In Sujay Singh's cartoon, shown to participants at the recent RNAi conference in Waltham, Mass., an impish red "Dicer" ball has its arms wrapped around the shoulders of a smiling electric blue character representing a half-ladder of siRNA. The siRNA half-ladder reaches its doughy arms out to grab a green mRNA character, and drags it to a roundish fellow, RISC, which is wielding a large pair of scissors to chop up the mRNA.
"I wanted to show how simple siRNA is," said Singh, president and CEO of Imgenex. The Dicer enzyme splits up the interfering RNA (or RNAi) into short 21-25 base-pair strands of siRNA, then the siRNA-protein complex is triggered to chop up the targeted mRNA at a certain places, inactivating it.
But Singh and others at the conference spent much of their time discussing what is not so simple about siRNA-how to deliver it effectively to cells so it can become a viable therapeutic. Imgenex has tried splicing the siRNA into retrovirus-baed plasmids and non-viral plasmids, as well as adenoviral vectors. The recombinant adenoviruses have been used "to reduce gene expression both in vitro and in vivo." So far, the company has been able to develop these viruses within two weeks.
Ken Reed, CEO of Australian RNAi company Benitec, noted that vectored delivery can "induce a series of artifacts." He discussed the company's work with lentiviruses to deliver RNAi to hippocampal neuron cells. These viruses, he said, do not have to be injected into the nucleus of the cell, but can just be injected in the cell. With more work on delivery vectors, Reed believes that researchers will find far more RNAi druggable targets than small molecule targets.
Reed also noted, however, that confirmation of gene suppression results is not always as easy as it looks. Northern blot analysis does not always work, and in using RT-PCR, "one must be very, very careful what region of the [sequence] you target."
Joanne Kamens of Abbott Research discussed her lab's work using RNAi for the past two years, which she depicted with a PowerPoint slide of a cartoon RNAi roller coaster. "We had a lot of problems getting reagents," she said. There are not enough protein reagents to test whether synthesis of a particular protein has been inhibited, and this shortage "has turned out to be quite a stopping point. In addition, transfection of recombinant plasmids with siRNA has not always worked as Kamens' group planned, although reagents from Sequitur have enabled the group to get small duplexes into cells within 48 hours. Still, some genes prove resistant to silencing, despite the use of different siRNAs. "I've had one gene where 25 different duplexes haven't been able to knock it down, much less out," she said. Additionally, there is the problem of "promiscuous silencing," where a duplex silences proteins other than the intended target.
However, Kamens noted, the technology has overall "worked the way we had hoped." She revealed that the group has found one unique siRNA that can knock out a particular protein but does not affect cytokines, intercellular signaling proteins. But she said the company can't talk about it further.
For those frustrated by siRNAs that don't work, Dharmacon CEO Steven Scaringe presented on the Lafayette, Colo., company's rational siRNA design methods. The company says that its Smart Selection tool designs siRNAs with a 99.99 percent chance of suppressing the activity in your chosen gene by at least 75 percent. The company does not reveal exactly how Smart Selection works, but Scaringe shed some light on the algorithms Dharmacon uses: through studying about 360 RNAi duplexes on four different genes, the company has honed 34 selection criteria. These criteria expand on a handful of rules laid out by siRNA inventor Tom Tuschl, (see http://www.mpibpc.gwdg.de/abteilungen/100/105/sirna.html). The researchers use weighted scoring to enhance the probability of silencing. The company then also pools four selected siRNAs for each target, in a manner that is also proprietary. "Mimicking natural siRNA pools [offers] much higher specificity," Scaringe said.
While the issue of whether selection algorithms actually enhance the utility of siRNA is controversial, Dharmacon must be doing something right, because the company is expanding rapidly. Dharmacon is currently moving into a 40,000-square foot facility that will have more room for manufacturing, as well as more office space (for new hires such as marketing VP Mike Deines), company executives told GenomeWeb. At the current facility, the work has to be divided into shifts along a 24-hour cycle in order to keep up with current demand, the executives said.
Nagesh Mahanthappa, who is director of corporate development at another rapidly growing company, siRNA therapeutic developer AlNylam, presented on the company's latest work with siRNA in mouse models, and the challenges of effective delivery to target cells. He outlined three leading delivery methods that have been discussed in the scientific literature, and the key publications on delivery of RNAi in the mouse model: the hydrodynamic method, which has been published on by McCaffrey (Nature 2002 Jul 4;418(6893):38-9), Lewis (Nature Genetics 2002 Sep;32(1):107-8), and E. Song (Nature Medicine 2003 Mar;9(3):347-51); lipid-based delivery, which has been reported by Sorenson (J.MB 2003 Apr 4;327(4):761-6); and naked siRNA, which Makimura and colleagues described in a recent article (BMC Neurosci. 2002 Nov 7;3(1):18).
Alnylam has found that hydrodynamic delivery "delivers the most robust results," Mahanthappa said. So far, company scientists have been able to knock out a particular gene with siRNA, and see a 50 percent drop in blood glucose levels. But for this type of siRNA to translate into a therapeutic, problems including off-target activity, including methylation of DNA, will have to be overcome, he said.
Later, Mahanthappa said that hydrodynamic delivery would have no clinical efficacy, and that the company is investigating in parallel methods of delivery that include conjugation of the siRNA to chemical moieties that can be metabolized by the liver, as well as polymer and other methods.
Alylam has also attracted attention for its IP position as the holder to two key patents for the exclusive license of siRNA for therapeutic uses: MIT, the Whitehead Institute, the Max Planck Institute, and the University of Massachusetts Medical School co-own the first patent, while MIT and Max Planck exclusively own the second. Interestingly, the licensing situation on these patents has not been sorted out, according to technology transfer officers from MIT and U. Mass Medical School who attended the conference. Under US law, each of the patent holders can license its rights to the patent as it sees fit, and the U. Mass Medical School has not agreed to exclusively license the first patent (one of whose inventors is Tuschl), to Lnylam. Instead, this patent is available for non-exclusive licensing to all companies that apply, the technology transfer officer said. Some believe, however, the two patents "play together", said MIT technology transfer officer Irene Abrams, so this license to just one patent for therapeutic use of siRNA would in that view be useless. Additionally, under European law, all of the patent holders must agree on the licensing arrangements, and so far they have been unable to agree on an exclusive license in Europe.
Mahanthappa, however, took a diplomatic position on the issue. "Whether we do or don't own siRNAs for therapeutic use, the question is, how to turn [the technology] into a drug," he said.