Transcriptional Profiling of Human Neural Differentiation Implicates Noncoding RNA and ASD-Associated Genes

Background: Increasing evidence suggests that noncoding RNAs hold diverse functions in various cellular processes and alterations in their expression may contribute to the etiology of several neuropsychiatric conditions, including autism spectrum disorder.  Despite being abundantly expressed in the brain, the functional properties of these non-protein coding RNAs in human neuronal cells remains largely unknown.  Exploring the complete transcriptional profile of human neurons may uncover gene networks underlying complex human disorders and may identify the noncoding RNAs possibly regulating these networks.

Objectives:  The present study aimed to identify the transcriptional landscape of two human neural progenitor cell lines as they differentiate into human cortical projection neurons.

Methods: The human neural progenitor cell lines, SK-N-SH and ReNcell CX, were used to measure gene expression as they undergo differentiation.  Cells were harvested at two stages of differentiation and RNA sequencing was performed to explore the transcriptional landscape of these cells.  Differential expression analysis and weighted gene co-expression network analysis (WGCNA) was performed to identify genes showing altered expression and to infer the functional properties of noncoding RNAs through their co-expressed genes. 

Results: Protein coding genes were found to account for 54.8% and 57.0% of expressed genes in SK-N-SH and ReNcell CX cells, respectively, and alignment of RNA sequencing reads revealed that only 25.5-28.1% mapped to exonic regions of the genome.  Differential expression analysis in the two cell lines identified altered gene expression in both protein coding and noncoding RNAs as they undergo neural differentiation with 222 differentially expressed genes observed in SK-N-SH cells and 19 differentially expressed genes in ReNcell CX.  Interestingly, genes showing differential expression in SK-N-SH cells are enriched in genes implicated in autism spectrum disorder, but not in gene sets related to cancer or Alzheimer’s disease.  Weighted gene co-expression network analysis (WGCNA) was used to detect modules of co-expressed protein coding and noncoding RNAs in SK-N-SH cells and found four modules to be associated with neural differentiation.  These modules contain varying levels of noncoding RNAs ranging from 10.7% to 49.7% with gene ontology suggesting roles in numerous cellular processes important for differentiation. 

Conclusions:   These results indicate that noncoding RNAs are highly expressed in human neural progenitor cells and likely hold key regulatory roles in gene networks underlying neural differentiation and neurodevelopmental disorders, such as autism spectrum disorder.