Increasing the conformational entropy of the Ω-loop lid domain in phosphoenolpyruvate carboxykinase impairs catalysis and decreases catalytic fidelity

Article abstract of DOI:10.1021/bi100399e

Many studies have shown that the dynamic motions of individual protein segments can play an important role in enzyme function. Recent structural studies of the gluconeogenic enzyme phosphoenolpyruvate carboxykinase (PEPCK) demonstrate that PEPCK contains a 10-residue Ω-loop domain that acts as an active site lid. On the basis of these structural studies, we have previously proposed a model for the mechanism of PEPCK catalysis in which the conformation of this mobile lid domain is energetically coupled to ligand binding, resulting in the closed conformation of the lid, necessary for correct substrate positioning, becoming more energetically favorable as ligands associate with the enzyme. Here we test this model by introducing a point mutation (A467G) into the center of the Ω-loop lid that is designed to increase the entropic penalty for lid closure. Structural and kinetic characterization of this mutant enzyme demonstrates that the mutation has decreased the favorability of the enzyme adapting the closed lid conformation. As a consequence of this shift in the equilibrium defining the conformation of the active site lid, the enzymes ability to stabilize the reaction intermediate is weakened, resulting in catalytic defect. This stabilization is initially surprising, as the lid domain makes no direct contacts with the enolate intermediate formed during the reaction. Furthermore, during the conversion of OAA to PEP, the destabilization of the lid-closed conformation results in the reaction becoming decoupled as the enolate intermediate is protonated rather than phosphorylated, resulting in the formation of pyruvate. Taken together, the structural and kinetic characterization of A467G-PEPCK supports our model of the role of the active site lid in catalytic function and demonstrates that the shift in the lowest-energy conformation between open and closed lid states is a function of the free energy available to the enzyme through ligand binding and the entropic penalty for ordering of the 10-residue Ω-loop lid domain.

Full text of DOI:10.1021/bi100399e

5
176 Biochemistry 2010, 49, 5176–5187
DOI: 10.1021/bi100399e
Increasing the Conformational Entropy of the Ω-Loop Lid Domain in
Phosphoenolpyruvate Carboxykinase Impairs Catalysis and Decreases
†,‡
Catalytic Fidelity
Troy A. Johnson and Todd Holyoak*
Department of Biochemistry and Molecular Biology, The University of Kansas Medical Center, Kansas City, Kansas 66160
Received March 15, 2010; Revised Manuscript Received April 26, 2010
ABSTRACT: Many studies have shown that the dynamic motions of individual protein segments can play an
important role in enzyme function. Recent structural studies of the gluconeogenic enzyme phosphoenolpyr-
uvate carboxykinase (PEPCK) demonstrate that PEPCK contains a 10-residue Ω-loop domain that acts as an
active site lid. On the basis of these structural studies, we have previously proposed a model for the mechanism
of PEPCK catalysis in which the conformation of this mobile lid domain is energetically coupled to ligand
binding, resulting in the closed conformation of the lid, necessary for correct substrate positioning, becoming
more energetically favorable as ligands associate with the enzyme. Here we test this model by introducing a
point mutation (A467G) into the center of the Ω-loop lid that is designed to increase the entropic penalty for
lid closure. Structural and kinetic characterization of this mutant enzyme demonstrates that the mutation has
decreased the favorability of the enzyme adapting the closed lid conformation. As a consequence of this shift
in the equilibrium defining the conformation of the active site lid, the enzyme’s ability to stabilize the reaction
intermediate is weakened, resulting in catalytic defect. This stabilization is initially surprising, as the lid
domain makes no direct contacts with the enolate intermediate formed during the reaction. Furthermore,
during the conversion of OAA to PEP, the destabilization of the lid-closed conformation results in the reaction
becoming decoupled as the enolate intermediate is protonated rather than phosphorylated, resulting in the
formation of pyruvate. Taken together, the structural and kinetic characterization of A467G-PEPCK
supports our model of the role of the active site lid in catalytic function and demonstrates that the shift in the
lowest-energy conformation between open and closed lid states is a function of the free energy available to the
enzyme through ligand binding and the entropic penalty for ordering of the 10-residue Ω-loop lid domain.
1
2þ
Phosphoenolpyruvate carboxykinase (PEPCK) catalyzes the
requirement for divalent cations for activity. Mn is the most
reversible decarboxylation and phosphorylation of OAA to form
PEP as shown in Scheme 1. While in vitro the reaction is freely
reversible, the overall consensus is that in most organisms
PEPCK operates primarily in the direction of PEP synthesis.
PEPCK is a metal-requiring enzyme demonstrating an absolute
activating cation in the GTP-dependent isoforms (1-4). In
addition, a second divalent metal ion is required for the reaction,
as the metal-nucleotide complex is the true substrate. In higher
eukaryotes, PEPCK is present as both a cytosolic (cPEPCK) and
mitochondrial (mPEPCK) isoform, with the relative distribution
of these two isoforms being species-dependent. In its biological
role, cPEPCK functions as a key cataplerotic enzyme; in addition
to its well-characterized role in gluconeogenesis, PEPCK parti-
cipates in glyceroneogenesis and triglyceride biosynthesis as well
as the synthesis of serine (reviewed in ref 5). In contrast, the role
of mPEPCK in metabolic function is less understood.
In general, the structural data on PEPCK illustrate the
presence of a specialized cationic active site, dominated by the
juxtaposition of the two aforementioned divalent metal ions and
the positioning of specific lysine and arginine residues, that is well
suited to carry out both the decarboxylyation/carboxylation and
phosphoryl transfer half-reactions as well as the stabilization of
the enolate intermediate postulated to form along the step-
wise reaction pathway (6). Because of the inherent reactivity
of the enolate intermediate, primarily its energetically favor-
able protonation resulting in pyruvate, we have suggested that
the protection or stabilization of the enolate by the enzyme is
key to the reversible nature of the PEPCK-catalyzed reaction
(Scheme 1) (6).
†
T.H. acknowledges support from National Center for Research
Resources (Grant P20 RR17708). Portions of this research were con-
ducted at the Stanford Synchrotron Radiation Laboratory (SSRL), a
national user facility operated by Stanford University on behalf of the
U.S. Department of Energy, Office of Basic Energy Sciences. The SSRL
Structural Molecular Biology Program is supported by the Department
of Energy, Office of Biological and Environmental Research, and by the
National Institutes of Health, National Center for Research Resources,
Biomedical Technology Program, and the National Institute of General
Medical Sciences.
‡
Coordinates and structure factors have been deposited in the RCSB
Protein Data Bank as entries 3MOE, 3MOF, and 3MOH.
*To whom correspondence should be addressed. Phone: (913) 588-
0
795. Fax: (913) 588-7440. E-mail: tholyoak@kumc.edu.
Abbreviations: ASU, asymmetric unit; βSP, 3-sulfopyruvate;
1
cPEPCK, cytosolic phosphoenolpyruvate carboxykinase; DTT, dithio-
0
0
threitol; GDP, guanosine 5 -diphosphate; GTP, guanosine 5 -tripho-
0
0
sphate; GTPγS, guanosine 5 -O-(γ-thio)triphosphate; IDP, inosine 5 -
0
0
diphosphate; ITP, inosine 5 -triphosphate; ITPγS, inosine 5 -O-(γ-
thio)triphosphate; NCS, noncrystallographic symmetry; OAA, oxaloa-
cetic acid; PEG, polyethylene glycol; PEP, phosphoenolpyruvate;
PEPCK, phosphoenolpyruvate carboxykinase; PGA, 2-phosphoglyco-
late; rmsd, root-mean-square deviation; TIM, triosephosphate isomer-
ase; TLS, translation/libration/screw.
pubs.acs.org/Biochemistry
Published on Web 05/17/2010
r 2010 American Chemical Society

Article
An informative aspect of the recent structural studies of the
Biochemistry, Vol. 49, No. 25, 2010 5177
of the reaction, resulting in a kcat/KM for OAA that is only 2% of
GTP-dependent isozyme from rat is the illumination of the
previously unappreciated role of conformational changes occur-
ring at the active site during the catalytic cycle (6-9). The most
prevalent mobile feature illustrated by the structural work is a 10-
residue Ω-loop lid domain, reminiscent of a similar domain in
TIM (10-13). An essential role for this domain in PEPCK-
mediated catalysis is suggested by the structural data for PEPCK
demonstrating that only upon closure of the lid domain are the
substrates positioned correctly for phosphoryl transfer to occur
the wild-type (WT) value.
MATERIALS AND METHODS
Materials. The nucleotides (GTP, GDP, IDP, ITP, and ADP)
and 2-phosphoglycolate were purchased from Sigma. PEP and
NADH were from Chem-Impex. DTT was from Gold Bio-Tech.
HEPES buffer was from Research Organics. Oxalic acid was
from Fluka. βSP was synthesized and purified as previously des-
cribed (17). Glutathione Uniflow Resin was purchased from
Clontech. HiQ, P6DG, and Chelex resins were from Bio-Rad. All
other materials were of the highest grade commercially available.
Enzymes. Malate dehydrogenase (22976 units/mL, 50%
glycerol solution) and lactate dehydrogenase (6000 units/mL,
(9). In addition to the essential role lid closure is proposed to
play in positioning the substrates for catalysis, closure of the
lid may additionally sequester the enolate intermediate, allow-
ing PEPCK-mediated catalysis to occur via the mechanism
illustrated in Scheme 1.
AmSO
kinase (10 mg/mL, AmSO
suspension) were from Calzyme Laboratories. Pyruvate
suspension) was from Roche.
4
The dynamic nature of lid opening and closing begs the
question of what the energetic driving force for lid closure is
and what specific role lid closure plays in the catalytic cycle. As the
structural data demonstrate that no new direct contacts are made
between the lid when it closes in the presence of substrates versus
in their absence, we have previously proposed a model consistent
with the notion originally proposed by Fersht (14). In this model,
the thermodynamic favorability of the enzyme adapting the closed
lid (active) state increases as ligands add to the enzyme, with a
portion of their Gibbs free energy of binding being partitioned to
the protein. This energy partitioning has the effect of remodeling
the free energy profile for the protein and offsetting the entropic
penalty associated with the lid assuming an ordered closed con-
formation rather than a conformationally dynamic open (inactive)
state (9). Previous solid-state and solution NMR studies on
TIM (15, 16) are consistent with the model we propose for PEPCK
lid motion (above and ref 9), as these studies on TIM suggest that
the position of the open-closed equilibrium is dependent on the
nature of the bound ligand.
To further probe the role of lid closure in PEPCK catalytic
function and to test our model, in which the penalty for the redu-
ction in configurational entropy of the enzyme upon lid ordering
is paid by the interaction free energy of substrates and/or inter-
mediates, we have generated a lid mutation [A467G (Figure 1)]
that is designed to increase the entropic penalty for ordering of
the lid in the closed conformation. Structural and kinetic char-
acterization of this mutant demonstrates that, consistent with our
model for lid closure, the A467G mutation results in a decrease in
the propensity for the lid to adapt its closed conformation. As a
functional consequence, this disruption in the energetics of
lid closure results in an observed decrease in the fidelity of the
conversion of OAA to PEP through the decoupling of the react-
ion to pyruvate. In addition, even though the closed lid makes no
direct contacts with the substrates, the kinetic data are consistent
with the destabilization of the closed lid conformation resulting in
PEPCK becoming less able to stabilize the enolate intermediate.
This loss of enolate stabilization significantly impacts the kinetics
4
PEPCK Expression and Purification. Recombinant, WT
rat cPEPCK was expressed and purified as previously described
(8) with the following alterations. All buffers in the purification
included 10 mM DTT and were treated with Chelex resin for
10 min prior to filter degassing. The molar extinction coefficient
-
1
used to determine protein concentrations is 1.65 mL mg as
determined experimentally (18).
Generation of A467G PEPCK. Recombinant rat cPEPCK
in PGEX-4T-2 was used as the starting vector to create the
0
A467G mutation. The forward primer (5 -GAGGCCACCGC-
0
TGGTGCAGAGCATAAGG-3 ) and its reverse complement
were utilized with the Stratagene Quik Change protocol. The
resultant mutated DNA was isolated with a Hurricane cleanup
kit from Gerard Biotech and subsequently sequenced to confirm
the presence of the desired mutation and the absence of any addi-
tional mutations introduced via the polymerase chain reaction
(PCR) protocol. The plasmid DNA was transformed into Esche-
richia coli BL-21(DE3) electro-competent cells for expression.
Expression and purification of the A467G mutant followed the
same protocol as the WT enzyme with the same modifications
described above.
Crystallization. Crystals of the complexes of A467G PEPCK
were obtained as previously described for the WT enzyme (9).
Data Collection. Data for the cryo-cooled A467G PEPCK-
2
þ
2þ
2þ
Mn -βSP-Mn GTP, and A467G PEPCK-Mn -oxalate-
2þ
Mn GTP complexes maintained at 100 K were collected on
beamline 11-1 of the Stanford Synchrotron Radiation Laboratory
2
þ
(Menlo Park, CA). Data for the A467G PEPCK-Mn
-
2þ
PGA-Mn GDP crystal maintained at 100 K were collected
using a RU-H3R rotating copper anode X-ray generator with Blue
F
IGURE 1: Sequence of the active site lid domain (residues 462-476).
The site of introduction of the glycine (position 467) is indicated
above the sequence.
Scheme 1: PEPCK-Catalyzed Interconversion of OAA and PEP

5
178 Biochemistry, Vol. 49, No. 25, 2010
Johnson and Holyoak
Table 1: Crystallographic Data and Model Statistics for the Structures of A467G PEPCK
A467G PEPCK-
A467G PEPCK-
A467G PEPCK-
2þ
2þ
2þ
Mn -βSP-
Mn -PGA-
Mn -oxalate-
2
þ
2þ
2þ
Mn GTP
Mn GDP
Mn GTP
˚
wavelength (A)
0.9
P2
a = 44.3 A
1.54
0.9
space group
unit cell
1
P2
a = 62.1 A
1
P2
a = 62.0 A
1
˚
˚
˚
˚
˚
˚
b = 119.1 A
b = 119.5 A
b = 119.6 A
˚
˚
˚
c = 60.1 A
c = 87.0 A
c = 87.1 A
R = γ = 90.0°
β = 111.2°
20.6-1.25
152044
R = γ = 90.0°
β = 107.2°
35.93-2.10
68697
R = γ = 90.0°
β = 106.9°
33.24-1.75
118295
˚
resolution limit (A)
no. of unique reflections
a
completeness (%) (all data)
95.0 (67.2)
6.9 (3.5)
97.2 (80.3)
7.0 (4.9)
10.0 (1.6)
0.1 (0.59)
2
96.7 (94.9)
7.1 (6.9)
a
redundancy
a
I/σ(I)
28.2 (2.1)
0.04 (0.46)
1
18.0 (2.3)
0.06 (0.55)
2
a
R
merge
no. of ASU molecules
a
R
R
free
17.4 (34.2)
14.8 (32.7)
0.03
24.7 (34.1)
19.0 (28.2)
0.16
22.8 (33.1)
17.9 (29.3)
0.10
a
work
estimated coordinate error based on
˚
maximum likelihood (A)
˚
bond length rmsd (A)
0.03
0.01
0.02
bond angle rmsd (deg)
Ramachandran statistics
2.6
1.5
1.5
96.9, 2.4, 0.7
96.4, 3.0, 0.6
96.5, 2.7, 0.8
(
preferred, allowed, outliers) (%)
state of active site lid
molecule A, open
molecule A, open
molecule B, open
molecule A, 70% closed
molecule B, open
a
Values in parentheses represent statistics for data in the highest-resolution shells.
þþ
Confocal Osmic mirrors and a Rigaku Raxis IV detector. All
concentrations of PEPCK utilized were 14-36 and 43-86 nM
data were integrated and scaled with HKL-2000 (19). Final data
statistics are presented in Table 1.
for the WT and A467G enzyme forms, respectively.
Assay A (OAA þ GTP f PEP þ CO þ GDP). The
2
Structure Determination and Refinement. The A467G
enzyme structures were determined by the molecular replacement
method using MOLREP (20) in the CCP4 package (21) and
the previously determined rat cPEPCK structures of the same
complexes [Protein Data Bank (PDB) entries 3DT4, 3DT7,
and 3DTB] (9). For each complex, the molecular replacement
solution was refined using Refmac5 followed by manual model
adjustment and rebuilding using COOT (22). Ligand, metal, and
water addition and validation were also performed in COOT.
A final round of TLS refinement was performed for all models
in Refmac5. A total of five groups per chain for the A467G
decarboxylation and phosphorylation of OAA to form PEP
were conducted in a reaction mixture containing 50 mM HEPES
(pH 7.5), 10 mM DTT, 4 mM MgCl , 500 μM GTP, 1 mM ADP,
2
200 μM MnCl , 150 μM NADH, 30 units of LDH, 50 μg of PK,
2
and 350 μM OAA. All enzyme-catalyzed rates that use OAA as
the substrate have been corrected for the nonenzymatic decar-
boxylation of OAA forming pyruvate (assays A, C, and D). This
assay was started by the addition of OAA to each cuvette. OAA
and GTP concentrations were varied in turn to yield the respec-
tive Michaelis constants. When OAA was varied, the reaction
was performed as described above. When GTP was the varied
2
þ
2þ
2þ
PEPCK-Mn -βSP-Mn GTP and A467G PEPCK-Mn
oxalate-Mn GTP models were used as refinements using more
-
substrate, the level of MgCl was also varied to keep a constant
2
2
þ
4:1 ratio of metal to nucleotide. The reaction mixture was
than five groups per chain did not significantly improve R/R
.
-
preincubated at 25 °C for 5 min prior to the addition of OAA.
free
2
þ
A total of 20 groups per chain for the A467G PEPCK-Mn
Assay B (PEP þ GDP þ CO f OAA þ GTP). The
2
2þ
PGA-Mn GDP model were used. The optimum TLS groups
were determined by submission of the PDB files to the TLSMD
server (http://skuld.bmsc.washington.edu/∼tlsmd/) (23-25). Final
model statistics are listed in Table 1.
dephosphorylation and carboxylation of PEP to form OAA
were assessed in a reaction mixture consisting of 50 mM HEPES
(pH 7.5), 10 mM DTT, 4 mM MnCl , 500 μM GDP, 150 μM
2
NADH, 2 mM PEP, 10 units of MDH, and 50 mM KHCO . This
3
Kinetic Experiments. All kinetic data were collected on a
Cary 100 spectrophotometer equipped with a multicell changer
and a temperature bath. PEPCK activity was assayed in both the
physiological direction of PEP formation and the reverse direc-
tion of OAA formation using modifications of previously publis-
hed assays (1, 18, 26). All assays were performed at 25 °C, in a
final volume of 1 mL, and monitored spectrophotometrically by
following the decrease in absorbance at 340 nm due to the oxi-
dation of NADH. The oxidation of NADH was recorded in real
time, to allow the calculation of a rate of enzyme activity from the
extinction coefficient for NADH (27). In all kinetic assays, the
assay was started by the addition of PEPCK to the reaction
mix in the cuvette. PEP, KHCO , and GDP concentrations were
3
varied in turn to yield the respective Michaelis constants. When
GDP was the varied substrate, the level of MnCl was also varied
2
to keep a constant metal to nucleotide ratio of 4:1. The con-
centration of dissolved CO was calculated from the amount of
2
added KHCO as previously described (28).
3
Assay C (OAA þ GTP f pyruvate þ GTP þ CO ). The
2
decoupling of the PEPCK-catalyzed reaction (assay A) through
the production of pyruvate was assessed in the same reaction mix
as assay A. The only exception was that the coupling enzyme PK

Article
Biochemistry, Vol. 49, No. 25, 2010 5179
and its required nucleotide (ADP) were omitted from the reaction
mix so that the direct production of pyruvate by PEPCK could
be monitored. Each cuvette contained 50 mM HEPES (pH 7.5),
10 mM DTT, 4 mM MgCl , 500 μM GTP, 200 μM MnCl ,
2 2
150 μM NADH, 30 units of LDH, and 350 μM OAA. Each assay
was initiated by the addition of OAA that was subsequently
varied to yield a Michaelis constant. The reaction mixture was
preincubated at 25 °C for 5 min prior to addition of OAA.
Assay D (OAA þ GDP f pyruvate þ GDP þ CO ). The
2
PEPCK-catalyzed decarboxylation of OAA without subsequent
phosphoryl transfer was conducted under reaction conditions
identical to those of assay C except that 500 μM GDP was sub-
stituted for GTP (18). A Michaelis constant for OAA was deter-
mined by varying the OAA concentration. Each assay was started
by the addition of OAA to each cuvette. The reaction mixture
was preincubated at 25 °C for 5 min prior to the addition
of OAA.
Inhibitor Assays. Inhibition experiments with oxalate and
PGA were conducted using assay B. A Michaelis constant for
PEP was determined at several different concentrations of each
inhibitor. The inhibition by βSP was carried out using assay
A because βSP is a substrate for MDH. A Michaelis constant for
OAA was determined at several concentration of βSP.
Data Analysis. All kinetic data were analyzed using Sigma-
Plot 11. In assays A-D, the slope of the line was used to calculate
a rate of enzyme activity. Because of the production of pyruvate
during the OAA f PEP reaction (assay A) in the A467G mutant,
the data were corrected for this activity using the KM and Vmax
values obtained for the enzyme from assay C for the A467G
enzyme. These two values were used in eq 1 to calculate the
theoretical rate of catalyzed pyruvate formation at each substrate
concentration used in assay A for the A467G enzyme. This
calculated rate was subtracted from the observed activity after
correction for nonenzymatic, spontaneous OAA decarboxyla-
tion as described above. All kinetic data were fit nonlinearly to
the Henri-Michaelis-Menten relationship (eq 1) to determine
KM and Vmax values. A kcat value was calculated from the Vmax
using a molecular mass of 69415 Da. For the inhibition studies,
the rate of the enzyme-catalyzed reaction was calculated from the
slope of the line for all data points. For the inhibition of PEPCK
by PGA and oxalate, the rate data were plotted together and
globally fit to eq 1, resulting in a singular value for Vmax and KM,
F
IGURE 2: Binding isotherms illustrating the binding of ITP to (A)
wild-type and (B) A467G PEPCK as measured by intrinsic protein
fluorescence quenching and described in Materials and Methods.
K
M
K
M;app
¼
½Iꢀ þ K
M
ð2Þ
Ki
app
values at each concentration of inhibitor. The resulting KM,app
K
M
values were replotted against the concentration of inhibitor
giving a linear fit to eq 2 that was used to calculate the K value
for each inhibitor. For βSP, the data at each concentration of
K
M;app
Vmax
K
M
i
¼
½Iꢀ þ
ð3Þ
2
V
max;app
K
i
V
max
inhibitor were fit individually to eq 1, resulting in KM,app and
Vmax,app values. From these values, the ratio KM,app/Vmax,app was
calculated, and these values were plotted as a function of βSP
where v is the initial velocity, Vmax is the maximal velocity
achieved, [S] is the concentration of substrate at each velocity,
KM is the Michaelis constant for each respective substrate, [I] is
concentration and fit to eq 3 to determine the K value for the
i
competitive binding effect.
the inhibitor concentration, and K is the inhibition constant.
i
Fluorescence Quenching. Fluorescence quenching experi-
ments were performed using a steady-state PTI instrument with
temperature control. Each assay was conducted in a triangular
cuvette at 15 °C with constant stirring. The samples contained
V
max½Sꢀ
v ¼
ð1Þ
K
M
þ ½Sꢀ
2
In our hands, βSP is observed to act as a mixed inhibitor of PEPCK
5
0 mM HEPES (pH 7.5), 10 mM DTT, 1 mM MnCl , 200 μM
2
against OAA that we have identified is due to contamination of our βSP
preparations with low levels of chloride. Structural studies of the
PEPCK-βSP complex demonstrate that this results in weak binding
of chloride near the active site of PEPCK and leads to the observed
mixed inhibition pattern (T. A. Johnson and T. Holyoak, unpublished
data). For the purposes of this study, we report only the K values on the
i
slope representing the competitive binding of βSP to the OAA site of
KCl, and PEPCK (1-5 μM). The emission was monitored at
330 nm after excitation at 295 nm. The sample was titrated with
an identical solution, which also contained IDP or ITP. Inosine
nucleotides were substituted for the guanosine nucleotides uti-
lized in the kinetic studies to reduce inner filter effects. The
binding constants and quenching maxima were determined by a
PEPCK.

5
180 Biochemistry, Vol. 49, No. 25, 2010
Johnson and Holyoak
a
Table 2: Characterization of Wild-Type and A467G Cytosolic PEPCK
(
A) OAA þ GTP f PEP þ GDP þ CO
2
-1
-1
K
M
(μM)
k
cat/K
M
(M
s )
-1
enzyme
OAA
GTP
k
cat (s
)
OAA
GTP
6
5
5
wild-type
A467G
52 ( 6
749 ( 67
68 ( 4
47 ( 17
54 ( 0.2
14 ( 0.1
1.0 ꢁ 10
7.9 ꢁ 10
2.7 ꢁ 10
4
1.9 ꢁ 10
(
B) PEP þ CO þ GDP f OAA þ GTP
2
-1
-1
K
M
(μM)
k
cat/K
M
(M
s
)
-1
enzyme
PEP
GDP
CO
2
k
cat (s
)
PEP
GDP
CO
2
4
4
5
4
wild-type
A467G
294 ( 16
63 ( 11
39 ( 2
70 ( 8
1194 ( 109
814 ( 98
19 ( 0.8
1 ( 0.1
6.6 ꢁ 10
1.5 ꢁ 10
5.1 ꢁ 10
1.6 ꢁ 10
1.2 ꢁ 10
4
3
1.4 ꢁ 10
(
C) Pyruvate Production during the OAA þ GTP f PEP þ GDP þ CO
2
Reaction
-1
-1 -1
enzyme
K
M
(OAA) (μM)
k
cat (s
)
k
cat/K
M
(M
s )
b
b
b
wild-type
A467G
ND
ND
ND
3
512 ( 43
2.6 ( 0.1
5.0 ꢁ 10
(
D) Decarboxylation Half-Reaction (OAA þ GDP f pyruvate þ GDP þ CO
2
)
-1
-1 -1
enzyme
K
M
(OAA) (μM)
k
cat (s
)
k
cat/K
M
(M
s
)
4
wild-type
A467G
117 ( 11
970 ( 118
2.3 ( 0.1
1.3 ( 0.1
2.0 ꢁ 10
1.0 ꢁ 10
3
(
E) Substrate and Substrate Analogue Affinities
enzyme
K
D
(ITP) (μM)
K
D
(IDP) (μM)
K
I
(βSP) (μM)
K
I
(PGA) (μM)
K
I
(oxalate) (μM)
wild-type
A467G
a
0.34 ( 0.01
0.50 ( 0.03
11.6 ( 0.4
8.0 ( 0.3
26 ( 1
50 ( 3
2530 ( 13
1460 ( 48
106 ( 6
860 ( 33
b
All experiments were performed at 25 °C. No pyruvate formation detected.
nonlinear least-squares fit of the titration data to eq 4 (IDP) or
eq 5 (ITP).
RESULTS
Kinetic experiments were performed to determine the effect of
the mutation upon the kinetic constants for the reaction in the
direction of both PEP formation and OAA formation as well as
the OAA decarboxylase activity (Table 2).
Kinetic Characterization of Recombinant WT and
A467G PEPCK. (i) OAA f PEP. In the physiological
direction, where OAA is converted to PEP, a mixed metal assay
Q ¼ Q
L
M
ꢂꢀ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiꢁ
ꢃ
2
K
D
þ L
T
þ P
T
-
ðK
D
þL
T
þP
T
Þ - 4P
T
L
T
=ð2P Þ
T
ð4Þ
(
assay A) was utilized, as the background rate of nonenzymatic
where Q is the quenching at each ligand concentration, Q is the
maximal quenching observed, KD is the dissociation constant of
the complex, LT is the ligand concentration titrated into the
L
M
OAA decarboxylation is lower in the presence of high concen-
trations of magnesium than in the presence of manganese as the
sole divalent cation. Utilizing this assay, A467G PEPCK has a
sample, and P is the protein concentration.
T
1
4-fold higher KM value for OAA than WT (Table 2A). This
increase in KM is coupled with a reduction in kcat (26% of the WT
value), resulting in a reduction in catalytic efficiency (kcat/K ) of
M T
Q L
Q
L
¼
ð5Þ
K
D
þ L
T
M
nearly 2 orders of magnitude (1.9% of the WT value). Con-
versely, there is little change in the KM for GTP (factor of 1.4),
resulting in the kcat/KM,GTP decreasing by a factor of <3. As
described below, A467G produces pyruvate during the turnover
of OAA to PEP. The data for the formation of PEP that is
coupled to pyruvate kinase and lactate dehydrogenase were
corrected for this pyruvate formation activity.
(ii) PEP f OAA. In the reverse direction, PEPCK catalyzes
the formation of OAA from PEP (assay B). The A467G mutant
exhibited a decrease in the KM value for PEP (21% of the WT
where Q is the quenching at each ligand concentration, Q is the
L
M
maximal quenching observed, KD is the dissociation constant of
the complex, and L is the ligand concentration titrated into the
T
sample. Equation 5 was utilized for the ITP binding data as poor
fits to the data were obtained using eq 4. We presume that these
poor fits were the result of contamination of the ITP solution
with low levels of IDP that did not allow for a discrete treatment
of the distribution of species using eq 4. Representative binding
isotherms for WT and A467G are shown in Figure 2.

Article
Biochemistry, Vol. 49, No. 25, 2010 5181
F
IGURE 4: Structure of molecule A of the A467G PEPCK-
oxalate-GTP complex. The partial occupancy (0.7) of the closed
F
IGURE 3: Structures ofthe OAA and PEP substratesand the enolate
lid in this molecule is illustrated by the 2F
1
o
c
- F density rendered at
intermediate alongside the corresponding analogue utilized in these
studies.
.3σ as a blue mesh with the corresponding lid domain rendered as a
green stick model colored by atom type. The active site manganese
ion and the bound oxalate molecule are rendered as a gray sphere and
green stick model, respectively. The location of the A467G mutation
is also labeled. Figures were generated using CCP4MG (47).
value) (Table 2B). In contrast, the A467G mutation had a kcat
reduced to 5% of the WT value. The combination of a decrease in
KM and a decrease in kcat results in a reduction in the catalytic
efficiency of A467G relative to WT by only a factor of 4. In
contrast, because of a slight increase in the KM value for GDP,
kcat/KM for GDP decreases to 3% of the WT value.
has little effect on the relative affinity of PEPCK for the nucle-
otide substrates. This conclusion was confirmed by the direct
measurement of the KD values for IDP and ITP by intrinsic pro-
tein fluorescence quenching experiments. The inosine nucleotides
were substituted in place of guanosine to reduce the inner filter
effect observed when guanosine nucleotides are utilized. PEPCK
utilizes inosine and guanosine nucleotides with similar catalytic
efficiency, while ITC experiments with the human enzyme have
demonstrated the exocyclic C-2 amino group of guanosine nucle-
otides increases the affinity of the enzyme for guanosine nucleo-
tides by approximately 1 order of magnitude compared to that of
the inosine analogue (31). As shown in Table 2E, WT PEPCK
has a KD for ITP of 0.34 μM while A467G has a KD of 0.50 μM.
Meanwhile, the dissociation constants for IDP are 11.6 and
8.0 μM for WT and A467G, respectively.
(
iii) Pyruvate Formation. It has been well documented that
PEPCK will catalyze the production of pyruvate from OAA
18, 27, 29, 30). With rat cPEPCK at 25 °C, this activity is
(
dependent on the presence of GDP (assay D). Upon comparison
of the mutant and WT enzymes, A467G has a KM value 7-fold
higher than that of WT, with a modest reduction in kcat (57% of
the WT value). This results in the WT enzyme being 20-fold more
efficient at catalyzing the decarboxylation of OAA than A467G
(
Table 2D).
As the A467G mutation was designed to reduce the favor-
ability of the enzyme adapting the lid-closed state and to test our
proposed role for the lid in protecting the enolate intermediate,
once formed, from favorable protonation (vide infra), we in-
vestigated the ability of the enzyme to catalyze the formation of
pyruvate during the catalyzed formation of PEP from OAA
(vi) Structures of A467G. To demonstrate that the substitu-
tion of A467 for glycine decreases the thermodynamic favor-
ability of the enzyme adapting the closed lid conformation but
does not alter the structure that the Ω-loop adapts upon closing,
we determined the structure of A467G PEPCK in complex with
βSP, PGA, oxalate, and either GTP or GDP (Table 1). Identical
structural studies of the WT enzyme demonstrated that as the
enzyme commits to catalysis the thermodynamic favorability of
the enzyme adapting the closed lid conformation increases (9).
In the βSP-GTP and PGA-GDP structures, mimicking the
Michaelis complexes for the forward and reverse reactions, res-
pectively, the WT enzyme contains two molecules in the ASU,
with one adapting a closed lid conformation and the other one
remaining open. In the oxalate-GTP complex with the WT
enzyme, the lid was always found to be in a closed conformation.
In contrast to those results, and consistent with the design of the
mutation, the structures of A467G in the βSP-GTP and
PGA-GDP complexes were found to adapt only the open
lid conformation, consistent with a destabilization of the closed
(
assay C). In the WT enzyme, there was no measurable formation
of pyruvate at 25 °C. However, the A467G mutant showed a
measurable amount of pyruvate formation, albeit with a KM
value similar to the elevated KM observed in both the decarbox-
ylation reaction and the overall reaction in the direction of PEP
formation, both described above (Table 2C).
(
iv) Inhibition of WT and A467G PEPCK by Substrate
Analogues. The competitive inhibition of PEPCK by substrate
analogues βSP and PGA and the enolate mimic oxalate has been
illustrated previously on both mitochondrial and cytosolic isoforms
of PEPCK (7, 26). The kinetic and structural data demonstrate that
all three inhibitors are excellent mimics of the respective substrate or
intermediate (Figure 3). We utilized the binding constants deter-
mined from the inhibition experiments to assess whether the A467G
mutation had any effect on substrate and enolate affinity. Both of
the substrate analogues, βSP and PGA, have similar affinity for WT
and A467G PEPCK (Table 2E). In contrast, the reduction in the
affinity of oxalate with A467G (a factor of approximately 8)
suggests that by extrapolation the lid mutation has decreased the
stabilization of the enolate intermediate by ∼5.3 kJ/mol.
2þ
lid state. In a similar fashion, the A467G-Mn -oxalate-
2þ
Mn GTP structure also contains two molecules in the ASU in
which molecule B possesses an open lid conformation. Molecule
A was found to partially adapt a closed lid conformation that was
manually adjusted to an occupancy of 70% (Figure 4). This
results in an overall occupancy of the closed lid conformation for
(
v) Nucleotide Binding by Fluorescence Quenching. The
KM values for GDP and GTP suggest that the A467G mutation

5
182 Biochemistry, Vol. 49, No. 25, 2010
Johnson and Holyoak
the crystal of 35%, compared to 100% in the WT enzyme. Upon
comparison of the closed lid conformation observed in molecule
A with the conformation of the closed lid in the WT PEPCK-
oxalate-GTP structure (PDB entry 3DT4), we observed that the
substitution of glycine for alanine does not change the low-energy
conformation of the closed lid (Figure 4), consistent with the
observed kinetic and binding effects resulting solely from the
perturbation of the thermodynamic favorability of the enzyme
adapting the closed lid conformation necessary for catalysis.
DISCUSSION
The motional properties of mobile loop regions have been
demonstrated in many diverse systems to be essential to enzyme
function (15, 16, 32-39). In particular, the Ω-loop lid domain of
TIM has been well characterized, and it has been demonstrated
that removal of a portion of the lid leads to an increase in the
production of the reaction side product methylglyoxal. Further,
the removal of a portion of the lid in TIM negatively impacts
catalytic function, as the lid is no longer able to stabilize the inter-
mediate or the transition states that precede or follow its form-
ation (40). In contrast, mutations in the hinge regions of the
Ω-loop of TIM, designed to increase the motional freedom of the
lid while negatively impacting catalytic function, do not result in
an increase in the production of methylglyoxal (32). In general,
these studies on TIM have demonstrated that the functional
nature of this mobile loop domain is an exquisite balance between
its rigidity and its flexibility, a property that may be a general
feature of all catalytically important mobile loops (16, 32, 37).
While EPR studies on S-adenosylmethionine synthetase (33)
suggest that the rates of open and closed conformation sampling
are identical in the presence and absence of substrates, NMR
studies on TIM (15, 16) as well as the crystallographic data on
TIM are consistent with ligand association influencing the most
stable, most highly populated conformation of the lid domain.
The latter data are also consistent with the recent structural
studies of PEPCK (9).
F
3
IGURE 5: Comparison of the active sites of WT (PDB entries 3DT2,
DT7, and 3DTB) and A467G PEPCK (PDB entries 3MOE, 3MOF,
2þ
2þ
and 3MOH) in the PEPCK-Mn -βSP-Mn GTP [(A) WT
2þ
2þ
and (B) A467G], PEPCK-Mn -PGA-Mn GDP [(C) WT and
2þ
2þ
(D) A467G], and PEPCK-Mn -oxalate-Mn GTP [(E) WT and
(F) A467G] complexes. All catalytic residues are rendered as ball-
and-stick models colored by atom type, whereas the ligands and nucl-
eotides (βSP, PGA, oxalate, GTP, and GDP) are rendered as thick
sticks colored by atom type and labeled. Potential hydrogen bonds
between the ligands and the active site residues are represented as
black dashed lines. The active site and nucleotide-associated manga-
nese ions are rendered as gray spheres, and the lid domain is colored
dark red.
While PEPCK and TIM catalyze two drastically different
chemical reactions, they both do so via unstable intermediate
enolate species that rapidly decompose in solution. On the basis
of the commonality of a lid domain, the simplest interpretation is
that the mobile lid is necessary to allow ligand association or
dissociation but during catalysis is required to provide an enviro-
nment that sequesters and/or stabilizes the reactive intermediate
and gives the enzyme the ability to follow a chemical pathway
that is not available in solution due to the instability of the inter-
mediate. Unlike the phosphate “gripper” lid domain of TIM, in
PEPCK the closed lid makes no direct contacts with the bound
substrates or the enolate intermediate. Therefore, in PEPCK it is
not possible that new interactions between the lid and the
substrates directly stabilize the lid in the closed state. In light of
this observation, an alternative mechanism for describing the
energetic driving force behind the transition between open and
closed lid states is needed. In general, our model of lid dynamics
in PEPCK proposes that not all of the free energy from substrate
association is used in the formation of a tight binding E-S
complex but that some of that energy is partitioned to the protein.
This input energy modifies the free energy landscape that defines
the conformation of the protein and results in the closed lid state
becoming more thermodynamically favorable as substrates as-
sociate (6, 9). This general idea has been presented previously
PEPCK is presently unknown, both the kinetic and structural
data presented in this work are consistent with this model. To
understand the role of the lid domain in the reaction mediated by
PEPCK, we have undertaken structural and kinetic studies of a
mutant of the enzyme that was designed to alter the equilibrium
of the open-closed transition.
Rationale behind the A467G Mutation. To test the model
of lid dynamics described above, we desired to increase the
entropic penalty for lid closure while keeping the available inter-
action energy fixed (i.e., using the same ligands). To achieve this
goal, we postulated that we could engineer a modified lid domain
that, while still operating as a singular rigid domain of an Ω-loop
(13), has an increased entropic penalty for lid ordering and
closure through the introduction of a conformationally more
flexible glycine residue (Figure 1). We decided that the mutation
of A467 to glycine would provide such an engineered change.
Upon expression and purification of this mutation (A467G), we
undertook structural and functional studies to determine if the
mutation provided data that were consistent with our model of
loop function.
Kinetic and Structural Characterization of the A467G
Mutation. (i) Structures of the Michaelis and Enolate
Intermediate-like Complexes. The A467G PEPCK structures
(41, 42). While the mechanism of this free energy partitioning in

Article
Biochemistry, Vol. 49, No. 25, 2010 5183
F
IGURE 6: Minimal kinetic schemes for the PEPCK-catalyzed decarboxylation of OAA in the presence of GDP. In red are shown those steps
o c
corresponding to the different steady-state kinetic complexes and constants. E and E represent the lid-open and lid-closed forms of the enzyme,
respectively.
mimicking the two Michaelis complexes and the enolate inter-
mediate state, which were determined under conditions identical
to those used for previous complexes with the wild-type enzyme
We are well aware that these schemes are an oversimplification.
A simple two-state model would result in an equilibrium defining
the open and closed lid conformations of each kinetic complex,
and a more realistic interpretation would be a reaction coordinate
surface that defines the ensemble of protein states in the different
kinetic complexes (43, 44). However, we think the lowest-energy
trajectory illustrated by the simple two-dimensional reaction co-
ordinate and the schemes is reasonable for the purpose of our
interpretation of the data.
(iii) Lid Closure Stabilizes the Enolate Intermediate. In
the WT PEPCK-oxalate-GTP complex, Y235 is observed to
occupy a rotomeric position that positions the phenolic side chain
forward in the active site toward oxalate and the active site
metal (9). In contrast, in molecule B of the A467G mutant struc-
ture, even though oxalate is bound at the active site, Y235
occupies a predominantly rearward rotomeric state. This state
is observed as the predominant conformation for Y235 in other
lid-open complexes of PEPCK. The coincident nature of lid
closure and the stabilization of the forward rotomer of Y235 in
the presence of oxalate are supported by the structure of molecule
A of the A467G mutant. Molecule A possesses a mixture of lid
open and closed states and also has a mixed occupancy of the
forward and rearward rotomers of Y235. While the forward
rotomer of Y235 cannot directly H-bond with the enolate, the
altered environment created by a shift in the lid to prefer the open
state in A467G PEPCK and the resulting shift in the low-energy
(9), illustrate that no gross structural changes are induced by the
lid mutation. Further, with the exception of the rotomeric state of
Y235 in the A467G-oxalate-GTP complex discussed below, the
enzyme-substrate contacts in the A467G complexes are identical
to those in the open complexes of the WT enzyme previously
determined (Figure 5). However, consistent with the design of the
mutation, the introduction of the glycine linker has resulted in the
lid-closed state being less thermodynamically favorable (Table 1).
This results in the closed lid state only being observed at a low
level in the A467G-oxalate-GTP complex, and not at all in the
two Michaelis-like complexes (A467G-βSP-GTP and A467G-
PGA-GDP). This contrasts identical experiments with the WT
enzyme in which the Michaelis-like complexes have a 50-50
distribution of lid-open and lid-closed states and the oxalate-
GTP complex possesses only a closed lid (9). Importantly, the
population of molecules in the A467G-oxalate-GTP struc-
ture that do adapt the closed conformation demonstrates that
the substitution of glycine for alanine has not changed the low-
energy conformation of the lid and confirms that we can interpret
our kinetic data in the context of the mutation functioning by
disrupting the low-energy lid state and not having other structur-
al effects (Figure 4). Unfortunately, the crystallographic studies
do not allow us to determine whether the shift in the equilibrium
for the lid conformation towardthe open state is due to a decrease
in the closing rate or an increase in the opening rate. Previous
studies with TIM have shown that hinge mutations in the Ω-loop
that result in a lid with more entropic freedom do not lead to an
increased rate of lid opening but do decrease the frequency of lid
closure (16). Therefore, further studies on PEPCK will be neces-
sary to see what effect increasing the entropy of the lid domain
has on the lid opening and closing rates.
position of Y235 appear to result in an increased K being obser-
i
ved for the A467G mutant (Table 2E). An explanation for this
loss of binding energy (∼5.3 kJ/mol) could be that the rearward
Y235 rotomer is no longer able to provide an anion-quadrapole
interaction with oxalate, an interaction that has been previously
suggested to be important for interactions of PEP with PEPCK
based upon an increased KM for PEP in Y235 mutants of human
cPEPCK (29). A related observation is that in the open A467G-
oxalate-GTP complex, the rearward Y235 rotomer coincides
with an altered hydration state of the bound oxalate molecule
(Figure 8). This altered hydration state may contribute to the
weakened affinity of the enzyme for the enolate or oxalate; how-
ever, the change in water structure could account for the form-
ation of pyruvate in A467G such that the repositioned water
(
ii) Kinetic Consequences of the Disruption of Lid Dyn-
amics. On the basis of the previously published structural data
demonstrating that PEPCK follows an induced-fit mechanism
that results in the true “E-S” complex not being formed until the
lid closes (9), we propose the schemes illustrated in Figures 6 and
7, to interpret the kinetic data for the WT and A467G enzymes.

5
184 Biochemistry, Vol. 49, No. 25, 2010
Johnson and Holyoak
F
IGURE 7: Minimal kinetic schemes for the PEPCK-catalyzedreactioninthe (A) OAA f PEP direction and (B) PEP f OAA direction. In red are
o c
shown those steps corresponding to the different steady-state kinetic complexes and constants. E and E represent the lid-open and lid-closed
forms of the enzyme, respectively.
molecules create an environment in which they are now able to
protonate the enolate intermediate, a feature not observed in the
WT enzyme at 25 °C. It is presently unknown whether the enolate
is protonated on the enzyme or is released and protonated in
solution. However, as pyruvate only weakly associates with
PEPCK [K = 9 mM (26)], the protonation of the tightly bound
i
enolate would provide a mechanism for its release after proton-
ation of the enzyme. Upon examination of the kinetic data, the
decrease in enolate stabilization by the shift in the equilibrium
position of the lid toward an open conformation and its subseq-
uent effect on the low-energy conformation for Y235 and the
solvation of the active site have the effect of increasing the activ-
ation barrier for both phosphoryl transfer and the carboxylation-
decarboxylation steps (Table 2 and Figure 9). The reduction in
oxalate affinity and, by extrapolation, enolate affinity by a factor
of ∼8 translates into a reduction in the decarboxylation rate
observed in the presence of OAA and GDP to approximately 57%
of the WT rate and a reduction in the kcat/KM for this process by
more than 1 order of magnitude (Table 2D). If, as discussed below,
in the direction of OAA synthesis (Table 2B) kcat represents the
F
IGURE 8: Changes in the rotomeric state of Y235 and the water
structure surrounding oxalate in the (A) lid-closed WT PEPCK-
2þ
2þ
2þ
Mn -oxalate-Mn GTP (9) and (B) lid-open A467G-Mn
-
2þ
oxalate-Mn GTP complexes. In panel A, the dashed lines repre-
sent potential hydrogen bonds with lengths of (a) 2.7, (b) 2.4,
˚
(c) 2.5, (d) 2.7, (e) 2.8, (f) 2.6, and (g) 3.0 A. Similarly, in panel B,
˚
the potential hydrogen bond lengths are (a) 3.3,A (b) 2.5, (c) 2.7,
˚
and (d) 3.4A. On the basis of structural work by Cotelesage et al. (48),
we propose that the water molecules separated by distance c in panel
A define the CO
site for CO is only present in the lid-closed conformation shown in
panel A.
2
binding site. As illustrated in panel B, this binding
2

Article
Biochemistry, Vol. 49, No. 25, 2010 5185
FIGURE 9: Cartoon reaction coordinatefor WT (bluesolid line) and A467G (orange dashed line) PEPCK based uponthe differentialinhibitionof
the two enzyme forms by the enolate intermediate analogue, oxalate.
phosphorylation of GDP by PEP to form the enolate intermediate,
this reduction in oxalate affinity would translate into a reduction
in the phosphoryl transfer rate to 5% of that for the WT enzyme
Previous structural studies of the PEPCK-OAA-GDP complex
demonstrate that this ligation state of the enzyme has an
increased propensity to form the closed lid conformation, similar
to the PEPCK-βSP-GTP and PEPCK-PGA-GDP Michaelis
complex mimics (8). (2) With the rat cPEPCK, at 25 °C, the
decarboxylation activity is observed only in the presence of dip-
hosphate nucleotide.
(
Table 2B). Interpreted in this way, the structural data for
2þ
2þ
the A467G-Mn -oxalate-Mn GTP complex agree with the
kinetic data. If the intermediate destabilization (∼5.3 kJ/mol) is
fully expressed in either transition state, this would result in a
reduction in the rate of catalysis by a factor of 10, a value similar in
magnitude to the results of the steady-state data presented above
Both of these pieces of data are consistent with the idea that lid
closure is required to initiate the decarboxylation reaction. Our
model for the decarboxylation activity is one, similar to forma-
tion of the Michaelis complexes, in which formation of the
PEPCK-OAA-GDP complex results in a shift in the open-
-closed equilibrium to the right and this leads to the generation
(Table 2). It should be noted that absolute discrepancies between
the oxalate affinity data and the kinetic data could be the result of
not knowing the extent to which oxalate mimics the actual enolate
intermediate [similar to studies of TIM (40)], and therefore, the
change in oxalate affinity is only a qualitative measure of the pot-
ential change in enolate affinity. Further, we do not know whether
the same destabilization of the enolate intermediate would be
fully expressed in either or both of the two transition states.
Finally, the exact nature of the rate-limiting step(s) in WT and
A467G is unknown. Our interpretation does provide a reason-
able explanation of how the destabilization of the closed lid
conformation leads to both a defect in catalysis and the propen-
sity to generate pyruvate.
of the enolate and CO . However, as there is no phosphoryl
2
donor present to phosphorylate the intermediate, pyruvate is
formed and released the next time the lid samples the open
conformation. Therefore, on the basis of the scheme illustrated in
Figure 6, the observed increase in the KM value could result from
either an increase in k , kopen, or koff or a decrease in kon or kclose
.
2
As the observed reduction in k from the decarboxylation data is
2
in the wrong direction to cause an increase in K , an alternate
M
explanation is needed. The inhibition of PEPCK by βSP suggests
that kon and koff are only modestly affected, since the binding of
βSP in WT and A467G differs by a factor of only ∼2. This
modest effect on binding could be the result of altering lid
dynamics in the PEPCK-GTP or -GDP state (Figure 6) such
The interpretation of the kinetic data given above also relies
on kcat not being limited by the lid open/close rate in PEPCK.
This would be in contrast to TIM. TIM, however, operates at a
much higher catalytic efficiency, and the open-closed transition
of the phosphate gripper loop occurs on that same time scale
that the altered opening and/or closing rates affect k /k as the
on off
4
-1
(
∼10 s ) (15, 45). This rate is approximately 3 orders of magni-
structural studies demonstrate that binding of substrates to
PEPCK cannot occur after lid closure (9). This type of perturbed
lid behavior, prior to formation of the Michaelis complex, has
also been used to explain the kinetic data of hinge mutations of
TIM (16). Therefore, the structural and kinetic data are consis-
tent with the observed increase in KM resulting from either a
reduction in the rate of the transition to the closed conformation
or an increase in the frequency of lid opening, which results in the
open-closed equilibrium shifting to the left, resulting in an
elevated KM value for OAA.
tude faster than the k in PEPCK, supporting our assumption
cat
that kcat does not represent the open-closed transition in
PEPCK. Further, as the effect of A467G on kcat is larger in the
direction of OAA synthesis than in the direction of PEP synthesis,
the asymmetry provides additional support for the possibility
that the phosphoryl transfer reaction is at least partially rate
limiting, consistent with the reaction coordinate diagram in
Figure 9.
(
iv) Decarboxylation of OAA. In addition to the modest
change in the decarboxylation rate observed in the decarboxyla-
(v) OAA f PEP Reaction. Similar to what is discussed
above for the decarboxylation half-reaction, a modest reduction
in kcat (26% of the WT rate) and an increase (14-fold) in the KM
for OAA are observed. On the basis of the interpretation
provided above, we again conclude that the change in the KM
value is due to either a decrease in the rate of lid closure or an
tion half-reaction, there is a modest increase (∼8-fold) in the K
M
for OAA in this reaction (Table 2D). As illustrated in Figure 6,
this KM value represents both binding of OAA to the
PEPCK-GDP complex and the opening and closing of the lid.
This conclusion is reached on the basis of two pieces of data. (1)

5
186 Biochemistry, Vol. 49, No. 25, 2010
Johnson and Holyoak
increase in the rate of lid opening, resulting in the equilibrium for
the closed lid conformation shifting to the left which is consistent
with both the design of the mutation and the structural data.
precedes enolate formation, and therefore, if as expected carbox-
ylation (k ) is significantly faster than the reopening rate, this
3
would result in the absence of pyruvate production in this dir-
ection (Figure 7B). Alternatively, the high concentrations of CO2
present in the reaction assay in this direction could result in an
Again, assuming that kcat ∼ k [the phosphorylation of the eno-
3
late by GTP (Figure 7A)], we conclude that this shift in the
favorability of lid closing has a modest effect on the activation
energy for the reaction as illustrated in Figure 9 and described
above. The observation of the formation of pyruvate by the
enzyme during turnover to PEP suggests that, while in the wild-
2þ
2þ
A467G-Mn -enolate-CO -Mn GTP complex that alters
2
the solvation of the enolate intermediate in much the same way as
the altered rotomeric state of Y235 that coincides with the lid
open enzyme does. On the basis of this interpretation, even though
the lid-closed conformation of PEPCK is destabilized by the
A467G mutation, the water structure in the A467G-enolate-
type enzyme k is fast relative to k
(no formation of pyruvate
3
reopen
observed), either the reduction in the phosphoryl transfer rate,
through the destabilization of the enolate, or an increase in the
rate of lid opening results in the rate of phosphoryl transfer
becoming closer to the rate of opening, such that now for every
five PEP molecules produced one pyruvate molecule is formed
CO -GTP complex is not conducive to enolate protonation.
2
This model would require that the enolate complex arising from
the decarboxylation of OAA rapidly lose CO during the transi-
2
2þ
2þ
tion to the closed A467G-Mn -enolate-Mn GTP complex.
This would be consistent with the weak affinity of the enzyme for
CO . In this direction, the resulting A467G-Mn -enolate-
Mn GTP complex would possess a water structure similar to
the A467G-Mn -oxalate-Mn GTP structure that we pro-
pose allows for protonation of the enolate (Figure 8B). Arguing
against this latter model is the observation that when using assay
(
Table 2 and Figure 7A). To address the possibility that a slowing
2þ
of the phosphoryl transfer rate results in the observed production
of pyruvate, we conducted kinetic studies with WT PEPCK using
2
2
þ
0
2þ
2þ
the slow substrate guanosine 5 -O-(γ-thio)triphosphate (GTPγS).
Consistent with previous studies using ITPγS (46), the use of
-
1
GTPγS reduces kcat (13 s ) and the KM for GTPγS is 39 μM.
Using GTPγS, we did not observe any pyrvuate formation during
the OAA f PEP reaction. Therefore, the data suggest that
making phosphoryl transfer solely or more rate-limiting does
not result in any decoupling of the reaction to pyruvate and that
simply slowing the phosphoryl transfer rate in the absence of a
destabilization of the lid-closed conformation is not sufficient to
decouple the reaction.
conditions with subsaturating concentrations of CO we are still
2
unable to detect pyruvate formation in the PEP f OAA direction
(data not shown). On the basis of all of the information, we
suggest it is most likely that the competing rates between phos-
phoryl transfer and lid reopening result in the decoupling of the
PEPCK reaction with the A467G mutation; however, more work
is obviously necessary to clarify the mechanism by which A467G
produces pyruvate during the OAA f PEP conversion and the
lack of this activity in the opposite direction.
(
vi) PEP f OAA Reaction. In contrast to the decarboxyla-
tion reaction and the reaction in the direction of PEP formation,
the reaction in the direction of OAA formation results in a
reduction in the KM for PEP (∼21% of the K for WT) and a
M
CONCLUSION
more drastic reduction in k (∼5% of that for WT). Again, we
cat
Perhaps not surprisingly, the lid domain of PEPCK has a
general function similar to that of the lid domain found in TIM. It
operates both to sequester the reactive intermediate from alter-
nate chemistries and to stabilize the enolate intermediate. In
contrast to TIM, this stabilization of the enolate by the closed lid
facilitates the chemical conversion of substrates to products with-
out the enzyme directly interacting with the substrates. The
structural and kinetic characterization of A467G is consistent
with the mutation causing a perturbation in the equilibrium
position of the lid toward the open state, consistent with the mut-
ation increasing the conformational entropy of the domain and
resulting in the decoupling of the reaction through the formation
of pyruvate. In addition, this shift in the conformational equi-
librium of the lid domain results in the loss of the Y235-enolate
interaction which hampers the ability of the enzyme to stabilize
the bound enolate and subsequently the transition states that
precede and follow its formation and results in catalytic defect.
can interpret the data in the context of the model presented in
Figure 7B. On the basis of the inhibition data using the PEP
analogue PGA, we see that the decrease in the KM value with
A467G is not due to solely to an increase in the affinity of the
enzyme for PEP as the affinity of the enzyme for PGA increases
by a factor of <2 (Table 2E). In the direction of OAA synthesis,
rather than lid closure initiating decarboxylation and the forma-
tion of the enolate, in this direction (PEP f OAA) lid closure
stimulates the transfer of a phosphoryl group from PEP to GDP,
forming the enolate intermediate since lid closure is necessary for
PEP to adopt a kinetically competent conformation for phos-
phoryl transfer (9). On the basis of the standard free energies for
the reaction, we envision this to be a more energetically costly
reaction (Figure 9) (6). As a consequence, the destabilization of
the enolate has a more drastic effect on the activation energy
barrier for the reaction in this direction and results in kcat being
reduced to 5% of the WT value. This reduction in kcat (k ) leads
2
to the observed reduction in K , as the observed K value for
M
M
ACKNOWLEDGMENT
A467G shifts in a direction toward the true KD for PEP. As the
other kinetic and structural data presented here suggest that lid
opening and closing are perturbed by the A467G mutation, the
true K for PEP should be lower than the K of 60 μM observed
We thank Dr. Aron Fenton and Ms. Sarah Sullivan for their
helpful suggestions during the preparation of the manuscript.
D
M
as the change in the position of the open-closed equilibrium for
the lid may still result in conditions that are not at rapid
equilibrium. This interpretation is also consistent with the lack
of observed pyruvate production during turnover in this
reaction direction. Again on the basis of our free energy
diagram (Figure 9), our model and data suggest that in the
PEP f OAA direction the slow step, phosphoryl transfer,
REFERENCES
1
. Lee, M. H., Hebda, C. A., and Nowak, T. (1981) The Role of Cations
in Avian Liver Phosphoenolpyruvate Carboxykinase Catalysis. Acti-
vation and Regulation. J. Biol. Chem. 256, 12793–12801.
2
. Colombo, G., Carlson, G. M., and Lardy, H. A. (1981) Phosphoe-
0
nolpyruvate Carboxykinase (Guanosine 5 -Triphosphate) from Rat
Liver Cytosol. Dual-Cation Requirement for the Carboxylation
Reaction. Biochemistry 20, 2749–2757.

Article
Biochemistry, Vol. 49, No. 25, 2010 5187
3
4
5
6
7
. Hebda, C. A., and Nowak, T. (1982) The Purification, Characteriza-
27. Noce, P. S., and Utter, M. F. (1975) Decarboxylation of Oxalacetate
to Pyruvate by Purified Avian Liver Phosphoenolpyruvate Carbox-
ykinase. J. Biol. Chem. 250, 9099–9105.
28. Holyoak, T., and Nowak, T. (2004) pH Dependence of the Reaction
Catalyzed by Avian Mitochondrial Phosphoenolpyruvate Carboxy-
kinase. Biochemistry 43, 7054–7065.
29. Dharmarajan, L., Case, C. L., Dunten, P., and Mukhopadhyay, B.
(2008) Tyr235 of Human Cytosolic Phosphoenolpyruvate Carboxy-
kinase Influences Catalysis through an Anion-Quadrupole Interac-
tion with Phosphoenolpyruvate Carboxylate. FEBS J. 275, 5810–
5819.
30. Jabalquinto, A. M., Laivenieks, M., Zeikus, J. G., and Cardemil, E.
(1999) Characterization of the Oxaloacetate Decarboxylase and
Pyruvate Kinase-Like Activities of Saccharomyces cerevisiae and
Anaerobiospirillum succiniciproducens Phosphoenolpyruvate Carbox-
ykinases. J. Protein Chem. 18, 659–664.
tion, and Activation of Phosphoenolpyruvate Carboxykinase from
Chicken Liver Mitochondria. J. Biol. Chem. 257, 5503–5514.
. Hebda, C. A., and Nowak, T. (1982) Phosphoenolpyruvate Carbox-
2þ
2þ
ykinase. Mn and Mn Substrate Complexes. J. Biol. Chem. 257,
515–5522.
5
. Yang, J., Kalhan, S. C., and Hanson, R. W. (2009) What Is the
Metabolic Role of Phosphoenolpyruvate Carboxykinase? J. Biol.
Chem. 284, 27025–27029.
. Carlson, G. M., and Holyoak, T. (2009) Structural Insights into the
Mechanism of Phosphoenolpyruvate Carboxykinase Catalysis. J. Biol.
Chem. 284, 27037–27041.
. Stiffin, R.-M., Sullivan, S. M., Carlson, G. M., and Holyoak, T.
(2008) Differential Inhibition of Cytosolic Pepck by Substrate Ana-
logues. Kinetic and Structural Characterization of Inhibitor Recogni-
tion. Biochemistry 47, 2099–2109.
8
. Sullivan, S. M., and Holyoak, T. (2007) Structures of Rat Cytosolic
Pepck: Insight into the Mechanism of Phosphorylation and Decar-
boxylation of Oxaloacetic Acid. Biochemistry 46, 10078–10088.
. Sullivan, S. M., and Holyoak, T. (2008) Enzymes with Lid-Gated
Active Sites Must Operate by an Induced Fit Mechanism Instead of
Conformational Selection. Proc. Natl. Acad. Sci. U.S.A. 105, 13829–
31. Dunten, P., Belunis, C., Crowther, R., Hollfelder, K., Kammlott, U.,
Levin, W., Michel, H., Ramsey, G. B., Swain, A., Weber, D., and
Wertheimer, S. J. (2002) Crystal Structure of Human Cytosolic
Phosphoenolpyruvate Carboxykinase Reveals a New Gtp-Binding
Site. J. Mol. Biol. 316, 257–264.
32. Xiang, J., Jung, J. Y., and Sampson, N. S. (2004) Entropy Effects on
Protein Hinges: The Reaction Catalyzed by Triosephosphate Isomer-
ase. Biochemistry 43, 11436–11445.
33. Taylor, J. C., and Markham, G. D. (2003) Conformational Dynamics
of the Active Site Loop of S-Adenosylmethionine Synthetase Illumi-
nated by Site-Directed Spin Labeling. Arch. Biochem. Biophys. 415,
164–171.
34. Hedstrom, L., Szilagyi, L., and Rutter, W. J. (1992) Converting
Trypsin to Chymotrypsin: The Role of Surface Loops. Science 255,
1249–1253.
35. Johnson, T. A., Qiu, J., Plaut, A. G., and Holyoak, T. (2009) Active-
Site Gating Regulates Substrate Selectivity in a Chymotrypsin-Like
Serine Protease the Structure of Haemophilus influenzae Immunoglo-
bulin A1 Protease. J. Mol. Biol. 389, 559–574.
36. Perona, J. J., and Craik, C. S. (1997) Evolutionary Divergence of
Substrate Specificity within the Chymotrypsin-Like Serine Protease
Fold. J. Biol. Chem. 272, 29987–29990.
37. Watt, E. D., Shimada, H., Kovrigin, E. L., and Loria, J. P. (2007) The
Mechanism of Rate-Limiting Motions in Enzyme Function. Proc.
Natl. Acad. Sci. U.S.A. 104, 11981–11986.
38. Sampson, N. S., and Knowles, J. R. (1992) Segmental Motion in
Catalysis: Investigation of a Hydrogen Bond Critical for Loop
Closure in the Reaction of Triosephosphate Isomerase. Biochemistry
31, 8488–8494.
9
1
3834.
1
1
0. Alber, T., Banner, D. W., Bloomer, A. C., Petsko, G. A., Phillips, D.,
Rivers, P. S., and Wilson, I. A. (1981) On the Three-Dimensional
Structure and Catalytic Mechanism of Triose Phosphate Isomerase.
Philos. Trans. R. Soc. London, Ser. B 293, 159–171.
1. Banner, D. W., Bloomer, A. C., Petsko, G. A., Phillips, D. C., Pogson,
C. I., Wilson, I. A., Corran, P. H., Furth, A. J., Milman, J. D., Offord,
R. E., Priddle, J. D., and Waley, S. G. (1975) Structure of Chicken
Muscle Triose Phosphate Isomerase Determined Crystallographically
at 2.5 Angstrom Resolution Using Amino Acid Sequence Data.
Nature 255, 609–614.
1
2. Lolis, E., and Petsko, G. A. (1990) Crystallographic Analysis of the
Complex between Triosephosphate Isomerase and 2-Phosphoglyco-
˚
late at 2.5-A Resolution: Implications for Catalysis. Biochemistry 29,
6
619–6625.
1
3. Joseph, D., Petsko, G. A., and Karplus, M. (1990) Anatomy of a
Conformational Change: Hinged “Lid” Motion of the Triosepho-
sphate Isomerase Loop. Science 249, 1425–1428.
4. Fersht, A. R. (1974) Catalysis, Binding and Enzyme-Substrate Com-
plementarity. Proc. R. Soc. London, Ser. B 187, 397–407.
5. Williams, J. C., and McDermott, A. E. (1995) Dynamics of the
Flexible Loop of Triosephosphate Isomerase: The Loop Motion Is
Not Ligand Gated. Biochemistry 34, 8309–8319.
1
1
1
6. Kempf, J. G., Jung, J. Y., Ragain, C., Sampson, N. S., and Loria,
J. P. (2007) Dynamic Requirements for a Functional Protein Hinge.
J. Mol. Biol. 368, 131–149.
7. Griffith, O. W., and Weinstein, C. L. (1987) β-Sulfopyruvate. Meth-
ods Enzymol. 143, 221–223.
8. Colombo, G., Carlson, G. M., and Lardy, H. A. (1978) Phosphoe-
nolpyruvate Carboxykinase (Guanosine Triphosphate) from Rat
Liver Cytosol. Separation of Homogeneous Forms of the Enzyme
with High and Low Activity by Chromatography on Agarose-Hex-
ane-Guanosine Triphosphate. Biochemistry 17, 5321–5329.
9. Otwinowski, Z., and Minor, W. (1997) Processing of X-Ray Diffrac-
tion Data Collected in Oscillation Mode. Methods Enzymol. 276,
39. Sampson, N. S., and Knowles, J. R. (1992) Segmental Movement:
Definition of the Structural Requirements for Loop Closure in
Catalysis by Triosephosphate Isomerase. Biochemistry 31, 8482–8487.
40. Pompliano, D. L., Peyman, A., and Knowles, J. R. (1990) Stabiliza-
tion of a Reaction Intermediate as a Catalytic Device: Definition of
the Functional Role of the Flexible Loop in Triosephosphate Iso-
merase. Biochemistry 29, 3186–3194.
41. Fisher, H. F. (1988) A Unifying Model of the Thermodynamics of
Formation of Dehydrogenase-Ligand Complexes. Adv. Enzymol.
Relat. Areas Mol. Biol. 61, 1–46.
1
1
1
42. Jencks, W. P. (1975) Binding Energy, Specificity, and Enzymic
Catalysis: The Circe Effect. Adv. Enzymol. Relat. Areas Mol. Biol.
43, 219–410.
3
07–326.
2
2
2
2
0. Vagin, A., and Teplyakov, A. (1997) Molrep: An Automated Program
for Molecular Replacement. J. Appl. Crystallogr. 30, 1022–1025.
1. Bailey, S. (1994) The Ccp4 Suite: Programs for Protein Crystal-
lography. Acta Crystallogr. D50, 760–763.
2. Emsley, P., and Cowtan, K. (2004) Coot: Model-Building Tools for
Molecular Graphics. Acta Crystallogr. D60, 2126–2132.
3. Painter, J., and Merritt, E. A. (2005) A Molecular Viewer for the
Analysis of Tls Rigid-Body Motion in Macromolecules. Acta Crystal-
logr. D61, 465–471.
43. Swint-Kruse, L., and Fisher, H. F. (2008) Enzymatic Reaction
Sequences as Coupled Multiple Traces on a Multidimensional Land-
scape. Trends Biochem. Sci. 33, 104–112.
44. Benkovic, S. J., Hammes, G. G., and Hammes-Schiffer, S. (2008) Free-
Energy Landscape of Enzyme Catalysis. Biochemistry 47, 3317–3321.
45. Rozovsky, S., Jogl, G., Tong, L., and McDermott, A. E. (2001)
Solution-State NMR Investigations of Triosephosphate Isomerase
Active Site Loop Motion: Ligand Release in Relation to Active Site
Loop Dynamics. J. Mol. Biol. 310, 271–280.
2
2
2
4. Painter, J., and Merritt, E. A. (2006) Optimal Description of a Protein
Structure in Terms of Multiple Groups Undergoing Tls Motion. Acta
Crystallogr. D62, 439–450.
5. Painter, J., and Merritt, E. A. (2006) Tlsmd Web Server for the
Generation of Multi-Group Tls Models. J. Appl. Crystallogr. 39, 109–
46. Konopka, J. M., Lardy, H. A., and Frey, P. A. (1986) Stereochemical
Course of Thiophosphoryl Transfer Catalyzed by Cytosolic Phos-
phoenolpyruvate Carboxykinase. Biochemistry 25, 5571–5575.
47. Potterton, L., McNicholas, S., Krissinel, E., Gruber, J., Cowtan, K.,
Emsley, P., Murshudov, G. N., Cohen, S., Perrakis, A., and Noble, M.
(2004) Developments in the Ccp4Molecular-Graphics Project. Acta
Crystallogr. D60, 2288–2294.
1
11.
6. Ash, D. E., Emig, F. A., Chowdhury, S. A., Satoh, Y., and Schramm,
V. L. (1990) Mammalian and Avian Liver Phosphoenolpyruvate
Carboxykinase: Alternate Substrates and Inhibition by Analogs of
Oxaloacetate. J. Biol. Chem. 265, 7377–7384.
48. Cotelesage, J. J. H., Puttick, J., Goldie, H., Rajabi, B., Novakovski,
B., and Delbaere, L. T. J. (2007) How Does an Enzyme Recognize
2
CO ? Int. J. Biochem. Cell Biol. 39, 1204–1210.