Embedded together: the life and death consequences of interaction of the Bcl-2 family with membranes

Abstract

Permeabilization of the outer mitochondrial membrane is the point of no return in most programmed cell deaths. This critical step is mainly regulated by the various protein-protein and protein-membrane interactions of the Bcl-2 family proteins. The two main models for regulation of mitochondrial outer membrane permeabilization, direct activation and displacement do not account for all of the experimental data and both largely neglect the importance of the membrane. We propose the embedding together model to emphasize the critical importance of Bcl-2 family protein interactions with and within membranes. The embedding together model proposes that both pro- and anti-apoptotic Bcl-2 family proteins engage in similar dynamic interactions that are governed by membrane dependent conformational changes and culminate in either aborted or productive membrane permeabilization depending on the final oligomeric state of pro-apoptotic Bax and/or Bak.

Figure 1

The Direct Activation Model An activator protein (black) such as tBid that contains a BH3 domain can bind to and activate Bax (powder blue). Activation of Bax leads to integration in the MOM and oligomerization. Oligomerization of Bax may displace the activator. Alternatively, one activated Bax may recruit additional Bax and therefore, there need be only one activator for each oligomer. Bak is predicted to function similarly but begins bound to the surface of MOM. Membrane permeabilization is inhibited by anti-apoptosis proteins (such as Bcl-XL shown) binding to the activator protein. Once bound to Bcl-XL the activator protein can no longer bind to or activate Bax. Sensitizer proteins compete with activator proteins for binding to Bcl-XL. Therefore, if the affinity for Bcl-XL or abundance of the sensitizer is greater than the activator it will displace it from Bcl-XL whereupon the activator can again bind to Bax.

Figure 2

The Displacement Model Anti-apoptosis proteins such as Bcl-XL, Mcl-1 and Bcl-w bind to and thereby inhibit active forms of Bax/Bak (for clarity only Bax is shown, in this model Bax and Bak are assumed to function similarly). BH3 only proteins initiate membrane permeabilization by displacing active Bax from the anti-apoptosis proteins by direct binding. The active Bax then inserts into the membrane and oligomerizes (the order is not specified by the model and was selected arbitrarily). Binding of BH3 proteins inhibit anti-apoptosis proteins by various mechanisms: Bcl-XL by sequestration, Bcl-w by embedding it in the membrane and Mcl-1 by initiating degradation. For simplicity each BH3 protein is shown binding a single anti-apoptosis protein however, the model proposes hierarchies of interaction based on the relative affinities of the interactions. Moreover, an active Bax displaced from Bcl-XL may be bound by Mcl-1 or some other anti-apoptosis protein. Therefore, initiating membrane permeabilization requires inactivating all of the relevant anti-apoptosis proteins. For simplicity the legend includes only one example of inhibition of Bax.

Figure 3

The Embedding Together Model - Initiation of membrane permeabilization. Bax is most frequently an inactive cytoplasmic protein. The binding affinity for cytoplasmic Bax for cytoplasmic BH3 proteins (tBid is shown) is low and cytoplasmic Bax is in equilibrium with form that binds with relatively low affinity to the surface of membranes. Therefore, other relatively low affinity interactions that might not survive immunoprecipitation can inhibit Bax in the cytoplasm by shifting the equilibrium away from membrane binding. Bax binds to membranes peripherally concomitant with a conformational change in Bax that increases the affinity of Bax/tBid interaction. In the diagram tBid is the only activator shown however, the model presupposes BH3 only proteins and other proteins that interact with Bax both in the cytoplasm and on the membrane can modulate one or more of the equilibriums shown. In the case of tBid the affinity for membranes is high and binding to membranes results in a conformation change that increases tBid affinity for Bax. The interaction between membrane bound tBid with peripherally membrane bound Bax triggers insertion of Bax into the membrane. The integral membrane form of Bax can recruit additional Bax proteins and oligomerize to permeabilize the outer mitochondrial membrane. The only step that is effectively irreversible is integration into the membrane. The model does not require that any one pro-apoptotic protein perform all of the functions ascribed to tBid, but rather envisions that most will perform only a subset of them. Bak is proposed to function similarly to Bax except that it is expected to bind the different modulators (e.g. tBid) with different affinities, its inactive conformation is constitutively membrane bound and is thus envisioned to be functionally similar to the peripheral membrane form of Bax.

Figure 4

The Embedding Together Model - Inhibition of membrane permeabilization. Bcl-XL inhibits both tBid and Bax. Although it may bind to both, the affinity of Bcl-XL for membrane bound tBid is higher than for cytoplasmic tBid. Binding of Bcl-XL to tBid tiggers insertion of Bcl-XL into membranes. Once inserted in membranes Bcl-XL recruits other Bcl-XL proteins to insert into membranes. Bcl-XL/tBid complexes are dead-end structures that neither prevent nor promote apoptosis. Membrane bound Bcl-XL prevents Bax inserting into the membrane by preventing the conformation change in Bax that occurs at the membrane. Membrane bound Bcl-XL also binds to and inhibits the integral membrane (oligomerization competent) form of Bax. The only step that is effectively irreversible for Bcl-XL is integration into the membrane. The model does not require that any one anti-apoptotic protein perform all of the functions ascribed to Bcl-XL, but rather envisions that most will perform only a subset of them. Bcl-2 is proposed to function similarly to Bcl-XL except that it is expected to bind the different modulators (e.g. tBid/Bax) with different affinities. As Bcl-XL competes with Bax/Bak in all steps it has no inactive conformation. Moreover, pro-apoptotic and anti-apoptotic multi-region Bcl-2 family proteins are envisioned as functioning similarly (competitively) except that the anti-apoptotic proteins are unable to support oligomerization and therefore, function as defective constituents of Bax/Bak oligomers inhibiting full oligomerization and membrane permeabilization.