Artificial model for cystathionine β-synthase: Construction of a catalytic cycle with a pyridoxal model compound having an ionophore function

Article abstract of DOI:10.1016/S0040-4020(03)00665-3

Catalytic transformation of serine-O-carbonate to S-aryl cysteine derivatives was successfully achieved in the presence of Li⁺ by the use of a pyridoxal model compound having an ionophore function, which is the first example mimicking cystathionine β-synthase, artificially.

Full text of DOI:10.1016/S0040-4020(03)00665-3

Tetrahedron 59 (2003) 4867–4872
Artificial model for cystathionine b-synthase: construction of a
catalytic cycle with a pyridoxal model compound having an
ionophore function
*
Kazuyuki Miyashita, Hidenobu Murafuji, Hiroshi Iwaki, Eito Yoshioka and Takeshi Imanishi
Graduate School of Pharmaceutical Sciences, Osaka University, 1-6 Yamadaoka, Suita, Osaka 565-0871, Japan
Received 12 March 2003; accepted 24 April 2003
Abstract—Catalytic transformation of serine-O-carbonate to S-aryl cysteine derivatives was successfully achieved in the presence of Liþ by
the use of a pyridoxal model compound having an ionophore function, which is the first example mimicking cystathionine b-synthase,
artificially. q 2003 Elsevier Science Ltd. All rights reserved.
1. Introduction
In previous papers, we reported a novel type of pyridoxal
model compound having an ionophore side-chain at the C-3
hydroxyl group, which had not been modified in previous
model compounds, and its application to the synthesis of
a,a-dialkyl amino esters by the a-alkylation.7,8 In these
studies, it was found that aldimine, prepared from 3 and an
amino ester, can capture Liþ specifically as shown in
Figure 1, and, as a consequence, the a-hydrogen is
activated.7 This interesting feature of 3 was expected to
be of use for other reactions mediated by PLP (1). As the
b-replacement reaction appears to be useful for the
synthesis of various b-modified amino acids by employing
various kinds of nucleophiles, we, at first, applied our model
compound 3 to the b-replacement reaction employing thiols
as a nucleophile, and successfully achieved the reaction
Pyridoxal 50-phosphate (PLP, 1) and pyridoxamine 50-
phosphate (PMP, 2) are important coenzymes related with
the biosynthetic and metabolic reactions of a-amino acids.
A number of reactions mediated by these coenzymes are
known.1 b-Replacement reaction of the serine hydroxyl
group with a nucleophile is one such reaction mediated by
pyridoxal, and is a biologically important reaction for
biosynthesis of amino acids. Tryptophan synthase and
cystathionine b-synthase are typical examples, in which the
nucleophiles are indole and homocysteine, respectively
(Figure 2). In order to clarify the mechanisms of the
biological reactions, and from the viewpoint of synthetic
organic chemistry, a number of studies mimicking the
biological reactions mediated by a pyridoxal–pyridoxamine
system have been conducted.2 In fact, pyridoxal model
studies mimicking tryptophan synthase have also been
reported,3 but, to the best of our knowledge, that mimicking
cystathionine b-synthase has not. Cystathionine b-synthase
is a biologically important enzyme, working as a homo-
cysteine scavenger,4 deficiency of which causes homo-
cystinuria, and is known to be related with cardiovascular
diseases, neural tube defects and Altzheimer’s disease.5
Cystathionine b-synthase is known to require two co-
factors, pyridoxal and heme. Although the role of pyridoxal
is apparent, that of heme has not been clarified, but it may be
a regulatory role.6
Keywords: amino acids and derivatives; coenzymes; pyridines; substitution.
*
Corresponding author. Tel.: þ81-6-6879-8200; fax: þ81-6-6879-8204;
e-mail: imanishi@phs.osaka-u.ac.jp
Figure 1. Structures of pyridoxal, pyridoxamine and model compound 3.
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00665-3

4868
K. Miyashita et al. / Tetrahedron 59 (2003) 4867–4872
under catalytic conditions. Herein we describe details of the
results.9
2. Results and discussion
At first, the reaction was examined under stoichiometric
conditions (Scheme 1). Namely, aldimine 5a, which had
been quantitatively prepared from 3 and serine ester 4a only
by mixing in CH2Cl2 at room temperature, was treated with
thiophenol in the presence of lithium perchlorate (LiClO4)
in acetonitrile. However, no reaction took place, and only
the starting aldimine 5a was recovered. As this appears to be
Figure 2. b-Replacement of reaction of Ser in a biological system.
Scheme 1.
due to the low eliminating ability of the hydroxyl group and/
or an effect of the chelation with the hydroxyl group rather
than with the ester moiety as shown in Figure 3, serine-O-
acylates 4b (acetate and benzoate) were examined in order
to increase the eliminating ability of the hydroxyl group and
to lower the chelation ability. However, as acyl migration of
4b took place easily, it was difficult to handle. In addition,
although the starting material 5b, as expected, disappeared
quickly on the TLC, a complex mixture was formed. In
these reactions, acetic acid or benzoic acid was formed by
the elimination, which was expected to interfere with the
reaction. Eventually, we selected serine-O-carbonate 4c as a
starting material, because the expected by-products were
methanol and carbon dioxide, both of which were neutral
and expected not to affect the reaction. In this case, the
reaction of 5c smoothly proceeded to give the desired
aldimine 6 in almost quantitative yield. To obtain cysteine
derivative 7a, the aldimine 6 was subjected to hydrolysis
under conventional acidic conditions. However, despite
extensive effort, hydrolysis of the aldimine 6 resulted in
poor yield, giving unidentified products and S-phenyl-
cysteine (7a) in 23% yield when p-toluenesulfonic acid was
employed. Although the reason is not clear, the nucleophilic
character of the neighboring sulfur atom of 6 might interfere
with the hydrolysis.
Figure 3. Possible chelation structure of 5a with Li.
Table 1. b-Replacement reaction of 4c with thiophenol catalyzed by
pyridoxal 3
Entry
3
(mol%)
LiClO4
(mol%)
Solvent
Time
(h)
Yield
(%) of 7a
Based
on 4c
Based
on 3
1
2
3
4
5
6
1
5
10
5
5
5
1
5
10
5
5
5
MeCN
MeCN
MeCN
CH2Cl2
AcOEt
THF
38
5
2
18
27
27
76
93
87
80
88
30
7600
1860
870
1600
1760
600
From the results described above, the O-carbonate function
was revealed to work as a good leaving group without
affecting the b-replacement reaction. The unexpected
resistance of the aldimine 6 towards hydrolysis was one
problem which remained. It occurred to us that, if imine

K. Miyashita et al. / Tetrahedron 59 (2003) 4867–4872
Table 2. b-Replacement reaction of 4c with thiols catalyzed by pyridoxal 3
4869
of 3 was used (entry 2). Increasing the amount of 3
shortened the reaction time, but lowered the yield as well
(entry 3). The solvent effect was also examined as
summarized in entries 4–6. Although the reaction pro-
ceeded in the solvents shown, acetonitrile was found to be
the best choice. The low yield in THF could be attributable
to solvation of Liþ with THF (entry 6).
Entry
R
Time (h) Product 7
Yield (%) of 7
Under the conditions of entry 2 in Table 1, aromatic thiols
reacted with 4c to afford S-aryl cysteines 7a–d in good
yields (entries 1–4 in Table 2). As the reactions of various
4-substituted thiophenols took place, it is noteworthy that
the thiophenol with an electron-withdrawing group is a good
substrate for the reaction (entry 4 vs entries 1–3). In
contrast, the reactions with alkane thiols proceeded, but
required longer reaction time, and the yields were lower
than those for aromatic thiols (entries 5–8).
Based on 4c Based on 3
1
2
3
4
5
6
7
8
Ph
5
5
5
a
b
c
d
e
f
g
hb
93
86
90
92
61
41
30
28
1860
1720
1800
1840
1220
820
4-MeC6H4
4-MeOC6H4
4-O2NC6H4
Bn
Et
HO(CH2)3
3
19
22
72
25
600
560
a
HS(CH2)3
a
Thiol (0.6 equiv.) was employed.
Both thiol groups were alkylated.
b
In our previous report dealing with a-alkylation of a-amino
esters employing the same model compound 3, a combi-
nation of 3 with Liþ was found to be essentially important
for the reaction to proceed, and the chelation structure
exchange reaction between the aldimine 6 and the starting
serine derivative 4c took place, we could desirably make
this reaction procedure catalytic. Based on this hypothesis,
we examined the reaction employing a catalytic amount of 3
and LiClO4, and the results are summarized in Table 1. The
reaction smoothly took place at room temperature only by
mixing serine-O-carbonate 4c and thiophenol, in the
presence of a catalytic amount of model compound 3 and
LiClO4 in acetonitrile, to give S-phenylcysteine 7a.10 It was
found that the reaction proceeds in the presence of only
1 mol% of 3 to give 7a in 76% yield based on 4c (entry 1). In
this case, the yield based on 3 was 7600%, which
simultaneously means that one molecule of 3 converts 76
molecules of serine derivative 4c to the corresponding
cysteine derivative 7a. However, from a practical view-
point, the best yield based on 4c was obtained when 5 mol%
1
including Liþ was also revealed by H NMR.7 From the
results, it was easily assumed that Liþ plays a crucial role in
the present reaction as well. In order to confirm the role of
Liþ, we studied the effect of metal ions on the elimination
1
procedure of the carbonate moiety using H NMR. In the
presence of Liþ, the aldimine 5c smoothly disappeared,
while two olefinic hydrogens appeared, indicating the
formation of unsaturated species 8 as expected. As is
obvious from Figure 4, this reaction specifically proceeded
only in the presence of Liþ, and did not proceed smoothly in
the presence of other metal ions or in the absence of Liþ.
These results clearly suggest that specific chelation of Liþ
between the iminoester moiety and the ethoxyethoxy group
plays an important role similar to our previous results.7
Figure 4. Elimination of aldimine 5c in the absence and presence of metal ions.

4870
K. Miyashita et al. / Tetrahedron 59 (2003) 4867–4872
Figure 5. Possible mechanism for b-replacement reaction of 4c catalyzed by a combination of Liþ and 3.
From the results described above, a possible reaction
mechanism is proposed in Figure 5. The first step should
3. Experimental
be that pyridoxal derivative 3 reacts with 4c to give the
aldimine 5c, which would capture Liþ specifically to give
complex 9. This phenomenon would induce restriction of
the conformation of the iminoester moiety in the same plane
as that of the pyridine ring, and consequently would activate
the a-hydrogen. As a result, elimination of the methyl
carbonate moiety is thought to take place as the next step,
which would produce carbon dioxide, methanol and
unsaturated species 10. This unsaturated compound 10
appears to be highly electrophilic, which would induce 1,4-
addition of thiol to give cysteine aldimine 11. Subsequent
aldimine exchange reaction of 11 with 4c via 12 would give
7a and 9, forming a catalytic cycle for the b-replacement
reaction. In general, alkane thiols are more nucleophilic
than aromatic thiols because of the electron-donating
property of an alkyl group, from which, the lower reactivity
of alkane thiols observed in the present experiments
appears to be strange. As aromatic thiols are more acidic
than alkane thiols, formation of more nucleophilic thiolate
from thiol and/or a protonation step to an enolate species
through the 1,4-addition procedure of thiol could be
important to complete the catalytic cycle. This is also
supported by the fact that the reaction with the most acidic
thiol, 4-nitrothiophenol, proceeded most smoothly (Table 2,
entry 4).
3.1. General
Infrared spectra were measured on a JASCO FT/IR-200
Fourier-transfer infrared spectrometer. 1H NMR spectra
were measured on a JEOL EX-270 (270 MHz) spectrometer
and tetramethylsilane (TMS) was used as an internal
standard. 13C NMR spectra were measured on the same
instrument (67.8 MHz) with CDCl3 as an internal standard
(77.0 ppm). Low and high resolution mass spectra (EI-MS
and HR-MS) were obtained by use of a JEOL D-300 mass
spectrometer. For silica gel column chromatography,
E. Merck Kieselgel 60 (0.063–0.200 mm) was used.
3.1.1. Stoichiometric b-replacement reaction of serine-
O-carbonate 4c. Aldimine 5c was prepared from serine
derivative 4c and pyridoxal 3 by the same method described
previously,7 and the resultant aldimine 5c was immediately
employed for the next reaction without purification.
b-Replacement reaction of 5c with thiophenol was carried
out as follows. To a stirred acetonitrile solution (5 mL) of
5c, which had been prepared from serine-O-carbonate 4c
(110 mg, 0.43 mmol) and pyridoxal 3 (142 mg, 0.43 mmol),
were added thiophenol (49 mL, 0.47 mmol) and LiClO4
(2.3 mg, 0.022 mmol). After being stirred at room tempera-
ture for 2 h, the reaction mixture was diluted with ethyl
acetate, washed with H2O and saturated NaCl solution, and
dried over Na2SO4. Concentration under reduced pressure
afforded crude product of 6, which was hydrolyzed without
purification according to the same procedure as described
previously,7 and purified by silica gel column chromato-
graphy (ethyl acetate/hexane¼1:1) to give S-phenylcysteine
In conclusion, b-replacement reaction of serine derivative
4c with aromatic thiols was successfully achieved under
catalytic conditions, which is the first artificial model
mimicking cystathionine b-synthase, and would serve as a
useful method for the preparation of S-aryl substituted
cysteines. Although the present model compound 3 cannot
catalyze the b-replacement reaction with alkane thiols
sufficiently, further effort to improve the model compound 3
to achieve the reaction with alkane thiols will be described
in the following article.
1
7a (28.4 mg, 23%). H NMR data for aldimines 5c and 6
taken in CDCl3 are as follows. 5c, d: 1.21 (3H, t, J¼7 Hz,
OCH2CH3), 2.56 (3H, s, C2-Me), 3.53 (2H, q, J¼7 Hz,
OCH2CH3), 3.62–3.65 and 3.93–3.97 (each 2H, m,
OCH2CH2O), 3.74 (3H, s, OMe), 4.35 (1H, dd, J¼4.5,

K. Miyashita et al. / Tetrahedron 59 (2003) 4867–4872
4871
8 Hz, a-H), 4.49 (1H, dd, J¼8, 11 Hz), 4.51 (2H, s,
OCH2Ph), 4.73 (1H, dd, J¼4.5, 8 Hz), 4.81 (2H, s, C5-
CH2), 5.18 (2H, AB type, J¼12 Hz, CO2CH2Ph), 7.27–7.37
(10H, m, aromatic H), 8.56 (1H, s, C6-H), 8.77 (1H, s, imine
H); 6, d: 1.15 (3H, t, J¼7 Hz, OCH2CH3), 2.56 (3H, s, C2-
Me), 3.31 (1H, dd, J¼8.5, 14 Hz, CH2SPh), 3.48 (2H, q,
J¼7 Hz, OCH2CH3), 3.63 (1H, dd, J¼4.5, 14 Hz, CH2SPh),
3.58–3.62 and 3.86–4.01 (each 2H, m, OCH2CH2O), 4.15
(1H, dd, J¼4.5, 8.5 Hz, a-H), 4.52 (2H, s, OCH2Ph), 4.82
(2H, s, C5-CH2), 5.15 (2H, AB type, J¼12.5 Hz, CO2CH2-
Ph), 7.18–7.37 (15H, m, aromatic H), 8.57 (1H, s, C6-H),
8.68 (1H, s, imine H).
resolution MS calcd for C17H19NO3S (Mþ): 317.1085.
Found: 317.1094.
3.2.4. Benzyl 2-amino-3-(4-nitrophenylthio)propanoate
(7d). A colorless oil. IR nmax (KBr): 3381, 3065, 1736,
1579, 1510, 1338 cm21. 1H NMR (CDCl3) d: 3.26 (1H, dd,
J¼7, 14 Hz, SCH2), 3.47 (1H, dd, J¼5, 14 Hz, SCH2),
3.80 (1H, dd, J¼5, 7 Hz, CH(NH2)CO), 5.15 (2H, s,
CH2Ph), 7.31–7.42 (7H, m, aromatic), 8.10–8.11 (2H, m,
aromatic H). 13C NMR (CDCl3) dC: 37.3, 53.9, 67.3,
123.9, 127.1, 128.3, 128.6, 128.7, 135.0, 145.4, 146.0,
173.1. FAB-MS m/z: 333 (MþHþ). High-resolution MS
calcd for C16H17N2O4S (MþHþ): 333.0909. Found:
333.0911.
3.2. General procedure for catalytic b-replacement
reaction
3.2.5. Benzyl 2-amino-3-benzylthiopropanoate (7e). A
Under a nitrogen atmosphere, an acetonitrile solution of 3
(0.1 M, 0.1 mL, 0.01 mmol) and an acetonitrile solution of
LiClO4 (0.1 M, 0.1 mL, 0.01 mmol) were added to a stirred
solution of serine-O-carbonate 4c (50.6 mg, 0.20 mmol) and
thiol (0.22 mmol) in acetonitrile (2 mL) at room tempera-
ture, and the whole was stirred at room temperature until 4c
disappeared on TLC (see Table 1). After concentration
under reduced pressure, the resultant residue was purified by
silica gel column chromatography (ethyl acetate/
hexane¼1:1) to afford S-substituted cysteines 7. The yields
are summarized in Table 1. Spectral properties of the
products 7a–h are as follows.
colorless oil. IR nmax (KBr): 3371, 3061, 1736, 1496 cm21
.
1H NMR (CDCl3) d: 2.66 (1H, dd, J¼7, 14 Hz, SCH2), 2.83
(1H, dd, J¼5, 14 Hz, SCH2), 3.63 (1H, dd, J¼5, 7 Hz,
CH(NH2)CO), 3.70 (2H, s, PhCH2S), 5.15 (2H, s, CH2Ph),
7.20–7.36 (10H, m, aromatic H). 13C NMR (CDCl3) dC:
36.4, 36.6, 54.1, 66.9, 127.1, 128.3, 128.4, 128.5, 128.6,
128.9, 135.4, 137.8, 173.8. EI-MS m/z: 301 (Mþ, 40.1), 210
(Mþ2Bn, 57.8), 166 (Mþ2CO2Bn, 78.3), 91 (Bnþ, 100).
High-resolution MS calcd for C17H19NO2S (Mþ): 301.1136.
Found: 301.1139.
3.2.6. Benzyl 2-amino-3-ethylthiopropanoate (7f). A
colorless oil. IR nmax (KBr): 3370, 3033, 2965,
3.2.1. Benzyl 2-amino-3-phenylthiopropanoate (7a). A
1
1737 cm21
.
1H NMR (CDCl3) d: 1.22 (3H, t, J¼7 Hz,
colorless oil. IR nmax (KBr): 3373, 3060, 1737 cm21. H
CH2CH3), 2.52 (2H, q, J¼7 Hz, CH2CH3), 2.80 (1H, dd,
J¼7, 14 Hz, SCH2), 2.96 (1H, dd, J¼5, 14 Hz, SCH2), 3.68
(1H, dd, J¼5, 7 Hz, CH(NH2)CO), 5.17 (2H, s, CH2Ph),
7.30–7.37 (5H, m, aromatic H). 13C NMR (CDCl3) dC:
14.7, 26.5, 36.8, 54.2, 66.9, 128.3, 128.4, 128.6, 135.4,
174.0. FAB-MS m/z: 240 (MþHþ). High-resolution MS
calcd for C12H18NO2S (MþHþ): 240.1058. Found:
240.1058.
NMR (CDCl3) d: 3.16 (1H, dd, J¼7, 14 Hz, SCH2), 3.35
(1H, dd, J¼5, 14 Hz, SCH2), 3.68 (1H, dd, J¼5, 7 Hz,
CH(NH2)CO), 4.95–5.13 (2H, AB type, J¼12 Hz, CH2Ph),
7.17–7.41 (10H, m, Aromatic H). 13C NMR (CDCl3) dC:
39.4, 53.9, 66.9, 126.8, 128.1, 128.3, 128.4, 128.5, 129.0,
130.5, 135.3, 173.5. EI-MS m/z: 287 (Mþ, 38.7), 152
(Mþ2CO2Bn, 73.3), 91 (Bnþ, 100). High-resolution
MS calcd for C16H17NO2S (Mþ): 287.0980. Found:
287.0969.
3.2.7. Benzyl 2-amino-3-(3-hydroxypropylthio)propano-
ate (7g). A colorless oil. IR nmax (KBr): 3370, 3033, 2965,
1
3.2.2. Benzyl 2-amino-3-(p-tolylthio)propanoate (7b). A
1737 cm21. H NMR (CDCl3) d: 1.80 (2H, qn, J¼7 Hz,
colorless oil. IR nmax (KBr): 3376, 3030, 2925, 1737 cm21
.
HOCH2CH2CH2S), 2.65 (2H, t, J¼7 Hz, HOCH2CH2CH2S),
2.90 (1H, dd, J¼7, 14 Hz, SCH2), 2.98 (1H, dd, J¼5, 14 Hz,
SCH2), 3.72 (2H, t, J¼7 Hz, HOCH2CH2CH2S), 3.70–3.74
(1H, m, CH(NH2)CO), 5.18 (2H, s, CH2Ph), 7.30–7.37 (5H,
m, aromatic H). 13C NMR (CDCl3) dC: 26.5, 36.8, 38.9,
54.2, 56.0, 66.9, 128.3, 128.4, 128.6, 135.4, 174.0. EI-MS
m/z: 269 (Mþ, 4.2), 91 (Bnþ, 100). High-resolution
MS calcd for C13H19NO3S (Mþ): 269.1085. Found:
269.1095.
1H NMR (CDCl3) d: 2.31 (3H, s, Ar-CH3), 3.11 (1H, dd,
J¼7, 14 Hz, SCH2), 3.38 (1H, dd, J¼5, 14 Hz, SCH2), 3.64
(1H, dd, J¼5, 7 Hz, CH(NH2)CO), 4.95–5.11 (2H, AB
type, J¼12 Hz, CH2Ph), 7.06–7.10 (2H, m, aromatic),
7.25–7.36 (7H, m, aromatic H). 13C NMR (CDCl3) dC:
21.0, 40.1, 53.9, 66.9, 128.3, 128.4, 128.6, 129.8, 131.1,
131.4, 135.4, 137.1, 173.6. FAB-MS m/z: 302 (MþHþ).
High-resolution MS calcd for C17H20NO2S (MþHþ):
302.1215. Found: 302.1211.
3.2.8. Dibenzyl 2,10-diamino-4,8-dithiaundecanedioate
(7h). A colorless oil. IR nmax (KBr): 3364, 3033, 2948,
3.2.3. Benzyl 2-amino-3-(4-methoxyphenylthio)propano-
ate (7c). A colorless oil. IR nmax (KBr): 3373, 3033,
1737 cm21. 1H NMR (CDCl3) d: 3.07 (1H, dd, J¼7, 14 Hz,
SCH2), 3.23 (1H, dd, J¼5, 14 Hz, SCH2), 3.62 (1H, dd,
J¼5, 7 Hz, CH(NH2)CO), 3.77 (3H, s, OCH3), 4.94–5.10
(2H, AB type, J¼12 Hz, CH2Ph), 6.80–6.83 (2H, m,
aromatic H), 7.25–7.39 (7H, m, aromatic H). 13C NMR
(CDCl3) dC: 41.2, 53.9, 55.3, 66.9, 114.7, 124.9, 128.2,
128.4, 128.5, 134.2, 135.4, 159.4, 173.6. EI-MS m/z: 317
(Mþ, 80.3), 182 (Mþ2CO2Bn, 37.7), 91 (Bnþ, 100). High-
1
1734 cm21. H NMR (CDCl3) d: 1.79 (2H, qn, J¼7 Hz,
SCH2CH2CH2S), 2.57 (4H, t, J¼7 Hz, SCH2CH2CH2S),
2.78 (2H, dd, J¼7, 14 Hz, SCH2), 2.90 (2H, dd, J¼5, 14 Hz,
SCH2), 3.68 (2H, dd, J¼7, 5 Hz, CH(NH2)CO), 5.17 (4H, s,
CH2Ph), 7.30–7.42 (10H, m, aromatic H). 13C NMR
(CDCl3) dC: 20.9, 26.5, 36.8, 54.2, 66.9, 128.3, 128.4,
128.6, 135.4, 174.0. EI-MS m/z: 462 (Mþ, 0.1), 91 (Bnþ,
100). High-resolution MS calcd for C23H30N2O4S2 (Mþ):
462.1647. Found: 462.1637.

4872
K. Miyashita et al. / Tetrahedron 59 (2003) 4867–4872
3.3. 1H NMR Analysis of the reaction in the absence and
presence of metal ion
Nakamura, K.; Kikuchi, J. Chem. Lett. 1990, 1765.
(e) Murakami, Y.; Hisaeda, Y.; Miyajima, T.; Sakata, H.;
Kikuchi, J. Chem. Lett. 1993, 645. (f) Murakami, Y.; Hisaeda,
Y.; Miyajima, T.; Kikuchi, J. Chem. Lett. 1993, 2093.
4. For example: Davis, L.; Metzler, D. E. The Enzymes; Boyer,
P. D., Ed.; Academic: New York, 1972; Vol. VII, pp 1–22.
5. (a) Refsum, H.; Ueland, P. M.; Nygard, O.; Vollset, S. E.
Annu. Rev. Med. 1998, 49, 31. (b) Mills, J. L.; McPartlin, J. M.;
Kirke, P. N.; Lee, Y. J.; Conle, M. R.; Weir, D. G. Lancet
1995, 345, 149. (c) Clarke, R.; Smith, A. D.; Jobst, K. A.;
Refsum, H.; Sutton, L.; Ueland, P. M. Arch. Neurol. 1998, 55,
1449.
A solution of 3 (17 mg, 0.05 mmol) and serine-O-carbonate
4c (12.7 mg, 0.05 mmol) in CH2Cl2 (2 mL) was stirred at
room temperature for 5 min. After being concentrated under
reduced pressure, the reaction mixture was azeotropically
evaporated with benzene for three times. The resultant
residue was dissolved in deuterated acetonitrile (0.5 mL). A
solution of metal salt (0.1 M, LiClO4, Zn(OAc)2 or NaClO4
in deuterated acetonitrile, 25 mL, 0.0025 mmol) was added
to the above solution, 1H NMR spectrum of which was taken
from time to time. Disappearance of 5c and appearance of 8
are summarized in Figure 4.
6. (a) Taoka, S.; Ojha, S.; Shan, X.; Kruger, W. D.; Banerjee, R.
J. Biol. Chem. 1998, 273, 25179. (b) Taoka, S.; Banerjee, R.
J. Inorg. Biochem. 2001, 87, 245. (c) Taoka, S.; Lepore, B. W.;
¨
Kabil, O.; Ojha, S.; Ringe, D.; Banerjee, R. Biochemistry
Acknowledgements
2002, 41, 10454.
7. (a) Miyashita, K.; Miyabe, H.; Kurozumi, C.; Imanishi, T.
Chem. Lett. 1995, 487. (b) Miyashita, K.; Miyabe, H.;
Kurozumi, C.; Tai, K.; Imanishi, T. Tetrahedron 1996, 52,
12125.
A part of this work was supported by SUNBOR Grant.
References
8. (a) Miyashita, K.; Miyabe, H.; Tai, K.; Kurozumi, C.;
Imanishi, T. Chem. Commun. 1996, 1073. (b) Miyashita, K.;
Iwaki, H.; Tai, K.; Murafuji, H.; Imanishi, T. Chem. Commun.
1998, 1987. (c) Miyashita, K.; Miyabe, H.; Tai, K.; Kurozumi,
C.; Iwaki, H.; Imanishi, T. Tetrahedron 1999, 55, 12109.
(d) Miyashita, K.; Miyabe, H.; Tai, K.; Iwaki, H.; Imanishi, T.
Tetrahedron 2000, 56, 4691.
1. For example: Emery, V. C.; Akhtar, M. In Enzyme
Mechanisms: Pyridoxal Phosphate Dependent Enzymes;
Page, M. I., Williams, A., Eds.; The Royal Society of
Chemistry: London, 1987; pp 345–389.
2. For recent review article: Murakami, K.; Kikuchi, J.; Hisaeda,
Y.; Hayashida, O. Chem. Rev. 1996, 96, 721.
9. A part of this work appeared in a preliminary communication:
Miyashita, K.; Murafuji, H.; Iwaki, H.; Yoshioka, E.;
Imanishi, T. Chem. Commun. 2002, 1922.
3. (a) Weinre, W.; Winkler, J.; Zimmerman, S. C.; Czarnil,
A. W.; Breslow, R. J. Am. Chem. Soc. 1985, 107, 4093.
(b) Murakami, Y.; Kikuchi, J.; Kitazaki, T. J. Chem. Soc.,
Chem. Commun. 1998, 143. (c) Murakami, Y.; Kikuchi, J.;
Hisaeda, Y.; Nakamura, K.; Kitazaki, T.; Kaya, H. Bull. Chem.
Soc. Jpn 1990, 63, 2339. (d) Murakami, Y.; Hisaeda, Y.;
10. Although we have employed optically active serine-O-
carbonate 4c, optical activity of the product 7a was completely
lost.