Posted by: Dan | February 5, 2007

Brain Tumor Stem-like Cells

(Entry #2 for the week of science)

I’ve previously mentioned that I’m rather skeptical on the stem cell theory of cancer, even if it may be clear that those pernicious tumor cells may acquire the capacity for self-renewal by co-opting mechanisms of dedifferentiation and pluripotency. One review from last summer makes a very strong argument, however, for stem cells as the source of some tumors, including some brain tumors in particular. In Nature Reviews Cancer, Angelo Vescovi et al. (2006) explain (emphasis mine):

The relevance of [numerous] findings is manifold. First, they confirm that different brain tumours contain transformed, undifferentiated neural precursors that respond to the same mitogens that activate adult neural stem cells. Second, they indicate that tumour stem-like cells possess some of the molecular features of neural stem cells. Third, they show that CD133, a 120 kDa cell-surface protein that is a marker of normal human neural precursors, can be used for the enrichment of tumour stem-like cells from brain tumours.

That strikes me as very strong support, showing that these brain tumor cells can indeed follow the pathways of multi-lineage differentiation observed in ‘true’ stem cells. Vescovi et al. support this with observations that CD133+ cells from human glioblastoma multiforme and medulloblastoma (the two principle types of brain cancer studied) can initiate tumors in mice, but CD133- brain tumor cells cannot.

But, as they note, the term “brain tumour stem cells is often loosely applied,” so they attempt to clarify the issue by suggesting a list of defined characteristics:

  • Cancer-initiating ability upon orthotopic implantation (tumours should be a phenocopy of the tumour of origin)
  • Extensive self-renewal ability, demonstrated either ex vivo (by showing both sequential-clonogenic and population-kinetic analyses) or in vivo (by serial, orthotopic transplantation)
  • Karyotypic or genetic alterations
  • Aberrant differentiation properties
  • Capacity to generate non-tumorigenic end cells
  • Multilineage differentiation capacity (Not a defining characteristic in all circumstances)

Again, all of this rests upon the capacity of these tumor stem cells to differentiate, in this case into neurons and glial cells. If this has been demonstrated, then case-closed, I think.

The authors note however that “the specific identity of [brain tumor stem cells] within the neurogenetic regions of the brain remain unresolved.” And damningly, I came across a paper in one of Nature‘s subjournals, Cell Research, that suggests that brain tumor stem cells resist differentiation (Zhang et al., 2006).

So what’s going on here? It seems likely that ‘true’ neural stem cells that have accumulated enough oncogenic mutations are indeed a primary source of brain tumor cells. At this point, however, they cease to be stem cells, and become tumor cells – retaining some characteristics, but becoming ‘damaged goods.’


  • Vescovi AL, Galli R, and Reynolds BA. 2006. Brain tumour stem cells. Nature Reviews Cancer 6, 425-436. Pubmed
  • Zhang QB, Ji XY, Huang Q, Dong J, Zhu YD, and Lan Q. 2006. Differentiation profile of brain tumor stem cells: a comparative study with neural stem cells. Cell Research 16(12):909-15. Pubmed


%d bloggers like this: