Genetic susceptibility to stuttering.

Disorders that disrupt the development of speech, language, or reading have substantial effects on social function. Researchers have implicated specific genetic variants in monogenic speech disorder,1 common language impairments,2,3 and dyslexia.4 With the report by Kang and colleagues in this issue of the Journal,5 stuttering joins the fray. Stuttering is a disorder in which speech fluency can be severely compromised. It is characterized by the involuntary repetition or prolongation of sounds, syllables, words, or phrases, as well as frequent pauses, impeding the rhythmic flow of speech. Onset is typically between 3 and 6 years of age, and approximately 5% of preschool children are affected.6 The majority of young children who stutter go on to make a full recovery. For a considerable number, however, the disorder continues unabated, resulting in a prevalence of about 1% among adults.6 The primary causes of stuttering and the high rates of spontaneous recovery, as well as the etiologic distinctions between persistent and resolved forms, have resisted explanation. Familial clustering is extensively documented, and twin studies carried out in different countries and cultures show a high degree of heritability.7 Nevertheless, it is clear that genetic susceptibility to stuttering is complex, multifactorial, and heterogeneous. Genomewide scans have yielded suggestive evidence of linkage at multiple chromosomal sites, with little overlap among independent data sets.8-11 A rare example of significant linkage (with a locus on chromosome 12q) was found in a study of stuttering in 46 consanguineous families from Pakistan.8 Kang and colleagues now describe an elegant follow-up to this familial study and conclude that persistent stuttering, at least in a subgroup of affected subjects, may be related to disturbances in a metabolic pathway. In pursuing the chromosome 12 locus, they focused on Family PKST72, the largest family described in their earlier report,8 and expanded their study by recruiting additional relatives and performing new linkage analyses. They carried out a search for mutations across the interval with strongest linkage, sequencing promoters, exons, and untranslated regions of 45 genes in several affected persons, including three members of Family PKST72. They zeroed in on a single nucleotide change (G3598A) in the GNPTAB gene, which encodes both alpha and beta subunits of GlcNAc-phosphotransferase (GNPT). This variant predicts the substitution of lysine at a glutamic acid residue, which is conserved across species, and showed the clearest evidence of cosegregation with stuttering in Family PKST72. The investigators went on to sequence GNPTAB in 123 unrelated Pakistani subjects who stuttered (one from each family that had been included in the previous study,8 together with newly ascertained singletons), and they identified 5 additional people who were heterozygous or homozygous for G3598A. In 96 matched Pakistani control subjects, only one person carried this variant. Kang and colleagues also found G3598A alleles in one person of Asian Indian ancestry from a panel of 270 North American and British people who stuttered but not in 276 North American control subjects. Furthermore, the screening uncovered three additional putative mutations that were present in four unrelated Pakistani and North American subjects who stuttered, but were absent in the control subjects. The authors went on to sequence GNPTG,