Annealing temperature and two-step PCR - Is the annealing step necessary? (Apr/24/2007 )

I pretty much always do my PCR cycles with a 3-step cycle (denature, anneal, elongate). But I've seen some programs with only 2 steps, skipping the anneal step.

My question is, when is the annealing step necessary? If the Tm of my primers is above 72 degrees, it means that at 72 degrees (elongation temperature) they will be annealed, right? So I could just skip the annealing step and save time?

Is the annealing step only necessary if the Tm of the primers is below 72 degrees? (My primers are all very high GC)

There is, to mynknowledge, no general rule for the decision, whether two step protocol work or not (please correct me, if I am wrong ). In many cases, you will use a very low annealing/extension temperature, very often about 60°C.
If it works for you is a matter of trying, I guess...

Krümel

After I posted that question here I read up on PCR theory and I think I can answer my own question

A common question is "if the annealing temperature is below 72 degrees, won't the primer come loose from the template when the thermocycler goes up to 72 degrees for the elongation step?"

The answer lies in the fact that Taq can still elongate even below 60 degrees, just not as fast. As soon as the primer anneals, the Taq starts elongating the new strand, and thus the Tm goes up (very rapidly). Taq only needs to add a few bases until the new strand has a Tm well above 72 degrees. At this point, the thermocycler can ramp up to 72 so the Taq can operate at full speed, with no chance of the primer/new strand coming loose.

So based on this I would say that the annealing step can be omitted if the Tm is above 72.

Now some of you may be thinking "you really only need 10-20 bases to make the Tm go above 72 degrees, so is a 30-second annealing step really necessary?" well apparently a 5-second annealing step works just as well.

(That Biotechniques article is very interesting; it claims that DNA can denature in 2 seconds at 94 degrees, annealing for 5 seconds is sufficient, and that the 1 min = 1kb elongation time is grossly unnecessary).

Hi Zouden,

You are right, but how you do a PCR depends on why you are doing it. Remember, all molecules do not behave the same at all times, and that any mixture of different chemicals, especially at varying temperatures, is a wild ride of associations, dissociations, extensions, aborted extensions, you name it. It's what the majority do that counts. If you just want to have a quick look at whether or not something is there, then a blitz programme is good enough. I have done 3-temperature 30 cycle PCRs in well under an hour. But if you want to clone and sequence something, I would be a bit more conservative. I once got a frankenstein clone from a PCR of a tetraploid organism where one allele was partially extended but not completely. This fragment then acted as a primer on a different allele during the next cycle, so the full-length molecule was part allele A and part allele B. Just my luck that this molecule ended up getting ligated into my cloning vector. So think of PCR products as populations. All populations have their freaks.

Agree with wbla3335, if you just need to confirm presence/absence for eg a pathogen, you could indeed try to minimize PCR time (in case you do not lose any specificity/sensitivity), otherwise i'd stick to optimised protocols, perhaps decreasing extension time a bit, or decreasing denaturation temp some degrees.

Hi swanny,

By sequencing many clones. At first we thought we had a case of gene conversion, but we ruled that out after further analysis.

This sounds like good advice.

Something I read today suggested very short cycles, (0 seconds denature, 0 seconds anneal and 0 seconds elongate!! The thermocycler simply ramps to that temperature then moves on) followed by a single long elongation step (10 minutes or so) to complete all the new strands.