Synthesis of arabinitol 1-phosphate and its use for characterization of arabinitol-phosphate dehydrogenase

Article abstract of DOI:10.1016/j.carres.2004.11.030

D-Arabinitol 1-phosphate (Ara-ol1-P), a substrate for d-arabinitol- phosphate dehydrogenase (APDH), was chemically synthesized from d-arabinonic acid in five steps (O-acetylation, chlorination, reduction, phosphorylation, and de-O-acetylation). Ara-ol1-P was used as a substrate for the characterization of APDH from Bacillus halodurans. APDH converts Ara-ol1-P to xylulose 5-phosphate in the oxidative reaction; both NAD⁺ and NADP⁺ were accepted as co-factors. Kinetic parameters for the oxidative and reductive reactions are consistent with a ternary complex mechanism.

Full text of DOI:10.1016/j.carres.2004.11.030

Carbohydrate
RESEARCH
Carbohydrate Research 340 (2005) 539–546
Synthesis of arabinitol 1-phosphate and its use for characterization
of arabinitol–phosphate dehydrogenase
a
Nikolai V. Soroka, Anna A. Kulminskaya, Elena V. Eneyskaya,
a,
a
*
a
Konstantin A. Shabalin, Andrei V. Uffimtcev, Mira Povelainen,
a
b
b
Andrei N. Miasnikov and Kirill N. Neustroev
a
a
Molecular and Radiation Biology Division, Petersburg Nuclear Physics Institute, Russian Academy of Science,
Gatchina 188300, Russia
b
Danisco Innovation, Sokeritehtaantie 20, Kantvik 02460, Finland
Received 8 June 2004; accepted 29 November 2004
Available online 2 February 2005
In memoriam of Dr. Kirill N. Neustroev
Abstract—D-Arabinitol 1-phosphate (Ara-ol1-P), a substrate for D-arabinitol–phosphate dehydrogenase (APDH), was chemically
synthesized from D-arabinonic acid in five steps (O-acetylation, chlorination, reduction, phosphorylation, and de-O-acetylation).
Ara-ol1-P was used as a substrate for the characterization of APDH from Bacillus halodurans. APDH converts Ara-ol1-P to xylu-
+
+
lose 5-phosphate in the oxidative reaction; both NAD and NADP were accepted as co-factors. Kinetic parameters for the oxi-
dative and reductive reactions are consistent with a ternary complex mechanism.
ꢀ
2004 Elsevier Ltd. All rights reserved.
Keywords: Pentitol metabolism; D-Arabinitol–phosphate dehydrogenase; Arabinitol 1-phosphate; Xylulose 5-phosphate
1
. Introduction
cell by permease, where a dehydrogenase oxidizes the
five-carbon sugar alcohol to its corresponding keto-
pentose. Then keto-pentoses are phosphorylated by
specific kinases, and enter the pentose–phosphate path-
Metabolic pathways of polyols, which are widespread in
nature, have been investigated extensively in fungi, bac-
teria, and higher organisms. One reason is their thera-
peutic significance. In the last decade, it was shown
that xylitol is an anti-cariogenic agent. This pentitol
can be transported into caries-causing oral bacteria,
and inhibits bacterial growth due to its influence on pen-
titol phosphate metabolism.
was discovered to be a quantitative diagnostic marker
4
for invasive candidiasis. L-Arabinosuria, an abnormal-
6
,7
way. More recently, an alternative metabolic path-
way of polyols was discovered, in which the alcohols
are first phosphorylated by the PTS before their oxida-
tion by polyol phosphate dehydrogenases. Xylitol
and ribitol were reported to be metabolized by this
1
2
,3
8
Likewise, D-arabinitol
route in several species of Lactobacilli. The xylitol 5-
phosphate dehydrogenase responsible for assimila-
tion of xylitol via this pathway was isolated and
9
described. Earlier, we had purified and characterized
ity of human polyol metabolism, has been reported to be
5
related to the deficiency of L-arabinitol dehydrogenase.
In general, metabolism of pentitols in both prokaryotes
and eukaryotes involves polyol transportation into the
a new polyol–phosphate dehydrogenase (E.C. 1.1.1.X)
from the cell lysate of Enterococcus avium involved
in a similar catabolic pathway for D-arabinitol. It exhib-
1
0
ited a new type of specificity. This NADH-dependent
arabinitol–phosphate dehydrogenase (APDH) cata-
lyzes the oxidation of D-arabinitol 1-phosphate and D-
arabinitol 5-phosphate into D-xylulose 5-phosphate
*
Corresponding author. Tel.: +7 81271 32 014; fax: +7 81271 32 303;
e-mail: kulm@omrb.pnpi.spb.ru
0
008-6215/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carres.2004.11.030

540
N. V. Soroka et al. / Carbohydrate Research 340 (2005) 539–546
D-Arabinitol extracellular
CH OH
2
HO
H
H
OH
H
OH
CH OH
2
Phosphotransferase system
D-Arabinitol 5-phosphate
D-Arabinitol 1-phosphate
CH OPO H
CH OH
2
3
2
2
HO
H
H
HO
H
H
OH
OH
OH
H
H
OH
CH OH
CH OPO H
2 3 2
2
D-Arabinitol 1-phosphate
dehydrogenase (E.C. 1.1.1.X)
D-Ribulose 5-phosphate
D-Xylulose 5-phosphate
CH OH
CH2OH
2
O
O
HO
H
H
H
H
OH
OH
OH
CH OPO H
3 2
CH2OPO3H2
2
Figure 1. Catabolic routes for D-arabinitol catalyzed by D-arabinitol 1-phosphate dehydrogenase.
(
1
Xul5P) and D-ribulose 5-phosphate, respectively (Fig.
).
To analyze the specificity and kinetic properties of
Enterococcus APDH, we originally used D-arabinitol
-phosphate produced by NaBH reduction of commer-
2. Results and discussion
2.1. Synthesis of Ara-ol1-P
1
In spite of extensive interest in enzymes of polyol
metabolism, there are only a few methods for the
production of pentitol phosphate substrates for
dehydrogenases involved into sugar metabolic path-
ways. The approaches have been mainly enzymatic,
and use a selection of kinases, isomerases, epimerases,
4
cial Xul5P followed by chromatographic separation of
the stereoisomeric products. However, this method is
expensive, tedious, and low yielding. We now report a
new, five-step chemical synthesis of Ara-ol1-P from D-
arabinonic acid, which can be carried out on the 10 g
scale. The Ara-ol1-P so produced was used for detailed
characterization of recombinant Bacillus halodurans
APDH expressed in Escherichia coli.
1
1,12
transketolases etc.
The advantages provided by the
selectivity of enzymatic synthesis is offset, though, by
the need for chromatographic separation of products

N. V. Soroka et al. / Carbohydrate Research 340 (2005) 539–546
541
HO
HO
Cl
O
O
O
HO
H
Ac-O
H
Ac-O
H
NaBH4
Ac O; HClO
PCl
5
2
4
H
H
OH
OH
H
H
O-Ac
O-Ac
H
H
O-Ac
O-Ac
OH
O-Ac
O-Ac
1
2
3
O
O
O
-
+
O
P
Cl
O
P
OH
O
P
O Na
+
CH OH
2
H C
-
2
H C
2
OH
H C
2
Cl
O Na
Ac-O
H
H
Ac-O
H
H
Ac-O
H
H
HO
H
H
POCl3
NH OH
H O
2
4
O-Ac
O-Ac
O-Ac
O-Ac
O-Ac
OH
OH
NaOH
H
O-Ac
H
H
H
O-Ac
O-Ac
O-Ac
OH
4
5
6
7
Figure 2. General scheme of Ara-ol1-P synthesis.
after each step, which also limits the scale of the
reaction.
revealing a single polypeptide band with an estimated
molecular mass of 36 ± 2 kDa. Analytical gel-filtration
on a Superose 12 and Sephacryl S200 columns revealed
a single symmetrical APDH peak corresponding to
140 ± 5 kDa. Thus, in its native state, the B. halodurans
arabinitol 1-phosphate dehydrogenase is a tetramer simi-
larly to the APDH from E. avium. The specificity of the
enzyme and its enzymatic properties were investigated
using the oxidative reaction of Ara-ol1-P as described
in Experimental section. The products of Ara-ol1-P oxi-
Investigations of pentitol metabolic pathways require
pentitol 1-phosphates and pentitol 5-phosphates. D-
Xylitol 5-phosphate and D-arabinitol 5-phoshate can
1
0
be obtained by NaBH reduction of D-xylofuranose
4
5
-phosphate and D-arabinopyranose 5-phosphate,
1
3
respectively, as reported earlier. However, chemical
approaches for production of pentitol 1-phosphates,
particularly Ara-ol1-P, have not been reported so far.
The method used for synthesis of arabinitol 1-phosphate
is set out in Figure 2. D-Arabinonic acid (readily
obtained from D-arabinose) was O-acetylated and con-
verted to the acid chloride. Reduction with NaBH4
afforded O-acetylated D-arabinitol selectively unpro-
+
dation by APDH with NAD were dephosphorylated by
phosphatase digestion and analyzed by HPLC. Only
xylulose and trace quantities of arabinitol were detected
demonstrating that the product of the reaction was D-
xylulose 5-phosphate (see scheme given in Fig. 1). Ana-
lysis of dephosphorylated products of the APDH-medi-
ated D-xylulose 5-phosphate reduction using NADH as
a co-factor revealed only D-arabinitol as a product after
the dephosphorylation. As expected, the only product of
the APDH-catalyzed oxidation of Ara-ol5-P was D-ribu-
lose 5-phosphate. Thus, substrate specificity of D-ara-
binitol phosphate dehydrogenase from Bacillus
halodurans seems to be limited to the oxidation of D-ara-
binitol 1-phosphate to D-xylulose 5-phosphate and D-
arabinitol 5-phosphate to D-ribulose 5-phosphate using
tected at position 1. Phosphorylation with POCl in pyr-
3
idine, just above the freezing point of the solvent gave
the O-acetylated D-arabinitol 1-phosphate, which could
be de-O-acetylated by 20% aqueous ammonia without
the phosphoryl migration observed under acid and neu-
1
4–16
tral conditions.
for the final product.
Chromatography was required only
2.2. Enzymatic properties of APDH and kinetic
parameters
+
either NAD or NADH as co-factors (Fig. 1).
APDH activity was assayed by NADH reduction of
xylulose 5-phosphate. The enzyme used in this study
was a recombinant APDH from B. halodurans produced
in E. coli and partly purified. The preparation of
APDH was essentially homogeneous on a SDS-PAGE
Maximal rates of the APDH in the forward (oxida-
tive) direction were found at pH 8.1–8.5 and in the
reverse (reductive) direction––at pH 6.8–7.2. Experi-
ments on pH––and temperature stability were per-
formed using Ara-ol1-P as a substrate. The enzyme
1
0

542
N. V. Soroka et al. / Carbohydrate Research 340 (2005) 539–546
Table 1. Kinetic data obtained for the purified APDH from B. halodurans
2
(mM )
Polyol substrate,
reaction direction
Substrate K
m
(mM)
Co-factor K
m
(mM)
Ternary complex K
m
V
max (lmol/min per mg)
þ
þ
þ
Ara-ol1-P
NAD
m
Ara-ol1-P;NAD
Ara-ol1-P;NAD
max
Ara-ol1-P, oxidative
Xul5P, reductive
Rul5P, reductive
Fru5P, reductive
Rib5P, reductive
K
K
K
K
K
4.0 ± 0.1
K
K
K
K
K
0.5 ± 0.02
0.02 ± 0.0004
0.02 ± 0.001
0.02 ± 0.001
2.8 ± 0.08
K
K
K
K
K
3.8 ± 0.1
0.01 ± 0.002
0.4 ± 0.005
0.02 ± 0.002
ND
V
V
V
V
V
5.9 ± 0.15
27.0 ± 0.5
3.6 ± 0.1
m
m
Xul5P;NADH
Xul5P
m
NADH
m
Xul5P;NADH
max
0.22 ± 0.004
4.0 ± 0.051
0.27 ± 0.005
40.0 ± 1.2
m
Rul5P;NADH
m
Fru5P;NADH
m
Rib5P;NADH
m
Rul5P
m
NADH
m
Rul5P;NADH
max
Fru5P
m
NADH
m
Fru5P;NADH
max
5.0 ± 0.15
1.2 ± 0.1
Rib5P
m
NADH
m
Rib5P;NADH
max
Abbreviations: Ara-ol1-P, arabinitol 1-phosphate; Fru5P, fructose 5-phosphate; Rib5P, ribose 5-phosphate; Rul5P, ribulose 5-phosphate; Xul5P,
xylulose 5-phosphate.
Figure 3. Initial velocities of APDH-catalyzed reactions with Ara-ol1-P as a substrate.
was stable at pH 7.0 during 48 h at 4 ꢁC (data not
shown). APDH activity was apparently optimal at
reactions. Firstly, D-xylulose was prepared by oxidation
of D-arabinitol by D-arabinitol dehydrogenase from
Acetobacter suboxidans. Secondly, D-xylulose 5-phos-
phate was obtained by phosphorylation with ATP and
xylulose kinase; the final product was isolated as its
5
0 ꢁC; however between 40 and 50 ꢁC, 50% of the en-
zyme activity was retained for 5 min. Above 55 ꢁC activ-
ity decreased drastically. Like Enterococcus APDH,
1
0
+
12
the enzyme accepted both NADP and NADHP as
co-factors. NADH was about 12 times more effective
over the tested concentration range (up to 3.5 mM).
Table 1 summarizes kinetic parameters obtained for
the forward and reverse directions of B. halodurans
APDH catalysis. Analysis of the double-reciprocal pri-
hydrazone. The suggested synthesis of Ara-ol1-P en-
ables its gram-scale production and therefore, synthesis
of xylulose 5-phosphate in a one-stage enzymatic reac-
tion using Bacilus APDH and synthetic Ara-ol1-P
makes such an approach quite attractive.
+
mary plots of velocity against Ara-ol1-P and NAD
shown in Figure 3 indicates the expected ternary com-
3. Experimental
3.1. Materials
1
7,18
plex mechanism. As can be seen from Table 1, the
values for ternary K for both substrates are of the same
m
order of magnitude while values for Vmax in the reverse
and forward directions differ about 4.5 times. Compari-
son of arabinitol–phosphate dehydrogenases from both
E. avium and B. halodurans reveals that the two enzymes
have rather similar kinetic characteristics in the reac-
tions with Ara-ol1-P or xylulose 5-phosphate. However,
B. halodurans APDH has a wider substrate specificity
than the Enterococcus enzyme and can use fructose 5-
phosphate and ribose 5-phosphate in the reductive reac-
tion (kinetic parameters are given in Table 1).
Xylulose 5-phosphate (Xul5P), arabinose 5-phosphate,
ribulose 5-phosphate (Rul5P), fructose 5-phosphate
(Fru5P), and ribose 5-phosphate (Rib5P) were pur-
chased from Sigma Chemical Co. (St. Louis, MO,
USA) and Fluka Chemie Gmbh (Buchs, Switzerland).
Xylitol 5-phosphate (Xyl-ol5-P) was synthesized accord-
1
3
ing to the protocol described earlier.
5-phosphate (Ara-ol5-P) was produced by NaBH
Arabinitol
4
1
reduction of arabinose 5-phosphate followed by a chro-
9
TM
matographic separation on a Dextro-Pak cartridge
The enzymatic synthesis of D-xylulose 5-phosphate
described earlier comprises two sequential enzymatic
column (8 · 100 mm) WATO85650 from Waters Co.

N. V. Soroka et al. / Carbohydrate Research 340 (2005) 539–546
543
1
Milford, USA) using isocratic elution in H O. D-Ara-
0
(
3.4. 2,3,4,5-Tetra-O-acetyl-D-arabinitol (4)
2
binonic acid was obtained by oxidation of D-arabinose
and isolated as Ca-salt form according to the procedure
previously described.
Acyl chloride 3 (9 g) was dissolved in 20 mL of dry tet-
rahydrofuran and added to a stirred solution of NaBH4
2
0
(3 g) in water (100 mL) at 0 ꢁC. After stirring the mix-
3
.2. 2,3,4,5-Tetra-O-acetyl-D-arabinonic acid (2)
ture for 1 min, a solution of 4 M HOAc (30 mL) in
water was added. After extraction with CHCl3
(3 · 200 mL), the organic phase was separated, washed
Calcium arabinonate (10 g) was added with stirring por-
tion wise to a solution of HClO (5% v/v) in Ac O
with KHCO (15 g in 100 mL of water) and water
4
2
3
(
100 mL) cooled to 0 ꢁC (ꢀ1 h). The reaction mixture
(100 mL) and dried (anhyd Na SO ). The solvent was
2
4
was heated to 40 ꢁC and stirring was continued until a
clear solution was obtained. After keeping the reaction
mixture at 40 ꢁC for 1 h, anhyd NaOAc (8 g) was added.
The solution was stirred for 5 min and cooled to 0 ꢁC;
removed by evaporation to yield an oil (7.5 g, 60%):
1
H NMR (CDCl ): d 5.37 (dd, 1H, J 2.3,J3,4 8.9 Hz,
3
2,3
H-3), 5.22 (ddd, 1H, J2,1a 5.1, J2,1b 2.1 Hz, H-2), 5.19
(ddd, 1H, J4,5a 2.5, J4,5b 4.5 Hz, H-4), 4.41 (dd, 1H,
J1a,1b 11.7 Hz, H-1a), 4.25 (dd, 1H, J5a,5b 12.5 Hz, H-
H O (400 mL) was poured into the reaction mixture,
2
1
3
which was subsequently allowed to warm to room tem-
perature (rt, 25 ꢁC) and was kept at rt for 1 h to com-
plete the reaction. The mixture was cooled to 0 ꢁC,
5a), 4.22 (dd, 1H, H-5b), 4.16 (dd, 1H, H-1b);
C
NMR (CDCl ): d 171.11, 170.67, 170.60, 170.42 (s,
3
1C, CO in OAc), 70.50 (s, 1C, C-2), 70.14 (s, 1C, C-3),
69.7 (s, 1C, C-4), 64.89 (s, 1C, C-5), 62.01 (s, 1C, C-1),
20.88 (s, 1C, Me in OAc), 20.75 (s, 3C, Me in OAc);
1
0 M HCl (20 mL) was added and extraction by CHCl3
was initiated immediately (4 · 100 mL). The organic
fractions were combined, dried (Na SO ), filtered, and
+
+
ESI MS [M+Na] m/z 343.1005 calcd for C H NaO ,
13 20 9
2
4
the solvent was removed by rotary evaporation. The resi-
due was dissolved in toluene (120 mL), heated to 50 ꢁC
and cooled to rt several times with stirring until crystals
were formed. The precipitate was filtered off and dried
observed 343.1051.
3.5. Phosphorylation of 4
1
under vacuum to give 15 g (80%) of compound 2: H
NMR (CDCl ): d 5.68 (dd, 1H, J 2.2, J3,4 9.2 Hz,
Oily compound 4 (1.7 g) was diluted in tetrahydrofuran
(36 mL) and diisopropylethylamine (7.2 mL) was added.
3
2,3
H-3), 5.32 (d, 1H, H-2), 5.22 (ddd, 1H, J4,5a 2.6, J4,5b
4
After the solution was cooled to À10 ꢁC, POCl
3
.4 Hz, H-4), 4.28 (1H, dd, J 12.6 Hz, H-5a), 4.13
5
(3.6 mL) was added; the resulting mixture was warmed
to rt and kept for 5 min. The reaction was cooled to
incipient solidification in liquid nitrogen and H O
a,5b
1
3
(
dd, 1 H, H-5b); C NMR (CDCl ): d 170.82, 170.77,
3
1
6
5
70.07, 169.67 (s, 1C, CO in OAc), 69.32 (s, 1C, C-2),
8.60 (s, 1C, C-3), 68.06 (s, 1C, C-4), 61.62 (s, 1C, C-
2
(54 mL) was poured in. The solution was adjusted to
pH 5 by gradual addition of water solution of KHCO3
(3 M, total volume 38 mL) at rt and the solvent (35 g
of tetrahydrofuran) was removed by rotary evaporation.
The impurities were separated by CHCl3 extraction
(2 · 30 mL). The water fraction containing 2,3,4,5-tet-
ra-O-acetyl-D-arabinitol 1-phosphate (6) was acidified
with 10 M HCl (4 mL) and extracted with EtOAc
(3 · 140 mL). The EtOAc fractions were re-extracted
with H O (40 mL) and a solution of NaHCO (0.5 M,
+
), 20.68, 20.63, 20.43, 20.29 (s, 1C, Me in OAc); ESI
+
MS [M+Na] m/z 379.0617 calcd for C H Na O ,
1
3
17
2
10
observed 379.0632.
.3. 2,3,4,5-Tetra-O-acetyl-D-arabinoyl chloride (3)
Diethyl ether (120 mL) was stirred with PCl (7.8 g) for
3
5
1
0 min and tetra-O-acetyl-D-arabinonic acid 2 (11 g)
was added. Stirring was continued until the solution be-
came clear. After additional stirring for 1.5 h, 400 mL of
hexane was added and a gel formed. The gel was crystal-
lized by repeated cycles of stirring, evaporation of small
2
3
5 mL) was slowly added to adjust the water phase to
pH 4–5. The water phase containing 6 was concentrated
by rotary evaporation to 7 mL and used for the synthe-
sis of 7 without isolation. The purity of 6 was analyzed
on an Agilent Zorbax Eclipse XDB-C8 column
(4.6 · 150 mm) using 0.1 M triethylamine phosphate as
a starting buffer (pH 6.7) and 90% MeOH as an eluting
buffer. Elution was carried out with linear gradient (7–
90%) of MeOH, flow rate was 0.8 mL/min. The HPLC
amounts of ether and addition of hexane, yielding 10 g
1
(
70%) of compound 3: H NMR (CDCl ): d 5.83 (dd,
3
1
(
H, J2,3 2.1, J3,4 9.1 Hz, H-3), 5.44 (d, 1H, H-2), 5.21
ddd, 1H, J4,5a 2.7, J4,5b 4.7 Hz, H-4), 4.28 (dd, 1H,
J5a,5b 12.7 Hz, H-5a), 4.11 (dd, 1H, H-5b), 2.18, 2.09,
1
3
2
1
1
6
2
.06, 2.06 (1H, Me in OAc); C NMR (CDCl ): d
3
70.44 (s, 1C, C-1), 169.62, 169.42, 169.39, 168.93 (s,
C, CO in OAc), 76.18 (s, 1C, C-2), 68.09 (s, 1C, C-3),
7.62 (s, 1C, C-4), 61.39 (s, 1C, C-5), 20.63, 20.59,
analysis revealed the only chromatographic peak corre-
1
sponding to compound 6: H NMR (CDCl ): d 5.42
3
(dd, 1H, J2,3 2.8, J3,4 7.5 Hz, H-3), 5.29 (dt, 1 H, J1,2
6.0 Hz, H-2), 5.21 (ddd, 1H, J4,5a 2.6, J4,5b 6.6 Hz, H-
4), 4.27 (dd, 1H, J5a,5b 12.3 Hz, H-5a), 4.10 (dd, 1H,
+
+
0.27, 20.09 (s, 1C, Me in OAc); ESI MS [M+Na]
m/z 375.0459 calcd for C H NaO Cl, observed
1
3
17
9
1
3
3
75.0471.
H-6b), 3.81 (dd, 2H, JH,P 6.3 Hz, CH -1); C NMR
2

544
N. V. Soroka et al. / Carbohydrate Research 340 (2005) 539–546
(
OAc), 69.71 (d, 1C, JC,P 8.8 Hz, C-2), 68.81 (d, 1C, JC,P
CDCl ): d 171.40, 171.09, 170.98, 170.40 (s, 1C, CO in
of 0.1 s was used in all MS modes. Solutions of sub-
stances under analysis (typical concentration 10–20 lmol
in 1:1 MeCN–water containing 0.5 mM NaCl) were in-
fused into the ion source at 2 lL/min (syringe pump).
For accurate mass determinations in the TOF MS
mode, centroid spectra were generated from continuum
3
3
5
2
.4 Hz, C-3), 68.57 (s, 1C, C-4), 62.84 (d, 1C, JC,P
.4 Hz, C-1), 58.24 (s, 1C, C-5), 20,73, 20.65,
+
0.54, 20,54 (s, 1C, Me in OAc). ESI MS [M+Na]
+
m/z 467.0308 calcd for C H Na O P, observed
1
3
19
3
12
+
4
67.0341.
spectra using the (M+Na) adduct of raffinose (a-Galp-
(
1!6)-a-Glcp-(1M2)-b-Fruf) as an internal standard
3
.6. De-O-acetylation of 6
(calcd mass 527.1588 u). The raffinose (raffinose penta-
hydrate, R-0250, Sigma–Aldrich, minimum purity
99%) concentration in each sample was adjusted empiri-
cally to give a peak intensity similar to that of the ara-
binitol analyte.
The solution 6 (7 mL) was treated with 20% (v/v) solu-
tion of NH OH (4 mL) for 3.5 h at rt and NH was
removed by rotary evaporation.
4
3
3
.7. D-Arabinitol 1-phosphate, disodium salt (7)
3.9. Purification of APDH
The solution was adjusted to pH 9 by step-by-step addi-
tion of NaOH (1 M, total volume 2.5 mL) and rotary
evaporation of forming NH , and the residue was con-
The coding sequence of APDH gene was amplified from
B. halodurans (JCM 9153) genomic DNA using oligonu-
0
cleotide primers oBHDH5 (5 -GAGGGAATTCATG-
0
3
centrated under vacuum to 3 mL. The resulting solution
was mixed with MeOH (45 mL) and left for 1 h at rt to
form a solid. The di-Na-salt of D-arabinitol 1-phosphate
was centrifuged, washed with MeOH (5 mL) and freeze
AAAGCATTAGTAAAAACACAACATGGC-3 ) and
0
oBHDH3 (5 -CAAAGATCTAGAAGCTTCGTCCA-
0
TACGGTCCTCCTTTTCCCTAAT-3 ). The PCR
product was digested with EcoRI and HindIII and
ligated to EcoRI and HindIII digested vector pTAC.
Construction of these vectors has been described
1
dried to give 0.5 g (34%) of pure title compound. H
NMR(D O): d 3.95 (dt, 1H, J 6.2, J2,3 1.9 Hz, H-2),
2
1,2
2
3
3
(
.80 (ddd, 1H, J1a,1b 12.8, J1a,P 6.5 Hz, H-1a), 3.78
ddd, 1H, J1b,P 6.5 Hz, H-1b), 3.57 (dd, 1H, J4,5a 3.0,
J5a,5b 11.7 Hz, H-5a), 3.68 (ddd, 1H, J4,5b 5.9, J3,4
earlier. The resulting plasmid was designated pTA-
C(APDH). E. coli (pTAC(APDH)) strain was cultivated
in Luria Bertani medium containing 100 mg/L ampicil-
lin. The expression of APDH was induced by addition
of IPTG (1 mM) when the cell density reached
OD600 = 0.2. After induction, cultivation was continued
overnight at 30 ꢁC. The cells were collected by centrifu-
gation and disrupted by sonication (by multiple 20 s
pulses with intermediate cooling on ice bath) until essen-
tially all cells were broken as judged by microscopic
examination. Ammonium sulfate precipitate containing
the APDH activity was dialyzed against 20 mM Tris
HCl buffer, pH 7.8, supplemented with 3 mM DTT,
applied to a DEAE 5PW (Pharmacia Biotech, Uppsala,
Sweden) column (21.5 · 159 mm) equilibrated with the
same buffer, and eluted with a linear gradient (0–1 M)
of NaCl. Fractions with APDH activity were pooled,
concentrated using an Amicon PM-30 membrane to
10 mL, and dialyzed against 20 mM Tris HCl buffer,
pH 7.8, supplemented with 3 mM DTT. Obtained pro-
tein solution was chromatographed on a Mono Q HR
(5/5) (Pharmacia) with a linear gradient (0–1 M) of
NaCl in the same buffer. The APDH activity was ap-
plied on a Sepharose Blue CL6B (Pharmacia) column
(10 · 50 mm) equilibrated with 50 mM Tris HCL buffer,
pH 8.0, 100 mM NaCl, 3 mM DTT, and eluted with
3 mM NADH in the same buffer. Purified enzyme was
stored in 50% glycerol–20 mM Tris HCl buffer solution
at À20 ꢁC. Aliquots of the enzyme were taken for all fur-
ther investigations. All purification steps were carried
out at 4 ꢁC.
8
.4 Hz, H-4), 3.61 (dd, 1H, H-3), 3.59 (dd, 1H, H-5b);
C NMR (D O): d 76.12 (s, 1C, C-4), 75.46 (s, 1C, C-
1
3
2
3
JC,P 4.2 Hz, C-2), 68.28 (s, 1C, C-5). ESI MS
), 74.44 (dd, 1C, JC,P 5.0 Hz, C-2), 70.36 (dd, 1C,
+
+
[
served 298.9925.
M+Na] m/z 298.9885 calcd for C H Na O P, ob-
5 11 3 8
3
.8. Analytical methods
The molecular mass of the protein under denaturing
conditions was estimated by SDS-PAGE on a 10% poly-
2
1
acrylamide gel according to the method of Laemmli
using molecular weight calibration standards kit LMW
14,400–94,000) from Pharmacia, Sweden. Isoelectric
(
focusing was performed on Servalyt PRECOATES
plates 3–10 (Serva Electrophoresis GmbH, Heidelberg,
Germany). Protein concentrations were determined by
2
2 1
the method of Lowry using BSA as a standard.
H
1
3
NMR spectra and C NMR spectra were recorded with
an AMX-500 Bruker spectrometer. Standard impulse se-
quence DEPT 135ꢁ from spectrometer library was used.
Mass spectrometric analysis was performed with a Q-
TM
Tof 2 mass spectrometer (Waters Corporation, Micro-
mass MS Technologies, Manchester, UK). Calibration
of the TOF analyzer (single-reflectron mode, resolution
>
10,000 FWHM) was obtained over the m/z range 50–
000 using a solution of NaI (1.5 g/L) in 1:1 2-propa-
nol–water. A scan time of 1.5 s with an interscan delay
1

N. V. Soroka et al. / Carbohydrate Research 340 (2005) 539–546
545
þ
NAD
M
Ara
M
-
ol1-P
3
.10. Enzymatic properties and kinetic parameters
þ
ol1 P;NAD
ÀK
ÀK
Ara
-
-
KM
¼
¼
:
ð2Þ
ðabscissa Þ ðabscissa Þ
1
2
The measurement of the enzyme activity in the reductive
reverse) reaction of Xul5P was performed according to
All kinetic experiments were performed in 20 mM Tris
HCl, pH 7.2, at 20 ꢁC. The same procedures were ap-
plied for the reductive reaction using Xul5P as a sub-
strate and NADH as a co-factor at pH 7.0 and 20 ꢁC.
(
Ref. 10 in 50 mM Tris HCl buffer, pH 7.0, 1 mM DTT,
with 0.07 mM NADH, 1 mM Xul5P. One unit of the en-
zyme activity was defined as an amount catalyzing the
oxidation of 1 lmol of NADH in 1 min under the con-
ditions described. Enzyme activity in the oxidative (for-
ward) direction was measured in 20 mM Tris HCl,
+
NADP and NADPH were tested in the same manner
in the forward and reverse reactions using Xul5P, Ara-
ol1-P, and Ara-ol5-P at corresponding pH optima
(20 mM Tris HCl, pH 7.1 and 8.3).
+
pH 8.5, 1 mM DTT, and 0.5 mM NAD using Ara-
Products of the oxidative and reductive reactions were
analyzed after the incubation of APDH (0.1–0.01 U)
with different substrates under standard reaction condi-
tions. The reaction mixture was then treated with a
ol1-P or Ara-ol5-P as a substrate. To determine the
pH optimum, purified APDH was incubated in the fol-
lowing 100 mM buffer solutions: pH 3.0j5.5––NaOAc
buffer; pH 5.5j7.0––MES buffer; pH 7.0j9.0––Tris HCl
buffer. Each buffer contained 1 mM DTT. The dehydro-
genase activity was measured under the standard assay
conditions. For pH stability assays, purified enzyme
1
0
phosphatase (Sigma Chemical Co., cat. No P4252)
and the resulted dephosphorylated polyols were ana-
lyzed on carbohydrate analysis column (size
.9 · 300 mm, Waters W0191H005).
a
3
(
0.04 U) was incubated for 18 h, at 4 ꢁC in the same buf-
fer solutions as described above. After incubation the
dehydrogenase activity was measured for the reverse
reaction under the standard conditions. The effect of
temperature on activity was measured in oxidative and
reductive reactions at a temperature range 20–65 ꢁC
and at optimal pH values for both reaction directions,
respectively. To evaluate temperature stability of the
APDH, the enzyme was incubated without a substrate
at various temperatures in 20 mM Tris HCl, pH 7.2,
Acknowledgements
The present work was supported by the grant of Pro-
gram for Basic Research in Molecular and Cell Biology
from the Presidium of Russian Academy of Sciences
MBC RAS). Authors thank Prof. Michael L. Sinnott
for his helpful assistance in critically reviewing the
manuscript.
(
1 mM DTT. Aliquots of the reaction mixture were with-
drawn at 5 min intervals with subsequent cooling to
2
sured in the reductive reaction under standard assay
0 ꢁC, and the residual activity of the enzyme was mea-
References
+
conditions. NADP and NADPH were tested as co-fac-
1
2
. Assev, S.; Stig, S.; Scheie, A. A. Oral Microbiol. Immunol.
002, 17, 95–99.
. Sahni, P. S.; Gillespie, M. J.; Botto, R. W.; Otsuka, A. S.
Gen Dent. 2002, 50, 340–343.
3. Assev, S.; Waler, S. M.; Rolla, G. Acta Pathol. Microbiol.
Immunol. Scand. [B] 1983, 91, 261–265.
4. Wong, B.; Leeson, S.; Grindle, S.; Magee, B.; Brooks, E.;
Magee, P. T. J. Bacteriol. 1995, 177, 2971–2976.
. Onkenhout, W.; Groener, J. E.; Verhoeven, N. M.; Yin,
C.; Laan, L. A. Mol. Genet. Metab. 2002, 77, 80–85.
tors in the oxidative and reductive reactions under stan-
dard assay conditions and at optimal pH values (7.2 and
2
+
.5) found for NAD and NADH as co-factors,
8
respectively.
Double-reciprocal primary plots of velocity against
+
Ara-ol1-P and NAD were obtained by fitting the experi-
mental data to equations developed for a ternary com-
1
plex mechanism.
5
5
V max
v
i
¼
;
ð1Þ
þ
þ
Ara
-ol1-P
þ
NAD
Ara
-
ol1
-
P;NAD
þ
1
þ KM
=½NAD ꢁ þ KM =½Ara-ol1-Pꢁ þ KM
=ð½Ara-ol1-Pꢁ½NAD ꢁÞ
where v is the initial velocity, V
is the maximum
is the ternary complex
6. Honeyman, A. L.; Curtiss, R., III. Infect Immun. 1992, 60,
369–3375.
7. Jennings, D. H. Adv. Microb. Physiol. 1984, 25, 149–193.
. London, J.; Chace, N. M. Proc. Natl. Acad. Sci. U.S.A.
977, 74, 4296–4300.
i
max
þ
ol1 P;NAD
Ara
M
-
-
3
reductive velocity, K
þ
NAD
Ara
-
ol1
-
P
constant, KM
Menten constants for Ara-ol1-P and NAD at saturat-
, and KM
are the Michaelis–
+
8
1
+
ing NAD and Ara-ol1-P concentrations, respectively.
9. Hausman, S. Z.; London, J. J. Bacteriol. 1987, 169, 1651–
1655.
þ
Ara
-
ol1-P;NAD
The ternary complex constant, KM
lated from the abscissa of the primary-plot common
+
intersection point at fixed [NAD ] or at fixed [Ara-ol1-
was calcu-
1
0. Povelainen, M.; Eneyskaya, E. V.; Kulminskaya, A. A.;
Ivanen, D. R.; Kalkkinen, N.; Neustroev, K. N.; Miasni-
kov, A. N. Biochem. J. 2003, 371, 191–197.
8
P] according to Eq. 2.

546
N. V. Soroka et al. / Carbohydrate Research 340 (2005) 539–546
1
1. Suzuki, S.; Sugiyama, M.; Mihara, Y.; Hashiguchi, K.;
Yokozeki, K. Biosci. Biotechnol. Biochem. 2002, 66, 2614–
620.
2. Simpson, F. J.; Bhuyan, B. K. Can. J. Microbiol. 1962, 8,
63–672.
17. Neuberger, M. S.; Patterson, A. R. A.; Hartley, B. S.
Biochem. J. 1979, 183, 31–42.
18. Pettersson, G. Eur. J. Biochem. 1976, 69, 273–278.
19. SavelÕev, A. N.; Eneyskaya, E. V.; Isaeva-Ivanova, L. S.;
Shabalin, K. A.; Golubev, A. M.; Neustroev, K. N.
Glycoconj. J. 1997, 14, 897–905.
20. Deulovev, G. J. Chem. Soc. 1932, 2973–2975.
21. Laemmli, U. K. Nature 1970, 227, 680–685.
22. Lowry, O. H.; Rosenbrough, N. J.; Farr, A. L.; Randall,
R. J. J. Biol. Chem. 1951, 193, 265–275.
23. Miasnikov, A. N.; Salusj a¨ rvi, T.; Neustroev, K. Int. Appl.
WO 03/038105, 2003.
2
1
1
6
3. Barnwell, J. L.; Saunders, W. A.; Watson, R. W. Can. J.
Chem. 1955, 33, 711–715.
4. King, E. J. Biochem. J. 1932, 26, 292.
1
1
5. Levene, P. A.; Raymond, A. L. J. Biol. Chem. 1933, 102,
3
17.
6. Levene, P. A.; Raymond, A. L. J. Biol. Chem. 1933, 102,
31.
1
3