Role of sarcoplasmic reticulum in mitochondrial permeability transition and cardiomyocyte death during reperfusion

Transcription

1 Am J Physiol Heart Circ Physiol 297: H1281 H1289, First published August 14, 2009; doi: /ajpheart Role of sarcoplasmic reticulum in mitochondrial permeability transition and cardiomyocyte death during reperfusion Marisol Ruiz-Meana, Arancha Abellán, Elisabet Miró-Casas, Esperanza Agulló, and David Garcia-Dorado Servicio de Cardiologia, Hospital Universitari Vall d Hebron, Barcelona, Spain Submitted 8 May 2009; accepted in final form 10 August 2009 Ruiz-Meana M, Abellán A, Miró-Casas E, Agulló E, Garcia- Dorado D. Role of sarcoplasmic reticulum in mitochondrial permeability transition and cardiomyocyte death during reperfusion. Am J Physiol Heart Circ Physiol 297: H1281 H1289, First published August 14, 2009; doi: /ajpheart There is solid evidence that a sudden change in mitochondrial membrane permeability (mitochondrial permeability transition, MPT) plays a critical role in reperfusion-induced myocardial necrosis. We hypothesized that sarcoplasmic reticulum (SR) Ca 2 cycling may induce partial MPT in microdomains of close anatomic proximity between mitochondria and SR, resulting in hypercontracture and cell death. MPT (mitochondrial calcein release), cell length, and sarcolemmal rupture (Trypan blue and lactate dehydrogenase release) were measured in adult rat cardiomyocytes submitted to simulated ischemia (NaCN/2- deoxyglucose, ph 6.4) and reperfusion. On simulated reperfusion, 83 2% of myocytes developed hypercontracture. In 22 6% of cases, hypercontracture was associated with sarcolemmal disruption [Trypan blue( )]. During simulated reperfusion there was a 25% release of cyclosporin A-sensitive mitochondrial calcein (with respect to total mitochondrial calcein content). Simultaneous blockade of SR Ca 2 uptake and release with thapsigargin and ryanodine, respectively, significantly reduced mitochondrial calcein release, hypercontracture, and cell death during simulated reperfusion. SR Ca 2 blockers delayed mitochondrial Ca 2 uptake in digitonin-permeabilized cardiomyocytes but did not have any effect on isolated mitochondria. Pretreatment with colchicine to disrupt microtubule network reduced the degree of fluorescent overlap between SR and mitochondria and abolished the protective effect of SR Ca 2 blockers on MPT, hypercontracture, and cell death during reperfusion. We conclude that SR Ca 2 cycling during reperfusion facilitates partial mitochondrial permeabilization due to the close anatomic proximity between both organelles, favoring hypercontracture and cell death. calcium; mitochondria; hypercontracture MYOCARDIAL REPERFUSION may precipitate cell death within minutes of restoration of blood flow (27). The occurrence of a sudden change in the permeability of the mitochondrial membranes (mitochondrial permeability transition, MPT), favored by oxidative stress and Ca 2 overload, may arrest mitochondrial respiration and induce necrotic death of cardiomyocytes (10, 12). This phenomenon has been shown to play a central role in reperfusion-induced necrosis, as demonstrated in studies in which pharmacological or genetic inhibition of MPT susceptibility limited infarct size (3, 7, 24). A recent pilot study demonstrated that intravenous administration of the MPT inhibitor cyclosporin A (CsA) immediately before percutaneous coronary intervention reduced infarct size in patients with acute myocardial infarction (26). Address for reprint requests and other correspondence: D. Garcia-Dorado, Servicio de Cardiologia, Hospital Universitari Vall d Hebron, Pg. Vall d Hebron, , Barcelona, Spain ( dgdorado@vhebron.net). Previous studies have indicated that in reperfused cardiomyocytes resumption of ATP synthesis may activate sarcoplasmic reticulum (SR) Ca 2 cycling (1, 36). SR Ca 2 cycling is favored by cytosolic Ca 2 overload and is due to consecutive Ca 2 uptake through SR Ca 2 pump [sarco(endo)plasmic reticulum Ca 2 -ATPase (SERCA)] and subsequent Ca 2 release through ryanodine receptors (RyR) when the storage capacity of the organelle is exceeded, resulting in cytosolic Ca 2 oscillations that propagate as Ca 2 waves and may induce arrhythmias and myofibrillar hypercontracture (1, 36, 40). Hypercontracture is energy dependent (2, 35) and has been shown to contribute to reperfusion-induced myocardial necrosis (4, 19, 27). The histopathology of reperfused myocardial infarcts is characterized by the presence of areas of contraction band necrosis, reflecting cardiomyocyte extreme cell shortening or hypercontracture occurring within minutes of reperfusion (4), and pharmacological contractile inhibition during the initial minutes of reperfusion markedly reduces infarct size in a variety of models (15, 33, 37, 39). Prevention of Ca 2 oscillations with blockers of SR Ca 2 uptake and release has been shown to protect against hypercontracture and cell death (36). Recent studies have demonstrated a close anatomic and functional relationship between SR and mitochondria (8, 16, 28). Mitochondrial Ca 2 uptake has been shown to be more dependent on high cytosolic Ca 2 microdomains around the contact areas between endoplasmic reticulum (ER) and mitochondria than on cytosolic Ca 2 concentration (28, 38). It has been proposed that the interplay between SR and mitochondria is involved in the propagation of Ca 2 waves in cardiomyocytes and in apoptotic and necrotic death in other cell types (6, 11). However, the role of SR-mitochondria interplay in reperfusion-induced cardiomyocyte death has not been previously reported. In the present study we investigated the hypothesis that SR Ca 2 cycling may directly favor MPT through SRmitochondria interplay. METHODS Simulated ischemia-reperfusion in adult rat cardiac myocytes. To obtain isolated cardiac cells, principles of laboratory animal care (Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health; NIH Pub. No , revised 1996) were followed, and the procedure was approved by the Research Commission on Ethics of the Hospital Vall d Hebron. Rat cardiac myocytes were isolated as previously described (30). Briefly, adult male Sprague-Dawley rats (300 g) were deeply anesthetized by an intraperitoneal injection of pentobarbital sodium (150 mg/kg), and the heart was rapidly excised, cannulated by the aorta in a Langendorff system, and retrogradely perfused for a 20 min with a modified Krebs buffer (in mmol/l: 110 NaCl, 2.6 KCl, 1.2 KH 2 PO 4, 1.2 MgSO 4,25 NaHCO 3, 11 glucose) containing 25 mol/l CaCl 2 and 0.03% type II collagenase. Cells from dissociated tissue were subjected to a progressive normalization of Ca 2 levels to a final concentration of /09 $8.00 Copyright 2009 the American Physiological Society H1281

2 H1282 mmol/l. Rod-shaped cells were selected by gravity sedimentation in a 4% BSA gradient and plated on laminin-coated glass-bottom culture dishes with medium 199-HEPES and 4% fetal calf serum. Isolated cardiomyocytes were submitted to 20 min of simulated ischemia at 37 C by incubation in a glucose-free buffer containing (in mmol/l) 2 NaCN, 20 2-deoxyglucose, 140 NaCl, 3.6 KCl, 1.2 MgSO 4, 2 CaCl 2, and 20 HEPES, at ph 6.4, and reperfused in a control buffer at ph 7.4 containing (in mmol/l) 140 NaCl, 3.6 KCl, 1.2 MgSO 4,2 CaCl 2, 20 HEPES, and 5 glucose. Changes in cell length. Changes in cell length were measured at the end of simulated ischemia and at 10 min of simulated reperfusion. Rigor shortening was defined as a 30% reduction of cell length with respect to baseline conditions with preservation of square-shaped geometry; hypercontracture was defined as 60% of shortening with respect to the initial cell length with concomitant morphological distortion of cell geometry (round-shaped morphology). The number of rod-shaped rigor and hypercontracted cells was expressed as a percentage with respect to the total cell number. Mitochondrial membrane potential during simulated ischemia and reperfusion. Isolated cardiac myocytes were loaded for 10 min at 37 C with 10 mol/l JC-1 (Molecular Probes) and excited at 488 nm with an Ar/Kr laser confocal system (Yokogawa CSU10, Nipkow spinning disk), set on an Olympus IX70 (VoxCell Scan, Visitech) at 60. Green and red emission fluorescence were detected at 520 and 590 nm, respectively, with two digital charge-coupled device (CCD) cameras (Hamamatsu). Changes in JC-1 ratio fluorescence were calculated with commercially available software (VoxCell Scan, Visitech) under baseline conditions, at the end of simulated ischemia, and throughout reperfusion, and mitochondrial membrane potential ( m) was expressed as percentage of JC-1 ratio fluorescence (590/ 520 nm) with respect to baseline conditions. As a reference, maximal ratio fluorescence decrease was obtained by 5-min incubation with 200 mol/l dinitrophenol (DNP, maximal depolarization). Mitochondrial permeability transition induced by simulated reperfusion. MPT was monitored in cardiomyocytes loaded for 30 min with 5 mol/l calcein-am at 37 C, followed by 10-min incubation with 1 mmol/l CoCl 2 to quench cytosolic and nuclear fluorescence, as previously described (25), and submitted to simulated ischemia and reperfusion. MPT was detected as a CsA-sensitive decrease in calcein fluorescence (excitation 488 nm/emission 525 nm) during reperfusion with an Ar/Kr laser confocal system (Yokogawa CSU10, Nipkow spinning disk) set on an Olympus IX70 (VoxCell Scan, Visitech) at 60. The contribution of SR Ca 2 uptake and release to MPT was analyzed after blocking both SERCA and RyR with thapsigargin (2 mol/l) and ryanodine (10 or 100 mol/l), respectively, added simultaneously during reperfusion. The contribution of mitochondrial Ca 2 uptake to MPT was analyzed by inhibiting the mitochondrial Ca 2 uniporter with ruthenium red (25 mol/l) during reperfusion. To avoid fluorescence artifacts imposed by hypercontracture, 20 mmol/l 2,3- butanedione monoxime (BDM) was present during reperfusion. In a subset of experiments, maximal mitochondrial calcein release was induced by osmotic cell swelling and Ca 2 overload (80 mosmol/ kgh 2O buffer with 5 mmol/l Ca 2 ), which resulted in sarcolemmal rupture and massive MPT. Calcein release was measured every 10 s for 15 min with commercially available software (VoxCell Scan, Visitech) and was expressed as both arbitrary units of calcein fluorescence and percentage of calcein fluorescence with respect to maximal calcein release. Cell death. Release of lactate dehydrogenase (LDH) into the extracellular medium was measured after 20-min simulated ischemia and 15-min reperfusion as an index of cell death under different pharmacological treatments. At the end of the experiment, extracellular buffer was replaced by 1 ml of distilled water for 60 min to induce sarcolemmal disruption and promote the release of the remaining LDH. LDH release (expressed as % of enzyme release with respect to total cell content) was assayed spectrophotometrically (SLT Spectra Vision) at 340 nm. In a subset of experiments, cell death was measured by the Trypan blue exclusion test to specifically determine sarcolemmal disruption in hypercontracted cells. Cytosolic Ca 2 oscillations during simulated reperfusion. Cytosolic Ca 2 oscillations were analyzed during simulated reperfusion with a ratio fluorescence imaging system (QuantiCell2000, Visitech). Cardiac myocytes were loaded with 5 mol/l of the acetoxymethyl ester of fura-2 (Molecular Probes) for 20 min at 37 C, washed, and briefly postincubated to allow hydrolysis of the acetoxymethyl ester within the cells. Cells were alternatively excited at 340 and 380 nm at 40 by means of a fast-speed monochromator (bandwidth was set at 15 nm). Emitted light was collected by an air-cooled intensified digital camera, and 340-to-380 nm ratios were calculated from individual wavelength intensities obtained in each pixel cluster. Changes in the average ratio for previously defined regions of interest were analyzed throughout the reperfusion period. Mitochondrial Ca 2 uptake in permeabilized myocytes. Because monitoring of mitochondrial Ca 2 uptake during reperfusion by nonratiometric fluorescent probes is subjected to artifacts (due to the concurrence of high cytosolic Ca 2 and partial mitochondrial membrane depolarization that can promote fluorochrome leak from mitochondria to cytosol), we investigated the contribution of SR to mitochondrial Ca 2 uptake under normoxic conditions. For this purpose, myocytes were loaded with 4 mol/l rhod-2 AM (30 min at 37 C), washed, and permeabilized for 1 min with 100 mol/l digitonin in an intracellular-like buffer (containing in mmol/l: 135 KCl, 20 HEPES, 0.5 KH 2PO 4, 1 MgCl 2, 2 ATP, 5 succinate, 1 EGTA, ph 7.2). Rhod-2 fluorescence was determined with an Ar/Kr laser confocal system as changes in light emission at 590 nm after excitation at 568 nm. The effect of a 25 mol/l Ca 2 pulse on fluorescence was monitored in the presence and absence of thapsigargin (2 mol/l), ryanodine (100 mol/l), or colchicine (1 mol/l) every 1 s for 10 min and analyzed with commercially available software (VoxCell Scan, Visitech). Results in fluorescent intensity were expressed as changes with respect to the initial value. In a subset of cells, the RyR agonist caffeine (10 mmol/l) was added after 10 min of Ca 2 pulse to further quantify the contribution of SR Ca 2 release to mitochondrial Ca 2 uptake. Ca 2 uptake in isolated cardiac mitochondria. Rat heart mitochondria were isolated by differential centrifugation (18) and additionally purified by a 20% Percoll gradient. Mitochondrial protein concentration, determined by Bradford assay, was adjusted to 0.2 mg/0.25 ml. Mitochondrial Ca 2 uptake was calculated from changes in extramitochondrial free Ca 2 measured at 25 C with the Ca 2 -sensitive fluorescent probe calcium green-5n (1 mol/l hexapotassium salt, membrane impermeant; Invitrogen). Fluorescence at 530 nm was recorded with a fluorometer (SpectramaX GeminiXS, Molecular Devices) after addition of 50 mol/l Ca 2 to the mitochondrial suspension in the presence or absence of 2 mol/l thapsigargin or 100 mol/l ryanodine. In a separate group, 25 mol/l ruthenium red was present during Ca 2 pulse to obtain a positive control for mitochondrial Ca 2 uptake inhibition. The assay buffer contained (in mmol/l) 200 sucrose, 6 MOPS, ph 7.2, 2 KH 2PO 4, and 1 MgCl 2. Mitochondrial Ca 2 uptake was initiated by the addition of 0.5 mol/l rotenone and 6 mmol/l succinate. Decrease in fluorescent values (arbitrary units) was converted into mitochondrial Ca 2 concentration by using reference calcium green curves [assay buffer for calibration: 0.15 mol/l KCl, 6 mmol/l MOPS ph 7.2, 0.1 mmol/l EGTA; Ca 2 concentration ranged between 0 and 398 mol/l using a 10 mmol/l CaCl 2 stock; calculation of free Ca 2 concentration was done with the aid of Calcium Titration Program RECIP (revised 1989)]. SR and mitochondria labeling in cardiac myocytes. SR and mitochondria from adult cardiac myocytes were simultaneously stained with 1 mol/l ER-Tracker red-am and 1 mol/l Mito-Tracker green-fm (30 min, 37 C), respectively. Stained myocytes were permeabilized with 100 mol/l digitonin for 1 min. Images were obtained with a spectral confocal microscope (FV1000, Olympus) at 100, under control conditions and after 20-min treatment with 1 mol/l

3 H1283 Hypercontracture and cell death during simulated ischemiareperfusion. After 20 min of simulated ischemia, 65% of rod-shaped myocytes developed cell shortening with preservation of square-shaped morphology (ischemic rigor contracture). During the first minutes of reperfusion, most cells with ischemic rigor contracture underwent further cell shortening with massive disorganization of cell architecture, fulfilling the criteria of hypercontracture (Fig. 1). Hypercontracted cells usually preserved sarcolemmal integrity; however, 20% of hypercontracted cells were Trypan blue( ) (Fig. 1). To investigate the role of hypercontracture in the genesis of cardiomyocyte death, simulated reperfusion was performed in the presence of the contractile blocker BDM (20 mmol/l). Contractile inhibition significantly reduced the rate of hypercontracture in isolated cardiac myocytes but did not modify cell death (Fig. 2A). These results suggest that in isolated cardiomyocytes the small proportion of cell death occurring during reperfusion is independent from hypercontracture. Fig. 1. A: sequential images of a couple of rat cardiac myocytes ( 60) over time to illustrate morphological changes from baseline conditions and after 20 min of simulated ischemia (SI) (rigor contracture shortening) and 10 min of reperfusion (R) (hypercontracture). B: % of cardiac myocytes with rod-shaped morphology, rigor contracture, and hypercontracture (HC) after 20 min of simulated ischemia and 10 min of simulated reperfusion (R). On simulated reperfusion, most of the cells that had developed rigor contracture during the preceding ischemic period underwent hypercontracture. C: % of Trypan blue( ) [TBlue( )] and Trypan blue( ) [Tblue( )] hypercontracted cells at 15 min of reperfusion. Although most hypercontracted cardiomyocytes preserved sarcolemmal integrity, in 20% of the cases hypercontracture was associated with cell death [Trypan blue( )]. Data are as means SE corresponding to 1,950 cells from 5 experiments. colchicine to promote partial dissociation between SR and mitochondria. Dyes were excited at 488 nm (Mito-Tracker green) and 568 nm (ER-Tracker red), and an average of 65 Z-planes were collected for each cardiomyocyte. The degree of overlap between SR and mitochondria fluorescent signals, volume reconstruction of the Z-planes, and three-dimensional reconstruction were obtained with Image J software (National Institutes of Health, Bethesda, MD). Statistics. All data are presented as means SE. Comparisons between two groups were done with a two-tailed Student s t-test for independent samples and were considered significant when P All analysis were performed with commercially available software (SPSS for Windows, v ). RESULTS Fig. 2. A: cell death [% of lactate dehydrogenase (LDH) release with respect to total cell content] at 15 min of reperfusion in cardiomyocytes previously submitted to 20 min of simulated ischemia, under control conditions and in the presence of the contractile blocker 2,3-butanedione monoxime (BDM). Inset: rate of hypercontracture at 10 min of reperfusion in the same experimental groups. Data are means SE from 3 experiments (396 total cell counting). *P vs. control; ns, not significant. B: changes in mitochondrial membrane potential ( m) in cardiac myocytes, detected as decrease in JC-1 ratio fluorescence with respect to the initial normoxic value (F/F 0), at the end of 20-min simulated ischemia and at 10 min of simulated reperfusion. As a reference, maximal mitochondrial depolarization obtained with the mitochondrial uncoupler dinitrophenol (DNP) is also shown.

4 H1284 Fig. 3. A: mitochondrial calcein fluorescence in cardiomyocytes during 15 min of simulated reperfusion under control conditions and in the presence of 100 mol/l ryanodine-thapsigargin (Ryano/Tg) or cyclosporin A (CsA). a.u., Arbitrary unit. B: mitochondrial calcein release at 15 min of simulated reperfusion in the treatment groups: control, 10 mol/l ryanodine-thapsigargin, 100 mol/l ryanodine-thapsigargin, ruthenium red (RuR), and CsA. Calcein release is expressed as % of calcein fluorescence loss with respect to maximal calcein fluorescence loss induced by hyposmotic stress and Ca 2 overload. Inset: kinetics of maximal calcein release, obtained in cells submitted to hyposmotic stress (80 mosmol/kgh 2O) and Ca 2 overload (5 mmol/l). Data are means SE of 8 or 9 cells per group from 4 experiments. Statistical differences indicated are with respect to control. that could be effectively prevented by CsA (Fig. 3). Simultaneous blockade of SR Ca 2 uptake and release with 2 mol/l thapsigargin and 10 or 100 mol/l ryanodine, respectively, also reduced mitochondrial calcein release (Fig. 3). To test whether the effect of SR blockade on MPT could be secondary to Ca 2 transfer from SR to mitochondria, the mitochondrial Ca 2 uniporter blocker ruthenium red (25 mol/l) was added during reperfusion. Addition of ruthenium red had a significant protective effect against calcein release (Fig. 3). Interestingly, in all cases, mitochondrial calcein release was partial, suggesting that MPT occurred in only a fraction of cell mitochondria, as demonstrated by comparison with total mitochondrial calcein release obtained by exposing the cells to hypotonic medium and Ca 2 overload (Fig. 3). Blockade of SR with thapsigargin/ryanodine during reperfusion reduced both cell death (LDH release) and hypercontracture rate (75 4% hypercontracted cells in control reperfusion vs. 59 3% in the presence of SR Ca 2 blockers; P 0.05), and this protective effect could be mimicked by CsA (Fig. 4), which also reduced hypercontracture by 47% (P 0.01). Addition of ruthenium red also protected against sarcolemmal disruption, and its protective effect was additive to that produced by SR blockade, suggesting that mitochondrial Ca 2 uptake by the uniporter could be involved in MPT induced by SR (Fig. 4). However, ruthenium red alone did not reduce the rate of reperfusion-induced hypercontracture (75 4%, P not significant), in agreement with previous observations (31). Altogether, these results suggest that partial MPT may facilitate cell death in hypercontracting cells. Cytosolic Ca 2 oscillations during reperfusion. Under the experimental conditions used in this study, reenergization by simulated reperfusion was associated with the occurrence of rapid cytosolic Ca 2 oscillations in about two-thirds of cells. Blockade of the mitochondrial Ca 2 uniporter with ruthenium red did not prevent the development of Ca 2 oscillations (Fig. 5). However, the SR Ca 2 uptake blocker thapsigargin and the SR Ca 2 release blocker ryanodine almost totally prevented the occurrence of cytosolic Ca 2 oscillations associated with reperfusion (Fig. 5). Changes in mitochondrial membrane potential during ischemia-reperfusion. At the end of the simulated ischemia period, m, analyzed by JC-1 ratio fluorescence, importantly decreased with respect to the baseline value, although it did not reach the level of maximal depolarization obtained after the addition of the mitochondrial uncoupler DNP. Reperfusion induced a rapid restoration of m (Fig. 2B). Contribution of sarcoplasmic reticulum to mitochondrial permeability transition and cell death during reperfusion. To assess MPT during reperfusion, mitochondrial calcein release was monitored in intact cardiac myocytes under baseline conditions, at the end of simulated ischemia, and throughout 15 min of reperfusion in the presence or absence of 1 mol/l CsA. Only minimal mitochondrial calcein release occurred during simulated ischemia ( % of total calcein release), probably because of the inhibitory effect of the low extracellular ph (6.4) on MPT. However, during the first 15 min of reperfusion there was a sustained mitochondrial calcein release Fig. 4. Cell death (% of LDH release with respect to total cell content) at 15 min of simulated reperfusion in the treatment groups: control, 10 mol/l ryanodine-thapsigargin, 100 mol/l ryanodine-thapsigargin, ruthenium red, 100 mol/l ryanodine-thapsigargin-ruthenium red, and CsA. Data are means SE of 4 6 experiments. Statistical differences indicated are with respect to control. Nx, normoxia.

5 H1285 Fig. 5. Representative cytosolic Ca 2 recordings (fura-2 ratio fluorescence) in individual cardiac myocytes during the first 10 min of reperfusion under control conditions (A) and in the presence of ruthenium red (B), thapsigargin (C), and ryanodine (D). E: quantification of cytosolic Ca 2 oscillations in these treatment groups. Ca 2 transfer from sarcoplasmic reticulum to mitochondria. The contribution of the interplay between SR and mitochondria to Ca 2 transfer was further analyzed in digitonin-permeabilized cardiomyocytes, in which the physical relations between SR and mitochondria are preserved (2). Addition of a Ca 2 pulse of 25 mol/l to these cells resulted in an important increase in mitochondrial Ca 2 uptake (rhod-2 fluorescence) under normoxic conditions. Mitochondrial Ca 2 uptake was significantly delayed when this maneuver was performed in the presence of SR blockers thapsigargin and ryanodine (Fig. 6A). Nevertheless, contrary to what was expected, exposure to a Ca 2 pulse in the presence of 1 mol/l colchicine (added to promote disruption between SR and mitochondria communication) did not delay mitochondrial Ca 2 uptake (Fig. 6A). Both colchicine and microtubules have been described to have effects on Ca 2 handling independent from their role in mitochondria-sr communication (5, 17, 34). Therefore, it cannot be ruled out that this result is secondary to a potential microtubule-mediated activation of RyR, as has been previously suggested (20), that counteracts the reduction in SR-induced mitochondrial Ca 2 uptake expected to be caused by reduced mitochondria-sr communication. Addition of the RyR agonist caffeine (10 mmol/l) to permeabilized cells 10 min after Ca 2 pulse induced further significant and sustained mitochondrial Ca 2 uptake, supporting a role for SR Ca 2 transfer to mitochondria (Fig. 6B). The effect of SR Ca 2 blockers on mitochondrial Ca 2 uptake could be interpreted as either a local transfer of Ca 2 from SR to mitochondria (susceptible to modification by SR blockers) or a direct effect of thapsigargin/ryanodine on mitochondria. To rule out a potential nonspecific effect of these drugs on mitochondria, experiments were performed with isolated rat cardiac mitochondrial preparations in which the contribution of SR is expected to be negligible. With this model, addition of a Ca 2 pulse of 50 mol/l induced a significant Ca 2 uptake in normoxic respiring mitochondria, but this effect could not be attenuated by either thapsigargin or ryanodine, while ruthenium red totally prevented it (Fig. 7B). These data argue against a direct effect of SR blockers on mitochondrial Ca 2 transport.

6 H1286 Disruption of interplay between sarcoplasmic reticulum and mitochondria. Simultaneous fluorescent labeling of SR and mitochondria with ER-Tracker red and Mito-Tracker green, respectively, in permeabilized cardiac myocytes revealed a pattern of close juxtaposition between these organelles (Fig. 8A). Pharmacological depolymerization of microtubules with 1 mol/l colchicine added over 20 min reduced the overlap between the fluorochromes (the overlap coefficient varied from 0.93 to 0.78 in an analysis including 4 independent cells with an average of 65 different planes per cell). This result suggests a discrete dissociation between SR and mitochondria as a consequence of disarrangement of the microtubule network, as previously described (23). The effect of colchicine was used to investigate the contribution of SR on reperfusion-induced MPT and cell death in cardiac myocytes. Although addition of colchicine during 20 min of simulated ischemia did not have any effect on the rate of reperfusion-induced MPT, it abolished the protective effect of SR blockers thapsigargin and ryanodine against mitochondrial permeabilization (Fig. 8B). The loss of protection of SR blockers on MPT in colchicine-pretreated cardiac myocytes was paralleled by a lack of effect against reperfusion-induced hypercontracture and cell death (Fig. 8C). Incubation of normoxic cardiomyocytes with colchicine did not have any effect on LDH release (Fig. 8C). DISCUSSION The present work describes a role for SR Ca 2 cycling in reperfusion-induced MPT and cardiomyocyte hypercontracture and death. This role appears to be dependent on the anatomicfunctional interplay between SR and mitochondria by a mechanism involving Ca 2 transfer between these organelles. These results may provide a new explanation for the protective effect of interventions aimed at preventing SR Ca 2 oscillations, and may help us to better understand the contribution of mitochondrial permeabilization and hypercontracture to reperfusioninduced cardiomyocyte death. Fig. 6. A: increase in rhod-2 fluorescence (reflecting mitochondrial Ca 2 uptake) in normoxic permeabilized cardiac myocytes after addition of a Ca 2 pulse of 25 mol/l (arrow) under control conditions, in the presence of sarcoplasmic reticulum (SR) Ca 2 blockers (ryanodine, thapsigargin, or both), and in the presence of colchicine. B: increase in rhod-2 fluorescence induced by addition of caffeine (arrow) in normoxic permeabilized cardiac myocytes previously submitted to a Ca 2 pulse. Fig. 7. A: decrease in calcium green-5n hexapotassium salt (CG5N) fluorescence (reflecting mitochondrial Ca 2 uptake) in isolated respiring mitochondria after the addition of a Ca 2 pulse of 50 mol/l (arrow) in the absence (control) or presence of SR Ca 2 blockers ryanodine (100 mol/l) and thapsigargin (2 mol/l) or in the presence of the mitochondrial Ca 2 uniporter blocker ruthenium red (25 mol/l). B: net mitochondrial Ca 2 uptake obtained from changes in calcium green-5n fluorescence using previously calibrated reference values at baseline conditions (0 min) and 5 min after the addition of the Ca 2 pulse in the 4 treatment groups. Data are means SE of 3 experiments. [Ca 2 ] i, intramitochondrial Ca 2 concentration.

7 H1287 Fig. 8. A: A1 and A2: confocal images ( 100) of a permeabilized cardiac myocyte simultaneously loaded with Mito-Tracker green and ER-Tracker red, and after merging both fluorochromes, under control conditions (A1) and 20 min after addition of 1 mol/l colchicine (A2). A volume reconstruction of a magnified region (framed by dashed yellow line) is shown at right. A3: 3-dimensional reconstruction of a cardiac myocyte submitted to the same labeling protocol under control conditions. A4: same as A3 after incubation with colchicine. B: mitochondrial calcein release at 15 min of simulated reperfusion, expressed as % of calcein fluorescence loss with respect to maximal calcein fluorescence loss, in the absence (Control) or the presence of SR Ca2 blockers in cells in which colchicine was present during the preceding simulated ischemia period. Open bar, control reperfusion without pretreatment with colchicine. Inset: kinetics of calcein fluorescence decrease during 15 min of reperfusion in cardiomyocytes in which colchicine was present during previous simulated ischemia in the absence (Control) and in presence of SR Ca2 blockers (ryanodine-thapsigargin). C: cell death (% of LDH release with respect to total cell content) at 15 min of simulated reperfusion under control conditions and in the presence of SR Ca2 blockers (ryanodine-thapsigargin), with or without pretreatment with colchicine during the preceding ischemic period. Inset: rate of hypercontracture at 10 min of reperfusion in the group pretreated with colchicine. Data are means SE of 3 experiments. Statistical difference indicated is with respect to control. AJP-Heart Circ Physiol VOL 297 OCTOBER

8 H1288 Reperfused infarcts are mainly composed of areas of contraction band necrosis reflecting cardiomyocyte hypercontracture (4, 14, 27). It has been proposed that during reperfusion energy restoration when cytosolic Ca 2 is overloaded activates SERCA, giving rise to cytosolic Ca 2 oscillations due to repetitive cycles of Ca 2 uptake and release. These SR-driven Ca 2 oscillations may lead to myofibrillar hyperactivation and development of hypercontracture. In the intact myocardium in which cardiac myocytes are tightly interconnected hypercontracture imposes a mechanical stress that results in sarcolemmal disruption (13, 19, 27). Transient contractile blockade during initial reperfusion has been shown to reduce myocardial necrosis in a variety of experimental models (15, 33, 37, 39). However, the contribution of hypercontracture as a causative mechanism of reperfusion-induced cell death has not been adequately quantified. Pharmacological or genetic inhibition of mitochondrial permeabilization effectively reduces infarct size both in animals and humans (3, 7, 24, 26). The relationship between hypercontracture and MPT is not clear, and it remains unexplained how MPT causing energy dissipation can occur in the presence of energy-dependent contractile activation leading to hypercontracture. One possibility is that MPT does not affect the whole population of mitochondria within the cell. It has been proposed that MPT in Ca 2 -overloaded mitochondria could worsen cytosolic Ca 2 handling and favor hypercontracture provided that sufficient intact mitochondria can sustain ATP synthesis (29). Because a fraction of SR is physically and functionally related to mitochondria (16), we hypothesized that SR Ca 2 cycling may increase the susceptibility for MPT in a fraction of mitochondria located in areas of close contact with SR, and that SR Ca 2 blockers may be cardioprotective independently of their effect on hypercontracture. Previous studies have provided extensive evidence of the interplay between SR and mitochondria (16, 28, 32). This interplay is the consequence of a close anatomic relationship between the organelles, with microdomains that allow the rapid exchange of molecules between them, including Ca 2 (16). It has been established by electronic tomography that the distance between the outer mitochondrial membrane and the ER in situ is nm (8), and that limited proteolysis with proteinase K in permeabilized cells disrupts the physical association between SR and mitochondria, abrogating mitochondrial Ca 2 increase in response to RyR agonists with no effect on cytosolic Ca 2 signal. Indeed, Ca 2 microdomains around the contact sites between ER and mitochondria are more determinant for mitochondrial Ca 2 uptake than cytosolic Ca 2 concentration (28), and its regulation has been implicated in energy supply and demand matching (21, 22). Although the molecular mechanisms controlling this interaction are not known, mitofusin 2 has been proposed as the protein that bridges the two organelles and provides the physical basis of their intercommunication during Ca 2 signaling (9). While the functional role of intracellular Ca 2 microdomains between SR (or ER) and mitochondria has been acknowledged in different cell types, the pathophysiological relevance of this interaction is beginning to be elucidated. It has been shown that SR vesicles can provide sufficient Ca 2 trigger for induction of mitochondrial membrane permeabilization in vitro (16) and that reducing the number of contact sites between ER and mitochondria by a genetically truncated isoform of SERCA1 may protect against apoptotic cell death (6). In accordance with these observations, a recent report showed that increasing SR Ca 2 content by transcriptional upregulation of the proapoptotic bcl-2 family protein NIX exacerbates SR-mitochondrial Ca 2 transfer, producing an apoptotic cardiomyopathy in mice (11). However, to our knowledge this is the first study describing a direct effect of SR in MPT and cell death in intact cardiac myocytes in the context of ischemia-reperfusion injury. We first provided evidence that alterations in SR Ca 2 handling were sensed by mitochondria, and that blockade of SR Ca 2 uptake and release with thapsigargin/ryanodine slowed mitochondrial Ca 2 uptake, an effect that cannot be explained by a direct action of these drugs on mitochondria because it was not observed when purified mitochondrial preparations were used. Addition of SR Ca 2 blockers effectively prevented cytosolic Ca 2 cycling during reperfusion, and this effect was paralleled by a reduction in MPT, hypercontracture, and cell death. Moreover, prevention of cytosolic Ca 2 oscillations did not reduce total cytosolic Ca 2 overload during the first minutes of reperfusion, in accordance with previous studies, supporting the hypothesis that there is a component of cell death triggered by a local mechanism not related to total cellular Ca 2 load. The role of direct SR-mitochondria interplay in MPT and cell death during reperfusion was further supported by experiments in which the anatomic relation between the two structures was partially disrupted with colchicine. It was shown previously that pharmacological depolymerization of microtubule network disrupts SR-mitochondria communication (23). In the present study this effect was confirmed by a reduction in the overlap coefficient obtained in confocal microscopy images. The results indicate that disruption of SR-mitochondria interplay with colchicine abolished the protective effect of SR Ca 2 uptake/release blockers on calcein release and cell death during simulated reperfusion. An interesting observation was that mitochondrial calcein measurements during simulated reperfusion suggested that only a fraction of the total cell mitochondria suffered MPT, supporting the hypothesis that MPT may be compatible with energy production sufficient to allow hypercontracture. Indeed, m rapidly recovered on reperfusion in most hypercontracting cells, and this was associated with the initiation of SRdriven cytosolic Ca 2 oscillations. Moreover, interventions reducing MPT during reperfusion, i.e., CsA and SR Ca 2 blockers, also reduced the rate of hypercontracture, suggesting that the phenomena may be interrelated. In summary, our results suggest that SR Ca 2 oscillations occurring during reperfusion may induce MPT and cell death through a direct interplay between SR and mitochondria at microdomains of close physical contact between these organelles. SR thus emerges as a target for prevention not only of reperfusion-induced hypercontracture but also of mitochondria-driven cell death. GRANTS This study was partially supported by Spanish Ministry of Science FIS- PI060996, Network for Cooperative Research on Cardiovascular Diseases (RETICS-RECAVA), and Comision Interministerial de Ciencia Y Technologia SAF (Spanish Ministry of Education).

9 H1289 REFERENCES 1. Abdallah Y, Gkatzoflia A, Gligorievski D, Kasseckert S, Euler G, Schluter KD, Schafer M, Piper HM, Schafer C. Insulin protects cardiomyocytes against reoxygenation-induced hypercontracture by a survival pathway targeting SR Ca 2 storage. Cardiovasc Res 70: , Altschuld RA, Wenger WC, Lamka KG, Kindig OR, Capen CC, Mizuhira V, Vander Heide RS. Structural and functional properties of adult rat heart myocytes lysed with digitonin. J Biol Chem , Baines CP, Kaiser RA, Purcell NH, Blair NS, Osinska H, Hambleton MA, Brunskill EW, Sayen MR, Gottlieb RA, Dorn GW, Robbins J, Molkentin JD. Loss of cyclophilin D reveals a critical role for mitochondrial permeability transition in cell death. Nature 434: , Barrabes JA, Garcia-Dorado D, Ruiz-Meana M, Piper HM, Solares J, Gonzalez MA, Oliveras J, Herrejon MP, Soler SJ. Myocardial segment shrinkage during coronary reperfusion in situ. Relation to hypercontracture and myocardial necrosis. Pflügers Arch 431: , Calaghan SC, Le Guennec JY, White E. Modulation of Ca 2 signaling by microtubule disruption in rat ventricular myocytes and its dependence on the ruptured patch-clamp configuration. Circ Res 88: E32 E37, Chami M, Oules B, Szabadkai G, Tacine R, Rizzuto R, Paterlini- Brechot P. Role of SERCA1 truncated isoform in the proapoptotic calcium transfer from ER to mitochondria during ER stress. Mol Cell 32: , Clarke SJ, McStay GP, Halestrap AP. Sanglifehrin A acts as a potent inhibitor of the mitochondrial permeability transition and reperfusion injury of the heart by binding to cyclophilin-d at a different site from cyclosporin A. J Biol Chem 277: , Csordas G, Renken C, Varnai P, Walter L, Weaver D, Buttle KF, Balla T, Mannella CA, Hajnoczky G. Structural and functional features and significance of the physical linkage between ER and mitochondria. J Cell Biol 174: , de Brito OM, Scorrano L. Mitofusin 2 tethers endoplasmic reticulum to mitochondria. Nature 456: , Di Lisa F, Menabo R, Canton M, Barile M, Bernardi P. Opening of the mitochondrial permeability transition pore causes depletion of mitochondrial and cytosolic NAD and is a causative event in the death of myocytes in postischemic reperfusion of the heart. J Biol Chem 276: , Diwan A, Matkovich SJ, Yuan Q, Zhao W, Yatani A, Brown JH, Molkentin JD, Kranias EG, Dorn GW. Endoplasmic reticulum-mitochondria crosstalk in NIX-mediated murine cell death. J Clin Invest 119: , Duchen MR, McGuinness O, Brown LA, Crompton M. On the involvement of a cyclosporin A sensitive mitochondrial pore in myocardial reperfusion injury. Cardiovasc Res 27: , Frank JS, Brady AJ, Farnsworth S, Mottino G. Ultrastructure and function of isolated myocytes after calcium depletion and repletion. Am J Physiol Heart Circ Physiol 250: H265 H275, Garcia-Dorado D, Inserte J, Ruiz-Meana M, Gonzalez MA, Solares J, Julia M, Barrabes JA, Soler-Soler J. Gap junction uncoupler heptanol prevents cell-to-cell progression of hypercontracture and limits necrosis during myocardial reperfusion. Circulation 96: , Garcia-Dorado D, Théroux P, Duran JM, Solares J, Alonso J, Sanz E, Muñoz R, Elizaga J, Botas J, Fernández-Avilés F, Soriano J, Esteban E. Selective inhibition of the contractile apparatus. A new approach to modification of infarct size, infarct composition and infarct geometry during coronary artery occlusion and reperfusion. Circulation 85: , Garcia-Perez C, Hajnoczky G, Csordas G. Physical coupling supports the local Ca 2 transfer between sarcoplasmic reticulum subdomains and the mitochondria in heart muscle. J Biol Chem 283: , Gomez AM, Kerfant BG, Vassort G, Pappano AJ. Autonomic regulation of calcium and potassium channels is oppositely modulated by microtubules in cardiac myocytes. Am J Physiol Heart Circ Physiol 286: H2065 H2071, Holmuhamedov EL, Jovanovic S, Dzeja PP, Jovanovic A, Terzic A. Mitochondrial ATP-sensitive K channels modulate cardiac mitochondrial function. Am J Physiol Heart Circ Physiol 275: H1567 H1576, Inserte J, Garcia-Dorado D, Ruiz-Meana M, Padilla F, Barrabes JA, Pina P, Agullo L, Piper HM, Soler-Soler J. Effect of inhibition of Na /Ca 2 exchanger at the time of myocardial reperfusion on hypercontracture and cell death. Cardiovasc Res 55: , Iribe G, Ward CW, Camelliti P, Bollensdorff C, Mason F, Burton RA, Garny A, Morphew MK, Hoenger A, Lederer WJ, Kohl P. Axial stretch of rat single ventricular cardiomyocytes causes an acute and transient increase in Ca 2 spark rate. Circ Res 104: , Liu T, O Rourke B. Enhancing mitochondrial Ca 2 uptake in myocytes from failing hearts restores energy supply and demand matching. Circ Res 103: , Maack C, O Rourke B. Excitation-contraction coupling and mitochondrial energetics. Basic Res Cardiol 102: , Mironov SL, Ivannikov MV, Johansson M. [Ca 2 ] i signaling between mitochondria and endoplasmic reticulum in neurons is regulated by microtubules. From mitochondrial permeability transition pore to Ca 2 - induced Ca 2 release. J Biol Chem 280: , Nakagawa T, Shimizu S, Watanabe T, Yamaguchi O, Otsu K, Yamagata H, Inohara H, Kubo T, Tsujimoto Y. Cyclophilin D-dependent mitochondrial permeability transition regulates some necrotic but not apoptotic cell death. Nature , Petronilli V, Miotto G, Canton M, Brini M, Colonna R, Bernardi P, and Di Lisa F. Transient and long-lasting openings of the mitochondrial permeability transition pore can be monitored directly in intact cells by changes in mitochondrial calcein fluorescence. Biophys J 76: , Piot C, Croisille P, Staat P, Thibault H, Rioufol G, Mewton N, Elbelghiti R, Cung TT, Bonnefoy E, Angoulvant D, Macia C, Raczka F, Sportouch C, Gahide G, Finet G, Andre-Fouet X, Revel D, Kirkorian G, Monassier JP, Derumeaux G, Ovize M. Effect of cyclosporine on reperfusion injury in acute myocardial infarction. N Engl J Med 359: , Piper HM, Garcia-Dorado D, Ovize M. A fresh look at reperfusion injury. Cardiovasc Res 38: , Rizzuto R, Pinton P, Carrington W, Fay FS, Fogarty KE, Lifshitz LM, Tuft RA, Pozzan T. Close contacts with the endoplasmic reticulum as determinants of mitochondrial Ca 2 responses. Science 280: , Ruiz-Meana M, Abellan A, Miro-Casas E, Garcia-Dorado D. Opening of mitochondrial permeability transition pore induces hypercontracture in Ca 2 overloaded cardiac myocytes. Basic Res Cardiol 102: , Ruiz-Meana M, Garcia-Dorado D, Hofstaetter B, Piper HM, Soler- Soler J. Propagation of cardiomyocyte hypercontracture by passage of Na through gap junctions. Circ Res 85: , Ruiz-Meana M, Garcia-Dorado D, Miro-Casas E, Abellan A, Soler- Soler J. Mitochondrial Ca 2 uptake during simulated ischemia does not affect permeability transition pore opening upon simulated reperfusion. Cardiovasc Res 71: , Rutter GA, Rizzuto R. Regulation of mitochondrial metabolism by ER Ca 2 release: an intimate connection. Trends Biochem Sci 25: , Schlack W, Uebing A, Schäfer M, Bier F, Schäfer S, Piper HM, Thämer V. Regional contractile blockade at the onset of reperfusion reduces infarct size in the dog heart. Pflügers Arch 428: , Shen JB, Pappano AJ. An estrogen metabolite, 2-methoxyestradiol, disrupts cardiac microtubules and unmasks muscarinic inhibition of calcium current. J Pharmacol Exp Ther 325: , Siegmund B, Koop A, Klietz T, Schwartz P, Piper HM. Sarcolemmal integrity and metabolic competence of cardiomyocytes under anoxiareoxygenation. Am J Physiol Heart Circ Physiol 258: H285 H291, Siegmund B, Schlack W, Ladilov YV, Balser C, Piper HM. Halothane protects cardiomyocytes against reoxygenation-induced hypercontracture. Circulation 96: , Sumida T, Otani H, Kyoi S, Okada T, Fujiwara H, Nakao Y, Kido M, Imamura H. Temporary blockade of contractility during reperfusion elicits a cardioprotective effect of the p38 MAP kinase inhibitor SB Am J Physiol Heart Circ Physiol 288: H2726 H2734, Szalai G, Csordas G, Hantash BM, Thomas AP, Hajnoczky G. Calcium signal transmission between ryanodine receptors and mitochondria. J Biol Chem 275: , Tani M, Hasegawa H, Suganuma Y, Shinmura K, Kayashi Y, Nakamura Y. Protection of ischemic myocardium by inhibition of contracture in isolated rat heart. Am J Physiol Heart Circ Physiol 271: H2515 H2519, Venetucci LA, Trafford AW, O Neill SC, Eisner DA. The sarcoplasmic reticulum and arrhythmogenic calcium release. Cardiovasc Res 77: , 2008.