Study Suggests Genome Instability Hotspots

Dr. Eggman writes "Ars Technica reports on a new study that suggests not only that certain areas of the mouse genome undergo more changes, but that changes to those areas are more tolerable by the organism than changes in other areas. Recently published in Nature Genetics, the study examined the certain copy number variations of the C57Bl/6 strain in mice that have been diverging for less than 1,000 generations. The results were a surprising number of variations. While the study does not address it, Ars Technica goes on to recount suggestions that genomes evolved to the point where they work well with evolution."

heh

[Kingrames (858416)]

"...evolved to the point where they work well with evolution"

ya think?

Re:heh

[KillerCow (213458)]

"It is not the strongest among the species that survive nor is it the most intelligent. It's those that are most adaptive to change." - Charles Darwin

Re:heh

[Rei (128717)]

The thing is, certain changes are more likely to be advantageous than others, so it only makes sense that certain parts of the genome would adapt more quickly. This isn't anything new. For example, bacterial plastids tend to evolve many times faster than the main bacterial genome. This helps them adapt more readily to changing food sources, threats, etc, without posing the higher risk of lethal mutations that changing arbitrary genes carries.

An extreme example of segments of DNA mutating faster than the rest comes from the mammalian immune system. Picture this: a mouse has less than 100k genes, but can make more than a million different antibodies. How? Each of the millions of B lymphocytes circulating in the bloodstream can only make one antibody, just one amino acid sequence. When the mouse is attacked by a particular disease, almost all of them will be useless against the disease. But the few that display antibodies that have any sort of ability to bind the disease do so, and this triggers those cells to undergo rapid mitosis. This produces many clones that can attack the disease; however, they're not exact clones. The gene that codes for the antibody has "C" (constant) regions and a "V" (variable) regions. Each antibody uses two identical "heavy" regions and two identical "light" regions. There are three parts of the "heavy" variable region -- VH, DH, and JH -- while there are two for the "light" variable region -- VL and JL. A molecule called AID changes a cytosine to a uracil, which isn't normally found in DNA. The body's DNA repair mechanism attempts to correct it, changing the gene in the process. There are about 50 possibilities for VH, 23 for DH, 6 for JH, 57 for VL, and 9 for JL. So, doing the math, you come out to a staggering 3 1/2 million antibody possibilities, all thanks to the extremely rapid evolution of the V regions. The more effectively the B lymphocite binds with the pathogen, the more its reproduction is activated, and the more copies of itself -- both identical and with slightly changed V regions -- it makes. This whole process is called "somatic hypermutation", and we couldn't survive without it.

Re:

[wizardforce (1005805)]

"...evolved to the point where they work well with evolution" ya think?

It doesn't mean what you think it means. What the article was talkin about in this regard is that a genome will tend to evolve in such a way that mutation rates will be at a good rate for the organism. Not too many as to cause irreversable damage to the genetic line and not too few as to cripple the organism's ability to adapt genetically to the environment.

Re:

[stranger_to_himself (1132241)]

What the article was talkin about in this regard is that a genome will tend to evolve in such a way that mutation rates will be at a good rate for the organism.

Indeed. Which is kind of related to this previous study Rate of Evolution Metrics Observed [slashdot.org] which showed that the optimum rate of evolution varied between small fast reproducing animals and larger slower reproducing ones.

What this adds is the news (or further evidence if it was already known) that the optimal rate might even vary across differ

Re:

[vegiVamp (518171)]

Meta-evolution, then - it evolves the process of evolving into it's most optimal state, or at least with a strong preference for the rate that can be sustained for the longest time by the most entities.

That's a bit over-interpreted, IMHO

[Moraelin (679338)]

That's a bit putting the carriage before the horse, IMHO.

What this really says is that the genome became, more or less, fault-tolerant. The ability to evolve really came out of that.

For starters, there is no part of the genome or ribosomes or whatever that actually produces mutations. On the contrary, most of the complexity in your cells is to prevent mutations, to the best of possibilities. It's the only way to have a coherent organism made of gazillions of cells. You don't want a cell in your palm to think it's supposed to grow into a nose, for example. And you really don't want cells to just start divided uncontrolled.

And you or the mouse have layers upon layers of defenses against that. The very reason why we're DNA based instead of RNA is to allow repairing single-strand mutations. But it goes on from there.

The very fact that you age is, pretty much, a defense against cancer: cells have a maximum division number counter, based on what tumor size still likely wouldn't kill you. (Hence also why larger species tend to live longer: they get a bigger limit there.) When more and more cells have reached that limit, then more and more damage can't be repaired, and you discover the fun of old age. And then you die.

Etc.

At any rate, the major thing is: there is no part in the genome that says you should evolve. Read: mutate. It actually tries to prevent mutations, hence evolution.

But mutations happen anyway, and some will happen in the sperm or eggs, or the first stages of embryo formation. You can't 100% prevent those. They _will_ happen. And the choices from there are basically two: either the result can still live with that mutation, or it dies.

Hence what they discovered here: natural selection favours the kind of genome that can tolerate mutations when they happen anyway. A species where the slightest change results in death will be at a disadvantage, compared to a species where more individuals survive even with mutations.

Sure, in the long term that also means being to evolve and cope with environment changes. No doubt. But I think there's a far stronger short-term pressure to achieve the same result. And most likely that's really what we're seeing there.

Re:

[dvice_null (981029)]

> Hence also why larger species tend to live longer: they get a bigger limit there.

Is that a fact? I thought that big species die older, because their cells devide less often. So instead of having a bigger counter, they just count slower.

Re:

[Lurker2288 (995635)]

"Hence what they discovered here: natural selection favours the kind of genome that can tolerate mutations when they happen anyway."

That's true more or less by definition, but I think you're overlooking something simple. A more complex organism has more opportunities for nonfatal mutations. That is, Mycoplasm genitalium, probably the simplest known bacterium, is extremely vulnerable to deleterious mutations. If it loses a gene that codes for a vital self-component, odds are it hasn't got a backup process fo

Re:

[wizardforce (1005805)]

For starters, there is no part of the genome or ribosomes or whatever that actually produces mutations. On the contrary, most of the complexity in your cells is to prevent mutations, to the best of possibilities.

I used to think the same way about it but the more I looked into it the more I realized that there is a way for species to regulate their evolution in a way. Genetic repair mechanisms to be exact. They have evolved to have some level of accuracy; this is different depending on what species you ar

That much is obvious, BUT...

[CarpetShark (865376)]

Yes, it seems fairly obvious that a creature might have adapted its evolutionary mutations to mutate things aren't quite proven yet, and to not mess with things that are "backbones" (literally or figuratively) of the design.

The question for me is... WHERE is that information stored? In another part of the DNA? If so, it becomes metaprogramming. Somewhere else? If so, there's a new type of DNA (unless RNA etc. qualify as that; I'm no geneticist, admittedly).

The Next Step

[Raindance (680694)]

not only that certain areas of the mouse genome undergo more changes, but that changes to those areas are more tolerable by the organism than changes in other areas.

I think a fascinating next step would be to see if, statistically speaking, viruses and transposons were channeled into jumping into these "safer to change" hotspots rather than other, more fragile areas of the genome.

It would seem to make some sense, given all the potential for genomic havok inherent in viruses and transposons' tendency toward hopping into the middle of genes.

Re:

[wizardforce (1005805)]

It depends on what kind of mutation you are introducing. regions that are more able to withstand point mutations which is much the case with substitutions of glutamic acid for aspartic acid and vice versa would not tolerate an insertion of a transposon or viral genetic material the same way. Point mutations are one thing, insertions of significant sections of genetic material is quite another. There is also the consideration of where exactly the genetic material is located. Genetic material located fart

Re:The Next Step

[mauthbaux (652274)]

I find it strange that organisms would allow *any* viruses etc. to tinker with its DNA.

That's kinda what viruses do. The virus by itself cannot reproduce (that's why it's normally not considered to be 'alive') - it has to hijack a cell's reproduction machinery to do the reproducing for it. In order to hijack the cell, it inserts its own viral DNA (or RNA - depends on the virus) into the cell's genomic DNA, and reprograms the cell to make more viruses.

Often, if the cell doesn't die from the infection, it passes on the viral genes as well when the cell reproduces. Our own human genome has a significant amount of viral DNA in it; most of it has been inactivated, but we still produce some viral proteins in very small amounts (reverse transcriptase for instance). I once heard the estimate that a full 15% of our genome has viral origins, but cannot find any reference to verify this claim at the moment - take it with a large grain of salt.

Now, cells do have several mechanisms that they use to defend against viral attacks. Most notably, restriction endonucleases. These are enzymes that chop up the DNA at certain sites. We use these enzymes all the time in genetics work. If you've seen images of agarose or acrylamide gels with patterns of lines on them, that's usually DNA that's been chopped into pieces by some of these endonucleases, and then separated by size. Restriction endonucleases are commonly found in bacteria, but can also be found in lower eukaryotes like yeasts.
Another method for defending against viral attacks is RNAses (enzymes that chew up RNA). This primarily works against viruses that use RNA as their genetic material. There's also the trick of marking your own genes with methyl groups so that you can tell the difference between it and foreign DNA, (if it's not marked, destroy it). Eukaryotes typically destroy any DNA found in the cytoplasm. So yeah, the cell does have several methods to defend against viral attack.

But I suppose it may spend more energy to defend the sensitive areas such that those areas that are more flexible to mutations are not as well protected; meaning they get hit more.

Once the virus genes have been inserted, removing them is quite difficult. Generally, viruses don't have a specific site that they insert to either, it's typically inserted at random. The reason that our own genes don't get significantly interrupted is that the majority of our genome doesn't code for anything; viruses insert themselves into areas we aren't using anyhow.

Re:The Next Step

[semiotec (948062)]

The reason that our own genes don't get significantly interrupted is that the majority of our genome doesn't code for anything; viruses insert themselves into areas we aren't using anyhow.

This is an outdated idea. It would be more correct to say that we don't know what the majority of our genome encode for. Currently, it is estimated that around 40-60% of our genomic DNA is transcribed into RNA. A small fraction of these are messenger RNAs, which encode for proteins. We have no idea why (or often even "how" as not all have apparent signals for transcription to begin) these sequences are expressed. One of the new-ish idea in evolution is that many novel micro/small RNAs are rapidly (on the evolutionary scale) evolved in and out of the genome.

Genome Hotspots

[mauthbaux (652274)]

It's been known for quite a while that certain sections of the genome mutate faster than others. Areas where the genes are less likely to mutate are typically referred to as 'conserved' regions, and most genome browsers will even indicate which regions they are. The UCSC genome browser is great for checking things like this (http://genome.ucsc.edu/). With that browser, you can look up genes and compare them to the coding sequences in other animals.

For very highly conserved genes such as the homeobox sequences, the degree of conservation is enormous. Nearly everything has the homeobox -or 'hox' sequence, and the sequence itself hasn't changed significantly (in comparison to most other genes). tRNA sequences as well don't change significantly; neither do ribosomal genes. Some stuff you simply can't change without experiencing lethal (or at least highly detrimental) results.

Other regions such as non-coding regions, and introns to a lesser extent, can be mutated significantly without any change to the phenotype of the organism. In fact, this is what a lot of DNA fingerprinting is based on - big variations in sequence lengths and other polymorphisms between individuals. These variations don't occur frequently enough within coding sequences to be of any use in identification. Rather, they check the non-coding areas and other mutational hotspots for differences. Conversely, changes in the protein-coding regions can be used to determine the relatedness between species (say, human and chimp differences, or rat and mouse) on a much longer scale.

Now, having said that, there are always exceptions. Some organisms have entirely novel mutation patterns. The influenza virus (admittedly, not an organism in the traditional sense) mutates almost exclusively in the coding areas of its envelope proteins. Even stranger, only 1 strain of the virus seems to survive every year to propagate the next. (See the 2001 article by Bull and Wichman entitled "Applied Evolution" in the journal 'Annual Review of Ecological Systems".)

Basically, what I'm saying is that the fact that some parts of the genome mutate faster than others is something we already know. This isn't necessarily news. The only way I can think that this would be significant is that lab mice are generally thought to be basically genetically identical. They're normally inbred for about 20 generations (most don't survive past 7) to ensure the homozygosity of the mice. Inbred mice like this are valuable because the way they react is consistent and reproducible (traits that are mainstays of science). If they're mutating faster than we expected, it may have an affect on the reliability of the studies done with these mice.

Re:

[stranger_to_himself (1132241)]

I haven't RTFA and frankly I don't plan to, but I think the significance here is that they've found regional variation at the genomic level, rather than the genetic level as you describe. That is, there are certain parts of the genome where genes themselves are more likely to be different. Having said that, I was always taught that areas near telomeres evolved more readily than other areas, but thinking about it that could have been speculation by my supervisor.

Re:

[jd (1658)]

Far as I know, you're correct, which is why only very specific markers are usable for genealogy, other specific markers are usable for ancient ancestry, and yet others are totally useless for either. Nor is this restricted to "junk" DNA - apples don't breed true, no matter what. The only way to copy an apple tree is to graft a cutting onto a root stock.

The only possibility I can think of is that the region(s) are abnormally variable for a lifeform as complex as a mouse - apples are hardly as sophisticated

Re:

[milamber3 (173273)]

I don't "allow" or "not allow" mutations in this area versus another area. It's just that mutations in other areas cause such a significant defect in the virus that it can not function and therefore no clones are created to carry on that mutation.

It's very similar to the metaphor about the thermos. How does it know to keep cold things cold and hot things hot. It doesn't, it just know, it just does.

Re:

[milamber3 (173273)]

Woops, was supposed to be "It doesn't" in the beginning of the first sentence.

Misinterpretation

[protobion (870000)]

Unfortunately, ars technica and by consequence Slashdot, have completely mis-interpreted the original paper, at least regarding the headline used. As many people have stated, there is no wonder in finding that there are genome instability hot-spots. This has been known for years. What was not obvious , is the existence of hot-spots leading to a specific kind of mutation - i.e , copy number variation (CNV). Even though CNVs are mutations in the classical sense, modern molecular biology reserves the term 'mutation' for single nucleotide or codon changes. Drastic changes at the genomic, chromosomal or transcript level are generally called by their specific names such as deletion, truncation, transposition, duplication etc. What this study seems to suggest is that certain regions of the genome (irrespective, it seems, if these regions are genes or have a known biological function) seem to have a fluctuating copy number in the genome, with the rate of fluctuation much higher than expected in a random process - suggesting the existence of a mechanism that allows for this fluctuation to occur. It implies, that evolution has caused these particular regions to become uncoupled from potential lethality or drastic abnormality that arises in organisms , when similar variations occur on other regions (for example: variation in X-chromosome number leads to Turner or Klinefelter's syndrome). The interesting question that I see, is if there is a mechanism that allows this "tolerance" to exist to variations in these particular regions, and if there is such a mechanism, can it be tailored to allow changes in other regions...leading to the possibility of creating strains of organisms specially suited for particular scientific experiments-with multiple copies of a gene etc. - animals that currently are simply impossible to create because these changes are lethal. A far shot would be therapeutics. There are certain diseases that arise simply because of a cells inability to tolerate certain changes in the genome, irrespective of whether those changes are the cause of the lethality. In other words, the cells defense system itself is the cause of the disease rather than the genetic change. This might be the case in several autoimmune diseases or developmental diseases where upon sensing a genetic change, cells undergo apoptosis - irrespective of whether the genetic change is detrimental during the natural life of the cell. So, if one reads the Nature article, there is really some news there

Re:

[SpeedyDX (1014595)]

Where's the pro-religion flamebait post?

I don't know, but it seems like the anti-religion flamebait posts are running right on schedule! :)

Re:

[bigstrat2003 (1058574)]

The irony is that while there is no pro-religion flamebait, you still had your anti-religion flamebait ready to go. Maybe we shouldn't drag religion needlessly into discussions?

Re:

[Heir Of The Mess (939658)]

It's 3pm in the afternoon where I am Taipei, but there's probably almost 0 creationists around here. I guess creationists all live on the other side of the flat earth.

Re:

[jd (1658)]

Gnome hotspots are unstable? I could have sworn that was one of the components that worked.