RNA-Seq Analyses of RORA-Deficient Neuronal Cells and Brain Tissues from Individuals with ASD Provide Support for RORA As a “Master Regulator” of Genes Impacted By Autism

Background:  We have previously shown that multiple cohorts of individuals with autism exhibit reduced expression of RORAboth in the brain as well as in lymphoblastoid cell lines derived from individuals with ASD. We further showed by RORA-dependent chromatin immunoprecipitation followed by microarray (ChIP-on-chip) analyses that RORA, a nuclear hormone receptor, can potentially regulate the transcription of over 2500 genes, including more than 400 genes associated with ASD through genetics and functional analyses. The latter study suggested that RORA may serve as a “master regulator” of genes involved in the pathogenesis of autism.

Objectives:  This study was conducted to investigate the genome-wide impact of targeted knockdown of RORA expression in a neuronal cell model as well as the transcriptomic profile of postmortem brain tissues from individuals with ASD that were previously determined to be RORA-deficient by RT-qPCR analyses. Moreover, we were interested in validating altered expression of predicted transcriptional targets of RORA in the RORA-deficient cells and tissues.

Methods:  Stable knockdown of RORA expression was induced by transfection of SH-SY5Y neuroblastoma cells (an experimental model for neuronal cells) with shRORA RNA, followed by selection of stable transfectants by maintaining the cultures in puromycin. Knockdown of RORA expression in the transfected SH-SY5Y cells relative to that in cells transfected with a negative control shRNA was verified by RT-qPCR analyses. The stable RORA knockdown and control cells as well as brain tissues from the frontal cortex (BA 9/10 region) of age-matched male cases and controls were submitted for RNA sequencing (RNA-seq) to investigate the transcriptional profiles associated with RORAdeficiency. Functional analyses of the differentially expressed genes were performed using Ingenuity Pathway Analysis software.

Results:  Stable knockdown of RORA expression in the SH-SY5Y cells resulted in altered expression of approximately 4500 genes with a fold-change ≥ 1.4. Among the differentially expressed genes, approximately 500 had been previously identified as putative targets of RORA by our ChlP-on-chip analysis. RNA-seq analysis of RORA-deficient postmortem brain tissues from autistic individuals in comparison to those from controls revealed 156 differentially expressed genes, with 45 of these genes overlapping those affected by stable RORAknockdown in the cell model. Network prediction and pathway analyses of the differentially expressed genes revealed significant over-representation of genes associated with neurological functions (e.g., neuronal migration and neuritogenesis) and canonical pathways (e.g., ephrin receptor and axonal guidance signaling) impacted by ASD.

Conclusions:  Collectively, the RNA-seq analyses of RORA-deficient neuronal cells and brain tissues validate a large number of transcriptional targets of RORA identified by our prior ChIP-on-chip analyses. Moreover, these results support our earlier findings and hypothesis that RORA is an autism risk gene whose deficiency in a subgroup of individuals with ASD can lead to much of the molecular pathophysiology currently associated with ASD.