Abstract:
The thesis presented here deals with four independent projects, which are briefly summarized in the following paragraphs.
Flagella enable bacteria to freely swim in solution granting them the motility to reach resourceful areas or to avoid an unfavorable environment. Therefore, the ability to produce a flagellum, is a major fitness factor for bacteria. Pathogens might rely on flagella to reach host surfaces and penetrating them. From a host point of view, flagella or their major protein FliC render an opportunity to detect bacteria and regulate a response appropriate to the bacterial threat. The regulation of the response can be dysfunctional, which can lead to inflammatory bowel disease. E. coli Nissle, a commensal showed positive effects in inflammatory bowel disease mouse models which partially depend on the major flagella protein FliC. We crystallized and solved the structure of the E. coli Nissle FliC. By comparison to known FliCs, we were able to identify two new structural features. First, the hypervariable region consists of three distinct domains and second, the hypervariable region is connected via a long linker to the constant region. Modeling the flagellum filament showed that the hypervariable regions might interact with each other forming an outer structure, which is connected via the linkers to the conserved flagellum core structure. Modeling the interaction with the immune receptor TLR5 that recognizes the conserved domain 1 of FliC showed that the domain should be well accessible and not sterically hindered by the three hypervariable domains. The dimerization of TLR5 with bound FliC to the active complex should also not be hindered by the hypervariable region.
Glycoprotein C (gC) is an attachment factor of Herpes simplex virus 1 that binds to heparan sulfate (HS) which is exposed on the cell surface of the host. This non-essential viral attachment protein allows lateral virus diffusion along cell surface and supports attachment of other viral proteins to their receptors on the cell surface, which ultimately results in membrane fusion or endocytotic uptake of the virus. The structure of gC was solved by experimental phasing and shows that gC consists of three distinct domains and presumably forms a dimer. We identify the putative heparan sulfate binding area ranging from the middle to the N-terminal domain. Another function of gC, the protection from the complement system, needs further investigation. Here, the structure might help to design experiments to investigate how gC binds C3b and prevents further interaction of C3b with the complement system. The gC structure from HSV-1, solved in this work, can serve as model for the closely related HSV-2 gC (Identity 73%) and migth help to explain differences in HS binding and C3b affinity.
We invented a nanobody (Nb)-based peptide tag system, which can be used for protein purification or to label proteins for super resolution microscopy. The nanobody binds a twelve amino acid long peptide with a low nanomolar affinity. We solved the structure Nb-peptide complex to explain the observed specificity for different amino acid sequences. We identified an essential tryptophan residue in the peptide sequence located in a deep hydrophobic pocket of the Nb, at another position basic residues are preferred. Some sidechains point towards the Nb and have to be rather small while other pointing away from the Nb with no preference for certain residues. The peptide is bound between the framework region and complementarity-determining region 3 (CDR3) of the Nb, mainly by backbone-backbone interactions.
In a side project, we contributed to the development of an adenovirus based vector system. The vector is based on the rare adenovirus serotype 43 with a low antibody prevalence in humans. The adenovirus was modified with affibodies changing the primary receptor to the ocotarget “epidermal growth factor receptor type 2” (Her2).
Print-on-Demand