FRAP Tutorial

EMBO course, Debrecen August 2011

Stefan Terjung
www.embl.de/almf
Overview
• Introduction
• Application examples and related techniques
• Instrument setups for FRAP
• Basic FRAP analysis procedures
Molecular Dynamics and Interactions

FRAP
FCS
Lightmicroscopy
FRET
1Å 1 nm 1 µm 1 mm 1 cm 1m
10-10m 10-9m 10-6m 10-3m 10-2m

Atoms organic Virus Bacteria, Organism

Molecules Organelles

Biopolymers eucaryotic Cells

Brief FRAP history
1970s: 1st applications of the FRAP method
mathematics for Gaussian beam profile bleaching
(Poo & Cone, Axelrod et al.)
 Custom built systems, mainly diffusion in membranes
1980s: first commercial confocal microscopes
1990s: revival of FRAP using GFP and confocal microscopes with
ROI-scanning (AOTF)
(Tsien, Cole et al., Lippincott-Schwartz..)
 Commercial confocal systems, virtually any protein fused
to a fluorescent protein
2000s: Computer modeling for quantitative FRAP analysis
Improvements of microscope systems for FRAP
Questions addressed by FRAP
• How fast do proteins move ?

• Which compartments are connected ?

• How long do proteins bind other proteins ?

• What is the percentage of proteins ‘immobilized’ by binding

to larger structures (e.g. DNA, Actin) ?

• How do drugs or mutations alter the binding of a protein ?

• How fast are molecules transported between compartments

(e.g. nuclear import) ?
 Molecular Dynamics

diffusion speed ?

exchange
rate ?

tsO45G-YFP traffic in HeLa cell

1. What is FRAP?

Fluorescence Recovery After Photobleaching (FRAP)

Let’s think of fluorescence molecules dispersed in a field. White circles

represent the molecules.
We focus a strong laser beam to a spot (red dotted circle).

Kota Miura
1. What is FRAP?

Fluorescence Recovery After Photobleaching (FRAP)

Then the strong irradiation BLEACHES the fluorescence at that spot.

Kota Miura
1. What is FRAP?

Fluorescence Recovery After Photobleaching (FRAP)

Since molecules are moving driven by diffusion or active transport,

bleached molecules exchange their place with un-bleached molecules.

Then the average intensity at the bleached spot recovers.

Kota Miura
 Scheme of a FRAP experiment
Mean value
inside the
bleaching ROI
over time

POSTBLEACH
PREBLEACH

FRAP monitoring ROI

-0.101s -0.051s -0.001s 0.000 s 0.001s 0.051s 0.101s 0.151s 0.201s 0.251s
Free diffusion vs. binding

Phair and Mistelli, Nature Reviews MolCellBio, 2001 Lippincott-Schwartz et al. Nature CellBio Supp. 2003

Multiple populations with differing diffusion rates

 multi-component equations
 The ‘Ideal’ FRAP Acquisition

…
T -0.101s T 0.000s T 0.001s T 10.001s
T -0.051s T 0.051s
T -0.001s T 0.101s

Prebleach Instantaneous Postbleach Series

Series Bleach pulse (“normal time-lapse” starting
(“normal time-lapse”) without delay after the bleach)
>20 data points until half-recovery
FRAP – basic parameters

‚immobile‘ fraction
intensity

‚mobile‘ fraction

Half-time of
equilibration (t½ )

time

Studying protein-protein interactions by advanced light microscopy and spectroscopy

EMBO course Debrecen, August 2011
 Bleaching
• The amount of fluorophore bleaching is dependent on the excitation
power
• Bleaching should be avoided during acquisition, but should be sufficient
and ‘instantaneous’ for the bleach pulse of a FRAP experiment
• Bleaching can also cause photodamage
 bleach only to ~50% of initial intensity
 repeat bleaching at same spot to check for differences due to photodamage)
Bleaching depends on excitation intensity
Fluorophore Saturation
1
0.9
0.8
Saturation

0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.00E+00 3.00E+24 6.00E+24 9.00E+24 1.20E+25 1.50E+25

Excitation Photons
FRAP conditions depend on the sample
• The bleaching ROI has to be matched to the protein/structure to
FRAP

Circle Half organelle whole organelle

e.g. cytoplasmic Protein e.g. nuclear protein e.g. protein import

• Conditions are highly instrument dependent (laser power, % AOTF,

PMT Voltage…) and cannot (easily) be compared between systems,
but should be kept constant on one system to compare different
conditions.
Controls

• Calibrate bleached volume in x,y and z

• Check dependence of recovery rate on different

bleaching power (cross-linking ?)

• Repeat FRAP on same spot (difference due to

photodamage ?)

• Dependence of recovery on ROI size ?

• Compare wiltype with mutant (e.g. non-binding mutant)

• Check for dark states / reversibility

Studying protein-protein interactions by advanced light microscopy and spectroscopy

EMBO course Debrecen, August 2011
Overview
• Introduction
• Application examples and Related techniques
• FLIP: Fluorescence Loss In Photobleaching
• iFRAP: inverse FRAP
• Photoactivation (paGFP…)
• Photoconversion (Kaede, Dendra, EOS…)

• Instrument setups for FRAP

• Basic FRAP analysis procedures
FRAP: Comparison of 2 conditions
ER-export with and without sterol depletion
Runz H et. al 2006:
Sterols regulate ER-
export dynamics of
secretory cargo
protein ts-O45-G.
EMBO Journal 25:
2953-2965
Fluorescence Loss in Photobleaching (FLIP)

Phair and Mistelli, Nature Reviews MolCellBio, 2001

 FLIP

• Probing organelle continuity by repetitive bleaching

GFP in the ER
Snapp E, Altan N, Lippincott-Schwartz J (2003) Measuring protein mobility by photobleaching
GFP chimeras in living cells. In Current Protocols Cell Biology, 21. Wiley
FRAP-FLIP
Phair et al. 2004
Mol Cell Biol 24: 6393-6402
 Combined FRAP and FLIP
fluorescence loss in photobleaching
(FLIP)

(FRAP)
fluorescence redistribution
after photobleaching

Adriaan
Houtsmuller
 D=4

D=7 + transiently immobile

Adriaan
Houtsmuller
Adriaan
Houtsmuller
 iFRAP
Inverse FRAP (iFRAP): All fluorescence except the region of interest
is bleached and the loss of fluorescence is monitored.

…
T -0.101s T 0.000s T 0.001s T 10.001s
T -0.051s T 0.051s
T -0.001s T 0.101s

Prebleach Bleach pulse Postbleach Series

Series (“normal time-lapse” starting
(“normal time-lapse”) without delay after the bleach)
iFRAP: Nuclear Pore Complex

Rabut G, Doye V, Ellenberg J (2004) Nat. Cell Biology 6: 1114-1121

Photoactivation and Conversion

From: Lukyanov, K. A. et al. (2005) Nat Rev Mol Cell Biol 6(11): 885-890.
Photoactivation and Conversion

www.olympusconfocal.com/applications/opticalhighlighters.html
 Photoactivation + Reference
Transport speed between compartments: Golgi ↔ plasma membrane

RAS (localized to plasma membrane and Golgi) tagged

with paGFP (green) and mCherry (red)
O.Rocks (Bastiaens Group)
 Reference Channel
• For long-term FRAP experiments it might be necessary to include a
reference channel, which shows the bleached structure.
first last
prebleach
postbleach postbleach
FRAP Channel

ROI alignment by
Reference Chn.

reference channel

line sequential frame sequential

• If sequential acquisition is necessary,
use line-by-line sequential!
Dendra2-Fibrillarin – Example from a course
Nucleolus highlighted with a
Dendra 2 fibrillarin fusion protein
is converted with tornado scan
(Olympus FV1000)

3
converted Nucleolus Ch1
converted Nucleolus Ch2
Ratio converted Nucleolus
Ratio Nucleolus 2
2.5
Ratio Nucleolus 3

Fluorescence intensity
2

1.5

0.5

0
0 10 20 30 40 50 60 70 80 90
Time (s)
Overview
• Introduction
• Application examples
• Related techniques
• Instrument setups for FRAP
• Basic FRAP analysis procedures
 FRAP on Epifluorescence microscope

• Fixed laser spot incoupling

• Laser incoupling via scanner

Olympus / Rapp OptoElectronic, 355 nm pulsed laser (cutting)

 FRAP on CLSM
• current commercial CLSMs can control
the laser intensity pixel by pixel with AOTFs
 scanning of Regions of interest (ROI)

• Bleach regions are very flexible

• Together with GFPs the commercial availability of ROI bleaching with

CLSMs was important for FRAP revival
• Pinhole can be adjusted appropriately
FRAP on Spinning disk
• Laser coupling via scanner

• Low bleaching during acquisition

• Fast frame rate
• A flipping mirror switches between imaging and bleaching

e.g. Ultraview VoX with PK unit

Overview
• Introduction
• Application examples and related techniques
• Instrument setups for FRAP
• Basic FRAP analysis procedures
• Retrieving raw data
• Qualitative vs. quantitative analysis
• Recovery time
• Mobile / immobile fraction
• Photobleaching correction
• Data normalization
• Curve fitting
Basic Frap analysis: retrieving raw data
Mean intensities inside regions of interest (ROI) are measured over time.

Kappel and Eils, Leica App.Letter 2004

Qualitative vs. quantitative analysis
• It is relatively easy to compare two conditions qualitatively by FRAP,
e.g. recovery of a wildtype vs. mutant protein.
 use exactly the same imaging/bleaching conditions!

• To determine quantitative data like the diffusion coefficient (inside

cells) or binding coefficients more information about the investigated
system is necessary.
 modeling can be used to compare the experimental data with
simulated data
 Modeling + Simulation

Monte Carlo Simulation Spatial modeling

Adriaan Houtmuller
www.erasmusmc.nl/pathologie/research/houtsmuller/?lang=en

A. Bancaud, S. Huet et al. (2010)

Live Cell Imaging: A Laboratory Manual, Second Edition
Cold Spring Harbor Laboratory Press

Studying protein-protein interactions by advanced light microscopy and spectroscopy

EMBO course Debrecen, August 2011
 Qualitative vs. quantitative analysis
Halftime of recovery

Amount of
‘mobile’ molecules
Recovery time

• Halftime of recovery 1/2 is a qualitative measure of the recovery speed

under constant experimental conditions.
• The Halftime of recovery is proportional to the bleach area size, if the
recovery is limited by diffusion.
diffusion coefficient D [µm2s-1]
 Mobile / immobile fraction
• Photodamage can also create an ‘immobile’ fraction
 bleaching creates radicals, which can cause crosslinking between proteins
by chemical reactions
• A simple test for photo-induced ‘immobile’ fraction is repeating the
FRAP experiment at the same position:
 higher recovery indicates
a real ‘immobile’ fraction
 similar ‘immobile’ fraction
indicates potential
photodamage

Lippincott-Schwartz et al. Nature CellBio Supp. 2003

 Spot bleaching of GFP in nuclei
before bleach after bleach
fixed

immobile
molecules

living

mobile
molecules

Adriaan Houtsmuller
Spot-FRAP on ERCC1-GFP fluorescence
ratio profiles
expressing cells
1.0
FIXED
0.8

after/before
0.6

0.4

0.2 n=32

0.0
0 2 4 6 8

1.0
LIVING, NO UV
0.8

after/before
0.6

0.4
n=64
0.2

0.0
0 2 4 6 8

2 1.0
LIVING, UV (8 J/m )
0.8
after/before

0.6

0.4

0.2 n=62

0.0 Adriaan
0 2 4 6 8

pre-bleach post-bleach distance to spot (m) Houtsmuller

Photobleaching correction

Photobleaching correction is important:

• to determine the ‘right’ relation Photobleaching corrected
of mobile/immobile fraction with photobleaching

arbitrary intensity
• Photobleaching influences
the recovery time determination
since it changes the plateau of
recovery

time
Photobleaching correction
Methods for photobleaching correction:

1. Acquisition photobleaching can also be corrected by using the total

cell intensity. This also corrects for fluorescence loss due to the
bleach pulse and laser fluctuations.
2. The amount of acquisition photobleaching can be determined by
measuring the intensity decay of neighboring cells (not ‘FRAPed’
cells).
3. Measuring bleaching in a time-lapse without FRAP, after the
fluorescence is back in steady state. This bleaching curve can be
fitted with an exponential decay function to correct the FRAP data.
4. Using the prebleach data to determine the acquisition bleach rate.
 Data normalization
• Raw Data
Normalization is important to
compare different experiments

With framerate above 1 fps the

prebleach should be at least
50 frames to reach the dark state equilibrium
Reversible Dark states

Living cell expressing H2b-GFP

Driving FPs into dark state
equilibrium.

Returning FPs into fluorescent

state after bleach pulse.

Returning FPs into fluorescent

state at slower frame rate.

Living cell expressing H2b-paGFP

Bancaud, A., Huet, S., RABUT, G., and Ellenberg, J.

2010 Live Cell Imaging: A Laboratory Manual, Second
Edition Cold Spring Harbor Laboratory Press
 Data normalization
• Raw Data
• background subtracted T (t )  Iback(t )  I (t )  Iback(t )
 Data normalization
• Raw Data
T (t )  Iback(t ) I (t )  Iback(t )
• background subtracted 
• normalized to prebleach Tprebleach  Iback Iprebleach  Iback
(single normalization)

Fluorescence loss due to

bleach pulse and
acquisition bleaching
 Data normalization
• Raw Data
I (t )  Iback(t ) Tprebleach  Iback
• background subtracted Inorm(t )  
• normalized to prebleach
Iprebleach  Iback T (t )  Iback(t )
(single normalization)
• Corrected for fluorescence loss
(double normalization;
Phair et al. 2003)
 Data normalization
• Raw Data Inorm(t )  Inorm(t1st postbleach)
• background subtracted Itriplenorm(t ) 
• normalized to prebleach
Inormprebleach  Inorm(t1st postbleach)
(single normalization)
• Corrected for fluorescence
loss (double normalization;
Phair et al. 2003)
• Additional normalization
to 1st postbleach set to zero
 experiments can easily
be compared with different
bleach depth.
Curve fitting
Using the appropriate
equations, data can be fitted to
obtain e.g. the effective
diffusion coefficient (Deff).

Circular ROI: Soumpasis 1983 Stripe ROI: Ellenberg 1997

Studying protein-protein interactions by advanced light microscopy and spectroscopy

EMBO course Debrecen, August 2011
 Acknowledgements
ALMF: Rainer Pepperkok
Yury Belyaev Christian Tischer
Kota Miura (CMCI)
 IgorPro FRAP analysis Macro:
http://cmci.embl.de/downloads/frap_analysis

Timo Zimmermann, CRG Barcelona

Adriaan Houtsmuller, Josephine Nefkens Institute, Rotterdam
Arne Seitz, EPFL, Lausanne
Jens Rietdorf, FMI, Basel

FRAPAnalyser http://actinsim.uni.lu/eng/Downloads

www.embl.de/almf/