Substrate profiling and aldehyde dismutase activity of the Kvβ2 subunit of the mammalian Kv1 potassium channel

Article abstract of DOI:10.1016/j.biocel.2010.09.002

Voltage-dependent potassium channels (Kv) are involved in various cellular signalling processes by governing the membrane potential of excitable cells. The cytosolic face of these α subunit-containing channels is associated with β subunits that can modulate channel responses. Surprisingly, the β subunit of the mammalian Kv1 channels, Kvβ2, has a high level of sequence homology with the aldo-keto reductase (AKR) superfamily of proteins. Recent studies have shown that Kvβ2 can catalyze the reduction of aldehydes and, most significantly, that channel function is modulated when Kvβ2-bound NADPH is concomitantly oxidized. As a result, the redox chemistry of this subunit is crucial to understanding its role in K⁺ channel modulation. The present study has extended knowledge of the substrate profile of this subunit using a single turnover fluorimetric assay. Kvβ2 was found to catalyse the reduction of aromatic aldehyde substrates such as 2, 3 and 4-nitrobenzaldehydes, 4-hydroxybenzaldehyde, pyridine 2-aldehyde and benzaldehyde. The presence of an electron withdrawing group at the position para to the aldehyde in aromatic compounds facilitated reduction. Aliphatic aldehydes proved to be poor substrates. We devised a simple HPLC-based assay to identify Kvβ2 reaction products. Using this assay we showed, for the first time, that Kvβ2 can catalyze a slow aldehyde dismutation reaction using 4-nitrobenzaldehyde as substrate and have identified the products of this reaction. The ability of Kvβ2 to carry out both an aldehyde reduction and a dismutation reaction is discussed in the light of current thinking on the role of redox chemistry in channel modulation.

Full text of DOI:10.1016/j.biocel.2010.09.002

The International Journal of Biochemistry & Cell Biology 42 (2010) 2012–2018
Contents lists available at ScienceDirect
The International Journal of Biochemistry
&
Cell Biology
journal homepage: www.elsevier.com/locate/biocel
Substrate profiling and aldehyde dismutase activity of the Kv␤ subunit of the
2
mammalian Kv1 potassium channel
a
a
b
a,∗
Kumari Alka , Barry J. Ryan , J. Oliver Dolly , Gary T.M. Henehan
a
School of Food Science and Environmental Health, Dublin Institute of Technology (DIT), Cathal Brugha Street, Dublin 1, Ireland
International Centre for Neurotherapeutics, Dublin City University, Dublin 9, Ireland
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 May 2010
Received in revised form 10 August 2010
Accepted 2 September 2010
Available online 15 September 2010
Voltage-dependent potassium channels (Kv) are involved in various cellular signalling processes by
governing the membrane potential of excitable cells. The cytosolic face of these ␣ subunit-containing
channels is associated with ␤ subunits that can modulate channel responses. Surprisingly, the ␤ sub-
unit of the mammalian Kv1 channels, Kv␤2, has a high level of sequence homology with the aldo–keto
reductase (AKR) superfamily of proteins. Recent studies have shown that Kv␤2 can catalyze the reduc-
tion of aldehydes and, most significantly, that channel function is modulated when Kv␤2-bound NADPH
is concomitantly oxidized. As a result, the redox chemistry of this subunit is crucial to understanding
Keywords:
␤
K
subunit
channels
+
+
its role in K channel modulation. The present study has extended knowledge of the substrate profile
of this subunit using a single turnover fluorimetric assay. Kv␤2 was found to catalyse the reduction of
aromatic aldehyde substrates such as 2, 3 and 4-nitrobenzaldehydes, 4-hydroxybenzaldehyde, pyridine
2-aldehyde and benzaldehyde. The presence of an electron withdrawing group at the position para to the
aldehyde in aromatic compounds facilitated reduction. Aliphatic aldehydes proved to be poor substrates.
We devised a simple HPLC-based assay to identify Kv␤2 reaction products. Using this assay we showed,
for the first time, that Kv␤2 can catalyze a slow aldehyde dismutation reaction using 4-nitrobenzaldehyde
as substrate and have identified the products of this reaction. The ability of Kv␤2 to carry out both an
aldehyde reduction and a dismutation reaction is discussed in the light of current thinking on the role of
redox chemistry in channel modulation.
Dismutation
Substrate specificity
Aldo–keto reductase
©
2010 Elsevier Ltd. All rights reserved.
1
. Introduction
Weng et al. (2006) showed that Kv␤2 was catalytically active
as an aldehyde reductase, identified substrates for this enzyme
activity, and demonstrated modulation of channel function when
Kv␤2-bound cofactor (NADPH) became oxidized (Weng et al.,
2006). This important finding focused attention on the redox prop-
erties of this channel subunit. It was suggested that oxidation of
the bound NADPH induces a conformational change in Kv␤, which
propagates to Kv1.1 to effect channel inactivation (Pan et al., 2008a;
Weng et al., 2006). Initially, two Kv␤2 substrates were identi-
fied (4-cyanobenzaldehyde and 4-carboxybenzaldehyde) as being
able to reversibly oxidize Kv␤2-bound NADPH (Weng et al., 2006).
Tipparaju et al. (2008) carried out a more detailed study of the Kv␤2
substrate specificity using a multiple turnover reaction and showed
that Kv␤2 could catalyse the reduction of aldehydes and ketones
with equal affinity.
Voltage-dependent potassium channels (Kv superfamily) play a
key role in the mammalian nervous system by governing the rest-
ing membrane potential of excitable cells thereby modulating the
frequency of action potentials and neurotransmitter release (Hille,
1
991). Four voltage sensing, integral membrane pore-forming Kv␣
+
subunits assemble to form the Shaker-related voltage-gated K
channels (Kv1 subfamily) (Kolb, 1990; Orlova et al., 2003). The aux-
illary ␤-subunit, Kv␤2, associated with the cytoplasmic face of Kv␣
proteins (Dolly et al., 1994; Parcej et al., 1992; Scott et al., 1994),
has a conserved catalytic core that shows a high level of sequence
homology with aldo–keto reductase (AKR) enzymes (Gulbis et al.,
1
999; Long et al., 2005). The catalytic C-terminal of Kv␤2 has a
tightly, but non-covalently, bound nicotinamide (NADPH) cofactor
and an aldehyde binding site. Of all the variants of Kv␤ genes (Kv␤1-
Understanding the details of the redox chemistry of Kv␤2 is
essential for elucidating its mode of action. It has been suggested
that it may respond to changes in the concentration of intracellu-
lar carbonyl substrates. Thus, an oxygen or metabolism-dependent
change in carbonyl substrate concentration could rapidly alter
the catalytic rate of Kv␤2, leading to cofactor oxidation and to
changes in Kv function (Tipparaju et al., 2008). The physiological
3
), Kv␤2 is the most widely distributed (Pongs et al., 1999; Xu and
Li, 1998) and it strongly associates with several Kv␣ proteins.
∗ Corresponding author. Tel.: +353 1 402 4408; fax: +353 1 402 4495.
E-mail address: gary.henehan@dit.ie (G.T.M. Henehan).
1
357-2725/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biocel.2010.09.002

K. Alka et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 2012–2018
2013
RCH(OH)2
ENADP
RCOOH
E
NADPH
2.2. Fluorescence assay of the single step hydride transfer rate
constant for Kvˇ2 substrates
RCH(OH)
2
RCOOH
ENADPH
RCHO
NADP
NADPH
The rates of single turnover hydride transfer reaction were mon-
E
ENADP
E
ψ
ψ
itored using a Perkin Elmer fluorescence spectrometer LS50B at
◦
2
2 C. Assays were carried out in a quartz cuvette containing 2.0 ␮M
RCH
2
OH
RCHO
ENADPH
Kv␤2 and varying substrate concentrations, in a final volume of
2.0 ml. All substrates were prepared in ethanol and then diluted
into 0.2 M potassium phosphate buffer, pH 7.5 to the required con-
centration, keeping the content of ethanol in the reaction mixture
equal to or below 1% (v/v). Ethanol at this concentration had no
effect on enzyme-bound NADPH. The excitation wavelength used
was 360 nm with a slit size of 15 nm and the emission spectra were
acquired from 300 to 600 nm at different times after initiating the
reaction by addition of substrate.
The decrease in emission at 454 nm due to enzyme-bound
NADPH followed a single exponential function over time, which
is consistent with the single-step hydride transfer reaction. Rate
constants were estimated from the slope of a plot of ln [NADPH]t
versus time (t) and expressed as the inverse of the exponential time
constant (see Fig. 3). The equation used was:
RCH
2
OH
ENADP
Scheme 1. Kinetic mechanism of aldehyde dismutation proposed for Kv␤2 at pH 7 5.
RCHO is an aldehyde, RCH(OH)2 is its corresponding hydrate, RCH2OH and RCOOH
are the corresponding alcohol and carboxylic acid respectively. The dismutation
of aldehyde substrate consists of two coupled half reactions. In the first half (the
upper pathway), hydrated aldehyde is oxidized irreversibly to the corresponding
carboxylic acid forming ENADPH. In the second half reaction (lower pathway), another
molecule of free aldehyde binds to the ENADPH complex and is reduced reversibly to
the corresponding alcohol. Hence, aldehyde is dismutated into equimolar concen-
tration of corresponding alcohol and carboxylic acid in a redox silent reaction with
no observable change in A340. ꢀ denotes the cofactor exchange steps. The steps
denoted by ꢀ are insignificant during dismutation as cofactor remains enzyme
bound throughout alternate oxidation and reduction.
substrates responsible for the reduction/oxidation of the enzyme-
bound cofactor are not known.
In addition to the well-characterized reversible oxidation of
alcohols, a number of pyridine-linked dehydrogenases/reductases
are known to catalyze the dismutation of aldehydes (Scheme 1)
to equimolar concentrations of the corresponding alcohol and
carboxylic acid (Daussmann et al., 1997; Henehan et al., 1995;
Mee et al., 2005; Oppenheimer and Henehan, 1995; Steinbuchel
and Schlegel, 1984; Winberg and McKinley-McKee, 1998). Kv␤2
bears some similarity to the microbial nicotinoprotein oxidore-
ductases in having a tightly bound, catalytically active NADPH
cofactor (Schenkels et al., 2000). Since nicotinoprotein oxidore-
ductases function as aldehyde dismutases, it seemed possible that
Kv␤2 might also catalyse aldehyde dismutation.
ln Ft = −kt + ln F0
where Ft the fluorescence intensity at time ‘t’, F0 is the initial fluo-
rescent intensity and ‘k’ is the rate constant. All assays were carried
out in duplicate.
2.3. HPLC assay to monitor Kvˇ2 catalysed dismutation of
4-nitrobenzaldehyde
Aldehyde dismutation was assayed in duplicate in 0.2 M potas-
+
sium phosphate buffer, pH 7.5 containing 0.2 mM NADP at
◦
37 C in the presence of ∼0.5 mg of Kv␤2 and 500 ␮M of 4-
Herein, we extend the information on the substrate profile for
this subunit in an attempt to identify compounds capable of inter-
acting with Kv␤2. We identify the reduced reaction products for its
optimum substrate, 4-nitrobenzaldehyde. In addition, the ability
of Kv␤2 to catalyze an aldehyde dismutation reaction is estab-
lished for 4-nitrobenzaldehyde and reduced and oxidized reaction
products identified.
nitrobenzaldehyde (final vol. of 2.0 ml). The reaction was quenched
by mixture with HPLC mobile phase (methanol/trifluroacetic
acid/water, 60:0.1:39.9) in the ratio 1:2. Aliquots (10 ␮l) of the
resultant mixture were analysed on a Nucleosil C18 (3.9 × 150 mm)
HPLC column, using a method adapted from Shearer et al. (1993)
with monitoring by a Millipore Waters (Mississauga, Canada)
liquid chromatography UV detector at 274 nm. Standard curves,
constructed for 4-nitrobenzaldehyde, 4-nitrobenzyl alcohol and 4-
nitrobenzoic acid were used to determine product concentrations
in the reaction mixtures.
2
. Materials and methods
2.1. Purification of Kvˇ2
3. Results
Kv␤2 gene from rat brain, cloned with an N-terminus His-
tag as described in Dolly et al. (1994), was provided by Dr. Oleg
3
.1. Substrate screening studies by fluorimetric assay
Shamotienko. Escherichia coli BL21 (DE3, plysS) transformed with
◦
pET15b-Kv␤2 construct were cultured at 37 C in LB medium
Kv␤2, when purified, contains bound NADPH. During purifica-
containing 50 ␮g/ml ampicillin. When the A600nm of the culture
tion care must be taken to isolate the protein with the dissociable
NADPH still bound. The presence of bound cofactor in recombi-
nant Kv␤2-NADPH was observed by UV and fluorimetric analysis
reached ∼0.8, the expression of Kv␤2 protein was induced for
1
4 h by the addition of IPTG to a final concentration of 1 mM
◦
at 25 C with constant shaking at 280 rpm (Stuart S1500 incuba-
tor/orbital shaker). Cells were resuspended in lysis buffer (20 mM
Tris–HCl, pH 7.9, 5 mM imidazole, 200 mM NaCl) and lysed by soni-
cation for 130 s on ice (VC-750 ultrasound generator, Sonics and
Materials Inc.). Cell debris were collected by centrifugation and
(Fig. 1). When excited at 360 nm Kv␤2 shows a prominent fluores-
cence peak at 454 nm corresponding to bound NADPH (Fig. 1 inset).
After addition of substrate to freshly purified Kv␤2 in 0.2 M phos-
phate buffer, pH 7.5, the transfer of a hydride from the Kv␤2-bound
NADPH to a substrate was observed as reduction in the 454 nm flu-
orescence peak (see Fig. 2A and B). This single turnover reaction
allowed an investigation of the ability of Kv␤2 to catalyze the reduc-
tion of various aldehydes. This assay has previously been described
for Kv␤1 (Pan et al., 2008a) and is more sensitive than the spec-
trophotometric assay first described for monitoring the reduction
of aldehydes by Kv␤2 (Weng et al., 2006).
the supernatant applied to a nickel-charged iminodiacetic acid
+
(
Novagen) column pre-equilibrated with lysis buffer. The Ni col-
umn was washed with 20 column volumes of the equilibration
buffer before elution with 20 mM Tris–HCl, pH 7.9, 200 mM NaCl,
and 300 mM imidazole. Fractions containing Kv␤2 were identified
by SDS-polyacrylamide gel electrophoresis, pooled and dialyzed
against 2 L of prechilled 0.2 M potassium phosphate, pH 7.5 for 36 h
with three changes.
A continuous spectrophotometric assay has also been described
for the aldehyde reducing activity of Kv␤2 (Tipparaju et al., 2008).

2
014
K. Alka et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 2012–2018
Fig. 3. Initial velocity of the hydride transfer reaction plotted versus Kv␤2 con-
centration for the single turnover reaction. The substrate (4-nitrobenzaldehyde)
concentration was kept constant at 0.5 mM and initial velocities were measured by
taking the slope of the linear part of the plot between fluorescence intensity and
time. The concentration of enzyme employed was between 0.04 and 0.14 mg/ml.
Fig. 1. UV/Vis and fluorescence spectra of freshly purified Kv␤2 protein. The UV
spectrum of Kv␤2 gave two absorbance peaks at 280 nm and 360 nm corresponding
to the protein and protein-bound NADPH, respectively (Liu et al., 2001; Schenkels
et al., 2000). The inset is a fluorescence emission scan of purified Kv␤2 showing an
emission band at 454 nm for bound NADPH. The excitation wavelength was 360 nm.
tures were reduced by Kv␤2, with aromatic substrates bearing a
strongly electron withdrawing substituent giving the largest rate
constants (Table 1). Of these, 4-nitrobenzaldehyde gave the high-
Such a multiple turnover assay is possible since the oxidized
cofactor resulting from aldehyde reduction will exchange with
externally added NADPH to recommence the catalytic cycle. In the
present study, the multiple turnover assay was only used to iden-
tify the reduced product, 4-nitrobenzylalcohol derived from the
reduction of 4-nitrobenzaldehyde. In our hands, multiple turnover
initial rates measurements were complicated by slow bleaching
of enzyme bound NADPH. Thus, multiple turnover initial rates
were not used for kinetic constant estimations. The use of a single
turnover reaction allows for the screening of a range of substrates.
More detailed studies of kinetic constants were not pursued due
to the low activity of Kv␤2 with the carbonyls tested as previously
noted (Tipparaju et al., 2008).
◦
est rate constant. All reactions were carried out at 22 C due to the
aggregation of Kv␤2 at higher temperatures. This assay is similar
to that previously used to identify substrates for Kv␤1 (Pan et al.,
2
008a).
The effect of the position of the electron withdrawing
group on the pseudo-first order rate constant was explored
by comparing 2-nitrobenzaldehyde, 3-nitrobenzaldehyde and 4-
nitrobenzaldehyde. It was found that the rate constants for
these substrates decreased in the order: 4-nitrobenzaldehyde > 3-
nitrobenzaldehyde > 2-nitrobenzaldehyde. All nitro-substituted
benzaldehydes had higher rate constants than any of the other
substrates tested. Thus, it was clear that the position of this group
strongly influenced catalysis, with the para position being optimal.
The single turnover rate constant for the hydride transfer reaction
When the initial rate of aldehyde reduction was measured by
taking the slope of the linear part of the plot of fluorescence inten-
sity at 454 nm versus time; the initial rate observed was linearly
dependent on Kv␤2 concentration (Fig. 3).
−1
was estimated to be 0.22 min for 0.5 mM 4-nitrobenzaldehyde
and assigned as 100%. Activities with all other substrates were
expressed as a percentage of 4-nitrobenzaldehyde reductase activ-
ity (Table 1). 4-Cyanobenzaldehyde, one of the original substrates
−
1
3.2. Substrate profiling
reported for Kv␤2, had a rate constant (0.014 min ) more than
0-fold lower than 4-nitrobenzaldehyde. The rate constant with 4-
1
Pseudo-first order rate constants were estimated for a range
cyanobenzaldehyde was similar to that for benzaldehyde showing
that the nitro group significantly accelerated the Kv␤2-mediated
aldehyde reduction. 2-Cyanobenzaldehyde had a rate constant half
that of 4-cyanobenzaldehyde, supporting the observation that the
position of the electron withdrawing group para to the aldehyde
of aromatic and aliphatic aldehydes, ketones and dicarbonyls. All
substrates were tested at a concentration of 0.5 mM except for cer-
tain aliphatic substrates that required a higher concentration to
yield observable activity. A range of aldehydes of various struc-
Fig. 2. Kv␤2 mediated reduction of 4-nitrobenzaldehyde. (A) Fluorescence emission spectra of purified Kv␤2 protein at different time points after mixing with 0.5 mM 4-
nitrobenzaldehyde. (B) Fluorescence intensity at 454 nm plotted versus time for Kv␤2 after mixing with 4-nitrobenzaldehyde. This plot was fitted to a first order exponential
decay to obtain a pseudo first order rate constant for aldehyde reduction.

K. Alka et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 2012–2018
2015
Table 1
Comparison of the rate constants (min−1) for the single turnover hydride transfer reaction (first order) for Kv␤2 substrates.
Group
Substrate
Concentration of substrate (mM)
Rate constant (min−1)
Relative rate constant (%)
Aromatic aldehydes
4-Nitrobenzaldehyde
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.220
0.116
0.048
0.011
0.017
0.044
0.007
0.014
-
100a
53
22
5
8
20
3
6
-
-
-
3
2
-Nitrobenzaldehyde
-Nitrobenzaldehyde
Pyridine 2 aldehyde
Benzaldehyde
4
2
4
4
4
-Hydroxybenzaldehyde
-Cyanobenzaldehyde
-Cyanobenzaldehyde
-Chlorobenzaldehyde
-Fluorbenzaldehyde
-
-
-
Tolualdehyde
Anisaldehyde
-
Aliphatic aldehydes
Formaldehyde
10
10
10
10
10
0.5
0.5
20
-
-
16
5
2
1
1
-
5
-
2-Methlybutyraldehyde
0.035
0.011
0.004
0.002
0.001
-
Butanal
Crotanaldehyde
Valeraldehyde
Hexanal
Octanal
Succinic semialdehyde
Cinnamaldehyde
0.012
-
Ketones
Isatin
0.5
0.5
-
-
2
4-Nitroacetophenone
0.004
Dicarbonyls
Methlyglyoxal
Phenylglyoxal
10
0.5
0.015
0.005
7
3
a
−1
The single turnover hydride transfer rate for 4-nitrobenzaldehyde was 0.22 min . Enzyme activity was measured as described in Section 2. Activities are expressed as a
percentage of 4-nitrobenzaldehyde reductase activity. All estimates of rate constants are the mean of triplicate measurements.
is important for optimal enzyme activity. This study indicates that
a range of known AKR substrates may be reduced by this protein.
Surprisingly, Kv␤2 showed no detectable activity with the aromatic
aldehydes 4-chlorobenzaldehyde and 4-fluorobenzaldehyde.
Pyridine-2-aldehyde, a well-known AKR substrate, gave a rate
valine) that can cross the blood–brain barrier (Oldendorf, 1971a,b)
where they serve as substrates for meeting parenchymal energy
demand (Pellerin, 2005; Yudkoff et al., 1996). Formaldehyde, cin-
namaldehyde andoctanal showedno detectableactivity withKv␤2.
−1
constant of 0.011 min . An aromatic alpha-keto aldehyde, phenyl-
3.3. HPLC-based assay for the identification of reduced product of
−1
glyoxal, had a rate constant (0.0054 min ) about half that of
the 4-cyanobenzaldehyde while an aliphatic alpha-keto aldehyde,
methylglyoxal was a poorer substrate than both phenylglyoxal and
multiple turnover reaction using 4-nitrobenzaldehyde as substrate
Kv␤2-mediated formation of reduced product using a phospho-
lipid aldehyde, 1-palmitoyl-2-oxovaleroyl phosphatidylcholine has
been reported (Tipparaju et al., 2008). To date, 4-nitrobenzaldehyde
is the most effective substrate for Kv␤2. Since this aldehyde has a
high extinction coefficient, it was used as the basis for the devel-
opment of a simple HPLC assay for aldehyde reduction/oxidation
The HPLC assay using this aldehyde proved to be a convenient and
4
-cyanobenzaldehyde. These data indicated the importance of acti-
vation of the aldehyde group for catalysis, and the importance of
the aromatic moiety, presumably for binding.
4
-Hydroxybenzaldehyde, with a moderately strong electron
donating group in position para to the aldehyde was 5-fold less
active than the 4-nitrobenzaldehyde but better than benzaldehyde.
Aromatic aldehydes with moderate electron donating groups such
as tolualdehyde and anisaldehyde showed no detectable activity.
These data seem to indicate a complex picture of substrate pref-
erence. Nitrobenzaldehyde derivatives are clearly among the best
aromatic substrates but other aromatic substrates, even those with
electron withdrawing substituents such as chlorobenzaldehyde,
showed no clear pattern of preference.
A para-substituted ketone, 4-nitroacetophenone, had an activity
5
0-fold less than 4-nitrobenzaldehyde indicating that Kv␤2 had
rather poor affinity towards ketones. This finding contrasts with
the conclusion of Tipparaju et al. (2008) who indicated that Kv␤2
could reduce aldehyde and ketone substrates with equal affinity.
However, these workers noted that when presented with a choice
between aldehyde and keto functional groups on a single substrate,
4
-oxo-nonenol, only the aldehyde was reduced.
Aliphatic aldehydes are poor substrates with rate constants
<100-fold or lower than aromatic aldehydes even at concentrations
up to 10 mM. In this study, 2-methylbutyraldehyde was identified
as a Kv␤2 substrate with a rate constant of ∼10-fold higher than all
other tested aliphatic aldehydes (Fig. 4). In mammals, isoleucine is
one of three branched chain amino acids (leucine, isoleucine and
Fig. 4. Kv␤2 mediated reduction of 2-methylbutyraldehyde. Fluorescence emis-
sion spectra of purified Kv␤2 at different time points after mixing with 10 mM
2-methylbutyraldehyde which had a rate constant of ∼10-fold higher than all other
tested aliphatic aldehydes.

2
016
K. Alka et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 2012–2018
Fig. 5. (A) Standard HPLC profile of a mixture of equimolar (1 mM) concentrations of 4-nitrobenzylalcohol (4-NBalc), 4-nitrobenzaldehyde (4-NBald) and 4-nitrobenzoic acid
4-NBacid) containing 200 ␮M of NADP; the retention times are 3.83, 4.33, 5.13 and 2.19 min, respectively. (B) Identification of products of Kv␤2 mediated multiple turnover
(
aldehyde reduction reaction. The data show formation of 4-nitrobenzylalcohol (reduced product) with a retention time of 3.83 min in a reaction mixture containing ∼10 ␮M
◦
Kv␤2, 0.2 M potassium phosphate buffer (pH 7.5), 0.2 mM NADPH and 0.5 mM of 4-nitrobenzaldehyde. This mixture was incubated for 15 min at 37 C. The peaks at 4.33 and
2
.17 min represent unreacted 4-nitrobenzaldehyde and NADPH, respectively.
sensitive way to monitor aldehyde reduction and dismutation and
will be broadly applicable to the assay of aldehyde dismutating
enzymes. Separation of 4-nitrobenzaldehyde, and its correspond-
ing alcohol and acid derivatives by HPLC is shown in Fig. 5A.
In the presence of externally added NADPH, Kv␤2 catalyses
a multiple turnover reaction (see above). When the HPLC assay
was used to monitor its reduction of 4-nitrobenzaldehyde, in the
presence of NADPH, the formation of reduced product, i.e., 4-
nitrobenzylalcohol (Fig. 5B) was clearly observed.
hence, this enzyme was chosen to validate the assay conditions in
this study (see below) (Oppenheimer and Henehan, 1995).
Purified Kv␤2.NADPH was first treated with 250 ␮M 4-
nitrobenzaldehyde to oxidize all enzyme bound cofactor before
dialysis for 24 h against 0.2 M potassium phosphate buffer, pH
7.5 to remove remaining substrates. This step ensured that all
Kv␤2 bound cofactor was in the oxidized form (Kv␤2.oxd). The
reaction mixture containing ∼10 ␮M Kv␤2.oxd, 200 ␮M of NADP,
0.2 M potassium phosphate buffer (pH 7.5) and 0.5 mM of 4-
◦
nitrobenzaldehyde was incubated at 37 C. Aliquots were removed
3
.4. Dismutation of aldehydes by Kvˇ2
from the reaction mixture every hour and the products analysed
by HPLC. This analysis revealed a clear peak for 4-nitrobenzoic
acid (oxidized product) along with an equivalent amount of 4-
nitrobenzylalcohol (reduced product) (see Fig. 6).
The HPLC based dismutation assay method was validated by
using horse liver alcohol dehydrogenase (HLADH) and oxidized
cofactor (NAD, 200 ␮M) under the same conditions for which dis-
mutation had been reported (Oppenheimer and Henehan, 1995)
A number of pyridine linked alcohol dehydrogenases/aldo–keto
reductases have been reported to catalyse the dismutation of alde-
hydes (Daussmann et al., 1997; Henehan et al., 1995; Mee et al.,
2
1
005; Oppenheimer and Henehan, 1995; Steinbuchel and Schlegel,
984; Winberg and McKinley-McKee, 1998). This reaction has
previously been described for horse liver alcohol dehydrogenase;
Fig. 6. Identification of the products of Kv␤2 mediated dismutation reaction. HPLC profile of the reaction mixture containing ∼10 ␮M Kv␤2.oxd, 0.2 M potassium phosphate
◦
buffer (pH 7.5), 0.2 mM NADP and 0.5 mM of 4-nitrobenzaldehyde incubated for 7 h at 37 C showing the formation of 4-nitrobenzylalcohol (reduced product) along with an
equimolar concentration of 4-nitrobenzoic acid (oxidized product, peak indicated in red) giving respective retention times of 3.85 and 5.09 min. The peaks at 4.35 represent
unreacted 4-nitrobenzaldehyde. This experiment was repeated several times with identical results. Control experiments with no enzyme showed no alcohol or acid formation
over the same time period.

K. Alka et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 2012–2018
2017
presumably slow and expected to be rate limiting (Oppenheimer
and Henehan, 1995).
4. Discussion
The substrate screening data in this study help to further define
Kv␤2 substrate specificity and establish that Kv␤2 has a preference
for aromatic aldehydes with strong electron withdrawing groups
at a position para to the aldehyde moiety. It has been suggested
that in vivo modulation of channel activity may be mediated by an
aldehyde substrate. The identity of such a substrate is difficult to
define when an enzyme has the broad specificity of Kv␤2. Most
of the substrates we and others have examined are unlikely to be
physiologically significant and physiological analogs are not easily
identified. It has been suggested that this subunit is involved in the
reduction of membrane-derived oxidized lipids (Tipparaju et al.,
2
008). However, it is difficult to envisage how the production of
such aldehydes could be controlled sufficiently to regulate channel
activity since they arise from a variety of processes (both enzymic
and non-enzymic) mostly as a result of oxidative stress and would
be expected to be competitive substrates.
The suggestion that aldehydes have a direct role in channel mod-
ulation (Tipparaju et al., 2008; Weng et al., 2006) presents some
challenges. Firstly, most aldehydes are reactive species that readily
react with, and modify, cellular proteins. Most cells contain efficient
enzyme systems to remove aldehydes in order to avoid such harm-
ful reactions (Kawamura et al., 2002; Svensson et al., 1996). Thus,
it would be necessary for such aldehydes to be generated in close
proximity to Kv␤2 in order to be effective. However, such a system
has not yet been identified. Secondly, there is the issue of com-
petition between aldehydes for Kv␤2 due to its broad specificity.
Many aldehydes generated in cells might be expected to compete
for the relatively open active site of Kv␤2; this necessitates very
tight control of aldehyde access to Kv␤2 for effective, controlled,
channel modulation. The requirement for tight control may indi-
cate a high degree of specificity for the true physiological substrate
for this reaction. As a result, the true substrate may prove difficult
to identify. A final difficulty lies in the requirement that oxidized
Kv␤2 bound cofactor must be reduced in order to reset the channel
modulation signal following its oxidation by an aldehyde substrate.
It might be argued that the slow rate of aldehyde reduction by
Kv␤2 makes it difficult to envisage that such a reaction could mod-
ulate channel activity in real time as has been proposed (Tipparaju
et al., 2008; Weng et al., 2006). A possible explanation lies in the
requirement thatthe true substrate, asyet unidentified, has asignif-
icantly higher turnover than those identified in this and previous
studies. Another possibility is that turnover may be upregulated
by association of Kv␤2 with the channel subunits or with other
proteins or indeed by other signaling events.
Fig. 7. (A) Progress curves showing 4-nitrobenzoic acid formation resulting from the
dismutation of 4-nitrobenzaldehyde by horse liver alcohol dehydrogenase (HLADH).
The data shows the formation of 4-nitrobenzoic acid at different starting concen-
trations of 4-nitrobenzaldehyde (250 ␮M, 500 ␮M, 1 mM and 5 mM). After addition
of the HLADH to the assay mixture aliquots were removed at 15, 30, 45, 60, 75,
9
0, 105 and 120 min and the amount of 4-nitrobenzoic acid formed was estimated
by HPLC. Results are expressed as the mean of duplicate measurements. The con-
centration of HLADH and NAD+ used was 20 ␮g/ml and 200 ␮M, respectively, in a
final reaction volume of 2 ml. (B) Dismutation of 4-nitrobenzaldehyde by Kv␤2. The
data show the Kv␤2 mediated dismutation of 4-nitrobenzaldehyde with externally
added NADP. After addition of substrate (500 ␮M 4-nitrobenzaldehyde) to the assay
mixture, aliquots were removed at 0.5, 1, 2, 3, 4, 5, 6 and 7 h and the amount of 4-
nitrobenzoic acid formed was estimated by HPLC. Results are expressed as the mean
of duplicate measurements. Assays were carried out as described in Section 2.
(
see Fig. 7A). Dismutation may be conveniently monitored by fol-
lowing the accumulation of 4-nitrobenzoic acid, a reaction product
unique to the dismutation reaction. Kv␤2 was shown to catalyze
a slow dismutation reaction where the amount of 4-nitrobenzoic
acid produced was found to be >1400 times less than HLADH in
2
.0 h (see Fig. 7B).
Dismutation of aldehydes has proven difficult to monitor since
it is a redox silent reaction (Henehan et al., 1995; Oppenheimer
and Henehan, 1995). In this study we have devised a sensitive
HPLC assay, based on the absorbance of nitrobenzaldehyde and its
derivatives, capable of detecting both net aldehyde reduction and
aldehyde dismutation. The assay will be generally useful for screen-
ing for dismutation, albeit confined to a single substrate. From these
studies it is clear that Kv␤2 can dismutate aldehydes but at a rate
far slower than HLADH. This is consistent with the slow turnover
of aldehyde substrates by aldo–keto reductases from a number of
sources when compared to corresponding alcohol dehydrogenases
(Barski et al., 2008). Although dismutation was slow it was clear
that acid formation was occurring. Control experiments showed
no acid formation under the same conditions thereby discounting
the possibility that acid formation was due to non-enzymatic alde-
hyde oxidation. Moreover, acid formation was time dependent and
Control reactions with no enzyme were incorporated in every
experiment. Equimolar concentrations of 4-nitrobenzaldehyde,
+
4
-nitrobenzoic acid, 4-nitrobenzyl alcohol and NADP were incu-
◦
bated at 37 C as a control to monitor any possible background
chemical reactions. The data clearly showed that the products of
the reaction observed were due to Kv␤2 catalyzed dismutation.
3
.5. Comparison of Kvˇ2-mediated multiple turnover and
dismutation reactions
The amount of reduced product (4-nitrobenzylalcohol) formed
from the multiple turnover reaction over the course of 30 min was
30 times higher than that seen for the dismutation reaction. This
was expected since dismutation involves the generation of Kv␤2-
bound NADPH by the formation of 4-nitrobenzoic acid which is
∼

2
018
K. Alka et al. / The International Journal of Biochemistry & Cell Biology 42 (2010) 2012–2018
acid was formed in equimolar concentrations to that of the alcohol
product.
chloral hydrate metabolism to trichloroacetic acid and trichloroethanol. BBA
991;1075(1):61–7.
Henehan GTM, Chang SH, Oppenheimer NJ. Aldehyde dehydrogenase activ-
ity of Drosphila melanogaster alcohol dehydrogenase: burst kinetics at high
pH and aldehyde dismutase activity at physiological pH. Biochemistry
1
If it were postulated that channel activity was controlled by
aldehyde dismutation it would obviate the necessity to regener-
ate reduced Kv␤2-bound cofactor since this reaction alternately
oxidizes and reduces NADP(H). Notwithstanding this point, it is
difficult to argue for a significant physiological role for the slow dis-
mutation observed here with 4-nitrobenzaldehyde but we cannot
discount the possibility that dismutation might be faster for other,
as yet untested, substrates. It is of interest for inhibitor design to
consider that Kv␤2 is capable of generating a carboxylate, presum-
ably from the hydrated form of the aldehyde, at its active site. A
similar dismutation reaction has previously been reported for the
carbonyl reductase from pig lung (Hara et al., 1991).
1995;34:12294–301.
Hille B. Ion channels of excitable membranes. 3rd ed. Sunderland, MA: Sinauer
Associates; 1991.
Kawamura M, Eisenhofer G, Kopin IJ, Kador PF, Lee YS, Fujisawa S, et al. Aldose
reductase: an aldehyde scavenging enzyme in the intraneuronal metabolism
of norepinephrine in human sympathetic ganglia. Auton Neurosci: Basic Clin
2002;96:131–9.
Kolb HA. Potassium channels in excitable and non-excitable cells. Rev Physiol
Biochem Pharmacol 1990;115:51–91.
Liu SQ, Jin H, Zacarias A, Srivastava S, Bhatnagar A. Binding of pyridine nucleotide
+
coenzymes to the ␤-subunit of the voltage-sensitive K channel. J Biol Chem
2001;15:11812–20.
Long SB, Campbell EB, MacKinnon R. Voltage sensor of Kv1.2: structural basis of
electromechanical coupling. Science 2005;309:903–8.
Mee B, Kelleher D, Frias J, Malone R, Tipton KF, Henehan GTM, et al. Characteri-
zation of cinnamyl alcohol dehydrogenase of Helicobacter pylori, an aldehyde
dismutating enzyme. FEBS J 2005;272:1255–64.
Oldendorf WH. Brain uptake of radiolabelled amino acids, amines, and hexoses after
arterial injection. Am J Physiol 1971a;221:1629–39.
Oldendorf WH. Uptake of radiolabelled essential amino acids by brain following
arterial injection. Proc Exp Biol Med 1971b;136:385–6.
Recent findings have shown that cortisone binds to Kv␤2 and
+
promotes its dissociation from the K channel (Pan et al., 2008b).
Structural studies identified two types of cortisone binding sites
on Kv␤2, one close to the bound NADPH and the other at the inter-
face of two neighbouring Kv␤ subunits (Pan et al., 2008b). Only
the interface binding site was reported to be required for chan-
nel modulation. These findings indicate that cortisone may act to
further regulate aldehyde-mediated channel modulation by pro-
moting Kv␤2 dissociation from ␣ subunits. It is conceivable that
fine control of channel activity could be modulated by a specific
aldehyde substrate but that cortisone could abolish such control
by dissociation of Kv␤2 from the ␣ subunits. The relationship of
Oppenheimer NJ, Henehan GTM. Horse liver alcohol dehydrogenase-catalyzed alde-
hyde oxidation. The sequential oxidation of alcohols to carboxylic acids under
NADH recycling conditions. Adv Exp Medical Biol 1995;372:407–15.
Orlova EV, Papakosta M, Booy FP, Van Heel M, Dolly JO. Voltage-gated K+ channel
from mammalian brain: 3D structure at 18A of the complete (alpha)4(beta)4
complex. J Mol Biol 2003;326:1005–12.
Pan Y, Weng J, Cao Y, Bhosle RC, Zhou M. Functional coupling between the
Kv1.1 channel and aldo–keto reductase Kv␤1.
J Biol Chem 2008a;283:
+
these two very different modes of K channel modulation remains
8634–42.
Pan Y, Weng J, Kabaleeswaran V, Li H, Cao Y, Bhosle RC, et al. Cortisone dissoci-
to be explored. In particular, data on the influence of low concen-
trations of cortisone on channel activity would be of great interest
since reports to date used concentrations of 500 ␮M and above that
may be difficult to attain physiologically. Interestingly, cortisone
was shown to bind to the active site of Kv␤2 but was not a sub-
strate. Thus, cortisone is expected to act as an inhibitor of aldehyde
reduction in vitro. Whether aldehyde substrates act in opposition
to cortisone or in synergy with cortisone as channel modulators
remains to be elucidated.
+
ates the Shaker family K channels from their beta subunits. Nat Chem Biol
2008b;4:708–14.
Parcej DN, Scott VE, Dolly JO. Oligomeric properties of alpha-dendrotoxin-
sensitive potassium ion channels purified from bovine brain. Biochemistry
1992;31:11084–8.
Pellerin L. How astrocytes feed hungry neurons. Mol Neurobiol 2005;32:59–72.
Pongs O, Leicher T, Berger M, Roeper J, Bahring R, Wray D, et al. Functional and
+
molecular aspects of voltage-gated K channel beta subunits. Ann N Y Acad Sci
1999;868:344–55.
Schenkels P, Duine JA, Nicotinoprotein. (NADH-containing) alcohol dehydrogenase
from Rhodococcus erythropolis DSM 1069 an efficient catalyst for coenzyme-
independent oxidation of a broad spectrum of alcohols and the interconversion
of alcohols and aldehydes. Microbiology 2000;146:775–85.
Scott VE, Rettig J, Parcej DN, Keen JN, Findlay JB, Pongs O, et al. Primary structure of
a beta subunit of alpha-dendrotoxin-sensitive K+ channels from bovine brain.
Proc Natl Acad Sci USA 1994;91:1637–41.
Shearer GL, Kim K, Lee KM, Wang CK, Plapp BV. Alternative pathways and reactions
of benzyl alcohol and benzaldehyde with horse liver alcohol dehydrogenase.
Biochemistry 1993;32:11186–94.
Steinbuchel A, Schlegel HG. A multifunctional fermentative alcohol dehydrogenase
from the strict aerobe Alcaligenes eutrophus: purification and properties. Eur J
Biochem 1984;141:555–64.
Acknowledgements
This work was supported by ABBEST research scholarship (to
K.A.) from the Dublin Institute of Technology and a Research Pro-
fessorship (to J.O.D) from Science Foundation Ireland. The authors
thank Dr. Oleg Shamotienko for help with expressing and purifying
Kv␤2.
Svensson S, Lundsjo A, Cronholm T, Hoog JO. Aldehyde dismutase activity of human
liver alcohol dehydrogenase. FEBS Lett 1996;394:217–20.
Tipparaju SM, Barski OA, Srivastava S, Bhatnagar A. Catalytic mechanism and
substrate specificity of the subunit of the voltage-gated potassium channel.
Biochemistry 2008;47:8840–54.
References
Barski OA, Tipparaju SM, Bhatnagar A. The aldo–keto reductase superfam-
ily and its role in drug metabolism and detoxification. Drug Metab Rev
Weng J, Cao Y, Moss N, Zhou M. Modulation of voltage dependant Shaker
2
008;40(4):553–624.
family potassium channels by an aldo–keto reductase.
006;281:15194–200.
Winberg JO, McKinley-McKee JS. Drosophila melanogaster alcohol dehydro-
genase: mechanism of aldehyde oxidation and dismutation. Biochem
998;329:561–70.
J Biol Chem
Daussmann T, Aivasidis A, Wandrey C. Kinetic data and new enzymatic activities
of Methanosarcina barkeri grown on methanol as sole carbon source. Water Sci
Technol 1997;36:175–82.
Dolly JO, Rettig J, Scott VE, Parcej DN, Wittkat R, Sewing S, et al. Oligomeric and
subunit structures of neuronal voltage-sensitive K+ channels. Biochem Soc Trans
2
J
1
Xu J, Li M. Auxiliary subunits of shaker-type potassium channels. Trends Cardiovasc
Med 1998;8:229–34.
Yudkoff M, Daikhin Y, Grunstein L, Stern J, Pleasure D, Nissim I. Astrocyte leucine
metabolism: significance of branched-chain amino acid transamination. J Neu-
rochem 1996;66:378–85.
1
994;22:473–8.
Gulbis JM, Mann S, MacKinnon R. Structure of a voltage-dependent K1 channel ␤
subunit. Cell 1999;90:943–52.
Hara A, Yamamoto H, Deyashiki Y, Nakayama T, Oritani H, Sawada H. Aldehyde
dismutation catalyzed by pulmonary carbonyl reductase: kinetic studies of