ElA MEDIATED ACTIVATION OF THE ADENOVIRUS …ElA MEDIATED ACTIVATION OF THE ADENOVIRUS E4 PROMOTER....

ElA MEDIATED ACTIVATION OF THE ADENOVIRUS E4 PROMOTER.

CHRISTOPHER JONES

A thesis submitted for the degree of

Doctor of Philosophy

at

The University of London,

1993.

Imperial Cancer Research Fund,

Clare Hall Laboratories,

Blanche Lane,

South Mimms,

Herts., EN6 3LD

and

University College London,

Gower Street,

London, WCIE 6BT

ProQuest Number: 10045891

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed,

a note will indicate the deletion.

uest.

ProQuest 10045891

Published by ProQuest LLC(2016). Copyright of the Dissertation is held by the Author.

All rights reserved.This work is protected against unauthorized copying under Title 17, United States Code.

Microform Edition © ProQuest LLC.

ProQuest LLC 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, Ml 48106-1346

ABSTRACT

An important function of adenovirus ElA proteins upon an

adenoviral infection of permissive human cells is the transcriptional

activation of viral early genes during the first phase of the lytic cycle. To

this end ElA targets a number of cellular transcription factors, many of

which are sequence specific DNA binding proteins that directly interact

with early viral promoters. Studies of the adenovirus E4 promoter have

identified a promoter element, the ATP binding site, that is critical for

fr^MS-activation by ElA. Several cellular transcription factors interact with

ATE binding sites and are therefore potential targets for ElA.

There is significant evidence that E4F and a member of the ATP

family (ATF-2) can independently function with E lA to activate the

E4 promoter. To assess the role of E4F, I have monitored the effects of

point mutations in the ATF/E4F binding sites on E4 promoter activity

in vivo using a transient expression assay in HeLa cells. I found that

whilst the core motif of these sites is the only upstream requirement for

activation of the E4 promoter, an adjacent point mutation that eliminates

E4F binding (but has no effect on ATE binding) does not affect E4 promoter

activity. These findings indicate that ATE binding sites are the sole

requirement for fraws-activation of the E4 promoter by ElA. I have also

found that some point mutations that strongly reduce E4 promoter

activity in vivo have no effect on binding of ATE-2 to these sites in vitro.

Therefore these results are inconsistent with the suggestion that ATE-2

alone functions with ElA to activate the E4 promoter. Together with the

results of previous studies, these results demonstrate that ElA has

evolved multiple options for activating transcription of early viral genes.

ACKNOWLEDGEMENTS

I would like to thank my supervisors Dr. Kevin Lee (at the Imperial

Cancer Research Fund) and Dr. David Latchman (at University College

London) for their help during my studies. Thanks to Susan Szymkowski

for her valuable comments on the first draft of this thesis and to my

colleagues Norma Masson and Andrew Brown for their help,

encouragement and friendship in the laboratory.

On a non-scientific note, thank you to the following LiDs (Doctors

of Life) for their special contributions: Spencer Howell for saving me from

"the rut"; Mark Hodson, Jonathan Hamill-Keays, Kevin Brotherton,

Christopher Chiverton and David Gaves for thinking that I'm actually

doing something useful; Dave Szymkowski and Christopher J. Jones for

listening to me drone on about ATF-2 and ElA; Philip and Alison

Mitchell for being the most fantastic inspiration (go diamond!); Alan

Tunnacliffe for being incredibly patient whilst I was writing about ElA

instead of thinking about fragile sites; Sharon (I've got a rock on my

finger) Mooney, the only person who would suffer three years of hell

while I further my academic career just because I said I'd marry her when

I'd finished; and of course last, but certainly not least, my parents.

Party on dudes; and be most excellent to each other!

All men dream: But not equally.

Those who dream by night in the dusty

recesses of their minds wake in the day to

find that it was vanity: But the dreamers of

the day are dangerous men, for they may

act their dreams with open eyes, to make it

possible.

T. E. Lawrence, The Seven Pillars of Wisdom.

The winner sees an answer for every

problem; the loser sees a problem for every

answer. The winner sees a green near every

sand trap; the loser sees two or three sand

traps near every green. The winner says "It

may be difficult, but it's possible"; the loser

says "It may be possible, but it's too

difficult". Your choice is, which one do you

want to be?

Don't let anyone steal your dream.

Dexter Yager.

This thesis is dedicated to my parents.

For their love and support, I am eternally grateful.

CONTENTS

ABSTRACT................................................................................................................. 2ACKNOWLEDGEMENTS....................................................................................... 3CONTENTS.................................................................................................................. 6

FIGURES............................................................................................................... 10TABLES................................................................................................................. 12

Chapter 1: Introduction............................................................................................13

1.0 General introduction.......................................................................................... 131.1 The adenovirus replication cycle......................................................................14

1.1.1 The ElA region........................................................................................... 141.1.2 Activation of adenovirus early genes...................................................18

1.2 Mammalian transcription..................................................................................181.3 Control of RNA polymerase II transcription.................................................19

1.3.1 The TATA box........................................................................................... 191.3.2 Upstream promoter elements.................................................................. 20

1.4 Cellular factors involved in transcription by RNApolymerase II........................................................................................................ 211.4.1 RNA polymerase II.....................................................................................211.4.2 General transcription (initiation) factors...............................................22

1.4.2.1 The TATA binding protein........................................................231.4.2.2 Other factors involved in assembly of the

preinitiation complex..................................................................251.4.3 Gene-specific transcription factors.........................................................25

1.5 Trans-activation by adenovirus ElA proteins...............................................271.5.1 A unique conserved region (CR3) of the large ElA protein

is essential for frans-activation by E lA .................................................301.5.2 CR3 contains one copy of a C4 zinc finger motif that is

essential and sufficient for frans-activation.......................................... 321.5.3 CR3 consists of two subdomains............................................................321.5.4 The activating domain of ElA specifically interacts with

TBP............................................................................................................... 351.5.5 ElA can activate transcription in vitro.................................................36

1.6 Characterisation of the adenovirus E4 promoter..........................................371.6.1 Deletional analysis of the E4 promoter................................................ 37

1.6.2 Identification of cellular factor binding sites within theE4 promoter................................................................................................. 39

1.7 Characterisation of proteins that interact with the E4 promoter.................411.7.1 The ATE family......................................................................................... 41

1.7.1.1 Characterisation of ATE binding sites in the viralearly promoters............................................................................ 41

1.7.1.2 Isolation of ATE proteins............................................................ 431.7.1.3 cDNA cloning reveals a multigene family of

transcription factors.................................................................... 431.7.2 E4E................................................................................................................ 471.7.3 Other proteins that bind to the E4 promoter...................................... 48

1.8 Trans-activation of the E4 promoter by E lA ................................................. 481.8.1 E4E mediated frans-activation of the E4 promoter by E lA ..............491.8.2 ATE-2 mediates frans-activation by E lA ............................................. 54

1.9 Aims of this thesis.................................................................................................57

Chapter 2: Analysis of the requirements for E4 promoter activityin vivo.........................................................................................................58

2.0 Methods and Materials....................................................................................... 602.0.1 General strategy of construction of new recombinants.................... 602.0.2 Electrophoretic analysis of nucleic acids.............................................. 62

2.0.2.1 Agarose gel electrophoresis of nucleic acids.......................... 622.0.2.2 Denaturing polyacrylamide gel electrophoresis of

nucleic acids.................................................................................. 622.0.2.3 Gel purification of nucleic acids................................................63

2.0.3 Preparation of plasmid D N A ................................................................. 642.0.3.1 'Maxiprep' DNA............................................................................642.0.3.2 'Miniprep' DNA............................................................................ 65

2.0.4 Preparation of competent bacteria and transformationwith recombinant plasmid D N A .......................................................... 66

2.0.5 Transient transfection assay................................................................... 672.0.5.1 CaCb/phosphate transfection of plasmid DNA into

mammalian cells......................................................................... 672.0.5.2 Assay of chloramphenicol acetyl transferase

activity............................................................................................. 68

2.1 Results......................................................................................................................692.1.0 An ATF binding site that does not bind E4F can mediate

trans-activation of the E4 promoter by E lA ......................................... 692.1.1 Spl binding sites are unable to mediate efficient trans-

activation of the E4 promoter by ElA.....................................................732.1.2 E4F binding is not required for activation of the intact

E4 promoter by E lA ...................................................................................772.1.3 Upstream promoter elements other than the ATF/E4F

binding sites are not required for activation by E lA ..........................832.1.4 E4F binding sites are not required for activation of the

E4 promoter by ElA in Vero cells............................................................84

2.2 Discussion............................................................................................................ 88

Chapter 3: Determination of the DNA binding specificity of anN-terminally truncated ATF-2 protein in vitro................................ 92

3.0 Methods and Materials....................................................................................... 933.0.1 In vitro transcription and translation of ATF proteins......................933.0.2 SDS/polyacrylamide gel electrophoresis of proteins.........................943.0.3 Analysis of the DNA binding activity of proteins using a

gel retardation assay..................................................................................95

3.1 Results.................................................................................................................... 963.1.0 ATF-1 and ATF-2 bind specifically to ATF binding sites

present in various adenovirus early promoters.................................963.1.1 The effect of point mutations on the binding specificity of

ATF-1 and ATF-2.......................................................................................983.1.2 The effect of point mutations on the stability of ATF-2

binding.........................................................................................................1023.1.3 The effect of point mutations on the binding of ATF-2 to

the E4 promoter in vivo............................................................................. 104

3.2 D iscussion............................................................................................................ 109

Chapter 4: Analysis of E4 promoter activity in vitro..........................................114

4.0 Methods and Materials....................................................................................... 1174.0.1 In vitro transcription of the E4 promoter...............................................1174.0.2 Expression and purification of recombinant proteins.......................118

4.1 Results.................................................................................................................... 1204.1.1 Reduction of E4 transcription by the addition of a

competitor oligonucleotide containing an ATF bindingsite ................................................................................................................ 120

4.1.2 Analysis of the effect of preincubation of the nuclear extract with the competitor oligonucleotide on depletionof E4 promoter activity.............................................................................120

4.1.3 The effect of glutathione, thrombin and imidazole uponthe in vitro transcription reaction.........................................................123

4.1.4 Recombinant CREB restores transcription of theE4 promoter by ATF depleted nuclear extracts.................................... 126

4.1.5 A mutant recombinant CREB protein, HôCRlAP, cannot restore transcription of the E4 promoter by ATF depletednuclear extracts...........................................................................................128

4.2 Discussion............................................................................................................ 133

Chapter 5: Discussion................................................................................................. 136

5.1 The relative roles of E4F and ATF-2 in trans-activation of theE4 promoter by ElA.............................................................................................. 1365.1.1 Problems associated with interpreting the results of this

study............................................................................................................. 1385.2 Factors affecting the requirements for trans-activation by E lA ................139

5.2.1 Regulation of other factors by ElA..........................................................1395.2.2 Dimerization of ElA inducible transcription factors......................... 140

5.3 The mechanisms of E4F and ATF-2 mediated trans-activation ofthe E4 promoter by ElA....................................................................................... 141

5.4 Other examples of multiple factors binding to individualpromoter elements................................................................................................ 143

5.5 The role of ATF proteins in other viral trans-activationpathways.................................................................................................................144

5.6 Conclusions......................................................................................................... 146

REFERENCES..............................................................................................................147

Appendix: Abbreviations not described in the text............................................. 168

FIGURES

Figure 1.1: The Ad2 cytoplasmic RNA transcripts, characterised byelectron microscopy of RNAiDNA heteroduplexes................................. 15

Figure 1.1.1: Structures of the transcribed and translated products ofthe Ad2 ElA region..........................................................................................17

Figure 1.4.2.2: Schematic representation of the transcriptionalinitiation of the adenovirus major late promoter.................................... 24

Figure 1.5: General structure of the products of the adenovirus ElA region. A; Structure of the 12S and 13S mRNAs and theirprotein products. B; Structure of CR3 of the large ElA protein.............29

Figure 1.6.2: Summary of results from DNase I footprinting and méthylation interference assay analysis of the adenovirusE4 promoter........................................................................................................40

Figure 1.7.1.3: Amino acid homology in the basic domain andleucine zipper regions of several members of the ATF family and selected other proteins that are known to interact with theE4 promoter........................................................................................................44

Figure 1.8.1: Méthylation interference analysis of E4F and ATFbinding to the -39/-56 site of the E4 promoter............................................. 51

Figure 2.1.0 A: Structure of the E4A38, E4CAT, VIPE4 and E2CATtest promoters......................................................................................................70

Figure 2.1.0 B: Activation of synthetic promoters by ElA;Representative CAT assays...............................................................................71

Figure 2.1.0 C: Activation of synthetic promoters by ElA;Quantitation of CAT assays.............................................................................72

Figure 2.1.1 A: Structure of the SplCAT test promoter...................................... 74Figure 2.1.1 B: Activation of the synthetic promoter SplCAT by

ElA; Representative CAT assays....................................................................75Figure 2.1.1 C: Activation of the synthetic promoter SplCAT by

ElA; Quantitation of CAT assays...................................................................76Figure 2.1.2 A: Structure of the IE4 and WE4 series of promoters.................79

10

Figure 2.1.2 B: Effects of point mutations on E4 promoter activity;Representative CAT assays.............................................................................80

Figure 2.1.2 C: Effects of point mutations on E4 promoter activity;Quantitation of CAT assays...........................................................................81

Figure 2.1.2 D: Sequence differences between pE4CAT andpWE4WT............................................................................................................ 82

Figure 2.1.4 A: Effect of cell type on the requirement for E4F for activation of the E4 promoter by ElA; Representative CATassays..................................................................................................................85

Figure 2.1.4 B: Effect of cell type on the requirement for E4F for activation of the E4 promoter by ElA; Quantitation of CATassays..................................................................................................................86

Figure 3.1.0: A) Binding of ATF-1 to adenovirus early promoters.B) Binding of ATF-2 to adenovirus early promoters..................................97

Figure 3.1.1 A: Effects of point mutations on ATF-1 binding.......................... 99Figure 3.1.1 B: Effects of point mutations on ATF-2 binding..........................100Figure 3.1.2: Stability of ATF-2 DNA binding to point mutated

ATF/E4F binding sites.................................................................................... 103Figure 3.1.3 A: Activation of the IE4xlO series of promoters by an

ATF-2/VP16 fusion protein; Representative CAT assays........................105Figure 3.1.3 B: Activation of the IE4xlO series of promoters by an

ATF-2/VP16 fusion protein; Quantitation of CAT assays....................... 106Figure 3.2: Comparison of the zinc finger motif of ATF-2 with

several representative C2 H2 zinc fingers...................................................... 112Figure 4.1.1: Transcription of the E4 promoter in vitro is inhibited

by the addition of a competitor DNA containing an ATFbinding site..........................................................................................................121

Figure 4.1.2: Analysis of the effect of preincubation of the nuclear extract with the ATF-BS competitor oligonucleotide ondepletion of E4 promoter activity..................................................................122

Figure 4.1.3 A: Effect of glutathione and thrombin on transcriptionof the E4 and MLP promoters in vitro..........................................................124

Figure 4.1.3 B: Effect of imidazole on transcription of the E4 andMLP promoters in vitro...................................................................................125

Figure 4.1.4: Recombinant CREB (H^CRla) restores transcription of the E4 promoter by ATF depleted nuclear extracts, but the CREB dependent stimulation of transcription is not enhanced by protein kinase A (PKA)...............................................................................127

11

Figure 4.1.5 A: Purification of recombinant CREB proteins.A; HéCRla. B; HôCRlAF.................................................................................129

Figure 4.1.5 B: Analysis of the DNA binding activity ofrecombinant CREB proteins............................................................................ 130

Figure 4.1.5: A mutant recombinant CREB protein, H^CRIAP, cannot restore transcription of the E4 promoter by ATF depleted nuclear extracts.................................................................................131

TABLES

Table 1.7.1.3: The relationship between cloned bZIP proteinsbelonging to the ATF family of transcription factors..............................45

Table 1.8.1: Binding specificities of ATF and E4F for binding sitesfrom various adenovirus early promoters................................................53

12

Chapter 1: Introduction.

1.0 General introduction.

The rate and duration of transcription most frequently underlies

the differential rate of protein synthesis in eukaryotes, as it does in

prokaryotes. The relative abundance of a particular RNA transcript

directly reflects the number of translational events occurring. The central

role of transcription in the process of gene expression makes it the most

attractive control point for regulating the expression of genes; it is now

clear that in the vast majority of cases where a particular gene is expressed

only in a particular tissue or in response to a particular signal, this control

is achieved at the transcriptional level (Darnell et ah, 1986).

To begin to unravel such complex events requires the use of

simpler systems that reproduce at least the basic principles of the more

complex regulatory pathways. Researchers began studying adenovirus

gene expression in the early 1970s because it provided a good model

system for the study of eukaryotic gene expression. The adenovirus-

infected cell was considered a particularly attractive system to investigate

because the viral genome is small enough to be analysed in detail (35 kb)

yet large enough that one can expect the expression of the genome to be

regulated during the course of infection. Secondly, adenoviruses such as

type 2 (Ad2) and type 5 (Ad5) were found to be easily propagated in spinner

cultures of HeLa cells and easy to purify. In addition relatively large

quantities of viral RNA are produced during an adenovirus infection.

Thirdly, it was expected that transcription of adenovirus genes would

involve cellular processes and would mimic the host cell in its regulation

of gene expression.

13

In effect the adenovirus genome was likened to a miniature

chromosome and information gathered from the study of adenovirus

gene expression was expected to be generally applicable to gene expression

in eukaryotes. These expectations have to a large extent been fulfilled,

w ith many fundamental discoveries regarding eukaryotic gene

organisation and expression being made as a result of research on the

adenoviruses.

1.1 The adenovirus replication cycle.

Expression of adenovirus genes during a productive infection of

permissive human cells proceeds through two distinct phases, early and

late, which are defined by their punctuation by viral DNA synthesis. This

latter event occurs between 6-9 h after infection under typical conditions of

infection. During the early phase of infection before the onset of viral

DNA replication, several regions of the viral genome are expressed which

give rise to five major classes of mRNA; ElA, ElB, E2A, E3 and E4 (Chow

et ah, 1979; Kitchingman and Westphal, 1980; Nevins et ah, 1979; Shaw

and Ziff, 1980; for review, see; Flint, 1986; see Figure 1.1).

1.1.1 The ElA region.

The first region of the adenovirus genome expressed immediately

following the entry of the viral DNA into the infected cell nucleus is the

E lA region (N evins et ah, 1979; Shaw and Ziff, 1980). RNA

complementary to this region is detected 45 min after infection, with a

maximal rate of RNA synthesis reached after 3 h and maintained

thereafter for at least 6 h (Nevins et ah, 1979). The ElA gene products are a

heterogeneous family of polypeptides that arise from both alternative

14

a i

Proteins

RNAs

13 K 21 K 2 6 K 16 K 32 K 55 K

IX

13.6 K

PTNase

52- penton hexon55 K ilia III pVII V pVI II 23 K

non-viriont *--------\

33 K

14K 14.5 K 11 K 10K

1 2

100 K pVIII 16 K

x y ;

ElAE1B_

fiberIV

L5

L4

L3

L2

L1 E3_ A _

r-strand 3’l-strand 5’

RNAs

[= îVA

I10

I20

I30

I40 50 60

I70

I80 90

3'100

- ]- ]

—Y—E2A

E2B

Proteins IVa2 140 K pol

87 K p-TP

72 K DBP

~E4

11 K 13K 17K 10K 14K(19K .21 K, 2 4 K .3 5 K )

Figure 1.0.2: The Ad2 cytoplasmic RNA transcripts, characterized by electron microscopy of RNAiDNA heteroduplexes. (From Flint, 1986)

splicing of the primary transcript and extensive post-translational

modification of the primary products (Berk and Sharp, 1978; Stephens

et al., 1986). Five different mRNAs from the Ad2 and Ad5 ElA gene have

been identified, having sizes of 9 S, 10 S, 11 S, 12 S and 13 S (Berk and

Sharp, 1978; Chow et ah, 1979; Kitchingman and Westphal, 1980; Stephens

and Harlow, 1987; Ulfendahl etah, 1987; see Figure 1.1.1). The four largest

mRNAs represent differentially processed mRNAs that encode different

portions of a single open reading frame. The most abundant of these are

the 12 S and 13 S transcripts, which are synthesised early in infection and

encode proteins of 243 and 289 amino acids respectively (Perricaudet et ah,

1979). The other products are detectable in the late phase (Stephens and

Harlow, 1987; Ulfendahl etah, 1987) whereas the product of the 9 S

transcript has not yet been found in infected cells (Virtanen and Peterson,

1983).

Viruses carrying mutations that impair or prevent the expression of

E lA gene products have been shown to produce much reduced

concentrations of viral early mRNAs (Berk et ah, 1979; Jones and Shenk,

1979). A similar situation is observed when protein synthesis in cells

infected with the wild type virus is inhibited from the beginning of

infection (Lewis and Mathews, 1980; Shaw and Ziff, 1982). Thus, a product

of the ElA region must be essential for the activation of transcription of all

other transcription units expressed during the early phase of infection.

Subsequent experiments have established that products encoded by the

ElA region greatly stimulate transcription from all of the adenovirus early

promoters, a phenomenon termed trans-activation (for reviews see; Berk,

1986; Flint and Shenk, 1989; Flint, 1986). The mechanisms of this function

of the ElA gene products will be discussed in more detail in a later chapter.

16

500 1000I

1500 J l _

13S 185 aa 104 aa AAAA

12S AAAA139 aa

AAAA87 aa 104 aa

lOS AAAA104 aa41 aa

U S 26 aa

26 aa

104 aa

9 S 126 aah aa ■AAAA

Figure 1.1.1: Structure of transcribed and translated products of the adenovirus ElA region. The structure of the five ElA mRNAs that are synthesized by the alternative splicing of a primary transcript are shown by the relative positions of their coding sequences in the viral genome (numbered). Boxes indicate open reading frames, and dotted lines represent sequences removed during splicing. The shaded box indicates a different reading frame to the C-terminus of the other mRNAs. Erom Dyson and Harlow, (1992).

17

1.1.2 Activation of adenovirus early genes.

Transcription of the other early regions begins slightly later than

that of ElA, starting with the ElB region approximately 60 min after

infection. The E3 and E4 regions are expressed simultaneously at about

90 min after infection and finally the last early region to be expressed is the

E2 region (Nevins et al., 1979). Full activation of E2 transcription is

dependent on the function of a 19 kDa protein encoded by E4 orf6/7 (Cutt

et al., 1987) and this product can function independently of other viral

gene products to frflws-activate the E2A promoter (Hardy et ah, 1989;

Huang and Hearing, 1989). RNA from the E2 region is produced

approximately 2 h after infection, 30 min later than that of the E4 region

(N e v in s etah, 1979). Transcription of all of the early regions then

accelerates during the next few hours of the early phase. The E3 and E4

regions reach a maximal rate of transcription approximately 3 h after

infection, the E2 region 7 h after infection, and the ElB region 8 h after

infection (Nevins et ah, 1979).

The regulation of adenovirus gene expression that is demonstrated

by the above observations is likely to rely upon cellular factors and

presumably reflects cellular modes of regulation. Therefore trans-

activation of the adenovirus early genes by ElA represents an ideal model

system for investigating the mechanisms by which gene expression can be

modulated.

1.2 Mammalian transcription.

Transcription of mammalian genes is performed by three different

RNA polymerases, designated I, II and III (Roeder and Rutter, 1969; Roeder

and Rutter, 1970a; Roeder and Rutter, 1970b). Each polymerase catalyses

18

transcription of genes encoding different classes of RNA. RNA

polymerase I is located in the nucleus and is responsible for synthesis of

ribosomal precursor RNAs. RNA polymerase III functions outside the

nucleus and transcribes the genes for tRNA, 5S rRNA, and an array of

small RNAs. RNA polymerase II is located within the nucleus and

catalyses transcription of all the protein-coding genes; that is, it functions

in the production of mRNAs. In addition, polymerase II also produces

several small RNAs that take part in RNA processing (for review, see;

Adams et al., 1986; Lewis and Burgess, 1982; Sentenac, 1985).

1.3 Control of RNA polymerase II transcription.

The rate of transcription of a gene is controlled by a region of DNA

5' (and occasionally 3') to the transcription initiation site called the

promoter. Mutagenesis studies have demonstrated that the promoters of

genes transcribed by RNA polymerase II usually consist of two functional

elements: A TA-rich region (the TATA box) and one or more upstream

promoter elements (for review, see; Breathnach and Chambon, 1981;

Maniatis et ah, 1987; Mitchell and Tjian, 1989).

1.3.1 The TATA box.

The TATA box is a TA-rich region (consensus TATA^/jA^/t)

located approximately 30 bp upstream from the transcription start site in

many but not all genes (for review, see; Breathnach and Chambon, 1981)

and is the binding site for a 38 kDa protein called the TATA binding

protein (TBP; Hoffmann et ah, 1990; Kao et al, 1990; Peterson et al, 1990).

The interaction between TBP and the TATA box is the first step in the

initiation of transcription, which results in the formation of a stable

(template-committed) and transcriptionally active complex that includes

19

RNA polymerase II (Davison et ah, 1983; Fire et al, 1984; see below).

Deletion or mutation of the TATA box results in a dramatic decrease in

the activity of many promoters in vivo (for example, see; Corden et ah,

1980; Wasylyk and Chambon, 1981) and experiments involving its

relocation have demonstrated that it is an important determinant in the

accurate positioning of the start site of transcription (for example, see;

C orden et ah, 1980). The TATA box is the minimal promoter element

necessary for the initiation of transcription in vitro and appears to be

sufficient to direct a low unregulated (basal) level of transcription (for

reviews, see; Roeder, 1991; Sawadogo, 1990; Weinmann, 1992).

Although most polymerase II promoters contain a consensus TATA

sequence, there are many examples of functional promoter elements that

either contain a TA-rich sequence that varies from the consensus or no

apparent TATA box at all (Breathnach and Chambon, 1981). However, the

yeast TFIID complex has been shown to bind with high affinity to such

non-consensus TATA sequences (Hahn et ah, 1989) and promoters that

lack any homology to the TATA box have been shown to nevertheless

require TBP for the initiation of transcription (Pugh and Tjian, 1991).

Hence the requirement for a consensus TATA box is gene specific and is

most likely dependent upon the presence of either additional promoter

elements such as Spl binding sites (Dynan and Tjian, 1983) or an initiator

element (Smale and Baltimore, 1989). This element was first identified in

some promoters that lack a TATA box and in its absence fixes the start site

of transcription and facilitates the binding of TBP to the promoter

(Carcamo et ah, 1991; Zenzie-Gregory et ah, 1992).

1.3.2 Upstream promoter elements.

Upstream promoter elements are sequences generally between

6 and 20 bp long, often have dyad symmetry and are the binding sites for

20

sequence-specific DNA binding proteins called transcription factors. These

cfs-control elements are located at varying distances upstream of the

TATA box and reversibly enhance or repress the basal level of

transcription (Maniatis et al., 1987; Mitchell and Tjian, 1989).

Upstream promoter elements that are involved in distinct signal

transduction pathways were amongst the first to be extensively

characterised, many of which have since been shown to be present in a

number of cellular and viral promoters. For example, analysis of the

signal transduction pathway activated by cyclic AMP (cAMP) identified an

element in the proenkephalin promoter called the cAMP response

element (CRE; Comb et ah, 1986). Examination of the promoters of many

genes induced by cAMP, such as somatostatin, vasoactive intestinal

polypeptide (VIP), tyrosine hydroxylase and c-fos reveals sequences

homologous to the enkephalin CRE that are limited to a 5 bp TGACG

motif and which are also functionally important for the activation of

those genes by cAMP (Harrington et al, 1987; Montminy and Bilezikjian,

1987; Sassone-Corsi et al., 1988; Tsukada et al., 1987).

1.4 Cellular factors involved in transcription by RNA polymerase II.

1.4.1 RNA polymerase II.

Eukaryotic RNA polymerase II contains two large polypeptide

subunits of 140 and 220 kDa and 10±2 smaller subunits of between 10 and

45 kDa (Armaleo and Gross, 1985; Engelke et al., 1983; Guilfoyle et al., 1984;

Renart et al., 1985; Sentenac, 1985). The two large subunits share several

conserved regions with polymerases I and III (in which 70-80% of the

amino acids are identical) and with the p and p' subunits of the E. coli

RNA polymerase (Allison et al., 1985; Memet et al., 1988; Sweetser et al.,

21

1987). Three of the smaller subunits (14 to 28 kDa) are also found

associated with polymerases I and III, and are hence called common or

shared subunits (Woychik et al., 1990; for review, see; Sawadogo, 1990;

Woychik and Young, 1990; Young, 1991).

The largest subunit of RNA polymerase II contains an unusual

repetitive domain that is not found in the homologous large subunits of

polymerases I and III, or in the E. coli p' subunit (Allison etal., 1985;

Corden et al, 1985). This C-terminal domain (CTD), or tail, consists of a

tandem heptapeptide repeat of the consensus sequence Tyr-Ser-Pro-Thr-

Ser-Pro-Ser. This domain is highly phosphorylated in a substantial portion

of the RNA polymerase II molecules in a mammalian cell and protein

kinases have been purified that appear to be involved in CTD

phosphorylation (Cisek and Corden, 1989; Feaver et ah, 1991; Lee and

Greenleaf, 1989; Stevens and Maupin, 1989). The CTD has been shown to

be essential for initiation from some (but not all) promoters tested, but

does not appear to be required for elongation (for review see; Young, 1991).

Phosphorylation of the CTD appears to occur during initiation and is

postulated to facilitate the transition between promoter binding and RNA

elongating forms of RNA polymerase II (Laybourn and Dahmus, 1989;

Payne et ah, 1989). In addition, several lines of evidence suggest that this

domain is one of several components of the transcription apparatus that

respond to activating signals from factors associated with upstream

promoter elements (Allison and Ingles, 1989; Sigler, 1988; for review, see;

Corden, 1990; Young, 1991).

1.4.2 General transcription (initiation) factors.

The general transcription factors are defined as those factors which

are required for transcription of all Pol II genes. Together with the RNA

22

polymerase, these factors interact with core promoter elements to form a

productive complex that directs a low unregulated (basal) level of

transcription. By reconstituting the transcriptional activity of a promoter

in vitro by the addition of increasingly purified fractions of whole cell or

nuclear extracts, a number of factors required for the initiation of

transcription by RNA polymerase II have been identified. By examining

the properties of these factors the formation of the preinitiation complex is

now well defined and involves the stepwise assembly of several factors,

including RNA polymerase II, into a transcriptionally active initiation

complex (for reviews, see; Roeder, 1991; Sawadogo, 1990; Weinmann, 1992;

see Figure 1.4.2.2).

1.4.2.1 The TATA binding protein.

The TATA binding protein (TBP) is a 38 kDa protein that binds to

the TATA consensus sequence motif (Peterson et al, 1990) and is tightly

associated with several other polypeptides (TBP associated factors, or

TAFs) to form the multiprotein complex TFIID (Dynlacht e t a l , 1991;

Tanese e ta l , 1991). TBP cannot substitute for native TFIID in activator

dependent transcription, but can mediate preinitiation complex assembly

and basal (core promoter) transcription in concert with other general

initiation factors (Peterson et al, 1990). It interacts directly with the general

initiation factors TFIIA and TFIIB, the C-terminus of the large subunit of

RNA polymerase II and an initiator binding factor (TFII-I) that may be

important for transcription from TATA-less promoters (Carcamo et a l ,

1991; Smale e t a l , 1990; Zenzie-Gregory e t a l , 1992; for review, see;

Weinmann, 1992; see Figure 1.4.2.2).

23

T fllD V______ /

TATACAP

TFIIA I J TFIIB

POL II7777TFIIDTFIIA TFIIB

T F I H

POL II

TFIIA I J TFIIB

POL IITFIIDTFIIA TFIIB

CAP

Figure 1.4.2.2: Schematic representation of the transcriptional initiation of the adenovirus major late promoter. The diagram summarises the order of events that lead to the formation of a functional intiation complex and does not represent the position, size or shape of the complex at the promoter. Adapted from Weinmann (1992).

24

1.4.2.2 Other factors involved in assembly of the preinitiation complex.

The first step in the assembly of the preinitiation complex involves

the binding of the TFIID complex (which includes TBP; see above) to the

TATA box (Davison et al , 1983; Fire e ta l , 1984; see Figure 1.4.2.2). This

interaction results in a template-committed complex (a stable complex

resistant to challenge by an equivalent protein-free template; Davison

et a l , 1983; Fire et ah, 1984) which nucleates the ordered assembly of the

other general transcription factors TFIIA, TFIIB, TFIIE and TFIIF, together

with RNA polymerase II, into an active transcription complex (Buratoski

et ah, 1989; Flores et ah, 1988; Inostroza et ah, 1991; Maldonado et ah, 1990;

for review, see; Roeder, 1991; Weinmann, 1992 ; summarised in

Figure 1.4.2.2). Subsequent events involving formation of the initiation

complex and leading to RNA elongation are less well defined, but may

involve at least another four activities (TFIIG (Sumimoto et ah, 1990),

TFIIH and TFIIJ (Flores et ah, 1992) and TFIIS (Reinberg and Roeder, 1987).

1.4.3 Gene-specific transcription factors.

Transcription factors that bind to upstream promoter elements are

gene-specific; that is they are involved in the selective activation of

particular genes through their exclusive interaction with a small subset of

these short DNA elements (Dynan and Tjian, 1985; Mitchell and Tjian,

1989). Particular gene-specific transcription factors are involved in a

variety of regulatory pathways, including tissue-specificity of gene

expression (for example, the lymphoid cell-specific expression of

immunoglobulin genes; Staudt et ah, 1986; Lenardo et ah, 1987; Pierce

et ah, 1988) and cellular response to outside stimuli such as heat-shock

(Pelham, 1982; Kingston et ah, 1987) or hormones (Miesfeld et ah, 1984).

25

Many transcription factors have now been cloned and analysis of

their cDNAs has revealed several structural features common to this

category of proteins. Typically, they can be dissected into two functional

regions; a domain responsible for sequence-specific DNA binding and in

many cases dimérisation; and one or several discrete domains involved in

transcriptional activation (for review, see; Mitchell and Tjian, 1989).

Studies of the yeast activator protein Gal4 have demonstrated that this

modular structure is remarkably flexible, with domains from different

proteins generally being inter-changeable with those of Gal4 with no

apparent loss of function (Brent and Ptashne, 1985; Keegan et al., 1986).

Many transcription factors bind to DNA as dimers and employ a

limited number of structural motifs to achieve dimérisation and DNA

binding. As a result, they are often classified according to which motifs

com pose the dimérisation and /or DNA binding domain. DNA

recognition motifs include the basic domain (Johnson et ah, 1987;

Landschulz et at., 1988) and zinc fingers (Miller et ah, 1985). Dimérisation

regions include the leucine zipper (Landschulz et ah, 1988) and

helix-loop-helix motifs (Steitz et ah, 1982) and both of these regions often

adjoin a basic (DNA recognition) domain (for review, see; Busch and P.,

1990; Struhl, 1989).

The activation domains of transcription factors are presently

classified according to their amino acid composition, and regions that are

rich in acid amino acids {e.g. VP16 (Triezenberg et ah, 1988)), proline

{e.g. CTF/NFI (Mermod et ah, 1989)) or glutamine {e.g. SPl (Courey and

Tjian, 1988)) have been identified. The best studied example of an

activation domain is the acidic activating domain of the herpes simplex

virus VP16 protein (Triezenberg et ah, 1988). Mutagenesis experiments

have confirmed that the acidic nature of VP16 is important for its

potential to stimulate transcription, but that several bulky hydrophobic

26

residues {e.g. Leu, lie, Phe) in the activation domain are required

absolutely (Cress and Triezenberg, 1991). The acidic activation domain of

VP16 has been shown to directly interact with the general transcription

factors TFIID (Stringer et al., 1990) and TFIIB (Lin et al, 1991) and these

interactions are not simply ionic in nature. Mutations of a phenylalanine

residue in the activation domain that have no net effect on the charge of

the region but which eliminate the activation function (Cress and

Triezenberg, 1991) are detrimental to the interaction between VP16 and

either TFIIB (Lin et a l , 1991) or TFIID (Ingles et a l , 1991). Hence these

results are interpreted to infer a direct mechanism for the activation of

transcription by acidic domains like that of VP16.

Other reports have suggested a role in activated transcription for a

class of non-DNA binding proteins that are neither transcriptional

activators nor general factors (Berger et al, 1990; Hoey et al, 1990; Kelleher

et a l , 1990; Pugh and Tjian, 1990). These novel molecules, variously

termed adaptors, coactivators, or mediators, are envisioned to bridge or

stabilise interactions between activators bound at upstream promoter

elements and the general factors bound at the TATA box (Pugh and Tjian,

1990; for review, see; Gill and Tjian, 1992).

1.5 Trflws-activation by adenovirus El A proteins.

Adenovirus ElA proteins represent another class of transcriptional

activators that neither bind DNA in a sequence-specific manner nor target

a unique promoter element in any other way; a property shared by several

other viral frflws-activating proteins {e.g. pseudorabies virus immediate

early proteins (Abmayr et al, 1985); HTLV-1 p40* (Fujisawa et a l , 1989);

HBV X gene product (Maguire et al, 1991).

27

The first indication that ElA regulates the expression of the early

adenovirus genes came from the analysis of mutant ElA viruses. Viruses

whose E lA regions are deleted, or more subtly mutated, support the

efficient expression of no regions of the viral genome other than ElA

itself, when present (Berk et ah, 1979; Jones and Shenk, 1979). Subsequent

experiments have established that this control is at the level of

transcription and that ElA acts in trans on the viral promoters (Flint,

1986). Trflns-activation by E lA is not limited to adenovirus genes;

transcription of some cellular genes ((3-globin and preproinsulin) is

stimulated when plasmid clones of the genes are co-transfected into HeLa

cells with a plasmid expressing ElA (Gaynor et ah, 1984; Green et ah, 1983).

Curiously, the endogenous counterparts of these cellular genes are not

activated (Gaynor et ah, 1984; Svensson and Akusjarvi, 1984) and other

experiments have concluded that transcription of most endogenous genes

is not induced by ElA (Beltz and Flint, 1979). However, it does appear that

ElA can stimulate transcription from the promoters of at least two

endogenous cellular genes, p-tubulin (Stein and Ziff, 1984) and hsp70 (Kao

and Nevins, 1983). The diversity of ElA's frans-activating properties is

further emphasised by its ability to stimulate transcription by RNA

polymerase III (Gaynor et ah, 1985; Hoeffler and Roeder, 1985).

In addition to activating transcription, ElA proteins have also been

demonstrated to effectively repress SV40 or polyoma virus enhancer-

mediated transcription (Borrelli et ah, 1984; Velich and Ziff, 1985). The

enhancers negatively regulated by ElA include not only those that are

active in a wide variety of cells, but also cell-specific enhancers, such as the

mouse immunoglobulin heavy and light chain enhancers, active only in

lymphoid IB cells (Hen et ah, 1985). In contrast, the same enhancer of the

mouse immunoglobulin heavy chain gene is activated by E lA in

nonlymphoid cells (Borrelli et ah, 1986).

28

12SmRNA

13SmRNA

sssssssCRl CR2

im ECRl CR2 CR3

243 aa Phosphoprotein

289 aaPhosphoprotein

6

CRl CR2 CR313S ElA 289 aa

139 154 157 171 174 188

Figure 1.5: General structure of the products of the adenovirus ElA region. A; Structure of the 12S and 13S mRNAs and their protein products. B; Structure of CR3 of the large ElA protein. The encircled Cs denote the cysteine residues that form the zinc finger motif, chelating a single Zinc ion (Zn). Numbers below the figure denote the amino acid position in ElA.

29

How the ElA protein is able to accomplish these diverse forms of

transcriptional control is not clear. The promoters activated by ElA do not

share one common czs-acting DNA sequence motif and consistent with

this lack of sequence specificity in function, ElA proteins have only a very

low affinity for DNA and no apparent sequence specificity (Chatterjee

et al., 1988; Zu et ah, 1992). In addition, mutation or deletion of the region

of ElA responsible for this weak DNA binding activity has no effect on the

ability of ElA to frans-activate the adenovirus E4 promoter (Zu et al.,

1992). Systematic mutagenesis of the adenovirus early promoters has not

revealed promoter sequences required exclusively for ElA trans-activation

(Berk, 1986). Instead, the promoter sequences required for the maximal

level of ElA activated transcription appear to be also required for the

maximal basal level of transcription in the absence of ElA (Berk, 1986).

Such observations have led to the conclusion that ElA acts by a

mechanism that must be both indirect and general.

1.5.1 A unique conserved region (CR3) of the large ElA protein is essential

for trans-activation by ElA.

During the early phase of a productive adenovirus infection, the

ElA region gives rise to two major, differentially spliced mRNAs (Berk

and Sharp, 1978; see Figure 1.5 A). These 12 S and 13 S mRNAs encode two

closely related nuclear-localised phospho-proteins of 243 and 289 amino

acids respectively, that differ only in an internal 46 amino acid region that

is unique to the larger ElA protein (Perricaudet et al, 1979; see Figure 1.5 A

and B). This region is one of three regions that are conserved among

distantly related human adenoviruses and has been termed conserved

region 3, or CR3 (Kimelman et al., 1985; van Ormondt et al , 1980).

Trans-activation of the early viral promoters was found to be an

activity of the product of the 13 S mRNA but not the 12 S mRNA (Carlock

30

and Jones, 1981; Esche et ah, 1980; Glenn and Ricciardi, 1985; Graham et ah,

1978; Lillie et ah, 1986; Montell et ah, 1982; Ricciardi et ah, 1981) and several

mutational studies have confirmed the importance of the unique region

of the 289 amino acid ElA protein (CR3) to its frans-activation activity.

Mutations in other regions of the large ElA protein, including the two

other conserved regions (CRl and CR2), have little or no effect on the

frflns-activating activity of the large ElA protein, whereas mutations in

CR3 have been found to severely impair frans-activation (Glenn and

Ricciardi, 1985; Lillie et ah, 1986; Moran et ah, 1986; for review, see; Moran

and Mathews, 1987).

However, some evidence for transcriptional activation by products

of the ElA 12 S mRNA has been reported. Although most viruses

expressing only the 12 S mRNA grow poorly on HeLa cells (Haley et ah,

1984; Montell et ah, 1984) the cDNA virus dl347 grows well and induces

the synthesis of larger quantities of early mRNA species than does the

virus dl312 (whose ElA region has been completely deleted), although

resulting in a fivefold reduced yield of virus compared to wild type

infection (Winberg and Shenk, 1984). The 12 S ElA mRNA product has

also been observed to stimulate transcription of the E2 and E3 regions

when plasmids containing these genes are co-transfected into HeLa cells,

although less efficiently than the ElA 13 S mRNA products (Leff et ah,

1984). In addition, the 243 amino acid ElA protein produced in Escherichia

coli can complement dl312, although to a lesser extent than the bacterially

produced 289 amino acid protein (Ferguson et ah, 1985). Hence it seems

clear that activation of early viral genes during a productive infection is

largely a function of the 289 amino acid ElA protein, although the 243

amino acid E lA protein can activate transcription under certain

circumstances.

31

1.5.2 CR3 contains one copy of a C4 zinc finger motif that is essential and

sufficient for trans-activation.

Four conserved cysteine residues of CR3 constitute one copy of a

motif that bears homology to a C4 zinc finger (see Figure 1.5 B). The 289

amino acid ElA protein binds one mole Zn^+/mole protein and point

mutations in the conserved cysteines that form the zinc finger prevent

zinc binding and trans-activation (Culp et ah, 1988), indicating that the

integrity of this domain is important to both of these processes.

Remarkably, a 49 amino acid synthetic peptide that comprises the 46

amino acids of CR3, plus three adjacent and conserved amino acids, is able

to stimulate transcription from the adenovirus E2A promoter in vivo

(Green et ah, 1988; Lillie et ah, 1987; Pusztai et ah, 1989) and from the MLR

and E3 promoters in vitro (Green et ah, 1988; Loewenstein and Green,

1989). Mutational analysis of this peptide has demonstrated that the

conserved cysteine residues present in CR3 are essential for its activity

(Green et ah, 1988; Loewenstein and Green, 1989).

1.5.3 CR3 consists of two subdomains.

Further analysis of the function of both the zinc finger and adjacent

regions of CR3 has been performed utilising transient expression assays.

These involve the co-transfection of a test promoter with a plasmid

expressing the ElA gene and enable a quantitative estimation of the effects

of mutations in these regions. Lillie and Green (1989) demonstrated that a

chimeric protein, consisting of the DNA binding domain of the yeast

activator Gal4 fused to amino acids 121-223 of ElA, can activate

transcription from a promoter containing Gal4 binding sites following

their co-transfection into CHO cells. In contrast, a Gal4-E1A fusion protein

containing amino acids 150-223 of ElA cannot activate transcription, but

this activity can be rescued by the addition of the VP16 acidic activating

32

domain to the fusion protein (Lillie and Green, 1989). These results

demonstrate that ElA behaves in a similar manner to a promoter bound

activator and suggest that the region between amino acids 121-150 of ElA,

which includes the N-terminus of CR3, includes an activating domain.

Utilising a similar assay, Martin et al. (1990) found that the

boundaries of the ElA activating domain coincide with the N-terminus of

CR3 and the C-terminus of the zinc finger, and characterised point

mutations in this region that reduce or eliminate the activating function

of ElA. In addition, a Gal4-E1A fusion protein was found to inhibit

activation of a promoter containing LexA binding sites by a LexA-ElA

fusion protein. This inhibition can be removed by the same mutations in

Gal4-E1A that affect activation of the promoter containing Gal4 binding

sites (Martin et ah, 1990). Hence ElA is postulated to contain an activating

domain that directly interacts with a component of the transcriptional

machinery and inhibition by an heterologous ElA fusion protein is likely

to be due to the specific sequestering (or squelching) of this cellular target

(Martin et al., 1990).

Like the wild type ElA protein, the Gal4-E1A fusion protein also

efficiently activates the E4 promoter in this assay (Lillie and Green, 1989).

However, a second class of mutation has been characterised that destroys

the ability of Gal4-E1A to activate the E4 promoter, but which does not

affect activation of the core promoter containing Gal4 binding sites (Lillie

and Green, 1989). In contrast to the effects of mutations in the activating

domain, the effects of these mutations cannot be rescued by the addition to

the fusion protein of the VP16 acidic activating domain. These mutations

include a non-conserved point mutation in the C-terminal region of CR3

at amino acid 180, which further distinguishes this domain from the

activating domain which is localised in the N-terminal region of CR3.

These observations are consistent with this domain being required for the

33

targeting of ElA to the promoter, and taken together with the above

results, ElA is suggested to be able to function in the vicinity of the

promoter by interacting with a factor(s) already bound to it.

These results are mirrored by the observation that mutations in the

postulated promoter targeting domain of ElA exhibit a negative trans-

dominant phenotype over the wild type protein (Glenn and Ricciardi,

1985; Glenn and Ricciardi, 1987; Webster and Ricciardi, 1991). Webster and

Ricciardi (1991) performed an extensive mutational analysis of CR3 and

defined several mutations in ElA that are defective in their trans-

activation of the adenovirus E3 promoter in vivo as determined by

co-transfection in a transient expression assay. These mutant proteins

were also tested for their ability to inhibit trans-activation by the wild type

protein upon their co-transfection, an effect termed trans-dom inance.

Trans-dominant mutations were found to map to a discrete region

C-terminal to the zinc finger (Webster and Ricciardi, 1991) that

corresponds to the promoter targeting domain described above (Lillie and

Green, 1989; Martin et al, 1990). Deletion of this region produces a very

strong trans-dominant mutation, indicating that the activating domain

alone is required for trans-dominance (Webster and Ricciardi, 1991).

In contrast to mutations in the promoter targeting domain,

mutations in the activating domain and zinc finger motif of ElA do not

exhibit trans-dominance. In addition, every mutation that destroys trans

activation without displaying trans-dominance, when introduced as a

second-site substitution into a trans-dominant mutant, destroys the trans

dominant phenotype (Webster and Ricciardi, 1991). The phenotypic

disparity between the activating domain and the promoter targeting

domain mutants indicates that they are distinct functional domains

within the trans-activating domain of ElA. In agreement with the

hypothesis described by Martin et al. (1990), the results of Webster and

34

Ricciardi (1991) strongly imply that the activating domain of ElA binds a

limiting cellular factor (such as a general initiation factor or an adaptor

protein) and are consistent with the suggestion that the promoter targeting

domain of ElA serves to localise ElA in the vicinity of the promoter by

interacting with a factor already bound to it.

1.5.4 The activating domain of ElA specifically interacts with TBP.

The suggestion that the activating domain of ElA might interact

with a component of the transcription complex is supported by

observations that certain TATA box sequences can mediate activation by

E lA (Simon et ah, 1988; Wu et ah, 1987), and that adenovirus infection

increases the transcriptional activity of a partially purified TFIID fraction

from HeLa cells (Leong et ah, 1988). Lee et al. (1991) have further

demonstrated that the large ElA protein binds specifically and stably to the

human TATA binding protein (TBP) using immunoprécipitation and far

western blotting analysis. In addition, using sedimentation velocity

analysis, Lee et ah (1991) demonstrated that these two proteins form a

discrete complex that is consistent with a one-to-one complex of the two

proteins {i.e., a heterodimer) and which appears to contain no other

proteins other than ElA and TBP (Lee et ah, 1991). The 243 amino acid ElA

protein, which is a much poorer trans-activator (for review, see; Flint and

Shenk, 1989), also binds to TBP although much more weakly than the

large ElA protein (Lee et ah, 1991). The interaction between the large ElA

protein and TBP is dependent upon the integrity of the activating domain

of ElA, since deletion of part of the activation domain (AN149; Martin

et ah, 1990) or two mutations in the activating domain that impair trans-

activation (148E^G and 151G—>A; Martin et ah, 1990; Webster and

Ricciardi, 1991) also diminish binding to TBP (Lee et ah, 1991). However,

another mutation in the activation domain that impairs trans-activation

35

(145D^A; Martin et al, 1990; Webster and Ricciardi, 1991) has no effect on

binding of TBP. Hence the ability of ElA to bind to TBP would appear to be

necessary but not sufficient for trans-activation by ElA.

1.5.5 ElA can activate transcription in vitro.

Several studies have examined the ElA dependent activation of

adenovirus genes in vitro. Leong and Berk (1986) used a run-off

transcription assay with whole cell extracts (Manley et al., 1980) to analyse

the activity of the major late promoter (MLP) in vitro (Leong and Berk,

1986). Extracts from adenovirus infected cells were found to be 5-15 times

more active in transcribing the MLP than extracts from mock-infected

cells, or cells infected with viruses that do not express the 13 S ElA

mRNA. Similar results were observed for transcription from the protein

IX and E3 promoters, whereas a modest 2-fold increase was observed for

the human p-globin promoter (Leong and Berk, 1986).

Spangler et al. (1987) used a similar run-off transcription assay with

whole cell extracts (Manley et al., 1980) to analyse transcription of the E2A

and MLP promoters in the presence and absence of a bacterially expressed

large ElA protein (Spangler et al., 1987). Transcription from the E2A

promoter was stimulated up to 7-fold following addition of the 289 amino

acid protein, compared to only 2-fold stimulation of MLP (Spangler et al.,

1987). Green et al. tested a 49 amino acid synthetic peptide (PD3)

comprising the 46 amino acids of CR3 plus three adjacent and conserved

amino acids for its ability to stimulate transcription of the major late

promoter in vitro. Using a run off transcription assay with nuclear extracts

prepared as described by Dignam et al. (1983) the major late promoter was

stimulated up to 20-fold following addition of PD3, whereas peptides

containing mutations in the conserved cysteine residues of CR3 were

found to be totally inactive (Green et al, 1988). This peptide was also found

36

to stimulate transcription from the E3 promoter (Loewenstein and Green,

1989) but not from several other cellular promoters (Green et ah, 1988;

Loewenstein and Green, 1989). Finally, a bacterially produced 289 amino

acid ElA protein has also been shown to stimulate transcription from the

RNA polymerase III transcribed adenovirus virus-associated (VAl) RNA

gene in vitro, using whole cell extracts from HeLa cells (Datta et at., 1991).

1.6 Characterisation of the adenovirus E4 promoter.

All of the early viral promoters have been analysed extensively and

have been found to be similar in structure to typical cellular promoters

that direct transcription by RNA polymerase II, consisting of a TATA box

and one or more upstream promoter elements. Several experimental

approaches have led to the characterisation of upstream promoter

elements that contribute to E4 promoter activity and hence could

potentially be involved in its trans-activation by ElA. Deletional analysis

has identified regions of the promoter that are important for its activity

in vitro and in vivo. DNAse I footprint and méthylation interference

assays have identified binding sites for cellular transcription factors, some

of which overlap or coincide with functionally defined regions. DNA

binding assays and in vitro transcription assays have been used to further

characterise the cellular factors that interact with these elements and

cDNAs encoding some of these proteins have been isolated.

1.6.1 Deletional analysis of the E4 promoter.

The effects of deletions of the E4 promoter upon its activity in vivo

have been extensively studied using transient transfection assays. Gilardi

and Perricaudet (1984) examined the effect of progressive deletion into the

37

upstream region of the E4 promoter. Little transcription was detected from

the E4 promoter when it was transfected into HeLa cells, whereas over

twenty fold more transcription was detected when it was co-transfected

with a plasmid encoding the large ElA protein (Gilardi and Perricaudet,

1984). Deletion of E4 promoter sequences up to -241 has little effect on

transcription (Gilardi and Perricaudet, 1984) and hence the NFl binding

site and two homologies to the SPl binding site (present between -330 and

-241; see Figure 2.1.0 A) are not required for activation of the E4 promoter.

Further deletion to -220 results in a reduction of E4 transcription to 30% of

the wild type level (Gilardi and Perricaudet, 1984) and this deletion

removes another potential SPl binding site. Deletion to -179 has no

further effect on transcription, but deletion to -158 reduces promoter

activity to less than 5% of the wild type level (Gilardi and Perricaudet,

1984). The region between -179 and -158 includes one half of a 13 bp

inverted repeat (between -172 and -160; the complement lies between -54

and -42) and a 10 bp homology to the core sequence of an enhancer

element in the ElA promoter (Hearing and Shenk, 1983).

Lee and Green (1987) showed that a region of the promoter located

from -100 to -200 can function as an ElA dependent enhancer element.

Using a transient transfection assay, there was found to be little effect of

deleting the promoter to -200. In the presence of ElA, further deletion of

the E4 promoter to -138 resulted in a 30 fold decrease in transcription,

whereas in the absence of ElA there was no effect of this deletion. When

E4 sequences from -100 to -200 were placed 600 bp upstream of the E4

TATA box, this region could still activate transcription to normal levels,

and its effect was found to be orientation independent. In addition, this

region was shown to stimulate transcription of heterologous promoters in

the presence of ElA in a similar orientation and distance independent

manner.

38

1.6.2 Identification of cellular factor binding sites within the E4 promoter.

Footprinting experiments have demonstrated the presence of

several binding sites for cellular proteins within the E4 promoter, some of

which lie within the designated enhancer region. Lee and Green (1987)

observed footprints with crude nuclear extracts in three distinct regions of

the promoter, between positions -135 to -172; -39 to -56; and around -110

(see Figure 1.6.2). The -135/-1 72 region was shown to consist of two

overlapping sites; a site distal to the TATA box (positions -150 to -172) is

protected at low protein concentrations whereas a site proximal to the

TATA box (positions -135 to -150) is only protected at high protein

concentrations. The distal site (-150/-172) and the region adjacent to the

TATA box (-39/-56) include the 13 bp inverted repeat sequences and

competition experiments demonstrate that they bind a common factor(s)

(Lee and Green, 1987). These two sites have a 5 bp motif (TGACG) in

common with the -135/-150 site and two other sites observed by Watanabe

et al. (1988) at positions -221 to -245 and -253 to -273 (see Figure 1.6.2). This

5 bp motif is also present in the adenovirus ElA, E2A and E3 promoters, as

well as several other viral and cellular promoters, suggesting that these

promoters could be activated by a common mechanism.

Raychaudhuri et al. (1987) analysed interactions between cellular

proteins and the E4 promoter utilising fractionated cell extracts.

Adenovirus infected extracts were fractionated by heparin-agarose

chromatography and fractions were tested for their ability to bind to the

E4 promoter. Fractions that yielded well resolved and reproducible

interactions with the promoter were examined further by footprint

39

-250

TATGATAATGAGGGGGTGGAGTTTGTGACGTGGCGCGGGGCGTGGGAACGGGGCGGGATACTATTACTCCCCCACCTCAAACACTGCACCGCGCCCCGCACCCTTGCCCCGCCC

-200

TGACGTAGGTTTTAGGGCGGAGTAACTTGTATGTGTTGGGAATTGTAGTTTTCTTACTGCATCCAAAATCCCGCCTCATTGAACATACACAACCCTTAACATCAAAAGAA

-150 • • • * * * # # " " 'AAAATGGGAAGTTACGTCACGTGGGAAAACGGAAGTGACGATTTGAGGAAGTTGTTTTTACCCTTCAATGCAGTGCACfCCTTTTGCCTTCACTGCTAAACTCCTTCAACA

-100

GGGTTTTTTGGCTTTCGTTTCTGGGCGTAGGTTCGCGTGCGGTTTTCTGGGTGTTTCCCAAAAAACCGAAAGCAAAGACCCGCATCCAAGCGCACGCCAAAAGACCCACAAA

-50

TTTGTGGACTTTAACCGTTACGTCATTTTTTAGTCCTATATATACTCGCTCTGCAAACACCTGAAATTGGCAATGCAGTAAAAAATCAGGATATATATGAGCGAGACG

Figure 1 .6 .2 : Summary of results from DNasel footprinting (from Watanabe et aL, 1988) and méthylation interference assay analysis (from Raychaudhuri et al., 1987) of the adenovirus E4 promoter. Regions that bind cellular proteins (as determined by DNasel footprinting) are indicated by black lines on each DNA strand. The thicker lines indicate footprints which include the 13 bp imperfect inverted repeat, and the thinner lines the footprint at -110 also observed by Lee and Green (1987). Residues that are directly involved in DNA/protein interactions (as determined by méthylation interference assays) are indicated by black dots. The TGACG (CGTCA) motifs and the TATA box are shown in bold type. The dashed lines indicate the E4F2 (at -150; Hai et al., 1988) and the E4TF1 (at -140; Watanabe et al., 1988) binding sites.

40

analysis to determine which sequences were protected by each heparin-

agarose fraction.

This approach demonstrated the presence of several interactions

between cellular proteins and the E4 promoter. A single factor was shown

to interact with the -39/-56 and -150/-172 sites, and was termed E4F (see

below). Other interactions were observed at positions -101, -149 to -150, -156

to -158 and -173 to -175 (Raychaudhuri et al, 1987; see Figure 1.6.2). The

interaction at position - 1 0 1 may correspond to the footprint observed with

crude cell extracts around position -110 (Lee and Green, 1987; Watanabe

et al., 1988); the interaction at -149/-150 is within the binding site of the

E4TF1 factor (Watanabe et al, 1988) and one of the homologies to the ElA

core-enhancer (Hearing and Shenk, 1983); the interaction at -156/-158

corresponds to the binding site of the E4F2 factor (Hai et al, 1988b); and

finally the interaction at -173/-1 75 is within another of the homologies to

the ElA core enhancer.

Finally, although a factor can associate with the region around

position -1 1 0 , this region can be deleted or replaced without affecting

transcription in vivo or in vitro (Handa and Sharp, 1984).

1.7 Characterisation of proteins that interact with the E4 promoter.

1.7.1 The ATF family.

1.7.1.1 Characterisation of ATF binding sites in the viral early promoters.

Lee and Green (1987) demonstrated that a single factor (or related

factors) binds to both the -150/-172 and -39/-56 sites of the E4 promoter (see

Figure 1.6.2), since footprints observed at these sites could be removed by

competition with DNA containing either site. In addition, they also

showed that the factor binding to these sites in the E4 promoter is the

41

same as (or related to) a factor binding to the E2A promoter, since

footprints at the sites in the E4 promoter can be removed by competition

with DNA containing E2A promoter sequences.

Analysis of the E2A and E3 promoters demonstrated that each of

these promoters contains a binding site for a cellular factor which includes

the 5 bp TGACG motif present in the E4 promoter. SivaRaman et al. (1986)

analysed linker scanning mutants of the E2A promoter utilising DNA

binding and footprinting assays, and demonstrated that a 17 bp sequence in

the E2A promoter, which includes the TGACG motif, is the binding site

for a cellular factor. This factor was also demonstrated to bind to the ElA,

E3 and E4 promoters, since DNA fragments containing these promoters

compete for binding of this factor to the E2A-probe DNA. Similar studies

of the E3 promoter (Hurst and Jones, 1987) demonstrated the presence of

three binding sites for cellular factors within this promoter, one of which

(BS2) includes the TGACG motif. BS2 competitor DNA was shown to

compete for binding of factors to an E2A probe, whereas competitor DNAs

containing the other E3 sites cannot. This demonstrated that the E3 BS2

site binds either the same (or a related) factor as the E2A promoter.

Both of these studies provided further evidence that the TGACG

sequence in each of the E2A, E3 and E4 promoters comprises the core motif

of the binding site for the same cellular factor. Lee et al. (1987) performed

extensive DNA binding assays with probes from each of the E2A, E3 and

E4 promoters and competitor DNAs to each of the characterised sites, and

demonstrated that the same (or related) factors bind to each of the sites

containing the TGACG motif. This factor has been termed activating

transcription factor (ATF; Lee et al., 1987a). These experiments were seen

as further evidence that these promoter may share a common mechanism

for activation by ElA.

42

1.7.1.2 Isolation of ATF proteins.

Gel mobility shift assays demonstrate that multiple factors bind

directly to ATF binding sites, and sequence-specific DNA affinity

chromatography has enabled the purification of some of these binding

activities. One family of immunologically related phosphoproteins of -43

and 47 kDa (referred to as ATF-43 and ATF-47, respectively) were purified

from HeLa cells and characterised by their ability to bind to ATF binding

sites present in the viral early promoters (Hai et ah, 1988b). A factor of

similar size, designated CREB (cAMP response element binding protein),

was purified from PC12 cells (Montminy and Bilezikjian, 1987) and rat

brain (Yamamoto et al., 1988) and characterised by its ability to bind to the

CRE (cAMP response element) present in the rat somatostatin promoter

(Montminy and Bilezikjian, 1987).

1.7.1.3 cDNA cloning reveals a multigene family of transcription factors.

Subsequent cloning of CREB revealed that it is a member of the

bZIP family of transcription factors (Gonzalez et ah, 1989; Hoeffler et ah,

1988a) and further analysis demonstrated that it corresponds to ATF-47

(Hurst et ah, 1990). Further cDNA cloning studies demonstrated that an

even more complex array of factors can interact with ATF binding sites.

Hai et ah (1989) isolated several cDNAs that encode distinct proteins which

bind to ATF binding sites. Like CREB, all of these proteins were found to

belong to the bZIP family of transcription factors, indicating that

collectively they belong to a multigene family (the ATF family).

Eight partial cDNAs were isolated (ATF-1 to -8 ) which bind with

varying efficiency to ATF binding sites. Of these, five clones (ATF-1,

ATF-2, ATF-3, ATF-4 and ATF-7) bind efficiently to a single ATF binding

43

acid basic RNR A2 -3 KCR basic spacer L L N L L L

CREB E E A A .R KREVRLM KNR EAAR ECR RKKK EYVKC L ENRVAV L ENQ N KT L lE EL K A LATF-1 TDDPQL KREIRLM KNR E A .R ECR RKKK EYVKC L ENRVAV L ENQ N KT L lE E L K T LATF-2 NEDPDE KRRKFLE RNR AAAS RCR QKRK VWVQS L EKKAED L SSL N GQ L QSEVTL L RNEVAQ LATF-3 APEEDE RKKRRRE RNK lA A A KCR NKKK EKTEC L QKESEK L ESV N AE L KAQLEE L KNEKQH LATF-4 KGEKLD KKLKKME QNK TAAT RYR QKKR AEQEA L TGECKE L EKK N EA L KERADS L AKEIQY LATF-5 ISRRRR EKENPKE RNK MAAA KCR NRRR ELTDT L QAETDQ L EDE K SA L QTELAN L LKEKEK LATF- 6 SDIAVL RRQQRMI KNR ESAC QSR KKKK EYMLG L EARLKA A LSE N EQ L KKENGR L KRQLDE LATF-a DEDPDE RRQRFLE RNR AAAS RCR QKRK LWVSS L EKKAEE L TSQ N IQ L SNEVTL L RNEVAQ Lc-JUN ESQ ERI KAERKRM RNR lA A S KCR KRKL ERIAR L EEKVKT L KAQ N SE L ASTANM L REQVAQ Lc-FOS E . . E . . KRRIRRE RNK MAAA KCR NRRR ELTDT L QAETDQ L EDE K SA L QTEIAN L LKEKEK LhXBP HLSPEE KALRRKL KNR VAAQ TAR DRKK ARMSE L EQQWD L EEE N QK L LLENQL L REKTHG LE4BP4 DEKKDA MYWEKRR KNN EAAK RSR EKRR LNDLV L ENKLIA L GEE N AT L KAELLS L KLKFGL IC/EBP VDKNSN EYRVRRE RNN lA V R KSR DKAK QRNVE T QQKVLE L TSD N DR L RKRVEQ L SRELDT L

W ENQ E

Figure 1.7.1.3: Amino acid homology in the basic domain and leucine zipper regions of several members of the ATF family and selected other bZIP proteins that are known to interact with the E4 promoter. Regions of homology are boxed, with conserved residues in bold. See Table 1.7.1.3 for references.

Equivalent (=) and like (-) proteins:

CREBi'2 = ATF-472.3; ~ CREM^ - ATF-lÂTF-15,6 = ATF-433'6 = TREB367; - CREB - CREMÂTF-25 = CREBPl» = mXBPl^ = TREBT = HB16^0; - ATF-aii ATF-45 = TREB6712 = CREB213; ~ mATF-4:4 ATF-55 - Pos:5 hXBPi6 = TREB57

Table 1.7.1.3: The relationship between purified and cloned bZIP proteins belonging to the ATF family of transcription factors. Several bZIP proteins that bind to the ATF consensus sequence have been cloned by more than one group and are hence referred to by more than one name. Above is a brief summary of the relationships between these proteins, adopting the notation according to Hai et a l (1988) as the standard.

References: ^Hoeffler etal . (1988a); Gonzalez etal . (1989); 2 Hurst et al. (1990); ^Hai et al. (1988b); Hurst and Jones (1987); ^Foulkes et al. (1991); ^Hai et al. (1989); 6 Hurst et al. (1991); ^Yoshimura et al. (1990); ^Maekawa et al. (1989); Îvashkiv et al. (1990); ^ Kara et al. (1990); ^^Gaire et al. (1990); i2Tsujimoto etal. (1991); ^^willems etal. (1992); ^^Mielnicki and Pruitt (1991); i^van Straaten et al. (1983); ^^Liou et al. (1990)

45

site, but not to AP-1 (Angel et ah, 1987), MLTF (Carthew et ah, 1985) or E4F2

(Hai et ah, 1988b) binding sites (Hai et ah, 1989). Clones ATF-1 to ATF- 6

were sequenced and were found to share significant amino acid similarity

within their basic domain and leucine zipper (see Figure 1.7.1.3) but shared

little homology outside these regions (Hai et ah, 1989). ATF-1 however,

was found to share extensive homology with CREB even outside the

conserved basic domain and leucine zipper regions (-75% amino acid

similarity; Hai et ah, 1989). Protein sequence analysis of tryptic fragments

of ATF-43 together with immunological analysis have since demonstrated

that ATF-1 corresponds to ATF-43 (Hurst et ah, 1991).

The basic domain/ leucine zipper (bZIP) motif was originally

described in the liver specific transcription factor C/EBP (Johnson et ah,

1987; Landschulz et ah, 1988). The leucine zipper is an amphipathic

a-helix, consisting of a heptad repeat of leucine residues that are aligned

along one face of the helix (Landschulz et ah, 1988) and the interaction

between two leucine zippers brings about the dimérisation of the two

proteins (Kouzarides and Ziff, 1988). The basic domain consists of two

clusters of the basic amino acids lysine and arginine, separated by an

alanine spacer and a conserved asparagine residue (Landschulz et ah,

1988). The leucine zipper also functions to position the two basic domains

of the dimerised proteins within the major groove of DNA (Gentz et ah,

1989) where the positively charged (basic) amino acids make direct contact

with the negatively charged DNA (Vinson et ah, 1989).

Independent investigations have identified binding sites containing

the ATF consensus sequence in several other viral and cellular promoters

and several proteins that bind to such sites have been cloned and found to

belong to the ATF family (see Figure 1.7.1.3). As a result there is duplicity

in the identification of some proteins, with different notations being used

in each case. For these proteins, I will adopt the notation described by Hai

46

et al. (1989); for reference the alternative names for these proteins are

listed in Table 1.7.1.3.

1.7.2 E4F.

Raychaudhuri et al. (1987) characterised an E4 binding activity that

is strongly induced upon adenovirus infection. Whole cell extracts from

mock and adenovirus-infected cells were fractionated using heparin-

agarose chromatography and the fractions were compared for their ability

to bind to regions of the E4 promoter in a gel-mobility shift assay. One

fraction contained a factor (E4F) that binds to E4 sequences from positions

+32 to -91 and which is strongly induced following adenovirus infection.

Using footprint analysis, E4F was shown to interact with the

E4 promoter (Raychaudhuri et al., 1987) within the -39/-56 site observed in

other studies (Lee and Green, 1987; Watanabe et al., 1988; see Figure 1.6.2).

E4F was also suggested to bind to the homologous site present at positions

-150/-172 (see Figure 1.6.2), since binding to the +32/-90 probe can be

inhibited by addition of a competitor containing E4 sequences from -91 to

-224 (Raychaudhuri et ah, 1987). In addition, E4F was shown to be an

E4-specific factor, since competitors containing sequences from the ElB, E2,

E3, MLP and hsp70 promoters do not affect binding of E4F to the

E4 promoter (Raychaudhuri et ah, 1987).

Raychaudhuri et al. (1989) purified E4F by fractionation and DNA

affinity chromatography and showed it to be a 50 kDa protein when

analysed by SDS gel electrophoresis and silver staining. Consistent with

previous suggestions that E4F may be involved in the activation of the

E4 promoter, the fractions from the DNA-affinity column that contained

E4F were shown to be able to stimulate transcription from the

E4 promoter, but not the MLP promoter, in vitro.

47

1.7.3 Other proteins that bind to the E4 promoter.

In addition to ATF and E4F, several other proteins have been found

to interact with the E4 promoter. EivF displays the same DNA binding

specificity as E4F and can specifically activate transcription of the

E4 promoter in vitro (Cortes et al, 1988). C/EBP (CCAAT-box/ enhancer

binding protein) can bind to ATF binding sites from various promoters

and activates transcription of a promoter containing the E4 -39/-56 and

-150/-172 sites in vivo (Bakker and Parker, 1991). E4BP4 displays a similar

DNA binding specificity to that of E4F and analysis of its activity in vivo

demonstrates that it functions as a transcriptional repressor (Cowell et al,

1992). E4TF1 specifically interacts with the E4 promoter in the region

between positions -132 and -152 (see Figure 1.6.2). The binding site for

E4TF1 overlaps with, but is distinct from the ATF binding site in this

region, since the binding specificity of E4TF1 is different from that of ATF

(Watanabe et al., 1988). Finally, E4F2 binds to the region between positions

-149 and -155 (see Figure 1.6.2). In the crude nuclear extract, this DNA

binding activity cannot be detected because its footprint is masked by the

ATF footprints in adjacent regions (Hai et ah, 1988b).

1.8 Trans-activation of the E4 promoter by ElA.

The above studies demonstrate that several different cellular

transcription factors interact with a limited number of binding sites in the

E4 promoter. Although the functional relationship between many of these

factors is unknown, the activities of two of these factors, E4F and a

member of the ATF family (ATF-2), has been analysed in detail in respect

to their individual roles in trans-activation of the E4 promoter by ElA.

Two models for the mechanism of ElA trans-activation of the

E4 promoter are postulated from the study of E4F and ATF-2 function. The

48

first model involves the post translational modification of a transcription

factor by an ElA associated or an ElA dependent protein kinase, that

increases its activity and hence stimulates transcription from the

E4 promoter. The second model involves a direct or indirect interaction

between ElA and a promoter bound transcription factor, that serves to

localise E lA at the E4 promoter allowing the activating domain to

stimulate transcription.

1.8.1 E4F mediated fraws-activation of the E4 promoter by ElA.

E4F is characterised as an E4 binding activity that is strongly induced

upon adenovirus infection (Raychaudhuri et ah, 1987). This suggests that

E4F may be involved in frans-activation of the E4 promoter by ElA. This

argument is strengthened by the observation that the adenovirus induced

increase in E4F DNA binding is ElA dependent, since extracts from cells

infected with the mutant ElA virus dl312 do not display induced E4F

activity (Raychaudhuri et ah, 1987). Analysis of the kinetics of induction of

E4F during infection with the wild type virus reveals a rapid increase

within 3 h of infection, and then maintenance of this level through 4.5 h

of infection. Thus, the level of E4F present in infected cells closely

corresponds to transcription of the E4 gene (Raychaudhuri et ah, 1987).

E4F DNA binding activity was also found to be regulated during

differentiation of mouse F9 teratocarcinoma cells (Raychaudhuri et ah,

1989), consistent with the presence and regulation of a cellular activity

similar to the viral ElA (Impériale et ah, 1984). Extracts of F9 cells were

found to contain substantial levels of E4F activity, independent of a viral

infection, whereas E4F activity can not be detected in extracts of

differentiated F9 cells. Adenovirus infection of the differentiated F9 cells.

49

which thus provides viral ElA function, was found to induce the

reappearance of E4F activity (Raychaudhuri et ah, 1989).

Raychaudhuri et al. (1989) showed that the DNA binding activity of

a preparation of purified E4F is inactivated upon its incubation with

agarose-bound calf intestinal phosphatase, suggesting that the regulation

of E4F activity may involve a phosphorylation event. This inactivation of

E4F DNA binding activity by phosphatase treatment is not an irreversible

event since an extract prepared from adenovirus-infected cells and

fractionated by heparin agarose chromatography to remove active E4F is

capable of reactivating E4F DNA binding activity upon incubation with

the phosphatase-treated sample. In contrast, a similarly fractionated extract

prepared from dl312- (E1A-) infected cells can not restore E4F activity.

Consistent with a phosphorylation dependent activation of E4F DNA

binding activity, a 50 kDa polypeptide that co-purifies with E4F was

labelled with [y-32p] aTP during this reactivation of E4F activity

(Raychaudhuri et ah, 1989).

Rooney et al. (1990) further characterised the DNA binding

properties of E4F and demonstrated that it can be distinguished from ATF

by their different DNA binding properties. Méthylation interference

analysis of the binding of purified E4F and ATF to the E4 -39/-56 site

clearly demonstrates that the DNA binding specificities of E4F and ATF

overlap but differ. A summary of the results of this analysis is presented in

Figure 1.8.1. Méthylation of the G residues at positions -46 and -49 on the

non-coding strand of the E4 promoter DNA and at position -48 on the

coding strand inhibit binding of both ATF and E4F to the E4 -39/-56 site.

However, méthylation of the G residue at -53 on the coding strand

prevents binding of E4F, but does not affect binding of ATF to this site (see

Figure 1.8.1). In addition, the effects of several point mutations in the

E4 -39/-56 site on the binding of purified E4F and ATF were examined in a

50

E4F

ATF

* *

AAAATGACGTAACGGTTTTTTACTGCATTGCCAA

* *

AAAATGACGTAACGGTTTTTTACTGCATTGCCAA

Figure 1.8.1: Méthylation interference analysis of E4F and ATF binding to the -39/-56 site of the E4 promoter. Methylated G-residues in the ATF/E4F binding site that inhibit binding of E4F (top) and ATF (bottom) are in bold type and are indicated by an asterix. From Rooney et al, 1990.

51

gel mobility shift assay. A summary of the results of this analysis is

presented in Table 1.8.1. A point mutation (PM4) strongly inhibits binding

of E4F, but has no effect on the binding of ATF (see Table 1.8.1). In

addition, purified E4F is unable to bind to ATF sites from the E2A and

E3 promoters (see Table 1.8.1). Hence the consensus binding site for E4F is

different to that of ATF, and the E4F binding sites from the E4 promoter

are considered to overlap with the ATF sites (Rooney et ah, 1990; see Table

1.8 .1).

Rooney et al. (1990) postulated that if ATF was responsible for

mediating this process, any ATF binding site might be able to confer ElA

inducibility. However, if E4F was the target of frans-activation by ElA,

then only E4F binding sites (which are a limited subset of ATF sites) would

be able to confer ElA inducibility. Utilising the differences in the sequence

recognition of ATF and E4F described above, Rooney et al. (1990)

attempted to define the nature of the factor involved in mediating ElA

frflns-activation of the E4 promoter by examining the relative roles of

these two activities on the activation of a heterologous promoter by ElA

in vivo. A simple promoter was constructed that contains only ATF or

E4F binding sites upstream of a TATA element and its response to ElA

was examined by co-transfection in a transient expression assay. The

TATTTAT sequence from the simian virus 40 (SV40) promoter was

chosen for this synthetic promoter (Rooney et al., 1990) since previous

experiments have shown that this element is not responsive to E lA

(Simon et ah, 1988).

Rooney et al. found that a promoter containing a duplicated

ATF/E4F binding site from the E4 promoter was ElA inducible when

assayed in Vero cells. However, promoters containing duplicated ATF

binding sites from either the E2A or E3 promoters were not ElA inducible.

That these test promoters were functional was indicated by the finding

52

a

ATF BINDING SITES SEQUENCE ATF E4F

ADENOVIRUS E3 (-55) T G A C G A A A G ++ +ADENOVIRUS E2 (-80) T G A C G T A G T ++ + -ADENOVIRUS E4 (-147) T G A C G A T T T ++ + -

(-230) T G A C G T A G G +++ -(-260) T G A C G T G G C + 4-4- -

ADENOVIRUS E4 (-50) T G A C G T A A C + + + +++(-163) T G A C G T A A C + + + +++

WT T G A C G T A A C + + + +++PMl T a A C G T A A C - -

PM2 T G A t G T A A C - +PM4 T G A C G T A t C + + + -PM5 T G A a G T A A C n /d n /dPM6 T G A C t T A A C n /d n /d

PM5/6 T G A a t T A A C n /d n /d

CONSENSUS ATF T G A C G TCONSENSUS E4F T G A C G T A A C

Table 1.8.1: Binding specificities of ATF and E4F for binding sites from various adenovirus early promoters. From Rooney et al. (1990). Bracketed numbers refer to the position of the binding site with respect to the start site of transcription.

that each was inducible by cAMP (Rooney et ah, 1990). These results were

interpreted to demonstrate that a promoter that can bind E4F (and ATF as

well) is ElA inducible, whereas a promoter that can only bind ATF is not.

To further test this hypothesis, Rooney et al. constructed another series of

promoters identical to those described above, but which contained

duplicated ATF/E4F binding sites harbouring the point mutations

described in Table 1.8.1. As described above, the promoter containing the

wild type ATF/E4F binding site was found to be ElA inducible. Promoters

containing the PMl and PM2 mutations, which inhibit both ATF and E4F

binding, were both found to be unresponsive to ElA. In addition a

promoter containing the PM4 mutation, which does not affect binding of

ATF but which eliminates binding of E4F, was also found to be

unresponsive to ElA. This promoter was shown to be functional by its

ability to respond to cAMP (Rooney et ah, 1990). This result was

interpreted as confirmation of the hypothesis described above, that a

promoter that can bind E4F is ElA inducible, whereas a promoter that can

only bind ATF is not. Hence in the context of this heterologous promoter,

binding of E4F but not ATF is essential for activation by ElA.

1 .8 . 2 ATF-2 mediates tmMS-activation by ElA.

Several of the adenovirus early promoters contain binding sites for

the ATF family of cellular transcription factors (Lee et al., 1987a),

suggesting the involvement of an ATF protein in the ElA transcriptional

response. Liu and Green (1990) examined the activities of two members of

the ATF family for their ability to recruit ElA to the promoter (as

postulated in the model described above; Martin et al., 1990) and

demonstrated that a specific member of the ATF family, ATF-2, can bind to

54

the E4 promoter and function as an ElA inducible transcription factor (Liu

and Green, 1990).

A chimeric protein consisting of ATF-2 fused to the Gal4 DNA

binding domain is able to activate transcription in vivo in a manner that

is dependent upon Gal4 binding sites in the test promoter and on the

presence of ElA (Liu and Green, 1990), as determined by co-transfection in

a transient expression assay. Mutations that destroy either the promoter

targeting domain, the activating domain, or the zinc finger motif of ElA

(Lillie and Green, 1989; Martin et ah, 1990) also prevent activation by ElA

in this assay (Liu and Green, 1990). Hence these results are consistent with

the model hypothesised by Martin et al. (1990) which requires that ElA

contains an activating domain and is able to interact with a promoter

bound protein.

As further evidence for an interaction between ElA and ATF-2, Liu

and Green demonstrated that the roles of these two proteins can be

switched. A Gal4-E1A fusion protein containing amino acids 150-223 of

ElA (and therefore lacking the activating domain) cannot stimulate

transcription from a test promoter containing Gal4 binding sites in the

same assay (Liu and Green, 1990; Martin et ah, 1990). However, this fusion

protein is able to stimulate transcription when co-transfected with an

ATF-2-VP16 fusion protein (Liu and Green, 1990). In this experiment, ElA

behaves as the promoter bound factor and ATF-2-VP16 the frans-activator.

Taken together, these results suggest an interaction between ATF-2

and the promoter targeting domain of ElA. The activating domain of ElA

is therefore localised at a promoter containing ATF binding sites, where it

is able to activate transcription. This interaction is specific to ATF-2 since

another member of the ATF family, ATF-1, does not mediate trans-

activation by E lA in this assay (Liu and Green, 1990). Another ATF

55

protein, CREE, also mediates trans-activation by ElA in this assay,

although less efficiently than ATF-2 (Flint and Jones, 1991).

Mutational analysis of ATF-2 has demonstrated that deletion of the

N-terminal 33 amino acids of this protein destroys its ability to mediate

trans-activation by ElA in the assay described above (Liu and Green, 1990).

Interestingly, this N-terminal region contains a potential zinc finger motif

that shares extensive homology with the C2 H2 zinc fingers found in

TFIIIA (Miller et al, 1985), Spl (Kadonaga et al, 1987), YYl (Shi et al, 1991)

and a variety of other regulatory proteins found in higher and lower

eukaryotes. These structures are characterised by spatially conserved pairs

of cysteine and histidine residues that can function to sequester a single

zinc ion (Miller et al, 1985). Further mutational analysis has demonstrated

that the integrity of this zinc finger motif is essential for mediating trans-

activation by ElA, since a non-conserved substitution of cysteine residues

known to be essential for zinc binding in the homologous structures of

other proteins also destroys the ability of ATF-2 to mediate trans-

activation by ElA (Flint and Jones, 1991; Zu et al, 1991). In addition the

N-terminal 112 amino acids of ATF-2, when fused to the DNA binding

domain of Gal4, were found to be sufficient to mediate frans-activation by

ElA, although further deletion of ATF-2 to leave the N-terminal 83 amino

acids also destroys this function (Flint and Jones, 1991). Hence it is possible

that a second structure in the 80-120 amino acid region is required,

together with the zinc finger motif, for ATF-2 to mediate trans-activation

by ElA.

56

1.9 Aims of this thesis.

Because multiple cellular transcription factors are able to interact

with the E4 promoter, it is difficult to assess which of these factors are

responsible for mediating frans-activation by ElA. An additional

complication is that one particular element in the E4 promoter that is

known to play an important role in this process has been shown to

interact with several distinct transcription factors which may individually

be able to co-operate with ElA to activate the E4 promoter.

The transcription factors E4F and ATF-2 have specific roles in

mediating trans-activation by ElA (Liu and Green, 1990; Rooney et ah,

1990), but their relative roles in activating the E4 promoter have not yet

been adequately addressed. Firstly, the ability of E4F and ATF proteins to

independently mediate trans-activation has only been examined in the

context of a heterologous promoter (Rooney et ah, 1990) and not in the

context of the whole E4 promoter. Secondly, the DNA binding specificity

of ATF-2 has not been examined in detail and hence cannot be compared

to that of E4F.

In order to properly assess the role of E4F in mediating trans

activation of the E4 promoter by ElA, I have introduced point mutations

into the ATF/E4F binding sites in the E4 promoter and monitored their

effects on E4 promoter activity in a transient expression assay in HeLa and

Vero cells. In addition, I have also examined the effects of these point

mutations on the binding of ATF-2 in vitro. Finally, I have explored the

possibility that recombinant ATF-2 could be utilised in an in vi tro

transcription assay to analyse its role in mediating trans-activation of the

E4 promoter by ElA.

57

Chapter 2: Analysis of the requirements for E4 promoter activity in vivo.

A recent study by Rooney et al (1990) demonstrated that a synthetic

promoter consisting of the SV40 TATA box and the two ATF/E4F binding

sites from the E4 promoter is inducible by ElA. However, when these sites

include a point mutation (PM4) which excludes binding of E4F but has no

effect on the binding of ATF, the promoter is no longer ElA inducible.

Rooney et a l therefore concluded that E4F binding to the E4 promoter was

essential for trans-activation by ElA, but that ATF binding was not

necessary.

Other studies do not appear to be consistent with this possibility. A

32 bp element from the vasointestinal peptide promoter (the VIP CRE) can

substitute for E4 promoter sequences upstream of the TATA box (Lee et al,

1989). The VIP CRE is responsible for the cyclic AMP inducibility of the

VIP promoter (Fink et al, 1988; Tsukada et al, 1987) but is not inducible by

ElA in its natural context (Lee e ta l , 1989). However a hybrid promoter

consisting of the VIP CRE fused to E4 promoter sequences including the

TATA box and CAP site (VIPE4CAT) has been shown to be ElA inducible

(Lee et al, 1989; see Figure 2.1.0). The VIP CRE consists of two ATF binding

sites, neither of which resembles the E4F binding site consensus sequence.

The activity of the VIPE4CAT construct therefore appears to be

inconsistent with the observations made by Rooney et al

The heterologous promoter used by Rooney et a l was designed to

isolate the ATF/E4F binding sites it contained as the only known

upstream promoter element (Rooney et al, 1990). However, the results of

Lee et a l (1989) suggest the presence of another element in the

E4 promoter (downstream of the TATA box) that can modify the activity

of ATF binding sites. Together with the observed inconsistency in these

results, this suggests that the conclusion of Rooney et a l that E4F binding

58

to the E4 promoter is essential for its trans-activation by E lA may be

incorrect. By analysing the effect of binding sites characterised for their

ability to bind ATF and E4F on the activity of the E4 promoter in vivo, I

will be able to elucidate the requirements for activation of the E4 promoter

by ElA.

59

2.0 Methods and Materials.

2 .0 . 1 General strategy of construction of new recombinants.

All new recombinant plasmids were constructed according to

standard protocols (Maniatis et al, 1982). Typically, 10 pg of vector was

prepared for ligation to insert DNA by 5-fold over-digestion with the

appropriate restriction enzyme(s) for 1-2 h {e.g. 50 units of enzyme were

used to digest 10 pg DNA in a 50 pi volume) in the reaction buffer

specified and/or provided by the manufacturer.

DN A was purified from the restriction reaction using

phenol/ chloroform extraction and glass milk (BiolOl/Stratech). Small

fragments (<100 bp) excised during the restriction digestion of vector DNA

are removed during this procedure due to the inability of glass milk to

bind DNA fragments smaller than 100 bp. DNA longer than 100 bp was

separated from vector DNA on a low melting point (LMP) agarose gel and

the vector fragment was purified using glass milk (see Chapter 2.0.2.3).

Insert DNA was prepared by digestion with the appropriate

restriction enzyme(s) and subsequent gel purification of the desired

fragment by one of two methods. If the insert DNA fragment was over

100 bp long, it was separated from contaminating DNA on a low melting

point (LMP) agarose gel and purified using glass milk (see Chapter 2.0.2.3).

If the insert DNA fragment was less than 100 bp long, it was separated

from contaminating DNA fragments on a 2% agarose gel and purified by

electroelution (see Chapter 2.0.2.3). Synthetic oligonucleotides were

prepared for ligation to vector DNA by their purification from a

denaturing acrylamide gel by passive elution (see Chapter 2.0.2.3).

Complimentary oligonucleotides were annealed together in 50 mM KCl by

heating to 65°C for 5 min and allowing to cool slow ly to room

temperature.

60

The prepared DNA fragments were checked by agarose gel

electrophoresis prior to their ligation. Typical ligations of vector and insert

DNA would contain 100 ng of vector, either 1 pi or 5 pi of the insert DNA

preparation and 1 pi (10 units) of T4 DNA ligase, in 10 pi of reaction buffer

specified and/or provided by the manufacturer. The products of the

ligation were transformed into competent bacteria (see Chapter 2.0.4)

which were subsequently plated onto L-agar (see Chapter 2.0.4) containing

the appropriate antibiotic for selection.

A number of colonies were picked off the plate using a sterile loop

and grown in 3 ml LB (see Chapter 2.0.3.1) + antibiotic for 5-6 hours.

"Miniprep' plasmid DNA was prepared from 1.5 ml of this culture (see

Chapter 2.0.3.2) and screened for the presence of the correct recombinant

DNA by digestion with appropriate restriction enzymes and analysis by

agarose gel electrophoresis (see Chapter 2.0.2.1). Alternatively, the products

of a restriction digest that yielded 5' overhangs were labelled with either

[32p]-dATP or P^P]-dCTP using AMV reverse transcriptase (see

Chapter 3.0.3). The labelled DNA bands were then separated on a

denaturing polyacrylamide gel (see Chapter 2.0.2.2) and visualised by

autoradiography.

The remainder of the culture used to make the "miniprep' DNA

that yielded a positive clone was used to innoculate 500 ml LB +

ampicillin. This culture was grown overnight and used to prepare a larger

stock of plasmid DNA ('maxiprep' DNA; see Chapter 2.0.3.1) that is

sufficiently pure to be used in a number of applications, including

transfection into human cells, DNA sequencing, in vitro transcription,

etc...

61

2 .0 .2 Electrophoretic analysis of nucleic acids.

2 .0 .2 .1 Agarose gel electrophoresis of nucleic acids.

Horizontal agarose gels were prepared by m elting agarose

(molecular biology grade) in Ix TAE buffer (40 mM Tris-acetate; 1 mM

EDTA pH 8.0) containing 0.5 pg/m l EtBr. Typically, 2% agarose gels were

used to separate DNA fragments less than 500 bp long, whereas 1 % agarose

gels were used to separate larger DNA fragments. The molten gel was

allowed to cool to <65°C before casting in a perspex tray and setting at

room temperature. The DNA sample was applied to the gel in a loading

buffer containing 40% (w /v) sucrose; 0.25% orange G dye and run at 100 V

for 30 min in Ix TAE buffer containing 0.5 pg EtBr. DNA bands were

visualised on a U.V. transilluminator and the gel photographed with a

Polaroid camera using polaroid '667' film.

2.0.2.2 Denaturing polyacrylamide gel electrophoresis of nucleic acids.

Denaturing polyacrylamide gels (10% acrylam ide at 20:1

acrylamide:bisacrylamide; 50% urea; in Ix TBE (90 mM Tris-borate; 2 mM

EDTA pH 8.0) were cross linked by the addition of l/lOO^k vol. 10%

ammonium persulphate and 1/2000*^ vol. of TEMED and poured into a

vertical gel assembly (Protean, BioRad) and allowed to set at room

temperature. The DNA sample was mixed with an equal volume of

loading buffer (90% deionised formalin; Ix TBE; 0.25% bromophenol blue;

0.25% xylene cyanol) and denatured by boiling for 2 min and quenching on

ice. The sample was then loaded on the gel and run at 45 mA for

approximately 2 h in Ix TBE. The gel was then wrapped in 'cling film' and

exposed to X-ray film at -70°C overnight.

62

2.0.2.3 Gel purification of nucleic acids.

DNA fragments larger than 100 bp were purified by separation on a

low melting point (LMP) agarose gel (prepared as for standard agarose gels,

except using low melting point agarose set at 4°C) and excision of the

desired band. The gel slice was melted by the addition of 1 ml saturated

Nal per gram of gel slice and heating at 65°C for 5 min. The DNA fragment

was then purified from the molten agarose using glass milk

(Biol01/Stratech), according to the manufacturers instructions.

DNA fragments smaller than 100 bp were purified by electroelution

from a standard 2% agarose gel (see above). DNA bands were visualised on

a U.V. transilluminator and a well cut in the gel behind the DNA band to

be purified. The well was filled with Ix TAE (minus EtBr) and the current

reversed such that the DNA band was eluted into the well. The current

was applied to the gel in 15-20 second pulses at full power (approximately

40 mA) with the buffer in the well being removed and saved after each

pulse of current. After several pulses, the DNA bands were again

visualised on a U.V. transilluminator to check that all of the required

band had been removed from the gel. The eluted DNA fragment was then

purified with phenol/chloroform and precipitated by the addition of

1 /lOth vol. NaOAc and 2.5 vols. EtOH.

Oligonucleotides were purified prior to their use for cloning by

separation and passive elution from a 15% denaturing polyacrylamide gel

(see above). DNA bands were visualised by U.V. shadowing against a U.V.

fluorescent TLC plate. The desired band was excised from the gel and was

'mashed' with a pipette tip in a 1.5 ml centrifuge tube. This was filled with

elution buffer (0.1% SDS; 0.5 M NH4 OAC; 10 mM MgOAc) and placed on a

rotating wheel for 16 h. The gel was pelleted by brief centrifugation and the

supernatant removed to a clean tube. DNA was precipitated by the

63

addition of 1/10^^ vol. NaOAc and 2.5 vols. EtOH and the DNA pellet was

resuspended in water.

2.0.3 Preparation of plasmid DNA.

2.0.3.1 'Maxiprep' DNA.

500 ml LB (1% w /v bacto-tryptone; 0.5% w /v yeast extract; 1% NaCl)

+ 50 |ig /m l ampicillin was inoculated with 1/1000*^ vol. of a saturated

bacterial culture harbouring the relevant plasmid, and the cells were

grown for 16 h at 37°C with shaking. Cells were harvested by

centrifugation at 3000 rpm for 15 min and cell lysates prepared by alkaline

lysis (Birnboim and Doly, 1979). Briefly, cells were resuspended in 100 ml

solution 1 (50 mM glucose; 25 mM Tris-HCl; 10 mM EDTA; pH 8.0); room

temperature for 5 min; 200 ml solution 2 (0.2 M NaOH; 0.1% SDS) added;

ice for 10 min; 150 ml solution 3 (3 M KO Ac; pH 5.0) added; ice for 30 min.

Cell debris was pelleted by centrifugation at 3000 rpm for 15 min and the

supernatant was strained through sterile gauze into a fresh bottle. DNA

was precipitated by the addition of 0 .6 vols, of isopropanol, incubation at

room temperature for 30 min and centrifugation at 3000 rpm for 15 min.

The supernatant was discarded and the DNA pellet resuspended in 7.5 ml

TE (10 mM Tris HCl; 0.1 mM EDTA; pH 8.0) and 2.5 ml of 10 M NH 4 OAC.

Following incubation on ice for 20 min, the precipitate formed was

pelleted by centrifugation at 3000 rpm for 10 min and the supernatant

transferred to a fresh tube. DNA was precipitated by the addition of

2.5 vols, of EtOH, incubation on ice for 30 min and centrifugation at

3000 rpm for 15 min. The supernatant was discarded and the DNA pellet

resuspended in 500 pi water. RNA was removed by the addition of

RNase A (boiled for 15 min to inactivate DNase) to 50 p g /m l and

incubation at 37°C for 15 min. The solution was adjusted to 1.5 M NaCl

64

and DNA was precipitated by the addition of 0.3 vols, of 30% PEG; 1.5 M

NaCl, incubation on ice for 30 min and centrifugation at 14000 rpm for

10 min. The DNA pellet was resuspended in 400 pi of a Ix proteinase K

buffer (10 mM Tris-HCl; 5 mM EDTA; 0.5% SDS; pH 7.8) and incubated at

55°C for 30 min. Proteinase K was removed by extraction with

phenol/chloroform and the DNA precipitated by the addition of 2.5 vols,

of EtOH, incubation on ice for 30 min and centrifugation at 3000 rpm for

15 min. The final DNA pellet was resuspended in 200 pi water and the

DNA concentration estimated by measuring the O.D. at 260 nm (O.D. of

1.0 = 50 pg/ml).

2.0.3.2 'Miniprep' DNA.

3 ml LB + 50 pg/m l ampicillin was inoculated with a single bacterial

colony and the cells were grown for 16 h at 37°C with shaking. 1.5 ml of

culture was transferred to a 1.5 ml centrifuge tube and cells were harvested

by centrifugation at 14000 rpm for 5 min. Cell lysates were prepared by

alkaline lysis (Birnboim and Doly, 1979). Briefly, cells were resuspended in

100 pi solution 1 (see Chapter 2.0.3.1); room temperature for 5 min; 200 pi

solution 2 (see Chapter 2.0.3.1) added; ice for 10 min; 150 pi solution 3 (3 M

NaOAc; pH 4.5) added; ice for 30 min. Cell debris was pelleted by

centrifugation at 14000 rpm for 10 min and the supernatant was pipetted

into a fresh tube. Proteins were removed by extraction with an equal

volume of phenol/chloroform and DNA was precipitated by the addition

of 2.5 vols, of EtOH, incubation at -20°C for 30 min and centrifugation at

14000 rpm for 10 min. The supernatant was discarded and the DNA pellet

washed in 1 ml 70% EtOH, air dried and resuspended in 20 pi TE (10 mM

Tris-HCl; 0.1 mM EDTA; pH 8.0). RNA was removed during subsequent

digestion with restriction enzymes by the addition of RNase A (boiled for

15 min to inactivate DNase) to 50 pg/m l to the reaction buffer.

65

2.0.4 Preparation of competent bacteria and transformation with

recombinant plasmid DNA.

Competent Eschericia coli strain MC1061 were prepared by the

standard CaCl2 procedure. Briefly, 100 ml of culture was inoculated with

1:1000* vol. of an overnight culture into LB. The cells were grown at 37°C

with shaking until the O.D. at 600nm = 0.5. The cells were cooled on ice for

15 min and harvested by centrifugation at 3000 rpm. The cell pellet was

resuspended in 20 ml ice cold 50 mM CaCl2 (pH 6.0) and the cells incubated

on ice for 30 min. The cells were again harvested by centrifugation at

3000 rpm and the cell pellet was resuspended in 50 mM CaCl2 ; 15%

glycerol. The cells were incubated on ice for between 15 min and 1 h, and

aliquots of 50 pi were snap-frozen using liquid nitrogen and stored at

-80°C.

For a typical transformation, an aliquot of competent cells were

thawed by warming in the palm of the hand and immediately placed on

ice. Approximately 10 ng of DNA (1 pi of a ligation) was added to the cells

and incubated on ice for 30 min. The cells were then heat shocked by

immersion in a water bath at 42°C for 90 sec. and returned to ice. 300 pi LB

was then added and the cells incubated at 37°C for 1 h to allow the cells to

express the antibiotic resistance gene present on the transformed plasmid.

Various amount of this mixture were then plated onto LB agar plates (1%

w /v bacto-tryptone; 0.5% w /v yeast extract; 1% NaCl; 1.5% bacto-agar)

containing 50 pg/m l ampicillin and the transformed cells grown at 37°C

overnight.

66

2.0.5 Transient transfection assay.

2.0.5.1 CaCl2 /phosphate transfection of plasmid DNA into mammalian

cells.

Mammalian cells (HeLa and Vero) were grown in E4 medium at

37°C and maintained by regular re-plating at 1/10* density from confluent

90 mm tissue culture plates. The cells were separated from the tissue

culture plates by washing with a 3:1 mix of versine/0.25% trypsin

following aspiration of the medium. Plasmid DNA containing the test

promoter linked to the CAT reporter gene was transfected into

mammalian cells by CaCl2 /phosphate precipitation. Typically, 5 pg of the

plasmid containing the test promoter and 15 pg of a carrier DNA (such as

pSP64); or 5 pg of the plasmid containing the test promoter, 5 pg of a

plasmid expressing the ElA gene (pH3G; see below) and 10 pg of a carrier

DNA, were transfected into cells split 1:8 from a confluent plate into

35 mm tissue culture plates 8 h previously. pH3G contains the Hindlll-G

fragment of adenovirus type 5 (nucleotide positions 1-2804) encoding the

entire ElA region cloned into pBR322 (Green et a l , 1983).

The precipitation of the DNA involved mixing 500 pi of a 240 mM

CaCl2 solution containing the DNA, with 500 pi of 40 mM Hepes;

275 mM NaCl; 1.4 mM Na2HP0 4 ; 1.4 mM NaH2P0 4 ; pH 7.1; dropwise and

with continual gentle mixing. The mixture was incubated at room

temperature for 30 min and 500 pi was added to the cells dropwise. The

cells were grown at 37°C for 16 h, after which the medium was changed

and the cells were grown for a further 24 h at 37°C. The cells were then

washed in 5 ml PBS and harvested by scraping off the plate in 1 ml PBS

into a 1.5 ml centrifuge tube. The cells were then pelleted by brief

centrifugation and resuspended in 50 pi 250 mM Tris-HCl; ImM EDTA;

67

pH 7.8. Crude cell extracts were prepared by three cycles of freeze/thaw and

the cell debris was pelleted by centrifuging at 14000 rpm for 2 min.

2.0.5.2 Assay of chloramphenicol acetyl transferase activity.

Crude cell extracts prepared from transfected cells was analysed for

CAT activity. 27 pi of the crude cell extract was incubated with 0.125 pCi

[ C] chloramphenicol in an 80 pi reaction containing 1.25 mM acetyl CoA;

215 mM Tris HCl; pH 7.5, for 1 h at 37°C. The products of the reaction were

extracted by vortexing with 2 0 0 pi ethyl acetate and brief centrifugation to

separate the aqueous and organic phases. 170 pi of the ethyl acetate phase,

which contains all forms of chloramphenicol, was carefully removed and

subsequently concentrated to approximately 1 0 pi by vacuum evaporation.

This was spotted onto a silica gel TLC ready foil (Schleicher and Schuell)

and allowed to dry before being subjected to ascending chromatography

with a 95:5 mixture of chloroformimethanol. Finally, the plate was

allowed to dry before being examined by autoradiography.

The activity of a promoter was deduced by excising the

[ C] chloramphenicol and acetylated-[^^C] chloramphenicol spots from the

TLC plate and counting the amount of radioactivity present in each using

a scintillation counter. The activity of the promoter is expressed as

% conversion of the chloramphenicol to its acetylated forms and is

calculated using the following formula;

% conversion = counts per minute acetylated chloramphenicol x 1 0 0

counts per minute total chloramphenicol

In each assay, this was calculated from the average of at least two

separate transfection experiments measuring the activity of multiple

preparations of the plasmid containing the test promoter.

68

2.1 Results.

2.1.0 An ATF binding site that does not bind E4F can mediate

trans-activation of the E4 promoter by ElA.

To address whether an ATF binding site that does not bind E4F can

confer ElA inducibility upon the E4 promoter, I constructed a hybrid

promoter (E2CAT) that consists of a duplicated ATF binding site from the

adenovirus E2A promoter placed upstream of the E4 TATA box and CAP

region. E2CAT therefore resembles the hybrid promoter VIPE4CAT, with

the ATF binding sites from the E2A promoter replacing the VIP CRE (see

Figure 2.1.0 A). Whereas it is unknown as to whether E4F binds to the VIP

CRE, the ATF binding site in the E2A promoter has previously been

shown not to bind E4F (Rooney et al., 1990; see Table 1.8.1).

ElA inducibility was tested in a transient expression assay in HeLa

cells, in which the test promoter is linked to the chloramphenicol acetyl

transferase gene {cat) as a reporter. The ability of ElA to induce expression

of cat was assayed in the presence and absence of a co-transfected ElA gene.

Representative autoradiograms of the assay are shown in Figure 2.1.0 B.

These illustrate the ElA inducibility, but not the absolute activity of the

test promoter, since different amounts of cell extract were used for each

individual assay. The amount of ^^C-chloramphenicol present in each of

the spots from the TLC was measured in a scintillation counter (data not

shown) and used to calculate the activity of the test promoter in the

presence and absence of ElA. This data has been adjusted to compensate

for the amount of extract used in each assay and is shown graphically in

Figure 2.1.0 C. The activity of the E2CAT promoter is expressed as a

percentage of the induced activity of a positive control (E4CAT) which

corresponds to that of the wild type E4 promoter.

69

Adenovirus E4 Promoter

EivF E4TF1 EivF

TATA

-325 -240 -174 -50 -38 +32

E4A38 ItataI 1 CAT

E4CAT TATA CAT

VIPE4

E2CAT

>TATA

TATA

Figure 2.1.0 A: Structure of the E4A38, E4CAT, VIPE4 and E2CAT test promoters. E4CAT (Lee et al , 1989) and VIPE4 (Lee et al , 1989) have been described elsewhere. E4CAT contains E4 promoter sequences (positions -240 to 4-32) fused to the chloramphenicol acetyltransferase (CAT) gene at position 4-32 relative to the E4 transcription start site. E4A38 contains E4 promoter sequences from -38 to 4-32 fused to the CAT gene. VIPE4 contains sequences between positions -94 and -61 of the vasoactive intestinal peptide (VIP) promoter (the cyclic AMP response element [CRED inserted into E4A38 at position -38 of the E4 promoter sequences.

E4A38Sac contains a S a d linker inserted at position -38 of the E4 promoter sequences in E4A38. E2CAT was obtained by cloning an oligonucleotide containing a duplicated ATE binding site from the adenovirus E2A promoter (cAACTACGTCATCTCCAACTACGTCATCT- CCgagct; E2A sequences are in uppercase, linker sequences are in lowercase, and the ATF binding sites are underlined) into the S a d site of E4A38SacCAT.

70

y / / /# # # # • • t •

E1A: - + - 4- - -k - 4-

Figure 2.1.0 B: Activation of synthetic promoters by ElA; Representative CAT assays. The test promoter is indicated above each lane and the presence (+) or absence (-) of ElA is indicated at the bottom of the figure. A 5 pg sample of the test plasmid was introduced into HeLa cells (grown on 60 mm plates at -30% confluence) with 10 pg of pSP64 (minus ElA) or 5 pg pSP64 plus 5 pg pH3G (plus ElA).

100 n

<

+E 1A

E4C A T E4A38 E2C A T

Promoter

Figure 2.1.0 C: Activation of synthetic promoters by ElA; Quantitation of CAT assays. Relative promoter activity (an arbitary scale reflecting percent converted ^^C-chloramphenicol per milligram protein) is charted for each promoter in the presence and absence of ElA, with the activity of E4CAT in the presence of ElA representing 100% activity.

72

As reported previously, ElA efficiently activates the E4 promoter in

this assay (31 fold activation by ElA). In addition, ElA also activates the

E2CAT promoter (9 fold activation), although to a lesser extent than

E4CAT. The corresponding promoter that is lacking the ATF binding sites

and contains only the E4 TATA box and CAP region (E4A38) is essentially

inactive in this assay.

Although E2CAT is induced by ElA, the activity of this promoter is

much lower than that of the E4 promoter (less than 10%). The difference

in the activities of these two promoters may be due to the presence of

binding sites for other cellular factors (such as SPl) and the additional ATF

binding sites that are present in E4CAT but not E2CAT (see Figure 2.1.0 A).

However, despite the differences in the overall activity of these two

promoters, these results demonstrate that an ATF binding site that does

not bind E4F can mediate frans-activation of the E4 promoter by ElA.

2 .1 .1 Spl binding sites are unable to mediate efficient trans-activation of

the E4 promoter by ElA.

A variety of elements other than ATF or E4F binding sites have

been shown to mediate trans-activation by ElA. Weintraub and Dean

(1992) demonstrated that both ATF and Spl binding sites could

individually confer ElA inducibility upon a heterologous promoter

containing the SV40 TATA box. In contrast, other studies have suggested

that Spl binding sites present in the E4 promoter are not necessary for its

activation by ElA, since deletion of these sites only reduces the overall

activity of the promoter and not its inducibility by ElA (Gilardi and

Perricaudet, 1984). However, this conclusion does not take into account

the possibility that more than one E4 promoter element is able to mediate

73

E4CAT ATF mm TATA CAT

E2CAT %BTATA I 1 C A T ^ ^

SplCAT TATA CAT

Figure 2.1.1 A: Structure of the SplCAT test promoter. E2CAT and E4CAT are shown for comparison and are described in Eigure 2.1.0 A. SplCAT was obtained by cloning an oligonucleotide containing a duplicated Spl binding site (AAATGGGGCGGGGCTTAGAAATGGGGCG- GGGCTTAgagct; linker sequences are in lowercase, the Spl binding sites are underlined) into the Sad site of E4A38SacCAT.

74

SplCAT E4CAT

# eElA: - + - +

Figure 2.1.1 B: Activation of the synthetic promoter SplCAT by ElA; Representative CAT assays. The test promoter is indicated above each lane and the presence (+) or absence (-) of ElA is indicated at the bottom of the figure. A 5 pg sample of the test plasmid was introduced into HeLa cells (grown on 60 mm plates at -30% confluence) with 10 pg of pSP64 (minus ElA) or 5 pg pSP64 plus 5 pg pH3G (plus ElA).

100 n-E lA

+E 1A8 0 -

> 6 0 -u<

40 -

2 0 -

E4C A T S P IC A T

Promoter

Figure 2 .1 .1 C: Activation of the synthetic promoter SplCAT by ElA; Quantitation of CAT assays. Relative promoter activity (an arbitary scale reflecting percent converted chloramphenicol per milligram protein) is charted for each promoter in the presence and absence of ElA, with the activity of E4CAT in the presence of ElA representing 100% activity.

76

frans-activation by ElA, and that SPl binding sites can indeed mediate this

response in the absence of other upstream promoter elements.

To address whether Spl binding sites are interchangeable with ATE

binding sites in the context of the E4 promoter, I constructed a synthetic

promoter that contains a duplicated binding site for Spl upstream of the

E4 TATA box and CAP region (SplCAT). SplCAT therefore corresponds to

E2CAT, with the Spl sites replacing the ATE binding sites from the E2A

promoter (see Figure 2.1.1 A).

SplCAT activity was assayed in HeLa cells in the presence and

absence of ElA. Representative autoradiograms of the assay are shown in

Figure 2.1.1 B and the data is shown graphically in Figure 2.1.1 C. Like

E2CAT, the induced activity of SPICAT is much lower than that of E4CAT

(less than 15% as active). However in contrast to E2CAT, SplCAT is only

induced 3 fold by ElA, compared to the 9 fold induction of E2CAT. This

result demonstrates that activation of E2CAT by ElA is specific to the ATE

binding sites from the E2A promoter and is not a general effect of placing a

promoter element upstream of the E4 TATA box. This would suggest that

the E4 promoter does have specific requirements for activation by ElA and

that these are distinguishable from the requirements of other E lA

inducible promoters.

2 .1 .2 E4F binding is not required for activation of the intact E4 promoter by

ElA.

The results described above suggest that E4F binding is not required

for activation of the intact E4 promoter by ElA. However, the activity of

E2CAT is far less than the E4 promoter (approximately 5% compared to

E4CAT) suggesting that this synthetic promoter may not be representative

of the activity of the intact E4 promoter. To properly assess the

77

requirements for activation of the E4 promoter whilst avoiding the

limitations of the experiments described above, I constructed a series of

promoters that correspond to the normal E4 promoter, but which contain

point mutations in their ATF/E4F binding sites. The mutations used in

this study have already been characterised for their effects on the binding

of E4F and ATE in an in vitro DNA binding assay (Rooney et ah, 1990) and

are summarised in Table 1.8.1.

pWE4WT corresponds to the normal E4 promoter and was

constructed by the ligation of synthetic oligonucleotides into a vector

containing the cat reporter gene, to recreate E4 promoter sequences

between +32 and -174. This region of the E4 promoter has previously been

shown to give activity equivalent to that of the entire E4 promoter (Gilardi

and Perricaudet, 1984; Lee and Green, 1987). Base substitutions in the E4

sequence were made to create unique restriction sites in the

oligonucleotides in order to facilitate the construction of the mutant

promoters and were chosen to minimise the sequence differences between

these promoters and the normal E4 promoter (see Figure 2.1.2 D).

WE4PM1, WE4PM2 and WE4PM4 correspond to pWE4WT but contain the

corresponding point mutation in each of the two ATF/E4F binding sites

present at positions -50 and -163 in the E4 promoter (see Figure 2.1.2 A).

The promoters were assayed in HeLa cells in the presence and

absence of FIA. Representative autoradiograms of the assay are shown in

Figure 2.1.2 B and the data is shown graphically in Figure 2.1.2 C. WF4WT

is comparable with F4CAT in overall activity and FIA inducibility.

Therefore minor changes to the F4 promoter sequence arising from the

introduction of novel restriction sites (Figure 2.1.2 D) have no effect on

F4 promoter activity. Consistent with previous studies, upstream

sequences to position -174 give activity that is comparable with that of

F4CAT. Mutation PM2 (WF4PM2) causes a substantial reduction in

78

E4CAT mm TATA CAT

IE4WT

IE4PM

CAT

—^ It a t a I 1 C A T ^ ^

WE4WT

WE4PM

TATA CAT^ (0 I^ IC A T ^

Figure 2.1.2 A: Structure of the IE4 and WE4 series of promoters. E4CAT is shown for comparison, and is described in Eigure 2.1.0 A. Restriction site overhangs were chosen so as to minimize the number of base changes introduced by the construction procedure (see Eigure 2.1.2 D). pE4Sac2CAT was constructed by inserting an oligonucleotide containing E4 promoter sequences from -33 to +32 and a Sad site at position -33, between the Hindlll and Ndel sites of RSVCAT. pIE4WT was obtained by inserting an oligonucleotide containing E4 promoter sequences from -38 to -60 and from -154 to -174 (separated by a Sall/Sacll linker) into the Sad site of pE4Sac2CAT. pIE4WT therefore contains the two ATF/E4E binding sites from the E4 promoter positioned upstream of the TATA box. pIE4PMl, 2 and 4 were constructed in the same manner with oligonucleotides harboring the corresponding point mutations. pWE4WT was obtained by ligating an oligonucleotide containing E4 promoter sequences from positions -67 to -148 to pIE4WT that had been digested with Sail and Sadi. pWE4PMl, 2 and 4 were constructed in the same manner but using pIE4PMl, 2 and 4 digested with Sail and Sadi as the source DNA.

79

y y y y• • • • # # # # # #

EMA: - + - 4 - - - h - - h - - f

Figure 2.1.2 B: Effects of point mutations on E4 promoter activity; Representative CAT assays. The test promoter is indicated above each lane and the presence (+) or absence (-) of ElA is indicated at the bottom of the figure. A 5 pg sample of the test plasmid was introduced into HeLa cells (grown on 60 mm plates at -30% confluence) with 10 pg of pSP64 (minus ElA) or 5 pg pSP64 plus 5 pg pH3G (plus ElA).

u<

120-

100-

8 0 -

6 0 -

4 0 -

20-

-E lA

+ E1A

IE4PM 1 IE4PM 2 IE4PM 4 IE4W T W E4PM 1 W E 4PM 2 W E 4P M 4 E 4C A T

Promoter

Figure 2.1.2 C: Effects of point mutations on E4 promoter activity; Quantitation of CAT assays. Relative promoter activity (an arbitary scale reflecting percent converted [ C] chloramphenicol per milligram protein) is charted for each promoter in the presence and absence of ElA, with the activity of E4CAT in the presence of ElA representing 100% activity.

81

-174 ATF/E4F *** ATF -136pE4CAT GGAAGTGACGTAACGTGGGA a a a CGGAAGTGACGATTTG

pWE4WT GGAAGTGACGTAACGTGGGA cca CGGAAGTGACGATTTG

SacII

-72 ATF/E4F ** TATA -22TTTTTTGT g GACTTTAACCGTTACGTCATTTTTTAG t C CTATATATACTCG

TTTTTTGT C GACTTTAACCGTTACGTCATTTTTTAG C t CTATATATACTCG

Sail

F igure 2.1.2 D: Sequence differences between pE4CAT and pWE4WT. Base changes introduced into the E4 sequence during the construction of pWE4WT are indicated by an asterix and lowercase letters, and the unique restriction sites created are underlined. ATF and ATF/E4F binding sites are in bold type and positions with respect to the start site of E4 transcription are noted.

82

activity (-10% compared to WE4WT) but remains ElA inducible, although

to a lesser extent (10 fold induction). Mutation PMl (WE4PM1) results in

very low activity (-5% compared to WE4WT) but remains somewhat ElA

inducible (-5 fold induction). In contrast the mutation PM4 (WE4PM4) has

no effect on overall promoter activity or ElA inducibility.

The effect of mutations PMl and PM2 demonstrates that the activity

of the promoter is dependant at least in part on the two ATF/E4F binding

sites and that their functioning is critical for normal promoter activity. In

addition, the lack of an effect of mutation PM4 on E4 promoter activity

demonstrates that binding of E4F to these sites is not required for

activation of the intact E4 promoter by ElA.

2.1.3 Upstream promoter elements other than the ATF/E4F binding sites

are not required for activation by ElA.

The reconstructed promoters described above contain other

upstream promoter elements in addition to the two ATF/E4F binding

sites, including an ATF binding site at position -147. To analyse the

requirement of these additional promoter elements, I tested the activity of

another series of promoters (IE4WT, PM l, PM2 and PM4). IE4WT

corresponds to WE4WT but lacks sequences between -67 and -148 and

therefore only contains downstream sequences inclusive of the TATA box

and CAP region, plus the two ATF/E4F binding sites (Figure 2.1.2 A).


absence of ElA. IE4WT is efficiently induced by ElA (Figure 2.1.2 C, 15 fold

induction) but is somewhat reduced in overall activity (-25% of E4CAT).

The effects of mutations PMl, PM2 and PM4 on this series of promoters

mirror the effects of these mutations on the WE4 series of promoters, but

are correspondingly lower in overall activity; PMl and PM2 give rise to

83

both a substantial reduction in promoter activity and ElA inducibility,

whereas PM4 has no effect on either promoter activity or ElA inducibility.

The reduction in the activity of this series of promoters caused by

removing the E4 promoter sequences between -67 and -148 (compare

IE4WT with E4CAT, Figure 2.1.2 C) could be due to the ATF site at -147 or

the SPl site at position -104 (Figure 2.1.0 A). These results demonstrate that

upstream elements other than the ATF/F4F binding sites are not necessary

for activation of the F4 promoter by FIA. However, these elements do

contribute to the overall activity of the promoter by stimulating both the

basal and induced activities.

2.1.4 F4F binding sites are not required for activation of the F4 promoter by

FIA in Vero cells.

As noted previously, a recent study by Rooney et al. concluded that

F4F binding sites are required for activation of the F4 promoter by FIA,

whereas ATF binding sites are insufficient. These conclusions were based

on the results of experiments using an heterologous promoter, consisting

of F4F binding sites placed upstream of the SV40 TATA box, which was

tested for its ability to mediate FIA inducibility in Vero cells. To determine

whether the different cell types might account for the differences between

my results and those of Rooney et ah, I analysed the activity of the

F4 promoter mutants in Vero cells.

The activity of the IF4WT and IF4PM4 promoters (see

Figure 2.1.2 A) was assayed in Vero cells in the presence and absence of

FIA. Representative autoradiograms of the assay are shown in

Figure 2.1.4 A and the data is shown graphically in Figure 2.1.4 B.

Mutation PM4 (IF4PM4) had no effect on either basal or FlA-induced

promoter activity in Vero cells when compared to IF4WT. This result

84

IE4WT IE4PM4

# •

# # * eElA: 4- - +

Figure 2.1.4 A: Effect of cell type on the requirement for E4F for activation of the E4 promoter by ElA; Representative CAT assays. The test promoter is indicated above each lane and the presence (+) or absence (-) of ElA is indicated at the bottom of the figure. A 5 pg sample of the test plasmid was introduced into Vero cells (grown on 60 mm plates at -30% confluence) with 10 pg of pSP64 (minus ElA) or 5 pg pSP64 plus 5 pg pH3G (plus ElA).

Vero cells

100-

80-

•S 60-u <

4 0 -

2 0 -

-E lA

i. 1 +E 1A

IE4W T IE4PM 4

Promoter

Figure 2.1.4 B: Effect of cell type on the requirement for E4F for activation of the E4 promoter by ElA; Quantitation of CAT assays. Relative promoter activity (an arbitary scale reflecting percent converted [^^C] chloramphenicol per milligram protein) is charted for each promoter in the presence and absence of ElA, with the activity of IE4WT in the presence of ElA representing 100% activity.

86

demonstrates that, as is the case in HeLa cells, E4F binding sites are not

required for activation of the E4 promoter by ElA in Vero cells. Therefore

the differences between the results presented here and those of Rooney

et al. are not due to the different cell lines used in these two assays.

87

2.2 Discussion.

Because the E4 promoter is able to interact with multiple cellular

transcription factors, it has been difficult to assess which of these factors

are responsible for mediating trans-a.ctivation of the E4 promoter by ElA.

Both E4F (Rooney et al, 1990) and ATF-2 (Liu and Green, 1990) have been

suggested to play critical roles in this process and I have examined

whether the binding of either of these factors correlates with E4 promoter

activity in HeLa cells. My results can be summarised as follows. The

ATF/E4F binding sites present upstream of the E4 TATA box are critical

for activation of the E4 promoter by ElA. However, the ability of these

sites to bind E4F does not correlate with trans-activation by ElA, whereas

ATF binding sites alone are sufficient to mediate trans-activation.

These results are in contrast to those of Rooney et a l (1990) who

demonstrated that in the context of a heterologous promoter containing

the SV40 TATA box, E4F binding sites are essential but ATF binding sites

are insufficient for mediating trans-activation by ElA. The overall activity

of these heterologous promoters is extremely low compared to that of the

E4 promoter (RE4WT is less than 4% as active in the presence of ElA;

Jones and Lee, 1991), which together with the contradiction between the

results presented here and those of Rooney et a l , demonstrates the

limitations of analysing the effects of a promoter element upon the

activity of a synthetic promoter. More significantly, a previous report has

demonstrated that ATF binding sites can confer quite distinct properties

when placed in different promoters (Lee et a l , 1989). Therefore the

rationale for using a heterologous promoter to investigate the specific role

of ATF/E4F binding sites in trans-activation of the E4 promoter by ElA

(Rooney et al, 1990) is highly questionable.

88

Despite their limitations, the results of Rooney et al. (1990)

demonstrate that E4F does play a role in this process under some

circumstances, and indeed there is significant circumstantial evidence

suggesting that E4F is involved in mediating frans-activation of the

E4 promoter by E lA (see Chapter 1; Raychaudhuri et ah, 1989;

Raychaudhuri et ah, 1987). Therefore the obvious interpretation of these

results is that ElA can target more than one cellular transcription factor to

activate the E4 promoter. E4F and ATF may be equally able to mediate

frflns-activation of the E4 promoter by ElA, but either factor may be

sufficient in the absence of the other.

There are several reasons for the possible differences in the

activities of the heterologous promoters examined by Rooney et ah and

the E4 promoter. Firstly, it is possible that ATF binding sites can only

confer ElA inducibility in a relevant promoter context, whereas E4F

binding sites do not share the same promoter dependent constraints. For

example, the presence of another E4 promoter element involved in

mediating frans-activation of the E4 promoter by ElA is suggested by the

results of Lee et ah (1989). The cyclic AMP response element (CRE) of the

vaso-intestinal polypeptide (VIP) promoter does not respond to ElA in its

natural context, but when placed upstream of the E4 promoter this

element mediates frans-activation by ElA (Lee et ah, 1989). Hence it is

possible that some other element within the E4 promoter affects the

activity of ATF binding sites such that they can mediate activation by ElA.

Alternatively, functional differences between the TATA boxes of these

promoters may affect the activity of ATF binding sites. The nature of

functional differences between different TATA boxes is not yet

understood, but is likely to reflect the DNA binding activity of general

factors other than TBP and to involve sequences surrounding the defined

TATA box. At least two different TFIID complexes have been described

89

(Timmers and Sharp, 1991) that differ in their complement of TBP

associated factors (TAFs) and it is possible that these factors may contribute

to the sequence recognition of the TATA region. Indeed, other factors

within the TFIID complex contact DNA (Purnell and Gilmour, 1993)

although the nature or specificity of these interactions has not yet been

defined. In addition, TAFs have been shown to act as adaptor proteins in

the activation of transcription by some classes of activator proteins (Tanese

et a l , 1991). Hence a requirement for a specific TFIID complex at a TATA

box would result in a particular complement of TAFs and hence specific

adaptor properties at that promoter. Different adaptor requirements for

ATF and E4F mediated trans-activation by ElA may therefore be

manifested as requirements for a particular TATA box.

Secondly, I cannot rule out the possibility that differences in the

experimental assays used to determine the activity of test promoters in

this study and by Rooney et al. may account for these conflicting results.

One such difference that I have not addressed experimentally is that

unlike the Vero cells used in this study, the assay performed by Rooney

et al. used Vero cells that had been grown under conditions of serum

starvation. Therefore it is possible that dependence on E4F binding for

trans-activation by ElA is linked to the cellular growth state.

It has been proposed that transcriptional activation of the

E4 promoter occurs through an ElA dependent increase in the DNA

binding activity of E4F (Raychaudhuri et al , 1987). This is thought to be

due to a change in the phosphorylation status of E4F that is mediated by an

ElA dependent cellular kinase (Raychaudhuri et al, 1989). Similarly, ElA

trans-activation of the polymerase Ill-transcribed VA RNA genes in vivo

occurs through an ElA dependent phosphorylation of TFIIIC that is

accentuated when cells are grown under low serum conditions (Hoeffler

et al., 1988b). This ElA dependent phosphorylation results in an increase

90

in the association rate of TFIIIC with the VA promoter and produces a

more salt resistant TFIIIC/promoter complex. In addition, these

physiological changes in the DNA binding activity of TFIIIC correlate with

the stimulation of transcription from the VA promoter by the

phosphorylated form of the protein. If TFIIIC and E4F are phosphorylated

by the same cellular kinase, these observations suggest that the activity of

E4F could be regulated by cellular growth factors and might therefore be

influenced by the proliferative state of infected cells.

91

Chapter 3: Determination of the DNA binding specificity of an

N-terminally truncated ATF-2 protein in vitro.

Previous studies have suggested that one member of the ATF

family, ATF-2, mediates activation of the E4 promoter by ElA (Flint and

Jones, 1991; Liu and Green, 1990). This effect is specific to ATF-2 since

another member of the ATF family, ATF-1, was found to be unable to

mediate an ElA-response in the same assay (Flint and Jones, 1991; Liu and

Green, 1990). The results described above are consistent with this

observation, since I have shown that ATF binding sites are required for

trans-activation of the E4 promoter by ElA. However, the binding affinity

of ATF-1 and ATF-2 for the sites characterised in vivo has not previously

been examined. By determining the specificity of binding of these two

members of the ATF family for the mutated ATF/E4F binding sites

described in the above experiments, I will be able to further establish the

nature of the factor(s) required for activation of the E4 promoter by ElA.

92


3.0.1 In vitro transcription and translation of ATF proteins.

ATF RNA was synthesised in vitro using the T7 transcription

system. A plasmid consisting of the relevant ATF cDNA cloned

downstream of a T7 promoter was linearised by cutting with a restriction

enzyme 3' to the translation stop signal. The cut plasmid was then

extracted with phenol/ chloroform and ethanol precipitated. The DNA was

resuspended in water that had been autoclaved, treated with 0 .2 % diethyl

pyrocarbonate (DEPC) and then re-autoclaved. All further manipulation of

the DNA and RNA were performed using fresh DEPC treated water,

double autoclaved pipette tips and tubes and whilst wearing clean gloves,

in order to minimise contamination of the RNA preparation with

RNases.

4 |ig of the linearised plasmid DNA was added to a transcription

reaction containing; 40 mM Tris-HCl pH 7.5; 6 mM MgCl2; 2 mM

spermidine; 100 |ig /m l BSA (RNase free; Boehringer Mannheim); 10 mM

DTT; 100 |iM GTP; 500 [iM each of ATP, CTP and UTP; 500 |iM CAP

analogue (m^GpppG; Boehringer Mannheim); 100 u RNasin (Boehringer

Mannheim); 10 u T7 RNA polymerase (Boehringer Mannheim) in a total

volume of 80 pi. The reaction was incubated for 1 h at 40°C, after which

the plasmid DNA was removed by the addition of 40 u of DNase I (RNase

free; Boehringer Mannheim) and incubation at 37°C for 10 min. 15 pg

tRNA and 1.5 vols, of 0.7 M NH 4 OAC were added, followed by extraction

with phenol/chloroform and ethanol precipitation of the RNA. The RNA

pellet was resuspended in 10 pi DEPC water and 1 pi was checked on a 2%

agarose gel.

Protein was synthesised from this RNA preparation in vitro using a

lysate from rabbit reticulocytes (Boehringer Mannheim). To test the RNA

93

preparation as a template for translation, 1 pi was incubated with 24 pi of

the lysate and 30 pCi P^SJ-methionine for 1 h at 30°C. The translation

reaction was then run on a 10% SDS/PAGE gel (see Chapter 3.0.2) and the

synthesised ATF protein was visualised by autoradiography. For a large

scale translation, 12 pi RNA was incubated with 380 pi of lysate and

120 pCi p5S]-methionine for 1 h at 30°C. Aliquots of 20 pi the translation

reaction were quick frozen in liquid nitrogen and stored at -80°C.

3.0.2 SDS/polyacrylamide gel electrophoresis of proteins.

SDS/polyacrylamide gels were prepared with a resolving gel of 10%

acrylamide (at 37.5:1 acrylamideibisacrylamide; in a buffer containing

375 mM Tris-HCl pH 8 .8 ; 0.1% SDS) and a stacking gel of 5% acrylamide (at

37.5:1 acrylamide:bisacrylamide; in a buffer containing 125 mM Tris-HCl

pH 6.8; 0.1% SDS) and were cross linked by the addition of 1 / 80th vol. 10%

ammonium persulphate; l/lOOO i vol. of TEMED, poured into a vertical

gel assembly (Protean, BioRad) and allowed to set at room temperature.

The protein sample was mixed with an equal volume of loading buffer

(125 mM Tris-HCl pH 6.8; 20% Glycerol; 4% SDS; 10% p-mercaptoethanol;

0.005% bromophenol blue; 0.25% xylene cyanol) and denatured by boiling

for 2 min. The sample was then loaded on the gel and run at 45 mA for

approximately 2 h in Ix running buffer (25 mM Tris; 190 mM Glycine;

0.1% SDS).

Proteins were visualised by staining the gel with coomassie blue.

The gel was soaked in staining solution (0.2% coomassie blue; 50%

methanol; 7.5% glacial acetic acid) for 10-20 min and then destained in 10%

methanol; 7.5% glacial acetic acid until protein bands were clearly

distinguished from the background staining.

94

3.0.3 Analysis of the DNA binding activity of proteins using a gel

retardation assay.

A DNA fragment containing an ATF binding site was excised from

a plasmid using relevant restriction enzymes and isolated from a 2 %

agarose gel by electroelution (see Chapter 2.0.2.3). 50 ng of the purified

DNA fragment was labelled with 20 pCi [a^^F]-dATP or [a^^P]-dCTP using

10 u AMV reverse transcriptase (Boehringer Mannheim) for 30 min at

37°C in 20 pi of a reaction buffer specified by the manufacturer. The

reaction mixture was made up to 50 pi total volume and extracted with

phenol/chloroform before use.

1 ng of the DNA probe was incubated with the protein sample for

15 min in 15 pi of a buffer containing 10 mM Tris-HCl pH 7.5; 60 mM

NaCl; 1 mM DTT; 1 mM EDTA; 5% glycerol, either in the presence or

absence of a competitor oligonucleotide. 2 pi of loading buffer (50%

glycerol; 0.25% bromophenol blue; 0.25% xylene cyanol) was then added

and the reaction was loaded on a neutral acrylamide gel (5% acrylamide at

50:1 acrylamide:bisacrylamide; 6.7 mM Tris-HCl pH 8.3; 1 mM EDTA;

3.3 mM NaOAc). Gels were run in Ix retard buffer (6.7 mM Tris-HCl

pH 8.3; 1 mM EDTA; 3.3 mM NaOAc) at 45 mA for approximately 2 h, with

continuous circulation of the electrophoresis buffer.

95

3.1 Results.

3.1.0 ATF-1 and ATF-2 bind specifically to ATF binding sites present in

various adenovirus early promoters.

To determine whether the requirement for an ATF binding site for

activation of the E4 promoter by ElA is consistent with a requirement for

ATF-2, I analysed the specificity of ATF-2 for ATF binding sites and

compared this to the binding specificity of another ATF protein, ATF-1. I

was not able to efficiently produce full length ATF-2 protein using a

T7 polymerase/rabbit reticulocyte lysate in vitro system (data not shown)

so an N-terminally truncated form of the protein (DL2; Benbrook and

Jones, 1990) which was efficiently produced in this system was instead

used for these assays. This truncated protein lacks amino acids 1-232, but

still contains the leucine zipper and basic domains (located between amino

acids 350-410) that are thought to be sufficient for sequence specific DNA

binding activity.

Studies of the binding specificity of ATF-1 and ATF-2 were

performed using an in vitro gel-mobility shift assay. ATF proteins were

produced by the in vitro transcription and translation of their cDNA and

bound to a 32p_labelled DNA probe containing an ATF binding site. When

run on a neutral polyacrylamide gel, the resulting D N A /protein

complexes are retarded compared to the unbound probe, due to their

slower mobility on the gel (Figure 3.1.0). No binding is observed when

ATF-1 (data not shown) and ATF-2 (Figure 3.1.0 B) mRNA is omitted from

the translation reaction, demonstrating that the protein/DNA complexes

formed are dependent upon ATF proteins.

Binding specificity was analysed by using a competition assay in

which unlabelled oligonucleotides were added to the binding reaction to

96

B

COMPETITOR

COMPLEXlATF1

UNBOUNDDNA

ex CM ^ ^ ^ LU UJ UJ UJ lü

nLU

i nf I

1 2 3 4 5 6 7 8

C - ^A T F 2

1 2

s 3 sa CM ^ ^ CO «

LU UJ LU UJ UJ UJ

3 4 5 6 7 8 9

Figure 3.1.0: (A) Binding of ATF-1 to adenovirus early promoters. Gel mobility shift assays were performed with ATF-1 synthesized in vitro in rabbit reticulocyte lysate and incubated with a ^^P-labelled DNA probe containing an ATF binding site. Unbound DNA and ATF-1/DNA complexes are indicated on the left. Assays were performed in the presence of a 200 fold excess of competitor DNA as indicated above the lanes. Competitor DNAs are described in the text.

(B) Binding of ATF-2 to adenovirus early promoters. ATF-2 protein was synthesized in vitro, and gel mobility shift assays were performed as described for ATF-1. Unbound DNA (U) and ATF-2/DNA complex (C) are indicated at the left. Lane 1, incubation with ATF-2 in the absence of competitor DNA; Lane 2, incubation of 32p-labelled DNA probe with rabbit reticulocyte lysate in the absence of ATF-2.

compete with the labelled probe for binding of ATF proteins. ATF-1

(Figure 3.1.0 A) and ATF-2 (Figure 3.1.0 B) bind to several competitor

oligonucleotides containing ATF binding sites found in the adenovirus

E2A (E2WT) and E3 (E3WT) promoters or a consensus ATF binding site

(E4WT). This is demonstrated by the absence of the shifted probe in these

lanes of the gels. However neither protein binds to the corresponding

competitor oligonucleotides carrying a mutation or deletion of the ATF

binding site (E2LS, E3M, E4M1 and E4M2), as is demonstrated by the

presence of the shifted probe. Thus ATF-1 and ATF-2 specifically bind to

all of the ATF binding sites tested. This further demonstrates the

difference in the binding specificity between E4F and specific members of

the ATF family, since E4F does not bind to ATF binding sites found in the

adenovirus E2A and E3 promoters.

3.1.1 The effect of point mutations on the binding specificity of ATF-1 and

ATF-2.

By characterising the DNA binding specificity of ATF-1 and ATF-2

using the point mutations that affect E4 promoter activity, I will be able to

determine whether the binding specificity of either protein is consistent

with their being involved in activation of the E4 promoter by ElA.

Binding specificity was analysed by adding increasing amounts of

unlabelled competitor oligonucleotides containing certain point

mutations, to a binding assay using a 32p_labelled DNA probe containing

an ATE binding site. Similar to the assay described above, an unlabelled

competitor that binds the protein efficiently will cause a reduction in the

amount of shifted probe on the gel with an increasing amount of

competitor. An unlabelled competitor containing a mutated site that

cannot bind the protein will have no effect on the shifted probe. The

98

ATF1

MOLAR EXCESS 60 , 240 , 960 X

#= liiwt

U

WT PM1 PM2 PM4 PM5 PM6

COMPETITOR

Figure 3.1.1 A: Effects of point mutations on ATF-1 binding. Gel mobility shift assays were performed as described in the legend to Figure 3.1.0 A. Unbound DNA (U) and ATF-1/DNA complexes (C) are indicated at the left. Assays were performed in the presence of increasing amounts of unlabelled oligonucleotide competitors containing either a wild type (WT) ATF binding site from the F4 promoter or the point mutations described in Table 1.8.1 as indicated at the bottom of the figure. The molar excess of competitor oligonucleotide is indicated at the top of the figure.

MOLAR EXCESS : 50150500.

ATF2

COMPETITOR: W T P M 4

0 20 60180

* - # - BC:. BBUD

PM 1 P M 2 P M 4 P M 5 P M 6 P M 5 /6

Figure 3.1.1 B : Effects of point mutations on ATF-2 binding. Gel mobility shift assays were performed as described in the legend to Figure 3.1.0 B. Unbound DNA (U) and ATF-2/DNA complexes (C) are indicated at the left. Assays were performed in the presence of increasing amounts of unlabelled oligonucleotide competitors containing either a wild type (WT) ATF binding site from the E4 promoter or the point mutations described in Table 1.8.1 as indicated at the bottom of the figure. The molar excess of competitor oligonucleotide is indicated at the top of the figure.

molar excess of competitor oligonucleotide containing the wild type ATF

binding site required to compete for binding of ATF proteins to the probe

DNA was deduced from the results of several preliminary experiments

(data not shown). A range of amounts of competitor DNA up to this pre

determined molar excess of the wild type competitor were then compared

for binding to competitor DNAs containing mutated the ATF binding

sites.

In this assay ATF-1 binds the competitor oligonucleotide containing

mutation PM4 as efficiently as that containing the wild type (WT) ATF

binding site (see Figure 3.1.1 A). Oligonucleotides containing mutations

PMl and PM2 compete very weakly and oligonucleotides containing

mutations PM5 or PM6 do not compete at all in the range tested (see

Figure 3.1.1 A). Thus PMl, PM2, PM5 and PM6 are strong down mutations

for ATF-1 binding whereas mutation PM4 has no effect. The lack of effect

of the PM4 mutation on ATF-1 binding further establishes the different

binding specificity of ATF-1 and E4F.

The effect of point mutations on ATF-2 binding are shown in

Figure 3.1.1 B. As is the case for ATF-1 binding, an oligonucleotide

containing mutation PM4 competes for ATF-2 binding as efficiently as the

wild type (WT) oligonucleotide. However, oligonucleotides containing

mutations PMl or PM2 also compete efficiently for ATF-2 binding. My

supervisor. Dr. Kevin Lee, also demonstrated that oligonucleotides

containing mutations PM5 or PM6 compete less efficiently for ATF-2

binding and that the effect of these mutations is confirmed by the inability

of an oligonucleotide containing the double mutation (PM5/6) to compete

for ATF-2 binding (Jones and Lee, 1991; see Figure 3.1.1 C). The lack of an

effect of PMl and PM2 on the binding of ATF-2 suggests that the binding

specificity of ATF-2 does not correlate with activation of the E4 promoter

by ElA in vivo.

101

3.1.2 The effect of point mutations on the stability of ATF-2 binding.

The above experiments demonstrate the affinity of a protein for a

particular binding site, but they give no indication of the stability of

binding. This can be assessed by binding the protein to a ^^P-labelled DNA

probe containing the relevant site in the absence of competitors. An

unlabelled competitor oligonucleotide containing a site known to bind the

protein efficiently is then added for varying lengths of time before loading

the binding reaction on the gel. The stability of binding can be expressed as

the time it takes the competitor to reduce the amount of shifted probe by

50% (tl/2 ).

To further assess the effect of these point mutations on ATF-2

binding my supervisor. Dr. Kevin Lee, examined the stability of ATF-2

binding to ^^P-labelled DNA probes containing mutations PMl, PM2 and

PM4 (Jones and Lee, 1991; see Figure 3.1.2). Before the addition of

competitor oligonucleotides (time=0) ATF-2 binds equally well to probes

containing the mutations PMl, PM2 and PM4, thus confirming the lack of

effect of mutations PMl and PM2 on ATF-2 binding. Upon addition of

competitor oligonucleotides, ATF-2 dissociates from each probe at a

similar rate, demonstrating that mutations PMl and PM2 do not affect

ATF-2 binding stability.

In addition to further demonstrating that mutations PMl and PM2

have no effect upon the binding affinity of ATF-2, this result also

establishes that these mutations have no effect upon the stability of

binding of ATF-2. Therefore the effects of these mutations on the DNA

binding specificity of ATF-2 do not correlate with their effects on the

activity of the E4 promoter in vivo.

102

TIME : 0 0 0 10 30120

ATF2

U

J L J L J L

PROBE : PM4 PM1 PM2 PM4

Figure 3.1.2: Stability of ATF-2 DNA binding to point mutated ATF/E4F binding sites. ^^P-labelled oligonucleotide containing mutations PM l, PM2 and PM4 (indicated at the bottom) were used to examine the effects of these mutations on the stability of ATF-2 binding. Gel mobility shift assays for ATF-2 were performed as described in the legend to Figure 3.1.0. At various times before the samples were loaded on the gel (time in minutes at the top of the figure), an excess of specific competitor DNA (PM4 oligonucleotide) was added to the binding reaction. Lane 1, control containing no competitor throughout the incubation period; Lane 2, control containing competitor before formation of stable complexes. Provided by Dr. Kevin Lee; from Jones and Lee (1991).

3.1.3 The effect of point mutations on the binding of ATF-2 to the

E4 promoter in vivo.

Contradictory with previous reports which demonstrate that ATF-2

can mediate frans-activation of the E4 promoter by ElA (Flint and Jones,

1991; Liu and Green, 1990), the results described above suggest that it is

unlikely that ATF-2 homodimers mediate activation of the E4 promoter

by ElA in HeLa cells. However, it is possible that the in vitro DNA binding

assay described above may not reflect the binding activity of ATF-2 to the

mutated E4 promoters in vivo, or alternatively that the DNA binding

specificity of the truncated ATF-2 protein is not the same as that of the

wild type protein.

Previous studies have demonstrated that ATF-2 binds to the

ATF/E4F binding sites in the E4 promoter in vivo (Liu and Green, 1990)

utilising a chimeric protein consisting of the full length ATF-2 protein

(Hai et al, 1989) fused to the acidic activation domain of VP16 (Triezenberg

et al., 1988). Activation of the E4 promoter by the ATF-2/VP16 fusion

protein was demonstrated in a transient expression assay in which the test

promoter (E4CAT) was co-transfected with a plasmid expressing

ATF-2/VP16 (pAct-CR2; Liu and Green, 1990). ATF-2 alone does not

efficiently activate transcription in this assay, and activation by the

ATF-2/VP16 fusion protein is dependent on the presence of ATF binding

sites in the test promoter (Liu and Green, 1990). In this assay, the DNA

binding domain of ATF-2 serves to recruit the acidic activation domain of

VP16 to the E4 promoter in order to activate transcription.

This assay was used to analyse the effects of point mutations PMl

and PM2 upon the DNA binding activity of ATF-2 in vivo, by measuring

activation by the ATF-2/VP16 fusion protein of a test promoter containing

the mutated ATF/E4F binding sites. Initial attempts to observe efficient

104

IE4WTxlO IE4PMlxlO IE4PM2xlO

#

ATF-2/VP16######

Figure 3.1.3 A: Activation of the IE4xlO series of promoters by an ATF-2/VP16 fusion protein; Representative CAT assays. The test promoter is indicated above each lane and the presence (+) or absence (-) of a plasmid expressing the ATF-2/VP16 fusion protein (pAct-CR2) is indicated at the bottom of the figure. A 5 ig sample of the test plasmid was introduced into HeLa cells (grown on 60 mm plates at -30% confluence) with 10 pg of pSP64 (minus ATF-2/VP16) or 5 |ig pSP64 plus 5 pg pAct-CR2 (plus ATF-2/VP16).

-A T F -2 /V P 1 6

Ü + A T F -2 /V P 1 6

IE4W T 1E4PM1 IE4PM 2 xlO xlO xlO

Promoter

Figure 3.1.3 B: Activation of the IE4xlO series of promoters by an ATF-2/VP16 fusion protein; Quantitation of CAT assays. Relative promoter activity (an arbitary scale reflecting percent converted [^^C] chloramphenicol per milligram protein) is charted for each promoter in the presence and absence of the ATF-2/VP 16 fusion protein, with the activity of IF4WTxlO in the presence of ATF-2/VP16 representing 100% activity.

106

activation of both the WE4WT and IE4WT promoters by the ATE-2/VP16

fusion protein were unsuccessful, giving rise to a very low level of

transcriptional activation (data not shown). In order to observe more

efficient activation by the ATF-2/VP16 fusion protein, I constructed a

series of promoters consisting of 5 duplicated pairs of ATE/E4F binding

sites placed upstream of the E4 TATA box and CAP region (IE4WTxlO,

PMlxlO and PM2xlO), supposing that the presence of multiple binding

sites in the test promoter may enhance activation by the ATF-2/VP16

fusion protein. These promoters was constructed by the cloning of

additional oligonucleotides containing the relevant binding sites into the

IE4 series of promoters, such that each construct contains the relevant

point mutation in each of its 10 ATF/E4F binding sites.


absence of a plasmid expressing the ATF-2/VP16 fusion protein described

above. Representative autoradiograms of the assay are shown in

Figure 3.1.3 A and the data is shown graphically in Figure 3.1.3 B.

IF4WTxlO is induced 6.5 fold by ATF-2/VP16 in this assay, demonstrating

that ATF-2 binds to the ATF/F4F binding sites in vivo. Mutation PM2

(IF4PM2xlO) causes a slight reduction in the inducibility by ATF-2/VP16

(~3 fold activation) and was less than 50% as active as IF4WTxlO in the

presence of ATF-2/VP16. Mutation PMl (IF4PMlxlO) also causes a

reduction in the inducibility by ATF-2/VP16 (4.5 fold activation) and in

addition substantially reduces the overall activity in the presence of

ATF-2/VP16 to less than 25% that of IF4WTxlO.

This experiment has only been performed once in duplicate and the

variability was such that in order to confirm the precise level of activation

of each of these promoters by the ATF-2/VP16 fusion protein this

experiment will need to be repeated. However, if there proves to be a

reproducible effect of these point mutations on activation by the

107

ATF-2/VP16 fusion protein as is suggested by the above results, I would

conclude that although ATF-2 binds to PMl and PM2, either the DNA

binding affinity and/or binding stability of ATF-2 to these mutations is

impaired to some extent in vivo. However, the small effect of these

mutations on ATF-2 binding in vivo does not correlate with their more

pronounced effects on activation of the E4 promoter in vivo. Therefore,

these results are further evidence that the DNA binding specificity of

ATF-2 does not correlate with activation of the E4 promoter in vivo.

108

3.2 Discussion

A previous study by Rooney et a l (1990) has analysed the effects of

several point mutations in the ATF/E4F binding sites of the E4 promoter

on the binding of E4F and an ATF binding activity purified from HeLa

cells. Utilising the effects of these point mutations upon the DNA binding

activity of E4F, I have shown that E4F binding is not required for trans-

activation of the E4 promoter by ElA. Liu and Green (1990) have

demonstrated that a specific member of the ATF family, ATF-2, can

mediate trans-activation by ElA and that this protein binds to the

E4 promoter in vivo. However, the precise nature of the DNA binding

specificity of ATF-2 has not previously been assessed. I have analysed the

DNA binding specificity of ATF-2 in vitro to determine whether binding

of ATF-2 to the E4 promoter correlates with its trans-activation by ElA

in vivo. My results can be summarised as follows. The DNA binding

specificity of an N-terminally truncated ATF-2 protein in vitro does not

correlate with trans-activation of the E4 promoter by E lA in vivo.

Although these point mutations may have a small effect on the DNA

binding activity of an ATF-2/VP16 fusion protein in vivo, these results are

nevertheless inconsistent with their more pronounced effects on

activation of the E4 promoter by ElA in vivo.

Previous reports have demonstrated that ATF-2 can function with

ElA to activate the E4 promoter (Liu and Green, 1990) and that ATF-2

mRNA is expressed in HeLa cells (Maekawa et a/., 1989). Therefore the

results described above which demonstrate that the DNA binding

specificity of ATF-2 does not correlate with activation of the E4 promoter

in HeLa cells is unexpected. One explanation for this result is that ATF-2 is

not present as a homodimer in HeLa cells, but only as a heterodimer with

another bZIP protein. ATF-2 is known to form heterodimers with ATF-3

109

(Hai et al., 1989) and c-jun (Benbrook and Jones, 1990) and it is possible that

either of these heterodimers could display a different affinity for the

mutated ATF/E4F binding sites to that of the ATF-2 homodimer.

It is also possible that the DNA binding assays used do not

accurately reflect the DNA binding specificity of ATF-2 in vivo, and

differences in the results from the in vitro and in vivo assays described

suggest that this may indeed be the case. These differences could be

attributed to one or more of several factors. Firstly, the in vitro gel

retardation assay utilised an N-terminally deleted ATF-2 protein which

may not display the same DNA binding properties as the full length

protein. This would suggest that another unknown functional domain,

distinct from the basic domain, can modulate the DNA binding activity of

ATF-2.

Recent evidence has demonstrated that the DNA binding activity of

CREB is influenced by its phosphorylation at a site distinct from the basic

domain (Nichols et ah, 1992). CREB contains a phosphorylation site for

protein kinase A (PKA) that is distant to the basic domain and leucine

zipper region and which is not required for DNA binding. However,

phosphorylation of this site in vivo increases the occupancy by CREB of a

weak, asymmetric binding site present in the tyrosine aminotransferase

(TAT) promoter (Nichols et ah, 1992). Hence although the PKA site is not

directly involved in DNA binding, it can influence the function of the

basic domain and indirectly affect the DNA binding specificity and/or

affinity of CREB for a particular site.

ATF-2 has been shown to contain several sites for cellular kinases

(Sakurai et ah, 1991) and hence the DNA binding activity of this protein

may also be regulated in a similar maimer. If this is the case, then different

phosphorylation patterns of the truncated ATF-2 protein made in vitro

and of the full length ATF-2 protein in vivo could account for differences

110

in the DNA binding specificity of these two proteins. Such differences in

the phosphorylation of the two proteins could be attributed either to

different compliments of kinases in HeLa cells and the rabbit reticulocyte

lysate, or to the lack of one or more phosphorylation sites in the truncated

ATF-2 protein.

Alternatively, the putative zinc finger present in the N-terminal

region may modulate the DNA binding specificity of ATF-2. The C2 H2 zinc

finger of ATF-2 shares homology to the zinc fingers of several other

proteins (see Figure 3.2) such as TFIIIA (Miller et al, 1985), Spl (Kadonaga

et a l , 1987) and YYl (Shi et al, 1991), and the zinc fingers of these proteins

are known to be involved in the sequence-specific recognition of DNA.

Unlike ATF-2, all of these other proteins contain multiple copies of the

C2H2 motif which each contribute to the overall DNA binding specificity

of the protein. Single units of these zinc fingers do not recognise specific

DNA sequences (Parraga et al, 1988) and therefore the single zinc finger of

ATF-2 may be expected not to impart sequence specificity upon the

protein. However it is possible that the zinc finger motif of ATF-2 may

contact DNA in a relatively non-specific manner such as to modulate the

DNA binding specificity conferred by the basic domain.

Secondly, it is possible that the DNA binding specificity of ATF-2

could be altered upon interaction with ElA, such that it cannot bind to the

mutated ATF/E4F binding sites. Interestingly, the hepatitis B virus (HBV)

X gene product (pX) has been reported to have a similar effect on the

binding of both ATF-2 and CREB to the HBV enhancer (Maguire et al,

1991). Neither pX, ATF2 or CREB alone can bind to the HBV enhancer;

however, pX has been shown to form complexes with either ATF-2 or

CREB and to specifically alter their DNA binding specificity such that these

complexes can then interact with the HBV enhancer (Maguire et al, 1991).

I ll

N>

ATF-2: K P F L C T A P 6 C G Q R F T N E D H L A V H K H K H E

S p l2 (1 ) K Q H I C H I Q G C G K V Y G K T S H L R A H L R W H T(2) R p F M C T W S Y C G K R F T R S D E L Q R H K R T H T

YY13 (1) R T I A c P H K G c T K M F R D N S A M R K H L H T H G(3) K P F Q c T F E G c G K R F S L D F N L R T H V R I H T(4) R P Y V c P F D G c N K K F A Q S T N L K S H I L T H A

TFIIIA4 (1) K R Y I c S F A D c G A S Y N K N W K L R A H L C K H T(2) K P F p c K E E G c D K G F T S L H H L T R H S I T H T(4) C V Y V c H F E G c D K A F K K H N Q L K V H Q F T H T(5) L P Y K c P H E G c D K S F S V P S C L K R H E K V H A(9) R P F A c E H A E c G K S F A M K K S L E R H S V V H D

Consensus K F c c K F L R H H

Figure 3.2: Comparison of the zinc finger motif of ATF-2 with several representative C2 H2 zinc fingers. Consensus residues are boxed, and homology to ATF-2 is in bold. Bracketed numbers denote the position of the zinc finger with respect to the other zinc fingers in the protein.

(1985)References: ^Hai et al. (1989); Maekawa et al. (1989); ^Kadonaga et al. (1987); ^Shi et al. (1991); ^Miller et al.

Finally, the in vivo binding assay that I have described is highly

artificial and I cannot discount either the effects of the binding of ATF-2 to

10 sites in such close proximity, or of the VP16 fusion on the binding

specificity of ATF-2. I have shown that PMl and PM2 have a strong

negative effect on binding of ATF-1 (and presumably CREB binding also).

Therefore these mutations may increase the opportunity for the

ATF-2/VP16 fusion protein to occupy these sites in vivo compared to

binding to the wild type site or to a site containing the PM4 mutation.

Hence any deleterious effect of these mutations on the DNA binding

specificity of ATF-2 in vivo may be masked rather than accentuated in the

absence of binding of other factors. Hence that I observe any effect at all of

the PMl and PM2 mutations on ATF-2 binding could be enough of an

indication that ATF-2 binding does in fact correlate with activation of the

E4 promoter in vivo.

To establish the nature of the differences in the DNA binding

specificity of these two proteins, we must first determine whether or not

the DNA binding specificity of full length ATF-2 is different from that of

the truncated protein in vitro. This was originally not possible because full

length ATF-2 could not be expressed efficiently in the in vitro system

described above or by overexpression in bacteria. However, I have since

constructed a His-tagged ATF-2 protein lacking the N-terminal 15 amino

acids that is efficiently overexpressed in E. coli (see Chapter 4.2). The effects

of functional domains of ATF-2 other than the basic domain (such as the

phosphorylation sites or the zinc finger) upon the DNA binding specificity

of the full length protein can hopefully be examined in vitro b y

mutational analysis of this recombinant ATF-2. Obviously it is possible

that this protein will still not reflect the binding specificity of the full

length protein, but since the missing 15 amino acids account for only 3%

of the proteins length, this is an unlikely possibility.

113

Chapter 4: Analysis of E4 promoter activity in vitro.

Because ATF binding sites are able to interact with multiple cellular

transcription factors (Hai et al, 1989) it is difficult to assess the role of an

individual factor in mediating trans-activation of the E4 promoter by ElA

in vivo. An in vitro transcription assay that faithfully reproduces trans

activation of the E4 promoter by ElA could be useful in examining the

role that ATF-2 plays in this process and in further analysing the

mechanisms involved. However, neither the ability of ATF-2 to activate

transcription in v it ro , nor the ability of ElA to t rans -ac t iva te the

E4 promoter in vitro, have been previously demonstrated.

Previous studies have demonstrated that the E4 promoter can be

accurately transcribed in vitro using crude nuclear extracts (Dignam et al,

1983; Lee and Green, 1987). Lee and Green (1987) demonstrated that the

E4 promoter was extremely transcriptionally active in such an assay and

was at least 1 0 -fold more active than the highly efficient adenovirus major

late promoter (MLP). When low amounts of promoter template were used

in this assay, the ratio of E4/MLP transcription was found to be particularly

high (Lee and Green, 1987).

It is now clear that the majority of the ATF binding activity present

in HeLa cells is attributable to ATF-1 and CREB. In vitro DNA binding

assays in which nuclear extracts from HeLa cells are incubated with a

32P-labelled DNA probe containing an ATF binding site give rise to

complexes that correspond to those produced by ATF-1, CREB and their

heterodimer (Hurst e ta l , 1991). DNA affinity purification of ATF binding

proteins from HeLa cell nuclear extracts results in high yields of both

43 kDa and 47 kDa proteins (relative to other proteins) that correspond to

ATF-1 and CREB by several criteria (Hai et al, 1988b; Hurst et al, 1990).

These affinity purified proteins and cloned CREB have been shown to

114

stimulate transcription in vitro (Gonzalez et al, 1991; Hai et al , 1988b) and

cloned CREB and ATF-1 have been shown to activate transcription

in vivo (Flint and Jones, 1991; Gonzalez e t a l , 1991; Lee e t a l , 1990;

R ehfuss e t a l , 1991). Therefore it is likely that ATF-1 and CREB are

responsible for the high level of transcriptional activity of the E4 promoter

in vitro observed by Lee and Green (1987). However, neither of these

proteins appears to be involved in mediating trans-activation by ElA

in vivo (Flint and Jones, 1991; Liu and Green, 1990) and this may explain

why trans-activation of the E4 promoter by ElA has not previously been

demonstrated in vitro.

Studies of trans-activation of the E2A and MLP promoters by ElA

in vitro have demonstrated that the addition of recombinant ElA to a

transcription assay is sufficient to bring about activation (Spangler et al,

1987). In addition, a 49 amino acid synthetic peptide that comprises the 46

amino acids unique to the larger ElA protein, plus 3 adjacent and

conserved amino acids, was found to be capable of stimulating

transcription from the E3 and MLP promoters in vitro (Green et al, 1988;

Loewenstein and Green, 1989). However, the effect of ElA in these assays

is not as pronounced as its effect on the same promoters in vivo (for

example see; Impériale et al, 1985; Lewis and Manley, 1985). Together with

the observation that the E4 promoter is extremely active in vitro (Lee and

Green, 1987) and that this high level of activity is most likely due to CREB

and ATF-1, this suggests that it is unlikely that ElA-dependent activation

of the E4 promoter could be observed in a standard in vitro transcription

assay utilising crude nuclear extracts.

The transcriptional activity of the E4 promoter in vitro can be

reduced significantly by the addition of a specific competitor

oligonucleotide containing an ATF binding site to the nuclear extract prior

to a transcription assay (Lee and Green, 1987), due to sequestering of ATF

115

binding proteins endogenous to the nuclear extract. Transcriptional

activity can then be restored by the subsequent addition of purified ATF

proteins (Hai etal., 1988a). Using this approach, it may be possible to

develop a transcription assay that demonstrates activation of the

E4 promoter by ElA and which is dependent on the addition of

recombinant ATF-2. ATF-2 has been shown to be a relatively poor

activator in vivo (Flint and Jones, 1991; Liu and Green, 1990) and hence

addition of recombinant ATF-2 to an ATF depleted in vitro transcription

assay may be expected to have little or no effect on the activity of the

E4 promoter. Hence I may be able to demonstrate ElA dependent

activation of an ATF depleted transcription reaction by either one of two

strategies. The simplest approach would be the comparison of activation

by recombinant ATF-2 and ElA together, to that of ATF-2 alone.

Alternatively, I could compare the effect of adding recombinant ATF-2 to

an ATF depleted transcription reaction using nuclear extracts from

adenovirus infected HeLa cells to one using nuclear extracts from

uninfected cells.

116


4.0.1 In vitro transcription of the E4 promoter.

Activity of the E4 promoter was analysed using pE4CAT as a

template for transcription. This consists of the E4 promoter from positions

+240 to -38 fused to the chloramphenicol acetyl transferase (CAT) gene

(Lee et a l , 1989). Activity of the MLP promoter was analysed using pMLP

as a template for transcription. This consists of the Ball-E fragment from

adenovirus type 2 (map units 14.7-21.5) including the major late promoter,

cloned into the BamHI site of pBR322 (Manley et a l , 1980). Nuclear

extracts were prepared as described by Dignam et a l (1983) with slight

modifications and were used to direct transcription from the E4 promoter

as described by Lee and Green (1987). Briefly, 1 pg of the promoter template

was incubated with 26 pi of nuclear extract in 40 pi of a buffer containing

0.5 mM each of ATP, CTP, GTP and UTP; 7.5 mM MgCl2 ; 50 u RNasin

(Boehringer Mannheim); at 30°C for 1 h. The transcription reaction was

treated with proteinase K (at 600 pg/m l) in 260 pi of a buffer containing

0.4% NP40; 100 mM NaCl; 7.75 mM Tris-HCl pH 8 .6 ; 1.15 mM MgCl2 ; at

55°C for 30 min. The mixture was then extracted with an equal volume of

phenol/ chloroform and ethanol precipitated.

Detection of the RNA transcribed from each promoter was by a

previously described method of primer extension, using ^^P-end-labelled

synthetic oligonucleotides as probes to either the CAT or MLP products.

10 pmoles of primer was labelled with 100 pCi [y32p]-ATP using 20 u

polynucleotide kinase in 15 pi of a buffer specified by the manufacturer.

The reaction was made up to 100 pi volume and extracted with

phenol / chloroform prior to use. The RNA pellet from the transcription

reaction was resuspended in 15 pi 0.5% SDS (in DEPC water) and

hybridised to 0.1 pmole of this labelled primer in 30 pi of hybridisation

117

buffer (40 mM PIPES pH 6.4; 1 mM EDTA; 400 mM NaCl; 80% formamide)

overnight at 37°C. The primer/RNA hybrid was then precipitated by the

addition of 400 |il of 60% isopropanol; 400 mM NH^OAc and

centrifugation at 14000 rpm. The RNA pellet was washed with 70%

ethanol and resuspended in 10 pi of DEPC water. The primer was extended

with 2 u AMV reverse transcriptase (Boehringer Mannheim) in 20 pi of a

buffer specified by the manufacturer, for 1 h at 42°C. The products of the

reaction were then resolved on a denaturing (urea) polyacrylamide gel (see

Chapter 2.0.2.2) and visualised using autoradiography.

4.0.2 Expression and purification of recombinant proteins.

Recombinant CREB proteins were overexpressed in E. coli by

cloning the rat CREB cDNA into the pET vector, such that the CREB open

reading frame was fused (in frame) to pET sequences encoding a poly

histidine tail. This fusion protein is expressed under the control of an

inducible T7 promoter and the BL21 host strain (a protease deficient strain

of E. coli) expresses the T7 polymerase. pET/CRla (encoding the H^CRla

fusion protein) was constructed as follows. The rat CREB cDNA was

completely digested with H ind l l l (which cuts the cDNA in the 3'

untranslated region) and then partially digested with Ncol (which cuts the

cDNA at the methionine start codon). The 'sticky' ends of the DNA created

by these restriction enzymes were blunted w ith AMV reverse

transcriptase) and the desired fragment isolated from a 1 % LMP agarose gel

using glass milk. This was then ligated to pET 15b that had been digested

with Xhol and blunted with AMV reverse transcriptase.

To construct a transcriptionally deficient CREB protein, a P v u l l

fragment (within the H gC R la coding sequences) was excised from

pE T /C R la. The resulting construct, pET/CRlAP, encodes the H^CRIAP

fusion protein, which lacks CREB amino acids 58 to 192 and hence all

118

regions of the CREB protein known to be involved in transcriptional

activation.

The pET constructs encoding the recombinant CREB proteins were

transformed into BL21, a protease deficient strain of E. coli. A single colony

containing the vector was used to inoculate 5 ml LB and grown overnight

at 37°C. This culture was used to inoculate 11 NZYCM (85 mM NaCl;

9.8 mM MgCli; 1% NZ amine; 0.5% yeast extract; 0.1% cas-amino acids) and

was grown at 37°C until the culture was in the log phase of growth. This

was determined by measuring the absorbance of the culture at 600 nm

(A5 0 0) until it reached A qo = 0.4. Expression of the recombinant fusion

protein was induced by the addition of IPTG to a final concentration of

0.5 mM and the cells were grown for a further 2 h. The cells were collected

by centrifugation, resuspended in 20 ml ice cold lysis buffer (Ix PBS; 0.5%

Triton XlOO; 1 mM EDTA; 1 mM PMSF; 5 p g /m l trasylol; 5 p g /m l

leupeptin), lysed by sonication on ice and the cell debris removed by

centrifugation at 4°C. 1 ml of a nickel-agarose column (Ni^+-NTA agarose;

Qiagen) was added to the supernatant and incubated on a rotating wheel at

4°C for 30 min. The histidine tail of the fusion protein forms a strong

interaction with the nickel column and the bound protein was washed

free of contaminating bacterial proteins with ice cold lysis buffer. The

fusion protein was eluted from the column by mixing with 1 ml of

250 mM imidazole in Ix PBS for a few minutes. The purified protein was

dialysed against 1000 volumes of 20 mM Hepes pH 8.0; 125 mM KCl; 1 mM

DTT; 15% glycerol overnight and aliquots of 20 pi were quick frozen on

liquid nitrogen and stored at -80°C. The yield and size of the purified

protein was checked by SDS/PAGE (followed by coomassie staining) and

gel retardation analysis.

119

4.1 Results.

4.1.1 Reduction of E4 transcription by the addition of a competitor

oligonucleotide containing an ATF binding site.

Previous reports have dem onstrated that a com petitor

oligonucleotide containing an ATF binding site can reduce transcription

from the F4 promoter in vitro. This is demonstrated in Figure 4.1.1.

pF4CAT was used as a template for transcription in this assay and a

plasmid containing the major late promoter (pMLP) was included as a

negative control, since the activity of this promoter is not affected by ATF

proteins. Various amounts of the competitor oligonucleotide (ATF-BS)

were added to the transcription reaction immediately before addition of

the template DNA. In this assay, 100 ng of the ATF-BS oligonucleotide

effectively reduces transcription of the F4 promoter, but has no effect on

transcription from the major late promoter.

4.1.2 Analysis of the effect of preincubation of the nuclear extract with the

competitor oligonucleotide on depletion of F4 promoter activity.

Transcription of the F4 promoter at lower amounts of competitor

DNA may be due to the activity of ATF proteins prior to their being

sequestered by the competitor. Preincubation of the nuclear extract with

the competitor oligonucleotide may ensure that upon initiation of the

transcription reaction, all ATF proteins are bound to the competitor DNA

and none to the promoter DNA. This may result in less competitor DNA

being required to reduce transcription and subsequently, the addition of

less recombinant protein being required to restore activity to the promoter.

120

ATF-BS (ng) 0 25 50 100 150 200

E4

MLP

Figure 4.1.1: Transcription of the E4 promoter in vitro is inhibited by the addition of a competitor DNA containing an ATF binding site. Lane 1, transcription reaction containing 0.5 pg pE4CAT and 0.75 pg pMLP as templates for transcription; Lanes 2-6, as Lane 1 but with increasing amounts of a competitor oligonucleotide containing an ATF binding site (ATF-BS) added to the transcription reaction.

No Freine. Freine.

ATF-BS (ng) 0 25 50 100 25 50 100

E4

MLF

Figure 4.1.2: Analysis of the effeet of preineubation of the nuelear extraet with the ATF-BS eompetitor oligonueleotide on depletion of E4 promoter aetivity. Lane 1, transeription reaetion eontaining 0.5 pg pE4CAT and 0.75 pg pMLF as templates for transeription; Lanes 2-4, as Lane 1 but with inereasing amounts of the ATF-BS eompetitor oligonueleotide added at the start of transeription (No Freine.); Lanes 5-6, as Lane 1 but with inereasing amounts of the ATF-BS eompetitor oligonueleotide ineubated with the nuelear extraet at 4°C for 15 min before the start of transeription (Freine.).

Various amounts of competitor were tested for their ability to

reduce transcription of the E4 promoter either with or w ithout

preincubation with the nuclear extract in the absence of promoter DNA.

Figure 4.1.2 shows that preincubation of the competitor DNA with the

nuclear extract has no effect on the amount of competitor DNA that is

required to reduce transcription of the E4 promoter.

4.1.3 The effect of glutathione, thrombin and imidazole upon the in vitro

transcription reaction.

In order to reconstitute the transcriptional activity of the ATF

depleted transcription reaction, recombinant ATF proteins must be added

to the transcription reaction. Two systems for the expression and

subsequent one step purification of recombinant proteins in E. coli

(his-tagged and glutathione-S-transferase (GST) fusion proteins) were

examined for use in this procedure. His-tagged fusion proteins are purified

by binding to a nickel-agarose column and eluting in 250 mM imidazole,

whereas GST fusion proteins are purified by binding to a glutathione-

agarose column and eluting in 5-15 mM glutathione. These fusion

proteins can be cleaved with thrombin to separate the recombinant

protein from its poly-histidine or GST fusion.

I examined the effect of the purification conditions of each of these

systems on the in vitro transcription reaction in order to determine

whether proteins purified in such a way could be added directly to the

transcription reaction, or whether they would require dialysis prior to

their use in the transcription assay.

Figure 4.1.3 A shows that addition of up to 3 mM glutathione does

not adversely affect the transcription reaction, whereas addition of

0.6 units of thrombin (Figure 4.1.3 A) or 25 mM imidazole (Figure 4.1.3 B)

123

Units thrombin mM glutathione

0 0.06 0.2 0.6 0 0.12 0.6 3.0

E4

MLP

Figure 4.1.3 A; Effect of glutathione and thrombin on transcription of the E4 and MLP promoters in vitro. Lanes 1 and 5, transcription reaction containing 0.5 jig pE4CAT and 0.75 pg pMLP as templates for transcription; Lanes 2-4, as Lane 1 but with increasing amounts of thrombin added to the transcription reaction; Lanes 6 -8 , as Lane 5 but with increasing amounts of glutathione added to the transcription reaction.

E4

MLP

mM Imidazole

6 12 25 37 50

Figure 4.1.3 B: Effect of imidazole on transcription of the E4 and MLP promoters in vitro. Lane 1, transcription reaction containing 0.5 pg pE4CAT and 0.75 pg pMLP as templates for transcription; Lanes 2-6, increasing amounts of imidazole added to the transcription reaction.

cause a substantial reduction in transcription. Hence GST-fusion proteins

eluted in 10 mM glutathione can be used directly in this assay, whereas

His-tagged fusion proteins will require dialysis before addition to the

transcription reaction. It is likely that more than 0.6 unit of thrombin

would be added to the transcription reaction when using cleaved fusion

proteins. Hence proteins purified using the His- and GST-tagged fusion

protein systems will be used uncleaved.

4.1.4 Recombinant CREB restores transcription of the E4 promoter by ATE

depleted nuclear extracts.

To restore transcription from the E4 promoter in the ATE depleted

assay, various amounts of a recombinant His-tagged CREB (HôCRla) were

added to the transcription reaction immediately after addition of

competitor and template DNA. Figure 4.1.4 shows that addition of 4.5 |xl

(approximately 500 ng) of HéCRla to the transcription reaction effectively

restores transcription from the E4 promoter.

It is possible that the transcriptional stimulation observed is not due

to the activity of H ôC R la , but due to the activity of ATE proteins

endogenous to the nuclear extract. Rather than activating transcription

directly, H^CRla may be competing with the endogenous ATE proteins for

binding to the competitor DNA, thereby releasing these proteins from the

competition effect and allowing them to stimulate transcription. In an

attempt to establish whether or not this is the case, I examined the effect of

pretreatment with protein kinase A (PKA) upon the activation of

transcription by HôCRla. Phosphorylation of CREB by PKA has been

shown to significantly increase activation of transcription of the

somatostatin promoter by CREB in vitro (Gonzalez et a l , 1991; Yamamoto

et al., 1988). Transcriptional stimulation upon addition of H^CRla and

126

HéCRla+PKA HéCRla PKA

0.5 1.5 4.5 0.5 1.5 4.5 0.5 1.5 4.5 ni

E4

ATF-BS (ng) 0 25 50 100

Figure 4.1.4: Recombinant CREB (HôCRla) restores transcription of the E4 promoter by ATE depleted nuclear extracts, but the CREB dependent stimulation of transcription is not enhanced by protein kinase A (PKA). Lane 1, transcription reaction containing 0.5 pg pE4CAT as the template for transcription; Lanes 2-4, as Lane 1 but with increasing amounts of competitor oligonucleotide (ATF-BS) added to the transcription reaction; Lanes 5-7, as Lane 4 (100 ng ATF-BS) but with increasing amounts of PKA- treated HsCRla added to the transcription reaction; Lanes 8-10, as Lane 4 but with increasing amounts of H éC Rla added to the transcription reaction; Lanes 11-13, as Lane 4 but with increasing amounts of PKA added to the transcription reaction.

PKA to the transcription assay, but not PKA alone, would demonstrate

that the recombinant CREB protein is specifically activating transcription.

Figure 4.1.4 shows that PKA has only a small stimulatory effect on the

activity of H^CRla in this assay. This activation is due to phosphorylation

of H ôCRla and not due to phosphorylation of other components of the

transcription reaction, since addition to the assay of PKA alone has no

effect on transcription.

4.1.5 A mutant recombinant CREB protein, HôCRlAP, cannot restore

transcription of the E4 promoter by ATE depleted nuclear extracts.

To further analyse the effect of adding HôCRla to the transcription

reaction, I examined the activity of a mutant CREB protein (HôCRlAP) that

lacks sequences known to be associated with transcriptional activation.

HeCRlAP has an internal deletion of 134 amino acids that include the

kinase inducible domain (KID; which contains PKA, PKC and casein

kinase II phosphorylation sites; Gonzalez e ta l , 1991; Lee et a l , 1990) and

the alternately spliced a-peptide (Yamamoto et ah, 1990). These domains

have previously been shown to be important for transcriptional activation

by CREB (Gonzalez et ah, 1991; Lee et ah, 1990; Yamamoto et ah, 1990) and

partial deletion of some of these sequences renders CREB transcriptionally

inactive (Gonzalez et ah, 1991; Lee et ah, 1990). Therefore any stimulation

of E4 promoter activity in the ATE depleted transcription reaction after

addition of HôCRIAP will almost certainly be due to the activity of ATE

proteins endogenous to the nuclear extract.

Firstly, I quantified the relative amounts of the two recombinant

CREB proteins that are to be added to the transcription reaction.

Figure 4.1.5 A demonstrates that 5 |xl of H ôC R la gives an equivalent

density band to 5 pi of H^CRIAP when analysed by SDS/PAGE and stained

128

— — 97.4 kDa— 6 6 .2 kDa

— — — 42.7 kDa

J — 31.0 kDa

1 2 3 4 M

H^CRIAP H^CRla

Figure 4.1.5 A: Quantitation of the relative amounts of recombinant CREB proteins. Proteins were run on an 8 % SDS/polyacrylamide gel and visualised by staining with coomassie blue. Lane 1, 5 pi H5 CRIAP; lane 2, 1 pi H^CRIAP; lane 3, 1.0 pi H^CRla; lane 4, 5 pi H^CRla; M, molecular

weight markers.

ComplexH^CRIAP

ComplexH^CRla

Unbound _D N A " —

1 2 3 4 5 6

H^CRIAP H^CRla

F igu re 4 .1 .5 B: Quantitation of the DNA binding activity of recombinant CREB proteins. Various amounts of H ^CRla and H^CRIAP

were incubated with a ^^P-labelled DNA probe containing an ATP binding site. Unbound DNA and protein/DNA complexes are indicated beside the figure. Lane 1, 0.25 pi H^CRla; lane 2, 0.5 pi H^CRla; lane 3, 1.0 pi H^CRla; lane 4, 0.25 pi H^CRIAP; lane 5, 0.5 pi H^CRIAP; lane 6,1.0 pi H^CRIAP.

HéCRla HôCRlAP

0.5 1.5 4.5 13.5 0.5 1.5 4.5 13.5

E4

MLP

ATF-BS (ng) 0 25 50 100

Figure 4.1.5: A mutant recombinant CREB protein, HéCRlAP, cannot restore transcription of the E4 promoter by ATP depleted nuclear extracts. Lane 1, transcription reaction containing 0.5 pg pE4CAT and 0.75pg pMLP as the templates for transcription; Lanes 2-4, as Lane 1 but with increasing amounts of competitor oligonucleotide (ATF-BS) added to the transcription reaction; Lanes 5-8, as Lane 4 (100 ng ATF-BS) but with increasing amounts of H eC R la added to the transcription reaction; Lanes 9-12, as Lane 4 but with increasing amounts of HeCRlAP added to the transcription reaction.

with Coomassie blue. This estimate of the relative concentrations of these

two proteins is confirmed by analysis in a gel mobility shift assay

performed as described in Chapter 3 and explained briefly in the legend to

Figure 4.1.5 B. Figure 4.1.5 B demonstrates that 1 pi of H^CRla binds an

equivalent amount of a ^^P-labelled probe containing the E4 -50 ATF

binding site as 1 pi of HôCRlAP.

Figure 4.1.5 C shows the effect of the addition of these two proteins

to the ATF depleted transcription reaction. As demonstrated in the

previous section, addition of 4.5 pi of H ^CRla results in the efficient

stimulation of E4 transcription. However, addition of an excess of

FîéCRlAP (13.5 pi) has no effect on E4 transcription. This demonstrates that

the stimulatory effect of adding HeCRla to the transcription reaction is not

due to HôCRla competing with endogenous ATF proteins for binding to

the competitor DNA, but is due to HôCRla binding to and activating the

E4 promoter directly.

132

4.2 Discussion.

In vitro transcription of the E4 promoter can be inhibited by the

addition of a specific competitor oligonucleotide containing an ATF

binding site to the transcription reaction (Lee and Green, 1987) and

transcriptional activity can be restored by the subsequent addition of

purified ATF proteins (Hai et a l , 1988a). I have demonstrated that

transcriptional activity can be effectively restored to a transcription assay

depleted of ATF in this way, by the addition of purified recombinant

CREB. In addition, I have demonstrated that this stimulation is specific to

the activity of the recombinant CREB rather than ATF proteins

endogenous to the nuclear extract, since a mutant CREB protein lacking all

known activation domains does not activate transcription in this assay.

Hence this strategy for depleting a transcription reaction of ATF proteins

will be suitable for the development of an in v i tro assay that will

hopefully demonstrate ElA activated transcription of the E4 promoter

mediated by recombinant ATF-2.

Further development of this assay requires the addition of

recombinant ATF-2 to the ATF depleted transcription reaction, but

unfortunately the GST- and His-tagged ATF-2 proteins that I initially

constructed were not efficiently expressed in E. coli. A recent report by

Furlong et al. (1992) determined several parameters that affect the

expression efficiency of proteins cloned into a pET vector. They observed

that the nature of the second amino acid of a protein affects the efficiency

of its expression, and concluded that the rules for protein expression were

complicated and require direct experimentation. I examined the possibility

that amino acids at the N-terminus of ATF-2 could affect its expression by

constructing a His-tagged ATF-2 protein lacking the N-terminal 15 amino

acids and testing its expression in E. coli. This protein was in fact efficiently

133

expressed and purified (data not shown), but this was unfortunately only

achieved immediately prior to the composition of this work and was not

tested in the assays described above.

Further development of this assay such as to demonstrate E lA

dependent activation of the E4 promoter could proceed by either of two

strategies. The simplest approach would be the comparison of activation

by recombinant ATF-2 and ElA together, to that of ATF-2 alone. This

approach demonstrates efficient ElA dependent activation of other

adenovirus promoters (Green e ta l , 1988; Loewenstein and Green, 1989;

Spangler et al, 1987) and hence in the absence of other ATF proteins that

may normally stimulate basal levels of transcription, this approach may be

sufficient to demonstrate activation of the E4 promoter. Alternatively, I

could compare the effect of adding recombinant ATF-2 to an ATF depleted

transcription reaction using nuclear extracts from adenovirus infected

HeLa cells to one using nuclear extracts from uninfected cells. The

possibility that other proteins, in addition to ATF proteins, may be

required for the ATF dependent activation of the E4 promoter by ElA has

been suggested by the contradictory activities of ATF binding sites in

differing promoter contexts (Jones and Lee, 1991; Lee et a l , 1989; Rooney

et a l , 1990 and this thesis). This approach would allow for the possibility

that such unknown activities may not be present in nuclear extracts from

uninfected cells.

Once an in vitro assay has been established that demonstrates

efficient trans-activation of the E4 promoter by ElA, the role of ATF-2

could be further characterised by mutational analysis of the recombinant

protein. In addition, the minimal requirements for this process can be

determined by fractionation of the nuclear extract into increasingly pure

components. This approach would be useful in identifying other proteins

that may be involved in trans-activation by ElA and would also improve

134

the usefulness of other methods of analysis in determining the roles of

proteins involved in this process.

Raychaudhuri et al. (1987) demonstrated that band shifts of the

whole E4 promoter using crude extracts of adenovirus infected HeLa cells

give rise to an extremely complex, unresolved band shift pattern. In

contrast, by using a smaller DNA probe and by performing a simple

fractionation of the extracts, a more defined and discrete band shift pattern

was obtained. Determining the minimal requirements for frflns-activation

of the E4 promoter by ElA in vitro would allow us to perform band shift

analysis upon the whole E4 promoter using the purified components

required for transcriptional activity. In parallel with the transcription

assay, this approach could be used to demonstrate functional interactions

between ElA and other factors (such as ATF-2 and/or TBP) present in the

transcription complex. Further fractionation of the components of this

assay could identify all factors that are required for such an interaction,

thereby further characterising the functional requirements for trans-

activation.

I have already discussed several possibilities that could explain the

conclusion that the DNA binding specificity of ATF-2 in vitro does not

correlate with activation of the E4 promoter in vivo when ATF-2 was

expected to mediate trans-activation by ElA. This conclusion could also be

examined at the functional level to determine whether mutations in the

ATF/E4E binding sites could affect the binding of ATF-2, such as to

interfere with its role in trans-activation by ElA. Such effects may not

necessarily be manifested as a change in the affinity of ATF-2 for these

mutated binding sites and hence would not be observed in the binding

assays employed in Chapter 3. More importantly, this possibility cannot be

addressed in vivo because of the large number of other proteins that

would be available to interact with this site.

135

Chapter 5: Discussion.

5.1 The relative roles of E4F and ATF-2 in trans-activaüon of the

E4 promoter by ElA.

Although the results presented in this study demonstrate that E4F

binding is not necessary for trans-activation of the E4 promoter by ElA,

they do not discount a role for E4F in this process. To date there are no

mutations known to eliminate ATF binding to the E4 promoter without

affecting E4F binding (Rooney et al, 1990) and it is therefore not possible to

determine the role of E4F in the absence of ATF binding. However, the

properties previously described for E4F (Raychaudhuri et al., 1989;

Raychaudhuri et al, 1987; Rooney et al, 1990) indicate that it is critical for

frans-activation by ElA under some circumstances. In addition, a recent

report suggests that trans-activation of the E4 promoter by ElA is primarily

mediated by E4F in HeLa cells (Bondesson et al, 1992).

Bondesson et al. (1992) demonstrated that the non-conserved

C-terminal exon of ElA contains two interchangeable elements, auxiliary

regions 1 and 2 (ARl and AR2), either one of which is required for

efficient CR3 and E4F dependent trans-activation of the E4 promoter in

HeLa cells. Bondesson et al. examined the requirements of these regions of

ElA for activation of the heterologous promoters described by Rooney

et al. (1990). ElA trans-activation of a heterologous promoter mediated by

E4F (RE4PM2) was found to be ARl and AR2 dependent. However,

neither ATF-2 mediated ElA trans-activation nor ElA trans-activation of a

heterologous promoter mediated by ATF (RE4PM4) require the integrity of

these two elements. Hence trans-activation of the E4 promoter by ElA

would appear to be primarily mediated by E4F (and not ATF-2) in HeLa

cells (Bondesson et al, 1992).

136

However, it is possible that the conclusions of Bondesson et al. are

incorrect. By assuming that trans-activation of the heterologous RE4

promoters by ElA (Rooney et al, 1990) represents trans-activation of the

E4 promoter, Bondesson et al. do not take into account the results of this

thesis, which demonstrate that these promoters are not functionally

comparable. Because of the different conclusions drawn from these results

and those of Rooney et al. (1990), it is apparent that the RE4 promoters do

not reflect the activity of ATF/E4F binding sites in the context of the

E4 promoter. Hence the dependence on the ARl and AR2 regions for

fraws-activation of either of these promoters by ElA may also not be

comparable.

However, the function of these regions and their effect on the

activity of ElA is extremely interesting. Ideally, any future examination of

the activity of these regions should be performed using mutated

E4 promoters such as those described in this thesis. Only by comparing the

activity of a mutated E4 promoter that only binds ATF (WE4PM4) with

that of a mutated E4 promoter that only binds E4F, can the nature of ARl

and AR2 dependent activation be determined in the context of the

E4 promoter. However, a mutation that prevents ATF binding without

affecting binding of F4F has not yet been described.

If the results of Bondesson et al. (1992) are correct, why does ATF

not contribute more significantly to fra ws-activation by FI A when it is

evident from my results that the F4 promoter is efficiently activated in the

absence of F4F binding? It is possible that the ATF protein that mediates

fraws-activation by FIA may have a lower affinity for ATF/F4F binding

sites than F4F. Rooney et al. (1990) demonstrated that F4F forms a stable

complex with the F4 promoter, whereas purified ATF forms much less

stable complexes that are rapidly dissociated. Although the precise nature

of the ATF proteins present in the ATF fraction used by Rooney et al. is

137

not known, their study does serve to demonstrate the high affinity that

E4F has for ATF/E4F binding sites in the E4 promoter. The dissociation

rates of E4F and ATF-2 binding to the ATF/E4F binding sites have not

been examined together and the individual analyses reported for ATF-2

(Jones and Lee, 1991 and described in this thesis) and for E4F (Rooney et al,

1990) used different probe DNA, competitor DNA and reaction conditions

and hence, unfortunately, are not comparable.

5.1.1 Problems associated with interpreting the results of this study.

Although the transient expression assay is an extremely useful tool

for defining the activity of a particular promoter, it does not necessarily

truly reflect the activity of the promoter in its natural context, whether a

cellular or viral promoter. This is very obvious when examining the effect

of ElA on the expression of cellular genes. Many plasmid-borne cellular

promoters are induced by ElA in a transient expression assay, whereas

their endogenous cellular counterparts are not. However, such differences

in activity are less obvious when examining the activity of viral

promoters. Mutational analysis of the adenovirus E2A promoter has

demonstrated different requirements for activation in the context of the

viral chromosome to activation of the plasmid-borne promoter in

transfection assays. In transient transfection assays, mutations in any one

of four promoter elements in the E2A promoter (the TAG A motif, the

ATF binding site and the two E2F binding sites) have no effect on ElA

induced transcription. In a stable transfection assay, a mutant promoter

with only the TAG A motif is sufficient for both basal and ElA induced

transcription. However, in its natural context in the viral chromosome,

mutation of any one of these elements substantially reduces both basal and

ElA induced transcription.

138

5.2 Factors affecting the requirements for trans-activation by ElA.

5.2.1 Regulation of other factors by ElA.

Other studies have demonstrated that ElA regulates the activity of

some cellular transcription factors by mediating their phosphorylation by a

cellular kinase (Hoeffler e ta l , 1988; Raychaudhuri et a l , 1989). Hence the

activity of such ElA dependent kinases in a cell may also affect the

transcription factor requirements in mediating trans-activation in a

particular cell type. For instance, if the ElA dependent kinase that is

responsible for the phosphorylation (and thence the increased DNA

binding activity) of E4F is not expressed or active in a particular cell type,

then trans-activation of the E4 promoter by ElA might be preferentially

mediated by an ATF protein in these cells.

Previous reports have demonstrated that ElA is associated with a

60 kDa polypeptide (Harlow et al., 1986) that has since been identified as

hum an cyclin A (Giordano et a l , 1989; Pines and Hunter, 1990).

Subsequent reports have demonstrated that the ElA complex containing

cyclin A also contains a cellular kinase that is related to the cell cycle

control protein (Giordano e ta l , 1991; Kleinberger and Shenk, 1991).

This kinase has since been cloned and is known as (Tsai et a l , 1991)

Interestingly, p3 4 ^ ‘ ‘ 2 kinase has been demonstrated to phosphorylate the

C-terminal domain of RNA polymerase II (Cisek and Corden, 1989), which

is postulated to be involved in the activator dependant stimulation of

transcription (see above).

It is intriguing to speculate as to the role of these proteins in both

normal cellular transcription and in transcriptional and cell cycle control

during an adenovirus infection. In an uninfected cell, it is possible that the

phosphorylation and dephosphorylation of the RNA polymerase II CTD is

a switch that is responsible for the arrest of transcription during M-phase.

139

In the adenovirus infected cell, the association of ElA with a cellular

kinase that can phosphorylate the CTD may allow the adenovirus to

circumvent the normal cellular controls on transcription during DNA

synthesis. In addition, this kinase may be responsible for the

phosphorylation of cellular factors involved in the transcription of

adenovirus genes (such as E4F) either as a result of its normal cellular

activity, or due to a change in context due to its association with ElA.

5.2.2 Dimérisation of ElA inducible transcription factors.

Another factor that is likely to affect the requirements for trans-

activation by ElA is the complement of proteins capable of forming

heterodimers with an ElA inducible transcription factor such as ATF-2

within a particular cell type. The bZlP family of transcription factors also

includes the Fos/Jun family (for review, see; Curran and Franza, 1988) that

bind to the TRE (TPA response element; TGACTCA; Lee et al., 1987b;

Angel et a l , 1987) which differs from the ATF consensus motif by only the

absence of one base pair. Initially, members of the Fos/Jun and ATF

families were regarded as distinct sets of transcription factors that share the

basic dom ain/leucine zipper region, but which have different DNA

binding specificities. However, more recent studies have indicated that

members of each family can form intrafamilial heterodimers through

interaction of their respective leucine zipper regions (Benbrook and Jones,

1990; Hai and Curran, 1991) which can alter the DNA binding specificity of

the proteins (Hai and Curran, 1991). Hence the availability of ATF-2 to

interact with ATF binding sites in the E4 promoter may not only rest upon

the relative abundance of ATF-2 within a particular cell type, but also

upon the nature of the dimerised form of the protein.

140

5.3 The mechanisms of E4F and ATF-2 mediated frans-activation of the

E4 promoter by ElA.

Independent studies have examined the mechanism of ATF-2

mediated fr^ns-activation by ElA and have demonstrated a requirement

for the potential zinc finger motif present at the N-terminus of ATF-2.

(Flint and Jones, 1991; Zu et al., 1991). A recent study has demonstrated

that ATF-a mediates frflws-activation by E lA through a similar

mechanism to ATF-2 (Chatton etal., 1993). ATF-a shares extensive

homology with ATF-2 (Caire et al., 1990), especially in the N-terminal zinc

finger region and like ATF-2, the integrity of this region is essential for

ATF-a to mediate trans-activation by ElA (Chatton et al., 1993). In

addition, ATF-a was found to associate with ElA and mutational analysis

demonstrated that two separate domains of ATF-a, in the N-terminal and

the C-terminal regions, can independently interact with ElA. The zinc

finger of ATF-a was shown to be responsible for the N-terminal

interaction with ElA, but the nature of the domain at the C-terminus that

interacts with ElA has not been elucidated (Chatton et a l, 1993).

Previous studies have suggested that the mechanisms for activation

mediated by either E4F or ATF-2 are different (Liu and Green, 1990;

R aychaudhuri et a l , 1989) and indeed the results of Bondesson et al.

would at first appear to confirm this suggestion since these two proteins

exhibit different requirements for the integrity of the ARl and AR2

regions of ElA (Bondesson e ta l , 1992). However, if the conclusions of

Bondesson et al. are correct, 1 propose that these two mechanisms are not

dissimilar. Previous studies have demonstrated that ElA functions at the

promoter complex in a similar manner to many cellular activators and

mutational analysis of ElA has demonstrated a requirement for the

activation and promoter targeting domains of ElA in CHO cells (Lillie and

141

Green, 1989; Martin e ta l , 1990). Bondesson e t al . confirmed the

requirement for the promoter targeting domain of E lA for trans-

activation of the E4 promoter in HeLa cells (Bondesson et a l , 1992).

Therefore, E4F activity correlates with the requirement for the promoter

targeting domain of E lA and would be predicted to mediate trans-

activation by ElA via a similar mechanism to that proposed for ATF-2

(Liu and Green, 1990). A direct interaction between E4F and ElA has not

previously been postulated since it has been assumed that stimulation of

the DNA binding activity of E4F by ElA would account for the ElA

induced increase in E4 promoter activity (Raychaudhuri et a l, 1989).

This revised model for the mechanism of E4F activity and the

conclusions of Bondesson et al. are mutually inclusive. E4F is only

predicted to function by recruiting ElA to the E4 promoter if trans-

activation of the E4 promoter by ElA is indeed primarily mediated by E4F.

If E4F does not function by such a mechanism, then conversely it is

unlikely that the observed AR1/AR2 dependent activation of the

E4 promoter is primarily mediated by E4F.

The suggestion that E4F functions to recruit ElA to the promoter

complex also presents a possible function for ARl and AR2. These two

domains are adjacent to the promoter targeting domain of ElA (Lillie and

Green, 1989) and could represent a functional extension of the activity of

this domain. Whereas the previously defined boundaries of the promoter

targeting domain may be a universal requirement for the interaction of

ElA with several transcription factors (such as ATF-2 and E4F), ARl and

AR2 may cooperate with this domain in a manner specific to an

interaction with E4F, but may not be required for the interaction with

ATF-2.

142

5.4 Other examples of multiple factors binding to individual promoter

elements.

The ATF/E4F binding sites in the E4 promoter are not the only

known example of a composite promoter elem ent consisting of

overlapping binding sites for distinct factors. Analysis of factors that bind

to the herpes simplex virus thymidine kinase promoter identified two

CCAAT binding activities, CTF and C/EBP (Jones et a l , 1985; Graves et al.,

1986). Although these proteins both interact with the CCAAT element,

they exhibit slightly different DNA binding specificities. A C-to-G

mutation at the first residue of the CCAAT sequence has opposing effects

on the binding of these two factors, increasing the affinity of C/EBP

(Graves et al., 1986) whilst reducing the affinity of CTF (Johnson and

McKnight, 1989). Subsequent cloning of C/EBP and CTF has demonstrated

that they are encoded by different genes, and somewhat surprisingly, that

they utilise different DNA binding motifs (Landschulz et a l , 1988; Santoro

et al., 1988). Together with the more recent cloning of a family of bZIP

proteins that are closely related to C/EBP, this situation is highly

analogous to that of ATF and E4F binding to the E4 promoter. Whether

E4F is a bZIP protein or utilises a different DNA binding domain will be

revealed when E4F is cloned.

This example demonstrates that the functional redundancy

characterised by multiple transcription factors binding to the E4 promoter

is not specific to viral promoters but is also common to cellular promoters.

This is consistent with the general structure of cellular and viral

promoters demonstrating essential similarities, together with the original

hypothesis that the regulation of viral transcription would reflect cellular

modes of regulation.

143

5.5 The role of ATF proteins in other viral frans-activation pathways.

The results of this and previous studies investigating the roles of

cellular transcription factors in viral trans-activation suggest that the ATF

family is commonly targeted by viral trans-activators in the activation of

viral and cellular genes. In addition to the adenovirus early promoters,

ATF binding sites are also present in the long terminal repeat (LTR) of

human T-cell leukaemia virus type 1 (HTLV-1) and the promoters of the

Krox-20 and Krox-24 genes (Giam and Xu, 1989; Alexandre et a l , 1991).

Recent studies have demonstrated that binding of CREB to the LTR is

enhanced by the HTLV-1 trans-activator, p40 ® , via a direct interaction

between the two proteins (Zhao and Giam, 1992). This is similar to the

mechanism by which CREB and ATF-2 are proposed to mediate activation

of the hepatitis B virus (HBV) enhancer by the HBV X gene product, pX

(Maguire et a l, 1991; see Chapter 3.2). Therefore in at least three distinct

viral systems, ATF proteins have been proposed to mediate tran s

activation by a direct interaction with the viral trans-activator. In addition

to p40 ^ , the bovine leukaemia virus (BLV) trans-activator p38‘® and the

cytomegalovirus (CMV) tegument protein pp71 also target promoters

containing ATF binding sites (Willems et a l , 1992; Hunninghake et a l ,

1989). Future studies will hopefully determine whether ATF proteins are

involved in mediating activation by these viral trans-activators and if so,

the mechanisms involved.

The use of similar pathways by unrelated viruses in the activation

of viral genes is probably a reflection of the mechanisms by which cellular

signal transduction pathways operate. Cellular equivalents of the viral

trans-activators are likely to be adaptor proteins (see Chapter 1.4.3); the

suggestion that these proteins function by bridging and/or stabilising

interactions between transcription factors and the general factors is similar

144

to the proposed mechanism of activation by ElA and other viral trans-

activators. However, whether these factors can activate transcription when

tethered to a DNA binding domain (as has been shown to be the case with

ElA; Lillie and Green, 1989) has yet to be deduced. Once the cellular

equivalent of ElA has been identified, the role of this factor can be studied

in relation to which cellular genes it regulates and what regulatory

pathway it is part of.

145

5. 6 Conclusions.

Together with studies of other viral systems, the study of trans-

activation of adenovirus early genes by ElA has identified many cellular

transcription factors and provided insights into how the cell modulates

transcription. However, the mechanisms by which E lA achieves the

regulation of early gene expression have proved to be more numerous

and varied than was initially expected. The characterisation of the early

viral genes led to the identification of the ATF family and it was

considered likely that ElA utilised a common mechanism, via ATF

binding sites, to activate these promoters. However, we now know that

trans-activation of the E2A promoter is mediated by the E2F transcription

factor, by a mechanism distinct to ATF mediated activation of the

E4 promoter (Hardy and Shenk, 1989; Hardy et al., 1989; Bagchi et ah, 1990).

In addition, the characterisation of E4F as an ElA inducible transcription

factor distinct from the ATF family has further demonstrated the diversity

of ElAs activity.

By targeting multiple transcription factors involved in the

regulation of adenovirus genes, ElA potentially increases the host range of

the virus such that it can efficiently replicate in a wide variety of cell types.

It is likely that the requirements for this process are a complex

combination of several factors, relating not only to the availability of an

individual transcription factor in a particular cell type, but also on the

availability of other proteins involved in trans-activation mediated by that

transcription factor. The different processes by which ElA activates

transcription can be directly related to normal cellular modes of regulation

and hence other factors required for trans-activation by ElA are likely to

also be involved in cellular signal transduction pathways.

146

REFERENCES

Abmayr, S.M., L.D. Feldman and R.G. Roeder (1985) In vitro stimulation of specific RNA polymerase Il-mediated transcription by the pseudorabies virus immediate early protein. Cell 43, 821-9.

Adams, R.L.P., J.T. Knowler and D.P. Leader (1986) The Biochemistry of the Nucleic Acids . Chapman and Hall, New York.

Alexandre, C., P. Charnay and B. Verrier (1991) Trans-activation of Krox-20 and Krox-24 promoters by the HTLV-1 Tax protein through common regulatory elements. Oncogene 6 , 1851-7.

Allison, L.A. and C.J. Ingles (1989) Mutations in RNA polymerase II enhance or suppress mutations in GAL4. Proc. Natl Acad. Sci. USA 86,2794-8.

Allison, L.A., M. Moyle, M. Shales and C.J. Ingles (1985) Extensive homology among the largest subunits of eukaryotic and prokaryotic RNA polymerases. Cell 42, 599-610.

Angel, P., M. Imagawa, R. Chiu, B. Stein, R.J. Imbra, H.J. Rahmsdorf,C. Jonat, P. Herrlich and M. Karin (1987) Phorbol ester-inducible genes contain a common cis element recognised by a TPA- modulated trans-acting factor. Cell 49, 729-39.

Armaleo, D. and S.R. Gross (1985) Purification of the three nuclear RNA polymerases from Neurospora crassa. /. Biol. Chem. 260, 16169-73.

Bagchi, S., P. Raychaudhuri and J.R. Nevins (1990) Adenovirus ElA proteins can dissociate heteromeric complexes involving the E2F transcription factor: A novel mechanism for ElA trans-activation. Cell 62, 659-69.

Bakker, O. and M.G. Parker (1991) CAAT/enhancer binding protein is able to bind to ATF/CRE elements. Nucl. Acids Res. 19, 1213-7.

Beltz, G.A. and S.J. Flint (1979) Inhibition of HeLa cell protein synthesis during adenovirus infection. Restriction of cellular messenger RNA sequences to the nucleus. J. Mol. Biol. 131, 353-73.

Benbrook, D.M. and N.C. Jones (1990) Heterodimer formation between CREB and JUN proteins. Oncogene 5, 295-302.

Berger, S.L., W.D. Cress, A. Cress, S.J. Triezenberg and L. Guarente (1990) Selective inhibition of activated but not basal transcription by the acidic activation domain of VP16: Evidence for transcriptional adaptors. Cell 61, 1199-208.

147

Berk, A.J. (1986) Adenovirus promoters and ElA t rans-a c t iv a t io n . Ann. Rev. Genet. 20, 45-79.

Berk, A.J., F. Lee, T. Harrison, J. Williams and P.A. Sharp (1979) Pre-early adenovirus 5 gene product regulates synthesis of early viral messenger RNAs. Cell 17, 935-44.

Berk, A.J. and P.A. Sharp (1978) Structure of the adenovirus 2 early mRNAs. Cell 14, 695-711.

Birnboim, H.C. and J. Doly (1979) A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucl. Acids Res. 7, 1513.

Bondesson, M., C. Svensson, S. Linder and G. Akusjarvi (1992) The carboxy-terminal exon of the adenovirus ElA protein is required for E4F-dependent transcription activation. EMBO J. 11, 3347-54.

Borrelli, E., R. Hen and P. Chambon (1984) Adenovirus-2 ElA products repress enhancer-induced stimulation of transcription. Nature 312, 608-12.

Borrelli, E., R. Hen, C. Wasylyk, B. Wasylyk and P. Chambon (1986) The immunoglobulin heavy chain enhancer is stimulated by the adenovirus type 2 E lA products in m ouse fibroblasts. Proc. Natl Acad. Sci. USA 83, 2846-9.

Breathnach, R. and P. Chambon (1981) Organization and expression of eukaryotic split genes coding for proteins. Ann. Rev. Biochem. 50, 349-83.

Brent, R. and M. Ptashne (1985) A eukaryotic transcriptional activator bearing the DNA specificity of a prokaryotic repressor. Cell 43, 729-36.

Buratoski, S., S. Hahn, L. Guarente and P.A. Sharp (1989) Five intermediate complexes in transcription initiation by RNA polymerase II. Cell 56, 549-61.

Busch, S.J. and P. Sassone-Corsi (1990) Dimers, leucine zippers and DNA- binding domains. T.I.G. 6 , 36-40.

Carcamo, J., L. Buckbinder and D. Reinberg (1991) The initiator directs the assembly of a transcription factor IID-dependent transcription complex. Proc. Natl Acad. Sci. USA 8 8 , 8052-6.

Carlock, L.R. and N.C. Jones (1981) Transformation-defective mutant of adenovirus type 5 containing single altered ElA mRNA species. /. Virol. 40, 657-64.

148

Carthew, R.W., L.A. Chodosh and P.A. Sharp (1985) An RNA polymerase II transcription factor binds to an upstream element in the adenovirus major late promoter. Cell 43, 439-48.

Chatter]ee, P.K., M. Bruner, S.J. Flint and M.L. Harter (1988) DNA-binding properties of an adenovirus 289R ElA protein. EMBO J. 7, 835-41.

Chatton, B., J.L. Bocco, M. Caire, C. Hauss, B. Reimund, J. Goetz andC. Kedinger (1993) Transcriptional activation by the adenovirus larger ElA product is mediated by members of the cellular transcription factor ATF family which can directly associate with ElA. M ol Cell Biol 13, 561-70.

Chow, L.T., T.R. Broker and J.B. Lewis (1979) Complex splicing patterns of RNAs from the early regions of adenovirus-2. J. M ol Biol 134, 265-303.

Cisek, L.J. and J.L. Corden (1989) Phosphorylation of RNA polymerase by the murine homologue of the cell-cycle control protein cdc2. Nature 339, 679-84.

Comb, M., N.C. Birnberg, A. Seasholtz, E. Herbert and H.M. Goodman (1986) A cyclic AMP- and phorbol ester-inducible DNA element. Nature 323, 353-6.

Corden, J., B. Wasylyk, A. Buchwalder, P. Sassone-Corsi, C. Kedinger and P. Chambon (1980) Promoter sequences of eukaryotic protein coding genes. Science 209, 1406-13.

Corden, J.L. (1990) Tails of RNA polymerase II. T.LB.S. 15, 383-7.Corden, J.L., D.L. Cadena, J.M. Ahearn and M.E. Dahmus (1985) A unique

structure at the carboxyl terminus of the largest subunit of eukaryotic RNA polymerase II. Proc. Natl Acad. Sci. USA 82, 7934-8.

Cortes, P., L. Buckbinder, M.A. Leza, N. Rak, P. Hearing, A. Merino andD. Reinberg (1988) EivF, a factor required for transcription of the adenovirus EIV promoter, binds to an element involved in Ela- dependent activation and cAMP induction. Genes Dev. 2, 975-90.

Courey, A. and R. Tjian (1988) Analysis of Spl in vivo reveals multiple transcriptional domains, including a novel glutam ine-rich activation motif. Cell 55, 887-98.

Cowell, I.G., A. Skinner and H.C. Hurst (1992) Transciptional repression by a novel member of the bZIP family of transcription factors. M ol Cell. Biol. 1 2 , 3070-7.

Cress, W.D. and S.J. Triezenberg (1991) Critical structural elements of the VP16 transcriptional activation domain. Science 251, 87-251.

149

Culp, J.S., L.C. Webster, DJ. Friedman, C.L. Smith, W.J. Huang, F.Y. Wu, M. Rosenberg and R.P. Ricciardi (1988) The 289-amino acid El A protein of adenovirus binds zinc in a region that is important for frflMS-activation. Proc, Natl Acad. Soi. USA 85, 6540-54.

Curran, T. and B.R. Franza (1988) Fos and jun: The AP-1 connection. Cell 55,395-7.

Cutt, J.R., T. Shenk and P. Hearing (1987) Analysis of adenovirus early region-encoded polypeptides synthesized in productively infected cells. /. Virol. 61, 543-52.

Darnell, J., H. Lodish and D. Baltimore (1986) Molecular cell biology. Scientific American Books, New York.

Datta, S., C.-J. Soong, D.M. Wang and M.L. Harter (1991) A purified adenovirus 289-amino-acid ElA protein activates RNA polymerase III transcription in vitro and alters transcription factor TFIIIC. J. Virol. 65, 5297-304.

Davison, B.L., J.M. Egly, V.R. Mulvihill and P. Chambon (1983) Formation of stable preinitiation complexes between eukaryotic class B transcription factors and promoter sequences. Nature 301, 680-6.

Dignam, J.D., R.M. Lebovitz and R.G. Roeder (1983) Accurate transcription initiation by RNA polymerase II in a soluble extract from isolated mammlian nuclei. Nucl. Acids Res. 11, 1475-89.

Dynan, W.S. and R. Tjian (1983) The promoter-specific transcription factor Spl binds to upstream sequences in the SV40 early promoter. Cell 35, 79-87.

Dynan, W.S. and R. Tjian (1985) Control of eukaryotic messenger RNA synthesis by sequence-specific DNA-binding proteins. Nature 316, 774-7.

Dynlacht, B.D., T. Hoey and R. Tjian (1991) Isolation of coactivators associated with the TATA-binding protein that m ediate transcriptional activation. Cell 6 6 , 563-76.

Dyson, N. and E. Harlow (1992) Adenovirus ElA targets key regulators of cell proliferation. Cane. Surv. 1 2 , 161-95.

Engelke, D.R., B.S. Shastry and R.G. Roeder (1983) Multiple forms of DNA- dependent RNA polymerases in Xenopus laevis. Rapid purification and structural and immunological properties. J. Biol. Chem. 258, 1921-31.

150

Esche, H., M.B. Mathews and J.B. Lewis (1980) Proteins and messenger RNAs of the transforming region of wild-type and mutant adenoviruses. J. M ol Biol 142, 399-417.

Feaver, W.J., O. Gileadi, Y. Li and R.D. Kornberg (1991) CTD kinase associated with yeast RNA polymerase II initiation factor b. Cell 67, 1223-30.

Ferguson, B., B. KrippI, O. Andrisani, N. Jones and H. Westphal (1985) Fla 13 S and 12 S mRNA products made in Escherichia coli both function as nucleus-localized transcription activators but do not directly bind DNA. M ol Cell Biol 5, 2653-61.

Fink, J.S., M. Verhave, S. Kasper, T. Tsukada, G. Mandel and R.H. Goodman (1988) The CGTCA sequence motif is essential for biological activity of the vasoactive intestinal peptide gene cAMP-regulated enhancer. Proc. Natl Acad. Sci. USA 85, 6662-6.

Fire, A., S. Samuels and P.A. Sharp (1984) Interactions between RNA polym erase II, factors, and template leading to accurate transcription. J. Biol. Chem. 259, 2509-16.

Flint, J. and T. Shenk (1989) Adenovirus FIA Protein Paradigm Viral Transactivator. Ann. Rev. Genet. 23, 141-61.

Flint, K.J. and N.C. Jones (1991) Differential regulation of three members of the ATF/CRFB family of DNA-binding proteins. Oncogene 6 ,

Flint, S.J. (1986) Regulation of adenovirus mRNA formation. Adv. Virus Res. 31, 169-228.

Flores, O., H. Lu and D. Reinberg (1992) Factors involved in specific transcription by mammalian RNA polymerase II: Identification and characterization of factor IIH. J. Biol. Chem. 267, 2786-2793.

Flores, O., F. Maldonado, Z. Burton, J. Greenblatt and D. Reinberg (1988) Factors involved in specific transcription by mammalian RNA polymerase II. RNA polymerase Il-associating protein 30 is an essential component of transcription factor IIF. J. Biol. Chem. 263, 10812-6.

Foulkes, N.S., F. Borrelli and P. Sassone-Corsi (1991) CRFM gene: Use of alternative DNA-binding domains generates multiple antagonists of cAMP-induced transcription. Cell 64, 739-49.

Fujisawa, J.-L, M. Toita and M. Yoshida (1989) A unique enhancer element for the trans-àctiwàioT (p40*^^) of human T-cell leukemia virus type I that is distinct from cyclic AMP- and 12-O-Tetradecanoyl- phorbol-13-acetate-responsive elements. J. Virol. 63, 3234-9.

151

Caire, M., B. Chatton and C. Kedinger (1990) Isolation and characterization of two novel, closely related ATF cDNA clones from HeLa cells. Nucl. Acids Res. 18, 3467-73.

Gaynor, R.B., L.T. Feldman and A.J. Berk (1985) Transcription of class III genes activated by viral immediate early poteins. Science 230, 447-50.

Gaynor, R.B., D. Hillman and A.J. Berk (1984) Adenovirus early region lA protein activates transcription of a non viral gene introduced into mammalian cells by infection or transfection. Proc. Natl Acad. Sci. USA 81,1193-7.

Gentz, R., F.3. Rauscher, C. Abate and T. Curran (1989) Parallel association of Fos and Jun leucine zippers juxtaposes DNA binding domains. Science 243, 1695-9.

Giam, C.Z. and Y.L. Xu (1989) HTLV-I tax gene product activates transcription via pre-existing cellular factors and cAMP responsive element. /. Biol. Chem. 264, 15236-41.

Gilardi, P. and M. Perricaudet (1984) The E4 transcriptional unit of Ad2: far upstream sequences are required for its transactivation by ElA. Nucl. Acids Res. 1 2 , 7877-88.

Gill, G. and R. Tjian (1992) Eukaryotic coactivators associated with the TATA box binding protein. Curr. Op. Genet. Dev. 2, 236-42.

Giordano, A., J.H. Lee, J.A. Scheppler, C. Herrmann, E. Harlow, U. Deuschle, D. Beach and B.R. Franza (1991) Cell cycle regulation of histone HI kinase activity associated with the adenoviral protein ElA. Science 253, 1271-5.

Giordano, A., P. Whyte, E. Harlow, B.R. Franza, D. Beach and G. Draetta(1989) A 60 kd cdc2-associated polypeptide complexes with the ElA proteins in adenovirus-infected cells. Cell 58, 981-90.

Glenn, G.M. and R.P. Ricciardi (1985) Adenovirus 5 early region lA host range mutants hr3, hr4, and hr5 contain point mutations which generate single amino acid substitutions. ]. Virol. 56, 66-74.

Glenn, G.M. and R.P. Ricciardi (1987) An adenovirus type 5 ElA protein with a single amino acid substitution blocks w ild-type ElA transactivation. Mol. Cell. Biol. 7, 1004-11.

Gonzalez, G.A., P. Menzel, J. Leonard, W.H. Fischer and M.R. Montminy(1991) Characterization of motifs which are critical for activity of the cyclic AMP-responsive transcription factor CREB. Mol. Cell Biol. 11, 1306-12.

152

Gonzalez, G.A., K.K. Yamamoto, W.H. Fischer, D. Karr, P. Menzel, W. Biggs, W.W. Vale and M.R. Montminy (1989) A cluster of phosphorylation sites on the cyclic AMP-regulated nuclear factor CREB predicted by its sequence. Nature 337, 749-52.

Graham, F.L., T. Harrison and J. Williams (1978) Defective transforming capacity of adenovirus type 5 host-range mutants. Virology 8 6 ,10-21.

Graves, B., P.F. Johnson and S.L. McKnight (1986) H om ologous recognition of a promoter domain common to the MSV LTR and the HSV tk gene. Cell 44, 565-76.

Green, M., P.M. Loewenstein, R. Pusztai and J.S. Symington (1988) An adenovirus ElA protein domain activates transcription in vivo and in vitro in the absence of protein synthesis. Cell 53, 921-6.

Green, M.R., R. Treisman and T. Maniatis (1983) Transcriptional activation of cloned human beta globin genes by viral immediate- early gene products. Cell 35, 137-48.

Guilfoyle, T.J., G. Hagen and S. Malcolm (1984) Immunological studies on plant DNA-dependent RNA polymerases with antibodies raised against individual subunits. /. Biol. Chem. 259, 640-8.

Hahn, S., S. Buratowski, P.A. Sharp and L. Guarente (1989) Yeast TATA- binding protein TFIID binds to TATA elements with both consensus and nonconsensus DNA sequences. Proc. Natl Acad. Sci. USA 8 6 , 5718-22.

Hai, T., M. Horikoshi, R.G. Roeder and M.R. Green (1988a) Analysis of the role of the transcription factor ATF in the assembly of a functional preinitiation complex. Cell 54, 1043-51.

Hai, T., F. Liu, E.A. Allegretto, M. Karin and M.R. Green (1988b) A family of immunologically related transcription factors that includes multiple forms of ATF and AP-1. Genes Dev. 2 , 1216-26.

Hai, T., F. Liu, W.J. Coukos and M.R. Green (1989) Transcription Factor ATF cDNA clones: An Extensive Family of Leucine Zipper Proteins Able to Selectively Form DNA Binding Heterodimers. Genes Dev. 3,2083-90.

Haley, K.P., J. Overhauser, L.E. Babiss, H.S. Ginsberg and N.C. Jones (1984) Transformation properties of type 5 adenovirus mutants that differentially express the ElA gene products. Proc. Natl Acad. Sci. USA 81, 5734-8.

153

Handa, H. and P.A. Sharp (1984) Requirement for distal upstream sequences for maximal transcription in vitro of early region IV of adenovirus. Mol. Cell. Biol. 4, 791-8.

Hardy, S. and T. Shenk (1989) E2F from adenovirus-infected cells binds cooperatively to DNA containing two properly oriented and spaced recognition sites. Mol. Cell. Biol. 9, 4495-506.

Hardy, S., D.A. Engel and T. Shenk (1989) An adenovirus early region 4 gene product is required for induction of the infection-specific form of cellular E2F activity. Genes Dev. 3, 1062-74.

Harlow, E., P. Whyte, B.R. Franza and C. Schley (1986) Association of adenovirus early-region lA proteins with cellular polypeptides. Mol. Cell. Biol. 6 , 1579-89.

Harrington, C.A., E.J. Lewis, D. Krzemien and D.M. Chikaraishi (1987) Identification and cell type specificity of the tyrosine hydroxylase gene promoter. Nucl. Acids Res. 15, 2363-84.

Hearing, P. and T. Shenk (1983) The adenovirus type 5 ElA transcriptional control region contains a duplicated enhancer element. Cell 33, 695-703.

Hen, R., E. Borrelli and P. Chambon (1985) Repression of the immunoglobulin heavy chain enhancer by the adenovirus-2 ElA products. Science 230, 1391-4.

Hoeffler, J.P., T.E. Meyer, Y. Yun, J.L. Jameson and J.F. Habener (1988a) Cyclic AMP-responsive DNA-binding protein: structure based on a cloned placental cDNA. Science 242, 1430-3.

Hoeffler, W.K., R. Kovelman and R.G. Roeder (1988b) Activation of Transcription Factor IIIC by the Adenovirus ElA Protein. Cell 53, 907-20.

H oeffler, W.K. and R.G. Roeder (1985) Enhancement of RNA polymerase III transcription by the ElA gene product of adenovirus. Cell 41, 955-63.

Hoey, T., B.D. Dynlacht, M.G. Peterson, B.F. Pugh and R. Tjian (1990) Isolation and characterization of the Drosophila gene encoding the TATA box binding protein, TFIID. Cell 61, 1179-86.

Hoffmann, A., E. Sinn, T. Yamamoto, J. Wang, A. Roy, M. Horikoshi and R.G. Roeder (1990) Highly conserved core domain and unique N-terminus with presumptive regulatory motifs in a human TATA factor (TFIID). Nature 346, 387-90.

154

Huang, M.M. and P. Hearing (1989) The adenovirus early region 4 open reading frame 6 /7 protein regulates the DNA binding activity of the cellular transcription factor, E2F, through a direct complex. Genes Dev 3, 1699-710.

Hunninghake, G.W., M.M. Monick, B. Liu and M.F. Stinski (1989) The promoter-regulatory region of the major immediate-early gene of human cytomegalovirus responds to T-lymphocyte stimulation and contains functional cyclic AMP-response elements. /. Virol. 63, 3026-33.

Hurst, H.C. and N.C. Jones (1987) Identification of factors that interact with the ElA-inducible adenovirus E3 promoter. Genes Dev. 1, 1132-46.

Hurst, H.C., N. Masson, N.C. Jones and K.A.W. Lee (1990) The cellular transcription factor CREB corresponds to activating transcription factor 47 (ATF-47) and forms complexes with a group of polypeptides related to ATF-43. Mol. Cell. Biol. 10, 6192-203.

Hurst, H.C., N.F. Totty and N.C. Jones (1991) Identification and functional characterisation of the cellular activating transcription factor 43 (ATF-43) protein. Nucl. Acids Res. 19, 4601-9.

Impériale, M.J., R. Hart and J.R. Ne vins (1985) An enhancer-like element in the adenovirus E2 promoter contains sequences essential for uninduced and ElA-induced transcription. Proc. Natl Acad. Sci. USA 82, 381-5.

Impériale, M.J., H.T. Kao, L.T. Feldman, J.R. Nevins and S. Strickland (1984) Common control of the heat shock gene and early adenovirus genes: evidence for a cellular ElA-like activity. Mol. Cell. Biol. 4, 867-74.

Ingles, C.J., M. Shales, W.D. Cress, S.J. Triezenberg and J. Greenblatt (1991) Reduced binding of TFIID to transcriptionally compromised mutants of VP16. Nature 351, 588-90.

Inostroza, J., O. Flores and D. Reinberg (1991) Factors involved in specific transcription by mammalian RNA polymerase II. ]. Biol. Chem. 266, 9304-8.

Ivashkiv, L.B., H.C. Liou, C.J. Kara, W.W. Lamph, I.M. Verma and L.H. Glimcher (1990) mXBP/CRE-BP2 and c-Jun form a complex which binds to the cyclic AMP, but not to the 12-O-tetradecanoylphorbol- 13-acetate, response element. Mol. Cell. Biol. 10, 1609-21.

Johnson, P.P., W.H. Landschulz, B.J. Graves and S.L. McKnight (1987) Identification of a rat liver nuclear protein that binds to the

155

enhancer core element of three animal viruses. Genes Dev. 1 , 133-46.

Jones, C. and K.A.W. Lee (1991) ElA-mediated activation of the adenovirus E4 promoter can occur independently of the cellular transcription factor E4F. M ol Cell. Biol. 11, 4297-305.

Jones, K.A., K.R. Yamamoto and R. Tjian (1985) Two distinct transcription factors bind to the HSV thymidine kinase promoter in vitro. Cell 42, 559-72.

Jones, N. and T. Shenk (1979) An adenovirus type 5 early gene function regulates expression of other early viral genes. Proc. Natl Acad. Sci. USA 76, 3665-9.

Kadonaga, J.T., K.R. Garner, F.R. Masiarz and R. Tjian (1987) Isolation of cDNA encoding transcription factor Spl and functional analysis of the DNA binding domain. Cell 51, 1079-90.

Kao, C.C., P.M. Lieberman, M.G. Schmidt, Q. Zhou, R. Pei and A.J. Berk (1990) Gloning of a transcriptionally active human TATA binding factor. Science 248, 1646-50.

Kao, H.T. and J.R. N evins (1983) Transcriptional activation and subsequent control of the human heat shock gene during adenovirus infection. Mol. Cell. Biol. 3, 2058-65.

Kara, G.J., H.G. Liou, L.B. Ivashkiv and L.H. Glimcher (1990) A cDNA for a human cyclic AMP response element-binding protein which is distinct from GREB and expressed preferentially in brain. Mol. Cell. Biol. 10, 1347-57.

Keegan, L., G. Gill and M. Ptashne (1986) Separation of DNA binding from the transcription activating function of a eukaryotic regulatory protein. Science 231, 699-704.

Kelleher, R.J., P.M. Flanagan and R.D. Kornberg (1990) A novel mediator between activator proteins and the RNA polymerase II transcription apparatus. Cell 61, 1209-15.

Kimelman, D., J.S. Miller, D. Porter and B.F. Roberts (1985) FIA regions of the human adenoviruses and of the highly oncogenic simian adenovirus 7 are closely related. J. Virol. 53, 399-409.

Kingston, R.F., T.J. Schuetz and Z. Larin (1987) Heat-inducible human factor that binds to a human hsp70 promoter. Mol Cell Biol 7, 1530-4.

Kitchingman, G.R. and H. Westphal (1980) The structure of adenovirus 2 early nuclear and cytoplasmic RNAs. J. Mol. Biol. 137, 23-48.

156

Kleinberger, T. and T. Shenk (1991) A protein kinase is present in a complex with adenovirus ElA proteins. Proc. Natl Acad. Sci. USA 88,11143-47.

Kouzarides, T. and E. Ziff (1988) The role of the leucine zipper in the fos- jun interaction. Nature 336, 646-51.

Landschulz, W.H., P.F. Johnson, E.Y. Adashi, B.J. Graves and S.L. McKnight (1988) Isolation of a recombinant copy of the gene encoding C/EBP. Genes Dev. 2 , 786-800.

Landschulz, W.H., P.F. Johnson and S.L. McKnight (1988) The leucine zipper: a hypothetical structure common to a new class of DNA binding proteins. Science 240, 1759-64.

Laybourn, P.J. and M.E. Dahmus (1989) Transcription-dependent structural changes in the C-terminal domain of mammalian RNA polymerase subunit Ila/o. /. Biol. Chem. 264, 6693-8.

Lee, C.Q., Y. Yun, J.P. Hoeffler and J.F. Habener (1990) Cyclic-AMP- responsive transcriptional activation of CREB-327 involves interdependent phosphorylated subdomains. EMBO J. 9, 4455-65.

Lee, J.M. and A.L. Greenleaf (1989) A protein kinase that phosphorylates the C-terminal repeat domain of the largest subunit of RNA polymerase II. Proc. Natl Acad. Sci. USA 8 6 , 3624-8.

Lee, K.A.W., J.S. Fink, R.H. Goodman and M.R. Green (1989) Distinguishable Promoter Elements are Involved in Transcriptional Activation by ElA and Cyclic AMP. Mol. Cell. Biol. 9, 4390-7.

Lee, K.A.W. and M.R. Green (1987) A cellular transcription factor E4F1 interacts with an ElA-inducible enhancer and mediates constitutive enhancer function in vitro. EMBO ]. 6 , 1345-53.

Lee, K.A.W., T.Y. Hai, L. SivaRaman, B. Thimmappaya, H.C. Hurst, N.C. Jones and M.R. Green (1987a) A cellular protein, activating transcription factor, activates transcription of multiple ElA- inducible adenovirus early promoters. Proc. Natl Acad. Sci. USA 84, 8355-9.

Lee, W., P. Mitchell and R. Tjian (1987b) Purified transcription factor AP-1 interacts with TPA-inducible enhancer elements. Cell 49, 741-52.

Lee, W.S., C.C. Kao, G.O. Bryant, X. Liu and A.J. Berk (1991) Adenovirus ElA activation domain binds the basic repeat in the TATA box transcription factor. Cell 67, 365-76.

Leff, T., R. Elkaim, C.R. Goding, P. Jalinot, P. Sassone-Corsi, M. Perricaudet, C. Kedinger and P. Chambon (1984) Individual products of the

157

adenovirus 12 S and 13 S mRNAs stimulate viral E2A and E3 expression at the transcriptional level. Proc. Natl Acad. Sci. USA 81, 4381-5.

Lenardo, M., J.W. Pierce and D. Baltimore (1987) Protein-binding sites in Ig gene enhancers determine transcriptional activity and inducibility. Science 236, 1573-7.

Leong, K. and A.J. Berk (1986) Adenovirus early region lA protein increases the number of template molecules transcribed in cell-free extracts. Proc. Natl Acad. Sci. USA 83, 5844-8.

Leong, K., L. Brunet and A.J. Berk (1988) Factors responsible for the higher transcriptional activity of extracts of adenovirus-infected cells fractionate with the TATA box transcription factor. Mol. Cell. Biol. 8,1765-74.

Lewis, E.D. and J.L. Manley (1985) Control of adenovirus late promoter expression in two human cell lines. Mol. Cell. Biol. 5, 2433-42.

Lewis, J.B. and M.B. Mathews (1980) Control of adenovirus early gene expression: a class of immediate early products. Cell 2 1 , 303-13.

Lewis, M.K. and R.R. Burgess (1982) Eukaryotic RNA polymerases, in: The Enzymes, (eds. P.O. Boyer) Academic Press, New York, 109-53.

Lillie, J.W., M. Green and M.R. Green (1986) An adenovirus ElA protein region required for transformation and transcriptional repression. Cell 46, 1043-51.

Lillie, J.W. and M.R. Green (1989) Transcription activation by the adenovirus El a protein. Nature 338, 39-44.

Lillie, J.W., P.M. Loewenstein, M.R. Green and M. Green (1987) Functional domains of adenovirus type 5 ElA proteins. Cell 50, 1091-100.

Lin, Y.-S., I. Ha, E. Maldonado, D. Reinberg and M.R. Green (1991) Binding of general transcription factor TFIIB to an acidic activating region. Nature 353, 569-71.

Liou, H.-C., M.R. Boothby, P.W. Finn, R. Davidon, N. Nabavi, N.J. Zeleznik-Le, J.P.-Y. Ting and L.H. Glimcher (1990) A new member of the leucine zipper class of proteins that binds to the HLA DRa promoter. Science 247, 1581-4.

Liu, F. and M.R. Green (1990) A Specific Member of the ATF Transcription Factor Family Can Mediate Transcription Activation by the Adenovirus ElA protein. Cell 61, 1217-24.

158

Loewenstein, P.M. and M. Green (1989) In vitro trans-actiwation by chemically synthesized adenovirus region 3 peptides. Reaction properties and mutational analysis. / Biol Chem 264, 21504-8.

Maekawa, T., H. Sakura, I.C. Kanei, T. Sudo, T. Yoshimura, J. Fujisawa, M. Yoshida and S. Ishii (1989) Leucine zipper structure of the protein CRE-BPl binding to the cyclic AMP response element in brain. EMBO /. 8,2023-8.

Maguire, H.F., J.P. Hoeffler and A. Siddiqui (1991) HBV X protein alters the DNA binding specificity of CREB and ATF-2 by protein-protein interactions. Science 252, 842-4.

Maldonado, E., I. Ha, P. Cortes, L. Weis and D. Reinberg (1990) Factors involved in specific transcription by mammalian RNA polymerase II - role of transcription factor-IIA, factor-IID and factor-IIB during formation of a transcription competent complex. Mol. Cell. Biol. 10, 6335-47.

Maniatis, T., E.F. Fritsch and J. Sambrook. Molecular Cloning: A laboratory manual. 1982.

Maniatis, T., S. Goodbourn and J.A. Fischer (1987) Regulation of inducible and tissue-specific gene expression. Science 236, 1237-45.

Manley, J.L., A. Fire, A. Cano, P.A. Sharp and M.L. Gefter (1980) DNA- dependent transcription of adenovirus genes in a soluble whole-cell extract. Proc. Natl Acad. Sci. USA 77, 3855-9.

Martin, K.J., J.W. Lillie and M.R. Green (1990) Evidence for interaction of different eukaryotic transcription activators with distinct cellular targets. Nature 346, 147-52.

Memet, S., M. Gouy, C. Marck, A. Sentenac and J.M. Buhler (1988) RPA190, the gene coding for the largest subunit of yeast RNA polymerase A. J. Biol. Chem. 262, 2830-9.

Mermod, N., E.A. O'Neill, T.J. Kelly and R. Tjian (1989) The proline-rich transcriptional activator of CTF/NF-I is distinct from the replication and DNA binding domain. Cell 58, 741-53.

Mielnicki, L.M. and B.C. Pruitt (1991) Isolation and nucleotide sequence of a murine cDNA homologous to human activating transcription factor 4. Nucl. Acids Res. 19, 6332.

Miesfeld, R., S. Okret, A.C. Wikstrom, O. Wrange, J.A. Gustafsson and K.R. Yamamoto (1984) Characterization of a steroid hormone receptor gene and mRNA in wild-type and mutant cells. Nature 312, 779-81.

159

Miller, J., A.D. McLachlan and A. Klug (1985) D N A-dependent transcription of adenovirus genes in a soluble whole-cell extract. EMBO /. 4,1609-14.

M itchell, P.J. and R. Tjian (1989) Transcriptional regulation in mammalian cells by sequence-specific DNA binding proteins. Science 245, 371-8.

M ontell, C., G. Courtois, C. Eng and A.J. Berk (1984) Complete transformation by adenovirus 2 requires both ElA proteins. Cell 26, 951-61.

Montell, C., E.F. Fisher, M.H. Caruthers and A.J. Berk (1982) Resolving the functions of overlapping viral genes by site-specific mutagenesis at a mRNA splice site. Nature 295, 380-4.

Montminy, M.R. and L.M. Bilezikjian (1987) Binding of a nuclear protein to the cyclic-AMP response element of the somatostatin gene. Nature 328, 175-8.

Moran, E. and M.B. Mathews (1987) Multiple functional domains in the adenovirus ElA gene. Cell 48, 177-8.

Moran, E., B. Zerler, T.M. Harrison and M.B. Mathews (1986) Identification of separate dom ains in the adenovirus E lA gene for immortalization activity and the activation of virus early genes. M ol Cell Biol 6 , 3470-80.

Nevins, J.R., H.S. Ginsberg, J.M. Blanchard, M.C. Wilson and J.E. Darnell (1979) Regulation of the primary expression of the early adenovirus transcription units. /. Virol 32, 727-33.

N ichols, M., F. Weih, W. Schmid, C. DeVack, E. Kowentz-Leutz, B. Luckow, M. Boshart and G. Schütz (1992) Phosphorylation of CREB affects its binding to high and low affinity sites: Implications for cAMP induced gene transcription. EMBO /. 11, 3337-46.

Parraga, G., S.J. Horvath, H. Eisen, W.E. Taylor, L. Hood, E.T. Young and R.E. Klevit (1988) Zinc-dependent structure of a single-finger domain of yeast ADRl. Science 241, 1489-92.

Payne, J.M., P.J. Laybourn and M.E. Dahmus (1989) The transition of RNA polymerase II from initiation to elongation is associated with phosphorylation of the carboxyl-terminal domain of subunit Ila. /. Biol Chem. 264, 19621-9.

Perricaudet, M., G. Akusjarvi, A. Virtanen and U. Petterson (1979) Structure of two spliced mRNAs from the transforming region of human subgroup C adenoviruses. Nature 281, 694-6.

160

Peterson, M.G., N. Tanese, B.F. Pugh and R. Tjian (1990) Functional domains and upstream activation properties of cloned human TATA binding protein. Science 248, 1625-30. [Published erratum appears in Science (1990) 249, 844].

Pierce, J.W., M. Lenardo and D. Baltimore (1988) Oligonucleotide that binds nuclear factor NF-kappa B acts as a lymphoid-specific and inducible enhancer element. Proc Natl Acad Sci U S A 85, 1482-6.

Pines, J. and T. Hunter (1990) Human cyclin A is adenovirus ElA- associated protein p60 and behaves differently from cyclin B. Nature 346,760-3.

Pugh, B.F. and R. Tjian (1990) Mechanism of transcriptional activation by Spl: Evidence for coactivators. Cell 61, 1187-97.

Pugh, B.F. and R. Tjian (1991) Transcription from a TATA-less promoter requires a multisubunit TFIID complex. Genes Dev. 5, 1935-45.

Purnell, B.A. and D.S. Gilmour (1993) Contribution of sequences downstream of the TATA element to a protein-DNA complex containing the TATA-binding protein. Mol. Cell. Biol. 13, 2593-603.

Pusztai, R., P.M. Loewenstein and M. Green (1989) Mutational analysis of autonomously functioning trans-actiwàting peptides encoded by adenovirus ElA: evidence for two subdomains. / Virol 63, 3516-8.

Raychaudhuri, P., S. Bagchi and J.R. Nevins (1989) DNA-binding activity of the adenovirus-induced E4F transcription factor is regulated by phosphorylation. Genes Dev. 3, 620-7.

Raychaudhuri, P., R. Rooney and J.R. Nevins (1987) Identification of an ElA-inducible cellular factor that interacts with regulatory sequences within the adenovirus E4 promoter. EMBO J. 6 , 4073-81.

Rehfuss, R.P., K.M. Walton, M.M. Loriaux and R.H. Goodman (1991) The cAM P-regulated enhancer-binding protein ATF-1 activates transcription in response to cAMP-dependent protein kinase A. J. Biol. Chem. 266, 18431-34.

Reinberg, D. and R.G. Roeder (1987) Factors involved in specific transcription of mammalian RNA polymerase II. Transcription factor IIS stimulates elongation of RNA chains. ]. Biol. Chem. 262, 3331-7.

Renart, M.F., L. Sastre, V. Diaz and J. Sebastian (1985) Purification and subunit structure of RNA polymerases I and II from Dictyostelium discoideum vegetative cells. Mol. Cell. Biochem. 6 6 , 21-9.

161

Ricciardi, R.P., R.L. Jones, C.L. Cepko, P.A. Sharp and B.E. Roberts (1981) Expression of early adenovirus genes requires a viral encoded acidic polypeptide. Proc. Natl Acad. Sci. USA 78, 6121-5.

Roeder, R.G. (1991) The complexities of eukaryotic transcription initiation: Regulation of preinitiation complex assembly. T.I.B.S. 16, 402-8.

Roeder, R.G. and W.J. Rutter (1969) Multiple forms of DNA-dependent RNA polymerase in eukaryotic organisms. Nature 224, 234-7.

Roeder, R.G. and W.J. Rutter (1970) Multiple ribonucleic acid polymerases and ribonucleic acid synthesis during sea urchin development. Biochemistry 9, 2543-53.

Roeder, R.G. and W.J. Rutter (1970) Specific nucleolar and nucleoplasmic RNA polymerases. Proc. Natl Acad. Sci. USA 65, 675-82.

Rooney, R.J., P. Raychaudhuri and J.R. Nevins (1990) E4F and ATF, two transcription factors that recognise the same site, can be distinguished both physically and functionally: A role for E4F in ElA frflws-activation. Mol. Cell. Biol. 10, 5138-49.

Sakurai, A., T. Maekawa, T. Sudo, S. Ishii and A. Kishimoto (1991) Phosphorylation of cAMP response element-binding protein, CRE-BPl, by cAMP-dependent protein-kinase and protein-kinase-C. Biochem. Biophys. Res. Commun. 181, 629-35.

Santoro, C., N. Mermod, P C. Andrews and R. Tjian (1988) A family of human CCA AT-box-binding proteins active in transcription and DNA replication: Cloning and expression of multiple cDNAs. Nature 334, 218-24.

Sassone-Corsi, P., J. Visvader, L. Ferland, P.L. Mellon and I.M. Verma (1988) Induction of proto-oncogene fos transcription through the adenylate cyclase pathway: characterization of a cAMP-responsive element. Genes Dev. 2 , 1529-38.

Sawadogo, M. (1990) RNA polymerase B (II) and general transcription factors. Ann. Rev. Biochem. 59, 711-54.

Sentenac, A. (1985) Eukaryotic RNA polymerases. Crit. Rev. Biochem. 18, 31-91.

Shaw, A.R. and E.B. Ziff (1980) Transcripts from the adenovirus-2 major late promoter yield a single early family of 3’ coterminal mRNAs and five late families. Cell 2 2 , 905-16.

Shaw, A.R. and E.B. Ziff (1982) Selective inhibition of adenovirus type 2 early region II and III transcription by an anisomycin block of protein synthesis. Mol. Cell. Biol. 2 , 789-99.

162

Shi, Y., E. Seto, L.-S. Chang and T. Shenk (1991) Transcriptional repression by YYl, a human GLI-Kruppel-related protein, and relief of repression by adenovirus ElA protein. Cell 67, 377-88.

Sigler, B.S. (1988) Selective inhibition of adenovirus type 2 early region II and III transcription by an anisomycin block of protein synthesis. Nature 333, 210-2.

Simon, M.C., T.M. Fisch, B.J. Benecke, J.R. Nevins and N. Heintz (1988) Definition of multiple, functionally distinct TATA elements, one of which is a target in the hsp70 promoter for ElA regulation. Cell 52, 723-9.

SivaRaman, L., S. Subramanian and B. Thimmappaya (1986) Identification of a factor in HeLa cells specific for an upstream transcriptional control sequence of an ElA-inducible adenovirus promoter and its relative abundance in infected and uninfected cells. Proc. Natl Acad. Sci. USA 83, 5914-8.

Smale, S.T. and D. Baltimore (1989) The "initiator" as a transcription control element. Cell 57, 103-13.

Smale, S.T., M.C. Schmidt, A.J. Berk and D. Baltimore (1990) Transcriptional activation by Spl as directed through TATA or initiator: Specific requirement for mammalian transcription factor TFIID. Proc. Natl Acad. Sci. USA 87, 4509-13.

Spangler, R., M. Bruner, B. Dalie and M.L. Harter (1987) Activation of adenovirus promoters by the adenovirus ElA protein in cell-free extracts. Science 237, 1044-6.

Staudt, L.M., H. Singh, R. Sen, T. Wirth, P.A. Sharp and D. Baltimore (1986) A lymphoid-specific protein binding to the octamer motif of immunoglobulin genes. Nature 323,

Stein, R. and E.B. Ziff (1984) HeLa cell P tubulin gene transcription is stimulated by adenovirus 5 in parallel with viral early genes by an ElA-dependent mechanism. Mol. Cell. Biol. 4, 2792-801.

Steitz, T.A., D.H. Ohlendorf, D.B. McKay, W.F. Anderson and B.W. Matthews (1982) Structural similarity in the DNA-binding domains of catabolite gene activator and cro repressor proteins. Proc. Natl Acad. Sci. USA 79, 3097-100.

Stephens, C., B.R. Franza, C. Schley and E. Harlow (1986) Heterogeneity of adenovirus ElA proteins is due to post-translational modification of the primary translation products of the 1 2 S and 13 S mRNAs. Cancer Cells 4, 429-34.

163

Stephens, C. and E. Harlow (1987) Differential splicing yields novel adenovirus 5 ElA mRNAs that encode 30 kd and 35 kd proteins. EMBO }. 6 , 2027-35.

Stevens, A. and M.K. Maupin (1989) 5,6-Dichloro-l-beta-D-ribofuranosyl- benzimidazole inhibits a HeLa protein kinase that phosphorylates an RNA polymerase Il-derived peptide. Biochem. Biophys. Res. Commun. 159, 508-15.

Stringer, K.F., C.J. Ingles and J. Greenblatt (1990) Direct and selective binding of an acidic transcriptional activation domain to the TATA- box factor TFIID. Nature 345, 783-6.

Struhl, K. (1989) Helix-turn-helix, zinc finger, and leucine zipper motifs for eukaryotic transcriptional regulatory proteins. T.I.B.S. 14, 137-40.

Sumimoto, H., Y. Ohkuma, T. Yamamoto, M. Horikoshi and R.G. Roeder (1990) Factors involved in specific transcription by mammalian RNA polymerase II: identification of general transcription factor TFIIG. Proc. Natl Acad. Sci. USA 87, 9158-62.

Svensson, C. and G. Akusjarvi (1984) Adenovirus 2 early region lA stimulates expression of both viral and cellular genes. EMBO J. 3, 789-94.

Sweetser, D., M. Nonet and R.A. Young (1987) Prokaryotic and eukaryotic RNA polymerases have homologous core subunits. Proc. Natl Acad. Sci. USA 84, 1192-6.

Tanese, N., B.F. Pugh and R. Tjian (1991) Coactivators for a proline-rich activator purified from the multisubunit human TFIID complex. Genes Dev. 5, 2212-24.

Timmers, H.T.M. and P.A. Sharp (1991) The mammalian TFIID protein is present in two functionally distinct complexes. Genes Dev. 5, 1946-56.

Triezenberg, S.J., R.C. Kingsbury and S.L. McKnight (1988) Functional dissection of VP16, the trans-actiwator of herpes simplex virus immediate early gene expression. Genes Dev. 2 , 718-29.

Tsai, L.-H., F. Harlow and M. Meyerson (1991) Isolation of the human cdk2 gene that encodes the cyclin A- and adenovirus FlA-associated p33 kinase. Nature 353, 174-7.

Tsujimoto, A., H. Nyunoya, T. Morita, T. Sato and K. Shimotohno (1991) Isolation of cDNAs for DNA-binding proteins which specifically bind to a tax-responsive enhancer element in the long terminal repeat of human T-cell leukaemia virus type I. J. Virol 65, 1420-6.

164

Tsukada, T., J.S. Fink, G. Mandel and R.H. Goodman (1987) Identification of a région in the human vasoactive intestinal polypeptide gene responsible for regulation by cyclic AMP. /. Biol. Chem. 262, 8743-7.

Ulfendahl, P.J., S. Linder, J.P. Kreivi, K. Nordqvist, C. Sevensson, H. Hultberg, G. Akusjarvi (1987) A novel adenovirus-2 ElA mRNA encoding a protein with transcription activation properties. EMBO J. 6,2037-44.

van Ormondt, H., J. Maat and R. Dijkema (1980) Comparison of nucleotide sequence of the early ElA regions for subgroups A, B and C of human adenoviruses. Gene 1 2 , 63-76.

van Straaten, P., R. Mueller, T. Curran, C. van Beveran, and I.M. Verma (1983). Complete nucleotide sequence of human c-onc gene: Deduced amino acid sequence of the human c-fos protein. Proc. Natl Acad. Sci. USA 80, 3183-87.

Velcich, A. and E.B. Ziff (1985) Adenovirus Ela proteins repress transcription from the SV40 early promoter. Cell 40, 705-15.

Vinson, C.R., P.B. Sigler and S.L. McKnight (1989) Scissor-grip model for DNA recognition by a family of leucine zipper proteins. Science 246, 911-6.

Virtanen, A. and U. Peterson (1983) The molecular structure of the 9 S mRNA from early region lA of adenovirus serotype 2. }. Mol. Biol. 165,496-9.

Wada, T., H. Watanabe, Y. Usuda and H. Handa (1991) Different biological activities of the hetero- and homodimers formed by the 47- and 43- kilodalton proteins of transcription factor ATF/E4TF3. J. Virol. 65, 557-64.

Wasylyk, B. and P. Chambon (1981) A T to G base substitution and small deletions in the conalbumin TATA box drastically decrease specific in vitro transcription. Nucl. Acids Res. 9, 1813-24.

Watanabe, H., T. Imai, P.A. Sharp and H. Handa (1988) Identification of two transcription factors that bind to specific elements in the promoter of the adenovirus early-region 4. Mol. Cell. Biol. 8 , 1290-300.

Webster, L.C. and R.P. Ricciardi (1991) Trans-dominant mutants of El A provide genetic evidence that the zinc finger of the trans-activating domain binds a transcription factor. Mol. Cell. Biol. 1 1 , 4287-96.

Weinmann, R. (1992) The basic RNA polymerase II transcriptional machinery. Gene Expr. 2 , 81-91.

165

Weintraub, SJ. and D.C. Dean (1992) Interaction of a common factor with ATF, Spl, or TAT A A promoter elements is required for these sequences to mediate transactivation by the adenoviral oncogene Ela. Mol. Cell Biol 1 2 , 512-7.

Willems, L., R. Kettmann, G. Chen, D. Portetelle, A. Burny and D. Derse (1992) A cyclic AMP-responsive DNA-binding protein (CREB2) is a cellular transactivator of the bovine leukemia virus long terminal repeat. /. Virol 6 6 , 766-72.

Winberg, C. and T. Shenk (1984) Dissection of overlapping functions within the adenovirus type 5 ElA gene. EMBO J. 3,1907-12.

Woychik, N.A., S.M. Liao, P.A. Kolodziej and R.A. Young (1990) Subunits shared by eukaryotic nuclear RNA polymerases. Genes Dev. 4, 313-23.

Woychik, N.A. and R.A. Young (1990) RNA polymerase 11: Subunit structure and function. T.I.B.S. 15, 347-50.

Wu, L., D.S. Rosser, M.C. Schmidt and A. Berk (1987) A TATA box im plicated in E lA transcriptional activation of a sim ple adenovirus 2 promoter. Nature 326, 512-5.

Yamamoto, K.K., G.A. Gonzalez, W Biggs and M.R. Montminy (1988) Phosphorylation-induced binding and transcriptional efficacy of nuclear factor CREB. Nature 334, 494-8.

Yamamoto, K.K., G.A. Gonzalez, P. Menzel, J. Rivier and M.R. Montminy(1990) Characterization of a bipartite activator domain in transcription factor CREB. Cell 60, 611-7.

Yoshimura, T., J. Fujisawa and M. Yoshida (1990) Multiple cDNA clones encoding nuclear proteins that bind to the fa%-dependent enhancer of HTLV-1: all contain a leucine zipper structure and basic amino acid domain. EMBO J. 9, 2537-42.

Young, R.A. (1991) RNA polymerase 11. Ann. Rev. Biochem. 60, 689-715.Zenzie-Gregory, B., A. O'Shea-Greenfield and S.T. Smale (1992) Similar

mechanisms for transcriptional initiation mediated through a TATA box or an initiator element. ]. Biol. Chem. 267, 2823-30.

Zhao, L.-J. and C.-Z. Giam (1992) Human T-cell lymphotropic virus type 1 (HTLV-1) transcriptional activator. Tax, enhances CREB binding to HTLV-1 21-base-pair repeats by protein-protein interaction. Proc. Natl Acad. Sci. USA 89, 7070-4.

166

Zu, Y.-L., T. Maekawa, S. Matsuda and S. Ishii (1991) Complete putative metal finger and leucine zipper structures of CRE-BPl are required for the ElA-induced trans-activation, ]. Biol. Chem. 266, 24134-9.

Zu, Y.-L., Y. Takamatsu, M.-J. Zhao, T. Maekawa, H. Handa and S. Ishii(1992) Transcriptional regulation by a point mutant of adenovirus-2 ElA product lacking DNA binding activity. /. Biol. Chem. 267, 20181-7.

167

Appendix: Abbreviations not described in the text.

Ad2 (Ad5) adenovirus type 2 (type 5)AMV avian myeloblastosis virus

ATP adenosine-5'-triphosphatebp base pairs

cDNA complementary DNACHO Chinese hamster ovary

C-terminus (-terminal) carboxy-terminus (-terminal)CTP cytidine-5’-triphosphate

dATP 2'-deoxy-adenosine-5'-triphosphate

dCTP 2’-deoxy-cytidine-5'-triphosphateDEPC diethyl pyrocarbonatedGTP 2'-deoxy-guanosine-5'-

triphosphateDNA deoxyribonucleic acid

DNase I deoxyribonuclease IDTI 1,4-dithiothreitol (Cleland's

reagent)EDTA ethylenediaminetetraacetic acidEtOH ethanolGTP guanosine-5’-triphosphate

Hepes N-[2-hydroxyethyl] piperazine- N'-[2-ethanesulfonic acid]

His histidinelie isoleucine

IPTG isopropyl p-D- thiogalactopyranoside

kb kilobaseskDa kilodaltonsLeu leucine

MgOAc magnesium acetateNaOAc sodium acetate

NH 4 OAC ammonium acetateN-terminal (-terminus) amino-terminal (-terminus)

PAGE polyacrylamide gel electrophoresis

168

PBS phosphate buffered saline PEG polyethylene glycol Phe phenylalanine

PIPES piperazine-N,N'-bis [2-ethanesulfonic acid]; 1,4-piperazine- diethanesulfonic acid

PMSF phenylmethylsulfonyl fluoride pol II RNA polymerase II

Pro proline RNA ribonucleic acid

RNasin ribonuclease inhibitor rRNA ribosomal RNA

SDS sodium dodecyl sulfate Ser serine

TEMED N,N,N',N'-tetramethylethylene- diamine

Thr threonineTLC thin layer chromatography Tris tris[hydroxymethy 1]amino-

methane tRNA transfer RNA

TIP thymidine-5'-triphosphate Tyr tyrosine

UTP uridine-5'-triphosphate U.V. ultraviolet

169

ElA MEDIATED ACTIVATION OF THE ADENOVIRUS …ElA MEDIATED ACTIVATION OF THE ADENOVIRUS E4 PROMOTER....

Documents

Transcript of ElA MEDIATED ACTIVATION OF THE ADENOVIRUS …ElA MEDIATED ACTIVATION OF THE ADENOVIRUS E4 PROMOTER....