"ladder" ligation result - cloning is the bane of my existance (Oct/26/2008 )
Hi everyone,
I would love, love help with this one. I am all out of ideas.
I am trying to add a 1kb insert into a 3kb vector. I do a double digest with XbaI and XhoI in NEbuffer 2 (50 ul rxn with 1.25ul of each enzyme). Digest for 1.5 hours, inactivate enzyme for 10 min at 65 deg, then add 1U Alkaline phospatase and cook for an hour more. Everything looks good on the gel.
I extract digested vector. (The insert is gel extracted PCR product).
Next, on to the ligation rxn: for a 10ul rxn, I use 60ng vector and insert ratios: 1:3, 1:5, and 1:10. I have tried ligating at 4 deg. overnight and at 16 deg. for 1-2 hrs.
The result is always the same: no nice "smear" of circular/coiled/nicked vector. Instead, I get a series of discrete bands spaced at 1kb. The original vector and insert band mostly disappear.
When I transform: no colonies. My positive control works; the negative control (ligase but no insert) also has no colonies.
I decided I was getting insert trains because one of the enzymes was not properly cutting my inser. My primers are designed:
5' TAG TCTAGA FWD PRIMER SEQUENCE 3' (XbaI)
5' TAG CTCGAG REV PRIMER SEQUENCE 3' (XhoI)
The "tag" part is to give a longer sequence for the enzyme to react with... I'm now not sure if 3 bp is long enough. I've done 6 in the past...
So, to test if the insert is being cut by both enzymes I singly digested the vector and dephosphorylated. I then ligated each single digested vector with a double-digested insert. The result looks like all my other ligations: the ladder. I can't tell if any of the vector is being taken up. What should I do next? What can I try? What am I missing? What is wrong??
(It's been two months now.)
Thanks in advance...
The first thing I would try is to omit the dephosphorylation. It usually causes more problems than it solves. I would cut vector and gel purify, then try the ligation, with an insert+ and insert- control. If you get substantially more colonies on your insert+ control, then pick some an go forward. Otherwise, you have to optimize your vector cutting and preparation. There are several approaches -- PCR of the vector, to dramatically reduce background circular DNA, dephosphorylating with a more user friendly enzyme (shrimp alkaline phosphatase or antarctic phosphatase). Make sure you are not trashing your DNA with short wave or long exposure to UV.
I don't think it makes a difference but keep the 6 bps on your primers next time so when you come to a situation like this you don't have to worry about it. I know it seems like a waste but it's only a few cents and it's a cheap way to eliminate a potential problem. Check out the NEB and Fermentas charts to check those number of overhanging bases are ok but i suspect they are fine.
Now, Xho I. I'm really fishy about this enzyme as a cloning enzyme. I've only cloned with it once i admit, and it is one of my favourite most reliable screening enzymes, but when it comes to cloning into it - i just don't know. I had real problems for no particular reason the first and only time i cloned into this site (and your cloning looks straight forward enough), i would avoid it as a cloning site. Sorry i don't have any more than a hunch for you on this but i just worry about it as a cloning enzyme. If you are going to re-order your primers with more overhanging bases, try swapping the Xho I for something else and see how you go - please let us know how it goes if you do! And you know what? It has the same overhang (5`-TCGA) as that pain Sal I - it is notorious for poor cloning - several people will vouch for that - and i've experienced that pain as well a couple of times (you may be wondering if i ever get anything cloned(!), well i do usually and often!). I suspect there is something fishy about that 5`-TCGA overhang - all hunches of course so i have very little substance to back up my claims but try swapping the site and see how that goes - it would be interesting to get some confirmation for my hunch.
If you're getting no colonies even from the vector only control, then use more vector, use as much insert as possible and use high concentrated ligase to really crowd that ligation and get it happening. By the way, i usually ligate for 1 hour at room temp before i go for the 16C O/N and 4C. I actually do the room temp for 1 hr, transform that and see how it goes, then put the ligation straight into 16C O/N so if the room temp was no good, well i have 16C O/N waiting to be transformed the next day. Keep doing the negative control - it's critical to seeing how well your vector is performing.
If you get nothing out of that, then as Phage says, drop the dephosporylation, it's not absolutely critical here. It's preferable to prevent re-ligation of singly-digested vectors but if you're not getting any background to start with and dephosphorylation could be the issue then give it a go without it - there's nothing to lose.
I'm assuming the Xba I site does not have an overlapping methylation site because it is in the MCS but check that. Any GATC overlapping the Xba I site (i.e gaTCTAGA or TCTAGAtc) and that's probably the issue solved right there. That's not an issue in your primers because they havent passed through the bacteria yet so havent had a chance to be methylated like the vector has.
Good luck,
Rob
Actually, I did not use AP for my first ligation. I didn't run it on a gel, just transformed. For that one, I got the same number of colonies on my experimental and my negative control. It looked grim, but I screened some colonies anyway: no insert. So I started de-phosphorylating after that... Come to think of it, that was the one and only time I got colonies. The AP must be nuking the vector ends. But how do I get the insert in there and prevent empty vector formation without the AP?
Thanks for the response!
usually if you do double digest you could omit the alkaline phosphatase step . because the ends can't anneal properly to each other anyway that's assuming those two enzymes don't produce compatible ends.
I would love, love help with this one. I am all out of ideas.
I am trying to add a 1kb insert into a 3kb vector. I do a double digest with XbaI and XhoI in NEbuffer 2 (50 ul rxn with 1.25ul of each enzyme). Digest for 1.5 hours, inactivate enzyme for 10 min at 65 deg, then add 1U Alkaline phospatase and cook for an hour more. Everything looks good on the gel.
I extract digested vector. (The insert is gel extracted PCR product).
Next, on to the ligation rxn: for a 10ul rxn, I use 60ng vector and insert ratios: 1:3, 1:5, and 1:10. I have tried ligating at 4 deg. overnight and at 16 deg. for 1-2 hrs.
The result is always the same: no nice "smear" of circular/coiled/nicked vector. Instead, I get a series of discrete bands spaced at 1kb. The original vector and insert band mostly disappear.
When I transform: no colonies. My positive control works; the negative control (ligase but no insert) also has no colonies.
I decided I was getting insert trains because one of the enzymes was not properly cutting my inser. My primers are designed:
5' TAG TCTAGA FWD PRIMER SEQUENCE 3' (XbaI)
5' TAG CTCGAG REV PRIMER SEQUENCE 3' (XhoI)
The "tag" part is to give a longer sequence for the enzyme to react with... I'm now not sure if 3 bp is long enough. I've done 6 in the past...
So, to test if the insert is being cut by both enzymes I singly digested the vector and dephosphorylated. I then ligated each single digested vector with a double-digested insert. The result looks like all my other ligations: the ladder. I can't tell if any of the vector is being taken up. What should I do next? What can I try? What am I missing? What is wrong??
(It's been two months now.)
Thanks in advance...
First of all, what vector are you using? How many bp are there between the Xba and Xho sites?
I would ligate at 16 overnight, or RT for a couple of hours.
The NEB website suggests that even 1 bp on the end of Xba and Xho sityes gives good digestion if the DNA in question is treated as linearised vector.
From your message, I presume the "ligation ladder" starts at ~1 kb, right? At least you know the insert is properly double-cut, and the ligase enzyme is OK!
For the vector, that's a bit harder. You could try ligating it at a high concentration (>100 ng/ul), because intermolecular interactions outweigh intramolecular interactions at that concentration and higher... Alternately, you could try transfecting cells with the digested vector and plating out large volumes of the competent cell, which should give you an indication of the completeness of digestion.
One thing you might try is to reduce the amount of vector in the ligation reaction, while maintaining the ratio of insert and vector. You might just tip things into a good concentration for ligation!
I checked on that: the vector is L4440 (c. elegans RNAi feeding vector) and there ~100bp b/w cut sites, so it should not be an issue. I also see a low m.w. smear drop out when I double digest.
Yes, there is actually a good, strong band at 2kb... 3kb, 4kb, etc. The 1kb band (my insert) pretty much disappears. It seems the insert pieces really like each other...
How high would you go? I use a protocol that suggest calculating 15 femtomol of "ends", adding that amount of vector to 10ul ligation reaction, and calculating the insert amounts based on that. From my calculations I get ~30ng of vector required. I've been using double that to try to shift the balance towards vector-insert interactions vs. vector-vector.
How low do you suggest I try?
Thanks!
I'm going to set up a couple of reactions now... and digest everything without the AP.
How high would you go? I use a protocol that suggest calculating 15 femtomol of "ends", adding that amount of vector to 10ul ligation reaction, and calculating the insert amounts based on that. From my calculations I get ~30ng of vector required. I've been using double that to try to shift the balance towards vector-insert interactions vs. vector-vector.
I was actually thinking about testing the vector for double-digestion, which was why I suggested you go gfor the higher concentration. If you are thinking about the actual ligation reaction, I would stay at 30 ng vector. By using the higher amount of vector, you are actually pushing the reaction away from inserting the gene.
I would go with 30 ng vector, as well as 15 ng to get the lower limit. I would still use 1:3 and 1:5 vector:insert ratios.
All the best with it.
I disagree with this. If there are no colonies on either plate i don't see the point in reducing the amount of vector. What would you be hoping to achieve here?
I disagree with this. If there are no colonies on either plate i don't see the point in reducing the amount of vector. What would you be hoping to achieve here?
Just trying to play around with the concentration of DNA, to push the reaction towards intermolecular interactions over intramolecular interactions. If the vector has been cut and treated well, the amount of vector from 60 to 30 ng will not make that much difference (only a twofold change). Apart from anything else, d.g. has not been having any joy following the current protocols. What did Einstein say? Madness is doing the same thing again and again, and expecting a different result, or something like that.