An efficient chemoenzymatic method to prepare optically active O-methyl-D-serine

Article abstract of DOI:10.1016/j.tet.2014.07.073

Lacosamide is an important anti-epilepsy drug and O-methyl-D-serine is a relevant intermediate in the synthesis of lacosamide. Optically active O-methyl-D-serine was prepared by using a chemoenzymatic method from inexpensive acrylamide. Our method is a four-step reaction sequence: bromination of acrylamide; etherification of dibromopropionamide; ammonolysis of α-bromo-β-methoxy-propionamide; enzymatic racemization and selective hydrolysis. The double-enzyme catalyst system, which consists of α-amino-ε-caprolactam racemase (Locus, E01594) and D-stereospecific amino-acid amidase (Locus, AB026907), was successfully applied to produce enantiopure O-methyl-D-serine (ee >99.8%) in high yield (>98.5%). Optically active O-methyl-D-serine was obtained with a total yield of 81.3%.

Full text of DOI:10.1016/j.tet.2014.07.073

Tetrahedron 70 (2014) 6991e6994
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
An efficient chemoenzymatic method to prepare optically active
O-methyl-
D
-serine
y
*
Zhi-Yuan Wang , Peng-Mei Lv, Zhen-Hong Yuan , Wen Luo, Shu-Na Liu
Key Laboratory of Renewable Energy and Gas Hydrate, Guangzhou Institute of Energy Conversion, CAS, Guangzhou 510640, Guangdong, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Lacosamide is an important anti-epilepsy drug and O-methyl-D-serine is a relevant intermediate in
Received 27 March 2014
Received in revised form 16 July 2014
Accepted 17 July 2014
the synthesis of lacosamide. Optically active O-methyl-
enzymatic method from inexpensive acrylamide. Our method is a four-step reaction sequence: bro-
mination of acrylamide; etherification of dibromopropionamide; ammonolysis of -bromo-
methoxy-propionamide; enzymatic racemization and selective hydrolysis. The double-enzyme cata-
lyst system, which consists of -amino-ε-caprolactam racemase (Locus, E01594) and -stereospecific
amino-acid amidase (Locus, AB026907), was successfully applied to produce enantiopure O-methyl-
D-serine was prepared by using a chemo-
a
b-
Available online 22 July 2014
a
D
Keywords:
D
-
O-Methyl-
Chemoenzymatic method
-Amino-ε-caprolactam racemase
-Stereospecific amino-acid amidase
D-serine
serine (ee >99.8%) in high yield (>98.5%). Optically active O-methyl-D-serine was obtained with
a total yield of 81.3%.
a
D
Ó 2014 Published by Elsevier Ltd.
Lacosamide
1. Introduction
O-methyl-
D
-serine.6,7 However, the two methods have their de-
fects. On the one hand,
source. On the other hand, the resolution of O-methyl-D,
to give O-methyl- -serine will lead to waste of the material and
increasing of the cost. Herein, we report a simple and efficient
method to obtain O-methyl-D-serine, which will eliminate the
above defects.
D
-serine is an expensive material and chiral
Lacosamide is an important one of the second generation
drugs.1 Lacosamide [(R)-2-acetamido-N-benzyl-3-methoxy pro-
pionamide] has an empirical formula of C13H18N2O3 with a mo-
lecular weight of 250.30. It is an active substance indicated for
adjunctive treatment of partial-onset seizures and diabetic neu-
ropathic pain. Based on experimental studies, lacosamide appears
to have a dual mode of action-enhancement of sodium-channel
slow inactivation and modulation of collapsing response media-
tor protein-2(CRMP-2),2 both of which are novel mechanisms for
an antiepileptic drug(AED). In the United States, lacosamide is
marketed under the name VIMPATTM for the treatment of epi-
L-serine
D
Soda et al. reported that the
(ACL racemase, Locus, E01594) does not act on other amides except
for -amino- -valerolactam and -amino-
-thia-ε-caprolactam.8e12
a-amino-ε-caprolactam racemase
a
d
a
b
However, Asano et al. discovered the racemizing activity to amino-
acid amide in ACL racemase (Locus, E01594) from Achromobacter
obae.13 At the same time, they reported the construction of an effi-
lepsy.3 O-Methyl-
thesis of lacosamide and the only chiral carbon atom of
Lacosamide is from O-Methyl-
-serine.4e6 However, only a few
methods have been reported so far for the synthesis of O-methyl-
D
-serine is a relevant intermediate in the syn-
cient system to produce
the combination of ACL racemase (Locus, E01594) and
peptidase.13,14 If an amino-acid amide racemase was used together
with these -stereospecific hydrolases, the remaining -amino-acid
amide would be racemized and the DL-amino-acid amide would be
completely hydrolyzed to form the -amino-acid. The system acted
D/L-amino-acids from amino-acid amides via
D/L-amino-
D
L
D
D-serine. There are two main synthetic routes. In the first method,
D-serine is used for the initial reactant and it can be converted into
L
O-methyl-
etherification reaction and amino deprotection. The second
method is by means of resolution of O-methyl-D, -serine to give
D
-serine after three-steps including amino protection,
broadly with various amino-acid amides, such as aminobutyric acid
amide, alanine amide, threonine amide, norvaline amide, norleucine
amide, leucine amide, methionine amide, serine amide, and phe-
L
nylalanine amide.14
D-stereospecific amino-acid amidase (DaaA, Lo-
cus, AB026907), is composed of 363 amino-acid residues (molecular
mass 40,082 Da), and from the Ochrobactrum anthropi SV3. The de-
duced amino-acid sequence exhibits homology to alkaline d-
* Corresponding author. Tel.: þ86 2087057735; fax: þ86 2087057737; e-mail
addresses: wangzy@ms.giec.ac.cn (Z.-Y. Wang), yuanzh@ms.giec.ac.cn (Z.-H. Yuan).
y
Tel./fax: þ86 020 37029689.
http://dx.doi.org/10.1016/j.tet.2014.07.073
0040-4020/Ó 2014 Published by Elsevier Ltd.

6992
Z.-Y. Wang et al. / Tetrahedron 70 (2014) 6991e6994
peptidase from Bacillus cereus DF4-B (32% identity), dd-peptidase
from Streptomyces R61 (29% identity), and other penicillin-
recognizing proteins. The DaaA protein contains the typical SXXK,
YXN, and H(K)XG active-site motifs identified in the penicillin-
had a higher conversion rate. One mol of the reaction mixture for
the second step was evaporated to dryness under vacuum, after
which the residue was placed in a stainless steel bomb and 340 g
(20 mol) of anhydrous ammonia was added. The bomb was closed,
heated, and rotated between 60 and 70 ꢀC for 3 h. By HPLC de-
tection, the conversion was 96%. At the same condition, the rate of
ammonolysis in concentrated ammonia and methanol is 78% and
85%.
binding proteins and b
-lactamases. It had maximal activity at 45 ꢀC
and pH 9.0, and was completely inactivated in the presence of phe-
nylmethanesulfonyl fluoride or Zn2þ. DaaA had hydrolyzing activity
toward d-amino-acid amides with aromatic or hydrophobic side
chains, but did not act on the substrates for the dd-peptidase and
b-
lactamase, despite their sequence similarity to DaaA.15
2.2. The enzymatic step
Herein, we report a simple three-step reaction sequence in-
volving brominating acrylamide 1 to form -dibromopropio-
namide 2, which was then reacted with sodium methylate,
to give -bromo- -methoxy-propionamide 3, ammonolysis
of -bromo- -methoxy-propionamide
methoxypropion amide 4. The
a,b
Racemic
a-amino-b-methoxy-propionamide 4 was synthe-
sized by three simple steps (Scheme 1). For compound 4, the
double-enzyme system could act on this new substrate effi-
ciently. Compound 4 was hydrolyzed by a combination of ACL
racemase (Locus, E01594) and
dase (Locus, AB026907), to give optically active O-methyl-
ine 5 in high yield (Scheme 1). The results showed that the
combination of ACL racemase (Locus, E01594) and -stereospe-
cific amino-acid amidase (Locus, AB026907) had a high activity
a
b
a
b
3
to give
-methoxy-propiona-
a-amino-b-
a
-Amino-
b
D
-stereospecific amino-acid ami-
mide 4 was hydrolyzed by a combination of ACL racemase (Lo-
cus, E01594) and -stereospecific amino-acid amidase
D
-ser-
D
(Locus, AB026907), to give optically active O-methyl-D-serine
(Scheme 1).
D
with racemic
strate. When
a
-amino-
b-methoxy-propionamide 4 as the sub-
D
-stereospecific amino-acid amidase (Locus,
O
Br
Br
Br
reflux
AB026907) and ACL racemase (Locus, E01594) were used at the
same time, 0.3 mol/l of -amino- -methoxy-propionamide 4 was
completely converted into O-methyl- -serine 5 after 10 h and the
final yield of O-methyl- -serine 5 was more than 98.5% (ee
>99.8%) (Fig. 1). The reaction conditions were optimized by using
DL- -amino-
-methoxy-propionamide as the substrate at 43 ꢀC
Br
NH
O
NH
baths of ice
CH ONa
NH
a
b
MeOH
O
O
D
1
2
3
D
NH
a
b
(Fig. 2) and pH 8.0 (Fig. 3). As you know, low concentration of
substrate will promote enzymatic reaction and high concentra-
tion of substrate will hinder enzymatic reaction, the optimal
substrate concentration of this double-enzyme system is
H N
NH
DaaA
O
NH
O
OH
ACL racemase
O
O
4
5
300 mmol/L by using DL-a-amino-b-methoxy-propionamide as
the substrate (Fig. 4). The double-enzyme system include two
different types of enzyme, different enzyme has different re-
action speed. There must be a optimal proportion between the
two enzymes and it is 1e1.5 (DaaA to ACL racemase) (Fig. 5).
Thus, amino-acid amide racemase is a key enzyme to realize the
dynamic kinetic resolution of amino-acid amides when used with
stereospecific amide hydrolases.
Scheme 1. Chemoenzymatic preparation of optically active O-methyl-D-serine.
2. Results and discussion
2.1. Preparation of a-amino-b-methoxy-propionamide 4
The preparation of
a-amino-b-methoxy-propionamide 4 in-
volves three steps: bromination of acrylamide 1, etherification re-
action of dibromopropionamide 2, and ammonolysis of
methoxy-propionamide 3.
a-bromo-b-
In the first step, bromine was quickly added dropwise into the
methanol solution of acrylamide. During the addition of bromine,
the mixture was heated spontaneously to the boiling temperature
of approximately 65 ꢀC. The mixture was heated at reflux for 2 h
until consumption of the bromine was complete. The yield of this
step was 99%. After cooling, the mixture was used immediately in
the next step.
The etherification of dibromopropionamide was conducted in
an ice bath. Due to the strong reactivity of the
propionamide 2, sodium methoxide could easily attack the
b
-Br of dibromo-
-car-
b
bon atom. Sodium methoxide can also catalyze the debromination
and polymerization as a strong alkaline catalyst. To avoid this
debromination and polymerization, a fed-batch reaction and bath
of ice were used in this step. Under these reaction conditions, the
yield of this step was 87%.
The third step was ammonolysis. To improve the ammonolysis
rate, three different reaction conditions were studied, including
ammonolysis in concentrated ammonia, liquid ammonia, and
methanol. The result indicated that the reaction in liquid ammonia
Fig. 1. Production of O-methyl-D-serine from DL-
D-stereospecific amino-acid amidase (Locus, AB026907) in the presence of ACL race-
mase (Locus, E01594). ( ): O-methyl-D-serine, ( ): L- -amino- -methoxy-pro-
pionamide, ( ): D- -amino- -methoxy-propionamide.
a-amino-b-methoxy-propionamide by
a
b
a
b

Z.-Y. Wang et al. / Tetrahedron 70 (2014) 6991e6994
6993
Fig. 5. Effect of enzyme proportion on the enzymatic activity. Reaction conditions:
0.3 mmol/L DL-a-amino-b-methoxy-propionamide, 1.0 mmg DaaA, 1.5 mmg ACL race-
mase, 20 ml reaction volume, 43 ꢀC, pH 8.0.
Fig. 2. Effect of temperature on the enzymatic activity. Reaction conditions: 0.3 mmol/
L DL- -amino- -methoxy-propionamide, 1.0 mmg DaaA, 1.5 mmg ACL racemase, 20 ml
a
b
reaction volume, pH 8.0.
Traditional methods are based on using D-serine as material or
the resolution of the DL-precursor by chemical or enzymic
methods,3,7 which would increase the cost and lead to the waste of
material. This new method will decrease the cost and avoid the loss
of substrate. Actually, other optical amino-acids and their de-
rivatives can be obtained by resolution of corresponding amino-
acid amides with the double-enzyme system. This will avoid the
waste of substrate and decrease the cost greatly.
3. Conclusion
Optically active O-methyl-
enzymatically by starting from inexpensive acrylamide 1. The
double-enzyme catalyst system, including -amino-ε-caprolactam
racemase (ACL racemase, Locus, E01594) and -stereospecific
D-serine 5 was prepared chemo-
a
D
amino-acid amidase (Locus, AB026907), was successfully applied as
an enantiopurity producing step via kinetic resolution of the ra-
cemic amino-acid amide precursor. The product was obtained with
excellent enantiopurity (>99.8%) and in high yields (>98.5%). This
method avoids the waste of half substrate comparing with tradi-
tional resolution methods. The overall yield of the whole chemo-
enzymatic process was 81.3%.
Fig. 3. Effect of pH values on the enzymatic activity. Reaction conditions: 0.3 mmol/L
DL-a-amino-b-methoxy-propionamide, 1.0 mmg DaaA, 1.5 mmg ACL racemase, 20 ml
reaction volume, 43 ꢀC.
4. Experimental part
4.1. General
4.1.1. Reagents and instruments. All chemicals were of analytical
grade and purchased from the China Medicine (Group) Shanghai
Chemical Reagent Corporation or other recognized commercial
suppliers. Melting points were recorded on a WRR melt apparatus.
1H NMR and 13C NMR were recorded on a Bruker DRX500
(500 MHz). Mass spectra were recorded on a Mariner ESI-TOF mass
spectrometer. Rotations were recorded on a WZZ-2B polarimeter.
The enantiomeric compositions of residual substrates were de-
termined as described by Zheng.16 Enantiomeric excess (ee) is de-
fined as (A1ꢁA2)/(A1þA2)ꢂ100, where A1 and A2 are peak areas of
the enantiomers; element composition was measured by trace el-
ement autoanalyzer of EA 3000 type.
4.1.2. Enzymes. ACL racemase (Locus, E01594) was obtained from
Toyobo Co. Ltd (Osaka, Japan). The
D-stereospecific amino-acid
amidase (Locus, AB026907) gene from Ochrobactrum anthropi SV3
was cloned into Escherichia coli BL21 (DE3) using plasmid pETDuet
as the vector in our laboratory. The strain was cultured in LB broth
Fig. 4. Effect of substrate concentration on the enzymatic activity. Reaction conditions:
DL-a
-amino-
b-methoxy-propionamide as the substrate, 20 ml reaction volume,
1.0 mmg DaaA, 1.5 mmg ACL racemase, 43 ꢀC, pH 8.0.
containing 50 mg ampicillin/ml for 4 h and induced with 0.4 mM

6994
Z.-Y. Wang et al. / Tetrahedron 70 (2014) 6991e6994
IPTG at 30 ꢀC. After further incubation for 8 h, the cells were har-
vested by centrifugation at 8000ꢂg for 20 min at 4 ꢀC. Harvested
cells were suspended in 20 ml of 20 mM TriseHCl (pH 7.5) and then
ultrasonicated at 4 ꢀC. The cell-free extracts were fractionated with
ammonium sulfate (30e70%). The precipitate obtained from am-
monium sulfate fractionation was dissolved in 50 mM of TriseHCl
(pH 7.5) containing 1 mM MnSO4, dialyzed against the same buffer,
and applied to a DEAE-Sepharose CL-6B column (1.5ꢂ55 cm)
equilibrated with the same buffer. The column was washed with
the same buffer, and then the enzyme was eluted with a gradient
formed between the buffer and that containing 0.4 M NaCl. The
active fractions were concentrated by ammonium sulfate pre-
cipitation and then applied to a Cellulofine GC-700 m column
(2.1ꢂ94 cm) (Seikagaku Kogyo, Japan) equilibrated with 50 mM
TriseHCl (pH 7.5) containing 1 mM MnSO4. The enzyme was eluted
with the same buffer. The final sample was homogeneous by
SDSepolyacrylamide gel electrophoresis. Ammonium sulfate was
added to the final enzyme sample to 70% saturation, and the sample
was stored at 4 ꢀC until it was used. In this case, the buffer used
throughout the purification procedures contained 1 mM MnSO4.17
D2O): 176.9 (CO), 76.2 (CH2), 60.7 (CH), 56.6 (OCH3). Anal. Calcd for
C4H10N2O2: C 40.68, H 8.47, N 23.73. Found: C 40.70, H 8.45, N 23.77,
ESI-MS (m/z): 119.1321 [MþH]þ.
4.5. Preparation of O-methyl-D-serine 5 by the double-
enzyme system
In a typical small-scale experiment, ACL racemase (Locus,
E01594) and -stereospecific amino-acid amidase (Locus,
AB026907) were added to a solution of the -amino- -methoxy-
D
a
b
propionamide 4 in water. The reaction mixture (50 ml), containing
2 mM of pyridoxal 50-phosphate (PLP), 0.1 M of potassium phos-
phate buffer (KPB) (pH 8.0), 0.3 mol/l of
a-amino-b-methoxy-pro-
pionamide 4, ACL racemase (Locus, E01594) (3.0 mg), and
D
-
stereospecific amino-acid amidase (Locus, AB026907) (2.0 mg) was
incubated at 43 ꢀC for 10 h. The progress of the reactionwas followed
by taking samples at intervals. The amounts of -amino-
D
/
L
-
a
b-
methoxy-propionamide and O-methyl- -serine 5 were de-
4
D
termined with an HPLC equipped with a Crownpak CR (þ) column at
a flow rate of 0.5 ml/min, using the solvent system of 50 mM HClO4.
Absorbance of the eluate was monitored at 210 nm. O-Methyl-D-
4.2. Preparation of a,b-dibromopropionamide 2
serine 5 was formed in the mixture and isolated by a procedure
involving deproteinization by trichloroacetic acid and column
chromatography on a Dowex-X8 (Hþ) column. The retention time
To a slurry of 71 g (1 mol) of crystalline acrylamide in 450 ml
methanol,160 g (1 mol) of bromine was added with stirring for over
10 min. During the addition of bromine, the mixture heated
spontaneously to the boiling temperature of approximately 65 ꢀC.
The mixture was heated at reflux for 2 h until consumption of
bromine was complete. The reaction mixture was then evaporated
to dryness, 228.2 g of white needle-like crystals was obtained
(yield: 99%). Mp 71e72 ꢀC (decomp.). 1H NMR (400 MHz, D2O):
for 5 is 17 min. The isolated O-methyl- -serine 5 (12.5 mmol, 1.5 g)
D
was recrystallized from wateremethanoleisopropyl alcoholeether.
The enantiomeric purity of the isolated O-methyl- -serine 5 was
D
greater than 99.8%. White crystals were obtained. Mp 208e210 ꢀC
(decomp.), ½a 2D5
ꢃ
ꢁ 14:2ꢀ (c 1, 6 mol/l HCl). The rotation of 5 is iden-
tical to Ref. 20 (ee >98.5%, ½a D25
ꢃ
ꢁ 13:7ꢀ (c 1, 6 mol/l HCl)).20 1H NMR
(400 MHz, D2O):
d
3.21 (s, 3H), 3.80 (d, J¼3.6, 1H), 3.85 (d, J¼9.2 Hz,
d
4.01 (d, J¼4.4 Hz, 2H), 4.65 (t, 1H), 13C NMR (100 MHz, D2O): 178.3
1H), 3.88 (dd, J¼3.6, 9.2 Hz, 1H), 13C NMR (100 MHz, D2O): 170.5
(CO), 73.2 (CH2), 58.4 (CH), 56.0 (OCH3). Anal. Calcd for C4H9NO3: C,
40.32, H, 7.58, N, 11.75. Found: C, 40.35, H, 7.54, N, 11.77. ESI-MS (m/
z): 120.1230 [MþH]þ.
(CO), 57.1 (CH2), 37.5 (CH). Anal. Calcd for C3H5NBr2O: C 15.56, H
2.20, N 6.10, Br 69.30. Found: C 15.58, H 2.16, N 6.06, Br 69.26. ESI-
MS (m/z): 232.1328 [MþH]þ.
The results of all spectroscopic analyses of compound 5 were
identical to those described in Refs. 7,18e20. Absolute configuration
of 5 was confirmed by comparing their optical rotation to that re-
ported in the above-mentioned literature.
4.3. Preparation of a-bromo-b-methoxy-propionamide 3
The above mixture was cooled to about 20 ꢀC and 360 ml of
a sodium methoxide solution, formed by reacting 26.5 g (1.15 g
atomic weight) of sodium with methanol, was added over 30 min.
During, and for 3 h after, the addition of the sodium methoxide, the
mixture was stirred and cooled to maintain it between 20 and 0 ꢀC
in a bath of ice. Afterward the reaction mixture was distilled to
a quarter of the volume and NaBr (reaction by-product) was re-
moved by filtration. The filtrate was evaporated to dryness to give
156.7 g of white powder (yield: 87%). Mp 83e85 ꢀC (decomp.). 1H
Acknowledgements
This study was supported by the National Science-Technology
Support Plan Projects (2014BAD02B07) and the Innovation Fund
Project of Guangzhou Institute of Energy Conversion, CAS
(Y307P11001).
NMR (400 MHz, D2O):
d
3.25 (s, 3H), 4.16 (dd, J¼3.5, 10.1 Hz, 1H),
References and notes
4.32 (dd, J¼6.0, 10.1 Hz, 1H), 4.51e4.75 (m, 1H), 13C NMR (100 MHz,
D2O): 177.3 (CO), 74.8 (CH2), 53.1 (CH), 55.6 (OCH3). Anal. Calcd for
C4H8NBrO2: C 26.37, H 4.40, N 7.69, Br 43.96. Found: C 26.35, H 4.39,
N 7.71, Br 43.93. ESI-MS (m/z): 183.1245 [MþH]þ.
1. Perucca, E. B. J. Clin. Pharmacol. 1996, 42, 531.
2. Beydoun, A.; Souza, J.; Hebert, D. Expert Rev. Neurother. 2009, 9, 33.
3. Ernesto, D. L.; Monica, B. V. US Patent 0034731A1, 2011
4. Stables, J. P.; Liu, R.; Kohn, H. Org. Biomol. Chem. 2010, 8, 2803.
5. Shridhar, V. A.; James, P. S.; Harold, K. Tetrahedron: Asymmetry 1998, 9, 3841.
6. Muddasani, P. R.; Madalapu, V.; Nannapaneni V. C. WO Patent 2011095995 A1,
2011.
4.4. Preparation of a-amino-b-methoxy-propionamide 4
7. Hamann, J.; Kanzler, R.; Merschaert, A.; Szczepaniak, C. J. WO Patent
2012065891 A1, 2011.
Compound 3 (150 g) was placed in a stainless steel bomb and
340 g (20 mol) of anhydrous ammonia was added. The bomb was
closed and heated, while rotating between 60 and 70 ꢀC for 3 h. The
conversion was determined to be 96% by HPLC detection. The bomb
was then cooled, opened to release the pressure, and the charge
removed. Afterward the reaction products were dissolved in 200 ml
of water and the solution was extracted with 400 ml of N-butanol
repeatedly. The N-butanol phase continued to be evaporated to
dryness and 90.4 g of light yellow oil was obtained. 1H NMR
8. Ahmed, S. A.; Esaki, N.; Tanaka, H.; Soda, K. Agric. Biol. Chem. 1983, 47, 1149.
9. Ahmed, S. A.; Esaki, N.; Tanaka, H.; Soda, K. FEBS Lett. 1984, 174, 76.
10. Ahmed, S. A.; Esaki, N.; Tanaka, H.; Soda, K. Agric. Biol. Chem. 1985, 49, 2991.
11. Ahmed, S. A.; Esaki, N.; Tanaka, H.; Soda, K. Biochemistry 1986, 25, 385.
12. Ahmed, S. A.; Esaki, N.; Tanaka, H.; Soda, K. Agric. Biol. Chem. 1983, 47, 1887.
13. Asano, Y.; Yamaguchi, S. J. Am. Chem. Soc. 2005, 127, 7696.
14. Yamaguchi, S.; Komeda, H.; Asano, Y. Appl. Environ. Microbiol. 2007, 73, 5370.
15. Komeda, H.; Asano, Y. Eur. J. Biochem. 2000, 267, 2028.
16. Zheng, R. C.; Zheng, Y. G.; Shen, Y. C. Biotechnol. Lett. 2007, 29, 1087.
17. Suzuki, H.; Kamatani, S.; Kumagai, H. Biosci. Biotechnol. Biochem. 2001, 65, 1549.
18. Nishiyama, A. JP Patent 2010037206-A, 2010.
19. Hodges, R. S.; Merrifield, R. B. J. Org. Chem. 1974, 39, 1870.
20. Xiong, J. B.; Liu, J. Z.; Liu, Q.; Jiao, Q. C. Chem. J. Chin. Univ. 2010, 31, 1560.
(400 MHz, D2O):
d
3.28 (s, 3H), 3.77 (dd, J¼3.6, 10.3 Hz, 1H), 3.89
(dd, J¼5.8, 10.3 Hz, 1H), 3.92e4.13 (m, 1H), 13C NMR (100 MHz,