Controlling animal behaviour with an optical on/off switch for neurons

Last week, I wrote about the work of Ed Boyden and his colleagues at MIT’s Media Lab. Boyden’s research group has developed a method by which light is used to control neuronal activity. The method involves the use of a light-activated protein called channelrhodopsin (ChR2); the gene encoding ChR2 was recently isolated from the extremophile archaebacterium Natronomonas pharaonis. (Incidentally, Boyden read my post and commented on it).

In today’s issue of Nature, former collaborators of Boyden report on a similar method, based on another photosensitive protein, called NpHR. The authors cloned the NpHR gene from N. pharaonis. The gene product is a protein called NpHR, which works in the same way as ChR2 – it is also a light-activated chloride channel. Photoactivation opens the channel, allowing an influx of chloride ions into the cell. Like ChR2, it can be used to knock out single action potentials in cultured cells. But the authors of the new study also describe experiments in which they expressed the protein in living nematode worms (Caenhorhabditis elegans), and used it to control the worms’ behaviour.

A construct consisiting of the NpHR gene fused to a muscle-specific promoter and a gene encoding a fluorescent protein was created, and a lentivirus vector was used to deliver the construct to cholinergic motor neurons in the nerve cord of the worms. These cells are located in the ventral aspect of the neamtode nerve cord; they innervate four longitudinal muscles that extend along the length of the worm’s body, and control the swimming movements. When the NpHR protein expressed in the motor neurons was activated with pulses of yellow light, the activity of the cells was inhibited, causing the worms to stop swimming.

This film clip shows the use of the system to control swimming behaviour in a nematode worm. When a pulse of light is directed at the worm, it activates the NhPR protein; this hyperpolarizes the motor neurons and inhibits their activity. The swimming movements cease about 600 milliseconds of illumination. Switching off the light pulse inactivates the protein, so that the motor neurons resume sending action potentials to the muscles. Illumination with yellow light is indicated by a yellow dot:

The authors also expressed NpHR in cultured neurons isolated from the CA1 and CA3 regions of the rat hippocampus. They used calcium imaging to show that photoactivation of NpHR causes a large increase in intracellular calcium ion concentration. The hippocampal cells were then made to co-express NpHR and ChR2. Because the two proteins are sensitive to different wavelengths of light, they can complement each other when used in the same system; a specific wavelength of light can be used to activate one of the proteins, but not the other (~580 and ~460 nanometres, for NpHR and ChR2, respectively).

Used together, NpHR and ChR2 will enable researchers to precisely control the activity of specific cells within neuronal circuits. Other research groups have developed similar systems using other halorhodopsin proteins to excite rather than inhibit cells. A combination of these proteins would therefore allow for a bidirectional switch; different wavelengths of light could be used to simultaneously activate or inhibit individual cells within a circuit.

Photo Sharing and Video Hosting at Photobucket

These methods will prove to be extremely powerful for researchers, but outside the lab, they will also have numerous clinical applications. I concluded last week’s post with a brief discussion of how these new methods could be used to develop optical neural prostheses for controlling the abnormal electrical activity associated with conditions such as epilepsy and Parkinson’s Disease. In a News and Views article accompanying the new paper, Michael Häusser and Spencer Smith reiterate this, and extend it:

…diseases such as epilepsy, which involves hyperactivity or altered excitation–inhibition balance in neurons, might be ameliorated by activating NpHR and/or ChR2 targeted to specific cell types. With implanted fibre-optic stimulation, these optical tools could also provide a more specific and durable form of deep-brain stimulation, which has shown some success in the treatment of Parkinson’s disease. Finally, one can envisage using this approach to create new types of optical neural prosthetic, such as in the retina or the sensory or motor cortex, for improving sensory perception and control of movement.


The paper discussed here, and the News and Views article by Häusser and Smith, are freely-accessible in a Nature web focus, which includes film clips of the researchers talking about the work. The journal’s weekly podcast also includes a discussion of the work.


Zhang, F, et al. (2007). Multimodal fast optical interrogation of neural circuitry. Nature 446: 633-639. [Full text]

Häusser, M. & Smith, S. (2007). Neuroscience: Controlling neural circuits with light. Nature 446: 617-619. [Full text]



3 thoughts on “Controlling animal behaviour with an optical on/off switch for neurons

  1. Hi, light (photons, in the broadest sense, of the electromagnetic spectrum) have been used for therapy for quite a long time. Cancer, psoriasis, kernicterus (convulsion in newborn due to excess bilirubin) are some examples where light has been traditionally used. They act by forming pyrimidine dimers, cis-trans isomerism etc…During solar eclipse, the behavior of birds and other animals, possibly suggest the role of some photon-gated ion channels (GABA/Glutamate etc.), as they DO NOT reflect simple diurnal/circadian phenomenon. Moreover, Vitamin A antagonists may be exploited to learn the contribution of retinene/retinal in this regard.
    Thanks for the article and the video.

  2. Pingback: Chasing Shadows

Comments are closed.