Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Double identity for proteins of the Bcl-2 family


Bcl-2 is an oncogenic protein that acts by inhibiting programmed cell death. The mechanisms used by this and related anti-apoptotic proteins to protect cells from cytotoxic stimuli are now emerging, with the discovery that Bcl-2 can function both as an ion channel and as an adaptor or docking protein.


Since its discovery over ten years ago as an oncogenic protein involved in human tumours1, Bcl-2 has defied attempts to determine the biochemical basis for its potent anti-apoptotic action. Recent findings, however, are beginning to reveal details of the mechanisms by which Bcl-2 and its homologous proteins, such as Bcl-XL, suppress cell death. A picture is emerging of a complicated protein, which in some ways resembles p53 in that it has multiple independent functions, rather like a Swiss army knife.

Although Bcl-2 family proteins lack significant amino-acid sequence homology with other proteins, the three-dimensional structure of Bcl-XL2 reveals a striking similarity to the pore-forming domains of certain bacterial toxins that act as channels for either ions or proteins. The closest in structure to Bcl-XL are diphtheria toxin, which transports a fragment of the toxin across the cell membrane, and the bacterial pore-forming colicins, which kill Escherichia coli by forming non-selective ion channels. If function follows form, the obvious prediction is that Bcl-2 family proteins can function as channels for ions, proteins, or both.

Circumstantial evidence from cell transfection studies indicated that Bcl-2 might have a membrane transport function, with reported effects on Ca2+flux and protein translocation across some of the intracellular membranes where Bcl-2 and its homologues are localized, namely the outer mitochondrial membrane, the endoplasmic reticulum (ER) and the nuclear envelope3,4,5. Now, direct evidence of ion-channel activity has been obtained from experiments in which the effects of recombinant Bcl-2 or Bcl-XL in synthetic lipid membranes were studied by using single-channel recordings from planar bilayers and by other approaches6,7. The mechanism by which Bcl-2 and Bcl-XL create channels in membranes has not been explored in detail, but preliminary indications are that at least some aspects of the process may be similar to the bacterial toxins. For example, the structure of Bcl-XL reveals a compact seven α-helical bundle, at the centre of which lies a hairpin comprised of two long hydrophobic core α-helices surrounded by five amphipathic α-helices2. By analogy to the bacterial toxins, the two core α-helices (α5 and α6) presumably insert perpendicularly across the lipid bilayer, with the protein undergoing profound conformational changes, perhaps opening like an umbrella, with the surrounding amphipathic helices folding upwards and resting on top of the membrane. Consistent with this idea, removal of α5 and α6 from Bcl-2 prevents channel formation in vitro and abolishes its anti-apoptotic effect in cells7.

Details about the structures of Bcl-2 and Bcl-XL channels are still to be resolved and already controversy has developed about the likely size (diameters) of these channels. For example, it has been reported that Bcl-2 and Bcl-XL form discrete ion channels in vitro which mostly assume a closed conformation, open only sporadically, and yield conductances ranging from 20 to 300 picosiemens (pS)6,7. However, evidence has been found for larger pores under some conditions, which assume a mostly open conformation and produce channels with conductances >1nS (S. Korsmeyer, personal communication). Related to the issue of pore diameter is the question of the stoichiometry of these channels. Although the predicted membrane-spanning α5 and α6 of Bcl-2 and Bcl-XL are mostly hydrophobic, as required for insertion into membranes, one face of the helices has hydrophilic residues, which presumably line the aqueous lumen of the channels. By analogy to other natural and synthetic α-helix-type channels (ref. 8), it follows that the membrane-inserted helices from two or more Bcl-2/Bcl-XL molecules should associate to form a ring that creates an aqueous lumen. How many molecules of Bcl-2 or Bcl-XL assemble in membranes to create channels is not known, but the existence of multiple Bcl-2/Bcl-XL conductance states suggests a range of possibilities, at least in vitro6,7. There is evidence that Bcl-XL can function as a monomer, but so far we do not have the methodology to determine the oligomerization state of this protein when it is integrated into membranes and so cannot discount the channel hypothesis. The challenge now is to demonstrate that Bcl-2/Bcl-XL channels exist in vivo and to determine their structures.

Complicating the question of the biological significance of these findings are data indicating that the pro-apoptotic Bcl-2 family member Bax, as well as Bcl-2 and Bcl-XL, have channel activity in vitro (J. C. Martinou, personal communication; S. Schendel, Z. Xie and J.C.R., unpublished results). As Bax can heterodimerize with either Bcl-2 or Bcl-XL when they are in their compact α-helical-bundle conformations in aqueous environments, these proteins may produce heteromeric channels when integrated into membranes. Alternatively, their heterodimerization may prevent Bcl-2 and Bax from undergoing the conformational changes required for integration into membranes, thus preventing channel formation. Both Bcl-2 and Bax might therefore form cytotoxic channels in cells, with Bcl-2/Bax heterodimerizaiton nullifying channel activity and thus promoting cell survival. This idea predicts that the Bcl-2:Bax ratio is critical, consistent with the finding that excess Bcl-2 paradoxically promotes cell death in some circumstances9 and with evidence that Bax and Bak unexpectedly exhibit cytoprotective functions in some cellular contexts10,11. Or Bcl-2/Bcl-XL and Bax might have different selectivities, providing conduits for different ions or proteins that have opposing effects on apoptosis—this possibility could be tested by mutagenesis in which the presumed channel-forming fifth and sixth helices of these proteins are swapped.

Although at present we have no direct evidence for in vivo channel formation, the fact that Bcl-2 family proteins can form channels in vitro suggests that they may participate in some of the cellular phenomena recently associated with apoptosis, particularly mitochondrial permeability transition (‘megapore’ opening) and release of apoptogenic protease activators (cytochrome c and apoptosis-inducing factor (AIF)) from mitochondria. Mitochondrial permeability transition involves the opening of a large (1.3nS, 2.9nm) channel in the inner membrane of the mitochondrion12,13. This occurs almost universally during apoptosis and has repercussions that may contribute to the induction of cell death, including the generation of oxygen free radicals, dumping of stored Ca2+into the cytosol, and the release of mitochondrial proteins into the cytosol in order to activate the cysteine proteases (caspases) that are the terminal effectors of apoptosis14. Overexpression of Bcl-2 inhibits mitochondrial permeability transition, whereas Bax overexpression induces it15,16. Bcl-2 is found on the outer mitochondrial membrane and is enriched at the contact sites where the inner and outer membranes abut. As some proteins that regulate the megapore reside in the outer membrane, Bcl-2 and Bax could somehow modulate or participate in megapore formation, possibly by controlling ion fluxes that influence the opening and closing of the megapore. On the other hand, some studies suggest that caspase-activating proteins, in particular cytochrome c, can leave the mitochondrial intermembrane space through the outer membrane and enter the cytosol where the cell-death proteases are found, irrespective of mitochondrial permeability transition17,18. If so, channels must exist in the outer mitochondrial membrane that control the release of cytochrome c during apoptosis. It remains to be determined whether Bax can create such channels and whether the suppression of cytochrome c release by Bcl-2 reflects an ability of Bcl-2 either to plug Bax channels or to transport cytochrome c back into mitochondria1718.

Although the new-found channel activity of Bcl-2 family proteins may prove to be crucial to their function as regulators of cell life and death, these proteins have other important ways of controlling apoptosis. In this regard, Bcl-2 or Bcl-XL has been reported to bind (or at least coimmunoprecipitate with) eleven proteins, including the protein kinase Raf-1, the protein phosphatase calcineurin, the GTPases R-Ras and H-Ras, the p53-binding protein p53–BP2, the prion protein Pr-1, and several proteins with unknown functions, including CED-4, BAG-1, Nip-1, Nip-2 and Nip-3 (reviewed in ref. 19, and see below). Conversely, the pro-apoptotic protein Bax apparently does not interact with these proteins, suggesting an important difference between the death-suppressing and death-promoting members of the Bcl-2 family.

Although the functional significance of these interactions of proteins with Bcl-2 is unclear, recent discoveries are reducing our ignorance while also raising many questions. The developing theme is that Bcl-2 and its anti-apoptotic homologues, which are normally anchored in intracellular membranes (by virtue of a stretch of hydrophobic residues at their carboxy termini) but oriented towards the cytosol, act at least in part as adaptor or docking proteins which pull other proteins out of the cytosol, either sequestering them to the membrane alongside Bcl-2 (and probably inactivating them) or targeting them for interaction with other membrane-associated proteins.

Of the Bcl-2/Bcl-XL-binding proteins, CED-4 or its functional equivalents in mammalian cells is probably the most important. CED-4 is a pro-apoptotic protein from the nematode Caenorhabitis elegans20. In a genetic sense, CED-4 functions upstream of the worm homologue of the caspases (CED-3) but downstream of the worm version of Bcl-2 (CED-2) (ref. 21). Although they are functionally connected, it was unclear whether CED-9, CED-4 and CED-3 were physically linked or whether other proteins might be required to create a bridge between them. This speculation was stopped by the recent demonstration that CED-9 as well as Bcl-XL can bind to worm CED-4, which in turn binds to CED-3 or other caspases22,23,24. Indeed, CED-9, CED-4 and CED-3 can form a trimolecular complex23. The mammalian homologue of CED-4 remains at large, but experiments demonstrating co-immunoprecipitation of Bcl-XL with caspases such as FLICE (caspase-8) and ICE (caspase-1) in human cells indicate that CED-4 homologues may exist in higher organisms. A race is now underway to clone mammalian CED-4 and to define its three-dimensional structure, as well as the mechanism by which it may activate the caspases. In the worm, two forms of CED-4 have been described that arise through alternative splicing of messenger RNA, with the shorter CED-4 promoting apoptosis and the longer CED-4 promoting cell survival25. If this is a sign of things to come in mammalian species, then we can expect an additional level of complexity at the CED-4-dependent step of the cell-death pathway.

Another burning question is how do anti-apoptotic Bcl-2 family proteins such as CED-9 and Bcl-XL interfere with CED-4 function? Overexpression of worm CED-4 in mammalian cells induces apoptosis in a CED-9- and Bcl-XL-suppressible manner. CED-4 is normally found in the cytosol, but when co-expressed with CED-9 or Bcl-XL, it is sequestered to the intracellular membranes where CED-9 and Bcl-XL are located24. Conversely, when pro-apoptotic members of the Bcl-2 family such as Bax, Bak and Bik are co-expressed with Bcl-XL, they dimerize with Bcl-XL, thereby displacing CED-4 (Fig. 1). Thus, it appears that anti-apoptotic Bcl-2 family proteins pull CED-4 out of the cytosol, but the question of exactly how they inhibit CED-4 remains unanswered.

Figure 1: Model for CED-4-mediated bridging of Bcl-2/Bcl-XL to caspases.

a, CED-4 binding to cell-death proteases (caspases) such as CED-3 promotes apoptosis. b, Bcl-2, tethered to mitochondrial and other intracellular membranes, binds to CED-4, pulling it from the cytosol and presumably preventing activation of caspases. c, By heterodimerizing with Bcl-2, pro-apoptotic proteins such as Bax displace CED-4 and presumably allow it to return to the cytosol and to activate proteases.

Interactions with other proteins may not be as fundamental to Bcl-2's function as CED-4 binding, but they may account for other activities previously associated with Bcl-2 and provide insight into how Bcl-2 family proteins are regulated. For example, Bcl-2 (but not Bax) can associate with Raf-1, pulling this kinase from the cytosol into proximity with the Bcl-2 on the mitochondrial membrane. Once there, Raf-1 induces phosphorylation of BAD, a pro-apoptotic Bcl-2 family protein that abrogates the cytoprotective functions of Bcl-2 and Bcl-XL by heterodimerizing with them. The phosphorylated BAD (which unlike Bcl-2 has no membrane-anchoring domain) then dissociates from Bcl-2/Bcl-XL, moving into the cytosol in a complex with 14-3-3, where it cannot interfere with Bcl-2/Bcl-XL. Although BAD is one potential substrate of Bcl-2/Raf-1, presumably others exist as artificially targeting active Raf-1 to mitochondria can also suppress apoptosis in cells that do not contain BAD protein26,27. Another Bcl-2-binding protein, BAG-1, can also interact with and activate Raf-1. Thus, BAG-1 probably collaborates with Bcl-2, providing a mechanism for activating Raf-1 locally at mitochondrial and other Bcl-2-containing membranes in a Ras-independent manner28.

Little is known about the regions of the Bcl-2 protein needed for interaction with other proteins. However, the three-dimensional structure of Bcl-XL in its compact, non-membrane-integrated conformation reveals the presence of a hydrophobic groove on the surface of the protein, which is partly formed by the well conserved BH1, BH2 and BH3 domains2. The second amphipathic helix of dimerized Bcl-2 family proteins inserts into this groove, like a peptide ligand binding to a receptor29,30. As the BH3 domain corresponds to this inserting α-helix (ligand) but also forms part of the surface pocket (receptor), the implication is that Bcl-2 family proteins assume at least two conformations even in aqueous environments, with one dimerizing partner playing the role of the receptor and the other the ligand. Where CED-4 binds on Bcl-2/Bcl-XL is unknown, but this surface pocket is a likely candidate, given that some pro-apoptotic Bcl-2 family proteins such as Bik can displace CED-4 from Bcl-XL and yet have only the BH3 domain (thus limiting their role to that of the ligand). Such ‘BH3 only’ proteins evidently lack the hydrophobic helices required for membrane integration, suggesting that they can induce apoptosis without forming channels in membranes.

The so-called ‘BH4’ domain near the N terminus of Bcl-2 is probably not required for binding to CED-4 but is needed for interaction with Raf-1 (refs 25, 26). This domain represents the first α-helix in Bcl-XL (ref. 2) and is present in all anti-apoptotic Bcl-2 family proteins but absent from nearly all of the pro-apoptotic proteins, implying that Bcl-2/Bcl-XL evolved an additional domain that provides a means for communicating with Raf-1. The Ca2+-dependent protein phosphatase calcineurin also binds to the BH4 domain of Bcl-2 (ref. 31). Overexpression of active calcineurin induces apoptosis in a Bcl-2-suppressible manner32, suggesting a role for this phosphatase during Ca2+-mediated apoptosis. Bcl-2 causes a redistribution of calcineurin from the cytosol to intracellular membranes, preventing interaction with phosphorylated NF–AT or other substrates of calcineurin in the cytosol. How calcineurin induces apoptosis is unclear, but dephosphorylation of BAD is one possibility. The Bcl-2/calcineurin interaction may also be relevant to the inhibition by Bcl-2 of cell proliferation and of translocation of NF-AT into the nucleus33,34, given that NF-AT-inducible genes are important for proliferation in some types of cell such as lymphocytes. Although it is not fundamental to its function as a suppressor of apoptosis35, the slowing of entry of quiescent cells into the cell cycle by Bcl-2 could provide additional cytoprotection, because proliferating cells can be more vulnerable to apoptotic stimuli. With kinases and phosphatases hovering around Bcl-2, it is notable that Bcl-2 can be phosphorylated and that this post-translational modification may inactivate Bcl-2 (refs 36, 37) (Fig. 2).

Figure 2: Multifunctional Bcl-2 protein.

Left, in addition to channel activity which may directly or indirectly influence mitochondrial megapore opening or release of cytochrome c and apoptosis-inducing factor (AIF) from mitochondria, Bcl-2 or Bcl-XL also binds to several proteins that can participate in cell-death regulation. CED-4 provides a physical connection to the cell-death proteases (caspases). See text for details. Right, the topology of the human Bcl-2 protein is depicted, showing the locations of the Bcl-2 homology (BH) domains, the predicted α-helical segments, and the transmembrane (TM) anchor. Binding of Raf-1 and calcineurin is BH4-dependent (first α-helix). A long flexible loop between the first and second α-helices is required for Bcl-2 phosphorylation and may represent a negative-regulatory domain. The BH3 domain (second α-helix) plays the role of ‘ligand’ during dimerization of Bcl-2 family proteins, whereas the combination of the BH1, BH2 and BH3 domains appears to be required for forming the hydrophobic groove into which the BH3 domain inserts. The fifth and sixth α-helices are predicted to participate in channel formation, presumably by penetrating lipid bilayers.

In addition to CED-4, BAG-1, Raf-1 and calcineurin, Bcl-2 can bind to a p53-binding protein, p53-BP2, conceivably explaining how overexpression of Bcl-2 can interfere with translocation of p53 from the cytosol into the nucleus5,38. Immunoelectron microscopy indicates that Bcl-2 is associated with nuclear pore complexes39, an ideal situation for catching proteins as they cross the nuclear envelope.

The available data thus suggest that Bcl-2 has two separate functions as a channel protein and as an adaptor/docking protein (Fig. 2). Although arguments could be raised against both models on the basis that truncation mutants of Bcl-2 without their carboxy-terminal membrane-anchoring domain retain anti-apoptotic function, these mutant proteins still associate with intracellular membranes, probably by virtue of interaction with endogenous Bcl-2 family proteins. The channel and adaptor protein activities may be mutually exclusive in some circumstances, because for channel formation Bcl-2 must integrate into membranes and drastically alter its conformation, which might prevent it from interacting with CED-4 or other key proteins. Conversely, some protein interactions may be assisted by this integration, for example those of the BH4 domain with Raf-1 and calcineurin. Which of the two functions will prove to be more important? The genetic data from C. elegans indicate that CED-4 is likely to be critical for Bcl-2 function, enabling Bcl-2 to regulate the caspases. Moreover, the finding that CED-9 is not required for survival of worms that lack CED-3 or CED-4 argues that channel formation is unnecessary21. However, Bax can induce cell death by means that do not depend on caspases, apparently by triggering mitochondrial permeability transition16.

The dual function of Bcl-2 could explain how Bcl-2 can inhibit both apoptosis and (in some cases) necrosis40 by interacting with CED-4 to modulate caspases and regulating motochondrial permeability transition through its ion-channel activity, respectively. Interaction of Bcl-2 with accessory proteins such as Raf-1, calcineurin, BAG-1 and p53-BP2 will likewise find their niches as additional cell-type- or circumstance-specific embellishments on the theme of Bcl-2/Bcl-XL as adaptor/docking protein, probably serving as supporting actors in a play in which channel activity and CED-4 binding star as the principal protagonists.

Figure 3: Multifunctional Bcl-2 protein.

Left, in addition to channel activity which may directly or indirectly influence mitochondrial megapore opening or release of cytochrome c and apoptosis-inducing factor (AIF) from mitochondria, Bcl-2 or Bcl-XL also binds to several proteins that can participate in cell-death regulation. CED-4 provides a physical connection to the cell-death proteases (caspases). See text for details. Right, the topology of the human Bcl-2 protein is depicted, showing the locations of the Bcl-2 homology (BH) domains, the predicted α-helical segments, and the transmembrane (TM) anchor. Binding of Raf-1 and calcineurin is BH4-dependent (first α-helix). A long flexible loop between the first and second α-helices is required for Bcl-2 phosphorylation and may represent a negative-regulatory domain. The BH3 domain (second α-helix) plays the role of ‘ligand’ during dimerization of Bcl-2 family proteins, whereas the combination of the BH1, BH2 and BH3 domains appears to be required for forming the hydrophobic groove into which the BH3 domain inserts. The fifth and sixth α-helices are predicted to participate in channel formation, presumably by penetrating lipid bilayers.


  1. 1

    Tsujimoto, Y. & Croce, C. M. Analysis of the structure, transcripts, and protein products of bcl-2, the gene involved in human follicular lymphoma. Proc. Natl Acad. Sci. USA 83, 5214–5218 (1986).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Muchmore, S. W. et al. X-ray and NMR structure of human Bcl-xL, an inhibitor of programmed cell death. Nature 381, 335–341 (1996).

    ADS  CAS  Article  Google Scholar 

  3. 3

    Baffy, G., Miyashita, T., Williamson, J. R. & Reed, J. C. Apoptosis induced by withdrawal of interleukin-3 (IL-3) from an IL-3-dependent hematopoietic cell line is associated with repartitioning of intracellular calcium and is blocked by enforced Bcl-2 oncoprotein production. J. Biol. Chem. 268, 6511–6519 (1993).

    CAS  PubMed  Google Scholar 

  4. 4

    Lam, M. et al. Evidence that Bcl-2 represses apoptosis by regulating endoplasmic reticulum-associated Ca2+fluxes. Proc. Natl Acad. Sci. USA 91, 6569–6573 (1994).

    ADS  CAS  Article  Google Scholar 

  5. 5

    Ryan, J. J. et al. c-myc and bcl-2 modulate p53 function by altering p53 subcellular trafficking during the cell cycle. Proc. Natl Acad. Sci. USA 91, 5878–5882 (1994).

    ADS  CAS  Article  Google Scholar 

  6. 6

    Minn, A. J. et al. Bcl-xLforms an ion channel in synthetic lipid membranes. Nature 385, 353–357 (1997).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Schendel, S. L. et al. Channel formation by anti-apoptotic protein, Bcl-2. Proc. Natl Acad. Sci. USA 94, 5113–5118 (1997).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Montal, M. Protein folds in channel structure. Curr. Opin. Struct. Biol. 6, 499–510 (1996).

    CAS  Article  Google Scholar 

  9. 9

    Chen, J. et al. bcl-2 overexpression reduces apoptotic photoreceptor cell death in three different retinal degenerations. Proc. Natl Acad. Sci. USA 93, 7042–7047 (1996).

    ADS  CAS  Article  Google Scholar 

  10. 10

    Middleton, G., Nunez, G. & Davies, A. M. Bax promotes neuronal survival and antagonises the survival effects of neurotrophic factors. Development 122, 695–701 (1996).

    CAS  PubMed  Google Scholar 

  11. 11

    Kiefer, M. C. et al. Modulation of apoptosis by the widely distributed Bcl-2 homologue Bak. Nature 374, 736–739 (1995).

    ADS  CAS  Article  Google Scholar 

  12. 12

    Bernardi, P., Broekemeier, K. M. & Pfeiffer, D. R. Recent progress on regulation of the mitochondrial permeability transition pore; a cyclosporin-sensitive pore in the inner mitochondrial membrane. J. Bioenerget. Biomembr. 26, 509–517 (1994).

    CAS  Article  Google Scholar 

  13. 13

    Zoratti, M. & Szabo, I. Electrophysiology of the inner mitochondrial membrane. J. Bioenerget. Biomembr. 26, 543–553 (1996).

    Article  Google Scholar 

  14. 14

    Petit, P. X., Susin, S.-A., Zamzami, N., Mignotte, B. & Kroemer, G. Mitochondria and programmed cell death: back ot the future. FEBS Lett. 396, 7–13 (1996).

    CAS  Article  Google Scholar 

  15. 15

    Susin, S. A. et al. Bcl-2 inhibits the mitochondrial release of an apoptogenic protease. J. Exp. Med. 184, 1331–1342 (1996).

    CAS  Article  Google Scholar 

  16. 16

    Xiang, J., Chao, D. T. & Korsmeyer, S. J. BAX-induced cell death may not require interleukin 1β-converting enzyme-like proteases. Proc. Natl Acad. Sci. USA 93, 14559–14563 (1996).

    ADS  CAS  Article  Google Scholar 

  17. 17

    Kluck, R. M., Bossy-Wetzel, E., Green, D. R. & Newmeyer, D. D. The release of cytochrome c from mitochondria: a primary site for Bcl-2 regulation of apoptosis. Science 275, 1132–1136 (1997).

    CAS  Article  Google Scholar 

  18. 18

    Yang, J. et al. Prevention of apoptosis by Bcl-2: release of cytochrome c from mitochondria blocked. Science 275, 1129–1132 (1997).

    CAS  Article  Google Scholar 

  19. 19

    Häcker, G. & Vaux, D. L. Asticky business. Curr. Biol. 5, 622–624 (1995).

    Article  Google Scholar 

  20. 20

    Yuan, J. Y. & Horvitz, H. R. The Caenorhabditis elegans genes ced-3 and ced-4 act cell autonomously to cause programmed cell death. Dev. Biol. 138, 33–41 (1990).

    CAS  Article  Google Scholar 

  21. 21

    Shaham, S. & Horvitz, H. R. Developing Caenorhabditis elegans neurons may contain both cell-death protective and killer activities. Genes Dev. 10, 578–591 (1996).

    CAS  Article  Google Scholar 

  22. 22

    Spector, M. S., Desnoyers, S., Heoppner, D. J. & Hengartner, M. O. Interaction between the C. elegans cell-death regulators CED-9 and CED-4. Nature 275, 1122–1126 (1997).

    Google Scholar 

  23. 23

    Chinnaiyan, A. M., O'Rourke, K., Lane, B. R. & Dixit, V. M. Interaction of CED-4 with CED-3 and CED-9: a molecular framework for cell death. Science 275, 1122–1126 (1997).

    CAS  Article  Google Scholar 

  24. 24

    Wu, D., Wallen, H. D. & Nunez, G. Interaction and regulation of subcellular localization of CED-4 by CED-9. Science 275, 1126–1129 (1997).

    CAS  Article  Google Scholar 

  25. 25

    Shaham, S. & Horvitz, H. R. An alternatively spliced C. elegans ced-4 RNA encodes a novel cell death inhibitor. Cell 86, 201–208 (1996).

    CAS  Article  Google Scholar 

  26. 26

    Wang, H. G., Rapp, U. R. & Reed, J. C. Bcl-2 targets the protein kinase Raf-1 to mitochondria. Cell 87, 629–638 (1996).

    CAS  Article  Google Scholar 

  27. 27

    Zha, J., Harada, H., Yang, E., Jockel, J. & Korsmeyer, S. J. Serine phosphorylation of death agonist BAD in response to survival factor results in binding to 14-3–3 not BCL-XL. Cell 87, 619–628 (1996).

    CAS  Article  Google Scholar 

  28. 28

    Wang, H.-G., Takayama, S., Rapp, U. R. & Reed, J. C. Bcl-2 interacting protein, BAG-1, binds to and activates the kinase Raf-1. Proc. Natl Acad. Sci. USA 93, 7063–7068 (1996).

    ADS  CAS  Article  Google Scholar 

  29. 29

    Sattler, M. et al. Structure of Bcl-xL–Bak peptide complex: recognition between regulators of apoptosis. Science 275, 983–986 (1997).

    CAS  Article  Google Scholar 

  30. 30

    Diaz, J.-L. et al. Acommon binding site mediates heterodimerization and homodimerization of Bcl-2 family members. J. Biol. Chem. 272, 11350–11355 (1997).

    CAS  Article  Google Scholar 

  31. 31

    Shibasaki, F., Kondo, E., Akagi, T. & McKeon, F. Suppression of signalling through NF-AT by interactions between calcineurin and Bcl-2. Nature 386, 728–731 (1997).

    ADS  CAS  Article  Google Scholar 

  32. 32

    Shibasaki, F. & McKeon, F. Calcineurin functions in Ca2+-activated cell death in mammalian cells. J. Cell Biol. 131, 735–743 (1995).

    CAS  Article  Google Scholar 

  33. 33

    Pietenpol, J. A. et al. Paradoxical inhibition of solid tumor cell growth by bcl-2. Cancer Res. 54, 3714–3717 (1994).

    CAS  PubMed  Google Scholar 

  34. 34

    Linette, G. P., Li, Y., Roth, K. & Korsmeyer, S. J. Cross talk between cel death and cell cycle progression: BCL-2 regulates NFAT-medicated activation. Proc. Natl Acad. Sci. USA 93, 9545–9552 (1996).

    ADS  CAS  Article  Google Scholar 

  35. 35

    Huang, D. C. S., O'Reilly, L. A., Strasser, & Cory, S. The anti-apoptosis function of Bcl-2 can be genetically separated from its inhibitory effect on cell cycle entry. EMBO J.(in the press).

  36. 36

    Haldar, S., Jena, N. & Croce, C. M. Inactivation of Bcl-2 by phosphorylation. Proc. Natl Acad. Sci. USA 92, 4507–4511 (1995).

    ADS  CAS  Article  Google Scholar 

  37. 37

    Chang, B. S., Minn, A. J., Muchmore, S. W., Fesik, S. W. & Thompson, C. B. Identification of a novel regulatory domain in Bcl-xLand Bcl-2. EMBO J. 16, 968–977 (1997).

    CAS  Article  Google Scholar 

  38. 38

    Naumovski, L. & Cleary, M. L. The p53-binding protein 53BP2 also interacts with Bcl-2 and impedes cell cycle progression at G2/M. Mol. Cell. Biol. 16, 3884–3892 (1996).

    CAS  Article  Google Scholar 

  39. 39

    Krajewski, S. et al. Investigations of the subcellular distribution of the bcl-2 oncoprotein: residence in the nuclear envelope, endoplasmic reticulum, and outer mitochondrial membranes. Cancer Res. 53, 4701–4714 (1993).

    CAS  PubMed  Google Scholar 

  40. 40

    Kane, D. J., örd, T., Anton, R. & Bredesen, D. E. Expression of Bcl-2 inhibits necrotic neural cell death. J. Neurosci. Res. 40, 269–275 (1995).

    CAS  Article  Google Scholar 

Download references


I thank M. Montal, S. Schendel and S. Choe for discussions on channel proteins, and D. Bredesen and G. Salvesen for critically reviewing the manuscript.

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Reed, J. Double identity for proteins of the Bcl-2 family. Nature 387, 773–776 (1997).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing