Blunt lig. help (NL)

Danuta Bois dbois at
Fri Sep 29 13:23:30 EST 1995

HAVILAND at KIDS.WUSTL.EDU (David L. Haviland, Ph.D.) wrote:

>> A method for cloning PCR products by blunt-end ligations was published
>> in BioTechniques last year.  It works very well for me. It really is
>> quite simple and fast.  No buffer changes, either! 
>>  Here's the reference:  Kai Wang, Ben F. Koop and Leroy Hood  "A
>> Simple Method Using T4 DNA Polymerase to Clone Polymerase Chain
>> Reaction Products", BioTechniques, vol. 17, no.2, pp.236-237, 1994.

>I wholeheartedly "second" Danuta's suggestion.  This technique is one of 
>the very few that worked the first time for me.  However, I have found some 
>perhaps obvious limitations.  First, from my experience, the PCR reaction 
>must be a pristine reaction - that being, no spurious bands or smears.  
>If it isn't a pristine reaction, you have a whole slurry of ligatable 
>"stuff" most of which isn't what you want.  Only on optimized PCR reactions 
>with defined products has this method worked for me. 

If I have a PCR reaction with a product I want to clone and some
background smear, I still clone the whole thing and then screen
colonies with the original primers (or internal primers, if you have

 If there are too many colonies to check individually, I pool them and
do the PCR on each pool to check if there is anything in there that I
want, then I may narrow it down to individual colonies.  I collect the
clones that generate the PCR fragment that I'm interested in and throw
away the rest. 

To verify that you've got the right fragment, you may compare
restriction digest of the original PCR fragment with the one from the

When checking sequences of such clones, I found that sometimes other
small fragments like primer-dimers, and the like,  may also be
inserted between your PCR fragment and the vector.  I guess that means
that the insert:vector ratio was too high.  Sometimes it is hard to

 Second, I have not 
>been able to succeed in ligating a purified fragment in which I took a 
>portion of the fragment and added all the reagents, including Taq buffer 
>and enzyme, and go through the protocol as if the fragment had been PCR'ed. 
> This latter point is one that I would like to overcome as it would help 
>many a ligation.

>On the PCR (actual product) where this protocol worked, it worked like 
>greased lightning.  I took a sample of the kinased product - per the 
>protocol and basically dropped it into pBluescript cut with EcoRV which had 
>been alk-phos treated.  It was one of the easier ligations that I had ever 
>conducted recently.

>I hope this helps,

More information about the Methods mailing list