Cloning, expression, purification, cofactor requirements, and steady state kinetics of phosphoketolase-2 from Lactobacillus plantarum

Article abstract of DOI:10.1016/j.bioorg.2008.03.002

The genes xpk1 and xpk2(Δ1-21) encoding phosphoketolase-1 and (Δ1-7)-truncated phosphoketolase-2 have been cloned from Lactobacillus plantarum and expressed in Escherichia coli. Both gene-products display phosphoketolase activity on fructose-6-phosphate in extracts. A N-terminal His-tag construct of xpk2(Δ1-21) was also expressed in E. coli and produced active His-tagged (Δ1-7)-truncated phosphoketolase-2 (hereafter phosphoketolase-2). Phosphoketolase-2 is activated by thiamine pyrophosphate (TPP) and the divalent metal ions Mg²⁺, Mn²⁺, or Ca²⁺. Kinetic analysis and data from the literature indicate the activators are MgTPP, MnTPP, or CaTPP, and these species activate by an ordered equilibrium binding pathway, with Me²⁺TPP binding first and then fructose-6-phosphate. Phosphoketolase-2 accepts either fructose-6-phosphate or xylulose-5-phosphate as substrates, together with inorganic phosphate, to produce acetyl phosphate and either erythrose-4-phosphate or glyceraldehyde-3-phosphate, respectively. Steady state kinetic analysis of acetyl phosphate formation with either substrate indicates a ping pong kinetic mechanism. Product inhibition patterns with erythrose-4-phosphate indicate that an intermediate in the ping pong mechanism is formed irreversibly. Background mechanistic information indicates that this intermediate is 2-acetyl-TPP. The irreversibility of 2-acetyl-TPP formation might explain the overall irreversibility of the reaction of phosphoketolase-2.

Full text of DOI:10.1016/j.bioorg.2008.03.002

Bioorganic Chemistry 36 (2008) 121–127
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Cloning, expression, purification, cofactor requirements, and steady state kinetics
of phosphoketolase-2 from Lactobacillus plantarum
*
Alejandro Yevenes, Perry A. Frey
Department of Biochemistry, University of Wisconsin–Madison, 1710 University Avenue, Madison, WI 53726, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 19 November 2007
Available online 21 April 2008
The genes xpk1 and xpk2(D1–21) encoding phosphoketolase-1 and (D1–7)-truncated phosphoketolase-2
have been cloned from Lactobacillus plantarum and expressed in Escherichia coli. Both gene-products
display phosphoketolase activity on fructose-6-phosphate in extracts. A N-terminal His-tag construct
of xpk2(D1–21) was also expressed in E. coli and produced active His-tagged (D1–7)-truncated phospho-
ketolase-2 (hereafter phosphoketolase-2). Phosphoketolase-2 is activated by thiamine pyrophosphate
(TPP) and the divalent metal ions Mg2+, Mn2+, or Ca2+. Kinetic analysis and data from the literature indi-
cate the activators are MgTPP, MnTPP, or CaTPP, and these species activate by an ordered equilibrium
binding pathway, with Me2+TPP binding first and then fructose-6-phosphate. Phosphoketolase-2 accepts
either fructose-6-phosphate or xylulose-5-phosphate as substrates, together with inorganic phosphate,
to produce acetyl phosphate and either erythrose-4-phosphate or glyceraldehyde-3-phosphate,
respectively. Steady state kinetic analysis of acetyl phosphate formation with either substrate indicates
a ping pong kinetic mechanism. Product inhibition patterns with erythrose-4-phosphate indicate that
an intermediate in the ping pong mechanism is formed irreversibly. Background mechanistic information
indicates that this intermediate is 2-acetyl-TPP. The irreversibility of 2-acetyl-TPP formation might
explain the overall irreversibility of the reaction of phosphoketolase-2.
Keywords:
Thiamine pyrophosphate
Phosphoketolase
Acetyl-thiamine pyrophosphate
Kinetics of phosphoketolase
Cloning phosphoketolase
Quaternary structure of phosphoketolase
Ordered equilibrium binding of metal-
thiamine pyrophosphate to
phosphoketolase
Ping pong kinetics of phosphoketolase
Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction
acetate kinase or as a source of acetyl-CoA by the action of phospho-
transacetylase. GAP produced from Xu5P by phosphoketolase is a
Phosphoketolases play significant roles in energy metabolism
by catalyzing the formation of acetyl phosphate (AcP)1 and ery-
throse-4-phosphate (E4P) from fructose-6-phosphate (F6P), an inter-
mediate in glycolysis, or of AcP and D-glyceraldehyde-3-phosphate
(GAP) from xylulose-5-phosphate (Xu5P), an intermediate in the
pentose phosphate pathway, according to Eqs. (1) and (2) [1]. AcP
is a high energy molecule that can serve as a source of ATP by the
action of
glycolytic intermediate.
Phosphoketolases are key enzymes of the phosphoketolase
pathway of obligatory and facultative heterofermentative lactic
acid bacteria and the F6P shunt of bifidobacteria [2]. Classification
of phosphoketolases follows substrate preferences, EC 4.1.2.9 when
specificity is for Xu5P, and EC 4.2.1.22 when both Xu5P and F6P are
accepted [3]. Phosphoketolase-2 from Lactobacillus plantarum is a
fructose-6-phosphate phosphoketolase that accepts Xu5P and so
is classified as EC 4.1.2.22.
Phosphoketolases are well known but little studied members of
the family of thiamine pyrophosphate (TPP)-dependent enzymes.
A limited literature on phosphoketolases describes activities in
various microbial and eukaryotic sources, and limited information
about molecular properties in several species is available [4–9].
Mechanistic information on the action of phosphoketolases is
sparse, but the outlines of a general reaction mechanism appear
to involve the transient formation of 2-(a,b-dihydroxyethylidene-
TPP (DHETPP) and 2-acetyl-thiamine pyrophosphate (AcTPP) as
intermediates [1,10–12]. As a TPP-dependent enzyme that pre-
sumably generates covalent substrate–TPP intermediates, the reac-
tion of phosphoketolase can be expected to follow ping pong
steady state kinetics, but no kinetic analysis is currently available.
F6P þ Pi ! E4P þ AcP þ H2O
Xu5P þ Pi ! d-glyceraldehyde-3-P þ AcP þ H2O
ð1Þ
ð2Þ
* Corresponding author. Fax: +1 608 265 2904.
E-mail address: frey@biochem.wisc.edu (P.A. Frey).
1
Abbreviations used: AcP, acetyl phosphate; AcTPP, 2-acetyl-thiamine pyrophos-
phate; ATP, adenosine 50-triphosphate; DEAE, diethylaminoethyl; DHETPP, 2-(a,b-
dihydroxyethylidene)-thiamine pyrophosphate; E4P, erythrose-4-phosphate; F6P,
fructose-6-phosphate; GAP, glyceraldehyde-3-phosphate; HETPP, 2-hydroxyethylid-
ene-thiamine pyrophosphate; IPTG, isopropyl-b-D-thiogalactoside; MES, 2-morpholi-
noethane sulfonic acid; MOPS, 3-(N-morpholino)propane sulfonic acid; NAD+,
nicotinamide adenine dinucleotide; NADH, reduced NAD+; PMSF, phenylmethane
sulfonylfluoride; Pi, phosphate ion; SDS–PAGE, sodium dodecylsulfonate–polyacryl-
amide gel electrophoresis; TPP, thiamine pyrophosphate.
0045-2068/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.bioorg.2008.03.002

122
A. Yevenes, P.A. Frey / Bioorganic Chemistry 36 (2008) 121–127
There is reason to expect at least one step in the hypothetical
mechanism, the formation of AcTPP, to be irreversible. An irrevers-
ible step in a mechanism can be revealed by an unusual product
inhibition profile. In this paper, we describe the cloning, expres-
sion, purification, and some molecular properties of (D1–7)-trun-
cated phosphoketolase-2 from L. plantarum. We also describe the
divalent metal ion preference, the kinetics for activation by diva-
lent metal ions, and steady state kinetics in the action this enzyme.
(ATCC: BAA-793DTM) by PCR and cloned in the expression vector
pET21a(+) (Novagen). Amplification of xpk1 employed the primers
xpk1–1: 50 TTA-AGA-ATT-CAA-CTG-CTT-ATT-TTA-AAC-CC 30 (EcoRI
site underlined) and xpk1–2: 50 TTT-CAT-ATG-ACA-ACA-GAT-TAC-
TCA-TCA-CC 30 (NdeI site underlined). The PCR product was cloned
into the expression vector pET21a(+) between the NdeI and EcoRI
sites, yielding pET21a-xpk1. The recombinant plasmids were iden-
tified by agarose gel electrophoresis followed by DNA sequence
analysis. Amplification of xpk2 employed the primers xpk2–1: 50
TGC-AAA-CAT-ATG-AGT-GAA-GCA-ATT-AAA-TCC 30 (NdeI site
underlined) and xpk2–2: 50 GCT-GAG-CTC-TTG-ATT-AAA-ACT-
CTA-CTT-CAA 30 (SacI site underlined). The PCR product was cloned
into the expression vector pET21a(+) between the NdeI and SacI
sites, yielding pET21a-xpk2. The recombinant plasmids were iden-
tified as described pET21a-xpk1.
The recombinant plasmid, pET21a-xpk1 or pET21a-xpk2, was
transformed into E. coli BL21(DE3) and expression induced by IPTG
at room temperature. The recombinant strains were grown in Lur-
ia–Bertani medium (10 g of tryptone per liter, 5 g of NaCl per liter,
5 g of yeast extract per liter) with 100 lg/mL of ampicillin at 37 °C
to optical density of 0.3–0.5 and then induced with different con-
centrations of IPTG for 3 h at room temperature.
2. Experimental
2.1. Materials
AcP, erythrose-4-phosphate, DEAE-Sepharose, F6P, 2-morpholi-
noethane sulfonic acid (MES), 3-(N-morpholino)propane sulfonic
acid (MOPS), NAD+, NADH, phenylmethane sulfonyl fluoride
(PMSF), S-300 (Sephadex-300), streptomycin sulfate, TPP, and
xylulose were purchased from Sigma Chemical Co. Other materials
and suppliers were Ampicillin and Kanamycin, Fisher Biotech; Bio-
gel P6, BioRad; isopropyl-b-D-thiogalactoside (IPTG, Promega; Ni–
NTA, Qiagen; pET21a and pET28b(+), Novagen; tryptone and yeast
extract, DIFCO. Xu5P was prepared by enzymatic phosphorylation
of xylulose using xylulokinase and MgATP. The pNCO-xylB for
overexpression of xylulokinase was kindly supplied by Professor
A. Bacher and Dr. F. Rohdich of the Technische Universität
München.
Both gene-products form inclusion bodies when expressed at
37 °C but are soluble when expressed at room temperature. Activ-
ities of phosphoketolases in cell extracts were maximal with
0.075 mM IPTG for phosphoketolase-1 and with 1 mM of IPTG for
phosphoketolase-2, as determined by the ferricyanide reduction
assay with F6P as the substrate.
2.2. Assays
The gene encoding phosphoketolase-2 from L. plantarum, ex-
cised from pET21a-xpk2 by digestion with NdeI and SacI, was
cloned into pET28b(+) between the NdeI and SacI sites. From
the resultant construct pET28b-xpk2, phosphoketolase-2 was
expressed with 6 histidine residues and recognition sites for
thrombin in an N-terminal 20-amino acid histidine-tail. The
recombinant plasmid pET28b-xpk2 was transformed into E. coli
BL21(DE3), and the recombinant strains were grown in Luria–Ber-
tani medium (10 g of tryptone per liter, 5 g of NaCl per liter, 5 g of
yeast extract per liter) with 50 lg/mL of kanamycin at 37 °C until
A600 reached 0.3–0.5, and expression of phosphoketolase-2 was
then induced with IPTG for 3 h at room temperature to produce
soluble protein. Maximum activity in cell-free extracts was ob-
served with 1 mM IPTG, similar to expression without the histi-
dine-tail.
Relative protein concentrations were measured by the Bradford
method using bovine serum albumin as the standard.
Initial rates of AcP formation were measured by the hydroxa-
mate method [13]. In the standard assay, the reaction mixture con-
sisted of 5 mM F6P, 1 mM TPP, 5 mM phosphate, 50 mM MES
buffer at pH 6, and 5 mM MgCl2 at 37 °C. The reaction was started
by addition of phosphoketolase and stopped at timed intervals
upon addition of 300 lL of 2 M hydroxylamine at pH 6.5, incuba-
tion 10 min at ambient temperature, and addition of 200 lL of
15% trichloroacetic acid, 200 lL of 4 M HCl, and 200 lL of 5% FeCl3
in 0.1 M HCl. Samples were centrifuged at 21000g for 5 min, and
the values of A505 of the supernatant fluids were measured.
In the ferricyanide reduction assay of phosphoketolase activity,
the reaction mixtures consisted of 5 mM F6P, 50 mM MES buffer at
pH 6, 1 mM TPP, 5 mM MgCl2, and 1 mM ferricyanide at 37 °C. The
reactions were started by addition of F6P and the rate of declining
A420 measured; rates were calculated using the extinction coeffi-
cient 1000 Mꢀ1 cmꢀ1 [14].
2.4. Purification of His-tagged phosphoketolase-2
The recombinant plasmid pET28b-xpk2 was transformed into
E. coli BL21(DE3), and the recombinant strains were grown in Lur-
ia–Bertani medium (10 g of tryptone per liter, 5 g of NaCl per liter,
5 g of yeast extract per liter) with 50 lg/mL of kanamycin at 37 °C
to A600 of 0.3–0.5. Expression was then induced with 1 mM IPTG
for 21 h at room temperature, and the cells were harvested by cen-
trifugation. The cells (6.65 g from 2 L of culture) were resuspended
in 7 mL of lysis buffer (50 mM MOPS pH 7.5; 200 mM NaCl;
0.2 mM PMSF), and sonicated four times for 15 s followed by 30 s
on ice. The sonicate was centrifuged at 11,000g for 15 min at
4 °C. Streptomycin sulfate (2 mL of 30%) was added to the superna-
tant fluid and stirred at 4 °C for 15 min, and the mixture was cen-
trifuged at 11,000g for 20 min at 4 °C. The supernatant fluid was
dialyzed against 500 mL of lysis buffer for 6 h. Dialysis was re-
peated twice more with changes of buffer, first for 6 h and then
for 12 h. The protein solution was then mixed with 5 mL of Ni–
NTA resin (previously equilibrated in 50 mM MOPS pH 7.5;
200 mM NaCl) and incubated on ice for 30 min with occasional
shaking. The mixture was centrifuged at 3000g and 4 °C and the re-
sin washed with 3 volumes of 50 mM MOPS at pH 6, 3 volumes of
2.3. Cloning and expression in Escherichia coli of L. plantarum genes
encoding phosphoketolase
Lactobacillus plantarum has two genes putatively encoding
phosphoketolase, xpk1 and xpk2 [15]. The encoded amino acid se-
quences are 67% identical, and the amino acid sequence of XPK1 is
98% identical with that of the enzyme from L. pentosus. The two
phosphoketolase genes from L. plantarum were cloned using PCR,
the primers being designed to amplify the genomic regions1 to
2412 bp of gene xpk1 encoding XPK1, and 22–2412 bp of xpk2
encoding XPK2(D1–7), comprising the conserved domains pre-
dicted by sequence alignments [7,9,13].Truncation of xpk2 deleted
the N-terminal 7-amino acids, allowing the change of the starting
codon UUG in the Lactobacillus genome to the first AUG as a start-
ing codon, initiation being most efficient in E. coli with AUG. UUG is
estimated to serve as initiator for only about 3% of the proteins in
E. coli [16]. The two genes were amplified from genomic DNA

A. Yevenes, P.A. Frey / Bioorganic Chemistry 36 (2008) 121–127
123
20 mM imidazole in 50 mM MOPS at pH 6, and 3 volumes of
50 mM imidazole in 50 mM MOPS at pH 6.
The assay mixtures contained 50 mM MES buffer at pH 6.0,
0.5 mM TPP, 15 mM phosphate (5 ꢁ Km) and varying F6P and
Me2+. Initial rates were determined in triplicate in activations by
MnTPP and MgTPP and in duplicate for CaTPP. Eq. (3) for the equi-
librium ordered binding mechanism and Eq. (4) for the equilibrium
The protein was eluted with 4 volumes of 150 mM imidazole in
50 mM MOPS at pH 6. The resultant solution (20 mL) was loaded
onto a column of Biogel P6 (1.5 ꢁ 47 cm), equilibrated and eluted
with 50 mM MOPS at pH 7. This step was necessary because the
protein lost activity in the presence of imidazole, a phenomenon
that has been traced to complexation of the activating divalent me-
tal cation by imidazole. The fractions containing phosphoketolase,
as determined by the ferricyanide reduction assay and the assay for
formation of AcP, were loaded on a column of Sephacryl-300
(2.5 ꢁ 46 cm), equilibrated and eluted in 50 mM MOPS at pH 7.
The protein was obtained in ꢂ90% of electrophoretic homogeneity
in a yield of 58 mg (Table 1).
m ¼ VAB=ðKiaKb þ KbA þ ABÞ
m ¼ VAB=ðKiaKb þ KbA þ KaB þ ABÞ
ð3Þ
ð4Þ
random binding mechanism were fitted to the data, where
A = [Me2+TPP] and B = [F6P]. In all cases Eq. (3) for the equilibrium
ordered mechanism gave the best fit of data; fits of Eq. (4) gave very
large values of error.
In the steady state kinetics of the phosphoketolase-catalyzed
reaction of F6P or Xu5P with phosphate to give AcP, the initial rates
of AcP were measured at varied concentrations of the sugar phos-
phate and phosphate. The assay mixtures contained 50 mM MES
buffer at pH 6.0, 5 mM CaCl2, 0.5 mM TPP and varying concentra-
tions of Pi and either F6P or Xu5P. The reactions were started with
10 lg of His-tagged phosphoketolase-2, and each assay was re-
peated in triplicate. The rates of AcP formation were measured col-
orimetrically by the hydroxamate/FeCl3 method. Eq. (4) for the
sequential binding mechanisms and Eq. (5)
2.5. Purification of phosphoketolase-2 from L. plantarum
To purify phosphoketolase-2 without the histidine-tail, the re-
combinant strains with pET21a-xpk2 were induced with 1 mM of
IPTG for 21 h at room temperature. The cells (3.32 g from 1 L)
were resuspended in 3 mL of lysis buffer (50 mM MOPS pH
7.0; 0.2 mM PMSF; 0.2 mM MgCl2), and sonicated four times in
bursts of 15 s followed by 30 s on ice. The mixture was centri-
fuged at 11,000g at 4 °C, and the supernatant fluid was loaded
on a DEAE-Sepharose column (1.5 ꢁ 20 cm), previously equili-
brated in 50 mM MOPS pH 7.0. The column was eluted with a
m ¼ VAB=ðKbA þ KaB þ ABÞ
ð5Þ
for the ping pong mechanism were fitted to the initial rate data,
where A and B were [sugar phosphate] and [Pi]. The best fits were
to Eq. (5) for the ping pong mechanism.
500 mL linear gradient increasing from
0 to 0.8 M NaCl in
50 mM MOPS at pH 7.0, and fractions containing phosphoketo-
lase activity by the ferricyanide reduction assay were loaded
In the product inhibition kinetics, the inhibiting product was
E4P, and the varied substrates were F6P or Pi, respectively. Initial
rates of AcP were measured by the hydroxamate/FeCl3 method,
with [F6P] as varied at several [E4P] and with [Pi] fixed at near
half-saturation (3 mM). Initial rates were measured with [Pi] var-
ied at several [E4P] and with [F6P] fixed at near half-saturation
(25 mM). Eqs. (6)–(8) for competitive, noncompetitive, and
uncompetitive inhibition were fitted to the data, where A repre-
sents concentration of the varied substrate, and I represents [E4P].
on
a
Sephadex-300 column (2.5 ꢁ 46 cm), equilibrated and
eluted in 50 mM MOPS pH 7. Phosphoketolase-2 was obtained
in about 80% electrophoretic homogeneity in a yield of 28 mg
(15% of the initial activity).
2.6. Steady state kinetics
Steady state kinetics methods were employed to determine the
kinetic mechanisms for activation of phosphoketolase by com-
plexes of TPP with divalent metal ions (Me2+TPP) and for the ki-
netic mechanism in the action of phosphoketolase in the
formation of AcP in the presence of saturating Ca2+. In preliminary
experiments, the initial rates of AcP formation were found to be
constant and zero order for at least 30 min at pH 6.0 and 37 °C in
MES buffer.
m ¼ VA=Kð1 þ I=KisÞ þ A competitive
ð6Þ
ð7Þ
ð8Þ
m ¼ VA=Kð1 þ I=KisÞ þ Að1 þ I=KiiÞ noncompetitive
m ¼ VA=K þ Að1 þ I=KiiÞ uncompetitive
3. Results
3.1. Purification and molecular properties of phosphoketolase-2
In studies of activation by Me2+TPP, initial rates of AcP were
measured varying the concentrations of F6P and the divalent metal
ions. The assay mixtures contained 50 mM MES buffer at pH 6.0,
0.5 mM TPP, 15 mM phosphate (5 ꢁ Km) and varying F6P and
Me2+. From the stability constants (KST) of MnTPP and MgTPP com-
plexes [17] the concentrations of Me2+TPP at each concentration of
divalent metal ion could be calculated. The stability constant for
CaTPP was not reported but was evaluated from a linear plot of
six values of KST for various divalent metal ions [17] against the
Me2+-ionic radii. The value of KST for CaTPP was taken from the po-
sition of the ionic radius of Ca2+ in this plot. Double reciprocal plots
of initial rates against [F6P] at several fixed [Me2+TPP] were
constructed.
The genome of L. plantarum incorporates two genes annotated
as phosphoketolases (xpk1 and xpk2 [15]. The encoded amino acid
sequences are 67% identical. To study phosphoketolase from
L. plantarum, both genes were cloned using PCR. Maximum activi-
ties for phosphoketolase-1 and phosphoketolase-2 in extracts were
obtained by induction with 0.075 and 1 mM of IPTG, respectively.
Both of the purified proteins catalyzed the TPP-dependent reduc-
tion of ferricyanide with F6P as the substrate, consistent with
phosphoketolase activity. The purified proteins also catalyzed the
production of AcP and E4P from F6P with TPP-dependence, verify-
ing that they are phosphoketolases.
Table 1
Purification of histidine-tail phosphoketolase-2 from L. plantarum
Procedure
Activitya (U/mL)
Protein (mg/mL)
Specific activity (U/mg)
Volume (mL)
Total activity (U)
Total protein (mg)
Purification factor
Crude extract
Biogel P6
Sephacryl-300
2.81
0.36
1.22
54
1.8
2
0.052
0.2
0.61
7
50
29
20
18
34
378
90
58
1
3.8
11.7
a
Activity was determined by the standard assay for formation of acetyl phosphate (see Section 2).

124
A. Yevenes, P.A. Frey / Bioorganic Chemistry 36 (2008) 121–127
Heterologous expression in E. coli of L. plantarum xpk2 from an
phosphoketolase-2, metal ion activation represents activation by
a divalent metal complex, Me2+TPP.
expression vector incorporating a His-tag to the N-terminus
(pET28b-xpk2), as described in Section 2 produces His-tagged
phosphoketolase-2, and it is conveniently purified by use of a
nickel chelating column (Table 1). Fig. 1 shows the analysis of
the protein by SDS–PAGE from the cell-free extract through nick-
el-chelation and elution with imidazole. The band of phosphoketo-
lase-2 migrates with an apparent molecular weight of 90 kDa,
similar to the calculated subunit molecular weight based on the
amino acid content.
The kinetics of activation by MnTPP, MgTPP, and CaTPP, as eval-
uated by steady state analysis in which the concentrations of F6P
and are varied at 15 mM Pi (ꢂ5Km), indicate ordered equilibrium
binding with Me2+TPP leading. The kinetic parameters are given
in Table 2, and the kinetic plots for activation by Ca, Mn, and
MgTPP appear in Fig. 2. The best fits are of Eq. (3) for ordered equi-
librium binding, with Me2+TPP leading. When the data are plotted
as reciprocal initial rates against [Pi] at several fixed [F6P], the lines
The N-terminal His-tag does not appear to interfere with activ-
ity because the untagged phosphoketolase-2 produced by expres-
sion from the plasmid pET21a-xpk2 and purified to about 70%
homogeneity displayed activity comparable to that of the His-
tagged enzyme.
Table 2
Kinetic parameters for activation of phosphoketolase-2 by divalent metal complexes
of TPP
Parametera
CaTPP
MgTPP
MnTPP
V (IU mg protꢀ1
Kia (lM)
)
1.0 0.1
115 46
1.3 0.1
0.27 0.07
1.0 0.1
24 0.12
3.2. Activation by divalent metal ions
KB (mM)
12
4
21
5
14
6
Like other TPP-dependent enzymes, the activity of phosphoke-
tolase-2 depends upon the presence of a divalent metal ion. The
enzyme is not highly specific for divalent metals, although it might
appear to be upon observation of relative activities and apparent
Km values with various divalent ions (appKmCa = 160 lM,
appKmMg = 32 lM, appKmMn = 0.4 lM). The apparent selectivity
can be traced to the differential binding affinities of divalent metal
ions to TPP. Values of stability constants (KST) for various Me2+TPP
complexes are available and range from 4.9 ꢁ 102 to 8.5 ꢁ 103 Mꢀ1
for the monocationic complexes, the dominant species at pH 6
a
A =, B = F6P, V = maximum velocity, Kia = dissociation constant for E., and
KB = Michaelis constant for F6P.
[17]. Manganous ion has
a much higher affinity for TPP
(KST = 3.6 ꢁ 103 Mꢀ1) than Mg2+ (KST = 6.6 ꢁ 102 Mꢀ1), and Mn2+
activates phosphoketolase-2 at much lower concentrations than
does Mg2+. However, the maximum activities at saturating metal
ions are comparable. The kinetics for reactions between divalent
metal ions and TPP show very fast complexation, with equilibrium
attained within milliseconds [17]. Therefore, in a typical assay of
Fig. 2. Kinetic plots for activation of phosphoketolase-2 by divalent metal com-
plexes of TPP. Shown are double reciprocal plots of initial rates for phosphoketo-
lase-2 at pH 6.0 and 37 °C in 50 mM MES buffer. (A) Activation by MgTPP: the
concentrations of F6P were 16, 24, 32, and 64 mM; and the concentrations of MgTPP
were j 2.5 lM, h 7.3 lM, and Ç 18 lM. (B) Activation by MnTPP: the concentrations
of F6P were 16, 24, 32, and 64 mM; and the concentrations of MnTPP were j
0.13 lM, h 0.65 lM, and Ç 3.2 lM. (C) Activation by CaTPP: the F6P concentrations
of F6P were 8, 16, 32, and 64 mM; and the concentrations of CaTPP were j 27 lM, h
90 lM, Ç 250 lM, and 450 lM. The data in (A), (B), and (C) are fitted in Eq. (3) for
equilibrium ordered binding, with binding first. The kinetic parameters for CaTPP,
MgTPP, and MnTPP are given in Table 2.
Fig. 1. Purification of His-tagged phosphoketolase-2 from L. plantarum. The SDS–
PAGE analysis in lane 2 of cell-free extract of E. coli expressing the gene xpk2 shows
the overexpression of phosphoketolase-2. Lane 3 shows the SDS–PAGE analysis of
His-tagged phosphoketolase-2 after purification by absorption to and elution from a
nickel chelator, and lane 4 shows the protein after S-300 chromatography. The
molecular weight markers are shown in lane 1.

A. Yevenes, P.A. Frey / Bioorganic Chemistry 36 (2008) 121–127
125
Table 3
converge at the left of the ordinate, in accord with Eq. (3). As
shown in Table 2, the affinities of phosphoketolase-2 for MnTPP,
MgTPP, and CaTPP are similar, though not identical, as are the max-
imum activities.
Kinetic parameters for reactions of F6P or Xu5P with P0 in the reaction of
phosphoketolase-2
Substrates
F6P and Pi
1.8 0.2
Xu5P and Pi
kcat [E] = V (IU mgꢀ1
Ka (mM)a
)
4.1 0.4
3.6 0.6
7.5 1.3
ndb
24
4
3.3. The steady state kinetic mechanism for reaction of
phosphoketolase-2
Kb (mM)a
2.9 0.5
E4P
Kis
Kii
ðmMÞ
8
4
ðmMÞ
4.4 0.9
ndb
E4P
The initial rates for the formation of AcP catalyzed by phospho-
ketolase-2 with F6P as the substrate are plotted in double recipro-
cal form in Fig. 3A and B. The data describe parallel lines, and Eq.
(5), corresponding to a ping pong bi bi kinetic mechanism is fitted
to the data. Xu5P as the substrate gives a similar pattern of parallel
lines shown in Fig. 3C and D. Kinetic parameters for the reactions of
F6P and Xu5P are included in Table 3.
a
F6P
Ka, Michaelis constant for F6P (Km
in Eqs. (9) and (10)) or Xu5P; Kb, Michaelis
P
constant for Pi (Km in Eqs. (9) and (10)).
b
Not determined.
Product inhibition by E4P is uncompetitive with respect to Pi
and noncompetitive with respect to F6P, as illustrated in Fig. 4.
This inhibition pattern is unusual for ping pong bi bi kinetic mech-
anisms. The first product to dissociate normally displays noncom-
Fig. 4. Product inhibition of phosphoketolase-2 by E4P. The reaction mixtures co-
ntained 50 mM MES at pH 6, 5 mM CaCl2, and 0.5 mM TPP together with varied
concentrations of F6P, Pi, and E4P. (A) F6P varied: the concentrations of F6P were 8,
16, 24, 32, and 48 mM; and the constant substrate (phosphate) was 3 mM. The
concentrations of E4P were 8 mM, 4 mM, Ç 2 mM, h 1 mM, and j 0 mM. (B) Pi
varied: the concentrations of the variable substrate (Pi) were 1, 1.5, 3, 6, and 12 -
mM; and the constant substrate (F6P) was 25 mM. The concentrations of erythrose-
4-phosphate (E4P) were 8 mM, Ç 4 mM, h 1 mM, and j 0 mM.
petitive inhibition with respect to both substrates [18], as is found
in the case of F6P as the varied substrate. With Pi as the varied sub-
strate, uncompetitive inhibition by E4P must be explained. The
rationale for the uncompetitive pattern in Fig. 4 is given in Section
4 in connection with a minimal kinetic mechanism.
4. Discussion
Fig. 3. Kinetic plots of initial rates for the overall reaction of phosphoketolase-2.
(A–D) Double reciprocal plots of initial rates AcP formation in the reactions of F6P or
Xu5P with Pi. (A) The varied concentrations of Pi were 1, 2, 4, and 8 mM, and the
changing fixed concentrations of F6P were j 8 mM, h 16 mM, Ç 32 mM, and
64 mM. (B) The varied concentrations of F6P were 8, 16, 32, and 64 mM; and the
changing fixed concentrations of Pi were j 1 mM, h 2 mM, Ç 4 mM, and 8 mM.
(C) The varied concentrations of Xu5P were 1.2, 2.3, 4.6, and 9.3 mM; and the
changing fixed concentrations of Pi were j 3 mM, h 9 mM, Ç 27 mM, and 54 mM.
(D) The varied concentrations of Pi were 3, 9, 27, and 54 mM, and the changing fixed
concentrations of Xu5P were j 1.2, h 2.3, Ç 4.6, and 9.3 mM. The parallel lines are
fitted in Eq. (5) for the ping pong bi bi kinetic mechanism, and he kinetic parameters
are given in Table 3.
4.1. Kinetics in the action of phosphoketolase-2
Scheme 1 shows the simplest and most likely ping pong kinetic
mechanism for the overall action of phosphoketolase-2 on F6P,
where TPP refers to Me2+TPP and Me = Ca2+, Mg2+, or Mn2+
.
The rate constants are numbered following the convention that
forward constants are odd-numbered and reverse constants are
even-numbered. In the absence of E4P, k[E4P] = 0, and the initial
rate law is Eq. (9), the same form as Eq. (5),

126
A. Yevenes, P.A. Frey / Bioorganic Chemistry 36 (2008) 121–127
kcat½Eꢃ½F6Pꢃ½Piꢃ
ibility in the dehydration of E.DHETPP, no reversible step connects
m ¼
ð9Þ
F6P
P
½F6Pꢃ½Piꢃ þ Km ½Piꢃ þ Km ½F6Pꢃ
these two enzyme forms, so there is no slope effect, and the inhi-
bition is uncompetitive. The product inhibition reveals a special
case of the ping pong bi bi mechanism, wherein an irreversible step
makes one inhibition pattern uncompetitive instead of the usual
noncompetitive pattern observed when all steps are reversible.
Noncompetitive inhibition by E4P with respect to F6P arises from
the binding of E4P and F6P to different enzyme forms E.DHETPP
and E.TPP, which are connected by reversible steps.
where kcat = k3k5k9/(k5k9 + k3k9 + k3k5), KmF6P = [k5k9(k2 + k3)]/[k1(k5k9 +
k3k9 + k3k5)], and KmP = [k3k5(k8 + k9)]/[k7(k5k9 + k3k9 + k3k5)]. In Eq.
(5), Kb = KmF6P and Ka = KmP. In the ping pong mechanism, E.DHETPP
and E.AcTPP are intermediates generated in the reaction of F6P with
E.TPP independently of the second substrate Pi. According to the
mechanism, F6P reacts with the complex of phosphoketolase-2
and to form enzyme-bound DHETPP and release E4P in the absence
of Pi. DHETPP undergoes dehydration irreversibly at the active site
to AcTPP, and Pi then binds to the enzyme, accepts the acetyl group
from AcTPP, and AcP undergoes dissociation.
Irreversibility in the dehydration of DHETPP to AcTPP explains
uncompetitive inhibition by E4P with respect to Pi. By Cleland’s
Rules [18], E4P and Pi bind to different forms of the enzyme,
E.DHETPP and E.AcTPP, respectively, giving an intercept effect in
the double reciprocal plots with varying [Pi]. Because of irrevers-
Accordingly, Eq. (10) is the initial rate law for the mechanism in
Scheme 1 when E4P is present and product inhibition occurs
kcat½Eꢃ½F6Pꢃ½Piꢃ
ꢀ
ꢁ
ꢀ
ꢁ
m ¼
ð10Þ
½E4Pꢃ
½E4Pꢃ½Piꢃ
F6P
P
½F6Pꢃ½Piꢃ þ Km ½Piꢃ 1 þ
þ Km ½F6Pꢃ 1 þ
E4P
E4P
Kis
Kii
In Eq. (10), kcat, KmF6P, and KmP are as in Eq. (9), and Kis = [k5(k2 + k3)]/k
2k4 and Kii = [k3k5(k8 + k9)]/k4k7k9. Eq. (10) takes the form of noncom-
petitive inhibition by E4P, when F6P is the varied substrate, and
uncompetitive inhibition when Pi is the varied substrate, in accord
with Fig. 4. Values for Kis and Kii are given in Table 3.
The kinetic mechanism in Scheme 1 predicts that phosphoketo-
lase catalyzes the exchange of E4P with F6P in the absence of Pi.
The literature is confused on this point. In one study, no correspond-
ing exchange of [14C]GAP/Xu5P could be observed [4] whereas in an-
other study DHETPP could be transformed with GAP into Xu5P
[1,19]. In the earlier study, the concentration of [14C]GAP was quoted
as ‘‘trace”, implying unknown and very low. A low concentration of
GAP would not react rapidly and could lead to a false negative result.
The product inhibition constants in this report support the kinetic
mechanism in Scheme 1 and the results of the transformation of
DHETPP and GAP into Xu5P, as reported [1,19].
Phosphoketolase from L. plantarum does not catalyze any rever-
sal of the formation of AcTPP, even enolization to enolacetyl-TPP,
Scheme 1. A kinetic mechanism for phosphoketolase-2.
Scheme 2. A chemical mechanism for phosphoketolase-2.

A. Yevenes, P.A. Frey / Bioorganic Chemistry 36 (2008) 121–127
127
because a study in [3H]H2O with arsenate as the acceptor led to one
References
equivalent of tritium in acetate. Based on this result, Goldberg and
Racker wrote: ‘‘This indicates that the formation of acetate does
not involve labialization of the hydrogen atoms on the b-carbon
of the thiamine-linked intermediate” [20], the intermediate being
DHETPP. Furthermore, phosphoketolase does not catalyze an ex-
change of 32Pi with AcP, showing that the reaction of Pi with AcTPP
at the active site is irreversible as well.
All things considered, available information from the literature
and this paper support the mechanism in Scheme 2 for the action
of phosphoketolase on Xu5P. In this mechanism, AcTPP arises from
dehydration of the initial adduct between the substrate and TPP,
followed by ketonization of the resultant enolacetyl-TPP. Both
dehydration and ketonization appear to be irreversible processes.
This mode of AcTPP formation differs from the oxidative processes
leading to AcTPP in the reactions of TPP-dependent oxidases and
oxidoreductases, in which electron transfer from 2-hydroxyethy-
lidene-TPP (HETPP) leads to AcTPP [21–24].
The dehydration of DHETPP in Scheme 2 is analogous to the
mechanism by which 3-fluropyruvate generates AcTPP in the active
site of pyruvate dehydrogenase, component E1 in the pyruvate
dehydrogenase complex [25,26]. Pyruvate dehydrogenase accepts
3-fluorpyruvate as a substrate and catalyzes its decarboxylation
and defluorination in the presence of dihydrolipoate to form
S-acetyldihydrolipoate. The mechanism follows that for the
decarboxylation of pyruvate to form b-fluoro-a-hydroxyethyli-
dene-TPP, which undergoes elimination of fluoride to generate
2-enolacetyl-TPP, analogous to the dehydration of DHETPP in
Scheme 2. Ketonization of 2-enolacetyl-TPP to AcTPP and acetyl
group transfer to dihydrolipoate produces S-acetyldihydrolipoate,
a step analogous to the reaction of phosphate with AcTPP in Scheme 2.
[1] L.O. Krampitz, Annu. Rev. Biochem. 38 (1969) 213–240.
[2] B. Pieterse, R.J. Leer, H.J. Schurem, M.J. van der Werf, Microbiology 151 (2005)
3881–3894.
[3] A. Bairoch, Nucleic Acids Res. 28 (2000) 304–305.
[4] E.C. Heath, J. Hurwitz, B.L. Horecker, A. Ginsburg, J. Biol. Chem. 231 (1958)
1009–1029.
[5] J.P. Grill, J. Crociani, J. Ballongue, Curr. Microbiol. 31 (1995) 49–54.
[6] K.G. Fandi, H.M. Ghazali, A.M. Yazid, A.R. Raha, Lett. Appl. Microbiol. 32 (2001)
235–239.
[7] J.M. Lee, D.W. Jeong, O.K. Koo, M.J. Kim, J.H. Lee, H.C. Chang, J.H. Kim, H.J. Lee,
Biotechnol. Lett. 27 (2005) 853–858 (Erratum in: Biotechnol Lett. 27 (16)
1247).
[8] C. Ingram-Smith, S.R. Martin, K.S. Smith, Trends Microbiol. 14 (2006) 249–253.
[9] C.C. Posthuma, R. Bader, R. Engelmann, P.W. Postma, W. Hengstenberg, P.H.
Pouwels, Appl. Environ. Microbiol. 68 (2002) 831–837.
[10] P.A. Frey, Biofactors 2 (2002) 1–9.
[11] I. Merkler, J. Retey, Eur. J. Biochem. 120 (2002) 593–597.
[12] W. Schöter, H. Holzer, Biochim. Biophys. Acta 77 (1963) 474–484.
[13] L. Meile, L.M. Rohr, T.A. Geissmann, M. Herensperger, M. Teuber, J. Bacteriol.
183 (2001) 2929–2936.
[14] G.A. Kochetov, Methods Enzymol. 89 (1982) 43–44.
[15] M. Kleerebezem, J. Boekhorst, R. van Kranenburg, D. Molenaar, O.P. Kuipers, R.
Leer, R. Tarchini, S.A. Peters, H.M. Sandbrink, M.W. Fiers, W. Steikema, R.M.
Lankhorst, P.A. Bron, S.M. Hoffer, M.N. Groot, R. Kerkhoven, M. de Vries, B.
Ursing, W.M. de Vos, R.J. Siezen, Proc. Natl. Acad. Sci. USA 100 (2003) 1990–
1995.
[16] F.R. Blattner, G. Plunckett 3rd, C.A. Bloch, N.T. Perna, V. Burland, M. Riley, J.
Collado-Vides, J.D. Glasner, C.K. Rode, G.F. Mayhew, J. Gregor, N.W. Davis, H.A.
Kirkpatrick, M.A. Goeden, D.J. Rose, B. Mau, Y. Shao, Science 277 (1997) 1453–
1474.
[17] H.B. Katz, K. Kustin, Biochim. Biophys. Acta 313 (1973) 235–248.
[18] W.W. Cleland, The Enzymes, third ed., vol. 2, pp. 1–65.
[19] R. Votaw, W.T. Williamson, L.O. Krampitz, W.A. Wood, Biochem. Z. 338 (1963)
756–762.
[20] M.L. Goldberg, E. Racker, J. Biol. Chem. 237 (1962) 3841–3842.
[21] K.J. Gruys, A. Atta, P.A. Frey, Biochemistry 28 (1989) 9071–9080.
[22] K. Tittmann, G. Wille, R. Golbik, A. Weidner, S. Ghisla, G. Hübner, Biochemistry
44 (2005) 13291–13303.
[23] K. Tittmann, R. Golbik, S. Ghisla, G. Hübner, Biochemistry 39 (2000) 10747–
10754.
Acknowledgment
[24] G. Wille, D. Meyer, A. Steinmetz, E. Hinze, R. Golbik, K. Tittmann, Nat. Chem.
Biol. 2 (2006) 324–328.
[25] L.S. Leung, P.A. Frey, Biochem. Biophys. Res. Commun. 81 (1978) 274–279.
[26] D.S. Flournoy, P.A. Frey, Biochemistry 25 (1986) 6036–6043.
This research was supported by Grant GM30480 from the
National Institute of General Medical Sciences.