What type of gel are you using? For larger fragments,

How do you reconstitute the primer sets? 

Use 1X TE to dissolve them and to dilute them to get required conc. stocks.

Set the temperature and time profile as per the recommendations of the taq enzyme's

(eg. a high fidelity taq) manufacturer you use. And also for the dNTP concentrations. 

Continue working with routine components of PCR, dont include DMSO or other enhancers in the initial experiments. 

Try a simple experiment with your fresh reagents. Use a sterile injection water (for PCR set up, reagent dilution TE preparation which can be bought from medical stores) dont trust your lab's deionized water for instance. 

Good luck with your PCR

i would visit the nested pcr again. Amplification is very efficient in nested pcr and a high dilution of the starting template and very few cycles are needed to get a good product eg 1ul of 1/100 dilution  and 10-15 cycles...10 cycles is 1000 times what you started with. The large product may be that you have so much product that the product is self annealing ( with other product molecules )and just getting longer as the polymerase fills in the double strands  because there is too much of it and the primers are no longer in huge excess as is usual with pcr. Also run product on a 0.8% agarose gel as it is a large product

Firstly, I'd try to make absolutely sure your electrophoresis conditions are fine. As suggested, around 0,8% agarose gel in TAE 1x, ~5V/cm should work well. Run a high-range ladder, wide range ladder - see how they behave. Given that you're seeing very strong bands from your product I'd also try running different amounts on the gel.

While I might be totally wrong, I think that time of elongation phase should mostly limit the size of PCR product you're getting. Though, if the electrophoresis seems fine for the ladders and your product is still not leaving the well then the self-annealing product might be the case.

may be the quantity of DNA loaded be the reason...quantify the PCR product and use required (ng) quantity to load the gel 

Dear,

Based on your explanation, I suggest you:

1. please, decrease your annealing temperature. In my opinion, usually annealing temperature is just in range from 50C to 60C although it can be increased in specific case such as because of using DNA extract from specific biota for example deep sea animal or cryptic animal.

2. please, check your gel concentration (Agarose/Acrylamide). Usually we use 1,5% - 2% Agarose Gel for PCR product. And please make sure that the gel have been solid in all of it's part after it is placed in it's mold.

3. please, check your loading dye. Based on your explanation, usually the problem is caused by loading dye.

4. please, check the concentration of genomic DNA extract. You can dilute it to make lower concentration. Lower concentration of DNA extract can also prevent the occurrence of smear in your DNA band.

Hopefully, all of them is useful for you.

Best regards,

Hi all,

Thanks a lot for your input, but the problem is still not solved

1) 0.8% agarose- the problem is that the product vastly exceeds 10kb and is entirely unexpected. I've been running 1% as it picks up best on the desired 3kb.

2) Too much starting material- I am using the lowest recommended concentration of 10ng/reaction. Should I go lower?

3) Old reagents- I only use high grade reagents and I actually recently replaced all of my aliquots with fresh reagents.

4) Paul- the nested PCR comment actually makes a lot of sense. I revisited my gels and I have one with an upward smear to the well. Given that I was using impure DNA before and I'm now using higher quality DNA which gives the strange band and given that my nested PCR gives the same band, I would say that this could be a too much product issue.

5) quantifiedPCR- 700 ng/uL, after dilution, I still see the same thing. 

6) Piotr- elongation time- this is very possible. I will try this Monday. I've been using max elongation time.

7) cycle number- also possible, will make sure to get an aliquot halfway through.

8) loading dye-I ran a positive control from a reaction that worked accidentally- the dye and gel are fine.

sample not leaving the well...are you sure this is dna not protein....does an equivalent dna sample pre amplification look similar?

also very often I associate a smear with only 1 primer working so chain elongation is lengthy ( at c100bases a second extension )  and random where the enzyme falls off. Might be interesting to see if a single primer can produce a smear in which case the other primer is simply not working.