1C)

1C). remains essentially cytosolic, or weakly associated to mitochondria1(2, 3for reviews). Following an apoptotic stimulus, Bax is relocated to the mitochondria, and most specifically to the outer mitochondrial membrane. Bax can then be organized as dimers and oligomers that form a large-sized pore in this membrane. This pore favors the release of proteins together known as ‘apoptogenic factors’ that are released from the mitochondrial intermembrane space to the cytosol, and that confer their apoptotic characteristics to the cells. In addition to TTP-22 extended biochemical evidence, this model of a large sized pore formed with Bax molecules is also widely supported by electrophysiology4, 5, 6, structural studies7, 8, biophysical approaches9and imaging data10, 11. As a central event in the apoptotic process, the translocation of Bax from the cytosol to the mitochondria has been the focus of a large number of studies, but is still not completely resolved. When located in the cytosol, Bax is a globular and mostly hydrophilic protein12. Bax can be activated through interacting with the BH3-domain of BH3-only proteins, such as tBid, Bim or Puma, that promotes major conformational changes to the protein13. Structural studies suggested that the interaction between Bax and a BH3-domain induced the formation of a head-to-tail dimer7, that is able to lay flat on the mitochondrial membrane8. The association of several dimers is thought to form the oligomer that constitutes the pore9. This hypothesis has been supported by microscopy experiments showing the formation of a large-sized pore both in membranes and in mitochondriain situ10, 11. However , this widely accepted model still contains a number of gray areas, including the TTP-22 role of the very hydrophobic C-terminal helix 9, that was absent from the structural data of the Bax dimer7, and of which the actual role in Bax interaction with mitochondria remains unclear: indeed, its absence does not prevent the mitochondrial localization of Bax, nor Bax-induced outer membrane permeabilization14, 15. One intriguing issue is the role of the serine residue in position 184 (S184). It is one of the few polar residues in this otherwise hydrophobic -helix. The deletion of the S184 (S184) converts 9 into a bona fide membrane anchor, that is able to drive the constitutive mitochondrial localization of Bax16. Furthermore, the S184 mutation prevented the regulation of mitochondrial Bax translocation by components of the TOM complex17. It has been established that S184 can be phosphorylated by different kinases, such as AKT18, 19and PKC20. This phosphorylation was shown TTP-22 to impair the mitochondrial relocation of Bax during apoptosis, that is consistent with the pro-survival function of AKT, including in cancer cells21. NPM1 However , AKT has multiple cellular targets, and it is therefore difficult to identify precisely the actual role of Bax phosphorylation in the survival effects induced by TTP-22 AKT activation. In recent experiments, we have co-expressed human Bax and AKT in yeast. Rather unexpectedly, we found that AKT increased both cellular and mitochondrial Bax content, and consequently increased the capacity of Bax to promote the release of cytochromec22. Also, substituted Bax mutants where S184 was replaced by non phosphorylatable Ala or Val residues, or by a phospho-mimetic Asp residue have been tested for their ability to interact with mitochondria22, 23. Like wild-type Bax, the phosphomimetic mutant S184D had the same weak mitochondrial localization that had already been observed in mammalian cells18. However , this weak mitochondrial localization was paradoxically associated to a high capacity to release cytochromec, suggesting that the mutation converted Bax into its active conformation22, 23. Conversely, the non-phosphorylatable mutants S184A and S184V had a much higher mitochondrial localization. Because of this, they induced a stronger release of cytochromecthan wild-type Bax. However , in spite of their much higher mitochondrial localization, they remained less active than the mutant S184D, suggesting that they adopted a poorly active conformation22, 23. Bax translocation to mitochondria also depends on other proteins, such as the anti-apoptotic protein.