CRISPR FAQs

• CRISPR is short for Clustered Regularly Interspaced Short Palindromic Repeats, a system of adaptive immunity first characterised in bacteria and archaea which has now been harnessed by the scientific community for a host of applications. Its most well-known application is to facilitate accurate and targeted cutting of DNA to enable editing of genes.

• A CRISPR/Cas complex comprises a CRISPR associated protein (Cas) and guide RNA molecule which is complementary to the target locus.

• The seminal Cas protein for genome-editing applications is Cas9 from Streptococcus pyogenes, although many more Cas proteins have been identified and are being studied, most notably Cpf1 (aka Cas12) and C2c2 (aka Cas13). Cas9 induces blunt ended double strand DNA breaks, whereas Cpf1 yields sticky ended DNA breaks (4-5 base pair overhang) and C2c2 targets RNA.

• The CRISPR system had first begun to be characterised in the 80s, but its potential was not realised until 2012 when Jennifer Doudna and Emmanuelle Charpentier developed a method to simplify the RNA guide component of the CRISPR complex. Later Feng Zhang and George Church, separately but simultaneously (in the same issue of Science), were the first groups to publish a method for its use in eukaryotes. This was followed very shortly after by the publication of independent work by other labs, including those of Jennifer Doudna and Jin-Soo Kim, which also described the successful use of the CRISPR system in eukaryotes.

• The realm of potential applications for the CRISPR system is very broad. It has already become a key investigational tool in basic science and the study of biological systems. Equally, its utility in both drug discovery applications as well as direct therapeutic applications has been well demonstrated, and a clinical trial is already underway in China in patients with aggressive lung cancer, using CRISPR to knock down PD-1 receptor expression in T-cells. Additionally the University of Pennsylvania has received approval in the US for a CRISPR related clinical trial (again targeting PD-1 receptors, amongst other targets).

• Aside from biomedical applications, there is significant potential for the use of CRISPR in agriculture, using genetic modification to introduce positive genetic traits to crops and livestock such as disease resistance, drought tolerance or improved nutritional properties. It also facilitates the engineering of human/humanised medical products or tissues in animals and the production of “green chemicals” such as biomaterials and biofuels from industrial bacterial, fungal and yeast cultures.

• Perhaps the most controversial application of CRISPR which is currently being pursued is in the creation of “gene drives” whereby incorporation of CRISPR system components into the inherited germ line of an organism drives the preferential expression of a particular genotype in a population. The example that has most captured the imagination is the use of gene drive technology to eradicate mosquitos which carry malaria, by driving the inheritance of a recessive female sterility gene. Serious regulatory and public policy questions are raised by the sheer power of such a technique and the challenge of controlling and containing a technology with the potential to alter an ecosystem or irreversibly change the genetic inheritance of an entire population.

• Research into the basic components of the CRISPR/Cas system continues at a rapid pace. Our understanding of the molecular mechanisms behind the CRISPR/Cas system in its myriad forms is expanding almost as fast as scientists can apply that understanding to find new applications and develop new molecular tools based on those principles.

• Much work has been done in silico (via computer simulation) to identify promising new Cas proteins to expand the flexibility of the CRISPR arsenal. Cpf1 and C2c2 have been identified as two of the most promising of the second wave of Cas proteins, and each has different properties and offers additional functionality to the first generation Cas9 protein. As additional Cas proteins are identified and characterised, this will expand the molecular toolbox available to researchers and mitigate the limitations of any single given Cas (in particular the specific PAM (Protospacer Adjacent Motif) required for target site binding and cleavage).

• It should also be noted that the utility of the CRISPR system is not limited to gene disruption and gene editing (through subsequent non-homologous end joining or homology directed repair); researchers are also taking full advantage of the simple ability of the crRNA (CRISPR RNA) to guide the CRISPR complex to bind its complement tightly and with specificity. For example Cas9 proteins with “dead” endonuclease activity can be fused with transcription factors to promote or repress gene expression, or can be used to bind directly to enhancer regions and interfere with transcription. This shows the power of the system as both a tool in mechanistic studies as well as a modulator of expression.

One of the most significant obstacles facing researchers looking to develop clinical and industrial applications for CRISPR systems is the challenge of delivering the system to the target cell or cells. Existing delivery vehicles for the CRISPR system (which must encode for each of the components of the system – Cas protein, single guide RNA (sgRNA) or crRNA and trans-activating CRISPR RNA (tracrRNA), etc.) include:

• various viral vectors (lentivirus, adenovirus)

• delivery of the RNA components through microinjection or electroporation

• lipid nanoparticle delivery of a mature CRISPR protein complex, plasmid encoding Cas protein and sgRNA gene, mRNA transcript for Cas protein, synthetic sgRNA, or some combination of the above

• Each of these routes of delivery has its own advantages and disadvantages, but all are limited by the size of the plasmid/transcript/protein which is able to be delivered, as well as the potential to trigger an immune response (in particular the viral vectors). There is much research being conducted into delivery technology as well as smaller Cas protein variants (such as Cpf1and Cas9 derived from S. aureus rather than S. pyogenes) as a result. In vivo delivery methods other than viral vectors will also be needed to expand the armoury and to avoid the need to go through the process of extracting patient cells, modifying them and reintroducing them into the patient (which is highly labour intensive and carries the added risk of contamination and mishandling).

• Potential off target effects of CRISPR systems is the other main challenge researchers face, both at a cellular level and DNA level. It is still very difficult to target particular cell types or organs with current delivery technology and indiscriminate CRISPR activity in vivo could give rise to a host of side effects. Similarly, off target cleavage of DNA sites even in target cells of interest could similarly yield harmful effects. This is a multifaceted issue which is being tackled from a variety of angles, from improving specificity of targeting to ensuring that the correct repair mechanisms are engaged following DNA cleavage.

• One technique being used is using modified Cas9 “nickase” proteins in which one of the two endonuclease domains of Cas9 has been neutralised (Cas9n proteins). Cas9n proteins will create a single stranded “nick” at the target site rather than the double stranded break that wild type Cas9 would create. A double stranded break can be created using nickases by using two separate guide RNAs complexed to Cas9n to form nicks at the target site – this ensures greater specificity as it requires the specific binding of both guide RNAs at the target site. A single, off target nick will be unlikely to cause any serious damage as it can be repaired by accurate single base excision repair mechanisms.

• In terms of using CRISPR for gene editing to insert, repair or correct genetic material (as opposed to disrupting a gene), the key challenge to overcome is obtaining high levels of homology directed repair. Homology directed repair (HDR) and non-homologous end joining (NHEJ) are the native DNA repair mechanisms which researchers take advantage of in order to perform deletions or insertions in DNA once a double stranded break has been formed (e.g. by the action of a CRISPR complex). HDR occurs much less frequently than NHEJ and is in practice the only way to make precise genome corrections at the desired site (NHEJ is fine for deletions). Much work is being done to try to improve HDR rates across a range of different cell types and researchers have found there is a lot of variability in HDR across different cell types. At this stage HDR rates seem to be nuclease agnostic but it is possible that different cleavage patterns characteristic of different Cas proteins could give rise to different rates of HDR.

Functional limitations of the system also exist, such as the requirement of the presence of a PAM downstream from the desired target site which corresponds to the Cas protein to be employed. As our repertoire of cloned and characterised Cas proteins expands this is likely to be less of an issue.

• This will depend on the nature of the company seeking access and the use for which access is being sought. Non-commercial entities engaging in basic and other non-commercial research will likely be able to take advantage of non-exclusive licences to the technology, for example through plasmids made available on Addgene. Addgene's model for plasmid sharing is detailed on its website here and its CRISPR resources are available here.

• Entities looking to develop commercial applications of CRISPR technology are in a more uncertain position. Pending the outcome of patent office disputes in the US Patent Trial and Appeal Board, European Patent Office and elsewhere in the world, the ownership of the foundational IP in the field is unclear. Further, there is a very real possibility that the outcome of that dispute will vary from jurisdiction to jurisdiction and that a patchwork landscape of protected territories for each of the different parties will emerge. Then there is the further consideration of how the second generation CRISPR patent portfolios (much of which is still in the application stage of prosecution) may cover the intended use of the technology. With at least 604 CRISPR patent families already filed around the world1 potential licensees face very real questions of:

• (a) whether to seek a licence now or wait until the landscape settles and patent portfolios are pooled together; and

• (b) if one seeks to take a licence now, from which party or parties should such a licence be sought – whose patent portfolio is likely to provide the freedom to operate one needs, both now and in the future.

• A still further layer of complexity is the web of exclusive and non-exclusive licences and sublicences over various fields and applications of the foundational CRISPR IP variously granted by the two founder camps (Doudna/University California Berkley (UCB)/Charpentier and Zhang/Broad/MIT/Harvard) to their spin out companies and industry collaborators (including such leading companies as Novartis, Bayer, Regeneron and others).

• Interestingly MPEG LA, the licence pooling and administration firm originally formed to administer the IP behind the MPEG-2 video codec standard, has announced an initiative to offer a one-stop licensing solution for CRISPR applications. Whether this is feasible given the existing commercial relationships and huge investments in play is another story. That said, the need for commercial certainty in respect of such a transformational technology may drive pressure from industry and government on patentees to offer such a solution.

1 See Egelie et al. “The emerging patent landscape of CRISPR-Cas gene editing technology” 2016 Nature Biotechnology 34(10) 1025.

• Both the Broad and UCB have made CRISPR plasmids available to the academic community via the non-profit organisation Addgene, which has a standard Uniform Biological Material Transfer Agreement, under which all materials are distributed.

There are two broad camps in the dispute over the foundational IP covering the CRISPR/Cas9 system currently in use:

• Jennifer Doudna and Emmanuelle Charpentier, co-authors on the seminal Jinek 2012 paper, together with UCB and University of Vienna and their various spin out companies, Caribou Biosciences, Intellia Therapeutics (both associated with Doudna and UCB), CRISPR Therapeutics and ERS Genomics (both associated with Charpentier). On 16 December 2016 Doudna, Charpentier and their associated institutions and start-ups signed a cross-licensing and patent prosecution co-operation agreement which cements their alliance.

• Feng Zhang, author of one of the first papers to demonstrate the use of the CRISPR system in eukaryotes, together with the Broad Institute, MIT and Harvard and the spin out company Editas Medicine (of which Doudna was also a co-founder, until a falling out when the Broad, MIT and Harvard were granted their key CRISPR patent ahead of UCB’s key patent application).

• As at April 2017, the Broad group has 9 granted patents in the EPO and 13 in the USPTO originating from Zhang's group (with another 4 originating from George Church’s group).

• The Doudna group and UCB have yet to have a patent go to grant in the EPO or USPTO, although their key European patent is due to grant on 10 May 2017.

• Many other groups have granted patents related to CRISPR in various jurisdictions around the world, including DuPont, Agilent Technologies, Institut Pasteur, University of Georgia Research Foundation and Toolgen.

A huge number of CRISPR related patent families have been filed across the globe since 2012. The Zhang group at the Broad leads the way with at least 56 different patent application families. By contrast, Doudna/Charpentier/UCB/Vienna have only 13 patent application families. Other significant patent filers include DuPont, the US National Institutes of Health, Cellectis, Sangamo Biosciences and Dow AgroSciences.

These later filings relate to various follow-on applications and particular aspects of the CRISPR technology, but most notably include specific applications of the technology, specific components of the CRISPR system (including newly discovered variants of those components) and mechanisms for the delivery of the necessary genetic components of the system into target cells.

• On 15 February 2017, the Patent Trial and Appeal Board (the PTAB) of the US Patent and Trade Mark Office (USPTO) delivered judgment in the interference proceedings between the University of California Berkeley (UCB)/University of Vienna/Charpentier and the Broad Institute of MIT and Harvard (the Broad). The judgement found that there was no interference-in-fact between the pending UCB application and 12 of the granted Broad patents, and one Broad application. On 12 April 2017, UCB, the University of Vienna and Charpentier filed an appeal to the Federal Circuit from the PTAB’s finding.

• This decision means that the PTAB considered the respective inventions claimed by UCB and the Broad to be sufficiently different and distinct from one another so as to be separately patentable. The Broad was able to convince the PTAB that its invention, being a method for use of the CRISPR/Cas9 system in eukaryotic cells, was not obvious over the invention of CRISPR/Cas9 in any environment (including in prokaryotic cells or in vitro), as claimed by UCB’s patent application. The PTAB considered that although the person of ordinary skill in the art would be motivated to try the CRISPR/Cas9 system in eukaryotes on the basis of UCB’s claims, they would have no reasonable expectation of success in doing so.

• Arguably the PTAB decision simply maintains the status quo between the parties. UCB’s patents still claim earlier priority dates and cover a broader range of applications, whereas the Broad still holds more patents in more jurisdictions than UCB (or any other group). The immediate impact of the decision predominantly related to the public perception of each party: the share prices of Intellia and CRISPR Tx suffered significant falls in the wake of the decision, whereas Editas’ share price rose sharply accordingly.

• Had the interference proceedings been maintained, the PTAB would have then considered whether the Broad or UCB was the first to invent – an analysis which would have come down to an examination of the work each side did to reduce the invention to practice.  This may have resulted in the loss of either side’s patents and patent applications, but UCB would have taken comfort from being the first to file and publish on the CRISPR/Cas9 system. 

• The PTAB decision and its reasoning may not be as influential or reflective of the likely outcome in other tribunals and jurisdictions as some first envisaged. This is due to factors including the limited scope of the PTAB's enquiry, the greater number of prior art references available to support an obviousness attack in proceedings other than an interference action and the different law and procedure in different courts and/or patent offices around the world.

• A&O's in-depth analysis of the decision and its implications for the CRISPR landscape in the short medium and long term is available here.

• The Broad’s eight granted patents are in various early stages of the opposition procedure in the EPO due to the different times they were granted and correspondingly the different deadlines for opposing each of the different patents. In addition, one of the Broad’s patents has only proceeded to grant very recently so the period in which oppositions may be filed has yet to expire. Thirteen different parties have opposed one or more of those granted patents, some openly (including CRISPR Therapeutics and Novozymes) and others using a strawman opponent (whereby the identity of the real party behind the opposition is hidden). The Broad and opponents of the various patents are still in the course of making written submissions of their arguments to the EPO and no preliminary opinion or oral hearing date has been issued for any of the opposed patents.

• UCB’s key patent is due to be granted on 10 May 2017, upon publication of the formal decision to grant in the European Patent Bulletin. Once the patent is granted the nine month opposition period will begin, during which it is likely that several opponents (including the Broad) will file notices of opposition to the patent.

• With good reason, life sciences industry press, companies large and small, and even the mainstream media have been keenly observing what has been billed as the biggest biotech patent dispute in history. The outcome of the dispute in the patent offices will have significant consequences and may come to define the developmental path of this transformational technology.

• The rapid rate of CRISPR-related research and discovery would seem to suggest that the uncertainty over the future control of the foundational CRISPR patents has not deterred the research community. However, commercial entities looking for certainty as to licensing arrangements and IP ownership may be deterred by the present complex and fragmented patent landscape.

• For those seeking to enter the CRISPR field, uncertainty over the ownership of the foundational IP may delay or deter their entry, thus having a chilling effect on innovation and participation in the CRISPR revolution. The lack of one clear winner will lead to a fragmented landscape globally or even within a single jurisdiction, with prospective industry entrants forced to take licences from multiple parties unless some single licensing platform or pool could be formed.

• The PTAB's decision that UCB and the Broad's claimed inventions are separatley patentable leaves open the possibility that ultimately both UCB and the Broad may retain valid CRISPR/Cas9 patents of broad scope which users of CRISPR/Cas9 technology (including UCB and the Broad themselves) will need to licence from each party. Should this come to pass, we may see more commercially attractive cross-licensing options emerge.

• For those who have already taken a licence, they will be keen to know whether they have backed the winner in the patent dispute, and how much their licence and/or collaboration is really worth. Clinical programs being developed in conjunction with a loser in the patent dispute could be jeopardised, and additional licences may need to be sought from the winners.

• For society as a whole, control over the CRISPR foundational patent estate could mean control over the direction of CRISPR research and development. In addition to the potential billions in royalties, the winners will have the power to carve up the exclusive and non-exclusive licensing landscape – reserving the most promising or commercially attractive applications of the technology for themselves and their collaborators and leaving the rest for academic institutions and industry players with non-exclusive licences (or potentially compulsory or Crown use or equivalent licences).

• For IP counsel and their external advisors the dispute will provide plenty to digest, and the way it is fought could set the tone for future breakthrough technologies, either as an example to be followed or a cautionary tale. By the end the case will no doubt have yielded plenty of lessons to be learned which can be applied more broadly, including:

• What is the optimal early filing strategy?
• How should secondary patent protection be developed for biotechnology platforms?
• How should big pharma approach collaborations with startups holding transformational IP (and vice versa)?
• How can the risk from such patent disputes be mitigated or accounted for in licensing/collaboration agreements?