Abstract
Recent advancement in the area of green fluorescent protein techniques coupled with microscopic imaging has significantly contributed in defining and dissecting subcellular changes of apoptosis with high spatio-temporal resolution. Although single cell based studies using EGFP and associated techniques have provided valuable information of initiation and hierarchical changes of apoptosis, they are yet to be exploited for multiparameter cell based real time analysis for possible drug screening or pathway defining in a high throughput manner. Here we have developed multiple cancer cell lines expressing FRET sensors for active caspases and adapted them for high throughput live cell ratio imaging, enabling high content image based multiparameter analysis. Sensitivity of the system to detect live cell caspase activation was substantiated by confocal acceptor bleaching as well as wide field FRET imaging. Multiple caspase-specific activities of DEVDase, IETDase and LEHDase were analysed simultaneously with other decisive events of cell death. Through simultaneous analysis of caspase activation by FRET ratio change coupled with detection of mitochondrial membrane potential loss or superoxide generation, we identified several antitumor agents that induced caspase activation with or without membrane potential loss or superoxide generation. Also, cells that escaped the initial drug-induced caspase activation could be easily followed up for defining long term fate. Employing such a revisit imaging strategy of the same area, we have tracked the caspase surviving fractions with multiple drugs and its subsequent response to retreatment, revealing drug-dependent diverging fate of surviving cells. This thereby indicates towards a complex control of drug induced tumor resistance. The technique described here has wider application in both screening of compound libraries as well as in defining apoptotic pathways by linking multiple signaling to identify non-classical apoptosis inducing agents, the greatest advantage being that the high content information obtained are from individual cells rather than being population based.