Problem with a genomic Southern blot

Sonya Clark szsclark at HAMLET.UCDAVIS.EDU
Wed Sep 1 18:37:28 EST 1993

Dear netters, 
I am attempting to perform a relatively straight forward genomic Southern
blot but am getting a strange result.  My probe is a 810 bp cDNA isolated
from a pea gt10 library which encodes one of the oxygen evolving subunits
of the chloroplast.  This gene is known from in vitro experiments to
encode a protein that is transported to the pea chloroplast and that acts as
expected once in the thylakoid membrane.  This pea gene also shows >80%
homology with the corresponding (and independently isolated) spinach cDNA. 
 My genomic Southern blots were performed with 36 ug of pea genomic DNA
digested to completion with 4 RE's (EcoRI, BamHI, HindIII and XbaI).  1 (7
pg), 2 and 3 copy reconstructions were run on the same gel from fragment
cut from the original plasmid, gel-purified, diluted and mixed with
equivalent loadings of herring sperm DNA.  Gels were blotted to Zetaprobe
membranes by alkaline tranfer. 

As I was using the BM Genius kit for the first time I ran a second gel
containing BM-supplied control DNA (cut pBR328 DNA) at several dilutions
(in herring sperm DNA) which was identically Southern blotted.  I also made
up a series of dot blots to act as controls - one of my probe DNA (10 pg to
0.1 pg) and one of the BM control DNA across the same range.  I then
labeled my probe (which had been made by PCR from the plasmid template using
Universal M13 forward and reverse primers) as well as some of the BM
Genius kit-supplied unlabeled control DNA and ran the two hyb's
side-by-side at 68oC overnight following BM's instructions.  Visualization
of the blots was by colorimetric detection using the Genius kit reagents.

My results were confusing.    All my controls were detected as
expected - ie. I could detect the BM pBR328 DNA down to 7 pg on the
Southern and down to 0.5 pg on the dot blot.  My own controls also worked
- dot blots detected to 0.5 pg, Southern copy reconstructions detected
at 3, 2 and 1 (7 pg) copy.   However no hybridizing bands were detected
within the digested pea DNA :-<.

This Southern was also carried out by a post doc in our lab a short time
ago and he got this exact same result.  To check his membranes he
followed up by reprobing with a different pea gene for another chloroplast
protein and got the expected Southern.  I therefore suspect that this is a
'real' result and not an artifact/technical problem.
 I cannot think of any logical reason for this outcome.  If anyone has
struck this type of result before, or has any ideas as to how it may come
about I would be delighted to hear from you.  Thanks in advance for your help.

Sonya Clark

Botany Department
University of California, Davis
szsclark at

More information about the Methods mailing list