Large-library virtual screens for novel amyloid ligands

• Applicant: Dr. Philipp Seemann
• Subject Area: Pharmacy
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 567438697

Neurodegenerative diseases such as Alzheimer's are increasing worldwide and place a significant burden on healthcare systems. Alzheimer's alone causes severe memory loss and dementia, already affecting around 50 million people. Due to rising life expectancy, this number is expected to triple by 2050. However, treatment options remain limited, as many fundamental aspects of disease development and molecular mechanisms are still not fully understood. A common feature of Alzheimer's is the formation of amyloid fibrils—protein deposits in the brain and nervous system. A major challenge in developing new diagnostic tools and treatments is understanding how molecules bind to these amyloid proteins. Recent advances in cryo-electron microscopy (cryo-EM) have provided new insights: many known PET tracers (molecules used in Alzheimer's diagnosis) bind in a stacked arrangement along the protein fibrils. Based on these findings, the SymDOCK method was developed in the Shoichet Lab. This method utilizes the symmetry of protein structures to predict how molecules bind. Initial tests have been promising, successfully reproducing the binding of known PET tracers. Building on these successes, the project pursues three main objectives: 1. Discovery of new drug candidates: Through computer-based screening of ultra-large molecular libraries, highly effective molecules binding to tau fibrils will be identified. These could serve as new diagnostic markers or even potential treatments. 2. Advancing the SymDOCK method: Currently, SymDOCK is limited to symmetric protein structures like amyloid fibrils. The goal is to extend the method to non-symmetrical proteins that also bind stacked ligands. 3. Application to other disease-relevant proteins: the improved method will be tested on a SARS-CoV-2 protein (Mac1 domain). This will evaluate whether the technique can be applied to proteins with flat, hard-to-target binding sites. If successful, it could facilitate the development of new antiviral drugs. In the long run, this research could not only accelerate the development of Alzheimer’s treatments but also significantly improve drug discovery for proteins with challenging binding sites across various diseases.

• DFG Programme: WBP Fellowship
• International Connection: USA
• Host: Professor Brian Shoichet, Ph.D.