All data except the untargeted metabolomics analysis were quantified by integrating the area under the curve of each metabolite and its isotopologue using MassHunter Quant (Agilent Technologies). Each metabolite’s accurate mass ion and subsequent isotopic ions were extracted (EIC) using a 10 ppm window. In the case of isotopic labeling with 13C6-glucose or 15N2-arginine, corrections for natural enrichments were generated for each metabolite (Ctotal−1 or Ntotal−1) using a previously developed algorithm42 (link). MIDs of the indicated metabolites were calculated as previously described42 (link) and expressed as percent of the untreated PBS control set at 1.0. Pathway enrichment, principal component analysis (PCA), fold change and ANOVA statistical analysis were performed using Metaboanalyst software (www.metaboanalyst.ca). Pathway analysis was performed comparing the log2 fold change data between two conditions per analysis, BAD SAHBA SD versus vehicle, and BAD SAHBA SD versus RO0281675, all in the presence of cytokines (Fig. 1i).