PRACTICAL MANUAL FOR

FLUORESCENCE
MICROSCOPY
Sohail Ahmed
Sudhaharan Thankiah
Radek Machán
TECHNIQUES Martin Hof
Andrew H. A. Clayton
Graham Wright
Jean-Baptiste Sibarita
Thomas Korte
Andreas Herrmann
chapter

6
 Fluorescence Recovery After
Photobleaching (FRAP)
Graham Wright1 and Jean-Baptiste Sibarita2,3
Institute of Medical Biology, A*STAR, Singapore
1

² University of Bordeaux, Interdisciplinary Institute for Neuroscience,

Bordeaux, France
3
CNRS UMR 5297, Bordeaux, France

Index
Abstract���������������������������������������������������������������������������������������������������������������������������������� 3
1. Principle of FRAP��������������������������������������������������������������������������������������������������������������� 3
2. Qualitative Determination of Protein Dynamics������������������������������������������������������������������ 5
3. Models for Quantification of Diffusion and Chemical Exchange ���������������������������������������� 6
Diffusion������������������������������������������������������������������������������������������������������������������������� 6
Point Bleaching with Gaussian Profile (Axelrod Model)������������������������������������������������� 6
Point Bleaching with Rectangular Profile (Soumpasis Model)��������������������������������������� 7
Line Bleaching���������������������������������������������������������������������������������������������������������������� 7
Chemical Interaction������������������������������������������������������������������������������������������������������� 7
4. Methods – FRAP Experiments������������������������������������������������������������������������������������������� 7
Instrumentation��������������������������������������������������������������������������������������������������������������� 7
Data Acquisition�������������������������������������������������������������������������������������������������������������� 8
Data Processing Prior to Quantification����������������������������������������������������������������������� 10
5. Conclusion������������������������������������������������������������������������������������������������������������������������ 10
Acknowledgements�������������������������������������������������������������������������������������������������������������� 10
Further Reading��������������������������������������������������������������������������������������������������������������������11
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-3

Abstract membrane diffusion, protein interactions and protein

dynamics[3-12].
Fluorescence recovery after photobleaching (FRAP) Photobleaching is a natural phenomenon that man-
is a microscopy technique capable of quantifying the ifests itself as decreasing intensity of fluorescence
mobility of molecules within cells. By exploiting the over time during fluorescence imaging. Exposing
phenomenon of photobleaching, fluorescent mole- a fluorophore to a high level of light intensity in the
cules within a region of interest can be selectively presence of molecular oxygen causes permanent
and irreversibly ‘turned off’. The analysis of the flu- and irreversible chemical changes to that molecule,
orescence recovery within the same region, due the rendering it non-fluorescent[11],[13],[14]. Under a constant
redistribution of the molecules, provides information absorption of light, the fluorescence intensity will
on their diffusion- and binding-dependent mobility. decrease over time following an exponential decay
Both qualitative and quantitative analysis can then law:
be applied to decipher the dynamic behavior of the
I (t )=I 0 e−Kt
molecules of interest.

where I0 represents the initial fluorescence intensity

1. Principle of FRAP and K the bleaching rate constant of the fluorophore
including the flux of illumination photons.
Fluorescence recovery after photobleaching (FRAP) Generally, photobleaching is considered a problem
is a popular fluorescence microscopy technique used for time-lapse and 3D imaging, leading to unwanted
to quantify the mobility of molecules within cells. The loss of the signal and resultant degradation of the
mobility is determined by the molecules’ proper- signal-to-noise ratio during acquisition. In FRAP ex-
ties of transport, diffusion and binding to immobile periments, however, the photobleaching phenome-
sites. Since the initial development by Axelrod et al.[1] non is exploited to selectively ‘turn off’ a subset of the
and Peters et al.[2] in the 1970’s, the technique has fluorescent molecules in the sample usually at a spot
been widely used in biological research to study cell or in a specific area of interest using a short pulse of

I( t)= I_0 func e^{-Kt}

Figure 1 Principle of FRAP experiments. A A cell or organelle is uniformly labeled with a fluorescent tag and a pre-bleach series of images
collected. B A ROI is selectively photobleached using a short pulse of intense laser light. C A post-bleach recovery time series of images is
collected and the intensity within the ROI monitored as the bleached dye diffuses out and new dye diffuses in. D Intensity changes within the
ROI are measured, corrected, normalized and plotted on a graph for further quantification.
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-4

high intensity laser light that can be positioned to, or in duration, is performed. When appropriate, this
scanned over, the region of interest (ROI). Monitoring can be done in up to 3 phases with a series of
the recovery of intensity in the bleached ROI yields fast acquisitions to cover the fast dynamics of
information on protein mobility (Figure 1). recovery, followed by phases of less frequent
The choice of fluorophore is an important consider- imaging as the recovery continues. Again, it is im-
ation for all fluorescence imaging based experiments, portant to reduce the amount of photobleaching
and FRAP is no exception. Ideally it should be stable caused by the imaging itself to a minimum (Figure
enough to undergo minimal photobleaching during 1C).
the imaging phases - the pre- and post-bleach acqui- 4. Data analysis: Following correction and normal-
sitions – but bleach quickly and permanently during ization steps, the intensity fluctuations during the
selective ROI photobleaching. Genetically encoded recovery phase within the ROI are analyzed quan-
fluorescent proteins, like GFP, are generally used to titatively (Figure 1D). This allows the extraction
label a molecule of interest with very high specifici- of the recovery rate as a measure for molecular
ty[3],[14],[15]. They provide enormous power and scope mobility as well as mobile and immobile fractions.
to study biomolecules in living cells through transient Advanced fitting using biophysical models can be
expression or through producing stable transgenic used under certain, precisely defined boundary
cell lines. Importantly, it must be taken into consider- conditions, to quantify the diffusion coefficient or
ation that these modifications may change the prop- chemical exchange rate.
erties of the tagged molecule and it is noteworthy
that high intensity illumination creates free radicals FRAP can be used qualitatively to identify whether a
which are highly reactive and thus cytotoxic[16]. The particular molecule is turning over, thereby undergo-
induced photodamage may affect cell viability and ing exchange with its environment, by basic analysis
result in artifactual results. It is therefore important of the recovery curve. It allows the determination of:
to minimize the extent of bleaching, even during the 1. the half-time of recovery (t1/2), and
bleaching step. 2. the mobile (M) and immobile (1-M) fractions
In practice, a FRAP experiment is a 3 step acquisi- (Figure 4).
tion process, followed by data analysis, as follows
(also refer to Figure 1)[11],[17],[18]: More advanced quantitative models allow the precise
1. Pre-Bleach: A time-lapse acquisition containing determination of the kinetics and molecular proper-
the cell or sample of interest and ideally some ties driving such dynamics. Fitting the experimental
empty background area. It is important to tune curve with advanced theoretical models allows the
the acquisition settings to reduce the amount of determination of additional parameters[19-27]:
photobleaching resulting from the imaging itself 1. the ratio between mobile (M) and immobile
(Figure 1A) (1-M) fractions (Figure 4),
2. Bleach: A ROI is selectively photobleached 2. the effective diffusion coefficient (D),
using a short pulse of intense laser light, the in- 3. the binding time of proteins to sufficiently
tensity modulation is typically controlled through immobile macromolecular complexes (kon, koff),
an acousto-optical tunable filter (AOTF). The and
ROI may be a single focused spot, or the laser 4. the interconnection of intracellular organelles.
beam may be scanned (in a raster or a whirlwind
pattern) over the area through the action of gal- A closely related technique to FRAP is Fluorescence
vanometric-mirrors. To avoid damaging the cell, Loss in Photobleaching (FLIP[11]; Figure 2). Again the
just a fraction of the overall pool of fluorescent experiment consists of a pre-bleach acquisition, but
molecules should be turned off (Figure 1B) then a ROI in the cell is repeatedly bleached whilst
3. Recovery: A time-lapse acquisition with similar a second ROI is analysed for the consequent loss of
parameters as the pre-bleach phase, but longer fluorescence. This technique gives information about

Figure 2 Principle of FLIP experiments. A cell or organelle is uniformly labeled with a fluorescent tag and a pre-bleach series collected.
A ROI is selectively photobleached using a short pulse of intense laser light repeatedly whilst a second ROI is analysed for the loss of
fluorescence.
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-5

Figure 3 Principle of iFRAP experiments. A cell or organelle is uniformly labeled with a fluorescent tag and a pre-bleach series collected. A
ROI is selected and the area outside of it photobleached using a scanned pulse of intense laser light. The ROI is then analysed for the loss of
fluorescence over a post-bleach acquisition. Due to the high likelihood of inducing phoodamage on live cells, this technique has largely been
replaced by photoactivation-type experiments.

molecules mobility and interconnection between 2. Qualitative Determination of

cellular compartments. It will not be covered further Protein Dynamics
within the scope of this chapter. This technique has
to be used with caution to avoid photodamage due to It is common, and quite straightforward, to charac-
repeated bleaching steps. terize molecule dynamics from FRAP experiment
iFRAP (inverse-FRAP; Figure 3[28]) consists of by the half-time of recovery (t1/2) and the mobile (M)
bleaching everything but the ROI, essentially the and immobile (1-M) fractions. Even if it has no direct
reciprocal experiment of FRAP. The sample’s fluo- relation with biophysical parameters, they provide
rescence, save for a small area, is photobleached a general semi-quanitative estimate on molecule
and the analysis concentrates on the loss of flu- dynamics and can be used to compare various bi-
orescence from the ROI, rather than the recovery. ological conditions. t1/2 gives information on the
This technique is typically more damaging to the average dynamics of moving molecules, whereas
sample due to the large areas that are exposed M quantifies the fraction of molecules which are
to high laser power. As a result it has been largely moving and 1-M describes the fraction of immobile
replaced by photoactivation (PA) experiments since molecules within the bleached area during the exper-
the discovery of photoactivable GFP and develop- iment. Immobile molecules strongly interact with a
ment similar proteins[13],[29-31]. With PA experiments structural component of the cell or be ‘trapped’ within
the fluorescence can be selectively “turned on” in an a multi-component protein complex, preventing them
ROI through a pulse of shorter wavelength light (typ- from moving away from the bleached ROI.
ically 405 nm). These techniques will not be covered From the recovery curves, t1/2 can easily be extract-
further within the scope of this chapter. ed (see Figure 4), provided the post-bleach recovery
imaging segment was sufficiently long and appro-
priate intensity corrections have been performed.
This value needs to be used with caution, since the

Figure 4 Recovery curve of a FRAP experiment and determination of half-time of recovery (t1/2), mobile (M) and immobile (1-M) fractions.
 Unbenannt
1

6 Fluorescence Recovery After Photobleaching (FRAP) | 6-6

cell geometry and bleached area properties (size, D=

kT
position with respect to the cell) can strongly influ- 6 πη R
ence t1/2 in addition to the molecules’ behavior[11].
The photobleaching depth B is given by the fraction Where k is the Boltzmann constant, T the tempera-
between the remaining signal and the original signal ture, η the viscosity and R the hydrodynamic radius
in the bleached ROI. It is given by of the molecule.

B=
I 1−I 0 In most FRAP experiments, D is defined as
I1
w²
D=
This fraction is important for quantitative analysis 4 τD
Unbenannt
1
and has to be less than 80% in practice.
The mobile fraction is given by where w is the waist of the bleached area, τD a char-
acteristic time constant extracted from mathematical
M=
I ∞−I 0 model fitting, and n the number of spatial dimen-
I 1−I 0 sions. In general τD, has no direct relation with t1/2
defined in Figure 4.
and gives information on proteins that are mobile
or interact transiently with immobile binding sites
Unbenannt
1
Point Bleaching with Gaussian Profile (Axelrod
during the observation time of the experiment. Again, Model)
caution needs to be taken since M may depend on It is possible to extract the diffusion coefficient by
acquisition parameters and bleaching dimensions. single-point
D =bleaching
{kT} over with a Gaussian
{6 πη profile,
R} using
the Axelrod model[1] (Figure 5). Recovery curves are
reconstructed by averaging the pixel intensity values
3. Models for Quantification of within a circular region of interest of w in diameter,
w being the waist of the laser at 1/e² (13.5%) of the
Diffusion and Chemical Exchange peak intensity of the Gaussian. Using imaging infor-
mation, the recovery sequence can be corrected by
Molecular mobility is mainly due to diffusion, flow normalizing mean pixel intensities in another ROI
B= {I_1
or chemical -I_0}
reaction over {I_1} to an
(association/dissociation located far from the bleaching area. This step makes
immobile molecular complex). The general equation it possible to take into account intensity fluctuations
of the fluorescence recovery is given by due toDobservational
= {w²} over {4%tau_D}
photobleaching or laser instabil-
ities. Once corrected, the recovery curves are fitted
∂ c(r , t)
=D ∇ 2 c(r ,t )−V
∂ c(r , t)
+(k off −k on )c(r , t ) with a 10th order limited development of the following
∂t ∂r equation.

The first term describes the diffusion, the second the

[ ][ )]
−1
M= {I_ infinity -I_0} over {I_1-I_0} (−K )n
(
−K ∞
1−e 2t
flow and the last term the association/dissociation F (t)=(1−M ) +M ∑ 1+n 1+ τ
K n!
processes. Unfortunately, this equation has no an- n=1

alytical solution, and it is easier to investigate each

contribution individually. It therefore may be neces- where M is the mobile fraction (accounting for the
sary to perform multiple experiments, bleaching for ability of the molecule to diffuse during the duration of
example areas with different sizes. the experiment), K is a bleaching constant parameter
and is the characteristic diffusion time. The diffusion
Diffusion coefficient D can be correctly estimated as follows:
The diffusion coefficient of a molecule is given by the
Stokes-Einstein relation:

(r,t)} over {∂t}=

Figure 5 Schematic D nabla^2
and model c( with
of point bleaching r,t)-V {∂c(r,t)}
F(a Gaussian
t )=( profile over over {K}+M sum from{n=1} to
1-Mand) recovery.
{1-e^-K}
} + ( k_off - k_on) c (r,t) infinity } left [ ( -K)^n over {n!}right ] left [ 1+n left (
1+ {2t} over { %tau }right ) right ]^-1
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-7

Unbenannt
1
w² Chemical Interaction
D= FRAP recovery curves of chemical reactions are the
4τ
simplest case, since they can be solved by the fol-
K can be obtained by fitting the following laser inten- lowing single exponential recovery function:
sity distribution on the bleaching profile:
1 k off
r² I (t )= A (1−e−t / τ ) with τ= and A=
− k on +k off k on + K off
I (r , t=0)=I 0 e−KB (r) with B(r)=B (0)e 2w ²

where kon and koff are the association and dissocia-

Point Bleaching with Rectangular Profile tion rates of the bleached molecules to their ligand.
(Soumpasis Model) In practice, this case is very rare in living cells, and
Bleaching with a rectangular profile (Figure 6) allows diffusion always plays a role[20]. Nevertheless, it is im-
avoiding the complexity to measure the bleaching portant to notice that in the case of pure binding, the
Unto
constant K required be na
solven nt
the 2
Axelrod equation[1],[32]. recovery rate is independent of the bleaching radius,
In this case, the recovery curve can be fitted with the while it is a function of w in the case of Brownian
following equation: diffusion.
−2 τ
I (t )=e t
(J 0
2τ
t
+J1
2τ
t ) 4. Methods – FRAP Experiments
Where J is the Bessel function and τ = w² / 4D. Instrumentation
The hardware required for FRAP experiments com-
Line Bleaching prises a fluorescence microscope equipped with light
The Soupmasis model requires shaping of the sources (arc lamps, LEDs or lasers) and filter sets for
beamD= with {
a square profile,
w²} over {4%tau} which is not commonly imaging as well as a light source for bleaching (typi-
available on standard FRAP equipments. A simpler cally lasers) with some A={methodk_off} over
of selectively {k_on
bleach- + K_
t=0)= solution consists of
I_0 e^ {-KB(r)} [33],[34] performing a line bleaching
with B( r)=B(0)e^-{{r²} with ing a region of interest (ROI).
over 1} over {k_on+k_off}Of course, a sample
a Gaussian beam profile (Figure 7). %tau={
I(t)= A(labeled with a fluorescent molecule attached to the
1-e^{-t/%tau})
w²}} In this case, the recovery curve can be fitted by the protein of interest is required and is one of the chal-
following equation, a 1D approximation of the diffu- lenges the biologist is facing. The fluorophore used
sion process: should be selected both for its spectral properties in
order to match the laser lines and filters available,
( √
I (t )=I ∞ 1−
w²
w ²+4 π Dt ) as well as for having favourable photophysical prop-
erties, i.e., negligible bleaching probability at low il-
lumination intensities and high bleaching probability
This model displays an accuracy of about 30%. at high intensities. As most biological experiments
involve living cells, there is a prerequisite for the
{-2%tau} over {t}} left ( J_0 {2%tau} over {t}

over {t}right )

Figure 6. Schematic and model of point bleaching with a rectangular profile and recovery.

)= I_infinity left (1- sqrt{ {w²} over {w²+4 %pi Dt} }

ght )

Figure 7. Schematic of line bleaching and recovery.

 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-8

Figure 8 Photographs of an implementation of the Roper iLas2 FRAP-3D system at the Institute of Medical Biology, A*STAR,
Singapore. The microscope stand is an inverted Nikon Eclipse Ti equipped with a Yokogawa CSU-22 spinning disk confocal head and
a liquid-cooled Photometrics Evolve EM-CCD camera. It is operated through the MetaMorph and iLas2 acquisition software. The lasers
(405/491/561 nm) inside the laser launch are used for both imaging and FRAP, with an AOTF to rapidly control the intensity of laser
reaching the sample, and a galvo mirror to rapidly switch between the spinning disk (mounted on the left of the microscope) and FRAP
light paths (mounted at the rear of the microscope).

Figure 9 Establishing and refining the imaging conditions in the MetaMorph software through selection of the correct camera mode,
objective lens, filter combination, laser power, camera gain and exposure conditions appropriate for the sample. Photobleaching and
photodamage, as a consequence of tuning these imaging conditions, should be reduced to a minimum – lower laser power and shorter
exposures, whilst maintaining sufficient dynamic range in the images for quantification and imaging frequency for sampling of the
recovery.

microscope to be equipped with an incubator and a nation light-path are independent from one another,
supply of humidity and CO2, as dictated by the cell offering very fast switching between, or even simulta-
type. neous, photobleaching and imaging. Typically, such
The most common way to achieve selective photo- systems will have some calibration routine to cor-
bleaching is to use galvanometer-driven mirrors to relate a given pixel’s position with the corresponding
steer the laser beam, which is momentarily switched galvo position. Some of the CLSM manufacturers
to a higher intensity, to a diffraction limited spot or to have also developed systems with dual scanners to
a scanned region (in raster or whirlwind pattern) over enable this (e.g. the SIM scanner on the Olympus
a pre-selected ROI. For this reason, many FRAP ex- FV1000 and FV1200 confocal systems).
periments have been carried out on confocal laser The short-lived increase in the laser intensity for pho-
scanning microscopes (CLSMs) as they are equipped tobleaching is achieved either by using an acous-
with high-power lasers and galvo mirrors [35-37]. FRAP to-optical tunable filter (AOTF), by modulating the
scanning heads are now also commercially available laser output power directly, or by using a fast switch-
for widefield fluorescence systems (e.g. DeltaVision ing mirror to steer the beam into an independent
Elite from GE), spinning disk confocal or TIRF light-path with reduced attenuation.
systems (e.g. iLas2 FRAP-3D from Roper (Figure
8), UltraView VoX from PerkinElmer and Revolution Data Acquisition
XD from Andor). In these setups the microscope’s Acquisition and photo-perturbation parameters need
excitation illumination light-path and the FRAP illumi- to be carefully adjusted in order to minimise errors
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-9

Figure 10 Establishing and refining the FRAP conditions in the iLas2 module through selection of the appropriate mode of operation
(On-Fly or MDA), ROI shape, size and location, duration/repetitions of bleach and laser intensity to be used, as well as the temporal
frequency and number of images in the pre- and post-bleach acquisition steps. They have to be adjusted according to the protein
dynamics and cell and fluorophore stability. This data, once established, is then sent back to the Multi Dimensional Acquisition tool in
MetaMorph.

in quantification and interpretation[27]. This is a rule until t = t1/2, then the acquisition frequency can be
of thumb for qualitative and quantitative FRAP ex- reduced to avoid observational photobleaching).
periments. Prior to carrying out a FRAP experiment, • recovery sequence duration must be about 10
one should perform empirical tests to establish time- times longer than t1/2.
lapse image acquisition settings that minimise photo-
bleaching and match protein dynamics. This usually When the size of the bleaching ROI affects the half
requires a compromise between sufficient dynamic time recovery, which is the case for example for diffu-
range for quantification and sufficient imaging speed sional or flow-induced mobility, it can be adjusted for
to properly sample the recovery dynamics. Acquisition the rule of ten to hold (e.g., bleaching a larger area
frequency and photo-perturbation duration need to in a fast mode will increase the t1/2, allowing slower
be adjusted according to the recovery speed as well. acquisition frequency).
This will involve selection of the appropriate objec- In the following example, we show the data acqui-
tive lens, fluorescence filter set, laser power, camera sition parameters that require consideration for the
settings, exposure duration (or corresponding iLas2 FRAP-3D system (Roper, France), integrated
confocal acquisition settings) and imaging frequen- onto the MetaMorph acquisition platform (Molecular
cy according to the sample preparation and protein Devices; Figure 9). Whilst the considerations are
dynamics (Figure 9). Subsequent empirical tests similar on other hardware/software platforms, the
should be carried out to tune the FRAP settings like terminology and implementation may differ.
the ROI shape, size and location, bleaching duration Once the appropriate imaging and FRAP conditions
and laser intensity (Figure 10). It is important that the are established, single or multiple ROIs are selected
duration of the photobleaching step should be short in a preview image and the iLas2 software sends the
enough to prevent any molecules from entering or data as a journal (macro) to the MetaMorph Multi
leaving the ROI during the bleach. In addition, ac- Dimensional Acquisition (MDA) tool. This handles
quisition of the post-bleach sequence must be fast the complex time-lapse acquisition routine estab-
enough to capture the fluorescence recovery. Finally, lished in the iLas2 window. When the time resolu-
the delay between the end of the bleaching and tion of the experiment is critical the FRAP-On-Fly
the beginning of the recovery must be as short as method can be used instead of the MDA. In this case
possible. In a general manner, one can use the rule a stream acquisition is setup in which the hardware
of ten: is pushed to its limits to maximize frame-rate. During
• bleaching duration must be at least 10 times the acquisition, the user can then manually select the
faster than the half time of recovery (t1/2). position of the FRAP ROI with the mouse whilst the
• delay between the end of the bleaching step and stream acquisition continues. The frame rate is then
the beginning of the recovery sequence must be limited by the exposure time and read-out speeds of
shorter than a 10th of t1/2. the camera/computer system.
• acquisition frequency of the recovery sequence
must be at least 10 times faster than t1/2 (at least
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-10

Figure 11 Image correction prior quantification. All the information for a given cell is used to measure the cell fluorescence eliminated
by bleaching and to compensate for observational bleaching over time throughout the experiment. A Three distinct ROI are defined: z1
is the region targeted by the laser bleaching pulses; z2 includes the whole cell and is assumed to contain a fixed number of fluorophores,
regardless of whether those fluorophores are bleached or fluorescent; z3 is defined for the estimation of background level, mostly due
to the CCD dark signal. B The average level over z3 is subtracted from all the images. Signal levels Z1 and Z2, in regions z1 and z2, are
analysed after background subtraction. They are processed to account for observational bleaching and for the limitation of recovery in z1
by the total loss over z2.

Data Processing Prior to Quantification recovery curves can be analyzed following the pro-
It is of major importance to process the acquired cedure introduced earlier.
data before quantification in order to correct for ob-
servational photobleaching and avoid artifacts in the
quantification[27]. Image sequences can be correct- 5. Conclusion
ed for observational photobleaching using the fact
that a closed volume (Z2 in Figure 11A) contains a FRAP is carried out on microscopes equipped with
finite number of fluorescent molecules[5]. Changes components that allow the user to selectively expose
in the average intensity inside this volume over time regions of interest (ROI) to intense pulses of laser
results from both bleaching pulses and observa- light. This is usually possible on research grade mi-
tional photobleaching, following a first-order decay croscopes such as confocal instruments. In doing so
with time constant τ. Observational photobleaching the fluorescent molecules (usually proteins of interest
can be assessed with Z2 before (t<t0) and after the tagged with fluorophores like GFP) within that region
pulse (t>t0), and corrected by multiplying Z2 by e(t−t0 are photobleached into an ‘off’ state. By subsequent-
)/τ
. In this normalization process, the step points A ly analysing that ROI the dynamics of the recovery
and B are fixed points, and the normalized Z2 curve of that fluorescence can be used to give a greater
(second plot) has two constants: Z2(t<t0) =A and understanding of the molecule being studied like:
Z2(t>t0) =B. (1) the half-time of recovery, (2) the ratio between
Let α be the ratio Z1/Z2, with Z1 the average inten- mobile and immobile fractions, (3) the effective dif-
sity over the volume z1 (see Figure 11A). Before fusion coefficient, (4) the binding time of proteins to
the bleaching pulse, at t<t0, α remains constant at immobile macromolecular complexes, and (5) the
steady state. Z1 (light grey) and the difference signal interconnection of intracellular organelles and the
ZR=Z2-Z1 over the complementary region (dark grey) existence of protein complexes.
therefore both decay like Z2. Consequently, Z1 and
ZR are constant before the pulse on the normalized
graph. Immediately after the photobleaching of z1, Acknowledgements
at t=t0, Z1/Z2 no longer equals α, but instead falls to
α−∂Z1 / Z2. If full recovery to the initial steady state Thanks to Emma Feng Yu (microLAMBDA Pte
occurs during the experiment, this proportion then Ltd, Singapore) for help and advice with the iLas2
returns to α1, and Z1 recovers at the expense of ZR. and MetaMorph software. Also, we thank Malte
With the exception of very limited bleached areas Wachsmuth for the critical reading and advice on the
and/or bleaching depths, full recovery in z1 is attenu- preparation of this manuscript.
ated by a loss factor B/A. Recovery curves are thus
compared to their asymptotic limit Z1(B/A), so that
the mobile fraction can still be measured correctly. To
enable comparisons between multiple experiments a
normalisation step is typically carried out.
Once the data are processed and normalized,
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-11

Further Reading
[1] Axelrod, D., et al., Mobility measurement by [16] Magidson, V. and A. Khodjakov, Circumventing
analysis of fluorescence photobleaching recovery photodamage in live-cell microscopy. Methods Cell
kinetics. Biophys J, 1976. 16(9): p. 1055-69. Biol, 2013. 114: p. 545-60.
[2] Peters, R., et al., A microfluorimetric study of [17] Swift, S.R. and L. Trinkle-Mulcahy, Basic prin-
translational diffusion in erythrocyte membranes. ciples of FRAP, FLIM and FRET. Proc Roy Microsc
Biochim Biophys Acta, 1974. 367(3): p. 282-94. Soc, 2004. 39: p. 3-10.
[3] White, J. and E. Stelzer, Photobleaching GFP [18] Terjung, S. and R. Pepperkok, FRAP-teaching
reveals protein dynamics inside live cells. Trends module, in European Advanced Light Microscopy
Cell Biol, 1999. 9(2): p. 61-5. Network (EAMNET). 2005.
[4] Snapp, E.L., N. Altan, and J. Lippincott- [19] Sprague, B.L. and J.G. McNally, FRAP analysis
Schwartz, Measuring protein mobility by photo- of binding: proper and fitting. Trends Cell Biol, 2005.
bleaching GFP chimeras in living cells. Curr Protoc 15(2): p. 84-91.
Cell Biol, 2003. Chapter 21: p. Unit 21 1. [20] Sprague, B.L., et al., Analysis of binding reac-
[5] Phair, R.D. and T. Misteli, High mobility of tions by fluorescence recovery after photobleaching.
proteins in the mammalian cell nucleus. Nature, Biophys J, 2004. 86(6): p. 3473-95.
2000. 404(6778): p. 604-9. [21] Bulinski, J.C., et al., Rapid dynamics of the
[6] Lippincott-Schwartz, J., T.H. Roberts, and K. microtubule binding of ensconsin in vivo. J Cell Sci,
Hirschberg, Secretory protein trafficking and organ- 2001. 114(Pt 21): p. 3885-97.
elle dynamics in living cells. Annu Rev Cell Dev Biol, [22] Lambert, N.A., Uncoupling diffusion and
2000. 16: p. 557-89. binding in FRAP experiments. Nat Methods, 2009.
[7] Houtsmuller, A.B., et al., Action of DNA repair 6(3): p. 183; author reply 183-4.
endonuclease ERCC1/XPF in living cells. Science, [23] Sullivan, K.D., A.K. Majewska, and E.B. Brown,
1999. 284(5416): p. 958-61. Single- and two-photon fluorescence recovery after
[8] Houtsmuller, A.B. and W. Vermeulen, photobleaching, in Imaging: A Laboratory Manual.
Macromolecular dynamics in living cell nuclei 2011, Cold Spring Harbor Press: Cold Spring
revealed by fluorescence redistribution after pho- Harbor.
tobleaching. Histochem Cell Biol, 2001. 115(1): p. [24] Gordon, G.W., et al., Analysis of simulated and
13-21. experimental fluorescence recovery after pho-
[9] Klonis, N., et al., Fluorescence photobleach- tobleaching. Data for two diffusing components.
ing analysis for the study of cellular dynamics. Eur Biophysical Journal, 1995. 68(3): p. 766-778.
Biophys J, 2002. 31(1): p. 36-51. [25] Phair, R.D. and T. Misteli, Kinetic modelling ap-
[10] Hoogstraten, D., et al., Rapid switching of proaches to in vivo imaging. Nat Rev Mol Cell Biol,
TFIIH between RNA polymerase I and II transcrip- 2001. 2(12): p. 898-907.
tion and DNA repair in vivo. Mol Cell, 2002. 10(5): p. [26] Lippincott-Schwartz, J., E. Snapp, and A.
1163-74. Kenworthy, Studying protein dynamics in living cells.
[11] Bancaud, A., et al., Fluorescence perturba- Nat Rev Mol Cell Biol, 2001. 2(6): p. 444-56.
tion techniques to study mobility and molecular [27] Weiss, M., Challenges and artifacts in quantita-
dynamics of proteins in live cells: FRAP, photoac- tive photobleaching experiments. Traffic, 2004. 5(9):
tivation, photoconversion, and FLIP. Cold Spring p. 662-71.
Harb Protoc, 2010. 2010(12): p. pdb top90. [28] Rabut, G. and J. Ellenberg, Photobleaching
[12] Reits, E.A. and J.J. Neefjes, From fixed to techniques to study mobility and molecular
FRAP: measuring protein mobility and activity in dynamics of protiens in live cells: FRAP, iFRAP
living cells. Nat Cell Biol, 2001. 3(6): p. E145-7. and FLIP, in Live-cell Imaging: A Labortory Manual,
[13] Lippincott-Schwartz, J., N. Altan-Bonnet, and R.D. Goldman and D.L. Spector, Editors. 2005,
G.H. Patterson, Photobleaching and photoactiva- Cold Spring Harbor Press: Cold Spring Harbor. p.
tion: following protein dynamics in living cells. Nat 101-127.
Cell Biol, 2003. Suppl: p. S7-14. [29] Lippincott-Schwartz, J. and G.H. Patterson,
[14] Lippincott-Schwartz, J. and G.H. Patterson, Fluorescent proteins for photoactivation experi-
Development and use of fluorescent protein ments. Methods Cell Biol, 2008. 85: p. 45-61.
markers in living cells. Science, 2003. 300(5616): p. [30] Patterson, G.H., Photoactivation and imaging
87-91. of photoactivatable fluorescent proteins. Curr Protoc
[15] Weber, W., et al., Shedding light on the dark Cell Biol, 2008. Chapter 21: p. Unit 21 6.
and weakly fluorescent states of green fluorescent [31] Patterson, G.H. and J. Lippincott-Schwartz, A
proteins. Proc Natl Acad Sci U S A, 1999. 96(11): p. photoactivatable GFP for selective photolabeling
6177-82. of proteins and cells. Science, 2002. 297(5588): p.
 6 Fluorescence Recovery After Photobleaching (FRAP) | 6-12

1873-7.
[32] Soumpasis, D.M., Theoretical analysis of
fluorescence photobleaching recovery experiments.
Biophys J, 1983. 41(1): p. 95-7.
[33] Ellenberg, J., et al., Nuclear membrane
dynamics and reassembly in living cells: targeting
of an inner nuclear membrane protein in interphase
and mitosis. J Cell Biol, 1997. 138(6): p. 1193-206.
[34] Braeckmans, K., et al., Line FRAP with the
confocal laser scanning microscope for diffusion
measurements in small regions of 3-D samples.
Biophys J, 2007. 92(6): p. 2172-83.
[35] McNally, J.G. and C.L. Smith, Photobleaching
by confocal microcopy, in Confocal and two photon
microscopy, A. Diaspro, Editor. 2001, Wiley-Liss:
New York. p. 525-538.
[36] Hauser, G.I., S. Seiffert, and W. Oppermann,
Systematic evaluation of FRAP experiments
performed in a confocal laser scanning micro-
scope--part II: Multiple diffusion processes. J
Microsc, 2008. 230(Pt 3): p. 353-62.
[37] Seiffert, S. and W. Oppermann, Systematic
evaluation of FRAP experiments performed in a
confocal laser scanning microscope. J Microsc,
2005. 220(Pt 1): p. 20-30.
PicoQuant GmbH
Rudower Chaussee 29
12489 Berlin
Germany

Material can be downloaded from www.picoquant.com

© 2016 PicoQuant GmbH

All rights reserved. No parts of it may be reproduced, translated or transferred to third parties without written
permission of PicoQuant GmbH.

www.picoquant.com