Catalytic reduction of carbonyl groups in oxidized PAPC by Kvβ2 (AKR6)

Article abstract of DOI:10.1016/j.cbi.2011.01.032

The β-subunits of the voltage-gated potassium channel (Kvβ) belong to the aldo-keto reductase superfamily. The Kvβ-subunits dock with the pore-forming Kv α-subunits and impart or accelerate the rate of inactivation in Kv channels. Inactivation of Kv currents by Kvβ is differentially regulated by oxidized and reduced pyridine nucleotides. In mammals, AKR6 family is comprised of 3 different genes Kvβ1-3. We have shown previously that Kvβ2 catalyzes the reduction of a broad range of carbonyls including aromatic carbonyls, electrophilic aldehydes and prostaglandins. However, the endogenous substrates for Kvβ have not been identified. To determine whether products of lipid oxidation are substrates of Kvβs, we tested the enzymatic activity of Kvβ2 with oxidized phospholipids generated during the oxidation of 1-palmitoyl-2-arachidonoyl-sn- glycero-3-phosphocholine (PAPC). Electrospray ionization mass spectrometric analysis showed that Kvβ2 catalyzed the NADPH-dependent reduction of several products of oxPAPC, including 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero- 3-phosphorylcholine (POVPC), 1-palmitoyl-2-(epoxycyclopentenone)-sn-glycero-3- phosphorylcholine (PECPC), 1-palmitoyl-2-(5,6)- epoxyisoprostane E2-sn-glycero-3-phosphocholine (PEIPC). These results were validated using high resolution mass spectrometric analysis. Time course analysis revealed that the reduced products reached significant levels for ions at m/z 594/596 (POVPC/PHVPC), 810/812 (PECPC/2H-PECPC) and 828/830 (PEIPC/2H-PEIPC) in the oxPAPC + Kvβ2 mixture (p < 0.01). These results suggest that Kvβ could serve as a sensor of lipid oxidation via its catalytic activity and thereby alter Kv currents under conditions of oxidative stress.

Full text of DOI:10.1016/j.cbi.2011.01.032

Chemico-Biological Interactions 191 (2011) 255–260
Contents lists available at ScienceDirect
Chemico-Biological Interactions
journal homepage: www.elsevier.com/locate/chembioint
Catalytic reduction of carbonyl groups in oxidized PAPC by Kv␤2 (AKR6)
Zhengzhi Xiea,1, Oleg A. Barskia,1, Jian Caib, Aruni Bhatnagara, Srinivas M. Tipparajua,∗
a Diabetes and Obesity Center, University of Louisville, Louisville, KY 40202, United States
b Department of Pharmacology and Toxicology, University of Louisville, Louisville, KY 40202, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 4 February 2011
The ␤-subunits of the voltage-gated potassium channel (Kv␤) belong to the aldo-keto reductase super-
family. The Kv␤-subunits dock with the pore-forming Kv ␣-subunits and impart or accelerate the rate
of inactivation in Kv channels. Inactivation of Kv currents by Kv␤ is differentially regulated by oxi-
dized and reduced pyridine nucleotides. In mammals, AKR6 family is comprised of 3 different genes
Kv␤1-3. We have shown previously that Kv␤2 catalyzes the reduction of a broad range of carbonyls
including aromatic carbonyls, electrophilic aldehydes and prostaglandins. However, the endogenous
substrates for Kv␤ have not been identified. To determine whether products of lipid oxidation are
substrates of Kv␤s, we tested the enzymatic activity of Kv␤2 with oxidized phospholipids generated
during the oxidation of 1-palmitoyl-2-arachidonoyl-sn-glycero-3-phosphocholine (PAPC). Electrospray
ionization mass spectrometric analysis showed that Kv␤2 catalyzed the NADPH-dependent reduction
of several products of oxPAPC, including 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-phosphorylcholine
(POVPC), 1-palmitoyl-2-(epoxycyclopentenone)-sn-glycero-3-phosphorylcholine (PECPC), 1-palmitoyl-
2-(5,6)- epoxyisoprostane E2-sn-glycero-3-phosphocholine (PEIPC). These results were validated using
high resolution mass spectrometric analysis. Time course analysis revealed that the reduced prod-
ucts reached significant levels for ions at m/z 594/596 (POVPC/PHVPC), 810/812 (PECPC/2H-PECPC) and
828/830 (PEIPC/2H-PEIPC) in the oxPAPC + Kv␤2 mixture (p < 0.01). These results suggest that Kv␤ could
serve as a sensor of lipid oxidation via its catalytic activity and thereby alter Kv currents under conditions
of oxidative stress.
Keywords:
Kv␤
Catalysis
Oxidized phospholipids
Potassium channel
Mass spectrometry
© 2011 Published by Elsevier Ireland Ltd.
1. Introduction
poorly understood. Recent work by Weng et al. shows that these
proteins display weak catalytic activity with model chemical sub-
The ␤-subunits of the voltage gated potassium channel belong
to the aldo-keto reductase (AKR) superfamily 6 (AKR6). The func-
tion of voltage-gated potassium channel (Kv) is essential for
several physiological processes, including muscle contraction, neu-
ronal excitation, and secretion. Despite extensive investigation,
the exact biological role of Kv␤-subunit remains unclear. Iden-
tification of Kv␤ substrates is important for understanding the
biological role of Kv channels, how they are regulated, and how
the activity of these channels could be therapeutically altered to
treat diseases such as hypertension, epilepsy, and arrhythmias.
Our previous studies have shown that Kv␤ subunits bind to pyri-
dine nucleotides with high affinity [1,2]. Sequence analysis of Kv␤s
shows that the proteins belong to the AKR superfamily [3], how-
ever, their catalytic properties and substrate specificities remain
strates such as 4-cyanobenzaldehyde and 4-carboxybenzaldehyde
[4]. In addition, investigations in our laboratory have shown that
Kv␤ can reduce both aldehydes and ketones, and that naturally
occurring compounds such as POVPC (1-palmitoyl-2-oxovaleroyl-
3-phosphatidylcholine), and PGJ2 have highest specific activity
with Kv␤2 [5].
Products of lipid peroxidation, such as POVPC, however,
are seldom generated in isolation. Lipid peroxidation generates
several structurally similar aldehydes and ketones, several of
which could also be potential Kv␤ substrates. Hence, to iden-
tify which products of lipid peroxidation are reduced by Kv␤, we
examined the activity of this protein with oxidized 1-palmitoyl-
2-arachidonoyl-3-phosphotidyl choline (oxPAPC). The oxPAPC is a
major component of minimally modified LDL (mmLDL) [6]. It has
been detected in the atherosclerotic lesions of animals and humans
[7,8]. Oxidation of PAPC generates several highly reactive carbonyl
compounds. Structurally, a typical oxidation product of PAPC con-
tains a glycerol backbone with a palmitoyl group esterified at the
sn1 position, an oxidized fatty acyl group at the sn2 position, and
a phosphatidyl choline head group at the sn3 position (Fig. 1A). In
a previous study, we have shown that aldose reductase (AKR1B1),
∗
Corresponding author at: Diabetes and Obesity Center, 580 South Preston Street
#421B, University of Louisville, Louisville, KY 40202, United States.
Tel.: +1 502 852 4221; fax: +1 502 852 3663.
E-mail address: smtipp01@louisville.edu (S.M. Tipparaju).
These authors made equal contribution.
1
0009-2797/$ – see front matter © 2011 Published by Elsevier Ireland Ltd.
doi:10.1016/j.cbi.2011.01.032

256
Z. Xie et al. / Chemico-Biological Interactions 191 (2011) 255–260
A
R
O
O
O
P
O
O
N+
O
H
HO
O
PAPC (782.6)
O2
Compound 1
Compound 2
O
Compound 3
O
O
O
O
O
OR
H
OR
OR
O
O
HO
POVPC (594.4)
PECPC (810.5)
PEIPC (828.5)
NADPH Kvβ2
NADPH Kvβ2
NADPH Kvβ2
O
OH
O
OH
O
O
OR
HO
OR
OR
O
HO
2H-PECPC (812.5)
2H-PEIPC (830.5)
PHVPC (596.4)
PAPC
782.6
Before air oxidation
100
%
B
0
POVPC
594.3
After air oxidation
100
%
PIEPC
828.6
846.7
650.3
PICPC
810.6
862.6
664.4
732.4
720
782.6
0
560
600
640
680
760
800
840
880
920
m/z
Fig. 1. Structural representation of air oxidation products of PAPC and its reduction by Kv␤2. (A) The chemical structure of the parent compound PAPC is depicted with the
non-substitutable group R represented in dotted box. Air oxidation of PAPC leads to the formation of reactive carbonyl groups at the sn2 position, and reduction of these
carbonyl groups by Kv␤2 in the presence of NADPH leads to the formation of corresponding alcohols. Compound 1 (POVPC), sn2 group: oxo-valeroyl group can be reduced
by Kv␤2 to form PHVPC (596.4 m/z). Compound 2 (PECPC), sn2 group: epoxy-isoprostaonic group, can be reduced to 2H-PECPC (812.5 m/z) by Kv␤2. Compound 3 (PEIPC),
sn2 group: hydroxyl-epoxyisoprostanoic, can be reduced by Kv␤2 to its corresponding 2H-PEIPC (830.5 m/z). (B) ESI-MS spectra of PAPC before air oxidation shows a well
resolved positive ion at m/z 782.6 as indicated, which was air oxidized for 72 h to generate carbonyl containing compounds: POVPC (m/z 594.3), PECPC (m/z 810.6), and PEIPC
(m/z 828.6).
and other members of AKR superfamily have high catalytic activity
for products of oxPAPC that contain aldehyde groups [9].
To determine which of the oxidation products of PAPC are
reduced by Kv␤, we preformed electrospray ionization mass spec-
trometric (ESI-MS) analysis of oxPAPC incubated with Kv␤2. Our
results show that Kv␤2 efficiently reduces both aldehydes and
ketones in oxidized phospholipids and therefore might play a reg-
ulatory role in sensing the carbonyl products that are generated
during lipid oxidation.
2. Materials and methods
2.1. Kvˇ2 expression and purification
The C-terminal domain of rat Kv␤2 (amino acids 39–367) with
a His tag at its N-terminus was expressed in the BL-21 strain of
Escherichia coli, as described earlier [1]. The His-tagged protein was
purified using Profinia Protein Purification System (Bio-Rad) with
a 1 ml Bio-Scale Mini Profinity IMAC cartridge (Bio-Rad) according

Z. Xie et al. / Chemico-Biological Interactions 191 (2011) 255–260
257
to manufacturer’s instructions. The purified protein was suspended
in 0.2 M potassium phosphate buffer (pH 7.4) immediately after
the purification using a 10 ml Bio-Gel P-6 Desalting cartridge (Bio-
Rad). The concentration of the purified protein was measured using
Bradford’s assay [10].
calculated based on the intensities of [M] and [M + 2] peaks and
reported as a percent conversion. One-way ANOVA analyses were
performed using SigmaStat 3, and time course figures were pre-
pared using Sigmaplot 10. Data was collected from 3 or more
experiments for each time point.
2.2. Preparation of air oxidized PAPC (oxPAPC)
3. Results
Air oxidation of synthetic 1-palmitoyl-2-arachidonoyl-3-
phosphotidyl choline (PAPC) was preformed as described before
[9]. Briefly, 100 ␮g of PAPC in chloroform (Avanti Polar Lipids)
was dried under N2 on the bottom of a 4 ml glass vial (National
Scientific). The dry lipid was exposed to air for 72 h in dark at room
temperature. The extent of the oxidation was determined using
ESI-MS. The oxidation product was stored in −80 ◦C until use.
3.1. Air oxidation of PAPC and generation of lipid-carbonyls
As shown in Fig. 1, air oxidation of PAPC for 72 h resulted in the
formation of oxPAPC containing several compounds with higher
and lower m/z values than the parent compound (m/z 782). As pre-
viously reported, the major ion at m/z 594 was ascribed to POVPC
(1-palmitoyl-2-oxo-valaroyl-3-phosphotidyl choline) compound 1
[9], whereas the ions at m/z 810 and 828 were epoxy isoprostane
2.3. Reduction of oxPAPC using Kvˇ2 and NADPH
compounds
2 and 3: 1-palmitoyl-2-(5,6)- epoxyisoprostane
E2-sn-glycero-3-phosphocholine (PEIPC) and 1-palmitoyl-2-
(epoxycyclopentenone)-sn-glycero-3-phosphorylcholine (PECPC)
respectively (Fig. 1A). All the three molecules contain reactive
carbonyl groups and POVPC has been previously reported to be
reduced by aldose reductase [9].
Reduction of carbonyl groups of lipid oxidation products in
oxPAPC to their corresponding alcohols by Kv␤2 was followed by
using ESI-MS analysis as described earlier [5] with some modi-
fications. The 250 ␮l reaction mixtures containing purified Kv␤2
protein (20 ␮M), NADPH (150 ␮M), and oxPAPC (88 ␮g/ml) in
0.2 M potassium phosphate buffer (pH7.4), were incubated in a
Teflon tube (Upchurch Scientific) at 37 ◦C. Additional NADPH was
added to the reaction mixture after 3 h, 6 h, and 12 h of incuba-
tion, and additional Kv␤2 was added into the reaction mixture
after 6 h of incubation. The reaction was stopped by freezing the
reaction mixture at −80 ◦C at 3 h, 6 h, 12 h, or 22 h, respectively.
The samples were stored at −80 ◦C until extraction and ESI-MS
analysis.
3.2. Reduction of carbonyl groups in oxPAPC mixture
To identify which of the carbonyl groups generated in oxPAPC
are reduced by Kv␤, and to determine the substrate prefer-
ence of Kv␤2, we incubated oxPAPC with recombinant Kv␤2.
Because enzymatic reduction leads to the formation of alco-
hols from aldehydes or ketones, we expected that the reduction
products generated in the reaction mixture would show an
increase in their m/z value by 2 atomic mass units over the
un-reduced substrates. Hence, we monitored the ratio between
intensities of [M] and [M + 2] peaks to identify reduction
products.
Incubation with Kv␤2 resulted in the appearance of several
new phospholipid species (Fig. 2A). For example, we found an
increase in the ratio of 596.3 (1-palmitoyl-2-hydroxyvaleroyl phos-
phatidylcholine, PHVPC) to 594.3 (POVPC), Fig. 2A and B. This is
consistent with our previous report that NADPH was consumed
in presence of Kv␤2 and purified reagent POVPC [5]. Similarly,
two dalton increments were observed for ions with m/z 812
and 830 which indicated the formation of 2H-PEIPC and 2H-
PECPC derived from reduction of PEIPC and PECPC, respectively
(Fig. 2C and D). These data clearly demonstrate that Kv␤2 con-
verts aldehydes and ketones in the oxPAPC to their corresponding
alcohols, and that Kv␤2 has a broad substrate activity required
to reduce several oxidized phospholipids containing carbonyl
groups.
2.4. Analysis of the reduced product using ESI-MS
Identification and quantification of carbonyl substrates and
alcohol products were performed as described earlier with
minor modification [5,9]. The reduced product was extracted
using the Bligh and Dryer procedure [11]. The chloroform
extract was dried under N2, and dissolved in 10 mM NH4OAc
in methanol–chloroform (2:1, v/v). The solution was injected
into a MicroMass ZMD 2000 mass spectrophotometer (Waters-
Micromass, Milford, MA) using a Harvard syringe pump (Harvard
Apparatus) at a flow rate of 10 ␮l per min. Positive ion spectra
were acquired over the mass range of 20–1000 atomic mass with
following operating parameters: capillary voltage 4.34 kV, cone
voltage 35 V, extractor voltage 9 V, RF lens voltage 0.9 V, source
block 100 ◦C, desolvation temperatures 200 ◦C.
2.5. High resolution mass spectrometry analysis
After extraction, the chloroform extract was dried under N2, and
dissolved in 7.5 mM NH4OAc in methanol. High resolution mass
spectrometry analysis was performed on a LTQ Orbitrap XL Mass
Spectrometer (Thermo) equipped with a TriVersa NanoMate ion
source (Advion BioSciences) and an electrospray chip (nozzle inner
diameter 5.5 ␮m). TriVersa NanoMate was operated in positive
mode at gas pressure 0.05 psi and ionization voltage of 1.2 kV. MS
scans were acquired in positive ion mode with a mass range from
100 to 1500 and resolution at 100,000.
3.3. High resolution mass spectrometry analysis
The relative peak intensity of [M + 2] was used to monitor
the progress of reduction. However, the ESI-MS analysis cannot
distinguish the [M + 2] peak which arises from reduction (e.g.
M + 2H) or the isotopic peak (e.g. M with 2 13C) which naturally
occurs even if the reduction does not occur. Fortunately, oxPAPC
compounds only contain C, H, O, and N, and their [M + 2] iso-
topic peaks are relatively small. So the lack of mass resolution
of ESI-MS analysis is not a serious problem if a large amount of
the reduction products is present. However, at the beginning of
the reaction, when the amount of reduction products is small,
the contribution of isotopic peak to [M + 2] cannot be ignored.
Therefore, we used high resolution MS analysis to overcome this
problem.
2.6. Data analysis
High-accuracy m/z values were obtained from LTQ-Orbitrap
spectra averaged over 5 min using Qual Browser Xcalibur 2.0. Peak
intensity values were acquired from ESI-MS spectra averaged over
3.5 min using Masslynx 3.2. Formation of alcohol products were

258
Z. Xie et al. / Chemico-Biological Interactions 191 (2011) 255–260
(A)
594.3
100
80
60
40
20
Control, 22h
664.4
648.3
828.6
610.3
810.5
783.2
632.3
0
100
830.7
(B)
80
60
40
20
0
596.4
Kvβ2, 22h
634.3
648.3
812.5
610.2
783.2
600
650
700
750
800
m/z
(C)
(i)
(iii)
PEIPC
POVPC
PECPC
812.6
Control
(ii)
810.5
828.6
830.6
594.3
100
100
%
100
%
Control
Control
%
596.4
596.4
0
0
0
(D)
(iii)
828.6
830.7
812.5
Kvβ2
(i) 594.3
(ii)
100
100
100
Kvβ2
Kvβ2
810.6
%
0
%
%
m/z
828 829 830 831 832
0
0
593 594 595 596 597
810 811 812 813 814
Fig. 2. Kv␤2 catalyzes the reduction of aldehydes generated from air oxidation of PAPC. Aliquots of oxidized PAPC were incubated without (A) or with Kv␤2 (B) in potassium
phosphate buffer (see Section 2). After the incubation, the phospholipids were extracted in chloroform/methanol/water and injected in to the electrospray. The oxidized
compounds POVPC, PEIPC and PECPC remain unaltered in the absence of Kv␤2. Incubation with Kv␤2 led to an increase in the intensity at m/z 596.3, 812.5, 830.5, which
represents the formation of the reduction products of POVPC, PECPC and PEIPC. Panels C and D, represent narrow-range spectra of the three products without (control; i–iii)
or with Kv␤2 (i–iii) respectively.
For high resolution mass analysis, oxPAPC was incubated with
Kv␤2 for 0, 3, and 6 h. After extraction, the reaction products were
applied on LTQ-Orbitrap MS for high resolution mass spectrometry
analysis. As shown in Fig. 3, when enzyme was absent, there was no
PHVPC peak (theoretical m/z: 596.39220). Instead, a peak with m/z
at 596.38256 was found, which was identified as POVPC 13C isotopic
peak (theoretical m/z: 596.38325). This peak did not increase with
Kv␤2 incubation at 3 and 6 h. In contrast to this, the PHVPC peak
(observed m/z 596.39139) was seen to be progressively increased
upon incubation with Kv␤2 at 3 and 6 h. This result clearly validates
the ESI-MS (Fig. 2) data, and demonstrates the reduction of the
aldehydes (POVPC) to an alcohol (PHVPC).
594.37558
(A)
100
%
30
(i)
Control
595.37893
Control
%
13C2-POVPC
596.38256
596.38256
0
100
0
30
594.37568
594.37568
(B)
(ii)
Kvβ2 3h
595.37903
Kvβ2 3h
596.39139
%
%
PHVPC
596.39139
596.38270
0
100
0
30
(C)
596.39139
(iii)
Kvβ2 6h
595.37910
Kvβ2 6h
PHVPC
596.39139
%
0
%
596.38307
596.38
0
596.40
594.5
595.0
595.5
m/z
596.0
596.5
597.0
m/z
Fig. 3. Identification of products by high resolution mass spectrometry. Aliquots of reaction mixture containing oxPAPC without or with Kv␤2 were analyzed by Orbitrap
high resolution mass spectrometer at 3 and 6 h time points. (A) In control group without Kv␤2, no PHVPC product at m/z 596.38 was identified. (B) After 3 or (C) 6 h of
incubation with Kv␤2 a progressive increase in the intensity at m/z 596.39139 was noted. Corresponding insets i, ii and iii (narrow-range spectra) representing the formation
of PHVPC (m/z 596.39) at 0, 3 and 6 h time points respectively.

Z. Xie et al. / Chemico-Biological Interactions 191 (2011) 255–260
259
(A)
(B)
(C)
*
*
POVPC
PECPC
PEIPC
60
40
20
0
*
65
55
45
35
*
55
Control
Control
Kvβ2
*
Kvβ2
45
Control
Kvβ2
35
0
0
5
10
15
20
25
5
10
15
20
25
0
5
10
15
20
25
Time (h)
Time (h)
Time (h)
Fig. 4. Time course of carbonyl reduction by Kv␤2 Reaction mixtures containing Kv␤2 and oxPAPC were incubated and monitored for the generation of reduced products at
different time points without or with Kv␤2. The reaction was stopped at 0, 3, 6, 12, and 22 h by freezing at −80 ◦C. Phospholipids were extracted by chloroform/methanol/water
mixture and analyzed by electrospray. The formation of reduced products is plotted as percentage conversion. Panel (A) shows the time course of PHVPC (m/z 596) product
formation compared with control. Panel (B) and (C) shows the 2H-PECPC (m/z 812) and 2H-PEIPC (m/z 830) product respectively. n = 3 separate reactions, *p < 0.01.
3.4. Time course effects on product formation
whereas modest reduction was monitored with 12-oxo-ETE and
steroid containing compounds.
To quantify the product formation over the time of incubation,
we monitored the major ion pairs at m/z 594/596, 810/812 and
828/830 at different time points of reaction. As shown in Fig. 4A, a
progressive increase in the conversion of POVPC (594 m/z) to PHVPC
(596 m/z) was observed. Indeed, the conversion could be detected
as early as 3 h. At 6 h, the differences between the reaction mixture
and no enzyme control were highly significant (p < 0.01). Similar
results were observed for the ion at m/z 828, which represents the
reduction of PIEPC to 2H-PIEPC (m/z 830). The reaction mixture con-
taining enzyme was significantly different at the 6 h and 22 h time
point compared with control (no enzyme). The relative increase
in the ion intensity at m/z 821, which indicates the conversion of
PECPC to its corresponding alcohol 2H-PECPC, was also found to
be higher in the sample incubated with the enzyme than in the
absence of the enzyme. However, the reaction had a delayed start
because the overall conversion did not alter even at 6 h time point.
For this reaction, the conversion was highest at the 22 h time point
and was statistically significant compared with the no-enzyme con-
trol group. These data demonstrate that the carbonyl groups in the
oxPAPC mixture are catalytically reduced to their corresponding
alcohols by Kv␤2, and that POVPC and PEIPC are preferred over
PECPC.
In comparison with other AKRs such as aldose reductase
(AKR1B), Kv␤2 is a much slower enzyme. In our previous publi-
cations we reported catalytic activity for Kv␤2 with kcat for POVPC
and PGJ2 of 0.078 and 0.088 min−1, respectively [5], whereas kcat
of aldose reductase with POVPC is 36.2 min−1 [9]. Based on these
measurements using POVPC as substrate, aldose reductase is esti-
mated to be 464 times more active than Kv␤2. Therefore, it is likely
that Kv␤2 is not a detoxification enzyme like some other AKRs.
However, the unique property of Kv␤ subunits is that it is the only
known AKR that binds to ion channels (Kv channels) and modulates
channel gating and trafficking [15]. Therefore, the catalytic activity
of Kv␤2 with the products of lipid oxidation may serve as a sensor of
oxidative changes in the cell membrane, which will allow the reg-
ulation of potassium currents in response to the metabolic/redox
status of the cell.
Our findings are in agreement with previous reports, which
show that oxidation of PAPC results in generation of several classes
of carbonyl compounds [6]. Among them, one of the most reactive
phospholipid aldehyde generated is POVPC and it has been sug-
gested that the generation of POVPC within the vessel wall could
contribute to atherosclerotic lesion formation and progression
[7,16,17]. In general, oxidized phospholipids have been implicated
in the pathogenesis of several chronic inflammatory diseases such
as arthritis, inflammatory bowel disease, and multiple sclerosis.
Nevertheless, the ability to modulate ion channels or their sub-
sidiary sub-units has not been examined.
The present study shows that Kv␤2 reduces several carbonyl
products generated from the oxidation of membrane lipids. We
observed significant catalytic activity of Kv␤ with POVPC, PEIPC,
as well as PECPC. These findings attest to the possibility that Kv␤2
subunit can sense metabolic changes in the cell and play a regula-
tory role in ion channel modulation. Although we used Kv␤2 for our
studies, because of high sequence homology among different Kv␤
subunits, it is likely that other Kv␤ subunits, i.e., Kv␤1 and Kv␤3,
can also catalyze the reduction of oxidized phospholipids, however,
future experiments are required to determine the catalytic prop-
erties of other ␤-subunits and how they are different from Kv␤2
[2,12]. Further work is also required to understand how individual
species generated within oxidized lipids regulate electrical activ-
ity by regulating Kv␤ catalysis and thereby the inactivation of Kv
currents.
4. Discussion
This study demonstrates that Kv␤2 catalyzes the reduction
of several structurally similar compounds in oxPAPC mixture,
which suggests that Kv␤2 subunits can play a regulatory role
in sensing the carbonyl products that are generated during lipid
oxidation. Sequence homology studies indicate that Kv␤ sub-
units belong to the AKR superfamily of proteins. The ␤-subunits
of shaker potassium channel have ␤/␣8 barrel structure with
tight NADPH binding and have 20–25% homology to other AKRs
[12]. Kv␤ 1 and 3 consist of the long N-terminus which acts
as a ball and inactivates the channel by plugging the pore of
the Kv channel, whereas in Kv␤2 the N-terminus is absent and
therefore it can only accelerate the self inactivating Kv channels
[13].
The Kv␤ subunits bind to the cytosolic domain of Kv channels
and modulate their gating properties. We have shown previously
that pyridine nucleotides can differentially regulate the Kv channel
inactivation in which oxidized pyridine nucleotides NAD(P)+ abol-
ish inactivation and reduced pyridine nucleotides NAD(P)H restore
inactivation [2,14]. We have also examined the catalytic activity
of Kv␤2 and have identified that the protein displays higher activ-
ity with naturally occurring compounds, such as PGJ2 and POVPC,
In summary, our studies show that several oxidized phospho-
lipid generated by the oxidation of PAPC are reduced by Kv␤2.
These results were validated using high resolution mass spectrom-
etry. We report that major carbonyl compounds with m/z 594, 810
and 828 are reduced by Kv␤2. High resolution mass spectrometry

260
Z. Xie et al. / Chemico-Biological Interactions 191 (2011) 255–260
evidence was used to validate the reduction of POVPC at differ-
ent time points, and the time-course of the reduction for the three
major species was monitored to demonstrate the reduction process
in oxPAPC mixture. Taken together, these data demonstrate that
Kv␤2 catalyzes the reduction of several oxidized phospholipids.
Based on these findings, we speculate that oxidized phospholipid
generated during oxidative stress might be able to decrease the
inactivation of Kv currents by oxidizing NADPH bound to Kv␤-
subunit and that changes in Kv conductance via Kv␤ may be one
mechanism by which cells respond to oxidative stress.
[5] S.M. Tipparaju, O.A. Barski, S. Srivastava, A. Bhatnagar, Catalytic mechanism
and substrate specificity of the beta-subunit of the voltage-gated potassium
channel, Biochemistry 47 (2008) 8840–8854.
[6] G. Subbanagounder, A.D. Watson, J.A. Berliner, Bioactive products of phos-
pholipid oxidation: isolation, identification, measurement and activities, Free
Radic. Biol. Med. 28 (2000) 1751–1761.
[7] G. Subbanagounder, N. Leitinger, D.C. Schwenke, J.W. Wong, H. Lee, C.
Rizza, A.D. Watson, K.F. Faull, A.M. Fogelman, J.A. Berliner, Determi-
nants of bioactivity of oxidized phospholipids. Specific oxidized fatty acyl
groups at the sn-2 position, Arterioscler. Thromb. Vasc. Biol. 20 (2000)
2248–2254.
[8] A.D. Watson, N. Leitinger, M. Navab, K.F. Faull, S. Horkko, J.L. Witztum,
W. Palinski, D. Schwenke, R.G. Salomon, W. Sha, G. Subbanagounder,
A.M. Fogelman, J.A. Berliner, Structural identification by mass spec-
trometry of oxidized phospholipids in minimally oxidized low
density lipoprotein that induce monocyte/endothelial interactions
and evidence for their presence in vivo, J. Biol. Chem. 272 (1997)
13597–13607.
Conflict of interest statement
None.
[9] S. Srivastava, M. Spite, J.O. Trent, M.B. West, Y. Ahmed, A. Bhatnagar, Aldose
reductase-catalyzed reduction of aldehyde phospholipids, J. Biol. Chem. 279
(2004) 53395–53406.
Acknowledgements
[10] M.M. Bradford, A rapid and sensitive method for the quantitation of micro-
gram quantities of protein utilizing the principle of protein–dye binding, Anal.
Biochem. 72 (1976) 248–254.
[11] E.G. Bligh, W.J. Dyer, A rapid method of total lipid extraction and purification,
Can. J. Biochem. Physiol. 37 (1959) 911–917.
[12] O.A. Barski, S.M. Tipparaju, A. Bhatnagar, The aldo-keto reductase superfamily
and its role in drug metabolism and detoxification, Drug Metab. Rev. 40 (2008)
553–624.
The authors thankfully acknowledge the research support pro-
vided by NIH (HL-54477, HL-59378, ES-11860; to AB), NCRR
(RR024489), University of Louisville Clinical and Translational Sci-
ence award 20017 and HL-089372 (to OAB), and American Heart
Association Beginning Grant-in-Aid 0865466D (to SMT).
[13] R. Peri, B.A. Wible, A.M. Brown, Mutations in the Kv beta
2 binding
site for NADPH and their effects on Kv1.4, J. Biol. Chem. 276 (2001)
738–741.
References
[14] S.M. Tipparaju, N. Saxena, S.Q. Liu, R. Kumar, A. Bhatnagar, Differen-
tial regulation of voltage-gated K+ channels by oxidized and reduced
pyridine nucleotide coenzymes, Am. J. Physiol. Cell Physiol. 288 (2005)
C366–C376.
[15] O. Pongs, J.R. Schwarz, Ancillary subunits associated with voltage-dependent
K+ channels, Physiol. Rev. 90 (2010) 755–796.
[16] B.G. Hill, P. Haberzettl, Y. Ahmed, S. Srivastava, A. Bhatnagar, Unsaturated
lipid peroxidation-derived aldehydes activate autophagy in vascular smooth-
muscle cells, Biochem. J. 410 (2008) 525–534.
[17] S. Srivastava, S.Q. Liu, D.J. Conklin, A. Zacarias, S.K. Srivastava, A. Bhatnagar,
Involvement of aldose reductase in the metabolism of atherogenic aldehydes,
Chem. Biol. Interact. 130–132 (2001) 563–571.
[1] S.Q. Liu, H. Jin, A. Zacarias, S. Srivastava, A. Bhatnagar, Binding of pyridine
nucleotide coenzymes to the beta-subunit of the voltage-sensitive K+ channel,
J. Biol. Chem. 276 (2001) 11812–11820.
[2] S.M. Tipparaju, S.Q. Liu, O.A. Barski, A. Bhatnagar, NADPH binding to beta-
subunit regulates inactivation of voltage-gated K(+) channels, Biochem.
Biophys. Res. Commun. 359 (2007) 269–276.
[3] T. McCormack, K. McCormack, Shaker K+ channel beta subunits belong to an
NAD(P)H-dependent oxidoreductase superfamily, Cell 79 (1994) 1133–1135.
[4] J. Weng, Y. Cao, N. Moss, M. Zhou, Modulation of voltage-dependent Shaker
family potassium channels by an aldo-keto reductase, J. Biol. Chem. 281 (2006)
15194–15200.