Simple explanation
Every editor described so far changes small amounts of text. Recombinases move paragraphs. They are enzymes that recognise a specific short sequence and swap the DNA around it — so if you first mark a safe spot in the genome, a recombinase can drop an entire working gene into that spot in one go.
Go deeper
Site-specific recombinases (Cre, Flp) and serine integrases (Bxb1, PhiC31) catalyse strand exchange between defined recognition sites, integrating, excising or inverting sequence without producing free double-strand breaks or depending on host repair pathways. Because natural recognition sites are rare in human genomes, therapeutic strategies pair a programmable editor that writes a landing site with an integrase that then inserts the cargo — the approach behind PASTE and related methods, which can integrate sequences of many kilobases.
Why large insertions matter
Some diseases cannot be treated by tweaking sequence because the gene is enormous or the mutations are scattered across it in hundreds of different places. Duchenne muscular dystrophy is the canonical example: the dystrophin gene is the largest in the human genome, and a per-mutation editing strategy would mean a different medicine for almost every family. Inserting one working copy at a defined location would be a single solution for everyone — which is why integrase approaches attract so much interest despite being preclinical.
Sources
- Nature Biotechnology (Yarnall et al.) · 2023
Drag-and-drop genome insertion of large sequences without double-strand DNA cleavage using CRISPR-directed integrases ↗