Bacteria sense a virus attack when the virus’s own enzyme cuts a tripwire, setting off self-destruct

Some viruses that infect bacteria carry an enzyme that gives them away. When that enzyme, a protease, cuts a particular sensor protein inside the bacterium, the cut itself sounds the alarm and switches on CBASS, a defense that ends with the infected cell killing itself before the virus can spread to its neighbors.

Sam Hobbs, an assistant professor of biochemistry at University of Utah Health, is first author of the paper, “Phage proteases activate CBASS antiphage immunity,” which appeared in Science on Oct. 1, 2026.

A protease that cuts the sensor it was meant to avoid

The work is listed in the Hobbs laboratory’s publication record as a Science paper, volume 394, issue 6819, pages 102 to 106, with Hobbs and Philip Kranzusch as co-corresponding authors. According to the University of Utah Health release, certain phages, the viruses that attack bacteria, make a protease, an enzyme whose ordinary job is to degrade other proteins. The finding is that this protease acts directly on a host protein, and that the cleavage of that protein is the signal the bacterium reads.

The sensed molecule is one the virus needs to complete its infection, so the trap is built around something the phage cannot do without. Hobbs said the result ran against his expectations, since he never would have guessed the defense worked this way, and he called the moment of discovery “a total eureka moment.” The cut works as a tripwire in the plainest sense: the bacterium does not need to see the virus’s genome or its coat, only to notice that its own sensor has been severed by a viral enzyme.

The sources read for this article do not name the sensor protein or the phage species used. The release’s illustration shows a phage infecting E. coli, but no text identifies the strain, and the lab’s own listing gives the paper a slightly different short title, “Phage protease enzymes activate CBASS immunity,” than the one on the journal citation.

CBASS: a last-resort defense that kills the infected cell

CBASS is a bacterial immune system whose full name, cyclic oligonucleotide-based antiphage signaling system, appears in Genetic Engineering and Biotechnology News. Its sensors are cGAS/DncV-like nucleotidyltransferases, known as CD-NTases, which sense phage infection and make nucleotide signals to start the antiviral response. The same report says a phage prohead protease cleaves the host sensor, and that the study found this to be a widespread way CD-NTases get activated.

What follows is the self-destruct step. The university describes CBASS as a “last resort” response: it kills the infected bacterium, which stops the virus from spreading to nearby cells. A single cell is lost, and the colony around it survives, a trade that makes sense for a population of bacteria facing a virus that would otherwise replicate inside one cell and burst out to infect the next.

That logic also separates CBASS from related antiviral pathways. Those pathways respond to the presence of viral genetic material, whereas CBASS in this case is triggered by a viral protein acting on a host protein. Hobbs called it “a totally new mechanism for how these host proteins are activated,” a quote carried in News-Medical’s account, which also quotes him on how common the defense is: “This is one of the most common forms of bacterial immunity.”

Phage therapy and the human cGAS connection

The practical stake the researchers name is phage therapy. Phages kill bacteria without harming human cells and can bypass antibiotic resistance, which makes them candidates against resistant infections. If bacteria detect certain phages through their own proteases, as the Bioengineer summary of the paper notes, then knowing the trigger could help researchers design phages that evade the bacterial defense, so that a therapeutic phage is not detected and the infected cell is not triggered to kill itself before the virus has done its work against the pathogen.

The paper also points back toward people. CBASS is related to an immune pathway in humans, which the university says suggests it has been conserved since bacteria and humans shared a common ancestor. Bacteria offer a fast, simple experimental system for studying that kind of immunity, so a mechanism worked out in a bacterial cell is treated as relevant to the study of immune signaling more broadly. The paper does not claim a direct human application; the connection the university draws is one of shared ancestry.

The study was supported by the Pew Biomedical Scholars program, the Burroughs Wellcome Fund, the G. Harold and Leila Y. Mathers Foundation, the Cancer Research Institute, the Parker Institute for Cancer Immunotherapy, the Massachusetts Consortium on Pathogen Readiness and the National Institute of General Medical Sciences at the NIH, according to the republished university summary.

Which phage proteases and which host sensors the finding extends to, beyond the cases in the paper, is left unnamed in every public summary of the work.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview