Prion diseases are a group of rare infectious neurodegenerative diseases caused by modifications in the prion protein (PrP) that prompt it to misfold.1 Invariably fatal, prion disease can be genetic, infectious, or sporadic with no known cause. Despite research efforts, there are currently no approved treatment options that can slow or stop disease progression.
At the 2026 annual meeting of the American Society of Gene and Cell Therapy, Jessie Davis, a chemical biologist in prion scientists Sonia Vallabh and Eric Minikel’s group at the Broad Institute, described the development of base editing (BE) strategies to tackle prion disease. Published in Nature Medicine, this approach was able to halt the production of PrP in a mouse model by inserting an early stop codon into the defective prion protein (PRNP) gene.2
“This should be a helpful therapeutic for prion disease, regardless of etiology,” Davis said during her talk at the “Advances in Cell and Gene Therapies for Chronic Infectious Diseases” session of the conference. “For genetic prion disease, there are benefits where we could potentially treat earlier, more prophylactically, but knockdown [of the PRNP gene] should be helpful for any form of prion disease.”
The team began by screening for positions in the PRNP gene where they could use cytosine base editors (CBEs) to convert a cytosine to uracil and create a premature stop codon. They homed in on one particular target to mutate, R37X. “It's high efficiency, there are low indels, and importantly, only silent bystander edits,” Davis remarked. “The other thing that's really fantastic about this edit is that this allele has been observed in heterozygous humans who do not have a propensity for prion disease, so we know it's compatible with healthy life and is not a prion disease-causing mutation.”
Chemical biologist Jessie Davis and her colleagues used a split-BE approach to move the system into an in vivo mouse model, incorporating a synapsin promoter and microRNA targeting sites to restrict editing to neurons.
Allison Colorado
To move the method in vivo, Davis and her colleagues first had to use a split-BE approach; base editor proteins are large, so packaging them inside a single adeno-associated viral vector (AAV) is a major challenge. By cutting the base editor protein BE3.9max in half and bundling each in a separate viral vector along with the single guide RNA, the team was able to deliver them in vivo to the brains of mice. Developed for therapeutic applications, BE3.9 max is commonly used in this type of split-BE approach. “When these are co-transduced in a cell, both are expressed,” Davis explained. “The protein halves come together by tertiary interactions, with help from the [Nostoc punctiforme PCC73102] Npu inteins, [which are] kind of like protein velcro, and this reconstituted protein is then active.”3
Using this approach, the team achieved 20 percent editing efficiency and a 31 percent reduction in PrP in the brains of humanized PrP mice. They also tested the approach in mice that were inoculated with human prion isolates, measuring the animals’ survival. “[With] the R37X installation, we see a 52 percent lifespan extension, and we were really encouraged by this,” said Davis.
To increase the safety profile and potency of the treatment, the team incorporated a human synapsin gene promoter and two microRNA targeting sites, restricting gene editing to neurons. “When we looked in the brain [of the mice], what we see is that we both maintain editing, and actually even maybe increase potency a little bit,” Davis said.
The researchers also investigated the possibility of reducing the high dose of AAV required by the original split-BE approach by using smaller Cas proteins or adenine base editors. Unfortunately, Davis said, these strategies had lower editing efficiencies and reductions in PrP than the original split-BE approach. “We are, however, still continuing to advance single AAV base editor strategies, and are having some good success,” she added.
The team is currently moving forward with the split-BE approach. Sharing some unpublished data, Davis explained that she and her colleagues are testing the use of AAV capsids that are engineered to better cross the blood-brain barrier. “What you can see, really excitingly, is, in addition to maintaining editing efficiency and protein knockdown with just a more relevant capsid. The other thing that's really nice to see is that we actually get a bit of a potency boost,” Davis said. “[That] will be really meaningful for clinical uses of this or similar technologies.”
- Prusiner SB. Prions. Proc Natl Acad Sci USA. 1998;95(23):13363-13383.
- An M, et al. In vivo base editing extends lifespan of a humanized mouse model of prion disease. Nat Med. 2025;31(4):1319-1328.
- Humberg C, et al. A cysteine-less and ultra-fast split intein rationally engineered from being aggregation-prone to highly efficient in protein trans-splicing. Nat Commun. 2025;16(1):2723.
