Tiny RNAs that regulate gene function

RNA — the little molecule that could

In the early 1980s, Ambros joined the laboratory of Robert Horvitz at the Massachusetts Institute of Technology as a postdoctoral fellow. He wanted to outline the means by which genes choreograph the construction of fully formed adults from single cells. Analyses of flies had revealed that certain genes instruct embryos where to place body parts — for example, wings belong on each side and legs belong on the bottom. But Ambros was intrigued by the notion that other genes might specify the timing — rather than the location — of developmental events; alterations in such genes might cause cells and tissues to adopt fates that are normally associated with earlier or later stages of development.

He directed his attention toward one of the first-known genes of this type, called lin-4, which had been identified earlier in the laboratory of Sydney Brenner (Lasker Special Achievement Award, 2000) and subsequently characterized by Horvitz, Martin Chalfie, and John Sulston. Ambros recognized that, during worms’ trek toward adulthood, those with inactive lin-4 get stuck repeating early larval stages. Consequently, they lack cell types and structures typical of fully formed animals and instead contain extra copies of cells ordinarily produced only at early stages. These observations suggested that normal lin-4 allows immature worms to advance past a particular developmental stage; animals with the defective version cannot overcome that hurdle. Ambros discovered that worms lacking a different gene — lin-14 — were the antithesis of those with inactive lin-4. The animals skip early steps in development and prematurely acquire characteristics that normally appear later. These and other results suggested that lin-4 and lin-14 exert opposite effects in worm cells.

To dig further into lin-14’s function and its possible relationship with lin-4, Ruvkun, who by this time (1982) had joined Horvitz’s laboratory as a postdoctoral fellow, collaborated with Ambros to isolate the lin-14 gene. After the investigators set up independent laboratories in the mid 1980s, Ruvkun, at Massachusetts General Hospital in Boston, established that the protein product of lin-14 is abundant during early larval stages and then its quantities plummet. Under conditions in which it unnaturally remains plentiful, early steps repeat, suggesting that the normal drop in the lin-14 protein allows worms to proceed to later stages. Ambros, at Harvard University, found that lin-4 dampens lin-14 activity and thus a picture emerged about how the genes collaborate. At the appropriate time, lin-4 blocks lin-14 and thus allows worms to continue their developmental trajectory.

Ruvkun sought to identify the portion(s) of lin-14 that lin-4 targets, so he tracked down certain genetic anomalies in lin-14’s sequence that underlie excess production of the lin-14 protein. He found that these alterations reside in the area of the gene that follows the protein blueprint, a span called the 3′ untranslated region (3′ UTR). The perturbations do not influence amounts of the protein’s messenger RNA (mRNA), the molecule that carries genetic information from DNA to the cell’s protein-making factory, Ruvkun showed. Rather, they alter protein quantities. Therefore, molecules that turn off lin-14 after early stages of development presumably exert their effects through the 3′ UTR region of the lin-14 mRNA and prevent the cell from translating its code into protein.

In the meantime, Ambros’s laboratory was isolating the lin-4 gene, which they assumed encoded a protein; although a few RNAs were known to control gene activity in bacteria, conventional wisdom held that, in animal cells, proteins alone enjoy such powers. The team homed in on smaller and smaller pieces of DNA from normal animals that restore typical developmental behavior to a worm that lacks lin-4. Stretches of DNA that were far shorter than standard genes worked. Eventually, the researchers began considering the possibility that its product was an RNA, but they still assumed that the regulatory molecule they sought would be a respectable size. The smallest RNAs known to do anything important in cells contained about 75 nucleotide (nt) building blocks. Eventually, however, their experiments led them to a tiny RNA, composed of about 22 nucleotides. A larger — 61 nt — molecule that contained the smaller RNA appeared as well and Ambros noticed that it could fold into a double-stranded ‘hairpin’ — a structure whose significance would become clear years later.

In an exciting exchange of data, Ambros and Ruvkun realized that the 22-nt lin-4 RNA matched sections within the 3′ UTR of the lin-14 mRNA: These sequences could bind one another by the same base-pairing rules that hold together the Watson and Crick DNA strands. In this view, the tiny lin-4 RNA settles on the target lin-14 mRNA — in its 3′ UTR — and the resulting double-stranded structure somehow interferes with translation of the lin-14 mRNA’s genetic information into protein (see illustration).

Small but mighty. This scheme shows how one type of tiny RNA, a microRNA (miRNA), silences genes. It is cut out of a precursor hairpin-shaped pre-miRNA to form a mature miRNA, which binds to the 3′ untranslated region (3′ UTR) of a target gene’s messenger RNA and turns off its activity. [Credit: Carin Cain. Based on an illustration from Victor Ambros]

Branching out to plants and beyond

Across the Atlantic, David Baulcombe, then of the Sainsbury Laboratory in Norwich, UK, was studying how plants resist viruses. When he and others added to viral-infected plants unusual versions of viral genes, the mRNA copies of the normal genes as well as the newly introduced ones disappeared. Similarly, experimentally added non-viral genes suppressed activity of plant genes that contained similar sequences. Baulcombe proposed that such gene silencing occurs when RNAs embrace target mRNA — through typical Watson-Crick base-pairing — and promote destruction of the mRNA or interfere with its translation into protein. However, no one could find such RNAs.

Baulcombe reasoned that the predicted RNAs might have eluded researchers because the molecules were shorter than anyone imagined and thus, experiments had not been designed to detect them. In 1999, he and a postdoctoral fellow in his laboratory, Andrew Hamilton, devised a hunt specifically for small RNAs. They added test genes to plants and found 25-nt-long RNAs that matched; furthermore, these small RNAs appeared only under conditions in which target mRNA activity was shut off. The stunning similarity in size between the plant and worm RNAs suggested that small regulatory RNAs exist in many organisms. Furthermore, it hinted at the presence of cellular machinery that dedicates itself to creating these precisely sized molecules and then uses them to quash gene activity.

In 2000, Ruvkun’s laboratory discovered a second tiny regulatory RNA in worms of exactly the same size as the lin-4 RNA and in the same genetic pathway. Similar to the lin-4 RNA, this let-7 RNA dampens activity of its target gene through its 3′ UTR. Furthermore, its sequence too resides within a larger molecule that folds up on itself to form a double-stranded hairpin structure. Later that year, Ruvkun found that many other creatures, including humans, fruit flies, chickens, frogs, zebrafish, mollusks, and sea urchins, carry their own versions of let-7, which could also fold into hairpins. The apparent binding site for let-7 RNA in its target was conserved in some of these organisms as well. Moreover, let-7 RNA appeared and disappeared at similar points during development in many of the animals.

The small RNAs, now called microRNAs (miRNAs), had broken through their designation as “worm curiosities.” Researchers realized that the miRNAs likely execute vital functions during growth and development of other creatures as well. Multiple teams raced to expose regulatory RNAs of approximately 22 nucleotides in length. In 2001, Ambros’s group, now at Dartmouth Medical School, in Hanover, as well as those of David Bartel (Massachusetts Institute of Technology) and Thomas Tuschl (Max Planck Institute for Biophysical Chemistry, Göttingen) discovered almost 100 of these small regulatory RNAs in flies, humans, and worms.

In addition to revealing that small regulatory RNAs dwell in organisms other than worms, Baulcombe’s finding caught many researchers’ attention because it seemed related to a process called RNA interference (RNAi), which had recently exploded onto the biological scene. In RNAi, long RNAs injected into cells hamper gene activity from similar sequences. No one knew why organisms possessed this ability, but presumably it played some role in normal physiology. In 1998, Andrew Fire (Carnegie Institution of Washington, Baltimore) and Craig Mello (University of Massachusetts Medical School, Worcester), published a watershed paper that defined the fundamental features of RNAi (which garnered them the Nobel Prize in 2006). That work yielded the surprising insight that the process depends on double-stranded RNA. However, the means by which double-stranded RNA triggered silencing remained mysterious.

Experiments from Baulcombe’s laboratory provided the crucial clues. Production of the silencing RNA strand depended on the presence of the other strand, he had noticed. This observation suggested that, at some point during manufacture of the small regulatory RNA, it exists as part of a double-stranded molecule. Suddenly it seemed possible that Baulcombe’s tiny RNAs arose by trimming longer molecules of the type that Fire and Mello had discovered. Furthermore, this notion suggested that the hairpin-like lin-4 and let-7 RNAs similarly gave rise to the mature, 22-nt entities.

Scientists wondered whether the cell deployed the same biochemical machinery to create and use RNA molecules that subdued gene activity in all of these gene-silencing systems. However, the mechanisms of the worm miRNAs seemed to differ from those of the plant molecules as well as RNAi. Unlike the system that Ambros and Ruvkun had been untangling, which allowed mRNA to accumulate but thwarted cells’ abilities to translate the information it contained into protein, the plant system and RNAi destroyed mRNA. For that reason and others, many people doubted that the processes were connected. Still the possibility that they shared a common mechanism and machinery tantalized researchers.

In 2001, the Mello, Ruvkun, and Fire groups collaborated to show that efficient liberation of the lin-4 and let-7 RNAs from the hairpin molecules relies on the C. elegans version of Dicer, an enzyme that Gregory Hannon (Cold Spring Harbor Laboratory) discovered and named for its ability to chop dsRNA into uniformly sized, small RNAs that direct mRNA destruction during RNAi. These results and others, including similar ones generated by Philip Zamore (University of Massachusetts Medical School, Worcester), cemented the connection between miRNAs and RNAi, thus providing one biological ‘reason’ for the RNAi machinery. Moreover, they identified the apparatus by which cells generate miRNAs and harness them for key pursuits.

Studies in the past several years have indicated that the human genome contains more than 500 and perhaps as many as 1000 miRNAs that could collectively control a third of all of our protein-producing genes. These regulatory molecules have been implicated in a wide range of normal and pathological activities. They play roles not only in embryonic development, but in blood-cell specialization, cancer, muscle function, heart disease, viral infections, and possibly neurological signaling and stem-cell behavior. Researchers are exploring the possibility of using miRNAs ‘signatures’ for diagnosis and prognosis and are considering manipulating their quantities for therapeutic purposes.

Looking where no one had looked before, Ambros, Baulcombe, and Ruvkun spied an unforeseen universe of potent molecules. Their work has elevated these hitherto unrecognized agents into the spotlight of biology and medicine.

by Evelyn Strauss

Key publications of Victor Ambros

Ambros, V. (1989). A hierarchy of regulatory genes controls a larva-to-adult developmental switch in C. elegans. Cell. 57, 49-57.

Lee, R., Feinbaum, R., and Ambros, V. (1993). The heterochronic gene lin-4 of C. elegans encodes small RNAs with antisense complementarity to lin-14. Cell. 75, 843-854.

Moss, E., Lee, R., and Ambros, V. (1997). Control of developmental timing by the cold shock domain protein Lin-28 and its regulation by the lin-4 RNA. Cell. 88, 637-646.

Olsen, P.H., and Ambros, V. (1999). The lin-4 regulatory RNA controls developmental timing in C. elegans by blocking LIN-14 protein synthesis after the initiation of translation. Dev. Biol. 216, 671-680.

Lee, R.C. and Ambros, V. (2001). An extensive class of small RNAs in Caenorhabditis elegans. Science. 294, 862-864.

Lee, R., Feinbaum, R., and Ambros, V. (2004). A short history of a short RNA. Cell. S116, S89-S92.

Key publications of David Baulcombe

Brigneti, G., Voinnet, O., Li, W.X., Ji, L.H., Ding, S.W., and Baulcombe, D.C. (1998). Viral pathogenicity determinants are suppressors of transgene silencing in Nicotiana benthamiana. EMBO J. 17, 6739-6746.

Dalmay, T., Hamilton, A.J., Rudd, S., Angell, S., and Baulcombe, D.C. (2000). An RNA-dependent RNA polymerase gene in Arabidopsis is required for posttranscriptional gene silencing mediated by a transgene but not by a virus. Cell 101, 543-553.

Hamilton, A.J. and Baulcombe, D.C. (1999). A species of small antisense RNA in posttranscriptional gene silencing in plants. Science. 286, 950-952.

Hamilton, A.J., Voinnet, O., Chappell, L., and Baulcombe, D.C. (2002). Two classes of short interfering RNA in RNA silencing. EMBO J. 21, 4671-4679.

Ratcliff, F., Harrison, B.D., and Baulcombe, D.C. (1997). A similarity between viral defense and gene silencing in plants. Science. 276, 1558-1560.

Baulcombe, D.C. (2006). Short silencing RNA: The dark matter of genetics? Cold Spring Harb. Symp. Quant. Biol. 71, 13-20.

Key publications of Gary Ruvkun

Ruvkun, G. and Giusto, J. (1989). The Caenorhabditis elegans heterochronic gene lin-14 encodes a nuclear protein that forms a temporal switch during development. Nature. 338, 313-319.

Wightman, B., Bürglin, T.R., Gatto, J., Arasu, P., and Ruvkun, G. (1991). Negative regulatory sequences in the lin-14 3′ untranslated region are necessary to generate a temporal switch during C. elegans development. Genes Dev. 5, 1813-1824.

Wightman, B., Ha, I., and Ruvkun, G. (1993). Post-transcriptional regulation of the heterochronic gene lin-14 by lin-4 mediates temporal pattern formation in C. elegans. Cell. 75, 855-862.

Reinhart, B.J., Slack, F.A., Basson, M., Pasquinelli, A.B., Bettinger, J.C., Rougvie, A.C., Horvitz, H.R., and Ruvkun, G. (2000). The 21 nucleotide let-7 RNA regulates C. elegans developmental timing. Nature. 403, 901-906.

Pasquinelli, A., Reinhart, B., Slack, F., Maller, B., Kurodo, M., Martindale, M., Srinivasan, A., Fishman, M., Hayward, D., Ball, E., Degnan, B., M�ller, P., Spring, J., Finnerty, J., Corbo, J., Levine, M., Leahy, P., Davidson, E., and Ruvkun, G., (2000). Conservation across animal phlylogeny of the sequence and temporal expression of the 21 nucleotide let-7 heterochronic regulatory RNA. Nature. 408, 86-89.

Ruvkun, G., Wightman, B., and Ha, I. (2004). The 20 years it took to recognize the importance of tiny RNAs. Cell. S116, S93-S96.