An Epigenetic Analysis of Myalgic Encephalomyelitis/Chronic Fatigue Syndrome

Abstract

Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS) is a debilitating disease affecting approximately 20,000 New Zealanders. Patients experience lifelong persistence of the disease with symptoms including a characteristic post exertional malaise, and dysfunctions in cognition, sleep and the autonomic nervous system. Symptoms are severe enough to prevent normal function, leaving a significant proportion of patients house or bed bound greatly reducing quality of life. Prior research and disease presentation indicates a multi-systemic pathophysiology primarily involving metabolic, immune and neurological dysfunctions. Susceptibility is believed to be a combination of a genetic predisposition in combination with environmental stressors. Recent investigative efforts have turned to epigenetics in order to further understand the disease. DNA methylation is a well-characterized epigenetic modification that is linked to changes in gene expression, especially when it is found within the regulatory regions of the genome such as promoters. Additionally, DNA methylation is a malleable, environmentally affected regulatory modification with the potential to provide insights into systematic changes linked to the disease. This investigation aimed to characterize, in depth, the DNA methylation patterns of ME/CFS patients compared with age and gender matched healthy controls.

In order to determine the DNA methylation variation in ME/CFS a cohort of 10 patients and 10 matched controls were investigated in the study. DNA was extracted from peripheral blood mononuclear cell fractions purified from whole blood. Following quality assessment of isolated DNA, DNA fragment libraries were prepared for Reduced Representation Bisulfite Sequencing (RRBS). The RRBS libraries were sequenced using high throughput next generation sequencing. The subsequent data were analysed using multiple bioinformatic platforms in order to determine patterns of differential methylation between the patient and control groups. RRBS designed MethylKit and DMAP analysis pipelines were utilised in order to investigate changes at each individual cytosine and at clustered cytosines within DNA 40-220bp fragments respectively. Additional methylation variation was investigated across genomic features of interest including promoters, enhancers and gene bodies. Genes identified associated with gene body differential methylation were then utilised for further pathway enrichment analyses.

With appropriate statistical significance thresholds, Methylkit identified 394 differentially methylated cytosines and DMAP identified 76 differentially methylated fragments. Manual inspection of the data identified four clusters of MethylKit cytosines overlapping or in close proximity to DMAP fragments. These clusters identified regions of regulatory importance for 16 protein-coding genes. Further independent regulatory feature analysis identified 22 promoter regions associated with 45 differentially methylated cytosines (utilising MethylKit data) and 11 promoters associated with 12 fragments (utilising DMAP data). Analysis of gene body differential methylation identified 91 genes associated with 121 individual differentially methylated cytosines and 31 genes associated with 31 fragments containing methylated cytosines. Functional pathway enrichment analysis with a FDR <0.05 identified 7 functional pathways through analysis of MethylKit identified genes, and 23 functional pathways through analysis of DMAP identified

The 16 genes associated with the regulatory regions identified by clusters of differential methylation largely falling into either immune or metabolic/mitochondrial related functions. Further gene body analysis identified a number of enriched functional pathways including a number of immune, metabolic and particularly neurological related functions. Together the pattern of DNA methylation in patients implicated components of a number of systems that may be differentially methylated and therefore potentially differentially regulated compared to a healthy population. Particularly the large number of enriched neurotransmitters and neuropeptide reactome pathways identified by DMAP genes implicate an irregular stimulation of the HPA axis through the ‘Stress Centre’ (Paraventricular nucleus), a link between an irregular neuroendocrine response to stress and the stress sensitivity of ME/CFS patients. Overall this work shows specific changes in ME/CFS methylation with compelling links to the pathophysiology of the disease