HELP!! Problem with cloning (self-ligated vector) (trying since 4 months (View forum version)


Posted 12 May 2014 - 12:28 AM

Hey Guys,

I've been having problems with cloning since 4 months now, the vector seems to self-ligate with itself for some reason without taking in the insert tried different res sites and different vectors but the problem is still not solved. I've given here the steps that I followed.


·         Insert (5.8kb) and Vector (3.5kb, pSL119)

·         Plasmid isolation (using Qiagen mini prep)

·         Phusion PCR amplification of Insert from plasmid using non phosphorylated primers to add                res sites to ends (7bp overhangs).

·         Gel extraction of the fragment (Using Thermo scientific gel ext kit)

·         Restriction of the Fragment using Nco1 and BamH1 (Fast Digest, Thermo Scientific ) (should            produce phosphorylated cohesive ends) also tried with (Sal1 and BamH1)

·         Reaction Purification (using Qiagen Reaction purification Kit)


·         Vector Dephosporylated using Fast AP


·         (Thinking of using kinase on the insert now, haven’t done it yet)


·         Overnight Ligation using T4 DNA Ligase (1:5, Vector:Insert) (also tried 1:2, 1:3, 3:1, 2:1)


·         Reaction 1: Vector=20ng; Insert: 100ng


           Reaction 2: Vector=100ng; Insert=500ng


·         Produces several colonies and all the colonies have self-ligated vector and no insert (colony PCR analysis and analysis by restriction digestion)


Getting really frustrated with this hope you guys can help.

Thank you


Posted 12 May 2014 - 12:45 AM

That is a large insert to be putting in a relatively small vector - this may be causing your problems in part.


You say you have 7 bp overhangs - how big is this before the restriction site starts - you need at least 3 bp and preferably 6 bp 5' of the restriction site to ensure that it will cut efficiently.


You shouldn't need to get extract the insert - just a PCR purification should be fine.  If you are worried about template coming through (e.g. if you used a plasmid as template) then Dpn1 digest the reaction, this should cut the methylated plasmid, but not the unmethylated PCR product.


I don't think your ends are compatible (I didn't check - but you should), so you don't need to dephosphorylate the vector.  Alkaline phosphatases often cause more problems than they solve, so avoid them if you can!  Also check that your digests are working by doing single digests on plasmid.


Your insert:vector ratios should be based on MOLAR ratios, not ng amounts.  Use the following formula:  ng insert = ratio x (insert bp/vector bp) x ng vector.  Where ratio is an integer e.g. 3 for a 3:1 insert:vector ratio.  Note that you may be better off using 1:3 (insert:vector) as your insert is much larger than your vector.  Try a range of ratios.


Posted 12 May 2014 - 09:20 AM

I agree with bob1's points above.

You don't need overnight ligations. If they are going to work, they will work in 5-30 minutes.

Please show us the 5' primer extensions you are using in your PCR.

You can test your insert ligation ability by taking your purified PCR product, cutting with ONE of the two renzymes, ligating (with only the insert fragment), and running the result on a gel. You should see a double length fragment. Do this for both enzymes, and verify that your insert can be ligated.

Test your vector digestion by looking for linearization with each of the two enzymes separately, which will verify that you can cut the vector with both enzymes.

Are the cut sites on your vector sufficiently far apart?


Posted 13 May 2014 - 02:39 AM


thank you for the tips.


I just checked, I have nearly 14bp on the either side from the start of the restriction site on the primers 

The cut sites on the vector are 38bp apart 

I have used molar rations to calculate the ligation mix, I had given the ng to show how much of dna there was in each reaction.

I did not dephosporylate the vector previously, but did this now because the vector was self-ligating for some reason (ends are not compatible).


I think the problem could be with improper restriction , will have to check this on the new vector and look for linearization.

also must check with single site ligation


Posted 13 May 2014 - 12:55 PM

The amounts you gave in your first post are not molar ratios!  they are ng ratios.  Follow the formula and you will see that for 20 ng vector a 1:5 ratio would require you to use 166 ng insert... insert ng =5 x (5.8/3.5) x 20.



However, it does sound like you are having problems with digestion - definitely check that it is working properly for both insert and vector.