Fig. 7. Metabolites of microbial neuroactive potential: integrated analyses of cerebral cortical brain, fecal and blood serum metabolomes.
a Variable importance plot of top 50 brain metabolites (y-axis) ranked by contribution to mean decrease accuracy of Gini coefficient (x-axis) in the random forest model for discerning group difference; embedded was a MetaMapp network view of all 58 metabolites altered in cerebral cortical brain tissues owing to microbiota, with nodes representing individual metabolites, edges for biochemical (Kyoto Encyclopedia of Genes and Genomes, i.e., KEGG reactant pairs) and chemical (Tanimoto coefficient > 0.7) relationships and lower transparency for lower p values (<0.05, two-sided Welch’s t test). b Box and Whisker plots of select metabolites exhibiting systemic alterations across feces, blood sera, and cerebral cortical brain tissues as mediated by microbiota, with the box ranging from the first quartile to the third while the whiskers going from each quartile to the minimum or maximum (n = 24), *p < 0.05, **p < 0.01, *** < 0.001, ****p < 0.0001, two-sided Welch’s t test; exact p values and adjusted p values (i.e., q values) are provided in Supplementary Data 4. (c) Structural annotation of ion feature m/z 212.002 at the retention time of 4.9 min as indoxyl sulfate that was highly enriched in CONV-R blood sera (4351.6-fold) and cerebral cortical brain tissues (26.8-fold) compared with GF mice. PAG N-(2-phenylacetyl)glycine, TMAO trimethylamine N-oxide, Gly-Phe glycine-phenylalanine dipeptide, HCD higher-energy C-trap dissociation, EIC extracted ion chromatogram, m/z mass-to-charge ratios, RTexp experimental retention time (from data), RTdb reference retention time (from chemical standard).