Purification and characterization of pyridoxine 5′-phosphate phosphatase from Sinorhizobium meliloti

Article abstract of DOI:10.1271/bbb.69.2277

Here we report the purification and biochemical characterization of a pyridoxine 5′-phosphate phosphatase involved in the biosynthesis of pyridoxine in Sinorhizobium meliloti. The phosphatase was localized in the cytoplasm and purified to electrophoretic homogeneity by a combination of EDTA/lysozyme treatment and five chromatography steps. Gel-filtration chromatography with Sephacryl S-200 and SDS/PAGE demonstrated that the protein was a monomer with a molecular size of approximately 29kDa. The protein required divalent metal ions for pyridoxine 5′-phosphate phosphatase activity, and specifically catalyzed the removal of Pi from pyridoxine and pyridoxal 5′-phosphates at physiological pH (about 7.5). It was inactive on pyridoxamine 5′-phosphate and other physiologically important phosphorylated compounds. The enzyme had the same Michaelis constant (K _m) of 385 μM for pyridoxine and pyridoxal 5′-phosphates, but its specific constant [maximum velocity (V_max)/K_m] was nearly 2.5 times higher for the former than for the latter.

Full text of DOI:10.1271/bbb.69.2277

Biosci. Biotechnol. Biochem., 69 (12), 2277–2284, 2005
Purification and Characterization of Pyridoxine 50-Phosphate Phosphatase
from Sinorhizobium meliloti
Masaaki TAZOE,y Keiko ICHIKAWA, and Tatsuo HOSHINO
,
*
**
*
Department of Applied Microbiology, Nippon Roche Research Center, Kamakura, Kanagawa 247-8530, Japan
Received April 15, 2005; Accepted July 29, 2005
Here we report the purification and biochemical
characterization of a pyridoxine 50-phosphate phospha-
tase involved in the biosynthesis of pyridoxine in
Sinorhizobium meliloti. The phosphatase was localized
in the cytoplasm and purified to electrophoretic homo-
geneity by a combination of EDTA/lysozyme treatment
and five chromatography steps. Gel-filtration chroma-
tography with Sephacryl S-200 and SDS/PAGE dem-
onstrated that the protein was a monomer with a
molecular size of approximately 29 kDa. The protein
required divalent metal ions for pyridoxine 50-phos-
phate phosphatase activity, and specifically catalyzed
the removal of Pi from pyridoxine and pyridoxal 50-
phosphates at physiological pH (about 7.5). It was
inactive on pyridoxamine 50-phosphate and other phys-
iologically important phosphorylated compounds. The
enzyme had the same Michaelis constant (Km) of 385 ꢀM
for pyridoxine and pyridoxal 50-phosphates, but its
specific constant [maximum velocity ðVmaxÞ=Km] was
nearly 2.5 times higher for the former than for the
latter.
1-deoxy-D-xylulose 5-phosphate. The former is synthe-
sized from D-erythrose 4-phosphate in three reaction
steps catalyzed by Epd, PdxB, and SerC (PdxF) proteins
in that order.4–7) The latter is formed from pyruvate and
D-glyceraldehyde 3-phosphate by 1-deoxy-D-xylulose 5-
phosphate synthase (Dxs).8) The two intermediates are
then combined by PdxA and PdxJ proteins to generate
the first vitamin B6 compound: pyridoxine 50-phosphate
(PNP).9) PNP is finally oxidized to the active form, PLP,
by a PNP/PMP oxidase (PdxH).10) PLP and PMP are
easily interconverted by ubiquitous transaminases.
Sinorhizobium (formerly known as Rhizobium) meli-
loti IFO 14782 produces large quantities of PN.11) Since
two compounds, 1-deoxy-D-xylulose (1-DX) and 4-
hydroxy-L-theronine (4-HT), have been found to act as
precursors of PN in a study of the biosynthetic pathway
12)
of vitamin B6 we examined the biosynthesis of PN
from 1-DX and 4-HT in vitro,13) and purified the
enzymes involved in its synthesis from the two
precursors (M. Tazoe, K. Ichikawa, and T. Hoshino,
unpublished results). BLAST searches of the five
purified enzymes required for the formation of PN from
the two compounds identified two putative xylulose
and homoserine kinases, 4-phosphohydroxy-L-threonine
dehydrogenase (PdxA), PNP synthase (PdxJ), and a
conserved hypothetical protein, based on analysis of
the chromosome sequence of S. meliloti strain 1021.14)
This suggests that PN is formed in the following
sequence of reactions (Fig. 2): Initially, 1-DX and 4-
HT are converted to 1-deoxy-D-xylulose 5-phosphate
and 4-phosphohydroxy-L-threonine by xylulose kinase15)
and homoserine kinase16,17) respectively. The latter
product is then oxidized by PdxA, followed by con-
densation with 1-deoxy-D-xylulose 5-phosphate by a
PdxJ protein to yield PNP, as it is in E. coli. The final
step requires a hypothetical protein that dephospho-
rylates PNP to PN and an enzyme with the required
Key words: phosphatase; pyridoxine 50-phosphate phos-
phatase; vitamin B6; pyridoxine; Sinorhi-
zobium (Rhizobium) meliloti
When pyridoxine (PN), which is also known as
***
vitamin B6 , and other compounds in this family are
taken up by living organisms, they are converted into
pyridoxal 50-phosphate (PLP) and pyridoxamine 50-
phosphate (PMP). These act as cofactors of vitamin B6
enzymes that are active in the metabolism of amino
acids1,2) and deoxysugars.3) Many microorganisms (bac-
teria, yeasts, fungi, and algae) and plants synthesize their
own vitamin B6. The biosynthetic pathway in Esche-
richia coli is well understood (Fig. 1). It contains two
key intermediates, 4-phosphohydroxy-L-threonine and
y
To whom correspondence should be addressed. Tel/Fax: +81-42-739-8682; E-mail: m.tazoe@lab.tamagawa.ac.jp
Present address: Mycology and Metabolic Diversity Research Center, Tamagawa University Research Institute, Machida, Tokyo 194-8610,
Japan
*
**
Present address: Research Planning and Coordination Department, Kamakura Research Laboratories, Chugai Pharmaceutical Co., Ltd., 200
Kajiwara, Kamakura, Kanagawa 247-8530, Japan
Abbreviations: 1-DX, 1-deoxy-D-xylulose; 4-HT, 4-hydroxy-L-threonine; p-NPP, p-nitrophenylphosphate; PLP, pyridoxal 50-phosphate; PMP,
pyridoxamine 50-phosphate; PN, pyridoxine; PNP, pyridoxine 50-phosphate
In this paper, the term vitamin B6 is used to denote the six compounds, pyridoxine, pyridoxal, pyridoxamine, and their 50-phosphate esters.
***

2278
M. TAZOE et al.
Fig. 1. Currently Proposed Biosynthetic Pathway for Vitamin B6 in Escherichia coli.
Materials and Methods
Materials. ATP, ADP, AMP, malate, and phospho-L-
serine were purchased from Wako Pure Chemical
(Osaka, Japan). PN, PLP, PMP, p-nitrophenylphosphate
(p-NPP), and most of the other biochemicals were
obtained from Sigma Chemical (St. Louis, MO.). PNP
was synthesized by reducing PLP with sodium borohy-
dride according to the method of Stock et al.18) Most of
the chromatographic media and prepacked columns
were purchased from Amersham Biosciences (Uppsala,
Sweden), and Ether Toyopearl was purchased from
Toyo Soda Manufacturing (Tokyo). All other chemicals
were of reagent grade and were purchased from
commercial sources.
Microorganisms, media, and cell growth. S. meliloti
IFO 14782 was cultured in a Pi-rich production medium
composed of 4% glucose, 2% polypeptone, 0.2% yeast
Fig. 2. Proposed Pathway for Pyridoxine from 1-Deoxy-D-xylulose
and 4-Hydroxy-L-threonine in S. meliloti.
.
extract, 0.05% MgSO4 7H2O, 0.05% MnSO4 5H2O,
and 0.001% FeSO4 7H2O (pH 6.8) at 28 C. The
.
ꢀ
.
properties, which is the subject of this paper. In
vitamin B6 biosynthesis in microorganisms, an enzyme
catalyzing dephosphorylation of PNP has not yet been
identified. With a view to understanding the dephos-
phorylation step in the biosynthesis of PN in S. meliloti
and our ultimate aim of gene engineering PN produc-
tion, it is important to characterize the enzyme catalyz-
ing the dephosphorylation of PNP. Here we show that
this enzyme is a cytoplasmic PNP/PLP-specific phos-
phatase that is active as a monomer at physiological pH.
Pi-limited chemically defined medium for the induction
of periplasmic marker proteins comprised the following:
1% glucose and L-arginine, glycine, L-histidine, L-
isoleucine, L-leucine, L-lysine, L-methionine, L-threo-
nine, L-valine, L-aspartic acid, L-glutamic acid, L-
phenylalanine, and L-tyrosine; pantothenic acid, inositol,
.
nicotinic acid, thiamine, and biotin; and MnSO4 5H2O,
.
.
MgSO4 7H2O, CaCl2 2H2O, KCl, and KH2PO4, which
was limited to 14 mg/l (pH 7.0).

Sinorhizobium Pyridoxine 50-Phosphate Phosphatase
2279
Fractionation of cellular proteins. S. meliloti IFO
14782 was grown in the production medium for 3 d.
After harvesting by centrifugation, the cells were
subcultured in the chemically defined medium and in
the production medium for an additional 6 h. Then they
were pelleted, washed twice with 0.85% NaCl, and
suspended in buffer A [30 mM Tris–HCl (pH 8.0) con-
taining 20% (w/v) sucrose and 1 mM EDTA]. Then
lysozyme (0.05%, w/v) was added at 23 ꢀC with slow
stirring for 20 min, and the suspensions were separated
into periplasmic proteins and cell pellets by centrifuga-
tion. The periplasmic fractions were dialyzed against
buffer B [10 mM Tris–HCl (pH 7.5) containing 15%
(w/v) sucrose, 1 mM dithiothreitol, and 0.1 mM phenyl-
methylsulfonyl fluoride], and alkaline phosphatase ac-
tivity was measured by recording the hydrolysis of p-
NPP.19) The activity from the Pi-limited culture was
1.56 mmol/min/mg protein, and that from the Pi-rich
culture was about 36-fold lower. When the cytoplasmic
marker enzyme malate dehydrogenase was assayed by
monitoring the rate of oxidation of NADH (A340),20) no
activity was found in the periplasmic fractions of either
culture.
dephosphorylase. The cell cake (59.5 g) formed by
centrifugation of a 3-d culture broth (3.4-liter) in
production medium was washed twice with 300 ml of
0.85% NaCl and resuspended in 340 ml of buffer A. The
cellular proteins were divided into periplasmic and
cytoplasmic fractions using the procedure described
above.
Step 2. Q Sepharose column chromatography (I). The
lysate of the cytoplasmic fraction was centrifuged, and
the supernatant was dialyzed against buffer B and
applied to a Q Sepharose column (4:4 ꢁ 17 cm). It was
developed with 600 ml of 400 mM KCl at a flow rate of
0.57 ml/min.
Step 3. Q Sepharose column rechromatography (II).
The active fractions, which were dialyzed against
buffer B, were then applied to a Q Sepharose column
(4:4 ꢁ 12:5 cm). Elution was carried out at a flow rate of
0.72 ml/min with a linear gradient of KCl (0–400 mM) in
buffer B (total volume, 2,200 ml).
Step 4. Ether Toyopearl column chromatography. The
sample was dialyzed against buffer B and adjusted to
1.3 M (NH4)2SO4, and then applied to an Ether Toyo-
pearl column (2:5 ꢁ 15 cm) previously equilibrated with
buffer B containing 1.3 M (NH4)2SO4. Elution was
carried out at a flow rate of 0.7 ml/min with a linear
gradient of (NH4)2SO4 (1.3–0.5 M) in buffer B (total
volume, 1,000 ml).
Step 5. Resource ISO column chromatography. The
sample was dialyzed against buffer B and applied to a
Resource ISO column (6 ml) equilibrated with buffer B
containing 1.2 M (NH4)2SO4. Elution was carried out at
a flow rate of 0.8 ml/min with a linear gradient of
(NH4)2SO4 (1.2–0.5 M) in buffer B (total volume,
120 ml).
Step 6. Sephacryl S-200 column chromatography. The
sample was dialyzed against buffer C [50 mM Tris–HCl
(pH 7.5), 15% (w/v) sucrose, and 150 mM KCl]. Finally,
the dialysate was concentrated to 300 ml by ultrafiltra-
tion. The solution was applied to a Sephacryl S-200
column (1:6 ꢁ 60 cm), which was developed with
buffer C.
To obtain the cytoplasmic proteins, pelleted cells
were resuspended in buffer B and passed through a
French pressure cell, and the suspensions were centri-
fuged at 37;500 ꢁ g for 1 h. Cytoplasmic extracts of both
cultures had substantial malate dehydrogenase activity
(0.55 and 0.58 mmol/min/mg protein in Pi-limited and
Pi-rich cultures respectively).
Enzyme assays. Dephosphorylase activity was initial-
ly assayed with NADP, NAD, ATP, MnCl2, and MgCl2
as cofactors. A mixture containing 0.38 mM NADP,
0.38 mM NAD, 5 mM ATP, 8.4 mM MnCl2, 32 mM
MgCl2, 50 mM Tris–HCl (pH 7.0), and enzyme solution
was pre-incubated for 1 min at 28 ꢀC. The reaction in a
total volume of 125 ml was then started by adding 40 mM
PNP, and was incubated at the same temperature for
30 min and then cooled in ice water. Once a cofactor
necessary for the enzyme activity and its optimum
concentration had been determined, the mixture con-
tained 1 mM MnCl2 as a cofactor. Enzyme activity was
calculated by quantifying the PN produced, or the Pi
released. First, the amount of PN was measured using
the turbidity method with Saccharomyces carlsbergensis
ATCC 9080, whose growth is supported with PN but not
with PNP in vitamin B6 assay medium, as described
previously.11) One unit of PNP dephosphorylase (or
phosphatase) activity produced 1 mmol of PN in 30 min.
Second, Pi was determined by the malachite-green
method of Geladopoulos et al.21) This method was used
to determine substrate specificity, the Michaelis constant
(Km), and the maximum velocity (Vmax).
Molecular size estimation and amino-terminal se-
quence analysis. The molecular size of the native
dephosphorylase was calculated using the Sephacryl S-
200 column calibrated with molecular mass standards
(Bio-Rad, Richmond, CA). SDS/PAGE was carried out
according to the method of Laemmli22) with 12%
acrylamide gels. Protein bands on the gels were
visualized with Coomassie brilliant blue R-250. The
apparent molecular size of the protein was estimated by
comparison with molecular mass standards (Amersham
Biosciences, Uppsala, Sweden). A purified protein was
identified on SDS/PAGE gels by the method of
Matsudaira,23) and the amino-terminal sequence was
determined by an amino acid sequence system.
Purification of PNP dephosphorylase. Unless other-
wise stated, all operations were performed at 0–4 ꢀC.
Step 1. Preparation of cell extract containing PNP
Removing divalent metal ions to test for divalent

2280
M. TAZOE et al.
Table 1. Localization of PNP Dephosphorylase
PNP dephosphorylase
(nmol/min/mg protein)
Alkaline phosphatase
(mmol/min/mg protein)
Malate dehydrogenase
(mmol/min/mg protein)
Cellular proteins
(Cells from Pi-limited culture)
Periplasmic proteins
Cytoplasmic proteins
2
137
1.56
0.07
0.01
0.55
(Cells from Pi-rich culture)
Periplasmic proteins
Cytoplasmic proteins
1
138
0.04
0.11
0.03
0.58
Reaction mixtures contained 40 mM PNP, 0.38 mM NADP, 0.38 mM NAD, 5 mM ATP, 8.4 mM MnCl2, 32 mM MgCl2, 50 mM Tris–HCl (pH 7.0), and periplasmic
proteins (3.23 mg) or cytoplasmic proteins (3.19 mg) in 125 ml.
metal requirements. Purified enzyme solution (1 mg/ml)
was dialyzed twice against buffer B containing 1 mM
EDTA for 15 h, and subsequently twice against buffer B
for 15 h to remove the remaining EDTA.
extracts of Pi-limited and Pi-rich cultures were divided
into periplasmic and cytoplasmic fractions after EDTA/
lysozyme treatment (see ‘‘Materials and Methods’’).
Alkaline phosphatase and malate dehydrogenase were
assayed as standard periplasmic and cytoplasmic marker
proteins respectively. The cytoplasmic fractions of both
the Pi-limited and the Pi-rich cultures had high dephos-
phorylase activity. By contrast, there was no activity in
the periplasmic fractions of either culture (Table 1). It
appears likely that the PNP dephosphorylase involved in
the de novo synthesis of PN is cytoplasmic.
Identification of the reaction product. A reaction
mixture (5 ml) consisting of 640 mM PNP, 1 mM MnCl2,
50 mM Tris–HCl (pH 7.5), and enzyme solution was
incubated at 28 ꢀC for 1 h and heated for 3 min in boiling
water, and then the denaturated proteins were removed
by centrifugation. The supernatant was applied to an
Amberlite CG-120 column (1:6 ꢁ 11 cm; Rohm and
Haas Co., Philadelphia, PA), which was washed with
water and developed with a 5% (w/v) ammonium
hydroxide solution. The eluate was concentrated under
reduced pressure, and the residue was dissolved in a
small volume of methanol and analyzed by high-
pressure liquid chromatography.11) The product was
identified as PN by reference to a pure sample.
Purification of PNP dephosphorylase
Protein samples for purification of PNP dephospho-
rylase were prepared initially by rupture of whole cells
from Pi-rich cultures, but the enzyme was difficult to
purify because of the high concentrations of other
proteins and its instability. Before the application of
chromatographic procedures, periplasmic proteins were
removed by treatment with EDTA/lysozyme and cen-
trifugation. PNP dephosphorylase was then purified
to apparent homogeneity from the cytoplasm in five
chromatographic steps. The purification procedure,
starting with the cytoplasmic fraction of S. meliloti, is
summarized in Table 2. The protein was purified 2,090-
fold with a yield of 18% based on the PNP dephospho-
rylase activity in cytoplasmic extracts. Since it was
enriched about two-fold with a yield of about 80% by
isolating the cytoplasmic proteins, the overall purifica-
tion was approximately 4,180-fold with a yield of
14.4%.
Results and Discussion
A protein involved in the last step of PN synthesis in
S. meliloti
PN was formed when 1-DX and 4-HT were incubated
with extracts of S. meliloti IFO 14782 in buffer with
15% (w/v) sucrose, but not when they were incubated
with extracts made in the absence of sucrose.13) PNP
was identified by high-pressure liquid chromatography
in the latter mixture; when treated with potato acid
phosphatase (Boehringer Mannheim, Munich, Germa-
ny), it supported the growth of a vitamin B6 indicator,
S. carlsbergensis ATCC 9080, which requires phos-
phate-free vitamin B6, PN, pyridoxal, or pyridoxamine
for growth in vitamin B6 assay medium (data not
shown). These observations suggest that sucrose extracts
of S. meliloti IFO 14782 generate PN by dephospho-
rylation of PNP and that the enzyme responsible is
inactivated by extraction in the absence of sucrose.
Molecular size estimation of PNP dephosphorylase
and amino-terminal amino acid sequence
In the final gel-filtration step described in the previous
section, PNP dephosphorylase activity eluted as a single
peak between chicken ovalbumin (44 kDa) and equine
myoglobin (17 kDa) (to the left in Fig. 3). The position
of the peak corresponded to a molecular size of 29 kDa,
which was similar to that seen on SDS/PAGE gels (to
the right in Fig. 3). The protein is thus a monomer with a
molecular size of approximately 29 kDa.
Localization of a PNP dephosphorylase
S. meliloti IFO 14782 accumulates large amounts of
extracellular PN, and the result presented above suggests
that a dephosphorylase participates in the final step of
PN synthesis. To determine the location of this enzyme,
The amino-terminal amino acid sequence was deter-
mined to be M(amino-terminal)–K–K–L–D–R–M–P–T–
H. The NCBI databases were predicted to be a

Sinorhizobium Pyridoxine 50-Phosphate Phosphatase
2281
Table 2. Purification of PNP Dephosphorylase Starting from Cytoplasmic Extracts
Total protein
(mg)
Total activity
(unit)
Specific activity
(unit/mg protein)
Purification
(-fold)
Yield
(%)
Steps
Cytoplasmic extracts
Q Sepharose (I)
Q Sepharose (II)
Ether Toyopearl
Resource ISO
8,570
5,700
316
4.1
3.4
2.5
1.6
1.2
0.74
0.00048
0.0006
0.0079
0.022
0.29
1.0
1.3
100
83
62
39
28
18
16.5
45.8
74
4.2
0.71
605
2,090
Sephacryl S-200
1.0
PNP dephosphorylase was assayed in mixtures containing (in 125 ml): 40 mM PNP, 0.38 mM NADP, 0.38 mM NAD, 5 mM ATP, 8.4 mM MnCl2, 32 mM MgCl2, 50 mM
Tris–HCl (pH 7.0), and desalted fraction in each chromatography.
Fig. 3. Molecular Size Estimation of PNP Dephosphorylase by Sephacryl S-200 Chromatography (left) and SDS–PAGE (right).
Left figure: molecular mass standard markers correspond to (1) thyroglobulin (669 kDa), (2) bovine gamma globulin (158 kDa), (3) chicken
ovalbumin (44 kDa), and (4) equnine myoglobin (17 kDa). Right figure: lanes correspond to (1) molecular mass markers: A, phosphorylase B
(92.5 kDa); B, bovine serum albumin (66.2 kDa); C, ovalbumin (45 kDa); D, carbonic anhydrase (31 kDa); E, soybean trypsin inhibitor
(21.5 kDa); and F, ꢀ-lactalbumin (14.4 kDa) and (2) 2.7 mg of purified PNP dephosphorylase.
conserved hypothetical protein consisting of 258 amino
acids [Sinorhizobium meliloti strain 1021] with 100%
identity from this sequence.14) A further search was
conducted using the 258 amino acid-sequence. It
resulted in predictions of putative hydrolases (many
bacteria), phosphoglycolate phosphatase (Brucella me-
litenis), and putative haloacid dehalogenase superfamily
hydrolase (many bacteria) with 40–70% identities.
These results suggest that our enzyme is yet unspecified.
(65%), Sn2þ (11%), and Ni2þ (7%). The enzyme should
thus be classified either as a phosphohydrolase or as a
phosphotransferase, depending on its reaction mecha-
nism. In the former case, Pi is released as PN is formed.
Hence we used the malachite-green method to determine
whether Pi is released upon incubation of PNP with a
purified protein. Pi was indeed released, indicating that
the reaction is hydrolytic. The enzyme is therefore likely
to be a phosphatase.
Many phosphatases have divalent metal ions,24) which
play an important role in protein structures to maintain
activity.19) An alkaline phosphatase of R. leguminosa-
rum, which incorporates divalent metal ions and forms
dimers, releases the metal ions upon EDTA treatment
and breaks down to inactive fragments, and its activity is
restored completely by adding an appropriate concen-
tration of divalent metal ions. Our PNP phosphatase was
Characterization of PNP dephosphorylase
PNP dephosphorylase activity did not require NAD,
NADP, or ATP, which are essential for PN formation
from 1-DX and 4-HT, but it did require either Mn2þ
or Mg2þ (Table 3). Activation by MnCl2 or MgCl2
followed saturation kinetics (Fig. 4). Activity was high-
est with Mn2þ (100%), followed by Mg2þ (88%), Co2þ

2282
M. TAZOE et al.
Table 3. Cofactor Requirements of PNP Dephosphorylase
PNP dephosphorylase
activity (unit/ml)
Cofactors present
Complete
minus NAD
0.32
0.32
0.31
0.32
0.32
0.30
0.32
0
minus NADP
minus ATP
minus NAD, NADP, ATP
minus MnCl2
minus MgCl2
minus MgCl2 and MnCl2
minus NAD, NADP, ATP and MnCl2
minus NAD, NADP, ATP and MgCl2
minus all
0.32
0.33
0
Mixtures for the complete system contained 40 mM PNP as substrate,
0.38 mM NADP, 0.38 mM NAD, 5 mM ATP, 8.4 mM MnCl2, and 32 mM
MgCl2 as cofactors, 50 mM Tris–HCl (pH 7.0), and a purified protein
(1.53 mg) in 125 ml.
Fig. 5. Effect of pH on PNP Phosphatase Activity.
Activity was measured in 50 mM Tris–maleate buffer (pH 5.5–
7.5) ( ) and 50 mM Tris–HCl buffer (pH 7.5–9.5) ( ) containing
40 mM PNP, 1 mM MnCl2, and a purified protein (1.53 mg) in 125 ml.
Table 4. Kinetic Parameters of PNP Phosphatase with Potential
Substrates
Vmax or activity
(mmol/min/mg
protein)
Km
Vmax=Km
(min/mg)
Compound
(mM)
Pyridoxine 50-phosphate
Pyridoxal 50-phosphate
Pyridoxamine 50-phosphate
D-Glucose 6-phosphate
D-(ꢂ)-3-Phosphoglyceric acid
2-Phosphoglycolic acid
ATP
385
385
105a
42a
2,180
870
<0:73b
<0:73b
<0:73b
<0:73b
<0:73b
<0:73b
<0:73b
<0:73b
ADP
AMP
Phospho-L-serine
Fig. 4. Dependence of PNP Dephosphorylase Activity on MnCl2 and
MgCl2.
The reaction was conducted in mixtures containing (in 125 ml) the indicated
compounds, 1 mM MnCl2, 50 mM Tris–HCl (pH 7.0), and a purified protein
(1.53 mg).
Reaction mixtures contained 40 mM PNP, MnCl2, or MgCl2,
50 mM Tris–HCl (pH 7.0), and a purified protein (1.53 mg) in 125 ml.
aVmax values were calculated using the indicated compounds at concen-
trations of 40, 80, 160, 320, and 640 mM.
Metal salts were tested at concentrations of 0.625–2 mM MnCl2 (
)
and 0.625–2 mM MgCl2 ).
(
bActivity with 640 mM of the indicated compounds.
purified in a buffer without metal ions, and its molecular
size was estimated using gel-filtration chromatography
in a buffer without metal ions. To determine the actual
molecular size of the active enzyme, we performed
chromatography on a Sephacryl S-200 column in a
buffer containing 1 mM MnCl2. The protein eluted at the
same position as that purified in buffer without a
divalent metal ion (to the left in Fig. 3). From these
results, we infer that this enzyme is likely to act as a
monomer. Our PNP/PLP phosphatase differs from the
human erythrocyte vitamin B6 phosphate phosphatase,
which is a dimer of 64 kDa.24)
Properties of PNP phosphatase
Next we examined the pH dependence of PNP
phosphatase activity. The optimal pH was about 7.5
(Fig. 5). The enzyme catalyzed the removal of Pi from
PNP and PLP, but not from PMP or other physiolog-
ically important phosphorylated compounds, such as
phosphosugars, nucleotides, and a phosphoamino acid
(Table 4). Phosphoglycolic acid, a substrate for phos-
phoglycolate phosphatase predicted in a NCBI database
search, was not a substrate for our phosphatase. This
substrate specificity indicates that it is a PNP/PLP

Sinorhizobium Pyridoxine 50-Phosphate Phosphatase
2283
Table 5. Effect of Various Compounds on PNP Phosphatase Activity
compounds
None
Relative activity (%)
100
21
Sodium molybdate
K-PO4
44
54
N-Ethylmaleimide
Idoacetate
NaF
48
76
Levamisole
L-Phenylalanine
L-(+)-Tartarate
74
100
100
Reaction mixtures contained 40 mM PNP, 1 mM MnCl2, 50 mM Tris–HCl
(pH 7.0), a purified protein (1.53 mg), and 5 mM of each compound in 125 ml.
finding that the highly specific PNP/PLP phosphatase of
S. meliloti is localized in the cytoplasm rather than in
the periplasm, where the well-known non-specific
alkaline and acid phosphatases are found. This is the
first report of a microbial phosphatase with this unique
characteristic. There have been many previous studies of
the regulation of PLP levels by interactions between
PLP phosphatase and PL kinase in mammals, which can
not synthesize their own vitamin B6.24,26–28) In Sinorhi-
zobium, which possesses a complete vitamin B6 bio-
synthetic pathway, PNP is the first vitamin B6 com-
pound synthesized de novo and represents the point of
divergence of the pathways leading to PN and PLP. PNP
phosphatase might play a major role together with a
PNP oxidase in the control of intracellular PLP levels.
In conclusion, we believe the results obtained in this
study will provide information that will help develop
approaches to genetic engineering of the PNP phospha-
tase of S. meliloti in order to produce PN.
Fig. 6. Kinetics of Dephosphorylation of Pyridoxine 50-Phosphate
and Pyridoxal 50-Phosphate.
The reaction mixtures contained substrate, 1 mM MnCl2, 50 mM
Tris–HCl (pH 7.0), and a purified protein (1.53 mg) in 125 ml.
Substrate concentrations were 40–2,400 mM pyridoxine 50-phosphate
(
) and 40–1,800 mM pyridoxal 50-phosphate ( ).
phosphatase. The velocities with PNP and PLP followed
Michaelis–Menten kinetics (Fig. 6). The Km and Vmax
(mM, mmol/min/mg protein) values for PNP and PLP
were 385 and 105, and 385 and 42, respectively
(Table 4). Thus it had the same Km for PNP and PLP,
but had a Vmax that was nearly 2.5-times higher for PNP
than for PLP; therefore the specific constant (Vmax=Km)
for PNP was higher than that for PLP. The human
erythrocyte vitamin B6 phosphate-specific phosphatase
shows substrate specificity similar to our enzyme, but it
has a lower Km value for PLP than for PNP.24)
References
1) Bender, D. A., The role of vitamin B6 in amino acid
metabolism. In ‘‘Amino Acid Metabolism’’, John Wiley
and Sons, New York, pp. 75–94 (1985).
To differentiate our phosphatase from human eryth-
rocyte phosphatase, the effects of inhibitors tested by
Fonda24) on our enzyme were examined in the presence
of 1 mM MnCl2, which gave a maximum activity for
PNP phosphatase (Fig. 4). EDTA was an effective
inhibitor (I50 ¼ 16 mM). From these results, we infer
that the Mn ion is not tightly bound to this enzyme.
Sodium molybdate was a poor inhibitor. Potassium
phosphate, N-ethylmaleimide, idoacetate, sodium fluo-
ride, and levamisole were even less inhibitory, and L-
phenylalanine and L-(+)-tartarate showed no inhibition
to concentration of 5 mM (Table 5). These features of
inhibition to our enzyme were similar to those of the
human phosphatase except for sulfhydryl reagents, N-
ethylmaleimide, and idoacetate, which are very sensitive
inhibitors of the latter phosphatase.24)
2) Dakshinamurti, K., ‘‘Annals of the New York Academy
of Sciences, Vol. 585, Vitamin B6’’, New York Acad-
emy of Sciences, New York, pp. 128–144 (1990).
3) Burns, K. D., Pieper, P. A., Liu, H.-W., and Stankovich,
M. T., Studies of the redox properties of CDP-6-deoxy-
L-threo-D-glycero-4-hexulose 3-dehydrase (E1) and
CDP-6-deoxy-L-threo-D-glycero-4-hexulose 3-dehydrase
reductase (E3), two important enzymes involved in the
biosynthesis of ascarylose. Biochemistry, 35, 7879–7889
(1996).
4) Dempsey, W. B., Synthesis of pyridoxal phosphate.
In ‘‘Escherichia coli and Salmonella typhimurium’’,
‘‘Cellular and Molecular Biology’’, eds. Neidhardt,
F. C., Ingraham, J. L., Low, K. B., Magasanik, B.,
Schaechter, M., and Umbarger, H. E., American Society
for Microbiology, Washington D.C., pp. 539–543 (1987).
5) Dreke, C., Notheis, C., Hansen, U., Leistner, E.,
Hemscheidt, T., Hill, R. E., and Spenser, I. D., Growth
response to 4-hydroxy-L-threonine of Escherichia coli
mutants blocked in vitamin B6 biosynthesis. FEBS Lett.,
318, 125–128 (1993).
There has been one previous report of an acidic
phosphatase specific for PLP in an anaerobic photo-
synthetic bacterium, but the enzyme was periplasmic.25)
An interesting aspect of our present work was the

2284
M. TAZOE et al.
6) Lam, H.-M., and Winkler, M. E., Metabolic relationships
derived antimetabolites with the enzymes of the threo-
nine biosynthetic pathway of Escherichia coli. J. Biol.
Chem., 259, 15331–15339 (1984).
between pyridoxine (vitamin B6) and serine biosynthesis
in Escherichia coli K-12. J. Bacteriol., 172, 6518–6528
(1990).
17) Farrington, G. K., Kumar, A., Shames, S. L.,
Ewaskiewicz, J. I., Ash, D. E., and Wedle, F. C.,
Threonine synthase of Escherichia coli: inhibition by
classical and slow-binding analogues of homoserine
phosphate. Arch. Biochem. Biophys., 307, 165–174
(1993).
18) Stock, A., Ortanderl, F., and Pfleiderer, G., Darstellung
von radioakiv markiertem pyridoxal 50-phosphat. Bio-
chem. Z., 344, 353–360 (1966).
19) Glenn, A. R., and Dilworth, M. J., The effect of metal
ions on the alkaline phosphatase of Rhizobium legumi-
nosarum. Arch. Microbiol., 126, 251–256 (1980).
20) Deng, S., Summers, M. L., Kahn, M. L., and McDermott,
T. R., Cloning and characterization of a Rhizobium
meliloti nonspecific acid phosphatase. Arch. Microbiol.,
170, 18–26 (1998).
21) Geladopoulos, T. P., Sotiroudis, T. G., and
Evangelopoulos, A. E., A malachite green colorimetric
assay for protein phosphatase activity. Anal. Biochem.,
192, 112–116 (1991).
7) Lam, H.-M., and Winkler, M. E., Characterization of
the complex pdxH-tyrS operon of Escherichia coli K-12
and pleiotropic phenotypes caused by pdxH insertion
mutants. J. Bacteriol., 174, 6033–6045 (1992).
8) Hill, R. E., Sayer, B. G., and Spenser, I. D., Biosynthesis
of vitamin B6: incorporation of D-1-deoxyxylulose.
J. Am. Chem. Soc., 111, 1916–1917 (1989).
9) Laber, B., Maurer, W., Scharf, S., Stepusin, K., and
Schmidt, F. S., Vitamin B6 biosynthesis: formation of
pyridoxine 50-phosphate from 4-(phosphohydroxy)-L-
threonine and 1-deoxy-D-xylulose 5-phosphate by PdxA
and PdxJ protein. FEBS Lett., 449, 45–48 (1999).
10) Zhao, G., and Winkler, M. E., Kinetic limitation and
cellular amount of pyridoxine (pyridoxamine) 50-phos-
phate oxidase of Escherichia coli K-12. J. Bacteriol.,
177, 883–891 (1995).
11) Tazoe, M., Ichikawa, K., and Hoshino, T., Production of
vitamin B6 in Rhizobium. Biosci. Biotechnol. Biochem.,
63, 1378–1382 (1999).
12) Tazoe, M., Ichikawa, K., and Hoshino, T., Biosynthesis
of vitamin B6 in Rhizobium. J. Biol. Chem., 275, 11300–
11305 (2000).
22) Laemmli, U. K., Cleavage of structural proteins during
the assembly of the band of bacteriophage T4. Nature,
227, 680–685 (1970).
13) Tazoe, M., Ichikawa, K., and Hoshino, T., Biosynthesis
of vitamin B6 in Rhizobium: in vitro synthesis of
pyridoxine from 1-deoxy-D-xylulose and 4-hydroxy-L-
threonine. Biosci. Biotechnol. Biochem., 66, 934–936
(2002).
14) Capela, D., Barloy-Hubler, F., Gouzy, J., Bothe, G.,
Ampe, F., Batut, J., Boistard, P., Becker, A., Boutry, M.,
Cadieu, E., Dreano, S., Gloux, S., Godrie, T., Goffeau,
A., Kahn, D., Kiss, E., Lelaure, V., Masuy, D., Pohl, T.,
Portetelle, D., Puehler, A., Purnelle, B., Ramsperger, U.,
Renard, C., Thebault, P., Vandenbol, M., Weidner, S.,
and Galibert, F., Analysis of the chromosome sequence
of the legume symbiont Sinorhizobium meliloti strain
1021. Proc. Natl. Acad. Sci. U.S.A., 98, 9877–9882
(2001).
23) Matsudaira, P., Sequence from picomole quantities of
proteins electroblotted onto polyvinylidene difluoride
membranes. J. Biol. Chem., 232, 10035–10038 (1987).
24) Fonda, M. L., Purification and characterization of
vitamin B6-phosphate phosphatase from human erythro-
cytes. J. Biol. Chem., 267, 15978–15983 (1992).
25) Doi, M., and Shioi, Y., A novel phosphatase specific for
pyridoxal 50-phosphate in an aerobic photosynthetic
bacterium, Erythrobacter sp. OCh 114. FEMS Micro-
biol. Lett., 55, 309–313 (1988).
26) Gao, G. J., and Fonda, M. L., Identification of an
essential cysteine residue in pyridoxal phosphatase from
human erythrocytes. J. Biol. Chem., 269, 8234–8239
(1994).
27) Gao, G. J., and Fonda, M. L., Kinetic analysis and
chemical modification of vitamin B6 phosphatase from
human erythrocytes. J. Biol. Chem., 269, 7163–7168
(1994).
15) Wungsintaweekul, J., Herz, S., Hecht, S., Wisenreich,
W., Feicht, R., Rordich, F., Bacher, A., and Zenk, M. H.,
Phosphorylation of 1-deoxy-D-xylulose by D-xyluloki-
nase of Escherichia coli. Eur. J. Biochem., 268, 310–316
(2001).
28) Jang, Y. M., Kim, D. W., Kang, T.-C., Won, M. H.,
Baek, N.-I., Moon, B. J., Choi, S. Y., and Kwon, O.-S.,
Human pyridoxal phosphatase. J. Biol. Chem., 278,
50040–50046 (2003).
16) Shames, S. L., Ash, D. E., Wedler, F. C., and
Villafranca, J. J., Interaction of aspartate and aspartate-