Measurement of 3-Dimensional cAMP Distributions in Living Cells using 4-Dimensional (x, y, z, and λ) Hyperspectral FRET Imaging and Analysis

This protocol is significant as it provides a method for measuring and visualizing FRET or cyclic AMP signals in three spatial dimensions in individual cells. Cyclic AMP distributions may occur axially in addition to laterally and it is important to acquire FRET cyclic AMP image data in three dimensions to sample these distributions. On the day of imaging, warm Tyrodes buffer to 37 degrees Celsius in a water bath and mount a cover slip seated with the transfected cells of interest into a cell chamber.

Secure the top of the chamber with a mounting gasket to prevent leaking and use a delicate task wiper to remove any excess medium or adherent cells from the bottom of the cover slip. Add 800 microliters of working buffer and four microliters of five millimolar nuclear label to the cell chamber and gently rock the chamber for five to 10 seconds. Then cover the cell chamber with aluminum foil for a 10 minute incubation at room temperature.

To image the cells, select the 60X water immersion objective on a confocal microscope equipped with a spectral detector and add a drop of water to the objective. Place the loaded cell chamber onto the microscope stage and tune the filter knob to select the appropriate filter set. Select the fluorescence widefield mode and use the eyepieces to select a field of view containing cells expressing the FRET cyclic AMP sensor.

Open the imaging software, select the confocal mode, and unlock the laser interlock button. Open the A1 settings menu. Check the boxes corresponding to the 405 and 561 nanometer laser lines and set these spectral detections to SD, the resolution to 10, and the channels to 32.

Select the start and end wavelength values to set the wavelength range. Open the A1 Settings menu to select the binning skip icon. Select the number 15 box and click OK.Set the laser intensities to eight percent for the 405 nanometer laser and to two percent for the 561 nanometer laser.

Set the detector gain to 149 and the pinhole radius to 2.4 Airy disk units. Set the scan speed to 25 spectral frames per second, and select the unidirectional scan direction. Set the count to four and the scan size to 1024 by 1024.

To define the z-stack acquisition parameters, click View, Acquisition Control, and ND Acquisition, and enter the file destination and name in the pop window to save the ND file. Click the z-series box and click Live in the A1 Plus Settings window to open a live viewing window. Adjust the focus knob to focus on the cells while looking in the preview window.

Adjust the focus knob on the microscope to select the top of the cell and click Top in the ND Acquisition window to set the current position as the top. Adjust the focus knob on the microscope to select the bottom of the cell and click Bottom to set the current position as the bottom. Enter one micrometer for the step size.

Select Top Bottom for the z-scan direction and click run to acquire a z-stack. When the z-stack acquisition is complete, use a pipette to gently add the reagent of interest to the chamber without disturbing the cells. After 10 minutes, change the file name and acquire a second z-stack image.

After imaging, create new folders with the same file names as the spectral z-stack images and open the spectral image file of interest. Click File, Import/Export, and Export ND Document. In the pop-up window, select the newly created folder, TIFF for the file type, Mono image for each channel, and keep bit depth.

Click export to export the individual files to the folder and open the linear spectral unmixing software. Open the Linear Unmixing. m file and click Run.

Browse and select the folder containing the exported TIFF file sequence and click OK to open the wavelength and z-slice window. Copy the file name of the first file without the z-slice and channel number from the folder and paste it into the first step of the Enter of the imagename box, enter the number of channels in the Enter the number of wavelength bands box, and the number of z-slices in the Enter the number of Z Slices box and click OK.Then in the pop-up window, browse and select the Wavelength. mat file and click Open.

Browse and select the Library. mat file in the new pop-up window and click Open again to initiate the slice unmixing. To calculate the FRET efficiency, open the multiFRRCF.

m programming script and enter the number of experimental trials to analyze in the How many folders to reslice box. Click OK.Browse to select the unmixed folders and click OK again. In the new pop-up window, set the scaling factor to 12.4, the threshold to 56, the X, Y, and Z frequency to five, five, and one respectively, and the smoothing algorithm to Gaussian.

Then click Run to perform the FRET measurements and reslicing. To map the FRET efficiency to cyclic AMP levels, open the Mapping_FRET.Efficiency_to_cAMP_concentration. m file and click Run.

Navigate to and select the first grayscale FRET image and click OK.Then open the FRET cyclic AMP images to inspect the distribution of the cyclic AMP signals in three dimensions. In these images, three-dimensional views of false colored raw hyperspectral image data acquired using confocal microscopy at baseline and 10 minutes after forskolin treatment can be observed. In this analysis, an acceptor fluorophore was completely photo destructed allowing spectral signatures of the donor and acceptor with one-to-one stoichiometry to be obtained.

Non-transfected cells labeled with a nuclear dye were utilized to obtain the pure spectrum of the dye fluorophore. Combining the spectra of the donor, acceptor, and nuclear dye fluorophores allows the creation of a three component library. Here, the sources of three background spectral signatures can be observed.

Although these signals are distributed non-uniformly within the sample and cannot simply be subtracted out, adding the spectral signatures of these signals to the spectral library and using linear unmixing to separate the signals provides an approach for removing these confounding signals from the donor and acceptor signals prior to calculating their FRET efficiencies. The six component spectral library can then be used to perform linear spectral unmixing for each slice in the axial image. Here, the changes in FRET efficiency and cyclic AMP levels in different XY plane slices can be observed allowing a comparison between the baseline conditions and the conditions 10 minutes after forskolin treatment.

This technique could also be used to quantify the FRET efficiency of other reporters in three dimensions in both cells and tissues.