Research Areas

Structural studies of virus proteins using mass spectrometry

To elucidate processing, assembly and interaction networks of viral proteins with native MS, we collaborate with internal and external research groups. Complex assembly of non-structural proteins is of high interest for drug development but so far poorly understood. Therefore, potential binding partners of a viral protein can be identified or verified using tandem affinity purification followed by MS-based proteomics. In native MS, complex formation can be characterised in detail as a function of the physical environment, posttranslational modifications and ligands such as nucleic acids. This can be compared to complexes formed in vivo. MS-based topology models can furthermore assist interpretation of EM data. Amongst others, we investigate proteins and protein complexes of adeno-, corona- and hepatitis C viruses.

In addition, hydrogen/deuterium exchange MS will be employed to localise conformational changes upon ligand binding and change of the physical environment; and cross-linking MS to map binding sites in protein complexes. These methods will be established in collaboration with the University Medical Centre Hamburg-Eppendorf (UKE).

MS-based methods at European XFEL

The development of an MS-based sample delivery system is performed in close collaboration with the European XFEL, namely the groups “Sample Environment” and “Single Particle and Biomolecules (SPB) Instrument”, into which the mass spectrometer will be integrated. The intense, ultra-short X-ray pulses of an XFEL enable imaging of individual protein complexes without any signal enhancement from a crystal. In the process, the sample is destroyed. To produce high resolution structures, many images of structural state have to be recorded. Classification and orientation of frames takes place under huge computational efforts. The mass spectrometer can enable introduction of protein complexes into the X-ray laser, which are size and conformation separated. On one hand, this reduces the required computational resources, on the other hand low populated transition states can be analysed selectively, which are invaluable for drug development but would normally be lost in the vast amount of data.