Researchers at the University of Illinois Urbana-Champaign have shown, in a rodent model, that CRISPR base editing can reduce the toxic fragmentation of the mutant huntingtin protein that drives Huntington’s disease — not by rewriting the disease-causing CAG repeat itself, but by using base editing to trigger exon skipping that removes a proteolytic cleavage site further along the protein. The work, led by Shraddha Shirguppe, Michael Gapinske and Devyani Swami with senior author Pablo Perez-Pinera, was published July 29, 2026 in Nature Biomedical Engineering. It is a preclinical, in vivo laboratory study, not a clinical trial — the relevant milestone here is a systematic engineering result, not a treatment.
A 141-variant screen, not a single lucky edit
Huntington’s disease is caused by an expanded CAG repeat in exon 1 of the huntingtin (HTT) gene. The expanded repeat produces a mutant huntingtin protein that is prone to proteolytic cleavage into smaller fragments, and it is these fragments — not necessarily the full-length protein — that aggregate and drive much of the neuronal toxicity observed in the disease. Rather than editing the repeat directly, the Illinois team set out to disrupt the region of the gene responsible for producing one of those toxic cleavage products.
To find an editing strategy that would do this reliably, the researchers screened 141 distinct CRISPR base editor variants before identifying ones that worked as intended. That screening scale is itself a data-transparency point worth noting: base editors vary in the enzyme deployed, the editing window, and the guide RNA used, and small differences in any of those can mean the difference between an edit that produces the intended splicing change and one that does nothing or edits the wrong site. Screening at this breadth, rather than reporting results from a single hand-picked construct, is what lets the eventual finding be checked and reproduced by other labs rather than taken on faith.
What “base editing” and “exon skipping” mean here
CRISPR base editors are a variant of CRISPR gene-editing tools that convert one DNA base to another (for example, an A to a G) at a targeted site without cutting both strands of the DNA double helix, the way the original Cas9 nuclease does. Because they don’t rely on the cell’s double-strand-break repair machinery, base editors tend to produce fewer unintended insertions or deletions at the edit site.
In this study, the researchers used base editors to alter the splice acceptor site of HTT exon 13 — the short DNA sequence a cell’s splicing machinery reads to decide where an exon begins during production of the mature messenger RNA. Disrupting that splice acceptor causes exon 13 to be skipped during splicing, so the resulting HTT protein is produced without the segment that exon encodes. According to the published abstract, this yields a “proteolysis-resistant” form of the HTT protein — one less prone to being cleaved into the fragments implicated in aggregation and toxicity. This is a distinct strategy from approaches that attempt to shorten or excise the CAG repeat itself, or from antisense oligonucleotide approaches that lower total huntingtin expression; here, the protein is still produced, just in a form more resistant to the cleavage step that generates toxic fragments.
What the in vivo results showed
The editing strategy was delivered to the striatum of rodents — the brain region most affected in Huntington’s disease. According to the published findings, treated animals showed reduced HTT fragment formation, decreased protein aggregation, improved performance on functional/behavioral measures, and attenuated brain atrophy compared to untreated controls. Those are the standard preclinical readouts used to gauge whether a Huntington’s disease intervention is doing what it is meant to do in an animal model — they are evidence that the mechanism works as designed in vivo, not evidence of safety or efficacy in humans, which would require a substantially longer path through further animal studies and formal clinical development.
From preprint to peer review
This result has a documented public record stretching back further than its formal publication date. An earlier version of this work was posted as a preprint on bioRxiv on July 6, 2024, roughly two years before the peer-reviewed version appeared in Nature Biomedical Engineering. That gap gave outside researchers an early, citable look at the approach and the underlying data well before formal peer review concluded — a practical example of the kind of reproducibility and open-science norm this site tracks: making methods and results available for scrutiny before, not only after, a journal’s own review process runs its course.
What this is, and isn’t
This is preclinical laboratory research: an engineered gene-editing tool that worked as intended in a rodent model of Huntington’s disease. It is not a treatment, is not in clinical trials, and its authors have not claimed either. The path from a positive rodent result to a human therapy — larger and longer-term animal safety studies, an Investigational New Drug application, and multiple phases of human clinical trials — is long, and most preclinical gene-editing results at this stage do not ultimately reach approved therapies. The value of this specific result is narrower and more concrete: a systematically screened, exon-skipping-based base-editing strategy that measurably reduced markers of Huntington’s disease pathology in a living animal, with the underlying data made public well ahead of formal publication.
Frequently asked questions
Does this base-editing approach target the CAG repeat that causes Huntington’s disease?
No. The CAG repeat itself, in exon 1 of the HTT gene, is left unedited. This approach instead disrupts the splice acceptor site of a different exon (exon 13), causing it to be skipped during splicing and producing a huntingtin protein that is more resistant to the proteolytic cleavage step that generates toxic fragments.
Is this a treatment for Huntington’s disease?
No. This is a preclinical, in vivo laboratory study in a rodent model, published in a peer-reviewed engineering journal. It demonstrates that the editing strategy reduces markers of disease pathology in animals; it has not been tested in humans and is not a clinical trial.
Why does screening 141 base editor variants matter?
Base editors differ in the enzyme used, the width of the editing window, and guide RNA design, and small differences among them can determine whether an edit lands on the intended site and produces the intended effect. Systematically screening a large panel, rather than reporting a single selected construct, is a data-transparency practice that makes the eventual result easier for other labs to evaluate and reproduce.
Sources
Primary source: Shirguppe, S., Gapinske, M., Swami, D. et al., “In vivo CRISPR base editing for treatment of Huntington’s disease,” Nature Biomedical Engineering, published July 29, 2026. DOI: 10.1038/s41551-026-01747-y. An earlier version of this work was posted as a preprint on bioRxiv on July 6, 2024. This article was last checked against the cited source on August 7, 2026.
