By Jennifer Crebs
It may not be televised, but the microRNA revolution is surely here. Over the course of the last 12 months, major findings have emerged that deepen the current understanding of how and why these small RNA molecules have developed their post-transcriptional silencing machinery.
MicroRNAs are non-coding RNAs, typically 21 to 24 nucleotides long, which technically made their debut in 1993, when Victor Ambros of Dartmouth University characterized lin-4 — a short, non-protein-coding strand of RNA essential for developmental timing in C. elegans. It took several years before homologs of lin-4 were found in higher organisms, including humans. Now, thanks to great strides made in the larger research field of RNA interference, the actual function of miRNAs is coming into focus.
In 2006 alone, the National Library of Medicine has indexed more than 110 papers sporting “microRNA” as a keyword. As of February, the Sanger Institute’s miRNA database, miRBase, contained more than 3,500 published entries, 326 of which were specific to humans.
The rush to stake claims in the field is marked with a fervor usually reserved for the discovery of ore or land. It’s no wonder: microRNAs have been implicated in everything from plant development to cancer, and the discoveries continue to pour into journals of every tier.
We touched base with several researchers whose miRNA investigations broke new ground over the past 12 months. They shared the stories behind their most notable findings, and also provided insight on what may be in store for the next phase of discovery.
REPRESSION AND EVOLUTION
Dave Bartel’s lab has been fanning the flames of microRNA research for years. Since 2001, the Whitehead Institute member has been studying RNA catalysts as well as RNAs thought to govern the state of gene expression. But it wasn’t until a little over a year ago that Bartel’s group, in collaboration with Christopher Burge’s team at MIT, reported the startling finding that more than a third of human genes are likely regulated by miRNAs. As they explained in Cell, the teams used a computational tool called TargetScan to predict conserved sequences across several species’ genomes, including mouse, chicken, and human.
“In the Cell paper, we were looking at different ways of finding miRNA target sites with greater sensitivity,” he says, adding that the results of this work raised a host of further questions.
Having established that some miRNA sites are conserved across genomes, Bartel wanted to know whether nonconserved sites are also able to perturb gene expression. To this end, the researchers experimentally exposed messenger RNAs featuring nonconserved target sequences to miRNAs, which showed no prejudice against binding to the 7-nucleotide-long matches.
When it comes to actual cell environments, as Bartel’s group reported in Science in late November, mRNAs with nonconserved sites are typically absent in mouse cells expressing matching miRNAs. “We found that conserved targets can be highly expressed at developmental stages before miRNAs are expressed,” explains Bartel, “and also that target expression levels fall as that miRNA begins to accumulate.” The results indicate that miRNAs may strongly influence the expression of mammalian mRNAs, and by extension, the very evolution of nonconserved sequences.
These days, Bartel’s group is “having a lot of fun doing high-throughput sequencing of small RNAs,” he says. With 454’s sequencing service, his team has already achieved coverage of 350,000 reads. By combining computational techniques with genomics technology, Bartel sees the possibility in finding miRNAs with lower expression levels, in addition to other classes of RNAs.
At the Institute of Plant Molecular Biology in France (IBMP Strasbourg), Olivier Voinnet directs the research of a lab dedicated to investigating RNA silencing in animal and plant systems. From yeast to Arabidopsis to cultured human cells, Voinnet’s lab studies the role of miRNA-directed regulation in RNA silencing. By spanning the gamut of organisms in which miRNA is found, Voinnet’s team has made substantial headway in clarifying miRNA’s role with a phenomenon common to nearly all life forms: viruses.
Last year, Voinnet published a paper in Science that effectively extends the role of miRNAs from regulatory to that of an antiviral defense mechanism. His team showed that a human miRNA, miR-32, targets the mRNAs of a retrovirus, the primate foamy virus type 1 (PFV-1), for translational silencing. Not to be outdone, the virus, it turns out, has developed a mechanism to reverse the silencing effect by way of a viral protein capable of suppressing miRNA function.
Previous studies showed that other types of virus also encode RNA silencing suppressors, Voinnet explains, and taken together with last year’s study, indicate that RNA silencing is a highly conserved antiviral defense mechanism. More broadly, the results also suggest that all miRNAs may have antiviral potential, independent of cellular function. “We are inclined to think that, at some stage, the miRNA would be beneficial to the virus in both humans and in plants,” he says.
Voinnet cites several research topics as important to the future of miRNA research. The composition of RISC, as well as any changes it goes through in vivo, is a valuable line of inquiry “especially as some of the proteins that form the RISC are very important to human disease, such as Fragile X syndrome,” he says. Another important challenge, he says, is deciphering the silencing mechanisms involved in miRNA function. “The ultimate aim is to get a grip on the exact amount of small RNAs in each cell,” he says, which could lead to diagnostic tools for a number of human diseases.
Because the pace of discovery in the miRNA field is fast and furious, labs everywhere run the risk of duplicating effort on intriguing questions. In order to prevent this from happening, Voinnet and other researchers in the European Union are striving to form a consortium with the goal of coordinating research efforts on small RNAs.
DEVELOPMENTAL TIMING AND AGING
For several years now, scientists have known that the lin-4 miRNA, along with its target (the transcription factor, lin-14) controls the timing of larval development in C. elegans. But it wasn’t until last December, when Frank Slack and Michelle Boehm published their report in Science, that lin-4 was found to also regulate aging and death in adult nematodes.
Slack, associate professor of biology at Yale University, says that the results of the study “lend strong credence to the notion of an intrinsic biological clock” at work in animals. “Despite a significant variation in life span from species to species, there are clear genetic aspects to the processes of development and aging,” he says. Specifically, reducing lin-4 activity was seen to accelerate aging and, ultimately, shorten life span in nematodes with a loss-of-function mutation. Conversely, overexpressing lin-4 (or reducing lin-14 activity) extended the life span of mutant worms.
Previous molecular genetic screens had identified insulin/IGF-1 signaling as critical for regulating nematode aging and longevity, which led Slack and Boehm to ask whether miRNAs were involved in the pathway. They found evidence that miRNAs indeed regulated gene expression via insulin signaling, at least in C. elegans. Slack’s lab is currently working on identifying other developmental timing genes in the insulin pathway, as well as the mechanism of mouse homologs. Using lin-4 and let-7 as models, Slack says he hopes to discover how miRNAs are expressed and function, which includes deciphering the roles they play in development and disease.
In order to correlate miRNA expression and targeting to biological function, Slack’s lab has pioneered a custom set of molecular, genetic, bioinformatic, and genomic tools and techniques. His lab makes extensive use of genetic screening, transgenic animal models, and has also linked GFP fusions to promoters of miRNAs.
“It has been a fun ride for a lot of years now,” Slack says in describing the state of miRNA research, which he considers quite vigorous. Although admitting that the field’s feverish pace may be leading to some sloppiness in published research, he remains confident that it is “strong and getting stronger.”
Isaac Bentwich, CEO of Rosetta Genomics, started working on miRNAs in 1998, several years before the current rage began. “It’s exciting to see how essential miRNAs have become,” he says. “It’s not every day that a new dimension of biology is unearthed and found to play a part in different facets of human health and disease. We’re very excited to play a role in moving on to new questions.” The specific role that Bentwich envisions for his company is one that spans basic research to more clinical applications, such as the development of diagnostics and therapeutics.
It was in Nature Genetics last year that Bentwich and others from Rosetta described the creation of a high-throughput validation procedure that integrates informatics, machine learning, and sequencing techniques for robust miRNA identification. With this technique in hand, researchers at Rosetta were able to identify a group of new primate-specific miRNAs, which holds obvious promise for further downstream clinical applications.
Rosetta’s recent focus, in addition to the discovery and validation of novel miRNAs, is on the development of diagnostics and therapeutics for human disease — especially for different types of cancer. This March, scientists at Rosetta contributed to a Proceedings of the National Academy of Sciences paper in which algorithmic predictions and experimental techniques were used to perform the largest analysis of human miRNAs to date. The authors of the paper were able to identify 112 previously uncharacterized miRNAs in human colorectal cells, indicating that many novel miRNAs remain to be identified.
Bentwich believes that most technical challenges in miRNA research have been resolved, as researchers are already able to identify large numbers of miRNAs and to isolate RNAs from different biological samples. He adds that the technology for using miRNAs for diagnostic purposes is available as well. “Interestingly, the main challenge when talking about diagnostics or the utilization of miRNAs is a psychological one of getting people to see the potential of miRNAs in molecular diagnostics,” he says.
“The writing is on the wall already: this is really a revolution in the making. In 2000, there were only four papers on miRNAs; now we know there are many, many hundreds of these genes that play a critical role in cell fate determination as well as a variety of diseases, and specifically cancer,” Bentwich says. To this end, Rosetta has inked a deal with Isis Pharmaceuticals to jointly develop a liver cancer drug, and has also signed with Asuragan to focus on the development of molecular diagnostics for prostate cancer.
REGULATORY NETWORKS AND FUNCTION
Bino John of the University of Pittsburgh hopes that his lab’s research on non-coding RNAs will help usher in new diagnostic, prognostic, and eventually therapeutic targets from the treatment of human diseases. His lab aims to develop methods to aid computer-aided drug design efforts.
To do so, John’s research strategy involves applying computational methods to biological problems to generate hypotheses, which are experimentally validated. “Our long-term objective is to use not only miRNAs, but also other non-coding RNAs, to try to identify cancer-specific RNAs,” John says.
Current projects in John’s lab include the discovery of novel miRNAs, identifying the functions of specific disease-associated miRNAs, identifying novel gene regulatory networks in cancer and other diseases, and developing ways to investigate protein and RNA structure.