News & Views | Published:

Cell biology

Divide and conquer

The discovery that cell death in nematode worms induces fragmentation of mitochondria reveals a new parallel to the death process in mammals, and may shed light on why mitochondria divide in death.

When mammalian cells die by the process of apoptosis, their mitochondria fragment into smaller pieces. Why these power-generating compartments should divide as the cell around them dies, and whether this fragmentation is important for the death process or simply an epiphenomenon, has so far largely remained unclear. But an answer is suggested by the paper from Conradt and colleagues on page 754 of this issue1. The authors show that mitochondria also fragment during apoptosis in the small nematode worm Caenorhabditis elegans. Moreover, experimental induction or prevention of mitochondrial fragmentation could respectively enhance or partially prevent apoptosis. These observations hint that mitochondrial fragmentation has an evolutionarily conserved, causative role in promoting apoptotic cell death.

The term apoptosis refers to a specific type of programmed cell death that occurs in all multicellular animals, from the lowly worm to the highly complex human. Apoptosis is characterized by specific morphological changes in the dying cell, and is mediated by several protein families2.

In mammals, most apoptotic cell deaths are mediated by a specific signalling pathway known as the mitochondrial pathway. As its name implies, this pathway requires the active participation of mitochondria — the organelles better known for their role in cellular respiration and the generation of the high-energy molecule ATP. In cells condemned to die, mitochondria release several dozen proteins into the cytosol, and they can then wreak havoc in the rest of the cell.

The best known of these mitochondrial expatriates — cytochrome c — interacts in the cytosol with the Apaf-1 protein, ultimately activating a group of proteases (protein-digesting enzymes) known as caspases3. These enzymes then cleave a selected set of target proteins, resulting in the controlled ‘implosion’ of the cell. How cytochrome c et al. manage to cross the outer lipid bilayer of the mitochondria to reach the cytosol is still hotly debated. What is clear, however, is that this release is regulated by proteins of the Bcl-2 family, many of which can bind directly to the outer mitochondrial membrane.

Recently, several groups have reported a second peculiar behaviour of mitochondria during apoptosis: not only do they release proteins, but they also fragment into smaller pieces4. That mitochondria can fragment is nothing new in itself — like bacteria, mitochondria divide by a process of fission, in which one long organelle is pinched in the middle to produce two shorter daughters. Unlike bacteria, mitochondria can also undergo the reverse process, and fuse together to form long filaments. Fission and fusion are tightly controlled, and are important for the proper distribution of mitochondria during cell division.

But why mitochondria should fragment during apoptosis is not clear. One possibility is that the release of mitochondrial proteins stimulates mitochondrial division. Indeed, conditions that compromise mitochondrial function have been reported to result in short, round mitochondria. An attractive alternative would be that fission is necessary (directly or indirectly) for the release of cytochrome c. Consistent with this idea, interfering with the fission process has been reported to delay cytochrome c release during apoptosis5. Whether this is a general phenomenon remains to be seen. Furthermore, given that mitochondrial fission occurs continuously in living cells, there must be more to the story than fission simply promoting death.

Enter C. elegans. Genetic studies in this species showed that most components of the apoptotic pathway have been conserved throughout evolution6. For example, C. elegans has a Bcl-2 counterpart (CED-9), an Apaf-1-like molecule (CED-4) and a caspase (CED-3). Surprisingly, however, mitochondrial proteins have so far played at best a minor role in the apoptosis saga in C. elegans. Evidence for the release of mitochondrial proteins has been rather limited7, and there are no data to suggest that activation of CED-3 requires cytochrome c. Rather, biochemical experiments led to the development of a model8 in which CED-9 inhibits apoptosis through direct interaction with and sequestration of CED-4. Although attractive in its simplicity, this model is still incomplete. It does not explain, for example, why CED-9 would need to be localized to mitochondria in order to function.

Conradt and colleagues1 now address this problem. Using live microscopy, the authors noticed that cells undergoing apoptosis in C. elegans embryos also showed fragmented mitochondria. Fragmentation was clearly caspase-independent (it still occurred in animals with mutant, inactive CED-4 or CED-3), indicating that it was a very early event in the apoptotic programme. Furthermore, fragmentation was abrogated in animals lacking the upstream protein EGL-1 (a so-called BH3-domain protein), confirming that fragmentation is an integral part of the death process. Surprisingly, both gain-of-function and loss-of-function muta- mutations in the Bcl-2 counterpart CED-9 also blocked fragmentation. This unexpected observation implies that CED-9 has at least two distinct, and apparently antagonizing, functions: sequestration of CED-4, which protects cells from death, and promotion of mitochondrial fission, which the authors suggest enhances apoptosis.

So how important is mitochondrial fragmentation for C. elegans apoptosis? To find out, Conradt and colleagues overexpressed either the wild-type form or a dominant-negative (poison) form of DRP-1 — a protein that participates in the normal fission process9. As has been reported by others, overexpression of the poison form prevented fragmentation. This treatment also resulted in a mild but significant increase in long-term cell survival. In contrast, overexpression of normal DRP-1 increased mitochondrial fragmentation. Using various assays, the authors conclude that this latter treatment also led to the death of at least some cells that normally would have lived. In other words, mitochondrial fragmentation is not only associated with apoptosis in C. elegans, but also contributes to it.

These results are clearly exciting, but a few notes of caution are warranted. First, although Conradt and colleagues make a convincing case for a causal involvement of mitochondrial fragmentation in apoptosis (see Fig. 4 on page 758), the overall effect on cell survival was rather weak. Fewer than 20% of cells could be rescued through expression of the poison form of DRP-1; even less could be killed by overexpression of the wild-type protein. These numbers are much lower than would be observed in animals with mutant CED-3 or CED-9, respectively. Second, because of technical limitations, the authors could only infer the extent of extra cell death caused by overexpression of wild-type DRP-1. A direct quantification is thus still missing. Third, as is the case in mammals, one must remember that mitochondrial fission occurs all the time, also in cells that live. How does the cell distinguish between normal fission and pro-apoptotic fission? Is CED-9 required for both processes, or only for the latter?

The final and perhaps most intriguing question is: how can mitochondrial division contribute to apoptosis? Conradt and colleagues1 posit that fission might promote the release from mitochondria of a cytochrome-c-like molecule, which would cooperate with CED-4 to activate the protein-digesting CED-3. If correct, this would imply that the apoptotic pathways in worms and mammals are much more similar than current dogma suggests. We are sure to see more mitochondria mining in the coming years.


  1. 1

    Jagasia, R., Grote, P., Westermann, B. & Conradt, B. Nature 433, 754–760 (2005).

  2. 2

    Danial, N. N. & Korsmeyer, S. J. Cell 116, 205–219 (2004).

  3. 3

    Hengartner, M. O. Nature 407, 770–776 (2000).

  4. 4

    Karbowski, M. & Youle, R. J. Cell Death Differ. 10, 870–880 (2003).

  5. 5

    Frank, S. et al. Dev. Cell 1, 515–525 (2001).

  6. 6

    Kinchen, J. M. & Hengartner, M. O. Curr. Top. Dev. Biol. 65, 1–45 (2004).

  7. 7

    Parrish, J. et al. Nature 412, 90–94 (2001).

  8. 8

    Horvitz, H. R. Cancer Res. 59, 1701s–1706s (1999).

  9. 9

    Labrousse, A. M., Zappaterra, M. D., Rube, D. A. & van der Bliek, A. M. Mol. Cell 4, 815–826 (1999).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.