Laboratory experiments led by Johns Hopkins scientists have revealed that so-called "jumping genes" create dramatic rearrangement in the human genome when they move from chromosome to chromosome. If the finding holds true in living organisms, it may help explain the diversity of life on Earth, the researchers report in the Aug. 9 issue of Cell.

"Jumping genes," or retrotransposons, are sequences of DNA that are easily and naturally copied from one location in the genome and inserted elsewhere, particularly in developing eggs and sperm. There are more than 500,000 copies in the human genome of the retrotransposon the scientists studied, accumulated over the millions of years of human evolution.

But the sheer quantity of these elements isn't as striking as what else they might be doing as they jump around, says Jef Boeke, a professor of molecular biology and genetics in the Institute for Basic Biomedical Sciences at the School of Medicine.

"Textbooks always show these elements inserting themselves cleanly into new places in the DNA," says Boeke, who headed the research team from Hopkins, "but we saw that about 10 percent of the time, in addition to inserting, it's taking out a big chunk of the chromosome. The interesting thing isn't where these elements are going, but what happens when they get there."

Jumping genes are tightly regulated, and the jumping process probably doesn't happen as often in living organisms as in laboratory dishes, Boeke notes. However, in cells that develop into egg and sperm, even in adults, jumping genes are active. If retrotransposons cause as much chaos in sperm as they do in the lab, they might allow new genetic changes to begin in the next generation, Boeke speculates.

"Assuming that what we see in the laboratory is also happening in real life, it suggests that these elements have been remodeling host genomes more than previously realized, with deletions, insertions and inversions," he says. "These changes were probably frequently disastrous, but occasionally they might have benignly increased genetic variation or even improved survivability or adaptability. Such remodeling probably happened thousands of times during human evolution."

Using a total of 44 man-made insertions in two types of human cancer cells, the scientists tracked where jumping genes plopped into the genome and examined the surrounding area for "collateral damage." DNA sequences from the Human Genome Project helped them identify the new location and any major alterations caused by the insertion, Boeke says.

Much to their surprise, the act of insertion caused chunks of existing DNA to be cut out and, in one location, caused neighboring DNA to be inverted, as though it had been removed and reinserted backwards, the researchers say.

"These things are happening by mechanisms never before described," says Boeke, who as a postdoc more than a decade ago began studying retrotransposons, which must be transcribed into RNA before they can "jump."

Because many aspects of the man-made jumps are similar to what happens naturally, the scientists are confident that this system is a good way to learn about retrotransposons, Boeke adds. And like every part of the genome, retrotransposons reflect the evolutionary links among species.

"Retrotransposons in humans have certain characteristics, but if you look deeply into the human genome sequence, you find elements common to our primate ancestors," he says. "If you keep looking, you find even older elements. Together, these elements provide a molecular fossil record of our evolutionary history."

One characteristic of old retrotransposon elements is that their "tail" of repeated adenines, one of the four building blocks of DNA, are shorter and even have other building blocks thrown into the pattern. Other researchers have suggested that newer (later) insertions have longer, purer tails, and indeed the man-made insertions had very long tails and no other bases in them, Boeke says.

"We just discovered a wealth of information by looking at where these retrotransposons go," Boeke says. "The sites of insertion are a microcosm of the human genome. It's just amazing."

A second paper in the same issue, by a group at the University of Michigan, documents similar findings in the 40 or so lab-induced insertions they studied.

The study was funded by the Howard Hughes Medical Institute and the National Institutes of Health.