Subcloning oligos into expression vectors

Kevin Mulcahy K.Mulcahy at
Mon Jan 20 09:36:15 EST 1997

David H. Gorski wrote:
> Second, in my previous cloning attempts, after cutting the vector with Bgl
> II and Nhe I serially and gel-purifying the fragment, I find that I'm
> getting a lot of self-ligation, as evidenced by lots and lots of colonies
> on my vector-only control plate. At least according to every restriction
> site list I read, Bgl II and Nhe I should not be compatible, but tell that
> to the vector. I've already verified that my enzymes are good by test
> digestions with each enzyme (I can't see the fragment that a double
> digestion should produce because it's only about a dozen nucleotides
> between the sites). Has anyone ever seen this or heard of it before? It
> looks like we're going to have to change our entire subcloning strategy if
> I'm going to get self-ligation between supposedly "incompatible" sites.

Dear David,

it is possible that you are getting incomplete digestion of the vector since the 
Nhe I and Bgl II sites are very close together. For efficient cutting some enzymes 
require quite a bit of flanking DNA sequence so that the enzyme molecule can 
attach and recognise the actual cut site. Therefore, if for example Nhe I requires 
at least 10 bp either side of its actual recognition sequence (i.e. not including 
the recognition sequence GCTAGC) but the length of intervening flanking sequence 
from the Nhe I cut site to the Bgl II cut site is only 8 bp then any vector 
molecules previously cut with Bgl II may not be subsequently cut with Nhe I. You 
would therefore end up with many vector molecules which were just singly cut with 
Bgl II which can therefore self-ligate.

You may therefore have to use much more of each enzyme and incubate for much 
longer in order to ensure sufficient digestion by both enzymes. If you still get 
self-ligation (indicating incomplete digestion) you could try CIAP-treating the 
vector molecules such that self-ligation cannot occur. In this way, vector 
molecules will only re-ligate if they contain your insert. If this still doesn't 
work then you may have to try different restriction enzyme sites which require 
much smaller flanking sequences.

Hope this helps!


Kevin Mulcahy.

More information about the Cellbiol mailing list