Chemistry & Biology
Review
Bridging the Gaps
between Synapses, Circuits, and Behavior
Pamela M. England1,*
1Departments of Pharmaceutical Chemistry and Cellular and Molecular Pharmacology, University of California, San Francisco,
San Francisco, CA 94158, USA
*Correspondence: england@picasso.ucsf.edu
DOI 10.1016/j.chembiol.2010.06.001
The decade of the brain may have come and gone, but the final frontier, cracking the neuronal code, still lies
ahead. Today, new technologies that allow precise spatiotemporal remote control over the activity of genetically
defined populations of neurons within intact neural circuits are providing a means of obtaining a functional
wiring diagram of the mammalian brain, bringing us one step closer to understanding precisely how
neuronal activity codes for perception, thought, emotion, and action. These technologies and the design
principles underlying them are reviewed herein.
Introduction
In 1990, the U.S. Congress designated the 1990s the ‘‘Decade of
the Brain,’’ and then-sitting president George H.W. Bush proclaimed
that ‘‘a new era of discovery is dawning in brain
research.’’ In the ensuing years, our understanding of the
mammalian brain greatly advanced. At the level of individual
neurons, tremendous progress was made identifying the molecular
mechanisms and components underlying signaling within
and between neurons (Sudhof and Malenka, 2008). At the
same time, functional imaging technologies were used to correlate
the activity of specific brain regions with particular brain
functions (Haber and Knutson, 2010; Maguire, 2001). Despite
these advances, the relationship between the activity of individual
neurons and higher brain function remained poorly understood.
Today, new tools are being developed that will build
a bridge between the activity of individual neurons and the
behavior of whole animals by providing remote control over the
functional neuronal circuitry that lies between the two.
These bridging technologies rely on a combination of genetic
engineering, to target functionally circumscribed populations of
neurons, and light and/or chemicals, to rapidly control the
activity of those neurons. Genetic strategies for targeting
specific neurons have recently been reviewed elsewhere
(Bernard et al., 2009; Luan and White, 2007; Luo et al., 2008).
Briefly, DNA sequences (promoters) unique to the desired population
of neurons are used to drive the expression of exogenous
proteins in those neurons. Inspired by key control points in
neuronal signaling, these exogenous proteins are designed to
place the neuronal circuitry governing behavior under the control
of exogenous light and/or chemicals. These control points, and
several of the technologies developed to rapidly and reversibly
manipulate them, are described in the following sections.
Key Control Points in Neuronal Signaling
A typical neuron is comprised of a cell body (soma), many short
processes called dendrites that receive information from other
neighboring cells, and one long process with many branches,
called the axon, that transmits information to other neurons
(Figure 1). Signaling along and between neurons depends
critically on the potential of the membrane, the release of
neurotransmitter, and the subsequent activation of postsynaptic
receptors.
Membrane Potential
All neurons maintain a voltage difference across the plasma
membrane known as the membrane potential. This membrane
typically rests at –70 mV, meaning the voltage inside the cell is
70 mV more negative than the outside of the cell. This polarization
is accomplished by ion channels and transporters that
move Na+
ions out of the cell and K+
ions into the cell (creating
a concentration gradient for these ions). Excitatory and inhibitory
inputs to a neuron cause the membrane potential to rise (depolarize)
or fall (hyperpolarize), respectively. If enough depolarization
accumulates to bring the membrane potential to approximately
–55 mV (threshold potential), a pulse of electricity called
the action potential is passed from the cell body, along the
axon, to the end of the neuron. At the biophysical level, action
potentials result from the coordinated opening and closing of
voltage-gated Na+
and K+
channels. When the membrane potential
crosses the threshold potential, voltage-gated Na+
channels
open, Na+
rushes down its concentration gradient into the cell,
and the membrane potential abruptly shoots upward, typically
reaching a peak at approximately +40 mV. The opening of Na+
channels is followed by the opening of voltage-gated K+
channels
that permit the exit of K+
out of the cell. Sodium channels
close at the peak of the action potential, while K+
continues to
leave the cell. The continued efflux of K+
briefly hyperpolarizes
the cell (–80 mV) before the K+
channels close to permit the
cell to return to the resting membrane potential (–70 mV). The
size and shape of the action potential does not vary, regardless
of the magnitude of the stimulus that caused it. Rather, axonal
information contained in the stimulus intensity is coded in the
frequency of successive action potentials, which can range
from about 1 to 100 per second.
Neurotransmitter Release
Axonal signals arriving at the synapse in the form of action potentials
are transformed by the synapse into a chemical signal.
When the action potential reaches the end of the axon (nerve
terminal), the associated change in membrane potential induces
the opening of voltage-gated calcium channels. The subsequent
influx of Ca+2
ions into the cell causes vesicles containing
Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved 607
neurotransmitter (synaptic vesicles) to fuse with the cell
membrane and release neurotransmitter into the synaptic cleft,
the space between the axon terminal and the dendrite of another
neuron.
Postsynaptic Receptor Activity
The frequency of action potentials arriving at the nerve terminal
determines the concentration of neurotransmitter in the synaptic
cleft, and postsynaptic neurotransmitter-gated receptors
compute this information. Neurotransmitters released by the
axon terminal rapidly diffuse (0.3 to 1.0 ms) across the synaptic
cleft to the postsynaptic membrane. Subsequent opening of
neurotransmitter-gated receptors allows the flow of ions into
the cell, which produces a small change in the membrane potential
called a postsynaptic potential. Glutamate, the principal
excitatory neurotransmitter in the mammalian brain, causes
postsynaptic AMPA receptors to open, and the resulting influx
of sodium ions into the cell depolarizes the membrane and
produces an excitatory postsynaptic potential (EPSP). GABA,
the principal inhibitory neurotransmitter causes postsynaptic
GABAA receptors to open, and the resulting flow of ClÀ
ions
into the cell hyperpolarizes the membrane and produces an
inhibitory postsynaptic potential (IPSP). Each axon has multiple
branches, forming synapses with many dendrites, and each
dendrite may receive inputs from multiple axons. The summed
activity of postsynaptic receptors across multiple synapses
may cause the membrane to reach threshold and fire an action
potential, depending on the nature (EPSP versus IPSP) and
sequence of postsynaptic events.
The following sections discuss how these three main control
points (membrane potential, neurotransmitter release, postsynaptic
receptor activity), combined with the selective activation
of genetically targeted exogenous proteins using light and/or
chemicals (Figure 2), are being used to provide precise control
over neuronal activity. The technologies discussed are limited
to those that have proven to provide reversible control over the
behavior of intact organisms.
Controlling Membrane Potential
Various portable systems for reversibly controlling the
membrane potential of neurons in vivo have been developed.
Some strategies rely on components of metabotropic signaling
cascades to stimulate or inhibit the firing of action potentials,
whereas others use ion channels and pumps that are directly
gated by light and/or chemicals. These approaches are discussed
below.
chARGe/Retinal
One of the first portable systems developed to provide rapid,
reversible remote control over membrane potential used three
components of the fly visual system to transduce an optical stimulus
into neural activity (Zemelman et al., 2002). The three
components—arrestin, rhodopsin (opsin complexed with
retinal), and the alpha subunit of cognate heterotrimeric
G-protein—comprise what is termed the chARGe system. In
this system, the light-gated generation of membrane depolarization
begins with the absorption of light by rhodopsin (430–550
nm absorption peak), which triggers the 11-cis to all-trans photoisomerization
of bound retinal. The resulting activated rhodopsin
catalyzes the exchange of GDP for GTP on the G-protein alpha
subunit, which then dissociates and activates phospholipase
C. This leads to the opening of two classes of endogenous
calcium-permeable channels, the transient receptor potential
(TRP) and TRP-like (TRPL) channels, which depolarize the
membrane. Binding of arrestin inactivates the rhodopsin, and
absorption of lower energy light ($580 nm) isomerizes bound
all-trans retinal to 11-cis retinal and releases the regenerated
rhodopsin from its complex with arrestin.
Ectopic expression of the chARGe system in cultured hippocampal
neurons, with alternating periods of light and darkness,
produced alternating episodes of electrical activity (peak firing
frequency of 7.5 Hz) and quiescence. However, action potentials
appeared and disappeared with unpredictable frequencies and
lag periods (hundreds of milliseconds to tens of seconds) after
the light stimulus was applied and removed. The normally tight
axon
dendrites
propagated
action potential
axon
branches
axon
terminal
1
1 2 3
2,3
threshold
stimulus Na+
Ca+2
presynaptic
neuron
membrane potential neurotransmitter release postsynaptic receptor
activity
graded receptor potentials
time
postsynaptic
neuron
synapse
synaptic
vesicle
neurotransmitter
+40
0
0 10 20 30 40 50
-55
-70
MembraneVoltage(mV)
Time (ms)
resting
potential
spike rate
Figure 1. Cartoon Depicting Synaptic Transmission in a Typical Neuron and the Three Control Points Used to Manipulate This Signaling
Action potentials are generated in the presynaptic neuron when the membrane potential is depolarized to a threshold potential. At the nerve terminal, this electrical
signal drives the release of neurotransmitter from synaptic vesicles. Binding of the neurotransmitter to ligand-gated ion channels on the postsynaptic
membrane induces graded postsynaptic receptor potentials that may sum to depolarize the membrane to its threshold value and, thereby, generate action potentials
at various frequencies.
608 Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved
Chemistry & Biology
Review
coupling between stimulus intensity and response frequency
characteristic of the native Drosophila photoreceptor system is
apparently relaxed in the transplanted three-component
chARGe system. Although this system would not prove to be
of practical value in intact animals, it nonetheless demonstrated
that synthetic neuronal signals could be inserted into neurons
and introduced the concept of achieving specificity by controlling
susceptibility to a stimulus, rather than the stimulus itself.
P2X2/ATP
Following the development of the chARGe system, it was recognized
that using ionotropic receptors for selective control over
designated populations of neurons would overcome many of
the limitations associated with using metabotropic receptors.
Instead of relying on a metabotropic cascade to trigger the
activity of an ion channel and transduce the signal, the excitatory
stimulus could be directly transduced by an orthogonal ion
channel, thus accelerating the response kinetics. Two ion channels
used for this purpose are the ATP-gated P2X2 receptor and
the capsaicin-gated TRPV1 receptor (Claridge-Chang et al.,
2009; Lima and Miesenbock, 2005; Miesenbock and Kevrekidis,
2005; Zemelman et al., 2003). Although these nonselective
cation channels are primarily expressed in the peripheral nervous
system of mammals, they are also present in the central nervous
425-550 nm
~ 380 nm
~ 500 nm
trans retinal cis retinal
trans MAG
DMNPE-ATP (caged ATP)
allatostatin ivermectin
zolpidem clozapine-N-oxide
DMNB-capsaicin (caged capsaicin)
CID AP20187
R = GluK2(L439C)
cis MAG
N
H
R
13 13
N
H
R
N
N
H
N
N
H
O
O
HO2C CO2H
NH2
N
H
O
N
O
O
R
N
N
N
H
O
N
O
O
R
NH
HN
O
O
CO2H
HO2C NH2
O
H
N
H
N
O
O
N
O
O
O
O
O
N
O
O
O
O
O
O
O
O
N
O
O
O
O
O
O
OO
OH
OH
R
OO
OCH3
OO
OCH3
N
N
N
O
OO
O
OCH3
NO2
H3CO
N
H
O
O
N
N
N N
NH2
OHHO
O
P
O
P
O
P
O
O
HO
O
HO
O
HO
NO2H3CO
H3CO
H2N
N
H
N
OH
O
O
H
NH2N
HN
NH
O
HN
NH
OH
O
OH
O
HN
NH
O
O
NH2
O
N
N
H
Cl
N
N
OFigure
2. Structures of the Chemicals Used
in Combination with Ectopically Expressed
Proteins to Manipulate the Activity of
Neurons
The light-activated isomerization of all-trans retinal
to cis retinal is used to gate the cation channel
ChR2 and the anion pump NpHR. Structure shown
is the Schiff’s base formed between retinal and
a lysine residue in the protein. The light-gated
isomerization of MAG, a cysteine reactive azobenzene-tethered
glutamate analog, is used to gate
the glutamate-gated ion channel GluK2. The
light-activated uncaging of caged ATP and caged
capsaicin are used to gate P2X2 and capsaicin
(TRPV1) receptors, respectively. The octapeptide
allatostatin activates the allatostatin receptor. Ivermectin
selectively activates a glutamate-gated
chloride channel. The synthetic ligand AP20187
induces the homodimerization of proteins containing
the FKBP ligand-binding domain. AL, allatostatin;
ATP,adenosine-50
-triphosphate;CID,chemical
inducer of dimerization; CNO, clozapine-N-oxide;
DMNB-capsaicin, 4,5-dimethoxy-2-nitrobenzylcapsaicin;
DMNPE-ATP, P3
-[1-(4,5-dimethoxy-2nitrophenyl)ethyl]-ATP;
IVM, ivermectin; MAG,
cysteine-reactive maleimide, photoisomerizable
azobenze, glutamate-based agonist; MIST,
molecular systems for inactivation of synaptic
transmission.
system, somewhat limiting the utility of
these systems in such animals.
Heterologous expression of either P2X2
or TRPV1 channels in cultured hippocampal
neurons, followed by application
of the cognate agonist, led to brief bursts
of action potentials at frequencies as high
as 40 Hz (Zemelman et al., 2003). The
current amplitudes and firing frequencies
could be tuned by varying the concentration
of agonist, though dose-response relationships were
complicated. The high firing rates were not sustainable over
extended periods of time, however; high-frequency activity
was followed by long-lasting plateaus, during which the
membrane remained depolarized but action potentials were
absent. These limitations associated with pharmacologically
activating the receptors were overcome by delivering pulses of
agonist using caged ATP and caged capsaicin combined with
flashes of light. In particular, ATP and capsaicin were photochemically
liberated from DMNPE-ATP and DMNB-capsaicin
with near-UV wavelength light. Flashes (1 s) of unfocused light
successfully triggered the firing of action potentials at frequencies
of 15–40 Hz. The onset of activity lagged behind the optical
stimulus ($5 s with TRPV1, $1 s with P2X2), and lasted slightly
longer than the light exposure ($2.6 s with TRPV1, 2.4 s with
P2X2). Additional studies have further characterized the ability
of TRPV1 and a related channel, TRPM8, to drive action potentials
in cultured neurons (Crawford et al., 2009).
The P2X2/ATP system was subsequently introduced into the
Drosophila central nervous system (which lacks purinoceptors)
where it was used to evoke action potentials and drive behaviors
in the intact fly (Claridge-Chang et al., 2009; Lima and Miesenbock,
2005). Specifically, P2X2 receptors were ectopically
Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved 609
Chemistry & Biology
Review
expressed in small subsets of central neurons (representing just
$0.05% of the total CNS population), and caged ATP was microinjected
through the eye into the brain. Brief illumination (150–
250 ms) drove escape movements, including flying. The efficacy
of photostimulation declined with a half-life of $75 min after the
injection of caged ATP, reflecting the clearance of the caged
compound from the CNS. Ectopic expression of TRPV1 in nociceptor
neurons of Caenorhabditis elegans combined with capsaicin
application was used to elicit ‘‘synthetic’’ avoidance behaviors
that are lacking in wild-type animals (Tobin et al., 2002).
ChR2/Retinal
The most widely used system thus far developed for remotely
controlling membrane potential combines a light sensor and an
ionic conductance within one protein (Boyden et al., 2005). The
protein is a light-gated cation channel called channel rhodopsin
2 (ChR2) derived from the green alga Chlamydomonas reinhardtii
(Nagel et al., 2003; Sineshchekov et al., 2002; Suzuki et al.,
2003). The photosensor is a molecule of all-trans retinal bound
to the protein core. Illumination with blue ($470 nm) light causes
the retinal to isomerize, inducing the channel to open. The resulting
influx of sodium ions into the cell depolarizes the membrane.
The ability of ChR2 to drive action potentials was initially characterized
in cultured hippocampal neurons (Boyden et al., 2005).
Photostimulation drove defined trains of action potentials with
frequencies ranging as high as 30 Hz, or subthreshold synaptic
events, depending on the light intensity. The delay between the
light flash and the beginning of the photocurrent is <50 ms and,
although the channel inactivates with sustained light, it rapidly
recovers in the dark (tau $5 s) due to the reisomerization of
cis-retinal to the all-trans ground state. Thus, multiple pulses of
blue light, interspersed by periods of darkness, could be used
to drive action potentials at physiological frequencies with millisecond
temporal resolution. A drawback to ChR2 is that the
number of channels that needs to be localized to the plasma
membrane in order to successfully drive action potentials is
rather high (100,000 to 1 million) owing to the low conductance
($50 femtosiemans) of the channel.
ChR2 has been used in vivo in several organisms, including zebrafish
(Douglass et al., 2008), worms (Nagel et al., 2005), flies
(Schroll et al., 2006), mice (Bi et al., 2006; Petreanu et al.,
2007; Wang et al., 2007), rats (Abbott et al., 2009), and monkeys
(Han et al., 2009), to investigate how brain states and behaviors
are generated by neural circuits. Notably, the endogenous
expression of retinal in mammals eliminates the need to introduce
the ligand exogenously. In addition to ChR2, several other
microbial opsin-based tools are being developed to interrogate
the organization and function of neuronal circuits, including a
yellow-light-sensitive cation channel (VchR1) (Zhang et al., 2008),
a yellow-light-sensitive proton pump (Arch) (Chow et al., 2010),
a blue-light-sensitive proton pump (Mac) (Chow et al., 2010),
andablue-light-sensitive chloridepump(NphR,videinfra)(Zhang
et al., 2007). Several reviews of these opsins are available
(Gradinaru et al., 2007; Gunaydin et al., 2010; Zhang et al., 2010).
GluCl/IVM
Paralleling the development of portable remote-controlled
systems for exciting neuronal activity has been the development
of systems for reversibly silencing the activity of specific
neurons. One strategy developed for this purpose uses pharmacological
manipulation of a glutamate-gated chloride channel
(GluCl) derived from Caenorhabditis elegans (Lerchner et al.,
2007; Slimko and Lester, 2003; Slimko et al., 2002). Glutamate-gated
chloride channels are found in several invertebrate
phyla, but not in mammals, and are the targets for various antiparasitic
drugs, such as ivermectin (IVM). Thus, targeted expression
of GluCl in mammalian neurons, combined with pharmacological
activation with IVM, could be used to hyperpolarize
neurons and, thereby, inhibit the generation of action potentials.
Indeed, in hippocampal neurons expressing GluCl, robust chloride
currents were induced in response to IVM, and the
membrane potential was fixed at values ranging between –50
and –65 mV. The time to activate the chloride current was
dose dependent (5–500 nM IVM), with the fastest time constant
(6 s) observed at the highest concentration. The silencing effect
of IVM reversed following washout of the drug, but required
several hours ($8 hr).
The GluCl/IVM system later proved to be an effective strategy
for manipulating the behavior of awake-behaving mice (Lerchner
et al., 2007). The two subunits of GluCl receptor (a and b) were
expressed in the mouse striatum, and IVM was introduced intraperitoneally.
Behavioral effects were observed 4 hr after a single
i.p. injection of IVM and reached maximal levels by 12 hr. The
effects were almost fully reversed by 4 days and multiple cycles
of silencing and recovery in a single animal were demonstrated.
Recently a modified human alpha1 glycine receptor (GlyR) was
reported as an improved ivermectin-gated silencing receptor
(Lynagh and Lynch, 2010). Like GluCl, gating of the glycine
receptor by ivermectin produces an inward chloride flux that
hyperpolarizes the membrane. The crucial improvement is that
the sensitivity of the receptor to glycine, an endogenous neurotransmitter,
was eliminated though the introduction of a single
point mutation (F207A). The sensitivity of GlyR to ivermectin
was rendered similar to that of GluCl through the introduction
of a second point mutation (A288G) in the receptor.
AlstR/AL
Another system developed for silencing neuronal activity in
genetically defined populations of neurons uses the allatostatin
(AlstR) receptor, a G protein-coupled receptor from Drosophila,
and its cognate ligand allatostatin (AL), an octapeptide (Birgul
et al., 1999; Lechner et al., 2002; Luo et al., 2008; Tan et al.,
2006; Wehr et al., 2009). Although the AlstR is structurally related
to the mammalian somatostatin/galanin/opiod receptor family,
there is no similarity between AL and mammalian peptides
such as somatostatin, galanin, or enkephalins. Thus, the activity
of the AlstR can be solely controlled by AL, which has a very high
apparent affinity (EC50 of 55 pM) for the receptor. Stimulation of
the AlstR leads to activation of G protein-gated inwardly rectifying
potassium channels (GIRKs) at subthreshold voltages, via
free G-protein bg subunits. The resulting efflux of K+
ions hyperpolarizes
the membrane, thus preventing the firing of action
potentials. The AlstR/AL receptor/ligand system has been used
to silence activity of cortical neurons quickly and reversibly. In
cortical slices from ferrets ectopically expressing the AlstR,
application of AL inhibited the firing of action potentials within
minutes, and the effect reversed within 15 min upon washout
of the ligand (Lechner et al., 2002).
The utility of the AlstR/AL system was subsequently demonstrated
in vivo in cortical and thalamic neurons of mice, rats,
ferrets, and monkeys (Tan et al., 2006; Wehr et al., 2009). AL
610 Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved
Chemistry & Biology
Review
was applied to anesthetized animals directly onto the cortical
surface (exposed by removing a portion of the skull). Neurons inactivated
within seconds or minutes (depending on the preparation)
upon application of AL and recovered within minutes of
saline washout.
RASSL/CNO
Another approach for controlling the activity of neurons uses G
protein-coupled receptors that have been reengineered such
that they are exclusively activated by synthetic ligands possessing
no biological activity (Conklin et al., 2008). In particular,
a family of muscarinic acetylcholine receptors (mAChRs) that
are potently activated by the pharmacologically inert compound
clozapine-N-oxide (CNO), but not by the native ligand acetylcholine
(ACh), were developed using directed molecular evolution
(Alexander et al., 2009; Armbruster et al., 2007). Double mutation
of the human M3 receptor (Y149C3.33
and A239G5.46
) yielded
a GPCR that is potently ($20–30 nM) activated by CNO, with
a sever reduction (>40,000-fold) in ACh potency compared
with wild-type receptors. Activation with CNO of similarly
mutated human M4 receptors, which couple to Gi/o and can
induce the activation of GIRKs, was shown to silence cultured
hippocampal neurons (Armbruster et al., 2007).
In transgenic mice expressing the mutated M3 receptors in
forebrain principal neurons, peripheral administration of CNO
(which readily crosses the blood–brain barrier) produced
changes in the activity of hippocampal neurons and in the
behavior of whole animals. Specifically, activation of these M3
receptors, which couple to Gq and stimulate phospholipase C,
increased the firing rate of hippocampal interneurons and, in
hippocampal slices isolated from these animals, increased the
firing rate of excitatory neurons (CA1 pyramidal cells). At low
doses,CNOincreasedtheactivity(locomotion, episodesofbeam
breaks) of the mice, and at higher doses evoked limbic seizures.
NpHR/Retinal
Halorhodopsin (NpHR) is a light-gated chloride pump from the
archaebacterium Natronomonas pharaonis (Kalaidzidis et al.,
1998; Schobert and Lanyi, 1982). This seven-transmembrane
protein, which has homology to the light-driven proton pump
bacteriorhodopsin, uses all-trans retinal as the photosensor. Illumination
with yellow ($570 nm) light causes the retinal to isomerize,
activating the pump. The resulting influx of chloride ions
into the cell hyperpolarizes the membrane, thus suppressing
the firing of action potentials. NpHR silencing has been demonstrated
electrophysiologically in cultured neurons (Zhang et al.,
2007; Zhao et al., 2008). Illumination with yellow light induced
rapid currents on the millisecond timescale ($26 ms delay
from light onset to 50% of the hyperpolarization peak). The
pump current exhibited a linear voltage dependence and was
active across the entire physiological voltage range.
NpHR has also been successfully used in vivo to paralyze
worms (Zhang et al., 2007) and to control the swimming of
zebrafish (Arrenberg et al., 2009). Used in combination with
ChR2 (vide supra), NpHR enables multiple-color optical activation,
silencing, and desynchronization of neural activity (Zhang
et al., 2007).
Controlling Neurotransmitter Release
Two systems have been developed for inhibiting neurotransmitter
release in genetically specified populations of neurons
in vivo. One method uses the inducible expression of tetanus
toxin light chain (TNT), which cleaves VAMP2/synaptobrevin,
a synaptic vesicle protein required for the release of neurotransmitters
(Kobayashi et al., 2008; Nakashiba et al., 2008). Although
this system can be used in vivo, the transcriptional induction has
a relatively slow onset of 14 days, and recovery only occurs when
the cell resynthesizes VAMP2/synaptobrevin (Nakashiba et al.,
2008). A second approach uses a small molecule-induced
crosslinking of genetically modified forms of VAMP2/synaptobrevin
to rapidly and reversibly inhibit neurotransmission and is
described below.
MIST/CID
The dimerization of proteins is a general control mechanism in
biological systems, which contributes to the activation of cell
membrane receptors, transcription factors, vesicle fusion
proteins, and other classes of proteins. In the 1990s, Schreiber
and colleagues developed cell-permeant synthetic ligands to
reversibly control the intracellular dimerization of modified intracellular
proteins (Rosen and Schreiber, 1992; Schreiber, 1991).
MIST, an acronym for molecular systems for inactivation of
synaptic transmission, uses this approach to enforce nonproductive
interactions between vesicle proteins (Karpova et al.,
2005; Tervo and Karpova, 2007). By sequestering essential
components of the neurotransmitter release machinery in this
way, the release of neurotransmitter is inhibited and the communication
between neurons is silenced. The method involves
fusing the ligand-binding domain of the immuophilin FKBP to
the intracellular domain of the target protein. Introduction of
a dimeric immunophilin ligand (e.g., AP20187), a so-called
chemical inducer of dimerization (CID), then enforces the homodimerization
of that target protein. Heterodimerization can also
be accomplished using an additional immunophilin ligandbinding
domain. This approach was used to homodimerize the
synaptic vesicle protein VAMP2/synaptobrevin. Expression of
the VAMP2/synaptobrevin-FKBP binding domain fusion in
hippocampal neurons, combined with application of the CID
produced rapid, specific, and reversible inhibition of synaptic
transmission. Postsynaptic currents were inhibited on the minute
timescale and the level of block remained constant after several
hours in the presence of the dimerizer. Inactivation was
completely reversed after 1 hr in medium lacking the CID.
The MIST system was further tested in vivo using transgenic
mice expressing the VAMP/synaptobrevin-FKBP fusion in Purkinje
neurons. Introduction of the CID by direct injection into
the brain led to deficits in motor learning and balance that
reversed within 48 hr. A disadvantage of this approach is that
the postsynaptic targets of the manipulated presynaptic neurons
must already be known in order to validate silencing electrophys-
iologically.
Controlling Postsynaptic Receptors
Three approaches have been developed for controlling the
activity of postsynaptic receptors in genetically specified
neurons in vivo. One method uses the expression of peptide
neurotoxins tethered to the surface of the neuron to inhibit target
receptors (Auer et al., 2010; Ibanez-Tallon et al., 2004; Stu¨ rzebecher
et al., 2010). However, as this system is controlled at
the level of transcription, it is effectively irreversible. A second
system uses the expression of genetically modified receptors
Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved 611
Chemistry & Biology
Review
that are activated and inactivated through the light-induced
isomerization of a tethered agonist. The third approach uses
the expression of GABAA receptors that are uniquely sensitive
to the agonist zolpidem in targeted neurons. The latter two
approaches are discussed in further detail below.
LiGluR/MAG
LiGluR is a cation-selective glutamate-gated ion channel engineered
to respond to light through a covalently attached analog
of glutamate containing a photoisomerizable azobenzene moiety
(Gorostiza and Isacoff, 2008; Szobota et al., 2007; Volgraf et al.,
2006). An early version of an optically controlled ligand-gated ion
channel was reported in 1980 in the form of a nicotinic acetylcholine
receptor that could be photoactivated via a tethered choline
analog (Lester et al., 1980). In both cases, light controls the cis/
trans isomerization of the azobenzene tether and thus the
binding and unbinding of the agonist to the receptor. LiGluR is
generated by attaching the thiophilic photoswitchable agonist
MAG (cysteine-reactive maleimide, photoisomerizable azobenze,
glutamate-based agonist) to an engineered cysteine
residue (L439C) located near the ligand-binding domain of
GluK2 (a kainite-type glutamate-gated channel). The cis conformation
of the azobenzene, which positions the tethered agonist
within the ligand-binding domain, is favored at 380 nm, whereas
the trans conformation, which directs the agonist away from the
binding site, is favored at 500 nm. Azobenzenes also thermally
isomerize to the lower energy trans state in the dark (cis state
t1/2 = 17.65 min). This thermal bistability of the MAG allows sustained
activation following a brief pulse of light, which may be
advantageous in behavioral experiments. In cultured hippocampal
neurons expressing LiGluR and treated with MAG, brief
(1–5 ms) pulses of 374-nm light triggered reproducible patterns
of action potentials at frequencies up to 50 Hz that persisted
for minutes in the dark until extinguished by a short pulse of
488-nm light. The amplitude of the responses could also be
reduced to induce EPSP-like depolarizations that mimic synaptically
evoked EPSPs by attenuating illumination intensity. These
subthreshold depolarizations were reliably evoked at frequencies
as high as 100 Hz.
The LiGluR system has successfully been used in vivo to
control the behavior of zebrafish. GluK2(L439C) was introduced
into sensory neurons in zebrafish larvae, and the receptor was
labeled with MAG upon incubation of the larvae (125 mM MAG
in 5% DMSO) for 30 min. Illumination of the larvae with low-intensity
365-nm light for 15 min using a handheld UV lamp inhibited
the animal’s touch response, an escape response evoked by
mechanical pressure. The touch response was restored by illumination
with 488-nm light for 30 s. When more intense UV light
was used, the touch response was reversibly blocked by 10 s of
illumination. A drawback to this approach is that a reactive extracellular
cysteine is required to attach the photoisomerizable
ligand to the receptor. In most cases this cysteine will need to
be engineered into the receptor, and the appropriate location
for activating the receptor will have to be identified by trial and
error. The design of LiGluR was guided by extensive structureactivity
relationship analyses that had been performed on ionic
glutamate receptor agonists and an X-ray structure of the
ligand-binding domain of GluK2 in complex with the agonist
(2S,4R)-4-methyl glutamate. An unexplored drawback of this
method is the cross-reactivity of the ligand with cysteine residues
on off-target proteins, potentially altering the functioning
of those proteins.
GABAA(F77I)/Zolpidem
Another strategy for manipulating the propagation of impulses
within neuronal circuits is to control the activity of postsynaptic
GABAA receptors. These ligand-gated chloride channels are
present on virtually all neurons in the mammalian brain and
when activated prevent membrane depolarization. GABAA
receptors are pentameric structures most commonly comprised
of two a and two b subunits, of which there are several variants,
and a single g subunit (g2). The g2 subunit confers sensitivity to
zolpidem, an allosteric modulator that increases the function of
GABAA receptors. When Phe77 of the g2 subunit is mutated to
isoleucine, the receptors function normally with respect to activation
by GABA, but they are no longer modulated by zolpidem
(Buhr et al., 1997; Wingrove et al., 1997). Pharmacological
control over genetically defined populations of neurons could,
therefore, be obtained by first generating knockin mice in which
the wild-type (Phe77) g2 subunit is universally replaced by the
zolpidem-insensitive (Ile77) g2 subunit, and then selectively
putting the wild-type subunit back into targeted neuronal populations
(Cope et al., 2004). The power of this method was first
demonstrated by expressing the g2F771I subunit in the cerebellum
of transgenic mice (Wulff et al., 2007). In these mice, zolpidem,
which effectively crosses the blood–brain barrier, rapidly
induced significant motor deficits and revealed how interneurons
regulate cerebellum-dependent behavior.
The Ideal System and Future Directions
The next decade of brain research will likely see improved iterations
of existing systems, as well as the development of new
strategies, for controlling the activity of neurons. A number of
features are important to consider when choosing an existing
system, or designing a new one.
Genetic Encoding
Every region of the brain contains complex circuits made up of
many different kinds of neurons. Deciphering the connectivity
and dynamics of those circuits requires being able to target
each type of neuron individually. Methods that are genetic in
nature allow targeting of functionally circumscribed, rather than
anatomically defined, populations of neurons. Systems that
rely on the expression of only one transgene simplify the creation
of genetically modified cells, tissues, and organisms, and
obviate the need to balance the relative expression levels of
more than one gene. However, it is unlikely that unique
promoters for every neuron subtype will be identified. Therefore,
approaches requiring the expression of multiple proteins (e.g.,
the GluCl/IVM system), and that use multiple promoters with
partially overlapping specificities, may ultimately be more effective
at targeting specific types of neurons.
Orthogonality
The nervous system is incredibly plastic, and as such it is important
that addition of the exogenous protein itself does not perturb
the normal functioning of neurons. This may be an issue with the
LiGluR/MAG and GluCl/IVM systems, as these receptors can be
activated by endogenous glutamate. However, the efficacy of
glutamate at the GluCl receptor was reduced by greater than
6-fold by introducing a single point mutation in the glutamatebinding
site (Li et al., 2002). The use of the P2X2/ATP and
612 Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved
Chemistry & Biology
Review
TRPV1/capsaicin systems may be problematic in mammals as
these receptors are expressed on some neurons in the brain.
Implementablity
The reversible systems developed thus far rely on the addition of
light and/or a chemical to control the activity of targeted neurons.
Delivering these reagents to the brain of an intact animal can be
challenging. For example, introducing light deep into a mammalian
brain (e.g., mouse, rat) requires surgically implanting optical
waveguides into the animal. The delivery of light into other
commonly studied organisms is comparatively straightforward.
For example, activation of the P2X2/caged ATP and the LiGluR/
MAG systems in flies and transparent zebrafish, respectively, is
accomplished noninvasively by shining light of the appropriate
wavelength on the intact animal. Introducing chemicals into the
mammalian brain can also be challenging, owing to the restricted
permeability of the blood–brain barrier, and require invasive
procedures. It is worth noting the accessibility of chemicals to
widely distributed populations of neurons will be greater than
that of light, due to scattering in tissue (see below).
The wavelength of light is another factor to consider. The
shorter the wavelength of light is, the greater the scattering is
in tissue and the closer the light output will have to be to the targeted
neurons. In addition, illumination with short wavelength
(e.g., ultraviolet) light is damaging to cells over prolonged periods
of time. The ChR2/retinal and NpHR/retinal systems are both
activated by long-wavelength light (i.e., blue, yellow). The
LiGluR/MAG system uses UV light to stimulate activity and
blue light to withdraw the tethered agonist from the receptor.
However, once activated by a brief pulse of light, the channel
remains on for minutes in the dark, thus enabling depolarizations
and trains of action potentials to be evoked with minimal exposure
to UV light. In the P2X2/ATP and TRPV1/capsaicin systems,
the ligands can be activated using either single photon or twophoton
uncaging (Geissler et al., 2003; Zhao et al., 2006). Photosensitive
caging groups have recently been reviewed (EllisDavies,
2007; Mayer and Heckel, 2006).
Introducing chemicals into the brain of an organism can also
be challenging, owing to the restricted permeability of the
blood–brain barrier. The most important advantage of ChR2
and NpHR is that the photosensor retinal is endogenously
expressed in mammals. In the GluCl/IVM system, although
IVM poorly penetrates the blood–brain barrier, administering
IVM systemically (i.e., i.p. injection) results in an effective
concentration in the mammalian brain. In the GABAA/Zolpidem
and RASSL/CNO systems, both ligands effectively cross the
blood–brain barrier following systemic administration. In the
other systems, the ligand either does not cross the blood–brain
barrier (e.g., the octapeptide in the AlstR/AL system, caged
ATP in the P2X2/ATP system, the glutamate-azobenzene agonist
in the LiGluR/MAG system) or has not been tested (e.g., the
chemical dimerizer in the MIST/CID system).
Time Course and Reversibility
Deciphering the temporal coding of neural systems requires
a system that is capable of mimicking normal physiological
patterns of spike activity and synaptic events—that is, the
system must have fast temporal kinetics and fine temporal resolution.
The temporal resolution of light is generally superior to
pharmacological activation. The LiGluR/MAG and ChR2/retinal
systems are $10003 faster than the other systems and can be
used to drive physiological patterns of neuronal activity. If exquisite
temporal control over the response is not essential, pharmacological
stimulation is a powerful alternative.
In conclusion, tools that offer the capacity to control the function
of genetically delineated neurons are expected to provide
insight into the connectivity of neural circuits and the relationship
between the activity of those circuits and behavioral content.
Along with similar approaches being developed to provide
remote control over intracellular signaling cascades (Airan
et al., 2009; Conklin et al., 2008; Pei et al., 2008), these tools
may ultimately be used to restore information in circuits corrupted
by disease or injury.
ACKNOWLEDGMENTS
Work in the lab of P.M.E. is supported by the National Institutes of Health,
the McKnight Foundation, and the Sandler Foundation.
REFERENCES
Abbott, S.B.G., Stornetta, R.L., Socolovsky, C.S., West, G.H., and Guyenet,
P.G. (2009). Photostimulation of channelrhodopsin-2 expressing ventrolateral
medullary neurons increases sympathetic nerve activity and blood pressure in
rats. J. Physiol. 587, 5613–5631.
Airan, R.D., Thompson, K.R., Fenno, L.E., Bernstein, H., and Deisseroth, K.
(2009). Temporally precise in vivo control of intracellular signalling. Nature
458, 1025–1029.
Alexander, G.M., Rogan, S.C., Abbas, A.I., Armbruster, B.N., Pei, Y., Allen,
J.A., Nonneman, R.J., Hartmann, J., Moy, S.S., Nicolelis, M.A., et al. (2009).
Remote control of neuronal activity in transgenic mice expressing evolved g
protein-coupled receptors. Neuron 63, 27–39.
Armbruster, B.N., Li, X., Pausch, M.H., Herlitze, S., and Roth, B.L. (2007).
Evolving the lock to fit the key to create a family of G protein-coupled receptors
potently activated by an inert ligand. Proc. Natl. Acad. Sci. USA 104,
5163–5168.
Arrenberg, A.B., Del Bene, F., and Baier, H. (2009). Optical control of zebrafish
behavior with halorhodopsin. Proc. Natl. Acad. Sci. USA 106, 17968–17973.
Auer, S., Stu¨ rzebecher, A.S., Ju¨ ttner, R., Santos-Torres, J., Hanack, C.,
Frahm, S., Liehl, B., and Iban˜ ez-Tallon, I. (2010). Silencing neurotransmission
with membrane-tethered toxins. Nat. Methods 7, 229–236.
Bernard, A., Sorensen, S.A., and Lein, E.S. (2009). Shifting the paradigm: new
approaches for characterizing and classifying neurons. Curr. Opin. Neurobiol.
19, 530–536.
Bi, A., Cui, J.J., Ma, Y.P., Olshevskaya, E., Pu, M.L., Dizhoor, A.M., and Pan,
Z.H. (2006). Ectopic expression of a microbial-type rhodopsin restores visual
responses in mice with photoreceptor degeneration. Neuron 50, 23–33.
Birgul, N., Weise, C., Kreienkamp, H.J., and Richter, D. (1999). Reverse physiology
in Drosophila: identification of a novel allatostatin-like neuropeptide and
its cognate receptor structurally related to the mammalian somatostatin/galanin/opioid
receptor family. EMBO J. 18, 5892–5900.
Boyden, E.S., Zhang, F., Bamberg, E., Nagel, G., and Deisseroth, K. (2005).
Millisecond-timescale, genetically targeted optical control of neural activity.
Nat. Neurosci. 8, 1263–1268.
Buhr, A., Baur, R., and Sigel, E. (1997). Subtle changes in residue 77 of the
gamma subunit of alpha 1 beta 2 gamma 2 GABA(A) receptors drastically alter
the affinity for ligands of the benzodiazepine binding site. J. Biol. Chem. 272,
11799–11804.
Chow, B.Y., Han, X., Dobry, A.S., Qian, X.F., Chuong, A.S., Li, M.J., Henninger,
M.A., Belfort, G.M., Lin, Y.X., Monahan, P.E., and Boyden, E.S. (2010). Highperformance
genetically targetable optical neural silencing by light-driven
proton pumps. Nature 463, 98–102.
Claridge-Chang, A., Roorda, R.D., Vrontou, E., Sjulson, L., Li, H.Y., Hirsh, J.,
and Miesenbock, G. (2009). Writing memories with light-addressable reinforcement
circuitry. Cell 139, 405–415.
Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved 613
Chemistry & Biology
Review
Conklin, B.R., Hsiao, E.C., Claeysen, S., Dumuis, A., Srinivasan, S., Forsayeth,
J.R., Guettier, J.M., Chang, W.C., Pei, Y., McCarthy, K.D., et al. (2008). Engineering
GPCR signaling pathways with RASSLs. Nat. Methods 5, 673–678.
Cope, D.W., Wulff, P., Oberto, A., Aller, M.I., Capogna, M., Ferraguti, F., Halbsguth,
C., Hoeger, H., Jolin, H.E., Jones, A., et al. (2004). Abolition of zolpidem
sensitivity in mice with a point mutation in the GABA(A) receptor gamma 2
subunit. Neuropharmacology 47, 17–34.
Crawford, D.C., Moulder, K.L., Gereau, R.W., Story, G.M., and Mennerick, S.
(2009). Comparative effects of heterologous TRPV1 and TRPM8 expression in
rat hippocampal neurons. PLoS One 4, e8166.
Douglass, A.D., Kraves, S., Deisseroth, K., Schier, A.F., and Engert, F. (2008).
Escape behavior elicited by single, channelrhodopsin-2-evoked spikes in
zebrafish somatosensory neurons. Curr. Biol. 18, 1133–1137.
Ellis-Davies, G.C. (2007). Caged compounds: photorelease technology for
control of cellular chemistry and physiology. Nat. Methods 4, 619–628.
Geissler, D., Kresse, W., Wiesner, B., Bendig, J., Kettnemann, H., and Hagen,
V. (2003). DMACM-caged adenosine nucleotides: ultrafast phototriggers for
ATP, ADR and AMP activated by long-wavelength irradiation. ChemBioChem
4, 162–170.
Gorostiza, P., and Isacoff, E.Y. (2008). Optical switches for remote and noninvasive
control of cell signaling. Science 322, 395–399.
Gradinaru, V., Thompson, K.R., Zhang, F., Mogri, M., Kay, K., Schneider, M.B.,
and Deisseroth, K. (2007). Targeting and readout strategies for fast optical
neural control in vitro and in vivo. J. Neurosci. 27, 14231–14238.
Gunaydin, L.A., Yizhar, O., Berndt, A., Sohal, V.S., Deisseroth, K., and Hegemann,
P. (2010). Ultrafast optogenetic control. Nat. Neurosci. 13, 387–392.
Haber, S.N., and Knutson, B. (2010). The reward circuit: linking primate
anatomy and human imaging. Neuropsychopharmacology 35, 4–26.
Han, X., Qian, X.F., Bernstein, J.G., Zhou, H.H., Franzesi, G.T., Stern, P.,
Bronson, R.T., Graybiel, A.M., Desimone, R., and Boyden, E.S. (2009). Millisecond-timescale
optical control of neural dynamics in the nonhuman primate
brain. Neuron 62, 191–198.
Ibanez-Tallon, I., Wen, H., Miwa, J.M., Xing, J., Tekinay, A.B., Ono, F., Brehm,
P., and Heintz, N. (2004). Tethering naturally occurring peptide toxins for cellautonomous
modulation of ion channels and receptors in vivo. Neuron 43,
305–311.
Kalaidzidis, I.V., Kalaidzidis, Y.L., and Kaulen, A.D. (1998). Flash-induced
voltage changes in halorhodopsin from Natronobacterium pharaonis. FEBS
Lett. 427, 59–63.
Karpova, A.Y., Tervo, D.G.R., Gray, N.W., and Svoboda, K. (2005). Rapid and
reversible chemical inactivation of synaptic transmission in genetically targeted
neurons. Neuron 48, 727–735.
Kobayashi, T., Kai, N., Kobayashi, K., Fujiwara, T., Akagawa, K., and Onda, M.
(2008). Transient silencing of synaptic transmitter release from specific
neuronal types by recombinant tetanus toxin light chain fused to antibody variable
region. J. Neurosci. Methods 175, 125–132.
Lechner, H.A.E., Lein, E.S., and Callaway, E.M. (2002). A genetic method for
selective and quickly reversible silencing of mammalian neurons. J. Neurosci.
22, 5287–5290.
Lerchner, W., Xiao, C., Nashmi, R., Slimko, E.M., van Trigt, L., Lester, H.A., and
Anderson, D.J. (2007). Reversible silencing of neuronal excitability in behaving
mice by a genetically targeted, ivermectin-gated Cl- channel. Neuron 54,
35–49.
Lester, H.A., Krouse, M.E., Nass, M.M., Wassermann, N.H., and Erlanger, B.F.
(1980). Covalently bound photoisomerizable agonist: comparison with reversibly
bound agonists at electrophorus electroplaques. J. Gen. Physiol. 75,
207–232.
Li, P., Slimko, E.M., and Lester, H.A. (2002). Selective elimination of glutamate
activation and introduction of fluorescent proteins into a Caenorhabditis elegans
chloride channel. FEBS Lett. 528, 77–82.
Lima, S.Q., and Miesenbock, G. (2005). Remote control of behavior through
genetically targeted photostimulation of neurons. Cell 121, 141–152.
Luan, H., and White, B.H. (2007). Combinatorial methods for refined neuronal
gene targeting. Curr. Opin. Neurobiol. 17, 572–580.
Luo, L., Callaway, E.M., and Svoboda, K. (2008). Genetic dissection of neural
circuits. Neuron 57, 634–660.
Lynagh, T., and Lynch, J.W. (2010). An improved ivermectin-activated Cl
channel receptor for inhibiting electrical activity in defined neuronal populations.
J. Biol. Chem. 285, 14890–14897.
Maguire, E.A. (2001). Neuroimaging, memory and the human hippocampus.
Rev. Neurol. (Paris) 157, 791–794.
Mayer, G., and Heckel, A. (2006). Biologically active molecules with a ‘‘light
switch’’. Angew. Chem. Int. Ed. 45, 4900–4921.
Miesenbock, G., and Kevrekidis, I.G. (2005). Optical imaging and control of
genetically designated neurons in functioning circuits. Annu. Rev. Neurosci.
28, 533–563.
Nagel, G., Szellas, T., Huhn, W., Kateriya, S., Adeishvili, N., Berthold, P., Ollig,
D., Hegemann, P., and Bamberg, E. (2003). Channelrhodopsin-2, a directly
light-gated cation-selective membrane channel. Proc. Natl. Acad. Sci. USA
100, 13940–13945.
Nagel, G., Szellas, T., Kateriya, S., Adeishvili, N., Hegemann, P., and Bamberg,
E. (2005). Channelrhodopsins: directly light-gated cation channels. Biochem.
Soc. Trans. 33, 863–866.
Nakashiba, T., Young, J.Z., McHugh, T.J., Buhl, D.L., and Tonegawa, S.
(2008). Transgenic inhibition of synaptic transmission reveals role of CA3
output in hippocampal learning. Science 319, 1260–1264.
Pei, Y., Rogan, S.C., Yan, F., and Roth, B.L. (2008). Engineered GPCRs as
tools to modulate signal transduction. Physiology (Bethesda) 23, 313–321.
Petreanu, L., Huber, D., Sobczyk, A., and Svoboda, K. (2007). Channelrhodopsin-2-assisted
circuit mapping of long-range callosal projections. Nat. Neurosci.
10, 663–668.
Rosen, M.K., and Schreiber, S.L. (1992). Natural products as probes of cellular
function: studies of immunophilins. Angew. Chem. Int. Ed. Engl. 31, 384–400.
Schobert, B., and Lanyi, J.K. (1982). Halorhodopsin is a light-driven chloride
pump. J. Biol. Chem. 257, 306–313.
Schreiber, S.L. (1991). Chemistry and biology of the immunophilins and their
immunosuppressive ligands. Science 251, 283–287.
Schroll, C., Riemensperger, T., Bucher, D., Ehmer, J., Voller, T., Erbguth, K.,
Gerber, B., Hendel, T., Nagel, G., Buchner, E., and Fiala, A. (2006). Lightinduced
activation of distinct modulatory neurons triggers appetitive or aversive
learning in Drosophila larvae. Curr. Biol. 16, 1741–1747.
Sineshchekov, O.A., Jung, K.H., and Spudich, J.L. (2002). Two rhodopsins
mediate phototaxis to low- and high-intensity light in Chlamydomonas reinhardtii.
Proc. Natl. Acad. Sci. USA 99, 8689–8694.
Slimko, E.M., and Lester, H.A. (2003). Codon optimization of Caenorhabditis
elegans GluCl ion channel genes for mammalian cells dramatically improves
expression levels. J. Neurosci. Methods 124, 75–81.
Slimko, E.M., McKinney, S., Anderson, D.J., Davidson, N., and Lester, H.A.
(2002). Selective electrical silencing of mammalian neurons in vitro by the
use of invertebrate ligand-gated chloride channels. J. Neurosci. 22,
7373–7379.
Stu¨ rzebecher, A.S., Hu, J., Smith, E.S., Frahm, S., Santos-Torres, J., Kampfrath,
B., Auer, S., Lewin, G.R., and Ibanez-Tallon, I. (2010). An in vivo tethered
toxin approach for the cell-autonomous inactivation of voltage-gated sodium
channel currents in nociceptors. J. Physiol. 588, 1695–1707.
Sudhof, T.C., and Malenka, R.C. (2008). Understanding synapses: past,
present, and future. Neuron 60, 469–476.
Suzuki, T., Yamasaki, K., Fujita, S., Oda, K., Iseki, M., Yoshida, K., Watanabe,
M., Daiyasu, H., Toh, H., Asamizu, E., et al. (2003). Archaeal-type rhodopsins in
Chlamydomonas: model structure and intracellular localization. Biochem.
Biophys. Res. Commun. 301, 711–717.
Szobota, S., Gorostiza, P., Del Bene, F., Wyart, C., Fortin, D.L., Kolstad, K.D.,
Tulyathan, O., Volgraf, M., Numano, R., Aaron, H.L., et al. (2007). Remote
614 Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved
Chemistry & Biology
Review
control of neuronal activity with a light-gated glutamate receptor. Neuron 54,
535–545.
Tan, E.M., Yamaguchi, Y., Horwitz, G.D., Gosgnach, S., Lein, E.S., Goulding,
M., Albright, T.D., and Callaway, E.M. (2006). Reversible inactivation of
mammalian neurons in vivo using the Drosophila allatostatin receptor. Neuron
51, 157–170.
Tervo, D., and Karpova, A.Y. (2007). Rapidly inducible, genetically targeted
inactivation of neural and synaptic activity in vivo. Curr. Opin. Neurobiol. 17,
581–586.
Tobin, D., Madsen, D.M., Kahn-Kirby, A., Peckol, E.L., Moulder, G., Barstead,
R., Maricq, A.V., and Bargmann, C.I. (2002). Combinatorial expression of TRPV
channel proteins defines their sensory functions and subcellular localization in
C. elegans neurons. Neuron 35, 307–318.
Volgraf, M., Gorostiza, P., Numano, R., Kramer, R.H., Isacoff, E.Y., and Trauner,
D. (2006). Allosteric control of an ionotropic glutamate receptor with an
optical switch. Nat. Chem. Biol. 2, 47–52.
Wang, H., Peca, J., Matsuzaki, M., Matsuzaki, K., Noguchi, J., Qiu, L., Wang,
D., Zhangn, F., Boyden, E., Deisserothl, K., et al. (2007). High-speed mapping
of synaptic connectivity using photostimulation in Channelrhodopsin-2 transgenic
mice. Proc. Natl. Acad. Sci. USA 104, 8143–8148.
Wehr, M., Hostick, U., Kyweriga, M., Tan, A., Weible, A.P., Wu, H., Wu, W.,
Callaway, E.M., and Kentros, C. (2009). Transgenic silencing of neurons in
the mammalian brain by expression of the allatostatin receptor (AlstR). J. Neurophysiol.
102, 2554–2562.
Wingrove, P.B., Thompson, S.A., Wafford, K.A., and Whiting, P.J. (1997). Key
amino acids in the gamma subunit of the gamma-aminobutyric acid(A)
receptor that determine ligand binding and modulation at the benzodiazepine
site. Mol. Pharmacol. 52, 874–881.
Wulff, P., Goetz, T., Leppa, E., Linden, A.M., Renzi, M., Swinny, J.D., Vekovischeva,
O.Y., Sieghart, W., Somogyi, P., Korpi, E.R., et al. (2007). From
synapse to behavior: rapid modulation of defined neuronal types with engineered
GABAA receptors. Nat. Neurosci. 10, 923–929.
Zemelman, B.V., Lee, G.A., Ng, M., and Miesenbock, G. (2002). Selective photostimulation
of genetically ChARGed neurons. Neuron 33, 15–22.
Zemelman, B.V., Nesnas, N., Lee, G.A., and Miesenbock, G. (2003). Photochemical
gating of heterologous ion channels: remote control over genetically
designated populations of neurons. Proc. Natl. Acad. Sci. USA 100,
1352–1357.
Zhang, F., Wang, L.P., Brauner, M., Liewald, J.F., Kay, K., Watzke, N., Wood,
P.G., Bamberg, E., Nagel, G., Gottschalk, A., and Deisseroth, K. (2007). Multimodal
fast optical interrogation of neural circuitry. Nature 446, 633–639.
Zhang, F., Prigge, M., Beyriere, F., Tsunoda, S.P., Mattis, J., Yizhar, O.,
Hegemann, P., and Deisseroth, K. (2008). Red-shifted optogenetic excitation:
a tool for fast neural control derived from Volvox carteri. Nat. Neurosci. 11,
631–633.
Zhang, F., Gradinaru, V., Adamantidis, A.R., Durand, R., Airan, R.D., de Lecea,
L., and Deisseroth, K. (2010). Optogenetic interrogation of neural circuits: technology
for probing mammalian brain structures. Nat. Protoc. 5, 439–456.
Zhao, J., Gover, T.D., Muralidharan, S., Auston, D.A., Weinreich, D., and Kao,
J.P.Y. (2006). Caged vanilloid ligands for activation of TRPV1 receptors by
1-and 2-photon excitation. Biochemistry 45, 4915–4926.
Zhao, S., Cunha, C., Zhang, F., Liu, Q., Gloss, B., Deisseroth, K., Augustine,
G.J., and Feng, G.P. (2008). Improved expression of halorhodopsin for lightinduced
silencing of neuronal activity. Brain Cell Biol. 36, 141–154.
Chemistry & Biology 17, June 25, 2010 ª2010 Elsevier Ltd All rights reserved 615
Chemistry & Biology
Review