The “point of no return” in apoptotic cell death is the rupture of the mitochondrial outer membrane leading to the release of apoptogenic factors such as cytochrome c. This crucial pore formation step requires either of two pro-apoptotic Bcl-2 relatives, Bak and Bax. Their ability to change conformation and oligomerise in the mitochondrial outer membrane appears to be key to understanding how they become active and perforate this membrane. Bak and Bax can be triggered to convert to the activated conformation by the “activator” BH3-only relatives (e.g. Bid and Bim), and blocked by the pro-survival relatives (e.g. Bcl-2 and Mcl-1). The core regions (BH3 domain and hydrophobic groove) of Bak and Bax allow the activated proteins to form homodimers, which then must coalesce to the high order oligomers needed for pore formation¹²³. The “symmetric BH3:groove dimer” model of Bak oligomerisation was recently verified by the first structure of an activated form of Bak or Bax⁴ .

To gain an overall view of Bak topology in the mitochondrial outer membrane before and after an apoptotic stimulus, we have employed site-directed cysteine mutagenesis and disulphide chemistry to the Bak N- and C-termini, in addition to the core regions.  Applying Blue Native PAGE and oxidant-induced disulphide linkage, we reveal an updated picture of Bak-to-Bak interfaces in oligomers present in mitochondrial fractions. These data support a “flexible extremity/constrained core” model of Bak oligomerisation in which the Bak homodimer consists of unstructured N- and C- termini and a structured BH3:groove core.