Ubiquinone Analogs: A Mitochondrial Permeability Transition Pore-Dependent Pathway to Selective Cell Death

Background Prolonged opening of the mitochondrial permeability transition pore (PTP) leads to cell death. Various ubiquinone analogs have been shown to regulate PTP opening but the outcome of PTP regulation by ubiquinone analogs on cell fate has not been studied yet. Methodology/Principal Findings The effects of ubiquinone 0 (Ub0), ubiquinone 5 (Ub5), ubiquinone 10 (Ub10) and decyl-ubiquinone (DUb) were studied in freshly isolated rat hepatocytes, cultured rat liver Clone-9 cells and cancerous rat liver MH1C1 cells. PTP regulation by ubiquinones differed significantly in permeabilized Clone-9 and MH1C1 cells from that previously reported in liver mitochondria. Ub0 inhibited PTP opening in isolated hepatocytes and Clone-9 cells, whereas it induced PTP opening in MH1C1 cells. Ub5 did not affect PTP opening in isolated hepatocytes and MH1C1 cells, but it induced PTP opening in Clone-9 cells. Ub10 regulated PTP in isolated hepatocytes, whereas it did not affect PTP opening in Clone-9 and MH1C1 cells. Only DUb displayed the same effect on PTP regulation in the three hepatocyte lines tested. Despite such modifications in PTP regulation, competition between ubiquinones still occurred in Clone-9 and MH1C1 cells. As expected, Ub5 induced a PTP-dependent cell death in Clone-9, while it did not affect MH1C1 cell viability. Ub0 induced a PTP-dependent cell death in MH1C1 cells, but was also slightly cytotoxic in Clone-9 by an oxidative stress-dependent mechanism. Conclusions/Significance We found that various ubiquinone analogs regulate PTP in different ways depending on the cell studied. We took advantage of this unique property to develop a PTP opening-targeted strategy that leads to cell death specifically in cells where the ubiquinone analog used induces PTP opening, while sparing the cells in which it does not induce PTP opening.


Introduction
Mitochondria are involved in several physiological processes including energy metabolism, calcium homeostasis and programmed cell death [1,2,3]. Numerous mitochondrial proteins, which have no pro-apoptotic activity when they remain inside mitochondria, promote cell death once released into the cytosol [4]. Both extra-mitochondrial and intra-mitochondrial signaling pathways can trigger the release of the mitochondrial proapoptotic proteins [2].
The mitochondrial permeability transition consists of a sudden non-specific increase in the permeability of the inner membrane [5,6]. A prolonged mitochondrial permeability transition results in a drastic ATP synthesis inhibition through the collapse of the proton-motive force, a dramatic increase in ROS production and the release of the mitochondrial pro-apoptotic proteins [7,8,9]. Permeability transition is due to the opening of an inner membrane channel [10]: the Permeability Transition Pore (PTP).
Matrix Ca 2+ is the single most important factor for PTP opening. The amount of matrix Ca 2+ required to open the pore is modulated by a number of factors. The ''PTP-inhibitors'' and the other so-called ''PTP-inducers'' designate factors that increase and decrease the amount of Ca 2+ required to induce PTP opening [6]. Cyclosporin A (CsA) is the reference PTP inhibitor. It inhibits PTP opening by detaching Cyclophilin D (CyP-D) from the other components of the pore [11].
In primary and cultured cells, several drugs known to inhibit PTP opening also decrease cell death in response to various cytotoxic insults [12,13,14]. In animal models, the inhibition of PTP opening by either CsA or genetic ablation of CyP-D provides strong protection from reperfusion injury [15,16,17]. In humans, the first clinical trial has recently shown that CsA treatment reduces infarct size after reperfusion of a coronary thrombosis [18]. These data suggest that PTP inhibition can be beneficial in particular pathological conditions, most likely through its effect on cell death. On the other hand, resistance to Doxorubicin has been shown to be related to PTP inhibition in a human chronic myelogenous leukemia cell line [19], while hepatocarcinogenesis with 2-acetylaminofluorene is preceded by PTP inhibition [20]. Thus, PTP inhibition might in some cases hamper cancer treatments or eventually participate in carcinogenesis.
Because prolonged PTP opening leads to cell death, the PTP represents a cellular target for the commitment to cell death [21]. Indeed, pharmacological agents used in anti-cancer therapy have been reported to target the PTP and to induce cell death via PTP opening [1,21]. Ideally, drugs used for the treatment of malignancies would be far more toxic for cancer cells than for normal cells. However, a PTP-targeted drug able to selectively open the PTP in cancerous cells only remains to be developed.
In a series of experiments conducted with isolated rat liver mitochondria, we have shown that several ubiquinone analogs regulate PTP opening [22,23,24,25]. Three functional classes of quinones were defined, the PTP-inhibitory quinones, the PTPinducing quinones and the PTP-inactive quinones that counteract the effects of both inhibitory and inducing quinones [24,25]. To date, few studies have reported a preventive effect of ubiquinone analogs in a model of cell death. DUb, a PTP-inhibitor quinone in the liver, has been shown to prevent PTP opening-induced cell death in HL 60 cells [26]. In contrast, although Ub 0 is more potent than CsA at PTP inhibition in liver and skeletal muscle mitochondria [27], Ub 0 was ineffective in preventing PTP opening-induced cell death in HL 60 cells [26].
To clarify this issue, and because PTP regulation can vary depending on the tissue studied [27], we have begun a comprehensive study of PTP regulation in different cell lines. This work presents the effect of four ubiquinone analogs on three different rat liver cell lines. We confirm that ubiquinone analogs regulate PTP opening in the different cells tested. However, we found that a number of ubiquinone analogs may regulate PTP in different ways depending on the cell studied. We took advantage of this unique property to develop a PTP opening-targeted strategy that leads to cell death specifically in cells where the ubiquinone analog used induces PTP opening.

Cells
Clone-9 and MH1C1 cells are, respectively, non-cancerous and cancerous rat hepatocyte cell lines. Clone-9 and MH1C1 cells were maintained in exponential growth phase using Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum for Clone-9, and Ham's F12K medium supplemented with 2.5% fetal bovine serum and 15% horse serum for MH1C1. Both media were supplemented with 2 mM glutamine, 1mM sodium pyruvate, 1% non-essential amino acids, 50 units/ml penicillin, and 50 mg/ml streptomycin. Hepatocytes were isolated according to Berry and Friend's methodology, modified by Groen et al [28].

Ca 2+ retention capacity
Non confluent cultured cells, harvested by trypsinization, washed with PBS or freshly isolated hepatocytes were permeabilized immediately before use by incubation of 5610 6 cells under agitation for 2 min at 25uC in a Ca 2+ free medium (Chelex resin, overnight, 4uC) containing 250 mM sucrose, 1 mM Pi-Tris, 10 mM Tris-MOPS (pH 7.4) and 50 mg/ml digitonin. Measurements of Ca 2+ were performed fluorimetrically at 25uC with a PTI Quantamaster C61 spectrofluorometer equipped with magnetic stirring and thermostatic controls. Extra-mitochondrial Ca 2+ was measured in the presence of 1 mM Calcium Green-5N with excitation and emission wavelengths set at 506 and 530 nm, respectively. The Ca 2+ uptake and Ca 2+ release of digitonin permeabilized cells were measured by loading cells with trains of Ca 2+ pulses at constant time intervals.

ROS production
Cells were incubated in a medium containing 250 mM sucrose, 1 mM Pi-Tris, 10 mM Tris-MOPS, and 5 mM H 2 DCFDA. Measurement of H 2 DCFDA oxidation were performed fluorimetrically at 37uC (with excitation and emission wavelengths set at 506 and 521 nm, respectively) with a PTI Quantamaster C61 spectrofluorometer equipped with magnetic stirring and thermostatic controls.

Cells Staining
Clone-9 cells were labeled with the lipophilic dye PKH26 according to the manufacturer's instructions. This non-toxic fluorescent dye binds irreversibly to the cell membrane without affecting cell growth. Therefore, upon cell division, the probe is partitioned equally between each daughter cell, but does not transfer to co-cultured cells [29].

Cells treatment and cell death analysis
Cells were exposed to the indicated concentrations of ubiquinone analogs for 30 min at 37uC in a serum-free medium. Cells were then harvested and incubated in a complete medium at 37uC for 24 h. Annexin V-positive cells were quantified by flow cytometry using a FACSCan flow cytometer (Becton-Dickinson). Cells (1610 6 /ml) were exposed to 5% v/v annexin V-FluoProbes Alexa 488 for 15 min at room temperature. For each sample, a minimum of 10,000 events were analyzed.

Statistics
Stastistical analyses were performed using two-tailed unpaired Student's t tests with equal variances.

Results
The Ca 2+ retention capacity (CRC) represents the minimum Ca 2+ load required to induce PTP opening in an entire population of mitochondria. Therefore, CRC measurement represents a suitable method to quantify and compare the potency of different PTP regulators. The CRC is measured by loading mitochondria with train of Ca 2+ pulses until a rapid Ca 2+ release occurs. This event is accompanied by mitochondrial swelling and membrane depolarization, and is prevented by CsA [27].
The CRC measurement can be performed equally well in isolated or in situ mitochondria (i.e. in digitonin permeabilized cells). However, because PTP regulation may be very sensitive to the conditions of incubation used [27,30], we first checked whether PTP regulation by ubiquinone analogs in permeabilized rat hepatocytes was identical to that previously measured in isolated rat liver mitochondria.
As shown in Figure 1A, Ub 0 and DUb increased the CRC (i.e. inhibited PTP opening) in rat hepatocytes in a concentration-dependent manner up to an optimal concentration beyond which Ub 0 and DUb became less potent at PTP inhibition. Ub 10 inhibited PTP opening at low concentrations, but activated PTP opening at high concentration in rat hepatocytes. Ub 5 did not affect PTP regulation in rat hepatocytes. These results are in total agreement with those previously found with isolated rat liver mitochondria [23,24,25], indicating that the model used (i.e. isolated mitochondria or in situ mitochondria) did not influence the effect of the tested ubiquinone analogs at PTP regulation.
We next studied the effects of Ub 0 , Ub 5 , Ub 10 and DUb on the CRC of two other digitonin permeabilized rat liver cells, namely cultured rat liver Clone-9 cells and cancerous rat liver MH1C1 cells. As shown in Figure 1, PTP opening in the absence of ubiquinone analogs occurred at approximately 60, 10 and 25 nmol Ca 2+ per million cells in isolated hepatocytes, Clone-9 and MH1C1, respectively. Such differences in the basal CRC are expected [31] and may be at least partly related to differences in In immortalized Clone-9 cells ( Figure 1B), Ub 0 and DUb were PTP-inhibitors as observed in permeabilized hepatocytes, although the optimal concentration varied with the cell line studied. Surprisingly, contrary to what occurred in permeabilized hepatocytes, Ub 5 favored PTP opening, whereas Ub 10 was ineffective in Clone-9 cells. In cancerous rat liver MH1C1 cells ( Figure 1C), Ub 0 favored PTP opening, Ub 5 and Ub 10 were ineffective, whereas DUb inhibited PTP opening. These data indicate that the effects of ubiquinone analogs on PTP regulation dramatically differed according to the type of cell used.
In isolated rat liver mitochondria, we have shown that PTPinactive quinones were able to counteract the effects of both inhibitory and inducing quinones, suggesting the existence of a common site of action for which the ubiquinone analogs compete [25]. In order to check whether these competitions persist in cells in which PTP regulation differs from that observed in liver mitochondria, permeabilized Clone-9 and MH1C1 cells were exposed to a combination of PTP-active plus PTP-inactive quinones. Results are presented in Figure 2. In Clone-9, the PTP-inactive quinone Ub 10 was able to counteract the effect of the PTP-inhibitory Ub 0 and of the PTP-inducing Ub 5 . Similarly, in MH1C1, the PTP-inactive quinone Ub 5 was able to counteract the inducing effect of Ub 0 . Therefore, despite the fact that PTP regulation by quinone changes according to the cell studied, this data suggests that the competition between 3 functionally classes of quinones is a ubiquitous phenomenon.
Because PTP opening is expected to induce cell death, we next verified whether the changes observed in PTP regulation were associated with consecutive changes in cell death regulation. In other words, we hypothesized that ubiquinone analogs were capable of inducing cell death selectively in cells in which they favored PTP opening.
In MH1C1 cells ( Figure 3A), PTP-inducer Ub 0 induced cell death in a concentration-dependent manner. Moreover, Ub 0induced cell death was prevented by PTP-inhibitor DUb ( Figure 3A) and by CsA (not shown), confirming that Ub 0 -induced cell death was due to PTP opening in that cell line. As expected, PTP-inactive Ub 5 and PTP-inhibitor DUb did not induce any significant toxicity.
In Clone-9 cells ( Figure 3B), PTP-inducer Ub 5 induced cell death in a concentration-dependent manner. As expected, Ub 5induced cell death was prevented by PTP-inhibitors DUb ( Figure 3B) and by CsA (not shown). Surprisingly, Ub 0 and DUb, which both inhibited PTP opening in that cell line (see Figure 1), affected cell viability in a different manner. DUb did not induce any significant toxicity, whereas Ub 0 induced Clone-9 cell death.
Ubiquinone analogs have been reported to either reduce or increase reactive oxygen species (ROS) formation [25,26]. In order to check whether Ub 0 -induced cell death in Clone-9 was related to oxidative stress, we next measured H 2 DCFDA oxidation (i.e., ROS production) in Clone-9 and MH1C1 cells before and after the addition of Ub 0 , Ub 5 , Ub 10 or DUb. As shown in Figure 4A, Ub 0 dramatically increased ROS production in Clone-9, whereas it slightly decreased ROS production in MH1C1 cells. Ub 5 and Ub 10 also acted as pro-oxidants in Clone-9 (although they were less potent than Ub 0 ), whereas they acted as antioxidants in MH1C1. Only DUb was antioxidant in the two cell lines. Confirming that Ub 0 toxicity in Clone-9 was due to oxidative stress, tocopherol or tiron hampered Ub 0 -induced cell death in Clone-9 ( Figure 4B).
Because Ub 5 induced cell death in Clone-9 but not in MH1C1, we next measured the toxicity of Ub 5 in co-cultures of MH1C1 plus Clone-9 cells. In order to distinguish the two populations of cells, Clone-9 were labeled with the fluorescent lipophilic dye PKH-26 before being co-cultured with MH1C1 in a complete F12K medium. In preliminary experiments, we first checked that PKH-26 did not affected PTP regulation or cell viability (data not shown). As shown in Figure 5A, Ub 5 induced cell death in the Clone-9 population, but it spared the MH1C1 population.
Because a same concentration of Ub 0 was more cytotoxic in cells in which it induced PTP opening (i.e. MH1C1) than in cells where it inhibited PTP opening (i.e. Clone-9), we also measured the toxicity of Ub 0 in co-cultures of MH1C1 plus Clone-9 cells. As shown in Figure 5B, Ub 0 induced cell death in the two populations but with a higher percentage of cell death in MH1C1 cells.

Discussion
In this work we have shown that several ubiquinone analogs are able to regulate PTP opening according to the cell type. Therefore, an important and practical conclusion of this work is that PTP regulation by ubiquinones is an unpredictable phenomenon that cannot be extrapolated from results observed with isolated liver mitochondria. However, this remarkable property led us to develop a PTP-oriented strategy that allows a selective death in cells in which the ubiquinone induces PTP opening, while sparing totally (Ub 5 in MH1C1) or partially (Ub 0 in Clone-9) the cells in which the ubiquinone does not induce PTP opening. If every PTP inhibiting ubiquinone did not necessarily prevent cell death, all the PTP inducing ubiquinones induced cell death, confirming that PTP opening triggers the cell suicide program.
Another important finding of this work is that ubiquinone analogs are able to modulate ROS production in different way according to the cell type. Note however that within a given cell line, the way a ubiquinone analog modulates ROS production does not correlate with the way it regulates PTP opening (compare Figures 1 and 4). Indeed, ubiquinone analogs can (i) inhibit PTP opening and stimulate ROS production (Ub 0 in Clone-9), (ii) favor PTP opening and inhibit ROS production (Ub 0 in MH1C1 cells), (iii) favor both PTP opening and ROS production (Ub 5 in Clone-9), (iv) inhibit both PTP opening and ROS production (DUb in Clone-9 and MH1C1 cells), (v) modify ROS production without obvious effect on PTP regulation (Ub 5 and Ub 10 in MH1C1 cells, Ub 10 in Clone-9). Therefore, the effects of the ubiquinone analogs on ROS production cannot account for the effects on PTP opening, and vice versa. Despite the fact that Ub 0 is the most potent PTP inhibitor discovered so far in rat liver mitochondria [27], this work shows that Ub 0 favors PTP opening in MH1C1 cells or induces oxidative stress in Clone-9 cells. It has been reported that Ub 0 was ineffective in preventing cell death in HL 60 cells [26]. This may be easily explained hypothesizing that Ub 0 may not inhibit PTP opening or may induce oxidative stress in HL 60 cells.
Some tissue-specificities in PTP regulation have been previously reported. For example, CsA does not inhibit PTP opening in a particular line of K562 cell resistant to doxorubicin [19]. The inhibition of respiratory chain complex 1 inhibits PTP opening in U937, KB and HMEC cells [31,32,33], whereas it does not have this effect in rat liver or rat heart mitochondria [6]. However, to the best of our knowledge, this is the first piece of evidence showing that a drug can regulate PTP opening in contradictory ways which are dependent on the cell line.
In previous works conducted with isolated rat liver mitochondria, we proposed that ubiquinone analogs might regulate PTP via a common site. This hypothesis was sustained by the fact that inactive quinones were able to counteract the effects of both inhibitory and inducing quinones. At that time, the biphasic effect of some quinones (inhibitory at low concentration and inactive or even activating at high concentration) was explained by hypothesizing (i) that quinones formed aggregates at high concentrations and (ii) that these aggregates were either inactive or activating compounds able to compete with the monomeric-inhibiting quinone.
The observation that the same concentration of ubiquinone analog can inhibit or activate PTP opening according to the cell line used no longer supports this model. These data are now more consistent with another proposed model [25] involving two regulatory sites: one responsible for inhibition and one for activation. The occupancy of a site by an active compound would, in turn, modulate PTP opening through secondary changes in the PTP Ca 2+ binding affinity, whereas binding by an inactive compound would not. In this model, the biphasic response observed with some quinones could easily be explained through the assumption that these quinones may bind the two sites: the inhibitory site with high affinity and the inducing site with a lower affinity.
Because the effect of some ubiquinone analogs depends on the cell line, we propose that the affinity of these sites for one particular quinone, as well as the secondary changes in the PTP Ca 2+ binding affinity, may change according to the cell line. These changes would depend on genetic or metabolic differences. Note, however, that the observed changes were not due to differences in the growth media (Figure 4), and that the CRC experiments were performed in the same experimental condition ( Figure 1). Therefore, the putative modifications may directly affect the PTP.
This study constitutes the first report of a PTP-targeted strategy able to selectively open the PTP and consecutively provoke cell death in some cells whilst sparing others. Therefore, this work should be viewed as the first proof of concept suggesting that the PTP is a key target for selecting cells that will commit the cell death program. Finding compounds that open PTP in one cancerous cell line (without any side effect on normal cells) needs further studies. This will benefit from an improved knowledge of the molecular nature of the PTP, which in turn will deepen our understanding of why PTP-regulation by quinones (and probably other compounds) changes according to the cell line.