Deconstructing Biological Function at the Host-Pathogen Interface through Constructive de novo Protein Design

• Applicant: Dr. Lukas Milles
• Subject Area: Biophysics ; Biochemistry ; Structural Biology
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 529882161

De novo protein design using highly accurate structure prediction neural networks presents a fundamentally novel approach to create diverse, complex, and nuanced protein structures beyond the capabilities of previous methods. Here, I propose to leverage, extend, and generalize this approach for multi state design to reconstruct complex biological functions at the host-pathogen interface, in order to deconstruct their underlying molecular mechanisms. We will focus on two specific systems: thioester domains (TEDs) and catch bonding interactions. Thioester domains are autocatalytic chemical "harpoons" that covalently bind to host targets. We will create synthetic, de novo designed TEDs by reconstituting the buried thioester chemistry and investigate key variables that govern autocatalytic intermolecular covalent targeting of both native and designed proteins. Ultimately, this will permit us to steer thioester reactivity towards arbitrary targets. Catch bonding interactions are atypical receptor ligand interactions that increase their bond lifetime under mechanical force allowing pathogens to persistently attach to their hosts even under high mechanical stress. Reconstituting these bonds using de novo design will reveal the minimal components required for their function. Consequently, this will allow us to decipher the relationship between thermodynamics, kinetics, and most importantly mechanics, as studied by single-molecule force spectroscopy, that shape catch bond energy landscapes. By aggregating the data generated through the proposed work, we will compare the biophysical properties of de novo designed proteins and their native templates to understand trade-offs in de novo designs and the shaping of protein biophysics by design methods. The reconstituted mechanisms could serve as versatile tools for applications such as: de novo catch-bond based novel biomaterials with strain-stiffening properties, or de novo thioester domains for covalent opsonization. A deeper understanding of these pathogen mechanisms will also open new routes to potential therapeutic approaches that inhibit these systems. In sum, I propose to deconstruct biological function through constructive de novo design to enable the precise isolation, reconstitution and thus study of complex protein functions beyond the space sampled by evolution.

• DFG Programme: Independent Junior Research Groups
• Major Instrumentation: HPLC Agilent 1260 Infinity II bioinert