Biomedical Engineering Reference
In-Depth Information
The observation that the cancer stem cell-derived xenografts are heteroge-
neous with respect to both differentiation grade and differentiated cell pheno-
type argues that cancer stem cells remain sensitive to signals that guide differ-
entiation. Indeed, nuclear b-catenin staining is only observed in a subset of cells
in colorectal cancers. This shows that although the genetic defect that leads to
Wnt signaling deregulation is present in all cells, maximal Wnt pathway activity
only occurs in a subset of cells in the cancer (Fodde and Brabletz, 2007).
Moreover, insights gained from the APCmin model for intestinal neoplasia
show that in adenomas differentiation patterns can be influenced by interfering
with the signals that guide normal differentiation in the intestine. When APC-
min mice are subjected to g-secretase inhibition, phenotypic changes are
observed that are comparable to the response of normal intestinal epithelium,
i.e., the induction of goblet cell differentiation (van Es et al., 2005b). This
illustrates that colorectal cancers cannot simply be viewed as monoclonal
expansions of malignantly transformed cells, but should be seen as heteroge-
neous, hierarchically organized tissues. Thus, although crucial homeostatic
control is lost regarding extensive proliferation and invasive growth, other
regulatory mechanisms that control cell differentiation are at least partially
functional.
4.4 Caveats
To date all evidence that exists for the cancer stem cell model of malignancies is
obtained using xenotransplantations (Hill, 2006; Vermeulen et al., 2008). In the
case of CRC all available data suggesting the existence of colon-cancer stem
cells involve complete tissue disruption to generate single cell solutions that can
be sorted based on cell surface molecules. The next step involves injection of the
cells into a non-orthotopic site, e.g., subcutaneously. Both experimental manip-
ulations, the tissue disruption and the xenotransplantation, can cause signifi-
cant biases in the interpretation of the data. For example, it is possible that
CD133
+
cells are better equipped than CD133
cells to survive the complete
tissue disruption process or are better capable to grow out in mice (Hill, 2006;
Vermeulen et al., 2008). It has been described, for example, that expression of
the CD44 molecule has important implications for the engraftment capacity of
cells in a wide variety of systems (Harada et al., 2001; Krause et al., 2006). Care
has to be taken in the interpretation of the identity of a cancer stem cell marker.
Is it a true marker in the sense that it is a read-out of a specific cellular state or,
alternatively, are the markers themselves involved in the process studied, in this
case engraftment and outgrowth of xenotransplants? In addition, the fact that
the exclusive capacity to initiate a new tumor resides in the CD133
+
cell
fraction does not necessarily tell us anything about the situation in an estab-
lished CRC. So, while the data are in perfect accordance with a hierarchical
organization of a CRC, other explanations are not convincingly ruled out at
