A newly discovered bacterial defense system works backward from every other known version

For three decades, microbiologists knew that a strange class of bacterial genetic elements called retrons could churn out loops of single-stranded DNA, but nobody could say why. A team at the Gladstone Institutes in San Francisco has now described a version of the system that works in reverse of every retron studied so far, flipping on a cell-killing toxin only after a virus attacks rather than keeping that toxin permanently suppressed. The findings, led by graduate student Karen Zhang and Gladstone investigator Seth Shipman, appear in a paper published this week in the journal Molecular Cell.

Retrons are small genetic modules, built around a reverse transcriptase, that many bacteria carry as part of their defense against bacteriophages, the viruses that infect bacteria. Until recently their job was a mystery. A 2022 study published in Nature by a team at the European Molecular Biology Laboratory, alongside an independent group at the Weizmann Institute of Science, first showed that retrons act as toxin-antitoxin pairs: the cell constantly produces a small stretch of DNA that keeps a lethal toxin locked down, and a phage infection disrupts that DNA, releasing the toxin and killing the infected cell before the virus can multiply and spread to its neighbors. One of that study's authors, Jacob Bobonis, said in an EMBL announcement at the time that "for more than 30 years, we'd had no clue why bacteria have retrons," a line that captures how long the function went unexplained even after the molecules themselves were first noticed in the 1980s.

A system that runs backward

"This is, by far, the strangest of the retron systems that we or anyone else has come across," said Seth Shipman, the Gladstone investigator who led the study.

The retron Zhang and Shipman's team examined, called Retron-Vpa2 and found in the marine bacterium Vibrio parahaemolyticus, does not fit that pattern at all. In uninfected cells, it produces no detectable single-stranded DNA and no measurable toxin, the opposite of the constant, low-level toxin production that other retrons hold in check. Only after a phage infects the cell does the retron spring into action: it reverse-transcribes a new DNA molecule, and that DNA then releases a block on the translation of a small protein that embeds itself in the cell's membrane, destabilizing it and killing the cell. The effect limits the phage's ability to spread, but the molecular logic runs opposite to every previously described retron.

The team traced the mechanism using Spacer-seq, a reporter tool built in the Shipman lab from repurposed CRISPR-Cas components, which let them measure how much of the retron's signature DNA was present in cells before and after phage exposure. They found a structured piece of RNA inside the bacterium, a hybrid transcript carrying both the template the reverse transcriptase copies and the gene for the toxin itself, with translation of that toxin gene kept shut off by the RNA's folded structure. Phage infection triggers a burst of newly made DNA, which binds the RNA and opens up the toxin's coding sequence to the cell's ribosomes. "It's almost the mirror-image version of how we traditionally view retrons, because the DNA switches on toxin production instead of inhibiting an already produced toxin," Zhang, the study's lead author, said of the finding. The paper lists fifteen co-authors across Gladstone, the University of California, San Francisco, and collaborating structural biology groups, and credits funding including the Gordon and Betty Moore Foundation, the Robert J. Kleberg Jr. and Helen C. Kleberg Foundation, and the National Science Foundation. A version of the work had circulated for a year as a preprint before this week's peer-reviewed publication, first posted online in October 2025.

Why a bacterial immune system matters beyond bacteria

The discovery lands at a moment when the search for new ways to fight drug-resistant bacteria has taken on new urgency. The World Health Organization's member states adopted an updated global action plan on antimicrobial resistance covering 2026 through 2036 earlier this year, reflecting concern that resistant infections are rising faster than the pipeline of new antibiotics can keep up. Retrons sit at the center of two separate efforts to respond to that pressure. One is phage therapy, the decades-old idea of using viruses to kill harmful bacteria, which has drawn renewed interest as conventional antibiotics lose effectiveness; understanding how bacteria disable phages, as this paper does, also clarifies how phage treatments might be designed to get around those defenses. The other is synthetic biology, where retrons' natural ability to convert a biological signal, such as the presence of a virus, into a precise genetic output has already made them useful tools for genome editing and cellular recording. Zhang said the newly described system "represents a unique way to potentially build a molecular sensor and switch encoded in DNA and RNA that no one has thought of before," pointing toward engineered circuits that could be triggered by a chosen signal rather than a phage.

The Gladstone group frames the practical payoff in two directions: diagnostic or control tools built from a retron that stays quiet until triggered, and new angles on antimicrobial design that exploit toxin-release mechanisms bacteria already carry. Neither application is close to deployment. The paper is a mechanistic study of one retron family in one species of marine bacterium, not a therapy or a product, and the authors' own framing is about expanding the catalog of parts available to synthetic biologists and infectious-disease researchers rather than announcing an immediate fix for resistant infections.

What is still unknown

Several questions remain open. The researchers do not yet know how widespread this reversed logic is among the hundreds of retron variants that genome surveys have turned up in bacteria across different environments, or whether related systems respond to triggers other than phage infection. The structural details of how the newly made DNA physically unlocks the folded RNA are still being worked out; two of the paper's co-authors, Guillermo Montoya and Arturo Carabias, work on structural biology and are likely candidates to pursue a molecular, atom-level picture of the switch in follow-up work. Reporting on the study by Phys.org, drawing on a release from Gladstone, also noted that researchers have not yet shown the mechanism works outside laboratory conditions or tested whether it could be repurposed into a working biosensor, steps that would need to come before any practical application.

The broader retron field has been moving quickly. Separate groups have spent recent years cataloging dozens of previously untested retrons from environmental bacteria and engineering others into tools for multiplexed genome editing, building on the realization that retrons defend against phages at all, which only came after their reverse transcriptase activity had puzzled microbiologists since the 1980s. Zhang and Shipman's description of a retron that inverts that basic defensive logic suggests the known diversity of these systems is still far from complete, and that bacteria have found more than one way to turn a once-mysterious stretch of DNA into a weapon against the viruses that prey on them.