Welcome to the Bellstedt Lab
Imagine a world where incurable diseases are a thing of the past, where a simple pill can halt the progression of cancer, or a new vaccine can prevent a global pandemic. Drug discovery is the gateway to such a future. It's the pioneering endeavor that delves deep into the complexities of biology to find solutions to humanity's most pressing health challenges. My research group at the University of Zurich (UZH) and the University Hospital Zurich (USZ) focuses on structure-based drug discovery and aims to contribute to medical challenges by applying state-of-the-art methods ranging from biophysics and molecular biology to chemoinformatics. We analyze the 3D structures of biological targets, typically proteins, to create compounds that bind effectively and modulate their activity. To identify initial compound candidates, we screen huge libraries of drug-like molecules in silico by applying state-of-the-art docking and molecular dynamics simulation procedures. After estimation of relative binding energies, selected compounds are tested and optimized for biochemical activity and metabolic stability using dedicated cell culture models as well as purified target proteins.
-
Click here to openOur Webpage at the University Hospital ZurichFind more details about our research and current team members.
-
Click here to openPublication recordHave a look at our latest publications at GoogleScholar.
Biophysical Characterisation
Our protein targets are typically overexpressed in E. coli cells, and purified with affninity/size-exclusion chromatography. Afterwards binding affinities or inhibitory concentrations of selected compound series are determined using a variety of different (and often orthogonal) methods including colorimetric bioassays, microscale thermophoresis (MST), localized surface plasmon resonance (OpenSPR) and NMR spectroscopy. State-the-art pipetting robots available in the lab ensure high reproducability of the results obtained.
Metabolic Stability and Biotransformation
Using S9 liver extracts isolated from different species and chromatography coupled mass spectrometry (LC-MS) we can analyze the time-dependend stability of compounds. Possible Phase I or Phase II metabolites are identified using collision induced fragmentation (MS/MS).
Cell culture models
We utilize different cell lines ranging from typical (HEK293, HeLa) to more specific cancer cell lines (Caco-2, DLD-1, HCT116, HT29) to test the activity of our compounds and to derive dose-response curves as well as IC50/EC50 values. Assay output depends on the protein target and ranges from cytotoxicity/apotosis assemssent to more specific assays like the detection of enzymatic reaction products. Automatic media dispensers and aspirators allow for high-throuput and minmize the influence of the experimentator on the results. Together with the Robert Lab, we are currently also developing protocols to mimic the blood-brain-barrier to estimate if the compound under investigation is able to pass, which is often not-intended. Furthermore, compounds can be tested for their ability to interact with major renal drug tansporters to estimate the risk for drug-drug-interactions (collaboration with the Visentin Lab).
Bio-/Chemoinformatics
We use and develop Python- and KNIME-based scripts to establish efficient workflows for our daily research. Parts of that work is extended to independent web services, that allow researchers across the world to easily analyze their own data without any need for coding skills. Please find the links below. Platforms, such as GitHub, provide an ecosystem for researchers to share, review, and improve code. Most importantly, it also allows us to work together from different places across the world. Indeed, my lab offers online internships as well as remote master projects, as in contrast to our work in the wet lab and cell culture, the actual working place does not matter for coding. DockerHub complements GitHub by offering a platform for distributing containerized applications, ensuring that software runs consistently across different computing environments, and allowing a high degree of reproducibility. As part of our strong belief in the power of open science and FAIR data, we publish portable container on DockerHub code on GitHub and provide structural data files that supplement our peer-reviewed publications on Zenodo.
-
Click here to openOur HSQC-2-Struct(ure) Web Service (manuscript submitted, PrePrint available)Our machine learning model predicts the secondary structure content based on unassigned 1H,15N-HSQC protein spectra.
-
Comming soonOur PDB-2-Struc(ure) Web Service (open source)Calculate the Q3 and Q8 secondary structure content of any protein structure deposited in the PDB.
-
Click here to openOur CAPITO Web Service (peer-reviewed)Analyze your Circular Dichroism (CD) spectrum to extract the secondary structure content of your protein sample.