Two teams publish DNA-guided CRISPR-Cas12 systems in Nature Biotechnology, inverting the RNA-guide paradigm for RNA diagnostics and antivirals

Overview

Two independent research teams have published companion advances in Nature Biotechnology that invert the conventional CRISPR architecture: instead of an RNA guide directing a Cas protein to a DNA target, both groups use a DNA guide to direct a Cas12 protein to an RNA target. A team at the University of Florida led by Piyush Jain reported its system in a paper titled “DNA-guided CRISPR–Cas12 for cellular RNA targeting”, while a separate team at the Hong Kong University of Science and Technology (HKUST) led by Prof. Hsing I-Ming reported a parallel system in “DNA-guided CRISPR–Cas12a effectors for programmable RNA recognition and cleavage”. The two papers, indexed under DOIs 10.1038/s41587-026-03129-w and 10.1038/s41587-026-03120-5 respectively, together describe a route to cheaper, more stable guides and reduced off-target cleavage when CRISPR is used to interrogate or cut RNA.

What We Know

The University of Florida system

The UF system was developed in the lab of Piyush Jain, an associate professor and Shah Rising Professor in the Department of Chemical Engineering at the University of Florida, according to a University of Florida News release dated May 15, 2026. Co-first authors on the paper are Carlos Orosco, Boyu Huang, and Santosh Rananaware, working in collaboration with researchers at UT Austin and Princeton University, the UF release reports. The work, first posted as a 2024 preprint, has now been formally published in Nature Biotechnology, according to Phys.org, which lists the article DOI as 10.1038/s41587-026-03129-w.

Jain framed the rationale by contrasting RNA copies of genetic information with their DNA originals. “Those RNA copies are like Xerox copies of the original manual, and sometimes those copies have errors,” Jain said in the UF release. “Existing RNA-targeting CRISPR systems rely on RNA guides to find their targets. While effective, they can sometimes affect unintended molecules, creating off-target effects. They can also be costly and less stable.” The new approach, he added, “gives us a way to fix or tune the instructions the cell is using in real time, without immediately changing the DNA.”

On the manufacturing case, Jain told UF News that “DNA guides are far cheaper and easier to manufacture than RNA guides, and they are far more stable. While RNA molecules degrade quickly, DNA can remain intact for long periods.” Orosco said the project “required a great deal of persistence and creativity because we were exploring an idea that challenged conventional thinking,” according to the same UF release.

On applications, the UF team reports detecting hepatitis C with 100% accuracy and demonstrating early detection of HIV, according to Phys.org. The UF release describes a drastic reduction in unintended effects compared with existing approaches and improved precision “by orders of magnitude in some cases,” as reported by Phys.org.

The HKUST system and the SLEUTH platform

The HKUST paper describes a DNA-guided Cas12a effector combined with isothermal amplification to form a diagnostic platform the team calls SLEUTH, short for Specific Locus Evaluation Utilizing Targeted Hydrolysis, according to the HKUST press release dated May 6, 2026. The work was led by Prof. Hsing I-Ming of HKUST’s Department of Chemical and Biological Engineering, with Prof. Zhai Yuanliang of HKUST’s Division of Life Science as collaborator. Co-first authors are PhD student Wu Xiaolong of Chemical and Biological Engineering and Dr. Lam Wai-Hei, a postdoctoral fellow in Life Science, with Dr. Cao Yumeng of Chemical and Biological Engineering also on the team, HKUST reports.

The SLEUTH platform uses a synthetic “CRISPR DNA,” or crDNA, molecule to reprogram Cas12a so that it recognises RNA targets via a DNA guide, according to Phys.org’s writeup of the HKUST paper. The team validated the platform with 31 clinical SARS-CoV-2 samples and reported attomolar-level sensitivity for both RNA and DNA targets, according to Phys.org and confirmed in the HKUST release. Hsing characterised the conventional architecture this way: “The RNA guide molecule is like the address you type in, and the Cas protein is the car,” as reported by Phys.org.

The HKUST team validated the system using AlphaFold-guided structural modeling, molecular dynamics simulations, and high-resolution cryo-electron microscopy, Phys.org reports. HKUST has filed two U.S. provisional patents covering the technology, according to the HKUST release, and plans over the next three years to expand SLEUTH to detect other respiratory viruses and to explore liquid biopsy applications for identifying circulating RNA biomarkers in cancer.

Shared advantages over RNA-guided RNA targeting

Both groups make a similar argument for replacing RNA guides with DNA guides when the target is itself an RNA molecule. RNA guides degrade quickly and are expensive to synthesize relative to DNA, as the UF team frames it via Jain’s quote in the UF release. The HKUST team independently reports that DNA guides eliminate cold-chain storage requirements and enable single-nucleotide discrimination, according to Mirage News and Phys.org. The HKUST system, Phys.org reports, also reduces off-target RNA cleavage in cells compared with Cas13-based RNA-targeting CRISPR tools, which the press release frames as a critical safety consideration for any future therapeutic development.

What We Don’t Know

Neither press release attaches quantitative numbers to the off-target reduction in animal models or in human cells beyond the diagnostic validation set. The UF release describes precision improvements “by orders of magnitude in some cases” as reported by Phys.org, without specifying which targets, conditions, or comparisons were used. The HKUST clinical validation rests on 31 SARS-CoV-2 samples, a small set whose generalisation to other RNA pathogens and to liquid biopsy contexts is presented as future work spread over three years, according to the HKUST release.

Whether the two DNA-guided architectures behave equivalently in therapeutic settings is also unsettled. Jain’s group emphasises ex vivo cell and tissue applications, with Jain estimating that early targeted uses could emerge “within a few years,” according to Phys.org. The HKUST group has filed two U.S. provisional patents and is pursuing diagnostic and antiviral routes first, per the HKUST release. The clinical and regulatory paths beyond diagnostics remain to be mapped.

The Machine Herald has previously covered an alternative RNA-triggered CRISPR architecture, Cas12a2, that shreds host DNA when activated by viral or cancer RNA. The two systems described here invert the same DNA–RNA boundary in the opposite direction: rather than letting RNA trigger DNA cleavage, they let DNA guides direct RNA cleavage. Whether the two strategies prove complementary in clinical pipelines or compete for the same RNA-modulation niche will depend on follow-up data neither team has yet published.