Subscribe to eTOC

New Study Suggests Rett Syndrome May be Reversible


Dr. Adrian Bird: “It was a fairly obvious experiment to do. Once you know its a single-gene disorder, with late onset, the question is whether, if you turn the gene back on, can you reverse the symptoms?”

Rett syndrome may be reversible — so concludes a new study in which restoring gene expression relieved neurologic symptoms and neuronal function in mice. Although these results don't bring a treatment any closer, they raise the possibility that should one be found, it could have profound effects on children with the disorder.

In humans, loss of MECP2 causes Rett syndrome, an X-linked autism spectrum disorder affecting mostly girls, causing delayed-onset mental retardation, mutism, movement disorder, and other symptoms. (For more on Rett syndrome, see “Understanding Rett Syndrome.”)

In the new study reported online on Feb. 8 in Science, investigators genetically modified mice by stopping the production of MECP2 for a period of time, creating the equivalent of a “temporary mutation;” this spurred the neurologic symptoms similar to human Rett syndrome. The mice in the experiment recovered after the MECP2 gene was turned back on.

“It was a fairly obvious experiment to do,” said Adrian Bird, PhD, who led the study. “Once you know it's a single-gene disorder, with late onset, the question is whether, if you turn the gene back on, can you reverse the symptoms?”

“Everyone thought it would be too late. We're not neuroscientists, and we picked up on that general perception,” she added. Dr. Bird, a molecular cell biologist, is a professor at the Institute of Cell Biology at the University of Edinburgh in Scotland.

Countering this skepticism, animal models have shown that symptoms of brain disease can be ameliorated when the primary gene disorder can be reversed, even in neurodegenerative diseases such as Huntington disease or spinocerebellar ataxia. There is no neurodegeneration in Rett syndrome, suggesting that it might be even more amenable to such a strategy.


To test this hypothesis, Dr. Bird prepared mice with one normally expressing MECP2 gene, and one normal but silenced MECP2 gene, which could be turned on with a molecular switch. Through the random process of X inactivation – when the extra X chromosome is turned off — girls with Rett syndrome have one good gene and one defective gene.

To create a silenced gene that could be switched back on, Dr. Bird began with mouse embryonic stem cells, into whose normal MECP2 genes he placed a “stop cassette,” a bit of DNA that shuts down gene expression. The stop cassette was flanked on either side by a short DNA sequence called a “loxP” sequence. loxP sequences have become a common tool in the genetic engineer's toolkit because they have a unique relationship with a bacterial enzyme called cre. When cre is exposed to a piece of DNA with flanking loxP sequences, it snips out the loxP sequences and everything between them, and stitches together the remaining DNA, restoring the original active gene.

If Dr. Bird had wanted cre to clip out the loxP sites right away, it would have been a simple matter to breed a standard cre gene into the same mice. But that would turn MECP2 back on right away, defeating the purpose of the experiment. Instead, he needed a way to control the timing of the cre-loxP interaction. To do this, Dr. Bird turned to another tool in the kit, a cre enzyme linked to a tamoxifen receptor, which he then bred into the loxP-containing mice.

Without tamoxifen, cre stays in the cytoplasm and away from the loxP-containing DNA. But when mice are exposed to tamoxifen, it links to the receptor, and the tamoxifen-receptor-cre complex is imported to the nucleus, where the enzyme can snip out the stop cassette and turn the gene back on.

“All of this was just to get us to the beginning of the experiment,” Dr. Bird said. “But to us, it was a huge amount of work.”


Thus, at the start of the experiment, his female mice began development with one normal and one silenced MECP2 gene. Development proceeded normally for the first month or so of life, and then symptoms progressed for several months. The mice displayed inertia, irregular breathing, abnormal gait, and hind limb clasping, as well as obesity (not a feature in humans); as in humans, the phenotype eventually stabilized.

Then Dr. Bird injected tamoxifen, which turned on normal MECP2 gene expression throughout the brain. As in humans, random X inactivation led to a mosaic pattern of gene expression in the mouse brain, and a disorder similar to human Rett syndrome.


Dr. Huda Zoghbi: “These results are exciting. I was hopeful this would happen, but I was surprised by how beautiful the recovery was.”

Remarkably, the animals progressively reverted to a normal or almost normal phenotype, becoming more active, walking normally, losing weight, and breathing regularly. In mouse models of Rett syndrome, long term potentiation – the cellular basis of learning and memory — is defective.Electrophysiologic abnormalities in long-term potentiation were also reversed, suggesting a cell-level restoration of function as a result of normalized gene expression.


The results, according to Dr. Bird, provide a new model for understanding the molecular underpinnings of Rett syndrome. The model, which he terms the neuronal maintenance model, “fits with what we know about the function of the protein.” The MECP2 protein binds to methyl groups on DNA. Methylation is a common way to reduce gene expression, and the MECP2 protein is thought to help maintain this inactive state. “It's a stabilization mechanism for keeping genes switched off,” Dr. Bird said. This mechanism is not needed early in life, but only as development progresses, hence the timing of the disease.

This suggests that “until symptoms begin, the neurons are fine,” Dr. Bird explained. But once MECP2's silencing stabilization function is needed for neuronal maintenance, but cannot be performed because of the Rett syndrome mutation, “everything gets too loose for the neurons to function properly. They functionally degenerate,” even while remaining viable cells. “They're not really proper neurons,” Dr. Bird said, but this new study shows they retain the ability to become proper neurons again. “The conclusion is that though you might think it's too late, it isn't.”

“These results are exciting,” according to Huda Zoghbi, MD, who was not involved with the study “I was hopeful this would happen, but I was surprised by how beautiful the recovery was.” Dr. Zoghbi, professor of developmental biology at the Baylor College of Medicine, led the discovery of the Rett syndrome gene in 1999.

“This tells us that the neurons in Rett syndrome are poised to recover if you give them a chance,” she continued. “The dysfunction is not permanent.”


By itself, this discovery does not bring researchers any closer to discovering a treatment for Rett syndrome. Repairing the mutation on the active copy of MECP2 is not feasible, and reactivating inactive, normal copies is also impractical. “It's not a simple process to dig a gene out of silence,” Dr. Bird said. “Silencing genes on an inactive X chromosome seems to be a multilayered and redundant process.”

Barring that, Dr. Zoghbi said, “the key issue is still to find out what the molecular changes are in that dysfunction. Once we understand those, we can find points of intervention. This result tells us that if you can modulate those, at least some of the features of the disease may not be permanent.”

Despite the challenge, she is optimistic. “Ten years ago, we didn't know what caused Rett syndrome. Within the past eight years, we've discovered the gene, developed an animal model, and now know it is at least partly reversible. This is amazing progress, and it gives us a handle on where to go next.”


✓ In mouse models of Rett syndrome, investigators were able to turn off the MECP2 gene and then turn it back on, reversing the symptoms of the disorder.


• Guy J, Gian J, A Bird, et al. Reversal of neurological defects in a mouse model of Rett syndrome. Science 2007; E-pub 2007 Feb 8.