sciencebriefs
13:00in productionCh. 1 · Sequences noticed before they were understood/ 13:00 · ceiling 15 min
Genetics

CRISPR

1987

A bacterial defence system that remembers past viral infections turned out, once its guide-RNA targeting was simplified in 2012, to be programmable enough to cut DNA anywhere a scientist chose.

CRISPR sequences were first noticed in E. coli DNA in 1987, though their purpose was not understood then; by 2007, experiments had shown they formed part of a genuine adaptive immune system, letting bacteria record fragments of viral DNA from past infections and use that record to recognise and destroy the same virus later. The mechanism relies on a protein, Cas9, guided to a matching DNA sequence by RNA molecules and a short adjacent marker called a PAM sequence, after which Cas9 cuts the DNA at that precise point. In 2012, Jennifer Doudna and Emmanuelle Charpentier's teams, working with Martin Jinek and Samuel Sternberg, showed that the two separate guide RNAs bacteria use could be fused into a single, synthetic guide RNA, making the targeting fully programmable rather than fixed to whatever sequence a bacterium happened to have stored. That simplification turned a bacterial immune mechanism into a general-purpose gene-editing tool, work recognised with the 2020 Nobel Prize in Chemistry, and one that has since moved into approved medical treatments as well as contested territory, including the case of edited human embryos.

Chapters & takeaways6
  1. 0:08
    Sequences noticed before they were understood

    CRISPR repeats were first seen in bacterial DNA in 1987, decades before their function as an immune system was established.

  2. 2:10
    A bacterium's memory of past infections

    By 2007, experiments confirmed bacteria store fragments of viral DNA to recognise and destroy the same virus on a repeat encounter.

  3. 4:20
    Guide RNA, PAM, and a precise cut

    Cas9 is directed to a specific DNA sequence by guide RNA and a short adjacent marker, then cuts the DNA at that exact location.

  4. 6:30
    Fusing two RNAs into one

    The 2012 innovation combined the bacterium's two natural guide RNAs into a single synthetic guide, making the target sequence fully programmable.

  5. 8:40
    From bacterial defence to lab tool

    The programmable system was quickly applied to editing genes in human cells and other organisms, not just bacteria.

  6. 10:50
    Therapies, and a line crossed

    CRISPR-based treatments have since reached approved clinical use, even as the He Jiankui embryo-editing case showed how far ahead application can run of ethical consensus.

Worth your time?

Yes. Study the whole thing.

4.5/ 5
What works
  • the twenty-five-year gap between first noticing CRISPR sequences and understanding their function is stated plainly, showing how slowly this kind of discovery can unfold
  • the single technical step, fusing two RNAs into one guide, is identified as the specific thing that made the system programmable
  • it connects the science directly to both an approved 2025 therapy and an unresolved ethical controversy without overstating either
What does not
  • the PAM sequence's role is named but not explained in terms of why it is necessary for targeting to work
  • the respective contributions of Jinek and Sternberg alongside Doudna and Charpentier are not distinguished in detail
Study it if
  • anyone who wants to understand what CRISPR actually is beyond the headline phrase gene editing
  • readers interested in how a natural bacterial mechanism became a laboratory tool through one specific technical simplification
  • people following the ethical debate over germline editing who want the underlying science first
Skip it if
  • readers wanting the structural biology of Cas9's nuclease domains explained at a molecular level
  • anyone looking primarily for the He Jiankui controversy rather than the science that made it possible
The written brief4 min read

Sequences noticed before they were understood

CRISPR sequences, short repeated stretches of DNA interspersed with unique spacer sequences, were first noticed in the genome of the bacterium E. coli in 1987, though at the time no one understood what function they served. It took two more decades of accumulating evidence before researchers established, by 2007, that these sequences form part of a genuine adaptive immune system in bacteria and archaea: when a virus infects a bacterium and the bacterium survives, it can capture a fragment of the virus’s DNA and insert it into its own CRISPR locus as a spacer, creating a stored record of that infection. If the same virus, or a closely related one, attacks again later, the bacterium can use that stored sequence to recognise and destroy the invader, a form of heritable immune memory built directly into the genome.

A bacterium’s memory of past infections

The molecular machinery that carries this out centres on a protein called Cas9, which is directed to a specific DNA sequence by RNA molecules that match the stored viral spacer, combined with recognition of a short adjacent DNA motif called a PAM sequence that helps the system confirm it has found a genuine target rather than cutting DNA indiscriminately. Once Cas9 locates a matching sequence next to the correct PAM, it cuts both strands of the DNA at that precise location, and the cell’s own repair processes then take over, in bacteria typically destroying the invading viral DNA in the process. This targeting system, evolved purely as bacterial defence, turned out to have a structure general enough that it could in principle be redirected toward any DNA sequence a researcher chose, given the right guide.

Guide RNA, PAM, and a precise cut

The pivotal step that made this redirection practical came in 2012, when Jennifer Doudna and Emmanuelle Charpentier’s research groups, working with Martin Jinek and Samuel Sternberg, showed that the two separate RNA molecules bacteria naturally use to guide Cas9 could be fused into a single synthetic guide RNA. That simplification meant researchers no longer needed to work within whatever sequence a bacterium’s own immune history happened to have recorded; instead, they could design a guide RNA to match essentially any DNA sequence of interest and expect Cas9 to cut there. The finding held up immediately and dramatically: within about a year, CRISPR-Cas9 editing had been demonstrated working in human cells, extending far beyond the bacterial context the system had evolved in.

Fusing two RNAs into one

What the 2012 breakthrough did not resolve, and what subsequent research has had to address separately, are questions of precision and safety in practical use. Off-target effects, in which Cas9 cuts DNA at unintended locations resembling but not matching the designed guide sequence, remain a real concern that researchers continue working to reduce through improved guide design and modified Cas9 variants. The technology’s rapid spread into human applications also outran, in at least one well-documented case, any settled ethical or regulatory consensus: in 2018, He Jiankui announced he had used CRISPR to edit the genomes of human embryos that were later born as twins, an act of germline editing that drew widespread condemnation from the scientific community for proceeding without adequate safety evidence, oversight, or informed consent.

From bacterial defence to lab tool

The broader significance of the 2012 targeting breakthrough is that it converted a narrow, bacteria-specific immune mechanism into arguably the most widely used gene-editing tool in modern biology, applied across basic research, agriculture, and medicine. That reach has continued to expand: by 2025, a CRISPR-based therapy had received approval for treating a metabolic disorder in a child, marking a concrete clinical application built directly on the programmable targeting system Doudna, Charpentier, Jinek, and Sternberg’s 2012 work established. Doudna and Charpentier were jointly recognised for the achievement with the 2020 Nobel Prize in Chemistry, and Doudna herself has since taken an active public role in bioethics debates, generally supporting gene editing in body cells for treating disease while opposing germline editing without far stronger safeguards than currently exist.

Therapies, and a line crossed

This is worth real time because it lays out, in a way headlines rarely do, the specific technical step, fusing two RNA molecules into one programmable guide, that separates a naturally occurring bacterial defence mechanism from a general laboratory tool capable of editing any organism’s genome. Understanding that distinction makes both the excitement around CRISPR therapies and the alarm around cases like He Jiankui’s easier to evaluate on their actual merits rather than as an undifferentiated blur of gene editing headlines. Readers wanting deep molecular structural detail on Cas9 itself will need a more technical source, but for grasping what CRISPR actually does and why it was worth a Nobel Prize, this is a solid and well-earned use of an hour.

Same field · Genetics4 of 57
Up next in Science

RNA splicing

· 13:00

In 1977 Phillip Sharp and Richard Roberts, examining adenovirus RNA under the electron microscope, found that genes are not continuous stretches of code but are broken up by non-coding introns removed during splicing.

13:00