A recently patented genome editing tool called PASTE holds genuine promise for expanding the universe of treatable genetic diseases. The approach combines elements of CRISPR and prime editing with a pair of enzymes designed to enable the integration of large segments of DNA without incurring double-stranded DNA breaks.
U.S. Patent No. 11,572,556, assigned to MIT, covers systems, methods, and compositions for programmable addition via site-specific targeting elements (PASTE). The patent describes site-specific integration of a nucleic acid into a genome, using a CRISPR–Cas9 nickase fused to a reverse transcriptase (RT) and a serine integrase. These enzymes target specific genome sequences known as attachment sites, binding to them before integrating their DNA payload.
PASTE can insert DNA fragments as large as 50,000 base pairs, which puts it on a different plane compared to other genome editing tools such as prime editing.
The credited inventors are Omar Abudayyeh, PhD, and Jonathan Gootenberg, PhD, two McGovern fellows at MIT’s McGovern Institute for Brain Research. The duo, who were both graduate students with CRISPR pioneer Feng Zhang, PhD, at the Broad Institute before moving across the road to the McGovern Institute in Kendall Square, Cambridge, MA, first detailed PASTE in a preprint posted on bioRxiv in 2021. The peer-reviewed paper, “Drag-and-drop genome insertion of large sequences without double-strand DNA cleavage using CRISPR-directed integrases,” was published in November 2022 in Nature Biotechnology.
PASTE entails the engineering of Cas9, RT, and integrase linkers to create a fusion protein capable of efficient integration (5–50%) of diverse cargos at precisely defined target locations within the human genome with small, stereotyped scars that can serve as protein linkers.
The serine integrases used in PASTE can insert DNA sequences as large as 50,000 base pairs by targeting specific “attachment” sites within the genome.
“It has been very difficult for the [CRISPR] field to really put large edits in the genome without relying on things like homology-directed repair,” Gootenberg told GEN Edge. “The concept of PASTE is actually quite simple: Instead of being able to try to do everything all at once in one conceptual aspect—and there are some technologies that approach that—we thought it would be easier to take something that can insert very efficiently into a constant sequence like an integrase, and then put the constant sequence into the genome.”
“It’s a two-step process where we first insert the constant sequence, which is a small letter, and it’s easier to do. Then we use that constant sequence to put in a larger sequence. That’s the concept in a nutshell,” Gootenberg added.
The $64-million question
Jacob S. Sherkow, JD, a professor at the University of Illinois College of Law and the Carle Illinois College of Medicine, and an authority on the long-standing CRISPR-Cas9 patent standoff, told GEN Edge that PASTE appears to address a key challenge in genome editing: How to achieve high fidelity site-specific insertion of an exogenous target sequence?
Citing the three-prong standard for patentability—novelty, utility, and non-obviousness—Sherkow said MIT and the PASTE inventors can likely show their technology is novel and useful. But it is less clear if they can show non-obviousness should the PASTE patent be challenged someday.
“That is the $64-million question,” said Sherkow. “To the extent that there’s patent litigation, I assure you that on the PASTE patent, at least some of those arguments are going to turn on non-obviousness.”
“You’ve got the older Liu prime editing patent that claims prime editing. Then you’ve got [the] PASTE patent filed later, which claims 80% of what the prime editing patent claims, plus the addition of using an interface and an RT, to insert a large piece of DNA in a site-specific site. That is not an overlap issue at all, such that the patent office is going to be asked to, or have the authority to, cancel one patent in favor of the other,” Sherkow said.
“Where the conflict is going to arise is not on the validity of these two patents. It’s going to be on who is going to pay for them, and who is going to use which technology over another technology,” Sherkow added.
PASTE and prime are among numerous technologies that have developed in the emerging genome editing field as researchers and startups pursue curative therapies that make changes to the genome without making double-stranded breaks, a known downside of traditional CRISPR-Cas9 gene editing that can be harmful to cells. Last year, Intellia Therapeutics shelled out up to $200 million to buy University of California, Berkeley, startup Rewrite Therapeutics and its DNA writing technology.
Patentability is among the key challenges for genome editing. For example, Tessera Therapeutics is pursuing a patent for its RNA gene writers technology, which uses a mechanism called target primed reverse transcription to write genes into the genome. The method involves four steps—binding RNA, binding DNA, nicking DNA, and priming reverse transcription. The company is among several whose intellectual property has been challenged in a recent STAT article as “bear[ing] strong similarity to prime editing—even if they use different terminology.”
PASTE editing could potentially treat diseases caused by genes harboring a large number of mutations, such as Leber’s congenital amaurosis or cystic fibrosis, where gene editing systems would have to be tailored for specific mutations and each subset of a disease population. In those diseases, programmable insertion of a wild-type gene could address most potential mutations and serve as a blanket therapy.
“Complete replacement of genes at their natural sites can be contemplated instead of creating variant-specific treatments,” Muhammad Arslan Mahmood, MPhil, and Shahid Mansoor, PhD, both of Pakistan’s National Institute for Biotechnology and Genetic Engineering, observed in a recent commentary on the PASTE report, published in The CRISPR Journal (a sister journal of GEN).
“These exciting results enhance the versatility of the CRISPR-based gene editing along with the LSRs [large serine recombinases] that provide the opportunity for genome engineering, combinatory screening of bulky DNA libraries aiding to spread these applications for treating the diseases in both animals and plants, which are caused by defective genes,” the authors concluded.
The Cystic Fibrosis Foundation is among the entities that have funded PASTE editing research, through a Pioneer Award designed to fund “ambitious basic research projects aiming to utilize cutting-edge techniques and strategies that have the potential to discover new genetic-based therapies for cystic fibrosis.”
“We’ve been working on actually being able to insert the genes at the locus, and things are looking promising with that. We’ll have more to share about that as things progress. I would say that’s one of the main funded applications we’re looking at right now,” Gootenberg said.
Carrying out genome editing without the need for double-stranded breaks in DNA was first laid out by David Liu, PhD, and colleagues at the Broad Institute of MIT and Harvard. In a pair of papers in Nature published between 2015–16, Liu’s lab developed base editing, a technique that can make certain classes of single-base substitutions without cleaving the double helix (unlike CRISPR-), two former postdocs in Liu’s lab, Alexis Komor, PhD (now of University of California, San Diego) and Nicole Gaudelli, PhD (now of Beam Therapeutics) recalled in a 2020 episode of The CRISPR Journal’s podcast series GuidePost, available via SoundCloud and Spotify.
Later, in a 2019 paper published in Nature, Andrew Anzalone, MD, PhD, and colleagues in Liu’s lab described a “search-and-replace” genome editing technology called prime editing. That approach supplied a desired edit in an extension to the guide RNA, which is then converted to DNA using the RT enzyme. The technology can introduce targeted insertions, deletions, and all 12 possible base-to-base substitutions.
Liu told GEN at the time that of the roughly 75,000 cataloged pathogenic mutations in human genetic diseases, prime editing had the versatility and potential to correct the majority (89%) of them.
Liu’s 2019 paper showed prime editing being used to install an integrase/recombinase landing site into a target DNA site. Earlier, in March of that year, Liu and colleagues filed for a U.S. patent that was granted in September 2022 (No. 11,447,770), which described several examples using prime editing to install integrase/recombinase landing sites, followed by targeted gene integration using an integrase/recombinase enzyme.
And in a 2021 paper published in Nature Biotechnology, Liu and co-authors reported the use of prime editing to install an integrase/recombinase landing site into a target DNA site, followed by an integrase/recombinase to catalyze integration of cargo DNA into that landing site.
Prime vs. PASTE
Speaking with GEN Edge, Liu asserted that one distinction between PASTE and Prime’s PASSIGE™ (prime-assisted site-specific integrase gene editing) approach is that PASTE fuses the prime editor and the integrase enzyme into a single protein chain, while PASSIGE typically uses them as two separate proteins.
“We have compared, side-by-side, fused and unfused prime editors plus integrases at several different target sites in human cells, and we have never observed any benefit from fusing the prime editor and integrase,” said Liu, who is also the director of the Merkin Institute of Transformative Technologies in Healthcare and a Howard Hughes Medical Institute investigator.
“Indeed, for most of the target sites and integrase enzyme combinations we’ve tested in human cells, we observed that the fused prime editor–integrase proteins as reported in the PASTE paper substantially underperform the separated prime editor and integrase proteins as used in PASSIGE,” Liu said. “That separate prime editor + integrase proteins perform better than fusing the two proteins together makes sense scientifically, because the prime editor must vacate the target landing site before the integrase enzyme can perform cargo DNA integration.”
Liu said fusing the two proteins together increases the chance that the target landing site is blocked by the prime editing protein and/or associated prime editing guide RNA (pegRNA), “because the integrase must compete with an always-nearby prime editor to access the target landing site when the two proteins are tethered.”
“In contrast, fusing the DNA-nicking Cas protein and the engineered reverse transcriptase together into a prime editor protein makes sense because the Cas protein must hold open—or melt—the two DNA strands in order for the nicked target DNA strand to prime reverse transcription, initiating the prime editing process,” Liu added. “In this case, fusing the Cas nickase to the engineered reverse transcriptase offers a benefit because it keeps the engineered reverse transcriptase nearby, poised to act on the opened and nicked target DNA site.”
“In other words,” Liu said, “the two proteins, in this case, act together, whereas in the case of PASSIGE/PASTE the two proteins must act separately.”
Liu co-founded Prime Medicine to commercialize prime editing based on Anzalone’s groundbreaking work when he was a postdoctoral fellow in Liu’s lab. Anzalone recently discussed the technology and the company on GEN’s “Close to the Edge” video interview series. (Anzalone is Prime’s lead developer of prime editing, and the company’s scientific co-founder and head of its prime editing platform.)
Search and replace
Prime says its editors can change any nucleotide to any other base, delete DNA sequences to correct insertion mutations, or insert DNA sequences to correct deletion mutations. Prime editors can also alter the regulatory regions of genes, insert or create premature stop codons, and modify splicing sequences.
Last year, Prime completed an initial public offering (IPO) that raised about $180 million despite the bear market for newly-public biotechs. In its prospectus, Prime revealed an 18-program pipeline and the in-licensing of U.S. Patent No. 11,447,770, which covers methods of using Prime Editors, was granted on September 20, 2022, and expires in 2040.
“We are confident in the strength of our patent portfolio,” Prime Medicine said in a statement to GEN Edge. “Broad Institute’s in-licensed IP includes an issued U.S. patent broadly covering Prime Editing methods and an allowed U.S. application, which is expected to issue shortly, covering pegRNAs. Our Prime Editing patent portfolio includes numerous in-licensed and Prime Medicine-owned patent applications in the U.S. and worldwide.”
Both the Broad Institute and Prime Medicine have filed for patent protection covering technological advancements that will “greatly” expand the scope of Prime Editing.
“We have in-licensed all Prime Editing improvements from Dr. Liu’s lab and anticipate continuing to do so in the future. We believe these investments, along with our continuing relationship with Dr. Liu, establish Prime Medicine as a clear leader in Prime Editing,” the company added. “We plan to continue investing in our Prime Editing technology with a focus on reinforcing our leadership position and making fundamental progress towards better therapies for patients.”
Prime has just announced the selection of its first development candidate—PM359, a treatment for chronic granulomatous disease, after observing long-term in vivo engraftment of Prime Edited hematopoietic stem cells (HSCs) in a mouse model of the disease.
“We look forward to advancing PM359 into investigational new drug (IND)-enabling studies later this year, while continuing to advance our broader portfolio toward additional preclinical proof-of-concept readouts,” Keith Gottesdiener, MD, Prime’s president and CEO, said in a statement.
Last January, Prime announced positive preclinical data for three of its development programs—candidates to treat Friedrich’s ataxia, cystic fibrosis, and Wilson’s disease. In addition, Prime said its PASSIGE platform achieved an approximately 60% precise insertion of a 3.5-kilobase transgene of interest at a single targeted site in primary human T cells, resulting in positive expression of the gene product.
Light bulb moment
Asked about the potential for a challenge to the PASTE patent, Gootenberg said: “It’s obviously going to be a complex situation with intersecting IP, as has been with Cas9 and other nucleases from the start. But that said, we’re really excited about how we can develop these technologies and move these technologies forward to develop actual cures for patients.”
Abudayyeh and Gootenberg are veterans of genome editing, having previously discovered Cas13, an RNA-targeting nuclease in the lab of Zhang, and later through their own lab, discovered the CRISPR effector protein Cas7-11. Abudayyeh and Gootenberg set up their joint lab at the McGovern Institute in 2019, bypassing the traditional postdoc path so they could keep working together.
“We had been basically interested in targeted gene integration for a long time. When we saw the paper, a light bulb went off in our heads,” Abuddayeh recalled. “We were like, whoa, wait a second! We saw these integrases that we were familiar with already. If we could somehow combine these systems together and basically use them to lay (at) the target site too.”
“When we first started down this path, Cas9 was the only thing people were focusing on for programmable double-stranded cleavage and editing. We asked the question of whether we can go beyond Cas9,” Abuddayeh added.
Gootenberg and Abudayyeh have licensed the technology behind PASTE editing to a startup they co-founded, Tome Biosciences, which has reportedly raised more than $95 million from several big-name investors, including Andreessen Horowitz (a16z), ARCH Venture Partners, and Polaris Partners. Other reported investors include Alexandria Venture Investments, Google’s GV, and Longwood Fund.
“Were it not for the weak market, I suspect (the PASTE patent) would have propelled CRISPR-related stocks higher. But this also demonstrates that—while markets are weak—innovation in the space is still going strong,” Jeff Brown, founder and chief investment analyst for Brownstone Research wrote last September on the investment firm’s website. “It’s advancements like this that will propel the entire industry forward—and move stock prices—when healthier markets return.”
Abudayyeh and Gootenberg would not discuss Tome Biosciences, which has said little about its operations publicly. Its website simply consists of a home page and a listing of more than a dozen available jobs. Tome was among companies that presented last year at GEN’s virtual presentation, “The State of Biotech”
On a roll
Tome is one of several companies co-founded by Gootenberg and Abudayyeh; the duo also co-founded Sherlock Biosciences, which recently gained rights to a U.S. patent for diagnostic use of a CRISPR system based on Cas12; Proof Diagnostics, a CRISPR-based COVID-19 molecular test developer; and Moment Biosciences, which has described itself as a precision microbiome therapy developer.
In the PASTE study published last year, Abuddayeh, Gootenberg, and a team of more than two dozen co-authors detailed how they delivered genes ranging from 779 to 36,000 base pairs—a range that would enable insertion of >99.7% of human cDNAs as transgenes—to three human cell lines, primary human T cells, and liver cells, as well as to liver cells in mice.
In all human cells studied, the researchers were able to insert genes with success rates that ranged from 5–60%, with “minimal” formation of indels at the integration sites. The researchers tested the delivery system with 13 payload genes, including some that could be therapeutically useful, and were able to insert them into nine different locations in the genome.
The researchers also reported inserting genes in “humanized” livers in mice, consisting of about 70% human hepatocytes, with PASTE showing a much lower percentage (as high as 2.5%) of successfully integrating new genes—with average insertion ranging from 0.3% or 1.4%.
Can PASTE transport genes larger than 36,000 base pairs?
“We’ve not done larger, but there’s no reason I think why you couldn’t do larger,” Abudayyeh said. “It becomes a delivery problem: How do we get 50,000 base pair templates into a human cell? The reason we’re able to max out at approximately 36 kb is because we’re using an adenoviral construct. Adenoviruses are approximately a 36-kb genome, so it’s easy to get that into the cell.”
Abudayyeh said they’ve been approached by scientists interested in inserts of 50–100 kb. “We just haven’t gone down that path,” he said. On its website, Tome Biosciences suggests it can handle DNA segments of that size range: “Using CRISPR, our technologies allow us to insert any genetic sequence of any size at any location into any genome.”
Beyond patent validity
While users of prime editing do not likely need a license covering PASTE, Sherkow added: “By contrast, if you are using the PASTE technology, there is, I think, an outstanding question as to whether you will also need to obtain a prime patent license.”
How companies answer that question will depend, he said, on how they use both PASTE and prime technology. “This is more just, frankly, garden-variety technology, improvement licensing stuff that happens in a variety of other fields,” Sherkow said. “It is not this cataclysmic dispute with respect to who got their first and validity, the way that we had in the CRISPR 1.0 context.”
The use of CRISPR-Cas9 in eukaryotic cells has been enmeshed in a nearly decade-long bitter battle royale over who invented the genome editing technology. The Patent Trial and Appeal Board (PTAB) last year decided a second patent interference process in favor of the Broad, MIT, and Harvard over the University of California (UC), the University of Vienna, and Nobel laureate Emmanuelle Charpentier, PhD. The first interference was decided in favor of the Broad, MIT, and Harvard in 2018.
“In the case of CRISPR 1.0, in some conceptual sense either the UC’s patents are valid and the Broad Institute’s are not, or vice versa. That’s just not necessarily true here between Harvard with prime editing and MIT with PASTE,” Sherkow said. “One could be valid, one could be not. One could be infringed, one could be not.
“It’s going to come down to whether a company that wants to develop a product in this area is going to be counseled to take a license to one or the other, or both. I think that’s going to be a function of what they think the best fit for their technology is,” Sherkow added.
As a result, he believes researchers will be inclined to use prime editing or PASTE—and thus pursue licenses for one or the other—based on the diseases they are working to treat.
Researchers in some therapeutic areas will be drawn to prime editing if they’re focused on gene correction—such as spinal muscular atrophy, most forms of which are caused by mutations of the survival motor neuron 1 gene.
“Unlike the CRISPR 1.0 situation, where who are we going to take a license to turns on, who do we think is going to win the patent fight, taking a license for prime [editing] or PASTE turns on what is our actual technology, which to be blunt is how we want the licensing system to work,” Sherkow said.
“We don’t want the licensing system to work where people are placing horse bets on winners, with some winners [betting] 90 cents to a dollar and coming in at $1.50, while others are just losing their shirts.”