Life With Longer Genetic Codes Seems Possible — but Less Likely

As wildly diverse as life on Earth is — whether it’s a jaguar hunting down a deer in the Amazon, an orchid vine spiraling around a tree in Congo, primitive cells growing in boiling hot springs in Canada, or a stockbroker sipping coffee on Wall Street — at the genetic level, it all plays by the same rules. Four chemical letters, or nucleotide bases, spell out 64 three-letter “words” called codons, each of which stands for one of 20 amino acids. When amino acids are strung together in keeping with these encoded instructions, they form the proteins characteristic of each species. With only a few obscure exceptions, all genomes encode information identically.

Yet, in a new study published last month in eLife, a group of researchers at the Massachusetts Institute of Technology and Yale University showed that it’s possible to tweak one of these time-honored rules and create a more expansive, entirely new genetic code built around longer codon words. In principle, their discovery points to one of several ways of expanding the genetic code into a more versatile system that synthetic biologists could use to create cells with novel biochemistries that make proteins found nowhere in nature. But the work also showed that an extended genetic code is hampered by its own complexity, becoming less efficient and even surprisingly less capable in some ways — limitations that hint at why life may not have favored longer codons in the first place.

It’s uncertain what these findings mean for how life elsewhere in the universe could be encoded, but it does imply that our own genetic code evolved to be neither too complicated nor too restrictive, but just right — and then ruled life for billions of years thereafter as what Francis Crick called a “frozen accident.” Nature opted for this Goldilocks code, the authors say, because it was simple and sufficient for its purposes, not because other codes were unachievable.

For example, with four-letter (quadruplet) codons, there are 256 unique possibilities, not just 64, which might seem advantageous for life because it would open opportunities to encode vastly more than 20 amino acids and an astronomically more diverse array of proteins. Previous synthetic biology studies, and even some of those rare exceptions in nature, showed that it’s sometimes possible to augment the genetic code with a few quadruplet codons, but until now, no one has ever tackled creating an entirely quadruplet genetic system to see how it compares with the normal triplet-codon one.

“This was a study that asked that question quite genuinely,” said Erika Alden DeBenedictis, the lead author of the new paper, who was a doctoral student at MIT during the project and is currently a postdoc at the University of Washington.

Expanding on Nature

To test a quadruplet-codon genetic code, DeBenedictis and her colleagues had to modify some of life’s most fundamental biochemistry. When a cell makes proteins, snippets of its genetic information first get transcribed into molecules of messenger RNA (mRNA). The organelles called ribosomes then read the codons in these mRNAs and match them up with the complementary “anti-codons” in transfer RNA (tRNA) molecules, each of which carries a uniquely specified amino acid in its tail. The ribosomes link the amino acids into a growing chain that eventually folds into a functional protein. Once their job is complete and the protein is translated, the mRNAs get degraded for recycling and the spent tRNAs get reloaded with amino acids by synthetase enzymes.

The researchers tweaked the tRNAs in Escherichia coli bacteria to have quadruplet anti-codons. After subjecting the genes of the E. coli to various mutations, they tested whether the cells could successfully translate a quadruplet code, and if such a translation would cause toxic effects or fitness defects. They found that all of the modified tRNAs could bind to quadruplet codons, which showed that “there’s nothing biophysically wrong with doing translation with this larger codon size,” DeBenedictis said.

But they also found that the synthetases only recognized nine out of 20 of the quadruplet anticodons, so they couldn’t recharge the rest with new amino acids. Having nine amino acids that can be translated with a quadruplet codon to some degree is “both a lot and a little,” DeBenedictis said. “It’s a lot of amino acids for something that nature doesn’t ever need to work.” But it’s a little because the inability to translate 11 essential amino acids strictly limits the chemical vocabulary that life has to play with.

Moreover, many of the quadruplet code translations were highly inefficient, and some were even detrimental to the cell’s growth. Without a major fitness advantage, it’s very unlikely nature would have selected a more complex code, especially once it had settled on a working code, DeBenedictis said. The authors concluded that the reason why nature didn’t select for a quadruplet code wasn’t because it was unachievable, but rather because the triplet code was simple and sufficient. After all, even if life needed to expand its repertoire of 20 amino acids, there’s still lots of room within the existing 64 codons to do so.

Triplet codons work well on Earth, but it’s not clear if that would be true elsewhere — life in the cosmos might differ significantly in its chemistry or in its coding. The genetic code is “presumably derivative and subservient to the biochemistry of peptides” that are required for life to work, said Drew Endy, an associate professor of bioengineering at Stanford University and president of the BioBricks Foundation, who was not involved in the study. In environments more complex than Earth, life might need to be encoded by quadruplet codons, but in much simpler settings, life might get by with mere doublet codons — that is, of course, if it uses codons at all.

The Entrenched Competition

No matter how life is encoded on our planet or on others, the real impact of the paper is that now we know it’s “totally possible to make a quad-code organism,” and the findings suggest it will be straightforward, Endy said. With one study, they’re almost halfway to getting it to work, he added, which is “an infinitely amazing accomplishment.”

Not everyone agrees that creating a full quad-coded life form will be simple. “I don’t think anything they show suggests that it’s going to be easy — but they do show it’s not impossible and that’s interesting,” said Floyd Romesberg, a synthetic biologist who co-founded the biotech company Synthorx. Getting something that works poorly to work better is a “very, very different game” than trying to do the impossible.

How much effort it will take to make a true quadruplet code work well is an open question, DeBenedictis said. She thinks you would also likely need to reengineer much of the translation machinery to work well with a larger code. She and her team are hoping to bring their work to the next level by adding an extra “tail” to the engineered tRNAs so that they will interact with a set of ribosomes designed to work with them alone. That might improve the efficiency of translation by reducing competition with any triplet-coding aspects of the system.

Overcoming the competition from the triplet code will always be a major challenge, she added, because it already works so well.