Carl Zimmer - Evolution - The Triumph of an Idea

Exactly what kind of body that common ancestor had is difficult to say. But paleontologists shouldn’t be surprised to unearth some inch-long creature that lived not long before the Cambrian explosion with a wormlike body; a mouth, a gut, and an anus; muscles and a heart; a nervous system organized around a nerve cord and a light-sensing organ; and, finally, some kind of out-growths on its body—if not actual legs or antennae, then perhaps appendages around its mouth to help it eat. It might be the creature that left those anonymous trails among the Ediacarans.

Paleontologists now believe that only after the genetic tool kit was complete could the Cambrian explosion take place. Only then was it possible for dozens of new animal body plans to emerge. Evolution did not build a new network of body-building genes from scratch in the process; it simply tinkered with the original genetic tool kit to build different kinds of legs, eyes, hearts, and other body parts. These animals took on dramatically different appearances, but they still held on to an underlying program for building bodies.

One of the most dramatic examples of this flexibility lies in the origin of our own nervous system. All vertebrates have a nerve cord running down their backs (the dorsal side, as biologists call it), while the heart and digestive tract are on the front (or ventral) side. Insects and other arthropods have an opposite arrangement: the nerve cord is on the ventral side, and the heart and gut are on the dorsal.

These mirror-image body plans inspired a fierce debate between Georges Cuvier and Geoffroy Saint-Hilaire in the 1830s. Cuvier found their anatomy so fundamentally different that he decided vertebrates and arthropods belonged in two completely distinct groups. But Geoffroy claimed that if you transformed the arthropod body plan drastically enough, you would end up with the vertebrate body plan. It turns out that Geoffroy was right, but in a way he couldn’t have imagined. The nervous systems of vertebrates and arthropods are indeed starkly different. But the genes that control their development are the same.

When a vertebrate embryo begins to form, the cells on both the dorsal and the ventral sides have the potential to become neurons. Yet we do not have spinal cords running down our bellies, because the cells on the ventral side of vertebrate embryos release a protein called Bmp-4, which prevents cells from becoming neurons. Gradually Bmp-4 spreads from the ventral cells toward the dorsal side of the embryo, blocking the formation of neurons as it goes.

If Bmp-4 spreads all the way to the other side, no neurons could form at all in a vertebrate embryo. But as the embryo develops, its dorsal cells release a protein that blocks Bmp-4. Known as chordin, it protects the dorsal side of the embryo from Bmp-4, leaving the cells there free to turn into neurons. Eventually they give rise to the spinal cord that runs along a vertebrate’s back.

Compare that sequence of events to what happens in a fruit fly. When a fruit fly embryo first forms, it can also form nerves on both its dorsal and ventral sides. But then a nerve-repressing protein called Dpp is made on its dorsal side, instead of the ventral side where Bmp-4 first appears in vertebrates. As Dpp spreads toward the ventral side of the fly, it is blocked by the protein sog. Protected from Dpp, a fly’s ventral side can form a nerve cord.

These sets of genes not only perform similar jobs in insects and vertebrates, but their sequences are nearly identical. The nerve-blocking gene Dpp and the nerve-blocking gene Bmp-4 are matches, as are their antagonists, sog and chordin. They are so similar, in fact, that if a sog gene from a fly is inserted into a frog embryo, a second spinal cord will start taking form in the frog’s belly. The same genes are building the same structures in insects and frogs, but they’re flipped.

Such similar genes doing such similar jobs must have a common ancestry. John Gerhart at the University of California at Berkeley has proposed how this transformation took place. The first animals with the genetic tool kit grew several small nerve cords running along the sides of their bodies rather than a single big one. These ancestral animals carried a gene that was the ancestor of both chordin and sog, and it promoted the growth of neurons at all the places where a nerve cord was to form in their embryos.

This common ancestor gave rise to all the lineages that appeared during the Cambrian explosion. In the lineage that led to arthropods, the nerve cords all coalesced into a single one running on their ventral side. In vertebrates, the cords all migrated to the back. But the original genes for building nerve cords didn’t disappear; the place where they became active changed. And so, over time, they became the mirror images that so impressed Geoffroy.

Vertebrates acquired more than just spinal cords running down their back during the Cambrian explosion. With some tinkering to their genetic tool kit, they evolved eyes, complex brains, and skeletons. In the process, vertebrates became powerful swimmers and excellent hunters and have remained the dominant predators of the ocean and land ever since.

The oldest known vertebrate fossils—lamprey-like creatures found in China—date back to the midst of the Cambrian explosion, 530 million years ago. In order to understand how those first vertebrates emerged from their ancestors, biologists have studied our closest living invertebrate relative. Known as the lancelet, it’s not a very impressive cousin. It actually looks like a headless sardine pulled from a can. The lancelet starts out life as a tiny larva, floating in shallow coastal waters and swallowing bits of food that drift past it. When it grows to be half an inch long, the adult lancelet burrows into the sand, sticks its head up into the water, and continues to filter-feed.

But as unassuming as the lancelet may look, it shares some key traits with vertebrates. It has slits near the front of its body that correspond to the gills of fish. It has a nerve cord running along its back, which is stiffened’ by a rod called a notochord. Vertebrates have notochords as well, but only while they are embryos. Over time, the notochord withers away as the spinal column grows larger.

In other words, certain pieces of the vertebrate body plan had already evolved in the common ancestor of lancelets and vertebrates. Yet lancelets also lack much of the anatomy that make vertebrates so distinct. They have no eyes, for example, and their nerve cord ends in a tiny bump, not in a proper mass of neurons one would, at first glance, call a brain.

But it’s possible to see precursors of brains and eyes in lancelets. The lancelet can detect light with a pit lined with photosensitive cells, and these cells are wired up like a retina in a vertebrate eye and connected to the front of the nerve cord in much the same way as our eyes are to our brains. That tiny bump at the front of the lancelet’s nerve cord may consist of only a few hundred neurons (our brains have 100 billion), but it is divided into simplified versions of parts of our own vertebrate brains.

The similarity between the lancelet’s nerve cord and vertebrate brains extends to the genes that build them. Hox genes and other master-control genes that map out the vertebrate brain and spinal cord do the same job in the lancelet embryo, in almost precisely the same head-to-tail order. In the cells of the developing lancelet eyespot, the genes are the same as those that build a vertebrate eye. It’s a safe bet that the common ancestor of lancelets and vertebrates had the same genes for constructing the same basic brain.

Once the ancestors of vertebrates and lancelets branched apart, our ancestors went through an extraordinary evolutionary experience. Lancelets have 13 Hox genes, but vertebrates have four sets of those genes, each arranged in the same head-to-tail order. Mutations must have caused the original set of Hox genes to be duplicated. After they evolved into four sets, the new genes met various fates. Some of them went on carrying out the jobs of the original Hox gene. But other copies of Hox genes evolved until they were able to help shape the vertebrate embryo in new ways.