Fly Genome Creates a Buzz.

Scientists try to make sense of an insect's myriad genes

Gerald M. Rubin expects to eat and drink well in the coming months, thanks to the gambling nature of some of his colleagues. A Howard Hughes Medical Institute investigator at the University of California, Berkeley, the biologist laid a few wagers on a subject he knows extraordinarily well: the common fruit fly, Drosophila melanogaster.

The bets, which center on the total number of genes used by the winged animal, include one for "a very nice bottle of wine" and one for a lavish dinner for six, says Rubin. Several years ago, he and a colleague predicted that the fruit fly would have around 12,000 genes. In the bets, Rubin shot a little higher but stuck with numbers that most fly scientists considered low. For the dinner, he bet that the fly genome contains less than 15,000 genes, while one opponent predicted between 15,000 and 20,000 genes and another said the fly has even more. For the wine, Rubin laid odds that D. melanogaster would come in with fewer than 18,000 genes.

Although one of the dinner wagerers still holds out hope, Rubin expects to collect on both bets. He and his colleagues finished sequencing the gene-rich areas of the fruit fly genome last year and recently reported that the insect appears to have between 13,000 and 14,000 genes (SN: 2/26/00, p. 132).

In addition to winning Rubin his bets, this completion of the fruit fly genome offers scientists their first comprehensive look at the genes of an animal with a true central nervous system, a complex body plan, a rudimentary immune system, and sophisticated behaviors. It should also help illuminate the contents of the human genome, which may itself be completed in a matter of weeks.

The fruit fly genome is "an incredible springboard to the future," says Kent G. Golic of the University of Utah in Salt Lake City.

Still, sequencing an entire genome is more an enabling piece of science than an explanatory one. The information in the genome sequence raises many more questions than it immediately answers, caution scientists. Biologists don't even have clues about the function of more than half the genes they've identified. Now that the grunt work of unraveling the fly's DNA has ended, researchers will turn to the more interesting challenge of figuring out what those insect genes do.

"It's a great tool to have the Drosophila sequence, but it doesn't tell us how a fruit fly works," stresses Rubin.

Given that caveat, what have biologists gleaned from the fruit fly genome in the little time they've had it? The low number of genes has startled most scientists--other than Rubin.

Investigators identified the approximately 14,000 genes by a variety of methods, including scanning the genome for DNA sequences similar to known genes in the fly and other organisms (SN: 4/29/00, p. 284). The fly tally is only about double the count for the single-cell yeast Saccharomyces cerevisiae. Considering that the lowly roundworm Caenorhabditis elegans makes use of more than 18,000 genes (SN: 12/12/98, p. 372), many fruit fly scientists had assumed that their favorite organism would require a far bigger genetic toolbox.

"It is surprising that flies do have fewer genes than worms because the prejudice is that worms are much simpler organisms," remarks Thomas C. Kaufman of Indiana University in Bloomington.

Perhaps, he continues, scientists shouldn't be so amazed. The fruit fly genome is only the second completed genome for a multicellular animal and only the third for a eukaryotic organism (the yeast S. cerevisiae, like the worm and fly, packages its genes inside a nucleus). "We're pretty naive about the structure of genomes at this point," says Kaufman.

They're naive about a lot more than that. As a first step in identifying a gene's role, scientists usually analyze the gene's DNA sequence to predict the structure of the protein it encodes. They then try to match parts of that molecule to features on proteins that they already know something about. Through this strategy, they can often tentatively decide that a gene encodes a cell-adhesion molecule, signaling protein, or some other type of molecule. But for 8,884 of the fruit fly's predicted proteins, scientists can't assign a function because the molecules don't match anything in protein databases.

The genetic inferiority of the fly, compared with the worm, may not be as hard to figure out. Many of the worm's individual genes have apparently expanded into large gene families during evolution. Consider collagens, fibrous proteins that play a role in connective tissues. Flies have just a few genes encoding subtly different collagens, whereas the worm has around 170. It's as if the fly can survive with one or two Phillips screwdrivers, while the worm stockpiles dozens with tips of slightly different sizes.

"The main take-home message [of the fruit fly genome] is that you don't build complexity in biological systems by increasing the gene number," says Rubin. "Counting gene number is something people like to do, but I don't think it has a lot of biological meaning."

Rubin argues that rather than overall gene number, the number of distinct gene families gives a better comparison between the fly and worm. In that, the two organisms turn out a bit more even. The fly has 8,065 distinct gene families, while the worm has 9,453, Rubin and his colleagues report in one of several articles on the fly genome in the March 24 SCIENCE.

The differing sizes of various worm and fly gene families have offered biologists several puzzles. For example, why does the fruit fly genome harbor 199 genes encoding proteins called trypsin-like peptidases? In contrast, the worm has 7, and yeast carries just a single such gene.

Although some fly gene families are unexpectedly crowded, others are surprisingly barren. Take the genes for olfactory receptors, the cell-surface proteins that animals use to detect smells. The worm has about a thousand such genes, and people have a similar number. Yet the fruit fly genome seems to hold just 50 to 100 olfactory-receptor genes.

With this genetic disparity now revealed, biologists can seek an explanation. "Worms are dependent on olfaction because they have no visual system, whereas flies have a highly developed visual system," suggests Rubin. "Maybe olfaction is less important to the fly."

Despite its seeming genetic simplicity, the fruit fly shares many cellular and physiological features with people, and fly investigators continue to seek ways to use the insect to model human illnesses. In the March 23 NATURE, for example, Mel B. Feany and Welcome W. Bender of Harvard Medical School in Boston describe creating a mutant fly strain that develops a neurodegenerative disease mimicking Parkinson's disease in many ways.

To quantify how the fly might illuminate human ailments, the investigators that sequenced the insect genome scanned it for fly versions of 289 human genes that play a role in a wide variety of diseases, ranging from cancers to diabetes. They found that 177 genes, 61 percent, had insect counterparts.

"It improves the argument that Drosophila is a good model for human health," says Golic.

Many scientists are eager to marry the massive amount of information within the fruit fly genome to a recently developed technology called DNA chips. Essentially microchips covered with thousands of distinct single-stranded DNA molecules, these devices enable researchers to easily determine the activity of many genes within a tissue sample (SN: 3/8/97, p. 144).

A report in the Dec. 10, 1999 SCIENCE illustrated the promise of DNA chips in fruit fly research. David S. Hogness of the Stanford University School of Medicine and his colleagues constructed a chip that monitors some 4,500 fly genes. Using the chip to examine the insect at various stages of its metamorphosis into a fly, they found that more than 10 percent of the genes significantly increased or decreased their action during this dramatic process. The chip, for example, documented the activity of genes involved in creating and breaking down muscle.

In a similar way, Thomas Brody of the National Institute of Neurological Disorders and Stroke in Bethesda, Md., is building a DNA chip that monitors the 1,500 or so genes active in the fly brain. And Louisa Wu of the University of Maryland Biotechnology Institute in College Park plans to use DNA chips to address how the insect's immune system functions.

"We want to see what genes are turned on or turned off in response to different kinds of infections," she says. "Ideally, it would be nice to find some novel antibiotics" that the fly makes within its body.

Although DNA chips encompassing fly genes will unleash a torrent of data, some scientists caution that the information won't make sense without years of follow-up research. "I have a certain amount of skepticism about DNA chips," says Gary Karpen of the Salk Institute for Biological Studies in La Jolla, Calif. "It's still going to come down to individual labs studying individual genes."

In the rush to use the fruit fly genome to study human disease and basic aspects of cell biology, it's easy to forget that the Drosophila genome also represents the first full genetic menu available for an insect, notes Wu. As such, she says, it should provide insight into the mosquitoes, locusts, and other insects that transmit many human diseases or destroy crops. The fruit fly genome, for example, may reveal insect genes or proteins that novel pesticides could target.

In a remarkable stroke of fortune, fly scientists appear to have developed an important new research technique just as they've announced completion of the genome. For years, they've been able to insert genes and random mutations into fruit flies' chromosomes, but investigators haven't known how to deactivate a specific gene. Mouse researchers have had such a knockout technology for the past decade and have exploited it to discover genes' functions and create mouse strains with humanlike illnesses.

At a recent Drosophila research conference, Golic described a strategy that he's developed to enable fly scientists to introduce a mutation into a chosen gene. The biologist reported his success with a single gene, but he declined to discuss details of the work because it has not yet been published.

Nevertheless, fly scientists are abuzz with the news and anxious to see if Golic's technique deactivate each of the genes revealed by the genome's sequencing. "Talk about timing!" exclaims Kaufman.

"It's exciting," adds Rubin. "That was the one thing that wasn't in our toolbox."

It's amazing how quickly fly scientists have advanced their trade, notes Karpen. The tradition of fruit fly research began in 1910 when Thomas Hunt Morgan chose D. melanogaster as the test subject for his studies on heredity and identified the first fly genetic mutant, a strain with white rather than red eyes. With that humble start, Morgan and his disciples began to unravel the mysteries of inheritance and the nature of genes.

While the sequencing of the fly genome may mark the end of one stage of these scientists' quest, it also signals the beginning of the next phase of fruit fly research. A century from now, biologists may have trouble imagining doing their work without knowing the full genomes of whatever species they study. "As much as people look back and laugh at pre-DNA biologists, we're going to be laughed at more robustly," predicts Karpen.