Contribution of Yeast Studies to the Understanding of BCL-2 Family Intracellular Trafficking

Abstract

BCL-2 family members are major regulators of apoptotic cell death in mammals. They form an intricate regulatory network that ultimately regulates the release of apoptogenic factors from mitochondria to the cytosol. The ectopic expression of mammalian BCL-2 family members in the yeast Saccharomyces cerevisiae, which lacks BCL-2 homologs, has been long established as a useful addition to the available models to study their function and regulation. In yeast, individual proteins can be studied independently from the whole interaction network, thus providing insight into the molecular mechanisms underlying their function in a living context. Furthermore, one can take advantage of the powerful tools available in yeast to probe intracellular trafficking processes such as mitochondrial sorting and interactions/exchanges between mitochondria and other compartments, such as the endoplasmic reticulum that are largely conserved between yeast and mammals. Yeast molecular genetics thus allows the investigation of the role of these processes on the dynamic equilibrium of BCL-2 family members between mitochondria and extramitochondrial compartments. Here we propose a model of dynamic regulation of BCL-2 family member localization, based on available evidence from ectopic expression in yeast.

Keywords: Bcl-2 family; Mitochondria-Associated Membranes; Translocase of Outer Membrane; apoptosis; endoplasmic reticulum; mitochondria.

Figure 1

Schematic representation of the differential effect of BCL-xL on BAX dynamics (reproduced from [38]). Left: BAX is under a dynamic equilibrium between mitochondria and cytosol, owing to the processes of BCL-xL-driven translocation and retrotranslocation. Mitochondrial BAX is primed but remains inactive because of the interaction with BCL-xL. The addition of a BH3-mimetic (or the expression of a BH3-only partner) enables the derepression of the interaction and the full activation of BAX. Right: BCL-xLΔC stimulates BAX translocation but not BAX retrotranslocation, leading to a higher mitochondrial BAX content. Furthermore, the interaction between BAX and BCL-xLΔC is weaker than with BCL-xL, and BAX can therefore be activated in the absence of BH3-only protein or BH3 mimetics.

Figure 2

Stimulation of BAX mitochondrial localization and activation in Sch9-deleted yeast. Different BAX mutants were expressed in the wild-type strain W303 or in a mutant carrying a deletion of the gene of the AKT-ortholog Sch9. BAX content was measured in isolated mitochondria with OMM protein VDAC (Voltage Dependent Anion Channel) as a loading control. As previously reported, the cellular content of the three mutants is about the same but they exhibit different mitochondrial localization and level of activity (see Figures 1 and 5 in ref. [81]). The mutant P168A has both a high activity and high mitochondrial localization, and the mutants P168A/S60A and S60D have a weak activity with a weak and intermediate mitochondrial localization, respectively [81]. The P168 residue is in the short loop between α8 and α9 helices and its substitution was shown to stimulate BAX in mammalian cells [89], yeast cells [90], and in vitro [91] by favoring the movement of the α9 helix. The S60 residue is part of a consensus sequence KKLS that might be targeted by a Protein Kinase A, and is in the α2 helix in close proximity to a negatively charged residue D33 located in the α1 helix (see ref. [81] for details). (A) Western-Blots showing the presence of BAX in isolated mitochondria. (B) Redox spectrophotometry measurements of Cyt.c and Cyt.b in mitochondria. Mitochondrial cytochromes absorb light differently in their oxidized and reduced states. The maximal difference between chemically reduced (with Sodium Dithionite) and oxidized (with Potassium Ferricyanide) cytochromes is at 550 nm for cytochrome c and 561nm for cytochrome b with isosbestic points at 540 and 575 nm., with Δε (reduced minus oxidized) = 18,000 M⁻¹·cm⁻¹ for both cytochromes, allowing the precise calculation of their remaining concentration in mitochondria. Cytochrome b is part of the inner membrane respiratory Complex III and is not released during OMM permeabilization, thus serving as an internal control. The decrease of the ratio mitochondrial Cyt.c/Cyt.b ratio reflects the capacity of BAX to release Cyt.c [15,81].

Figure 3

Effect of substitutions of potentially phosphorylatable residues on BAX activity. Different mutants of BAX carrying substitutions of potentially phosphorylatable residues have been expressed in yeast [80]. Potentially phosphorylatable residues were selected based on a mass spectrometry analysis of purified BAX incubated in the presence of AKT or GSK3β, or known data from the literature. (A) After mitochondria isolation, the mitochondrial ratio Cyt.c+c1/Cyt.b was measured by redox spectrophotometry [15]. The blue dashed line indicates the ratio measured on wild-type BAX. The red dashed line indicates the average value measured in W303 yeast strain not expressing BAX. The green dashed line indicated the theoretical ratio value if all cytochrome was released (corresponding to the ratio Cyt.c1/Cyt.b of 0.5 in Respiratory Complex III). Some mutants are significantly more active than wild-type BAX, such as S184D that has been characterized extensively [81,82], but some mutants also appeared to be less active than wild-type BAX. The level of expression of each BAX mutant might be different (B), but there is no obvious correlation between the content and the capacity to release cytochrome c: for example, the mutant BAX-S16D (inactive) is expressed at about two-times the level of BAX-WT, the mutant BAX-S135A (moderately active) is expressed at about half the level of BAX-WT and the mutant BAX-S184D (very active) is expressed at about the same level as BAX-WT.

Figure 4

Schematic representation of the link between BAX phosphorylation and Translocation. Phosphorylated BAX (by AKT and other survival kinases) is maintained away from mitochondria while non-phosphorylated BAX is in the OMM, both in yeast and mammalian cells [80,83]. The interaction of BAX with BCL-xL stimulate both BAX translocation [15] and BAX retrotranslocation [13,15]. Heterodimers between phosphomimetic BAX and BCL-xL are more present away from mitochondria while heterodimers between non-phosphorylatable BAX and BCL-xL are more present at the OMM [68]. The release of the interaction between mitochondrial BAX and BCL-xL by BH3-only proteins (or BH3 mimetics) triggers the oligomerization/activation process of BAX [15].

Figure 5

Schematic representation of the involvement of Mitochondria-ER contact sites (MERC) in BAX dynamics. Translocation and retrotranslocation processes are likely favored at MERC. The forced destabilization of the tethering complex ERMES in yeast led to a lower capacity of BAX to localize to the OMM and to promote the release of cytochrome c [24]. Components of the TOM complex Tom20 and Tom22 help the localization and/or the adequate conformation of BCL-2 [25,26] and BAX [27,28,29], respectively. Derepression by BH3-only proteins or BH3 mimetics allows the release of BAX under its active conformation able to form the pore [15,115].