The following protocol is for TFIID binding assays involving mixtures of DNA fragments. The protocol is based on one first worked out by Peter Emanual. Peter’s original protocol called for immunoprecipitation of TFIID in 100 mM HGKEDP (100 mM KCl, 25 mM HEPES pH 7.6, 10% glycerol, 0.1 mM EDTA, 1 mM DTT, 0.1 mM PMSF) but subsequent individuals and Peter have used 100 mM HGKEMNDP (100 mM KCl, 25 mM HEPES pH 7.6, 10 % glycerol, 0.1 mM EDTA, 12.5 mM MgCl₂, 0.1% NP40, 1 mM DTT, 0.1 mM PMSF). [Note, I have taken the liberty of changing the names of the buffers from what you might find in other places in order to be consistent in comparing the two buffers. The "100 mM" that precedes the acronym indicates the concentration of KCl.] The two major differences between these two buffers is the presence of 12.5 mM MgCl₂ and 0.1% NP40 in HGKEMNDP. There has not been a systematic comparison of the two buffers but there could be several advantages to the HEMGN. MgCl₂ might stabilize the TFIID complex as it has been found to stabilize the yeast TFIID complex (Joe Reese, personal communication). The 0.1% NP40 might produce cleaner precipitates. Also, on the basis of work in the Pugh lab, we would anticipate that NP40 might help dissociate the TFIID dimer. High salt should be avoided in the immunoprecipitates as 500 mM KCl was reported to dissociate both TAF150 and TFIIA.

I will describe the protocol as it pertains to a single sample. It is important to use siliconized tubes at all stages in order to reduce nonspecific binding of DNA and proteins. I siliconize tubes by splashing a drop of Sigmacote over the inside surface of the microfuge tube, briefly microfuging the tube, removing the excess solution with a pipette, air drying the tube, and finally rinsing the tube a few times with water. Also note the speed of the centrifugation steps when collecting the protein G sepharose. When you are collecting protein G beads, run the centrifuge at a setting of 2 to avoid crushing the beads and collecting unrelated aggregates of protein. When transferring the protein G beads, trim the tip of a yellow pipette with a razor blade so the opening is about 2 mm in diameter.

Once you have immunoprecipitated TFIID, you will incubate this with a mixture of radiolabelled DNA fragments and cold HaeIII-cut E.coli DNA. In looking over the information from various students, it appears that two different DNA binding buffers have been used successfully by various individuals in the lab. Emanuel and Li used something called CGSBB (10 mM HEPES pH 7.6, 5 mM MgCl₂, 1 mM DTT, 90 mM KCl, 10% glycerol) and Sypes used something quite similar. Taxman used 100 mM HGKEMNDP (see above) supplemented with 10 ng/ul poly dGdC and 50 ng/ul BSA for a binding reaction totaling 50 ul. Both systems exhibited specific binding. However, I favor the buffer of Emanuel and Li because this is more similar to the conditions that I used for previous studies with crude TFIID.

The proportions of materials are based on Bipasha’s experiments, and the source of TFIID is a DEAE fraction. All previous individuals used a phosphocellulose fraction that is prepared from DEAE fraction. Bipasha’s SDS-PAGE analysis of the material derived from the DEAE fraction indicates that it is very clean and I recall that Peter had made a similar observation.

Before beginning remember: low speed centrifugations for collecting protein G beads and siliconized tubes for binding assays. Also note that PMSF and DTT are shortly before use of the solutions. PMSF degrades within a few hours in aqueous solutions.

Step 1. Immunoprecipitating TFIID from the 0.25M DEAE fraction.

Note that samples are kept cold at all steps unless otherwise stated.

1. Centrifuge 40 ul of DEAE fraction for 5 minutes at top speed in a cold microfuge. This precipitates insoluble aggregates that can interfere with the immunoprecipitation. Transfer the supernatant to a new siliconized tube.
2. Combine 20 ul of monoclonal antibody (TAF250 or TBP) with 40 ul DEAE fraction and incubate on ice for 1 to 2 hours.

Note that in this case, the monoclonal antibody is a cell culture supernatant from a hybridoma cell line. We have one preparation of ascites fluid against TBP that is derived from a mouse which was injected with the hybridoma. The concentration of antibody in the ascites fluid is approximately 100 times greater than what is found in the cell culture supernatant. Be certain you know what antibody you are using so you can use the appropriate amount.
3. Prepare 10 ul of packed protein G sepharose. Put 10 ul of 100 mM HGKEMNDP in a siliconized 1.7 ml tube and mark the level. Add 20 ul of a 50/50 slurry of protein G sepharose. Add a few hundred microliters of 100 mM HGKEMNDP, mix briefly and centrifuge for 2 minutes at a setting of 2. Add or subtract protein G sepharose so that the final packed volume is approximately 10 ul. (Note that this is in excess over what is required to precipitate the antibody.) Finally, wash the beads with 200 ul of 100 mM HGKEMNDP. Collect the beads with a low speed spin and remove the supernatant. Note the use of a low speed spins throughout the procedure when working with the protein G sepharose.
4. Add the DEAE-antibody mixture to the packed beads and agitate to resuspend the beads. Rock the mixture in the cold room for 1 hour (a rotating wheel may be better than the rocker).
5. Wash the beads 3 times with 1 ml portions of 100 mM HGKEMNDP. For each wash, suspend beads and agitate for a few minutes. Then centrifuge in the cold at a setting of 2 for 4 minutes. Carefully remove the supernatant and avoid discarding the beads.
6. Suspend the beads in 100 ul of 100 mM HGKEMNDP and transfer the solution with a clipped yellow tip to a new siliconized tube(s).

In scaled up experiments, this may be the appropriate time to separate the beads into different portions. Fully suspend the beads and distribute into desired aliquots.
7. Spin down the beads at a setting of 2 for 2 minutes and discard supernatant.
8. Wash beads 1 time with 1 ml CGSBB. Spin down the beads at a setting of 2 for 4 minutes. Discard all but a total volume of 20 ul of material (20 ul of beads plus overlying solution).

1. Premix 100,000 cpm of labelled DNA with 1 ug of Hae III-cut E. coli DNA in a total volume of 30 ul of CGSBB. If may be useful to have 2X CGSBB to make this mixture.  The labelled DNA is typically a mixture of differently sized fragments.  It is good to include a fragment that shouldn't bind and a fragment that should bind as negative and positive controls.
2. Add the 30 ul mixture of DNA to the beads and rock at room temperature for 2 hours.
3. Spin down the beads at a setting of 2 for 2 minutes and transfer 30 ul of the supernatant to a new tube. The supernatant will be the source of unbound control so don't throw it away.
4. Wash the beads 3 times with 1 ml portions of cold CGSBB. All washes and centrifugation steps are done in the cold. Add the CGSBB and mix for about 1 minute before centrifuging for 4 minutes at a setting of 2.
5. Suspend the beads in 200 ul of 50 mM Tris pH 7.9, 10 mM EDTA, 0.5% SDS, 10 ng/ul salmon DNA and add 170 ul of this buffer to the unbound material described above. Add 20 ug of proteinase K to each sample and digest for 30 minutes at 37^oC.
6. Extract with 300 ul of phenol. After centrifuging the sample, carefully transfer 200 ul of the aqueous phase to a fresh tube. Sacrifice residual aqueous material so as to avoid the cloudy interface.
7. Add 20 ul of 3 M Sodium Acetate pH 6-7 and 500 ul of ethanol. Precipitate as usual, wash with 75% ethanol and finally suspend the dry pellet in sequencing gel loading buffer.