Evidence of the Dysbiotic Effect of Psychotropics on Gut Microbiota and Capacity of Probiotics to Alleviate Related Dysbiosis in a Model of the Human Colon

Abstract

Growing evidence indicates that non-antibiotic therapeutics significantly impact human health by modulating gut microbiome composition and metabolism. In this study, we investigated the impact of two psychotropic drugs, aripiprazole and (S)-citalopram, on gut microbiome composition and its metabolic activity, as well as the potential of probiotics to attenuate related dysbiosis using an ex vivo model of the human colon. After 48 h of fermentation, the two psychotropics demonstrated distinct modulatory effects on the gut microbiome. Aripiprazole, at the phylum level, significantly decreased the relative abundances of Firmicutes and Actinobacteria, while increasing the proportion of Proteobacteria. Moreover, the families Lachnospiraceae, Lactobacillaceae, and Erysipelotrichaceae were also reduced by aripiprazole treatment compared to the control group. In addition, aripiprazole lowered the levels of butyrate, propionate, and acetate, as measured by gas chromatography (GC). On the other hand, (S)-citalopram increased the alpha diversity of microbial taxa, with no differences observed between groups at the family and genus level. Furthermore, a probiotic combination of Lacticaseibacillus rhamnosus HA-114 and Bifidobacterium longum R0175 alleviated gut microbiome alterations and increased the production of short-chain fatty acids to a similar level as the control. These findings provide compelling evidence that psychotropics modulate the composition and function of the gut microbiome, while the probiotic can mitigate related dysbiosis.

Keywords: (S)-citalopram; aripiprazole; gut dysbiosis; gut microbiome; probiotics; psychotropics.

Figure 1

Modulation of microbiota composition following treatment with aripiprazole and probiotics. (A) observed features, (B) Shannon index, (C) Faith’s phylogenetic diversity, and (D) Pielou evenness of all treated samples (n = four biological replicates each). Results were calculated from rarefied 12,000 reads per sample using QIIME2. 2020.6 version. Middle lines represent the mean. Data were analyzed using the Kruskal- Wallis test and Two-stage Benjamini, Krieger, and Yekutieli FDR procedure; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Red dots represent the zero time points.

Figure 2

Modulation of microbiota composition following treatment with (S)-citalopram and probiotics. (A) observed features, (B) Shannon index, (C) Faith’s phylogenetic diversity, and (D) Pielou evenness of all treated samples (n = four biological replicates each). Results were calculated from rarefied 12,000 reads per sample using QIIME2. 2020.6 version. Middle lines represent the mean. Data were analyzed using the Kruskal- Wallis test and Two-stage Benjamini, Krieger, and Yekutieli FDR procedure; * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Red dots represent the zero time points.

Figure 3

Plots of Principal Coordinate Analysis (PCoA) based on Bray-Curtis distances among the identified microbiota in different samples. (A–C): effect of aripiprazole treatment; (D–F) effect of (S)-citalopram treatment. The PCoA figures show clustering based on source donor (A,D); experiment replicate (B,E) and treatments of each donor microbiota (C,F). The samples were colored as indicated in legends. PCoA1 and PCoA2 represent the top two coordinates that captured the highest microbial variability among samples, and the percentage shown indicates the fraction of variation represented by each coordinate.

Figure 4

Histograms of the linear discriminant analysis (LDA) scores showing microbial taxa that vary significantly in abundance between: (A) no-treatment control and aripiprazole treatment, (B) no-treatment control and aripiprazole + probiotic treatment, and (C) aripiprazole and aripiprazole + probiotic treatments.

Figure 5

Histograms of the linear discriminant analysis (LDA) scores showing microbial taxa that vary significantly in abundance between: (A) no-treatment control and (S)-citalopram treatment, (B) no-treatment control and (S)-citalopram + probiotic treatment, (C) (S)-citalopram and (S)-citalopram + probiotic treatments, (D) no-treatment control and probiotic treatment.

Figure 6

Short-chain fatty acids (SCFAs) concentration was measured by GC over 48 h with all treatment groups with aripiprazole (A–C) or Citalopram (D–F). (A,D) acetate, (B,E) butyrate, and (C,F) propionate. Each time point is represented with 4 biological replicates × 2 technical measures. Different treatments are compared statistically for each donor using one way ANOVA test followed by correction for multiple comparisons by controlling false discovery rate using the two-stage step up procedure of Benjamini Krieger and Yekutieli. * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 7

Experimental reactor set-up of the ex vivo fermentation model. IR: inoculum reactor, containing immobilized donor feces (30% v/v); CR: control reactor; TR1-TR3: test reactors 1–3; Stab: stabilization period (2 days); T: treatment period (2 days); W: wash period (1 day).