Computer illustration of bacteriophages attacking E. coli cell

Bacteriophage viruses — seen attacking an E. coli cell in this computer illustration — can have exotic chemistries in their DNA.Credit: Maurizio De Angelis/Science Photo Library

‘Alien’ genomes can be found on Earth. Some viruses that infect bacteria use an alternative genetic alphabet that’s distinct from the code used by nearly all other organisms — and, now, two teams have spelt out how the system works.

More than four decades in the making, the studies show how dozens of these bacteriophages (or just ‘phages’), as they are known, write their genomes using a chemical base called 2-aminoadenine, Z for short, instead of adenine — the A in the As, Ts, Cs and Gs of genetics textbooks.

“Scientists have long dreamed of increasing the diversity of bases. Our work shows that nature has already come up with a way to do that,” write Suwen Zhao, a computational biologist at ShanghaiTech University in China, and her team in a 29 April Science paper, showing how ‘Z-DNA’ is made1. Researchers in France described similar insights in a pair of papers in the same journal2,3.

The work is seminal, says Steven Benner, a synthetic biologist and founder of the Foundation for Applied Molecular Evolution in Alachua, Florida, who compares it to US microbiologist Carl Woese’s discovery of a new branch of single-celled life. “It represents the first discovery of a ‘shadow biosphere’ since Woese identified the Archaea a half century ago.”

Bond booster

Scientists in the Soviet Union were the first to discover Z-DNA, in the late 1970s, in a phage called S-2L, which infects photosynthetic bacteria4. They found that the phage DNA behaved oddly when its two helical strands were melted apart. The bond that forms between G and C bases breaks at a higher temperature, compared with that joining A and T, and the phage’s DNA behaved as if it was made primarily from G and C. But further analysis by the Soviet team showed that the phage had replaced A with Z, which formed a stronger bond with T.

Hardy genome: graphic that shows how the Z:T base pair forms three hydrogen bonds, compared to two in the A:T base pair.

“It looked like something transgressive,” says Philippe Marlière, an inventor and geneticist at the University of Evry, France, who led one of the Science studies. “Why did this phage have a special base like this?”

Follow-up studies showed that S-2L’s heartier genome was resistant to DNA-chomping enzymes and other anti-phage defences that bacteria wield. But researchers didn’t know how the Z-DNA system worked or whether it was common. Z-DNA is only one of a host of modifications known to exist in phage DNA.

To answer those questions, a team led by Marlière and Pierre-Alexandre Kaminski, a biochemist at the Pasteur Institute in Paris, sequenced the phage’s genome in the early 2000s. They found a gene that’s potentially involved in one step of making Z-DNA, but not in others. But the sequence had no matches in genomic databases at the time, and the team’s quest to understand the basis for Z-DNA hit a dead end.

Marlière and his colleagues patented the S-2L genome, but also made it public, and he continued to scour genomic databases. Finally, in 2015, the team got a hit: a phage that infects aquatic bacteria of the genus Vibrio harboured a gene that matched a stretch of S-2L’s genome. The gene encoded an enzyme that resembled one that bacteria use to make adenine. “It was an exhilarating moment,” says Marlière.

In 2019, Zhao’s team found similar database matches. Both teams showed that the phages all had a gene named PurZ. This codes for an enzyme that plays an early but crucial part in making the Z nucleotide from a precursor molecule that is present in bacterial cells. They then identified additional enzymes — encoded in the genomes of bacteria that the phages infect — that complete the pathway.

But a key question lingered. The enzymes that the teams identified produced the raw ingredient for Z-DNA — a molecule called dZTP — but that didn’t explain how phages insert the molecule into DNA strands, while excluding A bases (in the form of a chemical called dATP).

Here, the teams’ conclusions differed slightly. Alongside PurZ in the Vibrio phage’s genome sits a gene that makes an enzyme called a polymerase, which copies DNA strands. Marlière and Kaminski found that the phage polymerase incorporates dZTP into DNA, while cutting out any A bases that were introduced. “This explained to us why A was excluded,” says Kaminski. “This was really spectacular.”

Zhao thinks this isn’t the whole story. Her work suggests that another phage enzyme is needed, one that breaks up dATP but preserves dZTP inside cells. Her team found that increasing dZTP levels relative to those of dATP was enough to trick a cell’s own polymerase into making Z-DNA.

Missing links

“There’s a lot we don’t know,” Zhao says. It’s not clear how hosts keep Z out of their DNA. Nor it is apparent how cellular machinery that reads DNA to make proteins copes with Z-DNA, which forms a double helix that’s shaped slightly differently from ordinary DNA molecules. It’s also not fully understood how Z-DNA is copied (a process that might require specialized enzymes in addition to polymerase), Kaminski adds. “We still don’t know how the whole system works.”

The functionality of host enzymes could be improved or impaired when working on Z-DNA, says David Dunlap, a biophysicist at Emory University in Atlanta, Georgia, who has found that an E. coli enzyme struggles to coil and bend the exotic double helix5. The discovery of more phages with Z-DNA, and of genes involved in making the molecule, should help researchers to understand how phages benefit from using it.

Having these genes in hand could speed potential applications of Z-DNA, by making it easier and cheaper to make, says Zhao. Z-DNA’s hardiness could make the nascent technique of DNA data storage more stable and long-lasting. Nanomachines made of precisely arranged Z-DNA — known as DNA origami — might fold into shape faster. The French team is working on incorporating the molecule into bacterial genomes. “We have E. coli cells that are getting invaded with ‘Zed’. It’s not as toxic as I feared,” says Marlière.

Benner — whose research has expanded the genetic alphabet to include several artificial DNA bases6 — hopes that the new studies will rattle researchers into realizing the power of altering the genetic alphabet. “The fact that nature has taken a small step in the same direction may be the intellectual caffeine needed to get the molecular-biology community to understand that DNA can be improved, and beneficially so,” he says.