Unique cohesive overhangs cloning + cohesive/blunt ends cloning - molar ratio / use of PEG in cohesive-blunt ligation (Oct/29/2008 )

Hi everybody,

I am trying to clone a 2.1 PCRed cDNA in pMIG vector. I amplify using Expand High Fidelity Plus from Roche and PCR is completly specific and works well with enough cycles (the mRNA is a low level one).

I am trying two different approaches:

1.- 5' primer with BamHI/3' primer with BamHI too. Cutting the vector with BglII

2.- 5' primer with BamHI/3' primer with HpaI. Cutting the vector with BglII/HpaI

I am having troubles with cloning this cDNA and my main concerns are:

- When doing the approach number 1, what is your advice about molar ratio (vector:insert) during ligation? I am going to dephosphorilate the vector in the way tfitzwater commented in some post time ago (outstanding post, by the way): adjusting CIP to pmol of termini, purifying (qiagen in my case), and CIP again, then gel for band excision. Anyway I would like to know what molar ratio you consider the best as Bam ends in the insert will be phophorilated. I have read some opinions claiming a great insert excess is the best while others say just the opposite, this is, the lowest amount of insert the less posibilities of having it self-ligated.

- When doing the approach number 2, it is a good idea to use PEG for the blunt end? I mean, I know it is adviceable to use it when both ends are blunt, but this is a cohesive/blunt ligation, maybe PEG could interfere with the cohesive end ligation.

Thank you in advance.

(Sorry for my English)

Approach 2 is probably better because you are ligating into two different sites, which reduces the number of background colonies. A different overhang to BamHI (rather than a blunt end) would be even more preferable.

With regards to molar ratio, my feeling is the more insert the better, within reason. Concatamers (self-ligated inserts) are not that common. I also think that if you are struggling with your ligation, use more vector and more insert at the same time. More total DNA increases the chance of the vector and insert reacting with one another.

You can use PEG for the blunt end, it could help but it's not an absolute requirement. There is already some PEG in your ligation buffer. I've never heard of PEG inhibiting a ligation. It increases molecular crowding of all the reactants in your ligation, by taking up space in the reaction, which brings the reactants closer together and increases the chance they will react with one another. So i don't see how it could do any harm.

The secret to ligation is molecular crowding - more vector, more insert, more units of ligase (concentrated ligase can be purchased), smaller total volume. The more reactants are crowded together, the more likely they will react with one another.

Good luck,
Rob

Thanks for your reply killerkoz17, my opinion is just like yours, but anyway I am trying right now one condition with 1:1 molar ratio and other with 1:6, for approach 1.

Just one question, what would you consider a good concentration of molecules in a ligation? I have always done 30 ng of a 5-8 kb vector with 1:3 molar ratio in 10 µl of reaction and it has always worked perfect (always working with two different cohesive ends, which is not possible in this case).

I was thinking about doing it again the same way: 30 ng vector 1:1 or 1:6 in 10 µl. What is your advice for this issue?

Thanks again.

I disagree with this. There is competition between two competing reactions: ligation of multiple fragments into long chains, and circularization to form plasmids. Only the (single origin) circular plasmid form transforms. The concatamers are undesirable side reactions, which simply waste DNA and reagents. If you calculate the local concentration of the desired target end for recircularization, then it is quite low. You want the concentration of competing ends on other linear fragments to be below that concentration to favor circularization. On the other hand, you want lots of DNA to increase your chances of producing product. The sweet spot between these two issues is around 5-20 ng/ul of plasmid in your ligation reaction. More actually reduces the amount of circular plasmid produced.

I would recommend against blunt cloning, and would design a strategy to use two enzymes with incompatible overhangs.

PEG is not present in most ligation buffers, but only in the "quick ligation" buffers. I avoid those -- they cause more trouble in my hands.

Thanks Phage 434,

As I mentioned before I have tried both considerations with ligation ratios 1:1 and 1:6. I have seeded the DH5alpha tranformations right this afternoon so tomorrow I will have results of number of colonies in the vector-inserts versus the vector alone controls and I will have minipreps-restriction results on monday. I will let you know what condition worked the best. I hope at least one of them will work fine!

I cannot use an strategy of double-non compatible cohesive ends. This pMIG vector has a MCS of only 4 RE and all but two of them (considering also RE with compatible ends) cut inside my cDNA. One of them is BglII (cohesive, compatible with BamHI) and the other one is HpaI (blunt ends).

Thanks both of you guys,

I have now the colonies results from my last experiment (1:1 and 1:6 molucular ratios with 3 ng/µl vector). No good results at all: The control (only vector) ligations developed a huge number of colonies and plates of insert-vector ligations showed the same amount. Considering I double-desphosphorilated the vector (as I mentioned before), this makes me think the vector is able to recircularizate withouth phosphorilation due to it has been cut with only one RE (BglII). What is your point? I know it doesn't sound too orthodox but I cannot find another explanation: I think it is unlikely that after the two rounds of dephosphorilations the vector still harbours 5'-PO4

I plan two new approaches:

- Ligation with a very low concentration of vector and a great excess of insert (despite phage434 advice) with a first initial step of denaturation at 60 ºC and quick cool down to 0ºC right before adding the ligase.

- Trying again the cohesive-blunt end approach with two conditions: with and without PEG.

I finally achieved it.

Final Results, hope this can be useful for someone in the same situation:

- Approach 1: no good results. Anytime I tried to do it, control plates (only vector ligations) presented a lot of colonies. Complete ligation plates (vector + insert) displayed equal or even less colonies. I did not tried to check those colonies everytime but I have checked some of the conditions and I only obtained 1 positive construction over 24 minipreps. In my opinion, trying to clone a middle-large size insert in a vector with only one RE is not a worthy approach.

- Approach 2: good results with only some of the conditions.
Ligations with 1:12 molar ratio displayed a number of colonies higher than those from only vector ones. When miniprep/restriction checked, they show positive construction, specially if PEG was included in the reaction, where 8 out of 10 colonies were positive.
Ligations with low molar ratios did not worked at all (miniprep-RE tested too).

Best working condition (Ligation):
10 µl total reaction.
2.5 U Fermentas T4 ligase and 1 µl Fermentas PEG.
20 ng of vector and 1:12 molar ratio (vector:insert).

60ºC 3 minutes, cool down to 4ºC, then ligase added at 4ºC. 16ºC incubation Overnight in thermalcycler.

All the 10 µl of ligation transformed en DH5alpha E. coli.

Result: 10 colonies picked for miniprep, 8 of them positive in the Restriction assay.

Good luck if you are trying this!

Top job Mack! Well done. Thanks for keeping us up-to-date.