Extensive evaluation of sample preparation workflow for gas chromatography-mass spectrometry-based plasma metabolomics and its application in rheumatoid arthritis.

GC-MS platform has been proved to be an important analytical technique for plasma metabolomics, but lack of efficient sample preparation strategies prior to sample injection has limited its wide application. In the present work, twenty two extraction protocols including protein precipitation (PPT), liquid-liquid extraction (LLE), solid-phase extraction (SPE) and ultrafiltration were simultaneously evaluated using non-spiked and metabolite standards spiked human plasma. Sample-to-extraction solvent ratio and metabolites derivatization conditions were also investigated and optimised. The results demonstrated that stepwise LLE protocol (MeOH-H2O/CHCl3/CHCl3-MeOH) was the most efficient extraction strategy for the combination of untargeted and targeted metabolomics. Sample-to-extraction solvent ratio of 1:3 (v:v) was found to perform better than higher solvent ratios. Also, the maximum derivatization performance was obtained when using 30 μL N-methyl-N-tri-methylsilyltrifluoroacetamide (MSTFA) and 1% trimethyl-chlorosilane (TMCS) incubated at 60 °C for 120 min. In the end, the optimised sample preparation method was applied successfully to plasma untargeted metabolomics of rheumatoid arthritis (RA) and four metabolites (gamma-butyrolactone, MG (0:0/18:0/0:0), tocopherol and oxalic acid) were found to be potential biomarkers for the diagnosis of RA.