Fishy vision in the deep sea

11 May 2011

Interview with 

Shozo Yokoyama, Emory University


The Coelacanth, Latimeria chalumnae model in the Oxford University Museum of Natural HistoryAs different species of fish evolved over huindreds of millions of years, their vision became adapted to where they live in the ocean. Coelacanths for example, live about 200m below the surface, and have ditched the pigments in their eyes that would allow them to see red and green light, as they no longer need them - they live in a monochrome world. But how can we track how vision has evolved over time? Well, by using a combination of the fossil record, and genetic techniques, and Shozo Yokoyama explains...

Shozo - When you look at the retina, we have one type of pigment which is expressed in rods which we call rodopsin, and then four kinds of pigments are expressed in cones. For example one type of cone pigment detects either UV or violet, another type of pigment detects red or green, that's other extremes. The other two types absorb light between blue and green. So essentially four kinds of cone pigment we have, in fish.

Sarah - And are all four of those pigments seen in all fish? Or are some of them lost within some groups?

Shozo - Oh yes, their compositions can be very different, depending on where you live.

Sarah - When we look at living species of fish, when we compare the visual pigmets of different groups today, what sort of differences do we see between them?

Shozo - I think different groups of fish do not make much difference, but we can see more similarities among them. But the biggest difference comes from where they live in this case the depth of the ocean. And that effects them a lot.

Sarah - So, if we compare something like a shallow water, a fish that perhaps lives on a coral reef compared to something that lives a very long way down where there's not a lot of light at all?

Shozo - As you go down in the ocean, as you go deeper, then the amount of light will be reduced a lot and the specific light can do down can penetrate, that is because the water absorbs red colours and UV colours. So, depending on where they live, their vision can be significantly effected.

I give you one example is the coelacanth, which live 200m under the ocean. They live close to the Comoros Archipelago in the Indian Ocean and they get very special light. So coelacanth kept only rod pigments, rodopsins in rods, and one kind of pigment in cones, that's all and other genes have been eliminated.

We can study that at the molecular level and the function levels, so we know what they did to live in that environment.

But if you come up in the shallow waters, fish can see practically everywhere from UV to far red. So whereas this example with coelacanth can detect only very narrow strip of light around 480nm.

Sarah - So is it a case that with something like a coelacanth because they live at a depth of the ocean where only a certain amount of the light spectrum gets through to them there's no point in having visual pigments that would allow them to see the red end of the spectrum or the UV end?

Shozo - Yes, that's the way it works. Actually they did that very perfectly so, the coelacanth's ancestor, palaeontological data says they used to live near the shallow water. Then maybe 200 million years ago or so they went down, they started going down to 200m. Their ancestor could see all sorts of light, from UV to red, but today's coelacanth can see only around 480nm, that's around blue. And at the same time, what they did was that they eliminated the gene which detected UV and red, so their gene structure of their visual pigment is very, very simple.

Sarah - And if we look even deeper, so right down into the deep, deep sea, I suppose a lot of the fish species that live down there don't have any need for cones that will see red light, because red light attenuates so much in water that there basically isn't any left down there, so they have no need to see it. But there are some species that produce their own red light, to be able to see prey species. Do we see a specialization of cones in those species to be able to see red light?

Shozo - Actually that's what we'd like to know. There's one fish called the loose jaw, which creates two kinds of light. Many bioluminescence produce 480 but this particular fish produces both 480 and around 700nm, that's red, right.

And the genetics of this one, we really want to clone these genes which we are doing. But a function has not been determined yet from a molecular genetics point of view. So, that's why I cannot tell you too much about it.

Sarah - So I guess it's a case of, if you produce the light you want to make sure that you can see it?

Shozo - Right, I think that's what they would do. Of course another purpose of the bioluminescence is for camouflage, right. So, if you go down in the ocean and look up, then the sky is kind of light, so fish want to make their belly whiter, that's the reason they use bioluminescence.

But if some of the red light, as you mentioned, is to be used for hunting or communication then they have to be able to see it. So to establish it we have to really clone the gene but we're not really there yet.

Find out more

Shozo Yokoyama's lab


Add a comment