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

J. Biol. Chem., Vol. 275, Issue 35, 26925-26934, September 1, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/35/26925    most recent M002701200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend 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 Lemaire, M. Articles by Collart, M. A. Search for Related Content PubMed PubMed Citation Articles by Lemaire, M. Articles by Collart, M. A. Social Bookmarking                      What's this?

The TATA-binding Protein-associated Factor yTaf_II19p Functionally Interacts with Components of the Global Transcriptional Regulator Ccr4-Not Complex and Physically Interacts with the Not5 Subunit^*

Marc Lemaire and Martine A. Collart^§

From the Département de Biochimie Médicale, Céntre Médical Universitaire, 1 rue Michel Servet, 1211 Geneva 4, Switzerland

Received for publication, March 28, 2000, and in revised form, May 31, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The Saccharomyces cerevisiae HIS3 gene is a model system to characterize transcription initiation from different types of core promoters. The NOT genes were identified by mutations that preferentially increased transcription of the HIS3 promoter lacking a canonical TATA sequence. They encode proteins associated in a complex that also contains the Caf1 and Ccr4 proteins. It has been suggested that the Ccr4-Not complex represses transcription by inhibiting factors more specifically required for promoters lacking a TATA sequence. A potential target is the yTaf_II19 subunit of TFIID, which, when depleted, leads to a preferential decrease of HIS3 TATA-less transcription. We isolated conditional taf19 alleles that display synthetic growth phenotypes when combined with not4 or specific not5 alleles. Inactivation of yTaf_II19p by shifting these mutants to the restrictive temperature led to a more rapid and striking decrease in transcription from promoters that do not contain a canonical TATA sequence. We demonstrated by the two-hybrid assay and directly in vitro that yTaf_II19p and Not5p could interact. Finally, we found by the two-hybrid assay that yTaf_II19p also interacted with many components of the Ccr4-Not complex. Taken together, our results provide evidence that interactions between Not5p and yTaf_II19p may be involved in transcriptional regulation by the Ccr4-Not complex.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Transcription initiation by RNA polymerase II involves the assembly of a functional preinitiation complex on the core promoter (1). An essential step in this assembly is the recognition of the core promoter by general transcription factors. The TBP¹ subunit of TFIID plays a crucial role in this recognition event for TATA-containing promoters (2). For promoters that do not contain canonical TATA sequences (referred as TATA-less), other factors probably contribute to the correct positioning of the polymerase. Biochemical analyses indicate that the Taf_II subunits of TFIID make extensive contacts to the core promoter independently of the TATA element (3). This has implicated Taf_IIs themselves in participating in the core promoter recognition event and has suggested that Taf_IIs might be particularly important for recognition of TATA-less core promoters. In any event, this would define the general transcription factor TFIID as being the key player in core promoter recognition.

However, the role and mechanisms of action of the Taf_IIs still remain unclear. Recently Taf_II-containing complexes distinct from TFIID have been described in mammalian cells and in yeast (4-7). These complexes share some but not all TFIID Taf_II subunits. One example is the yeast SAGA histone acetyltransferase complex. These findings might call into question the presence of TFIID at all promoters. It could be that other Taf_II-containing complexes function at some promoters. Such a possibility has been further supported by studies of the in vivo role of many Taf_IIs in yeast (for reviews, see Refs. 8 and 9). Indeed, it appears that some Taf_IIs may be generally required for transcription (such as yTaf17_IIp), whereas others function only at some core promoters (such as yTaf145_IIp).

To understand in detail the mechanism of transcription initiation at TATA-less promoters, in vitro studies have not been very useful, because transcription initiation from promoters that lack both a TATA box and an initiator sequence is usually inefficient. In yeast, the HIS3 gene has been used as a model to investigate the differences between TATA-containing and TATA-less core promoter transcription initiation. Indeed, the HIS3 promoter contains both types of core promoters, which are functionally distinguishable (10, 11), because activation by upstream bound activators only functions through the TATA promoter. In Taf_II depletion assays, transcription from the HIS3 TATA-less promoter has been shown to decrease preferentially in some cases. This has led to the suggestion that yTaf_II19p, yTaf_II145p, yTaf_II40p, and yTaf_II67p are more specifically required for TATA-less transcription (12, 13). In contrast, recent work has suggested that in fact yTaf_II40p is generally required for transcription by RNA polymerase II (14). yTaf_II40p and yTaf_II19p are the yeast homologues of huTaf_II28p and huTaf_II18p, two subunits of the human TFIID complex interacting via histone-fold dimerization domains (15). yTaf_II145p is the yeast homologue of huTaf_II250 and was presumed to be the TFIID scaffold, although this belief has been challenged by recent work (discussed in Ref. 9). Not much has been described about yTaf_II67p.

In other studies, the five NOT genes have been identified by mutations that increase HIS3 transcription. The Not proteins preferentially repress transcription from the HIS3 TATA-less promoter and have been described as global regulators of transcription, as they also affect the transcription of many unrelated genes (16-18). The Not proteins are associated in one or multiple large complexes (16) that also contain the Ccr4 and Caf1 proteins, known to be required for nonfermentative gene expression (19). It has been suggested that the Ccr4-Not complex might function to repress transcription of TATA-less promoters by sequestering or inhibiting factors more specifically required for TATA-less transcription (18). A putative target of the Ccr4-Not complex is TFIID (or some of its subunits) because of its probable implication in core promoter recognition. At the present time, there is no experimental evidence to support this model. Not1p has been reported to co-immunoprecipitate with TBP (20), but other experiments have shown that transcriptional activity resulting from a functional Spt3p-TBP interaction is a target for repression by the Not1p (21). Furthermore, interactions between Not2p and the Ada proteins have also been reported (22). Taken together, these results might point to the SAGA complex rather than TFIID.

To further understand the mechanisms involved in transcription regulation by the Ccr4-Not complex, we sought to investigate whether yTaf_II19p, a TFIID subunit apparently preferentially required for transcription of the HIS3 TATA-less promoter, interacted in any way with the Ccr4-Not complex. Our interest in yTaf_II19p in particular was raised partly from the observations that (i) Spt3p and yTaf_II19p carry homologous sequences (15) and are thought to play similar roles in the SAGA and TFIID complexes, respectively (discussed in Ref. 9), and (ii) Spt3p appears to be a target for transcriptional regulation by the Ccr4-Not complex (21). We isolated temperature-sensitive taf19 alleles and found that they displayed, at permissive temperature, striking synthetic slow growth phenotypes with not4 and not5 mutants on minimal medium. In particular, two not5 alleles encoding truncated proteins of different lengths behaved differently when combined with mutant taf19 alleles but not with wild type TAF19. This suggested that yTaf_II19p and Not5p might interact, a hypothesis that could be confirmed both by two-hybrid experiments and by a direct interaction between Not5p and yTaf_II19p in vitro. Our results provide the first evidence that adequate transcription regulation by the Ccr4-Not complex may involve interactions with TFIID Taf_IIs.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Strains and Media-- All strains are described in Table I and were generated by classical genetic techniques. Media were standard. Strains carrying the kanMX4 module were selected for on YPD plates supplemented with G418 (200 mg/l, Life Technologies). Escherichia coli DH5 and BL21 (DE3) were used as cloning host and for recombinant protein expression, respectively.

TAF19 Gene Disruption-- TAF19 complete disruption was obtained through homologous recombination by transformation of PCR¹-synthesized marker cassettes with long flanking homology regions into MY542 (23). Briefly, by using two consecutive PCRs, upstream and downstream regions of TAF19 (containing start and stop codons, respectively) were placed at each end of the selectable kanMX4 cassette from pFA6a-kanMX4. The primers used for the TAF19 long flanking homology-PCR synthesis were as follows: P5', 5'-AAA AGT CGA CTC CTC TGC ACG TCC AAC ACC C-3' (the HincII site is underlined; the region starting 304 base pair upstream of the TAF19 start codon is in boldface); P5'L, 5'-GGG GAT CCG TCG ACC TGC AGC GTA CGC ATA TCT TAT CCA GCT CAC CC-3' (the reverse complement of the TAF19 start codon is in boldface and underlined; the TAF19 5' upstream region is in boldface; and the kanMX4 region is in plain text); P3'L, 5'-AAA CGA GCT CGA ATT CAT CGA TGA TAT GAT ATA GCT ACT TGG CAG GC-3' (the TAF19 stop codon is in boldface and underlined; the TAF19 3' downstream region is in boldface; and the kanMX4 region is in plain text); and P3', AAA ACT GCA GTA GGA GGC GCA CGT ACC TTC C (the PstI site is underlined; the region starting 398 base pair downstream of the TAF19 stop codon is in boldface). Correct integrations were verified by PCR by using P5', P3', and KanMX4 internal primers.

Plasmids-- pMAC195 is a URA3 centromeric plasmid that expresses a fully functional full-length yTaf_II19p from the DED1 promoter. pMAC186 is a pUCBM21 derivative carrying the TAF19 ORF cloned in the vector EcoRV-SacI sites. pML25 is a derivative of the pPC86 plasmid (24) containing the TAF19 ORF cloned between the promoter and 3' untranslated sequences of ADC1. pMAC197 is a pET15b derivative carrying the TAF19 sequences. It was created by the cloning of the EcoRV-BamHI fragment of pMAC186 into pET15b. The TAF19 sequences of pML25 were replaced with the taf19-1, taf19-7, and taf19-9 mutant sequences, leading to pML64, pML65, and pML66. pML98 is a pLex202 derivative (25) that encodes the LexA-yTaf_II19 fusion protein from the ADC1 promoter. pML135, pML132, and pML133 are the same fusion to the mutant yTaf_II19p proteins. pML69, pML70, and pML71 are pPC62 (24) derivatives expressing yTaf_II19p, yTaf_II19-1p, and yTaf_II19-9p from the ADC1 promoter. pML63 carries a GST-TAF19 fusion cloned under the control of the arabinose-inducible E. coli AraC promoter. pML136 was generated by cloning a Xba-HindIII fragment encoding a tagged version of yTaf_II40p from pRS415-TAF40 (26) into pRS424.

pMAC253 is a pET15b derivative that carries a 1.6-kilobase pair HindIII-BamHI fragment of CCR4, starting 472 nucleotides downstream of the ATG, thus expressing a truncated protein.

pMPM272 encodes GST from the arabinose-inducible E. coli AraC promoter (kindly provided by Mathias Mayer).

pU61 is a pET22b/pET15b derivative encoding a His₆-Not5 fusion protein (the fusion is at the N terminus of Not5p). A stop codon and 3' untranslated NOT5 sequences separate the NOT5 ORF and the pET22b His₆ sequences. Plasmids are summarized in Table II.

Cloning details are available upon request.

Isolation of taf19 Mutant Alleles-- pMAC186 was used as DNA template for TAF19 PCR mutagenesis by using the classical forward and reverse primers and the Taq polymerase (Life Technologies, Inc.). PCR products were pooled and cloned in pML25 to replace the TAF19 wild type allele. We obtained a library of 11 × 10³ independent transformants that was transformed into MLY187.

Analysis of Transcript and Proteins Levels in TAF19 and taf19-1 Strains-- After dilution from an overnight culture in rich medium, wild type and mutant cells were grown in rich medium to an A₆₀₀ of 0.3 at 30 °C. The cultures were then shifted to 37 °C. At each time point indicated thereafter, total cellular RNA from the equivalent of 10 A₆₀₀ of cells was extracted and analyzed by S1 nuclease protection assay as previously published (18, 27). The oligonucleotide for NOT5 mRNA analysis was 5'-GCG AGG CTG ATT CTA CAC CTG GCG CGA TTG GAG TCG TCG CCC TGT CTG ATA TAG AAA CAT CCC AAC AAC AA-3'. In parallel, equal amounts of cells (equivalent to 1 unit of A₆₀₀) were harvested by rapid centrifugation and washed with cold water, and total proteins were extracted according to Ref. 28. For this procedure, the frozen cell pellet was thawed on ice in 150 µl of lysis buffer (1.85 M NaOH, 7.4% -mercaptoethanol), vortexed, and left 10 min on ice. Proteins were then trichloroacetic acid-precipitated by the addition of 150 µl of trichloroacetic acid 50% and resuspended in 80 µl of SDS-PAGE sample buffer (40 µl of 0.1 M NaOH and 40 µl of 2× sample buffer). Equal amounts of total cell extracts (10 µl) were then fractionated by SDS-PAGE and analyzed by Western blot using chemiluminescence (Pierce). Antibodies against yTaf_IIs and TBP were kindly provided by Joe Reese and Anthony Weil. Antibodies against Not1p and Not5p were described previously (16, 21). Antibodies against yTaf_II19p and Ccr4p were raised. Briefly, recombinant yTafI_I19p was expressed from pMAC197, and recombinant Ccr4p was expressed from pMAC253, in BL21. The recombinant proteins were purified according to standard protocols (Qiagen) and were injected into rabbits (Elevage Scientifique des Dombes). Antibodies to yTaf_II19p were used at 1:3000, and those to Ccr4p were used at 1:8000.

Two-hybrid Interactions-- To test protein-protein interactions, pLex202 derivatives were cotransformed into EGY48 (25) with pJG4-5 derivatives encoding galactose-inducible B42 fusions to Not proteins (16, 18), as well as to Ccr4p and Caf1p (19). Protein-protein interactions were scored as function of growth on synthetic galactose but not glucose minimal medium lacking leucine.

-Galactosidase Assays-- Strains carrying the fusion proteins to be tested were grown overnight in 1 ml of glucose minimal medium supplemented with leucine. The cells were washed twice in water and resuspended in 10 ml of galactose minimal medium supplemented with leucine. After 24 h, the cultures were collected, and -galactosidase assays performed as described previously (17).

GST Pull-down Analysis-- Unless otherwise stated, protein manipulations were performed at 4 °C. E. coli BL21(DE3) was transformed either with pMPM272 (encoding GST), pML63 (encoding GST-yTaf_II19p), or pU61 (encoding His₆-Not5p). Transformed cells were grown at 30 °C to an A₆₀₀ of 0.6 in 100 ml of rich medium containing ampicillin (100 µg/ml), and recombinant proteins were induced by addition of arabinose (final concentration, 0.5% in the case of GST and GST-yTaf_II19p) or isopropyl-1-thio--D-galactopyranoside (final concentration, 0.5 mM for His₆-Not5p). Cells were allowed to grow at 37 °C for 4 h, harvested, washed with cold water, resuspended in 5 ml of Buffer A (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM -mercaptoethanol, 1 mM phenylmethylsulfonyl fluoride) supplemented with RNase A and DNase I (0.1 mg/ml, Sigma), and broken by sonication. Cellular debris and aggregates were discarded by centrifugation (20 min at 40,000 × g), and supernatants were supplemented with glycerol (final concentration, 10%) and kept frozen until used. Equal total protein amounts (200 µg) of His₆-Not5p extract were mixed with GST extract, GST-yTaf_II19p extract, or Buffer A, and the volume was brought to 100 µl with Buffer A and Tween 20 (final concentration, 0.1%), making Buffer B. The concentration of His₆-Not5p and GST-yTaf_II19p in these extracts was roughly similar, with perhaps a 3-fold higher level of the former, as determined by Coomassie staining. Tubes were incubated at 30 °C for 2 h. Reactions were centrifuged for 20 min at 40,000 × g, and 25 µl of a 50% gel slurry of glutathione-Sepharose beads (Amersham Pharmacia Biotech), previously equilibrated with Buffer B, was added to each supernatant. Suspensions were incubated at 30 °C for 1 h with a mild agitation and then centrifuged at 2000 × g for 1 min. Supernatant was kept on ice (fraction S), and beads were washed three times with 100 µl of cold Buffer B. The third wash was kept on ice (fraction W), and beads were resuspended with 100 µl of SDS-PAGE loading buffer (fraction B). Equal amounts (10 µl) of fractions S, W, and B were analyzed by Western blotting using antibodies directed against Not5p.

Yeast Extract Preparation-- For preparation of total cell extracts 12 liters of cells grown to an A₆₀₀ of 4.5 were pelleted and washed, and the pellet was frozen at 80 °C. This pellet was then slowly thawed and resuspended in 100 ml of lysis buffer (29), including 1 mM dithiothreitol and protease inhibitors. We additionally added a tip of DNase I (Sigma). Cells were broken in the cold with a French press (SLM AMINCO 20K Cell FA-073) three times at 1500 p.s.i. Then, the suspension was clarified first by 20 min at 16,000 × g and 1 h at 35000 rpm in a Beckman Ti35 ultracentrifuge.

For fractionation by ammonium sulfate precipitation, first 109 g/liter were added to the extract resulting in 30% ammonium sulfate. The precipitate was removed by ultracentrifugation (1 h centrifugation at 35,000 rpm in a Ti35 rotor) and 86 g/liter were then added to the supernatant leading to a 45% ammonium sulfate solution. The precipitate was collected by the same ultracentrifugation procedure, and 59 g/liter were added to the supernatant leading to 55% ammonium sulfate concentration. Finally, after collecting the precipitate the same way, 93 g/liter were added to the supernatant, leading to 70% ammonium sulfate. The pellets were resuspended in the minimal amount of lysis buffer carrying additionally 0.1% Nonidet P-40 and then dialyzed against 100 volumes of the same buffer. 300 µl of the dialyzed 45% cut of the cell extract was then analyzed by gel filtration (see below).

Gel Filtration Analysis-- For gel filtration analysis, 300 µl of total cell extracts were loaded on a Superose 6 gel filtration column equilibrated with 350 mM NaCl, 10% glycerol, 0.1% Tween 20, and 40 mM Hepes, pH 7.3. The column was run at 0.4 ml/min, and 400 µl fractions were collected starting at 16 min and analyzed by Western blot analysis for the presence of yTaf_II19p (10-15% Tris Tricine gel). The position of the void volume was determined by the elution of salmon sperm DNA.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Isolation of Conditional Alleles of TAF19-- We have suggested that the Ccr4-Not complex regulates transcription by sequestering or inhibiting transcription factors more specifically required for TBP function at TATA-less promoters. A potential candidate for such a factor is yTaf_II19p. To start investigating whether yTaf_II19p is associated with Ccr4-Not function, we used a genetic approach and first created a library of mutant TAF19 alleles. TAF19 was mutagenized by error-prone PCR amplification of TAF19, and the library of mutant alleles was cloned into a yeast TRP1 centromeric plasmid (pML25). This library was transformed into MLY187, a yeast strain carrying a complete disruption of the genomic TAF19 gene complemented by pMAC195, a URA3 centromeric plasmid carrying a wild type copy of TAF19. Transformants were streaked on 5-fluoroorotic acid to select for the loss of the episomal wild type copy of the TAF19 gene. The 5-fluoroorotic acid-resistant transformants were streaked on rich medium at 16, 30 and 37 °C to determine whether conditional mutants had been isolated.

Out of this screen, we further characterized three taf19 mutants, called 19-1, 19-7, and19-9. taf19-7 grew slowly on rich medium and at all temperatures, whereas the other two mutants grew well on rich medium at 30 °C but were temperature-sensitive (Fig. 1A). These phenotypes were recessive. The mutant alleles were sequenced. taf19-1 carries four mutations that result in four amino acid changes (D46G, L63H, L79D, and K98M), taf19-9 also carries four mutations that result in four amino acid changes (D24G, E38G, N57D, and F70S), and finally, taf19-7 is a point mutant that results in a single amino acid change (K13E). Fig. 1B shows the position of these mutations within the yTaf_I119p.

View larger version (49K): [in this window] [in a new window]   Fig. 1.   Characterization of taf19 mutant alleles. A, growth of wild type and mutant taf19 strains. Two temperature-sensitive mutants (taf19-1 and taf19-9) and one sick mutant (taf19-7) were isolated and streaked together with the isogenic wild type TAF19 strain on YPD plates at 30 and 37 °C as indicated. Strains used were MLY268 (TAF19), MLY270 (taf19-1), MLY272 (taf19-7), and MLY274 (taf19-9). B, schematic structure of yTafII19p with the position of amino acid substitutions in the mutant yTafII19 proteins. This scheme is based on the alignment in Ref. 15. C, growth curve of wild type or taf19 mutants at 37 °C. MLY268, MLY270, and MLY274 were grown in YPD at 30 °C to an A600 of 0.3, and the cultures were then shifted to 37 °C. In parallel, cultures that were kept at 30 °C grew similarly, independently of the TAF19 allele (data not shown).

Because taf19-1and taf19-9 carry multiple mutations, we tried to determine whether any individual mutation was responsible for the temperature-sensitive phenotype. Many of these mutations lie in the domain of yTaf_II19p that is thought to be involved in dimerization with yTaf_II40p (15). Particularly, residue Asp-24 is conserved between yTaf_I119p and orthologs from human, Drosophila, and Caenorhabditis elegans. In the three-dimensional structure (huTaf_II18-huTaf_II28 heterodimer structure), the huTaf_II18p Asp-45 residue (corresponding to yTaf_II19p Asp-24) is involved in a strong hydrogen bond network that stabilizes the heterodimer. On the other hand, this heterodimer is also stabilized by multiple hydrophobic interactions occurring at the crossover of two -helices (in huTaf_II18p, residues 60-70 of the 2-helix). A mutation lying in this area (yTaf_II19p Asp-46 corresponds to huTaf_II18p Glu-67) might induce a local disorganization of the 2-helix, resulting in the destabilization of the heterodimer. Surprisingly, in the case of taf19-1, neither D46G alone nor an allele carrying the other three mutations (L63H, L79D, and K98M) conferred a temperature-sensitive phenotype (data not shown). Thus, temperature sensitivity is conferred by D46G and at least one other mutation. Similarly, in the case of taf19-9, D24G alone does not confer temperature sensitivity. However, temperature sensitivity might still result from a disruption of this dimerization domain, which in turn might require more than one mutation. This hypothesis was confirmed by the fact that overexpressing TAF40 (from plasmid pML136) suppressed the temperature sensitivity of taf19-1 and taf19-9 (MY2268-MY2271; see Table I) at 36 °C (data not shown). In this context, it is interesting to note that temperature-sensitive taf40 alleles were also found to carry multiple mutations within the equivalent yTaf_II40p dimerization domain and that this temperature-sensitive phenotype was also alleviated by overexpressing yTaf_II19 (14).

Phenotypic Analysis of the taf19-1 and taf19-9 Mutants-- Not much is known about TAF19. We thus first further characterized the two conditional alleles that we had isolated, namely taf19-1 and taf19-9. Fig. 1C shows that both mutants rapidly arrest cell growth as the cells are shifted to the restrictive temperature. Viability of the taf19 mutants was assayed at given times after the temperature shift and was found to be very weakly affected only after 6 h at the restrictive temperature (data not shown).

Our interest in isolating taf19 mutant alleles was based upon a previous description of the preferential loss of transcription from TATA-less promoters upon yTaf_II19p depletion (12). However, similar results were described with yTaf_II40p, yet, as mentioned above, more recently, yTaf_II40p has been said to be generally required for transcription (14). To determine what phenotype our conditional mutants displayed at restrictive temperature upon loss of yTaf_II19p function, total cellular RNA was extracted from wild type and taf19-1 cells just prior to the shift to the restrictive temperature and at given times after the shift. This RNA was analyzed by S1 mapping first for the levels of the DED1 transcript expressed from a TATA-containing promoter, and the NOT5 and HIS4 transcripts that do not depend upon a canonical TATA sequence. It should be clarified that we claim that NOT5 has no canonical TATA based on the sequence of the promoter region. In the case of HIS4, this has been previously reported (30). In taf19-1 cells, within 4 h after the shift, the levels of the HIS4 and NOT5 transcripts decreased to a nondetectable level, whereas the level of the DED1 transcript remained stable (Fig. 2A). At later time points (6 h), the DED1 transcript also decreased (data not shown). Because the HIS4 and DED1 transcripts have been described to have similar half-lives (27), this result suggests that HIS4 transcription is more rapidly affected than DED1 transcription. Because we realized that the taf19 mutants isolated grew slowly on minimal medium lacking threonine or isoleucine, we also investigated the ILV1 transcript levels in wild type and mutant cells shifted to the restrictive temperature for 4 h, because the ILV1 gene lies in the pathway of the biosynthesis of both amino acids. This mRNA has been reported to be transcribed from a TATA-less promoter (31), and it was also dramatically decreased 2 h after the shift to the restrictive temperature (Fig. 2B). We then similarly analyzed the levels of an unstable tRNA (27) and found that transcription by RNA polymerase III was not measurably affected in either strain (Fig. 2B). Finally, we looked at HIS3 transcript levels and found that whereas transcription of HIS3 from both promoters decreased upon inactivation of yTaf_II19p, that from the TATA-less promoter decreased more rapidly and more severely (Fig. 2C, compare lane 4 to lane 1). Taken together these results suggest that the loss of yTaf_II19p function affects more rapidly transcription from the TATA-less promoters (HIS3, HIS4, NOT5, and ILV1) than from the canonical TATA-containing promoters (HIS3 and DED1).

View larger version (49K): [in this window] [in a new window]   Fig. 2.   TAF19 may generally affect transcription, but it is preferentially required for transcription from TATA-less promoters. A, MLY268 and MLY270 were grown in YPD at 30 °C to an A600 of 0.3, and cultures were then shifted to 37 °C. Total RNA was extracted from TAF19 and taf19-1 strains at the indicated time points after shifting to 37 °C. 50 µg of total RNA were analyzed by S1 nuclease protection assay for the levels of the indicated transcripts. Transcription of HIS4 depends upon a TATA-less core promoter in gcn4 strains (30), and sequence analysis shows that the NOT5 promoter does not contain a canonical TATA sequence. In contrast the DED1 promoter carries a canonical TATA sequence. The hybridizations were internally controlled by simultaneous hybridization of NOT5 and HIS4 with DED1. Similar results were obtained with the taf19-9 temperature-sensitive mutant (data not shown). B, the same experiment was performed, and the total cellular RNA was hybridized to measure the levels of ILV1 and WtRNA. The ILV1 is also thought to carry a TATA-less promoter. C, the same experiment was performed with strain MLY270 alone, and total cellular RNA was hybridized to measure the levels of the HIS3 and DED1 transcripts.

Decreased Steady State Level of TFIID Components upon yTaf_II19p Inactivation-- It has not been demonstrated that yTaf_II19p is definitively part of TFIID. It is thought to be, by homology with its human counterpart (15) and because it co-immunoprecipitates with both yTaf_II145p and TBP (12). Because it has been reported that depletion of given yTaf_IIs leads to degradation of other yTaf_IIs that are part of the same complex, we investigated whether inactivation of yTaf_II19p similarly affected the steady state levels of other yTaf_IIs. Total protein extracts were prepared in parallel to the RNA mentioned above, and they were analyzed by Western blot for the levels of different yTaf_IIs. As shown on Fig. 3, in taf19-1 mutant cells, the levels of specific TFIID yTaf_IIs (yTaf_II145p and yTaf_II40p) rapidly decreased upon inactivation of yTaf_II19p. The levels of other yTaf_IIs, shared between the TFIID and SAGA complexes, such as yTaf_II60p, yTaf_II68p, and yTaf_II90p, or even with the SWI/SNF complex, such as yTaf_II30p, decreased somewhat slower or were unaltered after 4 h at the restrictive temperature. TBP levels decreased rapidly to a lower stable level. The levels of yTaf_II19p itself were undetectable in up to 100 µg of total cell extract with the antibodies that we raised.

View larger version (65K): [in this window] [in a new window]   Fig. 3.   Decreased steady state levels of yTafIIs, TBP, and Ccr4-Not complex subunits upon inactivation of TAF19. MLY268 (TAF19) and MLY270 (taf19-1) were grown in YPD at 30 °C to an A600 of 0.3, and cultures were then shifted to 37 °C. At different time points (as indicated) after the shift at 37 °C, total proteins were extracted from equal amounts of cells (see under "Experimental Procedures"). Protein extracts from equal amounts of cells were then fractionated by SDS-PAGE and blotted to nitrocellulose. The levels of different TFIID and Ccr4-Not complex subunits (as indicated) were analyzed by Western blot.

These results are very similar to what has been described previously for temperature-sensitive taf40 mutants (14) and support the presence of yTaf_II19p in TFIID. They also suggest that the inactivation of yTaf_II19p leads to the destabilization of TFIID through the degradation of some specific TFIID Taf_IIs (e.g. yTaf_II40p and yTaf_II145p). Moreover, yTaf_II25p, a yTaf_II shared by TFIID and SAGA, was rapidly decreased upon yTaf_II19p inactivation. In that context, it is interesting to note that upon yTaf_II25p inactivation, yTaf_II19p was also rapidly degraded (32).

We also investigated the levels of some of the components of the Ccr4-Not complex. All of the components that were analyzed, namely Not1p, Not5p, and Ccr4p, also decreased, but not as dramatically, nor as quickly, as the yTaf_IIs. Because, at least in the case of NOT5, we have found that transcription is affected by the loss of yTaf_II19p, the decrease in Not5p may be a consequence of protein turnover. In any event, it is important to note that the levels of the different Ccr4-Not components were similar in the wild type and mutant strains at the permissive temperature.

taf19 Mutants Display Synthetic Phenotypes with Specific not Mutants-- We next used a genetic approach to determine whether the function of yTaf_II19p was related to that of the Ccr4-Not complex. We constructed a large number of double mutants by crossing taf19-1 and taf19-9 to many ccr4-not mutants, sporulating diploids and dissecting tetrads. The phenotypes of the double mutants were compared with the phenotypes of the single mutants to look for synthetic lethal interactions or suppression. In particular, growth on rich medium at 16, 30, and 37 °C as well as growth on minimal medium at permissive temperature was analyzed. not1-1, not1-2, not2-1, not2-4, not3::URA3, caf1::LEU2, and ccr4::URA3 mutants did not show any obvious genetic interaction with taf19-1 or taf19-9. In contrast, not4 and not5 mutants did. not4-1, but more dramatically not4::LEU2, showed a synthetic phenotype when combined with both taf19-1 and taf19-9 on minimal medium at 30 °C (Fig. 4). This same effect was observed when the minimal medium was complemented with histidine (data not shown). A slight synthetic growth phenotype could also be detected on rich medium at 30 °C (not shown). More interestingly, not5::LEU2 and not5-1, but not not5-2, displayed a dramatic synthetic growth phenotype on minimal medium when combined with either one of the two taf19 mutants (Fig. 4). The two alleles, not5-1 and not5-2, are similar in that they both carry nonsense mutations (16), but they can be distinguished by the fact that the protein encoded by not5-1 is shorter (see Fig. 4, bottom). Such an allele-specific synthetic phenotype strongly supports the fact that yTaf_II19p and Not5p functionally interact for growth on minimal medium and may even be physically associated. Alternatively, yTaf_II19p and the Ccr4-Not complex may participate independently in transcriptional regulation required for growth on minimal medium, and the contribution of the Ccr4-Not complex may be only seriously impaired when Not4p is absent or when Not5p is sufficiently truncated.

View larger version (103K): [in this window] [in a new window]   Fig. 4.   Synthetic slow growth of double taf19 and not4 or not5 mutants on minimal medium. Double mutants, isogenic single mutants and wild type strains were streaked together on minimal plates lacking histidine at 30 °C. Strains used were MLY268 (TAF19), MLY270 (taf19-1), MLY274 (taf19-9), YOU584 (not4), MLY365 (not4 taf19 + TAF19), MLY367 (not4 taf19 + taf19-1), MLY369 (not4 taf19 + taf19-9), MY1719 (not5), MLY329 (not5 taf19 + TAF19), MLY331 (not5 taf19 + taf19-1), MLY333 (not5 taf19 + taf19-9), YOU123 (not5-1), MLY347 (not5-1 taf19 + TAF19), MLY349 (not5-1 taf19 + taf19-1), MLY351 (not5-1 taf19 + taf19-9), YOU142 (not5-2), MLY353 (not5-2 taf19 + TAF19), MLY355 (not5-2 taf19 + taf19-1), and MLY357 (not5-2 taf19 + taf19-9). At the bottom of the figure are shown schemes of the Not5 wild type and mutant proteins. The growth of the various strains was no different when the plates were supplemented with histidine (data not shown).

TAF19 Interacts by Two-hybrid with NOT5-- To investigate further whether or not yTaf_II19p and Not5p may interact, we performed a two-hybrid analysis. A plasmid expressing a LexA-yTaf_II19 fusion protein was constructed (pML98, see under "Experimental Procedures") and found to complement a null mutation of TAF19. It was transformed into a strain carrying a leu2 gene under the control of LexA operators (EGY48; see Table I). Additional plasmids were co-transformed that expressed fusions of all the known components of the Ccr4-Not complex to the B42-activation domain (16, 18) under the control of the GAL1 promoter. Growth on glucose or galactose plates devoid of leucine was investigated. Fig. 5 shows that a positive two-hybrid interaction could be detected among TAF19 and NOT5, NOT3, NOT2, and CAF1 (the latter two to a lesser extent). Growth of all of these strains on galactose minimal medium supplemented with leucine was compared and found to be indistinguishable, except that the strains carrying B42-NOT2 and B42-NOT3 grew somewhat more slowly (data not shown). To confirm these results, the expression of -galactosidase was measured in the same strains after growth for 24 h in liquid galactose minimal medium supplemented with leucine. Table III summarizes these results. With this second reporter, an interaction between yTaf_II19p and Not2p, Not3p, and Not5p is confirmed, and an interaction between yTaf_II19p and Ccr4p is additionally detectable. No interaction can be measured between yTaf_II19p and Caf1p with this second reporter. Finally, no interaction between Not1p or Not4p and yTaf_II19p was detectable with either of the two reporters. These results support the existence of a physical interaction between yTaf_II19p and components of the Ccr4-Not complex, in particular Not5p, that was suggested by the allele-specific synthetic growth phenotypes presented above.

View larger version (40K): [in this window] [in a new window]   Fig. 5.   TAF19 interacts with NOT5 in the two-hybrid assay. A, EGY48 was transformed with pML98 (LexA-yTafII19p) together with pJG4-5 derivatives expressing either the B42-Not1, B42-Not2, B42-Not3, B42-Not4, B42-Not5, B42-Ccr4, or B42-Caf1 fusion proteins as indicated. B, EGY48 was transformed with the plasmid expressing the B42-Not5 fusion protein, together with plasmids encoding fusion proteins of LexA to either wild type or mutant forms of yTafII19p as indicated. In both cases, transformants were grown overnight in synthetic glucose medium supplemented with leucine. Cells were harvested, washed twice in cold water, and allowed to grow for 4 h in synthetic galactose medium supplemented with leucine. Equal amounts of cell culture (1 A600 unit) were washed with water, serially diluted, and spotted either on YPD, glucose minimal medium (Glu), or galactose minimal medium (Gal). The same strains were streaked on galactose minimal medium supplemented with leucine and grew indistinguishably, except the strains expressing B42-Not2p and B42-Not3p that grew somewhat slower (data not shown).

View this table: [in this window] [in a new window]   Table III -Galactosidase assays to measure two-hybrid interactions The indicated strains are described in Table I, and all carry the LexA-LacZ reporter gene. All of these strains carry LexA-TafII19p and the indicated B42 fusion protein. The values for -galactosidase activity indicated for two separate experiments were calculated in nmol × mg1 min1.

One way to investigate further whether the interaction between yTaf_II19p and Not5p is functionally relevant is to determine whether any of the taf19 mutants isolated are defective in this interaction. We thus introduced the taf19 mutations into the construct expressing LexA-yTaf_II19p (pML132, pML133, and pML135; see under "Experimental Procedures"). Except for LexA-yTaf_II19-7p, all new fusion proteins complemented the null mutation of TAF19 and were expressed at the same level as LexA-yTaf_II19 (data not shown). The new fusions were tested for a two-hybrid interaction with B42-Not5p (see Fig. 5, bottom panel). No interaction was detected between LexA-yTaf_II19-1p and B42-Not5p, whereas in contrast, the LexA-yTaf_II19-9 protein interacted with B42-Not5p, in a manner indistinguishable from LexA-yTaf_II19p. Thus, not all taf19 mutants isolated are detectably defective in yTaf_II19p-Not5p interaction, but the finding that one is, is additional support for a functional physical association of yTaf_II19p and Not5p in vivo. Indeed, yTaf_II19-1p and LexA-yTaf_II19-1p are functional at the permissive temperature, because they replace a wild type yTaf_II19p for vegetative growth. This indicates that the proteins are correctly folded. Hence, the absence of interaction with Not5p indicates that mutated residues in yTaf_II19-1p are important for interaction with Not5p.

yTaf_II19p and Not5p Associate Physically in Vitro-- Because the two-hybrid experiments described above are performed in vivo, they do not define whether or not yTaf_II19p and Not5p can interact directly. To address such a question, we prepared bacterial extracts from E. coli expressing GST-yTaf_II19p, GST alone, or His₆-Not5p recombinant proteins to test their interaction in vitro. The expression and solubility of all proteins was verified by analyzing the total soluble bacterial extracts by Western blotting with antibodies to the GST moiety or to Not5p (data not shown). Not5p was detected as multiple forms (as can be seen in S lanes of Fig. 6), most likely resulting from protein degradation. Most of these detectable forms were capable of binding Ni-nitrilotriacetic acid agarose (data not shown), suggesting that they were stable C-terminal truncations.

View larger version (44K): [in this window] [in a new window]   Fig. 6.   yTafII19p interacts directly with Not5p in vitro. Bacterial extracts containing His6-Not5p were mixed with bacterial extracts devoid of any recombinant protein (left three lanes), with bacterial extracts containing recombinant GST-yTafII19p (middle three lanes), or with bacterial extracts containing recombinant GST alone (right three lanes). After 2 h at 30 °C, the reactions were incubated with glutathione-Sepharose beads. Equivalent amounts of unbound fraction (S), the third bead wash (W), and the bound fraction (B) were analyzed by SDS-PAGE followed by Western blot analysis with antibodies to Not5p. The major visible forms of His6-Not5p were all capable of binding Ni-nitrilotriacetic acid agarose, suggesting that they are C-terminal Not5p truncations. Only the largest forms were found in fraction B, specifically in the case of incubation with recombinant GST-yTafII19p, and are labeled with an asterisk. The same results were obtained when binding was performed with 300 mM salt and 0.5% Tween 20 and were reproduced many times.

Bacterial extracts containing His₆-Not5p were incubated with Buffer A or with bacterial extracts containing either GST or GST-yTaf_II19 in the presence of 150 mM salt and 0.1% Tween 20 (see under "Experimental Procedures"), and glutathione-Sepharose beads were then added. The beads were washed, and then the unbound extract (Fig. 6, S), wash (W), and beads (B) were analyzed by Western blot for the presence of Not5p. His₆-Not5p was retained on the beads specifically in the presence of GST-yTaf_II19p. Interestingly, only the three longest forms of His₆-Not5p (all capable of binding Ni-nitrilotriacetic acid agarose) were retained on the beads carrying GST-yTaf_II19p. Similar results were obtained with higher stringency (300 mM salt and 0.5% Tween 20) (data not shown). These results demonstrate that yTaf_II19p and Not5p can directly associate in the absence of other yeast proteins. It appears that the interaction between yTaf_II19p and Not5p requires a minimal N-terminal Not5p fragment, a finding that might relate directly to the observation that the not5-1 allele, but not the not5-2 allele, which encodes a longer protein, displays synthetic growth phenotypes with taf19 mutants in vivo.

yTaf_II19p Is Found in Multiple Complexes in Vivo-- The results presented so far demonstrate allele- and gene-specific genetic interactions between TAF19 and genes encoding components of the Ccr4-Not complex. They further show that yTaf_II19p and Not5p physically interact. These results are in agreement with a model whereby components of the Ccr4-Not complex might be associated with and regulate yTaf_II19p. Our findings also confirm the hypothesis that yTaf_II19p is part of the TFIID complex, but it is not known whether it may be part of any other complexes, such as Ccr4-Not complexes. As mentioned above, with the antibodies that we raised, we could not detect yTaf_II19p in 100 µg of total cell extracts by Western blot analysis. However, we could detect yTaf_II19p very specifically in 45 and 55% ammonium sulfate cuts of total cell extracts (Fig. 7A). In contrast to this very specific fractionation of yTaf_II19p, yTaf_II145p (and other yTaf_IIs) fractionated with a very broad profile, from the 30% ammonium sulfate cut to the supernatant of the 70% ammonium sulfate cut (Fig. 7A). This material enriched for yTaf_II19p was analyzed by Superose 6 gel filtration to determine whether yTaf_II19p was associated in complexes other than TFIID. Fig. 7B shows that the majority of yTaf_II19p is associated in very large complexes. Fractionation by Sepharose 4B allowed us to clearly define that these yTaf_II19p complexes of a size apparently greater than 1 MDa were soluble complexes and not aggregated proteins (data not shown). When equal protein amounts of each fraction were loaded on the gel rather than equal volume equivalents of each fraction, small amounts of yTaf_II19p were also detectable as eluting with a very broad profile, as has been previously shown for yTaf_II145p (33) (Fig. 7C). This shows that yTaf_II19p also appears to elute in fractions 18, 22, and 26. 

View larger version (22K): [in this window] [in a new window]   Fig. 7.   Most yTafII19p has a very restricted distribution to large protein complexes. A, total protein extracts from wild type cells were incubated with 30% ammonium sulfate. The supernatant was brought to 45, 55, and finally 70% ammonium sulfate. The precipitates at each concentration were resusupended and dialyzed against the extract buffer, and 50 µg of each fraction (lanes 30C, 45C, 55C, and 70C), as well as 50 µg of the final supernatant (lane 70S), were analyzed by Western blot for the presence of yTafII19p and yTafII145p as indicated. B, 300 µl of fraction 45C was analyzed by Superose 6 gel filtration. Every second fraction was analyzed by SDS-PAGE followed by Western blot analysis with antibodies against yTafII19p. The position of the void volume and marker proteins of known size are indicated at the top. The position of yTafII19p and a cross-reactive band are indicated. Similar results were obtained with fraction 55C. C, to reveal the possible presence of yTafII19p in fractions other than fractions 1-4, the protein concentration of each fraction was measured, and equivalent amounts were similarly analyzed by Western blot for the presence of yTafII19. The fractions displayed on the figure are the only ones in which yTafII19 was reasonably detected (lane numbers correspond to fraction numbers). Fraction 18 corresponds to a size of 550 kDa, fraction 22 to 350 kDa, and fraction 26 to 228 kDa.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

yTaf_II19p is one of the less well characterized yTaf_IIs so far. It is the homologue of human Taf_II18p, which is known to be part of the human TFIID complex and to form dimers with human Taf_II28p according to a histone-fold type of structure (15). In yeast, TAF19 is essential for vegetative growth, and one report has demonstrated that not all transcription decreases similarly upon yTaf_II19p depletion but that transcription from TATA-less promoters is preferentially arrested (12). The same was reported for the yeast homologue of human Taf_II28p, namely yTaf_II40p (13), but a recent study claims that in fact yTaf_II40p is generally required for RNA polymerase II transcription (14). In the latter, Komarnitsky et al. show that the temperature-sensitive phenotype of the taf40 alleles isolated could be suppressed by overexpression of yTaf_II19p, suggesting that yTaf_II19p should also be generally required for transcription. We have isolated temperature-sensitive taf19 alleles. By analyzing the levels of transcripts of similar half lives (27), we found that not all transcription decreases with similar rapidity upon loss of yTaf_II19p function, supporting the selective requirement for yTaf_II19p function previously published (12). We did find that within 6 h of yTaf_II19p inactivation, all RNA polymerase II transcripts measured decreased. However, this general late effect could clearly be indirect, especially if one considers that yTaf_II19p is needed for the stability of one or more transcription complexes (e.g. TFIID). In fact, this late effect also correlates with the appearance of a drop in viability. All conditional alleles of TAF19 that we isolated carry multiple mutations, similarly to the taf40 temperature-sensitive alleles that were isolated (14). Most of these mutations also lie in the region of TAF19 that encodes the domain of yTaf_II19p thought to interact with yTaf_II40p. Our analysis demonstrated that multiple mutations are necessary to confer a temperature-sensitive phenotype. This might suggest that multiple mutations are necessary to disrupt a yTaf_II19p-yTaf_II40 interaction. Alternatively, temperature sensitivity might require the loss of yTaf_II19p interaction with multiple proteins. Because overexpression of TAF40 suppresses temperature sensitivity of the mutant taf19 alleles, the former seems a more likely explanation.

By the two-hybrid assay and GST pull-down analysis, we have demonstrated that yTaf_II19p can interact with Not5p, both in vitro and in vivo. This interaction is direct and does not require any other yeast protein. Interestingly, one of the taf19 mutants isolated, namely that encoding yTaf_II19-1p, was no longer able to interact with Not5p in the two-hybrid assay, even at 30 °C. At the permissive temperature and on rich media, the strain MLY270 (taf19-1) grew as well as the wild type, suggesting that the mutant protein is functional and has a global unaltered conformation. Hence, the fact that yTaf_II19-1p does not interact with Not5p indicates that one (or some, if not all) of the mutated residues is directly involved in this interaction and may define the yTaf_II19p domain interacting with Not5p. Within Not5p, we did not precisely localize the domain responsible for interaction with yTaf_II19p, but we could show that it was in the N terminus of the protein. In this most N-terminal part of Not5p, the first 207 amino acids are highly conserved in Not3p (40% identity), and we also detected a yTaf_II19p-Not3p interaction in our two-hybrid assay. Despite the fact that we did not try to confirm this interaction by GST pull-down analysis, it is tempting to suggest that this conserved domain might be the yTaf_II19p target. We are currently testing this hypothesis. The yTaf_II19p human homologue is quite well characterized, and complexes containing this Taf_II have been isolated (15). On the other hand, human homologues of the yeast genes encoding Ccr4-Not complex subunits have been isolated (34). Our study thus provides us with a tool to expand the comprehension of Taf_II function in the mammalian system, and it will be very interesting to determine whether the yTaf_II19p-Not5p interaction is also conserved in human.

Characterization of the level of components of the TFIID complex upon loss of yTaf_II19p function at the restrictive temperature demonstrates that, similarly to what has been reported for the depletion of other Taf_IIs, the steady-state level of a number of Taf_IIs specific to TFIID (yTaf_II40p and yTaf_II145p) decreases to undetectable levels very rapidly. The level of TBP also rapidly decreases to a lower level. In fact, we show that loss of yTaf_II19p has effects very similar to those of the depletion of yTaf_II40p, as described previously (14). Taken together with the fact that overexpression of TAF40 suppresses the mutant taf19 alleles that we isolated and with the fact that yTaf_II19p can coimmunoprecipitate with TBP and yTaf_II145p (12), these results support the idea that, like yTaf_II40p, yTaf_II19p is part of TFIID. Further support comes from our finding that yTaf_II40p and yTaf_II145p, so far only described in TFIID and not in SAGA, co-purified with GST-yTaf_II19p (data not shown).

It is not known how many different yTaf_II19p-containing complexes may exist. We have found that most of the yTaf_II19p from total cell extracts fractionates with a size greater than 1 MDa, as determined both by Superose 6 and Sepharose 4B gel filtration. One can also detect some yTaf_II19p in three separate peaks, one of which corresponds to a size that could be TFIID (fraction 18). Although the steady-state level of some yTaf_IIs (yTaf_II60p, yTaf_II68p, and yTaf_II90p) present in TFIID and SAGA remained relatively stable upon yTaf_II19p inactivation, that of yTaf_II25p (also found in both complexes) rapidly decreased. Conversely, yTaf_II19p disappears rapidly upon yTaf_II25p depletion (32). Thus, our results do not exclude the possible presence of yTaf_II19p in SAGA. In this regard, we have found that the taf19 alleles that we isolated are synthetic lethal with the null allele of SPT3 (data not shown). Spt3p contains histone-fold domains homologous to both those of yTaf_II19p and yTaf_II40p, and a model structure for Spt3p in which these two domains interact has been proposed (15). Nevertheless, one cannot exclude that yTaf_II19p interacts with the yTaf_II40-homologous histone-fold domain of Spt3p and thus may be present in the SAGA complex. This could account for the synthetic lethality mentioned above. Alternatively, yTaf_II19p is not in SAGA, and the taf19-spt3 synthetic lethality could be explained by the combined impairment of TFIID and SAGA function (or of yet other Taf_II-containing complexes). Further studies will be needed to elucidate this ambiguity. Whether or not yTaf_II19p is in SAGA as well as in TFIID (and maybe in other complexes), it appears that the mutant phenotype of the alleles that we isolated can be suppressed by increasing yTaf_II40p levels. Thus, the interaction with the Ccr4-Not complex that is suggested by the genetic interactions that we described probably involves complexes that carry both yTaf_II19p and yTaf_II40p. This would argue in favor of TFIID or another complex, but not SAGA.

The question of whether yTaf_II19p is associated in large Ccr4-Not complexes or only interacts transiently with components of the complex will require more work. Our findings clearly demonstrate that yTaf_II19p can associate with Not5p directly, and interacts with at least four other components of the complex by the two hybrid assay. Surprisingly Not4p is not one of these, yet not4 mutants display genetic interactions with the taf19 mutants. However, we know that the absence of Not4p dramatically decreases the association of Not5p in large Ccr4-Not complexes.²

Our present results are consistent with there being a functional interaction between the Ccr4-Not complex and TFIID, as has been previously suggested. Indeed, one can imagine that there is an equilibrium between the different yTaf_II19p and Not5p complexes. What elements regulate this balance, and how many different yTaf_II19p and Not5p complexes there are, are very interesting questions that can now be addressed.

	ACKNOWLEDGEMENTS

We thank Stéphane Jacquier for the purification of recombinant yTaf_II19p and Nicole Paquet for expert technical assistance in all of the biochemical experiments. We also thank Brice Petit and Caroline Raveraud for technical assistance. We thank Joe Reese and Anthony Weil for yTaf_II and TBP antibodies, Matthias Mayer for pMPM272, Ursula Oberholzer for strains and plasmids, and Clyde Denis for B42-Ccr4 and B42-Caf1 expression plasmids. We thank members of our laboratory for fruitful discussions.

	FOOTNOTES

^* This work was supported by Swiss National Science Foundation Grants 31-39690.93 and 31-49808.96 (to M. A. C.) and by Grant OFES96.0072 TMR (to M. A. C.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Present address: Université Claude Bernard, Unité de Microbiologie et Génétique, Génétique des Levures, Bt 405 R2, 43 Boulevard du 11 Novembre 1918, 69622 Villeurbanne Cedex, France.

^§ To whom correspondence should be addressed. Tel.: 41-22-702-55-16; Fax: 41-22-702-55-02; E-mail: martine.collart@medecine.unige.ch.

Published, JBC Papers in Press, June 22, 2000, DOI 10.1074/jbc.M002701200

² M. A. Collart, unpublished observations..

	ABBREVIATIONS

The abbreviations used are: TBP, TATA-binding protein; PCR, polymerase chain reaction; ORF, open reading frame; GST, glutathione S-transferase; PAGE, polyacrylamide gel electrophoresis; TFIID, transcription factor IID.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us