Mediation of the Antiapoptotic Activity of Bcl-xL Protein upon Interaction with VDAC1 Protein*

Background: Bcl-xL is overexpressed in cancer, contributing to resistance to chemotherapy. Results: Bcl-xL directly interacts with VDAC1 to mediate its antiapoptotic activity, activity that can be prevented by VDAC1-based peptides. Conclusion: Bcl-xL regulates apoptosis through direct interaction with VDAC1. Significance: Interfering with the interaction of Bcl-xL with VDAC1 can lead to apoptosis and potentiate the efficacy of conventional chemotherapeutics. The mitochondrial protein, the voltage-dependent anion channel (VDAC), is implicated in the control of apoptosis, including via its interaction with the pro- and antiapoptotic proteins. We previously demonstrated the direct interaction of Bcl2 with VDAC, leading to reduced channel conductance. VDAC1-based peptides interacted with Bcl2 to prevent its antiapoptotic activity. Here, using a variety of approaches, we show the interaction of the antiapoptotic protein, Bcl-xL, with VDAC1 and reveal that this interaction mediates Bcl-xL protection against apoptosis. C-terminally truncated Bcl-xL(Δ21) interacts with purified VDAC1, as revealed by microscale thermophoresis and as reflected in the reduced channel conductivity of bilayer-reconstituted VDAC1. Overexpression of Bcl-xL prevented staurosporine-induced apoptosis in cells expressing native VDAC1 but not certain VDAC1 mutants. Having identified mutations in VDAC1 that interfere with the Bcl-xL interaction, certain peptides representing VDAC1 sequences, including the N-terminal domain, were designed and generated as recombinant and synthetic peptides. The VDAC1 N-terminal region and two internal sequences were found to bind specifically, and in a concentration- and time-dependent manner, to immobilized Bcl-xL(Δ21), as revealed by surface plasmon resonance. Moreover, expression of the recombinant peptides in cells overexpressing Bcl-xL prevented protection offered by the protein against staurosporine-induced apoptosis. These results point to Bcl-xL acting as antiapoptotic protein, promoting tumor cell survival via binding to VDAC1. These findings suggest that interfering with Bcl-xL binding to the mitochondria by VDAC1-based peptides may serve to induce apoptosis in cancer cells and to potentiate the efficacy of conventional chemotherapeutic agents.

posed target, the voltage-dependent anion channel (VDAC) (11,12). VDAC1, one of the three isoforms of the protein, has been characterized at the structural level using NMR, crystallography, or a combination of the two and was found to comprise a 19-stranded ␤-barrel and an N-terminal ␣-helical region located inside the pore (13)(14)(15). Purified bilayer-reconstituted VDAC1 shows voltage-dependent conductance. At low voltages (Ϯ10 mV), the VDAC1 channel is stable in a long-lived open state, whereas at high positive or negative potentials (Ͼ40 mV), VDAC1 presents multiple substates with different ionic selectivities and permeabilities (16). At the OMM, VDAC1 assumes a crucial position in the cell, serving as the main interface between mitochondrial and cellular metabolisms and mediating transport of transport of anions, cations, ATP, Ca 2ϩ , and many metabolites between the mitochondria and the cytosol (17,18). Moreover, its location at the boundary between the mitochondria and the cytosol enables VDAC1 to interact with proteins, such as hexokinase, tubulin, actin, and others (18), that mediate and regulate the integration of mitochondrial functions with other cellular activities.
It is now recognized that VDAC1 not only plays a crucial role in regulating the metabolic and energetic functions of mitochondria but is also a key player in mitochondria-mediated apoptosis (18). VDAC1 is proposed to serve as a component of the apoptosis machinery participating in the release of cytochrome c and to mediate apoptosis regulation by Bcl2 and hexokinase (18 -21). Indeed, accumulated findings indicate that both antiapoptotic and proapoptotic proteins interact with VDAC1 to regulate mitochondria-mediated apoptosis. VDAC1 interacts with Bax, Bim, and Bcl-xL, in isolated mitochondria and in reconstituted membrane systems (12,21,22). Bax and VDAC reconstituted into liposomes were shown to form a new channel, with a conductance 4 -10 times larger than that of the individual proteins (21) and with such increase being prevented by Bcl-xL. The interaction of Bax and Bim with VDAC leads to cytochrome c release, with Bcl-xL and anti-VDAC antibodies preventing this release (23). A direct interaction between Bcl-xL and VDAC1 was demonstrated in NMR-based studies (12). Interaction of the putative loop region of VDAC1 with Bcl-xL was proposed (24). Bcl-xL was also shown to affect the oligomerization of VDAC, as revealed by chemical cross-linking of micelle-bound VDAC, shifting the equilibrium from the trimeric to the dimeric state (12).
In this study, we focused on the involvement of VDAC1 in the antiapoptotic activity of, Bcl-xL, a protein product of the longer Bcl-xL mRNA and one that protects cells from a wide variety of apoptotic stimuli (25). Bcl-xL is proposed to function either by directly inhibiting Apaf-1-mediated activation of caspase-9 or by attenuating release of mitochondrial cytochrome c through interaction with Bax (26). We demonstrate the direct interaction of Bcl-xL with VDAC1 and the involvement of this interaction in the antiapoptotic activity of Bcl-xL. Furthermore, we demonstrate that this interaction can be prevented by VDAC1-based peptides. Thus, these peptides, targeting the antiapoptotic activity of Bcl-xL, can serve as potential cancer therapy.
Cell Transfection-Logarithmically growing T-REx-293 cells were transiently transfected with plasmid pCDNA3.1 encoding Bcl-xL alone or with either one of the VDAC1-based peptideencoding pCDNA4/TO plasmids or plasmid pcDNA4/TO, encoding mutated rat VDAC1 (rVDAC1) or murine VDAC1 (mVDAC1), using calcium phosphate (CaH 2 PO 4 ). Protein expression was induced by tetracycline (0.5-1.5 g/ml) for 24 -48 h. For CaH 2 PO 4 -based transfection, cells (6 ϫ 10 5 ) were seeded onto a 60-mm culture dish plate and grown in 3 ml of DMEM plus supplements, as above. Plasmid DNA (0.2-1.0 g) was added to 250 l of sterile CaCl 2 (240 mM) and mixed with 250 l of sterile HEPES buffer (280 mM NaCl, 10 mM KCl, 1.5 mM Na 2 HPO 4 ⅐H 2 O, 12 mM glucose, 50 mM HEPES, pH 7.05). The mixture was applied to the T-REx-293 cells. Cells were maintained in a humidified atmosphere at 37°C with 5% CO 2 for 16 h, at which time the medium was replaced with 3 ml of fresh medium.
Cell Death Analysis-To determine the extent of apoptosis, cells were subjected to staining with acridine orange and ethidium bromide, as described previously (27). Cells were visualized by fluorescence microscopy (Olympus IX51). In each independent experiment, ϳ300 live, early, and late apoptotic cells were counted.
Bcl-xL(⌬C21) Expression and Purification-DNA encoding Bcl-xL(⌬C21) was cloned in the pET47bϩ vector and expressed in Escherichia coli BL21 cells. The bacteria were grown at 37°C to an A 600 of 0.4 and exposed to a heat shock of 40°C for 2 h followed by induction of expression with isopropyl-D-1-thiogalactopyranoside for 30 -60 min at 20°C. Following sonication and centrifugation, the soluble fraction (ϳ60% of the total expressed Bcl-xL(⌬C21)) was purified by chromatography using a nickel-nitrilotriacetic acid resin HiTrap column. Bcl-xL(⌬C21) was eluted from the column by a linear gradient of imidazole (10 -500 mM). Purified Bcl-xL(⌬C21) was dialyzed against 150 mM NaCl, 10 mM Tricine, pH 7.4, and 20% glycerol, concentrated using a 10-kDa molecular mass cutoff Centricon tube, and stored in aliquots at Ϫ80°C.
Purification of VDAC and Reconstitution into a PLB-VDAC was purified from sheep liver mitochondria and T-REx-293 cells that were silenced for hVDAC1 expression and transfected to express wild-type or mutated mVDAC1 as described previously (28). Sheep liver mitochondria in 10 mM Tris/HCl, pH 7.2, were incubated with 2% LDAO at 4°C for 30 min followed by centrifugation (20 min, 20,000 ϫ g), and the obtained supernatant was loaded onto a dry Celite:hydroxyapatite (2:1) column. VDAC was eluted with a solution containing 2% LDAO, 10 mM Tris/HCl, pH 7.2, 50 mM NaCl, and 22 mM NaH 2 PO 4 . The VDAC-containing fractions were dialyzed against 10 mM Tris/ HCl, pH 7.2, and subjected to second round of chromatography on a carboxymethyl cellulose column from which VDAC was eluted with a solution containing 10 mM Tris/HCl, pH 7.2, 0.3% LDAO, and 500 mM NaCl. The VDAC-containing fractions were collected and used for VDAC1 channel analysis.
Single and multiple channel current recording and data analysis were carried out as described previously (29). Briefly, a PLB was prepared from soybean asolectin dissolved in n-decane (50 mg/ml). Purified VDAC1 or N-terminal truncated VDAC1 (1-100 ng) was added to the chamber defined as the cis side containing 1 or 0.5 M NaCl. After one or a few channels were inserted into the PLB, excess protein was removed by perfusing the cis chamber with ϳ20 volumes of solution to prevent further channel incorporation. Following several recordings of channel activity at different voltages, purified Bcl-xL (⌬21) (ϳ1-2 g) was added to the cis chamber, and currents through the channel were again recorded. Recordings were made under voltage clamp using a bilayer clamp BC-535B amplifier (Warner Instruments, Hamden, CT). Currents were measured with respect to the trans side of the membrane (ground). The currents were low pass-filtered at 1 kHz and digitized on-line using a Digidata 1440 interface board and the Clampex 10.2 software (Axon Instruments, Union City, CA).
Real-time Surface Plasmon Resonance (SPR)-SPR, using the ProteOn-XPR36 system (Bio-Rad), was employed to study the interaction of VDAC1-based synthetic peptides (12.5-200 M) with purified Bcl2(⌬23), Bcl-xL(⌬21), and rabbit IgG. Proteins were immobilized onto a GLM sensor surface, according to the manufacturer's instructions. The peptides were diluted in running buffer (150 mM NaCl, 0.005% (v/v) Tween 20, 4% (v/v) DMSO, and 10 mM PBS, pH 7.4) and injected onto the sensor chip at varying concentrations, at a flow rate of 40 l/min. Response units were monitored using the ProteOn imaging system and related software tools. Signals were normalized using appropriate controls.
Gel Electrophoresis and Immunoblotting-SDS-PAGE was performed according to Laemmli (60). Gels were stained with Coomassie Brilliant Blue or electrotransferred onto nitrocellulose membranes for immunostaining. Membranes containing the transferred proteins were blocked with 5% nonfat dry milk and 0.1% Tween 20 in Tris-buffered saline and then incubated with monoclonal anti-VDAC1 antibodies (1:10,000) followed by incubation with HRP-conjugated anti-mouse IgG secondary antibodies (1:10,000). After treatment with the appropriate primary and secondary antibodies, enhanced chemiluminescence (Pierce) was performed.

RESULTS
Interactions between Bcl-xL and wild-type or mutated VDAC1 or VDAC1-based peptides were analyzed using a variety of approaches at both the purified protein and the cellular levels. Such studies demonstrate the relationship between Bcl-xL binding to VDAC1 and the conveyance of the antiapoptotic effect of Bcl-xL.
Bcl-xL(⌬21) Interacts with VDAC to Induce Channel Closure-Recombinant Bcl-xL(⌬21) was expressed and purified from E. coli BL21 cells on a HiTrap nickel column and analyzed in terms of its effect on the bilayer-reconstituted channel activity of purified mitochondrial VDAC (Fig. 1, A and B). To obtain soluble purified protein, we used Bcl-xL(⌬21), a C-terminally truncated version of the protein (Fig. 1C). Channel activity of bilayer-reconstituted VDAC was studied under voltage clamp conditions (29). The current produced in response to voltage steps from a holding potential of 0 mV to Ϫ10 mV was recorded before and 10 -15 min after the addition of Bcl-xL(⌬21). Channel conductance was reduced by Bcl-xL(⌬21), and the channel was stabilized in a low-conducting state (Fig. 1A). Removing Bcl-xL(⌬21) by perfusing the bath solution with Bcl-xL(⌬21)free solution restored channel conductance, as demonstrated in single channel (Fig. 1A) or multichannel (Fig. 1B) recordings. Bcl-xL(⌬21)-mediated reduction of VDAC conductance was only observed when Bcl-xL(⌬21) was added to the cis side of the bilayer in which VDAC was reconstituted, indicating that Bcl-xL(⌬21) interacts with the cytosolic face of VDAC. Bcl-xL(⌬21) decreased VDAC conductance in multichannel experiments at all voltages tested and stabilized VDAC at a low conductance state, regardless of the voltage gradient applied. Channel conductance was restored close to the level noted before Bcl-xL(⌬21) addition, upon removal of the added protein (Fig. 1B). The decrease in VDAC channel conductance upon Bcl-xL addition indicates that Bcl-xL directly interacts with VDAC and modifies its channel activity.
Bcl-xL(⌬21) Interacts with VDAC, as Demonstrated by MST-Although a Bcl-xL-VDAC interaction was previously demonstrated using NMR (12), here we were able to achieve quantitative analysis of the interaction and extract a binding affinity coefficient (K d ) for the interaction. For this, we used MST, a sensitive protein-protein interaction assay, to analyze the association of VDAC with fluorescently labeled Bcl-xL(⌬21) (Fig.   2). MST is capable of analyzing the binding of proteins or small molecules to an interacting partner as such interactions influence the thermal migration behavior of a labeled version of a binding partner such that fluorescence depletion in a heated spot of the protein solution is measured as a function of increasing interactor concentration, with K d values being derived from the depletion curves (30). Purified Bcl-xL(⌬21) (20 nM) was labeled with NT647 (at a ratio of labeling of about one lysine residue per one Bcl-xL(⌬21) molecule) and incubated with increasing concentrations of purified VDAC (2.4 nM to 20 M). MST analysis was then performed using a NanoTemper Monolith NT 115 apparatus ( Fig. 2A). By plotting the percentage of change in normalized fluorescence (⌬FNorm %) as a function of VDAC concentration, a curve fitted to the data points was obtained using the GraphPad prism software, and a dissociation constant of 0.67 M was calculated. For comparison, a similar experiment was carried out with Bcl2(⌬23), yielding a K d of 0.72 M (Fig. 2B). The results indicate that both C-truncated proteins, Bcl2(⌬23) and Bcl-xL(⌬21), interacted with VDAC with a similar binding affinity.
Identification of VDAC1 Amino Acid Residues and Domains Involved in the Interaction with Bcl-xL-To identify VDAC1 amino acid residues and domains involved in interaction with the Bcl-xL protein, site-directed mutagenesis of VDAC1 was carried out. The interaction of Bcl-xL with a bilayer-reconstituted mutated protein and modification of channel conduct- ance, as well as the antiapoptotic effect of Bcl-xL on cells expressing the mutant VDAC1, were analyzed (Fig. 3). Wildtype, E65Q-, E72Q-, or E202Q-VDAC1 proteins were ex-pressed in T-REx cells silenced for hVDAC1 expression by siRNA specific for human hVDAC1, purified, and reconstituted into a PLB. The effect of Bcl-xL(⌬21) on the channel activity of bilayer-reconstituted native, E65Q-, E72Q-, or E202Q-VDAC1 was studied under voltage clamp conditions (29). Channel activity was recorded before and 15 min after the addition of Bcl-xL(⌬21). As with wild-type VDAC1, Bcl-xL(⌬21) decreased the channel conductance of E65Q-VDAC1, yet had no effect on E72Q-, or E202Q-VDAC1 (Fig. 3A). Channel activity was recorded before and 15 min after the addition of Bcl-xL(⌬21). As with wild-type VDAC1, Bcl-xL(⌬21) decreased the channel conductance of E65Q-VDAC1 (Fig. 3A). However, Bcl-xL(⌬21) had no effect on E72Q-or E202Q-VDAC1 (Fig. 3A). The effects of these mutations in VDAC1 on Bcl-xL-mediated protection against cell death were analyzed by expression of native or mutated VDAC1 in T-REx-293 cells and in the same cells overexpressing Bcl-xL (Fig. 3B). T-REx-293 cells were transfected to express wild type, E65Q-, E72Q-, or E202Q-VDAC1 under the control of tetracycline, alone or together with Bcl-xL (Fig. 3). STS induced apoptotic cell death in about 50% of the cells. Such cell death was almost completely prevented upon expression of Bcl-xL. Protection against STS-induced apoptosis by Bcl-xL was observed only in cells expressing native or E65Q-VDAC1 but not in cells expressing E72Q-or E202Q-VDAC1 (Fig. 3). These results indicate that the interaction of Bcl-xL with VDAC1 is diminished upon introducing certain mutations into VDAC1 that most probably stabilize VDAC1 in conformations that are not recognized by Bcl-xL.

VDAC1-based Peptides Prevent Bcl-xL-mediated Protection against
Apoptosis-Based on the obtained results showing the interaction of Bcl-xL with wild type but not with certain VDAC1 mutants (Figs. 1-3), we designed several VDAC1based peptides and tested their effect on the antiapoptotic activity of Bcl-xL (Fig. 4). VDAC1-based peptides were expressed alone or together with Bcl-xL, and apoptosis, as induced by STS, was analyzed in the various transfected cells  ( Fig. 4). Bcl-xL overexpression prevented STS-induced apoptotic cell death. When cells expressing Bcl-xL were also transfected to express VDAC1-based peptides, designed according to the earlier proposed VDAC1 topology model (17,32), it was observed that those peptides corresponding to the N-terminal region and the first and fourth cytosol-facing loops, LP1 and LP4, but not peptides corresponding to the second and third cytosol-facing loops, LP2 and LP3, prevented the protection (70 -90%) afforded by Bcl-xL against STS-induced cell death (Fig. 4A).
Bcl-xL(⌬C21) Interaction with VDAC1-based Synthetic Peptides-Next, we generated synthetic peptides corresponding to the N-terminal region and the LP1, LP2, and LP4 peptide sequences and analyzed their direct interactions with purified Bcl-xL(⌬C21) using SPR technology (Fig. 5). Purified Bcl-xL(⌬C21) was coupled to a SPR biosensor, the ProteOn GLM chip, and increasing concentrations (50 -400 M) of the VDAC1-based peptides were injected onto the sensor chips. Peptide binding to the immobilized Bcl-xL(⌬C21) was then monitored. The N-terminal peptide and peptides LP1 and LP4 strongly bound to immobilized Bcl-xL(⌬C21) in a concentration-and time-dependent manner (Fig. 5, A-C).
In contrast, the VDAC1-based peptide, LP2, did not interact with Bcl-xL(⌬C21) (Fig. 5D). The binding of the LP1, LP4, and N-terminal peptides to Bcl-xL(⌬C21) was specific as no signal was obtained with an IgG-immobilized control chip (data not shown). Although all Bcl-xL-interacting VDAC1-based pep-   100, and 200 M). The peptides were run in parallel over surface strips containing immobilized Bcl-xL(⌬C21), and responses (resonance units, RU) as a function of peptide concentration were monitored using the ProteOn imaging system and related software tools. All experiments were carried out at 25°C. The solid and dashed arrows indicate runs with peptide-containing and peptide-free samples, respectively.
tides associated with immobilized Bcl-xL, the LP1 and N-terminal peptides showed rapid dissociation, whereas the LP4 peptide showed slow dissociation from immobilized Bcl-xL(⌬C21) upon exposure to peptide-free PBS-Tween buffer. These results are in agreement with the effect of these peptides on the antiapoptotic activity of Bcl-xL (Fig. 4A) and thus demonstrate the direct and specific interaction of selected VDAC1based peptides with Bcl-xL(⌬C21).
Channel Activity of ⌬(26)VDAC1 and Its Modulation by Bcl-xL(⌬21)-As the N-terminal region of VDAC1 interacted with Bcl-xL (Figs. 4A and 5), we further tested the interaction of Bcl-xL with bilayer-reconstituted N-terminally truncated VDAC1, ⌬(26)VDAC1. Wild-type VDAC1 or ⌬(26)VDAC1 was expressed in the mitochondria of porin-less yeast, purified, and reconstituted into a PLB, where VDAC channel activity was analyzed. As demonstrated in our previous study (19), the N-terminally truncated protein, like the native protein, showed a maximal conductance of 4 nanosiemens (at 1 M NaCl), yet lost its voltage-dependent conductance (Fig. 6B). The addition of Bcl-xL(⌬21) to bilayer-reconstituted native VDAC1 induced the channel to adopt a stable, long-lived low-conducting state (Fig. 6A). In contrast, Bcl-xL(⌬21) had no effect on the channel activity of ⌬(26)VDAC1 (Fig. 6A). The channel conductance, voltage insensitivity of the bilayer-reconstituted truncated protein, and its insensitivity to Bcl-xL at all voltages tested (Ϫ60 to ϩ60 mV) are shown (Fig. 6B). These results indicate that the VDAC1 N-terminal region is required for Bcl-xL(⌬21)-mediated modification of channel conductance.

DISCUSSION
Constitutively high levels of the antiapoptotic proteins Bcl-2 or Bcl-xL have been associated with a more aggressive malignant phenotype and/or drug resistance to various categories of chemotherapeutic agents in hematologic malignancies and solid tumors (33). It is well demonstrated that these proteins promote cancer cell survival by antagonizing apoptosis (34). As such, agents targeting antiapoptotic Bcl-2 family members were developed (35). However, targeting antiapoptotic proteins requires knowledge and understanding of their modes of action. Several possible mechanisms of action have been proposed (3,36,37). Here, we demonstrate that Bcl-xL antiapoptotic activity is mediated via direct interaction with VDAC1 and propose a new strategy for targeting this interaction.
Considering that the major cellular site targeted by Bcl-2 family proteins is the mitochondrion (36), several mechanisms were proposed to explain the action of such proteins at the mitochondria (38). The antiapoptotic proteins, such as Bcl2 and Bcl-xL, were proposed to neutralize the proapoptotic effector proteins, Bax and Bak, by forming heterodimeric complexes, with physical associations between Bcl-2 and Bax being reported in various experimental models (39,40). It was also suggested that the antiapoptotic Bcl2 proteins sequester activator BH3-only proteins and neutralize them (37). Bcl-2 and Bcl-xL were also proposed to inhibit autophagy by binding to the protein, Beclin 1, which is required for the initiation of autophagosome formation (41). It was further proposed that the primary targets of p53 at the mitochondria are the prosurvival Bcl2 proteins, with p53 interacting via its DNA-binding region with the antiapoptotic proteins, Bcl-xL and Bcl2 (42). The Bcl-xL-p53 DNA-binding domain protein-protein interface has been characterized (43).
The various Bcl2 and Bcl-xL interactor proteins discussed above are not constant residents of the mitochondria. Although Bcl-xL and Bcl-2 are mainly localized to the mitochondria, both can also control endoplasmic reticulum events associated with apoptosis (44). Their interaction with the OMM protein VDAC is not surprising (12,(21)(22)(23)(24)45). NMR-based studies showed Bcl-xL to interact with VDAC (12). Bcl-xL affects the oligomerization state of VDAC, shifting the equilibrium from the trimeric to the dimeric state (12). It has also been suggested that VDAC interacts with both Bax and Bcl-xL to form a tertiary complex and that the function of VDAC in mediating cytochrome c release could depend upon the ratio between Bax and Bcl-xL (24). Moreover, it has also been suggested that Bcl-xL interacts with VDAC via the putative loop region of VDAC1 (24) and that Bcl-2/Bcl-xL may interact with VDAC1 to sterically block its dimerization (45).
Here, we not only demonstrated a direct interaction of Bcl-xL with VDAC1 but also showed the function of this interaction in mediating the antiapoptotic activity of Bcl-xL and, furthermore, identified sequences and amino acids important FIGURE 6. The N-terminal domain of VDAC1 is required for its interaction with Bcl-xL(⌬C21). VDAC1 and ⌬(26)VDAC1 were expressed in porin-less yeast mitochondria, purified, and reconstituted into a PLB. A, Bcl-xL reduced conductance of VDAC1, yet had no effect on the channel activity of N-terminally truncated VDAC1. Currents through bilayer-reconstituted VDAC1 or ⌬(26)VDAC1 in response to a voltage step from 0 to Ϫ10 mV were recorded before and 10 min after the addition of Bcl-xL. The dashed lines indicate the zero and the maximal current levels. B, currents through the VDAC1 channel in response to a voltage step from 0 mV to voltages between Ϫ60 and ϩ60 mV were recorded. Relative conductance was determined as the ratio of conductance at a given voltage (G) to the maximal conductance (Go). A representative of four similar experiments is shown, using VDAC1 (•) and ⌬(26)VDAC1 (E) and 30 min after the addition Bcl-xL(⌬21) to ⌬(26)VDAC1 (‚).
for this interaction. The association of Bcl-2 and Bcl-xL with the mitochondria involves a hydrophobic C-terminal membrane-spanning domain serving as a membrane insertion device (46 -49). However, the C-terminal membrane-spanning domain of Bcl2 (11) or Bcl-xL (Figs. 1, 3, and 6) is not required for interaction with VDAC1, suggesting that other sequences in these proteins are involved in this interaction. Bcl-xL interacts and modifies channel conductance of purified wild type and E65Q-VDAC1 but not of N-terminally truncated ⌬(26)VDAC1 or E72Q-or E202Q-VDAC1 (Figs. 1, 3A, and 6).
The interaction of Bcl-xL with VDAC1 in the cell is supported by the inability of Bcl-xL to protect against STS-induced cell death in cells expressing E72Q-or E202Q-mutated VDAC1 (Fig. 3B). In previous studies addressing VDAC1 mutations, E65Q, E72Q, and E202Q were found to prevent the protection against apoptosis conferred by the mitochondria-bound hexokinase isoforms, HK-I and HK-II (19,32,50), as well as by Bcl2 (11). Here, we found that the E65Q-VDAC1 binds Bcl-xL and allows Bcl-xL-mediated protection against apoptosis. Further support for Bcl-xL interaction with VDAC1 is reflected in the interaction of Bcl-xL with VDAC1-based peptides both in vitro (Fig. 5) and in situ within the cell (Fig. 4). The ability of Bcl-xL to protect against STS-induced cell death was also prevented by VDAC1-based peptides (Fig. 4).
Although Bcl2 and Bcl-xL share a high degree in sequence homology and have been implicated in functioning in overlapping apoptotic pathways (44), there is accumulating evidence indicating that Bcl2 and Bcl-xL may differ in their modes of action. In other words, Bcl-2 and Bcl-xL may not antagonize apoptosis in exactly the same manner despite their close homology (51). Indeed, earlier studies (52) implied that Bcl-xL did not affect tBid insertion into mitochondrial membranes, whereas Bcl2 does (53). In addition, Bcl-xL and Mcl-1 but not Bcl-2 have been shown to target Bak (54). Both Bcl2 and Bcl-xL are overexpressed in colon and breast cancer, whereas only Bcl2 is expressed in prostate cancer, lymphoma, glioma, leukemia, and melanoma, and only Bcl-xL is expressed in gastric adenomas (3,6). Using VDAC1-based peptides or mutated VDAC1, we have demonstrated several similarities but also differences between Bcl2 and Bcl-xL, with respect to interactions with VDAC1. These include the VDAC1 mutants, E72Q and E202Q, which prevented the antiapoptotic activity of both proteins, and mutant E65Q, which affects the activity of Bcl2 but not of Bcl-xL (11) (Fig. 3). In addition, although the VDAC1-based peptide, LP4, and the N-terminal peptide interact with both proteins, LP2 only interacts with Bcl2, and LP1 interacts with Bcl-xL alone (11) (Fig. 5). In addition, differential effects of Bcl-2 and Bcl-xL on the regulation of inositol 1,4,5-trisphosphate receptors have been reported (55). These differences in activity might elucidate the path for discovery of the specific regulation of the antiapoptotic activities of Bcl2 and Bcl-xL.
In summary, as high levels of expression of antiapoptotic Bcl-2 family members are associated with resistance of many tumors to chemotherapy (4), targeting these proteins presents a possible strategy for cancer therapy. The availability of threedimensional structures of antiapoptotic Bcl-2 family member proteins has led to the design of novel small molecule inhibitors. BH3-mimetic therapy has been sought as anti-Bcl2 anti-cancer therapy (56), such as ABT-737 (57) or G3139/ Oblimersen, an 18-mer phosphorothioate oligonucleotide, targeted to Bcl-2 mRNA (58). Although many such molecules have been synthesized, rigorous verification of their specificity is often lacking (59). Further studies have revealed that many putative Bcl2 and Bcl-xL inhibitors are not specific and have other cellular targets, resulting in nonmechanism-based cell toxicity. Thus, based on the results presented in this study, VDAC1-based peptides interacting with Bcl2 and Bcl-xL and preventing their antiapoptotic activities can serve as a novel strategy for overcoming the apoptosis resistance conferred by overexpression of these prosurvival proteins in tumor cells. Targeting of VDAC1-based peptides to tumor cells overexpressing antiapoptotic proteins, such as Bcl-2, would minimize the self-defense mechanisms of the cancer cells, thereby promoting apoptosis and increasing sensitivity to chemotherapy. Thus, VDAC1-based peptides offer an attractive tool for further development as anticancer agents.