Ru₃₆₀, a specific mitochondrial calcium uptake inhibitor, improves cardiac post-ischaemic functional recovery in rats in vivo

Abstract

Background and purpose:

The mitochondrial permeability transition pore (mPTP), an energy-dissipating channel activated by calcium, contributes to reperfusion damage by depolarizing the mitochondrial inner membrane potential. As mitochondrial Ca²⁺ overload is a main inductor of mPTP opening, we examined the effect of Ru₃₆₀, a selective inhibitor of the mitochondrial calcium uptake system against myocardial damage induced by reperfusion in a rat model.

Experimental approach:

Myocardial reperfusion injury was induced by a 5-min occlusion of the left anterior descending coronary artery, followed by a 5-min reperfusion in anaesthetized open-chest rats. We measured reperfusion-induced arrhythmias and functions indicative of unimpaired mitochondrial integrity to evaluate the effect of Ru₃₆₀ treatment.

Key results:

Reperfusion elicited a high incidence of arrhythmias, haemodynamic dysfunction and loss of mitochondrial integrity. A bolus intravenous injection of Ru₃₆₀ (15-50 nmol kg⁻¹), given 30-min before ischaemia, significantly improved the above mentioned variables in the ischaemic/reperfused myocardium. Calcium uptake in isolated mitochondria from Ru₃₆₀-treated ventricles was partially diminished, suggesting an interaction of this compound with the calcium uniporter.

Conclusions and implications:

We showed that Ru₃₆₀ treatment abolishes the incidence of arrhythmias and haemodynamic dysfunction elicited by reperfusion in a whole rat model. Ru₃₆₀ administration partially inhibits calcium uptake, preventing mitochondria from depolarization by the opening of the mPTP. We conclude that myocardial damage could be a consequence of failure of the mitochondrial network to maintain the membrane potential at reperfusion. Hence, it is plausible that Ru₃₆₀ could be used in reperfusion therapy to prevent the occurrence of arrhythmia.

Introduction

Mitochondrial oxidative phosphorylation provides all the energy required for the contractile process. This energy accounts for more than 90% of that required by the myocardium (Mootha et al., 1997). Under pathological conditions such as ischaemia, mitochondrial ATP synthesis is abolished, resulting in severe damage to the integrity of heart cells. At reperfusion, abrupt re-oxygenation causes further cell damage by reactive oxygen-derived species (ROS) (Ferrari et al., 2004). ROS affect the sarcoplasmic reticulum and the sarcolemmal membranes, increasing the cytosolic calcium concentration ([Ca²⁺]_c) (Krause et al., 1989; Dixon et al., 1990) and therefore the mitochondrial calcium concentration ([Ca²⁺]_m) (Miyamae et al., 1996). At high [Ca²⁺]_m, mitochondria undergoes, energy-consuming futile cycles through calcium release and re-uptake, because the proton-driven energy from the respiratory chain is used for cation transport instead of mitochondrial ATP production (Saris and Carafoli, 2005). In addition, mitochondrial calcium overload triggers a nonspecific increase in the inner membrane permeability, which contributes to the uncoupling of oxidative phosphorylation and thereby to a diminished ATP synthesis. Recent findings also indicate that mitochondria undergoing nonspecific membrane permeability changes release intramitochondrial molecules that participate in apoptotic death signalling, that is, cytochrome c, Smac/DIABLO and apotosis-inducing factor (Regula and Kirshenbaum, 2005).

On the other hand, it has been suggested that an early mechanism by which ischaemic preconditioning exerts its beneficial effects in the reperfused heart, is the opening of mitochondrial K⁺-ATP channels, that dissipate the inner mitochondrial membrane potential and reduce the driving force for Ca²⁺ influx through the mitochondrial calcium uniporter (mCaU) (Yellon and Downey, 2003; O'Rourke, 2004). Clearly, this molecule has been a critical target for cardioprotective approaches. In this regard, ruthenium red (RR) a classical inhibitor of the mCaU, shows protective effects against reperfusion injury in rat hearts (Ferrari et al., 1982; Carry et al., 1989; Miyamae et al., 1996). However, it has been demonstrated that this compound interacts with many proteins related to the excitation-contraction cycle, altering the contractile response in normal hearts and affecting other excitable tissues (Velasco and Tapia, 2000; Zhou and Bers, 2002).

Recently we demonstrated that Ru₃₆₀, a RR analogue, exerts specific inhibition of the mCaU, preventing mitochondrial permeability transition pore opening when perfused into isolated heart. We found that [Ca²⁺]_m decreased dramatically in mitochondria obtained from Ru₃₆₀-treated reperfused hearts, correlating with a partial inhibition of the mCaU (García-Rivas et al., 2005).

Therefore, in this study we explored the ability of this compound to protect against ischaemia–reperfusion damage in an in vivo rat model. We present evidence that Ru₃₆₀ prevents the post-ischaemic electrical dysfunction induced by reperfusion, by inhibiting calcium overload and the opening of the mitocondrial permeability transition pore (mPTP).

Methods

Animal groups

All procedures and protocols were performed on male Wistar rats, weighing 250–300 g, in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NHI publication No. 85 (23) revised 1996). The rats were randomly divided into three groups: (1) The control group (n=20) underwent identical surgical procedures as the ischaemia/reperfusion group (I/R) group, without coronary artery ligation. (2) The I/R (n=23) received saline solution (0.9%) for 30 min before ischaemia and then was subjected to the reperfusion protocol. (3) The treated group (I/R+Ru₃₆₀, n=26), received a bolus injection of Ru₃₆₀ dissolved in saline solution 30 min before artery ligation and then was subjected to the reperfusion protocol.

Ruthenium complex synthesis

Ru₃₆₀ (μ-oxo) bis (trans-formatotetramine ruthenium), is a coordination complex that forms a near-linear structure containing two ruthenium atoms linked by an oxygen-bridge and surrounded by amine groups (Figure 1). To synthesize this complex, we followed the procedure described by Ying et al. (1991). The purified preparation was slightly yellowish and exhibited a single λmax at 360 nm. Ru₃₆₀ was obtained in 0.4 M ammonium formiate buffer, pH 5.5. Ru₃₆₀ concentration after chemical synthesis was calculated from the molar extinction coefficient of the complex at 360 nm (ɛ=2.6 × 10⁴ M⁻¹ cm⁻¹), as described by several groups including ours (Ying et al., 1991; Matlib et al., 1998; Zazueta et al., 1999). Ammonium formiate buffer alone, diluted in saline solution, was used in some experiments to discard any effect on reperfusion recovery. Commercial RR was purified by the technique described by Luft (1971). This preparation was not contaminated with Ru₃₆₀. A single absorption peak at 533 nm was observed with distilled water.

Figure 1.

Structures of the oxo-bridged amine dinuclear ruthenium complex: Ru₃₆₀ and RR. Modified from Ying et al. (1991).

In vivo reperfusion protocols

Rats anaesthetized with sodium pentobarbitone (55 mg kg⁻¹ i.p.) were intubated and air ventilated (10 ml kg⁻¹, 72 breaths min⁻¹) using a rodent respirator (model 683, Harvard Apparatus, Cambridge, MA, USA). The arterial pressure was measured through a cannula inserted into the femoral artery and connected to a hydrostatic pressure transducer. Electrocardiogram (ECG) was monitored by using three platinum electrodes placed at DII standard position. Arterial blood pressure (ABP) and ECG were recorded during the first 30 s of each minute, in a polygraph model 79-D (Grass Instrument Co. Quincy, MA, USA). The femoral vein was cannulated for the administration of the ruthenium complex. A bolus of the ruthenium compound (RR or Ru₃₆₀) or the corresponding saline solution volume was administered to the rats. The heart was exposed by lateral left thoracotomy. Regional ischaemia was produced by a ligature (6-0 silk) around the left coronary artery, approximately 2 mm from its origin, according to the method of Selye et al. (1960). Artery occlusion was performed by placing a short rigid tube over the vessel and tying both firmly with a silk thread. In I/R and I/R+ Ru₃₆₀ groups myocardial ischaemia was confirmed by the appearance of regional cyanosis, akinesia or bulging in the epicardium distal to the artery occlusion and ST segment elevation. After 5 min of ischaemia, the silk was removed by cutting it carefully over the tube to restore blood flow to the myocardium. Reperfusion was confirmed by the colour change in the ventricular surface, from cyanosis to hyperaemia and by the onset of ventricular tachycardia (VT). The heart was reperfused for 5 min, in accordance with previous studies, to induce cardiac damage, characterized by a higher incidence of reperfusion-induced VT and ventricular fibrillation (VF) (Manning and Hearse, 1984; Hagar et al., 1991; Arteaga et al., 1992; Bobadilla et al., 2001; Parra et al., 2005). The incidence and time course of arrhythmias were compared between groups and their classification was established in agreement with the Lambeth Convention (Walker et al., 1988).

Rats that developed arrhythmias before the ischaemia or VF immediately after ischaemia were discarded and replaced. Thus, all analyses only represent animals that survived the whole procedure.

Measurements of mitochondrial integrity

After reperfusion, heart tissue from the left ventricle was minced and homogenized in isolation medium, containing (in mM) KCl (125), ethylenediaminetetraacetic acid (1) and N-2-hydroxyl piperazine-N′-2-ethane sulphonic acid (HEPES)-HCl (10), pH 7.3. The mitochondrial fraction was obtained by differential centrifugation, as previously described, by using the protease Nagarse (García-Rivas et al., 2005). Mitochondrial oxygen consumption was measured using a Clark-type oxygen electrode (Yellow Springs Instruments, OH, USA). The experiments were carried out in 1.5 ml of assay medium, containing (in mM) KCl (125), HEPES-HCl (10) and KH₂PO₄-TRIS (3), pH 7.3. State 4 respiration was evaluated in the presence of 10 mM succinate, plus 1 μg ml⁻¹ rotenone. State 3 respiration was measured after addition of 200 μM ADP. Respiratory control index (RC) was calculated as the ratio between state 3 and state 4 rates. ADP/O ratio was calculated as (nmol) of added ADP per (ng) of oxygen consumed during state 3 respiration.

Mitochondrial aconitase activity [E.C.4.2.1.3] was determined spectrophotometrically, by monitoring the disappearance of cis-aconitate at 240 nm (ɛ=3.6 mM⁻¹ cm⁻¹) (Hoerter et al., 2004). One mIU was defined as the amount of enzyme that consumed 1 nmol cis-aconitate min⁻¹. Protein content was measured by the Lowry method (1951).

Measurement of mCaU activity

Mitochondrial calcium uptake was measured with ⁴⁵CaCl₂ (specific activity 1000 cpm nmol⁻¹) using the filtration technique. Briefly, 0.5 mg of mitochondria were incubated in assay medium at the indicated times. Aliquots were withdrawn and filtered through Millipore filters of 0.45 μm pore size. Non-entrapped ⁴⁵Ca²⁺ was washed with 0.1 M KCl and the radioactivity retained in the filter was measured in a scintillation counter (Beckman, CA, USA). The assay medium contained (in mM) KCl (125), HEPES-HCl (10), 10 succinate (10), KH₂PO₄-TRIS (3), ethyleneglycol tetraacetate (0.5), 1 μg ml⁻¹ rotenone and 50 μM free calcium, calculated by using the Chelator program (Th. Schoenmakers, Nijmegen, Netherlands), pH 7.3.

Determination of Ru₃₆₀ concentrations in blood and myocardial tissue

Anaesthetized control rats (not subjected to the I/R protocol) under assisted respiration were treated with the protective dose of Ru₃₆₀ (50 nmol kg⁻¹) then, at the indicated times blood aliquots (500 μl) were obtained from the left ventricular cavity, before the hearts were removed from the rat. The hearts were mounted in a Langendorff apparatus as previously described and washed for 10 min with cold Krebs–Henseleit Buffer (García-Rivas et al., 2005). Then, the hearts were lyophilized and digested using Suprapure HNO₃, HCl and 30% H₂O₂ (6:2:1) (Merck Darmstandt, Germany). For each inductively coupled plasma optical emission spectroscopy (ICP-OES) determination, 1 g of cardiac dry tissue was required (three or four different hearts).

Ruthenium content was analysed by ICP-OES at 240.272 nm single wavelength in a Simultaneous Optima 4300 DV apparatus (Perkin Elmer, CT, USA). Ruthenium standards (4–15 μg l⁻¹) were prepared from primary pure standards (1000 μg l⁻¹) (Perkin Elmer, CT, USA). For each sample group, independent calibration curves and blanks were prepared. No spectral interferences were detected at 240.272 nm. Ru₃₆₀ content was calculated on the basis of its reported molecular weight, that is, 550.8 g mol⁻¹.

Data expression and analysis

Data are expressed as the mean±s.e. Statistical analysis was by Student's t-test. The data for heart rate, blood pressure, duration time of arrhythmias, mitochondrial activities and calcium transport were compared between the control, I/R and I/R+Ru₃₆₀ groups. A P-value of <0.05 was considered statistically significant.

Results

Effect of Ru₃₆₀ on functional recovery of rat hearts after ischaemia

A striking feature in myocardial reperfusion is a considerable increase in the appearance of cardiac arrhythmias (Tsuchihashi and Curtis, 1991; del Monte et al., 2004). Particularly, VF has been used as a criterion of potential lethal damage induced by reperfusion injury (Roh et al., 2005). In this context, we found that I/R rats showed a 78% incidence of VF, whereas Ru₃₆₀ administration gradually diminished this incidence, until VF disappeared at doses between 25 and 50 nmol kg⁻¹ (Figure 2). Further experiments with Ru₃₆₀ were performed using a dose of 50 nmol kg⁻¹. Next, we compared the effectiveness of this compound against RR, a related and widely used inhibitor of calcium uptake, with known cardioprotective properties in different models (Ferrari et al., 1982; Carry et al., 1989; Miyamae et al., 1996). As observed in Figure 2, RR treatment decreased the incidence of VF and enhanced myocardial recovery only at higher doses (5 μmol kg⁻¹), in accordance with previous findings (Carry et al., 1989).

Figure 2.

Dose-dependent protective effect of ruthenium complexes on the incidence of VF in reperfused rat hearts. Incidence of VF at the fifth minute of reperfusion. The shaded column represents the values of I/R rats; solid columns represent the values from rats treated with Ru₃₆₀ (I/R+Ru₃₆₀). Open columns represent the values from rats treated with RR. n=20, for I/R and I/R+Ru₃₆₀ rats. For RR treatment, n=3.

Figure 3 shows the temporal ABP in Ru₃₆₀-treated rats. Negative numbers represent the last 3 min of the 30-min period after drug injection and before artery ligation. We show these time points to demonstrate that Ru₃₆₀ did not elicit haemodynamic or arrhythmic effects, at least during the 30 min before the ischaemia. In I/R and I/R+ Ru₃₆₀ groups a discrete drop in the ABP was observed during ischaemia. However, in I/R rats the ABP decreased approximately 60% during the reperfusion, whereas in I/R+Ru₃₆₀ rats, the ABP was maintained. Diminished ABP correlated with an increase in heart rate in I/R rats, indicating reperfusion-induced haemodynamic dysfunctions. There was no significant difference in the total number of arrhythmias that occurred during ischaemia between I/R and Ru₃₆₀-treated rats (data not shown). However, to determine the impact of Ru₃₆₀-treatment on the development of reperfusion-induced cardiac electrical abnormalities, we analysed the ventricular arrhythmias in I/R and I/R+Ru₃₆₀ rats; normal beats, VT and VF were studied. In the I/R+Ru₃₆₀ rats the incidence of arrhythmias was significantly modified and the sinus rhythm was recovered at the first minute of reperfusion. At 5 min of reperfusion normal beats reached about 90% of the total beats per minute, in contrast to the results in I/R rats in which normal beats represented 15% of the total number of beats (Figure 4). Notably, after 3 min of reperfusion, VT and VF were totally absent in I/R+Ru₃₆₀ rats.

Figure 3.

Effect of Ru₃₆₀ on ABP in reperfused rats. Time course analysis of ABP in control, I/R and I/R+Ru₃₆₀ (50 nmol kg⁻¹) rats. Values are the mean of at least 25 different experiments±s.e. ^*P⩽0.05, significantly different vs control and ^†P⩽0.05 vs I/R.

Figure 4.

Electric cardiac profile of I/R+Ru₃₆₀ rats. Analysis of the duration of sinus rhythm during reperfusion. Open columns represent I/R and solid columns correspond to Ru₃₆₀-treated rats. Values are the mean of at least 25 different experiments±s.e. ^*P⩽0.05, significantly different vs Ru₃₆₀-treated rats.

The effect of Ru₃₆₀ on cardiac mitochondrial integrity after reperfusion

A growing body of experimental evidence supports the idea that mitochondria contribute to cardiac dysfunction and myocyte injury in the pathophysiology of ischaemia–reperfusion (Lesnefsky et al., 2001). Hence, we investigated cardiac mitochondria integrity of Ru₃₆₀-treated rats subjected to reperfusion injury. Mitochondrial respiratory activity was measured in the presence of succinate as substrate (Table 1). Mitochondria from I/R ventricles exhibited a 45% reduction in state 3 respiration rate, compared to control mitochondria, whereas respiratory rates in mitochondria obtained from I/R+Ru₃₆₀ ventricles did not show any change. There was no significant difference between the two groups in state 4 respiration rates. RC, an indicator of the mitochondrial electron transport coupling to ADP phosphorylation, was calculated to determine mitochondrial integrity. RC value of control mitochondria was 6.0±0.8, whereas in mitochondria isolated from reperfused ventricles, this value diminished to 3.5±0.6. In contrast, reperfusion did not affect the RC in I/R+Ru₃₆₀ ventricles. The ADP/O indexes for mitochondria isolated from I/R and from I/R+Ru₃₆₀ ventricles were 0.73±0.3 and 1.4±0.5, respectively; this difference was statistically significant (P⩽0.05). ADP/O values for control and I/R+Ru₃₆₀ mitochondria did not change.

Table 1.

Respiratory activity in mitochondria isolated from control, I/R and IR+Ru₃₆₀ rat ventricles

	Control	I/R	I/R+Ru360
State 3 (nmol O min−1 mg−1)	372±34† (n=5)	205±18* (n=4)	358±45† (n=4)
State 4 (nmol O min−1 mg−1)	61±13 (n=5)	48±8 (n=4)	63±16 (n=4)
ADP/O	1.32±0.4† (n=5)	0.73±0.3* (n=4)	1.4±0.5† (n=4)
RC	6.0±0.8† (n=5)	3.5±0.6* (n=4)	5.6±0.7† (n=4)

Abbreviation: I/R, ischaemia/reperfusion group; I/R+Ru₃₆₀, Ru₃₆₀-treated group; RC, respiratory control.

Mitochondrial respiratory activity was determined in a standard buffer. Values are the mean±s.e.

^*P⩽0.05 significantly different vs control

^†P⩽0.05 vs I/R.

It is well known that the oxidative damage produced during reperfusion affects mitochondrial integrity, affecting important enzymatic activities, so we measured mitochondrial aconitase activity as evidence of such damage. Aconitase activity is inversely proportional to the amount of O₂^−• produced during oxidative stress (Hoerter et al., 2004). In I/R mitochondria, aconitase activity decreased significantly (35%) as compared to control mitochondria. Interestingly, aconitase activity was protected against oxidative damage in I/R+Ru₃₆₀ mitochondria (Figure 5). We determined that Ru₃₆₀ has no ROS-scavenger properties by assessing thiobarbituric acid reactive substances content in control mitochondria subjected to oxidative stress, as previously described (García et al., 2005) (data not shown).

Figure 5.

Effect of Ru₃₆₀ treatment on the mitochondrial aconitase activity of coronary artery-ligated rat hearts. Mitochondrial aconitase activity was measured in mitochondria obtained from control, I/R and I/R+Ru₃₆₀ (50 nmol kg⁻¹) hearts. Values given are the mean of at least four different experiments±s.e. ^*P⩽0.05, significantly different vs control and ^†P⩽0.05 vs I/R.

Effect of Ru₃₆₀ on mCaU activity

To assess the effect of Ru₃₆₀ on the mCaU activity, we investigated calcium uptake in heart mitochondria from control, I/R and Ru₃₆₀-treated rats. Linear fitting of initial uptake velocities showed a slower calcium influx in mitochondria isolated from Ru₃₆₀-treated hearts compared with control and I/R mitochondria, suggesting an interaction between Ru₃₆₀ and its mitochondrial target. A longer time course analysis showed that in mitochondria from I/R ventricles, calcium uptake was followed by a rapid release, probably due to the opening of the mPTP, whereas calcium was maintained inside mitochondria isolated from control and Ru₃₆₀-treated rats (Figure 6a). In parallel experiments, we measured mitochondrial membrane potential in mitochondria from each group. As expected, I/R mitochondria developed lower membrane potentials than control or Ru₃₆₀-treated mitochondria. Calcium addition induced a transitory mitochondrial membrane depolarization in control and I/R+Ru₃₆₀ mitochondria, whereas in I/R mitochondria it promoted an irreversible drop in the transmembrane potential, indicating the opening of the mPTP (data not shown).

Figure 6.

Ru₃₆₀ treatment diminishes the initial mitochondrial calcium uptake rate. (a) Time course analysis of mitochondrial calcium uptake. In mitochondria from control, I/R and I/R+Ru₃₆₀ (50 nmol kg⁻¹) rat hearts. The insert shows the statistical analysis of initial calcium influx rate (nmol Ca²⁺ mg⁻¹). ^*P⩽0.05 significantly different vs control and ^†P⩽0.05 vs I/R. (b) CSA inhibits the mPTP in mitochondria from I/R hearts. Calcium transport was measured in isolated heart mitochondria from I/R rats in the presence of 1 μM CSA and without CSA. The data represent the mean of at least five different hearts (a) and four experiments (b)±s.e.

Calcium release in I/R mitochondria was prevented by the addition of cyclosporine A (CSA) in the assay medium (Figure 6b). Under this condition, calcium accumulation increased by 60%.

Ru₃₆₀ accumulation in heart tissue

To quantify Ru₃₆₀ accumulation in blood and heart tissue, we measured total ruthenium content 4 and 30 min after Ru₃₆₀ administration. Ruthenium content, 4 min after Ru₃₆₀ administration was 0.41±0.03 μg ml⁻¹ (n=4) in blood; this value diminished to 0.32±0.01 μg ml⁻¹ (n=4), 26 min later. In the cardiac tissue, ruthenium was undetectable at early administration times (4 min), but at 30 min, ruthenium content increased to 1±0.35 μg g⁻¹ dry tissue (n=3). Similar to previous measurements of protein content and distribution in myocardial cells (Idell-Wenger et al., 1978; Vinnakota and Bassingthwaighte, 2004), we calculated a concentration of 2.1±0.45 pmol Ru₃₆₀ mg⁻¹ of protein of myocardial tissue.

Discussion and conclusions

Calcium homeostasis undergoes fluctuations in balance during reperfusion, largely owing to the release of calcium from intracellular stores, particularly from the sarcoplasmic reticulum (Krause et al., 1989; Temsah et al., 1999). Experimental observations of calcium signal transmission between endoplasmic reticulum and mitochondria suggest the existence of a stable mitochondria–reticulum interaction, where mitochondria could accumulate a large fraction of the calcium released through the ryanodine receptor and the IP₃ receptor (Hajnoczky et al., 2000). According to this hypothesis, confocal image analysis provides evidence of high calcium concentration microdomains, susceptible to being sensed by mitochondria (Filippin et al., 2003).

As calcium accumulation in mitochondria has been proposed to play a key role in triggering cellular damage in the reperfused heart (Miyata et al., 1992; Miyamae et al., 1996; García-Rivas et al., 2005), we suggest that interventions reducing mitochondrial calcium overload would prevent mPTP opening and hence, membrane potential depolarization, matrix swelling and abolition of ATP synthesis, events that concur with the incidence of arrhythmias and haemodynamic dysfunction elicited by reperfusion in a whole-rat model.

The compound used in this study is a mCaU inhibitor, which permeates slowly into the cell, and specifically inhibits mitochondrial calcium uptake in intact cardiomyocytes and in isolated heart. Matlib et al. (1998), showed that 1 μM ¹⁰³Ru₃₆₀ was taken up by myocardial cells and accumulated in the cytosol in a biphasic manner. A rapid accumulation phase was observed, possibly related to binding to the cell surface, whereas a second slow phase could be due to intracellular accumulation. They calculated a final concentration after the slow phase of 3 pmol ¹⁰³Ru₃₆₀/10⁶ cells. At this concentration total inhibition of calcium uptake into mitochondria was observed in situ, in single voltage-clamped myocytes. Experiments from our group in isolated hearts, showed that free mitochondrial matrix calcium from Ru₃₆₀-treated hearts is diminished as compared to [Ca²⁺]_m in control hearts, indicating that Ru₃₆₀ targets the mCaU (García-Rivas et al., 2005). Also, a quite recent report suggests the involvement of the mCaU in cardioprotection; Ru₃₆₀ (10 μM) treatment of isolated hearts provides cardioprotective effects and the mitochondria obtained from those hearts are resistant to calcium-induced swelling (Zhang et al., 2006). With regard to the permeation properties of this poli-charged compound and its ability to reach the mitochondrial membranes, policationic copper-based antineoplastic drugs have been demonstrated to affect mitochondrial metabolism when perfused into the isolated heart (Hernández-Esquivel et al., 2006).

In the present study, we showed that Ru₃₆₀ treatment suppressed arrhythmias and haemodynamic dysfunction elicited by reperfusion, and prevented mPTP opening, by a mechanism possibly related to the diminution of mitochondrial calcium overload (Figure 6). Although many mechanisms have been proposed to explain the development of reperfusion arrhythmias, calcium overload is one of the main factors promoting its generation. High intracellular calcium induces electrical effects on the action potential (AP), such as inward currents and delayed after depolarizations in the pacemaker cells, which lead to VT and VF (Bers, 2002). It has also been pointed out, that there is a direct connection between loss of mitochondrial function and alterations in the cellular AP (O'Rourke, 2000). Evidence for the involvement of post-ischaemic electrical dysfunction and mitochondrial bioenergetics has been obtained from experiments with inhibitors of the mitochondrial benzodiazepine receptor (a putative component of the mPTP), which block depolarization of the mitochondrial membrane potential and prevent reperfusion arrhythmias (Akar et al., 2005). These findings, together with our own results indicate that I/R-related arrhythmias could be, in part, a consequence of the failure of the mitochondrial network to maintain the membrane potential at reperfusion as a consequence of mPTP opening, induced by calcium overload.

A subpopulation of mitochondria that undergoes irreversible mPTP opening, would be totally disrupted and would not be recovered in the mitochondrial pellet. This would account for the mitochondrial yields always being lower in reperfused hearts than in control- or drug-treated hearts. The ‘surviving' mitochondria recovered from reperfused hearts showed a higher sensitivity to mitochondrial calcium overload, as shown in Figure 6. This increased sensitivity, compared to control- and drug-treated mitochondria, could be a reflection of their inability to regenerate the membrane potential, possibly due to loss of adenine nucleotides or to oxidative stress damage to the mitochondrial respiratory complexes. As the I/R mitochondria undergo more permeability transitions, the more the effect becomes additive, mainly reflecting calcium extrusion. This mechanism is in agreement with that proposed for the propagation of permeability changes, where the local liberation of calcium from mitochondria triggers propagating waves of calcium-induced Ca²⁺ release in the entire mitochondrial network. (Pacher and Hajnoczky, 2001). The finding that CSA addition to I/R mitochondria (panel b in the same figure) reversed this effect, indicates that the major release pathway involved is the mPTP. Mitochondria from drug-treated hearts did not show such sensitivity and were able to accumulate more calcium than I/R mitochondria, but as the mitochondrial uptake pathway remained partially blocked, calcium accumulation was diminished.

We detected 2.1 pmol Ru₃₆₀ accumulated per mg of protein of myocardial tissue in drug-treated hearts by using ICP-OES, this concentration is in the low range of K_D values reported for in vitro mCaU inhibition (Ying et al., 1991; Matlib et al., 1998; Zazueta et al., 1999), and should only produce a partial inhibition of the mCaU. In this respect, it is not always possible to compare data obtained from in vitro studies with those from in vivo models. It has been demonstrated that many factors are critical when correlating the true intrinsic potency of a drug or inhibitor in vivo with in vitro determinations; among others are the concentration of the inhibitor at the site of its metabolic activity, tissue specificity and drug metabolism (Prueksaritanont et al., 1997; Schmider et al., 1999).

However, the assumption that Ru₃₆₀ reaches the mitochondrial membranes has been supported by results from our own group that demonstrated that Ru₃₆₀ actually gets inside the mitochondria when perfused into isolated hearts. In that study, we perfused increasing concentrations of this compound and observed a dose-dependent response inhibition of the mitochondrial calcium uptake (García-Rivas et al., 2005).

Other drugs, such as diazoxide (Wang et al., 2001) and RR (Ferrari et al., 1982; Carry et al., 1989; Miyamae et al., 1996) have been used as modulators of the mitochondrial calcium content. Diazoxide is a mitochondrial K⁺-ATP channel opener, that exerts an impressive recovery from the effects of reperfusion in isolated hearts, when used at low concentrations (30–100 μM) (Garlid et al., 1997; Wang et al., 2001; Hausenloy et al., 2004). Whereas, in a whole-rat model, this compound only partially protects at concentrations up to 625 μM (Fryer et al., 2000). RR also exerts protection in isolated hearts at 0.025–10 μM (Ferrari et al., 1982; Miyamae et al., 1996), but in a whole-rat model it shows a protective effect at doses close to 30 μM (Carry et al., 1989). At these concentrations both compounds show collateral effects, not only in the heart, but in other organs (Balazs et al., 1975; Belmar et al., 1995; Silvani et al., 2004). Mitochondrial integrity can also be maintained after reperfusion by inhibiting the opening of the mPTP. In this context, a wide variety of molecules that inhibit this mega-channel, including CSA, have been used as protectors against reperfusion injury (Arteaga et al., 1992; Duchen et al., 1993). Other examples are sanglifehrin A (Clarke et al., 2002) and more recently NIM811 (Argaud et al., 2005) and octylguanidine (Parra et al., 2005). However, we proposed that prevention from calcium overload, instead of closing the mPTP could be a more effective strategy for the prevention of reperfusion injury, as ROS production in mitochondria appears to be mediated by an increase in [Ca²⁺]_m. The results obtained from measuring the activity of aconitase, a mitochondrial marker of oxidative stress, further support this idea. I/R mitochondria showed a diminution of mitochondrial aconitase activity and such inactivation was partially abolished by Ru₃₆₀-treatment, indicating that calcium accumulation into the mitochondrial matrix increases ROS production, an effect that has also been observed by other groups using RR (Petrosillo et al., 2004; Votyakova and Reynolds, 2005). Although the exact mechanism by which [Ca²⁺]_m induces ROS production is not clear (Brookes et al., 2004), one possible explanation is that calcium induces mitochondrial membrane depolarization, enhancing ROS production (Cadenas and Boveris, 1980; Turrens, 1997). Another possibility is that Ca²⁺-binding to cardiolipin molecules dissociates cytochrome c from the inner membrane, inhibiting the respiratory complex III (ubiquinol cytochrome c oxidoreductase) and increasing ROS generation in the ubiquinone cycle (Grijalba et al., 1999; Petrosillo et al., 2004).

In conclusion, our results indicate that Ru₃₆₀ increases the functional recovery of hearts subjected to ischaemia–reperfusion and maintains the mitocondrial integrity when perfused into a whole rat. The mechanism by which this compound prevented the damage could be related to the partial inhibition of the mitochondrial calcium transport. In this respect, it would be very interesting to explore the potential of Ru₃₆₀ as an alternative novel drug for use in reperfusion therapy.

Abbreviations

ABP

arterial blood pressure

AP

action potential

[Ca²⁺]_c

cytosolic calcium concentration

[Ca²⁺]_m

mitochondrial calcium concentration

CSA

cyclosporine A

ICP-OES

inductively coupled plasma optical emission spectroscopy

I/R

ischaemia/reperfusion group

I/R+Ru₃₆₀

Ru₃₆₀-treated group

mCaU

mitochondrial calcium uniporter

mPTP

mitochondrial permeability transition pore

RC

respiratory control

ROS

reactive oxygen-derived species

RR

ruthenium red

Ru₃₆₀

oxygen-bridged dinuclear ruthenium amine complex

VF

ventricular fibrillation

VT

ventricular tachycardia.

Conflict of Interest

The authors state no conflict of interest.