Báo cáo khoa học: Ca2+ rise within a narrow window of concentration prevents functional injury of mitochondria exposed to hypoxia ⁄reoxygenation by increasing antioxidative defence pdf

Ca 2+ rise within a narrow window of concentration prevents functional injury of mitochondria exposed to hypoxia ⁄ reoxygenation by increasing antioxidative defence Lorenz Schild 1 , Frank Plumeyer 1 and Georg Reiser 2 1 Bereich Pathologische Biochemie der Medizinischen Fakulta ¨ t der Otto-von-Guericke-Universita ¨ t Magdeburg, Germany 2 Institut fu ¨ r Neurobiochemie der Medizinischen Fakulta ¨ t der Otto-von-Guericke-Universita ¨ t Magdeburg, Germany It has been shown in animal models that transient isch- aemia in liver results in mitochondrial damage. The involvement of oxidative stress in the impairment of the organelle was demonstrated by the finding that glutathi- one (GSH) exerts a protective role [1]. A hallmark of ischaemia ⁄ reperfusion in liver is a significant increase in cytosolic and mitochondrial Ca 2+ concentration [2]. Oxidative stress and increase in the cytosolic Ca 2+ concentration favour opening of the mitochondrial permeability transition pore (MPTP) mediating mito- chondrial damage. In fact, cyclosporin A (CSA), a speci- fic inhibitor of MPTP, has been demonstrated to prevent mitochondrial and liver dysfunction in the re- perfusion phase [3,4]. Long-lasting ischaemia in liver was shown to induce cytochrome c release and necrosis, whereas short ischaemia with reperfusion results in the release of cytochrome c and apoptosis [5]. A further factor determining the outcome after liver ischaemia ⁄ reperfusion is nitric oxide (NO). However, reports about the effect of NO on mitochondrial and tissue damage are still controversial. Using either exo- genous NO donors, or endogenous NO precursors or inhibitors of NO synthesis, protective [6,7] as well as harmful effects [8,9] have been found with in vivo models of liver ischaemia. Investigations on isolated liver mitochondria have clearly shown that extramitochondrial Ca 2+ , reactive oxygen species (ROS), and NO, which are known to change in concentration during ischaemia ⁄ reperfusion, affect mitochondria. Elevation of Ca 2+ concentrations Keywords glutathione peroxidase; mitochondrial permeability transition pore; manganese super oxide dismutase; nitric oxide; oxidative stress Correspondence L. Schild, Bereich Pathologische Biochemie der Medizinischen Fakulta ¨ t der Otto-von- Guericke-Universita ¨ t Magdeburg, Leipziger Str. 44, 39120 Magdeburg, Germany Tel: +49 0931 6713644 Fax: +49 0931 6719176 E-mail: lorenz.schild@medizin. uni-magdeburg.de (Received 18 July 2005, revised 15 September 2005, accepted 19 September 2005) doi:10.1111/j.1742-4658.2005.04978.x Injury of liver by ischaemia crucially involves mitochondrial damage. The role of Ca 2+ in mitochondrial damage is still unclear. We investigated the effect of low micromolar Ca 2+ concentrations on respiration, membrane permeability, and antioxidative defence in liver mitochondria exposed to hypoxia ⁄ reoxygenation. Hypoxia ⁄ reoxygenation caused decrease in state 3 respiration and in the respiratory control ratio. Liver mitochondria were almost completely protected at about 2 lm Ca 2+ . Below and above 2 lm Ca 2+ , mitochondrial function was deteriorated, as indicated by the decrease in respiratory control ratio. Above 2 lm Ca 2+ , the mitochondrial membrane was permeabilized, as demonstrated by the sensitivity of state 3 respiration to NADH. Below 2 lm Ca 2+ , the nitric oxide synthase inhib- itor nitro-l-arginine methylester had a protective effect. The activities of the manganese superoxide dismutase and glutathione peroxidase after hypoxia showed maximal values at about 2 lm Ca 2+ . We conclude that Ca 2+ exerts a protective effect on mitochondria within a narrow concentra- tion window, by increasing the antioxidative defence. Abbreviations CSA, cyclosporin A; GPx, glutathione peroxidase; GSH, reduced glutathione; Mn-SOD, manganese superoxide dismutase; MPTP, mitochondrial permeability transition pore; L-NAME, nitro-L-arginine methylester; NO, nitric oxide; RCR, respiratory control ratio; ROS, reactive oxygen species; TPP + , tetraphenyl phosphonium cation. 5844 FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS into the low micromolar range causes reduction of state 3 respiration [10] and permeabilization of the mito- chondrial membrane [11–16]. ROS can impair function and integrity of mitochondria [17]. NO controls the electron flow through the respiratory chain by compet- itive inhibition of cytochrome oxidase [18]. There is a body of evidence showing that mitochon- dria are equipped with a constitutive and Ca 2+ sensi- tive NO synthase [19,20]. Thus, NO generated by mitochondria may influence oxidative phosphorylation. Moreover, mitochondria are major producers of super- oxide anion radicals within the respiratory chain in hepatocytes which in the presence of NO allow the formation of the highly reactive peroxynitrite. Peroxy- nitrite is known to contribute to functional damage of components of the mitochondrial electron transport chain [21]. Hypoxia ⁄ reoxygenation, an important com- ponent of ischaemia ⁄ reperfusion, leads to the impair- ment of isolated rat liver mitochondria with the involvement of oxidative stress [22]. In our previous work, we demonstrated that mitochondrially derived NO is significantly involved in the deterioration of iso- lated liver mitochondria upon hypoxia ⁄ reoxygenation [23]. In this way, mitochondria may exacerbate their own injury upon ischaemia ⁄ reperfusion. Although the effect of single factors on mitochondria, such as ele- vated extramitochondrial Ca 2+ concentration, hypo- xia ⁄ reoxygenation, and NO, have been elucidated, their interplay during ischaemia ⁄ reperfusion is still poorly understood. Here, we subjected isolated rat liver mitochondria to hypoxia and reoxygenation in combination with extra- mitochondrial Ca 2+ concentrations up to 5 lm. After- wards, we determined state 3 and state 4 respiration with glutamate and malate as substrates. We also meas- ured membrane permeability, and the activities of the antioxidative enzymes glutathione peroxidase (GPx) and manganese superoxide dismutase (Mn-SOD). Addi- tionally, the effect of permanent inhibition of mito- chondrial NO synthesis by nitro-l-arginine methylester (l-NAME) on respiration upon hypoxia ⁄ reoxygenation in combination with elevated Ca 2+ concentration was investigated. We found that within a narrow concentra- tion range at around 2 lm extramitochondrial Ca 2+ , mitochondria were almost completely protected against decrease in active respiration and increase in membrane permeability. The activities of the antioxidative enzymes GPx and Mn-SOD were stimulated by hypoxia ⁄ reoxy- genation and increase in Ca 2+ concentration, displaying maximal values at a concentration of about 2 lm.At this Ca 2+ concentration, the inhibition of NO synthesis with l-NAME did not affect state 3 respiration. From these data we conclude that extramitochondrial Ca 2+ at a narrow concentration window exerts a protective effect upon hypoxia ⁄ reoxygenation by increasing the activity of antioxidative enzymes in liver mito- chondria. Results Ca 2+ affects respiration and membrane permeability upon hypoxia ⁄ reoxygenation In liver ischaemia⁄ reperfusion the cytosolic Ca 2+ con- centration in hepatic cells is elevated into the low micro- molar range. In order to investigate the influence of extramitochondrial Ca 2+ on the impairment of mito- chondria by ischaemia ⁄ reperfusion, isolated rat liver mitochondria were subjected to 5 min hypoxia followed by 10 min reoxygenation in the continuous presence of Ca 2+ at concentrations varying from 0.2 up to 4.4 lm. Rates of respiration were determined after hypoxia ⁄ reoxygenation with 5 mm glutamate and 5 mm malate as substrates. Transient hypoxia in the presence of 0.2 lm Ca 2+ caused decrease in state 3 respiration to 45% of the normoxic control value (incubation in air saturated medium). The influence of Ca 2+ on state 3 respiration obtained with hypoxia ⁄ reoxygenation was characterized by a bell-shaped concentration depend- ence (Fig. 1A, upper part). Increasing the Ca 2+ concen- tration improved state 3 respiration. Almost complete protection was seen at 2 lm extramitochondrial Ca 2+ (91% of normoxic mitochondria). This was not observed when Ca 2+ uptake was inhibited by 10 lm ruthenium red (data not shown). Further increase in extramitochondrial Ca 2+ concentration resulted in decreased rates of state 3 respiration measured after hypoxia ⁄ reoxygenation. At the maximally used concen- tration of 4.4 lm Ca 2+ , no stimulation of oxygen con- sumption by ADP could be reached. The respiration determined in the absence of ADP (state 4) had no clear Ca 2+ dependence (lower part in Fig. 1A). In order to test whether the effect of hypoxia ⁄ reoxygenation and elevated Ca 2+ on state 3 respiration was due to opening of the MPTP, isolated rat liver mitochondria were exposed to hypoxia ⁄ reoxygenation and Ca 2+ in the additional presence of 2 lm of the MPTP inhibitor CSA. At this concentration, CSA completely prevented Ca 2+ -induced swelling of liver mitochondria (data not shown). CSA partially protected liver mitochondria against decrease in state 3 respiration at Ca 2+ concen- trations below and above 2 lm (Fig. 1A, upper part). Within the narrow concentration range at around 2 lm extramitochondrial Ca 2+ , no effect of CSA was observed. The rates of state 4 respiration were slightly higher in CSA-containing incubations in comparison to L. Schild et al. Ca 2+ protects mitochondria during hypoxia FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS 5845 CSA-free incubations. At 2 lm extramitochondrial Ca 2+ both values were equal (Fig. 1A, lower part). In order to evaluate precisely the functional injury of mitochondria, the coupling of oxidative phosphory- lation was quantified by calculating respiratory con- trol ratios (RCR), which are given by the ratios of state 3 and state 4 respiration. The resulting data are presented in Fig. 1B. Highest RCR values were found between 1 and 2 lm extramitochondrial Ca 2+ indica- ting a protective Ca 2+ concentration range. Below and above this concentration range, loss of mito- chondrial coupling was observed. Inhibition of pore opening by CSA had no significant effect on RCR over the whole Ca 2+ concentration range investigated. The RCR of freshly isolated mitochondria was 6.4 ± 0.7 (n ¼ 12). To investigate the possibility of a CSA-insensitive permeabilization of the mitochondrial membrane upon hypoxia ⁄ reoxygenation and Ca 2+ , we used a different approach. The membrane impermeable pyridine nucleotide NADH (5 mm) and cytochrome c (10 lm) were added to mitochondria respiring under state 3 conditions. Both compounds have no effect on state 3 respiration in intact mitochondria. However, permeabi- lization of the membrane allows access of NADH and cytochrome c to the respiratory chain resulting in sti- mulation of state 3 respiration. The relative changes of state 3 respiration without and with NADH addition plus cytochrome c, measured after exposure of mito- chondria to hypoxia ⁄ reoxygenation and various Ca 2+ concentrations, is depicted in Fig. 1C. Up to 2 lm extramitochondrial Ca 2+ , no stimulation of state 3 res- piration by NADH plus cytochrome c was observed, documenting the tightness of the mitochondrial mem- brane. Even in the presence of 2 lm CSA, elevation of the Ca 2+ concentration from 2 lm to 4.4 lm was par- alleled by an increase in the ratio of state 3 respiration without and with NADH addition plus cytochrome c clearly. This indicates permeabilization of the mito- chondrial membrane. The permeabilization (Fig. 1C) was associated with loss of mitochondrial function (Fig. 1B). Fig. 1. Influence of cyclosporin A on the Ca 2+ sensitivity of res- piration upon hypoxia ⁄ reoxygenation. Rat liver mitochondria (1 mgÆmL )1 ) were subjected to 5 min hypoxia and 10 min reoxy- genation with and without 2 l M CSA in the presence of various Ca 2+ concentrations at 30 °C. Afterwards, respiration was deter- mined in the presence of 5 m M glutamate, 5 mM malate either without (state 4) or with 200 l M ADP (state 3). Subsequently 5 mM NADH and 10 lM cytochrome c were added to demonstrate per- meabilization of the mitochondrial membrane. The rates of respir- ation (A), RCR (B) and the ratio of the rates of state 3 respiration before and after the addition of NADH and cytochrome c (C) are presented. The rate of state 3 respiration of freshly isolated mito- chondria was 82.3 ± 6.8 nmol O 2 Æmg )1 Æmin )1 . Data are presented as mean ± SEM of five mitochondrial preparations. Ca 2+ protects mitochondria during hypoxia L. Schild et al. 5846 FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS Ca 2+ and hypoxia ⁄ reoxygenation regulate antioxidative activity in liver mitochondria In our previous work we have shown that hypoxia ⁄ reoxygenation induces oxidative stress indicated by the formation of protein carbonyls (marker of oxidative protein modification) which depends on the Ca 2+ concentration [10]. To investigate how Ca 2+ modulates oxidative stress during hypoxia ⁄ reoxygenation, we measured the activity of the Mn-SOD in normoxic incu- bation and after hypoxia ⁄ reoxygenation. The enzyme activity in the normoxic incubation was maximal in the presence of 2 lm extramitochondrial Ca 2+ (Fig. 2). At 0.2 lm Ca 2+ , lower activity of Mn-SOD was found (73 at 0.2 lm Ca 2+ vs. 118 unitsÆmg )1 at 2 lm Ca 2+ ). Sim- ilar results were obtained with Mn-SOD from bovine erythrocytes. In the presence of 2 lm Ca 2+ , the enzyme activity was increased from 0.134 ± 0.021 unitsÆ mg )1 (at 0.2 lm Ca 2+ ) to 1.851 ± 0.056 unitsÆmg )1 . This sti- mulation could be reversed by the addition of 2 mm EGTA. The activity of the enzyme determined after hypoxia ⁄ reoxygenation also reached a maximum value in the presence of 2 lm Ca 2+ , but was significantly higher than in a normoxic incubation (157 unitsÆmg )1 after hypoxia⁄ reoxygenation vs. 118 unitsÆmg )1 without hypoxia ⁄ reoxygenation). Increase in the extramito- chondrial Ca 2+ concentration from 0.2 to 2 lm resulted in a 2.6-fold increase in the activity of Mn-SOD, whereas in normoxic incubations the elevation was only 1.6-fold. Thus, the combination of hypoxia⁄ reoxygena- tion and 2 lm extramitochondrial Ca 2+ caused a con- siderable increase in the activity of this antioxidative defence enzyme. In a further series of experiments we investigated whether the activity of a second antioxidative enzyme, that is GPx, is sensitive to hypoxia ⁄ reoxygenation and extramitochondrial Ca 2+ . The activity of this enzyme didnotdependontheextramitochondrialCa 2+ concentra- tion in the low micromolar range under normoxic con- ditions (Fig. 3, lower part). In Ca 2+ -free incubations, the activity of GPx was double after 5 min hypoxia fol- lowed by 10 min reoxygenation (819 vs. 405 unitsÆmg )1 at 0.2 lm extramitochondrial Ca 2+ ). The activity determined after hypoxia ⁄ reoxygenation was slightly Fig. 3. Influence of hypoxia ⁄ reoxygenation and Ca 2+ on the activity of glutathione peroxidase (GPx). Rat liver mitochondria (1 mgÆmL )1 ) were either incubated at various Ca 2+ concentrations and 5 mM glu- tamate plus 5 m M malate in the incubation medium or were subjec- ted to 5 min hypoxia and 10 min reoxygenation in the presence of various Ca 2+ concentrations at 30 °C. After the reoxygenation period 5m M glutamate and 5 mM malate were added. For the determin- ation of GPx activity, 500 lL samples were withdrawn from the incubations. The data are presented as mean ± SEM from five preparations of mitochondria. The differences between GPx activities of incubations with and without hypoxia ⁄ reoxygenation in the pres- ence of similar Ca 2+ concentrations were significant with P < 0.01. Fig. 2. Change of Ca 2+ -sensitivity of Mn-SOD activity by hypoxia ⁄ reoxygenation. Rat liver mitochondria (1 mgÆmL )1 ) were either incu- bated at various Ca 2+ concentrations and 5 mM glutamate plus 5m M malate in the incubation medium or were subjected to 5 min hypoxia and 10 min reoxygenation in the presence of various Ca 2+ concentrations at 30 °C. After the reoxygenation period 5 mM glutamate and 5 mM malate were added. For the determination of Mn-SOD activity, 500 lL samples were withdrawn from the incuba- tions. The data are presented as mean ± SEM from five prepara- tions of mitochondria. Additional student’s t-test analysis gave a significant difference in Mn-SOD activities between the values at 0.1 and 2.0 l M Ca 2+ (P < 0.01), both without and with hypoxia ⁄ reoxygenation. *Differences in Mn-SOD activities of incubations with and without hypoxia ⁄ reoxygenation were significant with P < 0.05. L. Schild et al. Ca 2+ protects mitochondria during hypoxia FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS 5847 sensitive to Ca 2+ with the tendency to reach the highest levels between 1 and 2 lm extramitochondrial Ca 2+ . Alternatively to the formation of H 2 O 2 by the Mn-SOD reaction, superoxide anion radicals produced within the respiratory chain can react with mitochond- rially generated NO to form the highly reactive per- oxynitrite. In order to estimate the involvement of NO and ⁄ or peroxynitrite in the impairment of mitochond- rial function by hypoxia ⁄ reoxygenation and extra- mitochondrial Ca 2+ , we studied to what degree the inhibition of NO synthesis by l-NAME affects respir- ation measured after 5 min hypoxia followed by 10 min reoxygenation. We could not find any signifi- cant difference in state 4 respiration by comparing incubations without and with l-NAME. Therefore, only the rates of state 3 respiration are depicted in Fig. 4. The continuous presence of l-NAME during hypoxia ⁄ reoxygenation performed in the presence of 0.2 lm extramitochondrial Ca 2+ (low Ca 2+ concentra- tion) partially protected rat liver mitochondria against decrease in state 3 respiration (72 ± 4.4 vs. 39 ± 3.1 nmol O 2 Æmin )1 Æmg )1 ). At this Ca 2+ concen- tration, the continuous presence of 50 lm haemoglobin was also protective (78 ± 3.9 vs. 39 ± 3.1 nmol O 2 Æ min )1 Æmg )1 ). When mitochondria were subjected to hypoxia ⁄ reoxygenation in the presence of 2 lm extra- mitochondrial Ca 2+ (protective Ca 2+ concentration), l-NAME did not affect respiration. Likewise, at 4.4 lm Ca 2+ , inhibition of enzymatic NO synthesis during hypoxia ⁄ reoxygenation did not lead to a change in state 3 respiration (high Ca 2+ concentration). Discussion Extramitochondrial Ca 2+ can amplify or attenuate the impairment of liver mitochondria by hypoxia ⁄ reoxygenation Isolated mitochondria have been successfully used to study the effect of distinct factors impairing mitochon- dria which are relevant in pathophysiological situations such as ischaemia ⁄ reperfusion [24–26]. Both the in vivo studies of ischaemia and the cell culture investigation on hypoxia ⁄ reoxygenation require a relatively long period of hypoxia to achieve significant injury. How- ever, in isolated mitochondria a few minutes of hypoxia are sufficient to cause dramatic damage. Differences in local oxygen concentration may be the reason for this different time required to reach injury either in vivo or in isolated mitochondria. We have found that at elevated extramitochondrial Ca 2+ con- centrations, ADP at physiological concentration protects mitochondria from hypoxia ⁄ reoxygenation- induced damage [27,28]. Only when all the ADP is converted into AMP, mitochondrial damage occurs. This finding may contribute to the fact that longer periods of ischaemia are required to achieve damage in tissue, in comparison with results obtained with iso- lated mitochondria, which have to be exposed only for a short period of time to hypoxia in order to induce damage. In previous papers we reported that hypoxia ⁄ reoxy- genation reduces state 3 respiration in isolated rat liver mitochondria [22] and that extramitochondrial Ca 2+ in the low micromolar range modulates mitochondrial damage and oxidative stress [10]. Now, by testing the action of CSA and measuring RCR we show that open- ing of the MPTP is not significantly involved in func- tional impairment of liver mitochondria exposed to hypoxia ⁄ reoxygenation and Ca 2+ . This is surprising as increases in extramitochondrial Ca 2+ concentration and oxidative stress are known to be major factors for increasing the probability for pore opening [29–33]. Reasons for the CSA-independent injury of mitochond- rial function might be the increase in oxidative stress below and above 2 lm Ca 2+ as demonstrated earlier [10], possibly causing damage to respiratory chain com- plexes, and ⁄ or CSA-insensitive permeabilization of the mitochondrial membrane. In fact, CSA-insensitive per- meabilization of the mitochondrial membrane was found after hypoxia ⁄ reoxygenation in the presence of Ca 2+ concentrations higher than 2 lm (Fig. 1C). Fig. 4. Modulation of the effect of NO on state 3 respiration after hypoxia ⁄ reoxygenation by extramitochondrial Ca 2+ . Rat liver mito- chondria (1 mgÆmL )1 ) were subjected to 5 min hypoxia and 10 min reoxygenation with and without 10 m ML-NAME in the continuous presence of either 0.2 l M,2lM or 4.4 lM extramitochondrial Ca 2+ at 30 °C. Afterwards, 5 mM glutamate, 5 mM malate and 200 lM ADP were added to stimulate state 3 respiration. Data are presen- ted as mean ± SEM of five preparations of mitochondria. *State 3 respiration with and without L-NAME is different with P < 0.05 according to Student’s t-test. Ca 2+ protects mitochondria during hypoxia L. Schild et al. 5848 FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS Ca 2+ affects the balance between oxidative and antioxidative processes during hypoxia⁄ reoxygenation As we have demonstrated earlier, hydrogen peroxide which is formed from superoxide anion radicals accu- mulates during reoxygenation in the presence of high Ca 2+ concentration [28]. However, the protection of mitochondria from hypoxia ⁄ reoxygenation-induced damage and the low amount of protein carbonyls [10] at 2 lm Ca 2+ suggest relatively low superoxide radical concentration. This is consistent with our finding of considerably increased activity of the Mn-SOD at this Ca 2+ concentration. At 2 lm Ca 2+ , no protection of state 3 respiration was seen during hypoxia⁄ reoxygena- tion in the presence of ruthenium red (data not shown). Therefore, it can be concluded that Ca 2+ has to enter the mitochondrial matrix in order to cause increase in Mn-SOD activity. Both Ca 2+ and hypox- ia ⁄ reoxygenation synergistically contribute to this effect. Under these conditions, protein levels of the Mn-SOD remained unchanged as determined by west- ern blot analysis (data not shown). This is not surpri- sing, as protein synthesis of this enzyme takes place within the cytosolic compartment [34]. Thus chemical modification is responsible for the change in the activ- ity of the enzyme. It has been shown that inactivation of the enzyme may result from tyrosine nitration by peroxynitrite [35]. In the in vitro model of isch- aemia ⁄ reperfusion applied here we did not observe decrease in the activity of Mn-SOD. Instead, increase in activity was found. The mechanism by which Ca 2+ in combination with hypoxia ⁄ reoxygenation mediates this antioxidant effect still remains unclear. It might be speculated that interaction of Ca 2+ with components of the active site as well as oxidation of certain amino acids in the catalytic domain may modulate enzymatic activity. It has been shown that replacement of His30 with Asn30 resulted in dramatic decrease of enzyme activity, because it did not participate in the hydrogen bond network of the active site [36]. Therefore, it is reasonable to assume that metal ions and ROS may modify the active site of the enzyme leading to changes in enzyme activity. Similarly, the activity of glutathione peroxidase was affected by hypoxia ⁄ reoxygenation. However, only a slight Ca 2+ dependence was observed. As the enzyme is synthesized within the cytosolic compartment, modi- fication of the enzyme should be responsible for increase in the activity caused by hypoxia ⁄ reoxygena- tion. The mechanism by which hypoxia ⁄ reoxygenation causes increase in the activity of mitochondrial gluta- thione peroxidase still has to be elucidated. In our previous study [10] we have demonstrated that increasing the extramitochondrial Ca 2+ concen- tration above 2 lm caused high levels of protein car- bonyls indicating a high degree of oxidative stress. On the other hand, we here report high activities of Mn-SOD and GPx in this range of Ca 2+ concentra- tion. The CSA-insensitive permeabilization of the mito- chondrial membrane occurring under this condition (Fig. 1C) may explain the apparent discrepancy. Per- meabilization of the membrane is known to be associ- ated with energetic failure and efflux of mitochondrial constituents such as glutathione. Consequently, the antioxidative defence decreases. At Ca 2+ concentrations lower than 2 lm, impair- ment of oxidative phosphorylation by hypoxia ⁄ reoxy- genation was caused by oxidative stress as high amounts of protein carbonyls were determined [10]. This fits well with our observation that in this range of Ca 2+ concentration relatively low activities of Mn-SOD and GPx were determined (Figs 2 and 3). Thereby, the mitochondrial respiratory chain is the source of reactive superoxide anion radicals generated within the complexes I and III. In experiments with isolated rat liver mitochondria, we have earlier demonstrated that NO accumulates during hypoxia ⁄ reoxygenation [23]. Subsequently, the highly reactive peroxynitrite can be formed which is known to injure components of the respiratory chain. It has been demonstrated that peroxynitrite is involved in Ca 2+ -induced impairment of liver mito- chondria [31,33]. Thus, the protective effect of the inhi- bition of NO synthesis at 0.2 lm Ca 2+ demonstrated here (Fig. 4) could be mainly attributed to the dimin- ished peroxynitrite formation. In contrast, at 2 lm Ca 2+ , superoxide anion radical concentration, but not NO production, is reduced, as the increase in Ca 2+ concentration stimulates NO synthesis. Here, no effect of l-NAME on state 3 respiration was seen. Conclusions We used an in vitro model of ischaemia ⁄ reperfusion of liver to study the effect of hypoxia ⁄ reoxygenation in combination with elevated extramitochondrial Ca 2+ concentration into the nonphysiological concentration range up to 4.4 lm. In this model we were able to clarify how mitochondrially generated ROS, NO and permea- bilization of the mitochondrial membrane are involved in mitochondrial damage. Our data demonstrate that hypoxia ⁄ reoxygenation and extramitochondrial Ca 2+ cause functional damage of isolated rat liver mitochon- dria. Essential steps involved in the cascade of mito- chondrial injury are CSA-insensitive permeabilization L. Schild et al. Ca 2+ protects mitochondria during hypoxia FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS 5849 of the mitochondrial membrane, production of ROS, generation of NO and peroxynitrite by mitochondria. We have found a distinct extramitochondrial Ca 2+ con- centration range around 2 lm, in which isolated rat liver mitochondria are almost completely protected against decrease in oxidative phosphorylation. Similar effects were also found in brain and heart mitochondria (data not shown). There was a clear correlation of increase in the activities of Mn-SOD, GPx, and insensitivity of inhi- bition of NO-synthesis at the 2 lm Ca 2+ concentration. Thus, we conclude that superoxide anion radicals and peroxynitrite play a pivotal role in damaging mitochon- dria upon hypoxia ⁄ reoxygenation at Ca 2+ concentra- tions up to about 2 lm. At higher Ca 2+ concentrations, the mechanism underlying the mitochondrial injury is the permeabilization of the membrane. As elevation of extramitochondrial Ca 2+ concentration into the low micromolar range appears to have protective effects, further investigations on the role of Ca 2+ in hypoxia⁄ reoxygenation at the cellular level and in in vivo studies of ischaemia should be performed. Experimental procedures Reagents Cyclosporin A, l-NAME and xanthin oxidase were pur- chased from Sigma (Deisenhofen, Germany). All other chemicals were of analytical grade. Preparation of mitochondria Liver mitochondria were prepared from 220 to 240 g male Wistar rats in ice-cold medium containing 25 mm sucrose, 20 mm Tris (pH 7.4), 2 mm EGTA, and 1% (w ⁄ v) bovine serum albumin using a standard procedure [37]. After the initial isolation, Percoll was used for purification of mito- chondria from a fraction containing some endoplasmatic reticulum, Golgi apparatus and plasma membranes [38]. This work was conducted in accordance with the regula- tions of the National Act for the use of Experimental Animals (Germany). Incubation of mitochondria Mitochondria (1–2 mg proteinÆmL )1 ) were incubated in a medium containing 10 mm sucrose, 120 mm KCl, 20 mm Tris, 15 mm potassium phosphate, 0.5 mm EGTA and 1mm free Mg 2+ at pH 7.4. Extramitochondrial Ca 2+ con- centrations were adjusted by using Ca 2+ ⁄ EGTA buffers. After preparation of the buffers, the free Ca 2+ concentra- tion was checked by means of a Ca 2+ -selective electrode. The actual concentration of Ca 2+ in the incubations was calculated considering the complex formation with other constituents of the medium such as Mg 2+ and adenine nucleotides. For the calculation, the complexing constants were used according to Fabioto et al. [39]. Hypoxia was produced by bubbling 2 mL of incubation medium with N 2 until an oxygen content of less than 1% of air saturation was reached. Afterwards, mitochondria were added to this oxygen-free medium and the incubation chamber was closed. The mitochondria themselves consumed most of the remaining oxygen resulting in very low oxygen concentra- tions reflected by collapse of the mitochondrial membrane potential (not shown). Reoxygenation was achieved by add- ing another 2 mL of incubation medium, which was air sat- urated, to the incubation tube [22]. Determination of mitochondrial respiration Oxygen consumption of mitochondria was measured in an incubation chamber equipped with a Clark-type electrode. The experimental approach was calibrated using the oxygen content of air saturated medium of 435 ng atomsÆmL )1 at 30 °C [40]. Determination of the mitochondrial membrane potential The mitochondrial membrane potential was calculated from the distribution of the lipophilic cation tetraphenyl phos- phonium (TPP + ) according to [41]. The extramitochondrial TPP + concentration was determined by means of a TPP + - sensitive electrode in the presence of 1 lm extramitochond- rial TPP + . For the calculation of the membrane potential a matrix volume of 1 lLÆmg )1 mitochondrial protein was assumed. The TPP + -sensitive electrode was calibrated by applying standard TPP + solutions. Determination of Mn-SOD activity The determination of Mn-SOD activity was based on the consumption of superoxide anion radicals generated by the xathine ⁄ xanthine oxidase system [42]. The reduction of cytochrome c was spectrophotometrically followed at 550 nm. 500 lL samples were withdrawn from mitochond- rial incubations and stored in liquid nitrogen. Before use samples were subjected to a threefold cycle of freezing and thawing. Determination of glutathione peroxidase activity Five-hundred microlitre samples were withdrawn from the mitochondrial incubations and stored in fluid nitrogen. After threefold freezing and thawing, samples were used for the determination of glutathione peroxidase activity. The Ca 2+ protects mitochondria during hypoxia L. Schild et al. 5850 FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS assay is based on the oxidation of glutathione and the sub- sequent oxidation of NADPH [43] which was followed photometrically at 340 nm. Determination of protein The protein content of the mitochondrial suspension was measured according to the Bradford method [44] using bovine serum albumin as the standard. Statistics Statistical analysis was performed by the Student’s t-test. Actual P -values are given in the legends of the figures. Data are presented as mean ± SEM. Acknowledgements This work was supported by Deutsche Forschungs- gemeinschaft (Project Re 847 ⁄ 3–1). References 1 Grattagliano I, Vendemiale G & Lauterburg BH (1999) Reperfusion injury of the liver: role of mito- chondria and protection by glutathione ester. J Surg Res 86, 2–8. 2 Isozaki H, Fujii K, Nomura E & Hara H (2000) Cal- cium concentration in hepatocytes during liver ischae- mia-reperfusion injury and the effects of diltiazem and citrate on perfused rat liver. Eur J Gastroenterol Hepatol 12, 291–297. 3 Leducq N, Delmas-Beauvieux MC, Bourdel-Marchasson I, Dufour S, Gallis JL, Canioni P & Diolez P (2000) Mitochondrial and energetic dysfunctions of the liver during normothermic reperfusion: protective effect of cyclosporine and role of the mitochondrial permeability transition pore. Transplant Proc 32, 479–480. 4 Leducq N, Delmas-Beauvieux MC, Bourdel-Marchasson I, Dufour S, Gallis JL, Canioni P & Diolez P (1998) Mitochondrial permeability transition during hypother- mic to normothermic reperfusion in rat liver demon- strated by the protective effect of cyclosporin A. Biochem J 336, 501–506. 5 Soeda J, Miyagawa S, Sano K, Masumoto J, Taniguchi S & Kawasaki S (2001) Cytochrome c release into cyto- sol with subsequent caspase activation during warm ischemia in rat liver. Am J Physiol Gastrointest Liver Physiol 281, G1115–G1123. 6 Rivera-Chavez FA, Toledo-Pereyra LH, Dean RE, Crouch L & Ward PA (2001) Exogenous and endogen- ous nitric oxide but not iNOS inhibition improves func- tion and survival of ischemically injured livers. J Invest Surg 14, 267–273. 7 Morisue A, Wakabayashi G, Shimazu M, Tanabe M, Mukai M, Matsumoto K, Kawachi S, Yoshida M, Yamamoto S & Kitajima M (2003) The role of nitric oxide after a short period of liver ischemia-reperfusion. J Surg Res 109, 101–109. 8 Meguro M, Katsuramaki T, Nagayama M, Kimura H, Isobe M, Kimura Y, Matsuno T, Nui A & Hirata K (2002) A novel inhibitor of inducible nitric oxide synthase (ONO-1714) prevents critical warm ischemia-reperfusion injury in the pig liver. Transplantation 73, 1439–1446. 9 Ozakyol AH, Tuncel N, Saricam T, Uzuner K, Ak D & Gurer F (2000) Effect of nitric oxide inhibition on rat liver ischemia reperfusion injury. Pathophysiology 7, 183–188. 10 Schild L, Plumeyer F, Reinheckel T & Augustin W (1997) Micromolar calcium prevents isolated rat liver mitochondria from anoxia-reoxygenation injury. Biochem Mol Biol Int 43, 35–45. 11 Andreyev A & Fiskum G (1999) Calcium induced release of mitochondrial cytochrome c by different mechanisms selective for brain versus liver. Cell Death Differ 6, 825–832. 12 Gogvadze V, Robertson JD, Zhivotovsky B & Orrenius S (2001) Cytochrome c release occurs via Ca 2+ -depend- ent and Ca 2+ -independent mechanisms that are regula- ted by Bax. J Biol Chem 276, 19066–19071. 13 Petronilli V, Cola C & Bernardi P (1993) Modulation of the mitochondrial cyclosporin A-sensitive permeability transition pore. II. The minimal requirements for pore induction underscore a key role for transmembrane elec- trical potential, matrix pH, and matrix Ca 2+ . J Biol Chem 268, 1011–1016. 14 Davies AM, Hershman S, Stabley GJ, Hoek JB, Peter- son J & Cahill A (2003) A Ca 2+ -induced mitochondrial permeability transition causes complete release of rat liver endonuclease G activity from its exclusive location within the mitochondrial intermembrane space. Identifi- cation of a novel endo-exonuclease activity residing within the mitochondrial matrix. Nucleic Acids Res 31, 1364–1373. 15 Xue L, Borutaite V & Tolkovsky AM (2002) Inhibition of mitochondrial permeability transition and release of cytochrome c by anti-apoptotic nucleoside analogues. Biochem Pharmacol 64, 441–449. 16 Kantrow SP & Piantadosi CA (1997) Release of cyto- chrome c from liver mitochondria during permeability transition. Biochem Biophys Res Commun 232, 669–671. 17 Wiswedel I, Tru ¨ mper L, Schild L & Augustin W (1988) Injury of mitochondrial respiration and membrane potential during iron ⁄ ascorbate-induced peroxidation. Biochim Biophys Acta 934, 80–86. 18 Shiva S, Brookes PS, Patel RP, Anderson PG & Darley-Usmar VM (2001) Nitric oxide partitioning into mitochondrial membranes and the control of respiration L. Schild et al. Ca 2+ protects mitochondria during hypoxia FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS 5851 at cytochrome c oxidase. Proc Natl Acad Sci USA 98, 7212–7217. 19 Giulivi C (2003) Characterization and function of mito- chondrial nitric-oxide synthase. Free Radic Biol Med 34, 397–408. 20 Ghafourifar P & Richter C (1997) Nitric oxide synthase activity in mitochondria. FEBS Lett 418, 291–296. 21 Cassina A & Radi R (1996) Differential inhibitory action of nitric oxide and peroxynitrite on mitochondrial elec- tron transport. Arch Biochem Biophys 328, 309–316. 22 Schild L, Reinheckel T, Wiswedel I & Augustin W (1997) Short-term impairment of energy production in isolated rat liver mitochondria by hypoxia ⁄ reoxygena- tion: involvement of oxidative protein modification. Biochem J 328, 205–210. 23 Schild L, Reinheckel T, Reiser M, Horn TF, Wolf G & Augustin W (2003) Nitric oxide produced in rat liver mitochondria causes oxidative stress and impairment of respiration after transient hypoxia. FASEB J 17, 2194– 2201. 24 Willet K, Detry O & Sluse FE (2000) Resistance of isolated pulmonary mitochondria during in vitro anoxia ⁄ reoxygenation. Biochim Biophys Acta 1460, 346–352. 25 Chang YJ & Chang KJ (2002) Calcium concentration in rat liver mitochondria during anoxic incubation. J Formos Med Assoc 101, 136–143. 26 Manzo-Avalos S, Perez-Vazquez V, Ramirez J, Aguil- era-Aguirre L, Gonzalez-Hernandez JC, Clemente- Guerrero M, Villalobos-Molina R & Saavedra-Molina A (2002) Regulation of the rate of synthesis of nitric oxide by Mg 2+ and hypoxia: studies in rat heart mito- chondria. Amino Acids 22, 381–389. 27 Schild L, Huppelsberg J, Kahlert S, Keilhoff G & Reiser G (2003) Brain mitochondria are primed by moderate Ca 2+ rise upon hypoxia ⁄ reoxygenation for functional breakdown and morphological desintegra- tion. J Biol Chem 278, 25454–25460. 28 Schild L & Reiser G (2005) Oxidative stress is involved in the permeabilization of the inner membrane of brain mitochondria exposed to hypoxia ⁄ reoxygenation and low micromolar Ca 2+ . FEBS J 272, 3593–3601. 29 Ichas F & Mazat JP (1998) From calcium signaling to cell death: two conformations for the mitochondrial per- meability transition pore: switching from low- to high- conductance state. Biochim Biophys Acta 1366, 33–50. 30 Brustovetsky N & Klingenberg M (1996) Mitochondrial ADP ⁄ ATP carrier can be reversibly converted into a large channel by Ca 2+ . Biochemistry 35, 8483–8488. 31 Ghafourifar P, Schenk U, Klein SD & Richter C (1999) Mitochondrial nitric-oxide synthase stimulation causes cytochrome c release from isolated mitochondria: evidence for intramitochondrial peroxynitrite formation. J Biol Chem 274, 31185–31188. 32 Kowaltowski AJ, Castilho RF & Vercesi AE (2001) Mitochondrial permeability transition and oxidative stress. FEBS Lett 495, 12–15. 33 Petronilli V, Costantini P, Scorrano L, Colonna R, Passamonti S & Bernardi P (1994) The voltage sensor of the mitochondrial permeability transition pore is tuned by the oxidation-reduction state of vicinal thiols: increase of the gating potential by oxidants and its reversal by reducing agents. J Biol Chem 269, 16638– 16642. 34 Sato M, Sasaki M & Hojo H (1995) Antioxidative roles of metallothionein and manganese superoxide dismutase induced by tumor necrosis factor-alpha and interleukin- 6. Arch Biochem Biophys 316, 738–744. 35 Macmillan-Crow LA & Cruthirds DL (2001) Invited review: manganese superoxide dismutase in disease. Free Radic Res 34, 325–336. 36 Ramilo CA, Leveque V, Guan Y, Lepock JR, Tainer JA, Nick HS & Silverman DN (1999) Interrupting the hydrogen bond network at the active site of human manganese superoxide dismutase. J Biol Chem 274, 27711–27716. 37 Johnson D & Lardy HA (1967) Isolation of liver or kidney mitochondria. Methods Enzymol 10, 94–96. 38 McCormack JG & Denton RMJ (1998) Influence of cal- cium ions on mammalian intramitochondrial dehydro- genases. Methods Enzymol 174, 95–118. 39 Fabiato A & Fabiato F (1979) Calculator programs for computing the composition of the solutions containing multiple metals and ligands used for experiments in skinned muscle cells. J Physiol (Paris) 75, 463–505. 40 Reynafarje B, Costa LE & Lehninger AL (1985) O 2 solubility in aqueous media determined by a kinetic method. Anal Biochem 145, 406–418. 41 Kamo N, Muratsugu M, Hongoh R & Kobatake Y (1979) Membrane potential of mitochondria measured with an electrode sensitive to tetraphenyl phosphonium and relationship between proton electrochemical poten- tial and phosphorylation potential in steady state. J Membr Biol 49, 105–121. 42 Flohe ´ L (1984) Superoxide dismutase assays. Methods Enzymol 105, 93–104. 43 Paglia DE & Valentine WN (1967) Studies on the quan- titative and qualitative characterization of erythrocyte glutathione peroxidase. J Laboratory Clin Med 70, 158–169. 44 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein util- izing the principle of protein-dye binding. Anal Biochem 72, 248–254. Ca 2+ protects mitochondria during hypoxia L. Schild et al. 5852 FEBS Journal 272 (2005) 5844–5852 ª 2005 The Authors Journal Compilation ª 2005 FEBS . permea- bilization of the mitochondrial membrane are involved in mitochondrial damage. Our data demonstrate that hypoxia ⁄ reoxygenation and extramitochondrial Ca 2+ cause functional damage of isolated rat liver. Ca 2+ rise within a narrow window of concentration prevents functional injury of mitochondria exposed to hypoxia ⁄ reoxygenation by increasing antioxidative defence Lorenz Schild 1 , Frank. From these data we conclude that extramitochondrial Ca 2+ at a narrow concentration window exerts a protective effect upon hypoxia ⁄ reoxygenation by increasing the activity of antioxidative enzymes