Trash to treasure: junk DNA influences eggs, early embryos.

The mammalian genome might have a good mason to hold on to its vast collection of what scientists call junk DNA. Some of this genetic clutter may control gene expression in eggs and the earliest embryos, according to a report in the October Developmental Cell.

About one-third of the total mammalian genome is made of king, repeated stretches of DNA known as retrotransposons. Researchers hypothesize that retrotransposons derived from viruses that infected cells early in animal evolution. Retrotransposons, also called jumping genes, duplicate themselves and insert the copies into random places in DNA. After many generations, thousands of copies of each retrotransposon now reside in the mammalian genome.

These repetitive sequences, studied primarily in adults, had seemed to have little or no functional role in mammals. Recently, however, Barbara Knowles, a developmental biologist at the Jackson Laboratory in Bar Harbor, Maine, and her colleagues examined mammalian gene activity in early development.

Knowles' team recorded which genes were turned on in mouse eggs and early embryos. Unexpectedly, the researchers found that about 10 percent of the active genes belong to a group of retrotransposons known as long terminal repeat class III.

"It was very surprising to us," Knowles says. The retrotransposons "were very abundant, so we thought, 'I wonder what they're doing?'" she recalls.

Taking their study in a new direction, the researchers discovered that some retrotransposons sit right in front of genes. That position is typically occupied by a gene's promoter, a segment of DNA that initiates the process of making a protein. The team observed that some retrotransposons act as promoters, and, because they're copies, they control simultaneous expression of multiple genes.

The researchers also found that the retrotransposons became permanently inactive before embryos reached the four-cell stage.

Knowles suggests that over the course of evolution, retrotransposons that had randomly inserted themselves into promoter sites turned on genes that otherwise might have stayed off. Other retrotransposons, which landed in the middle of genes, might have changed those genes' functions.

John Moran, a retrotransposon researcher at the University of Michigan in Ann Arbor, calls the finding "an awesome observation." However, he says the study doesn't answer the question of whether the genes turned on by retrotransposons are important for development.

"What are [the retrotransposons] doing there?" Moran asks. "It will be interesting to see the follow-up studies."

Knowles says that if retrotransposons of this type do indeed play a significant developmental role, they might lead to new theories about why so maw embryos die soon alter conception. The findings might also provide insight into prerequisites for normal development that often go awry in cloning experiments.