Purification of recombinant TAF-172. (A) Sf9 insect cells were infected with recombinant baculovirus containing the TAF-172 gene and six in-frame histidine codons at the amino terminus. Equal proportions of pooled fractions generated at each stage of the purification process were analyzed on an SDS–6% polyacrylamide gel in which the proteins were stained with silver. Molecular weight markers (M) are shown in lane 1. The crude cell lysate (CE; lane 2), the Ni²⁺ column elution (lane 3), the Sephacryl S300 gel filtration pool (lane 4), and the S Sepharose pool (lane 5) are shown. (B) Western blot of TAF-172. HeLa nuclear extract (NE; 40 μg, lane 1), immunopurified, P.3-derived TAF-172 (20 ng, lane 2), recombinant (rec.) TAF-172 (25 ng, lane 3), and a B-TFIID containing Superdex 200 PG fraction (39) (1.2 μg, lane 4) were electrophoresed on an SDS–6% polyacrylamide gel. Proteins were electroblotted to nitrocellulose and probed with affinity-purified TAF-172 antibodies.

Recombinant TAF-172 binds to TBP. TAF-172 (10 μg) was incubated alone (lane 3) or with 2 μg of GST fusions containing either amino-terminal residues 1 to 163 (lane 4, GST-TBP-N) or carboxyl-terminal residues 168 to 339 (lane 5, GST-TBP-C) and glutathione agarose as described in Materials and Methods. The resin was washed, and eluted proteins were subjected to SDS–6% polyacrylamide gel electrophoresis and analyzed for TAF-172 by Western blotting. Recombinant TAF-172 (50 ng) is present in lane 1 and represents 0.5% of the input. Molecular mass markers (M) are in lane 2.

TAF-172 is a TBP-stimulated DNA-dependent ATPase. (A and B) TAF-172 was assayed for ATPase activity in the presence or absence of TBP and a 361-bp G₆TI promoter DNA fragment, as indicated. Background rates of ATP hydrolysis determined in parallel reactions lacking TAF-172 were subtracted from the data; standard errors were summed. Background rates as a percentage of the rate determined in the presence of TAF-172 are indicated (% bkd).

TAF-172 does not inhibit TBP-TAF-promoted in vitro transcription in the presence or absence of TFIIA. (A) HeLa nuclear extracts (100 μg in lanes 1 to 10 and 80 μg in lanes 11 to 15) containing endogenous TAF-172 (endog. 172) as indicated, were used to transcribe 5 nM TI (lanes 1 to 5), G₆TI (lanes 6 to 10), or VA₁ (lanes 11 to 15) DNA in the presence of increasing concentrations of recombinant TAF-172, as indicated. Time of exposure to the phosphor screen was varied to obtain similar signal intensities among the different promoters. (B) HeLa nuclear extracts (120 μg) immunodepleted with TFIIA serum (lane 1) or an unrelated control (ctrl.) serum (lane 2) were electrophoresed on an SDS–6% polyacrylamide gel, electroblotted to nitrocellulose, and probed for the α subunit of TFIIA. (C) HeLa nuclear extracts (100 μg) depleted of TFIIA, TAF-172, or TFIIA and TAF-172 were used to transcribe 5 nM G₆TI (lanes 1 to 4) or VA₁ (lanes 5 to 8) DNA.

Possible ATPase cycle for TAF-172 and Mot1. ATP hydrolysis leads to either dissociation of TBP and TAF-172 from DNA (reaction 1) or just dissociation of TAF-172 (reaction 2). Note that the ATP-dependent dissociation of TAF-172 is not the reverse of the association reaction, in that ATP is not generated. Additional details are provided in the text.

(A) Alignment of yeast Mot1, human TAF-172, and Drosophila 89B helicase. Protein sequences were aligned by using the CLUSTAL W sequence alignment program. Identical amino acids have a black background, and conserved (I-L-M-V, T-S, D-E, Q-N, K-R-H, Y-W-F, and G-A) amino acids have a gray background. Peptide sequences obtained by internal microsequencing are underlined. Previously identified TPR motifs are indicated above the Mot1 sequence, and corresponding TAF-172 regions are indicated below the TAF-172 sequence. A black dot indicates a conserved fit to the TPR consensus, an open circle indicates no match, and a dash indicates no consensus. The sequence of TAF-172 is identical to that of TAF_II170 (41), except at position 945, which is a C in TAF_II170 and an S in TAF-172 (accession no. AF038362) and two GenBank expressed sequence tags of TAF-172 (accession no. R07413 and T78264). (B) Schematic alignment of Drosophila 89B helicase, human TAF-172, yeast Mot1, and human Snf2a. Gray blocks depict regions of high sequence similarity. The black box represents the putative ATPase-helicase domain found in all Snf2 family members. Percent similarities were calculated by using the identical and conserved amino acid changes described above. Blocks were generated by using the MACAW sequence alignment program.

Endogenous expression and coassociation of TAF-172 and TBP. (A) HeLa cells (85,000) were solubilized in protein sample buffer and electrophoresed on an SDS–6% polyacrylamide gel (lane 1) along with decreasing amounts of recombinant TAF-172 (80, 40, 20, 10, and 5 ng; lanes 2 to 6) and HeLa nuclear extract (NE; 50 μg; lane 7). Proteins were electroblotted to nitrocellulose and probed with affinity-purified TAF-172 antibodies. Equivalent amounts of HeLa cells and nuclear extracts were also electrophoresed on an SDS–7.8% polyacrylamide gel (lanes 8 and 15) along with recombinant TBP (4, 2, 1, 0.4, 0.2, and 0.1 ng; lanes 9 to 14), electroblotted, and probed with affinity-purified TBP antibodies. (B) HeLa nuclear extracts were chromatographed over phosphocellulose in H.15 buffer and serially step eluted with buffer containing 0.3, 0.5, 0.7, and 1.0 M KCl, as indicated. Equal portions of these fractions were separated on SDS–6% or–7.8% polyacrylamide gels and probed by Western blotting for TAF-172 or TBP, respectively, with affinity-purified antibodies. (C) HeLa nuclear extracts (50 μg; lanes 1 and 4), or nuclear extracts immunodepleted (depl.) of TBP (lanes 2 and 5) or TAF-172 (lanes 3 and 6) were separated on SDS–6% or–7.8% polyacrylamide gels and probed by Western blotting as indicated. Immunoprecipitates (Ip) from these depletions were eluted and treated similarly (lanes 7 to 10). The amount of immunoprecipitate used in the TBP Western blot (lanes 9 and 10) was 10 times that used for the TAF-172 Western blot (lanes 7 and 8).

ATP-mediated dissociation of TAF-172–TBP–DNA complexes as detected by EMSA. (A) Reaction mixtures contained human TBP (hTBP) (lanes 1, 2, and 5 to 15), radiolabeled TATA DNA (50 bp), and increasing concentrations of TAF-172 (lanes 5 to 10) or yeast Mot1 (lanes 12 to 15), as indicated. The letter A indicates that 0.1 mM ATP was included. Migration of TBP-DNA complexes is indicated by T·D and TAF-172/Mot1–human TBP–DNA complexes are indicated by 172·T·D. (B) TBP-DNA or TAF-172–TBP–DNA complexes were preassembled at the indicated concentrations. ATP (0.1 mM) and/or competitor (comp.) TATA DNA (250 nM) were then simultaneously added to the reaction mixtures in lanes 4 to 6, as indicated. Reactions were allowed to continue for 5 min before loading of the gel. In lane 2, competitor DNA was added at the same time as the labeled probe. (C) Reactions similar to those in panel A, except that a 106-bp TATA DNA probe was used. Where indicated, 0.6 μg of antigen affinity-purified TBP, TAF-172, or control TAF_II250 antibody was added to the reaction mixture.

Depletion of endogenous TAF-172 from pol II and pol III transcription reaction mixtures. (A) HeLa nuclear extract (NE; 120 μg, lane 1) and extracts immunodepleted with TAF-172 serum (lane 2) or an unrelated control serum (lane 3) were electrophoresed on an SDS–6% polyacrylamide gel, electroblotted to nitrocellulose, and probed for TAF-172. (B) TAF-172-depleted or control depleted nuclear extracts were used to transcribe 5 nM TI (lanes 1 and 2), G₆TI (lanes 3 and 4), VA₁ (lanes 5 and 6), or U6 (lanes 7 and 8) DNA as described in Materials and Methods. The time of exposure to the phosphor screen was varied to obtain similar signal intensities among the different promoters.

Inhibition of TBP-promoted transcription by TAF-172. (A) HeLa nuclear extracts (100 μg) were used to transcribe 5 nM U6 DNA (lane 1) in the presence of increasing concentrations of TAF-172, as indicated. All of the reaction mixtures contained 5 nM pure recombinant human TBP (rTBP) in addition to endogenous TBP-TAF complexes. (B) HeLa nuclear extracts (NE; 60 μg) were mock treated (lane 1) or heat treated (ht; lanes 2 to 7) at 47°C for 15 min to selectively inactivate endogenous TBP (25) and used to transcribe 5 nM TI promoter DNA in the absence (lane 2) or presence (lanes 3 to 7) of pure recombinant human TBP. TBP, TI DNA, ATP (10 μM), and increasing concentrations of TAF-172, as indicated, were preincubated at 30°C for 30 min prior to the addition of nuclear extract. (C) Reactions identical to those depicted in panel B, except that increasing concentrations of pure recombinant human TFIIA (rTFIIA) were included in the reaction mixtures shown in lanes 5 to 9, as indicated.