Truncated Brf2 proteins lacking the regions N or C terminal to the core domain are inactive for U6 transcription. (A) Structure of human TFIIB, Brf1, and Brf2 and of the recombinant “full-length” (FL) and truncated Brf2 proteins expressed in E. coli. The numbers indicate the first amino acid of the various highlighted regions. The region corresponding to the zinc ribbon structure determined by nuclear magnetic resonance in Pyrococcus furiosus TFIIB (44) and those modeled in Saccharomyces cerevisiae TFIIB and Brf1 (14) are indicated in light blue. The locations of the structured core domain of human TFIIB (2, 30) and the corresponding regions of human Brf1 and Brf2 are indicated in light purple. The histidine tag (HT) is indicated in red. (B) The full-length Brf2 protein (Brf2 FL) or the derivatives indicated above the lanes were produced in E. coli, purified on a nickel affinity column, fractionated on an SDS-polyacrylamide gel,and visualized by staining with the GelCode Blue Stain reagent (Pierce). (C) After normalization of the amounts, the same proteins were visualized by immunoblot with an anti-Brf2 antibody (CS1230, 1:5,000 dilution). (D) The transcription reactions contained 100 ng of TBP, 350 ng of Bdp1, 2 μl of SNAP_c, and 2 μl of Pol III. They were complemented with either buffer alone (lane 1) or 0.5, 1, and 2 μg of full-length Brf2 (lanes 2 to 4), Brf2(66-419) (lanes 5 to 7), or Brf2(2-289) (lanes 8 to 10). U6, correctly initiated U6 RNA; IC, internal control for RNA handling and recovery. In lane 11, only the internal control was loaded.

Activity of truncated versions of Brf2 in preinitiation complex assembly. (A) The C-terminal region of Brf2 is required for binding to the TBP/TATA box complex, whereas the N-terminal region is not required for preinitiation complex assembly. The probe contained a wild-type human PSE and a wild-type human TATA box. Where indicated above the lanes, the binding reactions contained 10.5 ng of TBP, 113 ng of Bdp1, 1 μl of 1:2 diluted SNAP_c, and 250 ng of full-length Brf2 (Brf2 FL) or derivatives as indicated above the lanes. The various complexes are labeled with the symbols indicated in Fig. 2E on the left of each lane or with the factors they contain on the sides of the panel. (B) The C-terminal region of Brf2 is required for cooperative binding with SNAP_c. The probe contained wild-type human U6 PSE and TATA box. Where indicated above the lanes, the binding reactions contained the same amounts of TBP, SNAP_c, and Brf2 or Brf2 derivatives as in panel A. The complexes labeled with an arrowhead on the left of lanes 4 and 5 contain SNAP_c and Brf2.

Activity of Brf2-TFIIB chimeras in U6 preinitiation complex assembly. (A) Activity of TFIIB and chimeras containing the C-terminal region of Brf2. The probe contained a wild-type human PSE and a human TATA box. Where indicated above the lanes, the binding reactions contained 10.5 ng of TBP, 113 ng of Bdp1, 1 μl of SNAP_c diluted 1:15, and 75 ng of Brf2, TFIIB, B/2/2, or 2/B/2, as indicated. The various complexes are labeled with the symbols summarized in Fig. 2E on the left of the lanes or with the factors that they contain on the sides of the panel. (B) Activity of chimeras lacking the C-terminal region of Brf2. The probe contained a wild-type human PSE and a human TATA box. Where indicated above the lanes, the binding reactions contained TBP, Bdp1, SNAP_c, and Brf2 or derivatives in similar amounts as in panel A. The various complexes are labeled with the symbols summarized in Fig. 2E on the left of the lanes or with the factors that they contain on the sides of the panel.

Functions of the N-terminal, core, and C-terminal regions of Brf2. The functions identified in the present study are indicated. A minus sign indicates that the region is dispensable for a given function; plus signs indicate that the region is required (but not necessarily sufficient). N.D, not determined; N.D (not sufficient), the function of the region was not tested but the region is known to be insufficient for function (for example, the contribution, if any, of the core domain of Brf2 to binding to the TBP/TATA box complex was not tested directly, but we know that the Brf2 core domain is insufficient for this function because Brf2 truncated at position 290 did not bind to the TBP/TATA box complex).

Assembly of the U6 preinitiation complex. (A) Assembly of the Brf2-TFIIIB complex on the U6 TATA box. Either a probe with a wild-type mouse U6 PSE and human U6 TATA box (lanes 1 to 13) or a wild-type mouse PSE and a mutant TATA box (lanes 14 to 17) was used. Lanes 1 and 14 show the probes alone. We added 10.5 ng of TBP, 75 ng of Brf2, and 113 ng of Bdp1 to the binding reactions as indicated above the lanes, as well as 1 μl of 1:3-diluted (lane 8) and 1 μl of undiluted (lanes 9 and 12) purified rabbit polyclonal anti-Bdp1 antibody (CS913) and 1 μl of 1:3 diluted (lane 10) and 1 μl of undiluted (lanes 11 and 13) purified mouse monoclonal anti-TBP antibody (SL30b). The locations of protein/DNA complexes containing TBP only, TBP and Bdp1, TBP and Brf2 (complex 2), or TBP, Brf2, and Bdp1 (complex 3) are indicated, as well as are the supershifts obtained after antibody addition. (B) Assembly of a four-factor complex. The probe contained a wild-type human PSE and a wild-type human TATA box. The amounts of TBP, Brf2, and Bdp1 were as described in panel A. In lanes 2 to 6 and lanes 7 to 11, 1 μl of a 1:54, 1:27, 1:9, 1:3, or 1:1 dilution of SNAP_c was added. The locations of the complexes containing SNAP_c alone, TBP and Brf2 (complex 2), TBP, Brf2, and Bdp1 (complex 3), or TBP, Brf2, Bdp1, and SNAP_c (complex 4) are indicated. (C) The four-factor complex is supershifted by anti-TBP, anti-Bdp1, and anti-SNAP190 antibodies. The probe contained a wild-type human PSE and a wild-type human TATA box. The amounts of TBP, Brf2, and Bdp1 were as described in panel A. The binding reactions contained, in addition, 1 μl of a 1:2 dilution of SNAP_c (lanes 2 to 8) and 1μl of 1:3 diluted (lane 3) or 1 μl of undiluted (lane 4) mouse monoclonal anti-TBP antibodies (SL30b), 0.5 and 1 μl of rabbit polyclonal anti-Bdp1 antibodies (CS913, lanes 5 and 6), and 1 μl of 1:3 diluted (lane 7) and 1 μl of undiluted (lane 8) rabbit polyclonal anti-SNAP190 antibody (CS696). The locations of the complexes containing SNAP_c alone, TBP, Brf2, and Bdp1 (complex 3) or TBP, Brf2, Bdp1, and SNAP_c (complex 4), as well as the supershifts obtained after antibody addition, are indicated. The dotted white line indicates the position of complex 4 across the gel and serves to reveal the very slight supershift obtained with the anti-TBP antibodies. (D) Subcomplexes of the four-factor complex. The probe contained a wild-type human PSE and either a wild-type human TATA box (lanes 1 to 6) or a mutated TATA box (lanes 7 to 10). Where indicated above the lanes, the binding reactions contained the same amounts of TBP, Brf2, and Bdp1 as in panel A and 1 μl of 1:15 diluted SNAP_c except for the reaction in lane 9, which contained half this amount. The locations of complexes containing SNAP_c, or TBP and Brf2 (complex 2), or TBP, Brf2, and SNAP_c (complex labeled with a white dot), or TBP, Brf2, and Bdp1 (complex 3), or TBP, Brf2, Bdp1, and SNAP_c (complex 4) are indicated. (E) Nomenclature used for some of the complexes.

Brf2-TFIIB chimeric proteins. (A) The structures of the various chimeras are shown. The numbers under the boxes correspond to amino acids. The C-terminal His tags are indicated in red, and the N-terminal FLAG tags are indicated in blue. (B) The chimeras indicated above the lanes were produced in E. coli, purified on a nickel affinity column, fractionated on an SDS-polyacrylamide gel, and visualized by staining with the GelCode Blue Stain reagent (Pierce). (C) After normalization of the amounts, the same proteins were visualized by immunoblotting with a mouse monoclonal anti-His tag antibody (QIAGEN) at a 1:3,000 dilution. (D) The transcription reactions contained 100 ng of TBP, 350 ng of Bdp1, 2 μl of SNAP_c, and 2 μl of Pol III. They were complemented with either buffer alone (lane 1); 1, 2, or 4 μg of the proteins indicated above the lanes (lanes 2 to 13); 1 μg of Brf2 (lane 14); or 12, 37, 110, 330, or 1,000 ng of 2/B/2 (lanes 15 to 19). U6, correctly initiated U6 RNA; IC, internal control for RNA handling and recovery.

Binding of Brf2, TFIIB, and chimeric proteins to TBP off DNA. GST alone, GST-TBP, or GST-PCMT were immobilized on glutathione-agarose beads as indicated above lanes 5 to 14. 3 μg of TFIIB, Brf2, B/2/−, or B/2/2 were then mixed with 10 μl of beads in the presence of 400 μg of ethidium bromide/ml. After incubation, the beads were washed, resuspended in 5× Laemmli buffer, and boiled; the eluted material was then loaded on an SDS-polyacrylamide gel. The proteins were visualized with a mouse monoclonal anti-His tag antibody (QIAGEN) at a 1:3,000 dilution. Lanes 1 to 4 show 20% of the input material.

Alignment of the human Brf2 sequence with Brf2 and Brf1 sequences from other vertebrate species. (A) Alignment of Brf2 sequences from Homo sapiens (Hs; NP_060780), Mus musculus (Mm; NP_079962), Rattus norvegicus (Rn; XP_224944), Danio rerio (Dn; NP_001003536), and Takifugu rubripes (Tr). This last protein sequence was assembled from expressed sequence tags. The alignment was performed with CLUSTAL W with the default parameters. (B) Alignment of part of the S. cerevisiae (Sc) and H. sapiens (Hs) Brf1 conserved region II with a region near the C terminus of H. sapiens (Hs), M. musculus (Mm), R. norvegicus (Rn), D. rerio (Dn), and T. rubripes (Tr) Brf2. The helical regions in the S. cerevisiae Brf1 conserved region II, as determined in a cocrystal containing S. cerevisiae Brf1 conserved region II, the conserved region of TBP, and a TATA box (19), are shown at the top. The first amino acids of each helix are indicated.