transgenic animals, phenotype and redundancy

S

spuriousmonkey

Banned
Banned
One of the most popular tool of the modern biologist is the transgenic animal. A gene is knocked out, knocked down, overexpressed, ectopically expressed, etc. You name it, scientists have done it.

What often happens with knockouts (where the gene in question has been made dysfunctional) is that there is no phenotype. The animal looks perfectly normal. Then the word redundancy is then brought forward. But lately I have been having discussion with other people, that seriously make me think that redundancy is a sign of a completely artificial look at nature. In fact, there often is no such thing as redundancy (there are always exceptions of course).

I'm not goint to eleborate yet on my ideas, but will start with an observation that someone else brought to my attention yesterday during a coversation in the hallways of our mighty institute.
She was very skeptical about the analysis of transgenic phenotypes in general, because laboratory animals are kept under very artificial conditions.

For instance, we just don't keep mice in a cage and that is it. They are carefully selected strains, inbred, and kept under strict conditions to keep them disease free. There is usually no strain on their lives. This alone could be a reason why we do not see a phenotype in some transgenice animals. The same animals might have severe problems in more natural circumstances. Also when you knock out a gene in one strain of mice you don't see any phenotype, but in another strain there suddently is.
 
If one is attempting to show the result on Mice in a natural habitat, and the mice are not in their natural habitat, it's an assertion that breaks the rules of Science.

I do not know who this "we" is that keeps the mice in an artificial habitat, and thus breaks this rule. If it takes place, then they surely are the scum of the science world.

But I'm not making the assumption that this is done, it was just something you said.
 
I'm more trying to raise the idea that the notion of redundancy is a highly artificial one. And that nature might work differently.
 
In the opinion of this layman you're right. And it's worse than that. We are only capable of detecting obvious changes in phenotype. There may be endless changes that we have not noticed (perhaps the transgenic mouse is short-sighted, but not to the extent that we notice. In the wild it might not survive, but in the lab we just see a normal mouse and assume a redundant gene has been altered). 'Redundancy' seems mostly a matter of perspective, and really just means 'redundant in terms of observable changes to the particular measurements that we normally make of laboratory mice'.

The genotype and its ontogeny is so complex and cross-related across functions and parts that that it must surely be impossible to be certain what is genetically redundant and what is not.

Also 'redundancy' suggests uselessness. In fact paedomorphosis and gerontomorphosis show that seemingly redundant genetic material may be vital to our future evolution even if it is not expressed in the phenotype currently, and that we have no test for long-term redundancy.
 
Originally posted by spuriousmonkey
I'm more trying to raise the idea that the notion of redundancy is a highly artificial one. And that nature might work differently.
Click to expand...

I do not contest the notion that the natural function of proteins/genes may be different than what is observable. Laboratory science is, by its very nature, artificial. If there were a better way to test these things, I would be all for it.
But as for your assertion that redundancy is artificial, I have to respectfully disagree.
Genetic duplications are natural and common. We can follow the evolution of genes and through sequences identify places were genes were duplicated. Eventually there may have been events that caused this gene to change, and the resulting protein took on a new role. However, there are a lot of genes out there (look at pseudotetraploids like Xenopus, and even more complex genomes like Maize) and you will see copies of genes whose proteins are virtually identical. I am not asserting that these proteins do not have specific functions, what I am saying is evolutionarily it is in an organisms best interest to have a back-up plan (which is why a diploid genome is so sucessful). If your only copy of a gene is damaged during replication, it is very useful to have another gene that encodes a closely related protein that can compensate for the other's absence. So in a lab when you knock out a gene and see no phenotype because of functional redundancy, this is artificial because you have removed one protein and therefore forced another to compensate. But this happens in nature, genes are damaged, deleted, viruses and transposons interupt the gene sequence. Therefore functional redundancy is a mechanism by which organisms can protect themselves from genetic damage in addition to repair enzymes and other methods. There are also alternate pathways that can accomplish the same thing. Redundancy as we observe in transgenics may be artificial, but I believe it is a real and useful method by which organisms cover all their bases.
 
that is the standard answer and I heard it many times, and i said it myself many times, but the fact is that I am getting truly uncomfortable with this answer. Science for me is more than doing experiments and giving objective interpretations. I want to know the true nature of life. And redundancy just doesn't feel right.

Gene duplication might start of as mere gene duplications, but Someone actually referred me to an article in nature quite some time ago (which i didn't look up yet) where the logical argument was put forward that truly redundant genes would disappear during evolution. So from that perspective I wouldn't really get excited about redundancy. Therefore they must have a function. I have an idea about it, but i might use it myself, so I am not going into it.
 
Wrmgrl - what you post clearly says is that even duplicated genes are not redundant (in that they are useful as back-up). There is no evidence that any genetic material is redundant - only that it hasn't been expressed in the phenotype this time around.

I don't have Spookz's view that nothing is redundant - but would argue that we cannot know what is redundant and what is not. We are in the position of the CEO of Ford (I forget which) who is quoted as saying "We know that half our advertising expenditure is wasted, but we don't know which half".

Did anyone ever manage to explain the ability of drosphilia (over a few generations) to regrow eyes again even when the necessary genes have been completely removed from the stock, and then only inbreeding allowed? Somehow mutant genes pair to create eyes again. This points to the dangers of arrogantly thinking we know what is genetically redundant and what is not.
 
reduancy

Originally posted by Canute

Did anyone ever manage to explain the ability of drosphilia (over a few generations) to regrow eyes again even when the necessary genes have been completely removed from the stock, and then only inbreeding allowed? Somehow mutant genes pair to create eyes again. This points to the dangers of arrogantly thinking we know what is genetically redundant and what is not.
Click to expand...

To the first part of your post I can only reply that redundancy is a major problem in my own research. And I do not believe your argument really counters what I have said. You are right that it would not be energetically favorable to maintain large amounts of redundant DNA. That is not what I suggest, it is more that when push comes to shove, some things can act in ways they wouldn't normally. Or certain pathways can be by-passed. Perhaps the whole thought on redundancy boils down to the problem that not all genes are essential, at least in terms of things we can analyze. Therefore we often assert "well it must be redundancy" instead of just admitting that maybe our favorite gene is not all that cool.

As for the Drosophila phenomenon, I have never heard/read of any such thing, so I can neither refute nor accept it. From and evolutionary development background I find it very hard to believe, and I would have to read the work myself to make a judgement. In theory, if there is functional redundancy in the genome, rearrangments, duplications, or modifications to already existing genes could cause the sort of phenomenon you describe. I think that argument more favors redundacy then refutes it.
 
Originally posted by spuriousmonkey
And redundancy just doesn't feel right.
Click to expand...
Hmm, there's a scientifically sound reason. However I must admit that I think all scientists have these sorts of reactions to things. Sometimes an answer just doesn't sit right with you and you can't explain why that is. I suppose there are more hunches that lead to great discoveries in science then we would admit.

Originally posted by spuriousmonkey
the logical argument was put forward that truly redundant genes would disappear during evolution. So from that perspective I wouldn't really get excited about redundancy. Therefore they must have a function. I have an idea about it, but i might use it myself, so I am not going into it.
Click to expand...

I do not disagree, it is not advantageous to maintain large amounts of DNA or produce proteins that do not have specific functions. It does not confer a selective advantage unless there is genetic damage, hence the diploid genome. I never said they don't all have a function, I just assert that those functions may overlap.
I will be the first to say that redundancy is not exciting. From a geneticist point of view it is more problematic, in terms of reserach. That doesn't mean its not useful for organisms.
I hope you study genomics or genetics and can solve this problem difinitively one way or another, until then I will continue to believe that to at least some degree redundancy is alive and well in genomes and proteomes.
 
wrmgrl - the fruit fly stuff comes from G. de Beer 'Embryos and Ancestors' and from A.C. Hardy 'The Living Stream' as quoted in Koestler's 'The Ghost in the Machine'. (Which also contains more examples of curious evolutionary anomalies). I think it's interesting enough to quote a piece.

"The fruit fly has mutant gene which is recessive. .... if two of these mutant genes are paired in the fertilised egg the offspring will be an eyeless fly. If now a pure stock of eyeless flies is made to interbreed, then the whole stock will have only the 'eyeless' mutant gene, because no normal gene can enter the stock to bring light into their darkness. Nevertheless within a few generations flies appear in the inbred eyeless stock with eyes that are perfectly normal."

I'm sure that you're right in saying that genomes contain redundant bits of information. My assertion/belief is just that we will never know with certainty which bits these are. To continue the management analogy, it is never clear in a complex business organisation which bits are completely redundant - both the internal and external dynamics of the thing are too complex to outguess. (A fact many companies came to realise in the 80's with all that simple-minded 're-engineering' that went on, and which so often came to grief through misjudgements over which operations could safely be got rid of).
 
I want to post a general reply to all. As for using inbred strains is highly important because they are genetically the same. We don't know the percise genotype, but we can be highly confindent that all the mice in one strain are genetically the same, which turn out to be really important in making genetic crosses (classical genetics). Because one knockout is not noticed in one strain, but is notice in another points to epistasis (gene/gene interactions). some genes phenoype will not be notices (or the genes expressed) when a different, unrelated gene is homozygous, or etc.... This can be alot more complicated than redundency. there is epistasise, epigenetics, penatrence, and all sorts of geneticst that needs to be considered.

As for redundency, there is some, and in plants, there is a lot. The Maize zein gene alone has 25 tandomely repeated copies. The rDNA genes have multiple copies. Interestingly, no matter how many copies, the number of rRNA per cell does not change. Add some more, delete some genes, the number doesn't change. So you can see how redundency can be selected for very import genes that are dosage sensitative. Through mutations, you lose a few genes, the others pick up the slack.

What about non important genes. There can still be redundency. This is clealy shown in the organizim that I do research on -> Maize. There are 2 genes, chalcone synthase, and White pollen. Both genes do the same thing at the same time. Why did one not diverge from the other. For some reason, evolution thought (not literally) that this was better.

not everything can be answer with redundency, but is a valid option along with everything else.

as for the eyeless flies, that could be a geat example of natural selection. You have a stock of eyeless flys. through sevel rounds of random mating, mutation gradualy accumulate. A mutation (lets say in that recessive gene maybe) that restores the eye. If it only takes one gene to get ride of the eye, you can see how mutating one gene could resore the eye (a hox gene maybe). This is favorable and is selected for, and eyelss flies are selected against. After awhile, eyeless flies are all but gone. Just an idea
 
You obviously know a lot more more than I do about this. But repitition is not necessarily redundancy, as in the word banana, or as when shouting 'help, help...' - so genetic redundancy can only be judged once we completely understand the genome, ontogeny etc - which certainly ain't for a while yet. .

Also I can't agree that the fruit-fly re-evolves its eyes. As you say it may only require one mutation. But that is one mutation amongst an infinity of possibilities, one mutation that exactly does the trick, no more or less. There's more chance of winning the lottery twice.
 
When I think of the word redundency (in the context of genetics), I think of a gene that does the same job as another. It is redundent. Repetition and redundency can be hand in hand. Maybe the problem is with my definition. I just want to say the redundency my not be an answer for everything, but it should always be an option along with everything else.

I agree that redundency will not be fully judged untill the genome is completely understood, which will take alot longer than most people realize. right now though, I think (my opinion) its for of a fact than fiction.

as for the fly thing, It was just a possibility, not an answer. If you really think about it, It could happen. It may be one mutation among an infinite number (well not really infinte, there is only 4 chromosomes counting the sex chr) of possibilities, but there is also a very, very large number of mutations. And depending on how many generations it took to revive the eye, the more generation, the increased possible number of mutations.

With the use of molecular biological techniques, this could even be investigated to see if that was the case. I don't know the contents of fly experiment that started this conversation, for instence, how many generations it took for the eye to come back, or even if this experiment was repeatable, the error rate of the fly DNA polymerase, was mutagins used, etc... I was jut giving a left field answer on a topic which I knew little info about. Maybe I shouldn't have said anything without knowing all the facts :)

-later
 
Originally posted by Chiasma
When I think of the word redundency (in the context of genetics), I think of a gene that does the same job as another. It is redundent. Repetition and redundency can be hand in hand.
Click to expand...
True. I was using the layman's definition, as in 'not useful'.

I was jut giving a left field answer on a topic which I knew little info about. Maybe I shouldn't have said anything without knowing all the facts[/B]
Click to expand...
If everyone did that there'd be no discussion here. (PS I'd like to know a bit more about this fruit fly experiment as well. I've posted the references earlier but aren't likely to follow them up myself).
 
da bomb

Oh Chiasma, you know you're da bomb.
You really know your schiznit with it comes to maize genetics (and most other aspects of molecular biology).
Stop apologizing for your opinion, you're entitled to it, you express it well, and you're usually right.
 
Re: da bomb

Originally posted by wrmgrl
Oh Chiasma, you know you're da bomb.
You really know your schiznit with it comes to maize genetics (and most other aspects of molecular biology).
Stop apologizing for your opinion, you're entitled to it, you express it well, and you're usually right.
Click to expand...


Damn, now I'm blushing ;)

thanks wrmgrl!!!
 
let's try to get a bit more intellectual again.

I was recently in a meeting in the US and saw an intersting presentation by Gary Odell (I apologize for all mistakes that I make her about his work. I have to work by memory). He was making computermodels/simulations of genetic networks, with all possible interactions between the different components. This model is based on real interactions during development when 2 rows of cells are formed with a clear boundary between them in drosophila.
A major problem with such a model is that basically you do not really know how strong the value of a specific interaction between for instance a signalling protein and a DNA binding region is. Especially compared to other interactions, be it protein-protein, protein-RNA, or protein-DNA interactions.
Odell therefore just used raw computerpower. He had this network of let's say 50 interactions that should give rise to a specific pattern of cellular arrangement. He then just tries to find values that will make the proper pattern.

And now we return to the theme of redundancy.
The computer has no problem finding values that will give rise to the proper pattern. That is, it used to have, when there were less interactions in the model. But then he added about 15 interactions more, that are based on published data, and suddenly the computers found many, many solutions. Apparently it is easier for a very complex set of interactions to make a proper pattern than for a simpler set of interactions.
These interactions would not really be redundant, but function to make the set of interactions more complex.

Another interesting feature was that a value for a single interaction was never limited to a narrow value range. No, the value could change several factors before the pattern was disrupted. This adds robustness to the system.

It is a bit difficult to describe, but if you would plot these values in a graph you would be able to see that they stay within the viable range quite long, and then suddenly they go straight up and shoor out of the viable range. But the next viable range of values that make up a slightly different pattern is always really close to the previous range. So if an individual wants a slight pattern change he only has to change the value of a single parameter/ genetic interaction slightly to jump from the viable range of the original parameter values to the viable range of the new parameter values.
From an evolutionary viewpoint this means that this complex, robust system also facilitates evolutionary changes in patterns and not necessarily has to stabilize changes.

in conclusion. Redundancy might not be a real phenomenom. We might have many genes with slightly different function and overlapping functions to increase complexity of the regulatory interactions, which makes the reproducibility of patterns more robust and at the same time makes it easier to hop from one viable pattern solution to another.
 
(Sounds of applause).

This seems to suggest that our genotype is a complex system in the same sense as is our weather or our ecosystem, and that it would share their attributes, in particular an extreme sensitivity to initial conditions. Who's to say which butterfly is redundant in terms of the future weather?
 
You must log in or register to reply here.
Back
Top