Training babies' brains and bodies might delay the onset of Rett syndrome, a devastating neurological disorder that affects about 1 in 10,000 girls worldwide.

In experiments with mice that replicate the genetic disorder, scientists discovered that intense behavioral training before symptoms develop staves off both memory loss and motor control decline. Compared to untrained mice, those trained early in life were up to five times better at performing tasks that tested their coordination or their ability to learn, the researchers report in the journal Nature.

Those data, from animals whose symptoms closely mimic the human disease, offers a clear rationale for genetically screening newborns for Rett syndrome, says the senior author.

Rett syndrome primarily affects girls, who are typically diagnosed around age three ¬- well after symptoms first appear. An earlier diagnosis could offer a window of opportunity for treatment, - potentially delaying disease progression in children, or even making them better able to benefit from future therapies. "We are losing precious time," the author says. "If we could screen these girls and put them through training, maybe the time we gain prior to overt onset of symptoms will also create more opportunity for other treatments to work."

There are no effective treatments for Rett syndrome, and the unrelenting barrage of symptoms is grim. After developing normally for roughly the first one to two years of life, children progressively lose skills they've learned. By 18 months, kids may have trouble using their hands. By two years, their ability to balance deteriorates and language skills fade. When symptoms are full-blown, almost every part of the brain is affected. In severe cases, girls cannot talk, feed themselves, or even open their mouths. They can also experience seizures, teeth grinding, and difficulty breathing.

In 1999, the team discovered that mutations in a gene on the X chromosome were to blame. A defective copy of the gene, called MECP2, disables about half the brain's neurons, so that they function at only about 50-70% of their normal capacity, the team reported in 2011 and 2016.

Correcting MECP2 via gene therapy would be an ideal treatment, the team writes, but delivering the right dose to the right neurons poses a challenge. Too much MECP2 can cause neurological problems, too. Scientists are currently pursuing a gene therapy candidate for Rett syndrome, though clinical trials have not yet begun.

The team took an alternative approach. What if researchers could somehow prod those sluggish neurons into action? Using electrodes implanted in the brains of "Rett mice," the team discovered that stimulating key neurons at the base of the brain activated hippocampal neurons and improved learning and memory, cranking neural activity back up to normal. The team reported the results of this deep brain stimulation in Nature in 2015. "That was really, really exciting," the author says. "It gave us the idea that boosting the neurons' activity could help them."

Implanting electrodes into the brain isn't ideal for children, though. For one thing, all brain regions are affected by Rett syndrome. So the team tried to mimic the effects of deep brain stimulation with something non-invasive - intensive behavioral training.

"We thought it might work," the author says, "because both techniques are doing the same thing - stimulating neurons."

The team trained Rett mice in two skills that wither in people with the disorder: coordination and learning ability. Instead of training only mice with symptoms, the authors tested an unconventional idea, too: They also trained mice before any symptoms had developed.

One type of training included a "rotarod" apparatus, a rolling log-like treadmill that requires mice to continuously walk to keep their balance. Twice a week, four times per day, researchers placed mice on the device for a five-minute session. Those trained early in life outperformed those trained later, and they stayed on the apparatus roughly five times longer than mice with no training.

The team saw something similar in a water maze test, where mice use pictures on the wall to learn where an underwater platform is hidden. Again, mice trained early performed better those trained after symptoms developed, the researchers found. "The results were really quite stunning," the author says. "The difference is huge." Training was task-specific: mice that received memory training did not make improvements in coordination, for example.

In lab experiments, the team traced improved performance to specific sets of neurons responsible for each training task. Under a microscope, those neurons actually looked different than those from late-trained mice, with more branches and connection points to other neurons. That difference suggests that intense early training physically changes the brain, in a way that counters the disease, the author says.

In fact, continued training of the animals even held off symptoms, the team discovered. These mice were symptom-free for up to four months longer than untrained mice. That's a hefty chunk of time for the animals, which, like healthy mice, live about two years. The senior author thinks a clinical trial in humans is the next big step for this work and imagines infant training could take many forms, including extra "tummy time" where babies lie on their stomachs to strengthen their core, or focused language training lessons to help the infants gain a few words.

Screening girls at birth could give doctors the heads-up they need to begin such training - and potentially buy these children some time before the disease begins to encroach. The MECP2 genetic test is widely available and covered by insurance, the senior author says.

Even a delay of six months would be "absolutely worthwhile," the author adds. "I'm hopeful that we have a path forward to make a difference in the lives of these individuals."

https://www.hhmi.org/news/behavioral-training-could-help-babies-with-rett-syndrome-mouse-study-suggests

https://www.nature.com/articles/s41586-021-03369-7