22 April 2002 -- The most comprehensive genetic study to date concerning the evolutionary relationships among the three animal species whose genes have been completely sequenced--the human, the fruit fly, and the nematode worm--has determined that the human species is more closely related to the fruit fly than to the nematode. "We compared 100 genes that are common among these three species--the largest data set ever used to address this question--and obtained a result that is unambiguous," says S. Blair Hedges, an evolutionary biologist at Penn State, whose research team includes other scientists from Penn State and Japan.

"Cost estimates for acquiring the human genome alone range between 300 million and 3 billion," Hedges comments. "After spending all that money and effort, we now should at least be able to know for sure how these animals are related." These three species, which were singled out for the extensive genome effort, each represent much larger groups of animals: vertebrates are represented by humans, arthropods are represented by the fruit fly, and nematodes are represented by one species of nematode.

The results of the study by Hedges and his colleagues overturn a popular recent hypothesis, based primarily on the study of a single gene, and have important implications for research in fields such as medicine and developmental biology. The study, published in the current edition of the web-based journal BMC Evolutionary Biology, also is expected to impact the content of biology textbooks.

"About five years ago, the journal Nature published an analysis of one gene common to these three species that inexplicably persuaded many scientists to abandon the classic hypothesis of their relationships, which was based on the long-standing method of comparing structural similarities," Hedges says. The new hypothesis, named "Ecdysozoa," argued that fruit flies and nematodes are more closely related to each other than to humans. The classic hypothesis, named "Coelomata," argued instead that humans and fruit flies are more closely related to each other than to the nematodes.

The name "Ecdysozoa" alludes to the fact that insects (and other arthropods) and nematodes both shed their outer covering, a process called ecdysis. The name "Coelomata" alludes to the presence of a true body cavity (coelom) in humans and fruit flies, whereas nematodes have a false body cavity (pseudocoelom).

"At first a few developmental biologists began using the new Ecdysozoa hypothesis of species relationships in studies involving the HOX genes, which control the development of body parts like fingers, toes, and wings. Soon afterwards, a lot of people started using it and now it is included in introductory biology textbooks even though it hasn't been adequately tested," Hedges says. "Many scientists are surprised by this uncharacteristically rapid abandonment of the long-standing Coelomata hypothesis and acceptance of the new Ecdysozoa hypothesis without the intense scrutiny that is typical in science," Hedges says. How textbooks arrange the relationships among HOX genes is particularly important, the researchers say, because it has an effect on how crucial events in the development of animals are understood by future generations of scientists.

To resolve the controversy, Hedges and his colleagues decided to apply that much-needed scrutiny by tapping into the wealth of data now available in the completely sequenced genomes of the three species. "We could have looked at more than the 100 genes we selected, but it takes a long time to inspect the genes as carefully as we wanted to in order to avoid the errors that can creep into automated analyses," comments Jaime E. Blair, a graduate student at Penn State and first author of the paper. "You want to be sure, for example, that you are comparing the nematode version of gene A with the human version of gene A with the fruit fly version of gene A--not with the fruit fly version of gene B."

The researchers also rigorously tested the genes to eliminate a number of possible biases that could occur by chance, or by natural selection, or for other reasons. "We did every relevant analysis known to molecular evolutionists, and took every known precaution to account for intrinsic biases in the data, "Blair says. Every test the researchers performed supported the classic Coelomata hypothesis, not the Ecdysozoa hypothesis. "With 100 genes, we could perform analyses that you just can't do with a single gene, which simply doesn't give you enough data," she explains.

Blair says most individual genes don't contain enough amino-acid molecules to provide the amount of data necessary for producing statistically valid relationship diagrams, known as trees. Working with 100 genes allowed the researchers to work with groups of 10 or 20 genes stuck together--the equivalent of several thousand amino acids--which they say is enough for the statistical analyses required to obtain a significant result.

"We also ordered the groups from the 10 slowest-evolving genes to the 10 fastest-evolving genes, which allowed us to test the claim of the Ecdysozoa supporters that the slowest-evolving genes should yield the correct tree because the faster-evolving genes incorporate more opportunities for mistakes and biases," says Blair. "What we found, instead, was that the slowest-evolving genes gave us the classic Coelomata tree, not the Ecdysozoa tree, at 100-percent significance. In addition, we analyzed the species of nematode designated by the Ecdysozoa supporters as slow-evolving and found that it, too, rejects the Ecdysozoa hypothesis," explains Blair. "Even the original emphasis on shedding in Ecdysozoans has been misleading, because the structure that is shed in arthropods is completely different from that in nematodes, and some vertebrates such as snakes also shed their skin."

The nematode and fruit fly are among the most widely used model organisms of medical and genetics researchers because they can be bred very easily and produce new generations quickly. "A lot of our understanding of human medicine is based on these species because we can do experiments with them that you wouldn't do with humans," Hedges says. Scientists' understanding of how those species are related will determine how that history is reconstructed and whether a mutational change is interpreted as relatively recent or ancient. "If you assume that the nematode and fruit fly are more closely related than either is to human, as under the Ecdysozoa hypothesis, we now know that you would probably be making a mistake in understanding the history of the mutations," Hedges explains.

The study impacts any field that is concerned with the inheritance of traits in major groups of animals, Hedges says. "Besides medicine and biology, these results also affect fields such as astrobiology, where scientists need to know how complex organisms developed on Earth to better understand how life develops elsewhere in the Universe." At the same time, he cautions, "we could be completely wrong. I prefer to view our result as the best supported, based on the weight of the evidence, rather than as a proven fact. It is always better to keep an open mind about these things, not to become married to one hypothesis or another, and to let the data speak for themselves."

This research was supported by the National Aeronautics and Space Administration (NASA) Astrobiology Institute and the U. S. National Science Foundation. In addition to Hedges and Blair, researchers involved in the study include Kazuho Ikeo and Takashi Gojobori, of the National Institute of Genetics in Japan.

CONTACTS: Barbara K. Kennedy (PIO), 814-863-4682, science@psu.edu

Source: Penn State University