HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 19/11/1504 04-3329fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by D’Alessio, M. Articles by Ghibelli, L. Search for Related Content PubMed PubMed Citation Articles by D’Alessio, M. Articles by Ghibelli, L.

Oxidative Bax dimerization promotes its translocation to mitochondria independently of apoptosis

M. D’Alessio^*, M. De Nicola^*, S. Coppola^*, G. Gualandi, L. Pugliese, C. Cerella^*, S. Cristofanon^*, P. Civitareale^*, M. R. Ciriolo^*, A. Bergamaschi, A. Magrini and L. Ghibelli^*,1

^* Dipartimento di Biologia;
Cattedra di Medicina del Lavoro, Universita’ di Roma "Tor Vergata," Rome, Italy;
Universita’ della Tuscia, Viterbo, Italy; and
S. A. F. AN. Bioinformatics, Torino, Italy

¹Correspondence: Dipartimento di Biología, Universita’ di Roma, "Tor Vergata," via Ricerca Scientifica 1, Roma 00133, Italy. E-mail: ghibelli{at}uniroma2.it

SPECIFIC AIMS

Bax translocation to mitochondria, probably induced by the conformational changes caused by Bax homodimerization, is the triggering event of the intrinsic pathway of apoptotic signaling; despite the importance of the issue and many efforts, the mechanisms that push Bax to mitochondria are still unknown. We have previously shown that redox imbalance is able to induce the release of cytochrome c, an event that is known to be a consequence of Bax translocation to mitochondria: this finding convinced us to explore whether oxidative conditions may be necessary and sufficient to cause Bax translocation.

PRINCIPAL FINDINGS

1. Oxidative alterations promote Bax translocation even in the absence of apoptosis
We have described conditions of nonapoptogenic oxidative stress on U937 monocytic and HepG2 hepatoma cells after glutathione depletion with buthioninesulfoximine (BSO). In these conditions, cells begin the apoptotic process, but this is aborted after cytochrome c release due to activation of survival pathways. Here we show that BSO and low, nonapoptogenic (i.e., 50 µM) H₂O₂ doses induce Bax to translocate to mitochondria on both cell types. This is a functional translocation, since it is accompanied by the release of cytochrome c. However, these oxidative treatments do not elicit caspase activation or apoptosis. We discovered another example of oxidative stress without apoptosis [i.e., a Burkitt lymphoma cell line, originally negative for Epstein Barr virus (BL41) in vitro converted with EBV (E2r clone)]. Unlike the parental BL41, E2r constitutively present high radical levels; these cells live and replicate with Bax constitutively present in mitochondria. This evidence implies that 1) Bax moves even in the absence of apoptosis; and 2) oxidation promotes Bax translocation.

2. Redox-independent apoptosis does not imply Bax translocation
If Bax automatically moves because of oxidation, it follows that apoptosis where Bax is not involved should evolve without oxidative changes. Receptor-induced apoptosis (i.e., with Fas) may evolve without Bax engagement. We proficiently induced apoptosis by stimulating Fas receptor of U937 cells, though Bax maintained a cytosolic localization (not shown). Unlike stress-induced apoptosis, Fas stimulation produced apoptotic cells that maintained a regular GSH content and failed to increase free radical levels (not shown), showing that in our system Fas-induced apoptosis does not involve redox alterations. This allows apoptosis uncoupling from oxidative alterations. These results show that Bax does not move in apoptosis occurring without redox alterations, indicating that oxidation is not only sufficient, but also necessary to promote Bax translocation.

3. Molecular modeling study: interchain disulfide may form and allow the exposure of the hydrophobic C-terminal
Next we explored the mechanism through which oxidative alterations promote Bax translocation. To date, homodimerization is the only model for Bax translocation, and the mechanism responsible for promoting dimerization is unknown. We explored whether it can be an oxidative dimerization. Bax possesses two cysteine residues that are exposed and that are possible targets of oxidation and disulfide bridging. To this purpose, we performed a molecular modeling study starting from pdb1f16.ent NMR structure of Bax in aqueous solution. In this structure, both cysteine residues of Bax (positions 62 and 126) are exposed, possibly being reactive. The analysis showed that disulfide bonds cys126-cys126 and cys62-cys126 are compatible with the Bax 3-dimensional structures, whereas cys62-cys62 is not compatible.

To explore whether formation of any of the two energetically possible bonds may promote the conformational change necessary for membrane insertion [i.e., exposure of the C-terminal tail (helix 9, H9)], we analyzed the position of H9 in the monomer and in the two possible dimers. The conformational search indicated that H9 is free to move in the monomer and in the cys126-cys126 dimer instead. In the cys62-cys126 disulphide bond, H9 is blocked in an exposed conformation.

These simulation studies indicate that 1) formation of two of three theoretical interprotein Bax disulfide bridges are energetically possible and 2) one of these bonds (cys62-cys126) can lead to a conformational change that allows translocation to the membranes.

4. Bax translocation is accompanied by disulfide dimerization
To check the structural bioinformatics predictions, we explored whether a Bax dimer may actually form upon oxidation. To this purpose, we omitted reducing agents in the standard protocol for Western blot analysis, with the goal of revealing a possible upshift of Bax migration in lysates of apoptotic and oxidative-stressed cells. We found such an upshift (in the mitochondrial fractions only) in a position compatible with dimerization with both oxidative treatments, as shown in Fig. 1 A. In the instance of apoptosis (PMC), we also found heavier bands recognized by the specific anti-Bax antibody, suggestive of the complex Bax supramolecular organization of apoptosis. The upshift of Bax migration disappeared in the presence of reducing agents (Fig. 1B ), thus demonstrating that disulfide bridging is the mechanism through which oxidation causes Bax dimerization.

View larger version (55K): [in this window] [in a new window]   Figure 1. Bax translocation is accompanied by disulfide dimerization. A)WB analysis of Bax with polyclonal Ab D21 on cytosolic (cyto) and mitochondrial (mito) fractions (purity was >70%, as assessed by anti-actin and cytochrome c oxidase Ab labeling, not shown) on U937 untreated (CTRL) or treated with BSO (1 mM, 3 h) or H2O2 (50 µM, 1 h) or PMC (4 h). Both cell lysis and the denaturing electrophoretic run were performed in nonreducing conditions (i.e., DTT and B-mercaptoethanol were omitted, respectively) to preserve eventual disulfides. An upshift in Bax migration is appreciated in the mitochondrial fraction upon the oxidative treatments (BSO and H2O2), in a position compatible with Bax homodimerization. In conditions of apoptosis with PMC (10 µg/mL, 4 h; 75% apoptotic cells in this experiment) Bax migrates as a multimer in the denaturing, nonreducing conditions. All the experiments were performed 3 times on U937, always with similar results. The same treatments on HepG2 cells gave similar results (not shown). B)WB analysis with polyclonal antibody D21 on the mitochondrial (mito) fractions performed in reducing conditions, on the same samples as above. Reduction completely abolishes the upshifs in Bax migration. C)WB analysis of Bax in nonreducing condition on cytosolic and mitochondrial fractions on U937 cells treated with 1 mM H2O2. In this condition, a Bax upshift (dimer and oligomer; see arrows) is appreciated only in the cytosolic fraction. D) WB analysis of Bax, performed in nonreducing electrophoretic conditions, on the pure cytosolic fraction of lysates treated with the indicated doses of H2O2 for 30 min, showing a dose-dependent dimerization of Bax (left). The presence of light membranes did not alter the pattern (not shown). The same samples run in reducing conditions (right side) essentially show the monomers. The histogram shows the ratio between dimers vs. monomers in the corresponding conditions (average of 3 independent experiments ±SD). E)Cytosolic fractions were treated or not with 50 µM H2O2 and analyzed for Bax dimerization as in panel D (left). After 30 min treatment, isolated mitochondria were added and incubated for further 30 min. The fractions were then separated and the mitochondrial fraction analyzed to evaluate the extent of Bax translocation (right). The oxidized cytosolic fraction supports much better translocation of Bax to mitochondria.

The cytosolic (plus light membrane) fraction of apoptotic or BSO- or 50 µM H₂O₂-treated cells only showed monomeric Bax. On the other hand, only a negligible amount of monomeric Bax was found in mitochondria (Fig. 1A ). The question then arose of whether Bax dimerizes before or after translocation: in the first instance, the lack of dimers in the cytosol may be due to a very rapid translocation after dimerization; in the second, a Bax monomer might immediately dimerize once in the mitochondria. We found a condition where Bax oligomers accumulate in the cytosol, (i.e., upon treatment with higher H₂O₂ doses). Indeed, 1 mM H₂O₂ produces Bax oligomers that translocate very poorly (Fig. 1C ), indicating that Bax dimers do form in a cytosolic environment, and that the dimers translocate only in "proper" conditions.

5. H₂O₂ induces dimerization of Bax in cytosolic lysates
We performed a direct oxidative treatment of cytosolic (±light membranes) cell extracts, by adding H₂O₂ on lysates from untreated healthy cells. Figure 1D , left, shows that disulfide upshift of Bax migration appears in an H₂O₂ dose-dependent fashion. The presence of reducing agents abolishes the upshift (Fig. 1D , center) confirming its disulfide nature; the ratios between the dimeric/monomeric form are shown in the right panel. The presence of the light membrane (i.e., ER in the cytosolic fraction) did not alter the monomeric/dimeric ratio, showing that dimerization can take place in a cytosolic environment devoid of membranes.

6. Oxidized lysates support Bax translocation to isolated mitochondria
To test the ability of a dimer vs. monomer to translocate, we set up an in vitro procedure consisting of two cycles of fractionation, remixing, and final fractionation of cytosol vs. mitochondria. We explored whether the dimers-enriched oxidized cytosolic fractions (ratio dimer/monomer=8.4) support translocation of Bax to an added mitochondrial preparation better than a nonoxidized cytosolic fraction (ratio dimer/monomer=0.7, Fig. 1E , "cyto"). After oxidation, the different cytosolic fractions were incubated for 30 min with purified untreated mitochondria. Mitochondria were then separated, and the cytosolic and mitochondrial resulting fractions were analyzed for quantification of Bax by Western blot. Figure 1E , "mito" shows that translocation is much more efficient (8.6-fold) when Bax comes from the oxidized, dimers-enriched cytosols, implying that dimers translocate better than monomers (if not exclusively). This experiment established a functional correlation between disulfide dimerization and ability to translocate to mitochondria.

CONCLUSIONS AND SIGNIFICANCE

Bax translocation is the most upstream event of the intrinsic apoptotic pathway described so far. The intrinsic pathway of apoptosis is mostly triggered by cell damage, involving gross environmental alterations such as uncontrolled Ca²⁺ or ROS overload. This suggests that its activation may be due not to sophisticated interactions (such as those involved in the receptor-mediated extrinsic apoptotic pathway), but rather to sudden environmental changes, thus rendering it quite difficult to identify the mechanisms involved. Thus, Bax may be a protein "sensing" intracellular chemical/physical alterations, rather than being activated by specific molecular interactions.

We show that oxidative alterations promote Bax translocation to mitochondria.

We have previously shown that redox alterations are necessary for stress-induced apoptosis, due to the ability of apoptosing cells to extrude glutathione in the reduced form through physiological carriers. Apoptotic GSH extrusion is associated with the generation of free radicals, and both types of redox alterations seem to contribute to triggering the downstream events of apoptosis. Here, we show that the mechanism through which this may occur is by convincing Bax to translocate. Bax becomes "competent" for translocation due to conformational changes leading to the exposure of the hydrophobic C-terminal BH4 domain, thus allowing membrane anchorage. However, the determinants of dimerization, attributed to the BH3 domain, were never individuated by amino acid substitution analysis.

Our model (i.e., that oxidative alterations promote Bax translocation) also implies a novel mechanism through which dimerization may occur. Reversible cysteine oxidation induced by ROS is a well-known means of protein dimerization. The process is controlled by both free radical and GSH levels. Radicals promote cysteine oxidation in a dose-dependent fashion. Whether this oxidation ends up with reversible glutathionylation (thus preserving protein structure and function) or interprotein disulfide bridging (leading to multimerization) may depend on the concentration of GSH. Thus, the two cysteine residues of Bax may be the primary targets of oxidative-induced dimerization. Indeed, both cysteine residues are exposed as results from published NMR modeling studies of Bax, and our computer modeling shows that the cysteines can react and participate in intermolecular disulfide bridging, producing a structure that blocks the BH4 domain in an exposed configuration, thus promoting insertion into membranes (Fig. 2 ).

View larger version (35K): [in this window] [in a new window]   Figure 2. Oxidative dimerization of Bax promotes its translocation: Oxidative stress may induce Bax dimerization via disulfide bridging; the cys62-cys126 bond blocks the C-terminal hydrophobic tail in an exposed configuration, thus favoring insertion in membranes of the dimer.

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/1096/fj.04-3329fje;

This article has been cited by other articles:

				J H-C Ho, K-C Tseng, W-H Ma, K-H Chen, O K-S Lee, and Y Su Thymosin beta-4 upregulates anti-oxidative enzymes and protects human cornea epithelial cells against oxidative damage Br. J. Ophthalmol., July 1, 2008; 92(7): 992 - 997. [Abstract] [Full Text] [PDF]

				C. Nie, C. Tian, L. Zhao, P. X. Petit, M. Mehrpour, and Q. Chen Cysteine 62 of Bax Is Critical for Its Conformational Activation and Its Proapoptotic Activity in Response to H2O2-induced Apoptosis J. Biol. Chem., May 30, 2008; 283(22): 15359 - 15369. [Abstract] [Full Text] [PDF]

				M. Tsai-Turton, B. N Nakamura, and U. Luderer Induction of Apoptosis by 9,10-Dimethyl-1,2-Benzanthracene in Cultured Preovulatory Rat Follicles Is Preceded by a Rise in Reactive Oxygen Species and Is Prevented by Glutathione Biol Reprod, September 1, 2007; 77(3): 442 - 451. [Abstract] [Full Text] [PDF]

				A. DORIO, C. CERELLA, M. DE NICOLA, M. D'ALESSIO, G. GUALANDI, and L. GHIBELLI Non-apoptogenic Ca2+-Related Extrusion of Mitochondria in Anoxia/Reoxygenation Stress Ann. N.Y. Acad. Sci., March 1, 2007; 1099(1): 512 - 515. [Abstract] [Full Text] [PDF]

				M DE NICOLA, S CORDISCO, C CERELLA, M.C ALBERTINI, M D'ALESSIO, A ACCORSI, A BERGAMASCHI, A MAGRINI, and L GHIBELLI Magnetic Fields Protect from Apoptosis via Redox Alteration Ann. N.Y. Acad. Sci., December 1, 2006; 1090(1): 59 - 68. [Abstract] [Full Text] [PDF]

				M DE NICOLA, C CERELLA, M D'ALESSIO, S COPPOLA, A MAGRINI, A BERGAMASCHI, and L GHIBELLI The Cleavage Mode of Apoptotic Nuclear Vesiculation Is Related to Plasma Membrane Blebbing and Depends on Actin Reorganization Ann. N.Y. Acad. Sci., December 1, 2006; 1090(1): 69 - 78. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 19/11/1504 04-3329fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by D’Alessio, M. Articles by Ghibelli, L. Search for Related Content PubMed PubMed Citation Articles by D’Alessio, M. Articles by Ghibelli, L.