Biology Reference
In-Depth Information
Application of FRET and Microscopy
To maintain total plasmid concentration, an equal amount of empty vector was trans-
fected together with receptor. To estimate the spectral contributions of light scatter-
ing and autofluorescence of cells, reference spectra of empty vector transfected cells
were used. Background and Raman spectra were unmixed along with references dur-
ing the data analysis. A CFP-YFP tandem construct with a 1:1 stoichiometry and
E
35%was used to obtain
R
TC
and as positive control. As negative control, we used
CFP and YFP, which are expressed in cytosol.
In contrast to FRET microscopy, where we use 440 and 514 nm excitation for
CFP-YFP pairs, excitation wavelength ex1:440 and ex2:488 nm was applied in
the Fluorolog. At 440 nm, the relative excitation of CFP to YFP is optimal. Ex2 how-
ever differs from microscope experiments of optimally 514 nm. To find best excita-
tion wavelengths, we test lux-FRET accuracy and error propagation by using various
couples of excitation wavelengths. We found that when CFP and YFP wavelengths
are longer than 490 nm it lead to serious unmixing errors since reference spectra of
acceptor and cell background have similar line shape above 500 nm and thus can
hardly be distinguished.
Acquisition conditions were set to excitation slit width 4 nm (emission was
recorded 450-600 nm for ex1 and 498-600 nm for ex2) and emission slit width
2 nm, with step width of 1 nm and integration time 1 s. In case of time lapse exper-
iments, acquisition protocol can be further optimized by sparsely recording second
excitation (e.g., only at start and end of time lapse).
¼
14.3.5
FRET imaging of receptor-receptor interaction for intact cells
with subcellular spatial resolution
14.3.5.1
Zeiss LSM 780
For pixel-based analysis and in order to achieve high spatial resolution in 3D in de-
tail, we perform the following calibration steps for best confocal overlap at both ex-
citation conditions: (1) MBS to pinhole calibration after thermal setting of 24 h is
especially required for CFP-YFP experiments with 440/514 nm excitation since
the two laser lines are introduced to the optical path by different “TwinGate” main
beam splitters, which could cause
x
-
y
-pixel shifts up to 200 nm due to thermal ef-
fects. For constant thermal settings, LSM part of the microscope is not switched off
on daily basis. (2) Correction collar adjustment to correct for coverslip thickness
mainly influences confocality in
z
-direction and thus influences brightness and leads
to spectral aberration at suboptimal settings. (3) Collimator adjustment for 440 nm
laser line is done to gain maximum intensity and best spatial overlap with 514 nm
excitation in
z
-direction. Calibration steps 2 and 3 can be performed best in the “fast
Z” mode, where
x
-
z
images can be recorded sufficiently fast.
After calibrating the system, live-cell FRET experiments were performed. FRET
images of N1E-115 cells expressing 5-HT
1A
-CFP and 5-HT
7
-YFP receptors were
acquired using a 40
/1.2 NA water immersion objective at frame size 1024
1024,
bit depth 16 bit, and pinhole setting of 1-1.5 AU. To obtain best spatial overlap of
both excitations, we applied line-wise switching of excitation. This could be
