UC Berkeley

There is profound interest in proteomimetic and peptidomimetic polymers containing monomers other than non-α-amino acids. Such hetero-oligomers have potential as improved biological therapeutics, renewable biomaterials that supplant petroleum-based commercial products, and for the engineering of next-generation biocatalysts with enhanced properties relative to current state-of-the art industrial enzymes. A rich body of work has detailed the advantages of such non-α-amino acid hetero-oligomers, however the traditional methods used to synthesize such materials come with non-trivial challenges including low-throughput production and upper limits on the terminal length of the hetero-oligomer. Leveraging genetic code expansion (GCE) to site-specifically incorporate new monomers into proteins and peptides using cellular translation machinery circumvents these synthetic challenges and represents a marked step towards the promised potential of these new biomaterials. Here, I present research detailing the programmed incorporation of multiple β2-hydroxy acid monomers into a model protein to produce the first mixed polyester/polyamide heteropolymers generated in cells. I also report ongoing efforts to enhance the yield and fidelity of heteropolymer biosynthesis via directed evolution of two essential translation factors: tRNA and EF-Tu. In chapter one I briefly review the field of genetic code expansion with special attention given to the strengths and limitations of using a cellular chassis to enable ribosomal heteropolymer synthesis (rather than in vitro or synthetic cell-based translation). In chapter two I report on my successful efforts to identify an aminoacyl-tRNA synthetase (aaRS)/tRNA pair (PylRS/tRNAPyl) that recognizes an expanded backbone monomer and to employ this aaRS/tRNA pair for genetic code expansion in cells. This work, adapted from an article published in ACS Central Science, also details the degree of misincorporation leading to a loss in polymer sequence fidelity and leverages metadynamics to propose that β2-hydroxy acids may be surprisingly good substrates for the ribosome. In chapter three I describe a directed evolution effort targeting tRNAPyl to enhance the yield and fidelity of non-canonical monomer incorporation. Using structure- and literature-guided design I generate two libraries of tRNAPyl variants and subjected the libraries to fluorescence-activated cell sorting (FACS) to identify library members that improved the incorporation of non-canonical monomers. Selections based on non-canonical α-amino acid monomers revealed random mutations in the plasmid sequence that unexpectedly enhance amber suppression for monomers containing α-NH2, α-OH, and (R)-β2-OH backbones. I provide a working hypothesis for this phenomenon and detail clear next steps as well as experimental design considerations for moving this project forward. In chapter four, I describe parallel efforts to evolve EF-Tu for enhanced recognition of tRNAs acylated with β2-OH acids. I demonstrate that overexpression of EF-Tu from a plasmid does not generally afford increased production of sfGFP with an in-frame amber codon – evaluated by bulk fluorescence experiments and validated by isolated protein yields and LC-MS characterization. In analogous experiments with a plasmid encoding EF1α and EF1β from M. alvi (the organism from which the PylRS/tRNAPyl derive) I also observe no difference in sfGFP production when compared to plasmids lacking EF1α and EF1β as measured by bulk fluorescence. I also describe and generate a large (3.2x106)-membered site-saturation mutagenesis library targeting 5 residues within the tRNA binding pocket of EF-Tu. This includes the development of a novel two-plasmid system that supports both high transformation efficiency and reporter signal. Finally, I describe the initial characterization of the library when it is split into two smaller libraries. Collectively the work described in this thesis details both the discovery and characterization of new platforms for the cellular synthesis of proteins containing β2-hydroxy acids and details ongoing efforts and challenges related to efficient evolution of translation factors downstream of the synthetase.