your target dna is now quite small so the quality of nested primers need not be as high as that of the original primers but nested primers where you move the primer sequences in by at least 10 bases from the original primers will give you lots of clean product. You can semi nest by having one new primer inside the original primer amplifying with the original other primer. Only run 20 cycles or fewer when re amplifying first round product and use 1ul of 1/100 to 1/1000 dilution of the first round product.  Simply re amplifying with the same primers will probably give you a smear of many sizes of product.

Yes. You can use the same set of primers (nested PCR) or change one by other that binds with amplicon's internal sequences   (seminested PCR). Did I answer you question? 

your target dna is now quite small so the quality of nested primers need not be as high as that of the original primers but nested primers where you move the primer sequences in by at least 10 bases from the original primers will give you lots of clean product. You can semi nest by having one new primer inside the original primer amplifying with the original other primer. Only run 20 cycles or fewer when re amplifying first round product and use 1ul of 1/100 to 1/1000 dilution of the first round product.  Simply re amplifying with the same primers will probably give you a smear of many sizes of product.

Thank you everyone for the opinions on nested PCR. Say I wish to use different primer to avoid smearing, is it going to make any difference if this different primer is assigned as forward or reverse primer? As additional information, I used GC-clamped forward primer which targeted D1/D2 region of 26S rDNA during the first PCR run. Thank you.

in general it does not matter which primer you change if you are hemi nesting but as one of your primers has a GC tail it will be long and expensive to make and also more expensive to purify when it was made so it will be cheaper to nest the shorter ( reverse) primer

I am particularly interested in the idea of semi-nesting PCR suggested here. As of my understanding, semi-nesting means using a new primer sequence to pair with one of the old primer (forward or reverse) rather than using entirely new primer set during re-amplification As mentioned earlier, I used GC-clamped forward primer (designated as gcNL1) and LS2 as reverse primer. Thus I need help to clarify the following:

1) If I were to use original gcNL1 to pair with a new reverse primer (designated as NL4) for re-amplification, does this imply 'semi-nested PCR'?

2) Regarding Paul's suggestion on distancing the new primer from the old one by 10 bp, does my new NL4's sequence sit sufficiently apart from LS2 given:

NL4: 5'-GGTCCGTGTTTCAAGACGG-3' and

LS2: 5’-ATTCCCAAACAACTCGACTC-3'

Thank you

without having the last 40 bases of template sequence it  is hard to say where these primers sit but there does not seem to be any common sequence between them so as long as NL4 lies completely inside ( nearer to gcnl1)  of the binding sequence for LS2 then it should be fine for semi nesting. The primer itself does not have to be excellent in terms of specificity because you have hugely enriched the target template with the first round of amplification

Thank you everyone. I shall attempt semi-nested PCR soon based on brilliant suggestions here.

as Paul said don't forget to dilute the first PCR products and reduce number of cycles 15-20

by the way for the DGGE analysis selection of right denaturing gradient is really important to get a clear band in your gel.

Alright, thanks for the reminder @Hossein Haghighi

Yes, you can use it, it worked for me.. maybe you can delute pcr product.

Absolutely, it is the sense of so called nested PCR.

I always Gel extract (in case of presence of non specific bands) or PCR purify the first PCR products, then I measure the concentration and take appropriate amount of purified product for second PCR. PCR purification will give you much better results