A new first: rewriting the genetic code

Imagine that you wanted to remove every instance of the letter Q from the English language without losing meaningful words spelled with Q, and without adding any new letters to the alphabet. You’d have to choose an alternate letter to take Q’s place—C or K, perhaps—then rewrite books with the new letter and re-teach people to spell and read using the new alphabet. Such an undertaking is what a team of Yale, Harvard, and MIT researchers have recently completed. Rather than altering the English language, however, they removed a letter from the genetic alphabet of a bacteria.

The re-engineered bacteria didn’t just contradict classic rules of biology: it also was able to better fight off invading viruses that normally relied on the host’s language to function.

In all living organisms, genes are made up of long strings of four nucleotides. Every set of three nucleotides in a row—64 different possible combinations—codes for one amino acid, the building block of a protein. Each of these so-called codons is translated into its corresponding amino acid through a unique tRNA molecule. At one end, the tRNA binds the three nucleotides; at the other it carries the associated amino acid. The team of scientists, which includes the School of Medicine’s Jesse Rinehart, Ph.D., assistant professor of cellular and molecular physiology, and Farren J. Isaacs, Ph.D., assistant professor of molecular, cellular, and developmental biology, wanted to remove one of these codons from a strain of bacteria.

“No one had ever entirely removed a codon from a genetic code,” Rinehart says. “But if we could remove a codon and the organism was fine, biologists could start utilizing that codon for their own engineering purposes.”

If Qs, for example, were no longer used in words like “quick” and “mosque,” then Q could be assigned a different meaning—like a new punctuation mark. For synthetic biologists, having an unassigned codon is key to adding new amino acids to a protein to give it new properties.

To show that removing a codon was possible, Rinehart, Isaacs, and George Church, Ph.D., professor of genetics at Harvard Medical School, set their sights on the least common codon: a string of the three nucleotides U-A-G. Rather than code for an amino acid in a protein’s structure, UAG is a stop codon: it tells the translation machinery that the end of a gene has been reached, like a period. But two other codons serve the same function: UAA and UGA. So, using precise gene editing techniques that they had previously developed, the scientists changed every occurrence of UAG in a strain of Escherichia coli, 321 in all, to UAA.

“What’s really powerful about these techniques is that we can take these oligonucleotides and insert them with high efficiency, and it allows us to simultaneously target many sites across the genome,” Isaacs says.

The codon replacement worked, but that wasn’t the end of the project. The team then introduced a genetic mutation into the protein that normally interprets the UAG as a stop codon—called release factor 1 (RF1). In a normal cell, deleting RF1 would lead to a jumble of misread genes: one protein would run into the next with no break, since the stop codon wouldn’t be read between genes. But in the newly engineered E. coli, there were no UAG sequences to be read. Unlike other strains of bacteria, removing RF1 from this strain had no effects. Or, at least, no negative effects: when the altered E. coli was infected with a bacteriophage, a type of virus that infects bacteria, the invading phage could no longer function.

“When the phage infects the cell, its genes contain stop codons, including UAG stop codons,” Rinehart says. “And it relies on the bacteria’s release factor to read those codons.” By removing a codon from the bacteria’s entire language, the scientists had given the bacteria a new defense against the virus. The research was published October 18 in the journal Science.

“This is an important advance in understanding the genetic code,” Rinehart says. “But it also shows that we are in an exciting new reality where we can take the lessons we’ve learned from biology, from understanding the genome and the proteome, and we can go forward into a more exciting time where we can engineer new properties into cells.”

The advance opens up the door to a new way of adding amino acids to proteins—by assigning UAG to a new tRNA, with a completely novel amino acid on the protein side.

The team plans to continue optimizing the techniques and pushing the boundaries of what’s possible in protein engineering. “We could now introduce entirely new properties into these organisms by assigning this codon to a new amino acid,” Isaacs says. “That we were able to change the code, as well as introduce new biological functions, is exciting and satisfying.”