Development of an ambient ionisation mass spectrometry method for metabolomic analysis of blood microsamples

Atmospheric Solids Analysis Probe mass spectrometry (ASAP-MS) has applications in food science, pharmaceuticals and toxicology but has not been applied to global metabolic profiling of biofluids. Coupled with dried blood microsampling techniques, ASAP-MS may provide rapid blood sample analyses benefitting medical and forensic sciences. This study aimed to develop a methodology utilising ASAP-MS for the metabolomic analyses of blood samples and comparison of venipuncture (serum and plasma) and capillary dried blood microsamples (DBM) using dried blood spot (DBS) and volumetric absorptive microsampling (VAM) matrices. Method development and optimisation on the Waters RADIAN-ASAP® assessed desolvation gas temperature, cone voltage, and corona current. The optimised parameters were applied to caffeine and lipid standards to determine the reproducibility of the instrument. The sample extraction methods (solvent, dilution, storage vial) and sampling regime for introducing samples into the ion source (sample volume, cooling mechanisms between acquisition) were assessed. The optimised methods were applied to venipuncture and capillary microsamples, and results were compared to ultra-high performance liquid chromatography-mass spectrometry {UPLC-MS). The coefficient of variation {CV) and principal component analysis (PCA) was used to assess analytical reproducibility. ASAP-MS parameters of 600 °C gas temperature, cone voltage of 15 V and corona current of 4 µA generate optimal signal intensity. Methanol extractions of caffeine in polypropylene microcentrifuge tubes produced the most reproducible data {CV = 4.7%), compared to extractions with acetonitrile {CV = 13.5%), isopropyl alcohol {CV = 15.2%) or methanol extractions in glass {CV= 8.1%). A methanol quench-bath to cool the glass sampler between acquisitions shortened the analytical run time while improving reproducibility {CV = 2.3%). UP LC-MS analysis of the lipid compound mixture detected 52 lipid species, of which ASAP-MS reproducibly detected 47 across both polarities. ASAP-MS detected fewer molecular features with a CV