Untargeted Metabolomics and targeted metabolite analysis of CarT cells

19BBz CAR T cells with TRBC1/2 stop mutation (T) or additional E81K modification (E) were twice stimulated with Nalm6 cells (E:T 2:1). 24h after last stimulation, absence of Nalm6 cells was controlled by flow analysis and 1e7 - 1.2e7 CAR-T cells were thoroughly washed followed by quenching with 400 µL ice-cold LC-MS grade methanol (≥99.9%) and stored at -80 °C. Metabolite extraction was performed on ice using a two-phase extraction with methanol, methyl tert-butyl ether (MTBE) and H₂O as described elsewhere in detail (https://doi.org:10.1186/s40170-025-00391-5). Briefly, samples were vortexed and ultrasonicated (Covaris) using two adaptive focused acoustics (AFA) cycles. After extraction, 300 µL LC-MS grade water was added, followed by centrifugation to induce phase separation of the aqueous and organic phase and the upper organic and lower polar phases were separated. Polar phase was evaporated in vacuum concentrator and reconstituted in 50 µL acetonitrile:water (1:1). LC-MS analysis was performed using a chromatography system (Elute Plus, Bruker Daltonics) coupled to a trapped ion mobility spectrometry time of flight (timsTOF) mass spectrometer equipped Pro 2 with a vacuum insulated pressure heated electrospray ionization (VIP-HESI) source (UPLC timsTOF Pro2 system, Bruker Daltonics). Hydrophilic interaction liquid chromatography (HILIC) was performed using a BEH Amide column (150 × 2.1 mm, 1.7 µm; Waters). To manually assess signal linearity, a pooled quality control (QC) sample was injected at 0.25 µl, 1.25 µl, 2.5 µl and 5 µL. A recalibration method was applied for both mass accuracies, using Natriumformate, and ion mobility, with Agilent’s ESIS Tuning Solution, as a calibrant was injected with each run. The MS analysis for non-targeted MS metabolomics was performed using Parallel Accumulation Serial Fragmentation (PASEF) mode with data-dependent MS/MS acquisition and Trapped Ion Mobility Spectrometry (TIMS) stepping, following the 4D metabolomics standard method in TimsControl software (Bruker Daltonics). HILIC was run in both ionization modes; RP in negative mode only.