EtBr in agarose gels vs. EtBr staining later

Karl Voss karl at
Sat Mar 4 14:23:53 EST 1995

Andy Bates (bates at wrote:

> The behaviour of supercoiled DNA on an agarose gel is a complex area. 
> However, in general, if you are interested in the state of
> supercoiling, then you have to stain afterwards.  Almost any amount of
> EtBr in the sample or the gel will result in the closed-circular DNA
> running with positive writhe on the gel, that is with high mobility
> relative to nicked, or open-circular DNA.  In other words different
> levels of supercoiling will not be resolvable.  If this is confusing,
> then get back to me privately if you like.

One interesting thing that can be done with EB in the gel is measure (or
at least estimate, the energy of supercoiling.  If a CCC dna is incubated 
with a topoisomerase until it is fully relaxed it will actually be mostly
fully relaxed molecules and some molecules varying by one or two linking
numbers in either the negative or positive sense.  This is just due to the
boltzman distribution and the energy of supercoiling.  If you run this 
sample on an agarose gel you can see the fully relaxed dna move slightly 
slower then the dnas with +1 or -1 supercoils.  But the +1 and -1 supercoiled
dna will move the same in the absence of EB.  With EB added all the dnas
will go to the maximum +ve superhelix density.  This will be different for
dnas with linking numbers of +1 and -1 and so they can be distinguished by
their mobilities in the gel.  Anyway the idea is that you can use EB in 
the gel to examine some of the topological properties of dna.

I run gels quite often with EB in them.  Particularly when checking clones
by cutting out small inserts (<100 bp).  I find that I can see these bands
better when I use EB in the gel.  I run the gel so that my bands should be 
farily close to the end.  Then I switch the buffer to a TBE without EB and
run some more.  Because EB is +ve it will migrate toward the cathode and
leave the end of the gel with out any EB background.  This makes it much
easier to see the bands.  I suppose adding EB to the sample would be 
the same but I always figured that there would be no EB left by the time
the gel was finished running.

Karl Voss
Department of Chemistry
University of Alberta
Edmonton, Alberta, Canada
T6G 2G2

karl at
phone 403-492-0222
fax   403-492-8231

More information about the Methods mailing list