Rett syndrome

Features

Rett syndrome affects approximately one in 10,000 births and was first recognized by Austrian physician Andreas Rett in 1966. Symptom onset is delayed, as toddlers may have learned to speak and walk before a regression that usually sees permanent loss of these skills and of purposeful hand use. Breathing problems, a need for feeding tubes, seizures, anxiety, gastrointestinal and orthopedic issues are common. Life expectancy is variable, but increasingly extends beyond 50 years of age.

Relevance to autism

Rett syndrome is frequently classified within the autism spectrum due to its delayed onset and the occurrence of repetitive movements, impaired motor coordination and social withdrawal. Variable presentation of these clinical features and the identification of its precise genetic cause have led some to suggest its removal from the bulk classification of autism, although this remains controversial.

Genetic cause

Rett syndrome rarely runs in families, as affected individuals do not reproduce. About 95 percent of cases are caused by new mutations in the gene encoding MECP2 protein. Random X chromosome inactivation in females heterozygous for a Rett mutation leads to mosaicism whereby approximately half of cells express only the mutant allele, while the other half are functionally wildtype. This admixture of cell types leads to Rett syndrome. Males with the same mutations on their single X chromosome, and therefore in all cells, rarely survive toddlerhood. Several clinical conditions caused by mutations in other genes (e.g. FOXG1) have overlapping symptoms, but are now considered to be separate disorders.

Molecular basis

Linkage analysis of rare pedigrees in which Rett syndrome shows familial inheritance pinpointed mutated MECP2 as the causal gene. The MECP2 protein had previously been identified as a nuclear factor that binds to sites of DNA methylation in genomic DNA and recruits transcriptional co-repressors, leading to inhibition of gene expression. Many tested Rett mutations directly interfere with this recruitment process, by either abolishing DNA binding or co-repressor recruitment. Alternatively, some mutations drastically reduce MECP2 abundance, for example by destabilizing the entire protein. DNA methylation is broadly spread across the genome, with the exception of CpG island promoters, which means that MECP2 is bound globally. Interpretation of the precise consequences of MECP2 loss is complicated by its somewhat uniform distribution and the small magnitude of its effects on gene expression. This has led to the proposal that other MECP2 functions may be at play, though as yet the evidence implicating these alternatives is incomplete.

Animal models

Mouse and human MECP2 orthologs are 95 percent identical in amino acid sequence suggesting functional conservation. Accordingly, molecular understanding of Rett syndrome has been greatly facilitated by the creation of mouse models that closely mimic the human condition. For example, the variable average clinical severity associated with specific Rett mutations is reproduced by introducing the equivalent mutations into mice. The use of models has revealed that Rett syndrome is primarily due to MECP2 deficiency in the brain, in particular neurons where the protein is extremely abundant. An unexpected finding was that the Rett-like phenotype of MECP2-null mice is reversed when the wildtype gene is activated late. This result shows that development in the absence of MECP2 does no lasting damage to the mouse brain. In this sense, Rett syndrome is not a “neurodevelopmental” disorder. Importantly, if extrapolated to humans, reversibility implies that this is a curable condition.

Prospects for therapy

Reversal of phenotype in the mouse model has stimulated the search for Rett syndrome therapies, ranging from attempts to treat “downstream” metabolic consequences to “upstream” technologies, such as gene therapy. Several downstream approaches have reached clinical trials, although so far these have yet to produce clear benefits. Proof of concept for gene therapy, mediated by adeno-associated virus vectors, has been achieved in mouse models, using both direct brain and peripheral administration protocols. There is current commercial interest in taking this approach to clinical trials.

Entry compiled by the Adrian Bird Lab, at The University of Edinburgh.