Restriction-free cloning to exchange gene in vector backbone - (Jul/02/2014 )
Hello everyone 
Having sneaked around for a long time for answers for several of my lab-questions (thank you), I am now hoping for your help for my latest cloning problems.
I am currently trying to replace a puromycin gene in my vector backbone with GFP. The only unique RE-site near the puromycin gene is approx. 150bp in the gene. Therefore, I have not been able to use classical RE-cloning.
I therefore desided to use homolgous recombination (with approx. 16bp homology to the vector), hoping that it would not be a problem that there would be a long way from the site of homology to the end of the vector. The PCR amplified insert with vector-homology overhangs was succesfully created. However, recombination was unssuccessful despite several attempts and trying different conditions. I was using highly recombination effective E. coli.
I then tried to use my PCR amplified insert as a megaprimer based on the MEGAWHOP technique: http://barricklab.org/twiki/bin/view/Lab/MEGAWHOP
Please note, that my homology sequence is not as long as that suggested in the procedure. This did not work either, as I was not able to detect a PCR amplified product of the whole plasmid on an agarose gel. I tried doing the transformation any way, but had no colonies.
Not really knowing where to go from here, I have now a PCR running with my vector linearized in the puromycin gene and using my PCR amplified GFP with vector overhangs as primer, hoping to amplify a knicked vector. However, my hopes of this succeeding are very low.
Does anyone have any suggestions of where to go from here or anything I did not think about when designing my cloning setup(s)?
Kind regards, Emilie
16 bp is probably not enough to get efficient homologous recombination. I would suggest 50+ bp.
Your best bet might be to amplify the whole vector with primers incorporating any restriction sites you would like to use to insert the GFP. Basically the primers would be on either end of the puro gene facing outwards, so that the puro won't be amplified.
Bob1's approach is probably going to work fine for you assuming you don't have too huge of a plasmid or have other crucial, complex sequences in the middle that will muck around with your PCR (ex, repeats in homology arm regions of targeting vectors). Which vector are you using? If you can share the sequence, I may be able to lend a second set of eyes/ideas for going about it. I love a good (molecular) challenge 
Thank you so much for the replies! :)
The reason I used a 'short' homology sequence was because I based my cloning on this system by System Bio: http://www.systembio.com/molecular-tools/cold-fusion-cloning/technical-details
I know other people in my Department use this method successfully, so I borrowed the competent E.coli as well as the buffer from them...
The problem is I do have (what I think to be) a huge plasmid! ;) I am using the pTRIPZ vector (http://www.addgene.org/vector-database/5561/). I was already considering doing what bob1 is suggesting, as this is a strategy I have used for cloning previously, just not in such big vectors, which is why I chose the recombination approach initially.
What is your opinion? Should I opt for the RE-overhangs or the increased homology sequences?
I use the TaKaRa PrimeSTAR polymerase for my PCR. This should be able to amplify the vector without mistakes, though I am not really experienced in the amplification of these really large sequences.
Btw, if I deside to amplify the entire plasmid, how long should the primer sequence be that anneals to the vector?
There are some experts here: http://www.rf-cloning.org/fluxbb/index.php
Its a nice technique but often its not that easy especially with bigger plasmids or when plasmids contain many repetitive sequences.
EmilieKB on Thu Jul 3 08:33:20 2014 said:
Btw, if I deside to amplify the entire plasmid, how long should the primer sequence be that anneals to the vector?
EmilieKB on Thu Jul 3 08:33:20 2014 said:
Btw, if I deside to amplify the entire plasmid, how long should the primer sequence be that anneals to the vector?
EmilieKB on Thu Jul 3 08:33:20 2014 said:
Btw, if I deside to amplify the entire plasmid, how long should the primer sequence be that anneals to the vector?
So there are two regular cloning options I can think of. One is longer (2 cloning steps) but in the long run will probably give a superior result and be more useful. The other is quicker (only one cloning step) and will probably be as efficient as the original pTRIPZ vector but not up to my personal standards =P
Approach 1 (quick and dirty):
The simplest way to go would be to linearize your vector at the BsiWI, dephosphorylate it, and then PCR-amplified GFP flanked by 2 BsiWI sites, digest it and ligate into the vector (make sure to keep the GFP in frame with the initial Puro amino acids). Select for insertions in the correct orientation. The GFP can tolerate quite a bit in regards to fusion proteins and should be fine.
The downside to this for me personally is that I don’t like IRESs. I don’t like that the post-IRES RNA is expressed at lower levels than the pre-IRES. If you like IRESs then this would work for you. That said, there is a (slightly) longer method that avoids the IRES by inserting a 2A spacer at the end of the rtTA3 and fusing it to GFP.
Approach 2 (what I would do):
Cloning #1
Step 1: PCR amplify the rtTA3 fragment using the original vector as template using primer set 1 (below) to add two sites at the 3’ end and remove the stop codon (this is important).
Step 2: Digest the PCR fragment and vector with the XcmI and BsiWI, and gel purify and ligate together.
Cloning #2
Step 3: PCR amplify your GFP using primer set two which contain P2A spacer peptide sequence before the GFP sequence. It’s going to be a big oligo but definitely doable (I’ve done it before). The reverse primer will simply be a BsiWI site-stop codon-GFP reverse sequence.
Step 4: Digest the PCR fragment and vector obtained from step 2 with AscI + BsiWI then gel purify and ligate into the vector.
The P2A spacer is a better option than the IRES in this case because it leads to an equimolar amounts of the GFP and rtTA2 whereas post-IRES sequences are expressed at a lower level than pre-IRES sequences. The only caveat is that the P2A peptide will leave a short tail on your gene of interest so make sure that this doesn’t affect its activity. If it does, you can swap the positions rtTA3 and GFP in which case the GFP will have a short tail and your rtTA3 will have only an N’terminal glycine added. GFP can definitely handle the tail.
You could use KpnI instead of BsiWI if you wanted to get rid of the now-nonfunctioning Puro but doing this will also remove the WPRE. If you’re ok with that use KpnI. Otherwise stick with BsiWI
Primers: (Bolded = restriction sites, underlined = template-binding sequence, Italics = P2A spacer. Double check the sequences in case I'm not perfect)
Approach 1 Primers:
BsiWI_GFP_Fwd: ggCGTACGgATGGTGAGCAAGGGCGAG
BsiWI_GFP_Rev: ggCGTACG tcaCTTGTACAGCTCGTCCATGCC
Approach 2 Primers:
Primer Set 1:
XcmI_rtTA3_Fwd: ggggCCATGTCTAGGCTGGACAAGAG (58C)
BsiWI_AscI_rtTA3_Rev: ggCGTACG-GGCGCGCCCCCGGGGAGCATGTCAAGGTC
Primer Set 2:
AscI_P2A_GFP_Fwd
ggGGCGCGCC-aGGAAGCGGAGCTACTAACTTCAGCCTGCTGAAGCAGGCTGGAGACGTGGAGGAGAACCCTGGACCT-ATGGTGAGCAAGGGCGAG
BsiWI_GFP_Rev: ggCGTACG tcaCTTGTACAGCTCGTCCATGCC
Thank you so much for all your replies! :)
I have gone ahead and ordered two sets of primers.
First, I used the link given by pito, and will try to do the RE-free cloning. It's a cool technique and would love to see it work.
I then also ordered primers for the first method suggested by Bio-Lad. Though I do see your point on the IRES , I am not really worried about that, as I know that my system is working perfectly despite it expressing puromycin instead of GFP :) But I will definitely remember this for future cloning projects! :)
Now I am just waiting for my primers to arrive so I can get on with this!
Awesome. Best of luck. Let us know how it works out =)