Skip to Content

Prescription RNA

RNA-based drugs offer hope for treating cancer, AIDS, and other diseases.
December 1, 2002

It’s champagne for everybody. Phil Zamore pops the cork from a bottle of Montaudon, drenching the brand-new carpet. Everyone in his lab fills a glass to toast the boss. “Unexpected good news,” he explains. Zamore, a biochemist at the University of Massachusetts Medical School in Worcester, has just received a national award worth $1 million over five years. “The budget of the lab just tripled.”

Zamore is understandably giddy, and it’s not just about the money. Zamore’s field, RNA interference, or RNAi, is only a few years old, but it has taken the world of biology by storm. “RNAi is the most exciting insight in biology in the past decade or two,” says Nobel laureate Phillip Sharp, a biologist at MIT. And Zamore’s lab is one of a handful moving the field forward at a dizzying pace. “I think everybody who works in the field feels a bit breathless from the progress,” Zamore says.

The sense of excitement shared by Zamore, Sharp, and other researchers is well-founded. For decades, researchers thought RNA was merely DNA’s messenger, slavishly delivering DNA’s protein blueprints. But it now appears that tiny double strands of RNA, introduced into lab-grown cells or animals, can quickly and efficiently turn off any given gene.

The implications are breathtaking, because living organisms are largely defined by the exquisitely orchestrated switching on and off of genes. Biologists, until now, have only been able to mimic this switching process in a slow, ponderous, and indirect way. But the ease with which RNAi can turn off genes, researchers say, seems almost mystical. Laboratory techniques using RNAi are already biologists’ methods of choice for discovering the functions of particular genes. And it promises a new way to treat disease directly by shutting down key genes involved in various ailments. Already, at least eight companies-including one founded by Zamore, Sharp, and colleagues-are working on RNAi therapies for everything from viral diseases to cancer.

“The Holy Grail is to develop all this into drugs,” says Zamore. “To be able to give you a small interfering RNA that would shut off expression of your high-cholesterol gene. That would lower the level of hepatitis C infecting your liver. Or maybe, I think in perhaps the biggest pie-in-the-sky application, that would hone in on a gene specific to tumor cells and kill the tumor.”

How soon this might happen is anybody’s guess. RNA interference burst into the consciousness of the scientific world at the annual meeting of the RNA Society in Banff, Alberta, in May 2001. There, Sayda Elbashir, a postdoc in the lab of biochemist Thomas Tuschl at the Max Planck Institute for Biophysical Chemistry in Gttingen, Germany, stunned his listeners with the news that tiny double-stranded RNA fragments quickly, easily, and specifically turned off genes in human cells, a role researchers had never before seen RNA play.

“Most of the audience was just sitting there saying to themselves, Science has just changed,’” recalls University of Michigan biochemist David Engelke. “The only thing that prevented pandemonium was that we’d been promised this sort of thing before.” Skeptical, Engelke waited a few months. “Then these reports started to trickle in: Gee, this really works!’”

The RNA Surprise

Until recently, the rather unglamorous role biologists had attributed to RNA was that of a passive messenger, delivering genetic information from DNA to the protein-making machinery of the cell. In this process, the DNA code of a gene is transcribed into an RNA copy, which the cellular machinery translates into a protein. In RNA interference, short bits of RNA block the process by destroying the message en route. The double-stranded RNA fragments lead cutting enzymes to the RNA that carries the genetic message. The messenger RNA is then chopped up and marked for destruction: the gene’s message is effectively “silenced.”

Biologists have known for years that single-stranded RNA molecules designed to pair with a messenger RNA could shut down protein production, but this artificial process is unreliable even in the lab. Nature, though, does regulate genes using RNA, specifically double-stranded molecules.

The first hints of the phenomenon appeared back in 1990, but at the time, researchers didn’t connect what they had observed with RNA. That year, plant biologist Rich Jorgensen, then at DNA Plant Technology in Oakland, CA, was trying to make purple petunias a deeper shade of purple. He inserted a new, supercharged copy of the gene that controls production of purple pigment. To his surprise, he got white petunias. Jorgensen recognized the importance of this paradoxical effect, but he could not explain why adding more of a gene had turned that gene off.

The next clue came in 1995, when geneticists at Cornell University cloned a gene in the microscopic soil worm C. elegans. To verify their discovery, they used a standard lab method to turn the gene off: they added a single strand of RNA that matched the messenger RNA. This complementary strand bound to the messenger, stopping it from being translated into a protein. Unexpectedly, a noncomplementary single strand of RNA they were using as an experimental control and which should have done nothing, also shut down the gene.

In 1998 biochemist Andrew Fire, then at the Carnegie Institution of Washington, and geneticist Craig Mello, at the University of Massachusetts Medical School, solved the mystery. Injecting complementary single strands of RNA into worms, they got an astonishingly potent silencing effect when the two strands combined. After demonstrating that double-stranded RNA was the real silencing agent, Fire and Mello coined the term “RNA interference,” and a new field was born. In retrospect, Jorgensen’s supercharged purple genes yielded double-stranded RNAs that had the same effect on the native purple genes, essentially shutting them off.

The double-stranded RNA seemed to provide a more stable and reliable means for shutting off specific genes than did the single strands, and labs that were studying organisms including plants, worms, and flies eagerly adopted the new method. RNA interference didn’t work in mammals, though: the immune system destroys cells that contain double-stranded RNA to defend against RNA viruses like those that cause hepatitis A and C. Then came Tuschl and Elbashir’s revelation in Banff that very short RNA segments, which they dubbed “small interfering RNA,” did work in human cells. At that point, says Sharp, “the whole field took off.”

Silent Treatment

Now investigators are looking for ways to turn this powerful new role for RNA into corporate profits. Virtually all drug companies already use RNA interference as a tool for drug discovery. One of the most popular strategies for finding new drug targets involves knocking out-or disabling-genes one by one to see what happens. If, for example, a diseased animal can be cured by knocking out a particular gene, that gene’s protein could make a good drug target. Using small interfering RNAs, it turns out, can radically speed this process. Instead of spending months or years to engineer a knockout, researchers use the RNAs to specifically and rapidly shut off a gene. They can also observe whether turning off the protein-as a drug would-causes side effects. The process takes place “in a matter of days, instead of a year,” says Christophe Echeverri, CEO of Cenix BioScience, a biotech company in Dresden, Germany.

In the ultimate application, small interfering RNAs might themselves be drugs: rather than blocking a particular protein, as standard drugs do, RNAi would prevent the protein from ever being made. Last June, MIT’s Sharp showed that such RNAs, targeted to key viral and human genes, could stop HIV infection in cells grown in the lab. In one experiment, the researchers mixed HIV-infected cells with small interfering RNAs targeted to viral genes. The RNAs halted viral reproduction. Sharp’s group also mixed uninfected cells with small interfering RNAs targeted to CD4, a protein on the surface of cells through which HIV gains entry. The researchers showed that the RNAs did decrease production of CD4. Two and a half days later, they exposed RNA-treated cells and untreated cells to HIV. The virus infected four times as many untreated cells.

Despite the encouraging results, for the time being RNAi drugs are still in the dream stage, says Sharp. But Sharp considered the early promise tantalizing enough to cofound-with Zamore, Tuschl, and two other scientists-Alnylam Pharmaceuticals in Cambridge, MA, to develop such drugs. The company was barely off the ground when it secured $17 million in venture capital funding last July.

Making RNAi drugs, though, won’t be easy. For one thing, no one has found methods suitable for administering the RNAs to humans. “There’s a delivery problem. It’s as simple as that,” says Harvard University chemist Stuart Schreiber. “Getting nucleic acids to their target tissues [is] an unsolved problem in medicine.” RNAi therapy is essentially gene therapy, Schreiber says, and it will face the same problems-inefficiency, ineffectiveness, and immunological side effects-that have stalled that field since 1999, when Jesse Gelsinger died during a gene therapy trial at the University of Pennsylvania. Doctors there used modified viruses as delivery vehicles, or “vectors,” to shuttle DNA into the teenager’s cells. Gelsinger’s immune system responded massively-and fatally.

Sharp says the hope is that small interfering RNA might not need vectors to reach its target, thus avoiding most of the pitfalls associated with DNA-based gene therapy. But that scenario is far from certain. “Can you modify RNAs to make them more stable [and] to make them be taken up more efficiently by cells?” Sharp asks. “We don’t know.”

Making Sense

Recent biomedical history doesn’t help settle the uncertainty. In fact, this isn’t the first time scientists have tried to make a drug based on silencing RNA. Single-stranded “antisense” RNA or DNA can also shut down genes-and doesn’t need a vector. Inside the cell, an antisense molecule finds its complementary messenger RNA and, like two sides of a zipper, they bind tightly, preventing the messenger RNA from going through the protein-making machinery of the cell. The result, in theory at least, is gene shutdown.

Antisense, though, has so far been a disappointment as the basis for new drugs. After more than a decade of intense developmental work, only one antisense drug-Isis Pharmaceuticals’ Vitravene, for the treatment of certain rare eye infections in AIDS patients-has won Food and Drug Administration approval. The first generation of antisense drugs, which were tested in the early 1990s, rapidly degraded in the body, were hard to get into cells, often failed to find their target, and caused severe side effects. More stable antisense drugs are now being tested in humans.

Can RNAi do better than antisense? Not anytime soon, predicts Frank Bennett, vice president for antisense research at Isis. “If you compare RNAi to the current version of antisense, there really is no advantage,” he says. “[Small interfering] RNA technology is really in its infancy. It’s somewhat equivalent to where antisense was 10 years ago, when we were just beginning to do experiments in animals.”

But RNAi people see their technology as fundamentally different from antisense. “The big advantage here of RNAi over antisense is that, lo and behold, this actually really works,” says Cenix CEO Echeverri. RNAi, he says, is far more potent and reliable than antisense. “Antisense projects were typically seen as suicide projects,” he says. “You could spend a lot of time getting it to work, and it would never work. You’d be left with nothing to show.” RNAi’s greater potency, Echeverri believes, should yield better therapies. And because less drug will be needed to silence a gene, there should be fewer side effects.

People have been struggling with antisense, and here’s a technology that comes along that really works,” agrees Jon Wolff, chief scientific officer of Mirus, an RNA therapeutics company in Madison, WI. “Antisense is hard to reproduce, but RNAi is something that works right out of the barrel.”

But could RNAi be just another overhyped technology? “The proof is in the pudding,” says Echeverri. “Over the last two, three years, RNAi has just completely taken over. Everyone is turning to it; in every organism they’re trying it. And it wouldn’t be this popular if it weren’t successful.”

Silencing Doubts

No one has yet tried RNAi in humans, but one company is close: Ribopharma, a biotech startup in Kulmbach, Germany. More than a year before Tuschl’s group stunned the scientific community with its news, Ribopharma’s founders, former Bayreuth University lecturers Roland Kreutzer and Stefan Limmer, discovered that small RNAs worked in mammalian cells. Or so Kreutzer and Limmer claim. They have never published their data.

Kreutzer and Limmer reasoned that it was physically impossible for the very long RNAs, such as those used by Fire and Mello, to bind all at once to their target RNAs. Only short segments would stick. So they tried silencing mammalian genes using RNAs short enough to evade the fatal immune response. “It wasgambling,” says Limmer. “And it turned out that it really works.” The researchers filed a patent application, quit their teaching jobs, and in June 2000 founded Ribopharma.

Ribopharma’s principals are planning to begin human trials next year, probably starting with tests of small interfering RNAs in the treatment of malignant melanoma and pancreatic cancer. Kreutzer and Limmer say their RNA constructs are stable enough to work without vectors and can be injected directly into the site of a tumor or into the bloodstream. The company has raised more than $18 million. But because Ribopharma has yet to publish its results, it’s difficult to evaluate its claims, say other RNA researchers. “They’ve been doing some things,” says MIT’s Sharp, “quite nicely….[But] it’s a long road.”

How long? Attitudes range from Ribopharma’s sanguine assurances to strong pessimism. David Beach, president of RNAi startup Genetica in Cambridge, MA, points to antisense’s decade-plus odyssey. “I don’t want to sit and argue deploying RNAi in a therapeutic mode would be any simpler,” he says.

What is far clearer is that RNAi is forcing biologists to rethink RNA’s role. In the last few years, researchers have found hundreds of genes that code for small RNA molecules, dubbed “microRNAs,” in organisms ranging from plants and worms to humans. Like their small interfering RNA cousins, microRNAs appear to silence genes, but their role in biology is mostly unknown. “Many of them have been very highly conserved during the course of evolution; [so] they must be doing something important,” says MIT biologist David Bartel. Meanwhile, the realization that RNAi is a natural-and probably fundamental-process in plants and animals has helped make it one of the most exciting mysteries in today’s biology.

“Tiny RNA genes may be the biological equivalent of dark matter-all around us but almost escaping detection,” wrote Gary Ruvkun, a Harvard Medical School molecular biologist, in 2001 in the journal Science. What are these mysterious genes doing? “I suspect what we’re looking at is a very ancient method of controlling gene expression,” says Zamore.

If microRNAs are switches that decide whether stem cells become neurons or muscle, or whether cancer cells grow or die, then RNA interference is a lot more important than anyone imagined just a few years ago. “We simply stumbled upon a whole new branch of molecular biology that we didn’t know about before,” says Michigan’s Engelke.

To the optimists, these breakthroughs portend the quick development of effective drugs. And even biomedical researchers made cynical by extravagant claims for magical cures think RNAi just may be the real thing. Last year, when RNAi first worked in human cells, “everyone woke up and said, I wonder if this is the silver bullet?’” says Engelke. “And it might be. It might be.”

Companies Developing RNA Interference COMPANY PRIMARY RNAi FOCUS Alnylam Pharmaceuticals
(Cambridge, MA)
Therapeutics Benitec
(Brisbane, Australia)
Intellectual property for genomics and therapeutics Cenix BioScience
(Dresden, Germany)
Drug target identification and therapeutics for cancer Devgen
(Ghent, Belgium)
Drug target identification and therapeutics for diabetes, depression, and Parkinson’s disease Genetica
(Cambridge, MA)
Drug target identification for cancer Mirus
(Madison, WI)
Therapeutics, using long double-stranded RNA Ribopharma
(Kulmbach, Germany)
Therapeutics for cancer and hepatitis C

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.