smearing on hyperamplification.

Basavaraju Shankarappa bsh at MED.PITT.EDU
Mon Nov 30 09:59:09 EST 1992

 >I amplify a fragment of DNA by PCR from genomic DNA and then take 
 >a small aliquot of the first reaction and add it as the template for a 
 >second PCR round. The result of the second reaction is a smear of 
 >generally large pieces. The target fragment is no where to be found. I 
 >have attempted a number of things to get the reaction optimized, ie, 
  >annealing temperature, [dNTP], [MgCl2], [primer], [template], etc. I'm 
 >aware of nested primers but haven't tried that. I was just wondering why 
 >this second reaction doesn't work with some templates and does with some 
 >others? I suspect that the template undergoes some kind of self priming 
 >and concatenates itself into long fragments. 

Almost all the time you can overcome this problem by diluting the original
PCR reaction to a very large extent.  It is often difficult to immagine that
few microliters contain billions of copies.  The reason for smears (I think)
is that when you do the first round of amplification, there will a large 
number of amplimers of diverse lengths which just do not show up on a 
agarose gel but invariably present.  I think it is these that end up 
getting amplified to give a smear.  I dilute my amplimers atleast 1000
to 10000 and load 5 ul in a 100 ul reaction.  YOu can in fact do a 
titration of different dilutions and can very dramatically see that as the
template is diluted, the specific signal starts to intensify.  The bottom
line I think is that your primer pairs are not sufficiently specific.  
In our lab I have seen this happen even when a band cut out from a gel is
used.  If you have to use this product my strategy is to stick that agarose
plug in another agarose gel and run it again.  I have seen it to reduce the 
smearing to a considerable extent.  

 >Then of course the question 
 >is why doesn't that occur during the initial PCR reaction?
Again, I think you just do not see it because of low copy numbers of these
I am sure others have had the same experience and may have better explanation
or strategies.
Good luck
Raj Shankarappa
bsh at

More information about the Methods mailing list