JCB

Mini-Review

Fluorescence resonance energy transfer (FRET)

microscopy imaging of live cell protein localizations
Rajesh Babu Sekar and Ammasi Periasamy
W.M. Keck Center for Cellular Imaging, Departments of Biology and Biomedical Engineering, University of Virginia,
Charlottesville, VA 22904

The current advances in fluorescence microscopy, coupled sufficient separation in excitation spectra for selective
with the development of new fluorescent probes, make stimulation of the donor GFP, an overlap (30%) between
fluorescence resonance energy transfer (FRET) a powerful the emission spectrum of the donor and the excitation spectrum
technique for studying molecular interactions inside living of the acceptor to obtain efficient energy transfer, and reason-
cells with improved spatial (angstrom) and temporal able separation in emission spectra between donor and acceptor
(nanosecond) resolution, distance range, and sensitivity GFPs to allow independent measurement of the fluorescence
and a broader range of biological applications. of each fluorophore (Pollok and Heim, 1999). GFP-based
The Journal of Cell Biology

FRET imaging methods have been instrumental in determining

the compartmentalization and functional organization of
Fluorescence resonance energy transfer (FRET)* is a distance-
living cells and for tracing the movement of proteins inside
dependent physical process by which energy is transferred
cells (Hanson and Kohler, 2001).
nonradiatively from an excited molecular fluorophore (the
In this paper, we give a brief description of FRET micro-
donor) to another fluorophore (the acceptor) by means of
scopic imaging techniques and data analysis. In addition,
intermolecular long-range dipole–dipole coupling. FRET
important biological applications of FRET microscopy, such
can be an accurate measurement of molecular proximity at
as protein interaction studies, Ca2 signaling, nucleic acid
angstrom distances (10–100 Å) and highly efficient if the
studies, characterization of gene expression, and real-time
donor and acceptor are positioned within the Förster radius
PCR assays, are discussed.
(the distance at which half the excitation energy of the donor
is transferred to the acceptor, typically 3–6 nm). The efficiency
Supplemental material available
of FRET is dependent on the inverse sixth power of inter-
molecular separation (Förster, 1965; Clegg, 1996; Lakowicz, More technical details of various intensity-based FRET and
1999), making it a sensitive technique for investigating a fluorescence lifetime imaging techniques are available online
variety of biological phenomena that produce changes in at http://www.jcb.org/cgi/content/full/jcb.200210140/DC1.
molecular proximity (dos Remedios et al., 1987). Technological A discussion about the successes of other FRET microscopy
advances in light microscopy imaging, combined with the techniques is also included.
availability of genetically encoded fluorescent proteins provide
the tools necessary to obtain spatial and temporal distribution FRET imaging microscopy
of protein associations inside living cells (Heim and Tsien, Whereas light microscopy initiated our understanding of
1996; Day, 1998; Elangovan et al., 2002, 2003). cellular structure and the associated function, molecular
The widely used donor and acceptor fluorophores for biological studies over the past few decades have shown that
FRET studies come from a class of autofluorescent proteins, cellular events, such as signal transduction and gene tran-
called GFPs. The spectroscopic properties that are carefully scription, require the assembly of proteins into specific
considered in selecting GFPs as workable FRET pairs include macromolecular complexes. Traditional biophysical or bio-
chemical methods did not provide direct access to the inter-
actions of these protein partners in their natural environment.
The online version of this article includes supplemental material. Intensity-based imaging techniques applying the method of
Address correspondence to Ammasi Periasamy, Ph.D., Center Director, FRET microscopy (wide field, confocal, and multiphoton
W.M. Keck Center for Cellular Imaging, Department of Biology, Gilmer [MP]) were subsequently developed, facilitating the study of
Hall (064), University of Virginia, Charlottesville, VA 22904. Tel.: (434) these interactions inside intact living cells (Periasamy, 2001).
243-7602/(434) 982-4869. Fax: (434) 982-5210. E-mail: ap3t@virginia.edu New imaging technologies, coupled with the development of
*Abbreviations used in this paper: CAM, calmodulin; EGFR, EGF receptor; new genetically encoded fluorescent labels and sensors and
FRET, fluorescence resonance energy transfer; MP, multiphoton; SBT,
spectral bleedthrough. the increasing capability of computer software for image
Key words: FRET microscopy; data analysis; cameleons; dimerization; acquisition and analysis, have enabled more sophisticated
FRET assays studies of protein functions and processes ranging from gene

 The Rockefeller University Press, 0021-9525/2003/03/629/5 $8.00

The Journal of Cell Biology, Volume 160, Number 5, March 3, 2003 629–633
http://www.jcb.org/cgi/doi/10.1083/jcb.200210140 629
 630 The Journal of Cell Biology | Volume 160, Number 5, 2003

Figure 1. Confocal FRET analysis dem-

onstrates that integrins induce local
Rac–effector coupling. NIH-3T3 cells
were microinjected with cDNAs encod-
ing the indicated GFP–V12-Rac fusion
proteins and then with Alexa–PBD pro-
tein (Pozo et al., 2002). Donor (A), un-
corrected FRET (B), and corrected FRET
(C) images are shown. In the color scale,
red represents a high FRET signal and
blue represents a low signal.

expression to second-messenger cascades and intercellular yields a wealth of spectral information with several advan-
signaling (Roessel and Brand, 2002). tages over a wide-field image, including controllable depth
FRET microscopy relies on the ability to capture fluores- of field and the ability to collect serial optical sections from
cent signals from the interactions of labeled molecules in thick specimens. Owing to its nanometer depth resolution
single living or fixed cells. If FRET occurs, the donor chan- and nonintrusiveness, confocal FRET provides a new
nel signal will be quenched and the acceptor channel signal approach to measure viscoelasticity and biochemical re-
will be sensitized or increased (Herman, 1998). With FRET sponses of living cells and real-time monitoring of cell
microscopic imaging, not only colocalization of the donor- membrane motion in natural environments (unpublished
and acceptor-labeled probes within 0.09 m2 can be seen, data). Confocal microscopy, however, is limited to stan-
but molecular associations at close distances can be verified. dard laser lines of defined wavelengths. MP-FRET micros-
Several FRET microscopy techniques exist, each with its copy overcomes this limitation by using a tunable laser
The Journal of Cell Biology

own advantages and disadvantages. They are used for vari- (range 700–1000 nm), allowing excitation of a wide variety
ous biological applications, including studies of organelle of fluorophores with higher axial resolution, greater sample
structure, conjugated antibodies, cytochemical identifica- penetration, reduced photobleaching of marker dyes, and
tion, and oxidative metabolism (www.cyto.purdue.edu). increased cell viability. These advantages allow investiga-
Wide-field microscopy is the simplest and most widely tions on thick living tissue specimens that would not other-
used technique. It is used for quantitative comparisons of wise be possible with conventional techniques. However,
cellular compartments and time-lapse studies for cell motil- all of these intensity-based FRET techniques require pro-
ity, intracellular mechanics, and molecular movement cessing software to remove the unwanted bleedthrough
(www.api.com). For example, new fluorescent indicators components in the FRET image (Fig. 2).
have allowed the measurement of Ca2 signals in the cyto- Other FRET imaging techniques. Temporal resolution of
sol and organelles that are often extremely localized the imaging modalities can be achieved by the technique of
(Miyawaki et al., 1997) and nondestructive imaging of dy- fluorescence lifetime imaging (FLIM). This technique moni-
namic protein tyrosine kinase activities in single living cells tors the localized changes in probe fluorescence lifetime and
(Ting et al., 2001). Wide-field microscopy, however, suf- provides an enormous advantage for imaging dynamic events
fers from a major drawback due to the generation of out of within the living cells. When combined with FRET, this ap-
focus signals. Laser scanning confocal and MP-FRET mi- proach provides direct evidence for the physical interactions
croscopy provide the advantage of rejecting out of focus in- between two or more proteins with very high spatial and
formation. They also allow associations occurring inside temporal resolution (Bastiaens and Squire, 1999; Elangovan
the cell to be localized in three dimensions. A confocal et al., 2002). Fluorescence correlation spectroscopy (FCS)
FRET image (Fig. 1), with improved lateral resolution, (www.probes.com), in which spontaneous fluorescence in-

Figure 2. Localization of CFP– and

YFP–C/EBP proteins expressed in live
mouse pituitary GHFT1-5 cells studied
using Bio-Rad Laboratories MP-FRET
microscopy. The donor (A), the uncor-
rected FRET (B), and processed FRET (C)
images and their respective histograms
(D, E, and F) representing the signal
strength of the selected protein (Elango-
van et al., 2003).
 FRET microscopy imaging | Sekar and Periasamy 631

tensity fluctuations are measured in a microscopic detection membranes and dissociating it from Rho-GDI (guanine nu-
volume, has greatly increased the utility and scope of FRET. cleotide dissociation inhibitors), and also that in spite of its
FRAP, based on the principle of observing the rate of recov- homogeneous distribution in the cell, constitutively active
ery of fluorescence due to the movement of a fluorescent Rac selectively interacts with effectors at specific regions of
marker into an area of the membrane, is widely used to assess the cell edge (Pozo et al., 2002) (Fig. 1).
the structure of biological membranes and to measure the The onset and termination of Ca2 signaling in specific
lateral diffusion of various membrane or cytoplasmic constit- cellular compartments like the cytoplasm, nucleus, or endo-
uents. A related technique, fluorescence loss in photobleach- plasmic reticulum can be observed by measuring the change
ing (FLIP), can reveal whether flow occurs between two in the ratio of the fluorescence intensities of acceptor and
compartments or not. FRAP and FLIP are valuable for stud- donor molecules in live cells (Truong et al., 2001). Came-
ies of Golgi/ER trafficking in animal cells (Cole et al., 1996). leons, a class of fluorescent indicators for Ca2 based on
With the development of new FRET modalities, like ho- GFPs and calmodulin (CAM), are useful tools in measur-
motransfer or energy migration FRET, photochromic FRET ing the free Ca2 concentrations in living cells. The tradi-
(Giordano et al., 2002), bioluminescence resonance energy tional yellow cameleon consists of a fusion of CFP, CAM,
transfer (BRET) (www.lifesciences.perkinelmer.com), and a the CAM-binding peptide of myosin light chain kinase
new class of luminophores, called long-wavelength long-life- (MLCKp), and a YFP. With the increase of free Ca2 in so-
time high-quantum yield luminophores, the future of FRET lution, the CAM module of the cameleon binds Ca2 and
is bright. More specifically, the future biological applications wraps around the fused MLCKp. This conformational
of FRET imaging microscopy may include cellular events change decreases the distance between CFP and YFP. Be-
coupled to specific molecular signaling processes and simul- cause FRET depends on both the proximity of the donor
taneous observation of a series of reversible molecular pro- and acceptor fluorophores and the orientation of their rela-
cesses in living cells (Truong and Ikura, 2001). tive dipoles, the interaction of cameleons with Ca2 leads to
The Journal of Cell Biology

changes in the degree of FRET between CFP and YFP. After

FRET data analysis calibration of the FRET response to known Ca2 concentra-
Intensity-based FRET imaging microscopy suffers from var- tions, the degree of FRET in vivo can theoretically reflect
ious drawbacks, including autofluorescence, detector noise, the absolute levels of Ca2 present in cellular compartments.
optical noise, and photobleaching. In addition, spectral Beyond intracellular ion sensing, FRET constructs have
bleedthrough (SBT) or contributions of donor and acceptor been used to investigate the downstream events of second
fluorescence emission into the FRET channel is a major messenger signaling (Adams et al., 1991). Targeting the
problem. Due to these effects, the observed FRET signal is cAMP signaling pathway, a sensor construct has been de-
higher than the actual signal. To correct for these problems, signed in which the kinase-inducible domain of cAMP-
various methods of FRET data analysis for wide-field responsive element–binding protein forms a linker between
microscopy have been developed (Gordon et al., 1998; blue fluorescent protein and GFP. cAMP-induced PKA-
Kraynov et al., 2000; Xia and Liu, 2001). These methods dependent phosphorylation is known to cause a conforma-
correct for SBT and for the dependence of FRET on donor tional change in the kinase-inducible domain, and FRET ef-
and acceptor concentrations. ficiency is altered in response to cAMP signaling.
Recently, a new algorithm that removes both the donor FRET establishes the possibility of studying on a localized
and acceptor SBT problems and corrects the variation in flu- spatial scale the interactions between a receptor–ligand pair,
orophore expression level for all intensity-based FRET tech- dimerization of individual receptors, as well as transbilayer
niques (wide field, confocal, and MP) has been developed to distribution of fluorescent lipid analogs and protein-medi-
calculate FRET efficiency (in percent) and estimate the dis- ated lipid transfer between vesicles. Changes in FRET effi-
tance (in angstroms) between donor and acceptor molecules ciency induced by fatty acids, phospholipids, and cholesterol
in a double-labeled cell. Both SBT and fluorophore expres- led to the identification of discrete binding sites for lipids on
sion level corrections are incorporated in mathematical cal- the membrane-bound nicotinic acetylcholine receptor pro-
culations (Elangovan et al., 2003). From the data collected, tein (www.kenes.com/cholinergic). FRET is used to detect
the bleedthrough component is evaluated based on the indi- EGF receptor (EGFR) dimerization and its conformational
vidual donor and acceptor specimens and is then eliminated state (Gadella and Jovin, 1995). Fluorescently labeled EGF
from the FRET data, pixel by pixel, to obtain the true (or molecules with fluorescein donor and rhodamine acceptor
precision) FRET signal (Fig. 2). were allowed to bind EGFR present on cells. The extent of
oligomerization of receptors was monitored by the spatially
FRET applications in cell biology resolved FRET efficiency as a function of the donor/accep-
FRET imaging using GFP spectral mutants provides the tor ratio and treatment conditions. The increase in average
ability to localize and monitor ion binding and molecular FRET efficiency indicated a minimal receptor dimerization
protein–protein interactions in living cells. For example, for the subpopulation of high-affinity receptors. These re-
FRET microscopy with a CFP/YFP FRET pair allows the sults proved that binding of EGF leads to a fast and temper-
detection of direct intermolecular integrin interactions in ature-dependent microclustering of EGFR, which is present
vivo. Integrins are important transmembrane receptor pro- in a predimerized or oligomerized state.
teins involved in cell signaling and adhesion. FRET-based FRET is also used to study the structure, conformation,
studies of Rac activation and localization revealed that inte- hybridization, and automated sequencing of nucleic acids.
grins induce local Rac–effector coupling by directing Rac to Chromosome FISH, based on hybridization of a nucleic
 632 The Journal of Cell Biology | Volume 160, Number 5, 2003

acid fragment to its complement, has become extremely im- information present in intrinsic or exogenous chromophores
portant for gene mapping, identification of mutations, clini- and fluorophores to differentiate tumors and dysplasias from
cal diagnostics, and studies of chromosomal and nuclear ar- normal tissues. Potential applications for tissue FRET imag-
chitecture. A homogeneous DNA diagnostic assay based on ing are also burgeoning. With recent advances in fluorescent
template-directed primer extension detected by FRET, probes, instrumentation, and methodologies, FRET is sure
named the template-directed dye-terminator incorporation to revolutionize scientific research in the near future.
assay, has been developed for mutation detection and high
throughput genome analysis (Chen et al., 1997). We wish to acknowledge Drs. Richard Day and Martin Schwartz for their
A more recent approach to the characterization of gene ex- helpful discussions. We thank Colten Noakes and Ms. Ye Chen for their
expert assistance.
pression involves the use of a “fluorescent timer,” a mutant This work was supported by the W.M. Keck Foundation.
of the dsRed fluorescent protein that shifts color from green
to red over time. Green fluorescence indicates recently trans- Submitted: 24 October 2002
Revised: 21 January 2003
lated protein, which over the course of hours undergoes an Accepted: 21 January 2003
oxygen-dependent autocatalytic reaction to generate a red
fluorescence, denoting matured protein. As the timer pro-
tein switches fluorescence over time, it can be used as a timer References
for gene expression. A tissue thus indicates its fluorescent Adams, S.R., A.T. Harootunian, Y.J. Buechler, S.S. Taylor, and R.Y. Tsien. 1991.
timer production history by its ratio of green to red fluores- Fluorescence ratio imaging of cyclic AMP in single cells. Nature. 349:694–697.
cence; tissues that have recently initiated gene expression ap- Bastiaens, P.I., and A. Squire. 1999. Fluorescence lifetime imaging microscopy: spa-
tial resolution of biochemical processes in the cell. Trends Cell Biol. 9:48–52.
pear green, those with continuous expression appear yellow Chen, X., B. Zehnbauer, A. Gnirke, and P.Y. Kwok. 1997. Fluorescence energy
to orange, and those that have ceased expression appear en- transfer detection as a homogeneous DNA diagnostic method. Proc. Natl.
tirely red. Acad. Sci. USA. 94:10756–10761.
The Journal of Cell Biology

FRET also finds significant application in membrane fu- Clegg, R.M. 1996. Fluorescence resonance energy transfer. In Fluorescence Imag-
sion assays and real-time PCR assays. In the lipid-mixing as- ing Spectroscopy and Microscopy. Vol. 137. X.F. Wang and B. Herman, ed-
itors. John Wiley & Sons Inc., New York. 179–251.
says based on NBD–rhodamine energy transfer (Struck et Cole, N.B., C.L. Smith, N. Sciaky, M. Terasaki, M. Edidin, and J.L. Schwartz.
al., 1981), membranes labeled with a combination of FRET 1996. Diffusional mobility of Golgi proteins in membranes of living cells.
donor and acceptor lipid probes are mixed with unlabeled Science. 273:797–801.
membranes. FRET decreases when the average spatial sepa- Day, R.N. 1998. Visualization of Pit-1 transcription factor interactions in the liv-
ing cell nucleus by fluorescence resonance energy transfer microscopy. Mol.
ration of the probes is increased upon fusion of labeled
Endocrinol. 12:1410–1419.
membranes with unlabeled membranes. In real-time PCR, dos Remedios, C.G., M. Miki, and J.A. Barden. 1987. Fluorescence resonance en-
the amount of fluorescence emission at each cycle is moni- ergy transfer measurements of distances in actin and myosin: A critical evalu-
tored as an indicator of amplicon production. Fluorescence ation. J. Muscle Res. Cell Motil. 8:97–117.
monitoring of PCR for detection and quantification has be- Elangovan, M., R.N. Day, and A. Periasamy. 2002. Nanosecond fluorescence reso-
nance energy transfer-fluorescence lifetime imaging microscopy to localize
come a standard method with many applications, including the protein interactions in a single living cell. J. Microsc. 205:3–14.
expression analysis and pathogen detection. Elangovan, M., H. Wallrabe, Y. Chen, R.N. Day, M. Barroso, and A. Periasamy.
FRET immunoassays, comprising a Cy5 NH2-terminally 2003. Characterization of one- and two-photon excitation fluorescence reso-
labeled phosphopeptide, which is recognized by an an- nance energy transfer microscopy. Methods. 29:58–73.
tiphosphotyrosine primary mouse antibody, followed by a Förster, T. 1965. Delocalized excitation and excitation transfer. In Modern Quan-
tum Chemistry. Vol. 3. O. Sinanoglu, editor. Academic Press Inc., New
Cy3-labeled secondary antibody, are useful in measuring York. 93–137.
specific antibody–antigen interactions. FRET occurs when Gadella, T.W.J., Jr., and T.M. Jovin. 1995. Oligomerization of epidermal growth
the components are sequentially bound together and excita- factor receptors on A431 cells studied by time-resolved fluorescence imaging
tion at Cy3 wavelengths produces emission at Cy5 wave- microscopy. A stereochemical model for tyrosine kinase receptor activation.
lengths. Disruption of the interaction between the phospho- J. Cell Biol. 129:1543–1548.
Giordano, L., T.M. Jovin, M. Irie, and E.A. Jares-Erijman. 2002. Dihet-
peptide and primary antibody will result in a reduction of eroarylethenes as thermally stable photoswitchable acceptors in photochro-
the FRET signal observed. FRET is also used in the design mic fluorescence resonance energy transfer (pcFRET). J. Am. Chem. Soc.
and synthesis of FRET-based fluorogenic enzyme substrates, 124:7481–7489.
useful in monitoring the enzymatic activity. Gordon, G.W., G. Berry, X.H. Liang, B. Levine, and B. Herman. 1998. Quantita-
tive fluorescence resonance energy transfer measurements using fluorescence
microscopy. Biophys. J. 74:2702–2713.
Conclusion Hanson, M.R., and R.H. Kohler. 2001. GFP imaging: methodology and applica-
Biology, imaging, and spectroscopy have been recently com- tion to investigate cellular compartmentation in plants. J. Exp. Bot. 52:529–
bined to provide powerful tools for research and clinical ap- 539.
plications. FRET techniques and fluorescent reagents such Heim, R., and R.Y. Tsien. 1996. Engineering green fluorescent protein for im-
proved brightness, longer wavelengths and fluorescence resonance energy
as GFP, which are widely used to probe for the presence and transfer. Curr. Biol. 6:178–182.
activity of genes, cellular components, and metabolic or sig- Herman, B. 1998. Fluorescence Microscopy. 2nd ed. Springer-Verlag New York
naling pathways, can become more powerful and flexible Inc., New York. 170 pp.
when multispectral imaging tools are employed. Spectral im- Kraynov, V.S., C. Chamberlain, G.M. Bokoch, M.A. Schwartz, S. Slabaugh, and
aging has started to improve chromosomal analysis and K.M. Hahn. 2000. Localized Rac activation dynamics visualized in living
cells. Science. 290:333–337.
genotyping methods with easy and accurate detection of Lakowicz, J.R. 1999. Principles of Fluorescence Spectroscopy. 2nd ed. Plenum
chromosomal rearrangements related to cancer and genetic Publishing Corp., New York. 692 pp.
abnormalities. Optical biopsy techniques use spectroscopic Miyawaki, A., J. Llopis, R. Heim, J.M. McCaffery, J.A. Adams, M. Ikura, and R.Y.
 FRET microscopy imaging | Sekar and Periasamy 633

Tsien. 1997. Fluorescent indicators for Ca2 based on green fluorescent pro- fer to monitor membrane fusion. Biochemistry. 20:4093–4099.
teins and calmodulin. Nature. 388:882–887. Truong, K., and M. Ikura. 2001. The use of FRET imaging microscopy to detect
Periasamy, A. 2001. Methods in Cellular Imaging. Oxford University Press, New protein-protein interactions and protein conformational changes in vivo.
York. 434 pp. Curr. Opin. Struct. Biol. 11:573–578.
Pollok, B.A., and R. Heim. 1999. Using GFP in FRET-based applications. Trends Truong, K., A. Sawano, H. Mizuno, H. Hama, K. Tong, T.K. Mal, A. Miyawaki,
Cell Biol. 9:57–60. and M. Ikura. 2001. FRET-based in vivo Ca2 imaging by a new calmodu-
Pozo, M.A.D., W.B. Kiosses, N.B. Alderson, N. Meller, K.M. Hahn, and M.A. lin-GFP fusion molecule. Nat. Struct. Biol. 8:1069–1073.
Schwartz. 2002. Integrins regulate GTP-Rac localized effector interactions Ting, A.Y., K.H. Kain, R.L. Klemke, and R.Y. Tsien. 2001. Genetically encoded
through dissociation of Rho-GDI. Nat. Cell Biol. 4:232–239. fluorescent reporters of protein tyrosine kinase activities in living cells. Proc.
Roessel, P.V., and A.H. Brand. 2002. Imaging into the future: visualizing gene ex- Natl. Acad. Sci. USA. 98:15003–15008.
pression and protein interactions with fluorescent proteins. Nat. Cell Biol. Xia, Z., and Y. Liu. 2001. Reliable and global measurement of fluorescence resonance
4:E15–E20. energy transfer using fluorescence microscopes. Biophys. J. 81:2395–2402.
Struck, D.K., D. Hoekstra, and R.E. Pagano. 1981. Use of resonance energy trans-
The Journal of Cell Biology