Neurodegenerative diseases such as Alzheimer's are caused by proteins in the brain misfolding and forming amyloid fibrils. A harmful property of some amyloids is their ability to secondary nucleate: the surface of a fibril acts as a template and triggers the formation of new, small and toxic aggregates. These soluble forms are thought to be crucial for nerve cell damage and memory loss. Among amyloids, Aβ42 is notorious for its strong proliferation activity and central role in Alzheimer's disease. In our research, we have made an unexpected discovery: spider silk proteins, which can also form amyloid-like fibrils and contain segments similar to Aβ42, completely lack the ability to drive proliferation. This contrast provides a unique model for understanding what makes an amyloid toxic. To investigate this, we combine biochemical analyses, cryo-electron microscopy (cryo-EM), targeted mutagenesis and electrophysiological measurements on mouse brain slices. We are also developing 'amyloid traps' and using Aβ42 variants to map the molecular properties that determine whether an amyloid is harmful or harmless. By identifying these differences, the project can increase understanding of brain disease mechanisms and pave the way for new strategies that slow or stop the spread of toxic protein aggregates. In the long run, this may contribute to the development of future diagnostics and treatments for Alzheimer's and other amyloid diseases, where current options are severely limited.