A Transcriptional Mediator Protein That Is Required for Activation of

MOLECULAR AND CELLULAR BIOLOGY, Aug. 1997, p. 4622–4632 0270-7306/97/$04.0010 Copyright © 1997, American Society for Microbiology

Vol. 17, No. 8

A Transcriptional Mediator Protein That Is Required for Activation of Many RNA Polymerase II Promoters and Is Conserved from Yeast to Humans YOUNG CHUL LEE, SOYOUNG MIN, BYUNG SOO GIM,

AND

YOUNG-JOON KIM*

Center for Basic Research, Samsung Biomedical Research Institute, Seoul 135-230, Korea Received 20 March 1997/Returned for modification 29 April 1997/Accepted 12 May 1997

A temperature-sensitive mutation was obtained in Med6p, a component of the mediator complex from the yeast Saccharomyces cerevisiae. The mediator complex has been shown to enable transcriptional activation in vitro. This mutation in Med6p abolished activation of transcription from four of five inducible promoters tested in vivo. There was no effect, however, on uninduced transcription, transcription of constitutively expressed genes, or transcription by RNA polymerases I and III. Mediator-RNA polymerase II complex isolated from the mutant yeast strain was temperature sensitive for transcriptional activation in a reconstituted in vitro system due to a defect in initiation complex formation. A database search revealed the existence of MED6-related genes in humans and Caenorhabditis elegans, suggesting that the role of mediator in transcriptional activation is conserved throughout the evolution. complex that is physically associated with RNA polymerase II (26). The mediator complex comprises some 20 polypeptides (including Srb2p, Srb4p, Srb5p, Srb6p, Gal11p, Rgr1p, and Sin4p), each of which has been identified as a transcriptional regulator in separate genetic studies (26, 28, 31, 33, 50). The mediator complex not only enables activated transcription, but also stimulates basal transcription and phosphorylation efficiency of the carboxy-terminal domain (CTD) of the RNA polymerase II largest subunit (encoded by RPB1) (26). Association of the mediator, TFIIF, and core-RNA polymerase results in the formation of RNA polymerase II holoenzyme (holo-polymerase), which is sufficient and necessary for in vitro transcriptional activation by RNA polymerase II (26). The association of additional transcription proteins, including TBP, TFIIB, TFIIH, several Srb proteins, and the SWI-SNF complex (14, 28, 56), is controversial. Genetic analysis of the mediator subunits revealed that mutations in SRB genes (40) reduce the level of transcription by RNA polymerase II (50, 51), while RGR1, SIN4, and GAL11 are required for regulated transcription of Gal4-regulated genes (21, 44, 49). The functional differences of mediator components as revealed by genetic and biochemical analyses of RGR1, GAL11, SIN4, and SRB genes suggest that a subset of the mediator proteins may be responsible for each of the mediator activities. Here, we report that a novel mediator gene, MED6 (previously designated MTR32 [31]), is absolutely required for transcriptional activation of RNA polymerase IItranscribed yeast genes involved in the carbon source utilization and the mating-type specification pathways. Mutations in MED6 inhibited activated transcriptions but had no direct effect on uninduced or repressed transcription. The identification of MED6-related genes in humans and Caenorhabditis elegans suggests that the role of mediator in transcriptional activation is conserved throughout evolution. From these observations, we propose that Med6 protein (Med6p) plays an essential role in mediating between activator signals and basal transcription machinery.

Regulation of mRNA synthesis requires intermediary proteins that transduce regulatory signals from upstream transcriptional activator proteins to basal transcription machinery at the core promoter. A reconstituted in vitro transcription system composed only of basal transcription machinery (coreRNA polymerase II and pure general transcription factors) does not respond to upstream transcriptional activator proteins even though it is fully functional for basal transcription (26, 36). Three types of intermediary factors that enable the basal transcription machinery to respond to transcriptional activator proteins bound to regulatory DNA sequences have been identified: (i) TAFIIs, which associate with TATA-binding protein (TBP) to form TFIID (12, 13; for a review, see reference 53); (ii) mediator, which associates with RNA polymerase II to form a holo-polymerase (11, 25, 26, 28; for a review, see reference 4); and (iii) coactivators such as human upstream stimulatory activity (USA) (36; for a review, see reference 23), mammalian CBP/P300 (for a review, see reference 19), yeast ADA complex (17), and HMG proteins (42, 46). The interaction of these multiprotein complexes with activators and general transcription factors is essential for transcriptional regulation (3, 12–14, 18, 19, 43, 48, 57). In this paper, we focus on the functional analysis of one intermediary factor from Saccharomyces cerevisiae called mediator. The existence of mediator was originally suggested from a squelching assay, in which the addition of one activator interferes with stimulation of transcription by another activator, suggesting that the two activators compete for a common target (11, 25). Squelching is specific for the activation domains of transcriptional activator proteins and could be relieved by addition of a partially purified yeast protein fraction but not by addition of known components of the basal transcription machinery. These results identified a novel activity termed mediator that is distinct from proteins required for basal transcription and that is essential for transcriptional activation. Fractionation of the yeast mediator activity yielded a mediator * Corresponding author. Mailing address: Center for Basic Research, Samsung Biomedical Research Institute, 50 Ilwon-dong, Kangnam-ku, Seoul 135-230, Korea. Phone: 82-2-3410-3638. Fax: 82-23410-3649. E-mail: [email protected].

MATERIALS AND METHODS Construction of the med6 ts mutant allele. To construct a host strain for plasmid shuffling, a URA3-based single-copy plasmid (pRS316 [47]) containing

4622

VOL. 17, 1997

MED6, A MEDIATOR OF TRANSCRIPTIONAL ACTIVATION

the wild-type MED6 gene was introduced into YPH500 (MATa ade2-101 ura3-52 lys2-801 trp1-D63 his3-D200 leu2-D1) to make the YCL3 strain. The chromosomal MED6 copy of the YCL3 strain was disrupted by replacing its open reading frame (ORF) with the LEU2 gene, resulting in strain YCL4 [MATa ade2-101 ura3-52 lys2-801 trp1-D63 his3-D200 leu2-D1 med6::LEU2 (MED6 on pRS316, URA3)]. DNA fragments containing random mutations in the MED6 gene were generated by an error-prone PCR protocol (30) with some modifications. Mutations in MED6 gene fragments containing approximately 250 bp of flanking sequences were generated by Mn21-induced infidelity of Taq polymerase during 30 rounds of PCR (94°C for 30 s, 55°C for 30 s, 72°C for 60 s). The PCR mixtures contained 100 pmol of each oligonucleotide (MED6-1 [59-GGGGGAATTCTTGTGGCT AATCCGGGAAGG-39] and MED6-5 [59-GGGGAGATCTCTTCGTCGTCC TGATCAATCAAC-39]), 20 ng of plasmid pBS-MED6 (32) as a template, 0.1 or 0.2 mM MnCl2, 2.5 mM MgCl2, 0.25 mM each deoxynucleotide triphosphate, 10 mM Tris-HCl (pH 9.0), 50 mM KCl, 0.1% Triton X-100, and 5 U of Taq polymerase in a volume of 100 ml. In order to make a med6 mutant library, we used a single-step method based on in vivo gap repair (38). The med6 mutant PCR products were cotransformed into strain YCL4 with a linearized HIS3based single-copy plasmid (pRS313) containing approximately 250 bp of MED6flanking regions at its termini. A total of 40,000 His1 transformants were isolated, and more than 80% resulted from homologous recombination between the gapped plasmids and mutagenized PCR products. Of these 40,000 transformants, only 10% were viable after removal of the wild-type MED6 gene by 5-fluoroorotic acid (5-FOA) selection. The 5-FOA-resistant colonies were grown on yeast extract-peptone-dextrose (YPD) medium, and mutants showing temperaturesensitive (ts) lethality were isolated. RNA preparation and analysis. Cells were grown in an appropriate medium to early exponential phase at 30°C (A600 5 0.3 to 0.4), divided into two aliquots, and then allowed to grow for another 2.5 h at either 30 or 37°C. Total RNA was prepared as described previously (7) and quantitated by absorbency at 260 nm, and the integrity was confirmed by ethidium bromide staining in an agarose gel. For S1 nuclease protection analysis, 30 mg of total RNA was hybridized to completion with a 100 M excess of the 32P-labeled oligonucleotide probes and digested with S1 nuclease as described by Cormack and Struhl (8). The products were resolved on a 10% denaturing polyacrylamide gel and quantitated with the use of a PhosphorImager and ImageQuant software (Molecular Dynamics). The sequences and references for the oligonucleotides used for S1 analysis are as follows: DED1, 59-GCTAAAGAAG CTGCCACCGC CACGGCCACC GTTG TAGCCG C-39 (6); HIS3, 59-GGTTTCATTT GTAATACGCT TTACTAG GGC TTTCTGCTCT GTCATCTTTG CCTTCGTTTA TCTTGCCTGC TCA TTT-39 (6); TRP3, 59-GGTAAAGGAA TCGTAGTTGT CAATTAAGAA CCACATGCTT ACCTTAG-39 (8); tRNA, 59-GGAATTTCCA AGATTTA ATT GGAGTCGAAA GCTCGCCTTA-39 (8); rRNA, 59-TGCCAGTACC CACTTAGAAA GAAATAAAAA ACAAATCGTT AGG-39 (8); MEDb, 59-G GATTTGTCT TCTTGCCGGA TCCACATAGG CAAAC-39; GAL1, 59-CAC CAATTAG ACTCTACCAG GCGATCTAGC AACAAAATCC GG-39; and SUC2, 59-GCCATTTGGC ATCTTTTTCA TCGTACCACA ACCCATTTGG GTCATTCATC-39. For Northern (RNA) analysis, poly(A)1 RNA (1 mg) from each sample was separated on a 0.8% agarose-formaldehyde gel and transferred to a Nytran membrane. The filter was probed with 32P-labeled antisense RNA as described by Sambrook et al. (45). To prepare probes for Northern analysis, coding regions of the yeast genes PYK1, MFa1, MATa1, MCM1, BAR1, and MATa2 were amplified by PCR and cloned into the multicloning sites of pBluescript II-SK1 (Stratagene). The cloned plasmids were then digested with appropriate restriction enzymes and used as templates for in vitro transcription with T7 or T3 RNA polymerase (Promega). Specifically hybridized signals were quantitated with the use of a PhosphorImager and the associated software. When necessary, the filter was stripped in distilled water for 10 min at 90°C and rehybridized with different probes. The descriptions of the PCR fragments of each gene used for probe preparation are as follows. (i) PYK1. A 1.4-kb DNA containing an ORF region was amplified by PCR with primers PYK1-1 (from position 11, the translation initiation site, to position 121, containing an XbaI site) and PYK1-2 (11393 to 11412, containing an EcoRI site). (ii) MFa1. A 508-bp DNA was amplified by PCR with primers MFa1-1 (13 to 125 tagged with an XbaI site) and MFa1-2 (1478 to 1498 with an EcoRI site). (iii) MATa1. A 537-bp DNA was amplified with primers MATa1-1 (17 to 127 tagged with an XbaI site) and MATa1-2 (1507 to 1528 with an EcoRI site). (iv) MATa2. A 624-bp DNA was amplified with primers MATa2-1 (121 to 143 tagged with an XbaI site) and MATa2-2 (1611 to 1633 with an EcoRI site). (v) MCM1. An 870-bp DNA fragment was amplified with primers MCM1-1 (16 to 128 with an XbaI site) and MCM1-2 (1840 to 1861 with a PstI site). (vi) BAR1. A 1.7-kb DNA was amplified with primers BAR1-1 (151 to 172) and BAR1-2 (11743 to 11764). Protein purification. Mediator-RNA polymerase II complex was purified from 800 g of MED6 wild-type cells (YCL10) [MATa ade2-101 ura3-52 lys2-801 trp1D63 his3-D200 leu2-D1 med6::LEU2 (MED6 on pRS313, HIS3)] and from the same amount of med6 ts mutant cells (YCL8) [MATa ade2-101 ura3-52 lys2-801 trp1-D63 his3-D200 leu2-D1 med6::LEU2 (med6 ts mutation on pRS313, HIS3)] through four chromatographic steps, including use of Bio-Rex70 (Bio-Rad), DEAE-Sepharose FF (Pharmacia), Bio-Gel-HTP hydroxyapatite (Bio-Rad), and MonoQ HR 10/10 (Pharmacia) as described by Kim et al. (26). Holo-polymerase

4623

was eluted with 1 M potassium acetate from the MonoQ HR 10/10 column, and an aliquot (1 ml) of the peak fractions (0.5 to 0.7 mg/ml) was used in each transcription assay. In order to make a Med6p bacterial expression construct, the MED6 ORF region containing a six-histidine tag at its 59 end was amplified by PCR with primers MED6-2 (59-GGGGGCCATG GCCCATCACC ATCACCATCA CAA CGTGACACCGTTGGATG AATTGC-39, with the NcoI site underlined) and MED6-4 (59-GGGGAGATCT TCTCATATGT AATTTGGGG-39, with the BglII site underlined) and pBS-MED6 used as a template. A 900-bp PCR product was digested with NcoI and BglII and cloned into the NcoI/BamHI sites of pET11d (Novagen) to create pET-MED6. The final clone for the bacterial expression construct was confirmed by DNA sequencing. Recombinant Med6p (r-Med6p) and recombinant Srb5p (r-Srb5p) were purified from bacterial strain BL21(DE3) (Novagen) carrying plasmids pET-MED6 and pET-Srb5, respectively, according to the protocol described by Thompson et al. (50). In vitro transcription and CTD phosphorylation. Nuclear extracts from yeast strains YCL8 and YCL10 were prepared as described by Lue and Kornberg (34). In vitro transcription using the nuclear extracts was performed as described by Wang et al. (55). Nuclear extracts containing 50 to 60 mg of proteins were preincubated for 10 min at the indicated temperatures. A supplemental mixture containing ribonucleotides, 150 ng each of pJJ470 and pS(GCN4)2CG2 (26), and 5 to 10 mCi of [a-32P]UTP with or without Gal4VP16 (30 ng) was then added to initiate transcription at 25°C. The reaction proceeded for 50 min. The reaction products were purified and analyzed by 7% denaturing polyacrylamide gel electrophoresis and quantitated with the use of a PhophorImager. Reconstituted in vitro transcription and CTD phosphorylation with holopolymerase were performed as described elsewhere (26). For a specific transcription assay, either wild-type or med6 ts mutant holo-polymerase (700 ng each; MonoQ fraction) was preincubated with other supplements (containing the two templates, general transcription factors, and 0.5 mM ATP, with or without Gal4VP16 activator) for 10 min at 25 or 37°C for formation of the initiation complex. [a-32P]UTP (10 mCi) and CTP were then added, and incubation was continued at 25°C for 30 min. To examine the defects after initiation complex formation, heat treatment (10 min, 37°C) was administered after an initiation period (10 min, 25°C), and the transcription reaction (30 min, 25°C) was initiated by the addition of the remaining nucleotides. All transcription reactions were performed with 180 mM potassium acetate. For the CTD phosphorylation assay, reaction mixtures containing TFIIH (40 ng), 0.3 mCi of [g-32P]ATP, and other supplements were added to core-polymerase (100 ng) or holo-polymerase (300 ng) with or without heat treatment, and the mixtures were incubated for 30 min at 25°C in a 15-ml reaction volume. Reactions were stopped by the addition of 5 ml of 43 sodium dodecyl sulfate (SDS) gel loading buffer, and the reaction products were analyzed on an SDS–7.5% polyacrylamide gel. Transcripts and 32 P-labeled Rpb1 were quantitated with a PhosphorImager. Immunoprecipitation. Affinity-purified anti-Srb5 antibody (100 mg) was conjugated with protein A-agarose (Sigma) beads (100 ml). The Srb5 antibody beads (30 ml) were incubated for 12 h at 4°C with 50 ml of holo-polymerase fraction (MonoQ column) and washed three times with 1 ml of IP buffer (20 mM Tris-HCl [pH 8.0]), 0.1 mM EDTA, 0.2% Nonidet P-40, 10% [vol/vol] glycerol) containing 0.8 M potassium acetate. The bound proteins were eluted twice with 30 ml of IP buffer containing 5 M urea. The eluted proteins were visualized on an SDS–10% polyacrylamide gel by silver staining. MED6 homolog cloning. The human and C. elegans expression sequence tag (EST) databases were screened with BLAST (1) for expression sequence tags similar to yMed6p. EST clones encoding human (231995) and C. elegans (yk60911) MED6 homologs were obtained from Research Genetics and National Institute of Genetics, Japan, respectively. The cDNA sequences of the homologs were determined, and their deduced amino acid sequences were aligned with the yeast Med6p (yMed6p) sequence by using ClustalW (52).

RESULTS Isolation of med6 ts alleles. Peptide microsequencing of purified holo-polymerase identified a novel mediator protein encoded by a gene, which we named MED6, which is essential for cell viability (32) (GenBank accession no. U78080). In order to examine the function of MED6 in vivo, we constructed a med6 mutant library in yeast by replacing the MED6 gene with a mutated med6 allele as described in Materials and Methods. To isolate ts mutants among the viable med6 mutant strains, 2,200 library cells were tested for their ability to grow at 30 and 37°C. From this screen, we isolated three med6 mutant strains that showed a growth defect at 37°C. The mutant isolate (YCL8) with the most severe ts phenotype was used for subsequent analyses (Fig. 1A). YCL8 grew normally at 30°C in a glucose medium but ceased to grow within 2 to 4 h after being shifted to 37°C. At this nonpermissive temperature, the morphology of YCL8 be-

4624

LEE ET AL.

FIG. 1. Isolation of the med6 ts allele conferring ts lethality. (A) Growth defect of the med6 ts mutant strain at the nonpermissive temperature. The MED6 wild-type (YCL10) and ts (YCL8) strains were streaked onto synthetic glucose medium and incubated for 2 or 3 days at the permissive (30°C) or nonpermissive (37°C) temperature. (B) The med6 ts gene contained six missense mutations in the ORF. Shown are the mutated amino acids and their positions along with the BamHI site used for a domain-swapping experiment.

MOL. CELL. BIOL.

came abnormal and a multibudded structure was observed (data not shown). Sequencing of the med6 ts allele revealed 10 nucleotide changes in the ORF, resulting in six amino acid substitutions (Fig. 1B). Results of a domain-swapping experiment with a wild-type MED6 gene indicate that the first three mutations were responsible for the ts lethality (data not shown). RNA polymerase specificity of the med6 ts mutation. Because Med6p is a component of the mediator-RNA polymerase II complex, the function of MED6 might be limited to transcription by RNA polymerase II. In order to investigate this possibility, we examined the effect of the med6 mutation on transcription by RNA polymerases I, II, and III by measuring the transcript levels of the rRNA, DED1, and tRNA genes, respectively (Fig. 2A). For rRNA and tRNA, which have a relatively long half-life compared to those of mRNAs, the synthesis rates were measured with probes specific to their precursor forms (8). As expected, the levels of both rRNA and tRNA precursors were not affected by the med6 mutation at the nonpermissive temperature; however, the mRNA level of DED1 was also unaffected (Fig. 2A). There remained the possibility that the med6 mutation affected only certain RNA polymerase II-transcribed genes. Therefore, we tested the effect of the med6 ts mutation on activated transcription of genes involved in amino acid metabolism (HIS3 and TRP3) and carbon source metabolism (GAL1 and PYK1), as well as on MED6 itself. The med6 ts mutation abolished activated transcription of GAL1 and diminished PYK1 transcription two- to threefold at the nonpermissive temperature. However, the med6 ts allele had no effect on transcription of TRP3 and MED6. In addition, the ratio of constitutive (11) to activated (113) transcription of HIS3 was not altered by med6 mutation, although the level of total HIS3 transcript in mutant cells was elevated due to an artificial condition placing both med6 and HIS3 genes in the

FIG. 2. In vivo analyses of the transcriptional defects of the med6 ts mutant. (A) Effects of the med6 ts allele on transcription. Wild-type (WT) and med6 ts mutant (TS) cells were grown in YPD or YP plus galactose medium (for GAL1) at 30°C to the early exponential phase, and then total RNA was prepared after cultivation at the indicated temperature for another 2.5 h. Total RNA (30 mg) or poly(A)1 RNA (1 mg) was used for S1 nuclease protection or Northern hybridization (for PYK1) assays, respectively, to measure the transcript levels of the indicated genes. S1 nuclease protection assay of HIS3 detected both the constitutively expressed (11) and the regulated (113) transcripts. (B) Effects of the med6 ts mutation on transcription of GAL1 and SUC2 genes. Total RNA was prepared and used for S1 nuclease protection assays from cells grown at the indicated temperatures in YPD (YPGlc), YP plus galactose (YPGal), or YP plus raffinose (YPRaf) medium, as described in Materials and Methods. To show the relative abundance of GAL1 and SUC2 transcripts, DED1 transcript levels are shown as an internal control.

VOL. 17, 1997


same plasmid (Fig. 2A). Therefore, the med6 mutation affected activated transcription only of specific genes transcribed by RNA polymerase II. Effects of the med6 ts allele on transcriptional regulation of the GAL1 and SUC2 genes. Consistent with the defect of the med6 ts mutation in activated transcription of GAL1, the med6 ts mutant strains showed a two- to threefold-lower growth rate than did the wild-type strain, even at the permissive temperature, when a carbon source other than glucose was supplied (data not shown). This result prompted us to examine the effect of the med6 mutation on transcription of the SUC2 gene as well. An S1 nuclease protection assay revealed that the expected activation of GAL1 transcription in galactose media and derepression of SUC2 transcription in raffinose media were completely inhibited in the med6 ts mutant when cultured at the nonpermissive temperature (Fig. 2B). In contrast to the effect on transcriptional activation and derepression, the repression of GAL1 and SUC2 transcription in glucose media was not affected by the med6 ts mutation (Fig. 2B). This result suggests that the ts activity of Med6p functions mainly in the activated transcription of GAL1, PYK1, and SUC2. Effect of the med6 mutation on mating-type-specific genes. Defects in carbon source metabolism and abnormal morphology are common phenotypes that the med6 mutant shares with other mediator mutant alleles (gal11, rgr1, and sin4). Thus, we examined whether the med6 ts allele also had a defect in transcription of mating-type specification genes, as observed in the gal11 null mutant (10, 39). First, we tested the amount of a-factor produced by wild-type and med6 mutant strains by use of a halo (growth inhibition) assay. Production of a-factor can be assayed by the size of the growth inhibition area of a tester strain (MATa sst1, the strain supersensitive to mating factor a). As shown in Fig. 3A, the size of the halo around the parental wild-type strain (YCL4) was about threefold greater than that around the med6 ts strain (YCL8), indicating a reduction in a-factor production by the med6 ts mutant, even at the permissive temperature. To determine whether the impaired a-factor production resulted at least in part from a defect in transcriptional activation of MFa1 (which encodes the majority of a-factor in the cell [29]), we measured the level of MFa1 transcripts. We also examined the expression of the transcriptional activator (Mata1p) and coactivator (Mcm1p) of MFa1 to distinguish between a direct and an indirect effect on the MFa1 transcription (16). Northern blot analysis revealed that the transcription of MFa1 was diminished substantially in the med6 mutant strain grown at the restrictive temperature, whereas the levels of MATa1 and MCM1 transcripts were not affected (Fig. 3B). Therefore, Med6p plays a crucial role in the activated transcription of MFa1. The regulation of cell type specification genes requires transcriptional repression as well as activation. In a-cells, the transcription of a-specific genes, such as BAR1, is inhibited by the repressor complex Mata2p-Mcm1p (16). Although the med6 mutation had no effect on the repression of GAL1 and SUC2 transcription, we investigated the role of Med6p in transcriptional repression of BAR1, since three mediator components, Gal11p, Rgr1p, and Sin4p, have been demonstrated to function as both positive and negative regulators of transcription (10, 20). As shown in Fig. 3B, the repressed level of BAR1 transcript was not increased but rather decreased slightly by the med6 mutation at the nonpermissive temperature. This result suggests that at least our med6 ts allele does not affect the transcriptional repression of a-specific gene BAR1, either. Effects of the med6 ts mutation on nuclear extract transcription. In vivo analyses described above revealed that Med6p plays a role in activated transcription of some sets of genes

4625

FIG. 3. Effects of the med6 ts mutation on a-factor production. (A) Mating factor assay. MED6 wild-type (wt) (MATa) and ts mutant (MATa) strains, along with the MATa and MATa cells as a positive and a negative control, respectively, were spotted onto a lawn of MATa sst1 tester cells (RC629W) grown on YPD medium and incubated for 3 days at 30°C. The size of the halo around the cell spots represents the amount of a-factor produced by the spotted strains. (B) Effect of the med6 ts mutation on transcription of genes involved in a-factor production. The indicated cells were grown in YPD medium at 30°C to early exponential phase and then grown for another 2.5 h at the indicated temperature. Poly(A)1 RNA (1 mg) prepared from each sample was analyzed by Northern hybridization with RNA probes complementary to the indicated gene. To show that the same amount of poly(A)1 RNA was used in each lane, the filter was stripped and rehybridized with a DED1 oligonucleotide probe.

transcribed by RNA polymerase II. To investigate the effect of the med6 ts mutation on activated transcription in vitro, we tested the transcriptional activities of nuclear extracts prepared from wild-type and med6 mutant cells. As shown in Fig. 4A, wild-type and mutant nuclear extracts exhibited similar activities in both activated and basal transcriptions when conducted at 25°C (lanes 1, 2, 6, and 7). Incubation of the nuclear extracts at 30 to 34°C prior to initiation of the transcription reaction at 25°C caused a gradual decrease of both basal and activated transcriptions without changing each enzyme’s activity as measured by fold activation (14- to 16-fold, based on quantitation using a PhosphorImager). However, preincubation of the mutant extract at 37°C caused a severe decrease in activated transcription rather than basal transcription (Fig. 4A, lane 10), resulting in only a 6-fold activation compared to the 16-fold activation by the wild-type nuclear extract under the same condition (Fig. 4A, lane 5). To further confirm that this transcriptional defect was due to the med6 mutation, we tested whether r-Med6p can rescue the transcriptional defect of the med6 mutant nuclear extract (Fig. 4B). The addition of

4626

LEE ET AL.

MOL. CELL. BIOL.

tracts support strongly the in vivo results implying that heatsensitive function of Med6p is required for transcriptional activation rather than basal transcription. Subunit composition of med6 mutant holo-polymerase. In order to address whether the med6 mutation was solely responsible for the defect in transcriptional activation, we isolated mediator-RNA polymerase II complex (holo-polymerase) from the med6 mutant strain and analyzed its subunit composition. Yeast whole-cell extracts from wild-type and mutant strains were fractionated by Bio-Rex70, DEAE-Sepharose, hydroxyapatite, and MonoQ column chromatographies as described by Kim et al. (26). The wild-type and mutant holo-polymerases possessed comparable biochemical characteristics during column chromatographies. Western blot (immunoblot) analyses demonstrated that Med6p always comigrated with other mediator components during the purification (Fig. 5A and data not shown), as described previously (31). Likewise, mutant Med6 protein (med6p) also fractionated together with other mediator subunits throughout the purification (Fig. 5B and data not shown). However, the amount of med6p as measured with antibodies to wild-type Med6p in the mutant holo-polymerase fraction was only about 10 to 15% of the amount of Med6p in the wild-type fraction (Fig. 6A). Because the antiMed6p antibodies cross-react equally well with both wild-type and mutant Med6 recombinant proteins (data not shown), med6p in the mutant fraction was substoichiometric to other mediator components. The deficiency of med6p in the mutant

FIG. 4. In vitro transcription with wild-type and med6 ts mutant nuclear extracts. Nuclear extracts containing 50 to 60 mg of protein were preincubated for 10 min at the indicated temperature. Additional supplements containing ribonucleotides, two templates [pJJ470 and pS(GCN4)2CG2 (26)], and [a-32P]UTP with and without Gal4VP16 were then added to initiate transcription at 25°C. The reaction mixture was then incubated for 50 min. Specifically initiated transcripts from templates containing Gcn4p-binding sites (GCN4:G2) or the Gal4pbinding site (GAL4:G2) are indicated. (A) Effects of the med6 ts mutation on transcriptional activation. Nuclear extracts made from wild-type cells (YCL10) or from med6 ts cells (YCL8) were tested for basal and activated transcription without (lanes 1, 2, 6, and 7) or with (lanes 3 to 5 and 8 to 10) preincubation at the indicated temperatures for 10 min as described in Materials and Methods. For activated transcription, Gal4VP16 (30 ng) was added to the reaction mixture. The amounts of wild-type (50 mg) and mutant (60 mg) nuclear extracts used in these experiments were normalized so that the same amount of transcription activity could be obtained from each reaction mixture when transcription was done without a heat treatment. (B) Rescue of the transcriptional defect in mutant nuclear extract by the addition of r-Med6p. Wild-type and med6 ts mutant nuclear extracts were assayed for transcription activity without (lanes 1, 2, 5, and 6) or with (lanes 3, 4, and 7 to 9) preincubation at 37°C. The effect of the addition of r-Med6p (80 ng; lanes 2, 4, 6, and 8) or r-Srb5p (80 ng; lane 9) to the indicated nuclear extracts was measured in the presence of Gal4VP16.

r-Med6p to the wild-type nuclear extract had no effect on either basal or activated transcription, regardless of whether the extract was preincubated at 37°C (Fig. 4B, lanes 3 and 4; 12-fold activation) or not (lanes 1 and 2; 11-fold activation). However, preincubation of mutant extract with r-Med6p rescued completely both its temperature-dependent defect in transcriptional activation (Fig. 4B, lane 4 [12-fold activation] versus lane 8 [10-fold activation]) and its temperature-independent defect in basal transcription to the level of wild-type extract (mutant holo-polymerase had a slightly lower transcription activity than wild-type enzyme even under the permissive condition; see Fig. 7A and text). This complementary activity was specific to Med6p; r-Srb5p had no effect on transcription when tested in an identical assay (Fig. 4B, lanes 7 and 9; fivefold activation). Therefore, in vitro data with nuclear ex-

FIG. 5. Copurification of wild-type and mutant Med6p with other mediator components. Hydroxyapatite column fractions (10 ml each) from wild-type (A) and med6 ts mutant (B) strains were analyzed by immunoblotting with antibodies specific for Rpb1, Srb4, Med6, and Srb2 proteins as indicated. The peak of holo-polymerase eluted at around 90 mM potassium phosphate for both wildtype and mutant fractions. Lane L, the sample loaded onto the column; numbered lanes, fraction numbers.

VOL. 17, 1997


4627

FIG. 6. Polypeptide composition of mediator-RNA polymerase II complexes from wild-type and med6 ts mutant strains. (A) Immunoblot analysis of holopolymerases. Holo-polymerases (3 mg; MonoQ fraction) purified from wild-type (WT) and mutant (TS) strains were subjected to Western blot analysis with antibodies specific to the proteins indicated at the right. (B) SDS-polyacrylamide gel electrophoresis of holo-polymerases. MonoQ fractions (50 ml) of wild-type and mutant holo-polymerases were immunoprecipitated with affinity-purified anti-Srb5p antibody coupled to protein A-agarose (30 ml) as described in Materials and Methods. Molecular masses (in kilodaltons) of the size marker proteins (Bio-Rad) are indicated on the right, and positions of the RNA polymerase II subunits and some known mediator components (including Med6p) are also indicated.

holo-polymerase did not result from the dissociation of the mutant protein during the purification procedure, because the total amount of med6p in the mutant nuclear extract was also about 10 to 15% of that in the wild-type nuclear extract (data not shown). In addition, no dissociated form of med6p was detected in any other fractions throughout the entire purification procedure (data not shown). Therefore, all of the med6p appears to be associated with mediator complex in vivo. To examine whether the decrease in the amount of mutant med6p disrupted the association of other mediator components in the mutant holo-polymerase, wild-type and mutant holo-polymerases were immunoprecipitated with antibodies to Srb5p, and the immune complexes were analyzed by silver staining. The polypeptide compositions of the two holo-polymerases were indistinguishable except for the large reduction in the amount of a 38-kDa polypeptide in the mutant holopolymerase (Fig. 6B). This 38-kDa polypeptide cross-reacts with antibodies to Med6p (data not shown). Immunoprecipitation of holo-polymerases demonstrated not only that Med6p is a component of holo-polymerase, but also that no other component of holo-polymerase was absent from the mutant holo-polymerase. In order to study how med6p affected transcription, we overexpressed med6p carried on a multicopy plasmid in med6 mutant cells. However, the increased number of copies of the med6 ts allele did not elevate the level of med6p in mutant cell; thus, the ts lethality of mutant cells was not suppressed (data not shown). These results suggest that ts mutation in med6p makes it loosely associated with holo-polymerase, which causes a rapid degradation of the mutant med6p regardless of its expression level. Biochemical properties of the med6 mutant holo-polymerase. Previous work has shown that mediator-RNA polymerase II complex (holo-polymerase) has three functional characteristics distinct from those of core-RNA polymerase II (26): (i) stimulation of basal transcription (5- to 10-fold), (ii) responsiveness to activator proteins, and (iii) a substantial increase in the phosphorylation efficiency of Rpb1 CTD by TFIIH (10- to

40-fold). We therefore examined the effect of the med6 mutation on each of the above activities of holo-polymerase in a defined system as described in Materials and Methods. In order to demonstrate ts activity of med6p, all the components of the transcription reaction mixture except CTP and UTP were preincubated at 25 or 37°C for 10 min, and then the transcripts were elongated at 25°C by the addition of the nucleotides. We examined the effect of the med6 ts mutation on basal transcription. When equal amounts of holo-polymerases were tested for basal transcription activities, the med6 mutant holopolymerase displayed 70% of the activity of wild-type holopolymerase (Fig. 7A, lanes 1 and 6). Preincubation at 37°C lowered the basal transcription activity of the mutant holopolymerase slightly (Fig. 7A, lanes 3 and 8) (less than a twofold decrease was observed consistently). On the other hand, an obvious temperature-dependent defect in transcriptional activation from med6 mutant holo-polymerase was observed. When transcription reaction mixtures were preincubated at 25°C in the presence of Gal4VP16 (Fig. 7A, lanes No), both mutant and wild-type holo-polymerases specifically activated transcription from the GAL4 enhancercontaining template more than 40-fold (43- and 40-fold, respectively; Fig. 7A, lanes 2 and 7). However, when a heat treatment (10 min at 37°C) was administered during initiation complex formation and followed by elongation at 25°C (Fig. 7A, lanes DI), activated transcription by the med6 mutant holo-polymerase was diminished (sixfold; Fig. 7A, lane 9). Under equivalent conditions, activated transcription by wild-type holo-polymerase was reduced only slightly, probably as a result of a nonspecific heat inactivation of the enzymes (28-fold; Fig. 7A, lane 4). We further confirmed that the crippled transcriptional activation by the mutant holo-polymerase resulted specifically from the mutant med6p by rescue of its defect with r-Med6p. Addition of r-Med6p to the mutant holo-polymerase during the heat treatment restored its activity to that of wildtype holo-polymerase (24-fold; Fig. 7A, lane 10). This rescuing activity was specific to Med6p function; r-Srb5p failed to ex-

4628

LEE ET AL.

MOL. CELL. BIOL.

FIG. 7. Comparison of wild-type and med6 ts mutant holo-polymerase activities in a reconstituted transcription system and CTD phosphorylation reaction. (A) Effects of the med6 ts mutation on a reconstituted in vitro transcription. At the top of the figure, the reaction conditions for initiation complex formation and transcription are illustrated. The effects of heat treatment on transcription during the initiation period (DI) (lanes 3, 4, and 8 to 10) and after the initiation period prior to elongation (AI) (lanes 5 and 11) were examined by comparison to the transcription reaction without heat treatment (No) (lanes 1, 2, 6, and 7). During initiation or heat treatment, holo-polymerase (holo-polII), general transcription factors (GTFs [TBP, TFIIB, TFIIE, and TFIIH]), two DNA templates, ATP, and buffer were incubated together at the indicated temperature. Activator (Gal4VP16; 30 ng) and r-Med6p (80 ng) were also added to some reaction mixtures (1). Only after initiation complex formation (No) or heat treatment (DI and AI), [a-32P]UTP and CTP were added to the reaction mixture to start elongation. Wild-type or med6 mutant holo-polymerases (700 ng each), which have the same level of nonspecific transcription activity, were used in all transcription reactions. GCN4:G2 and GAL4:G2 are as defined in the legend to Fig. 4. (B) Effect of the med6 ts mutation on CTD phosphorylation by TFIIH. Core-polymerase (C) (100 ng) and wild-type (W) (300 ng) and med6 ts mutant (T) (300 ng) holo-polymerases (equivalent amounts of each for nonspecific transcription activity) were incubated with yeast TFIIH (40 ng) and [g-32P]ATP, and phosphorylation of Rpb1 was analyzed by SDS-polyacrylamide gel electrophoresis as described in Materials and Methods. To examine the effect of heat treatment on CTD phosphorylation, the same amount of each polymerase was preincubated for 10 min at 37°C with (lanes 6 and 8) or without (lanes 4, 5, and 7) r-Med6p (30 ng) and subjected to the CTD phosphorylation reaction. The position of phosphorylated Rpb1 is indicated (arrow). (C) Quantitation of transcription and CTD phosphorylation of holo-polymerases. Basal and activated transcriptions of wild-type and med6 ts mutant holo-polymerases (left panel) were quantitated by using a PhophorImager, and their activities are shown as fold activation. Fold stimulation of CTD phosphorylation of holo-polymerase compared to that of core-RNA polymerase (right panel) was also quantitated.

hibit a similar effect when added to the mutant holo-polymerase, and r-Med6p had no effect on transcription when added to wild-type holo-polymerase (data not shown). Although heat treatment of mutant holo-polymerase during initiation complex formation caused an irreversible defect, heat treatment administered after initiation complex formation (Fig. 7A, lanes AI) caused no mutant-specific defect; the mutant holo-polymerase was almost as potent as the wild type in transcriptional activation (21-fold for the wild type and 15-fold for the mutant; Fig. 7A, lanes 5 and 11). Therefore,

Med6p activity is required for activated transcription during the initiation period, and once initiation complex is formed, the heat-sensitive Med6p function is dispensable. Finally, we examined the effect of the med6 mutation on CTD phosphorylation efficiency of holo-polymerases. It has been suggested that CTD phosphorylation is an important step in the transition from initiation to elongation, and a higher CTD phosphorylation efficiency may reflect a larger amount of transcription activity. When equal amounts of polymerases were phosphorylated by TFIIH, both holo-polymerases were

VOL. 17, 1997


4629

FIG. 7—Continued.

phosphorylated at least 10 times more efficiently than corepolymerase (13-fold for the wild type and 10-fold for the mutant; Fig. 7B, lanes 2 and 3), regardless of whether they were incubated at 37°C prior to phosphorylation at 25°C (22-fold for the wild type and 14-fold for the mutant; Fig. 7B, lanes 5 and 7). Although the CTD phosphorylation efficiency of the mutant holo-polymerase was independent of the heat-sensitive activity of Med6p, it was 30% less efficient than that of wild-type holo-polymerase under the same conditions. This defect appeared to result from an indirect effect of the med6 mutation, as it was not rescued by the addition of r-Med6p (15-fold; Fig. 7B, lane 8). Cloning of MED6 homologs from humans and C. elegans. Database searches for MED6-related genes using the yMed6p sequence identified expression sequence tags similar to yMed6p from the human and C. elegans EST databases. The sequences of the human and C. elegans EST clones were determined, and they predict a 157-amino-acid protein for humans and a 246-amino-acid protein for C. elegans. Comparison of the protein sequences of the human and C. elegans Med6p homologs to the yMed6p sequence showed 30 and 34% similarity, respectively (Fig. 8). Despite the sequence homology, neither the full-length human MED6 homolog nor several yeast-human chimeric genes complement the MED6 activity in yeast (data not shown). This suggests that species-specific interactions between MED6 and the transcriptional apparatus may be required for proper transcriptional regulation. In order

to confirm the functional similarity of the Med6p homolog, we are examining the occurrence of the homolog in human holopolymerase and its role in transcriptional activation. DISCUSSION From a genetic standpoint, Med6p defines a new class of yeast proteins that are not components of either an activatorenhancer assembly or of the basal transcription apparatus but which are nonetheless absolutely required for activated transcription. Furthermore, it is broad acting, as shown by the pleiotropic effect of the med6 ts allele. The absolute requirement of Med6p for activated transcription of several diverse yeast promoters demonstrates the general importance of the mediator complex in transcriptional activation in vivo. This general requirement of Med6p for activated transcription differs from that of other transcriptional cofactors, such as TAFIIs and USA (for reviews, see references 23 and 53). TAFIIs associate with the general transcription factor TBP, and extensive biochemical studies demonstrate their essential role in transcriptional activation and promoter selectivity. However, recent in vivo analysis of TAFIIs in yeast reveals that depletion of TAFIIs has no effect on transcription of most promoters tested, despite their requirement for cell viability (37, 54). Therefore, TAFIIs are dispensable for general transcriptional activation but are required for transcription of a

4630

LEE ET AL.

MOL. CELL. BIOL.

FIG. 8. Yeast, C. elegans, and human Med6 protein sequence alignments. Sequences of Med6p homologs from yeast (GenBank no. U78080), C. elegans (GenBank no. U78081), and humans (GenBank no. U78082) are aligned. Identity (black boxes), similar amino acids (shaded boxes), and gaps (dashes) are indicated.

subset of essential genes involved in cell cycle regulation (2) or cell differentiation (9). Another transcriptional cofactor, called USA (36), contains several independent cofactors, including PC4. PC4 appears to enable transcriptional activation by facilitating basal factoractivator interactions (24, 46). Functional analysis of the yeast homolog (Tsp1p/Sub1p) of mammalian PC4 showed that the protein interacted with TFIIB and stimulated transcriptional activation, but unlike Med6p, it was not required for cell viability (15, 27). In contrast to TAFIIs and Tsp1p/Sub1p, Med6p is essential for transcriptional activation of most of the promoters we tested in yeast. Among the genes tested (GAL1, SUC2, PYK1, MFa1, and HIS3), activated transcription of only the HIS3 gene was not affected by the med6 allele. Transcriptional activation of all other genes tested was reduced to the uninduced level. We also reproduced the med6 defect in transcriptional activation in a defined in vitro system in the absence of TAFIIs, USA, or the SWI-SNF complex. These results support the in vivo results indicating that mediator has a more general role in transcriptional activation in yeast than do other known cofactors. Identification of MED6-related genes in humans and C. elegans may suggest that a similar transcriptional regulation mechanism is used in higher eukaryotes as well. An RNA polymerase II complex has been purified from human cells, and a human homolog of yeast holo-polymerase component SRB7 has been identified (5, 35, 41). Although MED6 is required for transcriptional activation of most promoters we tested, the med6 ts mutation did not affect transcriptional activation from all cellular promoters. The effect of the med6 allele was observed for the GAL4- and MATa1-regulated genes but not for GCN4-regulated genes. The difference might result from allele specificity of the mutant we analyzed. However, it is also possible that distinct classes of activators may interact with different mediator subunits or even different complexes, such as TFIID or USA. The selec-

tivity in activator-mediator interaction also suggests that combinatorial interactions between activators and mediator subunits may be used to increase the diversity of transcriptional regulation. Another striking aspect of the med6 allele is its effect on only the activated, and not uninduced or repressed, transcription. In addition, neither transcription of constitutively expressed genes nor higher CTD phosphorylation efficiency, both of which correlate with higher basal transcriptional activity of holo-polymerase, was affected by the med6 mutation. Therefore, unlike the Srb proteins, which are required for general transcription by RNA polymerase II (51), Med6p is unique in that it is required only for activated transcription. On the other hand, another group of mediator subunits, including Rgr1p, Sin4p, and Gal11p, is involved in transcriptional repression as well as activation (10, 20, 22, 44). The observation that the various mediator functions (activation, repression, and stimulated basal transcription) appear to require different sets of mediator proteins suggests that the mediator complex is composed of several functional domains. The minimal number of mediator subunits required for transcriptional activation by RNA polymerase II remains to be deciphered. We demonstrated that Med6p is absolutely required for transcriptional activation both in vivo and in vitro. However, the inability of r-Med6p alone to mediate transcriptional activation in an in vitro system comprising homogeneous basal factors and core-RNA polymerase II indicates that additional mediator components collaborate with Med6p to achieve transcriptional activation (data not shown). Whether these mediator components have been identified (such as Rgr1p, Sin4p, and Gal11p) or are as yet uncharacterized can be determined by isolating proteins that interact with Med6p. ACKNOWLEDGMENTS We thank Hyenseung Shin, Yeonsoo Seo, Sungsook Rhee, Soonjung Park, Jeongsil Kim-Ha, and Kelly LaMarco for careful reading and


VOL. 17, 1997

critical comments on the manuscript. We offer special thanks to Changsoo Kang for help in sample preparation, Juri Kim for technical assistance, and Yuji Kohara for providing the yk60911 C. elegans EST clone. We thank Yang Li, Jasper Svejstrup, and Roger Kornberg for providing MED6 clones, TFIIH, and antibodies. We also thank Toshio Fukasawa and Rick Young for Gal11 and Srb antibodies. This work was supported by a grant (B95004) from SBRI to Y.K. REFERENCES 1. Altschul, S. F., G. Gish, G. Miller, E. W. Myers, and D. J. Lipman. 1990. Basic local alignment search tool. J. Mol. Biol. 215:403–410. 2. Apone, L. M., C. A. Virbasius, J. C. Reese, and M. R. Green. 1996. Yeast TAFII90 is required for cell-cycle progression through G2/M but not for general transcription activation. Genes Dev. 10:2368–2380. 3. Barlev, N. A., R. Candau, L. Wang, P. Darpino, N. Silverman, and S. L. Berger. 1995. Characterization of physical interactions of the putative transcriptional adaptor, ADA2, with acidic activation domains and TATA-binding protein. J. Biol. Chem. 270:19337–19344. 4. Bjorklund, S., and Y.-J. Kim. 1996. Mediator of transcriptional regulation. Trends Biochem. Sci. 21:335–337. 5. Chao, D. A., E. L. Gadbios, P. J. Murray, S. F. Anderson, M. S. Sonu, J. D. Parvin, and R. A. Young. 1996. A mammalian SRB protein associated with an RNA polymerase II holoenzyme. Nature 380:82–85. 6. Chen, W., S. Tabor, and K. Struhl. 1987. Distinguishing between mechanisms of eukaryotic transcriptional activation with bacteriophage T7 RNA polymerase. Cell 50:1047–1055. 7. Cismowski, M. J., G. M. Laff, M. J. Solomon, and S. I. Reed. 1995. KIN28 encodes a C-terminal domain kinase that controls mRNA transcription in Saccharomyces cerevisiae but lacks cyclin-dependent kinase-activating kinase (CAK) activity. Mol. Cell. Biol. 15:2983–2992. 8. Cormack, B. P., and K. Struhl. 1992. The TATA-binding protein is required for transcription by all three nuclear RNA polymerases in yeast cells. Cell 69:685–696. 9. Dikstein, R., S. Zuo, and R. Tjian. 1996. Human TAFII105 is a cell typespecific TFIID subunit related to hTAFII130. Cell 87:137–146. 10. Fassler, J. S., and F. Winston. 1989. The Saccharomyces cerevisiae SPT13/ GAL11 gene has both positive and negative regulatory roles in transcription. Mol. Cell. Biol. 9:5602–5609. 11. Flanagan, P. M., R. J. D. Kelleher, M. H. Sayre, H. Tschochner, and R. D. Kornberg. 1991. A mediator required for activation of RNA polymerase II transcription in vitro. Nature 350:436–438. 12. Goodrich, J. A., G. Cutler, and R. Tjian. 1996. Contacts in context: promoter specificity and macromolecular interactions in transcription. Cell 84:825– 830. 13. Goodrich, J. A., and R. Tjian. 1994. TBP-TAF complexes: selectivity factors for eukaryotic transcription. Curr. Opin. Cell Biol. 6:403–409. 14. Hengartner, C. J., C. M. Thompson, J. Zhang, D. M. Chao, S. M. Liao, A. J. Koleske, S. Okamura, and R. A. Young. 1995. Association of an activator with an RNA polymerase II holoenzyme. Genes Dev. 9:897–910. 15. Henry, N. L., D. A. Bushnell, and R. D. Kornberg. 1996. A yeast transcriptional stimulatory protein similar to human PC4. J. Biol. Chem. 271:21842– 21847. 16. Herskowitz, I., J. Rein, and J. N. Strathern. 1992. Mating type determination and mating-type interconversion in Saccharomyces cerevisiae, p. 583–656. In E. W. Jones, J. R. Pringle, and J. Broach (ed.), The molecular and cellular biology of the yeast Saccharomyces, vol 2. Cold Spring Harbor Laboratory Press, Plainview, N.Y. 17. Horiuchi, J., N. Silverman, G. A. Marcus, and L. Guarente. 1995. ADA3, a putative transcriptional adapter, consists of two separatable domains and interacts with ADA2 and GCN5 in a trimeric complex. Mol. Cell. Biol. 15:1203–1209. 18. Ingles, C. J., M. Shales, W. D. Cress, S. J. Triezenberg, and J. Greenblatt. 1991. Reduced binding of TFIID to transcriptionally compromised mutants of VP16. Nature 351:588–590. 19. Janknecht, R., and T. Hunter. 1996. A growing coactivator network. Nature 383:22–23. 20. Jiang, Y. W., P. R. Dohrmann, and D. J. Stillman. 1995. Genetic and physical interactions between yeast RGR1 and SIN4 in chromatin organization and transcriptional regulation. Genetics 140:47–54. 21. Jiang, Y. W., and D. J. Stillman. 1992. Involvement of the SIN4 global transcriptional regulator in the chromatin structure of Saccharomyces cerevisiae. Mol. Cell. Biol. 12:4503–4514. 22. Jiang, Y. W., and D. J. Stillman. 1995. Regulation of HIS4 expression by the Saccharomyces cerevisiae SIN4 transcriptional regulator. Genetics 140:103– 114. 23. Kaiser, K., and M. Meisterernst. 1996. The human general co-factors. Trends Biochem. Sci. 21:342–345. 24. Kaiser, K., G. Stelzer, and M. Meisterernst. 1995. The coactivator p15 (PC4) initiates transcriptional activation during TFIIA-TFIID-promoter complex formation. EMBO J. 14:3520–3527. 25. Kelleher, R. J. D., P. M. Flanagan, and R. D. Kornberg. 1990. A novel

26. 27. 28. 29. 30. 31.

32. 33. 34. 35.

36. 37. 38. 39.

40. 41. 42. 43. 44. 45. 46. 47. 48. 49.

50. 51. 52.

4631

mediator between activator proteins and the RNA polymerase II transcription apparatus. Cell 61:1209–1215. Kim, Y.-J., S. Bjorklund, Y. Li, M. H. Sayre, and R. D. Kornberg. 1994. A multiprotein mediator of transcriptional activation and its interaction with the C-terminal repeat domain of RNA polymerase II. Cell 77:599–608. Knaus, R., R. Pollock, and L. Guarente. 1996. Yeast SUB1 is a suppressor of TFIIB mutations and has homology to the human co-activator PC4. EMBO J. 15:1933–1940. Koleske, A. J., and R. A. Young. 1994. An RNA polymerase II holoenzyme responsive to activators. Nature 368:466–469. Kurjan, J., and I. Herskowitz. 1982. Structure of a yeast pheromone gene (MF a): a putative alpha-factor precursor contains four tandem copies of mature a-factor. Cell 30:933–943. Leung, D. W., E. Chen, and D. V. Goeddel. 1989. A method for random mutagenesis of a defined DNA segment using a modified polymerase chain reaction. Technique 1:11–15. Li, Y., S. Bjorklund, Y. W. Jiang, Y.-J. Kim, W. S. Lane, D. J. Stillman, and R. D. Kornberg. 1995. Yeast global transcriptional regulators Sin4 and Rgr1 are components of mediator complex/RNA polymerase II holoenzyme. Proc. Natl. Acad. Sci. USA 92:10864–10868. Li, Y., Y. C. Lee, Y. D. Yoo, S. Bjorklund, Y.-J. Kim, and R. D. Kornberg. Unpublished work. Liao, S. M., J. Zhang, D. A. Jeffery, A. J. Koleske, C. M. Thompson, D. M. Chao, M. Viljoen, H. J. van Vuuren, and R. A. Young. 1995. A kinase-cyclin pair in the RNA polymerase II holoenzyme. Nature 374:193–196. Lue, N. F., and R. D. Kornberg. 1987. Accurate initiation at RNA polymerase II promoters in extracts from Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 84:8839–8843. Maldonado, E., R. Shiekhattar, M. Sheldon, H. Cho, R. Drapkin, P. Rickert, E. Lees, C. W. Anderson, S. Linn, and D. Reinberg. 1996. A human RNA polymerase II complex associated with SRB and DNA-repair proteins. Nature 381:86–89. Meisterernst, M., A. L. Roy, H. M. Lieu, and R. G. Roeder. 1991. Activation of class II gene transcription by regulatory factors is potentiated by a novel activity. Cell 66:981–993. Moqtaderi, Z., Y. Bai, D. Poon, P. A. Weil, and K. Struhl. 1996. TBPassociated factors are not generally required for transcriptional activation in yeast. Nature 383:188–191. Muhlrad, D., R. Hunter, and R. Parker. 1992. A rapid method for localized mutagenesis of yeast genes. Yeast 8:79–82. Nishizawa, M., Y. Suzuki, Y. Nogi, K. Matsumoto, and T. Fukasawa. 1990. Yeast Gal11 protein mediates the transcriptional activation signal of two different transacting factors, Gal4 and general regulatory factor I/repressor/ activator site binding protein1/translation upstream factor. Proc. Natl. Acad. Sci. USA 87:5373–5377. Nonet, M. L., and R. A. Young. 1989. Intragenic and extragenic suppressors of mutations in the heptapeptide repeat domain of Saccharomyces cerevisiae RNA polymerase II. Genetics 123:715–724. Ossipow, V., J. P. Tassan, E. A. Nigg, and U. Schibler. 1995. A mammalian RNA polymerase II holoenzyme containing all components required for promoter-specific transcription initiation. Cell 83:137–146. Paull, T. T., M. Carey, and R. C. Johnson. 1996. Yeast HMG proteins NHP6A/B potentiate promoter-specific transcriptional activation in vivo and assembly of preinitiation complexes in vitro. Genes Dev. 10:2769–2781. Roberts, S. G., I. Ha, E. Maldonado, D. Reinberg, and M. R. Green. 1993. Interaction between an acidic activator and transcription factor TFIIB is required for transcriptional activation. Nature 363:741–744. Sakai, A., Y. Shimizu, and F. Hishinuma. 1988. Isolation and characterization of mutants which show an oversecretion phenotype in Saccharomyces cerevisiae. Genetics 119:499–506. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Shykind, B. M., J. Kim, and P. A. Sharp. 1995. Activation of the TFIIDTFIIA complex with HMG-2. Genes Dev. 9:1354–1365. Sikorski, R. S., and P. Hieter. 1989. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122:19–27. Stringer, K. F., C. J. Ingles, and J. Greenblatt. 1990. Direct and selective binding of an acidic transcriptional activation domain to the TATA-box factor TFIID. Nature 345:783–786. Suzuki, Y., Y. Nogi, A. Abe, and T. Fukasawa. 1988. Gal11 protein, an auxiliary transcription activator for genes encoding galactose-metabolizing enzymes in Saccharomyces cerevisiae. Mol. Cell. Biol. 8:4991–4999. (Erratum, 12:4806, 1992.) Thompson, C. M., A. J. Koleske, D. M. Chao, and R. A. Young. 1993. A multisubunit complex associated with the RNA polymerase II CTD and TATA-binding protein in yeast. Cell 73:1361–1375. Thompson, C. M., and R. A. Young. 1995. General requirement for RNA polymerase II holoenzymes in vivo. Proc. Natl. Acad. Sci. USA 92:4587–4590. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through

4632

LEE ET AL.

sequence weighting, position specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673–4680. 53. Verrijzer, C. P., and R. Tjian. 1996. TAFs mediate transcriptional activation and promoter selectivity. Trends Biochem. Sci. 21:338–341. 54. Walker, S. S., J. C. Reese, L. M. Apone, and M. R. Green. 1996. Transcription activation in cells lacking TAFIIs. Nature 383:185–188. 55. Wang, Z., S. Buratowski, J. Q. Svejstrup, W. J. Feaver, X. Wu, R. D. Kornberg, T. F. Donahue, and E. C. Friedberg. 1995. The yeast TFB1 and SSL1 genes, which encode subunits of transcription factor IIH, are required

MOL. CELL. BIOL. for nucleotide excision repair and RNA polymerase II transcription. Mol. Cell. Biol. 15:2288–2293. 56. Wilson, C. J., D. M. Chao, A. N. Imbalzano, G. R. Schnitzler, R. E. Kingston, and R. A. Young. 1996. RNA polymerase II holoenzyme contains SWI/SNF regulators involved in chromatin remodeling. Cell 84:235–244. 57. Xiao, H., A. Pearson, B. Coulombe, R. Truant, S. Zhang, J. L. Regier, S. J. Triezenberg, D. Reinberg, O. Flores, C. J. Ingles, and J. Greenblatt. 1994. Binding of basal transcription factor TFIIH to the acidic activation domains of VP16 and p53. Mol. Cell. Biol. 14:7013–7024.