Transient Exposure to Hydrogen Peroxide Causes an Increase in Mitochondria-Derived Superoxide As a Result of Sustained Alteration in L-Type Ca2+ Channel Function in the Absence of Apoptosis in Ventricular Myocytes
We sought to understand the effect of a transient exposure of cardiac myocytes to H2O2 at a concentration that did not induce apoptosis. Myocytes were exposed to 30 μmol/L H2O2 for 5 minutes followed by 10 U/mL catalase for 5 minutes to degrade the H2O2. Cellular superoxide was measured using dihydroethidium. Transient exposure to H2O2 caused a 66.4% increase in dihydroethidium signal compared with controls exposed to only catalase, without activation of caspase 3 or evidence of necrosis. The increase in dihydroethidium signal was attenuated by the mitochondrial inhibitors myxothiazol or carbonyl cyanide p-(trifluoromethoxy)phenyl-hydrazone and when calcium uptake by the mitochondria was inhibited with Ru360. We investigated the L-type Ca2+ channel (ICa-L) as a source of calcium influx. Nisoldipine, an inhibitor of ICa-L, attenuated the increase in superoxide. Basal channel activity increased from 5.4 to 8.9 pA/pF. Diastolic calcium was significantly increased in quiescent and contracting myocytes after H2O2. The response of ICa-L to β-adrenergic receptor stimulation was used as a functional reporter because decreasing intracellular H2O2 alters the sensitivity of ICa-L to isoproterenol. H2O2 increased the K0.5 required for activation of ICa-L by isoproterenol from 5.8 to 27.8 nmol/L. This effect and the increase in basal current density persisted for several hours after H2O2. We propose that extracellular H2O2 is associated with an increase in superoxide from the mitochondria caused by an increase in Ca2+ influx from ICa-L. The effect persists because a positive feedback exists among increased basal channel activity, elevated intracellular calcium, and superoxide production by the mitochondria.
Reactive oxygen species (ROS) can act as signaling molecules able to stimulate and modulate a variety of biochemical and genetic systems, including the regulation of signal transduction pathways, gene expression, proliferation, and cell death by apoptosis.1 The regulation of signaling pathways by hydrogen peroxide (H2O2) and superoxide has been linked to the development of various cardiovascular diseases including ischemic heart disease, hypertension, cardiomyopathies, cardiac hypertrophy, and congestive heart failure.2–4
Mitochondria play an integral role in cellular metabolism and oxidative phosphorylation but are also a source of superoxide and an important determinant of the fate of a cell. Increased ROS production by mitochondria has been reported after exposing mitochondria to ROS in cardiac myocytes.5,6 The synchronized release of ROS by the mitochondria has been shown to induce oscillations in action-potential duration and life-threatening postischemic arrhythmias.5,7,8 A persistent increase in intracellular ROS is associated with pathological remodeling and myocardial dysfunction.2,4,9
It has been suggested that increases in mitochondria-derived ROS are attributable to a direct effect of ROS on mitochondrial function.5,6 Another possible explanation for increased production of ROS by the mitochondria is enhanced Ca2+ uptake attributable to altered L-type Ca2+ channel (ICa-L) function. It is reasonable to postulate an involvement of the channel in oxidative stress because the α1C subunit of the channel protein contains a number of cysteines that could be modified under oxidizing conditions. In support of this, channel function can be acutely modified by thiol-oxidizing compounds including H2O2.10–13 Acute exposure to H2O2 or thiol-oxidizing agents has been shown to increase macroscopic basal ICa-L.10,11,13,14 In addition, adrenergic regulation of the channel is modified in response to alterations in the ROS production of the cell. A decrease in cellular production of superoxide or H2O2 has been shown to increase the sensitivity of the channel to β-adrenergic receptor stimulation.10,11,15 Therefore, there is good evidence that the activity of ICa-L is responsive to alterations in the redox state of the cell.
In this study, we sought to understand the effects of H2O2 on myocyte function at a concentration insufficient to cause apoptosis or necrosis. We found that transient exposure to H2O2 induces an increase in mitochondria-derived superoxide in ventricular myocytes. The increase in cellular superoxide is reversible and is associated with increased intracellular Ca2+ and influx of Ca2+ into the mitochondria as a result of an increase in basal ICa-L density. The increase in basal ICa-L persisted for several hours after exposure to H2O2. We propose a model to explain the data and the persistent response. We suggest that this may be a possible mechanism for pathophysiological remodeling associated with transient oxidative stress that involves elevated intracellular Ca2+ and ROS.
Materials and Methods
Detection of Superoxide
Guinea pig ventricular myocytes and rat neonatal ventricular myocytes were isolated using a collagenase digestion method, as previously described.10,16 Generation of superoxide was assessed in 24-hour guinea pig myocytes or 36-hour contracting neonatal myocytes using the fluorescent indicator dihydroethidium (DHE) (5 μmol/L, 515- to 560-nm excitation filter, 590 long pass emission; Molecular Probes), as previously described.17 Application of the superoxide scavengers N-tert-butyl-α-phenyl-nitrone and superoxide dismutase significantly decreased DHE signal, confirming specificity of the indicator for superoxide (n=4).17 Comparisons using inhibitors were made with cells exposed to 30 μmol/L H2O2 for 5 minutes followed by 10 U/mL catalase for 5 minutes and were performed on the same day. For further details regarding cell isolation and detection of superoxide, see the online data supplement, available at http://circres.ahajournals.org.
Quantitation of Apoptosis/Caspase 3 Activity Assay
Caspase 3 activity was measured using Ac-DEVD-AMC (Promega) as a substrate.18,19 For further details, see the online data supplement.
Determination of Intracellular Ca2+
Intracellular Ca2+ was monitored using the fluorescent indicator Fura-2 acetoxymethyl ester (1 μmol/L, Molecular Probes). Fluorescent ratios at 340/380 nm excitation, 510 nm emission were measured over 50 ms at 1-minute intervals on a Hamamatsu Orca ER digital camera attached to an inverted Nikon TE2000-U microscope. Metamorph 6.3 was used to quantify the signal by manually tracing myocytes. An equivalent region not containing cells was used for background and was subtracted. The fluorescent ratios recorded over 3 minutes before and immediately after addition of H2O2 and catalase were averaged. We performed calibrations to determine intracellular Ca2+ concentrations in guinea pig ventricular myocytes as previously described (see the online data supplement).
Data Acquisition for Patch-Clamp Studies
The whole-cell configuration of the patch-clamp technique was used to record L-type Ca2+ currents up to 9 hours after isolation of myocytes as described previously.10 For further details, see the online data supplement.
Transient Exposure to H2O2 Is Associated With an Increase in DHE Signal
We examined the effect of a transient exposure to H2O2 on DHE signal in adult ventricular myocytes. We chose DHE because it is a good indicator of changes in intracellular superoxide and does not interact with H2O2.20 However, intracellular superoxide is rapidly dismutated to H2O2, and we have shown that increases in cellular superoxide parallel increases in cellular H2O2.15,17 We began the experiments by titrating increasing concentrations of H2O2 until we detected an increase in cellular superoxide. We found that 20 to 30 μmol/L H2O2 was the minimum concentration required to reproducibly induce an increase in cellular superoxide. Myocytes were exposed to 30 μmol/L H2O2 for 5 minutes followed by 10 U/mL catalase for 5 minutes to degrade the peroxide. The concentration of catalase was sufficient to completely degrade extracellular H2O2 (n=5; see the online data supplement). In 45 cells, 30 μmol/L H2O2 caused a 66.4±8.6% increase in DHE signal (Figure 1A and 1B). The increase in superoxide did not cause necrosis in any of the 45 cells tested (assessed with propidium iodide uptake) or apoptosis determined by caspase 3 assay (Figure 1C).
Source of Increased Production of Superoxide Is the Mitochondria
We examined different sources for the increase in superoxide. In vascular smooth muscle cells, NAD(P)H oxidase is a prominent source of superoxide.21 Cells were exposed to 50 μmol/L gp91ds-tat peptide, a concentration of the peptide that is sufficient to inhibit angiotensin II–induced superoxide production in aortic rings by preventing association of gp47phox with gp91phox in NAD(P)H oxidase.22 The cells were then exposed to 30 μmol/L H2O2 for 5 minutes followed by 10 U/mL catalase, and superoxide production was measured (Figure 2A). In 5 cells, 30 μmol/L H2O2 caused a 91.0± 26.9% increase in DHE signal. This was not significantly different from cells exposed to H2O2 in the absence of the peptide (66.4±8.6% increase, n=45). Similar results were obtained when cells were exposed to NAD(P)H oxidase inhibitor apocynin (300 μmol/L) followed by H2O2 (59.9±17.6% increase, n=5; P=NS versus H2O2 in the absence of apocynin).
We examined whether xanthine oxidase was a possible source of superoxide by exposing the cells to 50 μmol/L allopurinol before H2O2. In 7 cells, 30 μmol/L H2O2 caused a 49.6±5.5% increase in DHE signal that was not significantly different from the increase in DHE signal recorded in the absence of allopurinol (66.4±8.6%, n=45; Figure 2B). Similarly when cells were exposed to the nitric oxide synthase inhibitor NG-nitro-l-arginine methyl ester (L-NAME) (100 μmol/L), there was no change in the DHE signal compared with cells exposed to H2O2 in the absence of L-NAME (74.4±22.6%, n=7 versus 66.4±8.6%, n=45; P=NS; Figure 2C).
We examined whether mitochondria were the source of superoxide by partially reducing mitochondrial membrane potential with 2 nmol/L carbonyl cyanide p-(trifluoromethoxy) phenyl-hydrazone (FCCP), an uncoupler of oxidative phosphorylation. In 10 cells, FCCP significantly attenuated the increase in DHE signal (P<0.05; Figure 3A). In addition, exposure of myocytes to 30 μmol/L H2O2 for 5 minutes followed by 10 U/mL catalase for 5 minutes caused a 11.6±5.6% increase in mitochondrial membrane potential assessed with the fluorescent indicator JC-1 (n=13; Figure 3B; see also the online data supplement). We performed additional experiments in which we partially inhibited the electron transport chain with 7 nmol/L myxothiazol. In 14 cells, myxothiazol significantly decreased DHE signal (P<0.05; Figure 3C). Consistent with previously published data, FCCP and myxothiazol decreased DHE signal in the absence of H2O2 (Figure 3A and 3C, inset, right).17 These data strongly suggest that the source of increase in superoxide following exposure to H2O2 is the mitochondria.
Increase in Cellular Superoxide Requires an Increase in Uptake of Ca2+ by the Mitochondria
We investigated whether an uptake of Ca2+ into the mitochondria is a requirement for the increase in superoxide. Ru360 (2 μmol/L), an inhibitor of the mitochondrial Ca2+ uniporter, applied before or after 30 μmol/L H2O2 significantly attenuated the increase in DHE signal (P<0.05; Figure 4A). To investigate the source of calcium, we exposed cells to the ICa-L inhibitor nisoldipine (2 μmol/L) before or after 30 μmol/L H2O2. Nisoldipine significantly attenuated the increase in DHE signal (P<0.05; Figure 4B), suggesting that calcium influx through ICa-L is a requirement for the increase in cellular superoxide. Oxidative stress can cause Ca2+ release from ryanodine receptors.23 However preventing ryanodine receptor activation with 20 μmol/L dantrolene did not change the DHE signal (n=6; P=NS versus exposure to H2O2 in the absence of dantrolene; Figure 4C).
Transient Exposure to H2O2 Is Associated With Sustained Alteration in L-Type Ca2+ Channel Function
Alterations in cellular production of H2O2 influence ICa-L. Exposing cardiac myocytes to hypoxia is associated with a decrease in cellular production of superoxide and H2O2 by the mitochondria that results in a decrease in basal channel activity, while increasing the sensitivity of the channel to β-adrenergic receptor stimulation.10,11,15,17 In addition, perfusing myocytes intracellularly with catalase (that specifically converts H2O2 to H2O and O2) mimics the increase in sensitivity of the channel to β-adrenergic receptor stimulation during hypoxia.15 We therefore used the response of the channel to β-adrenergic receptor stimulation as a functional reporter of a persistent oxidative stress. We exposed the cells to 30 μmol/L H2O2 for 5 minutes followed by 10 U/mL catalase for 5 minutes to degrade extracellular H2O2 and recorded the sensitivity of the channel to increasing concentrations of the β-adrenergic receptor agonist isoproterenol (Iso). In the absence of H2O2, 0.003 μmol/L and 0.01 μmol/L Iso elicited currents that were 28.8±7.1 and 74.1±8.6% of the current elicited by 1 μmol/L Iso, a maximally stimulating concentration of the β-adrenergic receptor agonist within the same cell (n=5; Figure 5A). However, after exposure to H2O2, 0.003 μmol/L and 0.01 μmol/L Iso elicited currents that were only 0.67±0.4 and 22.4±9.3% of the current recorded in response to 1 μmol/L Iso (n=9; Figure 5B). The K0.5 for activation of the channel by Iso was significantly increased from 5.8±0.3 to 27.8±0.1 nmol/L (P<0.05; Figure 6A). There was no difference in the response of the channel to a maximally stimulating concentration of Iso (1 μmol/L) before or after exposure to H2O2 (Figure 6B). However, the activity of ICa-L under non–β-adrenergic-stimulated conditions (basal channel activity) was significantly increased from 5.4±0.5 (n=7) to 8.9±0.7 pA/pF (n=25) at +10 mV following exposure to H2O2 (P<0.05). These data were recorded, on average, 4 hours after exposure to H2O2. The decrease in sensitivity of the channel to Iso and the increase in basal current activity persisted for at least 9 hours after exposure to H2O2 (and degradation of extracellular H2O2 with10 U/mL catalase). These data suggest that a transient exposure to H2O2 is associated with a persistent alteration in ICa-L. It would appear that existing channels are persistently activated with increased current after exposure to H2O2.
Persistently Altered ICa-L Is Mediated by Superoxide Produced by the Mitochondria
The persistently altered ICa-L could have been attributable to irreversible oxidation of the channel protein following a transient exposure to H2O2. Alternatively, it may have been maintained in an oxidized state as a result of increased cellular H2O2. We perfused cells intracellularly with 2000 U/mL catalase (which converts H2O2 to H2O and O2) and measured the response of the channel to Iso. Catalase significantly attenuated the decrease in sensitivity of the channel to Iso and the increase in basal current activity (see the online data supplement). These data suggest that after a transient exposure to H2O2, channel function is altered as a result of elevated cellular H2O2.
We confirmed that the mitochondria were the source of production of superoxide. Following exposure to 30 μmol/L H2O2, when cells were perfused intracellularly with FCCP or myxothiazol, the decrease in sensitivity of the channel to Iso and increase in basal current were significantly attenuated (see the online data supplement). In addition, the source of superoxide did not involve NAD(P)H oxidase, xanthine oxidase, or nitric oxide (see the online data supplement). These results confirm that channel function is persistently altered after a transient exposure to H2O2 but can be reversed when mitochondrial production of superoxide is inhibited.
Transient Exposure to H2O2 Is Associated With Elevated Intracellular Ca2+
We measured intracellular Ca2+ before and after exposure of guinea pig ventricular myocytes to 30 μmol/L H2O2 using the fluorescent indicator Fura-2. Figure 7A illustrates the persistent increase in intracellular Ca2+ after exposure to H2O2 and catalase in a guinea pig myocyte. Exposure of cells to H2O2 caused a significant increase in 340/380 fluorescence that was attenuated by 2 μmol/L nisoldipine before exposure to H2O2 but not with prior exposure to 2 μmol/L Ru360 (P<0.05; Figure 7A, inset at right), indicating the source of calcium was ICa-L.
Activation of ICa-L Is Required for an Increase in Mitochondria-Derived Superoxide
We examined whether activation of ICa-L was sufficient to increase intracellular superoxide. Application of 2 μmol/L Bay K, an L-type Ca2+ channel agonist, caused a 79.2±14.4% increase in DHE signal (n=7). This was comparable to the increase in DHE recorded after exposure to H2O2 (Figure 1A and 1B). The increase in DHE could be attenuated by application of 2 μmol/L Ru360 or 2 μmol/L nisoldipine before exposure to Bay K (Figure 7B). These data indicate that activation of ICa-L is sufficient for an increase in mitochondrial uptake of calcium and increased superoxide production by the mitochondria.
Transient Exposure to H2O2 Is Associated With an Increase in DHE Signal and Elevated Intracellular Ca2+ in Active Cycling Myocytes
We examined the effect of a transient exposure to H2O2 on DHE signal in spontaneously contracting neonatal rat ventricular myocytes. Exposure to 30 μmol/L H2O2 for 5 minutes followed by 10 U/mL catalase for 5 minutes caused a 8.1-fold increase in DHE signal that could be attenuated by exposure of cells to 2 μmol/L Ru360 before H2O2 (Figure 8A). In addition, exposure of spontaneously contracting myocytes to 30 μmol/L H2O2 for 5 minutes resulted in a small increase in diastolic 340/380 fluorescence (10.5±5.1% increase in 4 of 19 cells), but after 10 minutes, the diastolic 340/380 fluorescence was increased 71.1±9.5% (Figure 8B), without causing apoptosis or necrosis, as determined by caspase 3 assay and lactate dehydrogenase release 24 hours later (see Figure III in the online data supplement). Application of 2 μmol/L nisoldipine before 30 μmol/L H2O2 prevented the increase in diastolic 340/380 fluorescence (Figure 8B). Consistent with the results recorded in quiescent guinea pig ventricular myocytes, a transient exposure to H2O2 is associated with a significant increase in cellular superoxide and diastolic Ca2+ in active calcium cycling myocytes.
In this study, we examined the effects of a brief exposure of cardiac myocytes to H2O2 at a concentration that did not cause apoptosis or necrosis. We found that 5 minutes of exposure of cardiac myocytes to 30 μmol/L H2O2 was sufficient to cause persistent alterations in cellular superoxide production, ICa-L, and cellular Ca2+. The increase in cellular superoxide was dependent on activation of ICa-L, and it was reversible. One current view regarding the increased production of ROS by mitochondria following oxidative stress is that the mitochondria are the target of ROS.6,8 Our results differ from these studies in that we specifically applied an oxidative stress externally that is likely to mimic a burst of ROS associated with ischemia/reperfusion in vivo. We attenuated the oxidative stress with catalase and examined the effect on cellular function. At low concentrations (30 μmol/L) of H2O2, an increase in the activity of ICa-L was required for the increase in mitochondrial ROS production. Because Bay K alone increased cellular superoxide (Figure 7B), it would appear that the increase in superoxide occurs solely because of an increase in basal ICa-L activity and that low concentrations of H2O2 do not have to directly affect mitochondria. Persistently elevated intracellular Ca2+ and ROS are associated with induction of calmodulin and NFAT pathways that lead to pathological states such as cardiac hypertrophy and failure.24,25 Chronic in vitro exposure to low concentrations of H2O2 (10 to 30 μmol/L) have been shown to induce protein synthesis in adult cardiac myocytes, without affecting survival.26 Our results may represent the mechanisms that contribute to the development of cardiac pathology.
The L-type Ca2+ channel is responsive to alterations in cellular redox state. When intracellular H2O2 is decreased with exposure of myocytes to hypoxia or intracellular application of catalase, the sensitivity of the channel to β-adrenergic receptor stimulation is increased.10,11,15 This response is mimicked when cells are exposed to the thiol-specific reducing agent dithiothreitol, suggesting that direct redox modification of the channel or redox modification of a signaling intermediate such as protein kinase A is responsible for the response. We found that the K0.5 for activation of the channel by Iso significantly increased after exposure to H2O2, consistent with a response of the channel to an oxidized cellular environment (Figures 5B and 6⇑A). The response persisted for many hours after the insult as a result of a persistent increase in production of superoxide by the mitochondria because the mitochondria inhibitors myxothiazol and FCCP attenuated the decrease in sensitivity of the channel to Iso. If direct oxidation of the channel is necessary for the increase in basal current density (as acute exposure to hydrogen peroxide or thiol-oxidizing agents would suggest10–13), then the persistent increase in cellular superoxide appears to be necessary to maintain the channel in an oxidized state. In support of this, we found that basal current density was increased many hours after the transient exposure to H2O2 (Figure 6B), and this was attenuated by intracellular catalase, FCCP, and myxothiazol.
Consistent with persistent L-type Ca2+ channel activation, intracellular Ca2+ was increased in the myocytes (Figures 7A and 8⇑B). Our data indicate that Ca2+ influx through the L-type Ca2+ channel is required for the increase in superoxide by the mitochondria because nisoldipine attenuated the increase in DHE signal (Figure 4B) and the increase in intracellular Ca2+ (Figures 7A and 8⇑B). In addition Bay K alone was sufficient to increase Ca2+ uptake into the mitochondria and increase superoxide (Figure 7B). Dantrolene, an inhibitor of ryanodine release of Ca2+ from sarcoplasmic reticulum stores, did not alter the increase in DHE signal after exposure to H2O2 (Figure 4C). We propose therefore that the increase in intracellular Ca2+ and superoxide persists because a positive feedback exists between increased basal channel activity and superoxide production by the mitochondria (Figure 8C). The model does not preclude a direct effect of H2O2 on the mitochondria. Our data show that increased L-type Ca2+ channel activity is required for the response at low H2O2 concentrations. The L-type Ca2+ channel appears to be an important regulator of cellular Ca2+ and cellular superoxide production under conditions of oxidative stress that do not involve apoptosis or necrosis and have the potential to mediate cardiac pathology.
Sources of Funding
This study was supported by a grant from the National Health and Medical Research Council of Australia. L.C.H. is the recipient of a National Health and Medical Research Council Career Development Award.
Presented in part at the 3rd Annual Symposium of the American Heart Association Council on Basic Cardiovascular Sciences, July 31–August 3, 2006, Keystone, Colo (Circ Res 2006;99:E24).
Original received May 31, 2006; resubmission received November 8, 2006; revised resubmission received February 7, 2007; accepted February 27, 2007.
Thannickal VJ, Fanburg BL. Reactive oxygen species in cell signaling. Am J Physiol. 2000; 279: L1005–L1028.
Aon MA, Cortassa S, Marban E, O’Rourke B. Synchronized whole cell oscillations in mitochondrial metabolism triggered by a local release of reactive oxygen species in cardiac myocytes. J Biol Chem. 2003; 278: 44735–44744.
Zorov DB, Filburn CR, Klotz LO, Zweier JL, Sollott SJ. Reactive oxygen species (ROS)-induced ROS release: a new phenomenon accompanying induction of the mitochondrial permeability transition in cardiac myocytes. J Exp Med. 2000; 192: 1001–1014.
Akao M, O’Rourke B, Teshima Y, Seharaseyon J, Marban E. Mechanistically distinct steps in the mitochondrial death pathway triggered by oxidative stress in cardiac myocytes. Circ Res. 2003; 92: 186–194.
Hool LC. Hypoxia increases the sensitivity of the L-type Ca(2+) current to beta-adrenergic receptor stimulation via a C2 region-containing protein kinase C isoform. Circ Res. 2000; 87: 1164–1171.
Hool LC. Hypoxia alters the sensitivity of the L-type Ca(2+) channel to alpha-adrenergic receptor stimulation in the presence of beta-adrenergic receptor stimulation. Circ Res. 2001; 88: 1036–1043.
Hool LC, Arthur PG. Decreasing cellular hydrogen peroxide with catalase mimics the effects of hypoxia on the sensitivity of the L-type Ca2+ channel to beta-adrenergic receptor stimulation in cardiac myocytes. Circ Res. 2002; 91: 601–609.
Bogoyevitch MA, Andersson MB, Gillespie-Brown J, Clerk A, Glennon PE, Fuller SJ, Sugden PH. Adrenergic receptor stimulation of the mitogen-activated protein kinase cascade and cardiac hypertrophy. Biochem J. 1996; 314 (pt 1): 115–121.
Hool LC, Di Maria CA, Viola HM, Arthur PG. Role of NAD(P)H oxidase in the regulation of cardiac L-type Ca2+ channel function during acute hypoxia. Cardiovasc Res. 2005; 67: 624–635.
Griendling KK, Sorescu D, Ushio-Fukai M. NAD(P)H oxidase: role in cardiovascular biology and disease. Circ Res. 2000; 86: 494–501.
Rey FE, Cifuentes ME, Kiarash A, Quinn MT, Pagano PJ. Novel competitive inhibitor of NAD(P)H oxidase assembly attenuates vascular O(2)(-) and systolic blood pressure in mice. Circ Res. 2001; 89: 408–414.
McIntosh MA, Cobbe SM, Smith GL. Heterogeneous changes in action potential and intracellular Ca2+ in left ventricular myocyte sub-types from rabbits with heart failure. Cardiovasc Res. 2000; 45: 397–409.