The circular and linear transcriptome in Parkinson’s disease

Abstract

Neurodegeneration in Parkinson’s disease (PD) primarily affects dopaminergic neurons in the substantia nigra, often preceding clinical diagnosis by years. Biomarkers that reflect this pathology are needed for diagnosis, tracking disease progression and identifying at-risk individuals. RNA levels in PD patients might serve as measurable biomarkers in easily accessible tissues. While RNA produced at genetic loci is typically linear, circular RNA (circRNA) can also be expressed. CircRNAs are potential diagnostic biomarkers due to their increased stability and distinct regulation compared to cognate linear RNA. However previous RNA studies lack concordance, partly due to small cohort sizes, disease heterogeneity, and variable analysis methodologies. In this thesis, I analysed RNA sequencing data from blood samples of PD patients and controls from the PPMI and ICICLE-PD studies to measure gene, circRNA and cognate linear RNA expression. Using circRNA as a diagnostic biomarker in PD showed no clear improvement over linear RNA, minimising its potential clinical utility. Early-stage idiopathic PD patients exhibited a global reduction in circRNA expression, alongside increased expression of RNASEL and genes linked to the innate immune response. Similar global reductions in circRNA expression were observed in PPMI PD patients harbouring pathogenic LRRK2 and GBA variants. Asymptomatic PPMI participants at increased risk of developing PD, due to pathogenic LRRK2 and GBA variants or hyposmia, also had globally reduced circRNA expression. Conversely, RNA sequencing of dopaminergic neurons isolated from the substantia nigra of individuals from the BRAINcode study showed increased global circRNA expression in PD patients compared to controls. Overall, this work suggests that changes in global circRNA levels, potentially driven by the innate antiviral response, may play an important role in PD development and pathobiology.