Use of the second-generation antipsychotic, risperidone, and secondary weight gain are associated with an altered gut microbiota in children

Abstract

The atypical antipsychotic risperidone (RSP) is often associated with weight gain and cardiometabolic side effects. The mechanisms for these adverse events are poorly understood and, undoubtedly, multifactorial in etiology. In light of growing evidence implicating the gut microbiome in the host's energy regulation and in xenobiotic metabolism, we hypothesized that RSP treatment would be associated with changes in the gut microbiome in children and adolescents. Thus, the impact of chronic (>12 months) and short-term use of RSP on the gut microbiome of pediatric psychiatrically ill male participants was examined in a cross-sectional and prospective (up to 10 months) design, respectively. Chronic treatment with RSP was associated with an increase in body mass index (BMI) and a significantly lower ratio of Bacteroidetes:Firmicutes as compared with antipsychotic-naïve psychiatric controls (ratio=0.15 vs 1.24, respectively; P<0.05). Furthermore, a longitudinal observation, beginning shortly after onset of RSP treatment, revealed a gradual decrease in the Bacteroidetes:Firmicutes ratio over the ensuing months of treatment, in association with BMI gain. Lastly, metagenomic analyses were performed based on extrapolation from 16S ribosomal RNA data using the software package, Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt). Those data indicate that gut microbiota dominating the RSP-treated participants are enriched for pathways that have been implicated in weight gain, such as short-chain fatty acid production.

Figure 1

Differences in the fecal microbial communities in chronic risperidone (RSP)-treated participants vs psychiatric controls. PcoA of unweighted UniFrac distances between psychiatric control participants (green), chronic RSP-treated participants with significant body mass index (BMI) gain (blue, i.e., an increase in age−sex-specific BMI Z-score⩾0.5 units since starting RSP), and RSP-treated participants without significant BMI gain (yellow). Each point shows the average distance between individuals. Results are derived from bacterial V1−V2 16S rRNA data sets. ANOSIM R=0.5169, P=0.0001. PCoA, principal component analysis.

Figure 2

Phyla-level abundances. (a) Phyla-level abundances in psychiatric controls (green), chronic risperidone (RSP)-treated participants with significant body mass index (BMI) gain (blue, that is, an increase in age–sex-specific BMI Z-score ⩾0.5 units since starting RSP) and chronic RSP-treated participants without significant BMI gain (yellow). (b) Trajectory of change over time in the two major gut bacterial phyla, Bacteroidetes and Firmicutes, following the initiation of RSP. (c) Correlation of change in percent abundance of Bacteroidetes and Firmicutes, and change in BMI Z-score following the initiation of RSP.

Figure 3

Relative abundance of unique genus-level operational taxonomic units (OTUs) of the gut microbiota of chronic risperidone (RSP)-treated participants and psychiatric controls. Fifty unique OTUs were defined by three methods (Metastats comparison, random forests algorithm and linear discriminant analysis effect size analysis). Genus-level OTUs are colored by phyla (bottom, x-axis) and individual biological replicates (left, green or blue), where red hues denote increased relative abundance of a unique OTU. BMI, body mass index.

Figure 4

Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) predicted KEGG (Kyoto Encyclopedia of Genes and Gene Systems) orthologs. 1212 Significant KEGG orthologs were predicted with PICRUSt software. (a) PCoA of KEGG orthologs in chronic risperidone (RSP)-treated participants vs psychiatric controls. (b, c) Global and individual KEGG pathways of chronic RSP-treated participants (blue) vs psychiatric controls (green). PCoA, principal component analysis.