ZLive-cell calcium imaging of adherent and non-adherent GL261 cells reveals phenotype-dependent differences in drug responses

Background: The tumor-derived GL261 cell line is used as a model for studying glioblastoma and other high-grade gliomas, and can be cultured adherently or as free-floating aggregates known as neurospheres. These different culture conditions give rise to distinct phenotypes, with increased tumorigenicity displayed by neurosphere-cultured cells. An important technique for understanding GL261 pathobiology is live cell fluorescent imaging of intracellular calcium. However, live cell imaging of GL261 neurospheres presents a technical challenge, as experimental manipulations where drugs are added to the extracellular media cause the cells to move during analysis. Here we present a method to immobilize GL261 neurospheres with low melting point agarose for calcium imaging using the fluorescent calcium sensor fura-2. Methods: GL261 cells were obtained from the NCI-Frederick Cancer Research Tumor Repository and cultured as adherent cells or induced to form neurospheres by placing freshly trypsinized cells into serum-free media containing fibroblast growth factor 2, epidermal growth factor, and B-27 supplement. Prior to experiments, adherent cells were loaded with fura-2 and cultured on 8 well chamber slides. Non-adherent neurospheres were first loaded with fura-2, placed in droplets onto an 8-well chamber slide, and finally covered with a thin layer of low melting point agarose to immobilize the cells. Ratiometric pseudocolored images were obtained during treatment with ATP, capsaicin, or vehicle control. Cells were marked as responsive if fluorescence levels increased more than 30% above baseline. Differences between treatment groups were tested using Student's t-tests and one-way ANOVA. Results: We found that cellular responses to pharmacological treatments differ based on cellular phenotype. Adherent cells and neurospheres both responded to ATP with a rise in intracellular calcium. Notably, capsaicin treatment led to robust responses in GL261 neurospheres but not adherent cells. Conclusions: We demonstrate the use of low melting point agarose for immobilizing GL261 cells, a method that is broadly applicable to any cell type cultured in suspension, including acutely trypsinized cells and primary tumor cells. Our results indicate that it is important to consider GL261 phenotype (adherent or neurosphere) when interpreting data regarding physiological responses to experimental compounds.