On its own, cancer is a pretty disturbing disease. But over the last few years, we've identified an even more disturbing ability: in at least two cases, cancerous cells have evolved into a communicable disease. Now, researchers studying one of these infectious cancers have found that it's apparently stealing part of its host's cells when it infects them.
Currently, we know of two cases of transmissible cancer. In one case, a lethal oral cancer is pushing the Tasmanian devil towards extinction. But the other case is both less lethal and wider reaching: it strikes domestic dogs on all continents, but only causes a transient infection. The disease, canine transmissible venereal tumor, appears to have originated a few thousand years ago, based on the differences between its genome and that of domesticated dogs.
It's not quite clear what to call a cancer that's evolved in isolation from its host for thousand of years and has a lifestyle closer to that of a unicellular parasite—is it a new species? The authors of the new research settle for calling it an "asexual mammalian unicellular pathogen." They were apparently studying the evolution of the pathogen by sequencing the genome of its mitochondria, organelles in each cell that carry their own DNA. You can build an evolutionary tree of an organism's history by figuring out which DNA sequences are most similar and placing them on adjacent branches.
When they tried this with the mitochondrial sequences, the tree they got was a bit unusual, in that it didn't correspond to the tree obtained using the cancer's regular, nuclear genome. In fact, the cancer samples wound up scattered within a phylogenetic tree of the cancer as a whole. The authors offer a simple explanation: cancer cells have continued to steal mitochondria for their hosts, refreshing their own supply in a process that functions a bit like replacing worn-out parts.
They note that cancers are very metabolically active, which means that mitochondria are under a bit of stress to supply the cell with fuel. That makes mutations more probable, and there is evidence that they're likely to accumulate in the cancerous cells. Should one that harms fitness predominate, the cancerous cells would seem to have no option other than to grab a replacement. It would seem likely that, under these circumstances, competition between cells might allow those with harmful mutations to be weeded out, but the authors suggest that transmission between hosts might not create a bottleneck sufficient to create this sort of between-cell selection.
The cancer still seems to be evolving rapidly—the common ancestor of all the strains we've isolated is only a few hundred years old, despite the relative antiquity of the original cancer. At some point, the same lack of bottleneck that lets the mitochondrial genome go bad might affect the regular, nuclear genome of the cells. At which point we may see the cancer evolve a way to start stealing genes from the host it effects.
Science, 2011. DOI: 10.1126/science.1197696 (About DOIs).
reader comments
18