Southern blotting - high background, no clear signal? - (Apr/05/2012 )

Hi there,

New to the forums, just looking for some help in an area where I have been struggling.

My lab is trying to use Southern blotting for genotyping confirmation. At the point where I joined the lab and took over the blotting, we had already chosen and ordered primers for probes, and I cobbled together a protocol from my PI's protocols, the internet, and instructions that came with our solutions/membrane.

I am getting successful labeling of the Lambda HindIII ladder and of amplified PCR fragments, so I know that the problem isn't to do with labeling the probe or DNA transfer or gel runs or anything. What we're seeing is bright labeling on the ladder and some smearing and background in the sample lanes but no distinct bands anywhere.

Quick outline of protocol:
- Genomic DNA from whole tissue by phenol/chloroform extraction, ethanol precipitation, and resuspend in TE. Check on a Nanodrop for concentration - DNA is clean and pure.
- Digest DNA with Sac I up to 24 hours - have tried several different amounts, anywhere from 10 mg to 40 mg, changing volume accordingly.
- Run on 0.66% agarose gel at 50V for 8 hours; imaging shows clean digestion, no excess of undigested DNA.
- Denature in NaOH/NaCl solution for 1 hr
- Neutralize for 30 minutes
- Blot onto Hybond - N+ membrane overnight via neutral transfer.
- Crosslink blot, rinse, and let dry before storing.
- For probing, re-wet blot in 2x SSC.
- Lay blot on a bit of nylon mesh and place in a hybridization bottle.
- Pre-hyb with 7.5 mL Amersham Rapid-Hyb buffer for ~30 minutes (have tried this up to overnight with no difference on outcome) at 65 degrees C.
- Mix thawed probe (~15 ng) 50/50 with formamide and denature at 95 degrees C for 5 minutes, quick cool on ice, and pipette into hybridization bottle.
- Hybridize overnight at 65 degrees C.
- Rinse 3x 10 minutes in 2x SSC/0.1% SDS at RT, then 3x 15 minutes in 0.1x SSC/0.1% SDS at 65 degrees C.
- Wrap blot in plastic wrap and place in a storage phosopho cassette.
- Image ~24 hours later, then lay down again for 7-10 day before imaging again.

We have been trying this for a month with this particular genotype and have had no luck. Attached is an example of what we've mostly been seeing with this protocol. Can anyone provide me with some advice on what I might be doing wrong and what I can do that might increase my chances of getting a clear signal? We are expecting bands at 16.85, 10.04, and 6.96.

Thanks so much for your help, excited to get some groupthink from a community of people with much more experience than I have!

Have you tried different hyb temperatures?

Do you depurinate the DNA?

If you are crosslinking by UV, be careful not to over irradiate the membrane. If you can use a cross-linker, such as a stratalinker, set at 120 mJ. Crosslinking using short wave UV usually takes less than 1 min.

My denaturation is typically 2 x15 min in NaOH/NaCl and the neutralization is the same in tris-NaCl. Gel is then equilibrated in 20x SSC for 10 min.

Have a read of the DiG application manual for filter hybridization (Roche), it has full protocols, and a good trouble-shooting section.

Work backward from detection. Do serial dilutions of your probe, and spot on a membrane (crosslink). Detect the probe, to make sure the probe can be detected and to quantify its sensitivity.

Next, do serial dilutions of your DNA and spot those on the membrane (together with the spotted probe) and do detection of the serial dilutions of DNA. This will tell you if your hybridization is working and the sensitivity of its detection. You can optimize hybridization conditions and background with this, which is much easier than handling gels.

Only when this is working well do you run gels and transfer.

Depending on what kind of genotyping you are doing, you can also consider using other techniques. For comfirming homologous recombination of targeting vectors, I have found RT-PCR to be extremely powerful, for example, especially when combined with sequencing of the PCR products.

Great, thanks so much for the feedback! @phage424 I will try working backwards as you recommend - hopefully that gets us somewhere.