Expanding the molecular landscape of childhood apraxia of speech: evidence from a single-center experience

Abstract

Background: Childhood apraxia of speech (CAS) is a genetically heterogeneous pediatric motor speech disorder. The advent of whole exome sequencing (WES) and whole genome sequencing techniques has led to increased identification of pathogenic variants in CAS genes. In an as yet uncharacterized Italian cohort, we aimed both to identify new pathogenic gene variants associated with CAS, and to confirm the disease-related role of genes already reported by others. We also set out to refine the clinical and neurodevelopmental characterization of affected children, with the aim of identifying specific, gene-related phenotypes.

Methods: In a single-center study aiming to explore the genetic etiology of CAS in a cohort of 69 Italian children, WES was performed in the families of the 34 children found to have no copy number variants. Each of these families had only one child affected by CAS.

Results: High-confidence (HC) gene variants were identified in 7/34 probands, in two of whom they affected KAT6A and CREBBP, thus confirming the involvement of these genes in speech impairment. The other probands carried variants in low-confidence (LC) genes, and 20 of these variants occurred in genes not previously reported as associated with CAS. UBA6, ZFHX4, and KAT6A genes were found to be more enriched in the CAS cohort compared to control individuals. Our results also showed that most HC genes are involved in epigenetic mechanisms and are expressed in brain regions linked to language acquisition processes.

Conclusion: Our findings confirm a relatively high diagnostic yield in Italian patients.

Keywords: childhood apraxia of speech; exome sequencing; gene ontology and expression profile of CAS genes; high confidence genes; low confidence genes.

Figure 1

Chart of high-confidence (A) and low-confidence variants (B) in a CAS cohort.

Figure 2

(A) ToppGene enrichment analysis of the GO human phenotype ontology, molecular function and biological process. (B) RNA-seq RPKM values obtained from the BrainSpan database for the set of high-confidence (HC) genes. For each donor available in the BrainSpan Atlas of the Developing Human Brain database, the RNA-seq expression values of the HC genes are plotted using the ComplexHeatmap 2 R package. The heatmap is divided into submodules, each representing the genes expression values for a particular brain area, described at the top of the plot. Note that each donor is associated with one or more columns, since the RPKM values are available for multiple brain areas. Genes are also clustered based on their RPKM values, as represented by the dendogram on the left, in order to better show expression similarity between genes.