Back

Computational redesign of TALE proteins for DNA-templated assembly of protein fibers

de Haas, R. J.; Langowski, M. D.; Borst, A. J.; Pillai, V. V. S.; Hoffmann, G. E.; Bongers, M.; Cheng, S.; Treichel, C.; Leaf, E. M.; Wu, M.; Lynch, E. M.; Kollman, J. M.; Ruggeri, F. S.; Walkey, C.; de Vries, R.; King, N. P.

2024-10-21 biochemistry
10.1101/2024.10.21.619424 bioRxiv
Show abstract

Many viral proteins self-assemble into capsid structures, often using their genetic material as a template for assembly. To date, de novo designed capsid-like proteins do not require genetic material as a template for assembly, which can be both an advantage and a disadvantage depending on the use case. Templates are indispensable, for example, in the assembly of linear structures with well-defined lengths. As a first step towards fully de novo designed templated assembly, here we redesign proteins from the Transcription activator-like effector (TALE) family of transcriptional regulators to polymerize on double-stranded DNA (dsDNA) templates. Starting from natural TALE protein sequences, we created idealized repeat proteins with sequence-independent DNA binding properties that cooperatively self-assemble to form linear protein-DNA complexes with template-controlled lengths. We used high-resolution atomic force microscopy (AFM) and cryo electron microscopy (cryo-EM) to characterize the three-dimensional structures of the DNA-protein hybrid complexes. In these structures, a protein filament helically wraps around the dsDNA using a binding mode similar to that of natural TALE proteins. As an example application of these materials, we show the system can be used for repetitive peptide antigen display at precisely controlled repeat distances, and that such immunogens elicit robust antigen-specific antibodies in mice.

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.