Hi Esteban, no problem, it is much longer than a primer, it will anneal and be extended as if it were a primer, the 5' long tail won't matter much.

Only precaution, you should run the first 5 cycles without primers. Then you pause as the denaturing step starts and quickly add the primers.

Good luck.

pausing before denaturing and adding primers? that's new!. Good idea. Are you saying the annealed sequences would act as their own primers and then i can do the amplification with the primers? sounds terrific.

That's exactly the trick... for the first 5 cycles the two fragments act as primers on each other producing a number of entire templates. When you add the primers at the beginning of the 6th cycle, just as you reach the denaturing temp, you'll have your full/fused template ready to be amplified.

Of course this means you should pass on gel the two fragments to get rid of the first two pair of primers, otherwise you'll carry on whatever was not used in the next PCR.

yes, i'm cleaning up gels all along this procedure. that's the worst part because efficiency is very low (aprox 10 ng/ul). I'm excited to try this. I am ready near the thermocycler with my primers.

Hi Esteban,

if you have nice, distinct bands on an agarose gel, there is no need for gel purification, just purify with a PCR column purification step. In the one from Macherey-Nagel for example you can dilute your binding buffer to tune the size of fragments you want to get rid of (undiluted rescues almost everything, the farther you dilute it, the larger your washed away fragments get). But still with the original undiluted binding buffer you should get rid of your primers. The yield is much better for this method, if you follow every step in the most advanced fashion (i.e. pass your sample + binding buffer two times over the column, heat your elution buffer, also pass the elution buffer over your column twice etc..).

I changed to this, because i always had trouble getting a lot of DNA out of gel extraction. now my yields are over 100 ng/µl with the PCR cleanup. It also helps to do 35 PCR cycles in total (5 - 10 without primers, and then 30 or 25 with primers).

Another tip maybe, use the undiluted primer stocks, so you can minimize the volume you have to add to the PCR reaction. If you use the diluted primers, you have for example your reaction volume of 45 µl and add 2,5 µl of each primer (10µM) to get to your full 50 µl reaction. In turn that means, your polymerase has to work the first 5 cycles with 10% less volume than the buffer is meant to be, which means higher concentrations of salts and so on. So instead use 0,5 µl of each undiluted (100 µM) primer.

Good luck!

I haven't found a solution yet. I've tried carrying the reaction only with each band for 5 cycles and then adding the primers for the reminding 25. I tried using TAQ and Phusion under a TAQ thermocycling program (its the same as phusion's, but each step is doubled in time).

Since the combined band 400+2400=2800 is close to one of those, it's hard to know if it has been made or if its just the unreacted longer band. I've managed to find the extended band just above the longer one but in a very small amount.

Doing the reaction with the Phusion program (shorter in time, almost identical in temps) doesn't produce any results.

Tm = 56ºC, primers are 57ºC and 64ºC respectively.

Ok, there might be a few problems:

1) different fragment size:

Normally, the PCR works better, if the fragments are equal in size. Did you make sure, you adjust the elongation time? For the first cycles the limiting step is the length of the longer fragment, for the following 25 - 30 cycles you have to adjust it to the whole-length product.

2) Phusion needs a different temperature protocol as taq!

For Taq-based polymerases 5°C below the lowest primer Tm are suggested, while for Phusion polymerase (and also Q5 from NEB) 3°C ABOVE the lowest primer Tm are suggested. So you could try higher Tm (58 - 60°C) for the later cycles.

3) Try 10 cycles without primer and 25 with primer. In general, just to see if it is working overall, do 35 cycles in total. Of course there is a higher chance of introducing errors, but you first have to get your PCR product and sequence it. Afterwards you still can fine-tune your program.

4) For the first 10 cycles try 65°C Ta instead of 72, or do the two-step PCR at 68°C, just to be sure your fragments really anneal (they should, but if the 29 bp are very AT-rich, you are able to get Tm below 72°C). Also check with your primer design program (most of them are based on primer3 for calculations) the (theoretical) value for your 29 bp overlap region.

5) As a last option you could use 1 µl of the PCR where you could see your band and use that as a template in a normal PCR reaction with primers in it from the beginning on. Try to column-purify it, yields you really more than gel-purification.

Good luck!

Thanks Florian, I will try it out. I wasn't aware that phusion annealing temperatures must be 3ºC above the lowest primer Tm.

In 4) do you mean to lower the extension temperature to 65ºC for the 10 first cycles?

I just got my hands on phusion, I'll try it out as well and report.

EDIT: the GC content on the overlap is 62%

Thanks a lot!

Hi,

i made the experience that sometimes the annealing temperature is too high.. So if you are doing three steps (Denat-Anneal-elong) for the first cycles, you could try 65°C annealing and your normal 72°C for extension. Or you make a compromise and try a two-step cycle with 68°C for the annealing+extension step.

Also if you work with Phusion and high GC content, always try the GC-buffer parallel to the HF-buffer. And sometimes a pinch (1-5%) of DMSO works wonders ;)

Especially with the Phusion i always had a lot of tweaking to do with DMSO when having 72% GC content. Today i noticed that there is no pricing difference between Phusion and Q5 polymerase on the NEB homepage, so i would try to get my hands on Q5 polymerase! Also comes with a high GC enhancer (which probably is something like betaine or DMSO). Made it possible to amplify a 12,6 kb fragment first try and at least 7 kb with 72% GC content second try (after addition of GC enhancer it worked nicely). Q5 also has this "+3°C" rule, which is quite nice, since higher temperatures usually destabilize secondary structures.

Got carried away a bit.. I am not affiliated with NEB, i just had to do a lot of PCRs so far and Q5 always was my last resort and always did the job!

Good luck :)

It kinda worked. I don't know why it's so smeared.

Let me show you (picture): https://drive.google.com/open?id=0B7WtkMWfYxrpNU40SXg1RU5taGM

I did 10 cycles without  primers and then added them for the next 25. Ta was 60 because The lowest primer tm is 57ºC (I could use some DMSO and go for higher temps). The desired length is 2800 aprox. For comparision, the other bands in the gel are the overlapping segments that I use for the overlap extension, made with mutagenic primers and using TAQ.

Nice, looks good!

Smear is not uncommon in this kind of PCR technique, i wouldn´t worry too much about that.

I would say after a column purification you totally could try working with that.

If you want to tweak the method, maybe lower the temp of the 25 cycles to 57°C (still keeping the first 10 cycles at higher temp), since lower annealing temperature means higher yield. That´s always a trade-off between specificity and yield.

If you encounter errors in your final product, it may come from the usage of Taq-polymerase when producing the fragments itself. First, it has higher error rates compared to proof-reading polymerases and second, it adds an extra A to the 3'-end (if i remember correctly). There are mixtures of Taq together with proof-reading polymerases which have 3'->5' exonuclease activity to get rid of misincorporated bases and the extra A at the end.

I would give it a shot and furher work with it, but make sure to sequence the finished construct.

Good luck with the next steps!

Looks good, indeed! Since it seems a bit complicated to get, and since you used the standard Taq polymerase, I would TA clone... then you are safe, you can sequence different clones and possibly find one without unwanted point mutations (except yours) and reamplify easily from the plasmid with a proof reading polymerase...

Good luck!

I optimized further. Phusion with 5% DMSO and GC buffer seems to work best. 

There are two mutations in the gel. The first band is made with HF buffer + 5%dmso, and the second is GC + 5%dmso. Third and fourth are the same but with another mutation, seems to have worked equally bad for that one.

https://drive.google.com/open?id=0B7WtkMWfYxrpVG01VzMtbjVBWXc

Yeah, that´s most likely from the high DMSO content.. if you use GC buffer, maybe omit or at least reduce the DMSO.. 2% is mostly fine, but the less, the better..

Have you in parallel tried the cloning with the first successful PCR products? And most important, have you tried amplifying your starting fragments with Phusion or another proof-reading polymerase? Maybe the mutations already come from before the overlap extension..

I woult try lowering the DMSO content and maybe playing with higher temperatures instead. For each percent of DMSO you add, you normally must lower your annealing temp by 0.6°C, which in your case makes a difference of already 3°C.

Good luck!

Still haven't cloned yet, just wanted to have all mutations ready first and setting up the protocol for working with these primers. 5% DMSO made the best glowing bands, I don't want to lower it. If I rise annealing temps I would only get better specificity, not better glowing bands.