my strategy of ligation - (Dec/23/2008 )
Hi, I plan to do the ligation and transformation with another approach, based on my ... idea. I don't know that whether someone here tried it or not, but I think maybe it can work.
I will ligate by the normally used method, but increase the ligation volume to 40ul (somebody suggested don't mix the ligation exceed 20ul, so maybe the efficiency will be reduced) and time of incubation at 18 degree (my case is sticky end ligation) up to 12h or so. I also mix the control with the linear vector alone and ligase. Then I will run all the volume of ligation mixture and the control on the gel and excise out the band which is not the self-ligated product (I can compare with the control on the gel) nor the insert, purify by column and use the eluted solution for transformation.
Even there is the small amount of ligated plasmid, I think the transformation efficiency will be high. How about your opinion?
I try this strategy because I'm facing the problem with ligation. The plates containing ligated product and control have the equal number of colonies formed, and I can't detect any colony harboring the ligated plasmid by colony PCR (I screen hundreds of colonies with each cases, the positive controls were always good).
I digested my vector for the long time (10-12h) with 2 enzymes, treated it with CIP, antarctic phosphatase and then purify by gel extraction (I observed only 1 band on the gel, looked like it was completely digested). I digested my inserts from another plasmid (10-12h), with the same enzymes, of course, then purify by PCR purification. I used 50-70ng vector per ligation, with molar ratio of insert:vector=3:1 (even sometimes 5:1). My vector has the size of about 4kb, my inserts are 2.5Kb; 3.9kb in length.
I also checked the single digestion of all enzymes I used, they all worked.
I checked the remaining ligation mixture, I saw the insert band, the vector band and another bright band which was higher and looked like it has protein (maybe ligase) binding onto. So I think I still got the ligated plasmid after the ligation but somehow I can't got any positive clone 
.
I wished that I just have the ligated plasmid for the transformation, so this idea came to me.
I'm so tired of screening repeatedly with the blank gel plate everytime T__T. 1 month has gone, I have nothing with the work seems to be easy to finish.
I'll appreciate all of your suggestions, thank you in advance.
Why are you digesting for such long times?
Clare
assuming that the ends of your vector are uncompatible (since two restriction enzymes have been used)., you can skip the dephosphorylation step. Over exposure to CIP (calf alkaline phosphotase) can and will damage the ends of your vector making it unligatable. CIP if improperly used can be more trouble than it is worth.
Personally I would try this rather than attempting to isolate the ligated vector-insert molecule. How many colonies are you recovering on your vector only control?
Lastly, increasing the volume of the ligation mix, and thus the concentration of insert and vector molecules will decrease the ligation efficiency, because it makes it harder for the vector and insert molecules to meet.
@Clare: I just want to make sure they are completely digested, sometime I digest about 8h and I still got the non-linear band of my vector and the non-digested band of vector+insert (another vector containing my insert).
@perneseblue: You're right, the ligation efficiency will be much lowed down. For this, I will double the amount of them. I treated CIP for 2h, 3 times with 0.5ul each time (the interval times were 1h; 30m and 30m). Maybe it damaged the ends of my vector like you mentioned, because I got several results which were not interpretable with different digested bands. By the way, I wanna show you the photo I took to verify the ligation by running half the the ligation mixture on 1% agarose gel.
My inserts in lane 1 and 3 were the lowest bands and the vector is the band higher (~ 4kb); in lane 5 it has the similar size with the vector, so they showed the only band in the bottom (~ 4kb). Compare with the control of self ligation (lane 2 and 4); the third bands of ~6.5Kb (lane 1 and 3) and 8kb (lane 5) should be the hybrids. Moreover, they had something like protein binding to, maybe the ligase, I think. For the spreading on the plate after transformation, I inoculate 1h and spread 80ul on half of the plate; centrifuged the remaining and resuspended the cell pellets in 80ul remaining LB and spread on the other half of the plate (to make sure I don't have too much or too few number of colony). Both the control plate and the ligation plate, I got the equal numbers of colony (about 70 in total with lane 1 and 5 cases; about 25 with lane 3 case).
it should be noted that the desired band is not only insert+vector molecule but a insert+Vector which has been circularised. Circularised plasmid DNA runs slower than linear plasmid DNA.
I guess, you should go test the colonies. Use colony PCR if you can (always PCR amplify across the junction between the insert and vector. PCR is so sensitive that it can detect insert DNA leftover from the ligation reaction on the colony selection plate).
With colony PCR you can check all 200 colonies within 2-3 hrs.
PS: A very nice gel. It is a good example of what a ligated DNA mix should look like.
Thank you, perneseblue. Actually, I got tired of screening by colony PCR, I checked almost of the colonies I had (~30-50 colonies each time X 4 times) by both the primer pair of the insert and the 2 primers across the junction like you mentioned, I always have negative and positive control, so no problem with the PCR. I guessed the plasmid number/colony was low (I used pACYC - low copy number plasmid) so I inoculated them in 3ml LB overnight and use 1ul of the culture for PCR, only 1 time I found the correct transformant, the remaining cases didn't show anything different.
Yesterday, one of my senior suggested me to perform the "triple digestion" (which is similar to the strategy of badcell in this forum) because we both agreed that somehow the digestion was not good. I will use the 3rd RE between the 2 I used to prevent the non-digested vector (just increase the possibility of being digested). Another thing I'm trying is using this 3rd RE but after the ligation (badcell strategy): add 1ul of the digestion mixture (0.5ul RE + 10ul buffer) to 10ul remaining (of the ligation reaction I checked on the gel) and incubate for 30 minutes, then use 2ul to transform into 40ul competent cell. I hope I won't feel frustrated again 
.