YEAST VOL. 12: 1339-1350 (1996) Mutations in an Abflp Binding Site in the Promoter of Yeast RP026 Shift the Transcription Start Sites and Reduce the Level of RP026 mRNA SHAHRZAD NOURAINItf-, JIM HUtII, LINDA D. B. McBROOMfs A N D JAMES D. FRIESEN*tf* Toronto, Crmtrdu M5G 1 X 8 t,Depurtnient of' Genetics. Hospitu1,for Sick Chillr.cn, 555 University AIWZIAP, f Depurtmerrt of Moleciibi- und Medical Genetics, University of Toronto, Toronto, Cunuda Received 3 April 1996; accepted 3 June 1996 A binding site for the transcription factor Abflp was identified as an important promoter element of the gene that encodes RpoZ6, a subunit common to all three yeast nuclear R N A polymerases (RNAP). Mutations in the Abflp binding site were identified among a pool of rpo26 mutant alleles that confer synthetic lethality in combination with a temperature-sensitive mutation (rp021-4) in the gene that encodes the largest subunit of RNAPII (RpoZlp). In the presence of the wild-type allele of RP0-71 these r p 2 6 promoter mutations confer a cold-sensitive growth defect. Electrophoretic mobility-shift assays using purified Abflp demonstrated that Abflp binds to the RP026 promoter and that the promoter mutations abolish this binding in vitro. Quantitation of the amount of RPO26 mRNA showed that mutatioiis in the Abflp binding site reduce the expression of RP026 by approximately 60'%,.Mutations that affect Abflp binding also result in a shift of the RP026 transcriptional start sites to positions further upstream than normal. These results suggest that binding of the Abflp transcription factor to the RP026 promoter is important not only in establishing the level of transcription for this gene, but also in positioning the initiation sites of transcription. KEY WORDS - RP0-76; ABFl; transcription initiation INTRODUCTION Regulation of transcription is an important aspect of cellular responses to environmental cues. Transcriptional regulation can occur at various stages: initiation (Drapkin et a/., 1993). elongation (Kerpola and Kane, 1991), and/or termination (Richardson, 1993). or conceivably by controlling the intracellular amount of RNA polymerases (RNAPs). Eukaryotic nuclear transcription is carried out by three different RNAPs. RNAPI transcribes rRNA genes, RNAPII synthesizes primarily mRNA, and RNAPIII transcribes genes for tRNA, 5sRNA and the U6 small nuclear RNAs *Corresponding author, at Charles H. Best Institute, 112 College Street, Toronto, Ontario, Canada, M5G IL6. .Present address: Banting and Best Department of Medical Research, University of Toronto. 112 College St, Toronto, Ontario, Canada M5G 1L6. lipresent address: Department of Respiratory Research, Hospital for Sick Children. 555 University Avenue, Toronto. Canada M5G 1x8. $Present address: Molecular Oncology Group. Royal Victoria Hospital, Hersey Pavillion (H5), 687 Pine Ave West. Montreal, Canada M3A 1Al. CCC 0749-503X/96/131339-12 rC1 1996 by John Wiley & Sons Ltd (Archambault and Friesen, 1993). Eukaryotic nuclear RNAPs are complex enzymes composed of two large subunits and a number of smaller polypeptides (Archambault and Friesen, 1993). In Succhnromyces cerevisiue five of the smaller subunits are common to all three RNAPs: RpbSp, Rpb6p (called Rpo26p in this report), Rpb8p, RpblOp and RpclOp (Archambault and Friesen, 1993). It is known that these subunits are essential for the growth of yeast (Archambault and Friesen, 1993); however, little information is available concerning their specific roles. Since the common subunits are present in all three polymerases, regulation of the steady-state levels of these subunits could be an effective means f o r coordinate regulation of the amount of cellular RNAPs and, consequently. the level of cellular transcription. One way in which the level of common subunits can be regulated is through transcriptional regulatory sequences in their promoters. Identification of such regulatory sequences will be a valuable step in understanding the mechanism(s) of transcriptional responses to environmental cues. In this report we 1340 describe the identification of a promoter element, the binding site for the transcription factor Abflp, which is important for normal expression of the common subunit, Rpo26p. We provide evidence which suggests that Abflp not only stimulates transcription of RP026, but also sets the transcription initiation sites on the RP026 promoter. MATERIALS AND METHODS Plasmids and strains pRP026HAT (TRPI, CENIARJ? contains a 2.1 kb Sau3A RP026 fragment (S. Nouraini et al., In press) cloned into pFL39 (Bonneaud et ul., 1991). pRP026cDNA (Archambault et al., 1990) has been isolated from a cDNA library described by McKnight and McConaughy (1983). Plasmid pSN261 contains a HindIII( - 748)l HindIII(+754) RP026 genomic fragment in pFL39 (see Figure 1 for position of restriction sites numbered by choosing A of the RP026 ATG as +1). Plasmids pSN302, pSN303 and pSN304 contain an AccI ( - 534)/HindIII(+754) RP026 fragment in pFL39; they bear, respectively, the wild-type, double-mutant (rpo26-24) and single-mutant (rpo26-27) forms of the Abflp-binding site in the promoter. The construction of pSN303 and pSN304 was as follows: sequences corresponding to - 597 to +97 were amplified by polymerase chain reaction (PCR) from hydroxylamine mutagenized plasmids corresponding to rpo26-24 and rpo26-27, respectively. These fragments were then digested with Accl( - 534) and Bcll( - 5) and used to replace the corresponding sequences in pSN261. The plasmids thus constructed contain a wild-type RP026-coding sequence downstream of a mutagenized promoter. pSN302 was constructed similarly, except that the promoter fragment originated from unmutagenized pRP026HAT. The PCR primers used in the construction were A34, containing sequences - 597 to - 582 (5’-GCTGG CGAGTCGTCAC-3‘) in the TFC2 open readingframe (ORF), and PE2, corresponding to sequences +97 to +77 (5’-CCTGTATCACA ATGATAGAAG-3’) in the RP026 intron (See Figure 1). All PCR-generated fragments used in this study were sequenced following amplification and cloning to ensure that no extraneous mutations were introduced during amplification. In order to construct plasmids pSN305 and pSN306, a 214bp fragment of the RP026 S. NOURAINI ET AL. promoter (bp - 19 to - 232) was PCR-amplified from pSN302 and pSN304, respectively, and cloned into the BamHI site of pUC19. The primers used were Bandshift-l (5’-GGGGGATC CCTTTACACTGTTACC-3’) and Bandshift-2 (5’-GGGGGATCCGACAGCGATTTCAAC-3’). Yeast strain JAY472 (Archambault et al., 1990) has a temperature-sensitive (ts) allele of RP021 (rpo21-4). In strains JAY472 and JAY476, the chromosomal RP026 gene is partially duplicated such that a truncated and non-functional RP026 is expressed from the RP026lTFC2 promoter and a full-length RP026 is expressed conditionally from the GAL1 promoter (Archambault et ul., 1990). Note that in these strains the RP026JTFC2 divergent promoter is intact, therefore the expression of TFC2 is not affected. Strains SNYH167 through 169 are derivatives of JAY567 (S. Nouraini et al., In press) and contain plasmids pSN302 through pSN304, respectively. In JAY567 the chromosomal RP026 is missing the RP026 intron and nucleotides encoding the first 42 amino acids of the subunit (S. Nouraini et al., In press). This mutant RP026 allele (RP026A42) encodes a truncated but functional protein, which can support normal cell growth under a variety of conditions (S. Nouraini et al., In press). Genetic screen f o r synthetic lethality Plasmid pRP026HAT was mutagenized with hydroxylamine in vitro (S. Nouraini et al., In press) and, following passage through Escherichia coli strain JF1754 (Himmelfarb et al., 1987), the bank of mutagenized plasmids was introduced into the yeast strain JAY472. JAY472 transformants were selected on solid medium in the presence of galactose (chromosomal RP026 expressed) at 30°C. Then, individual transformants were tested for growth defects on solid medium containing glucose (chromosomal RP026 repressed) in order to identify plasmids that confer synthetic lethality in combination with rp021-4. Plasmids that conferred a stable and reproducible synthetic-lethal phenotype to JAY472 were chosen as candidates for mutant alleles of RP026. A more detailed description of this genetic screen will be reported elsewhere (S. Nouraini et al., In press). Mobility-shift analysis The mobility-shift assays were done as described by Buchman and Kornberg (1990) except that Abfl p was preincubated with the binding reaction 1341 ROLE OF Abflp IN THE EXPRESSION OF RP026 mixture (with or without unlabeled competitor oligonucleotides) for 10 min at room temperature prior to the addition of 6.7 fmoles of 32P-labeled probe. The probe was the BamH1 insert of plasmids pSN305 (RP026) or pSN306 (rpo26-27). Purified Abflp (Buchman and Kornberg, 1990) was a gift from Andrew Buchman. Two doublestranded oligonucleotides (1-32 and 33-63) were used for competition assays; their sequences were derived from a 63 bp BgZIIIAluI fragment of MATa (nucleotides 2001 to 2063, numbering according to Astell et al., 1981). Oligo 1-32 (5‘-GATCTAAATAAATTCGTTTTCAATGATT AAAA-3’, sequence of top strand) contains nucleotides 2001 to 2032 and an Abflp-binding site (bold). Oligo 33-63 (5‘-TAGCATAGTCGG GTTTTTCTTTTAGTTTCAG-3’, sequence of top strand) contains nucleotides 2033 to 2063, does not have an Abflp recognition sequence, and was used as a non-specific competitor (McBroom, 1993). produced from the truncated chromosomal allele of RP026 (expressed from a wild-type promoter), and also to the amount of mRNA produced from the ACTl gene. RESULTS In order to understand the function of Rpo26p, a genetic screen was used to isolate mutations in the gene that encodes this subunit. A bank of randomly mutagenized RP026 (S. Nouraini et al., In press) alleles was screened for mutant alleles that were lethal in combination with rp021-4, a ts allele of the gene encoding the largest subunit of RNAPII (Archambault et al., 1990). The phenotype conferred by rpo21-4 is very sensitive to the amount and functional state of RP026. This is based on two previous observations: (1) various RP026 mutant alleles support normal growth in a wild-type RP021 background, but cause lethality in the presence of the ts rpo21-4 mutation (Archambault et al., 1990) and (2) the ts phenotype Primer-extension analysis of rp021-4 is suppressed by an approximately Primers PEl and PE3 were used for primer- two-fold increase in the gene dosage of RP026 extension experiments. PE3 and PE1 hybridize to (Archambault et al., 1990). Plasmid pRP026HAT the RP026 mRNA and span sequences +273 to (RP026, T R P l , CENIARS) was mutagenized +25 1 (5’-CTTCTTATTTGCTCATGCTGTTG), in vitro with hydroxylamine and was introduced and +169 to +149 (5‘-TTCCTCATAAGTCT into yeast strain JAY472. This strain contains the CCTCATC-3‘), respectively, in the RP026 ORF. rpo21-4 ts allele, and the chromosomal RP026 The position of each primer is shown in Figure 4A. expressed from an inducible GA L l promoter. The Primer actin E, (5’-CGTCACCGGCAAAA library of mutagenized plasmids was tested in CCGGC-3’) hybridizes to the second exon of JAY472 for glucose-dependent (when expression ACT1 and allows measurement of ACTl mRNA of chromosomal RP026 is repressed) synthetic as an internal control. Analysis of transcription- lethality (S. Nouraini et al., In press; Figure 2A). initiation sites in the ACTl promoter has shown Sixty-four plasmids consistently conferred a stable one major and a number of minor start sites and recurrent lethal phenotype in the absence of (Munholland et al., 1990). In order to simplify the wild-type RP026 expression. quantitation of the ACT1 mRNA, primer actin E, was labeled with 32P and diluted 20-fold with Identification of promoter mutations unlabeled actin E, for the primer-extension experiments. Under these conditions, the only notably RP026 and TFC2, the gene encoding the detectable initiation site for ACTl is the major RNAPIII transcription initiation factor TFIIIA, are transcribed divergently (Archambault et al., start site, which was used for quantitation. Primer-extension assays were performed as 1992). The sequences that are both necessary and described (Hu et al., 1994) with 2 pg of polyA+ sufficient for expression of RP026 lie in this intermRNA. Quantitation of the amount of RP026 genic region (233 bp) (Mckune and Woychik, message was performed using a PhosphorImager 1994). In order to locate rp026 mutations, a frag(Molecular Dynamics, model 425E) and the ment was isolated from the mutagenized plasmids ImageQuant quantitation software. The amount of (Bcll - 5 to Hind111 +754, see Figure 1) that transcript from the plasmid-encoded RP026 contained the RP026 O R F and 200 bp of untrans(driven by either a wild-type or a mutant pro- lated downstream sequences. This fragment was moter) was normalized to the amount of mRNA then used to replace the wild-type counterpart in - 1342 S. NOURAINI ET AL. 100 bp U Hindlll Acc' 3'- A Wl AGAAGAATAlTATAlTP rpo26-24 rpo26-26 rpo26-27 T rpo26-28 Figure 1. Alleles of RP026 with mutations in an upstream Abflp-binding consensus sequence. The region of the genomic clone present on the mutagenized plasmid (Hind111 - 748 to Hind111 +754) is indicated. RP026 is transcribed divergently from TFC2, the gene encoding the RNAPIII transcription initiation factor, TFIIIA. Large arrows indicate the direction in which each gene is transcribed. In pRP026HAT, TFC2 is truncated at the Hind111 ( - 748) site. IVS in RP026 ORF refers to the intron in this gene. Small arrows designated PE2 and A34 indicate primers used for PCR amplification of the promoter fragment from the mutagenized plasmids. The sequence of the RP026 promoter containing the mutations is shown, and each mutant allele of RP026 is aligned with the corresponding base pair changes. The boxed region is the consensus binding-sequence for Abflp within the RP026 promoter. The T-rich sequence often found in association with Abflp binding-sites in a number of yeast promoters is underlined. Numbering of the nucleotides is based on assigning the A of the RP026 ATG as + I . an unmutagenized plasmid. As a result of this construction, expression of the RP026 coding region derived from the mutagenized plasmids is driven by a wild-type promoter. If the hydroxylamine-generated mutations lie outside of the R P 0 2 6 ORF and 200 bp of downstream untranslated sequences, this newly generated plasmid should not be able to confer a synthetic-lethal phenotype on JAY472 (rpo21-4). When tested, five RP026 mutant alleles were no longer synthetically lethal in combination with rpo21-4. Sequence analysis determined that these mutant alleles were wild-type in the R P 0 2 6 ORF, but had mutations in the 5'-upstream region. The R P 0 2 6 promoter was searched for matches to consensus sequences for binding of a variety of yeast transcriptional activators. The sequence 5'-ATCATACTA TACG-3' ( - 129 to - 140, bottom strand; Figure 1) matches the binding-site consensus-sequence (5'-RDCNYNNNNNACG-3'; Buchman and Kornberg, 1990) for the transcription factor Abflp. Two of the mutant alleles, upo26-24 and rpo26-27, had mutations in this consensus sequence (Figure 1). rpo26-26 and rpo26-28 contained multiple base changes that included the Abflp binding site (Figure l), and rpo26-25 had two that did not include the Abflp recognition sequence (not shown). Mutant alleles rpo26-24 and rpo26-27 were studied further in order to investigate the role of Abflp in regulation of R P 0 2 6 transcription. The phenotype conferred by these mutant alleles was tested in the presence of the wild-type allele of RPO2I in yeast strain JAY476 @GAL-RP026, RP021). These mutations generated a slowgrowth defect in a wild-type R P 0 2 1 background at a low temperature (Figure 2B). Although mutations were identified in the promoter of R P 0 2 6 for the above-mentioned mutant alleles, it was necessary to confirm that the ROLE OF Abflp IN THE EXPRESSION OF RP026 1343 A Galactose Glucose RP026 rpo26-24 rpo26-27 Vector B 1 5% 0 23 C 3 0°C 0 37 c RP026 rpo26-24 rp026-2 7 Vector Figure 2. Phenotypes generated by mutations in the Abflp binding sequence. (A) A growth test showing the synthetic-lethal phenotype conferred by the RP026 promoter mutations on JAY472 (rp021-4, pGALI-RP026), when the expression of chromosomal wild-type RP026 is repressed in the presence of glucose. A drop containing a suspension of -2000 cells w d S applied to solid medium containing either glucose or galactose and was tested for growth at 30°C. (B) Phenotype generated by the RP026 promoter mutations in the presence of wild-type RPO21 in strain JAY476 (RP021, pGAL1-RP026). Growth tests were performed at four different temperatures on solid medium in the presence of glucose (chromosomal RP026 is repressed). Each row indicates five-fold serial dilutions of a starting cell suspension of 2000 cells per 10 ~1 (volume of each drop). Vector refers to pFL39 (Bonneaud et al., 1991). phenotypes conferred by these mutants were not due to mutations in the vector sequences. Such mutations could lead to a reduction in the amount of RP026 gene product simply by lowering the plasmid copy-number. To test this possibility, the mutant promoter fragments from plasmids corresponding to RP026 (from pRP026HAT), rpo2627 and rpo26-24 were isolated and were used to replace corresponding sequences in the unmutagenized plasmid pSN261. The mutant promoter fragments were both necessary and sufficient to confer a synthetic-lethal phenotype to JAY472 and a cold-sensitive phenotype to JAY476 (not shown). Binding of Abflp to the RP026 promoter Mobility-shift analyses were performed with purified Abflp (Buchman and Kornberg, 1990) in combination with the wild-type or mutant promoter fragments containing the Abflp bindingsite. Abflp was able to bind to the wild-type promoter fragment (Figure 3, lane 3); binding was competed with a double-stranded oligonucleotide (1-32) containing a well-characterized Abflp binding-site from the MA Ta locus (McBroom and Sadowski, 1994; Figure 3, lanes 4 to 6). A similarsized double-stranded oligonucleotide (33-63), which lacked an Abflp binding site, was unable to compete in this binding reaction (Figure 3, lanes 7 to 9). A single mutation (rpo26-27) in the Abflp binding site prevented binding of Abflp to the promoter fragment (Figure 3, lane 12). These results suggest that Abflp may also bind to the RP026 promoter in vivo and that the mutations in rpo26-24 and rpo26-27 inhibit this interaction. S . NOURAINI ET AL. 1344 COMPETITOR COMP€llTOR n NON- n NON - SPECIFIC SPECIFIC x x xx a )umx ox v o0o0g0g0o 0 o .c I I Imlq*194 1 2 3 4 5 SPECIFIC SPECIFIC a r X X X X X X L L I1 I IYrlnl 1 - b - 6 7 8 ~0 ~ 0 0O ~ o )oo o o o m m a oooo v 9 10 11 12 1314 1516 17 18 L W 1 l - D TYPE rpo26-27 I Figure 3. Binding of Abflp to the RP026 promoter. Mobility-shift analysis was performed with purified Abflp (lanes 3 to 9 and 12 to 18) and 3ZP-labeled promoter fragments from RP026 and rpo26-27. As control, the experiment was also performed with no protein added (lanes 1 and 10) and BSA (lanes 2 and 11). Competition for binding of Abflp was performed with the indicated molar excess of an oligonucleotide containing a known Abflp binding site (specific competitor), and a similar-sized oligonucleotide lacking an Abfl p binding site (non-specific competitor) Quantitation of the RP026 mRNA Abfl p can activate the transcription of a variety of yeast promoters (Dhawale and Lane, 1993; Mager and Planta, 1990). Therefore, it was anticipated that a mutation in the Abflp binding-site upstream of RP026, which inhibits binding of this factor in vitro and leads to a growth defect, would lower the steady-state level of RP026 mRNA. The amount of RP026 mRNA was measured by primer-extension analysis of poly-A+ mRNA prepared from strains SNYH167 (RP026), SNYH168 (rpo26-24) and SNYH169 (rpo26-27; Figure 4B). In these strains a truncated functional form (S. Nouraini et al., In press) of RP026, RP026-A42, is expressed on the chromosome from a wild-type promoter, and full-length RP026 is transcribed on a yeast episomal plasmid from either wild-type or mutagenized promoters (rpo26-24 and vpo26-27; Figure 4A). The amount of mRNA produced from the ACTl gene was measured as an internal control in addition to mRNA produced from chromosomal RP026-A42. The steady-state level of RP026 mRNA was reduced in the presence of mutations in the Abflp binding-site (Figure 4B). Quantitation of the amount of RP026 mRNA (indicated by the single inner bracket in Figure 4B) normalized to the amount of ACTl mRNA (shown by arrow in Figure 4B), revealed an approximately 60% reduction (Figure 4C) in strains carrying rpo26-24 (Figure 4B, lane 3 ) and rpo26-27 (Figure 4B, lane 4), compared to wildtype (Figure 4B, lane 2). The reduction in the amount of the RP026 mRNA was essentially the same when the level of RP026-A42 was used as an internal control (not shown). ROLE OF Abflp IN THh EXPRESSION OF RP026 New transcription-initiation sites are generated by the promoter mutations 1345 (Archambault et al., 1996) and RNAPIII (Mosrin et a/., 1990). In the case of RNAPII, the expression The mutations in the Abflp binding-site led to of RPO2l was placed under the control of the the selection of new transcription-initiation sites repressible LEU2 promoter. When expression of in the RP026 promoter (Figure 4B). Primer- RP021 was repressed by adding leucine to the extension experiments were performed with two growth medium, yeast cells acquired a slowdifferent preparations of polyA+-mRNA; new ini- growth and a temperature-sensitive phenotype tiation sites of transcription in the RNA prepared (Archambault et al., 1996). Underproduction of from the mutant strains was observed consistently. Rpc3lp, the 31 kDa subunit of yeast RNAPIII, In order to map the initiation sites, primer exten- also led to a slow-growth phenotype and a desion analysis was performed using the PE1 primer, crease in the steady-state levels of tRNA in the cell which hybridizes only to full-length RP026 and (Mosrin et a/., 1990). The artificial underproducnot to RP026-A42 transcripts (see Figure 4A). A tion of Rpc3lp was achieved when a nonsense Bcll fragment of pRP026cDNA was used as a mutant allele of RPC31 was suppressed partially template for sequencing with the PE1 primer, using a tRNA suppressor (Mosrin et a/., 1990). Using electrophoretic-mobility-shift assays and which served as a size marker for the primerextension products (Figure 5A). The estimated purified Abflp, we have shown that Abflp binds positions of initiation-sites in the mutant (rp026- to the RP026 promoter in vitro and that mutations 24 and rpo26-27) and the wild-type (RP026) in the Abflp binding-site abolish protein binding. promoters are shown in Figure 5B. In wild-type The SPT2 promoter also contains an Abflp RP026, transcription was initiated at multiple binding-site (5’-ATCATGTTAAACG-3‘), which sites, which is characteristic of transcription initia- closely resembles the binding site in the RP026 tion in S. cerevisiae (Russell, 1983; Struhl, 1989), promoter (5’-ATCATACTATACG-3’; Buchman and spans sequences +1 (A of RP026 ATG) to and Kornberg, 1990). The affinity of Abflp for this -84. In the promoter mutants, the window of binding-site has been measured by mobility-shift initiation was shifted upstream to position - 108, assays (Buchman and Kornberg, 1990); a mutation 19 bp away from the Abflp binding site (under- analogous to that of rpo26-27 in the SPT2 prolined in Figure 5B). In this new window of initia- moter reduced the affinity of Abflp for its binding tion, previously unused or infrequently used start site by more than 500-fold, abolishing binding sites became major transcription initiation sites, (Buchman and Kornberg, 1990). This is consistent while previous major initiation sites that lie closer with the results of our mobility-shift assays and suggests that the Abflp binding-site in the proto the ATG were seldom used (Figure 5A). moter of RP026 is probably not occupied in vivo in SNYH168 (rpo26-24) and SNYH169 (rpo2627) strains. DISCUSSION Abflp alone is not a potent transcriptional Using a genetic screen based on synthetic lethality activator, normally requiring the cooperation of between rpo21-4 and mutant alleles of RP026, other weak activators (Buchman and Kornberg, we have identified mutations in the consensus 1990; Della Seta et a/., 1990b; Halfter et al., 1989; binding-sequence for the transcription factor Hamil eta/., 1988; Kraakman et a/., 1991; Sinclair Abflp in the promoter of RP026. In addition et al., 1994; Trawick et al., 1992; Willett et al., to the synthetic-lethal phenotype, the promoter 1993). For example, Abflp binding-sites are often mutants also give rise to a cold-sensitive growth associated with a T-rich sequence, which functions defect in the presence of wild-type RP021. The in a variety of yeast promoters as a transcriptional cold-sensitive phenotype conferred by the Abflp activator (Buchman and Kornberg, 1990; binding-site mutations in the RP026 promoter GonCalves et al., 1995; Iyer and Struhl, 1995; Lue could result from a reduction in the intracellular et al., 1989; Schultes and Szostak, 1991). In the amount of RNAPs. This in turn could lead to RP026 promoter the Abflp binding-site is adjasuboptimal expression of certain genes that may be cent to a T-rich sequence proximal to the RP026 required for growth of yeast at low temperatures. O R F (Figure 1). The Abflp binding-site mutations Change in the growth properties of yeast cells as in the RP026 promoter resulted in only a threea result of underproduction of a polymerase sub- fold reduction in the amount of RP026 mRNA unit has been observed previously for RNAPII (Figure 4C). This is consistent with the known 1346 S. NOURAINI ET AL. weak transcriptional activity of this activator and suggests that other promoter sequences, perhaps the T-rich region, contribute to the expression of RP026. The contribution, if any, of the T-rich region to the expression of RP026 would not be detected by our mutagenesis, since hydroxylamine alters only cytosines. Primer-extension analysis of RP026 mRNA showed that the mutations in the Abflp-binding sequence reduce the expression of this gene (Figure 4), and that the transcription start sites shift upstream in the promoter (Figure 5). The sites of transcription initiation from the RP026 promoter have been analysed previously by Woychik et al. (1990), who reported four major initiation sites ( - 10, - 15, - 25, - 26), only three of which ( - 15, - 25, - 26) were detected by our primerextension analyses (Figure 5). However, we did identify other initiation sites in addition to those previously reported (Woychik et al., 1990). This discrepancy might be due to differences in the yeast strains used for the preparation of mRNA in the two studies. For example, the number of transcription initiation sites has been shown to be different A ACCI II .. ' If . I TFC2 I ! RP026-A42 . _ ~ . Wild type I: I . ..// ~ I I A i Chromosome I Plasmid Wild type PE1 PE3 or mutant B 517 506 396 344 298 ,- 220 201 - 154 134 - 1 I 2 13 14 Mutant sites n \Wild-type initiation sites RPOZ 6-442 Figure 4 (A and B) 1347 ROLE OF Abflp IN THE EXPRESSION OF RP026 Quantitation of RP026 mRNA Normalized to actin 1.2 1 .o 0.8 0.6 0.4 0.2 0.0 -r RP026 rpO26-27 rpo26-24 Figure 4. Quantitation of the amount of RP026 message in the presence and absence of Abflp binding-site mutations. (A) Schematic representation of the RP026 locus on the chromosome and on the plasmid in the strain used for preparation of RNA. A truncated and functional form of RP026 (RP026-A42), which is missing the first 42 amino acids and the intron, is expressed from a wild-type promoter from the chromosome. A full-length RP026 is expressed from either a mutant or a wild-type promoter on the plasmid. The small arrows named PE3 and PE1 indicate the primers used for primer-extension experiments. (B) Results of the primer-extension analysis using p01y-A~mRNA. The products corresponding to the chromosomal and plasmid RP026 mRNA and the ACT1 mRNA are indicated. Lanes 2 to 4 represent results of primer-extension experiments performed with a mixture of actin E, and PE3 primers using polyA+-mRNA from strains SNYHl67 (RP026),SNYHl68 (rpo26-24) and SNYH 169 (rpo26-27), respectively. Lane 1 shows the results of primer-extension experiments performed with actin E, primer only, using mRNA from the wild-type strain. The position of migration of end-labeled 1 kb molecular weight DNA marker (from BRL) is shown on the left. (C) Quantitation. The amount of the plasmid-encoded RP026 message was normalized to the level of ACT1 message. The level of mRNA generated from each gene was measured using a Molecular Dynamics PhosphorImager. The normalization was performed by dividing the amount of RP026 mRNA by the amount of ACT1 mRNA. For quantitation of RP026 mRNA, the amount of signal generated from both mutant and wild-type initiation sites (indicated by brackets) were quantitated for lanes 2 to 4. 1348 S. NOURAINI ET AL. - -108 v -101 v -94 -51 c B t r t TCGTATAGTATGATATTTAATTGA A A A A A A A A TTTTTTTTTTITT AAGAA -141 rrr PPP P PP P wg2 TATCATTCAAAGGATCAATCACAACCTTGAAGAAA GGCT AAAAAGA CACA -91 -42 Be TTTTGC -4 1 L! GTAACAG P B TAAAGATTAAGGCTACAAGTGATC TGTCAGACTA +l + 1I Figure 5. Mapping of the new start sites generated due to the mutations in the Abflp binding-site. (A) Primer extension was performed with poly-A+ mRNA and primer PEl, which recognizes only the mRNA produced from the full-length RP026. The last four lanes are a sequencing ladder generated from sequencing of a Bcll fragment of RP026 cDNA with primer PEl. The initiation sites used in the wild-type promoter are shown by short arrows. Newly generated start sites due to rpo26-24 and rpo26-27 mutations are shown by long arrows. (B) Location of initiation sites used in the RP026 promoter. Initiation sites present in the wild-type promoter are shown by open circles; solid circles represent the initiation sites that are either enhanced or only used in the presence of the mutation in the Abflp binding-site. The Abflp binding consensus sequence is underlined. 1349 ROLE OF Abflp IN THE EXPRESSION OF RP026 for the ADHl gene amongst different wild-type yeast strains (Bennetzen and Hall, 1982; Pinto et al., 1992, 1994). The role of Abflp binding-sites in the expression of a variety of genes has been studied (Buchman and Kornberg, 1990; Della Seta et al., 1990a,b; Dequard-Chablat et al., 1991; Halfter et al., 1989; Hamil et al., 1988; Herruer et al., 1989; Kraakman et al., 1991; Mager and Planta, 1990; Sinclair et al., 1994; Trawick et al., 1992; Willett et al., 1993); however, to our knowledge, this is the first report of a shift in the initiation sites of transcription due to a mutation in an Abflp binding-site. In other studies, promoters have often been used to drive expression of the lacZ reporter gene and the effects of Abfl p binding-site mutations have been analysed by monitoring b-galactosidase activity in cellular extracts. Thus, possible shifting of the initiation sites of transcription due to mutations in an Abflp binding-site would have escaped detection. Abfl p is a highly abundant transcriptional activator and is required for constitutive o r regulated transcription of a large number of genes (Buchman and Kornberg, 1990; Della Seta et al., 1990b; Dhawale and Lane, 1993; GonCalves et al., 1995; Halfter et al., 1989; Hamil et al., 1988; Herruer et al., 1989; Kraakman et al., 1991; Mager and Planta, 1990; Sinclair et al., 1994; Trawick et al., 1992; Willett et al., 1993). For these genes Abflp appears to facilitate the access of other regulatory factors to promoters, thereby mediating their regulatory effects (GonCalves et al., 1995; Trawick et al., 1992; Willett et al., 1993). The ability of Abflp to create a sharp bend in DNA (McBroom and Sadowski, 1994) and to position nucleosomes on the promoter (De Winde et al., 1993) might underlie the mechanism by which it facilitates access to DNA for other regulatory factors. Whether the positioning of transcription initiationsites on the RP026 promoter represents another aspect of transcriptional regulation by Abf l p remains an open question. ACKNOWLEDGEMENTS We thank Alia Ahmed for help with preparation of poly-A' mRNA, and Jacques Archambault, David Jansma, Ian Donaldson, Christine Tennyson, Henry Krause and Jack Greenblatt for their valuable comments. This work was supported by a grant from the Medical Research Council to J. D. F. and a Steve Fonyo Studentship from the National Cancer Institute of Canada to S.N. REFERENCES Archambault, J. and Friesen, J. D. (1993). Genetics of eukaryotic RNA polymerases I, 11, and 111. Microbiol. Rev. 57, 511-779. Archambault, J., Jansma, D. B. and Friesen, J. D. (1996). Underproduction of the largest subunit of RNA polymerase I1 causes temperature sensitivity, slow growth, and inositol auxotrophy in Saccharomyces cerevisiae. Genetics 142, 737-747. Archambault, J., Milne, C. A,, Schappert, K. T., Baum, B., Friesen, J. D. and Segall, J. (1992). The deduced sequence of the transcription TFIIIA from Saccharomyces cerevisiae reveals extensive divergence from Xenopus TFIIIA. J. Biol. Chem. 267, 3282-3288. Archambault, J., Schappert, K. J. and Friesen, J. D. (1990). A suppressor of an RNA polymerase I1 mutation of Saccharomyces cerevisiae encodes a subunit common to RNA polymerases I, 11, and 111. Mol. Cell. Biol. 10, 6123-6131. Astell, C. R., Ahlstrom-Jonasson, L., Smith, M., Tatchell, K., Nasmyth, K. A. and Hall, B. D. (1981). The sequence of the DNAs coding for the mating-type loci of Saccharomyces cerevisiae. Cell 27, 15-23. Bennetzen, J. L. and Hall, B. D. (1982). The primary structure of the Saccharomyces cerevisiae gene for alcohol dehydrogenase. J. Biol. Chem. 257, 30183025. Bonneaud, N., Ozier-Kalogeropoulos, O., Li, G., Labouesse, M., Minvielle-Sebastia, L. and Lacroute, F. (1991). A family of low and high copy replicative, integrative and single-stranded S. cerevisiaelE. coli shuttle vectors. Yeast 7, 609-615. Buchman, A. R. and Kornberg, R. D. (1990). A yeast ARS-binding protein activates transcription synergistically in combination with other weak activating factors. Mol. Cell. Biol. 10, 887-897. De Winde, J. H., Van Leeuwen, H. C. and Grivell, L. A. ( 1 993). The multifunctional regulatory proteins ABFl and CPFl are involved in the formation of a nucleasehypersensitive region in the promoter of the QCR8 gene. Yeast 9, 847-857. Della Seta, F., Ciafre, S.-A,, Marck, C., et al. (1990b). The ABFl factor is the transcriptional activator of the L2 ribosomal protein genes in Saccharomyces cerevisiae. Mol. Cell. Biol. 10, 2437-2441. Della Seta, F., Treich, I., Buhler, J.-M. and Sentenac, A. (1990a). ABFl binding sites in yeast RNA polymerase genes. J. Biol. Chem 265, 15168-15175. Dequard-Chablat, M., Riva, M., Carles, C. and Sentenac, A. (1991). RPC19, the gene for a subunit common to yeast RNA polymerases A (I) and C (111). J. Biol. Chem. 266, 15300-15307. Dhawale, S. S. and Lane, A. C. (1993). Compilation of sequence-specificDNA-binding proteins implicated in transcriptional control in fungi. Nucl. Acids Res. 21, 5537-5546. 1350 Drapkin, R., Merino, A. and Reinberg, D. (1993). Regulation of RNAPII transcription. Curr. Opin. Cell Biol. 5, 469476. Gonqalves, P. M., Griffioen, G., Minnee, R., et ul. (1995). Transcription activation of yeast ribosomal protein genes requires additional elements apart from binding sites for Abflp or Raplp. Nucl. Acids Res. 23, 147551480. Halfter, H., Muller, U., Winnacker, E.-L. and Gallwitz, D. (1989). Isolation and DNA-binding characteristics of a protein involved in transcription activation of two divergently transcribed, essential yeast genes. EMBO. J. 8, 3029-3037. Hamil, K. G., Nam, H. G. and Fried, H. M. (1988). Constitutive transcription of yeast ribosomal protein gene TCMl is promoted by uncommon cis- and trans-acting elements. Mol. Cell. Biol. 8, 43284341. Herruer, M. H., Mager, W. H., Doorenbosch, T. M., Wessels, P. L. M., Wassenaar, T. M. and Planta, R. J. (1989). The extended promoter of the gene encoding ribosomal protein S33 in yeast consists of multiple protein binding elements. Nucl. Acids Res. 17, 74277439. Himmelfarb, H. J., Simpson, E. M. and Friesen, J. D. (1 987j. Isolation and characterization of temperaturesensitive RNA polymerase I1 mutants of Sacchuromyces cerevisiae. Mol. Cell. Biol. 7, 21 55-2164. Hu, J., Xu, Y., Schappert, K., et al. (1994). Mutational analysis of the PRP4 protein of Saccharomyces cerevisiue suggests domain structure and snRNP interactions. Nucl. Acids Rex 22, 17241734. Iyer, V. and Struhl, K. (1995). Poly(dA:dT), a ubiquitous promoter element that stimulates transcription via its intrinsic DNA structure. EMBO J. 14, 25702579. Kerpola, T. K. and Kane, C. M. (1991). RNA polymerase: regulation of transcript elongation and termination. FASEB J. 5, 2833-2842. Kraakman, L. S., Mager, W. H., Grootjans, J. J. and Planta, R. J. (1991). Functional analysis of the promoter of the gene encoding the acidic ribosomal protein L45 in yeast. Biochem. Biophys. Acta 1090, 204-210. Lue, N. F., Buchman, A. R. and Kornberg, R. D. (1989). Activation of yeast RNA polymerase I1 transcription by a thymidine-rich upstream element in vitro. Proc. Natl. Acad. Sci. USA 86, 486490. Mager, W. H. and Planta, R. J. (1990). Multifunctional DNA-binding proteins mediate concerted transcription activation of yeast ribosomal protein genes. Biochem. Biophys. Acta 1050, 351-355. McBroom, L. D. and Sadowski, P. D. (1994). Contacts of the ABFl protein of Succharomyces cerevisiue with a DNA binding site at MATa. J. Biol. Chem. 269, 16455-1 6460. S. NOURAINI ET AL. McBroom, L. D. B. (1993). PhD Thesis, University of Toronto. McKnight, G. L. and McConaughy, B. L. (1983). Selection of functional cDNAs by complementation in yeast. Proc. Nutl. Acad. Sci. USA 80. Mckune, K. and Woychik, N. (1994). Functional substitution of an essential yeast RNA polymerase subunit by a highly conserved mammalian counterpart. Mol. Cell. Biol. 14, 41554159. Mosrin, C., Riva, M., Beltrame, M., Cassar, E., Sentenac, A. and Thuriaux, P. (1990). The RPC31 gene of Saccharomyces cerevisiae encodes a subunit of RNA polymerase C (111) with an acidic tail. Mol. Cell. Biol. 10, 47374743. Munholland, J. M., Kelly, J. K. and Wildeman, A. G. (1990). DNA sequences required for yeast actin gene transcription do not include conserved CCAAT motif. Nucl. Acids Res. 18, 6061-6068. Pinto, I., Ware, D. E. and Hampsey, M. (1992). The yeast SUA7 gene encodes a homolog of human transcription factor TFIIB and is required for normal start site selection in vivo. CelZ 68, 977-988. Pinto, I., Wu, W. H., Na, J. G. and Hampsey, M. (1994). Characterization of sua7 mutations defines a domain of TFIIB involved in transcription start site selection in yeast. J. Biol. Chem. 269, 30569-30573. Richardson, J. P. (1993). Transcription termination. Crit. Rev. Biochem. Mol. Biol. 28, 1-30. Russell, P. R. (1983). Evolutionary divergence of the mRNA transcription initiation mechanism in yeast. Nature 301, 167-169. Schultes, N. P. and Szostak, J. W. (1991). A poly(dA.dT) tract is a component of the recombination initiation site at the ARG4 locus in Saccharomyces cerevisiue. Mol. Cell. Biol. 11, 322-328. Sinclair, D. A,, Kornfeld, G. D. and Dawes, I. W. (1994). Yeast intragenic transcriptional control: activation and repression sites within the coding region of the Succharomyces cerevisiae LPDl gene. Mol. Cell. Biol. 14, 21&225. Struhl, K. (1989). Molecular mechanisms of transcriptional regulation in yeast. Ann. Rev. Biochem. 58, 1051-1077. Trawick, J. D., Kraut, N., Simon, F. R. and Poyton, R. 0. (1992). Regulation of yeast COX6 by the general transcription factor ABFl and separate HAP2- and heme-responsive elements. Mol. Cell. Biol. 12, 2302-2314. Willett, C. E., Gelfman, C. M. and Holland, M. J. (1993). A complex regulatory element from the yeast gene E N 0 2 modulates GCRl-dependent transcriptional activation. Mol. Cell. Biol. 13, 2623-2633. Woychik, N. A,, Liao, S.-M., Kolodziej, P. and Young, R. A. (1990). Subunits shared by eukaryotic nuclear RNA polymerases. Genes. Dev. 4, 313-323.