Inverted repeat: three-way ligation or fusion PCR? - (May/06/2011 )

Hi!
I'm hoping someone can help me out with this. I've read through a number of threads now, but I think my case is quite unusual, so I decided to start a new one.
I'm trying to make an inverted repeat sequence by either fusion PCR or by cloning it into a vector, but haven't been successful either way.
So, I have a sequence "ABCD" of 1.5 kb and I want a sequence "DC" (an inverted repeat of the last ~300bp of "ABCD") to follow that, with a small spacer in between, so I end up with "ABCD-DC". I've amplified the individual fragments with primers that overlap the junction by 20bp, so that I can fuse the fragments by PCR by using only the outside primers. However, the PCR reaction never worked, although I tried a few different condition/enzymes. I actually didn't expect this to work, because 300 bp worth of inverted repeat should anneal to each other and prevent the reaction from working. The fact that this sequence is difficult to replicate is exactly the reason why I want to generate it, so no news here
My next try was to ligate the two PCR fragments in a three-way ligation. I have a non-palindromic restriction site, BseYI, in my spacer for that purpose. My idea is to cut the individual PCR fragments "ABCD-" and "-DC" with BseYI, so that they have a single A overhang on one side and a 4-nucleotide-sticky-end BseYI overhang on the other side and then do the three-way ligation into pGEMT. this way, theoretically, one side of each fragment ligates to pGEMT and the fragments ligate to each other on the BseYI overhang. Goes without saying that that didn't work either, otherwise I wouldn't be writing this...
so my questions are:
1) Is there still hope for my PCR fusion approach? do you know of anything that might make a reaction like that work?
2) Is there anything in my three-way ligation I'm not considering? Should I just keep trying until I get what I want? (I have to say I only screened 30 colonies, which was everything I got the first time I tried it)
3) Should I use mutant bacteria like SURE to clone this?

Please, any help is greatly appreciated!!

Maybe that one nucleotide (A/T) in combination with 3-way ligation was too much for the reaction to work. Can you clone your two fragments into pGem-T using different enzymes, eg SalI and SphI? So you can make your framents as SalI-ABCD-BseYI and BseYI-DC-SphI and digest pGEM-T with SalI and SphI.
Or if you need to clone ABCD-DC with BseYI, you could use different spacer sequence and add BseYI sites at the ends.

kajmak on Mon May 9 14:34:25 2011 said:

Hi Kajmak!
thanks for replying. I thought about doing that, but I'd have to order new primers for that, coz I don't have RE sites in them. Wouldn't be the end of the world, I guess. Do you think 3-way ligations don't work with A/T vectors? I guess a longer sticky-end would be better, but aren't people doing 3-way ligations with blunt ends? that'd be worse than A/T, right?
Do you know anything about SURE or STBL bacteria strains? I read that they are mutated in some recombination genes and some other DNA repair pathways to prevent them from recombining my inverted repeat, but we don't have them, so I don't want to order that if it's useless. I'm thinking that if my inverted repeat is difficult to replicate and you take out the pathway that the bacteria have to deal with that, they wouldn't grow at all...not sure if that's right...

Let me know your thoughts
Thanks

Honestly, blunt end ligation is something I run away from, and just create additional site.
I don't know about SURE and STBL strains. One of the students in our lab was making inverted repeat and he used XL.1 blue. But he had ABCD-DCBA.