Researchers report protein that writes DNA

Southern University of Science and Technology researchers have reported that a bacterial antiviral protein can build DNA without copying a DNA or RNA template, according to a bioRxiv preprint posted in April and circulating widely online this week. The study describes a defense-associated reverse transcriptase called DRT7 that the authors say makes sequence-specific DNA in vitro through a protein-templated mechanism rather than standard base-pair copying. The paper has not been certified by peer review, bioRxiv says. The work adds to a recent run of papers on bacterial defense systems that use reverse transcriptases in nonstandard ways. ### Which protein is at the center of the claim? DRT7 is the protein complex at the center of the new report, and the authors say it contains both a reverse transcriptase, or RT, domain and a primase-polymerase domain. In the preprint, Xin-Yi Song, Yushan Xia, Jun-Tao Zhang and colleagues at Southern University of Science and Technology wrote that DRT7 provides broad-spectrum anti-phage immunity through abortive infection and can be activated by a phage-encoded putative transcriptional regulator. (biorxiv.org) Ning Jia is listed as the corresponding author, and the paper says the team is based in Shenzhen, China. The authors reported that, once activated, DRT7 synthesizes long, protein-primed, palindromic poly(A)/poly(T)-rich duplex-like DNA. ### What exactly do the authors say breaks with the usual template rule? The core claim is in the abstract: DRT7’s RT domain initiates “protein-templated” and sequence-specific poly(T) synthesis without a complementary nucleic acid template. (biorxiv.org) The authors wrote that cryo-electron microscopy structures show an arginine-rich recognition pocket that enables that step and shifts DRT7 from an inactive closed dimer to an active open dimer. (biorxiv.org) The same preprint says the poly(T) product then becomes both primer and template for the primase-polymerase domain to extend poly(A), producing alternating poly(A)/poly(T) single-stranded tracts that fold back into duplex-like DNA. That means the paper is not claiming all DNA synthesis is template-free; it is claiming a specific initiating step can proceed without a nucleic-acid template in this bacterial defense system. (biorxiv.org) ### Is this the first report of protein-templated DNA synthesis? A Science paper published about four weeks ago reported a related but distinct bacterial defense system, DRT3, that also makes DNA in the absence of a nucleic-acid template. Pujuan Deng, Hyunbin Lee, Carlo Armijo and colleagues at Stanford wrote that one DRT3 subunit, Drt3b, synthesizes a complementary poly(AC) strand “in the complete absence of a nucleic acid template,” using conserved active-site residues to enforce base alternation. (biorxiv.org) That earlier paper matters for context because it shows the DRT7 report is not appearing in isolation. A separate 2025 bioRxiv preprint from Columbia University and collaborators also described DRT9 as an antiviral reverse transcriptase that makes poly-dA through another unusual mechanism, though that work still used an RNA template for the polymer product described there. ### What has been released publicly, and what has not? (science.org) The bioRxiv record shows the DRT7 manuscript was posted about three weeks ago and is labeled as a preprint that has not been peer reviewed. The abstract, author list and PDF are publicly available through bioRxiv. Public search results reviewed Thursday did not clearly surface a primary-source repository entry for the protocols or datasets referenced in social-media discussion of May 13 postings. (biorxiv.org) That does not mean such material does not exist; it means it was not readily verifiable from authoritative repository records in the sources reviewed here. The verified public source is the bioRxiv preprint itself. (biorxiv.org) ### What should readers watch next? Peer review is the next formal checkpoint because bioRxiv states the DRT7 paper has not been certified by peer review. Independent replication is the other concrete next step, especially for the claim that a protein-defined active site can direct sequence-specific DNA synthesis without a nucleic-acid template. Science’s April paper on DRT3 and bioRxiv’s April posting on DRT7 suggest that bacterial defense-associated reverse transcriptases are becoming an active research lane. (biorxiv.org) The next public milestones are likely to be journal review, updated manuscript versions, or follow-up studies from named groups including Ning Jia’s team in Shenzhen and Alex Gao’s team at Stanford. (science.org)