Q: How doing partial RE digests?

Sanjay Shirke sms1894 at is.nyu.edu
Sun Feb 2 13:32:40 EST 1997

Partial Digests require a large amount of time, and involve using large
amounts of enzyme and DNA, but they can work.

First, you have to ensure that you will be able to gel purify the desired
partially digested fragment from other fragments.  This is simple, you
just need to make sure that the species produced by digestion at the
desired site is of different length than the species produced by digestion
at the undesired site.

Basically, set up a large (~100 ul) digest, taking scrupulous care to be
very precise about the amounts of every reagent used.  Put all reagents
together in a tube, and pre heat the tube to 30 degrees C (or whatever).
Once preheated, add the enzyme, and mix with your pipette.  Take 10ul
(more or less) aliquots out every 10 minutes or so (remember that the
theoretical rate of digestion is 1 unit of enzyme digests 1 microgram of
plasmid in 1 hour.) When you remove the aliquots, add some EDTA (to final
conc. of 10-50 umol) and place them on ice.  Then, run all aliquots on a
gel, and determine at which time the desired fragment is most strongly

	At this point, you can either "scale up" and make a digestion
corresponding to the best time point which you've discovered, or you can
be lazy and use the aliquots you've already produced.  My personal
preference is to make a large second digest at the optimized conditions,
because this makes it less likely that you'll run out of insert and have
to repeat the whole exercise.

	I should point out that a partial digest with two sites will
produce four species: uncut, cut at one site, cut at the other site, and
cut at both sites.  Thus, you're starting with a base 25% yield, and
trying to increase that by the optimization above.  The point is, be
prepared to waste a lot of enzyme and DNA. It may be prudent to start from
a midi or maxi prep.

Good luck,


 s a n j a y   m .   s h i r k e

Consider this: If it had been discovered in America instead of Germany,
"spatzle" would have been called "cheetos".

More information about the Methods mailing list