Assembly of a 57-codon Escherichia coli genome and extensions towards mammalian synthetic genomics

The genetic code is one of the most fundamental hallmarks of life and is shared by almost all organisms. This set of instructions, defining how 64 triplet codons encode 20 canonical amino acids and three stop signals, links nucleotide sequence to protein function and thereby the genome to biological phenotype. Fundamentally reprogramming this code requires practical approaches for genome-wide rewriting and engineering. This thesis describes the design, debugging, and assembly of Syn57, an Escherichia coli operating on a 57-codon genetic code. By introducing 101,605 synonymous codon substitutions, we removed all instances of seven target codons from all annotated protein-coding sequences. Because codon choice is functionally consequential, rewriting the genetic code constitutes a substantial intervention into cellular physiology. Where direct synonymous replacement was incompatible with viability, we developed multiple distinct fixing strategies that exposed context-dependent constraints on recoding, including library-based approaches and repair of regulatory perturbations. During progressive consolidation of the recoded genome into a single strain, cumulative fitness defects emerged, which we localised and alleviated using targeted correction and evolution-based approaches. Full assembly of a deeply recoded organism allowed initial observations of physiological change, including global shifts in gene expression that are not readily explained by simple local recoding metrics, indicating that it remains difficult to predict how recoding propagates into system-level defects and how these defects can be effectively repaired. The resulting seven free codons provide a starting point for future genetic code expansion using non- canonical amino acids or other novel monomers. Finally, we bridge the gap from prokaryotic to eukaryotic life and extend genome synthesis approaches to mammalian systems. We assembled a bacterial artificial chromosome containing more than 500 kb of human sequence, and engineered mammalian cell lines to accommodate future integration of this payload. Together, these components establish a route towards the assembly and targeted integration of large DNA payloads in mammalian systems.