Apoptosis – programmed cell death – is one of those processes that are only supposed to go one way. Once begun, there is no turning back. The cell begins to degrade itself through the activation of caspases, proteolytic enzymes that cleave the structural elements of the cell, leading to a series of morphological and biochemical changes. The cell digests itself from the inside, turning itself to a soup of dead organic molecules inside the cell membrane. In the end, it fragments neatly into membrane-enclosed bags that are small enough for nearby cells to phagocytose, i.e. eat up.
Evidence for reversible apoptosis
In a recent paper in the British Journal of Cancer, Tang et al. show that apoptosis may be reversible. They have taken a panel of several cancer cell lines and treated them with an inducer of apoptosis for a few hours, and then washed it off. When they monitored the cells over the next hours, they found that many cells transiently exhibit morphological signs of apoptosis: they shrunk, the nuclei condensed, and mitochondria were degraded, but a few hours later the cells returned to their normal shape. The limiting event from which the cells had no chance of return was when degradation of the nucleus had begun. Caspases were activated at the same modest level from 3 up to 48 h after induction, even though the cells were back to their normal shape before 24 h.
Loss of bistability in the apoptosis signalling network
In terms of the signalling network, apoptosis should be a bistable process. The cell should very firmly be either apoptotic or non-apoptotic. Cells that start to digest themselves and then stop should have a strong fitness penalty. We would expect the evolutionary process to select strongly against a reversibility trait both in an organism and in a group of cells. Contrary to this expectation, the reversibility trait predominates in all the cell lines investigated in this study. Something has been lost, or gained, in their signalling network. But how can they survive a partial self-degradation?
Mitochondria – redundant in cancer cells?
The hallmarks of apoptosis that are investigated in the present study reflect degradation in three steps: First, the cytoskeleton, leading to cell shrinkage. Then, the mitochondria, shutting off the cell’s aerobic metabolism. And finally, the nucleus, with the entire gene regulatory system. But the cytoskeleton is made to be remodelled, so it’s not that surprising that the cells can cope. With the mitochondria, it’s a different story. Cancer cells almost universally switch to anaerobic metabolism during progression. This is known as the Warburg effect. It is a very curious phenomenon that still awaits explanation. Many of the mitochondria in tumour cells therefore lack any useful function. In a sense they are little more than decoration, present mainly for historical reasons.
The hen and the egg
Is it the case that tumour cells first acquire apoptosis resistance through mutations that repeal the bistability of their signalling network, and then are forced to rely less on their mitochondria because they are sometimes degraded? Or is it the other way around? Or are these two processes driven by unrelated factors, leading to reversibility of apoptosis by chance, as it were?
It is too early to draw any conclusions. Further studies should investigate reversibility in relation to biochemical and not just morphological parameters. I will write more on the properties of bistable signalling networks and their robustness during the coming week.
Tang, H., Yuen, K., Tang, H., & Fung, M. (2008). Reversibility of apoptosis in cancer cells British Journal of Cancer, 100 (1), 118-122 DOI: 10.1038/sj.bjc.6604802