Genetic implications of individual intervention and neuronal dysfunction in neurodevelopmental disorders

Abstract

Neurodevelopmental disorders (NDDs) are a group of conditions appearing in childhood, with developmental deficits that produce impairments of functioning. Autism spectrum disorder (ASD) is a common NDD with a high heritability affected by complex genetic factors, including both common and rare variants. Behavior interventions such as social skills group training (SSGT) have been widely used in school-aged autistic individuals to relieve social communication difficulties in a group setting. Studies have confirmed that intervention outcomes can be influenced by sex and age, but how the genetic risk contributes to the outcome variability remains elusive. Furthermore, although large population cohorts have been well studied and have found numerous genes associated with ASD and NDDs, the molecular and neuronal outcomes of risk variants and genes are unclear. Therefore, this thesis included four studies in which the effects of genetic factors on intervention outcomes and cellular level neuronal functions were investigated.

Results from this thesis may provide a genetic perspective for further studies to explore potential individualized treatments for ASD and other NDDs. Specifically, in STUDY 1-3, exome sequencing and microarray were performed on individuals from a randomized controlled trial of SSGT (KONTAKT®). Common and rare variants, including copy number variations (CNVs) and exome variants, were tested for association effects with SSGT and standard care intervention outcomes. Polygenic risk scores (PRSs) were calculated from common variants, and clinically significant rare CNVs and rare exome variants were prioritized. Molecular diagnoses were identified in 12.6% of the autistic participants. PRSs and carrier status of clinically significant rare variants were associated with intervention outcomes, although with varied effects on both SSGT and standard care. In addition, genetic scores representing variant loads in specific gene sets were obtained from rare and common variants in ASD-related pathways. Outcomes of interventions were differentially associated with genetic scores for ASD-related gene sets including synaptic transmission and transcription regulation from RNA polymerase II. After combining genetic information and behavior measures, a machine learning model was able to select important features and confirm that the intervention outcomes were predictable. In STUDY 4, genetic variants affecting Calcium/Calmodulin Dependent Serine Protein Kinase (CASK) gene, a risk gene for NDDs, were examined using human induced pluripotent stem cell-derived neuronal models to identify the cellular effects of these mutation consequences. CASK protein was reduced in maturing neurons from mutation carriers. Bulk RNA sequencing results revealed that the global expression of genes from presynaptic development and CASK network were downregulated in CASK-deficient neurons compared to controls. Neuronal cells influenced by CASK mutations showed a decrease of inhibitory presynapse size and changed excitatory-inhibitory (E/I) balance in developing neural circuitries.

In summary, this is the first study to investigate the association of genome-wide rare and common variants with ASD intervention outcomes. Differential variant effects were found for individuals receiving SSGT or standard care. Future studies should include genetic information at different levels to improve molecular genetic testing for diagnoses and intervention plans. Presynapses and E/I imbalance could be an option to be developed for the treatment of CASK-related disorders.