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

J. Biol. Chem., Vol. 275, Issue 30, 22678-22685, July 28, 2000

This Article Abstract Full Text (PDF) All Versions of this Article: 275/30/22678    most recent M001409200v1 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 Münsterkötter, M. Articles by Hörz, W. Search for Related Content PubMed PubMed Citation Articles by Münsterkötter, M. Articles by Hörz, W. Social Bookmarking                      What's this?

Transcriptional Regulation of the Yeast PHO8 Promoter in Comparison to the Coregulated PHO5 Promoter^*

Martin Münsterkötter, Slobodan Barbaric, and Wolfram Hörz^§

From the Adolf-Butenandt-Institut, Molekularbiologie, Universität München, Schillerstrasse 44, 80336 München, Germany

Received for publication, February 21, 2000, and in revised form, April 26, 2000

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Expression of the PHO8 and PHO5 genes that encode a nonspecific alkaline and acid phosphatase, respectively, is regulated in response to the P_i concentration in the medium by the same transcription factors. Upon induction by phosphate starvation, both promoters undergo characteristic chromatin remodeling, yet the extent of remodeling at the PHO8 promoter is significantly lower than at PHO5. Despite the coordinate regulation of the two promoters, the PHO8 promoter is almost 10 times weaker than PHO5. Here we show that of two Pho4 binding sites that had been previously mapped at the PHO8 promoter in vitro, only the high affinity one, UASp2, is functional in vivo. Activation of the PHO8 promoter is partially Pho2-dependent. However, unlike at PHO5, Pho4 can bind strongly to its binding site in the absence of Pho2 and remodel chromatin in a Pho2-independent manner. Replacement of the inactive UASp1 element by the UASp1 element from the PHO5 promoter results in more extensive chromatin remodeling and a concomitant 2-fold increase in promoter activity. In contrast, replacement of the high affinity UASp2 site with the corresponding site from PHO5 precludes chromatin remodeling completely and as a consequence promoter activation, despite efficient binding of Pho4 to this site. Deletion of the promoter region normally covered by nucleosomes 3 and 2 results in a 2-fold increase in promoter activity, further supporting a repressive role of these nucleosomes. These data show that there can be strong binding of Pho4 to a UAS element without any chromatin remodeling and promoter activation. The close correlation between promoter activity and the extent of chromatin disruption strongly suggests that the low level of PHO8 induction in comparison with PHO5 is partly due to the inability of Pho4 to achieve complete chromatin remodeling at this promoter.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Saccharomyces cerevisiae contains a set of genes coding for proteins involved in phosphate uptake and metabolism, the expression of which is coordinately regulated in response to changes in the inorganic phosphate (P_i) concentration of the growth medium. In P_i-containing medium, transcription of these genes is efficiently repressed, whereas phosphate starvation results in strong induction (1). The most strongly regulated gene is PHO5 (1), which codes for the major repressible nonspecific acid phosphatase isoenzyme (2), a heavily glycosylated protein localized in the periplasmic space (3). In addition, there are also alkaline phosphatases, a nonspecific vacuolar enzyme, encoded by the PHO8 gene (4) and secondly, the product of the PHO13 gene (5). PHO8 transcription is also increased as a consequence of phosphate starvation (1, 6), whereas PHO13 is constitutively expressed independently of the P_i concentration (5).

The expression of all P_i-regulated genes is negatively regulated by the products of the PHO80 and PHO85 genes, whereas the products of PHO2, PHO4, and PHO81 act as positive regulators except for the PHO8 gene, which was reported to be Pho2-independent (1). Pho4 is itself negatively regulated through phosphorylation by the Pho80/Pho85 cyclin/cyclin-dependent kinase complex, which in turn is regulated by the phosphate level (7). Under repressing conditions (high phosphate) Pho4 is hyperphosphorylated by Pho80/85, and its phosphorylated form is then exported into the cytoplasm (8) via the recently identified Msn5 receptor (9). Upon phosphate starvation the Pho80/85 complex is inactivated by the PHO81 gene product, which is activated in response to the phosphate starvation through an as yet unknown mechanism (7).

The molecular basis for Pho4-mediated transcriptional activation has been extensively studied on the PHO5 promoter. There are two major Pho4 binding sites at the PHO5 promoter, corresponding to two regulatory elements, UASp1 and UASp2 (10). Pho4 was shown to bind to these elements in vivo upon phosphate starvation but not under high phosphate conditions (11). The binding of Pho4 to the two UAS elements causes a massive transition of the chromatin structure at the promoter. Four positioned nucleosomes at the repressed promoter undergo a profound structural alteration, resulting in a 600-bp¹ region of the promoter becoming fully accessible (12, 13). This chromatin transition appears to be a prerequisite for transcriptional activation (14). Attempts to separate the chromatin remodeling function from the transcriptional activation function have not been successful thus far (15). Activation of the PHO5 promoter requires an additional activator, the homeodomain protein Pho2 (16), which binds to multiple sites at the PHO5 promoter in a cooperative manner with Pho4 (17). Pho2 plays a dual role, however, in the activation of PHO5. It is critically required for recruitment of Pho4 to UASp1, and in addition it enhances the Pho4 activation potential (18).

The PHO8 promoter is almost 10 times weaker than the PHO5 promoter (Ref. 18 and results of this paper). Deletion analysis of the PHO8 promoter indicated two regulatory regions (19), which correspond to two Pho4 binding sites UASp1 and UASp2, mapped by in vitro footprinting (20). UASp1 is a low affinity binding site with two mismatches to the Pho4 binding site consensus, whereas UASp2 is a high affinity site. No significant sequence homology between the PHO5 and the PHO8 UASp2 sites was found outside the consensus central hexanucleotide. Activation of the PHO8 promoter is also accompanied by chromatin remodeling (20). Under repressing conditions, there is a highly ordered chromatin organization with three hypersensitive regions, two of which contain the Pho4 binding sites, which were previously mapped in vitro. Upon induction, a labile nucleosome located between the two hypersensitive regions with the Pho4 sites is disrupted, and a 300-bp hypersensitive region is generated. However, the promoter region downstream of UASp2 acquires only intermediate accessibility to nucleases, consistent with the persistence of unstable, partially remodeled nucleosomes. This chromatin remodeling is fully Pho4-dependent but does not require Pho2 (20).

We have now extended our investigations to cis and trans factors involved in the regulation of the PHO8 promoter. We reasoned that a side by side comparison of PHO5 and PHO8, two promoters regulated by the same transcription factors that both undergo distinct chromatin transitions with characteristic differences, could yield new insights into the interplay between transcription factors, chromatin repression, and promoter regulation. We demonstrate here that the PHO8 promoter is activated through only one Pho4 binding site. Pho4 binding to this site is largely Pho2-independent, but Pho2 contributes to promoter activation mostly by increasing the activation potential of Pho4. We also show that replacement of the endogenous inactive UASp1 element by PHO5 UASp1 significantly increases promoter activity and leads to a more massive chromatin remodeling at the promoter. Removal of the nucleosomes in the promoter region that do not fully remodel upon activation similarly improves activation, suggesting that the lower level of PHO8 induction in comparison with PHO5 is at least partially due to the inability of Pho4 to fully remodel the repressive chromatin structure through its recruitment to a single binding site.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Yeast Strains and Media-- All S. cerevisiae strains used in this study are isogenic with strain YS18 (MAT his3-11 his3-15 leu2-3 leu2-112 can^r ura35). YS22 contains a disruption of PHO4, YS19 contains a disruption of PHO2, YS27 contains disruptions of PHO2 and PHO4, YS31 contains a disruption of PHO80, and YS32 contains disruptions of PHO80 and PHO2. YS45 contains a disruption of CPF1, YS46 contains disruptions of CPF1 and PHO4, YS42 contains disruptions of CPF1 and PHO2, and YS78 contains disruptions of CPF1, PHO2, and PHO4. Yeast strains were grown in YPDA or YNB medium (Difco, Detroit, MI) supplemented with the required amino acids (high phosphate repressing conditions) or in phosphate-free synthetic medium for induction (21).

Plasmids-- YCpP4 (22) and YCpP4int (18) have been described. The PHO8-lacZ reporter was constructed by exchanging the BamHI PHO5 promoter fragment of the previously described PHO5-LacZ reporter plasmid (23) against a polymerase chain reaction-generated PHO8 promoter fragment that contains 902 bp upstream of the start codon. All variants of the PHO8 promoter were made from this PHO8-lacZ reporter by the polymerase chain reaction megaprimer technique (24). For the UASp mutations, the central hexanucleotide consensus sequence was changed to a HindIII site (5'-AAGCTT-3'). For the UASp1 exchange, the PHO8 sequence 757 to 718 (starting with AGCA and ending with GTAA) was replaced by a PHO5 promoter sequence extending from 388 to 350 (CACA  ... GCAT). For the UASp2 exchange, the PHO8 sequence 543 to 522 (starting with ACGT and ending with CGAT) was replaced by the PHO5 sequence from 262 to 241 (TGGC  ... CTAG), and the adjacent NheI site (GCTAGC) was mutated to a SphI site (GCATGC).

The exchange of the proximal promoter was done by introducing a SpeI site (ACTAGT) at positions 143 to 138 into the PHO8 promoter and at positions 165 to 160 into the PHO5 promoter and making the SpeI site the point of transition from PHO8 to PHO5. For the deletion of nucleosomes 2 and 3, a 296-bp segment between 443 and 158 (starting with CCAG and ending with AAGC) was removed from the PHO8 promoter.

The PHO8-UAS CYC1-LacZ reporters were constructed as described previously (25) by using a 2-µm yeast vector containing a CYC1-LacZ gene fusion (26). A 30-bp promoter fragment extending from 747 to 718 was used as the UASp1 element and the UASp2 element was a 25-bp oligonucleotide ranging from position 543 to position 519. In both cases, reporter plasmids containing two tandem copies of the UAS oligonucleotide in an orientation reverse to that in the natural promoter were used to increase activity.

Integration of the Modified PHO8 Promotors-- The modified PHO8 promoter variants were integrated in a two-step procedure into the chromosomal locus. First the wt PHO8 promoter was replaced by the URA3 gene. In the second step, the URA3 gene was replaced by BamHI fragments from the variant PHO8 promoters using fluororotic acid selection.

Primers-- For dimethyl sulfate (DMS) in vivo footprint analysis of Pho4 binding to UASp2 in the wild type PHO8 promoter, we used the primer PHO8-UASp2-1 (5'-CCGTCCAGTCATGTCGTACAACGG-3'). For the PHO8 promoter variant with the substituted PHO5 UASp2 we used the primer PHO8-UASp2-2 (5'-TTGTTGCCGCTGCTGTTGACTAC-3'), and for the wild type PHO5 promoter we used primer 1 described in Ref. 11.

Functional Assays-- -Galactosidase activity measurements (23), nuclease digestion of isolated nuclei, and DMS in vivo footprint analysis were performed as described previously (21, 27). Alkaline phosphatase activity of permeabilized cells (2% chloroform + 0.005% SDS) was assayed at 30 °C in 1 ml of 50 mM Tris-HCl buffer, pH 8.8, containing 5 mM MgSO₄ and 4 mM p-nitrophenylphosphate. The reaction was stopped with 0.5 ml of 1 N NaOH, and the absorbance of liberated p-nitrophenol was measured at 410 nm as originally described (28). Enzyme activity is expressed in arbitrary units: A_410 nm × 1000/min/(OD_600 nm × ml of cells used).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Induction of the PHO8 Promoter Is Fully Dependent on Pho4 Binding to UASp2-- Alkaline phosphatase activity of yeast cells transferred to phosphate-free medium increases 2.5-3-fold, and this induction is Pho4-dependent (Table I). Significant alkaline phosphatase activity is, however, measured in pho4 cells, because of the presence of the P_i-independent alkaline phosphatase encoded by the PHO13 gene (5). By using a PHO8 promoter-lacZ construct, a 6-7-fold increase in activity was measured under inducing conditions that are Pho4-dependent (Table I), showing that the properties of the promoter are more accurately reflected through the use of the lacZ construct. The extent of induction is an order of magnitude lower than for PHO5 as measured by lacZ constructs or mRNA levels (18).²

View this table: [in this window] [in a new window]   Table I Induction of the PHO8 promoter by phosphate starvation is fully Pho4 dependent The activity of the PHO8 promoter in a wt and a pho4 strain was determined at repressive (+Pi) or inducing (Pi) conditions by measuring the endogenous alkaline phosphatase activity or by using a PHO8 promoter-LacZ construct. Activity of a PHO5 promoter-LacZ construct in a wt strain measured in parallel increased from 9 units at +Pi to 900 units at Pi conditions.

Our in vitro footprinting experiments had revealed two Pho4 binding sites at the PHO8 promoter, a high affinity site (521 to 540) designated UASp2, containing the Pho4 consensus hexanucleotide, 5'-CACGTG-3', and a low affinity site, UASp1, 200 bp further upstream containing two mismatches in the consensus hexanucleotide (20). Even in the repressed promoter, both sites are located in nonnucleosomal, hypersensitive regions (see schematic in Fig. 1). To examine the significance of each of the Pho4 binding sites in vivo, they were mutated, and the activities of the mutated promoter derivatives were determined. As shown in Fig. 1, mutation of the low affinity site UASp1 has no appreciable effect on promoter activity. On the other hand, mutating UASp2 completely abolishes inducibility of the promoter (Fig. 1). As expected, therefore, no further effect was observed by combining the mutations in both UASp1 and UASp2.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Mutations in the Pho4 binding sites UASp1 and UASp2 affect PHO8 promoter activity very differently. The activity of the wt and mutated PHO8 promoter variants fused to the LacZ gene was measured. The positions of UASp1 and UASp2 and the nucleosomal structure of the repressed promoter are schematically shown at the bottom. The shadowing of the nucleosomes reflects the level of DNA protection against restriction nuclease digestion. Black, 95%; gray, 80%; white, 50% protection (20).

The ability of the two UAS elements to activate transcription was also measured in heterologous constructs with a CYC1 minimal promoter driving lacZ. In agreement with the mutation analysis, UASp2 was found to activate transcription 100-fold upon phosphate starvation in a Pho4-dependent manner, whereas no activation at all was measured with UASp1 (Table II). The results of Fig. 1 and Table II show that induction of the PHO8 promoter upon phosphate starvation is completely dependent on Pho4 binding to the UASp2 element, whereas UASp1 does not seem to be relevant for promoter activity in vivo.

View this table: [in this window] [in a new window]   Table II PHO8 UASp1 is not active by itself when tested with a CYC1 minimal promoter The ability of the PHO8 UAS elements to activate transcription was determined by measuring -galactosidase activity in heterologous constructs with a CYC1 minimal promoter driving LacZ.

Full Activation of the PHO8 Promoter Requires Pho2-- It was reported that, in contrast to acid phosphatase, expression of alkaline phosphatase was Pho2-independent (1). However, our measurements of either alkaline phosphatase activity (not shown) or of a PHO8-lacZ construct show that Pho2 is required for full induction of PHO8 (Table III). In a pho2 strain the activity of the PHO8-lacZ reporter increases only 2-2.5-fold upon phosphate starvation, resulting in considerably lower activity than measured in a wt strain. A similar effect was obtained in a pho80 strain (Table III), which eliminates the possibility of nonspecific effects in the signal transduction pathway upstream of Pho80/85. An indirect effect of Pho2 via Pho4 expression has been ruled out by the results of Yoshida et al. (29), who demonstrated that Pho4 expression is constitutive and independent of the PHO regulatory system.

Pho2 Does Not Contribute Significantly to Pho4 Binding to the PHO8 Promoter-- To determine whether Pho2 participates in the activation of the PHO8 promoter through cooperative DNA binding with Pho4, as demonstrated at the PHO5 promoter (17), binding of Pho4 to UASp2 in a wt and a pho2 strain was examined by in vivo DMS footprinting (Fig. 2A). Under inducing conditions, a clear Pho4-dependent footprint was present. Pho4 binding induces a significant decrease in the reactivity of the G residue at the 3' end of the consensus hexanucleotide, 5'-CACGTG-3', and strongly enhances reactivity of the adjacent G (compare lane 3 with lane 1). Under repressing conditions an intermediate pattern is observed (lane 2) that is more similar but not identical to the pho4 pattern, thus suggesting weak Pho4 binding.

View larger version (61K): [in this window] [in a new window]   Fig. 2.   Pho4 binds to the PHO8 UASp2 element in a Pho2-independent manner. Pho4 binding under repressing (+Pi) and inducing (Pi) conditions (A) or Pho4int binding under inducing conditions (B) to UASp2 in a wt, and a pho2 strain was analyzed by the DMS footprint technique. Strains used are: YS45 (wt), YS46 (pho4), YS42 (pho2), and YS78 (pho2, pho4). All strains carry a CPF1 deletion to prevent binding of Cpf1 to this site (11). Where indicated, the strains express wild type Pho4 or Pho4int from centromere expression plasmids. The sequence of the Pho4 binding site determined by DNaseI footprinting (20) is shown in the box on the side. Guanines are marked by dots and arrows; medium arrows denote guanines whose reactivity with DMS is not changed, whereas the small arrow indicates a guanine that is protected by Pho4, and the big arrow a guanine that becomes hypersensitive to DMS. The lacZ activities of the PHO8 promoter in the combinations indicated for each lane at the top are shown at the bottom of the gel (B).

Pho4 binding under inducing conditions is quite strong in a pho2 strain, and occupancy is close to the level found for a wild type strain (compare lanes 4 and 3 in Fig. 2A). These experiments therefore show that the role of Pho2 in the activation of the PHO8 promoter is not primarily at the level of Pho4 DNA binding, which is in marked contrast to PHO5 (18).

The fact that Pho4 binding to the PHO8 promoter is largely Pho2-independent leads to the prediction that the Pho4 derivative lacking the Pho2 interaction domain (amino acids 200-247), Pho4int, which is unable to activate the PHO5 transcription (18), should bind and activate the PHO8 promoter. This is indeed the case. Pho4int binds strongly to the PHO8 promoter in a Pho2-independent manner and can activate the promoter almost as well as full-length Pho4 in a wt strain (Fig. 2B). In a pho2 strain, Pho4int activates more strongly than wt Pho4, demonstrating the higher activation potential of Pho4int, as previously demonstrated for PHO5 UASp2 (Ref. 6; also see Discussion). We therefore think that the predominant role of Pho2 at the PHO8 promoter is to increase the activation potential of Pho4.

Introduction of the PHO5 UASp1 Element Increases the Transcriptional Activity of the PHO8 Promoter-- The PHO8 promoter is activated essentially through only one UAS element (Fig. 1), whereas two UAS elements, UASp1 and UASp2, cooperatively activate the PHO5 promoter (18). Therefore, the difference in strength of the two promoters could be a consequence of the number and quality of their UAS elements. To address this question, PHO8 promoter derivatives were constructed by replacing the UAS elements with the corresponding elements from PHO5. Introduction of PHO5 UASp1 in place of the corresponding PHO8 element results in 2-fold higher promoter activity (Fig. 3). This result shows that Pho4 can bind to PHO5 UASp1 also in the context of the PHO8 promoter. However, introduction of PHO5 UASp1 into an otherwise inactive PHO8 promoter variant containing a mutated UASp2 element shows that the PHO5 UASp1 element by itself cannot activate transcription at all (Fig. 3). Therefore, the higher activity of the hybrid promoter must actually be the result of cooperative interactions between PHO5 UASp1 and PHO8 UASp2.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Activities of the PHO8 promoter variants containing UAS elements from the PHO5 promoter. The wild type promoter and promoter variants containing UAS elements from the PHO5 promoter are shown schematically at the bottom. The open and the solid rectangles represent PHO8 UASp1 and UASp2, respectively, and the circles represent the corresponding elements from PHO5.

Replacing UASp2 with the PHO5 UASp2 Element Weakens the PHO8 Promoter-- Although activity of the PHO8 promoter is significantly increased by introduction of the PHO5 UASp1 element, it is still far from the activity of the PHO5 promoter. It has been shown that UASp1 at the PHO5 promoter exhibits its full activity through cooperative interactions with UASp2 (18). Therefore, we wished to determine whether the introduction of both UASp1 and UASp2 from PHO5 would result in a stronger cooperative effect and correspondingly higher activity. We first replaced PHO8 UASp2 with PHO5 UASp2. Surprisingly, this substitution essentially inactivated the promoter (Fig. 3), and even the introduction of both PHO5 UAS elements into PHO8 gave only 20% of the activity obtained with the promoter variant containing PHO5 UASp1 and the native UASp2 element (Fig. 3). These results show that in the context of the PHO8 promoter, the PHO5 UASp2 element is much weaker than UASp2 from PHO8, although both elements contain the same consensus hexanucleotide and belong to the same class of high affinity Pho4 binding sites (30).

To determine whether the large difference in activation potential between the UASp2 elements from PHO5 and PHO8 is an intrinsic property of these elements or dependent on the context, both were tested in a minimal CYC1 promoter driving a lacZ reporter. PHO5 UASp2 gave 2.5-fold less activation than the PHO8 UASp2 element (not shown), which only partially explains the very poor activity of the PHO8 promoter derivative containing the PHO5 UASp2 element.

The Low Activity of the PHO5 UASp2 Element Placed in the Context of the PHO8 Promoter Is Not Due to Its Inability to Recruit Pho4-- The poor activity measured with the PHO8 promoter derivative containing the UASp2 element from PHO5 (Fig. 3) could be a result of inefficient binding of Pho4 to this element in the PHO8 promoter context. Therefore, this promoter variant was integrated into the chromosomal locus, and Pho4 binding was determined in vivo by DMS footprinting. As shown in Fig. 4, there is a strong binding of Pho4 to the PHO8 promoter derivative, indistinguishable from Pho4 binding to the same element at the native PHO5 promoter, showing that the inefficiency of this PHO8 promoter variant is not due to the inability of Pho4 to bind to its target.

View larger version (44K): [in this window] [in a new window]   Fig. 4.   Pho4 binds efficiently to PHO5 UASp2 also in the context of the PHO8 promoter. DMS footprint analysis of Pho4 binding to PHO5 UASp2 introduced into the PHO8 promoter (A) or to the same element in its natural location at the PHO5 promoter (B). For other details see Fig. 2.

Pho4 Binding Is by Itself Not Sufficient for Chromatin Remodeling and Activation of the PHO8 Promoter-- Induction of the PHO8 promoter results in chromatin remodeling at the promoter (20). However, in contrast to PHO5, remodeling at the PHO8 promoter is only partial, raising the possibility that the lower activity of the PHO8 promoter might be due to residual repression by chromatin. The introduction of PHO5 UAS elements into the PHO8 promoter gave rise to variants of quite different activities, thereby providing us with the opportunity of correlating Pho4 DNA binding with the extent of chromatin remodeling and promoter activation.

The chromatin structure of the promoter variants containing PHO5 UAS elements was examined by DNase I analysis, and the results are shown in Fig. 5. Introduction of PHO5 UASp1 into the PHO8 promoter resulted in a 2-fold higher activity (Fig. 3) and resulted also in more extensive chromatin remodeling compared with the native promoter (Fig. 5, lanes 9-11 versus lanes 5-7). The increased accessibility of chromatin is confirmed by analysis with restriction enzymes, which showed an approximately 20% increase in accessibility in the region covered by nucleosomes 3 and 2 (not shown). However, introduction of PHO5 UASp1 still did not result in a completely open chromatin structure, as is characteristic of the PHO5 promoter (13), because there is still a significant protection in the region covered by nucleosomes 2 and 3. On the other hand, the chromatin structure of the poorly active PHO8 promoter variant containing PHO5 UASp2 resembles the structure of the repressed promoter found in pho4 cells (Fig. 5, lanes 13-15 versus lanes 1-3), although the in vivo footprinting data showed good binding of Pho4 to the UAS element. Additional introduction of PHO5 UASp1 into this promoter variant increases its activity 2-3-fold and chromatin remodeling to the level of the wt promoter. These data, therefore, strongly support the notion that not Pho4 binding to the promoter but its ability to remodel chromatin is the critical step in PHO8 promoter activation.

View larger version (49K): [in this window] [in a new window]   Fig. 5.   Chromatin remodeling upon induction of the wt PHO8 promoter or of promoter variants containing PHO5 UAS elements. Nuclei isolated from the strains indicated that had been grown in the absence of phosphate were treated for 20 min at 37 °C with increasing DNaseI concentrations (0.25, 0.5, and 1 unit/ml in each case). DNA was isolated, digested with BglII, analyzed on a 1.5% agarose gel, blotted, and hybridized with the PvuII/XhoI fragment (20). The marker lanes contain restriction nuclease double digests of purified genomic DNA with BglII and either EcoR V (band 1), HpaI (band 2), PmlI (band 3), NheI (band 4), RsaI (band 5), HindIII (band 6), or XhoI (band 7). The nucleosomal structure of the promoter under repressive conditions is shown at the bottom with the positions of the restriction sites used to generate marker fragments indicated. Nucleosome 5 (black circle) does not undergo remodeling upon induction, nucleosomes 1, 2, and 3 (gray circles) undergo partial, and nucleosome 4 (white circle) complete remodeling (20).

Substituting the PHO5 for the PHO8 Core Promoter Increases the Activity of the Hybrid Promoter-- The overall activity of promoters not only depends on the number and quality of the UAS elements but also on the core promoter. To assess its contribution in the case of PHO8, we replaced the PHO8 by the PHO5 core promoter. The results presented in Fig. 6 show that this substitution increases the activity of the hybrid promoter by almost a factor of 2, both for the wt promoter as well as a variant containing the two PHO5 UAS elements. On the other hand, a promoter variant that is already very strong with the native core promoter, UASp1 from PHO5 and UASp2 from PHO8, benefits much less from the PHO5 core promoter, and its activity increases only an additional 20-30%. The weak promoter variant containing PHO5 UASp2 was not at all affected by the heterologous core promoter. The reason for this absence of any stimulation may be the difficulty of Pho4 to interact with the core promoter because of the persistence of repressive nucleosomes (Fig. 5, lanes 13-15). This result makes it unlikely that the positive effect of the PHO5 core promoter in the PHO8 context is due to a differential stability of the nucleosome forming over the TATA region.

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Substituting the PHO5 for the PHO8 proximal promoter increases promoter activity. The proximal promoter (positions 142 to 1) of the wt PHO8 promoter and variants thereof was replaced by a 159-bp proximal promoter fragment from PHO5, and activities of these constructs were measured.

The PHO8 Promoter Region Covered by Nucleosomes 3 and 2 Imposes a Repressive Effect-- The chromatin structure analyses and activity measurements of the promoter variants analyzed so far show a close correlation between the extent of chromatin remodeling and promoter activity. In particular, the persistence of nucleosomes 3 and 2 in contrast to nucleosome 4 (Ref. 4 and Fig. 5, lanes 5-7) might impose a repressive effect on the promoter. Therefore, we decided to investigate how a deletion of this promoter region would affect promoter activity. The results presented in Fig. 7. show that this deletion leads to an almost 2-fold increase in promoter activity. The activity of the more powerful promoter variant containing the PHO5 UASp1, which gives a higher level of remodeling at nucleosomes 3 and 2 than the wt promoter (Fig. 5, lanes 9-11 versus lanes 5-7), is also affected, but, as expected, to a lesser extent (25%), consistent with less chromatin repression in that construct. On the other hand, the activity of the weaker promoter derivative containing PHO5 UASp2, which undergoes almost no chromatin remodeling (Fig. 5, lanes 13-15), increases by this deletion 4-fold. These results demonstrate a good correlation between the resistance of chromatin to Pho4-mediated remodeling and the extent of stimulation because of the removal of nucleosomes 2 and 3. They are consistent with the notion that the effect of the deletion is indeed a consequence of alleviating chromatin repression rather than just a distance effect.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   Deletion of the PHO8 promoter region covered by nucleosomes 3 and 2 increases promoter activity. Activities of the wt promoter and variants containing PHO5 UAS elements were compared with corresponding promoter constructs lacking the region normally covered by nucleosomes 3 and 2, PHO8 296 (see schematic at the bottom).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Despite the coordinate regulation of PHO5 and PHO8, the PHO8 promoter is almost 10 times weaker than the PHO5 promoter. Upon induction, both promoters undergo significant remodeling of their chromatin structure, although remodeling at PHO8 results in a chromatin organization that is only partially disrupted (20). In this paper we report a detailed study of the cis- and trans-acting factor requirements for chromatin remodeling and activation of the PHO8 promoter, with the ultimate goal of uncovering the basis for the large difference in inducibility between the two promoters and obtaining new insights into the interplay between transcription factors and nucleosomes in regulating promoter activity.

A Single Pho4 Binding Site Is Responsible for Activation of the PHO8 Promoter-- Two Pho4 binding sites, a low affinity site (UASp1) and high affinity site (UASp2) were mapped at the PHO8 promoter in vitro (20). However, the in vivo mutation analysis described here shows that promoter activity is essentially unaffected by eliminating UASp1, whereas mutating UASp2 completely inactivates the promoter. Furthermore, in contrast to UASp2, UASp1 was unable to activate a heterologous promoter (Table II), supporting the conclusion that this element is not functionally relevant in vivo.

Activation of the PHO8 promoter is accompanied by chromatin remodeling at the promoter, which seems to be prerequisite for activation (20). As expected from the activity data, mutation of UASp1 was without effect on the extent of chromatin opening, whereas mutation of UASp2 completely abolishes chromatin remodeling at this promoter.³ These properties make the PHO8 promoter clearly distinct from PHO5, where both Pho4 sites need to cooperate for chromatin remodeling (14) and promoter activation (18).

The Strength of the PHO8 Promoter Is Determined by the Balance between Transcription Factor UAS Interactions and the Extent of Chromatin Repression-- The low level of activation of the PHO8 promoter could be due to the inability of Pho4 to fully disrupt chromatin through binding to a single site. At PHO5, it was shown that binding of Pho4 to both sites is required for chromatin remodeling, which ultimately results in a promoter activity 10 times higher than that measured when either of the two sites are absent (18), suggesting that the cooperativity between the two sites operates primarily at the level of chromatin remodeling (14). It was therefore interesting to see whether the introduction of an additional Pho4 binding site into the PHO8 promoter would substantially increase chromatin remodeling and promoter activity. The PHO8 promoter variant in which the inactive UASp1 element of PHO8 was replaced by UASp1 from PHO5 showed more extensive chromatin remodeling in the region corresponding to nucleosomes 3 and 2 (Fig. 5, lanes 9-11) but did not give the completely open chromatin structure typical of the PHO5 promoter (13). Also in terms of activity, this promoter variant was still much weaker than the PHO5 promoter and benefited only 2-fold from the additional UAS element. A 2-fold increase in wt PHO8 promoter activity was also observed when the DNA region normally covered by nucleosomes 3 and 2 was deleted, consistent with the repressive role of these nucleosomes. It would therefore be expected that the promoter variant containing UASp1 from PHO5, which shows a lower degree of chromatin repression, would benefit less from the deletion of the two repressive nucleosomes, and indeed activity increased by only 25% by their removal. These results are therefore in accord with the notion that the activity of the PHO8 promoter is determined by the balance between chromatin repression on one hand and the intensity of factor binding to the promoter on the other hand.

Surprisingly, replacing the PHO8 UASp2 element with the corresponding high affinity site from PHO5 almost completely inactivated the promoter. We did find that the PHO5 UASp2 elements was significantly weaker than the PHO8 UASp2 element when the elements were tested independently in front of a CYC1 minimal promoter. This cannot, however, fully explain the very low activity of this element once placed in the context of the PHO8 promoter. The poor activation potential of the PHO8 promoter containing the PHO5 UASp2 element is not due to inefficient Pho4 binding, because in vivo footprinting experiments revealed that the binding of Pho4 to the PHO5 UASp2 element introduced into the PHO8 promoter is indistinguishable from binding to the same element in the natural context. Chromatin analyses showed, however, that Pho4 is bound but unable to bring about chromatin remodeling at this promoter variant (Fig. 5, lanes 13-15). This is quite remarkable because it is the first demonstration that binding of a full-length Pho4 molecule to a UAS element does not result in disruption of the adjacent nucleosomes. We have previously shown for the PHO5 promoter that the ability of Pho4 to remodel chromatin is a property of its activation domain. (15). Therefore, the very poor activity of the PHO8 promoter variant containing the PHO5 UASp2 element might be explained by the inability of Pho4 to expose its activation domain properly, i.e. in the way required to trigger chromatin remodeling at this promoter. In this context, it is interesting to point out our recent finding that activation of the PHO8 promoter requires the SWI/SNF and the SAGA complex (31). There is considerable experimental evidence implicating these two protein complexes in chromatin remodeling (32-35). In the absence of Snf2, an essential component of the SWI/SNF complex, chromatin remodeling at the PHO8 promoter is indeed completely abolished, and only very limited remodeling was shown to occur in the absence of the SAGA component Gcn5 (31). Furthermore, efficient binding of Pho4 to UASp2 is observed in both mutant strains, showing that Snf2 and Gcn5 are required for chromatin remodeling at a step subsequent to Pho4 binding (31). In contrast, remodeling at the PHO5 promoter under fully inducing conditions does not require Snf2 (36) nor Gcn5 (34). The inability of Pho4 bound to the PHO5 UASp2 element in the PHO8 promoter to remodel chromatin might therefore be due to its inability to productively interact with components of the SWI/SNF and/or SAGA complexes.

At the PHO8 Promoter the Primary Role of Pho2 Is Not to Assist in Pho4 Binding-- Another feature that makes the PHO8 different from the PHO5 promoter is its significant activity in the absence of Pho2. We have previously shown that the PHO5 promoter is absolutely Pho2-dependent and that Pho2 plays a dual role in the activation process. It is critical for binding of Pho4 to UASp1, whereas at UASp2, where binding of Pho4 is not absolutely Pho2-dependent, Pho2 is mainly required for the ability of Pho4 to transactivate (18). Here we have shown that Pho4 can efficiently bind to PHO8 UASp2 without Pho2, but the presence of Pho2 significantly increases PHO8 activation (Table III). Therefore, the PHO8 promoter is partially Pho2-dependent, a conclusion also confirmed by activity measurements with Pho4int, a Pho4 derivative lacking the Pho2 interaction domain. Pho4int binds and activates the PHO8 promoter in a Pho2-independent manner almost to the same level as wt Pho4 (Fig. 2B). We have previously shown that activation by Pho4int for PHO5 UASp2 is higher than by wt Pho4 in the absence of Pho2. This phenomenon was explained by assuming the presence of a repressive domain in Pho4 that is counteracted by interactions with Pho2 under physiological conditions but has been excised in Pho4int (18). The PHO8 promoter is ideally suited to test this hypothesis, and indeed, the level of activation by Pho4int is in complete agreement with this concept.

It is important to realize that unlike the PHO5 promoter, PHO8 does not require Pho2 for chromatin remodeling (20). This is consistent with the ability of Pho4 to bind to the PHO8 but not the PHO5 promoter in the absence of Pho2. Furthermore, overexpression of Pho4 in a pho2 strain, which relieves the Pho2 requirement for Pho4 binding to its target and thus makes the PHO5 promoter more similar to the PHO8 promoter, results in fully open chromatin but leads to only 20-25% activity (37). Therefore, this finding is also in agreement with the concept that once Pho4 is bound to a promoter, Pho2 increases its activation potential rather than its ability to orchestrate chromatin remodeling.

The Differential Pho2 Requirement Might Play a Role in Fine Tuning the Levels of PHO5 and PHO8 Expression under Repressing Conditions-- The fact that PHO5 and PHO8 differ in the extent to which they require Pho2 can explain the significantly higher activity of PHO8 under repressing conditions (Table I). The recently published investigation of the role of particular serine residues in the regulation of Pho4 activity showed that phosphorylation regulates Pho4 not only by controlling its nuclear localization but also by a second mechanism, regulation of the interactions between Pho4 and Pho2. Phosphorylation of Ser¹¹⁴ and Ser¹²⁸ is necessary and sufficient for nuclear export of Pho4, whereas phosphorylation of Ser²²³ negatively regulates its ability to interact with Pho2 (38). At PHO5, these two mechanisms working together ultimately result in full repression of PHO5 transcription, whereas each one by itself brings about only partial repression. Our finding that activation of the PHO8 promoter is only weakly Pho2-dependent actually predicts that Pho4 phosphorylation should not result in full repression of PHO8 because the second mechanism would only have a slight effect on PHO8. The differential Pho2 dependence therefore explains the paradoxical finding that the much weaker promoter is more active at high phosphate conditions.

What Is the Basis for the Difference in Activation between PHO5 and PHO8?-- Clearly the most obvious difference between PHO5 and PHO8 that has come out of the present study is the absence of a functional UASp1 element at PHO8. This difference is partly responsible for the much lower activation of PHO8 compared with PHO5. However, even with the complete UASp1 element from PHO5 replacing its inactive counterpart, activation of PHO8 is still significantly less than of PHO5. In this promoter variant remodeling of nucleosomes 2 and 3 is still incomplete. These nucleosomes even persist under conditions of Pho4 overexpression.² It is important to realize that nucleosome 3 is immediately adjacent to the powerful UASp2 but is still refractory to complete remodeling under all these conditions. In contrast, the response of the PHO5 promoter follows an all or nothing course. For example, with only a single UAS element present at the PHO5 promoter (UASp2), chromatin remains completely closed at normal Pho4 expression level. Upon overexpression the promoter is remodeled, however, and then undergoes precisely the same four-nucleosome transition as the wt promoter at normal Pho4 level (11). At the PHO8 promoter nucleosome remodeling is incomplete and therefore presumably limiting even under optimal conditions. We therefore suggest that the most consequential difference between PHO8 and PHO5 is the presence of persistent nucleosomes that resist remodeling. Consistent with this interpretation is the absolute Snf2 requirement at PHO8 for chromatin remodeling and a significant Gcn5 requirement (31), whereas neither is required at PHO5 (34).

	ACKNOWLEDGEMENTS

We thank J. Svaren for help in the early stage of this work and continuous discussions, Philip Gregory for discussions and comments on the manuscript, and D. Blaschke for expert assistance.

	FOOTNOTES

^* This work was supported by Deutsche Forschungsgemeinschaft Grant SFB 190, Fonds der Chemischen Industrie (to W. H.), and by Pliva, Zagreb (to S. B.).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.

Present address: Lab. of Biochemistry, Faculty of Food Technology and Biotechnology, University of Zagreb, Pierottijeva 6, 10000 Zagreb, Croatia.

^§ To whom correspondence should be addressed: Adolf-Butenandt-Institut, Molekularbiologie, Universität München, Schillerstr. 44, 80336 München, Germany. Fax: 49-89-5996440; E-mail: hoerz@bio. med.uni-muenchen.de.

Published, JBC Papers in Press, May 2, 2000, DOI 10.1074/jbc.M001409200

² M. Münsterkötter, S. Barbaric, and W. Hörz, unpublished data.

³ M. Münsterkötter, unpublished data.

	ABBREVIATIONS

The abbreviations used are: bp, base pair(s); wt, wild type; DMS, dimethyl sulfate.

	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:

				P. Korber, S. Barbaric, T. Luckenbach, A. Schmid, U. J. Schermer, D. Blaschke, and W. Horz The Histone Chaperone Asf1 Increases the Rate of Histone Eviction at the Yeast PHO5 and PHO8 Promoters J. Biol. Chem., March 3, 2006; 281(9): 5539 - 5545. [Abstract] [Full Text] [PDF]

				C. B. Hertel, G. Langst, W. Horz, and P. Korber Nucleosome Stability at the Yeast PHO5 and PHO8 Promoters Correlates with Differential Cofactor Requirements for Chromatin Opening Mol. Cell. Biol., December 15, 2005; 25(24): 10755 - 10767. [Abstract] [Full Text] [PDF]

				C. D. Carvin, A. Dhasarathy, L. B. Friesenhahn, W. J. Jessen, and M. P. Kladde Targeted cytosine methylation for in vivo detection of protein-DNA interactions PNAS, June 24, 2003; 100(13): 7743 - 7748. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 275/30/22678    most recent M001409200v1 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