From FRET Imaging to Practical Methodology for Kinase Activity Sensing in Living Cells - Progress in Molecular Biology and Translational Science - page 200

method 145 or in tissue. 146 However, the biosensor expression level can
induce differences in the amplitude response. Indeed, the FRET probes
destabilize the reaction equilibrium between a kinase and its substrates.
We therefore recommend performing measurements on cells with equiva-
lent KAR expression levels. To avoid the variations caused by the sensor
expression levels, the creation of a stable cell line might be an alternative.

7.2.4 Photoinduced effects
Photoinduced effects are another source of artifacts. Among these effects, the
most obvious is cell death, which is easily noticeable by changes in cell mor-
phology. Photostress is more tricky to observe and can result in cell
autofluorescence. While autofluorescence spectra are broad and cannot be
corrected by filters, and autofluorescence lifetime is very short and can be
confused with a FRET event, both spectral and lifetime properties of
autofluorescence can create artifacts in measurements with both intensity
ratio- and lifetime-based techniques. The last photoinduced effect is photo-
bleaching, which can also falsify measurements with all techniques, thereby
leading to unacceptable errors in data interpretation. Extensive controls of
photoinduced effect on cells expressing a KAR negative reference are thus
indispensable (see Section 7.2.5 ).
These photoinduced effects depend on the laser power and the exposure
time. Thus, it will be less critical for techniques that require a lower illumi-
nation power ( P average measured using a power meter). For example, rati-
ometric measurements will be less affected than FD FLIM experiments,
and TSCPC is the most stressful method because of the time necessary to
scan an entire cell. Since most biosensors are based on the CFP/YFP FRET
pair (or their variants such as Cerulean, Turquoise, CyPET, Venus, circu-
larly permutated Venus, YPet, etc.), we will focus on these fluorophores to
exemplify the acquisition time and power required for each technique (per-
formed on our systems).
In ratiometric measurements, after exciting the donor, the exposure time
is set to 200 ms ( P average

<

1 mW) for CFP and YFP, respec-
tively, with our microscopy system. 48 As described in Section 5 , acquiring
a channel with excitation and observation of YFP is a prerequisite in inter-
molecular FRET. It is, however, not recommended in the case of sensor
imaging, because this optional correction step will increase both photoin-
duced effects and noise, without improving results. Acceptors excitation
can, however, be useful in another way. Indeed, acceptor photobleaching
can be a good negative control to recover a basal level after activation of

500 and