Three years ago, Stanley Qi took CRISPR—the world’s most versatile gene-editing tool—and stopped it from editing any genes.

CRISPR is a young and ferociously hyped technique that allows scientists to easily and precisely tweak almost any gene they want, opening up experiments that were once unfeasible or impractical, and triggering a new round of ethical debates about messing with the human genome. During these debates, CRISPR has become almost synonymous with the editing of genes. But gene-editing might not be its most promising use.

The technique relies on two components: an enzyme called Cas9 that cuts DNA like a pair of scissors, and a guide molecule that directs Cas9 to a specific target like a genetic GPS system. Qi, now at Stanford University, found a way of blunting the scissors, creating a “dead” version of Cas9 that can’t cut anything at all.

This seems perverse, but it’s actually quite brilliant. The dead enzyme can now act as a platform for other molecules, including activator molecules that switch genes on, repressors that turn them off, or glowing substances that reveal their locations. And with the right guide molecules, scientists can now direct these payloads to any gene they like.

Now, instead of a precise and versatile set of scissors, which can cut any gene you want, you have a precise and versatile delivery system, which can control any gene you want. You don’t just have an editor. You have a stimulant, a muzzle, a dimmer switch, a tracker.

This matters because much of biology depends on how genes are used, rather than the sequences of those genes. Think of the genome as a the script of a play: The same text can lead to vastly different productions depending on how lines are delivered, how sets are constructed, or how stage directions are interpreted. Likewise, we can use exactly the same sets of DNA to sculpt a muscle cell, a neuron, or a skin cell. By using CRISPR to finely control the activity of specific genes, we can better understand how our bodies do so naturally.

Scientists could turn on genes that cause heart muscles to expand after a heart attack, or silence genes that fuel the growth of cancers. “Or let’s say you’ve been exposed to a virus,” says Jonathan Weissman from the University of California, San Francisco. Viruses typically begin their invasions by latching onto receptor molecules on our cells, and we know the genes that make many of these receptors. “Turn those off, and now you’re immune to the virus. Your immune system can clear it. Then you turn the gene back on and you’re back to normal.”

These aren’t new concepts; scientists have long tried to perform similar feats using other tools. CRISPR just makes things easier. Its potential was clear right from the start, when Jennifer Doudna and Emmanuelle Charpentier showed that they could use specific guide molecules to point the snip-happy Cas9 scissors at a specific target. “We immediately thought: Well, let’s just break the scissors,” says Weissman.

By the time Doudna and Charpentier published their now-classic 2012 paper detailing CRISPR’s potential as a gene editor, Weissman, Qi, and their colleagues (Doudna included) had already developed the dead Cas9 and were racing to find ways of using it. While the world was chatting about editing, they were working on control.

Their speed is a testament to the value of basic research, Weissman says. Others had studied the structure of scissor proteins like Cas9, so the team already knew exactly what changes to make to blunt the enzyme. Others had studied enzymes that turn genes on or off, so the team could easily repurpose the relevant parts of these enzymes onto their own tools. “Something like CRISPR explodes because there’s all this work that was done beforehand,” says Weissman.

The team developed ways of using the blunted enzyme to switch genes off (CRISPRi, where the i stands for interference) or on (CRISPRa, where the a stands for activation), or to tune their activity over a 1,000-fold range. They used these techniques to quickly and thoroughly screen human cells for genes that they need to grow, or to deal with a bacterial toxin. They also affixed the dead Cas9 with a glowing molecule, so they could track the locations of specific genes and film them as they move about living cells.  

Most recently, they have started to write more complex “programs” in which several genes are simultaneously activated and others are repressed. George Church, another CRISPR pioneer, used similar programs to transform stem cells into neurons. And Wendell Lim, who was involved in developing CRISPRi and CRISPRa, wants to program customized cells, such as immune cells that attack cancers.

“I hope that the dead-Cas9 platform can one day partially replace the drugs that people have developed for treating cancers or other diseases,” says Qi. Many of these drugs are designed to block specific genes, but the dead Cas9 could do so more accurately, while also targeting many genes at once. “It’s equivalent to using a cocktail of drugs, but with much better specificity.”

As always, there are limits. “We still don’t have a good way of getting [the dead Cas9] into cells,” says Weissman. They don't have a good way of doing that regularly, either—and they need one. When CRISPR edits a gene, the change is permanent and only needs to be done once. But to turn a gene on or off, the dead Cas9 needs to be present. And that might be difficult since Cas9 comes from bacteria; if it stays in the body for too long, it might trigger a nasty immune response.

Then again, these transient changes would side-step many of the ethical debates around CRISPR, since they don’t involve permanent edits to the human genome that would affect future generations. “If we can get it to work, it’s fundamentally safer and better than cutting your DNA,” says Weissman.