New CRISPR-based screening method improves gene editors

Gene editing methods often involve breaking a strand of DNA in order to make specific changes to the sequence. They then rely on the cell’s DNA repair pathways to mend the break. These cellular repair pathways, however, are not completely understood, and introduce an element of chance to gene editing; for example, a repair mechanism may patch up the edited strand, but also leave behind an unwanted mutation. 

In a paper published online in Cell on October 20, 2021, a collaborative team of researchers in the lab of Whitehead Institute Member, Massachusetts Institute of Technology (MIT) biology professor and Howard Hughes Medical Institute Investigator Jonathan Weissman, the lab of Britt Adamson at Princeton University, and Cecilia Cotta-Ramusino, then a research scientist at Editas Medicine (now at Tessera Therapeutics), present a new experimental method that could help bridge this gap in our understanding. 

The method, called Repair-seq, allows researchers to find out which genes and genetic pathways are involved in DNA repair mechanisms. It provides a useful tool for fundamental research on gene repair, as well as a way to test the action of new genome editing methods as they are developed, says Weissman Lab postdoc and first author Jeffrey Hussmann. A companion paper published concurrently in Cell in collaboration with researchers at the Broad Institute of MIT and Harvard provides a glimpse into the utility of Repair-seq when applied to new gene editing technologies. 

“The field of gene editing has moved so quickly and people have been so creative developing new methods that our ability to apply them has dramatically outpaced our understanding of exactly how they work,” Hussmann said. “We think that Repair-seq will be a valuable tool going forward so that as new editing methods are developed, we can quickly do a better job of characterizing how they interact with different repair mechanisms.”

Studying repair mechanisms in one fell swoop

Cells have several different methods they use to repair breaks in DNA strands, and the path to any one method depends on a tree of decisions based on the circumstances. “Over decades, a huge number of people have worked out parts of these pathways  through focused experiments,” said Adamson, a senior author on both papers and an assistant professor at Princeton University. 

Together, the team of researchers saw an opportunity to harness existing CRISPR-based methods to take a broad look at repair pathways in the cell. The method they created combines several CRISPR-based technologies. First, the researchers used a method they previously developed called CRISPRi to inactivate hundreds of genes known to be involved in DNA repair across a cell population. Then, they induced double strand breaks in the cells’ DNA at specific places that the cell would need to heal. 

As the cells mended the breaks, the researchers used targeted sequencing to examine the ‘repair outcomes’ — mutations or the lack thereof — in the DNA strand resulting from different methods of repair. Finally, they were able to extrapolate which genes were essential to various repair mechanisms and how they were involved in producing or preventing each type of resulting mutation. They also posted their data online in an interactive format so others can use it to investigate DNA repair genes and pathways. 

“This combination of different CRISPR-based technologies has made it possible to, in one fell swoop, recapitulate a lot of the work that was done painstakingly over the past decades to study each repair pathway one at a time,” Hussmann said. “The high-level view of repair that our method produces shows us many of the things that people saw before, and at the same time reveals unexpected connections that we only get by having the comprehensive picture.”

These unforeseen relationships between repair genes may help fundamental researchers refine the decision tree of double strand break repair in the future, said Weissman. “One of the big themes that's come out of this is that outcomes that superficially look similar can actually have very different mechanisms,” Weissman said. 

A ‘prime’ example of Repair-seq’s utility 

As the team was developing their Repair-seq methodology, Broad Institute of MIT and Harvard Core Member David Liu’s lab was working on prime editing, a gene editing method that promises more precise control over genetic outcomes than traditional CRISPR methods. Instead of snipping both strands of DNA’s double helix, prime editing makes a ‘nick’ in only one of the strands and introduces a short sequence template containing the desired genetic change. 

“When Liu’s group came out with prime editing, our Repair-seq team realized that we had the perfect tool for quickly trying to understand exactly how it was working,” Hussmann said.

The three labs collaborated to use Repair-seq to identify which pathways were at play during the installation of mutations by prime editing, and identified one in particular, called the DNA mismatch repair pathway, that seemed to be interfering with the efficiency and accuracy of the method. When the researchers inhibited this pathway, the performance of Liu’s prime editing technique greatly improved. 

“Working with Britt, Jonathan, and their labs has been a beautiful integration of basic science, tool application, and technology development—a real testament to the power of multidisciplinary collaboration” said Liu, also an Investigator of the Howard Hughes Medical Institute. 

The researchers also applied Repair-seq to a base editor — a tool to swap specific bases in a DNA sequence — and were able to illuminate the DNA repair genes involved in swapping in particular base. 

In the future, the researchers plan to continue adapting the method to new sequencing methods and applying it to new editors as they are developed. “We think Repair-seq is a really practical way of making better genome editors,” Weissman said. 

 "It has been rewarding to see the efforts of our collaboration come together,” said Adamson. "We hope the insights from our study and tools that those insights have led to will be widely useful to the research community."