About half of all human cancers carry a broken version of a single protein called p53. It normally acts as a brake on runaway cell division. When it is mutated, that brake fails. You would think a target this common would have been drugged decades ago. It has not. The mutant protein is smooth and featureless, with no obvious pocket for a small molecule to grab. Drug hunters call proteins like this undruggable, and p53 is the most famous name on that list.
A team led by Jennifer Doudna, who shared the 2020 Nobel Prize for CRISPR, has published a different way in. Instead of trying to fix or block the broken protein, their engineered system waits for the cancer cell to reveal itself through its RNA, then destroys the cell from the inside. The work appeared this week in Nature.
Turning a bacterial defense into a trigger
The tool is built from CRISPR-Cas12a2, a nuclease that bacteria use to fight viruses. What makes Cas12a2 unusual is what it does after it recognizes a target. Most CRISPR enzymes make one clean cut at a specific spot. Cas12a2 does something messier. Once its guide RNA locks onto a matching sequence, the enzyme switches into a shredding mode and starts cutting nearby nucleic acids indiscriminately. In bacteria that is a scorched-earth tactic. If a virus gets in, the cell degrades its own genetic material to stop the infection from spreading.
Doudna's group repurposed that behavior. They programmed the enzyme to recognize a transcript that is specific to cancer cells, essentially a molecular fingerprint written in RNA. When Cas12a2 finds that signal, it flips into shredding mode and tears through the cell's chromatin, the packaged DNA in the nucleus. That damage sets off the cell's alarm systems, the DNA damage response fires, and the cell dies. Cells that lack the cancer signature never trigger the enzyme, so they are left alone.
The elegance is in the sensing step. The researchers are not editing a gene or repairing a mutation. They are using the cancer cell's own abnormal RNA as a password. A cell that speaks the wrong genetic language gets shredded. A healthy cell does not.
Why RNA sensing matters for undruggable targets
This flips the usual logic of cancer therapy. Conventional drugs need a physical handhold on a protein. That requirement is exactly what leaves p53 and many other tumor-suppressor mutations out of reach. By reading transcripts instead of proteins, the Cas12a2 approach sidesteps the whole problem of finding a binding pocket. In principle, any mutation that produces a distinctive RNA could become a target, whether or not the resulting protein has a druggable surface.
The authors frame this as a general strategy rather than a one-off trick. Transcript-activated chromatin shredding, as they call it, offers a route to precision treatments for the large fraction of cancer mutations that current medicine simply cannot address.
What the study does not settle
This is early work, and the paper reads like a proof of principle. The results describe engineered enzymes acting on cancer cells, not a therapy tested in patients or even, from the abstract, in whole animals. Two hard problems stand between a clever mechanism and a real drug. The first is delivery. Getting a large CRISPR enzyme into tumor cells throughout the body, and only tumor cells, is an unsolved challenge across the whole field. The second is specificity. A system that shreds chromatin is powerful and unforgiving, so the margin for error in target selection is thin. If the enzyme misreads a signal in a healthy cell, the consequences are not subtle.
None of that diminishes what has been shown. Programming a self-limiting bacterial nuclease to sense a cancer transcript and then commit the cell to death is a genuinely new idea. Whether it becomes a treatment will depend on the parts that always take longest: delivery, safety, and years of testing. For now it is a demonstration that the undruggable label may say more about our tools than about the biology itself.
Comments