Regulation of Adenovirus Alternative Pre-mRNA Splicing165076/FULLTEXT01.pdf · processing (capping, polyadenylation, splicing), RNA export, RNA stability, translation, to post-translational

Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 926

_____________________________ _____________________________

Regulation of Adenovirus Alternative Pre-mRNA Splicing

Functional Characterization of Exonic and Intronic Splicing Enhancer Elements

BY

BAI-GONG YUE

ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2000

Dissertation for the Degree of Doctor of Medical Science in Medical Virology presented atUppsala University in 2000

ABSTRACT

Yue, B-G. 2000. Regulation of adenovirus alternative pre-mRNA splicing. Functionalcharacterization of exonic and intronic splicing enhancer elements. Acta UniversitatisUpsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty ofMedicine 926. 55pp. Uppsala. ISBN 91-554-4712-0

Pre-mRNA splicing and alternative pre-mRNA splicing are key regulatory steps controllinggene expression in higher eukaryotes. The work in this thesis was focused on acharacterization of the significance of exonic and intronic splicing enhancer elements for pre-mRNA splicing. Previous studies have shown that removal of introns with weak and regulated splice sitesrequire a splicing enhancer for activity. Here we extended these studies by demonstrating thattwo “strong” constitutively active introns, the adenovirus 52,55K and the Drosophila Ftzintrons, are absolutely dependent on a downstream splicing enhancer for activity in vitro. Two types splicing enhancers were shown to perform redundant functions as activators ofsplicing. Thus, SR protein binding to an exonic splicing enhancer element or U1 snRNPbinding to a downstream 5´splice site independently stimulated upstream intron removal. Thedata further showed that a 5´splice site was more effective and more versatile in activatingsplicing. Collectively the data suggest that a U1 enhancer is the prototypical enhancer elementactivating splicing of constitutively active introns. Adenovirus IIIa pre-mRNA splicing is enhanced more than 200-fold in infected extracts.The major enhancer element responsible for this activation was shown to consist of the IIIabranch site/polypyrimidne tract region. It functions as a Janus element and blocks splicing inextracts from uninfected cells while functioning as a splicing enhancer in the context ofinfected extracts. Phosphorylated SR proteins are essential for pre-mRNA splicing. Large amountrecombinant SR proteins are needed in splicing studies. A novel expression system wasdeveloped to express phosphorylated, soluble and functionally active ASF/SF2 in E. coli.

Bai-Gong Yue, Department of Medical Biochemistry and Microbiology, Biomedical Center,Uppsala University, Box 582, SE-751 23 Uppsala, Sweden

� Bai-Gong Yue 2000

ISSN 0282-7476ISBN 91-554-4712-0

Printed in Sweden by University Printers, Ekonomikum, Uppsala 2000

To my parents

MAIN REFERENCES

This thesis is based on the following articles, which will be referred to in the text by their

Roman numbers.

I. Bai-Gong Yue, Göran Akusjärvi. 1999. A downstream splicing enhancer is essential for in

vitro pre-mRNA splicing. FEBS Letters, 451 10-14

II. Oliver Muhlemann*, Bai-Gong Yue*, Svend Petersen-Mahrt and Göran Akusjärvi. 2000.

A novel type of splicing enhancer regulating adenovirus pre-mRNA splicing. Molecular and

Cellular Biology, 20:2317-2325

(*OM and BGY contributed equally to this work)

III. Bai-Gong Yue, Paul Ajuh, Göran Akusjärvi, Angus I. Lamond and Jan-Peter Kreivi.

2000. Functional coexpression of serine protein kinase SRPK1 and its substrate ASF/AF2 in

E. coli. Nucleic Acids Research 28:E14

Reprints were made with the permission from the publishers.

TABLE OF CONTENTS

1. INTRODUCTION 6

2. PRE-MRNA SPLICING2.1. Different patterns of splicing:2.2 The splicing machinery:2.2.1 The consensus sequences required for splicing2.2.2 Splicing factors and spliceosome assembly2.2.2.1 U snRNPs2.2.2.2 RS domain-containing proteins2.2.2.2.1 SR proteins2.2.2.2.2 SR–related proteins (SRrps)2.2.3 The chemistry in splicing reaction

10101111131316161920

2.3 Regulation of alternative splicing; 5' splice site and 3' splice site selection2.3.1 Positive and negative cis-regulatory elements;2.3.1.1 Splicing enhancer elements2.3.1.2 Splicing repressor elements2.3.2 The role of SR protein phosphorylation for splicing catalysis andalternative splicing

2021212223

3 ADENOVIRUS—A MODEL SYSTEM FOR MECHANICSTIC STUDIES OFRNA SPLICING3.1 Background3.2 The adenovirus system3.2.1 The early genes3.2.2 The major late transcription unit (MLTU)

24

24242528

4. PRESENT INVESTIGATION4.1 A downstream splicing enhancer is essential for splicing of all types of introns(paper I)4.2 A U1 splicing enhancer is more versatile than an SR splicing enhancer in promotingpre-mRNA splicing: comparison of the properties of two types of splicing enhancers(paper I)4.3 Identification of a virus infection dependent splicing enhancer, the 3VDE (paper II)4.4 U2AF binding to the IIIa polypyrimidine tract does not correlate with IIIa splicingactivation in Ad-NE (paper II)4.5 Development of a new expression system for production of phosphorylated,

biologically active recombinant ASF/SF2 in E. coli (paper III)

3030

31

3335

36

5. CONCLUSION AND DISCUSSION 375.1 The universal role of splicing enhancers5.2 The models for the 3VDE function5.3 Perspective for further identification of the 3VDF5.4 The significance of modification of recombinant proteins expressed in E. coli andperspective for future applications

37 37 38 40 40

6. ACKNOWLEDGMENTS 42

7. REFERENCES 44

- 6 -

1. INTRODUCTION

Multicellular organisms have over millions of years of evolution and have developed a

plethora of functions to adapt to changes in the environment. Different types of cells perform

these functions. The genetic information, which is needed for cell development, is stored in

the genome. Even though all cells in an organism carry the same genetic information, their

shape and function are highly differentiated and specialized. This differentiation in function

and shape is controlled by regulation of gene expression, which not only provides organisms

with a basic structure and governs their life processes, but also supplements them with the

ability to adapt to inner and outer environmental changes.

Gene expression is regulated at multiple levels: ranging from transcription, pre-mRNA

processing (capping, polyadenylation, splicing), RNA export, RNA stability, translation, to

post-translational modifications (Fig. 1).

The first step in gene expression is the synthesis of an RNA from the DNA template, a

mechanism called transcription. Transcription is carried out by three distinct classes of RNA

polymerases (pol I, II, and III), each enzyme responsible for the transcription of separate

groups of genes. RNA polymerase I transcribes ribosomal RNA genes; RNA polymerase II

transcribes protein-encoding messenger RNAs (mRNAs); and RNA polymerase III

transcribes small cellular RNAs like tRNAs and the 5S RNA component of the ribosome.

Transcription can be divided into five steps: formation of the pre-initiation complex,

transcription initiation, promoter clearance, elongation and termination. These steps are

coordinated processes, and tightly controlled according to the demand of the developmental

stage, cell type or changes in the surroundings.

During and after transcription, the primary transcript from pol II genes undergoes

several steps of processing to produce a functional mRNA. The 5’ end of the RNA molecule

is capped by the addition of a methylated inverted G nucleotide immediately after

transcription initiation. The 5’ cap structure serves important functions in stimulating pre-

mRNA splicing (Schwer and Shuman, 1996), mRNA transport (Hamm and Mattaj, 1990),

protein translation (Sonenberg, 1993), 3’ processing (Gilmartin et al., 1988), and protection of

the growing RNA transcript from degradation (Furuichi et al., 1977).

Except for the cell cycle regulated histone mRNAs, the 3’ ends of all pol II transcripts

carry a poly-A tail. In this modification, the growing transcript is cleaved at a specific site and

- 7 -

- 8 -

a poly-A tail (150-250 residues of adenylic acid) is added by the poly-A polymerase (Keller,

1995). Some genes contain alternative sites for polyadenylation. The usage of alternative poly

A sites is an important mechanism to regulate gene expression in eukaryotic cells. It is

believed that the poly-A tail serves important functions by increasing mRNA stability (Lewis

et al., 1995), promoting translational efficiency (Sachs et al., 1997), and also having a role in

the transport of the mature mRNA from the nucleus to the cytoplasm (Colgan and Manley,

1997).

The protein coding sequence in a typical eukaryotic gene is interrupted by intervening

sequences (introns). These introns are transcribed into the precursor RNA (the pre-mRNA),

and are removed by a process called pre-mRNA splicing (Sharp, 1994). Splicing is an

important level for control of gene expression, and is subjected to a dynamic regulation in

cells, and during development. The work in this thesis is focused on the regulation of pre-

mRNA splicing which will be discussed in the next chapter.

RNA export from the nucleus to cytoplasm is also an important checkpoint in the

control of gene expression. From this point of view, it is interesting to note that pre-mRNA

splicing has been shown to be required for efficient mRNA export in vivo(Luo and Reed,

1999). For example, a mutation creating a defect in RNA splicing results in the retention of

the pre-mRNA in the nucleus. However, there are a few cellular genes that do not contain

introns and therefore are transported without splicing to the cytosol (Luo and Reed, 1999).

Control of RNA export is also used by some viruses to subvert the host cell gene expression

machinery to enhance virus multiplication. Thus, viruses like adenovirus and HIV have

evolved mechanisms resulting in the selective export of viral mRNAs to the cytoplasm,

meanwhile, blocking cellular mRNAs for export (Pilder et al., 1986).

After translation, the proteins produced are not always ready to carry out their

function. The activity of many proteins is regulated through post-translational modification,

such as phosphorylation, glycosylation, methylation or acetylation (Krishna and Wold, 1993).

Of particular interests for this thesis is the phosphorylation of a group of splicing factors, the

so called SR proteins. The properties and functions of SR protein will be discussed in the

following sections.

Recent studies have shown that regulation of gene expression is a highly integrated

process. For example, RNA pol II transcription has been shown to be tightly coupled to pre-

mRNA processing: capping, splicing and cleavage / polyadenylation. This functional coupling

is achieved by means of physical contacts between the protein machinery that performs

transcription and mRNA processing (Proudfoot, 2000). The capping enzymes have been

- 9 -

shown to bind to the phosphorylated pol II CTD (carboxyl-terminal domain). The association

of capping enzyme with phosphorylated pol II serves to recruit capping enzymes to the site of

RNA synthesis (Cho et al., 1997; McCracken et al., 1997). Thus, when the nascent pre-

mRNA is about 25 bases long, it is modified by capping. Subsequently, the cap is recognized

by the cap binding complex (CBC), which enhances the subsequent splicing and 3’-end

processing reactions (Coppola et al., 1983; Visa et al., 1996).

Like capping, cleavage and polyadenylation are also dependent on the RNA pol II

CTD (McCracken et al., 1997). Polyadenylation factors bind to pol II via the CTD, and

enhance the cleavage reaction. The three subunits of CPSF (cleavage polyadenylation

specificity factor) have been shown to first bind to the TATA-box binding protein (TBP) and

are therefore part of the basal transcription factor TFIID (Dantonel et al., 1997). After

initiation of transcription CPSF can be co-immunoprecipitated with RNA pol II, which

implies that it is transferred from TFIID to the CTD early during transcription initiation

(Dantonel et al., 1997). It has been proposed that CPSF remains stabely associated with the

transcribing polymerase and deposited at the site of polyadenylation.

Figure 2. A model of the coupling between RNA processing factors and the

transcription apparatus. After transcription initiation, the CTD is

phosphorylated and associates with capping enzymes, splicing factors (SR

proteins), cleavage and polyadenylation factors (CPSF and CstF). These

associations target the processing machinery to the site of RNA synthesis, and

regulate their activities.

Transcription and splicing are highly coordinated processes both at the functional and

structural level (Corden and Patturajan, 1997; Kim et al., 1997). It has been reported that

truncation of the CTD inhibits splicing (McCracken et al., 1997). In vitro splicing is also

inhibited either by an antibody directed against the CTD or by wild-type but not mutant CTD

SR

P o l II

bp

pP o l II

SR

P o l IIp

- 10 -

peptides (Yuryev et al., 1996), demonstrating that CTD is relevant for mRNA splicing. A

coupling between transcription and splicing has also been shown at the promoter level. Thus

dependent on the RNA pol II promoter which drives the transcription of the reporter gene, the

SR proteins ASF/SF2 and 9G8 were shown to differently activate an exonic splicing

enhancer, consequently controlling the alternative splicing of the enhancer dependent EDI

exon in the fibronectin gene (Cramer et al., 1999; Cramer et al., 1997). Collectively, all the

results from these reports suggest that the CTD may function as a liner platform that recruits

the factors required for capping, splicing, cleavage and polyadenylation of a pre-mRNA (Fig.

2).

2. PRE-mRNA SPLICINGIn 1977 two research groups working at the Cold Spring Harbor Laboratory and MIT

independently discovered that adenovirus mRNAs were encoded by discontinuous segments

on the viral genome (Berget et al., 1977; Chow et al., 1977). This finding of split genes

opened up a new era of research in biological sciences. A result that suddenly explained

inconsistencies in the data obtained by many research groups. Within half a year after the

description of split genes and RNA splicing in the adenovirus system, cellular genes were

similarly shown to be encoded by discontinuous gene segments (exons) (Rabbitts and Forster,

1978). The number of introns in eukaryotic genes differs much, ranging from zero to more

than 70. But the majority of the genes contain several introns.

During the process of RNA synthesis, both exons and introns are transcribed from the

DNA template. Before the mature mRNA is formed and exported from the nucleus to the

cytoplasm, the intron must be excised from the primary transcript and the flanking exon

sequences joined together in the process of pre-mRNA splicing.

2.1 Different patterns of splicing

The basic type of pre-mRNA splicing is called constitutive splicing, in which each

exon in a pre-mRNA is spliced with the most adjacent flanking exons. Only one mature

mRNA is formed from a given pre-mRNA in constitutive splicing. The majority of pre-

mRNA splicing in eukaryotes belongs to this category (65%).

In alternative splicing, the exons from a pre-mRNA can be ligated in different

combinations, resulting in the production of multiple mature mRNAs with different sequence

compositions (Fig. 3). It has been estimated that approximately 35% of eukaryotic genes

- 11 -

produce alternatively spliced mRNAs (Mironov et al., 1999), implicating its significant role in

the control of gene expression in higher eukaryotic cells.

Soon after the discovery of splicing and alternative splicing in the adenovirus system,

a further advance was demonstrated that the accumulation of alternatively spliced adenovirus

mRNAs was a regulated process (Imperiale et al., 1995). Different mRNAs were shown to

accumulate at different stages of the infectious cycle in a reaction requiring late viral protein

synthesis. This finding added a new dimension to gene regulation. Thus, the regulation of

gene expression was not confined only to on / off switches of transcription, but also extended

to accumulation of multiple mRNAs, with the capacity to produce alternative proteins, from a

transcription unit. In a single step of control, several mRNAs can be produced in a regulated,

competitive manner, such that the reduced accumulation of one group of mRNAs can be

compensated by the increased accumulation of another set of mRNAs, according to the

demand of the cell type, or developmental stage of the organism.

2.2 The splicing machinery

2.2.1 The consensus sequences required for splicing

Figure 3. Different patterns of alternative splicing.

5. Mutually exclusive exons

6. Alternative promoters/ alternative 5' splice site

7. Alternative poly (A)/alternative 3' splice site

AAAAAA

1. Intron retention;

3. Alternative 5' splice site;

2. Alternative 3' splice site;

4. Exon skipping/inclusion;

- 12 -

The accuracy in splicing is vital for gene expression in all eukaryotic cells. This

requires precise recognition of the 5' splice site and the 3' splice site at the intron-exon

boundaries. Based on consensus sequences (Fig. 4), introns can be sorted into two different

groups: the vast majority of pre-mRNA introns possess invariant GU and AG dinucleotides at

their 5’ and 3’ termini and are removed by the so called major spliceosome; the minor class of

introns with the non-canonical terminal dinucleotides AU and AC at their 5’ and 3’ splice

sites are spliced by the minor spliceosome. The major and minor spliceosomes vary by using

different sets of U snRNPs. The consensus sequences at introns are highly conserved in yeast,

but much more degenerate in higher eukaryotes.

The 3' splice site region consists of three sequence elements: the branch site, the

polypyrimidine tract, and the AG dinucleotides at the 3' splice site boundary (AC

dinucleotides in the case of the minor class of introns). These three sequence elements

together provide a tight control for 3' splice site recognition. At the same time the diversity in

the sequence composition of the branch site and the polypyrimidine tract in different pre-

AGGU RAGU UNCUR AC YAGG

AGGU AUGU(Y)

AUAU CCUU UCCUU AAC YCCAC

UACUA AC

(M)

AT-AC intron

Majorintrons

(Y6-11)

CAGG

5' splice

site

3' splice

site

Figure 4. Conserved sequence elements in pre-mRNAs. Nucleotides with90% or higher conservation are shown in bold, the branchpoint is markedwith an asterisk. Mammalian introns (M) usually contain apolypyrimidine tract near the 3' splice site, whereas yeast (Y) introns maylack such a sequence. (R, purines ; Y, pyrimidines ; Y6-11,polypyrimidines ; N, variable nucleotides).

�

�

�

- 13 -

mRNAs provide abundant resources for the regulation of 3' splice site selection in alternative

splicing.

2.2.2 Splicing factors and spliceosome assembly

Pre-mRNA splicing occurs in a macromolecular complex known as the spliceosome,

which consists of the five small nuclear ribonucleoprotein particles (snRNPs), designated U1,

U2, U4/U6 and U5 snRNPs, and a large number of non-snRNP protein splicing factors.

2.2.2.1 U snRNPs

The snRNPs are major structural and functional components of the splicing

machinery. In tandem with the assistance of other splicing factors, snRNP complexes can

precisely recognize and define the consensus sequences at intron-exon boundaries in part by

base pairing. This provides a basis for the accuracy of splicing. Each of these snRNPs are

composed of small nuclear, U-rich RNA (U snRNA), as well as 8 to 15 different

polypeptides. Of the polypeptides present in snRNPs, eight proteins are common to all

snRNPs; the rest are specific for individual U snRNP. The eight common polypeptides, called

Sm proteins, are recognized by antibodies produced in SLE (systemic lupus erythematosus)

patients. The Sm proteins form a ring structure and bind to the conserved Sm binding site in

each U snRNA.

With the exception of U6 snRNA (transcripted by RNA polymerase III), all U

snRNAs involved in splicing (U1, U2, U4, and U5 snRNA) are transcribed by RNA

polymerase II. Their secondary structures are highly conserved between species, although the

primary sequence may vary. Of the five spliceosomal snRNAs only three (U2, U5 and U6)

are thought to contribute functionally during the two transesterification reactions of splicing.

The U snRNAs play diverse roles in spliceosome assembly and splicing catalysis. One

of these functions is to recognize and bind to the short consensus sequences at the 5’ and 3’

boundaries of the intron.

U1 snRNP is composed of the U1 snRNA and 11 polypeptides. Three of these

polypeptides A, C, and 70K are U1 snRNP specific (Klein Gunnewiek et al., 1997). The U1

70K protein is an RS domain containing protein important for U1 snRNP recruitment to the 5'

splice site. U1 snRNP is the first snRNP that bind to the pre-mRNA during the early stage of

spliceosome assembly, generating the so-called E-complex, or commitment complex. (Black

et al., 1985; Rossi et al., 1996). With the assistance of SR proteins which interact both with

the 5' splice site and the U1 70K protein, U1 snRNP form a stable interaction with nucleotides

- 14 -

+1 to +6 of the intron at the 5' splice site, this results in the selection of this splice site for

spliceosome assembly (Staknis and Reed, 1994).

It is believed that formation of the E complex irreversibly defines the exon-intron

boundaries and commits the pre-mRNA for spliceosome assembly and splicing (Fu, 1993).

Thus, formation of the E complex is likely to be a key step at which alternative splicing is

regulated.

Two protein splicing factors, U2AF and SF1, have also been found in the E complex.

Another function of U1 snRNP is to target U2AF to the polypyrimidine tract (Hoffman and

Grabowski, 1992), and promote the association of the U2 snRNP complex with the branch

site region (Barabino et al., 1990). Although U2 snRNP will specifically bind a pre-mRNA in

the absence of U1 snRNP, for most pre-mRNAs this reaction is not efficient (Query et al.,

1997). So the assistance of U1 snRNP is important for the efficient recruitment of U2 snRNP

during spliceosome assembly.

U2 snRNP interacts with the branch site at the 3' splice site by a short base pairing,

which results in the selection of the 3' splice site during spliceosome formation. This step is

ATP-dependent, and enhanced by U2AF (Ruskin et al., 1988), SF1 (Arning et al., 1996) and

U1 snRNP interaction with the pre-mRNA. SF3a and SF3b, which are part of U2 snRNP,

bind upstream of the branch site, to the so-called anchoring sequence, and stabilizes U2

snRNP binding (Brosi et al., 1993; Gozani et al., 1994). The binding of U2 snRNP to the

branch site sequence generates the so called A complex (the pre-spliceosome).

U5 snRNP joins the spliceosomal complex in the form of the U4/U6-U5 tri-snRNP

particle (Konarska and Sharp, 1987). The interaction between U5 snRNP and pre-mRNA is

not only essential for the first step of the reaction, but is also required for the second step of

splicing. An invariant loop sequence in U5 snRNA makes contact with the nucleotide at

position -1 of the first exon, before and after 5' splice site cleavage. This same loop will also

interact with the nucleotide at the position +1 of the second exon, within the lariat intron-exon

2 intermediate. Most likely, U5 snRNP serves as the factor that holds the free exon 1 and the

intermediate in the spliceosome, and aligns it with the 3' splice site for the second catalytic

reaction (Sontheimer and Steitz, 1993; Teigelkamp et al., 1995; Wyatt et al., 1992).

U4 and U6 snRNAs are present in a single snRNP, associated with one another

through extensive base pairing (Black and Steitz, 1986). U4/U6 snRNP bind U5 snRNP to

form the 17S tri-snRNPs particle before they join the A complex. Incorporation of the tri-

snRNP particle converts the A complex to the B complex (the spliceosome).

- 15 -

SR

AG exon 2U5

A

U6 exon 1

(Yn) -3'

-5'

SR

AG(Yn)A exon 2GUexon 1

3' splice site

5' splice

site

pre-mRNA

complex E

complex A

complex B

complex C

mRNA

AG exon 2A -3'

-5'

exon 1

AG exon 2A -3'

-5'

ATPU 2

U5

U6

ATP

ATP

exon 2U5

A

U6

U2AF

U2AF

AG(Yn)

exon 1

A

U6

AG(Yn)

U5 exon 2exon 1 -3'5'-

5'- -3'

-3'

-5'

SR

SR

SRSR

U2AF

SRSRSR

U2AF

SR

Figure 5. The spliceosome assembly pathway and splicing process.For detail, see text.

U 2

U 2

U 2

U 2

- 16 -

Through conformational changes the B complex is converted to the C complex, the

active spliceosome. Thus, the 5' splice site-U1 snRNP base pairing is disrupted and replaced

by a U6 base pairing with the 5' splice site (Kandels-Lewis and Seraphin, 1993; Konforti et

al., 1993; Lesser and Guthrie, 1993; Sontheimer and Steitz, 1993). Simultaneously base

pairing between U4 and U6 is disrupted and U6 now interacts with U2 via base pairing

(Madhani and Guthrie, 1994). U5 and U6 snRNAs collaborate during these reactions and

align the two splice sites for the final exon-exon ligation reaction (Kandels-Lewis and

Seraphin, 1993; Lesser and Guthrie, 1993; Newman and Norman, 1992; Sontheimer and

Steitz, 1993). It is believed, that U6 snRNP catalyses the splicing reaction (Newman,

1994).(Fig. 5)

2.2.2.2. RS domain-containing proteins

In addition to snRNPs, the splicing process requires a large number of non-snRNP

splicing factors, containing domains rich in alternating arginine and serine residues (RS

domains). These include members of the SR protein family of splicing factors and SR-related

proteins (SRrp) that are structurally and functionally distinct from the SR family of splicing

factors.

2.2.2.2.1 SR proteins

The SR family of splicing factors are highly conserved in metazoan (Zahler et al.,

1993). The members of the family share the following properties: (1), they contain one or two

N-terminal RNA recognition motifs (RRMs), and a C-terminal RS domain; (2), they are

capable of activating splicing in splicing-defective cytoplasmic S100 extracts; (3), they

precipitate in the presence of 20 mM MgCl2; (4), they are recognized by the phospho-epitope

specific monoclonal antibody mAb 104.

The classical SR proteins consist of six members, SRp20, ASF/SF2, SC35, SRp40,

SRp55, and SRp75 ranging in size from 20—75 kDa (Fig. 6). However, the family is growing

and now also includes SRp30c , and 9G8.

The RRMs in SR proteins recognize and bind consensus sequences in the pre-mRNA,

such as the 5' splice site and splicing enhancer elements (Moras and Poterszman, 1995). The

observation that two SR proteins, ASF/SF2 and SC35, differ significantly in their ability to

commit specific pre-mRNAs to the splicing pathway provided the initial evidence for

divergent RNA binding specificities among SR proteins (Fu, 1993). Subsequent studies have

shown that each SR protein binds to a distinct RNA sequence element (Cavaloc et al., 1999;

- 17 -

Coulter et al., 1997; Tacke and Manley, 1995). Typically the consensus binding sites are

purine rich, 6—10 nucleotides long without obvious secondary structure (Tacke and Manley,

1999). The SELEX (Systemic Evolution of Ligands by EXponential enrichment) protocol has

been widely used to identify the consensus sequences mediating SR protein binding. In

functional SELEX (Liu et al., 1998; Schaal and Maniatis, 1999), a short random sequence

(about 20 nucleotides long) is inserted into a splicing enhancer dependent pre-mRNA to

replace the original enhancer sequence. A library of pre-mRNAs constructed in this way is

spliced in the presence of one type of SR proteins. The spliced product is isolated, and the

sequence corresponding to the random region amplified and rebuilt into pre-mRNA template

molecules for a new round of selection. An alternative SELEX approach is to isolate RNA

sequences showing the highest affinity for an SR protein (Tacke and Manley, 1995). It is

obvious that the sequences selected by functional SELEX are based on their capacity to

activate splicing rather than only RNA binding affinity.

SRP20RSRRM1

221 aaSRp30cRRM1 RRM2 RS

221SC35

9G8

SRp40

ASF/ SF2

SRp55

SRp75

Figure 6. The SR protein family comprises eight major polypeptides conservedbetween all higher eukaryotes, containing one or two RNA recognitionmotifs at the N-terminus and an RS domain of variable length and sequence composition at the C-terminus.

238

248

273

344

494

164 aa

aa

aa

aa

aa

aa

aa

- 18 -

The RS domain in SR proteins differs in their length, number of arginine-serine

dipeptides and the content of other amino acids. This region mediates protein-protein

interaction with RS domains in other SR proteins or in SR-related proteins. These protein-

protein interactions function in at least three steps in spliceosome assembly: (i) recruitment of

U1 snRNP to the 5' splice site; (ii) bridging the 5' splice site and 3' splice site by

simultaneously interacting with U1-70K and the U2AF35 component of U2AF; (iii)

regulating alternative splicing by interacting with splicing enhancer elements or splicing

repressor elements (Amrein et al., 1994; Kohtz et al., 1994; Wu and Maniatis, 1993; Xiao and

Manley, 1997). The RS domain is also required for proper subcellular (Colwill et al., 1996;

Koizumi et al., 1999) and subnuclear localization of SR proteins (Caceres et al., 1997; Hedley

et al., 1995).

SR proteins accumulate predominately in nuclear structures referred to as speckles,

which serves as storage and /or recycling sites for splicing factors (Misteli and Spector, 1998;

Singer and Green, 1997).

Early work suggested that SR proteins might be functionally redundant. However, a

number of subsequent studies have suggested that each protein probably perform at least

some non-redundant functions. For example, one SR protein, SRp55/B52, is essential for

proper development of Drosophila, and ASF/SF2, is essential for viability of a chicken B-cell

line (Tacke and Manley, 1999) and C. elegans development (Longman et al., 2000).

The RS domains of SR proteins are highly phosphorylated in vivo. Phosphorylation of

SR proteins increases their RNA binding affinity, reduces their unspecific binding (Misteli,

1999), and also affects their interactions with other proteins (Xiao and Manley, 1997).

Dephosphorylation of SR proteins occurs during the later stage of spliceosome assembly,

which appears to facilitate the configurational changes occurring during the spliceosome

maturation. Therefore both phosphorylation and dephosphorylation are vital for the function

of SR proteins in splicing (Cao et al., 1997). Through control of the phosphorylation status of

SR proteins, splicing and alternative splicing can be regulated (Kanopka et al., 1998; Xiao

and Manley, 1998), this has proven to be an important mechanism employed by both cells and

viruses. Several protein kinases and protein phosphatases are involved in this regulation;

examples include SRPK1, SRPK2, Clk/Sty, DNA topoisomerase I, cdc2, PP1, PP2A, PP2C

(Colwill et al., 1996; Gui et al., 1994; Mermoud et al., 1992; Okamoto et al., 1998; Rossi et

al., 1996) (Murray et al., 1999).

SR protein specific kinases have also been shown to regulate the subcellular

localization of SR proteins (Colwill et al., 1996; Koizumi et al., 1999). This might be an

- 19 -

alternative pathway for kinases to regulate the functions of SR proteins in splicing, e.g. to

control the activity of SR proteins by influencing their local concentration and availability,

instead of their behavior in protein-protein and protein-RNA interactions (Misteli and

Spector, 1997).

2.2.2.2.2 SR –related proteins (SRrps)

In addition to the SR family of splicing factors, some other proteins involved in

splicing also contain RS domains (SR-related proteins), for example snRNP-associated

splicing factor U1-70K, non-snRNP associated splicing factor U2AF, splicing regulators Tra,

Tra-2, SRm160 and SRm300 (Blencowe et al., 2000). These proteins also play important

roles in spliceosome assembly or in splicing catalysis. Similar to SR-SR protein interactions,

they interact with SR proteins through their RS domains.

SRrps are structurally and functionally distinct from SR proteins. They can not

complement splicing defective S100 extracts, which is the main characteristic of the classical

SR proteins. They contain one or two RS domains, but may lack an RNA recognition motif.

Some SRrps, like U2AF65, U1-70K, SRm160, and Sip1 are essential splicing factors

(Blencowe et al., 1998; Tazi et al., 1993; Zamore and Green, 1991; Zhang and Wu, 1998).

Antibody depletion of these proteins from nuclear extracts abolishes splicing. However, other

SRrps may be non-essential (at least for some pre-mRNAs), or only play regulatory roles in

pre-mRNA splicing, like U2AF35, Tra, and Tra2 (Guth et al., 1999; Tacke et al., 1998; Wu et

al., 1999).

U2AF is an essential splicing factor composed of a heterodimer of a 65 kDa

(U2AF65) and a 35kDa (U2AF35) subuint, which are present in approximately equimolar

amounts (Zamore and Green, 1989). U2AF is conserved between Drosophila and humans.

U2AF binds to the polypyrimidine tract at the 3' splice site and helps to recruit U2 snRNP to

the branch site during spliceosome assembly. The binding of U2 snRNP to the branch site is

the first ATP dependent step in spliceosome assembly. Purified U2 snRNP alone can not

stabely interact with the branch site sequence on the pre-mRNA (Ruskin et al., 1988). Instead,

stable binding of U2 snRNP requires assistance from U1 snRNP interacting with the 5' splice

site and at least three other protein factors SF1, SF3 and U2AF (Krämer and Utans, 1991;

Ruskin et al., 1988; Zamore and Green, 1989).

U2AF65 contains two functional domains: a sequence specific RNA-binding region

composed of three central RRMs, and an N-terminal RS domain (Zamore et al., 1992). All

three RRMs in U2AF65 are necessary for high affinity and sequence specific RNA binding

- 20 -

but are not sufficient for its in vitro splicing activity, which additionally requires the RS

domain (Zamore et al., 1992). After binding to the polypyrimidine tract, U2AF65 directs its

RS domain to contact the branch site, the positively charged surface of the RS domain

stabilizes base paring between U2 snRNP and the branch site (Gaur et al., 1995; Valcarcel et

al., 1996).

U2AF35 does not have an RRM but contains a C-terminal RS domain (Zhang et al.,

1992). Early work indicated that U2AF35 was dispensable for splicing. Thus, it was reported

that recombinant U2AF65 was sufficient to activate splicing of two constitutively active pre-

mRNAs in extracts that were chromatographically depleted of U2AF (Guth et al., 1999).

However, more recent data has been shown that U2AF35 is required for splicing of pre-

mRNAs containing introns with weak polypyrimidine tracts. These studies have demonstrated

a substract-specific requirement for U2AF35, and assigned this subunit a regulatory role in

alternative pre-mRNA splicing (Merendino et al., 1999; Wu et al., 1999; Zorio and

Blumenthal, 1999).

2.2.3 The chemistry of the splicing reaction

The splicing reaction consists of two transesterifications. The first step is initiated by

the 2’ hydroxyl group of the branch point adenosine attacking the 3'-5' phosphodiester bond at

the 5' splice site. This results in the cleavage at the 5’ intron-exon junction, generating two

splicing intermediates, a 5' exon with a free hydroxyl group at its 3’ terminus, and the so-

called lariat intermediate, in which the 5’ end of the intron is joined to the branch site

adenosine via a 5’-2’ phosphodiester bond (Adams et al., 1996; Newman, 1994). In the

second step, the free hydroxyl group of the 5’ exon attacks the 3’-5’ phosphodiester bond at

the 3' splice site, resulting in the release of the intron as a lariat, and the ligation of the 3’ end

of the 5’ exon to the 5’ end of the 3’ exon via a 5’-3’ phosphodiester bond, yielding the

spliced product.

2.3 Regulation of alternative splicing: 5' splice site and 3' splice site selection

Splicing factors and their binding sequences control alternative splice site selection.

SR proteins appear to play a central role in this regulation. SR proteins modulate the selection

of alternative splice sites in a concentration–dependent manner. In model pre-mRNAs

containing alternative 5' splice sites, higher concentrations of ASF/SF2 or SC35 promote

proximal 5' splice site usage (Ge and Manley, 1990; Krainer et al., 1990; Wang and Manley,

1995). Mechanistically this has been explained by a splice site occupancy model. Thus, at

- 21 -

high ASF/SF2 levels all competing 5' splice sites will be occupied by U1 snRNP. Under such

conditions the proximal splice site will be used due to its closeness to the 3’ splice site

(Eperon et al., 1993). ASF/SF2 has also been found to promote proximal 3' splice site usage

in vitro and in vivo (Bai et al., 1999; Fu et al., 1992). While other SR proteins, like SRp40 and

SRp55 have been found to promote distal 5' splice sites usage on some model transcripts

(Zahler et al., 1993; Zahler and Roth, 1995).

SR proteins function as both enhancer and repressor proteins of splicing (Kanopka et

al., 1996). Depending on where in the pre-mRNA they bind, SR proteins either stimulate or

interfere with the recruitment of other essential splicing factors to the adjacent splice site, and

to a large extent, decide whether or not the particular splice site will be selected by the

splicing machinery.

Other protein factors which can influence the binding of essential splicing factors to

the pre-mRNA also regulate splice site selection. For example, hnRNP A1 can modulate

splice site choice in a manner opposite to ASF/SF2, promoting selection of distal 5’ or 3'

splice sites (Bai et al., 1999; Caceres et al., 1994). The polypyrimidine tract binding protein

(PTB) and the Drosophila Sxl protein can compete with U2AF binding to a polypyrimidine

tract. Thereby, inhibiting 3' splice site selection by reducing U2 snRNP recruitment (Chan and

Black, 1997; Valcarcel et al., 1993).

2.3.1 Positive and negative cis-regulatory elements;

Splicing enhancer and splicing repressor elements mediate the interaction between

splicing factors and splice sites during spliceosome assembly. They are major cis-acting

elements determining splice site selection in alternative pre-mRNA splicing.

2.3.1.1 Splicing enhancer elements

Splicing enhancers are typically SR protein binding sequence elements that promote

the use of adjacent splice sites. Most of these sequences are composed of purine-rich

sequences embedded within exons, so called exonic splicing enhancers (ESEs). They are

often found downstream of introns that are considered to be weak (typically by a 5' or 3'

splice site that is a poor match to the consensus) (Manley and Tacke, 1996), and that are

subjected to alternative splicing. Members of the SR family of splicing factors have been

directly implicated in ESE function. By specifically interacting with SR proteins, ESEs aid in

the recruitment of other splicing factors to the nearby splice sites through protein-protein

interactions. This stabilized binding of splicing factors in turn promotes assembly of

- 22 -

spliceosomal complexes (Hoffman and Grabowski, 1992; Staknis and Reed, 1994; Wang and

Manley, 1995), and results in the selection of a relevant splice site by the splicing machinery.

Some ESEs appear also to act late during the splicing reaction, together with appropriate SR

proteins, to enhance the second catalytic step of splicing (Chew et al., 1999)

Different members of the SR protein family bind to distinct RNA sequence elements

and vary in their ability to commit specific pre-mRNAs to splicing. The effectiveness of

individual RNA binding sequences in promoting splice site activation correlate with their

binding affinity for SR proteins.

A considerable amount of data supports the conclusion that RNA sequences that form

both specific, high- affinity binding sites and degenerate, lower-affinity binding sites are

sufficient to function as exonic SR protein splicing enhancers (Sun et al., 1993; Watakabe et

al., 1993). Thus SR proteins binding to splicing enhancer sequences generally promote usage

of the adjacent splice sites in alternative splicing.

2.3.1.2 Splicing repressor elements

Splicing repressor, or silencer, elements are another type of regulatory elements

present in some pre-mRNAs. They inhibit splicing by interfering with recruitment of key

factors required for spliceosome assembly. One well-studied example is the regulation of

alternative splicing of the α-actinin gene. The polypyrimidine tract upstream of the SM exon

in the α-actinin gene functions as a repressor element by specifically binding PTB. Binding of

PTB to this repressor sequence excludes U2AF interaction with the polypyrimidine tract, and

thus represses 3' splice site usage. PTB has also been shown to repress α-tropomyosin and β-

tropomyosin pre-mRNA splicing (Gooding et al., 1998; Mulligan et al., 1992; Southby et al.,

1999). The Drosophila Sxl protein represents another example of a polypyrimidine tract

binding protein, similar to PTB, it inhibits splicing by competing with U2AF65 for binding to

the polypyrimidine tract on some pre-mRNAs involved in the maintain of sex in the fruit fly

(Granadino et al., 1997).

The abnormal positioning of snRNP binding sequences in pre-mRNAs can also resultin splicing repression. Thus, it has been shown that a sequence element upstream of thebranch point of the gag gene of Rous sarcoma virus (RSV) interacts with U11, U12, U1, U2snRNPs and represses splicing (Gontarek et al., 1993; McNally and Beemon, 1992; McNallyet al., 1991). A proposed mechanism for this repression is that recruitment of snRNPs to thewrong position in the pre-mRNA interferes with the normal spliceosome assembly process,and prevents the flanking splice sites being selected by the splicing machinery.

- 23 -

Furthermore, the results from our group have shown that a purine-rich sequence

element located just upstream of the branch site of adenovirus IIIa pre-mRNA represses IIIa

splicing in HeLa nuclear extracts, hence the name “the IIIa repressor element (3RE)”

(Kanopka et al., 1996). The 3RE binds all of the classical SR proteins found in HeLa cells. By

associating with SR proteins, the 3RE inhibits IIIa 3' splice site usage by interfering with

recruitment of U2 snRNP to the branch site, at least in part through steric hindrance.

Interestingly, substituting the 3RE with other SR protein binding sequences like ASF/SF2

consensus binding sites (Tacke and Manley, 1995), which function as splicing enhancer

elements when positioned in the downstream exon, results in ASF/SF2 dependent splicing

repression (Kanopka et al., 1996). Conversely, 3RE functions as a splicing enhancer when

moved to the downstream exon (Kanopka et al., 1996). These findings indicate that, not only

can SR protein binding sites enhance pre-mRNA splicing, but they can also inhibit splicing,

depending on where they are located in the pre-mRNA.

Some exons contain exonic splicing silencers. Their activities are usually balanced by

that of splicing enhancers. This double control mechanism ensures correct relative levels to be

reached for alternatively spliced mRNAs. HIV-1 tat exon 2 and K-SAM exon in human

fibroblast growth factor receptor 2 have been shown to contain hnRNP A1 binding sequences,

which function as splicing silencers. These silencers repress splicing by recruiting hnRNP A1

to the pre-mRNAs, and interfere with spliceosome assembly (Del Gatto-Konczak et al.,

1999).

In general, inappropriately positioned binding sites for essential splicing factors or

sequences binding specific inhibitory proteins function as splicing repressors.

2.3.2 The role of SR protein phosphorylation for splicing catalysis and alternative

splicing

Reversible phosphorylation of SR proteins is crucial in constitutive splicing, and also

plays an important role in alternative splice site selection. Phosphorylation significantly

increases the RNA binding specificity and affinity of SR proteins. At the early stage of

spliceosome assembly, this increased affinity enables SR proteins to bind to the 5' splice site

and facilitates U1 snRNP recruitment. During the transition of the early spliceosome to the

mature spliceosome, dephosphorylation of SR proteins may contribute to the release of U1

snRNP from 5' splice site, and facilitates the dynamic rearrangements of snRNPs pairing with

pre-mRNA inside the spliceosome (Cao et al., 1997; Misteli, 1999; Xiao and Manley, 1998).

- 24 -

Protein phosphorylation also increases the affinity of SR proteins for splicing

enhancer and repressor elements. This strategy turns out to be an important mechanism

involved in the regulation of alternative splice site selection in the adenovirus system. Thus,

late during an adenovirus infection SR proteins become functionally inactivated by a virus-

mediated dephosphorylation (Kanopka et al., 1998), resulting in the loss of their activities as

enhancer or repressor on splicing. In adenovirus L1 splicing, dephosphorylation of SR

proteins reduces their binding affinity for the 3RE, alleviating their repressive effect on IIIa

pre-mRNA splicing. As a result, IIIa splicing is dramatically enhanced.

3. ADENOVIRUS—A MODEL SYSTEM FOR

MECHANICSTIC STUDIES OF RNA SPLICING

3.1 Background

In this study, I have used adenovirus as a model system to investigate mechanisms

regulating pre-mRNA splicing. In general, viruses provide a very good tool for the study of

gene expression and its regulation. Thus, viruses are small, easy to manipulate and rely to a

large extent on the biosynthetic machinery provided by the host cell. Viral genomes are

adapted to be recognized and processed by the cellular machinery. Viruses encode for a few

regulatory proteins that interfere with key cellular factors regulating the cell cycle. Over the

years the significance of many cellular proteins have been discovered through their interaction

with viral proteins. For example, p53 was originally discovered because of its association

with the SV40 large T antigen (Lane and Crawford, 1979; Linzer and Levine, 1979).

Similarly, the function of the retinoblastoma tumor suppresser protein as a cell cycle regulator

was discovered through its interaction with the adenovirus E1A protein (Whyte et al., 1988).

It is likely that future studies will unravel additional mechanisms. In essence viruses have

been, and probably will continue to be valuable tools in genetic research.

3.2 The adenovirus system

Human adenovirus was first isolated from adenoid cell cultures in 1953 (Rowe et al.,

1953). Up to now, about 50 human adenovirus serotypes have been isolated (Horwitz, 1996).

They are divided into six subgroups (group A to F) based on hemagglutination, oncogenicity

in rodents and DNA sequence homology. Among them, Ad2 and Ad5 have been most

extensively studied. The whole genome sequence of several serotypes have been determined,

- 25 -

like type 2, 5, and 12 (Chroboczek et al., 1992; Roberts et al., 1986; Sprengler et al., 1995).

Adenoviruses are common pathogens, and cause primarily respiratory, ocular and

gastrointestinal diseases in humans, mostly in children.

The adenoviral virion is a non-enveloped, icosahedral particle, 70-100 nm in diameter.

It contains a liner, double-stranded DNA genome varying in length from 30,000 to 46,000

base pairs, depending on the serotype.(Fig. 7) The capsid is composed of 252 capsomers: 240

hexons and 12 pentons, which are buikt up by the penton base and the fiber. By attaching, via

the fibers, to the CAR receptor at the surface of the host cell, the virus enters the cell through

endocytosis. The capsid dissociates during passage from the endosome to the nuclear pore and

the viral DNA is injected into the cell nucleus. The lytic infection is divided into two phases,

the early and the late phase, which are separated by the onset of viral DNA replication which

start around 6-8 hours post infection. The virus life cycle lasts for 32-36 hours, and

approximately 10,000—100,000 progeny virions are released from each infected cell.

3.2.1 The early genes

The viral genome is organized into 9 transcription units, 5 early and 4 late. The early

region 1 (E1) is located at the 5’ end of the viral genome. E1A is the first region to be

transcribed during an adenovirus infection. The proteins encoded by E1A are multifunctional,

and in general encode for transcriptional regulatory proteins necessary for activation of other

viral transcription units. They stimulate and repress transcription of viral and cellular genes,

induce cell cycle progression, suppress differentiation, and mediate cellular transformation.

The proteins expressed from the E1B region inhibit p53-induced apoptosis (Debbas and

White, 1993). E1B proteins are also involved in the shut-off of host cell protein synthesis,

primarily by enhancing transport of viral mRNAs and blocking transport of host mRNAs

during a lytic viral infection (Pilder et al., 1986).

The E2 region encodes for the viral proteins required for viral DNA replication.

Transcription from this region is temporally regulated during the infection at the level of

promoter usage. At the early stage of infection, transcription initiates from an early promoter

and at late times of infection a second, late specific promoter is used. Furthermore, two

alternative poly-A sites are used to produce E2 mRNAs: thus, generating the E2A and E2B

mRNA families. E2A encodes for the single-stranded DNA binding protein (DBP) whereas

E2B encodes for the virus specific DNA polymerase, and the preterminal protein (pTP) which

functions as a primer protein for viral DNA replication (Chen et al., 1990).

- 26 -

Figure 7. Schematic presentation of transcription units and mRNAs encoded by the

adenovirus -2 genome. The physical locations of the early (dark thick arrows) and late

(light thick arrows) mRNA species and encoded proteins are indicated. This figure was

kindly provided by Dr. Göran Akusjarvi.

- 27 -

The proteins encoded by the E3 region antagonizes the host cell immune response

(Gooding et al., 1990; Wold et al., 1995). For example, the E3-19K protein prevents newly

synthesized major histocompatibility complex (MHC) class I antigen from being transported

to the cell surface, and this in turn help the infected cells escape cytotoxic T-cell recognition

(Wold et al., 1995). Through alternative splicing and alternative poly (A) site usage, at least

nine different mRNAs are expressed from E3. This region is dispensable for virus growth in

tissue culture cells.

The proteins expressed from the E4 region regulate transcription and RNA splicing.

Twenty-four alternatively spliced mRNAs are produced from this region. These mRNAs

encode at least seven different proteins (Freyer et al., 1984; Tigges and Raskas, 1984;

Virtanen et al., 1984). The proteins expressed from E4 are required for stable nuclear

accumulation of mRNAs derived from the major late transcription unit (Imperiale et al.,

1995).

Two proteins encoded by this region, E4-ORF3 and E4-ORF6, are essential for

efficient virus growth. A virus lacking both proteins show defects in viral DNA synthesis, late

viral mRNA accumulation and late viral protein synthesis, and a failure to shut-off host cell

protein synthesis (Bridge and Ketner, 1989; Halbert et al., 1985; Weinberg and Ketner, 1986).

Interestingly, E4-ORF3 and E4-ORF6 appear to perform redundant functions. Thus, mutant

viruses expressing either E4-ORF3 or E4-ORF6 can establish an essential wild-type virus

infection (Bridge and Ketner, 1989; Hemström et al., 1988; Huang and Hearing, 1989; Ketner

et al., 1989).

The E4-ORF3 and E4-ORF6 proteins are viral splicing factors. They regulate major

late tripartite leader assembly, and have opposite effects on accumulation of alternatively

spliced mRNAs (Nordqvist et al., 1994; Ohman et al., 1993). In transient transfection assays,

E4-ORF3 facilitates i-leader exon inclusion, whereas E4-ORF6 functions as an exon skipping

activity, enhancing the tripartite leader assembly. Furthermore, the E4-ORF6 protein also

stimulates i-leader exon skipping during lytic virus growth. The activities of these proteins are

not limited to mRNAs expressed from the major late transcription unit, since they show the

same effect on non-viral chimeric β-globin transcripts (Nordqvist et al., 1994), and similar

effects on transcripts from the early E1B transcription unit. These results suggest that E4-

ORF3 and E4-ORF6 proteins may be of global importance for the accumulation of

alternatively spliced mRNA from both the early and late viral transcription units (Ohman,

1995).

- 28 -

In addition, the E4ORF4 protein has been shown to regulate both transcription and

splicing. It binds to the cellular protein phosphatase 2A (PP2A). This complex is believed to

down regulate transcription through dephosphorylation of some transcription factors

(Bondesson et al., 1996; Kleinberger and Shenk, 1993; Mannervik et al., 1999; Muller et al.,

1992), and to regulate adenovirus alternative splicing by dephosphorylation of SR proteins

(Kanopka et al., 1998).

3.2.2 The major late transcription unit (MLTU)

The major late transcription unit generates a primary transcript of approximately

28,000 nucleotides that is processed into about 20 cytoplasmic mRNAs. These mRNAs are

grouped into five families (L1-L5, Fig. 7), where each family consists of multiple,

alternatively spliced species with co-terminal 3' ends. All mRNAs expressed from the MLTU

have a common 201-nucleotide tripartite leader sequence at their 5' end. A variant form of this

leader contains the 440-nucleotide i-leader exon. Splicing of the i-leader is temporally

regulated during infection. Thus, splicing of major late mRNA at early times of infection

usually leads to the inclusion of the i-leader exon (leader 1-2-i-3), but it is excluded at the late

times of infection in the majority of mRNAs (leader 1-2-3). The biological significance of the

i-leader exon inclusion/exclusion reaction is not clear. However, it has been shown that the i-

leader encodes for a 16 kDa protein (Virtanen et al., 1982), of unknown function. In general,

genes expressed from the MLTU encode for proteins necessary for viral progeny formation.

The expression of the mRNAs from MLTU is regulated at multiple levels during the

virus life cycle. Early after infection the major late promoter is active at a level comparable to

the early transcription units (Nevins and Wilson, 1981). But the majority of RNA

polymerases initiating at the major late promoter are prematurely terminated prior to the L2

poly(A) site. Only mRNAs from region L1 accumulate at this stage (Larsson et al., 1992).

After the onset of the late phase, the level of the mRNAs from MLTU increases dramatically.

One reason is probably the dramatic increase in template concentration caused by viral DNA

replication. At this stage, the transcription termination block is alleviated, and RNA

polymerase transcription extend through the entire late region to the end of the genome,

resulting in the production of the L1 to L5 family of mRNAs.

The accumulation of MLTU mRNAs is also regulated at the level of efficiency of poly

(A) site selection and alternative splicing of mRNAs within each late family (Akusjärvi and

Persson, 1981; Falck and Logan, 1989).

This work has focused on the L1 region (Fig. 8), which is the only part of the MLTU

- 29 -

that is expressed at both early and late times of infection (Nevins and Wilson, 1981). The L1

unit encodes for two major alternatively spliced mRNAs: the 52,55K, and the IIIa mRNAs.

The 52,55K mRNA encodes two related proteins of size 52 kDa and 55 kDa, which represent

differentially phosphorylated forms of a single 48-kDa polypeptide. The 52,55K protein is

required for viral DNA encapsidation, and also stimulates viral DNA replication, late protein

synthesis and virion assembly (Gustin and Imperiale, 1998; Hasson et al., 1992). The IIIa

mRNA encodes for a phosphoprotein, with an estimated molecular weight of 66 kDa. IIIa is a

structural protein, located in the vertex region of the capsid (Lemay et al., 1980). Splicing of

the 52,55K, and the IIIa mRNAs uses a common 5' splice site and two different 3' splice sites.

The 3' splice site of IIIa is located at 1266 nucleotides downstream of the 52,55K 3' splice

site.

L1 alternative splicing is strictly regulated during the infectious cycle, such that the

52,55K mRNA is spliced at all times of infection, whereas IIIa mRNA splicing is confined to

the late phase of infection. The study in this thesis has focused on the temporal regulation of

L1 alternative splicing. Previous work in our group had shown that: (1), the order of 3' splice

site presentation is important for the outcome of alternative L1 pre-mRNA splicing (Kreivi et

al., 1991); (2), Viral DNA replication and late protein synthesis play important roles in the

control of L1 alternative splicing (Larsson et al., 1991); (3), the IIIa repressor element, (3RE),

an SR protein binding sequence, located upstream of IIIa branch site inhibits IIIa splicing at

GOGTG GAGGA ATATGAC GAGGA CGATGAGTACGAGCCA GAGGA CGGCGAGTACT AAGCGGTGATGTTTCTGATCAG

3' ssEXON2

��

Figure 8. The L1 region sequence feature and splicing pattern inthe early and late phase. The lower panel shows the sequence ofthe IIIa 3' splice site containing two regulatory elements requiredfor IIIa splicing, the 3RE and the 3VDE.

1 2 3 IIIa52,55K5' ss 3' ss

Early splicing pattern

Late splicing pattern

3' ss

3RE 3VDE

bp

- 30 -

the early stage of infection (Kanopka et al., 1996); (4), during virus infection SR proteins

become functionally inactivated as IIIa splicing repressor proteins through a virus induced

dephosphorylation, hence an increase in IIIa splicing (Kanopka et al., 1998).

4. PRESENT INVESTIGATION

4.1 A downstream splicing enhancer is essential for splicing of all types of introns

(paper I)

Splicing enhancers are known to be required for the removal of introns with weak and

regulated splice sites. This type of introns contain short, frequently interrupted

polypyrimidine tracts, which are generally not consistent with the consensus sequence, and

therefore binds U2AF with a low affinity. In this context, SR proteins function as splicing

enhancer proteins by binding to the downstream exon and stabilizing the interaction of U2AF

with the polypyrimidine tract. Consequently the splicing machinery is able to select the weak

3’ splice site for spliceosome assembly. In the first part of paper I, we addressed the question

whether a splicing enhancer is necessary for splicing of introns with strong 3' splice sites.

In this experiment, two constitutively active pre-mRNAs, the adenovirus 52,55K and

Drosophila fushi tarazu (Ftz) were chosen as model pre-mRNAs. The introns in these two

pre-mRNAs contain so called “strong” 3' splice sites. To study the significance of a

downstream splicing enhancer for 52,55K splicing, three different sequences were used to

replace the second exon of the 52,55K pre-mRNA. These sequences were: (i,) the 49

nucleotide 3RE, which binds all the classical SR proteins; (ii,) a 28 nucleotide duplicated

ASF/SF2 binding splicing enhancer (2ASF); (iii,) a 49 nucleotide sequence from rabbit β-

globin gene that does not bind any SR proteins, we refer to this as an enhancer minus

sequence (E-). Three kinds of 52,55K pre-mRNA variants were created by this approach: the

52,55K(3RE), 52,55K(2ASF) and the 52,55K(E-). Their splicing properties were compared to

the wild-type 52,55K pre-mRNA by in vitro splicing in HeLa-NE (see paper I Fig.2).

To our surprise, replacing the second exon of 52,55K with the enhancer minus

sequence completely abolished splicing. At the same time 52,55K(3RE), 52,55K(2ASF)

transcripts were spliced as efficiently as the wild type 52,55K pre-mRNA. Based on this

finding we concluded that the downstream exon of 52,55K pre-mRNA must contain a splicing

enhancer element, which is required for 52,55K splicing. Since the 52,55K second exon can

be substituted by SR protein binding sequences, we speculated that the 52,55K second exon

- 31 -

might function through a similar mechanism, i.e. by binding SR proteins, and promoting

U2AF recruitment to polypyrimidine tract of 52,55K pre-mRNA. Collectively, these results

indicated that even a constitutively active intron with a strong 3' splice site like 52,55K was

absolutely dependent on a downstream splicing enhancer element for activity.

To determine whether the requirement of a downstream splicing enhancer was unique

to the adenovirus 52,55K pre-mRNA, we selected a second constitutively active pre-mRNA,

the Drosophila Ftz pre-mRNA to test its splicing enhancer dependency. The splicing of the

wild-type Ftz intron is very efficient with a conversion of almost 50% of input pre-mRNA to

spliced product after 90 minutes incubation under standard splicing conditions. Replacing the

downstream exon of the Ftz pre-mRNA with the ß-globin enhancer minus sequence (Ftz(E-)),

resulted in a complete loss of Ftz pre-mRNA splicing in vitro. This result further confirmed

our hypothesis that a downstream splicing enhancer element is essential for splicing of

“strong” contitutively active introns under in vitro splicing conditions.

To test if the rabbit β-globin enhancer minus sequence was inhibitory for splicing in

vitro, we attached an U1 splicing enhancer to the 3' end of 52,55K(E-), and Ftz(E

-)

transcripts. The splicing activity of the resulting pre-mRNAs, 52,55K(E-)-U1, and Ftz(E

-)-

U1, was restored to the same level as their wild-type counterparts. These results demonstrate

that the rabbit β-globin enhancer minus sequence functions as a neutral sequence and was not

inhibitory for splicing. Thus, the loss of splicing activity in transcripts of 52,55K(E-), and

Ftz(E-) was due to the absence of a downstream splicing enhancer element in these

transcripts.

Collectively our results extended previous observations by demonstrating that the

requirement of a downstream splicing enhancer is not restricted to splicing of weak and

regulated introns, but also essential for the splicing of strong and constitutively active introns.

4.2 A U1 splicing enhancer is more versatile than an SR splicing enhancer in promoting

pre-mRNA splicing: comparison of the properties of two types of splicing enhancers

(paper I)

In the second part of paper I, we studied the properties of two different types of

splicing enhancers for their ability to promote removal of an upstream intron.

In order to characterize the factors required for activation of IIIa-U1 splicing, we

separated HeLa-NE into three fractions of 40, 60, 90 ASP (the pellet precipitated from 40%,

- 32 -

60%, and 90% ammonium sulfate saturation) by stepwise precipitation with increasing

amounts of ammonium sulfate as described (Zahler et al., 1992). In a splicing assay, none of

these fractions were sufficient to activate IIIa pre-mRNA splicing. However, addition of 40

ASP, but not 60 ASP or 90 ASP, activated III-U1 splicing when supplemented with splicing

deficient S100 extracts. Since most SR proteins are enriched in the 90 ASP fraction, the result

excludes the possibility that SR proteins are the factors stimulating IIIa-U1 splicing. This

result is consistent with our previous finding that SR proteins function as repressor, not

activator proteins of IIIa pre-mRNA splicing (Kanopka et al., 1996).

By Western blot analysis, we demonstrated that U1 snRNP was selectively enriched in

40 ASP. To further test whether U1-snRNP was the enhancer factor in 40 ASP, required for

IIIa-U1 splicing, we functionally inactivated U1 snRNA by oligonucleotide-directed RNase H

depletion. The oligonucleotide used in this experiment was designed to hybridize to the 5' end

of U1 snRNA, which is the region that pairs with the 5' splice site of the pre-mRNA during 5'

splice site recognition (Black et al., 1985). RNase H cleavage in the presence of the U1

oligonucleotide resulted in the functional depletion of U1 snRNPs from the 40 ASP fraction.

The result further showed that incubation of 40 ASP with increasing amount of an U1

oligonucleotide during the RNase H treatment abolished the stimulatory activity of 40 ASP on

IIIa-U1 splicing, while pre-treatment of 40 ASP with an U2 snRNA specific oligonucleotide

had no effect on 40 ASP activation of IIIa-U1 splicing. From these results, we concluded that

U1 snRNP was the critical factor in 40 ASP stimulating III-U1 splicing.

Cytoplasmic S-100 extracts prepared from HeLa cells are inactive in splicing, either in

the presence of an U1 enhancer or an SR enhancer, suggesting that S-100 extracts contain

sub-optimal concentrations of both SR proteins and U1 snRNP. We found that 52,55K(E-)-U1

is activated in S-100 extracts supplemented either with 40 ASP or with purified HeLa SR

proteins. In contrast, 52,55K(3RE), the variant of 52,55K pre-mRNA with a second exon SR

enhancer was only activated in S-100 supplemented with SR proteins. This result implies that

two types of factors, U1 snRNPs or SR proteins can activate the U1 enhancer, while the SR

enhancer is only activated by SR proteins (Fig. 8). Most likely, this difference in activation

capacity results from the fact that an U1 enhancer can bind both U1 snRNP and SR proteins

(Crispino et al., 1994; Jamison et al., 1995) , while an SR enhancer can only bind SR proteins,

not U1 snRNP. The binding of either of these factors stabilize U2AF interaction with the

polypyrimidine tract, thus increasing splice site recognition and spliceosome assembly. In

- 33 -

other words, the U1 enhancer is more versatile than the SR enhancer in stimulating pre-

mRNA splicing.(Fig. 9)

4.3 Identification of a virus infection dependent splicing enhancer, the 3VDE (paper II)

The adenovirus IIIa pre-mRNA is spliced inefficiently in HeLa-NE. In contrast, IIIa

splicing is activated more than 200-fold in nuclear extract prepared from late adenovirus –

infected cells (Ad-NE). This finding implies that the IIIa pre-mRNA per se, contains an

endogenous splicing enhancer, which has evolved to function only in the context of an

adenovirus infection. We named this splicing enhancer the IIIa virus infection-dependent

splicing enhancer (3VDE). In paper II, we report the identification and functional

characterization of this novel type of splicing enhancer.

The rabbit ß-globin pre-mRNA is a constitutively active pre-mRNA. It is spliced

efficiently in HeLa-NE, but is slightly inhibited in Ad-NE; i.e. its splicing tendency is

Yn

A. Unstable U2AF binding topolypyrimidine tract ( Yn) .

Yn ESE

SR

�

B. ESE binds SR proteins. The interactionbetween SR proetins and U2AF stabilisesU2AF binding to the polypyrimidine tract.

Yn

SR

D. U1 enhancer can also bind SRproteins, which helps stabilise U2AFbinding to the polypyrimidine tract.

U1Yn

SR

7 0 KU1

snRNP

?

Figure 9. Schematic representation of possible mechanisms by whichan SR enhancer and a U1 enhancer promote splicing. 3' splice siteusage is triggered by U2AF binding to the polypyrimidine tract.

U1� �

�

C. U1 enhancer binds U1 snRNP,which helps stabilize U2AF bindingto the polypyrimidine tract.

U2AF U2AF

U2AFU2AF

- 34 -

opposite to that of the IIIa pre-mRNA. In our effort to identify the IIIa sequence elements

responsible for activated splicing in Ad-NE, we made chimeric transcripts by exchanging

sequence elements between the IIIa and the ß-globin transcripts. The splicing properties of

these hybrid pre-mRNAs were tested in standard in vitro splicing reactions in HeLa-NE and

Ad-NE.

In the first set of experiments, we replaced short sequences in the rabbit ß-globin pre-

mRNA with the corresponding sequences from IIIa, and tried to identify the minimal IIIa

element conferring an enhanced splicing phenotype to ß-globin in Ad-NE.

Transferring the 3RE to the ß-globin transcript resulted in a 5-fold reduction in the

splicing efficiency in HeLa-NE. This result is consistent with our previous work

demonstrating that the 3RE functions as a repressor element of IIIa splicing (Kanopka et al.,

1996). Interestingly, splicing of this pre-mRNA was still inhibited in Ad-NE, suggesting that

inactivation of SR protein binding to the 3RE is not sufficient to explain the enhanced

splicing phenotype of the IIIa pre-mRNA in Ad-NE.

Next we analyzed the splicing phenotype of a ß-globin transcript containing the 28 nt

encoding the IIIa branch site and polypyrimidine tract. In HeLa-NE, this sequence reduced ß-

globin splicing more than the 3RE. However, splicing of this transcript was enhanced about 5-

fold in Ad-NE compared to HeLa-NE, reaching approximately 70% of the splicing efficiency

of the wild type IIIa pre-mRNA. This result strongly suggested that the last 28 nt sequence of

IIIa intron encodes the critical IIIa sequence element, which when transferred to the ß-globin

transcript, was sufficient to confer an enhanced splicing phenotype to this pre-mRNA in Ad-

NE. We designated this 28 nt sequence the 3VDE (IIIa virus infection-dependent splicing

enhancer).

In order to test the contribution of the 3RE and the 3VDE on enhanced IIIa pre-mRNA

splicing in Ad-NE, we substituted sequence elements in the IIIa pre-mRNA with the

corresponding sequences from ß-globin (paper II fig. 6A). The results from in vitro splicing

assays showed that the basal level of IIIa(-3VDE) splicing in HeLa-NE was significantly

increased compared to the wild-type IIIa transcript, but that splicing activation was moderate

in Ad-NE (Paper II fig 6B). The small residual activation could be attributed to the reduced

binding of SR protein to the 3RE in Ad-NE. Collectively, our results show that the 3VDE is

the key element controlling IIIa 3' splice site activity, with the 3RE making a smaller, but

important contribution.

In an attempt to dissect further the 3VDE, we replaced the ß-globin branch site and

polypyrimidine tract with the corresponding sequences from IIIa; this resulted in transcripts of

- 35 -

glob(IIIa-bp) and glob(IIIa-py), respectively. As shown in paper II Figure. 3, splicing of

glob(IIIa-bp) in Ad-NE was not enhanced compared to HeLa-NE, indicating that the IIIa

branch site is not the critical element conferring an enhanced splicing phenotype to the IIIa

pre-mRNA in Ad-NE. The splicing of glob(IIIa-py) transcript was not detectable in HeLa-NE,

whereas a weak signal was consistently observed in Ad-NE. However, formation of the pre-

spliceosome (the A complex) was dramatically increased in Ad-NE compared to HeLa-NE,

suggesting that the IIIa polypyrimidine tract is, indeed, the primary element responsible for

enhanced IIIa 3' splice site recognition in Ad-NE, but that steps subsequent to A complex

formation are very inefficient with the glob(IIIa-py) transcript. Collectively, our results

suggest that: (i,) The splicing activity of transcripts in HeLa-NE is controlled by the

polypyrimidine tract strength; (ii,) The suboptimal IIIa polypyrimidine tract plays a central

role in enhanced splicing in Ad-NE by promoting efficient initiation of spliceosome

assembly; and (iii,) The integrity of the 3VDE (i.e. the native IIIa branch site and IIIa

polypyrimidine tract) is required for efficient splicing catalysis.

4.4 U2AF binding to the IIIa polypyrimidine tract does not correlate with IIIa splicing

activation in Ad-NE (paper II)

As discussed above, U2AF binding to the polypyrimidine tract is critical for 3' splice

site activation. U2AF shows a strong binding affinity for polypyrimidines, the longer the

better. Increase in U2AF binding leads to higher splicing efficiency in HeLa-NE. However,

the results in Ad-NE is just opposite. Thus, using a recombinant Gst-U2AF65 protein it was

previously shown that pre-mRNAs which bind U2AF efficiently are repressed in Ad-NE,

while pre-mRNAs with atypical polypyrimidine tracts and lower U2AF binding efficiency,

like IIIa, are enhanced in Ad-NE (Mühlemann et al., 1995). In paper II, we used UV cross-

linking combined with immunoprecipitation (using an anti U2AF65 antibody) to study the

binding affinity of endogenous U2AF65 in HeLa-NE and Ad-NE to the IIIa and β-globin

polypyrimidine tracts.

As seen in the paper II, Figure 7B, U2AF65 binds the IIIa polypyrimidine tract weakly

in HeLa-NE, which is consistent with the low splicing activity of the IIIa pre-mRNA. In

contrast, the binding of U2AF65 to the IIIa polypyrimidine tract is not enhanced in Ad-NE,

even though the IIIa splicing efficiency is dramatically increased. This unexpected result

suggests that the binding of U2AF65 to 3VDE is not the key factor, which contributes to the

enhanced splicing phenotype of the IIIa pre-mRNA in Ad-NE.

- 36 -

4.5 Development of a new expression system for production of phosphorylated,

biologically active recombinant ASF/SF2 in E. coli (paper III)

The mammalian SR family of splicing factors is highly phosphorylated predominately

within the RS domain at physiological conditions in cells. This phosphorylation is essential

for spliceosome assembly by promoting protein-protein interactions between different SR

proteins, and SR proteins and other splicing factors. Furthermore, control of the

phosphorylated status of SR proteins has been suggested to be a key mechanism in the

regulation of alternative splicing.

Functional studies of SR proteins require large amounts of pure protein. Typically

recombinant proteins are expressed in E. coli since this system provides an easy and

economical way to obtain large amounts of protein. However, expression in E. coli of an SR

protein usually results in the production of a protein with a low biological activity since

bacteria lack protein kinases capable of phosphorylating SR proteins. To overcome this

problem, we established a new expression system where the SR protein was co-expressed

with a protein kinase capable of phosphorylating SR proteins, in our case SRPK1 (SR

proteins kinase 1). We selected ASF/SF2 as a model SR protein in this study.

To co-express ASF/SF2 and SRPK1 in the same bacteria, we used two expression

vectors encoding different antibiotic resistant markers. The full-length SRPK1 was cloned

into pLysS (chloramphenicol resistance) generating plasmid pLysS-SRPK1, and ASF/SF2

was cloned into pQE-60 and pRSET A (ampicillin resistance) generating plasmids pQE-

ASF(-HIS) and pRSET A-ASF, respectively. By electroporation, the two plasmids expressing

ASF/SF2 and SRPK1 respectively were transformed into E. coli BL21 (DE3). Double

transformants were selected by growing cells in the presence of ampicillin and

chloramphenicol. ASF/SF2, without the SRPK1 plasmid, was also transformed and expressed

in E. coli in parallel.

ASF/SF2, with or without SRPK1 co-transformation, were expressed and purified by

Ni-agarose chromatography, or by the conventional two-step high salt precipitation procedure

used to purify SR proteins from mammalian cells. The phosphorylated status was determined

by SDS-PAGE and Western blot analysis. As shown in paper III (fig. 2), ASF/SF2 co-

expressed with SRPK1 (designated as ASF-P) migrated slower than ASF/SF2 expressed

without SRPK1. Treatment of ASF-P with calf intestinal alkaline phosphotase (CIP) shifted

the migration of ASF-P to the same level as ASF/SF2, demonstrating that ASF-P, indeed, was

successfully phosphorylated in E. coli.

- 37 -

Previous studies have shown that the monoclonal antibody 104 (mAb104)

specifically recognizes phosphorylated epitopes in the RS domain of SR proteins. In a

Western blot analysis, ASF-P was recognized by mAb104, whereas ASF/SF2 was not. This

finding further supports the conclusion that ASF-P was not only phosphorylated, but also

phosphorylated at the proper region within the RS domain. More importantly, in vitro splicing

assays demonstrated that ASF-P was more effective, compared to ASF/SF2, in

complementing 52,55K pre-mRNA splicing in S100 extracts, indicating ASF-P is functionally

active in committing the pre-mRNA to the spliceosome assembly pathway.

Proteins expressed in E. coli are typically insoluble. The expressed proteins form

inclusion bodies, which are easy to purify, but difficult to refold to their native conformation.

Interestingly, coexpression of SRPK1 and ASF/SF2 makes the SR protein more soluble in E.

coli. This simplifies the purification procedure, and as a bonus, generates a more biologically

active protein.

We conclude that coexpression of a protein kinase with its substrate in E. coli is an

efficient approach to obtain the phosphorylated version of a mammalian protein. In our model

system, the expressed ASF/SF2 protein is phosphorylated, soluble, and more biologically

active compared to its un-phosphorylated counterpart.

5. CONCLUSIONS AND DISCUSSION 5.1 The universal role of splicing enhancers

It has long been noticed that splicing enhancers are required for the removal of

introns containing weak, degenerate splicing signals in alternative splicing. However, a

similar mechanism could also be utilized by cells to ensure accurate splice site recognition in

constitutively active pre-mRNAs. For example, exons from constitutively spliced pre-mRNAs

have been shown to associate with SR proteins (Chiara et al., 1996), which have been shown

to promote 5’ and 3’ splice site activity (Wang et al., 1995).

The result from the paper I demonstrates, for the first time, that two constitutively

active introns also require a splicing enhancer for activity. In agreement with our hypothesis

that splicing of constitutively active pre-mRNAs requires a splicing enhancer for activity,

multiple splicing enhancer sequences were recently found to be embedded within the protein-

coding sequences of the constitutively spliced human β-globin pre-mRNA (Schaal and

Maniatis, 1999). These enhancers were shown to be capable of activating the splicing of a

heterologous enhancer-dependent pre-mRNA in vitro. Collectively, available data suggests

- 38 -

that splicing enhancers may be required for splicing all types of introns. Our data further

suggests that a downstream 5’ splice site is the main type of splicing enhancer stimulating

upstream intron removal. Thus, SR splicing enhancers probably play a minor role in

constitutive splicing of internal exons.

From previous data, it is known that splicing enhancers are the main cis-acting

elements regulating alternative splicing. Because alternatively spliced introns usually have

weak splicing signals, this leaves a large room for splicing enhancers to function, such that

splicing or not splicing, to a large extent, is decided by the type of splicing enhancer present

and the activity of the factors interacting with the enhancer.

5.2 Models for the 3VDE function

The 3VDE is a unique element controlling 3' splice site activity. It functions as a

”Janus” element, and inhibits pre-mRNA splicing in HeLa-NE, while enhancing pre-mRNA

splicing selectively in Ad-NE. It comprises of the 3' splice site region of the IIIa pre-mRNA;

the core enhancer element consists of the IIIa polypyrimidine tract. Current data suggests that

the 3VDE functions without efficient U2AF recruitment, i.e. it binds U2AF65 poorly in

HeLa-NE, and even worse in Ad-NE. The steady–state levels of U2AF in HeLa-NE and Ad-

NE is essentially the same, while available U2AF for each pre-mRNA in Ad-NE may be

lower than in HeLa-NE due to the large amount of viral pre-mRNA produced late in

infection. However, it appears that U2AF is not the critical factor required for enhanced IIIa

splicing in Ad-NE.

The work to isolate the key factor, the 3VDF (IIIa virus infection dependent splicing

factor) involved in 3VDE function is still ongoing. The results so far does not allow us to

discriminate between the possibilities that the 3VDF behaves as a positive or a negative

regulator. Therefore, I propose two working models to explain the possible mechanisms for

the 3VDE function.

In model A, the 3VDF is a cellular negative regulator (potentially an hnRNP, like

hnRNP A1). In HeLa-NE, the 3VDF specifically binds to the 3VDE and prevents U2 snRNP

recruitment. In Ad-NE, the 3VDF may be modified by viral factor, or sequestrated by the

large amount of viral RNA expressed late in infection. This would be predicted to cause a

release of 3VDF repression, and as a consequence lead to enhanced IIIa pre-mRNA splicing.

Since the IIIa branch point shows a close match to the branch site consensus sequence, the

IIIa pre-mRNA may not require U2AF for U2 snRNP recruitment. Therefore, splicing can be

dramatically enhanced as long as repression is released. (Fig. 10A)

- 39 -

In model B, the 3VDF functions as a positive factor. The 3VDE represses IIIa pre-

mRNA splicing in HeLa-NE due to its low affinity for U2AF. In Ad-NE the 3VDF (a viral

factor or viral modified cellular factor) binds to 3VDE, and helps to recruit U2 snRNP into

spliceosome. Consequently, IIIa pre-mRNA splicing is enhanced. (Fig. 10B)

pybp pybp

Early Late

3RE 3VDE3RE 3VDE

U2AF

U2AF

U2 U2

Figure 10. Two models for 3VDE function. Both of them work by influencing U2 snRNPrecruitment to the branch site. In model A, the 3VDF works as a negative regulator. Itinhibits IIIa pre-mRNA splicing in HeLa-NE by blocking U2 snRNP binding to the branchsite. In model B, U2 snRNP recruitment is low because the IIIa polypyrimidine tract bindsU2AF inefficiently in HeLa-NE. U2 snRNP recruitment is enhanced in Ad-NE by 3VDF.The contribution of 3RE to IIIa splicing is omitted. See the text for detail.

pybp

SR

pybp

3RE 3VDE3RE 3VDE

U2AF

U2AF

U2AF

Early LateB

A

U 2 U2

- 40 -

5.3 Perspective for further identification of the 3VDF

Since IIIa pre-mRNA splicing is controlled by the interaction between the 3VDE and

the 3VDF, the purification and characterization of the 3VDF is a key step towards our

understanding of how IIIa pre-mRNA splicing is regulated. Besides UV-crosslinking using

site specifically labeled 3VDE RNA and nuclear extracts (virus infected and uninfected), the

following experiments should also be considered:

Large scale RNA affinity chromatography. If the 3VDE is the sequence for specific

3VDF binding, this factor could be isolated by incubating the factor containing extracts with

gel immobilized 3VDE RNA under binding conditions, and then washed and eluted by

changes of pH or salt strength.

Using modified stable RNA oligos complementary to parts of 3VDE and flanking

sequences, to stepwise block 3VDF binding may provide further insight for 3VDE and 3VDF

interactions and help to identify the consensus sequence for 3VDF binding. Similarly, point

mutations of 3VDE may also produce useful information concerning its binding specificity.

Our previous data show that mutating pyrimidines to purines in the core part of 3VDE abolish

IIIa splicing. This implies that U2AF may be required, since such mutations reduces U2AF

binding. A better experiment to exclude the requirement of U2AF would be to mutate purines

to pyrimidines. If such mutations also abolish or reduce IIIa pre-mRNA splicing, the

requirement for U2AF becomes less likely.

To test the requirement for U2AF for IIIa pre-mRNA splicing, the use of U2AF

depleted HeLa-NE and Ad-NE may also produce valuable data. The depleted extract can be

used in splicing complex assembly, in virto splicing assays, or to check U2 snRNP binding.

Even though 3VDF has some unique features, it should be tested whether 3VDF is

PTB or a homologue of PTB or Sxl. Experimentally this could be done by using antibodies,

depletion or addition of purified protein in splicing assays in HeLa-NE or Ad-NE.

If the 3VDF is a cellular negative regulator, the reason for IIIa enhanced splicing in

Ad-NE is sequestration of 3VDF by viral RNA or DNA. This process could be simulated in

vitro by addition of purifyed viral RNA or DNA to HeLa-NE before splicing assay.

5.4 The significance of modification of recombinant proteins expressed in E. coli and

perspective for future applications

Recombinant protein produced in bacteria is used in many areas of basic research as

well as in the clinic. The first reason for this situation is that the natural resources of proteins

- 41 -

are often limited and therefore difficult to purify in sufficient quantities, for example

hormones, various growth factors, antibodies etc. Another important reason is the high risk of

contamination when proteins are purified from their natural sources, like the contamination of

trace amount of HIV or HBV / HCV (hepatitis B and C virus) in blood products. Problems of

this type make researchers and medical companies interested in using heterologous systems

for recombinant protein production. E. coli has proven to be a valuable production host for

these purposes.

As an expression system, E. coli has many advantages. It grows fast in inexpensive

media, the expression procedure is easy to set up, and to scale up. Since expression vectors

designed for E. coli are commercially available with various tags, the production and

purification of recombinant proteins has become routine in many laboratories. Most

importantly, the risk of contamination of other mammalian proteins or human viruses is

eliminated by using a bacteria for production of a recombinant protein.

A disadvantage with protein expression in E. coli is that this bacteria lack many of the

posttranslational modifications used by higher eukaryotic cells, such as the mammalian

enzymes required for phosphorylation, glycosylation, methylation or acetylation etc. This

results sometimes in production of nonfunctional proteins when they are expressed in E. coli.

This problem seriously restricts the use of E. coli as an expression system. Here we

use a co-expression strategy to produce phosphorylated ASF/SF2 in E. coli. It is likely that

the same strategy may be used for expression of proteins with other post-translational

modifications, such as methylation, acetylation etc.

In the paper III we used two plasmids to co-express an SR protein kinase and its

substrate. A further advance of our technique would be to reconstruct a plasmid encoding

both the modifying enzyme and its substrate either from two independent promoters, or an

expression cassette where expression of the two proteins are controlled by a bicistronic

transcription unit. These modifications may solve the plasmid incompatibility problem, which

is a major problem restricting the successful co-expression of proteins in E. coli.

In theory, our co-expression strategy may be used to more precisely define the

significance of different protein modifications for function. For example, several protein

kinases have been shown to phosphorylate SR proteins. By co-expressing an SR protein with

the different SR protein kinases, the significance for the individual enzymes for spliceosome

assembly may be tested. Similarly, the same strategy may be adapted to studies of other

biosynthetic machinery’s in the cell, like transcription and translation

- 42 -

6. ACKNOWLEDGEMENT

This thesis is based on the studies started from the Department of Cellular and Molecular

Biology, at the Karolinska Institute in Stockholm, finished at the Department of Medical

Biochemistry and Microbiology, Uppsala University, Sweden.

I would like to take this opportunity to express my sincere gratitude to all the people who

have helped, and supported me in this work. In particular, I would like to thank:

Professor Göran Akusjärvi, my supervisor, for introducing me to the RNA world, for your

stimulating training, encouragement, enthusiasm and patience, for creating the enjoyable lab

atmosphere.

Professor Göran Magnusson, my examiner, and people in your group for help and scientific

discussion.

Professor Jerker Porath, my former supervisor in the Department of Biochemistry, for

accepting me to study in your lab when I first came to Sweden, for sharing your knowledge

and ideas. And people in your Lab.

Professor Qian Lin-Fa, my mentor, at Medical Science Institute, Nanjing Railway Medical

College, China, for your help and generous support.

All the members in Professor Stefan Schwartz’s group and in Dr. Lars Hellman’s group.

The present and former members in our group: Jan-Peter Kreivi, Oliver Muhlemann, Arvydas

Kanopka, Maria Bondesson, Catharina Svensson, Karin Öhman-Forslund, Kerstin

Sollerbrant, Petra Olsson, Mattias Mannervik, Neil Portwood, Svend Petersen-Mahrt,

Christina Öhrmalm, Ann-Christine Ström, Camilla Estmer Nilsson, Edyta Bajak, Vita

Dauksaite, Dan Eholm, Shao-an Fan, David Huang, Cecilia Johnsson, Anette Lindberg,

Magnus Molin, Tanel Punga, Jodef Seibt, Anders Sundqvist, Saideh Berenjian, Martin

Luezelberger, for excellent collaboration and sharing the good and bad times.

- 43 -

Among those, special thanks to:

-Catharina, for your help, and encouragement.

-Peter for your collaboration, guidance and patience in the phosphorylation project.

-Oliver for introducing me to the splicing area, for stimulating discussion in work.

-Christina and John, for introducing me a lot knowledge about Sweden, for always being

ready to help.

-Shao-an for sharing the same language and culture.

Brian Collier for correcting my English (not for this sentence).

My Chinese friends:

Wang Shu and Zhou Yinghua, for sharing opinions, experiences, and Rome holiday. Li

Jinping and Zhang Xiao for help in those years; Wei Tao, Liu Aijie and Huang Jinfang, for

passing the God’s blessing.

Zhang Hongxue for help and care during the early stage of this work.

I thank my parents for their endless understanding, encouragement, support and love; think

my sister Baili and her family, for their support and help, taking care of our parents when I am

working on this thesis; my daughter Yue Kun, for giving me so much happiness.

- 44 -

7. REFERENCES

Adams, M. D., Rudner, D. Z., and Rio, D. C. (1996). Biochemistry and regulation of pre-mRNA splicing. Curr Opin Cell Biol 8, 331-9.

Akusjärvi, G., and Persson, H. (1981). Controls of RNA splicing and termination in themajor late adenovirus transcription unit. Nature 292, 420-6.

Amrein, H., Hedley, M. L., and Maniatis, T. (1994). The role of specific protein-RNA andprotein-protein interactions in positive and negative control of pre-mRNA splicing byTransformer 2. Cell 76, 735-46.

Arning, S., Gruter, P., Bilbe, G., and Krämer, A. (1996). Mammalian splicing factor SF1 isencoded by variant cDNAs and binds to RNA. RNA 2, 794-810.

Bai, Y., Lee, D., Yu, T., and Chasin, L. A. (1999). Control of 3' splice site choice in vivo byASF/SF2 and hnRNP A1. Nucleic Acids Res 27, 1126-34.

Barabino, S. M., Blencowe, B. J., Ryder, U., Sproat, B. S., and Lamond, A. I. (1990).Targeted snRNP depletion reveals an additional role for mammalian U1 snRNP inspliceosome assembly. Cell 63, 293-302.

Berget, S. M., Moore, C., and Sharp, P. A. (1977). Spliced segments at the 5' terminus ofadenovirus 2 late mRNA. Proc Natl Acad Sci U S A 74, 3171-5.

Black, D. L., Chabot, B., and Steitz, J. A. (1985). U2 as well as U1 small nuclearribonucleoproteins are involved in premessenger RNA splicing. Cell 42, 737-50.

Black, D. L., and Steitz, J. A. (1986). Pre-mRNA splicing in vitro requires intact U4/U6small nuclear ribonucleoprotein. Cell 46, 697-704.

Blencowe, B. J., Bauren, G., Eldridge, A. G., Issner, R., Nickerson, J. A., Rosonina, E.,and Sharp, P. A. (2000). The SRm160/300 splicing coactivator subunits. RNA 6, 111-20.

Blencowe, B. J., Issner, R., Nickerson, J. A., and Sharp, P. A. (1998). A coactivator of pre-mRNA splicing. Genes Dev 12, 996-1009.

Bondesson, M., Öhman, K., Manervik, M., Fan, S., and Akusjärvi, G. (1996). AdenovirusE4 open reading frame 4 protein autoregulates E4 transcription by inhibiting E1Atransactivation of the E4 promoter. J Virol 70, 3844-51.

Bridge, E., and Ketner, G. (1989). Redundant control of adenovirus late gene expression byearly region 4. J Virol 63, 631-8.

Brosi, R., Groning, K., Behrens, S. E., Luhrmann, R., and Krämer, A. (1993). Interactionof mammalian splicing factor SF3a with U2 snRNP and relation of its 60-kD subunit to yeastPRP9. Science 262, 102-5.

Caceres, J. F., Misteli, T., Screaton, G. R., Spector, D. L., and Krainer, A. R. (1997).Role of the modular domains of SR proteins in subnuclear localization and alternativesplicing specificity. J Cell Biol 138, 225-38.

- 45 -

Caceres, J. F., Stamm, S., Helfman, D. M., and Krainer, A. R. (1994). Regulation ofalternative splicing in vivo by overexpression of antagonistic splicing factors. Science 265,1706-9.

Cao, W., Jamison, S. F., and Garcia-Blanco, M. A. (1997). Both phosphorylation anddephosphorylation of ASF/SF2 are required for pre-mRNA splicing in vitro. RNA 3, 1456-67.

Cavaloc, Y., Bourgeois, C. F., Kister, L., and Stevenin, J. (1999). The splicing factors 9G8and SRp20 transactivate splicing through different and specific enhancers. RNA 5, 468-83.

Chan, R. C., and Black, D. L. (1997). The polypyrimidine tract binding protein bindsupstream of neural cell-specific c-src exon N1 to repress the splicing of the introndownstream. Mol Cell Biol 17, 4667-76.

Chen, M., Mermod, N., and Horwitz, M. S. (1990). Protein-protein interactions betweenadenovirus DNA polymerase and nuclear factor I mediate formation of the DNA replicationpreinitiation complex. J Biol Chem 265, 18634-42.

Chew, S. L., Liu, H. X., Mayeda, A., and Krainer, A. R. (1999). Evidence for the functionof an exonic splicing enhancer after the first catalytic step of pre-mRNA splicing. Proc NatlAcad Sci U S A 96, 10655-60.

Chiara, M. D., Gozani, O., Bennett, M., Champion-Arnaud, P., Palandjian, L., andReed, R. (1996). Identification of proteins that interact with exon sequences, splice sites, andthe branchpoint sequence during each stage of spliceosome assembly. Mol Cell Biol 16,3317-26.

Cho, E. J., Takagi, T., Moore, C. R., and Buratowski, S. (1997). mRNA capping enzyme isrecruited to the transcription complex by phosphorylation of the RNA polymerase II carboxy-terminal domain. Genes Dev 11, 3319-26.

Chow, L. T., Gelinas, R. E., Broker, T. R., and Roberts, R. J. (1977). An amazingsequence arrangement at the 5' ends of adenovirus 2 messenger RNA. Cell 12, 1-8.

Chroboczek, J., Bieber, F., and Jacrot, B. (1992). The sequence of the genome ofadenovirus type-5 and its comparison with the genome of adenovirus type-2. Virology 186,280-285.

Colgan, D. F., and Manley, J. L. (1997). Mechanism and regulation of mRNApolyadenylation. Genes Dev 11, 2755-66.

Colwill, K., Feng, L. L., Yeakley, J. M., Gish, G. D., Caceres, J. F., Pawson, T., and Fu,X. D. (1996). SRPK1 and Clk/Sty protein kinases show distinct substrate specificities forserine/arginine-rich splicing factors. J Biol Chem 271, 24569-75.

Colwill, K., Pawson, T., Andrews, B., Prasad, J., Manley, J. L., Bell, J. C., and Duncan,P. I. (1996). The Clk/Sty protein kinase phosphorylates SR splicing factors and regulates theirintranuclear distribution. EMBO J 15, 265-75.

- 46 -

Coppola, J. A., Field, A. S., and Luse, D. S. (1983). Promoter-proximal pausing by RNApolymerase II in vitro: transcripts shorter than 20 nucleotides are not capped. Proc Natl AcadSci U S A 80, 1251-5.

Corden, J. L., and Patturajan, M. (1997). A CTD function linking transcription to splicing.Trends Biochem Sci 22, 413-6.

Coulter, L. R., Landree, M. A., and Cooper, T. A. (1997). Identification of a new class ofexonic splicing enhancers by in vivo selection. Mol Cell Biol 17, 2143-50.

Cramer, P., Caceres, J. F., Cazalla, D., Kadener, S., Muro, A. F., Baralle, F. E., andKornblihtt, A. R. (1999). Coupling of transcription with alternative splicing: RNA pol IIpromoters modulate SF2/ASF and 9G8 effects on an exonic splicing enhancer. Mol Cell 4,251-8.

Cramer, P., Pesce, C. G., Baralle, F. E., and Kornblihtt, A. R. (1997). Functionalassociation between promoter structure and transcript alternative splicing. Proc Natl Acad SciU S A 94, 11456-60.

Crispino, J. D., Blencowe, B. J., and Sharp, P. A. (1994). Complementation by SR proteinsof pre-mRNA splicing reactions depleted of U1 snRNP. Science 265, 1866-9.

Dantonel, J. C., Murthy, K. G., Manley, J. L., and Tora, L. (1997). Transcription factorTFIID recruits factor CPSF for formation of 3' end of mRNA. Nature 389, 399-402.

Debbas, M., and White, E. (1993). Wild-type p53 mediates apoptosis by E1A, which isinhibited by E1B. Genes Dev 7, 546-54.

Del Gatto-Konczak, F., Olive, M., Gesnel, M. C., and Breathnach, R. (1999). hnRNP A1recruited to an exon in vivo can function as an exon splicing silencer. Mol Cell Biol 19, 251-60.

Eperon, I. C., Ireland, D. C., Smith, R. A., Mayeda, A., and Krainer, A. R. (1993).Pathways for selection of 5' splice sites by U1 snRNPs and SF2/ASF. EMBO J 12, 3607-17.

Falck, P. E., and Logan, J. (1989). Regulation of poly(A) site selection in adenovirus. JVirol 63, 532-41.

Freyer, G. A., Katoh, Y., and Roberts, R. J. (1984). Characterization of the major mRNAsfrom adenovirus 2 early region 4 by cDNA cloning and sequencing. Nucleic Acids Res 12,3503-19.

Frilander, M.J. and Steitz, J. A. (1999). Initial recognition of U12-dependent intronsrequires both U11/5' splice-site and U/12 branchpoint interactions. Genes Dev 13, 851-63.

- 47 -

Fu, X. D. (1993). Specific commitment of different pre-mRNAs to splicing by single SRproteins. Nature 365, 82-5.

Fu, X. D., Mayeda, A., Maniatis, T., and Krainer, A. R. (1992). General splicing factorsSF2 and SC35 have equivalent activities in vitro, and both affect alternative 5' and 3' splicesite selection. Proc Natl Acad Sci U S A 89, 11224-8.

Furuichi, Y., LaFiandra, A., and Shatkin, A. J. (1977). 5'-Terminal structure and mRNAstability. Nature 266, 235-9.

Gaur, R. K., Valcarcel, J., and Green, M. R. (1995). Sequential recognition of the pre-mRNA branch point by U2AF65 and a novel spliceosome-associated 28-kDa protein. RNA 1,407-17.

Ge, H., and Manley, J. L. (1990). A protein factor, ASF, controls cell-specific alternativesplicing of SV40 early pre-mRNA in vitro. Cell 62, 25-34.

Gilmartin, G. M., McDevitt, M. A., and Nevins, J. R. (1988). Multiple factors are requiredfor specific RNA cleavage at a poly(A) addition site. Genes Dev 2, 578-87.

Gontarek, R. R., McNally, M. T., and Beemon, K. (1993). Mutation of an RSV intronicelement abolishes both U11/U12 snRNP binding and negative regulation of splicing. GenesDev 7, 1926-36.

Gooding, C., Roberts, G. C., and Smith, C. W. (1998). Role of an inhibitory pyrimidineelement and polypyrimidine tract binding protein in repression of a regulated alpha-tropomyosin exon. RNA 4, 85-100.

Gooding, L. R., Sofola, I. O., Tollefson, A. E., Duerksen, H. P., and Wold, W. S. (1990).The adenovirus E3-14.7K protein is a general inhibitor of tumor necrosis factor-mediatedcytolysis. J Immunol 145, 3080-6.

Gozani, O., Patton, J. G., and Reed, R. (1994). A novel set of spliceosome-associatedproteins and the essential splicing factor PSF bind stably to pre-mRNA prior to catalytic stepII of the splicing reaction. EMBO J 13, 3356-67.

Granadino, B., Penalva, L. O. F., Green, M. R., Valcarcel, J., and Sanchez, L. (1997).Distinct mechanisms of splicing regulation in vivo by the Drosophila protein Sex-lethal. ProcNatl Acad Sci U S A 94, 7343-8.

Gui, J. F., Tronchere, H., Chandler, S. D., and Fu, X. D. (1994). Purification andcharacterization of a kinase specific for the serine- and arginine-rich pre-mRNA splicingfactors. Proc Natl Acad Sci U S A 91, 10824-8.

Gustin, K. E., and Imperiale, M. J. (1998). Encapsidation of viral DNA requires theadenovirus L1 52/55-kilodalton protein. J Virol 72, 7860-70.

Guth, S., Martinez, C., Gaur, R. K., and Valcarcel, J. (1999). Evidence for substrate-specific requirement of the splicing factor U2AF(35) and for its function after polypyrimidinetract recognition by U2AF(65). Mol Cell Biol 19, 8263-71.

- 48 -

Halbert, D. N., Cutt, J. R., and Shenk, T. (1985). Adenovirus early region 4 encodesfunctions required for efficient DNA replication, late gene expression, and host cell shutoff. JVirol 56, 250-7.

Hamm, J., and Mattaj, I. W. (1990). Monomethylated cap structures facilitate RNA exportfrom the nucleus. Cell 63, 109-18.

Hasson, T. B., Ornelles, D. A., and Shenk, T. (1992). Adenovirus l1 52-Kilodalton and 55-Kilodalton proteins are present within assembling virions and colocalize with nuclearstructures distinct from replication centers. J Virol 66, 6133-6142.

Hedley, M. L., Amrein, H., and Maniatis, T. (1995). An amino acid sequence motifsufficient for subnuclear localization of an arginine/serine-rich splicing factor. Proc Natl AcadSci U S A 92, 11524-8.

Hemström, C., Nordqvist, K., Pettersson, U., and Virtanen, A. (1988). Gene product ofregion E4 of adenovirus type 5 modulates accumulation of certain viral polypeptides. J Virol62, 3258-64.

Hoffman, B. E., and Grabowski, P. J. (1992). U1 snRNP targets an essential splicing factor,U2AF65, to the 3' splice site by a network of interactions spanning the exon. Genes Dev 6,2554-68.

Horwitz, M. S. (1996). Adenoviruses. In Fields Virology, B. N. Fields, D. M. Knipe, P. M.Howley and e. al, eds. (Philadelphia: Raven Press), pp. 2149-2172.

Huang, M. M., and Hearing, P. (1989). Adenovirus early region 4 encodes two geneproducts with redundant effects in lytic infection. J Virol 63, 2605-15.

Imperiale, M., Akusjärvi, G., and Leppard, K. (1995). Post-transcriptional control ofadenovirus gene expression. In The Molecular Repertoire of Adenoviruses. Current Topics inMicrobiology and Immunology, vol. 199/II, W. Doerfler and P. Böhm, eds.: Springer-Verlag,Berlin.), pp. 139-171.

Jamison, S. F., Pasman, Z., Wang, J., Will, C., Luhrmann, R., Manley, J. L., andGarcia-Blanco, M. A. (1995). U1 snRNP-ASF/SF2 interaction and 5' splice site recognition:characterization of required elements. Nucleic Acids Res 23, 3260-7.

Kandels-Lewis, S., and Seraphin, B. (1993). Involvement of U6 snRNA in 5' splice siteselection. Science 262, 2035-9.

Kanopka, A., Mühlemann, O., and Akusjärvi, G. (1996). Inhibition by SR proteins ofsplicing of a regulated adenovirus pre-mRNA. Nature 381, 535-538.

- 49 -

Kanopka, A., Mühlemann, O., Petersen-Mahrt, S., Estmer, C., Öhrmalm, C., andAkusjärvi, G. (1998). Regulation of adenovirus alternative RNA splicing bydephosphorylation of SR proteins. Nature 393, 185-7.

Keller, W. (1995). No end yet to messenger RNA 3' processing! Cell 81, 829-32.

Ketner, G., Bridge, E., Virtanen, A., Hemström, C., and Pettersson, U. (1989).Complementation of adenovirus E4 mutants by transient expression of E4 cDNA and deletionplasmids. Nucleic Acids Res 17, 3037-48.

Kim, E., Du, L., Bregman, D. B., and Warren, S. L. (1997). Splicing factors associate withhyperphosphorylated RNA polymerase II in the absence of pre-mRNA. J Cell Biol 136, 19-28.

Klein Gunnewiek, J. M., van de Putte, L. B., and van Venrooij, W. J. (1997). The U1snRNP complex: an autoantigen in connective tissue diseases. An update. Clin ExpRheumatol 15, 549-60.

Kleinberger, T., and Shenk, T. (1993). Adenovirus E4orf4 protein binds to proteinphosphatase 2A, and the complex down regulates E1A-enhanced junB transcription. J Virol67, 7556-7560.

Kohtz, J. D., Jamison, S. F., Will, C. L., Zuo, P., Luhrmann, R., Garcia-Blanco, M. A.,and Manley, J. L. (1994). Protein-protein interactions and 5'-splice-site recognition inmammalian mRNA precursors. Nature 368, 119-24.

Koizumi, J., Okamoto, Y., Onogi, H., Mayeda, A., Krainer, A. R., and Hagiwara, M.(1999). The subcellular localization of SF2/ASF is regulated by direct interaction with SRprotein kinases (SRPKs). J Biol Chem 274, 11125-31.

Konarska, M. M., and Sharp, P. A. (1987). Interactions between small nuclearribonucleoprotein particles in formation of spliceosomes. Cell 49, 763-74.

Konforti, B. B., Koziolkiewicz, M. J., and Konarska, M. M. (1993). Disruption of basepairing between the 5' splice site and the 5' end of U1 snRNA is required for spliceosomeassembly. Cell 75, 863-73.

Krainer, A. R., Conway, G. C., and Kozak, D. (1990). The essential pre-mRNA splicingfactor SF2 influences 5' splice site selection by activating proximal sites. Cell 62, 35-42.

Kreivi, J. P., Zerivitz, K., and Akusjärvi, G. (1991). Sequences involved in the control ofadenovirus L1 alternative RNA splicing. Nucleic Acids Res 19, 2379-2386.

Krishna, R. G., and Wold, F. (1993). Post-translational modification of proteins. AdvEnzymol Relat Areas Mol Biol 67, 265-98.

Krämer, A., and Utans, U. (1991). Three protein factors (SF1, SF3 and U2AF) function inpre-splicing complex formation in addition to snRNPs. EMBO J 10, 1503-9.

Lane, D. P., and Crawford, L. V. (1979). T antigen is bound to a host protein in SV40-transformed cells. Nature 278, 261-3.

- 50 -

Larsson, S., Kreivi, J. P., and Akusjärvi, G. (1991). Control of adenovirus alternative RNAsplicing: effect of viral DNA replication on RNA splice site choice. Gene 107, 219-27.

Larsson, S., Svensson, C., and Akusjärvi, G. (1992). Control of Adenovirus Major LateGene Expression at Multiple Levels. J Mol Biol 225, 287-298.

Lemay, P., Boudin, M. L., Milleville, M., and Boulanger, P. (1980). Human adenovirustype 2 protein IIIa. I. Purification and characterization. Virology 101, 131-43.

Lesser, C. F., and Guthrie, C. (1993). Mutations in U6 snRNA that alter splice sitespecificity: implications for the active site. Science 262, 1982-8.

Lewis, J. D., Gunderson, S. I., and Mattaj, I. W. (1995). The influence of 5' and 3' endstructures on pre-mRNA metabolism. J Cell Sci Suppl 19, 13-9.

Linzer, D. I., and Levine, A. J. (1979). Characterization of a 54K dalton cellular SV40tumor antigen present in SV40-transformed cells and uninfected embryonal carcinoma cells.Cell 17, 43-52.

Liu, H. X., Zhang, M., and Krainer, A. R. (1998). Identification of functional exonicsplicing enhancer motifs recognized by individual SR proteins. Genes Dev 12, 1998-2012.

Longman, D., Johnstone, I. L., and Caceres, J. F. (2000). Functional characterization of SRand SR-related genes in Caenorhabditis elegans. EMBO J 19, 1625-37.

Luo, M. J., and Reed, R. (1999). Splicing is required for rapid and efficient mRNA export inmetazoans. Proc Natl Acad Sci U S A 96, 14937-42.

Madhani, H. D., and Guthrie, C. (1994). Dynamic RNA-RNA interactions in thespliceosome. Annu Rev Genet 28, 1-26.

Manley, J. L., and Tacke, R. (1996). SR proteins and splicing control. Genes Dev 10, 1569-79.

Mannervik, M., Fan, S., Ström, A. C., Helin, K., and Akusjärvi, G. (1999). Adenovirus E4open reading frame 4-induced dephosphorylation inhibits E1A activation of the E2 promoterand E2F-1-mediated transactivation independently of the retinoblastoma tumor suppressorprotein. Virology 256, 313-21.

McCracken, S., Fong, N., Rosonina, E., Yankulov, K., Brothers, G., Siderovski, D.,Hessel, A., Foster, S., Shuman, S., and Bentley, D. L. (1997). 5'-Capping enzymes aretargeted to pre-mRNA by binding to the phosphorylated carboxy-terminal domain of RNApolymerase II. Genes Dev 11, 3306-18.

McCracken, S., Fong, N., Yankulov, K., Ballantyne, S., Pan, G., Greenblatt, J.,Patterson, S. D., Wickens, M., and Bentley, D. L. (1997). The C-terminal domain of RNApolymerase II couples mRNA processing to transcription. Nature 385, 357-61.

McNally, M. T., and Beemon, K. (1992). Intronic sequences and 3' splice sites control Roussarcoma virus RNA splicing. J Virol 66, 6-11.

- 51 -

McNally, M. T., Gontarek, R. R., and Beemon, K. (1991). Characterization of Roussarcoma virus intronic sequences that negatively regulate splicing. Virology 185, 99-108.

Merendino, L., Guth, S., Bilbao, D., Martinez, C., and Valcarcel, J. (1999). Inhibition ofmsl-2 splicing by Sex-lethal reveals interaction between U2AF35 and the 3' splice site AG.Nature 402, 838-41.

Mermoud, J. E., Cohen, P., and Lamond, A. I. (1992). Ser/Thr-specific proteinphosphatases are required for both catalytic steps of pre-mRNA splicing. Nucleic Acids Res20, 5263-9.

Mironov, A. A., Fickett, J. W., and Gelfand, M. S. (1999). Frequent alternative splicing ofhuman genes. Genome Res 9, 1288-93.

Misteli, T. (1999). RNA splicing: What has phosphorylation got to do with it? Curr Biol 9,R198-200.

Misteli, T., and Spector, D. (1997). Protein phosphorylation and the nuclear organization ofpre-mRNA splicing. Trends in Cell Bio. 7, 135-138.

Misteli, T., and Spector, D. L. (1998). The cellular organization of gene expression. CurrOpin Cell Biol 10, 323-31.

Moras, D., and Poterszman, A. (1995). RNA-protein interactions. Diverse modes ofrecognition. Curr Biol 5, 249-51.

Muller, U., Kleinberger, T., and Shenk, T. (1992). Adenovirus e4orf4 protein reducesphosphorylation of c-Fos and E1A proteins while simultaneously reducing the level of AP-1.J Virol 66, 5867-5878.

Mulligan, G. J., Guo, W., Wormsley, S., and Helfman, D. M. (1992). Polypyrimidine tractbinding protein interacts with sequences involved in alternative splicing of beta-tropomyosinpre-mRNA. J Biol Chem 267, 25480-7.

Murray, M. V., Kobayashi, R., and Krainer, A. R. (1999). The type 2C Ser/Thrphosphatase PP2Cgamma is a pre-mRNA splicing factor. Genes Dev 13, 87-97.

Mühlemann, O., Kreivi, J.-P., and Akusjärvi, G. (1995). Enhanced splicing ofnonconsensus 3' splice sites late during adenovirus infection. J Virol 69, 7324-7327.

Nevins, J. R., and Wilson, M. C. (1981). Regulation of adenovirus-2 gene expression at thelevel of transcriptional termination and RNA processing. Nature 290, 113-8.

Newman, A. (1994). RNA splicing. Activity in the spliceosome. Curr Biol 4, 462-4.

Newman, A. J., and Norman, C. (1992). U5 snRNA interacts with exon sequences at 5' and3' splice sites. Cell 68, 743-54.

- 52 -

Nordqvist, K., Öhman, K., and Akusjärvi, G. (1994). Human adenovirus encodes twoproteins which have opposite effects on accumulation of alternatively spliced mRNAs. MolCell Biol 14, 437-45.

Ohman, K. (1995). Function of adenovirus early region 4 proteins in RNA processing andcellular transformation (Stockholm: Karolinska Institute).

Ohman, K., Nordqvist, K., and Akusjärvi, G. (1993). Two adenovirus proteins withredundant activities in virus growth facilitates tripartite leader mRNA accumulation. Virology194, 50-8.

Okamoto, Y., Onogi, H., Honda, R., Yasuda, H., Wakabayashi, T., Nimura, Y., andHagiwara, M. (1998). cdc2 kinase-mediated phosphorylation of splicing factor SF2/ASF.Biochem Biophys Res Commun 249, 872-8.

Pilder, S., Moore, M., Logan, J., and Shenk, T. (1986). The adenovirus E1B-55Ktransforming polypeptide modulates transport or cytoplasmic stabilization of viral and hostcell mRNAs. Mol Cell Biol 6, 470-6.

Proudfoot, N. (2000). Connecting transcription to messenger RNA processing. TrendsBiochem Sci 25, 290-3.

Query, C. C., McCaw, P. S., and Sharp, P. A. (1997). A minimal spliceosomal complex Arecognizes the branch site and polypyrimidine tract. Mol Cell Biol 17, 2944-53.

Rabbitts, T. H., and Forster, A. (1978). Evidence for noncontiguous variable and constantregion genes in both germ line and myeloma DNA. Cell 13, 319-27.

Roberts, R. J., Akusjärvi, G., Aleström, P., Gelinas, R. E., Gingeras, T. R., Sciaky, D.,and Pettersson, U. (1986). A consensus sequence for the adenovirus-2 genome, Volume 8,W. Doerfler, ed. (Boston: Martin Nijhoff Publishing).

Rossi, F., Forne, T., Antoine, E., Tazi, J., Brunel, C., and Cathala, G. (1996). Involvementof U1 small nuclear ribonucleoproteins (snRNP) in 5' splice site-U1 snRNP interaction. J BiolChem 271, 23985-91.

Rossi, F., Labourier, E., Forne, T., Divita, G., Derancourt, J., Riou, J. F., Antoine, E.,Cathala, G., Brunel, C., and Tazi, J. (1996). Specific phosphorylation of SR proteins bymammalian DNA topoisomerase I. Nature 381, 80-2.

Rowe, w. r., r.j., H., l.k., G., r.h., P., and t.g., W. (1953). Isolation of a cytopathogenic agentfrom human adenpids undergoing sponataneous degenetation in tissue culture. Proc Soc ExpBiol Med 84, 570-573.

Ruskin, B., Zamore, P. D., and Green, M. R. (1988). A factor, U2AF, is required for U2snRNP binding and splicing complex assembly. Cell 52, 207-19.

Sachs, A. B., Sarnow, P., and Hentze, M. W. (1997). Starting at the beginning, middle, andend: translation initiation in eukaryotes. Cell 89, 831-8.

- 53 -

Schaal, T. D., and Maniatis, T. (1999). Multiple distinct splicing enhancers in the protein-coding sequences of a constitutively spliced pre-mRNA. Mol Cell Biol 19, 261-73.

Schaal, T. D., and Maniatis, T. (1999). Selection and characterization of pre-mRNA splicingenhancers: identification of novel SR protein-specific enhancer sequences. Mol Cell Biol 19,1705-19.

Schwer, B., and Shuman, S. (1996). Conditional inactivation of mRNA capping enzymeaffects yeast pre-mRNA splicing in vivo. RNA 2, 574-83.

Sharp, P. A. (1994). Split genes and RNA splicing. Cell 77, 805-15.

Singer, R. H., and Green, M. R. (1997). Compartmentalization of eukaryotic geneexpression: causes and effects. Cell 91, 291-4.

Sonenberg, N. (1993). Remarks on the mechanism of ribosome binding to eukaryoticmRNAs. Gene Expr 3, 317-23.

Sontheimer, E. J., and Steitz, J. A. (1993). The U5 and U6 small nuclear RNAs as activesite components of the spliceosome. Science 262, 1989-96.

Southby, J., Gooding, C., and Smith, C. W. (1999). Polypyrimidine tract binding proteinfunctions as a repressor to regulate alternative splicing of alpha-actinin mutally exclusiveexons. Mol Cell Biol 19, 2699-711.

Sprengler, J., Schmitz, B., Heuss-Neitzel, D., and Doerfler, W. (1995). The completenucleotide sequence of the DNA of human adenovirus type 12. In The Molecular Repertoireof Adenoviruses. Current Topics in Microbiology and Immunology, vol. 199/II, W. Doerflerand P. Böhm, eds.: Springer-Verlag, Berlin.), pp. 189-274.

Staknis, D., and Reed, R. (1994). SR proteins promote the first specific recognition of Pre-mRNA and are present together with the U1 small nuclear ribonucleoprotein particle in ageneral splicing enhancer complex. Mol Cell Biol 14, 7670-82.

Sun, Q., Mayeda, A., Hampson, R. K., Krainer, A. R., and Rottman, F. M. (1993).General splicing factor SF2/ASF promotes alternative splicing by binding to an exonicsplicing enhancer. Genes Dev 7, 2598-608.

Tacke, R., and Manley, J. L. (1999). Determinants of SR protein specificity. Curr Opin CellBiol 11, 358-62.

Tacke, R., and Manley, J. L. (1995). The human splicing factors ASF/SF2 and SC35possess distinct, functionally significant RNA binding specificities. EMBO J 14, 3540-51.

Tacke, R., Tohyama, M., Ogawa, S., and Manley, J. L. (1998). Human Tra2 proteins aresequence-specific activators of pre-mRNA splicing. Cell 93, 139-48.

Tazi, J., Kornstadt, U., Rossi, F., Jeanteur, P., Cathala, G., Brunel, C., and Luhrmann,R. (1993). Thiophosphorylation of U1-70K protein inhibits pre-mRNA splicing. Nature 363,283-6.

- 54 -

Teigelkamp, S., Newman, A. J., and Beggs, J. D. (1995). Extensive interactions of PRP8protein with the 5' and 3' splice sites during splicing suggest a role in stabilization of exonalignment by U5 snRNA. EMBO J 14, 2602-12.

Tigges, M. A., and Raskas, H. J. (1984). Splice junctions in adenovirus 2 early region 4mRNAs: multiple splice sites produce 18 to 24 RNAs. J Virol 50, 106-17.

Valcarcel, J., Gaur, R. K., Singh, R., and Green, M. R. (1996). Interaction of U2AF65 RSregion with pre-mRNA branch point and promotion of base pairing with U2 snRNA. Science273, 1706-9.

Valcarcel, J., Singh, R., Zamore, P. D., and Green, M. R. (1993). The protein Sex-lethalantagonizes the splicing factor U2AF to regulate alternative splicing of transformer pre-mRNA. Nature 362, 171-5.

Virtanen, A., Aleström, P., Persson, H., Katze, M. G., and Pettersson, U. (1982). Anadenovirus agnogene. Nucleic Acids Res 10, 2539-48.

Virtanen, A., Gilardi, P., Näslund, A., LeMoullec, J. M., Pettersson, U., and Perricaudet,M. (1984). mRNAs from human adenovirus 2 early region 4. J Virol 51, 822-31.

Visa, N., Izaurralde, E., Ferreira, J., Daneholt, B., and Mattaj, I. W. (1996). A nuclearcap-binding complex binds Balbiani ring pre-mRNA cotranscriptionally and accompanies theribonucleoprotein particle during nuclear export. J Cell Biol 133, 5-14.

Wang, J., and Manley, J. L. (1995). Overexpression of the SR proteins ASF/SF2 and SC35influences alternative splicing in vivo in diverse ways. RNA 1, 335-46.

Wang, Z., Hoffmann, H. M., and Grabowski, P. J. (1995). Intrinsic U2AF binding ismodulated by exon enhancer signals in parallel with changes in splicing activity. RNA 1, 21-35.

Watakabe, A., Tanaka, K., and Shimura, Y. (1993). The role of exon sequences in splicesite selection. Genes Dev 7, 407-18.

Weinberg, D. H., and Ketner, G. (1986). Adenoviral early region 4 is required for efficientviral DNA replication and for late gene expression. J Virol 57, 833-8.

Whyte, P., Buchkovich, K. J., Horowitz, J. M., Friend, S. H., Raybuck, M., Weinberg, R.A., and Harlow, E. (1988). Association between an oncogene and an anti-oncogene: theadenovirus E1A proteins bind to the retinoblastoma gene product. Nature 334, 124-9.

Wold, W. S. M., Tollefson, A. E., and Hermiston, T. W. (1995). E3 transcription unit ofadenovirus. In The Molecular Repertoire of Adenoviruses. Current Topics in Microbiologyand Immunology, vol. 199/I, W. Doerfler and P. Böhm, eds.: Springer-Verlag, Berlin.), pp.237-274.

Wu, J. Y., and Maniatis, T. (1993). Specific interactions between proteins implicated insplice site selection and regulated alternative splicing. Cell 75, 1061-70.

- 55 -

Wu, S., Romfo, C. M., Nilsen, T. W., and Green, M. R. (1999). Functional recognition ofthe 3' splice site AG by the splicing factor U2AF35. Nature 402, 832-5.

Wyatt, J. R., Sontheimer, E. J., and Steitz, J. A. (1992). Site-specific cross-linking ofmammalian U5 snRNP to the 5' splice site before the first step of pre-mRNA splicing. GenesDev 6, 2542-53.

Xiao, S. H., and Manley, J. L. (1997). Phosphorylation of the ASF/SF2 RS domain affectsboth protein-protein and protein-RNA interactions and is necessary for splicing. Genes Dev11, 334-44.

Xiao, S. H., and Manley, J. L. (1998). Phosphorylation-dephosphorylation differentiallyaffects activities of splicing factor ASF/SF2. EMBO J 17, 6359-67.

Yuryev, A., Patturajan, M., Litingtung, Y., Joshi, R. V., Gentile, C., Gebara, M., andCorden, J. L. (1996). The C-terminal domain of the largest subunit of RNA polymerase IIinteracts with a novel set of serine/arginine-rich proteins. Proc Natl Acad Sci U S A 93, 6975-80.

Zahler, A. M., Lane, W. S., Stolk, J. A., and Roth, M. B. (1992). SR proteins: a conservedfamily of pre-mRNA splicing factors. Genes Dev 6, 837-47.

Zahler, A. M., Neugebauer, K. M., Lane, W. S., and Roth, M. B. (1993). Distinctfunctions of SR proteins in alternative pre-mRNA splicing. Science 260, 219-22.

Zahler, A. M., and Roth, M. B. (1995). Distinct functions of SR proteins in recruitment ofU1 small nuclear ribonucleoprotein to alternative 5' splice sites. Proc Natl Acad Sci U S A 92,2642-6.

Zamore, P. D., and Green, M. R. (1991). Biochemical characterization of U2 snRNPauxiliary factor: an essential pre-mRNA splicing factor with a novel intranuclear distribution.EMBO J 10, 207-14.

Zamore, P. D., and Green, M. R. (1989). Identification, purification, and biochemicalcharacterization of U2 small nuclear ribonucleoprotein auxiliary factor. Proc Natl Acad Sci US A 86, 9243-7.

Zamore, P. D., Patton, J. G., and Green, M. R. (1992). Cloning and domain structure of themammalian splicing factor U2AF. Nature 355, 609-14.

Zhang, M., Zamore, P. D., Carmo-Fonseca, M., Lamond, A. I., and Green, M. R. (1992).Cloning and intracellular localization of the U2 small nuclear ribonucleoprotein auxiliaryfactor small subunit. Proc Natl Acad Sci U S A 89, 8769-73.

Zhang, W. J., and Wu, J. Y. (1998). Sip1, a novel RS domain-containing protein essentialfor pre-mRNA splicing. Mol Cell Biol 18, 676-84.

Zorio, D. A., and Blumenthal, T. (1999). Both subunits of U2AF recognize the 3' splice sitein Caenorhabditis elegans. Nature 402, 835-8.