integration of a gene per homologous recombination - (Nov/25/2010 )
Hi!
I am a student who has recently read that you can integrate a gene from a plasmid into a genome e.g. in a yeast genome.
Now my thoughts how this could probably work:
There are 2 sequences on both sides of the gene which you want to integrate.
On the yeast genome are 2 sequences which are homologous to these on the plasmid.
In this case the gene will be integrated per homologous recombination between these sequences.
Are my considerations right so far?
Now my major understanding problem:
How can I be sure that there won't be a second recombination event between the same loci so
my integration would be reversed?
How can I make sure that this will not happen?
I hope you'll be able to understand my question and help me. (Because I am from Germany my English might not be the best, I guess).
br,
Sebastian
Ikar on Thu Nov 25 17:03:45 2010 said:
Hi!
I am a student who has recently read that you can integrate a gene from a plasmid into a genome e.g. in a yeast genome.
Now my thoughts how this could probably work:
There are 2 sequences on both sides of the gene which you want to integrate.
On the yeast genome are 2 sequences which are homologous to these on the plasmid.
In this case the gene will be integrated per homologous recombination between these sequences.
Are my considerations right so far?
Now my major understanding problem:
How can I be sure that there won't be a second recombination event between the same loci so
my integration would be reversed?
How can I make sure that this will not happen?
I hope you'll be able to understand my question and help me. (Because I am from Germany my English might not be the best, I guess).
br,
Sebastian
The low probability of the integration reaction itself means that a second double recombination event to reverse the initial recombination event is very low. This is not a factor that you need to care about. What you do need to care about is how do you screen for the recombination event.
Not all your yeast cells will take in your DNA. And even flanked by homologous DNA sequences, it is more likely that your targeted gene will be integrated randomly into the yeast genome. Successful targeting can vary between 1/10 to 1/200 for yeast. Furthermore if I understand correctly, your yeast species is diploid. It is likely only one of the two chromosome will be targeted, creating a heterozygote.
What you need is a selection marker linked to your gene. You can then select for cell which have integrated the gene. Then you need sporulate the yeast to get haploids, which depending on the yeast your working on can quickly turn into homozygote diplods.... which are then selected for the selection marker again.
Then you need a screen (PCR, lost of gene function, southern blot) to identify actually homologous recombinant from random integration.
Whether the yeast is dipoid or haploid doesn't matter I think.
In this case I would prefer haploid yeast, because then it'll be easier to
identify successful integrations.
But why does the target sequence integrate at random?
Are transposons involved?
Imagine I have a mutant gene on my plasmid which I want to exchange with the
wildtype on the genome.
If the mutant allel would integrate somewhere into the genome, there won't be
a "normal" regulation of the gene's expression.
In this case the wildtyp gene would be expressed and maybe the mutant allel, too.
But if the mutant gene would be expressed it could also be expressed too much or
too less.
Even with a few base pair of homology, low levels of recombination can occur. This results in the background that you see. Transposons are not involved.
Can you select for the mutant allele? What I mean is can you select for cells which have taken up the DNA construct against those which never took up any DNA at all?
Is the homozygote of you mutant gene predicted to be lethal? If it is lethal then using a diploid is a must.
And depending on the yeast species you are using, the haploid state maybe the default state of the species or it could only occur very transiently in yeast life cycle.
perneseblue on Sat Nov 27 04:24:19 2010 said:
Even with a few base pair of homology, low levels of recombination can occur. This results in the background that you see. Transposons are not involved.
Can you select for the mutant allele? What I mean is can you select for cells which have taken up the DNA construct against those which never took up any DNA at all?
Is the homozygote of you mutant gene predicted to be lethal? If it is lethal then using a diploid is a must.
And depending on the yeast species you are using, the haploid state maybe the default state of the species or it could only occur very transiently in yeast life cycle.
In this case I would use a auxotrophy marker on my plasmid so I can see which cells have taken up the DNA.
In this case of a haploid individuum the wildtype allel on the plasmid should suppress the mutant phenotype.
But how can I maintain that the DNA will recombinate between the mutant allel and the wildtype on the plasmid? Because I only want to
have the wildtype beeing expressed. If the wildtype is integrated somewhere else into the genome it could happen that both will be expressed.
The mutant allel is only lethal under restrictive temperature.
you will have to screen colonies to identify ones where the targeted homology recombination has occurred. There are several screening methods available, PCR and southern blot