Problem with genomic PCR (2.4 kb) - (Oct/31/2011 )

I am making a knock-in mouse and trying to screen a batch of ES cells for a homologous recombinaiton event using PCR.

My reverse primer is in the Neo cassette in the construct sequence and my forward primer is outside the construct sequence in the genomic DNA so that the band of 2.5 kb is amplified only if the recombination occured. I also have another forward primer in the construct sequence, which I am using to optimise the reaction. It has similar melting temperature to the final forward primer and in combination with the Neo reverse primer it gives a band of 2.4 kb, which should be present it all Neo resistant clones, regardless of construct incorporation site.

I used such combinations of primers for the same purpose on same type of DNA extractions over 1.7-2.3kb before and never had a problem. The clone DNA concentrations are about 200ng/ul and the 260/280 ratios are 1.8-2.0.

I ran a short PCR (500bp product) first using optimised primers to a different gene to check for DNA quality and figure out genomic DNA dilutions. 2ul genomic DNA seemed to work best (although it does seem an awful lot of the template!) so I stuck to that amount for the next experiments.

I have also verified the presence of Neo in the clones by well established PCR, using construct DNA as a positive control.

Overall, I am able to amplify products up to 1.1kb from my genomic DNA, but I am struggling to amplify the 2.4kb band.
I optimised the 2.4kb PCR on plasmid DNA (20ng/ul original construct DNA used as template) in standard PCR conditions or substituted with 5% DMSO, 1.75 mM MgCl2 or 2.5mM MgCl2. It worked in 2.5mM MgCl2, giving a nice chunky band. However, on genomic DNA the band was very faint and sometimes didn't PCR again from the same clone. Furthermore, it only worked using 2ul of template DNA (400ng) and didn't work with 1ul or 0.5ul template.

Increasing the number of cycles from 30-40 didn't work, neither did increasing or decreasing the amount of template or increasing the amount of dNTPs. All primers have been blasted and are non-repetitve.

The only other thing I can think of is using the PCR product as a template and aplifying it again for 30 cycles, but my optimising PCR (the 2.4 kb one) should work (at least give a faint band) in every clone and at the moment it doesn't either.

Also, if I was to run another PCR based on my first PCR product, would I be able to do it without the purification step (i.e. just add fresh Taq, dNTPs and primers)? I need something that would work in 96-well format.

Any suggestions? Please help..

Which enzyme are you using and how long is your extension time? (can you give us your PCR cycle conditions?)