Deletion Mutant - Evolution - (Aug/20/2012 )

Can we predict evolution if we have deletion mutant strain of bacteria?
Thanks.

prabhubct on Mon Aug 20 06:31:54 2012 said:

WHat do you mean by evolution?

Can you predict what would happen with the bacterium if you delete a certain gene? Yes and no: if you delete a certain essential gene.. yeah, you can predict that the bacterium will not survive...
Same with deleting some specific gene with a known function, you can know what would happen.
But this is not evolution really...

Thanks for reply. What about Directed Evolution then? Isn't Deleting something mean to evolutionary tragectory change? How one is supposed to approach that question?

pito on Mon Aug 20 18:06:44 2012 said:

Directed evolution is not dealing with deletion mutants. Directed evolution uses techniques such as epPCR and DNA shuffling to introduce point mutations. To delete an entire gene, it is a bit difficult. Nature doesn't delete entire genes but either inactivates them or introduces stop codons in the middle of the protein or smth happens to the promoter.

Andreea

Thanks for reply. will same what you said applies for knock-in? Is Knock in evolution problem be considered in same line as knock-out?

ascacioc on Wed Aug 22 14:37:55 2012 said:

prabhubct on Wed Aug 22 15:49:42 2012 said:

Yes, you can either have a knock in or knock out.

Think about resistance genes in bacteria: often they become resistant because they got a gene due to uptate of a plasmid.

pito on Wed Aug 22 17:15:06 2012 said:

Thanks. How one is supposed to study Knock in Evolution? Will it be facing same problems as ascacioc mentioned above?

prabhubct on Wed Aug 22 17:47:00 2012 said:

You cant really study it , all you can do is try to copy it in a lab.
So you would indeed face problems.

You can use genomeshuffeling etc to generate mutations.. but are those some sort of evolution? I doubt it, its just something you create in a lab, something that might happen in real life.
You can call it evolution but it all depends on what you define as evolution.

Its pretty easy: in nature bacteria (I am just speaking of bacteria) can either take up DNA (plasmids for example) or lose DNA (due to plasmids causing homogues recombination for example). It can also have changes in DNA due to point mutations or mutations caused by a defect in copying the DNA ...
All kinds of possibilities. ANd sometimes a mutation just works fine and the bacteria becomes better....

You can recreate this by causing mutations (deliberately) and checking what happens..

@Pito: genome shuffling happens in nature: V(D)J recombination is DNA shuffling; this is the way your body makes antibodies by combining the alleles in the MHC class I loci; moreover in the next step, your body evolves the produced antibodies by epPCR-like process; both methods used by directed evolution happening in your body 4 days after infection with a virus.

The point of epPCR is that it actually recreates the same most probable mutations/errors done by DNA polymerases in bacteria. There are tons of papers calculating the probabilities of certain transitions/transversions and the hot spots for mutations and they are pretty much the same no matter what polymerase you are using. So what you get in the test tube is what nature produces in millions of years. This is why directed evolution is defined as fast-forwarding natural evolution in a test tube. (some people use this cliche over and over again in the directed evolution field)

@prabhubct: if you would like to read more about directed evolution: http://www.sesam-biotech.com/science-and-technology/directed-evolution A pretty good review of the state of the art 2-3 years ago (I wrote it:P) It also has references to material that paraallels directed evolution to natural evolution.

Andreea

ascacioc on Wed Aug 22 19:06:14 2012 said:

Of course genomeshuffeling happens in nature, but there is a difference between what you call genome shuffeling that happens in nature (there is a system behind it, I am talking about the VDJ recombination now) and genome shuffeling we do in the lab.

And I do not agree with what you state here, its a bit "risky" to state it like this: "your body evolves the produced antibodies by epPCR-like process; both methods used by directed evolution happening in your body 4 days after infection with a virus"
This is not entirely correct.
The body does not "evolve" like you state it.
The body has allready a bunch of antibodies present, by pure luck a few of those happen to bind the antigen, because they do, they will be favored and other that do not bind will not (or less) by enriched (recreated) by the body. This is called affinity maturation of antibodies. This happens because the B cells with the best binding receptors will bind the anitgen and will be selected (and survive/multiply and pass on their genes) because they are able to bind the follicular dendritic cells (those will bind the antigen, so the B cells bind indireclty) this is how affinity maturation happens and how "evolution" of antigens work. ANd yes, due to simple mutations in those cells, you will also create better cells in the end. But its bit different from just stating what you stated.


But at the start: its all a random proces, your body just "creates" random antibodies by VDJ recombination (random; but with a certain system).

This is completely different from the genome shuffling you are speaking of. In genome shuffling you cut DNA and create random new pieces of DNA by extending them again, ligating them.
+ genome shuffeling and then compare it with VDJ.. a bit weird esp since I was talking about bacteria, but perhaps I did not state this clear enough.
In bacteria there is no such thing as VDJ recombination.
And I dont like to link systems we use in bacteria/yeast for human or animal systems/genes.

Also: epPCR can indeed be used as a tool to study what happens with bacteria/yeast for example and you can indeed call it fast evolution, but it doesnt really represent 100% what happens in nature.
Its just a tool to cause mutation, nothing more.
ANd yes, you could state that those mutation (or some) would indeed also happen in nature , but nature is far more complex.

Also: linking genome shuffeling and epPCR is a bridge to far for me.
epPCR is much more controlled while genome shuffeling is (or can be) less controlled and you can get stranger results.
Altough, in the end its all about how you define certain stuff.

Also, and you said it yourself: its called "directed" evolution, thats just it: we (the researchers) direct evolution.. you cant simple say: aha, this is what would happen in nature.
Its a bit easy to state that.

What you create in the lab, I wouldnt call it evolution. I would call it: an observation of changes in DNA that cause a certain (observed/measured) effect, which in the end could indeed be a representation of a certain (possible) evolution.
You need to keep in mind that many of the so called "evolutions" caused by direct evolution techniques would not survive in nature or stay evolved like this because we want this evolution and keep in, while in nature the evolution might we "stupid" and not wanted and thus be lost in the end.

But this is more about semantics.

But do not agree with "directed evolution is defined as fast-forwarding natural evolution in a test tube" because for me this is not correct. A lot of the so called "direct evolution" are no evolutions that would happen in nature.