Functional validation of cancer stem cell markers in primary human colorectal cancer and established cell lines

In an increasing number of cancers, CSCs have been defined on the basis of the self-renewal and tumor initiation capacity by functional assays. It has been also suggested that CSC populations might be responsible for chemo- and radio-therapy resistance within tumors and ultimately for the post-therapeutic tumor recurrence. The identification of markers identifying CSC is fundamental for the validation of the CSC paradigm and for the development of new CSC-specific drugs and novel therapeutic approach. CD133, CD44 and CD166 have been proposed as putative CSC markers in CRC. These findings have opened the field for an extensive validation of the markers and their use for the development of specific anti-CSC therapy.
In this work, I addressed a) the prognostic relevance of the expression of CSC surface markers in CRC clinical specimens, b) the “in vivo” tumorigenicity of primary CRC derived cells, as related to their expression of putative CSC surface markers, c) the possibility of using cells derived from established CRC cell lines expressing CSC surface markers as CSC cellular model, and finally d) the development of innovative culture models of potential relevance for the screening of anti CRC compounds.
I have analyzed the surface markers expression in correlation with stem cell-like functional features, but no consistent results was found confirming stemness property associated with expression of those markers. These results obviously question the validity of putative surface CRC-SC markers. Taken together these data might suggest that their expression and CSC functional features might be associated with some degree of plasticity, potentially related to tumor microenvironmental characteristics being lost in conventionally cultured tumor cell lines and in primary tumor derived cell suspensions.
Based on this background I have investigated the possibility to perform 3D culture of CRC cell lines to assess whether these systems might provide useful insights for the interpretation of our data. My findings clearly document the plasticity of gene expression profiles of cultured CRC cells depending on their three-dimensional architectures. Most importantly, I demonstrate that major gene expression modulation events only occur when culture in 3D spheroids is associated with ischemia and necrosis. I also showed that the gene expression of putative CSC markers increases with the occurrence of ischemia and necrosis in the core of a 3D spheroid.