This month's question from the Molecular Biology Forums (online at molecularbiology. forums.biotechniques.com) comes from the “DNA and General PCR Methods” section. Entries have been edited for concision and clarity. Mentions of specific products and manufacturers have been retained from the original posts, but do not represent endorsements by, or the opinions of, BioTechniques.
Molecular Biology Techniques Q&A How can I calculate the specificity of my allele-specific PCR primers? (Thread 23566)Q Is there a method available for calculating the specificity of allele-specific PCR primers? I am thinking of testing oligonucleotides in 1000:1, 100:1, and 10:1 wild-type to mutant mixtures of DNA. From these mixtures, can I determine the minimum concentration of mutant sequence detectable by my PCR? Will this be an appropriate test, or is there a better option?
A If the primers are specific, they should fail to PCR amplify the mutant template regardless of the concentration in the mix. I think that good positive and negative controls would more appropriately tell you what you need to know.
Q I understand the importance of proper primer design, but I work with viruses and their sequences are very low in concentration. I need to obtain the viral product from my PCR. So to be practical, I can't use very stringent conditions for my PCR assays. But I still want specificity. What I need to know is if there is a method for measuring the level of specificity for an allele-specific PCR.
A The primers can be specific enough that they will not generate a product from the wrong template during PCR and still work at 100% efficiency. Specificity and efficiency are not contradictory, but you may have to optimize the assay to get the correct annealing temperatures for maximizing efficiency and specificity. If a primer set does not generate a PCR product in the presence of 100% allele A template, it will also fail when there is 99% of allele A present.
You can amplify the same template in reactions with two different primer sets. Try making a serial dilution of known amounts of each template and running your PCR to make a standard curve. This is usually done using qPCR.
A You will probably always have background amplification in your no-mutant template sample due to the abundance of wild-type sequence present. You will need to set up a standard curve for your experiment. I recommend that you look into Allele-specific Competitive Blocker PCR (ACB-PCR). There are some papers by B.L. Parsons that are informative and will help you get started. With ACB-PCR, you can measure as few as 5–10 copies of mutant sequences from mixed DNA populations.
Q I am trying to do multiplex allele-specific PCR to genotype several genes. So far, I am getting bands for the dominant and variant alleles with all templates. It looks like my primers are not discriminating between the two different alleles, so I need to increase the stringency of the PCR reactions. Can I do this with the same changes as used for increasing stringency in regular PCR reactions?
A At the risk of stating the obvious, you must use a non-proofreading polymerase such as Taq.
A The 3′ end of an allele-specific primer is mismatched for one allele and matched for the other allele. This can possibly reduce annealing to the other allele, but more importantly it will prevent Taq DNA polymerase from extending the mismatched primer because Taq needs a perfect match at the 3′ end or it can't extend. Since Pfu is a proofreading polymerase, when it binds to an incompletely matched primer-template pair, it cuts out the mismatched nucleotides at the 3′ end and starts polymerizing a new strand. If you are using Pfu, you will loose all specificity in your reactions.
A For troubleshooting reactions, I've mostly optimized the annealing temperature of the primers and concentration of magnesium using a hot start approach also helps. Lower dNTP concentrations of 100 µm each have been reported to improve specificity. I successfully run my reactions with 50 µm of each dNTP.
When designing your primers, the allele-specific nucleotide should ideally be the last, or at least one of the last two nucleotides at the 3′ end of the primer. If you still have specificity problems after making these changes, you might want to try primers with an additional intentional mismatch. In this way, a correct primer-template annealing pair will only be mismatched in one position, while the incorrect annealing pair will have two mismatches. This will impair the efficiency of the reaction, so it will only be useful in endpoint PCR and not qPCR.
A The annealing temperature is the main parameter I optimize. Co-solvents such as DMSO, formamide, or betaine can help improve specificity sometimes, but at other times, they make no difference and can even lower efficiency.
It could be difficult to optimize all of the primer sets for specificity in a multiplex PCR if you have only 1 or 2 mismatches between alleles. You might want to try making alternate versions of the primers that are a nucleotide longer or shorter and using the longer one if you need to increase PCR efficiency or the shorter one to increase stringency. These should be used with the same conditions that work well with the other primers.
Another option I've read about is using modified nucleotide bases at the mismatch area of the primer and several nucleotides around it. This will change the stacking of the hybridized bases and increase the difference in Tm between matched and mismatched primers. I think that locked nucleic acids are one option.
If by multiplex PCR you meant two alleles of only one gene at a time, your solution is easier than if you have multiple genes in one reaction. When multiplexing two alleles of one gene, I test various ratios of the two primer sets to find proportions where both amplifications work well with heterozygous templates.
