Central Dogma of Genetics May Not Be So Central 196
Amorymeltzer writes "RNA molecules aren't always faithful reproductions of the genetic instructions contained within DNA, a new study shows (abstract). The finding seems to violate a tenet of genetics so fundamental that scientists call it the central dogma: DNA letters encode information, and RNA is made in DNA's likeness. The RNA then serves as a template to build proteins. But a study of RNA in white blood cells from 27 different people shows that, on average, each person has nearly 4,000 genes in which the RNA copies contain misspellings not found in DNA."
Why is this news? (Score:5, Informative)
We have known for many years that the same DNA codes to different proteins, with the adjustments given the information in the non-coding regions AND the information in the epigenome. That people have discovered that the intermediate step is also adjusted can hardly be called a shock. The proteins have to get built differently somehow, so some alteration in the intermediate coding was inevitable. Honestly! If geneticists aren't even reading their own bloody papers, maybe the government grants should be issued to those Slashdot readers who do.
This is NOT what the central dogma says (Score:3, Informative)
What it does in fact say is that information flows from DNA to RNA to proteins, and not the other way around: proteins can't write DNA.
It's called an "error rate" (Score:4, Informative)
When DNA is copied to make new DNA, you get a certain number of copying errors, called mutations - most of them harmless. I assume everyone knows about those.
When DNA is copied to make a temporary-working-copy RNA, you get a larger number of these copying errors because, in general, they are one-shot non-critical deals. The need for stringency is much lower, the selective advantage for stringency is not so great, so it comes as no surprise that the level of proof-reading is also reduced.
Now, it's also possible that there are mechanisms by which these RNA molecules can be purposefully edited. As mentioned in the article, significant post-transcriptional editing (including in eukaryotes the readaction of big chunks, which are called "Introns".) But this finding doesn't speak much to that, although the rate is a *sconch* higher than I might expect for random errors. Even so, this doesn't shake the central dogma of molecular biology in any meaningful way, as for example Reverse Transcriptases did.
RTFA, the errors weren't random. (Score:4, Informative)
The amazing thing is not that there are mistakes, but the exact same mistakes occur in (almost) every strand of RNA! They aren't random errors, they occur the same way every time!
Not news. (Score:2, Informative)
From the article: The most common of the 12 different types of misspellings was when an A in the DNA was changed to G in the RNA. That change accounted for about a third of the misspellings.
This is a textbook example of RNA editing by adenosine deaminase. It will convert the Adenosine bases ('A') to Inosine ('I'). When they try to sequence the RNA the first step is to make a DNA copy. During the process the positions that contain 'I' are copied mostly as 'G'. This is because 'I' can pair with any base, but prefers 'C'. So in the first strand you will get 'C' paired with 'I'. When you build the second strand these 'C' positions will direct incorporation of 'G'.
Mystery solved
Re:RTFA, the errors weren't random. (Score:5, Informative)
Even that's not amazing. It would be amazing if it made a different mistake every time.
The simple model of transcription had always been that single nucleotides in DNA are matched to the complementary nucleotide on the RNA strand. But, of course, nobody thought the simple model was always correct. You've got the interaction of a DNA strand trying to fold back on itself and an RNA strand trying to fold back on itself, and a big honking RNA polymerase molecule with an extremely complicated electric field. It's to complicated for the simple model to work. Maybe on occasion the order of the codons a few hundred bases from the transcription site will interact with the RNA polymerase to insert a different base than expected. (Just throwing that out as a possibility. It could be any of a million things, like an induced change in the structure of RNA polymerase.) That's fine, as long as it happens the same way every time. In that case it's not an error in the DNA or the RNA. It's an error in our oversimplified model of how RNA transcription works. So now we need a better model that can predict how a DNA sequence will be transcrived. Don't look now, science is working the way it should!
I hate that they are even using the word dogma. Because actually dogma is never based on or swayed by evidence. And in this case the dogma was "it's simpler than any realistic biochemical system." I'd like to see a poll of how many biochemist, molecular geneticists, virologists and microbiologists actually believed this dogma.
Re:Mutations (Score:4, Informative)
Re:Central Dogma? (Score:3, Informative)
There are many, many twists to this sordid puzzle, but you are correct. The concept of a 1:1:1 translation has been dead for a very long time.
Re:Why is this news? (Score:2, Informative)
I agree that we've always suspected that transcription isn't a high fidelity process. In fact, there is evidence that leads us to this conclusion (ex. the lack of a 'spell-checker' mechanism). However, just because we have evidence that points to an effect doesn't mean that it shouldn't be tested. The thing is, we've been surprised before. We've had evidence of other phenomena/behaviour should exist but when actually tested, it turned out that it was not as expected. For example, in the past it was thought that during ischemic events it was the lack of oxygen and nutrients that did the most damage, now it is known that reperfusion and the immune response subsequent to ischemic injury has a significant role in the damage done. As pointless as some of these experiments must seem, they still have to be run to test the conclusions of those other 'bloody' papers that the geneticists are reading.
On a side note, the genetic code is built in such a way that small errors here and there during the transcription process may not have a huge effect (64 codons represent ~20 amino acids plus a few stop codons).
Re:Why is this news? (Score:5, Informative)
I don't see why you claim there isn't a spell checker. Using DNA for the long term storage itself increases fidelity over RNA. Putting the DNA in a nucleus to protect it from some chemical processes that can cause data malformation also means an increase in fidelity. Multicellularity means (admittedly among other things), moving the reproductive cells deep in the organism so they are again protected from some more sources of copying errors. Simultaniously, it allows apoptosis (as there's no advantage for cell death in a single celled organism), and that's a second spell checker of sorts for multicelled organisms only. A lot of the more complex organism's defenses against diseases such as cancer could all be described as spell checkers (for example, P53 tumor suppressor). The form of DNA polymerase used in the complex organisms itself improves copying accuracy by about 100fold over what's possible for the non-eukarotes and even some of the fairly complex bacteria, and it's been described in operation as 'wiggling the part it has just put together to make sure it hasn't allowed the wrong base to pair before it moves on to the next bit, and having a digestive capability to strip out such mistakes when it finds them'. (See "Our Molecular Nature", by David Goodsell for more on this). Then there's snRNPs (Small Nuclear Ribonucleoproteins, which are formed to snip out introns from RNA copies for those RNA strands that aren't self splicing ribozymes (and of course rybozymes themselves even in organisms too simple to have snRNPs). It looks to me like most of the major changes in organic complexity are also spell checkers of one sort or another. I don't really like to anthropomorphise evolution as having long term goals, but it's probably at least as fair to say evolution is trying to produce totally accurate transcription, as it is to say it is trying to make organisms more ideally suited to their environments.
Re:Why is this news? (Score:1, Informative)
There are four bases ("letters") in both DNA and RNA. In DNA they are Adenine, Thymine, Guanine, Cytosine. In RNA they are Adenine, *Uracil*, Guanine, Cytosine. Uracil performs essentially the same hydrogen-bonding in RNA as Thymine does in DNA, allowing it to base pair with both DNA and RNA. Get your biology straight.
Re:Central Dogma? (Score:3, Informative)
"In his autobiography, What Mad Pursuit, Crick wrote about his choice of the word dogma and some of the problems it caused him:
"I called this idea the central dogma, for two reasons, I suspect. I had already used the obvious word hypothesis in the sequence hypothesis, and in addition I wanted to suggest that this new assumption was more central and more powerful. ... As it turned out, the use of the word dogma caused almost more trouble than it was worth.... Many years later Jacques Monod pointed out to me that I did not appear to understand the correct use of the word dogma, which is a belief that cannot be doubted. I did apprehend this in a vague sort of way but since I thought that all religious beliefs were without foundation, I used the word the way I myself thought about it, not as most of the world does, and simply applied it to a grand hypothesis that, however plausible, had little direct experimental support."
It's worth noting that this kind of thing happens a lot in biology, where a name gets appropriated without the borrower fully understanding its meaning—or in some cases, the correct pronunciation. Classicists are frequently driven mad when they discover the plural of "locus" is pronounced "low-sigh".
Re:Central Dogma Barking Up Wrong Tree (Score:1, Informative)
Actually, you too are incorect, albeit closer.
The issue is not a change in accuracy (RNA copying is well known to be significantly lower in fidelity than DNA), but rather than the same deviations from the expected change are happening consistently. An example (simplified):
Lets say that the coding strand is these 20 bases:
AGGCATAGGC
Further, we'll say there's 1,000 bases in either direction of the translated sequences. We'll say, excluding what is shown below, in this 2,010 base sequence, there are an average of 1 base in each sequence other than that which is expected (a reasonable error rate). Now, lets say we get the following copies:
1) AGGCGTAGGC
2) AGGCGTAGGC
3) AGGCATAGGC
4) AGGCGTAGGC
5) AGGCATAGGC
6) AGGCGTAGGC
7) AGGCGTAGGC
8) AGGCGTAGGC
9) AGGCGTAGGC
10) AGGCGTAGGC
We have an average of 1 base per 1,000 error, IIRC, par for the course, well within the model. Nothing to write home about. Except:
Note the "G" in the fifth position in most strands. That is not what was expected, nor is it in every strand but it is still way to regular to be an accident. That kind of issue is what is being described in the article. For the most part our model works, but there are too many consistent errors for us to think we have a complete understanding.
This means that the change was not an accident, but it also does not fit with our model. Thus, we need to find out what it is. It's not a matter of quantity, but precision.