Single-cell metabolite annotation by tandem mass spectrometry imaging and <i>ab initio</i> molecular dynamics-based fragmentation

Reliable single-cell mass spectrometry (MS) imaging and metabolite annotation represent a challenge due to small cellular dimensions, small amount of desorbed materials and densely populated databases of the <i>m</i>/<i>z</i> values of endogenous compounds. To resolve these issues, a highly sensitive analytical approach was devised for metabolite annotation purposes, relying on the correspondence between the <i>m</i>/<i>z</i> values from single-cell MS/MS spectra and the <i>m</i>/<i>z</i> values of the molecular fragments calculated using <i>ab initio</i> molecular dynamics (AIMD). The approach was applied to the annotation of <i>m</i>/<i>z</i> 337.11 Da, which cellular content is increased in chronic lymphocytic leukemia (CLL). To evaluate the approach, five candidate compounds were selected by a metabolite database search using the given <i>m</i>/<i>z</i>. Matrix-assisted laser desorption/ionization ion trap-time-of-flight (MALDI IT-TOF) MS/MS spot analysis of S-nitrosoglutathione (GSNO), one of the candidate compounds, preceded single-cell MS/MS imaging: five fragments were present in the empirical and <i>in silico</i> spectra of the GSNO solution. The sensitivity of single-cell MS imaging was optimized, and MS/MS spectra were recorded for different lymphocytes containing 0-3 fragments that were present in the <i>in silico</i> spectra of three glutathione-related compounds; however, there was no match between the empirical and <i>in silico</i> spectra of the other candidate compounds. The lateral distribution of the selected fragment showed ∼3 μm shift with respect to the optical image of the lymphocyte. The novel concept developed for single-cell MS imaging enabled metabolite annotation in malignant lymphocytic clones and showed potential for metabolite annotation in other cell suspensions.