The creation of new drugs is a vital part of the health-care system. Researchers in academia and in industry are always searching for new ways to combat disease — more efficiently, with fewer toxicities, and less chance of rejection. Until recently, most medicines have been limited to the classic formulation of a small molecule targeting a protein to disrupt its function. But there are many targets and formulations that have yet to be fully exploited, particularly those involving RNAs.
Small interfering RNAs and microRNAs can be used both as targets for drugs and as compounds in drug formulations to disrupt the function of certain genes. By taking advantage of RNA interference, miRNAs and siRNAs can bind to specific messenger RNAs and either increase or decrease their expression to affect how much or how little a given target gene functions. "There are really quite a lot of different methods and cellular pathways that are exploited. In some of them the field is quite old, so you might consider antisense technology a form of RNA therapeutics," says the University of Massachusetts Medical School's Phillip Zamore, who co-directs the school's RNA Therapeutics Institute. "There are people who are engineering different kinds of cellular RNAs to alter splicing or to degrade messages. Most of those involve re-engineering longer RNA, and probably can be delivered as drugs."
But the real promise shown by RNA in the therapeutics field comes from small RNAs, he adds. "The new small RNA therapeutics are really the first time that RNA has shown some promise as a drug," Zamore says. "The secret is that they're small, they're generally double-stranded, and they need relatively little — although they need some — chemical modification to make them stable. The real advance has been the discovery of chemical formulations that allow them to be retained in the body instead of filtered out by the kidneys and delivered to cells. And those drugs, which generally take the form of siRNA, are now being tested in early-stage clinical trials."
Most miRNA-based drugs use RNAi pathways to bind argonaute proteins — the protein complexes responsible for RNAi — and, together, the combination of siRNA or miRNA and argonaute protein bind the target complementary mRNA and destroy it. "My lab has for the last 12 years studied the basic mechanisms of siRNAs and microRNAs," Zamore says. "We work in a variety of organisms including flies and mice, and out of our basic research efforts, we've been able to understand new ways in which one can use siRNAs to target genes that differ by as little as a single nucleotide. And in that case it would be a question of targeting a mutant versus a normal gene."
Zamore's primary focus is Huntington's disease, which has been shown to be caused by a mutated gene with an extended CAG repeat. "There has not been much success by any lab targeting the extended repeat," Zamore says. "But as we showed a couple of years ago, there's a polymorphism — a neutral base change — that is commonly associated with the disease gene. And because it creates a single nucleotide difference between the disease gene and the wild-type gene, one can target that and not the normal gene." Zamore's work on Huntington's disease is still in the preclinical stages, but he's hopeful that it will eventually make it to clinical trials.
The list of diseases that are potentially treatable with RNA drugs is long. "There are certainly clinical trials using siRNAs for cancer, and the kinds of targets that people are interested in are molecules, proteins, that normally would be considered non-druggable," Zamore says. "The pharmaceutical industry has a very short list of the types of proteins that they suspect will be amenable to inhibition by classical small-molecule drugs. Any gene that doesn't fall on that list — whose expression or over-expression contributes to a disease — would be a good candidate for RNA interference. So basically, if you can reduce the expression of the protein and do some good, it's a good target for RNAi."
Merck is one of the companies moving into RNA therapeutics. The company acquired San Francisco-based biotechnology company Sirna Therapeutics — which specialized in the development of siRNA drugs — in a 2006 deal worth $1.1 billion, because it believed RNA could "open up a whole new class of medications to treat patients with unmet medical needs," says Jeremy Caldwell, head of Merck's RNA therapeutics division. Caldwell says Merck is working to apply RNA drugs to the treatment of cancer as well as cardiovascular and respiratory diseases. He adds that he's "cautiously optimistic" that the company will be starting clinical trials on some therapeutics in the near future. "RNA is really going to be the modality that takes advantage of all the high-throughput sequencing and genomic information that identify potential drug targets, many of which are not druggable with classic approaches like small molecules," Caldwell says. "Classic small molecule targets are receptors and enzymes because they have a hydrophobic pocket that the small molecule can insert itself into and block the activity of the target. Biologics are similar, but only work on cell surface. What siRNA will be able to do is address both small molecule targets and biologic targets, but also targets such as adaptor proteins that regulate a catalytic intracellular event, for example." He says there are practically no diseases that cannot be targeted by RNA therapeutics.
There are some problems that must be resolved, however. For one thing, most RNA therapeutics are currently limited by their routes of administration — such as intravenous or subcutaneous approaches, Caldwell says. And since there are already plenty of drugs that can be administered orally, it's unlikely they would be replaced by comparable RNA therapeutics unless those, too, could be administered by mouth or were a significant improvement upon the standard treatments.
In addition, researchers are struggling with how to deliver these drugs to their intended targets once they're administered. Many companies developing RNA drugs, including Merck, are taking a route through the liver, which is generally quite efficient, but limits the number of diseases that can be treated with these drugs.
Silence Therapeutics, which specializes in the delivery of targeted RNAi therapeutics, aims to solve this problem with its new delivery system called AtuPlex — a lipid delivery technology that targets the vascular endothelium of different organs. "For the whole field, the biggest hurdle is the delivery. There are different ways to use RNAis or even antisense molecules, but I think in the last three or four years, most companies shifted to formulations — they realized you need delivery technologies for the nucleic acids," says Jörg Kaufmann, Silence's vice president of research. "Our delivery technology, AtuPlex, is a liposomal formulation complexed with nucleic formulations, and the difference is that our formulation targets the vascular endothelium of different organs, including the tumor vasculature." While some companies target the tumor cells themselves, Silence's method allows the therapeutic to directly enter the tumor, and to alter the vasculature of target organs in order to prevent secondary metastasis, Kaufmann adds.
The company's most advanced RNAi therapeutic, a compound called Atu027, has been shown to prevent metastasis to the lungs by modulating the organs' blood vessels, Kaufmann says. "Basically, if the cancer cells are in a solid tumor, at one point they will go into the bloodstream to start growing in different organs. Our system delivers the nucleic acids to the endothelium or the vasculature," he says. Then, the system works to change it enough to prevent cancer cells from taking hold and metastasizing. This approach differs from that of Silence's competitors — companies like Alnylam, which UMass's Zamore co-founded, or Tekmira — which generally target the liver, liver metastases or cancer in the liver, whereas Silence is attempting to target metastases throughout the body, Kaufmann says.
Currently, the company is at the end of Phase I testing on Atu027 and will start Phase II trials by the end of this year or in early 2012. Phase III trials are likely six to 10 years away, Kaufmann says, but he is optimistic about the drug's chances of making it to market. So far, the company is still escalating the dose, and is aiming to show Atu027 can be used as a monotherapy in Phase II testing. "After that, we will combine it, and I personally envision it that it will be used in combination with chemotherapy or maybe as an adjuvant therapy," Kaufmann says. "After the chemotherapy has taken care of the primary tumor, this would prevent metastasis."
RNA in the crosshairs
At the University of Michigan, chemistry and biophysics professor Hashim Al-Hashimi is taking a different approach: instead of using RNAi to create drugs, he's creating ways to target the RNAs themselves. "Antibiotics that bind RNA in the ribosome are really the only example of a bona fide drug that we have on the market that we know functions by binding RNA," says Al-Hashimi, who co-founded a company that specializes in RNA targeting, called Nymirum. "There may be more drugs out there that function by binding RNA that we don't know about, but the drugs that are currently known to bind RNA and have an effect are very few, and the ones that are known are the antibiotics, and those compounds tend to be positively charged. That presents a problem in general, for various reasons — they can be toxic, they can be difficult to take up by cells — but certainly these are compounds that demonstrate the proof of principle that one should be able to target RNA."
However, part of the challenge in finding compounds that bind RNAs is developing assays or technologies that will allow researchers to measure the exact effect of a compound on its target. "Most small molecule drugs target proteins and take advantage of the fact that many proteins are enzymatic," Al-Hashimi says. "When a molecule is enzymatic, you can have an enzymatic assay to read the effect of a drug, so you can screen to assess the inhibitory activity of small molecules by simply asking, 'How well does this enzyme do what it's supposed to do in the absence or presence of a small molecule?'" The challenge with RNA, however, is that the majority of RNA targets are not enzymatic, so there isn't an easy way to create a high-throughput assay to measure a compound's effect on the target RNA.
There are methods that involve tagging RNA with modified compounds like fluorescent tags, Al-Hashimi says. "But because RNA is very fickle and flexible — a very delicate structure — having these large tags attached to it can cause problems in terms of perturbing the RNA," he adds. "Also, with these techniques that rely on tagging the RNA, often what happens is that the molecules that you would like to test have features or optical properties that make them ill-suited to these types of experiments because they happen to absorb light at the same wavelengths as the tag does. So there's quite of a bit of limitation as to the types of molecules you can test with these types of approaches."
RNA, camera, action
To address this challenge, Al-Hashimi and his group have developed a new technology to test molecules and see if they will bind RNA. In a perfect world, this would be done with a computer program, Al-Hashimi says, but he notes that most computer programs assume that RNA is a rigid molecule when it is a mobile, flexible structure that's "wriggling and dancing around, and assumes many different shapes." Instead of taking static images of RNA and then asking a computer program to predict which compounds will bind to it, Al-Hashimi records videos of RNA to capture the structure's fluctuations.
"What we do is we take different frames from our movie, highlighting the lock in different shapes, and then ... we test keys," Al-Hashimi says. "We test them not just against one frame, but against all of the different frames we have, and that gives us more shots on target. So we will not, for example, miss the key that can specifically bind to an unusual shape of the lock. With this new technology we can find these keys, and screen them more effectively."
Using NMR spectroscopy coupled with computational techniques that can predict which agents will bind RNA, Al-Hashimi can visualize RNA in motion and screen for existing compounds that could target the RNA to treat disease. He's already successfully identified one compound — netilmicin — which inhibits HIV replication, and continues to screen existing compound libraries to see whether any of the molecules there could be used to target RNAs. "We know a lot about proteins — we know a lot about what molecules they like to bind — so we have a history that we've accumulated over many, many years," Al-Hashimi says. "With RNA, it's just an open field — we really don't know the kind of keys that RNA is going to like, and it might be keys that we have never synthesized. ... The advantage of this computational approach is that we can test molecules that don't even exist, and see if there's a class of molecules that we ought to spend some effort making, because they could be the next generation of small molecules targeting RNA."
Too soon to tell?
Of course, as with the development of any drug, there are questions as to how the body will react, whether there will be toxic side effects, and whether there is potential for the disease to become resistant to the treatment. Al-Hashimi says that it's still too early to tell whether drugs targeting RNA will create adverse reactions or treatment resistance, but says there is evidence indicating that RNA-targeting compounds may escape resistance more effectively than traditional drugs. "Because of the sheer amount of RNA that one has, there are more goals to shoot at," he says. "So the chance that you have one RNA that has more favorable properties might be quite large, simply because of the potential different ways you could attack a disease through targeting RNA."
And, he adds, the potential for beating resistance with combination therapies is high with an RNA targeting approach. "I think the sheer number of targets that are out there and the different strategies one can go about inhibiting a given disease, you can imagine cocktail strategies where you have drugs that bind not one but multiple elements and that could probably really help with the resistance issue, because you're hitting the disease from so many different ends. It's hard for a mutant to occur that can defeat all of them simultaneously," Al-Hashimi says.
Suppressing toxicities would be a matter of making sure the compound has "exquisite selectivity" for its target, so that it affects one specific target RNA and not similar RNAs as well, he adds.
When it comes to RNAi, Merck's Caldwell says that, although it's likely some diseases will evolve resistance to certain RNA therapeutics, the advantage of RNAi drugs is that it's easier to identify potential resistance mutations in the pre-clinical testing stages and be ready with a backup against the mutated version of the disease.
Overall, researchers say that there is a great deal of potential in RNA therapeutics. "The connection of RNA to diseases is literally unfolding as we speak," Al-Hashimi says. "We really have many decades to go to figure out RNAs and develop the technology needed, but now that the interest is there, we'll definitely be able to learn more and figure out how things work."