Surviving the Cytochrome Seas

were an order of magnitude higher than those required In the molecular drama of apoptotic cell death, perhaps to activate caspases in cell extracts (0.2–0.5 mM). Intact there is no character stranger than cytochrome c. This cells thus seem to have a way to regulate or suppress the familiar heme protein can be transformed, in Jekyll and response to cytosolic cytochrome c. When apoptosis is Hyde fashion, from life-sustainer into life-destroyer. Norinduced, the inhibitory effect is removed; this suppresmally, cytochrome c is restricted to the intermembrane sion is also apparently lost during the preparation of space of mitochondria, where it plays a well-known role cell extracts. Are the cell-free systems merely artifactual in the electron transport chain. In many cases, though, in this regard, or is there a hint of something still to be early in apoptosis, an ill-defined mitochondrial lesion learned here about apoptotic regulation in vivo? allows this protein to escape into the cytosol (Liu et al., This question has now been brought sharply into fo1996; Bossy-Wetzel et al., 1998; Kim et al., 1997; Kluck cus in a new study by Deshmukh and Johnson (1998 [this et al., 1997; Vander Heiden et al., 1997; Yang et al., 1997; issue of Neuron]) regarding the effects of cytochrome c Pastorino et al., 1998). What it does there, as suggested microinjection into primary sympathetic neurons. These by biochemical experiments (Liu et al., 1996; Li et al., cells require the continuous presence of nerve growth 1997b), is to interact with the cytosolic factors Apaf-1 factor (NGF) for survival; after the withdrawal of trophic and procaspase-9, leading to the aggregation and transfactor, they die by apoptosis within z24 hr. Expression activation of the latter. The active caspase-9, in turn, of the BAX protein is required for cell death following processes the “executioner” caspases (proteases imNGF deprivation, because neurons from BAX-deficient portant for the execution phase of apoptosis). This mice survive for at least several weeks in the absence “killer” role for cytochrome c is further supported by the of the growth factor. Cell death also requires protein phenotype of mice nullizygous for caspase-9 (Hakem et synthesis, as treatment with cycloheximide (CHX) resal., 1998; Kuida et al., 1998) or Apaf-1 (Cecconi et al., cues the cells from death following NGF withdrawal. 1998; Yoshida et al., 1998). These mice are defective in Since BAX expression is required for cell death following many (but not all) forms of apoptosis. The effects of NGF withdrawal, the effect of CHX may be at the level deficiency in caspase-9 (Hakem et al., 1998; Kuida et of expression of this protein. al., 1998) or the downstream target caspase-3 are severe In this new study, sympathetic neurons maintained in in the developing brain, which literally bursts through NGF were found to be resistant to killing by microinthe skull due to excess neuronal mass. In animals lackjected cytochrome c. On the other hand, neurons deing Apaf-1, there is a delay in the loss of the interdigital prived of NGF, but kept alive either because they lacked webs, hyperplasia of neurons and retinal cells, defects BAX or were treated with CHX, failed to release cytoin the formation of the lens and palate, and many other chrome c from their mitochondria but did die in response developmental consequences (Cecconi et al., 1998; to cytochrome c injection. Deshmukh and Johnson conYoshida et al., 1998). Unless the Apaf-1/caspase-9 pathclude that there are two processes contributing to the way can somehow be activated independently of cytoapoptotic death of sympathetic neurons. One of these chrome c, these results argue for a broad (although not is the release of cytochrome c from mitochondria, and universal) role of cytochrome c in apoptotic cell death. the other is the development of what they call “compeThe exquisite sensitivity of cell extracts to cytochrome tence-to-die,” defined as the ability of the cell to undergo c has provoked a simple hypothesis that all the cell apoptosis in response to the presence of extramitoneeds to do to regulate apoptosis is to control the efflux chondrial cytochrome c. NGF signals a block to both of of cytochrome c from its mitochondria. In support of these processes. When NGF is withdrawn, however, the this idea, studies both in intact cells and in cell-free two processes occur coordinately and may be regulated systems showed that the BCL-2 and BCL-XL proteins independently (see Figure 1). prevent cell death, at least in part, by preventing the What does competence-to-die entail, in molecular translocation of cytochrome c (Kim et al., 1997; Kluck terms? We don’t really know. One could guess that comet al., 1997; Vander Heiden et al., 1997; Yang et al., 1997; petence-to-die reflects the intracellular accumulation of Srinivasan et al., 1998). BAX, a pro-apoptotic counterdATP, a cofactor of Apaf-1. However, microinjection of part of BCL-2 and BCL-XL, may lead to cell death through dATP along with cytochrome c did not confer compethe opposite activity of promoting cytochrome c release tence-to-die on neurons cultured in NGF. Another possi(Juergensmeier et al., 1998; Pastorino et al., 1998). bility is that NGF signals the posttranslational modificaHowever, the full story is not so simple, as experition of one or more of the components of the Apaf-1/ ments in which cytochrome c was microinjected into caspase-9 pathway, at or downstream from the point