Smaller and sharper: Compact CRISPR scissors cut with precision

Gene editing has reshaped biology and produced its first approved medicine, yet CRISPR, the widely used gene-editing technology, faces two stubborn obstacles on its march into the clinic: the leading editors are too big to deliver with standard viral vectors, and they can cut DNA in the wrong place, raising fears about unintended genetic changes.

Two new studies from researchers in KAUST describe a new generation of genome editors designed to tackle both challenges at once. These editors are small enough to fit inside a single viral delivery vehicle, while also engineered to edit DNA with greater precision.

Currently, all approved CRISPR therapies require cells to be removed from patients, edited outside the body, and reinfused after intensive chemotherapy. This complex procedure can cost millions of dollars, carry substantial treatment-related risks and work only for conditions such as sickle-cell disease and beta-thalassemia that originate in blood-forming stem cells.

If the KAUST discoveries can be translated from laboratory experiments into clinical therapies, it may eventually become possible to deliver gene-editing treatments directly into the body using a single virus carrying a compact, highly specific editor. The authors of the studies, in collaboration with clinical and translational partners, are actively pursuing therapeutic applications of these new CRISPR technologies, with a focus on rare genetic disorders.

“Together, these results move the field one concrete step closer to a CRISPR therapeutic platform that is small enough to deliver, specific enough to trust, and effective enough to use,” says Magdy Mahfouz, head of the Laboratory for Genome Engineering and Synthetic Biology, who led the research.

One of the new studies, published in Trends in Biotechnology, focused on the problem of size. Mahfouz and his colleagues computationally combed through millions of viral genomes in search of sequences encoding Cas12j, a naturally tiny type of CRISPR enzyme originally discovered in bacteria-infecting viruses. In this way, they identified eight previously unknown Cas12j variants, expanding the known diversity of this promising family of miniature gene editors^[1].

In their natural form, these enzymes proved only weakly adept at editing human cells, but targeted engineering changed that. By fusing the Cas12j variants to a second enzyme that helps cells process broken DNA ends, the researchers transformed the new suite of CRISPR proteins into robust genome-editing tools. In laboratory tests, the engineered editors performed on par with Cas12a, one of the field’s workhorse CRISPR systems, despite being substantially smaller.

Along the way, the team uncovered an unexpected design rule of Cas12j editors: they work most efficiently when their DNA targets contain a specific three-letter sequence, TAC. “This is a previously unrecognized design principle that now guides how the entire Cas12j family should be deployed,” says Wenjun Jiang, a Ph.D. student in Mahfouz’s group, and the co-first author of the study.

The second study, published in Nucleic Acids Research, addressed the other major obstacle facing CRISPR therapeutics: specificity. The KAUST team began with a compact CRISPR protein called GoCas12m that binds DNA but does not cut it. They fused on to it a DNA-cleaving enzyme called FokI nuclease^[2]. Because FokI can cut DNA only when two copies of the enzyme come together, two complete GoCas12m-FokI complexes must bind adjacent genomic sites before editing occurs.

As a result, the system behaves like a molecular two-factor authentication scheme: “It only cuts DNA when two molecules find their target side by side, making accidental edits at the wrong location far less likely,” explains Tin Marsic, former Ph.D. student in Mahfouz’s group, who describes the GoCas12m-FokI editor that he helped build as “cautious by design.”

The researchers demonstrated highly specific genome editing at several clinically relevant targets, including the AIFM1 gene linked to mitochondrial disorders and the ABL gene implicated in leukemia, with no detectable off-target activity at the genomic sites examined.

The new editors remain at an early stage of development and have not yet been tested in mice, let alone people. However, they provide a foundation for what Mahfouz and his colleagues hope will become a new generation of precision genetic medicines that, owing to their small size, could one day be delivered directly into patients with a simple injection.

The implications also extend beyond human health. “The same GoCas12m–FokI architecture could, in principle, be adapted to edit crop genomes, to improve microbial production strains or to target pathogens, with similar gains in precision and deliverability,” says Sivakrishna Gundra, a postdoc who made major contributions to both studies.

For Mahfouz’s team, the goal is not merely to build better molecular scissors, but to create editors small and reliable enough to function as versatile tools across medicine, agriculture and industry.