I have always found that doing a partial digest on plasmid or short linear DNA is diffcult if
you use the standard methods that play with digest time and enzyme concentration. The
problem is that if you do an analytical run to test for the optimum conditions, these rarely
scale up or repeat consistently when you try the "real digest".
I have been using a technique based on ethidium bromide intercalation in DNA masking
the recognition site of the enzyme and have found it works beautifully. I normally make a
fresh stock of ethidium bromide at 0.1ug/ul (0.1mg/ml) and add volumes between 0 and
5ul to a 20ul restriction digest containing around 100 - 500ng DNA. Add around 0.5 - 1
unit of enzyme (plus the correct incubation buffer) and incubate at the specified temperature
for 30 min. The final ethidium concentration range is between 0 to 25ug/ml. On an
agarose gel you should see a lovely graded digest going from complete to almost uncut.
This method scales up well for more DNA if needed, provided that you respect the molar
I have used this many times now to partial plasmids with between 2 (to get a linear cut at
one site) and 15 sites with no problem.
I normally do 2 x a chloroform extraction of the final digest to remove the ethidium prior
I hope that this is of some help.
Dr. M.A. Blight,
Institut de Génétique et Microbiologie,
CNRS URA 1354,
Université de Paris XI,
91405 Orsay cedex,
Tel: +33 1 69156699
Fax: +33 1 69157808
e-mail: blight at igmors.u-psud.fr