Channelrhodopsin restores vision in blind mice

New research shows that a protein found in green algae can partially restore visual function when delivered into the retina of blind mice.

The work brings us one step closer to a genetic therapy for various conditions in which the degeneration of retinal cells leads to imapired vision or complete blindness.

Normally, light entering the eye falls upon the rods and cones at the back of the retina. These are the photoreceptors: they are packed with a light-sensitive protein called rhodopsin, which initiates an electrical signal when struck by photons (the particles which carry light).

The signals generated by the photoreceptors are transmitted to the bipolar cells, which in turn transmit the signals to ganglion cells, whose axons leave the retina to form the optic nerve. Thus, visual information is carried from the eye to the visual centres of the brain, via the lateral geniculate nucleus, a relay station in the thalamus.

In the retina, visual information passes along two separate and parallel channels, each one most sensitive to different types of light stimuli. Photoreceptors in the ON centre pathway are sensitive to bright stimuli against a dark background, whereas cells in the OFF centre pathway are sensitive to dark stimuli on a light background.

Bipolar cells receive converging inputs from several photoreceptors, and so have similar receptive field properties. ON bipolar cells are most responsive to a light spot on a dark background, and vice versa for OFF bipolar cells. The cells in the retina are therefore well-suited to detecting contrast, which is ideal for recognizing the edges of objects.

Contrast is further increased by a process called lateral inhibition, whereby activity in one pathway reduces activity in the other. In this way, the circuits in the retina carry out simple processing of the visual data entering the eye. They generate a very crude representation of the visual image, which is then sent to the brain for further processing.

The new research, led by Botond Roska of the Friedrich Miescher Institute for Biomedical Research in Basel, Switzerland, and Connie Cepko of Harvard Medical School, is reported online in an advance online publication in Nature Neuroscience. It uses a protein called channelrhodopsin-2 (ChR2), a light-sensitive ion channel that was recently isolated from the green algae Chlamydomonas reinhardtii.

ChR2 belongs to a family of proteins which are also found in bacteria, and are related to the rhodopsin proteins found in the retina of mammals. Unlike their mammalian counterparts, the channelrhodopsins are directly linked to a pore which spans the membrane of the cell, and can open or close in response to light of a specific wavelength and so alter the flow of electrical currents across the membrane. They can therefore confer light sensitivity onto cells that do not normally respond to light.

It is this property of the algal protein that was exploited in the new study. In conditions such as macular degeneration and retinitis pigmentosa, there is a progressive loss of photoreceptors. The researchers therefore reasoned that they could bypass the photoreceptors entirely, and restore visual function by using ChR2 to make the bipolar cells responsive to light.

The study was carried out on a strain of mutant mice which lack a gene that is required for the survival of photoreceptors, and which provide a good model for human retinal degenerative conditions. These animals lose most of their photoreceptors by 4 weeks of age, by which time, light stimulation does not elicit any detectable electrical activity in retinal cells. In other words, by that age, the mice are virtually blind.

The researchers first created a genetic construct consisting of the ChR2 gene fused to the gene encoding a fluorescent protein. This construct was then delivered to the animals’ retinas by in vivo electroporation, whereby cells are made transiently permeable to DNA by the application of a small external electric field.

The construct also contained a regulatory DNA sequence called a promoter, which normally drives the synthesis of a glutamate receptor in a specific cell type: the ON centre bipolar cells in the retina. So, when delivered to the retina, the construct may enter various types of cells, but will only be activated in the ON bipolar cells.


In mutant mice lacking photoreceptors, targeted expression of ChannelRhodopsin 2 (ChR2) confers light sensitivity upon bipolar cells. Bipolar cells expressing ChR2 are stained green. Scale bar = 10 micrometers. (From Lagali et al, 2008.)

Once inside, the construct is incorporated into the chromosomes, so that cells containing it synthesize both the fluorescent protein and ChR2. This should make the bipolar cells sensitive to light, and an electrode array which recorded electrical activity in the ganglion cells in response to light confirmed that this was indeed the case.

Thus, in the absence of photoreceptors, the bipolar cells synthesizing Chr2 were sensitive to light, which activated them, so that they generate an electrical signal which is then transmitted to the ganglion cells. The number of signals recorded was found to dependent on the intensity of light, so that the more intense the stimulus used, the more frequent were the bursts of ganglion cell activity.

This was further confirmed anatomically. The parallel ON and OFF bipolar cell streams have slightly different projections, with their processes terminating in two different layers of the retina, and confocal microscopy (above) showed that the cells targeted by the ChR2 construct project to the outer layer only, which is characteristic of ON bipolar cells.

Furthermore, electrophysiological recordings further showed that the responses of the bipolar cells are transmitted to the brain: when the experimental animals were presented with light stimuli, the electrodes detected corresponding activity in cells in the visual cortical areas.

The new ability of the previously-blind mice also affected their behaviour. When placed in a box that is divided into light and dark compartments, the movements of normal mice, but not of blind ones, increase. And when the mutant mice with ChR2-expressing bipolar cells were placed in this apparatus, their movements were comparable to those of normal animals.

Elsewhere, researchers are trying to develop retinal implants for the treatment of eye diseases. A therapy based on ChR2  would have the advantage of not requiring the implantation of foreign metallic objects (electrodes) into the retina. And, because it involves genetic targeting of a specific type of retinal cell, it is an improvement on previous similar work, which has been less selective.

Although the results of the present are very promising, there are several problems. First, the ChR2 protein was introduced into less than 10% of the bipolar cells in the retina, and was not evenly distributed; the bipolar cells were sensitive to a far narrower range of light intensities than are photoreceptors; and they required a far greater intensity to be stimulated into action.

Finally, it should not be assumed that the animal model faithfully reproduces all aspects of eye diseases in humans. It is unclear, for example, how the connections between retinal cells remain intact during the course of progressive degeneration. Undoubtedly, these issues will be addressed in future work.


Lagali, P. S., et al. (2008). Light-activated channels targeted to ON bipolar cells restore visual function in retinal degeneration. Nat. Neurosci. doi: 10.1038/nn.2117. [Abstract]

About these ads

4 thoughts on “Channelrhodopsin restores vision in blind mice

  1. Thank you for commenting on our work. I only want to point out one error. The electroporated gene most likely does not integrate into the chromosomes. This is important as it might limit the time during which the gene would be expressed. Alternatives, such as viral vectors that do integrate, or at least remain stably associated even if they do not integrate, would be preferable and will be tried.

Comments are closed.