A caspase sensor based on Förster resonance energy transfer between fluorescent proteins is reported. Enhanced cyan fluorescent protein anchored to the inner leaflet of the plasma membrane of living cells is optically excited by an evanescent electromagnetic field and transfers its excitation energy via a spacer (DEVD) to an enhanced yellow fluorescent protein. Upon apoptosis, DEVD is cleaved and energy transfer is disrupted, as proven by pronounced changes in fluorescence spectra and decay times. Fluorescence spectroscopy and lifetime imaging (FLIM) is combined with total internal reflection fluorescence microscopy (TIRFM) for selective detection of this membrane-bound caspase sensor. Fluorophores of the cytoplasm are thus excluded, and the signal-to-background ratio is increased considerably. In comparison with conventional or laser scanning microscopy, this permits long-term observation of apoptosis in live cell cultures using very low absorption and avoiding light-induced damages of the samples. '
2008International Society for Advancement of Cytometry Key terms caspase sensor; apoptosis; Förster resonance energy transfer; fluorescence lifetime microscopy; total internal reflection microscopy; live cell microscopy THE measurement of caspase activities is commonly used to detect and investigate programmed cell death. Cyan fluorescent protein (CFP) fused to yellow fluorescent protein (YFP) by a caspase-sensitive amino acid peptide (DEVD) has been developed as an indicator of caspase-3 activity in living cells (1,2). The peptide linker between CFP and YFP is short enough to bring the two fluorescent proteins in close proximity to each other resulting in Förster resonance fluorescence energy transfer [FRET; (3)]. Upon induction of apoptosis, FRET is interrupted due to cleavage of the linker by caspase-3, and thus loss of FRET is a direct indicator of caspase activity. The sensitivity of this genetically encoded caspase sensor has been improved by optimizing the amino acid sequence flanking DEVD for higher FRET efficiency and thus improving the dynamic range of signal generation (4,5). Caspase sensor constructs following this principle have since been used to investigate caspase activity in living cells. For example, the rate of apoptosis of whole cell populations was determined (6-8), or the temporally distinct onset of apoptosis in single cells within whole cell collectives was detected (8). Furthermore, cleavage sequences specific for different caspases were used to reveal the activation of these caspases in the same cell (9,10).Detection of events of such dynamic occurrence often requires a temporal and spatial resolution in data acquisition that are only obtainable by real-time or timelapse recordings. So far, changes in FRET of caspase sensor probes have been mostly detected by calculating the ratio of acceptor (YFP) and donor (CFP) emission using conventional standard fluorescence or confocal laser scanning microscopy (4,11). However, both microscopic methods can be harmful to living cells because...