Given that you can detect a much smaller quantity of your positive control there does not seem to be any reason to suspect a transfer problem, assuming your HBV bands are not significantly larger. If they're very large you'll need to fragment them in the gel prior to transfer (UV or HCl treatment) to have efficient transfer. Also, I recommend alkaline transfer for maximal efficiency. If you're seeing clear bands on your gel, there is no problem with digestion or migration of the DNAs. Also, the clear bands indicate the DNA migration is not affected by buffer salts, and salts would migrate ahead of the DNA. You did not specify the nature of your probe molecule. It sounds to me as though either [1] what you're pulling down is not really HBV, [2] variability in the sequences detected by your probe is too great, or [3] your HBV probe itself is not properly labeled with dig. Others may have other ideas, too.

Thanks David for your comment,

Actually my control lenght is as big as my HBV genome which I expect to detect (it's a full lenght PCR product) on the other hand this control is well detectable with my probe even in low concentrations (until 500pg), means DIG labeling was efficient. If I amplify the extracted DNA with real time PCR I have good amplification of HBV DNA so I Would expect that my extracts are real viral genome. Abput variablity of DNA I would say I get this extracts from transiently tranfected cells so there shouldn't be considerable differences between the DNA which I have with the one which I used for transfection. I almost tried to set up all the critical parameters and since its not still working I assumed it might be because of the purity of DNA !

Did you check the purity? A simple A260/280 will tell you if there is any significant contamination with protein. Lipids would not withstand phenol/chloroform extraction. Complex carbohydrates can interfere with initial isolation of DNAs in some systems, but once the DNA is isolated they are not a problem. I'm not sure what other contamination might give you trouble. It sounds from your description like your DNA is cutting and migrating normally, which also argues against any significant contamination. Transfer must be going well, as your control is there on the membrane and is similar in size. To my mind, that really leaves only a mismatch between DNA and probe for some reason. One last alternative: if you're using chemiluminescent detection on film and have TOO much of your target DNA on the blot, you may be overwhelming the film emulsion's ability to be reduced, which leads to a reversal of that part of the image. In other words, what you might expect to be black can actually show up clear (e.g. a direct photograph of the sun on b/w film gives a clear sun with dark halo around it). You didn't describe the nature of the probe or sequence being detected. If it is an oligonucleotide probe it is easy to miss a slightly altered target, especially if stringency is too great. If it targets sequences encoding either an immunologically-recognized part of a protein or a segment of the genome with no positive selection it may be too variant, in which circumstance even a longer probe may hybridize poorly.

One final comment: the ability to detect by qPCR may be misleading if your primers are to a conserved region of the genome. I study a variant multigene family, so issues like that always come to mind. If you have additional HBV primers you may be able to amplify a segment of the genome for sequencing to convince yourself it really is HBV and not something closely related but only 90% identical. Really, it sounds as though everything but the hybridization itself is going well, suggesting a mismatch between probe and target for some reason. I hope this helps.

Your comments are actually great.

Regarding the extraction method I doubt that I have something rather than my virus DNA at the end since first I do immune precipitation with specific Ab to separate viral particles from other cell components, then proteinase digestion, phenol-chloroform extraction and finally salt precipitation.

Anyhow, as you suggested, I'm now thinking of using a new probe for another part of the genome. I use DIG labeled RNA probes (about 800bp) and hybridization at 42C overnight. I also avoid stringent washes to keep any even miss matched hybrids (I think it's ok as long as I have no background or wrong positive signal).

The idea of sample purity came to my mind since I have a white pellet after salt precipitation (1:10 ratio of 3M NaAc) which disappears when I add water to dilute DNA. I checked the ratios, they are 1.2-1.5 for 260/280 and 1.3-2.2 but in some cases even 6 or 7 ! for 260/230. I know that they are somehow indicating not a good extraction; however I don't know how much they can interfere with transferring procedure. If this also could be the case I would do both modifications for next experiment,

Changing the probe and purifying DNA!

Those ratios and a white precipitate that easily dissolves in water strongly indicates you've got a lot of precipitated salt. That is very easily fixed by using ammonium acetate rather than sodium acetate at the precipitation step. Use 0.5x volumes of 7.5 M NH4Ac + 2x vol of EtOH (or 1.5x vol of isopropanol) to bring down the DNA. Wash once with cold 70-80% EtOH (depending on size of the DNA) and air-dry. NH4Ac has the advantages of precipitating far less than NaAc to begin with, and of being volatile. If you have a small amount of NH4Ac it will dissipate as NH3 and acetic acid, both of which are volatile. I use it routinely for those reasons. Drying under vacuum will get rid of it completely, but then the DNA is harder to get back into solution.

Sounds like you're on the right track in modifying your probe. RNA-DNA hybrids are more stable than DNA-DNA to begin with, and you're using low stringency washing, so it should be hybridized if it's going to. It really sounds like a sequence recognition problem. Good luck.

Is there any possibility to use P32 labeled probe? That is how I solved my Northern blot problem.

Thanks for sharing your experiences David. I have one more question, after adding NH4Ac to the samples which incubation temperature and time is appropriate for low concentration of DNA. I read in some manuals that room temp/ON + centrifugation at 14000rpm/room temperature gives the best amount of DNA while in our lab we incubate samples at -70C for 30 min and then cent at max speed at 4C (to protect from DNAse I guess).

Dear Li Song, actually a dot blot assay using P32 was already set up in our lab which works fine but I want to set up a southern blot with non-radioactive detection method.

Hi Elham- With NH4 acetate I have mainly used -20C for a half hour to precipitate the DNA, then centrifuged at 15,000 xg for 15 min, either at 4C or room temperature. Room temperature overnight would probably work fine, too. Either should result in nearly quantitative precipitation. Solutions of sodium acetate incubated at -70C do have more precipitated salts than if incubated at -20C, and the salt is much harder to get rid of than NH4 acetate. Good luck.

Hi ppl i have a problem in isolating DNA from a butterfly larval sample for southern blotting.I will be really glad if someone responds to this

Hi I have the same problem. I get good band with control where as the mutant bands is very thin or it not appear. Please, let me know who could you solve this problem.