MSIpattern – JPND project application – Pathways 2017

<< back


CIHR-logo BMBF-logo RCN

MSIpattern: Neurodegeneration metabolomics across time and space
National funding organisations: CIHR (Canada), BMBF (Germany), NFR (Norway), MRC (UK), SRC (Sweden), NRDIO (Hungary)

Jens Pahnke, University of Oslo (UiO), Medical Faculty, KlinMED, Translational Neurodegeneration Research and Neuropathology Lab, Oslo, Norway

Per Andrén, University Uppsala, Sweden
Zoltan Takats, Imperial College, London, UK
Theodore Alexandrov, EMBL, Heidelberg, Germany
Jörg Gsponer, UBC, Canada
Olaf Riess, University of Tübingen, Germany
László Vigh, BRC Szeged, Hungary

Maciej Lalowski, Meilahti Proteomics Center, University of Helsinki, Finland
Dan Frenkel, Tel Aviv University, Israel
Eleonora Aronica, AMC, The Nederlands
Masato Yasui, Keio University, Tokyo, Japan
Patricía Maceil, University of Minho, Braga, Portugal
Saleh Ibrahim, University of Lübeck, Germany
Iliya Lefterov, University of Pennsylvania, Pittsburgh, USA

Neurodegenerative diseases are denoted by the deposition of protein aggregates in specific locations of the brain. The depositions usually follow a characteristic spatial-temporal pattern and, thus, result in specific clinical symptoms due to the affection of the neuronal function or even local neuronal death. Interestingly, these pathological changes have been found associated with primary or secondary white matter changes and alterations in various metabolic pathways. Indeed, within the framework of the JPND NeuroGeM project, which focuses on the discovery of disease mechanisms that are common to neurodegenerative disorders, we discovered the involvement of lipid metabolism-related pathways in Parkinson’s disease as well as in SCA3. Furthermore, the key risk factor for Alzheimer’s disease is a protein that is involved in lipid transport and lipid signalling: Apolipoprotein E. We and others have also found that ABC transporters at the blood-brain barrier (ABCA1, A7), which are involved in the transport of lipids, are regulated by ApoE3 / E4 variants. These findings suggest a strong link between neurodegenerative changes in the human brain and altered metabolism, which may open up new avenues for the design of drugs that specifically interfere in metabolic processes and slow down or even prevent neurodegeneration. However, it is still unclear how changes in metabolic processes tie in with the prevalent concepts that see protein misfolding and aggregation as key pathogenesis factors in neurodegeneration. A major obstacle in establishing and characterizing this link has been the lack of appropriate technologies to map and quantify changes in metabolites at high resolution across time and space in disease and normal brains. Here, we propose to overcome this obstacle by using a new technology that combines an innovative ionisation method (Desorption Electrospray Ionization – DESI), which was invented by one of the consortium members (Zoltan Takats), with mass spectrometry (MS) in order to spatially map patterns of metabolites with high resolution (2D) directly from tissue samples of diseased and aged brains. Through the use of different animal models of neurodegenerative diseases as well as aging control animals, we intent to specifically map lipid and neurotransmitter pattern at different points in time of the diseases or the normally aging animals. This data will then be integrated with available “omics” information in directed gene networks in order to link the newly established metabolic changes upon disease progression or aging with those previously measured for the epigenome, transcriptome, and proteome. Our hypothesis is (i) that our approach will allow us to establish direct links between previously detected changes in metabolic pathway genes in neurodegenerative diseases with specific alterations in lipid as well as neurotransmitter patterns of disease brains and (ii) that some spatial and temporal metabolite changes are specific to certain neurodegenerative diseases whereas others are common to all of them. To test this hypothesis, we propose to pursue the following aims: i) to generate space/time specific lipidomics and neurotransmitter DESI-MSI data sets for each of the disease models, ii) to generate DESI-MSI data for aging, non-disease models, iii) to verify the location patterns with independent LCM-LCMS methods, iv) to generate graphical MSI data presentation and quantification of lipidomics and neurotransmitter data together with morphological data, v) to integrate the lipidomics and neurotransmitter data with previously collected and new transcriptomic, proteomic data in directed gene networks and (vi) to identify correlated metabolic and transcriptomic/proteomic changes that are specific or common, respectively to different neurodegenerative diseases. In order to achieve these ambitious aims, we have assembled a unique team with specialists in the field of MSI development (Zoltan Takats), MSI neurotransmitter metabolism (Per Andrén), MSI analysis (Theodore Alexandrov), lipidomics (Laszlo Vigh), animal studies of neurodegenerative diseases (Olaf Riess, Jens Pahnke), and computational analysis of omics data (Jörg Gsponer, Theodore Alexandrov). The consortium is complemented by collaborators who deliver additional expertise in MSI/MS (Maciej Lalowski), animal models of neurodegenerative diseases (Patricia Maciel, Matsato Yasui, Dan Frenkel), ApoE3/4 and lipid metabolism (Iliya Lefterov), aging (Saleh Ibrahim), and also deliver patient samples (Eleonora Aronica).
Our unique collaborative setup will allow us, for the first time, to provide detailed maps of the metabolomics changes occurring at different time points of normally aging brains as well as brains with progressing neurodegenerative disease. The integration of the collected metabolomics data with other “omics” datapoints from disease models will provide a clearer picture of the metabolic pathways altered in individual or multiple neurodegenerative diseases and how metabolomic changes are linked to known pathophysiological changes in the brain and disease stages, respectively. Such links are likely to help in the identification of new biomarkers (e.g. PET tracers for specific metabolites) that could help in diagnosis and treatment monitoring as well as in clinical studies of new therapies. In addition, a better knowledge of metabolic pathways involved in neurodegeneration will enable us to pinpoint potential new targets for disease modulation and treatment. We also would like to find explanations why the lipid-related protein ApoE is so far the best risk marker for AD.

Resources / Links:

Andrén lab

Alexandrov lab

Gsponer lab