Synthetic E. coli pushes the limits of gene synthesis
The synthetic organisms hall of fame has a new member: Syn61, an E. coli with all its genome replaced by a synthetic version.
Designing and building organisms from scratch could be the Holy Grail of synthetic biology. It therefore comes as no surprise that genome synthesis works get a lot of attention and are celebrated as high profile works form the community. In 2016 J. Craig Venter Institute scientists reported the synthesis of a reduced genome of Mycoplasma mycoides, in 2017 the synthetic yeast consortium reported the de novo synthesis of five yeast chromosomes (raising the count to six), and in April a team from ETH Zurich reported the rewriting and synthesis of a bacterial chromosome – though in the last instance the chromosome was not inserted in a bacterial cell, but stayed in a yeast host as a complete construct.
Last week, a team from the Medical Research Council Laboratory of Molecular Biology, a research institute in Cambridge UK, reported the complete synthesis and codon reassignment of the molecular biology workhorse genome, the one and only Escherichia coli.
The largest synthetic genome to date
The synthetic M. mycoides has a genome length of 531 kb. Caulobacter ethensis-2.0., the synthetic variety of Caulobacter crescentus, of around 700 kb. Syn61, the synthetic E. coli breaks the record, having a total of 4 mb (a million base pairs) of synthetic DNA sequence swapped in the native chromosome. This is an impressive feat of genome engineering, as scaling up poses significant challenges. And, as Benjamin Blout and Tom Elis note, the genome synthesis workflow is similar to the methodology of synthetic yeast, contributing to the standardization of the methods.
Codon reassignment leaves unused “plugs” for synthetic biologists
The genetic code has 64 codons, 61 encoding amino acids and three accounting for stop codons. As the (natural, encoded in DNA) amino acids are only 20, the code is redundant, meaning that several codons account for the same amino acid. It is therefore possible to reassign some codons to encode something else. This was previously shown when a research group from Tokyo, Japan, mutated E. coli to use only two stop codons, reassigning the remaining UAG codon to encode the unnatural amino acid iodotyrosine, or when researchers from University College London reassigned the UGA of the Chlamydomonas reinhardtii chloroplasts to encode for tyrosine.
In this work, Fredens and his collaborators decided to reassign three codons. When designing the synthetic genome, they replaced the triplets TCG and TCA (that encode serine) with AGC and AGT (that also encode serine), replaced TAG with TAA (both stop codons). The result was a genome with three “free plugs”, triplets that can be used by synthetic biologists to encode unnatural amino acids and insert new chemistry into biological systems.
What is the phenotype of the cells?
When doing such a scale of modifications, one would expect severe side-effects. After all, the organisms have evolved using all 64 codons, and modifications may affect structural characteristics of the chromosome. Reassigning codons may affect the balance of tRNA usage. The synthetic M. mycoides and the yeast incorporating the synthetic chromosomes display severe growth defect and odd phenotypes.
That is also true for Syn61, but nos quite as much. It grows 1.6 times slower than the control strain at 37 °C and the bacteria are slightly longer in shape. However, the differences in their proteome are not significant and the bacteria grow happily. It seems that E.coli has a degree of resilience and the genome designers pinpointed the design flaws correctly.
We are still a long way from building synthetic cells or writing synthetic eukaryotic genomes, but this study pushed us a bit closer. I am curious to see how the community will adopt Syn61 and use the opportunity to creatively use the unassigned codons. And congratulations to the Cambridge researchers for thier significant achievement!