Meta-analysis of shotgun sequencing of gut microbiota in Parkinson’s disease

We aimed to identify gut microbial features in Parkinson's disease (PD) across countries by meta-analyzing our fecal shotgun sequencing dataset of 94 PD patients and 73 controls in Japan with five previously reported datasets from USA, Germany, China1, China2, and Taiwan. GC-MS and LC-MS/MS assays were established to quantify fecal short-chain fatty acids (SCFAs) and fecal polyamines, respectively. α-Diversity was increased in PD across six datasets. Taxonomic analysis showed that species Akkermansia muciniphila was increased in PD, while species Roseburia intestinalis and Faecalibacterium prausnitzii were decreased in PD. Pathway analysis showed that genes in the biosyntheses of riboflavin and biotin were markedly decreased in PD after adjusting for confounding factors. Five out of six categories in carbohydrate-active enzymes (CAZymes) were decreased in PD. Metabolomic analysis of our fecal samples revealed that fecal SCFAs and polyamines were significantly decreased in PD. Genes in the riboflavin and biotin biosyntheses were positively correlated with the fecal concentrations of SCFAs and polyamines. Bacteria that accounted for the decreased riboflavin biosynthesis in Japan, the USA, and Germany were different from those in China1, China2, and Taiwan. Similarly, different bacteria accounted for decreased biotin biosynthesis in the two country groups. We postulate that decreased SCFAs and polyamines reduce the intestinal mucus layer, which subsequently facilitates the formation of abnormal α-synuclein fibrils in the intestinal neural plexus in PD, and also cause neuroinflammation in PD.