| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Department of Physiology, Faculty of Biology and Medicine, University of Lausanne, Lausanne; and 2Laboratory of Pharmaceutical Technology, School of Pharmacy Geneva-Lausanne, Geneva, Switzerland
Submitted 9 September 2004 ; accepted in final form 13 November 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
diazoxide; mitochondria; chick; hypoxia-reoxygenation; atrioventricular coupling
Despite recent advances in understanding of these mechanisms, little is known about the role that mitoKATP channel modulation could play in ROS production and functional recovery of the hypoxic-reoxygenated developing heart. Although the embryonic/fetal heart normally grows and functions in a relatively hypoxic environment, it rapidly reacts to oxygen deprivation and reoxygenation, e.g., as in the case of transient uteroplacental ischemia. We previously characterized the chrono-, dromo-, and inotropic responses of the isolated embryonic chick heart (4–5 days old) to anoxia-reoxygenation (30, 34). We also found that this preparation is subjected to an oxidative stress during reoxygenation (28) similar to reperfused cardiac cells obtained from 10-day-old chick embryo (41, 45). Interestingly, the postischemic protection of embryonic ventricular myocytes afforded by mitoKATP channel opening involves ROS and/or NO production (6, 19). On the other hand, we recently found that inducible NO synthase (NOS) is strongly expressed in the embryonic myocardium and generates NO during anoxia-reoxygenation (38), improving recovery of excitation-contraction (E-C) coupling in the ventricle (21). However, the underlying protective mechanism involving the mitoKATP channel remains to be explored in the whole developing heart.
In the present study, we asked whether pharmacological opening of the mitoKATP channel improves postanoxic recovery of pacemaking activity, cell-to-cell communication, myocardial contractility, and E-C coupling. Using a pharmacological approach, we also investigated the signaling pathways operating in the embryonic myocardium, with special emphasis on ROS-, NO-, and PKC-dependent pathways.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reagents
The fluorescence probe 2',7'-dichlorofluorescin diacetate (DCFH-DA), the mitochondrial complex III inhibitor myxothiazol (Myxo), the radical scavenger N-(2-mercaptopropionyl)glycine (MPG), the NOS inhibitor NG-nitro-L-arginine methyl ester (L-NAME), the mitoKATP channel opener diazoxide (Diazo), the mitoKATP blocker 5-hydroxydecanoate (5-HD), and the nonselective KATP channel blocker glibenclamide (Glib) were purchased from Sigma Aldrich (Buchs, Switzerland). The PKC inhibitor chelerythrine (Chel) was purchased from Calbiochem (JURO Supply, Lucerne, Switzerland).
In Vitro Mounting of the Heart
Fertilized eggs from Lohman Brown hens were incubated for 96 h at 38°C and 95% relative humidity to reach embryonic stage 24HH [according to Hamburger and Hamilton (13)]. The intact and spontaneously beating heart was carefully excised and placed in the culture compartment of an airtight chamber. The chamber was equipped with two windows for observation and measurements and maintained under controlled metabolic conditions on the thermostabilized stage (37°C) of an inverted microscope (model IMT2, Olympus, Tokyo, Japan) as previously described in detail (30, 38). Briefly, the culture compartment (300 µl) was separated from the gas compartment by a thin (15 µm) transparent and gas-permeable silicone membrane (RTV 141, Rhône-Poulenc, Lyon, France). The heart was slightly flattened by the silicone membrane, and the resulting thickness of the myocardial tissue facing the gas compartment was 
300 µm. Thus PO2 at the tissue level could be strictly controlled and rapidly modified (within <5 s) by flushing high-grade gas of selected composition through the gas compartment. At this developmental stage, the heart lacks vascularization and the myocardial oxygen requirement is met exclusively by diffusion.
The standard HCO3/CO2-buffered medium was composed of (in mmol/l) 99.25 NaCl, 0.3 NaH2PO4, 10 NaHCO3, 4 KCl, 0.79 MgCl2, 0.75 CaCl2, and 8 d-glucose. This culture medium was equilibrated in the chamber with 2.31% CO2 in air (normoxia and reoxygenation) or in N2 (anoxia), yielding pH 7.4.
MPG (1 mM), 5-HD (500 µM), L-NAME (50 µM), and Myxo (10 µM) were directly dissolved in this medium (control). Diazo (50 µM), Glib (1 µM), and Chel (5 µM) were first dissolved in DMSO (vehicle) and then diluted in the medium. The final DMSO concentration was 0.5%.
Measurement of Oxidant Stress
ROS generation was assessed using the nonfluorescent probe 2',7'-dichlorofluorescin (DCFH). The membrane-permeable diacetate form of DCFH (DCFH-DA) was dissolved in methanol at 40 mM (stock aliquots) and kept at –20°C. Before experimentation, the stock solution was diluted in culture medium to give a final concentration of 10 µM. Within the cell, esterases cleave the acetate groups of DCFH-DA, thus trapping DCFH intracellularly (23). DCFH is known to be preferentially oxidized by H2O2 or hydroxyl radical (OH·) but poorly by superoxide anion (O2–·) (12), yielding the fluorescent oxidized probe 2',7'-dichlorofluorescein (DCF).
Isolated hearts were loaded with DCFH-DA for 30 min at room temperature in the dark. Then the hearts were mounted in the culture chamber, and the medium was renewed with fresh DCFH-DA solution. The chamber was placed on the thermostabilized stage (37°C) of an epifluorescence microscope (Leitz) 15 min before the first fluorescence measurement. The fluorescence, expressed in arbitrary units, was measured on a restricted ventricular area (300 µm diameter) every 30 s throughout anoxia-reoxygenation with use of an excitation wavelength of 490 nm and an emission wavelength of 533 ± 3 nm. To avoid photobleaching of the fluorescent dye, illumination time was limited to 6 ms. The slope of the fluorescent signal vs. time represented the rate of production of oxyradicals and was expressed as arbitrary units per second.
To take into consideration interindividual variations in size and slight developmental differences of the hearts, end-diastolic wall area of the ventricle (mm2) was obtained from frames of video recordings of contracting heart obtained at the end of each experiment, as previously described (29). Moreover, the density of protein of the ventricular wall was obtained by dividing the protein content by the area and expressed as micrograms of protein per square millimeter.
Recording of Electrical and Mechanical Contractions
Electrical and contractile activities were recorded simultaneously and continuously throughout in vitro experiments as previously described (30, 34).
Electrical activity.
ECG activity of the spontaneously contracting hearts was recorded using two Ag/AgCl electrodes 1 mm apart (0.3 mm diameter) inserted into the window facing the culture compartment. Briefly, the atrial and ventricular regions were placed in the immediate vicinity of these electrodes, which were connected to a differential preamplifier (gain = 2,000), resulting in an output signal of 1 V from peak to peak. This signal was digitized and processed using an Apple Macintosh computer. The ECG displayed characteristic P, QRS, and T components, which allowed assessment of beating rate (beats/min) and atrioventricular (AV) conduction delay (PR interval). Additionally, the various types of arrhythmias (mainly sinoatrial arrest; brady- and tachycardia; 1st-, 2nd-, and 3rd-degree AV block; and Wenckebach phenomenon) and their duration could be precisely determined.
Contractile activity. Two adjustable phototransistors were positioned over the projected image of the sinoatrial pacemaking region and the apex of the ventricle, allowing detection of edge motion of the myocardial wall. The actual distance between investigated atrial and ventricular regions was the same in all groups: 2.3 ± 0.2 mm (n = 56). The mean propagation velocity of the wave of contraction was obtained by dividing this distance by the time between the peaks of the maximal shortening velocity in atrium and ventricle (i.e., mechanical AV delay).
The actual ventricular shortening at the apex (S, µm) was determined using video recordings obtained before anoxia and at the end of reoxygenation. The maximal shortening and relaxation velocities were obtained from the maximal positive and negative values of the first derivative of S (dS/dt, mm/s), respectively.
E-C coupling. The electromechanical delay (EMD, ms), reflecting the efficiency of E-C coupling, was determined at the level of the atrium (EMDa) and ventricle (EMDv) by measuring the delay between the electrical and mechanical events, i.e., the interval of time between the very initial phase of the P and QRS components and the initiation of contraction in the atrium and ventricle, respectively.
Anoxia-Reoxygenation Protocol
After 45 min of in vitro stabilization under normoxia at 37°C, the hearts were subjected to 30 min of strict anoxia followed by 60 min of reoxygenation. The pharmacological agents were present throughout anoxia-reoxygenation.
Protein and Glycogen Determination
At the end of each experiment, the hearts were carefully dissected into atria, ventricle, and conotruncus and stored at –20°C for subsequent determination of protein and glycogen content, as described previously (20).
Statistical Analysis
Values are means ± SD, unless otherwise indicated. The significance of any difference between the groups was assessed using Student's t-test or analysis of variance (ANOVA) with Tukey's post hoc test. The differences between the rates of functional recovery in the experimental groups were assessed by repeated-measures ANOVA during the first 20 min of reoxygenation. To take into consideration preanoxic variations between groups, all parameters are expressed as percent changes of their preanoxic value. Statistical significance was defined by P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Only spontaneously and regularly contracting hearts were used in this study. The preanoxic baseline parameters were not significantly altered by any of the agents used, either alone or in combination (Table 1).
|
Figure 1 shows that the time course of DCF fluorescence intensity measured in the ventricle essentially depended on the presence of oxygen. Preanoxic ROS production was not affected by any of the agents, alone or in combination (Fig. 2). On reoxygenation, the rate of ROS production was maximal after 10 ± 3 min in the control group (n = 6) and was reduced by 73 and 60% in the presence of Myxo and MPG, respectively (P < 0.02; Fig. 2).
|
|
Opening of the mitoKATP channel by Diazo doubled ROS production with respect to vehicle (P < 0.005). This Diazo-induced ROS production was abolished by 5-HD, MPG, or L-NAME but not by Chel. However, 5-HD, L-NAME, or Chel alone had no effect on ROS production (Fig. 2).
Recovery of AV Conduction and E-C Coupling Was Improved by Activation of the mitoKATP Channel
Although all parameters fully recovered at the end of reoxygenation, opening of the mitoKATP channel with Diazo significantly improved the rate of recovery of the PR interval and EMDv during the first 20 min of reoxygenation (Fig. 3). On the contrary, recovery of the other functional parameters was not affected by Diazo. 5-HD abolished the Diazo-induced protection of EMDv but not that of the PR interval. However, 5-HD alone did not have an effect on the rate of recovery of the functional parameters compared with the control group (Fig. 4). The nonselective KATP channel blocker Glib systematically induced atrial tachycardia during normoxia and reoxygenation associated with frequent second-degree AV blocks (not shown). The mechanical AV delay was 142 ± 22 ms (n = 56) under preanoxia and 227 ± 48 ms (n = 25) at the onset of reoxygenation and displayed the same time course as the electrical AV delay (PR interval), without regard to treatment. Furthermore, the mean propagation velocity of the wave of contraction between the atrium and ventricle passed from 16.6 ± 3.1 mm/s (n = 56) under preanoxia to 10.8 ± 2.5 mm/s (n = 25; P < 0.0001) at the onset of reoxygenation.
|
|
During reoxygenation, treatment with MPG alone decreased ROS production (Fig. 2) and increased EMDv (Fig. 4) but had no effect on other functional parameters. The Diazo-induced burst of ROS observed during reoxygenation was significantly lower in the Diazo + MPG group than in the Diazo group and comparable with that of vehicle group (Fig. 2). Moreover, scavenging ROS by MPG abolished the protective effect of Diazo on EMDv (Fig. 3). By contrast, heart rate, PR interval, EMDa, and ventricular shortening were not altered by the addition of MPG (Fig. 3).
During reoxygenation, Chel did not alter Diazo-induced ROS production (Fig. 2) but abolished the improvement in EMDv by the mitoKATP channel opener (Fig. 5). Furthermore, the Wenckebach phenomenon was observed during 52 ± 14 min in the Diazo + Chel group (n = 5) and during 19 ± 11 (n = 7), 19 ± 9 (n = 7), and 20 ± 4 min (n = 4) in the Diazo, vehicle, and Chel groups, respectively (P < 0.01, Diazo + Chel vs. others). Chel alone did not affect ROS production or functional parameters under steady normoxia or during reoxygenation (Figs. 2 and 5).
|
|
L-NAME alone did not affect ROS production throughout anoxia-reoxygenation (Fig. 2) but significantly delayed recovery of EMDv during reoxygenation (Fig. 4). Furthermore, L-NAME abolished the Diazo-induced ROS production (Fig. 2) and protection of AV conduction and EMDv during reoxygenation (Fig. 6). Contractile recovery was improved by L-NAME, alone and in combination with Diazo (Figs. 4 and 6).
Metabolic and Morphometric Parameters
The protein content of the whole heart, atria, ventricle, and conotruncus was 69.7 ± 14.3, 15.5 ± 5.4, 42.7 ± 9.1, and 11.1 ± 3.7 µg (n = 109), respectively, and did not vary from one group to another. Glycogen content was 1.2 ± 0.6, 0.7 ± 0.3, and 0.5 ± 0.3 nmol glucose unit/µg protein (n = 109) in atrium, ventricle, and conotruncus, respectively. In hearts on which fluorescence was measured, protein content, area, and density of protein of the ventricular wall were 44.7 ± 7.7 µg, 3.8 ± 0.8 mm2, and 11.9 ± 2.9 µg/mm2 (n = 45), respectively, with no statistically significant differences between the groups.
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Limitations of the Experimental Model
Detection of oxidant stress by DCFH fluorescence. Under our experimental conditions, the fluorescence measured in the ventricle was not influenced by the size of the hearts, because the protein density of the ventricular wall did not vary from one group to another. The decrease in fluorescence intensity observed during anoxia might be attributed to the unavoidable diffusion of the accumulated oxidized form of DCFH out of the cell (15). The complex kinetics underlying this backdiffusion are technically difficult to determine, and the changes in DCF fluorescence intensity vs. time (arbitrary units/s) might be slightly underestimated in all experiments to the same extent. The fact that the rate of DCFH oxidation in the absence of oxygen was negligible showed that only oxygen-derived radicals were able to alter fluorescence and that there was no illumination-induced DCFH autoxidation under our experimental conditions (Fig. 1).
Although the vehicle DMSO used in this study is also known as a cell-penetrating OH· scavenger, it affected neither ROS production nor the time course of functional parameters with respect to the vehicle-free control condition.
Contractile activity. In contrast to the chronotropic and dromotropic parameters, the ventricular shortening and maximal shortening velocity displayed important interindividual variations, even after in vitro stabilization (Table 1). Such variability of contractile activity might be due to slight differences in the developmental stage, variations of cardiac three-dimensional structure, and/or degree of flattening of the hearts in the culture compartment, as discussed previously (29, 30). Shortening determined just before anoxia (19 ± 17 µm, n = 56) and at the end of reoxygenation (20 ± 16 µm, n = 56), as well as the rate of contractile recovery, was similar in all experimental groups, indicating that contractile activity fully recovered. Furthermore, inotropy and lusitropy were altered to the same extent, because the ratio of the maximal contraction velocity to the maximal relaxation velocity was stable throughout anoxia-reoxygenation: 1.25 ± 0.25 (n = 56).
Opening of the mitoKATP Channel Increases ROS Production and Improves Functional Recovery
The maximal rate of ROS production was four times higher during reoxygenation than during preanoxia, indicating that the embryonic heart is subjected to an important postanoxic oxidative stress. This is corroborated by the 60–70% reduction in ROS production by MPG or Myxo, indicating also that mitochondrial complex III was the main source of ROS during reoxygenation.
It has been shown that treatment of adult human and mammalian cardiomyocytes with the mitoKATP channel opener Diazo before, but not during, ischemia-reperfusion increased ROS production and mimicked ischemic preconditioning (3, 7, 25, 31). However, in our experimental model, Diazo was present before and during anoxia-reoxygenation and increased ROS production during the early phase of reoxygenation, but not during normoxia. These differences between the adult and developing heart may be partly attributed to the degree of mitochondrial differentiation (35, 43) illustrated by a low oxidative metabolism (29) and to a limited antioxidant capacity at the stage investigated. Moreover, the fact that 5-HD alone affected neither ROS production nor functional recovery suggests that the mitoKATP channel remains closed throughout anoxia-reoxygenation in the control hearts. Interestingly, the Diazo-induced additional ROS production was associated with an improved recovery of ventricular E-C coupling during reoxygenation. The facts that Diazo is 1,000-fold more potent to activate mitoKATP than sarcolemmal KATP channels (9) and that the nonselective KATP channel blocker Glib, but not 5-HD, induced major atrial arrhythmias associated with second-degree AV blocks clearly indicate that the mitoKATP channels are directly involved in Diazo-induced protection of E-C coupling, whereas sarcolemmal KATP channels play no significant role. The obvious proarrhythmic effect of Glib observed in this work suggests, however, that the sarcolemmal KATP channels might play a crucial role in controlling the membrane potential, especially in the immature sinoatrial pacemaker cells.
The fact that MPG abolished Diazo-induced ROS production and E-C coupling protection during reoxygenation clearly indicates that ROS are also directly involved in the protective effect of Diazo in the embryonic heart. These results are in agreement with recent reports on pharmacological preconditioning using mitoKATP opener in adult mammalian (7, 27), adult human (3), and embryonic chick (19) cardiomyocytes. Thus the postanoxic protection of E-C coupling afforded by mitoKATP channel opening seems to be mediated by an additional ROS production, independent of the investigated species, the experimental models, or the selected protocol.
PKC Protects Selectively Ventricular E-C Coupling
During reoxygenation, Chel abolished the Diazo-induced protection of EMDv, suggesting that E-C coupling is modulated by PKC, which can be indirectly activated by mitoKATP channel opening via ROS production (Fig. 7). The E-C coupling machinery (mainly the plasmalemmal L-type Ca2+ channel, the ryanodine receptor channel, SERCA2, and phospholamban) is present and starts to be functional in the 4-day-old embryonic chick heart (37). It has previously been found that Ca2+ entry through a sarcolemmal L-type Ca2+ channel predominantly contributes to the Ca2+ transient during the cardiac cycle and strongly contributes to E-C coupling (39) but also participates in reoxygenation-induced Ca2+ overload and myocardial dysfunction at stage 24HH (37). These observations further strengthen the important role that L-type Ca2+ channel modulation by PKC may play in postanoxic recovery of the developing heart. Indeed, although controversial (17), PKC activity has been reported to restore Ca2+ homeostasis and play an important role in regulation of E-C coupling during postischemic reperfusion, whereas Chel, at the concentration used in the present study, significantly enhances postischemic Ca2+ overload in the adult heart (36). On the basis of our findings, we propose that such mechanisms operate also in the anoxic-reoxygenated embryonic heart. However, the role of different PKC isoforms, as well as their target proteins, deserves further investigation.
|
Endogenous NO Plays an Important Role in Diazo-Induced ROS Production and Cardioprotection
Recently, it has been shown that exogenous NO can activate the mitoKATP channel (33) and that NO and ROS are necessary to increase cell survival of 10-day-old chick cardiomyocytes via opening of mitoKATP channels (19). We recently showed that the embryonic heart produces NO during anoxia-reoxygenation mainly derived from inducible NOS, which is known to be strongly expressed early during cardiogenesis (38). The present data show that L-NAME alone delays recovery of EMDv, as in our previous work (21). These findings are in agreement with studies showing that NO can increase the open probability of the ryanodine receptor channel (46), which could accelerate Ca2+ release from the sarcoplasmic reticulum and, thus, shorten EMDv during reoxygenation (Fig. 7). The fact that NOS inhibition had no effect on the DCF signal during anoxia-reoxygenation indicates that production of ONOO– resulting from a possible reaction between NO and O2– is negligible in the embryonic heart under control conditions. Indeed, the DCFH technique is known to be sensitive not only to H2O2 and OH· but also to ONOO– (15). Moreover, myocardial levels of nitration of protein tyrosine by ONOO– (nitrotyrosine assessed by immunoblotting assay) were not increased after anoxia-reoxygenation (Raddatz and Markert, unpublished data).
Because L-NAME abolished Diazo-mediated ROS production and functional protection, activation of NOS appears to be a prerequisite for the beneficial effects of mitoKATP channel opening in the setting of anoxia-reoxygenation. It seems that cardiac mitochondria contain an NOS isoform (mtNOS) (18, 48) and that NO inhibits mitochondrial respiration at the complex IV level (2) and increases H2O2 production (32). On the basis of these observations and our findings, we hypothesize that endogenously produced NO could act directly on proteins involved in E-C coupling when mitoKATP channels are closed, as under control conditions. On the other hand, opening of the mitoKATP channel by Diazo could activate mtNOS and increase intramitochondrial NO production, resulting in increased ROS production at the level of complex III and I. This additional ROS production could activate PKC and result in phosphorylation of the L-type Ca2+ channel and/or ryanodine receptor channel, improving functional recovery (Fig. 7).
Our data showing that the PR interval was significantly prolonged and that the incidence of AV blocks was specially high during anoxia and reoxygenation clearly indicate that cell-to-cell communication was strongly altered. Indeed, although there is no specialized conduction system, various connexins (e.g., Cx43) are expressed and functional during early cardiogenesis, and their relative abundance (44), electrophysiological characteristics (4, 42), and regulation (5) vary throughout development. The fact that Diazo-induced protection of the PR interval was not suppressed by any of the agents, except L-NAME, suggests that NO, associated with Diazo, is necessary to protect cell-to-cell communication during anoxia-reoxygenation. Indeed, it has been shown that electrical uncoupling of cardiomyocytes during ischemia can be accelerated by KATP channel inhibition (16) and that L-NAME can decrease Cx43 phosphorylation and expression in vascular muscle (11), modulating cell-to-cell communication. However, information about a possible interaction of NO and Diazo with connexins is not available in the literature, and the exact mechanisms underlying this protection remain to be explored.
In conclusion, to the best of our knowledge, this study is the first to explore reoxygenation-induced alterations of electromechanical coupling in relation to pharmacological modulation of mitoKATP channels in the developing heart. Our results show that activation of the mitoKATP channel by Diazo throughout anoxia-reoxygenation improves recovery of AV conduction and ventricular E-C coupling without significantly affecting other parameters. Furthermore, protection of E-C coupling by mitoKATP channel opening requires NOS activation, ROS signaling, and PKC-sensitive pathways.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-Connexin (connexin43) gap junctions and activities of cAMP-dependent protein kinase and protein kinase C in developing mouse heart. Dev Dyn 223: 96–107, 2002.[CrossRef][Web of Science][Medline]
- and 
-PKC in cardiac ischemia and reperfusion: targeting the apoptotic machinery. Arch Biochem Biophys 420: 246–254, 2003.[CrossRef][Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  Y. Wang, N. Ahmad, B. Wang, and M. Ashraf Chronic preconditioning: a novel approach for cardiac protection Am J Physiol Heart Circ Physiol, May 1, 2007; 292(5): H2300 - H2305. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
P < 0.02 vs. control; *P < 0.003 vs. vehicle; #P < 0.01 vs. Diazo (Student's t-test).
