HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hayashi, K. Articles by Ohkuma, Y. Search for Related Content PubMed PubMed Citation Articles by Hayashi, K. Articles by Ohkuma, Y.

¹ Graduate School of Frontier Biosciences, and ² Graduate School of Pharmaceutical Sciences, Osaka University, Suita, Osaka 565-0871, Japan
³ Core Research for Evolutional Science and Technology, Japan Science and Technology Agency, Suita, Osaka 565-0871, Japan
⁴ Cellular Physiology Laboratory, RIKEN, Wako, Saitama 351-0198, Japan
⁵ Gene Engineering Division, BioResource Center, RIKEN, Tsukuba, Ibaraki 305-0074, Japan
⁶ Nippon Institute for Biological Science, Ome, Tokyo 190-0024, Japan

	Abstract

Top Abstract Introduction Results Discussion Experimental procedures DDBJ accession numbers Supplementary material References

 
The general transcription factor TFIIE plays essential roles in transcription by RNA polymerase II (PolII). Despite recent progress, the elucidation of its precise mechanisms including biological functions awaits further characterization. We report the isolation and characterization of Schizosaccharomyces pombe TFIIE (spTFIIE). Like human and other eukaryotic TFIIE proteins, spTFIIE consists of and ß subunits and the genes encoding both subunits are essential for viability. Chromatin immunoprecipitation assays demonstrated that spTFIIE localizes to promoters in vivo. Mutational analysis of the C-terminal basic helix-loop region of TFIIEß, which is involved in the transition from transcription initiation to elongation, revealed that transcription-defective mutants affected in this region are also cold sensitive. The spTFIIEß subunit binds both spTFIIEß and spTFIIE but spTFIIE binds only spTFIIEß. These results indicate that TFIIE forms an ₂ß₂ heterotetramer in which two ß heterodimers are connected via ß subunits. Further analysis of binding specificities showed that spTFIIEß binds the Rpb2 and Rpb12 subunits of PolII, whereas spTFIIE predominantly binds Rpb5, which is located at the clamp region and changes conformation upon transcription initiation.

	Introduction

Top Abstract Introduction Results Discussion Experimental procedures DDBJ accession numbers Supplementary material References

 
Recent progress in genome sequencing projects has made it possible to identify homologs of human proteins in other eukaryotes, in archaea, and in some cases in bacteria. Such studies have shown that the system that mediates transcription of protein-encoding genes is well conserved among eukaryotes and is catalyzed by RNA polymerase II (PolII). Transcription initiation by PolII requires five general transcription factors (TFIIB, TFIID, TFIIE, TFIIF, and TFIIH), which form a preinitiation complex (PIC) on promoters (see reviews, Ohkuma 1997; Hampsey 1998; Roeder 1998). The general transcription factor TFIIE, alone and in collaboration with TFIIH, plays roles both in transcription initiation and in the transition from initiation to elongation. TFIIE is recruited to a PIC precursor consisting of TFIIB, TFIID, TFIIF, and PolII by binding to all of these components as well as to double-stranded DNA (dsDNA) at around position –9 (nine bases upstream from the transcription initiation site, designated as +1), where the DNA helicase subunit XPB of TFIIH begins to promote melting at transcription initiation (Douziech et al. 2000; Yamamoto et al. 2001).

Human TFIIE (hTFIIE) consists of two subunits, hTFIIE and hTFIIE ß, which form an ₂ß₂ heterotetramer (Ohkuma et al. 1990). The larger (57-kDa) subunit hTIFIE consists of 439 amino acid residues and the smaller (34-kDa) subunit hTFIIEß consists of 291 amino acid residues (Ohkuma et al. 1991; Sumimoto et al. 1991). Recently, we have studied hTFIIEß and found that this subunit has two important regions, one located in the middle, which includes a winged-helix (or forkhead) motif and binds to dsDNA, and the other located in the C-terminus, which possesses a basic region-helix-loop (bHL) sequence and which binds to PolII, various general transcription factors and single-stranded DNA (ssDNA) (Okamoto et al. 1998; Okuda et al. 2000). In addition, our studies of point mutations affecting the bHL region of hTFIIEß revealed that this region is essential for transcription by mediating the transition from transcription initiation to elongation (Watanabe et al. 2003).

In contrast, less is known about TFIIE, and extensive structural and functional studies of this subunit have been undertaken only recently. An in vitro reconstituted transcription study showed that the amino (N)-terminal half of hTFIIE (residues 1–173) is sufficient for both basal and activated transcription (Ohkuma et al. 1995). This finding was genetically confirmed with the budding yeast Saccharomyces cerevisiae (Kuldell & Buratowski 1997). Although null mutation of TFA1, which encodes the S. cerevisiae TFIIE homolog (scTFIIE, 482 amino acids), confers lethality, mutations affecting the N-terminal half (residues 1–211) permit viability but confer cold-sensitivity. The N-terminal half shows good homology with recently identified archaebacterial TFIIE homologs (TFE), which lack a region corresponding to the carboxy (C)-terminal half of hTFIIE (Bell et al. 2001; Hanzelka et al. 2001). The acidic region near the C-terminus is the only functional region identified so far in the C-terminal half of hTFIIE that directly binds TFIIH and stimulates basal transcription and TFIIH-mediated phosphorylation of the C-terminal domain (CTD) of the largest subunit of PolII. The CTD consists of repeats of the heptapeptide YSPTSPS (Ohkuma et al. 1995). Recently, the most N-terminal region of archaeal TFE was reported to possess a novel extended winged helix (forkhead) motif that is well conserved among eukaryotic TFIIE homologs, although the functional role of this region is still obscure (Meinhart et al. 2003).

Judging from previous data, TFIIEß may play major roles in regulating the PIC at promoter melting upon transcription initiation by binding to both PolII and the promoter at around –10, where melting begins, as well as to the exposed ssDNA region of the promoter. The TFIIE subunit may modulate the PIC both structurally and functionally, but evidence is still limited; one proposal is that TFIIE recruits TFIIH and stimulates its CTD phosphorylation activity. Recently, it has become apparent that the phosphorylated residues Ser-2 and Ser-5 of the CTD heptapeptide of PolII function as a platform for the recruitment of factors involved in transcription-coupled events such as mRNA processing and nucleosome-histone modification (Orphanides & Reinberg 2002; Hampsey & Reinberg 2003). We demonstrated that TFIIE specifically stimulates CTD Ser-5 phosphorylation (Yamamoto et al. 2001), suggesting that TFIIE is a key regulator of not only transcription itself but also transcription-coupled events. To further investigate the functional roles of TFIIE both genetically and biochemically, we isolated two TFIIE subunit genes (spTFIIEA and spTFIIEB) from the fission yeast Schizosaccharomyces pombe. Both genes are essential for viability. The in vivo localization of spTFIIE on promoters was tested with chromatin immunoprecipitation (ChIP) assay, and spTFIIE was found to be associated with promoter regions that include the transcription initiation site, clearly confirming that TFIIE functions in transcription, as was suggested by in vitro analyses. We also studied mutations affecting the C-terminus of spTFIIEß, for which the corresponding mutations in hTFIIEß were demonstrated to prevent the transition from transcription initiation to elongation, and to confer cold sensitivity. For biochemical analyses, spTFIIE and spTFIIEß were expressed in bacteria. Using these proteins, we investigated the subunit–subunit interactions and searched for the target PolII subunits of the spTFIIE subunits. Our results suggest that TFIIE contributes to conformational changes in the active center of PolII at transcription initiation by pulling the clamp module to close its cleft and by supporting to maintain the opened state of the promoter region during transcription initiation.

	Results

Top Abstract Introduction Results Discussion Experimental procedures DDBJ accession numbers Supplementary material References

 
Isolation of S. pombe TFIIE cDNAs

We previously isolated human TFIIE homologs from Xenopus laevis and Caenorhabditis elegans to examine the relationship between TFIIE structure and its role in transcription (Ohkuma et al. 1992a, 1992b; Yamamoto et al. 2001). These homologs were useful in initial biochemical investigations, and we next extended our studies of the biological roles of TFIIE by isolating TFIIE cDNAs from S. pombe (spTFIIE and spTFIIEß). A TBLASTN homology search of the translated S. pombe database (NCBI) was carried out to identify regions with significant homology to human TFIIE (hTFIIE) subunits. This search yielded spTFIIE cDNA, about 1.5 kb in length, which encodes a highly acidic 434 amino acid protein (pI 4.5) with a calculated molecular weight of 49.1 kDa, and its gene maps to chromosome I. Inspection of spTFIIE revealed that it shares 19% identity and 45% similarity with human TFIIE (hTFIIE) and 26% identity and 50% similarity with S. cerevisiae TFIIE (scTFIIE) over the whole sequence, respectively (Fig. 1A) (Ohkuma et al. 1991; Feaver et al. 1994). The overall levels of sequence conservation were even higher for three N-terminal regions (Forkhead, ZF, and Hydrophobic) that are essential for transcription activity and are well conserved among eukaryotes and found even in archaea, and for a second acidic region near the C-terminus, which directly binds TFIIH. The spTFIIEß cDNA, about 1 kb in length, encodes a highly basic 285-amino acid protein (pI 9.5) with a calculated molecular weight of 32.2 kDa, and its gene maps to chromosome III. The spTFIIEß protein shares 27% identity and 52% similarity with hTFIIEß and 30% identity and 49% similarity with scTFIIEß, respectively (Fig. 1B) (Sumimoto et al. 1991; Feaver et al. 1994). In contrast to spTFIIE, the entire regions of spTFIIEß had higher homology over the whole sequence except for two regions in hTFIIEß, one between N-ter and Forkhead and the other between bHLH and bHL, and one region in scTFIIEß between N-ter and Forkhead.

View larger version (41K): [in this window] [in a new window]   Figure 1  S. pombe TFIIE subunits. (A) Comparison of spTFIIE (spIIE) with hTFIIE (hIIE) and scTFIIE (scIIE). An N-terminally extended forkhead (winged helix) motif (Forkhead), a zinc finger motif (ZF), a hydrophobic region (Hydrophobic), and the first and second acidic regions (2nd acidic) are indicated (Ohkuma et al. 1991; Feaver et al. 1994; Meinhart et al. 2003). The extent of identity and similarity shared by spTFIIE and hTFIIE (upper) and by spTFIIE and scTFIIE (lower) for each structural motif and characteristic sequence region is indicated as a percentage based on the spTFIIE sequence (%). The numbers above and below the diagram indicate amino acid residues that delimit each structure. (B) Comparison of spTFIIEß (spIIEß) with hTFIIEß (hIIEß) and scTFIIEß (scIIEß). An N-terminal highly conserved region (N-ter), a forkhead motif (Forkhead), a hydrophobic region (Hydrophobic), a basic region-helix-loop-helix motif (bHLH), and a basic region-helix-loop sequence (bHL) are indicated (Sumimoto et al. 1991; Feaver et al. 1994; Watanabe et al. 2003). The extent of identity and similarity shared by spTFIIEß and hTFIIEß (upper) and by spTFIIEß and scTFIIEß (lower) is indicated as described for (A) based on the spTFIIEß sequence. (C) SDS-PAGE of purified recombinant spTFIIE subunits. Independently expressed or co-expressed spTFIIE subunits were subjected to SDS-PAGE and stained with Coomassie Brilliant Blue. Lane 1: 300 ng 6H-spTFIIE; lane 2: 300 ng 6H-spTFIIEß; lane 3: 600 ng co-expressed spTFIIE and spTFIIEß. Arrows indicate the positions of each bands. Two bands were detected for the spTFIIEß subunit since for co-expressed spTFIIE, only the N-terminus of spTFIIE possesses a His tag, and spTFIIEß is untagged. The sizes of the molecular weight markers are indicated in kDa at left. (D) Western blot analysis of native spTFIIE subunits. Whole cell extracts (5 µg) were subjected to SDS-PAGE and each spTFIIE subunit was detected using rabbit polyclonal antisera raised against that subunit. Arrows indicate the positions of each subunit. The positions of prestained molecular weight markers are indicated at left (kDa).

To confirm that we had isolated bona fide spTFIIE cDNAs, spTFIIE and spTFIIEß were expressed independently in Escherichia coli with an N-terminal six histidine-tag (6H-spTFIIE and 6H-spTFIIEß) and purified on Ni-NTA agarose columns (Fig. 1C, lanes 1 and 2). 6H-spTFIIE probed to have low solubility (less than 10%), and it was therefore first solubilized in 4 M guanidine before passage through Ni-NTA agarose. However, both subunits were soluble when they were co-expressed in E. coli as a polycistronic construct which encodes an N-terminally His-tagged spTFIIE subunit (Fig. 1C, lane 3). Native spTFIIE subunits were detected in S. pombe nuclear extracts on Western blots using specific antibodies (Fig. 1D). The calculated molecular weight of the band detected with the anti-spTFIIE antibody was 56 kDa and that detected with the anti-spTFIIEß antibody was 34 kDa (Fig. 1D, lanes 1 and 2), both of which match well the estimated molecular weights of the recombinant spTFIIE and spTFIIEß bands determined on the SDS-PAGE gel shown in Fig. 1C.

The spTFIIE genes are essential in S. pombe

The genes for the two spTFIIE subunits (spTF2EA and spTF2EB) were isolated by PCR and disrupted by replacing each open reading frame (ORF) with the URA4 gene, as shown in Fig. 2A. We constructed diploid strains (FKH10 and FKH11) carrying a disrupted copy of each spTFIIE gene as described in Experimental procedures. Replacement of each ORF by the URA4 sequence was confirmed by Southern blotting with [³²P]-labeled probes as indicated in Fig. 2A, which clearly demonstrated that the disrupted strains were heterozygous mutants, spTF2EA/sptf2ea::URA4 and spTF2EB/sptf2eb::URA4 (Fig. 2B). These FKH10 and FKH11 cells were sporulated and subjected to tetrad analysis (Fig. 2C). Of nine tetrads dissected for each subunit gene disruption, all contained two viable and two inviable spores. Moreover, all of the viable spore colonies were ura4^–, indicating that sptf2e::URA4 spores were inviable. These results demonstrate that both spTFIIE genes, like their S. cerevisiae counterparts (Feaver et al. 1994), are essential for cell viability.

View larger version (32K): [in this window] [in a new window]   Figure 2  Tetrad analysis of spTF2EA and spTF2EB heterozygous mutants. (A) Replacement of the coding regions of the spTF2EA and spTF2EB genes with the URA4 gene. The restriction map of each subunit gene is shown. The position of each gene is boxed and the exons are indicated by black boxes. The URA4 gene is indicated by boxes with cross-hatching. The positions of gene-specific probes are indicated by the bars. (B) Confirmation of gene disruption. Chromosomal DNA was isolated from heterozygous mutant cells FKH10 (/wt) and FKH11 (ß/wt) as well as wild-type cells (wt/wt), and digested with the restriction enzyme PvuII to confirm spTF2EA disruption, and with ClaI to confirm spTF2EB disruption. As a control, the same treatments were performed with wild-type DNA. The digested DNA was subjected to Southern analysis with [32P]-labeled probes (P and URA4 for spTF2EA, and Pß and URA4 for spTF2EB) and detected by autoradiography. Arrows indicate bands specific for the disrupted locus, and arrowheads indicate bands specific for the intact locus. (C) Tetrad results of both heterozygous mutants. The segregants from nine dissected asci were grown on YPD plates at 30 °C for 3 days and photographed.

To examine the in vivo distribution of spTFIIE on PolII-transcribed genes we carried out chromatin immunoprecipitation (ChIP) assays after cross-linking nuclear proteins to DNA using formaldehyde (Komarnitsky et al. 2000). Two genes, adh1 and tef3, which encode alcohol dehydrogenase and translation elongation factor 3, respectively, were chosen since they possess a TATA box and are strongly and constitutively transcribed. As shown in Fig. 3B, four regions were monitored for each gene and relative amounts of PCR products were determined. Although the second largest spRpb2 subunit of spPol II was almost equally distributed in all regions, both spTFIIE subunits were observed predominantly at the promoter and promoter-proximal region of both adh1 and tef3, as also seen for spTBP (Fig. 3A). These results confirm functional data from human and S. cerevisiae indicating that TFIIE is involved in two sequential stages, transcription initiation and the transition from initiation to elongation.

View larger version (23K): [in this window] [in a new window]   Figure 3  Localization of spTFIIE subunits on actively transcribed genes as determined by chromatin immunoprecipitation. (A) Schematic of the actively transcribed genes adh1 and tef3. Open reading frames are represented by open boxes and the TATA boxes by black boxes. All numbers are relative to the first nucleotide of the initiation codon (+ 1). Bars above the two genes show the positions of the PCR products used in the ChIP analysis; the name of each PCR fragment is indicated. (B) ChIP assays. Samples were immunoprecipitated with rabbit polyclonal antisera specific to spTFIIE, spTFIIEß, spTBP or the spRpb2 subunit of spPol II. Preimmune sera was used as a negative control. The relative amount of each PCR product was determined on agarose gels with a Fuji LAS-1000 lumino image analyzer (Fujifilm); the amount of PCR product specific to the promoter region (P1 for adh1 and P for tef3) was normalized to 100%. Each lane of the agarose gel is labeled with the PCR product name indicated in (A). PCR products and combinations of primer oligonucleotides are as follows. adh1: P1, adh1pro-1T and adh1pro-0B; P2, adh1pro-1T and adh1pro-1B; CD1, adh1–2T and adh1–2B; CD2, adh1–3T and adh1–3B. tef3: P, tef3–1T and tef3–1B; CD1, tef3–2T and tef3–2B; CD2, tef3–3T and tef3–3B; CD3, tef3–4T and tef3–4B. These primer sequences are presented in Supplementary Table S2 at http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm.

In a previous study of the C-terminal point mutations affecting hTFIIEß, we identified residues essential for binding to PolII, to the general transcription factors TFIIB, hTFIIE, and hTFIIFß, and to ssDNA in both the bHLH and bHL regions, which are essential for transcription initiation (Watanabe et al. 2003). We also showed that the C-terminal helix of the bHL region is involved in the transition from transcription initiation to elongation. To assess the biological relevance of these residues, we created point mutations affecting the corresponding residues of spTFIIEß and determined their effects on cell viability by over-expressing the mutated spTFIIEß subunits (Fig. 4). Since all the mutants possess the hexa-histidine tag at the N-terminus, we confirmed that the expression level of each mutant at 30 °C was almost the same by Western blotting detecting with anti-histidine tag antibody (Qiagen) (data not shown). The residues mutated in the C-terminal bHLH and bHL regions are summarized in Fig. 4A. Six mutants showed cold-sensitivity, one possesses mutation in the bHLH region (W218A) and five others possess mutations in the bHL region (K258E, K260E K261E, R265A R266A, R265E R266E, and Y284A) (Fig. 4B). The mutant W218A corresponds to the human mutant W220A, which confers defects in both hTFIIE binding and transcription initiation (Watanabe et al. 2003). The mutants K258E and K260E K261E correspond to the human mutants R258E R259E and K260E K261E, respectively, which confer defects in binding to PolII, TFIIB, TFIIFß, and ssDNA as well as in transcription initiation. The mutants R265A R266A and R265E R266E correspond to the human mutants R268A R269A and R268E R269E, which confer only modest defects in transcription initiation and binding to PolII and TFIIB. However, hTFIIEß has an additional basic residue, K267, which does not exist in spTFIIEß, and therefore the effects of the spTFIIEß mutants R265A R266A and R265E R266E on transcription must be stronger than that of the human mutants R268A R269A and R268E R269E. The most notable mutant is Y280A, which corresponds to the human mutant Y284A and, in contrast to other mutants, all of which were associated with defects in transcription initiation, conferred a severe defect in the transition from transcription initiation to elongation (Watanabe et al. 2003). It is significant that none of the mutants tested were impaired at any defects at 37 °C (not shown), but six of them were cold sensitive despite their transcriptional defects, whether at initiation or in the transition to elongation.

View larger version (46K): [in this window] [in a new window]   Figure 4  Effects of point mutations in the C-terminal bHLH and bHL regions of spTFIIEß on cell viability. (A) Schematic of mutated residues in spTFIIEß, indicated by shadowing. The hTFIIEß sequence is aligned below. Residue numbers are provided above the spTFIIEß sequence and below the hTFIIEß sequence. The secondary structures of the C-terminal bHLH and bHL regions of spTFIIEß are indicated above the sequence. (B) Growth characteristics of the point mutants. The positions of the mutants are indicated on the panels at right. The effects of point mutations on cell growth were determined by over-expression. The left panels show plates incubated under normal conditions (30 °C) and the middle panels show plates incubated under cold conditions (20 °C). Arrows indicate six mutants with growth defects. Five of them with defects at transcription initiation were indicated in red and one with a defect in the transition from initiation to elongation was indicated in cyan. Names of these mutants are designated in the same colors in the panels at right.

Gel filtration analysis of both native and recombinant hTFIIE proteins indicated that hTFIIE forms an ₂ß₂ heterotetramer (Ohkuma et al. 1990; Peterson et al. 1991). It was thought that both subunits can bind themselves and each other but conclusive analyses had not been performed. Here we studied the binding specificities of both subunits by Far Western blotting analysis (Fig. 5). For probes, [³⁵S]-labeled 6H-spTFIIE and 6H-spTFIIEß were expressed in E. coli BL21(DE3) pLysS by the addition of [³⁵S]-methionine into the minimal essential media and purified as shown in Fig. 5A. The left panel of Fig. 5B shows that [³⁵S]-labeled 6H-spTFIIE bound only to 6H-spTFIIEß. In contrast, [³⁵S]-labeled 6H-spTFIIEß bound to both the 6H-spTFIIE and 6H-spTFIIEß subunits (Fig. 5B, right panel). Gel filtration analysis of recombinant spTFIIE protein was also carried out and the native molecular mass was around 180 kDa, similar to the case of human protein (data not shown). These results confirm that TFIIE forms a heterotetramer, and they also show that spTFIIE forms a heterotetramer in which the two -ß heterodimers are connected via a ß–ß interaction.

View larger version (16K): [in this window] [in a new window]   Figure 5  Subunit binding analysis of spTFIIE. (A) [35S]-labeled spTFIIE subunits. After purification, labeled proteins were subjected to 12.5% SDS-PAGE, and bands were detected by autoradiography. Arrows indicate full length [35S]-labeled spTFIIE subunits. An asterisk indicates a major contaminant in the [35S]-spTFIIE fraction. (B) Far Western blotting analysis of spTFIIE subunits. Arrows indicate the positions of either spTFIIE (about 60 kDa) or spTFIIEß (about 40 kDa). The positions of 10-kDa ladder marker proteins are indicated on the right side (in kDa).

As shown in Fig. 6A,B, the binding specificities of intact PolII with respect to general transcription factors were determined in a human system by using GST-pull down assays. These clearly demonstrated that hPol II binds to hTFIIEß, hTFIIFß (hRap30), and the XPB subunit of TFIIH (Fig. 6A, upper panel, lanes 5 and 7; lower panel, lane 3). To determine whether spPol II also binds to spTFIIEß, Flag-tagged PolII was purified from nuclear extracts of a S. pombe JY741/f-rpb3 cell line that express an N-terminally Flag-tagged spRpb3 through anti-Flag antibody (M2)-agarose (Sigma) (Kimura et al. 2002) (Fig. 6B). GST-pull down assays were carried out with purified Flag-PolII and GST-tagged spTFIIE subunits (Fig. 6C). Intact spPol II bound to spTFIIEß in crude bacterial lysate or in purified fractions (Fig. 6C, lanes 4 and 5, respectively), and it was also confirmed that intact spPol II cannot bind to spTFIIE (Fig. 6C, lane 3).

View larger version (39K): [in this window] [in a new window]   Figure 6  Binding assays of TFIIE to PolII. (A) GST pull-down assay of human general transcription factor subunits with intact human PolII. All of the GST-tagged human general transcription factors (300 ng each) except for the TAF subunits of TFIID were studied. The assay was carried out by addition of 1 µg hPol II. After SDS-PAGE on a 5.5% polyacrylamide gel and Western blotting, bound hPol II was detected with an antibody to the CTD of the hRpb1 subunit. Upper panel: lane 1, 10% input of hPol II (I); lane 2, GST alone (G); lane 3, GST-TFIIB (B); lane 4, GST-TFIIE (E); lane 5, GST-TFIIEß (Eß); lane 6, GST-TFIIF (F); lane 7, GST-TFIIFß (Fß); lane 8, GST-TBP (T); lane 9, GST-TFIIA (A); lane 10, GST-TFIIAß (Aß); and lane 11, GST-TFIIA (A). Lower panel: lane 1, 10% input of hPol II (I); lane 2, GST alone (GST); lane 3, GST-XPB (XPB); lane 4, GST-XPD (XPD); lane 5, GST-p62 (p62); lane 6, GST-p52 (p52); lane 7, GST-p44 (p44); lane 8, GST-p34 (p34); lane 9, GST-Cdk7 (Cdk7); lane 10, GST-CyclinH (CycH); and lane 11, GST-MAT1 (MAT1). The position of a 200-kDa molecular weight marker, myosin heavy chain, is indicated at left on each panel. Arrows indicate the position of the unphosphorylated form of the hRpb1 subunit of hPol II. (B) SDS-PAGE of purified spPol II. Flag-tagged spPol II (f-spPol II) was purified on an M2-agarose column from nuclear extracts of the S. pombe strain JY741/f-rpb3 (Kimura et al. 2002). Five hundred µg of purified f-spPol II was subjected to SDS-PAGE and the gel was silver-stained. Arrows indicate the positions of spPol II subunits. The positions of 10-kDa ladder marker proteins are shown at left. (C) GST-pull down assay of spTFIIE subunits with spPol II. Bacterial lysates containing 300 ng GST-tagged spTFIIE subunit or 300 ng purified GST-spTFIIEß was mixed with f-spPol II and precipitated with glutathione-Sepharose. Bound spPol II was detected by Western blotting with a rabbit antibody against the spRpb2 subunit. Lane 1, 10% input (Input); lane 2, GST alone (GST); lane 3, GST-spTFIIE (spIIE); lane 4, GST-spTFIIEß (spIIEß); and lane 5, purified GST-spTFIIE (purified spIIE). (D) Far Western blotting analysis of binding of spTFIIE subunits to the two largest subunits of spPol II. Arrows indicate the positions of spRpb2 and 6H-spTFIIEß. (E) GST-pull down assay of spPol II subunits. Bacterial lysates containing 300 ng GST-tagged spPol II subunit (spRpb3-spRpb12) were mixed with HA-tagged general transcription factor and precipitated with glutathione-Sepharose. Bound factor was detected by Western blotting with a mouse anti-HA monoclonal antibody 12CA5. Top panel, HA-spTFIIEß (spIIEß); second panel, HA-spTFIIE (spIIE); third panel, HA-spTFIIB (spIIB); bottom panel, HA-spTFIIFß (spIIFß). Lane 1, 10% input (Input); lane 2, GST alone (GST); lanes 3–12, GST-spRpb3 to GST-spRpb12, respectively (Rpb3-Rpb12); lane 13, GST-spTFIIE (IIE) for top panel, and GST-spTFIIEß (IIEß) for second, third and bottom panels. Arrows indicate bound input proteins.

To further describe interactions between spTFIIE and spPol II at the molecular level we examined the binding specificities of spTFIIE to spPol II subunits (Fig. 6D,E). Binding to the two larger subunits, spRpb1 and spRpb2, was monitored by Far Western blotting analysis with [³⁵S]-methionine labeled 6H-spTFIIE and 6H-spTFIIEß (Fig. 6D). As a positive control, 6H-spTFIIEß was run in parallel with purified Flag-PolII. As shown, spTFIIEß bound to spRpb2 and weakly to spRpb1 (Fig. 6D, lane 3) and spTFIIE bound weakly to both spRpb1 and spRpb2 (Fig. 6D, lane 1). GST-pull down assays were carried out to detect interactions with the smaller spPol II subunits, spRpb3 to spRpb12, each of which was N-terminally tagged with GST (Fig. 6E). HA-tagged spTFIIB, spTFIIE, spTFIIEß, and spTFIIFß were used for these binding assays. Since it was reported that hTFIIFß (hRap30) binds to the hRpb5 subunit and that S. cerevisiae (sc)TFIIB-related scPol III factor Brf binds to the scRpb4 subunit, we used spTFIIFß and spTFIIB as positive controls (Ferri et al. 2000; Wei et al. 2001). As a result, we confirm the binding specificities for the S. pombe counterparts. spTFIIFß also bound to spRpb5 and, additionally, to spRpb12 and weakly to spRpb3 and spRpb4 (Fig. 6E, bottom panel, lanes 5, 12, 3, and 4, respectively). spTFIIB bound to spRpb4 and, additionally, to spRpb12, and weakly to spRpb3 and spRpb11 (Fig. 6E, third panel from the top, lanes 4, 12, 3, and 11, respectively). When both subunits of spTFIIE were analyzed, spTFIIEß was found to bind to spRpb12 in addition to spRpb2 (Fig. 6E, top panel, lane 12), and intriguingly, spTFIIE was found to bind solely to spRpb5 (Fig. 6E, second panel from the top, lane 5). It is noteworthy that TFIIB and TFIIE can bind to individually expressed PolII subunits but not to intact PolII (Fig. 6A, lanes 3 and 4 and Fig. 6C, lane 3).

	Discussion

Top Abstract Introduction Results Discussion Experimental procedures DDBJ accession numbers Supplementary material References

 
Rapid progress in biochemical and structural techniques has made it possible to observe PolII and many other transcription factors at the atomic level. Recently, the X-ray structures of several different forms of scPol II (the 10-subunit core structure and its elongation form, and the 12-subunit holo structure) and the cocrystal structure of the general transcription factor scTFIIB and 10-subunit core scPol II were reported (Cramer et al. 2001; Gnatt et al. 2001; Amarche et al. 2003; Bushnell & Kornberg 2003; Bushnell et al. 2004). Thus, it is now known how the active center of PolII is guided by the N-terminal half of TFIIB to the transcription initiation site and how the clamp module of PolII closes when transcription commences. However, we still do not know how PolII changes in conformation when it binds to TFIIF before interacting with the PIC or when transcription is initiated by the addition of TFIIE and TFIIH to the PIC. In the present study, we isolated the spTFIIE genes and cDNA, and confirmed previous S. cerevisiae data that TFIIE is essential for cell viability (Feaver et al. 1994). To further elucidate the TFIIE functions both biochemically and genetically, we carried out chromatin immunoprecipitation (ChIP) assays and point mutation studies to examine the effects on cell viability of mutations affecting the C-terminal bHLH and bHL regions which correspond to point mutations in hTFIIEß that confer defects in transcription, either at initiation or at the transition from initiation to elongation (Watanabe et al. 2003). We also biochemically analyzed the TFIIE subunits by determining their binding specificities, both to themselves as well as to spPol II subunits.

spTFIIE forms a preinitiation complex on the promoters of transcriptionally active genes

Each spTFIIE subunit had several highly conserved regions when compared with human and S. cerevisiae, and both were essential for cell viability (Figs 1 and 2). It is intriguing that both subunits possess the forkhead domain but only that in TFIIEß showed a dsDNA binding activity. The situation is similar for TFIIF since both subunits of TFIIF (TFIIF and ß) possess the forkhead domain but only that of TFIIFß shows a dsDNA binding activity. The functions of the forkhead domain other than DNA binding might be quite important because this region is essential for transcription. As written in the paper which reported the crystallographic structure of the forkhead domain of TFA, the spTFIIE homolog in archaea, this domain might be essential for the protein–protein interactions (Amarche et al. 2003). We therefore examined, for the first time, whether both subunits were in the PIC in vivo by using the ChIP assay (Fig. 3). As a result, they were actually co-localized with spTBP but their localization profiles were completely different from spPol II. The previous paper reported that scTFIIEß located predominantly at the promoter in the budding yeast (Komarnitsky et al. 2000). Our data confirmed their result that spTFIIEß located at the promoter proximal regions with spTBP and, furthermore, demonstrated for the first time that the spTFIIE subunit also had overlapped localization with spTFIIEß and spTBP. Taken together with the tetrad result for cell viability, we conclude that spTFIIE is actually functional in S. pombe.

Cold sensitive growth correlated with transcriptional defects of spTFIIEß mutants

We recently identified the transcriptionally essential residues in the C-terminal bHLH and bHL regions of hTFIIEß, classified them in which transcriptional step they were involved, and found that the evolutionally conserved residues at the C-terminal helix region of bHL were essential for the transition from initiation to elongation and the rest of transcriptionally essential residues were basically involved in transcription initiation (Watanabe et al. 2003). To study the biological relevance of those residues, we mutated the residues in spTFIIEß corresponding to the transcriptionally defective residues in hTFIIEß and observed their phenotypes by over-expression (Fig. 4). Intriguingly, all defective mutants (W218A, K258E, K260E K261E, R265E R266E, R265A R266A, and Y280A) in the C-terminus of spTFIIEß showed cold sensitivities and their phenotypes were similar to the phenotypes of the previously reported C-terminally truncated mutants of scTFIIE which may fail to bind properly to scTFIIH (Kuldell & Buratowski 1997). A common feature of those mutants is that all those are transcriptionally active to some extent (more than 25% of the wild-type) (Ohkuma et al. 1995; Watanabe et al. 2003). This is in clear contrast to the features of the zinc finger mutants which showed lethal or temperature sensitive phenotypes in S. cerevisiae and possessed much lower transcription activities (less than 20% of the wild-type) in human by using human in vitro transcription system (Kuldell & Buratowski 1997; A. Tanaka & Y. Ohkuma, unpublished observation).

TFIIE is a heterotetramer

As expected from the gel filtration column profiles of both native and recombinant hTFIIE proteins (Ohkuma et al. 1990; Peterson et al. 1991), our Far Western studies demonstrated that spTFIIE formed a heterotetramer in vitro by interactions between and ß, and ß and ß subunits (Fig. 5). Judging from the protein-DNA photo-crosslinking data of the PIC components and the promoter DNA, both subunits of TFIIE bound to the wide range of promoter regions from –50 to +35 (Douziech et al. 2000; Forget et al. 2004) and, thus, this factor might be big enough to reach all sites. Our data that TFIIE is a heterotetramer with the molecular weight of 180 kDa and binds to all of the general transcription factors as well as PolII support these results and speculations (Ohkuma et al. 1990; Yamamoto et al. 2001).

PolII might change its conformation several times upon binding to various general transcription factors before transcription initiation

As shown in Fig. 6A,B, intact (no activation) hPol II bound predominantly to hTFIIEß, hTFIIFß and the XPB subunit of hTFIIH. The specific binding of PolII to TFIIEß was also confirmed in S. pombe (Fig. 6D). When the binding of the spTFIIE subunits to the individual spPol II subunit was tested, spTFIIEß mainly bound to spRpb2 and spRpb12 and quite faintly to spRpb1. The binding of spTFIIFß to spRpb5 was also confirmed but, additionally, the binding to spRpb12 was observed (Wei et al. 2001). The intriguing observations were that spTFIIE, though no binding was observed with intact spPol II as shown, bound well to spRpb5 and faintly to spRpb1 and spRpb2 as well and that spTFIIB, as suggested previously, bound to spRpb4 and was newly found to bind to spRpb12, in addition (Ferri et al. 2000). These results of spTFIIE and spTFIIB that they bound to the individual spPol II subunits but could not bind to intact spPol II (Fig. 6C and data not shown) immediately indicate a strong possibility of the intramolecular conformational changes of PolII upon binding of the general transcription factors. From those results together with the reported X-ray cocrystal structure of PolII-TFIIB in S. cerevisiae, spTFIIB becomes available to bind to spPol II when spPol II possibly changes its conformation by prebinding to spTFIIF (Bushnell et al. 2004). However, the opposite possibility may also be true since it was reported that spTFIIB changed its conformation to the open form upon binding to the transcriptional activator VP16, and possibly (we presume) upon formation of PIC with spTBP on the promoter (Glossop et al. 2004). This TFIIB open form is exactly the conformation of scTFIIB whose N-terminal half is inserted into the active center of scPol II in the cocrystal with scPol II (Bushnell et al. 2004).

Figure 7 represents our models of PolII conformational changes during PIC formation based on our results in addition to the reported structures of PolII and the general transcription factors under various conditions (Gnatt et al. 2001; Amarche et al. 2003; Chung et al. 2003; Bushnell et al. 2004). In Fig. 7A, the general transcription factor binding steps and accompanying PolII conformational changes are drawn step-wise (left, middle and right panels) on the space fill model of scPol II according to the order of PIC formation. The corresponding schematic models with conformational changes on the promoter DNA are shown in Fig. 7B. As described above, changes in PolII are first provoked by binding of TFIIF and TFIIB to PolII. TFIIFß will stimulate the clamp module to open by binding to the Rpb2 and Rpb5 subunits (corresponding to the clamp module) of PolII, and the latter binding will cause the kink at the joint of the Rpb4-Rpb7 module inward to the PolII clamp module as shown with a black arrow (Fig. 7A,B, middle panel) since the counterpart of TFIIFß, TFIIF, was demonstrated to bind extensively to the Rpb4-Rpb7 module (Chung et al. 2003). As a result, TFIIB and DNA can easily bind PolII; TFIIB binds to the PolII dock module at its N-terminal Zn ribbon motif and to the PolII cleft at its B finger region, maneuvering the promoter DNA around the transcription initiation site to the active center of PolII (Bushnell et al. 2004). TFIIB also binds to TBP and to the TATA box proximal region at its C-terminal half. Consequently, TFIIB shrinks in size on the surface of PolII because its N-terminal half is inserted into the active center and only the C-terminal half remains near the dock domain via interactions with the Rpb4-Rpb7 module. Therefore, TFIIB is drawn smaller in the panels at right of Fig. 7A,B than in the middle panels. As shown in the right panels of both Fig. 7A,B, the next PolII conformational change might be caused by the binding of TFIIE. Since TFIIEß can bind to intact PolII, two TFIIEß subunits will bind to the Rpb2 and Rpb12 subunits, without assistance from other factors, and at the same time to TBP, TFIIB, and TFIIFß as well as the promoter just upstream of the transcription initiation site. The PolII conformational changes induced by TFIIB and TFIIF will make TFIIE accessible to PolII at the Rpb5 subunit, which positions the edge of the clamp module at the time when TFIIEß binds to the Rpb2 and Rpb12 subunits. The closing of the clamp module of PolII upon transcription inititation will be triggered by TFIIE, since it binds to the clamp and bridges the clamp and jaw modules of PolII in a complex with TFIIEß, which binds to Rpb2 to form the jaw module of PolII. This model also fits protein-DNA photo-crosslinking results which show that TFIIE binds both upstream and downstream of the core promoter region and that TFIIEß binds just inside of the promoter region at approximately three regions, a region around the TATA box, a region thought to melt upon transcription initiation, and a region downstream of the initiation site (Douziech et al. 2000). This second PolII conformational change also enables TFIIH to bind to TFIIE near the PolII clamp module and to approach the CTD of Rpb1 for its phosphorylation at transcription initiation. Since the holo scPol II structure has been determined by X-ray crystallography and its cocrystal structures with other components have started to be determined (Amarche et al. 2003; Bushnell & Kornberg 2003; Kettenberger et al. 2003; Bushnell et al. 2004), the dynamic conformational changes in PolII during transcription will be elucidated in the near future. A further understanding of the effects of TFIIE binding to PolII on transcription from both structural and genetical viewpoints is now in progress.

View larger version (47K): [in this window] [in a new window]   Figure 7  Schematic models of PolII conformational changes upon the binding of general transcription factors. (A) A model of the space-filled structure of the whole PolII (PDB accession number: 1NT9 [PDB] ) from Cn3D. According to the factor addition, the estimated bound profiles of each factor were drawn. Red arrows indicate the movement of the clamp module and black arrows indicate the movement of the Rpb4-Rpb7 module. (B) Schematic model of PolII conformational changes upon binding of promoter DNA and general transcription factors (TFIIB, TFIIFß, and TFIIE). The positions of PolII subunits were drawn based on the PolII structures presented in (A). Red arrows indicate the movement of the clamp module and black arrows indicate the movement of the Rpb4-Rpb7 module as shown in (A).

	Experimental procedures

Top Abstract Introduction Results Discussion Experimental procedures DDBJ accession numbers Supplementary material References

The S. pombe strains used in this study are shown in Supplementary Table S1. NP1–6A and NP1–6D are parental strains (see Supplementary Table S1 at http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm). YE, EMM and ME media were prepared as previously described (Moreno et al. 1991). S. pombe was transformed by the lithium acetate method (Okazaki et al. 1990).

Isolation of S. pombe TFIIE cDNA clones

The putative spTFIIE coding sequence was identified in a TBLASTN homology search of the S. pombe translated expressed sequence tag (EST) databank (Sanger Centre, Cambridge, UK) to locate regions with significant homology to the hTFIIE amino acid sequence. For amplification of the full coding region of spTFIIE cDNA, the primer SPEAR6T, which has an NdeI site at the initiation codon, was used in conjunction with the primer SPEAR2B, which has a BamHI site behind the termination codon (see Supplementary Table S1 at http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm). PCR was performed using a S. pombe cDNA library as a template and a combination of the oligo cap-specific primer, Adaptor primer 1 (5'-CCATCCTAATACGACTCACTATAGGGC-3'; Clontech), and the primer SPEAR2B with Advantage cDNA polymerase (Clontech). A S. pombe cDNA library was prepared with total mRNA from TP4–5 A and each cDNAs have the oligo capping at the top of themselves. The spTFIIE cDNA was obtained using the first PCR product and the primers SPEAR6T and SPEAR2B with Pyrobest DNA polymerase (Takara). The PCR products were subcloned into the SmaI site of pBluescript II SK(–) (Stratagene). The nucleotide sequences of the cloned PCR products were determined using an ABI PRISM 310 Genetic Analyzer (Applied Biosystems).

The same strategy was used to identify several regions with high homology to the hTFIIEß amino acid sequence. Since the extent of homology was higher than the case of TFIIE and the putative N and C termini were identified, it was easier to identify the entire putative coding region of spTFIIEß from S. pombe genomic sequences. The oligonucleotide SPEB1T (5'-GATTCCATATGAGTTCACTAAGCGATC-3') contains an NdeI site (underlined) at the first methionine codon, and the oligonucleotide SPEB3B (5'-CGTTGACTCGAGAGGTTTCATGGAGCTATAATCACG-3') contains an XhoI site (underlined) after the stop codon. The PCR products were subcloned into the SmaI site of pBluescript II SK(–) (Stratagene). The nucleotide sequences of the cloned PCR products were confirmed using an ABI PRISM 310 Genetic Analyzer (Applied Biosystems).

Construction of spTFIIE expression vectors

The plasmid containing the spTFIIE cDNA was digested with NdeI and BamHI, and the relevant fragment was subcloned into the vectors 6HisT-pET11d and HA-pET11d to construct plasmids expressing six histidine-tagged spTFIIE (6H-spTFIIE) and hemagglutinin-tagged spTFIIE (HA-spTFIIE), respectively (Hoffmann & Roeder 1991; Okamoto et al. 1998). Similarly, the plasmid containing spTFIIEß cDNA was digested with NdeI and XhoI, and the relevant fragment was subcloned into the vectors 6HisT-pET11d and HA-pET11d to construct the expression plasmids 6H-spTFIIEß and HA-spTFIIEß, respectively.

A co-expression plasmid containing two spTFIIE subunit cDNAs was constructed essentially as previously described (Yamamoto et al. 2001). The spTFIIEß cDNA fragment was isolated by digestion with NdeI and XhoI and used to construct an untagged spTFIIEß-expressing plasmid (in pET3a). This plasmid was digested with XbaI and blunt-ended, and the resulting fragment was treated with calf intestine phosphatase. Finally, the 6H-spTFIIE expression plasmid was digested with XbaI and BamHI, and the ends of the fragment containing spTFIIE cDNA were blunt-ended. Finally, this 6H-spTFIIE cDNA fragment was subcloned into the XbaI (blunt) sites of the spTFIIEß expression plasmid such that both cDNAs were tandemly oriented to create the expression plasmid 6H-spTFIIE.

Glutathione S-transferase (GST) fusion constructs of both spTFIIE subunits were made in pGEX-2TL(+) as previously described by digestion with the same combinations of restriction enzymes (NdeI-BamHI for spTFIIE and NdeI-XhoI for spTFIIEß) (Okamoto et al. 1998). Fusions of GST constructs to spPol II subunits (spRpb3-spRpb12) were made as follows. The coding region of each subunit was excised from 6His-tagged subunit constructs in pET21d with appropriate restriction enzymes (NdeI-XhoI for 6H-spRpb3 and NdeI-BamHI for all others) and subcloned into pGEX-2TL(+). All subunits were GST-tagged at the N-terminus except that spRpb3 contains a 6His-tag immediately after the GST-tag at the N-terminus.

Expression and purification of recombinant proteins

Recombinant proteins were expressed in E. coli BL21(DE3)pLysS by induction with isopropyl-ß-D-thiogalactopyranoside (IPTG) (Okamoto et al. 1998). Soluble bacterial lysates were used for general purification. For miniscale preparations, lysates (1 mL) representing 50–100 mL culture were mixed directly with 1 mL buffer B (20 mM Tris-HCl (pH 7.9 at 4 °C), 10% (v/v) glycerol, 1 mM phenylmethylsulfonyl fluoride (PMSF), 2 µg/mL anti-pain, 2 µg/mL aprotinin, 1 µg/mL leupeptin, 0.8 µg/mL pepstatin, 10 mM 2-mercaptoethanol) containing 500 mM NaCl (BB500), and 100 µL Ni²⁺-nitrilotriacetic acid (NTA) agarose (Qiagen) and incubated for 4 h at 4 °C. The resin samples were washed twice with 1 mL BB500, twice with 1 mL buffer D (20 mM Tris-HCl (pH 7.9 at 4 °C), 20% (v/v) glycerol, 1 mM PMSF, 10 mM 2-mercaptoethanol) containing 500 mM KCl (BD500), and twice with 500 µL BD500 containing 40 mM imidazole-HCl (pH 7.9). Bound proteins were eluted twice with 300 µL BD500 containing 100 mM imidazole-HCl (pH 7.9). Typical preparations were > 80% pure as judged by Coomassie Brilliant Blue staining of a sodium dodecyl sulfate (SDS)-polyacrylamide gel. For purification of 6H-spTFIIE, the protein was first precipitated with 33% (v/v) saturated ammonium sulfate in 20 mM Tris-HCl (pH 7.9 at 4 °C) by ultracentrifugation at 20 000 g for 20 min at 4 °C using the 45 Ti rotor (Coulter-Beckman), resuspended with BB500 containing 10 mM imidazole-HCl (pH 7.9), and purified with Ni²⁺-NTA agarose as described above.

HA-tagged and GST fusion derivatives of spTFIIE subunits were expressed in E. coli BL21(DE3)pLysS by induction with IPTG. Cells were harvested from 50 mL culture, resuspended in 1 mL BB500, and sonicated. Soluble lysates were separated from insoluble debris by ultracentrifugation at 20 000 g for 20 min at 4 °C using the 50.2 Ti rotor (Coulter-Beckman) and stored at –80 °C until use for the GST pull-down assay. For purification of GST fusion proteins, soluble lysates were mixed with glutathione-Sepharose resin (Amersham Pharmacia) equilibrated with buffer C (20 mM Tris-HCl (pH 7.9 at 4 °C), 20% (v/v) glycerol, 0.5 mM EDTA, 1 mM PMSF, 10 mM 2-mercaptoethanol) containing 500 mM KCl (BC500) and then eluted with BC500 containing 20 mM glutathione.

Generation of antibodies against spTFIIE subunits

Both 6H-spTFIIE subunits were expressed independently in E. coli, solubilized by sonication and purified on a Ni-NTA agarose column. Since 6H-spTFIIEß was mostly soluble (> 80% in soluble lysate) and 6H-spTFIIE was mostly insoluble (> 90% in pellet), 6H-spTFIIEß was purified from bacterial lysates and 6H-spTFIIE from bacterial pellets after solubilization with 4 M guanidine-HCl (pH 7.5). Two mg of each purified protein was subjected to SDS-PAGE and the appropriate bands were excised from the gel after Coomassie Blue staining.

To raise rabbit polyclonal antibodies against spTFIIE subunits, 200 µg of each 6H-TFIIE subunit was mixed with complete Freund's adjuvant (Difco) and injected intramuscularly into a separate rabbit. Two weeks after the first injection, a second injection of 100 µg of each 6H-ceTFIIE subunit mixed with incomplete Freund's adjuvant (Difco) was given both intramuscularly and subcutaneously. The third and fourth injections, identical to the second, were given after a further two weeks. Blood was collected 8 days after the fourth injection. Each raised antibody recognized its corresponding spTFIIE subunit, whether natural or recombinant, in solution and on Western blots.

Preparation of yeast nuclei

S. pombe strain NP1–6 A was grown to OD₆₀₀ 1.0 in YEA medium. The cells were harvested and the pellet was washed twice with chilled Sorbitol buffer (1.4 M sorbitol, 40 mM HEPES (pH 7.5), 0.5 mM MgCl₂) containing 1 mM PMSF and 10 mMß-ME by centrifugation at 5000 r.p.m. for 5 min at 4 °C with a HB4 swinging bucket rotor. The pellet was resuspended in a 4-fold volume of Sorbitol buffer containing 1 mM PMSF and 2 mMß-ME and lyzed with zymolyase 100T (Seikagaku Kogyo). Spheroblasts were collected by centrifugation, washed twice with chilled sorbitol buffer containing 1 mM PMSF and suspended in Ficoll buffer (18% Ficoll 400, 20 mM PIPES (pH 6.5), 0.5 mM MgCl₂). The suspension was homogenized with a glass pestle, laid over Glycerol-Ficoll buffer (7% Ficoll 400, 20% (v/v) glycerol, 20 mM PIPES (pH 6.5), 0.5 mM MgCl₂ containing 1 mM PMSF) and centrifuged at 24 000 g for 30 min at 4 °C using a SW 28 swinging backet rotor (Coulter-Beckman). The pellet was resuspended in Ficoll buffer and the supernatant was collected by centrifugation as above. The supernatant was washed with Ficoll buffer and the nuclear pellet was prepared.

Disruption of the spTFIIE genes

A 2.8 kb genomic DNA fragment containing the spTFIIE gene (spTF2EA) from position –632 to +2196 (where +1 is the first nucleotide of the translation start codon) was amplified with Pyrobest DNA polymerase (Takara) and primers SPEAG1 and SPEAG2 (see Supplementary Table S2 at http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm) with S. pombe genomic DNA as a template. After treatment with T4 polynucleotide kinase (Roche), the fragment was subcloned into pBluescript II SK (–) at the SmaI site to construct pBS-spTF2EA. For disruption of spTF2EA, pBS-spTF2EA was digested with SplI and EcoRV, in which the entire coding region and the 1.7 kb URA4 fragment (blunt) were included, to construct pBS-sptf2ea::URA4. A 2.8 kb sptf2ea::URA4 fragment was produced by PCR using the same primer sets as above and pBS-sptf2ea::URA4 as the template. For disruption of spTFIIEß gene (spTF2EB), the same strategy was performed as above. A 2.3 kb genomic DNA segment including spTF2EB from positions –618 to +1705 was amplified using primers SPEBG1 and SPEBG2 (see Supplementary Table S2 at http://www.blackwell-publishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm) with S. pombe genomic DNA as a template, phosphorylated and subcloned into pBluescript II SK (–) at the SmaI site to construct pBS- spTF2EB. For disruption of spTF2EB, pBS-spTF2EB was digested with EcoRI and BstXI and blunt-ended, and 1.7 kb blunt-ending URA4 fragment was inserted to construct pBS-sptf2eb::URA4. A 2.8 kb sptf2eb::URA4 fragment was produced by PCR by using the same primer sets as above and pBS- sptf2eb::URA4 as the template. To generate the spTF2EA/sptf2ea::URA4 (FKH10) and the spTF2EB/sptf2eb::URA4 (FKH11) diploid strains, these fragments were used to transform a diploid by mating NP1–6A and NP1–6D. Disruption was confirmed by Southern blot analysis performed with genomic DNA extracted from the diploid transformants, FKH10 and FKH11. Genomic DNA from FKH10 was digested with PvuII and gene disruption was confirmed by hybridization with a [³²P]-labeled specific AflIII-SplI fragment (P in Fig. 1B) and the URA4 fragment as probes. Genomic DNA from FKH11 was digested with ClaI and hybridized with a [³²P]-labeled BstXI-SphI fragment (Pß in Fig. 1B) and the URA4 fragment as probes.

Tetrad analysis

The diploid strains spTF2EA/sptf2ea::URA4 (FKH10) and spTF2EB/sptf2eb::URA4 (FKH11) were grown on ME plates at 28 °C for 2 days and then streaked to a YPD plate and incubated at 30 °C for several days. The asci formed were isolated with a manipulator on a YE plate containing 100 µg/mL adenine, spores were dissected, and the plate was incubated at 30 °C for 3 days. All viable spore colonies were found to be ura4^– by replica plating to EMM2 plates containing uracil, adenine, leucine and histidine.

Chromatin immunoprecipitation

Chromatin immunoprecipitation (ChIP) was performed essentially as described (Komarnitsky et al. 2000) with minor modifications. A 50 mL-volume of S. pombe strain NP1–6A was grown to an optical density of 0.9 at 600 nm in YEA medium. Formaldehyde was added to a final concentration of 1% for 30 min at room temperature, and incubation was continued with 120 mM glycine at room temperature for 5 min. Cells were harvested by centrifugation, washed twice with chilled Tris-buffered saline, and lyzed with glass beads in lysis buffer (50 mM HEPES-KOH (pH 7.5), 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% Na-deoxycholate, 1 mM PMSF, 1 µg/mL leupeptine, 1 µg/mL pepstatin A). Chromatin was treated by sonication to yield fragments ranging from 200 to 800 bp. This cell lysate was allowed to react with antibody at 4 °C for 12 h and was incubated with protein G-Sepharose 4 Fast Flow (Amersham Pharmacia). The immunoprecipitate-bound resins were collected by centrifugation and washed successively with lysis buffer containing 140, 250 and 500 mM NaCl. After washing with LiCl/detergent wash buffer (1 mM LiCl, 10 mM Tris-HCl (pH 7.9 at 4 °C), 1 mM EDTA, 0.5% Na-deoxycholate, 0.5% Nonidet P-40) and then TE buffer (pH 8.0), the resins were incubated with elution buffer (50 mM Tris-HCl (pH 7.9 at 4 °C), 10 mM EDTA, 1% SDS) at 65 °C for 10 min. ChIP fractions were collected by centrifugation, treated with protease, and cross-linking was reversed. PCR was performed with 25 cycles using AmpliTaq Gold (Applied Biosystems). The relative amount of each PCR product was measured on an agarose gel with a Fuji LAS-1000 lumino image analyzer (Fujifilm), and normalized to the amount of PCR product of each precipitate at the promoter region, which was defined as 100%.

Preparation of spTFIIEß point mutants

Plasmids harboring mutated spTFIIEß cDNA were constructed with a site-directed mutagenesis system as previously described (Watanabe et al. 2003). A cDNA of full length spTFIIEß with hexa histidine-tag at the N-terminus (6H-spTFIIEß) was subcloned into the plasmid pREP1 (Maundrell 1990). A site-directed mutagenesis kit Mutan-K (Takara) was used to create various oligonucleotide-mediated point mutations with the plasmid pREP1–6H-spTFIIEß as a template (see Supplementary Table S2 at http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm). The mutants were then checked by sequencing as above. The mutant plasmids were transformed into the S. pombe strain TP4–5 A and cells over-expressing mutated spTFIIEß derivatives were constructed (see Supplementary Table S1 at http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm). The transformed strains were cultured in liquid EMM2 medium containing 100 µg of uracil and adenine, and 2 µM thiamine to 1 x 10⁷ cells/mL at 30 °C and were streaked on thiamine-free EMM2 plates containing 100 µg of uracil and adenine. Plates were incubated at 20 °C, 30 °C and 37 °C, and cold-sensitive mutants were identified after several days. The expression level of each mutated protein at 30 °C was checked by Western blotting of cell lyzate with mouse anti-Penta-His monoconal antibody (cat. no. 34660, Qiagen).

Labeling of spTFIIE subunits

Both 6H-spTFIIE and 6H-spTFIIEß were [³⁵S]-labeled by IPTG-induction in E. coli BL21(DE3)pLysS in the presence of 0.15 mCi/mL [³⁵S]-methionine (Amersham Pharmacia) and 200 µg/mL rifampicin. [³⁵S]-6H-spTFIIEß was purified through Ni²⁺-NTA agarose as described above. Because the recovery of [³⁵S]-6H-spTFIIE by Ni²⁺-NTA affinity chromatography was low, the labeled protein was concentrated by 33% (v/v) saturated ammonium sulfate in 20 mM Tris-HCl (pH 7.9 at 4 °C) by ultracentrifugation at 20 000 g for 20 min at 4 °C using the 50.2 Ti rotor (Coulter-Beckman). The precipitate was resuspended in buffer B without KCl (BB0) and adjusted to 100 mM KCl before use in Far Western blotting experiments.

Far Western blotting analysis

Proteins were electrotransferred from an SDS-polyacrylamide gel to a PVDF membrane (Immobilon P, Millipore) and denatured twice with 6 M guanidine-HCl in BC100 for 30 min at 4 °C, followed by successive 10-min treatments with 3.0, 1.5, 0.75, and 0.375 M guanidine-HCl in BC100. The membrane was washed twice with BC100 and treated with BC100 containing 5% skim milk for 2 h at 4 °C. The membrane was then soaked in BC100 containing 200 µg/mL BSA and [³⁵S]-labeled probe protein (6H-spTFIIE or 6H-spTFIIEß) for 6 h at room temperature. The membrane was washed with BC100 and the bound [³⁵S]-labeled proteins were detected by autoradiography using Fuji RX-U X-ray film.

Purification of f-Pol II

This purification method was based on that previously described (Kimura et al. 2002). The strain JY741/f-rpb3 was cultured in YE medium containing 75 µg/mL of adenine and uracil at 30 °C. Cells were harvested at the exponential phase, suspended in four times the cell weight of buffer E (62.5 mM Tris-HCl (pH 7.9 at 4 °C), 12.5% (v/v) glycerol, 0.125 mM EDTA, 1.25 mM DTT, 0.625 mM PMSF, 2 µg/mL anti-pain, 2 µg/mL aprotinin, 1 µg/mL leupeptin, 0.8 µg/mL pepstatin) containing 125 mM (NH₄)₂SO₄ (pH 8.0) and disrupted with a French Press (Otake, Tokyo, Japan). The cell suspension was sonicated and centrifuged at 18 000 g for 20 min at 4 °C. The supernatant was diluted 4-fold with buffer F (50 mM Tris-HCl (pH 7.9 at 4 °C), 10% (v/v) glycerol, 0.1 mM EDTA, 1 mM DTT, 0. 5 mM PMSF) containing 100 mM (NH₄)₂SO₄ (pH 8.0) (BF100) and 0.1% polyethylenimine was added. After incubation for 1 h, the precipitate was collected by centrifugation at 18 000 g for 20 min and extracted by 2-fold cell weight of buffer F containing 200 mM (NH₄)₂SO₄ (pH 8.0) (BF200). After centrifugation at 18 000 g for 20 min, 20 microliters of M2-agarose (Sigma) equilibrated with BF200 was added and gently mixed for 2 h at 4 °C. The M2-agarose was washed four times with 1 mL of BF200, and FLAG-PolII (f-PolII) was eluted with BF200 containing 100 µg/mL FLAG-peptide (Sigma).

GST pull-down assay

GST-fusion proteins were used for protein interaction assays. Two hundred ng of each protein to be tested (or 500 ng in the case of f-spPol II) was incubated with lysates containing 300 ng GST-fusion proteins together with 10 µL (packed volume) glutathione-Sepharose (Amersham Pharmacia) in a 500-µL reaction volume in BC100 containing 200 µg/mL BSA for 4 h at 4 °C with rotation. The glutathione-Sepharose resin was then washed twice with 500 µL BC200 and once with 500 µL BC100, and boiled in SDS sample buffer. The proteins released from the resin were separated by SDS-PAGE and detected by Western blotting.

	DDBJ accession numbers

Top Abstract Introduction Results Discussion Experimental procedures DDBJ accession numbers Supplementary material References

 
The DDBJ accession numbers for the spTFIIE and spTFIIEß cDNA sequences are AB176672 and AB176673, respectively.

	Supplementary material

Top Abstract Introduction Results Discussion Experimental procedures DDBJ accession numbers Supplementary material References

 
The following supplementary material is available at http://www.blackwellpublishing.com/products/journals/suppmat/GTC/GTC833/GTC833.htm.

Supplementary Table S1 Strains and their genotypes used in this study.

Supplementary Table S2 Oligonucleotides used in this study.

	Acknowledgements

 
We thank Shigeiro Osada for critical reading of the manuscript, Hiroshi Nojima for S. pombe cell lines and the tetrad analyses, Hideki Hiyama, Yukinobu Nakaseko and Mitsuhiro Yanagida for advice with S. pombe. We also thank our colleagues in Osaka University for helpful discussion. This work was supported in part by the Grants-in-Aid for Scientific Research (F.H. and Y.O.) and the Grants-in-Aid for Protein 3000 Project (Y.O.) from the Japanese Ministry of Education, Culture, Sports, Science and Technology, the Core Research for Evolutional Science and Technology (CREST) (F.H.), and the Biodesign Research Program of the Institute of Physical and Chemical Research (RIKEN) (F.H.).

	Footnotes

 
Communicated by: Hiroshi Handa

^aPresent address: Discovery Research Laboratories II, Pharmaceutical Research Division, Takeda Chemical Industries Ltd, Tsukuba, Ibaraki 300-4247, Japan

^bPresent address: Department of Biological Chemistry, Faculty of Pharmaceutical Sciences, Toyama Medical and Pharmaceutical University, 2630 Sugitani, Toyama 930-0194, Japan

^* Correspondence: E-mail: ohkumay{at}ms.toyama-mpu.ac.jp

	References

Top Abstract Introduction Results