The New York Times, April 26, 2010
Edward M. Marcotte is looking for drugs that can kill tumors by stopping blood vessel growth, and he and his colleagues at the University of Texas at Austin recently found some good targets — five human genes that are essential for that growth. Now they’re hunting for drugs that can stop those genes from working. Strangely, though, Dr. Marcotte did not discover the new genes in the human genome, nor in lab mice or even fruit flies. He and his colleagues found the genes in yeast.
“On the face of it, it’s just crazy,” Dr. Marcotte said. After all, these single-cell fungi don’t make blood vessels. They don’t even make blood. In yeast, it turns out, these five genes work together on a completely unrelated task: fixing cell walls.
Crazier still, Dr. Marcotte and his colleagues have discovered hundreds of other genes involved in human disorders by looking at distantly related species. They have found genes associated with deafness in plants, for example, and genes associated with breast cancer in nematode worms. The researchers reported their results recently in The Proceedings of the National Academy of Sciences.
The scientists took advantage of a peculiar feature of our evolutionary history. In our distant, amoeba-like ancestors, clusters of genes were already forming to work together on building cell walls and on other very basic tasks essential to life. Many of those genes still work together in those same clusters, over a billion years later, but on different tasks in different organisms.
Studies like this offer a new twist on Charles Darwin’s original ideas about evolution. Anatomists in the mid-1800s were fascinated by the underlying similarities of traits in different species — the fact that a bat’s wing, for example, has all the same parts as a human hand. Darwin argued that this kind of similarity — known as homology — was just a matter of genealogy. Bats and humans share a common ancestor, and thus they inherited limbs with five digits.
Some 150 years of research have amply confirmed Darwin’s insight. Paleontologists, for example, have brought ambiguous homologies into sharp focus with the discovery of transitional fossils. A case in point is the connection between the blowholes of whales and dolphins and the nostrils of humans. Fossils show how the nostrils of ancestral whales moved from the tip of the snout to the top of the head.
In the 1950s, the study of homology entered a new phase. Scientists began to discover similarities in the structure of proteins. Different species have different forms of hemoglobin, for example. Each form is adapted to a particular way of life, but all descended from one ancestral molecule.
When scientists started sequencing DNA, they were able to find homologies between genes as well. From generation to generation, genes sometimes get accidentally copied. Each copy goes on to pick up unique mutations. But their sequence remains similar enough to reveal their shared ancestry.
A trait like an arm is encoded in many genes, which cooperate with one another to build it. Some genes produce proteins that physically join together to do a job. In other cases, a protein encoded by one gene is required to switch on other genes.
It turns out that clusters of these genes — sometimes called modules — tend to keep working together over the course of millions of years. But they get rewired along the way. They respond to new signals, and act to help build new traits.
In an influential 1997 paper, Sean B. Carroll of the University of Wisconsin, Neil Shubin of the University of Chicago and Cliff Tabin of Harvard Medical School coined a term for these borrowed modules: “deep homology.”
Since then, scientists have gotten a far more detailed look at many examples of deep homology. Dr. Carroll and his colleagues, for example, recently figured out how the spots on a fly’s wing evolved through rewiring modules. A tiny fly called Drosophila guttifera sports a distinctive pattern of 16 polka dots on its wings. Dr. Carroll and his colleagues discovered that the module of genes that sets the location of the spots is the same module that lays out the veins and sensory organs in the wings of many fly species. The module was later borrowed in Drosophila guttifera to lay down dots, too.
Our own eyes are also the product of deep homology. The light-sensing organs of jellyfish seem very different from our eyes, for example, but both use the same module of genes to build light-catching molecules.
Scientists are also discovering that our nervous system shares an even deeper homology with single-celled organisms. Neurons communicate with each other by forming connections called synapses. The neurons use a network of genes to build a complete scaffolding to support the synapse. In February, Alexandre Alié and Michael Manuel of the National Center for Scientific Research in France reported finding 13 of these scaffold-building genes in single-celled relatives of animals known as choanoflagellates.
No one is sure what choanoflagellates use these neuron-building genes for. The one thing that is certain is that they don’t build neurons with them.
Until now, scientists have simply stumbled across examples of deep homology. Dr. Marcotte wondered if it was possible to speed up the pace of discovery.
The evidence for deep homologies, he reasoned, might already be waiting to be found in the scientific literature — specifically, in the hundreds of thousands of studies scientists have conducted on how various genes worked in various species.
Scientists have identified thousands of genes that can give rise to diseases in humans when they mutate. Other researchers have systematically mutated each of the 6,600 genes in yeast and observed how the mutant yeast fare under different conditions. If Dr. Marcotte could analyze data like these, he reasoned, he might find gene modules doing different things in distantly related species.
Dr. Marcotte and his colleagues amassed a database of 1,923 associations between genes and diseases in humans. They added more than 100,000 additional associations between genes and traits in species including mice, yeast and nematode worms.
The scientists then searched for related genes that produced different traits in different species. They discovered, for example, that five genes known to help build blood vessels were closely related to five genes that yeast cells use to fix their cell walls.
Discovering these shared genes then allowed Dr. Marcotte and his colleagues to make new discoveries. Their database had a total of 67 genes that fix cell walls in yeast. If yeast and humans inherited an ancient gene module, we might use related versions of other yeast genes to build blood vessels.
The scientists studied the 62 other wall-fixing yeast genes. To do so, they found related versions in frogs and watched how each one behaved in the developing frog embryo. The scientists discovered that five of the additional yeast genes also made proteins found in developing blood vessels. To see how important these proteins were for building blood vessels, the scientists shut down, one by one, the genes that carried the instructions for each protein, and observed how frog embryos developed.
“We ended up with a dramatic loss of blood vessels,” said John Wallingford, a University of Texas developmental biologist and co-author of the study. Dr. Marcotte wondered if humans might also share modules with much more distantly related organisms: plants. He and his colleagues expanded their database with 22,921 associations between genes and traits scientists have found in the mustard plant Arabidopsis thaliana.
To their surprise, the scientists discovered 48 modules shared by plants and people. “There was a lot of screaming in the halls for that one,” Dr. Marcotte said.
The scientists picked out one particularly strange module shared by plants and people for closer study. In humans, the genes have been linked to a rare genetic disorder called Waardenburg syndrome. It is caused by a disturbance in a group of cells in embryos called neural crest cells. Normally, the neural crest cells crawl through the embryo and form a strip running along the back. They then give rise to nerve cells, pigment-producing cells and some bones of the skull. People with Waardenburg syndrome have symptoms scattered across the parts of the body produced by neural crest cells. They may include deafness; widely spaced eyes; a white forelock of hair; and white patches on their face.
The scientists discovered that two Waardenburg-linked genes matched mustard plant genes for sensing gravity. If these genes are disabled by a mutation, a plant can’t grow upright.
Dr. Marcotte and his colleagues found three more gravity-sensing plant genes in their database. They decided to see if any of the three also played a role in Waardenburg syndrome.
The scientists found that one of the gravity-sensing plant genes became active in the neural crest cells of frog embryos. When they silenced the gene in those neural crest cells, the embryos became deformed.
Dr. Carroll (who also writes a science column for The New York Times) saw the new research as a logical progression from early studies. “It warms our hearts that deep homology is gaining traction like this,” he said.
“This is a very effective way to find human disease genes,” said David Platchetzski of the University of California, Davis, who was not involved in the study. “You can move forward much more quickly.”
Copyright 2010 The New York Times Company. Reprinted with permission.