4 way cloning into TA Vector - (Jul/20/2013 )

I've got 3 inserts cloned into pGem-T vectors (one insert per clone) that have been sequenced and confirmed as the correct sequence for the next stage of my cloning. I have attempted the following but it did not work. Here is the situation.....

Each insert have restriction sites added during the PCR with 6 extra nucleotides at the 5' and 3' ends so that the RE sites are not at the very end of the inserts. These were then TA cloned into the pGem-T vector.

What I want to do is remove these inserts and ligate them all together to then sub clone them back into pGem-T to make sure I've got the correct sequence to then clone into an expression vector for protein sysnthesis.

Firstly, all inserts have RE sites such that that the ligation can only go one way and hence the sequence integrity will be maintained and should be the same as the sequence I've designed.

Secondly, I cannot just digest out the inserts from the pGem-T vector as the two end inserts (1 and 3) will lose their RE site recognition sequences and hence will not be able to be finally digested out of the cloning vector and ligated directly into the expression vector. I have the correct primers to PCR the inserts out of the vectors with NEB's Q5 HF polymerase master mix which will leave the correct RE sites in place.

Once I have my inserts free of the vectors, I A-tailed the two outside inserts (1 and 3) to provide the A's for the TA cloning. Following this, I digested the junctions between insert 1 and 2, and 2 and 3 leaving the 5' end of the first insert and the 3'end of the third insert intact and undigested. This theoretically should have left me with 3 inserts, properly digested and with the correct A-Tails in the correct locations for the ligation reaction.

I then worked out a 3:1 molar ratio for all inserts:vector and added them all into a 10ul ligation reaction using 1ul of pGem-T vector in 10x ligation buffer (promega). The ligation was left at 4degC o/n. Transformation was then carried out using my in-house competent cells with a T/E of ~5x10e6. These cell usually work ok. Not the best T/E in the world, but not too bad.

The transformed cells were then plated out onto LB/Amp/IPTG/Xgal and grown o/n at 37degC. I used 100ul for one plate, 200ul for another and 500ul for a final plate to try and maximise the chance of getting positive clones (white cols). However, I only got blue cols. Quite a few of them.

I followed the correct protocols for every step and performed PCR clean ups after inactivation of the digests to remove extraneous reactant components.

I'm wondering what I can do to improve this. Is it possible that I've not provided enough vector (I used 1ul of pGem-T) to take into account the inefficiency of the 4 way ligation? Or is this workflow too convoluted to successfully do a 4 way ligation?

The inserts in question are pretty small. They are all within the range of 115-250bp giving a final ligated insert size of 482bp. pGemT is 3000bp.

Any help on this would be appreciated.

It seems very non-optimal to be using TA cloning for this purpose. Why don't you use PCR to ampllify your vector, with whatever REs you want at the cloning site. Then, you can easily use REs to cut the inserts (and the vector). You can then just ligate to form your product. TA cloning is expensive and overkill for this purpose.

Thanks so much for your reply Phage434.

I hadn't thought about the fact that using TA cloning would be non-optimal. Your suggestion is pretty obvious now you've mentioned it! I was of the opinion that once the inserts had been prepped accordingly, then it's just another ligation. By non-optimal do you mean that TA cloning itself is not ideal compared to restriction site cloning in general or do you mean because of the single nucleotide overhang, TA cloning is not as viable as sticky end restriction sites with 4 nucleotide overhangs and hence the bottleneck in the ligation is actually the TA vector? Or is it all of that plus the cost implication?

I will definitely try your suggestion. Hopefully I can get it to work! That sorts out my workflow issue.....

In the meantime, I can prepare all my inserts by digestion out of the vector using the appropriate restriction enzymes while I await for my primers for the vector. My question therefore still remains as thus: Will I need to use a much higher amount of backbone as I had in my prevcious attempt in order to make more total DNA available to the ligation reaction in order to increase my potential yield of positive clones?

Once again, thanks for your help on this.

Ligation is optimal with equimolar amounts of each part being inserted. You should keep the total concentration of DNA low -- high concentrations favor building concatamers rather than recircularizing intramolecularly. About 10-20 ng of a vector in a ligation reaction is typically a good place, with equimolar amounts of inserts. You can calculate this number yourself -- you want the probability of finding the other end of the same molecule for ligation to be greater than the probability of finding the matching end of a different molecule. Know your DNA length, you can do this calculation. But it is not a critical number.

Mostly I was thinking of the extra steps you avoid in a single ligation step vs. ligation, PCR, TA cloning, but there is also the cost and design issue.

Perfect. Many thanks Phage434.

I was using 3:1 molar ratios of each insert to vector (3:3:3:1 for Ins1:Ins2:Ins3:Vector). Would the liklihood of getting positive clones be increased if I added a higher volume of the ligation reaction to my competent cells? In my first attempt, I added 2ul of ligation reaction to 50ul of comp cells. I may increase this to see what happens when adding 2ul, 3ul and 5ul in seperate trasformations after performing the digestion and ligations as you suggested.

Additional ligation added to transformation seldom solves problems, and often increases them. More than 5% ligation in a chemical transformation usually reduces transformation efficiency.
Almost all "ligation" problems are due to incorrect DNA amounts or quality. Common problems are UV exposure and failures to cut with the RE. I'd recommend avoiding "quick ligase" unless you have blunt ligation -- and then, I would redesign my strategy to avoid blunt ligation.

Once again, thanks very much.

I am using the normal ligase, not the quick ligase. Thanks for your input. I'll let you know how I get on :-)

OK, I've now inserted two restriction sites to two separate vectors.  pGemT and pUC19.  I've basically chopped out the MCS regions of both vectors and inserted the appropriate restriction sites to allow insertion of my 3 fragments comprising my insert.  The lacz region is intact as it should be.

Now I'm just transforming a 3:3:3:1 Insert1:Insert2:Insert3:Vector molar ratio ligation and transformation of each vector into my in-house competent cells.  I'm plating out 100ul, 200ul and 500ul of each transformant onto LB/Amp/IPTG/X-Gal plates and see what comes up.  Hopefully this ligation will have worked and I get some nice positive clones.

Success.  I had a large number of white colonies and only a very small number of blues colonies.  I picked 5 of each transformants and performed colony PCR on them and all look to be the correct size.

Now for sequencing.

Great. It's normal to have a large number of correct colonies, since the PCR amplification of the plasmid backbone makes the dsDNA fragment much more common than the circular fragment. If you cut the result with your two RE enzymes, along with DpnI, you can remove most of the template circular DNA as well, further reducing background. If you take the opportunity to change antibiotic resistance, so that your plasmid backbone has a different resistance than the backbone of the parts you are cloning, then almost all of the transformants will be correct. This is the strategy used in three-antibiotic Biobrick assembly.