This article is reproduced with the permission of New Scientist for exclusive use by Nova users.

MicroRNAs: The cell's little emperors
25 June 2008
From New Scientist Print Edition.
Henry Nicholls

Enlarge Taking control

Last month, eight healthy volunteers in a special unit in Hvidovre University Hospital in Copenhagen, Denmark, made medical history when they received, via a drip, an experimental new drug meant to stop hepatitis C virus replicating in liver cells. There was no sign of any adverse reaction, so a slightly higher dose will soon be given to another batch of volunteers.

It will be months yet before we know whether the drug is safe, and longer still before we know whether it can help cure hepatitis C. But whatever happens, it will still be remembered as the first ever trial of an entirely new kind of drug. These drugs target minuscule snippets of RNA in our cells, snippets so small they went undetected for decades but which have turned out to have a huge influence on just about every process in our bodies. Blocking these tiny RNAs, or mimicking their effects, could lead to new treatments for everything from infections and cancer to ways to help repair damaged organs. What's more, such drugs will be far simpler to design, test and manufacture than conventional ones.

Just a couple of decades ago, no one suspected that RNA could wield such power. Back then, life seemed simple: genes code for proteins. When a protein is needed, an RNA copy of the DNA blueprint is made and sent to the protein-making factories. Since the RNA copies of genes - known as messenger RNAs - break down after a few hours, the level of protein production depends on the level of messenger RNA production.

According to this picture, the key players in a cell, beside genes, are proteins known as transcription factors. These proteins bind to the DNA next to a gene and can reduce or increase the production - transcription - of messenger RNAs. It seemed for a time that cells had an elegant and simple system for controlling gene activity, but biologists had reckoned without the messiness and madness of evolution.

One of the first hints that the control of gene expression is far more complex came early in the 1990s. Victor Ambros, now at the University of Massachusetts Medical School in Worcester, and colleagues had found that mutations in part of one chromosome caused developmental problems in nematode worms. They were trying to track down what they thought was a normal protein-coding gene located in that area, already named lin-4.

The lin-4 "gene", however, turned out to be tiny. What's more, it clearly did not code for a protein. A clue to its function, however, came from the fact that the RNA transcript of this sequence is complementary to part of the messenger RNA of another gene. In other words, the lin-4 RNA is an "antisense" sequence. This tiny bit of RNA - just 21 base pairs long - appeared to bind to a specific messenger RNA and jam up the protein-making factories. The team had stumbled upon a whole new mechanism for controlling gene activity (Cell, vol 75, p 843).

"It was really interesting and intriguing," Ambros says. However, the group put it down to a peculiarity of the nematode worm they were working on. They did look for similar RNAs but at the time the necessary technology to find them didn't exist.

In the late 1990s, the idea that some RNAs play an active role in controlling gene activity got a huge boost with the discovery of another, quite separate mechanism, known as RNA interference. It turns out that some cells produce short pieces of double-stranded RNA, known as siRNAs, which also target specific messenger RNAs containing antisense sequences. Unlike Ambros's tiny RNA snippet, though, siRNAs trigger a process that leads to the destruction of messenger RNAs before they can be used for making proteins. This system is thought to have evolved as a defence against viruses.

A weird worm thing

For the next few years RNA interference hogged the limelight. But in February 2000, a second gene for a tiny RNA similar to that found by Ambros was discovered in the nematode worm (Nature, vol 403, p 901). This tiny RNA, called let-7, turned out to block the production of not one but at least five proteins. "The reflex reaction everyone had was that this was a weird, worm-specific thing," recalls Gary Ruvkun of Massachusetts General Hospital in Boston, who led the team.

With the genomes of various species beginning to become available at the time, however, Ruvkun could search a few databases for the let-7 sequence on the off chance that it wasn't just a worm thing. "Bingo, up came the fly with a perfect match and the human," he says. "Within minutes I knew exactly what we were going to be doing for the next year." Soon Ruvkun showed that the let-7 gene could be found right across the animal kingdom (Nature, vol 408, p 86).

What this meant was that these these tiny RNAs could no longer be dismissed as worm-specific oddities. Since evolution had conserved them in so many different species, they must be doing something pretty important. And if there were two RNAs of this kind, there were likely to be many more, says Ruvkun. Even so, later discoveries completely surpassed his expectations.

Within a year other groups had discovered many other genes that coded for similar RNAs. The term microRNA, often shortened to miRNA, was coined in 2001 to describe these tiny regulatory RNAs, and soon dozens of teams - and companies - worldwide were studying them. MicroRNAs have turned out to be the biological equivalent of dark matter, says Ruvkun. "They were all around us but almost impossible to detect."

They might have long eluded us, but within cells they are bigwigs. Unlike siRNAs, microRNAs do not bind to just one or two messenger RNAs but many. In fact, some researchers think each microRNA can bind to hundreds of different messenger RNAs.

So far 678 microRNA genes have been identified in humans and some think there could be as many as 1000. If this is borne out, microRNAs would affect the expression of around a third of the human genome. The implications are truly enormous. "The miRNAs collectively have much of the genome under their influence," says Ambros. "They can affect almost any aspect of the biology of a cell or organism."

Bewilderingly complex

The world of microRNAs is proving to be bewilderingly complex, a real headache for those struggling to understand how cells work. However, it also brings enormous opportunities. There was already tremendous excitement about the potential of using artificial siRNAs to turn off individual genes, with several trials under way. Now, by inserting artificial versions of microRNAs into cells, researchers have shown that it is also possible to turn off whole sets of genes. What's more, by adding "anti-microRNA" molecules that bind to and block specific microRNAs, it is also possible to turn on whole sets of genes.

What makes such therapies revolutionary is that to design an artificial microRNA or anti-microRNA, you need only know the sequence of the microRNA whose action you want to block or mimic. There are still huge challenges associated with getting artificial microRNAs to specific organs or tissues in the body, but once researchers work out how to deliver one microRNA or siRNA, they will be able to deliver any small RNA in exactly the same way.

The leading company in the field is Santaris Pharma of Hørsholm in Denmark, which is behind the Copenhagen trial. The "drug" being tested is an anti-microRNA, or antagomir, designed to inhibit the action of miR-122, a microRNA found almost exclusively in the liver. Besides being exploited by the hepatitis C virus, it is also involved in the regulation of cholesterol, fatty acid and lipid metabolism.

Experiments in mice reveal that blocking miR-122 increases the expression of several hundred genes in the liver. Besides slowing replication of the hepatitis C virus, this can cause levels of cholesterol in the bloodstream to fall by up to 40 per cent. Importantly, says Keith McCullagh, the head of Santaris, his team has not seen any adverse side effects in mice or in monkeys, even at quite high doses of the inhibitor. It was this last finding that paved the way for the clinical trial.

MicroRNAs could help improve diagnosis as well as treatment. "All cancers are associated with their own particular pattern of altered miRNA expression," says Carl Novina of the Dana-Farber Cancer Institute and Harvard Medical School in Boston. "These patterns should help us to diagnose cancers more sensitively and work out the most efficient form of therapy."

There's also growing evidence that microRNAs play a direct role in cancer. Some white blood cell cancers, for instance, are associated with abnormally high expression of particular microRNAs. If these microRNAs help the cancer cells survive, anti-microRNAs might prove an effective treatment.

Conversely, several other cancers are associated with a lack of specific microRNAs. For example, chronic lymphocytic leukaemia cells are often characterised by the deletion of a cluster of microRNA genes that appear to suppress tumours, while levels of the let-7 microRNA discovered by Ruvkun are sometimes greatly reduced in lung cancers.

Frank Slack, a member of the team that discovered let-7, has recently shown that using artificial let-7 to boost levels of this microRNA in mice inhibits the growth of several kinds of lung cancer (Cell Cycle, vol 7, p 759). "We believe this is the first report of a miRNA being used to a beneficial effect on any cancer," says Slack, now at Yale University. "Let-7 may benefit a broad group of lung cancer patients."

It remains to be seen whether microRNA-based treatments will be effective on their own, but combining them with conventional therapies certainly looks promising. Work by Slack's team suggests that let-7 makes tumour cells more sensitive to radiation, while other teams have shown various microRNAs can make cancer cells far more sensitive to chemotherapies.

Besides having a role in cancer, microRNAs appear to be a key and previously unappreciated player in the immune system, says David Baltimore of the California Institute of Technology in Pasadena. "The more research the community does, the more important it sees microRNAs as being."

MicroRNAs play an important role in triggering an appropriate immune response. In mice, for example, miR-223 seems to control production and activation of the white blood cells that form part of the first line of defence against invading pathogens (Nature, vol 451, p 1125).

Some microRNAs directly target the genes of invading pathogens such as the flu virus, slowing their replication and helping reduce the damage. Others, however, are exploited by viruses, like the MiR-122 being targeted by Santaris. There are also plenty of viruses that produce microRNAs to thwart their host cell's defences, so anti-microRNAs should help to treat viral infections of this kind.

These master regulators of cellular activity have also been spotted at work in the very earliest stages of development, helping to determine which specific cell types embryonic stem cells will differentiate into. "It is likely that most lineage decisions will be influenced by at least one or more miRNAs," predicts Deepak Srivastava of the University of California at San Francisco.

Earlier this year, he and his colleagues showed that adding specific microRNAs to both mouse and human stem cells helped turn them into heart progenitor cells - the first clear evidence that specific microRNAs can direct the fate of embryonic stem cells. (Cell Stem Cell, vol 2, p 219). Although there are other ways of guiding stem cells down a particular pathway, the microRNA approach could be of immense value, Srivastava says.

Such findings raise the intriguing possibility that it might be possible to do the opposite: make specialised cells revert to a stem-cell-like state by adding microRNAs or anti-microRNAs. "Rewinding" cells was recently shown to be be possible by Shinya Yamanaka of Kyoto University in Japan, but his method involves genetically modifying cells. Doing it with microRNAs instead would be a huge step forward. "We are working on that right now with Dr Yamanaka," says Srivastava.

In spite of the explosion of interest in microRNAs, rushing towards miRNA-based therapies could be unwise, says Joan Steitz of Yale University School of Medicine. Just months ago, she and her colleagues reported that several human microRNAs assumed to block the production of specific proteins can instead boost protein production when conditions in the cell are different (Science, vol 318, p 1931). "This versatility completely changes the way we need to think about miRNA function," says Steitz. "It's early days, but it could turn out that all miRNAs have this kind of complex role. This is a warning flag for those designing therapeutics."

Ruvkun too is under no illusions. "In the long term I think it's a very good bet that there are going to be small RNA drugs, but is it going to come to fruition in the five-year horizon? I don't think so," he says. "I surely wouldn't invest any of my retirement funds in any of those companies."

And as pleased as Ambros is to have his name alongside the discovery of the first microRNA, he is not resting on his laurels. Instead, he is already hunting for the next undiscovered class of regulatory molecules. "miRNAs aren't going to be the final word," he predicts. "Biological systems are more complex than we can possibly imagine. It could be there are many modes of regulation yet to be discovered." Even so, you can expect to hear more - much more - about this small but mighty class of RNA.

Henry Nicholls is a science writer based in London and author of the book Lonesome George

From issue 2662 of New Scientist magazine, 25 June 2008, page 44-47

The secret of vertebrates' success What is it that makes vertebrates so special? Their backbone, of course, but was there anything about the genetic make-up of the very first vertebrates that set them apart from their spineless ancestors? A myriad of microRNAs is the surprising answer that Kevin Petersen of Dartmouth College in New Hampshire has come up with. As more and more microRNAs were discovered, Petersen noticed that mammals seemed to have far more of them than invertebrates. But were these microRNAs unique to mammals, he wondered? In fact, as his team reported earlier this year, it turns out that virtually all the microRNAs found in mammals are characteristic of vertebrates as a whole. This means that they appeared at the same time that complex vertebrate features arose, which was probably somewhere between 520 and 505 million years ago. Clearly, something spectacular happened in the genome of ancestral vertebrates, says Petersen. "It was an impressive rate of miRNA acquisition, almost an order of magnitude higher than most other places we've looked at in the animal evolutionary tree." He is prepared to bet that these miniature molecules are behind the appearance of ever-more-varied cell types characteristic of more complex organisms. Put simply, microRNAs made this explosion of complexity possible.

For the latest from New Scientiist visit www.newscientist.com