fenno annrev 2011

NE34CH17-Deisseroth ARI 21 May 2011 15:22

The Development andApplication of OptogeneticsLief Fenno,1,2 Ofer Yizhar,1 and Karl Deisseroth1,3,41Department of Bioengineering, 2Neuroscience Program, 3Departments of Psychiatry andBehavioral Sciences, 4Howard Hughes Medical Institute, Stanford University, Stanford,California 94305; email: [email protected]

Annu. Rev. Neurosci. 2011. 34:389412

The Annual Review of Neuroscience is online atneuro.annualreviews.org

This articles doi:10.1146/annurev-neuro-061010-113817

Copyright c 2011 by Annual Reviews.All rights reserved

0147-006X/11/0721-0389$20.00

Keywordschannelrhodopsin, halorhodopsin, bacteriorhodopsin,electrophysiology

AbstractGenetically encoded, single-component optogenetic tools have madea significant impact on neuroscience, enabling specific modulation ofselected cells within complex neural tissues. As the optogenetic toolboxcontents grow and diversify, the opportunities for neuroscience con-tinue to grow. In this review, we outline the development of currentlyavailable single-component optogenetic tools and summarize the appli-cation of various optogenetic tools in diverse model organisms.

389

Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


ContentsINTRODUCTION . . . . . . . . . . . . . . . . . . 390EARLY EFFORTS TOWARD

OPTICAL CONTROL . . . . . . . . . . . 390MICROBIAL OPSINS . . . . . . . . . . . . . . . 391OPTOGENETIC TOOLS FOR

NEURONAL EXCITATION . . . . . 393OPTOGENETIC TOOLS FOR

NEURONAL INHIBITION . . . . . . 396OPTOGENETIC TOOLS FOR

BIOCHEMICAL CONTROL . . . . . 397DELIVERING OPTOGENETIC

TOOLS INTO NEURONALSYSTEMS . . . . . . . . . . . . . . . . . . . . . . . . 398

TRANSGENIC ANIMALS. . . . . . . . . . . 399DEVELOPMENTAL AND

LAYER-SPECIFICTARGETING . . . . . . . . . . . . . . . . . . . . 399

CIRCUIT TARGETING . . . . . . . . . . . . 401LIGHT DELIVERY AND

READOUT HARDWARE FOROPTOGENETICS. . . . . . . . . . . . . . . . 401

OPTOGENETICS IN DIVERSEANIMAL MODELS. . . . . . . . . . . . . . . 402Caenorhabditis elegans . . . . . . . . . . . . . . . 402Fly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402Zebrafish . . . . . . . . . . . . . . . . . . . . . . . . . . 403Mouse. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 404Rat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405Primate . . . . . . . . . . . . . . . . . . . . . . . . . . . 405

OUTLOOK. . . . . . . . . . . . . . . . . . . . . . . . . . 405

INTRODUCTIONIn describing unrealized prerequisites for as-sembling a general theory of the mind, FrancisCrick observed that the ability to manipulateindividual components of the brain wouldbe needed, requiring a method by which allneurons of just one type could be inactivated,leaving the others more or less unaltered(Crick 1979, p. 222). Extracellular electricalmanipulation does not readily achieve true in-activation, and even electrical excitation, whileallowing for temporal precision in stimulatingwithin a given volume, lacks specificity for cell

type. However, pharmacological and geneticmanipulations can be specific to cells withcertain expression profiles (in the best case)but lack temporal precision on the timescale ofneural coding and signaling.

Because no prior technique has achievedboth high-temporal and cellular precisionwithin intact mammalian neural tissue, therehas been strong pressure to develop a new classof technology. As a result of these efforts, neu-rons now may be controlled with optogeneticsfor fast, specific excitation or inhibition withinsystems as complex as freely moving mammals[for example, with microbial opsin methods,light-induced inward cation currents may beused to depolarize the neuronal membrane andpositively modulate firing of action potentials,while optical pumping of chloride ions caninduce outward currents and membranehyperpolarization, thereby inhibiting spiking(Figure 1)]. These optogenetic tools of micro-bial origin (Figure 1) may be readily targetedto subpopulations of neurons within hetero-geneous tissue and function on a temporalscale commensurate with physiological ratesof spiking or critical moments in behavioraltests, with fast deactivation upon cessation oflight. With these properties, microbe-derivedoptogenetic tools fulfill the criterion set forthby Crick in 1979 (Deisseroth 2010, 2011).

EARLY EFFORTS TOWARDOPTICAL CONTROLThe microbial opsin approach is heir to along tradition of using light as an interventionin biology. With chromophore-assisted laserinactivation, light can be used to inhibittargeted proteins by destroying them [whata geneticist would call loss of function(Schmucker et al. 1994)]; conversely, lasers canbe used to stimulate neurons directly in a waythat could be adapted (in principle) to controlfluorescently labeled, genetically targeted cells[what a geneticist would call gain of function(Fork 1971, Hirase et al. 2002)]. Next, variouscascades of genes, and combinations of geneswith chemicals, were tested as multicomponent

390 Fenno Yizhar Deisseroth

Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


Na+

Na+ Na+

Na+

Na+

Na+

K+

H+

H+

H+H+

H+

H+

K+K+

K+

K+

K+

Na+Na+

Na+ Na+ Ca2+

Ca2+Ca2+

Na+

Na+

ChR

Cl

Cl

Cl

Cl

Cl

Cl

Cl

Cl

Cl

ClCl

Cl

Cl

Cl

Cl

Cl

HR

[IP3][DAG]

[cAMP] [cAMP]

GsGq Gi

OptoXR

Figure 1Optogenetic tool families. Channelrhodopsins conduct cations and depolarize neurons upon illumination (left). Halorhodopsinsconduct chloride ions into the cytoplasm upon yellow light illumination (center). OptoXRs are rhodopsin-GPCR (G proteincoupledreceptor) chimeras that respond to green (500 nm) light with activation of the biological functions dictated by the intracellular loopsused in the hybrid (right).

strategies for optical control; rhodopsin andarrestin genes from Drosophila photoreceptorswere combined to light-sensitize neurons(Zemelman et al. 2002); ligand-gated channels,combined with ultraviolet (UV)-light pho-tolysis of caged agonists, were developed forDrosophila experiments (Lima & Miesenbock2005, Zemelman et al. 2003); and UV lightisomerizable chemicals linked to geneticallyencoded channels were employed in culturedcells and in zebrafish (Banghart et al. 2004,Szobota et al. 2007, Volgraf et al. 2006). Theseefforts have been reviewed (Gorostiza & Isacoff2008, Miesenbock & Kevrekidis 2005) andwhile elegant, have thus far been found to belimited to various extents in speed, targeting,tissue penetration, and/or applicability becauseof their multicomponent nature. Here, wereview development and application effortsfocused on the distinct single-component opto-genetic tools, such as microbial opsins, over thepast six years since they were first implemented.

MICROBIAL OPSINSSpecies from multiple branches of the an-imal kingdom have evolved mechanismsto sense electromagnetic radiation in their

environments. Likewise many microbes, in theabsence of complex eye structures employedby metazoans, have developed light-activatedproteins for a variety of purposes. For some,this serves as a mechanism of homeostasis to re-main at a certain depth in the ocean (Beja et al.2000, 2001); for others, this helps maintainosmotic balance in a highly saline environment(Stoeckenius 1985). These and other diverseroles are, in many cases, fulfilled by a familyof seven-transmembrane, light-responsiveproteins encoded by opsin genes.

Opsin genes are divided into two distinctsuperfamilies: microbial opsins (type I) and an-imal opsins (type II). Opsin proteins from bothfamilies require retinal, a vitamin Arelatedorganic cofactor that serves as the antenna forphotons; when retinal is bound, the functionalopsin proteins are termed rhodopsins. Retinalcovalently attaches to a conserved lysineresidue of helix 7 by forming a protonatedretinal Schiff base (RSBH+). The ionic envi-ronment of the RSB, defined by the residuesof the binding pocket, dictates the spectraland kinetic characteristics of each individualprotein. Upon absorption of a photon, retinalisomerizes and triggers a sequence of confor-mational changes within the opsin partner.

www.annualreviews.org The Development and Application of Optogenetics 391

Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


The photoisomerized retinal is the triggerfor subsequent structural rearrangements andactivities performed by these proteins.

Although both opsin families encodeseven-transmembrane structures, sequence ho-mology between the two families is extremelylow; homology within families, however, ishigh (25%80% residue similarity) (Man et al.2003). Whereas type I opsin genes are found inprokaryotes, algae, and fungi (Spudich 2006),type II opsin genes are present only in highereukaryotes and are responsible mainly forvision (but also play roles in circadian rhythmand pigment regulation) (Sakmar 2002,Shichida & Yamashita 2003). Type II opsingenes encode G proteincoupled receptors(GPCRs) and, in the dark, bind retinal in the11-cis configuration. Upon illumination, retinalisomerizes to the all-trans configuration andinitiates the reactions that underlie the visualphototransduction second messenger cascade.After photoisomerization, the retinal-proteinlinkage is hydrolyzed; free all-trans retinal thendiffuses out of the protein and is replaced by afresh 11-cis retinal molecule for another roundof signaling (Hofmann et al. 2009).

In contrast, type I opsins more typicallyencode proteins that utilize retinal in theall-trans configuration, which photoisomerizesupon photon absorption to the 13-cis con-figuration. Unlike the situation with type IIrhodopsins, the activated retinal molecule intype I rhodopsins does not dissociate from itsopsin protein but thermally reverts to the all-trans state while maintaining a covalent bondto its protein partner (Haupts et al. 1997).Type I opsins encode several distinct subfami-lies of protein, discussed in more detail below.The central operating principle of these ele-gant molecular machines [established for thisbroad family of opsins since bacteriorhodopsin(BR) in 1971 (Oesterhelt & Stoeckenius 1971)and now including halorhodopsins and chan-nelrhodopsins (Figure 1)] is their unitary na-ture. They combine the two tasks of lightsensation and ion flux into a single protein(with bound small organic cofactor), encodedby a single gene. In 2005, one of these micro-

bial opsins was brought to neuroscience as thefirst single-component optogenetic tool (Boy-den et al. 2005), and the other microbial opsinsubfamilies followed close behind. For example,channelrhodopsin-1 (ChR1) (Nagel et al. 2002)and channelrhodopsin-2 (ChR2) (Nagel et al.2003) from Chlamydomonas reinhardtii are blue-light-activated nonspecific cation channels. Incommon with other type I opsins, these proteinsrequire retinal as the photon-sensing cofactorto function. In response to light stimulation, thechannel shuttles from the dark-adapted statethrough a stereotyped progression of functionaland conformational states, eventually (in the ab-sence of further light stimulation) reaching thedark-adapted state once again. These differentstates, which all have unique spectroscopic sig-natures, are collectively referred to as the pho-tocycle, which (as for BR and halorhodopsinin earlier work) has been extensively inves-tigated (Bamann et al. 2010, Berndt et al.2010, Ernst et al. 2008, Hegemann et al. 2005,Ritter et al. 2008, Stehfest et al. 2010, Stehfest& Hegemann 2010).

The size, kinetic properties, and wavelengthsensitivity of photocurrents resulting from acti-vation of an individual protein are a direct resultof its photocycle topology, ion selectivity, andactivation/deactivation/inactivation time con-stants (Ernst et al. 2008, Hegemann et al. 2005,Ritter et al. 2008). Typically, a transient peakphotocurrent, evoked at the onset of light stim-ulation, decays modestly to a steady-state pho-tocurrent even in the presence of continuouslight, owing in part to the desensitization ofa certain population of channels (Nagel et al.2003). The desensitized population can recoverin the dark with a characteristic time constanton the order of 5 seconds, giving rise to a simi-lar peak photocurrent if a second light pulse isapplied after sufficient time has elapsed (Nagelet al. 2003). The fraction of desensitized pro-teins is crucial for determining the reliability oflight stimulation during prolonged experiments(e.g., behavioral or long-term physiological ex-periments) where light is applied for extendedperiods. This issue is addressed in more detailbelow.


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


OPTOGENETIC TOOLS FORNEURONAL EXCITATIONMany years passed after the discovery of BRs,channelrhodopsins, and halorhodopsins, priorto the development of optogenetics. As notedabove, optogenetics with microbial opsins be-gan with a channelrhodopsin, introduced intohippocampal neurons in 2005 (Boyden et al.2005) where it was found to confer millisecond-precision control of neuronal spiking. A num-ber of additional reports followed over the nextyear (Bi et al. 2006, Ishizuka et al. 2006, Liet al. 2005, Nagel et al. 2005). Moreover, thisturned out to be a single-component system;through a remarkable twist of nature, investi-gators found that sufficient retinal is presentin mammalian brains (and as later established,in all vertebrate tissues tested thus far) to en-able functional expression of these optogenetictools as single components, in the absence ofany added chemical or other gene (Deisserothet al. 2006, Zhang et al. 2006). The optoge-netic toolbox has been vastly expanded sincethe original 2005 discovery to include dozensof single-component proteins activated by var-ious wavelengths of light, with various ion con-ductance regulation properties that operate inneurons over a wide range of speeds (from mil-liseconds to tens of minutes), enabling broadexperimental configurations and opportunities.

Initial improvements in ChR2 were carriedout in an incremental fashion and focusedon improving expression and photocurrentin mammalian systems. Human codon opti-mization for improved expression of ChR2(Boyden et al. 2005, Gradinaru et al. 2007) wascombined with substitution of histidine forarginine at position 134 to increase steady-statecurrent size, although the mechanism of thelatter effect (delayed channel closure) alsosignificantly impaired temporal precision andhigh-speed spiking (ChR2-H134R; Nagel et al.2005, Gradinaru et al. 2007). Better solutionswere needed, and further diversification of theoptogenetic toolbox via mutational engineer-ing has proven to be challenging but highlyproductive (Figure 2).

First, because many microbial tools do notexpress well in mammalian neurons, generalstrategies for enabling mammalian expressionwere required. Membrane-trafficking modi-fications turned out to be crucial, beginningwith the observation in 2008 that endo-plasmic reticulum (ER)-export motifs werehelpful for achieving high, safe expression ofhalorhodopsins (Gradinaru et al. 2008, Zhaoet al. 2008), a principle that turned out to beextendable [to other classes of membrane-trafficking motif (Gradinaru et al. 2010)] andgeneralizable [to most microbial opsins tested(Gradinaru et al. 2010)]. Next, chimera strate-gies (using hybrids of ChR1 and ChR2) werefound to be helpful in giving rise to improvedexpression and spiking properties for channel-rhodopsins (Lin et al. 2009, Wang et al. 2009).

Finally, much subsequent opsin engineer-ing for optogenetics was carried out on the ba-sis of hypothesized ChR structures. To date,there has been no reported crystal structureof any excitatory optogenetic tools, althoughstructures exist for halorhodopsin and for theproton pump BR (Luecke et al. 1999a,b), whichcan function to inhibit neurons when heterol-ogously expressed (Gradinaru et al. 2010); pro-ton conductance is a property shared by ChR2,whether through channel or possible pumpingmechanisms (Feldbauer et al. 2009). Capital-izing upon homology between BR and ChRs,we have introduced a number of mutations thatmodify various properties of the opsins (dis-cussed below) and have described a frameworkby which these improvements can be applied toother novel opsins [indeed, this direction hascome full circle, with insights from ChRs nowused to improve the optogenetic function of BRitself (Gradinaru et al. 2010)].

These structural hypothesis-based opsinengineering efforts were spurred by inherentlimitations of the channelrhodopsin system(Figure 2). Specifically, the deactivation timeconstant of ChR2 upon cessation of light(1012 ms in neurons) imposed a limit ontemporal precision, leading in some cases toartifactual multiplets of spikes after even singlebrief light pulses (as well as a plateau potential


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


Blue (off /peak activation wavelength)

ChR2

ChIEF

0 10 155 20 25off (ms)

~10 ms/470 nm

ChR2 (H134R) 18 ms/470 nm

~10 ms/450 nm

Blue ChETA (off /peak activation wavelength)

T159C/E123T

0 2 4 6 8off (ms)

E123A 4 ms/470 nm

E123T 4.4 ms/500 nm8 ms/500 nm

Red-shifted (off /peak activation wavelength)

C1V1

0 50 100 150off (ms)

120 ms/540 nm

VChR1 133 ms/545 nm

Red-shifted ChETA (off /peak activation wavelength)

0 10 20 30 40 50off (ms)

38 ms/545 nmC1V1 (E162T/E122T)C1V1 (E162T) 50 ms/535 nm

Bistable (off /peak activation wavelength/peak inactivation wavelength)

0 3 6 9 12 15 18 21 24 27 30off (min)

C128S 1.7 min/470 nm/560 nmChR2 0.000167 min

D156A 6.9 min/470 nm/590 nm

C128S/D156A 29 min/445 nm/590 nm

Biochemical signaling (off /peak activation wavelength)

1AR

0 1 2 203 off (s)

3 s/500 nm

2AR 0.5 s/500 nm

b-PAC 20 s/441 nmRh-CT 3 s/485 nm

Inhibitory (off /peak activation wavelength)

ArcheBR

0 2015105off (ms)

18.9 m/560 nm

eNpHR3.0 4.2 ms/590 nm

9 ms/566 nm


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


when pulsing blue light at frequencies above 40hz; the next pulse of light would occur before allthe ChRs had deactivated, leading eventuallyto failed deinactivation of native voltage-gatedsodium channels and thus missed spikes andfurther loss of fidelity). The latter problem wascompounded by the desensitization of ChRitself even in the presence of light, leading tofurther missed spikes late in trains. Addressingpart of this challenge, the chimeric opsins notedabove (e.g., ChIEF; Lin et al. 2009), demon-strated reduced desensitization in culturedneurons, allowing more robust spiking overtrains in culture as well as stronger currents. Inanother approach addressing both desensitiza-tion and deactivation, considering the crystalstructure of BR led to modification of the coun-terion residue E123 of ChR2 to threonine oralanine; the resulting faster opsin is referred toas ChETA (Figure 2) (Gunaydin et al. 2010).This substitution introduced two advantagesover wild-type ChR2. First, it reduced desen-sitization during light exposure, with the resultthat light pulses late in high-frequency trainsbecame as likely as early light pulses to drivespikes (a very important property referred to astemporal stationarity). Second, it destabilizedthe active conformation of retinal, speedingspontaneous isomerization to the inactive stateafter light-off and thus closing the channelmuch more quickly after cessation of light thanwild-type or improved ChR2 variants. Theresulting functional consequences of ChETAmutations are temporal stationarity, reducedextra spikes, reduced plateau potentials, andimproved high-frequency spike followingat 200 Hz or more over sustained trains,even within intact mammalian brain tissue(Gunaydin et al. 2010).

Whereas many experimental designs em-ploy optogenetic tools to initiate precise spik-ing, alternative paradigms may instead requirethe researcher to simply alter the excitabilityof a target neuronal population. Indeed, it isoften important to bias the activity of a par-ticular neuronal population without specificallydriving action potentials or synchronous ac-tivity. To facilitate experiments examining al-tered excitability, Berndt et al. (2009) devel-oped the step-function opsins (SFOs). SFOs area family of ChR mutants that display bistablebehaviororders-of-magnitude prolonged ac-tivity after termination of the light stimulusfirst instantiated as mutations in position 128in ChR2 (cysteine to serine, alanine, or threo-nine). Again based on homology between ChR2and BR (Peralvarez-Marin et al. 2004), we mu-tated this residue to manipulate the interac-tion between the opsin backbone and the cova-lently bound retinal photon sensor. In contrastto ChETA, the SFO mutations are designedto stabilize the active retinal isomer, the func-tional consequence of which is prolonging theactive state of the channel even after light-off.SFOs have inactivation time constants on theorder of tens of seconds or more instead of mil-liseconds (Figure 2) and can be activated by asingle 10-ms pulse of blue light (Berndt et al.2009). The SFOs can also be deactivated by apulse of yellow light; the yellow pulse drivesisomerization of retinal back to the noncon-ducting state. A subsequently engineered SFO,the ChR2(D156A) opsin (Bamann et al. 2010),displays an even longer inactivation time con-stant, which in our hands approaches eight min-utes. One potential noted use of opsins withextended time constants could be for scanningtwo-photon stimulation paradigms (Rickgauer

Figure 2Spectral and kinetic diversification of optogenetic tools. Deactivation time constants (off ) and approximate peak activation/inactivationwavelengths are shown for blue lightactivated opsins, blue ChETAs (including E123T/T159C) (Berndt et al. 2011, Mattis et al. 2011),red-shifted opsins, red-shifted ChETAs, bistable (SFO) opsins, and modulatory/inhibitory tools in a compact look-up table. ChR2 islisted among the bistable group for scale purposes only. Where precise values are not available, decay kinetics were measured (courtesyof J. Mattis, personal communication) or estimated from published traces; all values were recorded at room temperature (except foroptoXRs measured at 37C), with substantial acceleration in kinetics (50%) expected for all at 37C.


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


& Tank 2009), during which it would be help-ful to have persistent accumulating activity as asmall two-photon spot scans a cell or tissue ofinterest.

We have now engineered a third classof SFO by combining the D156 and C128mutations to produce a ChR2 variant that hasa spontaneous deactivation time constant ap-proaching 30 min; this stabilized SFO (SSFO)(O. Yizhar, L. Fenno, M. Prigge, K. Stehfest,J. Paz, F. Schneider, S. Tsunoda, R. Fudim, C.Ramakrishnan, J. Huguenard, P. Hegemann& K. Deisseroth, submitted) induces peakcurrents of >200 pA. An advantage of havingan opsin with such a long time constant is theability to conduct behavioral protocols in theabsence of tethered external light delivery de-vices (e.g., optical fibers). Because a single pulseof blue light is sufficient to induce activity for atime period extending beyond that of most be-havioral paradigms, the fiber may be removedbefore commencing the experiment. Just aswith the original SFO proteins, SSFO may beinactivated by yellow light, allowing for precisecontrol of network dynamics. A final advantageof the SFOs (which scales with their kinetic sta-bility) is orders-of-magnitude greater light sen-sitivity of transduced cells, particularly for longlight pulses, a direct result of the photon inte-gration bestowed by their prolonged deactiva-tion time constant. This property renders SFOsespecially attractive as minimally invasive toolsfor stimulating large brain regions (for example,in primate studies) and deep within tissue.

Separate from, but synergistic with, molec-ular engineering is the systematic genomicidentification of novel opsins. Adapting novelopsins activated by red-shifted wavelengthscould enable control of two separate popu-lations of circuit elements within the samephysical volume. To this end, we launched ge-nomic search strategies that led to identificationof an opsin from Volvox carteri (VChR1) (Zhanget al. 2008), which shares homology with ChR2and similarly functions as a cation channel.In contrast to ChR2, VChR1 is activated byred-shifted light. However, the relatively smallcurrents due to low expression in mammalian

neurons, as described in the initial report,have hampered in vivo adaptation of VChR1,even after codon optimization. To this end, wehave engineered a chimeric opsin, C1V1 (O.Yizhar, L. Fenno, M. Prigge, K. Stehfest, J.Paz, F. Schneider, S. Tsunoda, R. Fudim, C.Ramakrishnan, J. Huguenard, P. Hegemann& K. Deisseroth, submitted), composed ofthe first two and one-half helices of ChR1(a poorly expressing relative of ChR2 fromthe same organism) (Nagel et al. 2002) andthe last four and one-half helices of VChR1.The resulting tool retains the red-shiftedactivation spectrum of VChR1, but withnanoampere-scale currents that exceed thoseof ChR2. The large current of C1V1 allowsuse in vivo and also use of off-peak (redder)light wavelengths, together enabling trulyseparable control of multiple populations ofneurons when used in conjunction with ChR2(O. Yizhar, L. Fenno, M. Prigge, K. Stehfest,J. Paz, F. Schneider, S. Tsunoda, R. Fudim, C.Ramakrishnan, J. Huguenard, P. Hegemann& K. Deisseroth, submitted). In addition tocombinatorial experiments with ChR2, toolswith red-shifted activation wavelengths suchas C1V1 are also more amenable to use withsimultaneous imaging of genetically encodedcalcium indicators, such as GCaMP variants(Hires et al. 2008, Zhang & Oertner 2007).

OPTOGENETIC TOOLS FORNEURONAL INHIBITIONComplementing these tools for precise controlof neural excitation, certain light-activated ionpumps may be used for inhibition, althoughthus far only one ion pump has shown efficacyat modulating behavior in mammals (Tye et al.2011, Witten et al. 2010, Zhang et al. 2007a):the ER trafficking-enhanced version of ahalorhodopsin called NpHR (Gradinaru et al.2008, 2010; Zhang et al. 2007a) derived fromthe halobacterium Natronomonas pharaonis. Inthe context of optogenetic application, this yel-low lightactivated electrogenic chloride pumpacts to hyperpolarize the targeted neuron uponactivation (Figure 1) (Zhang et al. 2007a).


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


Unlike the excitatory channelrhodopsins,NpHR is a true pump and requires constantlight to move through its photocycle. Althoughoptogenetic inhibition with NpHR was shownto operate well in freely moving worms (Zhanget al. 2007a), mammalian brain slices (Zhanget al. 2007a), and cultured neurons (Han &Boyden 2007, Zhang et al. 2007a), several yearspassed before final validation of this (or any) in-hibitory optogenetic tool was obtained by suc-cessful application to intact mammals (Tye et al.2011, Witten et al. 2010) because of membranetrafficking problems that required additionalengineering (Gradinaru et al. 2008, 2010; Zhaoet al. 2008). Indeed, a number of modificationsto NpHR were required to improve its func-tion, initially codon-optimizing the sequencefollowed by enhancement of its subcellu-lar trafficking (eNpHR2.0 and eNpHR3.0)(Gradinaru et al. 2008, 2010), which resultedin improved membrane targeting and highercurrents suitable even for use in human tissue(Busskamp et al. 2010), as well as activation withred-shifted wavelengths at the infrared border(680 nm). Kouyama et al. (2010) publishedthe 2.0A crystal structure of halorhodopsinand illustrate that this protein has a highdegree of homology within the retinal bindingpocket with the proton pump BR. The two areproposed to distribute charge and store energyfrom absorbed photons in similar ways.

The diversification of the inhibitory opsintoolbox has been guided by bioinformatics ap-proaches to screen nature for novel inhibitoryion pumps with desirable properties, just as hadbeen successful previously for excitatory opsins(Zhang et al. 2008). In 2010, we and othersexplored the use of proton pumps (eBR, Mac,and Arch) as optogenetic tools (Gradinaruet al. 2010, Chow et al. 2010), finding evidencefor robust efficacy but leaving open questionsof long-term tolerability and functionality ofproton-motive pumps in mammalian neurons.A major concern is the extent to which pump-ing of protons into the extracellular space(especially in juxtamembranous compartmentsdifficult to visualize or measure) could havedeleterious or noncell-type-specific effects on

local tissue, which could show up as light-induced inhibition affecting all recorded units(more units than expected for a given transduc-tion efficiency) or with a slightly slower timecourse than expected for the fast proton pumps.

Although the proton pumps must be treatedwith caution until these issues are addressed,many opportunities exist; indeed, we haveimproved the ability of proton pumps to hy-perpolarize neurons following a methodologysimilar to that used for improving NpHR(Gradinaru et al. 2010), although none of theproton pumps yet described is as kinetically sta-ble or as potent as eNpHR3.0, especially at thesafe light levels (


Rhodopsin, the light-sensing protein in themammalian eye, is both an opsin, in that it iscovalently bound to retinal and its function ismodulated by the absorption of photons, and aGPCR, in that it is coupled on the intracellularside of the membrane to a G-protein, trans-ducin. Virtually all neurons can communicatevia GPCRs, which of course respond notonly to neuromodulators from dopaminergic,serotonergic, and adrenergic pathways, but alsoto fast neurotransmitters such as glutamateand GABA. Building both on our finding thatadequate retinal is present within mammalianbrain tissue (Deisseroth et al. 2006, Zhanget al. 2006) and on a long history of elegantGPCR structure-function work from Khorana,Kobilka, Caron, Lefkowitz, and others, wehave determined that GPCRs can be convertedinto light-activated regulators of well-definedbiochemical signaling pathways that functionwithin freely moving mammals. These pro-teins are referred to as optoXRs (Airan et al.2009). OptoXRs allow for receptor-mediatedintracellular signaling with temporal resolutionsuitable for modulation of behavior in freelymoving mice (Airan et al. 2009). OptoXRs aremodulated by 500-nm light and now include thealpha-1 and beta-2 adrenergic receptors (Airanet al. 2009), which are coupled to Gq and Gs sig-naling pathways, respectively, and the 5-HT1areceptor, which is Gi/o coupled (Oh et al. 2010).

Optical control over small GTPases wasnext achieved in cultured cells by several differ-ent laboratories (Levskaya et al. 2009, Wu et al.2009, Yazawa et al. 2009) using optically mod-ulated protein-protein interactions; althoughthese have not yet been shown to express or dis-play single-component functionality in freelymoving mammals, such capability is plausiblewhere flavin or biliverdin chromophores arerequired and present. Finally, investigatorshave recently described microbial adenylylcyclases with lower dark activity than earliermicrobial cyclases, and because they employ aflavin chromophore, these tools appear suitablefor single-component optogenetic control(Ryu et al. 2010, Stierl et al. 2010). Together,these experiments have extended optogenetic

capability to essentially every cell type, even innonexcitable tissues, in biology.

DELIVERING OPTOGENETICTOOLS INTO NEURONALSYSTEMSViral expression systems have the dual advan-tages of fast/versatile implementation and highinfectivity/copy number for robust expressionlevels. Cellular specificity can be obtained withviruses by specific promoters (if small, specific,and strong enough), by spatial targeting of virusinjection, and by restriction of opsin activationto particular cells (or projections of specificcells) via targeted light delivery (Zhang et al.2010, Diester et al. 2011). Lenti and adeno-associated (AAV) viral vectors have been usedsuccessfully to introduce opsins into the mouse,rat, and primate brain (Zhang et al. 2010). Ad-ditionally, these have been well tolerated andhighly expressed over long periods of time withno reported adverse effects. Lentivirus is easilyproduced using standard tissue culture tech-niques and an ultracentrifuge (see Zhang et al.2010 for protocol). AAV may be produced ei-ther by individual laboratories or through coreviral facilities. Neither AAV nor lentivirus werefound to be highly expressed in the zebrafish,for which Sindbis and rabies are more effective(Zhu et al. 2009). Viruses have been used totarget (among other cells) hypocretin neurons(Adamantidis et al. 2007), excitatory pyramidalneurons (Lee et al. 2010, Sohal et al. 2009,Zhang et al. 2007a), and astroglia (Gourine et al.2010, Gradinaru et al. 2009). For example, onegroup (Gourine et al. 2010) recently describedthe use of AAV-delivered ChR2 to controlastroglial activity in the brain stem of mice andto dissect a mechanism by which astroglia cantransfer systemic information from the bloodto neurons underlying homeostasis, in this casedirectly modulating neurons that manipulatethe rate of respiration. However, a major down-side of viral expression systems is a maximumgenetic payload length; only promoter frag-ments that are small (less than 4 kb), specific,and strong may be used, and these are rare. This


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


limitation may be skirted using Cre-driveranimals and Cre-dependent viruses, discussedbelow.

TRANSGENIC ANIMALSThe use of transgenic or knock-in animals ob-viates viral payload limitations and allows fortighter control of transgene expression usinglarger promoter fragments or indeed the en-dogenous genome in full via knock-in. Thefirst transgenic opsin-expressing mouse line wasgenerated using the Thy1 promoter (Arenkielet al. 2007, Zhao et al. 2008), with widespreadexpression throughout neocortical layer 5 pro-jection neurons as well as in some subcorticalstructures (Arenkiel et al. 2007). This mouseline has been widely used, for example, to exam-ine the roles of inhibitory neurons on corticalinformation processing (Sohal et al. 2009) andthe mechanism of action of deep brain stimu-lation for Parkinsons disease (Gradinaru et al.2009). Several other groups have subsequentlyalso generated transgenic mouse lines directlyexpressing opsin genes (Hagglund et al. 2010,Katzel et al. 2010, Thyagarajan et al. 2010).

Caveats to using transgenic mouse linesto directly express optogenetic tools includethe time, effort, and cost associated with theirproduction, validation, and maintenance. Toenable widespread use of the latest optoge-netic tools, investigators have designed opsin-delivering viruses for which opsin expression isdependent on the coexpression of Cre recom-binase (Figure 3). This doublefloxed invertedopen-reading-frame (DIO; reviewed in Zhanget al. 2010) strategy (Atasoy et al. 2008, So-hal et al. 2009, Tsai et al. 2009) situates theopsin gene (inhibitory or excitatory) in the in-verted (meaningless) orientation, but the geneis flanked by two sets of incompatible Cre re-combinase recognition sequences (Sohal et al.2009, Tsai et al. 2009). The recombinase recog-nition sequences are placed such that in thepresence of Cre, the ORF is inverted insteadof being excised. Reversing the sequence thenallows one of the Cre recognition sites to be ex-cised (Figure 3), locking the reading frame into

the correct direction and allowing for strongexpression of the opsin with (for example) theelongation factor 1-alpha (EF1) promoter.The specificity of this gene expression can thencome (for example) from the targeted expres-sion of Cre in driver rodent lines in whichCre is controlled with high specificity in thecontext of very large chromosomal promoter-enhancer regions; the DIO strategy thus en-ables versatile and widespread use of optoge-netics with the many (and growing numberof ) experimental systems selectively expressingCre recombinase (Geschwind 2004; Gong et al.2003, 2007; Heintz 2004). This strategy hasbeen used recently in many systems, for exam-ple to target dopamine-1 (D1) or dopamine-2(D2) receptorexpressing neurons of the stria-tum via transgenic D1-Cre and D2-Cre mouselines, to examine the effects of their stimulationin the classic direct/indirect-movement path-ways (Kravitz et al. 2010). With these sameCre lines, Lobo and colleagues (2010) exam-ined the roles of nucleus accumbens D1 andD2 neurons in modulating cocaine reward. TheCre-dependent optogenetic system allowed forthe first time a direct examination of the re-lationship between neuronal activity of specificneuronal populations and animal behavior, thuspaving the way for a deeper understanding ofdiseases such as Parkinsons disease, depression,and substance abuse.

DEVELOPMENTAL ANDLAYER-SPECIFIC TARGETINGThe ability to target specific neocortical layershas been a long-sought goal of neuroscience;this can now be achieved either with layer-specific Cre driver lines or with developmentaltargeting strategies such as in utero electropo-ration (IUE). As a result, multiple laboratorieshave now successfully teased apart the role oflayer-specific neurons in behavioral paradigmsand network dynamics. Optogenetic toolshave been well tolerated when electroporatedin utero into mouse embryos (Adesnik &Scanziani 2010, Gradinaru et al. 2007, Lewiset al. 2009, Petreanu et al. 2007); IUE may


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


+ Cre Cre

+ Cre Cre

Perc

enta

ge

of m

axim

um (%

)Pe

rcen

tage

of

max

imum

(%)

YFP intensity+Cre Cre

b c

eYFP

Parvalbumin

Transgenicparvalbumin::Cre

Overlay

30 m

a

Cre+

Cre+

Cre+

Cre+

Cre+

Cre

Cre

Cre

Viralinjection

eYFP

loxP loxP

ChR2eYFPEF1

ChR2 eYFPEF1

ChR2eYFPEF1

ChR2 eYFPEF1

ChR2 eYFPEF1

ChR2 eYFPEF1

lox2722

lox2722loxP

lox2722

ReversibleCre-mediated

recombination

PermanentCre-mediated

lox site excision

d Floxed stop

loxP loxP

ITR

hGHpolyA

hGHpolyA

ITR3X polyA WPRE

Double-floxed inverse ORFeITR

100 um

100

80

60

40

20

01010 1020 1030

100

80

60

40

20

01010 1020 1030

YFP intensity

ITRWPRE

loxP loxP

lox2722 lox2722

Figure 3Low-leak Cre-dependent expression using the doublefloxed inverted open-reading-frame (DIO) strategy. The combination of atransgenic mouse expressing Cre recombinase in specific neuronal subtypes and the injection of a virally encoded DIO opsin (a) resultsin the physical inversion of the open reading frame (ORF) in only that population (b,c), which may be transient and revert back to theoriginal state or undergo further recombination to be permanently anchored in the sense direction, resulting in functional expression ofthe opsin (c). The DIO strategy may be contrasted with the lox-stop-lox (floxed STOP) strategy (d,e). In the absence of Crerecombinase, lox-stop-lox (d ) allows for some level of expression leak as assayed by both enhanced yellow fluorescent protein (eYFP)expression and fluorescence-activated cell sorting (FACS) analysis. Because the ORF of an opsin in DIO configuration encodesnonsense (e), there is no functional expression in the absence of Cre recombinase. Adapted with permission from Sohal et al. (2009) andF. Zhang and K. Deisseroth.


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


be used to target specific layers of cortex byincorporating the DNA (with no promoter sizelimit) into neurons being born during a specificembryonic stage. A major advantage of usingIUE or transgenic mice over viral infection isthat opsins are expressed at the time of birth,allowing electrophysiological researchers toharvest acute slices at a younger stage. Counter-acting this advantage is the fact that transgenicanimals typically express lower levels of opsins,likely owing to the reduced gene copy number.

CIRCUIT TARGETINGAnother generalizeable strategy for targeting isreferred to as projection targeting, which cap-italizes on the efficient membrane traffickingof engineered opsin gene products, especiallydown axons to axon terminals. Light can bedelivered not to somata but to axons, therebyrecruiting cells defined by virtue of theirwiring without any genetic information aboutthe downstream target required (Gradinaruet al. 2007, 2009; Lee et al. 2010; Petreanu et al.2007). Neurons may also be targeted by projec-tion using viruses that transduce axon terminals,such as herpes simplex virus (HSV) familyviruses, certain serotypes of AAV, or pseudo-typed lentiviruses. Trans-synaptic targetingmay be achieved by exploiting the transcellulartrafficking of, for example, the wheat germ ag-glutinin (WGA) peptide sequence (Gradinaruet al. 2010), which can deliver Cre recombinaseto the site of a second Cre-dependent virusinjection. Combination strategies are also pos-sible; by crossing transgenic Drd2-GFP micewith mice expressing Cre under the controlof Emx1, then injecting DIO-ChR2-mCherryinto cortex, Higley & Sabatini (2010) were ableto localize synapses originating from cortex(red projections) onto neurons expressing thedopamine-2 receptor (D2) in striatum (greencell bodies) and use brief pulses of blue lightto elicit synaptic activity onto the D2 neurons.Combining optogenetic manipulation of thesynapse with pharmacology and two-photonglutamate uncaging allowed the investigatorsto elaborate precisely upon the role of D2 re-ceptors in glutamatergic synaptic transmission.

A noteworthy method pioneered by theCallaway group (Wickersham et al. 2007a,b)using a glycoprotein-deficient pseudotypedrabies virus is yet another technique formonosynaptic circuit tracing. Rabies virusis well known to travel trans-synapticallyfrom neuron to neuron; the virus used inthis technique is not able to produce viablepackaged copies of itself after moving trans-synaptically and thus will be stopped after onesynapse jump. Applying this method enablestracing of all neurons synaptically connectedto a single neuron of interest. Two plasmids,one containing the glycoprotein-deficientrabies payload and another containing theglycoprotein, are coelectroporated into a singleneuron in vivo. This neuron is then able toproduce competent rabies virus; the payload,however, does not encode the coat protein. Inthis case, the virus is stuck after moving onesynapse. By using red and green fluorophoresin the two components, the targeted neuronand its synaptic partners may be identified.This system has not yet been integrated withoptogenetics, however, and the extremely highlevels of expression resulting from rabies virusexpression may result in toxicity incompatiblewith typical optogenetic experiments.

LIGHT DELIVERY ANDREADOUT HARDWARE FOROPTOGENETICSOptogenetics fundamentally relies on light-delivery technology, the development of whichhas led to improved precision of modulationboth in vitro and in vivo. In vitro, optogenetictools are typically activated with filtered lightfrom mercury arc lamps (e.g., Berndt et al.2009, Boyden et al. 2005, Gunaydin et al.2010), lasers (Cardin et al. 2010, Cruikshanket al. 2010, Kravitz et al. 2010, Petreanu et al.2009), light-emitting diodes (LEDs) (e.g.,Adesnik & Scanziani 2010; Grubb & Burrone2010a,b; Wang et al. 2009), or LED arraysfor multisite stimulation (Grossman et al.2010). In vivo, stimulation of behaving animalshas been conducted mostly with laser lightdelivered to the transduced tissue via optical


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


fibers inserted through chronically implantedcannulas (Adamantidis et al. 2007, Aravaniset al. 2007, Zhang et al. 2010) or with fiber-coupled high-power LEDs (Wang et al. 2010).The chronic delivery of light using implantedinfrared-triggered LEDs (Iwai et al. 2011) is inthe early developmental stages but promises toopen a new direction in optogenetic research.

To achieve rich readouts from optogeneti-cally controlled tissue, major effort has been di-rected toward generating electrophysiologicalsystems that combine high-density single-unitrecordings with optogenetic stimulation inmice and other organisms. The first readoutsfrom in vivo optogenetic modulation wereobtained in anesthetized animals using a devicecomposed of a fiberoptic cable integrated witha tungsten electrode (Gradinaru et al. 2007),called an optrode (Cardin et al. 2009; Gradi-naru et al. 2007, 2009; Sohal et al. 2009; Zhanget al. 2009). More advanced strategies haveemerged recently, employing silicone multisiteelectrodes (Royer et al. 2010) and movabletetrode arrays combined with optical fibers formore flexible interrogation of neural activity invivo (Lima et al. 2009). Two-photon imagingis another avenue with which optogenetics maybe integrated to stimulate and record neuralactivity simultaneously. Several studies havemade progress toward this type of experiment(Andrasfalvy et al. 2010, Mohanty et al. 2008,Papagiakoumou et al. 2010, Rickgauer & Tank2009, Zhu et al. 2009); it seems that the majorlimitation that hampers optogenetic activationwith two-photon approaches is the combi-nation of rapidly decaying opsin-mediatedphotocurrents in the setting of typical slow2P raster-scanning techniques. Modifyingthe raster scan paradigm (Rickgauer & Tank2009) or modulating the laser light such thatfast activation is possible across wider regions(i.e., an entire cell soma; Papagiakoumouet al. 2010) can address this problem (Shoham2010). Appropriately rich readouts, when com-bined with optogenetic inputs, will powerfullyfacilitate fundamental studies regarding theorganization and function of intact, complexneural networks.

OPTOGENETICS IN DIVERSEANIMAL MODELS

Caenorhabditis elegans

In transgenic nematodes harboring the chan-nelrhodopsin gene, it is possible to controlmuscle wall motor neuron and mechanosen-sory neuron activity (Nagel et al. 2005). Zhangand colleagues (2007a) controlled body wallmuscle contraction bidirectionally with ChR2and NpHR, demonstrating the power of com-binatorial optogenetics. This concept has beenbuilt on with the description of three-colorLCD-based multimodal light delivery (Stirmanet al. 2011) and digital micromirror device(DMD)/laser light delivery (Leifer et al. 2011),each coupled with tracking software for use withthe behaving specimen. The facility of quanti-fying body-wall contraction and elongation inC. elegans has enabled large-scale investigationof various mutant strains for synaptic proteindefects (Liewald et al. 2008, Stirman et al. 2010)and nicotinic acetylcholine receptor function(Almedom et al. 2009). Finally, C. elegans wasalso used for combined light-based stimulationand readout of neural activity (Guo et al. 2009,Tian et al. 2009), fulfilling the promise ofall-optical physiological experiments usingoptogenetic tools and genetically encodedactivity sensors (Scanziani & Hausser 2009).

FlyFly lines expressing upstream activation se-quence (UAS):ChR2 (Zhang et al. 2007b) havebeen used to investigate the neuronal basis ofthe nociceptive response (Hwang et al. 2007)and appetitive/aversive odorant learning at thereceptor (Bellmann et al. 2010) or neurotrans-mitter (Schroll et al. 2006) level and to rescuephotosensory mutants (Xiang et al. 2010). Ad-ditionally, Hortstein et al. (2009) demonstratedGal4/UAS targeting of ChR2 to the larval neu-romuscular junction system. Creative uses ofoptogenetic tools in Drosophila include validat-ing neurons identified in a screen to probethe proboscis extension reflex (Gordon & Scott


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


2009), driving monoamine release to validatefast-scanning cyclic voltammetry detection ofserotonin and dopamine (Borue et al. 2009), andinvestigating the innate escape response (Zim-mermann et al. 2009). Special considerationsare required for this model organism (Pulveret al. 2009). Unlike mammals, flies and wormsdo not possess levels of endogenous retinal suf-ficient for the function of optogenetic tools,but food supplement can provide sufficientretinal to drive ChR2 function (Xiang et al.2010).

Flies also possess innate behavioral re-sponses to blue light that are developmentallydependent (Bellmann et al. 2010, Pulveret al. 2009, Suh et al. 2007, Xiang et al.2010), complicating behavioral studies usingopsins with blue activation spectra. Thisconfound may be partially rectified using flylines without vision, such as those lacking thenorpA gene (Bellmann et al. 2010), althoughnorpA-deficient lines remain sensitive to bluelight (Xiang et al. 2010). These issues could,in principle, be circumvented with red-shiftedoptogenetic tools for excitation and inhibition.Complementing eNpHR3.0 for red-shiftedinhibition, we have developed an optogenetictoolset for potent red-shifted excitation (O.Yizhar, L. Fenno, M. Prigge, K. Stehfest, J.Paz, F. Schneider, S. Tsunoda, R. Fudim, C.Ramakrishnan, J. Huguenard, P. Hegemann &K. Deisseroth, submitted). These have yet tobe tested in Drosophila, but their green peak ac-tivation wavelength is outside the range of keyDrosophila photosensory proteins (Xiang et al.2010); moreover, spiking may be driven withup to 630 nm light, improving the potential fordeep-penetrating excitation (Pulver et al. 2009).

ZebrafishThe short generational time and easy inte-gration of foreign DNA into zebrafish arecomplemented by ease of optogenetic manip-ulation owing to transparency of the organism(McLean & Fetcho 2011, White et al. 2008).The first use of optogenetic tools in zebrafish(Douglass et al. 2008) appeared in a study

examining the role of somatosensory controlof escape behavior. The use of ChR2 to drivesingle spikes in a genetically defined populationduring the course of movement took advantageof a number of properties of optogenetic toolsnot available with traditional pharmacologicalor electrophysiological methods. Neuronswere stimulated with a simple setup combin-ing a dissecting scope and epifluorescencesource, with light restricted by the micro-scope aperture. Recent advances (Arrenberget al. 2009) reported zebrafish lines witheNpHR- enhanced yellow fluorescent protein(eYFP) and ChR2-eYFP expression controlledby the Gal4/UAS system to allow the tools to beeasily targeted to specific neuronal subtypes viagenetic crosses with zebrafish expressing Gal4in various cell populations. Of note, Arrenberget al. (2009) compared various iterations ofNpHR and fluorophore and concluded thateNpHR2.0-eYFP had the most reliable andefficient expression. Photoconverting proteinsKaede and Dendra were used to approximatethe upper bound of light spread with lownumerical aperture, small-diameter (50 um)fibers in place of a microscope aperture, ob-serving that the combination of small-fiber andneuronal targeting allowed for the stimulationof an approximately 30-um-diameter spot.

Other groups have reported using Gal4/UAS systems driving optogenetic tools toexamine cardiac function and development(Arrenberg et al. 2010), transduction of sensoryneuron mechanoreception (Low et al. 2010),command of swim behavior (Arrenberg et al.2009, Douglass et al. 2008), and saccade genera-tion (Schoonheim et al. 2010). As an alternativeto producing stable transgenic lines, Zhu et al.(2009) undertook a systematic examination ofviral infection in zebrafish and found success-ful ChR2 delivery by the Sindbis and rabiesviruses. Of note, they also modulated ChR2 ex-pression using a Tet-inducible expression strat-egy. A specific technical consideration of im-plementing optogenetic tools in studies usingzebrafish is the stimulation of neurons thatexpress endogenous light-actived proteins;Arrenberg et al. (2009) found that 26% of


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


control neurons in zebrafish had a firing ratemodulated by yellow light. This percentage wasreduced to 14% in congenitally blind zebrafishlines. The remaining response was postulatedto be due to either expression of other opticallyactivated proteins or a thermal response, but itwas not investigated further.

MouseBy far, the most widely published optogeneticmodel organism to date has been the mouse.Mice represent the majority of transgenic an-imals, including a vast selection of transgeniclines expressing Cre recombinase in specificsubpopulations of neurons (Gong et al. 2007).Mouse embryonic stem cells have also beenamenable to expression and interrogation withoptogenetic technologies (Stroh et al. 2010).The first report to use channelrhodopsin inbehaving mammals examined the contributionof hypothalamic hypocretin (orexin) neuronsto sleep and wakefulness (Adamantidis et al.2007).

Optogenetic modulation in mouse has alsoyielded control of monoaminergic systems. Re-cently, in a study that used ChR2, eNpHR2.0,and TH::Cre transgenic mice to modulatethe locus coeruleus neurons bidirectionally,these noradrenergic neurons strongly modu-lated sleep and arousal states (Carter et al.2010). Using the same TH::Cre transgenicmice, causal relationships were identified be-tween activity patterns in VTA dopamine neu-rons and reward behavior in mice, showingthat phasic dopamine release is more effec-tive than tonic release in driving reward be-havior (Tsai et al. 2009). Using optogeneticstimulation of axonal terminals in the nucleusaccumbens, investigators recently discoveredthat dopamine neurons corelease glutamate(Stuber et al. 2010, Tecuapetla et al. 2010).DAT-Cre mice have been used in conjunc-tion with Cre-dependent ChR2 to examinemechanisms underlying dopamine-modulatedaddiction (Brown et al. 2010). And ChAT-Cretransgenic mice were used in combinationwith Cre-dependent ChR2 and NpHR viruses

to show that cholinergic interneurons ofthe nucleus accumbens are key regulators ofmedium spiny neuron activity and can mod-ulate cocaine-based place preference (Wittenet al. 2010). The connectivity of striatal mediumspiny neurons themselves has been describedusing a tetracycline-based ChR2 transgenic sys-tem (Chuhma et al. 2011).

Direct optogenetic modulation of principaland local-circuit inhibitory neurons in mousecortex and hippocampus has also enabledcontributions to understanding the complexityof mammalian neural circuit dynamics. Reportson the functions of parvalbumin-expressingfast-spiking interneurons demonstrated di-rectly their involvement in gamma oscillationsand information processing in mouse pre-frontal (Sohal et al. 2009) and somatosensory(Cardin et al. 2009, 2010) cortex. Focal stim-ulation of pyramidal neurons in Thy1::ChR2mice has enabled rapid, functional mappingof motor control across the motor cortex(Ayling et al. 2009, Hira et al. 2009), andaxonal stimulation in regions contralateral toinjected cortical areas has enabled the mappingof projection patterns in callosal corticalprojections (Petreanu et al. 2007). Within localcortical microcircuits, ChR2 has been used tocharacterize the spatial receptive fields of var-ious neuron types (Katzel et al. 2010, Petreanuet al. 2009, Wang et al. 2007) and to study thebasic properties of cortical disynaptic inhibi-tion (Hull et al. 2009). Optogenetics is alsobeing used to discern the possible therapeuticmechanism of cortical intervention in mousemodels of depression (Covington et al. 2010)and to develop novel strategies for control ofperipheral nerves (Llewellyn et al. 2010).

Mice have also been used to study amyg-dala circuits involved in fear and anxiety.Johansen and colleagues (2010) used ChR2 todemonstrate the sufficiency of lateral amygdalapyramidal neurons in auditory cued fearconditioning. In two recent reports, functionalcircuits within the central amygdala werefurther delineated, demonstrating that distinctsubpopulations of inhibitory central amygdalaneurons separately gate the acquisition and


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


expression of conditioned fear (Ciocchi et al.2010, Haubensak et al. 2010). Finally, Tyeand colleagues (2011) described the differ-ential effects of activating lateral amygdalaprojections onto central amygdala neurons inregulating anxiety behaviors. These studiesshed new light on fear and anxiety behaviorsand demonstrate the utility of optogenetictechniques in dissecting complex local neuronalcircuits.

RatRats are important for neuroscience researchbecause of their ability to perform complexbehavioral tasks, the relative simplicity of theirbrains (compared with human and nonhumanprimates), and the ability to perform high-density recordings of neural ensembles duringfree behavior. Recently, virally delivered opto-genetic tools were used in rats to examine bloodoxygen leveldependent (BOLD) responsesin functional magnetic resonance imaging(fMRI) (Lee et al. 2010). Driving ChR2 inexcitatory neuronal populations was sufficientto elicit a BOLD response not only in localcortical targets (where both the virus and lightdelivery optical fiber were targeted) but also indownstream thalamic regions, allowing globalmaps of activity causally driven by defined cellpopulations to be obtained within intact livingmammals. Optogenetic work in rats has beenlimited by the availability of viral promotersthat are capable of driving specific expressionin the absence of transgenic targeting, but theadvent of transgenic rat lines expressing Crerecombinase in specific neuronal subtypes (inaddition to projection targeting) will greatly ex-pand the potential for using rat models of neuralcircuit function in health and disease. Outsidethe central nervous system, optogenetic ma-nipulation in rodents is providing insights intodiverse physiological functions. ChR2 was usedto modulate rhythmic beating activity in rodentcardiomyocytes, demonstrating the potentialfor future applications in this field (Bruegmannet al. 2010), and several groups have used opto-

genetics to modulate cardiovascular function,breathing, and blood pressure (Abbott et al.2009a,b; Alilain et al. 2008; Kanbar et al. 2010)in both anesthetized and awake rats.

PrimateOptogenetic modulation of primate neurons(Han et al. 2009, Diester et al. 2011) has beenexplored by ChR2 delivery to cortical neuronsof macaques via lentiviral transduction, but be-havioral responses have not yet been observed.eNpHR2.0 has been delivered to human neu-ral tissue in the form of ex vivo human retinasand has shown optogenetic efficacy on physi-ological measures (Busskamp et al. 2010) withpossible relevance to retinitis pigmentosa (RP),a disease in which light-sensing cells degener-ate in the retina. By expressing eNpHR2.0 inlight-insensitive cone cells, normal phototrans-duction was restored, as well as center/surroundcomputational features, directional sensitivity,and light-guided behavior. Additionally, Weicket al. (2010) demonstrated the functionality ofChR2 in human embryonic stem cellderivedneurons.

OUTLOOKThe optogenetic toolbox has broadly expandedto include proteins that are powerful and di-verse in their ionic selectivity, spectral sen-sitivity, and temporal resolution. Combinedwith powerful molecular techniques for trans-genic and viral expression in rodents, zebrafish,and flies, the current generation of optogenetictools may be adapted to an extensive landscapeof questions within neuroscience. The currentgeneration of optogenetic tools has been op-timized for stronger expression, higher cur-rents, and spectral shifts to allow combinatorialcontrol within the same volume of space. On-going improvements to the toolbox will yieldmolecular tools targeted to subcellular com-partments [such as dendrites or axons (Lewiset al. 2009)], tools for two-photon activation,and tools that further expand the optical control


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


of biochemistry. At this moment in time, single-component optogenetics has become a staple in

neuroscience laboratories, even as many oppor-tunities remain yet untapped.

DISCLOSURE STATEMENTThe authors are not aware of any affiliations, memberships, funding, or financial holdings thatmight be perceived as affecting the objectivity of this review.

LITERATURE CITED

Abbott SB, Stornetta RL, Fortuna MG, Depuy SD, West GH, et al. 2009a. Photostimulation of retrotrape-zoid nucleus phox2b-expressing neurons in vivo produces long-lasting activation of breathing in rats.J. Neurosci. 29:580619

Abbott SB, Stornetta RL, Socolovsky CS, West GH, Guyenet PG. 2009b. Photostimulation ofchannelrhodopsin-2 expressing ventrolateral medullary neurons increases sympathetic nerve activity andblood pressure in rats. J. Physiol. 587:561331

Adamantidis AR, Zhang F, Aravanis AM, Deisseroth K, de Lecea L. 2007. Neural substrates of awakeningprobed with optogenetic control of hypocretin neurons. Nature 450:42024

Adesnik H, Scanziani M. 2010. Lateral competition for cortical space by layer-specific horizontal circuits.Nature 464:115560

Airan RD, Thompson KR, Fenno LE, Bernstein H, Deisseroth K. 2009. Temporally precise in vivo controlof intracellular signalling. Nature 458:102529

Alilain WJ, Li X, Horn KP, Dhingra R, Dick TE, et al. 2008. Light-induced rescue of breathing after spinalcord injury. J. Neurosci. 28:1186270

Almedom RB, Liewald JF, Hernando G, Schultheis C, Rayes D, et al. 2009. An ER-resident membraneprotein complex regulates nicotinic acetylcholine receptor subunit composition at the synapse. EMBO J.28:263649

Andrasfalvy BK, Zemelman BV, Tang J, Vaziri A. 2010. Two-photon single-cell optogenetic control ofneuronal activity by sculpted light. Proc. Natl. Acad. Sci. USA 107:1198186

Aravanis AM, Wang LP, Zhang F, Meltzer LA, Mogri MZ, et al. 2007. An optical neural interface: in vivocontrol of rodent motor cortex with integrated fiberoptic and optogenetic technology. J. Neural Eng.4:S14356

Arenkiel BR, Peca J, Davison IG, Feliciano C, Deisseroth K, et al. 2007. In vivo light-induced activation ofneural circuitry in transgenic mice expressing channelrhodopsin-2. Neuron 54:20518

Arrenberg AB, Del Bene F, Baier H. 2009. Optical control of zebrafish behavior with halorhodopsin. Proc.Natl. Acad. Sci. USA 106:1796873

Arrenberg AB, Stainier DY, Baier H, Huisken J. 2010. Optogenetic control of cardiac function. Science330:97174

Atasoy D, Aponte Y, Su HH, Sternson SM. 2008. A FLEX switch targets channelrhodopsin-2 to multiple celltypes for imaging and long-range circuit mapping. J. Neurosci. 28:702530

Ayling OG, Harrison TC, Boyd JD, Goroshkov A, Murphy TH. 2009. Automated light-based mapping ofmotor cortex by photoactivation of channelrhodopsin-2 transgenic mice. Nat. Methods 6:21924

Bamann C, Gueta R, Kleinlogel S, Nagel G, Bamberg E. 2010. Structural guidance of the photocycle ofchannelrhodopsin-2 by an interhelical hydrogen bond. Biochemistry 49:26778

Banghart M, Borges K, Isacoff E, Trauner D, Kramer RH. 2004. Light-activated ion channels for remotecontrol of neuronal firing. Nat. Neurosci. 7:138186

Beja O, Aravind L, Koonin EV, Suzuki MT, Hadd A, et al. 2000. Bacterial rhodopsin: evidence for a new typeof phototrophy in the sea. Science 289:19026

Beja O, Spudich EN, Spudich JL, Leclerc M, DeLong EF. 2001. Proteorhodopsin phototrophy in the ocean.Nature 411:78689


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


Bellmann D, Richardt A, Freyberger R, Nuwal N, Schwarzel M, et al. 2010. Optogenetically induced olfactorystimulation in Drosophila larvae reveals the neuronal basis of odor-aversion behavior. Front. Behav. Neurosci.4:27

Berndt A, Prigge M, Gradmann D, Hegemann P. 2010. Two open states with progressive proton selectivitiesin the branched channelrhodopsin-2 photocycle. Biophys. J. 98:75361

Berndt A, Schoenenberger P, Mattis J, Tye KM, Deisseroth K, et al. 2011. High-efficiency channelrhodopsinsfor fast neuronal stimulation at low light levels. Proc. Natl. Acad. Sci. USA Submitted

Berndt A, Yizhar O, Gunaydin LA, Hegemann P, Deisseroth K. 2009. Bi-stable neural state switches. Nat.Neurosci. 12:22934

Bi A, Cui J, Ma YP, Olshevskaya E, Pu M, et al. 2006. Ectopic expression of a microbial-type rhodopsinrestores visual responses in mice with photoreceptor degeneration. Neuron 50:2333

Borue X, Cooper S, Hirsh J, Condron B, Venton BJ. 2009. Quantitative evaluation of serotonin release andclearance in Drosophila. J. Neurosci. Methods 179:3008

Boyden ES, Zhang F, Bamberg E, Nagel G, Deisseroth K. 2005. Millisecond-timescale, genetically targetedoptical control of neural activity. Nat. Neurosci. 8:126368

Brown MT, Bellone C, Mameli M, Labouebe G, Bocklisch C, et al. 2010. Drug-driven AMPA receptorredistribution mimicked by selective dopamine neuron stimulation. PLoS One 5:e15870

Bruegmann T, Malan D, Hesse M, Beiert T, Fuegemann CJ, et al. 2010. Optogenetic control of heart musclein vitro and in vivo. Nat. Methods 7:897900

Busskamp V, Duebel J, Balya D, Fradot M, Viney TJ, et al. 2010. Genetic reactivation of cone photoreceptorsrestores visual responses in retinitis pigmentosa. Science 329:41317

Cardin JA, Carlen M, Meletis K, Knoblich U, Zhang F, et al. 2009. Driving fast-spiking cells induces gammarhythm and controls sensory responses. Nature 459:66367

Cardin JA, Carlen M, Meletis K, Knoblich U, Zhang F, et al. 2010. Targeted optogenetic stimulation andrecording of neurons in vivo using cell-type-specific expression of channelrhodopsin-2. Nat. Protoc. 5:24754

Carter ME, Yizhar O, Chikahisa S, Nguyen H, Adamantidis A, et al. 2010. Tuning arousal with optogeneticmodulation of locus coeruleus neurons. Nat. Neurosci. 13:152633

Chow BY, Han X, Dobry AS, Qian X, Chuong AS, et al. 2010. High-performance genetically targetableoptical neural silencing by light-driven proton pumps. Nature 463:98102

Chuhma N, Tanaka KF, Hen R, Rayport S. 2011. Functional connectome of the striatal medium spiny neuron.J. Neurosci. 31:118392

Ciocchi S, Herry C, Grenier F, Wolff SB, Letzkus JJ, et al. 2010. Encoding of conditioned fear in centralamygdala inhibitory circuits. Nature 468:27782

Covington HE 3rd, Lobo MK, Maze I, Vialou V, Hyman JM, et al. 2010. Antidepressant effect of optogeneticstimulation of the medial prefrontal cortex. J. Neurosci. 30:1608290

Crick FH. 1979. Thinking about the brain. Sci. Am. 241:21932Cruikshank SJ, Urabe H, Nurmikko AV, Connors BW. 2010. Pathway-specific feedforward circuits between

thalamus and neocortex revealed by selective optical stimulation of axons. Neuron 65:23045Deisseroth K. 2010. Controlling the brain with light. Sci. Am. 303:4855Deisseroth K. 2011. Optogenetics. Nat. Methods 8:2629Deisseroth K, Feng G, Majewska AK, Miesenbock G, Ting A, Schnitzer MJ. 2006. Next-generation optical

technologies for illuminating genetically targeted brain circuits. J. Neurosci. 26:1038086Diester I, Kaufman MT, Mogri M, Pashaie R, Goo W, et al. 2011. An optogenetic toolbox designed for

primates. Nat. Neurosci. 14:38797Douglass AD, Kraves S, Deisseroth K, Schier AF, Engert F. 2008. Escape behavior elicited by single,

channelrhodopsin-2-evoked spikes in zebrafish somatosensory neurons. Curr. Biol. 18:113337Ernst OP, Sanchez Murcia PA, Daldrop P, Tsunoda SP, Kateriya S, Hegemann P. 2008. Photoactivation of

channelrhodopsin. J. Biol. Chem. 283:163743Feldbauer K, Zimmermann D, Pintschovius V, Spitz J, Bamann C, Bamberg E. 2009. Channelrhodopsin-2 is

a leaky proton pump. Proc. Natl. Acad. Sci. USA 106:1231722Fork RL. 1971. Laser stimulation of nerve cells in aplysia. Science 171:9078


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


Geschwind D. 2004. GENSAT: a genomic resource for neuroscience research. Lancet Neurol. 3:82Gong S, Doughty M, Harbaugh CR, Cummins A, Hatten ME, et al. 2007. Targeting Cre recombinase to

specific neuron populations with bacterial artificial chromosome constructs. J. Neurosci. 27:981723Gong S, Zheng C, Doughty ML, Losos K, Didkovsky N, et al. 2003. A gene expression atlas of the central

nervous system based on bacterial artificial chromosomes. Nature 425:91725Gordon MD, Scott K. 2009. Motor control in a Drosophila taste circuit. Neuron 61:37384Gorostiza P, Isacoff EY. 2008. Optical switches for remote and noninvasive control of cell signaling. Science

322:39599Gourine AV, Kasymov V, Marina N, Tang F, Figueiredo MF, et al. 2010. Astrocytes control breathing through

pH-dependent release of ATP. Science 329:57175Gradinaru V, Mogri M, Thompson KR, Henderson JM, Deisseroth K. 2009. Optical deconstruction of

parkinsonian neural circuitry. Science 324:35459Gradinaru V, Thompson KR, Deisseroth K. 2008. eNpHR: a natronomonas halorhodopsin enhanced for

optogenetic applications. Brain Cell Biol. 36:12939Gradinaru V, Thompson KR, Zhang F, Mogri M, Kay K, et al. 2007. Targeting and readout strategies for

fast optical neural control in vitro and in vivo. J. Neurosci. 27:1423138Gradinaru V, Zhang F, Ramakrishnan C, Mattis J, Prakash R, et al. 2010. Molecular and cellular approaches

for diversifying and extending optogenetics. Cell 141:15465Grossman N, Poher V, Grubb MS, Kennedy GT, Nikolic K, et al. 2010. Multi-site optical excitation using

ChR2 and micro-LED array. J. Neural Eng. 7:16004Grubb MS, Burrone J. 2010a. Activity-dependent relocation of the axon initial segment fine-tunes neuronal

excitability. Nature 465:107074Grubb MS, Burrone J. 2010b. Channelrhodopsin-2 localised to the axon initial segment. PloS One 5:e13761Gunaydin LA, Yizhar O, Berndt A, Sohal VS, Deisseroth K, Hegemann P. 2010. Ultrafast optogenetic control.

Nat. Neurosci. 13:38792Guo ZV, Hart AC, Ramanathan S. 2009. Optical interrogation of neural circuits in Caenorhabditis elegans. Nat.

Methods 6:89196Hagglund M, Borgius L, Dougherty KJ, Kiehn O. 2010. Activation of groups of excitatory neurons in the

mammalian spinal cord or hindbrain evokes locomotion. Nat. Neurosci. 13:24652Han X, Boyden ES. 2007. Multiple-color optical activation, silencing, and desynchronization of neural activity,

with single-spike temporal resolution. PloS One 2:e299Han X, Qian X, Bernstein JG, Zhou HH, Franzesi GT, et al. 2009. Millisecond-timescale optical control of

neural dynamics in the nonhuman primate brain. Neuron 62:19198Haubensak W, Kunwar PS, Cai H, Ciocchi S, Wall NR, et al. 2010. Genetic dissection of an amygdala

microcircuit that gates conditioned fear. Nature 468:27076Haupts U, Tittor J, Bamberg E, Oesterhelt D. 1997. General concept for ion translocation by halobacterial

retinal proteins: the isomerization/switch/transfer (IST) model. Biochemistry 36:27Hegemann P, Ehlenbeck S, Gradmann D. 2005. Multiple photocycles of channelrhodopsin. Biophys. J.

89:391118Heintz N. 2004. Gene expression nervous system atlas (GENSAT). Nat. Neurosci. 7:483Higley MJ, Sabatini BL. 2010. Competitive regulation of synaptic Ca2+ influx by D2 dopamine and A2A

adenosine receptors. Nat. Neurosci. 13:95866Hira R, Honkura N, Noguchi J, Maruyama Y, Augustine GJ, et al. 2009. Transcranial optogenetic stimulation

for functional mapping of the motor cortex. J. Neurosci. Methods 179:25863Hirase H, Nikolenko V, Goldberg JH, Yuste R. 2002. Multiphoton stimulation of neurons. J. Neurobiol.

51:23747Hires SA, Tian L, Looger LL. 2008. Reporting neural activity with genetically encoded calcium indicators.

Brain Cell Biol. 36:6986Hofmann KP, Scheerer P, Hildebrand PW, Choe HW, Park JH, et al. 2009. A G protein-coupled receptor

at work: the rhodopsin model. Trends Biochem. Sci. 34:54052Hornstein NJ, Pulver SR, Griffith LC. 2009. Channelrhodopsin2 mediated stimulation of synaptic po-

tentials at Drosophila neuromuscular junctions. J. Vis. Exp. 25: doi:10.3791/1133. http://www.jove.com/details.stp?id=1133


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.

http://www.jove.com/details.stp?id=1133http://www.jove.com/details.stp?id=1133


Hull C, Adesnik H, Scanziani M. 2009. Neocortical disynaptic inhibition requires somatodendritic integrationin interneurons. J. Neurosci. 29:899195

Hwang RY, Zhong L, Xu Y, Johnson T, Zhang F, et al. 2007. Nociceptive neurons protect Drosophila larvaefrom parasitoid wasps. Curr. Biol. 17:210516

Ishizuka T, Kakuda M, Araki R, Yawo H. 2006. Kinetic evaluation of photosensitivity in genetically engineeredneurons expressing green algae light-gated channels. Neurosci. Res. 54:8594

Iwai Y, Honda S, Ozeki H, Hashimoto M, Hirase H. 2011. A simple head-mountable LED device for chronicstimulation of optogenetic molecules in freely moving mice. Neurosci. Res. In press

Johansen JP, Hamanaka H, Monfils MH, Behnia R, Deisseroth K, et al. 2010. Optical activation of lateralamygdala pyramidal cells instructs associative fear learning. Proc. Natl. Acad. Sci. USA 107:1269297

Kanbar R, Stornetta RL, Cash DR, Lewis SJ, Guyenet PG. 2010. Photostimulation of Phox2b medullaryneurons activates cardiorespiratory function in conscious rats. Am. J. Respir. Crit. Care Med. 182:118494

Katzel D, Zemelman BV, Buetfering C, Wolfel M, Miesenbock G. 2010. The columnar and laminar organi-zation of inhibitory connections to neocortical excitatory cells. Nat. Neurosci. 14:1007

Kouyama T, Kanada S, Takeguchi Y, Narusawa A, Murakami M, Ihara K. 2010. Crystal structure of thelight-driven chloride pump halorhodopsin from Natronomonas pharaonis. J. Mol. Biol. 396:56479

Kravitz AV, Freeze BS, Parker PR, Kay K, Thwin MT, et al. 2010. Regulation of parkinsonian motor be-haviours by optogenetic control of basal ganglia circuitry. Nature 466:62226

Lee JH, Durand R, Gradinaru V, Zhang F, Goshen I, et al. 2010. Global and local fMRI signals driven byneurons defined optogenetically by type and wiring. Nature 465:78892

Leifer AM, Fang-Yen C, Gershow M, Alkema MJ, Samuel AD. 2011. Optogenetic manipulation of neuralactivity in freely moving Caenorhabditis elegans. Nat. Methods 8:14752

Levskaya A, Weiner OD, Lim WA, Voigt CA. 2009. Spatiotemporal control of cell signalling using a light-switchable protein interaction. Nature 461:9971001

Lewis TL Jr, Mao T, Svoboda K, Arnold DB. 2009. Myosin-dependent targeting of transmembrane proteinsto neuronal dendrites. Nat. Neurosci. 12:56876

Li X, Gutierrez DV, Hanson MG, Han J, Mark MD, et al. 2005. Fast noninvasive activation and inhibition ofneural and network activity by vertebrate rhodopsin and green algae channelrhodopsin. Proc. Natl. Acad.Sci. USA 102:1781621

Liewald JF, Brauner M, Stephens GJ, Bouhours M, Schultheis C, et al. 2008. Optogenetic analysis of synapticfunction. Nat. Methods 5:895902

Lima SQ, Hromadka T, Znamenskiy P, Zador AM. 2009. PINP: a new method of tagging neuronal populationsfor identification during in vivo electrophysiological recording. PloS One 4:e6099

Lima SQ, Miesenbock G. 2005. Remote control of behavior through genetically targeted photostimulationof neurons. Cell 121:14152

Lin JY, Lin MZ, Steinbach P, Tsien RY. 2009. Characterization of engineered channelrhodopsin variantswith improved properties and kinetics. Biophys. J. 96:180314

Llewellyn ME, Thompson KR, Deisseroth K, Delp SL. 2010. Orderly recruitment of motor units underoptical control in vivo. Nat. Med. 16:116165

Lobo MK, Covington HE 3rd, Chaudhury D, Friedman AK, Sun H, et al. 2010. Cell type-specific loss ofBDNF signaling mimics optogenetic control of cocaine reward. Science 330:38590

Low SE, Ryan J, Sprague SM, Hirata H, Cui WW, et al. 2010. touche is required for touch-evoked generatorpotentials within vertebrate sensory neurons. J. Neurosci. 30:935967

Luecke H, Schobert B, Richter HT, Cartailler JP, Lanyi JK. 1999a. Structural changes in bacteriorhodopsinduring ion transport at 2 angstrom resolution. Science 286:25561

Luecke H, Schobert B, Richter HT, Cartailler JP, Lanyi JK. 1999b. Structure of bacteriorhodopsin at 1.55 Aresolution. J. Mol. Biol. 291:899911

Man D, Wang W, Sabehi G, Aravind L, Post AF, et al. 2003. Diversification and spectral tuning in marineproteorhodopsins. EMBO J. 22:172531

Mattis J, Tye KM, Ramakrishnan C, OShea D, Gunaydin LA, et al. 2011. Clarifying the optogenetics toolbox:a comprehensive analysis of new and existing opsins for scientific application. Nat. Meth. Submitted

McLean DL, Fetcho JR. 2011. Movement, technology and discovery in the zebrafish. Curr. Opin. Neurobiol.21:11015


Ann

u. R

ev. N

euro

sci.

2011

.34:

389-

412.

Dow

nloa

ded

from

ww

w.a

nnua

lrevi

ews.o

rgby

Sta

nfor

d U

nive

rsity

- M

ain

Cam

pus -

Rob

ert C

row

n La

w L

ibra

ry o

n 07

/06/

11. F

or p

erso

nal u

se o

nly.


Miesenbock G, Kevrekidis IG. 2005. Optical imaging and control of genetically designated neurons in func-tioning circuits. Annu. Rev. Neurosci. 28:53363

Mohanty SK, Reinscheid RK, Liu X, Okamura N, Krasieva TB, Berns MW. 2008. In-depth activation ofchannelrhodopsin 2-sensitized excitable cells with high spatial resolution using two-photon excitationwith a near-infrared laser microbeam. Biophys. J. 95:391626

Nagel G, Brauner M, Liewald JF, Adeishvili N, Bamberg E, Gottschalk A. 2005. Light activation ofchannelrhodopsin-2 in excitable cells of Caenorhabditis elegans triggers rapid behavioral responses. Curr.Biol. 15:227984

Nagel G, Ollig D, Fuhrmann M, Kateriya S, Musti AM, et al. 2002. Channelrhodopsin-1: a light-gated protonchannel in green algae. Science 296:239598

Nagel G, Szellas T, Huhn W, Kateriya S, Adeishvili N, et al. 2003. Channelrhodopsin-2, a directly light-gatedcation-selective membrane channel. Proc. Natl. Acad. Sci. USA 100:1394045

Oesterhelt D, Stoeckenius W. 1971. Rhodopsin-like protein from the purple membrane of Halobacteriumhalobium. Nat. New Biol. 233:14952

Oh E, Maejima T, Liu C, Deneris E, Herlitze S. 2010. Substitution of 5-HT1A receptor signaling by alight-activated G protein-coupled receptor. J. Biol. Chem. 285:3082536

Papagiakoumou E, Anselmi F, Begue A, de Sars V, Gluckstad J, et al. 2010. Scanless two-photon excitationof channelrhodopsin-2. Nat. Methods 7:84854

Peralvarez-Marn A, Marquez M, Bourdelande JL, Querol E, Padros E. 2004. Thr-90 plays a vital role in thestructure and function of bacteriorhodopsin. J. Biol. Chem. 279:164039

Petreanu L, Huber D, Sobczyk A, Svoboda K. 2007. Channelrhodopsin-2-assisted circuit mapping of long-range callosal projections. Nat. Neurosci.

fenno annrev 2011

Documents