University of Minnesota
Primordial-like proteins generated from random polypeptide libraries of ancient alphabets
Abstract
dc:description.abstractThe genetic code is an essential part of protein synthesis. There have been numerous hypotheses on the evolution of the modern standard code of 20 amino acids, which likely started from a smaller set of early amino acids. However, little is known about the functionality of proteins before forming the standard genetic code. How early proteins composed of fewer amino acids achieve their functions is critical to understanding early steps in biological evolution. This dissertation focuses on exploring the functionality of the primordial proteins (i.e., ligand/cofactor binding) before the completion of the modern genetic code. We utilized combinatorial libraries with trillions of random protein sequences with specific amino acid alphabet sets, each representing different time points of genetic code evolution. A high-throughput selection method, mRNA display, was used to synthesize trillions of protein sequences and subjected them through in vitro selection campaigns. First, we synthesized and characterized four random protein libraries with reduced amino acid alphabets, from the earliest 5 amino acids to 20 extant amino acids. Characterizing randomly expressed sequences showed that reduced amino acid alphabets can produce soluble and expressible proteins comparable to random proteins with 20 extant amino acids. Furthermore, we demonstrated that some proteins exhibit secondary structure under specific conditions. We aimed to perform in vitro selection rounds targeting proteins that bind pyridoxal 5’ phosphate (PLP), a ubiquitous and essential cofactor for all amino acid metabolism enzymes. We devised a safer and improved method to immobilize PLP with its catalytically active moiety to achieve this goal. Preliminary mRNA display rounds with the resin provided a solid groundwork for the future PLP binder selection. Along with PLP, we challenged primordial alphabets of their binding activity toward the universal energy currency and cofactors, ATP and GTP. After the mRNA display rounds, all four libraries provided highly enriched protein sequences with ATP/GTP binding affinity. Further binding analysis revealed proteins made with 9 of the earliest amino acids can bind ATP comparable to modern ATP binders. In summary, we adopted a bottom-up approach to recreate and test random proteins using primordial amino acid alphabets. We successfully identified and characterized proteins with primordial alphabets that can bind ligands. Our empirical approach, employing an in vitro selection method, showcased the robustness of the exponential enrichment technique in origin-of-life research.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Lee, Kun-Hwa
Subjects
dc:subject × 6Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Citation
- Lee, Kun-Hwa. (2024). Primordial-like proteins generated from random polypeptide libraries of ancient alphabets. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/280281.