The founders of CRISPR-Cas9 startup Editas Medicine have gotten a jump on the next version of CRISPR gene editing. Scientists from Harvard University, Massachusetts General Hospital, and the Broad Institute are behind Beam Therapeutics, which today announced a $87 million Series A round to turn a more precise version of CRISPR into human medicine.
The advance is called base editing, and it is aimed at fixing tiny genomic mistakes called single-point mutations. Imagine you want to fix a single-letter error in this paragraph. (Perhaps I have spelled peragraph wrong.) Unfortunately, your editing tools only allow you to insert or cut out several words. Seems like a blunt tool, right?
A similar problem with gene editing has pointed many researchers toward base editing, even as earlier work has fueled the first generation of CRISPR-Cas9 based medicines that companies like Editas (NASDAQ: EDIT), Intellia Therapeutics (NASDAQ: NLTA), and CRISPR Therapeutics (NASDAQ: CRSP) have, over the last four or five years, brought to the cusp of clinical trials.
Base editing zeroes in on the individual “letters” of the genetic code, called bases, represented by C, T, G, and A. The technology can replace one base with another in the genome. That’s important because there are more than 30,000 different point mutations, or single-letter typos, known to be linked to disease. Beam’s founders, including David Liu of Harvard, say that their technology can convert a faulty letter into the correct one, and transform a disease-causing gene into a normal one.
That’s something CRISPR-Cas9 can’t do; it cuts DNA and inserts or removes larger stretches of DNA.
“We want to turn a mutant letter back to normal with no other effect on the genome,” says Liu, whose lab pioneered the CRISPR base-editing technology for DNA. “That’s the ideal outcome.”
Beyond Beam and its founders, others are working on base editing, too. “It’s a paradigm-shifting technology,” says Branden Moriarty, whose lab at the University of Minnesota’s Center for Genome Engineering is focused on pediatric cancer and gene therapy. He and colleagues are studying base editing as a potential modification of human cells to use as therapeutics. They also have released a free software program for researchers to check their base editing work.
Beam CEO John Evans, formerly a business executive at Agios Therapeutics (NASDAQ: AGIO), won’t say when Beam will get to clinical trials. About a year old, the company has done lab work on diseases caused by point mutations like sickle cell anemia. The Series A round is from Arch Venture Partners, where Evans has been parked since leaving Agios in 2017, and F-Prime Capital Partners. That cash will also let Beam be ambitious. Evans declined to discuss specific diseases but says Beam is working on roughly a dozen different programs for genetic diseases across multiple tissue types.
Beam’s base editing system borrows basic parts from CRISPR-Cas9, but it’s a different beast. CRISPR-Cas9 naturally occurs in bacteria as a defense system against invading viruses. It uses molecular scissors—the Cas9 enzyme—to snip through both strands of DNA and disable a gene. The scissors are guided by a string of RNA code engineered to match the DNA target site. (A more complicated version of CRISPR-Cas9 also provides a piece of DNA to replace a mutated section; it is not nearly as far along in development as a human therapeutic.)
Even when CRISPR-Cas9 cuts in the right place, it does a sloppy job, leaving the cell’s natural machinery to stitch the two snipped ends back together. This repair effectively disables the targeted gene, but it also leaves open the chance for a bungled repair that creates new mutations. How dangerous they are remains a matter of debate in the field.
With base editing, the Cas is guided to the site, but it doesn’t snip. (It’s called deactivated Cas; some like to call it dead Cas.) It still binds to the targeted DNA, however, and when it grabs hold, the two strands of the DNA unravel. One strand forms a tiny loop that sticks out, like a bit of rope that has worked free from a braid.
Liu says this single strand of DNA is the key to the base editing system now under license to Beam. In his lab, post-doctoral fellow Nicole Gaudelli, who now works full-time for Beam, created a new version of an enzyme, cytidine deaminase. It converts C to T, and the cell then finishes the job, converting the paired G to an A. The key here is the new enzyme only works on single-stranded DNA. It leaves double-stranded DNA alone. (Liu’s lab has also produced a separate editor that converts A to G that did not require engineering a new enzyme.)
The loop of single-stranded DNA should also make for even more precision. At only five bases long, it offers an extremely narrow editing window. But it doesn’t mean base editing is fail-safe. In a recent paper that has not yet been peer reviewed, Beam cofounder and Massachusetts General Hospital researcher Keith Joung (also an Editas founder) acknowledges that existing base editors can still edit the wrong C if more than one C is present within the 5-base window, creating potentially harmful mutations. Imagine trying to change the misspelled cacch to catch but not knowing if you’ve created catch, tacch, or cacth.
In the same paper, Joung describes a base editor that converts C to T only when the C is part of a specific three-letter sequence. It seems to further reduce the chances of unwanted mutations and is “an important proof of principle” that should encourage the development of better base editors, says Joung.
When asked if Joung’s work could allay fears of editing the wrong letter, Andy May, senior director of genetic engineering at the Chan-Zuckerberg Biohub in San Francisco, says the first data “suggest it might work” but cautions that these are early days in the field.
May says there are other aspects of base editing to keep an eye on. The enzyme that converts C to T normally plays a specific role in the human immune system, creating diversity in the antibodies that patrol our bodies for invaders. There’s no evidence that using an engineered form of the enzyme to alter a range of cell types will cause problems, says May, “but what it does to the overall health and stability of those cells should be monitored.”
How to deliver these next-generation gene editors into patients’ cells remains an open question as well. No one has yet mastered the problem with the first CRISPR generation; the Cas enzymes and the rest of the machinery are, molecularly speaking, a lot to squeeze into a cell. Beam’s founders believe the company can turn to the work of outside labs for help. “There are hundreds of labs around the world dedicated to advancing delivery,” says Liu. “I’m optimistic there will be quite a few programs with well-matched solutions.” CEO Evans believes that Beam’s goal—treatments that only need to be administered once—will lessen the technological challenge.
Beam has also licensed technology from the Broad Institute lab of Feng Zhang, another Beam cofounder and a member of its board of directors. Beam has the rights to use for human therapeutics a CRISPR system from Zhang’s lab that uses the Cas13 enzyme instead of Cas9 to swap out bases in RNA instead of DNA.
Some of the technologies that Beam is using were previously licensed to Editas. Beam will sublicense some of those rights from Editas, and in exchange, Editas has taken an equity stake in the startup.