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

This Article Abstract Full Text (PDF) All Versions of this Article: 292/2/C708    most recent 00202.2006v1 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 ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (1) Citing Articles via Google Scholar Google Scholar Articles by Panov, A. Articles by Rosenfeld, J. Search for Related Content PubMed PubMed Citation Articles by Panov, A. Articles by Rosenfeld, J.

SPECIAL SECTION ON MITOCHONDRIAL MODELING AND FUNCTION

Species- and tissue-specific relationships between mitochondrial permeability transition and generation of ROS in brain and liver mitochondria of rats and mice

¹Carolinas Neuromuscular/ALS-MDA Center, Carolinas Medical Center, Charlotte, North Carolina; and ²Center for Neurodegenerative Disease, and ³Free Radicals in Medicine Core, Division of Cardiology, Emory University, Atlanta, Georgia

Submitted 24 April 2006 ; accepted in final form 11 October 2006

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
In animal models of neurodegenerative diseases pathological changes vary with the type of organ and species of the animals. We studied differences in the mitochondrial permeability transition (mPT) and reactive oxygen species (ROS) generation in the liver (LM) and brain (BM) of Sprague-Dawley rats and C57Bl mice. In the presence of ADP mouse LM and rat LM required three times less Ca²⁺ to initiate mPT than the corresponding BM. Mouse LM and BM sequestered 70% and 50% more Ca²⁺ phosphate than the rat LM and BM. MBM generated 50% more ROS with glutamate than the RBM, but not with succinate. With the NAD substrates, generation of ROS do not depend on the energy state of the BM. Organization of the respiratory complexes into the respirasome is a possible mechanism to prevent ROS generation in the BM. With BM oxidizing succinate, 80% of ROS generation was energy dependent. Induction of mPT does not affect ROS generation with NAD substrates and inhibit with succinate as a substrate. The relative insensitivity of the liver to systemic insults is associated with its high regenerative capacity. Neuronal cells with low regenerative capacity and a long life span protect themselves by minimizing ROS generation and by the ability to withstand very large Ca²⁺ insults. We suggest that additional factors, such as oxidative stress, are required to initiate neurodegeneration. Thus the observed differences in the Ca²⁺-induced mPT and ROS generation may underlie both the organ-specific and species-specific variability in the animal models of neurodegenerative diseases.

permeability transition; reactive oxygen species generation; interspecies difference

Experiments with isolated rat liver mitochondria (RLM) (12) and cultured hepatocytes (12, 31) have shown that peroxides, such as tert-butyl hydroperoxide (t-BOOH), promote mitochondrial PT (mPT) even in the presence of very small amounts of Ca²⁺. The current view is that oxidative stress plays an important role in promoting the Ca²⁺-induced mPT (12, 62), and t-BOOH was regarded as a useful tool to study the effects of oxidative stress on mitochondrial functions (12, 31). It was suggested that increased oxidative stress may be an important factor in pathogenesis of neurodegenerative diseases, such as Parkinson's disease (40), amyotrophic lateral sclerosis (47), and Huntington's disease (HD) (46).

The progress in studies of neurodegenerative diseases, such as HD and amyotrophic lateral sclerosis, was also promoted by creation of genetically engineered animal models of these diseases (reviewed in Ref. 65), and by introduction of the in vivo toxic models of Parkinson's disease (5, 14, 40) and HD (9). However, the animal models of the diseases revealed a basic problem: different species respond differently to the pathological agents used in the in vivo models of neurodegenerative diseases (20, 48). Besides, recent experiments have shown that during systemic intoxication with rotenone, the brain mitochondria were severely damaged, whereas liver mitochondria remained virtually unaffected (40).

Thus several fundamental questions arise that have to be answered to successfully use the toxic and genetic animal models of neurodegenerative diseases. Some of the uncertainties stem from the trivial fact that most of our current knowledge about mitochondrial functions, permeability transition in particular, was obtained in experiments with mitochondria from rat tissues, predominantly the liver. The genetic models of neurodegenerative diseases are, however, usually created with mice and involve studies of tissues other than liver.

The first question to consider examines the extent to which brain mitochondria are different from liver mitochondria. There is no consensus in this respect in the literature. The published respiratory activities for brain mitochondria may differ >10-fold, whereas some authors failed to find any difference in respiratory activity between liver and brain mitochondria (3, 64). However, it was also found that compared with rat liver mitochondria, rat brain mitochondria are not sensitive to a damaging effect of hydroperoxides (27), do not produce ROS such as superoxide (15, 54) and nitric oxide (30), or, on the contrary, generate more ROS than other tissue mitochondria (11). There are indications that brain mitochondria are less sensitive to a damaging effect of Ca²⁺ (27), and are less sensitive to the protective effect of cyclosporin A during calcium loads (39).

The second question regards the species differences in mitochondrial functions. There were reported species-specific differences in state 4 (resting respiration) respiratory rates, which depend on intrinsic proton conductivity of mitochondria, and thus determine the standard metabolic rate of an animal. Some of these differences were related to the size of the animals (8). To successfully use animal models of diseases, we have to understand the nature of these species-specific differences in mitochondrial functions. It is well established that responses of animals to pathological agents are genotype dependent (20, 48). However, little is known how these diversities are translated at the functional level in different organs. Because mPT and ROS production are thought to play important roles in neurodegeneration it was logical to focus our attention on these two mitochondrial functions.

Thus the purpose of this work was to study organ-specific and species-specific differences between rat and mouse liver and brain mitochondria in their major functions, namely permeability transition (PT) and generation of ROS. The abnormalities of these functions may contribute to the pathogenesis of neurodegenerative diseases, and the differences in the relationships between mPT and ROS generation in rats and mice may underlie the species-specific diversity in animal models of neurodegenerative diseases.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
Animals. Two- to three-month-old male Sprague-Dawley rats and male C57Bl/6J mice were used for isolation of the liver and brain mitochondria. The animals were housed and cared for in American Association for Accreditation of Laboratory Animal Care-accredited facilities at the Carolinas Medical Center and Emory University. All experiments involving animals were performed in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Isolation of the liver and brain mitochondria. Both liver and brain mitochondria were isolated in medium that contained (in mM) 225 mannitol, 75 sucrose, 20 MOPS (pH 7.2), 1 EGTA, and 0.1% BSA. Liver mitochondria (LM) were isolated by conventional differential centrifugation with a final spin at 8,600 g (40). Brain mitochondria were isolated from the pooled forebrains of three rats. We used the modified method of Sims (53) to isolate and purify brain mitochondria (BM) in a Percoll gradient. The modifications were as follows: brain tissue was homogenized with 15 strokes of a loose pestle in a Dounce homogenizer, and 5-ml volumes per tube of 15%, 23%, and 40% (vol/vol) of Percoll solutions were used to purify the brain mitochondria. After the final sedimentation of mitochondria at 8,600 g, the mitochondria were suspended in 250 mM sucrose and 10 mM MOPS (pH 7.2). Mitochondrial protein was determined using the Pierce Coomassie protein assay reagent kit.

Registration of PT and estimation of Ca²⁺ retention capacity. Recently, we introduced a quantitative parameter, Ca²⁺ retention capacity (CRC) that allows a meaningful comparison of the sensitivity to Ca²⁺ of mitochondria from different organs and species (39). CRC is the amount of Ca²⁺ that can be accumulated and retained by mitochondria until the PT occurs. It is expressed as nanomoles of Ca²⁺ per mg of mitochondrial protein. Ca²⁺ was added to 2 ml of mitochondrial suspensions (0.5 mg/ml) using aliquots of 5, 10, or 20 mM stock solutions of CaCl₂ to achieve final concentrations of Ca²⁺ of 12.5, 25, or 50 nmol/ml. At high CRC we switched to higher concentration of the CaCl₂ stock solutions to minimize volume changes. We utilized CaCl₂ of very high purity: 99.99% from Sigma.

The following three different methods were used to estimate CRC and PT.

First, potentiometric measurement of pH changes of the incubation medium during Ca²⁺ accumulation and release by the mitochondria, as described in Ref. 39. The pH measurements were performed with the use of a pH meter (model 440, Corning) equipped with a mono pH microelectrode from Lazar and an Ag/AgCl reference electrode connected to the incubation chamber by a KCl bridge.

Second, depolarization of mitochondria was registered by determining membrane potential () with a tetraphenyl phosphonium (TPP⁺)-sensitive electrode as described elsewhere (44). Because the volume of the matrix space and binding constants for TPP⁺ in brain mitochondria and mouse liver mitochondria are unknown, the values presented are approximate, and show only the dynamics and direction of changes.

Finally, swelling of mitochondria, registered as a decrease in optical density, was recorded at 545 nm using a Shimadzu spectrophotometer (Multispec model 1501). We measured simultaneously 2–3 parameters in mitochondria in various combinations.

The mitochondrial CRC and swelling were estimated in a medium (medium B) containing (in mM) 125 KCl, 10 NaCl, 0.5 MgCl₂, 3 glycyl-glycine, pH 7.2, 1 KH₂PO₄, and 20 glutamate plus malate 2 for brain and liver mitochondria. Mitochondrial protein was 0.5 mg/ml. The composition of medium B used in the present study to measure CRC is different from the sucrose-based medium we used in our previous studies (39, 40). We have found that the sucrose-based medium with only 20 mM KCl and absence of Mg²⁺ may inhibit maximal respiratory activities of the brain mitochondria (A. Panov et al., unpublished observations). Therefore, the current composition of the medium for measuring CRC only minimally differs from the medium A, which is optimal for maximal respiratory rates of brain mitochondria (40). We have found that the current composition of medium B, compared with the sucrose-based medium, gives somewhat higher and better reproducible values of CRC for the same type of brain mitochondria.

At the end of an experiment, the addition of 125 nmol/ml H⁺, that caused a pH of 0.07 pH units, served as an internal calibration. This allowed a comparison of the total pH changes and calculation of the H⁺/Ca²⁺ ratios. Similarly, additions of 0.5 µM TPP⁺ aliquots served as internal calibration for membrane potential changes. In experiments with liver mitochondria the final concentrations of TPP⁺ were 1.5 µM, with brain mitochondria –2 µM.

To determine maximal amplitudes of the mitochondrial swelling and complete release of CaP_i from the mitochondria after mPT, we added a bacterial toxin, alamethicin (5 µg/mg mitochondrial protein). Alamethicin forms a large conductance pore for low molecular mass hydrophilic substrates (21).

Measurements of H₂O₂ generation. H₂O₂ was determined using Amplex red (Molecular Probes) method. In the presence of horseradish peroxidase, the following reaction occurs: Amplex red + H₂O₂ resorufin + O₂. Resorufin is a stable and highly fluorescent compound with a wavelength spectra excitation/emission of 570/585 nm. The fluorescence of resorufin was determined in 1-ml incubations in a medium (medium A) containing (in mM) 125 KCl, 10 MOPS, pH 7.2, 2 MgCl₂, 2 KH₂PO₄, 10 NaCl, 1 EGTA, 0.7 CaCl₂, and 0.2 mg/ml mitochondrial protein, 5 µM Amplex red, and 3 units of horseradish peroxidase, as described in Ref. 40. At the EGTA/Ca²⁺ ratio of 1/0.7 the concentration of free [Ca²⁺] was close to 1 µM, as determined with the Fura-2 method. We measured H₂O₂ production in the presence of glutamate 20 mM + malate 2 mM, or succinate 5 mM. When present, the respiratory inhibitors and uncoupler were added to the incubation media prior to addition of mitochondria. Additions of 50 nM of resorufin or correspondingly diluted standard 3 mM solution of H₂O₂ (Fluka) were used for calibration of the fluorescence scale. The background fluorescence was subtracted from experimental fluorescence, and the scale between 4,000 and 6,000 fluorescence arbitrary units, which corresponded to 44 nM of H₂O₂ or resorufin, was used in calculations. There was a very close correlation between the fluorescence of freshly diluted H₂O₂ standard solution and the resorufin standard. However, H₂O₂ solutions decayed rapidly while resorufin solutions remained stable for a long time. Fluorimetric measurements were made using a fluorometer from C&L (Middletown, PA; www.fluorescence.com).

Data acquisition. The data acquisition was performed using hardware and software from C&L.

Chemicals. Chemicals were of highest purity available. All solutions were made using glass-bidistilled water.

Statistics. Data are presented as means ± SE of 4–6 separate experiments. For comparison of two groups, a two-tailed t-test was employed using Microsoft Excel software. Statistical significance was assumed when P < 0.05.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
PT in liver mitochondria. Figure 1 shows changes in , Ca²⁺ accumulation registered with the pH method (39) and optical density changes of the RLM (Fig. 1A) and MLM (Fig. 1B) during titration with calcium. The figures show that with RLM and MLM depolarization and slow rate swelling began before the large pore opening registered as alkalization of the medium (see also Ref. 39) clearly designating the moment of a large pore opening. These pH changes are bound to a dissociation of Ca²⁺ phosphate salts deposited in the matrix and binding of protons to the released HPO₄^2– and PO₄^3– anions to accommodate the medium's pH (15, 39).

View larger version (21K): [in this window] [in a new window]   Fig. 1. Changes in optical density, membrane potential, and medium pH during titration of mitochondria with calcium. A: rat liver mitochondria (RLM). B: mouse liver mitochondria (MLM). Incubation conditions for medium B: final volume 2 ml. TPP+ was added in 0.5 µM aliquots to achieve final concentration of 1.5 µM mitochondria. Calcium was added in 12.5 nmol/ml aliquots; 5 µg/ml alamethicin and 125 nmol/ml HCl caused pH of 0.07. See MATERIALS AND METHODS for other experimental details.

View larger version (22K): [in this window] [in a new window]   Fig. 2. Changes in optical density, membrane potential and medium pH during titration of mitochondria with calcium in the presence of 500 nM cyclosporin A (CsA) , 50 µM ADP, and 2 µg/mg mitochondrial protein of oligomycin. A: RLM. B: MLM. Ca2+ was added in 25 nmol/ml aliquots. Other additions are the same as in Fig. 1.

View larger version (22K): [in this window] [in a new window]   Fig. 3. Changes in optical density, membrane potential, and medium pH during titration of mitochondria with Ca2+. A: rat brain mitochondria (RBM). B: mouse brain mitochondria (MBM). Incubation conditions as in Fig. 1. TPP+ was added in 0.5 µM aliquots to achieve final concentration of 2 µM, brain mitochondria 0.5 mg/ml, Ca2+ was added in 25 nmol/ml aliquots. Other additions are the same as in Fig. 1.

View larger version (21K): [in this window] [in a new window]   Fig. 4. Changes in optical density, membrane potential, and medium pH during titration of mitochondria with Ca2+ in the presence of 500 nM CsA, 50 µM ADP, and2 µg/mg mitochondrial protein oligomycin. A: rat brain mitochondria. B: mouse brain mitochondria. Ca2+ was added in 50 nmol/ml aliquots. Other additions are the same as in Fig. 1.

View larger version (11K): [in this window] [in a new window]   Fig. 5. Effect of mitochondrial permeability transition pore (mPTP) inhibitors on Ca2+ retention capacity of mitochondria. Incubation conditions are the same as those in Figs. 1–4. A: rat liver and brain mitochondria. B: mouse liver and brain mitochondria. (1) Control; (2) ADP (50 µM) + oligomycin (2 µg/ml); (3) CsA (500 nM); (4) CsA + ADP + oligomycin. The P value over the control columns in A and B shows a statistical comparison between liver and brain mitochondria. *P < 0.05, **P < 0.01, ***P < 0.001, statistically significant difference of changes related to the corresponding controls for liver or brain.

The unprotected mouse brain mitochondria (MBM) did not change their optical density (OD) during incubation or addition of Ca²⁺ aliquots (Fig. 3B). When the large pore opened, as indicated by the collapse of and alkalization of the medium, the swelling of MBM also began, but the amplitude of swelling was small (0.2 OD). Unlike RBM, the opening of mPTP in MBM was transient and stopped spontaneously (see Fig. 3B), and was reopened again after addition of cyanide m-chlorophenylhydrazine (CCCP). This effect of CCCP was not associated with mitochondrial deenergization, but by the uncoupler directly promoting opening of mPTP (51). Figure 5, A and B, show that unprotected MBM sequestered two times (P < 0.001) more Ca²⁺ before opening of mPTP than RBM.

When RBM and MBM were protected with ADP (plus oligomycin) the RBM sequestered almost 5-fold and MBM 3-fold more CaP_i than the unprotected brain mitochondria. ADP-protected MBM sequestered 70% (P < 0.001) more CaP_i than the RBM (Fig. 5, A and B). As with the rat and mouse liver mitochondria, the RBM and MBM protected with CsA + ADP sequestered the same amounts of CaP_i (Fig. 5, A and B). In the RBM and MBM protected with ADP (not shown) or CsA + ADP (Fig. 4, A and B), collapse of membrane potential, alkalization of the medium and swelling occurred simultaneously. Before the opening of mPTP, the optical density of RBM and MBM may be noticeably increased, evidently due to sequestration of the optically dense CaP_i salts. remained relatively stable. The amplitudes and the rates of swelling of the RBM and MBM after opening of mPTP were significantly larger than in the unprotected mitochondria.

H⁺/Ca²⁺ ratios during Ca²⁺ and phosphate accumulation by mitochondria. Chalmers and Nicholls (15) suggested that the H⁺/Ca²⁺ ratio, the ratio of the net H⁺ extruded to the number of Ca²⁺ ions accumulated, reflects the type of the CaP_i salt sequestered by the mitochondria. A decrease in the H⁺/Ca²⁺ ratio would mean that more CaP_i is sequestered as CaHPO₄ x 2H₂O, which is relatively soluble. An increase in the H⁺/Ca²⁺ ratio to 1.0 would indicate that more CaP_i is sequestered as Ca₃(PO₄)₂, which is 15.9 times less soluble than the CaHPO₄ salt (17).

In the experiments presented in this paper for Sprague-Dawley rats (Figs. 1A–4A) and C57Bl/6J mice (Figs. 1B–4B), the H⁺/Ca²⁺ ratios for unprotected RLM and MLM were 0.77 ± 0.01 and 0.79 ± 0.01 correspondingly. When liver mitochondria were protected with CsA + ADP + oligomycin, the H⁺/Ca²⁺ ratio of both RLM and MLM increased after the first few additions of Ca²⁺ to 0.97 ± 0.02. With unprotected RBM and MBM, the H⁺/Ca²⁺ ratios were correspondingly 0.83 ± 0.01 and 0.85 ± 0.02. With protected brain mitochondria, the H⁺/Ca²⁺ ratio during titration with Ca²⁺ increased rapidly to 1.0 showing that CaP_i was sequestered as almost insoluble Ca₃(PO₄)₂. Thus sequestration of CaP_i is a highly dynamic process and can vary depending on the type of mitochondria and incubation conditions. Our data show that relatively small changes in the H⁺/Ca²⁺ ratios may result in large changes in the amounts of Ca²⁺ sequestered by the mitochondria.

Effects of PT modifiers on CRC. It is well established that the addition of 200 µM t-BOOH to RLM significantly decreases the amount of Ca²⁺ necessary to open mPTP (12, 61). However, we have found that t-BOOH was ineffective with MLM. CRC of the MLM remained at the control level even in the presence of 1 mM t-BOOH (not shown). RBM and MBM were also insensitive to the presence of this hydroperoxide (not shown). These results indicate that in RLM mPT is radically different compared with RBM, MBM, and MLM.

To analyze the properties of mPTP in rat and mouse liver and brain mitochondria, we used various compounds that are known to affect mPT. Figure 5, A and B, show that in the absence of mPT modifiers mouse brain and liver mitochondria sequester considerably (P < 0.001) more CaP_i than the corresponding rat mitochondria. Both rat and mouse brain mitochondria were much more sensitive to the protective effect of ADP than CsA alone. In the presence of ADP+CsA, however, there was no difference in CRC values between the rat and mouse mitochondria. Brain mitochondria sequestered two times more CaP_i than the liver mitochondria.

Mitochondrial generation of ROS. Most mitochondria have active superoxide dismutase, both in the matrix (Mn²⁺-SOD) and in the intermembrane space (Cu²⁺-Zn²⁺-SOD), therefore it is impossible to study intramitochondrial ROS production by following extramitochondrial O₂^– released by intact mitochondria (40, 55). The best way to follow generation of O₂^– in intact mitochondria is to determine formation of H₂O₂ (40, 55). In this study we used the Amplex red method and a highly sensitive fluorometer as described (40).

Generation of ROS by liver mitochondria. With RLM and MLM oxidizing succinate or glutamate + malate, there was almost no difference in the rates of H₂O₂ generation (correspondingly 92.1 ± 4 and 80 ± 4 pmol H₂O₂·min^–1·mg protein^–1), which was in stark contrast to the 4- to 6-fold difference between the two substrates in RBM and MBM shown in Figs. 6 and 7. We have also found that with the exception of antimycin A, addition of respiratory inhibitors rotenone, myxothiazol, and their combinations to LM had no effect on the observed rates of H₂O₂ production (not shown, see also Ref. 40). As we have suggested and explained earlier (40), measurements of the extramitochondrial H₂O₂ cannot be used for analysis of ROS generation by the intact liver mitochondria.

View larger version (11K): [in this window] [in a new window]   Fig. 6. Generation of H2O2 by RBM and MBM with glutamate as a substrate. A: response of H2O2 production by RBM to additions of CCCP and respiratory chain inhibitors. B: quantitative comparison of ROS generation by RBM and MBM oxidizing glutamate. Incubation conditions for medium A (see MATERIALS AND METHODS) were glutamate 20 mM + malate 2 mM, 5 µM Amplex red, 3 units of horseradish peroxidase (HRP), 0.2 mg rat brain mitochondria, volume 1 ml. Catalase (Roche Diagnostics) 30,000, in experiments when Antimycin A was present: 130,000 units; 5 µM Antimycin A, 5 µM Rotenone, 5 µM Myxothiazol, and 0.5 µM cyanide-m-chlorophenylhydrazone (CCCP). *P < 0.05, **P < 0.01, ***P < 0.001, statistically significant difference between rat and mouse brain mitochondria.

View larger version (11K): [in this window] [in a new window]   Fig. 7. Generation of H2O2 by RBM and MBM with succinate as a substrate (5 mM). Other incubation conditions and additions were as in Fig. 6. A: response of H2O2 production by RBM to additions of CCCP and respiratory chain inhibitors. B: a quantitative comparison of ROS generation by RBM and MBM oxidizing succinate. AU, arbitrary units.

Figure 6A shows that with glutamate + malate, the rate of ROS generation by BM was the same in the energized and in the uncoupled mitochondria, when membrane potential was collapsed. Thus with glutamate or pyruvate (not shown) as a substrate generation of ROS does not depend on mitochondrial energization.

Upon addition of rotenone there was a 4- to 5-fold increase in ROS generation (Fig. 6, A and B). Upon addition of antimycin A or myxothiazol, there was correspondingly a 5.5-fold and 3.3-fold increase in generation of ROS. A significantly larger effect of antimycin A indicates that some of the ROS was generated on complex III. With myxothiazol, which prevents reduction of the CoQ centers on complex III, and thus abolishes the effect of antimycin A, evidently all of the ROS was generated on Complex I. MBM generate (80 pmol H₂O₂/mg protein), that is 62% more ROS (P < 0.001) than RBM (49.3 pmol H₂O₂/mg protein) in metabolic state 4 (Fig. 6B); however, there was no difference in ROS production between MBM and RBM in the presence of rotenone or uncoupler CCCP. In the presence of myxothiazol or antimycin A, MBM generated more ROS than the RBM. The mechanism of this difference is unclear.

Figure 7A shows representatively how brain mitochondria respond to CCCP and respiration inhibitors when ROS generation was supported by succinate. Figure 7B gives a quantitative comparison of ROS generation by RBM and MBM. Figure 7B shows that under all conditions there was no difference between the two species in the rates of ROS production by BM oxidizing succinate. The addition of rotenone to BM inhibited ROS production by 80%. However, energization of the mitochondria was preserved. Antimycin A also inhibits respiration and energization of the mitochondria and thus prevents backward electron flow. However, in the presence of antimycin A, the CoQ sites of complex III become reduced and thus increase O₂^– generation (35). Because the CoQ_O site of complex III is located close to the outer surface of the inner mitochondrial membrane, a large portion of O₂^– is released from the mitochondria (35, 60). This is one reason why antimycin A dramatically increases generation of ROS with succinate as a substrate. The other reason for the 4- to 5-fold increase in ROS generation may be explained by the fact that mitochondria have 3 times more complexes III than complexes I (23, 49). Thus in the presence of antimycin A most of O₂^– was generated on complex III.

Myxothiazol, which inhibits respiration and prevents reduction of the complex III CoQ sites, also caused deenergization of the mitochondria and thus inhibited the energy-dependent backward electron flow driven by succinate. As a result, production of ROS was at the same level as with rotenone. In the presence of rotenone and myxothiazol, the electrons from the membrane pool of CoQ reduced by succinate cannot go downstream to complex III, or upstream because mitochondria are de-energized and rotenone would block reduction of the complex I redox sites. Therefore, in the presence of rotenone + myxothiazol generation of ROS may occur only on complex II (40). Figure 7, A and B, show that with rotenone, myxothiazol, or rotenone + myxothiazol, the rates of ROS production were practically the same, and thus complex II was responsible for this basic rate generation of ROS. When CCCP was added to mitochondria oxidizing succinate, the rate of ROS production was the same or lower than that observed in the presence of rotenone and myxothiazol.

Taken together, the data show that generation of ROS is strongly substrate dependent. With the NAD-dependent substrates, generation of ROS does not depend on mitochondrial energization or the functional state of the mitochondria, whereas with succinate it does. Importantly, with the NAD-dependent substrates MBM during resting respiration generate 50% more ROS than RBM (Fig. 6B).

Simultaneous measurements of ROS generation and mPTP opening. It was suggested by some researchers that mPT stimulates ROS generation by the mitochondria (14, 32). Figure 8 shows simultaneous registration of ROS generation and the Ca²⁺-induced mPT in RBM oxidizing either glutamate + malate (Fig. 8A) or succinate (Fig. 8B). Similar results were obtained with MBM (not shown). With glutamate + malate there was a slight inhibition of ROS generation during active Ca²⁺ consumption, which again slightly increased to the initial level upon opening of mPTP. These changes in the rate of ROS generation were small. Thus with the NAD-dependent substrates mPT does not affect dramatically the rate of ROS generation.

View larger version (19K): [in this window] [in a new window]   Fig. 8. Simultaneous measurements of H2O2 generation and mPT registered with membrane potential and pH method. A: RBM oxidizing 20 mM glutamate + 2 mM malate. B: RBM oxidizing succinate (5 mM). Incubation conditions and additions are the same as in Fig. 3A.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
PT in BM and LM. We employed simultaneously several methods to register mPT because mPTP opening is always accompanied by a collapse of , but mitochondrial deenergization does not always results in opening of mPTP (4, 39, 42). Regarding BM, there is still controversy as to whether mPT causes mitochondrial swelling or not, and the role of swelling in cytochrome c release (10, 15, 28, 64). We studied brain mitochondria from several strains of rats and mice and found that in the absence of mPT modifiers, BM from several species do not spontaneously open mPTP and do not undergo swelling (39). In general, the amplitudes of swelling of the unprotected brain mitochondria were not large, evidently not enough to rupture the outer membrane (39). This suggestion agrees with the observation made by Andreev and Fiskum (2) that unlike RLM, in RBM, cytochrome c was released by the mPT-independent mechanism. In the presence of ADP, which is a more physiological condition because ADP is always present in a cell, brain mitochondria sequester several times more CaPi (see Fig. 5, A and B) and always undergo mPT and large amplitude of swelling. We believe that in this case cytochrome c release may be associated with mPT as suggested in (10). Thus, it is very likely that the roles of mPT in cytochrome c release and initiation of apoptosis may depend on the species specific intrinsic properties of BM, as well as on assay conditions.

Figures 5, A and B, show that both unprotected and ADP protected MBM require correspondingly two times and 70% more Ca²⁺ to open mPTP than RBM under the same conditions. This might be associated with the fact that MBM have slightly higher H⁺/Ca²⁺ ratios than RBM during CaP_i sequestration. We have shown that even small changes in the H⁺/Ca²⁺ ratios dramatically change the CRC values (see also Ref. 7). A very large effect of ADP (plus oligomycin) on the CRC of BM (see Fig. 5, A and B) is bound to a decrease in the mitochondrial conductivity for H⁺ and K⁺ ions due to a change in the conformational state of ANT (39, 41). Decreased proton conductivity promotes mitochondrial energization, increases the H⁺/Ca²⁺ ratio, and thus increases the ability of mitochondria to sequester more CaP_i (39, 41). The liver mitochondria have a relatively small amount of ANT (reviewed in Ref. 38) compared with mitochondria from other organs, and therefore the protective effect of ADP is small.

Brain mitochondria require additional insults to induce mPT. Because ADP is always present in a cell, we suggest that under in situ conditions the BM may sequester very large amounts of CaP_i. In the presence of ADP, RBM sequester 867 ± 33, and MBM sequester 1,473 ± 79 nmol Ca²⁺/mg protein (see Fig. 5, A and B). Thus normally BM can withstand very large calcium insults without detrimental consequences for the neurons. However, if the cell and/or mitochondria were predisposed to mPT by some other hazardous effect, such as inhibition of the respiratory chain or oxidative stress, the detrimental events may occur. This ability of normal BM to withstand large calcium insults may explain the fact that in most neurodegenerative diseases clinical manifestations of a disease develop well after midlife when mitochondria have undergone significant oxidative stress associated with the aging process (34).

We have found that both rat and mouse BM and, surprisingly, MLM are insensitive to the oxidative stress caused by a peroxide tBOOH, which is in stark contrast with RLM (14). The mechanism of this BM and MLM resistance to this peroxide remains unclear. However, this fact suggests that the damaging effects of peroxides are much less general than was indicated earlier (14). Evidently, different species of ROS responsible for oxidative stress may vary in their mechanisms, and their adverse effects depend on the animal species and the type of mitochondria.

Because neuronal cells have very high respiratory activity and have a large life span (59), prevention of increased ROS generation by brain mitochondria is of paramount biological importance. There have to be evolutionary mechanisms present to lessen or prevent oxidative stress. One of these mechanisms is represented by the cytoplasmic and mitochondrial enzymatic and metabolic antioxidant defenses, such as catalase, SODs, glutathione reductases, and vitamin E, which vary from tissue to tissue (33). Because liver mitochondria have a relatively high antioxidant system, including catalase, this may explain why it is difficult to study generation of ROS in intact liver mitochondria (39).

Here we will outline the intrinsic mitochondrial structural mechanisms to minimize ROS production. For some reasons, these mechanisms have missed the attention of researchers who study oxidative stress, but the factual basis for our discussion was laid down by researchers in other fields of mitochondrial biology (23, 49).

Generation of ROS in brain mitochondria. In mitochondria, ROS can potentially be formed spontaneously at any protein involved in redox reactions, that may have a group capable of one electron-reduced state, such as ubiquinone and transition metals (1, 25). Currently, many authors consider semiubiquinones of the respiratory chain Complex I and Complex III as the major sources of O₂^– (13, 25). Recently, it was shown that mitochondrial -ketoglutarate dehydrogenase can generate ROS (56). However, there is no consensus regarding both the sites and regulation of ROS generation under normal in situ conditions.

In many studies the authors designate the role of a respiratory chain component in oxidative stress by the ability of this component to generate ROS in the presence of a specific inhibitor. For example, antimycin A dramatically increases generation of O₂^·– at Complex III in mitochondria oxidizing succinate, therefore some authors consider Complex III as one of the major sites of ROS production (35). The inhibitors show us only a potential capacity of a given respiratory component to generate ROS in the presence of this inhibitor. Some pathological situations may be compared with the effects of inhibitors. For example, hypoxia may resemble effects of rotenone or antimycin A, and the loss of cytochrome c after opening of mPTP may resemble the effect of antimycin A on ROS generation (29). But these are not normal, physiological situations. Under "normal or physiological conditions" we understand situations when mitochondria function in the absence of added inhibitors at and between the metabolic state 3 in an activated neuron, and closer to metabolic state 4 in a quiescent cell.

Supercomplexes of mitochondrial respiratory chain and ROS generation. From our earlier experiments we have concluded that pyruvate and also glutamate are the best substrates for brain mitochondria (39). This agrees with the current view that the major energy substrate for the brains of adult mammals is glucose (19). Moreover, recent research showed that there are complex interactions between astroglia and neuronal cells designed to optimize energy supply in the form of lactate for activated neurons (reviewed in Refs. 19 and 45).

In Fig. 6, we have shown that both in the energized and de-energized brain mitochondria the rates of ROS generation were at the same relatively low level (50–60 pmol·min·^–1 mg protein^–1 measured as H₂O₂). Thus with the NAD-dependent substrates the rate of ROS generation does not depend on the energy state of mitochondria. Only when the electron flow was prevented by any of the respiratory inhibitors did the generation of ROS increase several folds due to increased reduction of the CoQ sites of Complex I.

That ROS generation in BM oxidizing NAD-dependent substrates is at minimum and does not depend on the energy state of the mitochondria can be explained by the assumption that all mitochondrial components, which potentially can generate ROS, are maintained in the oxidized state. Evidently, the steady low level of ROS generation occurs at the initial rate-limiting step which is more likely either FMN or the (2Fe-2S)_N1a center of Complex I. We suggest that oxidation of the Fe-S clusters and CoQ centers in Complexes I and III is ensured by the superstructure of the respiratory chain.

It was shown that for the heart mitochondria the ratio for oxidative phosphorylation (OXPHOS) complexes I:II:III:IV:V is 1:2:3:6–7:3–5 (23), or more recently determined as 1:1.5:3:6:3 (37). The respiratory complexes interact with each other to form a supercomplex named the respirasome (49). On the basis of the above ratios of the OXPHOS complexes, Schägger et al. (49) suggested that the respirasome exists as a mixture of two large supercomplexes and one smaller complex. Each of the two large supercomplexes are comprised of a Complex I monomer, a Complex III dimer, and four copies of Complex IV. The smaller supercomplex contains two of Complex III and four of Complex IV (49, 50). The major advantages of the supercomplex structure of the mitochondrial respiratory chain are substrate channeling, catalytic enhancement, sequestration of reactive intermediates and structural stabilization (49, 50). The advantage of the substrate channeling is the use of localized substrate molecules, for example, quinone and cytochrome c, which can react independently of bulk properties of a quinone or cytochrome c pool (49).

We suggest that in addition to the benefits listed above, the respirasome is also an evolutionary adaptive mechanism designed to prevent excessive production of ROS. Evidently, this mechanism developed early during the evolution of the aerobic organisms because it is available in aerobic bacteria and yeast (49, 50). The initial reaction of NADH with the FMN of the Complex I is the rate limiting step in oxidation of the NAD-dependent substrates. The suggested composition of the respirasome ensures that all components proximal to Complex IV, which is in a 4-fold excess of Complex I, are kept oxidized regardless of the energy state of the mitochondria. In brain mitochondria, which respire at rates similar or even higher than heart mitochondria, the composition of the respirasome should be similar to that reported for beef heart mitochondria (49, 50).

Recently, we have found that unlike RLM, brain mitochondria have no intramitochondrial substrate storage pool (A. V. Panov, unpublished observations). This may represent yet another adaptive mechanism to minimize generation of ROS when brain mitochondria are inactive or de-energized.

Possible control of succinate-driven ROS generation. In Fig. 7, we have shown that in the energized brain mitochondria the rate of ROS generation with succinate is 4–6 times higher than with an NAD-dependent substrate. Inhibitor analysis showed that 80% of the total ROS with succinate is generated on Complex I. This conclusion agrees with previously published papers (40, 60, 63). Approximately 20% of the total ROS produced by the BM oxidizing succinate was independent of mitochondrial energization, and occurred on Complex II, presumably on the FAD moiety of the enzyme (40). As is shown in Figs. 6 and 7, in the absence of antimycin A there was no generation of ROS on Complex III whether BM oxidized glutamate or succinate.

Because in the presence of rotenone, rotenone + myxothiazol, and CCCP, the rates of ROS production are virtually the same, we may conclude that this basic level of superoxide generation occurs on Complex II, which does not depend on the energy state of the mitochondria. On the basis of the results of inhibitor analysis, we can also conclude that Complex III does not generate O₂^– in energized and de-energized BM. Complex III generates ROS only in the presence of antimycin A (Fig. 7, A and B; see also Ref. 37). The general agreement is that when energized mitochondria oxidize succinate, most of the ROS is generated on CoQ centers of Complex I that become reduced due to the energy-dependent backward electron flow (40, 61, 63).

In situ, when brain mitochondria utilize pyruvate or glutamate as respiratory substrates, succinate is formed during the functioning of the TCA cycle. Succinate dehydrogenase (SDH) is the only TCA cycle enzyme that is also part of the mitochondrial electron transport chain, designated as Complex II. During oxidation of succinate to fumarate complexes II feed electrons into the mitochondrial pool of CoQ with such a force (redox potential E = +90 mV; Ref. 37) that electrons can go not only downstream to Complex III, but also upstream reducing components of the Complex I (40, 61, 63). This catalyzed Complex II oxidation of succinate is an irreversible reaction, at least in mammalian mitochondria. Coupling of the SDH/Complex II to the thermodynamically irreversible electron transport reactions of the respiratory chain makes the TCA cycle also work irreversibly in the clockwise direction to ensure utilization of pyruvate and other intermediates as the source of hydrogen. However, the stoichiometric formation of succinate as a TCA cycle intermediate raises the question of how brain mitochondria protect themselves from excessive ROS generation when a neuronal cell is at rest and mitochondria become more energized?

Schagger et al. (49, 50), using a BN-PAGE method, and Bianchi et al. (7), using a flux control analysis, did not observe stable binding of Complex II to Complexes III and IV. However, other researchers, using different methods, isolated functional Complexes II and III (62). In submitochondrial particles prepared by sonication of the mitochondria, the activity of Complexes II and III can be readily assessed (58), although the SDH activity becomes almost totally inhibited (43). Schagger (49) suggested that Complex II and other FAD-dependent dehydrogenases can contact the smaller supercomplex comprising Complexes III and IV, which functions as a sink for electrons provided by the membrane pool of reduced CoQ.

We suggest further that because brain mitochondria do not have a storage pool of the TCA cycle intermediates (A. V. Panov, unpublished data), in active mitochondria the steady-state concentrations of the mitochondrial substrates, including succinate, must be also low. Thus unlike artificial in vitro experimental conditions, when brain mitochondria oxidize mM concentrations of succinate, the in vivo succinate concentrations must be low, and therefore the rates of reduction of the membrane pool of CoQ may be lower, compared with the rate of its oxidation by the smaller superstructure of Complexes II–IV. Therefore the succinate-driven backward electron flow could be prevented. In addition, brain mitochondria in vivo may never be in the highly energized state as in the in vitro state 4. Thus channeling of the substrate utilization by the respirasome increases the efficiency of mitochondria and prevents excessive production of ROS. In addition, our preliminary data have shown that there may be other means of suppression of the succinate-supported ROS generation that involve metabolic control over SDH activity. This suggestion requires further investigation.

Summary of species-specific differences in brain and liver mitochondria. Altogether, the data presented show that the observed differences in the Ca²⁺-induced mPT and ROS generation may underlie the organ and species-specific differences observed in animal models of neurodegenerative diseases. Interestingly, short-living liver cells and long-living neuronal cells have different strategies towards the Ca²⁺-induced mPT and ROS generation. Liver cells have strong antioxidant defense against ROS that were formed in the mitochondria (16). Whole liver mitochondria release very little O₂^– and H₂O₂ (40). Because liver mitochondria have much lower Ca²⁺ capacities, they evidently undergo mPT, followed by apoptosis much easier than brain mitochondria. Elimination of damaged hepatocytes and replacement with the new healthy cells is evidently the reason why during systemic intoxication of animals with a toxin the liver mitochondria remain almost normal (40). This probably reflects the high regenerative potential of the liver. Brain mitochondria, on the other hand, have "chosen" a strategy to prevent generation of ROS. In the brain, which has negligible regenerative potential and a very long life span, mitochondria can withstand enormous Ca²⁺ insults (see Fig. 5, A and B). Therefore, we suggest that interactions between ROS generation and mechanisms of CaP_i sequestration determine responses to pathological situations in animal models of neurodegenerative diseases.

However, the relationships between these functions require further investigation. Mouse brain mitochondria, for example, generate 50% more ROS with the NAD-dependent substrates than RBM, but required 70% more Ca²⁺ to undergo mPT. Quite unexpected was the insensitivity of MLM to t-BOOH, which dramatically decreases the amount of Ca²⁺ that cause mPT in RLM. Evidently, other strains of mice and rats may have different parameters of intrinsic mitochondrial functions. Further quantitative studies may help to reveal the subtle functional mechanisms that determine the animal's sensitivity to a pathogenic factor when modeling a neurodegenerative disease.

	GRANTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
This work was supported by National Institute of Environmental Health Sciences Grant 12068 and a grant from the Picower Foundation.

	FOOTNOTES

 

Address for reprint requests and other correspondence: A. V. Panov, Carolinas Neuromuscular/ALS-MDA Center, Carolinas Medical Center, 1000 Blythe Blvd., Charlotte, NC 28203 (e-mail: alexander.panov{at}carolinashealthcare.org)

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

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION GRANTS REFERENCES

 
1. Andreeva L, Crompton M. An ADP-sensitive cyclosporin-A-binding protein in rat liver mitochondria. Eur J Biochem 221: 261–268, 1994.[ISI][Medline]

2. Andreev A, Fiskum G. Calcium-induced release of mitochondrial cytochrome c by different mechanisms selective for brain versus liver. Cell Death Differ 6: 825–832, 1999.[CrossRef][ISI][Medline]

3. Berman SB, Watkins SC, Hastings TG. Quantitative biochemical and ultrastructural comparison of mitochondrial permeability transition in isolated brain and liver mitochondria: evidence for reduced sensitivity of brain mitochondria. Exp Neurol 164: 415–425, 2000.[CrossRef][ISI][Medline]

4. Bernardi P. Mitochondrial transport of cations: channels, exchangers, and permeability transition. Physiol Rev 79: 1127–1155, 1999.[Abstract/Free Full Text]

5. Betarbet R, Sherer TB, MacKenzie G, Garsia-Osuna M, Panov AV, Greenamyre JT. Chronic systemic pesticide exposure reproduces features of Parkinson's disease. Nat Neurosci 3: 1301–1306, 2000.[CrossRef][ISI][Medline]

6. Beutner G, Ruck A, Riede B, Brdiczka D. Complexes between porin, hexokinase, mitochondrial creatine kinase and adenylate translocator display properties of the permeability transition pore. Implication for regulation of permeability transition by the kinases. Biochim Biophys Acta 1368: 7–18, 1998.[Medline]

7. Bianchi C, Genova ML, Castelli GP, Lenaz G. The mitochondrial respiratory chain is partially organized in a supercomplex assembly. Kinetic evidence using flux control analysis. J Biol Chem 279: 36562–36569, 2004.[Abstract/Free Full Text]

8. Brand MD. The contribution of the leak of protons across the mitochondrial inner membrane to standard metabolic rate. J Theor Biol 145: 267–286, 1990.[CrossRef][ISI][Medline]

9. Brouillet E, Conde F, Beal MF, Hantraye P. Replicating Huntington's disease phenotype in experimental animals. Prog Neurobiol 59: 427–468, 1999.[CrossRef][ISI][Medline]

10. Brustovetsky N, Brustovetsky T, Jemmerson R, Dubinsky JM. Calcium-induced cytochrome c release from CNS mitochondria is associated with the permeability transition and rupture of the outer membrane. J Neurochem 80: 207–218, 2002.[CrossRef][ISI][Medline]

11. Brustovetsky N, Dubinsky JM. Limitations of cyclosporin A inhibition of the permeability transition in CNS mitochondria. J Neurosci 20: 8229–8237, 2000.[Abstract/Free Full Text]

12. Byrne AM, Lemasters JJ, Nieminen AL. Contribution of increased mitochondrial free Ca²⁺ to the mitochondrial permeability transition induced by tert-butylhydroperoxide in rat hepatocytes. Hepatology 29: 1523–1531, 1999.[CrossRef][ISI][Medline]

13. Cadenas E, Boveris A, Ragan CI, Stoppani AOM. Production of superoxide radicals and hydrogen peroxide by NADH-ubiquinone reductase and ubiquinol-cytochrome c reductase from beef-heart mitochondria. Arch Biochem Biophys 180: 248–257, 1977.[CrossRef][ISI][Medline]

14. Castilho RF, Kowaltowski AJ, Meinicke AR, Bechara EJH, Vercesi AE. Permeabilization of the inner mitochondrial membrane by Ca²⁺ ions is timulated by t-butyl hydroperoxyde and mediated by recative oxygen species generated by mitochondria. Free Radic Biol Med 18: 479–486, 1995.[CrossRef][ISI][Medline]

15. Chalmers S, Nicholls DG. The relationship between free and total calcium concentrations in the matrix of liver and brain mitochondria. J Biol Chem 278: 19062–19070, 2003.[Abstract/Free Full Text]

16. Chance B, Sies H, Boveris A. Hydroperoxide metabolism in mammalian organs. Physiol Rev 59: 527–605, 1979.[Free Full Text]

17. Chemical Rubber Company. Handbook of Chemistry and Physics, edited by Weast RC and Selby SM (48th ed.). city?: Chemical Rubber, 1967.

18. Crompton M. The mitochondrial permeability transition pore and its role in cell death. Biochem J 341: 233–249, 1999.

19. Deitmer JW. Strategies for metabolic exchange between glial cells and neurons. Respir Physiol 129: 71–81, 2001.[CrossRef][ISI][Medline]

20. Fujii M, Hara H, Meng WM, Vonsattel JP, Huang Z, Moskowitz MA. Strain-related differences in susceptibility to transitory brain ischemia in SV-129 and C57Bl.6 mice. Stroke 28: 1805–1811, 1997.[Abstract/Free Full Text]

21. Gostimskaya IS, Grivennikova VG, Zharova TV, Bakeeva LE, Vinogradov AD. In situ assay of the intramitochondrial enzymes: use of alamethicin for permeabilization of mitochondria. Anal Biochem 313: 46–52, 2003.[CrossRef][ISI][Medline]

22. Green DR, Reed JC. Mitochondria and apoptosis. Science 281: 1309–1312, 1998.[Abstract/Free Full Text]

23. Hatefi Y. The mitochondrial electron transport and oxidative phosphorylation system. Annu Rev Biochem 54: 1015–1069, 1985.[CrossRef][ISI][Medline]

24. Halestrap HA. Pore way to die. Nature 434: 578–579, 2005.[CrossRef][Medline]

25. James AM, Smith RAJ, Murphy MP. Antioxidant and prooxidant properties of mitochondrial coenzyme Q. Arch Biochem Biophys 423: 47–56, 2004.[CrossRef][ISI][Medline]

26. Kim JS, He L, Lemasters JJ. Mitochondrial permeability transition: a common pathway to necrosis and apoptosis. Biochem Biophys Res Commun 304: 4633–4470, 2003.

27. Kobayashi T, Kuroda S, Tada M, Houkin K, Iwasaki Y, Abe H. Calcium-induced mitochondrial swelling and cytochrome c release in the brain: its biochemical characteristics and implication in ischemic neuronal injury. Brain Res 960: 62–70, 2003.[CrossRef][ISI][Medline]

28. Kristal BS, Dubinsky JM. Mitochondrial permeability transition in the central nervous system: induction by calcium cycling-dependent and -independent pathways. J Neurochem 69: 524–538, 1997.[ISI][Medline]

29. Kushnareva Yu Murphy AN, Andreev A. Complex I reactive oxygen species generation: modulation by cytochrome c and NAD(P)⁺ oxidation-reduction state. Biochem J 368: 545–553, 2000.

30. Lacza Z, Horn TFW, Snipes JA, Zhang J, Roychowdhury S, Horvarth EM, Figueroa JP, Kollai M, Szabo C, Busija DW. Lack of mitochondrial nitric oxide production in the mouse brain. J Neurochem 90: 942–951, 2004.[CrossRef][ISI][Medline]

31. Lemasters JJ, Nieminen AL, Qian T, Trost LC, Elmore SP, Nishimura Y, Crowe RA, Cascio WE, Bradham CA, Brenner DA, Herman B. The mitochondrial permeability transition in cell death: a common mechanism in necrosis, apoptosis and autophagy. Biochim Biophys Acta 1366: 177–196, 1998.[Medline]

32. Maciel EN, Vercesi AE, Castilho RF. Oxidative stress in Ca²⁺-induced membrane permeability transition in brain mitochondria. J Neurochem 79: 1237–1245, 2001.[CrossRef][ISI][Medline]

33. Melov S. Therapeutics against mitochondrial oxidative stress in animal models of aging. Ann NY Acad Sci 959: 330–340, 2002.[Abstract/Free Full Text]

34. Melov S. Modeling mitochondrial function in aging neurons. Trends Neurosci 27: 601–606, 2004.[CrossRef][ISI][Medline]

35. Muller FL, Liu Y, Van Remmen H. Complex III releases superoxide to both sides of the inner mitochondrial membrane. J Biol Chem 279: 49064–49073, 2004.[Abstract/Free Full Text]

36. Nakagawa T, Shimizu S, Watanabe T, Yamaguchi O, Otsu K, Yamagata H, Inohara H, Kubo T, Tsujimoto Y. Cyclophilin D-dependent mitochondrial permeability transition regulates some necrotic but not apoptotic cell death. Nature 434: 652–658, 2005.[CrossRef][Medline]

37. Ohnishi ST, Ohnishi O, Muranaka S, Fujita H, Kimura H, Uemura K, Yoshida KI, Utsumi K. A possible site of superoxide generation in the complex I segment of rat heart mitochondria. J Bioenerg Biomembr 37: 1–15, 2005.[CrossRef][ISI][Medline]

38. Panov AV. Regulation of specific and nonspecific ion conductivities of mitochondria by adenine nucleotide translocator. Biochemistry (Mosc) 57: 325–332, 1994.

39. Panov A, Andreeva L, Greenamyre JT. Quantitative evaluation of the effects of mitochondrial permeability transition pore modifiers on accumulation of calcium phosphate: two modes of action of mPTP modifiers. Arch Biochem Biophys 424: 44–52, 2004.[CrossRef][ISI][Medline]

40. Panov A, Dikalov S, Shalbueva N, Taylor G, Sherer T, Greenamyre JT. Rotenone model of Parkinson's disease: Multiple brain mitochondria dysfunctions after short-term systemic rotenone intoxication. J Biol Chem 280: 42026–42035, 2005.[Abstract/Free Full Text]

41. Panov A, Filippova S, Lyakhovich V. Adenine nucleotide translocase as a site of regulation by ADP of the rat liver mitochondria permeability to H⁺ and K⁺ ions. Arch Biochem Biophys 199: 420–426, 1980.[CrossRef][ISI][Medline]

42. Panov A, Lund S, Greenamyre JT. Ca²⁺-induced permeability transition in human Lymphoblastoid cell mitochondria from normal and Huntington's disease individuals. Mol Cell Biochem 269: 143–152, 2005.[CrossRef][ISI][Medline]

43. Panov AV, Scaduto RC. Influence of calcium on NADH and succinate oxidation by rat heart submitochondrial particles. Arch Biochem Biophys 316: 815–820, 1995.[CrossRef][ISI][Medline]

44. Panov A, Scarpa A. Mg²⁺ control of respiration in isolated rat liver mitochondria. Biochemistry 35: 12849–12856, 1996.[CrossRef][Medline]

45. Pellerin L, Magistretti PJ. Neuroenergetics: calling upon astrocytes to satisfy hungry neurons. Neuroscience 10: 53–62, 2004.

46. Puranam KL, Wu G, Strittmatter WJ, Burke JR. Polyglutamine expansion inhibits respiration by increasing reactive oxygen species in isolated mitochondria. Biochem Biophys Res Commun 341: 607–613, 2006.[CrossRef][ISI][Medline]

47. Robberecht W. Oxidative stress in amyotrophic lateral sclerosis. J Neurol 247, Suppl 1: I-6, 2000.

48. Santos JB, Schauwecker PE. Protection provided by cyclosporine A against excitotoxic neuronal death is genotype dependent. Epilepsia 44: 995–1002, 2003.[CrossRef][ISI][Medline]

49. Schagger H. Respiratory chain supercomplexes. IUBMB Life 52: 119–128, 2001.[ISI][Medline]

50. Schagger H, Pfeiffer K. Supercomplexes in the respiratory chains of yeast and mammalian mitochondria. EMBO J 19: 1777–1783, 2000.[CrossRef][ISI][Medline]

51. Scorrano L, Petronilli V, Bernardi P. On the voltage dependence of the mitochondrial permeability transition pore. A critical appraisal. J Biol Chem 272: 12295–12299, 1997.[Abstract/Free Full Text]

52. Shimizu S, Narita M, Tsujimoto Y. Bcl-2 family proteins regulate the release of apoptogenic cytochrome c by the mitochondrial channel VDAC. Nature 399: 483–487, 1999.[CrossRef][Medline]

53. Sims NR. Rapid isolation of metabolically active mitochondria from rat brain and subregions using Percoll density gradient centrifugation. J Neurochem 55: 698–707, 1990.[ISI]