“In the land of the colorblind,
dichromatic color vision is king.“
– some Hebrew proverb (kinda)

Show Notes

Evolution of Opsins

We made the assumption that the number of unique opsins determines color vision, but is this true? If all we need are photoreceptors to be sensitive to different wavelengths of light, there are ways of tuning a photoreceptor besides changing the opsin protein.

For example, some fish and amphibians can tune their photoreceptors by changing the chromophore from retinal to 3-dehydroretinal. This happens simultaneously in all their photoreceptors, so it doesn’t change the nature of their color vision. However, an organism with one opsin could supposedly use retinal in some photoreceptors and 3-dehydroretinal in the others in order to qualify for color vision despite only having one opsin (Bowmaker 2008).

Another possibility could come from a feature of photoreceptors in birds and reptiles that use oil droplets to “pre-filter” light before it reaches the opsins. As with the retinal example, an organism with one opsin gene could have two classes of photoreceptors with the same opsins but different oil droplets and be able to therefore qualify for color vision (Goldsmith 1990).

We know that organisms use both of these strategies to alter their vision, but we have so far never seen an example of one using them to perceive color (Goldsmith 1990).

On the other hand, there are ways where an organism with multiple different opsins may still not have color vision. It is common in arthropods to have several different opsins sensitive to different parts of the spectrum, but all shoved into a single photoreceptor and therefore providing one combined signal. These are called coexpressed opsins.

In most vertebrates – excluding placental mammals – different cones often join together, sharing one potential. However, those same opsin classes are still expressed in non-double-cones and the doubled cones are believed to not contribute to color vision.

Even when there are distinct photoreceptors with distinct opsins, it’s possible that there are not the right neural mechanics or opponency channels to be able to, essentially “calculate” those colors. Retinal Bipolar Cells are responsible for making ‘comparisons’ on the output of many photoreceptors, and by comparing living species, these are almost as old as the duplications of the ciliary opsins.

Rhodopsin

At the time of the common vertebrate just before the Cambrian Explosion (550MYA), the Rh Opsin had not yet differentiated into Rh1 and Rh2. Only about 500MYA did the Rh differentiate into the Rh1 and Rh2 opsins. Only by that point was the common cone complement – with a suite of 5 vertebrate classes – complete. This includes four cone opsins (LWS, SWS1, SWS2, Rh2) and one rod opsin (Rh1), often called rhodopsin. This would likely lead quickly to the duplex retina (parallel photopic/bright and scotopic/dim vision) that is the hallmark of vertebrates.

Broad Spectrum

Lamb (2013) adds a bit of nuance to this, postulating that coexpressed opsins would not have lasted very long before color vision popped up:

“Currently it is not clear whether the initial driving force for the multiplicity of spectral classes of cone photoreceptor was simply in order to cover more of the ‘visible’ spectrum, given that each opsin absorbs over a relatively narrow region of the spectrum, or whether it was to provide ‘colour vision’. But it seems reasonable to think that, almost as soon new spectral information became available to an organism, it would have been utilized by the nervous system to provide colour information.”

Tunicate

The tunicate actually kinda behaves like a “reverse-butterfly”. It starts as the more “complex-looking”, free-swimming life form during its larval stage, which absolutely does look like something much more resembling a fish – complete with eyes and spinal cord – before transitioning into the mostly-blind, sessile, “rectum” stage that we usually see in pictures.

Flicker Theory

A more recent opinion paper than Maximov (2000) has argued that the Flicker Theory, as described in the video, is less likely than the simpler and less compelling Contrast Theory (Sabbah 2013). The contrast theory states that color vision would simply give the individual more colors, and therefore more chance that the foreground (e.g. predator) and background (e.g. blue ocean) colors would be differentiable. I don’t find the Contrast Theory significantly more convincing, so I presented the more compelling Flicker Theory in the video.

Transcript

On the first day, god created light… but there was nothing there to see it. The first single-celled life spent maybe half a billion years, ignoring photons, chilling, munching on methane and sulfur, the only Pringles flavor available at the time. It was when cyanobacteria evolved photosynthesis as their energy source, that you could say Earth saw its first light-”detecting” organisms.

The next evolutionary step for vision was Phototaxis, a series of simple mechanisms for cells to move towards – or away from – light. Minor improvements occurred throughout the dawn of eukaryotes and multicellular organisms, yada yada yada but it was the evolution of complex eyes which changed animal life forever.

That’s the story, right? The evolution of the eye – from a basic bag of pigment to the complex organ you are using right now. It’s a story more than sufficiently covered in thousands of articles, videos and junior high biology textbooks… but it only covers the evolution of image forming, which is literally only half the picture. The evolution of COLOR VISION is largely disconnected from the evolution of “the eye” and gets very little attention. Hell, the wikipedia page for the evolution of color vision is shorter than the average online recipe for sugar cookies… but unlike the eye, the evolution of color vision cannot so easily be summarized in one diagram and is a little harder to make a pithy youtube video on, but pith… I will bring.

Did you know that color vision existed even before eyes existed? How does that work? And why do we have worse color vision than a Lamprey? And why haven’t we evolved away from colorblindness? Today on Chromaphobe, we’ll be starting a 4 part series on the evolution of color vision, with part 1 covering how our ancestors first developed color vision: the Dawn of Color.

BEHAVIORAL TESTING

If I asked you how long ago the first animal with eyes lived, we could take a gander at the fossil record, and get a pretty good idea, because we know what eyes look like, even when they freaky lookin’. But if I asked you how long ago the first animal with color vision lived, what would you look for in the fossil record? Hell, even with living animals, determining whether they have color vision is stupid hard.

The only conclusive method is behavioral testing, where we determine whether an animal has color vision by making it hungry and only reward it with food when it can solve simple color-based puzzles, hundreds of them. We have performed these Behavioral Tests on everything from dogs to butterflies, but as you can imagine, there are some practical constraints when it comes to testing. For example, a great white shark, or… a T-rex… cuz like… they’re extinct. For now… “Keep absolutely still, his vision is based on color”

Did dinosaurs have vision? Actually, they almost definitely had tetrachromatic color vision, way more colorful than measly human trichromatic color vision… but we don’t know this through fossils or behavioral testing… We know it through phylogenetics.

PHYLOGENETICS

Phylogenetics is the study of the evolutionary family tree – basically looking at how animals are all related to each other. As you can probably guess with a name like phylogenetics, it didn’t exist 200 years ago. Back then, we were still categorizing animals based on their traits and morphology. For example, this Rock Hyrax looks like a rodent, so it must be… a rodent.

After Darwin, and the advent of widespread paleontology, we were now able to link living organisms through their extinct ancestors in the fossil record, and start to hypothesize how they may have evolved. It was missing-link fossils like this that made us realize that the Rock Hyrax’s closest living relative is actually this guy. What can I say? Phylogenetics be crazy.

Dear Rock Hyrax,

That’s sweet that the bullies stopped picking on you when they found out who your older brother was. Just make sure they don’t find out your sister is the manatee, cuz then you’ll never hear the end of it…

Sincerely Protan

Fossils are only half of it though, because phylogenetics got a huge boost when we discovered genetics. Once we could look at the DNA directly, it got a lot easier to establish relationships between species, and even track the evolution of a specific gene over millions of years. So that gives us two tools for understanding the grand evolution of color vision:

• Analysis of the fossil record
• Comparison of Genomes

PHOTORECEPTORS

Color vision is quite simply the ability to differentiate light of different wavelengths. The prerequisite for color vision is therefore to have different photoreceptors that are each TUNED to different wavelengths. In humans, these photoreceptors are the L, M and S cones. Each absorbs different bands of light, because their light catching proteins, the OPSINS, all differ slightly and are sensitive to different wavelengths. However, to qualify for color vision, only two different photoreceptors are needed, and this gives the sufficiently colorful dichromacy.

We have sequenced the genome of – and behaviorally tested – enough animals, that we’re pretty confident that all the animals that CAN detect color, do so in more or less the same way as us, and so we have made 2 assumptions about color vision in animals:

1. Multiple different opsin proteins are required for color vision.
2. Having multiple opsins is sufficient for color vision.

These would be very convenient assumptions, because then all we would have to do is count the number of different opsins in an organism’s retina, or even just in its genome, and we could determine whether the organism had color vision. I go into these assumptions much more in a shownote on my website – link in the description. I’ll be referencing a few more shownotes later with this number. Otherwise you can just take my word for it that the assumptions do hold, at least for vertebrates.

So once we know a vertebrate’s cone complement – that is: the set of opsins it has in its genome – it is a very good indicator of what kind of color vision they have. That means, an organism with one opsin type seems to always have monochromatic vision – i.e. no color vision – and an organism with 4 opsin types seems to always have tetrachromatic vision – i.e. excellent color vision. Furthermore, if we see that two animals have very similar opsins, we can assume that their common ancestor also had those opsins.

So the story of the Evolution of Color Vision as we know it is actually just the story of the Evolution of Opsins.

OPSINS

Opsins are a huge family of proteins that all follow this same general form, absorb light, are ubiquitous in all forms of life and are almost as old as life itself. There are two types of opsins. The first are present in single-celled organisms from every kingdom except animals, mostly in bacteria and archaea, so are known as Microbial Opsins. Microbes have their own interesting color vision story that I will cover in part 4 of this series, but for now, let’s continue with the second type of Opsin: the Animal Opsins, which together enable vision in animals.

Using some pretty basic phylogenetics, we see that Animal Opsins are present in every animals except sponges, which nails down their conception at just over 700MYA. The original Type 2 opsin gene quickly underwent 2 duplications, which meant 3 similar opsin genes mutating independently to be sensitive to different wavelengths of light. Considering our 2 assumptions from earlier, this COULD have made color vision possible in something not much more complex than a sponge.

Unfortunately, these assumptions kinda get thrown out the window when we start looking at pre-vertebrates, and would rely on our inability to behaviorally test modern versions of these ancestors, like this sea squirt… and I think I’d have an easier time testing the great white… So 700 MYA is a loose older limit for color vision.

These three opsins genes would later develop into three groups:

• Ciliary Opsins, which primarily includes the vertebrate visual opsins,
• Rhabdomeric Opsins, which primarily includes the invertebrate visual opsins, and
• Photoisomerase, which is like the Anti-opsin that basically cleans up after the other two.

CILIARY OPSINS

Over the next 150 million years, the initial Ciliary Opsin would itself duplicate and evolve into several different opsin classes. By the time that the common relative of all vertebrates was around 550 mya, there already existed the four Photopsin classes that give color vision to modern vertebrates. There is no theory for why these 4 opsins would have developed together if the organisms didn’t have the neural circuitry to support color vision, so for many, simply their presence is sufficient to assume color vision. Usually, such a bold claim requires a bit of supporting evidence from the fossil record. Luckily, only a few million years later, we’d get the most significant event in the history of fossils.

CAMBRIAN

The Cambrian Explosion, which happened 541million years ago, was a huge acceleration of evolution and the first time animals emerged that you could show a 10 year old and they’d say, “yup, that’s an animal”. Because before that, the most animal-like animal was just a tunicate, which is just a rectum attached to a rock. Imagine this evolutionary leap as an arms race between shrimp and fish that likely started with the development of the first complex eyes, which are visible in this fossil of a cambrian shrimp-like creature.

ACANTHODES

What kind of fossil evidence do we have that these opsins were actually being used for color at this point? A few years ago, a 300 million year old, surprisingly well preserved fish was dug up in Kansas. Examination of its eye tissue revealed individual rod and cone cells in the fish’s rocky eye sockets. These details are usually lost within DAYS of death… much less hundreds of millions of years. So let’s take a look… You see them? Circling these rods and cones reminds me of Schiaparelli seeing canals on the textured surface of Mars that we now know not to exist. Regardless, these cones – according to them – were sufficient proof that this fish had color vision. I found this terribly dubious though, because if I remember anything from the Wizard of Oz, it’s that Kansas – where this fish lived – is notably black and white.

STRUCTURAL COLOR

Actually, there is some additional fossil evidence, not in color vision directly, but in the DISPLAY of color. There are two ways a thing can BE a color.

The common way is through pigmentation, where certain molecules absorb some wavelengths of light and reflect others. Chlorophyll in leaves, melanin in your skin… any kind of paint… these are all pigments.

The other way is through structural color, where the textured pattern of a surface acts similar to a prism to diffract the various wavelengths – and therefore colors – of white light. Butterfly wings, peacock feathers and this iridescent beetle all display color structurally. A simple example of such a color is a diffraction grating, which is just a series of parallel ridges.

It turns out, such a pattern preserves exceptionally well in the fossil record, and we have found what are confidently assumed to be diffraction gratings in fossils from soon after the Cambrian explosion, 515MYA.
The question is, why would this animal have evolved structural color, if there was nothing around to SEE that color. In pigments, the coloration is often a side effect, as in chlorophyll. It’s gonna be green regardless of who’s looking. However, there is not really another reason something would need a diffraction grating besides coloration. The existence of structural color must therefore also infer that there existed some animals that perceived that color, be it a predator, prey or most likely a mate to the colorful animal.

No matter how convinced you are of any of these individual claims, through either phylogenetics or the fossil record, if you were to ask when our human ancestors evolved color vision, it would be somewhere between this fish and this sponge.

WHY?

A common question whenever evolution is brought up is, why? Why would evolution do this, or do this, or do this? We know the broad answer, in that it gives them an advantage over the rest of their species, but when it comes to asking, what advantage does slowly impaling yourself with your own teeth give, the answer is not so clear. Neither is the question of why color vision evolved. It may seem obvious, but complexity doesn’t mean better, otherwise bacteria would have died out long ago. However, according to one paper form 2000, color vision as we know it evolved because of RIPPLES.

We’ve been talking about some really old things in this video, but for this next topic, let’s go reeeeaaaally far back. Windows XP. If you were a preteen like me when Windows XP came out in 2001, you were probably blown away by the sick, photorealistic screensavers. If you were not born yet when windows XP came out, maybe you need to google “screensaver”.

You may notice that shimmer over the surfaces. As light passes from air to water, all of the small waves and ripples on the water surface lead to random refractions of light that create this flickering pattern. A creature chilling on the seabed hoping to not get eaten would therefore experience constant signals of changing luminosity that would be hard to distinguish from an approaching Anomalocaris coming to show down on your eyeballs.

This so-called Flicker Theory postulates that having two opsins essentially enables common mode noise rejection, such that the luminous noise from the ripples can be ignored, a steady color can be seen, and an approaching object can be seen through the flicker. The theory is therefore that prey animals likely first evolved color vision to avoid predators, so they could tell the difference between a trick of the light, and a trilobite.

CONCLUSION

Let’s end this video about 250 million years ago. The mammal lineage has just split off from dinosaurs and the very first mammals are enjoying the tetrachromatic beauty of the world. But soon, one ungrateful mammal would think… “nah, color sucks,” and proceed to get rid of it. Why would mammals devolve their color vision? That’s a question for part 2 of this series, where we’ll take a look at the nocbotnal turtleneck… nocturnal bottleneck.

This is Chromaphobe.