To study the properties of individual PP1 isoforms in living cells, including their localization and interaction with targeting subunits, we have taken advantage of the in vivo approach of fusing PP1 to fluorescent reporter molecules, in this case chromatic variants of the green fluorescent protein (GFP) derived from the jellyfish Aequorea victoria. This method effectively “tags” the intracellular pool of the protein and allows analyses of dynamic properties in living cells. Their changing spatio-temporal distributions can be monitored both throughout the cell cycle and following cellular perturbations by time-lapse fluorescence microscopy, and turnover rates of intracellular pools of the protein calculated by fluorescence recovery after photobleaching (FRAP). Interactions with targeting subunits can be visualized in vivo by fluorescence resonance energy transfer (FRET), using techniques as sensitized emission, acceptor photobleaching, or fluorescence lifetime imaging (FLIM/FRET).

For imaging live cells in glass bottom dishes or POC chambers, replace the standard growth medium with Phenol Red-free medium supplemented with FBS and penicillin-streptomycin. If using an environmental chamber supplied with 5% CO2, pH will be maintained. If not, add 20 mM HEPES to the medium as a buffer. Alternatively, a custom-made CO2-independent Phenol Red-free medium can be used, supplemented with FBS and penicillin-streptomycin. This medium does not require additional buffering. DNA can be stained in live cells, with the cells remaining viable and progressing through the cell cycle. To do this, pre-treat cells for 30 min with Hoechst No. 33342 dye. It is important to note that this dye cannot be used with ECFP-tagged proteins since the two cannot be spectrally resolved.

When imaging cells through the cell cycle, prophase cells can be found by their characteristic condensed DNA pattern. Minimize the amount of light the cells are subjected to during imaging by keeping exposure times as low as possible while maintaining adequate resolution. Hoechst No. 33342 can be imaged using a DAPI filter set, while EGFP can be imaged using a FITC filter set. EYFP and ECFP require special filter sets. Recently, we have used EGFP in combination with mCherry, which can be separated using a FITC/TRITC filter set.

Time-lapse imaging of FP-PP1. HeLa cells stably expressing FP-PP1γ and stained with the cell permeable DNA dye Hoechst No. 33342 were imaged as they progressed through mitosis. Images were taken every 3 minutes using low light levels and short exposure times to minimize photodamage. This cell starts in prophase, with the DNA condensing and FP-PP1γ still visible within nucleoli. Within 12 minutes it has progressed to metaphase, at which point FP-PP1γ is primarily diffuse but also shows an accumulation at kinetochores. As the cell progresses through anaphase (at 15-18 minutes), FP-PP1γ shows a relocalization to chromatin, where it remains until it starts to re-accumulate in nucleoli at late telophase (just visible at 27 minutes).

Although time-lapse imaging can only show the steady-state distribution of a protein over time, FRAP analyses can reveal its kinetic properties, such as whether it’s free to diffuse or immobilized to a scaffold, and if it’s exchanging between particular compartments and at what rate. In short, bleaching a fluorescent pool of protein in one region of the cell using a high-intensity laser pulse renders that protein “invisible”, and recovery can then be monitored as the unbleached molecules from neighboring regions of the cell move into the bleached region.

FRAP experiments are generally performed on cells expressing GFP- or EGFP-tagged proteins (EYFP shows a small but significant amount of spontaneous recovery of the molecule from photobleaching, which complicates the interpretation of FRAP results), and can be done using either laser scanning microscopes or widefield fluorescence microscopes equipped with an external laser for photobleaching. When analyzing proteins with very fast turnover rates, the microscope system used must offer the appropriate temporal resolution (i.e. image rapidly enough to capture early events).

To perform a FRAP experiment, take 2-3 images of the cell prior to photobleaching and then bleach a region of interest to approximately 50% of its original intensity, acquiring images over time after the photobleach period to monitor recovery of fluorescence signal within the bleached region. If recovery is observed, it indicates that the fluorescently-tagged proteins are mobile, and the rate of recovery is therefore an indication of the speed at which they’re moving. For comparison, the same photobleaching experiment can be conducted on fixed cells, in which no recovery is observed.

For qualitative FRAP, it may be enough to simply plot fluorescence intensity over time (known as the recovery curve), as a fraction of the initial fluorescence prior to photobleaching. Different pools of the same protein can be compared for relative mobility, or the same pool of protein can be assessed before and after a particular perturbation.

For quantitative FRAP, the most commonly published results are the mobile fraction (fraction of fluorescent molecules that are free to move within the bleached region) and the half-time for recovery of fluorescence (t1/2), which is the time required for half of the mobile fraction to recover. The t1/2 can then be used to calculate the diffusion coefficient (for reference see: Carrero G., McDonald D., Crawford E., de Vries G. and Hendzel, M.J. (2003) Using FRAP and mathematical modeling to determine the in vivo kinetics of nuclear proteins. Methods. 1:14-28.)

Example of a FRAP experiment in which the turnover of three different cellular pools of EGFP-PP1γ were compared (cytoplasmic vs. nucleoplasmic vs. nucleolar). A small circular region (hashed circle indicated by arrow) was photobleached in a region of the cell and recovery of fluorescent signal within that region monitored over time. By plotting recovery of fluorescence intensity versus time after photobleaching , half-times of recovery can be calculated for each pool of FP-PP1γ. Cytoplasmic FP-PP1γ shows the fastest recovery, with a t1/2 of approximately 1 sec (comparable to that for free GFP). Nucleoplasmic FP-PP1γ shows a slower recovery rate (t1/2 = 2 sec), as does nucleolar FP-PP1γ (t1/2 = 9 sec), but these rates still indicate a rapid flux of PP1 through these intracellular pools.

Interactions between proteins, such as PP1 and a targeting subunit, can be shown in vivo using various FRET measurement techniques. The basic principle of FRET is the transfer of energy from an excited donor fluorophore to an acceptor fluorophore in close proximity. FRET is strongly dependent on the distance between donor and acceptor, falling off with the sixth power of the distance between the two. Because of this, FRET can only occur when the proteins are within 1-10 nm of each other and in the proper orientation. A simplified diagram is presented here, which demonstrates how excitation of the donor sensitizes emission from the acceptor that ordinarily would not occur. FRET can therefore be detected as sensitized emission of the acceptor.

Energy transfer from donor to acceptor depletes or “quenches” the excited state population of the donor, and FRET will therefore reduce the fluorescence intensity of the donor. Photobleaching the acceptor to relieve this quenching of the donor (termed “acceptor photobleaching”) offers another option for detecting FRET in vivo. FRET-induced donor quenching is also observed as a decrease in the donor’s fluorescence lifetime, which is the average time that a molecule spends in the excited state before emitting a photon and returning to the ground state. Comparison of donor lifetime in the presence and absence of acceptor is the last FRET method presented here, termed the FLIM/FRET technique. Key advantages of this approach are the independence of measurement on probe concentration, and the use of infrared excitation wavelengths which are less damaging to cells. For all three of these FRET techniques, the efficiency of energy transfer may be used as a molecular ruler to determine the scale of a particular interaction (for reference see: Stryer, L. (1978) Fluorescence energy transfer as a spectroscopic ruler. Annu. Rev. Biochem. 47, 819-846).

This approach requires independent control of emission and excitation filters. Measurements are taken using a combination of three filter sets, ECFP (excite 436 nm, emit 470 nm), EYFP (excite 514 nm, emit 528 nm) and what we refer to as the FRET channel (excite 436 nm, emit 528 nm). The principle of FRET is detailed in this cartoon.

For the experimental condition cells should be co-transfected with EYFP-PP1 and ECFP-targeting subunit (or the inverse, ECFP-PP1 and EYFP-targeting subunit. For control measurements, transfect some cells with the EYFP construct alone, and some with the ECFP construct alone. As a negative control, co-transfect cells with PP1 and a mutant version of the targeting subunit that cannot bind PP1 (e.g. mutation of one or both hydrophobic residues in the RVXF motif). An alternate control is the ECFP-tagged protein co-expressed with an EYFP-tagged protein that shows the same localization but does not interact with it.

To correct for spectral bleed-through, first examine cells expressing either donor alone or acceptor alone with each of the three filter sets. For example, using our system, ECFP-NIPP1 alone gives a strong signal in the ECFP channel and no signal in the EYFP channel, but it does show significant spectral bleed-through into the FRET channel (~70% of the signal measured in the ECFP channel). EYFP-PP1 alone gives a strong signal in the EYFP channel and no signal in the ECFP channel, but also shows spectral bleed-through into the FRET channel (~20% of the signal measured in the EYFP channel).

Measure FRET in cells expressing both PP1 and targeting subunit by exciting at the ECFP wavelength (436 nm) and detecting at the EYFP emission wavelength (528 nm). The following equation is used for the correction of spectral bleed-through, in which FRET(N) is Net Energy Transfer:

Certain software programs, such as the SoftWorx analysis software used with the DeltaVision system, allow image subtraction, to obtain a final FRET image showing the signal remaining in the FRET channel following this correction.

The same analysis should then be applied to data collected from cells expressing PP1 and a mutant targeting subunit that cannot interact with it, or alternatively two proteins that show the same localization but do not interact. After the data have been corrected for spectral bleed-through, there should be no signal remaining in the FRET channel.

Positive control is shown on the left (FRET between PP1 and wild type NIPP1), and negative control is shown on the right (no FRET between PP1 and NIPP1 in which the PP1 binding site has been mutated). Note that in the latter case the mutant NIPP1 cannot retarget PP1 out of the nucleolus because it does not interact with it.

This approach, which is based on the increase in donor (ECFP) signal when FRET is disrupted by photobleaching the acceptor (EYFP) molecule, requires a laser line at approximately 532 nm to photobleach EYFP (without bleaching ECFP) and appropriate excitation and emission filter sets to monitor ECFP fluorescence before and after this bleaching. It can be done using either a laser scanning microscope such asa Zeiss LSM 51 510 or a widefield fluorescence microscope equipped with an external laser, such as the DeltaVision Spectris. The DeltaVision system offers slightly greater flexibility due to its improved temporal resolution, since the increased donor signal is only apparent until the photobleached pool of acceptor recovers and once again quenches the signal. If the protein happens to turn over rapidly, as does PP1, then the unquenched donor signal will only be observed at very early time points. This can be compensated for when using a system with slower post-bleach imaging resolution by photobleaching a larger area of the cell, to ensure that acceptor recovery does not mask the unquenched donor signal in the region of interest.

Cells should be co-transfected with EYFP-PP1 and ECFP-targeting subunit, or alternatively ECFP-PP1 and EYFP-targeting subunit. After obtaining several pre-bleach images of both the donor and the acceptor proteins, photobleach the acceptor in a region of the cell using the appropriate settings for the system used (e.g. 100% laser power and several iterations). After bleaching, collect images at the desired time intervals, using the same settings used to obtain the pre-bleach images.

Analyze the data utilizing the image analysis tools included in the imaging system’s software. Most systems now include FRAP analysis tools in their software packages. Spreadsheet and biostatistics programs such as Microsoft Excel (Microsoft Corp., California, USA) and GraphPad Prism (GraphPad Software Inc., California, USA) are also useful tools for the analysis of photobleaching data. In addition to the bleached region, include a region in the non-bleached portion of the same cell or in a neighboring cell in the data analysis as a control for bleaching due to imaging. A region of background fluorescence should also be defined outside the cell, and subtracted from both the bleached and control regions.

To briefly summarize the principle of lifetime imaging, the fluorescence of organic molecules is not only characterized by their excitation and emission spectra, but also by their lifetimes. When a fluorophore absorbs a photon it goes into the excited state and returns to the ground state by emitting a fluorescence photon, converting the energy internally, or by transferring the energy to the environment. Although fluorescence lifetime imaging systems can seem overly complex at first, in reality the basics are fairly easy to grasp (see Fig. 4). We use the BioRad Radiance Multiphoton Imaging System, which allows us to obtain both intensity and lifetime images for the fluorescent proteins of interest in live cells.

Cells should be co-transfected with either ECFP-targeting subunit and EYFP-PP1 or ECFP-PP1 and EYFP-targeting subunit. For control measurements, transfect cells with the ECFP construct alone (to measure the lifetime of unquenched ECFP), or with the ECFP construct and an EYFP-tagged protein that shows the same localization but does not interact with it. Excite ECFP at a wavelength of 840 nm, allowing the external detectors to collect the fluorescent emission and calculate lifetimes.

In the figure below, lifetime maps have been plotted (i.e. each pixel is color coded with a lifetime value to build up an image of differences in lifetime throughout the sample) for ECFP-NIPP1 in the presence of either EYFP-PP1 or EYFP-U1A. They are shown next to corresponding intensity images. Quenching of the donor (ECFP-NIPP1) causes a shift to a shorter lifetime, and FRET is thus observed as the appearance of a second, “quenched” lifetime (~1.6 ns, compared to the unquenched lifetime of ~1.9 ns). In the presence of EYFP-U1A, which shows the same localization pattern but does not interact with ECFP-NIPP1, only a single, unquenched lifetime of ~1.9 ns is observed. The quenched and unquenched lifetimes can also be color-coded to demonstrate more clearly where FRET is occurring within the cell. In this example, although NIPP1 is found throughout the cell nucleus, the predominant FRET signal is observed at nuclear speckles, where mRNA splicing factors are known to accumulate, suggesting a role for the complex in the regulation of pre-mRNA splicing.

ECFP-NIPP in the presence of EYFP-PP1 (same localization, FRET interaction indicated by quenching of donor by acceptor).
ECFP-NIPP1 in the presence of EYFP-U1A (same localization, no quenching of donor by acceptor).

Trinkle-Mulcahy, L., Chusainow, J., Lam, Y.W., Swift, S. and Lamond, A.I. Visualization of intracellular PP1 targeting through transiently and stably expressed fluorescent protein fusions. Methods in Molecular Biology. Vol.. 365:133-54; Protein Phosphatase Protocols, Ed. G. Moorhead, Humana Press, 2006.

Trinkle-Mulcahy, L.

Swift, S.R. and L. Trinkle-Mulcahy. Basic principles of FRAP, FLIM and FRET. Proc. Royal Mic. Soc., 39:3-10, 2004.