Mass spectrometry imaging (MSI) allows for the correlation of spatial localization and chemical information directly from biological surfaces. Typically untargeted MSI experiments are carried out in full scan MS. After mining of the MSI data, the next step is the identification of potential biomarkers which is usually a limited number of manual MS/MS experiments.
Recently a new Data Independent Acquisition (DIA) method called SONAR utilising a scanning quadrupole mass filter has been introduced with multiple functions: one with no collision energy (CE) and one with elevated CE to get precursor and fragment information within the same experiment. Here, we describe a new acquisition method that collects MS/MS results from a Data Independent acquisition (DIA) embedded into a DESI imaging workflow.
Experiments have been carried out on mouse brain tissue sections, produced using a cryotome and deposited on a standard microscope slide which was preserved at -80C degrees until analysis by mass spectrometry.
The tissues were directly mounted into the DESI source from the freezer, with no sample preparation or pre-treatment required. Data were acquired using a DESI source (Prosolia) which was mounted on a Xevo G2-XS mass spectrometer (Waters), with an acquisition mass range of m/z 50-1,200, in both positive and negative mode. DESI spray conditions were set at 2µl/min, 98:2 MeOH: water with nebulizing gas pressure of 5bar. DESI Imaging was set using High Definition Imaging (HDI) 1.4 and datasets were subsequently mined using MassLynx and DriftScope and processed/visualized HDImaging 1.4.
Here we have assessed the applicability of this new method and optimisation of settings for a DESI imaging analysis. The Sonar method for DESI imaging consisted of two alternating functions. In both cases the quadrupole was scanned multiple times across the mass range with a pre set quadrupole window. In the first function (precursor function) the collision energy was fixed at 6ev, in the second function collision energy was applied to fragment the ions (MS/MS function). The functions alternated between pixels to generate images of precursors and of fragments in a single experiment. The precursor and MS/MS functions were subsequently time aligned to relate the fragments to precursors for identification of multiple species from a single imaging run. Proof of concept experiments have been performed analysing a mouse brain tissue section in negative mode scanning the quadrupole from m/z 750-950 with a quad window of 8 Da. For the MS/MS function the collision energy was fixed at 30eV. Reviewing the data with Driftscope and HDImaging a number of time aligned precursors / fragments could be identified. One example was a PS(18:0_22:6) (m/z 834.52), from which fragments of the neutral losses of serine(m/z 747.49), the sn1 and sn2 RCOOH groups + serine (m/z 463.23 and m/z 419.26) were observed as well as the sn1 / sn2 RCOO- ions (m/z 327.23 and 283.26). In addition to the time aligned nature of the precursor/ fragment spectra, the spatial distribution of the precursors and fragments in the imaging data could also be used to further refine the precursor fragment assignments. Further work will be performed to optimise the quad settings to maximise the number of identifications and sensitivity / specificity.