Skip to Main Content

It’s the go-to phrase for biologists who know more than they’re telling. Ever since James Watson and Francis Crick ended their 1953 paper on the double helix by coyly saying “it has not escaped our notice” that the discovery might explain how DNA works as the molecule of heredity, other scientists have slipped that clause into papers and turned out to be just as prescient.

In a 2017 study, for instance, four biologists wrote that “it has not escaped our notice” that a funny little “jumping gene” might be harnessed for precision genome editing, giving classic CRISPR a hand at something it struggles to do: insert a string of healthy DNA in place of a disease-causing sequence, which for some genetic diseases might be the only path to a true cure.

The lead author of that study has been trying ever since to repurpose the jumping gene, but he (and other labs) was beaten to the punch on Thursday by CRISPR pioneer Feng Zhang of MIT and the Broad Institute. In a paper in Science that zoomed from submission to acceptance in just 25 days (four months is more usual), Zhang and his colleagues describe turning a jumping gene — aka a transposon, or mobile genetic element — into a mini TaskRabbit gig worker: With an assist from CRISPR enzymes, it zips over to the part of the genome whose address it is given and delivers a package of DNA, pronto.


Zhang’s team did all this in lowly bacteria, but other genome-editing biologists said the system could very well work in human cells, too, especially for repairing a disease-causing gene. “I think it’s something that could be used therapeutically,” said reproductive biologist Shoukhrat Mitalipov of Oregon Health and Science University, who was the first scientist in the U.S. to use CRISPR in human embryos. (He didn’t create pregnancies with them.) “It could even be very important” for treating disease via genome editing, he added, because “it seems more efficient and precise [than classic CRISPR]. It’s only a first step, but it’s really encouraging.”

This particular transposon, called Tn7, was discovered decades ago in bacteria. In general, transposons are pieces of DNA that sit within a genome, but for reasons that are somewhat mysterious have the ability to cut themselves out of their original site and jump to another.


Tn7 uses the CRISPR enzyme Cas12 to lead it to its next home. Rather than cutting DNA, as Cas12 usually does, when paired with Tn7 it keeps its molecular scissors sheathed. Ever since the “it has not escaped our notice” hint in the 2017 paper, scientists have been looking for ways to control where the jumping gene jumps to. Constructing a new guide molecule should allow scientists to control where the DNA inserts itself in the genome.

That’s essentially what Zhang’s team accomplished. Starting with Tn7 from the bacteria Scytonema hofmanni, they created new guide molecules to lead it to a specified address in the genomes of E. coli bacteria and insert its package of DNA there. Compared to ordinary CRISPR’s 1 percent success rate of DNA insertion, the jumping gene system scored about 80 percent.

Crucially, the insertion didn’t require slicing the genome. Such “double-stranded DNA breaks” often trigger genomic havoc, with entire chunks of chromosomes lifted up and dropped into new DNA neighborhoods — something that, in pre-CRISPR gene therapy, caused cancer in some patients. Any genome-editing technology that avoids double-stranded breaks might therefore have a safety advantage.

“As a proof-of-principle feasibility study, it does a great job showing the potential of these systems and should go down as a very important paper,” said microbiologist Joseph Peters of Cornell University, lead author of the 2017 paper predicting that transposons might become part of the CRISPR toolbox. “At this point it isn’t totally clear if this specific system will be useful for genome modification,” since some characteristics of Tn7 might limit how well it works in organisms other than bacteria, but in general, Tn7 “is very exciting as a potential genome editor.”

Zhang calls the system “CRISPR-associated transposase,” or CAST. He and one of his co-authors, Jonathan Strecker, have filed for a patent on it.

The jumping-gene version of CRISPR is most likely to best the classic version when curing a genetic disease requires making a gene function normally by replacing its misspelled DNA “letters.” CRISPR tries to do that by cutting out the mutation (like Word snipping out fi from orthografi) and offering up the correct letters (phy). Unfortunately, DNA is as reluctant to open up and accept the substitution as a toddler is to open up for peas.

CAST seems to have no such problem inserting DNA. When the scientists programmed it to 48 targets in the E. coli genome, CAST hit 29 of them. (In general, CRISPR guides don’t work all the time.) It produced the desired edit, inserting DNA, in 80 percent of the bacteria, a rate that leaves classic CRISPR in the dust.

One red flag was that CAST inserted DNA into some places it wasn’t supposed to, an “off-target” problem that also plagues classic CRISPR. The 99-to-1 ratio of on-target to off-target edits seems high, but more research will be needed to show whether even a single off-target DNA change is enough to break, say, an essential tumor-suppressor gene.

Comments are closed.