Hi Pieter,

If the efficiency of the positive control is also very low you would expect it is either the mix or the competent cells. You mentioned that you tested the cells with intact plasmid and got several hunderds of plasmid. IMO that is very low and I think your cells are therefore the problem.

Intact plasmid should transform very efficient. Good competent cells should at least yield 10^7 (preferably 10^8) cfu/ug DNA. So transforming 1 ng of plasmid should give a full plate with thousands of colonies of which it is very difficult to pick a single colony.

I would recommend to either make new competent cells and test them first or buy commercial cells. BTW, the HiFi assembly kit provides you already with efficient competent cells.

Thank you for your answer, Jurian. Do you really think the problem is the cells? The fact that I got 200-300 colonies (I didnt count them obviously, but that's what it looked like) when transforming a control plasmid, doesn't mean I should get at least 10 or so colonies with an actual transformation.

We order the HiFi mix without competent cells, so I'll have to borrow some commercial cells then, or order some and try with those.

Thanks for your answer!

Of course it can be a combination of factors. I also love the NEBuilder system but results can vary. A lot of times I got plenty of colonies but sometimes when the reaction is not efficient I only got a few. In that case it is important that at least the cells are good so you will get those few colonies.

My experience of self-made cells is that the quality is variable and that they become less competent in time. Therefore I prefer commercial cells which should always work. I always used XL1-blue cells from Agilent in the past and they are always very good. At the moment I use Top10 of Invitrogen which are also ok but not as efficient as XL1-blue.

Jurian Bronkhof I just wanted to respond that it was indeed the cells. I tried some commercial NEB 5 alpha cells and immediately started getting lots of colonies and 80% correct inserts.

It seems like the homemade cells had simply lost a lot of their competence.

So thanks for the advice!

Good to hear you can continue with your research. No problem and good luck :)

Pieter Van de Walle Thanks for the follow-up!

Hi Amy, if you give me more info I am happy to help you finding out what is going wrong.

So cloning strategy, hou did you design the primers? Backbone vector etc.

Hi Amy, so if I understand it correctly you want to make a linear fragment using NEBuilder out of 3PCR fragments without using a vector backbone/ selection in E.coli? So assemble the 3 fragments and pick it up with the LHA_fwd and 82b2_rev primers?

I have to say I never tried this because I think the NEBuilder system is designed to make circular vectors. The enzyme mix consists of three enzymes: An endonuclease that chews back the 5' end, a polymerase to fill in the gap after annealing and a ligase to seal the nicks. I can imagine that a lot can happen to the generated fragments if no circular fragments can be formed.

Therefore I would advise you to try it again using a (any) E.coli vector backbone (a linearized pUC19 will do).

Maybe it is possible to skip the E.coli part by adding a 4th fragment (either a vector backbone or any random DNA fragment) that circulizes your construct in the reaction and then do a PCR to generate the full length construct.

But on the other hand, if OE PCR also does the trick maybe use that for your application.

The principal is the same indeed. Maybe it has to do with the specificity and/or stability of the enzymes in the mix? I never tried it at least.

In an other discussion I read the advice to increase the overlap to 30-40bp and reducing the incubation time to 15-30 min using Gibson assembly. Maybe this will help. Good luck with it :)

Amy Johnson

Amy Johnson

Hi Amy,

I am having the same issue and getting incorrect insert with NEBuilder HiFi mix.. can you please tell me the protocol for OE PCR that works for you?