Kinetic and isotopic characterization of l -proline dehydrogenase from mycobacterium tuberculosis

Article abstract of DOI:10.1021/bi400338f

The monofunctional proline dehydrogenase (ProDH) from Mycobacterium tuberculosis performs the flavin-dependent oxidation of l-proline to Δ¹-pyrroline-5-carboxylate in the proline catabolic pathway. The ProDH gene, prub, was cloned into the pYUB1062 vector, and the C-terminal His-tagged 37 kDa protein was expressed and purified by nickel affinity chromatography. A steady-state kinetic analysis revealed a ping-pong mechanism with an overall k_cat of 33 ± 2 s^-1 and K _m values of 5.7 ± 0.8 mM and 3.4 ± 0.3 μM for l-proline and 2,6-dichlorophenolindophenol (DCPIP), respectively. The pH dependence of k_cat revealed that one enzyme group exhibiting a pK value of 6.8 must be deprotonated for optimal catalytic activity. Site-directed mutagenesis suggests that this group is Lys110. The primary kinetic isotope effects on V/K_Pro and V of 5.5 and 1.1, respectively, suggest that the transfer of hydride from l-proline to FAD is rate-limiting for the reductive half-reaction, but that FAD reoxidation is the rate-limiting step in the overall reaction. Solvent and multiple kinetic isotope effects suggest that l-proline oxidation occurs in a stepwise rather than concerted mechanism. Pre-steady-state kinetics reveal an overall k_red of 88.5 ± 0.7 s^-1, and this rate is subject to a primary kinetic isotope effect of 5.2. These data confirm that the overall reaction is limited by reduced flavin reoxidation in the second half-reaction.

Full text of DOI:10.1021/bi400338f

Article
pubs.acs.org/biochemistry
Kinetic and Isotopic Characterization of L‑Proline Dehydrogenase
from Mycobacterium tuberculosis
Hector Serrano and John S. Blanchard*
Department of Biochemistry, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, New York 10461, United States
ABSTRACT: The monofunctional proline dehydrogenase
(ProDH) from Mycobacterium tuberculosis performs the flavin-
dependent oxidation of L-proline to Δ1-pyrroline-5-carboxylate in
the proline catabolic pathway. The ProDH gene, prub, was cloned
into the pYUB1062 vector, and the C-terminal His-tagged 37
kDa protein was expressed and purified by nickel affinity
chromatography. A steady-state kinetic analysis revealed a ping-
pong mechanism with an overall kcat of 33 2 s−1 and Km values
of 5.7
0.8 mM and 3.4
0.3 μM for L-proline and 2,6-
dichlorophenolindophenol (DCPIP), respectively. The pH
dependence of kcat revealed that one enzyme group exhibiting a
pK value of 6.8 must be deprotonated for optimal catalytic
activity. Site-directed mutagenesis suggests that this group is Lys110. The primary kinetic isotope effects on V/KPro and V of 5.5
and 1.1, respectively, suggest that the transfer of hydride from L-proline to FAD is rate-limiting for the reductive half-reaction, but
that FAD reoxidation is the rate-limiting step in the overall reaction. Solvent and multiple kinetic isotope effects suggest that L-
proline oxidation occurs in a stepwise rather than concerted mechanism. Pre-steady-state kinetics reveal an overall kred of 88.5
0.7 s−1, and this rate is subject to a primary kinetic isotope effect of 5.2. These data confirm that the overall reaction is limited by
reduced flavin reoxidation in the second half-reaction.
Mycobacterium tuberculosis (Mtb), the causative agent of
tuberculosis (TB), is a global health concern.1 In 2011, there
were an estimated 8.7 million new cases of TB, with 1.4 million
deaths from TB. In spite of research into the eradication of TB,
the rise of drug-resistant strains of MTb is becoming alarmingly
common, and new drug targets need to be discovered.2
Recently, the oxidation of L-proline has come under scrutiny
for its possible protection of mycobacterial cells under hypoxic
conditions.3,4 Berney and Cook observed an increase in the
level of production of proline dehydrogenase (ProDH) at low
O2 concentrations and argued that the product of ProDH, Δ1-
pyrroline-5-carboxylate (P5C), reacts with methylglyoxal, a
toxic byproduct of several metabolic pathways, to form 2-acetyl-
1-pyrroline (2-AP).5−9
The oxidation of L-proline to L-glutamate follows a two-
enzyme pathway with Δ1-pyrroline-5-carboxylate as an
intermediate (Scheme 1). In most prokaryotes, the two
enzymes, proline dehydrogenase (ProDH) and Δ1-pyrroline-
5-carboxylate dehydrogenase (P5CDH), are a single fused
protein, named proline utilization A (PutA), while in
eukaryotes and some prokaryotes, they are separate enzymes.10
The L-proline dehydrogenase activity of fused multifunctional
enzymes and monofunctional enzymes have been studied.11,12
Crystal structures have also been determined for both forms of
the enzyme, suggesting two residues equivalent to K110 and
Y203 in the Mtb ProDH as potential active site bases.12,13 The
flavin-dependent ProDH activity of PutA from Escherichia coli
was recently characterized, in the steady state and pre-steady
state.11,14 It was found to follow a two-site ping-pong
mechanism utilizing various electron acceptors. Pre-steady-
state experiments allowed the kred and kox rates of PutA of 38
and 7.5 s−1, respectively, to be determined. No studies of M.
tuberculosis ProDH have been reported to date, and we present
here a kinetic and mechanistic analysis of the enzyme.
MATERIALS AND METHODS
■
Materials. Chemicals, biochemicals, buffers, and solvents
were purchased from Sigma-Aldrich Chemical Co. (St. Louis,
MO) or Fisher Scientific Inc. (Pittsburgh, PA), unless stated
otherwise. LB Broth, agar, Middlebrook 7H9, and BBL
Middlebrook ADC Enrichment were obtained from Becton,
Dickinson and Company (Franklin Lakes, NJ). Nucleic acid
primers were from Invitrogen. The pET-29a(+) vector was
from Novagen. The Mycobacterium smegmatis mc24517 cells
and pYUB1062 vector were a gift from W. R. Jacobs Jr. (Albert
Einstein College of Medicine). Hygromycin B and Ni-NTA
were purchased from Gold Biotechnology (St. Louis, MO).
Enzymes, reagents, and deoxynucleotide triphosphates
(dNTPs) used for the cloning procedures were purchased
from New England Biolabs (Ipswich, MA). Water (99.9%
deuterated) was from Cambridge Isotope Laboratories (And-
over, MA). [2,5,5-2H3]-L-Proline was from Medical Isotopes,
Inc. (Pelham, NH). Prepacked PD-10 Sephadex G-25 columns
Received: March 15, 2013
Revised: June 14, 2013
Published: June 19, 2013
© 2013 American Chemical Society
5009
dx.doi.org/10.1021/bi400338f | Biochemistry 2013, 52, 5009−5015

Biochemistry
Article
Scheme 1
were purchased from GE Healthcare Lifesciencs (Piscataway,
NJ).
internal PCR primers for K110A were oligonucleotides 5′-
GTGTCGCTCGCGCTGTCGGCGCTC-3′ and 5′-GAGCG-
CCGACAGCGCGAGCGACAC-3′, where the mutated co-
dons are bold. The internal PCR primers for Y203F were
oligonucleotides 5′-CAAGGGCGCCTTTGACGAACCCGC-
3′ and 5′-GCGGGTTCGTCAAAGGCGCCCTT-3′, where
the mutated codons are bold. Expression, purification, and
storage of the mutant forms of the enzyme followed the same
protocol that was used for the wild type.
Protein Concentration. The concentration of wild-type
(WT) ProDH and the two mutants was determined
spectrophotometrically using an ε451 of 11300 M−1 cm−1.
Measurement of Enzymatic Activity. Unless otherwise
stated, all steady-state kinetic assays were performed at 25 °C
and pH 7.1 in 20 mM Tris-HCl. ProDH activity was assayed by
measuring the reduction of the artificial electron acceptor 2,6-
dichlorophenolindophenol (DCPIP). The reaction mixture
consisted of 20 mM Tris-HCl, ProDH, DCPIP, and L-proline
in a total volume of 1 mL. The reaction was initiated by the
addition of L-proline. Reactions were monitored spectrophoto-
metrically until 10% conversion of substrate to product had
been achieved. Rates were calculated using the molar extinction
coefficient of DCPIP (ε600 = 18700 M−1 cm−1) and the total
enzyme concentration (Et).
Cloning, Overexpression, and Purification of ProDH.
The Mtb prub gene (Rv1188) was amplified via PCR from Mtb
H37Rv DNA using forward primer 5′-ATCCCGCTCATAT-
GGCCGGCTGGTTCGCGCAC-3′ and reverse primer 5′-
ATCCCGCTAAGCTTGCGCTCGGCGCACCCCCG-3′.
Amplification mixtures contained the appropriate synthetic
primers, deoxynucleotide triphosphates, genomic DNA, and
PCR reagents. The resulting PCR product and the pET29a(+)
vector were digested with NdeI and HindIII restriction
enzymes, purified, and ligated using T4 DNA ligase. An aliquot
of the ligation mixture was transformed into competent E. coli
DH5α cells. Transformants were selected at 37 °C on LB (10 g
of Bacto-tryptone, 5 g of yeast extract, and 10 g of NaCl) plates
with kanamycin (30 μg/mL) for selection. Plasmid DNA was
isolated from several randomly selected colonies and analyzed
by restriction analysis for the presence of the insert. The cloned
prub gene was sequenced to verify that no other mutations had
been introduced during the amplification of the gene. The
newly constructed expression vector was named pET29:prub.
Because of a lack of protein expression, the gene fragment
was subcloned into the mycobacterial pYUB1062 expression
vector using the same NdeI and HindIII restriction sites. After
the sequence had been verified, the recombinant plasmid,
pYUB1062:prodh, was transformed into M. smegmatis mc24517
cells. A single bacterial colony was selected and grown for 19 h
at 37 °C in medium containing 27 mL of Middlebrook 7H9
broth medium supplemented with 3 mL of BBL Middlebrook
ADC Enrichment, 0.05% Tween 80, 0.2% glycerol, kanamycin
(20 μg/mL), and hygromycin B (100 μg/mL). Cells were then
inoculated into 12 × 1 L of the medium described above and
grown to midexponential phase (OD600 = 0.5−0.6) at 37 °C
before being induced with acetamide (4 mL of a 0.5 g/mL
solution made in water) and allowed to grow for 18 h. Cells
were harvested by centrifugation, and the pellets were
resuspended in cold 50 mM Tris (pH 8.0) containing 300
mM NaCl, 10 mM imidazole, and 0.1% Triton X-100 (lysis
buffer), lysed using an EmulsiFlex-C3 homogenizer (Avestin,
Ottawa, ON), and centrifuged at 17000 rpm for 45 min to
remove cellular debris. Batch purification was performed with a
Ni-NTA agarose column. The column was washed with lysis
buffer, and the protein was eluted with 250 mM imidazole in
lysis buffer and stored at 4 °C.
Initial Velocity of WT, K110A, and Y203F ProDH. Initial
kinetic parameters were estimated by saturating with one
substrate and varying the concentration of the other substrate.
The resulting curve was fit to eq 1:
v = (VmaxA)/(Km + A)
(1)
where Vmax is the maximal velocity and A is the concentration
of the varied substrate. Initial velocity studies were conducted at
fixed, saturating concentrations of proline and varying
concentrations of DCPIP and vice versa. The resulting patterns
were fit globally to eq 2 for a ping-pong mechanism:
v = (VAB)/(KaB + KbA + AB)
(2)
where V is the maximal velocity, A and B are the substrate
concentrations, and Ka and Kb are the respective Michaelis
constants for each substrate.
pH Dependence of ProDH Activity. The pH depend-
encies of the kinetic parameters were determined in a mixed
buffer solution [MES, Bis-tris propane, and CHES (20 mM
each)] at the desired pH under saturating conditions of DCPIP
and varying concentrations of L-proline. The resulting kcat and
kcat/Kpro data were fit to eq 3:
K110A and Y203F ProDH Mutants. K110A and Y203F
ProDH were prepared using overlap extension PCR.15 The
forward primer (5′-ATCCCGCTCATATGGCCGGCTGGT-
TCGCGCAC-3′) is designated primer F1 and contains an NdeI
restriction site (underlined) followed by 18 bases correspond-
ing to the coding sequence of the prub gene. The reverse
primer (5′-ATCCCGCTAAGCTTGCGCTCGGCGCACCC-
CCG-3′) is designated primer R1 and contains a HindIII
restriction site (underlined) followed by 18 bases correspond-
ing to the complementary sequence of the prub gene. The
log kcat or log kcat/Kpro = log C/(1 + [H+]/K)
(3)
where C is the pH-independent plateau value, K is the observed
dissociation constant for the ionizing group, and [H+] is the
hydrogen ion concentration. This equation describes the pH
dependence at which the protonation of a single group
decreases the magnitude of the kinetic parameter.
5010
dx.doi.org/10.1021/bi400338f | Biochemistry 2013, 52, 5009−5015

Biochemistry
Article
Primary Kinetic Isotope Effects. Assays were performed
under saturating DCPIP conditions and varying concentrations
of either L-proline or [2,5,5-2H3]-L-proline in 20 mM Tris-HCl
(pH 8.0), where the enzymatic rate is independent of pH. The
results were globally fit to eq 4:
mutant forms of the enzyme yielded soluble protein of the
expected mass, ∼37 kDa.
Initial Velocity Studies. The parallel initial velocity pattern
obtained using L-proline and DCPIP (Figure 1), an artificial
v = (VA)/{K[1 + F(EV/K − 1)] + A(1 + FEv − 1)}
i
i
(4)
where V is the maximal velocity, A is the concentration of L-
proline or [2,5,5-2H3]-L-proline, Fi is the fraction of isotope (0
or 0.982), EV/K is the effect on kcat/Kpro, and EV is the effect on
kcat.
Solvent Kinetic Isotope Effects. Solvent kinetic isotope
effects were measured at saturating concentrations of DCPIP
and varying concentrations of L-proline and globally fit to eq 4,
where Fi = 0 for H2O and Fi = 0.92 for D2O. A viscosity control
of 9% glycerol did not show any effect on either V or V/Kpro. A
proton inventory was determined under saturating conditions
of both DCPIP and L-proline in 10% increments from 0 to 90%
D2O.
Stopped-Flow Measurements. All measurements were
conducted on a SX-20 stopped-flow spectrophotometer
operating in the absorbance mode at 451 nm (Applied
Photophysics). Assays contained 14 μM ProDH (after mixing)
and variable concentrations of L-proline or [2,5,5-2H3]-L-
proline, to ensure that binding is not contributing to the
observed rates. Measurements were performed in 50 mM Tris-
HCl (pH 8.4) containing 100 mM NaCl, 5 mM DTT, and 5%
glycerol made in water or D2O. The reduction of flavin by
substrate was monitored over time and fit to eq 5:
Figure 1. Bisubstrate kinetic analysis of ProDH. The initial velocity of
proline oxidation was measured in 50 mM Tris-HCl (pH 7.4) with
varying concentrations of L-proline (4−15 mM) and fixed concen-
trations of DCPIP: 1 ( ), 1.5 ( ), 2 ( ), 3.5 ( ), and 15 μM ( ).
Shown is the double-reciprocal plot of the initial velocity as a function
of DCPIP concentration. Global fitting of the data to eq 2 () gave a
kcat of 33 2 s−1, a KL‑pro of 5.7 0.8 mM, and a KDCPIP of 3.4 0.3
μM.
■
◆
●
▼
▲
electron acceptor, reveals that ProDH exhibits a ping-pong
kinetic mechanism, as observed for all other mono- and
bifunctional proline dehydrogenases11,12,14 (Figure 1). A kcat of
33 2 s−1 was determined. This value is similar to what was
found for E. coli ProDH measured with various other electron
acceptors.11 The Km values for L-proline and DCPIP were 5.7
0.8 mM and 3.4 0.3 μM, respectively (Table 1).
A = Af + A1e−k t + A2e−k t
1
2
(5)
where A is the absorbance at time t, Af is the final absorbance,
and A1 and A2 are the absorbance changes associated with each
16
of the first-order rate constants, k1 and k2, respectively
Table 1. Kinetic Parameters for ProDH and the Y203F
a
The kobs values obtained at various concentrations of
substrate were replotted as a function of enzyme concentration
and fit to eq 6:
Mutant
enzyme
kcat (s−1
)
KDCPIP (μM) Kpro (mM) kcat/Kpro (M−1 s−1
)
ProDH
Y203F
33.5 2.2
3.4 0.3
5.7 0.8
860 31
5877
117
kobs = (kHS)/(Kd + S)
b
(6)
100.6 2.1
ND
a
The steady-state kinetic parameters were measured in 20 mM Tris-
HCl buffer (pH 7.1) at 25 °C. Errors are standard deviations. Substrate
concentrations are described in Materials and Methods. Not
determined.
where kobs is the observed rate constant after the single-
exponential fitting at different concentrations of enzyme, kH is
the rate obtained with L-proline used as the substrate, S is the L-
proline concentration, and Kd is the dissociation constant for L-
proline. The isotope effect on the hydride transfer step or
proton extraction step (DkH) was obtained by dividing kH/kD,
using the values obtained after fitting with eq 6.
b
pH Dependence of the Kinetic Parameters. To probe
the potential involvement of general acid or general base
catalysis in the mechanism of ProDH, the pH dependencies of
kcat and kcat/Km were examined between pH 5.5 and 9.0, the
range in which the enzyme was found to be stable. The pH
profile was determined at a saturating concentration of DCPIP
and varying concentrations of L-proline. The pH dependencies
of kcat and kcat/Kpro are determined by a single ionizable group
exhibiting pKa values of 6.8 0.1 and 7.1 0.2, respectively
(Figure 2) that needs to be deprotonated for catalysis and
binding. The kcat/Kpro data at low pH values deviate somewhat
from the fit to a single ionizable group but clearly are not
consistent with two ionizable groups that must be deprotonated
(dashed line in Figure 2).
RESULTS AND DISCUSSION
■
Cloning, Expression, and Purification. The prub gene
was amplified from genomic DNA of M. tuberculosis H37Rv and
fused into the start codon of expression vector pET29a(+),
resulting in the pET29:prub construct. The construct was
sequenced to verify that no mutation was introduced by PCR.
The prub gene in pET29:prub is under the control of a T7
promoter, and expression of the enzyme was attempted in E.
coli T7 express, with no success. The gene was then subcloned
into the pYUB1062 vector, to generate pYUB1062:prodh, for
protein expression. The K110A and Y203F mutant genes were
amplified from the pYUB1062:prodh construct and sequenced
to confirm only the designated mutant was introduced by PCR.
Overexpression of the C-terminally His6-tagged WT and
The crystal structure of E. coli PutA in complex with the
competitive inhibitor, L-tetrahydro-2-furoic acid, provides clues
about which residue could function as an active site base,
5011
dx.doi.org/10.1021/bi400338f | Biochemistry 2013, 52, 5009−5015

Biochemistry
Article
primary kinetic isotope effects (KIEs) were measured at pH 8.1,
using L-proline and [2,5,5-2H3]-L-proline as substrates (Figure
4A). Normal KIEs were observed with a very large primary
effect on DV/Kpro of 5.6 0.1 and a small effect on DV of 1.1
0.2. The equations that describe the V and V/Kpro isotope
effects can be derived from the model shown in Scheme 2,
neglecting any contribution from the reverse commitment
factor. The commitment factor that determines the extent and
degree to which the intrinsic primary kinetic isotope effect is
observed for V/Kpro is k3/k2, that is, the ratio of the rate of the
chemical step defining the oxidation of L-proline to form Δ1-
pyrroline-5-carboxylate divided by the rate of dissociation of L-
proline from the binary Michaelis complex. The high value of
DV/Kpro argues that chemistry is much slower than dissociation,
and thus, L-proline is not a “sticky” substrate. The commitment
factor for V for a ping-pong reaction sequence is k3/k7, the ratio
of the two chemical steps that result in enzyme-bound flavin
reduction (k3) and enzyme-bound reduced flavin oxidation
(k7). The low value of DV, relative to either DV/Kpro or any
reasonable value for Dk3, requires that k7 be less than k3 and that
the oxidative half-reaction be the overall rate-limiting step of
the two chemical steps.
●
Figure 2. pH dependence of ProDH. kcat(app) ( ) and kcat/
K
▲
L‑pro(app) ( ), determined at saturating DCPIP and varying L-
proline concentrations. The solid lines are fits to eq 3, yielding pKa
values of 6.8 0.1 for kcat(app) and 7.1 0.2 for kcat/KL‑pro(app). The
dashed line in the kcat/KL‑pro−pH profile represents a fit for which two
groups must be deprotonated.
Lys329 or Tyr437.13 In the active site, Tyr437 is hydrogen-
bound to a water molecule, which is the closest neighbor to
what would be the amine group of proline. The ε-amino group
of Lys329 is 3.5 Å from L-tetrahydro-2-furoic acid, suggesting it
could also play a role in catalysis. A more recent structure of the
monofunctional ProDH from Deinococcus radiodurans similarly
shows the ε-amino group of Lys98 (equivalent to K110 in Mtb
ProDH) is 3.5 Å from the ring oxygen atom of the same
inhibitor.23 Sequence alignment of the bifunctional E. coli PutA
with the monofunctional ProDHs from D. radiodurans, Thermus
thermophilus, M. smegmatis, and M. tuberculosis shows both
these residues, Lys110 and Tyr203 (Mtb numbering), are
conserved in M. tuberculosis ProDH (Figure 3). We therefore
generated the K110A and Y203F mutant forms of the enzyme.
The K110A mutant has no measurable catalytic activity, while
the Y203F mutant exhibited an increase in the Kpro but no
change in kcat (Table 1). These data strongly argue that Y203
serves to bind and orient L-proline in the active site, but that the
active site base observed in the pH profile of kcat is K110.
Steady-State Kinetic Isotope Effects. To investigate the
rate-limiting nature of the catalytic oxidation of L-proline,
To investigate the involvement of proton transfer steps
occurring in the reactions, solvent KIEs were measured at pH
8.1, a region in which the kinetic parameters are not changing
as a result of the isotopic composition of the solvent. At
saturating concentrations of DCPIP and variable concen-
trations of L-proline, SKIEs for D OV of 1.3 0.1 and for D OV/
2
2
Kpro of 2.1
0.4 were determined (Figure 4B). A viscosity
control using 9% glycerol to mimic the relative viscosity of D2O
was conducted, and no effect of viscosity on the rate was
observed (data not shown). To determine the number of
protons involved in the solvent isotope-dependent step, a
proton inventory experiment was performed (Figure 4C).
D OV/H OV was determined in 10% increments between 0 and
∼100% D2O. The observed linear proton inventory is indicative
of a single solvent-derived proton involved in the step that is
solvent sensitive. Similar to the case of primary KIEs on V/Kpro,
only those steps involving proton transfer in the first half-
2
2
reaction influence the magnitude of D OV/Kpro. The oxidation
2
of L-proline requires both a deprotonation of the secondary
Figure 3. Sequence alignment of ProDHs from various organisms. Highlighted are the two residues believed to play a role as the general base.
5012
dx.doi.org/10.1021/bi400338f | Biochemistry 2013, 52, 5009−5015

Biochemistry
Article
●
Figure 4. Kinetic isotope effects. (A) Primary kinetic isotope effects at saturating DCPIP (80 μM) and varying (2−25 mM) L-proline ( ) or
2
D
[2,5,5- H3]-L-proline ( ) concentrations at pH 8.05. Data were fit to eq 4, which gave a V of 1.1 0.2 and a DV/KL‑pro of 5.6 0.1. (B) Solvent
▲
●
▲
kinetic isotope effects at saturating DCPIP (80 μM) and varying (2−25 mM) L-proline concentrations in H2O ( ) or D2O ( ) at pH 8.05. Data
were fit to eq 4, which gave a V of 1.3 0.1 and a V/KL‑pro of 2.1 0.4. (C) Proton inventory at saturating DCPIP (80 μM) and saturating L-
D
D
proline (20 mM) concentrations at pH 8.05 in 10% D2O increments from 0 to 100%. (D) Primary kinetic isotope effects at saturating DCPIP (80
2
●
▲
μM) and varying (2−25 mM) L-proline ( ) or (5−25 mM) [2,5,5- H3]-L-proline ( ) concentrations performed in D2O at pH 8.05. Data were fit to
eq 4, which gave a DV of 1.5 0.1 and a DV/KL‑pro of 3.4 0.4.
proceeds via an initial deprotonation of the protonated ring
nitrogen of L-proline (pK = 10.6) by K110, followed by the
transfer of a hydride from the C5 position of L-proline to FAD,
resulting in the formation of Δ1-pyrroline-5-carboxylate.17,18
Pre-Steady-State Kinetic Isotope Effects. To confirm
our interpretations of the steady-state isotope effects, we
measured the rate of the individual first half-reaction, the
reduction of flavin, using single-turnover stopped-flow
techniques. Using L-proline, a kred rate of 88.5 0.7 s−1 was
measured with an apparent Kd of 33.5 0.5 mM. This kred is
∼3-fold faster than that of the ProDH from E. coli.14 The kred
was then measured using [2,5,5-2H3]-L-proline as the substrate,
allowing for the calculation of the primary kinetic isotope effect
on Dkred of 5.2 (Figure 5A). These data confirm that L-proline is
not sticky and that the forward commitment factor is near zero.
It also confirms our interpretation, based on the much smaller
magnitude of the determined steady-state values of DV
Scheme 2
amine nitrogen of L-proline (by K110) and the transfer of a
hydride ion from C5 to N5 of FAD. The deprotonation step
will be sensitive to solvent isotopic composition, and the
magnitude of D OV/Kpro is relatively small (2.0), suggesting that
2
D
it is not rate-limiting, as least compared to the hydride transfer
compared to V/Kpro, that oxidation of the reduced flavin by
step reported by the primary KIE (DV/Kpro = 5.5). The
DCPIP is slower than flavin reduction by L-proline. For a ping-
pong mechanism, the reciprocal of the kcat is equal to the sum
of the reciprocal rates of the two half-reactions (1/kcat = 1/kred
+ 1/kox). Using a kcat value of 33 s−1 and a kred value of 88 s−1,
kox can be calculated as 52 s−1. Finally, kred was then measured
magnitude of D OV suggests that cleavage of the N5−D flavin
2
bond of reduced FAD is similarly not rate-limiting in the
oxidative half-reaction.
To probe the stepwise versus concerted nature of L-proline
oxidation, the primary kinetic isotope effect on L-proline and
[2,5,5-2H3]-L-proline oxidation was determined in D2O. The
in H2O and D2O, yielding a D Okred of 1.1, again confirming the
2
steady-state data (Figure 5B).
D
D
measured value of VD O was 1.5 0.1, while V/Kpro‑D O was
Chemical Mechanism. Various mechanisms have been
proposed for the oxidation of amines by flavoproteins.19,20
However, the mechanism of L-proline oxidation by ProDH is
different in two significant ways: oxidation occurs at a
2
2
3.4 0.4 (Figure 4D). The decrease in the primary V/K KIE
from 5.5 in H2O to 3.4 in D2O argues that the proton and
hydride transfer steps are uncoupled and that the reaction
5013
dx.doi.org/10.1021/bi400338f | Biochemistry 2013, 52, 5009−5015

Biochemistry
Article
Scheme 3
Figure 5. Stopped-flow kinetic isotope effects. (A) Primary KIEs. Each
point represents the average of three traces collected with 14 μM
2
●
ProDH and various concentrations of L-proline ( ) or [2,5,5- H3]-L-
▲
proline ( ) (5−60 mM). (B) Solvent KIEs. Each point represents the
average of three traces collected with 14 μM ProDH and various
concentrations of L-proline (5−60 mM) in H2O ( ) or D2O ( ).
●
▲
secondary amine, and oxidation involves the N1−C5 bond
rather than the typical N1−Cα bond. For years, the mechanism
of D-amino acid oxidase (DAAO) was thought to involve base
abstraction of the α-hydrogen, to form a carbanion, with
subsequent formation of the N1Cα bond after the transfer of
the hydride to FAD21 (Scheme 3A, shown for alanine). More
recent work on various amino acid oxidases has led to an
alternative mechanism in which the amino hydrogen is
removed first, followed by either a concerted hydride transfer
(Scheme 3B) or a stepwise transfer17,18,22 (Scheme 3C). On
the basis of the results reported here, most importantly the
multiple kinetic isotope effects, we propose a mechanism for
the reductive half-reaction (Scheme 3C). After L-proline enters
the active site, it interacts with Y203, positioning it to allow for
interaction of the amino ring nitrogen and the uncharged K110.
Deprotonation of the ring nitrogen by K100 gives the neutral
amino acid, which then collapses with donation of a hydride ion
from C5 to the FAD. The proton transfer step is less
energetically difficult than the hydride transfer step, as
supported by our observation of a modest SKIE but a very
large primary KIE on V/Kpro. It is this later step that is rate-
limiting in the reductive half-reaction, but it is reduced flavin
reoxidation that is rate-limiting for the overall reaction. This
same stepwise chemical mechanism has recently been proposed
for the D-arginine dehydrogenase reaction using similar isotopic
approaches.24
AUTHOR INFORMATION
■
Corresponding Author
*Department of Biochemistry, Albert Einstein College of
Medicine, 1300 Morris Park Ave., Bronx, NY 10461. E-mail:
john.blanchard@einstein.yu.edu. Phone: (718) 430-3096. Fax:
(718) 430-8565.
Funding
This work was supported by National Institutes of Health
Grants AI33696 and AI60899 to J.S.B. and support for H.S.
from Geographic Medicine and Emerging Infections Training
Grant NIH-NIAID T32 AI046985 awarded to the Albert
Einstein College of Medicine.
Notes
The authors declare no competing financial interest.
ABBREVIATIONS
■
DTT, dithiothreitol; LB, Luria broth; MES, 2-(N-morpholino)-
ethanesulfonic acid; Ni-NTA, nickel nitriloacetic acid; PCR,
polymerase chain reaction; SDS−PAGE, sodium dodecyl
sulfate−polyacrylamide gel electrophoresis; Tris, tris-
(hydroxymethyl)aminomethane; WT, wild type; FAD, flavin
adenine dinucleotide; Bis-tris propane, 1,3-bis[tris-
(hydroxymethyl)methylamino]propane; CHES, N-cyclohexyl-
2-aminoethanesulfonic acid.
5014
dx.doi.org/10.1021/bi400338f | Biochemistry 2013, 52, 5009−5015

Biochemistry
REFERENCES
Article
proline dehydrogenase provide insight into substrate recognition and
conformational changes associated with flavin reduction and product
release. Biochemistry 51, 10099−10108.
(24) Yuan, H., Xin, Y., Hamelberg, D., and Gadda, G. (2011) Insights
into the mechanism of amine oxidation by D-arginine dehydrogenase
through pH and kinetic isotope effects. J. Am. Chem. Soc. 133, 18957−
18965.
■
(1) Global tuberculosis report 2012 (2012) World Health
Organization, Geneva (http://www.who.int/tb/publications/global_
report/en/).
(2) The Stop TB Strategy (2006) World Health Organization,
Geneva (http://www.who.int/tb/strategy/en/).
(3) Berney, M., and Cook, G. M. (2010) Unique flexibility in energy
metabolism allows mycobacteria to combat starvation and hypoxia.
PLoS One 5, e8614.
(4) Berney, M., Weimar, M. R., Heikal, A., and Cook, G. M. (2012)
Regulation of proline metabolism in mycobacteria and its role in
carbon metabolism under hypoxia. Mol. Microbiol. 84, 664−681.
(5) Casazza, J. P., Felver, M. E., and Veech, R. L. (1984) The
metabolism of acetone in rat. J. Biol. Chem. 259, 231−236.
(6) Dhar, A., Desai, K., Kazachmov, M., Yu, P., and Wu, L. (2008)
Methylglyoxal production in vascular smooth muscle cells from
different metabolic precursors. Metabolism 57, 1211−1220.
(7) Huang, T. C., Huang, Y. W., Hung, H. J., Ho, C. T., and Wu, M.
L. (2007) Δ1-Pyrroline-5-carboxylic acid formed by proline
dehydrogenase from the Bacillus subtilis ssp. natto expressed in
Escherichia coli as a precursor for 2-acetyl-1-pyrroline. J. Agric. Food
Chem. 55, 5097−5102.
(8) Nemet, I., Turk, Z., Duvnjak, L., Car, N., and Varga-
Defterdarovic, L. (2005) Humoral methylglyoxal level reflects
glycemic fluctuation. Clin. Biochem. 38, 379−383.
(9) Weber, J., Kayser, A., and Rinas, U. (2005) Metabolic flux analysis
of Escherichia coli in glucose-limited continuous culture. II. Dynamic
response to famine and feast, activation of the methylglyoxal pathway
and oscillatory behaviour. Microbiology 151, 707−716.
(10) Tanner, J. J. (2008) Structural biology of proline catabolism.
Amino Acids 35, 719−730.
(11) Moxley, M. A., Tanner, J. J., and Becker, D. F. (2011) Steady-
state kinetic mechanism of the proline:ubiquinone oxidoreductase
activity of proline utilization A (PutA) from Escherichia coli. Arch.
Biochem. Biophys. 516, 113−120.
(12) White, T. A., Krishnan, N., Becker, D. F., and Tanner, J. J.
(2007) Structure and kinetics of monofunctional proline dehydrogen-
ase from Thermus thermophilus. J. Biol. Chem. 282, 14316−14327.
(13) Zhang, M., White, T. A., Schuermann, J. P., Baban, B. A., Becker,
D. F., and Tanner, J. J. (2004) Structures of the Escherichia coli PutA
proline dehydrogenase domain in complex with competitive inhibitors.
Biochemistry 43, 12539−12548.
(14) Moxley, M. A., and Becker, D. F. (2012) Rapid reaction kinetics
of proline dehydrogenase in the multifunctional proline utilization A
protein. Biochemistry 51, 511−520.
(15) Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K., and Pease,
L. R. (1989) Site-directed mutagenesis by overlap extension using the
polymerase chain reaction. Gene 77, 51−59.
(16) Czekster, C. M., Vandemeulebroucke, A., and Blanchard, J. S.
(2011) Kinetic and Chemical Mechanism of the Dihydrofolate
Reductase from Mycobacterium tuberculosis. Biochemistry 50, 367−375.
(17) Gaweska, H. M., Roberts, K. M., and Fitzpatrick, P. F. (2012)
Isotope effects suggest a stepwise mechanism for berberine bridge
enzyme. Biochemistry 51, 7342−7347.
(18) Fan, F., and Gadda, G. (2005) On the catalytic mechanism of
choline oxidase. J. Am. Chem. Soc. 127, 2067−2074.
(19) Fitzpatrick, P. F. (2007) Insights into the mechanisms of
flavoprotein oxidases from kinetic isotope effects. J. Labelled Compd.
Radiopharm. 50, 1016−1025.
(20) Fitzpatrick, P. F. (2010) Oxidation of amines by flavoproteins.
Arch. Biochem. Biophys. 493, 13−25.
(21) Ghisla, S., and Massey, V. (1989) Mechanisms of flavoprotein-
catalyzed reactions. Eur. J. Biochem. 181, 1−17.
(22) Sobrado, P., and Fitzpatrick, P. F. (2003) Solvent and primary
deuterium isotope effects show that lactate CH and OH bond
cleavages are concerted in Y254F flavocytochrome b2, consistent with
a hydride transfer mechanism. Biochemistry 42, 15208−15214.
(23) Luo, M., Arentson, B. W., Srivastava, D., Becker, D. F., and
Tanner, J. J. (2012) Crystal structures and kinetics of monofunctional
5015
dx.doi.org/10.1021/bi400338f | Biochemistry 2013, 52, 5009−5015