CRISPR that makes its own couriers

Inject a CRISPR editor into a tissue and it only works in the cells it physically reaches. That ceiling is gene therapy's quiet bottleneck: in a mouse liver, a single dose touches maybe one cell in twenty, and the rest are never edited.

The goal isn't better delivery — it's making suboptimal delivery therapeutically sufficient by amplifying it from within the tissue.

Jennifer Doudna's lab inverted the logic. Instead of injecting pre-made editor particles, they inject one piece of DNA that turns the cells it does reach into temporary factories. Those cells assemble the Cas9 editor inside small lipid bubbles, push them out, and the bubbles fuse with the neighbors and edit them too. In mouse liver, roughly 5% of cells were directly reached, yet 11% ended up edited — against 4% for a standard editor dose. In cultured cells the edited population grew to nearly three times the cells that started the process.

The bubbles are deliberately neutered: non-viral, and built so they cannot copy themselves. The editing spreads exactly one cell-generation outward and then stops, by design — amplification without a runaway chain. That bounded spread is both the safety story and the limit.

The target was transthyretin amyloidosis, a heart and nerve disease already in CRISPR trials, and the treated mice dropped the disease protein to 60% of normal. It is a preprint in mice, not a therapy. But it points at the field's deepest constraint from a new direction: rather than delivering to more cells, make the few you reach do the delivering.