Na+/H+ exchanger-1 inhibitors decrease myocardial superoxide production via direct mitochondrial action

Articles in PresS. J Appl Physiol (September 18, 2008). doi:10.1152/japplphysiol.90616.2008

Na+/H+ exchanger-1 inhibitors decrease myocardial superoxide production via direct mitochondrial action. Carolina D. Garciarena; Claudia I. Caldiz; María V. Correa; Guillermo R. Schinella; Susana M. Mosca; Gladys E Chiappe de Cingolani; Horacio E. Cingolani; Irene L. Ennis Centro de Investigaciones Cardiovasculares. Facultad de Ciencias Médicas, Universidad Nacional de La Plata. La Plata, Buenos Aires. ARGENTINA

Running title: NHE-1 inhibitors and myocardial ROS production Key words: NHE-1, reactive oxygen species, mitochondria

Mailing address:

Dr. Irene L. Ennis Centro de Investigaciones Cardiovasculares, Facultad de Ciencias Médicas, UNLP 60 y 120 (1900) La Plata, Argentina. Phone/FAX: (54-221) 483-4833 E-mail: [email protected]

GECC, SMM, HEC, ILE are Established Investigators of Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Argentina. CDG and MVC are Fellows of Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Argentina.

Copyright © 2008 by the American Physiological Society.

ABSTRACT The possibility of a direct mitochondrial action of Na+/H+ exchanger-1 (NHE-1) inhibitors decreasing reactive oxygen species (ROS) production was assess in cat myocardium. Angiotensin II and endothelin-1 induced an NADPH oxidasedependent (NOX) increase in anion superoxide (O2-) production detected by chemiluminescence. Three different NHE-1 inhibitors—cariporide, BIIB723, and EMD87580—with no ROS scavenger activity prevented this increase. The mitochondria appeared to be the source of the NOX-dependent ROS released by the “ROS-induced ROS release mechanism” that was blunted by the mitochondrial KATP channel (mKATP) blockers 5-hydroxydecanoate and glibenclamide, inhibition of complex I of the electron transport chain with rotenone and inhibition of the permeability transition pore (MPTP) by cyclosporin A. Cariporide also prevented O2- production induced by the opening of mKATP with diazoxide. Ca2+-induced swelling was evaluated in isolated mitochondria as an indicator of MPTP formation. Cariporide decreased mitochondrial swelling to the same extent as cyclosporin A and bongkrekic acid confirming its direct mitochondrial action. Increased O2- production, as expected, stimulated ERK1/2 and p90RSK phosphorylation. This was also prevented by cariporide giving additional support to the existence of a direct mitochondrial action of NHE-1 inhibitors in preventing ROS release. In conclusion, we report a mitochondrial action of NHE-1 inhibitors that should lead us to revisit or reinterpret previous landmark observations about their beneficial effect in several cardiac diseases, such as ischemia/reperfusion injury and cardiac hypertrophy and failure. Further studies are needed to clarify the precise mechanism and site of action of these drugs in blunting MPTP formation and ROS release.

2

INTRODUCTION The myocardial Na+/H+ exchanger-1 (NHE-1) plays a critical role in intracellular pH regulation during physiological and pathological processes. NHE-1 catalyzes the electroneutral exchange of intracellular protons for extracellular Na+ by a secondary active process that takes energy from the transmembrane Na+ gradient. Although NHE-1 is relatively quiescent under basal conditions, it is activated in many pathologic conditions, like ischemia/reperfusion injury and cardiac overload, and after pharmacological interventions like angiotensin II (Ang II), endothelin (ET), and α-adrenergic stimulation, which exert their effect through the activation of several kinases [for a review, see (5)]. NHE-1 hyperactivity leads to intracellular accumulation of Na+, which induces intracellular Ca2+ overload via the Na+/Ca2+ exchanger (NCX) (14). One of the mechanisms by which NHE-1 inhibitors protect against ischemia/reperfusion injury may be the prevention of the intracellular accumulation of Na+ and Ca2+ that occurs because of NHE-1 hyperactivity (4, 26); this mechanism may also explain the protective effects of NHE-1 inhibitors in cardiac hypertrophy and failure (6, 12) and the mitochondrial death pathway (51). An increase in reactive oxygen species (ROS) production has been associated with the cardiovascular pathologies that NHE-1 inhibition affects beneficially (3, 8, 15, 22, 45, 49, 55). Interestingly, there are several reports that suggest a link between ROS and NHE-1 activation (41, 46, 50). Moreover, the experiments performed by Wei et al. (54) indicate that ROS induce activation of the MEK-ERK1/2-p90RSK pathway, which increases NHE-1 activity via a phosphorylation process. In line with this hypothesis are the results reported by Ruiz-Meana et al. (47) suggesting that cariporide may inhibit the mitochondrial NHE (MNHE) delaying mitochondrial matrix acidification and ATP depletion. Therefore, downregulation of mitochondrial ROS induced by NHE-1 inhibitors could exert additional beneficial effects to those derived from the inhibition of the sarcolemmal NHE-1. In the present study we test the hypothesis that NHE-1 inhibitors have a mitochondrial effect leading to a decrease in ROS production.

3

Methods All procedures followed during this investigation conform to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996) and to the guidelines laid down by the Animal Welfare Committee of La Plata School of Medicine. Cats (body weight 3–4 kg) were anaesthetized by intraperitoneal injection of sodium pentobarbital (35 mg/kg body weight) and hearts rapidly excised when plane three of phase III of anaesthesia was reached. Preparation of cardiac slices: Cardiac tissue slices from the left ventricle (1 x 5 mm) were obtained and kept at 4 °C until assayed as previously described (10). Assay buffer consisted of Krebs-HEPES buffer of the following composition (in mmol/L): 118.3 NaCl; 4.7 KCl; 1.8 CaCl2; 1.2 Mg2SO4; 1.0 K2HPO4; 25 NaHCO3; 11 glucose; 20 HEPES; pH 7.4 at 37 °C. Cardiac slices were kept in the assay buffer during 30 min in the presence of different drugs in a metabolic incubator under 95% O2/ 5% CO2 before measuring anion superoxide (O2-) production. Measurement of O2- production: O2- production by cat cardiac slices was measured by the lucigenin-enhanced chemiluminescence method as previously described (10). The increase in O2- production was expressed as the difference from the control after 15 minutes in the presence of lucigenin. To explore whether the NHE-1 inhibitors have O2- scavenging activity, experiments were performed based on the Nishikimi et al. (42) method. O2- production was measured with 0.6 µmol/L phenazine methosulphate (PMS) in Krebs-HEPES buffer containing 5 μmol/L lucigenin with and without the NHE-1 inhibitors at the concentration used in our experiments. The reaction was started by addition of three different concentrations of NADH (0.5, 1 and 2 mmol/L). Determination of ERK1/2 and p90RSK phosphorylation: These experiments were performed using isolated cat ventricular myocytes instead of ventricular slices because by this way the variability was significantly reduced. Cat ventricular myocytes were isolated according to the technique described previously (1), incubated for 30 minutes in 35 mm culture dishes (~ 5 x 105 cells/dish) with or without the drugs being explored, and preincubated for 15 minutes with the

4

inhibitors being assayed (losartan, apocynin, 5-hydroxydecanoate (5-HD), glibenclamide, rotenone, and NHE-1 inhibitors). At the end of the incubation myocytes were homogenized in lysis buffer and the relative amounts of Pp90RSK and P-ERK1/2 were determined by Western blot as previously described (10). Total p90RSK and ERK-2 as a loading control were assayed. Isolation of cardiac mitochondria: Hearts were homogenized in ice-cold buffer containing (in mmol/L) 250 sucrose, 10 Tris·HCl, 1 PMFS, pH 7.4. The homogenate was centrifuged at 1000×g for 3 min at 2 °C, and then the supernatant was centrifuged at 10000×g for 10 min at 2 °C. The resulting pellets were used for determination of mitochondrial swelling. Mitochondrial swelling determination: The detection of changes in 90° light scattering in mitochondrial suspensions is a convenient method to study mitochondria permeability transition pore (MPTP) opening. Mitochondrial swelling caused by the influx of solutes through the opened MPTP decreases light scattering. After 5 min of preincubation at 37°C in a medium containing (in mmol/L) 120 KCl, 20 MOPS, 10 Tris–HCl and 5 KH2PO4 (pH 7.4); 200 μmol/L CaCl2 was added to induce MPTP opening (7). The decrease in light scattering was detected with a temperature-controlled Aminco Bowman Series 2 spectrofluorometer operating with continuous stirring at excitation and emission wavelengths of 520 nm. Light scattering decrease was calculated for each sample as the difference between the values before and after the addition of CaCl2. The experiments were performed in the absence and in the presence of the MPTP opening inhibitor cyclosporine A (CsA, 1 μmol/L), bongkrekic acid (BKA, 10 μmol/L) the NHE-1 inhibitor cariporide (10 μmol/L), or the combination of MPTP inhibition and cariporide. The effect of the inhibitors was expressed as the percentage of decrease in light scattering compared to that induced by 200 μmol/L CaCl2. Chemicals: All drugs used in the present study were analytical reagent. Ang II, ET-1, lucigenin, N-(2-mercaptopropionyl)-glycine (MPG), L-NAME, allopurinol, 5-HD, rotenone, CsA and BKA were purchased from Sigma; apocynin (Fluka); Losartan and EMD87580 (Merck); glibenclamide (RBI, USA); HOE642 (cariporide, Aventis); BIIB723 (Boehringer-Ingelheim). Either Krebs-HEPES

5

buffer or dimethyl sulfoxyde (DMSO) were used to prepare drug dilutions. The final DMSO concentration, when used, was kept < 0.5%. Statistics: Data are expressed as mean ± SEM. Differences between groups were assessed by one-way ANOVA followed by Student-Newman-Keuls test. P < 0.05 was considered significant.

6

RESULTS Ang II induces an NADPH oxidase (NOX)-dependent increase in ROS production A dose-dependent increase in myocardial O2- production induced by Ang II was detected by the lucigenin-enhanced chemiluminescence method. Average data for each Ang II concentration assayed is shown in Figure 1. No significant difference between background signal (lucigenin containing buffer alone) and non stimulated tissue (lucigenin containing buffer plus cardiac slices) was detected (data not shown), indicating that basal O2- production by feline myocardium is within the limits of detection of this lucigenin assay. The inability of this detection method to detect the basal production of O2- by non-stimulated tissue has been reported previously (19). To perform most of the experiments, we chose the concentration of 1 nmol/L Ang II on the basis of our previous experience with this drug concentration (16) and also because it is the one that better resembles the physiological concentration range of Ang II in interstitial myocardium (18). In the cardiovascular system, the main sources of O2- are the Nox-based reduced nicotinamide-adenine dinucleotide phosphate (NADPH) oxidases (NOX), xanthine oxidase, the mitochondrial electron transport system and, under certain circumstances, nitric oxide synthase. Figure 2A shows that the increase in O2- production induced by 1 nmol/L Ang II was attributable to the stimulation of NOX through the activation of the AT1 receptor, since it was canceled by the AT1 receptor blocker losartan (1 μmol/L) and by the NOX inhibitor apocynin (300 μmol/L). In addition, the stimulatory effect of Ang II on O2- production was not observed in the presence of the ROS scavenger MPG (2 mmol/L). Furthermore, the inhibition of nitric oxide synthase with L-NAME (1 mmol/L) and of the xanthine oxidase with allopurinol (10 mmol/L) did not affect the increase of O2induced by 1 nmol/L Ang II (Fig. 2B), indicating that it was not involved the uncoupling of nitric oxide synthase or the activation of xanthine oxidase in the Ang II-induced increase in ROS production under these experimental conditions. None of the inhibitors used had a detectable effect on the control chemiluminescence signal (insets of Figs. 2A and 2B).

7

Mitochondria emerges as the main source of the O2- detected The mitochondrial electron transport chain, especially at complex I and complex III, probably represents the major source of O2- production in cardiac myocytes. In vascular smooth muscle cells, mKATP channels have been implicated in Ang II-induced ROS production (33). To determine whether in the myocardium the opening of these channels is involved in the mitochondrial production of ROS induced by Ang II, cardiac slices were exposed to 5-HD (100 μmol/L) and glibenclamide (50 μmol/L), a specific and a non-selective mKATP channel blocker, respectively. The NOX-dependent increase in O2- production induced by Ang II was canceled by both 5-HD and glibenclamide (Fig. 2C). This finding clearly implicates mKATP channels in Ang II-mediated mitochondrial O2production, in agreement with previous reports (9, 57, 59). To further confirm the role of the mitochondrial as the source of O2generation, we evaluated the effect of Ang II in the presence of rotenone (10 μmol/L), an inhibitor of mitochondrial electron transport chain complex I. As shown in Figure 2D, Ang II did not increase O2- production in the presence of rotenone, indicating that mitochondrial electron transport chain complex I was responsible for the increased oxidative stress. We identified no effect of rotenone on O2- basal production (Fig. 2D). The experiment illustrated in Figure 2E shows that CsA, a potent inhibitor of MPTP opening, also prevented the rise in O2- production induced by 1 nmol/L Ang II, suggesting that MPTP opening is a necessary step for the NOX-induced mitochondrial release of ROS, as previously proposed (9). Therefore, Ang II, at the concentration used in the present experiments, appears to induce a NOX-dependent mitochondrial release of O2-, a find in agreement with recent reports (57). NHE-1 inhibitors prevent ROS release by directly targeting mitochondria Several reports from different authors, including our group, have suggested an interaction between mitochondria and NHE-1 inhibitors (22, 27, 28, 47, 51); in addition, the prevention of oxidative stress appears to be beneficial in most of the cardiovascular pathologies in which NHE-1 inhibitors

8

are effective (11, 15, 22, 35, 37, 38). Therefore, we analyzed the effect of three different NHE-1 blockers on the increased O2- production induced by Ang II. As shown in Figure 3A, administration of each of the NHE-1 blockers examined—cariporide (10 μmol/L), BIIB723 (10 μmol/L), and EMD87580 (5 μmol/L)—prevented the Ang II-induced rise in O2- production. None of these compounds had a detectable effect on O2- basal production (inset of Fig. 3A). However, to rule out the possibility that these compounds act as O2- scavengers, we induced O2- production in vitro by PMS and NADH in a range that included the values obtained with 1–100 nmol/L Ang II. Under these experimental conditions, we measured the chemiluminescence signal in the absence and presence of each of the above mentioned NHE-1 inhibitors. Figure 3B shows that there was no detectable change in O2- production when the NHE-1 inhibitors were added to the medium. Therefore, we can assume that these inhibitors did not exert a scavenger effect. Using the same experimental conditions as in the Ang II experiments, we examined the effect on myocardial O2- production of another Gq protein coupled receptor agonist, ET-1 (5 nmol/L), which is known to activate NOX (21, 36). We obtained results that were essentially the same as those elicited by Ang II. ET-1 induced a NOX-dependent mitochondrial release of O2- that was prevented by the non-selective ET receptor blocker TAK044 (5 μmol/L), the ETA receptor blocker BQ123 (10 μmol/L), the NOX inhibitor apocynin (300 μmol/L), the mKATP channel blockers 5-HD (100 μmol/L) and glibenclamide (50 μmol/L), and the NHE-1 inhibitor cariporide (10 μmol/L), as it is shown in Figure 4A. When we stimulated mitochondrial O2- production by directly inducing mKATP channel opening with diazoxide, we detected a dose-dependent rise in lucigenin chemiluminescence signal in cardiac slices (Fig. 4B). The increase in the chemiluminescence signal observed with 100 μmol/L diazoxide was of the same order of magnitude as that induced by 1 nmol/L Ang II and 5 nmol/L ET-1; in addition, it was prevented by the mKATP channel blocker 5-HD, by the NHE-1 inhibitor cariporide, and by CsA (Fig. 4C). These results are similar to those obtained by Kimura et al. (33) in rat vascular smooth muscle cells, in which they measured the effects of Ang II and diazoxide on ROS production.

9

Cariporide decreases Ca2+-induced mitochondrial swelling The above commented experiments in which the increase in O2production elicited by Ang II and diazoxide was blunted by CsA suggest that the opening of the MPTP was the source of the increase of ROS we detected. In order to get further evidence of the mitochondrial action of the NHE-1 inhibitors, experiments using isolated mitochondria were performed. The exposure of a mitochondrial suspension to 200 μmol/L CaCl2, which is known to induce MPTP opening and consequently mitochondrial swelling (7), evoked a large decrease in light scattering. As expected, the MPTP inhibitors CsA and BKA significantly decreased calcium-induced mitochondrial swelling, confirming that the increase in mitochondrial volume was attributable to the MPTP opening. Interestingly, cariporide also decreased mitochondrial swelling after the exposure to 200 μmol/L CaCl2 in a magnitude similar to that of CsA and BKA, again suggesting a direct mitochondrial action of the NHE-1 inhibitor. The combination of CsA or BKA with cariporide did not further decrease mitochondrial swelling, allowing us to speculate that these three drugs share a common mechanism of action on the mitochondria (Figure 5). The inhibition of mitochondrial ROS release prevents ERK1/2 and p90RSK phosphorylation There is evidence that ROS can act as signaling molecules to regulate cellular functions by activating protein kinase cascades (46, 48, 50). According to our data, in which the NHE-1 inhibitors blunted the Ang II-induced increase in ROS formation, the prevention of this kinase activation should be expected when Ang II is administered in the presence of the NHE-1 inhibitors. The data presented in Figure 6 show a significant increase in ERK1/2 and p90RSK phosphorylation after 30 minutes of incubation of isolated cat ventricular myocytes with Ang II. This effect was canceled by blocking AT1 receptors with losartan, by the NOX inhibitor apocynin, by scavenging ROS with MPG, by preventing mKATP channel opening with 5-HD or glibenclamide, by interfering with the mitochondrial electron transport chain with rotenone, and by the NHE-1

10

inhibitor cariporide. These data provide additional support for the idea that NHE1 blockers inhibit the NOX-dependent increase in mitochondrial O2- production, thereby preventing kinases activation.

11

Discussion To test the possible “anti-ROS effect” of NHE-1 inhibitors, we analyzed NOX-dependent mitochondrial O2- generation induced by two different Gcoupled receptor agonists, Ang II and ET-1. The NOX-dependent mitochondrial release of ROS is the basis of the so-called “ROS-induced ROS release” phenomenon proposed by Zorov et al. (59) and Kimura et al. (34). The latter group found evidence in smooth (33) and cardiac muscle (34) in which some effects of Ang II, including kinase activation, were suppressed by NOX inhibition with apocynin and by the mKATP blockade with 5-HD. We did not characterize the NOX isoform involved in the pathway leading to ROS formation; however, previous available evidence (23, 58) and the fact that apocynin suppressed ROS production and kinase activation in our study indicate that NOX 2 is the isoform involved. There is no clear evidence that sarcolemmal NOX-derived ROS interacts with the mitochondria, yet Zhang et al. (56) used reconstituted mKATP channels of bovine heart and demonstrated that O2- directly stimulates the opening of these channels. The three NHE-1 inhibitors we used blunted the increased mitochondrial ROS production and the redox activation of the kinases that are a known downstream target of ROS (46, 48). We demonstrated that these pharmacologic inhibitors did not have scavenger properties by performing experiments in which O2- was generated in vitro. The NHE-1 inhibitor cariporide blunted not only the increased O2- production induced by Ang II/ET-1 but also the production induced by opening the mKATP channel with diazoxide. In our experiments in cardiac muscle, the NHE-1 inhibitors suppressed the increased ROS formation induced by the opening of mKATP channels, a fact that favors a direct mitochondrial effect of these compounds. Even though we induced and measured an increase in O2- formation, O2- is quite unstable and is rapidly converted into H2O2 by several isoforms of superoxide dismutase (SOD), like EC-SOD, a membrane-bound extracellular SOD-, Mn-SOD in the mitochondrial matrix, and Cu-Zn-SOD in the interstitial space and cytosol. The activation of the previously mentioned kinases, ERK1/2 and p90RSK, probably occurs after O2- dismutation.

12

The increased O2- generation was reduced by CsA-triggered inhibition of MPTP formation in cardiac slices. The MPTP is a large conductance pore thought to be activated by ROS, by increased mitochondrial Ca2+ levels and by dissipation of the mitochondrial membrane potential (Δψm) (2, 30). The hypothesis of a direct mitochondrial action of NHE-1 inhibitors was confirmed by experiments performed in isolated mitochondria in which cariporide decreased CaCl2-induced mitochondrial swelling. The fact that quite similar results were obtained with the MPTP inhibitors CsA and BKA or with the combination of each MPTP inhibitor with cariporide allows us to speculate that these drugs share a common mechanism of action on the mitochondria. We would like to emphasize that a direct mitochondrial effect does not necessarily mean that the MNHE is targeted. Actions of NHE-1 inhibitors on mKATP channels (39), prevention of mitochondrial membrane potential loss (51), and recently actions on carbonic anhydrase (53) have already been reported. Further research is necessary to dissect the mitochondrial mechanism targeted by the NHE-1 inhibitors. Although the existence of MNHE is well documented, the presence of a specific mitochondrial isoform (NHE-6) is controversial (13, 43, 47). Regardless of the NHE isoform present in the mitochondria, whether the NHE-1 inhibitors target this mitochondrial exchanger remains controversial. Kapus and coworkers (31) reported a lacking or weak inhibitory action of amiloride analogues and derivatives on MNHE in isolated mitochondria. More recently, Hotta et al. (25) described a mitochondrial effect of NHE-1 inhibitors, and Ruiz-Meana et al. (47) suggested the possibility that cariporide may target MNHE. Karmazyn’s group in two recent publications (28, 29) concluded that the mitochondrial effect of NHE-1 inhibitors is indirect and mediated by the prevention of cytosolic Ca2+ overload through sarcolemmal NHE-1 inhibition (by altering intracellular Na+ concentration). Their most recent findings (28) reinforce their conclusion: they showed the lack of effect of an NHE-1 inhibitor on isolated mitochondria under carefully controlled conditions. We were unable to determine from the report whether or not under these experimental conditions the effect of NHE-1 inhibitors on mitochondrial Ca2+ and/or H+ is prevented. The experiments with isolated mitochondria in which we demonstrated the existence of a novel direct

13

action of the NHE-1 inhibitors were performed with cariporide, probably the most representative and widely used compound of this class and the one used by us to abolish the effect of Ang II on ERK1/2-p90RSK activation. Although the other two NHE-1 inhibitors used in the present work were not assayed in isolated mitochondria we did show that they shared with cariporide the inhibitory action on Ang II-induced mitochondrial ROS production. The explanation for the apparent contradiction with the lack of effect on isolated mitochondria reported for AVE-4890 is not apparent to us at present. We could speculate that NHE-1 inhibitors may act on the Ca2+ uniport, mitochondrial K+ channels (ATP and Ca2+), Na+/Ca2+ exchanger, or K+/H+ exchanger, or by some other mechanism (including MNHE) leading to changes in matrix Ca2+, H+, or ATP concentrations. Among these possibilities, the targeting of NHE-1 inhibitors to MNHE is obviously the first to be considered, although it is complicated to analyze. Under physiological conditions, the MNHE introduces cytosolic H+ into the mitochondrial matrix in exchange for mitochondrial Na+. Therefore, a decrease in exchanger activity should reduce H+ and increase Na+ concentration in the mitochondrial matrix and perhaps increase H+ concentration on the external side of the MPTP. The decrease in H+ matrix concentration would favor and not prevent MPTP formation (24), but the increase of H+ on the cytosolic side of the MPTP may inhibit pore formation. In connection with this scenario, the prevention of MPTP formation by acidosis in reperfusion after ischemia has been reported (17). An increase in mitochondrial Na+ would decrease the inwardly directed Na+ gradient, affecting the mitochondrial Na+/Ca2+ exchanger, other factors being constant (40). A decrease in Ca2+ efflux from the mitochondria would increase mitochondrial Ca2+ concentration and favor rather than reduce MPTP formation. Increasing the complexity of the analysis is the necessary consideration of whether inhibition of MPTP formation can be achieved by a direct inhibition of the MPTP rather than by the decrease in mitochondrial Ca2+ similarly to what occurs with CsA. Therefore, a careful analysis of the candidate mechanisms involved in MNHE inhibition and MPTP formation requires further research dissecting the different possibilities.

14

Although MPTP opening was initially related to irreversible myocardial injury, there is evidence that the duration and timing of the opening can determine the outcome of cell viability from necrosis to apoptosis (44) and perhaps can even trigger physiological intracellular signals (52). Whether or not to view ROS as “friend or foe” is a current subject of debate [for a review, see (20)]. ROS can be protective, inducing pre- and postconditioning (52), but they can also induce damage to the heart (32). This matter is out of the scope of the present study, but similarly to what was proposed for the MPTP opening, the effects may be related to the duration, timing, location, and extent of ROS release. Figure 7 schematizes the two-step mechanism previously proposed for the increase of ROS formation induced by a G-coupled receptor agonist and the possible mechanism involved in its reduction induced by NHE-1 inhibitors. In brief, we report here a direct mitochondrial “anti-ROS” action of NHE-1 inhibitors detected in cardiac slices and isolated mitochondria. Although we could not dissect the precise site of action of the inhibitors on the mitochondria, the knowledge of these effects opens new avenues of research and helps illuminate the reasons for some of the contradictory results.

15

Acknowledgements This work was partially supported by grants PICT 05-14565 (to I.L.E.); PICT 0525475 and PIP 5141 (to H.E.C.) and PICT 05- PICT 05-12480 (to G.E.Ch. de C.) from Agencia Nacional de Promoción Científica y Tecnológica and from Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Argentina.

16

Figure legends Figure 1: O2- production in cardiac tissue slices. 30 min incubation in the presence of Ang II induced a dose-dependent increase in O2- production. Values are the difference from the control after 15 minutes in the presence of lucigenin. Values are the mean ± SE. Figure 2: Source of Ang II-induced ROS production. A. Cardiac tissue slices incubated during 30 min in the presence of 1 nmol/L Ang II (n=34) induced a significant increase in O2- production that was prevented by losartan (Los, 1 μmol/L; n=8), apocynin (Apo, 300 μmol/L; n=7) and MPG (2 mmol/L; n=5). B: Inhibition of the nitric oxide synthase with L-NAME (1 mmol/L; n=17) and xanthine oxidase with allopurinol (Allo, 10 mmol/L; n=13) did not affect the rise in O2- production induced by Ang II. C: The Ang II-induced increase in O2production was canceled by two different mKATP channel blockers, the selective blocker 5-HD (100 μmol/L; n=10) and the non-selective glibenclamide (Gli, 50 μmol/L; n=6). D: No increase in O2- production was detected after stimulation with 1 nmol/L Ang II (n=8) in the presence of rotenone (Rot). E: MPTP formation inhibition suppressed the stimulatory action of Ang II on mitochondrial ROS production. CsA (0.5, 1 and 2 μmol/L) prevented the effect of Ang II (n=4). 2 μmol/L CsA did not affect control chemiluminiscence signal. Values are the difference from the control after 15 minutes in the presence of lucigenin expressed as the mean ± SE. None of the inhibitors used had an effect on the control chemiluminescence signal (inset in the figures). * p < 0.05 vs. all other groups, ANOVA. Figure 3: Effect of NHE-1 inhibitors on stimulated ROS production. A: The stimulatory effect of 30 min-incubation with Ang II on O2- production by cardiac tissue slices was prevented by three different NHE-1 inhibitors; cariporide (carip, 10 μmol/L; n=12), BIIB723 (BIIB, 1 μmol/L; n=3) and EMD87580 (EMD, 5 μmol/L; n=4). Values are the difference from the control after 15 minutes in the presence of lucigenin. The NHE-1 inhibitors did not have an effect on control

17

chemiluminescence signal (inset). * p < 0.05 vs. all other groups, ANOVA. B: O2- production was induced in vitro by PMS and NADH in a range that includes the values of chemiluminescence (in AU/min) obtained with 1–100 nmol/L Ang II. None of the NHE-1 inhibitors (BIIB, carip and EMD) had an effect on the detected levels of O2- induced by PMS and NADH (n=5). Figure 4: A: ET-1-induced ROS production. Incubation of cardiac tissue slices during 30 min in the presence of 5 nmol/L ET-1 (n=11) induced an increase of O2- that was prevented in the presence of BQ123 (BQ, 10 μmol/L; n=6), TAK044 (TAK, 5 μmol/L; n=6), apocynin (Apo, 300 μmol/L; n=4), 5-HD (100 μmol/L; n=6), glibenclamide (Gli, 50 μmol/L; n=5), and cariporide (carip, 10 μmol/L; n=6) * p < 0.05 vs. all other groups, ANOVA. None of the inhibitors used had an effect on the control chemiluminescence signal (inset) B: Opening of mKATP and ROS production. Diazoxide (Diaz), a mKATP channel opener, induced a dose-dependent rise in the lucigenin chemiluminescence signal. * p < 0.05 vs. 50 μmol/L Diaz , ANOVA. C: The increase in the chemiluminescence signal observed with 100 μmol/L Diaz (n=17) was of a similar magnitude to that induced by 1 nmol/L Ang II and 5 nmol/L ET-1, and it was prevented by 5-HD (100 μmol/L; n=5), carip (10 μmol/L; n=5) and CsA 2 μmol/L (n=5). * p < 0.05 vs. all other groups, ANOVA. Figure 5: Mitochondrial swelling induced by CaCl2. Panel A Typical experiment showing that cyclosporine A (CsA) and bongkrekic acid (BKA) significantly attenuated calcium-induced mitochondrial swelling and the decrease in light scattering in mitochondrial suspensions. Cariporide (carip) inhibited the decrease in light scattering in a similar magnitude to CsA (1 μmol/L) and BKA (10 μmol/L). Panel B Average results. The combination of both drugs, CsA or BKA with carip, did not show any greater effect (n=7). * p < 0.05 vs. CaCl2, ANOVA Figure 6: Ang II-induced phosphorylation of ERK 1/2 and p90RSK. Ang II (1 nmol/L) induced an increase in ERK1/2 (Panel A) and p90RSK (Panel B)

18

phosphorylation in isolated cat ventricular myocytes that was prevented by losartan (Los, 1 μmol/L), MPG (2 mmol/L), apocynin (Apo, 300 μmol/L), 5-HD (100 μmol/L), glibenclamide (Gli, 50 μmol/L), rotenone (Rot, 10 μmol/L), and cariporide (carip, 10 μmol/L) (n=4). No changes in total ERK-2 and p90RSK were observed. * p < 0.05 vs. all other groups, ANOVA.

Figure 7: Possible mitochondrial sites of action of NHE-1 inhibitors. The scheme shows the “two step” release of ROS through activation of G-coupled receptors and inhibition of the MPTP formation by NHE-1 inhibitors. These inhibitors may act upon different mitochondrial mechanisms, including MNHE. They may act through a decrease in mitochondrial Ca2+, H+, Δψ affecting the MPTP formation or altering the sensitivity to those factors to induce MPTP formation.

19

References

1.

Aiello EA, and Cingolani HE. Angiotensin II stimulates cardiac L-

type Ca(2+) current by a Ca(2+)- and protein kinase C-dependent mechanism. American journal of physiology 280: H1528-1536, 2001. 2.

Akao M, O'Rourke B, Kusuoka H, Teshima Y, Jones SP, and Marban

E. Differential actions of cardioprotective agents on the mitochondrial death pathway. Circulation research 92: 195-202, 2003. 3.

Amin JK, Xiao L, Pimental DR, Pagano PJ, Singh K, Sawyer DB, and

Colucci WS. Reactive oxygen species mediate alpha-adrenergic receptorstimulated hypertrophy in adult rat ventricular myocytes. Journal of molecular and cellular cardiology 33: 131-139, 2001. 4.

An J, Varadarajan SG, Camara A, Chen Q, Novalija E, Gross GJ, and

Stowe DF. Blocking Na(+)/H(+) exchange reduces [Na(+)](i) and [Ca(2+)](i) load after ischemia and improves function in intact hearts. American journal of physiology 281: H2398-2409, 2001. 5.

Avkiran M, and Haworth RS. Regulatory effects of G protein-

coupled receptors on cardiac sarcolemmal Na+/H+ exchanger activity: signalling and significance. Cardiovascular research 57: 942-952, 2003. 6.

Baartscheer A, Schumacher CA, van Borren MM, Belterman CN,

Coronel R, Opthof T, and Fiolet JW. Chronic inhibition of Na+/H+exchanger attenuates cardiac hypertrophy and prevents cellular remodeling in heart failure. Cardiovascular research 65: 83-92, 2005. 7.

Baines CP, Song CX, Zheng YT, Wang GW, Zhang J, Wang OL, Guo

Y, Bolli R, Cardwell EM, and Ping P. Protein kinase Cepsilon interacts with and inhibits the permeability transition pore in cardiac mitochondria. Circulation research 92: 873-880, 2003. 8.

Bolli R, Zughaib M, Li XY, Tang XL, Sun JZ, Triana JF, and McCay

PB. Recurrent ischemia in the canine heart causes recurrent bursts of free radical production that have a cumulative effect on contractile function. A pathophysiological basis for chronic myocardial "stunning". The Journal of clinical investigation 96: 1066-1084, 1995.

20

9.

Brandes RP. Triggering mitochondrial radical release: a new

function for NADPH oxidases. Hypertension 45: 847-848, 2005. 10.

Caldiz CI, Garciarena CD, Dulce RA, Novaretto LP, Yeves AM, Ennis

IL, Cingolani HE, Chiappe de Cingolani G, and Perez NG. Mitochondrial reactive oxygen species activate the slow force response to stretch in feline myocardium. The Journal of physiology 584: 895-905, 2007. 11.

Cappola TP, Kass DA, Nelson GS, Berger RD, Rosas GO, Kobeissi

ZA, Marban E, and Hare JM. Allopurinol improves myocardial efficiency in patients with idiopathic dilated cardiomyopathy. Circulation 104: 24072411, 2001. 12.

Cingolani HE, and Ennis IL. Sodium-hydrogen exchanger, cardiac

overload, and myocardial hypertrophy. Circulation 115: 1090-1100, 2007. 13.

Cingolani HE, Ennis IL, and Mosca SM. NHE-1 and NHE-6 activities:

ischemic and reperfusion injury. Circulation research 93: 694-696, 2003. 14.

Cingolani HE, Perez NG, Aiello EA, and de Hurtado MC. Intracellular

signaling following myocardial stretch: an autocrine/paracrine loop. Regulatory peptides 128: 211-220, 2005. 15.

Cingolani HE, Plastino JA, Escudero EM, Mangal B, Brown J, and

Perez NG. The effect of xanthine oxidase inhibition upon ejection fraction in heart failure patients: La Plata Study. Journal of cardiac failure 12: 491498, 2006. 16.

Cingolani HE, Villa-Abrille MC, Cornelli M, Nolly A, Ennis IL,

Garciarena C, Suburo AM, Torbidoni V, Correa MV, Camilionde Hurtado MC, and Aiello EA. The positive inotropic effect of angiotensin II: role of endothelin-1 and reactive oxygen species. Hypertension 47: 727-734, 2006. 17.

Cohen MV, Yang XM, and Downey JM. The pH hypothesis of

postconditioning: staccato reperfusion reintroduces oxygen and perpetuates myocardial acidosis. Circulation 115: 1895-1903, 2007. 18.

Dell'Italia LJ, Meng QC, Balcells E, Wei CC, Palmer R, Hageman GR,

Durand J, Hankes GH, and Oparil S. Compartmentalization of angiotensin II generation in the dog heart. Evidence for independent mechanisms in

21

intravascular and interstitial spaces. The Journal of clinical investigation 100: 253-258, 1997. 19.

Dikalov S, Griendling KK, and Harrison DG. Measurement of

reactive oxygen species in cardiovascular studies. Hypertension 49: 717727, 2007. 20.

Downey JM, and Cohen MV. Free radicals in the heart: friend or

foe? Expert review of cardiovascular therapy 6: 589-591, 2008. 21.

Duda M, Konior A, Klemenska E, and Beresewicz A. Preconditioning

protects endothelium by preventing ET-1-induced activation of NADPH oxidase and xanthine oxidase in post-ischemic heart. Journal of molecular and cellular cardiology 42: 400-410, 2007. 22.

Fantinelli JC, Cingolani HE, and Mosca SM. Na+/H+ exchanger

inhibition at the onset of reperfusion decreases myocardial infarct size: role of reactive oxygen species. Cardiovasc Pathol 15: 179-184, 2006. 23.

Griendling KK. Novel NAD(P)H oxidases in the cardiovascular

system. Heart (British Cardiac Society) 90: 491-493, 2004. 24.

Halestrap AP, Clarke SJ, and Javadov SA. Mitochondrial

permeability transition pore opening during myocardial reperfusion--a target for cardioprotection. Cardiovascular research 61: 372-385, 2004. 25.

Hotta Y, Ishikawa N, Ohashi N, and Matsui K. Effects of SM-20550, a

selective Na+-H+ exchange inhibitor, on the ion transport of myocardial mitochondria. Molecular and cellular biochemistry 219: 83-90, 2001. 26.

Imahashi K, Nishimura T, Yoshioka J, and Kusuoka H. Role of

intracellular Na(+) kinetics in preconditioned rat heart. Circulation research 88: 1176-1182, 2001. 27.

Javadov S, Baetz D, Rajapurohitam V, Zeidan A, Kirshenbaum LA,

and Karmazyn M. Antihypertrophic effect of Na+/H+ exchanger isoform 1 inhibition is mediated by reduced mitogen-activated protein kinase activation secondary to improved mitochondrial integrity and decreased generation of mitochondrial-derived reactive oxygen species. The Journal of pharmacology and experimental therapeutics 317: 1036-1043, 2006.

22

28.

Javadov S, Choi A, Rajapurohitam V, Zeidan A, Basnakian AG, and

Karmazyn M. NHE-1 inhibition-induced cardioprotection against ischaemia/reperfusion is associated with attenuation of the mitochondrial permeability transition. Cardiovascular research 77: 416-424, 2008. 29.

Javadov S, Huang C, Kirshenbaum L, and Karmazyn M. NHE-1

inhibition improves impaired mitochondrial permeability transition and respiratory function during postinfarction remodelling in the rat. Journal of molecular and cellular cardiology 38: 135-143, 2005. 30.

Javadov S, and Karmazyn M. Mitochondrial permeability transition

pore opening as an endpoint to initiate cell death and as a putative target for cardioprotection. Cell Physiol Biochem 20: 1-22, 2007. 31.

Kapus A, Lukacs GL, Cragoe EJ, Jr., Ligeti E, and Fonyo A.

Characterization of the mitochondrial Na+-H+ exchange. The effect of amiloride analogues. Biochimica et biophysica acta 944: 383-390, 1988. 32.

Kim JS, Jin Y, and Lemasters JJ. Reactive oxygen species, but not

Ca2+ overloading, trigger pH- and mitochondrial permeability transitiondependent death of adult rat myocytes after ischemia-reperfusion. American journal of physiology 290: H2024-2034, 2006. 33.

Kimura S, Zhang GX, Nishiyama A, Shokoji T, Yao L, Fan YY,

Rahman M, and Abe Y. Mitochondria-derived reactive oxygen species and vascular MAP kinases: comparison of angiotensin II and diazoxide. Hypertension 45: 438-444, 2005. 34.

Kimura S, Zhang GX, Nishiyama A, Shokoji T, Yao L, Fan YY,

Rahman M, Suzuki T, Maeta H, and Abe Y. Role of NAD(P)H oxidase- and mitochondria-derived reactive oxygen species in cardioprotection of ischemic reperfusion injury by angiotensin II. Hypertension 45: 860-866, 2005. 35.

Lassegue B, and Griendling KK. Reactive oxygen species in

hypertension; An update. Am J Hypertens 17: 852-860, 2004. 36.

Li L, Fink GD, Watts SW, Northcott CA, Galligan JJ, Pagano PJ, and

Chen AF. Endothelin-1 increases vascular superoxide via endothelin(A)-

23

NADPH oxidase pathway in low-renin hypertension. Circulation 107: 10531058, 2003. 37.

Li Q, Bolli R, Qiu Y, Tang XL, Murphree SS, and French BA. Gene

therapy with extracellular superoxide dismutase attenuates myocardial stunning in conscious rabbits. Circulation 98: 1438-1448, 1998. 38.

Minhas KM, Saraiva RM, Schuleri KH, Lehrke S, Zheng M, Saliaris

AP, Berry CE, Barouch LA, Vandegaer KM, Li D, and Hare JM. Xanthine oxidoreductase inhibition causes reverse remodeling in rats with dilated cardiomyopathy. Circulation research 98: 271-279, 2006. 39.

Miura T, Liu Y, Goto M, Tsuchida A, Miki T, Nakano A, Nishino Y,

Ohnuma Y, and Shimamoto K. Mitochondrial ATP-sensitive K+ channels play a role in cardioprotection by Na+-H+ exchange inhibition against ischemia/reperfusion injury. Journal of the American College of Cardiology 37: 957-963, 2001. 40.

Murphy E, and Steenbergen C. Mechanisms underlying acute

protection from cardiac ischemia-reperfusion injury. Physiological reviews 88: 581-609, 2008. 41.

Myers ML, Farhangkhoee P, and Karmazyn M. Hydrogen peroxide

induced impairment of post-ischemic ventricular function is prevented by the sodium-hydrogen exchange inhibitor HOE 642 (cariporide). Cardiovascular research 40: 290-296, 1998. 42.

Nishikimi M, Appaji N, and Yagi K. The occurrence of superoxide

anion in the reaction of reduced phenazine methosulfate and molecular oxygen. Biochemical and biophysical research communications 46: 849854, 1972. 43.

Numata M, Petrecca K, Lake N, and Orlowski J. Identification of a

mitochondrial Na+/H+ exchanger. The Journal of biological chemistry 273: 6951-6959, 1998. 44.

Petronilli V, Penzo D, Scorrano L, Bernardi P, and Di Lisa F. The

mitochondrial permeability transition, release of cytochrome c and cell death. Correlation with the duration of pore openings in situ. The Journal of biological chemistry 276: 12030-12034, 2001.

24

45.

Pimentel DR, Amin JK, Xiao L, Miller T, Viereck J, Oliver-Krasinski J,

Baliga R, Wang J, Siwik DA, Singh K, Pagano P, Colucci WS, and Sawyer DB. Reactive oxygen species mediate amplitude-dependent hypertrophic and apoptotic responses to mechanical stretch in cardiac myocytes. Circulation research 89: 453-460, 2001. 46.

Rothstein EC, Byron KL, Reed RE, Fliegel L, and Lucchesi PA.

H(2)O(2)-induced Ca(2+) overload in NRVM involves ERK1/2 MAP kinases: role for an NHE-1-dependent pathway. American journal of physiology 283: H598-605, 2002. 47.

Ruiz-Meana M, Garcia-Dorado D, Pina P, Inserte J, Agullo L, and

Soler-Soler J. Cariporide preserves mitochondrial proton gradient and delays ATP depletion in cardiomyocytes during ischemic conditions. American journal of physiology 285: H999-1006, 2003. 48.

Sabri A, Byron KL, Samarel AM, Bell J, and Lucchesi PA. Hydrogen

peroxide activates mitogen-activated protein kinases and Na+-H+ exchange in neonatal rat cardiac myocytes. Circulation research 82: 10531062, 1998. 49.

Seddon M, Looi YH, and Shah AM. Oxidative stress and redox

signalling in cardiac hypertrophy and heart failure. Heart (British Cardiac Society) 93: 903-907, 2007. 50.

Snabaitis AK, Hearse DJ, and Avkiran M. Regulation of sarcolemmal

Na(+)/H(+) exchange by hydrogen peroxide in adult rat ventricular myocytes. Cardiovascular research 53: 470-480, 2002. 51.

Teshima Y, Akao M, Jones SP, and Marban E. Cariporide (HOE642),

a selective Na+-H+ exchange inhibitor, inhibits the mitochondrial death pathway. Circulation 108: 2275-2281, 2003. 52.

Tissier R, Berdeaux A, Ghaleh B, Couvreur N, Krieg T, Cohen MV,

and Downey JM. Making the heart resistant to infarction: how can we further decrease infarct size? Front Biosci 13: 284-301, 2008. 53.

Villafuerte FC SP, Vaughan-Jones RD. Common inhibitors of

membrane H+-transport also inhibit carbonic anhydrase The FASEB Journal 21: 2007.

25

54.

Wei S, Rothstein EC, Fliegel L, Dell'Italia LJ, and Lucchesi PA.

Differential MAP kinase activation and Na(+)/H(+) exchanger phosphorylation by H(2)O(2) in rat cardiac myocytes. Am J Physiol Cell Physiol 281: C1542-1550, 2001. 55.

Xiao L, Pimentel DR, Wang J, Singh K, Colucci WS, and Sawyer DB.

Role of reactive oxygen species and NAD(P)H oxidase in alpha(1)adrenoceptor signaling in adult rat cardiac myocytes. Am J Physiol Cell Physiol 282: C926-934, 2002. 56.

Zhang DX, Chen YF, Campbell WB, Zou AP, Gross GJ, and Li PL.

Characteristics and superoxide-induced activation of reconstituted myocardial mitochondrial ATP-sensitive potassium channels. Circulation research 89: 1177-1183, 2001. 57.

Zhang GX, Lu XM, Kimura S, and Nishiyama A. Role of mitochondria

in angiotensin II-induced reactive oxygen species and mitogen-activated protein kinase activation. Cardiovascular research 76: 204-212, 2007. 58.

Zhang M, Kho AL, Anilkumar N, Chibber R, Pagano PJ, Shah AM,

and Cave AC. Glycated proteins stimulate reactive oxygen species production in cardiac myocytes: involvement of Nox2 (gp91phox)containing NADPH oxidase. Circulation 113: 1235-1243, 2006. 59.

Zorov DB, Filburn CR, Klotz LO, Zweier JL, and Sollott SJ. Reactive

oxygen species (ROS)-induced ROS release: a new phenomenon accompanying induction of the mitochondrial permeability transition in cardiac myocytes. The Journal of experimental medicine 192: 1001-1014, 2000.

26