Diversification—having railroads, hotels, and houses in prime locations—is one strategy for winning the board game Monopoly. Similarly, successful commercialization of technology may lie in extending it across different end uses.
Zinc finger proteins (ZFPs) are often called “game changing” because of the unprecedented way they precisely modify genes. Excitement about them is mirrored in the number of related scientific publications, which have climbed from hardly any 20 years ago to more than 360 in 2009, according to Chemical Abstracts Service. CAS data also show that U.S. patents and patent applications have risen from just a handful in the late 1990s to about 50 per year over the past decade.
Most of the intellectual property is in the hands of Richmond, Calif.-based Sangamo BioSciences—so much so that no one in the field can point to another company that competes directly in the ZFP space. Sangamo’s goal is to bring this distinctive technology to bear on therapeutics, drug production, agriculture, and basic research. Customers have translated the technology into products on the market, but so far Sangamo has none of its own.
Biotech executive Edward Lanphier founded Sangamo in 1995 after licensing patents from Johns Hopkins University, Massachusetts Institute of Technology, and Scripps Research Institute. He firmed up Sangamo’s hold on intellectual property with more university licenses and the $30 million acquisition in 2001 of Gendaq, a spin-off of the U.K.’s Medical Research Council founded by Nobel Laureate Aaron Klug.
As Lanphier recalls, the science was still at an early stage when he started accumulating technology, and academic technology-transfer offices were happy to strike deals. After many years of effort designing, making, and using ZFPs, Sangamo now has licensed or owns some 257 U.S. and foreign patents and 311 patent applications. It’s largely within the past few years that the technology has shown commercial promise.
“As the technology matured, the breadth and generality became clearer, and we saw that it could be applied essentially to any gene in any cell type in any organism,” Lanphier explains. “It’s a technology that is agnostic to the gene target and agnostic to the host.” Sangamo focuses its own efforts on using ZFPs to regulate and modify human genes for therapeutic needs, but it has reached out to Sigma-Aldrich and Dow AgroSciences to expand into research, bioprocessing, and agriculture markets.
Zinc fingers are proteins of about 30 amino acids in length, coordinated to a zinc ion, that recognize and bind specifically to DNA base-pair sequences. Linking three or four fingers creates an array that associates with a nine- to 12-base-pair target site. A pair of arrays will recognize sites that are likely long enough to be unique within a genome.
“The structure of these proteins always suggested that they could be engineered,” Lanphier says. Tinkering with the structure and mix of fingers alters the specificity and binding affinity of the arrays. The magic touch happens when an array is coupled to a functional entity, such as the bacterial FokI DNA-cutting domain, to form a zinc finger nuclease (ZFN). When a pair of ZFNs bind to DNA, the FokI pieces join like the halves of scissors to cause a double-strand break.
Enzymes in the cell try to fix the broken DNA, a process that often leads to mutations that knock out or inactivate genes. If donor DNA is supplied, cellular mechanisms will insert it to repair or add gene function. Although snipping and repair events take place naturally, ZFNs increase the frequency with which they occur by at least 1,000 times.
In essence, researchers can decide ahead of time what gene they want to alter, create a relevant ZFN, and then make that and only that change in meaningful amounts. The specificity and precision run counter to traditional methods of random mutagenesis and the insertion of transgenes. Because the process also works via the delivery of a ZFN-encoding DNA plasmid or mRNA transcript, intact cells or whole organisms can be modified directly.
Sangamo’s expertise in protein engineering and optimization has created a very general, automated approach to build highly effective ZFPs, Lanphier says. Its library of proteins, linkers, and functional domains, along with the rules for combining them, is proprietary.
Wanting to use ZFPs but frustrated by an inability to rapidly construct them, academic researchers have developed their own tools and disseminated the know-how. The different methods have had varying degrees of success. At Scripps, Carlos F. Barbas III and coworkers have set up a Web-based resource called Zinc Finger Tools. The software helps users determine valid targets, predict binding, and design ZFPs from an experimentally characterized database.
The Zinc Finger Consortium, with more than 15 academic members, was launched about five years ago to share information among groups. One of the consortium’s goals is to develop robust, user-friendly, freely available methods to overcome bottlenecks in ZFP engineering. It has launched a database and a software package and has offered protein modules through the nonprofit group Addgene.
In 2008, the consortium published a combinatorial method called OPEN, or oligomerized pool engineering (Mol. Cell 2008, 31, 294), in which customized arrays can be created from preselected zinc finger pools and subsequently tested for binding and activity. OPEN pools are supplied by the lab of consortium leader J. Keith Joung at Massachusetts General Hospital (Nat. Protoc. 2009, 4, 1471).
Although OPEN’s results are better than those of previous modular methods, the process still is technically challenging, says Daniel Voytas, director of the University of Minnesota’s Center for Genome Engineering and a consortium founder. “It takes a well-trained scientist three or so months to go from start to finish, but the end product is a high-quality reagent that has proven to work well in a large number of organisms,” he says.
In the past 18 months, numerous groups have started using the method. “Through our collective expertise, the protocols will get refined and streamlined and, as the body of ZFPs grows and we better understand the rules of how they interact with DNA, it will get simpler for academic scientists to make these proteins,” Voytas adds.
By and large, academics working with zinc fingers aren’t concerned about using patented technologies, according to recent analysis by Duke University law school professor Arti Rai and collaborators (Nat. Biotechnol. 2009, 27, 140). As in other areas, most scientists assume that companies won’t bother to come after them when the technology is employed in research. In this regard—and in its unwillingness to disclose proprietary know-how—Sangamo’s behavior, Raj points out, is typical of corporations.
Indeed, commercial and open-access interests have coexisted for many years in an interconnected community where joint meetings are common and many academics are inventors on Sangamo-licensed patents. “Patents are meant to create barriers to entry from a commercial perspective, and that is where we really want to rely on them,” Lanphier says. “While we are more than willing to, and certainly do, collaborate, we have made efforts to make the technology available broadly.”
David Smoller, president of Sigma-Aldrich’s Research Biotech business unit, had followed the technology for more than a decade, but it wasn’t until about 2006, he says, that it was becoming robust enough for Sigma to consider producing reagents. In July 2007, Sigma agreed to pay Sangamo up to about $40 million for rights to develop and sell zinc finger research reagents. Sangamo gets royalties on sales and a share of sublicensing revenues.
Meanwhile, scientific work showed the utility of zinc fingers in editing genes in insect, plant, and human cells. In late 2008, Sigma launched its CompoZr platform for providing customized ZFNs. Give Sigma about eight weeks and $30,000, and it will supply validated ZFNs designed for a specific gene target. Sigma has moved toward selling kits for gene insertion. It’s also readying off-the-shelf reagents, at about one-third the cost, for commonly studied gene targets and pathways.
Process improvements should reduce the cost of reagents, Smoller says, but at the same time, the technology can be employed in less expensive downstream products, such as engineered cell lines and transgenic research animals.
Sangamo previously had nonexclusive licenses with several major biotech and drug firms that wanted to improve their cell lines for manufacturing. In 2009, Sigma and Sangamo expanded their agreement to include rights to develop and sell transgenic animals and cell lines for commercial production. The deal brought Sangamo another $20 million up front, $25 million in potential milestones, and royalties.
Smoller has been surprised by how fast the technology is being adopted. For example, in transgenic animals, he says, “I didn’t think it was going to work right out of the box.” Within about six months of establishing its Sigma Advanced Genetic Engineering (SAGE) Labs early last year, the company and its collaborators had created the first knockout rats.
Knockout rats are desirable models of human diseases, but generating them had eluded scientists. The rats were produced through the single-step microinjection of ZFNs into developing embryos (C&EN, July 27, 2009, page 15). The desired mutations are permanent and heritable, and they can be generated in about four months—or less than one-third of the time it takes by conventional embryonic stem cell methods.
“It’s the fastest I’ve seen any technology go from start to finish,” Smoller says. The process is extremely efficient, he adds, and should help speed up drug R&D. SAGE has a partners program to evaluate and validate new rat models that target more than 20 genes, including those related to Parkinson’s and Alzheimer’s diseases, schizophrenia, and obesity. Sigma also is offering gene-modified rats to help predict drug absorption, metabolism, and toxicity.
Before Sigma, Sangamo’s first major licensing agreement was with Dow. In late 2005, the firms signed a three-year research pact to test the technology in plants. After meeting some “pretty challenging technical milestones,” Dow opted for a commercial license six months ahead of schedule, says Vipula K. Shukla, scientific leader for Dow’s ExZact Precision Technology business. So far, Dow has paid Sangamo about $32 million in fees and milestones.
In May 2009, two groups—one of Dow and Sangamo scientists and another led by Voytas—published papers reporting their initial success in directly modifying plant genes (Nature 2009, 459, 437 and 442). Targeted changes for herbicide resistance were made more precisely and much faster than could be done by the usual methods, Shukla explains.
Having moved past proof of concept, Dow is now deploying the technology in its own agbiotech products and is sublicensing it through ExZact to others for use in plants, algae, and fungi. Dow is the only provider of zinc finger reagents in the plant biotech industry, says Sharon Berberich, ExZact business development leader.
Shukla and Berberich attest to an enthusiastic response in the form of requests for collaboration and access to the technology. Shukla says Dow continues to work with Sangamo to find new applications and better ways of deploying the technology, and Berberich says Dow recognizes that it will need to find a third-party supplier to make reagents.
Although Sangamo strikes license agreements for near-term gain, its long-term goal is to develop drugs based on zinc finger technology. Its most advanced candidate is a ZFP that activates the VEGF-A gene to try to improve nerve health in diabetic patients. The company has started a Phase IIb trial with $3 million in support from the Juvenile Diabetes Research Foundation. It is studying the same ZFP in a Phase II amyotrophic lateral sclerosis trial.
Through gene modification, Sangamo is also trying to knock out a receptor used by HIV to infect T cells. The company’s HIV work is supported by a $14.5 million grant from the California Institute for Regenerative Medicine and $500,000 from the Bill & Melinda Gates Foundation.
Similarly, it is trying to eliminate the glucocorticoid receptor in cytolytic T cells engineered to fight a form of brain cancer. The company also has preclinical programs in neuropathic pain and Parkinson’s and in repairing DNA in some monogenic diseases, such as sickle cell anemia.
Sangamo believes ZFPs may work against a range of diseases because of their novel mechanisms of action. Along with clinical proof, development challenges will include effective delivery, expression, and the reduction of negative effects from off-target events. Potential products also will have to compete against small- and large-molecule drugs, gene therapy agents, and RNA interference technologies that also seek to regulate gene expression.
In addition, two younger firms, France’s Cellectis and Precision BioSciences, in Research Triangle Park, N.C., have been trying to engineer naturally occurring meganucleases to target and specifically cleave DNA. Like Sangamo, they are addressing the agbiotech world and have research collaborations with Monsanto, BASF, Bayer, and DuPont. Last month, Cellectis launched a plant sciences subsidiary in Minneapolis and named Voytas its chief scientific officer.
With significant income from grants and licenses, $21 million raised last year in a stock offering, and $85 million in the bank, Sangamo will continue to operate in a “fiscally conservative way,” Lanphier says. “Our hybrid model has not only given us a very different way of developing the science and getting it out to the scientific community but has allowed us to drive our therapeutic programs forward.” Lanphier anticipates eventually finding large pharma partners to help with late-stage development and commercialization.
An alliance with a large drug company would be a significant validation of Sangamo’s therapeutic approach, says Edward Tenthoff, senior research analyst with the investment bank Piper Jaffray. He believes the company faces challenges as it tries to follow an uncharted path through clinical and commercial development.
Although the firm’s drug targets offer big opportunities, Tenthoff is not convinced—in the case of the diabetic neuropathy agent, for example—that the data show Sangamo’s candidates are making a clinical difference. In the HIV work, he calls the firm’s ex vivo process of extracting, modifying, and reimplanting a patient’s cells “cumbersome” for practical use. Sangamo says it is working on in vivo approaches as well.
“The company has actually done a very good job of monetizing the ZFP assets in nontherapeutic applications and has been very thrifty in their consumption of capital,” Tenthoff says. Longer term, he sees “real revenue opportunity” if the Dow and Sigma businesses take off and generate further milestone, royalty, and sublicense payments.
If Sangamo succeeds, the universities and possibly the academic inventors on licensed patents would benefit as well. Lanphier calls the agreements “classic university licenses,” which include up-front payments or equity in the company, as well as possible milestone payments and royalties on commercial sales.
Sangamo and its licensees may come into closer competition with open-access sources as methods improve, Duke’s Rai notes, although commercial applications likely will require licenses to intellectual property controlled by Sangamo. Whether scientists engineer their own ZFPs or pay others to do it, there’s a general sense that these reagents are moving quickly into use and will become standard tools in industrial and academic labs. For Sangamo and the rest of the zinc finger community, the focus is shifting from how to make the materials to what can be accomplished with them.