Consider them the genomic jacks-of-all-trades, working behind the scenes to influence the inner workings of the human genetic code. While they remained undetected until the early 1990s, microRNAs — small, non-coding RNAs of 21 to 24 nucleotides in length — are now recognized as key regulators of nearly one-third of all protein-encoding genes in the human genome. Though miRNAs have been extraordinarily conserved throughout evolutionary history, researchers are only beginning to decipher how miRNAs have mediated — and even marshaled — the myriad intricate processes that have dictated human development, health, and disease for millions of years.
Researchers in Victor Ambros' lab at Harvard Medical School serendipitously discovered the first miRNA, lin-4, while working with Caenorhabditis elegans in 1993. Then in 2000, Harvard's Gary Ruvkun detected a gene in the nematode that coded for another miRNA, let-7. Though these findings went largely unnoticed at the time, John Rossi at City of Hope in Duarte, Calif., says that they were some "of the most important discoveries of the last 30 years — maybe since the double-stranded structure of DNA — because we didn't even know these molecules existed in mammals until about eight years ago, and now we're finding that they're incredibly important in terms of regulating gene expression."
Recent studies have shown that miRNA-mediated gene expression is constantly at the mercy of confounding influences and competing interests. "MicroRNAs have been shown to play really significant roles when they're up-regulated or down-regulated illicitly," Rossi says.
It's a delicate balancing act; while some miRNAs may function as oncogenes, others can act as tumor suppressors. "Dysregulation of microRNAs can have very profound effects," Rossi adds.
In fact, it's the powerful interplay between miRNAs and their targets that have made these molecules especially attractive as therapeutic candidates, according to Shobha Vasudevan, assistant professor at Harvard Medical School. MiRNAs are "playing a clear role" in human disease processes, "whether you're talking about fine-tuning effects or large effects," Vasudevan says. "Every piece of data points to their therapeutic potential."
But for all their achievements, researchers are still baffled by many aspects that underlie the biology of miRNA. And, judging by the throngs of investigators who've devoted their careers to demystifying these minuscule molecules, it's safe to say that MIT's Phillip Sharp was spot on in 2006 when he said that miRNA research would "occupy thousands of people for years."
"It will take decades to work out the specifics of many different microRNA-regulated processes and integrate those into whole-organism biology," Sharp told the Whitehead Institute's Paradigm magazine at the time.
Concatenation and chance encounters
For many miRNA researchers, the foray into investigations of small RNA-mediated gene expression was a result of the natural progression of their research programs. While he was a postdoc at the University of Texas Southwestern Medical Center, Da-Zhi Wang was working to decipher the transcriptional regulation of gene expression in cardiac, vascular, and skeletal muscle. When he left in 2002 to establish his own lab, he sought to continue studying tissue-specific gene expression. "At that time, microRNAs had just started to emerge as a very novel, new area of investigation," says Wang, who is now an associate professor of cardiology at Children's Hospital Boston. "I basically hypothesized that there must be tissue-specific expressed microRNAs" much like there are both tissue-specific and ubiquitous housekeeping transcription factors. In a 2006 Nature Genetics paper, Wang and his colleagues show that miR-1 and miR-133 play distinct roles in modulating skeletal muscle proliferation and differentiation in cultured myoblasts as well as in Xenopus laevis embryos.
Meanwhile, over at the University of California, San Francisco, Long-Cheng Li was working to unravel how DNA methylation is regulated in cancers. In re-focusing his group's research efforts in 2004, Li says that he and his team "turned our interest to RNA because at that time it was known that in plants there was an RNA-directed DNA methylation mechanism." Whether this process also occurred in human cells remained an unanswered question at the time. It was also the puzzle that would eventually put Li at the leading edge of small RNA research.
Investigators like Harvard's Vasudevan, on the other hand, say that they "fell into" the fledgling field much the same way they made their early miRNA discoveries: unexpectedly. When she moved to Joan Steitz's lab at Yale for postdoctoral work, Vasudevan picked up where her graduate studies had left off by investigating TNF-α AU-rich elements in quiescent cells. Using an in vivo cross linking-coupled affinity purification method, she and Steitz sought to isolate ARE-associated complexes from activated translation conditions and were surprised when they detected Argonaute 2 and fragile-X mental retardation-related protein 1, as both are associated with micro-ribonucleoproteins, or microRNPs. The pair published their results in 2007 in Cell. "We didn't talk about microRNAs at all in that paper. … We had not been on that track at all," Vasudevan says, though she adds that "when your research leads you in a different direction, then it becomes your direction."
While Vasudevan's surprise Ago2 discovery conferred a substantial shift in her research trajectory, it also established the presence of an miRNA in an unexpected regulatory context and consequently broadened the purview of small RNA research.
A plethora of reports published over the last 10 years shows that miRNAs work, primarily, to repress translation by binding the 3' untranslated regions of their target mRNAs. However, in a paper published in Science in 2007, Vasudevan, Steitz, and their colleague Yingchung Tong showed that miR-369-3 — the very miRNA they identified in their Ago complex work — directs the association of microRNPs with AU-rich elements during cell cycle arrest, and thereby activates translation. In the same report, Vasudevan et al. show that let-7 and the synthetic miRcxcr4 — which work to repress translation in proliferating cells — can also induce the translation of select mRNA targets during quiescence.
Around that same time, Li's team at UCSF was targeting specific promoter sequences with synthetic, 21-nucleotide double-stranded RNAs they'd designed in-house to deduce whether RDDM played a role in the management of human gene expression. It was quite by accident, Li says, that his group found that some of these dsRNAs, when transfected into human cell lines, appeared to activate gene expression by binding complementary promoter sequences. Li et al. published the story of their synthetic dsRNA-induced gene activation — or RNAa — in PNAS in 2006.
In building on this discovery — and equipped with the newfound notion of miRNA-induced activation of translation — Robert Place, an assistant researcher in the Li lab, sought to determine whether miRNAs could induce the expression of E-cadherin. Using a miRNA target prediction program, the team identified a "decent target" in miR-373, Li says. In subsequent transfection studies, the researchers found that miR-373 induced the E-cadherin expression, as well as that of CDSC2.
The resulting "paper was the first to show that a microRNA could induce gene expression through the same mechanism as [that of] RNAa," Li says, though he notes that this study did not address the question of whether miRNA-induced expression is an endogenous mechanism.
At present, Li and his colleagues are working to determine whether that's the case. So far, they've "collected sufficient evidence to support that microRNA-mediated gene activation is a naturally occurring mechanism, and … could play a role in biologic processes such as cancer," he says.
Vasudevan says that while "repression and decay are clearly the dominant pathways observed with microRNAs," she's seen an influx of papers that support the role of miRNA in the activation of select mRNA targets as well as other, more novel functions — such as miRNA-mediated relief of repression — which indicates that these molecules likely perform "further functions that we haven't yet given them credit for," she says.
She points to a paper published in Cell in March as an example of a novel miRNA function. In it, a team led by investigators at Ohio State University shows that the loss of miR-328 in blast crisis chronic myelogenous leukemia interferes with the function of hnRNP E2, a regulatory protein, and leads to the inhibition of CEBPA mRNA translation. In addition, the group showed the miR-328 functions as a decoy molecule that "releases CEBPA from the translation inhibitory effects of hnRNP E2," the authors write. Vasudevan says that these results indicate a "completely new mechanism of relief of repression, because normally, you think about base-pairing between a microRNA and its target, but here it was binding a protein and dragging it away," she says, adding that the specificity criteria that govern this type of miRNA-mediated expression "have yet to be discovered. … All of these things are mysterious right now. It's quite fascinating."
Children's Hospital Boston's Wang echoes Vasudevan's sentiments. He says that the miRNA community tends to be "very open-minded to new discoveries." Because of recent, novel findings — such as those made by Vasudevan, Li, and others — Wang says that researchers no longer "have a fixed, back-of-our-mind idea of what microRNAs are supposed to be doing."
Drivers of development and disease
As a developmental biologist by training, Wang has taken a broad, hypothesis-driven approach to his investigations of miRNA expression during satellite cell self-renewal, proliferation, and differentiation. In a Journal of Cell Biology paper published in September, Wang's team showed that the expression levels of two miRNA — miR-1 and miR-206 — correlate with satellite cell state. "When you over-express these two microRNAs in satellite cells, it inhibits the cells' self-renewal and proliferation and, in fact, [they] also promote those cells to go on to differentiation," Wang says.
At City of Hope, members of the Rossi lab are examining the roles that miRNAs play in blood cell development. Using short-hairpin RNA to knock out Drosha, a key enzyme in miRNA biogenesis, his team has found that the loss of specific miRNAs can perturb an immature red blood cell lineage, such that developing erythrocytes go on to become myeloid cells. Currently, Rossi says, his group is "studying which microRNAs are lost and why these are so important for red blood cell development," and whether they could have implications for blood disorders.
Wang says that, indeed, "microRNAs have been implied in many different diseases," but adds that they've been "most comprehensively studied in cancer."
Rossi says it was a 2005 Nature paper in which researchers at the Broad Institute "showed that in virtually every type of cancer there is dysregulation of microRNAs" that really "set the whole field afire."
Friend or foe?
While several reports published in the last few years have shown that certain miRNAs — dubbed oncomIRs — seem to promote tumorigenesis and proliferation in cancer, there are, perhaps, just as many papers that show that specific miRNAs can act like tumor suppressors. Still, researchers maintain that miRNAs benefit humans more than they inflict harm.
"They are definitely [our] friends," Vasudevan says of miRNAs. "The body wouldn't be making them if they weren't." However, she adds, in the event that miRNAs become aberrantly expressed, "too much of a good thing can be bad."
Rossi, too, says that miRNAs are usually only detrimental when they're dysregulated. "There's this very fine network of interactions that take place and when that's perturbed over a long period of time, it can lead to dysregulated growth. … MicroRNAs are really a balance," he says, adding that "things can go awry with any friend."
MicroRNAs as medicine
Given the demonstrable roles that miRNAs have been shown to play in the lab, it's no surprise that researchers are increasingly looking toward these molecules as indicators — and even modulators — of gene expression in the clinic.
Rossi says that these small molecules are best suited for diagnostic and prognostic purposes. Clinicians can detect circulating miRNA released from cancer cells in plasma. Further investigations of miRNA expression in normal versus cancer specimens, coupled to the clinical quantification of plasma-miRNA levels, could prove to be a powerful diagnostic tool and may even bolster physician's prognostications, he suggests.
More and more, though, investigators are exploring the possibilities of tailoring therapeutics to target miRNAs and are even considering using the molecules themselves as medicine. For the former scenario, much work has focused on how best to inhibit miRNAs — either by degrading them, knocking them out entirely, or blocking access to their targets with antagomirs or through other means. Work on the latter approach has thus far centered on re-supplying miRNAs that are ineffective or missing, through a mechanism similar to that which underlies short interfering RNA delivery. To that end, "perhaps restoring microRNAs and returning a cell back to a stem cell-like state might be one way of treating cancer, rather than trying to bludgeon it," Rossi says.
Wang at Children's Hospital Boston says that while he's "quite excited" about the therapeutic potential he's seen in miRNA knockout animal models, he's also "very cautious about it. ... One single microRNA can influence the expression of many genes."
UCSF's Li expresses similar reservations. "I'm concerned about the potential for adverse effects by manipulating a particular microRNA," he says.
Vasedevan's counter to these concerns has been to circumvent them with an increased understanding of the fundamental mechanisms behind the context-dependent base-pairing specificities of miRNAs and other small molecules. To that end, rather than knocking down a particular miRNA, her team has focused on blocking the ability to bind its target through antisense manipulations, much like those first demonstrated in zebrafish by Harvard's Alexander Schier.
Vasudevan expects the therapeutic angle to remain a hot-button topic in miRNA research for the next several years. How miRNAs are regulated and whether they play roles in gene expression at additional levels, such as that of epigenetic regulation, will likely become increasingly important, Vasudevan says.
For his part, UCSF's Li says he's dedicated his efforts to "understanding how microRNAs could affect the genome and epigenetics." In these areas, he says, "I believe we are just seeing the tip of the iceberg."
Rossi and his team have initiated investigations into miRNA biogenesis, and to what extent mutations in the genes that encode these small molecules affect their activities. He expects to see a growing interest in the activities of miRNAs encoded in viral genomes as well as the roles that small RNAs play in inflammation and other chronic conditions. Considering that miRNAs weren't even on the genomics radar until the early '90s, the strides researchers have made in miRNA research over the past few years indicate that "there's a huge future out there that is rapidly moving in the right direction," Rossi says.