A Changing Portrait Of DNA

Until recently, some scientists assumed that the rest of the genome was a hodgepodge of evolutionary leftovers that did very little of consequence. Part of it they called "junk DNA," and the rest of it they didn't even name. "I think some people were hoping that 99 percent of the genome could just be ignored," says biologist Eric Lander, founding director of the Broad Institute, a collaboration of Harvard University and the Massachusetts Institute of Technology. Over the last decade, though, researchers have realized that this forgotten part of the genome is, in fact, profoundly important. It contains the machinery that flips the switches, manipulating much of the rest of the genome.

Most of the machinery follows Mendel's laws. But not all of it does. Some of it violates the notion that both copies of a gene operate throughout life. During the tango of conception, the sperm and egg both try to lead: they argue over a small set of genes in which one copy, from the mother or the father, will be permanently switched off, leaving the other copy to work solo. This week, the journal Genome Research will publish a study, led by Jirtle, suggesting that there are 156 of these solo genes in the body. Many are responsible for regulating other genes.

Scientists have been studying gene regulation for decades, but in the past few years, since the Human Genome Project was completed, they have drastically accelerated their pace. There is still a great deal to be learned, but a new discovery now appears in a major journal almost every week. "Every time you think you've solved the way things get regulated, you realize there's yet another layer of complexity," says biologist Rick Young of MIT and the Whitehead Institute for Biomedical Research. "That can be frustrating, but it's also exciting. It's so beautifully complicated."

Researchers have explored and exploited several types of genetic switches in the last few years. "Small interfering" and "microRNAs"—tiny free-floating nucleic-acid strings that can fool genes into shutting themselves down—are some of the most intriguing. Scientists have figured out how to mimic them using artificially created versions to turn genes off. The technique, called RNA interference (RNAi), won the Nobel Prize last year and it has now moved from academia to industry. Several firms have invented locally injectable RNAi therapies, and three weeks ago Quark Pharmaceuticals began to test in humans the world's first systematically injected RNAi. It turns off a gene called p53 that can cause unnecessary cell death in the kidneys; once the drug's work is done it exits the body, and p53 turns back on.

Methyl groups, the four-atom configurations that tamped down the Agouti gene in Jirtle's mice, are another influential category of switches. They may interfere with the DNA directly, like monkey wrenches in its machinery or, instead, they may interact with histones, proteins that serve as yet another type of switch. Young, the MIT biologist, made a surprising discovery about histones in July: at least a third of our genes have histone switches that hover somewhere between on and off, allowing the genes to start manufacturing their signature proteins but not letting them finish. Young notes that the human embryo must initiate many complex developmental processes in a short time period. Maybe, he says, the body keeps some of the genes involved in development "poised for action" so it can kick-start them quickly, when there's "little time to waste."

That speed, however, may come at a cost. Some of the genes that are left half-on are crucial in early development. When they're fully turned on—and that could happen accidentally if they're already halfway there—they can wipe out the cell's entire machinery, turning it into a blank slate that looks dangerously like a cancer stem cell. Young is currently exploring the hypothesis that our half-on, half-off genes are directly linked to cancer—they're necessary for development, but they also may predispose us to tumors later on.