Scientists would love to map the activity of neural circuits that drive behavior, thought, and emotion. Doing that requires that one can direct neurons to fire or fall silent from afar, a bit like a lecturer might point to a spot on the screen with a laser pointer. In the living brain, that’s largely the stuff of science fiction. Now, Karl Deisseroth and colleagues from Stanford University in California have developed a new tool for accomplishing this task using flashes of colored light that control neuron activity quickly and precisely. Together with collaborators from the Johann Wolfgang-Goethe-University and the Max Planck Institute, both in Frankfurt, Germany, Deisseroth engineered neurons to express a light-sensitive ion channel or a chloride pump, or both. This allows the investigators to turn on neuronal firing with a flash of blue light, or turn it off with yellow. By expressing the channels in the worm C. elegans, the researchers showed they could control swimming behavior in live worms with pulses of light. The work appears in the April 4 Nature.



Artist's Rendering of Optical Remote-control in Action
Blue and yellow lights evoke or inhibit electrical activity in neurons. Image credit: Feng Zhang, Steve Dixon, and Karl Deisseroth

“The system provides researchers with a two-knob remote control for increasing or decreasing the activity of specific neurons using different colors of light,” write Michael Hausser and Spencer Smith at University College London, United Kingdom, in a commentary accompanying the paper. “The overwhelming advantages of this new approach should revolutionize the field,” they write, since having both stimulatory and inhibitory control over neuronal activity will allow researchers to prove when activity is either necessary or sufficient to trigger behavioral effects.

Another paper, out in the March issue of PLoS ONE from Xue Han of Stanford and Edward Boyden of MIT in Cambridge, Massachusetts, shows similar results with the same pump in cultured neurons.

Previously, Deisseroth and colleagues had shown that they can express the algal protein channel rhodopsin-2 (ChR2) in neurons, and that the protein rendered the neurons sensitive to depolarization by treatment with flashes of blue light (Boyden et al., 2005). That channel constituted a remote “on” switch for the genetically modified neurons. In the new work, the scientists introduce the companion “off” switch—a light-driven halorhodopsin chloride pump from the archaebacteria Natronomonas pharaonis (NpHR), which responds to yellow light. Expression of NpHR allowed the rapid and reversible hyperpolarization of cells, shutting down spiking activity. The responses to light were so fast, on the millisecond scale, that it was possible to block single action potentials in a train of spikes.

That the channel and pump register different wavelengths of light was a lucky break. It raised the possibility that cells expressing both proteins could be rapidly toggled between depolarization and hyperpolarization. The researchers showed that, indeed, they could use mixed light pulses on doubly expressing cells to generate predetermined on/off firing patterns either in cultured cells or in tissue slices from mouse brain. What’s more, by tagging the pump and channel with fluorescent proteins, and loading cells with a calcium dye, they could find modified cells, optically activate them, and measure the resulting calcium transients in cortical slices. Using these technologies together allowed the researchers to “identify, observe and control” intact living neural circuitry, the authors write.

To test their system in living animals, Deisseroth and coworkers turned to C. elegans. When they expressed the NpHR in either body wall muscle cells or cholinergic neurons, they were able to paralyze swimming worms with pulses of yellow light. The two proteins worked together, as well: in doubly expressing worms, ChR2 activation caused muscle contraction (the worms appeared shorter), which was reversed by NpHR activation.

Besides opening up a new frontier in circuit analysis, the technique could have clinical use, too. Deep brain stimulation, like that used to treat Parkinson disease, could be accomplished with fiber optic illumination of suitably engineered cells. The authors conjure up the idea of an “optical neuronal prosthetic,” for the treatment of disease.—Pat McCaffrey


  1. This article describes very concrete applications for a class of optically-activated ion channels/pumps based on the rhodopsin protein. These light-activated G-protein coupled receptors can be genetically manipulated and expressed in many cell types in order to control electrical excitability. They are therefore powerful tools for neurophysiology. This study describes the kinetics and functional output of the excitatory channel rhodopsin (ChR2) and the inhibitory halorhodopsin (NpHR); the former activates a non-selective cation channel while the latter actives a chloride pump. These and other genetically encoded rhodopsins were discussed in a recent Cold Spring Harbor meeting on Imaging Neurons and Neural Activity, and are neatly outlined in a review by Herlitze and Landmesser (2007).

    This novel approach allows for ‘hands free’ manipulation of neuronal activity simply by exposing the rhodopsin-expressing cells to light of a select wavelength. In certain proteins, such as the two discussed in this study, the excitation wavelengths are distinct, and therefore can be co-expressed and controlled independently. The beauty is in the design, and the details are being ironed out in studies such as this. Optically controlling neuronal activity, rather than direct electrical stimulation, is powerful for several reasons. Importantly, it removes the invasive and potentially confounding stimulating electrodes from the tissue preparation. It also permits precise spatial, as well as temporal, control when activating individual cells or circuits. By changing the light source aperture, individual cells, subcellular compartments, or entire regions of tissue can be selectively activated. This level of spatial resolution is one of the more intriguing, and novel, aspects of optical stimulation. Changing the temporal or intensity parameters can control the duration or amplitude of the ion flux, similar to extending the pulse duration, or increasing the current intensity, of conventional stimulating electrodes – yet without the spatial control. When combined with existing optical indicators, such as Ca2+- or voltage -sensitive dyes, many of the invasive techniques conventionally used to probe neuronal activity are obviated, such as electrodes and sensors. The authors here take the degree of manipulation one step further and control the movements of entire organisms – shedding light on much untapped potential with this technique.

    The widespread use of genetically-encoded rhodopsins is still, for the time being, limited by the techniques required to get these vectors into, and expressed by, neurons and other cells. Viral vectors, electroporation, and other transfection tools are not amenable to all preparations or protocols. Here it is a case of molecular techniques keeping up with the biological application. Hopefully it is just a matter of time before inserting these proteins into cells of choice becomes commonplace. Certainly, for experimental purposes, the development of transgenic mice will be invaluable. As a tool, many may feel that optogenetic techniques are inherently simpler to use and more precise in its ability to control cellular activity than conventional electrical stimulation and recording. Yet, imprecision exists in the variability of expression patterns and levels associated with all exogenously expressed proteins. In addition, the tunability of channel flux still appears rather limited, and the cellular output response can depend largely on the subcellular channel/pump distribution, as well as endogenously expressed regulators and buffers. Hopefully, genetic alterations to these optically activated channels can improve their functional characteristics, analogous to the development of eGFP and the XFP variants. Plasticity studies, transgenic animals, and human applications will hopefully soon follow.

    My feeling is that once some of the mundane and practical hurdles are overcome, optogenetic techniques will not only stay, but provide vastly new insight into neuronal circuitry. Its use as a therapeutic clinical tool are already being discussed as potential improvements to deep brain stimulation for Parkinson’s, recovery of neuronal function in select brain regions such as in Alzheimer’s, injury, and stroke, and activation of entire circuits to restore sensory and motor system impairments. In theory, these are viable applications for optical stimulation of light-sensitive receptors. And, because of the potential power of ‘optogenetics’ , many research groups will likely make the concerted effort to characterize and improve these proteins, and make them accessible for both basic research and clinical realms.


    . New optical tools for controlling neuronal activity. Curr Opin Neurobiol. 2007 Feb;17(1):87-94. PubMed.

    View all comments by Grace Stutzmann

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . Millisecond-timescale, genetically targeted optical control of neural activity. Nat Neurosci. 2005 Sep;8(9):1263-8. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Multimodal fast optical interrogation of neural circuitry. Nature. 2007 Apr 5;446(7136):633-9. PubMed.
  2. . Neuroscience: controlling neural circuits with light. Nature. 2007 Apr 5;446(7136):617-9. PubMed.
  3. . Multiple-color optical activation, silencing, and desynchronization of neural activity, with single-spike temporal resolution. PLoS One. 2007;2(3):e299. PubMed.