Cloning and characterization of an essential Saccharomyces cerevisiae gene, TAF40, which encodes yTAFII40, an RNA polymerase II-specific TATA-binding protein-associated factor.

In this report we describe the cloning and initial characterization of TAF40, a gene that encodes a yeast TATA-binding protein-associated factor (yTAF) of Mr = approximately 40,000. This gene has many similarities to other yTAFs described thus far in that it is present at a single copy per haploid genome, it is essential for viability, and the deduced protein sequence of yTAF40 exhibits similarity to previously described human and Drosophila TAFIIs. Immunological studies confirm that yTAF40 protein is a subunit of a large multiprotein TATA-binding protein-TAF complex that contains a subset of the total number of the yTAFs present in yeast cell extracts. Transcription reactions performed using yeast whole cell extracts reveal that of the three nuclear RNA polymerases only RNA polymerase II function is abrogated when yTAF40 and associated proteins are immunodepleted from solution, indicating that the functionality of the multiprotein complex containing yTAF40 is RNA polymerase II-specific. By these criteria yTAF40 appears to encode a bona fide RNA polymerase II-specific TAF, and thus the protein that it encodes has been termed yTAFII40.

In this report we describe the cloning and initial characterization of TAF40, a gene that encodes a yeast TATAbinding protein-associated factor (yTAF) of M r ‫؍‬ ϳ40,000. This gene has many similarities to other yTAFs described thus far in that it is present at a single copy per haploid genome, it is essential for viability, and the deduced protein sequence of yTAF40 exhibits similarity to previously described human and Drosophila TAF II s. Immunological studies confirm that yTAF40 protein is a subunit of a large multiprotein TATA-binding protein-TAF complex that contains a subset of the total number of the yTAFs present in yeast cell extracts. Transcription reactions performed using yeast whole cell extracts reveal that of the three nuclear RNA polymerases only RNA polymerase II function is abrogated when yTAF40 and associated proteins are immunodepleted from solution, indicating that the functionality of the multiprotein complex containing yTAF40 is RNA polymerase IIspecific. By these criteria yTAF40 appears to encode a bona fide RNA polymerase II-specific TAF, and thus the protein that it encodes has been termed yTAF II 40.
The TATA-binding protein (TBP) 1 is required by all three nuclear RNA polymerases for transcription (Refs. 1-3; reviewed in Refs. 4 -6). Inside the cell TBP exists complexed with a variety of TBP-associated factors (TAFs) (Refs. 1 and 7-14; see Refs. 15-17 for recent reviews). The collection of TAFs associated with TBP appear to dictate the polymerase specificity of a given TBP-TAF complex. Thus, the present day nomenclature of TAF I s, TAF II s, and TAF III s has arisen to allow demarcation of the RNA polymerase specificity of a given TAF. The exact number of TBP-TAF complexes that exist within the cell and the complement of TAFs that make up each specific complex are currently unclear and are actively being investigated by a number of laboratories.
TFIID is the TBP-TAF complex dedicated to RNA polymer-ase II (RNAP II) transcription. RNAP II is the polymerase responsible for the transcription of mRNA encoding genes. In the three major model systems studied to date (human, Drosophila, and yeast), the TFIID complex has been found to consist of TBP and 9 -12 tightly associated TAF II polypeptides ranging in size from M r ϭ 250,000 to 15,000 (11, 13, 14, 18 -27). Although all studies of TAF II function report an obligatory role for TBP in RNA polymerase II transcription, the exact role that the TBP-TAF II complex TFIID plays in transcription and transcriptional regulation remains to be elucidated (28 -31). Studies of Saccharomyces cerevisiae TBP-TAF complexes by our lab and others have suggested the presence of a TFIID complex that consists minimally of TBP and TAF II s of 150, 130, 90, 60, 40, 30, and 25 kDa (14,25,27). Here we extend these findings by reporting the cloning and characterization of TAF40, the gene encoding the ϳ40,000-Da moiety of the yeast TFIID complex. We find that the gene encoding yTAF40 is both single copy and essential for viability. The polypeptide sequence deduced for yTAF40 exhibits sequence similarity to two known metazoan TAF II s: hTAF II 28 (human) and dTAF II 30␤ (Drosophila). In the studies reported here we demonstrate that yTAF40 is in fact a bone fide TBP-associated factor that is part of the yeast TFIID complex. Further, by utilizing various in vitro transcription assays we show that the function of the yTAF40-containing TBP-TAF complex is RNA polymerase IIspecific, thus the protein encoded by TAF40 has been termed yTAF II 40.
yTAF40 Purification and Protein Sequencing-Yeast TAF proteins were preparatively purified by immunochromatography on columns containing anti-TBP IgG covalently coupled to protein A-Sepharose. Purified yTAFs (13,(25)(26)(27)33) were separated by SDS-PAGE as described, transferred to a nitrocellulose membrane, and visualized by Ponceau staining, and the band corresponding to yTAF40 was excised and cleaved with trypsin. The resulting tryptic peptides were purified by high pressure liquid chromatography and sequenced. The sequences of four of the resulting peptides were determined. These peptides had the following sequences: peptide 1, K/RLLVTNLDKDQTNRFEVFHR; peptide 2, K/RQMIDQVISEDQDYVTXK; peptide 3, K/RETTL-GNSLLQS; and peptide 4, K/RLQSDTLPNAYWR. BLAST searches (36) were used to scan data bases for protein sequence homologies, whereas Geneworks software (Intelligenetics Inc.) was used to generate and format sequence alignments.
TAF40 Gene Cloning-The gene encoding yTAF40 was cloned via a PCR strategy based upon the amino acid sequence of yTAF40 tryptic peptide 2. Two degenerate primers with HindIII (upstream) and XbaI (downstream) restriction endonuclease recognition sites (upstream, GAGAAAGCTTCA(A/G)ATGAT(T/A/C)GA(T/C)CA, HindIII site underlined; downstream, GAGATCTAGAAC(A/G)TA(A/G)TC(T/C)TG(A/ G)TC, XbaI site underlined) were synthesized. In a first round of PCR, the downstream oligo was used together with an oligo (GCAAAAACT-GCATAACCACTTTAACTAATAC) complementary to GAL1 promoter sequence downstream of the insertion site of cDNAs into vector pRS316-Gal1-cDNA (37), using total cDNA library DNA as the template. The products of this first PCR reaction were used as the template along with both degenerate oligonucleotides as primers in a second PCR reaction. The products of this second PCR reaction were subsequently digested with HindIII and XbaI, and the correct length (47 base pairs) product was cloned into HindIII/XbaI-digested pBSIIKS ϩ (Stratagene) for DNA sequence determination. From the determined nondegenerate sequence an oligonucleotide was designed (CAAGTCATCAGTGAGGA) that was then used to screen a yeast cDNA library. The cDNA clone with the longest insert resulting from this library screening was sequenced. A 1.3-kilobase pair fragment of this cDNA clone was then 32 P-labeled and used to screen a yeast genomic DNA library (ATCC 77164) to obtain a full-length plasmid clone of the gene encoding yTAF40. The insert of the resulting plasmid was sequenced in entirety using standard dideoxy sequencing methods.
Construction of a Yeast Strain Harboring a Chromosomal Null Mutation of TAF40 -Diploid strain SEY6210.5 was used as the parental strain for the disruption of TAF40. PCR was used to generate a knockout DNA fragment containing 50 base pairs of TAF40 upstream sequences and 50 base pairs of TAF40 downstream sequences bracketing an intact TRP1 gene. This DNA fragment was gel-purified and used to generate, via homologous recombination (38), a TAF40 null allele by transforming strain SEY6210.5 to tryptophan prototrophy; the resulting Trp ϩ strain was called yEK8. Proper integration of the taf40⌬1::TRP1 disrupting fragment at the TAF40 locus was verified by genomic Southern blotting. One of the cDNA plasmids found to contain the full-length TAF40 cDNA termed pRS316-Gal1-TAF40 cDNA (see above) was then used to transform yEK8 to uracil prototrophy to produce yeast strain yEK8/pRS316-Gal1-TAF40 cDNA. This strain was then sporulated, and the resulting tetrads were dissected onto galactose containing plates. The resulting spores were germinated and subjected to phenotypic testing. One of the resulting spore clones derived from this dissection was termed yEK31. yEK31 has the relevant genotype of taf40⌬1::TRP1 ura3 leu2 pRS316-Gal1-TAF40 cDNA.
Construction of a Yeast Strain Expressing HA 3 -tagged yTAF40 -A plasmid expressing HA 3 -tagged yTAF40 was constructed as follows. The genomic 2646-base pair EcoRV/XbaI TAF40 gene fragment was cloned into pBSIIKS Ϫ (Stratagene). The EcoRI site in the TAF40 promoter region was eliminated by digesting with EcoRI and filling in these ends with E. coli Klenow DNA polymerase I and then religating the ends. Subsequently an EcoRI site was introduced just 5Ј to the TAF40 ORF ATG using site-directed mutagenesis (39) to generate a plasmid termed pBSIIKS Ϫ -EcoRI-TAF40. A PCR fragment encoding three copies of the influenza hemagglutinin epitope (HA) tag (40) with EcoRI ends was inserted into the EcoRI site of pBSIIKS Ϫ -EcoRI-TAF40 to generate pBSIIKS Ϫ -HA 3 -TAF40. An XbaI/HindIII fragment from this plasmid was then cloned into similarly digested pRS415 (Stratagene) to generate pRS415-HA 3 -TAF40. This plasmid was introduced into haploid strain yEK31 via transformation, and this LEU2-marked CEN/ARS plasmid was exchanged for the URA3-marked pRS316-Gal1-TAF40 cDNA plasmid by overnight growth in appropriate selective medium. The resulting strain, termed yEK200, had the taf40⌬1::TRP1 null allele covered by the plasmid pRS415-HA 3 -TAF40.
Baculoviral Expression and Purification of yTAF40 -The TAF40 ORF (codons 32-347) was amplified using Pfu DNA polymerase (Stratagene) in PCR reactions using the following two oligonucleotides as primers: upstream GAGAGAATTCATTGACCAAGTCATCAGTG (EcoRI site underlined) and downstream GAGAGGTACCTATCTGAA-CATGGATCCC (KpnI site underlined). PCR reactions with these primers were templated with cloned genomic TAF40 sequences. The PCR product obtained was digested with EcoRI and KpnI and then cloned into similarly digested baculovirus transfer vector plasmid pAcHLT-A (Pharmingen). The resulting plasmid was cotransfected, along with linearized baculovirus DNA (BaculoGold-Pharmingen) into SF-9 insect cells grown in TMN-FH medium (JRH Scientific). Three rounds of virus amplification were then carried out in SF-9 cells to generate high titer virus stocks. For large scale yTAF40 protein production, virus at an multiplicity of infection of 5 was used to infect High-Five insect cells (Invitrogen) grown in spinner flasks using Ex-Cell 400 medium (JRH Scientific). Infected cells were harvested 64 h postinfection and then washed in ice-cold phosphate-buffered saline (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na 2 HPO 4 , 1.4 mM KH 2 PO 4 , pH 7.4) prior to making a lysate using a denaturing buffer (Buffer A, 6 M guanidine hydrochloride, 0.1 M NaH 2 P0 4, 0.01 M Tris, pH 8.0). His 6 yTAF40 fusion protein was purified under denaturing conditions using nickel nitrilotriacetic acid (Qiagen) resin (1 ml of resin/500 ml of infected cells). After batchwise washing (Buffer B, 8 M urea, 0.1 M NaH 2 P0 4 , 0.01 M Tris, pH 6.3), the resin was loaded into a syringe column, and bound His 6 TAF40 was eluted utilizing a pH jump (Buffer D, 8 M urea, 0.1 M NaH 2 P0 4 , 0.01 M Tris, pH 5.0). The approximate yield of purified protein was Ն1 mg of His 6 yTAF40 from 500 ml of infected High-Five insect cell culture. Protein purity (ϳ85%) was estimated by Coomassie staining of an SDS gel.
Antibodies-Polyclonal antibodies directed against yTAF40 were raised in rabbits by Bethyl Labs, Inc. (Montgomery, TX) using purified baculoviral expressed His 6 yTAF40 as antigen. Total rabbit IgG was purified from the resulting rabbit serum using protein A-Sepharose (Sigma) chromatography (41). Anti-influenza virus hemagglutinin (anti-HA) monoclonal antibody, 12CA5, which reacts with the HA epitope (YPYDVPDA), was purchased from Boehringer Mannheim.
Immunodepletion Experiments-Monoclonal anti-HA antibody 12CA5 cross-linked to protein A-Sepharose was used to deplete HA 3 -yTAF40, and associated proteins from 100-l aliquots of transcriptionally competent WCE were prepared from yeast strain yEK200 by the method of Woontner et al. (42). The procedures followed for the preparation of resins, and the immunodepletion reactions were as previously described (25). The supernatants from these immunodepletion reactions were frozen in 50-l aliquots at Ϫ70°C until assayed for both yTAF40 content and residual RNAP I-, II-, and III-specific transcriptional activity.
Immunoprecipitation and Immunoblotting-Immunoprecipitation and immunoblotting was performed as described previously (25). Where indicated, blots were subjected to quantitation by exposing the developed blot to a Bio-Rad high intensity imaging screen, which was then analyzed using the Bio-Rad GS-250 Molecular Imager and Molecular Analyst Software.
In Vitro Transcription Assays-RNA polymerase I-, II-, and IIIspecific transcription assays were carried out as described previously (Refs. 25, 43, and 44, respectively). Specific transcript production was quantitated using a Bio-Rad Imager as described above.

RESULTS AND DISCUSSION
Cloning of the Yeast Gene Encoding yTAF40 -Using conventional and anti-TBP immunoaffinity chromatography we have been able to purify TAFs from yeast whole cell extracts (13,(25)(26)(27)33). A yTAF of M r 40,000 is readily visualized on silverstained SDS gels from these preparations (e.g. see Fig. 1 in Ref. 25). Peptide sequence was obtained from this 40-kDa TAF as detailed under "Materials and Methods." At the time when this sequence was obtained GenBank and Saccharomyces Genome data base searches using the BLAST search algorithms (36) revealed that these sequences were novel.
In order to clone the gene encoding yTAF40, the amino acid sequence of one of the tryptic peptides derived from yTAF40 (peptide 2, (K/R)QMIDQVISEDQDYVTXK) was used to design degenerate oligonucleotide primer pairs for use in the polymerase chain reaction for amplification of the DNA encoding this peptide. Total yeast cDNA library DNA (37) was used as the template for this PCR reaction. As detailed under "Materials and Methods," two separate rounds of PCR amplification had to be performed to obtain significant quantities of the correct length product. This requirement for two separate rounds of PCR was probably due to the high degree of degeneracy of the oligonucleotides used as primers. The correct length PCR product was cloned, and the nucleotide sequence of the insert was determined. An oligonucleotide derived from this sequence was then synthesized, labeled, and used as a probe to screen a yeast cDNA library (37) to obtain a full-length cDNA encoding yTAF40. Sequence information derived from positive clones from this screen allowed the preparation of a longer labeled probe, which was then used to screen a yeast genomic library. Sequencing of a positive genomic plasmid clone revealed an ORF of 1038 nucleotides (GenBank accession number U85960). As shown in Fig. 1 this sequence contained all four of the peptides obtained from the microsequencing of the band corresponding to the M r 40,000 TAF polypeptide, suggesting that in fact we had cloned the gene encoding yTAF40. To further substantiate this fact, polyclonal antiserum raised in rabbits immunized with purified recombinant baculovirally ex-pressed yTAF40 protein was able to specifically recognize the 40-kDa protein in our TAF fraction (data not shown and see Fig. 3 below). This cloned gene has been termed TAF40 for TBP-associated factor 40,000. The predicted molecular weight of yTAF40 is 40620, which is similar to its apparent M r observed in our SDS-PAGE analyses. The protein is mildly acidic with an estimated pI of 5. 35.
Sequence Similarity of yTAF40 to Other Metazoan TAFs-Upon searching more recent editions of sequence data bases with the yTAF40-deduced protein sequence several matches to yeast and other metazoan proteins were noted. The yTAF40 sequence matched (p ϭ 8.1e Ϫ200 ) to Protein Information Resource entry S55104, a previously uncharacterized yeast ORF. Interestingly yTAF40 also matched to PIR entries S54780 (hTAF II 28; p ϭ 8.4e Ϫ13 , 27% identity 67% similarity over yTAF40 residues 122-181) and B49453 (dTAF II 30␤; p ϭ 9.3e Ϫ11 , 30% identity and 68% similarity over yTAF40 residues 122-181). Several other weaker matches were also noted, the most notable being a match of yTAF40 residues 236 -305 to human TFIIA ␣ chain polypeptide sequences 280 -349. Presented in Fig. 2 is an alignment of yTAF40 to both Drosophila TAF II 30␤ (23) and human TAF II 28 (19). Although overall identity between all three of these TAFs is relatively low (only 9% of yTAF40 sequence is identical to these other TAFs), in an 81-amino acid region between yTAF40 residues 111-192 there is 23% identity and 57% similarity among all three TAFs. It is important to note though that yTAF40 is significantly longer than either metazoan TAF II , thus overall identity must be somewhat low. The finding of similarity between yet another yTAF and metazoan TFIID subunits at the amino acid sequence level is not surprising. In fact all of the yTAFs identified thus far display significant sequence similarity to metazoan TAFs. However, it remains to be determined if these similarities reflect true functional homology. Regardless, the finding that yTAF40 has sequence similarity to TAF II s of other species is strong preliminary evidence that yTAF40 is in fact an RNA polymerase II-specific TAF.
TAF40 Is a Single Copy Essential Gene-Prior to performing gene disruption experiments, we examined TAF40 genomic DNA and RNA species to rule out the potential for multiple TAF40 encoding genes. Genomic DNA blots (not shown) indicated that TAF40 is present at a single copy per haploid genome, whereas RNA blots showed that only a single 1.1-kilobase RNA species anneals with TAF40-derived probes (not shown). The size of this mRNA is consistent with the size of the TAF40 ORF. When yeast strain yEK8 was sporulated and the resulting tetrads were dissected, viability segregated 2 ϩ :2 Ϫ , all viable spores were Trp Ϫ , suggesting that TAF40 is a single copy essential gene. When yEK8 was subsequently transformed to uracil prototrophy with the URA3-marked plasmid pRS316-Gal1-TAF40 cDNA and then sporulated upon tetrad dissection, all four spores were viable on galactose containing medium, all Trp ϩ spores were Ura ϩ , and the Trp ϩ phenotype segregated 2 ϩ :2 Ϫ as expected. These Trp ϩ Ura ϩ spore clones were also unable to grow on 5-fluoroorotic acid containing medium (45) indicating a requirement for the TAF40 covering plasmid in the spore clones carrying the taf40⌬::TRP1 null allele. These genetic and biochemical analyses prove that TAF40 is an essential gene. With the exception of yeast TAF30 (ANC1/TFG3) (26), all of the other characterized yeast TAF-encoding genes (TAF170, TAF150, TAF130, TAF90, TAF60, and TAF25) are essential (25,27,33). That all of these TAFs are essential underscores the vital role that they must serve in yeast cell function.
yTAF40 Is a Bona Fide yTAF That Is Present in a Multiprotein Complex with TFIID-specific TAFs-To show that yTAF40 is a bona fide TAF, we utilized the technique of coimmunoprecipitation. If yTAF40 is truly TBP-associated, then it should be detectable in an immunoprecipitate of a WCE treated with antibodies directed against TBP. The converse should also be true in that TBP should be detectable in an anti-yTAF40 im-FIG. 3. yTAF40 coimmunoprecipitates with TBP and TFIIDspecific TAF II s. Top panel, both HA 3 TBP and HA 3 -TAF moieties are specifically immunoprecipitable from WCEs prepared from yeast strains expressing the relevant HA 3 -tagged proteins. WCEs were prepared from yeast strains expressing the indicated HA-tagged versions of TBP or TAF. HA 3 -tagged proteins present in the WCEs were immunoprecipitated using anti-HA mAb 12CA5, and the immunoprecipitates were fractionated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane. Proteins containing the HA 3 epitope were detected by blotting with anti-HA mAb 12CA5. The arrows and labels indicate the various epitope-tagged proteins. Bottom panel, yTAF40 is only detectable in anti-HA mAb immunoprecipitates derived from strains expressing HA-tagged proteins that are TFIID-specific TAF II s. Immunoprecipitates were prepared using anti-HA antibody from the WCEs used in the panel above. The immunoprecipitates were fractionated via SDS-PAGE and transferred to a membrane, but in this case the blots were probed with polyclonal anti-yTAF40 IgG. The arrows and associated labels indicate either native yTAF40 or the more slowly migrating HA 3 -TAF40 polypeptides.
FIG. 2. Alignment of the deduced amino acid sequences of yTAF40, dTAF II 30␤, and hTAF II 28. Areas of sequence identity are contained within boxes. Gaps introduced to maximize alignment are indicated by dashes. Identical and similar amino acids are indicated by the shading. Although overall identity between these TAFs is only 9%, in the 81-amino acid region between yTAF40 residues 111 and 192, there is 23% identity and 57% similarity between these three sequences. munoprecipitate. For these coimmunoprecipitation studies we utilized our collection of yeast strains expressing HA 3 -tagged TAFs and/or HA 3 -tagged TBP as well as polyclonal rabbit anti-yTAF40 antibodies (see "Materials and Methods").
In the experiment depicted in Fig. 3 (upper panel) immunoprecipitates were formed using the anti-epitope antibody (anti-HA) and WCEs prepared from each of our immunotagged yeast strains along with an untagged control. An immunoblot of these immunoprecipitates with the anti-HA mAb shows that only the appropriate epitope-tagged TAF or epitope-tagged TBP are readily and specifically detected. Thus for each strain tested the only reacting species detected in the anti-HA immunoblot is the corresponding single, correct sized epitope-tagged polypeptide that the strain expresses. The background resulting from immunoblotting the immunoprecipitate from the same yeast strain, which lacks any HA 3 -tagged species, is shown in lane 1 of Fig. 3.
In Fig. 3 (lower panel) are presented the results of the portion of this immunoprecipitation experiment designed to examine whether yTAF40 is indeed both TBP-and yTAF II -associated. In this experiment immunoprecipitates were again formed using anti-HA mAb, but in this case, the SDS-PAGE fractionated and blotted immunoprecipitates were probed with rabbit polyclonal anti-yTAF40 IgG. As can be seen from the results presented in Fig. 3 (lower panel, lane 7), yTAF40 is indeed associated with TBP because yTAF40 is readily identified in the anti-HA immunoprecipitate from the strain that expresses HA 3 -TBP as its only form of TBP. The converse experiment utilizing anti-TAF40 IgG for the immunoprecipitation step and anti-HA mAb for the immunodetection step (data not shown) revealed that HA 3 -TBP is readily detected in the yTAF40 immunoprecipitate, indicating once again that TBP and yTAF40 are associated. Taken together, these results clearly demonstrate that TBP and yTAF40 are complexed, thus confirming that TAF40 encodes a bona fide yeast TAF. Again the control where the immunoprecipitate is derived from a strain lacking any HA 3 -tagged species demonstrates the specificity of the precipitation and detection (Fig. 3, lower panel,  compare lane 1 with lane 7).
To determine which of the other TAFs are associated with yTAF40, we extended the above analysis. Our previous bio- FIG. 4. Immunodepletion of transcriptionally competent WCEs demonstrates that a yTAF40-containing complex is uniquely required for RNA polymerase II transcription. A, immunodepletion efficiently removes yTAF40 from a yeast whole cell extract. A yeast WCE expressing HA 3 TAF40 as the only form of yTAF40 was depleted of yTAF40, and associated proteins using anti-HA mAb 12CA5 cross-linked to protein A-Sepharose as detailed under "Materials and Methods." Resin-bound mAb was either added directly to the WCE (mAb), or the resin-bound mAb was preincubated with either buffer alone (mAbϩBUFFER), or a 100-fold molar excess of specific peptide recognized by the antibody (mAbϩHA PEPTIDE) prior to the addition of WCE to the mAb resin. To determine the extent of immunodepletion, 2 l of the supernatants resulting after removal of immune complexes by centrifugation were compared with 2 l of either freshly thawed WCE (WCE) or WCE that had been incubated overnight, the length of time required for mAb depletion (WCE O/N). These samples were fractionated by SDS-PAGE, transferred to polyvinylidene difluoride, and immunoblotted using monoclonal anti-HA mAb 12CA5 IgG for detection. The signal on the immunoblot representing yTAF40 is indicated by the label and arrow. The amount of yTAF40 in each sample relative to the amount found in the freshly thawed untreated WCE after subtracting for background was determined using a Bio-Rad Imager and are as follows: WCE, 1.00; mAbϩHA PEPTIDE, 1.02; mAbϩBUFFER, 0.25; mAb, 0.25; WCE O/N, 0.98. B, RNA polymerase I-specific transcription is unaffected by depletion of yTAF40 containing complexes. A 4-l aliquot of the variably treated WCEs described for A were tested in an in vitro RNA polymerase I-specific transcription assay (25) to assess the effects of yTAF40 depletion on specific transcription. Results of the assay were scored by primer extension analysis. Following the transcription assay a constant amount of a 300-nucleotide 32 P-labeled DNA fragment (ϳ1000 cpm) was included in all the subsequent steps of the analysis as a recovery control. The autoradiograms presented depict the 50-nucleotide extension product expected for the rRNA primary transcript (labeled rRNA Primary Transcript). Also shown is the 300-nucleotide recovery control (Recovery Control) from a representative experiment. After subtracting for background and normalizing to the recovery control the GS-250 Imager units detected for each extract were as follows: untreated extract, 637 units; depleted extract, 433 units; and mock-depleted extract, 440 units. C, RNA polymerase II transcription is severely compromised as a result of depletion of yTAF40 and associated proteins from WCEs. In vitro transcription reactions were performed as described under "Materials and Methods." Two different G-minus cassette-containing transcription templates were included in each of these reactions, one generating a 400-nucleotide transcript, the other generating a 300-nucleotide transcript (25,49). Similar results were obtained from both templates, and only transcripts from one template are shown. In the autoradiogram presented the arrow indicates the 400-nucleotide transcript resulting from transcription of the longer G-minus template in response to increasing amounts (1, 3,  chemical studies suggested that yTAF40 is a subunit of a complex consisting minimally of yTAFs 150, 130, 90, 60, 30, and 25 and TBP (25,27). As can be seen in Fig. 3 (lower panel,  lanes 2-8 versus lane 1), yTAF40 is specifically detected in the immunoprecipitates formed using anti-HA mAb and WCEs derived from yeast strains expressing HA 3 -tagged TAFs 150, 130, 90, 60, 40, and 25 and TBP. Notably though, yTAF40 is not detected in immunoprecipitates formed from yeast strains expressing HA 3 -tagged Brf1p (yTAF III 70) (lane 9) or Mot1p (yTAF170) (lane 10). Both of these TAFs are components of distinct non-TFIID, TBP-TAF containing complexes (33, 46 -48). If the converse experiment is performed, that is immunoprecipitating yTAF40 (and associated polypeptides) from WCEs with polyclonal anti-yTAF40 antibodies and using anti-HA mAb for HA 3 -TAF and/or HA 3 -TBP immunodetection, the predicted and complementary result is obtained. In this case yTAF40 is found associated with yTAFs 150, 130, 90, 60, and 25 and TBP but not with either Mot1p or Brf1p (data not shown). That yTAF40 coprecipitates with each of the other known TAF II s individually confirms our previous supposition that yTAF40 is in fact part of the yeast TFIID complex that contains all of these TAF II s (13,25,27,33).
yTAF40 Containing Complexes Are Specific to RNA Polymerase II Function-Having demonstrated that yTAF40 is a bona fide yTAF that is associated with previously characterized yTAF II s, our next goal was to test if yTAF40 truly is a TFIID subunit. If this is true then yTAF40 functional activity should be RNA polymerase II-specific. The strategy that we took to test this idea again utilized coimmunoprecipitation. We immunodepleted a yeast WCE of yTAF40 and associated proteins and then used the depleted extracts for in vitro transcription assays to measure the ability of control, mock-depleted, and depleted extracts to support specifically initiated transcription by each of the three distinct nuclear DNA-dependent RNA polymerase systems. We reasoned that if yTAF40 was uniquely a subunit of TFIID, then only RNAP II-specific transcription should be altered in yTAF40-depleted WCEs.
Treatment of yeast WCE with the anti-HA mAb 12CA5 removed ϳ75% of the HA-tagged yTAF40 from the extract (Fig.  4A, compare lanes 3 and 4 with lanes 1, 2, and 5). This depletion was specific because when the anti-HA mAb was preincubated with the HA peptide prior to mixing with WCE, this peptide-blocked mAb preparation failed to remove yTAF40 from the WCE (Fig. 4A, compare lane 2 with lanes 3 and 4 versus lane 2 compared with lanes 1 and 5). The ability to block the depletion was specific to the peptide because preincubation of the mAb with peptide buffer alone generated mAb that was fully able to deplete as effectively as untreated mAb alone (compare lanes 3 and 4, Fig. 4A). As a further control there was as much yTAF40 in WCE that underwent overnight incubation as there was in freshly thawed WCE (compare lanes 1 and 5 in Fig. 4A).
When the depleted extracts were tested for their ability to support transcription by RNA polymerase I (Fig. 4B) or RNA polymerase III (Fig. 4D), only minimal effects of yTAF40 depletion were observed. The depleted extract was as competent for specific RNAP I and RNAP III transcription as was the mock-depleted extract (i.e. HA-peptide blocked extract; Fig. 4B, compare lane 1 with lanes 2 and 3; Fig. 4D compare lanes 1-3  with lanes 4 -6). In marked contrast, dramatic effects of depletion were observed when the depleted extract was tested for RNA polymerase II transcription. Basal transcription resulting from transcription from the G-minus cassette transcription template for the depleted extract was reduced to ϳ25% of that observed for the mock-depleted extract. These data clearly demonstrate that whether we deplete a yeast WCE of its TFIID-like TAF-TBP complex by removing yTAF40 (this report) or, as we previously demonstrated for yTAF II 25 (25), the result as expected is the same, only RNA polymerase II function is abrogated. Furthermore, because yTAF40 is associated with yTAF25 (see Fig. 3, lane 8), these data further corroborate the idea that the yTAF II 25/yTAF II 40-containing TBP-TAF complex represents the major (if not the only) form of TFIID in yeasts. If this were not the case, then one would not expect such a large proportional decrease in basal RNA polymerase IIspecific transcription upon removal of the complex by either yTAF II 25 or yTAF II 40 depletion.
In this report we have described the identification, cloning, and characterization of a gene termed TAF40 that encodes a novel yeast TBP-associated factor, yTAF II 40. The deduced amino acid sequence of yTAF II 40 bears significant similarity to that of both human and Drosophila TAF II s. The single copy TAF40 gene is essential for vegetative growth, a property this gene shares with other yeast TAF-encoding genes. We show that yTAF40 protein is TBP-associated and thus a bone fide TAF and that it is a subunit of a yeast TFIID multiprotein TBP-TAF II complex. Through this work we have added to the body of evidence supporting the existence of a discrete yeast TFIID complex by directly demonstrating that the yTAF40 containing TBP-TAF complex is specific to RNA polymerase II function. Further studies are needed to determine the exact composition and function of the TFIID complex.