Subcloning problems

Vladimir Svetlov svetlov at oncology.wisc.edu
Wed Mar 18 10:49:11 EST 1998


In article <01bd5256$bcd3bca0$dc5dfec2 at bio.univ-paris5.fr>, "Cynthia
Marie-Claire" <marie at pharmacie.univ-paris5.fr> wrote:


> I would like to submit you a problem. I need to subclone my favourite
> protein in an expression vector. it's basically a classic Nco1/EcoR1
> digestion of the cloning vector (A) and purification on agarose gel of my
> insert and then ligation in the expression vector (B) digested with the
> same enzymes and purified on agarose.

OK, how do you elute that? In using DE81 paper I block the "upstream"
fragments by inserting an additional piece of DE81 just above the band of
interest. Visually it appears to block the bigger chanks of DNA from
clamping onto the band I'm after. Run longer gels and cut out yer piece of
gel carefully, don't overload the wells and make sure they don't have a
hole in the bottom - both can cause your DNA to float above or below the
gel and contaminate the prep. 

> The problem is that 
> 1) I have quite the same number of colonies in my ligation and the negative
> control. I've screened the dishes by hybidization and pick the positive
> colonies. 

Make sure you digest is complete. In this case even you parent vector is
not supposed to religate. I dephosphorylate all my cloning vectors with SAP
right in the digestion mix. No background from cloning. Of course if some
vector goes uncut there is not much phospatase can do to it.

> 2) Among those 70% contain the cloning vector (A) and only 30% the
> expression vector (B). Unfortunately none of these had the insert.

Now, how'd you get that?! If your cloning/parent vector contained insert at
NcoI and EcoRI sites, it can't religate without the insert (these enzymes
do not produce compatible with each other ends). The only thing that can
make it to do so is the presence of an exonuclease that blunts the cohesive
ends (polymerases can do that too but nucleases are easier to come by). Or
you've simply pissed off the Cloning Fairy, and then nothing is gonna help
ya. Some commercial endinucleases have a substantial contaminations with
exo's and are not recommended for a prolonged or over- digestion. Make sure
that you are using some good, German-engineered enzymes from Boehringer
Mannheim. Hey, and quit using Perrier for the ligations, ddH2O would do
just fine. 
Hope this helps.
Regards,
V.

-- 
Vladimir Svetlov
McArdle Lab for Cancer Research
Dept. Oncology
UW-Madison
1400 University Ave.
Madison, WI 53706



More information about the Methods mailing list