Changes in specific lipids regulate BAX-induced mitochondrial permeability transition ´ ´ E Martınez-Abundis1, N Garcıa1, F Correa1, M Franco2 and C Zazueta1 ´ ´ ´ ´ Departamento de Bioquımica, Instituto Nacional de Cardiologıa Ignacio Chavez, Mexico ´ ´ ´ ´ Departamento de Nefrologıa, Instituto Nacional de Cardiologıa Ignacio Chavez, Mexico Keywords BAX; cholesterol; gangliosides; lipid microdomains; mitochondrial permeability transition pore Correspondence ´ C Zazueta, Instituto Nacional de Cardiologıa, ´ I Ch., Departamento de Bioquımica, Juan ´ Badiano No 1, Colonia Seccion XVI, ´ Mexico 14080, D.F Fax: +52 55 5573 0926 Tel: +52 55 5573 2911 (1465) E-mail: azazuetam@yahoo.com (Received 13 August 2007, revised 17 October 2007, accepted 25 October 2007) doi:10.1111/j.1742-4658.2007.06166.x Recent evidence suggests the existence of lipid microdomains in mitochondria, apparently coexisting as structural elements with some of the mitochondrial permeability transition pore-forming proteins and members of the Bcl-2 family The aim of this study was to investigate the relevance of the main components of membrane microdomains (e.g cholesterol and sphingolipids) in activation of the mitochondrial permeability transition pore (mPTP) by recombinant BAX (rBAX) For this purpose, we used chemically modified renal cortex mitochondria and renal cortex mitochondria from hypothyroid rats that show a modified mitochondrial lipid composition in vivo Oligomeric rBAX induced an enhanced permeability conformation in the mPTP of control mitochondria rBAX failed to induce mPTP opening when the cholesterol and ganglioside content of mitochondria were modified with the chelator methyl-beta-cyclodextrin Accordingly, hypothyroid mitochondria, with endogenously lower cholesterol and ganglioside content, showed resistance to mPTP opening induced by rBAX These observations suggest that enriched cholesterol and ganglioside domains in the mitochondrial membranes may determine BAX interaction with the mPTP An intriguing observation was that chemical extraction of cholesterol and ganglioside in control mitochondria did not have an effect on rBAX insertion Conversely, in hypothyroid mitochondria, rBAX insertion was diminished dramatically compared with control mitochondria The membrane and protein changes associated with thyroid status and their possible role in rBAX docking into the membranes are discussed BAX is a pro-apoptotic member of the Bcl-2 protein family that resides in an inactive state in the cytoplasm of normal cells Following an apoptotic stimulus, BAX undergoes conformational changes [1–4], becoming a mediator of the intrinsic phase of apoptosis by its insertion into mitochondrial membranes, a process that culminates in the release of cytochrome c and activation of effector caspases [5] Evidence from the literature indicates that BAX is activated by other BH3-only proteins, particularly Bid protein and Bim peptide [6,7] In this respect, it has been shown that Bid activation depends on its proteolytic processing into tBid and the translocation of tBid to mitochondria [8] In the mitochondria, tBid may form oligomers by itself and induce oligomerization of BAX and Bak [9] Although it remains unclear how tBid triggers such oligomerization, there is strong evidence to suggest that Bid-induced apoptosis depends on the presence of BAX and Bak [10] However, it has also been demonstrated that tBid accumulation in mitochondria is an essential, but not sufficient, event leading to mitochondrial disruption for cytochrome c leakage [6] The Abbreviations ANT, adenine nucleotide translocator; CSA, cyclosporine A; GST, glutathione S-transferase; MbCD, methyl-beta-cyclodextrin; mPT, mitochondrial permeability transition; mPTP, mitochondrial permeability transition pore; rBAX, recombinant BAX; VDAC, voltage-dependent anion channel; Dw, transmembrane potential 6500 FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS ´ E Martınez-Abundis et al molecular mechanism underlying such leakage is still a matter of debate One proposal is that oligomeric BAX forms pores in the outer mitochondrial membrane, providing a mechanism for cytochrome c release [11]; a second proposal is that BAX induces opening of the permeability transition pore (mPTP) [12] The opening of this mega-channel would lead to mitochondrial swelling and rupture of the outer membrane, thus explaining the release of cytochrome c from mitochondria Although several reports have suggested an interaction between BAX and the components of the mPTP, e.g the adenine nucleotide translocator (ANT) [13] and ⁄ or the voltage-dependent anion channel (VDAC) [14], the factors involved in BAX induction of mitochondrial permeability transition (mPT) are unknown By contrast, lipid microdomains or ‘lipid rafts’ were first described in the plasma membrane It is known that they are composed mainly of sphingolipids and cholesterol, and have been considered to be architectural domains, where specific proteins could interact to exert specific regulatory mechanisms [15] Relevant to this issue, there is recent evidence suggesting the existence of lipid microdomains in mitochondria which coexist as structural elements with some of the mPTP-forming proteins [16] Recent reports indicate that the disialoganglioside (GD3) contributes directly to the opening of the permeability transition pore complex in isolated mitochondria [17,18] and furthermore, that GD3-induced damage in mitochondrial membranes is confined to sites that can be restrained by Bcl-2 [19] Also, several lines of evidence converge on the assumption that the mPTP propagates the downstream stage of apoptosis, mediated by the lipid-dependent pathway, which includes ceramide and GD3 [20] In addition, cholesterol is known to induce changes in the phospholipid packing of the lipid bilayer and has been suggested as critical for microdomain formation [21] Finally, Keller et al [22] reported that changing the nature of the sterol in cellular rafts leads to changes in the protein composition of those domains Our results suggest that cholesterol and GD3 are relevant for BAX interaction with the mPTP These data provide evidence that specific lipids play a key role in cross-talk between rBAX and the mPTP Results rBAX-induced mitochondrial calcium release and modified cytochrome content The addition of oligomeric rBAX to isolated mitochondria induced the release of accumulated calcium Microdomain components – effect on BAX-induced mPT in a dose-dependent manner (Fig 1A) Figure also shows that lm carboxyatractyloside, a well-known mPTP inducer, promoted immediate depletion of the intramitochondrial calcium content The recombinant protein associated with mitochondrial membranes was resistant to alkaline extraction, indicating that it was deeply embedded in the bilayer (Fig 1B) A clear correspondence between cytochrome c release and calcium extrusion from the mitochondria was observed by inducing opening of the mPTP with either carboxyatractyloside or rBAX (Fig 1C) Cyclosporin A prevented cytochrome c release and mPTP opening induced by rBAX The immunosuppressant cyclosporin A (CSA) effectively inhibits mPTP opening under almost all conditions, and is therefore considered a marker for mPT To determine whether the altered mitochondrial permeability observed in the presence of rBAX in rat kidney mitochondria was related to the opening of this mega-channel, we measured the effect of CSA on calcium release induced by the addition of recombinant protein (Fig 2A) CSA inhibited calcium release, although rBAX remained attached to the membranes (Fig 2B) Interestingly, mitochondria incubated with rBAX only released cytochrome c when the permeability transition pore was opened, and this was abolished in a medium supplemented with CSA (Fig 2C) These results suggest active participation of the mPTP on cytochrome c release induced by rBAX The effect of rBAX on the mPT was entirely dependent on calcium Typical traces of mitochondrial swelling with and without calcium are shown in Fig 3A An important decrease in light scattering was detected when the medium was supplemented with a final calcium concentration of 50 lm, and was inhibited completely by lm CSA Also, pore opening can be estimated by following the discharge kinetics of the transmembrane potential (Dw) Figure 3B shows that the membrane potential developed by mitochondria added to the medium without any previous treatment (trace a) was similar to that developed in mitochondria incubated with rBAX in the absence of calcium (trace b) Conversely, addition of 50 lm CaCl2 to the medium led to a decrease in the Dw of mitochondria incubated with rBAX, which eventually resulted in its total collapse (trace c) We also determined mitochondrial NAD+ content, which has been related to mPTP gating in situ The NAD+ content of mitochondria incubated for 15 with rBAX, but not exposed to the calcium-containing FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS 6501 ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT rBAX (nM) A 0.30 50 100 200 BAX M± rBAX M Calciumrelease (nM) 200 100 A - B Time course graph 0.16 CAT μM ANT rBAX (nM) CAT (μM) 50 - 100 200 SS Cytc 0.02 1090 Time (seconds) M 2200 M Calcium release BAX A rBAX 0.18 SS Cytc M Control, rBAX + CSA 0.02 750 Time (seconds) 1500 medium, was similar to control mitochondria that had undergone incubation and further exposure to a medium supplemented with calcium, but not incubated with rBAX (9.66 versus 9.38 nmol NAD+ặmg protein)1, n ẳ 2) Conversely, NAD+ content decreased in rBAX-treated mitochondria when mPTP was induced in a medium supplemented with calcium (2.08 nmol NAD+Ỉmg protein)1, n ¼ 2) 6502 Std Mitos ± rBAX rBAX + CSA B Time Course Graph 0.35 rBAX A Control Fig Recombinant BAX induces mitochondrial calcium extrusion and cytochrome c release from isolated mitochondria (A) Calcium uptake in isolated mitochondria in the presence of different rBAX concentrations Mitochondrial protein (1.3 mg) was incubated with oligomerized rBAX, as described in Experimental procedures, for 15 and then added to mL of a medium containing 125 mM KCl, 10 mM Tris, 10 mM succinate, mM Pi, 200 lM ADP, 50 lM CaCl2, lgỈmL)1 rotenone and 50 lM Arsenazo III, pH 7.4 (B) BAX detection in mitochondria After maximal calcium release, 1.8 mL of the suspension was withdrawn and centrifuged at 18 000 g for 10 Mitochondrial samples were treated with Na2CO3 as described in Experimental procedures Mitochondria were dissolved in 50 lL NaCl ⁄ Pi, pH 7.0, and mixed with 25 lL of a 3· Laemmli’s loading buffer, boiled for 15 min, subjected to SDS ⁄ PAGE and evaluated by western blot Protein loading was determined by using anti-ANT polyclonal IgG (C) Cytochrome c content in mitochondria (M) and in supernatants (SS) recovered after rBAX incubation Proteins in the supernatants were precipitated with trichloroacetic acid and evaluated for cytochrome c content along with mitochondrial samples Results are representative of three independent experiments Fig CSA prevents cytochrome c release and mPTP opening induced by rBAX (A) After treatment of mitochondria with 150 nM rBAX, mPTP opening was evaluated as calcium release under the conditions described in Fig Where indicated, the assay medium was supplemented with lM CSA (B) rBAX and cytochrome c content was detected in mitochondria, as described in Fig Traces and blots are representative of at least four different experiments M, mitochondria; SS, supernatants; Std, purified rBAX and commercial cytochrome c, respectively Lipid modification in rat kidney mitochondria produced resistance to rBAX-induced permeability transition Methyl-beta-cyclodextrin (MbCD) is known to induce cholesterol efflux from membranes and, consequently, to promote microdomain disruption We evaluated the effect of this chelator on the ability of rBAX to induce FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT B Time course graph Time course graph 0.30 A a, b c d 0.70 e 0.60 (+) c 0.25 A 0.80 Mitochondrial swelling M±rBAX Membrane potential A 0.20 a 0.15 db 0.10 500 1000 Time (seconds) 0 500 Time (seconds) (–) 1000 Fig Effect of rBAX on mitochondrial swelling and Dw (A) Mitochondrial swelling induced by rBAX was monitored as absorbance changes at 540 nm Mitochondrial protein (1.3 mg) was incubated with 150 nM oligomerized rBAX and added to mL of basic medium, described in Experimental procedures, without calcium Where indicated, the medium was supplemented with 50 lM CaCl2 and lM CSA Trace a, mitochondria incubated only in the presence of the octyl glucoside (OG) concentration that promotes rBAX oligomerization; trace b, as in trace a, but the medium contained CaCl2; trace c, mitochondria incubated with 150 nM rBAX in the presence of calcium and CSA; trace d, mitochondria incubated with 150 nM rBAX without calcium; trace e, mitochondria incubated with 150 nM rBAX and calcium (B) Dw in rBAX-treated mitochondria Mitochondria added to the calcium-containing medium without any previous treatment (trace a) Mitochondria incubated with rBAX without calcium (trace b) Mitochondria incubated with rBAX and calcium (trace c) A Time course graph a A 0.10 0.08 0.06 Calcium release 0.12 rBAX B MβCD – – + BAX VDAC b, c 0.04 500 1000 2000 Time (seconds) Fig MbCD inhibits mPTP opening induced by rBAX in control mitochondria (A) Oligomerized rBAX was incubated with control mitochondria for 15 and then added to the basic medium described in Fig 1, in the presence of MbCD Trace a, 150 nM rBAX; trace b, 150 nM rBAX and 0.5 mM MbCD; trace c, 150 nM rBAX and 1.5 mM MbCD [Correction added after publication 26 November 2007: Fig 4A has been replaced with a corrected version] (B) Mitochondrial samples were withdrawn at the end of the tracing, centrifuged and subjected to alkaline extraction before SDS ⁄ PAGE fractioning and western blotting The horizontal line indicates the samples with added rBAX Membranes were incubated against anti-BAX mAb, stripped, and evaluated for VDAC content to check protein loading Blots are representative of two different experiments opening of the permeability transition pore in rat kidney mitochondria MbCD at doses of 0.5–1.5 mm prevented opening of the mPTP induced by 150 nm oligomeric rBAX (Fig 4A) Interestingly, rBAX remained inserted into MbCD-treated mitochondria (Fig 4B) These results indicate that altered cholesterol levels in mitochondrial membranes are important for rBAX cross-talk with components of the mPTP, and not influence rBAX insertion Kidney mitochondria from hypothyroid rats were refractive to mPTP-opening induced by rBAX To gain insight into the existence of putative mitochondrial lipid microdomains in vivo, we evaluated the inducing effect of rBAX on opening of the mPTP in kidney mitochondria from hypothyroid rats We hypothesized that if cholesterol and gangliosides are required for BAX-induced mPTP opening, and if such FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS 6503 ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT Time course graph A 0.20 B M±rBAX A a 0.10 b Calcium release Control 0.15 rBAX - - + Hypothyroid + - + + BAX M VDAC c d 0.05 1750 500 1000 Time (seconds) Fig Hypothyroid mitochondria are resistant to mPT induced by rBAX (A) Calcium release was measured in control and hypothyroid rat kidney mitochondria as described in Fig Trace a, control mitochondria incubated with 150 nM rBAX; trace b, control mitochondria incubated with medium without rBAX; trace c, hypothyroid mitochondria incubated in the presence of 150 nM rBAX; trace d, hypothyroid mitochondria incubated without rBAX Results are representative of at least three independent experiments in which different mitochondria and rBAX preparations were used (B) Mitochondrial samples were withdrawn at the end of the tracing, centrifuged and subjected to alkaline extraction before SDS ⁄ PAGE fractioning and western blotting Membranes were incubated against anti-BAX mAb, stripped, and evaluated for VDAC content to check loading Blots are representative of three different experiments M, mitochondria domains could be key sites for BAX docking and anchoring to the membrane, we should be able to find a different response from either hypothyroid or control mitochondria, on account of their modified mitochondrial lipid composition [23] Figure 5A shows that 150 nm rBAX did not have any effect on mPTP opening in hypothyroid mitochondria; furthermore, rBAX insertion was dramatically diminished in mitochondrial membranes (Fig 5B) Table Cholesterol content in control and hypothyroid mitochondria after MbCD treatment Mitochondria from control and hypothyroid rats were incubated with 1.5 mM MbCD, as indicated previously Total lipids were recovered and evaluated for cholesterol content as indicated in Experimental procedures Data represent mean ± SD of the indicated number of independent experiments Cholesterol (lgặmg protein)1) Control Hypothyroid (n ẳ 5) 70.55 ± 14.7 (n ¼ 5) 43.1 ± 1.6 46.54 ± 17.6* (n ¼ 5) 37.6 ± 1.9* Comparison between cholesterol and ganglioside content in chemically modified rat kidney mitochondria and hypothyroid kidney mitochondria Dimethylsulfoxide-treated mitochondria MbCD-treated mitochondria The cholesterol content of control and hypothyroid mitochondria was measured using gas chromatography In hypothyroid kidney mitochondria, cholesterol content was 39% lower than in control (euthyroid) mitochondria, i.e 43.1 ± 1.6 versus 70.5 ± 14.7 lgỈmg)1 protein) Diminished cholesterol levels in mitochondria correlated with inhibition of the ‘open state’ of the mPTP induced by rBAX Interestingly, chemical cholesterol depletion of control mitochondria membranes by MbCD (34%) produced resistance to rBAX-induced permeability transition (Table 1) These results suggest that specific lipid domains are required for BAX interaction with mPTP components; alternatively, cholesterol may be important for correct assembly of the pore components after rBAX induction of the ‘open state’ Along with cholesterol, glycosphingolipids are concentrated in specific lipid domains in the plasma *P < 0.05 versus mitochondria control 6504 membrane Recently, evidence has indicated that glycosphingolipids and their precursor, ceramide, are also associated with intracellular organelles, particularly mitochondria [24] Because GD3 is the main glycosphingolipid associated with mitochondria and it has been shown to have a role in mPTP regulation, we analyzed the polar glycosphingolipid fraction from total mitochondrial lipid extracts The obtained glycosphingolipids were dried and separated by TLC Only one positive band was found after sulfuric acid detection that comigrated with the GD3 standard in control mitochondria (not shown) Indeed, GD3 in hypothyroid mitochondria was almost undetectable after TLC To verify that GD3 signal variation was not due to variable loading, we performed densitometric scanning analysis of plates in which equal volumes of extracted FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT Control A Hypo A GD3 Control Hypo Resorcinol H2SO4 B Signal ratio (resorcinol/H2SO4) MβCD - - + + - - + CL + 1.0 1.5 CL MβCD (mM) 0.20 Hypo (µg CL/mg protein) 29.2 ± 5.2 15.3 ± 4.3* 0.5 0.25 Control (µg CL/mg protein) 0.30 25.9 ± 6.4 19.4 ± 2.0* 30.1 ± 1.9 17.8 ± 6.0* 1.5 31.1 ± 6.4 19.1 ± 5.2* * 0.15 * 0.10 n= n= Control 0.00 1.0 1.5 0.5 MβCD (mM) B 0.35 0.05 0.5 Control + MβCD * n= n= Hypothyroid Hypothyroid + MβCD Fig GD3 content in control and hypothyroid mitochondria after MbCD treatment Mitochondrial ganglioside content in which lg standard GD3 was run in parallel with mitochondrial extracted lipids (A) The plate is representative of three independent experiments (B) The densitometric ratio between sulfuric acid detection for total lipids and ganglioside-specific detection with resorcinol is presented A value of P < 0.05 was considered statistically significant glycosphingolipid fractions, obtained from the same amount of mitochondrial protein, were loaded and developed with sulfuric acid (unspecific) and resorcinol (ganglioside-specific stain) By increasing the volumes applied to the plates we detected resorcinol-positive signaling (Fig 6A) The ratio between the signals is plotted in Fig 6B The lower GD3 content observed in hypothyroid mitochondria is consistent with reports that hypothyroid conditions affect the biosynthesis and expression of gangliosides in specific tissues and cell types [23] This, along with lower cholesterol levels in the membranes, may translate into a diminished ability to form lipid microdomains in which specific proteins could be recruited and possibly, to interact Because cardiolipin is relevant for mPTP regulation and this inner membrane phospholipid is diminished in thyroid insufficiency, a condition that favors the closed mPTP conformation, we sought to determine the effect of MbCD treatment on cardiolipin content in control and hypothyroid mitochondria Figure 7A, shows the chromatogram of the phospholipids extracted from both mitochondrial types after MbCD incubation As expected, cardiolipin content in hypothyroid mitochondria was significantly diminished compared with control mitochondria, but no changes were observed in MbCD-treated mitochondria We conclude that changes in cardiolipin content could not account for Fig Mitochondrial cardiolipin analysis after MbCD-treatment (A) Cardiolipin was separated by TLC as described in the Experimental procedures from control and hypothyroid mitochondria treated with MbCD (B) Quantification was performed using acridine orange Values are expressed as the mean ± SD of three different experiments CL, cardiolipin *P < 0.005 versus respective control MbCD protection from mPTP opening, at least in control mitochondria, but may be important for rBAX insertion into the mitochondrial membranes Discussion The purpose of this study was to demonstrate the participation of lipids, specifically those related to microdomain components, in the activation of the permeability transition pore by the pro-apoptogenic protein BAX, using lipid-modified mitochondria in vitro and in vivo It is widely known that membrane lipids have major structural and functional roles modulating signaling pathways In the plasma membrane, apoptotic death mediated by lipids is propagated by activation of the surface receptor CD95 (FAS ⁄ APO-1), and includes the formation of ceramide and GD3 [24] Ceramide is produced by the hydrolysis of membrane sphingomyelin [25] and by means of a GD3 synthase (a2,8-sialyltransferase), rapidly converted to gangliosides [26] GD3 and cholesterol are relatively enriched in plasma membranes and are concentrated in specialized domains called rafts or lipid microdomains [15] In this sense, new evidence suggests that raft domains are not exclusive to the plasma membrane [16,27] In particular, the existence of GD3-enriched microdomains in mitochondria [16], and furthermore, data suggesting the FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS 6505 ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT existence of a metabolic pathway of sphingolipids, including several enzyme activities of sphingolipid metabolism, have been reported in these organelles [28] Thus, we measured the oligomeric rBAX-inducing effect on the mitochondrial permeability transition in chemically modified renal cortex mitochondria and renal cortex mitochondria from hypothyroid rats, which show an in vivo modified mitochondrial lipid composition According to other groups [20,29,30], our results indicate that rBAX promoted opening of the mPTP, and not only formed supramolecular openings in the outer mitochondrial membranes, as proposed previously [31] rBAX insertion into mitochondrial membranes induced calcium release from the matrix space, while cytochrome c content diminished The rBAX effect was similar to that exerted by carboxyatractyloside when used to induce mPTP opening (Fig 1) We also observed that rBAX-induced permeability was totally abolished by the classic mPTP inhibitor CSA (Fig 2) and depended on calcium overload (Fig 3) When the lipid mitochondrial composition was disrupted, a different response was observed Control mitochondria showed resistance to the rBAX-inducing action on the mPTP, following removal of cholesterol with MbCD Diminution in cholesterol and GD3 levels in treated membranes, correlated with low cholesterol and GD3 content in mitochondrial membranes isolated from hypothyroid rats that were naturally protected against mPTP opening induced by rBAX (Fig 5) It is conceivable that the cholesterol ⁄ gangliosides ratio may be relevant to maintaining the structure of the lipid microdomains, in such a way that disruption of one of the elements could be translated into a loss of the local formation of specific and functional raft-associated complexes Although the MbCD properties to chelate cholesterol are well known, there is, to our knowledge, no report indicating that it also decreases GD3 content Indeed, our results show that MbCD induced a marked decrease in GD3, which represent a novel finding In this respect, it has been described that MbCD– cholesterol extraction is accompanied by release of GM1 molecules and proteins from plasma membranes [32] With respect to the relevance of cholesterol in a mitochondrial lipid raft, it is known that addition of cholesterol to membrane models containing only phospholipids and sphingolipids permits the formation of a liquid-order phase in which saturated acyl chains are highly organized, as in a highly ordered gel phase, but exhibit lateral mobility more similar to that in the liquid-ordered crystalline phase [21] If cholesterol is the determinant for the generation of a liquid-ordered 6506 phase, it follows that alterations in the cholesterol content of membranes should lead to changes in their properties Our findings suggest that cholesterol decrease correlated with mPTP opening inhibition in the presence of rBAX, which is consistent with studies of cholesterol depletion that alters the function of a raft-associated potassium channel [33] Modifications in the cholesterol content in vivo may lead to alterations in the physical environment of the membrane, consequently changing the likelihood of certain proteins partitioning into these domains Furthermore, it has been reported that hypothyroid conditions affect the biosynthesis and expression of gangliosides in specific tissues and cell types [34] Our findings could be organized into three different scenarios First, cholesterol and GD3 in the mitochondrial membranes favored rBAX insertion and interaction with the mPTP (control mitochondria); second, a diminished cholesterol and GD3 content, resulted in a lower likelihood of raft formation, and hence in crosstalk disruption between rBAX and PTPm (control mitochondria + MbCD) This is supported by recent data indicating that the macromolecular complex conformed by GD3, the voltage dependent anion channel-1 (VDAC-1) and the fission protein hFis1 could be targets in which Bcl-2 family proteins are recruited [16] Relevant to this issue is that GD3 and ceramide have been shown to boost the ability of BAX to induce the mPTP [20] Finally, in membranes with endogenously low cholesterol and ganglioside content (hypothyroid mitochondria) the closed state of the mPTP was also favored It has been reported that hypothyroidism is associated not only with low cholesterol and ganglioside levels, but also with decreased mitochondrial functional activity, oxygen consumption, ATP synthesis and, relevant for mPTP regulation, with low cardiolipin content and a diminished amount of some of the mPTP components (e.g ANT) [35,36] The striking difference between rBAX association in control and hypothyroid mitochondria might be explained by the differences in cholesterol and ganglioside content, along with modifications characteristic of hypothyroidism, in particular cardiolipin content A potential function of GD3 could be to enhance translocation of the pro-apoptotic protein BAX to mitochondria An attractive hypothesis to explain the results obtained in hypothyroid mitochondria could be that cardiolipin is relevant for BAX docking into the membranes Regarding the relevance of cardiolipin and mPTP on BAX insertion, there are reports suggesting that BAX permeabilizes synthetic liposomes only if cardiolipin is present [31,37] and that oligomeric BAX does not permeabilize outer membrane vesicles if the FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS ´ E Martınez-Abundis et al contact sites have been removed [38] In this sense, cardiolipin clusters associated with apoptoptic and energy-flux process proteins have been found at contact sites in the mitochondrial membranes [39] In conclusion, our results indicate that cholesterol and GD3 are relevant for the interaction of BAX with the mPTP The data provide evidence that specific lipids play a key role in cross-talk between rBAX and the mPTP Indeed, more experimental data supporting the idea that BAX and mPTP converge into mitochondrial like-raft domains are imperative Experimental procedures Antibodies and reagents Chemicals were of reagent or higher grade from SigmaAldrich (St Louis, MO), unless otherwise specified Glutathione S-transferase (GST)–Sepharose, thrombin protease and enhanced chemioluminescence system detection were obtained from Amersham Biosciences (Chalfont St Giles, UK); protease inhibitors set was from Roche (Mannheim, Germany) silica gel 60F254 was purchased from Merck (Darmstadt, Germany); anti-ANT polyclonal IgG (N-19) was from Santa Cruz Biotechnology Inc (Santa Cruz, CA); anti-BAX mAb (Clone 6A7) was from Alexis Biochemicals (San Diego, CA), anti-(cytochrome c) mAb (Clone 7H8.2C12) and biotin-conjugated secondary antibodies were from Zymed Laboratories (San Francisco, CA) Induction of hypothyroidism All animal procedures were performed in accordance with the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health [publication no 85 (23) revised 1996] Male Wistar rats weighing 380 ± 17 g underwent surgical thyroidectomy with parathyroid re-implant, as described previously [40] Briefly, the trachea was exposed under anesthesia Parathyroid glands were visualized by means of a stereoscopic microscope (Wild M5, Wild Heerbrugg, Switzerland), dissected from the thyroid gland and re-implanted into the surrounding neck muscles The thyroid gland was then carefully dissected to avoid injury to the laryngeal nerves and completely excised The effectiveness of this procedure was assessed by determining the concentration of calcium, phosphorous and thyroxin in 10 sham-operated control and 10 hypothyroid rats, using standard techniques The results obtained 15 days after surgery were: Ca2+, 10.2 ± mm in control versus 10.3 ± 0.2 mm in hypothyroid; phosphorous, 6.5 ± 0.3 mm in control versus 6.3 ± 0.5 mm in hypothyroid and thyroxine 6.4 ± lgỈL)1 in control versus 11.8 ± 1.9 lgỈL)1 in hypothyroid, P < 0.05 The sham Microdomain components – effect on BAX-induced mPT group (375 ± 14 g) underwent a surgical procedure in which the animals were anesthetized, the trachea was exposed and the incision was closed, simulating thyroidectomy Purification of BAX-DC (rBAX) protein Recombinant GST–BAX-DC (protein lacking the C-terminal 20 amino acids) was prepared according to Xie et al [41] with slight modifications Briefly, Escherichia coli BL21(DE3)pLysS cells carrying the plasmid pGEX-4T1– BAX-DC were grown overnight at 37 °C in Luria Bertani medium supplemented with 100 lgỈmL)1 ampicillin, 25 lgỈmL)1 chloramphenicol and 1% glucose Cells were cultured overnight at 37 °C after induction with 0.4 mm isopropyl-b-d-thiogalactoside Harvested cells were disrupted with lysozyme in the presence of 1% TX 100 and sonicated in NaCl ⁄ Pi, pH 7.4 Unbroken cells were eliminated after centrifugation at 5000 g, for 10 High speed centrifugation (100 000 g for 30 at °C) was performed to clear the supernatant, before glutathione–Sepharose affinity chromatography Unbound protein was washed extensively using NaCl ⁄ Pi, pH 7.4, supplemented with 0.1% TX 100 and the protein was recovered by overnight proteolytic cleavage with thrombin at room temperature Eluted rBAX was dialyzed against 250 vol 10 mm Tris ⁄ HCl (pH 8.0), mm EDTA, and 0.1% (v ⁄ v) 2-b-mercaptoethanol rBAX was aliquoted and stored at )70 °C rBAX oligomerization rBAX oligomerization was induced by incubating the protein with 1% octylglucoside for 60 at °C The protein was then diluted 10· in the assay buffer and incubated with isolated mitochondria, before measurement of mitochondrial permeability transition Mitochondrial permeability transition pore opening induced by rBAX Mitochondria were isolated from rat kidney cortex by differential centrifugation in a sucrose-based medium, as described previously [42] Mitochondrial protein (1.3 mg) was incubated for 15 in the presence of oligomerized rBAX at the indicated concentration Opening of the mPTP was evaluated by measuring mitochondrial calcium movements, mitochondrial swelling and Dw in a basic medium containing, 125 mm KCl, 10 mm Hepes, 10 mm succinate, mm Pi, pH 7.4, plus 200 lm ADP, lgỈmL)1 rotenone and, where indicated, 50 lm CaCl2 Mitochondrial calcium uptake was evaluated spectrophotometrically in a double beam spectrophotometer at 675– 685 nm, by using the metallochromic dye Arsenazo III [43] Mitochondrial swelling was followed by changes in FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS 6507 ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT absorbance at 540 nm and Dw was determined by using the dye safranine, at 524–554 nm [44] Measurement of NAD+ content in mitochondria Mitochondrial NAD+ was measured after perchloric acid extraction as described previously [45], with minor modifications We added 0.5 mL of 21% (v ⁄ v) HClO4 to 10 mg mitochondrial protein per mL of suspension and incubated during 30 in an ice-cold bath The suspension was centrifuged at 8000 g and the supernatans neutralized NAD+ was determined fluorometrically at kex ¼ 340 nm and kem ¼ 460 nm, by measuring NAD+-dependent lactate dehydrogenase activity, in a medium containing lg of lactate dehydrogenase from rabbit muscle, 400 mm Hydrazine, 500 mm glycine and 10 mm l-lactate pH 9.0, at 25 °C rBAX insertion into mitochondria and cytochrome c release rBAX insertion and cytochrome c content in mitochondria were evaluated by western blot analysis, using a primary anti-BAX mAb (1 lgỈmL)1) or a mAb against cytochrome c (1 : 1000 dilution) Biotin-conjugated secondary antibodies and streptavidin–peroxidase conjugate were used followed by an enhanced chemiluminescence system detection After maximal calcium release, 1.8 mL of the suspension was withdrawn and centrifuged at 18 000 g for 10 To discard the protein loosely bound to the membranes, the mitochondrial pellets were incubated with 0.1 m Na2CO3, pH 11.5 (alkaline extraction), for 10 at room temperature, a procedure that assures that only full inserted BAX would be detected Mitochondria were recovered by centrifugation, washed once with 250 mm sucrose, 10 mm Tris, pH 7.4, and suspended in sample buffer for electrophoresis Total mitochondrial protein (50 lg) was loaded into each lane of the SDS ⁄ PAGE gels and transferred to poly(vinylidene difluoride) membranes for immunodetection To assess protein loading, the membranes were ‘stripped’ in a buffer containing 62.5 mm Tris ⁄ HCl, 100 mm b-mercaptoethanol, 2% SDS, pH 6.7, for 20 at 50 °C The membranes were incubated against anti-ANT or anti-VDAC polyclonal IgG Cholesterol and ganglioside measurements Control and hypothyroid mitochondria were incubated with the cholesterol-chelator MbCD for 25 at 25 °C and mild shaking After depletion, mitochondria were centrifuged at 8000 g and suspended in NaCl ⁄ Pi, pH 7.4, for further analysis Total cholesterol was determined from mitochondrial lipid extracts obtained with chloroform ⁄ methanol (2 : 1) in the presence of 0.02% butylhydroxy-toluene and using 50 lg stigmasterol as the 6508 internal standard, as described previously [46] Lipid extracts were incubated overnight at room temperature with hexamethyldisilazane and trimethylchlorsilane in dry pyridine Cholesterol was purified in a fused silica noncapillary column SE 54 (30 cm, 0.35 mm id) and quantified by gas chromatography using a gas chromatograph (Carlo Erba 2003), equipped with a flame ionization detector and a split-less inlet system Gangliosides were extracted from control and hypothyroid mitochondria previously treated with MbCD and analyzed by preparative TLC on silica gel plates with chloroform ⁄ methanol ⁄ 0.2% CaCl2 ⁄ acetic acid (7 : : 0.5 : 0.2 v ⁄ v ⁄ v ⁄ v), as described previously [47] The plates were sprayed with 10% H2SO4 (v ⁄ v) and heated for spot detection GD3 standard gave two equivalent spots (RF ¼ 0.85 and 0.81) The double band is known to result from the heterogeneity of fatty acid composition in gangliosides Positive signal in the samples correlated with the band presenting a retention factor of 0.85 from the GD3 standard No signal was detected comigrating with the GM1 standard Alternatively, gangliosides were specifically detected with resorcinol reagent [48] and compared with the signal developed by H2SO4 using a simple ‘dot’ assay Cardiolipin content was analyzed after inner mitochondrial phospholipid extraction with chloroform ⁄ methanol (2 : v ⁄ v), then individual phospholipids were separated by TLC on silica gel plates with chloroform ⁄ methanol ⁄ H2O (65 : 25 : v ⁄ v ⁄ v) and visualized with iodine Spots comigrating with the cardiolipin standard were scraped out from the chromatogram and quantified spectrophotometrically at 472 ⁄ 492 nm using acridine orange, as described [49] Statistical analysis Values are given as means ± SD and were evaluated by one-way ANOVA P < 0.05 was considered the threshold for statistical significance between the groups indicated Acknowledgments We greatly acknowledge the generous gift of pGEX4T-I BAX plasmid from Dr John Reed (Burnham Institute for Medical Research, La Jolla, CA) and partial support from the Doctorate Program in Biomedical Sciences of the National Autonomous University of Mexico (UNAM) to Eduardo Martı´ nez-Abundis ´ We also thank Mr Jose Santamarı´ a (Nephrology Department, National Institute of Cardiology, Ignacio ´ Chavez) for performing the surgical procedures This study was partially supported by grant 46456-M to CZ from the National Council of Science and Technology ´ (CONACYT), Mexico The authors state no conflict of interest FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT References Wolter G, Hsu Y, Smith C, Nechushtan A, Xi X & Youle K (1997) Movement of Bax from the cytosol to mitochondria during apoptosis J Cell Biol 139, 1281– 1292 Zhang H, Heim J & Meyhack B (1998) Redistribution of BAX from cytosol to membranes is induced by apoptotic stimuli and is an early step in the apoptotic pathway Biochem Biophys Res Commun 251, 254–259 Nechushtan A, Smith C, Hsu Y & Youle K (1999) Conformation of the BAX C-terminus region regulates subcellular location and cell death EMBO J 18, 2330– 2341 Schinzel A, Kaufmann T, Schuler M, Martinalbo J, Grubb D & Borner C (2005) Conformational control of Bax localization and apoptotic activity by Pro168 J Cell Biol 164, 1021–1032 Zou H, Li Y, Liu X & Wang X (1999) An Apaf-1–cytochrome c multimeric complex is a functional apoptosome that activates pro-caspase-9 J Biol Chem 274, 11549–11556 Yi X, Yin XM & Dong Z (2003) Inhibition of Bidinduced apoptosis by Bcl-2 tBid insertion, Bax translocation and Bax ⁄ Bak oligomerization supressed J Biol Chem 278, 16992–16999 Jiang M, Pabla N, Murphy RF, Yang T, Yin XM, Degenhardt K, White E & Dong Z (2007) Nutlin-3 protects kidney cells during cisplatin therapy by suppressing Bax ⁄ Bak activation J Biol Chem 282, 2636–2645 Li H, Zhu H, Xu CJ & Yuan J (1998) Cleavege of BID by caspase mediates the mitochondrial damage in the Fas pathway of apoptosis Cell 94, 491–501 Wei MC, Zong WX, Cheng EH, Lindsten T, Papautsakopoulou V, Ross AJ, Roth KA, MacGregor GR, Thompson CB & Korsmeyers J (2001) Proapoptotic BAX and BAK: a requisite gateway to mitochondrial disfunction and death Science 292, 727–730 10 Zong WX, Lindsten T, Ross AJ, MacGregor GR & Thompson CB (2001) BH-only proteins that bind prosurvival Bcl-2 family members fail to induce apoptosis in the absence of Bax and Bak Genes Dev 15, 1481– 1486 11 Basanez G, Sharpe J, Galanis J, Brandt T, Hardwick J ˜ & Zimmerberg J (2002) Bax-type apoptotic proteins porate pure lipid bilayers through a mechanism sensitive to intrinsic monolayer curvature J Biol Chem 277, 49360–49365 12 Gogvadze V, Robertson J, Zhivotovsky B & Orrenius S (2001) Cytochrome c release occurs via Ca2+-dependent and Ca2+-independent mechanisms that are regulated by Bax J Biol Chem 276, 19066–19071 13 Vyssokikh M, Zorova L, Zorov D, Heimlich G, Jurgensmeier J & Brdiczka D (2002) Bax releases cytochrome c preferentially from a complex between 14 15 16 17 18 19 20 21 22 23 24 25 porin and adenine nucleotide translocator Hexokinase activity suppresses this effect Mol Biol Rep 29, 93–96 Shimizu S, Ide T, Yanagida T & Tsujimoto Y (2000) Electrophysiological study of a novel large pore formed by BAX and the voltage-dependent anion channel that is permeable to cytochrome c J Biol Chem 275, 1231– 1235 Pike L (2004) Lipid rafts: heterogeneity on the high seas Biochem J 378, 281–292 Garofalo T, Giammarioli A, Misasi R, Tinari A, Manganelli V, Gambardella L, Pavan A, Malorni W & Sorice M (2005) Lipid microdomains contribute to apoptosis-associated modifications of mitochondria in T cells Cell Death Differ 12, 1378–1389 Scorrano L, Petronilli V, Di Lisa F & Bernardi P (1999) Commitment to apoptosis by GD3 ganglioside depends on the opening of the mitochondrial transition pore J Biol Chem 274, 22581–22585 ´ Garcı´ a-Ruiz C, Colell A, Parı´ s R & Fernandez-Checa J (2000) Direct interaction of GD3 ganglioside with mitochondria generates reactive oxygen species followed by mitochondrial permeability transition, cytochrome c release and caspase activation FASEB J 14, 847–858 Rippo M, Malisan F, Ravagnan L, Tomassini B, Condo I, Costantini P, Susin S, Rufino A, Todazo M, Kroemer G et al (2000) GD3 ganglioside directly targets mitochondria in bcl-2-controlled fashion FASEB J 14, 2047–2054 Pastorino J, Tafani M, Rothman R, Marcinkeviciute A, Hoek J & Farber J (1999) Functional consequences of the sustained or transient activation by Bax of the mitochondrial transition pore J Biol Chem 274, 31734– 31739 Brown D & London E (2000) Structure and function of sphingolipid- and cholesterol-rich membrane rafts J Biol Chem 275, 17221–17224 Keller R, Arnold T & Fliester S (2004) Formation of 7-dehydrocholesterol containing membrane rafts in vitro and in vivo, with relevance to the Smith–Lemli–Opitz syndrome J Lipid Res 45, 347–355 Paradies G, Ruggiero F & Dinoi P (1991) The influence of hypothyroidism on the transport of phosphate and on the lipid composition in rat-liver mitochondria Biochim Biophys Acta 1070, 180–186 Ardail D, Popa I, Bodennee J, Louisot P, Schmitt D & Portoukalian J (2003) The mitochondria-associated endoplasmic reticulum subcompartment (MAM fraction) of rat liver contains highly active sphingolipid-specific glycosyltransferases Biochem J 371, 1013–1019 Cifone M, Roncaioli P, De Maria R, Camarda G, Santoni A, Ruberti G & Testi R (1995) Multiple pathways originate at the Fas ⁄ APO1 (CD95) receptor: sequential involvement of phosphatidylcholine-specific phospholipase C and acidic sphingomyelinase in the programation of the apoptotic signal EMBO J 14, 5859–5868 FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS 6509 ´ E Martınez-Abundis et al Microdomain components – effect on BAX-induced mPT 26 Testi R (1996) Sphingomyelin breakdown and cell fate Trends Biochem Sci 21, 468–471 27 Nakayama J, Fukuda M, Hirabayashi Y, Kanamori A, Sasaki K, Nishi T & Fukuda M (1996) Expression cloning of a human GT3 synthase GD3 and GT3 are synthesized by a single enzyme J Biol Chem 271, 3684–3691 28 Birbes H, El Bawab S, Hannun Y & Obeid L (2001) A mitochondrial pool of sphingomyelin is involved in TNF-a-induced BAX translocation to mitochondria FASEB J 14, 2669–2679 29 Brenner C, Cadiou H, Vieira H, Zamzani N, Marzo I, Xie Z, Leber B, Andrews D, Duclohier H, Reed J et al (2000) Bcl-2 and Bax regulate the channel activity of the mitochondrial adenine nucleotide translocator Oncogene 19, 329–336 30 Guha M, Kumar S, Coubev V, Maity P & Bandyopadhyay U (2006) Apoptosis in liver during malaria: role of oxidative stress and implication of mitochondrial pathway FASEB J 20, 1224–1226 31 Kuwana T, Mackey MR, Perkins G, Ellisman MH, Latterich M, Schneiter R, Green DR & Newmeyer DD (2002) Bid, Bax and lipids cooperate to form supramolecular openings in the outer mitochondrial membrane Cell 111, 331–342 32 Ilangumaran S & Hoessli DC (1998) Effects of cholesterol depletion by cyclodextrin on sphingolipid microdomains of the plasma membrane Biochem J 335, 433–440 33 Martens J, Navarro-Polanco R, Coppock E, Nishiyama A, Parshley L & Grobaski T (2000) Differential targeting of shaker-like potassium channels to lipid rafts J Biol Chem 275, 7443–7446 34 Saito M & Sugiyama K (2000) Specific ganglioside changes in extraneural tissues of adult rats with hypothyroidism Biochim Biophys Acta 1523, 230–235 35 Brand MD, Steverding D, Kadenbach S, Stevenson P & Hafner RP (1992) The mechanism of the increased proton permeability induced by thyroid hormone Eur J Biochem 264, 7826–7830 ´ 36 Chavez E, Franco M, Reyes-Vivas H, Zazueta C, Ramı´ rez J & Carrillo R (1998) Hypothyroidism renders liver mitochondria resistant to the opening of membrane permeability transition pore Biochim Biophys Acta 1407, 243–248 37 Terrones O, Antonsson B, Yamaguchi H, Wang H, Liu Y, Lee R, Herrmann A & Basanez G (2004) Lipid pore ˜ formation by the concerted action of pro-apoptotic BAX and tBID J Biol Chem 279, 30081–30091 6510 38 Wieckowski M, Vyssokikh M, Dymkowska D, Antonsson B, Brdiczka D & Wojtczak L (2001) Oligomeric C-terminal truncated Bax preferently releases cytochrome c but no adenylate kinase from mitochondria, outer membrane vesicles and proteoliposomes FEBS Lett 505, 453–459 39 Epand RF, Schlattner U, Wallimann T & Epand RM (2007) Cardiolipin clusters and membrane domain formation induced by mitochondrial proteins J Mol Biol 365, 968–980 ´ ´ 40 Franco M, Bobadilla N, Suarez J, Tapia E, Sanchez L & Herrera-Acosta J (1996) Participation of adenosine in the renal hemodynamic abnormalities of hypothyroidism Am J Physiol 270, F254–F262 41 Xie Z, Schendel S, Matsuyama S & Reed J (1998) Acidic pH promotes dimerization of Bcl-2 family proteins Biochemistry 37, 6410–6418 ´ 42 Chavez E, Briones R, Michel B, Bravo C & Jay D (1985) Evidence for the involvement of dithiol groups in mitochondrial calcium transport: studies with cadmium Arch Biochem Biophys 242, 293–297 43 Kendrick N (1976) Purification of arsenazo III, a calcium sensitive dye Ann Biochem 76, 487–501 44 Zanotti A & Azzone GF (1980) Safranine as a membrane potential probe in rat liver mitochondria Arch Biochem Biophys 201, 255–265 45 Di Lisa F, Menabo R, Canton M, Barile M & Bernardi P (2001) Opening of the mitochondrial permeability transition pore causes depletion of mitochondrial and cytosolic NAD+ and is a causative event in the death of myocytes in postischemic reperfusion of the heart J Biol Chem 276, 2571–2575 ´ 46 Carvajal K, El Hafidi M, Marı´ n-Hernandez A & ´ Moreno-Sanchez R (2005) Structural and functional changes in heart mitochondria from sucrose-fed hypertriglyceridemic rats Biochim Biophys Acta 1709, 231– 239 47 Miyazaki K, Okamura N, Kishimoto Y & Lee Y (1986) Determination of gangliosides as 2,4-dinitrophenylhydrazides by high-performance liquid chromatography Biochem J 235, 755–761 48 Svennerholm L (1957) Quantitative stimation of sialic acids II A colorimetric resorcinol–hydrochloric acid method Biochim Biophys Acta 617, 97–109 49 Petit JM, Huet O, Gallet P, Maftah A, Ratinaud M & Julien R (1994) Direct analysis and significance of cardiolipin transverse distribution in mitochondrial inner membrane Eur J Biochem 220, 871–879 FEBS Journal 274 (2007) 6500–6510 ª 2007 The Authors Journal compilation ª 2007 FEBS ... mPTP-opening induced by rBAX To gain insight into the existence of putative mitochondrial lipid microdomains in vivo, we evaluated the inducing effect of rBAX on opening of the mPTP in kidney... glycosphingolipid fractions, obtained from the same amount of mitochondrial protein, were loaded and developed with sulfuric acid (unspecific) and resorcinol (ganglioside -specific stain) By increasing... Effect of rBAX on mitochondrial swelling and Dw (A) Mitochondrial swelling induced by rBAX was monitored as absorbance changes at 540 nm Mitochondrial protein (1.3 mg) was incubated with 150