Pure Pedantry

Andrew Fire and Craig Mello have won the Nobel Prize in Medicine for the discovery of RNA interference:

Americans Andrew Z. Fire and Craig C. Mello won the Nobel Prize in physiology or medicine Monday for discovering a powerful way to turn off the effect of specific genes, opening a new avenue for disease treatment.

”RNA interference” is already being widely used in basic science as a method to study the function of genes and it is being studied as a treatment for infections such as the AIDS and hepatitis viruses and for other conditions, including heart disease and cancer.

Fire, 47, of Stanford University, and Mello, 45, of the University of Massachusetts Medical School in Worcester, published their seminal work in 1998.

RNA interference occurs naturally in plants, animals, and humans. The Karolinska Institute in Stockholm, which awarded the prize, said it is important for regulating the activity of genes and helps defend against viral infection.

RNA interference is an only recently understood — and frankly peculiar — way of regulating the expression of genes. It was originally discovered because if you insert double stranded RNA (dsRNA) into cells you get a generalized destruction of RNAs and cessation of protein synthesis. This is because dsRNA in cells is associated with viruses — it shouldn’t be there and the cell knows it.

More interestingly, dsRNAs were shown to have some gene specific effects in cells where the viral suppression was not an issue — in stem cells and the like.

It works something like this. Figure is from here (click to enlarge).


dsRNAs whether artificially introduced (by the experimenter) or created by the cells are chopped up into tiny pieces by a protein called Dicer. They then go to bind a complex called RISC where they can have a variety of effects on mRNAs with sequences similar to them. They can suppress translation of new proteins. They can cause the degradation of those mRNAs. They can induce chromatin modification at the points in the genome where those mRNAs were generated, causing long lasting changes in transcription. All of these have the effect of downregulating the activity of particular genes with similar sequences to the tiny dsRNA.

This was a huge discovery for two reasons.

1) It produced an invaluable experimental tool. Using RNAi, researchers can specifically knockdown the production of any protein in a cell. This is important for use in cells for which transgenics that lack that protein cannot be easily created.

2) It opens up a huge new level of regulation in the human genome. For many years, we thought the regions of the genome that did not code for proteins — called introns — were just junk DNA. It turns out that much of this DNA codes for RNAi elements. Interestingly, each little RNAi sequence can downregulate not only one gene, but an entire class of genes through sequence similarity. This means that the expression of these tiny RNAs might have more to do with maintaining the developmental state of a cell than the expression of any individual gene. Scientists are, as we speak, trying to figure out how this whole business works.

Congratulations, Drs. Fire and Mello.


  1. #1 Monado
    October 2, 2006

    So if extrons are coding for something useful, that implies that their drift isn’t completely random and the genetic clock does not run at the same rate for all extrons. Interesting!

  2. #2 Jim Lund
    October 2, 2006

    Good post. I would clarify that in humans 25% of the genome sequence is genes (1% exons and 24% introns), and of the remaining 75%, only a small fraction, less than 1% is involved in RNAi or RNAi related gene regulation. Another small fraction, less than 1% are transcribed sequences with no known function. Most of the ‘junk’ DNA is still ‘junk’–no function known and possibly no function at all.

  3. #3 Pinko Punko
    October 2, 2006

    Jim, you haven’t been paying attention- a much higher percentage of the genome is actually transcribed, and much of the transcriptional noise may actually play a part in gene regulation, both epigenetic and otherwise.

    Also, let’s not forget that before the observations in worms about silencing in response to dsRNA, there were observations in fungi and plants about very strange epigenetic phenomena that all happen to now be understood in terms of dsRNA pathways.

  4. #4 Jim Lund
    October 2, 2006

    You are referring to the tiling experiments where polyA RNA is hybridized to arrays of probes covering the genome? Yes, I’m familiar with those studies. Johnson et al. 2005 has a good review. These studies find new transcribed sequences–effectively new exons that are evidence for new genes amounting to 50% to 100% more than standard predictions. For example, Bertone et al., 2004 found 10,595 new transcripts.

    The full size of the coding regions of these genes may be larger than the small total sequence identified to date as exon-intron structure of these new genes is not known, and it will likely add a few percent to the total transcribed seq. From our limited knowledge today I estimated this as an additional ‘1% are transcribed sequences with no known function’.

    There’s also evidence of substantial anti-sense transcription–a majority of human genes seem to have associated antisense transcripts.

    And there are transcribed repetitive DNA sequences which are substantial but have little or no known function.

    I’m familar with the RNAi story; I work with C. elegans worms. Don’t get me wrong, the RNAi discoveries made by Fire, Mello, and others are great work and are revealing a whole new layer of gene regulation in addition to the standard transcriptional/translational regulation.

    Still, at this point the extent to which the common antisense transcription found in tiling array experiments is regulatory (rather than being noise) is still not known. There’s little evidence for these transcripts have a function–the discovery is still too new. There are better characterized classes of non-coding RNA. siRNA and miRNA have emerged as important transcriptional regulatory mechanisms.

  5. #5 sparc
    October 2, 2006

    I do agree that RNA interference is indeed worth the award because its impact on the understanding of gene regulation and its applications in basic science are unquestionable. However, I think that one has to be careful when it comes to therapeutics. Like with many other technologies in the last 20 years there is quite some hype that does not mention the unsolved problems of the technique, e.g. off target effects, delivery methods, maintenance of expression and possible oversaturation of the RNAi machinery. For the latter please refer to:
    Grimm D, Streetz KL, Jopling CL, Storm TA, Pandey K, Davis CR, Marion P, Salazar F, Kay MA (2006): Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways.
    Nature 441(7092):537-41.

  6. #6 Melinda Wenner
    October 3, 2006

    Yes, I completely agree with Sparc. I did a lot of research on RNAi a few months ago. It has so much scientific potential, but as for its therapeutic potential… well, no one really knows. It’s a long way away, at the very least.

  7. #7 Pinko Punko
    October 4, 2006

    Jim, sorry to have sounded so aggressive- I’m also refering to transcription of regulatory regions such as Hox-gene upstream reagions and various LCRs as well as the basal level of transcription of supposedly silenced regions like centromeres and transgenes, transcription that may be required for the maintenance of the silenced state. I think there will be a much higher level of intergenic txn than is currently recognized. Much of these transcripts are most likely unstable and non-functional at the level of their existence, while the key point is that transcription has occured. RNA Polymerase II is likely to be the largest scale chromatin remodeler in eukaryotic genomes. I wasn’t trying to just on you, just throwing in a couple of points.

New comments have been disabled.