Portal de Periódicos da CAPES

Bid, but Not Bax, Regulates VDAC Channels

2004; Elsevier BV; Volume: 279; Issue: 14 Linguagem: Inglês

10.1074/jbc.m310593200

1083-351X

Tatiana K. Rostovtseva, Bruno Antonsson, Motoshi Suzuki, Richard J. Youle, Marco Colombini, Sergey M. Bezrukov,

ATP Synthase and ATPases Research

During apoptosis, cytochrome c is released from mitochondria into the cytosol, where it participates in caspase activation. Various and often conflicting mechanisms have been proposed to account for the increased permeability of the mitochondrial outer membrane that is responsible for this process. The voltage-dependent anion channel (VDAC) is the major permeability pathway for metabolites in the mitochondrial outer membrane and therefore is a very attractive candidate for cytochrome c translocation. Here, we report that properties of VDAC channels reconstituted into planar phospholipid membranes are unaffected by addition of the pro-apoptotic protein Bax under a variety of conditions. Contrary to other reports (Shimizu, S., Narita, M., and Tsujimoto, Y. (1999) Nature 399, 483-487; Shimizu, S., Ide, T., Yanagida, T., and Tsujimoto, Y. (2000) J. Biol. Chem. 275, 12321-12325; Shimizu, S., Konishi, A., Kodama, T., and Tsujimoto, Y. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 3100-3105), we found no electrophysiologically detectable interaction between VDAC channels isolated from mammalian mitochondria and either monomeric or oligomeric forms of Bax. We conclude that Bax does not induce cytochrome c release by acting on VDAC. In contrast to Bax, another pro-apoptotic protein (Bid) proteolytically cleaved with caspase-8 affected the voltage gating of VDAC by inducing channel closure. We speculate that by decreasing the probability of VDAC opening, Bid reduces metabolite exchange between mitochondria and the cytosol, leading to mitochondrial dysfunction. During apoptosis, cytochrome c is released from mitochondria into the cytosol, where it participates in caspase activation. Various and often conflicting mechanisms have been proposed to account for the increased permeability of the mitochondrial outer membrane that is responsible for this process. The voltage-dependent anion channel (VDAC) is the major permeability pathway for metabolites in the mitochondrial outer membrane and therefore is a very attractive candidate for cytochrome c translocation. Here, we report that properties of VDAC channels reconstituted into planar phospholipid membranes are unaffected by addition of the pro-apoptotic protein Bax under a variety of conditions. Contrary to other reports (Shimizu, S., Narita, M., and Tsujimoto, Y. (1999) Nature 399, 483-487; Shimizu, S., Ide, T., Yanagida, T., and Tsujimoto, Y. (2000) J. Biol. Chem. 275, 12321-12325; Shimizu, S., Konishi, A., Kodama, T., and Tsujimoto, Y. (2000) Proc. Natl. Acad. Sci. U. S. A. 97, 3100-3105), we found no electrophysiologically detectable interaction between VDAC channels isolated from mammalian mitochondria and either monomeric or oligomeric forms of Bax. We conclude that Bax does not induce cytochrome c release by acting on VDAC. In contrast to Bax, another pro-apoptotic protein (Bid) proteolytically cleaved with caspase-8 affected the voltage gating of VDAC by inducing channel closure. We speculate that by decreasing the probability of VDAC opening, Bid reduces metabolite exchange between mitochondria and the cytosol, leading to mitochondrial dysfunction. The crucial role of mitochondria in the initiation of apoptosis is well established. Triggered by a number of different stimuli, the mitochondrial outer membrane (MOM) 1The abbreviations used are: MOM, mitochondrial outer membrane; BH, Bcl-2 homology domain; tBid, truncated Bid; BaxFL, full-length Bax; tcBid, 15.5-kDa C-terminal fragment of caspase-8-cleaved Bid; DPhPC, diphytanoylphosphatidylcholine; nS, nanosiemens; PTP, permeability transition pore. becomes permeable to apoptogenic factors such as cytochrome c and Smac/DIABLO (4Du C. Fang M. Li Y. Li L. Wang X. Cell. 2000; 102: 33-42Google Scholar, 5Verhagen A.M. Ekert P.G. Pakusch M. Silke J. Connolly L.M. Reid G.E. Moritz R.L. Simpson R.J. Vaux D.L. Cell. 2000; 102: 45-53Google Scholar). The release of these factors leads to caspase activation, DNA fragmentation, and other characteristic changes associated with apoptotic cell death. So far, it remains unclear exactly how MOM permeabilization occurs, and published results are often contradictory (for reviews, see Refs. 6Desagher S. Martinou J.-C. Trends Cell Biol. 2000; 10: 369-377Google Scholar, 7Martinou J.-C. Green D.R. Nat. Cell Biol. 2001; 2: 63-67Google Scholar, 8Waterhouse N.J. Ricci J.-E. Green D.R. Biochimie (Paris). 2002; 84: 113-121Google Scholar). The Bcl-2 family of proteins regulates the permeabilization of MOM. Pro-apoptotic proteins such as Bax and Bid induce the release of apoptogenic factors, whereas anti-apoptotic proteins such as Bcl-2 and Bcl-xL prevent their release. There are two prevailing theories for the permeabilization of the outer membrane. One is that MOM is nonspecifically ruptured; the other proposes the formation of channels that allow cytochrome c release (6Desagher S. Martinou J.-C. Trends Cell Biol. 2000; 10: 369-377Google Scholar, 7Martinou J.-C. Green D.R. Nat. Cell Biol. 2001; 2: 63-67Google Scholar). One of the channel models is based on the ability of Bax to form channels when incorporated into lipid membranes (9Antonsson B. Conti F. Ciavatta A.M. Montessuit S. Lewis S. Martinou I. Bernasconi L. Bernard A. Mermod J.-J. Mazzei G. Maundrell K. Gambale F. Sadoul R. Martinou J.-C. Science. 1997; 277: 370-372Google Scholar, 10Schlesinger P.H. Gross A. Yin X.M. Yamamoto K. Saito M. Waksman G. Korsmeyer S.J. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 11357-11362Google Scholar). It has been reported that Bax can form oligomers of multiple sizes that have features of ion channels large enough to release cytochrome c (11Antonsson B. Montessuit S. Lauper S. Eskes R. Martinou J.-C. Biochem. J. 2000; 345: 271-278Google Scholar). Soluble monomeric Bax is not capable of forming channels in lipid membranes and does not trigger the release of cytochrome c from isolated mitochondria (11Antonsson B. Montessuit S. Lauper S. Eskes R. Martinou J.-C. Biochem. J. 2000; 345: 271-278Google Scholar). Thus, channel formation is associated with the formation of oligomers. Bax oligomerization has been proposed to be triggered by the BH3 domain-only protein Bid after cleavage by caspase-8 into the truncated form, tBid (12Wei M.C. Lindsten T. Mootha V.K. Weiler S. Gross A. Ashiya M. Thompson C.B. Korsmeyer S.J. Genes Dev. 2000; 14: 2060-2071Google Scholar, 13Eskes R. Desagher S. Antonsson B. Martinou J.-C. Mol. Cell. Biol. 2000; 20: 929-935Google Scholar). However, the situation has been complicated by a recent report that tBid itself homopolymerizes, and this process, without the participation of Bax, results in the release of cytochrome c from mitochondria (14Grinberg M. Sarig R. Zaltsman Y. Frumkin D. Grammatikakis N. Reuveny E. Gross A. J. Biol. Chem. 2002; 277: 12237-12245Google Scholar). In addition, it has been observed that tBid promotes leakage in planar lipid membranes and liposomes in the absence of other proteins (15Schendel S.L. Azimov R. Pawlowski K. Godzik A. Kagan B.L. Reed J.C. J. Biol. Chem. 1999; 274: 21932-21936Google Scholar, 16Zhai D. Huang X. Han X. Yang F. FEBS Lett. 2000; 472: 293-296Google Scholar, 17Kudla G. Montessuit S. Eskes R. Berrier C. Martinou J.-C. Ghazi A. Antonsson B. J. Biol. Chem. 2000; 275: 22713-22718Google Scholar). In all these studies, the channel/pore-forming ability was attributed to Bax and Bid; however, it is still an open question whether such channels are formed in vivo. There are two other channel models that associate the Bcl-2 family proteins with the existing major channel in MOM, the voltage-dependent anion channel (VDAC). This channel is known to be responsible for most of the metabolite flux across this membrane (18Colombini M. Curr. Top. Membr. 1994; 42: 73-101Google Scholar, 19Rostovtseva T. Colombini M. J. Biol. Chem. 1996; 271: 28006-28008Google Scholar, 20Rostovtseva T. Colombini M. Biophys. J. 1997; 72: 1954-1962Google Scholar, 21Hodge T. Colombini M. J. Membr. Biol. 1997; 157: 271-279Google Scholar). Thus, VDAC is a very attractive candidate for a pathway for cytochrome c release. When reconstituted into planar membranes, the 30-kDa VDAC monomer forms an aqueous pore 2.5-3 nm in diameter (18Colombini M. Curr. Top. Membr. 1994; 42: 73-101Google Scholar, 22Mannella C.A. Guo X.W. Cognon B. FEBS Lett. 1989; 253: 231-234Google Scholar, 23Song J. Midson C. Blachly-Dyson E. Forte M. Colombini M. J. Biol. Chem. 1998; 273: 24406-24413Google Scholar) that allows uncharged polymers (such as inulin, dextran, and polyethylene glycol with a molecular mass of ∼5000 Da) to cross membranes (24Colombini M. J. Membr. Biol. 1980; 53: 79-84Google Scholar, 25Zalman L.S. Nikaido H. Kagawa Y. J. Biol. Chem. 1980; 255: 1771-1774Google Scholar). However, the apparent size of the channel is very different for charged molecules. The channel even discriminates between different molecules based on their three-dimensional structure (26Rostovtseva T.K. Komarov A. Bezrukov S.M. Colombini M. J. Membr. Biol. 2002; 184: 147-156Google Scholar). Negatively charged anions such as ATP are favored due to the net positive charge within the channel. VDAC is permeable to anionic metabolites in the open state, but is essentially impermeable to anions such as ATP and phosphocreatine when the channel adopts a closed configuration, and the channel selectivity is reversed from favoring anions to favoring cations (19Rostovtseva T. Colombini M. J. Biol. Chem. 1996; 271: 28006-28008Google Scholar, 20Rostovtseva T. Colombini M. Biophys. J. 1997; 72: 1954-1962Google Scholar, 27Vander Heiden M.G. Li X.X. Gottleib E. Hill R.B. Thompson C.B. Colombini M. J. Biol. Chem. 2001; 276: 19414-19419Google Scholar). Certainly, the positively charged cytochrome c molecules (12 kDa) cannot permeate through this channel under normal conditions. If they did, cytochrome c would not be localized within the intermembrane space. Vander Heiden et al. (27Vander Heiden M.G. Li X.X. Gottleib E. Hill R.B. Thompson C.B. Colombini M. J. Biol. Chem. 2001; 276: 19414-19419Google Scholar, 28Vander Heiden M.G. Chandel N.S. Li X.X. Schumacker P.T. Colombini M. Thompson C.B. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 4666-4671Google Scholar) proposed a model of MOM permeabilization that involves regulation of VDAC channels. In this model, VDAC channel closure prevents the efficient exchange of ATP and ADP between the cytosol and the mitochondrial matrix. Loss of outer membrane permeability due to VDAC closure might result in the accumulation of the products of mitochondrial activity within the intermembrane space, generation of an osmotic gradient, and matrix swelling, followed by the rupture of the outer membrane. The anti-apoptotic protein Bcl-xL restores the ATP/ADP exchange. Experiments with VDAC channels reconstituted into planar phospholipid membranes demonstrated that Bcl-xL promotes the open configuration of mammalian VDAC channels (27Vander Heiden M.G. Li X.X. Gottleib E. Hill R.B. Thompson C.B. Colombini M. J. Biol. Chem. 2001; 276: 19414-19419Google Scholar). Another channel model postulates that the pro-apoptotic protein Bax directly interacts with VDAC, resulting in cytochrome c permeation through membranes (1Shimizu S. Narita M. Tsujimoto Y. Nature. 1999; 399: 483-487Google Scholar, 2Shimizu S. Ide T. Yanagida T. Tsujimoto Y. J. Biol. Chem. 2000; 275: 12321-12325Google Scholar, 3Shimizu S. Konishi A. Kodama T. Tsujimoto Y. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 3100-3105Google Scholar). It has been reported that experiments with liposomes and planar phospholipid membranes show that Bax induces a novel VDAC-containing channel that is larger than the channels observed when the proteins are used alone. It was concluded that, although the channels formed by Bax and VDAC alone are unable to allow cytochrome c translocation, the new larger pores are permeable to cytochrome c (2Shimizu S. Ide T. Yanagida T. Tsujimoto Y. J. Biol. Chem. 2000; 275: 12321-12325Google Scholar). In both models, the ability of Bcl-2 proteins to regulate the state and integrity of VDAC could thus account for their ability to be either anti- or pro-apoptotic. However, the mechanisms of VDAC channel regulation by Bcl-2 proteins proposed in these models are diametrically opposite. In the first model, the closure of VDAC channels leads to MOM rupture, and anti-apoptotic protein helps to maintain VDAC channels in their normal, functional open state. In the second model, the pro-apoptotic protein induces opening of the large VDAC-based channels permeable to cytochrome c. The mechanism of MOM permeabilization is undoubtedly very complex and most likely involves more than one protein to form the passages for cytochrome c release. Ceramides have also been proposed to form this permeability pathway (29Siskind L.J. Colombini M. J. Biol. Chem. 2000; 275: 38640-38644Google Scholar). The role of VDAC in this process is rather controversial; but considering that both BH1-3 proteins such as Bax and BH3 domain-only proteins such as Bid are most likely required for MOM permeabilization, it is important to test the effect of both pro-apoptotic proteins on VDAC channels. To address this question, we studied the effect of the proapoptotic proteins Bax and Bid on the properties of VDAC channels reconstituted into planar phospholipid membranes. We used the full-length Bax protein in the soluble monomeric and oligomeric forms and Bid cut with caspase-8 (referred to as cut Bid). Here, we show that the properties of VDAC channels isolated from mammalian mitochondria are unaffected by addition of Bax. We conclude that Bax does not induce cytochrome c release by acting on VDAC. In contrast, we demonstrate that cut Bid induces the closure of VDAC channels. By reducing the permeability of VDAC channels, the pro-apoptotic protein Bid might interfere with metabolite exchange between mitochondria and the cytosol, with the subsequent loss of outer membrane integrity. Protein Purification—VDAC from rat liver mitochondria was isolated and purified as described previously (30Parsons D.F. Williams G.R. Chance B. Ann. N. Y. Acad. Sci. 1966; 137: 643-666Google Scholar, 31Freitag H. Benz R. Neupert W. Methods Enzymol. 1983; 97: 286-294Google Scholar, 32Blachly-Dyson E. Peng S.Z. Colombini M. Forte M. Science. 1990; 247: 1233-1236Google Scholar). VDAC from Neurospora crassa MOMs was obtained using the standard procedure of isolation (31Freitag H. Benz R. Neupert W. Methods Enzymol. 1983; 97: 286-294Google Scholar, 33Mannella C.A. J. Cell Biol. 1982; 94: 680-687Google Scholar). Human recombinant wild-type monomeric full-length Bax (BaxFL) was obtained using two different protocols (34Suzuki M. Youle R.J. Tjandra N. Cell. 2000; 103: 645-654Google Scholar, 35Desagher S. Osen-Sand A. Nichols A. Eskes R. Montessuit S. Lauper S. Maundrell K. Antonsson B. Martinou J.-C. J. Cell Biol. 1999; 144: 891-901Google Scholar). The first sample of monomeric Bax was purified by a previously described procedure (34Suzuki M. Youle R.J. Tjandra N. Cell. 2000; 103: 645-654Google Scholar); Escherichia coli BL21(DE3) cells harboring pTYB1-Bax were cultured in LB medium; recombinant proteins were isolated from the cytosol by chitin affinity chromatography according to the protocol provided by the manufacturer (New England Biolabs Inc.) and further purified by ionexchange chromatography on a mono-Q column (Amersham Biosciences). The second BaxFL sample with a tag of six histidines at the N-terminus was expressed in the pBAD plasmid in E. coli (36Montessuit S. Mazzei G. Magnenat E. Antonsson B. Protein Expression Purif. 1999; 15: 202-206Google Scholar). Monomeric Bax was recovered in the soluble bacterial fraction and purified by chromatography on nickel-nitrilotriacetic acid-agarose, followed by Q-Sepharose. The protein was stored in 25 mm Tris-HCl, 100 mm NaCl, 0.2 mm dithiothreitol, and 30% (v/v) glycerol (pH 7.5) at -80 °C. Oligomeric BaxFL with an N-terminal His tag was expressed in and purified from E. coli (11Antonsson B. Montessuit S. Lauper S. Eskes R. Martinou J.-C. Biochem. J. 2000; 345: 271-278Google Scholar). The protein was recovered in the soluble fraction in the presence of 1% Triton X-100 and purified on nickel-nitrilotriacetic acid-agarose, followed by anion-exchange chromatography on a Q-Sepharose column. The purified protein was dialyzed against 25 mm HEPES-NaOH, 0.2 mm dithiothreitol, 1% (w/v) octyl glucoside, and 30% glycerol (pH 7.5) and stored at -80 °C. Other samples of oligomeric BaxFL were obtained by incubation of monomeric BaxFL with 1% octyl glucoside on ice for 30 min. Mouse full-length Bid was expressed with an N-terminal His6 tag in E. coli. Caspase-8-cleaved mouse wild-type Bid (cut Bid) was obtained from purified full-length Bid by cleavage with caspase-8 (37Chou J.J. Li H. Salvesen G.S. Yuan J. Wagner G. Cell. 1999; 96: 615-624Google Scholar, 38McDonnel J.M. Fushman D. Milliman C.L. Korsmeyer S.J. Cowburn D. Cell. 1999; 96: 625-634Google Scholar). Cut Bid was purified on a nickel affinity column to remove caspase-8 (39Antonsson B. Montessuit S. Sanchez B. Martinou J.-C. J. Biol. Chem. 2001; 276: 11615-11623Google Scholar). No caspase-8 activity was detected in the purified cut Bid sample. The resulting 15.5-kDa C-terminal fragment (tcBid, amino acids 60-195) was expressed as an N-terminal His6-tagged protein in E. coli (17Kudla G. Montessuit S. Eskes R. Berrier C. Martinou J.-C. Ghazi A. Antonsson B. J. Biol. Chem. 2000; 275: 22713-22718Google Scholar). Electrophysiological Recordings—Planar lipid membranes were formed from monolayers made from 1% (w/v) lipids in hexane on 70-80-μm diameter orifices in the 15-μm-thick Teflon partition that separated two chambers (modified after (40Montal M. Mueller P. Proc. Natl. Acad. Sci. U. S. A. 1972; 69: 3561-3566Google Scholar)). The lipid-forming solutions contained diphytanoylphosphatidylcholine (DPhPC); 90% asolectin (soybean phospholipids) and 10% cholesterol; or 42% asolectin, 42% DPhPC, 8% cardiolipin, and 8% cholesterol. DPhPC, asolectin, and cardiolipin were purchased from Avanti Polar Lipids (Alabaster, AL), and cholesterol was purchased from Sigma. The membrane potential was maintained using Ag/AgCl electrodes with 3 m KCl and 15% (w/v) agarose bridges (41Bezrukov S.M. Vodyanoy I. Biophys. J. 1993; 64: 16-25Google Scholar). VDAC channels insertion was achieved by adding 0.1-1.5 μl of a 1% Triton X-100 solution of purified VDAC to the 1.5-ml aqueous phase in the cis-compartment while stirring. Potential is defined as positive when it is greater at the side of VDAC addition (cis). The effect of the pro-apoptotic proteins on the VDAC properties was determined as follows. After VDAC channels were inserted and their parameters were monitored for at least 20-25 min, the pro-apoptotic protein was added to both sides of the membrane (if not indicated otherwise) under constant stirring for 2 min. The currents were recorded for at least 25 min, followed by a series of additions of the protein until a rise in conductance induced by the pro-apoptotic protein was obtained. Conductance measurements were performed using an Axopatch 200B amplifier (Axon Instruments, Inc., Foster City, CA) in the voltage clamp mode. Data were filtered by a low-pass 8-pole Butterworth filter (Model 9002, Frequency Devices Inc., Haverhill, MA) at 15 kHz, recorded on a chart recorder, and directly saved into the computer memory with a sampling frequency of 50 kHz. The membrane chamber and headstage were isolated from external noise sources with a double metal screen (Amuneal Manufacturing Corp., Philadelphia, PA). The voltage-dependent properties of the VDAC-containing membrane were assessed by applying a symmetrical 5-mHz triangular voltage wave from a Hewlett-Packard 33120A function waveform generator and recording the current using a Digidata 1322A (Axon Instruments, Inc.). Data were acquired at a sampling frequency of 1 Hz and analyzed using pClamp 8 software (Axon Instruments, Inc.). VDAC and Bax—One of the characteristic properties of VDAC is its voltage gating. The channels are in the high conductance open state at low voltages (<30 mV) and move to low conductance states ("closed" states) at high voltages (Fig. 1A, inset). The application of 50 mV causes the conductance of the single channel to drop by two-thirds of its open state level. A short exposure to 0 mV reopens the channel. After the conductance and voltage gating properties of the channel were measured, monomeric Bax was added to the membrane-bathing aqueous solutions on one or both sides of the membrane. In most experiments (7 of 10), we observed that monomeric Bax addition caused a 4-6-fold increase in membrane conductance. However, further analysis showed that this increase in membrane conductance was due to the insertion of additional VDAC channels and not to the formation of a novel larger channel. Fig. 1B illustrates the stepwise conductance decrease as individual channels closed in the presence of monomeric Bax. The individual steps (Fig. 1B, inset) are of the same size as in control records (Fig. 1A, inset). After 0 mV was applied, the channels reopened and closed again under a 50 mV applied potential. All this is normal VDAC behavior. These channels were still sensitive to König polyanion that induced their closure (42Colombini M. Yeung C.L. Tung J. König T. Biochim. Biophys. Acta. 1987; 905: 279-286Google Scholar) (data not shown). Sizing with polyethylene glycols (43Rostovtseva T.K. Nestorovich E.M. Bezrukov S.M. Biophys. J. 2002; 82: 160-169Google Scholar) gave similar results as for single VDAC channels. The conclusion that the observed increase in conductance was indeed the result of new insertion and not formation of larger channels (as previously reported (2Shimizu S. Ide T. Yanagida T. Tsujimoto Y. J. Biol. Chem. 2000; 275: 12321-12325Google Scholar)) is supported by the observation that perfusion of the chamber to remove excess VDAC prior to Bax addition resulted in the absence of the Bax-dependent conductance increment. Other characteristic properties of VDAC such as single channel conductance and selectivity were probed in the presence of monomeric Bax. No significant changes in these properties were observed (Table I). In principle, many additional factors may also influence Bax-VDAC interaction. The possibility that membrane lipids (44Basañez G. Sharpe J.C. Galanis J. Brandt T.B. Hardwick J.M. Zimmerberg J. J. Biol. Chem. 2002; 277: 49360-49365Google Scholar), especially charged lipids, might affect the observations was assessed using both the neutral DPhPC and a negatively charged natural mixture, asolectin. No effect of Bax on VDAC was observed with either of these lipids (Table I). The possibility that acidic pH might render the terminal α-helix of Bax accessible for interaction with VDAC was tested by performing the experiments at both neutral and acidic pH values. Ionic strength was also varied, as was the presence of either Ca2+ or Mg2+; and no influence of Bax on the properties of VDAC was observed under any of these conditions (Table I).Table ISummary of the experiments with VDAC channels and BaxBax sampleLipidMediumpHMe2+ (1 mm)Single channel conductanceSelectivityVoltage gatingMonomeric 16—3000 nmDPhPC1 m NaCl (12)aThe number of experiments is indicated in parentheses or 0.1 m7.0 (7), 5.0 (6)Ca2+ (13)NEbNE, no effect (13)KCl (1) 20—130 nmAsolectin1 m NaCl (6) or 0.1 m5.5 (8)Ca2+ (8)NE (1)NE (1)NE (7)NaCl (2)Oligomericwith 1% OGcOligomeric Bax was obtained by treating monomeric Bax with 1% octyl glucoside (OG) 30—60 nmAsolectin0.25 m NaCl (2), 0.25 m5.5 (7)Mg2+ (7)NE (3)NE (3)NE (4) [∼90 nmdThe concentration at which Bax addition resulted in membrane rupture]KCl (4), or 0.1 m KCl (1)OligomericeOligomeric Bax was obtained by isolating Bax in the presence of detergents 2—10 nmAsolectin0.25/0.05 m NaCl (2),5.5 (4), 7.0 (1)Mg2+ (5)NE (2)NE (2)NE (4) [15—20dThe concentration at which Bax addition resulted in membrane rupture nm]0.25 m NaCl (1), or0.25 m KCl (2)a The number of experiments is indicated in parenthesesb NE, no effectc Oligomeric Bax was obtained by treating monomeric Bax with 1% octyl glucoside (OG)d The concentration at which Bax addition resulted in membrane rupturee Oligomeric Bax was obtained by isolating Bax in the presence of detergents Open table in a new tab The possibility that the physical state of Bax (monomeric or oligomeric) is important for interaction with VDAC has been tested using both forms. As reported previously (11Antonsson B. Montessuit S. Lauper S. Eskes R. Martinou J.-C. Biochem. J. 2000; 345: 271-278Google Scholar), we observed that only Bax oligomers formed channel-like conductances. Examples of the representative current traces induced by oligomeric Bax are presented in Fig. 2 (b and c). Monomeric Bax itself was inactive (up to 3 μm) at neutral pH (Fig. 2a). Oligomeric Bax was obtained either by isolating Bax in the presence of detergents or by treating monomeric Bax with 1% octyl glucoside (see "Experimental Procedures"). Oligomeric Bax samples from both preparations formed channels in bilayer membranes, but at different concentration ranges (Fig. 3). The former increased membrane conductance at lower concentrations (5-20 nm). The latter required 30-80 nm to increase membrane conductance. The distributions of individual conductance steps produced by oligomeric Bax were very broad, with conductances from hundreds of picosiemens to tens of nanosiemens (nS). In a typical experiment, oligomeric Bax produced channels with a gradual increase in conductance during the experiment (45Roucou X. Rostovtseva T. Montessuit S. Martinou J.-C. Antonsson B. Biochem. J. 2002; 363: 547-552Google Scholar). Often, after reaching the high total conductance level of 10-20 nS, the membrane ruptured. The decrease in the lifetime of the membranes in the presence of Bax proteins has been described previously (46Basañez G. Nechushtan A. Drozhinin O. Chanturiya A. Choe E. Tutt S. Wood K.A. Hsu Y.-T. Zimmerberg J. Youle R.J. Proc. Natl. Acad. Sci. U. S. A. 1999; 96: 5492-5497Google Scholar).Fig. 3Pro-apoptotic proteins increase membrane conductance at different concentrations. The symbols represent the average conductance values obtained in individual experiments 10-20 min after protein addition. The inset shows, at a finer scale, the concentration dependence for oligomeric Bax from two different preparations. Oligomeric Bax obtained by isolation in the presence of detergent (▴) increased membrane conductance at 5-20 nm, whereas oligomeric Bax obtained by treatment of monomeric Bax with 1% octyl glucoside (shaded boxes) increased conductance at 30-80 nm. Cut Bid (•) induced membrane conductance at concentrations that were higher than those of tcBid (○). Monomeric Bax (⋄) was inactive at all concentrations. The other experimental conditions were as described in the legend to Fig. 2. The lines were drawn to guide the eye through the experimental points.View Large Image Figure ViewerDownload (PPT) To discriminate between VDAC- and Bax-induced conductances, both oligomeric Bax preparations were used at concentrations below or at the threshold at which they increased the membrane conductance. The concentrations at which oligomeric Bax formed channels were variable (Fig. 3, inset). Therefore, in a typical experiment, after the VDAC channel(s) was reconstituted into the membrane and its properties were recorded, oligomeric Bax was added in a series starting with concentrations below those at which Bax was found to form channels until a visible increase in the conductance, due to the formation of Bax channels, was observed. The Bax channels were distinctly different from VDAC channels (compare the noisy current traces produced by Bax in Fig. 2 (b and c) with the well defined stable currents through VDAC channels in Figs. 1 and 4A), which helped to avoid a confusing situation when a total current was actually a sum of two different conductances with different properties. Our task was to find the Bax-induced changes in VDAC properties under the conditions in which Bax did not form channels by itself. Fig. 4A shows an example in which the selectivity of VDAC channel was assessed before and after oligomeric Bax addition. The reversal potential (measured in a 5-fold NaCl salt gradient) was +17 mV before and after Bax addition. The single channel conductance also remained the same with and without Bax. Fig. 4A shows no effect of oligomeric Bax on either selectivity or single channel conductance. The probability of VDAC channels to be open or closed at different potentials in the presence of Bax was estimated by obtaining G/V plots as shown in Fig. 4B. The data were collected as follows. Slow triangular voltage waves (5 mHz) were applied to a multichannel membrane, and the current was recorded. Because VDAC channels have rapid opening rates (inverse submilliseconds) and slow closing rates (inverse seconds), a near-equilibrium G/V plot could be obtained by collecting data when the applied voltage opened the channels, changing from ±60 to 0 mV. In the membranes containing many VDAC channels there was always a possibility that a portion of the conductance was due to Bax channels. The number of VDAC channels could also vary during the recording. To take these effects into account, the G/V plot is expressed as a normalized conductance with the voltage-independent portion subtracted (this would include any Bax conductance). This plot is essentially an expression of the voltage dependence of the open probability. Fig. 4B shows the results of a typical experiment. There was no effect of Bax on VDAC voltage dependence. It should be mentioned that, in the experiment shown in Fig. 4A, oligomeric Bax was obtained by preincubation of monomeric Bax with 1% octyl