Screening, substrate specificity and stereoselectivity of yeast strains, which reduce sterically hindered isopropyl ketones

Article abstract of DOI:10.1016/j.tetasy.2006.12.013

Towards the synthesis of sterically hindered optically active secondary alcohol 2, yeast strains (Candida floricola IAM 13115 and Trichosporon cutaneum IAM 12206) with si-face hydride attack on isopropyl phenylsulfonylmethyl ketone 1 were developed by scr

Full text of DOI:10.1016/j.tetasy.2006.12.013

Tetrahedron: Asymmetry 17 (2006) 3358–3367
Screening, substrate specificity and stereoselectivity of yeast
strains, which reduce sterically hindered isopropyl ketones
Chihiro Hiraoka, Masaaki Matsuda, Yuya Suzuki, Shigeo Fujieda, Mina Tomita,
Ken-ichi Fuhshuku, Rika Obata, Shigeru Nishiyama and Takeshi Sugai*
Department of Chemistry, Keio University, Yokohama 223-8522, Japan
Received 30 October 2006; accepted 12 December 2006
Available online 16 January 2007
Abstract—Towards the synthesis of sterically hindered optically active secondary alcohol 2, yeast strains (Candida floricola IAM 13115
and Trichosporon cutaneum IAM 12206) with si-face hydride attack on isopropyl phenylsulfonylmethyl ketone 1 were developed by
screening. Strains with complementary re-facial selectivity (Pichia angusta IAM 12895 and Pichia minuta IAM 12215) were also found.
Based on the substrate specificity studies of these four strains, microbial reduction was applied to the synthesis of (3S,5S)-2,6-dimethyl-
3,5-heptanediol 12a.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
O
OH
H3C
L
H3C
Many kinds of enantiomerically enriched secondary alco-
hols are now available via biocatalytic methods.1–11 So
far, bakers’ yeast and other microbial strain-catalyzed
asymmetric reduction, and lipase-catalyzed enantiomeric
resolution are very effective towards substrates with one
small substituent, for example, methylcarbinols.
L
L
CH3
CH3
OH
OH
O O
S
(R)-2
(S)-2
yeast
strain
O
O O
S
On the other hand, the asymmetric reduction of highly ste-
rically hindered open-chain ketones still remains a chal-
O O
S
lenging task. Although there has been
a reported
1
reduction of isopropyl ketones (Scheme 1),12–15 the sub-
strates are limited to ketoesters. Herein, we report an
exploitation of yeast strains, which enable the preparation
of enantiomerically enriched forms of isopropyl carbinols
often encountered as the partial structure of naturally
occurring products and chiral auxiliaries in organic
chemistry.
Scheme 1.
ogy; IAM), in glucose medium were applied to the model
substrate 1 (Scheme 1). Ketone 1 was designed along the
following lines: (1) A bulky isopropyl group and a phen-
ylsulfonyl group are directly attached to both ends of the
carbonyl group. (2) The electron-withdrawing phenylsulfo-
nyl group is expected to promote reduction. (3) The UV-
absorbing phenylsulfonyl group would facilitate detection
of the progress of the reduction as well as determination
of the ee by HPLC analysis. (4) The sulfonyl group would
assist C–C bond formation. So far, only lipase-catalyzed
dynamic kinetic resolution of a racemic similar sulfonyl
alcohol has been reported.16
2. Results and discussion
2.1. Screening of yeast strain on sulfonylketone 1
Towards this end, grown cells of more than 30 yeast strains
from a culture collection (Institute of Applied Microbiol-
*
Corresponding author. E-mail: sugai@chem.keio.ac.jp
0957-4166/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2006.12.013

C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
3359
Table 1. Screening of the yeast strain on the reduction of 1
O
O
O
Yeast strain
IAM No. % ee Absolute
of 2a configuration
of 2
S
L
1
Trichosporon cutaneum
Candida floricola
12206
13115
12231
12885
92.8
89.1
87.4
71.0
61.3
52.6
29.7
28.3
23.1
9.2
98.0
95.0
85.9
49.4
45.4
35.0
25.7
15.7
12.6
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(R)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
Rhodotorula minuta
Rhodotorula aurantiaca
O
OMe
OMe
OH
OMe
OMe
(a)
Rhodosporidium sphaerocarpum 12261
H3C
Candida boidinii
Pichia henricii
Bullera alba
12269
12891
13475
04982
12214
12215
12895
04980
12883
10896b
13471
04947
04857
04125
S
3
4
Pichia finlandica
Pichia jadinii
Pichia minuta
Scheme 2. (a) Four yeast strains used in this study.
Pichia angusta
Williopsis californica
Candida nitratophila
Yamadazyma farinosa
Fellomyces polyborus
Yarrowia lipolytica
Candida kefyr
O
OH
(a)
(c)
(b)
(d)
SO2Ph
SO2Ph
(S)-2, 96.8% ee
1
OH
OH
Saccharomyces cerevisiae
a Daicel ChiralCel OD-H, see Section 4.4.
b NBRC No.
SO2Ph
SO2Ph
5
6
First, the preliminary screenings under the conditions with
50 mg of the substrate in test tube incubation were carried
out. From 33 strains, nineteen exhibited the progress of the
expected reduction (Table 1). Trichosporon cutaneum IAM
12206 and Candida floricola IAM 13115 yielded (R)-2 and
in contrast, both Pichia minuta IAM 12215 and Pichia
angusta IAM 12895 provided (S)-2, in highly enantio-
merically enriched states. Yamadazyma farinosa NBRC
10896,17 which had shown catalytic activities towards the
related sulfonyl methyl and trifluoromethyl ketones,
only resulted in as low as 45.4% ee. The yeast strains
showed diverse facial selectivity, as had been shown in a
sulfonyl ketone18 with different substituents and an a-
ketoester.19
OR
(R)-7a: R = H
7b: R = MTPA (S)
(e)
Scheme 3. Reagents and conditions: (a) P. minuta IAM 12215, 92%; (b)
LDA (2 equiv), THF, then allyl iodide (10 equiv), ꢀ78 ꢁC to rt, 76%; (c)
H2, Pd–C, EtOH, 99%; (d) Li (5 equiv), ethylenediamine (7 equiv), Et2O,
0 ꢁC to rt, 26%; (e) (R)-MTPA chloride.
ture was gradually increased from ꢀ78 ꢁC to room temp
and the allylated product 5 was afforded in 76% yield.
The smooth reaction required as much as 10 equiv of allyl
iodide, even under an elevated temperature and the reac-
tion was rather slow, probably due to the presence of a ste-
rically hindered isopropyl group. The terminal double
bond was then hydrogenated to provide saturated sulfone
6. Birch reduction using a combination of lithium
(5 equiv)–ethylenediamine (7 equiv) in diethyl ether21 was
applied to remove the phenylsulfonyl group of 6. The very
volatile product, 2-methyl-3-heptanol 7a was identified,
and (R)-absolute configuration was confirmed by a com-
parison of its (+)-sign of rotation with authentic data22
As already mentioned, isopropylketone 1 was designed for
the screening of yeast enzymes, which are able to reduce
sterically hindered carbonyl groups. A suggested high affin-
ity towards hydrophobic substrates15 was supported by the
following result, the attempted reduction of a less hindered
acyclic ketone 3 as the candidate (Scheme 2). Ketone 3 has
so far been a good substrate for whole-cell mediated reduc-
tion by Y. farinosa NBRC 10896 to give (R)-4 in 94.0%
ee.20 The above-mentioned four strains developed in this
study, however, did not work on this substrate 3.
1
and H NMR analysis23 of the corresponding (S)-MTPA
ester 7b. It was concluded that the absolute configuration
of alcohol 2 obtained by the P. minuta-mediated asymmet-
ric reduction was (S).
2.2. Determination of absolute configuration of product 2
Since the enantiomerically enriched forms of the resulting
alcohol 2 are not known in any literature, the absolute con-
figuration of the enantiomerically enriched form was deter-
mined by derivatization as shown in Scheme 3.
2.3. Towards the scale-up of the microbial reduction of
sulfonylketone 1
Needless to say, an important aspect is the high catalytic
activity, which is required for a whole-cell reduction en-
zyme. In this study, at the initial screening, a glucose con-
centration as high as 5% (w/v) was applied, so that yeast
strains, which can grow and provide a high amount of cell
(+)-Alcohol 2, which was obtained by reduction with P.
minuta IAM 12215, was treated with LDA (2 equiv) and
then allyl iodide (10 equiv) in THF. The reaction tempera-

3360
C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
mass, were selected. Next, the large-scale production condi-
tions for (R)- and (S)-alcohol 2 with high catalytic activities
were further elaborated upon. For the (R)-isomer, C. flori-
cola IAM 13115 was pre-incubated in glucose medium
(pH 6.0) and ketone 1 (substrate concentration: 0.5 w/v%)
was added. During the attempted scale-up reduction with
500 mg of the substrate, the incubation broth gradually
turned acidic (pH 3.9), with the yield as low as 16%. More-
over, the ee of product 2 declined slightly to 86%. Occa-
sionally adjusting the pH of the broth to 6.0 and
lowering the substrate concentration (0.2%) was effective
to enhance the conversion, so as to achieve 79% yield. In
this case, the lower ratio between the substrate and wet cell
weight (w/wcw, 0.03) also obviously affected the smooth
reaction, compared with the original value (0.2). Finally,
under a mechanically pH-controlled aerobic environment
by adding antifoam, with a plentiful oxygen supply, a high-
er yield (93%) with a higher 91.9% ee of (R)-2 was
obtained.
2.4. Substitution of the phenylsulfonyl group with alicyclic
and aliphatic groups
Next, we modified the phenylsulfonyl group in 1. Two new
ketones 8 and 9 were the candidates, which have alicyclic
and aliphatic ketones, respectively. A small modification
from the phenylsulfonyl group to a similar six-membered
ring as 8 brought about no reduction on the isopropyl ke-
tone at all, towards 10. In the case of C. floricola IAM
13115 the enzyme worked, however, the reduction on the
pro-R carbonyl group located in the cyclohexane ring pre-
dominated and two new products were isolated24 in a com-
bined yield of 88% (Scheme 4).
O
O
O
O
O
O O
S
O
9
1
8
There are some comments on another yeast strain, T.
cutaneum IAM 12206, for the production of the (R)-iso-
mer. An initial attempt for the use of the stationary phase
harvested cells of T. cutaneum after prolonged incubation
(48 h) only resulted in a low yield of the desired alcohol 2
(6%), and the ee of the product was unexpectedly low
(88%). In contrast, direct introduction of substrate 1 after
48 h incubation without any harvesting of the cells im-
proved the yield to 19% and the ee to 93.0%. In an inde-
pendent experiment, the growth curve determination of
this strain indicated the logarithmic growing phase termi-
nates after 24 h, at an OD (660 nm) of 1.4. This result sug-
gested that the desired reductive enzyme expresses at a
higher level in younger growing phase cells. The introduc-
tion of the substrate (0.2%) at 24 h and the continuing
incubation were indeed effective, to give 38% yield. Final-
ly, the addition of antifoam towards the acceleration of
oxygen uptake further enhanced the yield to 60% of (R)-
2 with 94.0% ee.
pro-R
H
OH
O
O
H
O
(a)
O
O
H
10
8
OH
O
OH
(b)
O
+
O
O
Scheme 4. (a) Four yeast strains in this study; (b) C. floricola IAM 13115.
In contrast, substitution by another bulky and electron-
withdrawing isopropylcarbonyl group, the diketone-type
substrate 9, was accepted by all four strains of yeast as de-
scribed above (Scheme 5). The reduction stopped at the
stage of hydroxyketone 11. This result means that the sub-
stitution of the phenylsulfonyl group in 1 with an isoprop-
ylcarbonyl group was acceptable, but with the
isopropylhydroxymethyl group in 11, it did not work.
The results were in contrast to the fact that a similar
hydroxyketone, 4-hydroxy-2-pentanone, was a good sub-
strate to Y. farinosa NBRC 10896 to give diol.25
In contrast, excessive oxygen exposure showed a deleteri-
ous effect on the growth of P. minuta IAM 12215, and it
was preferable to avoid the use of the antifoam at the stage
of pre-incubation. The scaled-up reduction was reproduc-
ible with several grams of substrate by applying harvested
wet cells (w/wcw, 0.13), and (S)-2 of 96.8% ee was obtained
in 92% yield. Although the pH of the medium turned acidic
during the reaction (ca. 3.9), high enzyme activity was
exhibited and mechanical pH adjustment was not neces-
sary. With these three strains, the preparative-scale reduc-
tion to both (R)- and (S)-2 was established, and the
results are summarized in Table 2.
As the resulting hydroxyketone 11 was a rather labile com-
pound, the enantiofacial selectivity was evaluated at the
Table 2. Comparison of facial selectivity between ketones 1 and 9
Strain (IAM No.)
Product 2 from ketone 1
Product 11 from ketone 9
Absolute configuration
Yield (%)
% ee
Absolute configuration
Yield (%)
% ee
C. floricola (13115)
T. cutaneum (12206)
P. angusta (12895)
P. minuta (12215)
(R)
(R)
(S)
(S)
93
60
32
92
91.9
94.0
95.0
96.8
(S)
(S)
(S)
(R)
36
12
12
36
92.2
97.6
43.2
59.4

C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
3361
C. floricola
T. cutaneum
O
O
OH
O
OH O
(a)
H
H
O
O
O
(S)-11
or
(R)-11
9
O
O
S
Ph
OH OH
OH OH
OH OH
(b)
1
9
(3S,5S)-12a
(3R,5R)-12a
meso-12a (syn)
P. angusta
(anti)
H
OBz OBz
OBz OBz
O
OH
O
O
O
(c)
O
O
S
Ph
(3S,5S)-12b
(3R,5R)-12b
11
H
1
9
Scheme 5. Reagents and conditions: (a) Four yeast strains in this study;
(b) Me4NHB(OAc)3, MeCN, AcOH, ꢀ20 ꢁC; (c) BzCl, pyridine.
P. minuta
O
O
O
O
O
later stages as follows: the partially purified product 11 was
immediately reduced with tetramethylammonium triacet-
oxyborohydride (Evans’ reagent)26 to give the enantiomeri-
cally enriched form of anti-diol 12a as the major product,
accompanied with a small amount of meso (syn)-12a (over
9:1). anti-Diol 12a was then converted to the corresponding
dibenzoate 12b and was analyzed by HPLC with Daicel
ChiralCel OD-H. The ee of 12b represents the enantiofacial
preference of the yeast-mediated reduction at the stage of
the formation of 11.
S
Ph
H
H
1
9
Figure 1. Change of enantiofacial selectivity in yeast-mediated reduction.
O
O
OH
O
(a)
The results of the enantiofacial selection at the step of yeast
enzyme-catalyzed reduction on 9 are summarized in Table
2, together with the results on the reduction of 1. Three of
the four yeast strains showed the same enantiofacial prefer-
ence; however, P. angusta IAM 12895 showed a reversal of
preference. In this case, it was supposed that plural
enzymes, which have opposite selectivity, compete in the
whole cell. One enzyme with re-facial selectivity on sub-
strate 1, and another with an inverted facial selectivity
worked preferentially on 9 in whole cells, which are incu-
bated under the same conditions (Fig. 1). At present, there
is no information with regards to the type of coenzymes,
such as NADH or NADPH.
11
9
OH OH
OH OH
(b)
meso-12a
(syn)
(3S,5S)-12a
(3R,5R)-12a
(anti)
Scheme 6. Reagents: (a) Four yeast strains in this study; (b) NaBH4.
strains was consistent (Table 3). C. floricola was the best,
from the standpoint of the feasibility of the incubation
and reproducibility in scaled-up conditions (63.4% ee,
61% yield) among the four strains.
So far, the chemical asymmetric reduction of 927–29 has
been elaborated towards 12a (DMHD), a useful chiral aux-
iliary.30,31 It was noted that the purification of the volatile
and considerably labile hydroxyketone 11, even in a partial
state, lowers the two-step combined yield of yeast-cata-
lyzed and Evans’ reduction. We then considered Singh’s
one-pot approach32 towards the diol, and an excess
amount of NaBH4 was directly added to the yeast-reduc-
tion broth after the consumption of starting material 9,
without any isolation of intermediate 11 (Scheme 6). Even
though the Evans’ reagent was switched with NaBH4 for
the second reduction, the preferential formation of anti-
diol 12a was still observed. All of the yields in the one-
pot approach were improved over those of the aforemen-
tioned step-by-step conversion. The ees of the product were
substantially lower, but the enantiofacial preference for
Towards the enantiomerically pure (3S,5S)-12a, a comple-
mentary use of microbial reduction and an enzyme-cata-
lyzed kinetic resolution was elaborated. There were no
attempts on the enzyme reaction of the diol 12a carrying
large isopropyl substituents attached on both hydroxy-
methyl carbons, and indeed, the enzyme-catalyzed acyl-
ation was very slow. We then introduced two electron-
withdrawing acyl groups33 to provide the bis(chloroace-
tate) 12c.
A mixture of (3S,5S)-12c and (3R,5R)-12c (81.7:18.3), pre-
pared from 12a (63.4% ee), was incubated with several

3362
C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
Table 3. Comparison between step-by-step and one-pot reduction of 9
3. Conclusion
Strain (IAM No.)
Step-by-step
One-pot
Yield (%)
So far, in both enzyme-catalyzed asymmetric reduction and
the resolution of racemate, difficulties have often been
experienced in the access to enantiomerically enriched
forms of hindered alcohols due to the low reactivity caused
by the sterically bulky group directly attached to the stereo-
genic centre. The present newly developed yeast strains, as
a method for the preparation of enantiomers, would be a
clue in how to overcome these problems.
Yield (%)
% ee
% ee
C. floricola (13115)
T. cutaneum (12206)
P. angusta (12895)
P. minuta (12215)
36
12
12
36
92.2
97.6
43.2
59.4
61
22
17
55
63.4
81.6
32.0
69.4
commercially available hydrolytic enzymes. Our first candi-
date was Bacillus licheniformis protease (subtilisin), and the
enzyme preferentially hydrolyzed the minor enantiomer,
(3R,5R)-12c. The hydrolysis, however, always resulted in
an incomplete manner, giving as low as 77.8% ee for the de-
sired (3S,5S)-isomer as the unaffected substrate (Scheme 6).
As from a preliminary independent experiment, the differ-
ence in the initial hydrolytic rate between (3R,5R)-12c
and (3S,5S)-12c was very high. We concluded that the less
reactive (3R,5R)-12c worked as the competitive inhibitor34
with almost no kcat. However, it showed a similar Km with
the preferred (3S,5S)-enantiomer.
4. Experimental
4.1. Materials and methods
Merck silica gel 60 F254 thin-layer plates (1.05744, 0.5 mm
thickness) and silica gel 60 (spherical and neutral; 100–
210 lm, 37560-79) from Kanto Chemical Co. were used
for preparative thin-layer chromatography and column
chromatography, respectively. Peptone and yeast extract
were purchased from Kyokuto Pharmaceutical Co., for
the cultivation of microorganism. Yeast strains are avail-
able from, the Institute of Applied Microbiology Culture
Collection; the Institute of Molecular and Cellular Biosci-
ences, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku,
Tokyo 113-0032, Japan.
Besides the enzymes with no enantioselectivity on 12c (pig
pancreatic lipase, Candida rugosa lipase: Meito OF), and
Candida antarctica lipase: Chirazyme L-2 with very low
activity on 12c, there was found Pseudomonas cepacia li-
pase PS-C showing the contrasting enantio-preference with
subtilisin.35 In this case, we were successful in isolating the
(3S,5S)-isomer in its enantiomerically pure state as diol
12a in 55% yield by enzyme-catalyzed hydrolysis (Scheme
7). Moderate catalytic activity and enantioselectivity of
yeast-catalyzed reduction of sterically hindered diketone
9 were overcome, by elaborating upon the combination
with the subsequent one-pot reduction procedure and
lipase-catalyzed kinetic resolution to give pure (3S,5S)-
12a.
4.2. Analytical methods
All mps are uncorrected. IR spectra were measured as films
for oils or KBr disks of solids on a Jasco FT/IR-410 spec-
trometer. H NMR spectra were measured in CDCl3 at
270 MHz on a Jeol JNM EX-270 or at 400 MHz on a Jeol
JNM GX-400 spectrometer. HPLC data were recorded on
Jasco PU-2080 and MD-2010 liquid chromatographs.
Optical rotation values were recorded on a Jasco DIP
360 and P-1010 polarimeter.
1
4.3. Preparation of substrate: 3-methyl-1-phenylsulfonyl-2-
butanone 1
OH OH
OCA OCA
OCA OCA
(a)
To
a mixture of methyl phenyl sulfone (100 mg,
0.640 mmol) and THF (2.0 mL) cooled to ꢀ78 ꢁC was
added n-BuLi (1.56 M in hexane, 0.86 mL, 1.3 mmol).
After stirring for 1 h, 2-methylpropanoyl chloride
(80.0 lL, 0.764 mmol) was added to this solution at
ꢀ78 ꢁC. Stirring was continued for 1 h at ꢀ78 ꢁC. The dis-
appearance of the starting material was confirmed by a
TLC analysis [silica gel, developed with hexane–AcOEt
(1:1)]. The reaction mixture was poured into saturated
aqueous NH4Cl solution and extracted with AcOEt. The
combined organic layer was washed with brine, dried over
anhydrous Na2SO4, and concentrated in vacuo. The resi-
due was charged on a silica gel column (8 g). Elution with
hexane–AcOEt (4:1) afforded 1 (132 mg, 85%).
(3S,5S)-12a
(3S,5S)-12c
(3R,5R)-12c
(63.4% ee)
81.7 : 18.3
OCA OCA
OH OH
(b)
CA =
ClCH2CO
(3S,5S)-12c
(3R,5R)-12a
(77.8% ee)
OH OH
OCA OCA
(3S,5S)-12c
(c)
81.7
:
18.3
Further purification by recrystallization from hexane–
AcOEt afforded 1 (114 mg, 73%) as colourless needles;
mp 63.5–64.5 ꢁC (lit.36 mp 66.5–67 ꢁC); IR mmax 2936,
(3R,5R)-12c
(3S,5S)-12a
(3R,5R)-12c
(100% ee)
1
1714, 1449, 1383, 1319, 1151, 784, 687 cmꢀ1; H NMR d:
Scheme 7. Reagents: (a) (ClCH2CO)2O, pyridine; (b) Bacillus licheniformis
protease (subtilisin); (c) Pseudomonas cepacia lipase PS-C.
1.11 (d, J = 6.9 Hz, 6H), 2.92 (sept, J = 6.9 Hz, 1H), 4.23

C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
3363
(s, 2H), 7.56 (dd, J = 7.1, 7.4 Hz, 2H), 7.65 (t, J = 7.4 Hz,
1H), 7.88 (d, J = 7.1 Hz, 2H). Its IR and NMR spectra
were identical with those reported previously.36
through a pad of Celite. The combined organic extracts
were washed with brine, dried over anhydrous Na2SO4
and concentrated in vacuo. The residue was charged on a
silica gel column (20 g). Elution with toluene–AcOEt
(12:1) and the further purification of some contaminated
4.4. Screening of microorganisms
fractions with preparative TLC with CHCl3–MeOH
20
The screening of microorganisms was performed as fol-
lows. The microorganisms from stock culture samples were
incubated in glucose medium [containing glucose (500 mg),
peptone (200 mg), yeast extract (50 mg), KH2PO4 (30 mg),
K2HPO4 (20 mg), at pH 6.5, total volume of 10 mL, in the
test tube] for 2 days at 30 ꢁC. Then ketone 1 (50 mg) and
glucose (500 mg) were added and the mixture was shaken
on a reciprocal shaker (250 cpm) for 2 days at 30 ꢁC. The
progress of the reduction was confirmed by TLC analysis
[silica gel, developed with hexane–AcOEt (1:1)]. Each reac-
tion mixture was separately worked up in the same way as
described below. The reaction mixture was filtered through
a Celite pad and extracted with AcOEt. The combined
organic layer was washed with saturated aqueous NaHCO3
solution and brine, dried over anhydrous Na2SO4 and con-
centrated in vacuo. The crude residue containing product 2
and unreacted starting material 1 was filtered through a
short column of silica gel, analyzed by the HPLC analysis
[column, Daicel Chiralcel OD-H, 0.46 cm · 25 cm; hexane–
i-PrOH (15:1); flow rate 0.5 mL/min]: tR (min) = 29.0 for
(R)-2, 32.9 for (S)-2, 36.7 for 1. Ee of the product 2 and
conversion were estimated by this HPLC analysis.
(19:1) afforded (R)-2 (189.9 mg, 93%). ½aꢁD ¼ ꢀ17:8 (c
0.98, EtOH). HPLC: tR (min) = 31.6 (96.0%), 37.7
(4.1%); 91.9% ee. IR: mmax 3521, 2964, 2877, 1446, 1304,
1146, 1086 cmꢀ1; H NMR: d 0.89 (d, J = 4.5 Hz, 6H),
1
1.69–1.81 (m, 1H), 3.18 (s, 1H), 3.20 (d, J = 2.1 Hz, 1H),
3.24 (d, J = 2.3 Hz, 1H), 3.93–4.00 (m, 1H), 7.57–7.72
(m, 2H), 7.93–7.96 (m, 3H). Anal. Calcd for C11H16O3S:
C, 57.87; H, 7.06. Found: C, 58.04; H, 7.031.
4.5.3. Incubation of T. cutaneum IAM 12206 and the
preparation of (R)-(ꢀ)-2. T. cutaneum IAM 12206 were
grown on the broth with the same ingredients with C. flori-
cola (total volume of 100 mL) on a gyratory shaker
(180 rpm) for 12 h at 30 ꢁC. Then the antifoam AF emul-
sion (100 mg/mL, 0.5 mL) was added and the mixture
was shaken for a further 12 h at 30 ꢁC. Then substrate 1
(205 mg, 0.91 mmol), which was pre-treated as above,
was introduced and the incubation was continued for
24 h. The extraction of products from the supernatant
and cell mass, workup and purification were carried out
in the same manner as described for C. floricola-mediated
20
reaction, to give (R)-2 (123.2 mg, 60%). ½aꢁD ¼ ꢀ16:7 (c
0.98, EtOH). The ee was estimated from HPLC analysis.
HPLC: tR (min) = 30.9 (97.0%), 36.7 (3.0%); 94.0% ee.
Its IR and NMR spectra were identical with those men-
tioned above.
4.5. Scaled-up reduction of 1
4.5.1. Pre-incubation of C. floricola IAM 13115. A small
portion of yeast cells of C. floricola IAM 13115 grown on
the agar-plate culture was aseptically inoculated to a glu-
cose medium [containing glucose (20 g), peptone (8.0 g),
yeast extract (2.0 g), KH2PO4 (1.2 g), K2HPO4 (0.8 g), at
pH 6.5, total volume of 400 mL] in four 500-mL baffled
Erlenmeyer cultivating flasks and then shaken on a gyra-
tory shaker (180 rpm) for 35 h at 30 ꢁC. The wet cells were
harvested by centrifugation (3000 rpm) and washed with
phosphate buffer (0.1 M, pH 6.5). The weight of combined
wet cells was ca. 13 g from 200 mL of the broth.
4.5.4. Incubation of P. minuta IAM 12215 and the prepa-
ration of (S)-(+)-2. The combined wet cells of P. minuta
IAM 12215 (23 g), incubated as described for C. floricola,
were re-suspended in a reaction medium [containing glu-
cose (72 g), phosphate buffer (0.1 M, pH 6.0), total volume
of 1438 mL] in a 5000-mL baffled Erlenmeyer cultivating
flask, together with 1 (2.88 g, 12.7 mmol), which was pre-
treated as above, and shaken on a gyratory shaker
(180 rpm) for 27.5 h at 30 ꢁC. Then the broth was centri-
fuged (3000 rpm). The extraction of products from the
supernatant and cell mass, workup and purification were
carried out in the same manner as described for C. flori-
4.5.2. Preparation of (R)-(ꢀ)-3-methyl-1-phenylsulfonyl-2-
butanonol 2 by C. floricola IAM 13115 mediated reac-
tion. The combined wet cells of C. floricola IAM 13115
(6.3 g) incubated as described above, were re-suspended
in a reaction medium [containing glucose (5.0 g), phos-
phate buffer (0.1 M, pH 6.0), total volume of 100 mL] in
a 200-mL round-bottomed flask, together with 1 (202 mg,
0.89 mmol), which was finely ground in advance. The mix-
ture was stirred for 42 h at 30 ꢁC. Throughout the reaction,
its pH was kept at 6.0 by the occasional addition of aque-
ous NaOH solution (2.0 M, total 10 mL) with an automatic
pH controller. The broth was then centrifuged (3000 rpm).
The supernatant was saturated with NaCl and mixed with
AcOEt (100 mL). The mixture was stirred for 1 h and fil-
tered through a pad of Celite. The organic layer of the fil-
trate was separated and the aqueous layer was further
extracted with AcOEt. On the other hand, the precipitated
cell mass at the stage of centrifugation was mixed with ace-
tone (5 mL). The mixture was stirred for 1 h and filtered
cola-mediated reaction, to give (S)-2 (2.67 g, 92%).
22
½aꢁD ¼ þ18:0 (c 0.97, EtOH). HPLC: tR (min) = 32.4
(1.6%), 36.3 min (98.4%); 96.8% ee. Anal. Calcd for
C11H16O3S: C, 57.87; H, 7.06. Found: C, 57.97; H, 7.038.
Its IR and NMR spectra were identical with those men-
tioned above.
4.6. Determination of absolute configuration of 2
4.6.1. (3S,4RS)-2-Methyl-4-phenylsulfonyl-6-heptene-3-ol 5.
To an LDA solution, separately provided by the addition
of n-BuLi (2.71 M in hexane, 7.0 mL, 19 mmol) to a solu-
tion of i-Pr2NH (2.9 mL, 21 mmol) in THF (26 mL) at
0 ꢁC, a solution of (+)-2 (2.1 g, 9.0 mmol, prepared by P.
minuta-mediated reduction) in THF (20 mL) was added
at ꢀ78 ꢁC and the mixture was stirred for 30 min. Then
\allyl iodide (8.2 mL, 91 mmol) was added and the mixture

3364
C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
stirred further at ꢀ78 ꢁC for 30 min, and allowed to warm
to room temperature over 2 h. The reaction was quenched
with saturated aqueous NH4Cl solution and the mixture
extracted with AcOEt. The organic layer was washed with
brine, dried over anhydrous Na2SO4 and concentrated in
vacuo. The residue was charged on a silica gel column
4.6.4. (2S,10R)-10-Isopropylpentyl 3,3,3-trifluoro-2-methoxy-
2-phenylpropanoate 7b. (R)-MTPA chloride (27 mg,
0.107 mmol) was added with stirring to a solution of (R)-
7a (6 mg, 0.046 mmol) in pyridine (0.46 mL). After stirring
overnight at room temp, the reaction was quenched with
water (5 mL) and the mixture was extracted with AcOEt.
The combined organic layer was washed with brine, dried
over anhydrous Na2SO4, and concentrated in vacuo. The
residue was dissolved in toluene (2 mL), concentrated in
vacuo twice for removing pyridine, and purified by pre-
parative TLC with hexane–AcOEt (9:1) to afford 7b
(14.4 mg, 99%). 1H NMR d: 0.83 (d, J = 6.4 Hz, 3H),
0.85 (d, J = 6.1 Hz, 3H), 0.88 (t, J = 7.0 Hz, 3H), 1.20–
1.40 (m, 4H), 1.52–1.72 (m, 2H), 1.85–1.97 (m, 1H), 3.56
(s, 1H), 4.94 (dt, J = 4.6, 6.0 Hz, 1H), 7.37–7.42 (m, 3H),
7.55–7.58 (m, 2H). An authentic sample from racemic 7a
showed the additional signals ascribable to the (10S)-iso-
mer: 0.82 (t, J = 7.0 Hz, 3H), 0.91 (d, J = 6.9 Hz, 3H),
0.92 (d, J = 6.9 Hz, 3H), 1.10–1.37 (m, 4H), 1.49–1.69
(m, 2H), 1.85–2.00 (m, 1H), 3.56 (s, 1H), 4.97 (m, 1H),
7.38–7.42 (m, 3H), 7.55–7.56 (m, 2H). The ee of product
7a from the microbial origin was 96%, judging from their
NMR spectra.
(170 g). Elution with hexane–AcOEt (4:1) afforded
24
(3S,4RS)-5 (1.8 g, 76%). ½aꢁD ¼ þ175:3 (c 1.04, EtOH).
IR mmax 3527, 2962, 1446, 1302, 1146, 1084 cmꢀ1 1H
;
NMR d: 0.72 (d, J = 6.8 Hz, 3H), 1.00 (d, J = 6.4 Hz,
3H), 1.81 (dqq, J = 9.1, 6.4, 6.8 Hz, 1H), 2.66 (m, 2H),
2.95 (d, J = 1.8 Hz, 1H), 3.20 (t, J = 5.4 Hz, 1H), 3.77
(dd, J = 1.8, 9.1 Hz, 1H), 4.95 (dd, J = 1.3, 9.7 Hz, 1H),
5.00 (dd, J = 1.5, 16.0 Hz, 1H), 5.70 (m, 1H), 7.64 (m,
3H), 7.91 (m, 2H).
4.6.2. (3S,4RS)-2-Methyl-4-phenylsulfonyl-3-heptanol 6.
The alcohol (3S,4RS)-5 (1.7 g, 6.5 mmol) was dissolved
in EtOH (64 mL) and then 10% Pd–C (340 mg) was added.
The mixture was vigorously stirred under H2 (1 atm) at
room temperature for 15 h. The reaction mixture was fil-
tered through a pad of Celite, and the Celite pad was
washed with AcOEt. The combined organic solution was
concentrated in vacuo to afford 6 (1.7 g, 99%) as a diaste-
24
reomeric mixture (ca. 3:1). ½aꢁD ¼ þ14:7 (c 1.01, EtOH).
4.7. Reduction of 9: step-by-step procedure
IR mmax 3525, 2962, 1300, 1146, 1084, 729 cmꢀ1 1H
;
NMR d: major diastereomer; 0.70 (d, J = 6.8 Hz, 3H),
0.86 (t, J = 7.3 Hz, 3H), 0.99 (d, J = 6.6 Hz, 3H), 1.38
(m, 2H), 1.74 (dqq, J = 9.6, 6.6, 6.8 Hz, 1H), 1.88 (m,
2H), 2.98 (d, J = 2.0 Hz, 1H), 3.08 (t, J = 5.1 Hz, 1H),
3.66 (dd, J = 2.0, 9.6 Hz, 1H), 7.55–7.71 (3H, m), 7.89–
7.92 (m, 2H); minor diastereomer; 0.79 (t, J = 7.3 Hz,
3H), 0.94 (d, J = 6.8 Hz, 3H), 0.99 (d, J = 6.6 Hz, 3H),
1.38 (m, 2H), 1.58 (m, 2H), 2.02 (dqq, J = 5.2, 6.6,
6.8 Hz, 1H), 3.18 (dt, J = 5.1, 5.0 Hz, 1H), 3.44 (d,
J = 5.4 Hz, 1H), 3.72 (ddd, J = 5.1, 5.2, 5.4 Hz, 1H),
7.55–7.71 (m, 3H), 7.89–7.92 (m, 2H).
4.7.1. (3S,5S)-(ꢀ)-2,6-Dimethyl-3,5-heptanediol 12a by C.
floricola IAM 13315. The combined wet cells of C. flori-
cola (10 g) incubated as described above were re-suspended
in a reaction medium [containing glucose (5 g), phosphate
buffer (0.1 M, pH 6.0), total volume of 100 mL] in a 300-
mL round-bottomed flask, together with 6 (200 mg,
1.28 mmol). The reaction mixture was stirred for 28 h at
30 ꢁC. Antifoam AF emulsion (100 mg/mL, 0.5 mL) was
then added to the reaction mixture and the mixture was
azeotropically distilled in vacuo. The distillate was satu-
rated with NaCl and extracted with Et2O. The combined
organic extracts were washed with brine, dried over anhy-
drous Na2SO4 and concentrated in vacuo. The residue
was employed for the next step immediately without fur-
ther purification.
4.6.3. (R)-2-Methyl-3-heptanol 7a. A solution of the alco-
hol (3S,4RS)-6 (694 mg, 2.6 mmol) in Et2O (5.2 mL) and
ethylenediamine (1.2 mL, 18 mmol) was cooled to 0 ꢁC.
The mixture was degassed to remove any trace of O2 by
applying ultrasonic (70–80 W) under evacuation. Lithium
shot (Wako, 126-04913, 97 mg, 14 mmol) was added under
the flow of N2 and the mixture was vigorously stirred at
0 ꢁC. The color of the reaction mixture immediately turned
brown. After the proper period, the reaction was quenched
by the addition of saturated aqueous NH4Cl solution and
6 M HCl to adjust its pH to be 6. The mixture was then ex-
tracted several times with Et2O. The combined extract was
washed with water, 2 M aqueous NaOH solution, and
brine and dried over anhydrous Na2SO4 with activated
charcoal. The solvent was carefully removed by fractional
distillation at atmospheric pressure. The residue was puri-
fied by preparative TLC with hexane–AcOEt (4:1) and fur-
A solution of the crude product in anhydrous MeCN
(0.85 mL) was added slowly with stirring under Ar to a
solution of Me4NHB(OAc)3 (1.11 g, 4.2 mmol) in anhy-
drous MeCN (3.4 mL) and anhydrous AcOH (3.4 mL) at
ꢀ20 ꢁC. After stirring for 2 h at ꢀ20 ꢁC, the reaction was
quenched by the addition of saturated aqueous sodium
potassium tartrate solution (2 mL) and stirred vigorously
for 30 min. The mixture was extracted with AcOEt, and
the organic layer was washed with saturated aqueous NaH-
CO3 solution and brine, dried over anhydrous Na2SO4 and
concentrated in vacuo. The residue was charged on a silica
gel column (7 g). Elution with hexane–AcOEt (6:1) and the
further purification of some contaminated fractions with
preparative TLC with hexane–AcOEt (2:1), which was
developed twice, afforded (3S,5S)-12a (73.7 mg, 36%) as a
colourless solid; mp 88.5–90.0 ꢁC; IR mmax 3336, 2958,
ther by micro-distillation apparatus to afford (R)-7a
24
(87 mg, 26%). Bp 120 ꢁC/80 mmHg; ½aꢁD ¼ þ25:7 (c 0.46,
EtOH) {lit.22 [a]D = +27.7 (c 10, EtOH)} IR mmax 3359,
1
1
2958, 2933, 2873, 1468, 991 cmꢀ1; H NMR: d 0.91 (m,
1471, 1333, 1146, 1038, 795, 704 cmꢀ1; H NMR d: 0.91
9H), 1.29–1.48 (m, 6H), 1.57–1.73 (m, 1H), 3.36 (br s,
1H); 13C NMR d: 14.2, 17.2, 19.0, 22.9, 28.3, 33.5, 33.9,
76.7.
(d, J = 6.8 Hz, 6H), 0.96 (d, J = 6.6 Hz, 6H), 1.59 (dd,
J = 5.6, 5.7 Hz, 2H), 1.72 (dqq, J = 5.9, 6.6, 6.8 Hz, 2H),
2.34 (s, 2H), 3.63 (ddd, J = 5.6, 5.7, 5.9 Hz, 2H); 13C

C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
3365
20
NMR d: 18.16, 18.77, 33.79, 36.54, 74.20; ½aꢁD ¼ ꢀ53:2 (c
500-mL baffled Erlenmeyer cultivating flasks) incubated
as described above were re-suspended in the same reaction
medium as above in a 500-mL round-bottomed flask,
together with 9 (400 mg, 2.56 mmol). After stirring for
14.5 h at 30 ꢁC, the reaction medium was cooled to 15 ꢁC
and NaBH4 (381 mg, 10.1 mmol) was occasionally added
over 20 min, while monitoring the internal reaction temper-
ature. Then the reaction mixture was centrifuged
(3000 rpm). The supernatant was saturated with NaCl
and mixed with AcOEt (ca. 130 mL). The mixture was stir-
red for 1 h and filtered through a pad of Celite. The organic
layer of the filtrate was separated and the aqueous layer
was further extracted with AcOEt. On the other hand,
the precipitated cell mass at the stage of centrifugation
was mixed with acetone (50 mL). The mixture was stirred
for 1 h and filtered through a pad of Celite. The combined
organic extracts were washed with brine, dried over anhy-
drous Na2SO4 and concentrated in vacuo. The residue
was charged on a silica gel column (4 g). Elution with hex-
ane–AcOEt (6:1) and further purification of some contam-
inated fractions, preparative TLC with hexane–AcOEt
(2:1, developed twice) afforded (3S,5S)-12a (249.1 mg,
61%) and syn-12a (58.3 mg, 14%).
25
1.01, MeOH) {lit.27 ½aꢁD ¼ ꢀ64:5 (c 1.00, MeOH)}. The
ee of (3S,5S)-12a was determined by HPLC analysis of
the corresponding dibenzoate 12b. HPLC [column, Daicel
Chiralcel OD-H, 0.46 cm · 25 cm; hexane–i-PrOH (19:1);
flow rate 0.5 mL/min]: tR (min) = 9.3 (96.1%), 10.3
1
(3.9%); 92.2% ee. Dibenzoate 12b; H NMR d: 0.89 (d,
J = 6.7 Hz, 6H), 0.92 (d, J = 6.7 Hz, 6H), 1.96 (dqq,
J = 5.9, 6.7, 6.7 Hz, 2H), 1.98 (dd, J = 5.5, 5.6 Hz, 2H),
5.21 (ddd, J = 5.5, 5.6, 5.9 Hz, 4H), 7.27 (dd, J = 7.3,
7.5 Hz, 4H), 7.41 (tt, J = 1.3, 7.3 Hz, 2H), 7.88 (dd,
J = 1.3, 7.5 Hz, 4H).
4.7.2. (3S,5S)-(ꢀ)-12a by T. cutaneum IAM 12206. T.
cutaneum grown on the broth with the same ingredients
with C. floricola (total volume of 100 mL) on a gyratory
shaker (180 rpm) for 12 h at 30 ꢁC. Then antifoam AF
emulsion (100 mg/mL, 0.5 mL) was added and the mixture
shaken for a further 12 h at 30 ꢁC. Then substrate 9
(200 mg, 1.28 mmol) was introduced and the incubation
was continued for 24 h. Extraction of the product and
the subsequent Evans’ reduction were carried out in the
same manner as described for C. floricola-mediated reac-
20
tion, to give (3S,5S)-12a (25.3 mg, 12%); ½aꢁD ¼ ꢀ52:3 (c
20
0.97, MeOH). The ee of (3S,5S)-12a was estimated from
HPLC analysis of 12b in the same manner as described pre-
viously; tR (min) = 8.6 (1.2%), 9.5 (98.8%); 97.6% ee. Its IR
and NMR spectra were identical with those described
above.
Compound (3S,5S)-12a; ½aꢁD ¼ ꢀ38:3 (c 1.00, MeOH);
HPLC of 12b tR (min) = 8.4 min (18.3%), 9.4 min
(81.7%); 63.4% ee. Its IR and NMR spectra were identical
with those described above. Compound syn-12a: IR mmax
3357, 2960, 1468, 1385, 1151, 1063, 852, 604 cmꢀ1 1H
;
NMR d: 0.92 (d, J = 6.9 Hz, 12H), 1.41 (dt, J = 14.0,
10.2 Hz, 1H), 1.59 (dt, J = 14.0, 2.0 Hz, 1H), 1.67 (m,
2H), 3.12 (s, 2H), 3.62 (ddd, J = 2.0, 5.1, 10.2 Hz, 2H).
4.7.3. (3S,5S)-(ꢀ)-12a by P. angusta IAM 12895. P. angu-
sta was grown on the broth with the same ingredients with
C. floricola (total volume of 200 mL) on a gyratory shaker
(180 rpm) for 36 h at 30 ꢁC. The combined wet cells (5 g)
were re-suspended in the same reaction medium as above,
in a 300-mL round-bottomed flask, together with 9
(200 mg, 1.28 mmol), and the incubation was continued
for 43 h. The extraction and the subsequent reduction were
carried out in the same manner as described for C. flori-
4.8.2. (3S,5S)-(ꢀ)-12a by T. cutaneum IAM 12206. T.
cutaneum was grown in 100 mL of the broth in total 24 h
at 30 ꢁC as described above. Substrate 9 (200 mg,
1.28 mmol) was incubated for 24 h at 30 ꢁC. The subse-
quent reduction with NaBH4, workup and purification
were done in the same manner as described above to give
cola-mediated reaction, to give (3S,5S)-12a (23.9 mg,
(3S,5S)-12a (45.6 mg, 22%) and syn-12a (7.3 mg, 4%).
20
20
12%); ½aꢁD ¼ ꢀ26:0 (c 0.99, MeOH); HPLC of 12b tR
Compound (3S,5S)-12a: ½aꢁD ¼ ꢀ40:2 (c 0.80, MeOH);
(min) = 8.6 (28.4%), 9.5 (71.6%); 43.2% ee.
HPLC of 12b tR (min) = 8.4 (9.2%), 9.4 (90.8%); 81.6%
ee. The NMR spectrum of syn-12a was identical with those
described above.
4.7.4. (3R,5R)-(+)-12a by P. minuta IAM 12215. P. min-
uta was grown on the broth with the same ingredients with
C. floricola (total volume of 200 mL) on a gyratory shaker
(180 rpm) for 36 h at 30 ꢁC. The combined wet cells (10 g)
were re-suspended in a reaction medium [containing glu-
cose (5 g), phosphate buffer (0.1 M, pH 6.0), total volume
of 100 mL] in a 300-mL round-bottomed flask, together
with 9 (200 mg, 1.28 mmol). The mixture was stirred for
36 h at 30 ꢁC. The extraction of the products and the sub-
sequent Evans’ reduction were carried out in the same
4.8.3. (3S,5S)-(ꢀ)-12a by P. angusta IAM 12895. Dike-
tone 9 (200 mg, 1.28 mmol) was incubated with the har-
vested wet cells of P. angusta (6 g from two 500-mL
baffled Erlenmeyer cultivating flasks) for 70 h at 30 ꢁC,
and then further reduced with NaBH4 affording (3S,5S)-
12a (34.0 mg, 17%) and syn-12a (12.5 mg, 6.1%). Com-
20
pound (3S,5S)-12a; ½aꢁD ¼ ꢀ20:6 (c 0.50, MeOH); HPLC
of 12b tR (min) = 8.5 min (34.0%), 9.5 min (66.0%);
manner as described for C. floricola-mediated reaction, to
32.0% ee.
20
give (3R,5R)-12a (74.2 mg, 36%); ½aꢁD ¼ þ39:5 (c 1.00,
MeOH); HPLC of 12b tR (min) = 8.8 (79.7%), 9.9
4.8.4. (3R,5R)-(+)-12a by P. minuta IAM 12215. Dike-
tone 9 (200 mg, 1.28 mmol) was incubated with the har-
vested wet cells of P. minuta (8 g from two 500-mL
baffled Erlenmeyer cultivating flasks) for 12 h at 30 ꢁC.
The subsequent reduction with NaBH4, workup and puri-
fication were done in the same manner as described above
to give (3R,5R)-12a (112.7 mg, 55%) and syn-12a (21.9 mg,
(20.3%); 59.4% ee.
4.8. Reduction of 9: one-pot two step procedure
4.8.1. (3S,5S)-(ꢀ)-12a by C. floricola IAM 13315. The
combined wet cells of C. floricola IAM (40 g from five

3366
C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
20
11%). Compound (3R,5R)-12a: ½aꢁD ¼ þ34:6 (c 1.00,
MeOH); HPLC of 12b tR (min) = 8.3 (84.7%), 9.2
(15.3%); 69.4% ee.
Acknowledgements
We thank Professors Tadashi Ema, Okayama University,
Mitsuru Abo, the University of Tokyo, and Jon D. Stew-
art, University of Florida, for valuable suggestions on this
work, and Amano Enzyme Inc. for lipase PS-C. This work
was accomplished as the 21st century COE Program
(KEIO LCC), also supported by ‘SORST: Solution-Ori-
ented Research for Science and Technology’ of Japan Sci-
ence and Technology Corporation, and we express our
sincere thanks to Professors Keisuke Suzuki, Takashi Mat-
sumoto and Dr. Ken Ohmori of Tokyo Institute of
Technology.
4.9. Enantiomeric enhancement of (3S,5S)-12a
4.9.1. (3S,5S)-2,6-Dimethyl-3,5-heptanediol bis(chloroacet-
yl)ester 12c. To a solution of (3S,5S)-12a (102.9 mg,
0.62 mmol) in anhydrous pyridine (1.0 mL) was added
chloroacetic anhydride (267.0 mg, 1.56 mmol) and the mix-
ture was stirred at room temperature for 1 h. Then the mix-
ture was poured into phosphate buffer (1.0 mL, 0.1 M,
pH 6.5) and extracted with AcOEt. The combined organic
layer was washed with 2 M HCl, saturated NaHCO3 solu-
tion and brine, dried over anhydrous Na2SO4 and concen-
trated in vacuo. The residue was charged on a silica gel
column (15 g). Elution with hexane–AcOEt (20:1) afforded
References
1. Fuhshuku, K.; Oda, S.; Sugai, T. Recent Res. Develop. Org.
Chem. 2002, 6, 57–74.
1
12c (190.7 mg, 94.8%); H NMR d: 0.91 (d, J = 6.8 Hz,
6H), 0.92 (d, J = 6.8 Hz, 6H), 1.80 (dd, J = 5.6, 5.6 Hz,
2H), 1.98 (dqq, J = 7.9, 6.8, 6.8 Hz, 2H), 4.04 (s, 4H),
4.86 (ddd, J = 5.5, 5.6, 7.9 Hz, 2H), 13C NMR d: 17.8,
18.1, 31.6, 32.0, 41.0, 76.1, 167.1. This was employed for
the next step without further purification.
2. Reetz, M. T. Tetrahedron 2002, 58, 6595–6602.
3. Pellissier, H. Tetrahedron 2003, 59, 8291–8327.
4. Nakamura, K.; Yamanaka, R.; Matsuda, T.; Harada, T.
Tetrahedron: Asymmetry 2003, 14, 2659–2681.
´
5. Palmies, O.; Ba¨ckvall, J.-E. Trends Biotechnol. 2004, 22, 130–
135.
6. Ghanem, A.; Aboul-Enein, H. Y. Tetrahedron: Asymmetry
2004, 15, 3331–3351.
4.9.2. Subtilisin-catalyzed hydrolysis of 12c. To a mixture
of (3S,5S)-12c [63.4% ee, (3S,5S):(3R,5R) = 81.7:18.3;
20.0 mg, total 0.064 mmol] phosphate buffer (0.5 mL,
0.2 M, pH 7.0) was added subtilisin (Sigma, P-5380,
50 mg) and the mixture stirred at 40 ꢁC for 2 days. Then
the insoluble materials were filtered off and the filtrate
was extracted with AcOEt. The combined organic layer
was washed with brine, dried over anhydrous Na2SO4
and concentrated in vacuo. The residue was purified by
preparative TLC [developed with hexane–AcOEt (4:1)] to
give diol 12a (1.1 mg, 11%), 12c (10.9 mg, 55%) and the
corresponding mono(chloroacetate) 12d (1.3 mg, 9%).
The recovered 12c was hydrolyzed by applying K2CO3 in
MeOH to give 12a, and then converted to dibenzoate
12b. HPLC tR (min) = 8.4 (11.1%), 9.4 (88.9%); 77.8% ee.
´
7. Garcıa-Urdiales, E.; Alfonso, I.; Gotor, V. Chem. Rev. 2005,
105, 313–354.
8. Wiktelius, D. Synlett 2005, 2113–2114.
9. Stewart, J. D. Adv. Appl. Microbiol. 2006, 59, 31–52.
10. Bode, S. E.; Wolberg, M.; Muller, M. Synthesis 2006, 557–
¨
588.
11. Bhowmick, K. C.; Joshi, N. N. Tetrahedron: Asymmetry
2006, 17, 1901–1929.
12. Ishihara, K.; Yamaguchi, H.; Hamada, H.; Nakamura, K.;
Nakajima, N. J. Mol. Catal. B: Enzyme 2000, 10, 419–428.
13. Ishihara, K.; Yamaguchi, H.; Hamada, H.; Nakamura, K.;
Nakajima, N. Biosci. Biotechnol. Biochem. 2002, 66, 588–597.
14. Inoue, K.; Makino, Y.; Itoh, N. Tetrahedron: Asymmetry
2005, 16, 2539–2549.
15. Zhu, D.; Yang, Y.; Hua, L. J. Org. Chem. 2006, 71, 4202–
4205.
4.9.3. P. cepacia lipase-catalyzed hydrolysis of 12c. To a
mixture of (3S,5S)-12c [63.4% ee, as above, 108.6 mg, total
0.68 mmol] phosphate buffer (2.5 mL, 0.2 M, pH 7.0) was
added lipase (Amano PS-C, 200 mg) and the mixture was
stirred at 40 ꢁC for 3 days. Then the enzyme was filtered
off and the filtrate was extracted with AcOEt. After a sim-
ilar workup as above, the residue was charged on a silica
gel column (5 g). Elution with hexane–AcOEt (10:1) affor-
ded diol 12a (30.3 mg, 55%), 12c (42.8 mg, 39%) and 12d
(3.2 mg, 4%); mp 89.5–90 ꢁC (lit.27 mp 89–91 ꢁC);
16. Kielbasinski, P.; Rachwalski, M.; Mikolajczyk, M.; Moe-
lands, M. A. H.; Zwanenburg, B.; Rutjes, F. P. J. T.
Tetrahedron: Asymmetry 2005, 16, 2157–2160.
17. Sugai, T. Curr. Org. Chem. 1999, 3, 373–406.
18. Robin, S.; Huet, F.; Fauve, A.; Veschambre, H. Tetrahedron:
Asymmetry 1993, 4, 239–246.
19. Bergo de Lacerda, P. S.; Ribeiro, J. B.; Leite, S. G. F.;
Ferrara, M. A.; Coelho, R. B.; Bon, E. P. S.; da Silva Lima,
E. L.; Antunes, O. A. C. Tetrahedron: Asymmetry 2006, 17,
1186–1188.
20. Yamazaki, T.; Kuboki, A.; Ohta, H.; Mitzel, T.; Paquette,
L. A.; Sugai, T. Synth. Commun. 2000, 30, 3061–3072.
21. Shindo, T.; Fukuyama, Y.; Sugai, T. Synthesis 2004, 692–700.
22. Pickard, R. H.; Kenyon, J. J. Chem. Soc. 1913, 103, 1923–
1959.
23. Kusumi, T.; Yabuuchi, T.; Takahashi, H.; Ooi, T. Yukigosei
Kagaku Kyokaishi (J. Synth. Org. Chem., Jpn.) 2005, 63,
1102–1114.
22
20
½aꢁD ¼ ꢀ64:3 (c 1.01, MeOH) {lit.27 ½aꢁD ¼ ꢀ64:5 (c 1.0,
MeOH)}. Anal. Calcd for C9H20O2: C, 67.45; H, 12.58.
Found: C, 67.06; H, 12.26. HPLC of 12b tR (min) = 9.4
(100%) and no peak was detected at 8.4 min; 100% ee. Its
IR and NMR spectra were identical with those of authentic
sample.
24. Fujieda, S.; Tomita, M.; Fuhshuku, K.; Ohba, S.; Nishiyama,
S.; Sugai, T. Adv. Synth. Catal. 2005, 347, 1099–1109.
25. Ikeda, H.; Sato, E.; Sugai, T.; Ohta, H. Tetrahedron 1996, 52,
8113–8122.
26. Evans, D. A.; Chapman, K. T. Tetrahedron Lett. 1986, 27,
5939–5942.
Bis(chloroacetate) 12c and mono(chloroacetate) 12d were
combined and hydrolyzed by applying K2CO3 in MeOH
to give 12a, and then converted to dibenzoate 12b. HPLC
tR (min) = 8.4 (46.1%), 9.4 (55.9%); 11.8% ee as (3S,5S)-
isomer.

C. Hiraoka et al. / Tetrahedron: Asymmetry 17 (2006) 3358–3367
3367
27. Sugimura, T.; Yoshikawa, M.; Yoneda, T.; Tai, A. Bull.
Chem. Soc. Jpn. 1990, 63, 1080–1082.
32. Ahmad, K.; Koul, S.; Taneja, S. C.; Singh, A. P.; Kapoor,
M.; Riyaz-ul-Hassan; Verma, V.; Qazi, G. N. Tetrahedron:
Asymmetry 2004, 15, 1685–1692.
33. Suzuki, M.; Nagasawa, C.; Sugai, T. Tetrahedron 2001, 57,
4841–4848.
34. Kuboki, A.; Okazaki, H.; Sugai, T.; Ohta, H. Tetrahedron
1997, 53, 2387–2400.
35. Ema, T. Curr. Org. Chem. 2004, 8, 1009–1025.
36. Thomsen, M. W.; Handwerker, B. M.; Katz, S. A.; Belser,
R. B. J. Org. Chem. 1988, 53, 906–907.
28. Tai, A.; Kikukawa, T.; Sugimura, T.; Inoue, Y.; Abe, S.; Osawa,
T.; Harada, T. Bull. Chem. Soc. Jpn. 1994, 67, 2473–2477.
29. Marinetti, A.; Geneˆt, J.-P.; Jus, S.; Blanc, D.; Ratoveloma-
nana-Vidal, V. Chem. Eur. J. 1999, 5, 1160–1165.
30. Sugimura, T.; Yoshikawa, M.; Futagawa, T.; Tai, A.
Tetrahedron 1990, 46, 5955–5966.
31. Sugimura, T.; Yamasaki, A.; Okuyama, T. Tetrahedron:
Asymmetry 2005, 16, 675–683.