| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Cardiovascular Research Laboratory, Departments of Medicine (Cardiology) and Physiology, David Geffen School of Medicine at University of California, Los Angeles, Los Angeles, California
Submitted 7 December 2005 ; accepted in final form 29 March 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
) and matrix volume from ADP-induced dissipation and shrinkage, 2) ADP-dependent increase in O2 consumption, and 3) enhanced rate of ATP synthesis. Removal of extramitochondrial H2O2 by catalase was required to stimulate endogenous substrate oxidation, as shown by the increase in O2 consumption and . This effect was greatly enhanced by addition of Nac or mercaptopropionyl glycine to suppress oxidation-induced ROS increases in the matrix. Theoretical considerations, as well as reversible inhibition of O2 consumption with 3-mercaptopropionic acid and pyruvate in state 3, indicate that these substrates are fatty acids. Under in vivo conditions in which powerful antioxidant conditions are maintained, this mechanism may be important in stimulation of 
-oxidation and ATP production at low levels of extramitochondrial fatty acids. Incapacitation of this mechanism may potentially contribute to mitochondrial dysfunction during oxidative stress.
isolated mitochondria; redox environment; endogenous substrates; adenosine diphosphate phosphorylation
), has been obtained from isolated mitochondria. In vitro studies permit more rigorous control of the environment and allow function to be followed with greater accuracy than in mitochondria in vivo. However, limitations in accurate replication of the in vivo extramitochondrial environment complicate extrapolation to the in vivo situation. These difficulties include quantitative, as well as qualitative, indexes. In cardiac myocytes, for example, mitochondrial volume is about four times cytoplasmic volume, whereas isolated mitochondria in a cuvette are suspended in an extramitochondrial volume at least three orders of magnitude larger. As a result, in vitro mitochondria are typically exposed to a vast reservoir of substrates and various regulatory factors that may not be characteristic of in vivo conditions. The relatively small cytoplasmic pool of substrates immediately accessible to mitochondria, coupled with more restricted diffusion and inhibited transport under certain conditions in vivo, may lead to situations in which substrate availability becomes transiently limited and ATP production must rely on more complete -oxidation of fatty acids already available in the matrix. Qualitatively, in vitro mitochondria are generally studied in an oxidizing environment, whereas in vivo mitochondria are generally well protected by powerful antioxidant systems.
Mimicking these conditions by limiting substrate availability and protecting the redox environment allowed us to demonstrate a surprisingly large capacity of in vitro heart mitochondria to enhance electron transport and support during ADP phosphorylation in the absence of added substrates. These results also suggest that, under oxidizing conditions, ROS produced by fatty acid oxidation do not allow complete oxidation of fatty acids in the matrix and consumption for ATP synthesis. Collectively, our findings show that a reduced redox environment in the cardiac cell could be an important condition for completion of fatty acid oxidation and, at the same time, for efficient consumption of the proton gradient for ATP synthesis, possibly due to inhibition of ROS-induced proton leak.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Mitochondria were isolated from rabbit hearts by homogenization and differential centrifugation as described previously (11, 29). The final pellet was resuspended in EGTA-free buffer to yield 20–30 mg/ml of mitochondrial protein by Lowry protein assay. Freshly isolated mitochondria were characterized by coupling ratios >8 measured after they were energized with 1.5 mM pyruvate, malate, and glutamate in the presence of 5 mM Pi and 0.4 mM ADP. Mitochondria were kept on ice and used within 5 h after isolation.
Experimental Conditions
All measurements were carried out using a fiber-optic spectrofluorometer (Ocean Optics) in a closed continuously stirred cuvette at room temperature (22–24°C). Mitochondria (0.4–0.6 mg/ml) and then 2.5 mM Pi, ADP, catalase, and Nac were added to the incubation buffer (100 mM KCl, 10 mM HEPES, pH 7.4, with Tris) containing 0.5 mg/ml BSA (<0.005% free fatty acids). In some experiments, 0.5 mM caproic acid, pyruvate, and/or citric acid cycle intermediates (5 mM each) were added. To determine the ability of mitochondria to sustain ADP phosphorylation, we added ADP in the presence or absence of exogenous substrates.
Mitochondrial O2 consumption.
For continuous measurement of mitochondrial O2 consumption, the decrease in buffer O2 content was monitored via a fiber-optic O2 sensor inserted through a hole in the cuvette cover. This method allows simultaneous determination of O2 consumption, , matrix volume changes, and Ca2+ uptake. The tip of the O2 sensor fiber was positioned in the center of the cuvette, where it reacted to changes in PO2. In most experiments, except those with N2 for calibration, the surface of the buffer was exposed to room air. This may have decreased O2 consumption rates to some extent, but because the stirring speed was constant, this small decrease was always a constant value. Also, under these conditions, addition of ADP accelerated O2 consumption significantly, indicating that the small O2 flux into the buffer had no major impact. Zero PO2 was obtained by direction of a stream of N2 through the hole in the cuvette cover at the buffer (2 ml), so that the stirred buffer had no contact with the air (11).
Mitochondrial .
Tetramethylrhodamine methyl ester (TMRM, 200 nM) was included in the cuvette solution, and mitochondrial was estimated from TMRM fluorescence at 580 nm as described previously (11, 22). When used at low concentrations, TMRM does not suppress respiration (22). Mitochondrial is expressed as percentage of the TMRM fluorescence in the presence of coupled mitochondria and substrates (100%) relative to that after addition of 0.5 µM cyanide p-trifluoromethoxyphenylhydrazone or alamethicine (5 µg/ml) to fully depolarize mitochondria (0%).
Mitochondrial swelling. Changes in matrix volume were estimated according to a standard procedure by measurement of 90° light scattering with excitation and emission wavelengths set at 520 nm. Changes in matrix volume were compared with maximum (100%) swelling induced by addition of alamethicine at the end of the experiment.
Mitochondrial ATP production.
Mitochondrial ATP production was evaluated by continuous monitoring of NADPH fluorescence in a coupled hexokinase-glucose-6-phosphate dehydrogenase assay as described elsewhere (2). Basic conditions for this assay were identical to those used to evaluate recovery of and matrix volume from ADP-induced dissipation and shrinkage, except enzymes/chemicals for ATP determination were present.
ROS production. ROS production was measured by the increase in fluorescence resulting from oxidation of 2',7'-dichlorodihydrofluorescein (DCFH) to fluorescent dichlorofluorescein (DCF). Membrane-permeant DCFH diacetate or a chloromethyl derivate of DCFH diacetate (10 µM) was incubated with mitochondria for 20 min at room temperature and then for 1 h on ice. DCFH-loaded mitochondria were pelleted by brief centrifugation, washed, and used for evaluation of ROS production as determined by an increased rate of DCF fluorescence. The precise ROS that directly mediates DCFH oxidation is difficult to define. Our experiments showing that DCFH oxidation in mitochondria requires catalytic iron support the role of hydroxyl radicals in DCFH oxidation, as reported by several groups (27 and references therein). However, hydroxyl radical generation in the Fenton reaction requires superoxide and subsequent H2O2 production; therefore, an increase in DCF fluorescence in mitochondria should also reflect an increase in production of these species.
Mitochondrial Ca2+ uptake. Mitochondrial Ca2+ uptake was measured with a Ca2+-selective minielectrode (World Precision Instruments) in conjunction with a reference electrode.
Chemicals and Data Analysis
TMRM and DCFH diacetate or its chloromethyl derivate were obtained from Molecular Probes, 3-mercaptopropionic acid from Fluka, manganase(III)tetrakis(1-methyl-4-pyridyl)porphyrin pentachloride (MnTMPyP) from A. G. Scientific, and all other chemicals from Sigma. Catalase, Nac, mercaptopropionyl glycine (MPG), and MnTMPyP stock solutions were prepared in incubation buffer before each experiment. Nac and MPG solutions were adjusted to pH 7.4 with Tris. Results are presented as original traces, with summary data shown as means ± SD. Student's t-test was used to assess statistical significance, with Bonferroni's correction for more than two groups.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Freshly isolated mitochondria were added to KCl buffer containing BSA to reduce membrane leakiness (proton leak), and 2.5 mM Pi was added, as required for ADP phosphorylation (Fig. 1A). Under these conditions, mitochondria respiring on endogenous substrates were unable to support during repeated ADP additions, a condition used throughout this study to test the readiness of mitochondria for sustained ADP phosphorylation. Full recovery of after ADP-induced dissipation signals that >99% of ADP has been phosphorylated; i.e., the ATP-to-ADP ratio is expected to be 
100 under such conditions (15). recovery also allows matrix volume, which shrinks in proportion to dissipation, to recover due to K+/P
uptake accompanied by osmotically driven water influx (9). After several additions of 10 µM ADP, however, remained dissipated, and the matrix contracted. Addition of 0.5 µM catalase had no effect, but subsequent addition of 5 mM Nac, a cell-permeant antioxidant (1, 18, 25) that can enter the matrix, resulted in a significant increase in , in parallel with acceleration of O2 consumption and recovery of matrix volume. Subsequently, these mitochondria were able to recover from dissipation and matrix shrinkage induced by repeated 10- to 20-fold higher ADP additions. At higher ADP loads, the recovery time from dissipation and matrix volume became prolonged, with less increase in O2 consumption. Rapid recovery of after oligomycin indicated that the inability to recover was related to consumption of the protonmotive force during ADP phosphorylation (Fig. 1A).
|
and matrix volume after ADP-induced dissipation than in Fig. 1A). These data indicate that the contaminant Ca2+ significantly decreased, rather than increased, mitochondrial ability to support during ADP phosphorylation. The ability of mitochondria to support ADP phosphorylation with only endogenous substrates in the presence of catalase and Nac was sensitive to Nac concentration. Figure 1C shows that, in EGTA buffer, addition of 50 µM Nac after catalase promoted less rapid mitochondrial recovery in response to ADP than addition of 2.5 mM Nac.
We speculated that the endogenous mitochondrial substrates supporting ADP phosphorylation under these conditions are fatty acids. In this case, addition of pyruvate, which is known to suppress fatty acid oxidation and vice versa (17), might be expected to inhibit the effects of catalase and Nac, as shown in Fig. 1C. After the ability of mitochondria to recover and matrix volume had been restored by catalase and Nac, addition of pyruvate inhibited the subsequent mitochondrial response to a second 150 µM ADP challenge. After addition of pyruvate, there was no increase in O2 consumption after addition of ADP, remained dissipated, and the matrix contracted. At this point, addition of the citric acid cycle intermediate malate (5 mM) to boost respiratory power led to full recovery. Indeed, when mitochondria were energized by citric acid cycle intermediates such as malate, fumarate, 
-ketoglutarate, and succinate in addition to endogenous substrates, recovery of and matrix volume after ADP additions was robust and even more enhanced by the presence of pyruvate.
The effects of catalase and Nac on the ability of mitochondria to recover from ADP loads in seven different mitochondrial preparations are summarized in Fig. 1D. The cumulative ADP load that mitochondria could phosphorylate using endogenous substrates was defined by their ability to recover from ADP-induced dissipation after successive ADP additions. Addition of 0.5 µM catalase and 2.5 mM Nac increased cumulative ADP load dramatically from 25 ± 8 to 1,461 ± 399 nmol/mg protein (P < 0.001). Attempts to demonstrate this effect in liver mitochondria under similar conditions, however, were unsuccessful (3 preparations, data not shown).
Catalase and Nac were both required to support during ADP phosphorylation (Fig. 2), but the order of addition was unimportant. In Fig. 2A, addition of 2.5 mM Nac alone had no effect, but subsequent addition of catalase promoted recovery, matrix volume recovery, and increased O2 consumption, allowing mitochondria to tolerate further ADP additions, similar to Fig. 1A. Once again, subsequent addition of pyruvate inhibited the ability of mitochondria to recover from ADP, unless a citric acid cycle intermediate, such as -ketoglutarate, was also added. The inhibitory effect of pyruvate was reversed with all citric acid cycle intermediates tested: malate and succinate were the most potent, followed by -ketoglutarate and fumarate.
|
and matrix volume or to increase O2 consumption in response to the same amount of ADP (200 µM) until malate was also added. In contrast, in the absence of caproic acid, pyruvate oxidation supported recovery reasonably well when relatively small concentrations (<100 µM) of ADP were added, but not at higher concentrations (Fig. 2C). The latter failure depended on the amount of mitochondria added and was rapidly corrected with malate (Fig. 2C), which accounts for why isolated mitochondria have almost always been energized with a combination of pyruvate and a citric acid cycle intermediate (most frequently malate).
Figure 2D confirms directly that Nac and catalase enhanced the rate of ATP synthesis. Mitochondria responded to the first addition of 10 µM ADP with a rapid increase in ATP synthesis, consistent with their ability to recover after the first ADP addition (Figs. 1A and 2A). However, ATP synthesis after subsequent ADP additions was depressed, corresponding to the incomplete recovery of and matrix volume. Addition of 2.5 mM Nac + 0.5 µM catalase then rapidly increased ATP synthesis (Fig. 2D). This increase (4-fold) in ATP synthesis (Fig. 2D, inset) was consistent with an effect of Nac + catalase on ADP-induced changes in , matrix volume, and O2 consumption presented above. After stimulation by Nac and catalase, increased ATP production was inhibited by pyruvate and then reversed by addition of malate (Fig. 2D).
Inhibition of Fatty Acid -Oxidation by 3-Mercaptopropionoic Acid Suppresses Effects of Nac and Catalase on State 3 Respiration
To further establish that the endogenous substrate utilization stimulated by Nac and catalase was endogenous fatty acid, we studied the effects of 3-mercaptopropionic acid. 3-Mercaptopropionic acid is known to inhibit -oxidation in rat heart mitochondria in state 3, because its metabolites reversibly inhibit acyl-CoA dehydrogenase, but it has no significant effect on the citric acid cycle or oxidative phosphorylation (21). In mitochondria incubated in the presence of BSA and EGTA, the ability of catalase and Nac to enhance recovery and O2 consumption from ADP loads (Fig. 3A) was significantly reduced when the same preparation was incubated with 200 µM 3-mercaptopropionic acid (Fig. 3B). The summary data in Fig. 3C show that the average O2 consumption after addition of 150 µM ADP in the presence of 3-mercaptopropionic acid was only <10% of control. Under those conditions, respiratory power was insufficient to restore , which remained dissipated (Fig. 3B). After addition of pyruvate, malate, and glutamate, rapidly recovered, with a parallel increase in O2 consumption. The response of energized mitochondria to addition of ADP was similar to that of controls (Fig. 3B), confirming previous findings that inhibitor has no significant effect on the citric acid cycle, electron transport, or ADP phosphorylation (21).
|
In the experimental settings described above (Figs. 1 and 2), catalase and Nac were required to restore after dissipation by repeated ADP additions. The relative role of these two antioxidants in the overall mobilization of endogenous substrates, however, remains unclear. To investigate this issue further, the effects of catalase and/or Nac on , O2 consumption, and matrix Ca2+ uptake were recorded under the conditions described in Figs. 1 and 2, except Pi was omitted, because Pi transiently is known to increase and contaminant Ca2+ uptake independently of substrates. Under these conditions, catalase still modestly increased (Fig. 4A) to an extent similar to that observed when Nac was also present (Fig. 4B). Although these changes were relatively small, in the absence of catalase (not shown) or in the presence of Nac alone (Fig. 4C), slowly decreased during prolonged incubation. Thus the effect of catalase on can be explained by a modest activation of electron transport, as demonstrated by increased O2 consumption after catalase addition, due to increased oxidation of endogenous substrates sufficient to support during accumulation of contaminant Ca2+. However, catalase-stimulated respiration failed to support during additional Ca2+ pulses (Fig. 4A), unless Nac was also present (Fig. 4B). Under the latter conditions, catalase resulted in a much greater increase in O2 consumption, which accounts for the ability of mitochondria to restore during subsequent Ca2+ pulses. Nac alone had no major effect on mitochondrial O2 consumption or the ability to support during Ca2+ accumulation (Fig. 4C). dissipation increased further with addition of 3 µM Ca2+, which apparently promoted some release of already accumulated Ca2+. However, subsequent addition of catalase resulted in activation of electron transport, recovery of , and enhancement of Ca2+ uptake. Collectively these results showed that catalase is responsible for activating electron transport and that its effect is significantly enhanced by membrane-permeable Nac, which by itself is not able to enhance electron transport and support .
|
We next explored the mechanisms by which catalase and Nac allowed increased -oxidation of endogenous fatty acids to support and ADP phosphorylation. Addition of membrane-impermeant catalase to isolated mitochondria is expected to scavenge H2O2 outside mitochondria, whereas membrane-permeant Nac is expected to prevent oxidation of important sulfhydryl (SH) groups in the matrix. In isolated mitochondria, the matrix of which was loaded with the fluorescent ROS sensor DCFH, catalase increased DCFH oxidation, indicating increased ROS production in the matrix. We determined ROS production by mitochondria with an ROS sensor located inside the mitochondrial matrix. A sensor outside the matrix would be expected to indicate reduced ROS from H2O2 depletion by catalase. Matrix DCFH oxidation increased further after addition of caproic acid (Fig. 5A). When the order was reversed, caproic acid increased DCFH oxidation, which was further potentiated by catalase (Fig. 5B). If it is assumed that DCFH oxidation accurately reflects ROS production, these results indicate that catalase and caproic acid increased matrix ROS production and that their effects were additive. A possible interpretation of these findings is that removal of H2O2 by catalase stimulates H2O2 efflux, decreasing matrix H2O2 and relieving its inhibition of electron transport chain/fatty acid oxidation. With increased electron transport, however, ROS production in the matrix increased. However, we cannot exclude direct effects of the probe itself on respiratory chain activity and enzymes in the matrix connected with ROS generation. Interestingly, the effect of catalase and caproic acid on matrix DCFH oxidation rate was significantly less when pyruvate and citric acid cycle intermediates were present (Fig. 5C).
|
recovery in the presence of ADP, unless the effects of ROS are also neutralized (e.g., by Nac). In contrast, when ample exogenous substrates are present, is more firmly supported and less sensitive to ROS accumulation.
To confirm the importance of ROS, we also tested the membrane-permeant ROS scavenger MPG. MPG (5 mM) was as effective as Nac in suppressing increased matrix DCFH oxidation induced by catalase or caproic acid (Fig. 5E) but less effective than Nac in restoring and matrix volume after an ADP load (Fig. 6). Once mitochondria failed to respond to ADP additions, Nac further improved their ability to enhance substrate oxidation in state 3 required for and matrix volume recovery (Fig. 6). Superoxide dismutase (SOD) was ineffective in replacing catalase in combination with Nac (data not shown), as might be expected, because SOD activity is high in the matrix already. It has been estimated that the steady-state matrix concentration of superoxide is about two orders of magnitude lower than that of H2O2 (5).
|
To investigate further the cause of the catalase-induced increase in DCF fluorescence, additional experiments were performed in isolated mitochondria loaded with a chloromethyl derivate of DCFH diacetate, which is better retained than DCFH and is more suitable for long incubations. Figure 7A shows that when mitochondria were energized with caproic acid, the subsequent addition of catalase increased DCF fluorescence, as shown in Fig. 5. Mitochondria were then subjected to anoxia by infusion of N2 into the tightly closed cuvette, resulting in a decrease in buffer O2 content (Fig. 7A, bottom trace). After O2 fell below the sensitivity of the O2 electrode, the DCF fluorescence rate decreased, consistent with O2 levels that are inadequate to support ROS generation (5, 8). With reoxygenation, the DCFH oxidation rate increased again (Fig. 7A). Similarly, in Fig. 7B, caproic acid-energized mitochondria were subjected to anoxia and then energized with catalase. In the O2-depleted state, catalase had a small effect on DCF fluorescence, which increased significantly once buffer O2 was increased. These findings further validate the belief that the changes in DCF fluorescence track ROS production and also indicate a modest nonspecific effect (4, 12, 16). Figure 7B also shows that, in the presence of O2, addition of the complex I inhibitor rotenone increased DCF fluorescence, suggesting that complex I was the source of increased ROS production. Further addition of complex IV substrates tetramethyl-p-phenylenediamine and ascorbic acid rapidly increased O2 consumption and inhibited further DCFH oxidation. DCFH oxidation remained inhibited after NaN3 addition, suggesting that a change in the redox environment after addition of exogenous reducing substrates was responsible.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
during ADP phosphorylation in the absence of exogenous substrates. This ability to oxidize endogenous substrates in response to ADP loading can be demonstrated only under specific conditions, i.e., after addition of catalase and Nac, which are expected to reduce H2O2 accumulation and preserve a reducing environment on both sides of the inner membrane. Although not investigated in the present study, a decrease in ROS production is expected to decrease also proton leak of the inner membrane (3), which could have a significant effect on ADP phosphorylation. Under normal resting conditions, heart mitochondria are known to oxidize primarily fatty acids, such that termination of respiration during rapid heart extraction and cooling is likely to leave significant amounts of activated fatty acids trapped in the matrix. Total intramitochondrial CoA is 3 mM, and during maximal -oxidation of fatty acids, 90–95% of CoA is acylated (7). If we assume a matrix volume of 1 µl/mg, 1 mg of mitochondria may contain 2 nmol (1.2 x 1015 molecules) of activated long-chain fatty acids. Complete oxidation of this amount would allow phosphorylation of 1.5 x 1017 ADP molecules, which is reasonably close to the amount we observed after addition of catalase and Nac (phosphorylation of 1.4 µmol of ADP translates to 8 x 1017 molecules of ADP; Fig. 1D). It seems plausible that the explanation for the effect of catalase + Nac (Fig. 1) is activation of intramitochondrial substrate oxidation related to addition of ADP. Apparently, redox conditions generated by catalase and Nac allow much more complete fatty acid oxidation in response to ADP, so that phosphorylation is inhibited in parallel with depletion of fatty acids trapped in the matrix. Although it is not possible to extrapolate our results directly to in vivo conditions, it is possible that a similar protection of functionally important SH groups exists inside healthy cardiac myocytes, in which powerful antioxidant systems such as catalase, glutathione peroxidase, and glutathione reductase maintain low H2O2 levels and a reduced environment. This could mean that fatty acid oxidation in cardiac cells is redox dependent, such that, in a reduced environment, low concentrations of activated fatty acids can be efficiently utilized for ATP production, whereas in an oxidized environment, fatty acid transport into the mitochondria is required to support the same ATP production rate. One caveat is that in vivo cytoplasmic O2 concentration is only a few micromoles (10); i.e., heart mitochondria are protected from the high O2 concentrations to which isolated mitochondria are typically exposed. On the basis of in vitro experiments, it is also difficult to understand how such high rates of respiration are maintained at low-micromolar O2 concentrations.
Mechanism of Catalase + Nac Effects
The finding that catalase + Nac was required to enhance ADP phosphorylation in mitochondria without exogenous substrates implies that removal of H2O2 and preservation of a reduced redox state were required. Catalase enhanced matrix DCF fluorescence associated with endogenous or exogenous fatty acid oxidation (Fig. 5). This was a rather unexpected finding. We confirmed that the major component of DCFH oxidation was directly related to ROS production, but we also found that DCFH oxidation in the matrix was somewhat sensitive to nonspecific factors (Fig. 7). Consistent with its effect on increasing DCFH oxidation, catalase also increased , which can be explained by relief of electron transport inhibition due to increased H2O2 efflux. St.-Pierre et al. (26) showed that isolated heart mitochondria respiring on palmitoyl carnitine generated H2O2 at a much higher rate than mitochondria respiring on pyruvate and malate. ROS production was enhanced without addition of any inhibitors and originated from the matrix, because addition of SOD had no significant suppressant effect. To account for enhanced ROS production during fatty acid oxidation, St.-Pierre et al. proposed greater steady-state reduction of complex I or additional superoxide production by electron transfer flavoprotein and electron transfer flavoprotein quinone oxidoreductase, which could act as potential sources of ROS production. Our findings are consistent with the belief that complex I is the site of increased ROS production by endogenous fatty acids in the presence of catalase, because rotenone enhanced the rate of DCFH oxidation by catalase (Fig. 7B). Because catalase is too large to enter the intermembrane space or matrix, it presumably acts indirectly by lowering extramitochondrial H2O2 concentration and creating a sink for H2O2 efflux from the matrix. Lowering matrix H2O2 per se would be expected to reduce DCFH oxidation, but because the opposite occurred, we hypothesize that the decrease in H2O2 in the vicinity of mitochondria accelerates ROS production in mitochondria.
Nac, on the other hand, is known to penetrate membranes (1, 18, 25) and can inhibit matrix DCFH oxidation (Fig. 5C). As demonstrated in Fig. 4, catalase by itself activated endogenous fatty acid oxidation, but Nac significantly enhanced the effect of catalase but had little effect on its own. One of the expected effects of Nac would be an increase in the amount of free CoA (CoA-SH) in the matrix, which would accelerate -oxidation flux and also acetyl-CoA oxidation in the citric acid cycle. A decrease in the available CoA-SH pool, coupled with insufficient concentration of citric acid cycle intermediates, would suppress fatty acid oxidation. The dramatic additive increase in matrix ROS by fatty acid oxidation and catalase would further decrease CoA-SH. As a reducing agent, Nac may relieve this inhibition by regenerating CoA-SH. It is interesting that the membrane-permeant ROS scavenger MPG was unable to fully substitute for Nac when catalase was present (Fig. 6). This may favor the CoA-SH explanation, because Nac directly regenerates CoA-SH, whereas MPG only indirectly inhibits CoA-SH oxidation by decreasing ROS and, therefore, may be less efficient.
A highly reduced state could also inhibit permeability transition pore flickering, such that the membrane would hyperpolarize and superoxide production would increase as proposed recently to explain the role of glutathione in the modulation of H2O2 production (24). Flickering would be stimulated by uptake of the contaminant Ca2+ (30), which could explain the effect of EGTA (Fig. 1B). However, cyclosporin A, which is expected to inhibit pore flickering, did not mimic the effect of catalase + Nac.
Inhibition of fatty acid oxidation by pyruvate, and vice versa, can be explained by the well-known reciprocal relation between these oxidation pathways, inasmuch as an increase in the mitochondrial ratio of acetyl-CoA to CoA-SH inhibits the pyruvate dehydrogenase complex (17, 20) and also -oxidation (7).
Physiological Implications
We found that endogenous fatty acids were unable to maintain and matrix volume required to support vigorous ADP phosphorylation unless a reduced redox environment was protected by catalase and Nac. A similar protection of functionally important SH groups inside healthy cardiac myocytes is normally provided by powerful antioxidant systems, such as catalase, glutathione, glutathione peroxidase, and glutathione reductase, to maintain low H2O2 levels and a reduced environment. Interestingly, a decrease in H2O2 production by overexpression of catalase targeted to mitochondria has recently shown to increase the life span of mice (23). The stimulation of -oxidation of endogenous fatty acids in phosphorylating mitochondria by catalase and Nac suggests that -oxidation flux is regulated by the redox environment. Moreover, in an oxidizing environment, much higher intramitochondrial fatty acid levels are required to phosphorylate the same amount of ADP. The extent to which this redox-dependent control is physiologically important, however, remains a matter of speculation, because fatty acid channeling and -oxidation regulation are complex.
It is interesting that, in isolated mitochondria, pyruvate alone had a limited ability to support during sustained ADP phosphorylation, which could be overcome by providing citric acid cycle intermediates, such as malate (Fig. 2C). Recently, studies in pyruvate-energized skeletal muscle mitochondria showed that the ADP-induced increase in O2 consumption was only 2% of the increase in mitochondria energized with pyruvate and malate (1 mM each). These findings suggest that isolated muscle mitochondria may contain little or no citric acid cycle intermediates (14). In cardiac cells, relative loss of citric acid cycle intermediates has shown to limit contractile function (20).
In summary, in the absence of exogenous substrates, cardiac mitochondria have a surprisingly large capacity to oxidize endogenous substrates in response to a decrease in , provided that the extramitochondrial environment removes H2O2 and a reduced environment is maintained in the matrix. This mechanism may be important for maintaining -oxidation flux and ATP production if mitochondrial uptake of fatty acids is acutely inadequate, which might occur during sudden increases in cardiac work or during ischemia. On the other hand, prooxidant conditions during ischemia-reperfusion are expected to disable this mechanism and, potentially, contribute to mitochondrial dysfunction in this setting.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-oxidation flux. Prog Lipid Res 41: 197–239, 2002.[CrossRef][ISI][Medline]
B and transcription of human immunodeficiency virus. Proc Natl Acad Sci USA 87: 9943–9947, 1990.
O2, NADH, and light scattering. J Biol Chem 276: 2586–2599, 2001.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
