A review of genome‐wide transcriptomics studies in Parkinson's disease

Abstract

Parkinson's disease (PD) is a progressive and incurable neurodegenerative disorder. Although numerous genetic and environmental factors have been linked to the aetiology of PD the underlying pathobiology remains poorly understood, hampering the development of improved therapies. Transcriptomics has the potential to reveal significant insights into disease processes. In this review, we focused on published transcriptomics studies on PD with the aim of summarizing studies and identifying common biological pathways. A total of 96 articles were identified as follows: 12 meta-analyses, 21 re-analyses of existing data and 63 original studies. Of the 63 original studies, 33 were performed on brain tissue, 26 on blood, three on cerebrospinal fluid and one on skin. In the brain studies, altered pathways identified included those involved in dopamine metabolism, mitochondrial function, oxidative stress, protein degradation, neuroinflammation, vesicular transport and synaptic transmission. Studies on blood samples revealed alterations in pathways involved in immune function, inflammation, RNA processing, protein chaperones, mitochondrial function and programmed cell death. Limitations of these studies include small sample sizes (generally <40 cases/40 controls) and the application of widely varying statistical analysis and parameters. Only eight studies used the RNA-Seq technique. This review highlights the need for harmonization of transcriptomic approaches and the statistical analyses, and for the data to be deposited into publicly available databases in a standardized format for meta-analyses. Notably, the concordance of several pathways such as mitochondrial function, protein degradation and inflammation, identified in both blood and brain tissues, suggests that the disease process is systemic and not restricted to neurological tissues.

Common gene expression signatures in Parkinson’s disease are driven by changes in cell composition

“…For instance, the signal from vesicle trafficking-and synaptic transmission-related processes [19][20][21][22][23][24] was significantly attenuated in both cohorts, suggesting that the signal was primarily driven by changes in neuronal proportions between PD and controls, rather than modulation of these pathways within neurons. Moreover, we observed an attenuation in the down-regulation of mitochondrial pathways, including the respiratory chain and oxidative phosphorylation, which are among the most consistent transcriptomic signatures in PD [2,4,19,21,[25][26][27][28]. The loss of transcriptional signal in these pathways is intriguing, because there is compelling evidence that decreased complex I protein levels occur in PD neurons [29].…”

“…Correcting for cell-type composition in our samples highlighted processes related to the endoplasmic reticulum, unfolded protein response and lipid/fatty acid oxidation as the top differential gene expression signatures in the PD prefrontal cortex. Unfolded protein response is indeed one of the most consistently reported transcriptomic signatures in PD [2,4,21,25,33,34].…”

“…These studies show surprisingly low replicability at the level of individual genes, however, and only partial concordance for pathways. The most consistent alterations have been found in pathways related to energy metabolism/mitochondrial function and protein degradation, followed by synaptic transmission, vesicle trafficking, lysosome/autophagy and neuroinflammation [2].…”

“…A recent systematic review identified 33 original genome-wide transcriptomic studies in the PD brain, of which 6 were performed on laser microdissected neurons from the substantia nigra pars compacta (SNc) and the remaining in bulk tissue from various brain regions [2]. These studies show surprisingly low replicability at the level of individual genes, however, and only partial concordance for pathways.…”

“…Access to high-quality brain tissue is generally limited, and thus an optimal choice of experimental platforms becomes paramount to maximize sensitivity. While RNA microarrays are being gradually superseded by RNA sequencing (RNA-Seq) technology, only 3 out of the 33 studies identified by an up-to-date review [2] used RNA-Seq, and all of them employed poly(A) capture, a widely used protocol to restrict the analysis to mature mRNA [4][5][6].…”