Faecal UPLC-MS/MS metabolomics to investigate the metabolic influence of the gut microbiome in an Australian Parkinson's Disease cohort

Jess A Spithoven

Background There is increasing evidence of an altered gut microbiome in individuals with Parkinson’s disease (PD). However, few studies have focused on the functional, metabolic impact of this dysbiosis. A reduction in microbial diversity, or change in enterotype, could alter the production of key health-promoting metabolites, and increase deleterious products, cultivating a gastrointestinal environment that activates the immune system and primes inflammation. Method To investigate the specific metabolic pathways influenced by the dysbiotic gut microbiome present in PD, this study collected faecal samples from an Australian cohort of individuals with idiopathic PD (n=58) at two time points for comparison as a matched pair, and alongside healthy, aged-matched controls (n=82). Following extraction, samples underwent high-throughput targeted metabolic phenotyping using a validated, ultraperformance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) approach. Results Following data pre-processing, 87 metabolites of interest were identified for analysis. Supervised orthogonal projection to latent structure-discriminant analysis (OPLS-DA) established separation both healthy controls (HC) and people with Parkinson’s disease (PwP) at timepoint 1 (R2X = 0.306, R2Y = 0.314, Q2 = 0.143); HC and PwP at timepoint 2 (R2X = 0.332, R2Y = 0.289, Q2 = 0.138); and the matched pair of PwP at timepoint 1 compared with PwP at timepoint 2 (R2X = 0.344, R2Y = 0.214, Q2 = -0.373). Univariate analysis highlighted 34 metabolites of interest between the groups. When comparing PwP at timepoint 1 to HC, we observed 13 significantly different metabolites, including lower concentrations of L-arginine (p = 3.21 x 10-5), L-histidine (p = 2.00 x 10-3), p-hydroxyphenylacetic acid (p = 7.81 x 10-4), cholic acid (p = 5.00 x 10-3), indole-3-propionic acid (p = 8.00 x 10-3), and nicotinic acid (p = 8.06 x 10-4), and higher concentrations of phenylpyruvic acid (p = 1.20 x 10-2), and isovaleric acid (p = 1.78 x 10-5). In PwP at timepoint 2 compared with HC, the results revealed 29 perturbed metabolites, including lower concentrations of L-lysine (p = 1.00 x 10-3), L-tyrosine (p = 1.00 x 10-2), and L-valine (p = 1.60 x 10-2) in PwP. The longitudinal analysis highlighted two altered metabolites, including a decreased level of indolepyruvate (p = 9.46 x 10-4) in PwP at timepoint 2 compared with timepoint 1. Conclusion This study provides further evidence of the functional impact related to gut dysbiosis present in PD. Future studies that validate the faecal phenotypes of PD will enable the identification of the disease's metabolic signatures, leading to enhanced understanding of the gut microbiome's mechanistic influence on host biochemical pathways in the pathogenesis of the disease.

Faecal UPLC-MS/MS metabolomics to investigate the metabolic influence of the gut microbiome in an Australian Parkinson's Disease cohort

Files and links (1)

Abstract

Details

Metrics