Optical Optogenetics

Biol. Cell (2013) 105, 1–22 DOI: 10.1111/boc.
201200087
Review
Optical developments for

optogenetics
Eirini Papagiakoumou 1
Wavefront-Engineering Microscopy Group, Neurophysiology and New Microscopies Laboratory, CNRS UMR 8154, Inserm S603, Paris
Descartes University, 75270 Paris Cedex 06, France
Brain intricacies and the difficulty that scientists encounter in revealing its function with standard approaches
such as electrical stimulation of neurons have led to the exploration of new tools that enable the study of neural
circuits in a remote and non-invasive way. To this end, optogenetics has initialised a revolution for neuroscience
in the last decade by enabling simultaneous monitoring and stimulation of specific neuronal populations in intact
brain preparations through genetically targeted expression of light sensitive proteins and molecular photoswitches.
In addition to ongoing molecular probe development and optimisation, novel optical techniques hold immense
potential to amplify and diversify the utility of optogenetic methods. Importantly, by improving the spatio-temporal
resolution of light stimulation, neural circuits can be photoactivated in patterns mimicking endogenous physiological
processes. The following synopsis addresses the possibilities and limitations of optical stimulation methods applied
to and developed for activation of neuronal optogenetic tools.
Introduction Electrophysiology suffers, however, from funda-

The attempts to explore the function of the nervous mental limitations. Implanting electrodes into the
system dates back more than 200 years to Luigi Gal- brain is invasive and difficult to accomplish in freely
vani’s (1737–1798) famous experiments in the late moving animals. Furthermore, there is a physical
18th century. By causing a frog’s legs to twitch by limit to the number of electrodes that can be simul-
connecting the lumbar nerve to an electrical current taneously inserted into brain tissue. Hence, intracel-
source (Piccolino, 2006), Galvani revealed the first lular patch-clamp recordings are limited to a small
and perhaps most fundamental principle of the neu- number of cells, compared with population size of
ral code: that information is transmitted in the form most cerebral circuits (Watt et al., 2009; Ko et al.,
of electrical impulses. Galvani’s approach of probing 2011). Although patch-clamp of principal dendrites
the nervous system with electrodes remains the prin- or axons has been demonstrated (Stuart and Sakmann,
ciple mode of neural circuit investigation today due 1994; Stuart et al., 1997), access to small compart-
to high signal-to-noise (S/N) ratios and temporal res- ments such as dendritic branches and spines continues
olution, as exemplified, for example, by patch clamp to prohibit direct electrical recording. Alternately,
recording (Hamill et al., 1981). extracellular electrodes permit simultaneous access
to several cells and populations (Buzsáki, 2004) but
at the cost of spatial resolution.
1 To whom correspondence should be addressed (email
The first and primary application of light to cel-
eirini.papagiakoumou@parisdescartes.fr) lular biology was for imaging. In the 1980s, it was
Key words: Digital holography, Generalised phase contrast, Illumination meth- exploited for trapping and manipulating of micro-
ods, Optogenetics, Spatio-temporal focusing.
Abbreviations: used AODs, acousto-optic deflectors; APs, action potentials; particles and cells (Ashkin et al., 1986; Ashkin and
CGHs, computer-generated holograms; ChR2, channelrhodopsin-2; DH, digital Dziedzic, 1987) and its use for investigation of cellu-
holography; DMDs, digital micromirror devices; fMRI, functional magnetic
resonance imaging; FWHM, full width at half-maximum; GPC, generalised lar structure and function has exponentially grown in
phase contrast; LC, liquid crystal; LCD, liquid crystal display; LCOS, liquid the 1990s with the discovery of genetically targetable
crystal on silicon; LED, light-emitting diode; NA, numerical aperture; PCF,
phase contrast filter; ROIs, regions of interest; SLMs, spatial light modulators;
fluorescent proteins (Heim et al., 1994, 1995). Over
TF, temporal focusing; VSFPs, voltage-sensitive fluorescent proteins. the past decade, light has become an ‘active’ tool for

C 2013 Société Française des Microscopies and Société de Biologie Cellulaire de France. Published by John Wiley & Sons Ltd 1
E. Papagiakoumou
neuronal stimulation. The idea of using light as an coalesced to form the field of optogenetics (Deisseroth
alternative way to stimulate neurons is not new, how- et al., 2006; Miesenböck, 2009).
ever, dating back to 1971 when Richard Fork used Excitation of the cells expressing such photosen-
laser light to stimulate nerve cells (Fork, 1971). Al- sitive compounds is most commonly accomplished
though Fork’s approach was not widely adopted at by illuminating the sample with a common epi-
that time, probably because the cellular response he fluorescence lamp or a light-emitting diode (LED).
got was considered a result of side effects, today a This illumination method is simple to implement
growing number of researchers choose light in place and easily adaptable to in vivo applications through
of electrodes as a less invasive wireless form of activa- an optical fibre. Its lack of specificity has been coun-
tion (Scanziani and Hausser, 2009). Moreover, light terweighed by targeting photosensitive proteins to
enables communication with many receivers in par- specific neuronal subpopulations with specific gene
allel with spatial and temporal resolution adequate to promoters. Thus, although the light is incident across
follow the dynamics of physiological processes. Since a large field, only targeted neurons are excited. The
1971, another work reporting laser irradiation of neu- majority of current studies utilise channelrhodopsin-2
rons was published in 1983 by Farber and Grinvald (ChR2), a cation channel highly sensitive to vis-
(1983), who used laser excitation of leech neurons ible light. In ChR2-expressing neurons, APs can
stained with a fluorescent dye to evoke action po- be evoked with excitation powers of approximately
tentials (APs) in the cells. In the 1990s, the develop- 1 mW/mm2 at the maximal absorption wavelength of
ment of caged glutamate enabled indirect photostim- 470 nm, and thus, wide-field illumination with blue
ulation of neurons through photolysis of glutamate light provides effective excitation. Indeed, the high
with flash lamps or lasers (Callaway and Katz, 1993; sensitivity of ChR2 has been critical for its success-
Katz and Dalva, 1994). Later, pulsed femtosecond ful utilisation, especially for in vivo studies (Aravanis
lasers were utilised for neuronal stimulation through et al., 2007; Gradinaru et al., 2007).
partial disruption of cell membranes (Hirase et al., Still, elaborated photosensitive tools for neuronal
2002). stimulation require elaborated methods to exploit
The real revolution, however, came when stud- their full potential. This is something that was un-
ies on rendering neurons actively responding to derstood from the beginning; Richard Fork finishes
light started coming up. The laboratories of Gero his article on neuronal photoactivation with the fol-
Miesenböck, while at Memorial Sloan-Kettering lowing phrase: ‘Attempts are being made to improve
Cancer Center, and of Ehud Isacoff, Richard H. this technique by limiting the area of stimulation
Kramer and Dirk Trauner, then all at the Univer- still further’. To individuate subsets of genetically
sity of California, Berkeley, employed methods to identical connected cells, for example, or establish
genetically engineer and target proteins that directly the role of specific spatiotemporal excitatory patterns
alter membrane conductivity upon light illumina- in guiding animal behaviour, innovative optical tech-
tion of specific wavelength (Zemelman et al., 2002; niques are needed, able to modulate stimulation for
Banghart et al., 2004). The revolution continued with reproducing physiological patterns of neuronal acti-
functional expression of natural photosensitive pro- vation. This is, however, a challenging technical issue
teins that can be used as cell actuators, such as chan- due to the complexity of neuronal systems in terms
nelrhodopsin (Nagel et al., 2003), and light-activated of spatiotemporal precision.
G-protein coupled receptors (GPCRs; Panda et al., The following review details optical methods used
2005; Qiu et al., 2005; Levitz et al., 2013), as so far for optogenetics as well as recent technical
well as genetically encoded fluorescent proteins, both developments and their potential applications. The
calcium-based (Nakai et al., 2001) and voltage-based techniques presented here will be discussed specifi-
(Siegel and Isacoff, 1997; Dimitrov et al., 2007), that cally in the context of optogenetic applications, but
can be used as functional reporters of neuronal ac- in principle can be extended to any application de-
tivity. Continued efforts to improve the performance manding spatially and temporally precise light stim-
of these molecular innovations for study of neuronal ulation, such as photolysis of caged neurotransmit-
function and treatment of diseases and disorders have ters, morphological and functional imaging.
2 www.biolcell.net | Volume (105) | Pages 1–22

Optical developments for optogenetics Review
Wide-field illumination approaches problem is to perform imaging with optical methods
Wide-field illumination is the first method used to that provide better spatial resolution and less proba-
stimulate optogenetic tools, as mentioned in the in- bility to excite both the functional reporters and the
troduction, due to its simplicity, requiring essentially actuators. In the studies described above, for exam-
inexpensive, off-the-shelf components, such as a flu- ple, 2P calcium imaging was used to excite the cal-
orescence lamp coupled to a fast shutter. In fact, the cium indicators Fluo-5F (Zhang and Oertner, 2007)
first photoactivation experiment of sensitised verte- or Rhod-2-AM (Bundschuh et al., 2012), whereas
brate neurons to light was done with white light il- near-membrane calcium imaging with evanescent-
lumination (Zemelman et al., 2002). Epi-fluorescent field excitation (Axelrod, 1981) of Fluo-4 or Xrhod-1
activation has been used for a first characterisation of was performed to monitor the calcium responses
optogenetic actuators (Zemelman et al., 2003; Bang- in astrocytes (Li et al., 2012). Recently, a family
hart et al., 2004; Zhang et al., 2007; Kleinlogel et al., of red-fluorescent calcium indicators has been pre-
2011; ). It has been also used in combination with sented (Collot et al., 2012) with longer excitation
fast CCD cameras for recording signals from voltage- and emission wavelengths, suitable for simultaneous
sensitive fluorescent proteins (VSFPs), as in that case, 2P calcium imaging and excitation of optogenetic
it facilitates collection of photons from large areas, es- actuators.
sential for VSFPs wherein the collected fluorescence In vivo applications demand further adaptation of
signal is very small (Chanda et al., 2005; Akemann optogenetics instrumentation, even for simple wide-
et al., 2010). In addressing more specific biological field excitation. For in vivo light delivery in freely
questions, epi-fluorescent activation of ChR2 at 460 moving animals, lasers or LEDs of sufficient power
to 470 nm was used, for example in mechanosensory are coupled to optical fibres (∼10–15 mW are re-
neurons of the nematode Caenorhabditis elegans (C. quired at the output of a fibre of 100 μm in
elegans) to trigger specific behaviours (Nagel et al., diameter) and attached to the animal’s skull with
2005), to map circuits between presynaptic ChR2- stereotactic surgery. The diameter of the fibre core
positive neurons and postsynaptic neurons (Petreanu (without plastic cladding) depends on the species and
et al., 2007), to investigate active synaptic contacts in the experiment. For mice, up to 300-μm fibres can
rat hippocampal slice cultures (Zhang and Oertner, be used without compromising animal movement,
2007) and long-term potentiation in combination whereas rats can tolerate up to 400-μm fibres (Zhang
with two-photon (2P) uncaging (Zhang et al., 2008). et al., 2010). Commercial vendors offer the lasers pre-
Wide-field excitation of ChR2 was also used to study coupled to optical fibres. To target deep brain struc-
Ca2+ elevation in astrocytic cultures in comparison tures, thin optical fibres can be used to efficiently
to other optogenetic actuators (Li et al., 2012), such transmit sufficient powers of light to the target area
as the Ca2+ -translocating channelrhodopsin (CatCh; (Adamantidis et al., 2007; Aravanis et al., 2007). Fur-
Kleinlogel et al., 2011) and the light-gated Ca2+ - thermore, fibreoptics fused to recording electrodes
permeable ionotropic glutamate receptor (LiGluR; (optrodes) have also been developed for simultaneous
Volgraf et al., 2006), or to examine the function of electrophysiological recording and photostimulation
dopamine in the olfactory bulb of adult zebrafish of genetically expressed photosensitive ion channels
(Bundschuh et al., 2012). and pumps (Gradinaru et al., 2007; Zhang et al.,
In some of these studies, photoactivation is com- 2009; Anikeeva et al., 2012; Wang et al., 2012)
bined with calcium imaging for monitoring the and GPCRs (Gutierrez et al., 2011) in freely mov-
neuronal activity. Combining photostimulation with ing mice. For superficial stimulation of cortical lay-
calcium or voltage imaging necessitates sufficiently ers, miniaturised LEDs can also be directly mounted
distinct excitation spectra for the photoexcitable neu- above the brain over a thinned skull area or cranial
ronal actuator and the functional reporter in or- glass window (Gradinaru et al., 2007; Huber et al.,
der to avoid light-driven neuronal excitation during 2008).
imaging and the inverse. Indeed, dyes with excita- Simple wide-field illumination of optogenetic
tion spectra overlapping little with common optoge- probes, despite its inherent lack of spatial speci-
netic tools, such as ChR2, are commercially available ficity, has nonetheless generated important in-
(Zhang et al., 2006). An alternative solution to this sights into neuronal function and dysfunction. For

E. Papagiakoumou
instance, by combining optogenetics with func- to combine photoactivation with fluorescent func-
tional magnetic resonance imaging (fMRI), inves- tional reporters wherein the use of wide-field excita-
tigators verified that the firing of local excitatory tion often reduces the S/N ratio, due to out-of-focus
neurons is sufficient to trigger downstream haemo- fluorescence signal.
dynamic phenomena detected by fMRI scanners (Lee
et al., 2010) or to facilitate appropriate plasticity Patterned visible excitation
by guiding the cortical reorganisation after nerve in- Light patterning techniques enable fast, remote
jury (Li et al., 2011). In addition, photostimulation modification of the spatial configuration of light de-
of the light-activated chloride pump halorhodopsin livered to tissue, without the need to intervene on
from Natronomonas pharaonis (eNpHR) (Zhang et al., the optical setup during the experiment. Such tech-
2007) or of the K+ -channel small molecule photo- niques can be useful for stimulating a subpopulation
switch AAQ (acrylamide-azobenzene-quaternary am- of genetically identical neurons or targeting subcel-
monium; Fortin et al., 2008) allows restoration of lular structures, such as dendrites, axons or dendritic
visual responses in mice and in human ex vivo retinas spines. Several schemes have been proposed to per-
(Busskamp et al., 2010; Polosukhina et al., 2012). form spatial shaping of the excitation beam, gener-
ChR2 activation was used to control heart muscle ally classified into two categories: scanning methods,
both in vitro and in vivo (Bruegmann et al., 2010), in which serial excitation of several regions of inter-
and breathing in anaesthetised mice (Depuy et al., est (ROIs) is performed by rapidly scanning the laser
2011). Optogenetic stimulation also shows promise focus, and parallel methods, in which all ROIs are
for treating mental diseases, already used to reveal simultaneously excited.
that cholinergic interneurons can block cocaine con-
ditioning (Witten et al., 2010) and to activate fear Scanning methods
memory recall (Liu et al., 2012). Moreover, optoge- The easiest way to create pattern-like excitation is
netic tools have been used to treat sleeping disor- by scanning a focused laser beam (Katz and Dalva,
ders by stimulating hypocretin neurons implicated 1994). This approach has been used to map synaptic
in narcolepsy (Adamantidis et al., 2007), to stimu- connectivity by photostimulating ChR2 with blue
late dopamine neurons in order to understand their laser radiation focused to a small spot (∼1 μm) and
function in depression and substance abuse (Tsai et al., scanned over different locations to assess light sensi-
2009), to acquire fundamental knowledge for Parkin- tivity (Wang et al., 2007a). Other investigators fo-
son’s disease and the mechanisms of action of ther- cused the laser to a near cylindrical beam of 6 to
apeutic interventions (Kravitz et al., 2010) and to 16 μm full width at half-maximum (FWHM),
control the activity of neocortical parvalbumin in- scanned over the dendrites of the recorded neuron
terneurons, implicated to mental dysfunctions char- (Petreanu et al., 2009). Single-cell scanning excita-
acteristic of schizophrenia or autism (Cardin et al., tion of ChR2 was also recently used to map network
2009; Sohal et al., 2009). connections between two distinct cortical inhibitory
In combination with the high sensitivity to visi- networks of the visual cortex in vivo (Wilson et al.,
ble wavelengths, the above-mentioned examples were 2012). In these examples, the position of the laser
made possible through a remarkable development beam was changed by rapidly deflecting its direc-
of gene delivery methods including viral expression, tion with a pair of galvanometer-driven mirrors that
transgenic mice, in utero electroporation (see Zhang dictated the position of the spot in x and y. Such scan-
et al., 2010 for a complete description of genetic ning systems are now commercially available and can
strategies to target specific neuronal populations). be easily adapted to standard microscopes by imag-
Both the resounding success and limitations of early ing the scanner with a telescope to the back aperture
optogenetic studies manifested the need for more flex- of the objective (Figure 1A). The main advantage
ible and specific illumination tools to differentially of scanning the beam with mirrors is that there are
activate cells of the same population or to target sub- no significant light losses in the optical path. How-
cellular structures. Thus, the development of optical ever, scan speed is limited by mechanical inertia and
techniques that offer higher spatial and temporal re- by the dwell time necessary to generate a sufficient
solution has become a necessity, especially for efforts physiological response (scanning frequencies of few

Figure 1 Laser illumination techniques

(A) Schematic illustrating the principle of scanning approaches. The scanner (galvanometer-driven mirrors or AODs) is imaged
onto the back focal plane of the microscope objective (blue dashed lines). In this manner, angular deflections induced by the
scanner to the beam, correspond to lateral translations in the focal plane of the objective. Light blue dotted lines show a different
deflection angle of the beam. (B) Schematic of a typical setup for amplitude modulation. Each pixel of the SLM (e.g. a DMD)
is imaged at the sample plane (blue dashed lines). The desired intensity pattern is addressed to the SLM and reproduced at
the focal plane of the objective. (C) Schematic of a phase modulation setup. Each pixel of the SLM is imaged at the back
focal plane of the objective (blue dashed lines). A pattern corresponding to the proper phase modulation (different grey levels
represent phase modulation from 0 to 2π) is addressed to the SLM. This phase modulation is imaged to the back focal plane
of the objective and transformed to the desired intensity pattern at the sample plane. Note that in phase-only modulation, the
intensity profile is characterised by speckle.
kHz have been reported; Gasparini and Magee, 2006). Nguyen et al., 2001) or acousto-optic deflectors
Thus, although scanning has been successfully ap- (AODs) (Reddy et al., 2008) to deflect the beam’s
plied to excite single cells, the scan speed limitation path much faster with frequencies reaching tens of
has hindered application of this method to neural kHz. Resonant scanning galvanometer mirrors are
network questions requiring simultaneous stimula- mirrors that oscillate in a sinusoid manner. These de-
tion of multiple neurons. vices cycle at a resonant frequency on the order of
Scan speed limitations can be addressed in part by 4 to 8 kHz, while bidirectional scanning increases
using resonant scanning mirrors (Fan et al., 1999; the effective scan rate to approximately 16 kHz

E. Papagiakoumou
(Fan et al., 1999). AODs are devices in which a manipulating the activity of neurons expressing the
propagating sound wave of radio frequency through ameliorated chimera variant of channelrhodopsin
an acousto-optic medium establishes a grating that ChIEF (Lin et al., 2009), achieving high temporal
diffracts a laser beam at a precise angle, which can resolution with single-photon excitation (1PE).
be modified within few microseconds by changing Scanning methods, even those using resonant scan-
the frequency of the sound wave. When a monochro- ners or AODs to achieve fast scan rates, remain serial
matic light beam traverses an AOD, a large fraction approaches limited by the dwell time, a parameter in-
of light is ‘deflected’ at an angle, θ, that is linearly dependent of how fast the beam is scanned from one
related to the sound frequency, f: θ = λf/v, where λ site to another. Exciting one or two cells (Wang et al.,
is the wavelength of the incident light and υ is the 2011) in a ‘simultaneous’ manner may be possible by
velocity of the propagating sound wave. This effect the long opening time of opsins (Lin et al., 2009).
can be used for highly versatile inertia-free scanning However, as temporal resolution of scanning meth-
schemes, because the deflection angle is controlled ods, TS , is determined by the sum of the scan time (ts )
by electronically generated frequencies without and the dwell time (tdwell ) for each visited position:
moving any mass-containing elements (i.e. mir- TS = N(tS + tdwell ), where N is the number of visited
rors). Consequently, a two-AOD scanning system en- sites, the number of cells that can be serially accessed
ables two-dimensional (2D) patterns in discontinuous for near-simultaneous stimulation is quickly limited.
scan modes, and more complex four-AOD configu-
rations, with counter-propagating chirped acoustic Parallel methods
waves enable in addition axial scanning of the laser Parallel excitation methods arose as a solution for the
beam (Reddy and Saggau, 2005; Reddy et al., 2008). ‘simultaneity’ problem. These methods utilise spa-
However, the excitation field that can be achieved tial light modulators (SLMs) to change the intensity
with AODs is small compared with scanning mir- profile of the laser beam. In this way, 2D or 3D dis-
rors, (typical values 150 × 150 – 200 × 200 μm2 ; tributions of multiple diffraction-limited spots and
Shoham et al., 2005; Reddy et al., 2008) depending arbitrary excitation patterns mimicking cellular mor-
on the available range of acoustic frequencies, which phology can be realised at the sample plane. Hence,
defines the maximum beam deflection angle. More- parallel methods enable high spatial flexibility, as il-
over, as AODs are diffracting elements, they have a lumination patterns can be easily adapted to different
certain diffraction efficiency (ratio of light that goes to experimental requirements within the same optical
the first diffraction order relative to the total amount configuration. The beam incident on the SLM can
of light entering the AODs) that ranges between 50% be described as an electromagnetic wave of a certain
and 70% for a two-AOD scanning system (Shoham amplitude, A, and spatial phase, : ε = Aei , where
et al., 2005; Losavio et al., 2009). Recently, new i is the imaginary unit. Thus, beam modulation can
design concepts that improve the performance of a be achieved either by modulating the amplitude or
4-AOD system have been presented, with the possi- the phase with the SLM.
bility to image neuronal somata within a volume of
700 × 700 × 1400 μm3 and dendrites within a vol- Amplitude modulation
ume of 290 × 290 × 200 μm3 (Katona et al., 2012). Amplitude modulation is performed by directly shap-
Resonant scanners have been used mainly for ing the intensity of light in a plane conjugated to
2P calcium imaging (Fan et al., 1999; Nguyen the sample plane (Figure 1B). The SLMs most com-
et al., 2001; Rochefort et al., 2009), whereas AOD monly utilised for this configuration are digital mi-
scanners have been widely used in 2D for uncaging cromirror devices (DMDs), which are arrays of mi-
applications (Shoham et al., 2005; Losavio et al., cromirrors (commercially available devices with up
2009) and 2P calcium imaging (Salomé et al., 2006; to 1024 × 768 or 1920 × 1080 individual micromir-
Otsu et al., 2008), also including three dimensions rors) suspended on air gap and controlled electrostat-
(3D) (Reddy and Saggau, 2005; Reddy et al., 2008; ically. Each mirror can be rotated along its diagonal
Katona et al., 2012). In addition, investigators by ±12◦ from the unpowered position via a control
recently used AODs to photostimulate optoge- signal (Knapczyk and Krishnan, 2005). In this way,
netic actuators (Wang et al., 2011), specifically light can be independently deflected by each mirror

creating spatially extended arbitrary patterns. DMDs pattern is created by redirecting unwanted light out
have been used widely for optogenetics in cell cultures of the excitation field in the case of DMDs or by
(Wang et al., 2007b), to control neuronal activity pat- switching off the corresponding emitters in micro-
terns in zebrafish (Wyart et al., 2009; Arrenberg et al., LEDs. For this reason, amplitude modulation tech-
2010; Zhu et al., 2012) and in C. elegans (Guo et al., niques are primarily utilised for single-photon exci-
2009; Leifer et al., 2011), and to probe odour-coding tation of optogenetic actuators, for which demands
mechanisms in vivo in the olfactory bulb (Dhawale in terms of power are modest compared with 2P ex-
et al., 2010). As an alternative to DMDs, digital citation (2PE).
amplitude modulation with a modified off-the-shelf
liquid crystal display (LCD) projector has been also Phase modulation
presented (Stirman et al., 2011, 2012). For light patterning through phase modulation, the
Works wherein high-power micro-LED arrays in phase wavefront of the laser beam is modulated at
a 1:1 imaging configuration or demagnified at the the back aperture of the microscope objective such
sample plane for excitation of dendrites were used that the desired excitation pattern is generated in fo-
to excite ChR2-expressing neurons in vitro (Degenaar cus at the sample plane. Phase modulation is most
et al., 2009; Grossman et al., 2010) have been pre- commonly performed with liquid crystal (LC)-based
sented, as well. However, as micro-LEDs have fewer SLMs, pixelated liquid crystal matrices that modu-
pixels than DMDs (64 × 64), a serious compromise late the spatial phase of an incident wavefront in a full
between the spatial resolution and the excitation field range of [0, 2π] by controlling the refractive index of
must be made in that case. For example, in an exci- each pixel. Refractive index changes are induced by
tation field of 3 × 3 mm2 , a single light spot is reorienting liquid crystal molecules with an electric
30 μm FWHM, although reducing the size of the field. The SLM typically occupies a plane conjugated
light spot to 3 μm FWHM limits the excitation with the back focal plane of the objective (Figure 1C).
field to 0.3 × 0.3 mm2 (Grossman et al., 2010). A variety of algorithms exist for calculation of phase
To tightly couple optical stimulation with 2D elec- modulation maps addressed to the SLM (also called
trical recording over multiple sites for in vivo studies computer-generated holograms, CGHs), the choice of
in mice and rats, tapered optical fibres were glued which depends on the application. For example, it-
to commercially available four-electrode or eight- erative Fourier-transform algorithms (Gerchberg and
electrode linear arrays (Royer et al., 2010). These Saxton, 1972) are used to generate extended excita-
2D electrical–optical arrays were used for stimulation tion patterns, as objective lens focusing can be sim-
and parallel recording of ChR2-transfected rat hip- ulated by a mathematical Fourier transform. For the
pocampal neurons. Other investigators extended this generation of multiple diffraction-limited spots in
approach for 2D control of light delivery with mi- independent axial planes, a similar, more efficient it-
crofabricated multiwaveguide probes (Zorzos et al., erative algorithm improves the intensity uniformity
2010) or probes with integrated micrometer-sized among the spots (Di Leonardo et al., 2007); however,
LEDs (McAlinden et al., 2013). simpler algorithms enabling control of spots’ lateral
The greatest advantage of amplitude modulation position with a grating phase effect and spots’ ax-
methods is the high temporal and, in many cases, ial position with a lens phase effect can also be used
high lateral resolution. DMDs, for example, can up- (Liesener et al., 2000).
date intensity patterns at rates of 1–10 kHz (Wilt Computer-generated hologram laser beam phase
et al., 2009) and micro-LEDs can switch on a nanosec- modulation, here forward referred to as digital holo-
ond time scale (Grossman et al., 2010), thus easily graphy (DH), was first utilised for generation of mul-
mimicking rapid physiological processes. The limit tiple optical traps with optical tweezers (Reicherter
on time resolution for amplitude modulation meth- et al., 1999; Curtis et al., 2002). Application of
ods is in fact the activation/deactivation kinetics of DH for neuronal photoactivation was first demon-
photoactive compounds, the primary determinant of strated for photolysis applications (Lutz et al., 2008;
dwell time. Nikolenko et al., 2008). Although DH found broad
The Achilles’ heel of amplitude modulation, how- application for 1P uncaging (Zahid et al., 2010;
ever, is the large fraction of light power lost, as the Anselmi et al., 2011; Yang et al., 2011; Santos

E. Papagiakoumou
et al., 2012), its utility for 1PE of optogenetic tools Figure 2 Single-photon holographic excitation of ChR2
was only recently demonstrated in stimulation of
(A) Wide-field fluorescence images of HEK cells express-
ChR2-expressing retinal ganglion cells of blinded
ing GFP-tagged ChR2-H134R (top) are superimposed with
retinas for bionic vision restoration (Reutsky-Gefen holographic illumination patterns imaged on a thin layer of
et al., 2013). Figure 2 illustrates another example of rhodamin-6G (in purple). The fluorescence excitation of each
1P holographic excitation of ChR2-expressing HEK spot is shown separately (inset). Underneath each image
cells and cultured neurons. Specifically, Figure 2A is the corresponding electrophysiology recorded whole-cell
shows holographic excitation extended over an entire current evoked by holographic photostimulation of ChR2.
cell’s surface, also demonstrating DH’s fine spatial (B) Wide-field fluorescence image of a cultured neuron ex-
specificity, as patched cells register no response when pressing GFP-tagged ChR2-H134R (top). Whole-cell currents
patterned light is delivered to neighbouring cells. (middle) evoked by holographic photostimulation of ChR2 on
Figure 2B illustrates modulation of the induced pho- the cell soma (black trace) or on the principal dendrite (red
tocurrent by exciting the whole soma (black trace) trace). The corresponding excitation patterns imaged on a
or the principal dendrite (red trace) of the recorded rhodamin-6G thin film are shown in the black and red boxes.
neuron. Excitation parameters: λ = 405 nm, 10 ms pulses, laser power
An advantage of phase over amplitude modulation 3.5 μW at the exit of a 40x, 0.8 NA objective. Scale bars in-
is that power losses are minimised by redistributing, dicate 20 μm. Unpublished data, courtesy of A. Bègue, F.
rather than deviating or blocking, light at the sample Anselmi and V. Emiliani (Wavefront-Engineering Microscopy
plane for formation of the desired intensity profile. Group, Neurophysiology and New Microscopies Laboratory).
Furthermore, DH offers an improved axial resolution
for extended spots in comparison with large Gaus-
sian beams (Lutz et al., 2008; Zahid et al., 2010;
Vaziri and Emiliani, 2012), arising from the abil-
ity to generate large excitation areas by overfilling
the objective back aperture, exploiting its full focus-
ing strength. In contrast, large area excitation with
Gaussian beams necessitates underfilling the objec-
tive back aperture at cost of the axial confinement of
the excitation volume. Moreover, as DH is a scan-less
method, the temporal resolution of photoactivation
is imposed only by the dwell time necessitated by
optogenetic actuator kinetics: TDH = tdwell .
DH suffers two primary drawbacks: light lost to
diffraction at the SLM and intensity inhomogeneities
caused by speckle. The spatially varying diffraction
efficiency of LC-SLMs causes light loss in the for-
mation of zero and higher diffraction orders, limit-
ing the excitation field (Arrizón and Testorf, 1997;
Palima and Daria, 2006). Light sent to zero or higher
diffraction orders can be minimised by using liquid
crystal on silicon (LCOS) SLMs and, in addition, sup-
pressed through adaptation of the original phase holo-
gram (Palima and Daria, 2007; Jesacher and Booth,
2010; Ronzitti et al., 2012). The maximum excita-
tion field size for DH depends on the SLM pixel size,
α, and its demagnification at the sample plane (Golan
et al., 2009; Yang et al., 2011). In the lateral direc-
tion, it can be calculated by the following equation:

λf (Golan et al., 2009). Although supplying sufficient
Xmax = Ymax = 2 2αobj ff12 , where λ is the illumi-
nation wavelength, fobj , f1 and f2 are the focal lengths light power for 1PE in optogenetics (Reutsky-Gefen
of the objective, the first and the second lens used to et al., 2013), significant losses to diffraction preclude
demagnify the SLM array to the back focal plane of implementation of this solution for applications de-
the objective (Figure 1C). For example, for a 40×, manding more power, such as 2PE.
0.8 NA objective, frequently used for biological ap- A different approach to project complex field holo-
grams, using a phase to intensity mapping technique,
plications, λ = 473 nm, α = 20 μm and ff12 = 1.5,
has been recently presented (Go et al., 2011). Inves-
the excitation field is roughly 160 × 160 μm2 , much tigators used the generalized phase contrast method
smaller than the excitation field achievable with scan- (see the ‘Phase modulation’ section in the ‘Two-
ning methods (∼600 × 600 μm2 for galvanometer- photon excitation’ paragraph) for amplitude mod-
driven mirror-based systems; Oron et al., 2012). The ulation in combination with DH for phase mod-
field size limitation is the primary reason that limits ulation, demonstrating efficient generation of near
the use of DH to photoactivation applications, and it diffraction-limited spots. Although not shown, the
is not preferred for imaging. projection of extended light patterns should be also
Speckle-induced intensity inhomogeneities within feasible.
the desired light pattern pose the second primary
disadvantage for DH. Speckles arise due to interfer-
ence of light caused by diffraction through adjacent Two-photon excitation for optogenetics
finite-aperture pixels and the fact that the iterative Although the above-described methods enabled sig-
algorithm modulates only the phase of the laser beam nificant gains in temporal resolution and lateral spa-
(Palima and Glückstad, 2008). Hence, when calculat- tial patterning for actuation of optogenetic tools with
ing the final solution for the electromagnetic field in visible wavelengths, limitations inherent to 1PE sys-
each iteration, the amplitude is ignored and replaced tems, that is scattering and lack of axial optical sec-
by the Gaussian beam incident onto the SLM (Oron tioning, prevent single cell targeting and limit exci-
et al., 2012), leading to an approximate solution in tation to shallow depths. Two-photon (or in general
which holographic patterns exhibit a speckled inten- multi-photon) fluorescence excitation remedies both
sity profile (Figure 2). These fluctuations can reach limitations (Denk et al., 1990). For 2PE, the proba-
20% in 1PE and 50% in 2PE. Ideally, one should bility of observing quasi-simultaneous absorption of
modulate both the phase and the amplitude of the two photons by a molecule falls off with the intensity
waveform in order to achieve a perfect reconstruc- squared, conferring optical sectioning. In addition,
tion of the desired pattern (Jesacher et al., 2008), the longer, near-infrared wavelengths used for 2PE
but, as previously mentioned, amplitude modulation are less scattered than visible wavelengths, enabling
results in loss of light power. Efforts to eliminate excitation hundreds of microns deep inside the scat-
holographic speckle include time averaging either tering tissue (Helmchen and Denk, 2005).
different speckle patterns with a rotating diffuser In optogenetics, 2PE is widely used for imaging
(Papagiakoumou et al., 2008) or multiple shifted ver- genetically encoded functional reporters (see ‘Exci-
sions of a single hologram (Golan and Shoham, 2009; tation by scanning approaches’). However, 2PE of
Matar et al., 2011). In the first case, however, there is a optogenetic actuators is not as trivial as for 1PE, even
significant loss of light power, and in the second case, with the large 2P absorption cross-sections demon-
the temporal resolution is limited by the refresh rate strated for probes such as ChR2 (260 Goeppert-
of LC-SLMs (60–200 Hz) for shifting between dif- Mayer at 920 nm empirically estimated by Rickgauer
ferent holograms. Ferroelectric liquid crystal SLMs, and Tank, 2009). Although some evidence incrimi-
capable of more than 1 kHz refresh rates, were used nates poor infrared absorption by the all-transretinal,
in the above-mentioned references to improve the the light-sensitive element of ChR2, rather than the
temporal resolution of shifting the CGH. However, features of the protein itself (Kasparov and Herlitze,
these devices can perform only binary [0, π] phase 2013), there are also other properties of these chan-
modulation, resulting in low diffraction efficiency nels hindering their use in high spatial resolution
(∼40%), consequently reducing the excitation field 2PE schemes. First, most opsins, including ChR2,

E. Papagiakoumou
have a small, single-channel conductance (∼80 fS; ChR2-expressing neurons. The obvious disadvantage
Feldbauer et al., 2009) much lower than the one of of this approach is limited temporal resolution, which
endogenous channels. Thus, the photocurrent gen- in principle could be increased by further decreasing
erated by the few channels occupying the femtoliter the NA, so that the spot focal volume simultaneously
volume of a 2P diffraction-limited spot is too small opens more channels. However, NA reduction also
to induce sufficient membrane depolarisation for the degrades axial resolution, defined as the 1/e width of
generation of APs. This limitation is especially salient the squared intensity profile in z, ωz , described for
for channels such as ChR2, for which low conduc- 2PE by Zipfel et al. (2003).
tance cannot be compensated by increasing the in- A similar approach, improving the temporal re-
tensity and duration of the illumination spot, due solution performance of scanning methods with low-
to the long lifetime of the excited conducting state. NA excitation beams, was published about a year later
Prolonged illumination with high light intensity re- (Andrasfalvy et al., 2010). Andrasfalvy et al. (2010)
sults in substantial depletion of the ground state, recuperated the lost optical confinement for the low-
with further increase yielding a saturated response NA excitation spots by temporally focusing the ex-
(Rickgauer and Tank, 2009). Furthermore, at high citation beams. Temporal focusing (TF), a technique
light intensities, out-of-focus excitation can induce originally introduced for wide-field 2P microscopy
unwanted photocurrents in other cellular compart- (Oron et al., 2005; Zhu et al., 2005), permits excep-
ments or neighbouring cells that can easily exceed tional depth confinement of light for large excitation
currents excited in the small focal volume, thus off- areas. As mentioned previously, the 2PE fluorescence
setting any gains in spatial specificity. Out-of-focus signal, S2PE , is proportional to the square of the peak
2
excitation can be avoided by scanning the laser spot intensity: S 2PE ∝ I 2 = τEA , where E is the pulse
over the membrane of a target neuron at a low in- energy, τ is the pulse duration and A is the area of
tensity still sufficient to excite most available current the excitation beam. Typical spatial focusing achieves
in-focus. However, the fast (∼millisecond) deacti- optical sectioning by tightly focusing the laser beam
vation time of ChR2 prevents the adequate temporal to a very small spot, such that if spot size increases,
summation of currents generated through serial scan- optical sectioning degrades with decreasing peak in-
ning. 2P activation of ChR2 was reported for the first tensity. In contrast, TF operates through pulse dura-
time in 2008 (Mohanty et al., 2008) by recording tion modification, that is the pulse is compressed as
neuronal responses in hippocampal brain slices with it propagates through the sample, reaches its shortest
2P calcium imaging. However, the small fluorescence duration at the focal plane and stretches again as it
reported by the calcium dye indicated small ampli- propagates beyond. Thus, lost optical sectioning due
tude, sub-threshold activation. Hence, the methods to the increase of the excitation spot size is regained
for efficient yet spatially confined 2PE of ChR2 and through increased pulse duration outside of the focal
other actuators needed to be reconsidered. volume.
TF is realised experimentally with a diffrac-
Excitation by scanning approaches tion grating imaged onto the sample via a tele-
The first article reporting rigorous 2PE of ChR2 in scope formed by a regular lens and the objective
cultured neurons used a laser-scanning approach with (Figure 3A). Briefly, the grating disperses frequencies
galvanometer-driven mirrors (Rickgauer and Tank, comprising the spectrum of the femtosecond pulse
2009). To improve excitation efficiency, laser spot (typically 100 fs pulses for 2PE microscopy; band-
size was increased by decreasing the effective nu- width ∼10 nm) towards different angular directions.
merical aperture (NA) of the objective with an iris Dispersed spectral components are then collimated
placed near the back focal plane, and the spot was by the first lens and recombined at the focal plane of
scanned over a spiral trajectory to cover the whole the objective. Rays corresponding to different spec-
cell surface. With an effective NA of 0.3 (instead of tral components propagate through the imaging sys-
the objective’s nominal NA 0.8) and overall scan- tem following different optical paths, eventually re-
ning time of approximately 30 ms, Rickgauer and combining with exactly the same phase at the objec-
Tank demonstrated the first 2PE AP generation in tive focal plane, recovering the original short pulse

Figure 3 Temporal focusing configuration

duration. In all other planes, the rays arrive with
a relative phase offset, resulting in illumination of
(A) Illustration of the experimental configuration for TF. The
this plane for a longer duration (spectral dispersion;
diffraction grating is imaged onto the sample via a 4f tele-
scope formed by a lens and the microscope objective. The
Figure 3B). As the pulse is broadened out of the focal
laser beam impinges on the grating at an angle such that the
plane, the probability of 2PE is low, thus restricting
central frequency of the pulse is diffracted perpendicular to
excitation to a small volume around the focal plane,
the grating, along the optical axis of the telescope. For large in which the pulse duration is short enough to evoke
illumination areas on the sample and consequently on the 2P absorption. It has been shown, both for multipho-
grating as well, this configuration helps to prevent illumination ton absorption (Oron et al., 2005) and multiphoton
of the sample planes tilted relative to the microscope axis. generation processes (Oron and Silberberg, 2005),
(B) Plot of the pulse duration in a TF setup for different sample that the axial resolution achieved for wide-field tem-
planes spaced up to ±10 μm from the focal plane of the ob- porally focused excitation is equivalent to the one
jective. The pulse duration was calculated after simulating TF of multiphoton line-scanning microscopy. For a de-
of 140-fs laser pulses centred at 780 nm (λ = 12 nm) imping- tailed description of the phenomenon in frequency or
ing on an 830 l/mm groove density grating and a telescope in time domain, the reader can refer to the review by
comprised of a 500-mm lens and a 60x, 0.9 NA objective Oron et al. (2012).
(see Papagiakoumou et al., 2008 for simulation details). Pulse Andrasfalvy et al. (2010) randomly scanned a tem-
width is plotted as a function of z-position for a flat-phase porally focused laser spot of 5 μm in diameter over the
beam (e.g. a Gaussian beam; red dots) and a holographic somata and processes of ChR2-expressing hippocam-
beam (black dots). Pulse broadening to a few picoseconds pal neurons in slices at saturation powers for ChR2
at 10 μm from the focal plane corresponds to an axial reso- (460 mW at the sample plane, i.e. >23 mW/μm2 ),
lution of few microns (∼3 μm for the Gaussian beam and 4 eliciting APs with 1–6 ms temporal resolution at
to 5 μm for the holographic beams for the setup described depths up to 150 μm. This approach was used
in Papagiakoumou et al., 2008). As evidenced by the calcu-
to study the cellular and circuit level mechanisms
lation, holographic phase distorts recombination of different
of hippocampal theta oscillations (Losonczy et al.,
spectral components, resulting in larger pulse duration at the
2010). However, the high excitation power rela-
focal plane than for a flat-phase beam. This effect increases
tive to Rickgauer and Tank, employed to trigger
the excitation volume by 1 to 2 μm for a holographic beam in
APs with high temporal resolution, cancelled exci-
comparison to a Gaussian one in a TF setup.
tation confinement gained with TF. Lateral and ax-
ial spatial resolution for this approach was deduced
by displacing a 5-μm size spot across a thin den-
drite and through different axial planes, respectively
(Figure 4A), evoking significant depolarisation more
than 10 μm away from the dendrite in the lateral
direction and more than 40 μm out of plane in the
axial direction, despite theoretical estimates for ax-
ial resolution of the spot itself of the order of the
micron.
Development of new opsins with slower clos-
ing times, custom expression and spectral proper-
ties adapted for typical raster scanning rates has
recently facilitated scanning approaches, without im-
plementation of TF or elaborate scanning trajectories.
Specifically, these advances enabled 2PE of C1V1
variants, red-shifted chimeric opsins composed of
channelrhodopsin-1 (ChR1) and Volvox carteri ChR1
(VChR1; Yizhar et al., 2011), by raster scanning a
1-μm size spot within more than 70 ms, total pho-
toactivation time, at 1064 nm (Packer et al., 2012)

E. Papagiakoumou
Figure 4 See Legend on next page
and approximately 1.3-μm size spots at 1040 nm in both cases necessitated excessively high excita-
(Prakash et al., 2012) with much shorter photoactivation power densities to efficiently excite neurons
tion time (2.3–9.5 ms). The large discrepancy in the (typical values: 38 mW/μm2 , Packer et al., 2012;
total scan time needed to evoke APs remains contra- 15–60 mW/μm2 depending on the objective used,
dictory given that both investigations used the same Prakash et al., 2012). The high excitation power de-
approach at similar excitation powers. Prakash et al. graded the realised axial resolution (Figures 4B and
demonstrated 2PE of C1V1 at depths up to 250 μm 4C), theoretically reaching micrometer precision for
in vivo, as well as 2PE of ChR bistable variants (ChR2- such a small excitation spot and increases the risk of
C128T or C128A) and the microbial rhodopsin pro- photodamage.
tein archaerhodopsin eArch3.0, an engineered proton Excitation with 2P diffraction-limited spots by
pump (Mattis et al., 2012) with kinetics slower than using mirror-based scanning systems has also been
the chloride pump eNpHR3.0, in cultures and brain applied to functional imaging of genetically en-
slices. The approximately 1.0-μm spot size employed coded calcium indicators, such as the widely known

Figure 4 Axial resolution of 2PE approaches in optogenetics

(A) Spatial resolution of ChR2 activation with temporally focused 5-μm-diameter spot scanned over a thin (1–2 μm) apical
dendrite of a CA1 pyramidal cell, adapted from Andrasfalvy et al., 2010. The image on the top left is a maximum intensity
projection of a z-series through a pyramidal cell loaded with Alexa594, showing the lateral locations of the activation spots.
The lateral and axial localization of responses induced by 488-nm light (blue curves) and temporally focused excitation (red or
black curves) are shown on the top right and bottom panels. The axial responses were measured for two different power values,
∼130 mW (red) and ∼260 mW (black; dashed lines indicate Gaussian fit). Individual responses at the focal plane of the dendrite
(colour-coded respectively) are shown on the inset. Note that the axial resolution improves with lower power. Nevertheless, the
photoactivation volume of the 5-μm spot extends in ∼35 μm and ∼15 μm FWHM in the axial and lateral dimension respectively.
(B) Axial (left) and lateral (right) resolution of scanning 2PE of C1V1 photocurrents from pyramidal cells in prefrontal cortex slices,
adapted from Prakash et al., 2012. Top panels show schematics of the areas where a ∼1.5-μm-diameter spot (5.7 μm in the
axial direction) was scanned relative to the center of the cell body. Blue triangles indicate a pyramidal neuron, and red boxes
indicate a typical ROI (10 × 10 μm to 15 × 15 μm). Graphs underneath show 2P photocurrent as a function of axial (left) and
lateral (right; black curve for ‘x’ direction and green curve for ‘y’) distance of the ROI from the center of a pyramidal cell. The
excitation wavelength was λ = 1040 nm, the laser intensity was 20 mW at the sample plane and the Gaussian fit gives a FWHM of
29.5 μm in the axial and 15 to 20 μm in the lateral dimension that roughly matches the size of the cell soma. (C) Spatial resolution
of AP probability acquired by raster scanning of ∼1-μm-diameter spot over a C1V1-expressing pyramidal cell in somatosensory
cortex, adapted from Packer et al., 2012. Lateral resolution of the photostimulation in relation to distance from soma is shown at
the top (λ = 1040 nm, 20 mW on sample). Although APs were produced in individual neurons with a lateral resolution of 6.5 μm
(13 μm FWHM), photocurrents are seen as far as 50 μm away from the soma (grey dashed line). The lateral resolution for spiking
is significantly higher due to the AP threshold. The axial resolution for AP probability generation is shown at the bottom. APs
were produced in individual neurons with axial resolution of 29.5 μm. The corresponding photocurrent curve vs. axial distance
from the soma is not shown. Note that the spatial resolution in this case is worse than in the data displayed in B, although the
excitation parameters are the same. This can possibly be attributed to the total scan time, which in Packer et al. is 73 ms, while in
Prakash et al. is ∼5 ms. (D) Spatial resolution of the TF-GPC method, adapted from Papagiakoumou et al., 2010. To demonstrate
lateral precision, we compared photocurrents excited by TF-GPC patterns of 4-μm axial resolution illuminating either the whole
soma (shape) or the surrounding area (antishape) of a layer V pyramidal neuron in somatosensory cortical brain slices. The top
right image shows wide-field fluorescence of a ChR2-YFP positive neuron filled with Alexa594 and superimposed excitation
patterns (red) with shaped and anti-shaped profiles with corresponding photocurrents (black for shape, grey for antishape;
10 ms illumination duration, 0.24 mW/μm2 at the sample plane). The bottom traces show the integrated photocurrent (area
above the inward current in pico-coulombs) evoked by a 10-μm-diameter spot centred on the cell body when displaced along
the z axis (0.30 mW/μm2 ). A FWHM of 27.5 μm can be inferred for the axial precision of photocurrents induction. The method’s
precision was also measured by the probability of generating APs on a neuronal dendrite. The left image shows fluorescence of a
ChR2-positive neuron filled with Alexa594 and the superimposed shaped excitation profile covering the apical dendrite (red), as
well as the photo-depolarisations evoked by the excitation shape at different z-axis positions (bottom panel; 10 ms illumination
duration, 0.30 mW/μm2 ). APs were generated only on a narrow band of 5 μm around the focal plane. Scale bars indicate 20 μm
and the excitation wavelength was λ = 920 nm.
GCaMP family (Mao et al., 2008; Tian et al., 2009; ventional laser scanning techniques, have hindered
Dombeck et al., 2010; Akerboom et al., 2012; Chen 2PE of genetically encoded voltage-sensitive proteins
et al., 2012), providing enhanced spatial resolution (Knöpfel, 2012).
compared with 1P scanning excitation. For volt-
age indicators, although some progress has been Parallel excitation
made with line (Fisher et al., 2008; Kuhn et al., Although parallel, patterned excitation methods
2008) and point scanning (Acker et al., 2011) and show immense promise to address the challenges
recent engineering of dyes more efficient for 2PE detailed above for both 2PE imaging and photoacti-
(Acker et al., 2011; Yan et al., 2012), photon- vation, these techniques are currently in their infancy.
counting limitations due to chromophore satura- The delay is driven by two primary causes, the first
tion during short dwell times, characteristic of con- emerging from the high complexity of the optical

E. Papagiakoumou
setup compared with mirror scanning systems, and scanning near diffraction limited spots in 2Ds while
the second from the reduced available power density. holographically changing their axial position.
As laser power is distributed across the excitation A similar approach recently has revealed benefits of
pattern, thus simultaneously illuminating larger ar- DH for optogenetics by simultaneously evoking APs
eas without loss of temporal resolution, the power from two C1V1-expressing neurons lying in planes
density is lower than if the laser was focused to a separated axially by 20 μm. The neurons were actu-
diffraction-limited spot. However, decreased power ated by raster scanning of 1-μm holographic spots in
density is easily compensated, and in fact surpassed, 2D, one in each plane (power: 30 mW per spot; Packer
as parallel excitation optimises the recruitment speed et al., 2012). The same experiments would have been
of optogenetic reporter molecules, thus decreasing impossible with objective scanning approaches com-
the excitation power density necessary to photoacti- monly utilised for 3D 2P calcium imaging (Göbel
vate, detailed as follows. et al., 2006). Indeed, objective scanning is too slow for
simultaneous stimulation of cells in separate planes,
Amplitude modulation even those expressing C1V1 or other optogenetic ac-
The low efficiency of amplitude modulation methods tuators with kinectics prolonged specifically for scan-
due to light rejection has thus far prevented their use ning systems.
with 2PE. Especially for DMDs, the maximisation Two-photon DH offers an alternative to single-spot
of diffraction efficiency requires their implementa- scanning stimulation of optogenetic actuators with
tion at a specific angle, which critically depends on extended arbitrarily shaped patterns. Although this
the illumination wavelength. Thus, although experi- illumination approach has yet to be demonstrated,
ments in two different wavelengths in 1PE have been preliminary experiments in our laboratory confirm its
successful (Wang et al., 2007b; Sakai et al., 2013), feasibility. Moreover, the efficiency of using extended
the use of DMDs in combination with broadband or patterns to activate ChR2 in 2P has been validated
tunable light sources is not straightforward. with a different phase modulation technique, gener-
alised phase contrast (GPC; Papagiakoumou et al.,
Phase modulation 2010).
Thus far, DH has been implemented in 2PE both for GPC (Glückstad, 1996) belongs to the category
the generation of multiple near diffraction-limited of the interferometric phase visualisation techniques
spots (Nikolenko et al., 2008; Daria et al., 2009) and for which the output image is obtained by the in-
extended excitation patterns (Papagiakoumou et al., terference between a signal and a reference wave,
2008; Dal Maschio et al., 2010). As large excita- travelling along the same optical axis. The most well-
tion areas suffer the disadvantage of degraded axial known example of interferometric phase visualisation
confinement, in 2008, we ameliorated the axial re- is Zernike’s phase contrast method (Zernike, 1955),
solution by combining DH with TF, although the which represented a breakthrough for medicine and
rapidly varying phase of the holographic beam wave- biology by rendering transparent organisms clearly
front slightly distorts colour recombination at the discernible under the microscope through the visu-
focal plane of the objective in comparison to smooth alisation of small phase perturbations with a Fourier
wavefronts, such as those of Gaussian beams (Fig- plane phase-shifting filter. The GPC method is an ex-
ure 3B). With this technique, we generated scan-less, tension of Zernike’s phase-contrast into the domain
depth-resolved excitation in arbitrary illumination of full range [0, 2π] of phase variations (Glückstad
configurations, independently of the excited area size and Mogensen, 2001). Briefly, a desired target inten-
(Papagiakoumou et al., 2008, 2009). Indeed, DH has sity map is converted into a binary [0, π] phase map
already proven useful for 2P uncaging applications that modifies the input beam wavefront via the LC-
(Nikolenko et al., 2008; Dal Maschio et al., 2010; SLM. The beam modulated by the LC-SLM is then fo-
Go et al., 2012) and furthermore shows potential cused on a patterned phase contrast filter (PCF) plate
for fast imaging applications (Papagiakoumou et al., imposing appropriate phase retardation between the
2009). Recently, a phase-only SLM was coupled to a on-axis focused component (non-diffracted light) and
standard 2P scanning microscope for inertia-free 3D the higher order Fourier components (Figure 5A).
morphological imaging (Dal Maschio et al., 2011), Lastly, the interference between these two beams

Figure 5 Generalised phase contrast

image projection is fast and direct, transforming
a pure phase modulation to an intensity pattern.
(A) Schematic layout of the 4f configuration for GPC. A bi-
Hence, GPC generates patterns free of speckle (Fig-
nary phase map of the desired excitation pattern is sent on
the SLM. A phase contrast filter (PCF) at the Fourier plane
ure 5B), ‘ghost’ images and zero order, as the zero
of the first lens introduces half-wave (π) phase retardance
order interferes directly with the signal wave. In ad-
between the component not diffracted by the SLM and the
dition, GPC is based on binary phase holograms;
high-frequency component (signal). Interference of the two thus, ferroelectric liquid crystal matrices can be used
light components at the output plane gives an intensity pat- to further improve the temporal resolution of the
tern according to the phase modulation at the input plane technique.
(constructive interference at the areas addressed with phase Despite the speed and homogeneity of GPC, DH
π and destructive interference at the areas addressed with remains more flexible in terms of the variety of ex-
phase 0). (B) Examples of GPC-generated patterns with 2PE. citation configurations that can be generated. GPC,
The pattern at the top mimics the dendrite of a Purkinje cell, for example, cannot generate spots in different ax-
while the one at the bottom shows simultaneous multi-spot ial planes. Moreover, conditions for maximum in-
excitation. 2PE is visualised on thin rhodamine-6G films. GPC terferometric contrast require a factor of 1:4 ratio
patterns, unlike DH, are free of speckle. (C) Schematic lay- between the number of SLM pixels modulated with
out of a setup for combining GPC with TF. Notably, the lenses phase at π (corresponding to bright regions in the
used to form the 4f system for GPC do not need to have equal final intensity map) and the total number of pix-
focal lengths. In this way, the excitation field can be adjusted els (Palima and Glückstad, 2008; Papagiakoumou
to the needs of the experiment. et al., 2010), thus restricting the size of the excita-
tion pattern area. GPC patterns also lack axial spa-
tial confinement, as out-of-focus planes contain an
interference pattern inherent to the interferometric
method (for a comparison between the axial propa-
gation of 2P holographic and GPC patterns see Oron
et al., 2012). Finally, the GPC excitation field can
be adjusted by selecting appropriate telescope lenses
for imaging the output plane to the sample plane,
as in DH. That is, for a circular aperture in front
of the SLM of √ radius RC , the excitation field, R,
f
will be R = X2 + Y2 = RC ff21 obj f
, where f,
f1 , f2 , fobj are the focal lengths of the lenses used
(Figure 5C). However, for 2P photoactivation, power
limitations practically restrict the excitation field to
tens of microns for a 40×, 0.8 NA objective.
We implemented GPC for 2PE of ChR2-expressing
layer V pyramidal neurons (Papagiakoumou et al.,
2010). In order to mitigate limitations imposed by
GPC’s fill factor requirement, a peripheral ring was
sent to the SLM, its weight adjusted for each exci-
tation pattern to maintain the 1:4 ratio. Light sent
to the ring was blocked near the GPC output plane.
generates the desired target intensity at the focal Although the maximum pattern area remained ul-
plane of a second lens (the output plane). timately limited, we were able to excite up to five
GPC surpasses DH both in terms of speed and neurons simultaneously within a circular 60 μm-
output pattern homogeneity. As for GPC the phase diameter excitation field. In addition, the issue of
hologram addressed to the SLM is a simple phase the axial confinement was addressed by combining
map transformation of the desired output inten- GPC with TF, imaging the GPC output plane on the
sity, there is no iterative calculation and thus the diffraction grating used for TF (Figure 5C).

E. Papagiakoumou
Temporally focused GPC (TF-GPC) patterns tation beam resulting in aberrated spots or, in the
mapped over the somatic surface reliably excited case of large excitation beams, interference of ballis-
neurons in mouse cortical slices, demonstrating for tic photons from adjacent illumination areas trans-
the first time the feasibility of simultaneously acti- form the intensity profile into a speckle pattern. The
vating multiple neurons or neuronal compartments latter effect can also decrease the axial resolution,
with patterns mapped onto separate dendrites, at especially for temporally focused beams (Dana and
low excitation power densities (0.3–0.6 mW/μm2 ). Shoham, 2011, 2012a). Adaptive optics can improve
The decreased power density enabled better preser- propagation of diffraction-limited spots through scat-
vation of the lateral and axial resolution than the tering samples (Rueckel et al., 2006; Vellekoop and
scanning methods that direct the power used to a Mosk, 2007; Débarre et al., 2009; Katz et al., 2011),
near-diffraction-limited spot (Figure 4D). Moreover, which has been especially useful for morphological
the broad focal volume of large-surface spots, and the imaging and other highly aberration-sensitive appli-
increased out-of-focus pulse duration by using TF cations. However, the limits on temporal resolution
reduces the chances to cause photodamage, even if imposed by lengthy computation times have thus far
we have to increase the total average power that is outweighed potential benefits of utilising adaptive
sent to the sample (Hopt and Neher, 2001) with cells optics for optogenetics applications.
sustaining to long-duration illumination protocols Interestingly, recent experimental and theoretical
(Papagiakoumou et al., 2010). studies on the propagation of temporally focused large
Gaussian beams (Dana and Shoham, 2011, 2012a)
and extended GPC-generated patterns through scat-
Perspectives and outlook tering media (Papagiakoumou et al., 2013) revealed
Optogenetics has been introduced in the field of neu- that even fine details of the excitation shape are
roscience during the last decade, and it has been well preserved deep in scattering tissue, maintaining
quickly adopted by scientists. Efforts to expand opto- good axial confinement up to one to two scattering
genetic tool capabilities have spurred innovations in lengths. The robustness to scattering of temporally
genomic tool discovery, molecular engineering, opsin focused patterns can be attributed to the spectral ‘self-
targeting and optical-device development. Although healing’: each spectral component dispersed by the
numerous optogenetic studies have provided impor- grating acquires its own speckle pattern, as it prop-
tant insights into neural function and dysfunction, agates through slightly different optical path than
complex interrogation and manipulation of neural the rest of the spectral frequencies in the scattering
circuits demands more flexible, controllable and spa- medium. These uncorrelated speckle patterns inter-
tiotemporally precise illumination methods, imple- fere, smoothing out the intensity profile (Oron et al.,
mentable in 2PE. Despite the progress in optical de- 2012; Papagiakoumou et al., 2013). Thus, tempo-
velopments for optogenetics over the past five years, rally focused patterns propose a powerful method to
scientists continue to improve their performances by overcome scattering without introducing extracom-
encountering the limitations of the techniques. putation time or power losses.
An important issue, for instance, when working An interesting application of patterned illumina-
in depth, both in vitro and in vivo, is scattering in- tion can be for in vivo activation in freely moving
duced by the tissue. 2PE has become the most widely animals with flexible fibre bundle image-transferring
implemented method thus far to overcome scatter- endoscopes (Bozinovic et al., 2008). Fibre transmis-
ing, as the scattered part of the beam, weak and sion of intensity patterns provides a more practical
temporally stretched outside the focus, contributes solution than transmitting the phase map because
little to the 2PE signal. In addition, the longer wave- multi-mode fibres randomise the phase, amplitude
length used decreases scattering (Oheim et al., 2001). and polarisation of the transmitted light. However,
Even so, the signal decreases exponentially with depth developers recently succeeded in transmitting CGHs
(S 2PE ∝ e −2z/ S , where s is the mean free path be- through multimode optical fibres by implementing
tween two scattering events, or else scattering length; a direct search algorithm to modulate the phase,
Helmchen and Denk, 2005). Moreover, scattering concentrating the output light power to one or
strongly distorts the optical wavefront of the exci- more spots (Di Leonardo and Bianchi, 2011). CGH

transmission with improved efficiency was also domain, that is, on the basis of the angular dispersion
achieved with an optimisation algorithm based on de- of the different spectral components, we can axially
composition of the initial laser field at the SLM plane shift the TF plane by adding a quadratic spectral
into a series of orthogonal modes each correspond- phase, analogous to the ‘lens effect’ for DH, in which
ing to a different square region on the SLM (Čižmár the focal plane is shifted via a quadratic spatial phase.
and Dholakia, 2011, 2012). Although interesting, A quadratic spectral phase involves introduction of
the phase map transfer approaches are more compli- a temporal delay, or ‘chirp’, between the frequencies
cated to implement than direct image pattern transfer comprising the spectral bandwidth of the femtosec-
and, in addition, highly sensitive to fibre movements, ond pulse before dispersing the spectral frequencies
hindering adaption for in vivo applications. with the grating. Pre-chirping the frequencies re-
Simultaneous 3D photoactivation and imaging of sults in recombination of the spectral components at
neuronal activity in separate planes poses another a plane offset from the nominal objective focal plane
frontier at the nexus of wavefront shaping and op- (Durst et al., 2006). This phenomenon has been ex-
togenetics. Decoupling the imaging and activation perimentally realised by chirping the beam with a
planes would grant access to circuits and structures pair of prisms (Durst et al., 2006, 2008), remotely
elaborated in three dimensions, such as dendritic ar- with a pair of AODs (Du et al., 2009), a folded grat-
bors, enabling optical activation and monitoring of ing pair and a piezo-bimorph mirror (Straub et al.,
responses simultaneously in separate planes. This is 2011) or by replacing the standard TF diffraction
not a trivial task, however, as typically both imag- grating with a dual-prism grating (DPGrism) com-
ing and photoactivation are performed through the posed of two prisms and a transmission diffraction
same objective lens. DH, once more, can provide an grating (Dana and Shoham, 2012b). Lastly, we can
elegant solution. Phase holograms can be modified to achieve more flexible control of the laser pulse shape
incorporate a ‘lens effect’, acting in practice as if we by modulating the spectral phase with an SLM at the
had placed an additional lens at the back focal plane Fourier plane of the TF telescope (Suchowski et al.,
of the objective, with which we can rapidly scan the 2006).
z-position of an excitation pattern without physi- Finally, in parallel with the development of opto-
cally modifying the optical components (Zahid et al., genetic tools, another class of genetically encoded
2010; Dal Maschio et al., 2011; Yang et al., 2011; Go molecular sensitisers has been presented that en-
et al., 2012; Packer et al., 2012). Hence, the excita- ables modulation of neuronal activity in response to
tion patterned can be holographically scanned in z to temperature changes, so-called ‘thermogenetic’ tools
precisely compensate the objective or stage scanning (Bernstein et al., 2011). Thermogenetic approaches,
motion used for 3D imaging. Alternatively, DH can similar to optogenetics, can be used for activation
be combined with remote focusing (Botcherby et al., or for inhibition of neuronal activity. As temper-
2007), in which fluorescence imaging is performed ature affects all physiological processes, its use as
by a second microscope objective symmetrical to the a tool to control neuronal activity poses particular
principal objective, recreating an aberration-free 3D challenges, especially for applications in mammalian
image of the sample. In the combined system, 3D models (Bernstein et al., 2011), which are not going
holographic photoactivation utilises the principal ob- to be discussed here. Thermogenetics show advan-
jective, although the plane imaged onto the camera tages, such as the possibility to stimulate simultane-
is selected independently, with a mirror moving in ously and non-invasively different regions of the brain
the remote space (Anselmi et al., 2011). This sys- at great depths, and limitations such as the lack of the
tem combining DH with remote focusing was used millisecond temporal resolution achieved by optoge-
for 1P uncaging, but remote focusing alone has been netics. Thermogenetics also lack excitation spatial
implemented for 2P imaging by coupling an infrared precision, as thermal stimulation, thus far, is done by
laser beam to the remote unit, scanning laterally with ambient changes of temperature. This stimulation
a pair of galvanometer-driven mirrors and axially by method has restricted thermogenetic application in
displacing the remote mirror (Botcherby et al., 2012). Drosophila. However, development of methods that
TF offers an alternative method to axially scan the could extend their application to other animals (e.g.
laser beam remotely. Considering TF in the frequency localised heating by using ultrasounds or lasers) could

E. Papagiakoumou
make them a complementary tool to optogenetic ap- Arrenberg, A.B., Stainier, D.Y.R., Baier, H. and Huisken, J. (2010)
proaches in order to enhance the characterisation and Optogenetic control of cardiac function. Science 330, 971–974
Arrizón, V. and Testorf, M. (1997) Efficiency limit of spatially quantized
manipulation of mammalian circuit properties and Fourier array illuminators. Opt. Lett. 22, 197–199
functions. Ashkin, A. and Dziedzic, J.M. (1987) Optical trapping and
manipulation of viruses and bacteria. Science 235, 1517
In this review, the illumination methods used so Ashkin, A., Dziedzic, J.M., Bjorkholm, J.E. and Chu, S. (1986)
far in optogenetic applications were examined, dis- Observation of a single-beam gradient force optical trap for
cussing the advantages and limitations of each, in or- dielectric particles. Opt. Lett. 11, 288–290
Axelrod, D. (1981) Cell–substrate contacts illuminated by total
der to help readers to choose strategies well adapted to internal reflection fluorescence. J. Cell Biol. 89, 141–145
their specific experimental parameters. As evidenced Banghart, M., Borges, K., Isacoff, E., Trauner, D. and Kramer, R.H.
throughout, the method selected from among those (2004) Light-activated ion channels for remote control of neuronal
firing. Nat. Neurosci. 7, 1381–1386
presented should be well adapted to the reporter and Bernstein, J., Garrity, P. and Boyden, E. (2011) Optogenetics and
desired scale (sub-cellular, cellular, networks, in vivo). thermogenetics: technologies for controlling the activity of targeted
cells within intact neural circuits. Curr. Opin. Neurobiol. 22, 61–71
Botcherby, E.J., Juskaitis, R., Booth, M.J. and Wilson, T. (2007)
Acknowledgements Aberration-free optical refocusing in high numerical aperture
microscopy. Opt. Lett. 32, 2007–2009
The author thanks A.J. Foust, A. Bègue, E. Ronzitti Botcherby, E.J., Smith, C.W., Kohl, M.M., Débarre, D., Booth, M.J.,
and V. Emiliani for useful comments on the Juskaitis, R., Paulsen, O. and Wilson, T. (2012) Aberration-free
manuscript and D. Oron for his contribution to three-dimensional multiphoton imaging of neuronal activity at kHz
rates. Proc. Natl. Acad. Sci. U. S. A. 109, 2919–2924
Figure 3. Bozinovic, N., Ventalon, C., Ford, T. and Mertz, J. (2008)
Fluorescence endomicroscopy with structured illumination. Opt.
Express. 16, 4603–4610
Conflict of interest statement Bruegmann, T., Malan, D., Hesse, M., Beiert, T., Fuegemann, C.J.,
Fleischmann, B.K. and Sasse, P. (2010) Optogenetic control of
The author has declared no conflict of interest. heart muscle in vitro and in vivo. Nat. Methods 7, 897–900
Bundschuh, S.T., Zhu, P., Schärer, Y.-P.Z. and Friedrich, R.W. (2012)
Dopaminergic modulation of mitral cells and odor responses in the
References zebrafish olfactory bulb. J. Neurosci. 32, 6830–6840
*Articles of special interest Busskamp, V., Duebel, J., Balya, D., Fradot, M., Viney, T.J., Siegert,
Acker, C.D., Yan, P. and Loew, L.M. (2011) Single-voxel recording of S., Groner, A.C., Cabuy, E., Forster, V., Seeliger, M., Biel, M.,
voltage transients in dendritic spines. Biophys. J. 101, L11–L13 Humphries, P., Paques, M., Mohand-Said, S., Trono, D., Deisseroth
Adamantidis, A.R., Zhang, F., Aravanis, A.M., Deisseroth, K. and De K., Sahel J.A., Picaud S. and Roska B. (2010) Genetic reactivation
Lecea, L. (2007) Neural substrates of awakening probed with of cone photoreceptors restores visual responses in retinitis
optogenetic control of hypocretin neurons. Nature 450, 420–424 pigmentosa. Science 329, 413–417
*Akemann, W., Mutoh, H., Perron, A., Rossier, J. and Knöpfel, T. Buzsáki, G. (2004) Large-scale recording of neuronal ensembles. Nat.
(2010) Imaging brain electric signals with genetically targeted Neurosci. 7, 446–451
voltage-sensitive fluorescent proteins. Nat. Methods 7, 643–649 Callaway, E.M. and Katz, L.C. (1993) Photostimulation using caged
Akerboom, J., Chen, T.-W., Wardill, T.J., Tian, L., Marvin, J.S., Mutlu, glutamate reveals functional circuitry in living brain slices. Proc.
S., Calderón, N.C., Esposti, F., Borghuis, B.G., Sun, X.R., Gordus, Natl. Acad. Sci. U. S. A. 90, 7661–7665
A., Orger, M.B., Portugues, R., Engert, F., Macklin, J.J., Filosa, A., Cardin, J.A., Carlén, M., Meletis, K., Knoblich, U., Zhang, F.,
Aggarwal, A., Kerr, R.A., Takagi, R., Kracun, S., Shigetomi, E., Deisseroth, K., Tsai, L.-H. and Moore, C.I. (2009) Driving
Khakh, B.S., Baier, H., Lagnado, L., Wang, S.S., Bargmann, C.I., fast-spiking cells induces gamma rhythm and controls sensory
Kimmel, B.E., Jayaraman, V., Svoboda, K., Kim, D.S., Schreiter, responses. Nature 459, 663–667
E.R. and Looger, L.L. (2012) Optimization of a GCaMP calcium Chanda, B., Blunck, R., Faria, L.C., Schweizer, F.E., Mody, I. and
indicator for neural activity imaging. J. Neurosci. 32, 13819–13840 Bezanilla, F. (2005) A hybrid approach to measuring electrical
*Andrasfalvy, B.K., Zemelman, B.V., Tang, J. and Vaziri, A. (2010) activity in genetically specified neurons. Nat. Neurosci. 8,
Two-photon single-cell optogenetic control of neuronal activity by 1619–1626
sculpted light. Proc. Natl. Acad. Sci. U. S. A. 107, 11981–11986 Chen, Q., Cichon, J., Wang, W., Qiu, L., Lee, S.-J.R., Campbell, N.R.,
Anikeeva, P., Andalman, A.S., Witten, I., Warden, M., Goshen, I., DeStefino, N., Goard, M.J., Fu, Z., Yasuda, R., Looger, L.L.,
Grosenick, L., Gunaydin, L.a, Frank, L.M. and Deisseroth, K. (2012) Arenkiel, B.R., Gan, W.B. and Feng, G. (2012) Imaging neural
Optetrode: a multichannel readout for optogenetic control in freely activity using Thy1-GCaMP transgenic mice. Neuron 76, 297–308
moving mice. Nat. Neurosci. 15, 163–170 Čižmár, T. and Dholakia, K. (2011) Shaping the light transmission
*Anselmi, F., Ventalon, C., Bègue, A., Ogden, D. and Emiliani, V. through a multimode optical fibre: complex transformation analysis
(2011) Three-dimensional imaging and photostimulation by and applications in biophotonics. Opt. Express. 19, 18871–18884
remote-focusing and holographic light patterning. Proc. Natl. Čižmár, T. and Dholakia, K. (2012) Exploiting multimode waveguides
Acad. Sci. U. S. A. 108, 19504–19509 for pure fibre-based imaging. Nat. Commun. 3, 1027
Aravanis, A.M., Wang, L.P., Zhang, F., Meltzer, L.A., Mogri, M.Z., Collot, M., Loukou, C., Yakovlev, A.V, Wilms, C.D., Li, D., Evrard, A.,
Schneider, M.B. and Deisseroth, K. (2007) An optical neural Zamaleeva, A.I., Bourdieu, L., Leger, J.-F., Ropert, N., Eilers, J.,
interface: in vivo control of rodent motor cortex with integrated Oheim, M., Feltz, A. and Mallet, J.M. (2012) Calcium Rubies: a
fiberoptic and optogenetic technology. J. Neural Eng. 4, family of red-emitting functionalizable indicators for two-photon
S143–S156 Ca2+ imaging. J. Am. Chem. Soc. 134, 14923–14931

Curtis, J.E., Koss, B.A. and Grier, D.G. (2002) Dynamic holographic Farber, I.C. and Grinvald, A. (1983) Identification of presynaptic
optical tweezers. Opt. Commun. 207, 169 neurons by laser photostimulation. Science 222,
Dal Maschio, M., De Stasi, A.M., Benfenati, F. and Fellin, T. (2011) 1025–1027
Three-dimensional in vivo scanning microscopy with inertia-free Feldbauer, K., Zimmermann, D., Pintschovius, V., Spitz, J., Bamann,
focus control. Opt. Lett. 36, 3503–3505 C. and Bamberg, E. (2009) Channelrhodopsin-2 is a leaky proton
*Dal Maschio, M., Difato, F., Beltramo, R., Blau, A., Benfenati, F. and pump. Proc. Natl. Acad. Sci. U. S. A. 106, 12317–12322
Fellin, T. (2010) Simultaneous two-photon imaging and photo- Fisher, J.A., Barchi, J.R., Welle, C.G., Kim, G.-H., Kosterin, P., Obaid,
stimulation with structured light illumination. Opt. Express. 18, A.L., Yodh, A.G., Contreras, D. and Salzberg, B.M. (2008)
18720–18731 Two-photon excitation of potentiometric probes enables optical
*Dana, H. and Shoham, S. (2011) Numerical evaluation of temporal recording of action potentials from mammalian nerve terminals in
focusing characteristics in transparent and scattering media. Opt. situ. J. Neurophysiol. 99, 1545–1553
Express. 19, 4937–4948 Fork, R. (1971) Laser stimulation of nerve cells in aplysia. Science 3,
Dana, H. and Shoham, S. (2012a) Numerical evaluation of temporal 907–908
focusing characteristics in transparent and scattering media: Fortin, D., Banghart, M. and Dunn, T. (2008) Photochemical control of
erratum. Opt. Express. 20, 28281 endogenous ion channels and cellular excitability. Nat. Methods 5,
Dana, H. and Shoham, S. (2012b) Remotely scanned multiphoton 331–338
temporal focusing by axial grism scanning. Opt. Lett. 37, Gasparini, S. and Magee, J.C. (2006) State-dependent dendritic
2913–2915 computation in hippocampal CA1 pyramidal neurons. J. Neurosci.
*Daria, V.R., Stricker, C., Bowman, R., Redman, S. and Bachor, H.A. 26, 2088–2100
(2009) Arbitrary multisite two-photon excitation in four dimensions. Gerchberg, R.W. and Saxton, W.O. (1972) A pratical algorithm for the
Appl. Phys. Lett. 95, 93701 determination of the phase from image and diffraction pictures.
Débarre, D., Botcherby, E. and Watanabe, T. (2009) Image-based Optik 35, 237–246
adaptive optics for two-photon microscopy. Opt. Lett. 34, Glückstad, J. (1996) Phase contrast image synthesis. Opt. Commun.
2495–2497 130, 225–230
Degenaar, P., Grossman, N., Memon, M.A., Burrone, J., Dawson, M., Glückstad, J. and Mogensen, P.C. (2001) Optimal phase contrast in
Drakakis, E., Neil, M. and Nikolic, K. (2009) Optobionic vision – a common-path interferometry. Appl. Optics 40, 268–282
new genetically enhanced light on retinal prosthesis. J. Neural Eng. Go, M.A., Ng, P.-F., Bachor, H.a and Daria, V.R. (2011) Optimal
6, 035007 complex field holographic projection. Opt. Lett. 36, 3073–3075
Deisseroth, K., Feng, G., Majewska, A.K., Miesenböck, G., Ting, A. Go, M.A., Stricker, C., Redman, S., Bachor, H.-A. and Daria, V.R.
and Schnitzer, M.J. (2006) Next-generation optical technologies for (2012) Simultaneous multi-site two-photon photostimulation in
illuminating genetically targeted brain circuits. J. Neurosci. 26, three dimensions. J. Biophotonics 5, 745–753
10380–10386 Göbel, W., Kampa, B.M. and Helmchen, F. (2006) Imaging cellular
Denk, W., Strickler, J.H. and Webb, W.W. (1990) Two-photon laser network dynamics in three dimensions using fast 3D laser
scanning fluorescence microscopy. Science 248, 73–76 scanning. Nat. Methods 4, 73–79
Depuy, S.D., Kanbar, R., Coates, M.B., Stornetta, R.L. and Guyenet, Golan, L., Reutsky, I., Farah, N. and Shoham, S. (2009) Design and
P.G. (2011) Control of breathing by raphe obscurus serotonergic characteristics of holographic neural photo-stimulation systems.
neurons in mice. J. Neurosci. 31, 1981–1990 J. Neural Eng. 6, 66004
Dhawale, A.K., Hagiwara, A., Bhalla, U.S., Murthy, V.N. and Albeanu, *Golan, L. and Shoham, S. (2009) Speckle elimination using
D.F. (2010) Non-redundant odor coding by sister mitral cells shift-averaging in high-rate holographic projection. Opt. Express.
revealed by light addressable glomeruli in the mouse. Nat. 17, 1330–1339
Neurosci. 13, 1404–1412 *Gradinaru, V., Thompson, K.R., Zhang, F., Mogri, M., Kay, K.,
Di Leonardo, R. and Bianchi, S. (2011) Hologram transmission Schneider, M.B. and Deisseroth, K. (2007) Targeting and readout
through multi-mode optical fibers. Opt. Express. 19, 1867–1869 strategies for fast optical neural control in vitro and in vivo. J.
Di Leonardo, R., Ianni, F. and Ruocco, G. (2007) Computer Neurosci. 27, 14231–14238
generation of optimal holograms for optical trap arrays. Opt. *Grossman, N., Poher, V., Grubb, M.S., Kennedy, G.T., Nikolic, K.,
Express. 15, 1913–1922 McGovern, B., Palmini, R.B., Gong, Z., Drakakis, E.M., Neil, M.A.,
Dimitrov, D., He, Y., Mutoh, H., Baker, B.J., Cohen, L., Akemann, W. Dawson, M.D., Burrone, J. and Degenaar, P. (2010) Multi-site
and Knöpfel, T. (2007) Engineering and characterization of an optical excitation using ChR2 and micro-LED array. J. Neural Eng.
enhanced fluorescent protein voltage sensor. PLoS One 2, e440 7, 16004
Dombeck, D.a, Harvey, C.D., Tian, L., Looger, L.L. and Tank, D.W. Guo, Z.V, Hart, A.C. and Ramanathan, S. (2009) Optical interrogation
(2010) Functional imaging of hippocampal place cells at cellular of neural circuits in Caenorhabditis elegans. Nat. Methods 6,
resolution during virtual navigation. Nat. Neurosci. 13, 891–896
1433–1440 Gutierrez, D.V, Mark, M.D., Masseck, O., Maejima, T., Kuckelsberg,
Du, R., Bi, K., Zeng, S., Li, D., Xue, S. and Luo, Q. (2009) Analysis of D., Hyde, R.a, Krause, M., Kruse, W. and Herlitze, S. (2011)
fast axial scanning scheme using temporal focusing with Optogenetic control of motor coordination by Gi/o protein-coupled
acousto-optic deflectors. J. Mod. Opt. 56, 81–84 vertebrate rhodopsin in cerebellar Purkinje cells. J. Biol. Chem.
Durst, M.E., Zhu, G. and Xu, C. (2006) Simultaneous spatial and 286, 25848–25858
temporal focusing for axial scanning. Opt. Express. 14, Hamill, O.P., Marty, A., Neher, E., Sakmann, B. and Sigworth, F.J.
12243–12254 (1981) Improved patch-clamp techniques for high-resolution
Durst, M.E., Zhu, G. and Xu, C. (2008) Simultaneous spatial and current recording from cells and cell-free membrane patches.
temporal focusing in nonlinear microscopy. Opt. Commun. 281, Pflügers Arch. 391, 85–100
1796–1805 Heim, R., Cubitt, A.B. and Tsien, R.Y. (1995) Improved green
Fan, G.Y., Fujisaki, H., Miyawaki, A., Tsay, R.K., Tsien, R.Y. and fluorescence. Nature 373, 663–664
Ellisman, M.H. (1999) Video-rate scanning two-photon excitation Heim, R., Prasher, D.C. and Tsien, R.Y. (1994) Wavelength mutations
fluorescence microscopy and ratio imaging with cameleons. and posttranslational autoxidation of green fluorescent protein.
Biophys. J. 76, 2412–2420 Proc. Natl. Acad. Sci. U. S. A. 91, 12501–12504

E. Papagiakoumou
*Helmchen, F. and Denk, W. (2005) Deep tissue two-photon Li, D., Hérault, K., Isacoff, E.Y., Oheim, M. and Ropert, N. (2012)
microscopy. Nat. Methods 2, 932–940 Optogenetic activation of LiGluR-expressing astrocytes evokes
Hirase, H., Nikolenko, V., Goldberg, J.H. and Yuste, R. (2002) anion channel-mediated glutamate release. J. Physiol. 590,
Multiphoton stimulation of neurons. J. Neurobiol. 51, 237–247 855–873
Hopt, a and Neher, E. (2001) Highly nonlinear photodamage in Liesener, J., Reicherter, M., Haist, T. and Tiziani, H.J. (2000)
two-photon fluorescence microscopy. Biophys. J. 80, Multi-functional optical tweezers using computer-generated
2029–2036 holograms. Opt. Commun. 185, 77–82
Huber, D., Petreanu, L., Ghitani, N., Ranade, S., Hromadka, T., Lin, J.Y., Lin, M.Z., Steinbach, P. and Tsien, R.Y. (2009)
Mainen, Z. and Svoboda, K. (2008) Sparse optical microstimulation Characterization of engineered channelrhodopsin variants with
in barrel cortex drives learned behaviour in freely moving mice. improved properties and kinetics. Biophys. J. 96, 1803–1814
Nature 451, 61–64 Liu, X., Ramirez, S., Pang, P.T., Puryear, C.B., Govindarajan, A.,
Jesacher, A. and Booth, M.J. (2010) Parallel direct laser writing in Deisseroth, K. and Tonegawa, S. (2012) Optogenetic stimulation of
three dimensions with spatially dependent aberration correction. a hippocampal engram activates fear memory recall. Nature 484,
Opt. Express. 18, 21090–21099 381–385
Jesacher, A., Maurer, C., Schwaighofer, A., Bernet, S. and Losavio, B.E., Iyer, V. and Saggau, P. (2009) Two-photon microscope
Ritsch-Marte, M. (2008) Full phase and amplitude control of for multisite microphotolysis of caged neurotransmitters in acute
holographic optical tweezers with high efficiency. Opt. Express. brain slices. J. Biomed. Opt. 14, 064033
16, 4479–4486 Losonczy, A., Zemelman, B.V, Vaziri, A. and Magee, J.C. (2010)
Kasparov, S. and Herlitze, S. (2013) Optogenetics at a crossroads? Network mechanisms of theta related neuronal activity in
Exp. Physiol. 98, 971–972 hippocampal CA1 pyramidal neurons. Nat. Neurosci. 13,
Katona, G., Szalay, G., Maák, P., Kaszás, A., Veress, M., Hillier, D., 967–972
Chiovini, B., Vizi, E.S., Roska, B. and Rózsa, B. (2012) Fast *Lutz, C., Otis, T.S., DeSars, V., Charpak, S., Digregorio, D.A. and
two-photon in vivo imaging with three-dimensional random-access Emiliani, V. (2008) Holographic photolysis of caged
scanning in large tissue volumes. Nat. Methods 9, 201–208 neurotransmitters. Nat. Methods 5, 821–827
*Katz, L.C. and Dalva, M.B. (1994) Scanning laser photostimulation: a Mao, T., O’Connor, D.H., Scheuss, V., Nakai, J. and Svoboda, K.
new approach for analyzing brain circuits. J. Neurosci. Methods (2008) Characterization and subcellular targeting of GCaMP-type
54, 205–218 genetically-encoded calcium indicators. PLoS One 3, e1796
*Katz, O., Small, E., Bromberg, Y. and Silberberg, Y. (2011) Focusing Matar, S., Golan, L. and Shoham, S. (2011) Reduction of two-photon
and compression of ultrashort pulses through scattering media. holographic speckle using shift-averaging. Opt. Express. 19,
Nat. Photon. 5, 372–377 25891–25899
Kleinlogel, S., Feldbauer, K., Dempski, R.E., Fotis, H., Wood, P.G., Mattis, J., Tye, K.M., Ferenczi, E.a, Ramakrishnan, C., Shea, D.J.O.,
Bamann, C. and Bamberg, E. (2011) Ultra light-sensitive and fast Prakash, R., Gunaydin, L.a, Hyun, M., Fenno, L.E., Gradinaru, V.,
neuronal activation with the Ca2+ -permeable channelrhodopsin Yizhar, O. and Deisseroth, K. (2012) Principles for applying
CatCh. Nat. Neurosci. 14, 513–518 optogenetic tools derived from direct comparative analysis of
Knapczyk, M. and Krishnan, A. (2005) High-resolution pulse shaper microbial opsins. Nat. Methods 9, 159–172
based on arrays of digital micromirrors. IEEE Photonic. Tech. L. 17, McAlinden, N., Massoubre, D., Richardson, E., Gu, E., Sakata, S.,
2200–2202 Dawson, M.D. and Mathieson, K. (2013) Thermal and optical
Knöpfel, T. (2012) Genetically encoded optical indicators for the characterization of micro-LED probes for in vivo optogenetic
analysis of neuronal circuits. Nat. Rev. Neurosci. 13, 687–700 neural stimulation. Opt. Lett. 38, 992–994
Ko, H., Hofer, S.B., Pichler, B., Buchanan, K.a, Sjöström, P.J. and *Miesenböck, G. (2009) The optogenetic catechism. Science 326,
Mrsic-Flogel, T.D. (2011) Functional specificity of local synaptic 395–399
connections in neocortical networks. Nature 473, 87–91 Mohanty, S.K., Reinscheid, R.K., Liu, X., Okamura, N., Krasieva, T.B.
Kravitz, A.V, Freeze, B.S., Parker, P.R., Kay, K., Thwin, M.T., and Berns, M.W. (2008) In-depth activation of channelrhodopsin
Deisseroth, K. and Kreitzer, A.C. (2010) Regulation of parkinsonian 2-sensitized excitable cells with high spatial resolution using
motor behaviours by optogenetic control of basal ganglia circuitry. two-photon excitation with a near-infrared laser microbeam cell
Nature 466, 622–626 culture. Biophys. J. 95, 3916–3926
Kuhn, B., Denk, W. and Bruno, R.M. (2008) In vivo two-photon Nagel, G., Brauner, M., Liewald, J.F., Adeishvili, N., Bamberg, E. and
voltage-sensitive dye imaging reveals top-down control of cortical Gottschalk, A. (2005) Light activation of channelrhodopsin-2 in
layers 1 and 2 during wakefulness. Proc. Natl. Acad. Sci. U. S. A. excitable cells of Caenorhabditis elegans triggers rapid behavioral
105, 7588–7593 responses. Curr. Biol. 15, 2279–2284
Lee, J.H., Durand, R., Gradinaru, V., Zhang, F., Goshen, I., Kim, D.-S., Nagel, G., Szellas, T., Huhn, W., Kateriya, S., Adeishvili, N., Berthold,
Fenno, L.E., Ramakrishnan, C. and Deisseroth, K. (2010) Global P., Ollig, D., Hegemann, P. and Bamberg, E. (2003)
and local fMRI signals driven by neurons defined optogenetically Channelrhodopsin-2, a directly light-gated cation-selective
by type and wiring. Nature 465, 788–792 membrane channel. Proc. Natl. Acad. Sci. U. S. A. 100,
*Leifer, A.M., Fang-Yen, C., Gershow, M., Alkema, M.J. and Samuel, 13940–13945
A.D.T. (2011) Optogenetic manipulation of neural activity in freely Nakai, J., Ohkura, M. and Imoto, K. (2001) A high signal-to-noise
moving Caenorhabditis elegans. Nat. Methods 8, 147–152 Ca(2+) probe composed of a single green fluorescent protein. Nat.
*Levitz, J., Pantoja, C., Gaub, B., Janovjak, H., Reiner, A., Hoagland, Biotechnol. 19, 137–141
A., Schoppik, D., Kane, B., Stawski, P., Schier, A.F., Trauner, D. and Nguyen, Q.T., Callamaras, N., Hsieh, C. and Parker, I. (2001)
Isacoff, E.Y. (2013) Optical control of metabotropic glutamate Construction of a two-photon microscope for video-rate Ca(2+)
receptors. Nat. Neurosci. 16, 507–516 imaging. Cell Calcium 30, 383–393
Li, N., Downey, J.E., Bar-Shir, A., Gilad, A.a, Walczak, P., Kim, H., *Nikolenko, V., Watson, B.O., Araya, R., Woodruff, A., Peterka, D.S.
Joel, S.E., Pekar, J.J., Thakor, N.V. and Pelled, G. (2011) and Yuste, R. (2008) SLM microscopy: scanless two-photon
Optogenetic-guided cortical plasticity after nerve injury. Proc. Natl. imaging and photostimulation with spatial light modulators. Front.
Acad. Sci. U. S. A. 108, 8838–8843 Neural Circuit. 2, 5

Oheim, M., Beaurepaire, E., Chaigneau, E., Mertz, J. and Charpak, S. Reddy, G.D. and Saggau, P. (2005) Fast three-dimensional laser
(2001) Two-photon microscopy in brain tissue: parameters scanning scheme using acousto-optic deflectors. J. Biomed. Opt.
influencing the imaging depth. J. Neurosci. Methods 111, 10, 64038
29–37 Reicherter, M., Haist, T., Wagemann, E.U. and Tiziani, H.J. (1999)
*Oron, D., Papagiakoumou, E., Anselmi, F. and Emiliani, V. (2012) Optical particle trapping with computer-generated holograms
Two-photon optogenetics. Prog. Brain Res. 196, 119–143 written on a liquid-crystal display. Opt. Lett. 24, 608–610
Oron, D. and Silberberg, Y. (2005) Harmonic generation with *Reutsky-Gefen, I., Golan, L., Farah, N., Schejter, A., Tsur, L., Brosh,
temporally focused ultrashort pulses. J. Opt. Soc. Am. B 22, I. and Shoham, S. (2013) Holographic optogenetic stimulation of
2660–2663 patterned neuronal activity for vision restoration. Nat. Commun. 4,
*Oron, D., Tal, E. and Silberberg, Y. (2005) Scanningless 1509
depth-resolved microscopy. Opt. Express. 13, 1468–1476 *Rickgauer, J.P. and Tank, D.W. (2009) Two-photon excitation of
Otsu, Y., Bormuth, V., Wong, J., Mathieu, B., Dugue, G.P., Feltz, A. channelrhodopsin-2 at saturation. Proc. Natl. Acad. Sci. U. S. A.
and Dieudonne, S. (2008) Optical monitoring of neuronal activity at 106, 15025–15030
high frame rate with a digital random-access multiphoton (RAMP) Rochefort, N.L., Garaschuk, O., Milos, R.-I., Narushima, M., Marandi,
microscope. J. Neurosci. Methods 173, 259–270 N., Pichler, B., Kovalchuk, Y. and Konnerth, A. (2009) Sparsification
*Packer, A.M., Peterka, D.S., Hirtz, J.J., Prakash, R., Deisseroth, K. of neuronal activity in the visual cortex at eye-opening. Proc. Natl.
and Yuste, R. (2012) Two-photon optogenetics of dendritic spines Acad. Sci. U. S. A. 106, 15049–15054
and neural circuits. Nat. Methods 9, 1202–1205 Ronzitti, E., Guillon, M., De Sars, V. and Emiliani, V. (2012) LCoS
Palima, D. and Glückstad, J. (2008) Comparison of generalized nematic SLM characterization and modeling for diffraction
phase contrast and computer generated holography for laser efficiency optimization, zero and ghost orders suppression. Opt.
image projection. Opt. Express 16, 5338–5349 Express. 20, 17843–17855
Palima, D. and Daria, V.R. (2006) Effect of spurious diffraction orders Royer, S., Zemelman, B.V, Barbic, M., Losonczy, A., Buzsáki, G. and
in arbitrary multifoci patterns produced via phase-only holograms. Magee, J.C. (2010) Multi-array silicon probes with integrated
Appl. Optics 45, 6689–6693 optical fibers: light-assisted perturbation and recording of local
Palima, D. and Daria, V.R. (2007) Holographic projection of arbitrary neural circuits in the behaving animal. Eur. J. Neursci. 31,
light patterns with a suppressed zero-order beam. Appl. Optics 46, 2279–2291
4197–4201 Rueckel, M., Mack-Bucher, J.a and Denk, W. (2006) Adaptive
Panda, S., Nayak, S.K., Campo, B., Walker, J.R., Hogenesch, J.B. wavefront correction in two-photon microscopy using
and Jegla, T. (2005) Illumination of the melanopsin signaling coherence-gated wavefront sensing. Proc. Natl. Acad. Sci. U. S. A.
pathway. Science 307, 600–604 103, 17137–17142
*Papagiakoumou, E., Anselmi, F., Bègue, A., de Sars, V., Glückstad, Sakai, S., Ueno, K., Ishizuka, T. and Yawo, H. (2013) Parallel and
J., Isacoff, E.Y. and Emiliani, V. (2010) Scanless two-photon patterned optogenetic manipulation of neurons in the brain slice
excitation of channelrhodopsin-2. Nat. Methods 7, 848–854 using a DMD-based projector. Neurosci. Res. 75, 59–64
*Papagiakoumou, E., Bègue, A., Leshem, B., Schwartz, O., Stell, Salomé, R., Kremer, Y., Dieudonné, S., Léger, J.-F., Krichevsky, O.,
B.M., Bradley, J., Oron, D. and Emiliani, V. (2013) Functional Wyart, C., Chatenay, D. and Bourdieu, L. (2006) Ultrafast
patterned multiphoton excitation deep inside scattering tissue. random-access scanning in two-photon microscopy using
Nat. Photon. 7, 274–278 acousto-optic deflectors. J. Neurosci. Methods 154,
Papagiakoumou, E., de Sars, V., Emiliani, V. and Oron, D. (2009) 161–174
Temporal focusing with spatially modulated excitation. Opt. Santos, M.D., Mohammadi, M.H., Yang, S., Liang, C.W., Kao, J.P.Y.,
Express. 17, 5391–5401 Alger, B.E., Thompson, S.M. and Tang, C.-M. (2012) Dendritic hold
*Papagiakoumou, E., de Sars, V., Oron, D. and Emiliani, V. (2008) and read: a gated mechanism for short term information storage
Patterned two-photon illumination by spatiotemporal shaping of and retrieval. PLoS One 7, e37542
ultrashort pulses. Opt. Express. 16, 22039–22047 Scanziani, M. and Hausser, M. (2009) Electrophysiology in the age of
Petreanu, L., Huber, D., Sobczyk, A. and Svoboda, K. (2007) light. Nature 461, 930–939
Channelrhodopsin-2-assisted circuit mapping of long-range Shoham, S., O’Connor, D.H., Sarkisov, D. V. and Wang, S.S. (2005)
callosal projections. Nat. Neurosci. 10, 663–668 Rapid neurotransmitter uncaging in spatially defined patterns. Nat.
Petreanu, L., Mao, T., Sternson, S.M. and Svoboda, K. (2009) The Methods 2, 837–843
subcellular organization of neocortical excitatory connections. Siegel, M. and Isacoff, E. (1997) A genetically encoded optical probe
Nature 457, 1142–1145 of membrane voltage. Neuron 19, 735–741
Piccolino, M. (2006) Luigi Galvani’s path to animal electricity. C. R. Sohal, V.S., Zhang, F., Yizhar, O. and Deisseroth, K. (2009)
Biol. 329, 303–318 Parvalbumin neurons and gamma rhythms enhance cortical circuit
Polosukhina, A., Litt, J., Tochitsky, I., Nemargut, J., Sychev, Y., De performance. Nature 459, 698–702
Kouchkovsky, I., Huang, T., Borges, K., Trauner, D., Van Gelder, Stirman, J.N., Crane, M.M., Husson, S.J., Gottschalk, A. and Lu, H.
R.N. and Kramer, R.H. (2012) Photochemical restoration of visual (2012) A multispectral optical illumination system with precise
responses in blind mice. Neuron 75, 271–282 spatiotemporal control for the manipulation of optogenetic
*Prakash, R., Yizhar, O., Grewe, B., Ramakrishnan, C., Wang, N., reagents. Nat. Protoc. 7, 207–220
Goshen, I., Packer, A.M., Peterka, D.S., Yuste, R., Schnitzer, M.J. Stirman, J.N., Crane, M.M., Husson, S.J., Wabnig, S., Schultheis, C.,
and Deisseroth, K. (2012) Two-photon optogenetic toolbox for fast Gottschalk, A. and Lu, H. (2011) Real-time multimodal optical
inhibition, excitation and bistable modulation. Nat. Methods 9, control of neurons and muscles in freely behaving Caenorhabditis
1171–1179 elegans. Nat. Methods 8, 153–158
Qiu, X., Kumbalasiri, T., Carlson, S.M., Wong, K.Y., Krishna, V., Straub, A., Durst, M.E. and Xu, C. (2011) High speed multiphoton
Provencio, I. and Berson, D.M. (2005) Induction of photosensitivity axial scanning through an optical fiber in a remotely scanned
by heterologous expression of melanopsin. Nature 433, 745–749 temporal focusing setup. Biomed. Opt. Express. 2, 80–88
Reddy, G.D., Kelleher, K., Fink, R. and Saggau, P. (2008) Stuart, G.J. and Sakmann, B. (1994) Active propagation of somatic
Three-dimensional random access multiphoton microscopy for action potentials into neocortical pyramidal cell dendrites. Nature
functional imaging of neuronal activity. Nat. Neurosci. 11, 713–720 367, 69–72

E. Papagiakoumou
Stuart, G., Schiller, J. and Sakmann, B. (1997) Action potential Yan, P., Acker, C.D., Zhou, W.-L., Lee, P., Bollensdorff, C., Negrean,
initiation and propagation in rat neocortical pyramidal neurons. J. A., Lotti, J., Sacconi, L., Antic, S.D., Kohl, P., Mansvelder, H.D.,
Physiol. 505(Pt 3), 617–632 Pavone, F.S. and Loew, L.M. (2012) Palette of fluorinated
Suchowski, H., Oron, D., Silberberg, Y., Suchowski Oron, D. and voltage-sensitive hemicyanine dyes. Proc. Natl. Acad. Sci. U. S. A.
Silberberg, Y.H. (2006) Generation of a dark nonlinear focus by 109, 20443–20448
spatio-temporal coherent control. Opt. Commun. 264, 482–487 *Yang, S., Papagiakoumou, E., Guillon, M., De Sars, V., Tang, C.M.
Tian, L., Hires, S.A., Mao, T., Huber, D., Chiappe, M.E., Chalasani, and Emiliani, V. (2011) Three-dimensional holographic
S.H., Petreanu, L., Akerboom, J., McKinney, S.a, Schreiter, E.R., photostimulation of the dendritic arbor. J. Neural Eng. 8, 46002
Bargmann, C.I., Jayaraman, V., Svoboda, K. and Looger, L.L. Yizhar, O., Fenno, L.E., Davidson, T.J., Mogri, M. and Deisseroth, K.
(2009) Imaging neural activity in worms, flies and mice with (2011) Optogenetics in neural systems. Neuron 71, 9–34
improved GCaMP calcium indicators. Nat. Methods 6, 875–881 Zahid, M., Velez-Fort, M., Papagiakoumou, E., Ventalon, C., Angulo,
Tsai, H.C., Zhang, F., Adamantidis, A., Stuber, G.D., Bonci, A., De M.C. and Emiliani, V. (2010) Holographic photolysis for multiple cell
Lecea, L. and Deisseroth, K. (2009) Phasic firing in dopaminergic stimulation in mouse hippocampal slices. PLoS One 5,
neurons is sufficient for behavioral conditioning. Science 324, e9431
1080–1084 Zemelman, B., Lee, G., Ng, M. and Miesenböck, G. (2002) Selective
Vaziri, A. and Emiliani, V. (2012) Reshaping the optical dimension in photostimulation of genetically chARGed neurons. Neuron 33,
optogenetics. Curr. Opin. Neurobiol. 22, 128–137 15–22
Vellekoop, I.M. and Mosk, A.P. (2007) Focusing coherent light through *Zemelman, B.V, Nesnas, N., Lee, G.a and Miesenbock, G. (2003)
opaque strongly scattering media. Opt. Lett. 32, 2309–2311 Photochemical gating of heterologous ion channels: remote
Volgraf, M., Gorostiza, P., Numano, R., Kramer, R.H., Isacoff, E.Y. and control over genetically designated populations of neurons. Proc.
Trauner, D. (2006) Allosteric control of an ionotropic glutamate Natl. Acad. Sci. U. S. A. 100, 1352–1357
receptor with an optical switch. Nat. Chem. Biol. 2, 47–52 Zernike, F. (1955) How I discovered phase contrast. Science 121,
*Wang, K., Liu, Y., Li, Y., Guo, Y., Song, P., Zhang, X., Zeng, S. and 345–349
Wang, Z. (2011) Precise spatiotemporal control of optogenetic *Zhang, F., Gradinaru, V., Adamantidis, A.R., Durand, R., Airan, R.D.,
activation using an acousto-optic device. PLoS One 6, e28468 De Lecea, L. and Deisseroth, K. (2010) Optogenetic interrogation
Wang, H., Peca, J., Matsuzaki, M., Matsuzaki, K., Noguchi, J., Qiu, of neural circuits: technology for probing mammalian brain
L., Wang, D., Zhang, F., Boyden, E., Deisseroth, K., Kasai, H., Hall, structures. Nat. Protoc. 5, 439–456
W.C., Feng, G. and Augustine, G.J. (2007a) High-speed mapping Zhang, Y., Holbro, N. and Oertner, T.G. (2008) Optical induction of
of synaptic connectivity using photostimulation in plasticity at single synapses reveals input-specific accumulation of
Channelrhodopsin-2 transgenic mice. Proc. Natl. Acad. Sci. U. S. alphaCaMKII. Proc. Natl. Acad. Sci. U. S. A. 105, 12039–12044
A. 104, 8143–8148 Zhang, J., Laiwalla, F. and Kim, J. (2009) Integrated device for optical
Wang, S., Szobota, S., Wang, Y., Volgraf, M., Liu, Z., Sun, C., Trauner, stimulation and spatiotemporal electrical recording of neural
D., Isacoff, E.Y. and Zhang, X. (2007b) All optical interface for activity in light-sensitized brain tissue. J. Neural Eng. 6,
parallel, remote and spatiotemporal control of neuronal activity. 055007
Nano Lett. 7, 3859–3863 Zhang, Y.P. and Oertner, T.G. (2007) Optical induction of synaptic
Wang, J., Wagner, F., Borton, D.a, Zhang, J., Ozden, I., Burwell, R.D., plasticity using a light-sensitive channel. Nat. Methods 4,
Nurmikko, A.V, Van Wagenen, R., Diester, I. and Deisseroth, K. 139–141
(2012) Integrated device for combined optical neuromodulation Zhang, F., Wang, L.P., Boyden, E.S. and Deisseroth, K. (2006)
and electrical recording for chronic in vivo applications. J. Neural Channelrhodopsin-2 and optical control of excitable cells. Nat.
Eng. 9, 016001 Methods 3, 785–792
Watt, A.J., Cuntz, H., Mori, M., Nusser, Z., Sjöström, P.J. and Zhang, F., Wang, L.P., Brauner, M., Liewald, J.F., Kay, K., Watzke, N.,
Häusser, M. (2009) Traveling waves in developing cerebellar cortex Wood, P.G., Bamberg, E., Nagel, G., Gottschalk, A. and Deisseroth,
mediated by asymmetrical Purkinje cell connectivity. Nat. K. (2007) Multimodal fast optical interrogation of neural circuitry.
Neurosci. 12, 463–473 Nature 446, 633–639
Wilson, N.R., Runyan, C.a, Wang, F.L. and Sur, M. (2012) Division Zhu, P., Fajardo, O., Shum, J., Zhang Schärer, Y.-P. and Friedrich,
and subtraction by distinct cortical inhibitory networks in vivo. R.W. (2012) High-resolution optical control of spatiotemporal
Nature 488, 343–348 neuronal activity patterns in zebrafish using a digital micromirror
Wilt, B.A, Burns, L.D., Wei Ho, E.T., Ghosh, K.K., Mukamel, E.A and device. Nat. Protoc. 7, 1410–1425
Schnitzer, M.J. (2009) Advances in light microscopy for *Zhu, G., Van Howe, J., Durst, M., Zipfel, W. and Xu, C. (2005)
neuroscience. Annu. Rev. Neurosci. 32, 435–506 Simultaneous spatial and temporal focusing of femtosecond
Witten, I.B., Lin, S.-C., Brodsky, M., Prakash, R., Diester, I., pulses. Opt. Express. 13, 2153–2159
Anikeeva, P., Gradinaru, V., Ramakrishnan, C. and Deisseroth, K. Zipfel, W.R., Williams, R.M. and Webb, W.W. (2003) Nonlinear magic:
(2010) Cholinergic interneurons control local circuit activity and multiphoton microscopy in the biosciences. Nat. Biotechnol. 21,
cocaine conditioning. Science 330, 1677–1681 1369–1377
Wyart, C., Del Bene, F., Warp, E., Scott, E.K., Trauner, D., Baier, H. Zorzos, A., Boyden, E.S. and Fonstad, C.G. (2010) A multi-
and Isacoff, E.Y. (2009) Optogenetic dissection of a behavioural waveguide implantable probe for light delivery to sets of
module in the vertebrate spinal cord. Nature 461, 407–410 distributed brain targets. Opt. Lett. 35, 4133–4135
Received: 9 December 2012; Accepted: 12 June 2013; Accepted article online: 18 June 2013

Optical Optogenetics

Uploaded by

Copyright:

Available Formats

Optical Optogenetics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Optical Optogenetics

Uploaded by

Copyright:

Available Formats

Biol. Cell (2013) 105, 1–22 DOI: 10.1111/boc.

Optical developments for

Introduction Electrophysiology suffers, however, from funda-

2 www.biolcell.net | Volume (105) | Pages 1–22

4 www.biolcell.net | Volume (105) | Pages 1–22

Figure 1 Laser illumination techniques

6 www.biolcell.net | Volume (105) | Pages 1–22

8 www.biolcell.net | Volume (105) | Pages 1–22

10 www.biolcell.net | Volume (105) | Pages 1–22

Figure 3 Temporal focusing configuration

Figure 4 See Legend on next page

12 www.biolcell.net | Volume (105) | Pages 1–22

Figure 4 Axial resolution of 2PE approaches in optogenetics

14 www.biolcell.net | Volume (105) | Pages 1–22

Figure 5 Generalised phase contrast

16 www.biolcell.net | Volume (105) | Pages 1–22

18 www.biolcell.net | Volume (105) | Pages 1–22

20 www.biolcell.net | Volume (105) | Pages 1–22

22 www.biolcell.net | Volume (105) | Pages 1–22

You might also like