Associations Between Gut Microbiota Composition and Impulse Control Disorders in Parkinson's Disease

Abstract

Impulse control disorders (ICDs) are a debilitating non-motor symptom of Parkinson's disease (PD), often associated with dopaminergic therapy. However, their occurrence in some patients but not others suggests additional biological mechanisms, including the gut microbiome. In this study, we analyzed 191 PD patients (14 with ICDs, 177 without) using 16S rRNA gene sequencing to explore the association between gut microbiota and ICDs. No significant differences were observed in alpha or beta diversity between groups, but several bacterial taxa showed differential abundances. Notably, Methanobrevibacter and Intestinimonas butyriciproducens were enriched in ICD patients. Functional pathway analysis revealed differences in metabolic pathways, including enrichment of xenobiotic degradation and nicotinate metabolism in the ICD group. These findings suggest that specific gut microbial taxa and their associated metabolic functions may contribute to ICDs in PD, highlighting a potential non-dopaminergic mechanism and opening new avenues for microbiome-targeted intervention.

Keywords: Parkinson’s disease; gut microbiota; gut–brain axis; impulse control disorders; microbiome metabolism.

Figure 1

Alpha and beta diversity of gut microbiota in PD patients with and without impulsive behaviors. (A) Boxplots of alpha diversity indices (Observed, Chao1, Shannon, and Simpson) show lower diversity in the impulsive group, but the differences were not statistically significant. (B) Beta diversity analysis using Canberra, Bray–Curtis, Unweighted UniFrac, and Weighted UniFrac metrics indicates no significant differences between the two groups, suggesting similar overall microbial composition.

Figure 1

Alpha and beta diversity of gut microbiota in PD patients with and without impulsive behaviors. (A) Boxplots of alpha diversity indices (Observed, Chao1, Shannon, and Simpson) show lower diversity in the impulsive group, but the differences were not statistically significant. (B) Beta diversity analysis using Canberra, Bray–Curtis, Unweighted UniFrac, and Weighted UniFrac metrics indicates no significant differences between the two groups, suggesting similar overall microbial composition.

Figure 2

Differential abundance analysis at the genus (A) and species (B) levels between impulsive and non-impulsive Parkinson’s disease patients, visualized by volcano plots. (A) At the genus level, several taxa exhibited significant differences before FDR correction. After correction, Methanobrevibacter remained significantly enriched in the impulsive group. (B) At the species level, Intestinimonas butyriciproducens was the only species that remained significantly enriched in the impulsive group following FDR correction. Genera and species that remained significant after correction are highlighted in bold. In the volcano plots, the horizontal dashed line marks the p-value threshold of 0.05, with taxa above this line considered statistically significant. The vertical solid line indicates the log2 fold change of zero. Therefore, taxa on the right are positively associated with impulsivity, while those on the left are negatively associated.

Figure 3

Network analysis illustrating associations gut bacterial genera in (A) impulsive group and (B) non-impulsive group. Each node represents a genus, with node color indicating whether that genus is enriched in the impulsive group (blue), enriched in the non-impulsive group (orange), or shows no differential abundance (gray). Edges depict Spearman’s rank correlations between genera: pink edges represent positive correlations, and blue edges represent negative correlations; thicker edges indicate stronger associations.

Figure 4

Comparison of metabolic pathway differences between impulsive and non-impulsive Parkinson’s disease patients. Relative abundances of predicted functional pathways were compared between groups using PICRUSt2 and STAMP. After false discovery rate (FDR) correction, six pathways were significantly associated with impulsive behaviors (q < 0.05). The most significant differences were observed in nicotinate and nicotinamide metabolism and biosynthesis of type II polyketide products. The 95% confidence intervals for differences in mean proportions are shown.