Mitochondrial Depolarization Accompanies Cytochrome cRelease During Apoptosis in PC6 Cells*

Cytochrome c is released from mitochondria into the cytosol in cells undergoing apoptosis. The temporal relationship between cytochrome c release and loss of mitochondrial membrane potential was monitored by laser-scanning confocal microscopy in single living pheochromocytoma-6 cells undergoing apoptosis induced by staurosporine. Mitochondrial membrane potential monitored by tetramethylrhodamine methyl ester decreased abruptly in individual cells from 2 to 7 h after treatment with staurosporine. Depolarization was accompanied by cytochromec release documented by release of transfected green fluorescent protein-tagged cytochrome c in these cells. The results show that mitochondrial depolarization accompanies cytochromec release in pheochromocytoma-6 cells undergoing apoptosis.

Cytochrome c is released from mitochondria into the cytosol in cells undergoing apoptosis. The temporal relationship between cytochrome c release and loss of mitochondrial membrane potential was monitored by laser-scanning confocal microscopy in single living pheochromocytoma-6 cells undergoing apoptosis induced by staurosporine. Mitochondrial membrane potential monitored by tetramethylrhodamine methyl ester decreased abruptly in individual cells from 2 to 7 h after treatment with staurosporine. Depolarization was accompanied by cytochrome c release documented by release of transfected green fluorescent protein-tagged cytochrome c in these cells. The results show that mitochondrial depolarization accompanies cytochrome c release in pheochromocytoma-6 cells undergoing apoptosis.
Substantial evidence implicates mitochondria in apoptotic cell death (1)(2)(3)(4). It is thought that proteins normally restricted to the mitochondrial intermembrane space, including cytochrome c and the 50-kDa apoptosis-inducing factor, are released to the cytosol where they initiate the apoptotic cascade (5)(6)(7)(8). The mechansim of protein release has not been established. One proposal is that protein release requires rupture of the outer mitochondrial membrane and that this is a consequence of the onset of the mitochondrial permeability transition (MPT) 1 (9,10). Onset of the MPT, which is an inner membrane process, would depolarize the inner membrane (11). Thus, the MPT hypothesis implies a direct temporal relationship between mitochondrial depolarization and cytochrome c release. Experimental data addressing this connection are limited.
Actually, two recent studies showed that cytochrome c release during apoptosis was not accompanied by mitochondrial depolarization (12,13). In these studies, mitochondrial membrane potential (⌬) was monitored by rhodamine 123 and DiOC 6 , compounds that are known to inhibit the ATP synthase and mitochondrial respiration (14 -16). In addition, cells respond asynchronously to toxic insults (17)(18)(19), and assessment of sequential events that occur rapidly becomes difficult when a population of cells rather than individual cells is studied.
To examine the temporal relationship between cytochrome c release and mitochondrial depolarization, we set out to monitor these events in the individual mitochondria of single cells undergoing apoptosis. To monitor mitochondrial ⌬, we used tetramethylrhodamine methyl ester (TMRM). TMRM has no apparent effect on metabolism, and it distributes into mitochondria in response to membrane potential without covalent binding (20). To monitor cytochrome c, we made a cytochrome c-green fluorescent protein (GFP) fusion construct for transfection so that release of cytochrome c could be observed in individual mitochondria by confocal microscopy (21). In the present report we show that mitochondrial depolarization is accompanied by cytochrome c release in staurosporine-induced apoptosis.

EXPERIMENTAL PROCEDURES
Plasmid Preparation-Full-length cDNA sequence of rat cytochrome c was synthesized by polymerase chain reaction amplification using Pfu polymerase (Stratagene, La Jolla, CA), and cloned into NheI and XhoI sites of the pEGFP-N1 plasmid (CLONTECH Laboratories, Inc., Palo Alto, CA). Polymerase chain reaction was performed using oligonucleotide primers designed to introduce the KOZAK sequence and remove the stop codon of cytochrome c. The KOZAK sequence of the GFP in pEGFP-N1 was removed to ensure expression of only the cytochrome c-GFP fusion protein.
Cell Line, Culture Conditions, and Transfections-Rat pheochromocytoma-6 (PC6) cells were cultured in Dulbecco's modified Eagle's medium/F12 medium supplemented with 15 mM HEPES, 4 mM glutamine, 17.5 mM glucose, heat-inactivated 10% fetal bovine serum, 5% horse serum, and penicillin/streptomycin. For transfection, cells (4 ϫ 10 5 ) were plated on poly-D-lysine-coated 40-mm glass coverslips. One day later, cells were transiently transfected using 4 g of plasmid DNA (cytochrome c-GFP or GFP) and 24 l of cationic lipid Lipofect-AMINE TM (Life Technologies, Inc.) in 2 ml of serum free Opti-MEM medium (Life Technologies, Inc.) per coverslip. After 3 h, an equal volume of regular medium containing 2ϫ normal serum was added. After 1 day, the cells were placed in normal growth medium without antibiotics.
Western Blot Analysis-Immunoblots were performed as described before (21,22). PC6 cells were harvested, washed twice with ice-cold phosphate-buffered saline, and lysed with ice-cold lysis buffer (150 mM NaCl, 1 mM EGTA, 1% Triton X-100, 1% sodium deoxycholate, 50 mM Tris-HCl, pH 8.0) containing protease inhibitors (0.2 g/ml pepstatin, 0.2 g/ml leupeptin, 1 g/ml aprotinin, and 0.2 mM phenylmethylsulfonyl fluoride). The cell lysates were centrifuged at 10,000 ϫ g for 10 min at 4°C. Supernatants were mixed with Laemmli sample buffer. Forty g of protein each of nontransfected cells and cytochrome c-GFP transfected cells, and 1 g of protein of GFP were separated on a 12% SDS-polyacrylamide gel electrophoresis gel. The proteins were transferred to a polyvinylidene difluoride membrane (Bio-Rad) and blotted with primary monoclonal anti-GFP antibody (CLONTECH) and secondary anti-mouse horseradish peroxidase antibody (Santa Cruz Biotechnologies, Santa Cruz, CA). Peroxidase activity was developed with ECL (Amersham Pharmacia Biotech).
Confocal Microscopy-To investigate localization of GFP and GFPfusion proteins, cells were transferred to a mounting chamber on the microscope stage in Dulbecco's modified Eagle's medium/F12 medium supplemented with 15 mM HEPES and 5% fetal bovine serum at pH 7.2 after 66 h of transfection. Subsequently, cells were loaded with 40 nM MitoTracker Red CMXRos (23) (Molecular Probes Inc., Eugene, OR) for 15 min in culture medium. A 63X N.A. 1.4 oil immersion planapochromat objective with a Zeiss 410 confocal microscope (Thornwood, NY) was used for all experiments. Confocal images of green cytochrome c-GFP and GFP fluorescence were collected using 488 nm excitation light from an argon/krypton laser, a 560-nm dichroic mirror and a 500 -550-nm band pass barrier filter. Images of red MitoTrackerRed CMXRos fluorescence were collected using a 568-nm excitation light from the argon/krypton laser, a 560-nm dichroic mirror, and a 590-nm long pass filter.
To measure mitochondrial membrane potential, cytochrome c-GFP transfected PC6 cells were loaded with 150 nM TMRM for 15 min in culture medium (24). Subsequently, cells were washed twice with fresh medium and transferred to a mounting chamber on the microscope stage in medium containing 50 nM TMRM. Temperature of the chamber was maintained at 35-37°C with a heat controller (Bioptechs, Butler, PA). Images of green cytochrome c-GFP fluorescence were collected as described above and red TMRM fluorescence was imaged using same filter settings as for MitoTrackerRed CMXRos. For measurements of cytochrome c-GFP and TMRM fluorescence, laser power was attenuated with neutral density filters at least 90% and 99.9%, respectively. In addition, the images were collected at 1 h intervals at the lowest zoom setting. These precautions were taken to avoid photodamage caused by intense laser illumination. After collecting basal images (0 h), 5 M staurosporine was added. Subsequently, images were collected at 2, 4, and 5 h. For the assessment of apoptotic cell nuclei, cytochrome c-GFP-transfected cells were stained with 500 nM SYTO-13 and imaged using the same filter settings as for cytochrome c-GFP.

RESULTS AND DISCUSSION
Rat PC6 cells were transiently transfected with plasmids bearing the cytochrome c-GFP fusion construct. To confirm that cytochrome c-GFP was targeted to mitochondria, cells were co-loaded with Mitotracker Red CMXRos, which localizes to mitochondria (23). Cytochrome c-GFP and Mitotracker Red CMXRos were imaged using laser scanning confocal microscopy. As shown in Fig. 1A, cytochrome c-GFP displayed a punctate pattern of fluorescence that was indistinguishable from that of MitoTracker Red. These confocal images demonstrated that the cytochrome c-GFP fusion protein was localized to mitochondria (Fig. 1, A and AЈ). By contrast, cells transfected with GFP cDNA alone displayed a diffuse fluorescence (Fig.  1B). The expression of the cytochrome c-GFP fusion protein was also confirmed by Western blot analysis using anti-GFP antibody (21, 22) (Fig. 1C).
To monitor changes in mitochondrial ⌬, cytochrome c-GFP- transfected cells were loaded with TMRM. TMRM is a cationic fluorophore that accumulates electrophoretically into mitochondria in response to the negative mitochondrial ⌬ (20). When mitochondria depolarize, they release TMRM and the red fluorescence inside mitochondria disappears. Accordingly, the green fluorescence of cytochrome c-GFP and the red fluorescence of TMRM were monitored simultaneously in single living cells by laser scanning confocal microscopy. Fig. 2A shows a field of PC6 cells transfected with cytochrome c-GFP (right panel) and loaded with TMRM (left panel). Before addition of staurosporine to induce apoptosis, all cells in the field displayed high mitochondrial ⌬, as indicated by bright punctate fluorescence of TMRM ( Fig. 2A, 0 h, left panel). Not all cells were transfected with cytochrome c-GFP, as indicated by the absence of green fluorescence, but all viable cells had polarized mitochondria as revealed by red fluorescence (Fig. 2A, 0 h, left  panel).
To induce apoptosis, cells were treated with 5 M staurospo- rine, a protein kinase inhibitor, which induces apoptosis in a wide variety of cell types (12,17,25,26). In the field of PC6 cells shown in Fig. 2A, TMRM fluorescence began to decrease in cell 1 after 2 h ( Fig. 2A, 2 h, left panel). The decrease of fluorescence was because of the disappearance of a portion of the bright fluorescent mitochondria indicating that some but not all mitochondria had depolarized and released their TMRM. Simultaneously, the pattern of cytochrome c-GFP fluorescence began to change from punctate to diffuse, indicating release of cytochrome c-GFP from the mitochondria ( Fig. 2A,  2 h, cell 1, right panel). As cytochrome c-GFP was released, green fluorescence increased diffusely in the cytosol and in the nucleus as well. At this early stage, cytochrome c-GFP fluorescence was not fully diffuse, and a punctate cytochrome c-GFP pattern remained in mitochondria that still retained TMRM fluorescence. After 3 h, however, all mitochondria depolarized, as indicated by the disappearance of virtually all TMRM fluorescence (data not shown). By this time cytochrome c-GFP fluorescence disappeared virtually completely from cell 1, which likely indicated onset of the secondary necrosis that often follows apoptosis. At the onset of cell necrosis, the plasma membrane ruptures and soluble cytosolic proteins, such as cytochrome c-GFP, leak out. However, apoptosis was the primary mode of cell death in the cells studied, as indicated by nuclear shrinkage, chromatin condensation, and formation of apoptotic bodies assessed by the nuclear stain SYTO-13 (Fig.  2B, Stauro).
Similarly, in cell 2 the number of TMRM-labeled mitochondria began to decrease after 2 h of exposure to staurosporine ( Fig. 2A, 2 h, cell 2, left panel). At this time, the pattern of cytochrome c-GFP fluorescence was unchanged ( Fig. 2A, 2 h,  cell 2, right panel). Subsequently, after 4 h, a greater portion of mitochondria released their TMRM fluorescence ( Fig. 2A, 4 h,  cell 2, left panel), and cytochrome c-GFP fluorescence pattern changed from punctate to partially diffuse, indicating a partial redistribution of cytochrome c-GFP from mitochondria to the cytosol and nucleus ( Fig. 2A, 4 h, cell 2, right panel). After 5 h, all mitochondria in cell 2 had released their TMRM, and cytochrome c-GFP fluorescence showed a fully diffuse pattern. All the cytochrome c-GFP transfected cells in the field behaved asynchronously regarding onsets of mitochondrial depolarization and cytochrome c release.
To confirm that the changes we observed regarding mitochondrial depolarization and cytochrome c-GFP fluorescence were indeed related to staurosporine treatment, we observed cells without staurosporine. Mitochondrial ⌬ and cytochrome c-GFP distribution did not change in the absence of staurosporine (Fig. 2C). Also the nuclear morphology of the cells remained normal (Fig. 2B, Cont). These results excluded the possibility that mitochondrial depolarization and cytochrome c-GFP redistribution would be related to photodamage induced by laser (27).
To quantitate the changes in TMRM fluorescence over time, we measured total TMRM fluorescence in each cell. Fig. 3A shows total TMRM fluorescence (x Ϯ S.E.) of all cells studied. After staurosporine, TMRM fluorescence decreased 50% in 5 h. Untreated cells showed no decline in TMRM fluorescence. Fig.  3B plots individual cellular TMRM fluorescence for each of the cells shown in Fig. 2A. All cells, except cell 7, showed 10 -40% decline in TMRM fluorescence after 2 h of exposure to staurosporine. Subsequently, TMRM fluorescence declined further in each cell in a variable manner. After 7 h, five of nine cells displayed no TMRM fluorescence. Because images of these same cells revealed that individual mitochondria released their TMRM fluorescence heterogeneously in an all-or-none fashion ( Fig. 2A), the decrease of total TMRM fluorescence in individual cells does not mean that all mitochondria within a cell were partially depolarized (Fig. 3B). Rather, a proportion of mitochondria within a cell were fully depolarized. Because the depolarized mitochondria had also lost their punctate cytochrome c-GFP fluorescence, our results indicate strongly that mitochondrial depolarization accompanied and probably preceded cytochrome c-GFP release.
Our fusion construct is a 43-kDa protein, whereas endogenous cytochrome c is a 15-kDa protein. Thus, endogenous cytochrome c might be released before mitochondrial depolarization and before release of the larger molecular weight fusion protein. To exclude this possibility, we monitored the release of endogenous cytochrome c in staurosporine-treated cells by immunocytochemistry (28). Like cytochrome c-GFP, endogenous cytochrome c displayed punctate fluorescence both in untreated cells and in cells treated with staurosporine for 1 h. After longer exposure to staurosporine, cytochrome c staining progressively became more diffuse (data not shown).
Our results show directly for the first time that mitochondrial depolarization and cytochrome c release occur simultaneously in individual mitochondria of single living cells during staurosporine-induced apoptosis. Although the present work does not attempt to define the mechanism of cytochrome c release from mitochondria to the cytosol, the results are compatible with the sequence of events proposed in the MPT hypothesis. Exposure of cells to staurosporine may cause the onset of MPT followed by mitochondrial depolarization. These events may result in matrix swelling because of uptake of water and solutes via the MPT. The expansion of matrix may rupture the mitochondrial outer membrane leading to the release of cytochrome c and other caspase-activating proteins from the intermembrane space (10, 29 -31).
When cytochrome c-GFP was released from the mitochondria into the cytosol, green fluorescence increased in the nucleus in all cells studied. In PC6 cells, the cytosol is a small proportion of cell volume that is mostly comprised of nucleoplasm. Once released from mitochondria into the cytosol, cytochrome c-GFP rapidly diffused into the nucleus, presumably through the nuclear pores. Thus, increases of nuclear cytochrome c-GFP reflected increases of cytosolic cytochrome c-GFP. The impor- tance of this nuclear cytochrome c-GFP in the apoptotic process remains to be elucidated.