Abstract
Alteration of tumor cell growth kinetics is the goal of nearly all current or proposed therapies for human neoplasms. The adenosine triphosphate (ATP) chemiluminescent assay has been used for some time as a surrogate marker of in vitro cell growth. Here we present data showing that three human glioblastoma cell lines (U87, U251, G55) demonstrate significantly different cell number to luminescence relationships when subjected to this assay. We plated progressively increasing numbers of cells per well; from 1000 to 50,000 were grown in Dulbecco's modified Eagle's medium without serum and cultured for 6 hours. Cells were then lysed and subjected to the chemiluminescent assay to measure ATP levels and a linear relationship between cell number and measured luminescence was found. Despite this, we found that the slope of the regression line (β) varied markedly between different cell lines (U251 [β = 0.968 ± 0.3] vs. U87 [β = 0.772 ± 0.2] vs. G55 [β = 0.757 ± 0.2]; p < 0.0001), suggesting a difference in ATP luminescence per cell between these cell lines. Thus, we have demonstrated that luminescence values are internally linear within a given cell population, but luminescence level per cell varies significantly between different glioma cell lines. Our findings suggest that different glioma cell lines have unique levels of ATP per cell.