Transcriptional Activation in Yeast Cells Lacking Transcription Factor IIA

Corresponding author: Kevin Struhl, Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, 200 Longwood Ave., Boston, MA 02115., kevin{at}hms.harvard.edu (E-mail)

Communicating editor: M. CARLSON

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The general transcription factor IIA (TFIIA) forms a complex with TFIID at the TATA promoter element, and it inhibits the function of several negative regulators of the TATA-binding protein (TBP) subunit of TFIID. Biochemical experiments suggest that TFIIA is important in the response to transcriptional activators because activation domains can interact with TFIIA, increase recruitment of TFIID and TFIIA to the promoter, and promote isomerization of the TFIID-TFIIA-TATA complex. Here, we describe a double-shut-off approach to deplete yeast cells of Toa1, the large subunit of TFIIA, to <1% of the wild-type level. Interestingly, such TFIIA-depleted cells are essentially unaffected for activation by heat shock factor, Ace1, and Gal4-VP16. However, depletion of TFIIA causes a general two- to threefold decrease of transcription from most yeast promoters and a specific cell-cycle arrest at the G2-M boundary. These results indicate that transcriptional activation in vivo can occur in the absence of TFIIA.

THE general transcription factor IIA (TFIIA) interacts with the TATA-binding protein (TBP) and with promoter DNA to form a TBP-TFIIA-DNA complex (reviewed in ORPHANIDES et al. 1996 ; ROEDER 1996 ). TFIIA stabilizes the association of TBP with the TATA element (IMBALZANO et al. 1994 ), and it expands the TBP-DNA footprint both upstream and downstream of the TATA element (LAGRANGE et al. 1996 ). The two subunits of yeast TFIIA, Toa1 and Toa2, have distinct roles in the TBP-TFIIA-TATA complex (GEIGER et al. 1996 ; TAN et al. 1996 ). Toa2 interacts with the aminoterminal stirrup of TBP, creating an expanded ß-sheet upstream of the TATA element, whereas Toa1 contacts the phosphate backbone of DNA within and upstream of the TATA element. Because the regions of TFIIA contacting TBP and DNA are minimal, large, solvent-exposed surfaces remain available for interactions with other transcription factors at the promoter. In the context of the TBP-containing complex TFIID, TFIIA affects the interaction of several TAF subunits with promoter DNA (OELGESCHLAGER et al. 1996 ) and synergistic binding of TFIID to the TATA and initiator elements (EMAMI et al. 1997 ). Essentially, TFIIA stabilizes the binding of TFIID to the TATA box, and it expands the surface area at the core promoter for interactions with other proteins.

In most reactions in vitro, TFIIA increases the level of both basal and activated transcription (ORPHANIDES et al. 1996 ; ROEDER 1996 ). However, TFIIA is not absolutely required for accurate transcriptional initiation in vitro, and it is dispensable under certain experimental conditions. In some reactions reconstituted with the TFIID complex, TFIIA preferentially enhances the response to transcriptional activators (OZER et al. 1994 ; SUN et al. 1994 ; YOKOMORI et al. 1994 ). The variable effect of TFIIA has been attributed in part to its ability to attenuate the repressive effects of certain factors (e.g., Mot1, NC2, and HMG1) that destabilize the association of TBP with promoter DNA and with other components of the preinitiation complex (AUBLE et al. 1994 ; GE and ROEDER 1994 ; GOPPELT et al. 1996 ; MERMELSTEIN et al. 1996 ; CHICCA et al. 1998 ). In the presence of these inhibitors of TBP function, TFIIA may stimulate transcription by functioning as an antirepressor. However, the biochemical properties of a TFIIA derivative indicate that antirepression is insufficient and, hence, unlikely to be the sole mechanism for transcriptional activation (MA et al. 1996 ).

A more direct role for TFIIA in transcriptional activation is suggested by direct biochemical interactions between TFIIA and a set of diverse activation domains (OZER et al. 1994 ; KOBAYASHI et al. 1995 ; CLEMENS et al. 1996 ; DAMANIA et al. 1998 ). These activators stabilize a complex containing TFIID and TFIIA bound at the TATA element, suggesting that TFIIA can function as a cofactor in the recruitment of TFIID to the promoter (LIEBERMAN and BERK 1994 ; KOBAYASHI et al. 1995 ; DAMANIA et al. 1998 ). Activators may also accelerate a rate-limiting step in transcriptional activation by facilitating the isomerization of a preassembled TFIIA-TFIID-DNA complex to an open promoter state (CHI and CAREY 1996 ). In the case of the ZEBRA activator, high-level-activated transcription involves both the recruitment of limiting components of the transcriptional machinery as well as this concentration-independent, activator-mediated isomerization of the TFIIA-TFIID-DNA complex. Moreover, this isomerization event appears to be limiting for activated transcription, as formation of an open TFIIA-TFIID-DNA complex permits activated transcription in the subsequent absence of the activator protein. Similarly, the nonhistone chromosomal protein HMG-2 potentiates activation at the level of the TFIID-TFIIA-TATA complex without increasing complex formation or altering the footprint on the promoter (SHYKIND et al. 1995 ). As with all biochemical studies, however, the physiological relevance of these observations is unclear.

Both subunits of yeast TFIIA are essential for cell growth, but the physiological role of TFIIA in transcription is unclear. Several studies have addressed this question using mutant derivatives of TBP or TFIIA that alter the TBP-TFIIA interface and severely inhibit TBP-TFIIA-TATA complex formation in vitro. First, a yeast TBP mutant defective for interacting with TFIIA specifically impairs the response to acidic activators, but does not generally affect Pol II transcription (STARGELL and STRUHL 1995 ). The phenotypes conferred by this yeast TBP derivative are largely suppressed by fusion to Toa2, indicating the importance of the TBP-TFIIA interaction; however, this TBP mutant does not address potential functions of TFIIA that are distinct from the TBP interaction. Second, human TBP mutants severely defective for interacting with TFIIA are transcriptionally incompetent in transiently transfected mammalian cells (BRYANT et al. 1996 ). However, this inability to support transcription might not be due simply to a defective TBP-TFIIA interaction because the mutations radically alter exposed surfaces of TBP and, hence, may prevent interactions with other proteins. Third, yeast Toa2 mutants that weaken the interaction with TBP have selective transcriptional effects, and they partially inhibit cell-cycle progression at cytokinesis (OZER et al. 1998A ). However, these Toa2 mutants support cell viability, indicating that they retain a significant degree of TFIIA function.

Because both subunits of TFIIA are essential for yeast cell viability, the most direct approach to analyze TFIIA function is to conditionally inactivate or deplete TFIIA in yeast cells. In this regard, it has been previously shown that a 10-fold reduction of TFIIA slows cell growth and produces only a modest 2-fold effect on transcription from a few promoters (KANG et al. 1995 ). The analyses in this report were limited, however, because of the partial nature of the conditional depletion, the limited set of genes analyzed, and the lack of experiments with activator-induced transcription after reduction of TFIIA levels. Here, we report a transcriptional analysis of cells that contain <1% of the wild-type level of TFIIA. Surprisingly, transcription of a variety of genes is generally reduced only 2- to 3-fold, and the response to a variety of transcriptional activators is largely unaffected. Nevertheless, TFIIA-depleted cells arrest specifically at the G2-M transition. Thus, TFIIA does not appear to be generally required for transcriptional activation, although it contributes to overall transcription levels and plays an important role at a subset of promoters.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Generation of toa1 temperature-sensitive allele:
The template for PCR mutagenesis of the TOA1 coding sequence pML3050 is a TRP1 centromeric plasmid containing a TOA1 allele with an engineered BamHI site upstream of the start codon. PCR mutagenesis was performed under the following conditions: 10 ng template DNA, 50 pmol/primer, 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 0.001% gelatin, 3 mM MgCl₂, 0.21 mM MnCl₂, 0.2 mM dATP, 0.2 mM dGTP, 1 mM dCTP, 1 mM dTTP and 5 units Taq polymerase (Boehringer Mannheim, Indianapolis) for 30 cycles of 94°, 30 sec; 40°, 1 min; 72°, 2 min. Mutagenic frequency in the resulting PCR product was 0.27% per nucleotide. The mutagenized PCR product was cotransformed with a BamHI-NheI-cleaved TRP1 centromeric plasmid bearing the TOA1 locus cleaved (overlap of the PCR product with the gapped plasmid at the 5' and 3' ends was in excess of 20 bp on each side) into BY40, a yeast strain constructed by Brendan Cormack. BY40 is a derivative of KY320 (CHEN and STRUHL 1988 ) in which TOA1 is present on a URA3 multicopy plasmid and the chromosomal locus is disrupted by ADE2. Trp⁺ transformants were tested for viable TOA1 alleles by plasmid shuffling; cells growing on medium containing 5-fluoroorotic acid at 30° were then screened for conditional lethality at 37°. Four of the eight candidates identified reproduced the temperature-sensitive (ts) phenotype upon retransformation of the recovered plasmids, and the derivative with the strongest phenotype, designated toa1-ts, was chosen for further study.

TFIIA depletion in vivo:
The toa1-ts allele was transferred into a TRP1 centromeric plasmid in which the FLAG epitope was engineered at the N terminus of Toa1, and the resulting FLAG-Toa1 DNA was then placed under the control of the GAL1 promoter. Cells containing this GAL1-toa1-ts allele as the sole source of TFIIA were grown at 30° in YP or synthetic complete (SC) medium containing 2% galactose + 0.6% glucose (permissive conditions) to A₆₀₀ = 0.15. Toa1 depletion was performed by shifting cells to conditions of growth in 2% glucose or 37° or both for 8 hr. To determine Toa1 levels, equal amounts of whole-cell protein extracts, as quantitated by Bradford assay, were separated by SDS-PAGE and transferred to an Immobilon P polyvinyldifluoride membrane (Millipore, Bedford, MA). FLAG-Toa1 was detected with an antibody (M5) against the FLAG epitope (IBI Biochemicals, New Haven, CT), using the Phototope-Star detection system (New England Biolabs, Beverly, MA) according to the manufacturers' instructions. In medium containing 2% galactose, Toa1 levels are much higher than are physiological (data not shown). Even under our permissive conditions of 2% galactose, 0.6% glucose, Toa1-ts levels are significantly higher than Toa1 levels in wild-type yeast strains. To analyze the terminal phenotype of TFIIA-depleted cells, wild-type and GAL1-toa1-ts cells were grown in double-shut-off conditions for 8 hr to OD₆₀₀ ~0.8, and cells (10⁷) were harvested, fixed, stained with 4',6-diamidino-2-phenylindole (DAPI), and visualized by fluorescence microscopy as described previously (TRUEHEART et al. 1987 ).

To examine whether TBP derivatives containing mutations on surfaces responsible for interacting with DNA, TFIIA, or TFIIB are synthetically lethal under conditions of limiting TFIIA, we took advantage of the fact that levels of Toa1 expression in strains containing the GAL1-toa1-ts allele can be titrated downward by increasing the amount of glucose in galactose-containing medium. Specifically, TBP alleles on LEU2 centromeric plasmids were introduced into yeast strain CC2, a derivative of BY40 that lacks the chromosomal copy of TBP but contains a TRP1 centromeric plasmid expressing TBP. Trp^- segregants containing the mutated TBP derivatives as sole source of TBP were then transformed with a TRP1 centromeric plasmid expressing Toa1-ts from the GAL1 promoter, and the wild-type TOA1 gene on a URA3 centromeric plasmid was removed by plasmid shuffling.

Transcriptional analysis:
For the initial experiments, cells were harvested after 8 hr in YP medium under double-shut-off conditions as described above. For analysis of Ppr1-dependent activation, cells were harvested following 8 hr under double-shut-off conditions in synthetic minimal medium containing 0.6% casamino acids with or without uracil. For analysis of heat shock factor (Hsf)-dependent activation, cells were grown in glucose-containing media at 30° for 12 hr and harvested following a 15-min heat shock at 37°. For analysis of Ace1-dependent activation, strains were grown in SC medium under double-shut-off conditions and induced with 100 or 400 µM CuSO₄ for 1 hr before harvesting. Gal4-VP16-dependent activation was assayed by growing strains expressing the activator from the ADH1 promoter on a centromeric URA3 plasmid, in synthetic minimal medium containing 0.6% casamino acids and lacking uracil, under double-shut-off conditions for 8 hr. Analysis of Ppr1- and Ace1-dependent activation of his3 transcription was performed in derivatives of BY40 in which the his3 locus is replaced by derivatives containing binding sites for the activators Ppr1 and Ace1 upstream of the his3 TATA region (IYER and STRUHL 1995 ).

For most experiments, total RNA (40 µg, as quantitated by A₂₆₀) was hybridized to completion with a 10- to 100-fold excess of the appropriate ³²P-labeled oligonucleotides and was treated with S1 nuclease as described previously (IYER and STRUHL 1996 ). Transcript levels were quantitated with respect to the tRNA^W internal control by PhosphorImage analysis. The TRP3, rRNA, RPS4, SSA4, HSP104, CUP1, DED1, GAL1, HIS3, and HIS4 oligonucleotides have been described previously (CHEN and STRUHL 1988 ; STARGELL and STRUHL 1995 ; IYER and STRUHL 1996 ). For Northern blot analysis, 25 µg of total RNA was hybridized with the appropriate ³²P-labeled, randomly primed oligonucleotide probes.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Depletion of TFIIA in vivo using a double-shut-off strategy:
Since both the Toa1 and Toa2 subunits of yeast TFIIA are essential for viability, analysis of TFIIA function in vivo requires a method for conditional depletion. Toward this end, we isolated a temperature-sensitive allele of TOA1 (see MATERIALS AND METHODS) that conferred normal growth at 30° while preventing colony formation at 37° (Figure 1A). Unfortunately, long incubation times (>12 hr) in liquid medium at 37° are required to fully deplete or inactivate Toa1, as assayed by both culture density and Western blot analysis of Toa1 levels (data not shown). Therefore, to produce a more efficient shut-off, the toa1-ts allele was placed under the control of the GAL1 promoter. As anticipated, cell growth supported by this GAL1-toa1-ts allele is regulated by glucose on solid medium (Figure 1A) as well as in liquid culture (data not shown). Importantly, cell growth is most efficiently inhibited by shifting the culture to glucose medium at 37° (Figure 1B). The combined effect of glucose repression and protein thermolability results in depletion of Toa1 to <1% of the wild-type level (the limit of detection) within 8 hr. In contrast, either of the single-shut-off protocols (i.e., glucose or temperature shift) results in detectable levels of Toa1 8 hr after glucose or temperature shift (Figure 1C), although depletion beyond the detection limit is achieved at later times.

View larger version (144K): In this window In a new window Download PPT slide   Figure 1. Characterization of a conditional allele of TOA1. (A) Growth of wild-type, toa1-ts, or GAL1-toa1-ts strains on medium containing glucose (glu) or galactose (gal) at 30° or 37°. (B) The GAL1-toa1-ts strain was grown in YP medium containing 2% galactose + 0.6% glucose medium at 30° and then either shifted (•) or unshifted () to medium containing 2% glucose at 37°; the zero time point corresponds to the time of the shift. (C) Western analysis of FLAG-Toa1 protein levels in wild-type or GAL1-toa1-ts strains grown in the indicated medium at the indicated temperatures. The panel on the left represents double-shut-off conditions, whereas the panels on the right indicate single-shut-off conditions. (D) Synthetic lethality. Strains containing the indicated derivatives of TBP and Toa1 were plated on glucose medium at 30°.

We used the toa1-ts allele, a condition of low TFIIA activity, to examine whether TBP mutants with specific defects in preinitiation complex formation would display synthetic lethality. The TBP-Y139A mutation that causes a 100-fold reduced interaction with TFIIA (STARGELL and STRUHL 1995 ; LEE and STRUHL 1997 ) is synthetically lethal with toa1-ts at 30°. In contrast, TBP mutations that weaken interaction with TFIIB (E188A and E189A; LEE and STRUHL 1997 ) or DNA (V161A; LEE and STRUHL 1995 ) are viable under these conditions (Figure 1D). Overexpression of TBP, TFIIB, Srb4 + Srb6 (coexpressed), TAF17, TAF90, TAF130, Ada1, Ada2, Ada3, or Gcn5 does not suppress the slow-growth phenotype caused by conditions of limiting TFIIA (data not shown). In fact, overexpression of TAF130 appears to exacerbate the slow growth caused by partial Toa1 depletion (data not shown), which is in accordance with a genetic interaction between these proteins (KOKUBO et al. 1998 ; OZER et al. 1998B ).

Depletion of TFIIA causes a modest decrease of transcription from most Pol II promoters:
Transcription of a variety of genes was examined in GAL1-toa1 ts cells that were shifted to glucose medium at 37° for 8 hr to eliminate TFIIA (Figure 2). As expected, Pol III transcription, as exemplified by the gene encoding tRNA^W, is unaffected by TFIIA depletion. In contrast, TFIIA depletion causes a consistent two- to fourfold decrease in Pol II transcription from 9 of the 10 different promoters analyzed. In the exceptional case of the ADH1 promoter, loss of TFIIA had no significant effect on transcription. The quantitatively modest but broad effect on Pol II transcription is in marked contrast to depletions of TBP (CORMACK and STRUHL 1992 ; MOQTADERI et al. 1996 ), TFIIB (MOQTADERI et al. 1996 ), the Kin28 subunit of TFIIH (CISMOWSKI et al. 1995 ; VALAY et al. 1995 ), and the Srb4 subunit of Pol II holoenzyme (THOMPSON and YOUNG 1995 ), all of which completely eliminate Pol II transcription. The TFIIA-depleted cells also show a small reduction in the levels of an rRNA precursor species, which is indicative of a slight defect in Pol I transcription. As general decreases in Pol II transcription indirectly reduce Pol I transcription (NONET et al. 1987 ; CORMACK and STRUHL 1992 ), it seems likely that TFIIA is not directly involved in Pol I transcription, but rather influences rRNA synthesis indirectly through its effects on Pol II promoters. These observations indicate that TFIIA has a general but quantitatively modest effect on Pol II transcription in yeast cells.

View larger version (56K): In this window In a new window Download PPT slide   Figure 2. TFIIA depletion broadly decreases transcription from Pol II promoters. Transcription from the indicated promoters was analyzed in wild-type (wt) and GAL1-toa1-ts (ts) strains grown for 8 hr in permissive (P) or double-shut-off (NP) conditions as follows: (A) S1 analysis and (B) Northern blotting.

Transcriptional activation by Ace1, Hsf, and Gal4-VP16 in the absence of TFIIA:
If TFIIA is required for transcriptional activation, cells lacking TFIIA should be unable to induce high levels of transcription in response to activator proteins. In considering this question, we were concerned about the possibility that TFIIA at highly active promoters might be sufficiently stabilized so as to be relatively immune to the depletion methods. To circumvent this potential problem, most of the experiments were performed by first depleting cells of Toa1 and then inducing the function of a particular activator. In the case of the Ace1 activator, TFIIA-depleted cells were treated with copper, which is required for folding of the DNA-binding domain and subsequent activation. It should be noted that the uninduced level of CUP1 transcription in Toa1-depleted cells is elevated relative to that in control cells, presumably reflecting the contribution of stress response elements in the CUP1 promoter (TAMAI et al. 1994 ). When normalized to uninduced levels of transcription, Ace1-dependent activation of CUP1 expression is unaffected in the absence of TFIIA (Figure 3A). This phenomenon is not restricted to the CUP1 promoter context, as TFIIA depletion does not alter the level of activated transcription of an Ace1-dependent his3 allele (Figure 3B).

View larger version (54K): In this window In a new window Download PPT slide   Figure 3. Ace1-dependent activation is unimpaired under conditions of TFIIA depletion. GAL1-toa1-ts cells were grown under permissive (galactose medium at 30°) or nonpermissive (glucose medium at 37°) conditions for 8 hr, whereupon the cells were subjected to copper induction (concentrations indicated) for 1 hr. Transcription was analyzed from (A) the native CUP1 promoter and from (B) an artificial promoter containing Ace1-binding sites upstream of the his3 core promoter and structural gene. Quantitation relative to tRNAW was performed by PhosphorImage analysis.

To extend this analysis to transcriptional activation by heat shock factor, we could not employ the double-shut-off method because the temperature shift induces the function of the Hsf activation domain. Thus, cells were depleted of TFIIA by glucose repression alone for 12 hr, a time at which Toa1 protein was reduced to undetectable levels (Figure 4A). When such TFIIA-depleted cells were subjected to a brief heat shock, Hsf-dependent activation of SSA4 and HSP104 was indistinguishable from that observed in cells grown in permissive conditions (Figure 4B). Thus, de novo activation by Hsf appears unaffected in cells previously depleted of TFIIA.

View larger version (26K): In this window In a new window Download PPT slide   Figure 4. TFIIA depletion does not affect Hsf-dependent activation. GAL1-toa1-ts cells grown in galactose or glucose medium (single-shut-off conditions) for 12 hr at 30° were subjected (+) or not subjected (-) to a 15-min heat shock at 39°, and were then analyzed for (A) FLAG-Toa1 levels by Western blotting or (B) SSA4 and HSP104 transcription by S1 analysis. Quantitation relative to tRNAW was performed by PhosphorImage analysis.

As a direct comparison with biochemical studies showing that the VP16 activation domain enhances TFIID-TFIIA-TATA complex formation (KOBAYASHI et al. 1995 , KOBAYASHI et al. 1998 ), we examined activation of the GAL1 gene by Gal4-VP16 (Figure 5). Under double-shut-off conditions, activation of GAL1 by Gal4-VP16 is unimpaired. Incidentally, GAL1 transcription is not observed in TFIIA-depleted cells lacking Gal4-VP16, indicating that TFIIA is not required to maintain glucose repression of GAL1.

View larger version (32K): In this window In a new window Download PPT slide   Figure 5. Activation by Gal4-VP16 is unaffected in TFIIA-depleted cells. Wild-type (wt) and GAL1-toa1-ts (ts) cells that do (+) or do not (-) contain a plasmid expressing Gal4-VP16 were grown under double-shut-off conditions and analyzed for transcription of the native GAL1 promoter by S1 analysis. Quantitation relative to tRNAW was performed by PhosphorImage analysis.

Taken together, these results indicate that TFIIA is not required or limiting for the function of the Ace1, Hsf, and Gal4-VP16 activators in vivo. Furthermore, in contrast to the vast majority of natural promoters whose function was reduced two- to fourfold upon loss of TFIIA (Figure 2), promoters responsive to these strong activators were essentially unaffected by TFIIA depletion. This observation suggests that strong activators such as Ace1, Hsf, and Gal4-VP16 can override the general limitation on transcription imposed under conditions lacking TFIIA.

Although promoters dependent on strong activators such as Ace1, HSF, and Gal4-VP16 appear unaffected by the loss of TFIIA, it is possible that promoters responsive to weaker activators might resemble typical yeast promoters and, hence, be sensitive to changes in TFIIA levels. We therefore analyzed a his3 promoter derivative whose function completely depends on a binding site for the weak activator Ppr1 (IYER and STRUHL 1995 ). In TFIIA-depleted cells, Ppr1-dependent activation of his3 transcription was reduced by a factor of three in comparison to control cells (Figure 6). As Ace1-dependent activation of the identical his3 core promoter is unaffected (Figure 3B), this result suggests that the importance of TFIIA can depend on the activator. However, these results do not distinguish whether TFIIA depletion affects the function of the Ppr1 activator per se, or whether it affects an aspect of core promoter function that can be overridden by Ace1, but not Ppr1.

View larger version (47K): In this window In a new window Download PPT slide   Figure 6. Ppr1-dependent transcription of an artificial his3 promoter is reduced under conditions of TFIIA depletion. Wild-type (wt) and GAL1-toa1-ts (ts) cells with a modified his3 promoter containing a single Ppr1-binding site upstream of the his3 TATA and initiator region were grown in synthetic minimal medium containing 0.6% casamino acids and lacking uracil under permissive and double-shut-off conditions. Under these conditions, his3 transcription (assayed by S1 analysis) depends almost entirely on the activity of Ppr1.

TFIIA-depleted cells arrest at the G2-M cell-cycle boundary:
The morphology of TFIIA-depleted cells was examined by fluorescence microscopy with DAPI staining. A significant percentage (40–50%) of the TFIIA-depleted cells appear as large-budded cells with DNA concentrated at the septum, indicating arrest at or near the G2-M cell-cycle boundary (Figure 7). Under similar conditions, <10% of wild-type cells appear with large buds. This observation is consistent with and extends a previous observation that viable Toa2 mutants deficient in TBP interaction show an increased number of cells with a large-budded phenotype (OZER et al. 1998A ). In apparent contrast to this previous report, we did not observe cell clumping in our TFIIA-depleted cells; we do not know if this apparent conflict reflects differences in the TFIIA derivatives or in the yeast strain background. However, the specificity of the cell-cycle phenotype and the selective transcriptional effects conferred by the viable Toa2 mutants (OZER et al. 1998A ) indicate that TFIIA is particularly important for a subset of promoters.

View larger version (15K): In this window In a new window Download PPT slide   Figure 7. TFIIA-depleted cells arrest at the G2-M cell-cycle transition. Fluorescence microscopy was performed on wild-type and GAL1-toa1-ts cells grown under double-shut-off conditions for 8 hr, which were then fixed and stained with DAPI.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Transcriptional activation can occur in the absence of TFIIA:
In vitro, TFIIA can interact directly with activation domains (OZER et al. 1994 ; KOBAYASHI et al. 1995 ; CLEMENS et al. 1996 ), and it is required for activator-dependent recruitment and isomerization of the TFIID-TATA complex (LIEBERMAN and BERK 1994 ; KOBAYASHI et al. 1995 ; CHI and CAREY 1996 ; DAMANIA et al. 1998 ). Furthermore, the ability of the VP16 activation domain to interact with TFIIA and to stimulate assembly of the TFIID-TFIIA-TATA complex is strongly correlated with the level of transcriptional activation in vitro (KOBAYASHI et al. 1998 ). Here, we show that cells containing <1% wild-type levels of TFIIA are essentially unaffected for transcriptional activation by Ace1, Hsf, and Gal4-VP16. Moreover, these activators appear to override the two- to fourfold decrease in transcription in TFIIA-depleted cells that is typical of the majority of natural yeast promoters (see below). Thus, in apparent contrast to many biochemical results, TFIIA is not required for transcriptional activation by several activator proteins in vivo.

It is generally believed that activators stimulate transcription by directly interacting with components of the Pol II machinery and/or by recruiting chromatin-modifying activities to the promoter. Thus, activators that do not require TFIIA for transcriptional stimulation must be able to function via other targets. Some activators simply might not interact with TFIIA or affect the TFIID-TFIIA-TATA complex; hence, they would function exclusively through other targets. For activators that do interact with TFIIA and affect the TFIID-TFIIA-TATA complex (e.g., VP16), these in vitro interactions might not occur under physiological conditions or they might be too weak to significantly contribute to transcriptional output in vivo. Alternatively, activator-TFIIA interactions might be physiologically significant but functionally redundant with activator interactions with other components of the Pol II machinery and/or chromatin-modifying complexes. Strong activators, such as Ace1, Hsf, and Gal4-VP16, might interact functionally with multiple targets in vivo such that the loss of any one target (e.g., TFIIA) would not have significant transcriptional consequences. Of course, these considerations do not exclude the possibility that some activators require TFIIA for activation in vivo.

TFIIA is broadly but not absolutely required for transcription in yeast:
TFIIA-depleted cells show a two- to fourfold decrease in the expression of 9 of 10 natural yeast genes examined, suggesting that TFIIA has an important function at most yeast promoters. However, TFIIA is not absolutely required for transcription in vivo because some promoters (e.g., ADH1 and those driven by strong activators) appear unaffected by the loss of TFIIA. Moreover, TFIIA does not behave as a general transcription factor because, unlike TBP (CORMACK and STRUHL 1992 ) or TFIIB (MOQTADERI et al. 1996 ), depletion does not abolish gene expression. In this regard, our results appear inconsistent with transcriptionally inactive mutants of human TBP that alter the TFIIA-interaction surface and eliminate the interaction with TFIIA (BRYANT et al. 1996 ). We suspect that the human TBP mutants also affect the interaction with proteins other than TFIIA, particularly because of the radical nature of the amino acid substitutions.

As yeast promoters differ considerably in their upstream sequences, we suggest that the broad role of TFIIA in transcription in vivo most likely involves the core promoter region. The biochemical properties of TFIIA suggest two models relating to TBP and TATA elements. In one model, TFIIA directly stabilizes the association of TBP at promoters by virtue of its ability to form TBP-TFIIA-TATA complexes. In the alternative model, TFIIA indirectly affects TBP association with promoters by blocking general negative regulators of TBP function, such as Mot1 (AUBLE et al. 1994 ), NC2 (GADBOIS et al. 1997 ; KIM et al. 1997 ), the N-terminal domain of TAF130 (KOKUBO et al. 1998 ; OZER et al. 1998B ), and perhaps the Not complex (COLLART and STRUHL 1994 ). We slightly prefer the latter model because negative regulators of TBP might have an effect similar to simply reducing the level of active TBP; this could account for why most promoters are affected in a quantitatively similar manner. On the other hand, the variability in the quality of TATA elements suggests that stabilization of the TBP-TATA complex by TFIIA might vary considerably depending on the promoter. These two models are not mutually exclusive and, in fact, may be functionally interrelated.

Although we favor the view that the broad role of TFIIA reflects a core promoter function, we cannot exclude the possibility that the general decrease in Pol II transcription is related to the response to activator proteins. In vivo, transcription requires activators because intact promoters are generally much more active than core promoter derivatives containing only the TATA and initiator elements. Thus, if TFIIA is a near-universal but not absolutely required target of activators, depletion of TFIIA could cause a wide reduction in Pol II transcription. In this regard, Srb4, a Pol II holoenzyme component, may be a direct activator target (KOH et al. 1998 ) even though it is generally required for transcription (THOMPSON and YOUNG 1995 ).

Promoter specificity and functional similarities between TFIIA and TAFs:
Although TFIIA broadly affects transcription, several lines of evidence suggest that TFIIA does not play an equivalent role in all genes. First, it seems unlikely that a nonspecific and quantitatively modest decrease in Pol II transcription, such as would occur in cells containing a partially defective Pol II subunit, would prevent cell growth. By this argument, the general two- to fourfold decrease in Pol II transcription in TFIIA-depleted cells would not explain why TFIIA is essential for cell viability. Second, TFIIA depletion does not affect transcription of ADH1 or certain genes responsive to strong activators. Third, Toa1 and Toa2 mutations that partially inhibit TFIIA function selectively affect gene expression (MADISON and WINSTON 1997 ; OZER et al. 1998A ). Fourth, depletion of TFIIA causes cells to arrest at a specific point in the cell cycle, the G2-M transition, suggesting that TFIIA is particularly important for transcription of some cell-cycle genes.

In several respects, the physiological role of TFIIA is reminiscent of that of the TAF subunits in the TFIID complex. As is the case for TFIIA, depletion or functional inactivation of such TAFs does not generally affect the response to transcriptional activators (MOQTADERI et al. 1996 ; WALKER et al. 1996 ), but rather reduces transcription of selected genes (MOQTADERI et al. 1996 ; SHEN and GREEN 1997 ; WALKER et al. 1997 ). Moreover, TAF130 exerts its effects primarily at the level of the core promoter (MOQTADERI et al. 1996 ; SHEN and GREEN 1997 ), and this also appears to be the case for other TAFs that are found in TFIID, but not in the SAGA complex (MOQTADERI et al. 1998 ). Finally, individual TAFs are important for progression through specific stages of the cell cycle (APONE et al. 1996 ; WALKER et al. 1997 ), and loss of TFIIA or TAF90 blocks cell growth at the G2-M transition. These broad similarities are not surprising given that TFIIA and TAFs interact with TBP and promoter DNA; in this sense, TFIIA is a TAF-like component of the Pol II machinery that associates with but is not stably part of the TFIID complex. However, the transcriptional phenotypes of TFIIA- and TAF-depleted cells are clearly different, indicating that TFIIA and TAFs perform distinct physiological functions.

	FOOTNOTES

¹ These authors contributed equally to this work.
² Present address: Howard Hughes Medical Institute, Stanford University Medical School, Stanford, CA 94305.

	ACKNOWLEDGMENTS

We thank Sanjoy Mahanty for assistance with fluorescence microscopy, and Aleida Leza and Seth Sadis for DNAs. This work was supported by grants to K.S. from the National Institutes of Health (GM-30186) and from the Lucille Markey Trust.

Manuscript received March 23, 1999; Accepted for publication August 11, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

APONE, L. M., C. A. VIRBASIUS, J. C. REESE, and M. R. GREEN, 1996  Yeast TAF_II90 is required for cell-cycle progression through G₂/M but not for general transcription activation. Genes Dev. 10:2368-2380[Abstract/Free Full Text].

AUBLE, D. T., K. E. HANSEN, C. G. F. MUELLER, W. S. LANE, and J. THORNER et al., 1994  Mot1, a global repressor of RNA polymerase II transcription, inhibits TBP binding to DNA by an ATP-dependent mechanism. Genes Dev. 8:1920-1934[Abstract/Free Full Text].

BRYANT, G. O., L. S. MARTEL, S. K. BURLEY, and A. J. BERK, 1996  Radical mutations reveal TATA-box binding protein surfaces required for activated transcription in vivo.. Genes Dev. 10:2491-2504[Abstract/Free Full Text].

CHEN, W. and K. STRUHL, 1988  Saturation mutagenesis of a yeast his3 TATA element: genetic evidence for a specific TATA-binding protein. Proc. Natl. Acad. Sci. USA 85:2691-2695[Abstract/Free Full Text].

CHI, T. and M. CAREY, 1996  Assembly of the isomerized TFIIA-TFIID-TATA ternary complex is necessary and sufficient for gene activation. Genes Dev. 10:2540-2550[Abstract/Free Full Text].

CHICCA, J. J. N., D. T. AUBLE, and B. F. PUGH, 1998  Cloning and biochemical characterization of TAF-172, a human homolog of yeast Mot1. Mol. Cell. Biol. 18:1701-1710[Abstract/Free Full Text].

CISMOWSKI, M., G. LAFF, M. SOLOMAN, and S. 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[Abstract].

CLEMENS, K. E., G. PIRAS, M. F. RADONOVICH, K. S. CHOI, and J. F. DUVALL et al., 1996  Interaction of the human T-cell lymphotropic virus type 1 tax transactivator with transcription factor IIA. Mol. Cell. Biol. 16:4656-4664[Abstract].

COLLART, M. A. and K. STRUHL, 1994  NOT1(CDC39), NOT2(CDC36), NOT3, and NOT4 encode a global negative regulator of transcription that differentially affects TATA-element utilization. Genes Dev. 8:525-537[Abstract/Free Full Text].

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[Medline].

DAMANIA, B., P. LIEBERMAN, and J. C. ALWINE, 1998  Simian virus 40 large T antigen stabilizes the TATA-binding protein-TFIIA complex on the TATA element. Mol. Cell. Biol. 18:3926-3935[Abstract/Free Full Text].

EMAMI, K. H., A. JAIN, and S. T. SMALE, 1997  Mechanism of synergy between TATA and initiator: synergistic binding of TFIID following a putative TFIIA-induced isomerization. Genes Dev. 11:3007-3019[Abstract/Free Full Text].

GADBOIS, E. L., D. M. CHAO, J. C. REESE, M. R. GREEN, and R. A. YOUNG, 1997  Functional antagonism between RNA polymerase II holoenzyme and global negative regulator NC2 in vivo.. Proc. Natl. Acad. Sci. USA 94:3145-3150[Abstract/Free Full Text].

GE, H. and R. G. ROEDER, 1994  The high mobility group protein HMG1 can reversibly inhibit class II gene transcription by interaction with TATA-binding protein. J. Biol. Chem. 269:17136-17140[Abstract/Free Full Text].

GEIGER, J. H., S. HAHN, S. LEE, and P. B. SIGLER, 1996  Crystal structure of the yeast TFIIA/TBP/DNA complex. Science 272:830-836[Abstract].

GOPPELT, A., G. STELZER, F. LOTTSPEICH, and M. MEISTERENST, 1996  A mechanism for repression of class II gene transcription through specific binding of NC2 to TBP-promoter complexes via heterodimeric histone fold domains. EMBO J. 15:3105-3116[Medline].

IMBALZANO, A. N., K. S. ZARET, and R. E. KINGSTON, 1994  Transcription factor (TF) IIB and TFIIA can independently increase the affinity of the TATA-binding protein for DNA. J. Biol. Chem. 269:8280-8286[Abstract/Free Full Text].

IYER, V. and K. STRUHL, 1995  Mechanism of differential utilization of the his3 T_R and T_C TATA elements. Mol. Cell. Biol. 15:7059-7066[Abstract].

IYER, V. and K. STRUHL, 1996  Absolute mRNA levels and transcriptional initiation rates in Saccharomyces cerevisiae.. Proc. Natl. Acad. Sci. USA 93:5208-5212[Abstract/Free Full Text].

KANG, J. J., D. T. AUBLE, J. A. RANISH, and S. HAHN, 1995  Analysis of yeast transcription factor TFIIA: distinct functional regions and a polymerase II-specific role in basal and activated transcription. Mol. Cell. Biol. 15:1234-1243[Abstract].

KIM, S., J. G. NA, M. HAMPSEY, and D. REINBERG, 1997  The Dr1/DRAP1 heterodimer is a global repressor of transcription in vivo. Proc. Natl. Acad. Sci. USA 94:820-825[Abstract/Free Full Text].

KOBAYASHI, N., T. G. BOYER, and A. J. BERK, 1995  A class of activation domains interacts directly with TFIIA and stimulates TFIIA-TFIID-promoter complex assembly. Mol. Cell. Biol. 15:6465-6473[Abstract].

KOBAYASHI, N., P. J. HORN, S. M. SULLIVAN, S. J. TRIEZENBERG, and T. G. BOYER et al., 1998  DA-complex assembly activity required for VP16C transcriptional activation. Mol. Cell. Biol. 18:4023-4031[Abstract/Free Full Text].

KOH, S. S., A. Z. ANSARI, M. PTASHNE, and R. A. YOUNG, 1998  An activator target in the RNA polymerase II holoenzyme. Mol. Cell 1:895-904[Medline].

KOKUBO, T., M. J. SWANSON, J. I. NISHIKAWA, A. G. HINNEBUSCH, and Y. NAKATANI, 1998  The yeast TAF145 inhibitory domain and TFIIA competitively bind to TATA-binding protein. Mol. Cell. Biol. 18:1003-1012[Abstract/Free Full Text].

LAGRANGE, T., T.-K. KIM, G. ORPHANIDES, Y. W. EBRIGHT, and R. H. EBRIGHT et al., 1996  High-resolution mapping of nucleoprotein complexes by site-specific protein-DNA photocrosslinking: organization of the TBP-TFIIA-TFIIB-DNA quaternary complex. Proc. Natl. Acad. Sci. USA 93:10620-10625[Abstract/Free Full Text].

LEE, M. and K. STRUHL, 1995  Mutations on the DNA-binding surface of TBP can specifically impair the response to acidic activators in vivo.. Mol. Cell. Biol. 15:5461-5469[Abstract].

LEE, M. and K. STRUHL, 1997  A severely defective TATA-binding protein-TFIIB interaction does not preclude transcriptional activation in vivo.. Mol. Cell. Biol. 17:1336-1345[Abstract].

LIEBERMAN, P. M. and A. J. BERK, 1994  A mechanism for TAFs in transcriptional activation: activation domain enhancement of TFIID-TFIIA-promoter DNA complex formation. Genes Dev. 8:995-1006[Abstract/Free Full Text].

MA, D., I. OLAVE, A. MERINO, and D. REINBERG, 1996  Separation of the transcriptional coactivator and antirepression functions of transcription factor IIA. Proc. Natl. Acad. Sci. USA 93:6583-6588[Abstract/Free Full Text].

MADISON, J. M. and F. WINSTON, 1997  Evidence that Spt3 functionally interacts with Mot1, TFIIA, and TATA-binding protein to confer promoter-specific transcriptional control in Saccharomyces cerevisiae.. Mol. Cell. Biol. 17:287-295[Abstract].

MERMELSTEIN, F., K. YEUNG, J. CAO, J. A. INOSTROZA, and H. ERDJUMENT-BROMAGE et al., 1996  Requirement of a corepressor for Dr1-mediated repression of transcription. Genes Dev. 10:1033-1048[Abstract/Free Full Text].

MOQTADERI, Z., Y. BAI, D. POON, P. A. WEIL, and K. STRUHL, 1996  TBP-associated factors are not generally required for transcriptional activation in yeast. Nature 382:188-191.

MOQTADERI, Z., M. KEAVENEY, and K. STRUHL, 1998  The histone H3-like TAF is broadly required for transcription in yeast. Mol. Cell 2:675-682[Medline].

NONET, M., C. SCAFE, J. SEXTON, and R. YOUNG, 1987  Eucaryotic RNA polymerase conditional mutant that rapidly ceases mRNA synthesis. Mol. Cell. Biol. 7:1602-1611[Abstract/Free Full Text].

OELGESCHLAGER, T., C. M. CHIANG, and R. G. ROEDER, 1996  Topology and reorganization of a human TFIID-promoter complex. Nature 382:735-738[Medline].

ORPHANIDES, G., T. LAGRANGE, and D. REINBERG, 1996  The general initiation factors of RNA polymerase II. Genes Dev. 10:2657-2683[Free Full Text].

OZER, J., P. A. MOORE, A. H. BOLDEN, A. LEE, and C. A. ROSEN et al., 1994  Molecular cloning of the small () subunit of human TFIIA reveals functions critical for activated transcription. Genes Dev. 8:2324-2335[Abstract/Free Full Text].

OZER, J., L. E. LEZINA, J. EWING, S. AUDI, and P. M. LIEBERMAN, 1998a  Association of transcription factor IIA with TATA-binding protein is required for transcriptional activation of a subset of promoters and cell cycle progression in Saccharomyces cerevisiae.. Mol. Cell. Biol. 18:2559-2570[Abstract/Free Full Text].

OZER, J., K. MITSOURAS, D. ZERBY, M. CAREY, and P. M. LIEBERMAN, 1998b  Transcription factor IIA derepresses TATA-binding protein (TBP)-associated factor inhibition of TBP-DNA binding. J. Biol. Chem. 273:14293-14300[Abstract/Free Full Text].

ROEDER, R. G., 1996  The role of general initiation factors in transcription by RNA polymerase II. Trends Biochem. Sci. 21:327-335[Medline].

SHEN, W. C. and M. R. GREEN, 1997  Yeast TAF_II145 functions as a core promoter selectivity factor, not a general coactivator. Cell 90:615-624[Medline].

SHYKIND, B. M., J. KIM, and P. A. SHARP, 1995  Activation of the TFIID-TFIIA complex with HMG-2. Genes Dev. 9:1354-1365[Abstract/Free Full Text].

STARGELL, L. A. and K. STRUHL, 1995  The TBP-TFIIA interaction in the response to acidic activators in vivo.. Science 269:75-78[Abstract/Free Full Text].

SUN, X., D. MA, M. SHELDON, K. YEUNG, and D. REINBERG, 1994  Reconstitution of human TFIIA activity from recombinant polypeptides: a role in TFIID-mediated transcription. Genes Dev. 8:2336-2348[Abstract/Free Full Text].

TAMAI, K. T., X. LIU, P. SILAR, T. SOSINOWSKI, and D. J. THIELE, 1994  Heat shock transcription factor activates yeast metallothionein gene expression in response to heat and glucose starvation via distinct signalling pathways. Mol. Cell. Biol. 14:8155-8165[Abstract/Free Full Text].

TAN, S., Y. HUNZIKER, D. F. SARGENT, and T. J. RICHMOND, 1996  Crystal structure of a yeast TFIIA/TBP/DNA complex. Nature 381:127-134[Medline].

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[Abstract/Free Full Text].

TRUEHEART, J., J. D. BOEKE, and G. R. FINK, 1987  Two genes required for cell fusion during yeast conjugation: evidence for a pheromone-induced surface protein. Mol. Cell. Biol. 7:2316-2328[Abstract/Free Full Text].

VALAY, J.-G., M. SIMON, M.-F. DUBOIS, O. BENSAUDE, and C. FACCA et al., 1995  The KIN28 gene is required for RNA polymerase II-mediated transcription and phosphorylation of the Rpb1p CTD. J. Mol. Biol. 249:535-544[Medline].

WALKER, S. S., J. C. REESE, L. M. APONE, and M. R. GREEN, 1996  Transcription activation in cells lacking TAF_IIs. Nature 382:185-188.

WALKER, S. S., W. C. SHEN, J. C. REESE, L. M. APONE, and M. R. GREEN, 1997  Yeast TAF_II145 required for transcription of G1/S cyclin genes and regulated by cellular growth state. Cell 90:607-614[Medline].

YOKOMORI, K., M. P. ZEIDLER, J.-L. CHEN, C. P. VERRIJZER, and M. MLODZIK et al., 1994  Drosophila TFIIA directs cooperative DNA binding with TBP and mediates transcriptional activation. Genes Dev. 8:2313-2323[Abstract/Free Full Text].

This article has been cited by other articles:

				T. Y. Erkina and A. M. Erkine Displacement of Histones at Promoters of Saccharomyces cerevisiae Heat Shock Genes Is Differentially Associated with Histone H3 Acetylation. Mol. Cell. Biol., October 1, 2006; 26(20): 7587 - 7600. [Abstract] [Full Text] [PDF]

				H. Singh, A. M. Erkine, S. B. Kremer, H. M. Duttweiler, D. A. Davis, J. Iqbal, R. R. Gross, and D. S. Gross A Functional Module of Yeast Mediator That Governs the Dynamic Range of Heat-Shock Gene Expression Genetics, April 1, 2006; 172(4): 2169 - 2184. [Abstract] [Full Text] [PDF]

				J. Zhao, J. Herrera-Diaz, and D. S. Gross Domain-Wide Displacement of Histones by Activated Heat Shock Factor Occurs Independently of Swi/Snf and Is Not Correlated with RNA Polymerase II Density Mol. Cell. Biol., October 15, 2005; 25(20): 8985 - 8999. [Abstract] [Full Text] [PDF]

				S. Gilfillan, G. Stelzer, E. Piaia, M. G. Hofmann, and M. Meisterernst Efficient Binding of NC2{middle dot}TATA-binding Protein to DNA in the Absence of TATA J. Biol. Chem., February 18, 2005; 280(7): 6222 - 6230. [Abstract] [Full Text] [PDF]

				A. M. Erkine and D. S. Gross Dynamic Chromatin Alterations Triggered by Natural and Synthetic Activation Domains J. Biol. Chem., February 28, 2003; 278(10): 7755 - 7764. [Abstract] [Full Text] [PDF]

				S. R. Bhaumik and M. R. Green Differential Requirement of SAGA Components for Recruitment of TATA-Box-Binding Protein to Promoters In Vivo Mol. Cell. Biol., November 1, 2002; 22(21): 7365 - 7371. [Abstract] [Full Text] [PDF]

				S. Solow, M. Salunek, R. Ryan, and P. M. Lieberman TAFII 250 Phosphorylates Human Transcription Factor IIA on Serine Residues Important for TBP Binding and Transcription Activity J. Biol. Chem., May 4, 2001; 276(19): 15886 - 15892. [Abstract] [Full Text] [PDF]