NIH News Release
National Institute of
Neurological Disorders and Stroke

Wednesday, January 16, 2002
2:00 p.m. EST

Natalie Frazin or Margo Warren
(301) 496-5751

Researchers Identify Potential Treatment for Learning Disability in Neurofibromatosis

Researchers studying learning disabilities associated with neurofibromatosis type 1, or NF1, have traced the problem to excessive activity of a crucial signaling molecule and have successfully reversed the disabilities in mice by giving them an experimental drug. The findings provide hope that these learning problems may one day be treatable in humans.

This study provides the first clear picture of what causes learning impairments in NF1, says study author Alcino J. Silva, Ph.D., of the University of California, Los Angeles (UCLA). NF1 is a genetic disorder that affects about one in every 4000 people. Patients with the disorder have an array of symptoms, including benign tumors called neurofibromas and light brown spots on the skin called café-au-lait spots. About half of the affected individuals have cognitive disabilities, which typically include problems with spatial learning (which affects organization and other abilities) and reading. The study appears in the January 16, 2002, electronic edition of Nature1 and was supported in part by the National Institute of Neurological Disorders and Stroke (NINDS).

In the new study, Dr. Silva, Rui M. Costa, and colleagues studied mice with a mutation in the gene for NF1. This gene produces a protein called neurofibromin that controls activity of a signaling protein called Ras. To investigate whether Ras plays a role in NF1-associated learning impairments, the researchers crossed the NF1 mice with mice that had another mutation (K-ras) that decreases Ras activity. They found that mice with mutations in either NF1 or K-ras had impaired performance on a test of spatial learning called the water maze test. Mice with both mutations, however, did as well as normal mice on this test. This suggests that the learning impairment in the NF1 mice was due to an excessive amount of Ras signaling and that this impairment can be surmounted by reducing the amount of Ras. The scientists also found that they could improve learning in the NF1 mice using an experimental drug that prevents Ras from carrying out its signaling role in cells.

The researchers also found evidence that explains how the abnormal NF1 gene leads to learning disabilities. Neurofibromin normally controls Ras activity and keeps it in check. Without this control, too much Ras signaling occurs. Experiments showed that the abnormal Ras activity increases nerve signals that inhibit activity-related changes (plasticity) in the synapses between neurons. Researchers believe these synaptic changes, called long-term potentiation or LTP, are the basis for learning in the brain.

The finding that drug therapy could reverse the learning impairment in adult mice is important because many scientists and physicians have thought that NF1-related learning problems are due to abnormal brain development, says Dr. Silva. The new study suggests that this may not be the case and provides hope for treating the thousands of NF1 patients worldwide who experience learning problems. The drug that reversed the NF1-associated learning impairments in mice is already in clinical trials to determine if it can help reduce the number of tumors in NF1, Dr. Silva says.

The researchers were surprised to find that the K-ras mutants, which have too little Ras signaling, had learning impairments just like the NF1 mutants, which have too much Ras, says Dr. Silva. This finding suggests that a specific amount of Ras activity — not too much and not too little — is important for learning. "It's like when you have too much acceleration or not enough braking — both are bad," Dr. Silva says.

This study is one of the first instances in which researchers have been able to reduce the symptoms of a neurological disorder in an animal model using their understanding of the molecular interactions that cause the disorder, says Dr. Silva. "This is the long-standing promise of the human genome project - the cloned genes can lead us to animal models, and the animal models to information that can be used in designing treatments." While these findings are encouraging, the researchers caution that they are not the entire story. Scientists need to confirm these findings and determine what cell types play a role in these learning impairments. However, "we now have something concrete that makes sense, with many lines of evidence converging. The hope is to exploit these findings to develop strategies to reduce these problems in humans," says Dr. Silva. He is now planning human studies of NF1-related learning problems in collaboration with a clinical lab at UCLA. Those studies should help develop measurements and tools that can be used in testing potential treatments.

The NINDS is a component of the National Institutes of Health in Bethesda, Maryland, and is the nation's primary supporter of biomedical research on the brain and nervous system.

This release will be posted on EurekAlert! at and on the NINDS website at

1.Costa RM, Federov NB, Kogan JH, et. al. "Mechanism for the learning deficits in a mouse model of neurofibromatosis type 1." Nature, January 16, 2002, early edition.