3 piece ligation - (Oct/30/2008 )

I am trying to insert a ~400bp piece cut by DraI and DrdI into a plasmid. The piece is part of a gene where I created various mutations. The plasmid I want to put this in is an expression vector, has the rest of the gene and 2 tags. Unfortunately, the vector backbone has 3 DraI and 3 DrdI sites. The way I could think of getting the work done is as follows:

The gene in the host plasmid has 2 MunI sites surrounding the DraI and DrdI insertion. I cut out that MunI piece, put it in a tube and digest with DraI and DrdI. The pieces should be:
MunI - DrdI : 1175 bp
DrdI - DraI : 389 bp
DraI - MunI : 809 bp
Gel purify 1175 and 809 bp pieces, put them in a tube with the 389 bp piece where I have mutations. Ligate them and screen.

Alternatively, I use oligos from MunI sites and PCR amplify that mixture.

Will any of the above scenarios work? If not why? Is there another (simple) way? Are you aware of a vector that does not have DraI and DrdI sites (or is there an easy way I can screen available vectors)?

thanks

pTR

-pTR-

This is a very difficult strategy you are suggesting here and i don't think it would work. These are my issues in order of importance:

1. There will be too many undesirable ligation products formed. Nothing is stopping each fragment ligating with itself and the two MunI ends ligating with each other.
2. The DrdI overhangs - are the two DrdI overhangs going to ligate? I've never used one of these N-containing restriction enzymes to clone before, but as far as i can see, just because an N-containing RE may cut the two sites you are using, doesn't mean both sites will contain the same overhangs because the "N" may be different at each site. So your DrdI overhangs may not be the same and therefore not ligate.
3. 3-way ligations are difficult to begin with
4. The blunt DraI ends are going to reduce the efficiency of the ligation

You could use PCR to amplify your desired fragment from the ligation (i think that's what you were suggesting) but there would be so many annealing sites in undesired fragments and your desired product would be such a low percentage of the total products that i don't think the PCR would be successful. The PCR would probably produce a smear or heaps of undesired bands. You might get a product around the correct size but it may also be an undesirable combination of fragments.

I think your best option is to join the three fragments using multitple splice-overlap PCRs. I can go into more depth on that but i have a couple of questions:

1. Why did you chose that specific 389 bp to mutate? Was it just a random fragment that contained the region you wanted to mutate or was there are more specific reason?
2. Why did you choose the DraI and DrdI sites on the outside of your mutated fragment?

Hope i've helped,
Rob

-killerkoz17-

Hi Rob,

Thanks for the reply.

To answer your questions, the 389 bp fragment flanked by DraI and DrdI sites was chosen based on the fact that the remaining parts of the gene does not have these sites. Within the gene those were the best single cutter sites to give a reasonable size for PCR based mutagenesis.

The DraI-DrdI piece will go back into the same gene, so the DrdI overhangs will be the same.

I thought about an oligo-based mutagenesis strategy by PCR (I am not sure if your strategy is same as mine), but the gene itself is 6 kb. With the plasmid sequence, that might become too big for that method. I guess I could also subclone a larger pice to TOPO and then do the oligo-based mutagenesis (the plasmid will be smaller).

I am open to all suggestions here.

Thanks,

pTR

QUOTE (killerkoz17 @ Oct 30 2008, 10:59 PM)

-pTR-

QUOTE (pTR @ Nov 6 2008, 05:47 AM)

I think you need an oligo-based strategy to join the 389 bp mutated sequence into your gene. There are too many combinations of fragments that can be made using ligation at your sites and you would want to amplify your desired ligation products by PCR any way to get enough to ligate the whole gene back into the vector. I think you are implying that you would need to PCR the whole vector to get the mutated fragment/s into the vector, but you can just PCR the fragment between the MunI-MunI site and ligate that into the vector (although MunI and another single site would be more preferable). So what i think you need to do is cut out your fragments as planned (MunI, DrdI, DraI, MunI - or other site to MunI) and design some primers that will join the fragments together using PCR. Your aim should be to do three PCRs:

-PCR 1 - join fragments 1 and 2
-PCR 2 - join fragments 2 and 3
-PCR 3 - join products from PCR 1 and 2

You need the internal primers to overlap the DrdI and DraI sites. At each site (DrdI and DraI) you need a forward primer and a reverse primer that both anneal to the same sequence (i.e. the primers will be a reverse complement of one another) as well as a forward primer at the first MunI site and a reverse primer at the second MunI site (so 6 primers in total). I don't know your experience in this area but if this type of joining fragments is foreign, draw it on a piece of paper and see how it works. I call it splice-overlap PCR but there are other names.

In the future, you need to check the sites you plan to use for chopping in and out of your clone are also not present in your vector, as well as the insert.

Hope that made sense.

Good luck, Rob

-killerkoz17-

Thanks, the PCR strategy you suggest might work. I will try.
And yes, next time I will check the host in addition to the insert for restriction sites.

-pTR-