Co-expression analysis of gene expression data to explore the differential expression network of a-synuclein in Parkinson’s disease

Abstract

Parkinson’s disease(PD) is the 2nd most common neurodegenerative disease which is characterised by motor and non-motor symptoms including muscle movement, learning and speech. PD starts from Substantia Nigra (SN) in the brain and spreads to Putamen, neocortex and eventually reaches the cortex. Degeneration of dopaminergic neurons occurs in PD in the presence of Lewy bodies rich in aggregated a-synuclein, a protein encoded by SNCA. Mutations in SNCA, duplication and triplication of its locus results in its aggregation. This process of aggregation of a-synuclein is absent in blood. There is a need for studying the co-expression pattern of a-synuclein in blood and brain to see the difference in its regulators. The progression of the disease from one part of the brain to the other part of the brain also needs to be studied to target the genes responsible for its progression which might help to halt the spread of the disease. Publically available microarray gene expression datasets were taken from different parts of the brain affected by the disease as it progresses including SN (stage-1), Putamen (stage-2) and Prefrontal Cortex (PFC)(stage-3) and whole blood. The genes most correlated with SNCA according to their expression were extracted to find the modules of genes using Weighted gene co-expression network analysis (WGCNA). The module of genes containing SNCA, the module with the highest positive correlation with the disease and the module with the highest negative correlation with the disease were used to create co-expression networks 1 using Cytoscape. Hub genes and transcription factors (TFs) of these genes were found using the Cytoscape plug-ins CytoHubba and iRegulon respectively. The results of each step were compared to see the similarities and uniqueness’s in different parts of the brain and blood. The overall expression pattern and correlation with SNCA shows similarity among different parts of the brain but difference were observed between blood and brain. Further analysis of TFs showed that the regulatory elements of these co-expression networks vary in blood and brain. Synaptotagmin 1(SYT1) and Histone deacetylase 1(HDAC1) are shared by all parts of the brain used in this study as hub genes of negatively correlated network and positively correlated network respectively which suggests that HDAC1 and SYT1 are probably involved with the spread of the disease. It was found out that Repressor element 1-silencing TF (REST) is the common TF of the negatively correlated module in blood and brain. Similarly, TEA domain family member 4 (TEAD4) is shared as a TF of SNCA containing co-expression network in blood, SN and PFC. Hence, REST and TEAD4 can be good biomarkers of PD diagnosis, since they are readily available in the blood as well.