site-specific recombination

Homologous recombination needs long matching sequences and can happen almost anywhere they line up. Site-specific recombination is the precise opposite in spirit: it is a controlled cut-and-rejoin that happens only at short, defined DNA sequences recognised by a dedicated enzyme. It is less like a general repair process and more like a programmable splice between two exact addresses in the genome.

A specialised enzyme, a site-specific recombinase, recognises a particular short sequence — its recognition site — wherever it appears, binds two such sites, cuts the DNA backbone cleanly at both, swaps the ends, and reseals them, all without losing or gaining a single base. The outcome depends on how the two sites are arranged: sites pointing the same way on one molecule cause the DNA between them to be looped out and deleted, sites pointing oppositely cause that stretch to be flipped (inverted), and sites on two separate DNA molecules cause them to be fused. The classic Cre-lox and FLP-FRT systems work exactly this way.

Cells and viruses use this for jobs that demand precision. Bacteriophage lambda integrates its genome neatly into a single chosen spot in the host chromosome by site-specific recombination, and some bacteria flip a piece of DNA back and forth to switch genes on or off. The same machinery has become a workhorse of the laboratory: by placing recognition sites around a gene, researchers can delete, invert, or activate it in a chosen tissue at a chosen time — a level of surgical control over the genome that ordinary recombination cannot provide.