After Embedding in Membranes Antiapoptotic Bcl-XL Protein Binds Both Bcl-2 Homology Region 3 and Helix 1 of Proapoptotic Bax Protein to Inhibit Apoptotic Mitochondrial Permeabilization

Jialing LinChengguo Xing, David W. Andrews and McNichol, Bo Huang, Xuejun C. Zhang,Zhang, Justin Kale, Domina Falcone, Jamie Jingzhen Ding, Blaine H. M. Mooers, Zhi Apoptotic Mitochondrial PermeabilizationProapoptotic Bax Protein to Inhibit Bcl-2 Homology Region 3 and Helix 1 ofAntiapoptotic Bcl-XL Protein Binds Both After Embedding in MembranesCell Biology:

doi: 10.1074/jbc.M114.552562 originally published online March 10, 20142014, 289:11873-11896.J. Biol. Chem.

10.1074/jbc.M114.552562Access the most updated version of this article at doi:

.JBC Affinity SitesFind articles, minireviews, Reflections and Classics on similar topics on the

Alerts:

When a correction for this article is posted•

When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/289/17/11873.full.html#ref-list-1

This article cites 61 references, 21 of which can be accessed free at

at Univ of O

klahoma H

ealth Science Library on O

ctober 2, 2014http://w

ww

.jbc.org/D

ownloaded from

at U

niv of Oklahom

a Health Science L

ibrary on October 2, 2014

http://ww

w.jbc.org/

Dow

nloaded from

http://affinity.jbc.org/

http://cell.jbc.org

http://memb.jbc.org

http://www.jbc.org/lookup/doi/10.1074/jbc.M114.552562

http://affinity.jbc.org

http://www.jbc.org/cgi/alerts?alertType=citedby&addAlert=cited_by&cited_by_criteria_resid=jbc;289/17/11873&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/289/17/11873

http://www.jbc.org/cgi/alerts?alertType=correction&addAlert=correction&correction_criteria_value=289/17/11873&saveAlert=no&return-type=article&return_url=http://www.jbc.org/content/289/17/11873

http://www.jbc.org/cgi/alerts/etoc

http://www.jbc.org/content/289/17/11873.full.html#ref-list-1

http://www.jbc.org/

http://www.jbc.org/

After Embedding in Membranes Antiapoptotic Bcl-XLProtein Binds Both Bcl-2 Homology Region 3 and Helix 1 ofProapoptotic Bax Protein to Inhibit Apoptotic MitochondrialPermeabilization*

Received for publication, January 22, 2014, and in revised form, March 7, 2014 Published, JBC Papers in Press, March 10, 2014, DOI 10.1074/jbc.M114.552562

Jingzhen Ding‡, Blaine H. M. Mooers‡§, Zhi Zhang‡, Justin Kale¶, Domina Falcone¶, Jamie McNichol¶, Bo Huang�,Xuejun C. Zhang�, Chengguo Xing**, David W. Andrews¶‡‡1, and Jialing Lin‡§2

From the ‡Department of Biochemistry and Molecular Biology and the §Peggy and Charles Stephenson Cancer Center, Universityof Oklahoma Health Sciences Center, Oklahoma City, Oklahoma 73126, the ¶Department of Biochemistry and BiomedicalSciences, McMaster University, Hamilton, Ontario L8N 3Z5, Canada, the �Institute of Biophysics, Chinese Academy of Sciences,Beijing 100101, China, the **Department of Medicinal Chemistry, University of Minnesota, Minneapolis, Minnesota 55455,and ‡‡Biological Sciences, Sunnybrook Research Institute and Department of Biochemistry, University of Toronto,Toronto, Ontario M4N 3M5, Canada

Background: Bcl-XL binds Bax at mitochondria, inhibiting Bax oligomerization and apoptosis.Results:Anchored inmembranes, Bcl-XL�Baxheterodimer is formed froma rigid helix-in-groove interface plus a flexible helicaldimer interface.Conclusion: Two interfaces contribute equally to the heterodimer stability required to inhibit Bax.Significance: This novel kind of protein-protein interaction stabilizes the membrane-bound heterodimer that is pivotal toapoptosis regulation.

Bcl-XL binds to Bax, inhibiting Bax oligomerization requiredfor mitochondrial outer membrane permeabilization (MOMP)during apoptosis. How Bcl-XL binds to Bax in the membrane isnot known. Here, we investigated the structural organization ofBcl-XL�Bax complexes formed in theMOM, including the bind-ing interface and membrane topology, using site-specific cross-linking,compartment-specific labeling,andcomputationalmodel-ing. We found that one heterodimer interface is formed by aspecific interaction between the Bcl-2 homology 1–3 (BH1–3)grooveofBcl-XLand theBH3helixofBax, asdefinedpreviouslybythe crystal structure of a truncated Bcl-XL protein and a Bax BH3peptide (Protein Data Bank entry 3PL7). We also discovered anovel interface in the heterodimer formed by equivalent inter-actions between the helix 1 regions of Bcl-XL and Bax whentheir helical axes are oriented either in parallel or antiparal-lel. The two interfaces are located on the cytosolic side of theMOM, whereas helix 9 of Bcl-XL is embedded in the mem-brane together with helices 5, 6, and 9 of Bax. Formation ofthe helix 1�helix 1 interface partially depends on the forma-tion of the groove�BH3 interface because point mutations inthe latter interface and the addition of ABT-737, a groove-binding BH3 mimetic, blocked the formation of both inter-faces. The mutations and ABT-737 also prevented Bcl-XL from

inhibiting Bax oligomerization and subsequent MOMP, sug-gesting that the structural organization in which interactions atboth interfaces contribute to the overall stability and function-ality of the complex represents antiapoptotic Bcl-XL�Bax com-plexes in the MOM.

Proteins of the B-cell lymphoma 2 (Bcl-2)3 family can inter-act in the mitochondrial outer membrane (MOM) to regulateits permeability. These proteins contain one or more Bcl-2homology (BH) motif and either promote or inhibit the MOMpermeabilization (MOMP) that initiates apoptosis. Bcl-2-asso-ciated X (Bax) and Bcl-2 antagonist/killer 1 (Bak) contain threeBH motifs (BH1–3) that are required for the MOMP and asix-residue structural motif similar to that found in the BH4motif of antiapoptotic family members (1). Bcl-2, B-cell lym-phoma-extra large (Bcl-XL), and other antiapoptotic familymembers contain four BHmotifs (BH1–4) and function as bothdirect and indirect inhibitors of Bax and Bak. BH3-only pro-teins, such as BH3 interacting domain death agonist (Bid), Bcl-2-like protein 11 (Bim), and Bcl-2-associated death promoter(Bad), contain only the BH3 motif and promote the MOMP by

* This work was supported, in whole or in part, by National Institutes of HealthGrants R01GM062964 (to J. L.) and R01AI088011 and P20GM103640 (toB. H. M. M.). This work was also supported by OCAST Grant HR10-121(to J. L.), and Canadian Institutes of Health Research Grant FRN 12517 (toD. W. A. and Brian Leber).

1 Holder of the Tier 1 Canada Research Chair in Membrane Biogenesis.2 To whom correspondence should be addressed: 940 Stanton L. Young Blvd.,

Biomedical Sciences Bldg. 935, P.O. Box 26901, Oklahoma City, OK 73126-0901. Tel.: 405-271-2227 (ext. 61216); Fax: 405-271-3092; E-mail: [email protected].

3 The abbreviations used are: Bcl-2, B-cell lymphoma 2; Bcl-XL, B-cell lymphoma-extra large; ANB, 5-azido-2-nitrobenzoyl; Bad, Bcl-2-associated death pro-moter; Bak, Bcl-2 antagonist/killer 1; Bax, Bcl-2-associated X; BH, Bcl-2homology; Bid, BH3 interacting-domain death agonist; tBid, truncated Bid;Bim, Bcl-2-like protein 11; BMH, bismaleimidohexane; CuPhe, Cu(II)(1,10-phenanthroline)3; IASD, 4-acetamido-4�-((iodoacetyl)amino)stilbene-2,2�-disulfonic acid; IEF, isoelectric focusing; IRES, internal ribosome entry site;MOM, mitochondrial outer membrane; MOMP, mitochondrial outermembrane permeabilization; Bak, Bcl-2 antagonist/killer 1; MEF, mouseembryo fibroblast; 6H, His6; PDB, Protein Data Bank.

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 289, NO. 17, pp. 11873–11896, April 25, 2014© 2014 by The American Society for Biochemistry and Molecular Biology, Inc. Published in the U.S.A.

APRIL 25, 2014 • VOLUME 289 • NUMBER 17 JOURNAL OF BIOLOGICAL CHEMISTRY 11873

at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

activating Bax and Bak, inhibiting the antiapoptotic familymembers, or both (2–4).Bax is inactive in the cytosol of normal cells but becomes

active in apoptotic cells by changing conformation in a multi-step process. First, Bax is recruited to themitochondria by trun-cated Bid (tBid) in the MOM (5, 6). The membrane-bound Baxrecruits more soluble Bax to the MOM by autoactivation (7).These Bax proteins bind to each other to form an oligomericpore in the membrane. The oligomerization most likely takesplace after Bax changes to a multispanning conformation withhelices 5, 6, and 9 embedding in the membrane (8). The struc-ture of the oligomeric Bax pore in the membrane is largelyunknown. However, the data from several recent studies sug-gest a possible consensus structure. Our photocross-linkingstudy revealed two interdependent interfaces in the Bax oligomerthat was formed in detergent micelles. The first one is formed bytheBH1–3 regions, and the secondonewas formedbyhelix 1 plusthe following loop and helix 6 (9). An electron paramagnetic reso-nance study of Bax oligomers that were formed in detergentmicelles and liposomalmembranes togetherwithhomologymod-eling of the Bax dimer using a Bak dimer model suggested anantiparallel helical dimer interface formed by the helix 2–3region of neighboring Baxmolecules in the oligomer (10, 11). Adisulfide cross-linking study indicated that the helix 2–3 inter-face was extended to include helix 4, which binds to other sideof helix 2, resulting in a BH3-in-groove dimer interface (12).This interface was also observed in a crystallographic studyusing a Bax fragment containing the helix 2–5 region (13). TheBH3-in-groove dimer interface together with another parallelhelical dimer interface formed by helix 6 of neighboring Baxmolecules could generate a Bax oligomer in the MOM (12).To inhibit Bax oligomerization, Bcl-XL could inhibit any one

or more of these Bax homointeractions, thereby interruptingthe activation process. In addition, it could also sequester tBid,preventing tBid from binding to and activating Bax. Neither ofthese potential inhibitory functions of Bcl-XL is mutuallyexclusive. Thus, interaction with Bax could inhibit the confor-mational change in Bax, preventing it from adopting the mul-tispanning state, the oligomerization of the multispanningmonomers, or both. Early mutational studies indicate that theBH1–3motifs of Bcl-XL and theBH3motif of Baxmediate theirinteraction. For example, replacement of Asp68 with arginine(D68R) in the BH3 region of Bax greatly impaired its interactionwith antiapoptotic family members (14). In addition, replace-ment of Gly138 with alanine (G138A) in the BH1 region ofBcl-XL abolished its binding to Bax and inhibition of apoptosis(15, 16). The crystal structure of a truncated Bcl-XL protein incomplexwith a peptide containing theBH3 region of Bax showsthat the peptide binds as an amphipathic� helix to a hydropho-bic groove formed by the BH1–3 regions of Bcl-XL (17).Although all of these studies provide clues, the complex struc-ture formedby the full-lengthBcl-XL andBax proteins inmem-branes remains elusive.Bcl-XL, like Bax, is a soluble cytosolic or peripheral mem-

brane protein in normal cells. Upon activation by membrane-bound tBid or Bax, Bcl-XL integrates into the MOM (5). Themembrane topology of the active Bcl-XL is not known. How-ever, the homologous Bcl-2 changes conformation during apo-

ptosis initiation, and this conformational change is required forits antiapoptotic activity (18–20).The following questions about Bcl-XL�Bax interaction are

important and have not been answered.Does Bcl-XL binding toBax prevent Bax from changing to the multispanning state, ordoes it prevent Bax oligomerization? Is the BH1–3 groove�BH3helix interface in the crystal structure relevant to the Bcl-XL�Bax complex formed in the MOM? Are there additionalregions of Bcl-XL and Bax involved in the complex formation,as we previously detected in the Bcl-2�Bax complex that wasformed in detergentmicelles (21)? If a second interface, in addi-tion to the groove�BH3 interface, forms in the membrane-bound Bcl-XL�Bax complex, can ABT-737, a BH3 peptidemimetic that binds to the BH1–3 groove of Bcl-XL in solution(22), still disrupt the Bcl-XL�Bax complex in themembrane? Toaddress these questions, we systematically mapped the inter-face and the topology of the Bcl-XL�Bax heterodimer that wasformed in membranes. We built structural models for the het-erodimer based on the experimental data, and we tested thesemodels with additional experiments that used mutational andfunctional assays. We also monitored the effect of ABT-737 onthe membrane-bound heterodimer, particularly its effect onthe formation of a non-groove�BH3 interface. Together, thesestudies revealed critical structural features of a functional Bcl-XL�Bax heterodimer that was formed in the biological mem-branes. Our results demonstrate that the membrane-boundheterodimer has two interfaces, the conventional groove�BH3interface seen in crystal and a novel helix1�helix 1 interface,which surprisingly can form in either a parallel or antiparallelfashion.

EXPERIMENTAL PROCEDURES

Materials—Phospholipids were purchased from AvantiPolar Lipids. The MOM liposomes consisting of MOM-char-acteristic lipids were made as described (7). The Bax�/�/Bak�/� mitochondria were isolated from the livers of Bakknock-out mice that also lack Bax as described (5). The BaxBH3 peptide that contains Bax residues 53–86, including theBH3 region, was synthesized by Abgent as described (7). Thefull-length human Bax protein with or without an N-terminalHis6 (6H) tag was expressed and purified as described (9, 23).The plasmid for expression of the N-terminal 6H-taggedhuman Bcl-XL in Escherichia coli was modified from thepCYB3-Bcl-XL plasmid by inserting six histidine codonsbetween the first two codons of Bcl-XL. The full-length humanBcl-XL protein with or without the N-terminal 6H tag wasexpressed and purified as described (5, 23), except that the6H-Bcl-XL eluted from the chitin column was purified using aNi2�-nitrilotriacetic acid-agarose column, and the resultingprotein was dialyzed in 20% (v/v) glycerol and 20mMTris/HCl,pH 8.0.Single-cysteine and Single-lysine Bax and Bcl-XL Mutants—

To construct plasmids for in vitro transcription and translationof Bax and Bcl-XL, we inserted the coding region of full-lengthhumanBax orBcl-XL into the vector pSPUTK (Stratagene).Wecreated lysine-null (K0) Bax and Bcl-XL mutant plasmids bychanging all of the lysine codons to arginine codons, and wecreated cysteine-null (C0) Bax and Bcl-XL mutant plasmids by

Architecture of Antiapoptotic Bcl-XL�Bax Dimer in Membranes

11874 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 289 • NUMBER 17 • APRIL 25, 2014

at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

changing all of the cysteine codons to alanine codons. We thencreated Bax and Bcl-XLmutant plasmids with a single lysine orcysteine codon at particular positions by mutating the corre-sponding codons in the K0 or C0 mutant plasmid to lysine orcysteine codon, respectively.MOMP (Cytochrome c Release) Assay—The assay was modi-

fied from that described previously (24).Wild-type andmutantBax and Bcl-XL proteins were synthesized using a transcrip-tion/translation-coupled SP6 RNA polymerase/reticulocytelysate system (Promega). The resulting reaction containing theBcl-XL protein (3 �l), the Bax protein (3 �l), or both was incu-bated with the Bax�/�/Bak�/� mitochondria (0.5 mg/ml totalprotein). Purified tBid protein (17 nM) (23) or Bax BH3 peptide(40 �M) was added to the reactions to activate the Bcl-XL andBax proteins. An adequate volume of buffer A (110 mM KOAc,1mMMg(OAc)2, 25mMHEPES, pH 7.5, and 2mM glutathione)was added to each reaction to bring the total volume to 15 �l.After incubation for 1 h at 37 °C, the samples were centrifugedat 10,000 � g for 10 min. The resulting pellet fractions wereresuspended with 0.5% (v/v) Triton X-100 in phosphate-buff-ered saline (PBS, pH 7.4) to the same volume as the supernatantfractions and centrifuged again at 16,100 � g for 10 min toobtain the second supernatant fractions as the detergent-solu-bilized mitochondrial pellet fractions. The amounts of cyto-chrome c in both supernatant and pellet fractions were mea-sured using an enzyme-linked immunosorbent assay (ELISA)with the antibody against mouse cytochrome c from R&D Sys-tems per its protocol. The fraction of cytochrome c release wascalculated using the formula, [cytochrome c in supernatant]/([cytochrome c in supernatant] � [cytochrome c in pellet]).Apoptotic Activity of Bax Mutants in bax�/�/bak�/� Mouse

Embryonic Fibroblasts (MEFs)—Phoenix cells were seeded in100-mm dishes and MEFs in 96-well plates. The pBabe-MN-Bax-IRES-GFP plasmids containing wild-type Bax, Lys-null, orsingle-lysinemutants were constructed as described previously(9). Each plasmid (10 �g) was transfected into the Phoenix cellswith Exgene500 (Fermentas) to package the plasmid into a rep-lication-incompetent murine virus. The media containing thevirus were harvested 24 h after the transfection, filtered with a0.2-�m filter, and then added into the MEFs. After 48 h ofinfection, theMEFs were treated with 0.5 or 2 �M etoposide for24 h, and three-channel images (green fluorescent protein(GFP), annexin V R-PE, and DRAQ5) were collected from fivefields of view in each well. In each field, cells were identified viathe DRAQ5 imaging, and measurements of emission intensityof GFP and annexin V were taken per cell from their respectiveimages. The total number of cells measured in the five fields ofany well was used to calculate the annexin V-positive percent-age score for that well. The fraction of annexin V-positive cellswas determined for green cells (infected) and non-green cells(uninfected) separately. Expression of GFP alone from theinternal ribosome entry site (IRES) sequence was somewhattoxic. The toxicity was shown in the bax�/�/bak�/� MEFs andpoorly inhibited by Bcl-2 (data not shown); hence, the toxicitywas probably not due to apoptosis. Because co-expression ofwild-type Bax significantly increased the apoptosis, as detectedby annexinV staining, the assay could clearly reveal the apopto-

tic activity of the Bax mutants relative to the wild-type Bax thatwas assayed in parallel.Disulfide and Bismaleimidohexane (BMH) Cross-linking—

[35S]Met-labeled single-cysteine Bax and Bcl-XLmutant proteinswere synthesized in thewheat germ-based in vitro translation sys-tem as described (25). The resulting reaction containing the Baxprotein (10 �l), the Bcl-XL protein (10 �l), or both was incubatedat 37 °C for 1.5 h with the MOM liposomes (4 mM total lipids) orfor 1 h with the Bax�/�/Bak�/� mitochondria (0.5–0.7 mg/mltotal proteins), 36–40�MBH3peptide or 1.2 nM tBidprotein, and1mMdithiothreitol (DTT), and an adequate volume of buffer Ato bring the total reaction volume to 40 �l. A subset of thereactions included 50 �M ABT-737. The resulting proteolipo-somes were isolated by a sucrose gradient centrifugation as250-�l fractions from the top of the gradient (20, 23). Themito-chondria with Bax, Bcl-XL, or both bound were pelleted bycentrifugation and resuspended in 100 �l of buffer A asdescribed (5). These membrane fractions were treated with 1mM NaAsO4 to decrease residual DTT and then with redoxcatalyst Cu(II)(1,10-phenanthroline)3 (CuPhe; consisting of 0.3mM CuSO4 and 1 mM 1,10-phenanthroline) to induce disulfidecross-linking. After incubation on ice for 30 min, the oxida-tion reactions were quenched by 20 mM N-ethylmaleimide(NEM) and 100 mM EDTA. For the “0 min” controls, NEMand EDTA were added to the samples to block the disulfidecross-linking before the addition of CuPhe. To producecross-linking with BMH, the proteoliposomes were incu-bated with 100 mM BMH at room temperature for 30 min,and then the cross-linking reactions were stopped by theaddition of 50 mM �-mercaptoethanol. Both disulfide- andBMH-cross-linked samples were solubilized with 1% (v/v)Triton X-100 before precipitation with trichloroacetic acid.The resulting proteins were analyzed by SDS-PAGE undernon-reducing or reducing conditions. The radioactive pro-teins and their adducts in the gels were detected by phos-phorimaging with a Fuji FLA-9000 image scanner.Photocross-linking—[35S]Met-labeledBax or Bcl-XLproteins

with a single N�-(5-azido-2-nitrobenzoyl)-lysine (ANB-lysine)incorporated at specific locations were synthesized from RNAof the corresponding single-lysine Bax or Bcl-XLmutants usingthe wheat germ-based in vitro translation system as described(25, 26). 10 �l of the resulting Bax or Bcl-XL proteins was incu-bated at 37 °C for 1.5 h with 2.2 �M purified 6H-Bcl-XL or6H-Bax protein, the MOM liposomes of 4 mM total lipids, and36–40�MBH3peptide in a 20-�l reaction adjusted by bufferA.The proteoliposomes were isolated as described above andphotolyzed to induce cross-linking via the 5-azido-2-nitroben-zoyl (ANB) probe. The resulting photoadducts of the [35S]Met-labeled protein and the 6H-tagged protein were solubilized in0.25% (v/v) Triton X-100, enriched on Ni2�-nitrilotriaceticacid-agarose, eluted, and analyzed with reducing SDS-PAGEand phosphorimaging.4-Acetamido-4�-((iodoacetyl)amino)stilbene-2,2�-disulfonic

Acid (IASD) Labeling—[35S]Met-labeled single-cysteine Bax orBcl-XL mutant proteins were synthesized by using either thewheat germ- or reticulocyte lysate-based in vitro translationsystem as described (25, 26). The resulting reactions containingthe Bax protein (10�l), the Bcl-XL protein (10�l), or bothwere



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

incubated at 37 °C for 1.5 h with the MOM liposomes with 4mM total lipids, 36–40 �M BH3 peptide, and an adequatevolume of buffer A to bring the total reaction volume to 40�l. Asubset of the reactions also included 0.5 �M purified rec-ombinant Bax or Bcl-XL protein. The resulting prot-eoliposomes were isolated as described above and labeled with0.2 mM IASD at room temperature in the dark for 30min in theabsence or presence of 2% (w/v) CHAPS, 4 M urea, or both. Thereactions were quenched by the addition of 10 mM �-mercapt-oethanol. For the “0 min” controls, the �-mercaptoethanol wasadded before the addition of IASD to prevent the labeling. Theradioactive IASD-labeled proteins were resolved from theunlabeled ones by isoelectric focusing (IEF) (8) and detectedbyphosphorimaging. The intensitiesof the IASD-labeledandunl-abeled bands for each protein under each condition were quant-ified using the program ImageQuant fromFuji. The integrated in-tensities were used to calculate the fraction of IASD labeling acc-ording to the formula, intensity of labeled band(s)/(intensity oflabeled band(s) � intensity of unlabeled band(s)).Modeling of Bcl-XL�Bax Heterodimer Interface and Mem-

brane Topology—The BH1–3 groove�BH3 helix interface forthe Bcl-XL�Bax heterodimer was modeled based on the crystalstructure of a Bcl-XL protein�Bax BH3 peptide complex (PDBentry 3PL7) (17). Themodel was used to predictmutations thatwould disrupt the heterodimer interface, and the predictionswere tested experimentally to verify the model.The helix 1�helix 1 interface for the heterodimer was mod-

eled based on the contact information of specific residues inthese regions provided by the disulfide cross-linking experi-ments. The structure of Bax helix 1 was extracted from theNMR structure of a Bax monomer (PDB entry 1F16) and thenpositioned in a parallel orientation with the Bcl-XL helix 1 inthe Bcl-XL protein�Bax BH3 peptide complex structure (PDBentry 3PL7) by interactive molecular modeling using the pro-gram PyMOL (Schrodinger, LLC). Bad contacts between thetwo helices were relieved by changing the rotameric state ofseveral side chains by using theDunbrack backbone-dependentrotamer library (27) via themutagenesis wizard in PyMOL. Themutagenesis wizard was also used to change several residues tocysteines in order to fit the model to the disulfide cross-linkingdata. We used geometry regularization in the program COOT(28) to remove any distortions in the backbone that were intro-duced while changing the amino acids. The stereochemicalquality of the backbone was checked with a Ramachandranplot, and the atom-atom clashes were scored using the programMOLPROBITY in the presence of hydrogen atoms added usingthe electron-cloud X–H bond lengths (29). Themodeling exer-cise was repeated with the Bax helix 1 rotated into an antipar-allel orientationwith the Bcl-XLhelix 1 to fit with another set ofdisulfide cross-linking data. The parallel and antiparallel mod-els were used as starting models in the Rosetta-based programFlexPepDock (30, 31) to optimize the position of the peptideand the conformation of the side chains in and near the inter-face. Two hundred trial structures were generated and scoredfor each model. The 10 trial structures with the most favorablescores were inspected with PyMOL. The parallel model andantiparallel model with the best agreement with the disulfidecross-linking data were selected for further analysis.

The overall structural organizationmodel for the Bcl-XL�Baxheterodimer bound to the membrane was constructed from acomposite of the BH1–3 groove�BH3 and parallel helix 1�helix 1interface models described above. Helix 9 of Bcl-XL was builtfrom residues 206–230 using the program PHYRE2 (32). Heli-ces 4, 5, 6, 7, 8, and 9 of Bax were extracted from the Bax mon-omer structure (PDB entry 1F16). Using PyMOL, these heliceswere added to the composite structure such that helix 9 ofBcl-XL and helices 5, 6, and 9 of Bax were located under avirtual plane representing the cytosol-MOM boundary,whereas helices 4, 7, and 8 of Bax were located above the planewith the composite structure. Further, the residues that wereburied in the membrane, as indicated by the IASD labelingresults, were placed under the plane, whereas the residues thatwere exposed to the aqueous milieu were placed above theplane. The positions of these helices relative to each other andto the composite structure were adjusted such that the dis-tances between the indicated residues were in the rangeexpected for the residues that could be cross-linked by BMH.

RESULTS

Experimental Design—The interface of Bcl-XL�Bax het-erodimer that was formed in membranes was mapped usingsite-specific disulfide cross-linking and photocross-linking.The former provided structural information about the interfaceat a higher resolution than the latter, because a disulfide bondcan formonly when theC1

�-C2� distance of the two cysteines is

�3–5 Å and the dihedral angle (C1�-S1�-S2�-C2

�) about thedisulfide bond is close to �90° (33). However, disulfide cross-linking occurs on a time scale of seconds tominutes (34, 35) andmay capture both the most abundant Bcl-XL�Bax complex thatis in the most stable conformation and the rare complexes thatare in less stable conformations and reached during the reac-tion time due to thermal fluctuations. In contrast, the photo-cross-linking occurs within nanoseconds at a very low effi-ciency (36), thereby providing a “snapshot” of only the moststable and abundant complex while not sampling the less stableand rare complexes.The topology of Bcl-XL�Bax heterodimer in membranes was

determined by labeling the membrane-bound proteins withIASD, a sulfhydryl-reactive and membrane-impermeant rea-gent with two negative charges. A cysteine in the protein that isexposed to the aqueous milieu will be labeled. On the otherhand, a cysteine that is buried in a hydrophobic environment(e.g. a lipid bilayer, a protein core, or a protein complex inter-face) will not be labeled, unless a detergent, such as CHAPS, ora denaturant, such as urea, is added to disrupt the bilayer ordenature the protein or protein complex, respectively (8).Two complementary in vitromembrane systems were used.

First, the synthetic liposomal membrane consisting of theMOM-characteristic lipids but not proteins (7) simplified thedata interpretation. Second, the isolated mitochondria pro-vided a native environment for protein interactions for verifi-cation of the results from the liposome system.To ensure that the structural information obtained from the

above assays was from functional Bcl-XL�Bax complexes inmembranes, only Bcl-XL and Bax mutants that were active inthe MOMP assay were used in the cross-linking and labeling



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

experiments. Moreover, mutations that could disrupt the het-erodimer formation were generated, and their effects on inhi-bition of Bax-mediated MOMP by Bcl-XL were determined todefine the structure-function relationship.The Bcl-XL and Bax Mutants Used in the Interface- and

Topology-mapping Experiments are Functional—In order tomap the interface and topology of the Bcl-XL�Bax complexformed in membranes, we made Bcl-XL and Bax mutants witha single cysteine or lysine at specific locations (Fig. 1) (9). Themutant proteins were synthesized in vitro using a coupled tran-scription/translation system, and their MOMP activities weremeasured in an ELISA-based cytochrome c release assay (24).Each Bax mutant was activated by either purified tBid proteinor a Bax peptide consisting of the BH3 region (BH3 peptide)and incubated with the mitochondria that lack both Bax andBak (Bax�/�/Bak�/� mitochondria) either in the absence orpresence of a Bcl-XL mutant. The resulting cytochrome crelease was quantified. Like wild-type Bax, each single-cysteineBax mutant released cytochrome c in a tBid-dependent (Fig.2A) or BH3 peptide-dependent (data not shown) manner. Likewild-typeBcl-XL, each single-cysteineBcl-XLmutant inhibitedthe cytochrome c release by the single cysteine Baxmutant thatwas paired with the Bcl-XL mutant in the disulfide cross-link-ing experiments (Fig. 2A). The additional single-cysteine Baxand Bcl-XL mutants used in IASD labeling and BMH cross-linking experiments were also active (Fig. 2B). All of the single-lysine Bcl-XL mutants used in the photocross-linking experi-ments had activities similar to that of the wild-type protein andthe lysine-null (K0)mutant (Fig. 2C). All of the single-lysine Baxmutants used in the photocross-linking experiments wereactive in the MOMP assay (Fig. 2C), in Bax and Bak doubleknock-out mouse embryonic fibroblast cells (bax�/�/bak�/�

MEFs) (Fig. 2D), in HEK293 cells (9), or in both cells.In Membranes, Bcl-XL and Bax Dimerize via the Canonical

BH1–3 Groove�BH3 Helix Interface after Activation—We per-formed disulfide cross-linking experiments to map the het-erodimer interface using the active single-cysteine Bcl-XL andBax proteins synthesized in vitro. We first did the experimentswith the MOM liposomal membranes to avoid complicationsfrom the other mitochondrial proteins. We also used the BaxBH3 peptide instead of tBid to activate the soluble Bcl-XL andBax proteins and generate the membrane-bound proteins,because tBid can bind to Bcl-XL and Bax in the membranes (5,6) and hence may interfere with the Bcl-XL�Bax interaction.Like tBid, the BH3 peptide could activate the Bcl-XL and Baxproteins (data not shown) but did not bind to the membranes,as shown previously (7). As shown in Fig. 3A, a disulfide-linkedproduct with an apparent molecular mass (Mr) close to thatpredicted for the Bcl-XL�Bax heterodimer was detected in themembranes when the indicated pairs of single-cysteine Bcl-XLand Baxmutants were targeted to themembranes. Particularly,the product (indicated by an arrow in lane 2 or 10 on the non-reducing gel) was formed when a Bcl-XL protein with the cys-teine in the BH1–3 groove, V126C or L194C, was paired with aBax protein with the cysteine in or near the BH3 region, L59Cor M74C, respectively.The following four results demonstrated that the arrow-in-

dicated product was a disulfide-linked heterodimer of the sin-

FIGURE 1. Sequences of Bax and Bcl-XL mutants. A and B, single-cysteine Baxand Bcl-XL mutants. Bax (A) and Bcl-XL (B) sequences are shown with BH motifshighlighted by dashed lines above, and helices are identified by arrows below. Thenative cysteines (underlined) were changed to alanine to create the cysteine-null(C0) mutants. Single-cysteine Bax and Bcl-XL mutants were created from therespective cysteine-null mutants by individually replacing the residues in bold-face type with cysteine. Arrowheads, Met74 of Bax and Gly138 and Arg139 of Bcl-XLthat were changed to glutamate, alanine, and aspartate in Bax M74E and Bcl-XLG138A and R139D mutants, respectively. C, single-lysine Bcl-XL mutants. Bcl-XL sequence is shown with BH motifs and helices indicated as in B. The nativelysines (underlined) were changed to arginine to create the lysine-null (K0)mutant. Single-lysine Bcl-XL mutants were created from the lysine-null mutantby individually replacing the residues in boldface type with lysine.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

gle-cysteine Bcl-XL and Bax after they were activated by theBH3 peptide and targeted to the membranes. (i) The productappeared on the non-reducing gel and disappeared on thereducing gel (Fig. 3A, lanes 2 and 10) and was not formed whenthe cysteine-null (C0) Bcl-XL andBaxmutantswere used (Non-reducing; lane 4) and was greatly reduced when NEM wasadded to block the sulfhydrylmoiety before oxidation byCuPhe(Non-reducing; lane 1 versus lane 2 and lane 9 versus lane 10).Therefore, it is a disulfide-linked product formed by the single-cysteine Bcl-XL and Bax. (ii) The product was formed whenboth proteinswere present in themembranes but notwhen oneprotein was omitted (Fig. 3A,Non-reducing, lane 2 versus lanes6 and 8 and lane 10 versus lanes 12 and 14). Therefore, it is nota Bcl-XL or Bax homodimer. (iii)When one of the proteins was

labeled by [35S]Met but the other was not, the product was stilldetectable, but the radiation intensity was less than that whenboth proteins were labeled (Fig. 3B, arrow-indicated bands inlane 2 versus lanes 4 and 6 and lane 8 versus lanes 10 and 12).These data suggest that the heterodimers detected in lanes 4and 10 were formed by the radioactive Bcl-XL (visible in thephosphorimages, indicated by filled circles) and the nonradioac-tive Bax (invisible in the phosphorimages), and the heterodimersdetected in lanes 6 and 12 were formed by the radioactive Bax(visible, indicated by open circles) and the nonradioactive Bcl-XL(invisible). In contrast, the heterodimers detected in lanes 2 and 8were formed by the radioactive Bcl-XL and Bax that were bothvisible in the phosphorimages. (iv) When the BH3 peptide wasomitted, although a substantial amount of Bcl-XL was bound

FIGURE 2. Activities of Bax and Bcl-XL mutants in Bax�/�/Bak�/� mitochondria and mouse embryonic fibroblasts. A-C, MOMP activity in the Bax�/�/Bak�/� mitochondria. Wild-type (WT) or the indicated mutant Bax proteins were synthesized in vitro, activated by tBid protein, and targeted to the Bax�/�/Bak�/� mitochondria in the absence or presence of WT or the indicated mutant Bcl-XL proteins that were synthesized in vitro. Cytochrome c release from themitochondria was measured using ELISA. Data shown are average fractions of cytochrome c release from two to three independent experiments with theranges indicated by error bars. D, the apoptotic activity of WT, lysine-null, and single-lysine mutant Bax in the bax�/�/bak�/� MEFs. The MEF cells were infectedwith retrovirus that co-expressed Bax and GFP from the same message using the IRES sequence between the two coding regions and treated with etoposideof the indicated concentrations. The cells were examined for apoptosis by annexin V staining. Both panels show the fraction of adherent annexin V-positive cellsin the GFP-positive (infected and expressing Bax; black bar) and GFP-negative (uninfected and not expressing Bax; white bar) populations. The type of Baxprotein expressed by the virus that was used to infect the cells is indicated below the plot. GFP, control virus that expressed only GFP; WT, K0, K21, etc., virus thatexpressed GFP and wild-type, lysine-null, or single-lysine Bax, respectively. Data shown are the averages from eight independent replicates. For each replicate,a minimum of 356 cells/well (average � 1084, maximum � 2613) or 309 cells/well (average � 849, maximum � 1442) was analyzed for the 2 or 0.5 �M

etoposide-treated cells, respectively. Error bars, S.D.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

FIGURE 3. Disulfide cross-linking of Bax and Bcl-XL proteins with a single cysteine in BH3 helix and BH1–3 groove, respectively. A, the in vitro synthesized[35S]Met-labeled cysteine-null and single-cysteine Bax and Bcl-XL proteins were activated by the BH3 peptide and targeted to the MOM liposomes. The resultingproteoliposomes were isolated and oxidized by CuPhe for 30 min. NEM and EDTA were then added to stop the oxidation. For the 0 min controls, NEM and EDTA wereadded prior to the addition of CuPhe. The resulting samples were analyzed by non-reducing and reducing SDS-PAGE and phosphorimaging. B, the in vitro synthesizedradioactive (R) and/or non-radioactive (N) Bax and Bcl-XL mutants were activated by the BH3 peptide and targeted to the MOM liposomes. The resulting proteolipo-somes were processed and analyzed as described in A. C, the in vitro synthesized radioactive Bax and Bcl-XL mutant proteins were incubated in the absence or presenceof the MOM liposomes and/or the BH3 peptide. The resulting proteoliposomes were processed and analyzed as described in A. D and E, in vitro synthesized radioactiveBax and Bcl-XL mutants were activated by tBid and targeted to the MOM liposomes (D) or activated by the BH3 peptide and targeted to the Bax�/�/Bak�/�

mitochondria (E). The resulting membranes were isolated, oxidized, and analyzed as described in A. F, the in vitro synthesized radioactive (R) and/or non-radioactive (N)Bax and Bcl-XL mutant proteins were activated by the BH3 peptide and targeted to the Bax�/�/Bak�/� mitochondria. The resulting mitochondria were isolated,oxidized, and analyzed as described in A. In all panels, protein standards are indicated on the side of phosphorimages with their molecular masses (Mr). Arrows,disulfide-linked Bax�Bcl-XL heterodimers; open or filled circles, Bax or Bcl-XL monomers, respectively.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

to the membranes, the amount of Bax bound to the mem-branes was negligible, and the disulfide-linked heterodimerwas undetectable in the membranes (Fig. 3C, lanes 6 and 14).In the absence of the membranes, virtually no Bcl-XL or Baxmonomer was recovered from the membrane fraction in thesucrose gradient; nor was the disulfide-linked heterodimer(lanes 4 and 12). Therefore, the heterodimer detected in Fig.3A was formed by a disulfide linkage of the indicated cys-teines in the Bcl-XL and Bax proteins that were activated bythe BH3 peptide and targeted to the membranes.To verify whether the heterodimer interface detected in the

liposomes after activation by the BH3 peptide is also formed ina more biological setting, we used tBid to activate the Bcl-XLand Bax proteins. Similar disulfide-linked heterodimers wereformed in this system (Fig. 3D). Further, we targeted both pro-teins that were activated by the BH3 peptide to the Bax�/�/Bak�/� mitochondria and detected similar disulfide-linkedheterodimers (Fig. 3,E and F). Therefore, the groove�BH3 inter-face is formed not only in the liposomal membranes but also inthe mitochondrial membranes after activation by either theBH3 peptide or the BH3 protein.We also conducted photocross-linking experiments using

Bcl-XL and Bax mutants with a single photoreactive lysineanalog located at specific positions. After activation by theBH3 peptide and targeting to the MOM liposomes, the invitro synthesized [35S]Met-labeled Bcl-XL or Bax proteinwith the photoreactive probe, ANB, attached to the �-aminogroup of the lysine located in or near the BH1–3 groove orthe BH3 region cross-linked with the purified His6-taggedBax or Bcl-XL protein, respectively (Fig. 4, A and B). Thesedata are consistent with those from disulfide cross-linkingexperiments and suggest the existence of the groove�BH3interface in the Bcl-XL�Bax heterodimer that is formed in themembranes.The above cross-linking data support a structural model in

which the BH1–3 groove�BH3 helix interface, as seen previ-ously by crystallography (17), exists in the Bcl-XL�Bax het-erodimer formed in theMOM after both proteins are activatedby the BH3 protein or peptide (Fig. 5). Clearly shown in themodel, Val126 and Leu194 in the BH1–3 groove of Bcl-XL are inclose proximity with Leu59 and Met74 in the BH3 helix of Bax,respectively (Fig. 5A). These hydrophobic residues interact tocontribute a significant portion of the binding energy that holdsthe complex together. For example, Leu59 andMet74 in the BH3helix of Bax bind to the first and the fifth hydrophobic pocket inthe BH1–3 groove of Bcl-XL, respectively (17). Val126 in thegroove of Bcl-XL has a van derWaals interaction with not onlyLeu59 but also Leu63, which are the first and the second con-served hydrophobic residues in the BH3 region of Bax, respec-tively. The side chain of Leu194 in the groove of Bcl-XL also hasfavorable van derWaals interactions with the aliphatic portionof the side chain of Arg78 in the BH3 helix.When these Bcl-XL/Bax residue pairs are replaced by the cysteine pairs (V126C/L59C and L194C/M74C), the model predicts that the S1�-S2�

distance between the paired cysteines and the C1�-S1�-S2�-C2

�

dihedral angle are close to the range for disulfide bond forma-tion (Fig. 5B) (33). This prediction is supported by the disulfidecross-linking data (Fig. 3). The model also predicts that the

C1�-C2

� distance between Cys62, a native cysteine located inthe BH3 helix of Bax, and the Val126 located in the groove of theBcl-XLmutant is 9.8 Å (Fig. 5A), which is too great a distance toresult in a disulfide bond when the Val126 was changed to cys-teine (V126C). Similar predictions were also made for the C1

�-C2

� distances between the other twomismatched cysteine pairsin the Bax�Bcl-XL complex: L59C/L194C and M74C/V126C(data not shown). These predictions were consistent withthe observation that no disulfide-linked dimer was detectedbetween the corresponding mismatched Bax/Bcl-XL mutantpairs (data not shown). It was also clear that the residues in theBH3 helix of Bax and the groove of Bcl-XL, which generatedphotocross-linked heterodimers when replaced by ANB-lysine(Fig. 4), were located in or near the BH3�groove interface in themodel (Fig. 5C).The Helix 1 Regions of Bcl-XL and Bax Interact with Each

Other in a Parallel Manner, Forming a Novel Interface in theHeterodimer—To determine if regions of Bcl-XL and Baxother than the BH1–3 regions are also involved in the het-erodimerization, we used photocross-linking to scan moreBcl-XL and Baxmutants with a single ANB-lysine positionedin the other regions. The photoreactive probes located inhelix 1 of both Bcl-XL and Bax generated the heterodimer-specific photoadducts in liposomes (Fig. 6), thereby indicatinginvolvement of helix 1 in the heterodimerization. To determineif the helix 1 regions from both proteins interact with eachother, we did disulfide cross-linking experiments with the fol-lowing single-cysteine Bcl-XL/Bax mutant pairs: E7C/M20C,S18C/F30C, and S23C/R34C. After these mutants were acti-vated by the BH3 peptide and targeted to the MOM liposomes(Fig. 7A) or the Bax�/�/Bak�/� mitochondria (Fig. 7B), multi-ple disulfide-linked products were detected, particularly in themitochondria. Among these products, the ones marked witharrows are the Bcl-XL�Bax heterodimers because they weregenerated only when both proteins were targeted to the mem-branes and because theirmolecularmasses were close to that ofthe heterodimer (Fig. 7, A and B, lanes 6, 12, and 18). Asexpected, the heterodimer-specific bands were still detectablewhen one of the two proteins was not labeled with [35S]Met,although the intensities of the bands were weaker than thosewhen both proteins were labeled in most cases (Fig. 7C, arrow-indicated bands). When the BH3 peptide was omitted,although a substantial amount of Bcl-XLwas bound to the lipo-somalmembranes, the amount of Bax bound to themembraneswas negligible, and the disulfide-linked heterodimers wereundetectable in themembranes (Fig. 7D, lanes 6, 14, and 22). Inthe absence of themembranes, virtually no Bcl-XL or Baxmon-omerwas recovered from themembrane fraction in the sucrosegradient; nor was the disulfide-linked heterodimer (Fig. 7D,lanes 4, 12, and 20). Therefore, the disulfide-linked het-erodimers detected in Fig. 7, A and B, were formed by theBcl-XL and Bax mutants with the indicated cysteines in theirhelix 1 after they were activated by the BH3 peptide and tar-geted to the membranes.When some single-cysteine Bcl-XL and Bax mutants were

individually targeted to the membranes, disulfide-linked prod-ucts were generated. Each of the products has amolecularmassclose to that of the corresponding homodimer (Fig. 7, A and B,



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

indicated by filled or open triangles, respectively), suggestingthat these products were disulfide-linked homodimers ofBcl-XL or Bax.Additional disulfide-linked products were detected when

certain single-cysteine Bcl-XL or Bax mutant was targeted tothe mitochondria (Fig. 7, B and C, indicated by stars andsquares). These products were probably heterodimers of therespective single-cysteine mutant and mitochondrial proteins,because treatment of the mitochondria with NEM before tar-geting the single-cysteine mutant to the mitochondria and oxi-dation by CuPhe blocked the formation of these products (Fig.7B, lane 20 versus lane 18; data not shown).Using the disulfide cross-linking data from the single-cys-

teine Bcl-XL/Bax mutant pairs, E7C/M20C, S18C/F30C, andS23C/R34C, we built a molecular model for the helix 1�helix 1interface in the Bcl-XL�Bax complex (Fig. 8A). We used the

FlexPepDock program (30) to dock the helix 1 peptide from theNMR structure of Bax monomer (PDB entry 1F16) againstthe helix 1 in Bcl-XL protein from the crystal structure of theBcl-XL protein�Bax BH3 peptide complex (PDB entry 3PL7).The two helix 1 regions, each with the three cysteine substitu-tions, were aligned in parallel and positioned such that the C1

�-C2

� distance and the C1�-S1�-S2�-C2

� dihedral angle in eachcysteine pair were close to 4 Å and �90°, respectively, thusfavoring the disulfide formation observed in the cross-linkingexperiments.Weapplieddistancerestraints in theFlexPepDockpro-gram between the C� atoms of the cysteine pairs, but the sidechain conformations were not constrained to any target values.The FlexPepDock program generated 200 output models after10 cycles of optimizing the rigid body position and the back-bone and side chain conformations of the Bax helix 1 peptideand the side chain conformations of the Bcl-XL protein. The

FIGURE 4. Photocross-linking of Bcl-XL or Bax mutant proteins with a single photoreactive ANB-lysine incorporated into the BH1–3 groove or BH3helix, respectively. The in vitro synthesized [35S]Met-labeled Bcl-XL (A) or Bax (B) mutant proteins without lysine (K0) and with a single photoreactive ANBprobe incorporated at the indicated lysine positions (F97K, M79K, etc.) were mixed with the purified 6H-Bax (A) or 6H-Bcl-XL (B) protein. The proteins wereactivated by the BH3 peptide and targeted to the MOM liposomes. The resulting proteoliposomes were isolated and photolyzed. The resulting photoadductsof the 6H-tagged and the 35S-labeled proteins were enriched and analyzed by reducing SDS-PAGE and phosphorimaging. Protein standards are indicated onthe side of phosphorimages with Mr. Filled or open circle, [35S]Bcl-XL or [35S]Bax monomer, respectively. Arrow, photoadduct of [35S]Bcl-XL and 6H-Bax (A) or[35S]Bax and 6H-Bcl-XL (B).



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

output models were ranked by their Rosetta energy scores (notshown). One of the top 10models is shown in Fig. 8A (left). Thedistribution of the scores was broad when plotted against theroot mean square deviation with the starting model. In otherwords, the bottom of the docking funnel was flat. The top 10models differed in their scores by only 1–2Rosetta energy units.In the selected model, the S1�-S2� distances and the C1

�-S1�-S2�-C2

� dihedral angles in the three cysteine pairs are near theranges that allow disulfide formation (33).We then replaced the three cysteine pairs in the structure

shown in the left panel of Fig. 8A with the corresponding wild-type residues and used the resulting structure as a startingmodel in the FlexPepDock program to generate another 200output models for the wild-type helix 1�helix 1 heterodimerinterface. The right panel of Fig. 8A depicts one of the top 10models, which predicts how Bcl-XL and Bax interact via theirhelix 1 when the two helices are parallel to each other. The

model also indicates how the two helices form a previouslyunreported dimer interface that has the three wild-type Bcl-XL-Bax residue pairs, Glu7-Met20, Ser18-Phe30, and Ser23-Arg34, in proximity to each other. The model predicted somestabilizing interactions, including a hydrogen bondbetween theside chainN� of BaxArg34 and the backbone carbonyl oxygen ofBcl-XLGln19, and complete or partial burial of the hydrophobicside chains of Bax Leu26, Leu27, and Phe30 and Ile31. Thesefavorable interactions are countered by the burial of the nega-tively charged side chain of Bcl-XL Glu7. This model was alsoconsistent with the photocross-linking data because the resi-dues that could form photoadducts, when replaced with ANB-lysine (Fig. 6), were also in or near the dimer interface (Fig. 8B).TheHelix 1 Regions of Bcl-XL andBaxCanAlso Interact in an

Antiparallel Fashion to Form an Alternative Interface in theHeterodimer—To determine whether the helix 1�helix 1 het-erodimer interface can form only in the parallel orientation, we

FIGURE 5. A model for the BH1–3 groove�BH3 helix interface in the membrane-bound Bcl-XL�Bax heterodimer. A, the crystal structure of Bcl-XLprotein�Bax BH3 peptide complex (PDB entry 3PL7) is shown with Val126/Leu59 and Leu194/Met74, the residue pairs that formed disulfide bonds when they werereplaced with cysteine pairs (Fig. 3), presented in stick form, and their �-carbon atoms (C1

� and C2�) linked by dashed lines with the distances in Å indicated.

Cys62, a native cysteine in the BH3 region of Bax, is also presented in stick form with its C� linked to the C� of Bcl-XL Val126 and the distance in Å indicated. B, thetwo Bcl-XL/Bax residue pairs, Val126/Leu59 and Leu194/Met74, shown in A are replaced by two cysteine pairs, which are presented in stick form with their �-sulfuratoms (S1

� and S2�) linked by virtual bonds and the distances in Å indicated. The C1

�-S1�-S2

�-C2� dihedral angles about the V126C/L59C and L194C/M74C

disulfide bonds are �52.2 and �22.0°, respectively. C, the crystal structure of the Bcl-XL protein�Bax BH3 peptide complex is shown with the residues thatgenerated the heterodimer-specific photoadducts when replaced by ANB-lysine (Fig. 4) presented in stick form. In all panels, the BH1, BH2, BH3, and BH4regions of Bcl-XL protein are colored blue, cyan, red, and orange, respectively, and the BH3 region of the Bax peptide is colored green.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

did disulfide cross-linking experiments using mismatched sin-gle-cysteine mutant pairs that were activated by the BH3 pep-tide and targeted to the MOM liposomes. Surprisingly, Bcl-XLE7C or S23C mutant with the cysteine located near the N or Cterminus of helix 1 formed a disulfide-linked dimer with BaxR34C or M20C with the cysteine located near the C or N ter-minus of helix 1, respectively (Fig. 9A). These disulfide cross-linking data support an “antiparallel dimermodel” in which thetwohelix 1 regions fromBcl-XL andBax are aligned antiparallelto each other (Fig. 9B). The model predicted stabilizing contri-butions from complete or partial burial of the hydrophobic sidechains of Bcl-XL Val10 and Trp24 and Bax Ile19, Leu27, Phe30,and Ile31. The Rosetta energy score (not shown) for the antiparal-lelmodel (Fig. 9B) was very close to that of the parallelmodel (Fig.8A), suggesting that thehelix1�helix1dimer interfaceequally sam-pled the twoarrangementsof theBaxpeptide. Further, theRosettaenergy scores calculated from the helix 1�helix 1 interface modelswere similar to that from the BH1–3 groove�BH3 helix interfacemodel (Fig. 5A; data not shown), implying similar contributionsfrom these interactions to the overall stability of the Bcl-XL�Baxcomplex. These Rosetta energy scores were also similar to thosereported previously for the groove�BH3 interfaces formed byBcl-XL and BH3 peptides from various BH3-only proteins (37),suggesting similar affinities for the binding of Bcl-XL to the Baxpeptides and the BH3 peptides. On the other hand, the flexiblebinding interaction at thehelix 1�helix 1 interface (Figs. 8A and9B)contrasted with the rigid binding interaction at the groove�BH3helix interface (Fig. 5A). We thus investigated their relationshipand contribution to the function of the Bcl-XL�Bax complex withthemutational analyses described below.The Helix 1�Helix 1 Interaction Is Largely Dependent on the

BH1–3 Groove�BH3 Helix Interaction, and Both InteractionsAre Important for Bcl-XL to Inhibit Bax-mediatedMOMP—Todetermine whether the novel helix 1�helix 1 interface can form

independently of the canonical BH1–3 groove�BH3 helix inter-face, we tested several mutations that would disrupt thegroove�BH3 interface and monitored their effect on both inter-faces. According to the groove�BH3 interface model (Fig. 5A),Gly138 in the groove of Bcl-XL has a favorable van der Waalsinteraction with Asp71 in the BH3 helix of Bax. Replacement ofthe Gly138 with alanine (G138A) would not only eliminate thevan der Waals interaction but also generate a severe steric col-lision between the �-carbon of Ala138 and the backbone car-bonyl oxygen of Gly67 in the BH3 helix of Bax (Fig. 10A, left).The next residue in the groove of Bcl-XL, Arg139, uses its sidechain to form two hydrogen bonds as part of a strong salt bridgewith the side chain of Asp68 in the BH3 helix of Bax. Replace-ment of the Arg139 with aspartate (R139D) would introduce astrong electrostatic repulsion with the Asp68 of Bax and elimi-nate the strong salt bridge (Fig. 10A, left). On the Bax side, theside chain of Met74 in the BH3 helix projects into the fifthhydrophobic pocket in the groove of Bcl-XL and forms favor-able van derWaals interactions with the side chain of Tyr195 inBcl-XL. Replacement of the Met74 with glutamate (M74E)would eliminate the hydrophobic interaction and generate acollision with the side chain of Tyr195 (Fig. 10A, right). As pre-dicted by the model, each mutation abolished the disulfidecross-linking of Bcl-XL with Bax in mitochondria via theV126C/L59C pair in the groove�BH3 interface (Fig. 10B, lane 2versus lanes 4 and 6).We then monitored the effects of the mutations in the

groove�BH3 interface on the helix 1�helix 1 interface formation.Each of the mutations inhibited the disulfide cross-linking ofBcl-XL and Bax via the S23C/R34C pair in the parallel helix1�helix 1 interface to a different extent (Fig. 10C). Thus, theM74E mutation in Bax BH3 helix completely abolished thecross-linking (lane 6 versus lane 8), whereas the R139D muta-tion in the Bcl-XL groove inhibited the cross-linkingmore thantheG138Amutation (lane 12 versus lanes 14 and 16). The threemutations also inhibited the heterodisulfide cross-linking viathe E7C/R34C and S23C/M20C pairs in the antiparallel helix1�helix 1 interface to similar extents as those via the cysteinepair in the parallel interface (Fig. 10D, lane 4 versus lane 6 andlane 12 versus lanes 14 and 16). Therefore, the formation of thehelix 1�helix 1 interface, no matter whether in the parallel orthe antiparallel conformation, is at least partially dependent onthe formation of the groove�BH3 interface.We did theMOMP assay to test whether the physical disrup-

tion of the interfaces by the mutations impacted the inhibitionof Bax-mediated MOMP by Bcl-XL. As shown in Fig. 10E,Bcl-XL with V126C in the BH1–3 groove greatly inhibited theMOMP activity of Bax with L59C in the BH3 helix, whereasBcl-XLwith S23C or E7C in helix 1 greatly inhibited the activityof Bax with R34C in helix 1 (columns 1–3 and 9–12). However,the single-cysteine Bcl-XLmutants barely inhibited theMOMPactivity of the corresponding single-cysteine Bax mutants thatalso have theM74Emutation (columns 4–6 and 13–16). Alongwith the great reduction in Bcl-XL activity against the Bax-mediated MOMP, the M74E mutation in Bax also greatlyimpaired the formation of both the groove�BH3 helix interfaceand the helix 1�helix 1 interface, either parallel or antiparallel, asshown by the disulfide cross-linking data (Fig. 10, B (lane 2

FIGURE 6. Photocross-linking of Bax and Bcl-XL mutant proteins with singlephotoreactive ANB-lysine incorporated into their helix 1. The [35S]Met-la-beled Bcl-XL (A) or Bax (B) mutant proteins with single photo-reactive ANB probeincorporated at the indicated lysine positions were synthesized in vitro and acti-vated together with the purified 6H-Bax (A) or 6H-Bcl-XL (B) protein, respectively,by the BH3 peptide. The proteins were targeted to the MOM liposomes, pro-cessed, and analyzed as described in the legend to Fig. 4.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

versus lane 4),C (lane 6 versus lane 8), andD (lane 4 versus lane6)). Similarly, when the R139D mutation was introduced intoBcl-XL V126C and S23C, the resulting Bcl-XL mutants par-tially lost the inhibitory activity on Bax L59C- and R34C-medi-ated orM20C-mediatedMOMP, respectively (Fig. 10E, column

3 versus column 7, column 11 versus column 17, or column 21versus column 22). Consistent with the partial loss of Bcl-XLactivity against the Bax-mediatedMOMP, the R139Dmutationin Bcl-XL blocked the helix 1�helix 1 interaction to a lesserextent compared with the M74E mutation in Bax (Fig. 10, C

FIGURE 7. Disulfide cross-linking of Bax and Bcl-XL proteins with a single cysteine in their helix 1. A and B, the in vitro synthesized [35S]Met-labeledsingle-cysteine Bcl-XL and Bax proteins were activated by the BH3 peptide and targeted to the MOM liposomes (A) or the Bax�/�/Bak�/� mitochondria (B). Theresulting membrane-bound proteins were processed and analyzed as described in the legend to Fig. 3. C, the in vitro synthesized radioactive (R) and/ornon-radioactive (N) single-cysteine Bcl-XL and Bax proteins were activated by the BH3 peptide and targeted to the Bax�/�/Bak�/� mitochondria. The resultingmitochondria-bound proteins were processed and analyzed as in B. D, the in vitro synthesized radioactive single-cysteine Bcl-XL and Bax proteins wereincubated in the absence or presence of the MOM liposomes and/or the BH3 peptide. The resulting proteoliposomes were processed and analyzed as in A. Inall panels, the labels are as described in the legend to Fig. 1. In addition, filled or open triangles indicate the disulfide-linked Bcl-XL or Bax homodimers, and starsor squares indicate the disulfide-linked Bcl-XL or Bax�mitochondrial protein complexes, respectively. NEM-Mito in B indicates that the mitochondria used in thereaction were pretreated with NEM to block the sulfhydryl moieties in the mitochondrial proteins, preventing their cross-linking with the sulfhydryl moietiesin the Bcl-XL and Bax proteins.

FIGURE 8. A model for the parallel helix 1�helix 1 dimer interface in the membrane-bound Bcl-XL�Bax heterodimer. A, left, to build the initial model, thehelix 1 extracted from the Bax monomer structure (PDB entry 1F16) was manually positioned onto the helix 1 of the Bcl-XL protein in the Bcl-XL protein�Bax BH3peptide complex structure (PDB entry 3PL7), such that the three cysteine pairs, E7C/M20C, S18C/F30C, and S23C/R34C, which resulted in disulfide-linkedBcl-XL�Bax heterodimers in Fig. 7, are in a geometry suitable for disulfide linkage. The resulting complex structure was input into FlexPepDock program, andone of the top 10 output models is shown with the cysteine pairs presented in stick form and their �-sulfur atoms (S1

� and S2�) linked by virtual bonds and the

distances in Å indicated. The C1�-S1

�-S2�-C2

� dihedral angles about the disulfide bond for the three cysteine pairs are �99.2°, �114.5°, and �178.6°, respec-tively. Right, to generate the final model, the cysteines in the left panel were changed back to the corresponding wild-type residues. The resulting complexstructure was the starting model in an automated peptide docking experiment with the FlexPepDock program. One of the top 10 output models from thedocking experiment is shown with the respective wild-type residue pairs presented in stick form and their �-carbon atoms (C1

� and C2�) linked by dashed lines

with the distances in Å indicated. B, a model for the parallel helix 1�helix 1 interface in the Bcl-XL�Bax heterodimer was built based on the model presented inthe right panel of A and the photocross-linking data shown in Fig. 6. The residues that generated the heterodimer-specific photoadducts when replaced byANB-lysine are presented in stick form. In all models, the BH1– 4 regions of Bcl-XL protein are colored as in Fig. 5, and the helix 1 of Bax is colored magenta. Forsimplicity, the BH3 helix of Bax was omitted from the models.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

(lane 6 versus lane 8 and lane 12 versus lane 16) and D (lane 4versus lane 6 and lane 12 versus lane 16)), although the twomutations blocked the groove�BH3 helix interaction equally(Fig. 10B, lane 2 versus lanes 4 and 8). Interestingly, the effect ofthe G138A mutation in Bcl-XL on its inhibition of Bax-medi-atedMOMP depended on the cysteine location in the mutants.Whereas the G138A mutation caused a partial loss in theBcl-XL V126C inhibition of Bax L59C (Fig. 10E, column 3 ver-sus column 8), it did not affect the Bcl-XL S23C inhibition ofBax R34C or M20C (Fig. 10E, column 11 versus column 18 orcolumn21versus column23).Thedifferential effects of theG138Amutationontheanti-BaxactivityofBcl-XL in theMOMPcouldbeexplainedby thedifferential effects of themutationon thephysicalinteraction of Bcl-XLwith Bax. Thus, as indicated by the disulfidecross-linking data, the G138A mutation largely disrupted the

groove�BH3 helix interaction between Bcl-XL V126C and BaxL59C (Fig. 10B, lane 2 versus lane 6) but only slightly inhibitedthe helix 1�helix 1 interaction between Bcl-XL S23C and BaxR34C or M20C (Fig. 10, C (lane 12 versus lane 14) and D (lane12 versus lane 14)). Therefore, the MOMP activity data fromthese Bcl-XL and Bax mutants are directly correlated with thephysical interaction data, supporting a conclusion that thephysical interactions in both interfaces are important forBcl-XL to inhibit Bax-mediated MOMP.The BH1–3 Groove of Bcl-XL Directly Competes with the

BH1–3 Groove of Bax for Binding to the BH3 Helix of Bax toInhibit BaxHomo-oligomerization in theMOM—The effects ofthemutations in the BH1–3 groove of Bcl-XL on its interactionwith Bax and inhibition of Bax-mediatedMOMP suggest a crit-ical role for the groove in the function of Bcl-XL. Conceivably,

FIGURE 9. Disulfide cross-linking of Bax and Bcl-XL proteins with a single cysteine in their helix 1 that supports the antiparallel helix 1�helix 1 dimermodel. A, the in vitro synthesized [35S]Met-labeled single-cysteine Bcl-XL and Bax proteins were activated by the BH3 peptide and targeted to the MOMliposomes. The resulting membrane-bound proteins were processed and analyzed as described in Fig. 7. B, the antiparallel helix 1�helix 1 dimer model wasbased on the disulfide cross-linking data from two single-cysteine Bcl-XL/Bax mutant pairs, E7C/R34C and S23C/M20C, shown in A. It was generated by theFlexPepDock program and presented similarly as described in the legend to Fig. 8A. Left, the two cysteine pairs are presented in stick form with their �-sulfuratoms (S1

� and S2�) linked by virtual bonds and the distances in Å indicated. The C1

�-S1�-S2

�-C2� dihedral angles about the disulfide bond for the two cysteine

pairs are �146.7° and �74.0°, respectively. Right, the two respective wild-type residue pairs are presented in stick form with their �-carbon atoms (C1� and C2

�)linked by dashed lines and the distances in Å indicated. The BH1– 4 regions of Bcl-XL protein and the helix 1 of Bax are colored as in Fig. 8. For simplicity, the BH3helix of Bax was omitted from the models.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

FIGURE 10. Effects of interfacial mutations on Bcl-XL�Bax heterodimer formation and function. A, a part of the crystal structure of Bcl-XL protein�Bax BH3 peptidecomplex (PDB entry 3PL7) is shown in the left panel with Bcl-XL Gly138 changed to alanine (G138A) and Arg139 changed to aspartate (R139D) and in the right panel withBax Met74 changed to glutamate (M74E). The mutated residues are shown in stick form. The residues in the respective binding partners that interact with these mutatedresidues are also illustrated in stick form with dashed lines linking the C� of Bcl-XL Ala138 to the backbone carbonyl O of Bax Gly67, O�1 of Bcl-XL Asp139 to O�1 of Bax Asp68, andtheC�ofBaxGlu74 toC�2 ofBcl-XLTyr195.ThedistancesbetweentheseatomsareindicatedinÅ.B–D, the in vitro synthesized[35S]Met-labeledsingle-cysteineBcl-XLandBaxproteins with or without the indicated interfacial mutations were activated by the BH3 peptide and targeted to the Bax�/�/Bak�/� mitochondria that were eitheruntreated (B) or pretreated with NEM (C and D). The mitochondria-bound proteins were processed and analyzed as described in the legend to Fig. 3. E, the indicatedBax proteins were synthesized in vitro, and their BH3 peptide-dependent cytochrome c release activities in the Bax�/�/Bak�/� mitochondria were assayed in theabsence or presence of the indicated in vitro synthesized Bcl-XL proteins as described in the legend to Fig. 2. Data shown are average fractions of cytochrome c releasefrom 2–4 independent experiments with the ranges indicated by error bars.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

the BH1–3 groove in Bcl-XL would compete with a similargroove in Bax that has been suggested to bind the BH3 helix ofBax nucleating Bax oligomerization in the MOM (13). To testthis scenario directly, wemonitored the effect of Bcl-XL on Baxhomodisulfide cross-linking via T56C in the BH3 helix andR94C in the BH1–3 groove that was previously used to capturea Bax homodimer formed via the BH3 helix�groove interface(12, 13). The single-cysteine Bax mutants formed a disulfide-linked homodimer after theywere activated by the BH3 peptideand targeted to the MOM liposomes (Fig. 11A, lane 2, opentriangle), consistent with the previous finding from the mito-chondria isolated from the apoptotic cells that expressed thesame Bax mutants (12). The addition of purified recombinantBcl-XL protein to the reaction inhibited the disulfide cross-linking of the Bax mutants (Fig. 11A, lane 4 versus lane 2). Todemonstrate that the inhibition is directly caused by the com-petition between the BH1–3 grooves of Bcl-XL and Bax for thesame BH3 helix of Bax, we added Bcl-XL E129C mutant with asingle cysteine in the BH1–3 groove to the Bax T56C with asingle cysteine in the BH3 helix. As predicted by thegroove�BH3 helix interface model for Bcl-XL�Bax heterodimer(Fig. 5), Bcl-XL E129C formed a disulfide-linked heterodimerwithBaxT56Cafter theywere activated by theBH3peptide andtargeted to the MOM liposomes (Fig. 11B, lane 2, arrow), sug-gesting that the similar BH1–3 grooves in Bcl-XL and Bax caninteract with the same residue in the BH3 helix of Bax, resultingin a direct competition.As expected, theR139Dmutation in thegroove of Bcl-XL E129C largely inhibited the heterodisulfidelinkage with Bax T56C (Fig. 11B, lane 6 versus lane 2). TheR139D mutation also partially abolished the inhibition of BaxT56C-mediated MOMP by Bcl-XL E129C (Fig. 11D, columns1–3 versus column 5). Altogether, these experiments demon-strated that the inhibition of Bax MOMP activity by Bcl-XL isdirectly mediated by the competition between the Baxhomointeraction and the heterointeraction with Bcl-XL in theMOM via the similar BH3 helix�BH1–3 groove interfaces.ABT-737 Dissociates the Membrane-bound Bcl-XL�Bax Het-

erodimer and Restores Bax Homo-oligomerization and MOMPActivity—ABT-737, a BH3 mimic compound, binds to theBH1–3 groove of Bcl-XL and inhibits its interaction with BH3-only proteins (22, 38–40). Because Bcl-XL binds to Bax notonly through the BH1–3 groove�BH3 interface but also thehelix 1�helix 1 interface, we tried to determine whether ABT-737 could fully inhibit the Bcl-XL interaction with Bax inmem-branes. We first tested the effect of ABT-737 on the formationof the two interfaces.When ABT-737 was included in the reac-tion with Bcl-XL E129C, Bax T56C, the BH3 peptide, and theMOM liposomes, it dramatically blocked the disulfide cross-linking of the two proteins in the membranes (Fig. 11B, lane 2versus lane 4), suggesting that ABT-737 outcompetes the BH3helix of Bax for interaction with the groove of Bcl-XL. As pre-dicted from the dependence of helix 1�helix 1 interface forma-tion on groove�BH3 interface formation, the compound alsopartially inhibited the disulfide cross-linking of Bcl-XL S23Cand Bax R34C with the cysteines located in the helix 1�helix 1interface (Fig. 11C, lane 2 versus lane 4). Moreover, the com-pound almost fully reversed the inhibition of the homodisulfidecross-linking betweenBaxT56C andBaxR94Cby recombinant

Bcl-XL (Fig. 11A, lanes 4 versus 6), indicating that ABT-737 hasfreed the BH3 helix of Bax from the groove of Bcl-XL, which inturn binds to the BH1–3 groove of other Bax. Consistent withthe ABT-induced reformation of the Bax homocomplex in thepresence of Bcl-XL, the addition ofABT-737 almost completelyreversed the Bcl-XL inhibition on the MOMP by BH3 peptide-activated Bax (Fig. 11D, column 3 versus column 4 and column 8versus column 9).Heterodimerization of Bcl-XL and Bax Does Not Alter Their

Membrane Topology—To determine the topology of the Bcl-XL�Bax heterodimer bound to the liposomal membrane afteractivation by the BH3 peptide, we used the compartment-spe-cific IASD labeling. We first did the experiments using Bcl-XLand Baxmutants with a single cysteine located in the interfacialregions. As shown in Fig. 12A, in the presence of excess purifiedrecombinant Bax protein, the in vitro synthesized Bcl-XL S18Cor S23C with the cysteine in helix 1 or V126C with the cysteinein the BH1–3 groove was labeled by IASD without using thedetergent CHAPS to solubilize the membrane or the denatur-ant urea to destabilize the proteins or their complexes. Theaddition of CHAPS, urea, or both did not increase the labelingsignificantly. Likewise, in the presence of excess purifiedrecombinant Bcl-XL protein, the CHAPS- and urea-indepen-dent IASD labeling was observed for the in vitro synthesizedBax F30C or R34C with the cysteine in helix 1 or L59C with thecysteine in the BH3 helix (Fig. 11B). Becausemost of the in vitrosynthesized Bcl-XL or Bax should be in complex with the puri-fied Bax or Bcl-XL in the presence of excess purified Bax orBcl-XL, respectively, the IASD labeling data suggest thatthese interfacial residues in the Bcl-XL�Bax heterodimer arelocated outside of the membrane and are accessible from theaqueous milieu. In contrast, the IASD labeling of Bcl-XLL194C with the cysteine in the groove near the C terminus ofhelix 8 (Fig. 12A) or Bax M74C with the cysteine near the Cterminus of the BH3 helix (Fig. 12B) was increased by theaddition of CHAPS, urea, or both, indicating that these res-idues are buried at least partially in the membrane, in theprotein, or in the protein complex.We next determined the effect of Bcl-XL on the location of

helices 5, 6, and 9 of Bax. We previously found that these threehelices are embedded in the MOM in the activated mitochon-drial Bax (8). As shown in Fig. 12C (top), in the presence ofexcess purified Bcl-XL, most of the in vitro synthesized, BH3peptide-activated, and MOM liposome-bound Bax V121C,T140C, and S184C with the cysteine located in helices 5, 6, and9, respectively,werenot labeledby IASDunlessCHAPSwasaddedto the reaction either by itself or together with urea. These resultssuggest that these three residues, and these three helices by infer-ence, were buried in themembrane at least partially. Similar IASDlabelingprofileswereobserved for the threeBaxmutantswhen theBcl-XL was absent (Fig. 12C, top), suggesting that the activatedBax, which was not in complex with Bcl-XL, had a membranetopology similar to that of the inactivated Bax, which was in com-plex with Bcl-XL. The fractions of IASD-labeled Bax in theabsence or presence of Bcl-XL with or without CHAPS from 2–5independent experiments were quantified and shown in Fig. 12C(bottom). These quantitative data indicate that Bcl-XL binding to



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Bax did not significantly alter the multispanning topology of theactivated Bax in themembrane.We also determined the effect of Bax on the location of hel-

ices 5, 6, and 9 of Bcl-XL, the three putative membrane-span-ning helices, based on the structural homology with Bax andwith the bacterial pore-forming toxins (41, 42). In the presenceand absence of excess purified Bax, most of the Bcl-XL T216CandV224Cwith the cysteine in helix 9were not labeled by IASDin the absence of CHAPS and urea, and the addition of eitherCHAPS or urea increased the labeling significantly (Fig. 12D,top). These results suggest that these residues, and helix 9 byinference, were buried partially in the membrane and partiallyin the protein or the protein complex. In contrast, most of theBcl-XL E153C and S164C with the cysteine in helices 5 and 6,respectively, were labeled by IASD in the absence of CHAPSand urea, and the addition of CHAPS, urea, or both did not alterthe labeling significantly. These results suggest that these tworesidues, and helices 5 and 6 by inference, were mostly exposedto an aqueous milieu. The average fractions of IASD-labeledBcl-XL in the absence or presence of Bax with or withoutCHAPS from three to four independent experiments (Fig. 12D,bottom) indicated that Bax binding to Bcl-XL did not changethe tail-anchored topology of Bcl-XL in the membrane. Basedon all of the IASD labeling data, we concluded that (i) the pres-ence of the other protein did not significantly alter the topologyof Bax andBcl-XL in theMOMliposomalmembrane; (ii) Bax inthe heterodimer was probably in the multispanning conforma-tion, whereas Bcl-XL was probably in the tail-anchored confor-mation; and (iii) the two interfaces in the heterodimer werelargely located outside of the membrane.In contrast to Bcl-XL, two bands were detected in the IEF gel

for the in vitro synthesized Bax proteins, even in the absence ofIASD labeling (Fig. 12, B and C, lane 1 in all panels). The twobands were also observed with the cysteine-null (C0) Baxmutant in the absence of IASD labeling (Fig. 12E). When apositively or negatively charged residue in Bax C0 was changedto an uncharged cysteine, the bands shifted accordingly in theIEF gel relative to the bands of Bax C0 (Fig. 12E). We suspectedthat a post-translational modification specific to the wheatgerm-based in vitro translation system altered the isoelectricpoint of the Bax proteins, causing the differential migrations inthe IEF gel, because the Bax C0 protein synthesized in thereticulocyte lysate-based systemmigrated as a single band (Fig.12E). Nevertheless, the modified and unmodified Bax proteinswere equally accessible to IASD labeling undermost conditions(Fig. 12, B and C).The Overall Structural Organization of Bcl-XL�Bax Het-

erodimer in theMembrane—Because bothBcl-XL andBax haveone or more helices embedded in the membrane, it is conceiv-able that these helices may interact to form an additional het-erodimer interface inside the membrane. We used various

cross-linking methods to test this possibility. The photocross-linking data from photoreactive probes located in these mem-brane-embedded helices indicate the proximity of these helicesto the other protein in the heterodimer (Fig. 13). However,the disulfide cross-linking experiments with cysteine placedthroughout these helices did not capture any heterodimer (datanot shown), suggesting that the distances between these heliceswere too long to be bridged by a disulfide bond. In support ofthis notion, chemical cross-linking with BMH, a membrane-permeant and sulfhydryl-reactive bifunctional reagent, gener-ated a heterodimer between Bcl-XL T219Cwith the cysteine inhelix 9 and Bax K123C, R134C, or L181C with the cysteine inhelix 5, 6, or 9, respectively (Fig. 14). Because the two sulfhy-dryl-reactive maleimides in BMH were spaced �13 Å apart,and the reactive nitrene in the photoreactive probe was �12 Åaway from the C� of the lysine, both photocross-linking andchemical cross-linking data were consistent with a model inwhich helix 9 of Bcl-XL is embedded together with helices 5, 6,and 9 of Bax in the membrane, without being close enough toform an additional heterodimer interface.We also used chemical cross-linking to assess the distance

between the cytosolic part and the membrane-embedded partof the Bcl-XL�Bax heterodimer. As shown in Fig. 14, Bcl-XLT216CorT219Cwith the cysteine in helix 9was cross-linked toBax Cys62, M74C, or M79C with the cysteine in the BH3 helixby BMH,whereas Bcl-XLL194Cwith the cysteine in helix 8wascross-linked to Bax R94C with the cysteine in helix 4. Thus, thedistances between these paired cysteines were �13 Å. Theseexperimentally determined distances, together with the infor-mation about the interface and the membrane topology, wereused to build a schematic model for the overall structural orga-nization of the Bcl-XL�Bax heterodimer that was partiallyembedded in the membrane (Fig. 15).

DISCUSSION

The BH1–3 Groove�BH3 Helix Interaction in the Membrane-bound Bcl-XL�Bax Heterodimer and the Functional Consequence—Accumulating mutagenesis and structure data indicate that aBcl-XLBH1–3 groove�BaxBH3helix interface is the exclusive siteof interaction between the two proteins and therefore responsi-ble for inhibition of Bax by Bcl-XL. However, it was still unclearwhether this interaction existed in the membrane-bound com-plex and how this interaction integrated into the inhibitoryfunction of Bcl-XL on Bax-mediated MOMP. We used site-specific cross-linking approaches to systematically map theinterface of the Bcl-XL�Bax heterodimer formed at the naturaland artificial MOM. Collectively, our data support a directphysical interaction between the hydrophobic BH1–3 groove ofBcl-XL and the amphipathic BH3 helix of Bax for the mem-brane-bound proteins, with an interface similar to that revealedby crystallography for the soluble Bcl-XL protein and Bax BH3

FIGURE 11. Effects of ABT-737 on interactions and MOMP activity of Bcl-XL and Bax. A-C, the indicated in vitro synthesized [35S]Met-labeled single-cysteineBax (A), Bax and Bcl-XL (B and C), or Bax and Bcl-XL R139D (B) proteins were activated by the BH3 peptide and targeted to the MOM liposomes in the absenceor presence of ABT-737, purified non-radioactive recombinant Bcl-XL protein (rBcl-XL), or both. The resulting proteoliposomes were processed and analyzedas described in the legend to Fig. 3. D, cytochrome c release from the Bax�/�/Bak�/� mitochondria by the indicated single-cysteine Bax proteins synthesizedin vitro was assayed in the absence or presence of the BH3 peptide, the indicated single-cysteine Bcl-XL protein synthesized in vitro, and/or ABT-737, asdescribed in the legend to Fig. 2. Data shown are average fractions of cytochrome c release from 2– 6 independent experiments with the ranges indicated byerror bars.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

peptide (17). These data included the following: (i) formation ofdisulfide bonds between cysteines in the groove of Bcl-XL andthose in the BH3 helix of Bax; (ii) formation of heterodimer-

specific photoadducts via ANB-labeled lysines in the grooveand the BH3 helix; and (iii) disruption of the groove�BH3 inter-action by the mutations in the groove and the BH3 helix and by

FIGURE 12. IASD-labeling of single-cysteine Bax and Bcl-XL proteins in liposomal membranes. The [35S]Met-labeled mutants with a single cysteinepositioned in helix 1 and BH1–3 groove of Bcl-XL (A), helix 1 and BH3 helix of Bax (B), or helices 5, 6, and 9 of Bax (C, top) or of Bcl-XL (D, top) were synthesizedin the wheat germ-based in vitro system. The resulting radioactive Bcl-XL or Bax protein, either alone or together with the purified non-radioactive recombinantBax (rBax) or Bcl-XL (rBcl-XL) protein, respectively, were activated by the BH3 peptide and targeted to the MOM liposomes. The resulting proteoliposomes wereisolated and treated with IASD in the absence or presence of CHAPS, urea, or both. After 30 min, the labeling reactions were stopped by �-mercaptoethanol.For the 0 min controls, the samples were pretreated with �-mercaptoethanol before the addition of IASD. The resulting radioactive proteins were resolvedusing IEF and detected by phosphorimaging. Circles and triangles, unlabeled and IASD-labeled Bcl-XL proteins, respectively; square and angle brackets,unlabeled and IASD-labeled Bax, respectively. The phosphorimaging data for IASD labeling of radioactive Bcl-XL and Bax mutants in the top panels of C and Dand the similar data from 1– 4 independent replicates were quantified. In the corresponding bottom panels, the average fractions of IASD labeling for themutants under the specified conditions are presented as bar graphs of the specified patterns with the ranges indicated by error bars. E, the indicated[35S]Met-labeled Bax mutants were synthesized in the wheat germ (WG)- or rabbit reticulocyte lysate (RRL)-based in vitro translation system and analyzed withIEF and phosphorimaging. The K119C or D142C mutant was constructed from the C0 mutant by changing Lys119 or Asp142 to cysteine, respectively.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

the groove-binding compound ABT-737. Interestingly, previ-ous studies showed that two of themutations, G138A in Bcl-XLand M74E in Bax, which inhibited the heterodimerization inmembranes in our study, abolished Bcl-XL inhibition of Bax incells (17, 43). In addition, ectopic expression of Bax in cellsenhances sensitivity to ABT-737 when Bcl-XL is present (44).Consistent with these findings, these mutations and ABT-737reversed the Bcl-XL repression of Bax-mediatedMOMP in ourcell-free system, suggesting that the BH1–3 groove�BH3 helixinteraction in the membrane-bound Bcl-XL�Bax heterodimerinhibits Bax.It was suggested that the BH3 motif of Bax is exposed upon

activation, resulting in Bax autoactivation and homo-oligomer-ization (5, 9, 10, 12, 13, 24, 45). The results presented here show

that the BH3motif of Bax is sequestered in the BH1–3 grooveof Bcl-XL and that this heterointeraction inhibits Baxhomointeraction in membranes via its BH3�BH1–3 grooveinterface. Therefore, the binding of Bcl-XL to Bax couldpotentially neutralize two activities mediated by the BH3motif of active Bax in membranes: (i) the recruitment ofcytosolic Bax to the membranes by preventing Bax autoacti-vation, facilitating retrotranslocation of Bax frommitochon-dria to cytosol (46), or both and (ii) the homo-oligomeriza-tion of membrane-bound Bax. Both of these activities areessential for Bax-mediated MOMP.The Helix 1�Helix 1 Interaction in the Membrane-bound

Bcl-XL�Bax Heterodimer and the Functional Implication—Al-though it was well documented that the BH3motifs of proapo-

FIGURE 12—continued

FIGURE 13. Photocross-linking of Bcl-XL and Bax proteins with a single photoreactive ANB-lysine incorporated into the membrane-embedded heli-ces. The in vitro synthesized [35S]Met-labeled Bcl-XL (A) or Bax (B) protein with the photoreactive probe attached to the indicated single lysine in helix 5, 6, or9 was activated together with purified 6H-Bax (A) or 6H-Bcl-XL (B) protein, respectively, by the BH3 peptide and targeted to the MOM liposomes. The resultingproteoliposomes were processed and analyzed as described in the legend to Fig. 4.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

ptotic members of Bcl-2 family and the BH1–3 groove ofantiapoptotic members dictate their binding affinities andspecificities in solution (47–49), it was speculated that regionsother than BH1–3 in Bax and Bcl-XL may contribute to theirhetero- and homointeractions and functions. First, Baxhomo-oligomerization is mediated not only by the BH1–3regions but also by helices 1 and 6 and the loop connectinghelices 1 and 2 (9, 10, 12). Second, Bax activation by BH3-only proteins and autoactivation by previously activated Baxare mediated by a BH3-binding triggering pocket composed

of helices 1 and 6 and the loop connecting helices 1 and 2 (7,45, 50, 51). Third, the helix 1 of Bcl-XL, which overlaps withthe BH4 region, is essential for binding to soluble Bax,thereby preventing Bax from inserting into membranes andperhaps retrotranslocating Bax from membranes to cytosoland is hence required for Bcl-XL to inhibit Bax-mediatedapoptosis (5, 52). Fourth, the 6A7 epitope of Bax, whichoverlaps with helix 1, is not accessible to the 6A7 antibody inthe Bax sequestered by Bcl-XL in cells (39, 53).Here, we demonstrated that the helix 1 regions in mem-

brane-bound Bax and Bcl-XL directly interacted with eachother and contributed to the functionality of the membrane-bound heterodimer. In contrast to the rigid groove�BH3interface, the helix 1�helix 1 interface is flexible. The grooveand the BH3 helix are formed by well organized structuralelements that are ideal for forming a rigid binding interface.In contrast, the helix 1 in both proteins is connected to therest of the protein by a long and unstructured loop that pro-vides the freedom for the helices to sample different bindinginterfaces to achieve a flexible interaction. It will be interest-ing to explore whether and how themultiple BH and non-BHregions of Bcl-XL and Bax are involved in the interactionswith other family members at membranes, because theseinteractions are expected to alter the functional Bcl-XL�Baxinteraction according to our embedded together model (3,54). The assays established here will be valuable tools forthese explorations.Beside disulfide-linking to each other, Bcl-XL and Bax with a

single cysteine in the BH4/helix 1 region could be disulfide-linked to a few unknown mitochondrial proteins (Fig. 7), sug-gesting that this region of Bcl-XL and Baxmight exert activitiesother than regulation ofMOMP. Bcl-XL was shown to regulate

FIGURE 14. Chemical cross-linking of single-cysteine Bcl-XL and Bax pro-teins. The indicated in vitro synthesized [35S]Met-labeled single-cysteineBcl-XL and Bax proteins were activated by the BH3 peptide and targeted tothe MOM liposomes. The resulting proteoliposomes were isolated and sub-jected to BMH cross-linking. The resulting radioactive proteins and theiradducts were analyzed by reducing SDS-PAGE and phosphorimaging.Arrows, BMH-cross-linked heterodimers. Other labels are the same as those inFig. 3.

FIGURE 15. A model for the overall structural organization of the Bcl-XL�Bax heterodimer in membranes. The cytosolic part of the heterodimer model wasassembled by merging the BH1–3 groove�BH3 helix interface model (Fig. 5A) with the parallel helix 1�helix 1 interface model (Fig. 8A, right). The MOM-embedded part was assembled with the �-helix 9 (�9) of Bcl-XL and the �-helices 5, 6, and 9 (�5, �6, and �9) of Bax. The distances between these membrane-embedded helices were set according to the BMH cross-linking data (Fig. 14). The distances between the helix �8 or �9 of Bcl-XL and the BH3 helix (�BH3) orhelix �4 of Bax, respectively, were further adjusted based on the BMH cross-linking data. Dashed lines link the residues that were cross-linked by BMH when theywere mutated to cysteines. The residues shown as spheres are embedded in the MOM, because their cysteine substitutions could not be substantially labeledby IASD unless CHAPS was added (Fig. 12). The residues shown as sticks are exposed to the cytosol, because their cysteine substitutions could be labeled byIASD in the absence of CHAPS and urea.



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

mitochondrial membrane potential by modulating the activity ofvoltage-dependent anion channel via an interaction that requiresthe BH4/helix 1 region (55). The soluble form of Bax, which wasunable to promote MOMP, positively regulated mitochondrialfusion through interaction with mitofusin-2 (56). On the otherside, the MOM-bound Bax, even when sequestered by Bcl-XLand unable to induce MOMP, promoted mitochondrial fission(39, 57). Interaction betweenBcl-XL andBaxmay regulate theirinteractions with these and other mitochondrial proteins andthe corresponding activities, some of which are intimately con-nected with apoptosis (58, 59).Does the StructuralDifference between the BaxHomocomplex

and the Bax�Bcl-XL Heterocomplex Explain Their FunctionalDifference?—Despite the striking similarity in their monomericstructure, it is puzzling how Bcl-XL interacts with Bax andthereby halts the homo-oligomerization of Bax. It is alsounclear why Bcl-XL cannot form large oligomeric pores inmembranes but Bax can.Now it becomes evenmore perplexingthat Bcl-XL and Bax interact at membranes via two interfaces,the canonical groove�BH3 helix interface and the novel helix1�helix 1 interface, which potentially canmediate the formationof a higher order hetero-oligomer. However, the disruptivemutations in the former interface also partially disrupt the lat-ter one, suggesting that the formation of the latter interfacepartially depends on the formation of the former one. Oneexplanation of this result is that both interfaces are used to forma more stable heterodimer, as shown in our overall structuralmodel (Fig. 15). Alternatively, the binding mediated by oneinterface is too weak to extend the heterodimer that is formedby the other interface to a higher order oligomer. In addition,Bax homo-oligomers are probably formed by the monomersthat are deeply embedded into membranes (8). In contrast, ourtopology data for the Bcl-XL�Bax heterocomplex (Fig. 12) sug-gest that although the embedding of the Bax in the hetero-complex in the membrane is similar to that of the Bax in thehomo-complex, the embedding of Bcl-XL seems to be shal-lower compared with that of Bax. Therefore, the different con-formations of Bcl-XL and Bax in themembranemay be anotherfactor that distinguishes the activities of the two proteins in themembrane.Are Bcl-XL and Bcl-2 Different?—Besides the subcellular

localization, the differences between Bcl-2 and Bcl-XL, twoantiapoptotic proteins with similar soluble domain structures,were implied in many ways. First, Bcl-2 and Bcl-XL have differ-ent binding spectrums to BH3-only proteins as well as to Baxand Bak (39, 48). Second, Bcl-XL is more effective than Bcl-2 ininhibition of apoptosis in some cell types (60). Third, our ISADlabeling data suggest that the Bcl-XL in complex with Bax maybe in the same tail-anchored conformation that it adopts whentargeted to membranes in the absence of Bax. In contrast, thefull antiapoptotic function of Bcl-2, including inhibition of Bax,requires a changed membrane-bound conformation (18, 20).Fourth, we found that Bcl-2 can be converted to a Bax-likemolecule through interaction with Bim and tBid, whereas asimilar conversion of Bcl-XL was not detected at least in lipo-somes (61).Conclusions—Our study revealed two interfaces in the

membrane-bound Bcl-XL�Bax complex. The first one is rigid

with the BH3 helix of Bax sequestered in the BH1–3 grooveof Bcl-XL, whereas the second one is flexible with the twohelices 1 interacting either in parallel or antiparallel. Forma-tion of the second interface largely depends on the first one,yet both contribute to the overall stability of the heterodimerthat inhibits homo-oligomerization of Bax and subsequentMOMP. Therefore, the rigid BH3�groove interface positionsthe two helices 1 such that they form a flexible interface,which contributes significantly to energetic stabilization ofthe membrane-bound Bcl-XL�Bax heterodimer that is piv-otal to apoptosis regulation.

Acknowledgments—We thank Jarkko Ylanko for helping with some ofthe data analysis and Kathy Kyler for editing the manuscript.

REFERENCES1. Kvansakul, M., Yang, H., Fairlie, W. D., Czabotar, P. E., Fischer, S. F.,

Perugini, M. A., Huang, D. C., and Colman, P. M. (2008) Vaccinia virusanti-apoptotic F1L is a novel Bcl-2-like domain-swapped dimer that bindsa highly selective subset of BH3-containing death ligands. Cell Death Dif-fer. 15, 1564–1571

2. Youle, R. J., and Strasser, A. (2008) The BCL-2 protein family: opposingactivities that mediate cell death. Nat. Rev. Mol. Cell Biol. 9, 47–59

3. Leber, B., Lin, J., and Andrews, D. W. (2007) Embedded together: the lifeand death consequences of interaction of the Bcl-2 family with mem-branes. Apoptosis 12, 897–911

4. Chipuk, J. E., Moldoveanu, T., Llambi, F., Parsons, M. J., and Green, D. R.(2010) The BCL-2 family reunion.Mol. Cell 37, 299–310

5. Billen, L. P., Kokoski, C. L., Lovell, J. F., Leber, B., and Andrews, D. W.(2008) Bcl-XL inhibits membrane permeabilization by competing withBax. PLoS Biol. 6, e147

6. Lovell, J. F., Billen, L. P., Bindner, S., Shamas-Din, A., Fradin, C., Leber, B.,and Andrews, D. W. (2008) Membrane binding by tBid initiates an or-dered series of events culminating in membrane permeabilization by Bax.Cell 135, 1074–1084

7. Tan, C., Dlugosz, P. J., Peng, J., Zhang, Z., Lapolla, S. M., Plafker, S. M.,Andrews, D. W., and Lin, J. (2006) Auto-activation of the apoptosis pro-tein Bax increases mitochondrial membrane permeability and is inhibitedby Bcl-2. J. Biol. Chem. 281, 14764–14775

8. Annis, M. G., Soucie, E. L., Dlugosz, P. J., Cruz-Aguado, J. A., Penn, L. Z.,Leber, B., andAndrews,D.W. (2005) Bax formsmultispanningmonomersthat oligomerize to permeabilize membranes during apoptosis. EMBO J.24, 2096–2103

9. Zhang, Z., Zhu, W., Lapolla, S. M., Miao, Y., Shao, Y., Falcone, M., Bore-ham, D., McFarlane, N., Ding, J., Johnson, A. E., Zhang, X. C., Andrews,D. W., and Lin, J. (2010) Bax forms an oligomer via separate, yet interde-pendent, surfaces. J. Biol. Chem. 285, 17614–17627

10. Bleicken, S., Classen, M., Padmavathi, P. V., Ishikawa, T., Zeth, K., Stein-hoff, H. J., and Bordignon, E. (2010) Molecular details of Bax activation,oligomerization, andmembrane insertion. J. Biol. Chem. 285, 6636–6647

11. Dewson, G., Kratina, T., Sim, H. W., Puthalakath, H., Adams, J. M., Col-man, P. M., and Kluck, R. M. (2008) To trigger apoptosis, Bak exposes itsBH3 domain and homodimerizes via BH3:groove interactions. Mol. Cell30, 369–380

12. Dewson, G., Ma, S., Frederick, P., Hockings, C., Tan, I., Kratina, T., andKluck, R. M. (2012) Bax dimerizes via a symmetric BH3:groove interfaceduring apoptosis. Cell Death Differ. 19, 661–670

13. Czabotar, P. E., Westphal, D., Dewson, G., Ma, S., Hockings, C., Fairlie,W. D., Lee, E. F., Yao, S., Robin, A. Y., Smith, B. J., Huang, D. C., Kluck,R.M., Adams, J.M., andColman, P.M. (2013) Bax crystal structures revealhow BH3 domains activate Bax and nucleate its oligomerization to induceapoptosis. Cell 152, 519–531

14. Fletcher, J. I.,Meusburger, S., Hawkins, C. J., Riglar, D. T., Lee, E. F., Fairlie,W. D., Huang, D. C., and Adams, J. M. (2008) Apoptosis is triggered when



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

prosurvival Bcl-2 proteins cannot restrain Bax. Proc. Natl. Acad. Sci.U.S.A. 105, 18081–18087

15. Desagher, S., Osen-Sand, A., Nichols, A., Eskes, R., Montessuit, S., Lauper,S., Maundrell, K., Antonsson, B., and Martinou, J. C. (1999) Bid-inducedconformational change of Bax is responsible for mitochondrial cyto-chrome c release during apoptosis. J. Cell Biol. 144, 891–901

16. Sedlak, T. W., Oltvai, Z. N., Yang, E., Wang, K., Boise, L. H., Thompson,C. B., and Korsmeyer, S. J. (1995) Multiple Bcl-2 family members demon-strate selective dimerizations with Bax. Proc. Natl. Acad. Sci. U.S.A. 92,7834–7838

17. Czabotar, P. E., Lee, E. F., Thompson, G. V., Wardak, A. Z., Fairlie, W. D.,and Colman, P. M. (2011) Mutation to Bax beyond the BH3 domain dis-rupts interactions with pro-survival proteins and promotes apoptosis.J. Biol. Chem. 286, 7123–7131

18. Dlugosz, P. J., Billen, L. P., Annis, M. G., Zhu, W., Zhang, Z., Lin, J., Leber,B., and Andrews, D. W. (2006) Bcl-2 changes conformation to inhibit Baxoligomerization. EMBO J. 25, 2287–2296

19. Kim, P. K., Annis, M. G., Dlugosz, P. J., Leber, B., and Andrews, D. W.(2004) During apoptosis Bcl-2 changes membrane topology at both theendoplasmic reticulum and mitochondria.Mol. Cell 14, 523–529

20. Peng, J., Tan, C., Roberts, G. J., Nikolaeva, O., Zhang, Z., Lapolla, S. M.,Primorac, S., Andrews, D. W., and Lin, J. (2006) tBid elicits a conforma-tional alteration in membrane-bound Bcl-2 such that it inhibits Bax poreformation. J. Biol. Chem. 281, 35802–35811

21. Ding, J., Zhang, Z., Roberts, G. J., Falcone, M., Miao, Y., Shao, Y., Zhang,X. C., Andrews, D. W., and Lin, J. (2010) Bcl-2 and Bax interact via theBH1–3 groove-BH3 motif interface and a novel interface involving theBH4 motif. J. Biol. Chem. 285, 28749–28763

22. Oltersdorf, T., Elmore, S. W., Shoemaker, A. R., Armstrong, R. C., Augeri,D. J., Belli, B. A., Bruncko, M., Deckwerth, T. L., Dinges, J., Hajduk, P. J.,Joseph, M. K., Kitada, S., Korsmeyer, S. J., Kunzer, A. R., Letai, A., Li, C.,Mitten, M. J., Nettesheim, D. G., Ng, S., Nimmer, P. M., O’Connor, J. M.,Oleksijew, A., Petros, A.M., Reed, J. C., Shen,W., Tahir, S. K., Thompson,C. B., Tomaselli, K. J.,Wang, B.,Wendt,M. D., Zhang, H., Fesik, S.W., andRosenberg, S. H. (2005) An inhibitor of Bcl-2 family proteins inducesregression of solid tumours. Nature 435, 677–681

23. Yethon, J. A., Epand, R. F., Leber, B., Epand, R. M., and Andrews, D. W.(2003) Interaction with a membrane surface triggers a reversible confor-mational change in Bax normally associated with induction of apoptosis.J. Biol. Chem. 278, 48935–48941

24. Kim, H., Tu, H. C., Ren, D., Takeuchi, O., Jeffers, J. R., Zambetti, G. P.,Hsieh, J. J., and Cheng, E. H. (2009) Stepwise activation of BAX and BAKby tBID, BIM, and PUMA initiates mitochondrial apoptosis.Mol. Cell 36,487–499

25. Lin, J., Liang, Z., Zhang, Z., and Li, G. (2001) Membrane topography andtopogenesis of prenylated Rab acceptor (PRA1). J. Biol. Chem. 276,41733–41741

26. Zhang, Z., Lapolla, S. M., Annis, M. G., Truscott, M., Roberts, G. J., Miao,Y., Shao, Y., Tan, C., Peng, J., Johnson, A. E., Zhang, X. C., Andrews, D.W.,and Lin, J. (2004) Bcl-2 homodimerization involves two distinct bindingsurfaces, a topographic arrangement that provides an effective mecha-nism for Bcl-2 to capture activated Bax. J. Biol. Chem. 279, 43920–43928

27. Dunbrack, R. L., Jr., and Karplus, M. (1993) Backbone-dependent rotamerlibrary for proteins. Application to side-chain prediction. J. Mol. Biol. 230,543–574

28. Emsley, P., Lohkamp, B., Scott,W. G., and Cowtan, K. (2010) Features anddevelopment of Coot. Acta Crystallogr. D Biol. Crystallogr. 66, 486–501

29. Chen, V. B., Arendall, W. B., 3rd, Headd, J. J., Keedy, D. A., Immormino,R.M., Kapral, G. J., Murray, L.W., Richardson, J. S., and Richardson, D. C.(2010) MolProbity: all-atom structure validation for macromolecularcrystallography. Acta Crystallogr. D Biol. Crystallogr. 66, 12–21

30. Raveh, B., London, N., and Schueler-Furman, O. (2010) Sub-angstrommodeling of complexes between flexible peptides and globular proteins.Proteins 78, 2029–2040

31. London, N., Raveh, B., Cohen, E., Fathi, G., and Schueler-Furman, O.(2011) Rosetta FlexPepDock web server: high resolutionmodeling of pep-tide-protein interactions. Nucleic Acids Res. 39,W249–W253

32. Kelley, L. A., and Sternberg, M. J. (2009) Protein structure prediction on

the Web: a case study using the Phyre server. Nat. Protoc. 4, 363–37133. Ozhogina, O. A., and Bominaar, E. L. (2009) Characterization of the krin-

gle fold and identification of a ubiquitous new class of disulfide rotamers.J. Struct. Biol. 168, 223–233

34. Braun, T. F., and Blair, D. F. (2001) Targeted disulfide cross-linking of theMotB protein of Escherichia coli: evidence for two H� channels in thestator complex. Biochemistry 40, 13051–13059

35. Trakselis, M. A., Alley, S. C., and Ishmael, F. T. (2005) Identification andmapping of protein-protein interactions by a combination of cross-link-ing, cleavage, and proteomics. Bioconjug. Chem. 16, 741–750

36. Brunner, J. (1993) New photolabeling and crosslinking methods. Annu.Rev. Biochem. 62, 483–514

37. London, N., Gulla, S., Keating, A. E., and Schueler-Furman, O. (2012) Insilico and in vitro elucidation of BH3 binding specificity toward Bcl-2.Biochemistry 51, 5841–5850

38. Aranovich, A., Liu, Q., Collins, T., Geng, F., Dixit, S., Leber, B., and An-drews, D. W. (2012) Differences in the mechanisms of proapoptotic BH3proteins binding to Bcl-XL and Bcl-2 quantified in live MCF-7 cells.Mol.Cell 45, 754–763

39. Llambi, F., Moldoveanu, T., Tait, S. W., Bouchier-Hayes, L., Temirov, J.,McCormick, L. L., Dillon, C. P., andGreen, D. R. (2011) A unifiedmodel ofmammalian BCL-2 protein family interactions at the mitochondria.Mol.Cell 44, 517–531

40. Lee, E. F., Czabotar, P. E., Smith, B. J., Deshayes, K., Zobel, K., Colman,P. M., and Fairlie, W. D. (2007) Crystal structure of ABT-737 complexedwith Bcl-xL: implications for selectivity of antagonists of the Bcl-2 family.Cell Death Differ. 14, 1711–1713

41. Muchmore, S. W., Sattler, M., Liang, H., Meadows, R. P., Harlan, J. E.,Yoon, H. S., Nettesheim, D., Chang, B. S., Thompson, C. B., Wong, S. L.,Ng, S. L., and Fesik, S. W. (1996) X-ray and NMR structure of humanBcl-xL, an inhibitor of programmed cell death. Nature 381, 335–341

42. Suzuki, M., Youle, R. J., and Tjandra, N. (2000) Structure of Bax: coregu-lation of dimer formation and intracellular localization.Cell 103, 645–654

43. Ottilie, S., Diaz, J. L., Chang, J., Wilson, G., Tuffo, K. M., Weeks, S.,McConnell, M.,Wang, Y., Oltersdorf, T., and Fritz, L. C. (1997) Structuraland functional complementation of an inactive Bcl-2 mutant by Bax trun-cation. J. Biol. Chem. 272, 16955–16961

44. Gautier, F., Guillemin, Y., Cartron, P. F., Gallenne, T., Cauquil, N., LeDiguarher, T., Casara, P., Vallette, F. M., Manon, S., Hickman, J. A., Gen-este, O., and Juin, P. (2011) Bax activation by engagement with, then re-lease from, the BH3 binding site of Bcl-xL.Mol. Cell Biol. 31, 832–844

45. Gavathiotis, E., Reyna, D. E., Davis, M. L., Bird, G. H., andWalensky, L. D.(2010) BH3-triggered structural reorganization drives the activation ofproapoptotic BAX.Mol. Cell 40, 481–492

46. Edlich, F., Banerjee, S., Suzuki, M., Cleland, M. M., Arnoult, D., Wang, C.,Neutzner, A., Tjandra, N., and Youle, R. J. (2011) Bcl-xL retrotranslocatesBax from the mitochondria into the cytosol. Cell 145, 104–116

47. Certo, M., Del Gaizo Moore, V., Nishino, M., Wei, G., Korsmeyer, S.,Armstrong, S. A., and Letai, A. (2006) Mitochondria primed by deathsignals determine cellular addiction to antiapoptotic BCL-2 family mem-bers. Cancer Cell 9, 351–365

48. Chen, L., Willis, S. N., Wei, A., Smith, B. J., Fletcher, J. I., Hinds, M. G.,Colman, P. M., Day, C. L., Adams, J. M., and Huang, D. C. (2005) Differ-ential trageting of prosurvival Bcl-2 proteins by their BH3-only ligandsallows complementary apoptotic function.Mol. Cell 17, 393–403

49. Kuwana, T., Bouchier-Hayes, L., Chipuk, J. E., Bonzon, C., Sullivan, B. A.,Green, D. R., and Newmeyer, D. D. (2005) BH3 domains of BH3-onlyproteins differentially regulate Bax-mediated mitochondrial membranepermeabilization both directly and indirectly.Mol. Cell 17, 525–535

50. Cartron, P.-F., Gallenne, T., Bougras, G., Gautier, F., Manero, F., Vusio, P.,Meflah, K., Vallette, F. M., and Juin, P. (2004) The first � helix of Bax playsa necessary role in its ligand-induced activation by the BH3-only proteinsBid and PUMA.Mol. Cell 16, 807–818

51. Gavathiotis, E., Suzuki, M., Davis, M. L., Pitter, K., Bird, G. H., Katz, S. G.,Tu, H. C., Kim, H., Cheng, E. H., Tjandra, N., and Walensky, L. D. (2008)BAX activation is initiated at a novel interaction site. Nature 455,1076–1081

52. Zhou, H., Hou, Q., Chai, Y., and Hsu, Y. T. (2005) Distinct domains of



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

Bcl-XL are involved in Bax and Bad antagonism and in apoptosis inhibi-tion. Exp. Cell Res. 309, 316–328

53. Hsu, Y. T., Wolter, K. G., and Youle, R. J. (1997) Cytosol-to-membraneredistribution of Bax and Bcl-XL during apoptosis. Proc. Natl. Acad. Sci.U.S.A. 94, 3668–3672

54. Leber, B., Lin, J., and Andrews, D. W. (2010) Still embedded togetherbinding to membranes regulates Bcl-2 protein interactions.Oncogene 29,5221–5230

55. Shimizu, S., Konishi, A., Kodama, T., and Tsujimoto, Y. (2000) BH4 do-main of antiapoptotic Bcl-2 family members closes voltage-dependentanion channel and inhibits apoptotic mitochondrial changes and celldeath. Proc. Natl. Acad. Sci. U.S.A. 97, 3100–3105

56. Hoppins, S., Edlich, F., Cleland, M. M., Banerjee, S., McCaffery, J. M.,Youle, R. J., and Nunnari, J. (2011) The soluble form of Bax regulatesmitochondrial fusion via MFN2 homotypic complexes. Mol. Cell 41,150–160

57. Sheridan, C., Delivani, P., Cullen, S. P., and Martin, S. J. (2008) Bax- orBak-induced mitochondrial fission can be uncoupled from cytochrome crelease.Mol. Cell 31, 570–585

58. Suen, D. F., Norris, K. L., and Youle, R. J. (2008) Mitochondrial dynamicsand apoptosis. Genes Dev. 22, 1577–1590

59. Tsujimoto, Y., and Shimizu, S. (2007) Role of the mitochondrial mem-brane permeability transition in cell death. Apoptosis 12, 835–840

60. Fiebig, A. A., Zhu, W., Hollerbach, C., Leber, B., and Andrews, D. W.(2006) Bcl-XL is qualitatively different from and ten times more effec-tive than Bcl-2 when expressed in a breast cancer cell line. BMCCancer6, 213

61. Zhao, L., He, F., Liu, H., Zhu, Y., Tian,W., Gao, P., He, H., Yue,W., Lei, X.,Ni, B., Wang, X., Jin, H., Hao, X., Lin, J., and Chen, Q. (2012) Naturalditerpenoid compound elevates expression of Bim protein, which inter-acts with antiapoptotic protein Bcl-2, converting it to proapoptotic Bax-like molecule. J. Biol. Chem. 287, 1054–1065



at Univ of O

klahoma H



ww

.jbc.org/D

ownloaded from

http://www.jbc.org/

After Embedding in Membranes Antiapoptotic Bcl-XL Protein Binds Both Bcl-2 Homology Region 3 and Helix 1 of Proapoptotic Bax Protein to Inhibit Apoptotic Mitochondrial Permeabilization

Documents