Nallonal Ubrary 81blloth~"Que n3tlon31e of Canada du Canada

Acquisitions and Direction des acq J1Sliions cl 81bllOgrç;phlC ServIces Branch des services blbllU\]raph\Qucs

395 Welllnqton StrL'c! 3~5. rUt,' \\'clhnqton OnJw,J, On:;1rIO OI:.lW~l (On13f10) K1AON': K1AON·; " ~ " ',.,' ,",' " " "

.'.., , " "'.' " ''''''''', " NOTICE AVIS

The quality of this microform is La qualité de cette microforme heavily dependent upon the dépend grandement de la qualité quality of the original thesis de la thèse soumise au submitted for microfilming. microfilmage. Nous avons tout Every effort has been made to fait pour assurer une qualité ensure the highest quality of supérieure de reproduction. reproduction possible.

If pages are missing, contact the S'il manque des pages, veuillez university which granted the communiquer avec l'université degree. qui a conféré le grade.

Sorne pages may have indistinct La qualité d'impression de print especially if the original certaines pages peut laisser à pages were typed with a poor désirer, surtout si les pages typewriter ribbon or if the originales ont été university sent us an inferior dactylographiées à l'aide d'un photocopy. ruban usé ou si l'université nous a fait parvenir une photocopie de quamé; inférieure.

Reproduction in full or in part of La reproduction, même partielle, this microform is governed by de cette microforme est soumise the Canadian Copyright Act, à la Loi canadienne sur le droit R.S.C. 1970, c. C-30, and d'auteur, SRC 1970, c. C-30, et subsequent amendments. ses a.nendements subséquents.

Canada 1 • Characterization of a binding site involved in the regulation of elongation within the murine c--mye gene

by

DANIEL DUFORT

Department of Medicine Division of Experimental Medicine McGill University Montreal, Quebec

December 1993

A Thesis submitted to the Faculty of Graduate Studies and Research in partial fulfilment of the requirements of the degree of Doctor of Philosophy

© Daniel Dufort 1993 • National Ubrary 8lbltoth~que nationale 01 Canada du Canada

AcquIsitions and DIrection des acqwslllons ct Bibliographie Services Branch des services bibliographiques

395 Weill; IQlon Streel 395. rue Wellington Ottawo..Ontano On;lwa (Ont~Hlo) K1A ON4 KlA ON·1

The author has grc:nted an L'auteur a accordé une licence irrevocable non-exclusive licence irrévocable et non exclusive allowing the National Library of permettant à la Bibliothèque Canada to reproduce, loan, nationale du Canada de distribute or sen copies of reproduire, prêter, distribuer ou his/her thesis by any means and vendre àes copies de sa thèse in any form or format, making de quelque manière et sous this thesis available to interested quelque forme que ce soit pour persons. mettre des exemplaires de cette thèse à la disposition des personnes intéressées.

The author retains ownership of L'auteur conserve la propriété du the copyright in his/her thesis. droit d'auteur qui protège sa Neither the thesis nor substantial thèse. Ni la thèse ni des extraits extracts from it may be printed or substantiels de celle-ci ne otherwise reproduced without doivent être imprimés ou his/her permission. autrement reproduits sans son autorisation.

ISBN 0-3IS-94613-X

Canada Ac-mye element involved in transcription elongation Il • Abstract The c-mye gene was previously shown to be regulated by a condilional block to transcription elongation and sequences from its promoter were implicated in this regulation. The objective of this project was to define the promoter elements involved in the control of transcription elongation. Using heterologous promoter constructs, the MElal protein binding site located in the c-mye promoter was shown to be required for the block to transcription elongation. From mutation analysis, a correlation was established between the binding of nuclear factors to MElal sites in vitro and the ability of these sites to confer block to transcription elongation in vivo, strongly implicating trans-acting factors in this process. Fractionation studies demonstrated that three nudear factors interact with the MElal site, thus generating three protein-DNA complexes termed "a", "b", and "c". These factors were characterized and a cDNA encoding the protein responsible for complex "c" was isolated. This protein was shown to represent the human homologue of the Drosophila Cut homeodomain protein (hu-Cut) and to repress expression from the c-mye promoter in transient transfection assays.

• 111 • Résumé L'expression du gène c-mye est régulée en partie au niveau de l'élongation de la transcription. Le but de ce projet était de définir les éléments du promoteur impliqués dans le contrôle de l'élongation de la transcription. J'ai montré que le blocage de la transcription était augmenté lorsque le site MElal, une séquence de 25 pb présente dans le promoteur de c­ mye, était introduit dans des constructions contenant un promoteur hétérologue lié à l'exon 1 de c-mye. Une analyse mutationnelle a permis d'établir une corrélation entre l'interaction avec des facteurs nucléaires in vitro et la capacité de conférer le blocage de la transcription in vivo. Ces résultats suggéraient que des facteurs de transcription se liant au site MElal pouvaient jouer un rôle dans le contrôle de l'élongation. Des études de fractionnement ont révélé que trois facteurs nucléaires pouvaient interagir avec le site MElal pour former les complexes "a", ''b'' et "c", Ces facteurs ont été analysés et le cDNA codant pour la protéine formant le complexe "c" a été isolé. J'ai montré que cette protéine était i'homologue humain de l'homéoprotéine Cut de la drosophile et qu'elle pouvait réprimer l'expression de gènes rapporteurs contenant le site MElal ainsi que l'exon 1 de c-mye.

• 1 \' • Acknowledgments 1 would like to thank my wife Celine and my little girl Alexandra for their love, understanding, and patience throughout these studies. 1 wish to express my gratitude to my supervisor, Dr, Alain Nepveu, for his advise and for giving me the opportunity to do this work. 1 would aIse like to thank the members of the Nepveu lab for their stimulating discussions and advise as weIl as for their friendship. Finally, 1 wish to thank the National Cancer Institute of Canada for financial support during these studies.

• v • Preface The Guidelines Concerning Thesis Preparation Issued By The Faculty Of Graduate Studies And Research At McGill University reads as follows:

''The candidate has the option, subject to the approval of their department, of including as part of the thesis, copies of the text of a papers(s) submitted for publication, or clearly-duplicated text of a published paper(s), provided that these copies are bound as an integral part of the thesis. If this option is chosen, connecting texts, providing logical bridges between different pages, are mandatory. The thesis must conform to all other requirements of the "Guidelines Concerning Thesis Preparation" and should be in a literary form that is more than a mere collection of manuscripts published or to be published. The thesis must include, as separate chapters or sections: (1) a Table of Contents, (2) a general abstract in English and French, (3) an introduction which clearly states the rationale and objectives of the study, (4) comprehensive general review of the background literature to the subject of the thesis, when this review is appropriate, and (5) a final conclusion and/or summary. Additional material (procedural and design data, as weIl as descriptions of equipment used) must be provided where appropriate and in sufficient detail (e.g.. in the appendices) to allow a clear and precise judgment to he made of the importance and originality of the research • reported in the thesis. YI

In the case of manuscripts co-authored by the candidate and others, the • candidate is required to make an explicit statement in the thesis of who contributed to such work and to what extent; supervisors must attest to the accuracy of such daims at the Ph.D. Oral Defense. Since the task of the Examiners is made more difficult in these cases, it is in the candidate's interest to make perfectly dear the responsibilities of the different authors of co-authored papers."

l have chosen to write my thesis according to to the above quoted option with one paper published and two submitted for publication. The thesis is organized into five chapters. Chapter l is a general introduction and literature review with references. Chapters II-IV contain furee manuscripts, each with its own abstract, introduction, methods, results and references. Chapter V is a general discussion of ail the results with references.

Severa! people have contributed to the work presented in this thesis. In the second cb.apter, "A Protein Binding Site from the Mutine C- Promoter Contributes to Transcriptional Block", Marc Drolet made the MElal-Sac construct and we both obtained stably transfected ceU population and we both performed the nueler run-on assays. Also, the MEC-MElal+ and MEC-MElal­ constructs were made by Harvey Miller, and the cell populations were established by Alain Nepveu who performed the nuelear run-on assays as weil as the SI mapping of the cytoplasmic RNA for these constructs. In the third chapter, ''The human Cut homeodomain protein represses transcription fron the c-myc promoter~', the MElal-eAT construct was made by Ryoko Harada, and the MElal-El-eAT construct was made by Ginette • Bérubé. Vil • Contributions to original research 1. The MElal protein binding site located in the promoter of the c-mye gene was shown to be important for transcriptional block.

2. A correlation was established between the ability of the MElal site to bind cellular factors in vitro and the ability to confer block to transcription elongation in vivo, strongly suggesting that cellular factors are involved in transcriptional block.

3. By fractionation of crude HeLa nuclear extracts three cellular factors were shown to bind to the MElal site generating three protein-DNA complexes

termed "a", "b", and "e".

4. Characterization of the factors responsible for complexes "a" and ''b''.

5. Isolation of ~e cDNA encoding the factor responsible for complex "c" which encoded the human homologue of the Drosophila Cut homeodomain protein (hu-Cut).

6. hu-Cut was shown to repress expression from the c-mye promoter. The MElal site as weil as exon 1 sequences were shown to be required for repression by hu-Cut. • VIlI • Publications arising from work of the thesis

1. Dufort, D., M. Drolet, and A. Nepveu. 1993. A protein binding site from the murine c-myc promoter contributes to transcriptional block. Oncogene 8:165­ 171.

2. Dufort, and A. Nepveu. 1993. The human Cut homeodomain protein represses transcription from the c-myc promoter. Mol. Cell. Biol. (in press)

3. Dufort, and A. Nepveu. 1993. Three factors can interact with the ME1a1 site within the c-myc promoter. Nucleic Acids Res. (submitted)

• ix • Table of content Abstract ii Résumé iii Acknowledgments iv Preface v Contributions to original research vii Publication arising from work of the thesis viii Table ofContents ix List of Figures xii List ofTables xv

Chapter1 Introduction 1 Transcription elongation and termination in 2 A Transcriptional pausing 3 B. Transcription termination 4

1 Intrinsic terminators 4

2 Rho-dependent terminators 6 C Elongation factors 7 D. Regulation of transcription elongation 9 and termination 1 Antitermination 10 2 Attenuation 15 3 Transcriptionalroadblocks 20 Transcription elongation and termination in eukaryotes 21 A. Termination by RNA polymerases 1 and m 22 • B. Termination by RNA polymerase II 23 x

C Termination signaIs recognized by 26 • RNA polymerase II in vitro D. RNA polymerase TI elongation factors 27 E. Regulation of Transcription Elongation in Eukaryotes 29

1 Antitermination in HIV 30 2 Transcriptional roadblock 34

3 Block to transcription elongation 35

4 Transcription elongation in the c-mye gene 40

References 44

Chaptern A Protein Binding Site from the Murine C-Myc 80 Promoter Contributes to Transcriptional Block Preface 80 Abstract 81 Introduction 82 Materials and Metl,ods 84 Results 87 Discussion 91 References 95 Figures 101

ChapterID The human Cut homeodomain protein represses 109 transcription fron the c-mye promoter Preface 109 Abstract 110 Introduction 111 • Materials ani Methods 113 xi

Results 117 • Discussion 120 References 125 Figures 134

Chapter IV TItree factors can interact l\

Chapter V General Discussion 165 References 173

• xii

List of Figures • ChapterI

Figure 1 Genetic map of lambda 10

Figure 2 Antitermination complexes 13 assembled by lambda N protein

Figure 3 Potential models for antitermination 16 by lambda Q protein

Figure 4 Map of the murine c-mye gene 42

ChapterII

Figure 1 Nliclear mn-on transcription analysis 101 of the constmct

Figure 2 SI nuclease mapping of the transcription 103 start sites utilized by the constmets

Figure 3 EMSA using the MElal and mutant MElal 105 double-stranded oligonucleotides

Figure 4 The MElal sequence functions either upstream 106 or downstream of the initiation site

Figure 5 Comparison of the transcription start sites 107 utilized by the MEC and MEC-Sac constmets

Figure 6 Promotion of termination at a distant site IDS • xiii

ChapterIII • Figure 1 Fractionation of HeLa nuclear extract 134 by size exclusion chromatography

FigureZ Binding of hu-Cut to MElal 135

Figure 3 Co-transfection experiments with c-myc 136 promoter constructs and hu-Cut

Figure 4 Co-transfections with minimal 137 promoter constructs

ChapterIV

Figure 1. Competition assay with MElal and 156 SpI oligonucleotides

Figure 2. Determination of the rate of dissociation of 157 the bound to the SpI and MElal sites

Figure 3. Binding of the cellular factor to the MElal 158 site is very stable

Figure 4. MElal oligonucleotides do not compete for 159 binding to the SpI probe in dissociation experiments

FigureS Fractionation of HeLa nuclear extract by 160 • size exclusion chromatography Xl\'

Figure 6. High molecular weight factors inhibit the 162 • binding of the factor responsible for complex b

Figure 7 uv cross-lin.1cing of the factor responsible for 164 the formation of complex b

• xv • List of Tables ChapterII

Table 1 Percentage of transcriptional block 102

Table2 Sequence comparison of the oligonucleotides 104

ChapterIV

Table1 Summary of the sequence requirements for the 161 binding of the cellular factor responsible for complex a

Table2 Summary of the sequence requirements for the 163 binding of the cellular factor responsible for complex b

• 1 • INTRODUCTION Transcription is a process whereby RNA is synthesized using DNA as a template. The process of transcription can be divided in three steps; initiation, elongation, and termination. The first step, initiation, involves the assembly of a functional transcriptional complex at the transcription initiation site. Elongation begins when the transcription complex has assembled and the first phosphodiester bond is formed. The elongating complex will continue RNA synthesis until termination signaIs are encountered. Termination occurs when the transcriptionaI complex stops RNA synthesis and dissociates from the DNA template. The liberated RNA polymerase can th.~n go on and initiate transcription at other promoters. Many genes have been shown to be regulated at the level of transcription. Regulation can occur either at the level of initiation or elongation. Regulation at the level of transcription initiation has been studied extensively, and we now have a better understanding of the various mechanisms involved. It has clearly been established that the amount of initiation depends on the interaction of trans-acting factors with upstream or downstream cis-acting regulatory elements. By regulating the interaction of trans-acting factors with these cis-acting regulatory elements, the levels of transcription initiation can be modulated. However, even if transcription has initiated, it does not mean that it will go to completion. That is, there is aIso regulation at the level of transcription elongation. Unlike the regulation of transcription initiation, little is known on how transcription is regulated at the level of transcription elongation. In prokaryotes, regulation of transcription elongation has long been recognized as being important in • regulating gene expression, but only recently has it been recognized in 2

eukaryotes. With the increasing number of genes found to be regulated at this level, more attention has turn towards understanding how transcription • elongation can be regulated in eukaryotes. Oruy now are we beginning to understand sorne of the mechanisms involved in the regulation of elongation. The following sections will review most of the important features on the regulation of transcription elongation and termination in prokaryotes and what is presently known in eukaryotes.

Transcription elongation and termination in prokaryotes

Oruy one known RNA polymerase is responsible for transcription in E. coli. Core RNA polymerase is a multisubunit complex composed of 2 a subunits, a ~ and a Wsubunit (Burgess et al., 1969; Chamberiin, 1976). The catalytic center of RNA polymerase is composed of both the ~ and ~. subunits which share extensive homology with the two largest subunits of eukaryotic RNA polymerase (Allison et al., 1985; Biggs et al., 1985). The ~ subunit interacts with DNA, the nucleotide and the RNA Gin and Gross, 1988; Kashlev et al., 1990; Landick et al. 1990; Lee et al., 1991; Atkinson and Gottesman, 1992), and mutations in the ~' subunit affect regulation by elongation factors Gin and Gross, 1989; !to et al. 1991). The a subunits are believed to be involved in the control of initiation by activators and are required for assembly of the core RNA polymerase (Russo and Silhavy, 1992; Ishihama, 1992). To better understand the processes of transcription elongation and termination, it is important to define some of the terms which are commoruy • used in the literature. The term transcription termination refers to the process 3

where the transcriptional complex stops transcription and dissociates from • the template, releasing the nascent transcript. The term transcriptional pausing is used when the transcriptional complex stops elongating but remains bound to the template. Under appropriate conditions, these complexes can resume elongation. The term transcriptional block or block to transcription elongation is used when transcription stops but it is not known whether this is due to pausing or to termination.

A. Transcriptional pausing The rate of transcription elongation by the core RNA polymerase, as determined by the time required for transcribing different genes, is approximately 43 nucleotides/second (Davis and Hyman, 1970; Mosteller and Yanofsky, 1970; Gotta et al., 1991). However, in vitro transcription studies have demonstrated that the rate of elongation is not constant but can vary by

up to 1000 fold over different regions of a gene (Morgan et al., 1983; Landick and Yanofsky, 1984). Regions where the rate of elongation is lower are termed pause sites. Regions constituting pause sites in prokaryotes are usually 40-60 nucleotides in length (Landick and Yanofsky, 1984; Levin and Chamberlin, 1987; Lee et al. 1990). Two types of pause sites have been described. The first type is present immediately downstream of a region of dyad symmetry which has the potential to form a stable stem and loop, or hairpin structure, in the nascent RNA. These sites have been called hairpin-dependent pause sites. The formation of a hairpin structure is important for pausing, since mutations which affect the stability of this structure, reduce pausing (Landick and Yanofsky, 1984; Chan and Landick, 1989). The mechanism by which the hairpin structure causes RNA polymerase to stop elongation is not known, • but it has been proposed that it may destabilize the RNA:DNA heteroduplex 4

(Landick and Yanofsky, 1987). The second type of pause sites show no obvious • secondary structure suggesting that the DNA sequence rather than the RNA secondary structure is important for pausing (Yager and von RippeI, 1987). These have been termed sequence-dependent pause sites. Sequences immediately downstream of several pause sites have aiso been shown to be important for pausing (Lee et al., 1990). A comparison of different downstream sequences has not revealed any potential secondary structures or other sorts of patterns to explain how these sequences affect pausing (Lee et al.,1990). B. Transcription termination Transcription elongation and termination are closely associated in prokaryot~s. That is, sites that are recognized as terminators are often used in the regulation of transcription elongation. By modulating the ternary complex or the termination signaIs, transcription can either stop or read through these termination sites. Therefore, to understand how transcription elongation can be regulated it is important to understand the signaIs which lead to transcription termination. Two types of terminators have been identified in prokaryotes, intrinsic terminators which function in the absence of cellular factors, and Rho dependent terminators which require the cellular factor Rho. (Richardson, 1993).

1. Intrinsic terminators

Intrinsic or rho-independent terminators are recognized in 'Ditro in the absence of accessory factors. Although purified RNA polymerase can terminate transcription at intrinsic terminators, severa! factors have been identified which increase the overall efficiency of termination at these sites • (Briat and Chamberlin, 1984; Briat et al., 1987; Schmidt and Chamberlin, 1987). 5

A comparison of the sequences of several intrinsic terminators have • revealed sorne distinct motifs. Intrinsic terminators are composed of an inverted repeat which when transcribed can form a stable stem and loop structure, followed by a stretch of adenine residues on the template DNA strand (Brendel et al., 1986; Cafara et al., 1990). Mutations in the stem-loop structure which affect the base-pairing of the stem also affects the efficiency of termination, suggesting that this structure is important for termination (Lynn et al., 1988; Cheng et al., 1991). Similarly, the run of adenine residues is also an important constituent of an intrinsic since mutations of these residues also affects termination (Lynn et al., 1988). Finally, the sequences following the run of adenine residues may also be an important constituent of intrinsic terminators. Although these sequences are not transcribed they have been shown to influence the efficiency of termination (Telesnitsky and Chamberlin, 1989; Reynolds and Chamberlin, 1992). The mechanism of termination at intrinsic terminators is not weil understood. Two models have been proposed to explain how the structure of intrinsic terminators promotes termination. The first model proposes that it is the relative stabilities of RNA-DNA and RNA-RNA hybrids which determine the efficiency of termination (von Hippie and Yager, 1991). During elongation, it has been proposed that the nascent RNA hybridizes to the DNA, and the resulting RNA-DNA hybrid is 12 bps in length. The formation of the RNA stem-loop structure reduces the length of the RNA-DNA hybrid resultin~ in a destabilization of the ternary complex. Thus the crucial feature of this mode! is the relative stability of the RNA-DNA hybrid. In support of this model is the requirement for adenine residues on the template DNA strand. The RNA-DNA hybrid , rU:dA, is the least stable of ail possible • nueleotide pairs (Martin and Tinoco, 1980). Therefore, this hybrid can 6

dissociate upon ::ormation of a stem-Ioop structure, leading to a destabilization of the transcriptional complex which can lead to termination. • Recent experiments have suggested that the RNA-DNA hybrid may not be 12

bps but actually 2 or 3 bps in length (Rice et al., 1991). In this situation, the formation of the stem-Ioop structure would not affect the stability of the RNA-DNA hybrid, and an alternative model has been proposeù by Chamberlin and co-workers to account for these resu1ts (Reynolds and Chamberlin, 1992). This model proposes that the RNA polymerase has an RNA binding site which can interact with the nascent RNA. The formation of the stem-Ioop structure would affect the interaction between the nascent RNA and the polymerase resulting in the release of the nascent transcript. In this mode1, the crucial feature is therefore the interaction of the RNA polymerase with the nascent RNA.

2. Rho-dependent terminators The second type of termination signais that are recognized by prokaryotic RNA polymerase, are rho-dependent terminators (platt and Richardson, 1992). As the name implies, the accessory factor rho is required for termination. Rho is a hexameric protein composed of identical 46 kDa subunits (Finger and Richardson, 1982; Pinkham and Platt, 1983). Rho is an RNA binding protein with RNA-dependent ATPase activity, as weil as a relatively slow activity (Lowery and Richardson, 1977a; Finger and Richardson, 1982; Brennan et al., 1987). Rho binds to single stranded RNA and has a strong affinity for C residues (Lowery and Richardson, 1977b). RhO dependent terminators do not share striking sequence similarities, but are composed of two funetional elements, a site on the nascent RNA • where rho binds, and a downstream site ,,:hich functions as a pause site. The ï

R.l\TA characteristics required for rho binding are not weil defined, but • sequences upstream of the termination site are devoid of secondary structure and usually rich in residues and poor in guanosine residues (Alifano et al., 1991). The high cytosine content probably enables rho to bind the nascent RNA. Rho promotes termination at sites which function as pause sites. These pause sites can be either hairpin-dependent or sequence­ dependent panse sites (Yager and von Hippel, 1987; Levin and Chamberlin, 1987). Rho promotes termination more efficiently at stronger pause sites and less efficiently at weaker pause sites (Rosenberg and Court, 1979; Court et al., 1980) suggesting that pausing of RNA polymerase is important for termination by rho. Two models have been proposed to explain how rho bound to upstream sequences can influence terminatioll at sites further downstream. In the first model, rho bound to upstream RNA sequences interacts with the RNA polymerase located at the pause site by looping of the RNA (Richardson, 1990). In the second model, rho translocates along the nascent RNA using energy from ATP hydrolysis, until it reaches the paused polymerase (Platt and Richardson, 1992). The mechanism by which rho causes termination is not known. It has been suggested that rho, through its helicase activity, dissociates the 12 nt RNA:DNA hybrid resulting in a destabilization of the ternary complex leacili.lg to transcript release (Brennan et al., 1987). Alternatively, rho may interact with the RNA polymerase and cause the dissociation of the RNA from the proposed RNA binding site on the polymerase.

C. Elongation factors Severa! cellular factors have been identified which affect the rate of • elongation by RNA polymerase. The best characterized of these elongation 8

factors in is NusA. NusA is a protein with a molecular weight • of 55 kDa (Ishii et al., 1984) which promotes or prolongs pausing at several naturally occurring pause sites (Kassavetis and Chamberlin, 1981; Kingston and Chamberlin, 1981; Lau et al., 1983). NusA binds to core RNA polymerase but not to the holoenzyme (Greenblatt and Li, 1981a; Gill et al., 1991) suggesting that it must bind to the same site as the , or to an adjacent site on the polymerase. NusA has two distinct effects on the elongation by RNA polymerase. One effect of NusA is to reduce the overall rate of elongation. At low dN1P concentrations NusA reduces the rate of elongation but its effect can be overcome by high concentrations of dNTP suggesting that NusA influences the rate of nucleotide binding by RNA polymerase (Schmidt and Chamberlin, 1984). The second effect of NusA is to promote pausing at pause sites where the RNA can form a stable stem loop structure (Landick and Yanofsky, 1984; Chan and Landick, 1989; Lau et al., 1983; Landick and Yanofsky, 1987). It is not known how NusA promotes pausing but two mechanisms have been proposed. Binding of NusA to the polymerase may stabilize the interaction of the RNA stem-loop with the polymerase, resulting in prolonged pausing. Altematively NusA may bind to the RNA stem-Ioop and stabilize its interaction with the transcriptional complex. However, binding of NusA to RNA has not yet been demonstrated. Aside from affecting elongation, NusA is also involved in termination at certain intrinsic terminators. For example, the termination efficiency of RNA polymerase at the tR2 terminator of lambda is approximately 45% but is increased to 90% at saturating NusA concentrations (Levin et al., 1987). NusA, as will be discussed further, is also involved in • antitermination in lambda. 9

Two elongation factors, GreA and GreB, have recently been isolated as • factors which confers transcript deavage activity to certain transcriptional complexes (Borukhov et al., 1992; Borukhov et al., 1993). Recently, a dass of arrest signaIs have been identified which lock the RNA polymerase on the template without releasing the nascent transcript. Those RNA polymerases which are arrested at these sites are termed dead-end complexes (Arndt and Chambertin, 1990). Certain complexes halted at these sites, in the presence of

Mg2+, catalyze an endonudeolytic deavage of the nascent transcript which allows the complexes to resume elongation (Surratt et al., 1991). Two types of cleavages have been detected, one in which 2-3 nts are deaved, and one where 10-11 nts are deaved from the 3' terminus of the nascent transcripts. Two factors have been identified, that associate with RNA polymerase and enhance this cleavage activity. These are the GreA protein which catalyses the deavage of 2-3 nts (Borukhov et al., 1992), and GreB which catalyses the deavage of up to 9 nts (Borukhov et al., 1993). GreA and GreB prevent the arrest of transcription elongation by different mechanisms. GreB functions when added after the polymerase complex has arrested by directly deaving the nascent transcript leading to the restart of elongation. The action of GreA is more complex. It cannot rescue arrested complexes, but can prevent their arrest if added before they have reached the state of arrest. These two factors are most probably the functional equivalents of the eukaryotic transcription elongation factor ns (TFllS) which will be discussed further.

D. Regulation ofTranscription Elongation in Prokaryotes.

Three general mechanisms are used to regu1ate transcription elongation: • antitermination, attenuation, and steric interference to polymerase progress. 10

Antitermination is a mechanism by which the transcriptional complex is modified to a form that can transcribe through terrnination signaIs (Da!', 1992; • Greenblatt, 1992; Roberts, 1992; Greenblatt et al., 1993). In attenuation, the ability of the transcriptional complex to read through specific sites is determined by the ability of the nascent RNA to adopt a specific conformation leading to the termination of transcription (Landick and Turnbough, 1992). Finally, binding of factors to downstream sequence elements can prevent the transcriptional complex from elongating past these sites, leading to premature termination of transcription.

1. Antitermination Antitermination has best been studied in bacteriophage lambda 0.. ). Downstream of the two early promoters, pL and pR, are several terminators which can efficiently terminate transcription in the absence of the bacteriophage encoded N protein (Fig. 1). However, N protein can modify the transcriptional complex to an antitermination form which can efficiently transcribe through the termination sites. Similarly in the late promoter, pR', the bacteriophage Q protein can prevent termination during the transcription of the late (reviewed in Das, 1992; Greenblatt, 1992; Roberts, 1992; Greenblatt et al., 1993). +Q l'''I---l''' nutL nutR -Q ...... ,.,.....J< exo clll N t cl cre t cil 0 P Q SR A~, tl3 tl2tl1 pL pR tR1 tR2tR3pR'tR' ~ 1 1 _ qut ~ ..~ -N ...... ·····I ~f-- I+N IL...- ...

Figure 1. Genetic map of lambda. Partial genetic map illustrating the • antitermination effects of N and Q proteins on transcription elongation. 11

The N protein can modify the transcriptional complex only once it has • transcribed a site termed N-utilization (nut) (Barik et al., 1987). The nut site can be subdivided into two elements, boxA and boxB (Friedman, 1988; Roberts, 1988; Greenblatt et al., 1993). BoxA is conserved in antiterminator

elements of other phages as weil as in the ml operon of E coli (Li et al., 1984; Franklin, 1985; Morgan, 1986). BoxB is specific to lambda and can form a stem­ loop structure to which N protein can bind (Lazinski et al., 1989; Oas, 1992). Aside From binding to the nascent RNA, N protein also interacts with NusA (Greenblatt and Li, 1981b). As described earlier, NusA interacts with the core

RNA polymerase shortly after the release of the initiation factor 0-7°. Therefore, N protein is thought to interact with the transcriptional complex through its association with NusA. ln vitro, N and NusA are sufficient to cause antitermination at rho-dependent and -independent terminators, provided that they are close to the nul site (Fig. 2a) (Whalen et al., 1988; Mason et al., 1992a). This non-processive antitermination complex is in contrast to the highly processive complex observed in vivo which can readthrough termination sites located as far as 5-10 kilobases downstream of the nul site (Friedman, 1988). Although N and NusA are sufficient for antitermination at a short distance From the nul site, other E coli proteins are required to assemble a processive ternary antitermination complex. These include NusB, NusG, and the ribosomal protein 510 (NusE) (Li et al., 1992; Mason and Greenblatt, 1991; Mason et al., 1992b). NusG and 510, like NusA, have been shown to bind weakly to RNA polymerase (Li et al., 1992; Mason and Greenblatt, 1991). Also, NusB can form heterodimers with 510 (Mason et al., 1992b) and the resulting NusB-510 can bind to the rrn boxA (Nodwell and Greenblatt, 1993). However, NusB-510 does not bind to boxA of lambda in the • absence of boxB (Nodwell and Greenblatt, 1993). Binding to Â. boxA by NusB- 12

S10 requires N protein as well as NusA and NusG (Nodwell and Greenblatt, 1991). • The current mode! for how the transcriptional complex is modified is as follows (Fig 2b): NusB-S10 binds to boxA and N protein binds to boxB of the

nut site, and by RNA looping, these factors interact with RNA po!ymerase. N proèein binds to NusA, and NusB-S10 binds directly to RNA polymerase through S10. Finally, NusG may stabilize the ribonucleoprotein complex (Greenblatt et al., 1993). The mechanism of antitermination by these proteins in not known. As described previously, pausing is an important step for both rho-dependent and -independent terminators. It has been demonstrated that N protein can suppress pausing at rho-dependent terminators, and it has been suggested that it is this activity which prevents termination (Mason et al., 1992a). It is unclear how N protein prevents pausing, but it has been suggested that either N or boxB RNA prevents the interaction of the nascent RNA with the RNA polymerase which is required for pausing. Alternatively, the N protein may stabilize a conformation of the polymerase which prevents pausing. It has recently been demonstrated that NusG binds to rho (Li et al., 1993), and is required for rho-dependent termination both in uïtro and in 'Di'DO (Sullivan and Gottesman, 1992; Li et al., 1993). Therefore, at rho-dependent terminators, rho may interact with NusG on the RNA polymerase to induce transcription termination. The presence of N protein on the transcriptional

complex in antitermination may affect NusG on the complex 50 that rho binding te it does not induce termination (Li et al., 1993). Antitermination by the 1.. Q protein is not as complicated as antitermination by 1.. N protein. RNA polymerase initiates transcription at the • pRO promoter and pauses for several minutes downstream of the initiation 13 •

Figure 2. Antitermination complexes assembled by lambda N protein. a. A non-processive complex containing N and NusA protein. b. A processive antitermination complex containing N, NusA, NusB, NusG, and the ribosomal protein 510. The potential interaction between rho and NusG is illustrated. Figure taken from Greenblatt et al., 1993.

• •

a ..

­,.".-- ..

..

• 14

site (Roberts, 1988). In the absence of Q, RNA polymerase leaving this site will • terminate transcription at the tR' terminator, located further downstrearn. The paused RNA polymerase can be recognized by Q which can modify the transcriptional complex to leave the pause site and prevent pausing at both rho-dependent and -independent terminators (Roberts, 1988; Yang and Roberts, 1989). This ability of Q protein to prevent pausing is believed to be important in the mechanism leading to antitermination. Unlike N protein which interacts with the nascent RNA, Q binds to DNA. The site required for antitermination by Q, termed qui, is located within the pR' promoter, spanning sequences both upstream and downstream of the transcription initiation site, between bases -26 to +18 (Yang et al., 1987). Mutations within this site abolishes Q function (Kainz and Roberts, 1992; Yamell and Roberts, 1992). In contrast to N which requires several host proteins, Q is sufficient to promote antitermination, however NusA is essential to prevent pausing at certain terminators ;Grayhack et al., 1985; Yang and Roberts, 1989). It is not known how Q can prevent termination at sites located further downstream, however, several models have been proposed (Fig. 3) (Greenblatt et al., 1993). Uke N protein which can contact the lranscriptional complex by RNA looping, Q may contact the complex by DNA looping (model a). However, Q protein bound to the promoter can interact with only one transcriptional complex at a time. Altematively, Q may physically associate with the RNA polymerase and become part of the transcriptional complex (model b), or without becoming part of the complex, modify the RNA polymerase to a forro which can read-through termination sites (model c). Antitermination has aiso been shown to occur during transcription of • the ribosomal (rrn) of E. coli (Albrechtsen et al., 1990). RNA 15

polymerases which transcribe the rm operon are modified in a way which • enables them to transcribe through certain terminators (Morgan, 1980), but not the strong intrinsic terminator at the end of the operon (Li et al., 1984).

Two sequences which are nearly identical to boxA of the nut site of À have beer:. identified (Li et al., 1984; Berg et al., 1989; Albrechtsen et al., 1990). One is located just downstream of the initiation site, and the other is located in the intergenic region between the 165 and 235 rRNA genes. The presence of a

boxA site suggests that the mechanism of antitermination of the Tm operon is

similar to that of À N protein, and involves the formation of a ribonucleoprotein complex (Albrechtsen et al., 1990; Nodwell and Greenblatt, 1993). 5upporting this suggestion is the recent demonstration that NusB-510 heterodimers bind to boxA of the rmG operon (Nodwell and Greel1blatt, 1993). Also, mutations in the nusB gene affect antitermination in the rrn operon (5harrock et al., 1985). Therefore, the mechanism of antitermination

in the Tm operon may be similar to that of N in À. The N protein may not be

required for the assembly of the ribonucleoprotein complex in the Tm operon,

because of the ability of the NusB-510 heterodimers to bind to Tm boxA. In À,

sinee NusB-510 heterodimers do not bind to À boxA, N protein binding to boxB may be required for the assembly and stabilization of the ribonucleoprotein complex.

2. Attenuation Attenuation, in the regulation of gene expression in prokaryotes, is a mechanism where the level of transcriptional termination at a specifie site within an operon, called an attenuator, is regulated in response to a • physiologieal signal (Bauer et al., 1983). Through the analysis of various 16 •

Figure 3. Potential models for antitermination by lambda Q protein. Q prcto.in bound to the qut site interacts with the RNA polymerase pause at +16 downstream of the initiation site. A. By DNA looping, the Q protein bound to the qut site contacts the RNA polymerase as it transcribes further downstream. B. Alternatively, Q protein may become part of the transcriptional complex. C. Q may modify the RNA polymerase to a form which can read-through termination sites. Figure taken from Greenblatt et al., 1993.

• • - ...... - """Iii'.. il Q bIndI ta ...DNA'" CMIIICII JDIA F'.-

- ."""""",,::....

c Gtm--~~-§Jm=- .... ---; _1lHA::=:;-_ wcltM

• li

bacterial operons several mechanisms of attenuation have been identified • (Landick and Turnbough, 1992). These mechanisms can be grouped in four different classes: 1) mechanisms where the position of the translating will govern the formation of either a secondary structure which will cause transcriptional termination or a secondary structure which precludes termination, 2) mechanisms where the position of the ribosome can physicaIly block the formation of the termination signal, 3) mechanisms where a trans-acting factor rather than a ribosome can interact with the nascent RNA and govern the formation of the termination signaIs, 4) mecr.anisms where the activity of rho-dependent terminators are modulated to control gene expression (Landick and Turnbough, 1992). Ali four mecnanisms are similar to the extent that in each case, it is t:te termination signal which is modulated and not the transcriptional complex which is modified. In many amino acid biosynthetic operons ( including trp, his, leu, tllr, ilvGMEDA, ilvBN), and in the pheST operon, which encodes phenylalanyl­ tRNA synthetase, attenuation is governed by the position of the translating ribosome which controls the formation of two mutually exclusive secondary structures that affect transcription termination. In ail of these operons, an intrinsic terminator is located in the leader region between the promoter and the first structural gene. The formation of the hairpin structure required for the function of the intrinsic terminator is determined by the position of the ribosome translating the leader peptide coding region. Within the coding region of the leader peptide are control codons which correspond to the

amino acid end product of the operon or, in the case of the pheST operon, the substrate. For example, in the , tryptophan codons serve as t~e • control codons. When the level of tryptophan is high, the ribosome can 18

quickly reach the end of the leader peptide coding region and al10w the • formation of the terminator hairpin structure which resu1ts in the termination of transcription. However, when the level of tryptophan is low, the ribosome stal1s at the control codons and allows the formation of an alternate readthrough conformation which prevents the formation of the terminator hairpin structure and allows transcription to proceed in the operon. A second mechanism of attenuation has been identified in transcription of the pryBI operon of E. coli which is regulated by the intracellular concentrations of UTP (Schartz and Neuhard, 1975; Tumbough, 1983). Located witt'.in the leader region is an intrinsic terminator which is preceded by a strong transcriptional pause site (Tumbough et al., 1983). The transcriptional pause site encodes a long uridine rich sequence which slows down transcription at low UTP concentrations. When the intracellular concentration of UTP is low, the transcribing RNA polymerase stalls in the uridine rich region. In this situation, the ribosome can translate up to the stalled polymerase. When the RNA polymerase leaves the pause site, the ribosome translates so close to the polymerase that the terminator hairpin of the intrinsic terminator is unable to form, allowing transcription of the structural genes. When the intracellular concentrations of UTP is high, the RNA polymerase does not pause and transcribes ahead of the translating ribosome. In this situation, the i:~rminator hairpin can form and transcription terminates. RNA binding proteins are also involved in attenuation, a mechanism termed regulatory factor-dependent attenuation. The best characterized RNA binding protein involved in attenuation is the BgIG protein, which can • inhibit termination at an intrinsic terminator in the bgl operon in E. coli 19

(Mahadevan and Wright, 1987; Houman et al., 1990). BgIG binds to a site on • the nascent bgl RNA which precedes or partially overlaps the hairpin structure of the intrinsic terminator. Binding of BgIG prevents the formation of the hairpin structure and allows expression of the operon (Amster-Choder and Wright, 1990; Houman et al., 1990). The Mtr protein has also been demonstrated to function in a similar manner in regulating expression of the tryptophan operon in Bacillus subtilis (Shimotsu et al., 1986; Gollnick et al., 1990). The three previous mechanisms of attenuation involved the regulation of the activity of intrinsic terminators. In several operons, induding tna and liv, the activity of a rho-dependent terminator is regulated (Landick and Turnbough, 1992). This is the fourth dass of attenuation mechanism, termed rho-dependent attenuation. The best characterized is the tna operon which is induced by high levels of tryptophan. The leader region of the tna operon codes for a 24 amine aàd leader peptide which contains a tryptophan residue at position 12 (Gollnick and Yanofsky, 1990). The leader region also contains a rho-dependent terminator. In the presence of tryptophan.. expression of the

tna operon is induced by preventing rho-dependent termination in the leader region (Stewart and Yanofsky, 1985). Although the exact mechanism of induction is not known, it is believed to be the result of ribosome occlusion of the rho binding site on the RNA. At low tryptophan concentrations, the ribosome translating the leader peptide stalls at position 12, allowing rho to bind to the RNA and terminate transcription. When tryptophan levels are high, the ribosome does not staIl and translates close to the polymerase preventing binding of rho to the RNA, thus allowing the polymerase to • continue transcription of the operon. 20

3. Transcriptional roadblocks In severa! operons, proteins bind to operator sites located in • the promoter and inhibit the initiation of transcription (Gralla, 1989a). Alternatively, may bind to operator sites located both upstream and downstream of the transcription initiation site, and by DNA looping prevent both the initiation and elongation of transcription (Gralla, 1989b; Hochschild, 1990). In severa! genes however, operator sites are located in the coding region, downstream of the transcription initiation site. Binding of a repressor to the operator site within these genes, results in a block to elongation of transcription upstream of the binding site. For these genes, transcription e!ongation is regulated by sterically hindering polymerase progress, a mechanism referred to as transcriptional roadblock. A transcriptional roadblock has been identified in two genes, the lac 1 gene and the purB gene of Escherichia coli (Deuschle et al., 1986; Sl!llitti et al., 1987; He and Zalkin, 1992). It is important to note that the simple binding of a factor to DNA is not sufficient to block transcription elongation. Transcription has been shown to displace proteins bound to DNA. Therefore, factors which interfere with polymerase progress not only bind to DNA, but also affect the transcriptional complex to stop elongation.

The gene encoding the (lac!) is located just upstream of the

, which encodes three structural genes, lacZ, lacY, and lacA. Between the end of the repressor coding region and the transcription initiation site of lacZ is an 84 nucleotide control region where RNA

polymerase, cAMP protein (eRP), and the lac repressor have been

shown to bind (Reznikoff and Abelson, 1980). Binding of the lac repressor to this site blocks transcription elongation and prevents readthrough • transcription into the lac operoJ;\ (Sellitti et al., 1987). In the presence of lac 21

repressor, the major lacI mRNA species synthesized have 3' ends which map • to positions upstream of the repressor binding site, whereas in the absence of lac repressor there is a 50 fold decrease in the abundance of these mRNAs (5ellitti et al., 1987). Lac repressor can also block elongation in a heterologous gene in which a single lac operator has been inserted (Deuschle et al., 1986). The Escherichia coli lac repressor can also block transcription elongation by eukaryotic RNA polymerase II (Deuschle et al., 1990). Therefore, lac repressor binding to the lac operator, in addition to blocking initiation of transcription from the lacZ promoter, can also serves as a in the lacI gene by acting as a roadblock and inhibiting transcription elongation. Recently, repression of the purB gene has been demonstrated to occur via a transcriptional roadblock mechanism (He and zalkin, 1992). The purB gene is one of 10 genes encoded by seven operons required for de novo synthesis of IMP. These seven operons are regulated by a repressor encoded by

purR (He et al., 1990; Meng et al., 1990). In aIl these operons, except for purB, the operator is located in the promoter region, and binding of the repressor inhibits transcription initiation. The operator, in the purB operon is located 224 bp downstream of the transcription initiation site (He et al., 1992). Binding

of the repressor to the operator in the purB operon, blocks transcription

elongation (He and Za1kin, 1992). Therefore, the purR repressor regulates the expression of purB by acting as a roadblock for elongating transcriptional complexes.

Transcription Elongation and Termination in Eukaryotes

Eukaryotic cells contain three RNA polymerases, each transcribing a • specifie subset of genes (Young, 1991). RNA polymerase 1 synthesizes 22

ribosomal RNA, RNA polymerase II transcribes aH mRNA genes as weH as most snRNA genes, and RNA polymerase III transcribes the 5S ribosomal • RNA and tRNA genes. Each polymerase is composed of two large subunits and several smaller associated polypeptides, sorne of which are shared among aH three polymerases. Each polymerase recognizes different termination signais and the mechanisms of termination for aH three polymerases are different.

A. Termination by RNA polymerases l and III. RNA polymerase m has been shown to terminate transcription both in vitro and in vivo at stretches of four or more T residues in the non-coding DNA strand (Bogenhagen and Brown, 1981; Cozzarelli et al., 1983; Adeniyi­ Jones et al., 1984; Mazabraud et al., 1987). The stretch of T residues seems to be the main determinant of the strength of the termination site, however, adjacent sequences have also been shown to influence the efficiency of termination (Bogenhagen and Brown, 1981; Allison and Hall, 1985; Mazabraud et al., 1987). Purified RNA polymerase m can distinguish between weak and strong terminators and terminate transcription at stretches of T residues with similar efficiencies as observed with crude nuclear extracts (Cozzarelli et al., 1983; Watson et al., 1984; Campbell and Setzer, 1992). These results suggest that RNA polymerase m alone is sufficient for termination, and acœssory factors are not required. Recent results however suggest that the cellular protein La may be involved in termination by RNA polymerase m (Gottlieb and Steitz, 1989a,b). La is a 50 kDa protein which binds to the 3' uridylate tail of RNAs synthesized by RNA polymerase m (Stefano, 1984; Mathews and Francoeur, 1984). It has been demonstrated that in the absence • of La, RNA polymerase stalls immediately upstream of the stretch of T 23

residues (Gottlieb and Steitz, 1989b). La protein may therefore facilitate • completion of the nascent transcript and promote release of the transcriptional complex from the DNA. Although purified RNA polymerase m can terminate transcription in the absence of La, when associated with several auxiliary factors in vivo, La may be required for efficient termination (Gottlieb and Steitz, 1989b; Campbell and Setzer, 1992). Transcription by RNA polymerase l terminates upstream of an 18 bp sequence element AGGTCGACCAGA/TI/ ANTCCG, termed the Sai-box (Grummt et al., 1985). Eight copies of this sequence element are found downstream of the 28S coding region. Mutations or deletions within the Sal­ box affect transcription termination by RNA polymerase l (Grummt et al., 1986). AlOS kDa protein, called TIF1, has been purified and shown to bind to the Sai-box sequence elements (Bartsch et al., 1988). TIF1 binding to the Sal­ box elements is required for transcription termination (Grummt et al., 1986; Bartsch et al., 1988). The mechanism by which TIF1 terminates transcription is not known. It has been demonstrated that DNA binding by TIF1 is not sufficient for termination suggesting that, in addition to a DNA binding domain, TIF1 contains a second domain which can interact with RNA polymerase l (Bartsch et al., 1988). Termination by TIF1 is specific for RNA polymerase l since transcription by eukaryotic polymerases il and m as weil as E. coli or bacteriophage 1'3 polymerases are not affected by a TIF1-Sal-box complex (Kuhn et al., 1990). This suggests that TIF1 may interact with a subunit which is specific to RNA polymerase 1.

B. Termination by RNA polymerase Il. The analysis of RNA polymerase il termination sites has proven to be • difficult since the 3' ends of ail mature mRNAs are generated by rapid RNA 24

processing rather than termination of transcription (reviewed in Wahle and

Keller, 1992). In prokaryotes, transcription termination has been studied using • in vitro transcription systems, however, no efficient in vitro system is available in eukaryotes to study termination by RNA polymerase II. Termination by RNA polymerase II has been analyzed by hybridization analysis of pulse-labeled RNAs (Ford and Hue, 1978; Nevins and Darnell, 1978), or by nudear run-on transcription analysis (McKnight and Palmiter, 1979; Groudine et al., 1981). Using these techniques, termination regions for many cellular genes have been mapped (Proudfoot and Whitelaw, 1988; Proudfoot, 1989). In aIl cases, termination of transcription occurs at sites located several hundreds to several thousands of nucleotides downstream of the 3' end processing signal. For sorne genes, termination occurs over a region of 1000-2000 bps, suggesting the presence of multiple weak termination sites, whereas for other genes, termination occurs at a single site within a region of 100-200 bps. Analysis of several termination regions has not revealed the existence of similar sequences or structures which could function as a common terminator element. In prokaryotes, pausing of RNA polymerase is important for termination at both intrinsic and rho-dependent terminators. In eukaryotes, accumulation of paused RNA polymerases at the site of termination has been observed for several genes induding the immunoglobulin kappa gene (Xu et

al., 1986), the adenosine deaminase gene (Maa et al., 1990), and the rabbit lX

and ~ globin genes (Vandenbergh et al., 1991). Studies on the human lX 2 globin gene has demonstrated that a 92 bp sequence element located in the termination region functions as a pause site and RNA polymerase terminates transcription with an efficiency of 80% at this site in the presence of an intact • upstream 3' end processing signal (Enriquez-Harris et al., 1991). Therefore, as 25

in prokaryotes, pausing of RNA polymerase may also be important for • termination in eukaryotes. Accumulating evidence suggests that 3' end processing of the primary transcript may also be necessary for efficient termination of transcription (Proudfoot, 1989). Thus, efficient termination of transcription requires both intact processing signais as weil as signais located further downstream (Maniey et al., 1989; Proudfoot, 1989) The cis-acting sequences and trans-acting factors involved in accurate 3' end formation have been weil characterized (reviewed in Whale and Keller, 1992). Two sequence elements are required for efficient cleavage and polyadenylation of the nascent RNA: the conserved hexanucleotide sequence AAUAAA located 20-30 nucleotides from the site of cleavage (proudfoot and Browniee, 1976), and the less conserved GU- or U­ rich element located from 10 to 50 nucleotides downstream of the cleavage site (Gil and Proudfoot, 1987). The AAUAAA sequence element has been

shown to be important for termination of transcription in the human alpha-2 globin gene (Whitelaw and Proudfoot, 1986), and the mouse l3-globin gene (Falck-Pederson et al., 1985; Logan et al., 1987). RNA polymerase transcribed unimpeded through the normal termination sites of these genes when mutations were introduced in the AAUAAA sequence element suggesting a link between termination and 3' end formation. A Hnk between 3' end formation and efficient termination has been shown for the late transcription unit of polyomavirus. Termination of transcription within this unit is inefficient and results in the formation of large multigenome-Iength RNAs (Acheson, 1984). However, the replacement of the weak polyomavirus poly(A) signal with the strong rabbit (3-globin poly(A) signal resulted in more • efficient termination (Lanoix and Acheson, 1988) 26

A link between 3' end formation and termination makes sorne physiological sense, since it would ensure that RNA polymerases terminate • transcription downstream of 3' end processing signais. Two models have been proposed to explain the link between 3' end formation and termination. One model proposes that upon transcribing the 3' end processing signais, a factor required for efficient elongation is released from the RNA polymerase (Logan, 1987). RNA polymerases which lack this elongation factor terminate transcription at sites located further downstream, whereas in the presence of this factor, RNA polymerases transcribe through these sites unimpeded. A second model proposes that upon cleavage of the nascent RNA, a rho-like termination factor recognizes the uncapped 5' end of the resulting RNA. This factor could then translocate up to the polymerase paused at a downstream site and promote termination (Proudfoot and Whitelaw, 1988; Proudfoot, 1989). Such a factor has not yet been identified.

C. Termination signals recognized by RNA polymerase II in "Ditro. Using a promoter-independent in vitro transcription system, sites recognized by purified RNA polymerase in the absence of accessory factors have been identified. Several of these intrinsic termination sites are found in regions that block transcription elongation in "Di"DO (Bentley and Groudine, 1988; Kerpolla and Kane, 1988). Generaily, termination occurs at stretches of T residues (Dendrick et al., 1987; Reines et al., 1987; Kerpolla and Kane, 1988). However, not ail stretches of T residues function as intrinsic terminator sites nor does the efficiency of termination correlate with the number of T residues (Dendrick et al., 1987; Reines et aL, 1987; Kerpolla and Kane; 1988). Also, RNA secondary structure is not involved in terminating • transcription at these site suggesting that other factors may be responsible. 27

Recent experiments suggest that bent DNA is an important structural • element in the signal for transcription termination by purified RNA polymerase II. Deletion analysis of the 3.3 gene has localized the minimal termination signal to the sequence TTTTTTTCCCTTTTTT (Kerppola and Kane, 1990). A similar sequence has also been identified at the m terrnination site within the human c-mye gene. Electrophoretic mobility shift assays have demonstrated that both sequence elements represent regions of bent DNA. Furthermore, SV40 sequences which had previously been shown to bend DNA also caused RNA polymerase II to terminate transcription. Taken together, these results suggest that bent DNA may be part of the signal for transcription termination.

D. RNA polymerase II elongation factors. Several cellular factors have been identified which affect transcription elongation by RNA polymerase II. The best studied are the Sil class of elongation factors (Hirashima et al., 1988) also known as TFIIS (Reinberg and Roeder, 1987) and RAP38 ($opta et al., 1985). These factors have been purified from several mammalian sources (Rappaport et al., 1987; Reinberg and Roeder, 1987; Hirashima et al., 1988), from Drosophila (Priee et al., 1987; Sluder et al., 1989), and from yeast (Sawadogo et al., 1980), and cDNAs have been isolated from mouse (Hirashima et al., 1988), human (Yoo et al., 1991), Drosophila (Marshall et aL, 1990) and from yeast (Clark et aL, 1991). Elongation factor Sil is a 35-38 kDa polypeptide which binds to RNA polymerase n (Horikoshi et al., 1984; Reinberg and Roeder, 1987). sn does not enhance the general rate of elongation (Bengal et aL, 1991; Wiest et al., 1992) but has been shown to enable RNA polymerase Il to readthrough strong • transcriptional pause sites found in the adenovirus late transcription unit 28

(Rappaport et al., 1987; Reinberg and Roeder, 1987), and in the human H3.3 gene (Reines et al., 1989; SivaRaman et al., 1990). SIl has also been shown to • relieve the obstruction to transcription elongation caused by a lac repressor­ operator complex (Reines and Mote, 1993). It has reCl'ntly been demonstrated that SIl stimulates a 3' to 5' cleavage activity associated with RNA polymerase II (Reines, 1992; Reines et al., 1992; Izaban and Luse, 1992a; Wang and Hawley, ~~93). This cleavage activity occurs in arrested ternary complexes resulting in a shortening of the nascent RNA chain which remains held by the transcriptional complex, and elongation can resume when supplied with nucleoside triphosphates. It has recently been demonstrated that sn stimulates a 3' to 5' exonuclease activity rather than an endonucleolytic or a pyrophosphorolYFtic activity (Wang and Hawley, 1993) Cleavage of the nascent RNA is required for the arr~sted transcriptional complexes to resume elongation. It has been proposed that cleavage of the nascent RNA enables the RNA polymerase to move backwards from a site of block and make repeated attempts at elongating through this site (Reines, 1992; Wang and Hawley, 1993). Alternatively, cleavage of the 3' end of the nascent RNA may release the transcriptional complex from an arrested conformation. A second elongation factor which binds to RNA polymerase II is TFIIF (Flores et al., 1989), also known as RAP30/74 (Burton et al., 1988). In contrast to SIl, TFIIF is required for transcription initiation (Flores et al., 1989; Price et al., 1989). TFIIF prevents RNA polymerase from binding to nonspecific sites on DNA (Conaway and Conaway.. 1990), and as a result, inhibits non-specific transcription by RNA polymerase II (Killen and Greenblatt, 1992). In contrast to SIl, TFIIF has been shown to stimulate the general rate at which RNA • polymerélSt~ II elongates (Bengal et al., 1991). TFIIF has also been shown, like 29

sn, to suppress pausing at certain sites in ,litra (Bengal et al., 1991; Izban and • Luse, 1992b). However, TFnF must be present before the transcriptional complex has stalled, since TFIIF has a limited ability to release the transcriptional complex from the !-'ause site if added after the complex has paused (Bengal et aL, 1991). It has recently been demonstrated that TFIIF also stimulates the shortening of nascent RNA transcripts in stalled RNA polymerase II complexes (Wang and Hawley, 1993). TFTIF, in the presence of pyrophosphate (PPi) stimulates a pyrophosphorolytic reaction, which results in the shortening of the nascent transcript (Wang and Hawley, 1993). Pyrophosphorolysis is the reverse of the polymerization reaction, and results in the generation of nuc1eotide triphosphates. The ability of TFIIF to stimulate a pyrophosphorolysis reaction in stalled transcriptional complexes is most probably important for its effect on transcription elongation. A third elongation factor that has been isolated is TFIIX. It was originally identified as a factor which stimulated transcription through the promoter­ proximal sequences of the adenovirus major late promoter (Reinberg et aL, 1987). lie TFTIF, TFIIX has been shown to stimulate the general elongation rate by RNA polymerase II and to release stalled transcriptional complexes (Bengal et al., 1991; Izban and Luse, 1992b).

E. Regulation of Transcription Elongation in Eukaryotes.

Most of our understanding of the mechanisms involved in the regulation of transcription elongation cornes from studies done in prokaryotes. It has been demonstrated that elongation can be regulated either • by modifying the transcriptional complex to a terminating or anti-terminating 30

form, or by modifying termination signais through the interaction of or trans-acting factors. Alternatively, DNA-binding proteins can • sterically interfere with polymerase progress by acting as a roadblock to transcription elongation. ln eukaryotes, increasing interest in the study of transcription elongation is due in part to the realization that many genes are regulated at this level. A growing number of genes have recently been identified which contain a conditional block to transcription elongation. In most of the genes identified, the extent of transcriptional block can be modulated depending on the cell type or the physiological state of the cells, demonstrating that control of transcription elongation contributes to the regulation of gene expression. Only now are we beginning to understand some of the mechanisms involved in the regulation of transcription elongation in eukaryotes.

1. Antitermination in HIV. The product of the human immunodeficiency virus (HIV) tat gene is required for efficient synthesis of the viral gene products. The tat protein induces a dramatic increase in the levels of viral mRNAs (Cullen, 1986; Peterlin et al., 1986; Musing et al., 1987). Nuclear run-on assays have demonstrated that tat protein increases the number of transcriptional complexes that reach the end of the transcription unit, whereas it has no effeet on transcription initiation. Initiation rates are high even in the absence of tat (Kao et al., 1987; Feinberg et al., 1991), however, using an adenovirus system, one report has suggested that tat may also increase transcription initiation (Laspia et al., 1989). The tat protein has been shown to increase the processivity of • transcriptional complexes having initiated transcription at the 5' LTR 31

promoter of ~::::::.; both in vivo and in vitro. (Kao et al., 1987; Feinberg et al., 1991; harciniak and Sharp, 1991; Kato et al., 1992). Nuclear run-on assays have • demonstrated that in the absence of tat, poorly processive transcriptional complexes are assembled which prematurely terminate within several hundred nucleotides from the transcription start site. In the presence of tat, highly processive complexes are assembled which efficiently transcribe the viral genes resulting in an increase in viral mRNAs (Kao et al., 1987; Feinberg et al., 1991; Marciniak and Sharp, 1991; Kato et al., 1992). Therefore, tat protein stimulates transcription by increasing the efficiency of elongation of the transcriptional complexes. Tat function is dependent on an RNA sequence element calIed TAR, located between nucleotides +18 to +44 relative to the transcription initiation site (Rosen et al., 1985; Jakobovits et al., 1988; Hauber and Cullen, 1988). The TAR element can form a hairpin structure containing a three nucleotide bulge in the stem as weIl as a six nucleotide locp at the top, both of which are required for tat binding (Dingwall et al., 1990; Roy et al., 1990; reviewed in Cullen, 1990). Mutations in the stem of the hairpin which affect base-pairing also affect transactivation by tat, however compensatory mutations that ::estores base-pairing aIse restore tat activity, demonstrating that the structure of the stem is more important than the sequence. It has been demonstrated that fusion proteins between tat and the RNA binding domains of other proteins can function as transactivators when the TAR element is replaced by sequences recognized by the fusion protein, suggesting that the only function of TAR is to serve as a recognition site for tat (Selby and Peterlin, 1990; Southgate et al., 1990). Aise supporting this view is the demonstration that tat • can function not only when bound to the TAR element located in the nascent 32

RNA, but aIse when targeted to the promoter by fusion to the GAL4 DNA binding domain (Southgate and Green, 1991). • Several proteins have been identified which affect transactivation by tat. Mutational analysis as well as functional assays have demonstrated that several cellular factors interact with the hairpin structure and affect transactivation by tat (Marciniac et al., 1990; Sheline et al., 1991). These proteins may stabilize the interaction between the tat protein and TAR. Two cellular proteins have been identified which bind to tat and affect transactivation. Screening of an expression library with labeled tat protein has identified a protein called TBP-1 which binds to tat (Nelbock et al., 1990). Overexpression of TBP-1 repressed transactivation by tat. Recently a protein sharing homology to TBP-1 called MSS1 has been isolated which binds to tat and greatly stimulates expression of a reporter construct driven by the HIV-1 LTR (Shibuya et al., 1992). These two proteins may modulate tat activity in the cell. Although the mechanism of tat function is not known, several models have been proposed. Because of the similarities between antitermination by tat and the lambda N protein, a model has been proposed in which tat, like N, assembles a ribonucleoprotein complex containing host factors and TAR RNA. In the absence of tat, transcriptional complexes which initiate transcription from the LTR may not be able to interact with elongation factor resulting in non-processive complexes which prematurely terminate transcription. As a result, few complexes transcribe to the end of the transcription unit and low levels of fu1llength viral mRNAs are synthesized. In the presence of tat, transcriptional complexes will transcribe the TAR element to which tat as weil as other host factors can bind. By RNA looping, • tat can interact with the transcriptional complex and either permit or stabilize 33

the interaction of elongation factors with the transcriptional complex. • Interestingly, the elongation factor TFIIF has been shown to be important for antitermination by tat, and high levels of TFIIF improve the processivity of the transcriptional complexes in the absence of tat (Kato et al., 1992). Therefore, tat may stabilize the interaction between TFIIF and RNA polymerase. It is also possible that tat may itself function as an elongation factor which can enhance the processivity of transcriptional complexes in a manner analogous to TFIIF. Recently, an alternative model has been proposed which suggests that tat activates transcription initiation rather than transcription elongation (Cullen, 1993). This model proposes that the LTR contains two overlapping promoters that direct the initiation at the same start site. One promoter has a high basal activity in the absence of tat but generates non-processive complexes which prematurely terminate transcription. The second promoter has a low basal activity but can generate highly processive transcriptional complexes. The less processive promoter would ensure the synthesis of the TAR element, so that in the presence of tat, the tat-TAR complex would activate the more processive promoter and down-regulate the less processive promoter, perhaps by promoter occlusion. This model proposes that tat has no direct effect on the processivity of the transcriptional complex but rather stimulates the initiation of transcription from a promoter which assembles highly processive complexes. Recent results by Sheldon et al. (1993) have given support to this hypothesis. They have demonstrated the presence a non-processive promoter which they have called, of short transcript (IST), located between -5 and +26 in the HIV-1 LTR. Mutations which significantly reduce expression from this promoter do not affect transactivation by tat supporting • the notion that tat may activate transcription initiation from a distinct 34

promoter generating processive complexes. Further experiment will be • required to fully understand the mechanism of transactivation by tat. 2. Transcriptional roadblock. In eukaryotes, two cases have been reported, where a DNA binding protein acts as a roadblock to transcription elongation. Connelly and Manly (1989) have studied transcription termination using a chimeric plasmid containing the adenovirus major late promoter (MLP) fused to the early region of SV40. They have demonstrated that transcription initiating from the MLP transcribes around the plasmid and terminates within a 75 bp region upstream of the MLP. Within the termination region is CCAAT box which binds a family of transcription factors. A 13 bp region containing this site was shown to be sufficient to block elongation in an orientation dependent manner, as determined by nuclear run-on analysis. Furthermore, mutations in the CCAAT box which were known to disrupt protein binding also abolished termination, suggesting that a protein binding to this site blocks transcription elongation. It has been suggested that polyomavirus large T antigen also blocks transcription elongation. Polyomavirus is a circular double-stranded DNA tumor virus. Nuclear run-on analysis has demonstrated that there is a high density of RNA polymerases transcribing the late strand in the region of the origin (Skarnes et al., 1988). The accumulation of stalled RNA polymerases was mapped on the late strand of polyomavirus to a region containing large T antigen binding sites (Bertin et al., 1992). The accumulation of stalled RNA polymerases suggested that there was a block to the elongation of transcription which may be due to the binding of large T antigen. Indeed, • functional large T antigen has been shown to be required for the 35

accumulation of stalled polymerases, strongly suggesting that the binding of • large T antigen bound to DNA acts as a transcriptional roadblock to elongating RNA polymerase complexes (Bertin et al., 1992).

3. Block to transcription elongation. A growing number of cellular genes and viral transcription units have been shown to contain a block to elongation of transcription. These include several proto-oncogenes including c-mye (Bentley and Groudine, 1986; Eick and Bornkamm, 1986; Nepveu and Marcu, 1986), l-mye (Krystal et al., 1988), N-mye (Morrow et al., 1992), c- (Bender et al., 1987; Watson, 1988), and c­ fos (Fort et al., 1987; Schneider-Schaulies et al., 1987), as weIl as genes encoding B-globin (Salditt-Georgieff et al., 1984), adenosine deaminase (Chinsky et al., 1989; Chen et al., 1990; Maa et al., 1990; Ramamurthy et al., 1990),the EGF­ receptor (Haley and Waterfield, 1991), the ribonucleotide reduetase R2 subunit (Bjorklund et al., 1992), the macrophage colony stimulating factor receptor (Yue et al., 1993), the Drosophila hsp70, hsp26, 131-tubulin, glyceraldehyde-3­ phosphate dehydrogenase, polyubiquitin, and copia (Rougevie and Lis, 1988; Rougevie et al., 1990). AIso the adenovirus major late promoter (Maderius and Chen-Kiang, 1984; Hawley and Roeder, 1985 ), the viral transcription unit of HIV-1 (Kao et al., 1987 ), HIV-2 (Toohey and Jones, 1989), the Minute virus of mouse (Reznikov et al., 1989), the SV40 late transcription unit (Hay et al., 1982) and the polyomavirus late transcription unit (Skarnes et al., 1988). More importantly, the extent of transcriptional block in many of these genes have been shown to be governed by metabolic, tissue-specific, and developmental signals (reviewed in Spencer and Groudine, 1990; Kerppola and Kane, 1991). Regulation of the extent of transcriptional block as a means of • modulating gene expression was first demonstrated in the human c-mye gene 36

(Bentley and Groudine, 1986; Eick and Bornkamm, 1986). Human HL60 cells can be induced to differentiate into granulocytes when treated with retinoic • acid. UpOI. differentiation, there is a 10 fold reduction in steady-state c-myc RNA levels. Nudear run-on assays demonstrated that the decrease in steady­ state c-myc RNA levels could be accounted for by an increase in the levels of transcriptional block (Bentley and Groudine, 1986; Eick and Bornkamm, 1986). Therefore, block to the elongation of transcription can control c-myc RNA levels during differentiation. In addition, a decrease in the levels of transcriptional block resulting in an increase in steady-state c-myc transcripts has been demonstrated in tonsillor mononudear cells stimulated with mitogens (Eick et al., 1987) as well as in normal peripheral blood T lymphocytes stimulated by PMA (Lindsten et al., 1988). Regulation of transcriptional block has also been described for the murine c-myc gene in Abelson murine leukemia virus transformed fibroblasts in which the gene was amplified (Nepyeu and Marcu, 1986). Mouse erythrolukemia (MEL) cells, like HL60 cells, can be chemically induced to differentiate. Increase in transcriptional block has aIso been demonstrated to account for the decrease in steady-state c-myc RNA levels in differentiating MEL cells (Mechti et al., 1986; Nepveu et al., 1987; Watson, 1988). Furthermore, some growth factors and mitogens decrease the extent of block as demonstrated in mouse spleen lymphocytes and antigen specifie T cells tre~ted with concanavalin A (Schneider-Schaulies et al., 1987) and EGF-treated mouse fibroblasts (Nepveu et ai., 1987). Other proto-oncogenes have been shown to be regulated at the level of transcription elongation during cell differentiation or upon mitogen stimulation. Fluctuations in the levels of steady-state c-myb transcripts in • Friend erythrolukemia cells induced to differentiate are controlled in part by block to transcription elongation (Watson et al, 1988). Also, the extent of block • to tr:-~scription elongation in the c-fos gene decreases when quiescent hamster fibroblasts are serum stimulated, or after treatment of mouse spleen lymphocytes with concanavalin A (Schneider-Schaulies et al., 198ï). Finally, transcription elongation has been shown to r"gulate cell type specifie expression of the adenosine deaminase gene (Chinsky et al., 1989; Maa et al., 1990) and the cell cycle regulated expression of the ribonucleotide reductase R2 gene (Bjorklund et al., 1992). Several approaches have been used to identify the sites at which

transcription is blocked in many genes. These include in VillO UV cross­ linking, nuclear run-on transcription assays, injections in Xenopus oocytes and in vitro transcription assays. The site of block in relation to the transcription initiation site appears to vary in different genes. Sites of block have been identified as close as 10-50 bps downstream of the initiation sites, as in the Drosophila hsp70 gene and the human c-myc gene (Rougvie and Lis, 1988; Strobl and Eick, 1992; Krumm et al., 1992), and as far as several thousand bps the c-myb gene (Bender et al., 1987; Watson, 1988). Many of the sites of block identified by injections in Xenopus oocytes and in vitro transcription assays have been shown to function as intrinsic termination elements (Bentley and Groudine, 1988; Kerpolla and Kane, 1988). It is not known if these sites al50 function as intrinsic terminators in the ceIl, since short prematurely terminated transcripts have not been detected. The mechanism regulating block to the elongation of transcription at these sites is not known. However, there is increasing evidence from experiments performed in Xenopus oocytes and by in vitro transcription, that transcriptional block can be modulated by modifications of the RNA • polymerase II complex. The addition of elongation factors such as TFIIS, 38

TFIIF, and TFIIX to in vitro transcription assays has been shown to allow RNA polymerase II to readthrough specifie sites of block (Rappaport et al., • 1987; Reinberg and Roeder, 1987; Reines et al., 1989; SivaRaman et al., 1990; Bengal et al., 1991; Izban and Luse, 1992b). Furthermore, recent titration experiments have demonstrated that the extent cf premature termination in Xenopus oocytes is governed by the overall levels of transcription initiation (Meulia et al., 1993; Spencer and Kilvert, 1993). Using the c-mye gene, it has been demonstrated that when initiation levels are low, most of the complexes transcribe through the site of block, however, when initiation levels are high, there is an increase in the number of complexes which prematurely terminate transcription. These observations suggest that in Xenopus oocytes, there is a limiting amount of a processivity factor which is depleted at high levels of transcription initiation. This would result in the generation of a class of transcription complexes with a reduced capacity for elongation. Similar

conclusions have recently been reported from studies on the Xenopus (1­ tubulin gene (Hair and Morgan, 1993). Thus, the ability to read-through sites of block may be regulated in Xenopus oocytes and in in vitro transcription assays, by the interaction of elongation or processivity factors with the transcriptional complex. Whether this also occurs in mammalian cells

remains to be determined. Although modification of the RNA polymerase complex may be responsible for the modulation of elongation at some sites of block, other mechanisms may also be involved. For example, block within the c-myb gene may be regulated by cellular factors interacting with sequences near the site of block. It has been demonstrated that severa! cellular factors bind in vitro to sequences near the site of block (Reddy and Reddy, 1989). Celllines expressing • high levels of c-myb mRNA and high readthrough transcription, contain 39

more of a DNA binding factor than cells expressing low levels of c-my/J • mRNA and low readthrough transcription (Reddy and Reddy, 1989). These results suggest that in the c-myb gene, transcriptional block may be regulated by the binding of cellular factors which modify the site of block rath':!r than the transcriptional complex. Therefore, several different mechanisms may be involved in the regulation of transcriptional block. Promoter elements have also been implicated in the control of transcription elongation. One of the best examples is the regulation of block to

transcription elongation in the Drosophila hsp70 gene. UV cross-linking of

protein-DNA complexes in vivo has demonstrated that RNA polymerase II is associated with the promoter-proximal region prior to heat induction (Gilmour and Lis, 1986). Upon heat shock, RNA polymerases are present throughout the gene. In non-heat shocked cells, RNA polymerase has been shown:o pause approximately 25 bps downstream of the initiation site. Upon heat shock, the paused transcriptional complex are released and can transcribe the gene (Rougvie and Lis, 1988). Deletion and mutational analysis have demonstrated that the GAGA element located in the promoter is required for generating a paused polymerase (Lee et al., 1992). The GAGA element has been shown to bind a cellular factor (Biggin and Tjian, 1988). Mutations in the GAGA element which affect protein binding also affect promoter-proximal pausing. Therefore, it has been proposed that the binding of the GAGA factor

to the hsp70 promoter leads to the assembly of an uninduced transcriptional complex which will pause downstream of the initiation site. Upon induction, the (HSF) allows the release of the transcriptional complex from the pause site. Regulation of transcription elongation by promoter­ • proximal pausing of RNA polymerase may be a general phenomenon since 40

several other Drosophila genes have been identified which have polymerases paused downstream of the initiation site (Rougevie and Lis, 1990). • Promoter sequences have also been shown to affect transcription termination in the snRNA genes VI and V2 (reviewed in Spencer and Groudine, 1990; KerpoUa and Kane, 1991). Transcription by RNA polymerase II initiates from the promoters of VI and V2, and terminates at an element termed the 3' box. Interestingly, if transcription is initiated from a heterologous promoter such as the ~-globin or the adenovirus major late promoter, the transcriptional complexes no longer recognize the 3' box and transcribe through the termination signais (Hernandez and Weiner, 1986; Neuman de Vegar et al., 1986). However, the VI and V2 promoters can substitute for one another. The elements which are required for recognition of the 3' box are indistinguishilble from the promoters themselves (Hemandez and Lucito, 1988; Parry et al., 1989). Therefore, elements in the VI and V2 proro.oters direct the assembly of a transcriptional complex, at the initiation site, which can recognize the 3' processing signais. The presence of these elements may allow RNA processing or termination factors to interact with the transcriptional complex, thereby permitting recognition of the 3' box.

4. Regulation of transcription elongation in the c-myc gene. e-mye was one of the first genes in which block to elongation of transcription was detected. Three approaches have been used to study transcriptional black in the c-mye gene. These are transfections into mammalian cells which have identified promoter elements involved in transcriptional block, and injections into Xenopus oocytes and in vitro transcription assays which have identified sequences within exon 1 which • could function as intrinsic terminators. 41

Using injection in Xenopus oocytes and in "itro transcription the site at • which transcription is block in the human and murine c-mye genes have been identified. In the human c-mye gene, the site of block has been mapped to T-rich sequences located near to the exon 1-intron 1 border (Bentley and Groudine, 1988; Kerppola and Kane, 1988). In the murine gene, the site of block has been mapped to two stretches of T residues (TS and T3) located within the 3' half of exon 1 ( Bentley and Groudine, 1988; Miller et al., 1989). When assayed in Xenopus oocytes, sequences from the 3' hait' of exon 1 including the sites of block of the human c-mye gene have been shown to cause premature termination when inserted downstream of heterologous promoters (Bentley and Groudine, 1988). However, it has recently been demonstrated that these sequences may not be required for block to the elongation of transcription in viv

initiated transcripts are blocked (Sp~ncer et al., 1990). In Burkitt's lymphoma cells, a correlation has been established between increaseè. Pl initiation and • decreased block to elongation of transcription suggesting that P2 initiated transcripts are preferentially blocked over Pl initiated transcripts (Spencer et al., 1990). Furthermore, a recent study has demonstrated that in Xenopus oocytes sequences downstream of the Pl initiation site allow efficient elongation of P1-initiated transcripts through the site of block at the 3' end of the exon 1 (Meulia et al., 1992).

c-myc ---.:~I' 1; 1 2 1 1 3 1- : ~-~------i ~A~2 ~ '+~631 Exon 1 Liri +55~ ME1a2 ME1a1 Pause sile Sites of in vitro termination

Figure 4. Map of the murine c-mye gene. The c-myc gene contains three exons and two transcription initiation sites termed Pl and P2. Exon 1 has been enlarged to illustrate important sites involved in transcription elongation. Protein binding sites located upstream of the P2 initiation site are illustrated. The site at which the transcriptional complex has been shown to pause in the human c-mye gene is illustrated (Strobl and Eck, 1992; Krumm et al., 1992). The stretch of 3 and 5 T residues where premature termination has been shown to occur invitro areidentified (Bentley and Groudine, 1988; Miller et al., 1989).. • Deletion analysis of the murine c-myc promoter has demonstrated that a • 16 bp deletion upstream of the TATA box of P2 significantly reduced transcriptional block as detected by nuclear run-on assays of transfected HeLa cells (Miller et al., 1989). This deletion partially removes a previously identified protein binding site called ME1a1 (Fig. 4) (Asselin et al., 1989). These results suggest that the ME1a1 protein binding site located within the P2 promoter may be important in the regulation of transcription elongation of the c-my' gene. The experiments presented in this thesis are aimed at determining whether the ME1a1 site is involved in the regulation of transcription elongation, and aise to determine if cellular factors binding to this site are important for transcriptional block.

• 44 • References Acheson, N.H. 1984. Kinetics and efficiency of polyadenylation of late polyomavirus nuclear RNA: generation of oligomeric polyadenylated RNAs and their processing into mRNA. Mol. Cell. Biol. 4:722-729.

Adeniyi-]ones, S., P.H. Romeo, and M. Zasloff. 1984. Generation of long­ readthrough transcripts in vivo and in vitro by deletion of 3' termination and processing sequences in the tRNAmet gene. Nucleic Acids Res. 12:1101-1115.

Albrechtsen, B., C.L. Squires, S. Li, and C. Squires. 1990. Antitermination of characterized transcriptional terminators by the Escherichia coli rrnG leader region. ]. Mol. Biol. 213:123-134.

Alfino, P., F. Rivellini, O. Limauro, C.B. Bruni, and M.S. CarIomagno. 1991. A consensus motif common to aIl rho-dependent prokarytic transcription } terminators. Cell 64:553-563.

Allison, O.S., and B.D. Hall. 1985. Effects of alterations in the 3' flanking sequence on in vivo and in vitro expression of the yeast 5UP4-o tRNATyr gene. EMBO]. 4:2657-2664.

Allison, LA., M. Moyle, M. Shales, and C.]. Ingles. 1985. Extensive homology among the largest subunit of eukaryotic and prokaryotic RNA polymerases. Cell 42: 599-610. • 45

Amster-Choder, O., and A. Wright. 1990. Regulation of activity of il • transcriptional antiterminator in E. coli by phosphorylation in vivo. Science 249:540-542.

Arndt, K.M., and M.J Chamberlin. 1990. RNA chain elongation by EscllcriclJÙI coli RNA polymerase: factors affecting the stability of elongating ternary complexes. J. Mol. Biol. 213:79-108.

Asselin, C, A. Nepveu, and K.B. Marcu. 1989. Molecular requirements for transcriptional initiation of the murine c-myc gene. Oncogene. 4:549-558.

Atkinson, B.L., and M.E. Gottesman. 1992. The Escherichia coli rpoB60 mutation blocks antitermination by coliphage HK022 Q-function. J. Mol. Biol. 227: 29-37.

Barik, S., B. Ghosh, W. Whalen, D. Lazinski, and A. Das. 1987. An antitermination protein engages the elongating transcription apparatus at a promoter-proximal recognition site. Cel! 50:885-899.

Bartsch, 1, C. Schonenberg, and 1 Grummt. 1988. Purification and characterization of TIF1, a factor that mediates termination of mouse ribosomal DNA transcription. Mol. Cel!. Biol. 8:3891-3897.

Bauer, C.E., J. Carey, L.M. Kasper, S.P. Lynn, and J.F. Gardner. 1983. Attenuation in bacterial operons. In gene function in prokaryotes. ed. J. Beckwith et al. pp65-89. Cold Spring Harbor Laboratory, Cold Spring Harbor, • New York. 46

• Bender, T.P., C.B. Thompson, and W.M. Kuehl. 1987. Differentiai expression of c-myb mRNA in murine B lymphomas by a block to transcription elongation. Science 237:1473-1476.

Bengal, E., O. Flores, A. Krauskopf, D. Reinberg, and Y. Aloni. 1991. Role of mammalian transcription factors IIF, lIS, and IIX during elongation by RNA polymerase II. Mol. Cell. Biol. 11:1195-1206.

Bentley, D.L., and M. Groudine. 1986. A block to elongation is largely responsible for decreased transcription of c-mye in differentiated HL60 cells. Nature 321:702-706.

Bentley, D.L., and M. Groudine. 1988. Sequence requirements for premature termination of transcription in the human c-mye gene. Cdl 53:245-256.

Berg, K.L., C. Squires, and C.L. Squires. 1989. Ribosomal RNA operon antitermination: function of leader and spacer region box B-box A sequences and their conservation in diverse microorganisms. J. Mol. Biol. 209:345-358

Bertin, J., N.-A. Sunstrom, P. Jain, and N.H. Acheson. 1992. Stalling by RNA polymerase II in the polyomavirus intergenic region is dependent on functional large T antigen. Virology 189:715-724.

Biggin, M.D., and R. Tjian. 1988. Transcription factors that activate the • Ultrabithorax promoter in developmentally staged extracts. Cell53:699-711. -l7

Biggs, J., L.L. Searles, and A.L. Greenleaf. 1985. Structure of the eukaryotic • transcription apparatus: Features of the gene for the largest subunit of Drosophila RNA polymerase II. Cell 42: 611-621.

Bjorklund, S., E. Skogman, and L. Thelander. 1992. An S-phase specifie release from a transcriptional block regulates the expression of mouse ribonucleotide reductase R2 subunit. EMBO J. 11:4953-4959.

Bogenhagen, O.F., and 0.0. Brown. 1981. Nucleotide sequence in Xenopus 55 DNA required for transcription termination. Cell 24:261-270.

Borukhov, S., A. Polyakov, V. Nikiforov, and A. Goldfarb. 1992. GreA protein: A transcription elongation factor from Eschericllia coli. Proc. Nat!. Acad. Sei. USA 89:8899-8902.

Borukhov, S., V. Sagitov, and A. Goldfarb. 1993. Transcript cleavage factors from E. coli. CeU 72:459-466.

Brennan, c.A., A.J. Oombroski, and T. Platt. 1987. Transcription termination factor rho is an RNA-ONA helicase. CeU 48:945-952.

Briat, J.F., and M.J. Chamberlin. 1984. Identification and characterization of a new transcriptional termination factors from Escherichia coli. Proc. Nat!. Acad. Sei. USA 81: 7373-7377.

Briat, J.F., G. Bollag, C.A. Keamey, I. Molineux, and M.J. Chamberlin. 1987. • Tau factor from Escherichia coli mediated accurate and efficient termination 48

of transcription at the bacteriophage T3 early termination site in vitro. J.Mol. • Biol. 198: 43-49. Burgess, R.R., A.A. Travers, J.J. Dunn, and E.K.F. Bautz. 1969. Factor stimulating transcription by RNA polymerase. Nature 221: 43-46.

Burton Z.F., M. Killeen, M. Sopta, L.G. Ortolan, and J. Greenblatt. 1988. RAP30/74: A general initiation factor that binds to RNA polymerase ll. Mol. Cell. Biol. 8:1602-1613.

Campbell, F.E., and D.R. Setzer. 1992. Transcription termination by RNA polymerase III: Uncoupling of polymerase release from termination signal release. Mol. Cell. Biol. 12:2260-2272.

Chan, c.L., and R. Landick. 1989. The Salmonella typltimurium ltis operon leader region contains an RNA hairpin-dependant transcription pause site. J. Biol. Chem. 264: 20796-20804.

Chen, Z., M.L. Harless, D.A. Wright, and R.E. Kellems. 1990. Identification and characterization of transcriptional arrest sites in exon 1 of the human adenosine deaminase gene. Mol. Cell. Biol. 10:4555-4564.

Cheng, S.W.c., E.C. Lynch, K.R. Lenson, D.L. Court, B.A. Shapiro, and D.I. Friedman. 1991. Functional importance of sequence in the stem-Ioop of a transcription terminator. Science 254: 1205-1207. • 49

Chinsky, J. M., M-C Maa, V. Ramamurthy, and R. E. Kellems. 1989. • Adenosine deaminase gene expression: Tissue-dependent regulation of transcriptional elongation. J. Biol. Chem. 264:14561-14565.

Clark, A.B., C.C. Dykstra, and A Sugino. 1991. Isolation. DNA sequence. and regulation of a Saccharomyces cerevisiae gene that encodes DNA strand transfer protein ex. Mol. Cel!. Biol. 11:2576-2582.

Conaway, J.W., and R.C. Conaway. 19SiJ. An RNA polymerase II transcription

factor shares functional properties with Escllerichia coli cr711 • Science 248:1550­ 1553.

Court, D., C. Brady, and M. Rosenberg. 1980. Control of transcription termination: A rho-dependent terminator site in bacteriophage lambda. J. Mol. Biol. 138:231-254.

Cozzarelli, N.R., S.P. Gerrard, M SehlisseI, D.D. Brown, and D.F. Bogenhagen. 1983. Purified RNA polymerase III accurately and efficiently terminates transcription of 55 RNA genes. Cell 34:829-835.

Cullen, B.R. 1986. Trans-activation of human immunodeficiency virus occurs via a bimodal mechanism. Cell 46:973-982.

CulIen, B.R. 1990. The HIV-l Tat protein: an RNA sequence-specifie processivity factor? Cell 63:655-657. • 50

CulIen, B.R. 1993. Ooes HIV-1 induce a change in viral initation rights ? Cell • 73:417-420. Das, A. 1992. How the phage lambda N gene product supresses transcription termination: Communication of RNA polymerase with regulàtory protein mediated by signais in the nacent RNA. J. Bacteriol. 174:6711-6716.

Davies, R.W., and R. Hyman. 1970. Physical locations of the in vitro RNA initiation site and termination sites of T7M DNA. Cold Spring Harbor Symp. Quant. Biol. 35: 269-281.

Dendrick, R.L., C.M. Kane, and M.J. Chamberlin. 1987. Purified RNA polymerase II recognizes specifie termination sites during transcription in vitro. J. Biol. Chem. 262:9098-9108.

Deuschle, U., R. Gentz, and H. Bujard. 1986. lac repressor blocks transcribing RNA polymerase and terminates transcription. Proc. Nat!. Acad. Sci. VSA 83:4134-4137.

Deuschle, V., R.A. Hipskind, and H. Bujard. 1990. RNA polymerase II transcription blocked by Escherichia coli lac repressor. Science 248:480-483.

Dingwàll, c., E. Emberg, M.J. Gait, S.M. Green, S. Heaphy, J. Kam, A.D. Lowe, M. Singh, and M.A. Skinner. 1990. HIV-1 tat protein stimulates transcription by binding to a V-rich bulge in the stem of the TAR RNA structure. EMBO. J. • 9:4145-4153. 51

Eick, D., and G.W. Bomkamm. 1986. Transcriptional arrest within the first exon is a fast control mechaIÙsm in c-mye gene expression. Nucleic Acids Res. • 14:8331-8346.

Eick, D., R. Berger, A. Polack, and G.W. Bomkamm. 1987. Transcription of c­ mye in human mononuclear ceUs is regulated by an elongation block. Oncogene 2:61-65.

Enriquez-Harris, P., N. Levitt, D. Briggs, and N.J. Proudfoot. 1991. A pause site for RNA polymerase II is associated with termination of transcription. EMBO J. 10:1833-1842.

Falck-Pederson, E., J. Logan, T. Shenk, and J.E. Damel1, Jr. 1985. Transcription termination within the ElA gene of adenovirus induced by insertion of the mouse ~maj-globin terminator element. Cell 40:897-905.

Feinberg, M.B., D. Baltimore, and A.D. Frankel. 1991. The role of Tat in the human immunodificiency virus life cycle indicates a primary effect on transcriptional elongation. Proc. Natl. Acad. Sei. USA 88:4045-4049.

Finger, LR., and J.P. Richardson. 1982. Stabilization of the hexameric form of Escherichia coli protein rho under ATP hydrolysis conditions. J. Mol. Biol. 156:203-219.

Flores, O., E. Maldonado, and D. Reinberg. 1989. Factors involved in specific transcription by RNA polymerase II. Factor lIE and IIF independently interaet • with RNA polymerase II. J. Biol. Chem. 264:8913-8921. 52

Ford, J.P., and M.T. Hsu. 1978. Transcription pattern of in vivo-labeled late • simian virus 40 RNA: equimolar transcription beyond the mRNA 3' terminus. J. Virol. 28:795-801.

Fort, P., J. Rech, A. Vie, M. Piechaczyk, A. Bonnieu, P. Jeanteur, and J.M. Blanchard. 1987. Regulation of c10s gene expression in hamster fibroblasts: initiation and elongation of transcription and mRNA degradation. Nucleic Acids Res. 15:5657-5667.

Franklin, N.C. 1985. Conservation of genome form but not sequence in the

transcription antitermination determinates of Â., $21, P22. J. Molec. Biol. 181:86-91.

Friedman, 0.1. 1988. Regulation of phage gene expression by termination and antitermination of transcription. In The Bacteriophages vo12, Calendar, R. ed., pp263-318. Plenum, New York.

Gil, A., and N.J. Proudfoot. 1987. Position-dependent sequence elements downstream of AAUAAA are required for effecient rabitt ~-globin mRNA 3' end formation. Cel! 49:399-406.

Gill, S.e., S.E. Weitzel, and P.H. von Hippel. 1991. Escherichia coli (170 and nusA proteins. 1. Binding interactions with core RNA polymerase in solution within the transcriptional complex. J. Mol. Biol. 220:307-324 • 53

Gilmour, O. S., and J. T. Lis. 1986. ln VitlO interactions of RNA polymerase Il • with genes of Drosopltila melanogaster. Mol. Cell. Biol. 5:2009-2018. Gollnick, P., and C. Yanofsky. 1990. tRNATrp translation of leader peptide codon 12 and other factors that regulate expression of the tryptophanase operon. J. Bacteriol. 172:3100-3107.

G!lllnick, P., S. Ishino, M.I. Henner, and C. Yanofsky. 1990. The mtr locus is a two-gene operon reguired for transcription att~nuation in the trp operon of Bacillus subtilis. Proc. Natl. Acad. Sei. USA 87:8726-8730.

Gotta, S.L., O.L. Jr. Miller, and S.L. French. 1991. rRNA transcription rate in Escherichia coli:. J. Bacteriol, 173: 6647-6649.

Gottlieb, E., and J.A. Steitz. 1989a. Function of the mammalian La protein: evidence for its action in transcription termination by RNA polymerase Ill. EMBO J. 8:851-861.

Gottlieb, E., and J.A. Steitz. 1989b. The RNA binding protein La influences both the accuracy and the effeciency of RNA polymerase il transcription in vitro. EMBO J. 8:841-850.

Gralla, J.O., ed. 1989a. DNA-protein interactions in transcription. Alan R. Liss, New York.

Gralla, J.O. 1989b. Bacterial gene regulation from distant DNA sites. CeU • 57:193-195. 54

Grayhack, E.J., X. Yang, L.F. Lau, ar.d J.W Roberts. 1985. Phage lambda gene Q • antiterminator recognizes RNA polymerase near the promoter and accelerates it through a pause site. Cel! 42:259-269.

Greenblatt, J. 1992. Protein-protein interactions as critical determinants of regulated initiation and termination of transcription. In Transcriptional Regulation, S.L. McKnight and K.R Yamamoto, eds.,pp 203-226. Cold Spring Harbor Press, Cold Spring Harbor.

Greenblatt, J., and J. Li. 1981a. Interaction of the sigma factor and the nusA gene protein of E. coli with RNA polymerase in the initiation-termination cycle of transcription. Cel! 24:421-428.

Greenblatt, J., and J. Li. 1981b. The nusA gene product of Escherichia coli: Its identification and a demonstration that it interacts with the gene N transcription antitermination protein of bacteriophage lambda. J. Molec. Biol. 147:11-23.

Greenblatt, J., J.R. NodweIl, and S.W. Masan. 1993. Transcriptional antitermination. Nature 364:401-406.

Groudine, M., M. Peretz, and H. Weintraub. 1981. Transcriptional regulation of hemoglobin switching in chicken embryos. Mol. Cel!. Biol. 1:281-288. • 55

Grummt, 1., H. Rosenbaurer, 1. Niedermeyer, U. Maier, and A Ohrlein. 1986. • A repeated 18 bp sequence motif in the mouse rRNA spacer mediates binding of a nuclear factor and transcription termination. Cell 45:837-846.

Grummt, 1., U. Maier, A. Ohrlein, N. Hassouna, and J.-P. Bachellerie. 1985.

Transcription of mouse rRNA terminates downst~eam of the 3' end of 285 RNA and involves interaction of factors with repe.lted sequences in the 3' spacer. Ce1l43:801-810.

Hair, A., and G.T. Morgan. 1993. Premature termination of tubulin gene transcription in Xenopus oocytes is due to promoter-dependent disruption of elongation. Mol Cell. Biol. 13:7925-7934.

Haley, J. D., and M. D. Waterfield. 1991. Contributory effects of de novo transcription and premature transcript termination in the regulation of human epidermal growth factor receptor proto-oncogene RNA synthesis. J. Biol. Chem. 266:1746-1753.

Hauber, J., and B.R. Cullen. 1988. Mutational analysis of the trans-activation­ response region of the human immunodeficiency virus type 1 long terminal repeat. J. Virol. 62:673-679.

Hawley, D. K., and R. G. Roeder. 1985. Separation and partial characterization of three functional steps in transcription initiation by human RNA polymerase ll. J. Biol. Chem. 260:8163-8172. • 56

Hay, N., H. Skolnik-David, and Y. Aloni. 1982. Attenuation in the control of • SV40 gene expression. Cell 29:183-193. He, B., A. Shiau, K.Y. Choi, H. Zalkin, and J.M. Smith. 1990. Genes of the Escherichia coli pur regulon are negatively controlled by a repressor-operator interaction. J. Bacteriol. 172:4555-4562.

He, B., and H. Zalkin. 1992. Repression of Escherichia coli purB is by a transcriptional roadblock mechanism. J. Bacteriol. 174:7121-7127.

He, B., J.M. Smith, and H. Zalkin. 1992. Escherichia coli purB gene: cloning, nucleotide sequence, and regulation by purR. J. Bacteriol. 174:130-136.

Hernandez, N., and A. M. Weiner. 1986. Formation of the 3' end of Ul snRNA requires compatible snRNA promoter elements. Cell 47:249-258.

Hernandez, N., and R. Lucito. 1988. Elements required for transcription initiation of the human U2 snRNA gene coincide with elements required for snRNA 3' end formation. EMBO J. 1:3125-3134.

Hirashima, S., H. Hirai, Y. Nakanishi, and S. Natori. 1988. Molecular cIoning and characterization of cDNA for eukaryotic SI!. J. Biol. Chem. 263:3858-3863.

Hochschild, A. 1990. Protein-protein interactions and DNA loop formation. In DNA topology and its biological effeets. NoR Cozzarelli and J.e. Wang eds. • ppl07-138. Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. Si

• Horikoshi, M., K. 5ekimizu, and 5. Natori. 1984. Analysis of the stimulatory factor of RNA polymerase II in the initiation and elongation complex. J. Biol. Chem.259:608-611.

Houman, F., M.R. Diaz-Torres, and A. Wright. 1990. Transcriptional antitermination in the bgl operon of E. coli is modulated by a specifie RNA binding protein. Ce1l62:1153-1163.

Ishihama, A. 1992. Role of the RNA polymerase alpha subunit in transcription activation. Mol. Microbiol. 6: 3283-3288.

Ishii, 5., M. Ihara, T. Maekawa, Y. Nakamura, H. Uchida, and F. Imamoto. 1984. The nudeotide sequence of the cloned nusA gene and its flanking region of Escherichia coli. Nucleic Acids Res. 12:3333-3342

Ito, K., E. Kohji, and Y. Nakamura. 1991. Genetic interaction between the 13' subunit of RNA polymerase and the argenine-rich domain of Escherichia coli nusA protein. J. Bacteriol. 173: 1492-1501.

Izaban, M.G., and O.S. Luse. 1992a. The RNA polymerase Il temary complex cleaves the nascent transcript in a 3'-5' direction in the presence of elongation factor SIl. Genes Dev. 6:1342-1356.

Izaban, M.G., and O.S. Luse. 1992b. Factor-stimulated RNA polymerase Il transcribes at physiological rates on naked DNA but very poorly on • chromatin templates. J. Biol. Chem. 267:13647-13655. 58

Jakobovits, A., D.H. Smith, E.B. Jakobovits, and D.J. Capon. 1988. A discrete • element 3' of human immunodificiency virus 1 (HIV-1) and HIV-2 mRNA initiation sites mediates transcriptional activation by an HIV trans activator. Mol. Cell. Biol. 8:2555-2561.

Jin, D.J., and C.A. Gross. 1988. Mapping and sequenceing of mutations in the Escherichia coli rpoB gene that leads to rifampicin resistance. J. Mol. Biol. 202: 45-58.

Jin, D.J., and C.A. Gross. 1989. Three rpoBC mutations that supress the termination defects of rho mutants also affect the function of nusA mutants. Mol Gen Genet. 216: 269-275.

Kainz, M., and J.W. Roberts. 1992. Structure of transcription elongation complexes in vivo. Science 255:838-841.

Kao, S.-Y., A.F. Calman, P.A. Luciw, and B.M. Peterlin. 1987. Anti-termination of transcription within the long terminal repeat of HIV-1 by tat gene product. Nature 330:489-493.

Kashlev, M., J. Lee, K. Zalenskaya, V. Nikiforov, and A. Goldfarb. 1990. Blocking the initiation-to-elongation transition by a transdominant RNA polymerase mutation. Science 248:1006-1009. • 59

Kassavetis, G.A., and M.J. Chamberlin. 1981. Pausing and termination of • transcription within the early region of bacteriophage T7 DNA in 'lifra. J. Biol. Chem. 256:2m-2786

Kato, H., H. Sumimoto, P. Pognonec, C.-H. Chen, C.A. Rosen, and R.G. Roeder. 1992. HIV-1 Tat acts as a processivity factor in vitro in conjunction with cellular elongation factors. Genes and Dev. 6:655-666.

KerpolIa, T.K., and C.M. Kane. 1988. Intrinsic sites of transcription termination and pausing in the c-mye gene. Mol Cell. Biol. 8:4389-4394.

Kerppola, T. K., and C. M. Kane. 1991. RNA polymerase: regulation of transcript elongation and termination. FASEB J. 5:2833-2842.

Kerppola, T.K., and C.M. Kane. 1990. Analysis of the signaIs for transcription termination by purified RNA polymerase II. Biochem. 29:269-278.

Killen, M.T., and J. Greenblatt. 1992. The general transcription factor RAP30 binds to RNA polymerase II and prevents it from binding non-specifically to DNA. Mol. Cell. Biol. 12:30-37.

Kingston, R.E., and M.J. Chanberlin. 1981. Pausing and attenuation of in vitro transcription in the rrnB operon of E. coli. Cell 27:523-531.

Krumm, A., T. Meulia, M. Brunvand, and M. Groudine. 1992. The bloc!< to transcriptional elongation within the human c-mye gene is determined in the • promoter-proximal region. Genes Dev. 7:2201-2213. 60

Krystal, G., M. Birrer, J. Way, M. Nau, E. SausviIIe, C. Thompson, J. Minna, • and J. Battey. 1988. Multiple mechanisms for transcriptional regulation of the myc gene family in small-cell lung cancer. Mol. Cell. Biol. 8:3373-3381.

Kuhn, A., I. Bartsch, and I. Grummt. 1990. Specifie interaction of the murine transcription terminator factor TIF! with class-I RNA polymerases. Nature 344:559-562.

Landick, R., and C. Yanofsky. 1984. Stability of an RNA secondary structure affects in vitro transcription pausing in the trp operon leader region. J. Biol. Chem. 259: 1155-11555.

Landick, R., and C. Yanofsky. 1987. Isolation and structural analysis of the Escherichia coli trp leader paused transcriptional complex. J. Mol. Biol. 196: 363-377.

Landick, R., and C.L. Jr. Turnbough. 1992. Transcriptional attenuation. In Transcriptional Regulation, S.L. McKnight and K.R. Yamamoto, eds.,pp 407­ 446. Cold Spring Harbor Press, Cold Spring Harbor.

Landick, R., J. Stewart, and D.N. Lee. 1990. Amino acid changes in conserved regions of the beta-subunit of Escherichia coli RNA polymerase alter tran:.cription pausing and termination. Genes Dev. 4: 1623-1636.

Lanoix, J., and N.H. Acheson. 1988. P. rabbit l3-globin polyadenylation signal • directs efficient termination of polyomavirus DNA. EMBO J. 7:2515-2522. 61

Laspia, M.F., A.P. Rice, and M.B. Mathews. 1989. HIV-1 Tat protein increases • initiation and stabilizes elongation. Cell 59:283-292.

Lau, L.F., J.W. Roberts, and R. Wu. 1983. RNA polymerase pausing and transcript release at the ÀtRl terminator in vitro. J. Biol. Chem. 258:9391-939ï

Lazinski, D., E. Grzadzielska, and A. Das. 1989. Sequence-specifie recognition of RNA hairpins by bacteriophage antiterminators requires a conserved argeninie-rich motif. CeU 59:207-218.

Lee, D.N., L Phung, J. Stewart, and :{. Landick. 1990. Transcription pausing by Escherichia coli RNA polymerase is modulated by downstream DNA sequences. J. Biol. Chem. 265: 15145-15153.

Lee, H., K.W. Kraus, M. F. Wolfner, and J. T. Lis. 1992. DNA sequence requirements for generating paused polymerase at the start of hsp7D. Genes Dev. 6:284-295.

Lee, J.Y., M. Kashlev, S. Borukhov, and A. Goldfarb. 1991. A ~ subunit mutation disrupting the catalytic function of Escherichia coli RNA polymerase. Proc. Nat!. Acad. Sci. USA 88: 6018-6022.

Levin, J.R., and M.J. Chamberlin. 1987. Mapping and characterization of transcriptional pause sites in the early genetic region of bacteriophage T7. J. Mol. Biol. 196:61-8/ 62

Levin, J.R., B. Krummel, and M.J. Chambe~lin. 1987. Isolation and properties of transcribing ternary complexes of Escherichia coli RNA polymerase • positioned at a single template base. J. Mol. Biol. 196:85-100.

Li, J.• R. Horwtz, S. McCracken, and J. Greenblatt. 1992. NusG, a new Escherichia coli elongation factor involved in transcriptional antitermination by the N protein of phage lambda. J. Biol. Chem. 267:6012-6019.

Li, J., S.W. Mason, and J. Greenblatt. 1993. Elongation factor NusG interacts with termination factor rho to regulate termination and antitermination of transcription. Genes Dev. 7:161-172.

Li, S.W., c.L. Squires, and C. Squires. 1984. Antitermination of E coli rRNA

transcription is caused by a control region segment containing lambda nut­ like sequences. Ce1l38:851-860.

Lindsten, T., C.H. June, and C.B. Thompson. 1988. Multiple mechanisms regulate c-myc expression during normal T cel! activation. EMBO J. 7:2787­ 2794.

Logan, J., E. Falck-Pederson, J.E. Damell Jr, and T. Shenk. 1987. A poly(A) addition site and a downstream termination region are required for effecient cessation of transcription by RNA polymerase II in the mouse 13 major-globin gene. Proc. Nat!. Acad. Sei. USA 84:8306-8310.

Lowery, c., and J.P. Richardson. 1977a. Characterization of the nucleoside • triphosphate phosphohydrolase (ATPase) activity of RNA synthesis 63

termination factor rho. 1. Enzymatic properties and effects of inhibitors. J. • Biol. Chem. 252:1375-1380. Lowery, c., and J.P. Richardson. 19ïïb. Characterization of the nucleoside triphosphate phosphohydrolase (ATPase) activity of RNA synthesis termination factor rho. II Influence of synthetic RNA homopolymers and random copolymers on the reaction. J. Biol. Chem. 252:1381-1385.

Lynn, S.P., L.M. Kasper, and J.F. Gardner. 1988. Contributions of RNA secondary structure and lenght of the thymidine tract to transcription termination at the thr operon attenuator. J. Biol. Chem. 263: 472-479.

Maa, M., J.M. Chinsky, V. Ramamurthy, B.D. Martin, and R.E. Kellem~ 1990.

Identification of transcription start sites at the 5' and ;j' ends of the murine adenosine deaminase gene. J. Biol. Che:n. 2:65:12513-12519.

Maderius, A., and S. Chen-Kiang. 1984. Pausing and premature termination of human RNA polymerase II during transcription of adenovirus in vivo and in vitro. Proc. Nat!. Acad. Sei. USA 81:5931-5935.

Mahadevan, S., and A Wright. 1987. A bacterial gene involved in transcription antitermination: Regulation at a rho-independent terminator in the bgl operon of E. coli. Cell 50:485-494.

ManIey, J.L., N.J. Proudfoot, and T. Platt. 1989.3' end formation. Gen~s Dev. • 3:2218-2222. 64

Marciniak, R.A., and P.A. Sharp. 1991. The HIV-1 Tat protein promotes • formation of more-processive elongation complexes. EMBO J. 10:4189-4196. Marciniak, R.A., M.A. Garcia-Blanco, and P.A. Sharp. 1990. Identification and characterization of a HeLa nuclear protein that specifically binds to the trans­ activation-response (TAR) element of human immunodeficiency virus. Proc. Nat!. Acad. Sci. USA. 87:3624-3628.

Marshall N.F., and D.H. Priee. 1992. Control of formation of two distinct classes of RNA polymerase II elongation complexes. Mol. Cell. Biol. 12:2078­ 2090.

Marshall, T.K., H. Guo, and D.H. Priee. 1990. Drosophila RNA polymerase II elongation factor Dm5-II has homology to mouse 5-II and sequence similarity to yeast PPR2. Nudeic Acids Res. 18:6293-6298.

Martin, F.H., and I.J. Tinoca. 1980. DNA-RNA hybrid duplexes containing oligo(dA:rU) sequences are exceptionally unstable and may facilitate termination of transcription. Nudeic Acids Res. 8: 2295-2305.

Mason, S.W., and J. Greenblatt. 1991. Assembly of transcription elongation

complexes containing the N protein of phage À. and the Escherichia coli elongation factors NusA, NusB, NusG, and 510. Genes Dev. 5:1504-1512.

Masan, S.W., J. Li, and J. Greenblatt. 1992a. Host factor requirements for processive antitermination of transcription and suppression of pausing by the • N protein of bacteriophage lambda. J. Biol. Chem. 267:19418-19426. 65

Mason, S.W., J. Li, and J. Greenblatt. 1992b. A direct interaction between two • Escherichia coli transcription antitermination factors, NusB and ribosomal protein S10. J. Mo!. Biol. 223:55-66.

Mathews, M.B., and A.M. Francoeur. 1984. La antigen recognizes and binds to the 3' oligouridylate tail of a small RNA. Mol. Cel!. Biol. 4:1134-1140.

Mazabraud, A., O. Scherly, F. Muller, O. Rungger, and S.G. Clarkson. 1987. Structure and transcription termination of a lysine tRNA gene from Xenopus Iaevis. J. Mol. Biol. 195:835-345.

McKnight, G.S., and R. Palmiter. 1979. Transcriptional regulation of the ovalbumin and conalbumin genes by steriod hormones in chick oviduct. J. Biol. Chem. 254:9050-9058.

Mechti, N., M. Piechaczyk, J.M. Blanchard, L Marty, A. Bonnieu. P. Jeanteur, and B. Lebeau. 1986. Transcriptional and post-transêriptional regulation of c­ myc expression during the differentiation of murine erythroleukemia Friend cells. Nucleic Acids Res. 14:9653-9666.

Meng, L.M., M. Kilstrup, and P. Nygaard. 1990. Autoreguhtion of purR repressor synthesis and involvment of purR in the regulation of purB, purC, purI, purMN and guaBA expression in Escherichia coli. Eur. J. Biochem. 187:373-379. • 66

Meulia, T., A. Krumm, and M. Groudine. 1993. Distinct properties of c-mye transcription elongation are revealed in Xenopus oocytes and mammalian • ceUs and by template titration, 5,6-didùoro-1-I3-o-ribofuranosylbenzimidazole (DRB), and promoter mutagenesis. Mol Cel! Biol. 13:5647-5658.

Meulia, T., A. Krumm, C. Spencer, and M. Groudine. 1992. Sequences in the human c-mye P2 promoter affect the elongation and premature termination of transcripts initiated from the upstream Pl promoter Mol. Cel!. Biol. 12:4590-4600.

Miller, H., C. Asselin, D. Dufort, J-Q. Yang, K. Gupta, K.B. Marcu, and A. Nepveu. 1989. Acis-acting element in the promoter region of the murine c­ mye gene is necessary for transcriptional bloc!<. Mol. Cel!. Biol. 9:5340-5349.

Morgan, E.A. 1980. Insertion of Tn10 into E. coli ribosomal RNA operon are incompletely polar. Cel! 21:257-265.

Morgan, E.A. 1986. Antitermination mechanisms in rRNA operons of Eseheriehia coli. J. Bact. 168:1-5.

Morgan, W.D., D.J. Bear, and P.H. von HippIe. 1983. Rho-dependant termination of transcription. II. Kinetics of mRNA elongation during transcription from the bacteriophage À.PR promoter. J. Biol. Chem. 258: 9565­ 9574. • 6ï

Morrow, M.A., G. Lee, S. Gillis, G.D. Yanacopoulos, and F.W. AIt. 1992. Interlukin-ï induces N-myc and c-myc expression in normal B lymphocytes. • Gene Dev. 6:61-ïO.

Mosteller, R.D., and C. Yanofsky. 19ïO. Transcription of the tryptophan operon in Eseheriehia coli: rifampicin as an inhibitor of initiation. J. Mol. Biol. 48: 525-531.

Musing, M.A., D.H. Smith, and D.J. Capon. 198ï. Regulation of mRNA accumulation by a human immunodificiency virus trans-activator protein. CeIl48:691-701.

Nelbock, P., P.J. Dillon, A. Perkins, and C.A. Rosen. 1990. A cDNA for a protein that interacts with the human immunodeficiency virus Tat transactivator. Science 248:1650-1653.

Nepveu1 A., and K.B. Marcu. 1986. Intragenic pausing and anti-sense transcription within the murine c-mye locus. EMBO J. 5:2859-2865.

Nepveu, A., R.A. Levine, J. Campisi, M.E. Greenberg, E.B. Ziff and K.B. Marcu. 1987. Alternative modes of c-mye regulation in growth factor­ stimuiated and differentiating cells. Oncogene 1:243-250.

Neuman de Vegvar, H. E., E. Lund, and J. E. Dahlberg. 1986. 3' end formation of Ul snRNA precursors is coupled to transcription from snRNA promoters. • Cell47:259-266. 68

Nevins, ].R., and] E. DarnelI. 1978. Steps in the processing of Ad2 mRNA: poly(A) nuclear sequences are conserved and poly(A) addition precedes • splicing. Cell15:1477-1493.

Nodwell, ].R., and ]. Greenblatt. 1991. The nut site of bacteriophage lambda is made of RNA and is bound by transcription antitermination factors on the surface of RNA polymerase. Genes Dev. 5:2141-2151.

Nodwell, ].R., and ]. Greenblatt. 1993. Recognition of boxA antiterminator RNA by the E coli antitermination factors NusB and ribosomal protein 510. Cell 72:261-268.

Parry, H.D., G. Tebb, and LW. Mattaj. 1989. The Xenopus U2 gene PSE is a single, compact, element required for transcription initiation and 3' end formation. Nucleic Acids Res. 17:3633-3645.

Peterlin, B.M., P.A. Luciw, P.J. Barr, and M.D. Walker. 1986. Elevated levels of mRNA can account for the trans-activation of human immunodeficiency virus. Proc. Nat!. Acad. Sei. USA. 83:9734-9738.

Pinkham, J.L, and T. Platt. 1983. Nucleotide sequence of the E. coli rho gene and its regulatory regions. Nucleic Acids Res. 11:3531-3545.

Platt, T., and J.P. Richardson. 1992. Escherichia coli rho factor: protein and enzyme of transcription termination. In Transcriptional Regulation, S.L. McKnight and K.R Yamamoto, eds.,pp 365-388. Cold Spring Harbor Press, • Cold Spring Harbor. 69

• Priee, D.H., A.E. Sluder, and A.L. Greenleaf. 1989. Dynamic interaction between a Drosophila transciption factor and RNA polymerase II. Mo!. Cel!. Biol. 9:1465-1475.

Priee, D.H., A.E. Slunder, and A.L. Greenleaf. 1987. Fractionation of transcription factors for RN:'. polymerase II from Drosophila Kc cell nuc1ear extracts. J. Biol. Chem. 262:3244-3255.

Proudfoot, N.J. 1989. How RNA polymèrase Il terminates transcription in higher eukaryotes. TrBS 14:105-110.

Proudfoot, N.J., and E. Whitelaw. 1988. Termination and 3' end processing of eukaryotic RNA. In Frontiers in Molecular Biology, B.D. Hames and D.M. Glover, eds. IRL Press, Oxford.

Proudfoot, N.J., and G.G. Brownlee. 1976.3' non-coding region sequences in eukaryotic mRNA. Nature 263:211-214.

Ramamurthy, V., M-C Maa, M.L. Harless, D.A. Wright, and R.E. Kellems. 1990. Sequence requirements for transcriptional arrest in exon 1 of the murine adenosine deaminase gene. Mol. Cell. Biol. 10:1484-1491.

Rappaport, J., D. Reinberg, R. Zandomeni, and R. Weinmann. 1987. Purification and functional characterization of transcription factor SIl from calf thymus. Role in RNA polymerase Il elongation. J. Biol Chem. 262:5227­ • 5232. 70

• Reddy, C.O., and E.P. Reddy. 1989. DifferentiaI binding of nuclear factors to the intron 1 sequences containing the transcriptional pause sites correlates with c­ myb expression. Proc Natl. Acad. Sei. USA 86:7326-7330

Reinberg, D., and R.G. Roeder. 1987. Factors involved in specifie transcription by mammalian RNA polymerase II. Transcription factor Ils stimulates elongation of RNA chains. J. Biol. Chem. 262:3331-3337.

Reinberg, O., M. Horikoshi, and R.G. Roeder. 1987. Factors involved in specifie transcription by mammalian RNA polymerase II: functional analysis of initiation factors lIA and !ID and identification of a new factor operating at sequences downstream of the initiation site. J. Biol. Chem. 262:3331-3337.

Reines, D. 1992. Elongation factor-dependent transcript shortning by template­ engaged RNA polymerase II. J. Biol. Chem. 267:3795-3800.

Reines, D., and J. Mote, Jr. 1993. Elongation factor SII-dependent transcription by RNA polymerase II through a sequence-specific DNA binding protein. Proc. Nat!. Acad. Sci. USA 90:11917-11921.

Reines, D., D. Wells, M.J. Chamberlin, and C.M. Kane. 1987. Identification of intrinsie termination sites in vitro for RNA polymerase II within eukaryotic gene sequences. J. Mol. Biol. 196:299-312. • 71

Reines, D., M. Chamberlin, and C.M. Kane. 1989. Transcription elongation • factor SII (TFIIS) enables RNA polymerdse II to elongate through a block to transcription in a human gene in vitro. J. Biol. Chem. 264:10799-10809.

Reines, D., P. Ghanouni, Q. Li., and J. Mote, Jr. 1992. The RNA polymerase II elongation complex. Factor-dependent transcription elongation involves nascent RNA cleavage. J. Biol. Chem. 267:15516-15522.

Reynolds, R., and M.J. Chamberlin. 1992. Parameters affecting transcription termination by Escheric1lia coli RNA polymerase. II. Construction and analysis of hybrid terminators. J. Mol. Biol. 224: 53-63.

Reznikov, O., M. Kessler, and Y. Aloni. 1989. RNA secondary structure is an integral part of the in vitro mechanism of attenuation in Simian virus 40. J. Biol. Chem. 264:9953-9959.

Rice, G.A., C.M. Kane, and M.J. Chamberlin. 1991. Footprinting analysis of mammalian RNA polymerase II along its transcript: an alternative view of transcription elongation. Proc. Nat!. Acad. Scï. USA 88:4245-4249.

Richardson, J. 1990. Rho-dependent transcription termination. Biochem. Biophys. Acta. 1048:127-138.

Roberts, J. 1992. Antitermination and the control of transcription elongation. In Transcriptional Regulation, S.L. McKnight and K.R Yamamoto, eds.,pp • 389-406. Cold Spring Harbor Press, Cold Spring Harbor. 72

Roberts, J.W. 1988. Phage lambda and the regulation of transcription • termination. Cell 52:5-6. Rosen, C.A., J.G. 50droski, and W.A. Haseltine. 1985. The location of cis-acting regulatory sequences in the human T-celllymphotropic virus type m (HTLV­ ill/LAV) long terminal repeat. Cel! 41:813-823.

Rosenberg, M., and D. Court. 1979. Regulatory sequences involved in the promotion and termination of RNA transcription. Annu. Rev. Genet. 13:319­ 353.

Rougvie A.E., and J.T. Lis. 1988. The RNA polymerase II molecule at the 5' end of the uninduced hsp70 gene of D. melanogaster is transcriptionally engaged. Cel! 54:795-804.

Rougvie A.E., and J.T. Lis. 1990. Postinitiation transcriptional control in Drosophila melanogaster. Mol. Cell. Biol. 10:6041-6045.

Roy, 5., U. Delling, C.-Ho Chen, C.A. Rosen, and N. 50nenberg. 1990. A bulge structure in ffiV-1 TAR RNA is required for Tat binding and Tat-mediated trans-activation. Genes Dev. 4:1365-1373.

Russo, F.O., and T.J. 5ilhavy. 1992. Alpha: The cinderella subunit of RNA polymerase. J. Biol. Chem 267: 14515-14518. • i3

Salditt-Georgieff, M., M. Sheffery, K. Krauter, J.E. Damell, Jr., R. Rifkind, and P.A. Marks. 1984. Induced transcription of the mouse ~-globin transcription • unit in erythroleukemia cel1s. J. Mo!. Bio!. 172:437-450.

Sarrock, R.A., R.t. Gourse, and M. Nomura. 1985. Defective antitermination of rRNA transcription and derepression of rRNA and tRNA synthesis in the nusBS mutant of Esc1leric11ia coli. Proc. Nat!. Acad. Sci. 82:5275-5279

Sawadogo, M., A. Sentenac, and P Fromageot. 1980. Interaction of a new polypeptide with yeast RNA polymerase B. J. Bio!. Chem. 225:12-15.

Schartz, M. and J. Neuhard. 1975. Control of expression of the pry genes in Salmonella typhimurium: Effects of variations in uridine and cytidine nucleotide pools. J. Bacterio!. 121:814-822.

Schmidt, M.C., and M.J. Chamberlin. 1987. NusA protein of Eschericlria coli is an efficient transcription termination factor for certain terminator sites. J. Mol. Biol. 195: 809-818.

Schmidt, M.C., and M.J. Chamberlin. Amplification and isolation of Escherichia coli nusA protein and studies of its effects on in vitro RNA chain elongation. Biochemistry 23:197-203.

. Schneider-Schaulies, J., A. Schimpl, and E. Wccker. 1987. Kinetics of cellular oncogene expression in mouse lymphocytes II. Regulation of c-fos and c-myc • gene expression. Eur. J. Immunol. 17:713-718. ï4

Selby, M.J., and B.M. Peterlin. 1990. Trans-activation bv HIV-l Tat via a • heterologous RNA binding protein. Cell 62:ï69-ïï6. Sellitti, M.A., P.A. Paveo, and D.A. Steege. 1987. lac repressor blacks in l'iI'O transcription of lac control region DNA. Proc. Nat!. Acad. Sei. USA 84:3199­ 3203.

Sheldon, M., R. Ratnasabapathy, and N. Hemandez. 1993. Characterization of the inducer of short transcripts, a human immunodeficiency virus type 1 transcriptional element that activates the synthesis of short RNAs. Mo!. Cel!. Biol. 13:1251-1263.

Sheline, C.T., L.H. Milocco, and K.A. Jones. 1991. Two distinct nuclear transcription factors recognize loop and buldge residues of the HIV-1 TAR RNA hairpin. Genes Dev. 5:2508-2520.

Shibuya, H., K. Irie, J. Ninomiya-Tsuji, M. Grobl, T. Taniguchi, and K. Matsumoto. 1992. New human gene encoding a positive modulator of HIV Tat-mediated transactivation. Nature 357:700-702.

Shimotsu, H., M.I. Kuroda, C. Yanofsky, and D.J. Henner. 1986. Novel form of transcription attenuation regulates e:::?ression of the Bacillus subtilis tryptophane operon. J. Bacteriol. 166:461-471.

SivaRaman, L., D. Reines, and c.M. Kane. 1990. Purified elongation factor sn • is sufficient to promote read-through by purified RNA polymerase II at 75

specifie termination sites in the human histone H3.3 gene. J. Biol. Chem. • 265:14554-14560. Skames, W. C, D. C Tessier, and N. H. Acheson. 1988. RNA polymerases stall and/or terminate nearby both early and late promoters on polyoma virus DNA. J. Mol. Biol. 203:153-171.

Sluder, A.E., A.L. Greenleaf, and D.H. Priee. 1989. Properties of a Drodophila RNA polymerase II elongation factor. J. Biol. Chem. 264:8963-8969.

Sopta, M., R.W. Carthew, and J. Greenblatt. 1985. Isolation of three proteins that bind to mammalian RNA polymerase II. J. Biol. Chem. 260:10353-10360.

Southgate, C., and M.R. Green. 1991. The HIV-1 Tat protein activates transcription from an upstream DNA-binding site: Implications for Tat function. Genes and Dev. 5:2496-2507.

Southgate, C, M.L. Zapp, and M.R. Green. 1990. Activation of transcription by HIV-1 Tat protein tethered to nascent RNA through another protein. Nature 345:640-642.

Spencer, C. A., and M. Groudine. 1990. Transcription elongation and eukaryotic gene regulation. Oncogene 5:777-785.

Spencer, CA., and M.A. Kilvert. 1993. Transcription elongation in the human c-rny' gene is govemed by overall transcription initiation levels in Xenopus • oocytes. Mol. Cell. Biol. 13:1296-1305. 76

Spencer, C.A., R.C. LeStrange, U. Novak, W.S. Hayward, and M. Groudine. • promot~r 1990. The block to transcription elongation is dependent in normal and Burkitt's lymphoma c-myc alleles. Genes Dev. 4:75-88.

Stefano, J.E. 1984. Purified lupus antigen La recognizes an oligouridylate stretch common to the 3' termini of RNA polymerase n transcripts. Cell 36:145-154.

Stewart, V., and C. Yanofsky. 1985. Evidence for transcription antitermination control of tryptophanase operon expression in Escherichia coli K-12. J. Bacteriol. 164:731-740.

Strobl, L., and O. Eick. 1992. Hold back of RNA polymerase n at the transcription start site mediates down-regulation of c-myc in vivo. EMBO J. 11:3307-3314.

Sullivan, S.L., and M.E. Gottesman. 1992. Requirement for E. coli NusG protein in factor-dependent transcription termination. Cell 68:989-994.

Surratt, c.K., S.c. Milan, and M.J. Chamberlin. 1991. Spontaneous cleavage of RNA in temary complexes of Escherichia coli RNA polymerase and its significance for the mechanism of transcription. Proc. Nat!. Acad. Scï. USA 88:7983-7987. • ï7

Telesnitsky, A., and M.J. Chamberlin. 1989. Termin..tor-distal sequences derermine the in vitro effeeieucy of the eariy terminators of bacteriophages T3 • and 17. B:ochemistry 28: 5210-5218.

Toohey, M. G., and K. A. Jones. 1989. In vitro formation of short RNA

polymerase II transcripts that terminate within the HIV-1 and HIV-2 promoter-proximal downstream regions. Genes Dev. 3:265-282.

Turnbough, C.L. Jr. 1983. Regulation of Escherichia coli aspartate transcarbamylase synthesis by guanosine tetraphosphate and pyrimidine ribonucleoside triphosphates. J. Bacterio!. 153:998-1007.

Vandenbergh, D.J., M. James-Peterson, and R.C. Hardison. 1991. An apparent pause site in the transcription unit of the rabbit a-globin gene. J. Mol. Bio!. 220:225-231.

von Hippel, P.H., and T.D. Yager. 1991. Transcription elongation and termination are competitive kinetie processes. Proe. Nat!. Aead. Sei. USA 88: 2307-2311.

Wahle, E., and W. Keller. 1992. The bioehemistry of 3' end cleavage and polyadenylation of messenger RNA precursors. Ann. Rev. Biochem. 61:419­ 440.

Wang, D., and D.K. Hawley. 1993. Identification of a 3'-5' exonuclease actiVÏty associated with human RNA polymerase II. Proe. Nat!. Aead. Sei. USA 90:843­ • 847. 78

• Watson, J.'3., D.W. Chandler, and J.D. Gralla. 1984. Specifie termination of in vitro transcription by calf thymus RNA polymerase III. Nucleic Adds Res. U:5369-5384.

Watson, R.J. 1988. <\ transcriptional arrest mechanism involved in controlling constitutive levels of mouse c-myb mRNA. Oncogene 2:267-272.

Whalen, W., B. Ghosh, and A Das. 1988. NusA protein is necessary and sufficient in vitro for phage À.N gene product to supress a p-independent terminator placed downstream of nutL. Proc. Natl. Acad. Sei. 85:2494-2498.

Whitelaw, E., and N.J. Proudfoot. 1986. Alpha thalassaemi:: caused by poly(A) site mutation reveals that transcription is linked to 3' end processing in the human tt-2 globin gene. Embo J. 5:2915-2922.

Wiest, D.K., D. Wang, and O.K. Hawley. 1992. Mechanistic studies of transcription arrest at the adenovirus major late attenuation site. J Biol Chem. 267:7733-7744.

Wright, S., and J.M. Bishop. 1989. DNA sequences that mediate attenuation of transcription from the mouse protooncogene mye. Proc. Nat!. Acad. Scï. USA 86:505-509.

Xu, M., M.B. Barnard, S.M. Rose, P.N. Cockeril1, S.-Y. Huang, and W.T. Garrard. 1986. Transcriptional termination and chromatin structure of the • active immunoglobulin li: gene locus. J. Biol. Chem. 261:3838-3845. 79

Yager, T.D., and P.H. von Hippie. 1987. In Escherichia coli and Salmonella • typhimurium: Cellular and Molecular Biology. edited by: Neidhardt,F.C., Ingraham, J.L., Low, K.B., Magasanik, B., Schaechter,M. and Umbarger, H.E. pp. 1241-1275, American Society for Microbiology, Washington, D.C.

Yang, X., and J.W. Roberts. 1989. Gene Q antitermination proteins of

Escherichia coli phages 82 and À. supress pausing by RNA polymerase at a p­ dependent terminator and at other sites. Proc. Nat!. Acad. Sei. USA 86:5301­ 5305.

Yang, x., C.M. Hart, E.J. Grayhack, and J.W. Roberts. 1987. Transcription

antitermination by phage À. gene Q protein requires a DNA segment spanning the RNA start site. Genes Dev. 1:217-226

YamelI, W.5., and J.W. Roberts. 1992. The phage À. gene Q transcription antiterminator binds in the late gene promoter as it modifies RNA polymerase. Cell 69:1181-1189.

Yoo, O., H. Yoon, K. Baek, C. Joen, K. Miyamoto, A. Ueno, and K. Agarwal. 1991. Cloning, expression and characterization of the human transcription elongation factor, TFIIS. Nudeic Acids Res. 19:1073-1079.

Yue, x., P. Favot, T.L Dunn, A.I. Cassady, and D.A. Hume. 1993. Expression of mRNA encoding the macrophage colony stimulating factor receptor (c-jms) is controlled by a constitutive promoter and tissue-specifie transcription • eiongation. Mol. Cell. Biol. 13:3191-3201. sa • CHAPTERII A Protein Bindins Site f~om the Murine C-Myc Promot-er Contributes to Transcriptional Block

Preface

Block to the elongation of transcription has been shown to be important in regulating c-myc gene expression. Analysis of the sequences required for transcriptional block suggested that sequences within the promoter were important. A 16 base pair deletion which removed part of a previously identified protein binding site termed ME1a1 was shown to significantly reduce block to transcription elongation suggesting that this site may be important for block. To demonstrate that this r.ite was involved in block to transcription elongation, oligonucleotides encoding this site were inserted in a heterologous promoter construct. This site was shown to significantly increase block to transcription elongation demonstrating that this site was indeed important for block. Furthermore, a correlation was established between the ability of this site to bind to cellular factors in vitro and the ability to confer block to transcription elongation in vivo strongly suggesting that proreins binding to this site are important for block.

• 81

Abstract

• sever~ Recent studies have revealed that the expression of eukaryotic genes can be regulated at the level of transcription elongation. As a first step to elucidate the mechanism by which tr:mscription elongation is modulated, several groups have identified sequences necessary for transcriptional block within the e-mye gene. These studies indicated that transcriptional block depends not only on sequences surrOll..'lding the sites of block, but also on sequences within the promoter: some deletions within the e-mye promoter eliminated transcriptional block and, with chimeric constructs, transcriptional block was observed when some heterologous promoters but not others were fused to the e-mye termination region. Using a chimeric construct containing the H-2Kb major histocompatibility class gene promoter linked to the e-mye first exon, we show that transcriptional block is increased by the addition of a 2S bp DNA sequence from the e-mye promoter. 5imilar results are obtained whether this sequence is inserted upstream or downstream of the transcription initiation site. We further show that nuclear factors interact with this sequence in vitro. Interestingly, when a mutated version of this sequence was tested, we observed decreased nuclear factor binding in vitro as weIl as redl:c·"':l transcriptional block in nuclear run-on transcription assays. These results suggest that interactions of protein factors with specifie nucleotide sequences near the transcription initiation site can affect elongation of transcription at sites located further downstream. • 82 • Introduction Control of gene expression in eukaryotes has been shewn to invol\'e modulation of transcription initiation rates. There is clear evidence that efficiency of transcription initiation depends on interactions of nuclear proteins with specifie binding sites in gene promoters. Spedfic gene regulation would depend on the unique combination of binding sites which have the potential to be recognized by various factors in different ceUs. Recent studies have indicated however that the rate of transcription initiation does not necessarily determine the quantity of mRNA that is synthesized. Indeed, for several eukaryotic genes, regulation is also exerted at the level of transcription elongation (reviewed in Spencer and Groudine 1990). This is manifested, in nuclear run-on transcription assays, by a block to elongation of transcription occurring at specific sites within a gene. The extent of transcriptional block within the c-mye gene has been found to vary with the physiological state of the cell, indicating that e-mye expression1s modulated by this mechanism (Bentley and Groudine 1986; Eick and Bomkamm 1986; Nepveu and Marcu 1986 ; Eick et al. 1987; Nepveu et al. 1987; Watson 1988). In the murine e-mye gene, the block to transcriptional elongation has been mapped to the 3' half of the first exon using various approaches. By nuclear run-on transcription assay, transcriptional block was localized about 100-200 bp upstream of the exon 1- intron 1 junction (Miller et al. 1989; Wright and Bishop 1989). In tlitro transcription of c-mye constructs in whole cell extracts revealed short stable transcripts whose 3' ends mapped within or very close to a stretch of five Ts on the sense DNA strand (Miller et al. 1989). Similar results were obtained in Xenopus oocytes (Bentley and • Groudine 1988). A minor site of in tlitro transcription terrnination was also 83

identified approximately 90 bp further downstream, close to another T-rich sequence (Miller et al. 1989). This site corresponds to the most 5' • termination/pause site (Tl) identified in the human c-mye gene (Bentley and Croudine 1988; Kerppola and Kane 1988). Studies in mammalian cells have indicated that two eis -acting elements are necessary for transcriptional block within the e-mye gene. One element is located in the 3' half of exon 1 and includes the sites of transcriptional block as well as adjacent sequences (Bentley and Croudine 1988; Miller et al. 1989; Wright and Bishop 1989). Another element is located between the two major transcription initiation sites, Pl and P2 (Miller et al. 1989; Spencer et al. 1990). Indeed, transfection studies have shown that deletions within this region effectively abolished transcriptional block (Miller et al. 1989). Further, a correlation was established between promoter usage and transcriptional block: P1-initiated transcripts were shown to be fully elongated, whereas P2-initiated transcripts were either elongated or terrninated at the end of exon 1 (Spencer et al. 1990). This was interpreted to mean that transcription complexes assembled at the P2 but not the Pl promoter were responsive to the signal for transcriptional block. In Xenopus oocytes the importance of promoter sequences was aise suggested from studies in which heterologous promoters were used to transcribe c-mye exon 1 sequences. Thus, with chimeric constructs, transcriptional block was observed when fragments from the 3' half of either the murine or human e-mye exon 1 were linked to some promoters (the herpes simplex virus (HSV) tk and human a-globin gene promoters and the Rous sarcoma virus long terminal repeat), but no transcriptional block was observed with other promoters (the Adenovirus • major late, the SV40 early or the H-2Kb major histocompatibility class gene promoters) (Bentley and Groudine 1988; Miller et al. 1989; Wright and Bishop • 1989). In this paper, we present evidence that control of transcription elongation, like transcription initiation, depends in part on the interaction of nuclear factors with specifie protein binding sites located in the promoter.

Materials and Method

Plasmid construction. AlI constructs originate from the plasmid pSVCAT(ô-EP), which is a derivative of pSV2CAT without the SV40 and promoter (Gorman et al. 1982; Yang et al. 1986). Construct MEC was made by insertion of the Bamffi-BglII fragment containing sequences from c-mye exon 1 into the BglII site of pSVCAT(Ô-EP). A Hinh III fragment containing sequences -1987 to +11 from plasmid pH2CAT (Kimura et al. 1986) was then inserted into the corresponding site. The HindIll site distal to the c-mye sequences was removed by partial digestion, filling the ends with Klenow polymerase and blunt end ligation. Constructs MElal+, ME1al-, mut.1, mut.2 and ME1a1-Sac were obtained by inserting the corresponding double-stranded oligonucleotides into the unique Hindill or SacII sites of plasmid MEC. The oligonucleotides were designed to have Hindill or SadI cohesive ends once annealed. Cell culture and transfection. HeLa and Mouse erythroleukemia (MEL) cells were grown in DMEM medium supplemented with 10% calf serum. Plasmid DNA was introduced • by the calcium phosphate precipitation technique (Wigler et al. 1979). 85

Typically, 0.2ug of pSVzNeo and 10ug of c-myc plasmid were added to 1x106 • cells in a 100mm plate. Transfections were performed in duplicate and 48 hours later cells were split 1:6 in selective medium containing 1mg/ml of G­ 418. Two weeks later, the resistant colonies were pooled. We usually obtain between 500 and 1000 G-418 resistant colonies in HeLa cells. Nuc1ear run-on transcription assays. Cells were scraped in PBS with 1mM EOTA, centrifuged for 5 minutes at 1500r.p.m. and lysed in 2ml of buffer A, adjusted to 0.1% NP40 (Schibler et al. 1983). Lysed cells were centrifuged for 5 minutes at 1500r.p.m. The supernatant was saved for cytoplasmic RNA purification. The nuc1ei were washed once in nuclei storage buffer and resuspended in the same buffer containing 100units/ml of RNasin (Promega) (Schibler et al. 1983). The in vitro elongation reactions were carried out at 26°C for 30 minutes in a cocktail containing 100mM Tris-HCl (pH 7.9), 50mM NaCl, 5mM MgC12, 5mM MnClz, 0.4mM EOTA, 300mM (Nl"4)zS04, O.lmM PMSF, 1.2mM OTT, 1mM each of ATP, CTP and GTP, 150uCi a-3Zp-UTP SOO Ci/mmol, 10mm creatine phosphate, 20 units/ml RNasin (promega), 29% glycerol and about 2x107 nuc1ei. The reaction was stopped by the addition of 50 mg of RNAse-free ONase from BRL and incubated at 37°C until the solution was not vi~cous. An equal volume of 2x proteinase buffer [20 mM Tris-HO, pH7.9, 20 mM EOTA, 1% SOS] was added and proteinase Kwas included at a concentration of 200 mg/ml. After incubation for 1 hr at 42°C, the solution was extracted twice with a mixture of phenol and chIoroform. The solution was divided in two halves and the unincorporated nucleotides were removed by centrifugation through Sephadex G-50 spin columns. The eluates were pooled and mixed with an equal volume of 10% TCA, 60 mM Sodium • Pyrophosphate in a microfuge tube. After 10 min on ice, the tubes were 1'6

centrifuged for 15 min. The RNA pellet was resuspended in 250 ml of 20 mM HEPES (pH7.5), 5 mM EOTA; 63 ml of lM NaOH was added, the tubes were • vortexed and incubated for 13 min on ice. Reactions \Vere terminated by adding 125 ml of lM HEPES (free acid) and 50 ml of 3M Na Acetate. The RNA was ethanol-precipitated and resuspended in 1.5 ml of hybridization buffer [50 mM Hepes pH 7,0.75 M Nad, 50% formamide, 0.5% SOS, 2mM EOTA, 10X Oeinhardt, 200 ug/ml salmon sperm ONA]. Prehybridization was done in sealed bag at 42oC, overnight or for a minimum of 4 hr, with 3 ml of hybridization buffer. Hybridization was performed at 420 C for two days (minimum of 40 hours). Washing was done at 62 oC in 0.1 x SSC, 0.1% SOS with at least 5 changes. RNA isolation. The cytoplasmic fraction obtained after cell lysis (see run-on protocol) was mixed with 4 volumes of guanidine thiocyanate and RNA was prepared by the guanidine-thiocyanate/hot phenol method essentially as described by Feramisco et al. 1982. 51 nuclease protection analysis. 51 nudease protection analysis was performed essentially as described by Yang et al. 1986. DNA probes were end-Iabeled using T4 kinase (Maxam and Gilbert 1980). Indicated amounts of cytoplasmic RNA was annealed to 2x1OSc.p.m. of end-Iabeled probe, at 56°C for 8 hours, in 80% formamide, 0.4M NaCI, 0.4M Pipes (pH6.4), 1mM EDTA. RNA-DNA hybrids were digested with 500 units of 51 nuclease (BRL) at 2SoC for 45 minutes and at 370C for 15 minutes. Electrophoretic mobility shift assays ŒMSA). Nuclear extracts were prepared as described by Shapiro et al. 1988. • EMSA were performed with either 10 ug of MEL or HeLa cell nudear extracts, 87

incubated in 25 mM NaCI, 10 mM Tris pH 7.5, 1 mM MgClz, 5 mM EDTA pH • 8.0, 5% glycerol, and 1 mM DTT, in a final volume of 20 ul, with 1 ug of poly dI-dC and 100 ng of spE'cific competitor, when specified. Samples were incubated at room temperature for 5 min and 20 000 cpm (100 pg) of end­ labeled probe was added and incubated for 15 min. Samples were loaded on a 5% polyacrylamide gel (30:1) and electrophoresed at 8 volts/cm for 3 hours in 50 mM Tris, 0.38 M glycine, and 1 mM EDTA, pH 8.5. Gels were dried and visualized by autoradiography.

Results

A sequence from the c-myc P2 promoter increases transcriptional block. Two protein binding sites, terIIled MElal and MEla2, have been identified by DNase 1 protection assays between the Pl and P2 transcription start sites of the murine e-mye gene (Asselin et al. 1989). Deletion analysis showed that removal of the MElal sequence decreased transcriptional block while reducing initiation of transcription at P2 (Miller et al. 1989). To determine the importance of the MElal protein binding site in the modulation of transcriptional elongation, double-stranded oligonucleotides encoding MElal were inserted in either orientation at the HindIIT site of the plasmid MEC (Fig.l). This plasmid contains the H-2Kb major histocompatibility class (MHC/H-2Kb) gene promoter linked to the e-mye first exon at the HindIII site, downstream of the P2 TATAA box (Fig.l). Therefore, even though the P2 start site is present, sequences necessary to drive initiation of transcription at this site are absent from this construct. Block to elongation of transcription was assayed by nuclear run-on transcription assay, • using populations of stably transfected HeLa cells (Fig.l). Comparison of the 88

hybridization signais obtained with probes A and B showed that there is low • levels of transcriptional block with construct MEC. The two signaIs are of equal intensity, but taking into account the uridine content of the region covered by probes A and B (95 and 156 uridines respectively) we ca1culate that there is still 40 to 45% block to the elongation of transcription with this construct (Table 2). This result therefore confirms that a fragment containing most of the c-myc first exon including the termination region, can confer a low level of transcriptional block. However, a stronger block to transcription elongation is observed when the ME1a1 sequence is inserted close to the transcription initiation site (Fig. 1 and Table 1). This effect is seen whether the ME1a1 sequence is inserted in its original orientation or in the reverse orientation. Thus the ME1a1 sequence can cause an increase in transcriptional block. Double-stranded oligonucleotides encoding mutated ME1a1 sequences were also inserted at the Hindill site of plasmid MEC. Table 2 shows a sequence comparison of the oligonucleotides used in this study. The sequence of ME1a1 has bcen established as being the protected sequence in DNase 1 footprinting assays (Asselin et al. 1989). The exact recognition sequence has not yet been defined. Therefore, in mutant 1 oligonucleotides, several mutations were introduced in the second half of the sequence, whereas in mutant 2 changes were present throughout the sequence. The resulting plasmids were ealled mutant 1 and 2 respectively. When assayed by nuclear run-on transcription analysis, both mutant 1 and 2 templates showed a lesser extent of transcriptional block as compared to the MEC-ME1a1+ construct (Fig. 1 and Table 2). Therefore, we conclude that introduction of point mutations in ME1a1 reduces the extent of transcriptional block, suggesting that specifie • sequences within ME1a1 are important for function. 89

The site of transcription initiation is not changed by the insertion of MElal oligonucleotides. • It has been suggested that initiation of transcription at the P2 site was necessary for transcriptional block within the e-mye gene (Spencer et ai. 1990: Spencer and Groudine 1990). The induction of transcriptional block could then result from a change at the level of transcription initiation. Both the MHC/H-2Kb and c-mye P2 transcription initiation sites are present in the constructs, but sequences from the P2 promoter, including the TATAA box, are lacking. With construct MEC, transcription has been shown to start at the MHC/H-2Kb site oruy (Miller et al. 1989 and Fig. 5). To determine if addition of the MElal sequence affected transcription initiation, we performed 51 nuclease protection assays with an end-Iabeled probe, using cytoplasmic RNA isolated from ceUs transfected with the MEC-MElal+ and MEC-MElal­ constructs (Fig. 2A, B). The protected fragments correspond to transcripts initiated at the MHC/H-2Kb site. No P2-initiated transcripts (175 nt) were detected. These results indicate that, with the MEC-MElal+ and MEC-MElal­ constructs, transcription is predominantly if not exclusively initiated at the MHC/H-2Kb start site. In conclusion, addition of the MElal sequence did not grossly alter the site at which transcription starts. Mutations in the MElal sequence reduce binding in tlitro. To verify whether nuclear factors interacted with the MElal sequence in

tlitro, an electrophoretic mobility shift assay (EMSA) was performed. Using the MElal double-stranded oligonucleotides and nuclear extracts prepared from HeLa cells, a specific protein-DNA complex was detected, which was competed with an excess of specific competitor (Fig. 3A, lanes 3, 4 and 5). Mutant 1 and 2 were used as specific competitors to determine their capacity • to compete for binding. Neither of the mutant MElal sequences competed 90

efficiently for protein binding to MElal (Fig. 3A lanes 6 and 7). Also, when • mutants 1 and 2 were used as probes in the binding assay, they showed lower binding activity compared to MElal (Fig. 3B). Thus, mutations in the MElal sequence reduces the ability of this element to bind nuc1ear factor(s). These results indicate that the MElal sequence is recognized by nuc1ear factor(s) in vitro and that this interaction requires specifie sequences. To determine whether similar factors are also present in mouse cells, a nuclear extract was prepared from mouse erythroleukemia (MEL) cells and used in the binding assay (Fig. 3C). In addition to the specifie protein-DNA complex observed with a HeLa nuclear extract, an additional slower migrating band was detected. We do not know whether the slower migrating band reflects binding to an additional factor or whether two different forms of the same factor are present in murine cens. The MElal sequence functions either upstream or downstream of the transcription initiation site. In the previous constructs, the MElal sequence was inserted at the Hindm site located l4bp downstream of the transcription initiation site. Therefore, our results did not exclude the possibility that the added oligonucleotides affect the secondary structure of the nascent transcripts and that this in turn affects the ability of RNA polymerase il to stop or readthrough when it reaches the signal for transcriptional block. To determine whether it functions as a DNA or an RNA element, we inserted the MElal sequence at the Sacll site of plasmid MEC, 33bp upstream of the transcription initiation site. Insertion of the MElal sequence at this location results in an even greater increase in the level of transcriptional block (Fig. 4 and Table 2). To compare the sites of transcription initiation in constructs • MEC and MEC-Sac, SI nuclease mapping analysis was performed using an 91

end-labeled probe. As illustrated in Fig. 5, transcription initiation is not changed folIowing insertion of the MElal sequence, but the steady-state level • of RNA is greatly reduced. The intensity of the bands are comparable, but it should be noted that 150 mg of poly A+ RNA and 30 mg of total RNA were used for MElal-Sac and MEC respectively.

Discussion

We have previously shown that mutations within the c-rny' promoter decreased transcriptional block in addition to reducing the level of initiation at the P2 start site. Similarly, decreased transcriptional block was observed when the c-rny' promoter region was replaced with the promoter for the H­ 2Kb major histocompatibility class gene. Altogether these results indicated that the decision to stop or read-through at the site of block could be influenced by sequences located several hundred base-pairs upstream. Reduced transcriptional block in the examples cited above could have resulted from one of the following: a shift in promoter usage in favor of Pl, an increase in the distance between the sites of transcriptional initiation and block, a modification of the sequence at the 5' end of the primary transcript or a modification of the DNA sequence around the transcription initiation site. In the present study, we show that a 25 bp DNA sequence from the c-rny' promoter can increase transcriptional block when inserted in the same chimeric construct containing the H-2Kb major histocompatibility class gene and most of the c-rny' first exon. In aU of the constructs tested, the MElal sequence has been placed in close proximity to the transcription initiation • site, either upstream or downstream. Our results confirm that sequences close 92

to the initiation site can affect transcriptional elongation at sites located several hundred nucleotides downstream. • In two of the constructs tested the MElal sequence was inserted downstream of the transcription initiation site. 51 nuclease mapping analysis of the 5' end of the transcripts revealed that insertion of the MElal sequence did not shift transcription initiation to the P2 start site. We can therefore exclude that a change in promoter usage was responsible for the difference in transcription elongation. However the primary sequence of the transcript was changed and the distance between the, sites of transcription initiation and block was increased by 30 nucleotides. The fact that the MElal sequence worked in both orientations downstream of the transcription initiation site, makes it unlikely but does not preclude that it functioned as an RNA sequence in these constructs. However an unequivocal interpretation of its mechanism of action can be derived from the experiment in which the MElal sequence was inserted upstream of the initiation site. That transcriptional block was increased with this construct excludes any effect of distance or RNA sequence and demonstrates that the MElal site functions as a DNA sequence. How DNA sequences close to the site of initiation can affect transcription elongation can be accounted by two sets of models: modification of transcriptional complexes or changes in DNA structure. Either of these models could involve trans-acting factors. Indeed, when mutated oligonucleotides were tested, decreased interaction with nuclear factors in vitro was observed together with a reduced ability to confer transcriptional block in vivo. However a formaI demonstration that trans-acting factors are • implicated in the control of transcription elongation will await the isolation 93

of such factors as weil as the development of a suitable in vitro transcription system. • In the first model the ME1a1 sequence would act as a recognition site for a factor which would modify the transcriptional complex or become part of it (see Fig. 6). This interaction would reduce the ability of the transcriptional complex to read-through terrnination signais located further downstream. This model implies that different classes of transcriptional complexes exist which differ in their elongation properties and that these properties are deterrnined, at least in part, at the promoter level. This is supported by the observation that the extent of transcriptional block varies downstream of different promoters. An ahemative set of models would involve structural changes in DNA resulting from interactions between trans-acting factors bound to promoter and termination elements. Precedents exist for this type of regulatory mechanism. Indeed, DNA looping out mediated by protein-protein interactions has been implicated in long-range transcriptional activation by enhancers (Muller et al. 1989), as weil as transcriptional repression of several bacterial operons (Besse et al. 1986; Mossing et al. 1986; Hochschild 1990). On the other hand, some transcription stop signaIs were recentIy shown to coincide with regions of bent DNA (Kerppola and Kane, 1990). One could therefore envision that binding of a factor to a site in the promoter may favor the occupancy of a downstream site which in turn would modify bending 0: DNA in that region. Interestingly, sequences identical to ME1a1 are not present within the herpes simplex virus (HSV) tk and human a-globin gene promoters, both of which display transcriptional block when linked to the c-mye termination • region (BentIey and Groudine 1988; Wright and Bishop 1989). This suggests 94

that there are other sequences within these promoters which substitute for • ME1a1. Different sequence motifs would therefore be involved in the control of elongation of transcription, allowing for independent regulation of different genes. The involvement of promoter sequences in the decision to terminate or not at specifie termination signaIs has previously been demonstrated for the U1 and U2 small nuclear RNA (snRNA) genes (Hernandez and Weiner 1986; Neuman de Vegvar et al. 1986; Hernandez and Lucito 1988; Lobo and Hernandez 1989). Effect at a distance of cis-acting elements has been most

extensively studied in the case of bacteriophage À. where elongation of transcription has been shown to depend on the presence of termination (Nun) or anti-termination (N and Q) factors which interact with the transcription apparatus at specific sequences (nut and qut) located upstream of the termination sites (Grayhack et al 1985; Barik et al. 1987; Horwitz, et al. 1987; Robert et al. 1987;). Our results indicate that in eucaryotic cells, interactions of nuclear factors with specific sequences in the promoter region can affect elongation of transcription at sites located further downstream. Identification of the protein factor(s) which bind to the ME1a1 sequence will represent a major advance in our understanding of how transcriptional elongation is modulated in eukaryotes.

Acknowledgements

We thank Harvey Miller for technical assistance; Yann Échelard, Morag Park and Rob Campbell for critical reading of this manuscript. D.D. and M.D. are supported respectively by a studentship from the National Cancer Insitute • of Canada and a fellowship from the Natural Sciences and Engineering 95

Research Council of Canada. This research was supported by a grant from the • National Cancer Institute of Canada to A.N. References

Asselin, C., A. Nepveu and K. B. Marcu. 1989. Molecular requirements for transcriptional initiation of the murine c-myc gene. Oncogene 4:549-558.

Barik, 5., B. Ghosh, W. WhaIen, D. Lazinski, and A. Das. 1987. An antitermination protein engages the elongating transcription apparatus at a promoter-proximal recognition site. Cell 50:885-899.

BentIey, D. 1., and M. Groudine. 1986. A block to elongation is largely responsible for decreased transcription of c-myc in differentiated HL60 cells. Nature 321:702-706.

BentIey, D. L., and M. Groudine. 1988. Sequence requirements for premature termination of transcription in the human c-mycgene. Cell 53:245-256.

Besse, M., B. Wilcken-Bergmann and B. Muller-Hill. 1986. Synthetic lac operator mediates repression through lac repressor when introduced upstream and downstream from lac promoter. EMBO J. 5:1377-1381.

Eick, D., and G. W. Bomkamm. 1986. Transcriptional arrest within the first exon is a fast control mechanism in c-myc gene expression. Nucleie Acids Res. • 14:8331-8346. 96

Eick, D., R. Berger, A. Polack and G.W. Bornkamm. 1987. Transcription of l'­ mye in human mononuclear cells is regulated by an elongation block. • Oncogene 2:61-65.

Feramisco, J.R., J.E. Smart, K. Burridge, D.M. Helfman and G.P. Thomas. 1982. Co-existence of vinculin and a vinculin-like protein of higher molecular weight ïn smooth muscle. J. Biol. Chem. 257:11024-11031.

Gorman, C. M., L. F. Moffat, and B. H. Howard. 1982. Recombinant genomes which express Chloramphenicol acetyltransferase in mammalian cells. Mol. Cell. Biol. 2:1044-1051.

Grayhack, E. J., X. Yang, L. F. Lau, and W. Roberts. 1985. Phage lambda gene Q antiterminator recognizes RNA polymerase near the promoter and accelerates it through a pause site. Ce1l42:259-269.

Hemandez, N., and A. M. Weiner. 1986. Formation of the 3' end of U1 snRNA requires compatible snRNA promoter elements. Cel! 47:249-258.

Hemandez, N., and R. Lucito. 1988. Elements required for transcription initiation of the human U2 snRNA gene coïncide with elements required for snRNA 3' end formatio.l. EMBO J. 1:3125-3134.

Horikoshi, M., K. Sekimizu, and S. Natori. 1984. Analysis of the stimulatory factor of RNA polymerase n in the initiation and elongation complex. J. Biol. • Chem. 259:608-611. 97

Horwitz, R.J., J. Li, and J. Greenblatt. 1987. An elongation control particle containing the N gene transcription antitermination protein of bacteriophage • lambda. Cell 51:631-641.

Hoschschild, A. 1990. in DNA topology and its biological effects., eds. Cozzarelli, N. R. and J. C. Wang ( Cold Spring Harbor Lab., Cold Spring Harbor, NY ), pp107-138.

Kerppola, T. K., and C. M. Kane. 1988. Intrinsic sites of transcription termination and pausing in the e-mye gene. Mol. Cell. Biol. 8:4389-4394.

Kerppola, T. K., and C. M. Kane. 1990. Analysis of the signaIs for transcription termination by purified RNA polymerase II. Biochemistry 29: 269-278.

Kimura, A., A. Israel, O. Le Bail, and P. Kourilsky. 1986. Detailed analysis of the Mouse H-2Kb promoter: enhancer-like sequences and their role in the regulation of class 1gene expression. Ce1l44:261-272.

Loba, S.M., and N. Hemandez. 1989. A 7 bp mutation eonverts a human RNA polymerase II snRNA promoter into an RNA polymerase ID promoter. Cell 58:55-67.

Maxam, A., and W. Gilbert. 1980. Sequencing end-Iabelled DNA with base­ specifie chemieaI cleavage. Methods Enzymol. 65:499-580. • 98

Miller, H., C. Asselin, D. Dufort, j.Q. Yang.. K. Gupta, K.B. Marcu, and A. • Nepveu. 1989. Acis-acting element in the promoter region of the murine c­ mye gene is necessary for transcriptional block. Mol. Cell. Biol. 9:5340-5349.

Mossing, M. C., and M. T. Record. 1986. Vpstream operators enhance repression of the lac promoter. Science 233:889-892.

Muller, H. P., S. M. Sogo, and W. Schaffner. 1989. An enhancer stimulates transcrption in trans when attatched to the promoter via a protein bridge. Cell 58:767-777.

Nepveu, A., and K. B. Marcu. 1986. Intragenic pausing and anti-sense transcription within the murine c-myc locus. EMBO J. 5:2859-2865.

Nepveu A., K. B. Marcu, A. 1. Skoultchi, and H. M. Lachman. 1987. Contributions of transcriptional and post-transcriptional mechanisms to the regulatioi", of c-myc expression in mouse erythroleukemia cells. Genes & Dev. 1:938-945.

Neuman de Vegvar, H. E., E. Lund, and J. E. Dahlberg. 1986. 3' end formation of VI snRNA precursors is coupled to transcription from snRNA promoters. Cell 47:259-266.

Piechaczyk, M., J. M. Blanchard, L Marty, C. Dani, F. Panatieres, S. El-Sabouty, P. Fort, and P. Jeanteur. 1984. Post-transcriptional regulation of glyceraldehyde-3-phosphate dehydorgenase gene expression in rat tissues. • Nucleic Acids Res. 12:6951-6963. 99

• Rappaport, J., D. Reinberg, R. Zandomeni, and R. Weinmann. 1987. Purification and functional characterization of transcription factor SI! from calf thymus. J Biol. Chem. 262:5227-5232.

Reinberg, D., and R.B. Raeder. 1987. Factors involved in specific transcription of mammalian RNA polymerase II. Transcription factor Ils stimulates elongation of RNA chains. J. Biol. Chem. 262:3331-3337.

Reines, D., D. Wells, M. Chambertin, and C. M. Kane. 1987. Identification of intrinsic termination sites in vitro for RNA polymerase II within eukaryotic gene sequences. J. Mol. Biol. 196:299-312.

Reines, D., M. Chamberlin, and C. M. Kane. 1989. Transcription elongation factor SIl (TFIIS) enables RNA polymerase to elongate through a block to transcription in a human gene in vitro. J. Biol. Chem. 264:10799-10809.

Robert, J., S. B. Sloan, R. A. Welsberg, M. E. Gottesman, R. Robledo, and D. Harbrecht. 1987. The Remarkable Specificity of a New Transcription Termination Factor Suggests That the Mechanisms of Termination and Antitermination Are Similar. Cell 51:483-492.

Shapiro, D. J., P. A. Sharp, W. W.Wahli, and M. J. Keller. 1988. A high efficiency HeLa nuclear transcription extract. DNA 7: 47 (1988). • 100

Spencer, c., R.C. Lestrange, U. Novak, W.S. Hayward, and M. Groudine. 1990 The block to transcription elongation is promoter dependent in normal and

• Burkitt's lymphoma c-myc alleles.. Genes & Dev. 4:75-88.

Spencer, C. A., and M. Groudine. 1990. Transcription elongation and eukaryotic gene regulation. Oncogene 5: 777-78S.

Schibler, W., O. Hagenbuchle, P. K. Wellauer, and A. C. Pittet. 1983. Two promoters of different strengths control the transciption of the mouse alpha­ amylase gene Amy_la in the parotid gland and the liver. Ce1l33:S01-S08.

Watson, R. J. 1988. Expression of the c-myb and c-myc genes is regulated independently in differentiating mouse erythroleukemia cells by common processes of premature transcription ~rrest and increased mRNA turn-over. Mol. Cel!. Biol. 8:3938-3942.

Wigler M., A. Pellicer, S. Silverstein, R. Axel, G. Urlaub, and L. Chasin. 1979. DNA-mediated transfer of the adenine phosphoribosyltransferase locus into mammalian cells. Proc. Nat!. Acad. Sei. USA 3:1373-1376.

Wright 5., and J. M. Bishop. 1989. DNA sequences that mediate attenuation of transcription from the mouse protooncogene mye. Proc. Nat!. Acad. Sei. USA 86:505-509.

Yang J. Q., E. F. Remmers and K. B. Marcu. 1986. The first exon of the c-mye proto-oncogene contains a novel positive control element. EMBO J. 5:3553­ • 3562. 101 •

Fig. 1. Nuclear run-on transcription analysis of the constructs. The constructs were stably transfected into Hela ceUs. Nuclei were isolated and used in run­ on assays. LabeUed nascent transcripts were hybridized to single-st.-anded M13 DNA probes immobilized on filters. The constructs are described in material and methods. The ME1a1 oligonucleotide was inserted at the Hind II site in either the correct orientation (ME1a1+), or the opposite orientation (ME1a1-). The sequences of mut.1 and 2 are illustrated in table 2. The broken line represents e-mye 5' fIanking sequences; the stripped and open boxes, e-mye exon 1 sequences and the chloramphenicol acetyl-transferase (CAT) gene, respectively. Arrows indicate the Pl (+1) and P2 transcription start sites; T5, a stretch of five Ts, where a site of transcriptional block was mapped. The small black box within exon 1 represents the ME1a1 protein binding site, nucieotides +95 to +120 relative to Pl. Hd, Hind m; Bg, Bgl II. Probe A is a single-stranded M13 DNA containing the BamHI-BglII fragment (Bm-Bg) of c-mye exon 1 (uridine content, 95). Probe B, the Bgl II - Neo I fragment (Bg-Nc) of the CAT gene (uridine content, 156) (Gorman et al. 1982). GAPDH, the PstI fragment of the rat glyceraldehyde-3-phosphate-dehydrogenase cDNA clone (Piechaczyk et al. 1984). The mp10 probe, the M13 vector mp10.

• •

1 2 3 4 5 MEC ME1 a1+ ME1 a1- mut.1 mut.2 A - - - .. .. 8 - - - GAPDH - - - .. mp10

A •• B • P1 Hel P2 T ~ SgCAT ,',., ~Exon1>k CAT ME1a1: 1

MEC 1 MHC-H2K 'Pz!: Exon 1>7El CAT 1 •1 MEC-oligo 1 MHC-H2K ~ Exon 1~ CAT

1 oligo 1 • 102 •

C""',""" MEC MEC-ME/al" MEC-ME/ar _./ _.2 MECoS« Pop. 1 41 IJ IZ '73 4li M ("l. 28, 45) (86. 75. 17) (82, 77. Il) (72. 70. 76) ("l. 34, 53) (96. 93. 93) Pop. 2 4li IZ ." 71 4f (41, 53. 44) (83. 80. 84) (76, 79. 13) (68. 74. 71) (42, S6, 48) <99. "94, 96)

Table L Percentage of transaiptional block. Two independant populations of stably transfected cells were isolated for each construet and three nuclear • run-on transcription assays were performed for each population. The intensity oi the signal for probes A and B have been measured by densitometric scanning. The values have been adjusted to take into account the number of uridine residues present in the corresponding portion of the construet (A, c-myc exon 1: 95 uridines; B, CAT: 156 uridines). The data have been expressed as the percentage of transcriptional block. The values determined for three run-on transaiption assays as well as the average value for each population are shown. • 103 •

Fig. 2. SI nuclease mapping of the transcription start sites utilized by the constructs. Cytoplasmic RNA was purified from populations of cells stably transfected with constructs MEC-MElal+ and MEC-MElal-. Fifty ug of RNA were hybridized to the DNA probes prior to SI nuclease digestion. The protected fragments are schematically represented in the diagrams below. A) Transcription initiation within construct MEC-MElal+. The probe was the Not 1 - Xba 1 ( Nt-Xb) fragment from this construct end-labeled at Not I. B) Transcription initiation within MEC-MElal-. The probe was the Not I-Xba 1 fragment from fuis construct end-labeled at Not 1.

• •

... A B ;;; w'" w Q) ::: Q) ::: ~ « U ~ « U 0 z w 0 Z w li: 5 ::: li: 5 :::

612 ". 612', •

.f .~ ~ ;..

291 291 271 271 ._253 -249 • _249 234 234

210 210

194 194

162 162

,XI> ttH:l ,NI ,XI> ~ ,Nt 1 MHC.H2" ~~~[SEx~o~nc:'a102:i1lH=:3C~AQl=::J 1 MHC.H2' 0{ Exon 1>@ CAl : ME1a,'" : MElal - : • 249 nt 1 )( 249 nt 1 JE 253 nt • 104 •

MElal GGAAAAAGAAGGGAGGGGAGGGA Mut.l GGAAAAAGAAGTGAGTTGATGTA Mut.2 GTACAATGACGTGAGTTGATGTA

Table 2. Sequence comparison of the oligonucleotides

• 105 •

Fig. 3. EM5A using the MElal and mu~ant MElal double-stranded oligonuc1eotides. A) EMSA was performed using end-JabeJed MElal doubJe­ stranded oligonucleotides and nuclear extracts purified from HeLa cells. Lane 1: probe aJone. Lane 2: the probe was allowed to interact with 10 ug of HeLa cell nuclear extract in the absence of any competitor. Lane 3: 1 ug of poJy dI-dC was added as non-specifie competitor. In Janes 4 to 7, specifie competitor was added as follows: Janes 4 and 5, MElal(lOO and 50 ng respectively); lanes 6 and 7, 100 ng of mut.l and 2 respectiveJy. B) EMSA using end-labeled MElal and mutant MElal doubJe-stranded oligonucleotides as probes. C) EM5A using end-labeled MElal double-stranded oligonucleotides and nuclear extracts purified from mouse erythroleukemia (MEL) cells. Lane 1: probe alone. Lane 2: the probe was allowed to interact with 10 ug of MEL cell nuclear extract in the absence of any competitor. Lane 3: 1 ug of poly dI-dC was added as non­ specifie competitor. In lane 4, 100 ng of MElal oligonucleotides was added as specifie competitor.

• • A. 8. C. -III - C\I W- :::> -:::> 1 2 3 4 5 6 7 ~ E E 1 2 3 4 - ••

~ ~ P2 T ~EXOn1~5 +1 ME1a1 +562 • 106 •

Fig. 4 The MElal sequence functions either upstream or downstream of the initiation site. The constructs were stably transfected into Hela cells. Nuclei were isolated and used in run-on assays. Labelled nascent transcripts were hybridized to single-stranded M13 DNA probes immobilized on filters. The map of the constructs and the probes have been described in the legend of Fig.l. Hd, Hind m; Sac, SadI.

• •

1 2 3 A - -. - B- GAPDH - - .- mp10

AB

salk 1. fvEC 1 MHC-H2K 2""W//////A"".,.,.,.,'"""M1r-----

I-ki 2. MEC-Hind 1r-':":M~HC"..-.,..,H2""'K.,....~~W.....W....h....W4i"'------

Sac 3. MEC-Sac 1 MHe-H2K~'----...J • 107 •

Fig. 5 Comparison of the transcription start sites utilized by the MEC and MEC-Sac constructs. Cytoplasmic RNA was purified from populations of cells stably transfected with constructs MEC-Sac and MEC. One hundred and fifty

mg of poly A+ RNA and 30 mg of total RNA from MEC-Sac and MEC respectively were hybridized to the DNA probe prior to 51 nuclease digestion. The probe was the Not 1- Xba 1 (Nt-Xb) fragment from the MEC construct end­ labeled at the Not 1 site. The probe and the protected fragment are schematically represented in the diagram below.

• •

al rn" CD oC Ù U 0 ;2 ~ w w CL 5 :::0 :::0 :::0 - - 603

- 310 ~ _ 281271 - 234 _. +215 - 194

rtHC;Ht~::~Je~~:iii~J'!:lYAtH==CA~t=::::J , 1 • Probe •...... 215 nt • IDS •

Fig. 6 Promotion of tennination at a distant site. Readthrough: transcription complexes assembled on many promoters would not recognize the signal for transcriptional bloc!< and would elongate beyond tIùs site. Termination: upon binding to the MElal sequence in the promoter region, a nuclear factor could modify or become part of the transcription complex, perhaps preventing the association of RNA polymerase II with an elongation factor like TFIIS. Following this interaction, the polymerase would adopt a conformation allowing recognition of the signal for transcriptional block. Alternatively, looping out of the DNA would permit interaction between factors linked to the promoter and the site of transcriptional bloc!<.

• • 1 ) Readthrough r;NA~~ ------\:~-=--..-:::~:....-_--~I-----

2)

~âÛf---- ,,,9;~ B --, .. ~f-----

ME1a1

c

• 109 • CHAYfERIII The Human Cut Homeodomain Protein Represses Transcription from the C-Myc Promoter

Preface

In the previous chapter, the MElal protein binding site was shown to be required for block to transcription elongation. Also mutational analysis demonstrated that factors binding to this site were important for block. To understand the mechanism of transcriptional block, the cellular factors which bind to this site were characterized. In this chapter, fractionation studies demonstrate that three cellular factors bind to the MElal site resulting in the formation of three protein-DNA complexes termed complex "a", "b", and "c". By screening a cDNA library, the cDNA encoding the cellular factor

responsible for complex "c" was isolated. This cDNA was shown to encode the human homologue of the Drosophila Cut homeodomain protein (hu-Cut).

Hu-Cut was shown to repress expression from the c-myc promoter and that MElal as well as exon 1 sequences were required for repression.

• 110 • Abstract 5tudies of the c-mye promoter have shown that efficient transcription initiation at the P2 start site as weil as block to elongation of transcription require the presence of the MElal protein binding site upstream of the P2 TATA box. Following fractionation by size exclusion chromatography, three protein-MElal DNA complexes were detected by electrophoretic mobility shift-assay (EM5A): complexes "a", "b" and "c". A cDNA encoding a protein present in complex "c" was isolated by screening of an expression library with an MElal DNA probe. This cDNA was found to encode the human homologue of the Drosophila Cut homeodomain protein. The bacterially expressed human Cut (hu-Cut) protein bound to the MElal site and antibodies against hu-Cut inhibited the MElal binding activity "c" in nudear extracts. In co-transfection experiments, the human Cut protein repressed transcription from the c-mye promoter and this repression was shown to be dependent on the presence of the MElal site. Using a reporter construct with a heterologous promoter, we found that c-mye exon 1 sequences were also necessary, in addition to the MElal site, for repression by Cut. Taken together these results suggest that the human homologue of the Drosophila Cut homeodomain protein is involved in the regulation of the c-mye gene.

• 111

Introduction

• The control of c-mye expression has been shown to be critical for both cellular proliferation and differentiation (reviewed in: Cole, 1986; Luscher and Eisenman, 1990; Spencer and Groudine, 1991; Marcu et al., 1992; Evans and Littlewood, 1993). The c-mye gene belongs to the family of immediate early response genes whose expression is activated by a variety of nùtogenic stimuli in quiescent fibroblasts and lymphocytes (Almendral et al., 1988). Inhibition of c-mye expression in proliferating cells leads to growth arrest (Waters et al., 1991; Dean et al., 1986; Coffey et al., 1988; Loke et al. 1988) suggesting that c-mye is required for continuous cell proliferation. Induction of c-mye expression in quiescent cells was also found to be essential for the transition from GO to Gl. However, in contrast to other immediate early response genes, c-mye expression is not restricted to this brief GO -io Gl transition period but rather remains constant through the cell cycle in proliferating cells (Hann et al., 1985; Rabbitts et al., 1985; Thompson et al., 1985). Upon induction of cellular differentiation in a variety of cell lines, c-mye expression was rapidly down­ regulated (Lachman and Skoultchi, 1984; Westin et al., 1982; Reitsma et al., 1983; Einat et al., 1985; Dotto et al., 1886; St-Arnaud et al., 1988). The importance of c-mye down-regulation during differentiation was further evidenced from the effect of introducing sense or antisense c-mye sequences into cells. The presence of antisense RNA or DNA, which reduced c-mye RNA levels, triggered the differentiation of HL60 and MEL cells, whereas constitutive c-mye expression from a transfected plasnùd blocked cellular differentiation (Yokoyama and Imamoto, 1987; Holt et al., 1988; Lachman et al., 1986; Prochownick et al., 1988; Dnùtrovsky et al., 1986; Prochownick and • Kukowska, 1986; Coppola et al., 1989). 112

Transcription of the c-mye gene is initiated from two major start sites termed Pl and P2, which are located 164 base pairs apart in the mouse. P2 is • the predominant initiation site, giving rise to 75-90% of c-ml/e mRNAs (Spencer and Groudine, 1991; Marcu et al., 1993). Expression of c-mye is controlled by intricate regulatory mecharùsms. Several positive and negative cis-acting elements have been identified both upstream and downstream of the transcription irùtiation sites and presumably these cis-acting elements coordinate c-mye expression in response to external stimuli. Regulation has been shown to occur at multiple levels including both transcription initiation and elongation, mRNA stability, translation, and protein stability (Spencer and Groudine, 1991). Several eis-acting regulatory elements have been identified between the Pl and P2 initiation sites. The ME1a1 protein binding site was defined by DNase footprinting analysis as a 25 bp region situated just upstream of the P2 TATA box (Asselin et al., 1989). From deletion analysis, the ME1a1 site was shown to be important for efficient irùtiation at the P2 start site (Miller et al., 1989; Bassone et al., 1992; Asselin et al., 1989). Deletion or mutation of this site strongly decreased the lever of P2 initiated transcripts (Miller et al., 1989; Bassone et al., 1992). Aside from its effect on transcription irùtiation, the ME1a1 site was also shown to be involved in the regulation of transcription elongation. Thus, increased readthrough transcription was observed following deletion of the ME1a1 site from the c-myc promoter (Miller et al., 1989). In addition, using a construct in which c-myc exon 1 sequences were placed downstream of a heterologous promoter, we showed that transcription elongation was greatly reduced following insertion of the ME1a1 site upstream of the TATA box (Dufort et al., 1993). Moreover, mutated versions • of the ME1a1 sites were employed to establish a correlation between in vitro 113

binding to cellular factors and the ability to confer block to transcription elongation in vivo, strongly implicating sequence-specifie transcription • factors in the control of transcription elongation (Dufort et al., 1993). In light of the importance of the MElal site in the regulation of both transcription initiation and elongation, we set out to characterize the cellular factors that interact with this site. In this report, we demonstrate that three cellular factors bind to MElal. By screerung a cDNA expression library with an MElal DNA probe, we isolated a cDNA encoding the human homologue of the Drosophila Cut protein. We show that the human Cut (hu-Cut) protein binds to the MElal binding site and represses transcription from the c-mye promoter.

MATERIALS AND METHODS

Preparation of nuclear extracts and size exclusion chromatography. HeLa cell nuclear extracts were prepared according to the procedure of Dignam et al. (1983). Extracts were fractionated on a Waters Protein Pak Glass 300 SW size exclusion column, using a Waters 625 HPLC. Usually, 1 mg of crude nuclear extract was applied to the column, run at a flow rate of 0.5 ml/min. and 0.5 ml fractions were collected.

Àgtll library screening. The human placenta Àgtlllibrary was a generous gift from Dr. Morag Park. Screemng was performed basically as described by Singh et al. (1988). Briefly, 2 X 1()4 pEu were plated on 150 mm plates and incubated at 370C for 3 hours. The plates were overlaid with IPTG-saturated nitrocellulose filters and incubated for 4 hours. The filters were then removed and a second­ • IPTG saturated nitrocellulose filter was overlaid on the plate and incubated 11-1

for another 4 hours. The filters were then transferred to BLOTTO (50 mM • Tris, pH 7.5, 50 mM NaCl, 1 mM EDTA, 1 mM DTT and 5% powdered milk) and incubated for 1 hr. They were transferred to binding buffer (25 mM NaCL

la mM Tris pH 7.5, 1 mM MgCI2 , 5 mM EDTA, pH 8.0, and 1 mM DTT) and incubated for 12 hours. The filters we:e then hybridized for 4 hours with 2 X 106 cpm/ml of concatemerized copies of the MElal binding site (Dufort et al., 1993) and 5llg/rnl of poly(dI-dC)-poly(dI-dC) (Pharmacia). They were washed 4 times for 8 min. with binding buffer, patted dry and exposed for 12-2-1 hours on Kodak X-AR film (Kodak) with intensifying screens at -70°C. Using this approach, one cDNA, called 76-1 was isolated. It contained sequences from nudeotide 3061 to 5374 according to the numbering of CDP cDNA (Neufeld et al., 1992). To obtain a longer cDNA, the library was re­ screened using cDNA 76.1 as a probe. A second cDNA, called 76.2 was isolated which is composed of sequences from nudeotide 1605 to 5374. This cDNA contained the coding regions for aU three Cut repeats as well as the homeodomain.

Plasmid construction. Ail mye constructs have been described previously: pMBgCAT (Yang et al., 1986),.M6 (Asselin et al., 1989), MEC and MEC-ME1a1 (Dufort et al., 1993). Constructs GaI4-E1b-CAT and Ga14-VP16 have been described by LUlie and Green (1989). The pSG-Cut expression vector was generated by doning cDNA 76-2 in the pSG424 vector digested with Bgrn and Bam HI which removed the Gal4 DNA binding domain. As a control, the BglII-Bqm HI-digested pSG424 vector was re-ligated, generating the pSG vector. The ME1a1-CAT construrt was made by cloning a MElal oligonudeotide into the Xba l site of the Gal4-E1b-CAT construct. Six ME1a1 • sites were introduced in this construct. ME1a1-E1-CAT was constructed by 115

cloning the blunt-ended Bam HI-BgI II fragment of exon 1 in the Sma l site of the MElal-CAT construct. The hu-Cut bacterial expression vector GST-Cut • was constructed by cloning cDNA 76-1 into the Eco RI site of the bacterial expression vector pGEX-2T (Pharmacia).

Bacterial expression and purification of recombinant protein. DH5 bacterial ceUs containing the GST-Cut expression vector were grown to an optical density of 0.5, and IPTG was added to a final concentration of 1 mM. The cells were grown for 90 min. and harvested by centrifugation. Bacterial cell lysates were prepared by sonication, and recombinant protein was purified with Glutathione Sepharose 4B beads (pharmacia) which were extensively washed with PBS. The recombinant protein was eluted with 25 mM Glutathione, pH 8.0. A Bradford assay (Bio Rad) was performed to determine the protein concentration.

Generation of hu-Cut antibodies. To generate polyclonal antibodies against hu-Cut, rabbits were injected with 500 Ilg of purified bacterial fusion protein in Freund's complete adjuvant. The animais were boosted twice with 250 Ilg of protein, and serum was collected la days after the last boost. The polyclonal antibodies were purified by affinity chromatography. Because antibodies directed against GST may be produced, the serum was passed through two GST affinity columns to remove them. The flow-through depleted of GST antibodies was then applied to a GST-Cut affinity col= to isolate antibodies against hu-Cut. Monoclonal antibodies were generated by Molecular Immunogenetics (San Andreas, CA), using the bacterially expressed fusion • protein. They were purified on a Protein A affinity col= (pharmacia). 116

Electrophoretic mobility shift assays (EMSA). EMSA were performed with

either 50 ng of purified bacterially-expressed fusion protein or 3 ~ 1 of • fractionated HeLa cell nuclear extracts. To deplete the extracts of hu-Cut, 50 ng of purified polyclonal antibodies, 150 ng of monoclonal antibodies, or 100 ng

of purified pre-immune antibodies were added to 30 ~I of fractionated extracts and incubated at room temperature for 45 min. Protein G-Sepharose (Pharmacia) was then included, incubated for 30 min., and the immune complex was removed by centrifugation. Binding reactions were performed in

25 mM NaCl, 10 mM Tris, pH 7.5,1 mM MgCl2 , 5 mM EDTA, pH 8.0, 5% glycerol, and 1 mM DIT, in a final volume of 20 ul, with 1 ~g of poly (dl-de) and 100 ng of specifie competitor, where specified. Samples were incubated at room temperature for 5 min., and 20 000 cpm (100 pg) of end-labeled probe were added and incubated for 15 min. The samples were then loaded on a 5% polyacrylamide gel (30:1) and electrophoresed at 8 volts/cm for 3 hours in 50 mM Tris, 0.38 M glycine, and 1 mM EDTA, pH 8.5. The Gel was dried and visualized by autoradiography.

CAT assays. 293 cells were grown in DMEM medium supplemented with 10% calf serum. Plasmid DNA was introduced by the calcium phosphate precipitation technique (Wigler et al., 1979). Typically, 5 llg of reporter construct and 6 llg of pSG or pSG-Cut werè added to 3x1OS cells in 100 mm plates. Where specified, 5 llg of Ga14-VP16 expression vector was transfected along with 5 llg of either pSG or pSG-Cut for a total of 10 llg of expression vectors. CMV-~-ga1 (5 llg) was added to aIl transfections as a control for transfection efficiency. The cells were harvested 48 hours after transfection. • Extracts were prepared and CAT assays performed as described previously 117

(Gorman et al., 1982). The CAT assays were quantitated on a Phosphoimager • (Fuji). Results

Multiple cellular factors bind to MElal The MElal binding site has been shown to play an important role in the regulation of both transcription initiation and elongation within the c­ mye gene. As a first step towards the purification of MElal binding proteins, a HeLa crude nuclear extract was fractionated on an HPLC size exclusion column and the fractions were assayed by EM5A for binding to the MElal DNA sequence. As a probe, we used double-stranded oligonucleotides encoding the 25 bp region previously shown to be protected in DNase footprinting assay (Asselin et al. 1989) Three protein-DNA complexes, termed "a", "b" and c were detected upon fractionation of the crude extract (Fig. 1). Complex "a" corresponds to the unique protein-DNA complex previously detected in crude HeLa nuclear extracts (Dufort et al., 1993). The factor responsible for complex "a" eluted in a broad peak and was detected in fractions 2 to 7. Complex "b" was present in fractions 6 to 8, whereas complex

"c" was detected in fractions 2 and 3, corresponding to molecular weight of 75­ 85 kDa and 180-200 kDa respectively. Thus, in fractionated extracts at least three proteins were found to form a complex with the ME1a1 binding site, suggesting that potentially three factors can bind to this site in the cell.

A cDNA encoding the human homologue of the Drosophila Cut homeodomain protein is isolated by screening an expression library with the • MElalsite 118

To isolate cDNAs encoding proteins that bind to the MElal site, a human placenta I..gt11 cDNA expression library was screened with • multimerized copies of the ME1a1 site. From one million plaques screened and after three rounds of purification, one positive clone was isolated. DNA sequencing of the cDNA and a search of GenBank DNA sequences demonstrated that the isolated cDNA was 100% homologous to the cDNA encoding the CCAAT displacement protein (CDP) (Neufeld et al., 1992), the human homologue of the Drosopltila Cut homeodomain protein. The human and Drosophila Cut proteins exhibit a similar structural organization with a Cut-type homeodomain as weIl as three 72 amine acid regions called Cut repeats (Fig. 2A).

A bacterially expressed GST/hu-Cut fusion protein binds to the MElal site To confirm that it binds to the ME1a1 site, a portion of the hu-Cut protein was expressed in bacteria as a glutathione-s-transferase (GST) fusion protein. The recombinant protei:t, schematically ilIustrated in Figure 2A, contained Cut repeat 3, the homeodomain and the carboxy terminus fused to GST. EMSA were performed using purified GST/Cut fusion proteins and labeled MElal oligonudeotides as a probe. One protein-DNA complex was detected which was not competed with a lOOo-fold excess of poly (dl-de) but was cOD'lplctely eliminated by a 10o-foid excess of cold MElal (Fig. 2B, lanes 2­ 4). Therefore a bacterially expressed hu-Cut protein can specifically bind to the MElal DNA sequence.

Antibodies against hu-Cut prevent the formation of complex "c" To test whether hu-Cut was one of the MElal binding proteins detected • in fractionated nudear extracts, polydonal and monoclonal antibodies were 119

generated against the bacterially expressed fusion protein. In western blots with HeLa crude nuclear extracts, both purified antibodies specifically • :ecognized a protein of the expected 200 kDa molecular weight (data not shown). When EMSA were performed with fraction 2 from size exclusion chromatography, protein-DNA complexes "a" and "c" were detected (Fig. 2C, lane 2). Prior incubation of fraction 2 with either monoclonal or polyclonal anti hu-Cut antibodies specifically eliminated comF!zx "c" (Fig.2C, lanes 3 and 5). This was not due to a general inhibitory effect, since the formation of complex "a" was not reduced in the presence of anti hu-Cut antibodies. The addition of pre-immune serum had no effect on either complex "a" or complex c (Fig. 2C, lane 4). We conelude that the hu-Cut protein is involved in the formation of complex "c" in fractionated nuclear extracts.

The hu-Cut protein represses transcription from the c-mye promoter in vivo To investigate the effect of the hu-Cut protein on c-mye expression in vivo, transient co-transfection experiments were performed in 293 cells using the vector pSGCut as an effector plasmid (Fig. 3D) and two c-mye/ CAT constructs as reporter constructs. The plasmid pMBgCAT contains c-mye exon 1 sequences plus 1141 bp of 5' flanking sequences. The plasmid 646 derives from pMBgCAT and contains a 16 bp deletion overlapping the ME1a1 site (Fig. 3d). Co-transfection of the hu-Cut expression veetor resulted in a two- to three-fold reduction in CAT activity from the wild type c-mye/CAT construct. Repression was dependent on the ME1a1 site since hu-Cut had no effect on 646 (Fig. 3a and c). However, the èeletion in 646 itself reduced CAT activity. This result confirms the importance of the ME1a1 site for transcription initiation and raises the possibility that hu-Cut had no effect on 646 because of • its low transcription level. Since expression was much lower with 646, the 120

absence of repression by hu-Cut may have resulted from the loI\' transcription level. To overcome this problem, we inserted the MElal site into a • heterologous promoter construct (Fig. 3d) in which the H-2Kb major histocompatibility class gene promoter w.:os linked to the e-mye first exon. Co­ transfection of hu-Cut had no effect on expression of the original MEC plasmid, however a three-fold repression in CAT activity was observed with the MEC-MElal construct (Fig. 3b and c). Taken together, these results indicate that hu-Cut can repress transcription from the c-mye promoter in t'it'o and that the MElal site is required for repression. The reporter constructs tested in Fig.3 ail contained exon 1 sequences between the initiation site and the CAT coding sequences. To test whether the MElal site was sufficient to repress transcription, we inserted 6 MElal sites in the (Gal4)s-Elb CAT construct which contains the minimal Elb promoter plus five binding sites for the Gal 4 DNA binding domain. Transcription was activated following co-transfection of an expression vector expressing GaI4/VP16. hu-Cut had no effect on basal transcription (Fig. 4, lane 1). In addition, hu-Cut did not affect GaI4/VP16-activated transcription (Fig. 4, lane 2 Vs lane 3). We the::! tested whether repression could be restored by the addition of c-mye exon 1 sequences. In the presence of the first exon, hu-Cut repressed CAT activity by 3- to 4-fold (Fig. 4, lanes 5 and 6). These results demonstrate that the MElal site alone is not sufficient for repression and that other sequences within the first exon are also required.

Discussion

The MElal protein binding site is a 25 bp region of the c-mye promoter • that plays an important role in the regulation of c-mye expression. MElal was 121

shown to be necessary for efficient transcription initiation at the P2 start site as weil as for the control of transcription elongation. In this report, fractionation • studies indicated that three different transcription factors could interact with the ME1a1 binding site. This 25 bp region therefore comprises three, possibly overlapping, recognition sites for transcription factors that are likely to compete for binding. By screening a cDNA expression Iibrary with an ME1a1 DNA probe we isolated a cDNA encoding the human homologue of the Drosophila Cut homeodomain protein (Neufeld et al., 1992). Several Iines of evidence suggest that the human Cut homeodomain protein binds to the ME1a1 sequence and is involved in the control of c-mye expression. Firstly, in EMSA, bacterially expressed recombinant human Cut proteins bound to the ME1a1 sequence. Secondly, incubation of nuclear extracts with either monoclonal or polyclonal anti-hu-Cut antibodies resu1ted in the specifie depletion of the ME1a1 binding activity "c". The same antibodies did not inhibit binding activity "a", and serum from non-immunized animais did not affect either binding activities. Thirdly, in transient co-transfection assays, the human Cut protein repressed expression from an intact c-mye promoter and from a heterologous promoter in which one copy of the MElal binding site had been inserted, indicating that repression by hu-Cut was dependent on the presence of the ME1al binding site. Co-transfection experiments using heterologous promoter constructs indicated that, in addition to the ME1a1 site, c-mye exon 1 sequences were necessary for the repression by hu-Cut. We do not currently know the basis for this requirement, however several hypotheses can be envisaged. It is possible that binding of hu-Cut to the MElal site is facilitated by protein­ protein interactions with other sequence-specific factors, possibly even hu-Cut • itself, that interact with other sites within c-mye exon 1. Alternatively, it is possible that hu-Cut repressed expression at the level of transcription elongation when c-mye exon 1 sequences were inserted between the • transcription start site and the CAT coding sequences. Indeed transcription elongation was previously shown to be subject to modulation and the ME1a1 site has been implicated in this modulation. In the present study, we did not obtain any evidence of modulation at the level of transcription elongation (data not shown), however block to elongation of transcription is not detected, at least in our hands, in run-on assays using nuclei prepareet from transiently transfected cells. Testing of this hypothesis will therefore await the establishment of stably transfected cells expressing hu-Cut under the control of an inducible promoter. Recently, two independent groups have isolated cDNAs encoding for zinc-finger proteins that bind to ME1al. These proteins were designated MAZ (MJlk. Associated Z.inc)(Bossone et al., 1992) and ZF87 (Pyrc et al., 1992). It remains to be determined whether these two isolates represent the same zinc­ finger protein and if either one or both of these are responsible for the formation of MElal complexes "a" and "b". ZF87 was doned as a protein that binds to the MEla2 site, upstream of ME1a1, however it can also bind to MElal. The factor responsible for complex "a" also can bind to both sites (Dufort and Nepveu, unpublished observation). Taken together these results would suggest that ZF87 is involved in the formation of complex "a". Several sequences have been demonstrated to serve as binding sites for mammalian Cut proteins. These include the FP sequence (GCTTTTCAGTTGACCAATGATTATTAGCCAATTTCTGATAAAAGAAAA GGAAACCGATTGC), the C3S sequence (AAAAGAAGCTTATCGATACCGT) (Harada et al., 1993) and the ME1a1 sequence • (GGAAAAAGAAGGGAGGGGAGGGATCC) for the human CCAAT 123

displacement protein/hu-Cut protein and the pe2 sub-element • (GATCTGTGAGCTGTGGAATGTAAGGGAGATC) for the canine Clox protein (eut like homeobm). We note that the FP, ME1a1 and C35 sequences shan: an AAAAGAA motif and that an AAGGGAG motif is present both ME1a1 and the pe2 sub-element. Whether these motifs are sufficient for

binding by Cut remains to be determined. In addition, we have recently found that Cut repeats 1 and 3 function as specifie DNA binding domains (Harada et al., 1993). The hu-Cut protein may therefore contain up to three specifie DNA binding domains: Cut repeats 1 and 3 and the Cut homeodomain. Future work will address which of these DNA binding domains is involved in targeting hu-Cut to the c-mye promoter. From studies in Drosophila, Cut was necessary for proper ceU type specification in several tissues including the external sense organs, the Malpighian t-ùbules, muscles, and the tracheal system. (Bodmer et al., 1987; Blochlinger et al., 1990; Liu et al., 1991; Blochlinger et al., 1993). Thus, in various eut mutants these structures did not develop and, when tested, ectopie expression of Cut did not induce ceU differentiation, but changed the ceU type specificity of differentiating ceUs. 5ince no target genes have been identified in Drosophila it is not known whether Cut acts as a positive or negative regulator of gene expression in specifying ceU identity. The recent isolation of two mammalian Cut genes, the CCAAT displacement protein (CDP)/hu-Cut protein and Clox (eut like homeob~, may provide us with sorne clues about the biochemical function of Cut (Neufeld et al., 1992; Andres et al., 1992, this report). CDP has been shown to bind to upstream regulatory sequences of several genes including the cytochrome ~ heavy chain (gp91­ phox) (Skalnik et al., 1991), the human y-globulin (Super-Furga et al., 1988, • 1989) and the sea urchin histone H2B (Barberis et al., 1987). Expression of these genes coincides with down-regulation of cor binding acti\'ity upon cellular differentiation, suggesting that cor may repress their expression in • undifferentiated cells. Clox has been shown, in co-transfection experiments. to repress transcription from a reporter construct harboring specific recognition sequences. In the present study we demonstrated that hu-Cut could reprcss expression from the c-mye promoter. Taken together, these results suggcst that Cut acts as a negative regulator of gene transcription. We cannot however exclude the possibility that Cut also acts as a positive regulator of gene expression in certain situations. The study of negative transcriptional regulators in recent years has uncovered multiple mechanisms by which repression can be exerted. Some repressors act by interacting with specific ONA binding sites overlapping those of transcriptional activators, thereby competing for ONA site occupancy and preventing the action of activators on the targeted promoters. Other repressors interact directly with an activator, thereby inhibiting its function, a mechanism referred to as "quenching". Alternatively, binding of the repressor may directly repress transcription, a mechanism termed active repression. Binding of COP to several regulator sequences has been shown to prevent the binding of CF1, a transcriptional activator. Thus, it has been proposed that COP/hu-Cut represses gene expression as a result of competitive binding (Neufeld et al., 1992; Andres et aL, 1992). In co­ transfection experiments with the c-mye promoter, it is conceivable that hu­ Cut also functions by preventing the interaction of a transactivator with the ME1a1 site. Indeed the low level of expression observed with a construct that has a 16 bp deletion within the ME1a1 sequence, suggests that ME1a1 is also a binding site for a transcriptional trans-activator (mutation 646: Miller et aL, • 1989; this study). On the other hand, we also observed repression by Cut when the ME1a1 site was inserted within the H-2Kb MHC promoter and the Ga14­ E1b minimal promoter. The fact that repression was effected when the ME1a1 • binding site was moved to two different heterologous promoters argues against competition with an activator as the mechanism for repression, but

instead suggests that Cut acts as an "active repressor". In accordance with this hypothesis, the carboxy-terminal region of hu-Cut contains an alanine-rich domain. Similar alanine-rich sequences have been identified within the repressor domain of the kruppel, even-skipped and homeodomain proteins (Licht et al., 1990, Han and Manley, 1993). Future experiments will determine whether the carboxy-terminal region of hu-Cut is involved in repression.

Acknowledgments

We thank Ginette Bérubé and Jean Paquette for excellent technical assistance, as weIl as Dr. Morag Park for the cONA library as weIl as for helpful advise. 0.0. is supported by a studentship from the National Cancer Institute of Canada. This research was supported by grant #3497 from the National Cancer Institute of Canada to A.N.

References

Almendral, J.M., D. Sommer, H. MacDonald-Bravo, J. Burckhardt, J. Perera, and R. Bravo. 1988. Complexity of the early genetic response to growth factors in mouse fibroblasts. Mol. Cell. Biol. 8:2140-2148. • 126

Andres, V., B. Nada1-Ginard, and V. Mahdavi. 1992. Clox, a mammalian

gene related to Drosop.llila CIl t, encodes DNA-binding regulatory • protein differential!y expressed during development. Development 116:321­ 334.

Asselin, C., A. Nepveu, and K.B. Marcu. 1989. Molecular requirements for transcriptional initiation of the murine c-mye gene. Oncogene. 4:549-558.

Barberis, A., G. Superti-Furga, and M. Busslinger. 1987. Mutual!y exclusive interaction of the CCAAT-binding factor and of a displacement protein with overlapping sequences of a histone gene promoter, Cel! 50:347-359.

Blochlinger, K., R. Bodmer, J.W. Jack, L.Y. Jan, and Y.N. Jan. 1990. Patterns of expression of Cut, a protein required for external sensory organ development in wild-type and eut nutant Drodophila embryos. Genes Dev. 4:1322-1331.

Blochlinger, K., L.Y. Jan, and Y.N. Jan. 1993. Postembryonic patterns of expression of eut, a locus regulating sensory organ identity in Drosophila. Development 117:441-450.

Bodmer, R., S. Barbel, S. Sheperd, J.W. Jack, L.Y. Jan, and Y.N. Jan. 1987. Transformation of sensory organs by mutations of the eut locus of D. melanogaster. Cell 51:293-307.

Bossone, S.A., C. Asselin, A.J. Patel, and K.B. Marcu. 1992. MAZ, a protein, binds to c-MYC and C2 gene sequences regulating transcription • initiation and termination. Proc. Natl. Acad. Sei. USA 89:7452-7456. 127

Coffey, Jr. J., C. C. Bascorn, N. J. Sipes, R. Graves-Deal, B. E. Weissrnan, and • H.L. Moses. 1988. Selective inhibition of growth-related gene expression in murine keratinocytes by transforming growth factor /3. Mol. Cell. Biol. 1988 8:3088-3093.

Cole, M.D. 1986. The mye oncogene: its role in transformation and differentiation. Annu. Rev. Genet. 20:361-384.

Coppola, J.A., J.M. Parker, G.D. Schuler, and M.D. Cole. 1989. Continued withdrawal from the cell cycle and regulation of cellular genes in mouse erythrolukemia cells blocked in differentiation by the c-mye Oncogene. Mol. Cell. Biol. 9:1714-1720.

Dean, M., R. A. Levine, W. Ran, M. S. Kindy, G.E. Sonenshein, and J. Campisi. 1986. Regulation of c-mye transcription and mRNA abundance by serum growth factors and cell contact. J. Biol. Chem. 261:9161-9166.

Dignam, J.O., P.L. Martin, B.S. Shastry, and R.G Roeder. 1983. Eukaryotic gene transcription with purified components. Methods Enzymol. 101:582-598.

Dmitrovsky, E., W.M. Kuehl, G.F. Hollis, I.R. Kirsh, T.P. Bender, and S. Segal. 1986. Expression of a transfected human c-mye oncogene inhibits differentiation of a mouse erythrolukaemia cellline. Nature 322:748-750. • 128

Dotto, G.P., M.Z. Gilman, M. Maryama, and R. A. Weinberg. 1986. c-myc and c-fos expression in differentiating mouse primary keratinocytes. EMBO. J. • 5:2853-2857.

Dufort, O., M. Drolet, and A. Nepveu. 1993. A protein biùding site from the murine c-rnyc promoter contributes to transcriptional block. Oncogene 8:165­ 171.

Einat, M., D. Resnitzky, and A. Kimchi. 1985. Close link between reduction of

c-rnyc expression by interferon and Go 1(;.. arrest. Nature 313:597-600.

Evan, G.I., and T.D. Littlewood. 1993. The role of c-rnyc in cell growth. Curr. Opin. Genet. Dev. 3:44-49.

Gorman. C.M., L.F. Moffat, and B.H. Howard. 1982. Recombinant genomes which express Chloramphenicol acetyltransferase in mammalian cells. Mol.

Cell. Biol. 2:1044-1051.

Han, K., and J.L. Manley. 1993. Functional domains of the Drosophila engrailed protein. EMBO J. U:2723-2733.

Hann, S.R., C. B. Thompson, and R.N. Eisenman. 1985. C-rnyc oncogene protein synthesis is independant of the cell cycle in human and avian cells.

Nature 314:366-369. • 129

Harada, R., D. Dufort, C. Denis-Larose, and A. Nepveu. 1993. Conserved Cut repeats in the human Cut homeodomain protein function as DNA binding • domains. J. Biol. Chem. in press.

Holt, J. T., R. L. Redner, and A.W. Neinhuis. 1988. An oligomer complementary to c-mye mRNA inhibits proliferation of HL-60 promyelocytic cells and induces differentiation. Mol. Cel!. Bio!. 8:963-973.

Lachman, H.M., and A. 1. SkouItchi. 1984. Expression of c-mye changes during differentiation of mouse erythrolukaemia cens. Nature 310:592-594.

Lachman, H.M., G. Cheng, and A.I. SkouItchi. 1986. Transfection of mouse erythrolukaemia ceUs with mye sequences changes the rate of induced commitment to differentiate. Proc. Nat!. Acad. Sei. USA 83:6480-6484.

Licht, J.O., M.J. Grossel, J. Figge, and U.M. Hanson. 1990. Drosophila krüppel protein is a transcriptional repressor. Nature 346:76-79

Lillie, J.W., and M.R. Green. 1989. Transcription activation by the adenovirus Ela protein. Nature 338:39-44.

Liu, S., E. McLeod, and J. Jack. 1991. Four distinct reguIatory regions of the eut locus and their effect on cen type specification in Drosophila. Genetics U7:151­ 159.

Loke, S.L., C. Stein, X. Zhang, M. Avigan, J. Cohen, and L.M. Neckers. 1988. • Delivery of c-mye antisense phosphorothioate oligodeoxnucleotides to 130

hemapoietic cells in culture by liposome fusion: specific reduction in c-myc protein expression correlates with inhibition of cell growth and DNA • synthesis. Curr. Top. Microbio!. Immuno!. 141:282-289.

Lüscher, B., and R.N. Eisenman. 1990. New light on Myc and Myb. Part 1. Mye. Genes Dey. 4:2025-2035.

Marcu, K.B., S.A. Bossone, and A.J. Patel. 1992. mye function and regulation. Annu. Rey. Biochem. 61:809-860.

Miller, H., C. Asselin, D. Dufort, J.Q. Yang. K. Gupta, K.B. Marcu, and A. Nepveu. 1989. A eis-acting element in the promoter region of the murine c­ mye gene is necessary for transcriptional block. Mol. Cell. Biol. 9:5340-5349.

Neufeld, E.J., D.G. Skalnick, P.M.-J. Lievens, and S.H. Orkin. 1992. Human CCAAT displacement proteir.. is homologous to the Drosoplzila homeodomain, eut Nature Genetics 1:50-55.

Prochownick, E.V., and J. Kukowska. 1986. Deregulated expression of c-mye by murine erythrolukemia cells preyents differentiation. Nature 322:848-850.

Prochownick, E.V., J. Kukowska, and C. Rodgers. 1988. Antisense transcripts accelerate differentiation and inhibit G1 progression in murine erythrolukemia cells. Mol. Cell. Biol. 8:3683-3695. • 131

Pyrc, J.J., K.H. Moberg, and D.J. Hall. 1992. Isolation of a novel cDNA Encoding a zinc-finger protein that binds to two sites within the c-mye promoter. • Biochenistry 31:4102-4110.

Rabbitts, P.H., J.V. Watson, A. Lamond, A. Forster, M. A. Stinson, G. Evan, W. Fischer, E. Atherton, R. Sheppard, and T.H. Rabbitts. 1985. Metabolism of c­ mye gene products: c-mye mRNA and protein expression in the cell cycle. EMBO. J. 4:2009-2015.

Reitsma, P.H., P.G. Rothberg, S.M. Astrin, J. Trail, Z. Bar-Shavit, A. Hall, S.L. Teitelbaum, and A.J. Kahn. 1983. Regulation of c-mye gene expression in HL­ 60 leukaemia cells by a vitamin 0 metabolite. Nature 306: 492-494.

Singh, H., J.H. LeBowitz, A.S. Baldwin, and P.A. Sharp. 1988. Molecular c10ning of an enhancer binding protein: Isolation by screening of an expression library with a recognition site DNA. Cell 52:415-423.

Skalnik, D.G., E.c. Strauss, and S.H. Orkin. 1991. CCAAT displacement protein as a repressor of the myelomonocytic-specific gp91-phox gene promoter. J. Biol. Chem. 266:16736-16744.

Spencer, C. A., and M. Groudine. 1991. Control of c-mye regulation in normal and neoplastic cells. Adv. Cancer Res. 56:1-48.

St-Arnaud, R., A. Nepveu, K. B. Marcu, and M.W. McBurney. 1988. Two transient increases in c-mye gene expression during neuroectodermal • differentiation of mouse embryonal carcinoma cells. Oncogene 3:553-559. 132

Superti-Furga, G., A. Barberis, G. Schaffner, and M. Busslinger. 1988. The -Il • mutation in greek HPFH affects the binding of the three nuclear factors to the CCAAT region of the y-globulin gene. EMBO J. 7:3099-3107.

Superti-Furga, G., A. Barberis, E. Schreiber, and M. Buss1inger. 1989. The protein CDP, but not CPl, Footprints on the CCAAT region of the y-globulin

gene in unfractionated B-ceU extracts. Biochem. Biophys. Acta. 1007:237-~42.

Thompson, C.B., P. B. Challoner, P. E. Neiman, and M. Groudine. 1985. Leveles of c-mye oncogene mRNA are invariant throughout the ceU cycle. Nature 314:363-366

Waters, C., T. Littlewood, D. Hancock, J. Moore, and G. Evan. 1991. c-mye protein expressionin untransformed fibroblasts. Oncogene 6:101-109.

Westin, E.H., F. Wong-Staal, E.P. Gelmann, R. Dalla-Favera, T.S. Papas, J.A. Lautenberger, A. Eva, E.P. Reddy, S.R. Tronick, S.A. Aaronson, and R.C. Gallo. 1982. Expression of cellular homologues of retroviral one genes in human hematopoietic cells. Proc. Nat\. Acad. Sei. USA 79:2490-2494.

Wigler, M., A. Pellicer, S. Silverstein, R. Axel, G. Urlaub, and 1. Chasin. 1979. DNA-Mediated transfer of the adenine phosphoribosyltransferase locus into mammalian cells. Proc. Nat!. Acad. Sei. USA 3:1373-1376, • 133

Yang, J.Q., E.F. Remmers, and K.B. Malcu. 1986. The first exon of the c-mye proto-oncogene· contains a novel positive control element. EMBO J. 5:3553­ • 3562.

Yokoyama, K., and F. Imamoto. 1987. Transcriptional control of the endogenous Mye protooncogene by antisense RNA. Proc. Nat!. Acad. Sci. USA 84:7363-7367.

• •

Fig. 1 Fractionation of HeLa nuclear extract by size exclusion chromatography. HeLa nuclear extract was fractionated on a Protein Pak Glass 300 SW size exclusion column and 0.5 ml fractions were collected. EMSA were performed using the MElal oligonucleotide (Dufort et al., 1993). The void volume was not collected, therefore fraction 1 corresponds to the first proteins eluting from the column. Fraction numbers are indicated above each lane. The three protein-DNA complexes, a, b, and c are indicated. The relative elution of several molecular weight markers are indicated below.

• •

al .c o IL 1 2 3 4 5 6 7 8 9 10 11 12 c- b- a-

1 1 1 1 M.W. (kDa) 200 150 66 29

• 135 •

Fig. 2 Binding of hu-Cut to MElal. A. Schematic representation of the human Cut protein as weil as the GsT fusion protein expressed in bacteria. The Cut repeats are depicted as hatched boxes and the homeodomain as a black box. B. EMsA using purified bacterially expressed fusion protein and the MEla1 binding site. Lane 1: probe alone. Lane 2: 50 ng of fusion protein. Lane 3: 50 ng of fusion protein plus 1 llg of poly (dI-dC). Lane 4: 50 ng of fusion protein plus 100 ng of cold MElal oligonucieotides. C. EM5A using fraction 2 of the size exclusion chromatography depleted of hu-Cut with monoclonal and polyclonal antibodies. Lane 1: probe alone. Lane 2: No antibodies. Lane 3: Incubated with purified monoclonal antibodies. Lane 4: Incubated with purified antibodies from the pre-bleed. Lane 5: Incubated with purified polyclonal antibodies. Complexes a and c are indicated.

• •

A

Hu-Cut ------:~----_i~~----

GST ~------

B c

1 2 3 4 1 2 3 4 5 D --a -(;10.'. {J ., - complex c

~ ~.; "'~'"--'>....• ..; '.' ',,' - complexa ; ;;.... :.\"r .'

• 136 •

Fig. 3 Co-transfection experiments with c-myc promoter constructs and hu­ Cut. A. CAT assay from co-transfection with pMBgCAT and 6046 with either pSG (-) or pSG-Cut (+). B. CAT assay from co-transfeetion with MEC and MEC-MElal with either pSG (-) or pSG-Cut (+). C. Graphie representation of relative CAT aetivity for the various e-myc promoter eonstruets. The results are an average of two independent transfeetions. CAT aetivities were eorreeted for transfeetion efficiencies by ~-gal assays. O. Schematie representation of the eonstruets used for transfeetion.

• •

A B pMBgCat .146 ~ MEC-ME1a1 + + + + • • • "'.,; "', '. • •... • • • • • • • • c o

30' .....------, P~P2 pMSgCAT ~i"--.Exon,M CAT ME'.' ...,.. .! 20 046 CAT <" 5 .. 16E-?K s Expn 'WA4 cer -! M ... 10 -;; a: MECoME1a1 1Mt:!C:2!S b ~ Exon , MI çAr MEl.' o pMBgCat 046 r.EC MEC-ME1.1 pSG-Cut 1SV40 ..q;.." C\riCQNA -.;:.;. ..'~'i

• pSG-Cld • 13ï •

Fig. 4 Co-transfections with minimal promoter constructs. Co-transfections were performed with either ME1a1-CAT or ME1a1-E1-CAT as reporter constructs and either 10 J,lg of pSG-Cut (lanes 1 and 4), 5 Ilg of Ga14­ VP16 plus 5 Ilg of pSG (lanes 2 and 5), or 5 Ilg of Ga14-VP16 plus 5 Ilg of pSG­ Cut (lanes 3 and 6). The reporter constructs are schematically represented below.

• •

ME1a1-CAT ME1a1-E1-CAT

1 2 3 4 5 6 .e _ 0

"," ..~. .. - ••• ME1a1~CAT ~5X ME1a1~ Gal4H6X CAT.

ME1 a1-E1-CAT ~5X Gal4H6X ME1a1rItëxon 1 ~

• 138 • ChapterIV Three factors can interact with the MElal site within the c-myc promoter

Preface

In the previous chapter, three cellular factors were shown to bind to the MElal site. The cDNA encoding the factor responsible for complex "c" was isolated. In this chapter, the factors responsible for complexes "a" and "b" are characterized. It had been suggested that the transcription factor SpI was responsible for the formation of complex "a". Here we demonstrate using several approaches that a factor other than SpI is responsible for complex a. Tlùs factor requires purine rich sequences for binding and forms a very stable complex with the MElal site. UV cross-linking experiments demonstrated that a single protein of 82 kDa is responsible for complex band requires sequences in the 5' half of the MElal site for binding. In crude HeLa nuclear, extracts complex 'b" is not detected but, upon fractionation of the extracts, this complex is very abundant. Mixing experiments demonstrated that the binding of the protein responsible for complex "b" was inhibited by lùgh molecular weight factors.

• 139

Abstract

• Studies of the c-myc promoter have shown that the ME1a1 protein binding site located upstream of the P2 TATA box is required for efficient transcription initiation at the P2 start site as weIl as block to elongation of transcription. Although one protein-DNA complex was detecred in crude HeLa extracts, fractionation studies demonstrate that at least three cellular factors bind to ME1a1 generating three protein-DNA complexes termed "a", "b", and "c", where complex "a" corresponds to the complex detected with crude extracts. It has been suggested that the transcription factor Sp1 or a closely related protein binds to ME1a1 (Asselin et al., 1989; Desjardins and Hay, 1993). Using rate of dissociation experiments and competition assays we demonstrate that a factor other than Sp1 is responsible for complex "a". This factor has an apparent molecular weight of 120-150 kDa and requîres purine rich sequences for binding. We have also characterized the factor responsible for complex ''b''. A single protein of 82 kDa, as determined by UV cross-linking, is responsible for complex "b" and recognizes sequences in the 5' portion of the ME1al site, From 96-113 relative to the Pl initiation site. The formation of complex "b" is inhibited by high molecular weight factors, suggesting that this binding activity may be regulated in the cell.

• 140

Introduction

• The control of c-mye expression has been shown to be critical for both cellular proliferation and differentiation (Cole, 1986; Luscher and Eisenman, 1990; Spencer and Groudine, 1991; Marcu et al., 1992; Evans and Littlewood, 1993). Expression of the c-mye gene involves intricate regulatory mechanisms which occur at multiple levels including both transcription initiation and elongation, mRNA stability, translation, and protein stability (Spencer and Groudine, 1991). Several positive and negative cis-acting elements have been identifi2d both upstream and downstream of the transcription initiation sites and presumably coordinate c-mye expression in response to external stimuli. Characterization of the trans-acting factors which interact with thesc eis-acting elements is necessary to understand the mechanisms regulating c-mye gene expression. Transcription of the c-mye gene is initiated from two major start sites termed Pl and P2, which are located 164 base pairs apart in the mouse. P2 is the predominant initiation site, giving rise to 75-90% of c-mye mRNAs (Spencer and Groudine, 1991; Marcu et al., 1992). Between the Pl and P2 initiation sites are several eis-acting elements which affect initiation at P2. These include the ME1a1, ME1a2, and protein binding sites. The E2F site was originally identified in the E2 promoter of adenovirus (Kovesdi et al.,

1986) and has been shown to be important for P2 initiation in Xenopus oocytes (Nishikura, 1986; Meulia et al., 1992). E2F has recently been shown to be important in the control of cell proliferation, and its activity is regulated by the retinoblastoma tumor suppressor protein Rb (reviewed in Nevins, 1992). The ME1a1 and ME1a2 sites were identified by DNase 1 footprinting analysis • (Asselin et al., 1989). The ME1a2 site does not contribute significantly to 141

initiation at P2, however, deietions or mutations in the ME1a1 site strongly deerease the levels of P2 initiated transeripts (Miller et al., 1989; Bossone et al., • 1992). Aside from its effeet on transcription initiation, the ME1a1 site was also shown to be involved in the regulation of transcription elongation. Deletion of the ME1a1 site from the e-myc promoter results in a decrease in block to transcription elongation as determined by nuclear run-on assays (Miller et al.,

1989). In addition, using a construct in which c-myc exon 1 sequences were placed downstream of a heterologous promoter, we showed that transcription elongation was greatly reduced following insertion of the ME1a1 site upstream of the TATA box (Dufort et al., 1993). Moreover, a correlation between in vitro binding to cellular factors and the ability to confer block to transcription elongation in vivo has been established, strongly implicating sequence-specifie transcription factors in the control of transcription elongation (Dufort et al., 1993). The cellular factors which bind to the ME1a1 site have not been characterized. One protein-DNA complex has been detected by electrophoretic mobility shift assays (EMSA) using the ME1a1 site as a probe and HeLa crude nuclear extracts (Hall, 1990; Bossone et al., 1992; Dufort et al., 1993). It has been suggested that the cellular factor SpI or a closely related protein binds to ME1a1 (Asselin et al., 1989; Desjardins and Hay, 1993). Partially purified as weil as bacterially expressed SpI protein protects the ME1a1 site from DNase l cleavage (Asselin et al., 1989), and methylation Interference assays have demonstrated that proteins in HeLa crude nuclear extracts and purified SpI protein contact identical residues (Desjardins and Hay, 1993). Furthermore,

addition of anti-Sp1 antibodies to HeLa nuc1ear extracts has been shown to reduce binding to ME1a1 and UV cross-linking experiments have suggested • that a 95-97 kDa protein binds to ME1a1 (Desjardins and Hay, 1993). The 142

screening of cDNA expression libraries with an MElal or an MEla2 probe has led to the isolation of cDNAs for the zinc finger proteins MAZ and ZF87 • (Bossone et al., 1992; Pyrc et al., 1992). Although these proteins were shown to bind to ME1a1 when expressed in bacteria, it rel."'Iains to be demonstrated whether they are responsible for the binding detected in crude nuclear extracts and whether they can affect c-mye expression. Fractionation of crude nuclear extracts by size exclusion chromatography indicated that at least three factors could interact with the ME1al binding site, generating three protein-DNA complexes termed, "a", "b", "c". Complex "a" corresponds to the unique protein-DNA complex previously detected in crude HeLa nuclear extracts. In this report, we characterize the cellular factors responsible for the formation of complexes "a" and "b". Using a variety of approaches, we demonstrate that a factor other than SpI is responsible for complex "a". This factor has an apparent molecular weight of 120-150 kDa and requires purine rich sequences for binding. We also show that the protein responsible for complex "b" is an 82 kDa protein and that the binding of this protein to ME1a1 is inhibited by high molecular weight factors.

Materials and Methods

Preparation of nuclear extracts and size exclusion chromatography. HeLa cell nuclear extraets were prepared according to the procedure of Dignam et al. (1983). Extraets were fractionated on a Waters Protein Pak Glass 300 SW size exclusion column, using a Waters 625 HPLC. Usually, 1 mg of crude nuclear

extract was applied to the column, run at a flow rate of 0.5 ml/min. and 0.5 ml • fractions were collected. 143

Electrophoretic mobility shift assays (EMSA). The sequence of the oligonucleotide used for the ME1a1 site was GAAAAAGAAGGGAGGGG­ • AGGGA, and the oligonucleotide used for the SpI consensus binding site was GATCCGGGCGGGCGG. EMSA were performed with either la Ilg of crude HeLa nuclear extracts or 3 III of fractionated HeLa ceU nuclear extracts. Binding reactions were performed in 25 mM NaCI, la mM Tris, pH 7.5, 1 mM MgC12' 5 mM EDTA, pH 8.0, 5% glycerol, and 1 mM DTT, in a final volume of 20 ul, with 1 Ilg of poly (dI-dC) and the indicated amount of specifie competitor, where specified. Samples were incubated at room temperature for 5 min., and 20 000 cpm (100 pg) of end-Iabeled probe were added and incubated for 15 min. The samples were then loaded on a 5% polyacrylamide gel (30:1) and electrophoresed at 8 volts/cm for 3 hours in 50 mM Tris, 0.38 M glycine, and 1 mM EDTA, pH 8.5. The Gel was dried and visualized by autoradiography.

Kinetic of dissociation. The kinetic of dissociation experiments were performed as described by Chodosh et al. (1986). Briefly, the binding reactions were performed as described above except that the reactions was scaled-up according to the number of time points. The samples were incubated for 20 min., and 1000-fold excess of competitor was added (time 0) and aliquots were removed at the indicated times and loaded immediately on a polyacrylamide gel.

uv cross-linking. To UV cross-lin!< the protein to the ME1a1 probe, standard binding reactions were scaled-up three fold. Reactions were performed in the presence or absence of specifie competitor, and incubated at room temperature for 12 min. The samples were spotted on a parafilm and exposed for la min. • to 254 nm UV light placed 2.5 cm from the sample. Following UV exposure, SOS sample buffer was added to each samples, heated to 95°C for 10 min. and run on an 8% SOS-PAGE gel. The gel was dried and visualized by • autoradiography.

Results

A cellular factor distinct from SpI is responsibie for the formation of complex a. It has been suggested that the transcription factor SpI or a related factor is responsible for the formation of complex a. As a first step in determining whether the transcription factor SpI or a related factor was responsible for the formation of complex a, we performed EMSA with crude nuclear extracts, using as a probe double-stranded oligonucleotides encoding a high affinity SpI binding site. To determine whether the MElal site could compete for binding to the SpI site, competition experiments were performed using increasing amOlLo."lts of unlabelled oligonucleotides corresponding to either the MElal or the SpI site. As illustrated in figure la, the MElal site competed as efficiently as the SpI site for binding to the SpI probe. Similarly, when using the MElal site as a probe, SpI oligonucleotides competed as efficiently as MElal oligonucleotides (Fig. lb). Thus, these results are in accordance with the notion that the SpI factor may be responsible for the formation of complex a. To further characterize the factor binding to the MElal site in crude extracts, the rates of dissociation of the proteins bound to the SpI and the MElal sites were determined. The time for half of the protein-ONA complexes (tl/2) to dissociate from the SpI site was estimated to be 6 min. (Fig. • 2a). However, with the MElal site, no dissociation was detected after 20 min. 145

(Fig. 2b). Interestingly, the protein binding to the MElal site formed a very stable complex and no dissociation was observed for periods of up to two • hours (Fig. 3). Two possible explanations can account for the difference in dissociation rates between the SpI and MElal probes. Either different factors bind to the MElal and the SpI sites, or the same factor binds to both sites, but interacts with the MElal site with a higher affinity. To distinguish between these two possibilities, dissociation experiments were performed using the SpI site as a probe and MElal oligonucleotides as competitor. If MElal is a high affinity binding site, then it should compete efficiently for binding to the SpI site and a t] /2 of approximately 6 min. should be obtained when MElal oligonucleotides are used as competitor. However, as illustrated in figure 4a, in this situation the MElal site did not compete for binding to the SpI probe, suggesting that the MElal site is not a higher affinity binding site and that the protein detected by EMSA with the SpI probe, presumably the SpI factor, indeed prefers the SpI site over the MElal site. Also, no dissociation of the protein bound to the MElal site was detected when SpI oligonucleotides were used as competitor (Fig 4b). Taken together, the difference in dissociation rate between the MElal and the SpI probe and the absence of competition between these two sites in dissociation experiments suggest that different factors are responsible for the binding detected with the MElal and the SpI sites.

Three faetors ean interaet with the MElal site. With crude HeLa nuclear extraets only one protein-DNA eomplex was deteeted in EMSA using the MElal site as a probe (Fig. 5; Dufort et al., 1993). However, when crude HeLa nuclear extracts are fractionated by size exclusion • chromatography, three protein DNA complexes termed "a", "b", and "e" are 1-16

detected (Fig. 1; Dufort and Nepveu, submitted). Complex "a" corresponds to the complex which was detected in crude nuclear extracts. The apparent • molecular weights of the factors, as estimated by size exclusion chromatography, are 120-150 kDa for complex "a", 66-90 kDa for complex "b", and 180-200 kDa for complex "c". Accumulating evidence suggests that the human Cut homeodomain protein is responsible for the formation of complex "c" (Dufort and Nepveu submitted).

A purine binding protein is responsible for the formation of complex "a" To characterize the sequences within the MElal site that were required for the formation of complex "a", competition experiments were performed using various oligonudeotides encoding either different regions of MElal or containing point mutations within the MElal site. Fraction 5 from size exdusion chromatography was incubated for 5 min. with 50 ng of competitor DNA. The MElal probe was then added and incubated for 15 min. The results are summarized in Table 1. AlI competitors competed to varying extents for binding to the MElal site. Although these oligonudeotides contained different sequences, aIl induded stretches of purine residues. To determine if purine rich sequences were sufficient for binding, an oligonucleotide containing random purine residues (PUR) was synthesized and used as competitor. This random purine oligonudeotide competed very weIl for binding to the MElal site. On the other hand, an oligonudeotide containing mutations throughout the MElal site which reduced its purine content, did not compete for binding (mut. 2). Taken together, these results suggest that the formation of complex "a" does not require specific sequences, but rather is dependent on the presence of a stretch of purine residues. Although the factor • bound random purine sequences, a preference for G residues was observed 147

since G rich oligonucleotides (109-117) competed with a rugher efficiency than • A rich oligonucieotides (96-104). High molecular weight factors prevent the binding of the cellular factor responsible for complex "b". It is striking ~hat complex "b", wruch is not detected in crude extracts, is very abundant once the extract has been fractionated. One possible explanation for trus is that the formation of complex "b" may be inhibited by cellular factors which are separated from this binding activity upon fractionation. Alternatively, in crude extracts there may be competition for binding to the ME1a1 probe between the factors responsible for complexes "a" and "b". To test these hypotheses, an equal quantity of fraction 7 was mixed with each fraction, and the resulting mixtures were used in EMSA. As shown in Fig. 5, a small amount of complex "a" is detected in fraction 7 in addition to complex "b". When fraction 7 was mixed with either fraction 1, 2, 3, or 4, complex "b" disappeared (Fig. 6) suggesting that factors present in these fractions prevented the formation of complex "b" since the small amount of complex "a" present in fraction 7 was not affected when mixed to fractions 1 tn 4. When fraction 7 was mixed with fractions 5 and 6, both complexes "a" and "b" were detected demonstrating that the absence of binding in crude extracts is not due to competition. Therefore, these res'.ùts suggest that in crude extracts complex "bu is not detected because of the presence of high molecular weight inhibitory factors. Upon fractionation by size exclusion, these inhioitory factors are separated from the ME1a1 binding factor and complex b is detected. Similar results are obtained whether the extracts are prepared using the method of Dignam et :.~. (1983), Shapiro et al. (1988), or • MaIÙey et al. (1983). Sequences from 96 to 113 of ME1a1 are required for binding. • To determine the sequences within MElal required for the formation of complex "b", EMSA were performed using fraction 7 and various oligonucleotides. The results are summarized in Table 2. Sequences from 96 to IDS were not sufficient to compete for binding. However, sequences from 96 to 113 efficiently competed for binding to MElal suggesting that sequences between IDS and 113 (GGGGA) are important for binding. Mutations of the two adenine residues at position 99 and 100 (99/100) abolished the ability of this oligonucleotide to compete, demonstrating that these two residues are important for binding. Oligonucleotides containing mutations at residues 103/104, 106/107, or 100/111 efficientl' competed for binding to MElal. Altogether, these results indicate that the cellular factor responsible for complex "b" recognizes sequences in the 5' portion of the MElal site, from +96 to +113 relative to the Pl start site, and that the two A residues at positions 99 and 100 are essential for binding.

An 82 kDa protein is responsible for complex b. The molecular weight of the factor responsible for complex b was determined by UV cross-linking. Binding reactions were performed using fraction 7 and labeled MElal oligonucleotides, in the presence or absence of specifie competitor. After cross-linking, the protein-DNA complexes were run on an SDS-PAGE gel. As illustrated in figure 7, in the absence of specifie competitor, a protein of approximately 82 kDa is detected. This value is in agreement with the apparent rnolecular weight estimated by size exclusion • chromatography (Fig. 5). 149

Discussion

• The MElal protein binding site is a 25 bp region of the c-mye promoter that plays an important role in the regulation of c-mye expression. MElal is

necessary for efficient transcription initiation at the pz start site as weil as for the control of transcription elongation. In HeLa crude nuclear extracts, only complex "a" is detected and several reports have suggested that this complex may be due to binding of the transcription factor SpI to the MElal site (Asselin et al., 1989; DesJardins et al., 1993). In this report, using several approaches, we have demonstrated that a factor other than SpI is responsible for the formation of complex "a". Firstly, the rate of dissociation of the factor bound to an SpI consensus binding site was determined to be approximately 6 min. whereas binding to the MElal site was very stable and no dissociation was detected after two hours. Secondly, in dissociation experiments the SpI consensus binding site and the MElal site did not compete with each other: once the protein-DNA complex was formed with the SpI site, the MElal site could not displace it. Thirdly, competition assays demonstrated that the factor binding to MElal is a purine binding factor since a random purine oligonucleotide competed for binding and mutations which decreased the purine content of the MElal site also decreased the ability to compete for binding. Finally, the apparent molecular weight of the purine binding factor was estimated by size exclusion chromatography to be 120-150 kDa while that of SpI is 95-105 kDa (Briggs et al., 1986). Taken together, these results demonstrated that a factor other than SpI binds to MElal. The MElal site, when added in excess, competed for binding to the SpI site, suggesting that SpI, under certain conditions, can bind to the MElal site. • This is supported by the observations that partially purified SpI protein can 150

protect the ME1a1 site from DNase l c1eavage when a reduced amount of non­ specific competitor is used (Asselin et al., 1989) and that the addition of anti­ • SpI antibodies reduced but did not eliminate binding to MElal (DesJardin et al.,1993). We have characterized the cellular factor responsible for complex "b". UV cross-linking experiments indicated that a single protein of 82 kDa was responsible for complex "b". Competition experiments showed that binding requires sequences in the 5' portion of the MElal site, from +96 to +113 relative to the Pl start site. Interestingly, this complex is not detected in crude HeLa extracts but is very abundant once the extract has been fractionated by size exclusion chromatography. We have demonstrated that high molecular weight factors inlùbit the binding of this factor in crude extracts. We do not know at this point the basis for this inhibition, but these results suggest tha: this binding activity may be regulated in the cell. Recently, two independent groups have isoiated cDNAs encoding zinc­ finger proteins that bind to MElal. These proteins were designated MAZ (Mw;. Associated Zinc)(Bossone et al., 1992) and ZF87 (Pyrc et al., 1992). ZF87 was cloned as a protein that binds to the MEla2 site, upstream of MElal, however it was shown to bind to MElal as weil. The factor responsible for complex "a" can also bind to both sites (Moberg et al., 1992; Dufort and Nepveu, unpublished observation). Therefore, ZF87 may be the cellular factor which is responsible for complex "a". However, the molecular weight of this factor is approximately 87 kDa which is lower that the 120-150 kDa estimated by size exclusion chromatography. It is possible that this protein interacts with other proteins to form a higher molecular weight complex. Further characterization of ZF87 will determine if this factor is indeed responsible for complex "a". • MAZ was cloned as a protein that binds to the MElal site. This factor does not 151

bind to ME1a2 nor to a high affinity SpI site, but has been shown to bind to sites containing a GGGAGGG core motif (Bossone et al., 1992). Therefore, this • factor appears to be different from the factors responsible for complexes "a" and "b", however, we cannot exclude the possibility that the binding conditions which we have used do not favor the binding of this factor. We have characterized two factors which bind to the ME1a1 site. Analysis of the sequence requirements for binding demonstrated that they bind to overlapping sequences. This raises the possibility that they may compete for binding to the ME1a1 site. Their effect on c-mye gene expression in not known, however, the ME1a1 site has been shown to be required for efficient initiation at the P2 start site as weIl as block to transcription elongation (Asselin et al., 1989; Miller et al., 1989; Bossone et al., 1992; Dufort et al., 1993). Thus, one factor may affect transcription initiation whereas another factor would affect transcription elongation, or alternatively, the same factor may affect both initiation and elongation. By introducing mutations in the ME1a1 site, a correlation has previously been established between protein binding to this site in vitro using crude nuc1ear extracts and the ability to confer block to transcription elongation in vivo (Dufort et al., 1993). In the present study, we have analyzed the binding specificity of the protein responsible for complex "a". Mutations within the MEla1 site which had previously been shown to reduce transcriptional block, also reduced the formation of complex "a", strongly suggesting that this factor is involved in transcriptional block. Analysis of the sequence requirements for binding of the factor responsible for complex 'b" demonstrated that this factor could not bind to mut.1, and this mutant had previously been shown to only slightly reduce transcriptional block (Dufort et al., 1993). Therefore, this factor is most • probably not involved in block to transcription elongation. However, it may 152

be involved in transcription initiation at the P2 start site. It is interesting to note that high molecular weight factors inhibited the binding of this factor in • crude extracts. Whether this also occurs in vh'o is not known but this may be a mechanism to regulate the activity of this factor. Several transcription factors have been shown to be regulated in this manner. Further characterization of the cellular factors binding to the MElal site and the isolation of the cDNAs encoding these factors will be necessary to determine their effect on c-mye gene expression.

Acknowledgments

0.0. is supported by a studentship from the National Cancer Institute of Canada. This research was supported by grant #3497 from the National Cancer Institute of Canada to A.N.

References

Asselin, c., A. Nepveu, and K.B. Marcu. 1989. Molecular requirements for transcriptional initiation of the murine c-mye gene. Oncogene. 4:549-558.

Bossone, S.A., C. Asselin, A.J. Patel, and K.B. Marcu. 1992. MAZ, a zinc finger protein, binds to c-MYC and C2 gene sequences regulating traru.cription initiation and termination. Proc. Nat!. Acad. Sei. USA 89:7452-7456.

Briggs, M.R., J.T. Kadonaga, S.P. Bell, and R. Tjian. 1986. Purification and biochemical characteization of the promoter-specific transcription factor SpI. • Science 234:47-52. 153

Chodosh, L.A., R.W. Carthew, and P.A Sharp. 1986. A single polypeptide • possesses the binding and transcription activities of the adenovirus magor late transcription factor. Mol. Cell. Biol. 6:4723-4733.

Cole, M.D. 1986. The myc oncogene: its role in transformation and differentiation. Annu. Rev. Genet. 20:361-384.

Desjardins, E., and N. Hay. 1993. Repeated CT elemerits bound by zinc finger proteins control the absolute and relative activities of the two principle human c-mye promoters. Mol Cell. Biol. 13:5710-5724.

Dignam, J.D., P.L. Martin, B.S. Shastry, and R.G Roeder. 1983. Eu.karyotic gene transcription with purified components. Methods Enzymol. 101:582-598.

Dufort, D., and A. Nepveu. The human cut homeodomain protein represses transcription fron the c-mye promoter. Submitted.

Dufort, D., M. Drolet, and A. Nepveu. 1993. A protein binding site from the murine c-mye promoter contributes to transcriptional block. Oncogene 8:165­ 171.

Evan, G.I., and T.D. Littlewood. 1993. The role of c-mye in eell growth. Curr. Opin. Genet. Oev. 3:44-49.

Hall, D.J. 1990. Regulation of e-mye transcription in vitro: dependenee on the • guanine-rieh promoter element ME1al. Oneogene 5:47-54. 15-1

Kovesdi, 1., R. Reichel, and ].R. Nevins. 1986. Identification of a cellular • transcription factor involved in ElA trans-activation. Cell 45:219-228.

Lüscher, B., and R.N. Eisenman. 1990. New light on Myc and Myb. Part 1. Myc. Genes Dev. 4:2025-2035.

ManIey, J.L., A. Fire, M. Samuels, and P.A. Sharp. 1983. ln vitro transcription: whole-cel! extract. Methods Enzymol. 101:568-582.

Marcu, K.B., S.A. Bossone, and A.J. Pate!. 1992. mye function and regulation. Annu. Rev. Biochem. 61:809-860.

Meulia, T., A. Krumm, C.A. Spencer, and M. Groudine. 1992. Sequences in the human c-mye P2 promoter affect the elongation and premature termination of transcripts initiateà from the upstream Pl promoter. Mol. Cell. Biol. 12:4590-4600.

Miller, H., C. Asselin, D. Dufort, J.Q. Yang, K. Gupta, K.B. Marcu, and A. Nepveu. 1989. Acis-acting element in the promoter region of the murine c­ mye gene is necessary for transcriptional block. Mol. Cel!. Biol. 9:5340-5349.

Nevins, J.R. 1992. E2F: a link between the Rb tumor suppressor protein andviral oncoproteins. Science 258:424-429. • 155

Nishikura, K. 1986. Sequences involved in accurate and effecient transcription of human c-mye genes microinjected into frog oocytes. Mol. Cell. Biol. 6:4093­ • 4098.

Pyrc, J.J., K.H. Moberg, and D.J. Hall. 1992. Isolation of a novel cDNA encoding a zinc-finger protein that binds to two sites within the c-mye promoter. Biochemistry 31:4102-4110.

Shapiro, D. J., P. A. Sharp, W. W.Wahli, and M. J. Keller. 1988. A high efficiency HeLa nudear transcription extract. DNA 7: 47 (1988).

Spencer, C. A., and M. Groudine. 1991. Control of c-myc regulation in normal and neoplastic cells. Adv. Canœr Res. 56:1-48.

• 156 •

Figure 1. Competition assay with ME1a1 and Sp1 oligonucleotides. EMSA were performed as described in the material and methods, with oligonucleotides encoding either the SpI site (A) or the MElal site (B) as probes. Varying quantities of either MElal or SpI oligonucleotides, as indicated above each lane, were added as competitor.

• ••

A. B. MElal SpI MElal SpI i Il 1 i Il 1 (F) 0 10 25 50 75 10 25 50 75 (no> (F) 0 10 25 50 75 10 25 50 75

.. - -

probe Sp1 probe ME1a1

•• 15ï •

Figure 2. Determination of the rate of dissociation of the proteins bound to the Spl and MElal sites. Binding reactions were incubated for 15 min., and at time 0, 1000fold excess of indicated competitor was added, and aliquots werc taken at the times indicated above each lane. A. The SpI site was used as a probe, and SpI oligonucleotides were used as competitor. B. The MElal site was used as a probe and MElal oligonucleotides were used as competitor . The time for l),alf of the protein-DNA complexes to dissociate (tl/2) is indicated below.

• A. fJ:A.Il.I Sp1 ollgos B. fJ:A.Il.I ME1.1 ollgo. Comool!!Q[ Spl 011905 Compel!!Q[ MEla1 011905

o 2 4 6 8 12 16 20 (min) o 2 4 6 8 12 16 20 (min)

...... :. -......

t1/2 = 6 min. t1/2 >20 min.

, 158 •

Figure 3. Binding of the cellular factor to the MElol site is very stable. Binding reactions were performed with the MElal site and MElal oligonudeotides were useà as competitor. Aliquots were taken at the time indicated above each lane.

..'

• •

.& :: 0.02468 12 16 20 25 30 35 40 45 50 60 70 80 90 120 (min)

• 159 •

Figure 4. MElal oligonucleotides do not compete for binding to the SpI probe in dissociation experiments. A. The SpI site was used as a probe, and at time a min., 1000 fold excess of ME1a1 oligonucleotides were added and aliquots taken at the time indicated above each lane. B. The MEtal site was used as a probe and SpI oligonucleotides were used as competitor.

• •

A. Spl 011905 B. ME1.l 01190. ComQ9lI!Qr MElal 01190' eompetnor Spl 011908 .8 ~ 2 "-0246 8 12 16 20 24 (min) ,,-0246 8 12 16 20 24 (min)

• 160 •

Figure 5. Fractionation of HeLa nuclear extract by size exclusion chromatography. HeLa nuclear extract was fractionated on a Protein Pak Glass 300 5W size exclusion column and 0.5 ml fractions were collected. EM5A were performed using the ME1a1 site. The void volume was not collected, therefore fraction 1 corresponds to the first proteins eluting from the column. Fraction numbers are indicated above each lane. The three protein-DNA complexes, a, b, and c are indicated. The relative elution of several molecular weight markers is indicated below.

• •

0 CO ~ -X Q)

Q) .0 co Fractions 0 ~ ~ -0 a.. ..r- 1 2 3 4 5 6 7 8 9 10 1 1 12 :3 14 complex c-

complex b-

complex a-

1 1 1 1 MW. 200 150 66 29

• 161 •

Table 1. Summary Jf the sequence :equirements for the binding of the cellular factor responsible for complex a. EM5A were performed with the MElal site as a probe and crude HeLa nuclear extracts. Competition experiments were performed with 50 ng of indicated oligonucleotides. As indicated, the sequence of the MElal site was composed of nucleotides 96 to 117 relative to the Pl initiation site. The sequences corresponding to MElal for each oligonucleotide are represented by solid lines. In the oligonucleotides containing point mutations, the residues which have been mutated are ilIustrated. The sequence of PUR is a random purine sequence.

• • Competition

ME1a1 GAAAAAGAAGGGAGGGGAGGGA +++

96-104 + 96-108 ++ 102-113 +++ 104-117 +++ 109-117 +++ 96-113 +++ 99/100 -TT------+++ 103/104 ----TT------+++ 106/107 ------'TT------+++ 110/111 ------'TT--- +++ Mut. 1 -----T--TT-T-T- + Mut. 2 T-C-T-C-T-TT-T-T- PUR AAGAGGAGAGGAAAG +++

Recognition Purine rich sequences site:

• 162 •

Figure 6. High molecular weight factors inhibit the binding of the factor responsible for complex b. An equal quantity of fraction 7 was mixed with each individual fractions, and the resulting mixtures were used in EM5A. In the lane labeled fraction 7, only extract from fraction 7 was used in the EM5A. The number above each lane corresponds to the fraction to which fraction 7 wasadded.

• •

l'-.

C 0

• 163 •

Table 2. 5ummary of the sequence requirements for the binding of the cellular factor responsible for complex b. EM5A were pe:.rormed with the MElal site as a probe and crude HeLa nuc1ear extracts. Competition experiments were performed with 50 ng of indicated oligonucleotides. As indicated, the sequence of the MElal site was composed of nucleotides 96 to 117 relative to the Pl initiation site. The sequences corresponding to MElal for each oligonucleotide are represented by solid lines. In the oligonucleotides containing point mutations, the residues which have been mutated are illush·ated.

• •

Competition

ME1a1 G~~GAAGGGAGGGGAGGGA +++

96-108 102-113 104-117 109-117 96-113 +++ 99/100 -TT------103/104 ----TT------+++ 106/107 ------TT------+++ 110/111 ------TT--- +++ Mut. 1 -----T-TT-T-T- Recognition ** site: GAAAAAGAAGGGAGGGGA

• 164 •

Figure 7. UV cross-linking of the factor responsible for the formation of complex b. Binding reactions were performed using the MElal site as a probe and proteins from fraction 7. The reactions were performed in the absence (-) or the presence (+) of specifie competitor. The mixture was exposed to UV and the complexes were separated by SOS-PAGE. The positions of different molecular weight markers are indicated.

• •

u.V. Cross Linking With Fraction 7

110- - +

84-

47-

33-

• 165

General Discussion

• Block to transcription elongation in the c-myc gene has been shown to occur within exon 1. This is illustrated in nuclear run-on assays as a higher density of RNA polymerases on the 5' end than on the 3' end of the gene. Furthermore, the extent of block can vary depending on the physiological state of the ceUs demonstrating that c-mye gene expression can be regulated at the level of transcription elongation. The mechanism by which transcription elongation is regulated is not known Three approaches have been used to study transcriptional block in the c­ myc gene. These are nuclear run-on transcription assays, injections into Xenopus oocytes, and in vitro transcription assays. These approaches have recently been shown to reveal different aspects of transcriptional block in the c-mye gene (Krumm et al., 1992; Meulia et al., 1993). Injections in Xenopus oocytes and in vitro transcription assays have identified sequences within exon 1 which can function as intrinsic pause-termination sites. In the human c-mye gene, termination sites have been mapped to T-rich sequences located near to the exon 1-intron 1 border (Bentley and Groudine, 1988; Kerppola and Kane, 1988). In the murine gene, termination sites have been mapped to two stretches of T residues (T5 and T3) located within exon 1 (Bentley and Groudine, 1988; Miller et al., 1989). Sequences from the 3' half of exon 1 including the sites of block of both the human and murine c-mye genes have been shown to cause premature termination when inserted downstream of heterologous promoters (Bentley and Groudine, 1988, Wright and Bishop, 1989). The importance of these sites for transcriptional block in mammalian • cells as detected by nuclear run-on assays is not known since recent studies 166

suggest that they may not be required. 5tudies in mammalian celis have

demonstrated that transcriptional block in the human C-17lYc gene is due to • promoter-proximal pausing of R.l\JA polymerase II complexes (5trobl and Eick, 1992; Krumm et al., 1992). Nuclear run-on assays using short oligonucleotide probes and potassium permanganate mapping of single­ stranded DNA regions have mapped the pause site near position +30 relative to the P2 initiation site (Strobl and Eick, 1992; Krumm et al., 1992). Sequences

downstream of +47 in the human C-17lYc gene, including the sites of block identified by in vitro transcription and microinjections in Xenopus oocytes are not required for promoter-proximal pausing of RNA polymerase II

complexes (Krumm et al., 1992). Sirnilar results have also been obtained from

studies performed on the murine C-17lYc gene (Henderson et al., 1993 submitted; Kolhuber et al., 1993). Nuclear run-on assays using probes spanning exon 1 demonstrated that the density of RNA polymerase declined gradually over distance and that the site of block previously identified to cause block to transcription elongation was not essential. A 3 fold reduction in the density of RNA polymerase was observed between the 5' haIf of exon 1 and the 3' half of exon 1 and a further 4 fold reduction between the 3' half of exon 1 and the proximal part of intron 1 was observed. Therefore, in the murine c-mye gene, although some complexes may pause in the promoter­ proximal region as in the human gene, many complexes transcribe past this region. These complexes which transcribe past the promoter-proximal region

pause or terrninate transcription at sites further downstream within exon l. Thus transcriptional complexes in both the human and murine c-mye genes pause or prematurely terminate transcription within exon l, and the sites of block identified in Xenopus oocytes or in in vitro transcription assays are not • essential for this te occur. 167

Analysis of the sequences required for transcriptional block in the murine c-myc gene have suggested that promoter sequences may be involved. • Deletion analysis have demonstrated that a 16 bp deletion within the P2 promoter significantly reduceè block to transcription elongation in exon 1 as determined by nuclear run-on assays (Miller et al., 1989). This deletion removed part of a previously identified protein binding site termed MElal (Asselin et al., 1989). Thus, these results suggested that the MElal site may be involved in the regulation of transcription elongation in the c-myc gene. The experiments presented in chapter II demonstrated that this site was indeed involved in transcriptional block. Using a chimeric construct containing the H-2Kb major histocompatibility class gene promoter linked to the c-myc first exon, we showed that the addition of the MElal site resulted in an increase in transcriptional block. The MElal site increased block when inserted in both orientations, and functioned when inserted either upstre'IID or downstream of the transcription initiation site. 51 nuclease mapping analysis demonstrated that the insertion of the MElal site in the chimeric construct did not affect the site of transcription initiation but resulted in a 50 fold decrease in the level of cytoplasmic RNA. 5ince the MElal site had been identified as a protein binding site, we investigated whether protein binding to this site was important for block. Mutations were introduced in MElal, and the ability ta interact with nuclear factors in vitro was determined, as weil as the ability to confer transcriptional block in vivo. These results demonstrated that there was a correlation between protein binding in vitro and the ability to confer transcriptional block in vivo, strongly implicating sequence-specific transcription factors in the • control of transcription elongation. 168

The mechanism by which the MElal site located in the promoter regulates transcriptional block is net known, however, a mode! can be • envisaged where factors bound to the MElal site could interact or affect the assembly of transcriptional complexes at the initiation site. In the absence of protein binding to the Melal site, highly processive transcriptional complexes may be assembled at the initiation site which can transcribe the gene, unaffected by pause or termination sites. However, when cellular factors are bound to the MElal site, these factors may directiy interact with the transcriptional complex or direct the assembly of a non-processive or abortive complex at the initiation site. Because these transcriptional complexes are non-processive, they would be affected by pause or termination sequences and

transcribe for oruy short distances within exon 1. The human c-rnyc gene may contain a stronger promoter proximal pause site than the murine gene, causing the majority of transcriptional complexes to pause at this site. However, in the murine c-rnyc gene, only a fraction of transcriptional complexes would pause at this site and the remaining complexes would pause or prematurely terminate transcription at sites further downstream, resulting in a progressive decrease in the density of RNA polymerases. How the presence of factors bound to the MElal site results in the generation of non­ processive complexes is not known. One possible mechanism is that factors bound to the MElal site prevents the interaction of elongation factors with the transcriptional complex. In chapters m and IV, the cellular factors which bind to the MElal site were characterized. EM5A with crude nuclear extracts detected one protein­ DNA complex. However, fractionation of crude HeLa nuclear extracts demonstrated that at least three factors cOuld bi.-1.d te the MElal site, resulting • in the formation of three protein-DNA complexes termed "a", 'b", and "c". 169

Complex "a" corresponded to the complex identified in crude extracts. The factor responsible for complex "a" has an appa,"ent molecular weight of 120­ • 150 kDa and is a purine binding protein which bind to MElal very stahly. The protein responsible for complex "b" is an 80-82 kDa protein whose binding in crude ex tracts is inhibited by high molecular weight factors. Finally, l have cloned the cDNA for the protein responsible for complex "c". The deduced amine acid sequence indicated that this protein was the human homologue of the Drosophila Cut homeodomain protein (hu-Cut). Incubation of nuclear extracts with either monoclonal or polyclonal anti-hu-Cut antibodies resulted in the specific depletion of complex "c". The same antibodies did not inhibit complex "a", and serum from non-immunized animaIs did not affect either binding activities demonstrating that hu-Cut is responsible for complex "c". The MElal site, aside from its effect on transcription elongation, has aIso been shown to be important for efficient initiation at the P2 start site (Miller et al., 1989; Bossone et al., 1992; Asselin et al., 1989). Deletion or mutation of this site strongly decreased the level of P2 initiated transcripts (Miller et al., 1989; Bossone et al., 1992). Thus, there may be factors which interact with the MElal site which affect transcription initiation whereas other factors affect transcription elongation, or alternatively the same factor may affect both initiation and elongation. Which factors are involved in transcriptional block? A correlation between protein binding to the MElal site in vitTo and the ability to confer block to transcription elongation in vivo was established using crude nuclear extracts. This suggested that the cellular factor responsible for complex "a" may be involved. The analysis of the binding specificity of the protein responsible for complex "a" using fractionated extracts, indicated that • the mutations affecting transcriptional block also reduced binding of the factor liO

responsible for complex "a", strongly suggesting that this factor is involved in transcriptional block. • Analysis of the sequence requirements for binding of the factor responsible for complex "b" demonstrated that there was no correlation between the ability to bind to different mutants and the ability of these mutants to confer block to transcription elongation. This factor could not bind to mut.1, and this mutant was shown to only slightly reduce transcriptional block. Therefore, this factor is most probably not involved in block to transcription elongation. However, it may be involved in transcription initiation at the P2 start site. It is interesting to note that high molecular weight factors inhibited the binding of this factor in crude extracts. Whether this also occurs in vivo is not known but this may be a mechanism to regulate the activity of this factor. Several transcription factors have been shown to be regulated in this manner. Further characterization of this factor will be required to determine its effect on the expression of the c-myc gene. The third factor which has been shown to bind to the MElal site is hu­ Cut. This protein contains a homeodomain as weil as three 72 amine acid repeats called Cut repeats. We have recently demonstrated that Cut repeats can function as DNA binding domains (Harada et al., 1993a). Therefore, this protein contains potentially four DNA binding domains. The specifie sequences required for binding of the different Cut repeats have been determined (Harada et al., 1993b). The consensus binding site can be summarized as an ATNNAT core motif. However, many selected sites diverged by one, two or even more nucleotides, and also sequences adjacent to the core motif were shown to affect binding. Each Cut repeat has distinct binding specificities, and combinations of Cut repeats or Cut repeats plus the • homeodomain in some cases changed the specificity of binding. Thus, these 171

results demonstrate that the sequence requirements of Cut repeats are flexible and multiple sequences can be recognized as DNA binding sites. Which Cut • repeat(s) are involved in binding to the MElal site is not known. ln transient co-transfection assays, the human Cut protein was shown to repress expression from an intact c-myc promoter as weil as from a heterologous promoter in which one copy of the MElal binding site had been inserted. However, we do not know how hu-Cut repressed expression from the c-myc promoter and whether repression occurred at the level of transcription initiation or elongation. There are several possible mechanisms by which hu-Cut could repress transcription. One possibility is that hu-Cut competes for binding with a transcriptional activator binding to MEla1. This mechanism is unlikely since repression was observed when the MElal binding site was moved to two different heterologous promoters. Also, sequences within exon 1 were also required for repression. Altematively, it is possible that binding of hu-Cut to the MElal site is facilitated by protein­ protein interactions with other sequence-specific factors, possibly even hu-Cut itself, that interact with other sites within c-myc exon 1. 5ince hu-Cut contains multiple binding domains, it may bind to both the MElal site and to sequences within exon 1. Either protein-protein interactions or the ~inding to multiple sites by hu-Cut may prevent transcription initiation. Finally, it is also possible that hu-Cut represses expression at the level of transcription elongation when exon 1 sequences are present downstream of the transcription initiation site. We have not been able to demonstrate this since block to elongation of transcription is not detected in transiently transfected cells. In order to determine the mechanism of repression by hu-Cut, a stably transfected population of cells expressing hu-Cut under the control of an • inducible promoter will be required. _., 1/-

In conclusion, experiments presented in this thesis have demonstrated that the MElal site located in the P2 promoter is involved in the control • transcription elongation, and that binding of cellular factors to this site is important for block. Three cellular factors have been shown to bind to the MElal site and the cDNA for one of these factors was isolated and shown to encode the human Cut homeodomain protein.

• 173

References

• AsseIin, A. Nepveu, and K.B. Marcu. 1989. Molecclar requirements for c., transcriptional initiation of the murine c-mye gene. Oncogene. 4:549-558.

BentIey, D.L., and M. Groudine. 1988. Sequence requirements for premature termination of transcription in the human c-mye gene. Cell 53:245-256.

Bossone, S.A., C. Asselin, A.J. Patel, and K.B. Marcu. 1992. MAZ, a zinc finger protein, binds to c-MYC and C2 gene sequences regulating transcription initiation and termination. Proc. Nat!. Acad. Sei. USA 89:7452-7456.

Harada, R., D. Dufort, C. Denis-Larose, and A. Nepveu. 1993a. Conserved Cut repeats in the human Cut homeodomain protein function as DNA binding domains. J. Biol. Chem. in press.

Harada, R., G. Bérubé, C. Denis-Larose, and A. Nepveu. 1993b. DNA binding speeificity of the Cut homeodomain and Cut repeats from the human Cut protein. Submitted.

Henderson, J.E., M. Drolet, and A. Nepveu. 1993. Analysis of transcription elongation within the murine e-mye gene. Submitted.

Kerpolla, T.K., and CM. Kane. 1988. Intrinsic sites of transcription termination and pausing in the e-mye gene. Mol Cell. Biol. 8:4389-4394. • 17-1

Kolhuber, F., L.J. Strobl, and D. Eick. 1993. Early down-regulation of c-mye in dimethylsulfoxide-induced mouse erythroleukemia (MEL) ceUs is mediated at • the i:'1/P2 promoters. Oncogene, 8:1099-1102.

Krumm, A., T. Meulia, M. Brunvand, and M. Groudine. 1992. The block to transcriptional elongation within the human c-mye gene is determined in the promoter-proximal region. Genes Dev. 7:2201-2213.

Meulia, T., A. Krumm, and M. Groudine. 1993. Distinct properties of c-mye

transcription elongation are revealed in Xenopus oocytes and mammalian cells and by template titration, 5,6-dichloro-l-l3-o-ribofuranosylbenzimidazole (DRB), and promoter mutagenesis. Mol Cell Biol. 13:5647-5658.

Miller, H., C. Asselin, D. Dufort,J-Q. Yang, K. Gupta, K.B. Marcu, and A. Nepveu. 1989. A eis-acting element in the promoter region of the murine c­ mye gene is necessary for transcriptional block. Mol. Cell. Biol. 9:5340-5349.

Strobl, L., and D. Eick. 1992. Hold back of RNA polymerase II at the transcription start site mediates down-regulation of c-mye in vivo. EMBO J. 11:3307-3314.

Wright, ;;., and J.M. Bishop. 1989. DNA sequences that mediate attenllation of transcription from the mouse protooncogene mye. Proc. Natl. Acad. Sei. USA 86:505-509. •