The effect of bovine serum albumin on the membrane potential and reactive oxygen species generation in succinate-supported isolated brain mitochondria

Neurochemistry International 50 (2007) 139–147 www.elsevier.com/locate/neuint

The effect of bovine serum albumin on the membrane potential and reactive oxygen species generation in succinate-supported isolated brain mitochondria Laszlo Tretter, Dora Mayer-Takacs, Vera Adam-Vizi * Department of Medical Biochemistry, Semmelweis University, Neurobiochemical Group, Hungarian Academy of Sciences and Szentagothai Janos Knowledge Center, P.O. Box 262, H-1444 Budapest, Hungary Received 4 May 2006; received in revised form 19 July 2006; accepted 21 July 2006 Available online 11 September 2006

Abstract Characteristics of the succinate-supported H2O2 formation were compared in mitochondria prepared from guinea-pig brain either by Percoll gradient centrifugation or using digitonin. The high rate of H2O2 generation measured in mitochondria prepared with digitonin (600.6  26.8 pmol/ min/mg protein) was inhibited by rotenone, consistently with a reverse flow of electrons via complex I. The rate of H2O2 formation was significantly smaller in Percoll-purified mitochondria (252.6  17.3 pmol/min/mg protein) and this was stimulated by rotenone. Since bovine serum albumin (BSA) is usually present in the isolation medium used in the digitonin method, systematic study was performed addressing the effect of BSA on H2O2 formation. Mitochondria prepared by the digitonin method (BSA present in the isolation medium) were highly polarized (185  3.2 mV) and addition of BSA (0.025%) to the assay medium increased H2O2 generation by only 50%. In Percoll-purified mitochondria DCm was more depolarized (171  2 mV) and BSA caused hyperpolarization by 10.7  1.9 mV. H2O2 formation, which was largely independent of DCm, was stimulated by 400%, became highly dependent on DCm and could be inhibited by rotenone in the presence of BSA. This shows that in Percoll-purified mitochondria ROS formation via reverse electron flow is preferred only when BSA is present in the assay medium. It is demonstrated that (i) the presence or absence of BSA could determine the mechanism by which ROS is generated in succinate-supported mitochondria and (ii) depolarization by about 10 mV eliminates reverse electron flow and the remaining ROS formation, which is smaller but still significant, is no longer dependent on DCm. # 2006 Elsevier Ltd. All rights reserved. Keywords: Mitochondria, Reactive oxygen species, ROS, BSA, Bovine serum albumin, Mitochondrial, Membrane potential, Reverse electron transfer

1. Introduction There are great varieties in the results concerning reactive oxygen species (ROS) generation in isolated mitochondria. The diverse results are partly due to species and tissue differences, but even results from different laboratories using mitochondria of the same tissue origin are contradictory as to the major site and mechanism of ROS generation. In recent years, several reports confirmed that isolated brain mitochondria produce ROS, the significance of which is given by the assumption that

Abbreviations: BSA, bovine serum albumin; FCCP, carbonyl cyanide-ptrifluoromethoxyphenylhydrazone; TMRM, tetramethylrhodamine methyl ester; TPP+, tetraphenylphosphonium ion; RCR, respiratory control ratio; ROS, reactive oxygen species; UCP, uncoupling protein * Corresponding author. Tel.: +36 1 266 2773; fax: +36 1 267 0031. E-mail address: [email protected] (V. Adam-Vizi). 0197-0186/$ – see front matter # 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuint.2006.07.010

ROS are involved in the pathogenesis of the neuronal deterioration associated with aging, neurodegenerative diseases or ischemia/reperfusion (for review see Halliwell and Gutteridge, 1998; Fiskum et al., 1999; Murphy et al., 1999; Nicholls and Budd, 2000; Tretter et al., 2004; Adam-Vizi, 2005). Therefore, understanding the mechanism of and conditions favoring ROS formation in brain mitochondria is of primary importance. A critical variable in studies addressing the quantitative features and supporting conditions of ROS formation in isolated brain mitochondria is the type of substrates used to fuel the respiratory chain. With the FADH-linked substrate, succinate significant amount of ROS generation has been detected in most of the studies (Patole et al., 1986; Zoccarato et al., 1988; Cino and Del Maestro, 1989; Kwong and Sohal, 1998; Votyakova and Reynolds, 2001; Liu et al., 2002) but values on the rate of ROS formation could differ by an order of magnitude in these studies.

140

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147

Furthermore, other groups detected no (Sorgato et al., 1974) or only very small amount of ROS formation with succinate as a substrate (Barja and Herrero, 1998). The rate of ROS formation was found to be smaller with NADH-linked substrates than with succinate in many studies (Patole et al., 1986; Kwong and Sohal, 1998; Gyulkhandanyan and Pennefather, 2004) with notable exceptions where either higher (Barja and Herrero, 1998; Arnaiz et al., 1999) or no ROS formation at all was detected (Cino and Del Maestro, 1989; Votyakova and Reynolds, 2001; Liu et al., 2002; Zoccarato et al., 2004). Although it has been reported recently that a key Krebs cycle enzyme, a-ketoglutarate dehydrogenase could be a significant source of ROS (Starkov et al., 2004; Tretter and Adam-Vizi, 2004, 2005; Chinopoulos and Adam-Vizi, 2006), the respiratory chain is generally viewed as the major site of ROS generation in mitochondria. This is based on the observations that inhibitors of the respiratory chain profoundly alter the rate of ROS production. However, there are also contradictions here: the complex III inhibitor, antimycin decreased (Votyakova and Reynolds, 2001), in other studies increased (Arnaiz et al., 1999; Liu et al., 2002) or did not influence (Barja, 1999) ROS formation in succinate-supported brain mitochondria. In the presence of the complex I inhibitor rotenone, the succinateevoked ROS generation was inhibited in some studies (Cino and Del Maestro, 1989; Kwong and Sohal, 1998; Votyakova and Reynolds, 2001; Gyulkhandanyan and Pennefather, 2004), but was not influenced in others (Barja and Herrero, 1998). The blockage of ROS generation by rotenone in succinatesupported mitochondria is consistent with a reverse flow of electrons from complex II and prevention of ROS production at complex I (Hinkle et al., 1967; Turrens and Boveris, 1980). It is generally accepted that generation of ROS in the respiratory chain is strongly dependent on DCm. This is based on observations showing a decreased ROS formation in response to depolarization of succinate-supported mitochondria prepared from the heart (Boveris and Chance, 1973; Korshunov et al., 1997; Hansford et al., 1997) or the brain (Cino and Del Maestro, 1989; Votyakova and Reynolds, 2001), but with NADH-linked substrates some results suggested that ROS formation was sensitive to changes in DCm (Starkov and Fiskum, 2003) others indicated the lack of dependence of ROS generation on DCm (Votyakova and Reynolds, 2001). We have also encountered these discrepancies when addressing ROS generation in isolated brain mitochondria and made an observation that the method of preparing mitochondria has a profound effect on the characteristics of ROS generation. There are two generally used techniques to isolate mitochondria. In one of them digitonin is applied to disrupt cholesterol containing plasma membranes leaving mitochondria functional, in the other, mitochondria are purified on a discontinuous Percoll gradient. In the present study characteristics of ROS generation were compared in mitochondria isolated from guinea-pig brain by these two methods. We report here that the presence of bovine serum albumin (BSA) could critically determine the mechanism and features of ROS production highlighting a critical membrane potential range in which ROS formation in succinatesupported brain mitochondria in highly sensitive to DCm.

2. Experimental procedures 2.1. Preparation of mitochondria using a discontinuous Percoll gradient Brain mitochondria were prepared from male guinea-pigs weighed 250– 300 g. The animals were decapitated, the procedure being in compliance with the Guidelines for Animal Experiments at Semmelweis University. The brain was homogenized in ‘‘buffer A’’ (in mM: 225 mannitol, 75 sucrose, 5 HEPES, 1 EGTA, pH 7.4 [KOH]) and centrifuged for 3 min at 1300  g. The supernatant was centrifuged for 10 min at 20,000  g, and then the pellet was suspended in 15% Percoll and layered on a discontinuous gradient consisting of 40 and 23% Percoll, respectively, which was then centrifuged for 8 min at 30,700  g without using brakes. After resuspension of the lowermost fraction in ‘‘buffer A’’, it was centrifuged at 16,600  g for 10 min, and then the pellet was resuspended in ‘‘buffer A’’, and centrifuged again at 6300  g for 10 min. After discharging the supernatant, the pellet was resuspended in ‘‘buffer B’’ (in mM: 225 mannitol, 75 sucrose, 5 HEPES, pH 7.4 [KOH]).

2.2. Preparation of mitochondria using digitonin The procedure in the first two centrifugation was slightly different from that described above (3000  g for 3 min and 15,000  g for 8 min), in particular ‘‘buffer A’’ was supplemented with 1 mg/ml fatty acid free BSA. The supernatant of the second centrifugation was resuspended in ‘‘buffer A’’ supplemented with BSA, digitonin was added (40 ml from 10% digitonin per guineapig brain; 0.04%) and after 1 min this was centrifuged at 15,000  g for 10 min. After removing the supernatant, the pellet was resuspended in ‘‘buffer A’’ containing BSA and centrifuged again at 15,000  g for 10 min and resuspended in the same buffer containing no EGTA. Experiments with both mitochondrial preparations were carried out in an incubation medium containing (mM): 125 KCl, 20 HEPES, 2 K2HPO4, 1 MgCl2, 0.1 EGTA, pH 7.0 (KOH).

2.3. Measurement of mitochondrial H2O2 production The assay is based on the detection of H2O2 in the medium using the Amplex Red fluorescent dye. In the presence of horseradish peroxidase, the Amplex Red reagent reacts with H2O2 with a 1:1 stoichiometry producing highly fluorescent resorufin. Horseradish peroxidase (5 U/2 ml) and Amplex Red reagent (1 mM) were added to the incubation medium described above, and then guinea-pig brain mitochondria (0.1 mg/ml) were added. H2O2 formation was initiated by addition of succinate (5 mM) and fluorescence was detected at 37 8C in a Deltascan fluorescence spectrophotometer (Photon Technology International, PTI; Lawrenceville, NJ). The excitation wavelength was 550 nm, and the fluorescence emission was detected at 585 nm. A calibration signal was generated with known amounts of H2O2 at the end of each experiment. We have not experienced any fluorescent signal in the Amplex Red assay in response to addition of BSA reported by Gyulkhandanyan and Pennefather (2004).

2.4. Detection of mitochondrial membrane potential with TMRM The method detailed by (Scaduto and Grotyohann, 1999) was followed. A 100 nM tetramethylrhodamine methyl ester (TMRM) was dissolved in the incubation medium, and then mitochondria (0.1 mg/ml) and succinate (5 mM) were added. Fluorescence was detected in the dual excitation ratiometric mode in a PTI Deltascan fluorescence spectrophotometer using 546 and 573 nm excitation and 590 nm emission wavelengths. The ratio of 546–590 and 573– 590 nm fluorescence was plotted.

2.5. Measurement of mitochondrial membrane potential with TPP+ electrode Mitochondrial membrane potential (DCm) was estimated from tetraphenylphosphonium (TPP+) ion distribution measured with a custom-made

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147 TPP+-selective electrode (Kamo et al., 1979). For these experiments, the incubation medium was supplemented with 2.0 mM TPP+Cl . The electrode was calibrated by sequential additions of TPP+Cl , and DCm was calculated utilizing the reported binding correction factor for brain mitochondria (Kamo et al., 1989) and assuming that the matrix volume for brain mitochondria is 1 ml/ mg protein (D.G. Nicholls, personal communication). For these experiments the mitochondrial protein concentration was 0.5 mg/ml.

2.6. Measurement of mitochondrial oxygen consumption Oxygen uptake by mitochondria (1 mg/1 ml) was measured with a Clarktype oxygen electrode at 37 8C using Hansatech Oxygraph Measurement System (Hansatech, Norfolk, UK). The quality of the mitochondrial preparation was estimated by measuring the respiratory control ratio (RCR) defined as ADP-stimulated (state 3) respiration divided by resting (state 4) respiration. State 3 respiration was initiated by addition of 1 mM ADP to the incubation medium. The effect of ADP was suspended and state 4 respiration was initiated by carboxyatractylate (1 mM), an inhibitor of the ADP/ATP translocator.

2.7. Materials Standard laboratory chemicals and TPP were obtained from Sigma (St. Louis, USA). The Amplex Red reagent and TMRM were from Molecular Probes (Eugene, OR, USA) and digitonin from Spectrum Chemical and Laboratory Products Inc. (New Brunswick, NJ, USA)

2.8. Statistics Statistical differences for Table 1 were calculated by Student’s t-test.

3. Results 3.1. Distinct features of succinate-supported ROS generation in mitochondria prepared with digitonin or Percoll gradient centrifugation

141

fluorescence an enhanced H2O2 generation was observed upon addition of succinate (5 mM). The rate of H2O2 generation was 2.5 times higher in mitochondria prepared with digitonin (600.6  26.8 pmol/min/mg) as compared to that measured in Percoll-purified mitochondria (252  17.3 pmol/min/mg; Table 1). An even more remarkable difference was observed in the effect of the complex I inhibitor, rotenone. In mitochondria prepared with the digitonin method, rotenone (0.5 mM) reduced the rate of H2O2 generation by about 50%, in agreement with earlier reports (Cino and Del Maestro, 1989; Kwong and Sohal, 1998; Votyakova and Reynolds, 2001; Gyulkhandanyan and Pennefather, 2004) and being consistent with a reverse flow of electrons via complex I. However, in mitochondria purified on a discontinuous Percoll gradient, H2O2 generation was not inhibited; instead, it was stimulated by rotenone (Fig. 1). Antimycin added after rotenone further increased the rate of H2O2 generation in both cases. Rotenone, itself gave a step-like fluorescent signal in the Amplex Red assay due to ethanol (<0.1%), in which it is dissolved. This is unrelated to H2O2 and was also observed by others (Gyulkhandanyan and Pennefather, 2004). The solvent, itself, was without effect on the rate of H2O2 formation (data not shown). It was obvious to control whether digitonin, per se is responsible for the distinct features of ROS production, thus we followed H2O2 generation in Percoll-purified mitochondria when digitonin (0.04%) was present in the assay medium. It was found that the characteristics of H2O2 generation shown in Fig. 1, curve b were not influenced by digitonin (data not shown). It was also considered whether the mitochondrial yield in the two methods could be different and the non-mitochondrial contamination contributes to the quantitatively and qualitatively distinct

Mitochondria were prepared using either the widely accepted digitonin method or a discontinuous Percoll gradient as described in Section 2. By continuous detection of Amplex Red Table 1 Respiratory control ratio (RCR) and rate of H2O2 formation measured in the presence and absence of BSA in mitochondria prepared by Percoll gradient purification or using digitonin RCR

H2O2

Percoll mitochondria +BSA

4.75  0.26 6.56  0.39*

252  17.3 1035  62.9*

Digitonin mitochondria +BSA

6.02  0.35 6.58  0.27

600.6  26.8 904.2  90 *

Mitochondria were prepared either by Percoll gradient purification or using digitonin as described in Section 2. During preparation with Percoll no BSA was present, whereas for preparation with digitonin BSA (0.1%) was present in the isolation medium. Respiratory control ratio (RCR) and H2O2 formation were measured as detailed in Section 2 and BSA (0.025%) was present in the incubation medium where indicated. For the calculation of RCR oxygen consumption measured in the presence of ADP (1 mM) and ADP + carboxyatractylate (1 mM) were taken into account. The rate of H2O2 generation is expressed in pmol/min/mg protein. Results are mean  S.E.M. from at least four (for RCR) or 20 experiments (for H2O2 formation). * Indicates significant differences from the corresponding values obtained in the absence of BSA ( p < 0.05). The values of H2O2 formation measured in the presence of BSA in Percoll vs. digitonin mitochondria are not significantly different.

Fig. 1. H2O2 formation in succinate-supported mitochondria prepared with digitonin (a) or using discontinuous Percoll gradient purification (b). The incubation medium was supplemented with Amplex Red and horseradish peroxidase and fluorescence was followed as detailed in Section 2. Mitochondria (m; 0.1 mg/ml), succinate (5 mM), rotenone (500 nM) and antimycin (250 nM) were added as indicated. Traces are representative of 12 independent experiments. At the end of the experiments 100 pmol H2O2 was added for calibration. Numbers indicate the rate of H2O2 formation in pmol/min/mg protein.

142

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147

features of ROS generation demonstrated in Fig. 1. However, the maximum, ADP-stimulated oxygen consumptions were found to be similar, no matter whether mitochondria were prepared with Percoll gradient purification (135  22 nanoatom O/mg protein/ min) or using digitonin (126.5  21 nanoatom O/mg protein/ min) indicating that the mitochondrial content in the two preparations is similar. 3.2. The effect of BSA on the generation of H2O2 in succinate-supported mitochondria By considering other differences in the two preparation methods it was established that when digitonin is used by different groups (and also in our laboratory), BSA is always present either in the isolation medium or during the experiments, in the incubation medium. In contrast, mitochondria prepared with Percoll centrifugation are often studied without the presence of BSA in the assay medium. We addressed the possibility that the presence of BSA could profoundly alter the features of H2O2 generation in succinatesupported mitochondria, therefore BSA (0.025%) was added to the incubation medium and the rate of H2O2 production was measured. Table 1 and Fig. 2a show that this increased the rate of H2O2 generation by only 50% in mitochondria prepared with digitonin, which was not surprising as, in this method BSA was already present in the isolation medium. However, when Percoll-prepared mitochondria were incubated in a medium containing BSA, the rate of H2O2 production was four times higher than that measured without BSA (Fig. 2b and Table 1). It is shown in Fig. 2c that upon addition of BSA, H2O2 generation was immediately stimulated. In addition, the effect of rotenone was also different in the presence of BSA; inhibition of complex I in the presence of BSA decreased the rate of H2O2 generation (Fig. 2c), whereas this resulted in an enhanced ROS production without BSA (Fig. 1, curve b). BSA augmented the RCR of mitochondria prepared with Percoll indicating that these mitochondria became more coupled in the

presence of BSA, whereas the RCR in the more coupled digitonin-prepared mitochondria was not altered in the presence of BSA (Table 1). 3.3. H2O2 generation in succinate-supported mitochondria becomes DCm-dependent in the presence of BSA It was remarkable that FCCP, in small concentration (10 nM), inhibited the generation of H2O2 when BSA was present, no matter whether mitochondria were prepared with digitonin (Fig. 2a) or Percoll (Fig. 2b), but apparently was without effect in the absence of BSA in Percoll mitochondria (Fig. 2b). This led us to study the dependence of ROS generation on DCm in the presence and absence of BSA. TMRM fluorescence was detected as a measure of DCm in the presence of different concentrations of FCCP and H2O2 generation was also determined under similar conditions. In mitochondria prepared with digitonin, addition of BSA did not influence TMRM fluorescence (Fig. 3a). FCCP added consecutively in 10 nM concentrations decreased TMRM fluorescence proportional with the FCCP concentrations, but in the presence of BSA (upper trace) the fall in TMRM fluorescence at each FCCP addition was smaller than that in the absence of BSA (lower trace). This is consistent with the ability of BSA to counteract the effect of a large number of uncouplers (Weinbach and Garbus, 1966). H2O2 generation was significantly decreased by FCCP both in the presence and absence of BSA. In Percoll-prepared mitochondria TMRM fluorescence was increased upon addition of BSA indicating hyperpolarization of mitochondria (Fig. 4a). The effect of FCCP was similar to that observed with digitoninprepared mitochondria: a stepwise decrease in TMRM fluorescence in response to addition of FCCP in 10 nM concentrations, which was always smaller in the presence of BSA (upper trace, in Fig. 4a). The rate of H2O2 production was greatly enhanced in the presence of BSA, as already demonstrated above (Fig. 2b), and this was drastically

Fig. 2. The effect of BSA on the H2O2 formation in mitochondria prepared with digitonin (a) or using Percoll gradient centrifugation (b and c). H2O2 in the medium was measured as for Fig. 1 except that BSA (0.025%) was present in the medium for (a and b) where indicated. For (c) BSA was given to the medium after succinate (5 mM). The concentrations of rotenone and antimycin were 500 and 250 nM, respectively. FCCP (10 nM) was added at the end of the measurements in (a and b). Traces are representative of at least five experiments.

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147

143

Fig. 3. Membrane potential as indicated by TMRM fluorescence (a) and the rate of H2O2 formation (b) in digitonin-prepared mitochondria. For (a) the incubation medium was supplemented with TMRM and the fluorescence was measured as detailed in Section 2. Mitochondria (m; 0.1 mg/ml), succinate (5 mM), BSA (0.025%, upper trace) and FCCP consecutively in 10 nM concentrations were given as indicated. (b) The rate of H2O2 formation as measured in the presence of different concentrations of FCCP. For the upper curve, BSA (0.025%) was present in the incubation medium. Points represent the average  S.E.M. from five independent experiments.

decreased by FCCP, again in a small concentration range (Fig. 4b). Without BSA, FCCP only slightly inhibited ROS generation. It is remarkable that generation of H2O2 was inhibited by FCCP (10–50 nM), when the change in TMRM fluorescence, in particular in the presence of BSA, was very small. We did not relate here quantitatively the fall in DCm by FCCP to the decrease in H2O2 generation but TMRM fluorescence assumed to be linear with the membrane potential (Scaduto and Grotyohann, 1999). Nonetheless, it is evident from Fig. 3b and more so from Fig. 4b that in the presence of BSA, small decrease in DCm is paralleled a significant reduction in H2O2 formation. Increase in FCCP concentration to higher than 50 nM did not reduce further the formation of H2O2, in spite of a further drop in TMRM fluorescence (Figs. 3 and 4). This indicated that ROS generation in succinate-supported mitochondria is dependent on DCm only in a small DCm range, and only at high DCm values.

3.4. Hyperpolarization of DCm by BSA In order to determine the extent of the BSA-induced hyperpolarization observed in Percoll-prepared mitochondria with TMRM fluorescence (Fig. 4a), we also measured DCm using a TPP electrode, which allows the quantification of DCm in mV. The calculation is based on the distribution of the lipophilic cation, tetraphenylphosphonium (TPP+) between the medium and mitochondria. It is demonstrated in Fig. 5a that in mitochondria prepared with digitonin (and BSA present in the isolation medium) DCm is set at a more hyperpolarized value ( 187 mV in this experiment; the average is 185  4 mV; n = 3) and addition of BSA caused further hyperpolarization by only a few mV. In Percoll-prepared mitochondria (Fig. 5b) the resting DCm is more depolarized ( 173 mV in this experiment, the average is 171  2 mV; n = 3) as compared to that measured in mitochondria prepared with digitonin. Addition of BSA caused a significant hyperpolarization (by 10.7  1.9 mV;

Fig. 4. Membrane potential as indicated by the TMRM fluorescence (a) and the rate of H2O2 formation (b) in mitochondria prepared with Percoll gradient centrifugation. Experiments were carried out and data are demonstrated as described for Fig. 3 except that mitochondria prepared with the Percoll method were used. Points represent the average  S.E.M from five independent experiments.

144

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147

Fig. 5. The effect of BSA on membrane potential of mitochondria prepared with digitonin (a) or Percoll gradient centrifugation (b) as measured with a TPP+ electrode. Membrane potential was established in mitochondria upon addition of succinate (5 mM). BSA (0.025%) and FCCP (1 mM) were given as indicated. Traces are representative of three independent measurements. The transient depolarization induced by ADP (40 mM) is also demonstrated.

n = 3) bringing DCm close to the value measured in digitoninprepared mitochondria (Fig. 5b). 3.5. The effect of palmitate on respiration, DCm and H2O2 generation; prevention of the effect of palmitate by BSA Since the effect of BSA on DCm and on the production of H2O2 demonstrated above is most likely due to preventing the effects of endogenous fatty acids, we applied exogenous palmitate and recorded respiration, DCm and generation of H2O2. As expected, palmitate (2–10 mM) accelerated respiration and induced depolarization of DCm (Fig. 6a, b) consistent with its uncoupling effect reported earlier (Andreyev et al., 1988; Brustovetsky et al., 1990; Schonfeld, 1990). Upon addition of BSA after palmitate, oxygen consumption was decreased and DCm started to recover, i.e., the effect of palmitate was reversed (Fig. 6a and b). We also demonstrate here that palmitate (2 or 10 mM) added to succinatesupported mitochondria decreased the production of H2O2 and addition of BSA (0.025%) reversed the effect of palmitate and stimulated the formation of H2O2 (Fig. 6c).

4. Discussion This study shows that the presence of BSA during isolation or incubation of mitochondria profoundly influences the mechanism by which ROS is generated in the presence of succinate. The key factor in the effect of BSA appears to be a significant hyperpolarization of mitochondria bringing DCm into a range in which reverse electron flow with succinate as a substrate is favorable, thus ROS formation becomes highly dependent on DCm. Below this DCm range, mitochondria still produce ROS, though in significantly smaller amount, but this is no longer related to a reverse electron flow and no longer dependent on DCm. An early systematic study demonstrated that BSA is able to restore the respiration in mitochondria treated with uncouplers of different chemical structure but does not influence the effect of inhibitors of respiration and ATP synthase (Weinbach and Garbus, 1966). In mitochondria untreated by added uncouplers, as in our study, the effect of BSA should be resulted from prevention of the effects of endogenous uncouplers, most likely,

Fig. 6. The effect of palmitate on the respiration (a), membrane potential (b) and H2O2 formation (c) in mitochondria prepared with Percoll gradient centrifugation. Palmitate (dissolved in ethanol) in 2 or 10 mM concentrations and BSA (0.025%) were added as indicated. For comparison, the effect of ADP (1 mM) and carboxyatractylate (CAT; 1 mM) on the respiration of mitochondria (1 mg/ml) is also shown (a). Numbers indicate oxygen consumption in nanoatom/min/mg protein. Membrane potential (b) was measured with a TPP+ electrode as detailed in Section 2. The experimental design for measuring the formation of H2O2 (c) was similar to that described for Fig. 1.

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147

fatty acids released from membranes during preparation or incubation of mitochondria. The uncoupling effect of free fatty acids in isolated liver and muscle mitochondria has been revealed by demonstrating an increased respiration in response to addition of palmitate (Andreyev et al., 1988; Brustovetsky et al., 1990; Schonfeld, 1990). The effect of palmitate was inhibited by BSA and carboxyatractylate (Brustovetsky et al., 1990) and a fatty acid cycling was hypothesized to mediate H+ translocation (Skulachev, 1991), in which protonated fatty acid are transported into mitochondria involving the adenine nucleotide translocase and fatty acid anions leave mitochondria through the lipid bilayer. Uncoupling by fatty acids may also involve other proteins such as UCPs or the dicarboxylate or glutamate/ aspartate antiporter (for review see Wojtczak and Schonfeld, 1993; Skulachev, 1998). Certain members of uncoupling proteins (UCP2, UCP4, UCP5) are expressed in the brain (Fleury et al., 1997; Kim-Han et al., 2001; Andrews et al., 2005) and it has been suggested that they could have a role in neuroprotection (Bechmann et al., 2002; Andrews et al., 2005; Conti et al., 2005). Kim-Han et al. (2001) reported that when BMCP1 (also referred to as UCP5) is overexpressed in GT-1 cells, DCm decreases. In their experiments the high state 4 respiration was reversed by BSA suggesting that BMCP1 may be activated by low concentrations of fatty acids. Inhibition of H2O2 formation in succinate-supported mitochondria by the uncoupler CCCP was first described by Boveris and Chance (1973) for mitochondria isolated from pigeon heart and later this was confirmed (Hansford et al., 1997) and also demonstrated for brain mitochondria (Cino and Del Maestro, 1989). A steep dependence of H2O2 formation on DCm in state 4 rat heart mitochondria fuelled with succinate was suggested by the result showing that about 13% decrease in DCm resulted in about 80% inhibition of H2O2 generation in the presence of an uncoupler (Korshunov et al., 1997). Similar observation was made by Votyakova and Reynolds (2001) for brain mitochondria using the same DCm-sensitive dye, safranin O to detect DCm. Our results are in agreement with these, and show that H2O2 formation in mitochondria respiring on succinate is dependent on DCm in a narrow DCm range. In addition, our quantitative data on DCm obtained with TPP+ electrode showed that this narrow range is about 10 mVand for H2O2 formation to be dependent on DCm, mitochondria have to be highly polarized. With BSA we could demonstrate that when mitochondria with high DCm value are depolarized by about 10.7  1.9 mV, the membrane potential falls below the range in which H2O2 is formed by a reverse flow of electrons and is sensitive to changes in DCm. Mitochondria prepared by the digitonin method with BSA present in the isolation medium are highly coupled, with DCm in the highly polarized range, where ROS is generated with a relatively high rate, through a reverse electron flow. In reverse electron flow, electrons are transferred from succinate to NAD+ via complex I leading to NADH formation (Chance and Hollunger, 1961a; Chance and Hollunger, 1961b; Hinkle et al., 1967). This mechanism is apparently lacking in mitochondria prepared by Percoll purification, without BSA, which are less coupled, have a more depolarized DCm and produce ROS with

145

smaller rate and via a forward flow of electrons in the respiratory chain. BSA added to these mitochondria, sets favorable conditions for a reverse electron transfer, by eliminating the uncoupling effects of endogenous fatty acids. Lambert and Brand (2004) have demonstrated recently that superoxide generation by complex I in succinate-supported skeletal muscle mitochondria are sensitive not only to DCm but also to DpH, the other component of the protonmotive force. We have not addressed DpH in this study, which together with DCm is totally abolished by FCCP. Reverse flow of electrons from complex II through complex I is inhibited by rotenone and inhibition of ROS formation by rotenone in succinate-supported brain mitochondria is an important feature of ROS generation occurring at complex I via a reverse electron transfer (Cino and Del Maestro, 1989; Votyakova and Reynolds, 2001; Liu et al., 2002). Below a critical DCm, ROS formation, which is smaller but still significant, is no longer related to reverse electron flow. Consistently, after depolarization by FCCP, the decreased H2O2 formation was no longer inhibited by rotenone, instead inhibition of complex I resulted in an increase in ROS generation (data not shown). Stimulation of ROS generation by rotenone would indicate forward flow of electrons via complex I and generation of ROS proximal from the rotenone binding site. This study highlights important features of ROS formation influenced by the presence of BSA in isolated brain mitochondria. The great deal of quantitative and qualitative diversities found in the literature (see Section 1) concerning ROS formation in brain mitochondria can be ascribed to many factors including species and tissue variability, different sensitivity of methods applied for detecting ROS formation and differences in the ‘quality’ of mitochondria prepared by different groups. The absence of BSA could explain the different characteristics of H2O2 formation in mitochondria prepared with Percoll gradient centrifugation as compared to that in mitochondria isolated by digitonin disruption. Digitonin per se, was without effect in our experiments, unlike in those by Brustovetsky et al. (2002) where addition of digitonin (>0.001%) to Percoll purified mitochondria produced swelling of slow kinetics and release of detectable amount of Cyt c. The characteristics of ROS formation in digitonin-prepared mitochondria in our study was clearly associated with the presence of BSA during the preparation as in mitochondria prepared by the same method but without BSA, ROS formation was of similar qualitative and quantitative features as that observed in Percoll-purified mitochondria (data not shown). This also eliminates a possible concern that with the two different methods mitochondria of different tissue origin (astrocytic versus neuronal) might be prepared having different characteristics. It has to be emphasized that the DCm-dependence of ROS generation in brain mitochondria described here and earlier studies is restricted only to ROS formation occurring in succinate-supported mitochondria and does not allow the conclusion that ROS formation generally, and in particular in vivo is DCm dependent. It is questionable, whether reverse electron flow is physiologically relevant and mitochondria in situ or even in vivo could ever have such a high DCm, where this

146

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147

is possible and where depolarization by a few mV would result in a significant decrease in ROS formation. Further detailed studies are required to determine the DCm-dependence of ROS generation in isolated brain mitochondria supported by the more physiological NADH-linked substrates and in mitochondria in situ, which receive the substrates from glucose metabolism. These studies could help to clarify whether a decrease in DCm could be expected to result in a reduced ROS formation in situ, and by this, would be a reasonable neuroprotective strategy, or the strong DCm-dependence of ROS generation would be characteristic for good quality isolated mitochondria respiring on succinate, as shown in this work, in which the non-physiological reverse electron flow is favored and responsible for ROS generation. Acknowledgements Authors are grateful to Katalin Takacs and Andrea Varnagy for the excellent technical assistance. Thanks are expressed to A. Starkov for the helpful discussion during the preparation of the manuscript. This work was supported by OTKA to VA-V and LT and by ETT and RET-NKTH and the Hungarian Academy of Sciences to VA-V. References Adam-Vizi, V., 2005. Production of reactive oxygen species in brain mitochondria: contribution by electron transport chain and non-electron transport chain sources. Antioxid. Redox Signal. 7, 1140–1149. Andrews, Z.B., Horvath, B., Barnstable, C.J., Elsworth, J., Yang, L., Beal, M.F., Roth, R.H., Matthews, R.T., Horvath, T.L., 2005. Uncoupling protein-2 is critical for nigral dopamine cell survival in a mouse model of Parkinson’s disease. J. Neurosci. 25, 184–191. Andreyev, A.Y., Bondareva, T.O., Dedukhova, V.I., Mokhova, E.N., Skulachev, V.P., Volkov, N.I., 1988. Carboxyatractylate inhibits the uncoupling effect of free fatty acids. FEBS Lett. 226, 265–269. Arnaiz, S.L., Coronel, M.F., Boveris, A., 1999. Nitric oxide, superoxide, and hydrogen peroxide production in brain mitochondria after haloperidol treatment. Nitric Oxide 3, 235–243. Barja, G., 1999. Mitochondrial oxygen radical generation and leak: sites of production in states 4 and 3, organ specificity, and relation to aging and longevity. J. Bioenerg. Biomembr. 31, 347–366. Barja, G., Herrero, A., 1998. Localization at complex I and mechanism of the higher free radical production of brain nonsynaptic mitochondria in the shortlived rat than in the longevous pigeon. J. Bioenerg. Biomembr. 30, 235–243. Bechmann, I., Diano, S., Warden, C.H., Bartfai, T., Nitsch, R., Horvath, T.L., 2002. Brain mitochondrial uncoupling protein 2 (UCP2): a protective stress signal in neuronal injury. Biochem. Pharmacol. 64, 363–367. Boveris, A., Chance, B., 1973. The mitochondrial generation of hydrogen peroxide. General properties and effect of hyperbaric oxygen. Biochem. J. 134, 707–716. Brustovetsky, N., Jemmerson, R., Dubinsky, J.M., 2002. Calcium-induced cytochrome c release from rat brain mitochondria is altered by digitonin. Neurosci. Lett. 332, 91–94. Brustovetsky, N.N., Amerkanov, Z.G., Yegorova, M.E., Mokhova, E.N., Skulachev, V.P., 1990. Carboxyatractylate-sensitive uncoupling in liver mitochondria from ground squirrels during hibernation and arousal. FEBS Lett. 272, 190–192. Chance, B., Hollunger, G., 1961a. The interaction of energy and electron transfer reactions in mitochondria. I. General properties and nature of the products of succinate-linked reduction of pyridine nucleotide. J. Biol. Chem. 236, 1534–1543.

Chance, B., Hollunger, G., 1961b. The interaction of energy and electron transfer reactions in mitochondria. IV. The pathway of electron transfer. J. Biol. Chem. 236, 1562–1568. Chinopoulos, C., Adam-Vizi, V., 2006. Calcium, mitochondria and oxidative stress in neuronal pathology. Novel aspects of an enduring theme. FEBS J. 273, 433–450. Cino, M., Del Maestro, R.F., 1989. Generation of hydrogen peroxide by brain mitochondria: the effect of reoxygenation following postdecapitative ischemia. Arch. Biochem. Biophys. 269, 623–638. Conti, B., Sugama, S., Lucero, J., Winsky-Sommerer, R., Wirz, S.A., Maher, P., Andrews, Z., Barr, A.M., Morale, M.C., Paneda, C., Pemberton, J., Gaidarova, S., Behrens, M.M., Beal, F., Sanna, P.P., Horvath, T., Bartfai, T., 2005. Uncoupling protein 2 protects dopaminergic neurons from acute 1,2,3,6methyl-phenyl-tetrahydropyridine toxicity. J. Neurochem. 93, 493–501. Fiskum, G., Murphy, A.N., Beal, M.F., 1999. Mitochondria in neurodegeneration: acute ischemia and chronic neurodegenerative diseases. J. Cereb. Blood Flow Metab. 19, 351–369. Fleury, C., Neverova, M., Collins, S., Raimbault, S., Champigny, O., LeviMeyrueis, C., Bouillaud, F., Seldin, M.F., Surwit, R.S., Ricquier, D., Warden, C.H., 1997. Uncoupling protein-2: a novel gene linked to obesity and hyperinsulinemia. Nat. Genet. 15, 269–272. Gyulkhandanyan, A.V., Pennefather, P.S., 2004. Shift in the localization of sites of hydrogen peroxide production in brain mitochondria by mitochondrial stress. J. Neurochem. 90, 405–421. Halliwell, B., Gutteridge, J.M.C., 1998. Free radicals, other reactive species and disease. In: Halliwell, B., Gutteridge, J.M.C. (Eds.), Free Radicals in Biology Medicine. Oxford UP, Oxford, pp. 617–784. Hansford, R.G., Hogue, B.A., Mildaziene, V., 1997. Dependence of H2O2 formation by rat heart mitochondria on substrate availability and donor age. J. Bioenerg. Biomembr. 29, 89–95. Hinkle, P.C., Butow, R.A., Racker, E., Chance, B., 1967. Partial resolution of the enzymes catalyzing oxidative phosphorylation. XV. Reverse electron transfer in the flavin-cytochrome beta region of the respiratory chain of beef heart submitochondrial particles. J. Biol. Chem. 242, 5169–5173. Kamo, N., Demura, M., Kobatake, Y., 1989. Linear-equation for calibration of the mitochondrial-membrane potential. Bioelectrochem. Bioenerg. 21, 33–46. 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 potential and phosphorylation potential in steady state. J. Membr. Biol. 49, 105–121. Kim-Han, J.S., Reichert, S.A., Quick, K.L., Dugan, L.L., 2001. BMCP1: a mitochondrial uncoupling protein in neurons which regulates mitochondrial function and oxidant production. J. Neurochem. 79, 658–668. Korshunov, S.S., Skulachev, V.P., Starkov, A.A., 1997. High protonic potential actuates a mechanism of production of reactive oxygen species in mitochondria. FEBS Lett. 416, 15–18. Kwong, L.K., Sohal, R.S., 1998. Substrate and site specificity of hydrogen peroxide generation in mouse mitochondria. Arch. Biochem. Biophys. 350, 118–126. Lambert, A.J., Brand, M.D., 2004. Superoxide production by NADH:ubiquinone oxidoreductase (complex I) depends on the pH gradient across the mitochondrial inner membrane. Biochem. J. 382, 511–517. Liu, Y., Fiskum, G., Schubert, D., 2002. Generation of reactive oxygen species by the mitochondrial electron transport chain. J. Neurochem. 80, 780–787. Murphy, A.N., Fiskum, G., Beal, M.F., 1999. Mitochondria in neurodegeneration: bioenergetic function in cell life and death. J. Cereb. Blood Flow Metab. 19, 231–245. Nicholls, D.G., Budd, S.L., 2000. Mitochondria and neuronal survival. Physiol. Rev. 80, 315–360. Patole, M.S., Swaroop, A., Ramasarma, T., 1986. Generation of H2O2 in brain mitochondria. J. Neurochem. 47, 1–8. Scaduto Jr., R.C., Grotyohann, L.W., 1999. Measurement of mitochondrial membrane potential using fluorescent rhodamine derivatives. Biophys. J. 76, 469–477. Schonfeld, P., 1990. Does the function of adenine nucleotide translocase in fatty acid uncoupling depend on the type of mitochondria? FEBS Lett. 264, 246– 248.

L. Tretter et al. / Neurochemistry International 50 (2007) 139–147 Skulachev, V.P., 1991. Fatty acid circuit as a physiological mechanism of uncoupling of oxidative phosphorylation. FEBS Lett. 294, 158–162. Skulachev, V.P., 1998. Biochemical mechanisms of evolution and the role of oxygen. Biochemistry (Mosc.) 63, 1335–1343. Sorgato, M.C., Sartorelli, L., Loschen, G., Azzi, A., 1974. Oxygen radicals and hydrogen peroxide in rat brain mitochondria. FEBS Lett. 45, 92–95. Starkov, A.A., Fiskum, G., 2003. Regulation of brain mitochondrial H2O2 production by membrane potential and NAD(P)H redox state. J. Neurochem. 86, 1101–1107. Starkov, A.A., Fiskum, G., Chinopoulos, C., Lorenzo, B.J., Browne, S.E., Patel, M.S., Beal, M.F., 2004. Mitochondrial alpha-ketoglutarate dehydrogenase complex generates reactive oxygen species. J. Neurosci. 24, 7779–7788. Tretter, L., Adam-Vizi, V., 2004. Generation of reactive oxygen species in the reaction catalyzed by alpha-ketoglutarate dehydrogenase. J. Neurosci. 24, 7771–7778. Tretter, L., Adam-Vizi, V., 2005. Alpha-ketoglutarate dehydrogenase: a target and generator of oxidative stress. Philos. Trans. R. Soc. Lond. B Biol. Sci. 360, 2335–2345.

147

Tretter, L., Sipos, I., Adam-Vizi, V., 2004. Initiation of neuronal damage by complex I deficiency and oxidative stress in Parkinson’s disease. Neurochem. Res. 29, 569–577. Turrens, J.F., Boveris, A., 1980. Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochem. J. 191, 421–427. Votyakova, T.V., Reynolds, I.J., 2001. DeltaPsi(m)-dependent and -independent production of reactive oxygen species by rat brain mitochondria. J. Neurochem. 79, 266–277. Weinbach, E.C., Garbus, J., 1966. The rapid restoration of respiratory control to uncoupled mitochondria. J. Biol. Chem. 241, 3708–3713. Wojtczak, L., Schonfeld, P., 1993. Effect of fatty acids on energy coupling processes in mitochondria. Biochim. Biophys. Acta 1183, 41–57. Zoccarato, F., Cavallini, L., Alexandre, A., 2004. Respiration-dependent removal of exogenous H2O2 in brain mitochondria: inhibition by Ca2+. J. Biol. Chem. 279, 4166–4174. Zoccarato, F., Cavallini, L., Deana, R., Alexandre, A., 1988. Pathways of hydrogen peroxide generation in guinea pig cerebral cortex mitochondria. Biochem. Biophys. Res. Commun. 154, 727–734.