Differential effects of synthetic psychoactive cathinones and amphetamine stimulants on the gut microbiome in mice

Abstract

The list of pharmacological agents that can modify the gut microbiome or be modified by it continues to grow at a high rate. The greatest amount of attention on drug-gut microbiome interactions has been directed primarily at pharmaceuticals used to treat infection, diabetes, cardiovascular conditions and cancer. By comparison, drugs of abuse and addiction, which can powerfully and chronically worsen human health, have received relatively little attention in this regard. Therefore, the main objective of this study was to characterize how selected synthetic psychoactive cathinones (aka "Bath Salts") and amphetamine stimulants modify the gut microbiome. Mice were treated with mephedrone (40 mg/kg), methcathinone (80 mg/kg), methamphetamine (5 mg/kg) or 4-methyl-methamphetamine (40 mg/kg), following a binge regimen consisting of 4 injections at 2h intervals. These drugs were selected for study because they are structural analogs that contain a β-keto substituent (methcathinone), a 4-methyl group (4-methyl-methamphetamine), both substituents (mephedrone) or neither (methamphetamine). Mice were sacrificed 1, 2 or 7 days after treatment and DNA from caecum contents was subjected to 16S rRNA sequencing. We found that all drugs caused significant time- and structure-dependent alterations in the diversity and taxonomic structure of the gut microbiome. The two phyla most changed by drug treatments were Firmicutes (methcathinone, 4-methyl-methamphetamine) and Bacteriodetes (methcathinone, 4-methyl-methamphetamine, methamphetamine, mephedrone). Across time, broad microbiome changes from the phylum to genus levels were characteristic of all drugs. The present results signify that these selected psychoactive drugs, which are thought to exert their primary effects within the CNS, can have profound effects on the gut microbiome. They also suggest new avenues of investigation into the possibility that gut-derived signals could modulate drug abuse and addiction via altered communication along the gut-brain axis.

Fig 1. Effects of study drugs on α-diversity.

The α-diversity metrics Chao-1 richness estimator (A), Shannon diversity index (B) and Simpson (1-D) index (C) were determined for 16S rRNA gene profiles of caecum contents harvested 1 day after treatment. The individual values for all subjects in each treatment group are included in each box plot. *, p < 0.05 or *** p < 0.001 compared to control; # p < 0.05, ## p < 0.01 or #### p < 0.0001 compared to Meph; § p < 0.05, or §§ p < 0.01 compared to MeCa.

Fig 2. Effects of study drugs on β-diversity.

Principal Coordinates Analyses (PCoA) illustrating differences in the structure (i.e. Bray-Curtis index) of gut microbiome profiles among mice treated with the different study drugs. Profiles were generated at 1 (A), 2 (B) or 7 days (C) after drug treatments.

Fig 3. Heat map illustrating the relative abundances of OTUs after treatment with study drugs.

The most prominent OTUs (≥ 1% average relative abundance) among treatment groups are plotted for each drug at 1, 2 h or 7 days after drug injections. Clustering was done using the Ward algorithm.

Fig 4. Bacterial taxa that were differentially abundant across study drug treatments.

Linear discriminant analysis effect size (LEfSe) was carried out and the results are presented for taxa with LDA scores of > 3.6 for the treatment groups at 1 (A), 2 (B) and 7 days (C) after treatments.

Fig 5. Relative abundances of phyla after treatment with study drugs.

Results are presented as % relative abundance of each phylum for each study drug. Stacked columns for the 7 most prominent phyla are included for the 1- (A), 2- (B) or 7- day (C) time points after drug injections.

Fig 6. Effects of study drugs on selected taxa below the level of phylum.

Results are presented as % relative abundance of taxa 1, 2 or 7 days after drug injections for Bifidobacteriales (A), Mucispirillum (B), Erysipelotrichia (C) and Enterobacteriales (D). * p < 0.05, ** p < 0.01, *** p < 0.001 and **** p < 0.0001 for the comparisons demarked by connecting lines above the bars.