By Malorye Branca
July 15, 2003 | It's so quick, easy, powerful, and cost-effective, RNA interference (RNAi) almost sounds like the too-good-to-be-true subject of a late-night television infomercial.
Certainly, in labs desperately hunting for better ways to study gene function, these tiny, exquisitely specific gene-silencing molecules seem nothing short of miraculous. RNAi promises to transform investigations of gene function in organisms from microbes to mammals.
RNAi rapture has also consumed the biotech and venture capital community like wildfire, inspiring resurgence in a previously struggling sector of the biotech industry that sees unlimited potential for RNAi in the clinic. A recent article in Fortune heralded "Biotech's Billion-Dollar Breakthrough." In 2002, Alnylam Pharmaceuticals, founded by a bevy of RNAi pioneers including Nobel laureate and Biogen founder Phillip Sharp, raised $17 million in early financing. And earlier this year, just a week after changing its name from Ribozyme Pharmaceuticals, reborn Sirna Therapeutics netted a cool $48 million in venture capital.
The color purple: The phenomenon of "co-suppression" was first described in petunias by Richard Jorgensen and Carolyn Napoli.
But now the tough questions loom large: Will RNAi prove to be the missing link needed for genomic drug discovery? And must RNAi-based drugs endure the same type of roller-coaster ride that earlier technologies, such as monoclonal antibodies and RNA antisense, endured? Or will past experiences in gene therapy and antisense make for an easier transition?
The first hints that there was a natural way to suppress genes emerged from the fertile field of plant genomics. Richard Jorgensen, currently at the University of Arizona, stumbled upon this realization in the late 1980s, when he tried adding more of a "purple" gene to deepen some petunias' purple hue. The flowers turned white instead.
The underlying phenomenon — RNAi — was described in 1998 by Andy Fire and Craig Mello at the Carnegie Institute in Washington, D.C. The researchers showed that small interfering double-stranded RNA (siRNA) fragments could specifically target and eliminate natural messenger RNA molecules.
Early RNAi work focused on plants and lower animals such as worms. Then, two years ago, a paper in Nature revealed that RNAi also worked in mammalian cells (Elbashir, S.M. et al. 411: 494-498; 2001). The precise role of RNAi in nature is still vague, but thousands of scientists are applying the tool to examine the effects of specifically shutting off genes. These experiments can take as little as a week — lightning speed compared to the months required to knock out a single gene in mice.
But Not Quite Magic
Despite its promising start, RNAi still has some rough edges. At Gene Expression Systems' recent RNAi 2003 meeting in Waltham, Mass., Joanne Kamens from Abbott Bioresearch Center described her struggles to use the technique in macrophages. These immune cells are notoriously resistant to the introduction of foreign genetic material. "Everyone agreed, surprisingly quickly, that we should do this, but then it wasn't as easy as we hoped," Kamens said.
Working in collaboration with Sequitur, the Abbott group was eventually able to shut down certain genes, but curiously the proteins they were trying to eliminate were still being expressed at high levels in the cell. "One possibility is that there are redundant mechanisms, and that shutting down one gene actually leads to increased protein production by other means," Kamens said.
Producing effective siRNAs — the actual gene-silencing molecules — can also be difficult. Companies including Dharmacon and Cenix BioScience are trying to work out the rules for what makes a good siRNA.
"Not all genes silence equally," said Stephen Scaringe, chief scientific officer at Dharmacon, who also spoke at the Waltham meeting. Certain sequences are just more difficult to shut down by RNAi. "You can change the siRNA by one base and suddenly gain or lose functionality," he said. Meanwhile, some siRNAs can interact with parts of the cellular machinery beyond their intended targets.
Single nucleotide polymorphisms (SNPs) and alternative splicing can also cause problems.
People who use poorly designed silencing molecules have a hard time getting good results. "Adding more bad siRNA does not make it work better," Scaringe said. Scientists are therefore relying on bioinformatics to help design the molecules and categorize them: Some siRNAs work, but poorly, while others are "superfunctional." Dharmacon's scientists have also found that using a specific set, or pool, of many siRNAs works best.
|A snapshot of drug-discovery companies now working with RNAi.
More drug companies are catching on to the "intelligent design" approach to siRNAs: Last month, Dharmacon announced a deal with Exelixis to produce a library of siRNA reagents targeted against 600 top drug candidates, including kinases. The library will be used for "high-throughput functional genomic studies designed to characterize drug targets and pathways in mammalian model systems," according to a joint announcement.
Pitfalls remain, but many companies are streamlining the technique, quickly making it easier.
But Can You Make Drugs with It?
RNAi is not just a spectacular lab tool — it offers a tantalizing mechanism for producing actual drugs as well. No sooner was RNAi's utility in mammals demonstrated than investors started pouring money into RNAi-based companies (see "RNAi-Fever Heats Up Novel Drug Category Funding," Oct. 2002 Bio·IT World, page 11). A string of recent successes in using RNAi to attack viruses such as HIV and hepatitis B and C in animal models has kept investors salivating.
One of the more glamorous startups in the field is Alnylam, with its advisory board of RNAi luminaries including David Bartel, Thomas Tuschl, Phillip Zamore, and Greg Hannon. The company is working on RNAi-based drugs for cancer, infectious diseases, and inflammation.
German biotech Cenix BioScience raised $4.9 million last year and also plans to make drugs. Cenix is working with Ambion to develop validated synthetic siRNAs for every gene in the human genome, according to David Brown, senior R&D scientist at Ambion.
Even more dramatic is the number of established companies suddenly embracing RNAi. Groups that have worked through some of the devilish details surrounding the use of other RNA-targeting technologies, such as antisense and ribozymes, have jumped on the RNAi bandwagon. Colorado-based Ribozyme Pharmaceuticals had nearly dissolved because its ribozyme drugs weren't passing clinical milestones. But applying its RNA insights to RNAi, capped off with a new name — Sirna Therapeutics — resurrected investor interest.
To produce RNAi-based therapeutics, companies need to develop both a molecule to specifically silence genes and a way to deliver that molecule inside the targeted cell. Several firms have one of those ingredients and are looking for the other.
Intradigm and Sequitur, for example, have just signed what will undoubtedly be one of many deals involving cross-licensing of RNAi technologies. Intradigm now has access to Sequitur's Stealth RNAi in certain disease areas, while Sequitur can use Intradigm's Targeted Synthetic Gene Vector technology.
Sequitur is an experienced service provider that uses antisense and RNAi to help drug companies confirm that they are hitting good molecular targets in the body with their new drugs. Stealth RNAi is a modified form of siRNA that is "more effective, more stable in serum, and less likely to generate an interferon [immune] response," according to Sequitur President Tod Woolf.
Intradigm, meanwhile, has several staffers from former Novartis subsidiary Genetic Therapy, where they were also tackling the daunting problem of introducing genetic material into human cells. Their solution for delivering siRNA is both technological and practical: Besides using a special chemical formulation that protects the drug as it is entering the body, they target contained areas, such as the eye, the joint space, and tumors.
Both companies want to make their own drugs as well as work with collaborators. "Our philosophy is to find groups to work with, rather than to fight," says Martin Woodle, Intradigm's founder and chief scientific officer.
But not everyone is going to work things out through deals. "If RNAi turns out to be as valuable as people anticipate, there will inevitably be litigation," says Irene Abrams, technology licensing officer at MIT, which is part owner of one of the key siRNA patents. The number of RNAi-related patents has, naturally, proliferated as the field heated up. Many of the major patents are pending, and many may overlap.
Because of the economic climate, the need for quick results, and the uncertainty over intellectual property, several deals are being negotiated. "With patents, you can never proceed with certainty until they are issued," Woodle says. Still, some bitter contests are likely. With so many players and so much at stake, "It's hard for people to compromise or fairly determine the value of their own intellectual property," Abrams says.
Even if the thorny patent issues are resolved, companies still have to get drugs into the clinic and then to the market. "Anybody who has experienced any kind of drug discovery and development realizes that there will be surprises ahead [for RNAi]," cautions Frank Bennett, vice president at Isis Pharmaceuticals.
Woolf believes the key to keeping the field hot long enough for bona fide drugs to get made is keeping the science high-quality. During the peak of antisense fever, he says, "Every type of chemistry tried was claimed to work in animals." But many of those results turned out to be spurious. Such "bad" science can proliferate when many inexperienced investigators are trying a new technique. Some companies also pushed hard too early with antisense- or ribozyme-based products that really weren't ready for the clinic. The result was a backlash, and only now is antisense technology getting a fresh chance in clinical trials.
Things could go differently with RNAi. Sequitur's Woolf points out: "The academic researchers who discovered [RNAi] are extremely high-grade scientists, and the work that has followed has been carefully done too."
And a decade of experience with related RNA technologies will prove invaluable as well. "I think antisense has helped RNAi a lot, in showing what the rate-limiting steps will be and how to solve some of the pharmacokinetic issues," Bennett says.
Although Isis is studying RNAi, Bennett is bullish on antisense. And with about a dozen antisense-based drugs still working their way through clinical trials, the older technology could yet eclipse any possible wave of RNAi-based drugs. Both techniques face the same major hurdle — delivery into the appropriate cells. But once that is solved, it would appear that RNAi will have finally arrived.
RNAi roundabout: Double-stranded RNA molecules are processed into small interfering RNAs (siRNAs) by the enzyme Dicer. These siRNAs pass to the RISC (RNA-induced silencing complex), where they become unwound and activated. Gene silencing probably occurs via several mechanisms, including RNA degradation, translational inhibition, and chromatin remodeling.
Adapted with permission from G. Hannon, et al., Nature 418, 244-251; (2002) (c) Nature Publishing Group