Preparation of (R)- and (S)-4-chloro-3-acetoxybutyronitrile using microbial resolution

Article abstract of DOI:10.1016/S0957-4166(01)00052-0

A new preparation of optically active 4-chloro-3-acetoxybutyronitrile (AcBN) was developed using the resting cells of bacteria. The resolution was based on enantioselective hydrolysis of the ester function of the substrate. (R)-AcBN was prepared using Pseudomonas sp. DS-K-717, and the resulting (R)-AcBN was obtained with high enantiomeric excess of >98% with a yield of 36% during the microbial resolution step. (S)-AcBN was prepared in the same manner using the resting cells of Pseudomonas sp. DS-K-19 and showed a high enantiomeric excess of >98% with a yield of 32%. The enzyme activity was enhanced and induced by the addition of AcBN, particularly the (R)-ester hydrolysis, which was enhanced 20-fold.

Full text of DOI:10.1016/S0957-4166(01)00052-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 369–373
Preparation of (R)- and (S)-4-chloro-3-acetoxybutyronitrile using
microbial resolution
a
b,
a
a
a,
Hideaki Idogaki, Naoya Kasai, * Motoko Takeuchi, Miki Hatada and Toshio Suzuki *
a
Research Laboratories of Daiso Co. Ltd., 9 Otakasu-cho, Amagasaki, Hyogo 660-0842, Japan
b
Faculty of Agriculture, University of Osaka Prefecture, 1-1 Gakuen-cho, Sakai, Osaka 599-8531, Japan
Received 28 August 2000; accepted 22 January 2001
Abstract—A new preparation of optically active 4-chloro-3-acetoxybutyronitrile (AcBN) was developed using the resting cells of
bacteria. The resolution was based on enantioselective hydrolysis of the ester function of the substrate. (R)-AcBN was prepared
using Pseudomonas sp. DS-K-717, and the resulting (R)-AcBN was obtained with high enantiomeric excess of >98% with a yield
of 36% during the microbial resolution step. (S)-AcBN was prepared in the same manner using the resting cells of Pseudomonas
sp. DS-K-19 and showed a high enantiomeric excess of >98% with a yield of 32%. The enzyme activity was enhanced and induced
by the addition of AcBN, particularly the (R)-ester hydrolysis, which was enhanced 20-fold. © 2001 Elsevier Science Ltd. All
rights reserved.
9
1
. Introduction
Nakamura et al. Also, we reported the production of
(
S)-BN by the enantioselective dehalogenation using a
10
Optically active cyanohydrins such as 4-chloro-3-
hydroxybutyronitrile (BN) and 4-chloro-3-alkanoyl-
oxybutyronitrile (acyl BN) are very important
synthetic intermediates for homochiral pharmaceuticals
and agrochemicals as a C4 chiral synthon. They are
especially suitable for the synthesis of carnitine, 4-
amino-3-hydroxybutyric acid and 4-chloro-3-hydroxy-
butanonate. In particular, (R)-BN is used for the
synthesis of (R)-carnitine (vitamin BT), which plays an
important role in the human metabolism and the trans-
port of long chain fatty acids on the mitochondrial
membrane. On the other hand, (S)-BN is used as a key
structural feature for the synthesis of the inhibitors of
bacterium. However, it was impossible to obtain both
(R)- and (S)-enantiomers using these biochemical
methods.
More recently, we have developed a new method for
the microbial resolution of racemic 4-chloro-3-acetoxy-
butyronitrile (AcBN) by enantioselective hydrolysis of
the ester function. It is possible to obtain both enan-
tiomers of the (R)- and (S)-AcBN by using two types
of microorganisms that preferentially degrade (S)-
AcBN and (R)-AcBN. In this paper, we discuss the
isolation and characterisation of these microorganisms
for the enantioselective hydrolysis of AcBN, and their
utility in the production of (R)- and (S)-AcBN.
1
2
3
HMG-CoA reductase, which is the rate limiting
enzyme in cholesterol biosynthesis. Also, optically
4
active 4-amino-3-hydroxybutyric acid (GABOB), 4-
chloro-3-hydroxybutanonate and 4-hydroxypyrrolidone
are easily prepared from optically active BN.
5
2. Results and discussion
In recent years, several preparation methods of opti-
cally active BN have been reported. In the chemical
preparation method, the synthesis from the optically
2
.1. Isolation of bacteria
Three strains were isolated from the soil samples as
optically active AcBN production bacteria.
6,7
active epichlorohydrin as the starting material has
8
been developed. In the case of the biochemical method,
the preparation of (R)-BN from racemic epichlorohyd-
rin and KCN using the bacterial lyase was reported by
Two strains, labelled strain DS-K-717 and strain DS-
mk3, showed enantioselective hydrolytic activity for
(
S)-AcBN and produced (R)-AcBN with a high enan-
tiomeric excess (e.e.) of >98% in yields of 43 and 28%,
respectively. One strain, labelled DS-K-19, showed
*
Corresponding authors. Fax: +81-6-6409-0754; e-mail: tsuzuki@
daiso.co.jp
0
957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00052-0

3
70
H. Idogaki et al. / Tetrahedron: Asymmetry 12 (2001) 369–373
Table 1. Substrate specificity for each acyl BN using resting cells of strains DS-K-19, DS-K-717 and DS-mk3
Strain
Substrate
Residual ratio of
acyl BN (%)
Enantiomeric excess of alkanoyloxy Amount of produced BN Enantiomeric excess of
a
BN (%)
(mg/mL)
BN (%)
DS-K-19
b
AcBN
45.0
Not degraded
25.8
99.2 (S)
–
99.1 (S)
6.6
–
9.0
81.3 (R)
–
34.4 (R)
PrBNc
d
BuBN
DS-mk3
AcBN
PrBN
BuBN
27.3
19.4
36.2
99.0 (R)
98.6 (R)
99.3 (R)
8.6
9.7
7.6
37.1 (S)
27.8 (S)
56.3 (S)
DS-K-717
AcBN
PrBN
BuBN
42.8
Not degraded
Not degraded
99.4 (R)
–
–
6.8
–
–
71.5 (S)
–
–
a
At initial time, residual ratio is indicated to be 100%.
AcBN, 4-chloro-3-acetoxybutyronitrile.
PrBN, 4-chloro-3-propionyloxybutyrontrile.
BuBN, 4-chloro-3-butyryloxybutyronitrile.
b
c
d
enantioselective hydrolytic activity for (R)-AcBN and
produced (S)-AcBN with an e.e. of 98% in 45% yield.
The principal bacteriological characteristics of these
three strains were as follows: gram negative, rods, no
spores, motile, 30°C in optimal growth and oxidative in
glucose OF-test. Based on Bergey’s Manual of Bacteri-
ology, these three strains were identified as belonging to
the genus Pseudomonas based on their morphological,
cultural and physiological characteristics.
AcBN. The hydrolytic activity of strain DS-K-19 was
increased 20-fold compared to the additive free
medium. The highest specific activities (mU/mg of wet
cells) were induced by addition of 0.3% (v/v) AcBN;
however, cell growth was inhibited by the increasing
AcBN concentration. The amount of cells was insuffi-
cient for the resolution reaction when the concentration
of AcBN exceeded 0.5%. The addition of 0.2% AcBN
was the most effective practical method for the produc-
tion of (S)-AcBN (Fig. 1). Also, strain DS-K-717
showed the most inducible effect at a concentration of
2.2. Substrate specificity for different types of acyl BN
0
.2% (v/v) AcBN.
The substrate specificity for the different types of acyl
BNs such as AcBN, 4-chloro-3-propionyloxybutyroni-
trile (PrBN) and 4-chloro-3-butyryloxybutyronitrile
2
.4. The production of (S)- and (R)-AcBNs using the
resting cells of strains DS-K-19 and DS-K-717
(
BuBN) by three different strains was investigated in
The preparation of (R)- and (S)-AcBNs was completed
order to check the effect of the ester chain length on the
enantioselectivities (Table 1). Strain DS-K-717 showed
hydrolysing activity only for AcBN while the (R)-form
remained. Strain DS-mk3 also showed hydrolytic activ-
ities for the substrates, and each (R)-ester remained,
but the enantioselectivities for the acyl BNs were not
sufficient for the production of (R)-acyl BN. Strain
DS-K-19 showed preferential hydrolysing activities for
the (R)-form of AcBN and BuBN, and as a result, each
on a 500 mL scale. The resting cells of strain DS-K-19
Table 2. Effect of additives on ester-hydrolysing activities
Additivesa
Relative activity (%)
DS-K-717
DS-K-19
None
100b
45
18
17
0
402
22
13
30
77
100c
0
16
43
0
2060
0
46
0
(
S)-form of the acyl BN remained. In particular, it
Olive oil (1.0% v/v)
Soybean oil (1.0% v/v)
Tributyrin (1.0% v/v)
Triacetyn (1.0% v/v)
AcBN (0.2% v/v)
Lecithin (1.0% v/v)
Triolein (1.0% v/v)
Sodium acetate (1.0% w/v)
Sodium citrate (1.0% w/v)
showed a higher enantioselectivity for AcBN. In the
reaction using resting cells of strain DS-K-717 and
DS-K-19, their enantioselectivities were not improved
by the increased ester chain length. Finally AcBN is
found to be a suitable substrate for the production of
optically active acyl BN using the resting cells of strains
DS-K-717 and DS-K-19.
75
a
Added to a nutrient medium consisting of 1% (w/v) each of
polypeptone, yeast extract and glycerol.
The relative activity corresponding to 10.8 mU/mg (wet cells) was
taken to be 100%.
The relative activity corresponding to 10.6 mU/mg (wet cells) was
taken to be 100%. One unit of enzyme activity was defined as the
quantity of enzyme which degraded 1 mmol of AcBN per minute.
Specific activity was expressed in units per milligram wet cells.
2.3. Effect of additives on ester-hydrolysing activities
b
c
The effects of various additives as inducers for
hydrolytic activity were examined. Table 2 shows the
activities of the resting cells of strains DS-K-717 and
DS-K-19 in the presence of additives. As a result, the
activities were strongly induced by the addition of

H. Idogaki et al. / Tetrahedron: Asymmetry 12 (2001) 369–373
371
in 32% yield, and the formed (R)-BN had an e.e. of 45.9%
with a conversion ratio of 66.3%. The purified (S)-AcBN
(
13.5 g, 98.2% e.e.) was isolated by solvent extraction and
distillation in vacuo (88.5°C/3 mmHg). The specific
rotation of the obtained (S)-AcBN was determined as
+
11.38 (c 1.181, CHCl ) (see Fig. 2a).
3
The production of (R)-AcBN was done using the resting
cells (12.7 g, wet weight) of strain DS-K-717. Racemic
AcBN (3 g) was added to the reactor; after stirring for
eight hours, the reaction was stopped. The residual
(
R)-AcBN had an e.e. of 99.5% in 36% yield and the
formed (S)-BN had an e.e. of 50.2% with a conversion
ratio of 64.3%. The purified (R)-AcBN (0.87 g) was
isolated with an e.e. of 99.2%. The specific rotation of
the obtained (R)-AcBN was determined as −11.43 (c
1
.187, CHCl ) (see Fig. 2b).
3
Figure 1. Influence of additional AcBN on the growth of cells
and the effect of induction for ester-hydrolysing activity in
strain DS-K-19. Total activity was evaluated in units per
volume of culture broth. Specific activity was expressed in
units per milligram of wet cells. Symbols: ꢀ, total activity;
Reduced yields of the enantiomerically pure (R)- and
(S)-AcBNs were observed in the more practical jar-scale
reaction when compared with the flask-scale reactions.
This is a result of using aqueous NaOH solution (25%
w/w) to neutralise the reaction in the jar scale, which
leads to some additional unwanted AcBN hydrolysis
forming acetic acid and BN. The yield of AcBN reduces
and the e.e. of the forming BN is lowered. The higher
yield of AcBN in the flask-scale reactions results from the
ꢁ
, specific activity; ꢂ, growth.
(
(
7.7 g, wet weight) were used for the production of
S)-AcBN. Racemic AcBN (50 g) was added to the
reactor, and the reaction was stopped after 24 h. The
use of the weaker CaCO as a base to neutralise the
3
residual (S)-AcBN had an e.e. of 98.5% and was isolated
flask-scale reactions, which does not affect the AcBN.
Figure 2. Estimation of enantiomeric excess of 4-chloro-3-acetoxybutyronitrile by gas chromatography. (a) (S)-4-Chloro-3-ace-
toxybutyronitrile (98.2% e.e.); (b) (R)-4-chloro-3-acetoxybutyronitrile (99.2% e.e.); (c) (R)- and (S)-4-chloro-3-acetoxybutyroni-
trile.

3
72
H. Idogaki et al. / Tetrahedron: Asymmetry 12 (2001) 369–373
3. Conclusion
water (1 L). Strains forming clear halos were selected,
positive stains had ester-hydrolysing activity). Sec-
(
In conclusion, we have developed a novel procedure for
the production of (R)- and (S)-AcBNs using microbial
resolution. Additionally, the remaining enantiomer of
AcBN had an e.e. of 98%. This method is economical
and practical because racemic AcBN is prepared from
epichlorohydrin, which is cheap and readily available
from the petrochemical industry. Generally, the enan-
tioselectivity of lipase and esterase for secondary alco-
hols and their esters depends on the relative sizes of the
ondly, the selected strains were cultivated on an agar
plate composed of 0.025% (w/v) of yeast extract,
0.025% (w/v) of polypeptone and 0.1% (v/v) racemic
AcBN. The degradation for AcBN was checked by
observing the pH change in these media containing
additional bromothymol blue as the pH indicator. The
microorganisms which lowered the pH of the medium
were isolated. Finally, their enantioselectivity was
checked for racemic AcBN using each resting cell. The
isolated strains were cultivated with 100 mL of a nutri-
ent medium containing 1% (w/v) each of polypeptone,
yeast extract and glycerol (pH 7.2) in a 500 mL Erlen-
meyer flask. After cultivation, 1 mL of racemic AcBN
was added to the culture broth and incubated at 30°C
with vigorous shaking. Calcium carbonate was added
to the culture broth as a neutraliser. After the reaction,
the amount of the remaining AcBN and the formed
BN, and their enantiomeric purities, were estimated by
gas chromatography. The microorganisms enantioselec-
tively hydrolysing racemic AcBN were isolated. Charac-
terisation of the isolated strain was principally done by
1
1
substituents attached to the stereocentre. Based on
this, the resolution of acyl BN with high enantioselec-
tivity would be difficult because of the slight difference
in size between the -CH Cl and -CH CN substituents.
2
2
Actually, it was not easy to isolate the microorganism
which can produce both enantiomerically pure AcBN
and BN in a one-pot reaction. We have isolated two
microbial strains which showed opposing enantioselec-
tivity for the hydrolysis of AcBN. Therefore, by using
resting cells with opposite enantioselectivity for AcBN,
the preparation of both enantiomers was established.
The enantiomerically impure BN formed by microbial
resolution can be recovered and recycled by converting
to AcBN and using it as the substrate again, allowing
both enantiomers to be more efficiently obtained. More
detailed research into their industrial production is now
in progress.
12
the standard method. Identification of the strain was
performed according to Bergey’s Manual of Systematic
Bacteriology, 9th edition.
4
.3. Substrate specificity for different types of acyl BN
using three kinds of bacteria
The resting cells were prepared in a 500 mL Erlenmeyer
flask with the nutrient medium (100 mL) at 30°C for 20
h with shaking using a rotary shaker (130 rpm). After
the cultivation, the wet cells were harvested by centrifu-
gation (26,000×g, 5°C), and washed with 20 mM
sodium phosphate buffer (pH 6.8, 2×100 mL) then used
as the resting cells. The resolution reaction was carried
out in a 500 mL Erlenmeyer flask with 0.5 M sodium
phosphate buffer (pH 6.0 or 7.0, 100 mL) containing
4
. Experimental
4
.1. Preparation of racemic AcBN
Racemic AcBN was prepared from epichlorohydrin via
BN. Racemic BN was synthesised from epichlorohydrin
which was reacted with potassium cyanide. To a solu-
tion of BN (400 g, 3.35 mol) in 1,2-dichloroethane (400
ml), acetyl chloride (309 g, 3.94 mol) was added. This
mixture was stirred at 10°C for 5 h and washed with
water. The organic layer was recovered, washed with
saturated aq. NaHCO and dried with MgSO . The
solvent was removed on a rotary evaporator to give an
oil containing the crude AcBN. The resulting AcBN
8
acyl BN (1 mL), CaCO (1 g) and bacterial cells (1 g,
3
wet weight). The flask was incubated at 30°C with
shaking (130 rpm) and the pH was adjusted to each
optimal pH for the reaction: strain DS-K-717, pH 6.0;
strains DS-K-19 and DS-mk3, pH 7.0. Estimation of
the residual acyl BN and formed BN were measured by
gas chromatography as described below. Their e.e.s
were measured by gas chromatography as described
below.
3
4
(
(
(
439.3 g, 2.72 mol) was purified by vacuum distillation
1
88.5°C/3 mmHg). H NMR (CDCl ): l=5.23–5.15
3
1H, m, 3-H); 3.73–3.71 (2H, m, 4-H); 2.87 (2H, dd,
J=5.67, 1.08, 2-H); 2.15 (3H, s, CH ). Racemic PrBN
3
and BuBN were prepared from BN in the same manner
as the racemic AcBN preparation with propionyl chlo-
ride and n-butyryl chloride, respectively. The racemic
PrBN (77.7 g, 0.44 mol, 98.5°C/3 mmHg) was prepared
from BN (90 g, 0.75 mol) and propionyl chloride (77 g,
4
.4. Effect of additives on ester-hydrolysing activities
Strains DS-K-717 and DS-K-19 were cultured with the
nutrient medium to which various additives were
added. The additives were of olive oil (1% v/v), soybean
oil (1% v/v), tributyrin (1% v/v), triacetyn (1% v/v),
lecithin (1% v/v), triolein (1% v/v), sodium acetate (1%
v/v), sodium citrate (1% v/v) or racemic AcBN (from
0
1
.83 mol). The racemic BuBN (255.8 g, 1.35 mol,
07°C/3 mmHg) was prepared from BN (200 g, 1.67
mol) and n-butyryl chloride (200 g, 1.91 mol).
0
.05 to 0.5% v/v). After cultivation for 20 h, the grown
4.2. Screening of microorganisms
cells were harvested by centrifugation (26,000×g, 5°C),
washed twice with 20 mM sodium phosphate buffer
(pH 6.8, 2×100 mL) and used as the resting cells. The
resolution reaction was completed with 0.5 M sodium
phosphate buffer (pH 6.0 or 7.0, 0.5 mL) containing 1%
In the first screening, the microorganisms were isolated
from soil samples on an agar plate of a tributyrin
medium at 30°C composed of tributyrin (10 mL),
polypeptone (5 g), yeast extract (3 g) and agar (15 g) in
(v/v) of AcBN, 1% (w/v) of CaCO and appropriate
3

H. Idogaki et al. / Tetrahedron: Asymmetry 12 (2001) 369–373
373
amounts of the resting cells. The degradation amount
of AcBN was estimated by gas chromatography as
described below.
240°C; column temperature, 180°C; carrier gas, nitro-
gen; flow rate, 50 mL/min; detector, flame ionisation
detector (FID). The amount of decreasing AcBN was
measured by gas chromatography. One unit of AcBN
hydrolysing activity was defined as the quantity of
enzyme which degraded 1 mmol of AcBN per minute.
Determination of the enantiomeric excess for acyl BN
and BN was done by gas chromatography analysis
4.5. Production of (S)- and (R)-AcBNs using the resting
cells of strains DS-K19 and DS-K-717 and their
purification
(
Model G-3000, Hitachi, Tokyo, Japan). The condi-
The resting cells of strain DS-K-19 were prepared in a
tions were as follows: sample size, 1 mL; column, CHI-
2
L jar fermenter (Model MDS-U, Marubishi Bioengi-
®
RALDEX G-TA capillary column (0.25 mm diameter,
neering Co. Ltd., Japan) with a culture medium (500
mL), containing 1% (w/v) each of polypeptone, yeast
extract, glycerol and 0.2% (v/v) AcBN (pH 7.2). The
cultivation was done under the following aerobic condi-
tions: agitation, 500 rpm; aeration, 0.2 L/min; tempera-
ture, 30°C. The seed culture (60 mL) was cultivated in
a 300 mL Erlenmeyer flask containing 60 mL of nutri-
ent medium and added into the culture medium in a 2
L jar fermenter. After 20 h, cultivation was stopped and
the wet cells (7.7 g) were harvested by centrifugation
3
0 m length, Astec Inc., NJ, USA); column tempera-
ture, 120°C; injection and detection temperature,
2
1
40°C; detector, FID; carrier gas, nitrogen; split ratio,
:100. Analysis of the enantiomeric excess was used for
the extracted fraction with ethyl acetate. The determi-
nation of the configuration was done using the pre-
pared (R)- and (S)-acyl BNs from the corresponding
(
R)- and (S)-epichlorohydrins via BN. Retention time
of the (S)-AcBN was 21.3 min and for (R)-AcBN it
was 26.3 min. Also, the retention time of (S)-BN was
(
15,000×g, 5°C), washed with 20 mM sodium phos-
3
0.4 min and of (R)-BN was 31.6 min under this
phate buffer (pH 6.8, 2×500 mL) and used as the
resting cells. The resulting cells were suspended in 20
mM sodium phosphate buffer (500 mL), and racemic
AcBN (50 g) was added for optical resolution. The
resolution reaction was done under the following condi-
tions: agitation, 200 rpm; temperature, 30°C; the pH
was controlled at 6.8 with 25% NaOH. The amount of
residual AcBN and formed BN, and their enantiomeric
purities and configurations, were measured by gas chro-
matography. After the reaction, the mixture was cen-
trifuged to remove the bacterial cells. The remaining
AcBN in the supernatant was extracted with ethyl
acetate. After removing the solvent by evaporation, the
AcBN was purified by distillation in vacuo (88.5°C/3
mmHg). The production of (R)-AcBN was carried out
using a similar procedure for the production of (S)-
AcBN. The resting cells of strain DS-K-717 were pre-
pared in the same manner as described above for the
condition. Specific rotation was measured using a high
speed automatic digital polarimeter (Jasco DIP-360)
with a thermo-controlled quartz cell. Chloroform was
used as the solvent for the isolated (S)- and (R)-
AcBNs.
References
1
. Kolb, H. C.; Sharpless, K. B. Tetrahedron: Asymmetry
993, 4, 133–141.
1
2
. Fraekel, G.; Friedman, S. In Vitamins and Hormones;
Harris, R. S.; Marrian, G. F.; Thimann, K. V., Eds.;
Academic Press: New York, 1957; Vol. 15, pp. 73–118.
. Brower, P. L.; Bulter, D. E.; Deering, C. F.; Le, T. V.;
Millar, A.; Nanninga, N.; Roth, B. D. Tetrahedron Lett.
3
1992, 33, 2279.
4. Grundy, S. M. New Engl. J. Med. 1988, 319, 24.
(
S)-AcBN production, and wet cells (12.7 g) were har-
5. Kanno, O.; Miyauchi, M.; Kawamoto, I. Heterocycles
vested by centrifugation (15,000×g, 5°C). Resolution
was carried out with 20 mM sodium phosphate buffer
2000, 53, 173.
6
7
8
9
. Kasai, N.; Tsujimura, K.; Unoura, K.; Suzuki, T. J. Ind.
Microbiol. 1992, 9, 97.
(
pH 6.0, 500 mL) containing racemic AcBN (3 g). The
recovery and purification procedures of (R)-AcBN were
carried out in the same manner as the (S)-AcBN
production.
. Kasai, N.; Tsujimura, K.; Unoura, K.; Suzuki, T. J. Ind.
Microbiol. 1992, 10, 37.
. Yoshimura, H.; Furukawa, Y.; Takehira, K. (DAISO
Co. Ltd.), Jpn. Kokai Tokkyo Koho JP05-310671.
. Nakamura, T.; Nagasawa, T.; Yu, F.; Watanabe, I.;
Yamada, H. Tetrahedron 1994, 50, 11821.
4.6. Analytical methods
The hydrolysis reaction was assayed by gas chromatog-
raphy (Model GC-14B, Shimadzu, Kyoto, Japan)
equipped with a PEG 20M-HP (5%, 60/80 mesh, GL
Science Co., Ltd., Tokyo, Japan) column (3.2 mm
diameter and 1 m length). The conditions for the gas
chromatography were as follows: sample size, 1 mL;
injection temperature, 200°C; detection temperature,
10. Suzuki, T.; Idogaki, H.; Kasai, N. Bioorg. Med. Chem.
Lett. 1996, 6, 2581–2584.
11. Kazlauskas, R. J.; Weissfloch, A. N. E.; Rappaport, A.
T.; Cuccia, L. A. J. Org. Chem. 1991, 56, 2656.
12. Komagata, K. In Classification and Identification of
Microorganisms; Hasegawa, T., Ed.; Gakkai Syuppan
Center: Tokyo, 1981; pp. 203–245.
.