Properties of Ca2+ Transport in Mitochondria of Drosophila melanogaster*

Background: We have studied the properties of Ca2+ transport in Drosophila mitochondria. Results: Drosophila mitochondria possess Ca2+ transport systems matching their mammalian equivalents but have a unique selective Ca2+ release channel that does not mediate swelling. Conclusion: The Drosophila Ca2+ release channel is involved in Ca2+ homeostasis rather than cell death. Significance: This channel may represent the missing link between the permeability transition pore of yeast and mammals. We have studied the pathways for Ca2+ transport in mitochondria of the fruit fly Drosophila melanogaster. We demonstrate the presence of ruthenium red (RR)-sensitive Ca2+ uptake, of RR-insensitive Ca2+ release, and of Na+-stimulated Ca2+ release in energized mitochondria, which match well characterized Ca2+ transport pathways of mammalian mitochondria. Following larger matrix Ca2+ loading Drosophila mitochondria underwent spontaneous RR-insensitive Ca2+ release, an event that in mammals is due to opening of the permeability transition pore (PTP). Like the PTP of mammals, Drosophila Ca2+-induced Ca2+ release could be triggered by uncoupler, diamide, and N-ethylmaleimide, indicating the existence of regulatory voltage- and redox-sensitive sites and was inhibited by tetracaine. Unlike PTP-mediated Ca2+ release in mammals, however, it was (i) insensitive to cyclosporin A, ubiquinone 0, and ADP; (ii) inhibited by Pi, as is the PTP of yeast mitochondria; and (iii) not accompanied by matrix swelling and cytochrome c release even in KCl-based medium. We conclude that Drosophila mitochondria possess a selective Ca2+ release channel with features intermediate between the PTP of yeast and mammals.

Mitochondria play a pivotal role in cellular Ca 2ϩ homeostasis and thereby participate in the orchestration of a diverse range of cellular activities. Indeed, the mitochondrial proton electrochemical gradient is used not only to synthesize ATP but also to accumulate cations into the mitochondrial matrix (1)(2)(3)(4). Consequently, when local cytoplasmic free Ca 2ϩ levels rise, mitochondria rapidly accumulate cytoplasmic Ca 2ϩ and then grad-ually release it as normal cytoplasmic levels are restored, amplifying and sustaining signals arising from elevation of cytoplasmic Ca 2ϩ , as well as protecting cells and neurons against transient elevation in intracellular Ca 2ϩ during periods of hyperactivity (1,5,6). As a result, the mechanisms controlling cellular and mitochondrial Ca 2ϩ homeostasis, metabolism, and bioenergetics must function as a tightly integrated system within the overall cellular Ca 2ϩ homeostatic network (2,(7)(8)(9).
The pathways responsible for mitochondrial Ca 2ϩ uptake and release have been intensely studied on a functional level for Ͼ50 years. In energized mitochondria, the Ca 2ϩ uniporter mediates Ca 2ϩ uptake across the inner mitochondrial membrane, whereas exchangers (Ca 2ϩ for Na ϩ and/or H ϩ ) are responsible for Ca 2ϩ efflux (9 -13). However, when the mitochondrial Ca 2ϩ load exceeds the capacity of inner membrane exchangers, an additional pathway for Ca 2ϩ efflux from mitochondria may exist through opening of the permeability transition pore (PTP). 3 The mitochondrial permeability transition (PT) describes a process of Ca 2ϩ -dependent, tightly regulated increase in the permeability of the inner mitochondrial membrane due to the opening of a high-conductance channel, the PTP (10). PTP opening causes collapse of the mitochondrial membrane potential (⌬) and Ca 2ϩ release through the pore itself, an event that for short "open" times may indeed be involved in physiological Ca 2ϩ homeostasis (14,15), as recently shown in mouse hearts (16) and adult neurons (17) consistent with a role of the PTP in cell signaling (18). Prolonged opening of the PTP, on the other hand, causes stable depolarization, loss of ionic homeostasis, depletion of pyridine nucleotides, respiratory inhibition, matrix swelling, release of cytochrome c, and cell death via apoptosis or necrosis depending on a variety of additional factors, among which cellular ATP and Ca 2ϩ levels play a major role (19).
Together with matrix Ca 2ϩ , P i is an essential inducer of PTP opening in mammals (19), whereas P i exerts an inhibitory action on the yeast permeability pathways triggered by ATP and energization (20 -24; see Ref. 25 for a recent review). In mammals, the PTP can be desensitized by submicromolar concentrations of the immunosuppressant drug cyclosporin A (26 -28) via an interaction with its matrix receptor cyclophilin D (29). Our recent discovery that the inhibitory effect of cyclosporin A and of cyclophilin D ablation on the pore requires P i (30) opens new scenarios. Indeed, this observation may bridge the gap between the pore of yeast and mammals, which we have hypothesized to be much closer than previously thought (31; see Ref. 32 for a review of earlier literature).
Despite its importance as a model organism, the characteristics of mitochondrial Ca 2ϩ transport have been little studied in Drosophila melanogaster. The present study demonstrates that Drosophila mitochondria possess Ca 2ϩ transport systems that are very close to those of mammals and that they can undergo a ruthenium red (RR)-insensitive Ca 2ϩ -induced Ca 2ϩ release through a selective channel that is insensitive to cyclosporin A and inhibited by P i , and whose general features may be intermediate between the properties of the PTP of yeast and that of mammals.
Cell Permeabilization-Cells were detached with a sterile cell scraper, centrifuged at 200 ϫ g for 10 min, and washed twice with Dulbecco's PBS without Ca 2ϩ and Mg 2ϩ , pH 7.4 (Euroclone). The resulting pellet was resuspended in 130 mM KCl, 10 mM MOPS-Tris, pH 7.4 (KCl medium), containing 150 M digitonin and 1 mM EGTA-Tris and incubated for 20 min on ice (6 ϫ 10 7 cells ϫ ml Ϫ1 ). Cells were then diluted 1:5 in KCl medium containing 10 M EGTA-Tris and centrifuged at 200 ϫ g in a refrigerated centrifuge (4°C) for 6 min. The final pellet was resuspended in KCl medium containing 10 M EGTA-Tris at 4 ϫ 10 8 cells ϫ ml Ϫ1 and kept on ice.
Fluorescent Staining of S 2 R ϩ Cell Mitochondria-In the experiments of Fig. 1A energization of mitochondria in both intact and permeabilized S 2 R ϩ cells was analyzed based on accumulation of the potentiometric probe tetramethyl rhodamine methyl ester (TMRM, Molecular Probes). Three days before the experiments, cells were seeded onto sterilized 24-mm round glass coverslips at 2 ϫ 10 6 cells per well in 2 ml of Schneider's medium supplemented with 10% FBS. On the day of experiment, cells were washed once with PBS and incubated for 20 min at room temperature with 1 ml of serum-free Schneider's medium supplemented with 1 g/ml cyclosporin H and 10 nM TMRM. Cyclosporin H is an inhibitor of the plasma membrane multidrug resistance pumps and allows an appropriate loading with the probe by preventing its extrusion at the plasma membrane (34). Images were acquired with an Olympus IX71/IX51 inverted microscope equipped with a xenon light source (75 watts) for epifluorescence illumination and with a 12-bit digital cooled CCD camera (Micromax). For detection of TMRM fluorescence, 568 Ϯ 25-nm bandpass excitation and 585-nm long pass emission filter settings were used.
In the experiments of Fig. 1C, mitochondrial membrane potential was measured using a Perkin-Elmer LS50B spectrofluorometer and evaluated based on the fluorescence quenching of Rhodamine 123. Two milliliters of 130 mM KCl, 10 mM MOPS-Tris, 5 mM Pi-Tris, 10 M EGTA, 0.15 M Rhodamine 123, pH 7.4, were added to the cuvette. The fluorescence of Rhodamine 123 was monitored at the excitation and emission wavelengths of 503 and 523 nm, respectively, with the slit width set at 2.5 nm. After a short incubation to reach stabilization of the signal, 2 ϫ 10 7 permeabilized S 2 R ϩ cells were added to the cuvette. Further additions were as indicated in the figure legends.
Electron Microscopy-S 2 R ϩ cells were washed with PBS and fixed in 2.5% glutaraldehyde in 0.1 M potassium phosphate buffer, pH 7.4, for 2 h at 4°C. After washing with 0.15 M potassium phosphate buffer, pH 7.0, cells were finally embedded in 2% gelatin as described previously (35). Gelatin-embedded samples were post-fixed with 1% osmium tetroxide in cacodylate buffer 0.1 M, pH 7.4, and embedded in Epon812 resin, sectioned, and stained following standard procedures (36). Ultrathin sections were observed with a Philips EM400 transmission electron microscope operating at 100 kV.
Mitochondrial Respiration-Rates of mitochondrial respiration were measured using a Clark-type oxygen electrode equipped with magnetic stirring and thermostatic control maintained at 25°C, and additions were made through a syringe port in the frosted glass stopper sealing the chamber. Intact S 2 R ϩ cells were incubated in Hank's balanced salt solution supplemented with 10 mM glucose and 5 mM P i -Tris, pH 7.4, whereas digitonin-permeabilized cells (see above) were incubated in 130 mM KCl, 10 mM MOPS-Tris, 5 mM P i -Tris, 5 mM succinate-Tris, 10 M EGTA, pH 7.4. In both cases, 2 ϫ 10 7 cells in 2 ml were used, and further additions are specified in the figure legends.
Light Scattering and Mitochondrial Ca 2ϩ Fluxes-Light scattering at 90°was monitored with a PerkinElmer LS50B spectrofluorimeter at 540 nm with a 5.5-nm slit width. Extramitochondrial Ca 2ϩ was measured with Calcium Green 5N (Molecular Probes) using either the PerkinElmer LS50B spectrofluorometer equipped with magnetic stirring (excitation and emission wavelengths of 505 and 535 nm, respectively) or a Fluoroskan Ascent FL (Thermo Electron Corp.) equipped with a plate shaker (excitation and emission wavelengths of 485 and 538 nm, respectively with a 10-nm band pass filter). The incubation medium contained 130 mM KCl, 10 mM MOPS-Tris, 5 mM succinate-Tris, 10 M EGTA, 2 M rotenone, pH 7.4, and P i -Tris as indicated in the figure legends. In the Ca 2ϩ measurements, 0.5 M Calcium-Green 5N was also added. Permeabilized cells (2 ϫ 10 7 in a final volume of 2 ml in the PerkinElmer spectrofluorometer and 2 ϫ 10 6 in a final volume of 0.2 ml in the Fluoroskan) were used. Further additions were made as indicated in the figure legends.
Western Blotting-Cell suspensions were centrifuged at 3000 ϫ g at 4°C. Proteins from the supernatants were precipitated in acetone at Ϫ20°C and centrifuged for 30 min at 18,000 ϫ g at 4°C. Pellets were washed twice in 20% methanol and finally solubilized in Laemmli gel sample buffer. Cell pellets were lysed in a buffer containing 150 mM NaCl, 20 mM Tris, pH 7.4, 5 mM EDTA-Tris, 10% glycerol, 1% Triton X-100, and supplemented with protease and phosphatase inhibitor cocktails (Sigma), and kept on ice for 20 min. Suspensions were then centrifuged at 18,000 ϫ g for 25 min at 4°C to remove insoluble materials. The supernatants were solubilized in Laemmli gel sample buffer. Samples were separated by 15% SDS-PAGE and transferred electrophoretically to nitrocellulose membranes using a Mini Trans-Blot system (Bio-Rad). Western blotting was performed in PBS containing 3% nonfat dry milk with monoclonal mouse anti-cytochrome c (BD Biosciences), monoclonal mouse anti-OxPhos complex IV subunit I (Invitrogen), or rabbit polyclonal anti-TOM20 (Santa Cruz Biotechnology) antibodies.
Reagents and Statistics-All chemicals were of the highest purity commercially available. Reported results are typical of at least three replicates for each condition, and error bars refer to the S.D.

RESULTS
We initially isolated mitochondria from Drosophila flight muscles after dissection of the thoraces to prevent contamination from the yeast on which Drosophila feeds and that may be present in the abdomen. Despite our great efforts mitochondria were of poor quality, as judged from the respiratory control ratios (results not shown). An additional problem we encountered was that the low yield of these preparations did not allow a reproducible analysis of the Ca 2ϩ transport properties of mitochondria. Thus, we characterized mitochondrial function in intact S 2 R ϩ Drosophila cells and then used digitonin permeabilization to access mitochondria in situ, an approach that we have successfully applied to mammalian cells (37) and to cells from 6-h-old embryos from Danio rerio (zebrafish) (38).
Mitochondria in both intact and permeabilized S 2 R ϩ cells were energized, as shown by fluorescence images after the addition of the potentiometric probe TMRM (Fig. 1A). Mitochondria appeared as bright bodies, and fluorescence was lost upon addition of an uncoupler (Fig. 1A). Ultrastructural analysis of intact S 2 R ϩ cells revealed round-shaped mitochondria with thin cristae aligned in parallel rows (Fig. 1B, left panel illustrates a typical example), which is strikingly similar to the morphology of mammalian mitochondria in situ and to the "orthodox" configuration of Hackenbrock (39). After digitonin treatment, most cells showed evidence of permeabilization as reflected by a change in the electron density of the cytoplasm and loss of chromatin definition (results not shown), but the overall morphology of organelles was retained (Fig. 1B, right panel). Mitochondria, however, now displayed a "condensed" configuration very similar to that of isolated mammalian mitochondria (39), which is characterized by an electron-dense matrix and evident and well preserved cristae and outer membrane (Fig. 1B, right  panel).
Digitonin-permeabilized cells are accessible to substrates, and this allows the study of their response to energization. Mitochondria readily developed a membrane potential (as  DECEMBER 2, 2011 • VOLUME 286 • NUMBER 48 judged on the basis of fluorescence quenching of Rhodamine 123) upon addition of the complex I substrates glutamate and malate (Fig. 1C). The sequential addition of rotenone, succinate, antimycin A, ascorbate plus tetramethyl-p-phenylene diamine, and finally cyanide caused the expected repolarization-depolarization cycles that indicate functioning of all respiratory complexes (Fig. 1C).

Mitochondrial Ca 2؉ Transport in Drosophila
Intact S 2 R ϩ cells displayed a good respiratory activity that was largely inhibited by oligomycin, indicating that a prevalent fraction of oxygen uptake was devoted to ATP synthesis. Basal respiration could be stimulated Ͼ5-fold by the addition of the uncoupler carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP), indicating a good reserve capacity of the respiratory chain (supplemental Fig. 1A). In addition, permeabilized cells displayed a good phosphorylation capacity after energization with succinate (supplemental Fig. 1B), and we used these conditions to study the properties of mitochondrial Ca 2ϩ transport.
Addition of RR alone after Ca 2ϩ uptake was followed by a slow process of Ca 2ϩ release (Fig. 2C, trace b), which suggests the existence of a Na ϩ -insensitive Ca 2ϩ release pathway as also found in mammalian mitochondria (10). Addition of FCCP after accumulation of Ca 2ϩ caused a fast process of Ca 2ϩ release (Fig. 2C, trace c), which was only partly inhibited by RR (Fig. 2C, trace d) without any additional inhibitory effect of cyclosporin A (Fig. 2C, trace e). These experiments suggest the presence of a voltage-dependent Ca 2ϩ release pathway (the RRinsensitive fraction of FCCP-induced Ca 2ϩ release) resembling the PTP of mammalian mitochondria except for its lack of sensitivity to cyclosporin A (50, 51). We screened additional compounds for potential inhibition of RR-insensitive, FCCP-induced Ca 2ϩ release, and we found a concentration-dependent inhibition by tetracaine (Fig. 2D, traces b-d), which also inhibits the PTP of mammalian mitochondria (52,53).
We next studied the Ca 2ϩ retention capacity (CRC) of Drosophila mitochondria by adding a train of Ca 2ϩ pulses to permeabilized cells (Fig. 3). Ca 2ϩ uptake was followed by spontaneous Ca 2ϩ release (Fig. 3A, trace a), which was accompanied by mitochondrial depolarization (results not shown) and delayed by tetracaine (Fig. 3A, traces b and c), which considerably increased the CRC (Fig. 3B). Note that the rate of Ca 2ϩ uptake was not affected by tetracaine, indicating that the Ca 2ϩ uniporter is not inhibited by this drug.
The CRC was strikingly affected by P i , in the sense that the threshold Ca 2ϩ load required for onset of Ca 2ϩ release increased at increasing concentrations of P i (Fig. 4). The rate of spontaneous Ca 2ϩ release decreased at increased P i concentrations despite the larger matrix Ca 2ϩ load (Fig. 4A). The halfmaximal effect of P i was seen at ϳ1 mM, which is similar to that required for inhibition by P i of the PTP of yeast (23,54) and of the PTP of mammals in cyclophilin D null mitochondria and in wild-type mitochondria treated with cyclosporin A (30).
We also tested the effect on the CRC of Ub0, a cyclophilin D-independent inhibitor of the mammalian pore (55,56) and of the combination of ADP plus oligomycin, which is very effective at desensitizing the PTP to Ca 2ϩ (57). No changes of CRC were observed with any of these PTP inhibitors, irrespective of whether the P i concentration was 1 or 5 mM (supplemental Fig. 2).   1 mM tetracaine (traces b and c, respectively). B, The amount of Ca 2ϩ accumulated prior to onset of Ca 2ϩ -induced Ca 2ϩ release in presence of the stated concentrations of tetracaine was normalized to that obtained in absence of tetracaine (CRC 0 ).
Mitochondrial Ca 2ϩ -induced Ca 2ϩ release could be induced by the dithiol oxidant diamide (Fig. 5A) in a process that was prevented by dithiothreitol (Fig. 5B). Ca 2ϩ release could also be induced by N-ethylmaleimide (NEM) (Fig. 6) after a lag phase that decreased as the concentration of NEM was increased (Fig.  6B) in the same range causing PTP opening in mammalian mitochondria (58).
We assessed mitochondrial volume changes in mitochondria subjected to an appropriate Ca 2ϩ load sufficient to cause spon-taneous Ca 2ϩ release at 0.1 mM P i . Parallel readings of Ca 2ϩ fluxes (Fig. 7A) and of light scattering at 540 nm (a sensitive measure of mitochondrial volume changes, Fig. 7B) revealed that after the small light scattering increase (matrix volume contraction) accompanying Ca 2ϩ uptake no matrix swelling (which should manifest itself as a decreased light scattering) could be detected after the onset of Ca 2ϩ release (Fig. 7B). It should be noted that mitochondria in permeabilized S 2 R ϩ cells can undergo swelling upon addition of the pore-forming peptide alamethicin, which also caused rapid release of residual matrix Ca 2ϩ (Fig. 7), or of the selective K ϩ ionophore valinomycin (supplemental Fig. 3). Mitochondrial Ca 2ϩ -dependent Ca 2ϩ release was not accompanied by cytochrome c release, which was instead readily elicited by the addition of alamethicin (Fig. 7C). This result is particularly striking because our experiments were carried out in KCl-based medium, which promotes ready cytochrome c removal if the outer membrane breaks following osmotic swelling of mammalian mitochondria (59). Electron microscopy fully confirmed that the condensed mitochondrial morphology was totally unaffected by a load of Ca 2ϩ able to induce full Ca 2ϩ release (compare the left and middle panels of Fig. 7D). This is a unique feature compared with the swelling response of mitochondria from all sources tested so far under similar conditions (19). On the other hand, mitochondrial swelling was readily detected after the addition of alamethicin (Fig. 7D, right panel).

DISCUSSION
In this work, we have characterized the pathways for Ca 2ϩ transport in mitochondria from digitonin-permeabilized Drosophila S 2 R ϩ cells. These cells were originally derived from late embryonic stages (20 -24 h), and selection was made based on the ability to adhere to tissue culture dishes (60). According to Schneider (60), they represent a variety of tissue precursors, and we assume that they are representative of Drosophila, although a full characterization of the Ca 2ϩ release channel will have to await its molecular definition.
We have found that mitochondria of S 2 R ϩ cells possess the classical pathways found in mammalian mitochondria, i.e. (i) the RR-sensitive Ca 2ϩ uniporter, which has been characterized by electrophysiology (12) and recently identified at the molecular level in mammals (40,41). The existence in the Drosophila    DECEMBER 2, 2011 • VOLUME 286 • NUMBER 48 genome of close orthologs of the mitochondrial Ca 2ϩ uniporter (40,41) and of the previously identified MICU1 (61) (CG18769 and CG4495, respectively) predicts the existence of a mitochondrial Ca 2ϩ uniporter in keeping with our findings. (ii) The Na ϩ -Ca 2ϩ antiporter recently identified as NCLX (49), whose ortholog also exists in Drosophila (CG14744) and is the likely mediator of the Na ϩ -dependent Ca 2ϩ release defined here. (iii) The putative H ϩ -Ca 2ϩ antiporter mediating Ca 2ϩ release at high membrane potential, which can be unmasked by the addition of RR (10). Notably, it recently has been proposed that LETM1 (and its Drosophila ortholog CG4589) mediates H ϩ -Ca 2ϩ exchange by catalyzing RR-sensitive Ca 2ϩ uptake in mitochondria (62). However, this contrasts with the well established role of LETM1 as a K ϩ -H ϩ antiporter (63)(64)(65)(66) and with the fact, confirmed here, that the putative H ϩ -Ca 2ϩ antiporter is insensitive to RR. (iv) A tetracaine-sensitive, RR-insensitive release pathway that opens in response to matrix Ca 2ϩ loading or to depolarization and mediates Ca 2ϩ release. The tetracainesensitive pathway, which displays unique features that appear to be intermediate between those of the PTP of yeast and mammals (31), is the main focus of the present manuscript.

Mitochondrial Ca 2؉ Transport in Drosophila
Disequilibrium between Distribution of Ca 2ϩ and Its Electrochemical Gradient-Ca 2ϩ uptake is an electrophoretic process driven by the Ca 2ϩ electrochemical gradient, ⌬ Ca. In respiring mitochondria, the inside-negative ⌬ favors uptake of Ca 2ϩ (67,68); and with a ⌬ of Ϫ180 mV, the Ca 2ϩ accumulation ratio at equilibrium (i.e. at ⌬ Ca ϭ 0) should be 10 6 (69). This is never reached because at resting cytosolic Ca 2ϩ levels, the rate of Ca 2ϩ uptake is comparable with that of the efflux pathways, and Ca 2ϩ distribution is governed by a kinetic steady state rather than by the thermodynamic equilibrium (69,70). The activity of the mitochondrial Ca 2ϩ uniporter and of the antiporters indeed creates a Ca 2ϩ cycle across the inner membrane, whose energy requirement is very low (71) because the combined maximal rate of the efflux pathways is ϳ20 nmol Ca 2ϩ ϫ mg Ϫ1 protein ϫ min Ϫ1 (10). On the other hand, because the V max of the uniporter is ϳ1400 nmol Ca 2ϩ ϫ mg Ϫ1 protein ϫ min Ϫ1 , and its activity increases sharply with the increase of extramitochondrial [Ca 2ϩ ] (72), this arrangement exposes mitochondria to the hazards of Ca 2ϩ overload when cytosolic [Ca 2ϩ ] increases. We have argued that the PTP may serve the purpose of providing mitochondria with a fast Ca 2ϩ release channel (10,14). This hypothesis is consistent with the effects of cyclosporin A on Ca 2ϩ distribution in rat ventricular cardiomyocytes (73), with a PTP activating response to the combined action of two physiological stimuli increasing cytosolic [Ca 2ϩ ] without detrimental effects on cell survival (17), and with the demonstration that cyclophilin D ablation causes mitochondrial Ca 2ϩ overload in vivo, which, in turn, increases the propensity to heart failure after transaortic constriction, overexpression of Ca 2ϩ /calmodulin-dependent protein kinase II␦c or swimming exercise (16; see Ref. 75 for discussion).
Properties of Drosophila Ca 2ϩ -induced Ca 2ϩ Release-The properties of the Drosophila Ca 2ϩ release system described here appear to be intermediate between those of the PTP of mammals and yeast. Like the mammalian pore, Drosophila Ca 2ϩ release is inhibited by tetracaine (52) and opens in response to matrix Ca 2ϩ loading (76), inner membrane depo- larization (77), thiol oxidation (78), and treatment with relatively high concentrations of NEM (58); like the yeast PTP (and at variance from the mammalian pore), it is inhibited by P i (22,23) and insensitive to cyclosporin A (23). The latter observations may be strictly related. P i is a classical inducer of the mammalian PTP, yet P i is essential for PTP inhibition by cyclosporin A and cyclophilin D ablation (30), suggesting that cyclophilin D masks an inhibitory site for P i (79). It is interesting to note that a Drosophila mitochondrial cyclophilin has not been found and that even Drosophila Cyp1-PA, which according to the primary sequence, has a high probability of import into mitochondria, could not be found in the organelle after tagging with GFP and expression in S 2 R ϩ and KC cells. 4 It is tempting to speculate that lack of mitochondrial cyclophilin leaves the P i inhibitory site unhindered and that the PTP-stimulating ability of P i has developed after the evolutionary divergence of Drosophila and vertebrates.
At the onset of Ca 2ϩ -dependent Ca 2ϩ release, Drosophila mitochondria undergo depolarization, suggesting that the putative channel is also permeable to H ϩ . On the other hand, no matrix swelling is observed in KCl-based medium, indicating that the channel is not permeable to K ϩ (and Cl Ϫ ), despite the fact that the hydrated radius of Ca 2ϩ is larger than that of K ϩ . Lack of swelling, which was confirmed by lack of cytochrome c release and by ultrastructural analysis, is not due to peculiar features of Drosophila mitochondria because matrix swelling and cytochrome c release readily followed the addition of the K ϩ ionophore valinomycin or of the pore-forming peptide alamethicin. We conclude that the putative Ca 2ϩ release channel of Drosophila mitochondria is also permeable to H ϩ . This is an essential feature because the Ca 2ϩ diffusion potential created by efflux through a Ca 2ϩ -selective channel would otherwise oppose Ca 2ϩ release (10).
Mitochondrial Ca 2ϩ -dependent Ca 2ϩ Release as Mediator of Cell Death in Drosophila?-Available evidence points to persistent activation of the PTP as a prime mediator of apoptotic or necrotic cell death in a variety of situations (19). Indeed, unregulated opening of the PTP and ensuing mitochondrial and cellular dysfunction may be responsible for the pathology that characterizes a variety of human diseases (19). Although many of the proteins important for apoptosis in mammalian cells are conserved in Drosophila, the role that mitochondria play in cell death in this organism remains controversial (74,80). The apparent absence of a regulatory role for a mitochondrial cyclophilin in the function of the "Drosophila PTP" prevents an investigation based on the effects of cyclosporin A in cells. However, our functional studies pave the way for the application of the sophisticated genetic strategies available in Drosophila to define the molecular nature of the channel and its role in pathophysiology of Ca 2ϩ homeostasis.