Bacterial enzyme reads eight-letter genetic code, structure shows

All known life writes its genetic information using four letters: the bases that pair up inside DNA. Researchers at UC San Diego have now shown that RNA polymerase, one of the cell's essential enzymes, can accurately read and copy a DNA alphabet expanded to eight letters, four natural and four synthetic.

RNA polymerase is the enzyme that reads a DNA sequence and produces a matching RNA copy, the first step a cell takes in turning genetic information into action. The team studied RNA polymerase from the bacterium E. coli and found that it recognizes the four added, synthetic letters using the same biochemical and structural cues it already uses for the natural four. That means the enzyme does not need to be redesigned to handle the expanded alphabet; it already treats the new letters as it treats the old ones.

To see how this recognition happens, the researchers combined biochemical experiments with high-resolution cryo-electron microscopy, an imaging method that let them observe structural detail smaller than the width of a single atom. Using this approach, they captured snapshots of RNA polymerase in the act of incorporating two synthetic base pairs, pairings of chemical letters that do not occur in nature, into a growing RNA strand. The images showed the enzyme engaging these synthetic pairs through the same molecular contacts it uses for natural base pairs, which helps explain why the copying process stays accurate.

The same research group reported a related result in a separate study, published in the journal PNAS, showing that RNA polymerase can also recognize a different synthetic base pair that has no hydrogen bonds holding it together at all. Hydrogen bonds are the weak chemical attractions that normally hold the two strands of natural DNA's base pairs to each other. That a hydrogen-bond-free synthetic pair can still trigger the enzyme's normal copying machinery adds further evidence that RNA polymerase's recognition process is flexible enough to work with chemistry unlike anything found in nature.

The eight-letter system used in this study is often called Hachimoji DNA, and it builds on years of separate work by a team led by Steven Benner at Firebird Biomolecular Sciences. That group designed synthetic base pairs by rearranging the chemical groups that let bases bond to each other, rather than inventing new chemistry from scratch, so the added letters could sit inside a DNA strand without disrupting its double-helix shape. Because the new pairs fit into the structure the same way natural pairs do, they can, in principle, be placed anywhere in a sequence and swapped around, a property needed for the DNA to be copied and evolved like natural genetic material.

Researchers describe this expanded alphabet as a way of testing whether earth's four-letter, protein-based version of life is the only workable one, or just one outcome among many. With eight letters instead of four, a short stretch of DNA can encode far more possible variations than the same length of natural DNA can, which researchers suggest could be useful for storing information in DNA form. Separately, past work has already used pieces of eight-letter DNA to bind to cancer cells, an early sign of possible use in diagnostics and treatments.

Getting an enzyme to transcribe expanded DNA is only part of the challenge; scientists also need reliable ways to sequence it, that is, to read back the order of letters in a strand of expanded DNA once it exists. A separate line of work has developed a method called ESEGA that converts some of these synthetic letters into standard DNA letters through chemical and enzymatic reactions, so that ordinary high-throughput sequencing machines, which are not built to read the synthetic letters directly, can still determine the sequence indirectly. That study's authors note that current sequencing approaches for expanded alphabets remain limited: some conversions are described as straightforward, while others are called difficult and produce mixed, hard-to-interpret results depending on the exact chemical conditions used.

Outside researchers see promise in the direction of this work but also identify unfinished steps. One genetic-systems researcher, not involved in the new UC San Diego study, has said that the next major task is to engineer enzymes that can not just read but also replicate and mutate the synthetic DNA, a capability that would let it be evolved the way natural DNA is. The UC San Diego findings show that a natural enzyme can already read the expanded code faithfully, but building full biological systems around expanded genetic alphabets, including reliable sequencing and self-directed replication, remains a work in progress.