@Trof: do you have any idea how this is possible? If you use the primer mix in a PCR, your first step is also a denaturation step so any primer dimers made in the primer mix would be expected to be denatured at that time, no? I'm asking because we recently ran into problems with one of our PCRs. When we make a primer mix of our 4 primers (it's a duplex reaction), the PCR doesn't work, but when we add them separately, it does work. The first time we tried out the primer mix (not freshly made but after freezing it once), it worked, but following experiments always failed, even when using a different aliqout of the primer mix. I must say it is a difficult template (GC rich) so we need DMSO in the reaction as well, but this is only added during PCR setup.
I think I've seen this as an offcial Roche recommendation in their protocols using primer-probe mix, there is a note there "Due to possible primer/primer interactions that occur during storage it may be necessary to preheat PCR primer-probe mix for 1 min 95 C before starting the reaction. This extra step will ensure optimum sensitivity."
I don't really searched for the reason, but I already encountered this situation once.
I may just guess that maybe the behaviour of primers is different in the final reaction mixture, so they maybe less willing to get out of any structure they created in storage.
I think it could be due to the primer design. Too high annealing temperatures or too large negative ΔG values, giving rise to huge reduction in product yield as a result of formation of primer secondary structures. Having said that, it becomes a headache when attempting to design primers. I haven't actually tried denaturing the primer mix before adding them as I haven't done a duplex reaction as of yet. But yeah, like you said it would be interesting to do so. Also, is there really a need to add DMSO in PCR? Thank you.
Well the problem is those primers don't form any undesirable structure when added separately, but some probably do upon a combined storage. This in not easily explained or predicted in the primer design, and it won't be unless there is an exact knowledge of what lies behind this, given that it is not very common thing.
About that primer headache, you probably need to see some of the "bad" primers in action (like, some published ones, ones you "inherited") to have less headaches about deltaG values that softwares spit out
And I just design all my primers to slightly higher annealing tempreratures (60 C) because it increases specificity.
DMSO is a PCR additive used for decreasing the annealing temperature needed for specific binding, also it relaxes secondary structures on target DNA so it's commonly used for GC-rich templates. On the other hand they decrease activity of polymerase significantly with and increasing concentration.
So no you don't need it in normal PCR, but you do on GC-rich one, that wouldn't amplify otherwise (5% DMSO is currently only optimization I do, either if reaction doesn't work, or I can guess from the look at the amplified sequence that it might not, but I have a hell of a PCR Master mix ).