Bacterial Butyrate in Parkinson's Disease Is Linked to Epigenetic Changes and Depressive Symptoms

Abstract

Background: The gut microbiome and its metabolites can impact brain health and are altered in Parkinson's disease (PD) patients. It has been recently demonstrated that PD patients have reduced fecal levels of the potent epigenetic modulator butyrate and its bacterial producers.

Objectives: Here, we investigate whether the changes in the gut microbiome and associated metabolites are related to PD symptoms and epigenetic markers in leucocytes and neurons.

Methods: Stool, whole blood samples, and clinical data were collected from 55 PD patients and 55 controls. We performed DNA methylation analysis on whole blood samples and analyzed the results in relation to fecal short-chain fatty acid concentrations and microbiota composition. In another cohort, prefrontal cortex neurons were isolated from control and PD brains. We identified genome-wide DNA methylation by targeted bisulfite sequencing.

Results: We show that lower fecal butyrate and reduced counts of genera Roseburia, Romboutsia, and Prevotella are related to depressive symptoms in PD patients. Genes containing butyrate-associated methylation sites include PD risk genes and significantly overlap with sites epigenetically altered in PD blood leucocytes, predominantly neutrophils, and in brain neurons, relative to controls. Moreover, butyrate-associated methylated-DNA regions in PD overlap with those altered in gastrointestinal (GI), autoimmune, and psychiatric diseases.

Conclusions: Decreased levels of bacterially produced butyrate are related to epigenetic changes in leucocytes and neurons from PD patients and to the severity of their depressive symptoms. PD shares common butyrate-dependent epigenetic changes with certain GI and psychiatric disorders, which could be relevant for their epidemiological relation. © 2022 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Keywords: DNA methylation; Parkinson's disease; epigenetics; gut brain axis; microbiome.

FIG 1

Outline of this study. [Color figure can be viewed at wileyonlinelibrary.com]

FIG 2

Butyrate and butyrate‐producing microbes are associated with PD depressive symptoms. (A) Short‐chain fatty acids change with Geriatric Depression Scale (GDS) total scores in PD patients (robust linear regression, adjusting for age, sex, smoking status, and BMI [body mass index]. Benjamini–Hochberg FDR [false discovery rate] q < 0.05 is used as the significant threshold). (B) The butyrate‐producing bacterial genera Roseburia and Romboutsia were negatively associated with GDS (Geriatric Depression Scale) total scores in PD patients. Benjamini–Hochberg FDR (false discovery rate) q < 0.05, signed logP: SLP. [Color figure can be viewed at wileyonlinelibrary.com]

FIG 3

Links of butyrate to PD epigenome in brain and blood. DNA methylation analysis was performed by Illumina EPIC array in PD and control whole blood (n = 55 PD, 55 controls). (A) DNA methylation changes in the blood of PD patients associated with butyrate levels are identified (robust linear regression, 3195 cytosines at Benjamini–Hochberg FDR [false discovery rate] q < 0.05, after adjusting for age, sex, smoking status, BMI [body mass index], and blood cell types). PD risk genes (identified by GWAS [genome‐wide association studies]) with differential methylation are highlighted. The percentage of hypermethylated and hypomethylated cytosines is plotted (Fisher's test, P < 0.05), and butyrate level is positively associated with cytosine methylation in PD. (B) Genes in PD blood (295 genes) and PD prefrontal cortex (107 genes) are significantly converged on the genes linked to butyrate levels (Fisher's exact test, P < 0.05). (C) Epigenetic alterations in blood and brain converge on those methylation sites linked to butyrate. The log(odds ratio) of genes in PD blood is significantly correlated with the log(odds ratio) of those genes in PD prefrontal cortex (Fisher's exact test, P < 0.05). (D) Enrichment analysis of genes epigenetically linked to butyrate and altered in PD relative to control. Nodes are q < 0.05 pathways merged by EnrichmentMap in Cytoscape. [Color figure can be viewed at wileyonlinelibrary.com]

FIG 4

Blood cell types are differentially linked to butyrate. (A) Significant methylation sites (related gene numbers) linked to butyrate in the epigenome of neutrophils, monocytes, CD8⁺T cells, CD4⁺T cells, and B cells, respectively. (B) Epigenetically altered genes linked to cytokines overlap with those epigenetically linked to butyrate in monocytes and neutrophils. Fisher's exact test, P < 0.05. (C) Gene set enrichment analysis of the monocyte and neutrophil genes that are epigenetically altered and linked to butyrate. PD polygenic risk pathways, including innate immune systems, metabolism of lipids and neutrophil degranulation, are significantly altered. Significant threshold Benjamini–Hochberg FDR q < 0.05. [Color figure can be viewed at wileyonlinelibrary.com]

FIG 5

Epigenetic–genetic correlation between butyrate‐associated mDNA (methylated DNA) regions and the GWAS (genome‐wide association studies) of other diseases. (A) LD (linkage disequilibrium) score regression was performed between butyrate‐associated mDNA regions and the GWAS summary statistics of Parkinson's disease, ulcerative colitis, Crohn's disease, rheumatoid arthritis, bipolar disorder, Alzheimer's disease, schizophrenia, and major depression, respectively. P < 0.05 is used as the significant threshold. (B) Gene set enrichment analysis of the location of common genetic regions in Parkinson's disease, butyrate‐associated mDNA regions, and those diseases in panel A. The log(FDR) of six PD polygenetic risk pathways without known PD risk loci was plotted. Benjamini–Hochberg FDR (false discovery rate) q < 0.05 is considered as significant threshold; white squares indicate nonsignificant associations. [Color figure can be viewed at wileyonlinelibrary.com]