3.7. Untargeted Metabolomics by GC–MS

Primary metabolites were extracted by published protocol [75] with modifications on the extraction solvent volume. The samples (50 mg) were weighed in 2 mL microcentrifuge tubes, followed by addition of 80 μL of ribitol (10 mg/mL) as an internal standard, then extracted with 1.4 mL of methanol at 75°C. After cooling, the sample extracts were centrifuged at 15,000× g for 3 min, and 0.7 mL of supernatants was transferred to new 2 mL microcentrifuge tubes. To fractionate polar compounds, 0.375 mL of cold chloroform (−20 °C) and 0.7 mL cold water (4°C) were added. After vigorous mixing, the extracts were centrifuged at 15,000× g for 3 min, and then 50 μL supernatant was transferred to 1.5 mL microcentrifuge tubes. The extracts were dried using Vacufuge^TM concentrator (Eppendorf^TM, Thermo Fisher Scientific, Waltham, MA, USA) with 10 μL of methanol to facilitate water evaporation. The dried extracts were derivatized with 50 μL of methoxyamine hydrochloride (40 mg/ml in pyridine) for 90 min at 37°C, then with 70 μL MSTFA + 1%TMCS at 37°C for 30 min. The metabolites were analyzed using a GC–MS (Trace 1310 GC, Thermo Fisher Scientific, Waltham, MA, USA) coupled to an MS detector system (ISQ QD, Thermo Fisher Scientific, Waltham, MA, USA) and an autosampler (Triplus RSH, Thermo Fisher Scientific, Waltham, MA, USA). A capillary column (Rxi-5Sil MS, Restek, Bellefonte, PA, USA; 30 m × 0.25 mm × 0.25 µm capillary column w/10 m Integra-Guard Column) was used to detect polar metabolites. After an initial temperature held at 80°C for 2 min, the oven temperature was increased to 330°C at 15°C/min and held for 5 min. The injector and detector temperatures were set to 250°C and 250°C, respectively. An aliquot of 1 μL was injected with a split ratio of 70:1. The helium carrier gas was kept at a constant flow rate of 1.2 mL/min. The mass spectrometer was operated in positive electron impact mode (EI) at 70.0 eV ionization energy at m/z 40–500 scan range.

The acquired chromatograms were converted to mzXML using the RawConverter [84]. Peak detection and alignment were performed by XCMS package in R language with default settings [85]. All data were normalized to unique ion peak (319) from the internal standard in the online platform MetaboAnalyst [86], and further statistical analysis was conducted after Pareto scaling. Metabolite identification was based on standard compounds (STD) in comparison with the mass spectra present in The National Institute of Standards and Technology (NIST) and retention times (Table S4).