ethidium bromide: add to gel before pouring or stain afterwards

David L. Haviland, PhD dhavilan at IMM2.IMM.UTH.TMC.EDU
Sun Jun 28 07:41:40 EST 1998

On Sun, 28 Jun 1998 vargaa at wrote:


> My lab collegues and I have been debating this for a while now...

No surprise, most every lab has some sort of methodological debate...

> To detect RFLP's, we currently add ethidum-bromide to a melted, high
> resolution gel, then re-nuke it so to pour an even gel.  I, however, think
> this is a highly mutagenic way to do things, particularly since the gel is
> volatile, the ethidium following with the vapour...  I would prefer to stain
> the gel afterwards, following by a destaining bath in water, of which I have
> noted to work perfectly fine...
> I have visited other labs doing gel electrophoresis, and most seem to stain
> the gels after running.. not prior to running into the gel before pouring...
> this is sometimes hard, especially when using large gels (minimum of 30
> wells)...
> Please, end our constant in-fighting!  Does anyone else add the ethidium prior
> to nuking and pouring?  or do most of you RFLP'ers etc stain the gel after
> running?

You've ommitted option #3... Add EtBr to gel *after* micrwaving as it is
cooling down...

What would skew this would be if your lab, or part thereof, has a "master
solution" of agarose (say 300 mls) which is repeatedly microwaved for gels
until it is gone.  Personally, this is a practice that I do not allow in
my small lab as the gels won't be consistient due to water loss.  Also,
and I don't know if it is true, but I've been told by more than one
individual that EtBr does not tolerate the heat needed to melt agarose
(this could be molecular-biological "folklore" for all I know).  People in
other labs have nuked the EtBr but they seem to use more of it that I do.
(Jim_G, David_S, Dom_S any thoughts here?)

Staining post run never appealed to anyone in our group as it is time
intensive.  The one "hold-out in our group who was staining afterwards
eventually started adding EtBr *after* nuking as destaining was taking too
much time.  

Sure, EtBr can alter the migration of DNA but so what? Wouldn't migration
be altered proportional to the length of DNA and concentration of EtBr
used?  Just for fun, I ran a gel with two lanes - one with cut bluescript,
and the other was lamdaXHindIII markers.  Into these samples I put 0.5 ug
EtBr into the loading buffer, none anywhere else in the gel (gel itself or
buffer).  The next two lanes were the same but had 1ug EtBr, the next had
5, and the last two had 10 ug -- all samples were run on the same 0.8%
gel. What I saw was a difference in how the four groups ran. Yes, the 10
ug EtBr ran slower relative to the 0.5 ug EtBr samples.  However, the
relative spacing between the bands among all the LH3 samples was very much
the same from what I could tell.  I've never had the inclusion of EtBr
(post-microwaving only!) alter the interpretation of a gel - without an
insert, bluescript has never shown up as a 5 kb band!  I guess it could if
one were to put different amounts of EtBr in separate lanes but that sort
of person would be better of not being in science!

I put about 2-3 ug (from 1 mg/ml stock) into a 125 ml gel and I have my
ficoll-based loading buffer with 0.5 ug/ml of EtBr.  I always add EtBr to
the gel after microwaving and as I've said, it has never been a problem
with regard to interpretation.

Hope this helps,

David L. Haviland, PhD
Assistant Professor, Immunology
University of Texas, HSC - Houston
Institute of Molecular Medicine
2121 W. Holcombe Blvd.
Houston, TX  77030

More information about the Methods mailing list