© 1995 American Heart Association, Inc.
Articles |
Electrophysiological Effects of 4-Hydroxynonenal, an Aldehydic Product of Lipid Peroxidation, on Isolated Rat Ventricular Myocytes
From the Department of Physiology and Biophysics and the Department of Human Biological Chemistry and Genetics, University of Texas Medical Branch, Galveston.
Correspondence to Dr Aruni Bhatnagar, Department of Physiology and Biophysics and Department of Human Biological Chemistry and Genetics, University of Texas Medical Branch, Galveston, TX 77555.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Abstract Aldehydic products of lipid peroxidation, such as 4-hydroxynonenal (4-HNE), have been implicated in the etiology of pathological changes under oxidative stress. To identify the mechanism by which 4-HNE alters cellular excitability, its effects on isolated rat ventricular myocytes were studied. Superfusion with 100 to 880 µmol/L 4-HNE led to a time- and concentration-dependent rigor shortening of myocytes. A reduction in [Ca2+]o and inhibition of transsarcolemmal Ca2+ transport by 1 mmol/L La3+ did not affect either the magnitude or the time course of 4-HNE–induced myocyte rigor. Superfusion of myocytes with 400 µmol/L 4-HNE led to an increase in the action potential duration, progressive depolarization of the resting membrane potential, and an increase in the input resistance (Rin) of the myocyte (phase I), followed by a loss of electrical excitability. Continued superfusion with 4-HNE resulted in membrane hyperpolarization and a prominent decrease in the Rin (phase II). The decrease in Rin coincided with myocyte rigor. In whole-cell voltage-clamp experiments, superfusion with 4-HNE inhibited current through the inward rectifier K+ channel (IK1). 4-HNE had no effect on either the magnitude or the rate of "rundown" of L-type Ca2+ currents. Exposure to 4-HNE led to an increase in the magnitude of the fast inward Na+ current (INa). The voltage dependence of the steady state parameters for activation and inactivation of INa shifted to more positive potentials, with a resultant increase in the window current. 4-HNE–induced myocyte rigor was accompanied by a large increase in time-independent currents that displayed linear dependence on the membrane potential and were inhibited by glibenclamide, suggesting activation of the ATP-sensitive K+ channel. Steady state currents recorded in Cs+-containing Ringer's solution with La3+ and tetrodotoxin and Cs+-containing internal solution (leak currents) were not affected by 4-HNE. Superfusion with 4-HNE resulted in a significant decrease in the cellular concentration of nonprotein thiols and a severe decrease in [ATP]i. The energy charge of the myocytes fell from 0.9 to 0.3. These observations indicate that 4-HNE–induced membrane depolarization may be due to an inhibition of IK1. Changes in voltage dependence of INa, inhibition of IK1, and membrane depolarization appear to contribute to the prolongation of the action potential, observed during phase I. Depletion of [ATP]i may be responsible for changes observed during phase II, ie, activation of the ATP-sensitive K+ channels, membrane hyperpolarization, decrease in Rin, and rigor shortening of the myocytes. These results suggest that stable products of lipid peroxidation, such as 4-HNE, are proarrhythmic and may contribute to the cytotoxic effects of oxidative stress.
Key Words: 4-hydroxynonenal • myocytes • oxidative stress
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Alterations in cellular oxidative metabolism under pathological conditions, such as ischemia/reperfusion or drug toxicity, have been suggested to be responsible for the observed tissue injury.1 2 3 Part of such oxidative stress has been found to be due to rapid autocatalytic oxidation of polyunsaturated fatty acids and accumulation of the products of lipid peroxidation, initiated by oxygen-derived free radicals.3 4 5 6 ß-Cleavage of alkoxy radicals, formed during lipid peroxidation, leads to the formation of stable aldehyde products.7 Unlike short-lived free radical species, aldehydes are relatively stable and can in principle, because of their high reactivity, alter cellular structure and function. Therefore, accumulation of aldehydes could account, at least in part, for the toxicity of oxygen-derived free radicals and lipid peroxidative reactions.
Of the number of aldehydes generated during lipid peroxidation, malondialdehyde and 4-hydroxynonenal (4-HNE) are currently a focus of much interest because they accumulate in high concentrations and participate in a number of potentially cytotoxic reactions.8 9 10 Previous investigations suggest that 4-HNE, formed on oxidation of omega-6 polyunsaturated fatty acids such as 18:2 and 20:4,10 may be one of the most toxic aldehydes produced during lipid peroxidation.11 12 13 Exposure to 4-HNE has been shown to lead to physiological and structural changes in a number of cell types.10 These changes include depletion of cellular glutathione,13 14 15 alterations in nucleic acid synthesis,16 17 disturbances in Ca2+ homeostasis,15 respiration,18 19 and eventual cell lysis and death.10
Administration of 4-HNE to perfused hearts leads to coronary vasodilation20 and progressive decrease in the peak systolic pressure.21 Sustained insult results in complete failure of contraction.22 However, electrophysiological changes underlying the cardiotoxic effects of 4-HNE have not been investigated. In the present study, the effects of 4-HNE on isolated cardiac myocytes are reported. Electrophysiological investigations show that inhibition of the inward rectifier K+ current (IK1) may lead to "early" membrane depolarization and loss of excitability. "Late" effects of the aldehyde include a pronounced depletion of cellular ATP, activation of the ATP-sensitive K+ channel, and rigor shortening of myocytes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
4-HNE diethyl acetate (4-HNE-DEA) was a gift from Dr H. Esterbauer, University of Graz, Austria. ATP, ADP, and AMP were purchased from Sigma Chemical Co. Glibenclamide was purchased from Research Chemicals Inc. All other chemicals were of the highest purity available.
Isolation of Adult Rat Ventricular Cells
Rat ventricular myocytes were isolated from male
Sprague-Dawley rats weighing 200 to 260 g. Rats were injected
intraperitoneally with heparin (300 U) 30 minutes or more before
pentobarbital anesthesia (20 mg). The hearts were rapidly removed and
perfused in a retrograde fashion through the aorta with Tyrode's
solution containing 1.0 mmol/L CaCl2 at 37°C until all
signs of blood were removed with gentle squeezing of the heart. The
hearts were then perfused with a nominally Ca2+-free
Tyrode's solution (free [Ca2+], 3 µmol/L) for 5
minutes, followed by perfusion with collagenase solution containing 40
to 45 mg of collagenase (Boehringer, lot BJA119) in 50 mL of
low-Ca2+ Tyrode's solution (free [Ca2+], 10
µmol/L). All solutions were oxygenated. After 5 minutes of enzyme
perfusion, the hearts were removed from the cannula, the atria and
right ventricles were discarded, and the left ventricular wall and
septum were cut vertically into four to six pieces and allowed to
incubate in the enzyme solution for an additional 5 to 7 minutes.
After variable time periods (2.5 to 15 minutes) of additional digestion at 37°C, the pieces were dipped in a KB-like solution23 containing (mmol/L) KCl 85, K2HPO4 30, MgSO4 5, K2-EGTA 1, Tris2-ATP 2, pyruvate 5, creatine 5, taurine 20, and glucose 20, (pH 7.2) or an ATP-free variant of it (mmol/L: KCH3SO3 106, KCl 3.9, MgSO4 2.4, glucose 22, taurine 22, creatine 6, pyruvate 5, and potassium phosphate 8, [pH 7.3, with osmolarity adjusted to 300 mosm/kg]), gently swirled in 3 to 3.5 mL of the same solution until the other pieces had been similarly treated, then triturated, filtered through a 200-µm nylon mesh, and centrifuged at 22g for 3 to 5 minutes. The supernatant was drained, and cells were resuspended in 2 mL of the same solution and centrifuged again. Finally, the cells were resuspended in 4 mL of the same solution for storage in a refrigerator until use. At this time, 10 µL of cells were transferred to 1.0 mL of 1 mmol/L CaCl2–containing Tyrode's solution for counting purposes. After 15 to 30 minutes at room temperature in this Tyrode's solution, isolated cell preparations generally exhibited 50% to 60% rod-shaped quiescent cells. Experiments were performed on the preparations yielding the highest proportion of rod-shaped cells. Cells obtained by this method retained rod-shaped morphology on exposure to 2 mmol/L Ca2+ and remained viable for 20 to 25 hours at 4°C as tested by trypan blue exclusion.
Determination of Protein and Nonprotein Thiols
A 0.5-mL aliquot of superfused cells was precipitated with 0.5
mL of 10% perchloric acid. The mixture was centrifuged at 5000 rpm for
5 minutes to settle the precipitate. The pellet was dissolved in 1.2 mL
Tris-EDTA-SDS (0.5 mol/L Tris, 5.0 mmol/L EDTA, and 1% sodium dodecyl
sulfate), and a 0.5-mL aliquot of this mixture was added to 0.5 mL of
either 50 mmol/L N-ethylmaleimide (blank) or 500 µmol/L
5,5'-dithiobis-(2-nitrobenzoic acid) (DTNB) (thiols). A 25-µL aliquot
was used to determine protein by the method of Lowry et
al.24 The supernatant (0.9 mL) of the perchloric acid
(PCA) extract was neutralized with 0.2 mL of 2 mol/L KOH+2.5 mol/L
KHCO3. To 0.8 mL of the neutralized extract, 0.1 mL of 0.1
mol/L Tris-HCl (pH 8.0) and 0.1 mL of 500 µmol/L DTNB (in 1% sodium
citrate) were added. Thiol concentrations were determined by the
absorbance of the mixture at 412 nm; an extinction coefficient of
13 600 (mol/L)-1 · cm-1 was
used.25
Nucleotide Assay
Superfused myocytes were collected and centrifuged at
22g for 3 minutes. The cells were then resuspended in normal
Ringer's solution (n-Ringer), and an equal volume of cold 10% PCA was
added to the mixture (final volume, 500 µL). The protein precipitate
was removed by centrifugation, and the supernatant was neutralized by
the addition of 50 µL of 2 mol/L KOH+2.5 mol/L KHCO3. The
precipitated salts were removed by centrifugation, and the neutralized
extract was filtered through a 0.22-µm filter.
Nucleotides in the extract were assayed by using reverse-phase high-pressure liquid chromatography. Solvent A consisted of 0.1 mol/L potassium phosphate and 8 mmol/L tetrabutylammonium hydrogen sulfate; solvent B consisted of solvent A with the addition of 30% methanol. The column (C-18, Whatman; ODS, 5 µm; 4.6 mmx25 cm) was equilibrated with solvent A, and the gradient was started at the time of injection (t=0) to 100% solvent B programmed linearly, in 12 minutes, by using Beckman 110A pumps. This was followed by 6 minutes of an isocratic period, after which the concentration of solvent A was increased in 4 minutes from 0% to 100% by using a linear gradient. Pump control, peak integration, and data analysis were performed by SYSTEM GOLD software (Beckman). The absorbance of the eluate was monitored at 254 nm by a Beckman 406 detector. Peak identity was confirmed by the retention times. Standard curves were constructed with solutions of known concentrations of ATP, ADP, and AMP. Concentrations of the nucleotides were determined by the peak area. Protein was determined by Bradford's dye-binding method.26
Electrophysiological Measurements
The whole-cell mode of the patch-clamp technique27
was used to measure membrane currents and voltage. Cells were layered
on a coverslip and superfused with extracellular solution at a rate of
0.5 to 1 mL/min. All experiments were conducted at room temperature.
Patch pipettes were made of borosilicate glass (Qwikfil, WPI) by a
three-stage pull on a Flaming-Brown horizontal puller (P80/PC, Sutter
Instruments Co) and fire-polished. For recording membrane voltage and
action potentials, patch pipettes with resistances of 7 to 8 M
were
used. For recording K+ and Ca2+ currents
(IK and ICa, respectively), electrodes of 5
M were used; for measuring Na+ current
(INa), electrodes of 1 to 1.5 M were used. Ag/AgCl
electrodes, connected via an agar bridge, were used to establish
contact between the pipette and the bathing solution.
The electrode holder and the pipette were mounted on the head stage of
a List EPC-7 patch-clamp amplifier (List-Electronic). Voltage and
current commands were programmed with PCLAMP software.
Gigaohm seals were developed between the cell membrane and the patch
pipette. The patch of the membrane in the orifice of the pipette was
ruptured by suction. After the membrane was ruptured, the resting
membrane potential (RMP) was allowed to stabilize for 2 to 3 minutes.
Stable RMPs (±1 mV) could be recorded for >1 hour. The voltage or
current tracings were digitized by an Axolab 1100 computer interface
(Axon Instruments) at a sampling rate of 20 or 40 KHz. To record
INa, part of the Na+ in the Ringer's solution
was substituted with Cs+ (Cs-Ringer). Large
INas, recorded in 135 mmol/L
[Na+]o in mammalian cells, prevent adequate
voltage clamp by use of single suction pipettes.28 The
total series resistance was compensated before recording by using the
series-resistance controls of the EPC-7 amplifier. The series
resistance before compensation was estimated from the cell capacitance
and the time constant (
) for decay of the capacitative transient.
The initial capacitative transient for small voltage displacements (10
mV) was well approximated by a single exponential
(R2=.996 to .998) with a value of 116±12
µs. The falling phase of the capacitative transient and the tail
currents were smooth and without notches, indicating that the cell was
adequately voltage-clamped. The current-voltage (I-V) relations of
INa did not show an abrupt rise near the threshold.
Nevertheless, even in experiments in which the negative limb of the I-V
relation appeared graded, the establishment of voltage control may be
slow. Therefore, as suggested by Hanck and Sheets,29 only
experiments in which the slope factor of the inactivation curve of
INa (k in Equation 4
) was >6 mV were included in the
analysis.
Changes in cell shape were monitored with a Cohu CCD camera (model 4815-3100) connected to a video monitor (VM 4512, Sanyo) and a VHS video recorder (DX-900, Toshiba). Cells were monitored continuously, and the entire experiment was videotaped. Alterations in cell length were analyzed with IMAGE 1 software (Universal Imaging Corp). For determinations of time to rigor (TR), changes in the resting cell length of 60 to 80 cells per preparation were monitored. Data from three to five hearts were pooled, and the average values of TR are reported.
Solutions
The composition of n-Ringer solution was (mmol/L) NaCl 150, KCl
5.4, MgCl2 1, CaCl2 2, and HEPES 10. The
composition of La3+-containing Ringer's solution at pCa
6.0 was as follows (mmol/L): NaCl 150, KCl 5.4, MgCl2 1.0,
LaCl3 1.0, and HEPES 10. Although no Ca2+ was
added to this solution, the free [Ca2+] in this solution
was 1 µmol/L. In the text, this solution is referred to as pCa
6.0–La-Ringer. The pH of both these solutions was adjusted to 7.4 with
0.1 mol/L NaOH. The free [Ca2+] was measured with
Ca2+ electrodes (model 93-20, Orion Research Inc). The
internal solution used to record membrane potential and IK
and ICa contained (mmol/L) potassium aspartate 120, KCl 30,
MgCl2 1.0, Na2-ATP 1.0, HEPES 10, and EGTA
0.005. For recording INa, the composition of the internal
solution of the patch electrode was (mmol/L) CsCl 134, NaCl 10,
MgCl2 1, Na2-ATP 1, HEPES 10, and
Cs+-EGTA 10 (pH 7.2). Cs2+ was substituted for
K+ in the internal solution because it eliminates and
blocks all K+-dependent conductances. The cells were
superfused with Cs-Ringer containing (mmol/L) NaCl 20, CsCl 129,
CaCl2 1.8, MgCl2 1.0, and HEPES 10 (pH 7.4,
adjusted with CsOH). Ca2+ conductances were blocked by 500
µmol/L LaCl3.
To prepare a solution of 4-HNE, 0.5 mL of 4-HNE-DEA (50 mg/mL), dissolved in chloroform, was bubbled with nitrogen. Chloroform was evaporated, and 5 mL of 0.1 mmol/L HCl was added to the oily residue that remained. The turbid solution obtained became clear as saponification continued. Conversion of 4-HNE-DEA to 4-HNE was complete in 60 minutes. The concentration of the aldehyde was determined by measuring the absorbance of the sample at 223 nm with an extinction coefficient of 13 750 (mol/L)-1 · cm-1 (H. Esterbauer, Universität Graz, Austria; unpublished data).
Data Analysis
Equation 1 was used to analyze the rate of decrease in currents
on superfusion with the aldehyde to determine of the decay:
 | (1) |
) of INa, INa was elicited by
depolarization from a holding potential of -95 mV to various test
potentials ranging from -60 to +20 mV. The sodium conductance
(gNa) of the maximal inward current INa max at
each potential (V) was calculated according to Equation 2:
 | (2) |
were calculated by taking
the cubic root of the normalized gNa values and were fitted
to the following equation30 :
 | (3) |
, which is
the ratio of the current obtained with and without conditioning pulse,
was fitted by the same model as m:
 | (4) |
and k is the slope factor.
Data in the text are presented as mean±SD of the population. For
comparision of population means, the SEM is presented in the
tables and figures. Statistical significance was determined by using
Student's t test for comparing two population means. For
multiple comparisions of the control mean to other group means,
Dunnett's test was performed after repeated-measures
ANOVA31 when changes were comparable to the value of the
parameter before exposure to the aldehyde (Table 1 and Figs 5 and 9).
When both the treated and control groups were varying in time (Figs 7 and 8), data were analyzed by using an ANOVA procedure for a two-factor
experiment with repeated measures on one factor (time). The two factors
were treatment (control and treated) and time. Data were considered
statistically significant at P<.05. Linear and nonlinear
curve fitting were performed using NFIT (Island Products).
In all cases, the best fit to the data was chosen on the basis of the
regression coefficient R and the 
2 test.
|
|
) and in normal Ringer's solution
containing 400 µmol/L 4-hydroxynonenal (4HNE, 
). B, Rate of
decline in the maximal transient inward current elicited by a 500-ms
depolarization to 0 mV from a holding potential of -40 mV in normal
Ringer's solution (
|
|
) and 20 (
)
minutes of superfusion with 400 µmol/L 4-HNE. C and D, Current
tracings obtained on 25-ms depolarizations from a holding potential of
-95 mV in Cs-Ringer before (C) and after (D) 20 minutes of superfusion
with 4-HNE. *P<.05.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Isolated myocytes superfused with n-Ringer retained rod-shaped morphology for an observation time of 3 hours. When isolated myocytes were superfused with n-Ringer containing 4-HNE, the myocytes underwent a concentration- and time-dependent reduction in resting cell length (rigor shortening). When the decrease in cell length was complete, the resting length of the myocytes in rigor was 58.3±7.5% (mean±SD, n=550) of cell length before superfusion with 4-HNE. Myocytes in rigor excluded trypan blue, indicating that the barrier functions of the plasma membrane were unaffected by rigor shortening. Fig 1A
shows TR of myocytes superfused with
different concentrations of 4-HNE. Increasing 4-HNE concentration from
100 to 300 µmol/L decreased TR from 100 to 60 minutes. At
concentrations >400 µmol/L, little further decrease in
TR was observed. The time course of rigor shortening of a
myocyte exposed to 400 µmol/L 4-HNE is shown in Fig 1B. The initial
phase of superfusion with the aldehyde resulted in a progressive, but
slight, decrease in the resting cell length. At 41 minutes, the resting
cell length of the myocyte decreased abruptly, after which it remained
unchanged for the rest of the experiment (60 minutes). The average
TR of myocytes exposed to 400 µmol/L 4-HNE (in n-Ringer,
pCa 3.0) was 44.7±6.4 minutes. When isolated myocytes were superfused
with 400 µmol/L 4-HNE in pCa 6.0–La-Ringer, the myocytes underwent
rigor shortening with a TR of 42±8.5 minutes (Fig 1C). The
resting cell length of the myocytes in rigor was 65±10.1% of the
initial cell length. Neither TR nor the total reduction in
the cell length in Ringer's solution at pCa 3.0 (n-Ringer) or pCa
6.0–La-Ringer was statistically different (P<.05).
|
Electrophysiological effects of 4-HNE, on both the passive and the
active electrical properties of myocytes, were studied. In n-Ringer,
the mean RMP of cardiac myocytes measured within 1 to 2 minutes of
rupturing the membrane was -71.3±3.8 mV (n=81). The mean input
resistance (Rin) of myocytes, estimated by small current
injections from the resting potential, was 86.3±20.5 M (n=39).
Superfusion with 400 µmol/L 4-HNE in n-Ringer led to a rapid increase
in Rin with attendant membrane depolarization (Fig 2). After 30 minutes of superfusion with 4-HNE, a rapid
and large decrease in Rin of the myocyte was observed. In
five similar experiments, Rin decreased by 43.5±9.8 M
(compared with Rin before exposure to 4-HNE). Decrease in
Rin was accompanied by prominent membrane hyperpolarization
and myocyte rigor. Sustained superfusion with 4-HNE led to progressive
depolarization of the membrane and a parallel increase in
Rin of the myocyte (Fig 2).
|
To evaluate the effects of 4-HNE on the active properties, myocytes,
stimulated at a constant rate of 1 Hz, were superfused with 400
µmol/L 4-HNE. The earliest observable effects of the aldehyde were
membrane depolarization and an increase in the action potential
duration (Table 1
). Within 10 minutes of exposure, the
increase in the action potential duration (APD) was evident, both at
50% and 90% repolarization. No statistically significant change in
the maximal velocity of the upstroke of the action potential was
observed. Superfusion with 4-HNE also led to depolarization of the RMP,
which could contribute to the prolongation of the action potential (Fig 3A through 3D). Parameters of the action potential,
recorded after different times of superfusion with the aldehyde, are
shown in Table 1. Changes in the action potential were rapid, and
myocytes superfused for more than 25 minutes with the aldehyde were
electrically unexcitable. The time at which myocytes superfused with
400 µmol/L 4-HNE became inexcitable was 27±8 minutes (n=5). The
4-HNE–induced loss of excitability of the myocytes was followed by
gradual hyperpolarization, a decrease in Rin, and myocyte
rigor.
|
To investigate the mechanism of early changes in 4-HNE–exposed
myocytes, whole-cell currents of the myocytes were studied under
voltage clamp. Hyperpolarization to -85 mV from a holding potential of
-40 mV elicited slowly inactivating inward currents, which reversed
around -70 mV (steady state currents). These currents decreased as the
test pulse became more positive (Fig 4A). At test
potentials positive to the holding potential, the current became inward
again. These transient inward currents were completely inhibited by 2
mmol/L Mn2+ or 10 µmol/L verapamil (data not shown),
suggesting that they represent ICas through the
L-type Ca2+ channels (transient currents). The I-V relation
of the whole-cell currents obtained at the beginning of the test pulse
shows a characteristic N shape (Fig 4D, filled circles). Superfusion
with 400 µmol/L 4-HNE led to a decrease in the magnitude of both the
steady state and the transient inward currents (Fig 4B), and the
reversal potential of these currents shifted to more positive
potentials (Fig 4D, open circles). This shift in the reversal potential
corresponded in time to the early membrane depolarization. Further
superfusion with 4-HNE abolished the time dependence of whole-cell
currents (Fig 4C). Whole-cell currents obtained after 45 minutes of
superfusion with 4-HNE show a linear dependence on the membrane
voltage, and the reversal potential of these currents shifted to more
negative potentials (Fig 4D, open triangles). This shift in the
reversal potential corresponded in time to membrane hyperpolarization,
observed under current-clamp conditions.
|
Since whole-cell recordings change in time because of dilution of the
cell cytoplasm with the internal solution of the patch pipette,
time-dependent changes obtained in the presence of 4-HNE were compared
with time-dependent changes in control experiments, in which the cells
were superfused with n-Ringer alone. Under superfusion with n-Ringer,
the steady state currents (obtained on hyperpolarization to -85 mV
from a holding potential of -40 mV) did not show a large rundown, and
the rate of linear decrease was 0.15±0.03% per minute (Fig 5A). On superfusion with 4-HNE, these currents decreased
exponentially with a value of 15.7±3.2 minutes (n=5) (Fig 5A).
However, the ICas, even under control conditions, showed
appreciable exponential rundown (=16.9±3.9, n=5, Fig 5B), and this
rundown was not significantly (P>.05) affected by the
presence of 4-HNE (=11.5±3.8, n=4, Fig 5B). No increase in the
amplitude of ICas was observed at any time during
superfusion with 4-HNE.
Fig 6 shows whole-cell currents obtained in response to
test pulses applied from a holding potential of -40 mV in n-Ringer
containing 2 mmol/L Mn2+ (to inhibit ICa).
These currents show both time- and voltage-dependent kinetics and
prominent rectification, with a reversal potential of -75 mV (Fig 6C,
open circles). The rundown of these currents was small, 0.18±0.06%
per minute (n=3). On superfusion with 4-HNE, these currents decreased
with a value of 18.2±3.91 minutes (n=5). Currents obtained after
superfusion with 4-HNE for 32 minutes are shown in Fig 6B. The I-V
relation of these currents obtained after a 32-minute exposure to 4-HNE
shows a reversal potential of -45 mV (Fig 6C, filled circles), which
was also the RMP of the cell recorded at the same time. The I-V
relation of the inhibited current (obtained on subtraction of the I-V
relations of these currents before and after the addition of 4-HNE, Fig 6D) shows rectification at positive potentials and a reversal potential
of -82 mV, which is close to the K+ equilibrium potential,
suggesting that 4-HNE inhibits IK1.
|
Fig 7 shows current tracings obtained on depolarizations
from a holding potential of -95 mV, recorded in Cs-Ringer, in which
Ca2+ conductances were blocked by La3+. The
currents show both time- and voltage-dependent kinetics and were
totally inhibited by 50 µmol/L tetrodotoxin, suggesting that they
represent the voltage-dependent fast inward Na+
conductance. These currents showed a time-dependent decline on
superfusion with Cs-Ringer. For a 30-minute observation period,
INa that was elicited on depolarization to -10 mV from a
holding potential of -95 mV showed a linear decline, with a slope of
0.020±0.001 per minute (n=7, Fig 7A, open circles). When the cells
were superfused with 400 µmol/L 4-HNE, a significant
(P<.05) increase in the current was observed after 10
minutes of exposure to the aldehyde (Fig 7A, filled circles) compared
with the magnitude of the current observed at the corresponding time in
Cs-Ringer alone. The I-V relation of INas obtained in
n-Ringer showed a graded increase in current on depolarization, with a
reversal potential of 14.9±2.3 mV (n=9), which is close to the
reversal potential predicted by the Nernst equation under the
present experimental conditions (=17 mV). Superfusion with 4-HNE
led to no significant change in the reversal potential (Fig 7B),
although shifts in the voltage dependence of INa were noted
(see below). In five separate experiments, the reversal potential of
INa after 20 minutes of superfusion with 4-HNE was
15.6±2.7 mV. Furthermore, the kinetics of activation or decay of these
currents appear to be unaffected by the presence of the aldehyde
(compare Fig 7C and 7D).
The increase in the magnitude of INa on exposure to the
aldehyde appears to be due to a change in the voltage dependence of the
steady state parameters for activation (m) and
inactivation (h) of the channel. In the absence of the
aldehyde, the values of membrane voltage at half maximal
m and h (Vm and
Vh, respectively, in Equations 3 and 4) shifted to
hyperpolarizing potentials, without a change in the slope factors (s
and k). Such time-dependent shifts in the voltage dependence of the
Na+ channel during a series of voltage-clamp recordings
have been reported previously.29 The rate of shift of
Vm (0.28±0.13 mV/min, n=5) was not statistically different
(P<.05) from the rate of shift of Vh
(0.39±0.15 mV/min, n=5). Moreover, in agreement with previous
observations,29 the shifts appeared linear (Fig 8B and 8C), with no sign of saturation. Exposure to 400
µmol/L 4-HNE initially led to a shift in the voltage dependence of
m and h (values of both Vm
and Vh shifted to depolarizing potentials) (Fig 8B and 8C),
which would at positive potentials result in greater availability of
the current. After this initial depolarizing shift, both Vm
and Vh shifted back to hyperpolarizing potentials, at a
rate roughly comparable to that of untreated cells. Nevertheless, the
values of Vm and Vh in 4-HNE–superfused
myocytes remained, at most times, more positive than the corresponding
values of these parameters in untreated cells. No significant change in
slope factors k and s of these parameters was observed. The change in
the voltage dependence of the activation and inactivation parameters of
INa resulted in a larger voltage range of overlap of the
two curves. Fig 9 shows the window current
(Iw) calculated from the overlap of the activation and
inactivation parameters of INa
(=m3h) at different times of
superfusion with Cs-Ringer (filled circles) or Cs-Ringer containing
4-HNE (open circles). Myocytes exposed to the aldehyde displayed
twofold to fourfold higher Iw after 10 minutes of
superfusion with 4-HNE.
To investigate the mechanism of 4-HNE–induced decrease in
Rin, myocytes were superfused with 4-HNE, and outward
currents obtained on depolarization to 15 mV from a holding potential
of -40 mV were recorded. In the experiment shown in Fig 10, a small decrease in the outward current was
observed on superfusion with the aldehyde for 30 minutes, after which a
large increase (threefold) in the current was observed. In four similar
experiments, the time to increase in the outward current was 37.8±14
minutes, and the total increase in current was threefold to sevenfold.
The increase in the magnitude of this outward current coincided in time
with rigor shortening of myocytes and was completely reversed by the
addition of 10 µmol/L glibenclamide to the superfusate (Fig 10). This
observation suggests that the increase in the amplitude of
glibenclamide-sensitive current may underlie the sharp decrease in
Rin observed on rigor shortening of the myocytes.
|
Since exposure of myocytes to 4-HNE affected the membrane potential,
the effects of the aldehyde on the leak current were examined. Myocytes
were superfused with Cs-Ringer containing 500 µmol/L La3+
and 50 µmol/L tetrodotoxin, and the steady state currents were
recorded with Cs+-containing internal solution (for
composition see "Materials and Methods"). Fig 11
shows the I-V relation of the steady state currents obtained under
these conditions. These currents reversed near 0 mV, indicating that
the channel is largely unselective. In four separate experiments, the
average slope resistance of these currents (calculated from the I-V
relations) was 249.2±52.3 M. Superfusion with 4-HNE for 50 to 55
minutes did not cause a significant change in either the magnitude of
these currents, the slope resistance (=235±55 M, n=3), or the
reversal potential (Fig 11).
|
Table 2 lists the concentration of phosphorylated
nucleotides in neutralized extracts of myocytes that were superfused
with n-Ringer or in n-Ringer containing 400 µmol/L 4-HNE. Extracts of
myocytes superfused with n-Ringer were found to contain high
concentrations of ATP with an energy charge close to 1.0 (Table 2). The
ATP-to-ADP ratio of these myocytes was also high (>6.0), and the
concentrations of AMP were low. On superfusion of the myocytes with 400
µmol/L 4-HNE, the concentration of ATP in the extracts decreased from
34 to 2 nmol/mg protein (P<.05) (Table 2). Statistically
significant decreases were also obtained in the ATP-to-ADP ratio,
whereas the concentration of AMP increased >10-fold. No statistically
significant change (P>.05) in the ADP concentration was
observed (Table 2).
|
In addition to a decrease in ATP concentration, neutralized extracts of myocytes superfused with 400 µmol/L 4-HNE also showed a decrease in the nonprotein thiol concentration. The concentrations of nonprotein and protein thiols in cells incubated with n-Ringer alone for 60 minutes were 10.5±2.4 and 220±24 nmol/mg protein, respectively (n=3). In contrast, the concentrations of nonprotein and protein thiols of cells incubated with 4-HNE were 3.59±1.8 and 219±53 nmol/mg protein, respectively (n=3). This constitutes a 65% decrease in the concentration of nonprotein thiols in the presence of 4-HNE. The difference in nonprotein thiols between myocytes that were or were not exposed to 4-HNE was statistically significant (P<.05), whereas no statistical difference in the concentration of protein thiols was observed (P>.05).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Free radical–mediated lipid peroxidation reactions have been implicated in the genesis of cardiac injury under a variety of conditions, including ischemia/reperfusion, drug toxicity, and aging. Lipid peroxidation is a complex collection of autocatalytic reactions that are as yet poorly understood. Peroxidation of unsaturated fatty acids leads to the formation of hydroperoxides, which are catalytically decomposed by iron to alkoxy and peroxy radicals. These transient free radical species on oxidation form stable aldehydes such as 4-HNE.3 4 5 6 Although the cardiotoxic effects of transient free radicals generated during lipid peroxidation have received considerable attention, the effects of stable products of lipid peroxidation have not been investigated in detail, even though recent evidence suggests that some of the toxic effects of "free radicals" may be mediated by lipid aldehydes.10 The present study was therefore designed to identify lipid aldehyde-induced changes in cellular electrophysiology and metabolism.
Isolated cardiac myocytes exposed to 4-HNE, a product of lipid peroxidation, showed time- and concentration-dependent rigor shortening. A saturable concentration dependence of TR suggests that a critical concentration of the aldehyde is necessary for the development of its toxic effects. An increase in the concentration of the aldehyde of >400 µmol/L did not result in a large decrease in TR, suggesting that the delay in TR at lower concentrations of the aldehyde could be due to the time required for accumulation of the aldehyde. 4-HNE is a lipophilic aldehyde, and although its partition coefficient between cell cytoplasm and membrane lipids is not known, its partition coefficient between water and chloroform is 0.04.32 By use of this value, the expected concentration of 4-HNE in the membrane lipids of cells exposed to 100 µmol/L 4-HNE would be 2.5 mmol/L and to 880 µmol/L 4-HNE would be 22 mmol/L. Lipid peroxidation of rat liver microsomes by ADP/Fe3+ results in the accumulation of 417 nmol of lipid aldehydes per milligram microsomal protein, which corresponds to a local aldehyde concentration of 100 mmol/L.33 Estimates of lipid aldehyde concentration in the liver of rats intoxicated with haloalkanes range from 4 to 11 mmol/L,34 depending on the chemical nature of the haloalkane. Even during early stages of lipid peroxidation, the expected local concentration of 4-HNE was estimated to be 4.5 mmol/L or more.35 For most of the experiments reported in the present study, a near-saturating concentration of 400 µmol/L was used, which would result in the accumulation of 10 mmol/L 4-HNE in the lipid membrane and which is within the expected range of concentrations of the aldehyde accumulated during lipid peroxidation.
Electrophysiological investigations of myocytes exposed to 400 µmol/L 4-HNE show two distinct phases of injury. Early changes, or phase I of the injury, included alterations in the action potential configuration and membrane depolarization. Continued superfusion with the aldehyde resulted in the development of more severe alterations, late effects or phase II of the injury. Prominent changes observed during phase II include a sharp decrease in Rin, activation of large outward currents, and rigor shortening. Changes during phase II appear to be related to the aldehyde-induced decrease in [ATP]i and a loss of energy charge.
Changes in the action potential configuration were the earliest observable effects of the aldehyde. It is clear, from the data presented above, that the increase in the APD is not due to an increase in L-type ICas, because no change in these currents was observed on superfusion with 4-HNE, at times when the APD was markedly increased. Prolongation of the action potential by 4-HNE could, however, be due to changes in the voltage dependence of INa. The observation that the maximal velocity of the upstroke of the action potential was not affected, even when the RMP was depolarized, suggests changes in the voltage dependence of this current. The voltage dependence of both the activation and the inactivation were affected; this occurrence led to an increase in the magnitude of the total current available on depolarization to membrane potentials >-95 mV, and an increase in the steady state Iw current was observed. A similar increase in Iw has been observed on exposure of cardiac myocytes to lipid peroxide–generated oxidative stress.36 The hydrogen peroxide–induced increase in APD of cardiac myocytes has been shown to be sensitive to tetrodotoxin.37 However, unlike myocytes exposed to tert-butyl hydroperoxide36 or hydrogen peroxide,37 no decrease in the rate of inactivation of INa was observed with 4-HNE.
Depolarization-induced activation of the fast inward INa is largely responsible for the upstroke of the action potential, but most of the INa is inactivated at the times and the voltage range of the action potential plateau. However, a small, but significant, portion of the Na+ channels opens late on depolarization. Such late INa has been shown to contribute to the configuration of the cardiac action potential.38 39 40 41 Also, a tetrodotoxin-sensitive Iw resulting from an overlap of the voltage dependence of the activation and the inactivation of the current has been observed.42 This window extends well into the range of the action potential plateau. An inhibition of the late or the window INa has been suggested to be responsible for the change in the APD caused by tetrodotoxin and local anesthetics.39 40 41 42 The 4-HNE–induced changes in Iw could therefore affect the APD, thereby altering the effective refractory period, and in depolarized ischemic heart may contribute to the genesis of triggered activity or ectopic automaticity. However, since changes in the APD caused by 4-HNE occurred at times when IK1 was inhibited, prolongation of the action potential could be also partly due to the observed decrease in the outward currents through IK1 and accentuated by the attendant depolarization of the RMP.
Another early effect of 4-HNE exposure was time-dependent depolarization of the myocyte. Aldehyde-induced depolarization could be due to progressive inhibition of the Na+ pump, inhibition of IK1, or an activation of the [Ca2+]i-activated cation channel. Oxygen metabolites have been shown to inhibit the Na+,K+-ATPase in isolated sarcolemmal vesicles43 and the Na+-K+ pump current in whole-cell voltage-clamp studies.44 However, the observation that membrane depolarization by 4-HNE was accompanied by an increase in Rin suggests that inhibition of the Na+ pump may not be responsible for membrane depolarization. The Na+ pump is considered a near-zero conductance current source, and inhibition of the Na+ pump does not affect Rin at rest.45 46 Oxidative stress generated by photoactivation of rose bengal has been found to activate a steady state membrane conductance with a reversal potential near 0 mV, which may be due to an increase in Ca2+-activated nonselective cation channels.47 The 4-HNE–induced membrane depolarization was, however, accompanied by an increase in Rin, which is inconsistent with an increase in the nonspecific membrane conductance or leak. Direct estimations of the leak current in the presence of La3+, Cs+, and tetrodotoxin do not show significant effects of the aldehyde. The I-V relation of the current inhibited by 4-HNE does not reverse near 0 mV but displays a reversal potential around the K+ equilibrium potential and a prominent rectification at positive potentials, suggesting that the inhibited current is most likely IK1 (the inward rectifier K+ current). In cardiac myocytes, IK1 determines the RMP,48 and inhibition of IK1 could lead to membrane depolarization.
Free radical injury, induced by exogenously generated free radicals, has been shown to be associated with membrane depolarization,47 49 and membrane depolarization has also been observed on reperfusion of ischemic myocardium.50 Recent investigations show that myocardial depolarization during hemorrhagic shock is also due to free radical–induced oxidative stress.51 Voltage-clamp studies suggest that depolarization caused by free radical–initiated reactions47 or lipid peroxides49 may be due to an inhibition of IK1. A similar effect of 4-HNE, which is a product of lipid peroxide decomposition, thus suggests the possibility that oxidative stress–induced depolarization during ischemia/reperfusion or hemorrhagic shock may be due to inhibition of IK1 by the products of lipid peroxidation, such as 4-HNE.
Loss of excitability of myocytes was followed by a sudden decrease in Rin, which coincided with rigor shortening. Previous investigators have observed myocyte rigor after metabolic inhibition, ischemia, and free radical injury, conditions that lead to a decrease in [ATP]i.52 53 54 55 Myocyte rigor observed on ischemia56 or reperfusion of ischemic myocytes57 has been found to be independent of [Ca2+]i. The observation that 4-HNE–induced rigor was not prevented by a reduction in [Ca2+]o and by the addition of La3+ in the superfusing Ringer's solution does not support the view that increased Ca2+ influx is responsible for rigor shortening. Moreover, myocyte rigor was associated with a pronounced decrease in Rin. Voltage-clamp experiments showed that a large increase in outward currents accompanied the fall in Rin. The outward currents activated by 4-HNE were time independent, displayed a linear dependence on the membrane potential, and were sensitive to glibenclamide. These characteristics suggest that the decrease in Rin during rigor shortening may be due to the activation of the ATP-sensitive K+ channels.
Glibenclamide-sensitive ATP-sensitive K+ channels are activated by low [ATP]i and have been implicated in hyperpolarization of cardiac cells during metabolic inhibition or ischemia58 59 60 and, significantly, under conditions of oxidative stress.61 62 The ATP-sensitive K+ channels have a high unit conductance and are capable of generating large outward currents in metabolically inhibited cardiac cells.60 Activation of these currents in 4-HNE–exposed myocytes suggests that either the aldehyde has a direct effect on these channels or that the aldehyde causes a depletion of [ATP]i, which in turn leads to the activation of these channels. Direct measurement of ATP concentrations in the extracts of myocytes exposed to 4-HNE revealed that the aldehyde causes a severe depletion of [ATP]i, with a corresponding loss of the cellular energy charge. Therefore, rigor shortening and activation of outward currents appear to be due to 4-HNE–induced depletion of [ATP]i, although direct stimulatory/inhibitory effects of 4-HNE on the ATP-sensitive K+ channel cannot be ruled out and require further investigations.
Although exposure of a number of cell types to high concentrations of
4-HNE has been shown to lead to inhibition of cellular respiration and
a decrease in [ATP]i,10 the specific
metabolic pathways inhibited by the aldehyde are not known. Exposure to
free radicals generated by lipid peroxides or hydrogen peroxide has
been shown to result in an inhibition of glyceraldehyde-3-phosphate
dehydrogenase and pyruvate dehydrogenase.63 64 If 4-HNE
displays a similar inhibition profile, the exposure to the aldehyde
would result in an inhibition of glycolysis. Since the ATP-sensitive
K+ channels are relatively more sensitive to the glycolytic
rather than mitochondrial ATP,65 an inhibition of
glycolysis may result in activation of this channel, even though the
bulk (cytoplasmic) ATP may be high. In the whole-cell experiments
described above, 1 mmol/L ATP was included in the patch pipette;
nevertheless, 4-HNE caused activation of the ATP-sensitive
K+ channels, probably because the pipette ATP was not in
rapid equilibrium with the cytoplasmic ATP. The maximal available
conductance through the ATP-sensitive K+ channels in rat
cardiac myocytes has been estimated to be 
1150 nS per
cell.66 The 4-HNE–induced myocyte rigor was associated
with at least a 40-M decrease in Rin, which corresponds
to an increase in the cell conductance of 20 nS or the activation of
2% of the total number of ATP-sensitive K+ channels in
the cell. This could occur if [ATP]i decreases to
slightly less than 1 mmol/L. Exposure to 4-HNE was found to result in
an 93% decrease in [ATP]i. When total
[ATP]i is assumed to be 5 mmol/L, the
[ATP]i under rigor would be 0.3 to 0.35 mmol/L, which
would result in the activation of a substantial (5% to 10%) fraction
of the ATP-sensitive K+ channels, thus accounting for the
observed decrease in Rin. Rigor contracture of cardiac
myocytes, however, does not occur if the intracellular concentrations
of Mg2+ and ATP are >0.1 mmol/L.67 It is
likely that in 4-HNE–treated myocytes undergoing rigor contracture,
most of the residual ATP is located in the mitochondria, as occurs
during glucose-free anoxia,68 and that the cytoplasmic
concentrations of Mg2+-ATP are <0.1 mmol/L.
In agreement with previous reports,15 16 17 18 exposure of myocytes to the aldehyde led to depletion of intracellular nonprotein thiols. No change, however, was observed in the concentration of protein thiols. Depletion of nonprotein thiols such as glutathione does not, by itself, affect the viability of cardiac myocytes. Experiments of Timmerman et al69 show that total glutathione could be depleted from 11 to 1 nmol/mg protein without affecting [ATP]i, rod-shaped morphology of the myocytes, or the integrity of the sarcolemma. Therefore, it appears that the partial depletion of glutathione observed in these experiments could not account for ATP depletion or for the rigor shortening of myocytes. Since no significant oxidation of protein thiols was observed, the effects of 4-HNE could also not be attributed to nonspecific oxidation of functional sulfhydryl groups of cytoplasmic proteins. However, selective modification of a few membrane sulfhydryl or lysine groups cannot be ruled out. Further experiments are necessary to identify specific modification in the inward rectifier channel protein and the ATP-generating metabolic processes that are sensitive to 4-HNE.
In summary, the studies described above show that exposure of cardiac myocytes to 4-HNE leads to early effects on the action potential and the RMP (phase I). Sustained exposure to the aldehyde results in hyperpolarization, decrease in Rin, reduction in cell length, and activation of ATP-sensitive IKs (phase II), changes that seem to result from depletion of [ATP]i. It is likely that such electrophysiological changes underlie the observed vasodilation, loss of systole, and contractile failure20 21 22 of isolated hearts exposed to 4-HNE. Furthermore, the toxic effects of the aldehyde, such as depletion of [ATP]i, increase in the Na+ Iw, activation of the ATP-sensitive IK, and inhibition of IK1, are similar to those observed under free radical–induced oxidative stress.36 37 47 49 61 62 63 64 On the basis of these observations, it is suggested that stable products of lipid peroxidation, such as 4-HNE, may contribute to or play a critical role in the evolution of free radical injury and the genesis of oxidative stress.
|
Acknowledgments |
|---|
This study was supported by National Institutes of Health grant HL-44676. The technical assistance by Gregory J. Castro and Jeff Almrud is gratefully acknowledged. Special thanks are due to Si-Qi Liu for his help in determining ATP levels and to Philip T. Palade for his helpful comments and suggestions. The gift of 4-hydroxynonenal, generously provided by Dr H. Esterbauer, is gratefully acknowledged.
|
Footnotes |
|---|
Previously published as a preliminary report in abstract form (FASEB J. 1994;8:A607).
Received October 6, 1994; accepted October 27, 1994.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- Halliwell B, Gutteridge JMC. Free Radicals in Biology and Medicine. Oxford, England: Clarendon Press; 1985.
- Halliwell B, Gutteridge JMC. Role of free radicals and catalytic metal ions in human disease: an overview. Methods Enzymol. 1990;186:1-85. [Medline] [Order article via Infotrieve]
- Sies H, ed. Oxidative Stress. New York, NY: Academic Press Inc; 1985.
- Vigo-Pelfrey C, ed. Membrane Lipid Oxidation, I and II. Boca Raton, Fla: CRC Press; 1990.
- Chan HWS, ed. Autoxidation of Unsaturated Lipids. New York, NY: Academic Press Inc; 1987.
- Slater TF, ed. Free Radicals, Lipid Peroxidation and Cancer. London, England: Academic Press Inc; 1982.
- Esterbauer H, Zollner H, Schaur RJ. Aldehydes formed by lipid peroxidation: mechanisms of formation, occurrence and determination. In: Vigo-Pelfrey C, ed. Membrane Lipid Oxidation, I. Boca Raton, Fla: CRC Press; 1990.
- Esterbauer H. Lipid peroxidation products: formation, chemical properties and biological activities. In: Poli G, Cheeseman KH, Dianzani MU, Slater T, eds. Free Radicals in Liver Injury. Washington, DC: IRL Press; 1985:29-47.
- Grosch W. Reactions of hydroperoxides: products of low molecular weight. In: Chan HWS, ed. Autoxidation of Unsaturated Lipids. New York, NY: Academic Press; 1985:95-139.
- Esterbauer H, Schaur RJ, Zollner H. Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med. 1991;11:81-128. [Medline] [Order article via Infotrieve]
- Oarada M, Miyazawa T, Kaneda T. Distribution of 14C after oral administration of (U-14C) labelled methyl linolate hydroperoxides and their secondary oxidation products in rats. Lipids. 1986;21:150-154. [Medline] [Order article via Infotrieve]
- Kaneko T, Honda S, Nakano SI, Matsuo M. Lethal effects of a linoleic acid hydroperoxide and its oxidation product, unsaturated aldehydes, on human diploid fibroblasts. Chem Biol Interact. 1987;63:127-137. [Medline] [Order article via Infotrieve]
- Poot M, Verkerk A, Koster JF, Esterbauer H, Jongkind JF. Influence of cumene hydroperoxide and 4-hydroxynonenal on the glutathione metabolism during in vitro ageing of human skin fibroblasts. Eur J Biochem. 1987;162:287-291. [Medline] [Order article via Infotrieve]
- Cadenas E, Muller A, Brigelius R, Esterbauer H, Seis H. Effects of 4-hydroxynonenal on isolated hepatocytes. Biochem J. 1983;214:479-487. [Medline] [Order article via Infotrieve]
- Griffin DS, Segall HJ. Role of calcium homoeostasis in toxic liver injury induced by the pyrrolizidine alkaloid senecionine and the alkenal trans-4-OH-hexanal. J Biochem Toxicol. 1987;2:155-167. [Medline] [Order article via Infotrieve]
- Hauptlorenz S, Esterbauer H, Moll W, Pumpel R, Schauenstein E, Puschendorf B. Effects of the lipid peroxidation product 4-hydroxynonenal and related aldehydes on proliferation and viability of Ehrlich ascites tumor cells. Biochem Pharmacol. 1985;34:3803-3809. [Medline] [Order article via Infotrieve]
- Clark IA, Butcher GA, Buffinton GD, Hunt N, Cowden WB. Toxicity of certain products of lipid peroxidation to human malaria parasite plasmodium falciparum. Biochem Pharmacol. 1987;38:543-546.
- Schauenstein E, Esterbauer H. Formation and properties of reactive aldehydes. In: Submolecular Biology and Cancer. Amsterdam, Netherlands: Excerpta Medica; 1979:225-244. Ciba Foundation Series 67.
- Bickis IJ, Schauenstein E, Taufer M. Effects of hydroxypentenal on the metabolism of tumor cells and normal cells. Monatsh Chem. 1968;100:1077-1104.
- van der Kraaj AMM, de Jonge HR, Esterbauer H, de Vente J, Steinbusch HWM, Koster JF. Cumene hydroperoxide, an agent inducing lipid peroxidation and 4-hydroxy-2,3-nonenal, a peroxidation product, cause coronary vasodilation in perfused hearts by cyclic nucleotide independent mechanism. Cardiovasc Res. 1990;24:144-150. [Medline] [Order article via Infotrieve]
-
Ishikawa T, Esterbauer H, Seis H. Role of cardiac glutathine
transferase and the glutathione s-conjugate export system in
biotransformation of 4-hydroxynonenal in the heart. J Biol
Chem. 1986;261:1576-1586.
[Abstract/Free</