Here is one of our squirrel monkeys, Dalton, who was treated for red-green color blindness enjoying a feast of colored fruits and vegetables. The image on the left was digitally altered to simulate what the scene would look like to a person (or monkey) with red-green color blindness.
While gene therapy has successfully allowed red-green color-blind monkeys to see new colors that they have never seen before, we still don’t know what their internal perceptions of those colors are like, or if any psychological side-effects might result from humans suddenly being able to see a new dimension of color. [Note: There have been no indications of psychological distress in the monkeys.] Gene therapy also involves risks associated with the viral vector and therapeutic transgene being injected, and with the surgical procedure itself (subretinal injection). Therefore, the first step in moving the treatment forward will be determining its safety for use in human patients.
Human experiments involving gene therapy must first be reviewed and approved by the National Institute of Health’s (NIH) Office of Recombinant DNA Activities (ORDA)/ Recombinant DNA Advisory Committee (RAC) and by the Food and Drug Administration (FDA). In addition to approval of an Investigational New Drug Application (IND) from the FDA, approval from an Institutional Review Board (IRB) where the study will take place must also be obtained.
We would also need to recruit subjects who would be willing to be pioneers in the cure for red-green color blindness; that is, willing to accept the risks involved knowing that it may not work.
The same gene therapy virus vector and injection procedures are currently being used in human gene therapy trials for a blinding disorder known as Leber’s congenital amaurosis, or LCA. Thus far, no serious adverse events have been reported, even after one year post-treatment. A key difference in our experiments for color blindness is the therapeutic gene that is carried by the virus vector. Because a human visual pigment gene was used to replace the missing visual pigment of the monkeys, and no adverse side effects have been observed, we are optimistic that this transgene will also be safe to use in humans. The most critical barrier in moving the treatment forward will be insuring its safety for human patients.
While red-green color blindness is generally not considered to be a debilitating visual disorder, many affected individuals would disagree. Normal color vision is required for employment as a police officer, fire fighter, commercial/public transit driver, or pilot. In other professions, the requirement for normal color vision is not as obvious, and some spend years training for careers as designers, geologists, chemists, or ophthalmologists before being excluded by their color vision deficiency. Everyday difficulties associated with red-green color blindness are presented in the slide show, “colorblind world.”
Our laboratory and the gene therapy project were recently included in an award-winning documentary film series, “Cracking the Colour Code,” produced by Electric Pictures (Perth, Western Australia). Please see the here for more details.
We used a computerized test for human color blindness that was similar to the well-known testing books in which colored numbers or symbols are concealed in a pattern of dots. Prior to treatment, the monkeys were trained to touch the location of a colored patch hidden among the gray dots. Correct choices were rewarded with a small amount of white grape juice and a positive dinging sound. Following incorrect choices, no juice was delivered and a negative buzzer tone sounded.
Similar to color-blind humans, the monkeys could not distinguish red or green, but following treatment they passed the test easily for all colors. This is a movie of Dalton performing the test for a reddish color that he was not able to see prior to treatment.
Because the new dimension of red-green color vision was closely timed with the appearance of expression of the new visual pigment transgene, we conclude that neural rewiring was not associated with the change in color vision. Rather, the new visual pigment took advantage of pre-existing visual circuitry and altered its spectral response characteristics to automatically give rise to a new dimension of color.
What is the neural circuitry for color vision? Hundreds of years of color vision and color matching experiments have established that the four main hue percepts (blue, yellow, red, and green) involve contributions from all three cone types, short- (S), middle- (M), and long- (L) wavelength-sensitive. Theories of color vision have focused on cell types recorded physiologically in the retina and LGN with S vs. (L+M) and L vs. M signatures. As a result, textbooks have attributed blue-yellow and red-green perceptions to the respective cells containing these signals. Importantly it must be emphasized the physiology of these cell types do not match the spectral characteristics of human perception. It is possible hue perception is based on cells in the retina matching the spectral signatures of human color vision.
In understanding color vision it helps to first consider reduced systems with fewer cone types and fewer percepts. In dichromatic animals, retinas are composed of either S- and M-cones (protanope) or S- and L-cones (deuteranope) as shown in the figure below. When behaviorally tested these animals demonstrate the ability to discriminate blues and yellows, whereas greens and reds are indistinguishable from gray.
The relative percentage of S-cones when compared to the total number of cones retina remains similar across mammals, at around 5-7% of the whole. This creates a unique sampling mosaic that is conserved across many species, in which chromatic information is available only from the M vs. S (or L vs. S) receptive fields. For example, the dichromatic squirrel monkeys only had M and S cones. Their M vs. M center-surround receptive fields carry only light-dark edge information. Therefore, color signals leaving the dichromatic retina are carried down two pathways, those with S-cone centers and those with M cone centers with S-cone inputs to the surround.
In the gene therapy experiment the pre-treated retinas contained only S- and M-cones. By adding a third class of cone the existing M center vs. S surround circuitry was split into two pathways; an M vs. (S+L) and an L vs. (S+M) thus creating new, uncorrelated, activity patterns leaving the retina. Receptive fields that do not have S cone inputs do not contribute to hue perception in dichromats. If the cells in trichromats without S cone input do not contribute to hue perception, then it is not necessary to propose that an entire visual pathway that was previously dedicated to spatial vision was converted to a new purpose of color vision. Gaining a new dimension of color vision becomes a simple matter of splitting the preexisting blue-yellow pathway into two systems, one for blue-yellow and a second for red-green color vision.
By applying these basic rules, we have constructed a flash based circuit demonstration which can be found here. Please follow the link and play around with the circuit. Instructions are on the left to perform a virtual version of the gene therapy project. Also available is a pdf of the circuits described here. Click on the picture below to begin downloading.