Advancing Toward Construction of a Bacterial Recoded Genome

Initial testing of a synthetic bacterial genome that uses 57 of the 64 natural codons showed minimal fitness impairment.

The Science

Taking advantage of the genetic code’s redundancy, a collaborative project led by researchers at Harvard University synthesized a bacterial genome that uses 57 of the 64 natural codons; the remaining seven codons were reassigned to nonstandard amino acids that can be used to develop novel protein functions. So far, DNA segments that span over 60% of the synthetic genome and contain over half the essential genes have been introduced into living cells to test for deleterious effects, and only very few cases showed significant growth defects.

The Impact

Reassigning several of the 64 natural codons in a bacterial genome enables the development of microbial strains with multiple combinations of proteins that can perform novel functions, while preventing the engineered strain from surviving if it escapes laboratory conditions. This large-scale, genome-wide recoding required developing design tools that must be fine-tuned after testing and gaining knowledge of rules that must be observed to synthesize functional genetic elements and networks. This research shows that recoding essential genes is possible and has uncovered fundamental design principles.


To construct a completely recoded Escherichia coli genome, the researchers first used computational tools to design a genomic sequence lacking all instances of seven redundant codons and synthesized the genome in 87 fragments spanning 50 kb each. Testing of 55 of these fragments, which contain 63% of the genome and 52% of essential genes, showed that most of them caused limited or no change in growth and transcription levels. The recoded version of one gene resulted in severe fitness impairment, but the researchers were able to redesign the gene, allowing the strain to survive. At the same time, the researchers were able to optimize the design tools to further reduce potential growth defects in recoded microbes. This research demonstrates the feasibility of high-level recoding of microbial organisms to confer new functionality such as the development of new bioproducts. It also shows that genome-wide engineering approaches provide new knowledge on the fundamental principles that drive biological systems.

Principal Investigator

George M. Church
Harvard University
[email protected]

BER Program Manager

Pablo Rabinowicz

U.S. Department of Energy, Biological and Environmental Research (SC-33)
Biological Systems Science Division
[email protected]


This work was supported by the Office of Biological and Environmental Research within the U.S. Department of Energy’s Office of Science (award DEFG02-02ER63445). Authors also acknowledge support from the U.S. Department of Defense and National Science Foundation.


Ostrov, N., M. Landon, M. Guell, G. Kuznetsov, J. Teramoto, N. Cervantes, M. Zhou, K. Singh, M. Napolitano, M. Moosburner, E. Shrock, B. Pruitt, N. Conway, D. Goodman, C. Gardner, G. Tyree, A. Gonzales, B. Wanner, J. Norville, M. Lajoie, and G. Church. 2016. “Design, Synthesis, and Testing Toward a 57-Codon Genome,” Science 353(6301), 819–22. DOI: 10.1126/science.aaf3639.