Rett syndrome (RS) is among the most physically disabling of the autism spectrum disorders, affecting one in every 10,000 to 15,000 females born, according to the National Institutes of Health. After its outset—usually within 18 months of birth—young girls suffering from the illness begin to display asocial symptoms similar to those of autism. RS primarily affects the nervous system and can eventually lead to problems with speech and movement, often leaving patients with a stiff gait or confined to a wheelchair. Symptoms can also include tremors and irregular breathing.

In 1999 Huda Zoghbi, a specialist in pediatric neurological disorders at Baylor College of Medicine in Dallas, found that RS was likely caused by a mutation in the gene Mecp2, located on the X chromosome. Mecp2 had been discovered nine years earlier by molecular biologist Adrian Bird of the University of Edinburgh in Scotland. It is believed to have an epigenetic function, regulating other genes by switching them off when necessary.

Now Bird's lab reports in this week's issue of Science that it has reversed the debilitating symptoms associated with RS by targeting the gene it discovered nearly two decades ago.

"The point of our study is to show that it's never too late to reactivate the gene," Bird says. "We could activate the gene in about 80 percent of all brain cells and this seemed to be enough to reverse the symptoms."

Bird and his colleagues created mutant mice with a roadblock in the Mecp2 gene that prevented it from being expressed. Using the drug tamoxifen—an estrogen receptor modulator used to treat breast cancer—the scientists were then able to prompt an enzyme suspended by an estrogen receptor in the cytoplasm of the mouse's neuronal cells to migrate into the nucleus and restore the gene's function. After determining the proper dosage of tamoxifen—an early trial resulted in a number of mouse deaths due to overactivation of Mecp2—researchers settled on a four-week regimen of ramping up the gene's function.

The team scored the mice on a scale based on the symptoms they presented, such as awkward gait, lethargy and tremors, to monitor improvement. Both mouse and human males typically die early from the mutation in Mecp2, because their Y chromosome does not supply a normal copy of the gene. In the mouse model, the scientists were able to activate Mcep2, drastically reduce aberrant behaviors and prolong the lives of the mice well beyond the usual 17-week maximum life span of males with blocked Mcep2. Human females can live a normal life span with the disease, as can female mice—the latter species also frequently becomes obese. All of the symptoms, including increased weight, were erased in the female mouse model. The team cemented its finding by noting the mice's neurons' ability to undergo long-term potentiation—a type of response to stimulation, which has been implicated in learning processes—after gene function was restored.

"It's like a dream result, to me," marvels biologist Rudolf Jaenisch of the Massachusetts Institute of Technology's Whitehead Institute. "Even if you have the disease—and these animals were almost moribund—you can still rescue them."

Bird says that previous studies suggest that the mouse model studied here could be applicable to humans. "It encourages the belief that human Rett syndrome is also reversible, if only we can identify the appropriate therapy," he says. Schahram Akbarian, a psychologist at the University of Massachusetts Medical School, agrees the finding is exciting, but is not as optimistic about its application. "We don't know if the observed reversibility of the disease symptoms as observed in the mouse," he says, "exists in humans who have a much longer period of pre- and post-natal brain development than mice—months and years in humans, weeks in mice."

In addition to the new work's potential for RS, there is speculation that it could pave the way to treatments for other neurological disorders, such as learning disabilities, schizophrenia, autism and newborn encephalopathy as well as some mental retardation that has also been linked to the Mecp2 gene.

"The successful restoration of normal function demonstrated in the mouse models suggests that if we can develop therapies to address the loss of Mecp2," Baylor's Zoghbi says, "we may be able to reverse neurological damage in children and adults with Rett, autism and related neuropsychiatric disorders."

Bird is cautiously optimistic. "Our data only refers to Rett," he says, "but it makes one wonder about autism and other human brain disorders."