Skip to main content
Premium Trial:

Request an Annual Quote

Pharma Runs Interference Part three in a multi-part GT series on RNA interference


By Adrienne J. Burke

When the methods of RNA interference in human cell cultures were first described in the scientific press in 2001, many pharmaceutical industry researchers who wanted to try the approach merely followed the publication like a recipe and were stunned at how easily results came. Now, research papers emerging from their laboratories indicate that, in just two years, RNAi has been widely adopted and universally accepted by the pharmaceutical industry as the most important new tool for drug target discovery and validation.

Garret Hampton, director of cancer biology at the Genomics Institute of the Novartis Research Foundation, sums up the RNAi buzz this way: “When you’ve had a favorite gene for many years and you’ve tried antisense and you’ve tried over-expressing the gene — you’ve gone through all that to functionally identify this gene that you’re emotionally attached to — you get really excited about [being able to] knock down every gene and ask what the cell does as a consequence.”

Mastery of siRNA techniques might someday make for a competitive advantage in the cutthroat world of drug discovery. But for now, the technology is still new and thrilling enough that many pharma scientists are not only willing but also eager to exchange information about their experiences using it. Conferences on the subject abound, and pioneering pharma researchers are advancing the field with publications and public lectures. For instance, Aventis sponsored an RNAi meeting in Maine earlier this year, a team of Merck researchers published their findings on off-target effects of siRNAs in May, and Stephen Fesik, divisional VP of cancer research at Abbott Laboratories, organized a Keystone meeting on RNAi in January this year. His team also published a paper on siRNA specificity in PNAS in May.

To be sure, a few pharmas remain behind the traditional shroud of secrecy. GlaxoSmithKline and Eli Lilly turned down requests for interviews for this article — a spokeswoman for the latter company explaining, “It’s too competitive an area for us to comment.”

But the more pervasive attitude seems to be that sharing knowledge will be good for all by advancing the field more quickly. One of Wyeth’s resident RNAi experts who is a proponent of that knowledge-sharing approach declined to speak only because his company’s public relations department denied him permission. But genomics experts at Abbott, Aventis, Bristol-Myers Squibb, Incyte, and Novartis were able to describe for GT the status and applications of their RNAi programs. What follows is a survey of the RNAi activities of those organizations.

Building Libraries

Like many of its competitors, Abbott’s global pharmaceutical R&D department adopted RNAi in 2001 after Tom Tuschl published his proof of the technique in mammalian cells. “We gathered together the cancer biologists and applied the tools to test the method and it worked extremely well in knocking down our genes of interest,” recalls VP Fesik. From that point, the group commissioned Dharmacon to build it a library of 507 siRNAs to target the kinase family of enzymes. Ever since, scientists in Abbott’s oncology and metabolic disease groups, as well as those on an immunosciences team, have been using the library to knock down genes.

The relatively small investment that RNAi research requires — no new analytical instruments or platforms — seems to pay off quickly. Says Fesik, “We have been able to identify new targets using this methodology that we wouldn’t have identified otherwise. Also, we are learning much more on a target-by-target basis by having this at our disposal.”

In their PNAS paper, Fesik and his colleagues recommend siRNA design methods, transfection conditions, and used microarray experiments for comparing gene expression signatures among siRNA-targeted genes. They conclude that their data “establish siRNA-mediated gene silencing as a reliable and valuable approach for large-scale screening of gene function and drug target identification and validation.”

Now Fesik is in negotiations to purchase a bigger library of siRNAs. Explaining his choice to outsource the chore instead of building such a collection in-house the way some pharmas are doing, Fesik says, “We investigated making vector-based siRNAs in-house, but decided to purchase a library. They’re available on the outside, it’s more cost effective for someone else to make them, and I’d rather put our people on running assays.”

Fesik sees a few areas where, he says, “We could extend the utility of siRNAs.” One way would be to use siRNAs in vivo, in transgenic animals. Indeed, some Abbott researchers are studying that approach now. Another improvement Fesik would like is a more stable siRNA — one that won’t break down as easily, he says. More stable siRNAs would have a longer half-life in animals and possibly penetrate certain difficult cell types.

Switched On to Switching Off

Mark Cockett, vice president of applied genomics for Bristol-Myers Squibb, says his group has been “very switched on to [the RNAi] approach” since 1998 when Andy Fire and Craig Mello first described interference in model organisms. In collaboration with Exelixis, BMS screened Drosophila and C. elegans with long, double-stranded RNA. As a result, Cockett says, the group now has whole-genome sets that it applies to key model organism assays.

Asked if the interference method has been perfected, Cockett says, “It’s fairly routine. There are improvements you’d like to make, but to do RNAi in C. elegans, you just soak the worms in long, double-stranded RNA and knock genes out and you can see phenotypes.” For Drosophila, BMS now has siRNAs for most of the 13,000 genes in 96-well plates.

Cockett categorizes siRNA as a “new capability area” — one of two major technologies in which he is investing budget dollars. (The other is antibodies and peptides.) The goal, he says, is to create assays for human cells, similar to those the group owns for Drosophila, in order to have a mammalian surveillance system. To start, he’ll focus on capturing the main druggable target classes. “We’re taking a targeted approach, but the definition of what is druggable is going to change. We want to ultimately have the complete set, but that’s more of an aspiration right now,” he says.

Cockett acknowledges that RNA interference is in its early days. “It’s still unclear how it’s really going to be broadly applied. But it opens the way forward for genome-wide datasets on function. We’re driving it toward where we hope we’ll get functional information on key areas en masse such that we can build mass functional databases and lead to a richer depth of information.”

As for BMS’s chosen method of synthesizing siRNAs, Cockett says, “We haven’t locked ourselves in.” The company has an alliance with Sequitur to create siRNA libraries, but also purchases oligos from Ambion and Dharmacon. BMS is also funding research in the area at the Whitehead Institute, and the BMS bioinformatics group recently released an open-source siRNA design algorithm.

Parallel Approaches

At the Genomics Institute of Novartis, RNAi has been in use for target validation on novel genes for nearly a year. “Like everybody else, as soon as it became apparent what the potential might be, we jumped on board,” says Garret Hampton, director of cancer biology. Now at least two research papers by members of his lab are awaiting publication.

While typical pharmaceutical company labs need to be more pragmatic and think about the end use, Hampton says his team, which is more like a halfway house between pharma research and basic biomolecular research, has more freedom to experiment. The institute has developed two parallel RNAi approaches, Hampton says.

One team plans to develop a complete set of either siRNAs or short hairpin DNAs — the vector-cloned equivalent of siRNAs — for the entire human genome. Novartis researchers have been using a pilot collection of about 700 siRNAs in screening experiments, and now the institute is in the process of “identifying the best partner to move forward with to build a genome-wide resource” that would serve the entire Novartis organization, Hampton says.

“We want to have a plasmid-based collection of shDNAs, because that’s a renewable resource,” he says. That’s the same as a cDNA collection, where you can always make more of it — we want a permanent resource.”

The other RNAi approach is comprised of individual lab usage. “We have a certain number of genes we think are potentially interesting targets. We’ve developed siRNAs against those targets — that’s basic functional genomics screening … more on an individual basis like you would find in an academic lab,” Hampton says.

As for the investment of his scientists’ time in RNA interference studies, Hampton says, “We do spend a lot of time at it, because in many ways, this is the holy grail of functional genomics. We’ve never had a tool that doesn’t require specialized equipment; we’ve never been able to knock down functionality. This couldn’t have come at a better time.”

And Hampton agrees with others’ assessments of the value of this new tool saying that it’s enabling his group to identify new genes involved in disease processes. “The sense of discovery is thrilling. You have something in your hands that you never had before,” he says.

Adopted Over Antisense

Katherine Call, the global head for genomics technology transfer at Aventis, says RNAi was adopted rapidly there, and has now been deployed at the company’s laboratories worldwide. “We were doing antisense, and we had some success, but it took a lot of optimization and did not have a high enough success rate,” Call says. RNA interference had neither problem and achieved longer-term expression knockdown than antisense.

Call says she alerted Aventis’ model organism researchers in 1999 to RNA interference studies in the scientific literature. Company scientists in Frankfurt started working on it and were able to transfect double-stranded RNAs in worms by simply feeding E. coli vectors to them. “It was fantastic,” Call says, adding, “I didn’t realize that a breakthrough for mammalian would come so quickly.”

Now, Call mentors monthly meetings of a company-wide RNAi network of about 15 people at labs in Cambridge, Mass., Frankfurt, and Paris. The group usually gathers by teleconference or WebEx to discuss issues such as transfection methods, vector and viral-based approaches, working with hairpins, and “who’s got what working where and how is the effect,” Call says. “We’re trying to do as high-throughput as possible, so we’re thinking along those lines.”

While the company isn’t doing any whole-genome RNAi analyses, it does have some gene-family screening projects, and is putting in place a database for capturing and sharing RNAi records. And Call says that although it has in-house oligo-making capabilities, Aventis chooses to outsource siRNA synthesis. “A lot of companies are marketing good products and have spent a lot of time and effort [perfecting oligo design]. We’re not here for technology development.”

Ubiquitous as PCR

Most of Incyte’s drug discovery scientists who are transplants from DuPont Pharmaceuticals began using RNAi techniques soon after the breakthrough in mammalian RNA interference was published, says Greg Hollis, VP of the applied technology group. “The technique worked well enough that the first experiment we did based on the publication worked in our hands.”

And now that those scientists are at Incyte, they are able to use the company’s proprietary transcriptome database to design siRNAs. “It started out as a tool in one or two people’s hands, but like many other technology tools, it has been disseminated across [the company],” Hollis says, not just as a tool for validating targets, but also for invalidating a small number of targets found in the public domain.

“I don’t think it would be fair to say we’ve perfected it, but it’s a useful and powerful tool and we’re using it in a focused way,” Hollis says. “Like in the very early days PCR was only used by a few, and now it’s ubiquitous.”


The Scan

Suicidal Ideation-Linked Loci Identified Using Million Veteran Program Data

Researchers in PLOS Genetics identify risk variants within and across ancestry groups with a genome-wide association study involving veterans with or without a history of suicidal ideation.

Algorithm Teases Out Genetic Ancestry in Individuals at Biobank Scale

Researchers develop an algorithm known as Rye to tease apart ancestry fractions in admixed individuals at a biobank-scale, applying it to 488,221 UK Biobank participants in Nucleic Acids Research.

Multi-Ancestry Analysis Highlights Comparable Common Variants at Complex Trait-Linked Loci

Researchers in Nature Genetics examine common variants implicated in more than three dozen conditions, estimating genetic effect similarities across ancestry tracts in admixed individuals.

Sick Newborns Selected for WGS With Automated Pipeline

Researchers successfully prioritized infants with potential Mendelian conditions for whole-genome sequencing or rapid whole-genome sequencing, as they report in Genome Medicine.