Novel Transcripts Identified from iPSC-Derived Cortical Neurons Generated from Individuals with Idiopathic Autism

Background: Autism spectrum disorder (ASD) is a heterogeneous, neurodevelopmental condition characterized by irregularities in social interaction, verbal and non-verbal communication, and repetitive stereotyped behaviors. Understanding of ASD has been restricted by a lack of model systems that faithfully replicate all of the features of autism pathophysiology. Induced pluripotent stem cells (iPSCs) provide an attractive model to examine the cellular and molecular changes that underlie ASD. Previous RNA-seq analysis in our laboratory from cortical neurons derived from iPSC lines of individuals with idiopathic ASD and controls has identified sets of coding transcripts that are differentially expressed (DE). These results demonstrated an enrichment of DE genes in pathways including neuronal fate specification, extracellular matrix interactions, and axonal functionality. However, these coding RNAs represent only a small fraction of the total transcripts identified.

Objectives: The purpose of our current study is to identify novel transcripts, such as spliced isoforms and noncoding transcripts including long intergenic non-coding (LINC) RNAs, antisense transcripts, and circular RNAs that are differentially regulated in ASD-specific iPSC-derived cortical neurons compared to control neurons.

Methods: iPSC lines were created from peripheral blood mononuclear cells (PBMCs) and developed from a number of individuals with autism and controls. These iPSCs lines were differentiated into cortical neurons and RNA was extracted at several time points post initiation of differentiation (day 35, day 85 and day 135). RNA-seq analysis was performed using the HiSeq 2500 and significantly differentially expressed genes were identified at each of the time points using edgeR software. Further informatics analyses were used to identify novel spliced isoforms (eg. MapSplice) and noncoding transcripts (eg. lncrnadb.com and circbase.org) differentially expressed in ASD compared to control neurons.

Results:  A wide variety of novel transcripts were identified that did not match the predominant coding RNAs, including alternatively spliced isoforms of coding genes and non-coding RNAs. Differential expression of alternative spliced isoforms that distinguish ASD and control neurons were identified at each time point. These can be categorized into different groupings such as missed exons, alternative exons, and retained introns. Noncoding transcripts were differentially expressed between the neurons from ASD and control neurons, including antisense RNAs (e.g. DLGAP1-AS1 and POU6F2-AS2) and long intergenic noncoding RNAs (e.g. LINC01139). In addition, we were able to identify circular RNAs in our dataset. This included the well characterized CDR1-AS, also known as ciRS-7, a circular RNA highly expressed in the central nervous system that functions as a sponge for miR-7 (Memczak, et. al., 2013, Nature).

Conclusions: The results of this study show that there is a diversity of transcripts that are differentially expressed in cortical neurons derived from ASD-specific and control iPSC lines. This enrichment for novel transcripts, particularly noncoding transcripts, suggests potential novel gene regulatory roles in ASD.