Scientists cut PRNP 50% in mice

Prion disease is one of the bleakest problems in neurology. A misfolded version of one protein — prion protein, or PrP — can trigger a chain reaction in the brain, and once that starts, patients usually decline fast. The obvious idea has been to lower the normal protein before it can misfold and spread. What changed is that a team from the Broad Institute, Mass General Brigham, and collaborators showed that a one-time base-editing treatment can do exactly that in mice carrying human PRNP — and the animals lived substantially longer. ### What is PRNP, exactly? PRNP is the gene that encodes PrP, the protein prions corrupt. That makes it a very unusual drug target — you are not trying to fix the bad protein after it appears, you are trying to reduce the supply of raw material the disease needs. Human genetics has pointed this way for years, because people with naturally lower PrP levels seem protected, and animal work has repeatedly shown that less PrP usually means slower disease. (nature.com) ### What did the team actually edit? They used base editing, which is a CRISPR-derived method that changes single DNA letters without cutting both strands. In this case the editor installed a stop signal called R37X in PRNP. Basically, the edit tells the cell to stop making full-length PrP. The system was packaged in two AAV vectors and delivered systemically into mice engineered to carry the human gene. (nature.com) ### How strong was the effect? The headline numbers are clean. The edit landed in about 37% of PRNP copies on average. That was enough to reduce PrP in the brain by about 50% versus untreated animals. In mice challenged with two common human prion isolates — one sporadic and one genetic — median lifespan increased by 52%. That is a big effect for a disease model this hard to move. (nature.com) ### Why does 37% editing become 50% protein lowering? Because biology is not a one-to-one scoreboard. Each edited gene copy can completely shut down production from that copy, and the treatment does not need to hit every cell equally to shift total protein levels meaningfully. The practical point is the important one — partial editing was enough to cut overall PrP roughly in half, and half appears to be biologically meaningful. (nature.com) ### Is this the first proof that PRNP lowering works? Not really — but it is one of the strongest versions. Other groups have used epigenetic silencing, zinc-finger repressors, and RNA approaches to push PrP down and extend survival in mice. What makes this paper stand out is the combination of a humanized mouse model, human prion isolates, and a one-time gene-editing intervention that directly rewrites PRNP. (nature.com) ### So why isn’t this ready for people? Delivery is the catch. The mice got an AAV system that worked well enough for a proof of concept, but human brains are bigger, harder to reach evenly, and much less forgiving about off-target editing or immune reactions. The team also had to optimize the editor to improve efficiency and reduce unwanted activity outside the target. That is progress, but it is still preclinical progress. (science.org) ### What does this mean for prion drug development? It sharpens the field’s direction. The main strategic question is no longer whether lowering PrP helps — that case keeps getting stronger. The real question is which modality gets there first in humans: gene editing, gene silencing, or repeated RNA drugs. Each makes a different tradeoff between durability, control, and safety. (broadinstitute.org) ### Bottom line This is not a cure, and it is not a human trial result. But it is a serious signal. Cut PRNP by about half, and prion disease in these mice slows down a lot. For a disease with no approved treatment, that is the kind of result that can change what the field builds next. (nature.com) (biorxiv.org)