Synthetic Genome Construction Techniques to Edit the Genetic Code of Escherichia coli

The expansion of the genetic code to allow for site-specific incorporation of non-standard amino acids (nsAAs) — proteogenic amino acids other than the 22 found across life — is an important step in the development of new experimental and industrial proteins. Genome recoding is a process by which all instances of a given codon are replaced with synonymous alternatives. This frees the codon for reassignment and to code uniquely for the incorporation of an nsAA while preserving the preexisting proteome. The Church lab has embarked on a project to construct a strain of E. coli with 7 of its codons simultaneously recoded: rE.coli-57. This is a recoding effort not yet achieved at this scale, requiring 62,214 codon changes compared to the wild-type genome. To achieve this, the rE.coli-57 genome was first purchased in thousands of short, synthetic DNA fragments which were then assembled into 87 different ~50 kb episomal segments. Since joining this project, I have developed a pipeline for the precise, iterative, in vivo replacement of the wild-type E. coli genome with all of these recoded segments, and I led the construction for the first third of the 4 Mb genome. To enable rE.coli-57’s completion, I developed a method to combine recoded portions of multiple strains edited in parallel into a single, final strain. I also optimized the in-lab production of a useful bacterial selective agent which enables complex genetic engineering tasks without the need for specialized equipment. While rE.coli-57 itself aims to be a platform to test virus resistance and site-specific translation of multiple nsAAs, the publication of an effective technology for synthetic, prokaryotic genome construction could allow for the invention of other biocontained and bioisolated bacterial species.