Apoptosis Regulation by BCL2 Family

The BCL2 family regulates apoptosis via protein-protein interactions between the anti- and pro-apoptotic proteins. Previously, BH3-only proteins were generally assumed to bind indiscriminately to all anti-apoptotic proteins. Recently, however,a comprehensive in vitro analysis using recombinant anti-apoptotic proteins and synthetic peptides of BH3 domains documents marked differences that could exist in the binding affinity between pro- and anti-apoptotic proteins. Among the BH3 peptides derived from human BAD, BID, BIK,BIM, HRK, NOXA, PUMA, and mouse BMF proteins, only BIM, PUMA, and tBID peptides have comparable strong affinities to all the anti-apoptotic proteins tested.BAD and BMF peptides preferably binds to BCL2, BCL-XL, and BCL-W; NOXA peptide binds strongly to MCL1 and BFL1; and BIK and HRK peptides interact more strongly with BCL-XL, BCL-W, and BFL1.

Despite the reported affinity of NOXA for MCL1 as opposed to BCL-XL, the sequestration of NOXA by BCL-XL but not MCL1 has recently been shown to dictate sensitivity to chemotherapeutic agents and DNA damage. Although these studies do not dispute the theory that BCL2 family members bind to each other with various affinities, they do suggest that binding affinities determined with BH3 peptides in an in vitro environment may not fully reflect the binding affinities found in cells.

The variable affinities, combined with the spatial and temporal variations of these proteins within a cell, tightly regulate apoptosis and give rise to two models to explain how BH3-only molecules trigger the activation of BAX and BAK to induce mitochondrial outer membrane permeabilization: the direct and indirect activation models.


Direct Activation Model

The direct model purposes that there are two types of BH3-only proteins: direct activators (principally BIM and tBID) and sensitizers (the remaining BH3-only proteins, including BAD and NOXA). Direct activators can interact with BAX/BAK to induce their oligomerization, whereas sensitizers cannot. Hence, in a viable cell, activators can either be sequestered away from the mitochondria or neutralized by the anti-apoptotic proteins at the mitochondria. Upon an apoptotic stimulus, sensitizer BH3-only proteins are mobilized to inhibit the anti-apoptotic proteins, thereby displacing the activators that were originally sequestered by the anti-apoptotic proteins. The apoptotic stimulus can also mobilize activators from their subcellular compartment. Released activators can then interact with BAX/BAK to induce their oligomerization and the onset of apoptosis. For example, in a viable cell, the activator BIM can either be sequestered by an anti-apoptotic protein or be localized to the microtubular network. Upon apoptotic stimulation, BIM can be displaced from its anti-apoptotic partner by the sensitizer BAD and/or released from the dynein motor. Free BIM can then translocate to the mitochondria and induce BAX/BAK oligomerization.

Mutagenesis studies that used differential binding mutants of BH3-only proteins defective in binding either the anti-apoptotic proteins or BAX/BAK provide support for the direct activation model. For example, a BID BH3 mutant incapable of binding anti-apoptotic proteins activates BAX to release cytochrome c and induce apoptosis,suggesting the induction of apoptosis by BID does not require the neutralization of anti-apoptotic proteins BCL2, BCL-XL and MCL1.

A major concern for the direct activation model is that endogenous interactions between BAX/BAK and BIM or BID have never been confirmed in response to an apoptotic stimulation. The “hit and run” theory was thus proposed,suggesting that a “hit” occurs when an activator binds to BAX and displaces the BAX transmembrane domain. Subsequent BAX conformational change may then disrupt the hydrophobic groove to immediately displace the activator, which defines the “run”. Additionally, mutations in BIMS and tBID that prevent their binding to BAX but not to anti-apoptotic proteins kill cells as potently as wild-type tBID and BIMS. Apoptosis is also not abrogated in cells lacking both BIM and BID in response to several apoptotic stimuli. These findings indicate that the presumed direct BAX/BAK activation by BIM or tBID may not be essential for the onset of apoptosis.


Indirect Activation Model

The indirect activation model suggests that all the BH3-only proteins interact only with their anti-apoptotic partners, negating the requirement for any interaction between the BH3-only proteins and BAX/BAK. The primary role of the BH3-only proteins is thus to prevent the anti-apoptotic proteins from antagonizing BAX/BAK activation. BIM, tBID, and PUMA are considered “potent inducers” of apoptosis because they can engage all the anti-apoptotic proteins. The remaining BH3-only proteins, which bind to a subset of the anti-apoptotic proteins, are thus termed “poor inducers” of apoptosis. Commitment to apoptosis is ensured when all anti-apoptotic proteins are neutralized by BH3-only molecules.

Biochemical and functional studies have provided findings consistent with the indirect activation model. In the absence of BIM and tBID, the co-expression of BAD and NOXA, a combination that neutralizes all anti-apoptotic proteins expressed in mouse embryonic fibroblasts (MEFs), induced cytochrome c release and apoptosis. Furthermore, when PUMA expression was knocked down with RNA interference technology in bim–/–bid–/– MEFs, the inhibition of all anti-apoptotic members still efficiently induced cytochrome c release and apoptosis. Lastly, BAX and BAK were shown to be held in check by anti-apoptotic members. These findings suggest that BH3-only proteins can initiate apoptosis without binding to BAX/BAK and that they function primarily to disrupt the binding between anti-apoptotic proteins and BAX/BAK.

A problematic aspect of the indirect activation model is that only a fraction of cellular BAX and BAK are bound to anti-apoptotic proteins in viable cells,suggesting there is a pool of inactive BAX and BAK that is not sequestered by the anti-apoptotic proteins. In fact, BAX and BAK are known to associate with proteins outside the BCL2 family, and it is currently unclear how BH3-only proteins affect these interactions. For example, BAX and BAK can interact with the voltage-dependent anion channel (VDAC) at the mitochondria to regulate the release of cytochrome c independent of BCL2/BCL-XL. BAX and BAK have also been found to co-localize with dynamin-related protein 1 and mitofusin 2 to regulate mitochondrial fission, a process that can be uncoupled from its role in cytochrome c release. Lastly, even if all the BAX and BAK in a viable cell is sequestered by the anti-apoptotic proteins, the indirect model does not address how BAX/BAK is activated once the BH3-only proteins displace them from the anti-apoptotic proteins.

Despite the contradictory data supporting either of the two models, the indirect and direct activation models may not be mutually exclusive and may share points of convergence. For example, BAX and BAK exist as two conformers within cells: (1) with their BH3 domain exposed (active) and (2) with their BH3 domain hidden (inactive). The anti-apoptotic proteins bind to the active conformer to inhibit its pro-apoptotic activity, since the exposed BH3 domain has been shown to be required for this interaction. There could thus be a pool of inactive BAX/BAK that does not need to be neutralized by the anti-apoptotic proteins. Currently, the apoptotic signal that directly induces the shift between the inactive and active BAX/BAK conformers has not been defined. One possible mechanism for this inducing event could be the activator BH3-only proteins, as suggested by the direct activation model.



1. Chen L, Willis S N, Wei A, et al. Differential targeting of prosurvival Bcl-2 proteins by their BH3-only ligands allows complementary apoptotic function[J]. Molecular cell, 2005, 17(3): 393-403.

2. Hagenbuchner J, Ausserlechner M J, Porto V, et al. The anti-apoptotic protein BCL2L1/Bcl-xL is neutralized by pro-apoptotic PMAIP1/Noxa in neuroblastoma, thereby determining bortezomib sensitivity independent of prosurvival MCL1 expression[J]. Journal of Biological Chemistry, 2010, 285(10): 6904-6912.

3. Gavathiotis E, Suzuki M, Davis M L, et al. BAX activation is initiated at a novel interaction site[J]. Nature, 2008, 455(7216): 1076-1081.

4. Kim H, Rafiuddin-Shah M, Tu H C, et al. Hierarchical regulation of mitochondrion-dependent apoptosis by BCL-2 subfamilies[J]. Nature cell biology, 2006, 8(12): 1348-1358.

5. Willis S N, Fletcher J I, Kaufmann T, et al. Apoptosis initiated when BH3 ligands engage multiple Bcl-2 homologs, not Bax or Bak[J]. Science, 2007, 315(5813): 856-859.

6. Sheridan C, Delivani P, Cullen S P, et al. Bax-or Bak-induced mitochondrial fission can be uncoupled from cytochrome C release[J]. Molecular cell, 2008, 31(4): 570-585.

7. Adams J M, Cory S. The Bcl-2 apoptotic switch in cancer development and therapy[J]. Oncogene, 2007, 26(9): 1324-1337.