Alternative approaches to characterizing the genetic architecture of autism spectrum disorder

Author(s)

Claudia Ismania Samogy Costa

• Document type: Doctoral Thesis
• Press: São Paulo.
• Institution: Universidade de São Paulo (USP). Instituto de Biociências (IBIOC/SB)
• Defense date: 2024-06-18

Examining board members

Maria Rita dos Santos e Passos Bueno; Michel Satya Naslavsky; Marcos Leite Santoro

• Advisor: Maria Rita dos Santos e Passos Bueno

Abstract

Autism spectrum disorder (ASD) represents a clinically heterogeneous group characterized by persistent deficiencies in social interaction and communication, repetitive behaviors and restricted interests. ASD affects about 1% of the global population, with a prevalence four times higher in boys than girls. It presents a high heritability (38%-90%), with a complex and heterogeneous genetic architecture. Although hundreds of genes have been associated with ASD, up to 30% of cases carry a known pathogenic variant with significant clinical effect. The remaining cases seem to be associated with more complex inheritance models, such as the oligogenic and polygenic/multifactorial models. Despite advances in identifying rare risk variants for ASD (associated with the oligogenic model), the number and combinations of such variants necessary for phenotype manifestation remain unknown. Furthermore, due to the low frequency of these variants, the sample size required to detect causal combinations often exceeds the large size of established autism cohorts. Therefore, this thesis aimed to contribute to the characterization of ASD\'s genetic architecture using two different approaches: a) Characterizing the genetic background of individuals carrying a known rare risk variant to establish a genetically more homogeneous cohort. Pathogenic variants in the dystrophin gene (DMD), a well-known cause of muscular dystrophies, were used as a model. From a sample of 83 individuals with dystrophinopathies, we demonstrated that DMD-ASD individuals carry a higher average number of de novo and inherited risk variants compared to DMD-controls. Additionally, we described an enrichment of extracellular matrix-related genes, especially collagens, and Ehlers-Danlos syndrome genes. These findings support an oligogenic model for ASD in dystrophinopathies; b) Determining the rate of de novo variants (DNVs) in two generations of families with an ASD proband, which had not been characterized for the Brazilian population until then. Studying these 33 families enabled us to demonstrate that individuals with ASD carry a higher number of DNVs than their unaffected parents and controls. Moreover, we observed that 40% of the DNVs identified in parents transmitted to probands are in genes associated with or candidates for ASD, representing recently emerged risk variants in these families. It is worth noting that, in both studies involving different ASD samples, an effect of paternal age on the increased average number of DNVs in individuals affected by ASD was observed. In summary, our studies addressed the role of rare risk variants in the genetic architecture of autism, emphasizing the effectiveness of using genetically more homogeneous populations. We explored rare variants in dystrophinopathies and DNVs in multigenerational families, observing an enrichment of risk variants in both scenarios. (AU)

• FAPESP's process: 19/19521-7 - Characterization of the genetic architecture of neurodevelopmental disorders using Duchenne Muscular Dystrophy as a model
• Grantee: Claudia Ismania Samogy Costa
• Support Opportunities: Scholarships in Brazil - Doctorate