2S,4R
2S,4R
2S,4R
4
11
11
19
49
4
R. glutinis KCh 242
C. wiswanathi KCh 120
C. sake KCh 908
C. parapsilosis KCh 909
C. pelliculosa ZP22
Y. lipolytica KCh 71
2
1
10
24
Table 3
Microbial transformation of trans-flavan-4-ol (7).
Microorganism
t [days]
trans-flavan-4-ol (7)
Flavanone (4)
ee
cis-flavan-4-ol (5)
%
ee
Config.
%
Config.
%
ee
Config.
R. rubra KCh 82
6
6
6
6
3
87
84
84
74
48
6
14
16
14
93
2S,4R
2S,4R
2S,4R
2S,4R
2S,4R
6
31
30
43
41
85
R
R
R
R
R
7
4
0
0
0
30
31
–
–
–
2R,4R
2R,4R
R. glutinis KCh 242
C. wiswanathi KCh 120
C. parapsilosis KCh 909
Y. lipolytica KCh 71
12
16
26
52
OH
O
O
OH
O
O
K2Cr2O7
H2SO4
K2Cr2O7
H2SO4
O
O
(2R,4R)-5
(R)-4
(2S,4R)-7
(S)-4
Scheme 2. Oxidation of (2R,4R)-cis-flavan-4-ol (5) and (2S,4R)-trans-flavan-4-ol (7) recovered from the bioreduction mixture.
the culture of this strain we observed almost complete oxidation of
(2S,4S)-enantiomer of the substrate to (S)-flavanone (4) (ee = 93%).
This highly enantioselective oxidation allowed us to obtain (2R,4R)-
cis-flavan-4-ol (5) with the enantiomeric excess of over 97%. The
capability of the C. parapsilosis strains for enantioselective oxida-
tion of S-alcohols was reported earlier for racemic 1-arylethanols
and allylic alcohols, which were deracemised to the (R)-enantiomer
(ee up to >99%) by whole cells of C. parapsilosis ATCC 7330 culture
[36–38].trans-Flavan-4-ol (7) underwent oxidation at much lower
rate compared to the cis-alcohol in the cultures of almost all tested
biocatalysts. In Table 3 we presented the results for five strains that
were able to transform trans-flavan-4-ol (7). For the other seven
species the substrate conversion was below 5%, even after 10 days
of incubation. The only strain that performed effective oxidation
of trans-flavan-4-ol (7) was Y. lipolytica KCh 71. In the culture of
this biocatalyst we observed the effective enantiospecific oxidation
of (2R,4S)-trans-flavan-4-ol (7) to (R)-flavanone (4) (ee = 85% after
three days). Due to the high enantioselectivity of oxidation by this
strain, after six days of biotransformation we identified also (2S,4R)-
trans-flavan-4-ol (7) with ee = 97%. This result was astonishing,
because when ( )-cis-flavan-4-ol (5) was subjected to biotrans-
formation by this strain, the selectivity of oxidation process was
low.
oxidized with Jones reagent to give (R)-flavanone and (S)-
flavanone, respectively. Enantiomeric excesses of the starting
alcohols and the obtained ketones were identical.
During biotransformation of trans-flavan-4-ol we observed a
selective oxidation of the alcohol with S configuration at C-4
(CH-OH). The same selectivity was observed in deracemiza-
tion of 1-aryl ethanol in the cultures of C. parapsilosis ATCC
7330 [36], Candida albicans CCT 0776 [37] and some bacterial
strains [39,40], which proceeded via enantioselective oxida-
tion of the (S)-enantiomer. In our previous research concerning
bioreduction of propiophenone we observed that during incubation
of propiophenone in the cultures of all tested microorgan-
isms, longer biotransformation time resulted in
a drop in
a percent content of the S-alcohol, along with an increase in a per-
cent content of the R-alcohol. This time-dependence of products
concentration was observed for microorganisms that possessed
either S-specific or R-specific dehydrogenases. Employment of
racemic 1-phenylpropan-1-ol as a substrate confirmed these find-
ings [41]. The ability to perform enantioselective oxidation of
R-alcohol was also observed for some plant biocatalysts (Arracacia
xanthorrhiza, Beta vulgaris, Zingiber officinale) [42] and few strains
of filamentous fungi (Aspergillus terreus CCT 4083) [43]. Such an
opposite enantioselectivity we found also in the case of oxidation
of racemic cis-flavan-4-ol (5) in the cultures of Rhodotorula strains.
It is also worth noting, that when ( )-trans-flavan-4-ol (7)
served as a substrate for the strains C. wiswanathi KCh 120, C.
parapsilosis KCh 909 and the strains of the genus Rhodotorula, oxi-
dation of (2R,4S)-trans-flavan-4-ol (7) was observed. Moreover, in
the cultures of the Rhodotorula strains concentration of (2R,4R)-cis-
flavan-4-ol (5) increased, as a result of reduction of (R)-flavanone
(4) (Scheme 2).
4. Conclusions
Twelve yeast strains were tested with respect to their capabil-
ity for enantioselective reduction of flavanone (4). The majority
of the strains demonstrated high enantioselectivity towards this
substrate. In the cultures of R. rubra KCh 4, R. rubra KCh 82, R.
glutinis KCh 242, Z. bailii KCh 907, and Candida sake KCh 908 the
(2R,4R)-cis-Flavan-4-ol (5) and (2S,4R)-trans-flavan-4-ol (7)
recovered from the preparative biotransformation mixture were

52
T. Janeczko et al. / Journal of Molecular Catalysis B: Enzymatic 109 (2014) 47–52
reduction rate was higher for the R-enantiomer of the substrate,
whereas in the culture of Candida pelliculosa ZP22 (S)-flavanone was
reduced faster. Reduction of the carbonyl group proceeded with a
very high enantioselectivity, leading to formation of (2R,4S)-trans-
flavan-4-ol (7) with ee > 99% in the culture of R. rubra (41% of the
reaction mixture after one day of biotransformation); with ee = 98%
in the culture of R. glutinis KCh 242 (42% after three days); and with
ee = 92% in the culture of C. wiswanathi KCh 120 (51% after three
days). Additionally, in the culture of the last strain after one day
of biotransformation we obtained also (2S,4S)-cis-flavan-4-ol (5)
(43%) with the enantiomeric excess of 96%. The same compound
was also identified in biotransformation with Candida pelliculosa
ZP22 (33% after three days with ee = 75%). Only in the culture of Z.
bailii KCh 907 (2R,4R)-cis-flavan-4-ol (5) was obtained in a decent
amount (44% after three days of substrate incubation, ee = 61%).
The strain C. parapsilosis KCh 909 was proven to be the catalyst
capable to separate ( )-cis-flavan-4-ol (5). Due to its enzymatic
system activity, after a one-day biotransformation it is possible to
obtain (S)-flavanone (4) with 49% of conversion and ee = 93%, along
with 49% of unreacted (2R,4R)-cis-flavan-4-ol (5) with ee = 97%.
Whereas, the strain Y. lipolytica KCh 71 was the only one which
effectively and enantioselectively oxidized ( )-trans-flavan-4-ol
(7). After three days of transformation we identified (R)-flavanone
(4) (ee = 85%), and after six days (2S,4R)-trans-flavan-4-ol (7) (45%)
with a high enantiomeric excess (ee = 97%). The results obtained
prove that C. parapsilosis KCh 909 and Y. lipolytica KCh 71 may
find application for separation of racemic mixtures of synthetic
flavanols.
[9] T. Morikawa, K. Funakoshi, K. Ninomiya, D. Yasuda, K. Miyagawa, H. Matsuda,
M. Yoshikawa, Chem. Pharm. Bull. 56 (2008) 956–962.
[10] H. Halbwirth, S. Kahl, W. Jaeger, G. Reznicek, G. Forkmann, K. Stich, Plant Sci.
170 (2006) 587–595.
[11] H. Halbwirth, Int. J. Mol. Sci. 11 (2010) 595–621.
[12] E.C. Bate-Smith, Phytochemistry 8 (1969) 1803–1810.
[13] F. Spinelli, J.-B. Speakman, W. Rademacher, H. Halbwirth, K. Stich, G. Costa, Eur.
J. Plant Pathol. 112 (2005) 133–142.
[14] H. Wei, G. Wu, D. Shi, S. Song, X. Zhang, Y. Lei, J. Ruan, Food Chem. 134 (2012)
1959–1966.
[15] J. Jiang, L. Tian, L. Wang, Y. Liu, Y. Chen, Biochem. Syst. Ecol. 50 (2013) 136–138.
[16] Z. Zhao, J. Jin, J. Ruan, C. Zhu, C. Lin, W. Fang, Y. Cai, J. Nat. Prod. 70 (2007)
1683–1686.
[17] M.-K. Lemke, P. Schwab, P. Fischer, S. Tischer, M. Witt, L. Noehringer, V.
Rogachev, A. Jäger, O. Kataeva, R. Fröhlich, P. Metz, Angew. Chem. Int. Ed. 52
(2013) 11651–11655.
[18] D. Zhao, B. Beiring, F. Glorius, Angew. Chem. Int. Ed. 52 (2013) 8454–8458.
[19] K.J. Hodgetts, Tetrahedron Lett. 42 (2001) 3763–3766.
[20] T. Izumi, K. Suenaga, J. Heterocyclic Chem. 34 (1997) 1535–1538.
[21] S. Ramadas, G.L.D. Krupadanam, Tetrahedron-Asymmetry 15 (2004)
3381–3391.
[22] T. Izumi, T. Hino, A. Kasahara, J. Chem. Soc. Perkin Trans. 1 (1992) 1265–1267.
[23] E. Kostrzewa-Susłow, J. Dmochowska-Gładysz, A. Białon´ ska, Z. Ciunik, J. Mol.
Catal. B-Enzym. 52–53 (2008) 34–39.
´
[24] E. Kostrzewa-Susłow, J. Dmochowska-Gładysz, T. Janeczko, K. Sroda, K. Micha-
lak, A. Palko, Z. Naturforsch. 67c (2012) 411–417.
[25] E. Kostrzewa-Susłow, J. Dmochowska-Gładysz, A. Białon´ ska, Z. Ciunik, W.
Rymowicz, J. Mol. Catal. B-Enzym. 39 (2006) 18–23.
[26] J.-E. Yeom, M.R. Kumar, S. Lee, J.-B. Lee, H.-R. Park, Bull. Korean Chem. Soc. 32
(2011) 4092–4094.
[27] N. Yadav, S.K. Dixit, A. Bhattacharya, L.C. Mishra, M. Sharma, S.K. Awasthi, V.K.
Bhasin, Chem. Biol. Drug Des. 80 (2012) 340–347.
[28] I. Zaleska, M. Piegza, Acta Sci. Pol. Biotechnol. 7 (2008) 27–43.
[29] T. Janeczko, J. Dmochowska-Gładysz, E. Kostrzewa-Susłow, World J. Microb.
Biol. 26 (2010) 2047–2051.
[30] L. Cappaert, C. Larroche, Biocatal. Biotransform. 22 (2004) 291–296.
[31] R.S. Martins, D.S. Zampieri, J.A.R. Rodrigues, P.O. Carvalho, P.J.S. Moran, Chem-
CatChem 3 (2011) 1469–1473.
[32] T. Janeczko, W. Gładkowski, E. Kostrzewa-Susłowa, J. Mol. Catal. B-Enzym. 98
(2013) 55–61.
Acknowledgments
[33] M.R.B. Chaves, P.J.S. Moran, J.A.R. Rodrigues, J. Mol. Catal. B-Enzym. 98 (2013)
73–77.
[34] E. Kostrzewa-Susłow, T. Janeczko, Molecules 17 (2012) 14810–14820.
[35] E. Kostrzewa-Susłow, T. Janeczko, Sci. World J. (2012), ID 254929, 8 p.
[36] T. Kaliaperumal, S.N. Gummadi, A. Chadha, Biocatal. Biotransform. 29 (2011)
262–270.
This research was supported financially by the European Union
within the European Regional Development Found (grant no.
POIG.01.03.01-00-158/09).
[37] S.M. Amrutkar, L. Banoth, U.C. Banerjee, Tetrahedron Lett. 54 (2013)
3274–3277.
References
[38] D. Titu, A. Chadha, Tetrahedron-Asymmetry 19 (2008) 1698–1701.
[39] S.M. Mantovani, C.F.F. Angolini, A.J. Marsaioli, Tetrahedron-Asymmetry 20
(2009) 2635–2638.
[40] L.S. Araújo, E. Kagohara, T.P. Garcia, V.H. Pellizari, L.H. Andrade, Mar. Drugs 9
(2011) 889–905.
[1] B. Winkel-Shirley, Curr. Opin. Plant Biol. 5 (2002) 218–223.
[2] S. Das, J.P.N. Rosazza, J. Nat. Prod. 69 (2006) 499–508.
[3] N.C. Cook, S. Samman, J. Nutr. Biochem. 7 (1996) 66–76.
[4] B.H. Havsteen, Pharmacol. Ther. 96 (2002) 67–202.
[5] P.C.H. Hollman, M.B. Katan, Food Chem. Toxicol. 37 (1999) 937–942.
[6] H. Arakawa, M. Nakazaki, Chem. Ind. 12 (1960) 73.
[41] T. Janeczko, E. Kostrzewa-Susłowa, Tetrahedron-Asymmetry (2014), http://dx.
doi.org/10.1016/j.tetasy.2014.08.006 (in press).
[7] E. Hardegger, H. Braunschweiger, Helv. Chim. Acta 44 (1961) 1413–1417.
[8] M. González-Cortazar, A.M. Maldonado-Abarca, E. Jiménez-Ferrer, S. Marquina,
E. Ventura-Zapata, A. Zamilpa, J. Tortoriello, M. Herrera-Ruiz, Molecules 18
(2013) 13260–13270.
[42] L.H. Andrade, R.S. Utsunomiya, A.T. Omori, A.L.M. Porto, J.V. Comasseto, J. Mol.
Catal. B-Enzym. 38 (2006) 84–90.
[43] J.V. Comasseto, L.H. Andrade, Á.T. Omori, L.F. Assis, A.L.M. Porto, J. Mol. Catal.
B-Enzym. 29 (2004) 55–61.