Regular readers will know about the advantages of T4 DNA polymerase-mediated ligation independent cloning. The fact that it is faster, more efficient and allows easier parallel cloning than conventional cloning has made it my method of choice in the lab.
But the technique does have it’s downsides – not least the requirement that existing vector multiple cloning sites be modified to convert them into ligation independent cloning vectors.
This paper by Li and Elledge recently flagged up in a comment by Max (thanks Max!) looks like it could change all that. It turns out that no sequence modification is required at all and LIC (or sLIC – sequence and ligation independent cloning as the authors call it) can be performed at any site in any vector of your choice.
If you are familiar with T4-mediated ligation independent cloning you will know that the vector sequence needs a specific LIC site containing restriction site flanked by regions that lack one of the nucleotides (e.g. adenine) for a 13-14nt stretch (read this first if you are not familiar with it). After linearising at the restriction site, the vector is incubated with dATP + T4 DNA polymerase, which chews back the 3′ end of the DNA until it stops after 13-14 nucleotides due to the presence of the dATP. This creates a single-stranded region to with a similarly treated insert can be annealed.
In this paper, Li and Elledge showed that the specific vector LIC site was not required. Treating the vector with T4 DNA polymerase and no dNTPs for a certain length of time (30 minutes was optimal for them) was sufficient for vector preparation. The single stranded stretch this creates is longer than required for annealing an insert, but single stranded gaps like these are apparently repaired very efficiently by E.coli after transformation so this is not a problem.
They also demonstrated that inserts prepared in one of three ways could be successfully annealed and transformed, which considerably increases the versatility of the process. The insert prep methods were:
1. T4 DNA polymerase treatment. Just like the vector, the insert could be subjected to T4 DNA polymerase treatment (without dNTPs) to create single stranded regions that will anneal to the prepared insert.
2. iPCR. Non-treated PCR fragments could also be annealed to prepared vectors, albeit with much lower efficiency. They showed that this is because a subset of fragments were synthesised incompletely, resulting in 5′ overhangs. Although this is was relatively in efficient, the authors found that it was robust and recommend it for routine cloning.
3. Mixed PCR. This was the most efficient method. It involves amplifying the insert using two separate reactions. In the first reaction, the forward primer has a 30nt tail homologous to the vector ss region, while in the second reaction, the reverse primer has the homologous tail. After amplification the two reactions are mixed, denatured and annealed to yield a subset of inserts that have both the forward and reverse primer single stranded tails that can be annealled into the prepared vector.
The authors showed that efficient annealing needed only 20-30 nt (single stranded) regions of homology at each end of the vector and insert. Amazingly, the homologous regions didn’t even need to be at the ends of the insert – the authors showed that non-homologous regions of up to 20 nucleotides could be tolerated as the branched products produced after annealing are efficiently trimmed and repaired by the cell.
I have not had a chance to try out this method yet, but it certainly looks very exciting. It looks to me like once this I have this protocol is up and running there will be no need at all for restriction enzyme-mediated cloning… which is a day I long for!
When I have some results of my own from this I’ll publish them, and my protocol here so watch this space.
Digg
Stumble
Reddit
Delicious
TwitterAbout the Author
Nick Oswald
Nick is a molecular biologist-turned-publisher. After a PhD in Developmental Biology and an eclectic seven years in biotech he is now Editorial Manager of Neuroendocrinology and the founder and Editor-In-Chief of Bitesize Bio. You are welcome to connect with Nick on LinkedIn
Wow! that is pretty exciting information Nick, i had planned to try the LIC method for an ongoing cloning, but dropped the idea since i had to design the vector – but i guess i could possibly adapt this method for immediate use.
SLIC is definitely promising but could be tricky sometimes, I spent around one month on that and succeeded finally. My suggestion is to avoid gel elute your insert and vector, while you can use restriction enzyme digestion to suppress the background (your PCR template or parental vector containing the resistant marker). It seems to me that the agarose left in the gel elution might inhibit T4 polymerase exonuclease activity (otherwise you might want to try using cycle purification kit to further purify the gel elution, if things don’t work out).
For the PCR part, just skip the final extension step and do the T4 polymerase treatment anyway. I used 15 nT overhang and it works OK (3 positive clones out of 10 randomly-picked colonies). I will be grateful if someone could also share his experiences here.
Thanks for the tips Yan. I’ll be giving SLIC a go soon and I’ll post my results here. If anyone else has any experiences with this technique it would be great to hear your thoughts.
I’m not quite clear if there’s an advantage to do it in a recA strain?
Hi Kurt,
There is no advantage in doing this in a strain with active recA. As the paper mentions, there are two mechanisms of homologous recombination in E.coli – recA mediated and single strand annealing.
In this protocol the homologous recombination (i.e. the sticking together of the exposed homologous single stranded regions) occurs in vitro, before transformation. This is done in the paper either by recA or by single strand annealling.
The authors showed that at low DNA concentrations recA-mediated recombination was best, but at higher concentrations (above about 1ng/ul) single strand annealling was equally as good.
Either way, by the time the construct is transformed, the recombination part has already been done – any linear DNA that has not recombined will not transform – so the presence or absence of recA ince the hose strain makes no difference.
Hello,
I arrived to this site some time ago using the keyword LIC. I have been using LIC cloning for one year and could make it work but with difficulty.
In my cloning experiments I wanted to take advantage of LIC cloning to introduce tags or restriction sites, so my procedure was to introduce LIC sequences in both insert and vector by PCR. I made up a software called LIC generator in order to generate the primers.
Recently I have used LIC certified T4 DNA polymerase from Novagen with very little success. I take the idea of a preceding comment that agarose might inhibit the polymerase. I tried sLIC cloning by simply suppressing the final elongation step at 72°C to finish the strands after the PCR and in my last experiments it was the only thing that worked.
So the conclusion is that LIC and sLIC work but there might be some missing tricks.
PS : thanks for the site, I am visiting regularly
Hi Marie-Claire
That’s strange. I have found ligation independent cloning to be quite successful – much more so than traditional restriction/ligation cloning. I wonder what the problem could be – which protocol did you use?
Your LIC primer generator looks very useful (and comprehensive!). Here is the link so everyone can try it out – http://noxtoolbox.ibs.fr/LICgenerator/
Hi guys,
to those of you that think that agarose might be problem in LIC: If you look at the protocol of Clontech’s <a href=”http://www.clontech.com/images/pt/PT3941-1.pdf”In-fusion (version 2), you will notice they don’t do any agarose purification of the insert and claim that this enhances cloning efficiency. Instead they do a Exo SAP cleaning (=I guess! At least their “cloning enhancer” smells a lot like Exo/SAP) of the PCR digest. Could that be a crucial step?
I think they say somewhere in the brochure that this enhances cloning efficiency.
Anyways, Exo/SAP might be faster if you have many clonings to do (one distant day, is this whole protocol really works) than gel-extraction…
@Maximilian
An ExoSAP cleaning will only remove primers leftover nucleotides of your PCR reaction … for that one could just do a normal PCR purification. No need for agarose gel purification.
The point is, though, that you might have unspecific PCR products or you want to get rid of your uncut vector. For that you have to do a agarose gel purification, right?
Regarding the agarose gel purification… at least the Qiagen kit states that for removing of all traces of agarose one should do an additional washing step with buffer QG.
So Nick, what kit are you using for your gel purification since you seem to have no problems?
Marc
I have used Qiagen’s gel purification kit and Zymo’s without any problems. I normally do all of the recommended washes twice (e.g. for qiagen, 2xQG washes, 2xPE washes) and I elute in a relatively high volume – normally 100ul.
I think the high elution volume is crucial. I have certainly seen in problems when I elute in low volumes e.g. 30ul for quiagen or 6-10ul for zymo. I always took this to be because the higher volume diluted out any impurities that came off the column, and agarose is definitely one of the possibilities.
If I want a really clean prep I will ethanol precipitate after the gel extraction. For other applications (I have not tried it with LIC) I have found that this gives a much cleaner prep that gives generally better downstream results.
Hi Nick (et al.),
I started to have problems when I suppressed clean-up of the T4 treated fragments, and simultaneously I shifted to LIC certified T4 from Novagen instead of T4 from NEB. I am not quite sure of what protocol to use (in what buffer, what temp, what incubation time for T4 treatment) and what reagents to add when mixing the two products together. There are many different protocols and maybe my guesses were wrong.
Many thanks for the discussion going on here.
Marie-Claire
@Marc: My PCRs are usually from genomic DNA. If not, an alternative to get rid of plasmids might be adding DpnI to the Exo/SAP mix, but I’ve never tried this. Clontech’s cloning enhancer might be a Exo/SAP/DpnI-mix.
@Marie-Claire: I use NEB’s T4 DNA polymerase, but have no experience of Novagen’s. For the protocol, take a look at http://bitesizebio.com/2008/01/17/ligation-independent-cloning-protocol/
I think that has the parameters you are looking for
@Max: I think Marc is referring to the vector prep, rather than the insert prep.
So, has anybody tried to use SLIC in a “real lab” cloning experiment yet?
I just ordered some oligos (25 bp homology as a compromise… I hope that’s enough) and I’ll report my results in a week or so.
It would be great if we could share tips and tricks for the best protocol here!
Cheers,
Marc
@Marc. I have not had the need to do any cloning recently so I’ve not given this a go in the lab. I think tomorrow I’ll order some oligos for an experiment specifically to try out this technique.
@Jon. I don’t think that there should be any problem with the vector/strain you used. What did you try
@Nick, Sorry, I did not see your comment yet!
I used E. coli CC118 lambda pir cells (the vector I am cloning into is lambda pir based suicide vector).
I tried variations with RecA, without RecA. Varying incubation times from 0min to 30 min. And I also tried both electrical and chemical transformation. I am certain that my overlaps are correct.
Nevertheless, I will try it again over the next few weeks (this is a background project). Additionally, I am going to make chemical competent DH5 alpha that contains a plasmid which allows the cell to support lambda pir based vector.
I will let you know how it goes.
Hi, just a useless idea if someone is playing around anyways: Could using recA bacteria (like NEB’s turbo) improve the efficiency? Perhaps it’s enough if the recA is in the bacteria?
@Max – Good idea….
Hi,
everyone. SLIC works for me like a dream so far.
Have done two Fragments in two days:
3500 Bp into 9000 Bp Vector and 1800 Bp into 11000 Bp Vector.
Cleanup by Zymo-Kit in low Elution Volume and quantification by 260nm.
T4-Polymerase by NEB
Reaction:
15,67 µl DNA
2 µl NEB Buffer 2
2 µl 10x BSA
0,33 µl T4-Pol.
Annealing:
Equimolar amounts of Vector and insert
(I never reach the in the Paper suggested 1µg Vector, works still)
30 min. 37°C
Transformation by Electroporation
Always transform same amount of Vector as used in Annealing reaction as negative control
One of the Transformation took two days to grow.
I guess one of the issues with the mixed PCR method is that 4 primers are needed instead of 2. And with regards to how many overhanging bases to use, how ever many bases are required for a sufficient Tm at room temperature would be suffice wouldn’t it? So this number would vary depending on the sequence you used.
Another thought. Can any polymerase with 3` exonuclease activity be used? How about Phusion for example?
Hi, I have tried SLIC method in our lab for two months, I successfully made two construct with this method ( my overhang is 20-23 nts), but when I tried more cloning with this method, I always get too many negative clones or nothing. Is there anybody know why?
Li
Hi, Johannes,
Could you tell me how much vector and insert you used for annealing?
Thanks
Li
Hi guys,
I have just found complete sLIC protocols in this paper
(see supplementary materials p.14)
http://www.nature.com.gate1.inist.fr/nmeth/journal/v6/n6/suppinfo/nmeth.1326_S1.html
Not tested yet but I have been trying hard to find a robust protocol
before and was not very successful.
Good luck,
FYI,
I have found a protocol that looks even better than sLIC. It’s called SLIM (site-directed, Ligase independent mutagenesis). You can use it for insertions, deletions, and point mutations. I haven’t tried it, but plan to soon.
Here’s the pubmed link.
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC535700/?tool=pubmed