The Cas9 enzyme, key to the CRISPR system, has a lot of fat that it can lose and still serve as a powerful tool.

Meletios Verras/

CRISPR is too fat for many therapies, so scientists are putting the genome editor on a diet

COLD SPRING HARBOR, NEW YORK—The genome editor CRISPR has morphed over the past 6 years from an obscure bacterial immune mechanism into the rock star tool of biology, allowing researchers to alter DNA with greater precision and ease than ever before. But the most popular version of CRISPR is simply too big, which complicates reaching some targets—and limits the ability of this powerful technology to create new therapies. Now, researchers have devised a way to put CRISPR on a diet and still retain its core functions.

Standard CRISPR methods have appropriated a DNA-snipping protein called SpCas9 from the Streptococcus pyogenes bacterium. Another CRISPR component guides the enzyme to targeted places on the genome. SpCas9 binds the DNA and its molecular scissors clip the double-stranded helix. But this lab darling, which has 1368 amino acids, is too chunky for many biomedical applications. So a team led by David Savage of the University of California, Berkeley, has devised a huge library of slimmer Cas9s using a “directed evolution” scheme.

“This is an amazing story because it’s a reversal of the actual evolutionary process,” says Kira Makarova, a pioneering CRISPR researcher at the National Center for Biotechnology Information in Bethesda, Maryland, who was not involved in the new work.

Savage, a structural biologist who presented his group’s work last week at the annual Cold Spring Harbor Laboratory CRISPR meeting here, calls the protein engineering method Minimization by Iterative Size-Exclusion Recombination (MISER). The technique uses two enzymes to systematically snip the DNA of the SpCas9 gene, pulling out chunks encoding different parts of the protein. Savage and colleagues then test those genetic sequences to see whether their resultant proteins still retain Cas9’s ability to bind to DNA targets. They then combine the ones that succeed, to add to the unique truncated options. So far, they have made half a million variants. “Shockingly, it works really well,” Savage says. “I didn’t expect it to be so flexible that it could tolerate enormous deletions and those could be stacked together.”

The MISER mutants won’t necessarily be able do everything that the typical CRISPR-Cas9 system can. One handicap is that some of the mutant Cas9s can lock onto an exact spot in the genome but cannot cut the DNA. But researchers earlier found that these “dead” Cas9s are handy tools, too, as they can ferry other molecules to specific destinations; one particularly powerful CRISPR technology called base editing exploits this to shuttle an enzyme to a target site that can convert one DNA base into another. The smallest MISER Cas9 mutant created to date—which can’t cut—has only 880 amino acids, about two-thirds the size of the original SpCas9.

Harvard University chemist David Liu, whose lab invented the base editor system, says Savage’s work with MISER is an “an outstanding early application of this exciting new method—and moves the genome editing field closer to a long-standing goal.”

Many investigators using CRISPR to design biomedical treatments package the genes for Cas9 and its other component inside a harmless virus that can shuttle them to specific cells to repair genetic defects. But the viruses have a limit to how much genetic cargo they can carry, and that’s where the skinny Cas9 could help tremendously—especially if its scissors work. “We have to finish this story,” says Savage, whose team is now sifting through its creations to find out which ones get the biggest bang for the smallest size.