Southern blot troubleshoot - (Jun/24/2010 )

Hi

I have been trying to do some blots with GFP probe. I would like to know how many viral integrants are there in a transgenic mouse. I have tried it 5 times so far and I even have trouble getting clear marker bands. I do this with people from a neighboring lab all the time and they always get clear blots. So I don't think its a problem with the reagents. The probe is just GFP around 700bps which I cut out from a vector. I digest the samples with EcoRI. I am at my wit's end now

So what is the actual problem? Probe labelling failure? Detection? High background? non-specific binding?...

Does EcoRI chop somewhere in the probe sequence? Should it give you more than one band if there is more than one integrant?

I think I wasn't clear enough. Sorry. I don't see anything on the film after exposure. I can sometimes see a few marker bands but usually there is nothing for the samples and there is no background. I don't think there is anything wrong with the probe labeling or the reagents. I do everything together with other people and they usually end up with nice blots. The GFP sequence does not have any EcoR1 sites. And if there are several viral integrants, I should see as many bands.

I was told by my boss that I could use the same membrane that I use for Western blot and yesterday a colleague told me maybe that the problem as they use special membranes. But I had used theirs the first time and then there was too much background. I am going to try again with their membrane and see if it works. But if you could think of anything else, please let me know

Nylon membranes work much better for Southerns than PVDF or nitrocellulose.

It sounds like you may be having transfer issues otherwise. Make sure that the stack is tall enough (10 cm or so should be enough) and that you have 200-400g weight on the top. Also make sure that the stack is not short-circuiting (i.e. stack not touching the bridge).

For lowering background, ensure that you block the membrane properly and use the correct hybridisation temperature for the probe. If you have experienced people in the lab, especially if they do this sort of blot routinely, they should be able to help. Even observing as they do one may give you a hint as to something you are missing.

Positively charged nylon membranes are the thing for southerns and northerns but from what you've said your issue is untraceable. IMO this sentence of Bob's is worth addressing; "For lowering background, ensure that you block the membrane properly and use the correct hybridisation temperature for the probe."

If the probe and hybridization conditions aren't running routinely and the conditions used 100% replicated by you, perform a couple of slot blots to get that side of things running well, it's faster than a full blot and success there would indicate that you have transfer issues. If you're using DIG I'd recommend their easy Hyb granules, in my hands they outperformed a number of other buffers and were much easier to use.

I prefer shorter probes too (150-300bp).

It sounds like you've got experienced people to hand so I'd sit down with one for 30 minutes.

Maybe I wasn't clear enough. But I always do the blot with someone who is doing a blot at the same time and so we do everything together starting from the membrane cross linking. And when we develop the film, they always have good blots while mine doesn't have anything. And I have never had the background problem since the first time. Now the blots are clear with nothing on them except for maybe 2 bands of the marker.

I asked several of the Southern blot experts in the other lab and the only thing they could think of was the membrane. So this time I have done it with all the same reagents and protocol and used their membrane. But the membrane had a very low radioactivity signal compared to my colleague's and it was only around 40. The marker was around 50 and I am not sure whether it was just the marker which was beaming. So I have to wait till next week and see whether it works. Keeping my fingers crossed. If not, I guess I will be back to square 1. Maybe I can try to cut my GFP probe to a smaller size. But my colleague said 700bp is fine. Any other suggestions?

Check your probe - do a dot or slot blot to ensure that it is properly labelled. Otherwise PCR label part of the GFP gene and use that as a probe, it should be a lot hotter than an end labelled probe.

bob1 on Jul 7 2010, 03:49 AM said:

Make some PCR primers for your gene target (GFP in your case). Using alpha phosphate 33or32P labeled nucleotide (i.e. one of 32P labeled dATP, dCTP, dGTP, or dTTP) in your PCR in the place of the normal dNTP, amplify the fragment and purify the PCR product to get rid of the parent material.

You have to use alpha labeled dNTP for this as the beta and gamma phosphates are lost during the polymerisation into the DNA PCR product.

Better still would be to move away from using radiolabels altogether and use something like DIG which works extremely well, the probes never go off and you don't have to take all the precautions like you do with radioactivity.