A role for "junk DNA"

[Paul C. Anagnostopoulos]

In a recent Scientific American, there is an article about a possible role for junk DNA. Turns out that the amount of junk DNA in a genome relates fairly well to the "complexity" of the organism. This and various other observations have led people to suggest that introns (supposed junk in the middle of DNA sequences for genes) may be transcribed into RNA that then serve some kind of regulatory role. Since the RNA can bind with DNA and other RNA, these bits of RNA are perfect for regulating the transcription and translation of DNA/RNA.

Prokaryotes do not have much junk DNA. It is hypothesized that this is the case because there is not enough time between transcription and translation to accomodate the splicing out of introns.

On hindsight, this is so obvious. Here is a bunch of RNA that is spliced out of a longer RNA sequence to make messenger RNA and ultimately some protein. Would nature just let this stuff float around and simply be recycled? Of course not! It would co-opt the stuff for some function, just by pure accident of evolution. How can you not love this stuff?

~~ Paul

 

[Dr. Imago]

Yeah, I saw something about this too, Paul. As the old adage goes which is especially important in science, "live and learn."

It will be interesting to see what future ramifications this has concerning our better elucidating the inner workings of molecular genetics.

-TT

 

[Soapy Sam]

I've rather wondered if all the padding simply makes the "real" genes harder for a cosmic ray to hit. If you could "defragment" the genome, so it contained only coding data, then any damage would be serious.

 

[Paul C. Anagnostopoulos]

Also, any actual junk DNA is useful fodder for mutation.

~~ Paul

 

[Zombified]

I've rather wondered if all the padding simply makes the "real" genes harder for a cosmic ray to hit. If you could "defragment" the genome, so it contained only coding data, then any damage would be serious.

Hmm... this would imply that either the seperation is to reduce the number of genes damaged by a single ionizing particle, or that there's redundancy in what's currently considered non-coding DNA.

The former implies that the non-coding DNA should be more or less evenly distributed and that there are no large uninterrupted coding regions. Is this so?

Has the idea that non-coding DNA represent redundant copies of other genes been ruled out? I thought it had...

 

[Dymanic]

Originally posted by Paul C. Anagnostopoulos

This and various other observations have led people to suggest that introns (supposed junk in the middle of DNA sequences for genes) may be transcribed into RNA that then serve some kind of regulatory role.

This suggestion is nothing new. One of the best known proponents of this idea is none other than the infamous Michael Behe himself.

It isn't just that we haven't found functional reasons for these non-coding segments being present; we have found very good reasons to believe that they (at least some of them) cannot be functional.

The response (from talkorigins faqs):

Sections of DNA can be cut out or replaced with randomized sequences with no apparent effect on the organism.

Some sections of DNA are corrupted copies of functional coding DNA, but mutations in them, such as stop codons early in the sequence, show that they cannot have retained the same function as the coding copy.

The fugu fish has a genome that is about one third as large as its close relatives.

Mutations in functional regions of DNA show evidence of selection -- non-silent changes occur less often that one would expect by chance. In other sections of DNA, there is no evidence that any changes are selected against.

 

[Tez]

Sections of DNA can be cut out or replaced with randomized sequences with no apparent effect on the organism.

Some sections of DNA are corrupted copies of functional coding DNA, but mutations in them, such as stop codons early in the sequence, show that they cannot have retained the same function as the coding copy.

Interesting.

One of my whackier ideas, that I've yet to really try and quantify, is that junk DNA acts as a store of "random strings", which are then used to actively create mutations in the non-junk sections (perhaps triggered by stress). In this way mutations could be somewhat controlled so as to not differ too much from the original (as well as there being obvious(?) advantages in having the ability to speed up or slow down your own evolution..)

I guess I just like the idea that organisms may have evolved in a way that allows them to somewhat control their own evolution...

 

[Anders]

Hmm... this would imply that either the seperation is to reduce the number of genes damaged by a single ionizing particle, or that there's redundancy in what's currently considered non-coding DNA.

The former implies that the non-coding DNA should be more or less evenly distributed and that there are no large uninterrupted coding regions. Is this so?

Has the idea that non-coding DNA represent redundant copies of other genes been ruled out? I thought it had...

There are really no longer uninterrupted exon sequences. All exons varies in size from tenths of base pairs up to a few hundred not much longer than that.

No no, there are redunant copies of genes, or at least sequences that have at some point been genes and some that could develop to genes. They have usaually lost their gene stutus due to the unwelcome visit by a transposon or by mutation, multiple SNPs or the like.

We have long know that the "junk" DNA is full of regulatory sequences, we just don't know how they work. We know fairly well how they work in procaryotes but in eucaryotes, we have just not gotten that far yet.

 

[Zombified]

Thanks. I take it "exon" refers to what I've ineptly called "coding" DNA, that's actually used to make functional proteins? I'm not that familiar with the terminology.

(In fact, the more I read this forum, the more I realize I know very little about biology.)

 

[Anders]

Thanks. I take it "exon" refers to what I've ineptly called "coding" DNA, that's actually used to make functional proteins? I'm not that familiar with the terminology.

(In fact, the more I read this forum, the more I realize I know very little about biology.)

Thats correct, exons are coding sequences. I just wanted to show off a bit...

 

[Soapy Sam]

Zombified- Chromosomes, by molecular standards, are big .

I just wonder if, whatever the source of the non coding material (whether viral or an internal repeat "error" for example), it all acts as shielding against some radiation (low energy stuff , clearly- but every little helps) and other damaging elements such as viral / bacterial or chemical attackers. Like you, I'm too ignorant of the subject to make such assertions, but it seems intuitively sensible to me that , so long as the extra DNA has no deleterious effect, a big molecule is a harder thing to damage than a small one.

 

[athon]

Also, any actual junk DNA is useful fodder for mutation.

~~ Paul

Maybe, but don't forget there are regions of highly conserved introns.

The regulation theory has been floated for a while, and is now gaining popularity as the evidence piles up. We're now seeing more and more instances of physical (non-coding) roles played by nucleic acid, to the point that there are thoughts of putting it to use in nano-engineering. The protein-like twists and turns of regions of nucliec acid polymer can do anything from prevent regions from being read to possibly (as you said) preventing 'important' regions from being damaged.

I was always intrigued by the very precise damage done by peroxide to particular chromosomal regions during Down Syndrome development. Again, we are learning more and more about the physical positioning of nucleic acid polymers within cells.

I love living in the future!

Athon

 

[Dymanic]

Selfish gene theory posits that evolution is the result of competition between genes, expressed as differential rates of success between organisms. Producing beneficial changes in its host organism is an obvious way that a gene may come to be better represented in a population, but the competition between genes may be viewed from another perspective, that being competition for space on the DNA strand itself. Obviously, when such a competition is won by a gene having a detrimental effect on the organism both winner and loser alike suffer, but to the extent that the effects are neutral, a lot of what is seen may simply be the result of such jockying for position on the strand.

 

[Zombified]

Maybe, but don't forget there are regions of highly conserved introns.

Introns? Since I've already outed my lack of biology knowledge, I may as well ask all the dumb questions...

 

[Dr Adequate]

You are BLINDED by your MATERIALIST RELIGION. Junk DNA was put there by SATAN to prove evolution by BLIND LOGIC. You are all DARNED to the UNCOMFORTABLE WARMTH of HECK.

I won this argument.

 

[Anders]

Introns? Since I've already outed my lack of biology knowledge, I may as well ask all the dumb questions...

Introns = Non-coding sequences. Not that complicated, just naming and definitions.

 

[Dymanic]

Originally posted by Dr Adequate


I won this argument.

Sorry, but so far, there has been no sign of the troll you're looking for here. Maybe you should try dangling your hook elsewhere.

 

[athon]

Selfish gene theory posits that evolution is the result of competition between genes, expressed as differential rates of success between organisms. Producing beneficial changes in its host organism is an obvious way that a gene may come to be better represented in a population, but the competition between genes may be viewed from another perspective, that being competition for space on the DNA strand itself. Obviously, when such a competition is won by a gene having a detrimental effect on the organism both winner and loser alike suffer, but to the extent that the effects are neutral, a lot of what is seen may simply be the result of such jockying for position on the strand.

Selfish gene theory is one of my favourite theories. I use it all the time in biology; however, it's one that unfortunately is hard to support with evidence. To falsify it, you'd have to look for cases of genetic altruism (unless you assume that 'related' genes are like family members that can self-sacrifice on behalf of a related gene). I'm not sure if any exist, or if there is evidence of this. It would be fun to work on as a thesis, though.

To make matters worse, the more recent concepts of 'modules' (related clusters of genes that work together to create a complete trait) muddy the waters. Can you have 'selfish' gene clusters? The 'one gene one trait' idea is long dead.

Bah! I love Dawkin's work, and the Selfish Gene is brilliant as far as the abstract idea of 'life' goes....but it sure does my head in after a while.

Athon

 

[Dymanic]

Originally posted by athon

I love Dawkin's work, and the Selfish Gene is brilliant as far as the abstract idea of 'life' goes....but it sure does my head in after a while

I understand. I think what makes it tricky is that you have to keep in mind what level you're looking at, because some concepts lose their explanatory value when applied at the wrong level.

At the level of the organism, we do see altruistic behavior, the social insects being among the clearest examples. Worker bees commit what might be considered the ultimate act of altruism by relinquishing their own reproductive opportunities in order to devote their lives to the care of their sisters. This is difficult to explain from the perspective of individual success for the individual worker. Selfish gene theory undertakes its explanation by noting that the peculiarities of reproduction in haplodiploid species make each worker more closely related to her sisters than she would be to her own offspring, and plots the different strategies in a payoff matrix to see which strategy results in the most copies of an individual's genes.

I'm not sure what an analogue to this at the level of the gene itself would be. The water is indeed quite muddy when one begins to consider the interactions between genes such as the 'gene modules' you describe, and how those interactions propagate up to the level of the organism, and then trickle back down again.

A metaphor Dawkins used is to consider how the various attributes posessed by an individual athlete in a team sport might contribute to the team's success; a baseball team with nine pitchers, for example, would not be very competitive no matter how skilled they were at pitching. But as individuals, their membership in a 'cluster' would be a temporary status; the winning teams almost certainly would have a better mix of player attributes, which means that either an excess or a shortage of attributes of a particular type would tend to be more balanced in the next generation.

But that gets muddy again when it is considered that at any one time a gene pool may include numerous types of team-level strategy; one football team has a strong passing game, another a better running game, yet another is known for its defense, etc. If your head hurts, it might mean you're getting it.

 

[pupdog]

There's an article in Science a few weeks back (22 October 2004, p. 632-635; E. Pennisi) dealing with the "other code"--that part of the genome that controls the actions of genes (enhancers and silencers). The point of the article was that it's not just a difference in genes that results in physical differences, but differences in how the same genes are activated. Obviously, the timing during development would have significant effect.