Systematic Identification of circRNAs in Alzheimer's Disease.

Kyle R Cochran,Christopher Coletta,Gautam Kundu,Krystyna Mazan-Mamczarz,Kirtana Veeraraghavan,Myriam Gorospe,Supriyo De,Madhav Thambisetty

doi:10.3390/genes12081258

Kyle R Cochran, Christopher Coletta + Show 6 more

Open Access

https://doi.org/10.3390/genes12081258

Copy DOI

Abstract

Mammalian circRNAs are covalently closed circular RNAs often generated through backsplicing of precursor linear RNAs. Although their functions are largely unknown, they have been found to influence gene expression at different levels and in a wide range of biological processes. Here, we investigated if some circRNAs may be differentially abundant in Alzheimer’s Disease (AD). We identified and analyzed publicly available RNA-sequencing data from the frontal lobe, temporal cortex, hippocampus, and plasma samples reported from persons with AD and persons who were cognitively normal, focusing on circRNAs shared across these datasets. We identified an overlap of significantly changed circRNAs among AD individuals in the various brain datasets, including circRNAs originating from genes strongly linked to AD pathology such as DOCK1, NTRK2, APC (implicated in synaptic plasticity and neuronal survival) and DGL1/SAP97, TRAPPC9, and KIF1B (implicated in vesicular traffic). We further predicted the presence of circRNA isoforms in AD using specialized statistical analysis packages to create approximations of entire circRNAs. We propose that the catalog of differentially abundant circRNAs can guide future investigation on the expression and splicing of the host transcripts, as well as the possible roles of these circRNAs in AD pathogenesis.

Highlights

Mammalian circularRNAs arise from the looping of the 5 and 3 ends of an RNA to form a covalent bond
Decreasing the levels of ciRS-7 caused an increase in miR-7 available in the cell for repression of miR-7-specific mRNA targets; one of these targets encodes the protein UBE2A, which is responsible for the removal of amyloid peptides present in Alzheimer’s Disease (AD) brains [11]
We set out to identify circRNAs selectively associated with AD pathology

Summary

Introduction

Mammalian circular (circ)RNAs arise from the looping of the 5 and 3 ends of an RNA to form a covalent bond. With widespread use of high-throughput sequencing technologies, tens of thousands of circRNAs have been identified, typically based on the detection of their unique junction sequences, the RNA sequence where the 5 and 3 ends are covalently ligated [4,5] Given their closed-loop structure, circRNAs are believed to be more stable than linear RNAs [7]. Many circRNAs have gained attention as reliable biomarkers of organ function, dysfunction, and disease [23,24,25,26,27,28] Given their intrinsic stability, even if circRNAs originate in other tissues, they often eventually reach the circulation and might be found in plasma. We propose that the circRNAs identified in this study can guide the analysis of specific processes aberrant in the AD environment

RNA Sequencing Datasets from Brain Tissue and Plasma

Gene Expression Analysis in R

Relating circRNAs to AD Pathology

Analysis of AD-Associated RNA-Seq Datasets to Identify circRNAs