Neuron, Vol. 48, 189–199, October 20, 2005, Copyright ª2005 by Elsevier Inc. DOI 10.1016/j.neuron.2005.10.

003

Innovations in the Imaging of Brain Primer

Functions using Fluorescent Proteins

Atsushi Miyawaki* probes have advanced our understanding of the spatio-

Laboratory for Cell Function Dynamics temporal regulation of biological functions inside the
Advanced Technology Development Group neuron and the brain and their technical limitations. I
Brain Science Institute, RIKEN will also discuss how these approaches will continue
2-1 Hirosawa to improve due to various features of FPs. The direct ob-
Wako-city, Saitama, 351-0198 servation of morphological changes in the developing
Japan nervous system using FPs will not be covered, as that
has been recently reviewed elsewhere (Niell and Smith,
Fluorescence imaging has enabled us to decipher 2004).
spatiotemporal information coded in complex tissues.
Genetically encoded probes that enable fluorescence More-Sensitive Fluorescence Imaging
imaging of excitable cell activity have been con- Because genetically encoded probes using FPs can be
structed by fusing fluorescent proteins to functional introduced by gene-transfer techniques, they can ex-
proteins that are involved in physiological signaling. tract neuronal signals from an intact brain more effi-
The probes are introduced into an intact organism ciently than conventional organic dyes. For instance,
and targeted to specific tissues, cell types, or subcellu- whereas conventional optical imaging of brain tissue
lar compartments, thereby allowing specific signals to stained with voltage-sensitive organic dyes is a noninva-
be extracted more efficiently than was previously pos- sive technique for recording the activities of a number of
sible. In this primer, I will describe how this approach neurons simultaneously, this technique collects signals
has met neuroscientists’ demands and desires. from all of the cell types, including glial cells, which rep-
resent a large fraction of the total membrane surface of
the brain. By contrast, the selective introduction of genet-
Green fluorescent protein (GFP) was originally isolated ically encoded probes into certain neurons has enabled
from the light-emitting organ of the jellyfish Aequorea the elimination of glial signals. Moreover, it is possible to
victoria in 1962 (Shimomura et al., 1962). About 30 years place the probes within specific subcellular compart-
passed, however, before the cDNA encoding the protein ments where the desired signals predominate. Theoret-
was characterized (Prasher et al., 1992). Although spec- ically speaking, the selective introduction of genetically
tral variants with blue, cyan, and yellowish-green emis- encoded probes into neurons should greatly increase
sions have been successfully generated from the the sensitivity of the optical imaging of neuronal firings.
Aequorea GFP (Tsien, 1998), none exhibit emission max- Highly sensitive fluorescence imaging requires bright
ima longer than 529 nm. Fortunately, the discovery of fluorescence signals. Poor folding of a fused FP variant
novel ‘‘GFP-like’’ proteins from Anthozoa (coral animals) results in a nonfluorescent chimera. Accumulation of
has significantly expanded the range of colors available a large amount of such a protein inside cells will de-
for biological applications (Matz et al., 1999). Despite crease the fluorescence signal and potentially perturb
only a modest degree of sequence similarity, these cellular homeostasis if the labeled host protein retains
GFP-like proteins probably share the b-can fold structure its original function. It is therefore imperative to use
that is central to the fluorescence of GFP. In this primer, FP variants that mature efficiently. After protein synthe-
the term ‘‘fluorescent proteins’’ (FPs) is used to describe sis, many FPs mature quite slowly by means of a multi-
those proteins that can become spontaneously fluores- step process that consists of cyclization, dehydration,
cent through the autocatalytic synthesis of a chromo- and oxidation (Tsien, 1998). Several bright FPs that ma-
phore. The FPs can be divided into two families: the Ae- ture efficiently, however, have been developed. They in-
quorea GFP variants and the GFP-like proteins (Figure 1). clude Citrine (Griesbeck et al., 2001) and Venus (Nagai
In the nervous system, intracellular signaling events et al., 2002), two yellow-emitting variants of Aequorea
are closely linked with electrical activities and play es- GFP (YFP), and several red-emitting variants of DsRed
sential roles in information processing. To identify and (Bevis and Glick, 2002; Terskikh et al., 2002) (Figure 1).
characterize the mechanisms by which signals are orga- When incorporated into probes, these variants may
nized inside cells, it is necessary to analyze spatiotem- also facilitate protein folding. Moreover, their rapid mat-
poral patterns of signaling pathways. On the other hand, uration allows the immediate detection of fluorescence
neural circuitry operates as an ensemble in the nervous signals after gene introduction in freshly prepared bio-
system. To investigate the patterns of neuronal firing, it logical samples, such as brain slices (Nagai et al., 2002).
is necessary to monitor multiple transmembrane vol- Among the genetically encoded fluorescent probes
tages or signals that result from electrical activity in for Ca2+, yellow cameleons (YCs) employ fluorescence
complex tissues or intact animals. In the past several resonance energy transfer (FRET) from a cyan-emitting
years, various probes have been generated principally variant of Aequorea GFP (CFP) to YFP. YCs are com-
using variants of Aequorea GFP (Zhang et al., 2002; posed of a linear combination of CFP, calmodulin
Miyawaki, 2003; Miesenböck, 2004; Miesenböck and (CaM), a glycylglycine linker, the CaM binding peptide
Kevrekidis, 2005) (Table 1). Here, I describe how the of myosin light chain kinase (M13), and YFP (Miyawaki
et al., 1997). Binding of Ca2+ to the CaM moiety of the
YC initiates an intramolecular interaction between the
*Correspondence: [email protected] CaM and M13 domains, causing the chimeric protein
Neuron
190
Primer
191

to shift from an extended conformation to a more com- Ca2+ concentration in the ER. Another attempt to elimi-
pact one, thereby increasing the efficiency of FRET from nate the sensitivity of genetically encoded Ca2+ indica-
CFP to YFP. FRET is highly sensitive to the relative ori- tors to endogenous CaM and CaM-binding proteins in-
entation and distance between the two fluorophores volved the use of troponin C proteins from skeletal and
(Miyawaki, 2003). This technique is amenable to emis- cardiac muscle (Heim and Griesbeck, 2004). Because
sion ratioing, which is more quantitative than single- troponin C is specifically expressed in muscle, these
wavelength monitoring, and is also an ideal readout novel probes should not be perturbed by endogenous
for fast imaging by laser-scanning confocal micros- proteins when expressed in nonmuscle tissues. A vari-
copy. Due to their emission of ratiometric responses, ant of the troponin C-based Ca2+ indicators, TN-L15,
YCs are the best choice for observing Ca2+ dynamics was targeted to the plasma membrane of primary hip-
in motile animals. Calcium transients in neurons have pocampal neurons to monitor submembrane Ca2+ lev-
been observed in the process of gentle touch sensation els. These levels were found to be in equilibrium with
of C. elegans (Suzuki et al., 2003) and during escape be- the concentration of bulk cytosolic Ca2+, which indi-
haviors of zebrafish (Higashijima et al., 2003). The ratio- cates the absence of a standing Ca2+ gradient from
metric approach using YCs is also of great advantage the membrane toward the cytosol.
for quantitative Ca2+ measurements. A version of YC
(Synapcam) was localized to the postsynaptic terminals Fluorescence Imaging of Multiple Populations
at the Drosophila larval neuromuscular junction to se- of Neurons across Synapses
lectively and quantitatively measure Ca2+ influx through The expression of a genetically encoded probe is driven
glutamate receptors with single-impulse and single- in a certain population of neurons by the use of a specific
bouton resolution (Guerrero et al., 2005). The Ca2+-influx promoter. Visualizing the connectivity between two or
signals (i.e., the transmission strength) varied along the more different (sub)populations of neurons is more
axonal branches with a gradient of weak at the proximal challenging. For example, in vivo time-lapse imaging
boutons to strong at the distal ones. of mouse neuromuscular junctions has revealed axon-
A large dynamic range is an important factor for de- branch dynamics associated with naturally occurring
tecting subtle but significant signals. Expansion of the synapse elimination (Walsh and Lichtman, 2003).
Ca2+ responses of YCs has been achieved by combining Additionally, the olfactory system of Drosophila pro-
FRET and circular-permutation techniques. To achieve a vides one of the most interesting and accessible sys-
larger Ca2+-dependent change in the relative orientation tems for examining neuronal connectivity. Functional
and distance between the fluorophores of CFP and YFP, imaging of multiple synaptically coupled populations
the most-recent approach has been to use circularly of neurons in the antennal lobe has been performed us-
permuted FP variants. In this technique, the amino and ing two genetically encoded fluorescent probes. The
carboxyl portions of a protein are interchanged and re- first example of this approach employed synapto-
connected by a short spacer between the original ter- pHluorin (Ng et al., 2002). This protein reports the re-
mini (Baird et al., 1999). Circular permutation was con- lease of neurotransmitter, i.e., the exocytosis of synap-
ducted on YFP, and new termini were introduced into tic vesicles. A pH-sensitive variant of GFP, pHluorin
surface-exposed loop regions of the b-barrel. Compared (Miesenböck et al., 1998), is fused to the lumenal aspect
to conventional YCs, YC3.60, in which YFP has been re- of VAMP. As the lumenal pH increases from acidic to
placed by a circularly permuted YFP (cpYFP), produces neutral, the signal of synapto-pHluorin increases by
equally bright signals while showing a 5- to 6-fold larger about 20-fold. Changes in fluorescence reflect the net
dynamic range (Nagai et al., 2004). Thus, YC3.60 gives accumulation of VAMP on the presynaptic surface.
a greatly enhanced signal-to-noise (S/N) ratio. Synapto-pHluorins have revealed important aspects of
It has been pointed out that the dynamic range of YCs information processing in specific neural networks of a
is damped down in neuronal cell types that have a large variety of animals, including C. elegans (Samuel et al.,
amount of CaM and CaM-associated proteins. Thus, the 2003), Drosophila (Yu et al., 2004), Aplysia (Kim et al.,
interface between the CaM and M13 peptide has been 2003), and mice (Bozza et al., 2004). Initially, synapto-
redesigned to generate highly specific protein/peptide pHluorin was expressed in the three classes of neurons
pairs that are not perturbed by endogenous CaM or that form synapses in the Drosophila antennal lobe:
CaM-binding proteins (Palmer et al., 2004). A mutant olfactory receptor neurons (ORNs), projection neurons
CaM/peptide pair, D1, was cloned between CFP and (PNs), and inhibitory local interneurons. Visualization
YFP to yield a reengineered YC. The D1-containing YC of population responses to natural odors revealed that
is indifferent to a large excess of CaM. Moreover, it dis- the glomerular synaptic relay from the ORN to the PN
plays a low-Ca2+ affinity with a Kd value of 60 mM and transmitted spatially invariant but temporally restruc-
has proven to be an ideal probe for measuring the tured patterns of activity and that the synaptic relay

Figure 1. Wild-Type Fluorescent Proteins and Their Variants

Colors indicate emission maxima. Aequorea GFP and its variants are enclosed with a box. Oligomeric states are indicated in square brackets.
Bidirectional and unidirectional arrows indicate reversible and irreversible photomodulating properties, respectively. Commercial names are
given in the parentheses: yEVRUGEN, zSTRATAGENE, #Clontech (TAKARA), *MBL (Medical Biological Laboratory) (Amalgaam). xPromega, $In-
vitrogen, &Q-BIOgene. References are a(Tsien, 1998), b(Griesbeck et al., 2001), c(Nagai et al., 2002), d(Zacharias et al., 2002), e(Rizzo et al.,
2004), f(Patterson and Lippincott-Schwartz, 2002), g(Shagin et al., 2004), h(Chudakov et al., 2004), i(Matz et al., 1999), j(Bevis and Glick,
2002), k(Terskikh et al., 2002), l(Terskikh et al., 2000), m(Campbell et al., 2002), n(Shaner et al., 2004), o(Wang et al., 2004a), p(Verkhusha and
Sorkin, 2005), q(Wiedenmann et al., 2002), r(Gurskaya et al., 2001), s(Chudakov et al., 2003), t(Ando et al., 2002), u(Karasawa et al., 2004), v(Kar-
asawa et al., 2003), w(Wiedenmann et al., 2004), x(Ando et al., 2004), y(Tsutsui et al., 2005), z(Shkrob et al., 2005).
Neuron
192

Table 1. Genetically Encoded Probes that Detect Physiological Signaling

a
Re-engineered to be indifferent to endogenous calmodulin and calmodulin-binding proteins. References are
b
(Siegel and Isacoff, 1997), c(Guerrero et al., 2002), d(Ataka and Pieribone, 2002), e(Sakai et al, 2001), f(Baird
et al., 1999), g(Nakai et al., 2001), h(Nagai et al., 2001), i(Miyawaki et al., 1997, 1999; Griesbeck et al., 2001; Nagai
et al., 2002, 2004), j(Truong et al., 2001), k(Palmer et al., 2004), l(Heim and Griesbeck, 2004), m(Takao et al., 2005),
n
(Okamoto et al., 2004), o(Kunar and Augustine, 2000), p(Miesenböck et al., 1998), q(An and Almers, 2004).

also received inhibitory input from the local interneu- ular activity that were conserved in different flies, and
rons (Ng et al., 2002). the specific responsivity of a given glomerulus was
In a second example, G-CaMP was successfully used found to be a consequence of the specificity of a single
to study the representation of olfactory information in odorant receptor expressed by the incoming ORNs
the Drosophila antennal lobe (Wang et al., 2003). (Wang et al., 2003). These observations are consistent
G-CaMP, a single-wavelength intensity-modulating probe with the ‘‘one neuron-one receptor’’ (Wang et al.,
for Ca2+, is based on a circularly permuted GFP (Nakai 1998) and ‘‘glomerular convergence’’ (Mori et al., 1999)
et al., 2001). The probe was expressed in ORNs or principles. This type of study utilizing G-CaMP has
PNs, and patterns of glomerular activity in the antennal been extended to the analysis of the mushroom body.
lobe were imaged presynaptically or postsynaptically, Spatially specific odor-evoked activity was clearly
respectively. Odors elicited specific patterns of glomer- observed in mushroom body neurons (Wang et al.,
Primer
193

2004b). Recently, flies that expressed G-CaMP were as short as 2 ms and does not become inactive during
successfully used to map the primary sensory neurons extended depolarizations. Another approach to the de-
responsible for an innate avoidance behavior of the an- velopment of voltage sensors uses FRET between CFP
imal. In this study, it was determined that CO2 activates and YFP. The two variants of Aequorea GFP were tan-
only a single glomerulus in the antennal lobe, the V glo- demly linked and fused to the C terminus of the S4
merulus (Suh et al., 2004). segment of a potassium channel to generate a voltage-
Compared to the olfactory system of the Drosophila, sensitive fluorescent protein (VSFP1) (Sakai et al., 2001).
the anatomy of brain cortex is more complicated. Anal- Membrane depolarization results in a rotational move-
ysis of neural activity in complex regions may be facili- ment of S4. This structural change probably alters the
tated by the use of functional probes that carry photo- relative angle between the dipole moments of CFP
modulating FPs (Figure 1). Since the probes can be and YFP, leading to a change in the FRET signal of
photoactivated, the specific neurons to be analyzed VSFP1. When expressed in HEK293 cells, the sensitized
could be highlighted by an optical means, such as YFP signal changed by 1.5% upon a depolarizing step
two-photon excitation (see below). between 280 and 20 mV, with a time constant of 0.7
ms. To date, however, there has been no report of the
Fluorescence Imaging with Higher successful use of genetically encoded probes for mem-
Temporal Resolution brane-potential measurement in complex tissues or in-
As mentioned in the review by Miesenböck (2004), the tact animals. The in vivo application of sensors that de-
real ‘‘clock speed’’ of neural computation is estimated tect fast signals such an AP firing requires that the S/N
to be in the range of 100 Hz to 1 kHz; thus, the probes ratio be maximized (Zochowski et al., 2000). Because
that monitor signals resulting from electrical activity, the S/N ratio improves as the number of photons mea-
such as free-Ca2+ concentrations and pH, instead of sured is increased, future efforts should focus not only
transmembrane voltages, function as low-pass filters. on increasing the signal change but also the amount
Two papers have pointed out the difficulty in monitoring of sensors that are correctly localized to the neuronal
low-rate spiking activity by using such secondary bio- membrane.
logical readouts. First, Fernández-Alfonso and Ryan
(2004) employed synapto-pHluorin to monitor the bal- Fluorescence Imaging with Higher
ance between exocytosis and endocytosis at hippo- Spatial Resolution
campal terminals under various conditions. They found A synaptic vesicle has a typical diameter of 30–50 nm.
that endocytosis is sufficiently fast to avoid vesicle-pool Individual vesicles that gather in a presynaptic terminal
depletion during continuous action-potential (AP) firing can be discerned by electron microscopy. It is, how-
and that the probe overlooks neural activity below the ever, desirable to visualize the dynamics of synaptic
threshold frequency of AP firing (10 Hz at 35ºC). Second, vesicles in the pool, including their movement and fu-
Wilson et al. (2004) studied the representation of olfac- sion during presynaptic activation. Even if the vesicles
tory information in the antennal lobe by applying are fluorescently labeled with FPs localized on the
whole-cell recordings to live flies. They observed that membrane or inside the lumen, individual vesicles are
PNs displayed broader and more-complex responses not discernible by conventional fluorescence micros-
than their respective ORNs, which is different from the copy. The resolution of conventional fluorescence mi-
findings obtained using G-CaMP (Wang et al., 2003) croscopy is limited by diffraction to about 180 nm in
and synapto-pHluorin (Ng et al., 2002) in that PNs mir- the focal plane. In the range near the diffraction limit,
rored the activity of their monosynaptic ORN inputs. It spatial resolution can be improved only by increasing
was suggested that the genetically encoded probes the semiaperture angle of the objective lens.
failed to measure the low-frequency spiking activity Recently, however, the diffraction barrier has been
that has been implicated in lateral interactions within broken by a new concept (Hell, et al., 2004). Stimulated
the antennal lobe. emission depletion (STED) microscopy has been real-
Thus, many researchers await genetically encoded ized as a point-scanning system; at each point, excita-
voltage sensors. The prototype of a voltage-sensitive tion and STED are accomplished using two synchro-
fluorescent protein, ‘‘FlaSh,’’ was made by inserting nized ultrashort pulses. The first pulse shifts the
wild-type Aequorea GFP after the sixth transmembrane molecules into the excited state, and the immediately
domain of a nonconducting mutant of the Shaker potas- succeeding STED pulse, the wavelength of which corre-
sium channel (Siegel and Isacoff, 1997). In the con- sponds to the red-edge of the emission spectrum,
struct, voltage-driven rearrangement of the channel is transfers the molecules back to the ground state. Be-
converted into a change in the fluorescence intensity cause the STED beam inhibits fluorescence in a nonlin-
of the FP. This protein shows relatively complex and ear manner (saturated depletion) and is delivered in
slow kinetics (toff > 85 ms), suggesting that the struc- a doughnut shape on the focal plane, the excitation
tural rearrangement is related to the C-type inactivation spot is squeezed sharply to a subdiffraction size,
of the channel. Recently, the response speed and/or the thereby allowing nanoscale resolution.
dynamic range of FlaSh have been significantly im- Although STED microscopy can be theoretically ap-
proved (Guerrero et al., 2002). ‘‘SPARC,’’ another genet- plied to any fluorophore, it requires a great deal of ex-
ically encoded voltage probe, was generated by insert- pertise and experience. Based on the basic principle
ing wild-type Aequorea GFP into an intracellular loop of of reversible saturable optical fluorescent transitions
a reversibly nonconducting form of the rat mI skeletal (RESOLFT) (Hell et al., 2004) from which the STED ap-
muscle voltage-gated sodium channel (Ataka and Pier- proach stems, however, it has been predicted that any
ibone, 2002). The probe can report depolarizing pulses fluorescent protein that can be switched ‘‘on’’ and
Neuron
194

‘‘off’’ by different wavelengths of light can be used to More-Quantitative Fluorescence Imaging

break the diffraction barrier, possibly with less expertise Quantitative fluorescence imaging requires that the FPs
and experience than is required for STED microscopy. are not affected by changes in environmental factors,
Possible candidates include KFP (Chudakov et al., such as pH. Neuronal activity can give rise to cytosolic
2003) and Dronpa (Ando et al., 2004), both of which acidification during depolarization and repetitive firing
are variants of Anthozoan FPs (Figure 1). in hippocampal neurons. Thus, the resistance of FPs
to pH changes is often required for effective quantitative
Fluorescence Imaging Deep into Tissues neuronal imaging. The chromophore of an FP is sur-
The penetration of visible light into brain tissue is mostly rounded by a hydrogen-bond network within the b-bar-
prevented by scattering. Since longer-wavelength light rel (Tsien, 1998); the FP is pH sensitive when this net-
scatters less, near-infrared (NIR) light at 650–950 nm work can be affected by external protons. Whereas
can travel further than visible light. Furthermore, NIR wild-type Aequorea GFP, Renilla GFP, and DsRed ex-
light is less absorbed by hemoglobin. Thus, fluoro- hibit pH-insensitive fluorescence, green- and yellow-
phores that show excitation and emission maxima in emitting variants of Aequorea GFP and many other
the NIR range are desirable for deep fluorescence imag- GFP-like proteins show acid-sensitive molar extinction
ing. With regard to NIR fluorescence, organic dyes are coefficients with a wide variety of pKa values (Miyawaki
currently the best option. For example, diaminocya- and Tsien, 2000). Particularly, the circularly permuted
nines (DACs) are newly developed chemical probes constructs tend to be more pH sensitive; the cpGFP-
based on tricarbocyanine and o-phenylenediamine based Ca2+ probes, such as camgaroo, G-CaMP, and
(Sasaki et al., 2005). Upon binding to nitric oxide, these pericam (Nagai et al., 2001), suffer from the problem of
probes emit NIR fluorescence; their excitation and pH sensitivity. On the other hand, blue- and cyan-
emission maxima are 755 nm and 790 nm, respectively. emitting variants of Aequorea GFP have pH-sensitive
By contrast, FPs with NIR fluorescence have not yet fluorescence quantum yields (Miyawaki and Tsien, 2000).
been achieved. One of the far-red-emitting FPs is For effective use of FRET in studies of environments
mPlum, which was created from monomeric red fluores- that experience pH extremes, it is crucial that both the
cent protein (mRFP1) via iterative somatic hypermuta- donor’s quantum yield and the acceptor’s molar extinc-
tion (Figure 1) (Wang et al., 2004a). mPlum shows exci- tion coefficient are indifferent to pH changes. It is unfor-
tation and emission maxima at 590 nm and 649 nm, tunate that CFP and YFP, a frequently used donor/
respectively. Nevertheless, an important advantage of acceptor pair, have an acid-sensitive quantum yield and
FPs over organic dyes is their ability to be genetically in- molar extinction coefficient, respectively. When pH de-
troduced into brain tissues regardless of the depth of creases, these negative effects reinforce each other, re-
the target area. sulting in a loss of FRET between CFP and YFP. A new
Penetration into tissue is markedly improved by the donor/acceptor pair that should enable pH-insensitive
use of multiphoton excitation microscopy with modern FRET measurements has emerged; a cyan-emitting FP
nonlinear optics, because the fluorescence excitation (MiCy) and a monomeric orange-emitting FP (mKO) dis-
is conducted with longer-wavelength light in the NIR play a pH-stable fluorescence quantum yield and molar
range and because in the absence of pinholes the emit- extinction coefficient, respectively (Karasawa et al.,
ted photons can be maximally collected even if they are 2004) (Figure 1). Alternatively, many FPs except for
scattered inside thick tissue (Denk and Svoboda, 1997; blue- and cyan-emitting variants of Aequorea GFP
Helmchen and Denk, 2002). Multiphoton excitation of show pH-stable fluorescence quantum yields. These
FPs is gaining popularity, particularly for in vivo imag- FPs, including EGFP, YFP, and mRFP, should make
ing, in which the imaging points are several-hundred mi- ideal donors for the red and far-red FPs that have
crons deep in the tissue. In an extreme case, Theer et al. emerged recently (Gurskaya et al., 2001; Shaner et al.,
(2003) imaged GFP-labeled neurons down to a depth of 2004; Wang et al., 2004a; Shkrob et al., 2005) (Figure 1).
1 mm below the brain surface of an intact mouse. This Quantitative fluorescence imaging with repetitive im-
depth of penetration was achieved by using a regenera- age acquisition also requires that FPs be photostable.
tive amplifier instead of an oscillator, which is usually An excitation-dependent decrease in the fluorescence
used as the excitation source in multiphoton excitation intensity of a fluorophore usually involves photobleach-
microscopy. The two-photon action spectra of five ing. In some FPs such as Dronpa, however, photochro-
common FPs (EGFP, CFP, YFP, wild-type Aequorea mism dominates over photobleaching (Ando et al.,
GFP, and DsRed) were measured in vitro (Zipfel et al., 2004). Photobleaching is a complex phenomenon; it
2003), which should help adjust the optimal wave- depends on the oxygen concentration and the mode
lengths of the femtosecond ultrashort pulses. Consider- of illumination (wide-field versus point-scan). Among
ing that some special fluorophores have been designed the GFP-like proteins, mPlum (Wang et al., 2004a) and
and synthesized for two-photon excitation (Albota et al., mKO (Karasawa et al., 2004) have proven to be photo-
1998; Mongin et al., 2002), it will be challenging to stable. The intensity of excitation light, however, should
develop FPs that show large two-photon cross- always be reduced while preserving a good signal-
sections. Also promising is the combination of multi- to-noise ratio.
photon excitation with photomodulating FPs (Figure 1), For example, in FRET experiments using CFP and
which enables three-dimensionally restricted highlight- YFP, laser illumination in the point-scan mode, under
ing (Lukyanov et al., 2005). Selective photobleaching which most LSCMs (laser-scanning confocal micro-
of DsRed through three-photon excitation (l < 760 nm) scopes) operate, produces smaller responses than
(Marchant et al., 2001) can be used as an optical high- those obtained with wide-field illumination, probably
lighter. due to the vulnerability of both CFP and YFP to the
Primer
195

transient but high-power illumination of the laser lines. inputs; tetanus shifted the equilibrium toward F-actin,
The intermediate illumination systems between wide- whereas low-frequency stimulation shifted it toward
field and point-scan include the slit-scan mode, multia- G-actin. It is likely that actin dynamics contributes to
perture confocal microscopes that use a Nipkow disk the remodeling of dendritic spines. In fact, the equilib-
(Maddox et al., 2003), and the fringe-projection tech- rium was positively correlated with the size (width) of
nique (Neil et al., 1997). These systems seem to produce the spines. Third, a probe for monitoring the formation
better signals with lower rates of photobleaching than of the complex of syntaxin, SNAP25, and synaptobrevin
commonly used LSCMs, despite more-relaxed optical was constructed and termed the SNARE complex re-
sectioning strengths or lower spatial resolution. porter (SCORE) (An and Almers, 2004). CFP and YFP
Measurement of the lifetime of donor fluorescence by were incorporated into SNAP25. Although the steps of
fluorescence lifetime imaging microscopy (FLIM) is the the core complex formation are important because their
most-reliable method for the quantification of FRET. Im- regulation may contribute to short-term synaptic plas-
portantly, this measurement is not affected by photo- ticity, the sequential order of the interactions had not
bleaching. Although CFP shows a complex decay time been studied in live cells. The SCORE-based probes
course with multiple components, probably due to the combined with a mutagenesis approach have identified
tryptophane in the chromophore-forming tripeptide the Ca2+-sensitive precursor form of the complex. Inter-
(Tsien, 1998), and makes FLIM data complicated, a re- estingly, in this study, the SCORE signal was enhanced
cently developed variant of CFP, Cerulean (Figure 1), when cell samples were observed under evanescent
has a fluorescence lifetime that is best fit by a single ex- field illumination.
ponential (Rizzo et al., 2004). MiCy (Karasawa et al., 2004), Camuia and SCORE utilize intramolecular FRET,
which has a tyrosine residue in the chromophore-forming whereas CFP-actin/YFP-actin use irregular intermolecu-
tripeptide, also shows a simple fluorescence lifetime, lar FRET (the same molecular species are labeled with
which is amenable to FLIM. either the donor or the acceptor). Although the stoichi-
The innovations in microscopy that serve to overcome ometry between donor and acceptor (CFP and YFP) is
the problem of photobleaching include multiphoton ex- supposed to be 1:1 in an intramolecular FRET system,
citation microscopy. Compared with single-photon ex- it is variable in space and time in intermolecular FRET,
citation microscopy, multiphoton excitation microscopy which usually examines interactions between two differ-
shows a reduced net bleaching rate of fluorophores ent species. The quantitative measurement of FRET is
because fluorescence excitation is limited to the focal an important issue for many researchers. The funda-
point during scanning. It has been shown, however, mental, instrument-independent measure of FRET is
that the photobleaching rate of fluorophores per unit the FRET efficiency (E), which is defined as the propor-
excitation is increased supralinearly with the Fourier- tion of energy transferred from the donor to the accep-
transform-limited (FTL) pulse intensity (Patterson and tor. The measurement of E for intermolecular FRET gives
Piston, 2000). Whereas conventional multiphoton exci- the apparent FRET efficiency (Eapp), which provides in-
tation microscopy uses the FTL pulse of the ultrashort formation about donor occupancy. E and Eapp can be
pulsed laser, a recent study (Kawano et al., 2003) suc- obtained by the acceptor bleaching method (Miyawaki
cessfully reduced the bleaching rate of several Aequorea and Tsien, 2000) or more directly by FLIM. A second ap-
GFP variants using coherent control methods (Dela Cruz proach defines FRET indices to make the best use of the
et al., 2004). sensitized emission from the acceptor. Various methods
that introduce different observation strategies for FRET
Fluorescence Imaging of Signaling Events indices, including some that are being implemented on
Related to Synaptic Regulation commercial wide-field and confocal microscopy sys-
Three recent studies attempted to investigate the mech- tems, have been reported (Erickson et al., 2003; Berney
anisms underlying synaptic regulation using FRET be- and Danuser, 2003; Zal and Gascoigne, 2004).
tween CFP and YFP. First, a new probe was developed
for visualizing Ca2+/calmodulin-dependent protein ki- Fluorescence Imaging in a
nase II (CaMKII) activity (Takao et al., 2005). The probe, More-Physiological Context
which is called Camuia, detects the conformational Neural activity can be analyzed in dissociated neuronal
change in the enzyme during activation. In combination cultures, slice cultures, acute slices, and intact animals.
with two-photon excitation microscopy, CaMKII activity Gene-transfer techniques, including liposome-medi-
was imaged at single-dendrite and single-spine resolu- ated transfection, various viral vectors, electroporation,
tion. Because CaMKII is highly enriched in excitatory and the gene gun, have shown significant progress in
synapses and involved in synaptic plasticity, the spatio- recent years. There is a trend toward the understanding
temporal patterns of its activation are of great interest. of brain functions in a physiological context. In this re-
Ideally, Camuia is expected to give us more information gard, the ultimate approach is to construct transgenic
than what was obtained in a previous study where the lines, to which two-photon excitation microscopy can
dynamic translocation of GFP-labeled CaMKII was visu- be applied. This approach has proven to be very power-
alized (Shen and Meyer, 1999). Second, FRET between ful for morphological studies. For example, long-term
actin monomers (CFP-actin and YFP-actin) was mea- in vivo imaging using transgenic mice expressing Ae-
sured to monitor the F-actin/G-actin equilibrium in den- quorea GFP variants in a subset of cortical neurons
dritic spines (Okamoto et al., 2004). Again two-photon (Feng et al., 2000) has revealed experience-dependent
excitation microscopy was used. The postsynaptic appearance, disappearance, and persistence of den-
actin equilibrium was found to modulate rapidly and dritic spines (Trachtenberg et al., 2002; Grutzendler
bidirectionally depending on the frequency of synaptic et al., 2002; Mizrahi and Katz, 2003; Oray et al. 2004;
Neuron
196

Holtmaat et al., 2005). The dynamic turnover varied de- Aequorea GFP as well as GFP-like proteins, an increas-
pending on the region of the brain and the age of the an- ing number of researchers have looked for a tool that
imal, which helps explain how structural plasticity with enables the direct visualization of biological functions.
synapse formation and elimination is involved in the In the last decade, the marriage of FP and FRET has
functional rewiring of neural circuits. Another example made a significant impact on many molecular and cellu-
is recent work in which dendrite growth and synapto- lar investigations. Needless to say, further progress of
genesis were imaged over days in tectal neurons of liv- this technology will require additional marriages or
ing transgenic zebrafish that expressed an Aequorea crossovers. Technological innovations in fluorescence
GFP variant postsynaptically (Niell et al., 2004); the im- imaging of brain functions using FPs will promote multi-
aging revealed that synapse formation directs dendrite disciplinary research in neuroscience, physiology, op-
arborization. The Aequorea GFP variants used in these tics, cell biology, developmental biology, chemistry,
morphological studies will hopefully be replaced with molecular biology, and biochemistry.
improved genetically encoded probes to allow for a bet-
ter understanding of dynamic brain functions. Compar- Acknowledgments
ative studies on the properties of genetically encoded
Ca2+ indicators have been performed using stable The author thank Satoshi Karasawa and Takako Kogure for assis-
tance in figure preparation.
transgenic mouse lines producing inverse pericam
(Nagai et al., 2002) or camgaroo-2 (Baird et al., 1999) un-
References
der the control of the tetracycline-inducible promoter
(Hasan et al., 2004). A similar approach was taken to Albota, M., Beljonne, D., Bredas, J.L., Ehrlich, J.E., Fu, J.Y., Heikal,
characterize G-CaMP, inverse pericam, and camga- A.A., Hess, S.E., Kogej, T., Levin, M.D., Marder, S.R., et al. (1998).
roo-2 in cultured hippocampal slices (Pologruto et al., Design of organic molecules with large two-photon absorption
2004). To date, the performance of these molecules for cross sections. Science 281, 1653–1656.
detecting neural activity seems to vary greatly with the An, S.J., and Almers, W. (2004). Tracking SNARE complex forma-
cell types and the animals used (Reiff et al., 2005). Trans- tion in live endocrine cells. Science 306, 1042–1046.
genic mouse lines that express the updated probes pro- Ando, R., Hama, H., Yamamoto-Hino, M., Mizuno, H., and Miya-
vide promising avenues for future studies. Very recently, waki, A. (2002). An optical marker based on the UV-induced
reliable signals for stimulus-induced Ca2+ transients green-to-red photoconversion of a fluorescent protein. Proc. Natl.
Acad. Sci. USA 99, 12651–12656.
were obtained from parallel fibers in cerebellar slices
Ando, R., Mizuno, H., and Miyawaki, A. (2004). Regulated fast nu-
prepared from transgenic mice expressing G-CaMP2,
cleocytoplasmic shuttling observed by reversible protein highlight-
an improved version of G-CaMP (Dı́ez-Garcı́a et al., ing. Science 306, 1370–1373.
2005).
Ataka, K., and Pieribone, V.A. (2002). A genetically targetable fluo-
rescent probe of channel gating with rapid kinetics. Biophys. J. 82,
Future Perspectives 509–516.
Here, I would like to propose that it is time to evaluate Baird, G.S., Zacharias, D.A., and Tsien, R.Y. (1999). Circular permu-
the present limitations of fluorescence imaging of brain tation and receptor insertion within green fluorescent proteins.
functions using fluorescent proteins. For example, the Proc. Natl. Acad. Sci. USA 96, 11241–11246.
signals from currently available genetically encoded Berney, C., and Danuser, G. (2003). FRET or no FRET: A quantita-
voltage sensors are not yet strong enough to take ad- tive comparison. Biophys. J. 84, 3992–4010.
vantage of the fact that they can be targeted to specific Bevis, B.J., and Glick, B.S. (2002). Rapidly maturing variants of the
cell types. On the other hand, most functional imaging Discosoma red fluorescent protein (DsRed). Nat. Biotechnol. 20,
based on secondary readouts, such as Ca2+ levels 83–87.
and pH, may miss sporadic action-potential firings. I Bozza, T., McGann, J.P., Mombaerts, P., and Wachowiak, M.
would also like to propose that neuroscientists should (2004). In vivo imaging of neuronal activity by targeted expression
of a genetically encoded probe in the mouse. Neuron 42, 9–21.
be particular about the approaches they take to imag-
Campbell, R.E., Tour, O., Palmer, A.E., Steinbach, P.A., Baird, G.S.,
ing. Although the technology for gene transfer, including
Zacharias, D.A., and Tsien, R.Y. (2002). A monomeric red fluores-
construction of transgenic animals, has been greatly cent protein. Proc. Natl. Acad. Sci. USA 99, 7877–7882.
advanced, the introduction of any type of indicator of
Chudakov, D.M., Belousov, V.V., Zaraisky, A.G., Novoselov, V.V.,
brain functions may enhance or suppress related signal- Staroverov, D.B., Lukyanov, S., and Lukyanov, K.A. (2003). Kindling
ing events, thereby perturbing the system being ob- fluorescent proteins for precise in vivo photolabeling. Nat. Biotech-
served. Thus, the process of visualization should be nol. 21, 191–194.
viewed as a potential influence on the observed objects. Chudakov, D.M., Verkhusha, V.V., Staroverov, D.B., Souslova, E.A.,
It is therefore worthwhile to evaluate the ways in which Lukyanov, S., and Lukyanov, K.A. (2004). Photoswitchable cyan
the imaging method impacts the results and the degree fluorescent protein for protein tracking. Nat. Biotechnol. 22,
to which these effects can be minimized. 1435–1439.
Although GFP has been known as a fluorescent pro- Dela Cruz, J.M., Pastirk, I., Comstock, M., Lozovoy, V.V., and
Dantus, M. (2004). Use of coherent control methods through scat-
tein for more than 40 years, it is only in the past dozen
tering biological tissue to achieve functional imaging. Proc. Natl.
years that it has been used as a tool for biological ex- Acad. Sci. USA 101, 16996–17001.
periments. In much the same way, FRET, which was first
Denk, W., and Svoboda, K. (1997). Photon upmanship: why multi-
theorized in 1948 (Förster, 1948), had been primarily photon imaging is more than a gimmick. Neuron 18, 351–357.
used by a small proportion of biologists as a spectro-
Dı́ez-Garcı́a, J., Matsushita, S., Mutoh, H., Nakai, J., Ohkura, M.,
scopic ruler (Stryer, 1978). To push FRET into the larger Yokoyama, J., Dimitrov, D., and Knopfel, T. (2005). Activation of
biological community, a technological advance was cerebellar parallel fibers monitored in transgenic mice expressing
needed. With the emergence of spectral variants of a fluorescent Ca2+ indicator protein. Eur. J. Neurosci. 22, 627–635.
Primer
197

Erickson, M.G., Liang, H., Mori, M.X., and Yue, D.T. (2003). FRET Lukyanov, K.A., Chudakov, D.M., Lukyanov, S., and Verkhusha,
two-hybrid mapping reveals function and location of L-type Ca2+ V.V. (2005). Photoactivatable fluorescent proteins. Nat. Rev. Mol.
channel CaM preassociation. Neuron 39, 97–107. Cell Biol., in press.
Feng, G., Laskowski, M.B., Feldheim, D.A., Wang, H., Lewis, R., Maddox, P.S., Moree, B., Canman, J.C., and Salmon, E.D. (2003).
Frisen, J., Flanagan, J.G., and Sanes, J.R. (2000). Roles for ephrins Spinning disk confocal microscope system for rapid high-resolu-
in positionally selective synaptogenesis between motor neurons tion, multimode, fluorescence speckle microscopy and green fluo-
and muscle fibers. Neuron 25, 295–306. rescent protein imaging in living cells. Methods Enzymol. 360, 597–
Fernández-Alfonso, T., and Ryan, T.A. (2004). The kinetics of syn- 617.
aptic vesicle pool depletion at CNS synaptic terminals. Neuron Marchant, J.S., Stutzmann, G.E., Leissring, M.A., LaFerla, F.M., and
41, 943–953. Parker, I. (2001). Multiphoton-evoked color change of DsRed as an
optical highlighter for cellular and subcellular labeling. Nat. Bio-
Förster, Th. (1948). Intermolecular energy migration and fluores-
cence. Ann. Phy. (Leipzig) 2, 55–75. technol. 19, 645–649.
Matz, M.V., Fradkov, A.F., Labas, Y.A., Savitsky, A.P., Zaraisky,
Griesbeck, O., Baird, G.S., Campbell, R.E., Zacharias, D.A., and
A.G., Markelov, M.L., and Lukyanov, S.A. (1999). Fluorescent pro-
Tsien, R.Y. (2001). Reducing the environmental sensitivity of yellow
teins from nonbioluminescent Anthozoa species. Nat. Biotechnol.
fluorescent protein. Mechanism and applications. J. Biol. Chem.
17, 969–973.
276, 29188–29194.
Miesenböck, G. (2004). Genetic methods for illuminating the func-
Grutzendler, J., Kasthuri, N., and Gan, W.B. (2002). Long-term den-
tion of neural circuits. Curr. Opin. Neurobiol. 14, 395–402.
dritic spine stability in the adult cortex. Nature 420, 812–816.
Miesenböck, G., and Kevrekidis, I.G. (2005). Optical imaging and
Guerrero, G., Siegel, M.S., Roska, B., Loots, E., and Isacoff, E.Y.
control of genetically designed neurons in functioning circuits.
(2002). Tuning FlaSh: Redesign of the dynamics, voltage range,
Annu. Rev. Neurosci. 28, 533–563.
and color of the genetically encoded optical sensor of membrane
potential. Biophys. J. 83, 3607–3618. Miesenböck, G., De Angelis, D.A., and Rothman, J.E. (1998). Visual-
izing secretion and synaptic transmission with pH-sensitive green
Guerrero, G., Rieff, D.F., Agarwal, G., Ball, R.W., Borst, A., Good-
fluorescent proteins. Nature 433, 741–745.
man, C.S., and Isacoff, E.Y. (2005). Heterogeneity in synaptic trans-
mission along a Drosophila larval motor axon. Nat. Neurosci. 8, Miyawaki, A. (2003). Visualization of the spatial and temporal dy-
1188–1196. namics of intracellular signaling. Dev. Cell 4, 295–305.
Gurskaya, N.G., Savitsky, A.P., Yanushevich, Y.G., Lukyanov, S.A., Miyawaki, A., and Tsien, R.Y. (2000). Monitoring protein conforma-
and Lukyanov, K.A. (2001). GFP-like chromoproteins as a source of tions and interactions by fluorescence resonance energy transfer
far-red fluorescent proteins. FEBS Lett. 507, 16–20. between mutants of green fluorescent protein. Methods Enzymol.
327, 472–500.
Hasan, M., Friedrich, R.W., Euler, T., Larkum, M.E., Giese, G.G.,
Both, M., Duebel, J., Waters, J., Bujard, H., Griesbeck, O., et al. Miyawaki, A., Llopis, J., Heim, R., McCaffery, J.M., Adams, J.A.,
(2004). Functional fluorescent Ca2+ indicator proteins in transgenic Ikura, M., and Tsien, R.Y. (1997). Fluorescent indicators for Ca2+
mice under TET Control. PLoS Biol. 2, E163. based on green fluorescent protein and calmodulin. Nature 388,
882–887.
Heim, N., and Griesbeck, O. (2004). Genetically encoded indicators
of cellular calcium dynamics based on troponin C and green fluo- Miyawaki, A., Griesbeck, O., Heim, R., and Tsien, R.Y. (1999). Dy-
rescent protein. J. Biol. Chem. 279, 14280–14286. namic and quantitative Ca2+ measurements using improved camel-
eons. Proc. Natl. Acad. Sci. USA 96, 2135–2140.
Hell, S.W., Dyba, M., and Jakobs, S. (2004). Concepts for nanoscale
resolution in fluorescence microscopy. Curr. Opin. Neurobiol. 14, Mizrahi, A., and Katz, L.C. (2003). Dendritic stability in the adult ol-
599–609. factory bulb. Nat. Neurosci. 6, 1201–1207.
Helmchen, F., and Denk, W. (2002). New developments in multipho- Mongin, O., Porres, L., Moreaux, L., Mertz, J., and Blanchard-
ton microscopy. Curr. Opin. Neurobiol. 12, 593–601. Desce, M. (2002). Synthesis and photophysical properties of new
conjugated fluorophores designed for two-photon excitation. Org.
Higashijima, S., Masino, M.A., Mandel, G., and Fetcho, J.R. (2003).
Lett. 4, 719–722.
Imaging neuronal activity during zebrafish behavior with a geneti-
cally encoded calcium indicator. J. Neurophysiol. 90, 3986–3997. Mori, K., Nagao, H., and Yoshihara, Y. (1999). The olfactory bulb:
coding and processing of odor molecule information. Science
Holtmaat, A.J.G.D., Trachtenberg, J.T., Wilbrecht, L., Shepherd,
286, 711–715.
G.M., Zhang, X., Knott, G.W., and Svoboda, K. (2005). Transient
and persistent dendritic spines in the neocortex in vivo. Neuron Nagai, T., Sawano, A., Park, E.S., and Miyawaki, A. (2001). Circu-
45, 279–291. larly permuted green fluorescent proteins engineered to sense
Ca2+. Proc. Natl. Acad. Sci. USA 98, 3197–3202.
Karasawa, S., Araki, T., Yamamoto-Hino, M., and Miyawaki, A.
(2003). A green-emitting fluorescent protein from Galaxeidae coral Nagai, T., Ibata, K., Park, E.S., Kubota, M., Mikoshiba, K., and Miya-
and its monomeric version for use in fluorescent labeling. J. Biol. waki, A. (2002). A variant of yellow fluorescent protein with fast and
Chem. 278, 34167–34171. efficient maturation for cell-biological applications. Nat. Biotechnol.
20, 87–90.
Karasawa, S., Araki, T., Nagai, T., Mizuno, H., and Miyawaki, A.
(2004). Cyan-emitting and orange-emitting fluorescent proteins as Nagai, T., Yamada, S., Tominaga, T., Ichikawa, M., and Miyawaki, A.
a donor/acceptor pair for fluorescence resonance energy transfer. (2004). Expanded dynamic range of fluorescent indicators for Ca2+
Biochem. J. 381, 307–312. by circularly permuted yellow fluorescent proteins. Proc. Natl.
Acad. Sci. USA 101, 10554–10559.
Kawano, H., Nabekawa, Y., Suda, A., Oishi, Y., Mizuno, H., Miya-
waki, A., and Midorikawa, K. (2003). Attenuation of photobleaching Nakai, J., Ohkura, M., and Imoto, K. (2001). A high signal-to-noise
in two-photon excitation fluorescence from green fluorescent pro- Ca2+ probe composed of a single green fluorescent protein. Nat.
tein with shaped excitation pulses. Biochem. Biophys. Res. Com- Biotechnol. 19, 137–141.
mun. 311, 592–596. Neil, M.A.A., Juskaitis, R., and Wilson, T. (1997). Method of obtain-
Kim, J.H., Udo, H., Li, H.L., Youn, T.Y., Chen, M., Kandel, E.R., and ing optical sectioning by using structured light in a conventional mi-
Bailey, C.H. (2003). Presynaptic activation of silent synapses and croscope. Opt. Lett. 22, 1905–1907.
growth of new synapses contribute to intermediate and long-term Niell, C.M., and Smith, S.J. (2004). Live optical imaging of nervous
facilitation in Aplysia. Neuron 40, 151–165. system development. Annu. Rev. Physiol. 66, 771–798.
Kuner, T., and Augustine, G.J. (2000). A genetically encoded ratio- Niell, C.M., Meyer, M.P., and Smith, S.J. (2004). In vivo imaging of
metric indicator for chloride: capturing chloride transients in cul- synapse formation on a growing dendritic arbor. Nat. Neurosci. 7,
tured hippocampal neurons. Neuron 27, 447–459. 254–260.
Neuron
198

Ng, M., Roorda, R.D., Lima, S.Q., Zemelman, B.V., Morcillo, P., and Suzuki, H., Kerr, R., Bianchi, L., Frokjaer-Jensen, C., Slone, D., Xue,
Miesenböck, G. (2002). Transmission of olfactory information be- J., Gerstbrein, B., Driscoll, M., and Schafer, W.R. (2003). In vivo
tween three populations of neurons in the antennal lobe of the imaging of C. elegans mechanosensory neurons demonstrates
fly. Neuron 36, 463–473. a specific role for the MEC-4 channel in the process of gentle touch
Okamoto, K., Nagai, T., Miyawaki, A., and Hayashi, Y. (2004). Rapid sensation. Neuron 39, 1005–1017.
and persistent modulation of actin dynamics regulates postsynap- Takao, K., Okamoto, K., Nakagawa, T., Neve, R.L., Nagai, T., Miya-
tic reorganization underlying bidirectional plasticity. Nat. Neurosci. waki, A., Hashikawa, T., Kobayashi, S., and Hayashi, Y. (2005). Visu-
7, 1104–1112. alization of synaptic Ca2+/calmodulin-dependent protein kinase II
Oray, S., Majewska, A., and Sur, M. (2004). Dendritic spine dynam- activity in living neurons. J. Neurosci. 25, 3107–3112.
ics are regulated by monocular deprivation and extracellular matrix Terskikh, A., Fradkov, A., Ermakova, G., Zaraisky, A., Tan, P.,
degradation. Neuron 44, 1021–1030. Kajava, A.V., Zhao, X., Lukyanov, S., Matz, M., Kim, S., et al.
Palmer, A.E., Jin, C., Reed, J.C., and Tsien, R.Y. (2004). Bcl-2-me- (2000). ‘‘Fluorescent timer’’: protein that changes color with time.
diated alterations in endoplasmic reticulum Ca2+ analyzed with an Science 290, 1585–1588.
improved genetically encoded fluorescent sensor. Proc. Natl. Terskikh, A.V., Fradkov, A.F., Zaraisky, A.G., Kajava, A.V., and An-
Acad. Sci. USA 101, 17404–17409. gres, B. (2002). Analysis of DsRed mutants. Space around the fluo-
Patterson, G.H., and Piston, D.W. (2000). Photobleaching in two- rophore accelerates fluorescence development. J. Biol. Chem. 277,
photon excitation microscopy. Biophys. J. 78, 2159–2162. 7633–7636.
Patterson, G.H., and Lippincott-Schwartz, J. (2002). A photoactivat- Theer, P., Hasan, M.T., and Denk, W. (2003). Two-photon imaging
able GFP for selective photolabeling of proteins and cells. Science to a depth of 1000 mm in living brains by use of a Ti:Al2O3 regener-
297, 1873–1877. ative amplifier. Opt. Lett. 28, 1022–1024.
Pologruto, T.A., Yasuda, R., and Svoboda, K. (2004). Monitoring Trachtenberg, J.T., Chen, B.E., Knott, G.W., Feng, G., Sanes, J.R.,
neural activity and [Ca2+] with genetically encoded Ca2+ indicators. Welker, E., and Svoboda, K. (2002). Long-term in vivo imaging of
J. Neurosci. 24, 9572–9579. experience-dependent synaptic plasticity in adult cortex. Nature
420, 788–794.
Prasher, D.C., Eckenrode, V.K., Ward, W.W., Prendergast, F.G., and
Cormier, M.J. (1992). Primary structure of the Aequorea Victoria Truong, K., Sawano, A., Mizuno, H., Hama, H., Tong, K.I., Miyawaki,
green fluorescent protein. Gene 111, 229–233. A., and Ikura, M. (2001). FRET-based in vivo Ca2+ imaging by a new
Reiff, D.F., Ihring, A., Guerrero, G., Isacoff, E.Y., Joesch, M., Nakai, calmodulin-GFP fusion molecule. Nat. Struct. Biol. 8, 1069–1073.
J., and Borst, A. (2005). In vivo performance of genetically encoded Tsien, R.Y. (1998). The green fluorescent protein. Annu. Rev. Bio-
indicators of neural activity in flies. J. Neurosci. 25, 4766–4778. chem. 67, 509–544.
Rizzo, M.A., Springer, G.H., Granada, B., and Piston, D.W. (2004). Tsutsui, H., Karasawa, S., Shimizu, H., Nukina, N., and Miyawaki, A.
An improved cyan fluorescent protein variant useful for FRET. (2005). Semi-random engineering of a coral fluorescent protein into
Nat. Biotechnol. 22, 445–449. an efficient highlighter. EMBO Rep. 6, 233–238.
Sakai, R., Repunte-Canonigo, V., Raj, C.D., and Knopfel, T. (2001). Verkhusha, V.V., and Sorkin, A. (2005). Conversion of the mono-
Design and characterization of a DNA-encoded, voltage-sensitive meric red fluorescent protein into a photoactivatable probe.
fluorescent protein. Eur. J. Neurosci. 13, 2314–2318. Chem. Biol. 12, 279–285.
Samuel, A.D., Silva, R.A., and Murthy, V.N. (2003). Synaptic activity Walsh, M.K., and Lichtman, J.W. (2003). In vivo time-lapse imaging
of the AFD neuron in Caenorhabditis elegans correlates with ther- of synaptic takeover associated with naturally occurring synapse
motactic memory. J. Neurosci. 23, 373–376. elimination. Neuron 37, 67–73.
Sasaki, E., Kojima, H., Nishimatsu, H., Urano, Y., Kikuchi, K., Hirata, Wang, F., Nemes, A., Mendelsohm, M., and Axel, R. (1998). Odorant
Y., and Nagano, T. (2005). Highly sensitive near-infrared fluorescent receptors govern the formation of a precise topographic map. Cell
probes for nitric oxide and their application to isolated organs. 93, 47–60.
J. Am. Chem. Soc. 127, 3684–3685. Wang, J.W., Wong, A.M., Flores, J., Vosshall, L.B., and Axel, R.
Shagin, D.A., Barsova, E.V., Yanushevich, Y.G., Fradkov, A.F., (2003). Two-photon calcium imaging reveals an odor-evoked map
Lukyanov, K.A., Labas, Y.A., Semenova, T.N., Ugalde, J.A., Meyers, of activity in the fly brain. Cell 112, 271–282.
A., Nunez, J.M., et al. (2004). GFP-like proteins as ubiquitous meta- Wang, L., Jackson, W.C., Steinbach, P.A., and Tsien, R.Y. (2004a).
zoan superfamily: evolution of functional features and structural Evolution of new nonantibody proteins via iterative somatic hyper-
complexity. Mol. Biol. Evol. 21, 841–850. mutaion. Proc. Natl. Acad. Sci. USA 101, 16745–16749.
Shaner, N.C., Campbell, R.E., Steinbach, P.A., Giepmans, B.N., Wang, Y., Guo, H.F., Pologruto, T.A., Hannan, F., Hakker, I.,
Palmer, A.E., and Tsien, R.Y. (2004). Improved monomeric red, or- Svoboda, K., and Zhong, Y. (2004b). Stereotyped odor-evoked ac-
ange and yellow fluorescent proteins derived from Discosome sp. tivity in the mushroom body of Drosophila revealed by green fluo-
red fluorescent protein. Nat. Biotechnol. 22, 1567–1572. rescent protein-based Ca2+ imaging. J. Neurosci. 24, 6507–6514.
Shen, K., and Meyer, T. (1999). Dynamic control of CaMKII translo- Wiedenmann, J., Schenk, A., Rocker, C., Girod, A., Spindler, K.D.,
cation and localization in hippocampal neurons by NMDA receptor and Nienhaus, G.U. (2002). A far-red fluorescent protein with fast
stimulation. Science 284, 162–166. maturation and reduced oligomerization tendency from Entacmaea
Shimomura, O., Johnson, F.H., and Saiga, Y. (1962). Extraction, pu- quadricolor (Anthozoa, Actinaria). Proc. Natl. Acad. Sci. USA 99,
rification and properties of aequorin, a bioluminescent protein from 11646–11651.
the luminous hydromedusan, Aequorea. J. Cell. Comp. Physiol. 59, Wiedenmann, J., Ivanchenko, S., Oswald, F., Schmitt, F., Rocker,
223–240. C., Salih, A., Spindler, K.D., and Nienhaus, G.U. (2004). EosFP,
Shkrob, M.A., Yanushevich, Y.G., Chudakov, D.M., Gurskaya, N.G., a fluorescent marker protein with UV-inducible green-to-red fluo-
Labas, Y.A., Poponov, S.Y., Mudrik, N.N., Lukyanov, S., and Lukya- rescence conversion. Proc. Natl. Acad. Sci. USA 101, 15905–15910.
nov, K.A. (2005). Far-red fluorescent proteins evolved from a blue Wilson, R.I., Turner, G.C., and Laurent, G. (2004). Transformation of
chromoprotein from Actinia equina. Biocem. J., in press. olfactory representations in the Drosophila antennal lobe. Science
Siegel, M.S., and Isacoff, E.Y. (1997). A genetically encoded optical 303, 366–370.
probe of membrane voltage. Neuron 19, 735–741. Yu, D., Ponomarev, A., and Davis, R.L. (2004). Altered representa-
Stryer, L. (1978). Fluorescence energy transfer as a spectroscopic tion of the spatial code for odors after olfactory classical condition-
ruler. Annu. Rev. Biochem. 47, 819–846. ing; memory trace formation by synaptic recruitment. Neuron 42,
Suh, G.S., Wong, A.M., Hergarden, A.C., Wang, J.W., Simon, A.F., 437–449.
Benzer, S., Axel, R., and Anderson, D.J. (2004). A single population Zacharias, D.A., Violin, J.D., Newton, A.C., and Tsien, R.Y. (2002).
of olfactory sensory neurons mediates an innate avoidance behav- Partitioning of lipid-modified monomeric GFPs into membrane mi-
iour in Drosophila. Nature 431, 854–859. crodomains of live cells. Science 296, 913–916.
Primer
199

Zal, T., and Gascoigne, N.R.J. (2004). Photobleaching-corrected

FRET efficiency imaging of live cells. Biophys. J. 86, 3923–3939.
Zhang, J., Campbell, R.E., Ting, A.Y., and Tsien, R.Y. (2002). Creat-
ing new fluorescent probes for cell biology. Nat. Rev. Mol. Cell Biol.
3, 906–918.
Zipfel, W.R., Williams, R.M., and Webb, W.W. (2003). Nonlinear
magic: multiphoton microscopy in the biosciences. Nat. Biotech-
nol. 21, 1369–1377.
Zochowski, M., Wachowiak, M., Falk, C.X., Cohen, L.B., Lam, Y.-W.,
Antic, S., and Zecevic, D. (2000). Concepts in imaging and micros-
copy. Imaging membrane potential with voltage-sensitive dyes.
Biol. Bull. 198, 1–21.