FILM-FRET is a powerful technique that makes it possible to confirm suspected or predicted protein-protein interactions. In this protocol, we present a way to analyze data in particular cases of unbalanced donor and acceptor quantities. FILM-FRET is able to provide information about the protein-protein interactions directly in live cells.
It is important to investigate personal-protein interactions in native cellular environment, in particular when fluorescent protein are expressed in the endogenous level. Begin by growing P.aeruginosa, E.coli TOP10, and E.coli helper bacteria each in five milliliters of LB without antibiotic at 30 degrees Celsius while shaking overnight. On the next day, measure the optical density of the cultures and mix an equal quantity of P.aeruginosa, E.coli TOP10 and E.coli helper in a 1.5 milliliter microtube.
Centrifuge the tube for five minutes at 9, 300 times G to pellet the bacteria, then discard the supernatant and resuspend the pellet in 50 microliters of LB.Plate a spot of the mixture on preheated LB agar and incubate it for five hours at 37 degrees Celsius. After the incubation, scrape the spot with a sterile inoculation loop and resuspend it in one milliliter of LB.Plate 100 microliters of this bacterial suspension on LB agar containing 10 micrograms per milliliter chloramphenicol and 30 micrograms per milliliter gentamycin and incubate two days at 37 degrees Celsius to eliminate E.coli TOP10 and E.coli helper bacteria. Resuspend one colony in one milliliter of LB and incubate it at 37 degrees Celsius with orbital shaking for four hours, then centrifuge the tube for three minutes at 9, 300 times G and discard 950 microliters of supernatant.
Resuspend the pellet in 50 microliters of LB and isolate the mixture on LB agar containing sucrose and no sodium chloride. Incubate the plate overnight at 30 degrees Celsius. Spot isolated colonies on LB agar and LB agar containing 15 milligrams per milliliter gentamycin in order to check for gentamycin sensitivity.
Place a microscope glass slide on a flat horizontal surface and arrange two glass slides topped with two layers of adhesive tape around it. Pipette a 70 microliter droplet of 1%melted agarose onto the glass slide and place a fourth slide on top to flatten the agarose droplet. Press down gently for about a minute.
Take off the upper slide and drop about three microliters of bacteria in three to four spots at different locations on the agarose pad. Cover the agarose pad with a microscopy glass coverslip and fix it with melted paraffin to seal it onto the glass slide starting with the four corners. Place the microscopy slide on the stage with the coverslip facing the objective.
Turn the filter cube turret to select the eGFP cube and open the fluorescence lamp shutter, then send the fluorescence light towards the eyepiece of the microscope. Focus the objective on the bacteria using the microscope knob. Select a region of interest in the sample using the joystick that controls the motorized stage.
Send the fluorescence emission path back towards the detector, then turn back the filter cube turret to select the dichroic cube for the 930 nanometer laser and set the laser power to 20 milliwatts. Set the size of the region of interest to 30 micrometers which adjusts the voltage operating the galvo mirrors and defines the range of their movements. Turn on the detector and start scanning the sample.
Choose the field of view for imaging by finely moving the stage from the computer interface. This can be done on the setup by moving the cross on the image in the microscope control software which will define the new center of the image and pressing move stage. A good field of view for acquisition should contain 10 to 30 immobile bacteria all in focus.
If interested in extracting single cells FILM-FRET data, ensure that the bacteria are well-individualized. Open the imaging software and check that the photon count rate is not too high to avoid a pileup effect that can influence lifetime measurements. If necessary, lower the laser intensity to keep the photon count rate low.
Adjust acquisition parameters including the acquisition collection time. Press the start button and wait for the acquisition to complete. To analyze the data, open RStudio and create a new project.
Create a new folder in the main project folder and name it data, then move all analysis ask files into this folder. Open a new script file or the supplementary script flim_analysis.r. Install the dedicated FlimDiagRam package for film data analysis and call the package in the workspace.
Empirical cumulative distribution functions of the fluorescence lifetimes measured for the different bacterial strains are shown here. If FRET occurs, the functions are shifted towards the shorter lived lifetimes. It is important to note that FRET can't occur when the interaction of the two proteins results in a long distance between the two flurophores, which can't be distinguished from the absence of interaction.
Therefore, it may be necessary to label the proteins at different positions to maximize the chances of probing the interaction. Due to the large difference in protein expressions between PvdA and PvdL, the same complex does not result in similar FILM-FRET data. Unbalanced stoichiometries lead to differences in the contribution of the free as compared to the bound donor-labeled proteins in the recorded fluorescence lifetime distribution.
The diagram plot can be used to provide critical information about the stoichiometry. In the PvdA-eGFP mCherry-PvdL mutant, the quantity of donor-labeled PvdA is much higher than the quantity of mCherry-PvdL. Amongst all donors present in the sample, only a few of them are interacting with PvdL.
The single tau one value centered at approximately 2.3 nanoseconds suggests that each PvdA-eGFP donor can only transfer with one mCherry-PvdL acceptor. When the labeling is reversed, most of the eGFP-PvdL proteins are expected to interact with PvdA-mCherry which is confirmed by the higher alpha one values. The tau one values became more distributed suggesting that multiple PvdA proteins may bind to a single PvdL protein.
The FLIM setup is becoming available in an increasing number of imaging facilities. Imaging samples in a FILM-FRET is no more complicated than imaging in video microscopy.
