Address correspondence and reprint requests to Dr. T. Kristián at Center for the Study of Neurological Disease, The Queen's Medical Center, 1356 Lusitana Street, UH Tower 8th Floor, Honolulu, HI 96813, U.S.A.
Abstract: The objective of the present study was to assess the capacity of nonsynaptic brain mitochondria to accumulate Ca2+ when subjected to repeated Ca2+ loads, and to explore under what conditions a mitochondrial permeability transition (MPT) pore is assembled. The effects of cyclosporin A (CsA) on Ca2+ accumulation and MPT pore assembly were compared with those obtained with ubiquinone 0 (Ub0), a quinone that is a stronger MPT blocker than CsA, when tested on muscle and liver mitochondria. When suspended in a solution containing phosphate (2 mM) and Mg2+ (1 mM), but no ATP or ADP, the brain mitochondria had a limited capacity to accumulate Ca2+ (210 nmol/mg of mitochondrial protein). Furthermore, when repeated Ca2+ pulses (40 nmol/mg of protein each) saturated the uptake system, the mitochondria failed to release the Ca2+ accumulated. However, in each instance, the first Ca2+ pulse was accompanied by a moderate release of Ca2+, a release that was not observed during the subsequent pulses. The initial release was accompanied by a relatively marked depolarization, and by swelling, as assessed by light-scattering measurements. However, as the swelling was <50% of that observed following addition of alamethicin, it is concluded that the first Ca2+ pulse gives rise to an MPT in a subfraction of the mitochondrial population. CsA, an avid blocker of the MPT pore, only marginally increased the Ca2+-sequestrating capacity of the mitochondria. However, CsA eliminated the Ca2+ release accompanying the first Ca2+ pulse. The effects of CsA were shared by Ub0, but when the concentration of Ub0 exceeded 20 μM, it proved toxic. The results thus suggest that brain mitochondria are different from those derived from a variety of other sources. The major difference is that a fraction of the brain mitochondria, studied presently, depolarized and showed signs of an MPT. This fraction, but not the remaining ones, contributed to the chemically and electron microscopically verified mitochondrial swelling.
In addition, under adverse conditions, such as gross mitochondrial Ca2+ accumulation or oxidation of thiol groups, a mitochondrial permeability transition (MPT) pore may be formed that is permeable to molecules with a mass of <1,500 Da. As a result, the electrochemical H+ gradient (which is essential for oxidative phosphorylation) is dissipated, and mitochondrial swelling occurs due to entry of water from the cytosol (Szabo and Zoratti, 1991; Bernardi et al., 1994, 1999; Gunter and Gunter, 1994; Bernardi and Petronilli, 1996). Evidence has been obtained that an MPT occurs during reperfusion of cardiac muscle and liver cells, and that this MPT is a major factor causing reperfusion damage in these tissues (Crompton and Costi, 1990; Griffiths and Halestrap, 1995; Nieminen et al., 1995).
Most of the in vitro data on the assembly of an MPT pore have been obtained by using mitochondria isolated from tissues such as heart, liver, and skeletal muscle, and it has even been questioned if an MPT can be elicited in brain mitochondria (Kristal and Dubinsky, 1997; Berman et al., 1998). However, recent studies give support to the hypothesis that, at least under some circumstances, an MPT pore is also assembled in brain mitochondria. Kristal and Dubinsky (1997) found that isolated nonsynaptic brain mitochondria underwent swelling when exposed to Ca2+ or Pi, and similar results were obtained when cultured astrocytes were exposed to Ca2+ loads. However, in both situations, CsA was only partially protective. In another study (Andreyev et al., 1998), a mixed population of brain mitochondria (both synaptic and nonsynaptic) were found to undergo an MPT, but only in the absence of ATP and Mg2+, i.e., under conditions in which only small amounts of Ca2+ can be accumulated by the mitochondria.
To elucidate further the mechanisms of MPT formation in brain tissue, we examined the ability of isolated nonsynaptic brain mitochondria to accumulate repeated Ca2+ loads, and the ability of these Ca2+ loads to elicit mitochondrial swelling by measuring Calcium Green-5N fluorescence and light scattering, respectively. The results were compared with those obtained on muscle mitochondria. In addition, to support our spectrophotometric studies, we examined mitochondrial swelling by electron microscopy. Furthermore, we studied the effects of not only CsA, but also ubiquinone 0 (Ub0) on Ca2+ uptake by brain nonsynaptosomal mitochondria and on MPT pore opening. The latter was done because Ub0 has been shown to increase markedly the capacity of mitochondria from other sources to accumulate Ca2+ and to block the Ca2+-triggered MPT (Fontaine et al., 1998a, b).
MATERIALS AND METHODS
Isolation of nonsynaptic brain mitochondria
We isolated nonsynaptic brain mitochondria from 300-g male Wistar rats (Simonsen Laboratories Inc., Gilroy, CA, U.S.A.) using a modification of the method of Sims (1990). Following decapitation, brains were rapidly removed and placed in ice-cold isolation buffer containing 320 mM sucrose, 1 mM K2EDTA, and 10 mM Tris (pH 7.4). All subsequent homogenization and centrifugation steps were carried out at 4°C. The cerebellum and underlying structures were removed, and the remaining brain tissue was used. Each hemisphere was minced separately with scissors in 3 ml of isolation buffer and homogenized by hand in a Potter-Elvehjem homogenizer. The resulting homogenate was centrifuged at 1,330 gav for 3 min. The supernatant was decanted, and the pellet was resuspended in half of the original volume and recentrifuged as above. The pooled supernatant was centrifuged at 21,200 gav for 10 min. The supernatant from this centrifugation step was decanted, and the pellet obtained was resuspended in 15% Percoll (3 ml) and layered into centrifuge tubes containing a preformed two-step discontinuous density gradient consisting of 3.5 ml of 23% Percoll on top of 2.5 ml of 40% Percoll. The gradients were centrifuged at 30,700 gav for 10 min. The mitochondrial fraction, located at the interface between the bottom two layers, was removed and diluted 1:4 in isolation buffer and recentrifuged at 16,700 gav for 10 min. The supernatant was decanted and the pellet resuspended in buffer containing 320 mM sucrose, 0.1 mM K2EDTA, and 10 mM Tris-HCl (pH 7.4). Aliquots were removed for protein determinations, then bovine serum albumin (10 mg/ml) was added to the mitochondria, and the suspension was centrifuged at 6,900 gav. The supernatant was removed, and the mitochondrial pellet was then gently resuspended in an equal volume of buffer used for mitochondrial Ca2+ uptake experiments (see below) and stored on ice.
Isolation of heart mitochondria
After decapitation, hearts were rapidly removed and placed in ice-cold isolation buffer containing 210 mM mannitol, 70 mM sucrose, 1 mM EDTA, 50 mM Tris (pH 7.4). The tissue was finely minced with scissors and then homogenized in 10 ml of isolation buffer using a Tissuemizer motorized homogenizer (Tekmar Co., Cincinnati, OH, U.S.A.). The homogenate was centrifuged at 1,300 gav for 3 min. The supernatant was decanted, and the pellet was resuspended in 5 ml of isolation buffer and centrifuged as above. The pooled supernatant was centrifuged at 10,000 gav for 10 min. The resulting supernatant was removed together with the top layer of the pellet (which contains “light” mitochondria and/or damaged mitochondria), and the dark-brown “dense” mitochondrial pellet was resuspended in isolation buffer with a lower EDTA concentration (0.1 mM EDTA). After aliquots were removed for protein measurements, the mitochondria were centrifuged at 6,800 gav for 10 min. The mitochondrial pellet obtained was then gently resuspended in buffer and stored on ice prior to the Ca2+ uptake experiments.
Mitochondrial protein measurements
Mitochondrial protein concentrations were measured using a Lowry DC kit (Bio-Rad) with bovine serum albumin used as a concentration standard.
Mitochondrial oxygen consumption
The measurement of oxygen consumption rates was performed using a custom-made incubation chamber with a water jacket and a micro-Clark electrode (Yellow Springs Instrument Co., Yellow Springs, OH, U.S.A.) fitted into the top of the chamber. The incubation medium was stirred constantly using an electromagnetic stirrer and magnetic stirring bar. The oxygen consumption studies were conducted at 37°C in respiration medium consisting of 100 mM KCl, 75 mM mannitol, 25 mM sucrose, 5 mM phosphate-Tris, 10 mM Tris-HCl, and 50 μM EDTA (pH 7.4). Approximately 0.2 mg of mitochondrial protein was added into an incubation chamber in a total volume of 400 μl of respiration medium. The concentrations of substrates used were 5 mM glutamate and 5 mM malate. State 3 respiration was induced by the addition of ADP, and the respiratory control ratio was calculated from the ratio of the state 3/state 4 oxygen consumption rates, with and without ADP, respectively (Chance and Williams, 1956).
Measurement of mitochondrial Ca2+ uptake
Extramitochondrial free Ca2+ was measured according to the methods described by Murphy et al. (1996). Isolated mitochondria (0.5 mg of protein) were resuspended in 2 ml of buffer containing 125 mM KCl, 2 mM K2HPO4, 1 mM MgCl2, 1 μM EGTA, 20 mM Tris (pH 7.2 at 37°C), 5 mM glutamate, and 5 mM malate. Free Ca2+ was monitored by use of a hexapotassium salt of Calcium Green-5N (0.1 μM). Fluorescence was recorded continuously in a water-jacketed cuvette holder at 37°C using a Perkin-Elmer LS-50B fluorescence spectrometer with excitation and emission wavelengths of 506 and 532 nm, respectively. At the end of each experiment, calibrations were performed to establish (a) a zero (by adding 20 μl of 0.1 M EGTA) and (b) a saturated Ca2+ level (by adding 20 μl of 1 M CaCl2). Free Ca2+ concentration was calculated from the following equation:
where F is the fluorescence of the indicator at experimental Ca2+ levels, Fmin is the fluorescence in the absence of Ca2+ (after adding EGTA), and Fmax is the fluorescence of the Ca2+-saturated probe (1 M CaCl2 added). A Calcium Green-5N KD for Ca2+ of 4.3 was used (Rajdev and Reynolds, 1993). To calculate the total amount of Ca2+ taken up by mitochondria, a calibration pulse was recorded by addition of a known quantity of CaCl2 to the medium with mitochondria where the Ca2+ uptake was prevented by 5 μM rotenone or 0.5 μM antimycin. As the buffer contains EGTA and phosphate, which can bind Ca2+, the ordinate axes on the figures are labeled as added nanomoles of Ca2+ per milligram of protein, which correspond to the fluorescence level associated with a known quantity of Ca2+ added to the medium, normalized to 1 mg of mitochondrial proteins.
Monitoring of the mitochondrial membrane potential
Changes in mitochondrial membrane potential were measured in the presence of 0.2 μM rhodamine (Rh) 123 as described by Emaus et al. (1986). The excitation and emission wavelengths for Rh 123 were 503 and 525 nm, respectively.
Measurement of mitochondrial swelling
Mitochondrial swelling was estimated from the changes of light scattering (at 90° to the incident light beam) at 540 nm (for both excitation and emission wavelengths) measured in mitochondrial suspensions (0.5 mg of protein in 2 ml) using the same buffer as for the mitochondrial Ca2+ experiments above. The experiments were performed in a water-jacketed cuvette holder at 37°C using a Perkin-Elmer LS-50B fluorescence spectrometer. Each experiment was terminated by the addition of alamethicin (40 μg/mg of protein) to induce a maximal swelling of the whole mitochondrial population. This was confirmed to be the case with the electron micrographs obtained (see below). In the figures, light intensity after alamethicin treatment represents 0, and the light intensity before Ca2+ addition is marked as 100 on the ordinate.
Fixation of mitochondria for electron microscopy
For electron microscopy, mitochondria were incubated under the specific conditions and centrifuged at 13,000 g for 5 min. The pellet obtained was fixed overnight with 4% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.4) at 4°C, postfixed with 1% osmic acid in 0.1 M cacodylate buffer, dehydrated, and embedded in LX-112 (Ladd) epoxy resin. Thin sections were observed with a Zeiss 10/A transmission electron microscope.
We will describe changes in the capacity of brain mitochondria to sequester Ca2+, as well as the accompanying alterations in membrane potential and in mitochondrial volume. The changes observed will be correlated to the electron micrograph appearance of Ca2+-loaded mitochondria. After that, we will consider the effects of CsA and Ub0. However, it seems reasonable that we begin by obtaining control data by illustrating how heart mitochondria respond to Ca2+ loading under conditions similar to those used in the experiments on brain mitochondria. The mitochondria used in this study were well coupled. The respiratory control ratios of the nonsynaptic brain mitochondria and heart mitochondria used were always ≥5 and ≥8, respectively. The experiments where qualitative changes are reported were repeated three or four times to confirm reproducibility.
Ca2+ loading of heart mitochondria
Figure 1 shows that heart mitochondria buffered two consecutive Ca2+ loads (40 nmol of Ca2+/mg of mitochondrial protein) before they had been saturated and released the Ca2+ accumulated, probably because an MPT pore was opened. Measurement of light scattering did not suggest that an MPT pore opened after a single Ca2+ addition (curve a). However, when the Ca2+ dose was increased to 100 nmol of Ca2+/mg of protein, progressive swelling occurred (curve b). By increasing the added Ca2+ to 400 nmol of Ca2+/mg of protein, the swelling rate increased further (curve c). The results thus suggest that both the swelling and the release of Ca2+ reflect the opening of an MPT pore. This contention was also supported by electron micrographs (Fig. 2), which show that following the addition of 400 nmol of Ca2+/mg of protein, the majority of the heart mitochondria showed high amplitude swelling with damaged cristae.
As Fig. 3 (upper panel) shows, brain mitochondria had a seemingly better capacity to sequester repeated Ca2+ loads than heart muscle mitochondria. There were two additional features that the brain mitochondrial preparations displayed. First, the initial Ca2+ uptake was followed by a small release; this is illustrated more clearly in the middle panel of Fig. 3. Second, when the mitochondria were saturated, they did not release the Ca2+ accumulated. However, when carbonyl cyanide m-chlorophenylhydrazone (CCCP; 3 μM) was added to the buffer, a release of Ca2+ was observed, indicating that mitochondria are still loaded with the Ca2+.
Mitochondrial Ca2+ accumulation was blocked by ruthenium red (RR) (Fig. 3, lower panel). Thus, when RR was present in the buffer, Ca2+ added to the incubation medium greatly increased the initial Ca2+ concentration in the buffer, as no uptake was observed. As RR is a relative specific inhibitor of mitochondrial Ca2+ uptake, these data suggest that the added Ca2+ was accumulated by the mitochondria.
Changes in mitochondrial membrane potential, as measured by Rh 123 fluorescence, are shown in Fig. 4. The upper part of the figure illustrates an experiment in which Ca2+ was added at 3-min intervals. Addition of Ca2+ caused a rapid, but small, depolarization, which was followed by a gradual loss of membrane potential. Additional Ca2+ pulses resulted in progressive depolarization until the mitochondrial membrane potential reached the residual Donnan potential. This was confirmed at the end of experiments by the induction of complete depolarization with 3 μM CCCP.
The lower panel of Fig. 4 illustrates an experiment in which 10 min was allowed to pass before a second addition of Ca2+ was made. The results demonstrate that the first addition caused a substantial loss of membrane potential (cf. effect of CCCP).
The results illustrated in Fig. 4 should be compared with those of Fig. 5, which shows the effect of a single and of repeated additions of Ca2+ on the light scattering measured at 540 nm. A decrease in light scattering is believed to reflect mitochondrial swelling caused by an MPT. The results demonstrate that most of the swelling observed after repeated Ca2+ additions was the effect of the first addition, i.e., the one that caused some Ca2+ release and that substantially depolarized the membranes.
As alamethicin caused at least twice as much change in light scattering as Ca2+ additions, the combined results suggest that the first Ca2+ wave is associated with an MPT in a subpopulation of the mitochondria that leads to limited mitochondrial depolarization and to partial swelling.
To shed further light on this issue, mitochondrial preparations were examined in the electron microscope before and after Ca2+ addition. As shown in Fig. 6, Ca2+ caused swelling of part of the mitochondrial population. This was in contrast to alamethicin, which triggered swelling of the whole mitochondrial population. As the swelling was similar whether one or several Ca2+ additions were made, the results support our conclusion that the first Ca2+ pulse triggers an MPT. Additional support is provided by the effect of CsA (see below).
The immunosuppressant drug CsA is reported to have two effects on isolated mitochondria from tissue other than the brain: it is a virtually specific blocker of the MPT, and it increases the capacity of the mitochondria to sequester Ca2+ (for reviews, see Bernardi et al., 1992; Duchen et al., 1993; Murphy et al., 1996). Recently, Ub0 has been shown to have a similar effect on mitochondria from liver and muscle tissue (Fontaine et al., 1998a, b).
Figure 7 shows that CsA had little effect on Ca2+ accumulation by the mitochondria; in fact, CsA increased the capacity of the mitochondria to sequester Ca2+ by only 10% (from ∼ 210 to ∼ 230 nmol of Ca2+/mg of protein). However, CsA prevented the moderate release of Ca2+ following the first Ca2+ pulse. When added at a dose of 5 μM, Ub0 had very similar effects. When the concentration was increased to 20 and 40 μM, however, the Ca2+ sequestration capacity was progressively decreased, and when the uptake system was saturated, Ca2+ was released from the mitochondria (see Fig. 8).
Both CsA and Ub0 stabilized the membrane potential following repetitive additions of Ca2+ (Fig. 9). Thus, in the presence of CsA, the first three Ca2+ pulses induced a small decrease in membrane potential, which quickly recovered. Further Ca2+ additions then caused progressive depolarization until finally the mitochondria were fully depolarized, as confirmed by the addition of CCCP.
As Fig. 10 shows, both CsA and Ub0 reduced Ca2+-induced swelling in a dose-dependent manner and, at optimal concentrations, the swelling was prevented completely. These concentrations were ∼ 1 μM for CsA and 20-40 μM for Ub0. When swelling had been prevented (or minimized), it appeared (or was augmented) when additional Ca2+ pulses were delivered (data not shown).
As it was suggested that Ub0 affects the electron flow via complex I, thereby modulating the Ca2+ retention capacity and sensitivity of mitochondria to Ca2+-induced MPT (Fontaine et al., 1998a, b), we also measured the Ca2+ uptake by nonsynaptic brain mitochondria that were energized with the complex II-linked substrate succinate. With succinate as substrate, the mitochondrial Ca2+ accumulation following Ca2+ additions was more rapid than with malate plus glutamate (see Fig. 11). However, after four Ca2+ pulses, there was a marked transient increase in extramitochondrial Ca2+ concentration, and further Ca2+ additions led to a gradual saturation of the uptake system (Fig. 11, upper panel). In the presence of Ub0, the retention capacity of mitochondria using succinate as substrate was also increased (see Fig. 11). Thus, Ub0 was effective when either complex I-linked or complex II-linked substrates were used for mitochondrial respiration.
In summary, CsA at a concentration of 1 μM had only a small effect on the Ca2+ sequestration capacity of brain mitochondria, but prevented the partial release of Ca2+ accompanying the first Ca2+ pulse, reversed the small depolarization of mitochondrial membranes following the first two or three Ca2+ pulses, and prevented the mitochondrial swelling that was elicited by a single Ca2+ pulse. Ub0 had, in general, similar effects, and we could not identify any qualitative or quantitative differences between CsA and Ub0. It seems clear, however; that Ub0 at high concentrations was toxic because it reduced the Ca2+ sequestration capacity of the mitochondria and caused massive release of Ca2+ when the mitochondria were saturated with Ca2+. We recognize that when this occurred, brain mitochondria behaved as those isolated from heart tissue.
Mechanisms of Ca2+ uptake by isolated mitochondria, or by mitochondria in isolated cells, have attracted considerable interest mostly because such uptake is considered a trigger of membrane depolarization and of the assembly of an MPT. The majority of such studies have been carried out on mitochondria from liver, heart, and skeletal muscle (Fontaine et al., 1998a, b); however, comparable studies have been performed on the neuronal cell line GT1-7 (Murphy et al., 1996). In all of these studies, the mitochondria could absorb a limited number of Ca2+ pulses before they were saturated with Ca2+. At that point, they released the Ca2+ accumulated, probably because an MPT pore was opened in the Ca2+-loaded mitochondria. Much interest has been devoted to the pore-blocking drug CsA and the similarly acting Ub0. Both of these drugs increase the capacity of the mitochondria to sequester Ca2+, probably because they reduce the mitochondrial depolarization caused by Ca2+ accumulation (see Results and Fontaine et al., 1998b). Similar results on CsA were reported by Murphy et al. (1996) on GT1-7 cells. These authors also studied cells overexpressing Bcl-2 and found that Bcl-2 expression potentiated the capacity of mitochondria to sequester Ca2+.
As stated in the introductory section, several groups have found that brain mitochondria are less prone to assemble an MPT, and it has been questioned if an MPT acts as an important modulator of ischemic cell death in the brain (Kristal and Dubinsky, 1997; Andreyev et al., 1998; Berman et al., 1998). However, it has remained a challenge that CsA, when allowed to pass the blood-brain barrier, is markedly neuroprotective. This is what inspired us to study in some detail how a nonsynaptosomal population of brain mitochondria responds to repeated Ca2+ pulses.
The primary objective of this study was to explore how nonsynaptosomal brain mitochondria differ from those derived from other tissues, such as liver and heart, when subjected in vitro to Ca2+ loads that normally would be expected to lead to an MPT, and to cause release of Ca2+ from Ca2+-loaded mitochondria. An equally important objective was to assess how CsA, an established antiischemic drug, affects Ca2+ accumulation of isolated mitochondria and the assembly of an MPT pore. A secondary objective was to establish if Ub0 mimics the effect of CsA, or if it acts by different mechanisms. Another, secondary objective was to explore whether energization by a complex II substrate (succinate) would give results similar to those obtained with glutamate plus malate.
Our results proved that brain mitochondria behave in a radically different way from liver, heart, and skeletal muscle mitochondria, studied by others, and heart mitochondria, as examined in the present study. Thus, under the experimental conditions used in our present studies, following saturation of the Ca2+ uptake system, brain mitochondria did not respond with a large release of their sequestrated Ca2+. As stated, however, there was a small transient release of mitochondrial Ca2+ after the first Ca2+ pulse, suggesting that some part of the mitochondrial population underwent an MPT and that Ca2+ released from this subpopulation of mitochondria was then accumulated by other mitochondria within the incubation medium. It is of note, however, that this second population of mitochondria did not release the Ca2+ accumulated.
The effect of CsA on the ability of nonsynaptic brain mitochondria to accumulate Ca2+ was only moderate compared with the effects of CsA on liver or muscle mitochondria (Fontaine et al., 1998a, b). Similar effects of CsA were also reported by Murphy et al. (1996) on their GT1-7 cell line. However, our studies unequivocally showed that CsA prevented the partial efflux of Ca2+ that accompanied the first Ca2+ pulse, which depending on the CsA dose entirely prevented the Ca2+-induced swelling of the mitochondria, and which reversed the depolarizations during the first two or three Ca2+ pulses.
The changes in Rh 123 fluorescence cannot be readily interpreted in our study as depolarization or recovery of mitochondrial membrane potential. This is because of the heterogeneity of the nonsynaptic brain mitochondrial population. Following addition of 40 nmol of Ca2+/mg of protein, some of the mitochondria depolarized due to an MPT, but the rest of the population could still have maintained their membrane potential, which is reflected in their ability to accumulate additional Ca2+. Thus, a moderate increase in Rh 123 fluorescence intensity may reflect complete depolarization of a subfraction of the mitochondria, without significant changes in membrane potential in the rest of the mitochondrial population. Data on mitochondrial membrane potential suggest that the saturation of the Ca2+ uptake after several Ca2+ pulses is due to complete depolarization of mitochondria to the resting Donnan potential. The fact that no Ca2+ release was observed suggests that membrane integrity was preserved.
Our data, showing only incomplete swelling of the nonsynaptic mitochondrial population, together with the fact that the mitochondrial accumulation of Ca2+ after the first Ca2+ pulse was always preceded by a transient increase in Ca2+ in the incubation medium, suggest that there is a subpopulation of nonsynaptic mitochondria that undergo the MPT at lower Ca2+ concentrations than other nonsynaptic mitochondria. Nonsynaptic mitochondria are derived from a variety of cell types representing particularly neurons and glia. Thus, such a mitochondrial population may give rise to a heterogeneous response to Ca2+ exposure. This hypothesis is also supported by the fact that within the rat brain there are different subpopulations of nonsynaptic and synaptic mitochondria that have different oxygen consumption rates with NAD-linked and FAD-linked substrates, as well as different activities of primary dehydrogenase enzymes, such as pyruvate dehydrogenase and glutamate dehydrogenase (Lai and Clark, 1979). In addition, it is known that synaptic and nonsynaptic mitochondria have different activities of the respiratory chain complexes (Bates et al., 1994; Almeida et al., 1995), and that mitochondrial membrane potentials in a single cell can be heterogeneous (Smiley et al., 1991). Thus, these different mitochondrial populations may have different resting membrane potentials and depolarize to different extents when exposed to the same amount of Ca2+. This could explain why some mitochondria may be able to take up only a small amount of Ca2+ before they are overloaded, whereas other mitochondria have a much larger capacity for Ca2+ uptake.
The recent results reported by Fontaine et al. (1998a) raise the possibility that electron flow through mitochondrial complex I is a major modulator of MPT pore opening, and that this is the site where Ub0 acts. One could argue that if Ub0 only affects electron flow via complex I, it should have no effect on mitochondria energized with a complex II substrate. However, in our experiments, mitochondria incubated in the presence of succinate and Ub0 were relatively resistant to Ca2+ challenges and, at least at a concentration of 20 μM Ub0, the partial release of Ca2+ after repeated pulses was blocked. This suggests that Ub0 suppresses the MPT induced by Ca2+ in succinate-respiring mitochondria.
These data thus suggest that nonsynaptic brain mitochondria behave differently from mitochondria from liver or muscle, particularly with respect to MPT pore opening, and also support the concept of mitochondrial heterogeneity with respect to Ca2+ accumulation and MPT pore opening. The data also show that although CsA can increase the capacity of brain mitochondria to accumulate Ca2+ and to resist Ca2+-induced mitochondrial swelling, the concentration of CsA that the mitochondria are exposed to is critical. The data obtained using succinate and Ub0 suggest that electron flow via respiratory complexes has a controlling/modulatory role in MPT formation in nonsynaptic brain mitochondria, as well as in liver and skeletal muscle mitochondria (Fontaine et al., 1998a, b).
The studies of Andreyev et al. (1998) were carried out on a mixture of synaptic and nonsynaptic mitochondria. They reported the following results. First, when mitochondria took up Ca2+, they released up to 40% of their cytochrome c content in a cyclosporin-insensitive manner. When ATP and Mg2+ were present, this process was not accompanied by mitochondrial swelling, as assessed by absorbance measurements. The conclusions drawn were that cytochrome c released by brain mitochondria does not require an MPT; however, in the absence of ATP and Mg2+, Ca2+ triggers a CsA-sensitive high-amplitude swelling and release of matrix solutes. The results reported again raise the question whether an MPT occurs in vivo. It seems particularly justified to examine mitochondrial behavior in the presence of Mg2+ and phosphate, without and with ATP or ADP, because these are the conditions prevailing in ischemic and postischemic tissues, respectively.