This article was originally published in The Daily UW on March 14, 2019. Read it here.
To about 8 percent of men and 0.5 percent of women, the world doesn’t look quite the same as it does to others. Red — the color of blood, cherries, lipstick, anger, love — is blocked from the brain’s visual input from birth, rendered indistinct from various hues of blue and yellow. The resulting world looks uniformly like a scene from a David Fincher film: dull, desaturated, neutral. People with colorblindness experience the same things as everyone else, but not everything looks as vivid as it may feel.
Being colorblind can have consequences on a number of tasks as well.
“Colorblindness affects [my work] in unexpected ways,” Eric Banisadr, a UW freshman who intends to study computer science, said. “There are many times in CS where you have to go out of your way to develop accessibility features for people with various disabilities, [including] making a program accessible for a colorblind person. Those kinds of things take me a lot of extra work to develop.”
Colorblindness is genetically inherited and has long been considered something that people just have to learn to live with. However, Jay and Maureen Neitz, two researchers at the UW department of ophthalmology, beg to differ.
For the past 10 years, they’ve been working on a cure for colorblindness that uses gene therapy to induce trichromatic color vision in squirrel monkeys. Not only was the experiment successful, but it represents new breakthroughs in understanding the science of color vision — and has the potential to work on humans, too.
In 2009, the Neitzes tested an experimental virus on two male squirrel monkeys named Sam and Dalton. Male squirrel monkeys, like most mammals, are dichromats, which means they have only two photoreceptors for processing color – neither of which is naturally sensitive to longer wavelengths. In other words, they are red-green colorblind. The virus developed by the Neitz Lab was able to reprogram the photoreceptors in the monkeys’ retinas to detect longer, redder wavelengths of light, effectively turning them into trichromats.
“It all started with us really figuring out what it is that causes colorblindness,” Jay Neitz said when I visited him at his lab. “We're continuing to study lots of different kinds of genes that cause people's vision to go wrong.”
The Neitz Lab is a maze of workspaces in the futuristic UW Medicine building at South Lake Union, a kind of Toys“R”Us of test tubes, lab gadgetry, and various eye exams. Neitz showed me a booklet of “colored dots” vision tests that he helped design, which improve upon the popular Ishihara test for colorblindness. Patients are tasked with discerning dot patterns of various hues from a field of grayish, desaturated dots.
“It turns out that if a colorblind person — or monkey — looks at this, these dots that are colored look exactly like the gray background,” Neitz said.
In this sample from the Neitz Test of Color Vision, a person with red-green color blindness would distinguish no pattern in the dots. A person with normal vision would see a pink square.
The Neitzes gave a digital version of this exam to Sam and Dalton, who Jay Neitz said were born without a gene that codes for the red photopigment. For the first five months after the experimental virus was administered, they were still unable to tell the colored dots apart from the gray ones.
But soon after, the monkeys were passing these tests with flying colors (pun intended). The treatment worked, proving a remarkable payoff to nearly a decade of effort. The discovery was celebrated in publications worldwide, and was named TIME Magazine’s third-best scientific discovery of 2009.
However, what truly stunned the scientific community was how the Neitzes’ research flew in the face of conventional wisdom about color processing. It was thought that primates stop developing critical neural faculties, such as color vision, by adolescence. The Neitzes, however, tested their modified gene on adult squirrel monkeys to successful results. This showed that the adult brains of colorblind monkeys are still capable of accepting new sensory input.
The Neitz Test for Color Vision. Courtesy of the Neitz Lab.
The biological reasons behind this ability remain a mystery. When I asked Jay Neitz, he chuckled and replied, “that’s a $64,000 question.”
What’s important, though, is that the Neitzes’ virus used a human gene for the red photopigment to treat Sam and Dalton’s colorblindness. The neural wiring that squirrel monkeys use to make color distinctions is already quite similar to ours. If the treatment worked on them, Jay Neitz said, chances are it’ll work on us.
But how did the Neitzes pull any of this off in the first place? To answer this, it’s essential to understand how the brain lets us see color.
What is color?
In the 17th century, the study of color was born after Isaac Newton discovered the spectrum of light. However, it was in the 1800s that our modern understanding of color theory got its roots, through two defining insights about how photoreceptors work.
The first, known as the Young-Helmholtz trichromatic theory, posits that the eye’s photoreceptors are sensitive to three distinct wavelength ranges within the visible light spectrum: red, blue, and green.
Steven Buck, a color vision researcher at the UW and longtime colleague of the Neitzes, said the evidence for Young-Helmholtz can be found in the mixing of colored lights.
“We discovered that you can match any color with the right combination of three additional lights,” Buck said. “That said that our color vision was a three-variable system.”
Trichromacy, however, isn’t the full picture. The other major theory from the 19th century was the opponent process theory. Developed by Ewald Hering, this theory says human color vision is governed by two opponent systems: red-green and blue-yellow, combinations that we are unable to see.
“Red and green are opponent to each other,” Buck explained. “If you add red to green you could cancel the greenness, and if you add green to red you could cancel the redness, but you can’t have both at the same place at the same time.”
When light travels to the eye, the lens focuses it onto the retina, a layer of tissue at the back of our eyes containing millions of photoreceptor molecules. Photoreceptors are divided into two categories: rods, which are sensitive to dimmer light and contain only one photopigment, and cones, which respond to brighter light.
An illustration of the rods and cones in the retina.
By Milo Nguyen
The cones contain light-sensitive proteins known as photopigments, which come in three forms: one each for long (red), medium (green), and short (blue) wavelength detection, as described by Young and Helmholtz.
The most common form of colorblindness results in a defect in the gene that produces either the red or green cone photopigment. The result in Hering’s terms is an “inability to perceive the opponency axis of red and green,” according to Buck.
The genetics of color vision
A cone’s wavelength sensitivity is determined by three gene locations, each corresponding to a different size of wavelength.
The gene for short-wavelength pigments does not exist on a sex chromosome, which means both males and females have two copies of this gene.
As for long and mid-wavelength color pigments, the respective genes only exist on X chromosomes, which means they are sex-linked.
“Women have two X-chromosomes, so they get two versions of the long-wavelength cone pigment and the mid-wavelength cone pigment,” Buck explained. “Men only have one X chromosome, so we only get one shot at getting [the red and green pigments].”
This is why red-green color blindness is more common among men than women.
Buck elaborated further: “[The] common inherited form of red-green color deficiency that you find in humans is due to one of those genes — either the mid-wave pigment gene or the long-wave pigment gene — being dysfunctional, absent, mutated, or malformed. And what Jay and Maureen did was put a gene for this missing pigment [in] a virus that’s able to carry that gene into the eye.”
Where the magic happens: making a virus
To create a protein, Jay Neitz said, what normally happens is that DNA is transcribed to messenger RNA, which is then translated into a protein. The transcription of DNA to messenger RNA is catalyzed by an enzyme aptly called “transcriptase.”
He explained that there are non-mammalian organisms that can run this process backwards, using an enzyme called — wait for it — reverse-transcriptase. This enzyme converts RNA back into a complementary DNA molecule, cDNA for short. cDNA molecules are capable of integrating into their host’s genome in the form of a virus.
Neitz recounted how when he and Maureen were getting their Ph.D.s at UC Santa Barbara in the ‘80s, they met Jeremy Nathans — now a renowned molecular geneticist — who was doing his graduate studies at Stanford. Nathans gathered a large sample of freshly-discarded human retinas and managed to use the reverse-transcription process to isolate the cDNAs of the red and green photopigment genes.
Nathans grew a colony of the red-pigment cDNA in E. coli bacteria, and gave the Neitzes a sample to work with. However, they weren’t fully satisfied.
“It turns out there’s variability in the sequences of people’s red genes, and Maureen didn’t like the particular one that [Nathans’ sample] was,” Neitz recounted. “Some of the [red photopigment variants] are shifted a little bit towards green. And she wanted one that was the most perfect shade of red she could find.”
Maureen made modifications to the initial photopigment sample, enabling it to absorb a purer wavelength of red. She and Jay then packaged the resulting sample into a virus, whose DNA could self-replicate inside a cell — to a point. Neitz likened it to a soap bubble.
“[The virus] can make more and more of itself until the cell bursts, and the virus gets released, goes to many places, and becomes an infection,” he said.
For the Neitzes’ virus to work, the infection had to spread to enough cells for the red-pigment gene to be sufficiently expressed. According to Jay Neitz, about 30 percent of the monkeys’ cones were activated at most. As it turns out, this was enough for Sam and Dalton to see red, though Neitz admitted that their vision still fell short of female squirrel monkeys that were born with full color vision.
“But it was enough for them to pass the test,” I pointed out.
“Yes, that’s right,” he said with a laugh.
Soundbite: Neitz - Acquiring the cDNA Sample
––––
In the 10 years since the treatment was first shown to work, the Neitzes have continued tweaking their methodology to better ready their flagship experiment for human testing. However, part of what they’re currently investigating isn’t red, blue, or green, but very gray. It’s a question that’s loomed over the study since the beginning:
Were the monkeys really trichromatic?
Of course, Sam and Dalton could correctly identify red sensory stimuli after their receptors were modified, but actual vision is much more complex than just a linear relationship between receptors and input. The neural wiring that enables those receptors to interact with one another contains millions of different retinal cells that are all uniquely interconnected.
Human color vision results from differential stimulations of this complex receptor network that combine to produce a retinal image. According to Neitz, squirrel monkeys do share many properties of this neural infrastructure with humans, but it’s near-impossible to pinpoint how their vision differs from ours with respect to the subtle, internecine stimulus responses within the retina.
However, those subtleties, as Buck pointed out, make the difference between the functional trichromacy that Sam and Dalton showed, and the full-fledged, human sensation of color vision. There’s no clear answer as to how this experience maps onto the monkeys’ internal visual state.
To Neitz, this dilemma speaks to an age-old problem of consciousness. How can we be sure that what we see is what someone else sees? Or, for that matter, that what we see is what others feel? Color, after all, is not just a visual sensation but a psychological one.
“Red is really a collective experience of what we associate it with,” Neitz said. “But different people have different things that they associate colors with.”
Neitz revealed that people with colorblindness are still able to form common psychological associations with colors they can’t see. For example, people who have normal color vision generally prefer hues of red and blue and dislike brown. For people with colorblindness, however, brown and red appear nearly the same, yet red color associations can still be invoked in brown hues.
“I associate brown with dirt and feces and overcooked spinach,” Neitz said. “But for the colorblind, brown is their favorite color, because often what I call brown, they’d call carmen.”
However, even if these overlapping associations are rooted in emotion, they have important consequences for the visual sensation itself in ways that have yet to be understood.
The final item that Jay Neitz showed me from his lab was an image from a documentary about colorblindness for which he served as a consultant. The top frame is an image of two stuffed animals: a red monkey and the famously green Kermit the Frog. The bottom frame is the image run through a filter to mimic what people with colorblindness see. One of the colorblind subjects in the documentary, he said, couldn’t tell the difference between the two images.
“But when you ask [the subject] what colors these are,” Neitz said, pointing to the bottom frame, “he says that the monkey is red and the frog is green, even though there’s no red or green in the image.
Neitz believes that exploring such mysteries about our vision will yield useful insights on how to translate this research to human subjects. But he and Maureen are a busy pair, and in the time they’ve spent refining this brainchild study of theirs, they’ve also taken on a number of other ambitious projects.
These include a set of glasses that could prevent myopia in adolescents, experiments to cure blue-cone monochromacy in adults, and even attempts to induce tetrachromacy in macaques.
“Ultimately we want to use this kind of gene therapy platform to cure other kinds of eye diseases, not just colorblindness,” Neitz said.
