Neuronal gene variant associated with language deficiency

UK scientists have made the first association between a gene variant and a language disorder prevalent in pre-school children. Dr Simon Fisher and his team, of the Wellcome Trust Centre, University of Oxford published their findings last week in the New England Journal of Medicine. The work is the first to link a genetic variation with the developmental condition, termed Specific Language Impairment (SLI), which affects seven per cent of five to six year olds. SLI is the most common language disorder in children.

In spite of its high incidence, SLI can be difficult to identify. Just eight per cent of children with SLI have a conspicuous speech delay. Symptoms can be less apparent, like poor comprehension of syntax and grammar, or an inability to grasp pragmatics. Affected children inevitably have problems expressing themselves.

The story started in 2001, with the discovery that a mutated version of a gene called FOXP2 causes severe language impairment. Yet, FOXP2 itself does not make a component of the neuronal architecture per se; its product is a master regulator, exerting powerful downstream effects by turning other genes on or off. 'What's great about FOXP2', explained Dr Fisher, 'is that even though the mutations are very rare, we know that when people have a mutation it severely messes with their speech. It gives us a window into the biology'.

Hence the Oxford scientists screened all the stretches of DNA that are bound by the protein made by FOXP2, following which another gene, already implicated in schizophrenia, epilepsy, and - of most relevance - language impairment in autism, revealed itself. It was, according to Fisher's paper, the 'most compelling of candidates'.

The gene, called CNTNAP2, codes for a protein called neurexin, initially discovered as the target of the black widow spider toxin. Neurexins span the neuronal cell membrane and mediate cross talk between one cell and another. It is highly active in the cortex, a brain region implicit in language ability in humans. The team looked at the DNA sequences of CNTNAP2 in 184 affected families, and found a significant association between a particular variant and 'non-sense word repetition', a behavioural marker of SLI. The discovery opens the way for a genetic test to predict which children are at a higher risk of developing SLI. Intriguingly, this variant is seen in both autism and SLI, suggesting the two may indeed share a common cause.