Footprints of amyloid aggregation in vivo

Structural biology, the branch of the life sciences that deals with characterization of 3D-structures of proteins and other biomolecules, is currently undergoing a rapid transformation. The “resolution revolution” in cryo-electron microscopy (EM) enabled a step-change in understanding complex structures, such as membrane proteins in native lipid environments and fibril structures of amyloid protein. By providing molecular snapshots with intricate detail, structural biology methods have thus greatly accelerated our mechanistic understanding of protein function – their internal movements (conformational changes) in response to external stimuli and their interactions with other proteins, DNA/RNA, lipids and glycans. These insights are key for understanding cellular function in health and disease and form the basis for the rational design of therapies. Recent advances in experimental methodology have been paralleled by innovations in computational modelling and structure prediction using e.g., Alphafold. Nevertheless, structural approaches usually regard the sample in isolation, without the context of its complex, diverse and often crowded cellular environment. Few methods exist to study molecules directly in cells (“in-situ structural biology”), notably super-resolution microscopy and electron tomography (ET). As exciting and powerful these methods are, they usually require labelling or “carving-out” of molecules, hence targeting specific features of interest. Structural-MS methods on the other hand are uniquely capable of identifying many different proteins and their modifications in complex and heterogeneous samples, while also providing information on conformations and interactions under certain circumstances. Here we propose to develop and test a novel approach for in-situ structural biology based on fragmentation patterns in secondary ion mass spectrometry (SIMS) – a high-resolution imaging method in surface science which we apply to biological questions such as amyloid aggregation in vivo. Mass spectrometry-based imaging can already scan across tissues to map distributions of small molecules, lipids and peptides; but we postulate now that SIMS can extend this approach to larger and insoluble structures such as proteins and amyloid fibrils in their native state and at subcellular resolution. Rather than releasing these biomolecules whole from tissue via e.g. liquid extraction, we will exploit their fragmentation patterns and how they correlate with the native 3D-structure. Using a combination of spectral interpretation and clustering, differential analysis and A.I. approaches, we can distinguish different structural features, characterize conformations and attempt to derive 3D-structures de-novo from experimental data and computational modelling. Native SIMS-imaging addresses key challenges of in-situ biology, a current frontier of structural molecular biology, with the ability to obtain unbiased molecular profiles of whole tissue sections at high spatial resolution while also providing footprints of native structures. We will develop this approach using reference proteins in defined conformational and oligomeric states and evaluate the ability of SIMS to detect ligand binding. We will then use this strategy to study amyloid aggregation in tissue samples, map hotspots of aggregation and identify fibril isomorphs by their distinct fragment patterns, but also investigate the micro-environment around plaques – thus complementing high-resolution cryo-ET structures of A-beta and tau fibrils recently obtained by project partner René Frank (Leeds). Native SIMS-imaging, based on structural interpretation of fragmentation patterns, will have wide applications in pharmaceutical research, medicine and biomaterials development. It promises to transform capabilities in medical imaging by investigating molecular structures directly in tissues and providing corresponding chemical context, to the benefit of human health.