THE JOURNAL OB BIOLCGICAL CHEMISTRY 0 1994 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 269, No. 13, Issue of April 1, pp. 9790-9797, 1994 Printed in U.S.A.

The Upstream Repression Sequence from the Yeast Enolase Gene ENOl Is a Complex Regulatory Element That Binds Multiple Tkans-acting Factors Including REBl*

(Received for publication, June 1, 1993, and in revised form, December 6 , 1993)

Andrew A. Carmen$ and Michael J. Holland From the Department of Biological Chemistry, School of Medicine, University of California, Davis, California 95616

Cis-acting sequences that modulate ENOl URS (up- stream repression site) element activity were identified by base pair substitution mutagenesis. Base substitu- tion mutations within three distinct regions of the 125- base pair URS element caused partial loss of URS activ- ity in vivo. AURS element containing all three mutations was inactive. A binding site for the yeast REBl protein was identified near the 5‘ terminus of the ENOl URS element. Base substitution mutations that disrupted REBl binding in vitro caused a 30% loss of URS activity in vivo. A second DNA binding activity was identified which also bound near the 5‘ terminus of the URS ele- ment. This latter binding activity was not antigenically related to REBl nor was binding of this activity affected by base substitution mutations that abolished REBl binding. Base substitution mutations within a second region of the ENOl URS element caused a 38% loss of URS activity in vivo. The nucleotide sequence of this latter region is very similar to essential sequences within the URS elements from the yeast CARl and SSAl genes, respectively. Base substitution mutations within a third region near the 3’ terminus of the ENOl URS element caused a 70% loss of URS activity in vivo. These latter sequences bound a partially purified factor that was distinct from REB1. These results showed that ENOl URS element activity was modulated by multiple cis-acting sequences that bound distinct trans-acting factors.

Transcription of the yeast enolase genes ENOl and EN02 is activated by complex cis-acting regulatory sequences (UAS el- ements)’ (1, 2) that bind distinct trans-acting factors (3, 4). ENOl gene expression is also repressed in cells grown in a medium containing glucose through the action of an upstream repression site (URS element) (2). Previous deletion mapping analysis of the ENOl URS element showed that sequences located between positions -181 and -143 relative to the ENOl transcriptional initiation site are essential for URS element activity and that sequences between positions -213 and -181 are required for maximum URS activity (2).

* This research was supported in part by United States Public Health Service Grant GM30307 from the National Institutes of Health and a grant from the March of Dimes Birth Defects Foundation. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Recipient of a predoctoral fellowship from the McKnight Founda- tion. Present address: Molecular Biology Institute, University of Cali- fornia, Los Angeles, CA 90024. ’ The abbreviations used are: UAS, upstream activation site; URS,

upstream repression site; EBF, enolase binding factor; wt, wild-type; HMRB, the ABFl binding site within the yeast silent mating type locus HMR.

The DNA sequence of a portion of the ENOl URS element is quite similar to a sequence found to be required for the activity of URS elements located in the 5’-flanking regions of the yeast CARl gene (5, 6) and SSAl gene (7). A double-stranded oligo- nucleotide corresponding to this conserved URS sequence i n the SSAl gene (7) specifically binds an activity in yeast whole cell extracts. A binding activity has also been reported that interacts specifically with an oligonucleotide corresponding to the CARl URS element (6). Deletion mapping analysis of the ENOl URS element showed that sequences that are similar to those found in the SSAl and CARl URS elements are required for but are not sufficient for maximum ENOl URS element activity (2). In this report we show that the ENOl URS element is composed of at least three cis-acting sites that interact with distinct DNA binding activities. One of these DNA binding activities corre- sponds to REBl protein (8), a factor that has been implicated in activation of transcription by RNA polymerase I1 as well as ter- mination of transcription by RNA polymerase I.

EXPERIMENTAL PROCEDURES Material~-’~~I-Labeled donkey anti-rabbit IgG, 32P-labeled nucleoti-

des, and terminal deoxytransferase were purchased from Amersham Corp. Thermus aquaticus DNA polymerase (Amplitaq) for polymerase chain reactions was purchased from Perkin-Elmer Cetus. Polyclonal antibody directed against REBl (8 ) and preimmune sera were a gener- ous gift from Dr. Jon Warner and Dr. Bernice Morrow, Albert Einstein College of Medicine.

Strains and Growth Conditions-Saccharomyces cereuisiae strain S173-6B (a leu23 leu2-112 his3-A1 trpl-289 ura3-52) was provided by F. Sherman, University of Rochester. S. cereuisiae strain S173-LA was identical to strain S173-6B except that it carried a deletion of 90% of the ENOl coding sequences (9). Yeast strains were grown at 30 “C in YP medium (1% yeast extract, 2% peptone) or in a defined medium con- taining 0.67% Difco yeast nitrogen base without amino acids were supplemented with 2 mg/ml uracil, leucine, and tryptophan. Histidine (2 mg/ml) was included for strains lacking a functional HIS3 gene. Carbon sources were 2% glucose or 2% glycerol plus 2% lactate. Cells were harvested in early log phase of growth (A,,, ”~ = 1.0) or as noted.

DNA Probes-A 125-base pair Sal1 restriction endonuclease cleavage fragment containing ENOl URS element sequences extending from position -241 to -126 was isolated from plasmid peno46/HIS3 (-301/ -241, -126/-121) and subcloned into the Sal1 site in the polylinker region of plasmid pUC12. This latter plasmid was cleaved in the polylinker at sites upstream or downstream from the ENOl URS ele- ment with either SacI or PstI, respectively. Cleaved DNA was 3’ labeled with terminal deoxytransferase and [a-32PldATP according to the pro- tocol provided by Amersham Corp. The 32P-labeled DNA was digested with either PstI or SacI and the fragment containing the ENOl URS element was purified by standard protocols. These latter DNA probes were used for gel mobility shift assays, DNase I footprinting, and methylation interference analysis as described below.

binding site within the UAS, element of the EN01 gene (ENOlAJAS2) Double-stranded oligonucleotides corresponding to the EBFl (REBl)

(RIB011 (5’-CTATGATCCGGGTAAAAACATGT-3’), the REBl binding the REBl binding site from the yeast ribosomal RNA enhancer element

site from the yeast ribosomal RNA enhancer element containing the in-

(3) (5”CTCGGATTAGCAGATAACCCGCCTAGAAGCGGTATTTTT-3‘),

9790

Yeast ENOl URS Element 9791

dicated (underlined) transversion mutations that block REBl binding in vitro (RIBO11) (5’-CTATGATCATTGTAAAAACATGT-3’), portions ofthe EN01 URS element corresponding to sequences extending from position -228 to -193 (URSl), -191 to -162 (URSZ), and -170 to -142 (URSS), and URS3-mut3, a mutant version ofURS3 containing transversion mu- tations described below, were synthesized. One strand ofeach oligonucle- otide was 5‘ labeled with T, polynucleotide kinase and [y-”P[ATP and then annealed with a 1.2-fold molar excess of the unlabeled complemen- tary strand. These double-stranded oligonucleotide probes were used in gel mobility shift assays. Unlabeled double-stranded oligonucleotides were used in competition DNA binding assays.

The 125-base pair EN01 URS element was cloned into the phage vector M13-mplO to generate site-directed mutations as described be- low. DNA probes containing specific mutations within the ENOI URS element were labeled to identical specific activity utilizing a 5’ “P- labeled M13 universal primer and the polymerase chain reaction and a second unlabeled primer for double-stranded DNA synthesis. These probes were used in gel mobility shift assays.

Mono s Chromatography ofS100 Extracts-Whole cell SI00 extracts were prepared by ammonium sulfate precipitation as described previ- ously (10). Aliquots of SlOO extracts containing 8-10 mg of protein, as determined by the method of Warburg and Christian (111, were frac- tionated by Mono S cation exchange chromatography as described pre- viously (3). Chromatography was performed at room temperature using a Pharmacia Mono S HR5/5 fast protein liquid chromatography column equilibrated with buffer containing: 20 m v HEPES, pH 7.4,50 mM KCl, 0.2 mM EDTA, 2 rn dithiothreitol, and 17% glycerol. DNA binding activities were eluted with a 27-ml linear KC1 gradient (50-500 mu KCI) in the same buffer. The column was then washed with 2 ml of buffer containing 1.0 M KCl. One-milliliter fractions were collected and stored a t -80 “C. DNA binding activities were localized by gel mobility shift assay as described previously (4) using the indicated probes and

A. URS-wt.

electrophoresis in 2.0% agarose gels. DNAIProtein Binding Assays-Gel mobility shift assays, DNase I

footprint analysis, and methylation interference assays were performed as previously described (4) with the indicated DNA probes. In some cases, the DNA binding activities obtained after Mono S chromatogra- phy were concentrated by ultrafiltration using an Amicon Centricon 10 microconcentrator. Gel mobility shift DNA competition assays were as previously described (3). Gel mobility supershift assays were performed as described for gel mobility assays (4) except that following the 15-min binding reaction, the indicated dilutions of polyclonal antisera directed against REBl or preimmune sera were added to the binding reactions and the reactions were incubated for an additional 10 min. Protein concentration was determined using the Bio-Rad protein assay reagent according to the manufacturer’s protocol. DNA sequencing reactions were performed according to the procedure of Maxam and Gilbert (12).

Site-directed Mutagenesis-The 125-base pair Sal1 restriction endo- nuclease cleavage fragment containing the EN01 URS element was subcloned into the phage vector M13-mplO and propagated as described by Messing (13). The URS-mutl, URS-mut2, and URS-mutl+2 muta- tions were generated by the in uitro site-directed mutagenesis tech- nique described by Innis and McCormick (14) except that T, DNA poly- merase was used instead of DNA polymerase I Klenow fragment. URS- mut 1 contains a triple transversion mutation that converts the wild- type sequence CCC between positions -209 and -207 to AAA. URS- mut2 contains a triple transversion mutation that converts the wild- type sequence CCT between positions -173 and -171 to AAA. URS- mut3 contains a triple transversion mutation that converts the wild- type sequence CCC between positions -155 and -153 to AAA. The URS-mut3, URS-mutl+3, URS-mut2+3, and URS-mutl+2+3 mutations were generated utilizing the polymerase chain reaction. M13-mplO con- taining the wild-type EN01 URS fragment and the URS-mutl, URS- mut2, and URS-mutl+2 fragments were used as templates in the pres-

E. IJRS1

3 3 4 3 6 38 40 42 44 46 48 50 52 54 5 6 58 6 0 SlOO 36 38 40 42 44 46 48 M 52 54 56 58 60 :”’

B. URS-mutl F. URS2

32 34 3 6 38 4 0 42 44 46

C. RIB01

48 50 52 54 5 6 58 60 SI00 36 38 40 42 44 46 48 M 52 Y 56 58 60

G. URS3

I ”.

1- 4 6

D. E N O l l U A S 2

4 8 5 0 5 2 I 4 56 5 8 6 0 5 1 0 0 36 38 40 42 44 46 48 S 52 Y 56 58 €J’ :* ’

H. URS3-mut3

” - 0 . ..

32 34 36 30 40 42 44 4 6 40 50 52 54 56 50 60 SI00 36 38 40 42 44 46 a8 M 52 54 56 58 6 0 ’ : : ‘

Fraction Number Fraction Number FIG. 1. Mono S chromatography of ENOl U R S element binding activities. Mono S chromatography was performed with SlOO extracts

isolated from S. cereuisiae strain S173-6B grown in media containing glucose as carbon source. DNA binding activities were eluted from a Mono S column with a linear KC1 gradient (50-500 mM), followed by step elution with 1 M KC1 (fraction 60). DNA binding activities were monitored by gel mobility shift assay using the indicated double-stranded DNA probes. Gel mobility shift assays contained 2 fmol of 5’-end labeled DNA probe (specific activity, 5 x lo6 dpdpmol), 0.1-0.2 pg of Mono S fractionated protein or 7 pg of crude SlOO protein, and 12.5 pg/ml poly(dI-dC) DNA. A, URS-wt wild-type ENOIAJRS element sequences extending from position -241 to -126. B, URS-mutl: ENOIAJRS element sequences extending from position -241 to -126 containing triple transversion mutations at positions -207, -208, and -209 (Fig. 4). C, RIBO1: a double-stranded oligonucleotide containing the REBl binding site within the yeast ribosomal RNA enhancer element. D, ENOIIuAS2: a double-stranded oligo- nucleotide containing the EBFl (REB1) binding site from the ENOI UAS2 element. E, F, and G , URS1, URS2, and URS3: double-stranded oligonucleotide probes corresponding to portions of the EN01 URS element extending from position -228 to -193, -191 to -162, and -170 to -142, respectively. H , URS-mut3: a double-stranded oligonucleotide probe corresponding to URS element sequences extending from position -170 to -142 containing a triple transversion mutation a t positions -153, -154, and -155 (Fig. 4). The arrows indicate the positions of complexes formed with EBFl (REB1) binding activity.

9792 Yeast ENOl URS Element ence of an M13 universal primer located upstream from the 5’ terminus of the EN01 URS sequences and a mutagenic primer (5”AGGTCGAC

the triple point mutation is underlined) comzmentary to sequences near the 3‘ terminus of the ENOl URS element. The polymerase chain reactions were performed according to the conditions suggested by Per- kin-Elmer Cetus. The presence of all mutations in the respective ENOl URS fragments was confirmed by DNA sequencing according to the method of Sanger et al. (15).

Each of the mutant ENOl URS SalI fragments was subcloned into the unique SalI site in plasmid peno46/HIS3(-301/-121) as described previously (2). Plasmids containing the URS fragment mutations in the correct orientation relative to the EN01 transcriptional initiation site were isolated. Integration of these latter plasmids at the enol locus of S. cereuisiue strain S173-LA and analysis of ENOl gene expression were as previously described (2). Synthetic double-stranded oligonucleotides designated; URS1-mutl, URSB-mut2, and URS3-mut3 were identical to URS1, URS2, and URS3, respectively, except that they contained the indicated triple point mutations shown in Fig. 4. These latter oligo- nucleotides were used in gel mobility shift assays.

CATGTGTGCCAGAAAAGGCCACAGT”CCTTTGAGAGGCATAC-3‘,

RESULTS

Binding of Yeast Cellular Factors to the ENOl URS Element-Sequences that are sufficient for EN01 URS activity were mapped between positions -241 and -126 relative to the ENOl transcriptional initiation site (+1) (Ref. 2; data not shown). A DNA fragment corresponding to these sequences was isolated from plasmid peno46/HIS3(-301/-241:-126/-121) and end-labeled as described under “Experimental Procedures.” To identify factors that bind to the URS element, yeast whole cell extracts were fractionated by Mono S chromatography and the fractions were analyzed by gel mobility shift assays using the 32P-labeled URS probe (Fig. lA). Several DNA binding activi- ties were identified. A major binding activity eluted a t 0.35 M KC1 (fractions 49-53) which formed a complex with the URS probe. Complexes also formed between the URS probe and binding activities that eluted at 0.17 (fractions 3 9 4 1 ) and 0.4 M KC1 (fractions 52-55).

DNase I Footprint and Methylation Interference Analysis- Gel mobility shift assays of the URS probe incubated with unfractionated yeast whole cell extract (S100) or fraction 50 obtained after Mono S chromatography (Fig. lA) are shown in Fig. 2A. DNase I protection analysis of the URS probe incu- bated with either whole cell extract (S100) or Mono S fraction 50 revealed similar footprints between positions -195 and -223 on one strand (Fig. 2B) and -197 to -226 on the other strand (data not shown). Single base protections were also observed for unfractionated cell extract (S100) and Mono S fraction 50 at positions -190, -182, -176, -167, -163, -157, and -146 (Fig. 2B 1. The single base protection pattern was highly reproducible and occurred commensurate with formation of the extended footprint between positions -195 and -226.

Methylation interference assays were performed with the U R S probe incubated with unfractionated whole cell extracts isolated from cells grown in media containing glucose (S100d) and glycerol plus lactate (SlOOgl) and with fraction 50 obtained after Mono S chromatography (Fig. IA). Critical guanine resi- due contact sites were identified a t positions -204, -205, -207, -208, -209, and -214 with fraction 50 after Mono S chroma- tography. The same guanine residue contact sites were visible with SlOO extract prepared from cells grown in a medium con- taining glycerol plus lactate as carbon source but difficult to visualize with SlOO extract prepared from cells grown in a medium containing glucose (Fig. 3). All of these guanine resi- due contact sites are within the extended DNase I footprint shown in Fig. 2B. The positions of the DNase I protected se- quences and the guanine residue contact sites are summarized in Fig. 4.

The URS Element Binds a Factor That Binds to the ENOl UASZ Element and the Yeast rRNA Enhancer Element-A factor

A .

B. - 2 2 6

- 1 95- I .190-

-182-

- 1 7&

- 1 6 L - 1 6 L

- 1 51_,

interacts with the ENOl URS element. Panel A, gel mobility shift FIG. 2. DNase I footprinting of a major binding activity that

assays were performed with a ““P-labeled DNA probe corresponding to the ENOl URS element extending from position -241 to -126 (URS-wt) prepared as described under “Experimental Procedures.” DNA binding reactions contained 8 fmol of probe and approximately 0.7 pg of protein from fraction 50 after Mono S chromatography as described in the legend to Fig. 1 or 20 pg of a yeast SlOO extract isolated from S. cereuisiae strain S173-6B grown in media containing glycerol plus lac- tate as a carbon source. Panel B, aliquots of each of the DNA binding reactions described in panel A were subjected to DNase I protection analysis as described under “Experimental Procedures.” DNase I diges- tion and polyacrylamide gel electrophoresis were performed as de- scribed by Holland et al. (4). Aliquots of DNA binding reactions lacking protein extract ( - E ) were digested with DNase I and electrophoresed in parallel lanes on the polyacrylamide gel along with DNA sequencing reactions. The location of a DNase I footprint extending from position -226 to -195 is indicated by the bar. Single base protections are indi- cated by the arrows.

designated EBFl (enolase binding factor) binds specifically to EN01 sequences extending from position -452 to -472 (3). These latter sequences are required for the activity of an up- stream activation site designated ENOIIUAS, (3). Comparison of the EBFl binding site within the ENOIIUAS, element and the binding site within the URS element extending from posi- tion -226 to -195 revealed significant similarities (Fig. 5). Five of six critical G contact sites found in the URS element are also present in the EBFl binding site in the ENOl/UAS, element (3). Interestingly, the nucleotide sequences of the factor binding sites within the URS element and the ENOlIUAS2 element were similar to the nucleotide sequence of the REBl binding site within the yeast ribosomal RNA enhancer element (Fig. 5) (8,

Yeast ENOl URS Element

Bottom Strand TOP St rand

"""- m e r m

""""

"0""

9793

f 00% U B U B U B G G U B U B U B 000, z. i, - - - "- 9 9 % a

9 d: $ % 3, 06,

% 00 ??? 0,

FIG. 3. Identification of critical guanine residue contact sites for a major binding activity that interacts with the ENOl URS element. Methylation interference analysis was performed with the URS-wt probe using fraction 50 obtained after Mono S chromatography of a yeast whole cell extract as described in the legend to Fig. 1 and yeast SlOO extracts isolated from strain S173-6B grown in a medium containing glucose (S100d) or glycerol plus lactate (SlOOgl). DNA binding reactions and gel mobility shift assays were as described in the legend to Fig. 2. Unbound (U) probe and bound ( B ) probe were extracted from agarose gels, cleaved, and analyzed on polyacrylamide gels as described by Holland et al. (4). Sequencing reactions were electrophoresed in parallel lanes as indicated. Top and bottom strand refers to the labeled strand of the DNA probe. The positions of guanine residues that are important for complex formation are indicated by the arrows.

S ' - T C G A C T C T C C C T ~ G G A A C G T G SSAliURS

A G G T A T G C C T C CLCZCEGGAAACTGTG 3 ' ()I

I I I I I I I I I -225 -220 -210 -200 - 1 9 0 - 1 80 - 1 70 - 1 60 - 1 l 3

- f o o l p r l n l 7 t l n g l m b a n d p r o t m c t l o n

s l te -d l rec ted mutant

FIG. 4. Summary of the DNase I footprinting and methylation interference data for the ENOl URS element. The extended DNase I footprint, single base protections, and guanine residue contact sites for a major binding activity that interacts with the ENOl URS element are indicated. The bored sequences correspond to the URSl consensus sequence ( 6 ) within the ENOl URS element and within the URS element from the yeast SSAl gene. The asterisks indicate the positions a t which transversion mutations were introduced as described under "Experimental Procedures."

16), suggesting that all three sites may bind REBl protein. Con- sistent with this hypothesis, the major binding activity that in- teracted specifically with the URS-wt fragment and the URSl double-stranded oligonucleotide (Fig. 1, A and E ) co-chromato- graphed after Mono S chromatography with EBFl which bound the ENOINAS, oligonucleotide (Fig. 1 D ) and REBl which bound the RIBOl double-stranded oligonucleotide (Fig. 1C).

To determine if REBl binds to the ENOl URS element, com- petition DNA binding assays were performed with a 32P-labeled

RIBOl double-stranded oligonucleotide probe which contains the REBl binding site from the yeast ribosomal RNA enhancer element (8, 16). As illustrated in Fig. 6 A , a single gel mobility shift complex was formed between the RIBOl probe and a binding activity in fraction 50 after Mono S chromatography (Fig. IA). Formation of this complex was competed by the un- labeled oligonucleotides RIBO1, URS1, and ENOl/UASZ (Fig. 6A). RIBO11, a derivative of RIBOl which contains a triple transversion mutation that abolishes REBl binding in vitro

9794 Eas t ENOl URS Element

ENOUURS -199 G A C T B ~ T A G C -218 * * * * *

E N O I / U A S Z -470 T T C T F T G C T -451

R I B O l -162 T A T G I A T C C G G G T A A A N A C A T -143 FIG. 5. Nucleotide sequence comparison of the factor binding

site within the ENOl U R S element, the EBFl binding site within the ENOl UAS2 element, and the REBl binding site within the yeast ribosomal RNA enhancer element. The nucleotide sequences of the binding sites within the EN01 URS element (ENOIAJRS) and UAS2 element (ENOIAJAS2) were aligned with the nucleotide se- quence of the REBl binding site within the yeast ribosomal RNA en- hancer element (RIBOl). The boxed region indicates conserved se- quences within the REBl binding sites. Asterisks indicate the positions of guanine contact sites for factor binding determined by methylation interference analysis.

(data not shown), and URS1-mutl, a derivative of URSl which contains a triple transversion mutation that abolished binding of the major binding activity to the URS element (Fig. lB), failed to compete complex formation (Fig. 6A). These results suggest that REBl binds to the URS element and the ENOlI UAS2 element.

Similar competition DNA binding assays were performed with 32P-labeled URSl oligonucleotide. Two complexes were formed between the URSl probe and binding activities in frac- tion 50 after Mono S chromatography (Fig. 6B). Formation of the larger complex was competed by unlabeled RIBO1, URS1, and ENOlIUASB oligonucleotides, whereas formation of the smaller complex (I) was competed by unlabeled URSl and ENOlKJAS2 oligonucleotides but not the RIBOl oligonucle- otide (Fig. 6B ). These results suggest that REBl and a second binding activity bind to the URSl and ENOlIUAS2 elements.

Gel mobility supershift assays were performed with 32P-la- beled RIBOl double-stranded oligonucleotide probe, binding activities from fraction 50 after Mono S chromatography, and a polyclonal antibody directed against REBl protein. As illus- trated in Fig. 7, the gel mobility shift complex formed with the RIBOl probe was supershifted by the REBl polyclonal anti- body. The same dilutions of preimmune sera failed to supershift the complex and the REBl antibody failed to supershift a com- plex formed between the HMRB probe and partially purified ABFl(3,4). The REBl polyclonal antibody, but not preimmune sera, supershifted the larger complex formed with the URSl and ENOlIUAS2 oligonucleotides (Fig. 7). These results con- firm that REBl binds to the URS and ENOIIUASS elements. The REBl polyclonal antibody failed to supershift the smaller complex formed with the URSl and ENOIIUAS2 probes, sug- gesting that the binding activity responsible for formation of the smaller complex is unrelated to REB1.

Substitution Mutations within Three Regions of the ENOl URS Element Reduced URS Element-dependent Repression of ENOL Expression in Viuo-Deletion mapping analysis of the ENOl URS element showed that removal of the sequences required for REBl binding caused only partial loss of URS element activity in uiuo (2). To further define the relationship between REBl binding and URS activity, transversion muta- tions were introduced at three guanine bases implicated in contacting REBl by methylation interference assays (Fig. 3). A URS element probe carrying this triple transversion mutation (URS-mutl) failed to bind REBl in vitro (Fig. 1B). Since dele- tion mapping analysis showed that sequences extending down- stream from the REBl binding site to position -143 were also required for ENOl URS element activity in vivo (2), triple transversion mutations were introduced into ENOl URS se- quences that are similar to the SSAl (7) (Fig. 4) and CAR1 (5, 6) URS elements and a third site shown by deletion mapping

analysis (2) to be essential for ENOl URS element activity in uiuo. These triple transversion mutations, designated URS- mutl , URS-mut2, and URS-mut3, respectively, are described under “Experimental Procedures.”

Sal1 restriction fragments corresponding to ENOl URS ele- ment sequences extending from positions -241 to -126 contain- ing each triple transversion mutation alone or in combination were ligated at a unique SalI cleavage site between positions -301 and -121 in plasmid peno46/HIS3(-301/-121). Plasmid peno46/HIS3(-301/-121) lacking a functional URS element and the plasmid derivatives containing mutant URS elements were integrated at the ENOl locus of strain S173-LA which carries an enol null mutation. Expression of these mutant ENOl genes was monitored by Western blotting analysis. The steady state level of enolase 1 polypeptide synthesized from the mutant ENOl genes was measured relative to the enolase 2 polypeptide synthesized from the chromosomal EN02 gene which is induced 20-fold in cells grown in a medium containing glucose uersus glycerol plus lactate by densitometry as previ- ously described (2).

An ENOl gene containing a deletion mutation extending from position -301 to -121 which removed the URS element was expressed at a level 17-fold higher than the wild-type ENOl gene in cells grown in a medium containing glucose (Fig. 8). When a SalI fragment containing the wild-type ENOl URS element was ligated between positions -301 and -121, the gene was expressed at a level 1.5-fold higher than the wild-type ENOl gene in cells grown in a medium containing glucose. The URS-mutl, URS-mut2, and URS-mut3 mutations alone caused 6-, 7.5-, and 14-fold glucose-dependent induction of ENOl ex- pression, respectively (Fig. 8). These latter mutations had only modest effects on ENOl expression in cells grown in a medium containing glycerol plus lactate (Fig. 8). Since REBl binding to the URS element in vitro was abolished by the URS-mutl mu- tation (Fig. 1B ), the results shown in Fig. 8 suggest that REBl plays an important role in modulating URS element activity. Glucose-dependent induction of ENOl expression was further elevated (15-17-fold) when the URS element carried combina- tions of two mutations (Fig. 8). Finally, URS element activity was abolished when all three mutations were present in the URS element. The affects of the URS element base substitution mutations are consistent with those obtained by deletion map- ping analysis (2) and demonstrate that the ENOl URS element is a complex regulatory element.

To determine if additional binding activities interact with the URS element, gel mobility shift assays were performed with the URSB oligonucleotide probe corresponding to URS se- quences extending from position -191 to -162 and fractions obtained after Mono S chromatography of a yeast whole cell extract. This latter probe includes sequences which are similar to the URS element from the yeast SSAl gene (Fig. 4). As illustrated in Fig. lF, the URSB probe formed minor complexes with several fractions. Similar assays were performed with a URS2-mut2 probe which is a derivative of URS2 containing the transversion mutations at positions -173, -172, and -171. None of the minor binding activities observed with the URS2 probe were altered by the URS2-mut2 triple transversion mu- tation (data not shown). None of the gel mobility shift com- plexes formed with the URS2 probe were supershifted with the REBl polyclonal antibody (data not shown).

Gel mobility shift assays were also performed with the URS3 oligonucleotide probe corresponding to URS sequences extend- ing from position -170 to -142. As illustrated in Fig. lG, the URS3 probe formed a complex with a binding activity in frac- tions 52 and 53 after Mono S chromatography. This latter com- plex was not supershifted with REBl polyclonal antibody (data not shown). When similar assays were performed with the

A.

REBl+

pro-

B.

Yeast ENOl URS Element

RIB01

9795

0 4x 2ox looX 4x 2ox 1oox 4x 2ox looX 4x 2ox looX P 0 4x 2ox lWX RlBOl RlBOl 1 URSl URSl-mutl ENOlIUAS2

URSl

0 4x 20x 1 oox 4x 20x 1oox P 0 4x 20x 1oox

RIB01 URS1 ENOlIUAS2 FIG. 6. The major binding activity that interacts with the ENOl URS element and the ENOl UAS2 element is REBl. Competition

DNA-protein binding analysis was performed with "'P-labeled RIBOl (panel A) , a double-stranded oligonucleotide containing the REBl binding site from the yeast ribosomal RNA enhancer element or URSl (panel B ) , a double-stranded oligonucleotide corresponding to ENOl URS element

fraction 50 after Mono S chromatography (Fig. l.4). P indicates binding reactions containing probe without added fractionated cell extract. Binding sequences extending from position -228 to -193. DNA binding reactions contained 0.25 fmol of probe and approximately 0.1 pg of protein from

reactions contained 0, 4-, 20-, or 100-fold molar excess of the indicated competitor DNA. Double-stranded oligonucleotide competitor DNAs included: RIBO1; RIBOl1, RIBOl sequences containing a triple transversion mutation which abolished REBl binding in vitro; URS1; URS1-mutl, URSl sequences containing a triple transversion mutation a t positions -207, -208, and -209 (Fig. 4); and ENOlILTAS2, an oligonucleotide containing the EBFl binding site from the ENOl UAS2 element. The position ofthe REBl (EBFlbdependent complex is indicated as is the position of a second complex (I) formed with the URSl probe.

URS3-mut3 oligonucleotide probe which is a derivative of URS3 containing transversion mutations a t positions -155, -154, and -153, complex formation with the binding activity in fractions 52 and 53 was greatly diminished (Fig. lH). Since this latter triple transversion mutation caused a large decrease in URS activity in vivo (Fig. 8), these results suggest that the binding activity in fractions 52 and 53 after Mono S chroma- tography plays a role in URS activity.

DISCUSSION

The ENOl URS element, originally identified by deletion mapping analysis (2), bound multiple DNA binding activities. URS element sequences that bound an abundant DNA binding activity were identified by DNase I footprinting and methyla- tion interference analyses using crude whole cell extracts and partially purified binding activity. On the basis of DNA binding site sequence similarity, co-chromatography, competition DNA binding assays, and gel mobility supershift assays with a poly- clonal antibody directed against the yeast REBl protein, we concluded that this abundant binding activity is REB1. Similar analyses showed that EBF1, a binding activity that binds spe-

cifically to sequences within the ENOl UAS, element (3), is identical to REB1.

REBl binding to the URS element resulted in a DNase I footprint extending from position -226 to -195. Interestingly, a series of single base protections a t positions -190, -182, -176, -167, -163, -157, and -146 were observed commensurate with formation of the extended footprint between positions -226 and -195. This latter observation suggests that REBl might con- tact sequences extending across the entire URS element. Alter- natively, other URS binding activities that co-chromatograph with REBl after Mono S chromatography might bind commen- surate with REBl resulting in the observed extended interac- tions with URS element sequences. Interestingly, a second binding activity present in fraction 50 after Mono S chroma- tography formed a gel mobility shift complex, designated I com- plex, with the URSl and ENOlmAS2 oligonucleotides but not the RIBOl oligonucleotide. This latter binding activity was not antigenically related to REBl nor was this binding activity affected by base substitution.mutations within the URS ele- ment that abolished REBl binding. Based on these observa- tions, we conclude that the binding activity responsible for I

9796 Yeast ENOl URS Element

- - 0 0 .)

RIBOl RIBOl HMRB(ABF1)

URSl ENOINASZ FIG. 7. Gel mobility supershift analysis using an antibody directed against REBl. Gel mobility shift assays were performed using the

following 32P-labeled double-stranded oligonucleotide probes: RIBOI, URSI, ENOIAJASZ, and HMRB (contains an ABFl binding site from the yeast HMR silencer element). RIBOI, URSI, and ENOIAJASP DNA binding reactions contained 0.25 fmol of probe and approximately 0.1 pg of protein from fraction 50 after Mono S chromatography (Fig. IA). HMRB binding reactions contained 0.25 fmol of probe and approximately 0.1 pg of protein from a fraction after Mono S chromatography containing ABFl binding activity. Following a 15-min binding reaction, the indicated dilutions of sera directed against REBl or preimmune sera were added and the reaction was allowed to proceed for an additional 10 min.

complex formation is not related to REB1. Deletion of 5' terminal ENOl URS element sequences which

include the REBl binding site extending to position -181 caused only a 3-5-fold increase in ENOl expression in cells grown in a medium containing glucose (2). Since deletion mu- tations would be expected to alter spacing of multiple regulator elements andor remove multiple factor binding sites, the or- ganization of cis-acting sites within the URS element was ana- lyzed using site-directed base pair substitution mutations. Base substitution mutations that abolished REBl binding in vitro caused a 6-fold increase in ENOl expression in cells grown in a medium containing glucose. This observation is consistent with the earlier deletion mapping analysis (2) and strongly suggests that REBl binding is required for full URS element activity. REBl (8, 161, also known as, factor Y (17), factor Q (18), and GRF2 (19), binds to upstream activation sites (UAS elements) from several yeast genes (19-21). We show here that REBl is identical to EBF1, a factor that binds a UAS element from the yeast enolase gene ENOl (2). Since REBl binds essential sequences in both URS and UAS elements, it would appear the REB1, like RAP1 (22,23), ABFl(22,24), and MCMl (25,26,27), plays a role in activation and repression of yeast gene expression. REBl binding to the enhancer element from yeast ribosomal cistrons has been implicated in termina-

tion of transcription in vivo (28) and in vitro (29) by RNA polymerase I. Although the precise biological activity of REBl protein is not known, it is clear that the activity of this abun- dant DNA-binding protein can influence a number of seemingly unrelated regulatory elements.

Deletion of URS element sequences located downstream from position -181 caused 8-13-fold increases in ENOl expres- sion in cells grown in a medium containing glucose (2), sug- gesting that the URS element contains multiple cis-acting ele- ment. Base substitution mutations were introduced at two other sites implicated in URS element activity by deletion map- ping analysis (2). One of these sites corresponded to a region of the ENOl URS element that shares extensive sequence simi- larity with a URS element identified within the 5"flanking regions of the yeast CARl structural gene (5, 6) and the yeast SSAl structural gene (7). Expression of an ENOl gene contain- ing base substitution mutations in this latter site was in- creased 7.5-fold relative to a wild-type ENOl gene in cells grown in a medium containing glucose. This observation sug- gests that the URS elements from ENOl, CARl, and SSAl may bind a common factor. Expression of an ENOl gene containing a base substitution mutation in a third site near the 3' termi- nus of the URS element caused a 14-fold increase in ENOl expression relative to a wild-type ENOl gene in cells grown in

Yeast ENOl URS Element 9797

A E N 0 1 / EN02

glycerol glucose +lactate

w t 1/20 1/1

+ .I41 -126 1.5/20 1/1

... A . URS-mutl c 6/20 0.7/1

6 . URS-mut2 c 7.5/20 1.7/1

C . URS-mut3 14/20 1.7/1

D . URS-mutl + Z c 15/20 1/1

E . URS-mutl+3 17/20 1/1

F. URS-mutZ+3 17/20 1.8/1

G . URS-mutl+2+3 20/20 2.5/1

... ...

......

... ... ......

. . . . . . . . .

B GLUCOSE

w t + - A B C D E F G

1 2=

GLYCEROL +LACTATE

W t + - A B C U E F G FIG. 8. Site-directed mutational analysis indicated that at

least three regions of the ENOl URS element are important for repression of ENOl expression in uiuo. Site-directed mutagenesis of the ENOl URS element was performed as described under “Experi- mental Procedures.” Triple transversion mutations were introduced at three sites within the URS element as indicated in Fig. 4. 125-base pair Sal1 fragments that include complete URS elements containing the indicated mutations were ligated in the wild-type orientation between positions -301 and -121 within the ENOl 5’-flanking region in plasmid peno46/HIS3(-301/-121). These latter plasmids were integrated at the ENOl locus of a strain carrying an enol deletionlURA3 insertion mu- tation. ENOl gene expression was monitored relative to expression of the wild-type chromosomal EN02 gene by Western blotting analysis using whole cell extracts derived from log-phase cells grown on a me- dium containing either glucose or glycerol plus lactate. Extract from strain S173-6B (wt) served as controls. The enolase 1 polypeptide was quantitated relative to the glucose inducible enolase 2 polypeptide in cells grown in a medium containing glucose, 20, or glycerol plus lactate,

S173-6B containing a wild-type chromosomal EN01 gene; +, an ENOl 1, by densitometry as previously described (3). Panel A, wt, strain

gene containing a 125-base pair U R S fragment without mutations li- gated between positions -301 and -121; -, an ENOl gene containing a

tions -301 and -121. A-G, ENOl genes containing 125-base pair URS deletion mutation that removed URS element sequences between posi-

elements carrying the indicated base substitution mutations ligated between positions -301 and -121 within the ENOl 5’-flanking region. Panel B , Western blotting analysis of extracts prepared from strains carrying the mutant ENOl genes described in panel A. The positions of the enolase 1 and enolase 2 polypeptides are indicated.

a medium containing glucose. An oligonucleotide corresponding to these latter wild-type sequences (URS3) bound a novel factor present in fractions 52 and 53 following Mono S chromatogra- phy of a yeast whole cell extract. The ability of this binding activity to form a gel mobility shift complex was greatly dimin- ished by the URS-mut3 triple transversion mutation suggest- ing that this binding activity plays a role in modulating URS element activity.

The activity of URS elements containing double mutations (15-25% of wild-type URS element activity) was lower than ob- served for either single mutation alone, whereas the triple mu- tation completely abolished URS activity. These results clearly showed that the activity of the ENOl URS element was modu- lated by multiple cis-acting elements. The approximately addi- tive effects of double and triple mutations uersus single muta- tions on URS activity indicates that the three sites function together to modulate URS activity. These results are in contrast to those reported for the CAR1 and SSAl URS elements. For these latter genes, URS activity was abolished by point muta- tion within the single conserved URS sequence (5-7).

Acknowledgments-We thank Catherine Willett and Alan Pepper for helpful discussions. We are also indebted to Teresa Yokoi for superb technical assistance.

REFERENCES 1. Cohen, R., Holland, J. P., Yokoi, T., and Holland, M. J. (1986) Mol. Cell. Biol.

6.2287-2297 2. Cohen, R., Yokoi, T., Holland, J. P., Pepper, A. E., and Holland, M. J . (1987)

3. Brindle, P. K., Holland, J. P., Willett, C. E., Innis, M. A., and Holland, M. J.

-, - ~ - -

Mol. Cell. Biol. 7, 2753-2761

4. Holland, J. P.. Brindle, P. K., and Holland, M. J. (1990) Mol. Cell. Biol. 10, (1990) Mol. Cell. Biol. 10, 46724885

4863-4871

39974001

3884-3895

5. Sumrada, R. A,, and Cooper, T. G. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,

6. Luche, R. M., Sumrada, R., and Cooper, T. G . (1990) Mol. Cell. Biol. 10,

7. Park, H. O., and Craig, E. A. (1989) Mol. Cell. Biol. 9,2025-2033 8. Morrow, B. E., Ju, Q.. and Warner, J. R. (1990) J. Biol. Chem. 265, 20778-

9. McAlister, L., and Holland, M. J. (1982) J. Biol. Chem. 257, 7181-7188 29783

10. Swanson, M. E., and Holland, M. J. (1983) J. Biol. Chem. 258,3242-3250 11. Layne, E. (1957) Methods Enzymol. 3,447-454 12. Maxam, A., and Gilbert, W. (1980) Methods Enzymol. 65,499-560 13. Messing, J. (1983) Methods Enzyrnol. 101,20-78 14. Innis, M. A,, and McCormick, F. (1986) Methods Enzymol. 119,397-403 15. Sanger, F., Nicklen. S., and Coulson, A. R. (1977)Proc. Natl. Acad. Sci. U. S. A.

16. Morrow, B. E., Johnson, S. P., and Warner, J. R. (1989) J. Biol. Chem. 264,

17. Fedor, M. J., Lue, N. F., and Kornberg, R. D. (1988) J. Mol. Biol. 204,109-127 18. Chen, W., and Struhl, K. (1989) EMBO J. 8.261-268 19. Chasman. D. I., Lue, N. F., Buchman, A. R.. LaPointe, J. W., Larch, Y., and

20. Nishizawa, M., Araki, R., and Teranishi, Y. (1989) Mol. Cell. Biol. 9,442-451 21. Scott, E. W., Allison, H. E., and Baker, H. V. (1990) Nucleic Acids Res. 18,

22. Buchman, A. R., Kimmerly, W. J., Rine, J., and Kornberg, R. D. (1988) Mol.

23. Huet, J., and Sentenac, A. (1987) Proc. Natl. Acad. Sci. U. S. A. 84,3648-3652 24. Brand, A. H., Micklem, G., and Nasmyth. K. (1987) Cell 51,709-719 25. Keleher, C. A,, Passmore, S., and Johnson, A. D. (1989) Mol. Cell. Biol. 9,

26. Passmore, S., Elble, R., a n d Q e , B. K. (1989) Genes & Deu. 3,921-935 27. Jarvis, E. E., Clark, K. L., and Sprague, G . E , Jr. (1989) Genes & Deu. 3,

28. Mestel, R.,Yip, M., Holland. J. P., Wang, E., Kang, J., and Holland, M. J. (1989)

29. Lang, W., and Reeder. R. H. (1993) Mol. Cell. Biol. 13, 649-658

74,5463-5467

9061-9068

Kornberg, R. D. (1990) Genes & Deu. 4,503-514

7099-7107

Cell. Biol. 8, 210-225

5228-30

936945

Mol. Cell. Biol. 9, 1243-54