A Multimodal Desorption Electrospray Ionisation Workflow Enabling Visualisation of Lipids and Biologically Relevant Elements in a Single Tissue Section

Abstract

The colocation of elemental species with host biomolecules such as lipids and metabolites may shed new light on the dysregulation of metabolic pathways and how these affect disease pathogeneses. Alkali metals have been the subject of extensive research, are implicated in various neurodegenerative and infectious diseases and are known to disrupt lipid metabolism. Desorption electrospray ionisation (DESI) is a widely used approach for molecular imaging, but previous work has shown that DESI delocalises ions such as potassium (K) and chlorine (Cl), precluding the subsequent elemental analysis of the same section of tissue. The solvent typically used for the DESI electrospray is a combination of methanol and water. Here we show that a novel solvent system, (50:50 (%v/v) MeOH:EtOH) does not delocalise elemental species and thus enables elemental mapping to be performed on the same tissue section post-DESI. Benchmarking the MeOH:EtOH electrospray solvent against the widely used MeOH:H₂O electrospray solvent revealed that the MeOH:EtOH solvent yielded increased signal-to-noise ratios for selected lipids. The developed multimodal imaging workflow was applied to a lung tissue section containing a tuberculosis granuloma, showcasing its applicability to elementally rich samples displaying defined structural information.

Keywords: biological tissue analysis; correlative imaging; desorption electrospray ionisation mass spectrometry; ion beam analysis; multimodal imaging.

Figure 1

PIXE maps taken post-DESI from 3 sequential liver homogenate sections using an ROI scanned over the edge of the DESI ROI using (A) 95:5 (%v/v) MeOH:H₂O and (B) 50:50 (%v/v) MeOH:EtOH. The DESI (dotted line) and PIXE ROIs (solid line) are indicated on the far-left image. (C) Average ratio (n = 3) of the elemental peak areas measured by PIXE from ROIs with/without prior DESI. See Figure S1 for further information on the analysis locations for each technique.

Figure 1

PIXE maps taken post-DESI from 3 sequential liver homogenate sections using an ROI scanned over the edge of the DESI ROI using (A) 95:5 (%v/v) MeOH:H₂O and (B) 50:50 (%v/v) MeOH:EtOH. The DESI (dotted line) and PIXE ROIs (solid line) are indicated on the far-left image. (C) Average ratio (n = 3) of the elemental peak areas measured by PIXE from ROIs with/without prior DESI. See Figure S1 for further information on the analysis locations for each technique.

Figure 2

Spectra taken from regions of interest capturing tissue-only areas acquired with DESI using the two solvent systems—95:5 (%v/v) methanol/water (blue) or 50:50 (%v/v) methanol/ethanol (green).

Figure 3

(A) Number of lipid features (out of the top 50) detected per lipid classes for MeOH:H₂O and MeOH:EtOH spray solvents, respectively; (B,C) Top 10 most abundant peaks and their respective intensities (normalised to the total ion count) for (B) MeOH:H₂O and (C) MeOH:EtOH. * refers to peaks detected using both solvents in liver homogenates.

Figure 4

Red, green and blue (RGB) overlay of m/z 953 (TG (58:8) [M+Na]⁺), m/z 832 (PC (38:4) [M+K]⁺) and m/z 780 (PC(36:5) [M+H]⁺) obtained using DESI and two spray solvents—(A) 95:5 (%v/v) MeOH:H₂O and (B) 50:50 (%v/v) MeOH:EtOH.

Figure 5

Chlorine (Cl), potassium (K) and iron (Fe) PIXE maps taken from snap-frozen lung tissue sections after DESI analysis using 95:5 (%v/v) MeOH:H₂O or 50:50 (%v/v) MeOH:EtOH. A third section (control) was also analysed—no DESI measurements were taken on this sample. Optical images taken before analysis highlighting the PIXE/EBS analysis areas (red squares) on each tissue.