Purifying small DNA from polyacrylamide gel

tfitzwater at gilead.com tfitzwater at gilead.com
Fri Nov 10 17:17:27 EST 2000

>Zhongtang Yu (zyu at interchange.ubc.ca)
>Wed 08 Nov 2000 - 22:35:07 GMT >I want to purify small dsDNA (40-80 bp)
from 12% native polyacrylamide gel. I want to
>recover as much as possible. Does anyone have simple methods to do that?
What's the
>recovery? Thanks.

>Zhongtang Yu, Ph.D. Dept. of Microbiology, UBC

Gel Purification of Nucleic Acids

This is a general method for purifying all types of small nucleic acids
from acrylamide gels.  In my hands, this method yields 85-94% recovery of
RNA, ssDNA or dsDNA.  Electroelution allows recovery of an additional 1% of
the nucleic acids.

1.   DNase I-treated RNA transcription reaction should be concentrated by
ethanol precipitation or chloroform extraction (if transcription buffer
contains PEG or HEPES) followed by use of a Microcon 30K cartridge to
reduce the volume to about 30 µL. A concentrated 500 µL transcription can
be loaded onto a 1 inch-wide well of a denaturing 8% gel as described

Crude desalted oligonucleotide synthesis product is resuspended in 400 µL
of Type I water and the concentration is determined by OD260, preferably
using the extinction coefficient, but otherwise using 33 ug ssDNA/mL = 1
OD260 unit.  Alternatively, if the synthesis yield in pmoles is known,
resuspend to 1 nmole/µL.  Oligonucleotides > 40 nucleotides long are
purified by gel electrophoresis on a denaturing 8% acrylamide, 8 M urea
(leave out urea if purifying dsDNA), 1x TBE slab gel with wells that are
0.75 mm thick and 1 inch wide.  Primers and truncated ligands < 40
nucleotides in length are purified on denaturing 12% gels.  Kinased
material or other smaller amounts of nucleic acid may be purified in 1-cm
width wells.

Thirty nanomoles of desalted synthetic oligonucleotide is mixed with 2x
formamide gel loading buffer lacking xylene cyanol and bromphenol blue
because these dyes generate UV shadows that interfere with band
identification.   Bromphenol blue is omitted whenever bands in the
neighborhood of 20-30 nucleotides are important.  The oligonucleotide
mixture is heated at 95°C for 3 minutes, quickly cooled on ice, briefly
centrifuged and loaded.  Five microliters of 1x formamide gel loading
buffer containing xylene cyanol and bromphenol blue are added to one corner
of the sample well as a running marker.  When loading dsDNA, mix the DNA
with the appropriate volume of gel loading buffer and heat inactivate
restriction enzymes at 65°C if appropriate before loading the gel.  The
denaturing gel is run at constant wattage sufficient to hold the gel at
50-55°C until the Orange G dye is near the bottom of the gel (typically 18
watts when the gel plates are not thermally equilibrated in a buffer-backed
gel box).  Native gels for dsDNA are run at a constant 200 volts.  Gels
containing non-radioactive samples are wrapped in Saran Wrap in preparation
for UV-shadowing.  It is helpful to gently blot the wells with a Kimwipe to
remove excess buffer prior to transfer.  Residual buffer allows the gel to
slide around inside the Saran Wrap making registration imprecise.  In
addition, when transferring radiolabeled gels to GelBond PAG, this residual
buffer interferes with the transfer of the gel to the hydrophilic support.
When loading > 100 µCi of 32P on a gel, the use of a lead shield between
the gel box and ½ inch Plexiglas is recommended in order to reduce the
production of bremsstrahlung.

Place the Saran wrapped gel on a 254 nm TLC plate (Merck Art. 11798
Kieselgel 60 F254) and use a hand-held UV (254 nm epi-illumination) to
locate the nucleic acid bands by UV shadow.  The dye(s) in the loading
buffer will also generate UV shadows.  Mark the full-length UV shadow bands
on the reverse side of the gel with a thin line marker and cover the band
with labeling tape.  Flip the gel over to the correct loaded orientation
and use a #11 scalpel blade to cut out the desired bands.  Marking the back
of the gel allows you to still see your lines and taping it prevents
harvesting the bottom layer of Saran.  Care must be taken to avoid the
lower third of the UV shadow band of oligo synthesis products if it is
noticeably more transparent, as this contains significant amounts of
shorter sequences (n-1, etc.).  The UV shadow can be photographed with a
Polaroid MP-4  camera using Polaroid Type 667 film and a Wratten 15 deep
yellow filter at f/5.6 with two 1/2 second exposures using the hand-held UV
lamp at a height of about 12 inches from the left and right sides of the

2.   Passive elution:  transfer the gel pieces with the back of the blade
in 5 mm x 10 mm pieces to microfuge tubes containing Elution Solution
composed of 0.3 M sodium acetate, pH 7.5-7.8 and 2 mM Na2EDTA.  0.75 mm
thick bands from 1-cm wells elute in 330 µL and bands from 1-inch wells
elute in 1 ml of Elution Solution with 94% and 85% efficiency,
respectively, due to the fact that about 5% of the nucleic acid is lost on
the walls of each siliconized microfuge tube during ethanol precipitation.
Nucleic acid is passively eluted at least 4 hours (the elution is finished
in 4 hours, but we usually go overnight for convenience) on a room
temperature rotator.  After elution, the tubes are spun briefly in a
microfuge, the aqueous contents are carefully pipetted off and ethanol
precipitated as described below.  If the gel slice fragmented during
elution (a problem when you use too much TEMED), transfer the sample to a
Spin-X cartridge and spin briefly in a picofuge as described below.
Passive elution works for 6-15% acrylamide gels.  Bands from higher
concentration gels should be extracted by a crush/soak technique.  Passive
elution is recommended for radioactive sample in order to reduce exposure.

3.   Crush/soak:  the gel slices are placed in microfuge tubes, frozen in
dry ice-ethanol for 5 minutes and thawed (optional, this doesn't really do
anything but it makes some people feel more comfortable because that's the
way they learned it), 200 µL of Elution Solution is added, and the gel
slice is crushed with a microfuge tube pestle (Kontes, Vineland, NJ, but
available from VWR or Fisher) or with the plunger from a 1 mL disposable
syringe (which takes more practice) until a fine slurry is formed.
B.Beutel and L. Gold, J. Mol. Biol. 228, 803 (1992).  Care must be taken to
avoid splashing during this procedure, particularly if the sample is
radioactive.  The sample is diluted with an additional 400 µL of elution
solution, the gel slurry is spun through a 0.45-µM sterile cellulose
acetate Spin-X filter (Corning Costar) or pushed through a more risky 25 mm
diameter cellulose acetate syringe filter (Millipore, Bedford, MA) to
remove gel particles.  The slurry may be rinsed with a small volume of
elution solution.  The eluted material is then ethanol precipitated.

4.   Ethanol precipitation may be performed in the manner to which the
investigator has become accustomed.  My particular method is listed below.
Ethanol precipitate the eluted oligonucleotides in 330 µL aliquots in tubes
containing 3 µL of 0.5% (w/v) linear polyacrylamide carrier.   C. Gaillard
and F. Strauss, Nucleic Acids Res. 18, 378 (1990). and 2.5 volumes of
ethanol.  Incubate the mixture at -20°C in a pre-chilled rack for 20-60
minutes, transfer the tube to dry ice/ethanol for 20 minutes and spin in a
microfuge at maximum speed for 12 minutes.  Carefully aspirate the
supernatant.  Wash the pellet once with 950 µL of room temperature 70%
ethanol.  Chill the tube in dry ice/ethanol for 6 minutes and centrifuge
for 3 minutes.  Carefully aspirate the supernatant and spin the pellet to
dryness in a SpeedVac (Savant Instruments, Inc.; Farmingdale, NY).
Resuspend the pellets in 25 µL of water, pool the fractions and determine
the OD260.

Tim Fitzwater
Principal Research Associate
Gilead Sciences


More information about the Methods mailing list