Neurosphere and adherent culture conditions are equivalent for malignant glioma stem cell lines

Certain limitations of the neurosphere assay (NSA) have resulted in a search for alternative culture techniques for brain tumor-initiating cells (TICs). Recently, reports have described growing glioblastoma (GBM) TICs as a monolayer using laminin. We performed a side-by-side analysis of the NSA and laminin (adherent) culture conditions to compare the growth and expansion of GBM TICs. GBM cells were grown using the NSA and adherent culture conditions. Comparisons were made using growth in culture, apoptosis assays, protein expression, limiting dilution clonal frequency assay, genetic affymetrix analysis, and tumorigenicity in vivo. In vitro expansion curves for the NSA and adherent culture conditions were virtually identical (P=0.24) and the clonogenic frequencies (5.2% for NSA vs. 5.0% for laminin, P=0.9) were similar as well. Likewise, markers of differentiation (glial fibrillary acidic protein and beta tubulin III) and proliferation (Ki67 and MCM2) revealed no statistical difference between the sphere and attachment methods. Several different methods were used to determine the numbers of dead or dying cells (trypan blue, DiIC, caspase-3, and annexin V) with none of the assays noting a meaningful variance between the two methods. In addition, genetic expression analysis with microarrays revealed no significant differences between the two groups. Finally, glioma cells derived from both methods of expansion formed large invasive tumors exhibiting GBM features when implanted in immune-compromised animals. A detailed functional, protein and genetic characterization of human GBM cells cultured in serum-free defined conditions demonstrated no statistically meaningful differences when grown using sphere (NSA) or adherent conditions. Hence, both methods are functionally equivalent and remain suitable options for expanding primary high-grade gliomas in tissue culture.