I understand (vaguely) that at some stage in our ontogenesis our mNRA transmits some of the information encoded in our DNA to our ribosomes so they can build the right proteins etc. However mRNA only takes a subset of that information. What gets left out, and why? Please keep it simple if you can.
Well, it's Saturday night here and I just popped open a beer, but here's what happens (and it's happening all of the time). This is the bare bones account. If you crave more detail, maybe somebody will fill that in, or you can check this out , now that you'll have the basic idea, in a Biology text: DNA makes mRNA (transcription). One side of a DNA strand is the template. A-U, C-G, G-C, T-A for the DNA to mRNA respectively. The mRNA is then processed. Some segments are cut out by enzymes. These are the introns. The remaining segments, the exons, are spliced together. This is the mRNA that will be translated into an amino acid sequence by a ribosome or ribosomes. I've read different ideas, speculation I think mostly, about why this processing occurs. One idea I read is that the DNA that transcribes the introns may represent viral DNA that got spliced into the host cell genome. Some of the Molecular Biology types here could probably give you some info on that.
I will repeat some from the previous post and add some: In prokaryotes there is no splicing of the pre mRNA. They go straight to the ribosomes and are translated into a protein. There isn't also much of post translational modification. In Eukaryotes most genes contain introns. Some only a few, some a lot. The introns need to be sliced out by an enzyme complex that actually contains mostly enzymatic RNAs instead of Proteins. Interestingly this splicing can produce different mRNAs from a single transcript. The transcript is then usually modified even further, with for instance adding of a poly A tail and capping the Messenger. It then gets translated into a protein (or not actually), and still at this point there are many opportunities for further modifications to the protein. In fact a single transcript can sometimes give rise to many different end products.
I still don't really understand this. What does the portion of human DNA that does not get copied by the mRNA do, and what proportion of the DNA gets copied?
http://www.ncbi.nlm.nih.gov:80/books/bv.fcgi?call=bv.View..ShowSection&rid=genomes.box.5303 of 3200 Mb about 48Mb seem to code for mRNA
Spurious Thanks for the link but it doesn't quite give me an answer. Perhaps I'm asking the wrong question as usual. If, as your figures suggest, only about 15% of DNA gets transmitted to create proteins etc and thus kick off and guide our ontogenesis (right word?) then what's the other 85% all about? What does it all do?
Wow. We really don't know? I'll take your word for it but I find it hard to believe. I thought we knew a lot more.
na ja..we know a bit. But as a biologist I can interpret that correctly as we know shit. but as you can see from the schematic, there are some broad categories and hence broad ideas on the function on non-coding genomic DNA.
There are two clear roles for the DNA that isn't transcribed (into mRNA). transcriptional regulation - ie non coding regions that are functional and control in which cells and at what time in those cells a specific gene is transcribed. These include enhancers (involved in turning on transcription), basal promoters (where the transcriptional machinery binds and starts copying), insulator elements (block interactions between enhancers and basal promoters), and silencers (turn off transcription at nearby promoters or shut down specific enhancers). There are others, but most of them are not well characterized. Structural - one type of regulation of transcription is to condense local DNA into heterchromatin a highly conmpacted form of DNA that isn't accessible to transcription (Indeed fully extended the DNA in a cell would not come close to fitting in it (I don't remember the details offhand, but I'm fairly sure all the DNA in every cell in your body set end to end could go roughly to the moon and back). However some DNA is always compacted into heterchromatin. To make an analogy DNA is essentially a huge scroll one line of text wide (with complementary characters on each side) and it's so long that to manage it in a cell certain regions must be bound up into complexes for ease of manipulation. For instance there are extensive interconnections between DNA and the nuclear scaffold a protein network that lines the interior of the nucleus. This structural organization is not well understood, but it is certainly part of the function this DNA plays. I would say we know a lot more that spuriousmonkey said, but are only scratching the surface. Since many of the functions are dynamic and at a structural size scale that is not experimentally tractable with current techniques, we know very little relative to other areas of cell and molecular biology.
Thanks. How much mutating goes on in the transcription of DNA to RNA? That is, are the copies exact, given the occasional quantum surprise, or are the RNA copies often different to each other?
The error rate of RNA polymerase is roughly 1E-5 (compared to for DNA polymerase 1E-9). An average transcript is shorter than 100,000 bp so there is less than an error per transcript. Given the redundancy of the genetic code at the third codon, nonstop mediated decay, and chaperone mediated breakdown of transcripts that just won't fold the system is quite well buffered from mutations in RNA processing having a significant impact. There are however certain transcripts that get edited and I'm not sure what the error rate is in editing. It comes in different forms in different organisms and the purpose is not well understood (so far as I know). From what I know it is plausible that the manner and details of editing give rise to transcript diversity that is useful, but that's highly speculative. Interestingly, the first places such editting was found was T. brucei a parasite that goes to great lengths to very it's coat proteins to evade host rejection. It also comes up in the nervous system (in a different form) in vertebrates another place where diversity is potentially important. Very speculative idea that might not hold up to what more informed people than I know though ...
Mutations in transcription of mRNA is irrelevant: each gene is transcribed repeatedly so functional version of mRNA are always produced even though a few may be defective, also the mRNA is eventually destroyed and recycled so a defective copy will not be in circulation indefinably, last but not least it will not be inheritable from one cell to the next, as long as the master template DNA is not mutatated. As for the purpose of our genome's DNA its been cover over by scilosopher and spuriousmonkey. Most of it is structural and is made of nearly endless small repeating sequences that are not transcribed. A yet unknown percentage is epigenetic and RNA-only genes, the details of which is still cutting edge of research.
Fetus, I agree to some extent with that point, but defective copies could have serious effects in some cases. Some prematurely truncated molecules function as dominant negatives (meaning they block the function of the full length molecule). This is most often the case with proteins that function as part of a complex. Also mRNAs for some genes are at a low enough copy number that one is a significant proportion of the mRNA and if the error rate was high enough it would have significant consequences if the effects were not infrequent and specifically buffered. I also noticed I made a mistake in my last post - the degredation by chaperones is at the protein not mRNA level.
It still would not matter as the existence of both mRNA and protein is brief and limited in production a few defective proteins and mRNA is not going to cause a serious problem, in fact a normal cells has quite a efficient degradation and recycling mechanism for dealing with such used and defective material. Again for the event to be of serious effect it must happen chronically which implies mutation to the DNA its self. it does not have to be a mutation directly to the gene in production. let say there is mutation in the gene for ubiquitin or a mutation for a subunit in 26/20S proteasome or in any gene for a enzyme needed for these complexes production and function. Without the ability to degraded proteins perhaps the effects of such mRNA mutations would be easily observable, of course the cell would most likely die from a build up of aggregated proteins first.
I said I agreed to some extent and then gave compounding cases and examples where it was more of an issue. In the limiting case where there are few transcripts and defective transcripts have a significant effect may not be negligible. Indeed transciption and translation cycles are slower than many biological events so mRNA and protein lifetime is not brief in all cases. Signalling for instance happens over much faster time scales and if a key component is not functioning correctly one could get changes in state that were maintained in the cell. For instance APC levels are very low in Wnt signalling and as it is involved in a complex it is not inconcievable that a defect could have negative consequences. Wnt signalling can over-ride checkpoints in the cell cycle and not stopping at a check point could lead to division when there is DNA damage which could result in cancer. I'm not saying this example is likely, the main point was that in some cases a mutation in an mRNA could have serious consequences. Obviously evolution will have tended to ensure this isn't a severe issue in most cases and I agree it isn't a significant problem, but I would gaurantee mRNA defects do have significant impacts at low frequency.
Were talking about one bad mRNA in a series of several dozen per hour at minimum, one mRNA only produces a few hundred proteins on average with a half life of one-three dozen hours, unless that group of defective proteins has some horribly virulent ability the rate of functional protein produced will dip by only a few percentage points, and this most likely happens constantly because of the instability of RNA with a half life of only a few hours in eukaryotes and a few minutes in prokaryotes.
Several dozen per hour is not a minimum by any means. Some mRNAs are present at a level of 1-2 copies per cell (according to a talk I attended given by Micheal Elowitz) and they do not have such short half lives that there are 12 made per hour (In fact I think numbers are more on the scale of 10 per day). It was in this type of situation I was thinking. I would say that the probability of a a mutation occuring in an mRNA is inversely proportional to the number of RNAs so maybe it's a stupid point anyways. Generally I should have phrased it differently anyhow - I agree with you in the vast majority of cases. All I intended to say was that mutations in mRNA could in some cases have phenotypic effects in response to your stating it's irrelevant. The proper way to express my point would have been that evolution pushes error rates to a level where there are still phenotypic (at the level of a cell in this case) consequences at some acceptable rate and/or level.
Well I'm basing my numbers off peer review reports: http://arep.med.harvard.edu/pdf/Selinger03.pdf http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=116431 http://www.tiem.utk.edu/~gross/bioed/webmodules/mRNA.htm http://mcb.asm.org/cgi/content/full/19/8/5247?view=full&pmid=10409716 http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=3018930&dopt=Abstract Please Register or Log in to view the hidden image! Also a half-life is how long on average until at least 1/2 of a substance is still left, a half-life value is not related to production but to the longevity of the product. if you can cite that either it was transcribed once every 10 days or that it had a half-life of 10 days, which ever one your trying to say, then I'll beleive you.
