March 24, 2000 -- The common fruit fly, Drosophila melanogaster, has been the workhorse of biology and genetics laboratories for the past 90 years. Now the entire Drosophila genome has been sequenced by a collaborative effort involving researchers from the Drosophila Genome Project Group, led by Howard Hughes Medical Institute (HHMI) vice president Gerald Rubin at the University of California, Berkeley, and researchers led by J. Craig Venter at the Celera Genomics.
The Drosophila genome sequence was published in the March 24, 2000, issue of Science. The researchers report that they have sequenced 97 to 98 percent of the genome and perhaps 99 percent of the estimated 13,600 genes. The sequence data will be accessible to scientists worldwide through Genbank, the National Institutes of Health genetic sequence database.
In an accompanying editorial in Science, Thomas Kornberg at the University of California, San Francisco, and HHMI investigator Mark Krasnow at Stanford University, report that the Drosophila sequence will be a "critical resource" for research in genetics, biology and medicine.
Over the years, Drosophila has been one of the most influential model systems for geneticists. "The conservation of biological processes from flies to mammals extends the influence of Drosophila to human health," write Kornberg and Krasnow. "When a Drosophila homology of an important but poorly understood mammalian gene is isolated, the arsenal of genetic techniques in the Drosophila system can be applied to its characterization."
The Drosophila sequencing project was launched in 1991 when Rubin and HHMI investigator Allan Spradling at the Carnegie Institution decided, says Rubin, that the time was right to begin a fly genome project. In May 1998, the Berkeley Drosophila Genome Project was one year into a three-year NIH grant and had finished 20 percent of the sequencing, when Rubin was approached by Venter with what Rubin calls "an offer that was too good to turn down."
Venter proposed that his newly-formed company, Celera, would sequence the Drosophila genome free-of-charge using a controversial technique known as whole genome shotgunning. The technique requires sheering the Drosophila DNA into three million random clones with overlapping ends. These clones are then sequenced by automated DNA sequencing machinesÐat Celera, some 300 sequencers, each costing $300,000Ðand then massive computing power is put to work to assemble the complete genome sequence in a process similar to reconstructing a jigsaw puzzle.
Venter formed Celera with backing from PE Corporation (formerly known as Perkin Elmer Corporation), which makes the DNA sequencing machines, as a commercial venture to sequence the human genome by 2001, several years before the date projected for completion by the international Human Genome Project. While promising the data would be made available to researchers, Venter was also betting that Celera could make money by licensing early looks at the sequencing data to the pharmaceutical industry.
The Drosophila genome, says Mark Adams, Celera1s vice president for genome programs, would be "a proof-of-principle" for the whole genome shotgun strategy. "It seemed like a good idea to do a medium-sized organism in which there was extensive scientific interest," he says, "and in which there was already a lot of good information available in terms of map and sequence data that we could use to validate the strategy."
While Rubin says he had some concern about working with Celera, he was delighted by the offer nonetheless. "Anyone who would help me get the Drosophila sequence done and out of the way was my friend," says Rubin. "They were offering to do all this work in a collaborative way and not expecting any money for it."
Celera started the sequencing last April and finished collecting the raw data in early September. "Since then," says Rubin, "we1ve been putting all the pieces together, which is not trivial. It1s the big challenge of the whole genome shotgun approach."
The finished genome already seems to be remarkably revealing. Of the 289 genetic flaws known to cause disease in humans, says Rubin, they have found Drosophila homologues for 60 percent and for 70 percent of the genes involved in human cancers. Among the genes that have already been identified are Drosophila homologues of genes involved in Parkinson1s disease, and the long-sought Drosophila homologue of the p53 tumor suppressor gene, which is implicated in a host of human cancers.
The biggest surprise to come out of the Drosophila sequencing project, says Rubin, is that flies have only twice as many genes as yeast. "Yeast is a simple, single-cell fungus, " says Rubin, "and yet flies only need twice as many genes to make an animal that can fly around without crashing into walls, has tissues, nerves, muscles, memories and other kinds of complicated behaviors like circadian rhythms. The take-home message is that the higher complexity in animals like flies and humans comes without needing a lot of new parts. You can build them with the same parts listÐwith more of the same parts organized togetherÐin much the same way a supercomputer can be built from a bunch of desktop PCs hooked together in parallel."
Rubin sees the genome drastically changing the pace of his research. With less than 15,000 genes in Drosophila, and some 5000 researchers worldwide working on the organism, he says, "that1s one human being for every three genes. If you give those people very efficient tools for figuring out the functions of genes, you can do it in a massively parallel way." Moreover, the full Drosophila sequence allows researchers to look at multiple genes simultaneously to understand the complex signal transduction pathways that regulate cellular processes. "That is where the genome project really comes into play," he says. "It enables us to know all the genes so we can look at all of them at once and see what they1re doing. "
Tak Mak, for instance, director of the Amgen Institute at the University of Toronto, says he has been working to understand the signal transduction pathways involved in cancer formation. "The easiest way to understand that would be some kind of a genetic screen." As a result he has recently dedicated half of his laboratory to Drosophila genetics in anticipation of the publication of the sequence. "It will make Drosophila genetics relatively easy," he says.
Whether the whole genome shotgun technique will work as impressively for the human genome is now the next question. Celera1s Adams says the Drosophila work is obviously encouraging, and that Celera1s human sequencing work has already begun and should "start to look like a genome" toward the end of the year. Rubin says, "It worked better in Drosophila than most people expected it would. I think it will work for humans. But the problems are more complex for humans, so we1ll have to wait and see."
Journal
Science