Site-Directed Mutagenesis (megaprimer) of internal sequence - (Jan/25/2007 )
I have been trying to insert a 2 base pair mutation into a gene for several months now. I have the target gene inserted into pUC19 plasmid. Since the sites I am interested in mutating are internal in the gene sequence I cannot use a standard site-directed mutagenesis protocol because I want to amplify the entire gene with the mutation--not just from the mutated sequence onward. I have also tried some megaprimer protocols. I can make megaprimer but I cannot obtain full length gene sequence during the second step of megaprimer mutagenesis. I think that after denaturation the megaprimer may be re-annealing back to itself. Does anyone have a suggestion as how to perform this second step of megaprimer mutagenesis? or else is their another reliable method of site directed mutageneis that might work in this situation?
The easiest approach is probably the use of the Quikchange mutagenesis kit from Stratagene (or the protocol, with your own Pfu and DpnI).
If you want to do the megaprimer approach, you can make single stranded product of each megaprimer, mix them together, and then PCR using outer primers. To make the ssDNA, set up a PCR reaction with small amounts of the left PCR product and the left outer primer in one tube. Setup another PCR reaction with small amounts of the right PCR product and right outer primer in a second tube. Cycle these tubes for 5-10 cycles, producing ssDNA from the single primer. Stop the reaction, and mix the products. Continue cycling for another 20-25 cycles to produce a final product (which will amplify using both the left and right outer primers).
I'm assuming this "megaprimer" approach is another way of saying splice-overlap PCR. If so, I have a few questions. Firstly, are you attempting to insert 2 bases or mutate 2 bases? Secondly, how many bases are in the overlap and what is the Tm of the overlap? And thirdly, how many bases are on either side of the mutation you are interested in creating? I use this technique quite often and may be able to provide some advice with these answers.
Thanks for posting. I am attempting to mutate 2 bases. This to my knowledge is not the same as splice-overlap PCR. In the first step I use a flanking primer and a mutagenic primer such as attcggctgctxxacttgat so that I am provided with a "megaprimer" that has the desired mutation at the far end, in the second step I am trying to amplify the entire gene sequence, which is to say everything that is upstream of the mutation along with the part of the sequence already amplified in the first round of PCR. This is where I am running into problems. I trust this clarifies my previous post. Thanks.
I use a megaprimer protocol to insert a base pair mutation in a gene.
I used a protocol based on “Rapid an efficient Site-directed Mutagenesis by the single-tube Megaprimer PCR Method” described in Molecular Biology by Sambrook (The Bible! 
).
How much primers are you using??
I used 10 pmoles (or less) of mutagenic primer (internal)
And 100 pmoles of flanking primers (with a low Tm).
If you are using a polimerase who add an A like Taq pol, be sure the internal primer is located just next to a T. I don’t know if I was clear. 
What is your PCR program?
Thanks for posting. I am attempting to mutate 2 bases. This to my knowledge is not the same as splice-overlap PCR. In the first step I use a flanking primer and a mutagenic primer such as attcggctgctxxacttgat so that I am provided with a "megaprimer" that has the desired mutation at the far end, in the second step I am trying to amplify the entire gene sequence, which is to say everything that is upstream of the mutation along with the part of the sequence already amplified in the first round of PCR. This is where I am running into problems. I trust this clarifies my previous post. Thanks.
Oh, I see. Splice-overlap PCR works way better than this approach. In some ways SOE could be considered a double megaprimer strategy, so you're essentially doubling your chances if I can say that. Order another primer and perform the SOE, I think you'll find much friendlier results.