The CRISPR Patent Wars Hit Japan: A High-Stakes Battle Over Gene-Editing's Future

"This brings us to Japan, where patent law has its own unique peculiarities—and where this particular battle just reached a crucial milestone. . . ."

The Nobel Prize That Started a War

In October 2020, Jennifer Doudna received an out-of-the-blue, 2:53 a.m. phone call from a reporter at Nature Magazine that would change her life forever. She learned that she'd just won the Nobel Prize in Chemistry together with her colleague, Emmanuelle Charpentier, for developing CRISPR—a revolutionary gene-editing technology that's proving to be the most important biological breakthrough of the 21st century. See UC Berkeley News (Oct. 7, 2020) (https://www.youtube.com/watch?v=RbNI_V0P574&t=140s).

But the celebration in Stockholm a year later was tempered by events unfolding in courtrooms across the globe. While Doudna and Charpentier were celebrating in Stockholm, lawyers around the world were locked in a multi-billion-dollar battle over who actually owns the rights to this world-changing invention.

Fast forward to June 26, 2025, when a Japanese appeals court handed down a decision that could reshape the global CRISPR patent landscape. In ToolGen Inc. v. The Regents of the University of California et al., Case No. Reiwa 5 (Gyō-Ke) 10147 (IP High Court, June 26, 2025) ("Decision"), Japan's Intellectual Property High Court sided with the University of California ("UC") in a closely watched patent dispute. The stakes are high—control over a technology that can cure genetic diseases, create drought-resistant crops, and potentially eliminate hereditary conditions before birth.

You might be wondering; how can there be a dispute over who invented something that already won a Nobel Prize? The answer to that question reveals the complex and often unforgiving interplay between cutting-edge scientific innovations and patent law's award of ownership rights in those achievements—and why a seemingly technical legal decision in Tokyo matters for anyone who might benefit from gene therapy in the future.

From Bacterial Bodyguards to Genetic Revolution

To understand this legal battle, we need to start with one of nature's oldest and most elegant defense systems. Imagine that you're a bacterium living in a world full of viruses trying to hijack your cellular machinery—with the goal of injecting their DNA to turn you into a virus factory. If you survive the attacks, you need a way to remember those past attacks and fight them off if they return, protecting both yourself and later generations of bacteria. That's exactly what CRISPR does—it's essentially a bacterial immune system with a memory.

Here's how it works in nature. When a virus attacks a bacterium, the bacterial cell takes a molecular "mugshot" of the invader's DNA and files it away in a special genetic filing cabinet called CRISPR (which stands for "Clustered Regularly Interspaced Short Palindromic Repeats"—basically, nature's way of organizing these mugshots). If that same virus shows up again, the bacterium pulls out the mugshot, makes a copy called "guide RNA," and sends it out with a molecular pair of scissors, a protein called "Cas9." The guide RNA finds the matching viral DNA, and Cas9 cuts it up, defeating the viral invasion.

Think of it like having security guards (Cas9) with a photo (guide RNA) of someone you want to keep out of your building. The guards routinely patrol the building until they find someone matching the photo, then—snip—the threat is removed.

Now here's where Doudna and her team made their breakthrough; they realized they could engineer new "photos"—new guide RNAs—that would direct Cas9 to cut any DNA sequence they wanted. Suddenly, this bacterial defense system became a programmable gene editor. Instead of just fighting viruses, scientists can now use it in ways that make old science fiction movies look like yesterday's news.

The Billion-Dollar Question: Bacteria vs. Everything Else

Now here's where the patent fight gets interesting, and why that Japanese court decision matters so much. There's a huge difference between making CRISPR work in bacteria (called "prokaryotes") and making it work in everything else—plants, animals, humans (called "eukaryotes").

And here's why this distinction matters. Imagine you've patented a revolutionary new car engine that works perfectly in a simple go-kart. Now someone asks, "Does your patent cover this engine working in a Formula 1 race car?" The basic principle might be the same, but a Formula 1 car has sophisticated computer systems, complex aerodynamics, and safety features that a go-kart doesn't have. Making your engine work in that more complex environment might require solving entirely new problems.

That's essentially the situation with CRISPR. Bacterial cells are relatively simple—like biological go-karts. But eukaryotic cells (which include everything from yeast to redwood trees to humans) are vastly more complex. They have a nucleus where DNA is packaged with proteins, sophisticated cellular machinery, and multiple compartments—all things that bacteria themselves don't have.

In this way, the CRISPR patent wars boil down to a fight over the genomic pie—comprised of the prokaryote and eukaryote genomes. And here's the striking irony; Jennifer Doudna and her colleagues, the undisputed inventors of the programmable gene-editing tool and Nobel Prize winners for their breakthrough, find themselves in patent battles around the world where they could potentially lose rights to the most valuable pieces of that pie—patent rights to CRISPR gene editing of eukaryotic DNA. The question therefore isn't just who invented CRISPR, but who gets the biggest and most desirable slices of the multi-billion-dollar market it's created—with the real possibility that the original inventors might end up with the smallest slice.

The question that would determine billions of dollars in patent rights was this—When UC and its co-inventors filed their foundational CRISPR patent application in May 2012, did they adequately describe how to make this modified bacterial system work within the complex internal environment of eukaryotic cells? Or did they just have a great idea that still needed years of additional work to become the universal gene-editing tool we know today?

The Race That Never Ended

The CRISPR patent story reads like a scientific thriller with multiple plot twists, evoking a foot race for the grand prize. In one lane, the UC team led by Doudna and Charpentier, who filed their foundational patent application on May 25, 2012. In the other lane, a team at the Broad Institute (a collaboration between MIT and Harvard) led by Feng Zhang, a PhD biochemist, who filed a later patent application, claiming to be the first to actually make CRISPR work in eukaryotic cells.

This wasn't just a two-horse race. Research teams around the world were all chasing the same prize, filing patent applications in multiple countries with slightly different claims and timing. Some teams worked alone, and others collaborated. It quickly resembled a global relay race where everyone's running simultaneously in different lanes of the same track, and the finish line keeps moving as each of the new breakthroughs surrounding the CRISPR technology keep piling up.

Back at the starting blocks, UC's strategy was bold. UC claimed that its original work provided the foundation for using CRISPR in all types of cells. Their patent application described the basic CRISPR-Cas9 system and included the crucial insight that you could program it to target any DNA sequence by changing the guide RNA.

But here's the fundamental problem that dogs the team to this day. They demonstrated the programmable CRISPR-Cas9 system working in vitro (in test tubes) and in bacterial cells, but—and this is the critical detail—they didn't include any examples of it working in eukaryotic cells. To be clear, the patent said that the invention can be applied to eukaryotic cells using conventional means but offered no real-world examples of having done so.

This fundamental question—whether UC's patent adequately described eukaryotic applications—has been litigated around the world, with courts and patent offices reaching strikingly different conclusions. While some jurisdictions have sided with UC, others have denied UC's claim to priority and awarded separate patents to competitors like the Broad Institute for eukaryotic applications. The result is a fragmented global landscape where the same scientific breakthrough receives different patent treatment depending on which country's laws apply—setting the stage for the crucial battle that would unfold in Japanese courts.

Enter the Japanese Patent Battlefield

This brings us to Japan, where patent law has its own unique peculiarities—and where this particular battle just reached a crucial milestone. There's also a certain historical irony to this legal battle playing out in Japanese courts. CRISPR itself was first discovered in Japan in the 1980s by Yoshizumi Ishino at Osaka University, when he stumbled upon these strange repetitive DNA sequences while studying bacteria. He had no idea at the time that he'd discovered the foundation of what would become the most powerful gene-editing tool in history. Now, decades later, Japanese courts are helping determine who owns the rights to the revolutionary technology that grew from that original discovery.

That's so because Japanese law provides third parties an opportunity to challenge issued patents. In Japan, once a patent's granted, anyone can challenge it through a process called a "patent invalidation trial" conducted in the Japanese Patent Office ("JPO"). It's essentially a formal legal proceeding in the JPO designed to double-check whether a patent should have been granted in the first place.

That's exactly what happened to UC's CRISPR patent. A competitor, ToolGen, initiated an invalidation trial at the JPO, asserting the fundamental dispute plaguing UC's patent: "The patent should be limited to using the invention in prokaryotes because the original application didn't adequately describe how to make CRISPR work in eukaryotes." The JPO examined the evidence presented by ToolGen and sided with UC, concluding the full scope of the patent should remain valid. But ToolGen wasn't finished—it appealed to Japan's IP High Court.

What makes this dispute particularly significant is that Japan represents a massive biotechnology market. Japanese pharmaceutical companies are major players in developing CRISPR-based therapies, and the patent landscape in Japan could influence licensing deals worth hundreds of millions of dollars. And as we've seen, different countries sometimes reach different conclusions about the same patents, creating a complex global puzzle where a certain institution owns patent rights to the technology in some places but not others.

The Technical Heart of the Legal Battle

Now we get to the really fascinating part—where cutting-edge molecular biology meets the intricate and sometimes messy world of patent law. The legal question appears simple; did UC's original patent application provide enough detail for a skilled scientist to use CRISPR in eukaryotic cells? But answering that question required the court to dive deep into the technical field of cellular biology.

First, the court had to define who counts as a "person of ordinary skill in the art"—or "POSA," as patent lawyers call them. This isn't just academic hairsplitting. It's the predicate question that often determines who wins and who loses in a patent fight. Imagine you're describing how to bake a cake to someone. The amount of detail you need to include depends entirely on who you're talking to. If you're talking to a professional pastry chef, you might just say "make a basic sponge cake." But if you're talking to someone who's never baked before, you need to explain everything from how to crack an egg to what "fold in gently" means.

In this case, ToolGen argued that the relevant POSA should be a general molecular biologist or even a university student—someone who might have used the gene editing tools predating CRISPR but isn't necessarily an expert in the field of gene editing. Think of them as competent home cooks who can follow a recipe but might struggle if crucial steps are left unexplained. UC, on the other hand, argued that the relevant POSA would have more specialized knowledge about gene-editing systems, akin to the professional pastry chef.

This distinction mattered enormously. Under Japanese patent law, what counts as "common technical knowledge" can be left out of the patent's description of the invention. In other words, a patent need not explain "common technical knowledge" relevant to the invention because Japanese patent law presumes that the POSA, like our master chef, is already proficient in that "common technical knowledge."

The Missing Puzzle Pieces—Common Knowledge or Critical Elements

ToolGen then pointed to specific technical hurdles that it claimed were essential to making the leap from prokaryotes to eukaryotes much harder than UC's patent suggested. These weren't minor details—they were fundamental make-or-break challenges a POSA would have faced in making the CRISPR-Cas9 system succeed in eukaryotic cells.

First was the "PAM" sequence problem. In nature, Cas9 doesn't just cut DNA anywhere—it needs a specific molecular "landing pad" called a PAM ("Protospacer Adjacent Motif") sequence located right next to its target. It's like a molecular zip code that tells Cas9 "you can cut here." Actually, PAM sequences serve an even more crucial biological purpose—they're how bacteria distinguish between "self" and "non-self" DNA whenever a virus attacks. Bacterial cells don't have PAM sequences in their own DNA, so Cas9 never attacks the bacteria's own DNA. But invading viruses do have PAM sequences, making them fair game for Cas9's molecular scissors. No PAM sequence, no cut—it's nature's safety mechanism to prevent friendly fire.

While scientists understood this self/non-self distinction in bacterial systems, ToolGen argued that it wasn't obvious these same PAM requirements would work the same way in eukaryotic cells, or that researchers would know which PAM sequences to use.

Then there was the "nuclear localization signal" issue. Bacterial cells are simple—their DNA just floats around inside the cell. But eukaryotic cells keep their DNA locked inside a nuclear fortress with sophisticated security. One way of getting Cas9 and other proteins into that nucleus involves special "passport" sequences called nuclear localization signals ("NLS"). Without them, Cas9 might be like a delivery person stuck outside a gated community with no way to reach the address unless he knows the pin code that opens the gate.

Third was codon optimization—essentially a translation problem. Although all DNA share common chemical instructions for making proteins, different types of cells prefer different versions of those instructions (like regional dialects of the same language). The Cas9 protein was originally designed to work with a bacterial "dialect," but making it work efficiently in eukaryotic cells might require translating it into the local "dialect" of eukaryotes.

ToolGen's argument was essentially this: "Sure, UC showed that the patent's CRISPR system can be programmed to work in bacteria and even explained the underlying principles of how that system operates. But jumping to eukaryotic cells required solving all these additional problems that weren't addressed in the patent."

ToolGen further argued that, when confronted with these significant hurdles at the outset, a POSA would likely have been discouraged from even attempting to work the invention in eukaryotes. Under Japanese patent law, these kinds of hurdles are referred to as "barriers," and if the barriers are determined to be too great, a court can invalidate the patent.

Alternatively, ToolGen argued that even if POSAs tried to implement the invention in eukaryotes, they would have needed to conduct extensive trial-and-error experimentation to make it work. Again, under Japanese patent law, if using the claimed invention requires undue experimentation involving techniques and elements left out of the patent, a court can likewise invalidate the patent.

The Courtroom Molecular Detective Work

Here's where the Japanese judges had to become part scientist, part legal scholar, and part fortune teller. They needed to transport themselves back to May 2012 and ask, "What would a POSA have reasonably understood from UC's patent application?"

ToolGen painted a picture of scientific uncertainty and technical roadblocks. It pointed to emails exchanged between the UC inventors themselves, showing that even the people who invented the programmable CRISPR-Cas9 system were struggling to make it work in eukaryotic cells—even after filing their patent. "Look," ToolGen essentially argued, "if the inventors themselves couldn't figure it out right away, how can you conclude that the patent adequately taught a POSA how to do it?"

ToolGen highlighted a series of UC's failed experiments in 2012 where the inventors tried to get CRISPR working in human and other eukaryotic cells and hit wall after wall. RNA degradation was destroying their guide RNAs. The Cas9 protein wasn't getting into cell nuclei efficiently. The complex cellular environment of eukaryotes seemed to interfere with the system in unpredictable ways. It was like trying to perform delicate surgery in a crowded, noisy railway station instead of a quiescent operating room.

"These weren't just minor tweaks," ToolGen argued, "These were fundamental barriers that would have discouraged any reasonable scientist from even attempting to use CRISPR in eukaryotic cells, or if they did, would have required substantial trial-and-error experimentation to overcome."

UC's Counterattack: "The Tools Were Already There"

But UC had a powerful counter-narrative. Its argument was essentially: "All the pieces of the puzzle were already sitting on the table. Any competent molecular biologist could have put them together."

They pointed to the scientific literature from before May 2012, showing that knowledge about PAM sequences, nuclear localization signals, and codon optimization was already well-established in the field. These weren't mysterious unknowns—they were standard tools in any molecular biologist's toolkit.

Thinking of it like a cookbook again, UC's lawyers essentially argued, "We provided the core recipe and explained the underlying chemistry. We didn't need to separately explain how to adjust the cooking time for different ovens, or how to substitute ingredients for people with allergies, because any competent chef already knows how to make those adjustments."

More to the point, UC revealed what happened immediately after their research was published for the first time in Science Magazine's June 2012 publication. The disclosure of a programmable CRISPR-Cas9 gene editing tool was a seismic event. Everyone in the field immediately recognized its profound potential to revolutionize gene editing with a tool that can be "programmed" by simply substituting Cas9's guide RNA with a sequence correlated to the target DNA. The June 2012 publication thus promised a gene editing tool that could be programmed in a matter of hours, unlike existing tools that often required months of preparation.

Specifically, UC presented evidence that in the six months following their June 2012 publication in Science, multiple independent research groups around the world successfully adapted CRISPR for use not only in prokaryotic cells but also in eukaryotic cells. Those research teams reported success in a wide range of eukaryotic cells, even in the grand prize of them all—human cells.

This was UC's smoking gun: "If it was so difficult and uncertain, how did all these different labs figure it out so quickly?" More startling yet, UC presented evidence that the research teams accomplished their work using known, conventional techniques to introduce their programmed CRISPR-Cas9 constructs into eukaryotic cells. Nothing special or out of the ordinary was required. UC essentially argued, "The fact that multiple groups succeeded almost simultaneously using conventional technology proves that the technical barriers weren't as insurmountable as ToolGen claims."

The Court's Verdict: A Detailed Legal and Technical Analysis

The Japanese IP High Court's decision turned on a fundamental question of patent law under both the Paris Convention and Japanese Patent Law Article 29-2—whether UC's May 2012 patent application provided sufficient technical details to support its priority claim. This wasn't merely about whether CRISPR works in eukaryotic cells—it was about whether UC's original application adequately taught a POSA how to implement the invention without undue experimentation.

Understanding the Priority Puzzle

Think of patent priority like calling "dibs" on an invention, but with strict rules about what you need to prove. To understand the court's analysis, it's crucial to know that UC's patent is based on a series of three patent applications it filed successively over a period of several months. The first application, filed on May 25, 2012, described the basic CRISPR system but only included examples in bacteria and in vitro experiments. The second application added more detail but no concrete examples of using the invention in eukaryotes. UC's third application, filed on January 28, 2013, indisputably included actual working examples in eukaryotic cells—but by then, competitors like the Broad Institute had already filed their own applications claiming eukaryotic uses of the invention.

Here's the legal puzzle—if UC can only claim priority from its third application (the one with eukaryotic examples), then competitors who filed patent applications between May 2012 and January 2013 would have superior patent rights. But if UC can successfully claim priority all the way back to its first application from May 2012, it beats everyone to the patent office. The entire case hinged on whether that original May 2012 application—what the court calls "the first application"—contained enough detail to support eukaryotic uses of the invention, even though UC hadn't yet demonstrated those uses in actual experiments.

In essence, the court had to determine whether UC's May 2012 filing provided a legally sufficient blueprint for eukaryotic gene editing, or merely a promising proof-of-concept that still needed substantial additional development.

The Legal Framework: International Priority Rules

The court applied international patent law, grounding its analysis in Article 4 of the Paris Convention, specifically noting that Article 4A(1) grants priority rights to those who've duly filed patent applications in member countries. Crucially, the court cited Article 4H, which provides that "priority cannot be denied on the ground that a component part of the invention to which the priority claim pertains is not stated as within the claims in the initial application, provided that the component part is made clear in the entire application for the initial application." Decision at 14.

The legal test the court applied was straightforward in concept but complex in application. The court established the controlling legal standard; for UC to maintain its May 25, 2012 priority date, "a POSA must have been able to implement the invention without excessive trial and error based on the description of the entirety of the first application document and common technical knowledge as of the priority date." Id at 15. If the invention couldn't have been implemented by a POSA even when combining the first application's disclosure with the common technical knowledge as of the priority date, then "the invention is not considered to be a matter described in the entire first application document and the effect of the priority claim under the Paris Convention should not be recognized." Id.

Adopting ToolGen's POSA Standard—And Still Ruling for UC

In what should have been a victory for ToolGen, the court essentially adopted its restrictive definition of the relevant POSA. Rather than defining the POSA as a gene editing expert, the court accepted that the appropriate standard was "general researchers and students in the field of molecular biology who used gene editing tools existing at the time of the first application." Id. at 38.

This was a significant procedural victory for ToolGen, as it meant that what counts as "common technical knowledge" was judged, not by the knowledge held by experts in the field, but by the more mundane knowledge and expertise of "generalists" engaged in gene editing at the time of the first application. But remarkably, even applying this higher bar, the court systematically found that all the technical knowledge ToolGen claimed was missing was actually accessible to these general practitioners. See id. at 38-39 ("even if those individuals who are claimed by the plaintiff are considered as persons of ordinary skill in the art, we are not restrained from considering the teachings of PAM sequences published before the priority date as well-known art and technical common knowledge understood by a person of ordinary skill in the art").

PAM Sequences: Universally Known, Even to General Practitioners

Remember those PAM sequences—the molecular "zip codes" that tell Cas9 where it's allowed to cut? ToolGen argued that figuring out how these worked in complex eukaryotic cells was a major unsolved puzzle that would have stumped scientists in 2012.

The court disagreed, conducting a thorough and technically sophisticated analysis of the PAM sequence problem. The judges found that "PAM sequences had been mentioned in many documents before the priority date, including documents cited by [UC]," and they specifically referenced key publications that established knowledge of PAM sequences in the field before UC's first application. Id. at 36.

The court didn't just rely on general assertions—it dove deep into the scientific literature to make this determination. The court recognized that a large body of literature predating UC's first application had meticulously elucidated the role of PAM in Cas9's targeting of DNA sequences.

For example, the court referenced several articles that identified PAM as a prerequisite to Cas9 targeting and cleaving its target DNA. Those references include the work of Michael P. Terns et al., CRISPR-Based Adaptive Immune Systems, Current Opinion in Microbiology, Vol. 14(3), 321-327 (June 2011) (B25) and F.J.M. Mojica, Short motif sequences [PAM] determine the targets of the prokaryotic CRISPR defence system, Microbiology, Vol. 155(3), 733-740 (March 2009) (B20). See Decision at 27 & 29.

Kira S. Makarova's influential paper Evolution and classification of the CRISPR/Cas9 systems, Nature Reviews Microbiology, Vol. 9, 467-477 (May 9, 2011) (B16) supported that notion, stating, "In the type II system, Cas9 bound to CrRNA presumably targets the invading DNA directly, and the process requires PAM." Decision at 25.

The court was further persuaded by publications confirming the essential role of PAM by showing that Cas9's cleavage function fails in situations where the PAM sequence is altered by mutation. For example, Philippe Horvath's foundational work CRISPR/Cas, the Immune System of Bacteria and Archaea, Science, Vol. 327(5962), 167-170 (January 8, 2010) (B19) established that "phages can evade the CRISPR/Cas system by mutating the CRISPR motif [PAM], indicating that the CRISPR motif is involved in the immunity encoded by CRISPR." Decision at 25-26.

Devaki Bhaya's comprehensive review CRISPR/Cas9 Systems in Bacteria and Archaea: Versatile Small RNAs for Adaptive Defense and Regulation, Annual Review of Genetics, Vol. 45, 273-97 (December 2011) (B15) likewise demonstrated that "mature CrRNAs integrate with Cas9 and perform homology-driven cleavage in the immediate vicinity of the PAM of the protospacer sequence" and that "mismatches at the PAM allow phages and plasmids to evade CRISPR-encoded immunity." Decision at 24; see also Rimantas Sapranauskas et al., The Streptococcus Thermophilus CRISPR/Cas system provides immunity in Escherichia coli, Nucleic Acids Research, Vol. 39(21), 9275-9282 (August 3, 2011) (B21) ("mutations near or within the protospacer adjacent motif (PAM) allow plasmids to escape CRISPR-encoded immunity"); Hélène Deveau, Phage Response to CRISPR-Encoded Resistance in Streptococcus thermophilus, Journal of Bacteriology, Vol. 190(4), 1390-1400 (December 7, 2007) (B22) (mutations in PAM sequences allowed "phage to escape CRISPR1-mediated resistance"). Decision at 26-27.

In other words, by 2012, the role of PAM sequences wasn't a mystery—it was well-established and accessible science, even under ToolGen's restrictive POSA standard, at the time of UC's first application. The court concluded that "even general researchers and students in molecular biology would have had knowledge obtainable through the literature that had already been published for a certain period concerning the system." Decision at 38.

But here's another twist that really strengthened UC's case. The court also found that UC's first application actually contained implicit evidence supporting PAM sequence functionality in eukaryotic cells. Although UC hadn't explicitly identified PAM sequences or described their necessity, the court noted that the target DNA sequences shown in UC's experimental figures (Figures 3C and 5B) actually contained the well-known "NGG" PAM sequence adjacent to the target sites—structures that were "consistent with the literature findings" about PAM requirements. Id. at 36-37.

Moreover, consistent with that finding, UC presented evidence that these target sequences "had greater homology with eukaryotic DNA sequences," meaning there was no reason to doubt that PAM sequences would function in eukaryotic cells just as they did in prokaryotic cells. Id. at 12-13. This gave UC additional support for their argument that a POSA would naturally expect the CRISPR system to function in eukaryotic cells.

NLS and Codon Optimization: Well-Known & Unnecessary

For the other technical challenges ToolGen raised—getting proteins into cell nuclei and optimizing genetic "dialects" for different cell types—the court reached similar conclusions. These weren't unsolved mysteries but routine techniques that molecular biologists were already using for other applications.

For nuclear localization signals, the court referenced a long list of prior art publications establishing this as routine technology. These included Yannick Doyon's work in Nature Biotechnology, Vol. 26(6), 702-708 (May 25, 2008) (B36); Claudio Mussolino's paper in Nucleic Acids Research, Vol. 39(21), 9283-9293 (August 3, 2011) (B69); Leaf Huang's chapter in Nonviral Vectors for Gene Therapy, Chapter 7, 139-142, 147-153 (1999) (B72); Edward J. Rebar's publication in Nature Medicine, Vol. 8(12), 1427-1432 (November 4, 2002) (B77); and Tomas Cermak's work in Nucleic Acids Research, Vol. 39(12), e82, 1-11, (April 14, 2011) (B78). Decision at 30. This body of work illustrated that "the technique of adding one or more nuclear localization signals (NLS) to proteins in order to efficiently transport the proteins into the nucleus where chromosomes exist was a well-known conventional technology as of the priority date." Id. at 30-31.

But more importantly, the court found that NLS wasn't even necessary. UC had presented evidence that CRISPR could be delivered to eukaryotic cells through alternative methods—"nucleic acids can be introduced into cells as RNA by known techniques for introducing nucleic acids into cells (microinjection, electroporation, etc.)." Id. at 31. Moreover, the court noted ToolGen's admission that post-priority literature confirmed that "target DNA cleavage (genome editing) by Cas9 was possible without having NLS, so it's difficult to conclude that it was an essential technology for the CRISPR Cas9 system." Id. at 39.

For codon optimization, the court referenced numerous publications, including: S.B. Primrose's textbook Principles of Gene Manipulation and Genomics, 233-35, (Seventh ed. 2006) (B75); Claes Gustafsson's review in Trends in Biotechnology, Vol. 2(7), 346-353 (July 4, 2004) (B79); Stacey S. Patterson's work in Journal of Industrial Microbiology & Biotechnology, Vol. 32(3), 115-123 (March 11, 2005) (B80); Huirong Gao's paper in The Plant Journal, Vol. 61(1), 176-187, (November 9, 2009) (B82); Vipula K. Shukla's publication in Nature, Vol. 459(7245), 437-441 (April 29, 2009) (B84); and Alan Villalobos' work in BMC Bioinformatics, Vol. 7(285) (2006) (B180). Decision at 31-32.

Again, the court applied identical reasoning. It found this was "well-known conventional technology" that was discussed extensively within prior art publications. Id. at 32 & 39. And again, the court emphasized that codon optimization, while preferred in many situations, wasn't essential; the prior art showed that "genome editing was possible without codon optimization." Id. at 39.

UC's Original Application: More Complete Than It First Appeared

The court's most significant finding was that UC's original application contained much more than ToolGen acknowledged. Critically, the court expressly found that "the technical concept of applying the CRISPR/Cas9 system to target DNA in eukaryotic cells was disclosed in the first application, and the invention's detailed disclosure was sufficient to enable its implementation in combination with the known technology as of the priority date." Id. at 36.

Rather than just describing a bacterial gene-editing system, UC had disclosed the fundamental concept for universal gene editing. The court rejected ToolGen's argument that UC's priority rights should be denied because the first application included no concrete examples of using the invention in eukaryotic cells. Having found that the first application's "disclosure, when combined with the known prior art, was sufficiently specific to enable the invention," the court concluded that "the absence of specific examples provides no grounds to deny priority based on the [the first application]." Id. at 41.

The court was likewise unmoved by ToolGen's evidence that the inventors' contemporaneous email communications suggested that they'd tried, but failed, to implement the invention in eukaryotes before filing their first application. In analyzing this issue, the court essentially recognized that the priority issue should be judged against the skill level of a POSA rather than the inventors themselves. Specifically, the court found:

Id. at 40.

Real-World Validation: Rapid Success After Publication

Perhaps the most compelling evidence for UC's position was what happened immediately after the inventors published their research. In further support of those findings, the court recited the undisputed evidence that other research groups rapidly implemented the CRISPR Cas9 gene editing techniques in eukaryotes. Notably, those successes were informed by the inventors' June 2012 Science article, which contained even less information about the invention than the then-unpublished first application. Regarding those successes, the court emphasized:

Id.

If it had been as difficult and uncertain as ToolGen claimed, how did multiple independent research teams figure it out so quickly? This was significant because it showed that UC hadn't just stumbled upon a gene editing system that works in bacteria—it had conceived of that system as a replacement for existing gene-editing technologies that were already being used in both prokaryotic and eukaryotic cells. As the court observed, pre-existing techniques like "ZFN" ("zinc-finger nucleases") and "TALEN" ("transcription activator-like effector nucleases") were in use before the first application but each of those techniques involved the time-consuming "construction of chimeric endonuclease enzymes by fusing a sequence-nonspecific DNA endonuclease domain to an engineered DNA binding domain." Id. at 23.

The court specifically noted that the first application was not expressly limited to prokaryotic gene editing. Instead, the first application presented a novel alternative to ZFN and TALEN by "providing a technique that solved the problem of enabling the accurate targeting of nuclease activity to target DNA without requiring the design of new proteins (nucleases) for each new target sequence." Id. at 33.

The Bottom Line: A Systematic Victory

Having satisfied itself that the invention disclosed in the first application could be implemented in eukaryotes with then-existing common knowledge, the court concluded that the invention is entitled to priority for the entirety of its stated purpose—as a novel alternative to ZFN and TALEN, in both prokaryotes and eukaryotes. See generally id. at 41 ("[I]t is obvious to this court that this invention was disclosed based on the description of the entire first application and the common general knowledge at the time of filing [and] is entitled to the benefit of priority based on the first application").

This wasn't simply a close call—it was a systematic determination that even under ToolGen's restrictive POSA standard, all the necessary technical knowledge was accessible, much of it wasn't even required, and the rapid success of multiple research groups provided compelling objective evidence of enablement. In this way, the decision represents a significant clarification of Japanese patent law and guidance to lower courts in how they should analyze patents on breakthrough technologies where the fundamental insight may precede the detailed engineering solutions for specific applications claimed in the patent.

Conclusion

UC's victory in Japan is significant, but it represents just one battle in a much larger global war. The CRISPR patent landscape remains fragmented across different countries, with various institutions claiming overlapping rights to different aspects of the technology. For biotechnology companies developing CRISPR-based therapies, this creates a complex web of licensing requirements. There's also a generational irony at play; while patent lawyers continue fighting over who invented what in 2012, a new generation of scientists has grown up using CRISPR as a routine laboratory tool while unaware of the ongoing courtroom drama—much like how today's computer programmers might find it difficult to imagine fighting over who invented the Internet.

But the Japanese IP High Court's decision provides important clarity about how courts should analyze breakthrough biotechnology patents where fundamental insights may precede detailed applications. By systematically applying the Paris Convention priority framework and finding that UC's original disclosure, combined with pre-existing common technical knowledge, adequately enabled eukaryotic applications, the court established a significant precedent. The decision demonstrates that even when inventors haven't yet solved every engineering challenge, they might still secure broad patent protection if their core insight is sound and the remaining steps are routine for skilled practitioners. For the ongoing global CRISPR patent disputes, this Japanese ruling offers a roadmap for how courts might balance the competing demands of rewarding pioneering discoveries while ensuring those discoveries are adequately disclosed to the scientific community.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.