The genetics of speech and language impairments.

Without instruction, most children master the complexities of spoken language by the age of 6 or 7 years. About 5% of apparently healthy children, however, struggle to acquire basic competence in one or more aspects of spoken language and are classified as having specific language impairment. Genetic factors have an important role in many such cases.1,2 Children with specific language impairment are four times as likely to have a family history of the disorder as are children who do not have such an impairment,3 and the concordance rate for the disorder is almost twice as great for monozygotic twins as for dizygotic twins.4 More than 10 susceptibility loci have been identified. More often than not, loci that are robustly linked to specific language impairment in one study show no linkage in other studies, and all these loci have been linked to other neurodevelopmental disorders.5 Are these reported associations real? If so, which genes underlie these linkages, and what is their mechanistic effect? Perhaps one reason that linkage studies have implicated different loci for specific language impairment is that each group of investigators has used different case definitions. Since the subjects are selected in different ways and different measures are used to define language impairment, the discovery of different loci would not be unexpected. Studies of relatively homogeneous groups of children with specific language impairment, whose disorders would appear to have a common cause, would seem to be more likely to yield a robust genetic result. Causes that have been proposed for receptive specific language impairment include deficits in short-term auditory memory, auditory sequencing, and rapid auditory processing. A second way of dealing with the clinical heterogeneity is to define a measurable biologic marker, or endophenotype, that is present in children with specific language impairment regardless of how the disorder is manifested clinically. However, even carefully delineated endophenotypes may have different causes. For example, children might perform poorly in the repetition of nonsense words because they have poor auditory abilities related to short-term memory, sequencing, processing, or perception or because they have oral motor apraxia or are simply inattentive during testing. Moreover, the same genotype can result in different endophenotypes. A person carrying a mutation that affects the coordination of complex oral motor movements could present as someone who performs poorly when asked to repeat nonsense words, is mute or selectively mute, omits phonologically unstressed elements and hence appears to have a grammatical deficit, or has speech dyspraxia, dysfluency, or stutter. A third and powerful technique for elucidating the genetics of specific language impairment is to identify cases of language impairments that show mendelian transmission. Virtually all cases of familial language impairments have multifactorial, polygenic transmission. However, in 1990, Hurst et al.6 discovered a three-generation British family with autosomal dominant transmission of oral motor and speech dys praxia.7,8 Affected members of this family (called the KE family) carry a mutation in FOXP2, which encodes a transcription factor.8 However, it is not clear how the mutation results in speech dyspraxia in affected