Coenzyme q10 prevents apoptosis by inhibiting mitochondrial depolarization independently of its free radical scavenging property.

The permeability transition pore (PTP) is a mitochondrial channel whose opening causes the mitochondrial membrane potential (deltapsi) collapse that leads to apoptosis. Some ubiquinone analogues have been demonstrated previously to modulate the PTP open-closed transition in isolated mitochondria and thought to act through a common PTP-binding site rather than through oxidation-reduction reactions. We have demonstrated recently both in vitro and in vivo that the ubiquitous free radical scavenger and respiratory chain coenzyme Q10 (CoQ10) prevents keratocyte apoptosis induced by excimer laser irradiation more efficiently than other antioxidants. On this basis, we hypothesized that the antiapoptotic property of CoQ10 could be independent of its free radical scavenging ability and related to direct inhibition of PTP opening. In this study, we have verified this hypothesis by evaluating the antiapoptotic effects of CoQ10 in response to apoptotic stimuli, serum starvation, antimycin A, and ceramide, which do not generate free radicals, in comparison to control, free radical-generating UVC irradiation. As hypothesized, CoQ10 dramatically reduced apoptotic cell death, attenuated ATP decrease, and hindered DNA fragmentation elicited by all apoptotic stimuli. This was accompanied by inhibition of mitochondrial depolarization, cytochrome c release, and caspase 9 activation. Because these events are consequent to mitochondrial PTP opening, we suggest that the antiapoptotic activity of CoQ10 could be related to its ability to prevent this phenomenon.

doreductase), II (succinate-ubiquinone oxidoreductase), and III (ubiquinone-cytochrome c oxidoreductase) of mitochondrial respiratory chain and a ubiquitous free radical scavenger (1,2). Our preliminary observations that CoQ 10 is endowed with antiapoptotic activity as a free radical scavenger (3) prompted us to challenge the ability of CoQ 10 to prevent keratocyte apoptosis induced by excimer laser photorefractive keratectomy. Apoptotic death of keratocytes is the earliest pathogenetic event of haze formation, the main drawback of photorefractive keratectomy (4,5). We demonstrated that a 2-h pretreatment with CoQ 10 lowered the number of apoptotic keratocytes in response to excimer laser irradiation to a much higher extent than other free radical scavengers both in vitro (6) and in vivo (5). The antiapoptotic effects of CoQ 10 as free radical scavenger have been demonstrated consistently by others (2,(7)(8)(9)(10)(11) in parallel or in successive works.
The mitochondrial permeability transition pore (PTP) is a complex, large conductance channel that plays a pivotal role in triggering apoptosis (12). The opening of PTP is responsible for disruption of the mitochondrial transmembrane electrochemical gradient (⌬ from Ϫ180 to 0 mV). The oligomerization of apoptotic members of the bcl-2 family follows, thus forming mitochondrial transmembrane channels through which a number of apoptotic factors are released into the cytoplasm to trigger the mitochondrion-dependent (intrinsic) pathway of apoptosis execution (13)(14)(15). This is orchestrated by caspase cascade activation, which leads to cleavage of cellular substrates. Mitochondrion-dependent (intrinsic) and membrane death receptor-dependent (extrinsic) apoptotic pathways are known to share the same caspases with two exceptions: caspase 9 for the former (16) and caspase 8 for the latter (17). Factors and mechanisms involved in PTP formation are still unclear. Nevertheless, three proteins, the adenine nucleotide translocator, the voltage-dependent anion channel, and cyclophilin-D, appear to play the main structural role (13)(14)(15)18). Besides a number of ancillary structural/regulative proteins, such as benzodiazepine peripheral receptor, creatine kinase, hexokinase, and Bax, that are recruited by the voltage-dependent anion channel-adenine nucleotide translocator-cyclophilin-D complex, several molecules are positive (Bax, Ca 2ϩ , ROS, and atractyloside) or negative (Bcl-2, ATP, bongkrekic acid, and cyclosporin A) determinants of PTP opening (15,18).
Some authors (19 -24) demonstrated in isolated mitochondria that the PTP harbors a ubiquinone-binding site and is regulated by complex I of the mitochondrial respiratory chain. Irrespective of the method used to induce permeability transition, the opening of PTP was strongly inhibited by some ubiqui-none analogues, including Ub 0 , decyl-Ub, and Ub 10 , whereas other tested quinones were ineffective or had antithetic effects (23). Specific structural features have been demonstrated to be necessary for regulation of PTP by ubiquinone analogues, independently of their ability to decrease mitochondrial ROS production to the same extent. These results supported the idea that PTP regulation by ubiquinones is exerted through binding to a common site rather than through redox reactions (23). Interestingly, the natural CoQ 10 has a very strict structural analogy with the synthetic Ub 10 (CoQ 2 ); as shown in Fig. 1, their difference is restricted to the number of side isoprenoid units within the side chain, which has been reported to not influence substantially the folding of the molecule (25). These considerations prompted us to hypothesize that what was observed with ubiquinone analogues applied to isolated mitochondria (19 -24) could underlie a natural phenomenon operating in the whole cell. In line with this possibility, the respiratory chain complex I has been suggested to be a constituent of mitochondrial PTP (26). Armstrong and co-workers (27,28) have designed a new model for mitochondrial PTP, in which the ROS producer respiratory chain complex III is functionally linked to the permeability transition. The ability of complex III to generate ROS and to open mitochondrial PTP could override the antiapoptotic function of Bcl-2. Noteworthy, both respiratory chain complexes I and III share CoQ 10 as electron acceptor or electron donor, respectively. These observations led us to explore the possibility that, besides its free radical scavenging property, CoQ 10 may be endowed with antiapoptotic activity as a modulator of PTP opening.
In this study we demonstrate that CoQ 10 provides effective protection against apoptosis of rabbit keratocytes not only in response to the free radicals generated by UVC irradiation but also in response to three apoptotic stimuli generally known to act independently of free radical generation such as chemical hypoxic drug antimycin A (29), the C 2 -ceramide, a cell-permeable analogue of natural lipidic apoptotic messenger ceramide (30), and survival factor withdrawal (serum starvation) (18). This evidence emerged from examination of two classes of different parameters: 1) ROS, MDA, and superoxide dismutase (SOD) activity levels for assessment of the effect of our treatments on free radical generation; and 2) cellular morphology, number of living and apoptotic cells, ATP levels, and DNA status for assessment of the effect of our treatments on cell life and death. In contrast, another well known free radical scavenger, vitamin E (␣-tocopherol) (31)(32), whose effect in association with CoQ 10 has been described in our previous work (5), provides effective protection against apoptosis only in response to UVC irradiation.
The antiapoptotic activity of CoQ 10 was mediated by hindering mitochondrial depolarization, cytochrome c release to cytoplasm, and procaspase 9 activation. This suggests that the mechanism by which CoQ 10 prevents apoptosis may involve inhibition of mitochondrial PTP opening that triggers the intrinsic execution pathway of apoptosis. This possibility and the hypothesis that this novel functional activity may underlie physical localization of CoQ 10 in the PTP, possibly in association with other components of the complex, is now under study.
Treatments-Each of the four damaging agents was applied at doses experimentally established to induce apoptosis: UVC irradiation (254 nm) at 15 mJ/cm 2 , the respiratory chain blocker antimycin A at 200 M concentration, the apoptotic signaling lipid C 2 -ceramide (a synthetic cell-permeable analogue of endogenous ceramides) at 20 M concentration, and FBS restriction to 0.5%. Treatments with 10 M CoQ 10 or with 10 M vitamin E (␣-tocopherol), both dissolved in 0.04% Lutrol F127 used as vehicle to ensure cellular uptake of this hydrophobic molecule (5, 6), were commenced 2 h prior to application of apoptotic stimuli. Vehicle alone-treated cells were used as controls. In some experiments vitamin C (ascorbic acid) was also used.
Identification of Apoptosis on the Basis of Cellular Morphology and DNA End Labeling by Klenow-FragEL TM Assay-Apoptotic RCE cells were identified by light microscopy following end labeling of DNA fragments by Klenow-FragEL TM kit (Oncogene Research Products, Boston). Briefly, following apoptotic treatments, the keratocytes were detached from substrate and fixed in 4% formaldehyde and 80% ethanol at a concentration of 10 6 cells/ml. Prior to detection, the cells (3 ϫ 10 5 ) were affixed to glass slides by cytocentrifugation, air-dried, and processed according to the manufacturer's instructions. In this assay Klenow binds to exposed ends of DNA fragments and catalyzes the templatedependent addition of biotin-labeled deoxyribonucleotides, which are then detected using streptavidin-horseradish peroxidase conjugate. Diaminobenzidine reacts with the labeled fragmented DNA of apoptotic cell nuclei, generating a dark brown insoluble chromogen that contrasts with counterstained methyl green cytoplasms. In contrast, viable cells appear uniformly green or even unstained.
Measurement of ROS and MDA-4-HNE Levels and of SOD Activity-The extent of free radical generation by cultured cells exposed to UVC, antimycin A, C 2 -ceramide, or FBS restriction to 0.5% was measured by means of three independent indexes: reactive oxygen species (ROS), lipid peroxidation level, and superoxide dismutase activity. Generation of ROS was assessed immediately after treatments, using 2Ј,7Ј-dichlorofluorescein diacetate (Sigma) (33), by a confocal laser scanning microscope (Bio-Rad MRC 1024 ES scanning microscope) equipped with a krypton/argon laser source. The emitted fluorescence was monitored at 488 and 568 nm wavelengths with a Nikon plan Apo ϫ 60 oil-immersion objective. To this purpose, series of optical sections (512 ϫ 512 pixels) were taken through the depth of cells with a thickness of 1 m at intervals of 0.8 m. Twenty optical sections for each samples examined were projected as a single composite image by superimposition. To monitor the rate of lipid peroxidation, the levels of typical end products of the process, which are MDA plus 4-hydroxy-2-alkenals, exemplified by 4-hydroxy-2-nonenal (4-HNE), were determined in the cellular extracts prepared and analyzed at 2 h following treatments. Measurements were made using a colorimetric method at 586 nm, according to the reaction of a chromogenic reagent, N-methyl-2-phenylindole, with MDA and 4-HNE in the presence of methanesulfonic acid at 45°C (34). SOD activity was determined at 2 h following treatments using a spectrophotometric assay (Bioxytech, Portland, OR). For this purpose, RCE cells were lysed in 4 volumes of water added to 1 volume of cells, and the cellular lysates were extracted with ethanol/chloroform 62.5: 37.5 (v/v) and vortexed for 30 s. Cellular membranes were separated by centrifugation at 3000 ϫ g at 4°C for 10 min. The upper phase was collected to determine SOD activity. The absorbance of samples was measured spectrophotometrically at 525 nm.
Quantification of Living and Apoptotic Cells-The number of living cells was evaluated by the MTT (C,N-diphenyl-NЈ-4,5-dimethyl thiazol-2-yl tetrazolium bromide) colorimetric assay (reduction of tetrazolium salt to formazan as described previously (35)). The cumulative apoptotic events were scored by the time-lapse videomicroscopy using a Zeiss inverted phase contrast microscope equipped with a 10ϫ objective, Panasonic CCD cameras, and JVC BR9030 time-lapse video recorders. After cell detachment from the substrate, an apoptotic event was counted the moment the cell had shrunk completely and blebbing started (36).
Analysis of ATP Levels-Cellular levels of ATP were evaluated 24 h following apoptotic stimuli, using untreated cells as control as reported previously (36). The cells (2-3 ϫ 10 5 ) were pelletted, resuspended in distilled water, and boiled for 3-5 min. Samples were then cooled to room temperature and stored frozen at Ϫ20°C for later measurements. ATP in the extracts was quantified by a bioluminescence assay with an ATP determination kit (Molecular Probes), using a liquid scintillation analyzer (Camberra Packard) for bioluminescence analysis, according to the manufacturer's instructions.

Detection of Change in Mitochondrial Transmembrane
Potential (⌬)-The change in ⌬ occurring during apoptosis was detected by fluorescence-based assay. The RCE cells were cultured on coverslips in Dulbecco's modified Eagle's medium containing the lipophilic cationic probe 5,5Ј,6,6Ј-tetrachloro-1,1Ј3,3Ј-tetraethylbenzimidazol-carbocyanine iodide (JC-1, 5 mg/ml, Molecular Probes, Eugene, OR) for 15 min at 37°C. This dye has a unique feature: at hyperpolarized membrane potentials (to Ϫ140 mV) it forms a red fluorescent J-aggregate, whereas at depolarized membrane potentials (to Ϫ100 mV) it remains in the green fluorescent monomeric form. Prior to detection, cells were washed in phosphate-buffered saline and placed in an open slide-flow loading chamber that was mounted on the stage of a confocal scanning microscope (Bio-Rad) equipped with a krypton/argon laser source. The emitted fluorescence was monitored at 488 and 568 nm wavelengths with a Nikon plan Apo ϫ60 oil-immersion objective (37). Imaging acquisition of the optical sections through the cell was performed as described above.
Western Blot Analysis of Cytoplasmic Cytochrome c-RCE cells were evaluated 24 h following application of apoptotic stimuli. Cytosolic fractions were prepared as reported previously (38). Proteins in the cytosolic extracts were quantified by the BCA Protein Assay Reagent (Pierce). Proteins (25 g/lane) were electrophoresed through SDS-polyacrylamide 12.5% gel and electroblotted onto nitrocellulose membrane (Schleicher & Schuell) using a transblotter (Bio-Rad). The nonspecific signals were blocked with blocking buffer (5% w/v instant nonfat milk powder in phosphate-buffered saline) and incubated overnight at 4°C with 1 g/ml of anti-cytochrome c monoclonal antibody (Pharmingen). The membrane was washed and subsequently incubated with goat anti-mouse IgG horseradish peroxidase conjugate (Sigma). Detection was carried out using a commercial chemiluminescence procedure (Amersham Biosciences).
Analysis of Caspase 9 Activity-Caspase 9 activity was determined by the Caspase-9 Colorimetric Protease Assay (BioSource Europe, S.A., Nivelles, Belgium). Cytosolic extracts, prepared by lysing cells with Cell Lysis Buffer provided in the kit, were incubated with the colorimetric substrate Leu-Glu-His-Asp (LEHD) conjugated to the chromophore pnitroanilide, in 50 l of 2ϫ Reaction Buffer containing 10 mM dithiothreitol. After a 2-h incubation at 37°C, the absorbance of samples was measured at A 405 nm in Bio-Rad enzyme-linked immunosorbent assay reader.
Analysis of DNA (Nucleosomal Laddering)-Apoptotic internucleosomal DNA fragmentation was evaluated by classical assay, detecting electrophoretically a separated ladder of fragmented DNA. The genomic DNA was extracted from RCE cells as described by Blankenberg et al. (39). The fragments were separated by gel electrophoresis in 0.8% agarose containing ethidium bromide (0.2 g/ml), UV-visualized, and photographed.
Statistical Analysis-The statistical evaluation of the data was performed with the two-tailed Student's t test for unpaired values. Differences were considered statistically significant when p Ͻ 0.05. The data are reported as percentage of the maximal value. Free Radical Generation Was Restricted to UVC-Irradiation-We have then analyzed the effects of the apoptotic doses of UVC irradiation, antimycin A, C 2 -ceramide and serum starvation on free radical generation. For this purpose, three different parameters, which are cellular levels of ROS, MDA-4-HNE levels, and SOD activity, were analyzed after application of apoptotic stimuli (Fig. 2). As shown in Fig. 2A, UVC irradiation was the only apoptotic stimulus able to generate ROS as visualized, immediately after treatment, by confocal microscopy following ROS labeling with the fluorescent indicator (2Ј,7Ј-dichlorofluorescein diacetate) as described by Formigli et al. (36). In keeping with this observation, Fig. 2B shows that only UVC irradiation markedly enhanced MDA-4-HNE levels (left panel) and SOD activity (right panel), evaluated at the 2nd h after treatment, compared with untreated cells (from 5.6 nmol/mg of protein to 19.4 nmol/mg of protein, and from 4.8 units/mg of protein to 17.2 units/mg of protein, respectively). Increase of ROS, MDA-4-HNE, and SOD activity induced by UVC irradiation was substantially lowered by a 2-h pretreatment with either 10 M CoQ 10 or vitamin E (Fig. 2C).

but Not Vitamin E Increased Cell Survival in Response to Free Radical Independent Apoptotic Stimuli-Once
ascertained that apoptosis by antimycin A, C 2 -ceramide and serum starvation was not consequent to free radical generation, the effects of a 2-h pretreatment with 10 M CoQ 10 or vitamin E on cell survival at the 24th h following application of apoptotic stimuli was evaluated by the MTT assay. As shown in Fig. 3, UVC irradiation, antimycin A, C 2 -ceramide, and serum starvation induced a dramatic decrease in the number of living cells compared with untreated controls, which was significantly attenuated by pretreatment with CoQ 10 (upper panel: from 82 to 49%, from 56 to 29%, from 61 to 24%, and from 51 to 80%, respectively). On the contrary, pretreatment with vitamin E was able to attenuate significantly this decrease only in response to UVC irradiation (lower panel: from 82 to 58%).
The effects of the 2-h pretreatment with 10 M CoQ 10 or vitamin E on cell apoptosis induced by UVC irradiation, antimycin A, C 2 -ceramide, and serum starvation were then evaluated using light microscopy and ultramicroscopy (not shown) as reported previously (3, 6, 36). Identification of apoptotic cells with fragmented DNA was also carried out by end labeling with the Klenow-FragEL TM , by which apoptotic cells are easily recognized by the presence of a dark brown stain in contrast to viable cells that appear green or even unstained. At the 24th h following treatment with UVC irradiation, antimycin A, C 2ceramide, and serum starvation, a significant number (about 30 -60% depending on type of apoptotic stimulus) of RCE cells contained brown-stained fragmented DNA (apoptotic cells). Treatment with 10 M CoQ 10 2 h before application of apoptotic stimuli dramatically reduced the number of brown-stained cells, so that the number of cells stained green (viable cells) overcame 90% of all cells. Treatment with 10 M vitamin E significantly reduced brown-stained (apoptotic) cells in response to UVC irradiation but not in response to the other, free radical unrelated, apoptotic stimuli. Results obtained with UVC and serum starvation, either pretreated or not with CoQ 10 or vitamin E, are shown in Fig. 4. Quite similar results have been obtained using 10 M vitamin C (ascorbic acid) as antioxidant (not shown).
CoQ 10  then evaluated by time-lapse videomicroscopy at 24 h following UVC irradiation or treatment with antimycin A, C 2 -ceramide, or serum starvation. Following all treatments, a high number of apoptotic RCE cells was observed in cultured plates, which dramatically decreased if cells were pretreated with CoQ 10 (not shown). Corresponding quantitative data are reported in Fig. 5. When compared with untreated or CoQ 10 -pretreated controls, the number of cumulative apoptotic events scored by timelapse videomicroscopy markedly increased, following application of all apoptotic stimuli, but to a significantly lower extent if the cells were pretreated with CoQ 10 for 2 h before the induction of apoptosis. These results clearly indicated that CoQ 10 was able to increase cell survival by preventing apoptosis even in response to stimuli that do not generate free radicals.
CoQ 10 Attenuated Lowering of ATP Levels in Response to All Apoptotic Stimuli-Execution of the apoptotic death program requires massive consumption of ATP and, consequently, is accompanied by a dramatic reduction in ATP cellular levels. As shown in Fig. 6, UVC irradiation, antimycin A, C 2 -ceramide, and serum starvation markedly lowered ATP cellular levels as compared with untreated controls. Nevertheless, this lowering was significantly attenuated by a 2-h pretreatment with CoQ 10 (from 65 to 28%, from 76 to 41%, from 81 to 51%, and from 60 to 27%, respectively, as compared with respective controls). In further experiments the effects of CoQ 10 in response to UVC irradiation were omitted.
CoQ 10 Counteracted Mitochondrial Depolarization, Cytochrome c Release, Caspase 9 Activation, and DNA Fragmentation in Response to All Apoptotic Stimuli-The effects of antimycin A, C 2 -ceramide, and serum starvation on mitochondrial membrane charge was quantified by the uptake of JC-1 (Fig. 7). The shift in membrane charge was observed as disappearance of fluorescent red-orange-stained mitochondria (large negative ⌬) and an increase in fluorescent green-stained mitochondria (loss of ⌬). Following application of apoptotic stimuli most cells underwent mitochondrial membrane depolarization; indeed, cells with red-orange-stained mitochondria dropped to ϳ20% of total cells. Treatment with CoQ 10 before application of apoptotic stimuli prevented significantly mitochondrial membrane depolarization, because more than 60% of cells examined maintained red-orange-stained mitochondria. CoQ 10 was most efficient in serum-starved cells and had less effect in response to antimycin A or C 2 -ceramide.
Cytochrome c, caspase 9 activity, and DNA status were analyzed as indexes of apoptosis execution by means of the intrinsic (mitochondrion-dependent) pathway (14) (Fig. 8). Fig.  8A shows the release of cytochrome c to cytoplasm, evaluated by Western blotting analysis; at 24 h after application of all free radical unrelated apoptotic stimuli, cytoplasmic levels of cytochrome c markedly increased but remained substantially unaffected if treatments were preceded by CoQ 10 administration.
Similarly, Fig. 8B shows that caspase 9 activity undergoes a 6 -8-fold enhancement at 24 h after application of apoptotic stimuli. This enhancement was dramatically lower when apoptotic stimuli application was preceded by CoQ 10 administration. Fig. 8C shows that pretreatment with CoQ 10 also prevented DNA internucleosomal fragmentation elicited by all free radical unrelated apoptotic stimuli (40).
The ability of CoQ 10 to prevent cytochrome c release and caspase 9 activation in response to free radical unrelated apoptotic stimuli, two events that are triggered by PTP opening, suggests that, independently of its free radical scavenging property, CoQ 10 inhibits apoptosis by directly maintaining mitochondrial PTP in the closed conformation. Finally, prevention by CoQ 10 of DNA internucleosomal fragmentation indicates that blocking of intrinsic apoptotic pathway by pretreatment with CoQ 10 suffices to prevent the "ignition" of the entire apoptotic machinery triggered by apoptotic stimuli. DISCUSSION We have shown previously (5, 6) that apoptosis of rabbit keratocytes induced by excimer laser irradiation, a free radicalgenerating treatment, was prevented both in vitro and in vivo by the free radical scavenger CoQ 10 dissolved in Lutrol F127 and administered 2 h before treatment.
By using the same keratocyte cell line (RCE), we show here that CoQ 10 prevents apoptosis also in response to apoptotic stimuli that do not generate free radicals, which are antimycin A, C 2 -ceramide, and serum starvation. UVC irradiation, a well known free radical-generating damaging agent, has been used as control. We also demonstrate that the mechanism underlying the free radical scavenging independent antiapoptotic properties of CoQ 10 is inhibition of mitochondrial depolarization.
Multiple and often interrelated mechanisms have been described to account for the effects of a single apoptotic stimulus. The damaging effect of UVC irradiation is consequent to direct alteration of molecular substrates, such as proteins and nucleic acids, or mediated by generation of free radicals, which are mainly reactive oxygen species (41). Antimycin A is a chemical hypoxic drug that has been reported to prevent free radical generation (27,(42)(43)(44)(45)(46)(47). Antimycin A induces apoptosis by poi-soning the mitochondrial respiratory chain complex III (29) but also by mimicking a cell death-inducing Bcl-2 homology domain (48). C 2 -ceramide is a cell-permeable analogue of physiological ceramides, a small group of membrane sphingolipids involved in cell growth, differentiation, and apoptosis (30,49,50). Ceramide-induced apoptosis is mainly consequent to its ability to collapse mitochondrial ⌬ (51), either by direct inhibition of mitochondrial respiratory chain complex III (52) or by formation of large transmembrane channels that raise mitochondrial permeability (53,54). Withdrawal of survival factors, achieved in cultured cells by serum starvation, commit cells to apoptosis by different mechanisms, such as mitogen-activated protein kinase induction (55), ceramide release (56), and COX-2 activation (57), whose common target is triggering the mitochondrion-dependent apoptotic pathway (18). On the other hand, some authors have reported that treatments with antimycin A (58, 59), C 2 -ceramide (60), and serum starvation (11) can also lead to ROS generation. Because apoptosis execution is accompanied by ROS generation (27,28,41,61,62), whether a ROS increase during apoptosis is a cause or an effect of this phenomenon is not easy to establish. Because this point is very crucial for our study, in which it was critical to exclude antimycin A, C 2 -ceramide, and serum starvation as directly generating free radicals, we quantified three different free radical generation indexes: ROS, MDA, and SOD activity levels. Quantification was performed early after application of apoptotic stimuli, which is prior to commencement of apoptosis execution as monitored by time-lapse videomicroscopy. A marked increase in ROS, MDA, and SOD activity levels occurred in response to UVC irradiation but not to treatment with antimycin A or C 2 -ceramide or serum starvation, which indicated that, at least in our experimental system, the three latter apoptotic stimuli do not generate free radicals.
The administration of CoQ 10 2 h prior to apoptotic stimuli prevents apoptosis not only in response to UVC irradiation but also to antimycin A, C 2 -ceramide, or serum starvation, i.e. independently of the ability of apoptotic stimuli to trigger or not trigger free radical generation. This protective effect was clearly demonstrated by several evidences, including changes of cell morphology detected by light microscopy and ultramicroscopy, quantification of living and apoptotic cells, and analysis of ATP cellular levels. Indeed, CoQ 10 significantly enhances the number of living cells evaluated by the MTT analysis and lowers the number of cumulative apoptotic events scored by time-lapse videomicroscopy in response to any apoptotic stimulus, whereas the free radical scavenger vitamin E was effective only against the free radical-generating UVC irradiation. In line with this assertion, pretreatment with CoQ 10 was able to prevent the massive reduction in ATP levels induced by all apoptotic stimuli, a phenomenon that is typically associated with the energy-consuming apoptosis execution and in particular with ⌬ collapse consequent to PTP opening.
We suggest that the mechanism by which CoQ 10 exerts its antiapoptotic activity is associated with inhibition of PTP open- ing. This possibility is supported by a body of literature. First, Fontaine et al. (20 -22) have demonstrated that Ca 2ϩ -dependent opening of PTP in isolated mitochondria can be prevented by two synthetic quinone analogues, ubiquinone 0 and the decyl-ubiquinone, whereas the other quinones tested are ineffective; on this basis, they have proposed a model of mitochondrial PTP that harbors a ubiquinone-binding site. Second, Walter et al. (23) have identified the structural features required for regulation of the mitochondrial PTP by ubiquinone analogues and defined three functional classes of ubiquinone analogues, PTP inhibitors, PTP inducers, and PTP-inactive quinones, on the basis of their ability to keep the mitochondrial PTP open or closed or to counteract the effects of both inhibitors and inducers. Because all ubiquinones used decrease mitochondrial ROS production to the same extent, they claim that their different regulative effect on PTP, when present, is mediated by their binding to a common site rather than by their free radical scavenging activity, possibly through secondary changes in the PTP Ca 2ϩ binding affinity. Our result indicating that CoQ 10 , but not vitamin E, prevents mitochondrial depolarization, cytochrome c release, and caspase 9 activation induced by apoptotic stimuli strongly support this possibility. Indeed, these three sequential events are specifically associated with opening of PTP and lead to mitochondrion-dependent apoptosis execution. The PTP opening-dependent mitochondrial permeability transition collapses the mitochondrial ⌬ and induces a massive ATP hydrolysis, an early event in the apoptotic pathway. The consequent release of cytochrome c into cytoplasm and caspase 9 activation triggers caspase cascade that culminates in DNA laddering and apoptosis (12). All these apoptosis execution-related events occurred in our experimental model and were substantially prevented by treatment with CoQ 10 . On the basis of these considerations, CoQ 10 may be a candidate as a functional analogue of the Bcl-2 protein (63). Moreover, association of complexes I and III of the mitochondrial respiratory chain with mitochondrial PTP (19,20,(23)(24)(25)(26)(27)(28) and the integration of CoQ 10 in both complexes suggest the role of CoQ 10 as a structural element and modulator of mitochondrial PTP. Experiments are now in progress to obtain direct evidence of this possibility and to "pin-point" localization of CoQ 10 in the PTP complex.