NAD(P)H oxidase v from Lactobacillus plantarum (NoxV) displays enhanced operational stability even in absence of reducing agents

Article abstract of DOI:10.1016/j.molcatb.2011.04.013

Active pharmaceutical ingredients (APIs) such as l-sugars and keto acids are favorably accessed through selective oxidation of sugar alcohols and amino acids, respectively, catalyzed by NAD(P)-dependent dehydrogenases. Cofactor regeneration from NAD(P)H conveniently is achieved via water-forming NAD(P)H oxidases (nox2), which only need molecular oxygen as co-substrate. Turnover-dependent overoxidation of the conserved cysteine residue in the active site of water-forming NADH oxidases is the presumed cause of the limited nox2 stability. We present a novel NAD(P)H oxidase, NoxV from Lactobacillus plantarum, with specific activity of 167 U/mg and apparent kinetic constants at air saturation and 25 °C of k_cat,app = 212 s^-1 and K_M,app = 50.2 μM in the broad pH optimum from 5.5 to 8.0. The enzyme features a higher stability than other NAD(P)H oxidases against overoxidation, as is evidenced by a higher total turnover number, in the presence (168,000) and, most importantly, also in the absence (128,000) of exogenously added reducing agents. While the native enzyme shows exclusively activity on NADH, we engineered the substrate binding pocket to generate variants, G178K,R and L179K,R,H that accommodate and oxidize both NADH and NADPH as substrates.

Full text of DOI:10.1016/j.molcatb.2011.04.013

Journal of Molecular Catalysis B: Enzymatic 71 (2011) 159–165
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
NAD(P)H oxidase V from Lactobacillus plantarum (NoxV) displays enhanced
operational stability even in absence of reducing agents
Jonathan T. Parka, Jun-Ichiro Hiranoa,1, Vaijayanthi Thangavela,2
,
Bettina R. Riebelb, Andreas S. Bommariusa,c,∗
a School of Chemical & Biomolecular Engineering, Parker H. Petit Institute of Bioengineering and Biosciences, Georgia Institute of Technology, 315 Ferst Drive, Atlanta, GA 30332-0363,
USA
b Department of Pathology, Whitehead Building, Emory University, 615 Michael Drive, Atlanta, GA, 30322, USA
c School of Chemistry and Biochemistry, Georgia Institute of Technology, 901 Atlantic Drive, Atlanta, GA 30332-0400, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Active pharmaceutical ingredients (APIs) such as l-sugars and keto acids are favorably accessed through
selective oxidation of sugar alcohols and amino acids, respectively, catalyzed by NAD(P)-dependent dehy-
drogenases. Cofactor regeneration from NAD(P)H conveniently is achieved via water-forming NAD(P)H
oxidases (nox2), which only need molecular oxygen as co-substrate. Turnover-dependent overoxidation
of the conserved cysteine residue in the active site of water-forming NADH oxidases is the presumed
cause of the limited nox2 stability.
Received 4 January 2011
Received in revised form 9 April 2011
Accepted 11 April 2011
Available online 17 April 2011
Keywords:
We present a novel NAD(P)H oxidase, NoxV from Lactobacillus plantarum, with specific activ-
ity of 167 U/mg and apparent kinetic constants at air saturation and 25 ◦C of kcat,app = 212 s−1 and
KM,app = 50.2 ␮M in the broad pH optimum from 5.5 to 8.0. The enzyme features a higher stability than
other NAD(P)H oxidases against overoxidation, as is evidenced by a higher total turnover number, in the
presence (168,000) and, most importantly, also in the absence (128,000) of exogenously added reducing
agents. While the native enzyme shows exclusively activity on NADH, we engineered the substrate bind-
ing pocket to generate variants, G178K,R and L179K,R,H that accommodate and oxidize both NADH and
NADPH as substrates.
NADH oxidase
Cofactor
Cofactor regeneration
Cofactor specificity
Overoxidation
Total turnover number
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
number of l-nucleoside-based pharmaceutical compounds, such as
the hepatitis B drug Emtriva® and the human immunodeficiency
Enzymatic synthesis steps are used to produce active pharma-
ceutical ingredients (APIs) at higher yields, greater selectivity, and
greater productivity than is possible through isolation of natural
sources or chemically catalyzed routes. Interest in the nicotinamide
cofactor aided production of rare sugars, namely l-nucleosides,
for example l-ribose, l-mannose and l-gulose, has arisen for a
virus drug Clevudine®. A number of such rare sugar-based phar-
maceuticals are currently approved or in clinical trials [1]. Another
example of cofactor-assisted API synthesis is the production of
keto acids, which are used to treat mild chronic renal insufficiency
of hemodialysis patients and hyperphosphatemia [2,3]. Thus, ␣-
ketoglutarate can be synthesized enzymatically from mono sodium
l-glutamate (MSG) using l-glutamate dehydrogenase and a nicoti-
namide cofactor [4].
High costs of nicotinamide cofactors (NAD+: $30/g; NADP+:
$230/g from laboratory suppliers) rule out the option of adding
equimolar amounts for large-scale processes. A regeneration sys-
tem is mandatory for advantageous economics but also often
alleviates product inhibition, and acts as the driving force to
overcome thermodynamic equilibrium limitations [6]. Enzymatic,
chemical, electrochemical, photochemical, and biological methods
have been proposed for cofactor regeneration [5–7].
Among different systems that can be chosen for cofactor regen-
eration, NAD(P)H oxidases feature a number of benefits, as they
utilize only NAD(P)H and oxygen as co-substrates, both of which
are available intracellularly (Fig. 1), thus obviating exogenous addi-
Abbreviations: API, active pharmaceutical ingredient; BME, ␤-mercaptoethanol;
DTT, dithiothreitol; NoxV, NADH oxidase from L. plantarum V; SDS-PAGE, sodium
dodecyl sulfate polyacrylamide gel electrophoresis; TTN, total turnover number;
T5030, temperature exhibiting half of the original activity after 30 min; Enzymes,
E.C. 1.6.–.–NADH oxidase.
∗
Corresponding author at: School of Chemical and Biomolecular Engineering and
School of Chemistry and Biochemistry, Parker H. Petit Institute for Bioengineering
and Bioscience, Georgia Institute of Technology, 315 Ferst Drive, Atlanta, GA 30332-
0363, USA. Fax: +1 404 894 2291.
E-mail address: andreas.bommarius@chbe.gatech.edu (A.S. Bommarius).
Current address: Mitsubishi Chemical Corporation, 14-1 Shiba 4-chome,
1
Minato-ku, Tokyo, 108-0014, Japan.
2
Current address: Department of Chemistry, Graduate School of Science, Kyoto
University, Kitashirakawa-Oiwakecho, Sakyo-Ku, Kyoto 606-8502, Japan
1381-1177/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2011.04.013

160
J.T. Park et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 159–165
OH
OH
restriction sites used for restriction and ligation. The PCR
Mannitol dehydrogenase
product was gel purified with a Qiagen gel extraction kit and
cloned into a pET-28a vector (Novagen; Darmstadt, Germany).
The plasmid was then transformed into E. coli BL21(DE3)pLysS for
expression. Constructs were confirmed with sequencing through
Eurofins-mwg|operon Biosciences (Huntsville, AL).
HO
OH
OH
O
OH OH
OH OH
Ribitol
L-Ribose
NAD+
NADH
2.3. Site-directed mutagenesis
H2O
O2
Single and double mutations were performed on residues
G178 and L179 into K, R and K, R, H, respectively, through over-
lap and Quikchange® PCR protocol using the following forward
primers and their complementary reverse primers; G178ARR (5ꢀ-
GCAAGGTAAGGAAGTCACACTAATTGATARRTTACCACGGATTTTAA-
ATAAATACTTAGACAA-3ꢀ), L179CRY (5ꢀ-AGGTAAGGAAGTCAC-
ACTAATTGATGGTCRYCCACGGATTTTAAATAAATACTTAGACAAAG-
3ꢀ), L179K(5ꢀ-AGGTAAGGAAGTCACACTAATTGATGGTAAACCACGG-
ATTTTAAATAAATACTTAGACAAA-3ꢀ), G178ARR/L179CRY (5ꢀ-G-
CAAGGTAAGGAAGTCACACTAATTGATARRCRYCCACGGATTTTAAA-
TAAATACTTAGACAAAG-3ꢀ), G178ARR/L179K (5ꢀ-GCAAGGTAAGG-
AAGTCACACTAATTGATARRAAACCACGGATTTTAAATAAATACTTAG-
ACAAA-3ꢀ). Degenerate codons were used to generate multiple
mutants with single transformations. ARR codes for amino acids R
and K, and CRY codes H and R. The DNA from PCR was transformed
into E. coli XL1-blue. Colonies were picked from agar plates and sent
for sequencing to Eurofins-mwg|operon Biosciences (Huntsville,
AL).
nox2 NADH oxidase
Fig. 1. Schematic conversion of ribitol to l-ribose through mannitol-1-
dehydrogenase from Apium graveolens complemented with NADH cofactor
regeneration using nox2 NADH oxidase.
tion, and, in the case of nox2 NAD(P)H oxidases, produce only water
next to NAD(P)+ [8]. The reaction is carried out in a purely aqueous
phase (in the absence of harsh solvents and metal catalysts).
All previously discovered nox2 NAD(P)H oxidases utilize FAD as
a cofactor, and have a conserved cysteine residue that is catalyti-
cally active [9–12]. During catalysis, the thiol (–SH) or thiolate (–S−)
is oxidized to sulfenic acid (–SOH) and reduced back to the thiol/ate
as a part of the NAD(P)H reduction mechanism. During this redox
cycle the cysteine can also be overoxidized, producing a sulfinic
(–SO2H) and then a sulfonic acid (–SO3H), and thereby deactivat-
ing the enzyme [13]. Overoxidation of this cysteine residue can
be decelerated by employing exogenous reducing agents such as
dithiothreitol (DTT) or ␤-mercaptoethanol (BME) [4]. This overox-
idation of the catalytically active cysteine is the presumed cause of
turnover-limited operational stability of nox2 proteins [4,14]. Thus,
reducing agents have shown to extend the useful active enzyme
lifetime and its total turnover number (TTN), a measure of opera-
tional stability.
The NoxV gene codes for an annotated NADH oxidase (NoxV)
from Lactobacillus plantarum consisting of 1350 bp with a predicted
size of 49 kDa. This study investigates the enzymatic properties
of NoxV. A higher enzymatic productivity is achieved compared
to its analogs, as evidenced by NoxV’s higher TTN and increased
stability against over-oxidation. Furthermore, while the wildtype
enzyme solely accepts NADH as a substrate, we engineered the sub-
strate binding pocket so that the enzyme also accepts and oxidizes
NADPH.
2.4. Overexpression
Growth and overexpression was carried out in MagicMediaTM
E. coli Expression Medium (Invitrogen; Carlsbad, CA) with a final
concentration of 30 ␮g mL−1 each of kanamycin and chlorampheni-
col. A dual temperature protocol was used for growth, starting out
at 30 ◦C for 6 h, and then continued at room temperature (25 2 ◦C)
for an additional 22 h. Cultures were harvested by centrifugation in
a Beckman centrifuge at 4050 g for 20 min in 50 mL conical tubes.
The resulting cell pellet was either frozen, and stored at −80 ◦C or
purified directly as described below.
2.5. Purification
2. Experimental
Purification of NoxV was carried out at 4 ◦C or on ice to pre-
vent denaturation of the enzyme. Cell pellets were resuspended
in 15 mL of 10 mM Tris–Cl buffer, pH 7.5 with 5 mM DTT (Buffer
A) and sonicated at 14 W for 30 s nine times. Sonicated cells were
centrifuged at 18,500 g for 30 min. The clarified cell lysate was then
dialyzed against 250 mL of buffer A with 50% ammonium sulfate for
2 h. The dialysis membrane was transferred to fresh buffer and fur-
ther dialyzed for 2 h. The solution was then centrifuged at 18,500 g
for 30 min. The resulting supernatant was filtered through a 0.8 ␮m
and 0.2 ␮m microfiltration membranes in series. The filtrate was
loaded onto a HiPrep 16/10 butyl hydrophobic interaction column
on an ÄKTAexplorerTM. A gradient separation was performed start-
ing from buffer A with 30% ammonium sulfate to 15%. The fractions
with the highest activity were collected and dialyzed against buffer
A for 2 h. After exchanging to fresh buffer the sample was dialyzed
for an additional 2 h. The sample was loaded on a HiPrep 16/10
DEAE weak anionic exchange column on the ÄKTAexplorerTM. Sep-
aration was achieved with buffer A containing NaCl, a gradient of
150–250 mM. The resulting fractions were assayed and the ones
with highest activity were collected as pure protein. The purified
protein was either stored in 4 ◦C or in −20 ◦C with 25% glycerol.
2.1. Enzymes and other materials
The gene of NADH oxidase V (NoxV) from L. plantarum 10S was
obtained via gene probe from genomic DNA from the American
Type Culture Collection (Manassas, VA), ATCC 10012. NADH and
NADPH were obtained from Amresco (Solon, OH) and EMD chem-
icals (Gibbstown, NJ), respectively. MagicMediaTM Escherichia coli
Expression Medium was purchased from Invitrogen (Carlsbad, CA).
Molecular biology reagents such as DTT were obtained from JT
Baker (Phillipsburg, NJ), all other materials such as various salts
used for buffers, BME and NAD+ were obtained from Sigma–Aldrich
(St. Louis, MO). Oligonucleotides for cloning were purchased from
Eurofins-mwg|operon Biosciences (Huntsville, AL).
2.2. Cloning
The gene was cloned through PCR using the following primers;
forward (5ꢀ-TGCATGCATGCCATGGTTATGAAAGTTATTGTAATTGGT-
TGTACCCA-3ꢀ)and reverse(5ꢀ-CCGCCGCCGCCGCTCGAGTTATTCAG-
TGACAGCTTCGGCC-3ꢀ) with NcoI and XhoI, underlined, as

J.T. Park et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 159–165
161
detected with an extinction coefficient ε of 54,000 M−1 cm−1 via
fluorescence spectroscopy with maximal emission at 587 nm and
indicates H2O2 production with strict 1:1 stoichiometry. Various
amounts of substrates were reacted and assayed to detect the pres-
ence of H2O2. The reactions were carried out in the provided 50 mM
sodium phosphate buffer pH 7.4. Standards for the calibration curve
were prepared with the same reaction buffer. Steady-state emis-
sion and excitation spectra were recorded with a PTI fluorimeter
(Birmingham, NJ).
2.6. Enzyme assay and protein determination
Standard assays to detect enzymatic activity were performed
in 100 mM triethanolamine (TEA) buffer pH 7.5 with 5 mM DTT in
cuvettes with either 1 cm or 1 mm path length, depending on the
concentration of substrate. Excluding studies of enzymatic activ-
ity and stability temperature dependence, all activity assays were
performed with samples that were either pre-equilibrated to 25 ◦C,
using a thermomixer (Eppendorf; Hamburg, Germany) or made and
used directly at room temperature. Initial activity was measured
by following the absorbance change using a Beckman Coulter DU
800 UV/Vis spectrophotometer at 340 nm. Activity of the enzyme
was calculated using an extinction coefficient ε of NAD(P)H as
6.22 mM−1 cm−1 [15]. Unless otherwise noted, a substrate con-
centration of 0.2 mM NAD(P)H and enzyme concentration of 4 nM
was used for assays. One unit (U) of activity is defined as ␮mol of
NAD(P)+ produced per minute.
2.12. Total turnover number (TTN)
Standard kinetic assays at pH 7.5 and 25 ◦C in air-saturated solu-
tion were performed with 0.25, 0.5 and 1.0 nM of enzyme. 0.2 mM
NAD(P)H substrate concentration was used. Assays were carried
out for 2–3 h until there was no more enzymatic conversion of
the substrate. Calculations of TTN were performed by dividing the
change of NAD(P)H concentration until activity reached a standstill
by the (necessarily very low) concentration of enzyme subunits that
was used for each assay. As the stock solution of enzyme contained
DTT, the buffer was exchanged to 10 mM Tris–Cl pH 7.5.
The protein concentration was estimated by a Bradford assay
[16]. BSA was used as standards and the absorbance was measured
on a biophotometer (Eppendorf; Hamburg, Germany). SDS-PAGE
analysis was performed to confirm the purity.
2.7. pH activity
3. Results and discussion
Enzyme pH activity profiles were obtained at 25 ◦C using
100 mM buffers containing one of the following salts: sodium cit-
rate from pH 4.0 to pH 6.5; sodium phosphate from pH 6.0 to pH 8.0;
TEA from pH 7.0 to pH 8.0; Tris–Cl from pH 7.0 to pH 9.0; glycine
from pH 9.0 to pH 10.0.
3.1. Cloning, expression, and purification
NoxV was identified through a sequence blast search. It has
high similarity and identity levels with many other water-forming
NADH oxidases (Table 1). The NoxV gene was cloned from genomic
DNA of ATCC 10012 and inserted into a pET-28a vector using restric-
tion sites NcoI and XhoI. Sequencing results of the gene showed a
silent mutation where the nucleotide at position 45 changed from
C to T (accession number: Q88SH4).
NoxV was overexpressed in E. coli BL21(DE3)pLysS constitut-
ing 8% of cell protein. Purification of NoxV resulted in a yield of
14% of the total units and a 12.7-fold increase of specific activity
to 167.5 U mg−1 (Table 2). The pure protein was identified as a sin-
gle band at 49 kDa in sodium dodecyl sulfate polyacrylamide gel
electrophoresis (SDS-PAGE) analysis (lane 5, Fig. 2).
2.8. Temperature activity
The dependence of the enzymatic activity on temperature was
studied by preheating the buffer to different temperatures, and
then adding the enzyme. After 1 min of incubation, the standard
assay was carried out. A temperature range of 10–55 ◦C was chosen
for this study.
2.9. Temperature stability
Temperature stability was studied by incubating the enzyme at
various temperatures for 30 min. The enzyme solution was then
cooled down and assayed at 25 ◦C. This study covered a range of
15–55 ◦C.
3.2. Enzyme activity and stability
NAD(P)H oxidases are used exclusively in cofactor regeneration
because their byproducts, water and hydrogen peroxide, are not of
interest. As utility of NADH oxidases can only be derived through
coupling with other enzymes, their operating conditions must be
able to match those of the cofactor consuming enzyme. Conse-
quently, studies of reaction conditions, especially in terms of pH
value and temperature, are required to maximize production and
stability of both enzymes.
NoxV showed a rather broad pH activity range. Maximum activ-
ity was found at pH 7.5, independent of buffer type. The enzyme’s
optimal activity range was from pH 5.5 to 8.0, a common range for
NAD(P)H oxidases (Fig. 3). The upper limit is compatible with most
dehydrogenases, so appropriate coupling seems feasible.
The temperature activity profile showed that maximum instan-
taneous activity was found at 40 ◦C, and the activity quickly
declined to zero beyond that temperature (Fig. 4). Using an Arrhe-
nius model and data between 10 and 38 ◦C, the activation energy
Ea was calculated as 32.7 kJ mol−1, and the deactivation energy
(40–55 ◦C) Ed was calculated as −93.6 kJ mol−1. The temperature
that exhibited half of the original activity after 30 min, T5030, was
estimated to be 45 ◦C (Fig. 5). At 55 ◦C the enzyme was completely
inactive. The enthalpy of deactivation, ꢀHd, was calculated using
the Van’t Hoff equation to be 5.0 kJ mol−1 (Fig. 5).
2.10. Kinetic parameters
Depending on the enzyme, a substrate concentration range from
1.5 ␮M up to 984 ␮M was investigated to determine the kcat and
KM values for NAD(P)H. This was conducted at atmospheric con-
centrations of oxygen, 0.25 mM, present in the system. Doubly
concentrated solutions were prepared by adding each the substrate
andenzymeto100 mMTEA bufferpH7.5. Thereaction was initiated
by mixing the two solutions. The specific activity was measured,
and the kinetic parameters were calculated from that data.
Inhibition effects were measured by incubating the enzyme with
NAD+ for 30 min before the assay. A range of 0.2, 0.3, 0.4 and 0.6 mM
were chosen and investigated.
2.11. Amplex Red Assay (H2O2 presence)
An Amplex Red Hydrogen Peroxide/Peroxidase Assay Kit (Invit-
rogen; Carlsbad, CA) was used to determine the amount of
hydrogen peroxide (H2O2) produced during turnover. The pres-
ence of H2O2 was determined by incubating the standard assay
mixture with Amplex Red and peroxidase. Produced resorufin is

162
J.T. Park et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 159–165
Table 1
Sequence identity and similarity comparison of water forming NADH oxidases from different bacteria with respect to NoxV from L. plantarum (accession number: Q88SH4).
Bacteria
Identity (%)
Similarity (%)
Accession number
Reference
Lactobacillus sanfranciscensis
Lactococcus lactis
Lactobacillus brevis
Enterococcus faecalis
Streptococcus mutans
59.4
33.1
64.0
33.0
9.4
72.2
56.1
74.2
42.9
19.8
Q9F1X5
A2RIB7
Q8KRG4
P37061
Q54453
[18]
[9]
[12]
[27]
[28]
Table 2
Table of purification.
Volume (ml)
Units (U)
Protein (mg)
Sp. ac. (U mg−1
)
Purification fold
Yield (%)
Lysate
20
7
37.5
20
1843.5
930.0
379.3
261.9
139.3
88.5
4.7
13.24
10.51
80.72
167.5
1.0
0.8
6.1
100
50
21
AS50 dia.
Butyl dia.
DEAE
1.6
12.7
14
Lysate: clarified lysate; AS50 dia.: supernatant from centrifuged sample dialyzed against 50% ammonium sulfate; butyl dia.: fractions collected from HiPrep 16/10 butyl
column dialyzed against 10 mM Tris-Cl and 5 mM DTT; DEAE: fractions collected from HiPrep 16/10 DEAE column.
300
250
200
150
100
50
0
10
20
30
40
50
Temperature (°C)
Fig. 4. L. plantarum NoxV activity profile at various temperatures with 100 mM TEA
buffer pH 7.5.
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
Fig. 2. SDS-PAGE analysis of NoxV. Lane 1: 15 ␮g lysate, lane 2 and 3: 6 and 3 ␮g,
respectively, of dialyzed protein collected from HiPrep 16/10 butyl column, lane 4
and 5: 3 and 1 ␮g, respectively, of collected protein from HiPrep 16/10 DEAE column.
The desired protein band is approximately 49 kDa.
200
180
160
Citrate
140
Phosphate
TEA
120
100
80
60
40
20
0
15
25
35
45
55
Temperature (°C)
Tris-HCl
Glycine
30
Fig. 5. T50 plot. Checking stability of L. plantarum NoxV by incubating at different
temperatures for 30 min each.
3.3. Kinetic parameters
4.0
6.0
8.0
10.0
pH
Apparent kinetic data were obtained at constant oxygen con-
centration (air saturation) at 25 ◦C and pH 7.5 over a concentration
range of NADH from 5 to 200 ␮M. The data were fitted with four
different models (non-linear Michaelis–Menten, Lineweaver–Burk,
Eadie–Hofstee, and Hanes–Woolf), all of which were in good
agreement. Through non-linear fitting with a least squares approx-
Fig. 3. L. plantarum NoxV activity profile in different buffers (100 mM) at different
pH values at 25 ◦C.

J.T. Park et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 159–165
163
Table 3
Comparison of TTN with and without DTT, for NADH oxidases from L. plantarum, L. sanfranciscensis and L. lactis.
Organism
Enzyme
NADH
TTN
NADPH
TTN
TTN with DTT
TTN with DTT
NoxV WT
L179R
G178R/L179R
Nox2
128,000
181,000
123,000
5000
168,000
148,000
162,000
112,500
78,480
–
–
107,000
–
L. plantarum
100,000
–
L. sanfranciscensis [4]
L. lactis [9]
Nox2
38,740
For all the TTN measurements the relative standard deviation (%RSD) was less than 5%, except for the TTN of mutant G178R/L179R with NADH, which was 12%.
Table 4
Cell lysate activity of mutants.
Mutant
NADH activity (U/mg)
NADPH activity (U/mg)
Wild type
G178K
G178R
L179K
L179R
10.0
3.92
1.51
5.14
7.32
1.37
0.84
2.00
1.24
0.94
2.64
0.64
0.00
0.46
0.23
1.03
1.76
0.64
3.11
5.68
3.85
3.56
6.00
0.52
L179H
G178K/L179K
G178K/L179R
G178K/L179H
G178R/L179K
G178R/L179R
G178R/L179H
1.4
1.2
1
Fig. 6. Nonlinear fitting to a Michaelis–Menten kinetic to determine the kinetic
parameters kcat and KM at pH 7.5, 25 ◦C.
No enzyme
0.25 nM [E]
0.5 nM [E]
3.5
3.0
0.8
0.6
0.4
0.2
2.5
600 μM NAD+
2.0
400 μM NAD+
300 μM NAD+
200 μM NAD+
200 μM NAD+
0 μM NAD+
1.5
1.0
0.5
0.0
1 nM [E]
0
0
50
100
150
Time (min)
200
250
300
-50
50
150
Fig. 8. Total turnover number analysis with different amounts of enzyme. The reac-
tion was carried on until there was no consumption of substrate ([E] = 0.25, 0.5 and
1.0 nM, 25 ◦C, pH 7.4).
[NADH] (µM)
Fig. 7. Hanes–Woolf plot to identify NAD+ inhibition.
the analogs from Lactobacillus sanfranciscensis (0.2%) and Lactococ-
cus lactis (0.4–0.7%) [9,18], thus strongly suggesting that NoxV is a
water-forming oxidase.
imation, the apparent kcat and KM values of the wild type were
measured to be 211.6 s−1 and 50.2 ␮M, respectively, with an R2 of
0.998 (Fig. 6).
Possible product inhibition was investigated by incubating the
enzyme with different concentrations of NAD+. A non-competitive
inhibition pattern was observed where the vmax value decreased
with increasing inhibitor concentration, while KM,app was constant.
The product inhibition constant, KIP, was calculated to be 289 ␮M
for NAD+ (Fig. 7). The apparent inhibition ratio KM,app/KIP is cal-
culated to be 0.17. As this ratio is less than unity, it indicates that
complete conversion is possible despite the non-competitive prod-
uct inhibition [17].
3.5. Total turnover number (TTN)
Total turnover number (TTN) is a measure of catalyst produc-
tivity, defined as the total amount of product produced over the
lifetime of an enzyme [19]. The TTN, in presence of DTT, was found
to be approximately 168,000, at 25 ◦C and pH 7.5 (Fig. 8).
Three potential causes of TTN limitation, which are commonly
associated with enzyme deactivation, were investigated: (i) ther-
mal deactivation at 25 ◦C, (ii) deactivation caused by H2O2, and (iii)
non-competitive product (NAD+) inhibition in the system. Thermal
stability at 25 ◦C (i) was measured by incubating the enzyme for
extended periods of time. The enzyme was then assayed to deter-
mine the remaining activity. As there was no enzymatic activity
loss after 2 h (data not shown), which is longer than the reaction
completion timescale (Fig. 8), this study proved that the enzyme
3.4. Water/H2O2 formation
The Amplex Red assay was performed to determine whether
NoxV is a water- or hydrogen peroxide-producing enzyme. H2O2
was formed in 2.63% of the catalytic events, slightly higher than

164
J.T. Park et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 159–165
Table 5
Kinetic parameters of wild-type(WT) and mutant NAD(P)H oxidases. Data were fitted with least squares approximation to Michaelis–Menten kinetics with an R2 of 0.96 or
higher.
Enzyme
NADH
NADPH
kcat (s−1
)
KM (␮M)
kcat/KM (␮M−1 s−1
)
kcat (s−1
)
KM (␮M)
kcat/KM (␮M−1 s−1
)
WT
L179R
G178R/L179R
211.6
122.0
34.0
50.2
6.56
2.57
4.22
18.6
13.2
–
–
–
146.4
114.1
489.6
9.76
0.30
11.7
became catalytically inactive before it began to thermally deacti-
vate. The effect of H2O2 was studied (ii) at 25 ◦C and pH 7.5 within
a concentration range from 25 to 200 ␮M. Even at high concen-
trations of H2O2 the specific activity did not change over 1 h (data
not shown), thus demonstrating that the enzyme is indeed stable
against H2O2. The presence of NAD+ inhibition (iii) and its pattern
had been elucidated (see Section 2.10). However, as expected from
the apparent inhibition ratio of 0.17, even at high concentrations
of NAD+ there was a reasonable amount of residual activity (data
not shown), so it cannot be concluded that NAD+ had inhibited the
reaction completely. The cause of TTN-coupled deactivation is still
not fully elucidated but it is not related to temperature nor either
H2O2 or NAD+ concentration.
To investigate the effect of reducing agents on the TTN, mea-
surements were taken with and without reducing agents. The TTN
without reducing agents, with 5 mM DTT, and with 5 mM BME were
128,000, 168,000, and 107,000, respectively. The presence of DTT
had a positive effect but it was not as dramatic compared to the
NADH oxidase from L. sanfranciscensis, which has an increase of
over 20-fold [4]. However, addition of BME decreased the TTN by
∼15%. The reason for the different influence of the reducing agents
on the TTN is unknown, and will be assessed in future studies. One
intriguing but speculative thought focuses on the different stoi-
chiometry of the sulfenic acid reduction for DTT and BME: a single
molecule of DTT is capable of performing reduction, whereas two
molecules of BME are required. In contrast, the first molecule of
BME initiates thiol-disulfide interchange and occupies the water
channel, thus blocking access to the second BME molecule and
leaving the residue inactive.
NoxV has shown a higher TTN compared to previously stud-
ied NADH oxidases from L. sanfranciscensis and L. lactis, both with
and without exogenously added reducing agents (Table 3). Such a
higher TTN is an indication of improved intrinsic stability against
overoxidation at the catalytically active cysteine residue because
DTT, a known enhancer of TTN in the other NADH oxidases from L.
sanfranciscensis and L. lactis, has little effect on the TTN in L. plan-
tarum. As overoxidation of a catalytic cysteine is known as limiting
catalysis over time in enzymes such as d-amino acid oxidase from
Trigonopsis variabilis and xenobiotic reductase A (xen A) from Pseu-
domonas putida, this increased intrinsic stability is of great interest
[20,21]. Since there are known inhibition effects of DTT on certain
rare sugar-producing dehydrogenases, such as mannitol dehydro-
genase from Apium graveolens, it is crucial for a generally useful
NAD(P)H oxidase to be stable without DTT [22]. Also, improved
stability without reducing agents would be a significant advantage
in industries where the use of reducing agents is avoided.
the negative charge of the 2ꢀ-phosphate, but NoxV consists of only
small or hydrophobic residues in that area. Based on this knowl-
edge, these residues were targeted for mutation with basic residues
such as histidine, lysine and arginine [24,25,29].
The single mutations of residues G178 and L179 into K, R and K,
R, H, respectively, and double mutations 178/179 were performed
and investigated for substrate specificity. The mutation of G178H
was excludedbecausethen two histidines wouldbepositioned next
to each other, causing steric hindrance within the binding pocket.
The resulting mutants were expressed on small scale and assayed
at the cell lysate level (Table 4). All the mutants showed activity
with both NADH and NADPH as substrates. Among the single and
double variants, L179R shows the highest specific activity at 25 ◦C
and pH 7.5 with NADH (7.32 U/mg) and G178R/L179R with NADPH
(6.00 U/mg), compared to wild-type at 10.0 U/mg. We surmise that
introduction of an additional positive charge at G178R stabilizes
the positioning of L179R, decreasing the ability of free rotation of
L179R side-chain bonds through hydrogen bonding. Arginine may
be expected to provide the greatest NADPH activity because it has
the most positions for hydrogen bonding. The low activity of L179H
can be rationalized by the presence of an adjacent proline causing
steric hindrance.
3.7. Study of variants L179R and G178R/L179R
L179R and G178R/L179R were selected for further purification
and kinetic characterization (Table 5). L179R was not as active
as the wild-type (lower kcat,NADH). However, the KM,NADH value
was also much lower, improving the specificity kcat,app/KM,NADH
more than 4-fold. The mutant also showed NADPH activity but
with a very high KM,NADPH value. The double mutant G178R/L179R
shows a trend similar to L179R: decreased kcat,app,NADH and
KM,NADH values, resulting in an improved specificity kcat,app/KM,NADH
of 13.2 ␮M−1 s−1
. For NADPH activity kcat,app/KM,NADPH was
11.7 ␮M−1 s−1. Overall, both variants yielded improved specificities
kcat,app/KM for both NADH and NADPH. The TTN of the two variants
behaved similarly to the wildtype, in that the presence of reducing
agents did not affect the processing stability (TTN) to a great extent
(Table 3).
4. Conclusion
Starting from an annotated sequence, NADH oxidase V from L.
plantarum (ATCC 10012) was developed and demonstrated to be
a very active enzyme in air-saturated aqueous buffer at pH 7.5
and 25 ◦C (kcat,app = 212 s−1 and KM,app = 50.2 ␮M). The temperature
and pH optima, 45 ◦C and pH 5.5–8.0, respectively, overlap with
relevant dehydrogenases that might be coupled for cofactor regen-
eration with NoxV. With total turnover numbers (TTN) of 128,000
and 168,000, respectively, in the absence and presence of DTT, L.
plantarum NoxV demonstrated high processing stability regardless
of the presence of reducing agents, a first among NADH oxidases.
After inspection of the homology model of L. plantarum NoxV on
the structure of the L. sanfranciscensis analog, mutations in the sub-
strate binding pocket to basic amino acid residues to accommodate
the negative charge of the 2ꢀ-phosphate of NADPH were intro-
3.6. Mutation for NADPH activity
Wild type NoxV had activity exclusively towards NADH and
not towards NADPH. To introduce NADPH activity, substrate bind-
ing pocket mutations were carried out. Homology modeling of the
sequence of NoxV from L. plantarum onto the crystal structure of
NAD(P)H oxidase from L. sanfranciscensis [23] revealed electro-
static differences in the substrate binding pocket. Nox2 from L.
sanfranciscensis features histidine (His 179) able to accommodate

J.T. Park et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 159–165
165
duced to broaden the substrate specificity to NADPH. All single and
double variants G178K,R/L179K,R,H exhibited significant NADPH
and NADH activity. Novel and more stable cofactor regeneration
enzymes, in conjunction with novel methods of immobilization,
such as on functionalized nanotubes [26], stand to further broaden
the application of oxidative reactions with dehydrogenases/NADH
oxidases.
[11] C.M. Gibson, T.C. Mallett, A. Claiborne, M.G. Caparon, J. Bacteriol. 182 (2000)
448–455.
[12] W. Hummel, B. Riebel, Biotechnol. Lett. 25 (2003) 51–54.
[13] L.B. Poole, A. Claiborne, J. Biol. Chem. 264 (1989) 12330–12338.
[14] A. Claiborne, J.I. Yeh, T.C. Mallett, J. Luba, E.J. Crane, V. Charrier, D. Parsonage,
Biochemistry-Us 38 (1999) 15407–15416.
[15] R.M.C. Dawson, Data for Biochemical Research, 3rd ed., Clarendon Press, Oxford,
1986.
[16] M.M. Bradford, Anal. Biochem. 72 (1976) 248–254.
[17] L.G. Lee, G.M. Whitesides, J. Org. Chem. 51 (1986) 25–36.
[18] B.R. Riebel, P.R. Gibbs, W.B. Wellborn, A.S. Bommarius, Adv. Synth. Catal. 344
(2002) 1156–1168.
[19] A.S. Bommarius, B.R. Riebel, Biocatalysis, Wiley-VCH, Weinheim/Cambridge,
2004.
[20] A. Slavica, I. Dib, B. Nidetzky, Appl. Environ. Microb. 71 (2005) 8061–8068.
[21] Y. Yanto, H.H. Yu, M. Hall, A.S. Bommarius, Chem. Commun. 46 (2010)
8809–8811.
[22] J.M.H. Stoop, J.D. Williamson, M.A. Conkling, J.J. MacKay, D.M. Pharr, Plant Sci.
131 (1998) 43–51.
[23] G.T. Lountos, R.R. Jiang, W.B. Wellborn, T.L. Thaler, A.S. Bommarius, A.M. Orville,
Biochemistry-Us 45 (2006) 9648–9659.
References
[1] R.D. Woodyer, N.J. Wymer, F.M. Racine, S.N. Khan, B.C. Saha, Appl. Environ.
Microb. 74 (2008) 2967–2975.
[2] E. Riedel, M. Nundel, H. Hampl, Nephron 74 (1996) 261–265.
[3] V. Teplan, O. Schuck, M. Votruba, R. Poledne, L. Kazdova, J. Skibova, J. Maly,
Wien Klin Wochenschr 113 (2001) 661–669.
[4] P. Odman, W.B. Wellborn, A.S. Bommarius, Tetrahedron – Asymmetry 15 (2004)
2933–2937.
[5] M.D. Leonida, Curr. Med. Chem. 8 (2001) 345–369.
[6] H.K. Chenault, G.M. Whitesides, Appl. Biochem. Biotechnol. 14 (1987)
147–197.
[7] W.A. van der Donk, H.M. Zhao, Curr. Opin. Biotechnol. 14 (2003) 421–426.
[8] M. Higuchi, Y. Yamamoto, L.B. Poole, M. Shimada, Y. Sato, N. Takahashi, Y. Kamio,
J. Bacteriol. 181 (1999) 5940–5947.
[9] R.R. Jiang, B.R. Riebel, A.S. Bommarius, Adv. Synth. Catal. 347 (2005) 1139–1146.
[10] B. Geueke, B. Riebel, W. Hummel, Enzyme Microb. Technol. 32 (2003)
205–211.
[24] P. Bubner, M. Klimacek, B. Nidetzky, FEBS Lett. 582 (2008) 233–237.
[25] R. Woodyer, W.A. van der Donk, H.M. Zhao, Biochemistry-Us 42 (2003)
11604–11614.
[26] L.A. Wang, L. Wei, Y.A. Chen, R.R. Jiang, J. Biotechnol. 150 (2010) 57–63.
[27] R.P. Ross, A. Claiborne, J. Mol. Biol. 227 (1992) 658–671.
[28] J. Matsumoto, M. Higuchi, M. Shimada, Y. Yamamoto, Y. Kamio, Biosci. Biotech-
nol. Biochem. 60 (1996) 39–43.
[29] O. May, Presentation at Biocat 2010, Hamburg, Germany.