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

J. Biol. Chem., Vol. 275, Issue 41, 31914-31920, October 13, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/31914    most recent M003760200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Xie, Y. Articles by Kodadek, T. Search for Related Content PubMed PubMed Citation Articles by Xie, Y. Articles by Kodadek, T. Social Bookmarking                      What's this?

Biochemical Characterization of the TATA-binding Protein-Gal4 Activation Domain Complex^*

Yueqing Xie, Carilee Denison, Sang-Hwa Yang, David A. Fancy, and Thomas Kodadek^§

From the Departments of Internal Medicine and Biochemistry, Ryburn Center for Molecular Cardiology, University of Texas Southwestern Medical Center, Dallas, Texas 75390-8573

Received for publication, May 3, 2000, and in revised form, August 4, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

It has been suggested that complexes between gene-specific activators and the TATA-binding protein (TBP) play an important role in the expression of many genes. However, few detailed studies of well defined activator-TBP complexes have been reported. An analysis of the biochemical properties of the complex formed by the acidic activation domain (AAD) of the yeast activator Gal4 and TBP is presented here. This is shown to be composed of two AAD and one TBP molecule. DNA binding experiments reveal that TATA-containing DNAs and the Gal4 AAD bind TBP competitively, suggesting that the AAD and TATA boxes recognize overlapping surfaces of TBP. The kinetics of the formation and dissociation of the AAD₂-TBP complex is also probed. The impact of these findings on models for Gal4-mediated transcriptional activation is considered.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

The TATA-binding protein (TBP)¹ is a critical factor in RNA polymerase II-mediated transcription in eukaryotic cells. TBP recognizes the TATA box present in many promoters and binds in the minor groove, bending and kinking the DNA sharply into an unusual geometry that is probably important for assembly of the complete preinitiation complex (for a review, see Ref. 1). Cross-linking studies in yeast have shown that TBP associates with the promoters of many genes only when the activators that regulate them are induced (2, 3), suggesting that many activators stimulate TBP-promoter binding directly or indirectly. Consistent with a direct role for activator-TBP contacts in transcriptional activation are the large number of reports of activation domain (AD)-TBP contacts (4-10). However, there has been very little detailed biochemical analysis of highly purified, well defined AD-TBP complexes. The type of information obtained in such studies is important for evaluating various mechanistic models for how activators might affect TBP function. In this study, we probe the biochemistry of a complex formed by the acidic activation domain (AAD) of the Gal4 protein (11, 12), a potent yeast activator, and TBP. The ability of mutant Gal4 AADs to function in vivo correlates well with their ability to bind TBP in vitro (6, 7), supporting, but not proving, the idea that this complex is biologically relevant. It is demonstrated here that this complex is composed of 2 eq of the AAD and 1 eq of TBP. Competitive binding studies show that the Gal4 AAD and TATA box-containing DNA compete for limiting TBP, suggesting that the AAD and DNA bind overlapping sites on TBP. Finally, the association and dissociation kinetics of the AAD₂-TBP complex are probed. The implications of these results for models of Gal4-activated transcription are considered.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

Proteins-- Recombinant wild-type yeast TBP (13), the C-terminal 180-residue fragment of yeast TBP (TBP180) (14), GST-Gal4 AAD (fusion of the Gal4 activation domain (residues 841-875) to glutathione S-transferase)(GST) (6), and Gal4p-(1-93 + 768-881) (15) were purified from overexpressing Escherichia coli strains as described.

For the determination of the stoichiometry of the Gal4 AD-TBP complex, a derivative of GST-Gal4 AAD that also included a C-terminal His₆ tag was constructed, expressed, and purified. This was done to obtain very pure full-length protein by double affinity purification over sequential glutathione-Sepharose and nickel-saturated chelating-Sepharose columns. This protocol eliminated proteolysis products completely, which otherwise constituted a significant portion of the preparation. The expression plasmid was constructed as follows. The activation domain of Gal4p was polymerase chain reaction-amplified using the following primers: 5'-CCG GGA TCC TTA GTA GTG GTG ATG GTG ATG GTG CGA TCC TCT CAT GGT ATC TTC ATC ATC GAA TAG-3' and 5'-GGT TGG ACG CCA TGG ACG ACC AAA CTG CGT ATA ACG CG-3' and pGEXcs34 (16) as the template. The polymerase chain reaction product was digested with NcoI and BamHI and cloned into the plasmid pGEXcs digested with NcoI and BamHI. This provided the expression plasmid pGEXcs34(AD)-His₆.

To purify GST-Gal4 AAD-His₆, the plasmid pGEXcs34(AD)-His₆ was transformed into the E. coli strain BL21(DE3)pLysS. The fresh transformant was grown overnight at 37 °C in 10 ml of Luria broth containing ampicillin and chloramphenicol (75 and 25 mg/liter, respectively) and then transferred to 1 liter of Luria broth containing ampicillin and chloramphenicol and grown to an A₆₀₀ of 0.5. Expression was induced by adding isopropyl-1-thio--D-galactopyranoside to a final concentration of 1 mM. The culture was grown at 37 °C for another 2 h, after which the cells were collected by low speed centrifugation and resuspended in 1× PBS (20 mM sodium phosphate, 150 mM NaCl, pH 7.3) in the presence of protease inhibitors phenylmethylsufonyl fluoride, pepstatin A, and leupeptin (final concentration of 1 mM, 1 mg/liter, and 1 mg/liter, respectively). Cells were lysed by sonication and centrifuged at 16,000 × g for 45 min. The cleared lysate was loaded on glutathione-Sepharose (Amersham Pharmacia Biotech) pre-equilibrated with 1× PBS and washed with PBS and eluted with 30 mM glutathione. The eluate was diluted 10-fold with 1× binding buffer and further purified on chelating-Sepharose (Amersham Pharmacia Biotech).

The concentrations of the GST-Gal4 AAD-His₆ and TBP180 stocks employed in the experiment to determine the stoichiometry of the complex (see below) were determined using amino acid analysis (performed on an ABI 420 amino acid analyzer).

Determination of the GST-Gal4 AAD-His₆/TBP Stoichiometry-- GST-Gal4 AAD-His₆ (1 µM) was mixed with glutathione-Sepharose beads, and the indicated amount of TBP180 (see Fig. 1) was added in a total volume of 400 µl in 1× HBS buffer (10 mM HEPES, pH 7.4, 150 mM NaCl, 15% glycerol). After incubation on a rotator at 4 °C for 30 min, the beads were washed three times with 1× HBS (with 300 mM NaCl) containing 0.1% Triton X-100, followed by washing once with 1× HBS (300 mM NaCl) without detergent. The beads were resuspended in SDS-PAGE loading buffer, heated at 95 °C for 3 min, and loaded onto the SDS-polyacrylamide gel. TBP180 and GST-Gal4 AAD-His₆ standards of known concentration were run on the same gel. The intensities of these bands were analyzed by video densitometry.

Fluorescence Polarization Assays-- A 14-mer double-stranded oligonucleotide containing a consensus TATA box (5'-GCT ATA AAA GGG CA-3') was 5'-labeled with fluorescein. 10 nM TBP, the indicated amount of GST-Gal4 AAD or GST (see Fig. 2), and 5 nM of the labeled TATA DNA were mixed in 500 µl of buffer (20 mM Tris acetate, pH 7.4, 4 mM magnesium acetate, 1 mM dithiothreitol, 0.1 mM EDTA, 5% glycerol, 150 mM potassium glutamate, and 100 µg/ml bovine serum albumin), and the solution was allowed to come to equilibrium by incubating at room temperature for 45 min. The sample was then placed into the cavity of a fluorescence spectrometer equipped to measure anisotropy (Panvera Beacon 2000), and the polarization of the emitted light was recorded.

Filter Binding Assays-- 20 nM TBP and 50 nM ³²P-labeled TATA-containing DNA (double-stranded oligonucleotides containing a consensus sequence (5'-GCT ATA AAA GGG CA-3')), or a "T6-substituted" sequence (5'-GCT ATA ATA GGG CA-3') were mixed in 50 µl of reaction buffer (20 mM Tris acetate, pH 7.4, 4 mM magnesium acetate, 1 mM dithiothreitol, 0.1 mM EDTA, 5% glycerol, 75 mM potassium glutamate, and 100 µg/ml bovine serum albumin) at room temperature. The GST-Gal4 AAD concentration was 2 µM if present. After the desired time of incubation, 140 µl of competitor solution (containing 1 µM TATA DNA (5'-GCT ATA AAA GGG CA-3') and 200 µg/ml sonicated salmon sperm DNA) was added to the reactions. This served to remove any TBP nonspecifically bound to the labeled DNA. The reaction mixtures were then loaded onto a nitrocellulose membrane (BA85, Schleicher & Schuell) in a BioDot apparatus (Bio-Rad). All wells were washed with 500 µl of reaction buffer. The dried membrane was scanned with a PhosphorImager (Molecular Dynamics) to quantitate the amount of labeled DNA retained. The data were analyzed with ImageQuant software.

Competitive Binding of Gal4p-(1-93 + 768-881) and DNA to TBP-- 40 nM Gal4-(1-93 + 768-881), 10 nM F-UAS21(5'-GAC GGA GGACTG TCC TCC GAG-3', end-labeled with fluorescein), and 0.1 mg/ml bovine serum albumin was mixed in GD buffer (20 mM HEPES, pH 7.5, 200 mM NaCl, 1 mM -mercaptoethanol, 20 µM ZnSO₄, 50 µM EDTA, and 10% glycerol) at room temperature for 15 min to allow complex formation. TBP and TATA DNA were mixed at a molar ratio of 1:3 at 4 °C for 30 min. This TBP-TATA mixture (in parallel experiments, either TBP alone or TATA DNA alone) was then mixed with the Gal4-UAS complex solution at the indicated concentration and incubated at room temperature for 5 min. The samples were placed into the cavity of a fluorescence spectrometer equipped to measure anisotropy (Panvera Beacon 2000), and the polarization of the emitted light was recorded.

Testing the Competition between GST-Gal4 AAD and DNA for TBP Binding Using Pull-down Assays-- Purified His₆TBP or yeast crude lysate was incubated with excess TATA-containing double-stranded DNA (5'-GGA ATT CGG GCT ATA AAA GGG GGA TCC G-3') in GTD buffer (20 mM HEPES, pH 7.5, 75 mM potassium acetate, 20 µM zinc sulfate, 4 mM magnesium chloride, 1 mM -mercaptoethanol, 50 µM EDTA, and 10% (w/v) glycerol) at 4 °C for 1 h. Then GST-Gal4 AAD-bound glutathione beads were added, and the reactions were further incubated at 4 °C for 45 min. The final concentrations of GST-Gal4 AAD, TATA DNA, purified TBP, and TBP in yeast lysate were 12 µM, 6 µM, 12 nM, and 6 nM, respectively. The beads were collected by centrifugation and washed with GTD buffer three times. The beads were then resuspended in SDS-PAGE loading buffer, heated at 95 °C for 10 min, and loaded onto a SDS-polyacrylamide gel. The amount of TBP was compared by Western blot.

Protein Cross-linking-- Chemical cross-linking experiments were carried out as described (17). The concentrations of GST-Gal4 AAD, GST, and TBP used in this experiment were 1.3, 1.3, and 0.16 µM, respectively.

Determination of the Association Rate of the GST-Gal4 AAD-TBP Complex-- TBP180 (0.3 µM) was added to a solution containing GST-Gal4 AAD (1 µM) immobilized on glutathione beads in a final volume of 400 µl in 1× HBS buffer (see above). After various time increments (0, 3, 10, 15, 20, and 30 min of incubation on a rotor at 4 °C) aliquots were removed, the beads were pelleted by centrifugation, and the supernatant was discarded. As described for the stoichiometry experiment, the beads were washed three times with 1× HBS buffer (300 mM NaCl) containing 0.1% Triton X-100 and then a final time with 1× HBS buffer that did not contain detergent. The resulting beads were then resuspended in 1× SDS-PAGE sample buffer and heated at 95 °C for 10 min. The degree of complex association as a function of time was assessed by separating the resulting protein mixtures with SDS-PAGE and then quantifying the appropriate bands with densitometry. As described under the stoichiometry experimental section, standards of known concentration of both GST-Gal4 AAD and TBP180 were also run on the gel to standardize the intensities of the bands for the pull-down lanes. An additional control lane represents a GST pull-down experiment (instead of GST-Gal4 AAD) to demonstrate that the binding of TBP is specific for the activation domain.

Determination of Gal4p-(1-93 + 768-881)-TBP Dissociation Rate by a Fluorescence Polarization Assay-- Reaction mixture containing 40 nM Gal4p-(1-93 + 768-881), 10 nM F-UAS21, and 720 nM TBP in GD buffer was incubated at room temperature for 60 min to allow triple complex formation. GST-Gal4 AAD was then added to a final concentration of 10 µM, and the sample was immediately placed into the fluorescence polarization analyzer and the polarization value was recorded at 30-s intervals for 18 min.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

The Gal4 AD and TBP Form a Complex of 2:1 Stoichiometry-- Gal4p binds to DNA as a dimer (18), and we have previously shown that a single Gal4p dimer bound stably to a promoter can maximally activate transcription in vivo (19). Thus, two AADs are sufficient to satisfy whatever requirement there is for activated transcription in yeast, at least for the GAL genes. This raises the interesting question of whether each of the two AADs in the dimer contact the same or different factors at any particular stage of the transcription cycle. Specifically, for the purposes of this study, when Gal4p binds TBP, are both AADs occupied in this interaction or is one free to bind a different transcription factor?

To determine the stoichiometry of the Gal4 AAD-TBP complex, a protein composed of GST fused to the 34-residue core region (residues 841-875) of the Gal4 AAD (GST-Gal4 AAD) and a C-terminal 6-histidine tag was titrated with the conserved carboxyl 180-residue fragment of yeast TBP (TBP180) (14). Each protein was purified to apparent homogeneity, and their concentrations were determined accurately by amino acid analysis. The experiment employed GST-Gal4 AAD at a concentration of 10⁶ M to ensure stoichiometric binding (the reported K_D of this complex is approximately 10⁷ M (6, 7)).

As shown in Fig. 1, binding of TBP180 to GST-Gal4 AAD saturated cleanly at a molar ratio of two GST-Gal4 AAD molecules for each TBP180. Addition of more than 0.5 eq did not increase the amount of TBP180 retained by the AAD, even at these high protein concentrations. Similar results were obtained with full-length TBP (data not shown). Although this experiment was conducted with an artificial AAD-containing fusion protein, it seems likely that GST-Gal4 AAD is a reasonable model for how the AAD is presented to TBP normally, because both GST and Gal4p are native dimers. This 2:1 stoichiometry argues that if DNA-bound Gal4p binds to TBP in vivo, both AADs are involved in the association.

View larger version (36K): [in this window] [in a new window]   Fig. 1.   The stoichiometry of the Gal4 acidic activation domain-TBP complex is 2:1. Bottom, Coomassie Blue-stained gel showing the results of the titration of bead-bound GST-Gal4 AAD with TBP180 (the conserved C-terminal piece of TBP). GST-34-His6 represents the fusion of the 34-residue core Gal4 AAD to GST on its N terminus and 6 histidines on its C terminus. Far left lane, mass standards. Lane 1, pelleted material from the addition of 20 µg of TBP180 to 10.8 µg of GST lacking the Gal4 AD. No TBP180 was retained. Lanes 2-7, pelleted material from the addition of different amounts of TBP180 (0.4, 0.8, 4, 8, 20 and 40 µg, respectively) to 10.8 µg of GST-Gal4 AAD. Top, graphical representation of the gel derived from scanning, densitometry and comparison to standards.

It is important to point out that these data cannot distinguish between a 2:1 and a 4:2 complex because both would have a 2:1 ratio of proteins. Because TBP can form a stable homodimer (20), the 4:2 species must be considered. Unfortunately efforts to resolve this issue using analytical ultracentrifugation or gel filtration, which require the complex to remain intact for long periods of time, failed to provide clear-cut results (data not shown). However, on the basis of cross-linking and other data presented below, it seems likely that the complex has only one molecule of TBP in it.

The Gal4 AAD and DNA Bind to TBP Competitively-- A common model for how an activator might stimulate TBP function is that it enhances TBP-TATA binding, for example, through cooperative binding to the promoter. The effect of the Gal4 AAD on TBP-TATA interactions has not been investigated previously, so it is unclear if the Gal4 AAD-TBP complex has properties consistent with this kind of model. Indeed, previous mutagenesis studies (6, 21) have indicated that the TBP residues most critical for Gal4 AAD binding are located on the DNA binding surface of the protein, suggesting that the AAD and DNA might compete for TBP.

To probe this point, a fluorescein-labeled oligonucleotide containing a consensus TATA sequence was mixed with TBP and increasing amounts of GST-Gal4 AAD (or GST as a control). TBP-DNA binding was monitored by fluorescence polarization (22). When the free fluorescein-labeled DNA is excited with polarized light, little of the emitted light retains the original polarization due to relatively rapid tumbling in solution. However, binding of the much larger TBP molecule reduces the tumbling rate of the DNA and results in a large increase in polarization of the emitted light. Thus, if the Gal4 AAD competes with DNA for limiting TBP, then the polarization would be expected to decrease as the AAD concentration increases. As shown in Fig. 2A, this is exactly the result observed. A semilog plot of the degree of polarization plotted against the GST-Gal4 AAD concentration is linear, consistent with a simple competition between GST-Gal4 AAD and the labeled DNA (Fig. 2B). However, no effect was observed when the TBP-TATA complex was challenged with GST alone (Fig. 2A). This experiment demonstrates that the Gal4 AAD and TATA-containing DNAs compete for TBP. Combined with the mutagenesis data (21), this argues that the AAD and DNA both bind to the convex underside of the TBP "saddle" (23), although we cannot absolutely rule out an allosteric competition model.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   The Gal4 activation domain competes with TATA DNA for binding to TBP. A, a complex composed of TBP bound to a fluorescein-labeled, TATA-containing oligonucleotide was titrated with increasing amounts of GST or GST-Gal4 AAD. The result was monitored by fluorescence polarization spectroscopy. The loss of polarization when the AAD was added is indicative of dissociation of the TBP-DNA complex. B, a semilog plot of the data shown in A. The linear nature of the plot is consistent with a simple, reversible competition between DNA and the Gal4 AAD for TBP.

Fig. 3 shows an experiment that corroborates the conclusion that the Gal4 AAD and DNA bind TBP competitively. In this case, TBP was preincubated with GST-Gal4 AAD (or GST as a control) to allow for complex formation, then a radiolabeled TATA-containing oligonucleotide was added, and the rate of TBP-DNA complex formation was measured by nitrocellulose filter binding. Experiments were carried out with either a consensus TATA sequence (TATAAAA) or one that binds TBP with an approximately 2-fold lower affinity (TATAATA). In each case, the presence of GST-Gal4 AAD inhibited the rate of association of TBP with the TATA-containing oligonucleotides.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   The Gal4 AAD reduces the rate of association of TBP with DNA. Nitrocellulose filter binding was employed to monitor the rate of association of TBP with TATA-containing, radiolabeled oligonucleotides with or without preincubation of the TBP with a 40-fold excess of GST-Gal4 AAD. MLP represents adenovirus major late promoter TATA box, which includes the consensus TATA sequence 5'-TATAAAA. T6 represents a substitution of T at position 6 of the 7-nucleotide TATA sequence.

This competition between the Gal4 AAD and TATA DNA for TBP binding was also found in experiments using a different derivative of the activator, Gal4-(1-93 + 768-881), that contains a much larger fragment of the native C terminus as well as the N-terminal DNA-binding domain. It binds to TBP in vitro with a K_D of 200 nM.² When this Gal4 derivative was mixed with a fluorescein end-labeled 21-base pair oligonucleotide containing a consensus Gal-UAS, an increased polarization value seen relative to free DNA was observed, as expected (not shown). Titration of the Gal4 derivative-UAS complex with TBP resulted in a further increase in the fluorescence polarization, reflecting binding of TBP to the AAD (TBP alone did not bind to the UAS DNA at such concentrations, data not shown). It should be noted that although the change in polarization upon TBP addition is modest, this is expected. Because of the lifetime of the fluorescein excited state, the magnitude of changes in the polarization value become very small as the mass of the molecule approaches about 100 kDa and the mass of the Gal4 derivative-DNA complex is already 60 kDa. In any case, the data presented were highly reproducible, and careful measurements do allow this technique to be employed in the 60-100-kDa molecular mass range. In contrast to adding TBP alone to the Gal4 derivative-DNA complex, if the TBP was first saturated with a TATA box-containing oligonucleotide prior to addition to the Gal4 derivative-UAS complex, no such increase was detected (Fig. 4). This indicates that TBP-Gal4p-(1-93 + 768-881) association was blocked by the TATA-containing DNA. We conclude that both the core 34-residue AAD and the much larger C-terminal 768-881 fragment compete with DNA for TBP and that this result is likely to reflect the behavior of native Gal4p

View larger version (22K): [in this window] [in a new window]   Fig. 4.   TATA DNA competes with Gal4p-(1-93 + 768-881) for TBP binding. A complex was formed between Gal4-(1-93 + 768-881) and a fluorescein-labeled 21-base pair oligonucleotide containing the consensus Gal4p binding site. The complex was then incubated with TBP, a TATA-containing oligonucleotide, or the preformed complex between TBP and the TATA-containing DNA. An increase in polarization upon TBP addition indicates binding to the Gal4 derivative, but preincubation of TBP and TATA blocks this association. A, schematic representation of the possible molecular species formed. B, graph of the polarization data.

These experiments utilized purified, recombinant TBP. A concern is that the results might not reflect the properties of TBP in its native environment, where it is associated with other transcription factors such as TBP-associated factors (24, 25). To address this point, the effect of TATA-containing DNA on the ability of GST-Gal4 AAD to bind purified TBP was compared with its effect on TBP binding out of a whole cell extract. A pull-down experiment was employed in which the GST-Gal4 AAD was immobilized on glutathione-Sepharose beads. As shown in Fig. 5, bead-bound GST-Gal4 AAD retained TBP from an extract, consistent with previous findings. But this binding was almost completely abrogated when a TATA-containing oligonucleotide (approximately equal in concentration to the AAD) was added to the extract prior to exposure to the bead-bound AAD. The result was similar to that obtained using purified, recombinant TBP (Fig. 5). These data indicate that the binding experiments using purified proteins are representative of Gal4 AAD-TBP interactions when the latter can associate with other proteins.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   The Gal4 activation domain and TATA-containing DNA compete for TBP derived from a crude whole cell extract. A yeast lysate, or purified TBP, was incubated with glutathione bead-bound GST-Gal4 AAD with or without an oligonucleotide containing a consensus TATA box. The TBP associated with GST-Gal4 AAD was then pulled down and detected by SDS-PAGE and Western blotting using anti-TBP antibodies. No binding of TBP to GST alone was observed.

Kinetics of Gal4 AAD-TBP Association and Dissociation-- It was also of interest to examine the kinetics of the formation and the dissociation of the Gal4 AAD-TBP complex. To determine the association rate, GST-Gal4 AAD and TBP180 were incubated together for various time intervals, and then the complexes were pulled down using glutathione-agarose beads. The concentrations of GST-Gal4 AAD (1 µM) and TBP180 (0.3 µM) were above the reported K_D of the complex (2 × 10⁷ M) to encourage complex formation. As shown in Fig. 6, the amount of TBP pulled down by the GST-Gal4 AAD increased slowly with time, reaching half-saturation in about 10-11 min. GST lacking an AAD fusion did not bind detectable amounts of TBP even after 30 min (data not shown).

View larger version (8K): [in this window] [in a new window]   Fig. 6.   The core Gal4 AAD and TBP associate slowly. GST-Gal4 AAD (1 µM) attached to glutathione beads was incubated with TBP protein (0.3 µM). At different time intervals, aliquots were removed from the sample, and the beads were washed to remove unbound TBP. The amount of TBP associated with GST-Gal4 AAD as a function of time was determined by SDS-PAGE and densitometry. 100% bound refers to the amount of TBP that can be bound under these conditions.

A potential complication with this experiment is that the TBP must associate with a bead-bound AAD, and this heterogeneous aspect of the reaction might result in kinetics that do not reflect the true solution phase association rate of the proteins. Therefore, a chemical cross-linking experiment was used to monitor association of GST-Gal4 AAD and TBP in solution. We employed chemistry developed recently in our laboratory in which a water-soluble Ru(II) complex, tris(2,2'-bipyridyl)ruthenium(II) chloride hexahydrate, is activated by photolysis with visible light in the presence of ammonium persulfate (26). The resultant Ru(III) complex mediates very rapid and efficient cross-linking of closely associated proteins. In many cases, good yields of cross-linked products can be obtained in 1 s or less, making this technique useful for monitoring association reactions that occur over several seconds or minutes. GST-Gal4 AAD (1.3 µM) and TBP (0.16 µM) were mixed together in a buffer containing the Ru(II) complex and ammonium persulfate at time 0. At the times indicated in Fig. 7, the sample was irradiated for 0.5 s, then quenched immediately by the addition of a reducing buffer. The degree of GST-Gal4 AAD-TBP cross-linking was analyzed by SDS-PAGE and Western blotting using an antibody raised against TBP. An identical experiment was done using GST (1.3 µM) in place of GST-Gal4 AAD as a control. As shown on the left side of Fig. 7, cross-linking in the presence of GST lacking an AAD produced only two TBP-containing bands, the monomer and the homodimer. No spurious bands due to GST-TBP cross-linking were observed. The production of TBP homodimers was expected because this protein has previously been reported to dimerize (20). When GST-Gal4 AAD was mixed with TBP, there was a reduction of the TBP dimer band, and at least two new bands of higher molecular mass were produced at early times (Fig. 7, right side). One had the mobility expected of the (GST-Gal4 AAD)₂-TBP cross-linked complex, and the other was of higher apparent molecular mass. The latter species gradually disappeared over the course of the experiment, and the band representing the two GST-Gal4 AAD molecules and one TBP molecule cross-linked together intensified. Finally, a less intense band corresponding to a GST-Gal4 AAD-TBP cross-link also appeared and intensified over the course of this experiment. The latter species presumably results from incomplete cross-linking of the (GST-Gal4 AAD)₂-TBP complex. Finally, at later times, only bands resulting from a complex containing two molecules of GST-Gal4 AAD and one molecule of TBP are detectable by cross-linking. This suggests that the GST-Gal4 AAD-TBP complex indeed has a 2:1, rather than 4:2, stoichiometry, although we cannot absolutely rule out the latter based on cross-linking data alone.

View larger version (47K): [in this window] [in a new window]   Fig. 7.   The association of the core Gal4 AAD and TBP analyzed by photoinitiated chemical cross-linking. GST (left) or the GST-Gal4 AAD fusion protein (right) was incubated with TBP. At the times indicated, aliquots were withdrawn and cross-linked using a rapid and efficient light-initiated reaction (17). The aliquots were electrophoresed through an SDS-PAGE gel, and the TBP-containing products were identified by Western blotting.

The pull-down and cross-linking experiments were done under similar conditions except that the latter employed a lower TBP concentration (0.16 versus 0.3 µM). Although both reveal a modest rate of association between the Gal4 AAD and TBP, it is interesting that the cross-linking bands assigned to the AAD₂-TBP complex seem to appear with a rate faster than that observed in the pull-down experiment, which employed a higher TBP concentration. This could be due to the differences between heterogeneous and homogeneous reaction conditions. Alternatively, this apparently counterintuitive result could be explained by the well known phenomenon of TBP dimerization. It is the dissociation of these dimers that limits the association of TBP with DNA (27, 28). Since the Gal4 AAD and DNA appear to bind similar surfaces of TBP, we speculate that AAD-TBP association is also limited by TBP dimer dissociation. There may be more monomeric TBP present at the lower concentration employed in the cross-linking reaction. Invoking the TBP dimer as the initial form of the protein in this experiment also provides a possible rationalization for the high molecular mass product formed initially in the cross-linking reaction. This might be ascribed to a weaker, transient association of the GST-AAD dimer with the TBP homodimer, although we cannot assign this species in an unequivocal fashion. It can be detected by the highly efficient Ru(II)-mediated cross-linking reaction, but not by pull-down experiments in which the pellet is washed several times. In summary, if one considers the complications of the TBP homodimer, it is difficult to say what the association rate of TBP with the Gal4 AAD really is because the value obtained will be highly condition-dependent. This makes it is difficult to say how well any number measured would reflect the in vivo situation when an unknown fraction of the TBP exists in dimeric form.

To measure the kinetics of dissociation of the Gal4 AAD-TBP complex, a complex was formed between the Gal4-(1-93 + 768-881) derivative and a fluoresceinated 21-mer oligonucleotide containing a consensus, high affinity 17-base pair Gal4p binding site. This complex was then incubated with TBP, resulting in an increase in the polarization of the fluorescent signal. The DNA-Gal4-(1-93 + 768-881)-TBP complex was then challenged with an excess of GST-Gal4 AAD. The dissociation of TBP from the DNA-Gal4-(1-93 + 768-881) complex was monitored by the resultant decrease of the fluorescence polarization. The data obtained from this experiment (Fig. 8A) reveal that the TBP-Gal4p complex has a half-life of 4 min under these conditions (k_diss = 3.2 × 10³ s¹). The Gal4 derivative-DNA complex was very stable over the life of the experiment (Fig. 8B) and therefore, all of the drop in polarization observed in Fig. 8A can be ascribed to TBP dissociation from the Gal4 derivative-DNA complex.

View larger version (10K): [in this window] [in a new window]   Fig. 8.   The dissociation rate of the complex formed between a large AAD-containing Gal4 fragment and TBP. A, a complex was formed that included Gal4p-(1-93 + 768-881) bound to a fluorescein-labeled UAS containing DNA and TBP. This was challenged with GST-Gal4 AAD. A semilog plot showing the decay of the complex as monitored by the change in fluorescence polarization is shown. The derived t1/2 was 3.5 min (kdiss = 3.2 × 103 s1). B, the Gal4p-(1-93 + 768-881)-UAS complex (lacking TBP) was challenged with excess GST. No effect was observed, indicating that the Gal4 derivative-DNA complex was stable over the lifetime of the experiment.

Implications for the Role of Gal4 AAD-TBP Contacts in Vivo-- The experiments described here demonstrate that the Gal4 AAD and TBP form a well defined complex of 2:1 stoichiometry that is relatively slow to form and has a half-life of approximately 4 min. If the Gal4 AAD binds monomeric TBP, as the data suggest, then the slow association rate probably reflects, at least in part, the relatively slow dissociation rate of TBP dimers. The most striking result obtained in this study is that the Gal4 AAD and DNA compete for binding to TBP. This finding is consistent with earlier studies, which demonstrated that mutations that alter binding of the AAD to TBP are located on the DNA binding surface of the basal factor (6, 21). We note that the AAD of the herpes simplex virus VP16 has also been shown to recognize the DNA binding surface of TBP (29). This finding is apparently at odds with a common model for activator-stimulated TBP function, which is that the activator and TBP bind cooperatively to promoters. Indeed, recent studies in our laboratory have demonstrated that the Gal4 activator and TBP do not bind DNA cooperatively in vitro or in vivo.³ This is not to say that the Gal4 activator might not stimulate TBP function in some other fashion. However, mechanistic models will have to take into account this observation that DNA and the Gal4 AAD cannot co-occupy TBP.

The impact of the kinetic measurements reported here is harder to judge. It is known that the GAL genes respond rapidly upon induction in yeast, so if the Gal4 protein does somehow stimulate TBP binding, this must be a relatively fast process. We observe relatively slow association and dissociation kinetics of the complex, which would seem to be at odds with rapid induction of transcription. However, there remains the possibility that the kinetics of binding of the intact Gal4p to TBP in vivo might differ quantitatively from those observed in our in vitro experiments, because one suspects that TBP dimerization plays a major role in the association rate at least. It is also possible that the dissociation rate could vary significantly in vivo from the value measured in vitro. However, this value should be relatively unaffected by TBP dimerization. Furthermore, a very large AD-containing C-terminal fragment was employed in the off-rate experiment (Fig. 8), making it less likely that a complex containing intact Gal4p would exhibit substantially different kinetics. Nonetheless, it remains possible that a putative complex between TBP and the native Gal4 protein associates and dissociates rapidly enough to play a kinetic role in the formation of a transcription complex, for example by competing an inhibitor from TBP, then "handing off" TBP to the DNA (30).

Alternatively, the possibility must be considered that the Gal4 AAD-TBP association may not be biologically relevant. While it has been shown here that the proteins do form a well defined 2:1 complex, the biochemical properties of the complex do not seem to obviously support any of the common models for activation. While it is clear that induction of Gal4p activity in vivo does stimulate TBP-TATA binding (2, 3), this could be an indirect effect not involving direct Gal4 AAD-TBP contacts.

	FOOTNOTES

^* This work was supported in part by American Cancer Society Grant NP-935 and Welch Foundation Grant I-1299.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Partially supported by National Institutes of Health Training Grant HLO7360.

^§ To whom correspondence should be addressed: Dept. of Internal Medicine and Biochemistry, Ryburn Center for Molecular Cardiology, University of Texas Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-8573. Tel.: 214-648-1239; Fax: 214-648-1415; E-mail: thomas.kodadek@utsouthwestern.edu.

Published, JBC Papers in Press, August 7, 2000, DOI 10.1074/jbc.M003760200

² Y. Xie and T. Kodadek, unpublished data.

³ Y. Xie, L. Sun and T. Kodadek, submitted for publication.

	ABBREVIATIONS

The abbreviations used are: TBP, TATA-binding protein; AAD, acidic activation domain; GST, glutathione S-transferase; PBS, phosphate-buffered saline; PAGE, polyacrylamide gel electrophoresis; UAS, upstream activation sequence.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				S. Takahata, H. Ryu, K. Ohtsuki, K. Kasahara, M. Kawaichi, and T. Kokubo Identification of a Novel TATA Element-binding Protein Binding Region at the N Terminus of the Saccharomyces cerevisiae TAF1 Protein J. Biol. Chem., November 14, 2003; 278(46): 45888 - 45902. [Abstract] [Full Text] [PDF]

				S. Dai, S. Petruccelli, M. I. Ordiz, Z. Zhang, S. Chen, and R. N. Beachy Functional Analysis of RF2a, a Rice Transcription Factor J. Biol. Chem., September 19, 2003; 278(38): 36396 - 36402. [Abstract] [Full Text] [PDF]

				L. Sun and T. Kodadek Removal of impurities from transcription factor preparations that alter their DNA-binding properties Nucleic Acids Res., August 15, 2002; 30(16): e88 - e88. [Abstract] [Full Text] [PDF]

				S. R. Bhaumik and M. R. Green SAGA is an essential in vivo target of the yeast acidic activator Gal4p Genes & Dev., August 1, 2001; 15(15): 1935 - 1945. [Abstract] [Full Text] [PDF]

				Y. Xie, L. Sun, and T. Kodadek TATA-binding Protein and the Gal4 Transactivator Do Not Bind to Promoters Cooperatively J. Biol. Chem., December 22, 2000; 275(52): 40797 - 40803. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/41/31914    most recent M003760200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Xie, Y. Articles by Kodadek, T. Search for Related Content PubMed PubMed Citation Articles by Xie, Y. Articles by Kodadek, T. Social Bookmarking                      What's this?