one question on the 4-way ligation - (Oct/05/2006 )

Hi, all,
a couple of days ago, I posted a question on the ligation of 3 three PCR products into a plasmid vector. And i did receive informative feedbacks from you guys.

Here I get another question to clarify: what would be the molar ratio of three pieces over plasmid vector? YOu guys suggested the 1:1:1 for the three pieces to be ligated in. But I am not exactly sure the ratio of inserts over the plasmid. My own first guess is 3:1 (e.g. 1 molar of plasmid, 3 molar of each PCR products). How does it sound?

another minor question--if I wanna PCR screening the correct transformants--what primers I shall use? if I use the vector-specific primers, I am worring if the Taq could go that long (totoal insert for three pieces would be around 4.1kb). But if I use insert-specific primers, I am also worrying about the false-positives...what ideas do you guys have?

thanks so much!

Paula

---

3:1 ratio (e.g 1 molar of plasmid and 3 molar of each PCR product) sound fine. That is okay.



4.1kb is right at the edge of Taq's amplification ability. I wouldn't attempt to use Taq for a colony PCR screen with a product that large. Taq will do most PCR reaction with a 2kb product (but I am a bit iffy on this because some reaction will need optimisation. Which is too much of a bother for a screen) In my experience anything around 1.1kb and below will definately work.

As for false-positives.... it rarely happens, especially when amplfying small products. The signal is STRONG... very STRONG. There is no mistakening it for anything else. The lane is cleared of any miss primed bands.. it is clear of any smearing with eexception of a single very strong band.

However there is a possibility of the vector picking up only 1 or 2 of the PCR fragments (rather then all three)

So for the screening , I would suggest the use of two PCR screens to confirm the presence of all three PCR fragments.

An example
Screen 1
junction between PCR fragment 1 and 2

Screen 2
junction between PCR fragment 2 and 3.

A colony that is possitive for both screens would probably be correct.

To save time, you can run both screens at the same time. Though to save reagents, one can run one screen first, then pick only the positives to be used for the second screen.

Of course, after the screen, pick 2-3 colonies and midiprep them, and tested them by restriction enzyme. On the rare occation something odd may actually get through. So this would be another layer of caution.

---

thanks so much for your rpely to my questions--I lost my last user name and password---so I cannot re-log in until today (after registering a new account).

Yesterday, I setup the 4-way ligation reaction mix using Promega DNA Fast Ligation Kit and let it sit in 4C for 16hour. Then this morning, I transformed the Maxi Competent cells (Invitrogen) with only 1ul of the reaction mix which is suggsted in the transformation protocol. So I was wondering if such a low amount of ligation mix will give any positive transformants...what is your ideas on this? I once heard some people say that sometimes too much ligation mix or DNA added will lower the efficiency of transformation--so in the case of 4-way ligation, which way I shall go--add only 1ul of mix or add about 5-8ul(about the half of the total volumne setup)?


And another question--since the 2nd piece to be ligated carries an antibiotic resistance marker--shall I select the transformants based on this marker or on the antibiotic marker in the vector backbone? I did both anyway today---but I would prefer to use the marker in the insert--in theory, only the correct ligation will yield a complete recombinant plasmid that carries and expresses this marker. Is this reasoning right?

I will keep you guys posted of the transformation results tomorrow morning....

thanks so much!

---

Well, the amount of DNA transformed is based on
-how much DNA vector was used in the ligation reaction
-how efficient are the competent cells in use.
-and to a lesser extend how efficient the ligation reaction is.

I am uncertain how much vector was used in the ligation. So I can't comment if 1 ul is too little. Your cells are however very competent (if Invitrogen advert is true) so it might be okay. The thing to do would be to try 1ul and see what happens. (Which is already being done). Remember to keep the ligated DNA. If there are too few colonies, transform more DNA.

I personally would try using more DNA. (maybe 2.5ul) Mainly because a 4 way isn't that efficient.

I have not heard anything about using too much DNA that it lowers transformation efficiency. However I tend not to use company competent cells, but competent cells made in house/lab. Those cells aren't very efficient (I use 3.5ul to 5ul DNA) but they get the work down (usually)



Yes, you should select transforments based on the marker on the insert. This is far better then selecting for the marker on the vector.

(Provided there is no synegestic effect between the two antibiotics in use) It is even better to actually select for both antibiotic resistent markers on the same plate at the same time. That way it is guaranteed the colony contains a vector whcih carries the 2nd insert.




Ummm... In an idealised situation.. But there is a chance that some template vector, used to clone the 2n insert (with the resistence marker) will come thru. So a few false positives will show up if only this one antibiotics is used.

Secondly and more importantly, there is no certainty that a complete plasmid will show up. The probability is good, but things happen and you'll find some vectors will pick up only 1 or 2 inserts and not all 3. Even though their ends aren't compatible.

There is no skipping the PCR screen. (Although the size can be toned down). But if you use a 96 well PCR plate and a multichannel pipette... does it actually make a difference between 24, 48 and 96? One plate is still going to be used.

---

Perneseblue: if you don't like the competency of your in-house competent cells, I recommend trying out the prep protocol here, which (after a lot of experimentation) yields results similar to the Invitrogen store-bought versions. And you get to make a year's supply at a time.
http://openwetware.org/wiki/TOP10_chemically_competent_cells

I agree with most all of the advice you give, with the exception of the molar ratios, which should be ideally 1:1:1:1. But it will work in any case very likely. There is an inhibitory effect in adding ligation mix to competent cells, but it is not the DNA, but the dilution effect of the ligation mix on the competent cell buffers which is inhibitory. Also, some restriction enzyme buffers (EcoRI notably) have Triton-X100 as a component. The carry over of these into a ligation buffer and then into a competent cell transformation is very inhibitory (I know, it is hard to believe, but "trust me").

---

Hi, all,

I am writing to let you know the transformation results I got this morning:

1) on LB+Cm25 plate (which is selective for antibiotic resistance marker on the vector backbone): I got around 100-200 colonies per plate. I had 8 plates in total.

2) on BHI+Erm150 plate (which is selective for the presence of 2nd piece): I got only 9 colonies out of 8 plates.

3) no colony on negative control (H2O added to competent cells) and tons of colonies on positive controls (a plasmid I used in the past).

4) less than 10 blue colonies on the plates for non-kill-digest-treatment ligation sample and 1 or 2 blue colonies for kill-digest treatment sample.

5) plates were incubated at 37C for 16hours and all the colonies showing up look healthy and around 2-3mm in diameter. Of course I also saw tiny satellite colonies around the big ones, I ignored them when counting and patching.

Yeah, that is what I got so far--I agreed with what you guys said--vector may pick up only 1 or two piecies--otherwise there should not be so many colonies on the Cm plates while so few on the Erm plates. Right? What I did just now--is to patch white colonies on Erm plate onto a new Erm plate and Cm plate and white ones on Cm plate onto a new Erm plate. I will let them grow for about 8-10 hour and see if those guys are resistant to both Cm and Erm. If yes, I will use them for colony PCR, miniprep and RE pattern...

does it sound like a plan? I am sort of scared of large-scale of colony-PCR process since I need to test two junctions for each colonies. So I would have tons of PCR reactions to run--I do not know how to use the multichannel pipett yet... I am thinking of shrinking the size of candidate pool for colony-PCR first....

---

each time before the transformation, I cleaned up the DNA using columns--so hopefully, the cleanup process get rid of the inhibitory stuff in the buffers....

and when I setup the ligation mix 2 days ago--I used 300ng of vector DNA in a total volume of 34ul (since I need to add another three pieces and 2X ligation buffer)--My ligation mix seems to be a large volume. In theory, the amount of vector DNA in 1ul of such mix would be around 10ng.


The efficiency of Maxi competent cells seem high--but I am just wondering why so few colones showed up on the Erm plate....Is the possibility of correct ligation of three pieces that low?


thanks for attention and discussion!

---

Yes, the efficiency of a 3 way is lower then 2 way ligation. But not to this low! There should still be a reasonable number of colonies... more then you can test anyway.

If the 9 colonies on the erythromycin plate don't turn out to be the right thing,

I would redo the ligation reaction. The ligation mix is, as you mentioned on the diluted side of things. Also I have a feeling that there isn't enough vector DNA for this large a ligation mix

Assuming it is too much trouble to reconcentrate the inserts, I would suggest that you follow phage434's recommendation and move the vector:inserts ratio to 1:1.

I would certainty transform more then 1ul, maybe around 3ul or 3.5ul (assuming this doesn't cause arcing... )

Then select on a chloamphenicol plus erythromycin plates. I believe that there is no interaction between the two antibiotics, but please check with your local postdoc/supervisor. The double selection would take care of any stray unligated vector and stray ermR only plasmids.


Ooo, and thanks Phage434 i've printed the protocol and I'll give it a try. Anything to improve the sad lot of cells I have now.

PS: The size of a colony PCR reaction mix has a total volume of 10ul (My ratio is 7.5ul PCR mix + 2.5ul cell solution from 50ul).... It doesn't need to be any larger.

As for the multichannel, play with it. Find out how it work. Move water from place to place.

---

oh, oh, your words made me worry now although I am still waiting for the growth on the new antibiotic-containing plates yet....I do not care how many positive hits I got--only one will be good enough!

One concern on the double-antibiotic plate would be--I heard from my boss that Erm is not effective in LB --that is why we tend to use BHI+Erm (the reasons I cannot recall now....). So I am not sure if it is possible to make BHI+Erm+Cm or LB+Cm+Erm plates


If it turned out that I failed to get any desired ones, I will definitely re-do the ligation, following you guys' advice....By the way, I am using chemical heat-shock transformation protocol--no worry about arching so far....


thanks...

Paula

---

---only one will be good enough

One of my labmates shares the same out look, "You only need one."

chemical trans... well use more DNA then..

Also I would suggest that the ligated DNA be concentrated into a small volume <10ul. Use this concentrated DNA, rather then the 1ul of 34ul for transformation.

EtOH percipitate the DNA, then wash with 100ul DNA free 70% EtOH, air dry and resuspend DNA in sterile distilled water, (no more then 10ul, maybe less).

I do my ligation in 0.5ml centrifuge tubes. The tube's small size makes it easier to see the DNA, as it is less smeared.


Fishersci sells BHI + Cm plates. So that combo is doable. So perhaps give BHI + Cm + Erm a try.

---