© 1999 American Heart Association, Inc.
Original Contribution |
Chloride Channel Inhibition Blocks the Protection of Ischemic Preconditioning and Hypo-Osmotic Stress in Rabbit Ventricular Myocardium
From the Divisions of Cardiovascular Research and Pathology (R.J.D., G.D.M., M.K.F., G.J.W.), Research Institute, The Hospital for Sick Children; Division of Cardiovascular Surgery (G.J.W.) and Department of Medicine (P.H.B.), The Toronto Hospital; and Department of Physiology (P.H.B., G.J.W.), Department of Pathobiology and Laboratory Medicine (G.J.W.), Department of Surgery (G.J.W.), and Institute of Medical Science (V.A.L., G.J.W.), The University of Toronto, Toronto, Ontario, Canada.
Correspondence to Dr Gregory J. Wilson, Cardiovascular Labs, Rm CCRW1-885, The Toronto Hospital, General Division, 200 Elizabeth St, Toronto, Ontario, Canada, M5G 2C4. E-mail diazport{at}sickkids.on.ca
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Abstract—The objective of this study was to examine the role of chloride (Cl–) channels in the myocardial protection of ischemic preconditioning (IP). Isolated rabbit ventricular myocytes were preconditioned with 10-minute simulated ischemia (SI) and 20-minute simulated reperfusion (SR) or not preconditioned (control). The myocytes then received 180-minute SI or 45-minute SI/120-minute SR. Indanyloxyacetic acid 94 (IAA-94, 10 µmol/L) or 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB, 1 µmol/L) was administered before IP or before SI or SI/SR to inhibit Cl– channels. Electrophysiological studies indicate that these drugs, at the concentrations used, selectively abolished Cl– currents activated under hypo-osmotic conditions (215 versus 290 mOsm). IP significantly (P<0.001) reduced the percentage of dead myocytes after 60-minute (30.8±1.3%, mean±SEM), 90-minute (35.3±1.3%), and 120-minute (39.2±1.7%) SI compared with controls (44.7±1.6%, 54.5±1.3%, and 58.9±1.8%, respectively) and after 45-minute SI/120-minute SR (36.3±0.6%) compared with control (56.6±2.2%). Hypo-osmotic stress also produced protection similar to IP. IAA-94 or NPPB abolished the protection of both IP and hypo-osmotic stress. In buffer-perfused rabbit hearts preconditioned with three 5-minute ischemia/10-minute reperfusion cycles given before the 40-minute long ischemia and 60-minute reperfusion, IP significantly (P<0.0001) reduced infarct size (IP+vehicle, 4.7±0.9%, versus control+vehicle, 26.6±3.3%; mean±SEM). Again, IAA-94 or NPPB abolished the protection of IP. Our results implicate Cl– channels in the IP protection of the myocardium against ischemic/reperfusion injury and demonstrate that hypo-osmotic stress is capable of preconditioning cardiomyocytes.
Key Words: ischemic preconditioning • hypo-osmotic stress • chloride channel • myocardial infarction • cardiomyocyte
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
During ischemia, metabolites accumulate in cardiomyocytes, thus increasing intracellular osmolarity.1 The development of an osmotic gradient between the intracellular and extracellular milieu, in turn, causes cell swelling. It is known that cell swelling and the loss of cell volume regulation play important roles in ischemic injury in the myocardium.2 3 Cell swelling activates a variety of transport pathways that result in the net efflux of K+, Cl–, organic anions, and organic osmolites.4 Cell swelling activates an outwardly rectifying conductance that has an anion selectivity sequence of CN–>I–>NO3–>Br–
Cl–>F–>gluconate
and is characterized by a single-channel conductance of 40 to 50 pS (at
+120 mV).5 Duan et al6 have reported a
tamoxifen-sensitive outwardly rectifying swell-activated
Cl– current
(ICl,swell) in rabbit atrial myocytes.
ICl,swell has also been reported in
dog,7 8 guinea pig,7
chick,9 and human10
cardiomyocytes. Furthermore,
ICl,swell has been reported in a variety of
noncardiac cells and demonstrates biophysical and pharmacological
properties similar to those seen in cardiac
myocytes.11 12 The regulation of Cl– channels has much in common with the controlling factors associated with the phenomenon of ischemic preconditioning (IP) of the myocardium. A Cl– channel conductance is activated by angiotensin II via AT1 receptors in sinoatrial cells13 and by adenosine in guinea pig ventricular myocytes,14 whereas activation of angiotensin II AT1 receptors15 and adenosine A1/A3 receptors16 17 is thought to trigger IP. In addition, transient exposure to phorbol 12-myristate 13-acetate, which has been shown to mimic the protection of IP via protein kinase C (PKC) activation in isolated cardiomyocytes,18 activates a Cl– current in feline19 and guinea pig ventricular myocytes20 but not in rabbit atrial myocytes.6 It is noteworthy that a transient increase in [Ca2+]i resulting from Ca2+ entry during the initial preconditioning ischemia is thought to mediate the protection of IP by triggering the activation of PKC21 and to activate a Ca2+-dependent Cl– conductance in rabbit ventricular myocytes.22 23 Furthermore, Sorota24 has shown that swell-activated Cl– currents are triggered via protein tyrosine kinases that have recently been implicated in IP protection.25
Because signaling pathways associated with IP also regulate the activity of Cl– channels, we postulated that IP activates Cl– channels, including swell-activated Cl– channels, and thereby reduces myocardial necrosis during prolonged ischemia and reperfusion. In the present study, we demonstrated in the rabbit the participation of Cl– channels in the protection of IP against myocardial necrosis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Isolated ventricular myocytes and whole hearts were obtained from New Zealand White rabbits (weight range, 3.0 to 3.5 kg). Animal protocols conformed with the Guide for the Care and Use of Laboratory Animals published by NIH (NIH publication No. 85-23, revised 1996) and was approved by the Animal Care Committee of the Research Institute, The Hospital for Sick Children.
Preparation of Ventricular Myocytes
Ventricular myocytes were isolated by enzymatic
dissociation using a method previously reported.26 27
Briefly, donor rabbits were anesthetized with a mixed solution
of pentobarbital (60 mg/kg) and heparin (200 IU/kg). The heart was
excised and immediately perfused on a nonrecirculating Langendorff
apparatus at 75 mm Hg of perfusion pressure with
oxygenated (95% O2–5%
CO2) Joklik-modified minimal essential medium
solution (S-MEM, Gibco) at 37°C, pH 7.4, and supplemented with
(in mmol/L) CaCl2 1.2,
MgSO4 0.2, creatine 20, taurine 60,
NaHCO3 24, and HEPES 10 (280 to 300 mOsm). Hearts
were perfused with calcium-containing S-MEM solution (2 to 3 minutes)
before perfusing with nominally calcium-free S-MEM solution
supplemented with EGTA (0.016 µmol/L) and 0.1% BSA (Pentex
fraction V, Sigma) for 6 to 8 minutes. Subsequently, all hearts were
perfused in recirculating mode, with the same calcium-free S-MEM
solution (coronary flow rate, 10 to 15 mL/min) supplemented
with 200 IU/mL of collagenase type II (Worthington Inc;
specific activity range, 215 to 225 IU/mg) for 15 to 25 minutes.
Isolated myocytes were filtered through a nylon mesh,
centrifuged (twice at 45g for 3 minutes), and
resuspended in calcium-free S-MEM solution containing 2% BSA. The
calcium concentration was then increased in steps over a 30-minute
period, to a final concentration of 1.07 mmol/L. The cell
suspension was centrifuged again for 2 minutes at
45g, the supernatant discarded, and the cell pellet
resuspended in 12 to 15 mL of calcium-containing (1.2 mmol/L)
S-MEM solution supplemented with 0.1% BSA. Using this procedure, we
generally obtained 8 to 9 million cells per gram of tissue with a yield
of 70% to 80% rod-shaped myocytes.
Single-Cell Electrophysiology Studies
Electrophysiological recordings were
made using the whole-cell patch-clamp technique (Axopatch 200A, Axon
Instruments) on single myocytes isolated as described above. After
isolation, myocytes were suspended in high K+
buffer containing, in mmol/L, glutamate 120, KCl 20, HEPES 20,
MgCl2 1, glucose 10, and EGTA 0.3. An aliquot of
cells was transferred into a 1-mL perfusion bath situated on the stage
of an inverted microscope and perfused with external recording
solution at a rate of 1 to 2 mL/min. To isolate
Cl– current we used the following pipette
solution (in mmol/L): N-methyl-D-glucamine
aspartate 100, NMDG-Cl 30, Mg-ATP 5, and HEPES 10 (pH adjusted to 7.4
with NMDG-OH), and osmolarity of 290 mOsm was achieved by adding
mannitol. The hypo-osmotic bath solution contained the following
(in mmol/L): NaCl 80, KCl 4, BaCl2 2,
tetraethylammonium chloride 10, CdCl2 0.2,
4-aminopyridine 5, MgCl2 0.8,
CaCl2 1.0,
NaH2PO4 0.33, HEPES 10, and
glucose 5.5 (pH adjusted to 7.4 with NaOH) and osmolarity of 215 mOsm.
The iso-osmotic bath solution was the same as the hypo-osmotic
solution, but the osmolarity was adjusted to 290 mOsm by adding 75
mmol/L mannitol. The temperature was 20°C to 24°C.
Recording pipettes were prepared from thin-walled borosilicate
glass (1.5-mm diameter, World Precision Instruments Inc) using a
Flaming-Brown micropipette puller (Sutter Instruments Inc). The pipette
tip was heat polished and, when filled with intracellular solution, had
a resistance of 1 to 2 M
. After membrane rupture, the cell
capacitance was estimated by integrating the capacity currents. The
series resistance was typically 2 to 4 M and was compensated by 60%
to 80%. For all current recordings, myocytes were held at a
membrane potential of –80 mV, and a 100-ms prepulse to –40 mV was
used to inactivate Na+ current.
Voltage steps ranged between –80 and +60 mV and were 300 ms in
duration.
Nonspecific effects of 1 µmol/L NPPB or 10 µmol/L indanyloxyacetic acid 94 (IAA-94) on ATP-sensitive K+, inward rectifier, and L-type Ca2+ channels were also assessed. To isolate ATP-sensitive K+ (IKATP) and K+ rectifier (IK1) currents, we used a bath solution containing the following (in mmol/L): NaCl 130, KCl 5.4, CaCl2 1.8, MgCl2 1, CdCl2 0.3, and HEPES 10 (pH 7.4 with NaOH). The pipette solution contained, in mmol/L, potassium aspartate 100, KCl 30, MgCl2 1, EGTA 5, and HEPES 10 (pH 7.4 with KOH). To activate IKATP, 2-deoxyglucose (10 mmol/L) and cyanide (2 mmol/L) were added to the external solution. To isolate L-type Ca2+ current (ICa), we used a bath solution containing the following (in mmol/L): NaCl 130, CsCl 5, MgCl2 1, CaCl2 1.8, HEPES 10, and glucose 10 (pH 7.4 with NaOH). The pipette solution contained, in mmol/L, CsCl 130, MgCl2 1, EGTA 5, and HEPES 10 (pH 7.4 with CsOH).
Isolated Ventricular Myocyte Studies
Simulated Ischemia (SI) and Simulated Reperfusion
(SR)
Ischemia was simulated in myocytes with the use of a
previously described method.27 28 Briefly, 1.5 mL of the
cell suspension was placed in a 1.8-mL Eppendorff tube and
centrifuged (45g for 2 minutes) to form an 8- to
10-mm-thick cell pellet. The supernatant was discarded, except for a
volume equivalent to about one third of the pellet thickness. The cell
pellet and supernatant were covered by a 3- to 4-mm–thick mineral oil
layer and incubated at 37°C.
Reperfusion was simulated using a method we have previously characterized.29 Briefly, 100 to 150 µL of the ischemic cell pellet was resuspended in 1.0 mL of oxygenated calcium-containing buffer supplemented with 0.1% BSA and incubated on a Nunclon multidish cell culture plate (Irvine Scientific, Inc) at 37°C with agitation in an O2 atmosphere.
Experimental Protocol for Ventricular Myocytes
Isolated ventricular myocytes (n=7 hearts in each
set) were studied to determine the role of Cl–
channels in the protection of IP against cell death caused by
ischemia and reperfusion. After an initial 30-minute
stabilization period (incubation in a 95%
O2–5% CO2 atmosphere at
30°C without agitation), myocytes were either preconditioned using a
10-minute period of SI (37°C) followed by 20-minute SR or not
preconditioned (control). Next, both control and preconditioned
myocytes were subjected to either 180-minute SI (ischemia
protocol) or 45-minute SI/120-minute SR (ischemia/reperfusion
protocol) (Figure 1
).
|
Control, preconditioned, and oxygenated baseline (240-minute incubation in an O2 atmosphere at 37°C) myocytes were simultaneously treated with either of 2 selective Cl– channel blockers that have distinct molecular structures and inhibitory mechanisms, 10 µmol/L IAA-94 or 1 µmol/L 5-nitro-2-(3-phenylpropylamino)benzoic acid (NPPB) (Research Biochemicals International Inc),30 31 or they were treated with the vehicle of either drug (0.03% and 0.09% ethanol for IAA-94 and NPPB, respectively). The blocker was initially added to the myocyte suspension either 10 minutes before IP or just before the long SI (IAA-94 only). When added before IP, neither blocker was washed out before the long SI. The concentrations of IAA-94 and NPPB were selected to be 10 times the IC50.31 32
We also determined the effect of IAA-94 (10 µmol/L) and NPPB (1 µmol/L) on PKC in an in vitro PKC activity assay as previously reported.9 We found that IAA-94 and NPPB resulted in 94% and 96% retention of PKC activity, respectively. PKC inhibition by 10, 25, and 50 µmol/L chelerythrine resulted in 55%, 40%, and 10% retention of PKC activity, respectively, whereas inhibition by 10 and 25 µmol/L polymyxin B resulted in 16% and 0%, respectively, validating the assay. Therefore, Cl– channel inhibitors have no significant effect on PKC activity at the concentrations used.
Because cell swelling is known to activate swell-activated Cl– channels, we also assessed whether activation of these channels by transient hypo-osmotic stress could mimic the protection of IP. After stabilization, myocytes were incubated in a hypo-osmotic (80-mOsm osmotic gradient) buffer identical to the iso-osmotic buffer except for the removal of creatine (20 mmol/L) and taurine (60 mmol/L) for 10 minutes, in place of IP, and then resuspended in iso-osmotic buffer for 20 minutes just before the long SI. IAA-94 (10 µmol/L) and NPPB (1 µmol/L) were given 10 minutes before the long SI to myocytes transiently subjected to hypo-osmotic stress. Control myocytes were also incubated in the same taurine- and creatine-free buffer supplemented with 80 mmol/L mannitol. To determine whether the concentrations of IAA-94 and NPPB used in these studies have a nonspecific effect on mitochondrial KATP channels, the effects of IAA-94 and NPPB on protection in myocytes by diazoxide33 34 were also examined. Diazoxide was added at a concentration (100 µmol/L) that selectively activates mitochondrial KATP channels.33 Each Cl- channel inhibitor (IAA-94 or NPPB) was also coadministered with diazoxide before SI in separate groups of myocytes.
Assessment of Ventricular Myocyte Viability
All histological examinations were performed
with a light microscope at x100 magnification. Myocyte viability was
assessed in each group at different time points (Figure 1
).
Hypotonic (85 mOsm) trypan blue–modified Tyrode staining solution
containing (in mmol/L) KCl 2.68, CaCl2 1.8,
NaH2PO4 0.42,
NaHCO3 11.9, MgSO4 0.83,
glucose 5.55, and amylobarbitone 3.0; 0.5%
glutaraldehyde; and 0.5% trypan blue was used to
assess myocyte viability and osmotic fragility. A 15-µL cell pellet
was obtained under normoxic or ischemic conditions and
resuspended in 200 µL of staining medium for 1 to 2 minutes
for cell counting. Dead cells stain dark blue, whereas viable cells are
not stained. As described previously, amylobarbitone, a mitochondrial
inhibitor, and the fixative glutaraldehyde
were added to the counting medium to preclude further cell rounding or
squaring.27 28 35 The morphology (rod, square, and round
shape) of isolated myocytes was determined, as reported by Vander Heide
et al.28 Myocytes were considered to have rod shape if the
length/width ratio was >3:1 and square shape if the ratio was >1:1
but <3:1. We counted 250 to 300 myocytes per sample in <15 minutes to
minimize variability associated with changes in the ratio of
stained/unstained cells over time. From these data, the percentage of
dead cells was calculated and compared. Only experiments with >70% of
viable myocytes (rod-shaped, square, and round myocytes) observed at
the end of the stabilization period were considered acceptable for this
study. Viable round myocytes (cells excluding trypan blue, 
5%) were
only found at the end of stabilization or before the long SI.
Isolated Heart Studies
Surgical Preparation
Rabbits were prepared as previously reported.15
Briefly, hearts were excised, mounted on a modified nonrecirculating
Langendorff apparatus, and immediately perfused with
Krebs-Henseleit buffer solution containing (in mmol/L) NaCl 118.5,
KCl 4.7, MgSO4 1.2, CaCl2
2.5, NaHCO3 24.8,
KH2PO4 1.2, and glucose 10
(pH 7.4 by oxygenation with 95%
O2–5% CO2) at
37°C and constant pressure of 75 mm Hg. To induce regional
ischemia, a branch of the left coronary artery was
intermittently occluded. An intraventricular latex
balloon connected to a pressure transducer was placed into the left
ventricle to assess left ventricular developed pressure
(LVDP; systolic minus diastolic pressure) and heart
rate. A probe was also placed in the heart to monitor myocardial
temperature. Once instrumented, hearts were placed in a water-jacketed
chamber and stabilized before each experiment began.
Experimental Protocol for Isolated Hearts
The stability of our preparation has been previously
demonstrated using isolated buffer-perfused rabbit hearts in which
LVDP, heart rate, and coronary flow were
measured.15 Using the same model, 30 hearts (n=5 per
group) were initially subjected to 15-minute aerobic perfusion
(stabilization period) followed by 40-minute normothermic
(37°C) regional ischemia and 60-minute reperfusion. Control
hearts were also subjected to an additional 45-minute aerobic perfusion
(total stabilization period, 60 minutes) before the long
ischemia so as to equalize the total length of the experimental
protocol in all groups. Preconditioned hearts were also subjected to 3
cycles of 5-minute regional ischemia followed by 10-minute
reperfusion before the long ischemia. To determine whether
Cl– channels play a role in the protection of IP
during the long ischemia, hearts were subjected to the same
control and IP protocols with a 10-minute exposure to IAA-94 or NPPB,
at the same concentrations used in the isolated myocyte model, before
the long ischemia.
As in isolated myocyte studies, we explored whether IAA-94 (10 µmol/L) could block the protection induced by stimulation of mitochondrial KATP channels by diazoxide (100 µmol/L) in 21 additional buffer-perfused hearts. The following 3 groups of hearts were studied: those treated with control+vehicle (0.05% DMSO; n=5), those treated with diazoxide (n=8), and those treated with diazoxide+IAA-94 (n=8) hearts. Diazoxide was given 15 minutes before the long ischemia. IAA-94 was concurrently given with diazoxide 10 minutes before the long ischemia.
Infarct Size Measurements
At the end of each experiment, the coronary artery
subjected to occlusions was reoccluded. The heart was then perfused
with 5- to 10-µm zinc-cadmium sulfide yellow fluorescent
particles (Duke Scientific, Inc) to identify the area at risk (areas
without particles). Next, hearts were cross-sectioned into 4 or 5
slices and incubated in a 1.25% solution of triphenyl tetrazolium
chloride, made with 0.2 mol/L Tris buffer (pH 7.4), at 37°C for 10
minutes. Using this staining method, viable tissue stains brick-red,
and necrotic tissue looks white or tan. The necrotic, risk, and total
areas from each heart slice were then traced onto an acetate sheet and
computer planimetered to calculate the percentage of the
biventricular area that was at risk (risk area/total area)
and the area at risk that was necrotic (necrotic area/risk area).
Statistical Analysis
All data shown are expressed as mean±SEM. The isolated myocyte
and buffer-perfused heart data were first tested for normality
(Kolmogorov-Smirnoff test) and homogeneity of variance (Levene test).
Because the criteria for parametric analysis were not
met, we performed a nonparametric analysis using
the multigroup-comparison Kruskal-Wallis test to assess for differences
among the groups, followed by the post hoc Dunn procedure to determine
whether a statistically significant difference (P<0.001)
existed between 2 groups. ANOVA was performed in the whole-heart
studies, in which the data were found to meet the criteria for
parametric tests to assess for differences among the groups.
Where appropriate, the Scheffé F test was then applied to
determine whether a statistically significant difference
(P<0.05) existed between 2 groups. A regression
analysis was performed to assess for any association between 2
measurements in each study. For all
electrophysiological studies, the paired
Student t test was used.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Single-Cell Electrophysiology Studies
In Figure 2A
, the magnitude of the
outward current recorded from a representative
myocyte is shown. Initially, the myocyte was superfused with the
iso-osmotic external solution (290 mOsm). After 5 minutes, the
hypo-osmotic external solution (215 mOsm) was applied, and this evoked
a 2-fold increase in outward current after 4 minutes. Application of
NPPB significantly (P<0.001) reduced outward current to
below that observed in iso-osmotic conditions. Washout of NPPB with
hypo-osmotic external solution resulted in complete normalization of
swell-activated outward current.
|
in panel A), a modest
Cl– current (ie, ICl,b) was
observed. Hypo-osmotic conditions resulted in a 2-fold increase in
Cl– current (
in panel A). NPPB (1 µmol/L; + in
panel A) blocked both Cl– currents. Blockade by NPPB was
completely reversed by washout of Cl– channel
inhibitor (
in panel A). Dotted line
represents 0 current. C and D, Mean (±SEM) I-V
relationship curves for NPPB and IAA-94, respectively. Iso-osmotic
conditions (
) normalized current to basal
level, whereas washout with hypo-osmotic solution (
Raw traces are shown in Figure 2B and mean (±SEM)
current-voltage (I-V) plots in Figure 2C. Under
iso-osmotic conditions, the currents were outwardly rectifying, as
reported previously for Cl–
currents.6 The current reversed at –27.8 mV, close
to the calculated Cl– reversal potential of
–30.4 mV. After a 10-minute hypo-osmotic period, the current increased
at all voltages, continued to outwardly rectify, and reversed at –25.2
mV. The outward current density increased by 46.3% at +60 mV (n=4,
P<0.01). This swell-activated current was time
dependent, taking 5.6±2.2 minutes (mean±SEM) to activate,
consistent with a previous report.6
Application of 1 µmol/L NPPB inhibited both iso-osmotic and
hypo-osmotic currents. Washout of NPPB with iso-osmotic external
solution restored outward current to baseline, while washout with
hypo-osmotic solution restored outward current to a
swell-activated level. Similar results were observed when
IAA-94 was applied (Figure 2D). These results are
consistent with the presence of a basal
Cl– current (ICl,b)
and ICl,swell, as previously reported in
cardiac myocytes.6 7 10 36
Previous studies have reported that NPPB (10 to 40 µmol/L)
activates ATP-sensitive K+ currents
(IKATP),37 whereas NPPB
(100 µmol/L) and IAA-94 (200 µmol/L) partially block
Ba2+ current through Ca2+
channels,38 albeit at concentrations much higher than
those used in this study. The mean I-V plots shown in Figure 3A and 3B summarize the effects of NPPB
and IAA-94 on sarcolemmal IKATP. In the
absence of glucose and exogenous ATP, typical inward rectifier-like
K+ currents40 were observed.
External application of the metabolic
inhibitors cyanide and 2-deoxyglucose activated a
K+ conductance typical of
IKATP.39 Application of 1
µmol/L NPPB or 10 µmol/L IAA-94 did not block or further
activate sarcolemmal IKATP
at any of the voltages tested (n=4). However, 10 µmol/L
glibenclamide blocked the current activated by
metabolic inhibition consistent with the induced
current arising from sarcolemmal IKATP
channels. To determine whether the Cl– channel
blockers could directly activate sarcolemmal
IKATP, K+ current
(ie, IK1) was recorded in the absence
of glucose and ATP. External application of NPPB (Figure 3C;
n=4) or IAA-94 (Figure 3D; n=4) had no effect on the current
even after 20 minutes, which suggests that these blockers do not
activate sarcolemmal IKATP nor
block IK1 currents at the concentrations
used in this study. Also, NPPB or IAA-94 did not affect calcium current
(ICa), as demonstrated by the mean
I-V plots in Figure 3E (n=4).
|
, in panels A and B,
respectively) did not have an effect on
IKATP. In contrast, application of 10
µmol/L glibenclamide (
) significantly reduced
IKATP. C and D, In the absence of
metabolic inhibition, IK1 was
observed (
). External application of NPPB (C) or
IAA-94 (D) did not activate IKATP
(
Whether ICl,swell is regulated by PKC in
rabbit ventricle has not been previously determined. The plot in Figure 4A shows the time course of current
magnitude in a typical cell under iso-osmotic conditions, whereas
Figure 4B shows corresponding raw traces in response to voltage
steps from –80 to +60 mV. Replacement of external iso-osmotic solution
with a hypo-osmotic solution again activated
ICl,swell. Application of chelerythrine
(20 µmol/L) in the continued presence of hypo-osmotic solution,
a concentration sufficient to block PKC,35 inhibited
both ICl,b and
ICl,swell in ventricular
myocytes by 85% and 88%, respectively, at +60 mV (n=4,
P<0.01). The effects of chelerythrine were observed at all
of the voltages tested (Figure 4C).
|
Isolated Ventricular Myocyte Studies
Myocyte viability expressed as percentage (mean±SEM %) of dead
myocytes for each group of cells during 180-minute SI and during
45-minute SI combined with 120-minute reperfusion are presented
in Figures 5 through 8. There was no difference in the percentage of dead myocytes
among untreated control myocyte groups or among untreated IP myocyte
groups, with or without the vehicle, in all cell experiments at any
time point (Figures 5 through 8). Because
there was no difference in the percentage of dead myocytes between
untreated and treated (with drug) oxygenated baseline
myocytes, in each set of experiments, we have pooled the data for
oxygenated baseline and presented it in a single
line graph (Figures 5 through 8).
|
|
|
|
) completely blocked the
protection induced by hypo-osmotic stress. B, Effect of
Cl– channel inhibition on the hypo-osmotic stress
protective effect against myocyte death caused by combined 45-minute SI
and 120-minute SR. Hypo-osmotic stress (
Furthermore, there was no difference between untreated control and
untreated IP myocytes, in terms of percentage of dead myocytes, either
before the long SI or after 30-minute SI in each set of experiments
(Figures 5 through 8). In the first set of
experiments (Figure 5A), IP significantly (P<0.001)
limited the percentage of dead myocytes after 60-, 90-, or 120-minute
SI, when compared with controls. A similar protective effect was also
observed in the second set of experiments (Figure 5B). This
protective effect of IP was not detected after 180-minute SI (Figure 5A and 5B). Inhibition of Cl– channels
with 10 µmol/L IAA-94 or 1 µmol/L NPPB, each drug given
before IP, completely abolished (P>0.05) the protection of
IP against myocyte death caused by SI alone, whereas it did not
increase myocyte mortality in treated controls, as shown in Figure 5A and 5B.
Next, we tested IAA-94 and NPPB using a model of SI combined with SR
for isolated ventricular myocytes.29 In this
model, IP significantly (P<0.001) reduced the percentage of
dead myocytes caused by a combined 45-minute SI/120-minute SR (Figure 6A and 6B). Furthermore, this IP protection was completely
abolished (P>0.05) when either 10 µmol/L IAA-94 or
1 µmol/L NPPB was added into the cell suspension 10 minutes
before IP (Figure 1) to inhibit Cl–
channels (Figure 6A and 6B). IAA-94 and NPPB had no effect on
treated controls (Figure 6A and 6B).
To investigate the specific role of Cl– channels
during the maintenance phase of IP and rule out the possibility
that Cl– channel inhibition might have blocked
the induction of IP, we added IAA-94 10 minutes before the long SI in
both the SI-alone and combined SI/SR protocols. When the SI-alone
protocol was used (Figure 7A), IP significantly
(P<0.0001) reduced the percentage of dead myocytes after
60-, 90-, or 120-minute SI as compared with untreated controls.
Similarly, IP significantly (P<0.0001) reduced the
percentage of dead myocytes after 120-minute SR when the combined SI/SR
protocol was used (Figure 7B). Complete inhibition of
Cl– channels by IAA-94 produced a total blockade
of the protective effect of IP (Figure 7A and 7B). In the same
group of myocytes in which a combined SI/SR protocol was used, we
assessed the morphology of myocytes (rod, square, and round shape) in
each group. The percentage of rod-shaped myocytes was significantly
(P<0.05) higher after 45-minute SI/120-minute SR in the
preconditioned group (IP+vehicle, 58.8±3.3%, mean±SEM) as compared
with the control group (C+vehicle 27.8±4.1%). The percentages of
square and round myocytes were significantly (P<0.0001)
lower in preconditioned groups (IP+vehicle 6.9±1.8% and 34.0±2.6%,
respectively) as compared with control groups (C+vehicle 17.3±4.1%
and 53.9±2.6%, respectively). This IP effect on myocyte morphology
was completely abolished by 10 µmol/L IAA-94 (IP+IAA-94
19.0±3.8% square and 56.3±1.5% round myocytes) as compared with
untreated IP (IP+vehicle).
Moreover, we found no correlation (P>0.05) between the percentage of dead myocytes for either IAA-94–treated or NPPB-treated control and IP myocytes as compared with oxygenated baseline myocytes at 60-, 90-, and 120-minute SI in the ischemia-alone protocol, and at 15-, 60-, and 120-minute SR in the ischemia/reperfusion protocol. Thus, the increase in the percentage of dead myocytes in IP cells treated either with IAA-94 or NPPB, was, in all likelihood, due to the inhibition of the IP protective effect rather than the consequence of increased myocyte mortality rate with the preparation.
We explored the effect of IAA-94 and NPPB at concentrations that completely blocked IP, on diazoxide-induced protection against myocyte death during ischemia.33 34 Diazoxide significantly (P<0.001) reduced myocyte mortality after either 90 (24.7±2.9%) or 120 (27.7±1.4%) minutes of SI when compared with controls (50.0±2.9% and 49.0±2.0%, respectively). Blockade of Cl– channels either with IAA-94 or NPPB did not inhibit diazoxide protection (diazoxide+IAA-94, 24.8±3.8/26.6±1.5% [90-/120-minute SI]; diazoxide+NPPB, 27.1±3.1/27.8±1.5%, respectively), indicating that IAA-94 and NPPB have no effect on mitochondrial KATP channels.
The results above show that Cl– channels
blockers can inhibit preconditioning, whereas
Cl– channels can be activated by
hypo-osmotic stress. Therefore, we examined whether hypo-osmotic stress
could induce protection. Suspending myocytes in hypo-osmotic buffer
(80-mOsm osmotic gradient) for 10 minutes and then resuspending them in
iso-osmotic buffer for 20 minutes before the long SI produced
protection similar to IP. The percentage of dead myocytes was
significantly (P<0.0001) reduced in myocytes subjected to
hypo-osmotic stress after 60-, 90-, or 120-minute SI compared with
control (Figure 8A). Furthermore, pretreatment with 10-minute
hypo-osmotic stress prevented a significant (P<0.0001)
increase in the percentage of dead myocytes after combined 45-minute SI
and 120-minute SR compared with control (Figure 8B). Blockade of
Cl– channels either with IAA-94 or NPPB
completely abolished the protection induced by hypo-osmotic stress
(Figure 8A and 8B).
Isolated Heart Studies
There was no difference among all groups, in terms of LVDP, heart
rate, and coronary flow, after 15-minute stabilization or
before ischemia or after 40-minute regional myocardial
ischemia. Preconditioned hearts treated with IAA-94 or control
(with vehicle) hearts showed significantly higher heart rate and lower
coronary flow, respectively, at the end of reperfusion, when
compared with preconditioned (with vehicle) hearts (data not shown).
However, a regression analysis performed on the function (LVDP
and heart rate) and the infarct size data between these groups
indicated that there was no association (P>0.05) between
either heart rate or coronary flow and infarct size.
IP significantly (P<0.0001) reduced infarction within
the myocardium at risk (control+vehicle versus IP+vehicle),
as shown in Figure 9. Administration of
the Cl– channel inhibitors, either
IAA-94 (10 µmol/L) or NPPB (1 µmol/L), before the long
ischemia, completely blocked the protection against infarction
by IP, whereas they did not alter infarct size in controls (Figure 9). The area at risk did not differ among the groups (data not
shown).
|
Pretreatment with 100 µmol/L diazoxide for 15 minutes
before ischemia significantly (P<0.001) reduced
infarct size as shown in Figure 9. This protective effect of
diazoxide was not inhibited by blockade of Cl–
channels with 10 µmol/L IAA-94. No significant differences were
observed in LVDP and heart rate after 60-minute reperfusion between
diazoxide-treated and control hearts. Diazoxide significantly
(P<0.001) increased coronary flow, as previously
reported by Garlid et al,34 in the presence or
absence of Cl– channel inhibition by IAA-94 or
NPPB (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The principal aim of the present study was to assess the participation of Cl– channels in the protection of IP against ischemic injury alone and combined ischemic/reperfusion injury. In isolated rabbit cardiomyocytes IP reduced myocyte death during a 180-minute SI period (Figure 5
). These results are in agreement with
previously published reports by other investigators in the same
model.18 40 A reduction in cell death after 45-minute
SI/120-minute SR was also observed with IP (Figure 6A and 6B).
Similarly, in buffer-perfused isolated rabbit hearts, IP (3 cycles of
5-minute ischemia followed by 10-minute reperfusion)
significantly reduced necrosis after 40-minute ischemia and
60-minute reperfusion, as reported previously.15 Our
results show that the protection of IP was abolished by
Cl– channel inhibition with IAA-94 or NPPB in
isolated cardiomyocytes (Figures 5 through 7) and isolated hearts (Figure 8).
Chloride channel blockers were shown in our experiments to abolish the
protection of IP when added either before IP or before the long
ischemia. Previous studies have separated the protection of IP
into an initiation phase, which involves the activation of signal
transduction during the short IP ischemia, and a
maintenance phase, involving those events related to IP
protection during the long ischemia.41 It is not
clear whether Cl– channels are involved in the
initiation. To test this, Cl– channels must be
blocked only during the initiation phase, requiring complete washout of
blockers during the 20-minute reperfusion period before the long
ischemia. Unfortunately, the ability of NPPB and IAA-94 to
block the initiation of IP could not be tested, because washout of
Cl– channel blockers required at least 20
minutes as assessed in our
electrophysiological experiments with
continuous superfusion. Nevertheless, the protection achieved by a
single brief episode of hypo-osmotic stress in place of IP supports a
role for Cl– channel activation in protection
during the initiation phase of IP, since this hypo-osmotic stress
results in Cl– channel activation (Figure 2). In addition, Cl– channels clearly
play a role in the maintenance phase, since
Cl– channel inhibition before and during the
long ischemic period abolished the protection of IP in both
isolated cardiomyocytes (Figure 7) and
buffer-perfused hearts (Figure 9).
In the present study, we used the whole-cell patch-clamp technique to demonstrate that rabbit ventricular myocytes possess currents that were inhibited by the specific Cl– channel blockers NPPB and IAA-94. Under iso-osmotic conditions designed to isolate Cl– currents, the current recorded reversed close to the reversal potential for Cl–, did not inactivate, and was outwardly rectifying, as reported previously.6 7 36 Under hypo-osmotic conditions, the Cl– current, with properties otherwise similar to those of the ICl,b, increased 2-fold, as previously reported in cardiac myocytes.6 7 36 The recorded currents were largely abolished by IAA-94 (10 µmol/L) or NPPB (1 µmol/L), as expected if the basal and swell-activated conductances are Cl– currents (ICl,b and ICl,swell, respectively).
IAA-94 and NPPB have been used to selectively inhibit
Cl– channels in different cell
types.47 48 It is noteworthy that NPPB and IAA-94 applied
at concentrations much greater than those used in this study affect
sarcolemmal IKATP38 and
ICa.39 This is important, because
inhibition of these channels has been shown to block
IP.42 43 However, at the concentrations used in this
study, NPPB and IAA-94 had no effect on sarcolemmal
IKATP and ICa.
Furthermore, these blockers had no effect on
IK1, PKC activity, and the protection
produced by selective stimulation of mitochondrial
KATP channels with diazoxide (Figures 3, 4, and 9), which supports the notion that the abolition
of IP protection against necrosis was the result of
Cl– channel blockade.
The conclusions from the isolated cardiomyocyte studies
depend on correctly distinguishing viable from necrotic
cardiomyocytes. The trypan blue staining solution used to
assay cell death was hypo-osmotic (85 mOsm) to compensate for the lack
of mechanical forces on cardiomyocytes generated by
intercellular attachments in the heart that play a major role in
ischemia/reperfusion injury.35 44 Because an
osmotic gradient is used to test the fragility of the cell membrane
during the trypan blue viability assay, it could be argued that
cardiomyocytes exposed to Cl–
channel blockers might be more susceptible to hypo-osmotic stress.
This, being distinct from relative injury to the cell membrane during
the long SI, might therefore influence the results. However, in the
present studies, the cardiomyocytes were immediately
(<1 minute) fixed in glutaraldehyde and rendered
metabolically inactive by amylobarbitone at the same time
as exposure to trypan blue. In contrast, Cl–
channel activation by hypo-osmotic stress typically requires 5
minutes (Figure 2). This suggests that the
Cl– channel blockers are affecting events
occurring during the ischemia, and these results are not
artifacts of the cellular model. This is consistent with the
results in the isolated hearts, which mirrored the findings in isolated
ventricular myocytes.
Our inclusion of SR after the long SI in the isolated
cardiomyocyte model is novel. In the absence of IP, the
extent of cardiomyocyte death seen after 120-minute SR
after 45-minute SI was similar to that observed in rabbit hearts on
reperfusion in vivo after 30-minute regional
ischemia.45 In the presence of IP,
cardiomyocyte death during SR as assessed by trypan blue
staining was significantly reduced compared with control (Figures 6, 7B, and 8B). Rounded cardiomyocytes
are generally not viable, and the IP protection corresponded to a
significant reduction in the proportion of round cells on SR.
Interestingly, IP also promoted the recovery during SR from
square-shaped myocytes after 45-minute SI to rod-shaped myocytes,
resulting in a significantly higher proportion of rod-shaped myocytes
after 120-minute SR. This was not observed either under control
conditions or when IP was blocked by Cl– channel
inhibition.
In the present study cardiomyocyte swelling, which occurs during ischemia, activates PKC, whereas IP is abolished by PKC inhibition. Consistent with this, the specific PKC inhibitor chelerythrine inhibited ICl,swell in rabbit ventricular myocytes, as in previous reports in noncardiac tissue,46 47 but in contrast to rabbit atrial myocytes.48 Differences in PKC regulation of this current between atrial and ventricular myocytes may be accounted for by regional differences within the heart.
During ischemia, cardiomyocytes swell3 and respond by extruding K+, Cl–, organic solutes, and water, in an attempt to restore cell volume.49 Three major ion-transport mechanisms have been postulated to contribute to cell volume regulation, including cation-coupled cotransport (K+/Cl– or Na+/K+/2Cl–), K+/H+ exchange coupled to Cl–/HCO3- exchange, and both K+ and Cl– diffusion pathways.4 50 51 Furthermore, given that activation of sarcolemmal IKATP results in K+ movement, this channel might also play a role in volume regulation of the myocyte and mitochondria. Swell-activated and other Cl– channels are known to play an important role in cell volume regulation.4 52 However, if volume regulation depends on Cl– channels alone, it would be anticipated that ischemic and/or reperfusion-induced necrosis would be greater in tissue treated with Cl– channel inhibitors compared with nontreated tissue. In this study, this was not the case either in isolated cardiomyocytes or isolated hearts. This does not necessarily mean that Cl– channels are not involved in cell volume regulation. Rather, it would be advantageous for cells to have several volume-regulatory mechanisms working in concert or independently.
In the rabbit ventricle, protein kinase
A–activated53 and
Ca2+-activated22 23
Cl– currents are known to exist. Furthermore, we
demonstrate the presence of ICl,swell.
Since available Cl– channel blockers are not
specific for any 1 type of Cl– channel, specific
Cl– channel involvement in IP cannot be
distinguished pharmacologically. However, we demonstrate that
hypo-osmotic stress activated
ICl,swell in rabbit ventricular
myocytes and resulted in protection from ischemic and
reperfusion injury (Figures 2 and 8). This implies that
ICl,swell, in particular, might play an
important role in IP. Given that cardiomyocyte swelling is
associated with ischemic injury,1 2 3
swell-activated Cl– channels are
involved in cell volume regulation,4 and both
swell-activated Cl– channels (Figure 4) and IP54 are linked to PKC activity, we
postulate that IP affects swell-activated
Cl– channels such that they are
activated earlier and/or to a greater extent during the long
ischemia. If this is correct, swell-activated
Cl– channels may act as an end effector of IP
through improved cell volume control during the long ischemia.
We believe that cell swelling may prove to be a useful unifying
perspective from which to examine IP.
In summary, the present study demonstrates the presence of IAA-94–sensitive and NPPB-sensitive swell-activated and basal Cl– currents, which are inhibited by PKC blockade, in ventricular myocytes. Inhibition of these Cl– currents completely abolishes IP protection against necrosis induced by ischemia and reperfusion in isolated rabbit ventricular myocytes and buffer-perfused rabbit hearts. Conversely, Cl– channel activation by hypo-osmotic stress mimics the protection of IP, and this protection is also abolished by Cl– channel blockade.
|
Acknowledgments |
|---|
This study was supported by Ontario Heart and Stroke Foundation Grant B-3536. Funds for equipment were received from the Alan Tiffin Trust. V.A.L. is supported by a Heart and Stroke Foundation of Ontario Studentship. P.H.B. is a Medical Research Council Scholar. G.J.W. is the recipient of a Career Investigator Award of the Heart and Stroke Foundation of Ontario. We thank Usha Thomas for laboratory assistance and Dr Amar K. Sen for advice.
|
Footnotes |
|---|
1 Both authors contributed equally to the study.
Received August 27, 1998; accepted January 11, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
- Jennings RB, Ganote CE. Structural changes in myocardium during acute ischemia. Circ Res. 1974;34(suppl III):III156–III172.
- Jennings RB, Reimer KA. The cell biology of acute myocardial ischemia. Annu Rev Med. 1991;42:225–246.[Medline] [Order article via Infotrieve]
- Jennings RB, Ganote CE, Kloner RA, Whalen DA, Hamilton DG. Explosive swelling of myocardial cells irreversibly injured by transient ischemia. In: Albrecht F, George R, eds. Pathophysiology and Morphology of Myocardial Cell Alteration. Vol 6. Baltimore, Md: University Park Press;1975:405–413.
- Vandenberg JI, Rees SA, Wright AR, Powell T. Cell swelling and ion transport pathways in cardiac myocytes. Cardiovasc Res. 1996;32:85–97.[Medline] [Order article via Infotrieve]
- Strange K, Jackson PS. Swelling-activated organic osmolyte efflux: a new role for anion channels. Kidney Int. 1995;48:994–1003.[Medline] [Order article via Infotrieve]
-
Duan D, Hume JR, Nattel S. Evidence that outwardly
rectifying Cl– channels underlie
volume-regulated Cl– currents in heart.
Circ Res. 1997;80:103–113.
[Abstract/Free Full Text] -
Sorota S. Swelling-induce chloride-sensitive current
in canine atrial cells revealed by whole-cell patch-clamp method.
Circ Res. 1992;70:679–687.
[Abstract/Free Full Text] -
Du X-Y, Sorota S. Cardiac swelling-induced chloride
current depolarizes canine atrial myocytes. Am J
Physiol. 1997;272:H1904–H1916.
[Abstract/Free Full Text] -
Zhang J, Rasmusson RJ, Hall SK, Lieberman M. A
chloride current associated with swelling of cultured chick heart
cells. J Physiol. 1993;472:801–820.
[Abstract/Free Full Text] -
Oz MC, Sorota S. Forskolin stimulates swelling-induced
chloride current, not cardiac cystic fibrosis transmembrane-conductance
regulator current, in human cardiac myocytes. Circ Res. 1995;76:1063–1070.
[Abstract/Free Full Text] - Voets T, Manolopoulos V, Eggermont J, Ellory C, Droogmans G, Nilius B. Regulation of a swelling-activated chloride current in bovine endothelium by protein tyrosine phosphorylation and G proteins. J Physiol. 1998;506(pt 2):341–352.
-
Crepel V, Panenka W, Kelly ME, MacVicar BA.
Mitogen-activated protein and tyrosine kinases in the
activation of astrocyte volume-activated chloride current.
J Neurosci. 1998;18:1196–1206.
[Abstract/Free Full Text] - Bescond J, Bois P, Petit-Jacques J, Lenfant J. Characterization of an angiotensin II-activated chloride current in rabbit sino-atrial cells. J Membrane Biol. 1994;140:153–161.[Medline] [Order article via Infotrieve]
-
Matsuura H, Ehara T. Activation of chloride current by
purinergic stimulation in guinea pig heart cells. Circ Res. 1992;70:851–855.
[Abstract/Free Full Text] - Diaz RJ, Wilson GJ. Selective blockade of AT1 angiotensin II receptors abolishes ischemic preconditioning in isolated rabbit hearts. J Mol Cell Cardiol. 1997;29:129–139.[Medline] [Order article via Infotrieve]
-
Liu GS, Thornton J, Van Winkle DM, Stanley AWH, Olsson
RA, Downey JM. Protection against infarction afforded by
preconditioning is mediated by A1 adenosine receptors in rabbit
heart. Circulation. 1991;84:350–356.
[Abstract/Free Full Text] -
Liu G, Richards S, Olsson R, Mullane K, Walsh R, Downey
J. Evidence that the adenosine A3 receptor may mediate the
protection afforded by preconditioning in the isolated rabbit heart.
Cardiovasc Res. 1994;28:1057–1061.
[Abstract/Free Full Text] -
Armstrong SC, Ganote CE. Preconditioning of isolated
rabbit cardiomyocytes: effects of glycolytic blockade,
phorbol esters, and ischaemia. Cardiovasc Res. 1994;28:1700–1706.
[Abstract/Free Full Text] -
Zhang K, Barrington PL, Martin RL, Ten Eick RE. Protein
kinase-dependent Cl– currents in feline
ventricular myocytes. Circ Res. 1994;75:133–143.
[Abstract/Free Full Text] - Shuba LM, Asai T, McDonald TF. Phorbol ester activation of chloride current in guinea-pig ventricular myocytes. Br J Pharmacol. 1996;117:1395–1404.[Medline] [Order article via Infotrieve]
-
Miyawaki H, Ashraf M. Ca2+ as
mediator of ischemic preconditioning. Circ Res. 1997;80:790–799.
[Abstract/Free Full Text] - Zygmunt AC, Gibbons WR. Calcium-activated chloride current in rabbit ventricular myocytes. Circ Res. 1991;68:424–437.