Simple question that is hard to answer

Peter Ellis via (by pjie2 from
Sat Feb 17 03:34:54 EST 2007

peter wrote:
> On Feb 16, 6:45 pm, "Peter Ellis" <p... from> wrote:
>> peter wrote:
>>> Peter,
>>> Mismatch will be corrected indeed, but using what template - it
>>> seems that it will be corrected again 50/50 and still end up as
>>> mixed population?
>> Doesn't matter what template.  When you do the transformation,
>> you'll only have *ONE* copy of the plasmid in each bacterium.  Then,
>> when it copies that for the first time, it'll correct using one
>> strand or the other, and you'll have two identical copies of one
>> allele or the other.
>>> If that is the case and I isolate DNA and re-transform, I
>>> should be able to get 50/50 of homozygous plasmids in the cells and
>>> have colonies that have one or the other allele.
>> Um, yes.  If you have a mixed population and you transform it, some
>> of the colonies will bear one allele and some will bear the other.
>> What you will *not* find are single colonies that carry both
>> alleles, or single bacteria that carry both alleles, or bacteria
>> carrying uncorrected heteroduplexes.
>> I'm not quite certain why you expect to have heteroduplex DNA in your
>> transformation mixture in any case: how are you generating your
>> plasmid for the transformation?
>> Peter
> Making heteroduplex on purpose...

Why?  What's the goal of your experiments?  Why is it important to you to 
have both alleles present in one bacterium, and why can't you simply 
transform first with one allele and then the other?  Because that's the only 
way I can think of reliably getting both into the one cell, and even then 
it's hit and miss as to whether you'll retain both.

It doesn't look to me as though cloning heteroduplex DNA is going to get you 
where you want to go.  It may be that in a tiny minority of cells you don't 
repair the mismatch, and end up heteroplasmic for the two alleles.  Or, if 
you just transform the mixture using a large excess of plasmid, a minority 
of cells will pick up more than one plasmid, and of that minority a small 
proportion will end up with both alleles.

In each case, while you may get a very few colonies that carry both plasmids 
(but with no guarantee of preserving them both), you'll have a massive 
background level of colonies that only carry one or other of the alleles - 
and there won't be ably way or telling which is which, short of genotyping a 
gazillion colonies.


More information about the Methods mailing list