General principles for gold tagging a protein
The method that has previously been used to gold-label proteins at
the molecular level uses the following strategy (RecA protein bound to DNA
will be used as an example). A reaction mixture is made that will produce a
high yield of DNA-RecA filaments. If such a reaction is now prepared by
conventional techniques for EM, the filaments can be seen to be complete and often
show repeated striations along the complex (0007a.jpg). Once the filaments have been prepared, they
can be purified away from unreacted materials or, if one is willing to put up
with a somewhat dirty background, the gold tagging can be carried out on a
crude reaction mixture. Antibody to RecA (raised in rabbit) is added followed
by protein A-gold with the final result that gold spheres are now coupled with
RecA and due to the high contrast of the gold, their location is unmistakable.
The experiment can be performed in other ways; for instance the protein A-gold
conjugate could be reacted with the antibody, which in turn could be reacted
with RecA monomers before the filaments are formed.
Whichever way is chosen, the procedure, as outlined above, requires
compromises to the optimal concentrations of the RecA, antibodies and the
protein A-gold because of the persistent aggregating tendency of the antibodies
in these reactions. Although concentrations can usually be found that work
around this problem, it is a time consuming process to find these conditions
and, in our hands, different proteins require different compromises in these
concentrations.
New modified technique.
As outlined above, the conventional gold tagging reaction is performed in true
solution and as a result there is always a likelihood of aggregation. Our
modification avoids this problem by simply performing the reactions on
the protein-DNA complex while it is immobilized on the carbon film used to
support the molecules during visualization in the electron microscope. In
brief, the reaction mixture, is prepared in exactly the same way as if the
sample is to be stained or shadowed for electron microscopy, but once the
sample is absorbed onto a carbon film it is first gold tagged before staining
or shadowing by simply floating the carbon films on drops of the various
reagents. Because the DNA-protein complex is immobilized on a surface before
the addition of the antibody, the aggregation problem is avoided.
Details of this procedure are given below, with gold
labeling of RecA coated single stranded DNA molecules as an example.
(1) RecA filaments are made by incubating the following reaction mixture.
Water 10.38 ul
4xHepes (4xHepes at pH 7.2 : Hepes 100mM, NaCl 800mM, Mg Acetate 40mM, K glutamate 12 mM) 6.25 ul
Phosphocreatine (150 mM) 2.00 ul
Creatine phosphokinase (0.2u/ul) 1.25 ul
ssDNA (316 uM) 1.65 ul
RecA (190 uM) 0.92 ul
Incubate 10m 37C
ATP (91.6 uM) 1.65 ul
SSB (55.6 uM) 0.90 ul
Incubate 20m 37C. Total volume during incubation is 25 ul
ATPgammaS (0.1M) 0.65 ul
Incubate 10m RT
Glutaraldehyde (7% in 1xHepes) 3.00 ul
Incubate 10m RT then 20m on ice
The final concentrations of ssDNA, RecA and SSB are 21uM, 7uM and 2uM respectively.
This solution can be diluted 1:100 in 20 mM NaCl - 5 mM EDTA and used
immediately or dialyzed against the same salt solution and is then stable for
at least 10 days. This dialyzed solution is also diluted 1:100 as above before
continuing with the next step.
(2) Prepare carbon coated grids.
Carbon is evaporated onto freshly cleaved mica and the film floated off
the mica at an air-water interface and picked up from below on EM
grids. We store these carbon coated grids in a desiccator and immediately before use they are washed in
chloroform followed by drying under a heat lamp.
(3) Adsorb sample to carbon film.
Float the carbon coated grid (carbon side down) for 1 minute on 10 ul
of the diluted sample described above. The drop should be placed on a
clean hydrophobic surface such as a teflon block. The carbon film is now
washed by floating on a 100 ul drop of 50 mM ammonium acetate for
15-20 min. The sample can then be used immediately as described below or it
can be dried by holding the grid in a vertical position and blotting the bottom edge against filter paper and dried under a heat lamp; such samples can then be used at a later time if they are stored in a desiccator.
NOTE: In all the steps described below, the carbon film is never allowed to
dry out between the various steps: transfer between the solutions was done as quickly as possible.
(4) Attach RecA antibody to the adsorbed RecA filament.
Float the carbon film (now containing adsorbed RecA filaments) on a 50 ul drop
of 0.1% BSA in PBS (150 mM NaCl, 8.3 mM Na2HPO4, 1.85 mM NaH2PO4). Float and
shake (see note below) on a 20 ul drop of the antibody solution (diluted
1:5000), This polyclonal antibody was raised in rabbit and contained about 1.9
mg/ml total antibody; probably only 10% of this being directed against RecA
protein.
(5) Attach the protein A-gold conjugate to the antibody-RecA-DNA filament.
Float on a 50 ul drop of 0.1% BSA in PBS for 3 min as before. Float and shake (see note
below) on a 20 ul drop of protein A-gold diluted 1:20 in 0.1% BSA in PBS for
20 min. The protein A-gold was purchased from Sigma (0.031 mg protein A/ml,
10nm Gold). Float again on a 50 ul drop of 0.1% BSA in PBS for 3 min.
(6) Staining (and/or shadowing).
Float for 2 min on two successive 100 ul water drops. Touch to a 20 ul drop of 5% uranyl acetate
for 1 sec. Float on a second drop of uranyl acetate for 25 sec. Touch the film to 6 successive 100 ul drops of water, each for 1 sec. Withdraw water from the grid using a capillary pipette. If needed the sample can now be
shadowed in the conventional way.
NOTE on Shaking: During the incubation with antibody and also with the
protein A-gold, we try to have good movement of the liquid across the
immobilized RecA-DNA filaments. The liquid drop, on which is floating the EM
grid containing the RecA filaments immobilized on the carbon film, is on a
teflon block and this is held against a vibrator. We set the vibrator to a
level which just causes a visible low frequency vibration of the EM grid with
respect to the drop.
NOTE on crosslinking: In this protocol, at the end of the incubation
that produces filaments (see (1) above), the protein-DNA complex is crosslinked
with glutaraldehyde. Depending on the particular protein involved,
crosslinking may not be needed to stabilize the structure. In these cases we
omit the crosslinking step and add glycerol (5-15%) to all solutions mentioned
in steps (2) and (3) above.
Examples of the procedure:
The electron micrographs shown in 00009a.jpg and 0009c.jpg are RecA-ssDNA filaments which have been
treated as described above, the gold label can be seen as very dense 10nm
spheres that are preferentially bound to the filaments. An unlabelled recA
filament is shown in 0007a.jpg for comparison. The
repeat structure seen along this filament is not as easily observed when the
structure is gold labelled because the bound antibody tends to obscure the
fine structure.
When single stranded linear DNA is reacted with RecA and SSB using the
incubation conditions described in (1) above, a very high proportion of the
filaments have one end that is free from RecA as shown in
0027b.jpg or 0027c.jpg. When such filaments are
treated as described above, using antibody against SSB, then it can be shown
that the end not coated with RecA is labelled with gold and therefore contains
SSB ( 0027d.jpg or 0027e.jpg).
Return to Electron Micrograph Library
For questions about application of this technique, or improvements you might
like to have added to this protocol, contact
mschnos@facstaff.wisc.edu
