HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 281, Issue 52, 39766-39775, December 29, 2006

This Article Abstract Full Text (PDF) All Versions of this Article: 281/52/39766    most recent M608268200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by O'Malley, Y. Articles by Sivitz, W. I. Search for Related Content PubMed PubMed Citation Articles by O'Malley, Y. Articles by Sivitz, W. I. Social Bookmarking                      What's this?

Reactive Oxygen and Targeted Antioxidant Administration in Endothelial Cell Mitochondria^*

From the Department of Internal Medicine, Division of Endocrinology and Metabolism and the College of Pharmacy, Division of Medicinal and Natural Products Chemistry, Iowa City Veterans Affairs Medical Center and the University of Iowa, Iowa City, Iowa 52242

Received for publication, August 29, 2006

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
We used fluorescent probes and EPR to study the mechanism(s) underlying reactive oxygen species (ROS) production by endothelial cell mitochondria and the action of mitoquinol, a mitochondria-targeted antioxidant. ROS measured by fluorescence resulted from complex I superoxide released to the matrix and converted to H₂O₂. In contrast, EPR largely detected superoxide generated at complex III and effluxed outward. ROS fluorescence by mitochondria fueled by the complex II substrate, succinate, was substantial but markedly inhibited by rotenone. Superoxide, detected by EPR, in succinate-fueled mitochondria was not inhibited by rotenone and likely derived from semiquinone formation at complex III. Mitoquinol decreased H₂O₂ fluorescence by succinate-fueled mitochondria but had little effect on the EPR signal for superoxide. This was not associated with a detectable decrease in membrane potential. Mitoquinol markedly enhanced ROS fluorescence in mitochondria fueled by the complex I substrates, glutamate and malate. Inhibitor studies suggested that this occurred in complex I, at one or more Q binding pockets. The above effects of mitoquinol were determined in mitochondria isolated and subsequently exposed to the targeted antioxidant. However, similar effects were observed in mitochondria after antecedent exposure to mitoquinol/mitoquinone in culture, suggesting that the agent is retained after isolation of the organelles. In conclusion, ROS production in bovine aortic endothelial cell mitochondria results largely from reverse transport to complex I and through the Q cycle in complex III. Mitoquinol blocks ROS from reverse electron transport but increases superoxide production derived from forward transport. These effects likely occur at one or more Q binding sites in complex I.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Control of reactive oxygen species (ROS)² in endothelial cells is particularly important, because endothelial dysfunction, which may be triggered by ROS (1), is a major factor contributing to the development of atherosclerotic heart disease (2–4). Moreover, mitochondrial oxidative damage may underlie problems, including the vascular and neurologic complications of diabetes, cell damage in degenerative diseases, and aging (5–9). Reactive oxygen derived from endothelial cell mitochondria has also been implicated in the metabolic pathways leading to the microvascular complications of diabetes (6).

Mitochondrial electron transport generates substantial amounts of superoxide derived from electron leaks as substrates are metabolized (10). The process may have adverse consequences as evidenced by observations in mice lacking the mitochondrial enzyme manganese superoxide dismutase. These mice develop dilated cardiomyopathy and live less than 2 weeks (11). The major sites of superoxide production have been somewhat controversial, but there is evidence that most derive from complexes I and III (12). There is also evidence that complex I superoxide is released exclusively to the matrix side of the inner membrane, whereas complex III likely generates superoxide to both the matrix and intermembrane space (13–15).

The likely role of ROS in human disease has led to efforts at prevention through antioxidant therapy. However, the therapeutic use of antioxidants in vivo, particularly in human studies directed at vascular events, has been disappointing (16). Possible reasons for this include inadequate dosing due to concerns over toxicity and inability to deliver agents to target sites of ROS production. Hence, efforts are underway to develop effective antioxidant compounds targeted to mitochondria. One approach to this involves the synthesis of compounds linking agents such as redox forms of Q compounds (ubiquinol and ubiquinone) or vitamin E to compounds like triphenylphosphonium, a lipophilic cation avidly taken up into the relatively negative mitochondrial matrix (17, 18). In fact, we and several others have used the tetraphenylphosphonium ion to measure mitochondrial membrane potential (19, 20).

One such compound, an alkyltriphenylphosphonium cation, termed mitoQ (mitoquinol or mitoquinone or a mixture of these redox forms), consists of compound Q covalently bound to the cation triphenylphosphonium. By virtue of its delocalized positive charge, mitoQ accumulates several hundredfold in mitochondria (18, 21) and has been used to modulate ROS in the mitochondrial matrix (22).

Although mitoQ appears to have protective effects in certain cell types, the mechanism of action is not well defined and appears complex. A major action may be to decrease lipid peroxidation by virtue of the quinol moiety acting as a chain-breaking antioxidant (23). Mitochondria-targeted antioxidants protected Friedreich ataxia fibroblasts, in which glutathione synthesis was blocked, from oxidative stress (24), and mitoQ reduced telomere shortening in fibroblasts exposed to oxidative stress (25). In bovine aortic endothelial (BAE) cells, mitoQ reduced oxidative damage in cells stressed by 25 mM glucose and glucose oxidase (26). Moreover, mitoQ also reduced ROS and ERK2 activation in endothelial cells after hypoxic stress (27).

There is currently no information concerning the mitochondrial sites of ROS production or about the effects of mitochondria-targeted antioxidants in mitochondria of vascular cells. There is also no information concerning the effects of these targeted compounds in mitochondria isolated from cells subject to prior antioxidant treatment in culture.

To better understand the mechanism(s) and site(s) of ROS production by endothelial mitochondria we examined ROS production under different substrate and inhibitor conditions. We assessed ROS in two ways; as fluorescence produced by hydrogen peroxide and as superoxide specifically determined by EPR. As our data will show, these two methods differ in regards to what they measure. To learn more of the action of targeted quinols/quinones in endothelial mitochondria, we examined ROS production upon addition of these compounds directly to mitochondria as well as after antecedent exposure of BAE cells to these compounds in culture. Control compounds included non-targeted ubiquinone and decyltriphenylphosphonium (dTPP), which is identical to mitoquinone and mitoquinol but lacks the quinone/quinol group (28).

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Reagents and Supplies—Mitoquinone and mitoquinol were synthesized from commercially available 11-bromoundecanoic acid and 2,3-dimethoxy-5-methyl-1,4-benzoquinone as described previously (18, 21). Other reagents, kits, and supplies were as specified or purchased from standard sources.

Animals—Animals were fed and maintained according to standard National Institutes of Health guidelines, and the protocol was approved by our institutional Animal Care Committee. Normal C57Bl/6 mice (age 99 ± 19 days, 8 males and 2 females) were sacrificed by intraperitoneal injection of pentobarbital followed by cardiac puncture to obtain hind limb muscle tissue for preparation of mitochondria.

Cell Culture—BAE cells were grown in medium M199 (Invitrogen) supplemented with minimal essential medium amino acids (Invitrogen), penicillin/streptomycin (Invitrogen), minimal essential medium vitamins (Sigma), and 20% fetal bovine serum (HyClone, Logan, UT) as described (29). Cells were grown to near confluence in 150-cm² flasks between passages 6 and 12. For experiments involving mitochondria-targeted antioxidants administered to intact cells, 1 µM concentrations of mitoquinone, mitoquinol, ubiquinone, dTPP, or methyltriphenylphosphonium (mTPP), or vehicle (ethanol) were then added, and the cells incubated for 24 h before isolation of mitochondria.

Isolation of Mitochondria—Cells were washed with phosphate-buffered saline and scraped. Collected cells were homogenized using a Dounce homogenizer in ice-cold homogenization buffer (0.25 M sucrose, 5 mM HEPES, 0.1 mM EDTA, pH 7.2) with 0.1% fatty acid-free bovine serum albumin. For muscle mitochondria, the tissue was minced and processed using a Potter-Elvehjem tissue grinder and subsequently by a ground glass homogenizer. The homogenate was centrifuged at 1,000 x g for 10 min. The pellet was discarded, and the supernatant was centrifuged again at 10,000 x g for 10 min to obtain the mitochondrial pellet. The resulting pellet was then washed three times in homogenization buffer without bovine serum albumin and resuspended in media as described below. Protein was determined by the Bradford method (Bio-Rad). Mitochondria prepared in this way were of good quality as documented by an increase in respiratory rate of 4- to 6-fold after addition of carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone (FCCP) as an uncoupler.

Mitochondrial Hydrogen Peroxide—Mitochondria were studied during state 4 respiration, under which circumstance oxygen radical formation is enhanced as electron flow leads to high potential unmitigated by ATP generation (30). H₂O₂ production was assessed using the fluorescent probes Amplex Red (Invitrogen) and 2',7'-dichlorodihydrofluorescein diacetate (DCF, H₂DCF-DA, Invitrogen) using a Fluostar Optima fluorescence spectrophotometer (BMG Labtechnologies, Inc., Durham, NC) at 37 °C with filter wavelengths of 544 nm absorbance and 590 nm emission. Samples were prepared in 96-well plates containing 0.06 ml per well of respiratory buffer (220 mM mannitol, 70 mM sucrose, 2.5 mM KH₂PO₄, 2 mM MgCl₂, 1 mM EDTA, 2 mM HEPES, pH 7.4 with 0.1% fatty acid-free bovine serum albumin, 2 µm oligomycin) plus 20 µM Amplex Red or DCF, 5 units/ml horseradish peroxidase (Sigma) and mitochondria at a concentration of 0.05–0.1 mg/ml. When DCF was included, the compound was subject to chemical hydrolysis of the diacetate groups by exposure to 0.2 M KOH for 1 h before use. Fluorescence in all wells was measured in relative units once every 44 s (one cycle) and carried out for 50 cycles. For quantification, a hydrogen peroxide standard curve ranging from 0 to 12 µM was prepared in respiratory buffer and included on each plate. Addition of catalase, 500 units/ml, to incubations performed under the conditions studied reduced fluorescence to below the detectable limit indicating specificity for H₂O₂.

Mitochondria Superoxide Generation—As in the case of fluorescence detection, mitochondria were studied during state 4 respiration. EPR spectra were recorded using a Bruker EMX EPR spectrometer operating in X band and equipped with a high sensitivity resonator ER 4119HS. Samples were prepared at room temperature in 0.3 ml of respiratory buffer with 0.0688 M 5,5-dimethyl-l-pyrroline-N-oxide (DMPO), and 0.12 mg of isolated mitochondria. Respiration was initiated with the addition of substrate (succinate or glutamate plus malate), and samples were preincubated for 5 min in a 37 °C water bath. Reactions were transferred to a flat aqueous EPR cell, and the spectra were recorded at room temperature using the following instrument settings: microwave power 40 milliwatt, modulation amplitude 2 G, receiver gain 2 x 10⁵, conversion time 40.96 ms, time constant 81.92 ms, and scan rate 80G/41.92 s. Spectra shown are the average of 7–10 scans.

Oxygen Utilization and Mitochondrial Membrane Potential—The rate of respiration of isolated mitochondria (0.2–0.4 mg of protein/ml) was determined as we previously described (19, 31). Concurrently, we assessed mitochondrial inner membrane potential based on the distribution (inside and external to the mitochondrial matrix), of the lipophilic cation triphenylphosphonium (TPP⁺) as we previously described (19). To determine the concentration of TPP⁺ inside the mitochondrial matrix, it is necessary to know the mitochondrial matrix volume and the extent to which TPP⁺ is bound to mitochondrial protein as opposed to freely present within the matrix. Mitochondrial matrix volumes and the TPP+ binding correction, under the conditions of incubation, were determined as we have previously reported (20).

Kinetic Relationship of Proton Flux to Membrane Potential—The relationship of H⁺ flux to mitochondrial membrane potential (kinetics of the proton leak) has been considered an optimal assessment of mitochondrial uncoupling (32, 33). This was accomplished through simultaneous recording of oxygen consumption and potential as we have previously carried out using electrodes sensitive to oxygen and to the distribution of the tetraphenylphosphonium ion (TPP⁺) (19, 20). Mitochondria were incubated in respiratory buffer plus 5-µM rotenone to inhibit electron entry at Complex-I, and 0.1-µM nigericin to abolish the pH (34) across the mitochondrial membrane. H⁺ flux, under these conditions, is proton leak-dependent and, with succinate as fuel, follows a 6:1 stoichiometry (H⁺/O) with oxygen consumed (32, 35–37). Succinate (5 mM) was added followed by malonate in incremental amounts to final concentrations of 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 1.0, and 2.0 mM to inhibit succinate dehydrogenase, thereby decreasing electrons available for the transport system and creating a range of membrane potentials (38).

Measurement of Membrane Potential by Isotopic Assessment of TPP⁺ Distribution—Reaction tubes contained 40 µg of mitochondria, 2 µM oligomycin, 6 µM tetraphenylphosphonium chloride, and 40 nM tetra[³H]phenylphosphonium bromide (1.4 µCi/ml) in a volume of 100 µl of respiration medium plus 0.1% bovine serum albumin. The microcentrifuge tubes were incubated for 5 min at 37 °C in the presence of 5 mM succinate or 5 mM succinate plus 2.5 µM FCCP. The tubes were then centrifuged at room temperature for 4 min at 10,000 x g. 90 µl of each supernatant was removed for liquid scintillation counting. Succinate-driven mitochondrial TPP⁺ uptake was assessed as the difference in counts recovered between the succinate plus FCCP supernatant and the succinate alone supernatant. Potential was then calculated based on the Nernst equation. Each individual assay was carried out in duplicate, and mean values were determined.

Measurement of Complex I (NADH-Ubiquinone Oxidoreductase) Activity in BAE Cell Mitochondria—Activity was determined as described previously (39) by following the decrease in absorbance due to NADH oxidation at 340 nm. Activity was measured for 5 min in the presence of excess NADH and ubiquinone. Mitochondria were freeze-thawed three times to disrupt membranes and to release activity.

View larger version (36K): [in this window] [in a new window]   FIGURE 1. H2O2 production measured as relative DCF (A) or Amplex Red (B) fluorescence in isolated BAE mitochondria incubated in respiration buffer (see "Experimental Procedures") with the additions shown. Data represent mean (±S.E.) values of four wells within single mitochondrial preparations; *, p < 0.01 compared with succinate. C, H2O2 production by BAE mitochondria as affected by different substrates with or without rotenone as indicated. Here H2O2 was quantified based on Amplex Red fluorescence of a standard curve. Data represent mean values (±S.E.) for five separate mitochondrial preparations with each individual value determined as the mean of three or more replicate wells; *, p < 0.01 compared with all other groups. D, H2O2 production by mouse hind limb muscle mitochondria as affected by different substrates with or without or rotenone as indicated. Data represent mean values (±S.E.) for mitochondrial preparations from 10 separate mice with each individual value determined as the mean of two or more replicate wells; *, p < 0.01 compared with succinate; , p < 0.01 compared with glutamate plus malate. Additions consisted of 5 mM succinate (succ), 5 mM glutamate plus 1 mM malate (glut + mal), 5 µM rotenone (rot), and 1 µM antimycin A(Ant A).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
H₂O₂ Production from Isolated BAE Cell Mitochondria Assessed by Fluorescence—H₂O₂ was assessed as relative Amplex Red (Fig. 1A) or DCF (Fig. 1B) fluorescence in mitochondria isolated from BAE cells and fueled by succinate, a complex II substrate. Using either fluorescence probe, ROS production was markedly inhibited by the complex I inhibitor rotenone. Addition of the complex III inhibitor, antimycin A, mildly decreased ROS as assessed in this manner. H₂O₂ production in succinate-fueled mitochondria was compared with that in the presence on the complex I substrates, glutamate plus malate (Fig. 1C). In this case H₂O₂ was quantified in molar terms based on fluorescence of Amplex Red in the presence of a standard curve. H₂O₂ production was much less in the presence of the complex I substrates. Rotenone did not alter H₂O₂ production in this circumstance. The experiments in Fig. 1C were also carried out in mouse hind limb muscle mitochondria (Fig. 1D). Succinate-driven H₂O₂ production was greater in muscle mitochondria. In contrast, relatively little H₂O₂ was generated by glutamate/malate-fueled mitochondria unless rotenone was added.

View larger version (34K): [in this window] [in a new window]   FIGURE 2. EPR spectra (A–F) generated using the spin trap TMPO, which is specific for superoxide or the hydroxyl radical. Spectra were determined in the presence of mitochondria incubated in respiratory buffer plus the substrates and/or compounds indicated. Spectral signals were abolished by addition of manganese superoxide dismutase demonstrating specificity for superoxide. Note the scale difference for the succinate plus antimycin A condition. Additions consisted of 5 mM succinate, 5 µM rotenone, 1 µM antimycin A, or 200 µg/ml manganese SOD.

Superoxide Production from Isolated BAE Cell Mitochondria Assessed by EPR Spectroscopy—H₂O₂ production as measured by our fluorescence probes (see above) should derive largely from matrix superoxide released at complex I and converted to H₂O₂ by matrix superoxide dismutase (SOD) (13). Superoxide, effluxed outward to the cytoplasmic side, should not be detected by our fluorescence probes unless converted to H₂O₂ by exogenous SOD, which was not present in the incubation medium. On the other hand, superoxide effluxed directly outward should be detectable by EPR spectroscopy.

Superoxide was quantitatively assessed by EPR as the decay product of the superoxide-DMPO adduct, DMPO/·_OOH, which rapidly and spontaneously decays to the hydroxyl radical adduct DMPO/·_OH (40). Representative EPR spectra (Fig. 2) are those of DMPO/·_OH adducts as determined based on their characteristic splitting pattern (four lines of intensity 1:2:2:1) and hyperfine splitting constants a_N = 14.9 G and a_H = 14.9 G. These DMPO/·_OH spectra could possibly arise directly from the hydroxyl radical rather than the rapid decay of the superoxide adduct, DMPO/·_OOH. However, addition of SOD abolished the spectra (Fig. 2) indicating that formation of DMPO/·_OH was entirely mediated by the superoxide radical.

Fig. 2 reveals marked differences in reactive oxygen as assessed by EPR measurement of superoxide compared with fluorescence measurement of H₂O₂. As shown in Fig. 2, succinate-fueled mitochondria generated superoxide substantially above that in the absence of substrate, but, as opposed to fluorescent detection of H₂O₂ (Fig. 1), there was no inhibition by rotenone. Moreover, whereas antimycin A markedly increased the EPR signal (Fig. 2F), antimycin decreased H₂O₂ by fluorescence (Fig. 1). The spectra shown in Fig. 2 are representative of at least five repetitions for each condition studied (see below).

Effects of Mitochondria-targeted Antioxidant Compounds Added to Isolated BAE Mitochondria on H₂O₂ Production Detected by Fluorescence—ROS were assessed both as H₂O₂ fluorescence and as superoxide by EPR. Fig. 3A shows that mitoquinol and mitoquinone reduced H₂O₂ production by succinate-fueled mitochondria. The non-targeted compound, ubiquinone, had no effect. Fig. 3B shows that, in sharp contrast to the effect of mitoquinol in succinate-fueled mitochondria, mitoquinol markedly increased H₂O₂ production by mitochondria fueled by the complex I substrates, glutamate and malate. When added to mitochondria in the absence of substrate mitoquinol, mitoquinone, dTPP, and ubiquinone, all increased Amplex Red fluorescence, but by <10% of that observed in succinate-fueled mitochondria. Moreover, there was essentially no difference between the effects of the individual compounds, and therefore, nonspecific fluorescence cannot account for the observed differences between these compounds in the presence of substrate.

View larger version (32K): [in this window] [in a new window]   FIGURE 3. H2O2 production (mean ± S.E.) quantified by Amplex Red fluorescence in isolated BAE mitochondria incubated in respiration buffer and fueled with succinate (A) or glutamate plus malate (B) with the additions shown. mQL, mitoquinol (1 µM); mQN, mitoquinone (1 µM); CoQ, ubiquinone (1 µM); and dTPP, decyl-TPP. Otherwise abbreviations and concentrations are as in Figs. 1. *, p < 0.01 compared with succ (A) or compared with glut + mal (B). Numbers in parentheses represent the number of mitochondrial preparations with each individual value representing the mean of at least two replicate wells.

View larger version (26K): [in this window] [in a new window]   FIGURE 4. H2O2 production (mean ± S.E.) quantified by Amplex Red fluorescence relative to known standards in isolated BAE mitochondria incubated in respiration buffer and fueled with glutamate plus malate in the presence of 1 µM concentrations of the compounds indicated on the x axis. For each group along the x axis, mitochondria were exposed to vehicle (veh), 5 µM rotenone (rot), or 80 nM stigmatellin (stig). *, p < 0.01 compared with glut + mal in mitochondria treated with a Q compound or dTPP; , p < 0.01 compared with veh and compared with stig in mitochondria treated with mQL; , p < 0.05 compared with veh and compared with stig in mitochondria treated with mQN.

The effect of the complex I inhibitor, rotenone, to decrease H₂O₂ fluorescence by succinate-fueled mitochondria (Fig. 3) suggests that the H₂O₂ fluorescent signal derives from reverse electron transport (41, 42). Mitoquinol had a similar effect to reduce fluorescence by succinate-fueled mitochondria. So it could be that mitoquinol reduced H₂O₂ fluorescence by reducing potential, to which reverse electron transport is highly sensitive (5, 43). However, as shown in Fig. 5, this does not appear to be the case, because mitoquinol did not alter membrane potential in succinate-fueled mitochondria incubated under the conditions of Fig. 4. dTPP, unlike mitoquinol, did reduce potential in succinate-fueled mitochondria (Fig. 5). So, for dTPP, its effect on ROS in succinate-fueled mitochondria (Fig. 3) may reflect reduced potential. Notably, dTPP, unlike mitoquinol or mitoquinone, had no effect to increase H₂O₂ production by glutamate/malate-fueled mitochondria (Fig. 4).

Effects of Mitochondria-targeted Antioxidant Compounds Added to Isolated BAE Mitochondria on Superoxide Production by EPR Spectroscopy—Fig. 6 depicts superoxide production by EPR under several different conditions. Thirty-one mitochondrial preparations were examined in individual experiments (each on a separate day) with each experiment including the succinate alone condition as well as mitochondria exposed to the various conditions indicated. These data were normalized to the mean EPR signal intensity for the succinate alone condition. The results show that mitoquinol as well as ubiquinone and rotenone had no significant effects to alter superoxide production in succinate-fueled BAE mitochondria. Antimycin markedly enhanced superoxide by succinate-fueled mitochondria, whereas stigmatellin decreased superoxide. These effects of antimycin and stigmatellin were as expected and serve as controls validating the methodology and responsiveness of the mitochondria. In glutamate/malate-fueled mitochondria, as in succinate-fueled mitochondria, mitoquinol did not significantly alter superoxide by EPR. Rotenone decreased superoxide in glutamate/malate-fueled mitochondria as expected for the complex I substrates, wherein rotenone should limit electron transport through complex I and secondarily to complex III.

View larger version (37K): [in this window] [in a new window]   FIGURE 5. Inner membrane potential (mean ± S.E.) determined by the distribution of the lipophilic cation, tetra[3 H]phenylphosphonium bromide, as described under "Experimental Procedures." Mitochondria were incubated for 5 min in the presence of succinate (5 mM) in respiratory buffer with vehicle or 1 µM concentrations of mTPP, dTPP, or mitoquinol (mQL) as indicated on the x axis. *, p < 0.01 compared with vehicle and mTPP and <0.05 compared with mQL, n = 4 for each condition.

View larger version (33K): [in this window] [in a new window]   FIGURE 6. Superoxide production (mean ± S.E.) quantified as integrated EPR spectra normalized to the succinate condition. Representative examples of these spectra are shown in Fig. 2. Values represent mean ± S.E. Numbers above bars indicate repetitions, each from individual mitochondria preparations. Otherwise abbreviations and concentrations are as indicated in Figs. 1 and 3. **, p < 0.001; *, p < 0.01 compared with succinate; , p < 0.05 compared with glutamate (glut) and malate (mal).

As was the case for mitochondria exposed to mitoquinol after isolation, Fig. 7A suggests that succinate-driven H₂O₂ generation in mitochondria of BAE cells resulted from reverse electron transport (41, 42). As discussed above, reverse electron transport is very sensitive to mitochondrial membrane potential and to "mild" uncoupling (5, 43). So the effect of mitoquinol and mitoquinone in Fig. 7A could have been due to reduced potential and "mild" uncoupling. To examine this, and to address the possible effect of uncoupling (5, 43), we assessed the kinetics of respiration and potential in BAE mitochondria isolated from cells after 24-h treatment (of the cells) with either vehicle, dTPP, mitoquinone, or the methylated form of TPP (mTPP). As shown in Fig. 7C, mitoquinone had no effect on the kinetics of respiration and potential. Only dTPP reduced potential and shifted the curve of respiration versus potential up and to the left consistent with mitochondrial uncoupling (32, 33). Because we used a TPP⁺ electrode to detect mitochondrial potential, a question arose as to whether the targeted antioxidant compounds, which contain a TPP moiety, may have interfered with electrode potential. This would not seem very likely, because the mitochondria were isolated from cells previously treated with the targeted compounds and washed after isolation. Moreover, these results are consistent with mitochondrial inner membrane potential by distribution of the radiolabeled triphenylphosphonium ion as determined in mitochondria exposed to the targeted antioxidants after isolation (Fig. 5).

View larger version (37K): [in this window] [in a new window]   FIGURE 7. A, intact BAE cells were incubated for 24 h in the presence of 1 µM concentrations of the compound indicated (bar legends). Subsequently, mitochondria were isolated and incubated in the presence of respiratory buffer and fueled with succinate in the presence or absence of rotenone as indicated. H2O2 production (mean ± S.E.) was determined as Amplex Red fluorescence quantified relative to known standards. *, p < 0.01; p < 0.05 compared with vehicle, n = 3–8 mitochondrial preparations for each condition with individual data points representing the mean of two or more replicate wells. B, intact BAE cells were incubated for 24 h in the presence of 1 µM concentrations of the compound indicated. Subsequently, mitochondria were isolated and incubated in the presence of respiratory buffer and fueled with glutamate plus malate. *, p < 0.01 compared with vehicle, n = 4 mitochondrial preparations for each condition with individual data points representing the mean of two or more replicate wells. C, kinetic relationship between inner membrane potential and oxygen consumption as a measure of respiratory uncoupling in mitochondria isolated from BAE cells after 24 h prior incubation with the compound indicated. Data points represent mean ± S.E. for both oxygen use and potential. n = 4 for each condition.

View larger version (24K): [in this window] [in a new window]   FIGURE 8. Schematic diagram depicting electron transport, the action of substrates and inhibitors, proposed sites of ROS production, and proposed sites of action of mitoquinol/mitoquinone in BAE mitochondria. Straight lines with arrows represent forward electron transport, dashed lines represent reverse transport, and dashed-dotted lines represent diffusion of Q compounds in complex III or of H2O2 out from the matrix as shown. Dotted boxes represent complexes I, II, and III. X depicts sites of inhibition by rotenone (ROT), stigmatellin (STIG), or antimycin A (ANT A). Dark arrows show direction of superoxide release or conversion to H2O2, and ovals represent proposed sites of action of mitoquinol and mitoquinone.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Fig. 8 schematically depicts electron transport and the effects of the substrates and inhibitors used in our studies. Based on our current findings, the figure also depicts proposed sites of ROS production in BAE mitochondria and sites of action for mitoquinol and mitoquinone. The rationale for this is as follows.

Our results show that fluorescent H₂O₂ probes and EPR spectroscopy detected ROS in different fashion. This was evident in that the complex I inhibitor, rotenone, markedly decreased succinate-driven ROS production as detected by fluorescence but had no effect on superoxide production by EPR spectroscopy. Moreover, antimycin markedly increased succinate-driven superoxide by EPR but decreased ROS by fluorescence detection. These results are not discrepant. Amplex Red and DCF detect superoxide indirectly. When added to isolated mitochondria, these probes will detect H₂O₂ generated from superoxide by manganese superoxide dismutase, which is abundant in the mitochondrial matrix. H₂O₂ generated in this way will diffuse out of mitochondria and react with horseradish peroxidase in the incubation medium to trigger fluorescence. H₂O₂ produced in this way derives largely from matrix superoxide released at complex I (13). Superoxide effluxed outward and away from the matrix will not be detected by this fluorescence method in the absence of SOD. In contrast to our fluorescent probes, our EPR spin trap, DMPO, should directly detect superoxide effluxed outward. Superoxide produced in this fashion should largely derive from the Q cycle at complex III (13). The EPR signal so detected is highly specific for superoxide or the hydroxyl radical both of which react with DMPO to generate the same EPR spectra. However, the spectra reported here represent superoxide, because they could be completely abolished by added SOD. Because DMPO will not easily penetrate mitochondria and because matrix superoxide is rapidly converted to H₂O₂, our spin trap should not detect superoxide released to the matrix as occurs at complex I.

Thus, both the data in Figs. 1 and 2 and the theoretical considerations above imply that EPR spectroscopy and fluorescent probes can be used in complementary fashion, as we have done, to comprehensively examine endothelial ROS generation from mitochondrial complex I (fluorescence) and complex III (EPR). Theoretically we could have assessed complex III superoxide release to the cytoplasmic side of isolated mitochondria simply by measuring H₂O₂ production as fluorescence in the presence and absence of added SOD. However, we found marked nonspecific interference with fluorescence when SOD was added to isolated mitochondria. We do acknowledge that H₂O₂ production in the presence and absence of SOD has been effectively carried out in assessing the topology of muscle, heart, and liver mitochondria (13). However, that required mathematical correction for considerable fluorescence interference.

Our data (Fig. 1) show that BAE mitochondria consuming the complex II substrate, succinate, generate matrix superoxide released at complex I and detected as H₂O₂ fluorescence. This suggests that ROS so produced resulted from reverse electron transport, which is confirmed by the marked inhibition by the complex I inhibitor rotenone (Figs. 1 and 3). Similar findings have been reported for muscle, heart, and brain mitochondria (41, 44, 49).

In contrast to the amount of complex I ROS produced by succinate-fueled BAE mitochondria, consumption of the complex I substrates, glutamate and malate, generated considerably less (Figs. 1 and 2). However, this amount was clearly detectable, an observation that which differs from isolated muscle, heart, and brain mitochondria (41, 44, 49), which generate very little or no H₂O₂ when complex I substrates are oxidized. Complex I substrates in muscle, heart, and brain mitochondria generate more H₂O₂ when rotenone is added, possibly due to downstream inhibition of a Q-cycle type mechanism in complex I, similar to the known effect of antimycin at complex III (44). Our studies contrasted in that we did not see an increase in H₂O₂ production when rotenone was added. This was not due to technical differences, because, using the same methodology, we did see a stimulatory effect of rotenone in mitochondria isolated from normal mouse muscle tissue (Fig. 1D). Hence a complex I rotenone-sensitive Q-cycle mechanism may not be operative, or is operative in a rotenone-insensitive manner, in endothelial cell mitochondria.

Our EPR data (Figs. 2 and 6) show that BAE mitochondria fueled with either complex I or complex II substrates generate superoxide effluxed outward at complex III. This likely results from semiquinone formation in the complex III Q-cycle (Fig. 8). This is supported by the stimulatory effect of antimycin A, which inhibits downstream reduction and consequently prolongs the lifetime of the semiquinone radical and the inhibitory effect of stigmatellin, which blocks electron transport through complex III (Fig. 8).

Because mitochondrial ROS appear important in endothelial cell-mediated vascular disease, efforts are underway to develop effective means of antioxidant treatment. As discussed above, one means is to develop mitochondria-targeted antioxidants such as mitoQ. Our current findings provide new insight as to how these compounds modulate ROS in endothelial cell mitochondria. Based on our EPR results, there appears to be little effect of mitoquinol on superoxide generated at complex III (Fig. 6). This is compatible with a report indicating lack of restoration of respiration by MitoQ in ubiquinone-deficient yeast mitochondria due to the inability of mitoQ to be oxidized by complex III (28).

Although our EPR results showed that mitoquinol did not alter superoxide production at complex III, mitoquinol markedly reduced H₂O₂ generation in succinate-fueled mitochondria (Figs. 3 and 4). Thus, from the above discussion, mitoquinol appears to inhibit reverse electron transport to complex I. On the other hand, mitoquinol strongly enhances forward transport through complex I as manifest by the effect on H₂O₂ generation in the presence complex I substrates (Figs. 3 and 4).

How might mitoquinol and mitoquinone act to manifest the above effects on complex I superoxide formation? The answer to this is somewhat limited in that the physiology of electron transport at complex I is incompletely understood. However, our data do suggest certain mechanistic possibilities. Mitoquinol and mitoquinone and related compounds are felt to have antioxidant properties initiated by formation of a semiquinone as an initial step in scavenging an electron from peroxide species, thereby, initiating a chain-breaking effect on lipid peroxidation (23). Moreover, mitoquinol and mitoquinone are believed to be mobile but situated in the mitochondrial inner membranes with the nonpolar (TPP) moieties within lipid bilayers and the quinone/quinol portions extending inward (28). So, it is possible that the quinone/quinol portions react during forward electron transport at a Q binding site wherein conversion of mitoquinol to the reactive semiquinone radical would lead to enhancement of superoxide production with release to the matrix and conversion to H₂O₂. In this respect, mitoquinol is envisioned as an analogue of ubiquinol but, upon semiquinone formation, would be less susceptible to further reduction leading to a longer half-life and, hence, greater superoxide generation. The effect of stigmatellin is compatible with this. Stigmatellin, by blocking electron transport through complex III, without inhibition of complex I, should lead to reduction of all upstream electron carriers back to NADH. This would limit semiquinone formation by preventing conversion from mitoquinol and, for that matter, ubiquinol. The effect of the Q site, complex I inhibitor, rotenone, to inhibit mitoquinol/mitoquinone-induced ROS generation during forward transport is compatible with this and supports an effect at a Q-binding pocket. However, rotenone would not work in a way that blocks reduction of the semiquinone, as is the case for antimycin in the Q cycle of complex III, because that would increase rather than decrease ROS. Rather, rotenone must block the interaction of mitoquinol/mitoquinone with the Q binding site in some other fashion.

Thus, our data on the action of mitoQ and ROS during forward electron transport implies a pro-oxidant rather than antioxidant effect. However, it is important to note that this does not imply deleterious effects of mitoQ in terms of cellular and mitochondrial oxidative stress. This is because, as stated above, formation of a semiquinone by mitoQ would scavenge an electron from peroxide species impairing lipid peroxidation (23). This would provide benefit at the expense of superoxide molecules against which cellular defenses efficiently exist.

Although rotenone markedly reduced the effect of mitoquinol/mitoquinone on ROS generation by forward transport, this effect of rotenone was less than that of stigmatellin (Fig. 4). Thus, the effect of rotenone to block interaction at a Q binding pocket is either not as potent as the effect of stigmatellin or mitoquinol/mitoquinone may act at an additional site, likely upstream, because the rotenone site is felt to be in the distal position of the complex (Fig. 8).

Completely opposite to their effect in glutamate and malate-fueled mitochondria, mitoquinol (Figs. 3 and 7) and mitoquinone (Fig. 7) reduce ROS generation in succinate-fueled mitochondria. Unfortunately, we cannot easily explain the mechanism underlying this opposite action. We considered that the action of mitoQ on ROS during reverse electron transport might occur through reduction in mitochondrial inner membrane potential and "mild" uncoupling to which reverse electron transport is highly sensitive (43, 50). However, this was not evident either as uncoupling of respiration as assessed by electrochemical means (Fig. 7) or by distribution of a radiolabeled lipophilic cation (Fig. 5). We do acknowledge that we cannot completely rule out that mitoQ does reduce potential to an undetectable extent and that reverse transport happens to be exquisitely sensitive to this reduction. However, another possibility considers that mitoQ is driven to the reduced state by reverse electron flow and prevented form forming a semiquinone. It would then seem feasible that reduced mitoQ (mitoquinol) could act in some way, possibly like rotenone, at a Q site, to block reverse transport.

We did observe that the compound, dTPP, which represents mitoQ with the 10 carbon chain but no quinol/quinone moiety did reduce potential (Figs. 5 and 7) and did reduce succinate-driven ROS (Figs. 3 and 7). Notably dTPP, unlike mitoQ, had no effect to enhance ROS during forward electron transport (Figs. 4 and 7) demonstrating that the quinol/quinone moiety is essential for action on ROS during forward transport.

Our results also suggest that some cycling of mitoQ must occur within BAE mitochondria. This is because our studies revealed similar effects of mitoquinol compared with mitoquinone on ROS generation (Figs. 4 and 7A). So, these agents, in effect, appear to be "buffered" with respect to redox state by the respiratory chain through redox cycling between mitoquinol and mitoquinone.

As seen in Fig. 7, the effects of mitoquinone and mitoquinol are evident in mitochondria isolated after antecedent exposure of the intact cells in culture to the targeted antioxidants. In this regard, mitoquinone and mitoquinol added to the culture media had actions similar to those evident when added directly to isolated mitochondria (Figs. 3, 4, and 7). Because the mitochondria of the antecedent-treated cells were washed after isolation (and the cells were washed prior to isolation of the mitochondria), this suggests that the targeted compounds remained in mitochondria after isolation. Although, we have not ruled out concurrent effects on gene expression or other cellular processes, the similarity of the effects of mitoquinol and mitoquinone whether added to cells in culture or to isolated mitochondria suggests that the effects were due to persistence of the compounds in mitochondria.

In summary, our data suggest the following. Superoxide production at complex I in BAE mitochondria results from reverse electron transport and, to a lesser extent, by forward transport. Superoxide is also generated at the Q cycle in complex III. mitoQ acts in complex I to block ROS generated by reverse transport but markedly enhance superoxide production derived from forward electron transport. This likely results from action at a rotenone-sensitive Q binding site and, possibly, one or more other upstream sites. Moreover, our data show that mitoQ has persistent effects evident in mitochondria isolated from cells after antecedent exposure in culture suggesting that the targeted antioxidant is taken up and maintained in mitochondria for at least 24 h.

	FOOTNOTES

 
^* This work was supported by Veterans Affairs Medical Research Funds and National Institutes of Health Grant DK25295. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ To whom correspondence should be addressed: Dept. of Internal Medicine, Division of Endocrinology and Metabolism, The University of Iowa Hospitals and Clinics, 422GH, 200 Hawkins Drive, Iowa City, IA 52242. Tel.: 319-353-7812; Fax: 319-353-7850; E-mail: william-sivitz{at}uiowa.edu.

² The abbreviations used are: ROS, reactive oxygen species; BAE, bovine aortic endothelial; DCF, 2',7'-dichlorodihydrofluorescein diacetate; dTPP, decyltriphenylphosphonium; mTPP, methyltriphenylphosphonium; TPP, tetraphenylphosphonium; SOD, superoxide dismutase; FCCP, carbonyl cyanide p-(trifluoromethoxy)phenylhydrazone; DMPO, 5,5-dimethyl-l-pyrroline-Noxide; mitoQ, mitoquinol or mitoquinone or a mixture of these redox forms; ERK, extracellular signal-regulated kinase.

	ACKNOWLEDGMENTS

 
We thank Judy Herlein and Brian Dake for technical assistance.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 

1. Ogita, H., and Liao, J. K. (2004) Endothelium 11, 123–132[CrossRef][Medline] [Order article via Infotrieve]
2. Pitkanen, O. P., Nuutila, P., Raitakari, O. T., Ronnemaa, T., Koskinen, P. J., Iida, H., Lehtimaki, T. J., Laine, H. K., Takala, T., Viikari, J. S. A., and Knuuti, J. (1998) Diabetes 47, 248–254[Abstract]
3. Nitenberg, A., Valensi, P., Sachs, R., Dali, M., Aptecar, E., and Attali, J. R. (1993) Diabetes 42, 1017–1025[Abstract]
4. Yokoyama, I., Momomura, S. I., Ohtake, T., Yonekura, K., Nishikawa, J., Sasaki, Y., and Omata, M. (1997) J. Am. College Cardiol. 30, 1472–1477[Abstract]
5. Brand, M. D., Buckingham, J. A., Esteves, T. C., Green, K., Lambert, A. J., Miwa, S., Murphy, M. P., Pakay, J. L., Talbot, D. A., and Echtay, K. S. (2004) Biochem. Soc. Symp. 71, 203–213
6. Nishikawa, T., Edelstein, D., Du, X. L., Yamagishi, S., Matsumura, T., Kaneda, Y., Yorek, M. A., Beebe, D., Oates, P. J., Hammes, H. P., Giardino, I., and Brownlee, M. (2000) Nature 404, 787–790[CrossRef][Medline] [Order article via Infotrieve]
7. James, A. M., and Murphy, M. P. (2002) J. Biomed. Sci. 9, 475–487[Medline] [Order article via Infotrieve]
8. Green, K., Brand, M. D., and Murphy, M. P. (2004) Diabetes 53, S110–S118[Abstract/Free Full Text]
9. Beckman, K. B., and Ames, B. N. (1998) Physiol. Rev. 78, 547–581[Abstract/Free Full Text]
10. Skulachev, V. P. (1996) Q. Rev. Biophys. 29, 169–202[Medline] [Order article via Infotrieve]
11. Li, Y., Huang, T. T., Carlson, E. J., Melov, S., Ursell, P. C., Olson, J. L., Nobile, L. J., Yoshimura, M. P., Berger, C., Chan, P. H., Wallace D. C., and Epstein, C. J. (1995) Nat. Genet. 11, 376–381[CrossRef][Medline] [Order article via Infotrieve]
12. Raha, S., and Robinson, B. H. (2000) Trends Biochem. Sci. 25, 502–508[CrossRef][Medline] [Order article via Infotrieve]
13. St-Pierre, J., Buckingham, J. A., Roebuck, S. J., and Brand, M. D. (2002) J. Biol. Chem. 277, 44784–44790[Abstract/Free Full Text]
14. Miwa, S., St-Pierre, J., Partridge, L., and Brand, M. D. (2003) Free Radic. Biol. Med. 35, 938–948[CrossRef][Medline] [Order article via Infotrieve]
15. Han, D., Williams, E., and Cadenas, E. (2001) Biochem. J. 353, 411–416[CrossRef][Medline] [Order article via Infotrieve]
16. Yorek, M. A. (2003) Free Radic. Res. 37, 471–480[CrossRef][Medline] [Order article via Infotrieve]
17. Smith, R. A., Porteous, C. M., Coulter, C. V., and Murphy, M. P. (1999) Eur. J. Biochem. 263, 709–716[Medline] [Order article via Infotrieve]
18. Kelso, G. F., Porteous, C. M., Coulter, C. V., Hughes, G., Porteous, W. K., Ledgerwood, E. C., Smith, R. A., and Murphy, M. P. (2001) J. Biol. Chem. 276, 4588–4596[Abstract/Free Full Text]
19. Fink, B. D., Hong, Y. S., Mathahs, M. M., Scholz, T. D., Dillon, J. S., and Sivitz, W. I. (2002) J. Biol. Chem. 277, 3918–3925[Abstract/Free Full Text]
20. Fink, B. D., Reszka, K. J., Herlein, J. A., Mathahs, M. M., and Sivitz, W. I. (2005) Am. J. Physiol. 288, E71–E79
21. Asin-Cayuela, J., Manas, A. R., James, A. M., Smith, R. A., and Murphy, M. P. (2004) FEBS Lett. 571, 9–16[CrossRef][Medline] [Order article via Infotrieve]
22. Echtay, K. S., Murphy, M. P., Smith, R. A., Talbot, D. A., and Brand, M. D. (2002) J. Biol. Chem. 277, 47129–47135[Abstract/Free Full Text]
23. James, A. M., Smith, R. A., and Murphy, M. P. (2004) Arch. Biochem. Biophys. 423, 47–56[CrossRef][Medline] [Order article via Infotrieve]
24. Jauslin, M. L., Meier, T., Smith, R. A., and Murphy, M. P. (2003) FASEB J. 17, 1972–1974[Abstract/Free Full Text]
25. Saretzki, G., Murphy, M. P., and von Zglinicki, T. (2003) Aging Cell 2, 141–143[CrossRef][Medline] [Order article via Infotrieve]
26. Dhanasekaran, A., Kotamraju, S., Kalivendi, S. V., Matsunaga, T., Shang, T., Keszler, A., Joseph, J., and Kalyanaraman, B. (2004) J. Biol. Chem. 279, 37575–37587[Abstract/Free Full Text]
27. Schafer, M., Schafer, C., Ewald, N., Piper, H. M., and Noll, T. (2003) Circ. Res. 92, 1010–1015[Abstract/Free Full Text]
28. James, A. M., Cocheme, H. M., Smith, R. A., and Murphy, M. P. (2005) J. Biol. Chem. 280, 21295–21312[Abstract/Free Full Text]
29. Moser, D. R., Lowe, W. L., Jr., Dake, B. L., Booth, B. A., Boes, M., Clemmons, D. R., and Bar, R. S. (1992) Mol. Endocrinol. 6, 1805–1814[Abstract]
30. Boss, O., Hagen, T., and Lowell, B. B. (2000) Diabetes 49, 143–156[Abstract]
31. Hong, Y., Fink, B. D., Dillon, J. S., and Sivitz, W. I. (2001) Endocrinology 142, 249–256[Abstract/Free Full Text]
32. Porter, R. K., Joyce, O. J., Farmer, M. K., Heneghan, R., Tipton, K. F., Andrews, J. F., McBennett, S. M., Lund, M. D., Jensen, C. H., and Melia, H. P. (1999) Int. J. Obes. Relat. Metab. Disord. 23, Suppl. 6, S12–S18
33. Brand, M. D., Brindle, K. M., Buckingham, J. A., Harper, J. A., Rolfe, D. F., and Stuart, J. A. (1999) Int. J. Obes. Relat. Metab. Disord. 23, S4–S11
34. Lionetti, L., Iossa, S., Liverini, G., and Brand, M. D. (1998) Arch. Biochem. Biophys. 352, 240–246[CrossRef][Medline] [Order article via Infotrieve]
35. Nicholls, D. G. (1997) Biosci. Rep. 17, 251–257[CrossRef][Medline] [Order article via Infotrieve]
36. Chavin, K. D., Yang, S., Lin, H. Z., Chatham, J., Chacko, V. P., Hoek, J. B., Walajtys-Rode, E., Rashid, A., Chen, C. H., Huang, C. C., Wu, T. C., Lane, M. D., and Diehl, A. M. (1999) J. Biol. Chem. 274, 5692–5700[Abstract/Free Full Text]
37. Brown, G. C., and Brand, M. D. (1988) Biochem. J. 252, 473–479[Medline] [Order article via Infotrieve]
38. Brand, M. D. (1990) Biochim. Biophys. Acta 1018, 128–133[Medline] [Order article via Infotrieve]
39. Birch-Machin, M. A., Briggs, H. L., Saborido, A. A., Bindoff, L. A., and Turnbull, D. M. (1994) Biochem. Med. Metab. Biol. 51, 35–42[CrossRef][Medline] [Order article via Infotrieve]
40. Zwicker, K., Dikalov, S., Matuschka, S., Mainka, L., Hofmann, M., Khramtsov, V., and Zimmer, G. (1998) Arzneimittelforschung 48, 629–636[Medline] [Order article via Infotrieve]
41. Liu, Y., Fiskum, G., and Schubert, D. (2002) J. Neurochem. 80, 780–787[CrossRef][Medline] [Order article via Infotrieve]
42. Votyakova, T. V., and Reynolds, I. J. (2001) J. Neurochem. 79, 266–277[CrossRef][Medline] [Order article via Infotrieve]
43. Skulachev, V. P. (1998) Biochim. Biophys. Acta 1363, 100–124[Medline] [Order article via Infotrieve]
44. Lambert, A. J., and Brand, M. D. (2004) J. Biol. Chem. 279, 39414–39420[Abstract/Free Full Text]
45. Brownlee, M. (2001) Nature 414, 813–820[CrossRef][Medline] [Order article via Infotrieve]
46. Murphy, M. P. (1997) Trends Biotechnol. 15, 326–330[CrossRef][Medline] [Order article via Infotrieve]
47. Murphy, M. P. (2004) Biochem. Soc. Transact. 32, 1011–1014[CrossRef][Medline] [Order article via Infotrieve]
48. Talbot, D. A., Lambert, A. J., and Brand, M. D. (2004) FEBS Lett. 556, 111–115[CrossRef][Medline] [Order article via Infotrieve]
49. Ohnishi, S. T., Ohnishi, T., Muranaka, S., Fujita, H., Kimura, H., Uemura, K., Yoshida, K., and Utsumi, K. (2005) J. Bioenerg. Biomembr. 37, 1–15[CrossRef][Medline] [Order article via Infotrieve]
50. Miwa, S., and Brand, M. D. (2003) Biochem. Soc. Transact. 31, 1300–1301[Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This arti