If the sequences really are a perfect match, then there really isn't much you can do in terms of adjusting the PCR conditions.  You could try to increase the extension phase of the PCR reaction to try to improve the efficiency of synthesis of the "longer" reaction product, but the shorter product  will always be the majority of what is produced.  Otherwise, you need to try to redesign the primers to take advantage of any differences that exist between the two similar regions--focus on designing a primer that has a nucleotide mismatch near the 3' end with the undesirable priming site.

If the sequences really are a perfect match, then there really isn't much you can do in terms of adjusting the PCR conditions.  You could try to increase the extension phase of the PCR reaction to try to improve the efficiency of synthesis of the "longer" reaction product, but the shorter product  will always be the majority of what is produced.  Otherwise, you need to try to redesign the primers to take advantage of any differences that exist between the two similar regions--focus on designing a primer that has a nucleotide mismatch near the 3' end with the undesirable priming site.

I believe there are a few options, but to decide which one would be the best choice it is necessary to know what is this product for - cloning? direct sequencing? ..?

First, I would try to lower a stringency of the reaction and, as David Barker already suggested, increase the extension time.

Dear Miriam

Optimization of PCR condition and mastermix components is the first and foremost thing we need to perform for any PCR analyses with a newly designed primer. As per your problem I could see you are getting a non-specific amplification. First check the primer’s concentration and decrease the concentration, if necessary. Followed by you can go for the optimization of MgCl2 concentration, annealing temperature, elongation time (as suggested by Prof. David) and number of cycles to prevent nonspecific priming.

All the best

I agree with David and Andrei. If, however, the sequence match is incomplete for the internal annealing site (smaller PCR product), try decreasing the concentration of your forward primer to 0.1uM and increase the annealing temp to improve specificity. In the case of a perfect match, as David said, there's not much that can be done.

Dear Miriam,

In consensus with the suggestions above, the only concern I have is "no - detection" of the larger fragment. If at all, the primer binds at two sites, shouldn't you be getting both the fragments instead of one? I believe a little troubleshooting on the primer sequence homology with the two putative sites might present you with a solution. 

Miriam - If you are sequencing or cloning your PCR-products, you may be able to live with the production of multiple bands from your PCR. Even, if you are absolutely not able to change your primers and all optimisations - as suggested by others - do not improve your results. If sequencing, you may be able to sequence your products, using specific internal sequencing-primers. You may lose efficiency of production of the target you want, i.e., by producing also non-target PCR-products. But, we have been able to obtain good sequences by direct sequencing PCR-products with multiple bands - sometimes more than 2 bands. If you are cloning your PCR-products, you may need to use some form of screening of your libraries for the target products, e.g., hybridisation probes or sequencing. Good luck.

I am not sur to understand. If your reverse primer binds at only one site, you should have only one product of the desired length (except if you forward primer binds at two sites in the studied sequence, is that the case ?). Just extend your forward primer to gain in specificity. Then adjust PCR conditions for this extended primer. If unpossible, I would what Alexei said.

Hi Miriam,

Sorry this answer is so late, but if you are still struggling with the amplification then I have one suggestion, which depends upon the nature of what you are trying to amplify.  As others have said there is little you could do if the primers are a perfect match, so you could try an old fashioned nested PCR.  Design primers downstream and upstream of your intended target and amplify a product that has only the primer site you are interested in, then with enough dilution this could be used as template for your PCR removing the unwanted binding site of the forward primer from the equation.  This will involve 2 PCRs which increases the potential error rate if you are cloning the product, but as long as the amplicon is not too long and you use a proofreading polymerase you should be OK.