- Email: [email protected]
Dissociation of superoxide production by mitochondrial complex I from NAD(P)H redox state
FEBS Letters 582 (2008) 1711–1714
Dissociation of superoxide production by mitochondrial complex I from NAD(P)H redox state Adrian J. Lambert*, Julie A. Buckingham, Martin D. Brand MRC Dunn Human Nutrition Unit, Hills Road, Cambridge CB2 0XY, UK Received 26 March 2008; revised 14 April 2008; accepted 15 April 2008 Available online 28 April 2008 Edited by Barry Halliwell
Abstract The relationship between the rate of superoxide production by complex I and NAD(P)H redox state was investigated in rat skeletal muscle mitochondria. A high rate of superoxide production was observed during succinate oxidation; the rate during pyruvate oxidation was over fourfold lower. However, the NAD(P)H pool was significantly less reduced during succinate oxidation than during pyruvate oxidation. We conclude that there is no unique relationship between superoxide production by complex I and the reduction state of the NAD(P)H pool. Our data suggest that less than 10% of the superoxide originates from the flavin site during reverse electron transport from succinate. Ó 2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: NADH:ubiquinone oxidoreductase, Mitochondria; Superoxide; NAD(P)H; Hydrogen peroxide; Reactive oxygen species
1. Introduction NADH-ubiquinone oxidoreductase (complex I) is an important proton-pumping enzyme of the mitochondrial electron transport chain which can also convert molecular oxygen to the superoxide anion radical [1]. The discovery of superoxide dismutase (SOD), a naturally occurring enzyme that catalyses the dismutation of superoxide to hydrogen peroxide demonstrated the relevance of superoxide in vivo [2]. This relevance was confirmed by the demonstration that mice null for the mitochondrial form of SOD (MnSOD or SOD2) exhibit severe pathologies and have a markedly curtailed lifespan [3]. In addition, several pathologies are associated with defects in complex I and elevated production of superoxide [4]. It is, therefore, of great importance to understand the sites, mechanisms and regulation of superoxide production by complex I. Complex I contains a number of redox centres that are potentially capable of producing superoxide. However, as yet there is no consensus view on the site or sites of production of superoxide by complex I under physiological conditions (such as in vivo or in intact cells) or conditions that approach * Corresponding author. Fax: +44 (0) 1223 252805. E-mail address: [email protected] (A.J. Lambert).
Abbreviations: SOD, superoxide dismutase; FCCP, carbonylcyanide-ptrifluoromethoxyphenylhydrazone; ROS, reactive oxygen species
physiological (such as in energised, intact mitochondria). In the isolated enzyme, the fully reduced flavin, FMNH has been identified as the most likely electron donor to oxygen to produce superoxide [1]. Whether the reduced flavin remains the major site of production in a membrane-bound, protonpumping complex I is an unresolved, yet pressing question. There is additional evidence that the flavin produces superoxide [5], but other sites such as the iron–sulfur centres and the quinone site have been suggested also [6–10]. In intact mitochondria, the situation is complicated by the ability of complex I to undergo reverse electron transport, which results in very high rates of superoxide production compared to the rate observed during forward electron transport [11]. In this study, we investigated whether the flavin is the main site of superoxide production in intact mitochondria during reverse electron transport. We based our experiments on the observation that in isolated complex I, the rate of superoxide production from the flavin at fixed [NADH + NAD+] is dependent on the NADH/NAD+ ratio, with the reaction being strongly inhibited by NAD+ [1]. A high NADH/NAD+ ratio results in a highly reduced flavin (FMNH ) which leads to a high rate of superoxide production. Therefore, if superoxide originates exclusively from the flavin under all conditions, there should be a unique relationship between NADH redox state and superoxide production. Our data demonstrate that there is not a unique relationship: under conditions where superoxide production is high (during reverse electron transport), the NADH/NAD+ ratio is significantly lower than under conditions where superoxide production is low (during forward electron transport). By inference, superoxide is not originating from the flavin during reverse electron transport in our system.
2. Materials and methods Mitochondria (with an average respiratory control ratio of 2.0 ± 0.2 on succinate) from skeletal muscle of female Wistar rats (aged between 5 and 8 weeks) were isolated as described [12]. Superoxide production rate was assessed indirectly by measurement of hydrogen peroxide generation rate, determined fluorometrically by measurement of oxidation of amplex red to fluorescent resorufin. Mitochondria were incubated at 0.35 mg mitochondrial protein ml 1 in buffer containing 120 mM potassium chloride, 3 mM HEPES, 1 mM EGTA and 0.3% BSA (w/ v) (pH 7.2 and 37 °C). All incubations also contained 50 lM amplex red, 4 U ml 1 horseradish peroxidase, 30 U ml 1 SOD and 1 lg ml 1 oligomycin. The reaction was initiated by addition of respiratory substrates (either 5 mM succinate or 2.5 mM pyruvate + 2.5 mM malate) and the increase in fluorescence at an excitation of 563 nm and an emission of 587 nm was monitored. Standard curves were generated using known amounts of hydrogen peroxide, these standard curves
0014-5793/$34.00 Ó 2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2008.04.030
1712 were not affected significantly by substrates or inhibitors. We assessed superoxide production by more direct methods (aconitase inhibition and dihydroethidine fluorescence) and found qualitatively similar findings to the amplex red assay (data not shown). NAD(P)H fluorescence was determined at excitation and emission wavelengths of 365 and 450 nm, respectively [13]. Incubation conditions were identical to those used for hydrogen peroxide measurement. We measured changes in fluorescence in this study to track changes in NADH, but we refer to what we measured as NAD(P)H. This is because NADH and NADPH exhibit the same fluorescent spectra and like other studies that measured the NAD(P)H reduction state in mitochondria, we assumed that the changes in fluorescence we measured were reporting changes in NADH as opposed to changes in NADPH. As discussed previously, this is a reasonable assumption to make [14], but at this stage we cannot completely rule out the possibility that there are complex factors that influence the fluorescence of the NADH and NADPH pools that may affect interpretation of our results. Calculations of matrix pH and values of DpH were based on data from [8].
3. Results We first compared hydrogen peroxide production and NAD(P)H fluorescence during succinate oxidation to NADH-linked oxidation induced with pyruvate and malate
A.J. Lambert et al. / FEBS Letters 582 (2008) 1711–1714
in the presence of complex I inhibitor rotenone (Fig. 1). As previously reported by this laboratory and others [8,15–17], the rate of hydrogen peroxide production was substantially greater during succinate oxidation than during NADH-linked oxidation (Fig. 1A and C). It is generally understood that the high rate observed with succinate originates as superoxide production from complex I during reverse electron transport, since about 95% of the signal can be abolished by rotenone. Under identical conditions, the levels of NAD(P)H fluorescence were significantly lower during succinate oxidation than during NADH-linked oxidation (Fig. 1B and D). To test whether the results obtained with rotenone were specific to that inhibitor, we examined other conditions during NADH-linked oxidation. In the presence of complex I inhibitor piericidin, the rate of hydrogen peroxide production (Fig. 2, point ‘‘c’’) was significantly lower than the rate observed during succinate oxidation (Fig. 2, point ‘‘b’’). The rate of hydrogen peroxide production with the complex III centre ÕoÕ inhibitor stigmatellin (Fig. 2, point ‘‘d’’) was also significantly lower than the rate seen with succinate. The lowest rate of hydrogen peroxide production was seen in the absence of any inhibitor (Fig. 2, point ‘‘e’’). In all three cases, the levels of NAD(P)H fluorescence were higher than the levels observed
Fig. 1. Rates of hydrogen peroxide production and levels of NAD(P)H fluorescence in rat skeletal muscle mitochondria. Panels A and B show traces of resorufin and NAD(P)H fluorescence over time, respectively. Dashed lines, no additions; solid lines, succinate added at t = 35 s; dotted lines, pyruvate + malate + 2 lM rotenone added at t = 35 s. Rates of hydrogen peroxide production were calculated from slopes of the traces between 100 and 150 s in conjunction with a standard curve. The average readings between 100 and 150 s were used to calculate the level of NAD(P)H fluorescence. Fully oxidised NAD(P)H was determined in the presence of 2 lM carbonylcyanide-p-trifluoromethoxyphenylhydrazone (FCCP) (not shown). The experiments in A and B were repeated and the means ± S.E. are shown in panels C and D. *Significant (p < 0.05) difference vs. succinate (n = 4 separate mitochondrial preparations).
A.J. Lambert et al. / FEBS Letters 582 (2008) 1711–1714
Fig. 2. Rates of hydrogen peroxide production and levels of NAD(P)H fluorescence. Mitochondria were incubated as described in Materials and Methods and the rates of hydrogen peroxide production and NAD(P)H fluorescence levels were calculated as described in Fig. 1 legend. Point a, pyruvate + malate + 2 lM rotenone (NAD(P)H = 100%); point b, succinate; point c, pyruvate + malate + 1 lM piericidin; point d, pyruvate + malate + 80 nM stigmatellin; point e, pyruvate + malate. Each point is the mean ± S.E. of n = 4 separate mitochondrial preparations.
with succinate. Taken together with point ‘‘a’’ derived from Fig. 1, these results show no evidence for a single, unique relationship between superoxide production rate from complex I and the level of NAD(P)H fluorescence during both forward and reverse electron transport. However, we can assume that there is a unique relationship between superoxide production from the flavin and NAD(P)H reduction level during forward electron transport specifically [1]. By drawing a curve through zero and points a, c, d and e, we estimate (by dropping a vertical line from point ‘‘b’’ to that curve) that the rate of superoxide generation from the flavin during reverse electron transport at 90% reduction of NAD(P)H is about 0.2/2.3, or <10% of the total rate. During succinate oxidation, the electron transport chain pumps protons from the mitochondrial matrix to the intermembrane space resulting in alkalinisation of the matrix and generation of an electrochemical proton gradient. Under our
1713
conditions the chemical component of the gradient, DpH, is about 0.5 pH units [8]. Therefore, when the pH of the buffer is 7.2 the pH of the mitochondrial matrix is approximately 7.7. In the presence of NADH-linked substrates and inhibitors however, there is no proton pumping and the matrix pH is equal to the external pH which is determined by the buffering system. Therefore, we tested the effects of matrix pH on hydrogen peroxide production rate and NAD(P)H fluorescent levels to see if the differences between succinate and NADH-linked oxidation could be explained by differences in matrix pH. For experiments with succinate, buffers with pH 6.5, 7 and 7.5 were used to give matrix pH values of about 7, 7.5 and 8, respectively (based on data in [8]). For experiments with NADH-linked oxidation, buffers with pH of 7, 7.5 and 8 were used. The results are displayed in Fig. 3, with the data for pH 7.2 and 7.7 included. Panels A and B indicate that changes in matrix pH do not have any significant effects on hydrogen peroxide production rate or NAD(P)H fluorescence. Importantly, the differences in hydrogen peroxide production rate between succinate and NADH-linked oxidation were significant at all values of matrix pH. In terms of NAD(P)H fluorescence, the differences were significant at 2 out of the 3 values of matrix pH. Therefore, changes in matrix pH are not responsible for the observed differences in superoxide production rates and NAD(P)H levels between forward and reverse electron transport.
4. Discussion As yet, there is no clear demonstration of the site, or sites, of superoxide production by mitochondrial complex I under physiological conditions. There is strong evidence that in the isolated enzyme (which is distinct from a membrane-bound, proton-pumping enzyme subject to protonmotive force and other regulators), the site is the fully reduced flavin, FMNH , and that the rate depends on the NADH/NAD+ ratio [1]. At low concentrations, NADH and NAD+ are in equilibrium with the flavin site. At the higher concentrations of NADH + NAD+ in the mitochondrial matrix, inhibition by NAD+ may prevent equilibrium, but even so, at fixed total NAD+ +
Fig. 3. Effect of mitochondrial matrix pH on hydrogen peroxide production rate and NAD(P)H fluorescence levels. Mitochondria were incubated as described in Section 2, except that the pH of the buffer was varied between 6.5 and 8. Values of DpH were based on data in [8]. Closed symbols, succinate; open symbols, pyruvate + malate + 1 lM piericidin. Panel A shows matrix pH vs. hydrogen peroxide production rate, B shows matrix pH vs. NAD(P)H fluorescence level. *Significant (p < 0.05) difference vs. succinate at the same pH (n = 3 separate mitochondrial preparations).
1714
NADH, the NADH/NAD+ ratio will determine the superoxide production rate. If superoxide originates at the flavin under all conditions, then a single, unique relationship should exist between the NADH/NAD+ ratio and the superoxide production rate. Such relationships have been reported for complex I in intact mitochondria, but for forward electron transport only [18,19]; in these experiments the superoxide is most likely coming from the flavin. However we report here a condition that is clearly an exception: reverse electron transport where superoxide production rate is very high compared to forward electron transport, yet the NADH/NAD+ ratio is relatively low. Therefore, there is not a unique relationship between the NADH/ NAD+ ratio and superoxide production rate, and this implies that most superoxide does not originate at the flavin site during reverse electron transport. Our estimate from Fig. 2 is that less than 10% of it does. It is well established that the rate of superoxide production during succinate oxidation is much greater than the rate during oxidation of succinate in the presence of rotenone [11]. Therefore, the superoxide production rate from sites other than complex I (such as complex III) is low compared to the rate originating from complex I during reverse electron transport. We conclude that any small contribution from complex III (or any other sites) to the overall rate of superoxide production is negligible and does not alter the conclusion of our study. Our data are supported by previous work that measured NAD(P)H and superoxide production, although the aims of those studies were different from ours. Hansford et al. reported a rate of hydrogen peroxide production from heart mitochondria oxidising glutamate, malate and rotenone (where the NAD(P)H pool is fully reduced) of 0.22 ± 0.03 nmol/min/mg [15]. During succinate oxidation the rate was higher (0.63 ± 0.12 nmol/min/mg), yet the NAD(P)H pool was 22% less reduced. Qualitatively similar findings were demonstrated for heart mitochondria by Kushnareva et al. [7]. Therefore, our findings are not specific to our system, and are likely to apply to mitochondria from different tissues under a variety of different conditions (such as buffer composition). Other isolated mitochondrial enzymes, such as pyruvate dehydrogenase and a-ketoglutarate dehydrogenase, have been demonstrated to produce reactive oxygen species (ROS) [20]. Like the flavin site of complex I, ROS production from these enzymes also depends on the NADH/NAD+ ratio, with NAD+ being inhibitory. In general, we can conclude that the site of superoxide production in intact mitochondria during succinate oxidation is unlikely to be any of the dehydrogenases upstream of the electron transport chain, since we show in our system that the rate of ROS production is dissociated from the NADH/NAD+ ratio. In summary, we demonstrate that there is not a unique relationship between superoxide production rate from complex I and the NAD(P)H reduction state, suggesting that the flavin is not the main site of superoxide production during reverse electron transport. The iron–sulfur centres of complex I (except for centre N1a) are also in equilibrium with the NADH/ NAD+ ratio [21], thus these sites are not likely to be producing superoxide during succinate oxidation. The most likely site is, therefore, the quinone binding site. Acknowledgements: A.J.L. is supported by a Research into Ageing Fellowship, M.D.B. is supported by the Medical Research Council and the Wellcome Trust (066750/B/01/Z).
A.J. Lambert et al. / FEBS Letters 582 (2008) 1711–1714
References [1] Kussmaul, L. and Hirst, J. (2006) The mechanism of superoxide production by NADH:ubiquinone oxidoreductase (complex I) from bovine heart mitochondria. Proc. Natl. Acad. Sci. USA 103, 7607–7612. [2] McCord, J.M. and Fridovich, I. (1969) Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein). J. Biol. Chem. 244, 6049–6055. [3] Li, Y., Huang, T.T., Carlson, E.J., Melov, S., Ursell, P.C., Olson, J.L., Noble, L.J., Yoshimura, M.P., Berger, C., Chan, P.H., Wallace, D.C. and Epstein, C.J. (1995) Dilated cardiomyopathy and neonatal lethality in mutant mice lacking manganese superoxide dismutase. Nat. Genet. 11, 376–381. [4] Raha, S. and Robinson, B.H. (2000) Mitochondria, oxygen free radicals, disease and ageing. Trends Biochem. Sci. 25, 502–508. [5] Galkin, A. and Brandt, U. (2005) Superoxide radical formation by pure complex I (NADH:ubiquinone oxidoreductase) from Yarrowia lipolytica. J. Biol. Chem. 280, 30129–30135. [6] Genova, M.L., Ventura, B., Giuliano, G., Bovina, C., Formiggini, G., Parenti Castelli, G. and Lenaz, G. (2001) The site of production of superoxide radical in mitochondrial complex I is not a bound ubisemiquinone but presumably iron–sulfur cluster N2. FEBS Lett. 505, 364–368. [7] Kushnareva, Y., Murphy, A.N. and Andreyev, A. (2002) Complex I-mediated reactive oxygen species generation: modulation by cytochrome c and NAD(P)+ oxidation–reduction state. Biochem. J. 368, 545–553. [8] Lambert, A.J. and 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. [9] Lambert, A.J. and Brand, M.D. (2004) Inhibitors of the quinonebinding site allow rapid superoxide production from mitochondrial NADH:ubiquinone oxidoreductase (complex I). J. Biol. Chem. 279, 39414–39420. [10] Ohnishi, S.T., Ohnishi, T., Muranaka, S., Fujita, H., Kimura, H., Uemura, K., Yoshida, K. and Utsumi, K. (2005) A possible site of superoxide generation in the complex I segment of rat heart mitochondria. J. Bioenerg. Biomembr. 37, 1–15. [11] Brand, M.D., Affourtit, C., Esteves, T.C., Green, K., Lambert, A.J., Miwa, S., Pakay, J.L. and Parker, N. (2004) Mitochondrial superoxide: production, biological effects, and activation of uncoupling proteins. Free. Radic. Biol. Med. 37, 755–767. [12] Rolfe, D.F.S., Hulbert, A.J. and Brand, M.D. (1994) Characteristics of mitochondrial proton leak and control of oxidative phosphorylation in the major oxygen-consuming tissues of the rat. Biochim. Biophys. Acta 1188, 405–416. [13] Ernster, L. and Lee, C.-P. (1967) Energy-linked reduction of NAD+ by succinate. Meth. Enzymol. 10, 729–738. [14] Katz, L.A., Koretsky, A.P. and Balaban, R.S. (1987) Respiratory control in the glucose perfused heart. A 31P NMR and NADH fluorescence study. FEBS Lett. 221, 270–276. [15] Hansford, R.G., Hogue, B.A. and Mildaziene, V. (1997) Dependence of H2O2 formation by rat heart mitochondria on substrate availability and donor age. J. Bioenerg. Biomembr. 29, 89–95. [16] Liu, Y., Fiskum, G. and Schubert, D. (2002) Generation of reactive oxygen species by the mitochondrial electron transport chain. J. Neurochem. 80, 780–787. [17] Votyakova, T.V. and Reynolds, I.J. (2001) DWm-Dependent and independent production of reactive oxygen species by rat brain mitochondria. J. Neurochem. 79, 266–277. [18] Kudin, A.P., Bimpong-Buta, N.Y., Vielhaber, S., Elger, C.E. and Kunz, W.S. (2004) Characterization of superoxide-producing sites in isolated brain mitochondria. J. Biol. Chem. 279, 4127–4135. [19] Starkov, A.A. and Fiskum, G. (2003) Regulation of brain mitochondrial H2O2 production by membrane potential and NAD(P)H redox state. J. Neurochem. 86, 1101–1107. [20] Starkov, A.A., Fiskum, G., Chinopoulos, C., Lorenzo, B.J., Browne, S.E., Patel, M.S. and Beal, M.F. (2004) Mitochondrial alpha-ketoglutarate dehydrogenase complex generates reactive oxygen species. J. Neurosci. 24, 7779–7788. [21] Vinogradov, A.D. (2001) Respiratory complex I: structure, redox components, and possible mechanisms of energy transduction. Biochemistry (Moscow) 66, 1086–1097.
Dissociation of superoxide production by mitochondrial complex I from NAD(P)H redox state Adrian J. Lambert*, Julie A. Buckingham, Martin D. Brand MRC Dunn Human Nutrition Unit, Hills Road, Cambridge CB2 0XY, UK Received 26 March 2008; revised 14 April 2008; accepted 15 April 2008 Available online 28 April 2008 Edited by Barry Halliwell
Abstract The relationship between the rate of superoxide production by complex I and NAD(P)H redox state was investigated in rat skeletal muscle mitochondria. A high rate of superoxide production was observed during succinate oxidation; the rate during pyruvate oxidation was over fourfold lower. However, the NAD(P)H pool was significantly less reduced during succinate oxidation than during pyruvate oxidation. We conclude that there is no unique relationship between superoxide production by complex I and the reduction state of the NAD(P)H pool. Our data suggest that less than 10% of the superoxide originates from the flavin site during reverse electron transport from succinate. Ó 2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: NADH:ubiquinone oxidoreductase, Mitochondria; Superoxide; NAD(P)H; Hydrogen peroxide; Reactive oxygen species
1. Introduction NADH-ubiquinone oxidoreductase (complex I) is an important proton-pumping enzyme of the mitochondrial electron transport chain which can also convert molecular oxygen to the superoxide anion radical [1]. The discovery of superoxide dismutase (SOD), a naturally occurring enzyme that catalyses the dismutation of superoxide to hydrogen peroxide demonstrated the relevance of superoxide in vivo [2]. This relevance was confirmed by the demonstration that mice null for the mitochondrial form of SOD (MnSOD or SOD2) exhibit severe pathologies and have a markedly curtailed lifespan [3]. In addition, several pathologies are associated with defects in complex I and elevated production of superoxide [4]. It is, therefore, of great importance to understand the sites, mechanisms and regulation of superoxide production by complex I. Complex I contains a number of redox centres that are potentially capable of producing superoxide. However, as yet there is no consensus view on the site or sites of production of superoxide by complex I under physiological conditions (such as in vivo or in intact cells) or conditions that approach * Corresponding author. Fax: +44 (0) 1223 252805. E-mail address: [email protected] (A.J. Lambert).
Abbreviations: SOD, superoxide dismutase; FCCP, carbonylcyanide-ptrifluoromethoxyphenylhydrazone; ROS, reactive oxygen species
physiological (such as in energised, intact mitochondria). In the isolated enzyme, the fully reduced flavin, FMNH has been identified as the most likely electron donor to oxygen to produce superoxide [1]. Whether the reduced flavin remains the major site of production in a membrane-bound, protonpumping complex I is an unresolved, yet pressing question. There is additional evidence that the flavin produces superoxide [5], but other sites such as the iron–sulfur centres and the quinone site have been suggested also [6–10]. In intact mitochondria, the situation is complicated by the ability of complex I to undergo reverse electron transport, which results in very high rates of superoxide production compared to the rate observed during forward electron transport [11]. In this study, we investigated whether the flavin is the main site of superoxide production in intact mitochondria during reverse electron transport. We based our experiments on the observation that in isolated complex I, the rate of superoxide production from the flavin at fixed [NADH + NAD+] is dependent on the NADH/NAD+ ratio, with the reaction being strongly inhibited by NAD+ [1]. A high NADH/NAD+ ratio results in a highly reduced flavin (FMNH ) which leads to a high rate of superoxide production. Therefore, if superoxide originates exclusively from the flavin under all conditions, there should be a unique relationship between NADH redox state and superoxide production. Our data demonstrate that there is not a unique relationship: under conditions where superoxide production is high (during reverse electron transport), the NADH/NAD+ ratio is significantly lower than under conditions where superoxide production is low (during forward electron transport). By inference, superoxide is not originating from the flavin during reverse electron transport in our system.
2. Materials and methods Mitochondria (with an average respiratory control ratio of 2.0 ± 0.2 on succinate) from skeletal muscle of female Wistar rats (aged between 5 and 8 weeks) were isolated as described [12]. Superoxide production rate was assessed indirectly by measurement of hydrogen peroxide generation rate, determined fluorometrically by measurement of oxidation of amplex red to fluorescent resorufin. Mitochondria were incubated at 0.35 mg mitochondrial protein ml 1 in buffer containing 120 mM potassium chloride, 3 mM HEPES, 1 mM EGTA and 0.3% BSA (w/ v) (pH 7.2 and 37 °C). All incubations also contained 50 lM amplex red, 4 U ml 1 horseradish peroxidase, 30 U ml 1 SOD and 1 lg ml 1 oligomycin. The reaction was initiated by addition of respiratory substrates (either 5 mM succinate or 2.5 mM pyruvate + 2.5 mM malate) and the increase in fluorescence at an excitation of 563 nm and an emission of 587 nm was monitored. Standard curves were generated using known amounts of hydrogen peroxide, these standard curves
0014-5793/$34.00 Ó 2008 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2008.04.030
1712 were not affected significantly by substrates or inhibitors. We assessed superoxide production by more direct methods (aconitase inhibition and dihydroethidine fluorescence) and found qualitatively similar findings to the amplex red assay (data not shown). NAD(P)H fluorescence was determined at excitation and emission wavelengths of 365 and 450 nm, respectively [13]. Incubation conditions were identical to those used for hydrogen peroxide measurement. We measured changes in fluorescence in this study to track changes in NADH, but we refer to what we measured as NAD(P)H. This is because NADH and NADPH exhibit the same fluorescent spectra and like other studies that measured the NAD(P)H reduction state in mitochondria, we assumed that the changes in fluorescence we measured were reporting changes in NADH as opposed to changes in NADPH. As discussed previously, this is a reasonable assumption to make [14], but at this stage we cannot completely rule out the possibility that there are complex factors that influence the fluorescence of the NADH and NADPH pools that may affect interpretation of our results. Calculations of matrix pH and values of DpH were based on data from [8].
3. Results We first compared hydrogen peroxide production and NAD(P)H fluorescence during succinate oxidation to NADH-linked oxidation induced with pyruvate and malate
A.J. Lambert et al. / FEBS Letters 582 (2008) 1711–1714
in the presence of complex I inhibitor rotenone (Fig. 1). As previously reported by this laboratory and others [8,15–17], the rate of hydrogen peroxide production was substantially greater during succinate oxidation than during NADH-linked oxidation (Fig. 1A and C). It is generally understood that the high rate observed with succinate originates as superoxide production from complex I during reverse electron transport, since about 95% of the signal can be abolished by rotenone. Under identical conditions, the levels of NAD(P)H fluorescence were significantly lower during succinate oxidation than during NADH-linked oxidation (Fig. 1B and D). To test whether the results obtained with rotenone were specific to that inhibitor, we examined other conditions during NADH-linked oxidation. In the presence of complex I inhibitor piericidin, the rate of hydrogen peroxide production (Fig. 2, point ‘‘c’’) was significantly lower than the rate observed during succinate oxidation (Fig. 2, point ‘‘b’’). The rate of hydrogen peroxide production with the complex III centre ÕoÕ inhibitor stigmatellin (Fig. 2, point ‘‘d’’) was also significantly lower than the rate seen with succinate. The lowest rate of hydrogen peroxide production was seen in the absence of any inhibitor (Fig. 2, point ‘‘e’’). In all three cases, the levels of NAD(P)H fluorescence were higher than the levels observed
Fig. 1. Rates of hydrogen peroxide production and levels of NAD(P)H fluorescence in rat skeletal muscle mitochondria. Panels A and B show traces of resorufin and NAD(P)H fluorescence over time, respectively. Dashed lines, no additions; solid lines, succinate added at t = 35 s; dotted lines, pyruvate + malate + 2 lM rotenone added at t = 35 s. Rates of hydrogen peroxide production were calculated from slopes of the traces between 100 and 150 s in conjunction with a standard curve. The average readings between 100 and 150 s were used to calculate the level of NAD(P)H fluorescence. Fully oxidised NAD(P)H was determined in the presence of 2 lM carbonylcyanide-p-trifluoromethoxyphenylhydrazone (FCCP) (not shown). The experiments in A and B were repeated and the means ± S.E. are shown in panels C and D. *Significant (p < 0.05) difference vs. succinate (n = 4 separate mitochondrial preparations).
A.J. Lambert et al. / FEBS Letters 582 (2008) 1711–1714
Fig. 2. Rates of hydrogen peroxide production and levels of NAD(P)H fluorescence. Mitochondria were incubated as described in Materials and Methods and the rates of hydrogen peroxide production and NAD(P)H fluorescence levels were calculated as described in Fig. 1 legend. Point a, pyruvate + malate + 2 lM rotenone (NAD(P)H = 100%); point b, succinate; point c, pyruvate + malate + 1 lM piericidin; point d, pyruvate + malate + 80 nM stigmatellin; point e, pyruvate + malate. Each point is the mean ± S.E. of n = 4 separate mitochondrial preparations.
with succinate. Taken together with point ‘‘a’’ derived from Fig. 1, these results show no evidence for a single, unique relationship between superoxide production rate from complex I and the level of NAD(P)H fluorescence during both forward and reverse electron transport. However, we can assume that there is a unique relationship between superoxide production from the flavin and NAD(P)H reduction level during forward electron transport specifically [1]. By drawing a curve through zero and points a, c, d and e, we estimate (by dropping a vertical line from point ‘‘b’’ to that curve) that the rate of superoxide generation from the flavin during reverse electron transport at 90% reduction of NAD(P)H is about 0.2/2.3, or <10% of the total rate. During succinate oxidation, the electron transport chain pumps protons from the mitochondrial matrix to the intermembrane space resulting in alkalinisation of the matrix and generation of an electrochemical proton gradient. Under our
1713
conditions the chemical component of the gradient, DpH, is about 0.5 pH units [8]. Therefore, when the pH of the buffer is 7.2 the pH of the mitochondrial matrix is approximately 7.7. In the presence of NADH-linked substrates and inhibitors however, there is no proton pumping and the matrix pH is equal to the external pH which is determined by the buffering system. Therefore, we tested the effects of matrix pH on hydrogen peroxide production rate and NAD(P)H fluorescent levels to see if the differences between succinate and NADH-linked oxidation could be explained by differences in matrix pH. For experiments with succinate, buffers with pH 6.5, 7 and 7.5 were used to give matrix pH values of about 7, 7.5 and 8, respectively (based on data in [8]). For experiments with NADH-linked oxidation, buffers with pH of 7, 7.5 and 8 were used. The results are displayed in Fig. 3, with the data for pH 7.2 and 7.7 included. Panels A and B indicate that changes in matrix pH do not have any significant effects on hydrogen peroxide production rate or NAD(P)H fluorescence. Importantly, the differences in hydrogen peroxide production rate between succinate and NADH-linked oxidation were significant at all values of matrix pH. In terms of NAD(P)H fluorescence, the differences were significant at 2 out of the 3 values of matrix pH. Therefore, changes in matrix pH are not responsible for the observed differences in superoxide production rates and NAD(P)H levels between forward and reverse electron transport.
4. Discussion As yet, there is no clear demonstration of the site, or sites, of superoxide production by mitochondrial complex I under physiological conditions. There is strong evidence that in the isolated enzyme (which is distinct from a membrane-bound, proton-pumping enzyme subject to protonmotive force and other regulators), the site is the fully reduced flavin, FMNH , and that the rate depends on the NADH/NAD+ ratio [1]. At low concentrations, NADH and NAD+ are in equilibrium with the flavin site. At the higher concentrations of NADH + NAD+ in the mitochondrial matrix, inhibition by NAD+ may prevent equilibrium, but even so, at fixed total NAD+ +
Fig. 3. Effect of mitochondrial matrix pH on hydrogen peroxide production rate and NAD(P)H fluorescence levels. Mitochondria were incubated as described in Section 2, except that the pH of the buffer was varied between 6.5 and 8. Values of DpH were based on data in [8]. Closed symbols, succinate; open symbols, pyruvate + malate + 1 lM piericidin. Panel A shows matrix pH vs. hydrogen peroxide production rate, B shows matrix pH vs. NAD(P)H fluorescence level. *Significant (p < 0.05) difference vs. succinate at the same pH (n = 3 separate mitochondrial preparations).
1714
NADH, the NADH/NAD+ ratio will determine the superoxide production rate. If superoxide originates at the flavin under all conditions, then a single, unique relationship should exist between the NADH/NAD+ ratio and the superoxide production rate. Such relationships have been reported for complex I in intact mitochondria, but for forward electron transport only [18,19]; in these experiments the superoxide is most likely coming from the flavin. However we report here a condition that is clearly an exception: reverse electron transport where superoxide production rate is very high compared to forward electron transport, yet the NADH/NAD+ ratio is relatively low. Therefore, there is not a unique relationship between the NADH/ NAD+ ratio and superoxide production rate, and this implies that most superoxide does not originate at the flavin site during reverse electron transport. Our estimate from Fig. 2 is that less than 10% of it does. It is well established that the rate of superoxide production during succinate oxidation is much greater than the rate during oxidation of succinate in the presence of rotenone [11]. Therefore, the superoxide production rate from sites other than complex I (such as complex III) is low compared to the rate originating from complex I during reverse electron transport. We conclude that any small contribution from complex III (or any other sites) to the overall rate of superoxide production is negligible and does not alter the conclusion of our study. Our data are supported by previous work that measured NAD(P)H and superoxide production, although the aims of those studies were different from ours. Hansford et al. reported a rate of hydrogen peroxide production from heart mitochondria oxidising glutamate, malate and rotenone (where the NAD(P)H pool is fully reduced) of 0.22 ± 0.03 nmol/min/mg [15]. During succinate oxidation the rate was higher (0.63 ± 0.12 nmol/min/mg), yet the NAD(P)H pool was 22% less reduced. Qualitatively similar findings were demonstrated for heart mitochondria by Kushnareva et al. [7]. Therefore, our findings are not specific to our system, and are likely to apply to mitochondria from different tissues under a variety of different conditions (such as buffer composition). Other isolated mitochondrial enzymes, such as pyruvate dehydrogenase and a-ketoglutarate dehydrogenase, have been demonstrated to produce reactive oxygen species (ROS) [20]. Like the flavin site of complex I, ROS production from these enzymes also depends on the NADH/NAD+ ratio, with NAD+ being inhibitory. In general, we can conclude that the site of superoxide production in intact mitochondria during succinate oxidation is unlikely to be any of the dehydrogenases upstream of the electron transport chain, since we show in our system that the rate of ROS production is dissociated from the NADH/NAD+ ratio. In summary, we demonstrate that there is not a unique relationship between superoxide production rate from complex I and the NAD(P)H reduction state, suggesting that the flavin is not the main site of superoxide production during reverse electron transport. The iron–sulfur centres of complex I (except for centre N1a) are also in equilibrium with the NADH/ NAD+ ratio [21], thus these sites are not likely to be producing superoxide during succinate oxidation. The most likely site is, therefore, the quinone binding site. Acknowledgements: A.J.L. is supported by a Research into Ageing Fellowship, M.D.B. is supported by the Medical Research Council and the Wellcome Trust (066750/B/01/Z).
A.J. Lambert et al. / FEBS Letters 582 (2008) 1711–1714
References [1] Kussmaul, L. and Hirst, J. (2006) The mechanism of superoxide production by NADH:ubiquinone oxidoreductase (complex I) from bovine heart mitochondria. Proc. Natl. Acad. Sci. USA 103, 7607–7612. [2] McCord, J.M. and Fridovich, I. (1969) Superoxide dismutase. An enzymic function for erythrocuprein (hemocuprein). J. Biol. Chem. 244, 6049–6055. [3] Li, Y., Huang, T.T., Carlson, E.J., Melov, S., Ursell, P.C., Olson, J.L., Noble, L.J., Yoshimura, M.P., Berger, C., Chan, P.H., Wallace, D.C. and Epstein, C.J. (1995) Dilated cardiomyopathy and neonatal lethality in mutant mice lacking manganese superoxide dismutase. Nat. Genet. 11, 376–381. [4] Raha, S. and Robinson, B.H. (2000) Mitochondria, oxygen free radicals, disease and ageing. Trends Biochem. Sci. 25, 502–508. [5] Galkin, A. and Brandt, U. (2005) Superoxide radical formation by pure complex I (NADH:ubiquinone oxidoreductase) from Yarrowia lipolytica. J. Biol. Chem. 280, 30129–30135. [6] Genova, M.L., Ventura, B., Giuliano, G., Bovina, C., Formiggini, G., Parenti Castelli, G. and Lenaz, G. (2001) The site of production of superoxide radical in mitochondrial complex I is not a bound ubisemiquinone but presumably iron–sulfur cluster N2. FEBS Lett. 505, 364–368. [7] Kushnareva, Y., Murphy, A.N. and Andreyev, A. (2002) Complex I-mediated reactive oxygen species generation: modulation by cytochrome c and NAD(P)+ oxidation–reduction state. Biochem. J. 368, 545–553. [8] Lambert, A.J. and 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. [9] Lambert, A.J. and Brand, M.D. (2004) Inhibitors of the quinonebinding site allow rapid superoxide production from mitochondrial NADH:ubiquinone oxidoreductase (complex I). J. Biol. Chem. 279, 39414–39420. [10] Ohnishi, S.T., Ohnishi, T., Muranaka, S., Fujita, H., Kimura, H., Uemura, K., Yoshida, K. and Utsumi, K. (2005) A possible site of superoxide generation in the complex I segment of rat heart mitochondria. J. Bioenerg. Biomembr. 37, 1–15. [11] Brand, M.D., Affourtit, C., Esteves, T.C., Green, K., Lambert, A.J., Miwa, S., Pakay, J.L. and Parker, N. (2004) Mitochondrial superoxide: production, biological effects, and activation of uncoupling proteins. Free. Radic. Biol. Med. 37, 755–767. [12] Rolfe, D.F.S., Hulbert, A.J. and Brand, M.D. (1994) Characteristics of mitochondrial proton leak and control of oxidative phosphorylation in the major oxygen-consuming tissues of the rat. Biochim. Biophys. Acta 1188, 405–416. [13] Ernster, L. and Lee, C.-P. (1967) Energy-linked reduction of NAD+ by succinate. Meth. Enzymol. 10, 729–738. [14] Katz, L.A., Koretsky, A.P. and Balaban, R.S. (1987) Respiratory control in the glucose perfused heart. A 31P NMR and NADH fluorescence study. FEBS Lett. 221, 270–276. [15] Hansford, R.G., Hogue, B.A. and Mildaziene, V. (1997) Dependence of H2O2 formation by rat heart mitochondria on substrate availability and donor age. J. Bioenerg. Biomembr. 29, 89–95. [16] Liu, Y., Fiskum, G. and Schubert, D. (2002) Generation of reactive oxygen species by the mitochondrial electron transport chain. J. Neurochem. 80, 780–787. [17] Votyakova, T.V. and Reynolds, I.J. (2001) DWm-Dependent and independent production of reactive oxygen species by rat brain mitochondria. J. Neurochem. 79, 266–277. [18] Kudin, A.P., Bimpong-Buta, N.Y., Vielhaber, S., Elger, C.E. and Kunz, W.S. (2004) Characterization of superoxide-producing sites in isolated brain mitochondria. J. Biol. Chem. 279, 4127–4135. [19] Starkov, A.A. and Fiskum, G. (2003) Regulation of brain mitochondrial H2O2 production by membrane potential and NAD(P)H redox state. J. Neurochem. 86, 1101–1107. [20] Starkov, A.A., Fiskum, G., Chinopoulos, C., Lorenzo, B.J., Browne, S.E., Patel, M.S. and Beal, M.F. (2004) Mitochondrial alpha-ketoglutarate dehydrogenase complex generates reactive oxygen species. J. Neurosci. 24, 7779–7788. [21] Vinogradov, A.D. (2001) Respiratory complex I: structure, redox components, and possible mechanisms of energy transduction. Biochemistry (Moscow) 66, 1086–1097.