CRISPR was a bacterial immune system before it was a tool
Scientists had noticed those odd repeated sequences in bacterial DNA for years. Nobody knew what they were for.
Clustered regularly interspaced short palindromic repeats—CRISPR—were first observed in bacterial genomes in 1987, when a Japanese researcher noticed a peculiar pattern near a gene in E. coli: identical sequences of DNA separated by unique spacers. The observation was published, and mostly ignored. It took fifteen more years before anyone understood what those repeats actually did.
They were memory. When a bacterium survived a viral attack, it cut out a fragment of the virus's DNA and stored it between the repeats. The next time that virus appeared, the bacterium produced RNA matching the stored fragment, used it to locate the viral DNA, and cut it apart with a protein called Cas9. Bacteria had been running a rudimentary immune system—one that remembered its enemies by sequence and could find them again.
In June 2012, Jennifer Doudna, Emmanuelle Charpentier, and colleagues published a paper in Science showing that Cas9 could be programmed with a single engineered guide RNA to cut any DNA sequence of interest. The two-RNA system bacteria used naturally could be simplified into one synthetic molecule pointing at any target in any genome. What bacteria had been doing to fight off viruses for billions of years, researchers could now do deliberately—to rewrite the code of any living thing.
Doudna and Charpentier shared the Nobel Prize in Chemistry in 2020.