The answer of Rakesh is correct concerning the nucleotide lying at the 3' en of the primer, which will be poorly or not extended anyway if it doesn't match the sequence of the template. However, there is no problem in using a proofreading polymerase for extending a primer that contains an internal mismatch. Indeed, it is even recommended to use a proofreading polymerase, since its high fidelity will avoid the creation of spurious mutations. The only precaution required is to lower the hybridization temperature, at least for the first cycles, since the mismatch will lower the melting point of the hybridized oligo.

http://www.sciencedirect.com/science/article/pii/S0027510702002518

This process of proofreading of the polymerase detects that a wrong (incorrectly paired) nucleotide has been added, it will remove and replace with the correct one. but its not good idea to extend degenerate primers using a proofreading polymerase. 

The answer of Rakesh is correct concerning the nucleotide lying at the 3' en of the primer, which will be poorly or not extended anyway if it doesn't match the sequence of the template. However, there is no problem in using a proofreading polymerase for extending a primer that contains an internal mismatch. Indeed, it is even recommended to use a proofreading polymerase, since its high fidelity will avoid the creation of spurious mutations. The only precaution required is to lower the hybridization temperature, at least for the first cycles, since the mismatch will lower the melting point of the hybridized oligo.

well, I think we can do it