| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Laboratoire de Physiologie
Lyon-Nord, We sought to determine whether stretch-induced
preconditioning may be related to activation of adenosine receptors,
ATP-sensitive K+
(K+ATP) channels, and/or protein
kinase C (PKC) in the rabbit heart. Anesthetized rabbits underwent 30 min of coronary artery occlusion followed by 3 h of reperfusion.
Ischemic preconditioning was induced by one episode of 5 min of
ischemia followed by 5 min of reperfusion, and stretch
preconditioning was induced by a transient volume overload. The
abilities of gadolinium (Gd3+),
a blocker of stretch-activated channels, glibenclamide (Glib), a
blocker of K+ATP channels,
8-(p-sulfophenyl)-theophylline (8-SPT), a blocker of adenosine receptors, and polymyxin B (PMXB), an
antagonist of PKC, to prevent the infarct size-limiting effect of
stretch-induced preconditioning were evaluated. Because the infarct
size-reducing effect of stretch occurred in the absence of
ischemia and was prevented by previous administration of
Gd3+, Glib, 8-SPT, and PMXB, we
propose that activation of mechanosensitive ion channels protects the
rabbit heart from subsequent sustained ischemic insult, likely through
a mechanism that involves downstream activation of PKC, adenosine
receptors, and/or K+ATP channels.
stretch-activated channels; infarction; protein kinase C; adenosine
receptors; adenosine 5'-triphosphate-sensitive potassium
channels
ONE OR MORE BRIEF EPISODES of coronary artery occlusion
render the myocardium more resistant to a subsequent sustained ischemic insult (16). Numerous studies have shown that this protection, termed
ischemic preconditioning, can be mimicked by pharmacological stimulation of various plasma membrane receptors or ion channels (1, 5,
14). Moreover, one previous study (17) has suggested that, in canine
myocardium, this enhanced tolerance toward subsequent ischemia
may also be triggered by transient mechanical deformation of the heart.
Mechanosensitive ion channels are proposed to be involved in cell
volume regulation, electromechanical transduction, and regulation of
growth and development in various tissues and species, but their exact
role in the mammalian cardiovascular system is unknown (4, 6). During
ischemia, mechanosensitive channels located in the sarcolemma
of ventricular myocytes are likely activated because of dilatation and
passive elongation within the risk region. In addition, edema and cell
swelling caused by increased myocardial osmolality may stretch the
inside of the plasma membrane (25).
We previously demonstrated (17) that gadolinium
(Gd3+), a blocker of
stretch-activated ion channels, prevented both stretch- as well as
ischemia-induced protection in the dog heart.
However, the mechanism by which activation of these mechanosensors may protect the heart from a subsequent ischemic insult is a major unresolved issue. Activation of adenosine
A1 receptors and opening of
K+ATP channels have both been shown to
play a role in ischemic preconditioning (5, 14). Because gating of some
voltage- or ligand-activated receptors may be modulated by stretch and,
conversely, mechanogated channel activity may be regulated by
(nonmechanical) biochemical stimuli, we sought to determine whether
"cross talk" between these receptors and stretch-activated ion
channels might occur in the preconditioned heart (10, 27). In addition,
there is evidence that activation of protein kinase C (PKC) plays a key
role in signal transduction in the preconditioned rabbit heart (31). We
therefore investigated whether protection afforded by mechanical
deformation of the heart might be mediated by PKC. We employed the in
vivo rabbit model to determine whether
8-(p-sulfophenyl)-theophylline
(8-SPT), a blocker of adenosine receptors, glibenclamide (Glib), a
blocker of K+ATP channels, and/or
polymyxin B (PMXB), a blocker of PKC, prevents infarct size reduction
due to transient myocardial stretch.
All experiments performed in this study conform to the "Guiding
Principles in the Care and Use of Animals" approved by The American
Physiological Society.
Surgical Preparation
ABSTRACT
Top
Abstract
Introduction
Methods
Results
Discussion
References
INTRODUCTION
Top
Abstract
Introduction
Methods
Results
Discussion
References
METHODS
Top
Abstract
Introduction
Methods
Results
Discussion
References
The rabbits were ventilated with room air through a tracheotomy tube, and tidal volume and rate were adjusted to provide physiological pH and blood gases. Fluid-filled catheters were inserted into the right jugular vein (for administration of drugs and fluids) and the left carotid artery (for measurement of blood pressure). Body temperature was monitored via a rectal thermometer and kept constant by means of a heating pad.
After an intravenous injection of 10 µg/kg of fentanyl, the chest was opened through the fourth left intercostal space. The pericardium was incised and the heart exposed. The major anterolateral branch of the left circumflex artery was encircled with a 4-0 silk suture. The ends of the suture were threaded through a piece of tubing, forming a snare that could be tightened to occlude the artery. Electrocardiographic limb leads and arterial pressure were monitored continuously throughout the experiment on a Gould recorder (Cleveland, OH). The animals were allowed 15 min after these surgical procedures to stabilize.
Protocol I: Confirmation of Stretch-Induced Protection in Rabbit Heart
We previously demonstrated that stretch preconditions the canine heart, likely through activation of mechanosensitive ion channels (17). Our first objective was to establish a rabbit model of stretch-induced preconditioning and document that 1) acute volume overload can stretch the rabbit myocardium, and 2) this mechanical stress results in significant infarct size limitation following a subsequent prolonged coronary artery occlusion. Because acute volume overload may, in theory, result in hypoxia/ischemia due to a decrease in myocardial blood flow, increased left ventricular (LV) end-diastolic pressure, increased workload, or simple hemodilution, we also sought to confirm that protection due to volume loading 3) did not impair myocardial perfusion or result in lactate release and 4) could not be attributed to hemodilution, but 5) was a consequence of activation of mechanosensitive ion channels.Protocol I(A): Stretch-Induced Protection
All animals underwent 30 min of sustained coronary artery occlusion followed by 3 h of reperfusion. Before the occlusion was sustained, the first 65 rabbits enrolled in the study were assigned to a 10-min "treatment" period (Fig. 1) consisting of 1) no intervention (control group), 2) 5 min of preconditioning ischemia and 5 min of reperfusion (PC), 3) saline volume overload (SVO), or 4) blood volume overload (BVO).
|
SVO was induced by infusing 40 ml/kg of isotonic saline warmed to 38°C via the jugular vein over 5 min. This volume and rate of infusion were calculated on the basis of our previous work (17) in the dog and pilot experiments in the rabbit model showing a resulting 10-15% regional LV dilatation and a significant reduction of infarct size. At the end of this 5-min intervention, saline infusion was stopped and followed by 5 min without any intervention.
BVO was induced by mixing 40 ml/kg of blood, freshly obtained from the carotid artery of support rabbits, with 1,000 IU heparin and administering it over 5 min into the jugular vein of the recipients. As with the SVO group, BVO was followed by a 5-min period without any intervention before the sustained occlusion.
So that all animals would receive a similar total volume of exogenous liquid (saline), both the control and PC groups received 40 ml/kg of saline, administered as an intravenous perfusion starting at the onset of the treatment period and ending at the completion of the experiment.
Quantification of myocardial stretch. The amount of regional myocardial stretch produced by either 5 min of ischemia and 5 min of reperfusion or by volume overload was assessed in an additional group of 17 (4 control, 3 PC, and 10 SVO) rabbits by means of sonomicrometry.
Detection of hypoxia or ischemia during stretch. The effect of acute volume overload on regional myocardial blood flow was investigated in an additional group of 16 (3 control, 3 PC, and 10 volume-overloaded) rabbits using the radioactive microsphere technique. Radioactive microspheres labeled with either 141Ce or 103Ru (NEN, Boston, MA) were injected at baseline (n = 8), during acute volume overload (n = 7), during the 5-min preconditioning ischemia (n = 3), or 10 min into the 30-min coronary artery occlusion (n = 9), as previously described (17).
Electrocardiographic changes during the treatment period were analyzed in all SVO and BVO rabbits to detect any S-T segment shift indicative of myocardial ischemia during acute volume overload. As further confirmation, in two cohorts of rabbits not used for infarct size determination we withdrew blood from the coronary vein for measurement of lactate concentration. Animals underwent either a 5-min SVO or 5 min of ischemia (n = 8 in each group), and measurements were made at baseline and 5 min into ischemia or volume overload.Influence of hemodilution during stretch. Hematocrit, used as the index of hemodilution, was measured at baseline, just before the 30-min coronary occlusion, at the end of sustained ischemia, and after 3 h of reflow for all animals in protocol I(A).
Protocol I(B): Role of Stretch-Activated Channels
To determine whether Gd3+, a blocker of stretch-activated ion channels, attenuates infarct size limitation due to SVO and/or PC, 38 rabbits, all subjected to a 10-min treatment period [as described in Protocol I(A): Stretch-Induced Protection] followed by 30 min of sustained ischemia and 3 h of reperfusion, were assigned to three groups: 1) Gd3+ + control, 2) Gd3+ + PC, or 3) Gd3+ + SVO. In each group, 40 µmol/kg Gd3+ dissolved in 3 ml saline was administered 1 min before the onset of PC ischemia, volume overload, or the 10-min waiting period (Fig. 1).Protocol II: Role of K+ATP Channels
The aim of protocol II was to determine whether activation of ATP-sensitive ion channels may be involved in the protection afforded by stretch. After the surgical procedures described in Surgical Preparation were completed, 32 rabbits were subjected to either PC, ischemia, SVO, or a 10-min waiting period. All rabbits received an intravenous bolus of Glib (0.3 mg/kg) immediately before the 10-min treatment phase (Fig. 1).Protocol III: Role of Adenosine Receptors
The objective of protocol III was to assess whether adenosine receptors may play a role in the stretch-induced protection. The design of protocol III was comparable to that of protocol II, except that control, PC, and SVO groups (total n = 39) received 7.5 mg/kg of 8-SPT in lieu of Glib (Fig. 1).Protocol IV: Role of PKC
We sought to determine whether PKC might be involved in stretch-induced preconditioning. Control, PC, and SVO groups (total n = 22) received PMXB. The intravenous infusion of PMXB (24 mg/kg) was started 1 min before the treatment period and stopped just before the sustained coronary occlusion (Fig. 1).At the end of the 10-min treatment period, rabbits in all protocols underwent 30 min of coronary artery occlusion followed by 3 h of reperfusion (Fig. 1).
Chemicals
Gadolinium(III) chloride hexahydrate (99.999%; Aldrich, Milwaukee, WI) was dissolved in saline (0.9%). 8-SPT (Research Biochemicals International) was dissolved in saline (0.9%), and pH was adjusted to 7.4 by adding 0.1 N NaOH. Glib (Sigma) was dissolved in a vehicle prepared on the day of the experiment; 10 ml vehicle contained 0.5 ml NaOH (1.0 N), 0.5 ml propylene glycol, 0.5 ml ethanol, and 8.5 ml saline (0.9%). PMXB sulfate (Sigma) was dissolved in saline (24 mg/kg in 3 ml of saline).Analysis
Hemodynamics (all protocols). Heart rate and arterial blood pressure were measured and averaged over three cardiac cycles in sinus rhythm for each sample period. Measurements of heart rate and arterial blood pressure were made at baseline (i.e., before treatment) and immediately before the sustained occlusion in all groups. In the volume-overload and PC groups, hemodynamics were measured during acute infusion and during the 5-min PC ischemia, respectively. In all groups, hemodynamics were then monitored throughout the sustained occlusion and at frequent intervals after reperfusion.
Measurement of regional myocardial blood flow (protocol I). For the cohort of rabbits in protocol I in which radiolabeled microspheres were injected, tissue samples were cut from the center of the ischemic and nonischemic zones, weighed, and counted with the reference blood samples in a gamma well counter. Blood flow was then computed using standard methods and expressed in milliliters per minute per gram (17).
Measurement of regional myocardial stretch (protocol I). For the group of rabbits in protocol I in which stretch was quantified, one pair of ultrasonic crystals used to assess regional LV dilatation was positioned in the center of the soon-to-be ischemic area as previously described (17). Crystals were inserted via small scalpel incisions into the subepicardium at a separation of 4-5 mm and oriented parallel to the minor axis of the heart. Maximal segment length (MSL; expressed as a percentage of baseline values), our index of stretch, was defined as the maximal separation between the two ultrasonic crystals, irrespective of the time point of the cardiac cycle (17). Under baseline conditions, MSL was equal to end-diastolic length. During ischemia or after reperfusion, however, MSL was larger than end-diastolic length, reflecting the greatest extent of bulging of the myocardial segment. Regional stretch was measured at baseline, during the treatment period, and during the subsequent 30-min coronary artery occlusion.
Measurement of blood lactate and hematocrit (protocol I). Blood lactate was measured by means of a colorimetric technique (Lactate PAP, BioMérieux, Lyon, France). Hematocrit was measured by means of a centrifuge technique. These measurements were performed at baseline, just before the sustained ischemia, at the end of the sustained ischemia, and after 3 h of reperfusion.
Area at risk and infarct size determination (all protocols). At the end of each protocol, after 3 h of reperfusion, the coronary artery was reoccluded and 0.5 mg/kg Unisperse blue pigment (Ciba-Geigy, Hawthorne, NY) was injected intravenously to delineate the in vivo area at risk, as previously described (17). With this technique, the previously nonischemic myocardium appears blue, whereas the previously ischemic myocardium (area at risk) remains unstained. The deeply anesthetized rabbits were then killed by an intravenous injection of 4 meq KCl. The heart was excised and cut into six or seven transverse slices, parallel to the atrioventricular groove. After right ventricular tissue had been removed, the heart slices were weighed. The basal surface of each slice was photographed for later measurement of the area at risk. Each slice was then incubated for 15 min in a 1% solution of triphenyltetrazolium chloride at 37°C. This method reliably identifies necrotic myocardium, which appears pale, from viable myocardium, which stains brick red (28). The slices were then rephotographed. Enlarged projections of these slices were traced for determination of the boundaries of the area at risk and area of necrosis. Extent of the area at risk and area of necrosis was quantified by computerized planimetry and corrected for the weight of the tissue slices. Total weights of the area at risk and the area of necrosis were then calculated and expressed as percentages of total LV weight.
Exclusion criteria. We decided prospectively that hearts with a risk region of <10% of the LV weight and volume-overloaded rabbits that exhibited S-T segment shift on electrocardiograms during the treatment period would be excluded from the study.
Statistics
Because protocols I-IV were performed consecutively and not concurrently, separate statistical comparisons were performed for each of the four components of the study. Comparison of infarct size, hemodynamics, regional LV dilatation, regional myocardial blood flow, blood lactate, and hematocrit among groups was performed by analysis of variance and post hoc Tukey's test. Bonferroni correction was applied when multiple comparisons were performed. Differences in the relationship between infarct size and area at risk (both expressed as percentages of the LV weight) were evaluated by analysis of covariance (ANCOVA) and post hoc Tukey's test, with infarct size as the dependent variable and area at risk as the covariant. Measurements are expressed as means ± SE, and P values <0.05 were considered statistically significant.| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
Protocol I
Mortality and exclusions. Among the 103 rabbits that entered the infarct size arm of protocol I, 15 rabbits developed ventricular fibrillation and 13 died from heart failure, defined here as a progressive decrease of blood pressure to a systolic value <50 mmHg with obvious global LV dilatation and inefficient cardiac contraction. These latter events, manifest either during the 30-min ischemia or the following reperfusion, were likely related to increased workload secondary to volume overload. Among the 33 rabbits that underwent saline overload, 2 rabbits displayed an upward S-T segment shift during the treatment period and, according to our criteria, were excluded from the study. Four rabbits were excluded because they had an area at risk of <10% of the LV weight. Two animals in the BVO group died during blood infusion, one from an unexplained rapid drop in blood pressure and the other because of accidental air injection. Overall, infarct size data are presented for 42 untreated (8 control, 9 PC, 13 SVO, and 12 BVO) and 25 Gd3+-treated (8 control, 10 PC, and 7 SVO) rabbits (Table 1).
|
Characteristics of volume overload. REGIONAL LV DILATATION. Regional LV dilatation was measured in 17 (4 control, 3 PC, and 10 SVO) rabbits. Acute SVO resulted in a significant regional elongation: MSL increased to 112 ± 2% of baseline in the SVO group (P < 0.05 vs. baseline and control group) (Fig. 2). PC hearts exhibited comparable regional dilatation during the brief episode of coronary artery occlusion: at that time, mean MSL increased to 110 ± 2% [P < 0.05 vs. baseline and control group and P = not significant (NS) vs. the SVO group]. At the end of the treatment period, MSL remained significantly increased compared with baseline values in both PC and SVO groups. During the 30-min sustained occlusion, control, PC, and SVO hearts displayed similar significant increases in MSL that averaged from 110 to 114% of baseline values (P < 0.05 vs. baseline and P = NS among groups) (Fig. 2).
|
|
1 · g1,
respectively (P < 0.05 vs. baseline
and P = NS among
groups). In contrast, SVO as well as BVO increased
(rather than decreased) regional myocardial blood flow in both the
soon-to-be ischemic and nonischemic territories (Table
3).
|
|
|
Infarct size with volume overload. Area at risk was comparable among the four groups with mean values ranging from 27.0 ± 3.0 to 28.0 ± 2.0% of the LV weight (P = NS among groups; Table 4). As expected, PC significantly reduced infarct size to 13.1 ± 5.2% of the area at risk (P < 0.05 vs. 59.5 ± 7.5% in the control group). Interestingly, both SVO and BVO rabbits also developed significantly smaller infarcts than the control rabbits, averaging 20.6 ± 6.0 and 25.2 ± 5.9% of the area at risk, respectively (P < 0.05 vs. control group; Table 4).
|
|
Effect of blockade of stretch-activated channels by Gd3+ on infarct size limitation. Both mean area at risk and mean infarct size were comparable among the three Gd3+-treated groups (Table 4). That is, Gd3+ prevented infarct size limitation in both preconditioned and volume-overloaded hearts: infarct size averaged 46.2 ± 9.0, 45.1 ± 7.7, and 41.9 ± 7.5% of the area at risk in Gd3+ + control, Gd3+ + PC, and Gd3+ + SVO groups, respectively (P = NS). This finding was further confirmed when infarct size was plotted as a function of area at risk (Fig. 5C). Regression lines for all three groups did not differ by ANCOVA, indicating that Gd3+ prevented infarct size limitation in PC and SVO hearts irrespective of the size of the risk region.
Protocol II
Mortality and exclusions. Of the 32 rabbits that entered this component of the study, 4 were excluded because they developed ventricular fibrillation and 3 because of heart failure (Table 1). Thus data are presented for 25 Glib-treated (8 control, 8 PC, and 9 SVO) rabbits.
Hemodynamics. As in protocol I, SVO resulted in a significant but transient increase in blood pressure and a slight but not significant increase in heart rate (Table 2).
Infarct size. Area at risk was comparable among the three groups, with mean values ranging from 28.6 ± 3.1 to 30.3 ± 2.8% of the LV weight (P = NS among groups) (Table 4). Mean infarct size (expressed as a percentage of the area at risk) was 52.5 ± 6.9% in the Glib + control group, 54.2 ± 8.8% in the Glib + PC group, and 49.3 ± 7.5% in the Glib + SVO group (P = NS), and the regression relationship between infarct size and risk region did not differ by ANCOVA (Table 4 and Fig. 5D). That is, Glib prevented infarct size limitation by ischemic preconditioning or volume overload.
Protocol III
Mortality and exclusions. Of the 39 rabbits enrolled in this part of the study, 5 were excluded because they developed ventricular fibrillation, 1 because of heart failure, and 4 because of an area at risk of <10% of the LV weight. Data are thus presented for 29 8-SPT-treated (9 control, 10 PC, and 10 volume-overloaded) rabbits (Table 1).
Hemodynamics. As in protocols I and II, SVO resulted in a significant but transient increase in blood pressure and a slight but not significant increase in heart rate (Table 2).
Infarct size. Area at risk was comparable among the three groups, averaging 28.5 ± 2.7, 22.0 ± 2.4, and 25.9 ± 2.8% of the LV weight in the 8-SPT-treated control, PC, and volume-overloaded groups, respectively (P = NS among groups) (Table 4). Treatment with 8-SPT prevented infarct size reduction by either PC or SVO; this was documented both by comparison of mean values of infarct size (60.4 ± 4.9, 43.3 ± 7.1, and 49.1 ± 9.1% of the area at risk in SPT + control, SPT + PC, and SPT + SVO groups; Table 4) and ANCOVA (Fig. 5E).
Protocol IV
Mortality and exclusions. Of the 32 rabbits that entered protocol IV, 5 were excluded because they developed ventricular fibrillation and 5 because of heart failure. Data are thus presented for 22 PMXB-treated (5 control, 10 PC, and 7 volume-overloaded) rabbits (Table 1).
Hemodynamics. As in protocols I to III, SVO resulted in a significant but transient increase in blood pressure and a slight but not significant increase in heart rate (Table 2).
Infarct size. Area at risk was similar among groups, averaging 22.1 ± 2.9, 27.9 ± 2.7, and 24.7 ± 3.2% of the LV weight in the control, PC, and volume-overloaded groups, respectively (P = NS among groups) (Table 4). PMXB prevented infarct size limitation in both the PC (59.2 ± 7.2%) and the SVO (55.0 ± 7.1%) groups (P = NS vs. 54.7 ± 7.7% in the control group) (Table 4). This was further confirmed by ANCOVA (Fig. 5F).
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
In the present study, we have demonstrated that myocardial stretch per se (i.e., in the absence of hypoxia or ischemia) protects the rabbit heart from subsequent sustained coronary artery occlusion. We have further determined that this cardioprotection is mediated by cardiac receptors sensitive to mechanical stress, which may interact with K+ATP channels and/or adenosine receptors, via activation of PKC.
Myocardial Protection Mediated by Stretch
No evidence of confounding hypoxia/ischemia. We previously demonstrated (17) that myocardial stretch, induced by an acute infusion of saline, can precondition the canine heart through activation of stretch-activated ion channels. The first objective of the present study was to confirm that stretch can also protect the ischemic rabbit heart. Clearly, acute volume overload significantly dilated the rabbit myocardium: regional elongation averaged 110-115% of the initial segment length, which closely resembles the amount of stretch induced by ischemia in the preconditioned hearts. Hearts that underwent this mechanical stress during the treatment period developed significantly smaller infarcts as a consequence of a subsequent 30-min coronary artery occlusion than did hearts in controls. It is worth noting that the amount of infarct size reduction afforded by stretch was comparable to that observed in the ischemia-preconditioned hearts. This protective effect was likely not a consequence of the modest and transient increase in arterial pressure during volume loading. Area at risk, the major determinant of infarct size in the rabbit model, was comparable among groups and likely did not play a role in the protective effect of stretch, as suggested by ANCOVA performed in each of the four infarct size protocols of the study. Collateral flow, which may be a confounding factor when measuring infarct size, cannot be used to explain the observed protection because it is well established that the rabbit heart lacks coronary collaterals, as seen in the six animals in which we measured regional myocardial blood flow during ischemia (14).
Obviously, a major concern is whether the apparent stretch-induced protection was simply due to a confounding hypoxia/ischemia caused by acute volume overload. After rapid volume overload, either hemodilution, increased LV end-diastolic pressure, or increased workload might produce an imbalance between O2 supply and demand and thereby induce preconditioning. To address this issue, we endeavored to detect any sign of ischemia or hypoxia during the treatment period in volume-overloaded hearts. We excluded 2 of the 18 hearts that underwent SVO in protocol I because they exhibited S-T segment upward shift during the treatment period. Regional myocardial blood flow measurements revealed that volume loading of the heart increased (rather than decreased) myocardial perfusion. However, in the present study, we measured transmural rather than subendocardial and subepicardial blood flow and therefore cannot exclude a redistribution of flow resulting in subendocardial ischemia. Yet, in a previous investigation using the canine model (17), we observed that volume overload resulted in an increase in both subendocardial and subepicardial blood flow with no redistribution of flow. Coronary venous lactate concentration, measured directly in the blood draining the area at risk, increased after regional ischemia in control hearts but was unaffected by volume overload, suggesting that stretch did not cause myocardial hypoxia/ischemia. One might argue that hemodilution might have artificially prevented an increase in coronary venous lactate in the volume overloaded group. However, hemodilution in the SVO group was moderate, as reflected by the 10% decrease in hematocrit, and likely cannot fully account for the 35% reduction in coronary venous lactate displayed by the SVO group compared with the control group at 5 min of ischemia. Finally, infarct size was measured in a subset of rabbits in which volume overload was performed using blood instead of isotonic saline to prevent hemodilution-induced hypoxia. BVO did not decrease the hematocrit, yet it reduced infarct size to a similar extent than did SVO, strongly suggesting that hemodilution was an unlikely confounding factor in this model. Overall, we found no evidence of myocardial ischemia or hypoxia during acute volume overload. Rather, the first component of this study demonstrates that mechanical stretch per se protects (or preconditions) the rabbit heart from subsequent sustained ischemic insult. It further indicates that acute SVO is a valid model for studying the mechanism of stretch-induced cardioprotection.Role of mechanosensitive ion channels.
Stretch-activated ion channels, studied in large part using patch-clamp
preparations, have been identified in a variety of tissues and species,
including ventricular myocytes (3, 21). The class of mechanogated
channels, like voltage-gated or ligand-gated channels, displays
different ion selectivities (e.g., cation, anion,
K+, and nonselective) and responds
to membrane stress by changes in open probability without significant
changes in either single-channel conductance or ion selectivity (6).
Their role in the cardiovascular system remains unclear, although it
has been suggested that they may be involved in cell volume regulation
and growth (4). Gd3+ is the most
commonly used blocker of mechanogated channels and is often used as a
pharmacological tool for testing the putative role of mechanogated
channels in various physiological processes. In the present study, we
extend our previous observation in the dog by showing that
Gd3+ prevented infarct size
limitation in both ischemic and stretch-activated hearts in the rabbit
model. In the present study, rabbits received ~100 µM of
Gd3+ intravenously; although the
myocardial concentration of Gd3+
was not measured, it was likely sufficient to block mechanogated channels. Takano and Glantz (22), using an in vivo canine model, showed
that an intravenous injection of 76 µM/kg of
Gd3+ was sufficient to reduce the
upward shift of the LV diastolic pressure-volume relation during
pacing-induced ischemia, probably by blocking stretch-activated
ion channels. One might argue that Gd3+ is not highly specific for
stretch-activated ion channels, because in vitro studies suggest that
it may also, at low concentrations, block voltage-dependent
Ca2+ channels in some experimental
preparations (2). One cannot exclude the possibility that the dose used
in the present study blocked these
Ca2+ channels. However,
Suleymanian et al. (21) recently demonstrated, using isolated rabbit
ventricular myocytes, that cell volume regulation by
Gd3+-sensitive ion channels is not
blocked by the Ca2+ antagonist
verapamil and persists in the absence of
Ca2+ in the bathing medium. Hansen
et al. (8), using an in vitro canine heart preparation, found that
stretch-induced arrhythmias could be prevented by
Gd3+ but not by the
Ca2+ antagonists verapamil or
nifedipine. In addition, Wallbridge et al. (29) recently reported that
the Ca2+ antagonist nisoldipine
fails to prevent ischemic preconditioning in the in vivo pig heart.
Thus, although the specific site of action of
Gd3+ remains unknown, results from
protocol I strongly implicate
stretch-activated ion channels as playing an important role in
preconditioning. We cannot exclude the possibility that cell swelling
during ischemia-reperfusion affects other osmotically modulated
channels and transporters in cardiac cells, including
Cl channels,
delayed-rectifier K+ channels,
Na+-K+-ATPase,
and the
Na+/Ca2+
exchanger (26). However, their possible activation or inactivation during cell swelling is not known to be affected by
Gd3+ and likely does not not
explain the present results. Because mechanosensitive ion channels have
been identified in neurons, endothelial cells, and smooth muscle cells,
the possibility cannot be excluded that these nonmyocyte cell types
were also involved in the protection. Volume overload might have also
activated pathways that do not directly depend on mechanosensitive ion
channels, similar to neural reflexes, resulting in the induction of
classic preconditioning. The type of ion that penetrated the
cardiomyocyte via the
Gd3+-sensitive channels activated
by volume overload was not investigated in the present study. One may
hypothesize that these channels were permeable to
Ca2+, which could have activated
different intracellular enzymes, including phospholipases
and/or PKC, and thereby influenced the cardioprotection.
Stretch-Related Protection and K+ATP Channels
There is evidence that activation of K+ATP channels plays a role in the protection afforded by ischemic preconditioning in the dog and pig hearts (5, 19). However, the role of K+ATP channels in ischemic preconditioning in the rabbit heart is still debated. Thornton et al. (23) failed to prevent ischemic preconditioning in pentobarbital-anesthetized rabbits. In contrast, Toombs et al. (24) were able to abolish the infarct size-limiting effect of ischemic preconditioning in ketamine/xylazine-anesthetized rabbits. Although the reason for this discrepancy is not clear, and different types of anesthesia have been implicated, the protective role of the activation of K+ATP channels has been demonstrated in in vitro models of preconditioning using rabbit myocytes not exposed to anesthetic agents (9). Our results are in agreement with those of Toombs et al. (24), because we observed that Glib could prevent ischemic preconditioning in the in situ rabbit heart.The first novel observation of the present study is that Glib also prevented protection afforded by stretch. Glib likely did not block stretch-activated channels, given that Suleymanian et al. (21) recently reported that stretch-activated channels are insensitive to Glib in isolated rabbit ventricular myocytes (21).
One possibility is that mechanical deformation of the myocyte might directly activate K+ATP channels. In such a direct mechanism, mechanical energy is directly coupled to the mechanosensitive channel protein without the intervention of biochemical reactions, although energy may be focused onto the channel via cytoskeletal and/or extracellular elements. This hypothesis is supported by a study by Van Wagoner (27), who demonstrated in isolated atrial myocytes that stretch can directly activate K+ATP channels in the absence of any variation in the ATP content. However, direct mechanical activation of K+ATP channels by stretch cannot fully account for the protection observed in overloaded hearts in the present study because it was prevented by Gd3+, an agent that has no known effect on K+ATP channels. Alternatively, one can hypothesize that mechanosensitive protection would be indirect, with intervening biochemical steps between the initial mechanical event (activation of stretch-activated channels) and an end effector (K+ATP channels).
Stretch-Related Protection and Adenosine Receptors
In the present study, we make the second novel observation that 8-SPT, a blocker of adenosine receptors and well-established inhibitor of ischemic preconditioning, also prevents stretch-induced protection. 8-SPT is a nonspecific blocker for adenosine receptors, with an inhibition constant (Ki) of 4.5 and 6.3 µM for A1 and A2 receptors, respectively (14). Therefore, although many studies support a role for A1- rather than A2-adenosine receptors in ischemic preconditioning, we cannot affirm which receptor subtype is involved in stretch-induced protection.How can one explain that 8-SPT blocks stretch-induced protection? Although it has been demonstrated that the gating of some ligand-activated receptors may be directly modulated by stretch, whether this applies to adenosine receptors is currently unknown (7, 30). Whittaker et al. (30) recently demonstrated in the rat heart that intramyocardial injections of isotonic saline, causing local mechanical stress of ventricular myocytes, reduced infarct size following a subsequent 60-min coronary artery occlusion, a protection that could be prevented by addition of Gd3+. Interestingly, intramyocardial injections of adenosine reduced infarct size to an extent similar to that resulting from saline, indirectly suggesting that stretch and adenosine may share a common mechanism to protect the ischemic heart (29). However, whether the protection afforded by intramyocardial injections of adenosine could be blocked by Gd3+ was not investigated in the latter study.
One may first hypothesize that activation of
A1-adenosine receptors is an
initial event that modulates gating of
Gd3+-sensitive stretch-activated
channels, because it is well established that adenosine receptors can
be coupled to ion channels. Schwiebert et al. (20) have demonstrated
that stimulation of the
A1-adenosine receptors by
N6-cyclohexyladenosine
activates a cellular pathway involving an A1-adenosine receptor,
phospholipase C, diacylglycerol, PKC, and a G protein, which regulates
a Cl channel known to be
activated elsewhere by cell swelling and membrane stretch. This latter
scheme would, however, apply only to ischemia-induced
preconditioning and probably not to stretch-induced protection in which
there is no evidence of hypoxia/ischemia and no apparent reason
for adenosine to be released before mechanogated sensors are
stimulated. Rather, our results suggest that in the stretch-protected
heart, activation of mechanosensitive ion channels occurs upstream with
respect to activation of adenosine receptors.
Stretch-Related Protection and PKC
Previous in vitro experiments indicate that stretch can activate protein kinase C; Komuro et al. (12) demonstrated in a primary culture of rat ventricular myocytes that stretch activates both phospholipases C and D and increases the production of diacylglycerol, the natural activator of PKC. This is the first report that stretch-induced protection is mediated by PKC in the in situ rabbit heart.One may speculate on the link between stretch-activated ion channels, PKC, and K+ATP channels or adenosine receptors. Regarding K+ATP channels, Hu et al. (9) and Light et al. (13) recently reported that PKC may regulate such channels in rabbit ventricular myocytes, whereas Liu et al. (15) showed that PKC activation increases ATP-sensitive K+ currents induced by metabolic inhibition or pinacidil, both thought to be able to trigger preconditioning. Together with these results, our present data indirectly support sequential activation of mechanosensitive ion channels, PKC, and K+ATP channels. With respect to adenosine receptors, Kitakaze et al. (11) reported that PKC may stimulate ectosolic 5'-nucleotidase that transforms adenosine 3',5'-cyclic monophosphate into adenosine. Extrapolation of these data suggests that stretch might activate PKC and thereby enhance adenosine production, further leading to endogenous protection. However, further work is needed to clarify this signal transduction pathway in stretch-induced preconditioning and the link between stretch-activated ion channels, PKC, K+ATP channels, and adenosine receptors.
In conclusion, the present study demonstrates that stretch per se can trigger preconditioning in the rabbit heart through activation of Gd3+-sensitive mechanogated ion channels. The results further suggest that stretch- and ischemia-induced protection may share a common mechanism, including activation of PKC, K+ATP channels, and/or adenosine receptors.
| |
ACKNOWLEDGEMENTS |
|---|
We gratefully thank Prof. Karin Przyklenk for carefully reading this manuscript.
| |
FOOTNOTES |
|---|
Anne Gysembergh was a recipient of the Allocation Jeune Chercheur 1996 from the Hospices Civils de Lyon (Lyon, France).
Address for reprint requests: M. Ovize, Hôpital Cardiovasculaire et Pneumologique Louis Pradel, Unité 81, 59 Bd Pinel, 69394 Lyon Cedex 03, France.
Received 29 July 1997; accepted in final form 1 December 1997.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
1.
Banerjee, A.,
C. Locke-Winter,
K. B. Rogers,
M. B. Mitchell,
E. C. Brew,
C. B. Cairns,
D. D. Bensard,
and
A. H. Harken.
Preconditioning against myocardial dysfunction after ischemia and reperfusion by an 
1-adrenergic mechanism.
Circ. Res.
73:
656-670,
1993
2.
Biagi, B. A.,
and
J. J. Enyeart.
Gadolinium blocks low and high threshold calcium currents in pituitary cells.
Am. J. Physiol.
259 (Cell Physiol. 28):
C515-C520,
1990
3.
Christensen, O.
Mediation of cell volume regulation by Ca++ influx through stretch-activated channels.
Nature
330:
66-68,
1987[Medline].
4.
Craelius, W.,
V. Chen,
and
N. El-Sherif.
Stretch-activated ion channels in ventricular myocytes.
Biosci. Rep.
8:
407-414,
1988[Medline].
5.
Gross, G. J.,
and
J. A. Auchampach.
Blockade of ATP-sensitive potassium channels prevents myocardial preconditioning in dogs.
Circ. Res.
70:
223-233,
1992
6.
Guharay, F.,
and
F. Sachs.
Stretch-activated single ion channel currents in tissue-cultured embryonic chick skeletal muscle.
J. Physiol. (Lond.)
352:
685-701,
1984
7.
Hamill, O. P.,
and
D. W. McBride, Jr.
The pharmacology of mechanogated membrane ion channels.
Pharmacol. Rev.
48:
231-252,
1996[Abstract].
8.
Hansen, D. E.,
M. Borganelli,
G. P. Stacy, Jr.,
and
L. K. Taylor.
Dose-dependent inhibition of stretch-induced arrhythmias by Gadolinium in isolated canine ventricles. Evidence for a unique mode of antiarrhythmic action.
Circ. Res.
69:
820-831,
1991
9.
Hu, K.,
D. Duan,
G. R. Li,
and
S. Nattel.
Protein kinase C activates ATP-sensitive K+ current in human and rabbit ventricular myocytes.
Circ. Res.
78:
492-498,
1996
10.
Kirber, M. T.,
R. W. Ordway,
L. H. Clapp,
J. V. Walsh, Jr.,
and
J. J. Singer.
Both membrane stretch and fatty acids directly activate large conductance Ca++-activated K+ channels in vascular smooth muscle cells.
FEBS Lett.
297:
24-28,
1992[Medline].
11.
Kitakaze, M.,
T. Minamino,
K. Node,
K. Komamura,
Y. Shinozaki,
M. Chujo,
H. Mori,
M. Inoue,
M. Hori,
and
T. Kamada.
Role of activation of ectosolic 5'-nucleotidase in the cardioprotection mediated by opening of K+ channels.
Am. J. Physiol.
270 (Heart Circ. Physiol. 39):
H1744-H1756,
1996
12.
Komuro, I.,
H. Katoh,
T. Kaida,
Y. Shibazaki,
M. Kurabayashi,
E. Hoh,
F. Takaku,
and
Y. Yazaki.
Mechanical loading stimulates cell hypertrophy and specific gene expression in cultured rat cardiac myocytes. Possible role for protein kinase C activation.
J. Biol. Chem.
266:
1265-1268,
1991
13.
Light, P. E.,
A. A. Sabir,
B. G. Allen,
M. P. Walsh,
and
R. J. French.
Protein kinase C-induced changes in the stoichiometry of ATP binding activate cardiac ATP-sensitive K+ channels.
Circ. Res.
79:
399-406,
1996
14.
Liu, G. S.,
J. Thornton,
D. M. Van Winkle,
A. W. H. Stanley,
R. A. Olsson,
and
J. M. Downey.
Protection against infarction afforded by preconditioning is mediated by A1 adenosine receptors in rabbit heart.
Circulation
84:
350-356,
1991
15.
Liu, Y.,
W. D. Gao,
B. O'Rourke,
and
E. Marban.
Synergistic modulation of ATP-sensitive K+ currents by protein kinase C and adenosine. Implications for ischemic preconditioning.
Circ. Res.
78:
443-454,
1996
16.
Murry, C. E.,
R. B. Jennings,
and
K. A. Reimer.
Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium.
Circulation
5:
1124-1136,
1986.
17.
Ovize, M.,
K. Przyklenk,
and
R. A. Kloner.
Stretch preconditions the canine myocardium.
Am. J. Physiol.
266 (Heart Circ. Physiol. 35):
H137-H146,
1994
18.
Sackin, H.
Stretch-activated ion channels.
Kidney Int.
48:
1134-1147,
1995[Medline].
19.
Schulz, R.,
J. Rose,
and
G. Heusch.
Involvement of activation of ATP-dependent potassium channels in ischemic preconditioning in swine.
Am. J. Physiol.
267 (Heart Circ. Physiol. 36):
H1341-H1352,
1994
20.
Schweibert, E. M.,
K. H. Karlson,
P. A. Friedman,
P. Dietl,
W. S. Spielman,
and
B. A. Stanton.
Adenosine regulates a chloride channel via protein kinase C and a G protein in a rabbit cortical collecting duct cell line.
J. Clin. Invest.
89:
834-841,
1992.
21.
Suleymanian, M. A.,
H. F. Clemo,
N. M. Cohen,
and
C. M. Baumgarten.
Stretch-activated channel blockers modulate cell volume in cardiac ventricular myocytes.
J. Mol. Cell. Cardiol.
27:
721-728,
1995[Medline].
22.
Takano, H.,
and
S. A. Glantz.
Gadolinium attenuates the upward shift of the left ventricular diastolic pressure-volume relation during pacing-induced ischemia in dogs.
Circulation
91:
1575-1587,
1995
23.
Thornton, J. D.,
C. S. Thornton,
D. L. Sterling,
and
J. M. Downey.
Blockade of ATP-sensitive potassium channels increases infarct size but does not prevent preconditioning in rabbit hearts.
Circ. Res.
72:
44-49,
1993
24.
Toombs, C. F.,
T. L. Moore,
and
R. L. Shebuski.
Limitation of infarct size in the rabbit by ischemic preconditioning is reversible with glibenclamide.
Cardiovasc. Res.
27:
617-622,
1993
25.
Tranum-Jensen, J.,
M. J. Janse,
J. W. T. Fiolet,
W. J. G. Krieger,
C. Naumann d'Alnoncourt,
and
D. Durrer.
Tissue osmolality, cell swelling, and reperfusion in acute regional myocardial ischemia in the isolated porcine heart.
Circ. Res.
49:
364-381,
1981
26.
Vandenberg, J.,
S. A. Rees,
A. R. Wright,
and
T. Powell.
Cell swelling and ion transport pathways in cardiac myocytes.
Cardiovasc. Res.
32:
85-97,
1996[Medline].
27.
Van Wagoner, D. R.
Mechanosensitive gating of atrial ATP-sensitive potassium channels.
Circ. Res.
72:
973-983,
1993
28.
Vivaldi, M. T.,
R. A. Kloner,
and
F. J. Schoen.
Triphenyltetrazolium staining of irreversible ischemic injury following coronary artery occlusion in rats.
Am. J. Pathol.
121:
522-530,
1985[Abstract].
29.
Wallbridge, D. R.,
R. Schulz,
C. Braun,
H. Post,
and
G. Heusch.
No attenuation of ischemic preconditioning by the calcium antagonist nisoldipine.
J. Mol. Cell. Cardiol.
28:
1801-1810,
1996[Medline].
30.
Whittaker, P.,
R. A. Kloner,
and
K. Przyklenk.
Intramyocardial injections and protection against myocardial ischemia. An attempt to examine the cardioprotective actions of adenosine.
Circulation
93:
2043-2051,
1996
31.
Ytrehus, K.,
Y. Liu,
and
J. Downey.
Preconditioning protects ischemic rabbit heart by protein kinase C activation.
Am. J. Physiol.
266 (Heart Circ. Physiol. 35):
H1145-H1152,
1994
This article has been cited by other articles:
|
|
 |
|
  S. M. Mosca Cardioprotective effects of stretch are mediated by activation of sarcolemmal, not mitochondrial, ATP-sensitive potassium channels Am J Physiol Heart Circ Physiol, August 1, 2007; 293(2): H1007 - H1012. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Argaud, A.-F. Prigent, L. Chalabreysse, J. Loufouat, M. Lagarde, and M. Ovize Ceramide in the antiapoptotic effect of ischemic preconditioning Am J Physiol Heart Circ Physiol, January 1, 2004; 286(1): H246 - H251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Chiari, V. Piriou, G. Hadour, C. Rodriguez, J. Loufouat, J.-J. Lehot, M. Ovize, and R. Ferrera Preservation of ischemia and isoflurane-induced preconditioning after brain death in rabbit hearts Am J Physiol Heart Circ Physiol, November 1, 2002; 283(5): H1769 - H1774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Harada, S. Sakamoto, Y. Niwa, and Y. Nakaya Involvement of adenosine in vascular contractile preconditioning Am J Physiol Heart Circ Physiol, June 1, 2001; 280(6): H2911 - H2919. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Vassort Adenosine 5'-Triphosphate: a P2-Purinergic Agonist in the Myocardium Physiol Rev, April 1, 2001; 81(2): 767 - 806. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||
P < 0.05 vs. control.
) group of protocol I(A).
A and
B: data points for PC (
), SVO
(
), and BVO (
) groups are shifted downward vs. control regression
line, indicating that they developed significantly smaller infarcts for
any size AR (P < 0.05 by analysis of
covariance). C-F: data points for
PC and SVO hearts that had been pretreated with
Gd3+, glibenclamide, 8-SPT, or
PMXB, respectively, lie close to control regression line, indicating
that these agents prevented infarct size reduction by ischemic PC or
SVO.