My first concern about your protocol is your long primers. In practice long primers have an increased tendency to hybridize to a template despite mismatches making them, in fact, less specific. Generally, best specificity is achieved with primers of 18-25 bases with rich GC content at the 5'-prime end and low at the 3'prime end.

If you are concerned about wasting your sample you could try lower sample amount in your reaction. 20-50 nmol should be enough. Keep in mind that your amplicon is long, you have 10x more DNA in your 6kb band compared to an equimolar amount of a 600 bp PCR product. If you decrease your sample amount you may also need to decrease primer amount to maintain your primer/sample ratio (assuming it is ideal in your current protocol).

Try also to keep the number of cycles moderate. 40 cycles may be too much, in later cycles non-specific binding of one of your long primers to your PCR product could increse, perhaps explaining the 5kb band.

You are already using a proofreading polymerase so there may be no need to try another brand. Phusion Hot Start II is highly processive and works with shorter extension times (3-4 min should be ok) so using that you could shorten your protocol. It works well for long templates.

Pl use high fidelity taq or long acting taq

It would be easier to pinpoint possible reasons for your problem if you gave more details on what amplification conditions you have tried, but facing the same situation I'll list a few parameters I would look into. As for polymerase I strongly recommend a hot start enzyme with proofreading activity and high processivity (such as Phusion Hot Start II High-Fidelity Polymerase). For reducing non-specific amplification you may decrease primer concentration, use less polymerase or use shorter annealing times and consider template amount. Touchdown-PCR is also helpful in many cases. Nonspecific bands and wrong side bands may sometimes be controlled by changing extension times (shorter time to avoid long bands and vice versa).

Thanks very much Meeta and Peter for your reply.

Regarding PCR conditions:

PCR reaction using ExTaq Hot start: I use 100 ng of gDNA template, primer concentration: 0.2 uM, dNTPs 200 uM, Mg+ 2 mM, 50 ul total volume. Cycling 98 for 1 min , 40X (98 C for 10sec, 59-70 C for 30 sec, 72 for 6.5 min), 72 for 10 min, primers are 35 bp (to make them specific). The band should be 6 Kb; I get two sharp 6 Kb with 5 Kb bands of similar intensity. I tried two step cycles 35 X (98 for 10 sec, 68 C for 5 min) I got same result.

What about performance of Phusion Hot Start II for Long PCR like 6 Kb?

As I mentioned, My sample is tiny, this is why I am doing maximum cycling numbers, and being careful about not to make a lot of optimization reactions

My first concern about your protocol is your long primers. In practice long primers have an increased tendency to hybridize to a template despite mismatches making them, in fact, less specific. Generally, best specificity is achieved with primers of 18-25 bases with rich GC content at the 5'-prime end and low at the 3'prime end.

If you are concerned about wasting your sample you could try lower sample amount in your reaction. 20-50 nmol should be enough. Keep in mind that your amplicon is long, you have 10x more DNA in your 6kb band compared to an equimolar amount of a 600 bp PCR product. If you decrease your sample amount you may also need to decrease primer amount to maintain your primer/sample ratio (assuming it is ideal in your current protocol).

Try also to keep the number of cycles moderate. 40 cycles may be too much, in later cycles non-specific binding of one of your long primers to your PCR product could increse, perhaps explaining the 5kb band.

You are already using a proofreading polymerase so there may be no need to try another brand. Phusion Hot Start II is highly processive and works with shorter extension times (3-4 min should be ok) so using that you could shorten your protocol. It works well for long templates.

Mohamed, I suggest using some other template(s) for your reaction optimizations and then go for your samples after the conditions are optimized. However, you should also be concerned considering the GC content of your samples. As viral sequences may contain high GC that may make your reactions hard to be optimized especially for long run.

Thanks Peter & Kemal,

I will try again to optimize my protocol considering your advice. I will try shorter primers also.

I will give feed back about my results.

Phusion High-Fidelity PCR Master Mix (ThermoScientific) or KOD Hot Start (Novagen) work perfect.

If sensitivity is the problem,perhaps nested or seminested PCR may help to some extent. Try to design the third (forth) internal primer and run two rounds of PCR: The first reaction using external primers and genomic DNA. In the second reaction, you amplify a small 0.5-1.0 ul aliquot from the first reaction as a template using internal primers. Keep annealing temperature as high as possible. Use enzymes of reliable proveniance, such as those from Roche (Expand system).

Try to reproduce the advices of Peter & Kemal, i think that they give you the optimal conditions, Good work

for your PCR where 6 kb is amplicon size high fidelity taq or long acting taq like pfu polymerase is needed with optimal condition

All enzymes listed work great. I suggest using Illustra Ready-To-Go GenomiPhi DNA Amplification Kit to amplify your template prior to PCR.

Thanks José, This is very nice idea to amplify my limited DNA sample to properly optimize my conditions.

I have also had similar problems in the past with other organisms and with sample running out. I found two solutions for different genes.

One solution was using the Advantage polymerase mix from Clontech. It has very good specificity, efficiency and good for long-range. That would also work with your long primers as it requires primers with high Tm to do the annealing and extension at 68 degC.

The other solution was to use Hot-Start with a 15 minutes initial denaturation (I used the old QIAGEN HotStarTaq - not the Plus) and 1 ul of DMSO in the 50 ul reaction. The overcame the difficulties with my most difficult template when I only got smears when I used any other enzyme/protoco, not even the above.

Also, I agree with Peter, I also found that increasing cycles to 40 sometimes made things worse and it was better to do the optimization of other things while sticking with 35.

I hope this helps a bit, good luck!

The best polymerase is Taq and pfu polymerase

In my lab we use the "Thermo Scientific Phire Plant Direct PCR Kit" for fungal DNA and we have had great results. The kit makes use of Thermo Scientific Phire Hot Start II DNA Polymerase, which is incredibly efficient and has thus far been very reliable. The nice thing about the kit is it provides the buffer with the dNTPs added etc. so no additional work need be done.

I want to thank Michelle Buck (what a funny surname) her advice on the Thermo Scientific Phire Plant Direct PCR Kit. I work on animals, the lab besides mine had the enzyme and it has just solved me out a problem. Thank you very much, I am getting old and still in the lab. I am very grateful indeed.

Dear Mohamed,

I had some amplifications problems with herbarium samples, and the best solution was use the new Go Taq G2 Hot Start Polymerase from Promega. It cost 5x more than the usual Hot Start Taq, but the amplification sucsess was much better than the expected.

Hi, I think that you already have many good answers, but I tell you my experience with different polymerases and difficult PCRs.

For your purpose I think that you need a Hi Fidelity polymerase:

A Pfu Turbo is a really good option, I had good results with it, but last year I tried the Phusion HotStart II and it is incredible! I was trying to amplify a dificult fragment to 14kb, and whereas the Pfu amplified a weak band, with Phusion I had obtained a strong band. I repeated the amplification many times and always had best results with Phusion.

Last week I saw a "new" polymerase from NEB (Q5) that look so good but I didn't try it.

Another things:

- When I have problems with extra bands, I always use DMSO to improve the reaction (final concentration al 5%).

- Even when I need to much amplification product, I prepare the mix that I consider necessary, I take some for the negative control. After that I add the DNA in the mix and them I divide the mix in tubes with no more than 50ul each.

I hope this information will be usefull, and you can amplify and sequence your precious sample!!

Phusion worked well for me. It was successful to amplify from high GC template while many others failed. Since then, I switched to Phusion and it's been working well and time saving too. It is recommended to have 15 second polymerizing instead of 1 minutes per kb.

Another option would be use a whole-genome amplification kit so that you have more sample. http://www.qiagen.com/knowledge-and-support/spotlight/wga/. I would also try nested PCR (in addition to using one of enzymes others suggest). This will vastly increase sensitivity. The increase in sensitivity means that you'll need to be careful about contamination.

Your amplicons are to long, 200-400 bp amplicons sizes works most efficient.

We have good results with PicoMaxx High Fidelity PCR System from Agilent Technologies. Hope it will work for you. Best Nives

I also agree with Martin. Unless your DNA is extremely high quality, you will not be able to amplify this length. Shorter lengths with a nested PCR will help. Using a long annealing and extension time in the PCR cycle can also help.

Mohamed, try QIAGEN multiplex kit. To obtain the maximum specificity use 2 Pmol of your primers and proceed with the touch down PCR with 15 minutes initial denaturation, 90 sec annealing time for each cycle and 60 minutes final extension. Or choose the best protocol from the manual for your case.

Hi, try Q5 (New England Biolabs). It's designed to amplify long (

I like KOD polymerase. It is very robust and with really good fidelity. Do you really need to amplify 3-6kb? That is big. Your life would be much easier if you could get away with

In order to save your material why can you use genome wide amplification kit . You can increase your template DNA using small quantity of DNA.

I like Phusion High-Fidelity DNA Polymerase (http://www.thermoscientificbio.com/pcr-enzymes-master-mixes-and-reagents/phusion-high-fidelity-dna-polymerase/), it worked for me perfectly, when Taq and AmpliTaq Gold® 360 DNA Polymerase where not working or bands were very weak and Phusion works with much less amount of template than Taq's :)

I too agree with Asta Stapulionyte, I`m using Phusion High-Fidelity DNA Polymerase for most of my PCR standardization and it is working perfectly.

Generally speaking, there are 3 factors to consider when selecting an enzyme: fidelity, processivity, and elongation rate.

ExTaq and LATaq appear to be taq-based recombinants given a proofreading domain (from pfu would be my guess) to increase fidelity, but it's not much of an improvement (5xtaq) compared compared to pfu- and kod-based recombinants (typically 50xtaq or more). While fidelity is always an issue insofar as we all want our PCR products to be as free of misincorporated nucleotides as possible, it is not a reason for reaction failure.

A high elongation rate is nice in that it shortens your reaction time, but again I don't see an inferior rate as a reason for reaction failure. It's compound quantity, anyway: the elongation rate is a product of enzyme processivity and nucleotide incorporation rate. So let's look at the more fundamental parameter: processivity. For such a long template, it is a serious consideration, as the longer a template is, the more likely the enzyme is to "fall-off" before finishing its job.

It can be tough to compare published rates, because of inconsistencies in methodology between manufacturers (they will always choose the method that makes their product look best); if you're looking for a comparison, check out the article that I linked; it goes into more detail than I have on fidelity, processivity, etc, and/or look for comparisons performed by independent labs.

Personally, I use Thermo's Phusion Hi-Fidelity Hot-Start II enzyme, which has ~50xTaq fidelity (thanks to pyrococcus furiosus's proofreading domain) and high processivity (thanks to sulfolobus solfataricus's DNA-binding domain). NEB's Q5 is a comparable enzyme, but I chose Phusion for the price (my institution has a supplier agreement with Thermofisher).

Hope that helps!

http://www.biocompare.com/Editorial-Articles/148146-Picking-a-PCR-Polymerase-No-Problem/

If the purpose is just sequencing your template, it is not neccesary to use a high fidelity enzyme. Normal Taq polymerase with a hotstart procedure gives the highest amount of PCR product. 

The size of your amplicons are to long for efficient amplification, 400bp amplicons gives the best performance. Sequencing reads are not longer then about 800bp, so you need to order multiple primers anyway.

To increase the amount of template you could use a whole genome amplication kit (WGA).

It would also be worth using a nested PCR.   This can vastly increase the sensitivity of a PCR reaction (in my experience more than using different polymerases), although for this reason will also be more prone to contamiation.