Cloning woes - Running out of ideas! - (Nov/12/2009 )

OK. I am at my wit's end with this plasmid I've been attempting to construct. Here's what's going on in a nutshell:

Insert = 2775kb, cut from Promega pGEMt vector using BglII
Vector = 5100bp, cut with BglII and dephosphorylated with AP

After digestion, both insert and vector are visualized on a gel (next to uncut plasmids) and gel purified using Qiagen spin columns. The digestion does not seem to be an issue as I routinely get 2 bands when cutting the insert from the pGEMt vector and get 2 bands if I double digest the vector. Ligations (using a number of insert:vector rations) are carried out using NEB T4 ligase and are incubated overnight at 16C. Transformations are done using heat shock. I generally add 3-5 ul of ligation reaction to 25 or 50ul of cells. Cells are placed in SOC and allowed to recover for 1.5 hours before being plated on amp resistant plates. The positive control plate consistently has hundreds of colonies while the experimental plates have none - no background, nothing. I should also add that I have done the same procedure skipping the gel extraction step for my vector but still purifying it. I have done ethanol washes after the purifications just incase salt was carrying over (although I know this is not very likely considering I use Qiagen kits). And importantly, I have run a gel containing my ligation product in one lane, the same concentration of vector I added to my ligation in a second lane, and the same concentration of insert I added to the ligation in a third. In the ligation lane I did not see an 8kb fragment as expected. Instead, I saw a band for the vector, which was the same intensity as the just-vector lane, and a very faint band for the insert, which was much less intense with the just-insert lane. Hmmm...this last piece is puzzling. What's also puzzling is my lack of background colonies.

If you have any ideas, I'd really appreciate hearing them.

kbd on Nov 12 2009, 10:14 AM said:

How much total DNA are you adding to your reactions? What volume are your reactions, and how much ligase are you adding? How are you quantifying your insert and vector concentrations, and are you ratios based on molar concentrations or something else?

Following ligation, you can use 1 ul of your reaction as template for PCR using primers in the vector to amplify across the junction to check the ligation for success.

My ligation reactions are 10ul total and contain 1ul ligase (400,000 units/mL). I quantify my DNA using a nanodrop spec. My ratios are based on molar concentrations and have ranged from 1:1 (insert:vector) to 10:1 (insert:vector). Thus, the total DNA I have added to the ligations ranges from ~38 - ~155ng.

Oops. Disregard the units I indicated for ligase.

And Fishdoc, thank you for your comments. I will probably do a PCR as you mentioned. I have been holding off on doing that since it will mean ordering new primers. Time is money though, and at this point, $30 for primers is worth it.

kbd on Nov 12 2009, 11:48 AM said:

I recently had 5 ligations that did not work. Each of them was done in a 10-15 ul total volume. I repeated the ligations with the same conditions, added a little bit more DNA (total between 15 and 30 ng/ul) and increased the volume to 20 ul. Each of them worked the second time with basically only increasing the volume of the reaction. That led me to believe the ligase was at too high of a concentration (I used 1 ul of ligase in 10 ul total reaction volume). I'm not sure if that is actually what happened, but it's the only thing that was changed. NEB recommends a final DNA concentration of between 1 and 10 ug/ml (ng/ul) for ligation:

Q7: How much DNA should be used in a ligation using T4 DNA Ligase?

A7: The unit definition uses 0.12 μM (300 μg/ml) lambda HindIII fragments. The high DNA concentration can be used for linker ligation. To promote circle formation, useful in transformation, a lower total DNA concentration should be used. The overall concentration of vector + insert should be between 1-10 μg/ml for efficient ligation. Insert:vector molar ratios between 2 and 6 are optimal for single insertions. Ratios below 2:1 result in lower ligation efficiency. Ratios above 6:1 promote multiple inserts. If you are unsure of your DNA concentrations, perform multiple ligations with varying ratios.
http://www.neb.com/nebecomm/products/faqproductM0202.asp#346

I think I've also read somewhere that using too much DNA (or too much insert) can result in an abundance of linear constructs.

As I said, I've had good luck using between 15 and 30 ng/ul. A colleague of mine refuses to use less than 30 ng/ul, and she still has good luck, but looking at the concentration of DNA in your reaction may be a starting point. I will typically aim for 100 ng of vector DNA in the reaction, and then add enough insert DNA for a 3:1 molar ratio. I can't say I've had a problem doing it that way except for the 5 rxns described above, which were fixed after increasing the total volume of the rxn.

I suspect you might be damaging your vector during the alkaline phosphatase reaction -- how are you performing this step?

We routinely used 1/5 - 1/10 the recommended amount of AP, and add it directly to the vector digestion at the end. We let the AP reaction go for 5 minutes, then load it immediately on an agarose gel for gel purification.

We had some cloning problems in the past, and traced it to this step -- too vigourous a reaction with AP and you've ruined your vector and won't get anything out of the cloning. Since we went way back on both the amount used and the time allowed, and stopped trying to heat-inactivate it in favor of loading the sample right on a gel after five minutes, our success rate went way up.

You may also be damaging your DNA by exposing it to UV during the gel purification step. We add 0.28 g/L (~1 mM) guanosine (e.g. Sigma G6752) to 1x TAE used for casting and running the gel as a "sunblock", which also increased our success rate dramatically (see Grundemann, D., and E. Schomig. 1996. Protection of DNA during preparative agarose gel electrophoresis against damage induced by ultraviolet light. Biotechniques 21(5):898-903. There's a pdf here).

Hello Fishdoc and Homebrew,

Thanks for your advice. I am hoping to set up a series of ligations tomorrow and fortunately have some vector that has not yet been dephosphorylated. I'll give your suggestions a shot and let you know how it works out.

Dear Homebrew and Fishdoc,

I just want to let you know (5 months after the fact, sorry) that I combined your suggestions regarding the AP and reaction volume and FINALLY, MY LIGATIONS WORKED!!!

Thanks so much for your help.