Our study was approved by the Institutional Review Board of the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences. All human subjects provided informed consent before participating in our study. The first morning urine (midstream) was collected at around 7:00 and 9:00 a.m. on an empty stomach from a cohort of 315 clinically healthy adults and 348 clinically healthy children. Their detailed demographics shown in Supporting Information Table S1.

Acetonitrile (200 ml) was added to each urine sample (200 ml), then the mixture was vortexed for 30 sec and centrifuged at 14,000g for 10 min. The supernatant was dried under vacuum and then reconstituted with 200 ml of 2% acetonitrile. Urinary metabolites were further separated from larger molecules using 10 kDa molecular weight cut-off ultracentrifugation filters (Millipore Amicon Ultra, MA) before transferred to the autosamplers. The final samples of children and adults was a pooled urine sample prepared by mixing aliquots of 315 adults samples and 348 children samples. The quality control (QC) sample was a pooled urine sample prepared by mixing aliquots of fifty representative samples across different groups to be analyzed and was therefore globally representative of the whole sample set. The QC samples were injected every 10 samples throughout the analytical run to provide a set of data from which method stability and repeatability can be assessed. The 348 children metabolites samples and 315 adults metabolites samples were separately pooled with equal amounts of metabolites into one sample for 1D, 2D analyses .

The metabolites mixtures were dried under vacuum, redissolved in 2% acetonitrile and fractionated with a HPLC column from Waters (4.6 mm × 250 mm, Xbridge C18, 3 μm). Each metabolites mixture was loaded onto the column in buffer A2 (H2O, pH 10). The elution gradient was 5−30% buffer B2 (90% ACN, pH 10; flow rate, 1 mL/min) for 30 min. The eluted metabolites were collected as one fraction per minute. The dried 30 fractions were resuspended by 2% acetonitrile and pooled into 15 samples by combining fractions 1 and 16, 2 and 17,3 and 18 and so on. A total of 15 fractions from one sample were analyzed by LC−MS/MS.

Ultra-performance LC-MS analyses of urine samples were conducted using a Waters ACQUITY H-class LC system coupled with a LTQ-Orbitrap mass spectrometer (Thermo Fisher Scientific, MA). Urinary metabolites were separated with a 29-min gradient on a Waters HSS C18 column (3.0 3 100 mm, 1.7 lm) at a flow rate of 0.3 ml/min. The mobile phase A was 0.1% formic acid in H2O and the mobile phase B was acetonitrile. The gradient was set as follows: 0–2 min, 2% solvent B; 2–5 min, 2–55% solvent B; 5–15 min, 55– 100% solvent B; 15–20 min, 100% solvent B; 20–20.1 min, 100–2% solvent B; 20.1–29 min, 2% solvent B. The column temperature was set at 50℃. Full MS acquisition scanned from 100 to 1000 m/z at a resolution of 60 K. Automatic gain control (AGC) target was 1× 106 and maximum injection time (IT) was 100 ms. UPLC targeted-MS/MS analyses were acquired at a resolution of 15 K with AGC target of 5× 105, maximum IT of 50 ms, and isolation window of 3 m/z. Collision energy was optimized as 20, 40, and 60 for each target with higher-energy collisional dissociation (HCD) fragmentation. A total of 1008 LC-MS/MS runs were conducted, including 547 runs for the urine samples of adults and 461 runs for the samples of children.

The data were processed based on a previously described strategy performed by our lab[14]. Raw data files were processed by the Progenesis QI (Waters, Milford, MA) software based on a previously published identification strategy. The detailed workflow for QI data processing and metabolites identification is given in Supporting Information. The identification results were exported as .csv files for subsequent compound confirmation and multivariate statistical analysis. Confirmation of the differential compounds was performed by the parameters, including Score, Fragmentation score given by Progenesis QI. Score ranging from 0 to 60, is used to quantify the reliability of each identity. According to the score results of the reference standards, the fragment score threshold was set at 20.0. The identified metabolites were screened by package R to distinguish endogenous and exogenous metabolites. Metabolic pathways were analyzed using the “Mummichog” algorithm based on the MetaboAnalyst 3.0 platform. Further functional classification of the metabolites identified in urine was performed using the Ingenuity Pathway Analysis (IPA) tool.