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

J. Biol. Chem., Vol. 277, Issue 37, 34003-34009, September 13, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/37/34003    most recent M206168200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hannum, C. Articles by Rolfes, R. J. Search for Related Content PubMed PubMed Citation Articles by Hannum, C. Articles by Rolfes, R. J. Social Bookmarking                      What's this?

Functional Mapping of Bas2

IDENTIFICATION OF ACTIVATION AND Bas1-INTERACTION DOMAINS^*

Charles Hannum, Olga I. Kulaeva^§, Helen Sun, Jennifer L. Urbanowski, Ashley Wendus, David J. Stillman^¶, and Ronda J. Rolfes

From the  Department of Biology, Georgetown University, Washington, D. C. 20057-1229 and ^¶ Division of Cell Biology and Immunology, Department of Pathology, University of Utah Health Sciences Center, Salt Lake City, Utah 84132-2501

Received for publication, June 20, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The transcriptional activator protein Bas2 is required to express more than 20 genes in pathways for purine nucleotide and histidine biosynthesis, phosphate utilization, and the HO endonuclease by acting with co-regulator proteins Bas1, Pho4, and Swi5. The role that Bas2 plays in transcriptional activation may be to unmask latent activation domains in the co-regulator and to promote ternary complex formation between Bas2, the co-regulator, and DNA. We show that Bas2 also contributes to transcriptional activation by providing an activation domain. We localize this domain in Bas2 to the C-terminal 156 amino acids using deletion analysis and fusion to a heterologous DNA binding domain. Additionally, we show that Bas2 makes direct contacts with Bas1. This interaction is detected by co-immunoprecipitation and by two-hybrid analysis. We localize the interaction region to the central portion of Bas2, from amino acids 112 to 404.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The transcriptional activator protein Bas2 (also known as Pho2 and Grf10) is required to express genes in several pathways by acting in conjunction with one of several different transcription factors, each providing specificity to a different regulon. Bas2 is required for expression of secreted phosphatases with Pho4 (reviewed in Ref. 1) for basal expression of HIS4 and derepression of the purine nucleotide biosynthetic genes with Bas1 (2-4) and for expression of HO with Swi5 (5, 6). Thus, Bas2 has been implicated in the regulation of more than 20 genes, the majority of these in conjunction with Bas1.

The Bas2 co-regulators are thought to provide an activation domain that is able to interact with a component of the transcriptional machinery, Swi/Snf, and Mediator in the case of Swi5 and TFIIB for Pho4 (7-10). The role of Bas2 in transcriptional activation may be to increase the accessibility of masked activation domains, as shown for Pho4 and Bas1 (11-13). Alternatively, Bas2 may interact with a component of the transcriptional machinery different from its co-regulators, although Bas2 is not considered to have an activation domain (14).

Another role for Bas2 is to promote formation of the ternary complex between both activators and DNA. Cooperative binding between Bas2 and Pho4 occurs at PHO5 (15) and between Bas2 and Swi5 at HO (16). However, Bas2 and Bas1 did not exhibit binding cooperativity at HIS4 (4). Cooperative binding suggests physical interaction, although other explanations are possible (17, 18); however, two-hybrid analyzes detect an interaction of Bas2 with Pho4, Bas1, and Swi5 (12-14, 19).

In contrast to a report which indicated that Bas2 lacks an activation region (14), our previous work suggested that Bas2 does provide activation function (12). In this report, we clearly show that Bas2 has an activation domain and localize this activity to the C-terminal 156 amino acids. Additionally, we demonstrate that Bas2 makes a direct contact with Bas1, and we map this region for interaction along the central portion of Bas2, from amino acids 112 to 404.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Strains and Plasmids-- For this work, we used Saccharomyces cerevisiae strain RR88 (MAT his3 trp1 ura3-1 leu2::lexA_op-LEU2 bas1-2 bas2-2), which has been previously described (12), and strains DY6977 (MAT ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 lys2 ura3-1::<lexA_op-lacZ URA3> bas1::KanMX PHO4 SWI5), DY6975 (MAT ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 lys2 ura3-1::<lexA_op-lacZ URA3> bas1::KanMX pho4::ADE2 SWI5), DY6981 (MAT ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 lys2 ura3-1::<lexA_op-lacZ URA3> bas1::KanMX PHO4 swi5::hisG), and DY6980 (MAT ade2-1 can1-100 his3-11,15 leu2-3,112 trp1-1 lys2 ura3-1::<lexA_op-lacZ URA3> bas1::KanMX pho4::ADE2 swi5::hisG).

All of the plasmids used in this work are listed in Table I. Plasmids pCB841 (3) and p2331 (12) carry wild-type and mutant alleles of BAS2, respectively. Vector pEG202 was used to construct proteins fused with LexA (20), and pHQ346 was used for fusions with an activation domain (21). Plasmids expressing LexA-Bas1 and LexA-Bas2 (12), LexA-bicoid (20), LexA-Gal4 (22), the high copy (2 µm) lexA_op-lacZ reporter (23) and the integrating lexA_op-lacZ reporter (24) have been previously described.

Plasmids carrying the activation domain fused with portions of Bas2 are derivatives of pHQ346 or pOS1. Plasmid pOS1 was made by filling in the EcoRI site of pHQ346 to generate the polylinker in another reading frame. Eight plasmids were constructed in pHQ346: pOS9 contains the 1905-bp EcoRI-ClaI fragment of BAS2, pOS3 contains the 324-bp EcoRI-BglII fragment of BAS2, pOS2 contains the 773-bp EcoRI-NcoI fragment of BAS2, pOS4 contains the 1092-bp EcoRI fragment of BAS2, pOS5 contains the 443-bp BglII-NcoI fragment of BAS2, pOS10 contains the 326-bp PCR fragment of BAS2 made using oligonucleotides OS-142 (5'-GCTCGGATCCGGCCAAATTTGAACAATCAGTGGTCC-3', BamHI site is underlined) and OS-144 (5'-GCTCCTCGAGCTCGATAACAGATCATGATCG-3', XhoI site underlined), pOS11 contains the 606-bp PCR fragment of BAS2 made using oligonucleotides OS-142 and RO-102 (5'-GCTCCTCGAGTATCCATCTATGCTCGTCAG-3', XhoI site is underlined), and pOS12 contains the 307-bp PCR fragment of BAS2 made using oligonucleotides OS-143 (5'-GCTCGGATCCGCGATCATGATCTGTTATCGAG-3', BamHI site is underlined) and RO-102. Three plasmids were made in pOS1: pOS6 contains the 319-bp NcoI-NdeI fragment of BAS2, pOS7 contains the 680-bp NcoI-HincII fragment of BAS2, and pOS21 contains the 872-bp BglII-EcoRI fragment of BAS2.

Plasmids carrying the lexA-protein fusions are derivatives of pEG202 (20): pCH10 contains the 1905-bp EcoRI-ClaI fragment of BAS2 with the mutated homeodomain, pCH3 contains the 773-bp EcoRI-NcoI fragment of BAS2 containing the mutated homeodomain, pCY2 contains the 319-bp NcoI-NdeI fragment of BAS2, pCY3 contains the 872-bp BglII-EcoRI fragment of BAS2, pCH5 contains the 606-bp PCR fragment of BAS2 made using oligonucleotides OS-142 and HO-1 (5'-GCTCCTCGAGTCATATCCATCTATGCTCGTC-3', XhoI site is underlined), pCH8 contains the 330-bp PCR fragment of BAS2 made using oligonucleotides OS-142 and CO-162 (5'-GCTCCTCGAGTCACTCGATAACAGATCATGATCG-3', XhoI site is underlined), pCH6 contains the 307-bp PCR fragment of BAS2 made using oligonucleotides OS-143 and HO-1, pAW1 contains the 475-bp PCR fragment of BAS2 made using oligonucleotides RO-171 (5'-GCTCGGATCCATTCTCTAATTCTAGACTATAAATCATCG-3', BamHI site is underlined) and HO-1, pJU1 contains the 288-bp PCR fragment of BAS2 made using oligonucleotides RO-171 and RO-198 (5'-GCTCCTCGAGCTACGGTAGTATCCAGTAAATTGAGCG-3', XhoI site is underlined), and pJU2 contains the 226-bp PCR fragment of BAS2 made using oligonucleotides RO-171 and CO-162.

Immunoprecipitation-- A 5-ml overnight cell culture in SC medium containing 2% glucose was used to inoculate 50 ml of SC medium containing 1% raffinose and 2% galactose, and the cells were grown for 4-6 h until the A₆₀₀ was between 0.5 and 1.0. Cells were harvested by centrifugation, washed with 5 ml of ice-cold BB50 buffer (20 mM Tris-HCl, pH 8.0, 50 mM KCl, 0.1% Triton X-100), and repelleted. The pellets were frozen overnight at 80 °C after removing the buffer. Cell extracts were prepared in 100 µl of BB50 buffer containing the protease inhibitor mixture Complete (Roche Molecular Biochemicals) by shaking with glass beads for six 1-min pulses in a Tomy shaker at 4 °C with 1-min cooling on ice between shaking. Crude extracts were transferred to a new tube, the glass beads were washed once with 100 µl of BB50 plus protease inhibitor, and the wash was combined with the supernatant. Cellular debris was pelleted by centrifugation for 10 min at 4 °C at 16,000 × g, and the supernatant was transferred to a clean tube. Protein concentrations, generally between 15 and 20 mg/ml, were determined using the Bradford reagent and bovine serum albumin as the standard.

Extracts were precleared by mixing 200 µl of protein-A-Sepharose beads (prepared in BB50) with 2 mg of yeast extract and rocking for 2 h at 4 °C. Antibodies were prebound to protein-A-Sepharose beads by mixing 4 µl of rabbit anti-LexA antiserum (a gift of Dr. Erica Golemis) with 100 µl of beads and rocking them at 4 °C for 2 h. At the end of the 2-h pretreatment, the beads were pelleted from the extract by centrifugation at 8000 × g for 30 s. The entire supernatant was transferred to a clean tube and the divided in half, by volume; half was added to the tube containing the beads prebound with antibody, and the other half was added to the tube with beads but lacking antiserum. Both tubes were rocked at 4 °C overnight. Samples were subjected to centrifugation at 8000 × g for 1 min, and the supernatants were transferred into clean tubes. The pellet fractions were washed 4 times using 500 µl of BB50 + protease inhibitor.

Western Analysis-- The entire pellet fraction from the immunoprecipitation was mixed with 20 µl of 2× SDS sample buffer, and 20 µl of the supernatant (~20%) was mixed with 5 µl of 6× sample buffer (25). Samples were loaded onto 10 or 12% SDS-polyacrylamide gels and subjected to electrophoresis at 100 V (25). Protein was electroblotted onto HYBOND membranes at 4 °C in transfer buffer (10% methanol in 1× SDS running buffer) for 1 h at 100 V. Filters were blocked overnight at 4 °C in PBS-milk-Tween (1× PBS,¹ 2% Carnation nonfat dry milk, 0.1% Tween 20, 0.02% NaN₃).

Filters were washed briefly with PBS-Tween and incubated at ambient temperature for 1 h in PBS-milk-Tween containing 12CA5 monoclonal anti-hemagglutinin (HA) antibody at a 1/2000 dilution (Roche Molecular Biochemicals). Filters were washed four times for 5 min in 20 ml of PBS-Tween. Filters were incubated for 1 h in PBS-Tween containing horseradish peroxidase-conjugated goat anti-mouse Ig at 1/5000 dilution (Amersham Biosciences). Filters were washed 3 times for 5 min each in PBS-Tween. The ECL detection kit (Amersham Biosciences) was used for detection.

-Galactosidase Analysis-- Strains to be assayed were grown at 30 °C for about 48 h to stationary phase in synthetic dextrose medium containing only the supplements (26) dictated by the auxotrophic requirements plus 0.15 mM adenine. For induction of the activation domain (AD)-domain fusion proteins, cells were grown in supplemented synthetic dextrose medium and transferred into synthetic medium with 2% galactose and 1% raffinose as the carbon sources. Cultures were diluted 1/25 into 50 ml of the same medium containing or lacking adenine (as indicated in the text) and grown to mid-log phase. Extracts were prepared, and -galactosidase assays were performed as previously described (27).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Bas2 Contains an Activation Domain Located between Amino Acids 404 and 559-- Our previous results suggest that Bas2 contains an activation domain (12), and we wanted to characterize this domain in greater detail. We integrated the lexA_op-lacZ reporter into the genome and expressed a LexA fusion protein that contained the full-length Bas2 protein, LexA-Bas2-(1-559). The LexA-Bas2-(1-559) fusion promoted expression of the reporter to 49 units (Fig. 1, construct 1), 4-fold above the 12 units for the negative control LexA-bicoid. Because the lexA_op-lacZ reporter was integrated, expression was detected at levels significantly lower than previously reported for episomal constructs (12). To eliminate potential DNA binding by Bas2, we incorporated the double-point mutations (tryptophan 124 to alanine and asparagine 127 to alanine), which disrupt the secondary structure of the homeodomain and essentially eliminate DNA binding activity (28). This protein construct, LexA-Bas2-(4-559m) exhibited the same level of expression as LexA-Bas2-(1-559), indicating that activation function does not require a functional homeodomain (construct 2). These results confirm our earlier report (12) and indicate that Bas2 carries an activation domain.

View larger version (28K): [in this window] [in a new window]   Fig. 1.   The C-terminal 156 amino acids of Bas2 provides an activation domain. Depicted along the top of the figure is a representation of the Bas2 protein showing the relationship of restriction sites used in constructing the fusion proteins and the corresponding amino acid positions (numbered). The homeodomain is depicted as the solid box from amino acids 77 to 136. The striped box located at the N terminus of each chimera represents the LexA protein, amino acids 1-202. The solid black bar depicts portions of Bas2 found in each fusion protein, with an X depicting a homeodomain mutation. -Galactosidase activities (±S.E.) are on the right. -Galactosidase activities of the negative control (LexA-bicoid) and the positive control (LexA-Gal4) are 12 ± 2 units and 2300 ± 500 units, respectively. Assays were performed on extracts from strain RR88 with the integrated lexAop-lacZ reporter. Units (U) = 1 nmol o-nitrophenyl(D)galactopyranoside hydrolyzed/min/mg of protein.

To localize this activation domain, we generated fusion proteins containing the LexA DNA binding domain and various portions of Bas2. As shown in Fig. 1, we made nine new constructs containing N-terminal, central, and C-terminal portions of Bas2. LexA fusion proteins containing N-terminal (LexA-Bas2-(4-263m) and central portions (LexA-Bas2-(262-368) and LexA-Bas2-(112-404)) of Bas2 were unable to stimulate reporter expression (Fig. 1, constructs 3-5). We were able to detect expression of these proteins by Western analysis; however, the molecular weight of LexA-Bas2-(262-368) was the same as LexA alone, suggesting that this fusion protein was unstable and truncated to the more stable LexA domain (data not shown).

LexA fusion proteins containing C-terminal portions of Bas2 exhibited very high reporter expression. Fusion proteins LexA-Bas2-(360-467) and LexA-Bas2-(460-559) led to -galactosidase activities of 120 and 170 units, respectively. These values were 10- and 16-fold higher than the negative control (12 units from LexA-bicoid). A fusion protein containing the equivalent of both of these pieces together, LexA-Bas2-(360-559), strongly stimulated expression to 890 units (construct 6). However, the highest level of expression resided in LexA-Bas2-(404-559) (construct 9). This fusion protein led to 1400 units of -galactosidase activity, a ~120-fold increase. To more finely map this activation region, we generated two additional C-terminal deletion constructs based on this latter construct (Fig. 1, constructs 10 and 11). LexA-Bas2-(404-496) and LexA-Bas2-(404-467) both promote -galactosidase expression to ~290 units but are significantly decreased from the 1400 units in LexA-Bas2-(404-559). Additionally, they are detected at about 2-fold higher levels than LexA-Bas2-(404-559) in immunoblot analysis (data not shown). Taken together these results indicate that a minimally bipartite activation domain resides in the C-terminal portion of Bas2. Because constructs 10 and 11 activate expression to similar levels, the first domain resides between residues 404 and 467, and no activation function can be attributed to residues 468 to 496. Construct 8 (LexA-Bas2-(460-559)) reveals the position of the second domain between 460 to 559; however, because no activation function is present between 468-496, the second domain can be narrowed to 496-559. Importantly, the high level of activation observed with 404-559 indicates that the two independent domains cooperate synergistically to promote transcription. Furthermore, because expression from LexA-Bas2-(404-559) is ~30-fold higher than LexA-Bas2-(1-559) (compare 1400 units to 49 units), it is likely that other portions of the protein normally mask exposure of this domain.

Bas2 Makes Direct Contact with the Transcriptional Machinery-- The experiment described above using the LexA-Bas2 fusion proteins suggests that the C-terminal activation domain makes direct contact with one or more components of the transcriptional machinery. An alternative interpretation is that the region of Bas2 from amino acids 404 to 559 is making an indirect contact with the transcriptional machinery through one of its partners, Bas1, Pho4, or Swi5. Both Pho4 and Swi5 interact with Bas2, and they also make direct contacts with the transcriptional machinery; Pho4 contacts TFIIB (8), and Swi5 contacts the Swi/Snf and Mediator complexes (7, 10). In the experiment above, we used a bas1 strain to eliminate effects from Bas1. To determine whether Pho4 or Swi5 or both are contributing to the expression detected in Fig. 1, strains were constructed that carried deletion alleles of PHO4 and SWI5 as single alleles or together as a double mutant in a strain that already carried a deletion of bas1. Plasmids encoding LexA-bicoid, LexA-Bas2-(1-559), and LexA-Bas2-(404-559) were transformed into these four strains, and extracts were assayed for expression of the lexA_op-lacZ reporter construct. As shown in Table II, both LexA-Bas2-(1-559) and LexA-Bas2-(404-559) express approximately the same level of -galactosidase activity independent of the presence or absence of Swi5 or Pho4. These results indicate that the activation domain of Bas2 is not indirectly interacting with the transcriptional machinery via Swi5 or Pho4, although it remains possible that it does so through an uncharacterized activator. Instead, it appears that LexA-Bas2-(404-559) contacts the transcriptional machinery directly.

View this table: [in this window] [in a new window]   Table II -Galactosidase expression

Bas2 Directly Binds Bas1-- We previously demonstrated that a LexA-Bas1 fusion protein promoted adenine-regulated, high level expression when cells expressed the native Bas2 protein or a derivative that carried mutations disrupting the homeodomain, but only basal levels of expression were observed in the absence of Bas2 (12). We interpreted those data to indicate that LexA-Bas1 tethered Bas2 to DNA by direct protein interaction. To determine whether Bas1 and Bas2 directly associate with each other, we performed a co-immunoprecipitation using the two-hybrid reagents we developed. Yeast extracts were prepared from strains co-expressing either LexA-Bas1 or LexA-alone along with a Bas2 fusion protein composed of Bas2-(4-559), the B42 AD, the hemagglutinin epitope, and a nuclear localization signal. We performed immunoprecipitations using an antiserum generated against LexA, a control antiserum generated against eIF2, or no antiserum. The pellet and supernatant fractions were subjected to SDS-PAGE, and the proteins were blotted to a membrane filter. An anti-HA antibody was used to detect the hemagglutinin epitope in the AD-Bas2 fusion protein.

The results of this analysis are shown in Fig. 2A. No AD-Bas2 protein was detected in the pellet fraction of cells that expressed LexA protein without Bas1 (lane 6). When the immunoprecipitation antibody was omitted (lanes 1 and 2) or when anti-eIF2 was used (lanes 7 and 8), AD-Bas2 protein was detected only in the supernatant fraction. AD-Bas2 protein was found in the pellet fraction only when anti-LexA antiserum was added to the immunoprecipitation of cell extracts expressing LexA-Bas1 (lane 4). Thus, we detect a direct interaction between Bas1 and Bas2 by co-immunoprecipitation.

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Anti-LexA antiserum co-immunoprecipitates an AD-Bas2 fusion protein independent of adenine. A, the LexA-Bas1 (lanes 1-4 and 7-8) or LexA (lanes 5 and 6) proteins were immunoprecipitated using a rabbit polyclonal anti-LexA antiserum (lanes 3-6) or a rabbit polyclonal anti-eIF2 antiserum (lanes 7-8). The pellet (P) and supernatant (S) fractions were subjected to SDS-PAGE and blotted to membranes, and the position of AD-Bas2 was detected by Western analysis using a monoclonal antibody against the hemagglutinin epitope in this protein. Ab, antibody. B, the co-immunoprecipitation assay was performed as described above on extracts prepared from cells cultured in the presence or absence of adenine. Anti-LexA antiserum was added to the reactions in lanes 1-2 and 5-6. The arrow indicates the position of the AD-Bas2 protein detected through its HA epitope. P indicates the pellet fraction, and S indicates the supernatant fraction, ~20% of the total volume.

Bas2 Contains a Region That Interacts with Bas1-- To characterize the interaction between Bas1 and Bas2, we used a two-hybrid approach. A plasmid was constructed to express nearly the entirety of Bas2 fused to the B42 AD. Transformed cells carrying plasmids that expressed both the LexA-Bas1 fusion and either the AD-Bas2 or AD-vector were grown in media lacking adenine (derepressing conditions) and containing adenine (repressing conditions). The LexA-Bas1 chimera with AD vector promoted expression of the lexA_op-lacZ reporter to only the low level of ~220 units under derepressing conditions and 170 units under repressing conditions, as shown in Fig. 3 (AD vector, construct 1). The fusion protein containing Bas2 (AD-Bas2-(4-559), construct 2) stimulated expression from the reporter to 1300 units under derepressing conditions and 260 units under repressing conditions. If we subtract out the background of the AD vector construct and divide to obtain a repression ratio, the AD-Bas2-(4-559) protein exhibits a 12-fold adenine repression of the lexA_op-lacZ reporter. Thus, these results demonstrate that adenine regulation occurs when Bas1 and Bas2 are in a two-hybrid context and that adenine limitation is required for the interaction between Bas1 and Bas2, consistent with our previous results (12).

View larger version (30K): [in this window] [in a new window]   Fig. 3.   The central region of Bas2 is required to interact with LexA-Bas1. Depicted along the top of the figure is a representation of the Bas2 protein showing the relationship of restriction sites used in constructing the fusion proteins and the corresponding amino acid positions. The homeodomain is depicted as the solid box from amino acids 77-136. The striped box located at the N terminus represents the B42 activation domain, the nuclear localization signal, and the HA epitope. The solid black bar depicts portions of Bas2 found in each fusion protein. On the right side of the figure, we indicate the -galactosidase activities (±S.E.) and the repression ratio (calculated as the derepressed level (ade) divided by the repressed level (+ade) after subtracting out the vector levels). Assays were performed on extracts from strain RR88 carrying the lexAop-lacZ reporter on plasmid pSH18-34. -Galactosidase values that are statistically significant (p < 0.05 by Student t test) are shown in bold. AD vector values are the combined values for pHQ346 and pOS1.

Co-immunoprecipitation Does Not Reflect Growth Conditions-- Because growth in the presence and absence of adenine led to regulated expression of the two-hybrid reporter and regulated interaction between LexA-Bas1 and AD-Bas2, we wanted to determine whether we could detect a difference in the ability to co-immunoprecipitate AD-Bas2 with LexA-Bas1 under these two growth conditions. Whole cell extracts were prepared from cells grown in media containing or lacking adenine. Anti-LexA antiserum was mixed with extracts, which were separated into pellet and supernatant fractions, as described above. We detected approximately the same amount of AD-Bas2 in the pellet fraction under both conditions (Fig. 2B). Thus, when cells are grown in adenine, we can detect repression of reporter expression and regulated interaction in the two-hybrid assay but no difference in the ability to co-immunoprecipitate. This result suggests that the adenine regulatory signal, as communicated to the transcription factors, is lost during preparation of the extracts for co-immunoprecipitation analysis.

Mapping the Interaction Domain of Bas2-- To map the region in Bas2 that interacts with Bas1, we constructed 10 plasmids to carry truncations of Bas2 fused to the AD. -Galactosidase activities were determined from extracts prepared from cells expressing LexA-Bas1 and one of the AD-Bas2 constructs.

We generated a set of C-terminal deletions in Bas2 (Fig. 3, constructs 3-5) and compared expression from these proteins with the full-length fusion protein (construct 2). The fusion proteins carrying the longer two portions of Bas2, AD-Bas2-(4-404) and AD-Bas2-(4-263), each expressed high levels of -galactosidase under derepressing conditions (1010 units and 940 units, respectively). They also expressed a significant level of -galactosidase under repressing conditions, measured at about 400 units for both constructs. The smallest construct made in this set, AD-Bas2-(4-112), failed to promote expression significantly above the empty vector. These results indicate that the C terminus is dispensable for interaction with LexA-Bas1, and the N terminus is insufficient to promote interaction, indicating that portions of the center of Bas2 are important for interaction with Bas1.

We generated a set of three large constructs that covered segments in the middle and C terminus (Fig. 3, constructs 6-8). Only one of these fusion proteins produced elevated -galactosidase activities. AD-Bas2-(112-404) stimulated expression to 440 units under derepressing conditions, marginally significant with a p value of 0.054. Interestingly, the portion of Bas2 found in this construct contains the entire middle region of Bas2 from the homeodomain to the activation domain we defined above (Fig. 1).

Finally, we made a set of four small fusions that, together with the N-terminal construct discussed above, cover the entire Bas2 protein (Fig. 3, constructs 5 and 9-12). The N-terminal (AD-Bas2-(4-112)) and C-terminal (AD-Bas2-(460-559)) fragments failed to promote expression of the lexA_op-lacZ reporter (Fig. 3, constructs 5 and 12). Interestingly, three fragments that essentially do not overlap with each other are each able to stimulate significant expression under both repressing and derepressing conditions; these constructs are AD-Bas2-(112-263), AD-Bas2-(262-368), and AD-Bas2-(360-467) (constructs 9-11). The finding that non-overlapping portions of Bas2 can individually stimulate reporter expression indicates that the region of Bas2 that interacts with Bas1 may be an extensive region in the center of the molecule, localized between amino acids 112 and 404. The homeodomain extends from position 71 to 136; thus, we can provisionally limit the region for interaction from amino acids 137 to 404.

We also observed that the ability to repress expression with adenine was impaired in all of the truncation constructs. As described above, expression of the lexA_op-lacZ reporter by LexA-Bas1 and AD-Bas2-(4-559) is regulated by growth in adenine, resulting in a repression ratio of 12. When the repression ratios are determined for the six constructs that promoted significant expression (constructs 3-4, 6, 9-11), the repression ratios fall between 1.2 and 3.3. Thus, the truncated Bas2 fusion proteins have lost much of the ability to achieve repression, and full repression may require the entire Bas2 protein.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Bas2 is a transcriptional activator that participates in the regulation of more than 20 genes involved in multiple pathways. This homeodomain protein binds to the proposed consensus sequence 5'-(T/C)TAA(T/A)T(T/G)AAT-3' (15) at sites located adjacent to binding sites for one of three co-regulators, Bas1, Pho4, or Swi5. A schematic of the structure of Bas2 is given in Fig. 4. The homeodomain is localized near the N terminus at amino acid positions 77-136 (29). As described in greater detail below, we define an activation domain at the C terminus from 404 to 559, with two smaller regions that function synergistically. We also identify a region in the center of the protein that is important for Bas2 to interact with Bas1.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Bas2 functional domains. Depicted is a schematic of Bas2 showing the functional domains. The homeodomain (HD) localizes to amino acids 77-136 and is shown as the closed rectangle. The activation domain, from 404 to 559, is subdivided into two domains, AD-1, with amino acids 404-467, and AD-2, with amino acids 496-559, each shown separately as the striped boxes. Above the activation domains are the sequences of the clusters of aromatic and bulky hydrophobic residues; each cluster of consecutive residues ends with the position number of the last residue, and they are separated by periods. The bars under this schematic indicate the three portions of Bas2 found in the constructs that interacted with LexA-Bas1. Shown as the open boxes are regions in which mutations affect interaction with the partner proteins2; interaction region 1 (IR-1) encodes positions 244-270, and interaction region 2 (IR-2) encodes positions 343-390. The area labeled as acidic is a 40-amino acid hydrophilic region from 286-326 in which 15 positions are acidic.

The Activation Domain of Bas2-- Using a domain swap approach, we found that the Bas2 protein carries an activation domain in the C-terminal 156 amino acids (positions 404-559 of Bas2). This region was a better activation region than a fragment carrying the C-terminal 200 amino acids (positions 360-559 of Bas2). Interestingly, when two fragments of ~100 amino acids were examined, 107 amino acids of Bas2 from 360-466 and 100 amino acids of Bas2 from 460-559, neither of these fragments was able to promote expression to the very high levels seen in the larger proteins, although these did stimulate expression significantly above background. We generated two smaller constructs starting from AD-Bas2-(404-559); one encodes the 93 amino acids from 404 to 496, and the other encodes the 64 amino acids from 404 to 467. Both of these fusion proteins have lower activation than the parent construct. Taken together, the activation function has been localized to two 64-amino acid regions, AD-1, from 404 to 467, and AD-2, from 496 to 559, that act synergistically to stimulate expression.

We initially hypothesized that an acidic region in the center of the protein, from 286 to 326, may be important for activation function, because acidic activation domains have been previously characterized (30). However, the LexA-Bas2-(112-404) protein was unable to promote expression of the lexA_op-lacZ reporter, and instead, we uncovered the importance of the C terminus for activation. The activation domain found in the C-terminal 156 residues is slightly acidic in character, although not overwhelmingly so; acidic residues account for only 14% of the total. Other types of activation domains have been identified, including glutamine-rich and proline-rich activation domains (30), although these latter amino acids are not abundant in the C terminus of Bas2. Clusters of aromatic and bulky hydrophobic amino acids have been shown to be important for activation function of the VP16 and GCN4 transcription factors (30-34). Using these criteria, several potential aromatic/bulky hydrophobic sequence motifs are found in the two C-terminal activation regions. In AD-1 we find Leu⁴⁰⁷-Asp⁴⁰⁸-Tyr⁴⁰⁹, Phe⁴³⁷-Leu⁴³⁸, and Leu⁴⁵⁷-Phe⁴⁵⁸-Phe⁴⁵⁹-Asp⁴⁶⁰, and in AD-2 we see Asp⁵⁴⁴-Phe⁵⁴⁵-Leu⁵⁴⁶ and Trp⁵⁵⁸-Ile⁵⁵⁹ (Fig. 4). If there is a minimal number of these sequence motifs critical for function of the Bas2 activation domain, perhaps separating them into the two fragments dramatically decreases their effectiveness.

Which component of the transcriptional machinery does Bas2 target? Using a glutathione S-transferase (GST) pull-down assay, GST-Bas2 and GST-Pho4 interacted with recombinant T7 fusion proteins of TATA-binding protein, TFIIB, and TFIIE in vitro (9). Thus, these proteins are candidates for interaction with the Bas2 activation domain in vivo. Interestingly, Bas2 may interact with another component of the transcriptional machinery, NuA4. NuA4 is a histone acetyltransferase complex, and mutations in NuA4 subunits affect HIS4 and PHO5 expression (35). Interestingly, Bas2 stimulates expression of both HIS4 and PHO5 with Bas1 and Pho4, respectively. If Bas2 does indeed interact with NuA4, then perhaps Bas2 recruits a different co-activator from that recruited by Pho4 and Swi5.

Bas2 Interacts Directly with Bas1-- We demonstrated a direct interaction between Bas1 and Bas2 using a co-immunoprecipitation assay. In this assay, LexA-Bas1 was immunoprecipitated from a yeast extract, and AD-Bas2-(4-559) was detected via its HA tag in Western analysis. Our results, showing an interaction between Bas1 and Bas2, should be contrasted to the experiments showing Bas2 interacting with Swi5 (5) or with Pho4 (9). In those studies, the presence of an added DNA fragment containing the cognate binding sites was required to detect the interaction between the two proteins. It may be argued that our assay conditions favored interaction due to high protein concentration. However, we note that the AD-Bas2 fusion protein appeared to be poorly expressed, because it was considerably more difficult to detect by Western analysis, and we were unsuccessful in obtaining a co-immunoprecipitation using the anti-HA antibody. Together, these results suggest a weak interaction between Bas2 and its partners, Pho4 and Swi5, in the absence of DNA but a stronger interaction with Bas1.

This intriguing result may reflect a bone fide relationship that occurs in vivo. Justice et al. (28) find that two mutations in the Bas2 homeodomain, W124A and N127A, eliminated expression of a Swi5-based reporter (HO-lacZ) and a Pho4-based reporter (PHO5-lacZ) yet retained ~30% of expression from the Bas1-based reporter (ADE1-lacZ). These results echoed those we found at ADE5,7-lacZ, where the Bas2-(W124A, N127A) and Bas2-(114-136) mutations also retained a limited ability to derepress expression (~2-fold) when compared with the wild-type protein, which derepressed this reporter 9-fold (12). However, both the Bas2-(W124A, N127A) and Bas2-(114-136) were able to stimulate expression of lexA_op-lacZ to nearly the same level as wild-type Bas2 when combined with LexA-Bas1. These results indicate that interaction of Bas2 with Bas1 occurs when the structure and function of the Bas2 DNA binding domain have been disrupted.

Mapping the Interaction Domain-- The central portion of Bas2 is able to interact with LexA-Bas1 in a two-hybrid assay. Several constructs expressing fusions of Bas2 with the B42 activation domain stimulated reporter expression above that produced by LexA-Bas1 alone, although none of these expression levels was as high as the full-length fusion. This reduction in expression may be due to the absence of the identified activation domain from Bas2 (Fig. 1) that is found only in the full-length construct or due to an inability of the truncated proteins to open the masked activation domain in Bas1 (12). The three protein fragments that promoted the highest degree of expression, AD-Bas2-(112-263), AD-Bas2-(262-368), and AD-Bas2-(360-467), do not carry an overlapping region, suggesting that disparate regions in the primary sequence may fold to produce a face for interaction with Bas1.

We tested the three constructs (AD-Bas2-(112-263), AD-Bas2-(262-368), and AD-Bas2-(360-467)) that activated expression with LexA-Bas1 in our co-immunoprecipitation assay to see if we could detect a direct interaction biochemically. Unfortunately, the HA-tagged AD-Bas2 proteins were detected in the pellet fractions when antibody was not added to the immunoprecipitation reactions, suggesting that the truncated proteins aggregated in solution (data not shown). Although we were unable to confirm our two-hybrid results through a biochemical assay, in a related study we were able to show the importance of this region by mutational analysis. We have isolated point mutations in Bas2 that affect the expression of two Bas1-dependent reporters, HIS4-lacZ and ADE5,7-lacZ, as well as Pho4- and Swi5-dependent reporters.² These mutations cluster in two regions, Region 1, from 244 to 270, and Region 2, from 343 to 390. Interestingly, the positions of these point mutations are coincident with the fragments that are functional in the two-hybrid assay, providing support for the interpretation that the central region of Bas2 is important for partner interaction.

The full-length Bas2 fusion protein was capable of adenine-regulated transcriptional activation in this two-hybrid context. This was not unexpected because adenine-regulated expression occurs with LexA-Bas1 and native Bas2 (12). However, adenine regulation is lost when Bas1 and Bas2 are fused into a single protein (13) or when Bas1 lacks its C terminus (LexA-Bas1-(1-630); Ref 12). We now show here that adenine regulation is also lost when Bas2 is truncated (Fig. 3). We propose that unmasking of activation domains or ternary complex formation depends on a complex set of interactions between the two proteins.

	ACKNOWLEDGEMENTS

We thank Hongfang Qui for providing plasmid pHQ346 and Erica Golemis and Thomas Dever for providing antisera recognizing LexA and eIF2, respectively. We thank Heidi Elmendorf, Anne Rosenwald, and Thomas Dever for critically reading the manuscript and Catherine Yankelovitch for technical assistance.

	FOOTNOTES

^* This work was supported by National Science Foundation Career Grant MCB-9734170 (to R. J. R.) and National Institutes of Health Grant GM48624 (to D. J. S.).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.

^§ Current address: Wayne State University, Karmanos Cancer Institute, Detroit, MI 48201.

To whom correspondence should be addressed: Dept. of Biology, Box 571229, Georgetown University, Washington, D. C. 20057-1229. Tel.: 202-687-5881; Fax: 202-687-5662; E-mail: rolfesr@georgetown.edu.

Published, JBC Papers in Press, July 10, 2002, DOI 10.1074/jbc.M206168200

² Bhoite, L. T., Allen, J., Garcia, E., Thomas, L., Gregory, I. D., Voth, W. P., Rolfes, R. J., and Stillman, D. J. (2002) J. Biol. Chem., in press.

	ABBREVIATIONS

The abbreviations used are: PBS, phosphate-buffered saline; AD, activation domain; HA, hemagglutinin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				R. N. Koehler, N. Rachfall, and R. J. Rolfes Activation of the ADE Genes Requires the Chromatin Remodeling Complexes SAGA and SWI/SNF Eukaryot. Cell, August 1, 2007; 6(8): 1474 - 1485. [Abstract] [Full Text] [PDF]

				I. Som, R. N. Mitsch, J. L. Urbanowski, and R. J. Rolfes DNA-Bound Bas1 Recruits Pho2 To Activate ADE Genes in Saccharomyces cerevisiae Eukaryot. Cell, October 1, 2005; 4(10): 1725 - 1735. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/37/34003    most recent M206168200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted 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 Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Hannum, C. Articles by Rolfes, R. J. Search for Related Content PubMed PubMed Citation Articles by Hannum, C. Articles by Rolfes, R. J. Social Bookmarking                      What's this?