Certain cells bind so tightly to each other that, on occasion, one cell ends up inside another, usually with fatal consequences for the ingested cell. This involuntary cell death might help protect us from cancer.
The epithelial cells that cover most of the surfaces of our bodies create tight physical barriers that protect us from the outside world. To do this effectively, these cells need to stick to each other very well — which they do, thanks to molecular Velcro proteins known as cadherins. In a provocative study published in Cell, Overholtzer et al.1 show that unless epithelial cells are physically restrained, their strong affinity for each other can turn into a deadly embrace, with one cell ending up inside the other. To make the story even more intriguing, the authors show that the internalized cell usually dies by means of a molecular mechanism unlike any other known.
Normally, epithelial cells sit on an extracellular matrix — a complex meshwork of molecules that acts as a support net for cells to crawl on and attach to. Interaction with this matrix is essential not only for epithelial-cell function, but also for these cells' very survival; epithelial cells that detach from the extracellular matrix rapidly activate a cell-suicide programme known as apoptosis, which in animals leads to the elimination of cells that are in excess, in the wrong place or potentially dangerous2,3,4,5.
Once a cell decides to die, it is rapidly recognized by specific 'eat-me' signals on its surface, taken up and degraded by a neighbour6. This clearance process is by necessity highly selective; living cells lack such markers and are left alone. There are, however, some exceptions to this rule. For example, pathologists have often reported the occurrence in cancer tissues of 'cannibalistic' cells, which can apparently ingest other normal-looking cells7.
What could be the basis for this odd behaviour? Overholtzer and colleagues1 propose an attractive answer. They report that, when mammary epithelial cells are separated from the extracellular matrix and left to float in a suspension, they frequently form 'cell-within-cell' structures, whereby one epithelial cell is either partly or completely inside another.
Surprisingly, the cell-within-cell phenomenon the authors observed is not triggered by apoptosis. First, many of the ingested cells looked to be alive and well, and those that were dead had clearly died from something other than apoptosis. Second, the internalized cells did not present the eat-me signals typical of apoptotic cells. Rather, the internalization process seems to depend on the cadherin system, and the authors suggest that ingestion is the result of a tragic mistake, a friendly hug gone awry.
When epithelial cells meet, because of the strong self-adherence properties of cadherins, they will rapidly try to maximize their surface interactions8,9. Normally, the extent of such cadherin-mediated interaction is kept in check by the cells' attachment to the extracellular matrix. The resulting balance of forces leads to the formation of a neat cobblestone pattern characteristic of epithelial sheets (Fig. 1a).
In the absence of solid support, however, there is little to oppose cadherin-based self-adhesion. Consequently, epithelial cells' urge to maximize their surface interactions can be pushed to its logical extreme, whereby, over the course of a few hours, one of the two cells actively 'covers itself' with the other. The topological consequence is that the overeager cell ultimately finds itself, probably quite unwittingly, fully inside its partner (Fig. 1b). The authors name this unusual, unexpected internalization process entosis, from the Greek entos, meaning inside or within.
That entosis is neither premeditated murder nor suicide is further supported by Overholtzer and colleagues' observation of what happens immediately after internalization — namely, not much. Unlike apoptotic cells, which are rapidly degraded after being engulfed, the internalized cell — let's call it an entocyte — lives on inside its host, blissfully ignorant of its precarious condition. The host cell, in turn, also seems to have little idea what to do with its guest, or perhaps even that it is hosting one. Neither side is fully committed to any further step — indeed, the internalization process is reversible, and a small fraction of entocytes are eventually released back into the 'wild' of the Petri dish, with no apparent long-term damage.
For most entocytes, however, the story has a tragic ending. At some point, often many hours after internalization, the ingesting cell suddenly switches from gentle host to ogre, and kills its guest entocyte. This 'entocide' uses a novel, crude, but effective strategy. The host simply fuses the entocyte with its lysosomes — acidic bags of digestive enzymes used by cells to degrade and recycle large molecules or, in this case, the unsuspecting guest. And the entocyte is literally digested alive. It is not currently known what causes this switch in behaviour.
These observations are spectacular, but many sceptics will question their physiological relevance. Are there ever free-floating epithelial cells in our bodies? As mentioned above, cell-in-cell structures can often be detected, admittedly at low frequency, in various cancers. But whether these are the result of self-adhesion-mediated entosis or some other mechanism is not known.
And what about possible functions? Clearly, a mechanism such as entosis could help eliminate mislocalized epithelial cells. Such situations might arise during development, or during cancer progression. Indeed, Overholtzer et al. posit that entosis functions as a barrier to tumour formation by eliminating metastatic cancer cells that have escaped from their physiological niche. But one could also imagine the opposite. Could cancer cells take advantage of entosis, and transiently crawl into their healthy neighbours to escape immune surveillance or chemotherapy?
Finally, it will be of great interest to determine how widespread entosis is in other species, and to identify genes involved in this process. After apoptosis, necrosis and autophagy, entosis is not only the latest addition to the Greek-derived, cell-death-associated jargon, but also a new and provocative cell–cell interaction process, which clearly merits further investigation.
Overholtzer, M. et al. Cell 131, 966–979 (2007).
Frisch, S. M. & Francis, H. J. Cell Biol. 124, 619–626 (1994).
Jacobson, M. D., Weil, M. & Raff, M. C. Cell 88, 347–354 (1997).
Vaux, D. L. & Korsmeyer, S. J. Cell 96, 245–254 (1999).
Galluzzi, L. et al. Cell Death Differ. 14, 1237–1243 (2007).
Savill, J. & Fadok, V. Nature 407, 784–788 (2000).
Abodief, W. T., Dey, P. & Al-Hattab, O. Cytopathology 17, 304–305 (2006).
Adams, C. L., Chen, Y.-T., Smith, S. J. & Nelson, W. J. J. Cell Biol. 142, 1105–1119 (1998).
Pokutta, S. & Weis, W. I. Annu. Rev. Cell. Dev. Biol. 23, 237–261 (2007).
About this article
Scientific Reports (2017)
Biochemistry (Moscow) (2015)
International Journal of Molecular Medicine (2013)
Cell Death & Disease (2013)
Cell Death & Differentiation (2009)