MSACL 2023 Abstract

1.Introduction
Alzheimer’s disease (AD) is the most frequent neurodegenerative disorders worldwide representing the first cause of dementia in the world. Biomolecular hallmarks of the pathology include intracellular neurofibrillary tangles of hyperphosphorylated TAU proteins and extracellular accumulation of senile plaques made of Amyloid β (Aβ) peptides oligomers. The multimodality levels of AD make challenging to decipher and unveiled the metabolic process behind the disease. Focusing on a particular molecular environment such as Aβ plaques surrounding is the first step to understand any metabolic change happening in the disease development. Previous worked showed that amyloid plaque production and formation is regulated by a crosstalk with lipids of neurons membrane. Moreover, the presence of plaques exacerbates neurocellular stress and production of reactive oxygen species (ROS) which will target surrounding membrane lipids causing lipids peroxidation (LPO), which will eventually lead to the disruption of neurons lipids membrane and the production of neurotoxic aldehydes. LPO consequences speed up the decline of neuron cells population and increase the production of plaques, therefor LPO contributes to the massive cognitive decline observed in AD. Being able to identify and correlate markers of LPO within plaques environment could be beneficial to understand how different marker of the pathology act altogether and how the disease spread. In order to be able to correlate the interaction of plaques with their environmental lipids, Desorption Electrospray Ionisation-Mass spectrometry imaging (DESI-MSI) is the most suited techniques as it combines sensitive, high lipids ionisation and rapidity of analyses in ambient condition and does not require sample preparation. We have established a multimodal approach to resolve and corollate the spatial lipidomic markers of lipids alteration within the Aβ plaques surrounding by combining DESI-MSI imaging with fluorescence microscopy (MF) imaging using ThioflavineS on post-DESI section to locate Aβ plaques. Imaging data were combined to target LC-MS method from the same brain samples to double confirm the identity of the pathological features.

2.Materials and Methods
Entorhinal cortex (EC) region from human AD brain (n=9) and control (n=9) were sectioned at 10um and analysed on a XEVO-G2S-QTOF (Waters, USA) with a DESI stage and source from (Waters) in both negative and positive mode at a rate of 250um/s, pixel size of 5um and 5 scans/s. For RP-LC-MS sections (20mg) from the same samples were collected and extracted using IPA/water and run on XEVO-G2S-QTOF(Waters, USA) with an ESI source in both modes RP-LC-MS for structural confirmation spectra were acquired using Masslynx V4.1 (Waters, USA) . Fluorescence staining was performed on post-desi slides using Propidium Iodine red (50mM) for nuclei staining and ThioFlavineS (Sigma Aldrich, UK) 0.2% in water, washed, and mounted. Fluorescence microscopy was done using a Leica Thunder widefield (Leica Microsystems Ltd, UK). Imaging data were analysed using an inhouse algorithm program on Matlab for spectral data pre-processing and Jupyter Notebook for classification, statistical analyses and features extraction. Fusion imaging of MSI and MF were also done using deconvolution and spatial mask overlay of the different lipid features. Network correlation was performed using cytoskape (3.9.1).
Aldehydes were extracted from the sample tissue (60mg) and labelled with 2,4-DNPH (0.2M) and quantify using deuterium standards. Analyses was done on Triple-Quadrupole (Waters,USA).

3.Results
Spatially resolved lipids showed a significant increase of LysoPls such as LysoPA (18:0) (m/z 419.2), LysoPE(18;0) (m/z 462.2), hydroxy-SM (d17:2/20:4) (m/z 749.5) in negative mode and an increase of PA (538.6), PC (16:0) (m/z 782.5) was observed in positive mode. SM 36:2 (m/z 727.5), while overall PI, plasmalogens and d18:1/18:0-CPE (m/z 689.3) levels were significantly depleted in AD. This was confirmed within the LC-MS dataset showing an overall decrease of PI, PS and an unbalanced of plasmalogen PE/PE ratio in AD compared to the control and increase of PAs and PCs species. HydroxyPLs can be linked to LPO. Structural analyses confirmed the accurate mass’s identity. Distribution of the Lipids species with the Aβ plaques showed a high a high correlation factors and confirm the network colocalization analyses showing a strong interaction of these lipids and hydroxy lipids within the plaque’s environment allowing to follow the progression of AD within the EC region. Degradation of the lipids seems to follow the density of amyloids plaques and is higher around diffused plaques than in single plaques, while poorly observed in the control region. Marker of aging brain like 2-amino-3-oxo-hexanedioic acid was also stronger in AD brain than in the control, confirming that environment of plaques speeds up the pathological process in AD. While aldehydes quantification showed a significant increase of 4-HNE and MDA, however quantification was not possible in every sample due to aldehydes volatile nature as they tend to decrease over time.

4.Conclusion
Our multimodal approach combining DESI-MSI and MF allowed to resolve and identify in a very short time pathological markers of LPO within Aβ plaques environment in AD EC region.