The scope and limitation of the regio- and enantioselective hydrolysis of aliphatic epoxides using Bacillus subtilis epoxide hydrolase, and exploration toward chirally differentiated tris(hydroxymethyl)methanol

Article abstract of DOI:10.1016/j.tetasy.2010.07.014

The substrate specificity of an engineered Bacillus subtilis epoxide hydrolase, which so far had shown high activity and enantioselectivity with 1-benzyloxymethyl-1-methyloxirane, has been studied by altering the methyl substituent into hydrogen, oxygen-containing functionalities, and unsaturated homologs. High enantioselectivity (E = 44) was observed with 1-benzyloxymethyl-1-vinyloxirane with a proper catalytic activity. The elaboration of the reaction conditions and work-up procedures enabled a preparative-scale kinetic resolution, to give (R)-2-benzyloxymethyl-3-butene-1, 2-diol and its antipodal (R)-epoxide in high ees.

Full text of DOI:10.1016/j.tetasy.2010.07.014

Tetrahedron: Asymmetry 21 (2010) 2043–2049
Contents lists available at ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
The scope and limitation of the regio- and enantioselective hydrolysis
of aliphatic epoxides using Bacillus subtilis epoxide hydrolase, and exploration
toward chirally differentiated tris(hydroxymethyl)methanol
*
Ken-ichi Shimizu, Maki Sakamoto, Manabu Hamada, Toshinori Higashi, Takeshi Sugai, Mitsuru Shoji
Faculty of Pharmacy, Keio University, 1-5-30, Shibakoen, Minato-ku, Tokyo 105-8512, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The substrate specificity of an engineered Bacillus subtilis epoxide hydrolase, which so far had shown high
activity and enantioselectivity with 1-benzyloxymethyl-1-methyloxirane, has been studied by altering the
methyl substituent into hydrogen, oxygen-containing functionalities, and unsaturated homologs. High
enantioselectivity (E = 44) was observed with 1-benzyloxymethyl-1-vinyloxirane with a proper catalytic
activity. The elaboration of the reaction conditions and work-up procedures enabled a preparative-scale
kinetic resolution, to give (R)-2-benzyloxymethyl-3-butene-1,2-diol and its antipodal (R)-epoxide in high ees.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 9 June 2010
Accepted 2 July 2010
Available online 10 August 2010
1. Introduction
H3C
BnO
H3C OH
O
BnO
OH
Recent extensive studies on epoxide hydrolases (EH) of various
microbial origin promote new robust entries for providing enanti-
omerically enriched epoxides themselves and the corresponding
diols which would serve as the starting materials and intermedi-
ates for medicines and many biologically active substances. We
have investigated a recombinant enzyme from Bacillus subtilis
(BSEH), which hydrolyzes racemic epoxide 1 in a highly enantiose-
lective manner (E = 73).1
In this epoxide, (S)-1 is more reactive than the (R)-isomer and
diol (R)-2 is produced with a total retention of absolute configura-
tion at the tertiary stereogenic center, due to the attack of a nucle-
ophile such as water at the less hindered terminal position
(Scheme 1). The products, enantiomerically pure epoxide (R)-1
and antipodal diol (R)-2, were used as the starting materials for
the synthesis of (R)-bicalutamide and taurospongin A.1
:OH2
1
2
(S)-
(R)-
(a)
fast isomer
( )-
H3C
H3C
O
O
BnO
BnO
(R)-1
(R)-1
slow isomer
Scheme 1. Enzyme-catalyzed hydrolysis of racemic epoxide 1. Reagents and
conditions: (a) Bacillus subtilis epoxide hydrolase (BSEH), 30 °C, 7 d, 51% conv.,
E = 73.
sis on the specific substrate 1. This enzyme works even at high con-
centration [1.0 M of ( )-1] on the (S)-isomer, and is weakly
inhibited by (R)-1, the antipode of the proper substrate (S)-1.
Our first examination was the BSEH-catalyzed hydrolysis of the
less hindered secondary epoxides 3–5, by substituting the methyl
group with hydrogen in original substrate 1. Epoxides ( )-3–5 were
subjected to the reaction with harvested whole cells of B. subtilis
expressing epoxide hydrolases (Tamy 2 strain). We were very sur-
prised to see that the enantioselectivity was quite low when com-
pared with 1, with only a small change in the structure between 1
and 3–5.
One of the most important aspects for any catalyst is its scope
and limitation. To clarify, we prepared several aliphatic epoxides
3–8 (Fig. 1) by altering the substituents on the oxirane ring and/
or the protective group on the hydroxymethyl group, and describe
the substrate specificity study in the BSEH-catalyzed hydrolysis.
2. Results and discussion
2.1. Attempts at the kinetic resolution of secondary epoxides
The source of the present epoxide hydrolase was found in B.
subtilis by screening microorganisms which carry out the hydroly-
The order of susceptibility of the enzyme-catalyzed hydrolysis
for each substrate is summarized in Table 1. It is reasonable to
state that the rate of hydrolysis of the sterically less hindered
(R)-3 (Scheme 2) was higher than that of (R)-1, because of the de-
crease in steric hindrance (entries 4 and 6). In contrast, it was
* Corresponding author. Tel./fax: +81 3 5400 2695.
E-mail address: shoji-mt@pha.keio.ac.jp (M. Shoji).
0957-4166/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2010.07.014

2044
K. Shimizu et al. / Tetrahedron: Asymmetry 21 (2010) 2043–2049
H
H
OH
H
O
O
BnO
4-H3C(C6H4)H2CO
BnO
OH
20% ee
2% ee
3
4
(R)-9
H
(a)
O
H
BnO
O
TBDMSO
H
3
( )-
O
5
BnO
(R)-3
HO
BnO
AcO
BnO
O
O
O
O
Scheme 2. Enzyme-catalyzed hydrolysis of racemic epoxide 3. Reagents and
conditions: (a) BSEH, 30 °C, 6 d, 10% conv., E = 1.5.
6a
7
6b
8
A slight modification at the opposite end of the reactive site in
(S)- and (R)-4, where a CH3 group was attached on the p-position of
the benzyl protective group brought about an increase in the reac-
tion rate for (S)-4 (entries 2 and 3). Disappointingly, the difference
in the reaction rates between the enantiomers was still as low as
E = 3.5 (entries 2 and 5). Finally, both enantiomers of bulky sub-
strates, (S)- and (R)-5, with a TBDMS instead of a benzyl group
were very poor toward our BSEH. In summary, our attempts for
the enantioselective hydrolysis of the secondary epoxides bearing
a similar aliphatic skeleton to the original tertiary epoxide were
unsuccessful.
BnO
BnO
Figure 1. Seven designed substrates as probes for the analysis of the reaction
pathway and enantioselectivity.
interesting to note that the rate of hydrolysis of the sterically less
hindered (S)-3 (fast isomer) was lower than that of the original ter-
tiary epoxide (S)-1 (entries 1 and 3), which was confirmed by an
independent experiment in which (S)- or (R)-3 were separately
incubated with BSEH. This observation would be ascribable to
the ‘mobility’ of the substrates when they occupied the catalytic
site of BSEH. In the case of the ‘fast’ enantiomer (S)-3, its sterically
less hindered property may help the terminal oxirane ring to ‘es-
cape’ from the aspartate residue which is responsible for the
hydrolysis. On the other hand, even in the case of slow isomer
(R)-3, the terminus can meet the aforementioned catalytic site,
through the flexible nature of the substrate.
2.2. Tertiary epoxides with further advanced functionalities
The above insufficient results prompted us to design new ter-
tiary epoxides with further advanced functionalities in order to
consider new approaches to chirally differentiated tris(hydroxy-
methyl)methanol and synthetic equivalents 10b, 11, and 12
(Fig. 2). These diols would be derived by the enantioselective
hydrolysis with B. subtilis on epoxides 6b (for 10b), 7 (for 11),
and 8 (for 12). All the substrates are tertiary epoxides and have
certain advanced functional group introduced on the methyl group
of 1.
Table 1
The order of susceptibility of enzyme-catalyzed hydrolysis
AcO
BnO
OH
OH
OH
Entry
Substrate
Relative ratea
E-valueb
OH
BnO
OH
BnO
OH
H3C
O
O
O
O
O
O
*
*
*
C6H5H2CO
1
73
73
10b
11
12
(S)-1
p-CH3(C6H4)H2CO
(S)-4
Figure 2. Tris(hydroxymethyl)methanol equivalents to be obtained by enzyme-
catalyzed hydrolysis.
H
H
H
H
2
3
4
5
6
20
8.6
5.7
5.7
1
3.5
1.5
—
The preparation of the racemic forms of epoxides 6–8 is shown
in Scheme 3. The key intermediate was one of the substrates itself,
6a. Starting from benzyl glycidyl ether 3,2 the ring opening reaction
with p-methoxybenzyl alcohol and the subsequent 2-azaadaman-
tane-N-oxyl (AZADO) oxidation3 of the resulting 13 provided an
unsymmetrically protected dihydroxyacetone 14 in 73% yield.
Although the Wittig reaction of ketone 14 followed by an epoxida-
tion did not proceed efficiently, epoxide ring formation with a con-
comitant one carbon elongation was achieved by the action of
dimethyloxosulfonium methylide.4 The PMB ether ( )-15 was trea-
ted with DDQ under neutral conditions to give the requisite alcohol
( )-6a in 91% yield. Acetylation and Parikh–Doering oxidation5
produced 6b and 16 in 84 and 90% yield, respectively. Aldehyde
16 was further derived to alkenyl epoxide 7 by Wittig reaction in
92% yield. Toward the alkynyl epoxide 8, the first step of Corey–
Fuchs acetylene formation, that is, the formation of the terminal
dibromoalkene was successful by the action of PPh3 and CBr4.
The next step, dehydrobromination under basic conditions did
not provide any satisfactory results. We then switched to an alter-
native route; the introduction of a two carbon acetylene unit prior
C6H5H2CO
(S)-3
C6H5H2CO
(R)-3
p-CH3(C6H4)H2CO
(R)-4
—
H3C
C6H5H2CO
—
(R)-1
a
Based on the progress of the reaction and enantioselectivity for racemic sub-
strates 1, 3, and 4.
b
Calculated for substrates (S)-1, 3, and 4 over their enantiomers.

K. Shimizu et al. / Tetrahedron: Asymmetry 21 (2010) 2043–2049
2045
O
OH
(a)
(b)
O
BnO
BnO
OPMB
BnO
OPMB
13
3
14
(c)
PMBO
BnO
RO
BnO
(d)
O
O
( )-15
6a
( )-
R = H
(e)
( )-6b R = Ac
(f)
(g)
OHC
BnO
O
O
6a
14
BnO
( )-
16
( )-
7
H
TMS
(k)
(h)
OH
O
BnO
BnO
OR
8
( )-
17a
17b
( )-
( )-
R = PMB
R = H
(i)
(j)
17c
( )-
R = Ms
Scheme 3. Preparation of racemic epoxides 6–8. Reagents and conditions: (a) PMBOH, 40% NaOH aq, THF, 50 °C, 43 h; (b) 1-Me-AZADO, KBr, NaClO, CH2Cl2, satd NaHCO3 aq,
0 °C, 30 min (73% over two steps); (c) trimethylsulfoxonium iodide, n-BuLi, DMSO, rt, 25 min (93%); (d) DDQ, CH2Cl2, 0.1 M phosphate buffer (pH 7.0), rt, 2 h (91%); (e) Ac2O,
pyridine, rt, 45 min (87%); (f) sulfur trioxide-pyridine complex, Et3N, DMSO, CH2Cl2, 0 °C, 2 h (90%); (g) methyltriphenylphosphonium bromide, t-BuOK, THF, 0 °C, 15 min
(92%), (h) trimethylsilylacetylene, n-BuLi, THF, ꢀ78 to 0 °C, 1.5 h (97%); (i) CAN, CH3CN, H2O, rt, 3 h (94%); (j) MsCl, Et3N, CH2Cl2, ꢀ20 to 0 °C, 1 h; (k) K2CO3, MeOH, rt, 30 min
(70% over two steps).
to the epoxide ring formation. Ketone 14 was treated with lithium
trimethylsilylacetylide to give alkyne 17a. The selective removal of
the PMB ether6 followed by mesylation on the primary alcohol of
17b and cyclization furnished the desired epoxide 8 in 66% yield
from 17a.
Our first attempts were the incubation of the hydroxymethylat-
ed epoxide ( )-6a and the corresponding acetate 6b (Scheme 4).
The enzyme-catalyzed reaction, however, was quite slow.
lytic residues,7 for example, aspartate and the molecular substrate.
Insertion of the methylene group at the opposite or inside position
on the secondary epoxide (S)-3, ‘pushes’ the terminal of the sub-
strate molecule closer to the catalytic residue, to give (S)-4 and
(S)-1 higher reactivity. We expected further enhancement of the
reactivity by the insertion of –O– and –OCOCH2– groups between
the methyl C–H bond in (S)-1, but the results were disappointing.
We assumed that some hydrogen bond formation between polar
functional groups and the enzyme protein makes the molecules
far removed from the catalytic site, in the case of 6a, 6b, and 16.
The above hypothesis was partially correct as was shown by the
experiment when the hydrophobic substrates 7 and 8 were ap-
plied. In these cases, the enzyme-catalyzed hydrolysis was satisfac-
tory, and the reactions proceeded in enantioselective manners as
shown in Scheme 5.
RO
BnO
(a)
RO
BnO
OH
O
OH
10a R = H
( )-6a
R = H
10b R = Ac
( )-6b R = Ac
Under the same conditions where the conversion was 51%
(E = 73) for ( )-1a, the hydrolysis proceeded in 29.6% conversion
for ( )-7 (E = 44) and in 12.5% for ( )-8 (E = 28). Although the steri-
cally less hindered alkynyl epoxide 8 was anticipated to show a
higher reactivity than that of alkenyl epoxide 7, the hydrolysis of 7
proceeded faster than that of 8. These results indicated that the vinyl
group could reduce steric repulsion between the enzyme protein
due to C–C bond rotation compared with a linear and rigid alkynyl
group. The absolute configuration of diol 11 from ( )-7 was deter-
mined to be (R), after conversion to the saturated diol 18 by chemo-
selective diimide reduction of the double bond (Scheme 6).8 The
resulting diol 18 showed a negative sign for the specific rotation,
whose (R)-stereochemistry has been established.9 This transforma-
tion provides an alternative path for the BSEH-catalyzed direct
hydrolysis to optically active 18 with an ethyl substituent. The use
Scheme 4. Enzyme-catalyzed hydrolysis of ( )-6a and 6b. Reagents and conditions:
(a) Bacillus subtilis, 30 °C, 4 d for ( )-6a and 7 d for ( )-6b.
The relative activity of 6a toward BSEH was as low as 7% when
compared with 1. The hydrolysis was very sluggish and after pro-
longed incubation, the hydrolyzed product 10a was obtained in as
low as 5.1% yield. We realized that a spontaneous, non-enzymatic
hydrolysis competed even in the neutral buffer solution. By sub-
tracting the non-enzymatic effect, which was estimated from an
independent ‘blank’ experiment the enantioselectivity of the ‘net’
enzyme-catalyzed hydrolysis was evaluated; E = 3 at 6% conver-
sion. Attempts to apply both acetate 6b and aldehyde 16 resulted
in worse, or almost no reaction.
Reetz et al. demonstrated that the rate of the reaction catalyzed
by epoxide hydrolase depends upon the proximity between cata-

2046
K. Shimizu et al. / Tetrahedron: Asymmetry 21 (2010) 2043–2049
H
OH
(a)
(b)
O
H
OH
BnO
BnO
OH
OH
BnO
(R)-
H
7
11
71.2% ee
(R)-
BnO
OH
H
82.9% ee
H
11 93.4% ee
(R)-
H
(a)
O
OH
BnO
(R)-7
H
82.9% ee
H
O
( )-7
H
(S)-11
82.9% ee
BnO
7
39.2% ee
(R)-
Scheme 7. Ring opening of (R)-7. Reagents and conditions: (a) H2SO4, H2O, 0 °C,
70 min (39%); (b) CsOHꢂH2O, H2O, reflux, 12 h (78%).
H
OH
3. Conclusion
H
BnO
OH
Herein, we have disclosed a new scope and limitation in the B.
subtilis-catalyzed enantioselective hydrolysis. For the scope, prob-
ing the enzyme catalytic site of B. subtilis by changing the structure
of substituents, an appropriate substrate bearing an advanced
functional group was elaborated. (R)-1-Benzyloxymethyl-1-viny-
12
(R)-
92.3% ee
(a)
O
BnO
( )-8
H
loxirane
7
and (R)-2-benzyloxymethyl-3-butene-1,2-diol 11
O
BnO
became available in high ees; both of them are promising equiva-
lents for chirally differentiated tris(hydroxymethyl)methanol by
proper synthetic transformations, and the efforts to apply them
as the starting materials for the synthesis of bioactive natural
products will be reported in due course.
8
(R)-
13.2% ee
Scheme 5. Enzyme-catalyzed hydrolysis of ( )-7 and ( )-8. Reagents and condi-
tions: (a) BSEH, 30 °C; 4 d for ( )-7, 29.6% conv., E = 43.8; 7 d for ( )-8, 12.5% conv.,
E = 28.4.
4. Experimental
4.1. Materials and methods
Et OH
OH
(a)
BnO
OH
BnO
OH
Merck Silica Gel 60 F254 thin-layer plates (1.05744, 0.5 mm
thickness) and Silica Gel 60 (spherical and neutral; 100–210 lm,
(R)-18
11
92.0% ee
37,558–84) from Kanto Chemical Co. were used for preparative
thin-layer chromatography and column chromatography, respec-
tively. Preparation of wet cells of B. subtilis Tamy 2 was according
to the reported procedure.1
(R)-
Scheme 6. Determination of the absolute configuration of (R)-11. Reagents and
conditions: (a) hydrazine monohydrate, H2O2, EtOH, 0 °C, 3 h (quant.), ½a D25
¼ ꢀ8:5.
ꢁ
4.2. Analytical methods
IR spectra were measured as ATR on a Jeol FT-IR SPX60 spec-
trometer. The 1H NMR and 13C NMR spectra were measured in
CDCl3 or acetone-d6 at 400 and 100 MHz on a VARIAN 400-MR
spectrometer or 600 and 150 MHz on a Jeol JNM-ECP 600 spec-
trometer. High resolution mass spectra were recorded on a Jeol
JMS-700 MStation spectrometer. HPLC data were recorded on Jasco
MD-2010 multi-channel detectors and SHIMADZU SPD-20 Å diode
array detector. Optical rotation values were recorded on a Jasco
P-1010 polarimeter.
of ( )-7 as the substrate for enzyme-catalyzed reaction would be
advantageous, in terms of the syntheticutility of product 11 contain-
ing a double bond.
An increase in the total amount of enzyme in the hydrolysis of
( )-7 was straightforward to achieve due to the engineered nature
of the enzyme. For example, by doubling the enzyme amount the
conversion and ee(S) were improved from 29.6% to 38.0%, and from
39.2% to 56.5%, respectively, without sacrificing either ee(P) (92–
93.4% ee) or the E-value (42–44). Large-scale hydrolysis also pro-
ceeded with good reproducibility. If the conversion reaches near
60%, ee(S) would be over 95%, by a simple calculation under the
conditions of E = 44.
4.3. 1-Benzyloxymethyloxirane 3
A mixture of NaOH aq solution (40% w/w, 60 mL), benzyl alco-
hol (5.00 g, 46.2 mmol), and tetrabutylammonium bromide
(745 mg, 2.31 mmol) was vigorously stirred at room temperature
and then cooled to 0 °C. To the mixture was added epichlorohydrin
(17.1 g, 185 mmol) over 20 min. After stirring for 9 h, the reaction
mixture was poured into ice-water (30 mL). The organic materials
were extracted with Et2O twice. The combined organic phases
were washed with brine and dried over anhydrous Na2SO4, then
concentrated in vacuo. The residue was purified by distillation un-
der reduced pressure to afford 3 (6.90 g, 91%) as a colorless oil, bp
The repetition of the BSEH-catalyzed kinetic resolution on the
recovered epoxide (R)-7 (48.5% ee) enhanced its ee (82.9%), by
removing (S)-7, which had been contaminated. Treatment of (R)-
7 (82.9% ee) with diluted H2SO4 aq gave (R)-11 with inversion of
the stereochemistry and a slight decrease in ee (71.2%) which
was suggested by our previous observation with similar substrate
1.1 On the other hand, ring opening of (R)-7 with CsOH gave (S)-11
without decreasing its ee (Scheme 7). In all cases, as diols 11 were
crystalline, simple recrystallization from cold diethyl ether could
increase the ees of both (R)- and (S)-11.

K. Shimizu et al. / Tetrahedron: Asymmetry 21 (2010) 2043–2049
2047
75 °C/11 Torr. Its NMR spectra were identical to that reported
added a solution of 14 (3.51 g, 11.7 mmol) in anhydrous DMSO
(30 mL) via cannula and stirred for 25 min. The reaction was
quenched with saturated NH4Cl aq solution and the organic mate-
rials were extracted with EtOAc three times. The combined organic
phases were washed with brine and dried over anhydrous Na2SO4,
then concentrated in vacuo. The residue was purified by silica gel
column chromatography (80 g) with hexane-EtOAc (5:1) to afford
15 (3.40 g, 93%) as a colorless oil. Its NMR spectra were identical
with that reported previously.14
previously.2
4.4. B. subtilis-catalyzed hydrolysis of ( )-3
To epoxide ( )-3 (21.9 mg, 133
lmol) was added B. subtilis,
[0.4 g (wet cell weight)/mL, 60 L] and glycerol (30
l
lL) and stirred
for 6 days at 30 °C. The reaction mixture was saturated with NaCl
and diluted with EtOAc. The organic phase was separated and con-
centrated in vacuo. HPLC [Daicel Chiralcel OD-H, 0.46 cm ꢃ 25 cm;
hexane/i-PrOH (10:1), 0.5 mL/min; detected at 213 nm], tR
(min) = 15.6 [(S)-3, 43.9%], 17.7 [(R)-3, 46.1%], 29.2 [(R)-9, 6.0%],
37.1 [(S)-9, 4.0%]. The NMR spectra of 32 and 910 were identical
to that reported previously.
4.8. ( )-1-Benzyloxymethyl-1-hydroxymethyloxirane 6a
To a solution of 15 (11.6 g, 36.9 mmol) in CH2Cl2 (50 mL) and
0.1 M phosphate buffer (pH 7.0, 50 mL) was added 2,3-dichloro-
5,6-dicyano-1,4-benzoquinone (10.9 g, 47.9 mmol) and vigorously
stirred at room temperature. After 3 h, the reaction was quenched
with Na2S2O3 aq solution and organic materials were extracted
with CHCl3 twice. The combined organic phases were washed with
brine and dried over anhydrous Na2SO4, then concentrated in va-
cuo. The residue was purified by silica gel column chromatography
(200 g) with hexane/EtOAc (10:1–1:3) to afford ( )-6a (6.52 g, 91%)
as a pale yellow oil. Its NMR spectrum was identical with that re-
ported previously.15
4.5. B. subtilis-catalyzed hydrolysis of ( )-1-(4-methylbenzyl-
oxymethyl)oxirane 4
In a similar manner as described for the hydrolysis of ( )-3, ( )-
411 (22.5 mg, 126
lmol) was treated with B. subtilis and glycerol
(30
lL) for 6 days. The reaction mixture was saturated with NaCl
and diluted with EtOAc. The organic phase was concentrated in va-
cuo and purified by preparative TLC (hexane/EtOAc = 4/3) to afford
4 and 3-(4-methylbenzyloxy)-1,2-propanediol 19. (R)-4: HPLC
[OD-H; hexane/i-PrOH (10:1), 0.5 mL/min, detected at 213 nm]:
tR (min) = 14.6 [(S)-, 44.3%], 16.0 [(R)-, 55.7%]; 11.4% ee.
4.9. ( )-[1-(Benzyloxymethyl)oxiranyl]methyl acetate 6b
½
a 2D6
ꢁ
¼ þ1:0 (c 0.30, EtOH) {lit.12
½
a 2D5
ꢁ
¼ þ8:6 (c 0.40, EtOH), for
To a solution of ( )-6a (51.5 mg, 0.27 mmol) in pyridine (1 mL)
was added acetic anhydride (50.2 lL, 0.53 mmol) and the mixture
(R)-3}. The NMR spectrum was identical with that of ( )-4. (R)-
19: HPLC [OD-H; hexane/i-PrOH (10:1), 0.5 mL/min; detected at
212 nm]: tR (min) = 25.5 [(R)-, 75.7%], 41.5 [(S)-, 24.3%]; 51.4% ee.
was stirred for 45 min at room temperature. After concentration in
vacuo, the residue was purified by silica gel column chromatogra-
phy (2 g) with hexane/EtOAc (5:1) to afford ( )-6b (57.3 mg, 87%)
as a colorless oil. 1H NMR (400 MHz, CDCl3): d = 2.03 (3H, s), 2.79
(2H, s), 2.79 (1H, d, J = 11.4 Hz), 2.80 (1H, d, J = 11.4 Hz), 4.17
(1H, d, J = 12.1 Hz), 4.33 (1H, d, J = 12.1 Hz), 4.53 (1H, d,
J = 11.9 Hz), 4.56 (1H, d, J = 11.9 Hz), 7.27–7.35 (5H, m); 13C NMR
(100 MHz, CDCl3): d = 170.5, 137.6, 128.4, 127.8, 127.7, 73.4, 69.8,
63.9, 56.4, 49.2, 20.7; IR mmax 2861, 1743, 1454, 1367, 1228,
1095, 1037, 736, 698, 601 cmꢀ1. Anal. Calcd for C13H16O4: C,
66.09; H, 6.83. Found: C, 66.21; H, 6.83.
½
a 2D6
ꢁ
¼ ꢀ1:6 (c 0.30, CHCl3) {lit.10
½
a 2D1
ꢁ
¼ ꢀ1:0 (c 1.02, CHCl3), for
(R)-9}. 1H NMR (400 MHz, CDCl3): d = 2.25 (1H, br s), 2.33 (3H, s),
2.72 (1H, br s), 3.48 (1H, ddd, J = 0.6, 6.3, 9.6 Hz), 3.51 (1H, ddd,
J = 0.6, 3.9, 9.6 Hz), 3.61 (1H, br dd, J = 4.9, 11.4 Hz), 3.67 (1H, br
d, J = 11.4 Hz), 3.86 (1H, br s), 4.49 (2H, s), 7.14 (2H, d, J = 8.0 Hz),
7.19 (2H, d, J = 8.0 Hz).
4.6. 1-Benzyloxy-3-(4-methoxybenzyloxy)propan-2-one 14
A mixture of NaOH aq solution (40% w/w, 15 mL), THF (5 mL),
p-methoxybenzyl alcohol (2.97 mL, 24.0 mmol),
3
(2.62 g,
4.10. ( )-1-Benzyloxymethyl-1-formyloxirane 16
16.0 mmol), and tetrabutylammonium bromide (257 mg,
0.80 mmol) was vigorously stirred at 50 °C for 43 h. The residue
was purified by silica gel column chromatography (60 g) with hex-
ane/EtOAc (10:1–3:1) to afford a 1:1 mixture of 13 and p-methoxy-
benzyl alcohol (5.82 g) as a yellow oil.
To a solution of the above mixture (5.82 g) in CH2Cl2 (32 mL) was
added saturated NaHCO3 aq solution(16 mL), KBr(2.29 g, 19.2 mmol),
1-methyl-2-aza-adamantan-N-oxyl (3.2 mg, 0.02 mmol), and the
mixture was vigorously stirred at 0 °C. To the mixture was added
5% NaClO aq solution (60 mL, 48.1 mmol) at 0 °C and vigorously stir-
red for 30 min. The reaction mixture was quenched with Na2S2O3 aq
solution and extracted withEtOAc three times. The combined organ-
ic phases were washed with brine and dried over anhydrous Na2SO4,
and then concentrated in vacuo. The residue was purified by silica
gel column chromatography (300 g) with hexane/EtOAc (1:0–3:1)
to afford 14 (3.51 g, 73% over two steps) as a colorless oil. Its NMR
spectra were identical with that reported previously.13
To a solution of ( )-6a (96.9 mg, 0.50 mmol) in anhydrous
DMSO (0.5 mL) and anhydrous CH2Cl2 (0.5 mL) were added trieth-
ylamine (206 lL, 1.50 mmol) and the sulfur trioxide-pyridine com-
plex (175 mg 1.10 mmol) at 0 °C and stirred for 2 h. The reaction
was quenched with saturated NaHCO3 aq solution and the organic
materials were extracted with EtOAc. The organic phase was
washed with brine and dried over anhydrous Na2SO4, and then
concentrated in vacuo. The residue was purified by silica gel col-
umn chromatography (5 g) with hexane/EtOAc (1:0–5:1) to afford
( )-16 (86.5 mg, 90%) as a colorless oil. Its NMR spectra were iden-
tical with that reported previously.14
4.11. ( )-1-Benzyloxymethyl-1-vinyloxirane 7
To
a suspension of methyltriphenylphosphonium bromide
(444 mg, 1.24 mmol) in anhydrous THF (2.1 mL) was added potas-
sium t-butoxide (92.9 mg, 0.83 mmol) at 0 °C under an argon
atmosphere. After stirring for 15 min, to the mixture was added a
solution of ( )-16 (79.6 mg, 0.41 mmol) in anhydrous THF
(1.2 mL) via cannula and stirred for 15 min. The reaction was
quenched with saturated NH4Cl aq solution and organic materials
were extracted twice with EtOAc. The combined organic phases
were washed with brine and dried over anhydrous Na2SO4, then
concentrated in vacuo. The residue was purified by silica gel col-
4.7. ( )-1-Benzyloxymethyl-1-(4-methoxybenzyloxymethyl)-
oxirane 15
To
a
solution of trimethylsulfoxonium iodide (3.08 g,
14.0 mmol) in anhydrous DMSO (30 mL) was added n-butyllithium
(4.90 mL, 12.8 mmol, 2.6 M in hexane) at room temperature under
an argon atmosphere and stirred for 30 min. To the mixture was

2048
K. Shimizu et al. / Tetrahedron: Asymmetry 21 (2010) 2043–2049
umn chromatography (10 g) with hexane/EtOAc (1:0–5:1) to afford
( )-7 (37.0 mg, 92%) as a colorless oil. 1H NMR (400 MHz, CDCl3):
d = 2.69 (1H, d, J = 5.5 Hz), 2.92 (1H, d, J = 5.5 Hz), 3.62 (1H, d,
J = 11.4 Hz), 3.74 (1H, d, J = 11.4 Hz), 4.56 (1H, d, J = 11.9 Hz), 4.58
(1H, d, J = 11.9 Hz), 5.26 (1H, dd, J = 1.2, 11.0 Hz), 5.44 (1H, dd,
J = 1.2, 17.4 Hz), 5.85 (1H, dd, J = 11.0, 17.4 Hz) 7.26–7.33 (5H,
m); 13C NMR (100 MHz, CDCl3): d = 52.9, 57.5, 71.3, 73.2, 117.4,
127.6, 127.7, 128.4, 134.9, 137.8; IR mmax 3033, 2858, 1452, 1367,
1095, 925, 736, 696 cmꢀ1, HRMS (FAB): calcd for C12H15O2:
[M+H]+: 191.1072; found: 191.1065.
phases were washed with brine and dried over anhydrous Na2SO4,
and then concentrated in vacuo. This crude 17c was employed for
the next step without further purification.
To a solution of the above crude 17c (125 mg) in MeOH
(3.2 mL) was added potassium carbonate (87.5 mg, 0.63 mmol)
and vigorously stirred for 30 min at room temperature. The reac-
tion was quenched with saturated NH4Cl aq solution and organic
materials were extracted with EtOAc twice. The combined organic
phases were washed with brine and dried over anhydrous
Na2SO4, then concentrated in vacuo. The residue was purified by
silica gel column chromatography (2 g) with hexane/EtOAc
(1:0–10:1) to afford ( )-8 (41.4 mg, 70% over two steps) as a yel-
low oil. Its NMR spectra were identical with that reported
previously.16
4.12. ( )-2-Benzyloxymethyl-3-(4-methoxybenzyloxymethyl)-4-
trimethylsilylprop-3-yn-2-ol 17a
To a solution of trimethylsilylacetylene (283 lL, 2.00 mmol) in
anhydrous THF (5 mL) was added n-butyllithium (0.96 mL,
1.50 mmol, 1.6 M in hexane) at ꢀ78 °C under an argon atmosphere
and stirred for 30 min. To the mixture was added a solution of 14
(300 mg, 1.00 mmol) in anhydrous THF (5 mL) via cannula at
ꢀ78 °C. After stirring for 1 h, the reaction mixture was warmed
to 0 °C and stirred for 30 min. The reaction was quenched with sat-
urated NH4Cl aq solution and the organic materials were extracted
with EtOAc. The organic phase was washed with brine and dried
over anhydrous Na2SO4, then concentrated in vacuo. The residue
was purified by silica gel column chromatography (20 g) with hex-
ane/EtOAc (1:0–2:1) to afford 17a (386 mg, 97%) as a yellow oil. 1H
NMR (400 MHz, CDCl3): d = 0.16 (9H, s), 1.51 (1H, s), 3.56 (1H, d,
J = 9.6 Hz), 3.58 (1H, d, J = 9.6 Hz), 3.59 (1H, d, J = 9.6 Hz), 3.61
(1H, d, J = 9.6 Hz), 4.55 (2H, s), 4.62 (2H, s), 6.84 (2H, d,
J = 8.6 Hz), 7.23 (2H, d, J = 8.6 Hz), 7.26–7.32 (5H, m); 13C NMR
(100 MHz, CDCl3): d = ꢀ0.17, 55.2, 70.2, 73.2, 73.4, 73.5, 89.9,
105.2, 113.7, 127.5, 127.6, 128.3, 129.3, 130.0, 137.9, 159.2; IR mmax
3450, 2956, 2904, 2861, 1737, 1612, 1585, 1511, 1454, 1247, 1172,
1091, 1031, 917, 840, 757, 736, 698 cmꢀ1; HRMS (FAB): calcd for
4.15. B. subtilis-catalyzed hydrolysis of ( )-7
In a similar manner as described for the hydrolysis of ( )-3, ( )-
7 (34.2 mg, 0.18 mmol) were treated with B. subtilis, [0.5 g (wet cell
weight)/mL, 139 lL] and glycerol (43 lL) for 4 days, to give (R)-7
(18.8 mg, 55%) as a colorless oil along with (R)-11 (10.9 mg, 29%)
as a colorless solid.
(R)-7: HPLC analysis [OD-H; hexane/i-PrOH (90:1), 0.5 mL/min,
detected at 207 nm], tR (min) = 22.8 [(S)-, 30.4%], 25.6 [(R)-, 69.6%];
39.2% ee. Its NMR spectrum was identical with that of ( )-7.
(R)-11: HPLC [AD-H; hexane/i-PrOH (15:1), 0.5 mL/min, de-
tected at 210 nm], tR (min) = 30.9 [(R)-, 96.7%], 34.3 [(S)-, 3.3%];
93.4% ee. 1H NMR (400 MHz, CDCl3): d = 2.32 (1H, br s), 3.47 (1H,
d, J = 9.4 Hz), 3.50 (1H, d, J = 11.2 Hz), 3.58 (1H, d, J = 9.4 Hz), 3.66
(1H, d, J = 11.2 Hz), 4.53 (1H, d, J = 12.1 Hz), 4.57 (1H, d,
J = 12.1 Hz), 5.24 (1H, dd, J = 1.4, 10.9 Hz), 5.41 (1H, dd, J = 1.4,
17.4 Hz), 5.82 (1H, dd, J = 10.9, 17.4 Hz) 7.24–7.36 (5H, m); 13C
NMR (100 MHz, CDCl3): d = 67.2, 73.8, 74.6, 74.7, 116.1, 127.7,
127.9, 128.5, 137.6, 138.0; IR mmax 3419, 2925, 2861, 1454, 1089,
1076, 1043, 1027, 995, 925, 732, 696 cmꢀ1. Anal. Calcd for
C
23H30O4SiNa: [M+Na]+: 421.1811; found: 421.1796.
C
12H16O3: C, 69.21; H, 7.74. Found: C, 68.92; H, 7.80.
For the large-scale hydrolysis of ( )-7, B. subtilis [1.0 g (wet cell
4.13. ( )-2-Benzyloxymethyl-4-trimethylsilyl-but-3-yne-1,2-
diol 17b
weight)/mL, 19.8 mL] and glycerol (6.1 mL) were added to epoxide
( )-7 (4.87 g, 25.6 mmol). The broth was centrifuged (3000 rpm)
and the supernatant was saturated with NaCl and diluted with
EtOAc. The mixture was stirred 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 EtOAc. To the cell debris
precipitated by the above centrifuge was added acetone (50 mL).
The mixture was sonicated for 15 min and filtered. Both of the
EtOAc and the acetone extracts were washed with brine, dried over
anhydrous Na2SO4, and then concentrated in vacuo. The residue
was purified by silica gel column chromatography (100 g) with
hexane/EtOAc (10:1–1:3) to afford (R)-7 (3.00 g, 61.6%, 48.5% ee)
as a colorless oil along with (R)-11 (1.97 g, 36.7%, 93.3% ee) as a col-
orless solid.
To a solution of 17a (6.4 mg, 16
(0.1 mL) was added ceric ammonium nitrate (20.3 mg, 36.9
l
mol) in CH3CN:H2O (9:1)
mol)
l
and the reaction was vigorously stirred at room temperature. After
3 h, the reaction mixture was diluted with EtOAc (7 mL) and
quenched with saturated NaHCO3 aq solution. The organic materi-
als were extracted with EtOAc. The organic phase was washed with
brine and dried over anhydrous Na2SO4, then concentrated in va-
cuo. The residue was purified by silica gel column chromatography
(1 g) with hexane/EtOAc (1:0–3:1) to afford 17b (4.2 mg, 94%) as a
colorless solid, mp 88.0–88.5 °C: 1H NMR (400 MHz, CDCl3):
d = 0.15 (9H, s), 3.61 (1H, d, J = 9.6 Hz), 3.67 (1H, d, J = 9.6 Hz),
3.69 (2H, s), 4.61 (1H, d, J = 12.3 Hz), 4.63 (1H, d, J = 12.3 Hz),
7.24–7.36 (5H, m); 13C NMR (100 MHz, CDCl3): d = ꢀ0.17, 67.2,
70.7, 73.7, 74.2, 90.9, 103.9, 127.7, 127.9, 128.5, 137.6; IR mmax
3398, 3222, 2919, 2861, 1403, 1247, 1045, 840 cmꢀ1; HRMS
The secondary hydrolysis of (R)-7 (2.92 g, 15.3 mmol, 48.5% ee)
was preformed with B. subtilis, [1.0 g (wet cell weight)/mL,
11.9 mL] and glycerol (3.6 mL) for 4 days, to give (R)-7 (2.21 g,
75.9%, 82.9% ee) as a colorless oil along with (R)-11 (0.72 g,
(FAB): calcd for
301.1221.
C
15H22O3SiNa: [M+Na]+: 301.1236; found:
22.5%, 78.4% ee) as a colorless solid. ½a D24
¼ þ29 (c 1.12, MeOH)
ꢁ
for (R)-7 (82.9% ee).
4.14. ( )-1-Benzyloxymethyl-1-ethynyloxirane 8
The hydrolyzed product (1.1 g, 90.8% ee) was recrystallized
from diethyl ether at ꢀ50 °C to give (R)-11 (192.2 mg, 20% recov-
To a solution of 17b (88.1 mg, 0.32 mmol) and triethylamine
ery, 97.7% ee); mp 30.0–30.5 °C, ½a D24
¼ ꢀ28 (c 0.65, CH2Cl2).
ꢁ
(88.4
lL, 0.63 mmol) in anhydrous CH2Cl2 (1.6 mL) was added
methanesulfonyl chloride (29.4
l
L, 0.38 mmol) at ꢀ20 °C under
4.16. B. subtilis-catalyzed hydrolysis of ( )-8
an argon atmosphere. After stirring for 30 min, the reaction mix-
ture was warmed to 0 °C and stirred for 30 min. The reaction was
quenched with saturated NH4Cl aq solution and the organic mate-
rials were extracted with EtOAc twice. The combined organic
In a similar manner as described for the hydrolysis of ( )-3, ( )-
8 (31.3 mg, 0.17 mmol) was treated with B. subtilis, [0.5 g (wet cell

K. Shimizu et al. / Tetrahedron: Asymmetry 21 (2010) 2043–2049
2049
weight)/mL, 128
l
L] and glycerol (40
lL) for 7 days, to give (R)-8
4.19. (S)-2-Benzyloxymethyl-3-butene-1,2-diol 11
(23.6 mg, 75%) as a colorless oil along with (R)-12 (4.0 mg, 12%)
as a colorless solid.
(R)-8: HPLC analysis [AD-H, hexane/i-PrOH (90:1), 0.5 mL/min,
detected at 207 nm], tR (min) = 22.3 [(R)-, 56.6%], 26.4 [(S)-,
43.4%]; 13.2% ee. The NMR spectrum was identical with that of
( )-8.
A mixture of (R)-7 (16.1 mg, 0.08 mmol, 82.9% ee), CsOHꢂH2O
(92.3 mg, 0.55 mmol), and H2O (0.4 mL) was refluxed. After 12 h,
the reaction was quenched with saturated NH4Cl aq solution and
the organic materials were extracted twice with Et2O. The com-
bined organic phases were washed with brine and dried over anhy-
drous Na2SO4, then concentrated in vacuo. The residue was
purified by silica gel column chromatography (1 g) with hexane/
EtOAc (2:1–1:1) to afford diol (S)-11 (13.7 mg, 78%) as a colorless
(R)-12: HPLC [AS-H, hexane-i-PrOH (7:1), 0.5 mL/min, detected
at 210 nm], tR (min) = 27.7 [(R)-, 96.1%], 30.9 [(S)-, 3.9%]; 92.3% ee.
1H NMR (600 MHz, acetone-d6): d = 2.85 (1H, s), 3.61 (1H, d,
J = 9.7 Hz), 3.63 (1H, d, J = 9.7 Hz), 3.66 (2H, d, J = 6.6 Hz), 3.88
(1H, t, J = 6.3 Hz), 4.41 (1H, s), 4.62 (1H, s), 7.24–7.41 (5H, m);
13C NMR (150 MHz, acetone-d6): d = 67.1, 71.2, 74.0, 74.0, 74.4,
85.9, 128.2, 128.3, 129.0, 139.7; IR mmax 3423, 3259, 2923, 2867,
solid; ½a 2D3
¼ þ24 (c 0.65, CH2Cl2).
ꢁ
Acknowledgments
2362, 2343, 1396, 1253, 1178, 1093, 1045, 1025, 1010, 954 cmꢀ1
.
The authors thank Drs. Masaya Ikunaka, Hitomi Yamaguchi and
Mr. Naoki Shirasaka of the Research & Development Center, Nagase
& Co., for their support on Bacillus subtilis epoxide hydrolase. This
work was partly supported by the ‘High-Tech Research Center’ Pro-
ject for Private Universities: matching fund subsidy 2006–2011
from the Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan, and acknowledged with thanks.
4.17. (R)-2-Benzyloxymethyl-butane-1,2-diol 18
To a solution of (R)-11 (30.5 mg, 0.15 mmol, 92.0% ee) in EtOH
(1.5 mL) were added hydrazine monohydrate (178 L, 3.66 mmol)
and H2O2 (239 L, 30% w/w in water, 2.34 mmol) at 0 °C. After 3 h,
l
l
the reaction was quenched with Na2S2O3 aq solution and the or-
ganic materials were extracted twice with Et2O. The combined or-
ganic phases were washed with brine and dried over anhydrous
Na2SO4, and then concentrated in vacuo. The residue was purified
by silica gel column chromatography (1 g) with hexane/EtOAc
(2:1) to afford diol (R)-18 (33.7 mg, quant.) as a colorless oil:
References
1. (a) Fujino, A.; Asano, M.; Yamaguci, H.; Shirasaka, N.; Sakoda, A.; Ikunaka, M.;
Obata, R.; Nishiyama, S.; Sugai, T. Tetrahedron Lett. 2007, 48, 979–983; (b)
Fujino, A.; Sugai, T. Adv. Synth. Catal. 2008, 350, 1712–1716.
2. Sun, F.; Xu, G.; Wu, J.; Yang, L. Tetrahedron: Asymmetry 2006, 17, 2907–2913.
3. Shibuya, M.; Tomizawa, M.; Suzuki, I.; Iwabuchi, Y. J. Am. Chem. Soc. 2006, 128,
8412–8413.
4. Corey, E. J.; Chaykovsky, M. J. Am. Chem. Soc. 1962, 84, 867–868.
5. Parikh, J. R.; Doering, W. v. E. J. Am. Chem. Soc. 1967, 89, 5505–5507.
6. Johansson, R.; Samuelsson, B. J. Chem. Soc., Perkin Trans. 1 1984, 2371–2374.
7. Reetz, M. T.; Bocola, M.; Wang, L.-W.; Sanchis, J.; Cronin, A.-L.; Arand, M.; Zou,
J.; Archelas, A.; Bottalla, A.; Naworyta, A.; Mowbray, S. L. J. Am. Chem. Soc. 2009,
131, 7334–7343.
½
a 2D5
ꢁ
¼ ꢀ8:5 (c 1.07, CH2Cl2) {lit.9
[
a]
D = ꢀ2.4 (c 1.52, CH2Cl2)};
The NMR spectrum was identical with that reported previously.9
4.18. (R)-2-Benzyloxymethyl-3-butene-1,2-diol 11
8. (a) Corey, E. J.; Mock, W. L.; Pasto, D. J. Tetrahedron Lett. 1961, 2, 347–352; (b)
Nagendrappa, G.; Devaprabhakara, D. Tetrahedron Lett. 1970, 11, 4243–4244.
9. Hale, K. J.; Manaviazar, S.; Peak, S. A. Tetrahedron Lett. 1994, 35, 425–428.
10. Tsujigami, T.; Sugai, T.; Ohta, H. Tetrahedron: Asymmetry 2001, 12, 2543–2549.
11. Sun, F.; Xu, G.; Wu, J.; Yang, L. Tetrahedron: Asymmetry 2007, 18, 2454–2461.
12. Pandey, S. K.; Pandey, M.; Kumar, P. Tetrahedron Lett. 2008, 49, 3297–3299.
13. Walker, L. F.; Bourghida, A.; Connolly, S.; Wills, M. J. Chem. Soc., Perkin Trans. 1
2002, 965–981.
14. Russo, F.; Wangsell, F.; Savmarker, J.; Jacobsson, M.; Larhed, M. Tetrahedron
2009, 65, 10047–10059.
15. Tanner, D.; Somfai, P. Tetrahedron 1986, 42, 5985–5990.
16. Lo, C.-Y.; Guo, H.; Lian, J.-J.; Shen, F.-M.; Liu, R.-L. J. Org. Chem. 2002, 67, 3930–
3932.
A mixture of (R)-7 (22.2 mg, 0.12 mmol, 82.9% ee), H2O (0.7 mL),
and conc. H2SO4 (53 lL, 0.96 mmol) was vigorously stirred at 0 °C.
After stirring for 70 min, the reaction mixture was quenched with
saturated NaHCO3 aq solution and the organic materials were ex-
tracted twice with EtOAc. The combined organic phases were
washed with brine and dried over anhydrous Na2SO4, then concen-
trated in vacuo. The residue was purified by preparative TLC with
hexane/EtOAc (1:2) to afford diol (R)-11 (9.5 mg, 39%, 71.2% ee)
as a colorless solid. Its NMR spectrum was identical with that of
(R)-11 (as shown in Section 4.15).