Fragile X: a class of its own

Just over a decade ago, fragile X syndrome — a common form of hereditary mental retardation — rose to notoriety by being the first disorder to be caused by the expansion of a trinucleotide repeat. Fragile X sufferers have more than 200 CGG repeats in the 5′ untranslated region of the fragile X mental retardation 1 ( FMR1 ) gene, compared with 60 or fewer in normal individuals. As in other similarly inherited conditions, the expanded disease alleles derive from phenotypically normal individuals that carry an intermediate number of unstable repeats, which in the case of fragile X ranges from 60 to 200.

Although this inheritance model has held true for years, some of these so-called 'premutation' carriers have now been found to develop a new form of progressive neurodegeneration — a phenotype that is unrelated to fragile X. This observation provided the starting point for the study by Peng Jin and colleagues. By modelling the condition in Drosophila they show that human premutation repeats alone can lead to neurodegeneration, and that the effect depends on the abundance and length of the repeat. This is also the first time that neurodegeneration has been attributed to RNA alone.

To test whether the neurodegeneration phenotype could be pinned down to changes in CGG-repeat size, Jin and colleagues looked at the effect of expressing human repeats of different length in several Drosophila tissues. What they found was clear cut — moderate expression of 60 CGG repeats (rCGG₆₀) had no effect, but moderate expression of the longer premutation rCGG₉₀, or overexpression of either rCGG₆₀ or rCGG₉₀, causes progressive degeneration specifically in neuronal tissue. This tells us that longer repeats are more toxic than shorter (normal length) ones and that toxicity increases with transcript abundance. The result also tallies nicely with what is seen in premutation patients, who have high levels of FMR1 mRNA.

Fragile X patients lack any FMR1 message, so the mechanism that underlies the progressive neurodegeneration must be distinct from that which causes fragile X. Instead, it could resemble that of other neurodegenerative disorders, such as some forms of spinocerebellar ataxia, which are caused by repeat expansions in noncoding regions and have been blamed on RNA.

But are we any closer to understanding what these repeat RNAs do? In both the experimental flies and human premutation patients, nuclear aggregates are seen that also contain ubiquitin and so it is possible that here the RNA repeats sequester vital proteins from their normal functions. The fact that the formation of these clumps can be reversed by overexpressing the chaperone heat shock protein 70 (Hsp70), which normally unravels or destroys badly folded proteins, is intriguing: the involvement of the protein-degradation machinery could link these RNA-mediated defects to the larger class of protein-based neurodegenerative disorders, many of which are also reversed by Hsp70.

The creation of a Drosophila model for this disease has already paved the way for targeted genetic studies. Meanwhile, it looks like fragile X is back where it started — in a class of its own.