VDAC as a voltage-dependent mitochondrial gatekeeper under physiological conditions.

Mitochondria, composed of two membranes, play a key role in energy production in eukaryotic cells. The main function of the inner membrane is oxidative phosphorylation, while the mitochondrial outer membrane (MOM) seems to control the energy flux and exchange of various charged metabolites between mitochondria and the cytosol. Metabolites cross MOM via the various isoforms of voltage-dependent anion channel (VDAC). In turn, VDACs interact with some enzymes, other proteins and molecules, including drugs. This work aimed to analyze various literature experimental data related to targeting mitochondrial VDACs and VDAC-kinase complexes on the basis of the hypothesis of generation of the outer membrane potential (OMP) and OMP-dependent reprogramming of cell energy metabolism. Our previous model of the VDAC-hexokinase-linked generation of OMP was further complemented in this study with an additional regulation of the MOM permeability by the OMP-dependent docking of cytosolic proteins like tubulin to VDACs. Computational analysis of the model suggests that OMP changes might be involved in the mechanisms of apoptosis promotion through the so-called transient hyperpolarization of mitochondria. The high concordance of the performed computational estimations with many published experimental data allows concluding that OMP generation under physiological conditions is highly probable and VDAC might function as an OMP-dependent gatekeeper of mitochondria, controlling cell life and death. The proposed model of OMP generation allows understanding in more detail the mechanisms of cancer death resistance and anticancer action of various drugs and treatments influencing VDAC voltage-gating properties, VDAC content, mitochondrial hexokinase activity and VDAC-kinase interactions in MOM.

Apparent "mild depolarization of the inner mitochondrial membrane" as a result of a possible generation of the outer membrane potential.

Recently reported kinase-linked mild depolarization of mitochondria, which prevents the generation of the reactive oxygen species (ROS) and disappears in various organs of the old mice, has been assumed to represent a crucial component of the mitochondrial anti-aging program. To measure mitochondrial inner membrane potential (IMP), the authors used fluorescent probe safranin O+. It is widely accepted that the accumulation of such cationic probes in the mitochondrial matrix depends exclusively on IMP, thus completely ignoring the possibility of the outer membrane potential (OMP) generation. However, computational analysis performed in the presented work suggests that the kinase-linked generation of the positive OMP might take place under the described conditions, because the measured potential includes the algebraic sum of both IMP and OMP. Alternatively to the suggested mild depolarization of mitochondria, the reported experimental data might reflect mainly a change of the positive OMP generated by the VDAC-kinase complexes. We also demonstrate that the reported in the literature mitochondrial hyperpolarization induced by erastin (known to prevent VDAC-tubulin interactions) and the depolarization caused by the mitochondrial VDAC knockdowns in the cancer cells might actually represent a decrease or increase, respectively, of the magnitude of the kinase-linked positive OMP. This is consistent with our hypothesis that VDAC voltage gating by the kinase-linked metabolically-dependent OMP plays a very important physiological role in regulating the cell energy metabolism under normal and pathological conditions, in the maintenance of the cell death resistance and even in the genetic aging program.

Cytolytic activity of peptides derived from the Cry11Bb insecticidal toxin of B. thuringiensis subsp. medellin.

A few Bacillus thuringiensis Cry proteins, known as parasporins, have demonstrated cell proliferation inhibition of human cancer cells in vitro after protease activation. In this work, eight peptides derived from the Cry11Bb protoxin produced by B. thuringiensis subsp. medellin were selected and evaluated to investigate their membrane permeabilization and cytolytic activities, using red blood cells and cancer cell lines A549, MCF-7 and Caco-2, respectively. The most active peptides permeabilized red blood cells in a membrane potential-dependent manner. Half maximal inhibitory concentration in cancer cells was in the range 0.78-7.63 µM. At the same time, at peptides concentration of 25 µM, the hemolysis percentage varied in the range of 4.6-32.4%. The peptides BTM-P1 and BTM-P4 in D form had the lowest IC50 values on the MCF-7 cell line and they are considered as the most promising peptides among the evaluated. Fluorescence microscopy using AnnexinV-FLUOS staining indicates that the possible cause of MCF-7 cell death by peptide BTM-P1, is apoptosis. Real time PCR analysis showed an increased transcription of p53 in MCF-7 cells, thus confirming the probable pro-apoptotic effect of the peptide BTM-P1. In general, this study suggests that the cytolytic activity of the polycationic peptides derived from the Cry11Bb protoxin could be mediated by a pro-apoptotic mechanism that might include potential-dependent membrane permeabilization. Further studies might be accomplished to establish whether the peptides are cytolytic to other cancer cell lines and to solid tumors.

Electrical control of the cell energy metabolism at the level of mitochondrial outer membrane.

Energy, generated by the mitochondrial oxidative phosphorylation system, is transferred to the cytosol across the mitochondrial outer membrane (MOM), through the voltage-dependent anion channels (VDACs). The role of the VDAC's voltage-gating process to control the transfer of ATP, creatine phosphate and other negatively charged metabolites across MOM might be crucial for the cell energy metabolism regulation. However, it depends on the probability of the outer membrane potential (OMP) generation by a currently undefined mechanism that has usually been considered doubtful, based on the assumption that VDACs always stay in the electrically open state. Nevertheless, computational analysis of various possible metabolically-dependent mechanisms of OMP generation suggests that MOM is not a "coarse sieve", but in fact it functions as an electrical gatekeeper of cell energy metabolism, due to a probable OMP-dependent VDAC's gating. OMP generation could also be involved in the control of cell death resistance and mechanisms of various diseases.

VDAC electronics: 5. Mechanism and computational model of hexokinase-dependent generation of the outer membrane potential in brain mitochondria.

Glycolysis plays a key role in brain energy metabolism. The initial and rate-limiting step of brain glycolysis is catalyzed mainly by hexokinase I (HKI), the majority of which is bound to the mitochondrial outer membrane (MOM), mostly through the mitochondrial inter-membrane contact sites formed by the voltage-dependent anion channel (VDAC, outer membrane) and the adenine nucleotide translocator (ANT, inner membrane). Earlier, we proposed a mechanism for the generation of the mitochondrial outer membrane potential (OMP) as a result of partial application of the inner membrane potential (IMP) to MOM through the electrogenic ANT-VDAC-HK inter-membrane contact sites. According to this previous mechanism, the Gibbs free energy of the hexokinase reaction might modulate the generated OMP (Lemeshko, Biophys. J., 2002). In the present work, a new computational model was developed to perform thermodynamic estimations of the proposed mechanism of IMP-HKI-mediated generation of OMP. The calculated OMP was high enough to electrically regulate MOM permeability for negatively charged metabolites through free, unbound VDACs in MOM. On the other hand, the positive-inside polarity of OMP generated by the IMP-HKI-mediated mechanism is expected to protect mitochondria against elevated concentrations of cytosolic Ca2+. This computational analysis suggests that metabolically-dependent generation of OMP in the brain mitochondria, controlled by many factors that modulate VDAC1-HKI interaction, VDAC's voltage-gating properties and permeability, might represent one of the physiological mechanisms of regulation of the brain energy metabolism and of neuronal death resistance, and might also be involved in various neurodegenerative disorders, such as Alzheimer's disease.

VDAC electronics: 4. Novel electrical mechanism and thermodynamic estimations of glucose repression of yeast respiration.

Inhibition of cell respiration by high concentrations of glucose (glucose repression), known as "Crabtree effect", has been demonstrated for various cancerous strains, highly proliferating cells and yeast lines. Although significant progress in understanding metabolic events associated with the glucose repression of cell respiration has been achieved, it is not yet clear whether the Crabtree effect is the result of a limited activity of the respiratory chain, or of some glucose-mediated regulation of mitochondrial metabolic state. In this work we propose an electrical mechanism of glucose repression of the yeast S. cerevisiae, resulting from generation of the mitochondrial outer membrane potential (OMP) coupled to the direct oxidation of cytosolic NADH in mitochondria. This yeast-type mechanism of OMP generation is different from the earlier proposed VDAC-hexokinase-mediated voltage generation of cancer-type, associated with the mitochondrial outer membrane. The model was developed assuming that VDAC is more permeable to NADH than to NAD+. Thermodynamic estimations of OMP, generated as a result of NADH(2-)/NAD+(1-) turnover through the outer membrane, demonstrated that the values of calculated negative OMP match the known range of VDAC voltage sensitivity, thus suggesting a possibility of OMP-dependent VDAC-mediated regulation of cell energy metabolism. According to the proposed mechanism, we suggest that the yeast-type Crabtree effect is the result of a fast VDAC-mediated electrical repression of mitochondria due to a decrease in the outer membrane permeability to charged metabolites and owing their redistribution between the mitochondrial intermembrane space and the cytosol, both controlled by metabolically-derived OMP.

Computational models for monitoring the trans-membrane potential with fluorescent probes: the DiSC3(5) case.

Fluorescent permeant charged probes are commonly used for monitoring the trans-membrane potential in lipid vesicles and biological membranes, which has been earlier described by various mathematical models. In the present study, we developed a more complex model based on the computational step-by-step analysis of the influence of various factors, such as the membrane surface potential, ionic strength, and the aggregation properties of cationic cyanine probe DiSC3(5) in the membrane and aqueous phases, in addition to the Nernstian distribution of the probe across the membrane and the hydrophobic interaction with the lipid bilayer. The final full model allows prediction of the optimal experimental conditions for monitoring the trans-membrane potential, such as the probe/lipid ratio and the concentration of liposomes, with a given percentage of negatively charged phospholipids in the membrane, the ionic strength of the aqueous media, the "membrane-water" partition coefficient and the aggregation properties of the probe, as well as the most adequate mode of fluorescence measurement. In agreement with many experimental studies, this model showed high voltage sensitivity of the quantity of the aqueous phase DiSC3(5) monomers, showing its almost exponential decrease with an increase in the trans-membrane potential value. The model also demonstrated the highest voltage sensitivity of the ratio of the quantity of DiSC3(5) monomers in the aqueous phases to that in the membrane phase. A new combined parameter, the logarithmic function of this ratio, demonstrated almost linear changes within a wide range of the trans-membrane potential changes.

VDAC electronics: 3. VDAC-Creatine kinase-dependent generation of the outer membrane potential in respiring mitochondria.

Mitochondrial energy in cardiac cells has been reported to be channeled into the cytosol through the intermembrane contact sites formed by the adenine nucleotide translocator, creatine kinase and VDAC. Computational analysis performed in this study showed a high probability of the outer membrane potential (OMP) generation coupled to such a mechanism of energy channeling in respiring mitochondria. OMPs, positive inside, calculated at elevated concentrations of creatine are high enough to restrict ATP release from mitochondria, to significantly decrease the apparent K(m,ADP) for state 3 respiration and to maintain low concentrations of Ca(2+) in the mitochondrial intermembrane space. An inhibition by creatine of Ca(2+)-induced swelling of isolated mitochondria and other protective effects of creatine reported in the literature might be explained by generated positive OMP. We suggest that VDAC-creatine kinase-dependent generation of OMP represents a novel physiological factor controlling metabolic state of mitochondria, cell energy channeling and resistance to death.

VDAC electronics: 2. A new, anaerobic mechanism of generation of the membrane potentials in mitochondria.

Mitochondrial hexokinase (HK) and creatine kinase (CK) known to form complexes with a voltage dependent anion channel (VDAC) have been reported to increase cell death resistance under hypoxia/anoxia. In this work we propose a new, non-Mitchell mechanism of generation of the inner and outer membrane potentials at anaerobic conditions. The driving force is provided by the Gibbs free energy of the HK and CK reactions associated with the VDAC-HK and the ANT (adenine nucleotide translocator)-CK-VDAC complexes, respectively, both functioning as voltage generators. In the absence of oxygen, the cytosolic creatine phosphate can be directly used by the ANT-CK-VDAC contact sites to produce ATP from ADP in the mitochondrial matrix. After that, ATP released through the fraction of unbound ANTs in exchange for ADP is used in the mitochondrial intermembrane space by the outer membrane VDAC-HK electrogenic complexes to convert cytosolic glucose into glucose-6-phosphate. A simple computational model based on the application of Ohm's law to an equivalent electrical circuit showed a possibility of generation of the inner membrane potential up to -160mV, under certain conditions, and of relatively high outer membrane potential without wasting of ATP that normally leads to cell death. The calculated membrane potentials depended on the restriction of ATP/ADP diffusion in narrow cristae and through the cristae junctions. We suggest that high inner membrane potential and calcium extrusion from the mitochondrial intermembrane space by generated positive outer membrane potential prevent mitochondrial permeability transition, thus allowing the maintenance of mitochondrial integrity and cell survival in the absence of oxygen.

Competitive interactions of amphipathic polycationic peptides and cationic fluorescent probes with lipid membrane: experimental approaches and computational model.

The electrostatic interaction of polycationic peptides with negatively charged biomembranes has been recognized as the first and very important step of their selective binding to many bacteria and transformed cells. In this work we demonstrated the phenomenon of competition of some earlier designed polycationic peptides and fluorescent probes for their binding to the negatively charged inner membrane of mitochondria and to the PC/PG (9:1) liposomes. Rat liver mitochondria swelling induced by the antimicrobial polycationic peptide BTM-P1 (VAPIAKYLATALAKWALKQGFAKLKS) and by the retro-BTM-P1 was significantly diminished in the presence of 10µM fluorescent probe safranin O. In experiments with liposomes, the polycationic peptides BTM-P1 and P7-5 (IYLATALAKWALKQGF-GG-RRRRRRR) at the concentrations of 2-3µM completely displaced the membrane-bound fluorescent probe DiSC3(5) in a low ionic strength medium. The developed computational model allowed a mathematical description of such interactions, predicting membrane surface concentrations of bound peptides as the function of the membrane surface charge and lipid quantity in the sample, the peptide charge, hydrophobicity and concentration, the ionic strength of incubation medium and of the presence of a charged fluorescent probe used for monitoring the membrane surface potential under real-time peptide-membrane interactions.

VDAC electronics: 1. VDAC-hexo(gluco)kinase generator of the mitochondrial outer membrane potential.

The simplest mechanism of the generation of the mitochondrial outer membrane potential (OMP) by the VDAC (voltage-dependent anion channel)-hexokinase complex (VHC), suggested earlier, and by the VDAC-glucokinase complex (VGC), was computationally analyzed. Even at less than 4% of VDACs bound to hexokinase, the calculated OMP is high enough to trigger the electrical closure of VDACs beyond the complexes at threshold concentrations of glucose. These results confirmed our previous hypothesis that the Warburg effect is caused by the electrical closure of VDACs, leading to global restriction of the outer membrane permeability coupled to aerobic glycolysis. The model showed that the inhibition of the conductance and/or an increase in the voltage sensitivity of a relatively small fraction of VDACs by factors like tubulin potentiate the electrical closure of the remaining free VDACs. The extrusion of calcium ions from the mitochondrial intermembrane space by the generated OMP, positive inside, might increase cancer cell resistance to death. Within the VGC model, the known effect of induction of ATP release from mitochondria by accumulated glucose-6-phosphate in pancreatic beta cells might result not only of the known effect of GK dissociation from the VDAC-GK complex, but also of a decrease in the free energy of glucokinase reaction, leading to the OMP decrease and VDAC opening. We suggest that the VDAC-mediated electrical control of the mitochondrial outer membrane permeability, dependent on metabolic conditions, is a fundamental physiological mechanism of global regulation of mitochondrial functions and of cell death.

Electrical potentiation of the membrane permeabilization by new peptides with anticancer properties.

New polycationic peptides were designed on the basis of 16-mer and 14-mer fragments of the peptide BTM-P1, derived from the Cry11Bb protoxin. The peptides caused mitochondrial, but not red blood cell membrane permeabilization. Conjugation of the cell penetrating hepta-arginine vector to their N- or C-termini through two glycine residues resulted in more active peptides, which also permeabilized the red blood cells with a relatively high plasma membrane potential generated in the presence of valinomycin. The efficiency of the peptides was remarkably higher in the lower ionic strength media. The capability of the plasma membrane permeabilization of the normal red blood cells by the designed conjugated peptides and by known anticancer peptide R7-KLA was also strongly potentiated by the external electrical pulses applied to the cell suspension. These results open the new avenues of the local destruction of solid tumors using the combined "peptide--electrical pulses" synergistic treatment. The designed peptides were active against the human leukemia Jurkat cells but not against the normal wild type CHO cells.

Electrical hypothesis of toxicity of the Cry toxins for mosquito larvae.

Many electrical properties of insect larval guts have been studied, but their importance for toxicity of the Cry-type toxins has never been reported in the literature. In the present work, we observed potential-dependent permeabilization of plasma membrane by several polycationic peptides derived from the Cry11Bb protoxin. The peptide BTM-P1d, all D-type amino acid analogue of the earlier reported peptide BTM-P1, demonstrated high membrane-permeabilizing activity in experiments with isolated rat liver mitochondria, RBC (red blood cells) and mitochondria in homogenates of Aedes aegypti larval guts. Two larger peptides, BTM-P2 and BTM-P3, as well as the Cry11Bb protoxin treated with the protease extract of mosquito larval guts showed similar effects. Only protease-resistant BTM-P1d, in comparison with other peptides, displayed A. aegypti larval toxicity. Taking into account the potential-dependent mechanism of membrane permeabilization by studied fragments of the Cry11Bb protoxin and the literature data related to the distribution of membrane and transepithelial potentials in the A. aegypti larval midgut, we suggest an electrical hypothesis of toxicity of the Cry toxins for mosquito larvae. According to this hypothesis, the electrical field distribution is one of the factors determining the midgut region most susceptible for insertion of activated toxins into the plasma membrane to form pores. In addition, potential-dependent penetration of short active toxin fragments into the epithelial cells could induce permeabilization of mitochondria and subsequent apoptosis or necrosis.

Redox state-dependent aggregation of mitochondria induced by cytochrome c.

Cytochrome c is known to play central role in apoptosis. Here, it is shown that ferricytochrome c, but not ferrocytochrome c is able to directly induce the aggregation of rat liver mitochondria, similar to the effect caused by magnesium ions at high concentrations. The aggregation was revealed by a decrease in light dispersion of mitochondrial suspension and it was confirmed by the optical microscopy. In the medium containing NADH and cytochrome c, mitochondrial aggregation was initiated only after exhaustion of NADH leading to oxidation of cytochrome c. The aggregation induced by 30 µM ferricytochrome c, but not by 5 mM MgCl(2), was completely inhibited by 30-100 µM ferricyanide, thus indicating that ferricyanide-cytochrome c specific interaction prevents mitochondrial aggregation. After completion of the aggregation caused by ferricytochrome c, this effect cannot be readily reversed by subsequent reduction of cytochrome c. The aggregation induced by ferricytochrome c and/or magnesium ions explains masking of the external NADH-oxidase activity of mitochondria in vitro reported in the literature. This new cytochrome c redox state-dependent phenomenon might also be involved in more complex mechanisms controlling aggregation (clustering) of mitochondria in vivo under the influence of pro-apoptotic factors and requires further study.

Permeabilization of mitochondria and red blood cells by polycationic peptides BTM-P1 and retro-BTM-P1.

Mitochondrial and plasma membrane permeabilization by polycationic peptides BTM-P1 and retro-BTM-P1 were studied. BTM-P1 was more active than its retro-analog. In the sucrose medium, the capacity of BTM-P1 to permeabilize mitochondria was lower than in salt media. In contrast, retro-BTM-P1 showed the lowest activity in the KCl medium. The efficacy of both peptides to permeabilize red blood cells was higher in the sucrose medium and depended on the nature of salt in high ionic strength media. BTM-P1, but not retro-BTM-P1, induced biphasic change in light dispersion of red blood cells with artificially generated high transmembrane potential: the initial phase of fast cell shrinkage preceded the subsequent phase of cell swelling. The shrunken red blood cells demonstrated increased sensitivity to BTM-P1 that might be explained by the cell suicide mechanism via phosphatidylserine exposure at the cell surface. As a working hypothesis, we assume that some peptide topology characteristics, such as the orientation and values of the total and local electrical dipole moments, interacting with the membrane dipole potential, as well as the asymmetric distribution of polar and non-polar side chains are important factors affecting the membrane-permeabilizing activity of polycationic peptides.

Red blood cell permeabilization by hypotonic treatments, saponin, and anticancer avicins.

Plasma membrane permeabilization by saponin and anticancer avicins was studied using light dispersion measurements, since high correlation between light dispersion changes and hemolysis has been demonstrated. Nevertheless, we observed that rat red blood cell swelling in moderately hypotonic media was accompanied by up to 20% decrease of light dispersion, when hemolysis was not yet detectable. Avicin G and avicin D were significantly more efficient than saponin in inducing cytotoxicity in PC3 human prostate cancer cells. We found that the preincubation of avicins with the plasma membrane, but not with the cytosolic fraction of previously lysed red blood cells, completely protected fresh cells against permeabilization. The data suggest that the plasma membrane can tightly bind the avicins, but not the saponin. Using the "osmotic protection" method with 100mOsm PEGs of increasing molecular weight in isotonic media, the size of the pores generated by avicin G and avicin D in the plasma membrane was estimated to be higher than the hydrodynamic radius of PEG-8000. The obtained results indicate that the anticancer activity of avicin G and avicin D could be related, at least partially, to their high ability to permeabilize biological membranes. These data might represent interest for possible applications of these anticancer drugs in vivo.

Potential-dependent membrane permeabilization and mitochondrial aggregation caused by anticancer polyarginine-KLA peptides.

The anticancer activity of the polycationic peptide (KLAKLAK)(2), as a possible mitochondria-damaging agent, named KLA (l-form) or kla (d-form), has been increased by the fusion with hepta-arginine cell delivery vectors r7 and R7 (peptides r7-kla and R7-KLA, respectively), as shown in the literature. We demonstrated that 3.6muM r7-kla or R7-KLA, but not kla, caused significant permeabilization of the inner and the outer membranes of energized rat liver mitochondria. In addition, r7-kla or R7-KLA induced mitochondrial aggregation, thus causing the inhibition of metabolic activity. Potential-dependent mechanism of permeabilization of the inner mitochondrial membrane by these peptides was also observed for the plasma membrane of red blood cells. The obtained results suggest that polyarginine cell delivery vectors of anticancer polycationic peptides not only increase their direct potential-dependent permeabilization of biological membranes, but also create the capacity to cause aggregation of mitochondria, as a new mechanism of cytotoxic action of these peptides.

Potential-dependent permeabilization of plasma membrane by the peptide BTM-P1 derived from the Cry11Bb1 protoxin.

The peptide BTM-P1, which is derived from the amino acid sequence of the Cry11Bb1 protoxin, is able to permeabilize mitochondrial membranes and reveals antimicrobial activity. In this work we demonstrated that the permeabilizing activity of BTM-P1 for the plasma membrane of rat red blood cells increased in a dose-dependent manner for the concentration range of 1-4 microg/ml. Using osmotic protectants, the radius of pores formed at 4 microg/ml BTM-P1 was determined as 0.8 nm for 5 min hemolysis data, 0.7 nm for 5 min decrease in light dispersion of the cell suspension and 0.5 nm for the light dispersion slope measurements. The permeabilizing activity of 1 microg/ml peptide was increased by valinomycin-induced plasma membrane potential, especially under moderately hypotonic conditions. These results might explain the antimicrobial activity of BTM-P1 and support the hypothesis of potential-dependent and pro-apoptotic character of toxicity of naturally proteolysed Cry11Bb1 protoxin for epithelial cells of mosquito larvae midgut.

Hemoglobin precipitation by polyethylene glycols leads to underestimation of membrane pore sizes.

The size of pores formed in the plasma membrane by various substances is frequently determined using polyethylene glycols as osmotic protectants. In this work, we have found that the size of pores formed by saponin in the red blood cell membrane determined by hemolysis versus molecular weight of polyethylene glycol was different to that estimated by light dispersion of cell suspensions. After complete swelling of cells induced by saponin in semiisotonic salt media containing 150 mOsm PEG-4000 or PEG-3000, a significant increase in the light absorbance at 640 nm was developed resulting from the formation of hemoglobin precipitates. Easily sedimenting aggregates were also formed when the supernatant of lysed cells was added to the equiosmotic solutions of polyethylene glycols with molecular weight higher than 1000. We suggest that the real size of large pores could be underestimated due to the phenomenon of hemoglobin precipitation by polyethylene glycols.

Synergistic inhibition of mitochondrial respiration by anticancer agent erucylphosphohomocholine and cyclosporin A.

Alkylphosphocholines are a new class of anticancer agents. The mechanisms by which these drugs display their antitumor activities are not known. In this work, we show that erucylphosphohomocholine, a new antineoplastic compound, significantly decreased ATP synthesis in isolated rat liver mitochondria at a concentration of 50 microm or higher via permeabilization of the inner membrane. At a concentration of 25 microm, it induced a moderate swelling of mitochondria, a slight decrease of the inner membrane potential, and an increase in state 4 respiration without an essential influence on state 3 respiration or the outer membrane permeability to cytochrome c. We found that cyclosporin A did not prevent mitochondrial swelling induced by 25-100 microm erucylphosphohomocholine. Moreover, cyclosporin A induced a fast drop of the inner membrane potential in the presence of 25-50 microm erucylphosphohomocholine that seems to be due to a strong synergistic inhibition of the respiratory activity. The ratio of uncoupled to state 3 respiration rates increased from 1.3 +/- 0.1 with 25 microm erucylphosphohomocholine and from 1.5 +/- 0.1 with 1 microm cyclosporin A to 4.5 +/- 0.3 in the presence of both drugs. On the other hand, oligomycin or cyclosporin A protected certain cancer cell lines against erucylphosphohomocholine-induced apoptosis. This protection might be related to a prevention of cellular ATP hydrolysis by permeabilized mitochondria and to the inhibition of the classical permeability transition pore, respectively. Our findings provide new insight into the mechanisms by which these unusual alterations of mitochondria might be involved in anticancer activity of alkylphosphocholines.