by Neil Shubin
Our camera-like eye is common to every creature with a skull, from fish to mammals. In other groups of animals we find different eyes, ranging from simple patches of cells specialized to detect light, to eyes with compound lenses such as those found in flies, to primordial versions of our own eye. The key to understanding the history of our eyes is to understand the relationship between the structures that make our camera-eye and those that make all the other kinds of eyes. To do this, we will study the molecules that gather light, the tissues we use to see, and the genes that make the whole thing.
Eyes come into focus: from primitive light-capturing devices in invertebrates to our camera-type eye with a lens. As eyes evolve, visual acuity increases.
LIGHT-GATHERING MOLECULES
The really important work in the light-gathering cells happens inside the molecule that actually collects light. When this molecule absorbs light, it changes shape and breaks up into two parts. One part is derived from vitamin A, the other from a protein known as an opsin. When the opsin breaks off, it initiates a chain reaction that leads to a neuron sending an impulse to our brain. We use different opsins to see in black and white and in color. Just as an inkjet printer needs three or four inks to print in color, we need three light-gathering molecules to see in color. For black-and-white vision, we use only one.
These light-gathering molecules change shape in the light, then recharge in the dark and go back to their normal state. The process takes a few minutes. We all know this from personal experience: go from a bright place into a dark room and it is virtually impossible to see faint objects. The reason is that the light-gathering molecules need time to recharge. After a few minutes, vision in the dark returns.
Despite the stunning variety of photoreceptor organs, every animal uses the same kind of light-capturing molecule to do this job. Insects, humans, clams, and scallops all use opsins. Not only can we trace the history of eyes through differences in the structure of their opsins, but we have good evidence that we can thank bacteria for these molecules in the first place.
Essentially, an opsin is a kind of molecule that conveys information from the outside of a cell to the inside. To pull off this feat, it needs to carry a chemical across the membrane that encircles a cell. Opsins use a specialized kind of conductor that takes a series of bends and loops as it travels from the outside to the inside of the cell. But this twisted path the receptor takes through the cell membrane is not random—it has a characteristic signature. Where else is this twisted path seen? It is identical to parts of certain molecules in bacteria. The very precise molecular similarities in this molecule suggest a very ancient property of all animals extending all the way to our shared history with bacteria. In a sense, modified bits of ancient bacteria lie inside our retinas, helping us to see.
We can even trace some major events in the history of our eyes by examining opsins in different animals. Take one of the major events in our primate past, the development of rich color vision. Recall that humans and our closest ape relatives, the Old World monkeys, have a very detailed kind of color vision that relies on three different kinds of light receptors. Each of these receptors is tuned to a different kind of light. Most other mammals have only two kinds of receptors and so cannot discriminate as many colors as we can. It turns out that we can trace the origin of our color vision by looking at the genes that make the receptors. The two kinds of receptors most mammals have are made by two kinds of genes. Of our three receptor-making genes, two are remarkably like one of those in other mammals. This seems to imply that our color vision began when one of the genes in other mammals duplicated and the copies specialized over time for different light sources. As you’ll remember, a similar thing happened with odor receptor genes.
This shift may be related to changes in the flora of the earth millions of years ago. It helps to think what color vision was likely good for when it first appeared. Monkeys that live in trees would benefit because color vision enabled them to discriminate better among many kinds of fruits and leaves and select the most nutritious among them. From studying the other primates that have color vision, we can estimate that our kind of color vision arose about 55 million years ago. At this time we find fossil evidence of changes in the composition of ancient forests. Before this time, the forests were rich in figs and palms, which are tasty but all of the same general color. Later forests had more of a diversity of plants, likely with different colors. It seems a good bet that the switch to color vision correlates with a switch from a monochromatic forest to one with a richer palette of colors in food.
TISSUES
Animal eyes come in two flavors; one is seen in invertebrates, the other in vertebrates, such as fish and humans. The central idea is that there are two different ways of increasing the light-gathering surface area in eye tissue. Invertebrates, such as flies and worms, accomplish this by having numerous folds in the tissue, while our lineage expands the surface area by having lots of little projections extending from the tissue like tiny bristles. A host of other differences also relate to these different kinds of designs. Lacking fossils at the relevant phase of history, it would seem that we would never be able to bridge the differences between our eyes and those of invertebrates. That is, until 2001, when Detlev Arendt thought to study the eyes of a very primitive little worm.
Polychaetes are among the most primitive living worms known. They have a very simple segmented body plan, and they also have two kinds of light-sensing organs: an eye and, buried under their skin, a part of their nervous system that is specialized to pick up light. Arendt took these worms apart both physically and genetically. Knowledge of the gene sequence of our opsin genes and the structure of our light-gathering neurons gave Arendt the tools to study how polychaetes are made. He found that they had elements of both kinds of animal photoreceptors. The normal “eye” was made up of neurons and opsins like the eye of any invertebrate. The tiny photoreceptors under the skin were another matter altogether. They had “vertebrate” opsins and cellular structure even with the little bristle-like projections, but in primitive form. Arendt had found a living bridge, an animal with both kinds of eyes, one of which—our kind—existed in a very primitive form. When we look to primitive invertebrates, we find that the different kinds of animal eyes share common parts.
GENES
Arendt’s discovery leads to yet another question. It is one thing for eyes to share common parts, but how can eyes that look so different—such as those of worms, flies, and mice—be closely related? For the answer, let us consider the genetic recipe that builds eyes.
At the turn of the twentieth century, Mildred Hoge was recording mutations in fruit flies when she found a fly that had no eyes whatsoever. This mutant was not an isolated case, and Hoge discovered that she could breed a whole line of such flies, which she named eyeless. Later, a similar mutation was discovered in mice. Some individuals had small eyes; others lacked whole portions of the head and face, including their eyes. A similar condition in humans is known as aniridia; affected individuals are missing large pieces of their eyes. In these very different creatures—flies, mice, and humans—geneticists were finding similar kinds of mutants.
A breakthrough came in the early 1990s, when laboratories applied new molecular techniques to understand how eyeless mutants affected eye development. Mapping the genes, they were able to localize the bits of DNA responsible for the mutations. When the DNA was sequenced, it turned out that the fly, mouse, and human genes responsible for eyelessness had similar DNA structures and sequences. In a very real sense, they are the same gene.
What did we learn from this? Scientists had identified a single gene that, when mutated, produced creatures with small eyes or no eyes at all. This meant that the normal version of the gene was a major trigger for the formation of eyes. Now came the chance to do experiments to ask a whole other kind of question. What happens when we mess with the gene, turning it on and off in the wrong places?
Flies were an ideal subject for this work.
During the 1980s, a number of very powerful genetic tools were developed through work on flies. If you knew a gene, or a DNA sequence, you could make a fly lacking the gene or, the reverse, a fly with the gene active in the wrong places.
Using these tools, Walter Gehring started playing around with the eyeless gene. Gehring’s team was able to make the eyeless DNA active pretty much anywhere they wanted: in the antenna, on the legs, on the wings. When his team did this, they found something stunning. If they turned on the eyeless gene in the antenna, an eye grew there. If they turned on the eyeless gene on a body segment, an eye developed there. Everywhere they turned on the gene, they would get a new eye. To top it all off, some of the misplaced eyes showed a nascent ability to respond to light. Gehring had uncovered a major trigger in the formation of eyes.
Gehring didn’t stop there; he began swapping genes between species. They took the mouse equivalent of eyeless, Pax 6, and turned it on in a fly. The mouse gene produced a new eye. And not just any eye—a fly eye. Gehring’s lab found they could use the mouse gene to trigger the formation of an extra fly eye anywhere: on the back, on a wing, near the mouth. What Gehring had found was a master switch for eye development that was virtually the same in a mouse and a fly. This gene, Pax 6, initiated a complex chain reaction of gene activity that ultimately led to a new fly eye.
We now know that eyeless, or Pax 6, controls development in everything that has eyes. The eyes may look different—some with a lens, some without; some compound, some simple—but the genetic switches that make them are the same.
When you look into eyes, forget about romance, creation, and the windows into the soul. With their molecules, genes, and tissues derived from microbes, jellyfish, worms, and flies, you see an entire menagerie.
CHAPTER TEN
EARS
The first time you see the inside of the ear is a letdown: the real machinery is hidden deep inside the skull, encased in a wall of bone. Once you open the skull and remove the brain, you need to chip with a chisel to remove that wall. If you are really good, or very lucky, you’ll make the right stroke and see it—the inner ear. It resembles the kind of tiny coiled snail shell you find in the dirt in your lawn.
The ear may not look like much, but it is a wonderful Rube Goldberg contraption. When we hear, sound waves are funneled into the outside flap, the external ear. The sound waves enter the ear and make the eardrum rattle. The eardrum is attached to three little bones, which shake along with it. One of these ear bones is attached to the snail-shell structure by a kind of plunger. The shaking of the ear bone causes the plunger to go up and down. This causes some gel inside the snail shell to move around. Swishing gel bends nerves, which send a signal to the brain, which interprets it as sound. Next time you are at a concert, just imagine all the stuff flying around in your head.
This structure allows us to distinguish three parts to the ear: external, middle, and inner. The external ear is the visible part. The middle contains the little ear bones. Finally, the inner ear consists of the nerves, the gel, and the tissues that surround them. These three components of ears enable us to structure our discussion in a very convenient way.
Of the three parts of our ear—the outer, middle, and inner—the inner ear is the most ancient and the part that controls the nerve impulses sent to the brain.
The part of the ear we can see, the flap on which we hang our glasses, is a relatively new evolutionary addition to bodies. Confirm this on your next trip to the aquarium or zoo. How many sharks, bony fish, amphibians, and reptiles have external ears? The pinna—the flap of the external ear—is found only in mammals. Some amphibians and reptiles have visible external ears, but they have no pinna. Often the external ear is only a membrane that looks like the top of a drum.
The elegance of our connection to sharks and bony fish is revealed when we look inside our ears. Ears might seem an unlikely place for a human-shark connection, especially since sharks don’t have ears. But the connection is there. Let’s start with the ear bones.
THE MIDDLE EAR—THE THREE EAR BONES
Mammals are very special. With hair and milk-producing glands, we can easily be distinguished from other creatures. It surprises most people to learn that some of the most distinctive traits of mammals lie inside the ear. The bones of the mammalian middle ear are like those of no other animal: mammals have three bones, whereas reptiles and amphibians have only one. Fish have none at all. Where did our middle ear bones come from?
Some anatomy: recall that our three middle ear bones are known as the malleus, incus, and stapes. As we’ve seen, each of these ear bones is derived from the gill arches: the stapes from the first arch, and the malleus and incus from the second arch. It is here that our story begins.
In 1837, the German anatomist Karl Reichert was looking at embryos of mammals and reptiles to understand how the skull forms. He followed the gill arches of different species to understand where they ended up in the various skulls. As he did this again and again, he found something that appeared not to make any sense: two of the ear bones in the mammals corresponded to pieces of the jaw in the reptiles. Reichert could not believe his eyes, and his monograph reveals his excitement. As he describes the ear-jaw comparison, his prose departs from the normally staid description of nineteenth-century anatomy to express shock, even wonderment, at this discovery. The conclusion was inescapable: the same gill arch that formed part of the jaw of a reptile formed ear bones in mammals. Reichert proposed a notion that even he could barely believe—that parts of the ears of mammals are the same thing as the jaws of reptiles. Things get more difficult when we realize that Reichert proposed this several decades before Darwin propounded his notion of a family tree for life. What does it mean to call structures in two different species “the same” without a notion of evolution?
Much later, in 1910 and 1912, the German anatomist Ernst Gaupp picked up on Reichert’s work and published an exhaustive study on the embryology of mammalian ears. Gaupp provided more detail and, given the times, interpreted Reichert’s work in an evolutionary framework. Gaupp’s story went like this: the three middle ear bones reveal the tie between reptiles and mammals. The single bone in the reptilian middle ear is the same as the stapes of mammals; both are second-arch derivatives. The explosive bit of information, though, was that the two other middle ear bones of mammals—the malleus and the incus—evolved from bones set in the back of the reptilian jaw. If this was indeed the case, then the fossil record should show bones shifting from the jaw to the ear during the origin of mammals. The problem was that Gaupp worked only on living creatures and didn’t fully appreciate the role that fossils could play in his theory.
Beginning in the 1840s a number of new kinds of fossil creatures were becoming known from discoveries in South Africa and Russia. Often abundantly preserved, whole skeletons of dog-size animals were being unearthed. As they were discovered, many of them were crated and shipped to Richard Owen in London for identification and analysis. Owen was struck that these creatures had a mélange of features. Parts of their skeleton looked reptile-like. Other parts, notably their teeth, looked like mammals. And these were not isolated finds. It turns out that these “mammal-like reptiles” were the most common skeletons being uncovered at many fossil sites. Not only were they very common, there were many kinds. In the years after Owen, these mammal-like reptiles became known from other parts of the world and from several different time periods in earth history. They formed a beautiful transitional series in the fossil record between reptile and mammal.
Until 1913, embryologists and paleontologists were working in isolation from one another. At this time, the American paleontologist W. K. Gregory, of the American Museum of Natural History, saw an important link between Gaupp’s embryos and the African fossils. The most reptilian of the mammal-like reptiles had only a single bone in its middle ear; like other reptiles, it had a jaw composed of many bones. Something remarkable was revealed as Gregory looked at the successively more mammalian mammal-like rep
tiles, something that would have floored Reichert had he been alive: a continuum of forms showing beyond doubt that over time the bones at the back of the reptilian jaw got smaller and smaller, until they ultimately lay in the middle ear of mammals. The malleus and incus did indeed evolve from jawbones. What Reichert and Gaupp observed in embryos was buried in the fossil record all along, just waiting to be discovered.
Why would mammals need a three-boned middle ear? This little linkage forms a lever system that allows mammals to hear higher-frequency sounds than animals with a single middle ear bone. The origin of mammals involved not only new patterns of chewing, as we saw in Chapter 4, but new ways of hearing. In fact, this shift was accomplished not by evolving new bones per se, but by repurposing existing ones. Bones originally used by reptiles to chew evolved in mammals to assist in hearing.
So much for the malleus and incus. Where, though, does the stapes come from?
If I simply showed you an adult human and a shark, you would never guess that this tiny bone deep inside a human’s ear is the same thing as a large rod in the upper jaw of a fish. Yet, developmentally, these bones are the same thing. The stapes is a second-arch bone, as is the corresponding bone in a shark and a fish—the hyomandibula. But the hyomandibula is not an ear bone; recall that fish and sharks do not have ears. In our aquatic cousins, this bone is a large rod that connects the upper jaw to the braincase. Despite the apparent differences in the function and shape of these bones, the similarities between the hyomandibula and the stapes extend even to the nerves that supply them. The key nerve for the functioning of both bones is the second-arch nerve, the facial nerve. We thus have a situation where two very different bones have similar developmental origins and patterns of innervation. Is there an explanation for this?
