High background on Northern blot

Fri Sep 23 10:34:52 EST 1994

T. Andrew Gu wrote:

[Doing Northern blots with Nytran membranes and Dupont's chemiluminescence
> ...the background of the autoradiograph was very high.  It seems 
> that my plasmid probe also crosshybridized to the
> 28S ribosomal RNA. 

Do you mean to say that this one extraneous <?> band is the only problem,
or are you telling us that you also have signal spread all over the filter?
You're going to get a lot of advice about blocking agents, etc. that is
only relevant if you have background appearing outside of the lanes.  Also,
with chemiluminescence the image of the whole filter will be black if you
expose long enough.  This is not a malfunction as long as your signal to
noise is good.  You just take a shorter exposure.  Are you telling us
that you see the signal over the 28S RNA instead of, or in addition to
the expected signal?

> Several factors may be responsible for the problem. One possi-
> bility is that I used the whole plasmid as a template in probe labelling. 

First find out if vector is a problem by probing a control filter
with a vector-only probe.
The vector can raise a background in several ways: 1) It might actually
cross hybridize to some RNA species.  Usually this can be blocked by
including some nonspecific cold competitor DNA in the probe, and/or by raising
stringency.  2) You could have actually contaminated your sample with some
plasmid.  I hate to tell you what all I've found in people's loading dyes.
3) You can get background all over the filter or in smears by using trays
or gel boxes that are loaded with DNA from someone's plasmid prep.  I've
even seen a clear image of a ruler come through that someone laid on
the gel during photography.

If you have problme 2 or 3, then your technique is not clean enough to
work with RNA in the first place.

> I also heard that nylon membrane usually generates more noise 
> than nitrocellulose membrane. 

This is relevant to background appearing outside of lanes, but not to
cross hybridization type problems.

> Finally, I am wondering if I should isolate poly A mRNA to avoid the
> nonspecific hybridization problem. 

I'd say as a last resort.  If the band over the 28S RNA hangs in through
the above controls, then you have to consider that your mRNA really runs
there.  It also could be an unspliced precursor or an alternatively 
spliced species.  Then you'd probably have to do the poly(A) prep to be 
sure.  Before I'd do that, I'd run a longer gel and take careful measurements
to see if I couldn't just resolve the signal from the 28S RNA.  

Some people have claimed that the heavy rRNA bands can actually trap some
portion of the smaller mRNA species during the gel run.  I find that
a little hard to believe, but I don't know.

Hope this helps,
Steve Hardies, Dept. of Biochemistry, Univ. of Texas HSC at San Antonio
Hardies at thorin.uthscsa.edu

More information about the Methods mailing list