Awesome Daniel, thanks for the response.

Hi Rachael,

My answer to this question would vary depending on the ends of your fragments you are ligating. If you are preforming a ligation where the only way for the vector to close properly is to have all 3 fragments together, then I would just add all 3 fragments and let it go (ie if your vector was cut say NcoI - XbaI, and you were trying to put in a fragment Nco-BamHI, and another BamHI-XbaI...then you need all 3 for your fragments to close your vector). With other combinations or with blunts I would do different strategies. If you are using sticky restriction sites I would not leave your fragments together too long in the absence of vector, as the ligated inserts are still available as substrate for the ligase, and it will start to assemble concatemers of your insert which can be really difficult to sort through in your vector. I usually wouldn't allow these to sit together for longer than 15min before adding my vector. I would only use this strategy when the restriction fragments used would allow my vector to close without one of my inserts, and in this case I would normally add the vector and the insert that will not close, then I would add the third fragment that would allow the vector to close.

Cheers

Ron

Hi Ron, thanks for your advice. I have cut both of my amplicons with BamHI and Bgl1. I then cut my plasmid just with BamHI. They all will have sticky ends.

Hi Rachael,

Ok, this makes it easier! Don't forget that you will need to dephosphorylate your vector so it does not just reclose on itself. then I would add your vector (BamHi-BamHI) with one of your BamHI-BglI inserts and allow it to ligate for no longer than 15 minutes. Also I would use a ratio of 1:1 vector:insert here so that you don't have an excess of insert to form concatemers and seal up your vector. Then I would add the second fragment at a 1:3 ratio (V:I) so that you have an excess of this one to seal up your vector.  Allow it to ligate for at least an hour. There is no need for a reduced temp/long duration ligation even though BglI is a blunt, as you have the BamHI sticky to bring it into the vector...then it just becomes a zero order kinetics reaction and your blunt ends will seal (because they will be part of the same molecule at this point). This is all assuming you are using a standard ligase, and not one of the 5min ligation super fast ligases with PEG in the buffer. Also note that you will have to sort through your clones quite a bit, as your BamHI containing fragments can ligate to either side of your vector, so your inserts might end up backwards or with 2 of either of the inserts. Usually a PCR colony screen is the easiest way to sort through a lot of clones that are like this.

Good luck

Ron

Hi Ron, just for the record, I was under the impression that Bgl1 produced sticky ends as well. 

Hi Rachael,

No Bgl I is a blunt cutter (it cuts in the middle of GCCN-NGGC). Maybe you are getting confused with Bgl II? (it cuts A-GATCT). Whatever you do...do not try ligating the 2 (BamHI-Bgl I) inserts first...In this case they will ligate at the sticky BamHI sites every time, and will not give you the Bgl I to Bgl I connection that you are hoping for.

Good Luck (I am sure it will work for you)

Ron

 Hi Ron, I have just looked at both BglI and Bgl II and they both produce sticky ends. The BglI is GCCNNNN-NGGC and the BglII is what you said earlier. But that is beside the point, I am wondering why you said something about the super fast ligase with PEG because that is what I am using.  

Hi Rachael,

Sorry about that...It totally is a sticky cutter! I remembered there was something weird about Bgl I so when I quickly looked at it I missed the subscript 4 on the first N so thought it was blunt. Now that you have brought my attention to it, I remember what the issue is with Bgl I...because it cuts the multi sequence GCCNNNN-NGGCC, you have to take care that the Bgl I sites are exactly the same otherwise the ends will not be compatible. I am assuming that you have engineered your Bgl I sites so this shouldn't be a problem in your case. What I was saying about the super fast ligase is that the entire ligation reaction proceeds in 5 minutes, so if you add sticky fragments together and let them sit for 15min before adding the final fragments...your free BamHI and Bgl I sites will start to grab onto each other and form concatemers. In this case I would probably just add all 3 fragments at the same time, and I would keep the molar ration of the two inserts equal to each other but 3 times the vector. You could try ligating the 2 inserts first for a minute or two before adding the vector. Either way you will need to sort through a bunch of clones, as both inserts can anneal to themselves and close your vector. PCR colony screen is going to be your best friend for sure!

Good Luck

Ron

Ok, Ron. I appreciate your help. I will probably do some where I add all three at the same time and some where I put the two fragments in for about 2 minutes before I put the plasmid in. 

Hi Rachael..Can you plz let me know adding all the three fragments worked for you or not? I am using a similar method where I want to join two fragments along with the plasmid to prepare a knockout..

Hai Rachael...

I have the same problem as you had.. But this time I digest my amplicon with NotI and PacI, while digest my vector with NotI.. I did triple ligation but have never succeed yet..

I know that this discussion is already long time ago. But I wonder if you could give me some suggestions or kindly tell me how you did last time..

I am quite frustating by always failed ligating my 2 inserts to my vector.

thank you very much

do some people try ligate plasmid backbones to 3 DNA fragments?

thanks

Hi Yuxi,

Yes people will often try to ligate multiple fragments all at once, but it really depends on what your fragments are, and how big of a total construct size you have. Generally if the construct isn't too big, and there is only one way for the fragments to piece together, then it is worth a shot, but if the construct can go together multiple ways, it is best to break it down into smaller steps.

good luck

Ron