Transcriptome Analysis of Neurons Differentiated from Patient-Specific Blood Derived Induced Pluripotent Stem Cells Reveals Convergent Pathobiology in Idiopathic Autism Spectrum Disorders
To date, numerous candidate genes have been associated with autism spectrum disorder (ASD) with many of these genes known to have important roles in synaptic function and the development of neural circuits. This suggests that certain neurobiological processes could be commonly altered in ASD. Therefore, although there is a great deal of clinical and genetic heterogeneity in ASDs, there may be convergent deficits in key molecular mechanisms which underlie the disease. Nonetheless, Lack of appropriate human-based models of complex neurodevelopmental disorders has greatly hindered investigations of convergent neurobiology in ASDs. Hence, induced pluripotent stem cells (iPSCs) offer the opportunity to further unravel the complex biology underlying ASD.
This study aims to determine the convergent biological pathways underlying ASD by probing the transcriptome of cortical neurons derived from ASD-specific iPSCs.
We derived patient-specific iPSCs from the whole blood of six individuals with idiopathic ASD and of five control individuals as well. Subsequently, each of these stem cell lines were differentiated into cortical neurons for 135 days. RNA was extracted from these neurons at day 35, 85 and 135 of differentiation and whole transcriptome analysis (RNA-Seq) was performed. Significantly differentially expressed genes were identified at each of the time points using edgeR software. Furthermore, gene networks and pathways were analyzed via multiple approaches: Ingenuity Pathway Analysis (IPA), Weighted Correlation Network Analysis (WGCNA), Short Time-series Expression Miner (STEM) and Gene Ontology (GO ontology).
Analysis of siginificantly differentially expressed genes at each time point highlighted deregulation in brain developmental, cytoskeletal and metabolic processes. Furthermore, analysis of overlapping differentially expressed genes between time points revealed five consistently differentially expressed genes at all three time points: FAR2P1, HS3ST4, MAB21L2, POTEF and TFF3. WGCNA analysis of the RNA Seq data generated a combined total of 52 gene modules and many of these modules presented an enrichment in the deregulation of pathways associated with neuronal differentiation, transcription and DNA and RNA metabolic processes. Finally, the analysis of changes in gene expression over time using the STEM software revealed neuronal cytoskeletal, proliferation and metabolic processes to be deregulated across the whole time course.
In our study, different analytical approaches revealed multiple pathways and processes that have been previously reported to underlie ASD biology. These mainly consist of brain developmental, neuronal differentiation and cytoskeletal processes. Of particular note, we have found that DNA, RNA and other metabolic processes may be important contributors to ASD pathobiology. We have shown that patient-specific iPSCs can be used to model brain region-specific neuronal development permitting the identification of common molecular mechanisms disrupted in ASD and identifying important candidate targets for therapeutic intervention.