As the CRISPR approach to gene editing moves from the lab to the marketplace, it promises to become yet another example of a disruptive innovation. At first, such innovations take hold because they offer so much convenience that it hardly matters whether they lack all the advantages of more established (and highly developed) alternatives. Then, having democratized a pursuit once confined to a select few, a disruptive innovation gathers strength, simultaneously building and benefiting from a diverse constituency of users and developers. Ultimately, this constituency is what makes an innovation truly disruptive. Users and developers essentially take over, enhancing the original innovation, pushing it toward increasingly demanding applications.
In the case of CRISPR, which stands for Clustered Regularly Interspaced Short Palindromic Repeats, the initial appeal is that it makes gene editing fast, easy, and cheap. A CRISPR-based system, even in the hands of relative newcomers, may home in on nearly any gene of interest. It allows users to avoid resorting to alternative technologies such as TALENs (transcription activator-like effector nucleases) and ZFNs (zinc-finger nucleases).
Both of these alternatives rely on proteins to recognize particular DNA sequences, unlike CRISPR, which relies on RNA. While TALENs and ZFNs are well characterized and capable of great specificity, they oblige researchers and developers to engineer new proteins in order to target new DNA sequences. With CRISPR technology, one need only prepare a new scrap of RNA, called guide RNA. This, needless to say, is a relatively trivial task. It can be accomplished in days or weeks rather than months or years.
And so CRISPR has democratized gene editing. But fledgling democracies tend to have growing pains, and CRISPR is no exception. CRISPR, it is often noted, may give rise to a wide range of off-target effects. That is, the guide RNA used in CRISPR gene editing may lock onto a stretch of DNA that is only approximately (not perfectly) complimentary, and a CRISPR-associated protein (the Cas9 nuclease) may snip DNA in unfortunate places, silencing genes meant to be left alone, or even permitting unintended additions of genetic material.
Where CRISPR originated—as an adaptive immune system for bacteria—it may not matter terribly if CRISPR-Cas9 complexes are a bit overzealous in seizing and cutting apart DNA from infectious viruses. But where CRISPR is headed—the creation of model organisms and the development of therapeutics—an extremely high degree of specificity is mandatory.
At present, such specificity is available with TALENs and ZFNs, which have been refined for decades and are beginning to show promising clinical results. Good news, it would seem, for them. They may, however, find that they are becoming sustaining innovations.
Both sustaining innovation and disruptive innovation are terms of art coined by Clayton Christensen, a professor at Harvard Business School who is best known for writing The Innovator’s Dilemma. According to Christensen, sustaining innovations risk becoming isolated in their own value networks while disruptive innovations create new ones that ultimately prove larger and richer. Prominent examples of sustaining/disruptive pairs include telegraphy/telephony and mainframe computing/PC computing.
In either case, the “sustainer” and the upstart “disruptor” had their respective advantages. Yet the sustainer’s advantage was subject to erosion, while the disruptor’s advantage was difficult, if not impossible, for the sustainer to emulate. (For example, telegraphy had unequaled geographical reach, whereas telephony introduced voice communications.) In the case of CRISPR, a disruptive advantage (other than speed and ease of application) is the ability to simultaneously introduce mutations in multiple target genes. In general, this kind of efficiency eludes TALEN-based and ZFN-based platforms.
The Sustainers Strike Back
So, in the face of a disruptor’s challenge, what is a sustainer to do? One possible course of action is FUD, the cultivation of fear, uncertainty, and doubt. In a FUD campaign, a sustainer might highlight a potential disruptor’s shortcomings. As it turns out, CRISPR may be vulnerable to FUD. For example, executives of TALEN- and ZFN-promoting companies have been quick to note CRISPR’s specificity challenges.
Other responses available to sustainers are co-option and accommodation. Some mixture of the two may account for a quote that appeared April 8 in Nature Biotechnology, in an article entitled “Gene Editing at CRISPR Speed.” This article reports that Jean-Charles Epinat, deputy CEO of Cellectis bioresearch, a provider of custom-designed TALENs, said, “People will start evaluating their projects with CRISPR, screen targeted sequences, make their proof of concept, and then, probably, will come back to ‘cleaner’ nucleases like TALENs for the ‘real’ work.”
The same article quotes Gregory Davis, principal scientist at Sigma Aldrich in St. Louis, as saying, “For people who want something fast and cheap, CRISPR is a good option. If people want to go down a path that has more of a proven track record, then ZFNs are the way to go.”
Disruption Lives
Despite such sentiments, CRISPR still inspires. It has been taken up by thousands of laboratories. But will the enthusiasm last? That depends on how successful researchers and (increasingly) commercial organizations are in improving CRISPR technology. Curiously, to address specificity concerns and lessen the problem of off-target effects, CRISPR researchers borrowed an idea from ZFN and TALEN systems, which rely on dual DNA-binding modules. For example, a research group led by Feng Zhang, Ph.D., a neuroscientist at MIT’s McGovern Institute for Brain Research, combined a Cas9 nickase mutant with paired guide RNAs to introduce targeted double-strand breaks.
The Cas9 enzyme may be modified to contain a single inactive catalytic domain, resulting in a so-called nickase. With only one active nuclease domain, the nickase cuts only one strand of the target DNA, creating a single-strand break. If dual nickases are deployed, each may be targeted with a different guide RNA. Such nickases may be designed to nick opposite strands of DNA, with the nicks so close to each other that they effectively act as a double-strand break.
The nickase approach developed by Dr. Zhang’s group was described in an article (“Double Nicking by RNA-Guided CRISPR Cas9 for Enhanced Genome Editing Specificity”) that appeared August 29, 2013, in Cell. The authors reported that they were able to use the double nicking of DNA to “reduce off-target mutagenesis by 50- to 1,000-fold.”
A Closer Look at Nickases
Since then, several researchers have explored double-nicking while introducing their own refinements. For example, two independent research groups recognized a key limitation of the original double-nicking approach, and they developed similar workarounds.
The problem is that each component of a dual-nickase system remains catalytically active and may create weakly mutagenic single-strand breaks, which occasionally result in unwanted off-target mutations. If each component in a dual-nickase system is, essentially, monomeric, one solution would be to create a truly dimeric system. Basically, the idea is to create monomers that are catalytically active only when they are part of a larger complex. Then, cleavage of DNA would depend on dimerization, not mere co-localization.
One research group that found a way to exploit dimerization was led by Keith Joung, M.D., Ph.D., associate professor of pathology at Harvard Medical School and director of the Molecular Pathology Unit at Massachusetts General Hospital. This group published its findings April 25 in Nature Biotechnology, in an article entitled “Dimeric CRISPR RNA-guided FokI nucleases for highly specific genome editing.”
The authors described how the cleavage activity of dimeric RNA-guided dead Cas9 (dCas9)-FokI nucleases, named RFNs, depends on the binding of two guide RNAs to DNA with a defined spacing and orientation. The published findings show robust activities compared to monomeric wild-type Cas9 nucleases and nickases, in addition to improved genomic specificities with no evidence of unwanted mutagenesis at off-target sites.
The paper indicates that Dr. Joung is also affiliated with Editas Medicine and Transposagen Biopharmaceuticals. He is, in fact, a cofounder of Editas Medicine and a member of Transposagen’s scientific advisory board. Transposagen indicates that Dr. Joung’s work has contributed to the company’s NextGEN CRISPR technology, which it characterizes as the first truly dimeric RNA-guided FokI nuclease.
By incorporating FokI, an enzyme that functions only when two copies of the molecule are paired, the technology essentially doubles the length of DNA that must be recognized for cleavage, substantially improving the precision of genome editing.
A similar approach was developed by a group of researchers led by David R. Liu, Ph.D., a Howard Hughes Medical Institute investigator and professor of chemistry and chemical biology at Harvard University. (Dr. Liu, like Dr. Joung, is a founder of Editas Medicine.) Dr. Liu’s group published its results April 25 in Nature Biotechnology, in a paper entitled “Fusion of Inactivated Cas9 to FokI Nuclease Improves Genome Modification Specificity.”
Dr. Liu’s group reported that it had engineered a dimerization-dependent Cas9-FokI nuclease fusion, named fCas9 that requires the simultaneous DNA binding and association of two FokI-dCas9 monomers to cleave DNA. The engineered FokI-dCas9, said the researchers, modified target DNA sites with efficiency comparable to that of nickases, but with greater than 140-fold higher specificity than wild-type Cas9 in human cells.
The specificity of fCas9 was at least four-fold higher than that of paired nickases at human genome loci with highly similar off-target sites elsewhere in the genome. Additionally, unwanted binding was reduced further by the requirement that only sites flanked by two guide RNAs approximately 15 or 25 base pairs apart are cleaved.
Other Alternatives
Editas Medicine and Transposagen are not the only commercial organizations boasting ties with prominent CRISPR researchers. CRISPR Therapeutics, a biopharmaceutical company that just raised $25 million in a Series A investment from Versant Ventures, has announced a founding team comprising experts in diverse fields of science including CRISPR-Cas9, genome editing, stem cell biology, advanced drug delivery technologies, RNA interference, and gene silencing.
One of the company’s scientific founders is Emmanuelle Charpentier, Ph.D., a professor at the Helmholtz Centre for Infection Research and Hannover Medical School, Germany and the Laboratory for Molecular Infection Medicine at Umeå University, Sweden. Of late, Dr. Charpentier has been investigating the evolution of CRISPR-Cas systems, including systems that could serve as alternatives to the usual CRISPR system, which relies on Cas9 from Streptococcus pyrogenes.
In Nucleic Acids Research, in a paper printed in February with the title “Phylogeny of Cas9 determines functional exchangeability of dual-RNA and Cas9 among orthologous type II CRISPR-Cas systems,” Dr. Charpentier’s team reported that evolution of dual-RNA and Cas9 in bacteria produced remarkable sequence diversity. Moreover, the team demonstrated that these two components are interchangeable “only between closely related type II systems when the PAM sequence is adjusted to the investigated Cas9 protein.” The reported collection of dual-RNA:Cas9 with associated PAMs, added the researchers, “expands the possibilities for multiplex genome editing and could provide means to improve the specificity of the RNA-programmable Cas9 tool.”
Beyond Specificity
Even though researchers are continuing to improve the specificity of CRISPR-based systems, other challenges are still outstanding. For example, such systems will only prove effective in therapeutic applications if effective delivery mechanisms are developed. That is, genetic changes will need to be induced in all of the cells that need it, or at least a sizable subset of targeted cells. Doing so via plasmid transfection, a common delivery method can be challenging, particularly in certain cell types such as stress-sensitive human embryonic stem cells. Delivery vehicles could be adapted from gene therapy applications, but these have had only mixed success.
Nonetheless, if delivery challenges are attacked with as much gusto as specificity challenges, CRISPR may continue its advance from laboratory tool to therapeutic modality. Also, in Christensen’s language, it may move progress from “low-end disruption” to a “high-end disruption.”