Biochimica et Biophysica Acta 1857 (2016) 1506–1513

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbabio

Direct electrochemistry of nitrate reductase from the fungus

Neurospora crassa
Palraj Kalimuthu a, Phillip Ringel b, Tobias Kruse b, Paul V. Bernhardt a,⁎
a
School of Chemistry and Molecular Biosciences, University of Queensland, Brisbane 4072, Australia
b
Department of Plant Biology, Braunschweig University of Technology, 38106 Braunschweig, Germany

a r t i c l e i n f o a b s t r a c t

Article history: We report the first direct (unmediated) catalytic electrochemistry of a eukaryotic nitrate reductase (NR). NR
Received 14 February 2016 from the filamentous fungus Neurospora crassa, is a member of the mononuclear molybdenum enzyme family
Accepted 1 April 2016 and contains a Mo, heme and FAD cofactor which are involved in electron transfer from NAD(P)H to the (Mo)
Available online 7 April 2016
active site where reduction of nitrate to nitrite takes place. NR was adsorbed on an edge plane pyrolytic graphite
(EPG) working electrode. Non-turnover redox responses were observed in the absence of nitrate from holo NR
Keywords:
Molybdenum
and three variants lacking the FAD, heme or Mo cofactor. The FAD response is due to dissociated cofactor in all
Enzyme cases. In the presence of nitrate, NR shows a pronounced cathodic catalytic wave with an apparent Michaelis con-
Voltammetry stant (KM) of 39 μM (pH 7). The catalytic cathodic current increases with temperature from 5 to 35 °C and an ac-
Nitrate reductase tivation enthalpy of 26 kJ mol−1 was determined. In spite of dissociation of the FAD cofactor, catalytically activity
is maintained.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction Our focus has been on the mononuclear molybdenum oxidoreduc-

The immobilization of proteins on electrode surfaces is an
established methodology for achieving direct heterogeneous elec-
tron transfer of redox proteins and larger molecular weight oxidore-
ductase enzymes [1–5]. When successful, it provides a foundation for
the fabrication of new kinds of electrochemical biosensors, bio-fuel cells
and biomedical devices without use of any artificial electron transfer
mediators [6,7]. The direct electrochemistry of small redox proteins
(MW b 20 kDa) such as cytochrome c, hemoglobin and myoglobin has
a long history and is now quite well understood [8–11]. However, the
successful direct electrochemistry of larger oxidoreductase enzymes is
much less common. Unless there is a specific part of the enzyme's sol-
vent exposed surface that has an affinity for the electrode surface, the
orientation of the protein will be unknown and most likely random.
Most of these random orientations will be inappropriate for interfacial
electron transfer due to the sheer size of the enzyme and the remote-
ness of the redox active cofactor(s) from the electrode surface. Another
issue is denaturation due to adsorption on the electrode surface
[12–14]. These problems can be avoided by modification of the elec-
trode surface with (redox inert) promoters including self-assembled
monolayers [15,16], nanoparticles [17], surfactants [18], carbon nano-
tubes [19,20] and polymers [21] which favor the orientation of enzyme
on the electrode surface for electron transfer.
⁎ Corresponding author.
E-mail address: p.bernhardt@uq.edu.au (P.V. Bernhardt).
tases, which are usually classified into three sub-families on the basis
of the first coordination sphere of their Mo-containing active site (Fig.
1) [22]. However this primary classification belies the diversity of struc-
ture within each family including the presence of other redox active co-
factors that relay electrons to or from the Mo active site during catalysis.
NR from the fungus Neurospora crassa is a complex homodimeric en-
zyme (MW ~200 kDa) that catalyzes the initial step in inorganic nitro-
gen assimilation [23]. Each NR monomer bears three redox cofactors;
the Mo cofactor (Moco), a heme and FAD (Scheme 1) [24–26]. The N
terminal domain includes the Mo cofactor active site while the C termi-
nal domain contains the FAD and NAD(P)H binding sites. The heme do-
main of NR has homology with cytochrome b5 and the FAD domain is
similar to cytochrome b5 reductase [23,25,27,28].
Based on the crystal structure of NR from the yeast Pichia angusta and
biochemical studies, the catalytic mechanism for eukaryotic nitrate re-
duction is as illustrated in Scheme 2 [27]. NADPH provides the FAD cofac-
tor with two electrons that are subsequently transferred (one at a time)
via the heme to the Mo active site. Nitrate binds to the active MoIV form
of NR and O-atom transfer to Mo is coupled with oxidation of MoIV to
MoVI releasing nitrite as the product. After completion of the reductive
half-reaction, MoVI is returned to its active MoIV form by the reduced
FAD and heme cofactors.
Very recently, we reported the mediated catalytic voltammetry of a
truncated (FAD-free) form of NR from the plant Arabidopsis thaliana [29,
30]. We were not able to achieve direct catalytic voltammetry of this en-
zyme but instead used artificial electron transfer mediators (viologens

http://dx.doi.org/10.1016/j.bbabio.2016.04.001
0005-2728/© 2016 Elsevier B.V. All rights reserved.
 P. Kalimuthu et al. / Biochimica et Biophysica Acta 1857 (2016) 1506–1513
Fig. 1. The three mononuclear Mo enzyme families.
and anthraquinol) to accomplish electro-catalysis. To date no direct
electrochemistry of any eukaryotic NR has been reported.
There are several known Mo-containing nitrate reductases but most
electrochemical studies have focused on bacterial NR enzymes, which
possess an entirely different active site and belong to the DMSO reduc-
tase family (Fig. 1). The prokaryotic NR enzymes investigated electro-
chemically specifically are those from Paracoccus pantotrophus [31],
Escherichia coli [32] (NarGH) and Rhodobacter sphaeroides (NapAB)
[33,34]. In contrast, the eukaryotic NR enzymes found in plants, fungi
and algae, belong to the sulfite oxidase family (Fig. 1) and share a high
degree of sequence homology. The enzymes from higher plants and
algae use NADH as their electron donor while those from fungi use
NADPH as the electron donor for nitrate reduction [24].
Herein we demonstrate direct catalytic electrochemistry of recombi-
nant NR from the filamentous fungus N. crassa [23]. The holo NR
employed in this study lacks the first 112 amino acids from its N-
terminus (K113-holo NR, Scheme 3A) and this deletion enhances
protein yield and purity without loss of functionality relative to full-
length holo NR [23]. Furthermore this recombinant system is well un-
derstood and a number of key variants may be expressed which lack
each of the three cofactors (Mo, heme and FAD). The H654A/H677A
double mutation (heme-free NR, Scheme 3B) lacks the heme binding
histidines and so the heme cofactor is absent altogether from this vari-
ant. Similarly the R778E mutant (FAD-free NR, Scheme 3C) does not
bind FAD due to the loss of this key arginine residue. The so-called cyto-
chrome c reducing fragment only contains amino acids 618–984 and
lacks the Mo cofactor (Mo-free NR, Scheme 3D). This fragment has
been shown to catalytically reduce cytochrome c (a non-physiological
function in this case) using NADPH as the reductant [35] but in this trun-
cated form it has no capacity to reduce nitrate [23].
1.1. Mo-free NR
The four proteins shown in Scheme 3 provide a unique opportunity
to study the electro-catalytic properties of these variants and to gauge
the relative importance of each redox active cofactor. As will be
shown it has also enabled a systematic study of the redox cofactors
under non-turnover conditions.
2. Materials and methods
2.1. Cloning of N. crassa K113 nitrate reductase and variants
For creation of N terminal truncated nitrate reductase variant
K113, the first 336 codons were deleted. PCR-based amplification of
Scheme 1. Domain structure of Neurospora crassa nitrate reductase. Nitrate reductase variant K113 lacks the first N-terminal 112 amino acids. Molybdenum cofactor (Moco) and
dimerization domain share a common sequence stretch comprising 11 amino acids [23].
1507
NR variant K113 coding sequence was carried out using Phusion
High-Fidelity polymerase (NEB) with primers (5′-ATACACGTGAAA
CCAGCCTACCCCCTCCC-3′) and (5′-ATTACTAGTTCAAAAAACTAATAC
ATCCTCATCCTTCC-3). As a template the readily available full length
N. crassa nitrate reductase coding sequence [23] was used. As tem-
plates for PCR based cloning of NR variants K113 H654A/H677A
(hereafter referred to as heme-free NR) and K113 R778E (FAD-free
NR) we used the respective coding sequences described earlier
[23]. The CloneJET™ PCR Cloning Kit (Thermo Scientific) has been
used for subcloning according to the manufacturer's instructions.
Upon sequence confirmation, SpeI and PmlI based subcloning into
the expression vector were performed. As expression vector we
chose a pQE-80L (Qiagen GmbH) based vector, allowing the N-
terminal fusion of a 6 × His-tag and the C-terminal fusion of a Twin-
Strep-tag® [23,36] to NR K113 variants.
2.2. Enzymes and materials
K113 nitrate reductase variants were expressed and purified from
E. coli strains TP1000 [37] and BL21, respectively. Heterologous gene
expression and affinity purification were performed as described
earlier [23]. SDS-PAGE analyses of all proteins studied herein are
shown in the Supporting Information. Sodium nitrate, sodium sul-
fite, polyethyleneimine and polymyxin B were purchased from Al-
drich and were used as received. All other reagents used were of
analytical grade purity and used without any further purification.
All solutions were prepared in purified water (Millipore, resistivity
18.2 MΩ·cm). Phosphate buffer was prepared using equal amounts
of Na 2HPO4 /NaH2 PO 4 to give a total phosphate concentration of
100 mM. The mixture of buffers (20 mM citric acid, 20 mM MES,
20 mM Bis-Tris, 20 mM Tris and 20 mM CHES) was used for pH de-
pendent experiments in the range 4.5 b pH b 10 and the desired pH
was obtained by addition of dilute acetic acid or NaOH.
2.3. Electrochemical measurements and electrode cleaning
Cyclic voltammetry (CV) was carried out with a BAS 100B/W elec-
trochemical workstation using a three-electrode system consisting of
edge-plane pyrolytic graphite (EPG) working electrode, a platinum
wire counter electrode, and a Ag/AgCl reference electrode (+196 mV
vs the normal hydrogen electrode, NHE). All potentials are cited versus
NHE. Unless otherwise stated, electrochemical solutions were purged
with argon for at least 30 min. Prior to the series of experiments and
all experiments were performed under a blanket of argon gas. Volt-
ammetry carried out at different temperatures ranging from 5 to 35 °C
1508 P. Kalimuthu et al. / Biochimica et Biophysica Acta 1857 (2016) 1506–1513

Scheme 2. Electron flow and atom transfer within holo NR. The intermediate MoV form is not shown for clarity.

was achieved using a Huber, Ministat-125 temperature controlled 2.4. Enzyme electrode preparation
water bath connected to a water jacketed electrochemical cell.
The variation of the observed limiting catalytic current (ilim) as a The EPG working electrode (surface area ̴ 0.1 cm2) was prepared by
function of nitrate concentration followed Michaelis–Menten kinetics cleaving several layers from the face of the electrode with a microtome
and the data were fit to Eq. (1) and then cleaning by sonication in Milli-Q water. No abrasives were
used. The cleaned electrode was dried in a nitrogen atmosphere. A
  3 μL droplet of NR solution (141 μM in Tris–HCl buffer pH 8) was
i max NO− added to the conducting surface of the inverted EPG electrode surface
i lim ¼  3−  ð1Þ
K M þ NO3 along with a mixture of 1.5 μL polyethyleneimine (PEI, 5%) and 1.5 μL
polymyxin B (PM, 5%). This mixture was allowed to dry to a thin film
over ca. 1 h at 4 °C (in a refrigerator). In some cases (as indicated in
where imax is the limiting current at saturating concentration of nitrate the text and figure captions) the enzyme coated electrode was covered
and KM is the apparent Michaelis constant. with a dialysis membrane (MW cutoff 3.5 kDa), presoaked in water, to
The pH dependence of the catalytic current was modeled with prevent protein loss to the bulk solution. The dialysis membrane was
Eq. (2), which is applicable for an enzyme that is deactivated by either carefully pressed onto the electrode with a Teflon cap and fastened
deprotonation at high pH (pKa1) or protonation at low pH (pKa2) and with a rubber O-ring to prevent leakage of the protein solution under
iopt is the maximum current at the pH optimum [38]. the membrane. The electro-active enzyme was always confined to the
electrode surface (not under diffusion control) while nitrate and nitrite
were able to diffuse to and from the electrode surface (across the mem-
iopt brane if present). The resulting enzyme-modified electrode was stored
i lim ðpHÞ ¼ : ð2Þ
1 þ 10ðpH−pK a1 Þ þ 10ðpK a2 −pHÞ at 4 °C in 100 mM phosphate buffer solution (pH 7.0) when not in use.

Scheme 3. Neurospora crassa nitrate reductase variants examined in this study. The first and last residues of the NR domains are indicated. An asterisk indicates the position of modified
amino acids which remove cofactor binding ability.
 P. Kalimuthu et al. / Biochimica et Biophysica Acta 1857 (2016) 1506–1513
Fig. 2. Square wave voltammetry of the four nitrate reductase variants (see Scheme 3) at
pH 8 (step 5 mV, amplitude 10 mV, frequency 10 Hz). EPG electrode preparation as given
in the Experimental section (a dialysis membrane covers the electrode).
3. Results and discussion
3.1. Enzyme film composition
The greatest challenge in direct enzyme electrochemistry is estab-
lishing heterogeneous (interfacial) electron transfer; electrochemical
communication with the redox active cofactors. This requires deliberate
adsorption of the enzyme onto the working electrode to remove the re-
striction of enzyme diffusion becoming rate limiting. The enzyme film
must fulfill three requirements: (i) retention of native enzyme function;
(ii) enable the redox cofactors to be in proximity of the electrode surface
and (iii) physi-sorption must be sufficiently strong that the film is in
contact with the electrode. Several different enzyme films were exam-
ined before the combination of polymyxin and polyethyleneimine was
chosen; both cationic hydrophilic compounds that have an affinity for
Fig. 3. (A) Square wave voltammetry of FAD-free NR at different pH values (a) 4, (b) 4.54, (c) 4.98, (d) 5.48, (e) 6, (f) 6.46, (g) 6.93, (h) 7.44, (i) 7.96, (j) 8.54, (k) 9 and (l) 9.49; step 5 mV,
amplitude 10 mV, frequency 10 Hz) and (B) the redox potentials as a function of pH. EPG electrode preparation as given in the Experimental section.
1509
the edge-plane pyrolytic graphite surface, which is negatively charged
at neutral pH due to surface carboxylate and phenolic functional groups.
Although heterogeneous electron transfer from the NR enzyme film was
demonstrated (see below) desorption of the film does occur over time
with the enzyme eventually being lost to the bulk solution. To avoid
this, the enzyme may be entrapped beneath a dialysis membrane. This
ensures that enzyme film desorption is a reversible process (dynamic
equilibrium) and under these conditions electrochemical activity was
sustained.
3.2. Non-turnover redox responses
We firstly examined the voltammetry of holo-NR, FAD-free NR;
heme-free NR and Mo-free NR (see Scheme 3) in the absence of nitrate.
Under ideal conditions electrochemical responses from the individual
cofactors may be identified. Although weak responses were identified
by conventional cyclic voltammetry, square wave voltammetry, a
more sensitive experiment, was employed to better resolve Faradaic re-
sponses above the background charging current (Fig. 2).
It is clear that the lower potential response is common to all en-
zymes bearing the FAD cofactor but absent in the FAD-free form,
while the higher potential peak is only found in NR enzymes bearing a
heme but conspicuously absent from the heme-free NR. With some con-
fidence we may assign the FAD response (at pH 8) to the peak at
− 270 mV vs NHE and the heme couple to − 140 mV vs NHE. No re-
sponses attributable to either the MoVI/V or MoV/IV couple were identi-
fied. This is not unexpected given that the Mo cofactor is the least
accessible of all and in this case is too remote from the electrode surface
for direct redox responses to be resolved. It also cannot be ruled out that
the MoVI/V and MoV/IV redox couples overlap with the heme response as
these redox potentials have not been determined independently before.
The pH dependence of the heme redox potential of FAD-free NR
varies linearly with pH (Fig. 3) but with a slope of only − 32 mV/pH
which, given the obligate one electron stoichiometry, is too small to
be consistent with proton associated electron transfer reaction.
The pH-dependence of the FAD response in both Mo-free NR
(Supporting Information Figure S1) and holo NR (Figure S2) shows
two distinct linear regions. Below pH 7 the slope of the plot is
− 59 mV/pH which is consistent with a 2e−/2H+ reaction (FAD/
FADH2) while above pH 7 the slope is about −26 mV/pH, which is con-
sistent with a 2e−/1H+ reaction (FAD/FADH−) i.e. the hydroquinone is
1510 P. Kalimuthu et al. / Biochimica et Biophysica Acta 1857 (2016) 1506–1513
Fig. 4. CVs obtained for EPG/PEI-PM-holo NR in the (a) absence and (b) presence of 1 mM
of nitrate in 100 mM phosphate buffer solution (pH 7) at a scan rate of 5 mV s−1. Solutions
were oxygen free.
singly deprotonated in this range. The break in the plot yields a pKa of
6.7 for the hydroquinone from this analysis. However, the FAD cofactor
is weakly (non-covalently) bound to NR and a dissociation constant of
Kd(FAD) = 0.61 μM has been reported [23]. The published assay for
NR [23] is conducted in the presence of excess FAD to ensure that the
enzyme has a full complement of cofactors so it was important to exam-
ine whether the FAD electrochemical response is from enzyme-bound
FAD or simply from dissociated FAD. A control experiment was carried
out using only FAD (enzyme free) and the same electrode modification
protocols used in Figs. 2 and 3 (and Figures S1 and S2). The data are
shown in Supporting Information Figure S3 and clearly the behavior of
the enzyme-free FAD response is indistinguishable from that seen in
Figures S1 and S2; the only discriminating feature is the weaker heme
response which naturally is absent from Figure S3. So we conclude
that dissociation of the FAD cofactor is facile under these conditions
and the low potential couple is simply of free FAD, not enzyme bound
FAD.
Fig. 5. CVs obtained for the increasing concentration of nitrate (a) 0, (b) 20, (c) 40, (d) 80, (e) 160, (f) 320, (g) 640, (h) 960 and (i) 1280 μM at the EPG/PEI-PM-holo NR electrode (no
membrane) in 100 mM phosphate buffer solution (pH 7) at a scan rate of 5 mV s−1. (B) Plot of the nitrate concentration dependence of the cathodic current at −250 mV vs NHE and
curve fit to Eq. (1).
3.3. Catalytic voltammetry
Fig. 4 illustrates the catalytic voltammetry of holo NR at an EPG
working electrode modified with polyethyleneimine (PEI) and poly-
myxin (PM) in 100 mM phosphate buffer. The CV trace does not reveal
any clear responses in the absence of nitrate (Fig. 4, curve a) Upon intro-
duction of 1 mM of nitrate into the electrochemical cell, a pronounced
sigmoidal cathodic wave emerges (Fig. 4, curve b) superposed on a slop-
ing baseline.
Both holo NR and the PEI/PM promoters must be present for cataly-
sis. Omission of the enzyme (Supporting Information, Figure S4) or the
PEI/PM promoters (Figure S5) leads to no nitrate reduction current. It is
possible that NR denatures at the unmodified graphite electrode surface
while the PEI/polymyxin coating on the electrode provides a more pro-
tein compatible interface. Alternatively it may be that NR alone does not
adsorb sufficiently strongly to the EPG electrode, a prerequisite for elec-
trochemical activity, without the PEI and polymyxin co-adsorbates. The
observed cathodic current is due to electrochemically driven enzymatic
reduction of nitrate by NR adsorbed on the electrode. It is noted that the
composition of promoters (5% polyethyleneimine and 5% polymyxin
B) plays a vital role facilitating electron transfer between NR and the
EPG electrode and the relative amounts of PEI and polymyxin are im-
portant. Many combinations were tried and we found that doubling
the amounts of the two promoters led to a 30% decrease in catalytic ac-
tivity which perhaps indicates electrode fouling by excess promoter. In-
terestingly catalytic activity is still seen if only PEI or only PM is used as a
promoter (supporting information, Figure S6) but the combination of
the two gives superior catalysis.
3.4. Substrate concentration dependence
Fig. 5A illustrates the effect of increasing nitrate concentration at the
EPG/PEI-PM-holo NR electrode in 100 mM phosphate buffer solution
(pH 7). In this case no dialysis membrane is present and nitrate may dif-
fuse freely to the enzyme film on the electrode. A well-defined sigmoi-
dal cathodic wave is observed around − 175 mV vs NHE upon
increasing additions of nitrate into the electrochemical cell (Fig. 5A).
This wave grows in amplitude with nitrate concentration and saturates
around 400 μM nitrate (Fig. 5B). Fig. 5B shows the cathodic reduction
current at −250 mV as a function of nitrate concentration. An apparent
Michaelis constant of KM = 39 μM was obtained from a fit to Eq. (1). This
value is significantly lower than reported from a conventional solution
 P. Kalimuthu et al. / Biochimica et Biophysica Acta 1857 (2016) 1506–1513
Fig. 6. CVs obtained in the presence of 1.2 mM of nitrate for the four different NR variants
adsorbed on the EPG/PEI-polymyxin electrode in 100 mM phosphate buffer solution
(pH 7) at a scan rate of 5 mV s−1.
assay (290 μM–440 μM) with NADPH the reductant [23]. The system
here is inherently simple and only comprises two reacting components;
holo NR continuously reactivated by the working electrode and nitrate
diffusing to the electrode. The higher apparent KM value seen in the
NADPH assay (a ternary system where holo NR, nitrate and NADPH
must all combine) may reflect rate limitations involving the NADPH
reductant.
If the same electrochemical experiment is carried out with a holo NR
modified EPG electrode covered with a dialysis membrane then very
similar results are obtained (Supporting Information Figure S7). The
only notable difference is a slight increase in the apparent Michaelis
constant from 39 μM (no membrane) to 72 μM (membrane). Assuming
that the membrane has no specific interaction with the enzyme this ob-
servation is consistent with mass transport of nitrate to the enzyme film
being slowed due to its passage across the membrane. Under these con-
ditions Eq. (1) does not strictly apply and the apparent KM value is no
longer the true Michaelis constant. We recently reported a similar
membrane influence on the current/substrate concentration profile of
Fig. 7. (A) CVs obtained for 1 mM nitrate at different temperature in 100 mM phosphate buffer (pH 7) at a scan rate of 5 mV s−1. (B) Arrhenius plot of the natural logarithm of maximum
catalytic current versus the reciprocal of absolute temperature (K).
1511
a different Mo enzyme in a mediated electrochemical system [39]. The
presence of the membrane does enhance the lifetime of the enzyme
electrode as any NR which dissociates from the electrode surface is
retained under the membrane rather than being irreversibly lost into
the bulk solution. This is illustrated in Supporting Information
Figure S8 where some loss of activity is seen in the absence of a mem-
brane over periods of hours.
The same experiments were repeated with the FAD-free, heme-free
and Mo-free NR enzymes. In the presence of nitrate it is apparent that
only holo NR and FAD-free NR are able to catalytically reduce nitrate
(Fig. 6). Indeed the current/nitrate concentration profile of FAD-free
NR is essentially the same as that of holo NR. This is an interesting fea-
ture and illustrates that the FAD cofactor is redundant in this electro-
chemically driven catalytic system. Electrons may enter through the
heme and then on to the Mo active site. The nitrate concentration profile
for the FAD-free NR (Supporting Information Figure S9) is similar to
holo NR and again the use of a membrane to cover the electrode leads
to a small increase in the apparent KM value.
3.5. Temperature dependence
The catalytic activity of NR was investigated at varying tempera-
tures in the range of 5 to 35 °C. It is apparent that the catalytic current
increases significantly with temperature as shown in Fig. 7A. The ap-
proximate activation energy E a was estimated from the Arrhenius
plot (natural logarithm of saturation catalytic current versus the recip-
rocal of absolute temperature (K)) as displayed in Fig. 7B. From this plot
(gradient −Ea/R), an activation energy of 26 kJ mol−1 was obtained.
3.6. pH dependence
We investigated the activity of the EPG/PEI-PM-holo NR electrode
towards the reduction of nitrate within the range 4.5 b pH b 10. A
bell-shaped profile is seen with an optimum of pH 6.8. The data were
modeled assuming that catalytic activity is lost by deprotonation
(pK1 = 8.4) or protonation (pK2 = 5.1) of an acid or base, respectively.
Further, the pH profile was independent of the direction of titration be-
tween pH 5 and 10 and reversible. As nitrate has no pKa in this range
these acid/base reactions most likely involve amino acid side chains in
proximity to the active site. Further, we found that the enzyme modified
electrode is stable while between pH 5 to 10 and the activity is
completely restored when the pH is returned to its optimal value. How-
ever, the electrode cannot be reactivated if the pH is taken below 4.5,
1512 P. Kalimuthu et al. / Biochimica et Biophysica Acta 1857 (2016) 1506–1513

Fig. 8. (A) CVs obtained for 1 mM nitrate at EPG/PEI-PM-NR electrode in different pH of mixed buffer solution at a scan rate of 5 mV s−1. (B) pH dependence of the maximum reduction
current at the EPG/PEI-PM-NR electrode. The solid curve is obtained from a fit to the experimental points using Eq. (2).

which suggests that the enzyme is denatured at acidic pH or that the
film is irreversibly dispersed at this pH.
In addition, it is apparent from Fig. 8A that the potential of the cata-
lytic wave is also pH dependent. The inflection points of the sigmoidal
curves (the catalytic wave potential) gradually shift to higher potential
at the pH is lowered but the decreasing catalytic currents at lower pH
make an accurate analysis of this pH dependence difficult. At best we
can say that the catalytic potentials seem to match the pH dependence
of the heme redox potential seen in Fig. 3B.
3.7. Oxygen dependence
The NR catalyzed reduction of nitrate at a high potential (ca.
− 150 mV vs NHE) offers the possibility for carrying out the electro-
chemical reaction even in the presence of oxygen. The data shown in
Fig. 9A (curves a (no nitrate) and b (1.2 mM nitrate)) are in the ab-
sence of oxygen and illustrate the enhancement of cathodic current
due to catalytic nitrate reduction. When the same solutions are ex-
posed to air (curves a′ and b′) there is no discrimination between
the two CVs which are both dominated by a dioxygen reduction cur-
rent. However sulfite is capable of acting as an in situ scavenger of
dioxygen [40]. Addition of 1 mM sulfite to the air saturated solutions
led to effective deoxygenation of the solutions and clear discrimina-
tion of the CVs in the absence (Fig. 9B (curve a)) and the presence of
nitrate (Fig. 9B (curve b)) was achieved. Interestingly there is a slight
enhancement of the cathodic current in the presence of increasing
amounts of sulfite (Supporting Information Figure S10). This is not
due to any catalytic reaction with sulfite as all forms of NR show no
activity for sulfite oxidation (Supporting Information Figure S11) de-
spite the active site similarities of eukaryotic nitrate-reducing and
sulfite-oxidizing Mo enzymes. The slight increase in cathodic current
may perhaps be due to the changes in ionic strength as sulfite con-
centration increases.
4. Conclusions
The unmediated catalytic electrochemistry of eukaryotic NR from
the fungus N. crassa was demonstrated for the first time on an EPG

Fig. 9. CVs of the EPG/PEA-PM-NR electrode (A): (curve a) in the absence of nitrate and absence of oxygen; (curve a′) in the absence of nitrate and presence of air (no purging); (curve b) in
the absence of oxygen and plus 1.2 mM nitrate and (curve b′) in the presence of 1.2 mM nitrate and air (B) obtained in the presence of oxygen and 1 mM sulfite (curve a) without and
(curve b) with 1.2 mM of nitrate. 100 mM phosphate buffer solution (pH 7), scan rate of 5 mV s−1.
 P. Kalimuthu et al. / Biochimica et Biophysica Acta 1857 (2016) 1506–1513
electrode with the mixture of promoters polyethyleneimine and poly-
myxin B. The enzyme modified electrode doesn't show any catalytic
current in the absence of promoters and the amount of promoter was
also vital in terms of maintaining keeping native function and electrical
communication with the electrode. The catalytic nitrate reduction cur-
rent increased non-linearly and an apparent Michaelis constant (KM)
was found to be 39 μM. A bell shaped pH profile was obtained with
pH optimum of pH 6.8. The catalytic cathodic wave is significantly in-
creased upon increasing the temperature and an activation energy of
26 kJ mol−1 was obtained. The FAD cofactor is shown to be nonessential
for electrocatalysis and evidently direct reduction of the heme cofactor
is sufficient for catalytic activity. The electrode could be used to detect
nitrate even without purging the solution with an inert gas but the
use of sulfite to deoxygenate the solution was necessary.
Conflicts of interest
The authors have no conflicts of interest in this submission.
Acknowledgements
Support from the Australian Research Council (DP150103345) is
gratefully acknowledged.
Appendix A. Supplementary data
FAD voltammetry (in the absence and presence of NR). Control ex-
periments for catalytic activity towards the nitrate reduction at EPG/
PEI-PM and unmodified EPG/NR electrodes as well as sulfite oxidation
experiments in the absence or presence of NR. Catalytic voltammo-
grams comparing nitrate reduction activity at EPG/PEI-NR, EPG/PM-
NR and EPG/PEI-PM-NR electrodes. SDS-PAGE analysis of all recombi-
nant NR forms is also provided. Supplementary data associated with
this article can be found in the online version, at http://dx.doi.org/10.
1016/j.bbabio.2016.04.001.
References
[1] J. Hirst, Elucidating the mechanisms of coupled electron transfer and catalytic reac-
tions by protein film voltammetry, Biochim. Biophys. Acta, Bioenerg. 1757 (2006)
225–239.
[2] F.A. Armstrong, G.S. Wilson, Recent developments in faradaic bioelectrochemistry,
Electrochim. Acta 45 (2000) 2623–2645.
[3] Y. Degani, A. Heller, Direct electrical communication between chemically modified
enzymes and metal electrodes. 2. Methods for bonding electron-transfer relays to
glucose oxidase and D-amino-acid oxidase, J. Am. Chem. Soc. 110 (1988) 2615–2620.
[4] E. Katz, I. Willner, Nanobiotechnology: integrated nanoparticle-biomolecule hybrid
systems: synthesis, properties, and applications, Angew. Chem. Int. Ed. 43 (2004)
6042–6108.
[5] P.V. Bernhardt, Enzyme electrochemistry — biocatalysis on an electrode, Aust. J.
Chem. 59 (2006) 233–256.
[6] D. Zheng, S.K. Vashist, K. Al-Rubeaan, J.H.T. Luong, F.-S. Sheu, Mediatorless ampero-
metric glucose biosensing using 3-aminopropyltriethoxysilane-functionalized
graphene, Talanta 99 (2012) 22–28.
[7] Y. Wang, L. Ge, C. Ma, Q. Kong, M. Yan, S. Ge, J. Yu, Self-powered and sensitive DNA
detection in a three-dimensional origami-based biofuel cell based on a porous Pt-
paper cathode, Chem. Eur. J. 20 (2014) 12453–12462.
[8] C. Wang, C. Yang, Y. Song, W. Gao, X. Xia, Adsorption and direct electron transfer
from hemoglobin into a three-dimensionally ordered macroporous gold film, Adv.
Funct. Mater. 15 (2005) 1267–1275.
[9] C. Mu, Q. Zhao, D. Xu, Q. Zhuang, Y. Shao, Silicon nanotube array/gold electrode for
direct electrochemistry of cytochrome c, J. Phys. Chem. B 111 (2007) 1491–1495.
[10] S.M. Dengale, A.K. Yagati, Y.-H. Chung, J. Min, J.-W. Choi, An electrochemical H2O2 de-
tection method based on direct electrochemistry of myoglobin immobilized on gold
deposited ITO electrode, J. Nanosci. Nanotechnol. 13 (2013) 6424–6428.
[11] T.d.F. Paulo, I.C.N. Diógenes, H.D. Abruña, Direct electrochemistry and
electrocatalysis of myoglobin immobilized on l-cysteine self-assembled gold elec-
trode, Langmuir 27 (2011) 2052–2057.
[12] G.-X. Ma, Y.-G. Wang, C.-X. Wang, T.-H. Lu, Y.-Y. Xia, Hemoglobin immobilized on
whisker-like carbon composites and its direct electrochemistry, Electrochim. Acta
53 (2008) 4748–4753.
[13] R. Huang, N. Hu, Direct electrochemistry and electrocatalysis with horseradish per-
oxidase in Eastman AQ films, Bioelectrochemistry 54 (2001) 75–81.
1513
[14] X. Chen, X. Peng, J. Kong, J. Deng, Facilitated electron transfer from an electrode to
horseradish peroxidase in a biomembrane-like surfactant film, J. Electroanal.
Chem. 480 (2000) 26–33.
[15] P. Kalimuthu, J. Tkac, U. Kappler, J.J. Davis, P.V. Bernhardt, Highly sensitive and stable
electrochemical sulfite biosensor incorporating a bacterial sulfite dehydrogenase,
Anal. Chem. 82 (2010) 7374–7379.
[16] P. Kalimuthu, M.D. Heath, J.M. Santini, U. Kappler, P.V. Bernhardt, Electrochem-
ically driven catalysis of Rhizobium sp. NT-26 arsenite oxidase with its native
electron acceptor cytochrome c552, Biochim. Biophys. Acta, Bioenerg. 1837
(2014) 112–120.
[17] S. Li, J. Xia, C. Liu, W. Cao, J. Hu, Q. Li, Direct electrochemistry of cytochrome c at a
novel gold nanoparticles-attached NH2+ ions implantation-modified indium tin
oxide electrode, J. Electroanal. Chem. 633 (2009) 273–278.
[18] K.-F. Aguey-Zinsou, P.V. Bernhardt, A.G. McEwan, J.P. Ridge, The first non-turnover
voltammetric response from a molybdenum enzyme: direct electrochemistry of
dimethylsulfoxide reductase from Rhodobacter capsulatus, J. Biol. Inorg. Chem. 7
(2002) 879–883.
[19] J.J. Gooding, Nanostructuring electrodes with carbon nanotubes: a review on
electrochemistry and applications for sensing, Electrochim. Acta 50 (2005)
3049–3060.
[20] C. Gao, Z. Guo, J.-H. Liu, X.-J. Huang, The new age of carbon nanotubes: an updated
review of functionalized carbon nanotubes in electrochemical sensors, Nanoscale 4
(2012) 1948–1963.
[21] G.-X. Wang, Y. Qian, X.-X. Cao, X.-H. Xia, Direct electrochemistry of cytochrome c on
a graphene/poly (3,4-ethylenedioxythiophene) nanocomposite modified electrode,
Electrochem. Commun. 20 (2012) 1–3.
[22] R. Hille, J. Hall, P. Basu, The mononuclear molybdenum enzymes, Chem. Rev. 114
(2014) 3963–4038.
[23] P. Ringel, J. Krausze, J. van den Heuvel, U. Curth, A.J. Pierik, S. Herzog, R.R. Mendel, T.
Kruse, Biochemical characterization of molybdenum cofactor-free nitrate reductase
from Neurospora crassa, J. Biol. Chem. 288 (2013) 14657–14671.
[24] W.H. Campbell, Nitrate reductase structure, function and regulation: bridging the
gap between biochemistry and physiology, Annu. Rev. Plant Physiol. Plant Mol.
Biol. 50 (1999) 277–303.
[25] W.H. Campbell, J.R. Kinghorn, Functional domains of assimilatory nitrate reductases
and nitrite reductases, Trends Biochem. Sci. 15 (1990) 315–319.
[26] C. Gonzalez, N. Brito, G.A. Marzluf, Functional analysis by site-directed mutagenesis
of individual amino acid residues in the flavin domain of Neurospora crassa nitrate
reductase, Mol. Gen. Genet. 249 (1995) 456–464.
[27] K. Fischer, G.G. Barbier, H.-J. Hecht, R.R. Mendel, W.H. Campbell, G. Schwarz, Struc-
tural basis of eukaryotic nitrate reduction: crystal structures of the nitrate reductase
active site, Plant Cell 17 (2005) 1167–1179.
[28] C. Probst, P. Ringel, V. Boysen, L. Wirsing, M.M. Alexander, R.R. Mendel, T. Kruse, Ge-
netic characterization of the Neurospora crassa molybdenum cofactor biosynthesis,
Fungal Genet. Biol. 66 (2014) 69–78.
[29] P. Kalimuthu, K. Fischer-Schrader, G. Schwarz, P.V. Bernhardt, Mediated electro-
chemistry of nitrate reductase from Arabidopsis thaliana, J. Phys. Chem. B 117
(2013) 7569–7577.
[30] P. Kalimuthu, K. Fischer-Schrader, G. Schwarz, P.V. Bernhardt, A sensitive and stable
amperometric nitrate biosensor employing Arabidopsis thaliana nitrate reductase, J.
Biol. Inorg. Chem. 20 (2015) 385–393.
[31] L.J. Anderson, D.J. Richardson, J.N. Butt, Catalytic protein film voltammetry from a re-
spiratory nitrate reductase provides evidence for complex electrochemical modula-
tion of enzyme activity, Biochemistry 40 (2001) 11294–11307.
[32] S.J. Elliott, K.R. Hoke, K. Heffron, M. Palak, R.A. Rothery, J.H. Weiner, F.A. Armstrong,
Voltammetric studies of the catalytic mechanism of the respiratory nitrate reductase
from Escherichia coli: how nitrate reduction and inhibition depend on the oxidation
state of the active site, Biochemistry 43 (2004) 799–807.
[33] S. Dementin, P. Arnoux, B. Frangioni, S. Grosse, C. Leger, B. Burlat, B. Guigliarelli, M.
Sabaty, D. Pignol, Access to the active site of periplasmic nitrate reductase: insights
from site-directed mutagenesis and zinc inhibition studies, Biochemistry 46 (2007)
9713–9721.
[34] V. Fourmond, M. Sabaty, P. Arnoux, P. Bertrand, D. Pignol, C. Leger, Reassessing the
strategies for trapping catalytic intermediates during nitrate reductase turnover, J.
Phys. Chem. B 114 (2010) 3341–3347.
[35] R.H. Garrett, A. Nason, Involvement of a B-type cytochrome in the assimilatory nitrate
reductase of Neurospora crassa, Proc. Natl. Acad. Sci. U. S. A. 58 (1967) 1603–1610.
[36] T.G.M. Schmidt, L. Batz, L. Bonet, U. Carl, G. Holzapfel, K. Kiem, K. Matulewicz, D.
Niermeier, I. Schuchardt, K. Stanar, Development of the twin-strep-tag® and its ap-
plication for purification of recombinant proteins from cell culture supernatants,
Protein Expr. Purif. 92 (2013) 54–61.
[37] T. Palmer, C.-L. Santini, C. Iobbi-Nivol, D.J. Eaves, D.H. Boxer, G. Giordano, In-
volvement of the narJ and mob gene products in distinct steps in the biosynthe-
sis of the molybdoenzyme nitrate reductase in Escherichia coli, Mol. Microbiol.
20 (1996) 875–884.
[38] M.S. Brody, R. Hille, The kinetic behavior of chicken liver sulfite oxidase, Biochemis-
try 38 (1999) 6668–6677.
[39] P. Kalimuthu, U. Kappler, P.V. Bernhardt, Catalytic voltammetry of the
molybdoenzyme sulfite dehydrogenase from Sinorhizobium meliloti, J. Phys. Chem.
B 118 (2014) 7091–7099.
[40] D. Quan, J.H. Shim, J.D. Kim, H.S. Park, G.S. Cha, H. Nam, Electrochemical determina-
tion of nitrate with nitrate reductase-immobilized electrodes under ambient air,
Anal. Chem. 77 (2005) 4467–4473.