Best way to get DNA from gel bands?

Hiranya Roychowdhury hroychow at NMSU.EDU
Tue Jul 25 10:46:31 EST 1995

On 24 Jul 1995, Allen Gathman wrote:

> We have been extracting DNA from bands in agarose gels by electroeluting
> into a trough in the gel, filled with 15% PEG.  While this works
> reasonably well for many people in the lab, it requires some manual
> dexterity to cut the troughs, and recovery is not always very good from
> the PEG solution.
> I know there are numerous kits, spin columns, etc. for this purpose.
> What is your experience with these?  What's the best (and reasonably
> cheap) method for getting DNA out of a band in a gel?
> ***********************************
> |    Allen Gathman                |
> |    Biology Department           |
> |    Southeast Missouri St. U.    |
> |    Cape Girardeau, MO 63701     |
> ***********************************

DEAE-membrane - reasonably priced, lasts a long time and is probably the 
most gentle of all. Very large (>7kbp) DNA is difficult to recover 
completely due to the very principle involved. I have successfully 
recovered up to 60% of a 7kbp fragment by increasing the ionic conc and 
by heating. 

Glass fines - cheap, but is bad on very large pieces and very small 
pieces. I am not particularly fond of this one.

Electroelution - best recovery and least damaging, depending on the 
method being followed and the dexterity of the worker, of course.
It is probably the only method of choice for very small fragments. 

Squeeze-through - very good for very large sized DNA and probably works 
better than any for the small fragments. I have routinely extracted phage 

DNA from gels by this method with recovery of up to 80%. However, one 
needs to clean the DNA further after the elution.

I have used all of the above, and I presently stick to the first and the 
last methods because of their ease and speed.

			  Hiranya S. Roychowdhury
   			  Plant Genetic Engineering Lab.
			  Box 3GL, NM State Univ.
			  Las Cruces, NM 88003
			  Phone: (505) 646-5785
			  hroychow at

More information about the Methods mailing list