de-phosphorylating a vector - (Feb/22/2005 )

Hi all,
I am cloning my cDNA into a vector with one end being a sticky EcoRI end, and blunting the other end which will originally be cut with SalI.

As these ends Shouldn't re-ligate, should I still de-phosphorylate the vector? Do you de-phosphorylate so the vector will not re-ligate, or so the vector and insert WILL ligate? I am a little confused

Thanks

---

hi
i don't think it's necessary to dephosphorylate your vector due to the fact you have uncompatible ends. But if you want to dephosphorylate, and increase the specificity of religation, don't forget to phosphorylate your cDNA.

Hope that helps.

Fred

---

Hi Nay.
This has been discussed recently, Here's the link
Cheers!

---

Well in theory the vector will still be able to religate. I am a bit bevildered to see that ppl say that having either two different sticky ends or one sticky plus one blunt, could not result in vector ligation. This is infact not true, since TWO vectors would be able to religate, i.e. the two vectors would "twist" into ligating to each other. Infact this happens quite often.

The dephosporylation of the vector is important in some cases but not always required - I will however suggest you always to do this, and only skip the step if you have no positive colonies growing. Also the dephosporylation is pretty easy, since it can actually be done in the restriction buffer - if you use CIP you will have to add a little bit of zinc ions first though (I usually add a bit of the CIP 10x buffer to my restriction reaction).

Okay - and to your last question. The ligation requires a phosphoryl group in order to ligate two pieces of DNA (just think of the backbone structure of DNA). If you where to dephosphorylate the vector these missing phosphate groups would be missing and (in theory) vector religation or vector-vector ligation shouldn't occur.

So when you do the actual ligation reaction with product + vector, your product would provide 2 phosphate groups (one for each strand), but since your vector is not able to provide two as well, you will actually get "single-stranded" DNA ligation - i.e. your ligation product would have two nicks - one in each end. Usually this is not a problem however, since the bug machinery will quickly fix this when you transform them. However, probably due to this fact, you will get fewer colonies when dephosphorylating your vector.....

Also don't waste time on trying to digest your ligation product to see if you have the right thing - it seldom works. Just stick the thing into your bugs, pick 3-20 colonies and do a digestion (i.e. don't waste time on a PCR either - they result in billions of false positives).

---

I'm not sure this is so usual. In seven long years doing ligations, it never happened to me. I would believe than a high-MW plasmid with two replication origins would be unstable and/or easily lost by bacteria.
Cheers!

---

hmm. interesting post. I didn't know that dephosphorylating would lead to a doubly nicked DNA. I just acquired some of NEB's new Antarctic Phosphatase (for free) and tried it out.

My regular transformation protocol didn't give me any colonies.

Granted, my final construct size is around 18kB, which would surely make my transformation efficiency lower. I spun my transformation culture and resuspended in less SOC, to make density higher. Hopefully, I can get some colonies out of it.

---

Hmmm this is a bit odd - are you sure you have never seen double plasmids badcell? If I have to give a rough estimate I see this around 1:10....

---

happened to me!
On one plate, had the entire bacterial population take on the multiple vector plasmid, and not the insert: vector plasmid (at all!!!!). Oh, the things i've screwed up.

---