The parasite is one of a group of bacteria called Wolbachia that has been found to live inside a range of insects, from wasps to cockroaches, disrupting the insects' sex lives to increase the number of female offspring and suppress or eliminate males.

Now, Thomas W. Cline, professor of molecular and cell biology at UC Berkeley, and graduate student Diana J. Starr, have discovered that the bacterium Wolbachia pipientis allows a sterile female fruit fly to lay eggs, circumventing a genetic mutation in a gene that is the key to determining the sex of offspring.

"It's stunning that we've created in the lab, without intending to, the same situation that people have found in the wild," Cline said. "This is very exciting, because now we can study how the bacterium interacts with the genes of a model organism we know a lot about, the fruit fly. This will help us find out what determines the ratio of males to females, and how host-parasite interactions drive evolution."

A natural parallel is a parasitic wasp, Asobara tabida, that depends on Wolbachia infection to reproduce. When "cured" of its parasites, researchers reported last year, it is unable to reproduce.

"The remarkable thing in this case is that infection with the parasite makes a bad fly mutation more benign - but at the cost of potentially addicting the species to the parasite," Cline said.

Starr and Cline publish their findings in the July 4 issue of Nature.

Various species of Wolbachia have been found to infect from 25 to 75 percent of all insect species in the wild, as well as spiders, nematodes and some crustaceans. The bacteria pass from one host to another only through eggs - which means only through the female line - and have developed a very aggressive strategy to improve their chances of success. In mosquitoes, infected males can mate successfully only with females infected with the same strain of the bacteria, thereby ensuring that the number of infected mosquitoes increases with each generation. Similar techniques exploiting "cytoplasmic incompatibility" have been found in beetles, wasps, moths and fruit flies.

In some wasps, the bacteria allow females to reproduce without the assistance of males. And in several species, the bacteria kill males during early development.

Though Wolbachia apparently don't infect vertebrates, it can have an impact on human health. The nematode that causes river blindness in tropical countries is able to lay eggs and reproduce in humans only thanks to the Wolbachia bacteria infecting it. Wolbachia, not the nematode itself, trigger the immune response that damages the eye.

No one knows how Wolbachia achieve this, but now that they have been shown to affect a specific fruit fly gene, called sex-lethal (Sxl), Cline said researchers have their first handle on solving the problem.

"The fact that the interacting gene in this case has been studied so extensively and belongs to a model experimental organism can be exploited to yield further insights into the mechanism by which this parasite takes advantage of its various arthropod hosts," the authors write in their paper.

Cline has studied the genetics of sex determination in fruit flies for 30 years, and discovered the genes that set the sex ratio - what percentage of offspring is male and what percentage is female. These genes work by controlling sex-lethal, and Cline has mapped out the molecular details of this control.

"Interestingly, the genetic mechanism that regulates sex ratio changes dramatically in very closely related organisms," said Cline. "Sex ratio seems so fundamental, though, it raises a big question: Why does the mechanism change so fast across species lines?"

Hence his interest in a parasitic bacterium that changes the sex ratio in fruit flies. The existence of such a parasite suggests the important role that host-parasite interactions have in evolution, including the rise of new species.

"There's a very real possibility that species often get in a bind with a parasite that changes the sex ratio in a way that doesn't benefit the species," he said. "To get out of the bind, the species has to evolve further. Most of life is a race between host and parasite, plus change resulting from interaction between the sexes."

Cline's focus is on the sex-lethal gene, which he named because mutations in the gene kill one sex or the other. Over many years of research, he has established that Sxl is the master gene determining an offspring's sex. The activity of Sxl is determined by the number of X-chromosomes allotted to an egg: females have two X chromosomes, which switch on Sxl and generate all the characteristics that make females female; males have one X, which leaves Sxl silent.

The gene also plays a role in the physical process of egg development. While some mutations in sex-lethal knock out the entire gene and are lethal to female offspring, a few point mutations affect only the egg-production machinery, creating viable but sterile offspring.

The fruit flies observed in this experiment had a point mutation in both copies of their sex-lethal gene, which made the females sterile. (Some flies had a mutation - an altered base pair - in one area of the gene, while others had a single base-pair change at a second, nearby site. Both mutations had the same effect.) While preparing to irradiate some infertile flies to try to create mutations that reverse the infertility, a researcher in Cline's lab found an unusual group of flies that already appeared to have lost this trait. In this group of flies, the females produced large numbers of eggs, though few offspring.

Starr found an inherited difference between the fertile flies and the supposedly identical and still infertile mutants, but showed that this genetic factor was not located on any fly chromosome. Cline immediately suspected Wolbachia infection. Starr searched with the polymerase chain reaction and found evidence of the bacteria. She subsequently crossed a Wolbachia-free strain of mutant fly with a wild fly infected with Wolbachia, and found most of the resulting females fertile. The clincher was that treating these flies with the antibiotic tetracycline to kill Wolbachia made them sterile again.

The strange results have already told them that Wolbachia probably produce some protein that interacts with the protein encoded by the sex-lethal gene. The bacteria apparently do not affect expression of the gene or bypass the protein in the pathway leading to egg production. In other insects, however, the bacteria might work differently, Starr cautions.

"We can't tell if what we find in fruit flies will be related to how Wolbachia operate in wasps or other insects, but we know it will be interesting," said Starr, who continues to track down the details of how the bacterium interacts with fruit fly genes or proteins. "We'll discover something about how this bacterium works, but it will tell us a lot about sex determination in general."