On co-transcriptional splicing and U6 snRNA …€¦ · On co-transcriptional splicing and U6 snRNA...
Transcript of On co-transcriptional splicing and U6 snRNA …€¦ · On co-transcriptional splicing and U6 snRNA...
On co-transcriptional splicing and
U6 snRNA biogenesis
DISSERTATION
zur Erlangung des akademischen Grades
Doctor rerum naturalium
(Dr.rer.nat.)
vorgelegt
der Fakultät Mathematik und Naturwissenschaften
der Technischen Universität Dresden
von
Imke Listerman
Master of Science in Biology
geboren am 23. September 1977 in Bremerhaven, Deutschland
Gutachter:
Prof. Dr. Francis Stewart Technische Universität, Dresden
Dr. Karla Neugebauer Max Planck Institut für Molekulare Zellbiologie und Genetik, Dresden
Prof. Dr. Neus Visa Stockholms Universitet, Schweden
Table of Contents
Table of Contents
Table of Contents ........................................................................................................1
Summary .....................................................................................................................3
Part I ............................................................................................................................5
Introduction.................................................................................................................7
5’ end capping and the cap binding complex..............................................................7
Pre-mRNA splicing .................................................................................................10
Chromatin-Immunoprecipitation..............................................................................14
How does the spliceosome assemble co-transcriptionally?.......................................14
Transcription and mRNA processing .......................................................................16
Does Pol II couple splicing to transcription?............................................................20
Aim of part I of the thesis ........................................................................................21
Results........................................................................................................................22
Pol II and CBC80 accumulate on promoter-proximal positions of constitutively active
genes .......................................................................................................................22
Pol II and CBP80 accumulate on 5’ ends of uninduced c-fos and HSP70 but are
robustly detectable throughout the induced genes ....................................................25
Establishing splicing factor ChIP in mammalian cells..............................................26
Co-transcriptional detection of hnRNP A1...............................................................29
Splicing factors accumulate on the 5’ end of β-actin................................................29
Splicing factor accumulation on c-fos is dependent on transcription and can be
enhanced by camptothecin.......................................................................................31
Splicing factors do not accumulate on intronless HSP70..........................................34
Interactions between splicing factors and Pol II .......................................................38
Development of Chromatin-RNA-Immunprecipitation (ChRIP) ..............................40
Co-transcriptional splicing of c-fos is enhanced by camptothecin.............................43
Discussion ..................................................................................................................44
Materials and Methods .............................................................................................51
Cell culture and treatments ......................................................................................51
Chromatin Immunoprecipitation (ChIP) and Real Time PCR...................................51
Table of Contents
2
RNA extraction and RT-PCR ..................................................................................52
Standard Immunoprecipitations and Western Blot Analysis.....................................53
Indirect immunofluorescence ..................................................................................53
Chromatin-RNA-Immunoprecipitation (ChRIP) ......................................................54
Part II ........................................................................................................................57
Introduction...............................................................................................................59
The U6 snRNA .......................................................................................................59
Recruitment of Pol II to mRNA and snRNA promoters ...........................................61
Recruitment of Pol III to the U6 snRNA promoter...................................................64
Evidence for U6 snRNA transcription by Pol III......................................................65
U6 snRNA genes.....................................................................................................68
Chromosome...........................................................................................................69
Aim of part II of the thesis.......................................................................................70
Results .......................................................................................................................71
Pol II and III accumulate on the U6-1 snRNA gene promoter ..................................71
Pol II and Pol III accumulate on other U6 snRNA genes..........................................76
α-amanitin inhibits U6 maxigene transcription in vivo.............................................80
Endogenous U6 snRNA is affected through α-amanitin treatment ...........................82
Endogenous U6 snRNA is polyadenylated ..............................................................84
Discussion ..................................................................................................................85
Materials and Methods .............................................................................................91
Cell culture and treatments ......................................................................................91
Mapping of potential human U6 snRNA genes........................................................91
Chromatin Immunoprecipitation and Real Time PCR..............................................91
RNA extraction and RT-PCR ..................................................................................93
References..................................................................................................................96
Appendix .................................................................................................................109
Acknowledgements..................................................................................................112
Declaration ..............................................................................................................113
Summary
3
Summary
Messenger RNA (mRNA) is transcribed by RNA polymerase II (Pol II) and has to
undergo multiple processing events before it can be translated into a protein: a cap
structure is added to its 5’ end, noncoding, intervening sequences (introns) are removed
and coding exons are ligated together and a poly(A) tail is added to its 3’end. Splicing,
the process of intron removal, is carried out in the spliceosome, a megacomplex
comprehending up to 300 proteins. The core components of the spliceosome that
directly interact with the pre-mRNA are the small nuclear ribonucleoprotein particles
(snRNPs). They consist of one of the U-rich snRNAs U1, U2, U4, U5 or U6 together
with several particle-specific proteins and core proteins. All mRNA processing events
can occur co-transcriptionally, i.e. while the RNA is still attached to the gene via Pol II.
The in vivo studies of co-transcriptional RNA processing events had been possible only
in special biological systems by immunoelectron microscopy and only recently,
Chromatin Immunoprecipitation (ChIP) made it possible to investigate co-
transcriptional splicing factor assembly on genes.
My thesis work is divided into two parts: Part I shows that the core components
of the splicing machinery are recruited co-transcriptionally to mammalian genes in vivo
by ChIP. The co-transcriptional splicing factor recruitment is dependent on active
transcription and the presence of introns in genes. Furthermore, a new assay was
developed that allows for the first time the direct monitoring of co-transcriptional
splicing in human cells. The topoisomerase I inhibitor camptothecin increases splicing
factor accumulation on the c-fos gene as well as co-transcriptional splicing levels,
which provides direct evidence that co-transcriptional splicing events depend on the
kinetics of RNA synthesis.
Part II of the thesis is aimed to investigate whether Pol II has a functional role in
the biogenesis of the U6 snRNA, which is the RNA part of the U6 snRNP involved in
splicing. Pol III had been shown to transcribe the U6 snRNA gene, but ChIP
experiments revealed that Pol II is associated with all the active U6 snRNA gene
promoters. Pol II inhibition studies uncovered that U6 snRNA expression and probably
3’end formation is dependent on Pol II.
Part I
4
Part I
5
PART I
Splicing is not a biochemist’s dream.
Ian Eperon
6
Part I: Introduction
7
Introduction
The life of a messenger RNA (mRNA) molecule starts with its synthesis by RNA
polymerase II (Pol II) and awaits multiple processing events, before it is mature and can
be translated into a protein. At the transcription unit (TU), the region of DNA that is
transcribed to produce a single primary RNA transcript, transcription and RNA
processing steps occur in parallel (see Figure 1). Pol II has to be recruited to the
promoter to enter the pre-initiation complex, initiate transcript synthesis, elongate the
nascent transcript and terminate transcription appropriately. The transition from
initiation to elongation demands modifications of the transcription complex. Without
these signals, elongation complexes may pause, arrest or even terminate prematurely.
The nascent RNA has to undergo 5’ capping, splicing and 3’end formation before it
represents a translatable mRNA. Each of these RNA processing events occurs
independently of one another. However, a growing body of evidence suggests
functional relationships between these processes that occur, at least partly, co-
transcriptionally. But what is the significance of co-transcriptional RNA processing?
The following paragraphs will give a brief introduction to mRNA processing and
interactions between the transcription and RNA processing machineries.
5’ end capping and the cap binding complex
A distinct feature of all RNAs transcribed by RNA Pol II is the presence of a cap
structure at their 5’ end that protects the transcript from 5’ exoribonucleases. This
modification occurs co-transcriptionally when the nascent transcript is approximately
20-30 nt long (Rasmussen and Lis, 1993). In humans, capping is accomplished by the
sequential action of three enzymatic activities that reside in the two proteins Hce1 and
Hcm1: RNA triphosphatase removes the γ-phosphate from the 5’end of the transcript,
RNA guanylyltransferase transfers GMP to the diphosphate end which results in the G
cap, and RNA methyltransferase methylates the G, resulting in a 7-methyl guanosine
(m7G) cap (Cougot et al., 2004; see Figure 2). In fact, the enzymes performing these
reactions were shown in vitro to bind to the phosphorylated form of the C-terminal
Part I: Introduction
8
Figure 1. Co-transcriptional pre-mRNA processing.Schematic of Pol II transcription and co-transcriptional RNA processing events. Pol II (black ball)initiates transcription at the promoter (arrow), elongates along the transcription unit (TU), terminatesafter passage through the polyadenylation signal and releases from the DNA template, followed byrecycling. Polyadenylation factors, such as CPSF and CstF (yellow hexagons) bind directly to Pol II andtravel with Pol II along the TU. Additional polyadenylation factors (brown hexagon) are recruited to Pol IIat downstream regions. Capping enzymes (light green oval) bind to Pol II as it enters elongation phaseand then dissociate from the TU.The RNA is capped at its 5’ end shortly after transcription initiation, symbolized by the baseball cap.Splicing factors recognizing the 5’ and 3’ splice site (red and light blue ball, respectively) associate to theTU and splicing occurs in the completely assembled spliceosome (orange oval). The RNA is cleaved andpolyadenylated, the mRNP is released from the template and is transported to the cytoplasm. Attermination, the fragment of cleaved nascent RNA remaining is degraded (adapted from Neugebauer,2002).
Figure 2. 5’ capping.The addition of the 7-methyl-guanosine cap occurs after 20-30 nucleotides have been synthesized. AnRNA 5’ triphosphatase hydrolyzes the triphosphate of the first nucleotide to a diphosphate. Then, aguanylyltransferase catalyzes the addition of a GMP to the diphosphate via a 5’-5’ triphosphate linkage.In humans, both the triphosphatase and guanylyltransferase activity reside in the Hce1 protein. Finally,the methyltransferase Hcm1 methylates the N7 position of the GMP. The cap structure is bound by thecap binding complex (CBC) in the nucleus (based on Alberts, 2002).
Part I: Introduction
9
domain (CTD) of the large subunit of RNA Pol II (Cho et al., 1997; McCracken et al.,
1997a). In the yeast S. cerevisiae as well as in human cells, the capping enzymes were
reported to associate to promoter regions of actively transcribed genes in vivo,
consistent with their recruitment to the TU by the Pol II CTD (Cheng and Sharp, 2003;
Komarnitsky et al., 2000). Splicing is directly dependent on the presence of the cap
structure. Only precursor mRNAs (pre-mRNAs) with an m7G cap are efficiently spliced
in HeLa whole cell or nuclear extracts, thus underlining the interconnection of different
RNA processing events (Edery and Sonenberg, 1985; Konarska et al., 1984; Krainer et
al., 1984).
The m7G cap is bound in the nucleus by the cap binding complex (CBC), a
heterodimer of a 20 kDa and 80 kDa subunit (Izaurralde et al., 1994). Early in vitro
studies suggested that not only the m7G cap, but also the CBC facilitates splicing.
Immunodepletion of CBP80 in splicing extracts severely inhibited the splicing of
capped reporter pre-mRNA and only back-addition of both recombinant CBC subunits
restored splicing efficiency (Izaurralde et al., 1994). The proposal that both CBC
subunits are required for efficient binding to the cap structure was confirmed recently
by X-ray crystallography, revealing that CBP20 is the cap-contacting subunit while
CBP80 is binding to CBP20 (Calero et al., 2002; Mazza et al., 2001). But how does the
CBC support splicing? CBC depletion or addition of cap analogues to in vitro splicing
assays indicated that the earliest steps of spliceosome assembly were impaired
(Izaurralde et al., 1994). More detailed studies revealed that the CBC is required for U1
snRNP binding to the proximal 5’ splice site (Lewis et al., 1996). In the yeast S.
cerevisiae, studies with CBP20 (Mud13p) deletion mutants confirmed that CBC acts in
the early steps of spliceosome formation (Colot et al., 1996). In vivo studies performed
in our lab revealed that deletion of the CBC inhibits the co-transcriptional binding of the
U1 snRNP to the pre-mRNA in the majority of intron-containing yeast genes. In a
minority of intron-containing genes with long first exons, CBC seems to stimulate the
displacement of the U1 snRNP from the 5’splice by U6 snRNP, thus allowing the
formation of the active spliceosome (Gornemann et al., 2005). The last result is also in
agreement with experiments conducted in HeLa nuclear extracts showing that the cap
structure is required for efficient interaction of U6 snRNA with the 5’ splice site
(O'Mullane and Eperon, 1998).
Part I: Introduction
10
Co-transcriptional binding of the CBC to nascent RNA has been shown to occur
in yeast (Gornemann et al., 2005; Zenklusen et al., 2002) and in higher eukaryotes, as
shown by immunoelectron microscopy at Balbiani ring genes in C. tentans (Visa et al.,
1996). The CBC may also bind to mammalian genes co-transcriptionally, but due to a
lack of appropriate in vivo methods, this has not been shown.
Pre-mRNA splicing
Removal of introns is a crucial process in the production of RNA in all eukaryotes. The
average human gene contains 8 introns with an average length of 3.4 kb, interspersed by
exons that average less than 300 bp in length (ConsortiumInternational, 2004; Lander et
al., 2001).
Splicing is mediated by the spliceosome, a complex of over 300 proteins and
five RNA molecules (Jurica and Moore, 2003, and references therein). This perplexing
megacomplex is highly dynamic and changes its structure and composition during the
process of splicing. The small nuclear ribonucleoprotein particles (snRNPs) belong to
the basic components of the spliceosome. They consist of one of the U-rich snRNA U1,
U2, U4, U5 or U6 together with several particle-specific proteins and either Sm core
proteins or, in the case of U6, like-Sm (Lsm) core proteins. In addition, many non-
snRNP protein factors participate in each step of the splicing reaction.
The selection of intron and exon boundaries is an early step in spliceosome
formation, in which the 5’ and 3’ ends of the intron are recognized through basepairing
interactions between the pre-mRNA with the U1 and U2 snRNAs, respectively (Black,
2003). The mammalian 5’ splice site consensus is AG/GURAGU, whereas the 3’ splice
site resembles CAG/G. Other important sequences for splice site recognition are the
branchpoint sequence (YNYURAC) which is usually 20 to 100 nt upstream of the 3’
splice site and an immediately adjacent stretch of pyrimidines termed polypyrimidine
tract (Soller, 2006, see Figure 3A). The first step in spliceosome assembly is the
recognition of the 5’ splice site by the U1 snRNP and binding of splicing factor 1 (SF1)
to the branchpoint and U2 auxiliary factor (U2AF35 and U2AF65) to the poly-pyrimidine
tract, thereby enhancing U2 snRNP recruitment (Valcarcel and Green, 1996, see Figure
4). The specificity of SF1 for relevant branchpoint sequences is refined by interactions
between the U2AF65 subunit that also directly contacts the polypyrimidine tract
(Goldstrohm et al., 2001, and references therein). The SF1/U2AF complex at the 3’
Part I: Introduction
11
splice site and U1 snRNP at the 5’ splice site define the borders of the intron. Binding
of U2 snRNP to the branchpoint sequence via base-pairing between the pre-mRNA and
the U2 snRNA requires the displacement of SF1 and U2AF from the branchpoint,
although they may remain transiently associated to the spliceosome (Das et al., 2000).
Figure 3. The splicing reaction.(A) Most of the vertebrate pre-mRNA introns start with a 5’-GU-3’ and end with a 5’-AG-3’. They aretherefore called ‘GU-AG’ introns and all members of this class are spliced in the same way. These splicesites are parts of longer consensus sequences that span the 5’ and 3’ splice sites. In vertebrates, the 5’splice site consensus is 5’-AG↓GUAAGU-3’, whereas the 3’ splice site consensus is 5’-PyPyPyPyPyPyNCAG↓-3’. The polypyrimidine tract and the branchpoint sequence 5’-YNYURAC-3’ arelocated immediately upstream of the 3’ splice site. Py or Y, pyrimidine (U/T or C); R, A or G; N, anynucleotide; ↓ exon-intron boundary. (B) In the first transesterification reaction, the 2’-hydroxyl group ofthe branchpoint adenosine attacks the 5’ splice site junction, leading to 2’-5’ phosphodiester branchwithin the intron sequence, the lariat structure. In the second transesterification reaction, the 3’-OHgroup of the upstream exon attacks the phosphodiester bond of the 3’ splice site, inducing cleavage andthe release of the lariat intron. This enables the two exons to be ligated together (based on Alberts,2002).
A
B
Part I: Introduction
12
Spliceosome formation requires the additional recruitment of the U4/U6⋅U5 tri-snRNP
particle. The spliceosome becomes catalytically active upon rearrangement and
destabilization of U1 and U4 snRNPs, which leads to the recognition of part of the 5’
splice site by U6 snRNA (Staley and Guthrie, 1998) (Goldstrohm et al., 2001, and
references therein) and the addition of the Prp19 associated complex (Chan et al., 2003;
Tarn et al., 1994). Intron removal is carried out in two transesterification steps: During
the first step, the branchpoint adenosine attacks the phosphodiester linkage of the 5’
splice site of the intron, followed by the ligation of the 5’ end of the intron to the 2’-
hydroxyl group of the branchpoint adenosine, forming a lariat structure. In the second
step, the phosphodiester linkage of the 3’ splice site is attacked by the 3’-hydroxyl
group of the upstream exon, resulting in joining of the two exons and release of the
lariat (Moore and Sharp, 1993, see Figure 3B). Although it is assumed that the lariat is
rapidly degraded (Padgett et al., 1986), more detailed studies revealed much longer
intron half lives (Clement et al., 2001; Clement et al., 1999).
The faithful selection of splice sites to produce a functional mRNA is of utmost
importance, and an highly complicated matter considering the vast amount of intronic
sequence that contains many cryptic splice sites. It is stupendous how well controlled all
splicing decisions are accomplished in the cell. It has been estimated that at least fifteen
percent of all point mutations that result in human genetic disease create an RNA
splicing defect (Faustino and Cooper, 2003; Krawczak et al., 1992). The spliceosome in
higher eukaryotes has not only the daunting task of assembling from its parts onto each
individual intron, thereby recognizing only the true splice sites and ignoring all cryptic
splice sites. On top of this, the spliceosome also has the choice between alternative
splice sites, which, selected at the right time in the appropriate context, results in
inclusion of alternative exons into the mRNA. Alternative splicing is very common in
higher eukaryotes - about 50 – 70 % of human genes are alternatively spliced - and
greatly enhances the proteome diversity in the cell (Black, 2003). Many splicing factors
have been identified that aid in splice site recognition, among them the serine-arginine
(SR) rich proteins, polypyrimidine-tract binding protein (PTB) or Nova. They bind to
rather ill defined sequences close to splice sites and generally either facilitate or block
the binding of the core spliceosome components (Soller, 2006, and references therein,
see Figure 4).
Part I: Introduction
13
Figure 4. Spliceosome assembly in vitro.An in vitro synthesized pre-mRNA is added to a nuclear extract. Splicing activity is initiated by formationof the commitment complex. This complex comprises the U1 snRNP binding to the 5’ splice site, splicingfactor 1 (SF1) and U2 auxiliary factor (U2AF35 and U2AF65), that bind to the branchpoint, polypyrimidinetract and 3’ splice site, respectively. Binding of these factors might also be enhanced by alternativesplicing factors (like SR proteins) bound to exonic splicing enhancers (ESEs). U2 snRNP recruitmentleads to replacement of SF1 and U2AF and the formation of the pre-spliceosomal complex. Thespliceosome is formed when the U4/U6⋅U5 tri-snRNP particle joins. Upon rearrangement anddestabilization of the U1 and U4 snRNPs, the 3’ splice site is brought close to the 5’ spice site and thebranch point and the two transesterification reactions occur, leading to ligation of the two exons and therelease of the lariat intron (based on Jurica and Moore, 2003).
Part I: Introduction
14
Chromatin-Immunoprecipitation
The direct examination of co-transcriptional association of RNA processing factors to
the nascent RNA had been for a long time only been possible in special biological
systems such as the Balbiani ring genes in C. tentans. Work from our lab and others
have established a new method to look at co-transcriptional recruitment events
(Gornemann et al., 2005; Kotovic et al., 2003; Lacadie and Rosbash, 2005; Zenklusen et
al., 2002). Since the nascent transcript and its associated proteins lie adjacent to the
DNA axis (Wetterberg et al., 2001), formaldehyde can be used to rapidly induce
reversible 2 Å crosslinks between protein-protein and protein-nucleic acid reactive
groups (Orlando, 2000), i.e. between RNA binding proteins to the nascent RNA via Pol
II to the DNA axis. After cell lysis and chromatin shearing, the crosslinked
RNA/protein/DNA complexes can be immunopurified, followed by DNA extraction
and PCR amplification. This method called Chromatin Immunoprecipitation (ChIP)
allows to monitor simultaneously the in vivo association of Pol II and RNA processing
factors along the length of a gene and has been used successfully in our lab (Gornemann
et al., 2005; Kotovic et al., 2003; see Figure 5). So far, this method was only applied in
yeast, but was enormously helpful for our understanding of co-transcriptional RNA
processing events.
How does the spliceosome assemble co-transcriptionally?
Spliceosome assembly has been characterized biochemically, using nuclear extracts,
which allows for the investigation of intermediate splicing complexes. Spliceosome
assembly in these in vitro systems is uncoupled from transcription. Most of this in vitro
work is consistent with biochemical and genetic analysis of spliceosome formation in
yeast extracts (Reed, 2000, and references therein). It has been assumed that the in vivo
assembly of the spliceosome occurs according to the in vitro data in a step-wise fashion,
with the U1 snRNP binding to the 5’ splice site and U2AF, SF1 and the U2 snRNP
binding the polypyrimidine tract and branch point and 3’ splice site of the pre-mRNA,
respectively, followed by the U4/U6⋅U5 tri-snRNP (Black, 2003; Reed, 2000).
However, another hypothesis had been formulated against the stepwise assembly of the
spliceosome but instead argues for the recruitment of a preformed spliceosomal
complex to the pre-mRNA.
Part I: Introduction
15
Figure 5. Chromatin immunoprecipitation.By applying formaldehyde crosslinking to happily growing cells (1+2), a snapshot of the transcriptionunit can be taken in which the nascent RNA (and its associated processing factors) are still attached tothe chromatin via Pol II. After crosslinking, the cells are lysed and the chromatin is sheared bysonication (3), followed by immunoprecipitation with antibodies against endogenous RNA processingfactors (4). The crosslinks between the precipitated DNA/protein/RNA complexes are reversed and theDNA is purified (5) and Real-Time PCR amplified with primers along the gene of interest (6).
This so called penta-snRNP could be isolated as a 45S particle from yeast nuclear
extracts and contains all spliceosomal snRNPs (Stevens et al., 2002). In vitro evidence
from experiments in metazoan support this penta-snRNP hypothesis: In HeLa nuclear
extracts, a complex of five spliceosomal snRNPs binds a specific short RNA containing
a 5' splice site through base pairing with the 5' end of U1 (Malca et al., 2003). It was
furthermore shown that the U1 and U5 snRNPs contact the 5’ splice site without prior
3’ splice site recognition (Maroney et al., 2000) and before splicing intermediates could
be detected (Wyatt et al., 1992). Complexes of 200S that contain all of the spliceosomal
snRNPs, most non-snRNP splicing factors, and pre-mRNA could be isolated by
gradient centrifugation of HeLa nuclear extracts (Miriami et al., 1995; Raitskin et al.,
Part I: Introduction
16
2002). These complexes were termed supraspliceosome and were shown by cryo-
electron microscopy to consist of 4 connected subcomplexes that might harbor one
spliceosome each (Azubel et al., 2006).
The investigation of spliceosome assembly in vivo had been a difficult task. It
has been known for quite some time that splicing and consequently spliceosome
assembly can start co-transcriptionally, meaning while the nascent transcript is attached
to the DNA axis by Pol II (Bauren and Wieslander, 1994; Osheim et al., 1985;
Wetterberg et al., 2001). However, the in vivo detection of specific splicing factors
associating co-transcriptionally to the pre-mRNA in other systems than insect cells was
until recently not possible due to the lack of specific assays. Work from our lab
demonstrated that ChIP can be used not only to directly monitor the association of
splicing factors to the gene as it is being transcribed but also the spatial and temporal
resolution of their binding to the nascent RNA (Kotovic et al., 2003). Indeed, the
individual components assemble in a stepwise fashion on an intron-containing gene in
yeast, starting with U1 snRNP as the 5’ splice site emerges from Pol II, followed by the
branch point binding protein (BBP), yeast U2AF65 (Mud2) and U2 snRNP as the 3’
splice site is available. In agreement with the in vitro model of spliceosome assembly,
Mud2 and BBP are less well detectable in regions downstream of the 3’ splice site,
while U2 snRNP detection is unchanged, indicating that binding of the U2 snRNP leads
to a displacement of these two factors. Furthermore, U5 snRNP and other components
of the active spliceosome could be detected at the downstream region, but not at
“earlier” sequences close to the 5’ splice site, demonstrating that the active spliceosome
can form in vivo co-transcriptionally and assembles from its individual parts
(Gornemann et al., 2005; Lacadie and Rosbash, 2005).
Transcription and mRNA processing
A large body of excessively reviewed evidence demonstrates that RNA undergoes
processing events while it is still transcribed by Pol II (Proudfoot et al., 2002; Bentley,
2005; Neugebauer, 2002). Although it seems instinctively plausible that RNA
processing events can happen as Pol II moves along the gene (Pol II transcribes with a
speed of about 1-1.5 kb/min, and in vivo intron removal occurs with t(1/2) of 0.4 min;
(Audibert et al., 2002), the relevance of coupling these events is not completely clear.
As mentioned above, the physical coupling of 5’capping to transcription by direct
Part I: Introduction
17
binding of the capping enzymes to the polymerase ensures the immediate addition of
the protecting cap to the growing nascent RNA (Neugebauer, 2002). In line with the
earlier proposal (Greenleaf, 1993), the CTD was shown to be the determining factor in
directly binding the capping enzymes (Cho et al., 1997; McCracken et al., 1997a).
The CTD of the largest of the 12 Pol II subunits is a highly conserved domain,
consisting of 52 heptad repeats in mammals (26 in yeast) with the consensus sequences
Y1S2P3T4S5P6S7 (Cramer et al., 2001; Palancade and Bensaude, 2003). In mammals, the
CTD is 100 % conserved, indicating its important role in the Pol II holoenzyme. The
CTD is essential; total deletion leads to cell death in yeast and mammalian cells. Yeast
is viable with a minimum of 8 repeats (West and Corden, 1995); deletion to 31 repeats
interferes with mammalian cell viability (Meininghaus et al., 2000), but mutant mice
homozygous for a deletion to 39 repeats are smaller, but viable (Litingtung et al., 1999).
Deletion of the non-consensus 1st to 3rd and and/or 52nd repeat leads to proteolytic
degradation of the CTD in vivo, but all the other internal repeats can be deleted without
inducing degradation (Chapman et al., 2005; Chapman et al., 2004).
Phosphorylation of the CTD is a modification associated with the elongating
form of the Pol II holoenzyme. CTD phosphorylation leads to a remarkable shift in its
electromobility, leading to an increase in molecular weight from 210 (the hypo-
phosphorylated or IIa form) to 240 kDa (the hyper-phosphorylated for IIo form: Cadena
and Dahmus, 1987). The major CTD phosphorylations occur at two Serines at the 2nd
(Serine 2) and 5th (Serine 5) position. The three kinases CDK7, CDK8 and CDK can
phosphorylate the CTD, and modify in this way the elongation properties of Pol II
(reviewed in Dahmus, 1995). CDK7, the kinase subunit of the general transcription
factor TFIIH, can phosphorylate Serine 5 (Feaver et al., 1994; Rodriguez et al., 2000;
Valay et al., 1995). Also CDK8 had been shown to phosphorylate Serine 5, and is
implicated to play a role in transcriptional repression by phosphorylating the CTD
prematurely, thereby preventing the formation of the transcription initiation complex
(Hengartner et al., 1998). Serine 2 is phosphorylated by CDK9, the kinase activity of
the positive elongation factor (Kim et al., 2002; Ramanathan et al., 2001; Zhou et al.,
2000).
Antibodies recognizing the different phosphoepitopes of the CTD have been
generated, although they are not 100 % solely specific for one CTD phosphoepitope
(Palancade and Bensaude, 2003). Early in vitro experiments and recent ChIP studies
Part I: Introduction
18
established that the Pol II CTD is hypo-phosphorylated when it is recruited to the
promoter. Upon transcription initiation, Serine 5 is phosphorylated, while the Serine 2
gets phosphorylated as Pol II entries into elongation mode further down the gene
(Cadena and Dahmus, 1987; Cheng and Sharp, 2003; O'Brien et al., 1994; Payne et al.,
1989).
Additionally to phosphorylation, other modification of the CTD exists, such as
isomerization and glycolysation (Buratowski, 2003). All these modifications may
influence the binding of other transcription and RNA processing factors to the CTD.
There are a growing number of reports of interaction of RNA polyadenylation factors
with Pol II and the CTD. In HeLa cell extracts, PSF and p54/nrbNonO, factors that have
been implicated in splicing, 3’ end cleavage and transcriptional control, are found to
bind hypo- and hyper-phosphorylated CTD (Emili et al., 2002; Liang and Lutz, 2006;
Rosonina et al., 2005). Cells expressing a CTD with only five repeats resulted in
inhibition of 3’ end processing in vivo, and in fact subunits of cleavage polyadenylation
specificity factor (CPSF) and cleavage stimulation factor (CstF) were found to bind the
wild type CTD (McCracken et al., 1997b). The same study also revealed that expression
of the CTD deletion mutant led to decreased transcription and splicing of reporter gene
constructs. Other 3’ processing factors have also been found to bind the CTD, such as
subunits of the cleavage/polyadenylation factor IA in yeast and the integrator complex
that mediates 3’ processing of snRNAs (Baillat et al., 2005; Barilla et al., 2001).
Unlike for capping enzymes or polyadenylation factors, there is mostly only
indicative data for the interaction of splicing factors with the CTD. The yeast U1
snRNP factor Prp40 has been found to bind to a phosphorylated CTD peptide in a Far
Western experiment, which supported the proposal that the Pol II CTD is generally
responsible for recruiting splicing factors to transcription units (Greenleaf, 1993; Morris
and Greenleaf, 2000). However, in vivo ChIP studies revealed that neither of the U1
snRNP proteins Prp42 and Prp40 is recruited to intronless genes in S. cerevisiae.
Moreover, deletion of the Prp40 protein domain responsible for the CTD interaction did
not interfere with Prp40 recruitment to intron-containing genes in yeast (Kotovic et al.,
2003 and K. Kotovic, unpublished results). These data suggest that co-transcriptional
Prp40 recruitment does not necessarily rely on interactions with the CTD, nevertheless
stabilizing interactions after U1 snRNP recruitment could be possible.
Part I: Introduction
19
Furthermore, it was shown that arginine-rich proteins, termed SR-like CTD
associated factors (SCAFs), interact with the CTD (Yuryev et al., 1996). The authors
furthermore showed that SR proteins co-immunoprecipitate with Pol II, which led to the
belief that SCAFs and SR proteins are recruited to the sites of transcription by Pol II.
However, for Western analysis of the immunoprecipitated complexes, the authors used
the 16H3 antibody that recognizes alternating arginines in SR proteins as well as other
arginine-rich proteins (Neugebauer et al., 1995, and K. Neugebauer, personal
communication), which does not allow the conclusion that SR proteins interact with the
CTD. Until now, it has not been investigated whether the SCAFs are splicing factors.
There is evidence that SR proteins reactive with the mAb 104 (which recognized
alternating arginine domains) co-immunoprecipitate with the phosphorylated Pol II
CTD in an RNA-independent manner (Kim et al., 1997). However, it should be
considered that the CTD antibody H5 that reacts with phosphorylated Serine 2 cross-
reacts with phosphorylated SR proteins and in turn an SR protein specific antibody
(mAb SC35) recognizes both Serine 2 and 5 phosphorylated CTD peptides (Doyle et
al., 2002). The possibility exists that also other antibodies raised against phosphorylated
SR proteins (like mAb 104) cross-react with Pol II CTD.
The splicing factor U2AF65 was found to co-immunoprecipitate with the hypo-
phosphorylated form of the CTD in an RNA-independent manner (Robert et al., 2002).
In vitro crosslinking studies revealed that U2AF65 binds directly to the nascent RNA as
it emerges from the putative exit groove 1 of Pol II during the transition from initiation
to elongation. The fact that the addition of recombinant U2AF65 assisted Pol II in
overcoming a pause site suggest that U2AF65 might be involved in Pol II elongation
(Ujvari and Luse, 2004; Ujvari and Luse, 2006). The authors proposed that the Pol II
subcomplex Rpb4/Rpb7 close to the exit groove 1 has a role in anchoring the RNA and
serving as a platform for the recruitment of RNA processing factors to the transcript.
Further studies were performed to identify additional CTD-binding proteins. A
recent proteomic analysis to identify CTD-binding proteins by phospho-CTD and Far
Western resulted only in a handful of proteins. When a synthetic CTD peptide with
three phosphorylated repeats was used for affinity chromatography, about 100 proteins
could be identified (Phatnani et al., 2004). However, careful investigation is needed to
determine whether the observed interactions support true functional relevance in vivo.
Part I: Introduction
20
Does Pol II couple splicing to transcription?
Notwithstanding the lack of evidence for direct binding of splicing factors to the CTD,
it is the prevailing model in the field that Pol II CTD binds splicing factors that can be
transferred to the nascent RNA when they are needed (Bentley, 2005). This theory is
based on several in vitro studies that demonstrated that the addition of Pol II or
phosphorylated CTD can stimulate splicing in vivo and in vitro (Fong and Bentley,
2001; Hirose et al., 1999; Zeng and Berget, 2000). However, whether these stimulatory
effects might come about due to direct interactions between the CTD and the splicing
machinery is not understood. ChIP experiments have shown that in yeast, an organism
where introns appear only in a small subset of genes, the U1, U2 and U5 snRNP as well
as BBP and the active spliceosome component, Prp19, are not recruited to intronless
genes in vivo whereas all these factors could be detected at intron-containing genes
(Gornemann et al., 2005; Kotovic et al., 2003; Lacadie and Rosbash, 2005). Mud 2, the
yeast homolog of U2AF65, on the other hand, has been found to associate to actively
transcribed intronless genes (Gornemann et al., 2005), supporting the observation made
in mammalian in vitro studies that U2AF65 can bind Pol II (Robert et al., 2002; Ujvari
and Luse, 2004; Ujvari and Luse, 2006). In the Balbiani Ring genes of C. tentans,
snRNPs are concentrated in intron-rich regions and are rare in intronless regions
(Kiseleva et al., 1994).
Taken together, the in vivo results suggest that snRNPs and Prp19 do not
necessarily travel with Pol II in every TU. It rather seems that signals in the nascent
RNA (i.e. splice sites) might be the determining factor that recruit splicing factors to the
site of transcription. Considering that only ~ 4 % of all protein coding genes in yeast
contain introns (Spingola et al., 1999), it makes sense that Pol II would not bind all
these factors if they were of no use in 96 % of the TUs. Mammalian genomes, however,
are different from the yeast genome; almost all mammalian genes contain introns. In
addition, the mammalian CTD is considerably longer and the spliceosome is more
complex compared to yeast, thus a general mechanism of recruiting splicing factors via
Pol II would be more reasonable. Although co-transcriptional splicing has been
documented in humans (Tennyson et al., 1995; Wuarin and Schibler, 1994), no in vivo
evidence of co-transcriptional splicing factor recruitment to mammalian genes has been
reported.
Part I: Introduction
21
Aim of part I of the thesis
For many years, the binding of splicing factors to pre-mRNA has been investigated by
in vitro splicing assays. In these experiments, in vitro transcribed pre-mRNA is added to
nuclear extracts, allowing the formation of splicing complexes. Unfortunately, splicing
is uncoupled from transcription in this experimental setup, and the question of
recruitment cannot be asked. Although it has been shown that purified CTD or Pol II
added to nuclear extracts stimulate splicing (Fong and Bentley, 2001; Hirose et al.,
1999; Zeng and Berget, 2000), the molecular mechanisms for this stimulation is
unknown and in vivo assays to directly investigate co-transcriptional splicing
requirements had to be developed.
Earlier work from our lab and others demonstrated that ChIP can be successfully
applied to examine the co-transcriptional recruitment of splicing factors to transcription
units in yeast (Gornemann et al., 2005; Kotovic et al., 2003; Lacadie and Rosbash,
2005). So far, splicing factor ChIP has only been achieved in S. cerevisiae where
endogenous RNA processing factors could be epitope-tagged to facilitate
immunoprecipitation. In mammalian cells, such tagging is more difficult to perform
without overexpressing the protein of interest. However, a vast number of antibodies
against endogenous mammalian splicing and RNA processing factors have been
generated over the years. The feasibility of the splicing factor ChIP assay in yeast
encouraged me to extend this approach to the mammalian system.
It is unknown how splicing factors are recruited to mammalian genes and to
what extend splicing occurs co-transcriptionally. Although widely assumed, there is no
satisfying in vivo evidence that Pol II is the determining factor in recruiting splicing
factors to mammalian genes. In my thesis I aimed to apply ChIP and related assays in
mammalian cells, to answer the following questions:
1. Are RNA processing factors recruited co-transcriptionally to mammalian
genes in vivo?
2. What is the dependence of RNA processing factor recruitment?
3. Are introns in human genes co-transcriptionally spliced in vivo and how is
this process linked to transcription?
Part I: Results
22
Results
Pol II and CBC80 accumulate on promoter-proximal positions of
constitutively active genes
For successful detection of co-transcriptionally bound splicing factors, a sufficient
amount of nascent RNA along the length of genes is required. It was crucial to
determine the accumulation of Pol II along actively transcribed genes, since the
abundance of Pol II at a given position in a gene directly correlates with the amount of
nascent RNPs. The highly expressed β-actin gene was chosen as a first attempt for Pol
II ChIP in mammalian cells. The fast growing C2C12 mouse myoblast cell line was
selected for initial ChIP studies with antibodies directed against the hypo-
phosphorylated form of Pol II. As expected, the Pol IIa form accumulated at the
promoter-proximal regions as it had been reported in other studies (Cheng and Sharp,
2003; Komarnitsky et al. 2000; Figure 6A). In order to examine total Pol II along the β-
actin gene, irrespective of the CTD phosphorylation state, ChIP with the Pol 3/3
antibody specific for the internal region F of Pol II large subunit was performed.
Surprisingly, the accumulation profile was very similar to that observed with the hypo-
phosphorylated Pol II, with total Pol II only poorly detectable in downstream regions
(Figure 6B). Previous studies of mammalian transcription units reported that Pol II is
particularly concentrated in promoter-proximal regions, suggesting that many
constitutively active genes may undergo promoter-proximal pausing (Boehm et al.,
2003; Brodsky et al., 2005; Cheng and Sharp, 2003). However, the observed Pol II
accumulation at promoters could also arise from Pol II molecules in pre-initiation
complexes.
The cap-binding complex (CBC) is bound co-transcriptionally in yeast and
insect cells (Visa et al., 1996; Zenklusen et al., 2002). It has not been reported that the
CBC binds to Pol II (see also Figure 15 and 16), so co-transcriptional detection of CBC
would indicate promoter-proximal paused nascent RNPs. In order to test if the CBC can
be detected at the mouse β-actin TU, ChIP with an antibody against the CBP80 subunit
of the CBC was performed (Figure 6B). Indeed, CBP80 is concentrated at the promoter
region, indicating that CBP80 is co-transcriptionally bound to the nascent RNA,
Part I: Results
23
however detectability decreases strongly in downstream regions of the gene, paralleling
the Pol II distribution.
Figure 6. Pol II and CBC accumulation along the β-actin gene.Schematics representing the β-actin gene. Black lines indicate the regions amplified by primer sets,identified by the nucleotide at the center of the amplified region. In the panels aligned immediatelybelow, histogram bars are placed according to the positions of the PCR products along each gene. Allvalues are relative to non-immune background ChIP experiments and normalized to an upstream controlregion. ChIP analysis was done in C2C12 mouse myoblasts, using antibodies specific for hypo-phosphorylated Pol II (8WG16; Pol IIa; A) or CBP80 and the internal region F of the Pol II large subunit(B). The accumulation profile reveals that the vast majority of Pol II and CBC are bound to the 5’ regionof the gene. The data represent the average of at least three independent experiments. Upper panel: PolIIa, n=5. Lower panel: Pol II, n=4. CBP80, n=3. Error bars represent the SEM.
Part I: Results
24
Figure 7. Accumulation pattern of Pol II and CBP80 along actively transcribed genes indicatespromoter-proximal pausing.Schematics of constitutively expressed genes (PGK1, LDHA and histone H2A.m) along with PCR primerpositions are shown, corresponding results of ChIP experiments are shown as histograms (upper panels,lower left panel). All values are relative to non-immune background ChIP experiments and normalized tointergenic control region set to 1. ChIP analysis was done in human A431 cells with antibodies specificfor total Pol II and CBP80. Lower right panel: Comparison of the results for all three genes tested,including β-actin from Figure 6 showing the dependence of detectability of Pol II and CBP80 on distancefrom the promoters. Levels at the most promoter-proximal regions were set to 100%. Interrupted linesrepresent the intergenic values. The data represent the average of at least three independentexperiments. LDHA: Pol II, n=4; CBP80, n=4. histone H2A.m: Pol II, n = 3; CBP80, n = 3. PGK1: Pol II,n=3; CBP80, n=3. Error bars represent the SEM.
Part I: Results
25
The fact that Pol II and CBP80 are almost undetectable at downstream gene regions
(although β-actin is highly expressed) clearly suggests that Pol II pauses shortly after
transcription initiation. To further address this hypothesis, the distribution of Pol II and
CBP80 was tested along several other constitutively active human genes by ChIP.
Figure 7 shows that both Pol II and CBC were concentrated in promoter-proximal
regions of the histone H2A.m, PGK1 and LDHA genes. Detectability of both Pol II and
CBP80 decreased in parallel in downstream regions of each gene, reaching intergenic
levels ~1kb downstream of the transcriptional start site (see Figure 7, lower right). This
provides direct evidence for Pol II undergoing promoter-proximal pausing of
constitutively active genes.
Pol II and CBP80 accumulate on 5’ ends of uninduced c-fos and HSP70 but
are robustly detectable throughout the induced genes
It has been reported that during the transcription of some cellular (c-fos, c-myc, c-myb,
HSP70) and viral genes, Pol II elongation is blocked immediately downstream of the
promoter (Lis, 1998). It is common understanding that promoter-proximal pausing
comprises a regulatory step of gene expression, however the molecular mechanisms of
pausing regulation are poorly understood. To determine whether the distribution of
CBC and Pol II on LDHA, PGK1, histone H2A.m and β-actin resembles bona fide
promoter-proximal pausing, the c-fos and HSP70 genes were analyzed by ChIP.
Pausing is expected to occur in both genes after 20-40 nucleotides of transcription when
uninduced (Boehm et al., 2003; Fivaz et al., 2000; Rasmussen and Lis, 1993), and each
can be induced with specific treatments: c-fos transcription is very rapidly induced with
the calcium ionophore A23187 (Bravo et al., 1985), while HSP70 is inducible with
sodium arsenite (Yih et al., 2002). In uninduced A431 cells, hypo-phosphorylated and
total Pol II and CBP80 were robustly detectable at the promoters of both genes and
undetectable in downstream regions (Figure 8). This shows that the CBC is co-
transcriptionally recruited to very short nascent RNAs and is consistent with the
interpretation of the previous results that CBC concentration at the promoters of
constitutively active genes reflects promoter-proximal pausing.
Upon HSP70 gene induction with sodium arsenite, both Pol II and CBP80 become
detectable in downstream gene regions (Figure 8 right panel). Similar results were
obtained upon c-fos induction with calcium ionophore (Figure 8 left panel). In
Part I: Results
26
agreement with others (Brodsky et al., 2005; Cheng and Sharp, 2003), the Pol 3/3
antibody specific for a non-CTD Pol II epitope was more effective in ChIP of
downstream regions than the hypo-phosphorylated CTD-specific antibody 8WG16,
likely due to phosphorylation of the CTD during initiation and elongation.
The downstream accumulation of Pol II and consequently nascent RNA is a prerequisite
for studying co-transcriptional recruitment of nascent RNA binding proteins like the
CBC. Interestingly, detection of CBP80 in downstream regions did not parallel the Pol
II distribution, suggesting that as the nascent RNP grows co-transcriptionally, CBP80
epitopes may become unavailable for antibody binding. This indicates that epitope
availability could be an issue, particularly in downstream regions, when assaying RNA
processing factors that associate later to the nascent RNA.
Establishing splicing factor ChIP in mammalian cells
The finding that Pol II and CBC do not accumulate at equal levels along constitutively
active mammalian genes is in striking contrast to the results obtained in yeast, where
Pol II is robustly detectable throughout the genes (Gornemann et al., 2005; Kotovic et
al., 2003). For successful detection of splicing factors, however, a sufficient amount of
nascent RNA along the length of genes is required. Furthermore, the epitopes must be
available for antibody binding after formaldehyde crosslinking. Since stable epitope
tagging is more difficult to achieve in mammalian cells than in yeast, I concentrated on
finding antibodies specific for endogenous splicing factors that work in ChIP. All
antibodies tested in mammalian ChIP are assembled in Table 1. During the course of
optimizing the protocol it became apparent that enormous amounts of antibodies were
needed for splicing factor ChIP. I normally used 30 µg for one ChIP experiment, which
is much more than the 8 µg of mAbs against the HA epitope used in yeast splicing
factor ChIP. Not all antibodies working in ChIP are available in unlimited amounts in
our lab, but were also provided from other labs in form of polyclonal rabbit antibodies,
which imposed limitations to particular experiments. The recovered DNA fragments
from ChIP experiments with specific and unspecific antibodies were detected by Real
Time PCR, which has the advantage over conventional PCR that even minor differences
in template amounts can be successfully resolved in the linear range.
Part I: Results
27
Figure 8. Induction of c-fos and HSP70 transcription leads to downstream accumulation ofRNA Pol II, CBP80, and hnRNP A1.Schematics of c-fos and HSP70, along with PCR primer positions are shown. PCR products are identifiedby the nucleotide at the center of the amplified region. C-fos transcription was induced in A431 cells with5 µM calcium ionophore A23187 in DMSO for 15 min (red bars), control cells were treated with DMSOalone (blue bars). HSP70 gene transcription was induced with 250 µM sodium arsenite for 1 h (red bars)and control cells were untreated (blue bars). All values are relative to non-immune background ChIPexperiments and normalized to an intergenic control region set as 1. Accumulation profile ofunphosphorylated Pol II (8WG16; Pol IIa) or total Pol II (Pol3/3) on c-fos and HSP70 reveals paused PolII at promoter regions on genes before induction and no accumulation on downstream regions. Upongene induction, Pol II accumulates on downstream regions. CBP80 accumulates solely on promoterregions when genes are uninduced and is detectable in downstream gene regions upon induction. HnRNPA1 accumulates co-transcriptionally on the induced c-fos gene but not significantly on induced HSP70.No hnRNP A1 accumulation is detected on uninduced genes.The data represent the average of at least three independent experiments. Number of experiments:uninduced; induced. C-fos: n = 4; 4. Pol II, n = 3; 3. CBP80 n = 4; 3. hnRNP A1, n = 4; 3. HSP70: PolIIa, n = 4; 4. Pol II, n = 4; 8. CBP80, n = 3; 5. hnRNP A1, n = 3; 4. Error bars represent the SEM.
Part I: Results
28
Antibody Epitope/protein Class Source Works Referencesc-899 (N20) Pol II Rpb1 N-term rabbit Santa Cruz yessc-17798(A10)
Pol II Rbp1 1-224 IgG2b Santa Cruz no
Pol3/3 Rbp1 internal regionF
IgG D. Eick yes (Kramer et al., 1980)
8WG16 Pol II Rpb1 CTD IgG2a Neoclone yes (Thompson et al.,1989)
H4 Acetylated HistoneH4
rabbit Upstate yes
α-CBP80 CBP80 rabbit E. Izaurralde yes (Izaurralde et al.,1994)
α-TMG 2,2,7-trimethylguanosine
IgG1 Oncogene no (Krainer, 1988)
Y12 Sm proteins IgG3 no (Lerner et al., 1981)MC3 U2AF65 IgG2b M. Carmo-
Fonsecayes (Gama-Carvalho et
al., 1997)PTB clone 1 C-term PTB IgG1 Zymed no (Chou et al., 2000)PTB clone 3 N-term PTB IgG1 Zymed no (Chou et al., 2000)BB7 PTB IgG2b D. Black no (Chou et al., 2000)hnRNP A1 hnRNP A1 ascites D. Black yes1H4 SR proteins IgG1 Zymed no (Neugebauer and
Roth, 1997)16H3 Alt. arginine repeat
domainIgG1 Zymed no (Neugebauer et al.,
1995)7B4 SRp20 linker region IgG1 Zymed no (Neugebauer and
Roth, 1997)SF2/ASFclone 103
N-te rm (RRM1)SF2/ASF
IgG1 Zymed no (Hanamura et al.,1998)
SF2/ASFclone 96
N- te rm (RRM1)SF2/ASF
IgG2b Zymed no (Hanamura et al.,1998)
CPSF 100 CPSF 100 kDasubunit
mAb W. Keller yes (Jenny et al., 1994)
PSF (A20) Internal PSF goat Santa Cruz no (Patton et al., 1993)U5-116k U5 snRNP 116 kDa
protein, N-termrabbit R. Lührmann yes (Fabrizio et al., 1997)
U5-40k U5 snRNP 40 kDaprotein
rabbit R. Lührmann no (Achsel et al., 1998)
U4/U6- 61k U4/U6 snRNP 61kDa protein
rabbit R. Lührmann no (Makarova et al.,2002)
U1 U1 snRNP 70kDaprotein
ascites D. Black yes
U1 U1 snRNP 70kDaprotein
hybridoma sup.
D. Black yes
SF1 SF1 mAb A. Krämer no (Tanackovic andKramer, 2005)
SF3a66 SF3a66 mAb A. Krämer no (Brosi et al., 1993)
Table 1. List of ChIP-tested antibodies.
Part I: Results
29
Co-transcriptional detection of hnRNP A1
To begin establishing splicing factor ChIP, an antibody specific for hnRNPA1 was
used. This protein is abundant in the nucleus and takes part in transcription, pre-mRNA
splicing and nuclear export (Dreyfuss et al., 2002). HnRNP A1 was robustly detectable
at c-fos in induced but not in un-induced cells, indicating that hnRNP A1 association
with c-fos is transcription-dependent (Figure 8 lower left panel). However, hnRNP A1
was not detectable on HSP70 with or without transcriptional activity, suggesting that
hnRNP A1 associates with c-fos but not HSP70 nascent RNA (Figure 8 lower rightpanel). Although hnRNP A1 binding sites do not conform to a strict consensus (Burdand Dreyfuss, 1994; Hutchison et al., 2002; Zhu et al., 2001), scanning of c-fos pre-
mRNA sequence revealed at least four potential hnRNP A1 binding sites
(UAGNNNUAG or UAGGGA) with the first occurring in intron 1. HSP70 mRNA did
not contain any putative hnRNP A1 binding sites, suggesting that the detection of
hnRNP A1 on c-fos may reflect a direct binding to c-fos nascent RNA. Finally,
detection of hnRNPA1 did decrease in downstream regions of c-fos; this could be due to
either a decrease in epitope availability or a loss of hnRNP A1 as the nascent RNP
matures. Taken together, the results of the gene induction experiments indicate that co-
transcriptional association of RNA splicing factors can indeed be detected by ChIP.
Splicing factors accumulate on the 5’ end of β-actin
In order to investigate whether splicing factors can be detected at the β-actin gene,
antibodies specific for the U1 70K component of the U1 snRNP and U2AF65 were used
in ChIPs of normally growing C2C12 cells. As shown in Figure 9A, significant
accumulation of these factors could be detected on the promoter-proximal position,
however only U2AF65 showed significant accumulation on downstream regions. Since
Pol II and CBB80 were poorly detectable in downstream regions (see Figure 6), it is a
fair assumption that the majority of all nascent RNPs are concentrated in the 5’ region
of β-actin. The promoter-proximal primer pair with the center point at 71 bp is
overlapping with intron 1 sequence, which could account for U1 70k accumulation due
to the presence of the 5’ splice site in the nascent RNA. However, the polypyrimidine
binding tract is not synthesized at that early position in the gene, so U2AF65
accumulation could be explained by its recruitment to the TU by Pol II as previously
Part I: Results
30
suggested (Robert et al., 2002; Ujvari and Luse, 2004; Ujvari and Luse, 2006).
Alternatively, U2AF65 could be bound loosely to suboptimal polypyrimidine stretches
in the first intron and would thereby be captured by ChIP.
Figure 9. Splicing factor accumulation on the β-actin and histone H4 genes.Schematics of β-actin and histone H4, along with PCR primer positions are shown. ChIP analysis wasdone in C2C12 mouse myoblasts, using antibodies specific for splicing factors U1 70k and U2AF65 (A andC) or total Pol II and CBP80 (B). Splicing factors accumulate significantly on the promoter-proximalregion of β-actin but not on the intronless, constitutively expressed histone H4 gene. In the upper right,an illustration of β-actin exon1 (black capitals) and intron1 (grey lower case) 5’ to 3’ sequence is givenwith polypyrimidine stretches in red.The data represent the average of at least three independentexperiments, except lower left panel, where one representative experiment is shown. Upper panel: U170k, n=6. U2AF65, n=5. Lower panels: Pol II, n=1. CBP80, n=1. U1 70k, n=3. U2AF65, n=3. Error barsrepresent the SEM.
Part I: Results
31
Scanning of the first β-actin intron for polypyrimidine stretches revealed that indeed a
CU rich region is present in the beginning of intron 1 (Figure 9A right).
Furthermore, U1 70k and U2AF65 accumulation on the constitutively transcribed
but intronless murine histone H4 gene was investigated. Pol II and CBP80 accumulate
considerably on the histone H4 5’ end which suggests the presence of nascent RNPs
(Figure 9C). However, U1 70k and U2AF65 do not significantly accumulate on this gene
(Figure 9D), indicating that the presence of Pol II, nascent RNA and the CBC is not
sufficient for splicing factor accumulation on actively transcribed genes.
Splicing factor accumulation on c-fos is dependent on transcription and can
be enhanced by camptothecin
In order to test if pre-mRNA splicing factors accumulate on c-fos with a promoter-
proximal paused or elongating Pol II, the uninduced (DMSO treated) and induced c-fos
gene was investigated with antibodies specific for U1 70K, U2AF65, and the 116K
component of the U5 snRNP. None of these three factors was detectable on c-fos in
uninduced cells at any position (Figure 10A left panel). Because CBC and Pol II are
abundant in the promoter proximal region under these conditions (see Figure 8), this
suggests that the U1 and U5 snRNPs and U2AF65 do not detectably associate with the
paused polymerase, the CBC, or the short nascent transcript. However, upon
transcriptional induction, all three factors were significantly detectable throughout the
transcription unit, with the exception that U2AF65 was not detectable in the promoter-
proximal region corresponding to exon 1. None of these factors accumulated on the
intronless histone H2A.m gene, which is constitutively active (Figure 10A right panel).
The c-fos gene is a target of the drug camptothecin, which inhibits
topoisomerase I when complexed with DNA (Hsiang et al., 1985). More than 30 sites of
topoisomerase I activity within the c-fos gene have been mapped with camptothecin,
which inhibits c-fos transcription by stalling the progress of Pol II through the gene
(Ljungman and Hanawalt, 1996; Stewart et al., 1990). Pol II and CBC ChIPs in cells
treated with a short pulse of camptothecin in combination with induction indicate that
nascent RNPs do not “pile up” at internal gene regions but instead are trapped in
positions characteristic of induction alone (Figure 10B). Therefore, the possibility was
considered that camptothecin treatment might amplify splicing factor signals, by
stalling nascent RNPs and giving the nascent RNA more
Part I: Results
32
Figure 10A. Splicing factors accumulate co-transcriptionally on the intron-containing c-fosgene.Figure text see Figure 11B.
Part I: Results
33
Figure 10B. Pol II and CBC accumulation at the intron-containing c-fos gene.Distribution of total Pol II (Pol3/3) and CBP80 (figure 11b): U1 70K, U2AF65, and the 116K component ofthe U5 snRNP (figure 11a) on c-fos and histone H2A.m genes, under various conditions. A431 cells wereeither treated with DMSO (blue bars), calcium ionophore (red bars) or calcium ionophore andsubsequently camptothecin (yellow bars). Schematics of c-fos and histone H2A.m are shown abovecorresponding data panels. Black lines indicate the gene regions amplified by primer sets, identified bythe nucleotide at the center of the amplified region. All values are relative to non-immune backgroundChIP experiments and normalized to an intergenic control region set as 1. The data represent theaverage of at least three independent experiments. Number of experiments denoted: DMSO; calciumionophore; calcium ionophore + camptothecin. C-fos: Pol II, n = 3; 4; 5. CBP80, n = 4; 3; 3. U1 70K, n= 3; 4; 3. U2AF65, n = 4; 7; 4. U5 116K, n = 3; 4; 3. Histone H2A.m: Pol II, n = 3; 4. CBP80, n = 4; 3.U1 70K, n = 3; 4. U2AF65, n = 4; 3. U5 116K, n = 4; 4. Error bars represent the SEM.
Figure 11. c- myc and c-fos are spliced upon calcium ionophore and camptothecin treatment.A431 cells were either treated with DMSO, calcium ionophore or calcium ionophore and subsequentlycamptothecin, RNA was extracted and treated with DNAse I. Gene-specific primers were used for reversetranscription and spliced and unspliced message was PCR amplified. The PCR products were run on a 2%agarose gel, band intensities were measured and the ratio of spliced/unspliced message was calculated.
Part I: Results
34
time to bind splicing factors. Indeed, camptothecin treatment resulted in enhanced
splicing factor signals along c-fos but not on histone H2A.m, indicating that
camptothecin treatment did not artificially induce splicing factor association with
intronless RNA (Figure 10A right panel). C-fos and c-myc splicing after the different
induction treatments was analyzed to confirm that splicing per se was unchanged
(Figure 11).
Examination of the distributions of U1 snRNP, U2AF65 and U5 snRNP in the
presence or absence of camptothecin indicates that the intron/exon structure of the
nascent c-fos RNA likely plays a role in the transcription-dependent accumulation of
these factors on the gene. Although the U1 and U5 snRNPs were only modestly
detectable within the induced c-fos gene, levels were significantly elevated upon
camptothecin treatment at the first two positions tested, in exon 1 and intron1. Taking
into account a probable loss of detectability in downstream regions, the data suggest
that U1 and U5 snRNPs accumulate on c-fos at the earliest measurable position. This is
consistent with co-transcriptional U1 snRNP recruitment to the 5’ splice site, since the
5’ splice site of intron1 is close to the exon1 primer set and cannot be resolved from the
exon1 within the limits of the assay. Surprisingly, the U5 snRNP also accumulates at
this early position, since the 3’ splice site, thought to be required for U5 snRNP
recruitment to pre-mRNA by the step-wise model of spliceosome assembly, is ~1kb
downstream of the exon1 region. One possible explanation for these observations is that
splicing factors may be recruited to the transcription unit by elongating Pol II, as
previously suggested (Goldstrohm et al., 2001; Greenleaf, 1993).
Splicing factors do not accumulate on intronless HSP70
If elongating Pol II is generally recruiting splicing factors to transcription units, they
should also be detectable at actively transcribed intronless genes. The HSP70 gene,
which is also a camptothecin target with multiple topoisomerase I sensitive sites
distributed throughout the gene (Collins et al., 2001; Kroeger and Rowe, 1992) serves
as a robust test for splicing factor recruitment to an intronless gene. As shown in Figure
12A, HSP70 exhibits a massive downstream concentration of RNA Pol II upon
induction with arsenite. The splicing factor U2AF65 as well as the U1 and U5 snRNPs
were not significantly detected at any position along induced HSP70. Additional
treatment with camptothecin did also not lead to an accumulation of splicing factors
Part I: Results
35
(Figure 12B). In order to verify that arsenite treatment did not impair splicing, the ratio
of spliced versus unspliced c-myc pre-mRNA was quantitated by reverse transcription
(RT)-PCR (Figure 13).
Arsenite treatment induces oxidative stress in cells, and stress conditions can
cause RNA binding proteins, such as hnRNP A1 to concentrate in cytoplasmic stress
granules (SGs; (Kedersha et al., 2005; van der Houven van Oordt et al., 2000). SGs
contain mRNA encoding most cellular proteins, but exclude mRNAs encoding for heat
shock proteins (Kedersha and Anderson, 2002) and are thought to be triage centers that
sort, remodel and export specific mRNA transcripts for reinitiation, decay or storage
(Anderson and Kedersha, 2006). To determine whether the failure of splicing factor
accumulation on HSP70 after arsenite treatment is due to their relocalization, the
subcellular distribution of splicing factors upon arsenite treatment was tested.
Immunostaining revealed that the U1 70K protein, U2AF65, snRNAs in general, and
hnRNP A1 are distributed throughout the nucleoplasm of A431 cell nuclei (Figure
14A). The relative intensity of nuclear staining was unchanged upon arsenite treatment,
with the exception that a small proportion of the total hnRNP A1 signal became
detectable in cytoplasmic foci. The cytoplasmic foci of hnRNP A1 were shown to be
SGs, by double immunofluorescence with an antibody against TIA-1, an SG marker
protein (Figure 14B). Nevertheless, splicing factors remained predominantly nuclear
after arsenite treatment. Therefore, HSP70 induction represents a valid test for splicing
factor recruitment to an induced, intronless gene. Like the intronless histone H2A.m
transcription unit, HSP70 fails to accumulate splicing factors. Taken together, the data
suggest that introns are necessary for the co-transcriptional accumulation of the splicing
factors examined here.
Part I: Results
36
Figure 12. Splicing factors do not accumulate on induced HSP70, in the presence or absenceof camptothecin.Accumulation profile of total Pol II (Pol 3/3; dark blue bars), U1 70k (red bars), U2AF65 (light blue bars)and U5 116k (green bars) along induced HSP70. A431 cells were treated for 1 h with sodium arsenite(A) or 45 min + additional 15 min with camptothecin (B), prior to crosslinking and ChIP. All values arerelative to non-immune background ChIP experiments and normalized to an intergenic control region setas 1.The data represent the average of at least three independent experiments. Upper panel: Pol II, n =8. U1 70K, n = 3. U2AF65, n = 5. U5 116K, n = 3. Lower panel: Pol II, n = 3. U1 70K, n = 3. U2AF65, n =3. U5 116K, n = 3. Error bars represent the SEM.
Figure 13. C-myc is spliced upon arsenite treatment.A431 cells were treated with sodium arsenite, RNA was extracted and treated with DNAse I. Gene-specific primers were used for reverse transcription and spliced and unspliced message was PCRamplified. The PCR products were run on a 2% agarose gel, band intensities were measured and theratio of spliced/unspliced message was calculated.
Part I: Results
37
Figure 14. Sodium Arsenite treatment does not alter the nuclear localization of U1 70k,U2AF65 and snRNAs.(A) Immunostaining with antibodies specific for U1 70K, U2AF65, TMG cap (snRNAs), and hnRNP A1shows that these factors are distributed throughout the nucleoplasm of A431 cell nuclei. (B) The relativeintensity of nuclear staining was unchanged upon arsenite treatment, with the exception that the hnRNPA1 protein became detectable in cytoplasmic foci, which were shown to be stress granules (SGs), bydouble staining with an antibody against TIA-1, an SG marker protein (in collaboration with K.Neugebauer).
Part I: Results
38
Interactions between splicing factors and Pol II
To investigate the possibility that the splicing factors examined bind to Pol II, two
standard co-immunoprecipitation protocols were carried out that do not involve
crosslinking. In the first approach, A431 cells were metabolically labeled with 32P-
orthophosphate; immunoprecipitates were analyzed by fluorography for the presence of
specific snRNAs (Figure 15). As expected, α-U1 70K pulled down only the U1 snRNA,
and α-U5 116K pulled down U4, U5, and U6 snRNAs. α-U2AF65 pulled down lower
but significant amounts of U2, U4, U5, and U6 snRNAs, consistent with the presence of
U2AF65 in spliceosomes (Zhou et al., 2002). In contrast, antibodies specific for CBP80,
the internal region of Pol II, and the hypo-phosphorylated or hyper-phosphorylated Pol
II CTD did not pull down any detectable snRNA. Thus, interactions between snRNPs
and Pol II or CBC were not detected.
In the second approach, α-CBP80, α-U1 70K, α-U2AF65, and α-U5 116K were
used to immunoprecipitate unlabeled A431 cell extracts; phosphorylated forms of Po II
large subunit in each immunoprecipitate were detected by Western blotting, using the
H14 and H5 monoclonal antibodies against Pol II CTD repeats predominantly
phosphorylated on Serine 2 and Serine 5 residues, respectively. Neither form of Pol II
was detected in CBP80, U1 70K, or U5-116K immunoprecipitates (Figure 16). Taken
together with the results of the metabolic labeling experiment, these data do not support
a strong physical interaction between Pol II and the CBC, the U1 snRNP or the U5
snRNP. However, Pol II did co-immunoprecipitate with U2AF65 (Figure 16 lower
right). Interestingly, RNase A treatment prior to immunoprecipitation nearly abolished
detection of Pol II, suggesting that the association between active Pol II and U2AF65 is,
at least in part, mediated by RNA.
Part I: Results
39
Figure 15. Splicing factors do not co-immunoprecipitate with hypo-phosphorylated Pol II.Immunoprecipitation of metabolically labeled snRNAs from 32P-orthophosphate-labeled A431 cells withY12 (α -Sm ) as a positive control, α -U1 70k, α-U5 116k, α-U2AF65, α-CBP80, α-Pol II hypo-phosphorylated CTD (8WG16), and α-Pol II internal region (Pol3/3). Samples were run on a 10% ureagel and detected with a phosphorimager.
Figure 16. Spicing factors and CBP 80 do not associate co-immunoprecipitate withphosphorylated Pol II.Extracts of A431 cells were immunoprecipitated with antibodies indicated above each panel underidentical conditions to those in Figure 16. The precipitate and 1% of the input material was analyzed byWestern blot, using antibodies specific for Ser2 or Ser5 phosphorylated forms of the Pol II CTD (mAbsH5 and H14, respectively). In the U2AF65 experiment, samples were treated with or without RNAse Aprior to immunoprecipitation. Instead of input, 0.3% of supernatant is analyzed by Western blot tocontrol for possible effects of the additional incubation on phosphoepitope availability.
Part I: Results
40
Development of Chromatin-RNA-Immunprecipitation (ChRIP)
The observation that splicing factors, including the U5 116K protein most likely
recruited in the U4/U6•U5 tri-snRNP active in splicing, accumulate on induced c-fos
suggests that splicing itself may also occur co-transcriptionally. To directly monitor co-
transcriptional splicing, I designed a new assay termed Chromatin-RNA-
Immunprecipitation (ChRIP), in which active chromatin and the attached nascent RNA
was immunopurified with antibodies specific for acetylated histone 4 (AcH4; Figure
17). Acetylated histones were robustly detectable within c-fos by ChIP (Figure 18). As
background control for ChRIP, unspecific mouse IgG was used for
immunoprecipitation. The purified RNA from AcH4 and unspecific IgG
immunoprecipitations was excessively treated with DNAse I and subjected to gene-
specific RT with reverse primers located in the intron 2 or exon 4 of c-fos. The cDNA
and -RT controls were then amplified by Real Time PCR with primers amplifying either
spliced exon1/exon2 and intron 1 (from cDNA generated with intron 2 reverse primer)
or spliced exon3/exon4 and intron 3 (from cDNA generated with exon 4 reverse
primer). Spliced cDNA generated from unspecific IgG immunopurified RNA was on
average 8 times less abundant than cDNA from AcH4 immunopurified RNA (see
Figure 19). To determine whether fully processed, polyadenylated mRNA could
contaminate the beads, RNA purified with anti-AcH4 and unspecific IgG was subjected
to reverse transcription with an oligo(dT) primer and Real Time PCR amplification with
a primer pair amplifying c-fos exon 4. Equal amounts of exon 4 were recovered with
specific and unspecific antibody, indicating that cytoplasmic mRNA is not
contaminating the beads.
ChRIP was applied to cross-linked extracts of A431 cells induced with calcium
ionophore, revealing that both spliced and unspliced c-fos RNA was chromatin
associated. The ratio of the signals for spliced versus unspliced RNA at exon1/exon2
was ~17 upon induction, while the analogous ratio for exon3-exon4 splicing was ~45
(Figure 20). Taken together, this data suggest that c-fos splicing occurs co-
transcriptionally and that ChRIP can be used as a new tool to directly observe co-
transcriptional splicing in mammalian cells.
Part I: Results
41
Figure 17. Chromatin-RNA immunoprecipitation (ChRIP).Schematic of the experimental approach, illustrating the expectation that antibodies against acetylatedHistone H4 (AcH4) pull down nascent c-fos RNA, attached to the chromatin through Pol II.
Figure 18. Acetylated histone 4 (AcH4) distribution along c-fos by ChIP.Schematics of c-fos, along with PCR primer positions are shown. PCR products are identified by thenucleotide at the center of the amplified region. A431 cells were either treated with calcium ionophore(red bars) or calcium ionophore and subsequently camptothecin (red bars). ChIP was performed withanti acetylated histone H4 (AcH4). All values are relative to non-immune background and normalized toan intergenic control region set as 1. Accumulation profile of AcH4 on c-fos reveals that histone H4 isacetylated throughout the gene in both conditions. Number of experiments denoted: calcium ionophore;calcium ionophore + camptothecin. AcH4, n = 3; 3. Error bars represent the SEM.
Part I: Results
42
Figure 19. Characterization of the ChRIP assay.(A) Comparison of raw Real Time PCR results from anti-AcH4 ChRIPs, using primers specific for spliced(blue plot) and unspliced (red plot) c-fos cDNA. Amplification plots are displayed as emitted fluorescenceversus cycle number with the threshold fluorescence indicator line (blue horizontal line). The differencebetween the individual threshold cycles (Cts; the fractional cycle number at which the fluorescencepasses the fixed threshold) of the spliced c-fos plot and the unspliced c-fos plot is greater forcamptothecin treatment, indicating a decrease in the relative abundance of unspliced c-fos messageattached to chromatin. Reactions carried out in the absence of reverse transcription (-RT, green plot)indicate that the signals are not due to DNA contamination of the samples. A representative experimentis shown. (B) Comparison of raw real-time PCR results for anti-AcH4 (blue plot) and non-immune IgG(red plot) ChRIPs, using primers specific for spliced c-fos cDNA. The difference between the individualCts of anti-AcH4 ChRIP plots from the non-immune IgG ChRIP plots indicates a 11.9 and 11.4 foldenrichment of c-fos nascent RNA over background. A representative experiment is shown and results forex3/ex4 primers were similar. (C) Comparison of raw real-time PCR results for anti-AcH4 (blue plot) andnon-immune IgG (red plot) ChRIPs, using primers specific for c-fos exon 4, following reversetranscription with oligo dT. The similarity of the anti-AcH4 ChRIP amplification plot from the non-immuneIgG ChRIP amplification plot indicates that poly-adenylated c-fos mRNA is not trapped on the cross-linked chromatin.
Part I: Results
43
Figure 20. Co-transcriptional c-fos pre-mRNA splicing detected by ChRIP is enhanced bycamptothecin.The ratio spliced vs. unspliced c-fos intron 1 was determined with RT primers in intron 2 and PCRprimers amplifying either exon1-exon2 or intron 1. The ratio spliced vs. unspliced c-fos intron 3 wasdetermined with RT primers in exon 4 and PCR primers amplifying either exon3-exon4 or exon3-intron3.The data represent the average of at least three independent experiments, in which cells were treatedwith calcium ionophore for 15 minutes, followed by an additional 15 minutes of ionophore with orwithout camptothecin. Error bars represent the SEM.
Co-transcriptional splicing of c-fos is enhanced by camptothecin
Since camptothecin treatment amplified splicing factor signals by ChIP, probably by
stalling Pol II and the nascent RNPs (see Figure Figure 10A left panel), it was tempting
to find out if camptothecin enhanced c-fos co-transcriptional splicing. Acetylated
histones were robustly detectable within c-fos by ChIP also after camptothecin
treatment (Figure 18). ChRIP in A431 cells induced with calcium ionophore and
camptothecin, revealed that camptothecin treatment led to significant increases in the
level of co-transcriptional splicing for both introns, with ratios of spliced/unspliced
signals 3- and 2-fold higher than in calcium induction alone (Figure 20). These data
suggest that co-transcriptional c-fos splicing is promoted by stalling the progress of Pol
II with camptothecin, thus providing direct evidence that co-transcriptional spliceosome
assembly and splicing is influenced by the kinetics of transcription by Pol II.
Part I: Discussion
44
Discussion
The in vivo studies of co-transcriptional RNA processing events had been possible only
in special biological systems such as in the Balbiani ring genes in C. tentans or the
chorion genes of D. melanogaster (Neugebauer, 2002). Only recently, work from our
lab and others showed that a modified ChIP method can be used to detect
simultaneously Pol II and RNA-processing factors at the sites of transcription in yeast
(Gornemann et al., 2005; Kotovic et al., 2003; Lacadie and Rosbash, 2005). From
theses studies, a detailed picture of co-transcriptional splicing factor assembly was
obtained. However, the yeast genome is rather simple compared to the complex
genomes of mammalians. Introns are only present in ∼4 % of the yeast genes (Spingola
et al., 1999), whereas almost all mammalian genes contain multiple introns and up to 70
% of all genes undergo alternative splicing (Black, 2003). My PhD work was aimed
towards understanding the requirements of co-transcriptional splicing and splicing
factor recruitment in mammalian cells. For this purpose, I set up the splicing factor
ChIP method for use in mammalian cells and established experimental conditions for
monitoring co-transcriptional RNA splicing events.
The abundance of Pol II at a given position in a gene is directly correlated with
the amount of nascent RNA and therefore I started to evaluate the Pol II distribution at
constitutively active genes by ChIP. As expected, the hypo-phosphorylated (Pol IIa)
form accumulated on the 5’ ends of the constitutive gene β-actin. However, ChIP with
an antibody specific for a non-CTD epitope of Pol II revealed the same 5’ end
accumulation on β-actin and other constitutively expressed genes such as, LDHA,
PGK1 and histone H2A.m. Pol II accumulation on 5’ ends in mammalian genes had
been previously observed by other studies, suggesting promoter-proximal pausing of
Pol II (Brodsky et al., 2005; Cheng and Sharp, 2003). In these studies, ChIP was applied
to map Pol II along the length of human genes with antibodies specific for either the Pol
IIa form or the total Pol II. However, if Pol II accumulation on 5’ ends is due to pausing
shortly after transcription initiation or if Pol II abounds in pre-initiation complexes
could not be resolved satisfactory.
I could address this question by applying ChIP to detect the co-transcriptional
accumulation of the CBC. The CBC binds the 7-methyl-guanosine cap at the 5' end of
Part I: Discussion
45
Pol II transcripts co-transcriptionally in insect cells and yeast (Gornemann et al., 2005;
Visa et al., 1996; Zenklusen et al., 2002), and has not been reported to be bound directly
by Pol II (see also Figure 15+16). Thus, co-transcriptional detection of the CBC
provides an excellent marker for the formation of nascent RNPs and consequently
successful transcription initiation. ChIP with an antibody specific for CBP80, a subunit
of the CBC, resolved that the CBC followed the Pol II accumulation profile at the
constitutive active genes β-actin, LDHA, PGK1 and histone H2A.m. These results
provide direct evidence that Pol II indeed is paused at promoter-proximal position in
these genes.
I could demonstrate that CBP80 binds co-transcriptionally to nascent c-fos and
HSP70 RNA associated to paused transcription complexes. These results indicate that
the CBC can readily bind to very short nascent RNAs, since Pol II undergoes pausing
after ∼40 nucleotides of transcription at c-fos and HSP70 (Fivaz et al., 2000; Rasmussen
and Lis, 1993). This observation is consistent with the notion that capping occurs after
20-30 nt of transcription (Rasmussen and Lis, 1993). Upon rapid transcription
activation, CBP80 was also detectable in downstream regions of HSP70 and c-fos,
indicating the movement of Pol II into the genes and the growing RNP.
But why is Pol II detectable at downstream regions in both inducible genes but
not in constitutively active genes? One possible explanation for this observation is that
in a population of cells (for one ChIP experiment, 107 cells are used), Pol II
permanently assembles into a pre-initiation complex, initiates transcription and pauses
promoter-proximal. Many polymerases may not switch to elongation mode successfully
and even less will do so at the same time, which leads to Pol II molecules being spread
out along the length of a gene (see Figure 21). Polymerases paused or initiating at
(rapid) inducible genes, however, start to elongate as the induction signal is provided.
The consequence would be that in a cell population, many Pol II molecules move in
parallel into the gene, which leads to improved detectability by ChIP.
The decrease of CBC downstream accumulation on induced c-fos and HSP70,
which in the case of HSP70 did not parallel the Pol II distribution, suggest that the CBC
epitope might become unavailable as the RNP grows. Immunoelectron microscopy of
the C. tentans Balbiani ring genes revealed that the CBC was bound co-transcriptionally
throughout the transcription loops, with greater abundance in the 5’ proximal region
(Visa et al., 1996). Furthermore, the growing RNP became packed into a solid globular
Part I: Discussion
46
structure in middle and distal regions of the Balbiani ring genes, indicating that indeed
major RNP remodeling was occurring.
Figure 21. Model of Pol II distribution along the same constitutive active gene in a populationof cells.Pol II (blue balls) predominantly accumulates at promoter-proximal regions in a pre-initiated or pausedcomplex. Not every promoter fires Pol II at the same rate in a cell population, and not every Pol IIswitches successfully into elongation mode. This is why Pol II holoenzymes that successfully escaped thepromoter and pause are spread out trough the gene.
There is no report that CBC binding is lost from the nascent RNA during transcription
or splicing. In fact, the CBC binds m 7GpppG cap analogues with a Kd of ∼13 nM
(Mazza et al., 2001), some 100-fold better than unmethylated GpppG and at least 1000
fold better then other nucleotides (Izaurralde et al., 1992). The above results are in
striking contrast to the ChIP results obtained in the yeast system, where Pol II and CBC
are detectable at high levels along the length of genes (Gornemann et al., 2005; Kotovic
et al., 2003). Promoter-proximal pausing has not been reported to occur in yeast,
Part I: Discussion
47
suggesting that this mechanism might be an evolutionary consequence of the
development of the complex protein coding genes found in mammals.
Having established the experimental conditions for co-transcriptional detection
of the CBC as bona fide RNA binding factor in human genes, I could move on to test
the recruitment requirements of RNA splicing factors to transcription units. By testing
many different antibodies specific for mammalian splicing factors, I developed the
experimental conditions for splicing factor ChIP in mammalian cells (see Table 1). In
fact, most of the antibodies tested did not result in any recovery of DNA. This might be
due to loss of epitope availability after formaldehyde crosslinking. It is also conceivable
that the epitopes for several splicing factors antibodies are simply hidden in the massive
nascent RNPs that form during transcription (Osheim et al., 1985; Visa et al., 1996;
Wetterberg et al., 2001). Ongoing experiments in our lab are aimed to stably express
GFP-tagged alternative splicing factors on Bacterial artificial chromosomes (BACs) in
human tissue culture cells. The advantage of using the BAC technology is the avoidance
of massive overexpression of the splicing factors that could otherwise interfere with
alternative splicing decisions. Furthermore, only one GFP antibody it needed for ChIP
experiments with all tagged proteins, which facilitates the comparison of the abundance
of different splicing factors.
By performing ChIP with antibodies specific for components of the U1 and U5
snRNPs as well as U2AF65 and hnRNP A1, I could show that co-transcriptional
recruitment of these factors to transcription units is dependent on the presence of
introns. The position of splicing factor accumulation is meaningful, as it indicates the
potentially important regions of the corresponding transcribed RNA. HnRNP A1
accumulation was detected in upstream regions of active c-fos, consistent with the
presence of a putative hnRNP A1 binding site early in intron 1. In contrast, U2AF65 was
best detected in downstream regions of c-fos, following transcription of the first 3'
splice site; thus, the U2AF65 distribution is consistent with its role in 3' splice site
definition.
Recent ChIP studies in yeast have begun to shed light on how the spliceosome
assembles co-transcriptionally (Gornemann et al., 2005; Kotovic et al., 2003; Lacadie
and Rosbash, 2005). Comparison of the mammalian and yeast data indicates several key
differences. First, the U1 and U5 snRNPs were robustly detectable in upstream c-fos
regions, in exon 1 and intron 1. While the presence of the U1 snRNP in this position is
Part I: Discussion
48
consistent with the presence of the 5' splice site in this region, the detection of the U5
snRNP was surprising. Similar studies in yeast revealed a clear separation in U1 and U5
snRNP accumulation, with the appearance of the U5 snRNP following 3' splice site
synthesis in all genes examined (Gornemann et al., 2005; Lacadie and Rosbash, 2005).
Because U5 116K protein did not co-immunoprecipitate with Pol II or associate with
intronless genes, it seems unlikely that the U5 snRNP is brought to upstream regions of
c-fos by Pol II. These results are in agreement with the yeast studies, showing that U1
and U5 snRNPs are not recruited to intronless genes by Pol II or any other mechanism
(Gornemann et al., 2005; Kotovic et al., 2003). Instead, the mammalian data suggest
that either the U5 snRNP makes early and relatively stable contacts with the 5’ splice
site as suggested by biochemical studies (Maroney et al., 2000; Wyatt et al., 1992) or
that the U5 snRNP is recruited to the gene in the context of a pre-assembled penta-
snRNP complex (Malca et al., 2003; Stevens et al., 2002).
Second, the yeast homologue of U2AF65, Mud2, accumulates on intronless as
well as intron-containing genes. In contrast, human U2AF65 was not detectable on either
intronless gene, histone H2.m or induced HSP70. In vitro experiments suggests that
U2AF65 binds directly to Pol II during the transition from initiation to elongation (Ujvari
and Luse, 2004; Ujvari and Luse, 2006). The finding that U2AF65 co-
immunoprecipitated with hyper-phosphorylated forms of Pol II is consistent with this
possibility; however, the fact that this interaction was sensitive to RNase A treatment
suggests that U2AF65 binding to Pol II alone is not stable. Because U2AF65 also co-
immunoprecipitates significantly with snRNPs, the interaction of U2AF65 with Pol II
may be direct or indirect. U2AF65 also co-immunoprecipitates hypo-phosphorylated,
transcriptionally inactive Pol II (Robert et al., 2002); however, U2AF65 could not be
detected by ChIP in promoter regions where hypo-phosphorylated Pol II is abundant.
Therefore, the data allows for the interpretation that U2AF65 binding to intron-
containing nascent RNA is promoted and/or stabilized by cooperative interactions with
Pol II. These observations point to species differences in co-transcriptional spliceosome
assembly.
Recent speculation has focused on the possibility that splicing factors, like
capping enzymes, may be brought to transcription units by Pol II (Bentley, 2005;
Goldstrohm et al., 2001; Maniatis and Reed, 2002). The rationale is that splicing factor
binding to Pol II would increase the local concentration of splicing factors at the sites of
Part I: Discussion
49
RNA synthesis and increase splicing efficiency and/or fidelity. However, most splicing
factors are expressed in cells at high concentrations, generally in the micromolar range,
so this recruitment mechanism may be unwarranted (Neugebauer, 2002). Indeed,
mammalian genes are characterized by poorly conserved and cryptic splice sites, such
that locally elevated concentrations might even reduce splicing fidelity. Perhaps less
abundant splicing factors other than those studied here are directly bound to Pol II,
accounting for the effects of CTD deletion on pre-mRNA splicing (Fong and Bentley,
2001; McCracken et al., 1997b). The present data indicate that intron-containing
nascent RNA is required for hnRNP A1, U2AF65, U1 and U5 snRNP accumulation,
because none were detected on the highly transcribed intronless genes, histone H2.m or
induced HSP70, on which elongating Pol II was robustly detectable. If Pol II plays a
role in recruitment of these factors, relatively low-affinity binding between splicing
factors and Pol II must be involved. This is in contrast to the stable interactions
observed between the Pol II CTD and capping enzymes and may allow for more
flexibility in when and where along a gene splicing factors are stably recruited.
Accumulation of U2AF65, U1 and U5 snRNPs on the induced gene was
enhanced by treatment with the topoisomerase I inhibitor camptothecin. Importantly,
the overall distribution of factors along the gene was not altered, and the levels of Pol II
and CBC detected in downstream regions of c-fos was unchanged by camptothecin.
This indicates that the drug stalls the polymerase at various positions along the length of
the gene (Collins et al., 2001; Ljungman and Hanawalt, 1996; Stewart et al., 1990) and
allows splicing factors more time to bind to the nascent RNA. Camptothecin did not
cause splicing factors to accumulate on induced HSP70, which is also a camptothecin
target. Therefore, it is conclusive that the splicing factor signals observed in the
presence and absence of camptothecin reflects the potential of splicing factors to bind to
the nascent RNAs present in the indicated gene regions.
The fact that core components of the splicing machinery accumulate in a transcription-
dependent manner on the c-fos gene opens the possibility that c-fos pre-mRNA could be
co-transcriptionally spliced. The wide-spread belief that pre-mRNA splicing is co-
transcriptional has previously relied on experimental evidence obtained with
immunoelectron microscopy of the Balbiani ring genes in C. tentans or the chorion
genes of D. melanogaster (Neugebauer, 2002). Because nascent RNA is generally such
a small fraction of any given RNA species, it has been difficult to extend studies of co-
Part I: Discussion
50
transcriptional pre-mRNA splicing to mammalian cells. By considering splicing in the
context of chromatin, one gains access to these rare RNA molecules, which lead to the
development of the novel assay ChRIP, in which active chromatin is immunopurified
and co-purifying nascent RNA is analyzed.
The ChRIP assay was used on the c-fos gene and allowed for the first time the
direct monitoring of co-transcriptional splicing in human cells. Approximately 17 fold
more RNA with spliced intron 1 is present at the calcium induced c-fos gene than
unspliced RNA. Camptothecin treatment led to a three-fold increase of the
spliced/unspliced ratio for intron 1. Enhanced splicing factor accumulation on c-fos by
camptothecin correlated with higher levels of co-transcriptional splicing, providing
direct evidence for kinetic competition between transcription and co-transcriptional
splicing rates. Interestingly, alternative pre-mRNA splicing is influenced by promoter
identity, transcription rates and the presence of transcriptional pause sites, in a manner
consistent with coordination between splicing and transcription (Cramer et al., 1997; de
la Mata et al., 2003; Howe et al., 2003; Kadener et al., 2001; Roberts et al., 1998). The
fact that the constitutive splicing of c-fos is also influenced by transcription time
emphasizes that even transcripts with strong splice sites are not fully spliced co-
transcriptionally, and that further splicing can occur co-transcriptionally, given
additional time before transcription termination. This scenario indicates that pre-mRNA
splicing occurs within the context of transcription unit activity and may be regulated
within chromatin by diverse cellular mechanisms.
The ChRIP assay can be a useful tool not only to detect co-transcriptional
splicing, but also other RNA processing events that are thought to occur during
transcription or in close vicinity to the gene, such as RNA editing and polyadenylation.
With my thesis work I could show that the core components of the splicing
machinery are recruited co-transcriptionally to mammalian genes in vivo. This co-
transcriptional splicing factor recruitment is dependent on active transcription and the
presence of introns in genes. The finding that the topoisomerase I inhibitor
camptothecin increases splicing factor accumulation on c-fos as well as co-
transcriptional splicing levels provides direct evidence that co-transcriptional splicing
events depend on the kinetics of RNA synthesis.
Part I: Materials and Methods
51
Materials and Methods
Cell culture and treatments
A431 human epidermoid carcinoma cells were grown in DMEM supplemented with 5
mM HEPES (pH 7.2), 100 U/ml penicillin, 100 µg/ml streptomycin and 10% fetal calf
serum. Two hours before c-fos induction (Stewart et al., 1990), the medium was
replaced with serum-free medium. Cells were either control treated with 0.2% DMSO
for 15 min, induced with 5 µM calcium ionophore A23187 (Molecular Probes) for 15
min or, and for camptothecin treatment, cells were incubated for an additional 15 min
with 10 µM (S)-(+)-camptothecin (Sigma). For HSP70 induction, cells were treated
with 250 µM sodium (meta)arsenite (Sigma) for 1h.
Chromatin Immunoprecipitation (ChIP) and Real Time PCR
We used a modification of the technique described (Kuo and Allis, 1999). Briefly,
approximately 108 cells were crosslinked with a final concentration of 1%
formaldehyde added directly to the medium. Cells were washed 2 times with cold PBS,
scraped and collected. Cell pellets were resuspended in 2 ml of SDS lysis buffer (1%
SDS, 10 mM EDTA, 50 mM Tris-HCl, pH 8.1) containing complete protease inhibitor
cocktail (Roche) and incubated for 10 min on ice. Cell extracts were sonicated with a
Branson sonifier W-450 D at 30 % amplitude with 15 times 10 s bursts resulting in
∼500 bp chromatin fragments and then centrifuged for 10 min at 14,000 rpm. 50 µl of
the supernatant was saved as input DNA and the remainder was diluted 1:10 in ChIP
dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl,
pH 8.1, 167 mM NaCl) containing protease inhibitors. 2 ml chromatin solution was pre-
cleared at 4°C with sepharose beads for 1 h before overnight incubation (4°C) with
either 10 µg of 8WG16, 20 µg of Pol3/3, 30 µg MC3 (α-U2AF65), 30 µg CB7 (α-U1
70K), 3 µl α-U5 116K serum, 4 µl α-hnRNP A1 ascites fluid and 10 µl α-CBP80
serum. 10-30 µg non-immune mouse IgG (Sigma) was used as a control. Complexes
were immunoprecipitated with GammaBind G sepharose beads (Pharmacia Biotech)
blocked with 0.2 mg/ml salmon sperm DNA and 0.5 mg/ml BSA for 1 h at 4°C. The
beads were washed rocking for 4 min in each of the following buffers: Low Salt
Immune Complex Wash Buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM
Part I: Materials and Methods
52
Tris-HCl, pH 8.1, 150 mM NaCl), High Salt Immune Complex Wash Buffer (0.1%
SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 500 mM NaCl), LiCl
Immune Complex Wash Buffer (0.25 M LiCl, 1% NP-40, 1% deoxycholic acid, 1 mM
EDTA, 10 mM Tris-HCl, pH 8.1) and twice in TE. The immune complexes were eluted
in 1%SDS and 50mM NaHCO3 and crosslinks reversed for 6 h at 65°C. Samples were
digested with proteinase K for 1 h at 45°C and the DNA extracted with the Qiagen PCR
purification kit.
DNA templates retrieved by ChIP were analyzed by quantitative real-time PCR
on a Stratagene MX3000, using the SYBR Green method (ABsolute QPCR SYBR
Green Rox Mix, AB Gene). The reaction volume was 20 µl, with 4 µl DNA template
(input 1:10) and 90-900 nM of each primer according to individual optimization.
Cycling was done for 15 min at 95°C, followed by 40 cycles for 30s at 95°C, 1 min at
60°C, and 30s at 72°C, with measuring at the end of the annealing step. Dissociation
curves were obtained by heating the samples to 95°C followed by cooling down to 55°C
and successive heating of the samples to 95°C with continuous measurement. Primer
sets distinguishing between different regions of the genes are available upon request.
The relative proportions of co-immunoprecipitated gene fragments were determined
based on the threshold cycle (Ct) for each PCR product. Data sets were normalized to
ChIP input values, then the Ct values from pol II and splicing factor ChIP were
subtracted from the Ct values obtained from templates derived from ChIP with
unspecific antibody (2 [Ct(unspec)-Ct(input) ] – [Ct(spec)-Ct(input) ] (Chakrabarti et al., 2002). The
fold difference over background obtained for gene regions was further normalized to the
value obtained with a primer pair amplifying an intergenic region on chromosome 10
where no annotated genes could be found. For every gene fragment analyzed, each
sample was quantitated in duplicate and from >3 independent ChIPs. SEM was
determined for each fold difference above non-immune control and intergenic control
region.
RNA extraction and RT-PCR
Total RNA was extracted from A431 treated with calcium ionophore, calcium
ionophore and camptothecin or sodium arsenite with TRIZOL (Invitrogen) according to
the manufacturer´s recommendation. Total RNA was briefly exposed to RNase-free
DNase I (Ambion). RNA was reverse transcribed to cDNA using oligo (dT)18 primer
Part I: Materials and Methods
53
and SuperScript III Kit (Invitrogen). cDNA was amplified by PCR with primers listed
in Table 2. PCR reactions were as follows: 5µl PCR buffer; 10 mM dNTPs; 10 µM
primers, 1 µl cDNA and 0,4 µl Taq polymerase. Cycling was performed using a
ThermoCycler (PTC-200, MJ Research). Cycling parameters were 94°C for 4 min
followed by (94°C for 1 min, 60°C for 1 min, 72°C for 1 min) x 27 cycles. Products
were electrophoretically separated on a 2% agarose gel, DNA bands stained with Gel
Star (Cambrex BioScience), and bands detected and analyzed with ArgusX1 (Biostep
GmbH) and TotalLab (Nonlinear Dynamics).
Standard Immunoprecipitations and Western Blot Analysis
Whole cell extracts from semi-confluent A431 cells were prepared in NET-2 buffer (50
mM Tris-HCl pH 7.5, 150 mM NaCl, 0.05% Nonidet P-40) containing protease and
phosphatase inhibitors (1mM NaF, 10 mM b-glycerophosphate). Immunoprecipitation
was carried out for 3 h at 4°C with gamma-bind beads coupled to either 10 µl a-CBC80
serum, 500 µl MC3 or CB7 hybridoma supernatant, 5 µl of a -U5 116K serum or a -
hnRNP A1 ascites fluid. The beads were washed 5 times in NET-2 buffer and proteins
were eluted in 50 µl in SDS sample buffer. For RNase A digestions, starting extracts
were treated with 100 µg/ml RNase A for 30 min at RT. 10 µl of IP and 0,3% of starting
material were analyzed by Western blot after electrophoresis on a 7.5% SDS-
polyacrylamide gel, H5 or H14. Metabolic labeling was carried out by incubating 15 cm
dishes in phosphate-free medium + 1%FCS and 100 mCi 32P-orthophosphate overnight,
followed by lysis and immunoprecipitation as described above. RNA was extracted
from the final pellets with phenol/chloroform, precipitated, resolved on a 10% urea gel,
and analyzed by phosphorimager.
Indirect immunofluorescence
A431 cells were grown on cover slips and treated with sodium arsenite for 1 h. Cells
were fixed in 4% PFA (Sigma-Aldrich) for 10 min, permeabilized for 5 min with 0.2%
Triton X-100 (Sigma-Aldrich), and incubated with antibodies specific for TMG cap
(snRNAs), U1 70K, U2AF65, hnRNP A1 and TIA-1. Secondary anti–mouse antibodies
conjugated with TRITC and anti-rabbit antibodies conjugated with FITC (Jackson
ImmunoResearch Laboratories) were used. Images were collected using the DeltaVision
microscope system (Applied Precision) coupled with Olympus IX70 microscope. The
Part I: Materials and Methods
54
images are projections of z-sections encompassing the whole nucleus and were
subjected to mathematical deconvolution (SoftWorx; Applied Precision).
Chromatin-RNA-Immunoprecipitation (ChRIP)
For ChRIP, cells were crosslinked and harvested as in ChIP, but pellets were
resuspended in RIPA buffer (50 mM Tris–Cl, pH 7.5, 1% Nonidet P-40 (NP-40), 0.5%
sodium deoxycholate, 0.05% SDS, 1 mM EDTA, 150 mM NaCl) containing complete
protease inhibitors and RNasin (Promega). Extracts were sonicated and insoluble
material was pelleted. 250 µl of supernatant was precleared with sepharose beads for 30
min at 4°C before 9 µl of anti-acetyl Histone H4 ab (Upstate) or 10 µg non-immune IgG
was added for incubation overnight. Immunoprecipitation and washing was done as in
ChIP, chromatin was eluted for 15 min at 65°C in 1%/ SDS (w/v) in TE following
proteinase K treatment for 1 h at 45°C. Crosslinks were reversed at 65°C for 5 h. RNA
was extracted with Phenol/Chloroform and treated extensively with DNase I. cDNA
was prepared from 1/3 of RNA from AcH4 or IgG IP with SuperScript III (Invitrogen).
Primers located in c-fos intron 2 and exon 4 and oligo (dT) were used for reverse
transcription. The cDNA and no RT controls were analyzed by quantitative PCR with
primers spanning exon1/2 or exon3/4 and intron 1 and exon3/intron3, respectively. As
control, the oligo (dT) cDNA was amplified with primers located in c-fos exon4. The
relative proportions of co-immunoprecipitated RNA fragments were determined based
on the threshold cycle (Ct) for each PCR product. All AcH4 values were at least 3
cycles (i.e. 8 fold) more enriched than non-immune controls. The AcH4 Cts for spliced
and unspliced product were substracted from each other and squared to yield the fold
difference spliced/unspliced. For every RNA fragment analyzed, each sample was
quantitated in duplicate and from >3 independent ChRIPs. SEM was determined for
each fold difference.
Part I: Materials and Methods
55
Primer Sequence Where?GW10 1 5'- GGCTAATCCTCTATGGGAGTCTGTC -3' Chromosome 10, contig AL392045GW10 2 5'- CCAGGTGCTCAAGGTCAACATC -3'
PGK 9a 5'- GTAGTGTGGGCCCTGTTCC -3' PGK1 promoter (-55 → 97)PGK 10 5'- TCCAGCGTCAGCTTGTTAGA -3'
PGK 3 5'- TCTGTGCTCTGTCGCAAACCTC -3' PGK1 intron 1PGK 4 5'- AATCTCAGCCCTTCCTCAGTGG -3'
PGK 7 5'- ACATTACCAGAACCCCAAAGCC -3' PGK1 in1-ex2-in2PGK 8 5'- AAACCCAGCCCAAGGATTAGC -3'
LDHA 19 5'- CGTCAGCATAGCTGTTCCAC -3' LDHA prom-ex1-in1LDHA 20 5'- CAGAGGCAGTTGGCTCTACC -3'
LDHA 17 5'- AACTGCCTCTGGTTCTGCTG -3' LDHA intron 1LDHA 18 5'- ATCCTTGTTCCGCATCCAC -3'
LDHA 3 5'- GGAGCCCATAGAGCCAAAAAAG -3' LDHA intron 1LDHA 4 5'- GAGCGTCCCAAGAGAAAAATGC -3'
LDHA 5 5'- CCTTTCAACTCTCTTTTGGCAACC -3' LDHA in 1-ex2LDHA 6 5'- AATCTTATTCTGGGGGGTCTGTTC -3'
H2 1 5'- ATTCCTCACAGCCTACCTCCAGTC -3' Histone H2A.m promoter-exonH2 2 5'- TTTACCGCCTTGTTTGCCG -3'
His 3 5'- TAAACTCTTGGGGCGTGTGACC -3' Histone H2A.m downstreamHis 4 5'- TGCTGTTAGGCTGATTTTGTCTGC -3'
HSP70 5 5'- AGGGTCCGCTTCGTCTTTCG -3' HSP70 exonHSP70 6 5'- TCTCCACCTTGCCGTGTTGG -3'
HSP70 11 5'- TTTGAGGGCATCGACTTCTAC -3' HSP70 midHSP70 12 5'- ACCAGGTCGTGAATCTGGG -3'
HSP70 15a 5'- CCATTGAGGAGGTGGATTAGGG -3' HSP70 endHSP70 4 5'- AAACAGCAGCAAAGTCCTTGAGTC -3'
Fos 17 5'- GCATCTGAGAAGCCAAGACTGAGC -3' c-fos exon 1Fos 18 5'- TGAAGCCCGAGAACATCATCG -3'
Fos 23 5'- CAAACCCCCTTTCAAGCAAGTG -3' c-fos intron 1Fos 24 5'- TTTTCCTCCCAGGCATTCCG -3'
Fos 13 5'- AAGGTGGAACAGGTGAGGAACTC -3' c-fos ex2-in2Fos 14 5'- TTGGGATGGAATGGGCTTG -3'
Fos 9 5'- CGGAGATGTAGCAAAACGC -3' c-fos exon 4Fos 10 5'- AAAGAGAAAAGAGACACAGACCC -3'
BA unProm1 F 5'- TCCCAGGACTGAGAGGTGACAAAG -3' Upstream β-actinBA unProm1 R 5'- AGCCCAGAGTATTCCCAGGAGAAC -3'
BA 1 5'- TGGTGTCCGTTCTGAGTGATCCTC -3' β-actin ex1-in1BA 2 5'- GAGCACAGCTTCTTTGCAGCTC -3'
BA 5 5'- AAGCCTGGGGTTTTCTTGGG -3' β-actin in2-ex3BA 6 5'- CTGGGTCATCTTTTCACGGTTG -3'
BA 15 5'- CCTTCCTTCTTGGGTAAGTTGTAGC -3' β-actin ex4-ex5BA 16 5'- CAGCACTGTGTTGGCATAGAGGTC -3'
BA 27 5'- TGATGTATGAAGGCTTTGGTCTCC -3' β-actin exon 6BA 28 5'- GCTGTTTGTGTAAGGTAAGGTGTGC -3'
Part I: Materials and Methods
56
Primer Sequence WhereHis2 5 5'- AAAAACCCCCCTCCCACTTC -3' Upstream mouse Histone H4His2 6 5'- CTTGCCCTGAAACTTGTTGAGC -3'
His2 1 5'- TTTCAGTCCTCTAAAAGGTCCGC -3' Mouse Histone H4 prom-exHis2 2 5'- TGGTGATGCCCTGGATGTTATC -3'
Fos e 5'- TCTTACTACCACTCACCCGCAGAC -3' c-fos spliced ex1/ex2Fos f 5'- GGAATGAAGTTGGCACTGGAGAC -3'
Fos k 5'- CCGAAGGGAAAGGAATAAGATGG -3' c-fos spliced ex3/ex4Fos l 5'- TAGTTGGTCTGTCTCCGCTTGG -3'
Fos i 5'- TCTGTGGGTTGCTCCTTTTT -3' c-fos unspliced in3/ex4Fos j 5'- AGGTTGGCAATCTCGGTCT -3'
Myc 11 5'- TGCTCCCTTTATTCCCCCAC -3' c-myc unspliced in1-ex2Myc 12 5'- GGTCATAGTTCCTGTTGGTGAAGC –3'
Myc F 5'- GCGACTCTGAGGAGGAACAAGAAG -3' c-myc spliced ex2-ex3Myc R 5'- ACTCTGACCTTTTGCCAGGAGC -3'
Table 2. Primers used in this study.
Part II
57
PART II
58
Part II: Introduction
59
Introduction
Transcription of eukaryotic genes requires three different kinds of RNA polymerases,
Pol I, II and III. Each holoenyzme transcribes a different set of genes, and each RNA
polymerase must recognize a specific DNA region important for transcription initiation,
which is termed promoter. The recruitment of RNA polymerases to the promoter
depends on accessory factors, so-called transcription factors that recognize their cognate
promoter sequence. Pol I transcribes the multicopy repeat units containing the 28S, 5.8S
and 18S ribosomal RNA (rRNA) genes. It is unique among nuclear RNA polymerases
by only having to recognize one type of promoter structure. Most research attention has
been dedicated to Pol II, since this polymerase transcribes protein-coding genes into
messenger RNA (mRNA). Besides mRNA, Pol II also synthesizes several non-coding
RNAs, including the spliceosomal U1, U2, U4 and U5 small nuclear RNAs (snRNAs)
and microRNAs (Lee et al., 2004; Orphanides et al., 1996). Pol III transcribes only non-
coding RNAs, among them the genes for transfer-RNAs (tRNAs), 5S rRNA and
snRNAs including H1, 7SK and spliceosomal U6 snRNA.
The U6 snRNA
The formation of the spliceosome results from the stepwise assembly of U1, U2, U4,
U5 and U6 small nuclear ribonucleoprotein particles (snRNPs) and many non-snRNP
splicing factors onto the pre-mRNA. In every snRNP, core proteins and snRNP specific
proteins are organized with one snRNA or two in the case of the U4/U6 di-snRNP.
U1, U2, U4 and U5 snRNAs are upon synthesis by Pol II exported to the
cytoplasm where they assemble with snRNP specific proteins and Sm (core) proteins.
Upon re-import to the nucleus, these snRNAs assemble into mature, splicing competent
snRNPs (reviewed in Will and Luhrmann, 2001). U6 snRNA, on the other hand has
been shown to be synthesized by Pol III. U6 snRNP maturation is also different since
U6 snRNA is not complexed with Sm proteins like the other spliceosomal snRNAs but
U6 snRNA instead associates with Like-sm (Lsm) proteins in the nucleus (reviewed in
Will and Luhrmann, 2001). These Lsm 2-8 proteins form a seven-member ring structure
near the U-rich 3’ end of U6 snRNA (Vidal et al., 1999, see Figure 1).
Part II: Introduction
60
A B
B
C
Figure 1. U6 snRNA alone or in U4/U6 di-snRNP.(A) Sequence and proposed secondary structure of naked human U6 snRNA (Achsel et al., 1999). (B)Proposed structure of human U6 snRNA in the U6 snRNP (Karaduman et al., 2006). Nucleotides that are100% evolutionarily conserved between yeast U6, tomato U6, nematode U6, fly U6, mouse U6, human U6 andhuman U6atac snRNAs are shown in red. Nucleotides that are 70% conserved are shown in green. In eachpart of the Figure, nucleotides that are involved in the formation of stems I and II with the U4 snRNA, arehighlighted with blue and red lines, respectively. (C) Schematic of U4/U6 di-snRNP. RNA parts of snRNPare denoted as black lines, U4 snRNP specific proteins are 61K, 15.5, hPrp4, USA-Cyp and Sm proteins.U6 specific Lsm proteins are bound in a ring-like structure to the 3’ end of the snRNA (Stanek andNeugebauer, 2004).
Figure 2. Structure of the human snRNA promoters.U1 and U2 snRNA promoters exemplify the prototypic Pol II snRNA promoter whereas U6 snRNA servesas the prototypic Pol III snRNA promoter. Both genes contain the PSE and DSE, which are centered atapprox. 60 bp and 220 bp upstream of the transcription start site, respectively. The U6 snRNA promotercontains furthermore at position –25 a TATA box which determines polymerase specificity (adapted fromHernandez, 2001).
The 107 nts U6 snRNA is the shortest and most highly conserved of the snRNAs
involved in pre-mRNA splicing. It is a very dynamic molecule undergoing multiple
conformation changes during assembly and splicing and is found in the U6 snRNP, the
U4/U6 di- snRNP base-paired with the U4 snRNA and in the U4/U6⋅U5 tri-snRNP
(Vidal et al., 1999, see Figure 1). U6 snRNA is an exceptional member of the
Part II: Introduction
61
spliceosomal snRNAs: Apart from its synthesis by Pol III, its 5’ terminus contains a γ-
monomethyl phosphate cap instead of a 2,2,7-trimethylguanosine (TMG) cap
characteristic for Pol II transcribed spliceosomal snRNAs. Furthermore, the U6 snRNA
3’ end can be post-transcriptionally uridylated and is blocked with a 2',3'-cyclic
phosphate (Kunkel et al., 1986; Lund and Dahlberg, 1992; Singh and Reddy, 1989).
Aside from its role in splicing, the U6 snRNA gene grew famous in the recent
field of RNA interference (RNAi). The human and mouse U6 snRNA promoter is
widely used in vectors driving small hairpin RNA (shRNA) expression for gene knock-
down studies in mammalian cells (Paul et al., 2002; Sui et al., 2002; Yu et al., 2002). In
shRNA expression, Pol III promoters are preferred due to their well-defined
transcription initiation and termination sites, producing distinct non-poly(A) transcripts.
Although the U6 snRNA gene is believed to be transcribed by Pol III, I found evidence
Pol II is involved in its biogenesis. This work reveals that Pol II is bound to the U6
snRNA promoter in vivo and elucidates possible functions of Pol II in U6 snRNA
generation and maturation.
Recruitment of Pol II to mRNA and snRNA promoters
Pol II promoters consist of the core promoter, defined as the minimal region capable of
directing transcription in vitro, and a regulatory region that can stretch for several kb
upstream. Generally, these regulatory regions are responsible for the recruitment of
activator or repressor proteins, which modulate the levels of transcription.
The classical core Pol II promoter itself consists of 2 segments, an A/T rich
sequence 25-30 bp upstream of the transcription start site called TATA box and the
initiator (Inr) with the consensus sequence 5’-YYCARR-3’, the adenosine marking the
transcription start site. The TATA box was historically considered to be a strictly
conserved sequence element for all Pol II transcribed genes (Breathnach and Chambon,
1981). Today however, as the human genome sequence is known and promoters of
many genes are analyzed, the prevalence of the TATA box diminished. It is noteworthy
that cellular Pol II promoters can contain one element, both elements, or neither.
Transcription from TATA box-containing mRNA promoters is well understood
and has been reconstituted with in vitro experiments. The TATA-box and Inr are
recognized by components of the transcription machinery, the general transcription
factors (GTFs). In vitro studies showed that they assemble sequentially on the core
Part II: Introduction
62
promoters to form the transcription initiation complex. The TATA-Binding-Protein
(TBP), by itself or together with TBP associated factors (TAFS) in the TFIID complex
binds first to the TATA box, followed by TFIIB, a TFIIF-Pol II complex, TFIIE and
TFIIH. TFIIA can enter the complex at any stage, but its role in transcription initiation
is still controversial (Hernandez, 2001, and references therein; Orphanides et al., 1996).
TFIIB is the key player for recruiting Pol II to the promoter since it tethers Pol II/TFIIF
to the promoter by binding to the TFIID/DNA complex (see Figure 3A)
The Pol II dependent spliceosomal snRNA genes represent a group of genes that
lack both the TATA box and the Inr. Instead, the core promoter region contains a
proximal sequence element (PSE), which is sufficient for basal transcription and located
50-70 bp upstream of the transcription start site (reviewed in Hernandez, 2001). The
regulatory region of snRNA promoters contains the distal sequence element (DSE),
which is located 220 bp upstream of the transcription start site and functions as an
enhancer of transcription (see Figure 2).
The initiation complex that forms on the TATA-less snRNA promoters is somewhat
different to that of mRNA genes. The snRNA activator protein complex (SNAPc), a
multisubunit complex also known as PTF, binds to the PSE and nucleates the assembly
of the transcription initiation complex. Stepwise assembly of GTFs on the snRNA
promoters has not been shown in vitro, and so the order of recruitment and position in
the initiation complex is less well understood. Depletion and reconstitution experiments
revealed that TBP, TFIIB, TFIIA and TFIIE are required for U1 snRNA transcription
(Kuhlman et al., 1999, see Figure 3A). Although TFIID is not required for basal
transcription of U1 snRNA in vitro, Chromatin Immunoprecipitation experiments
revealed that the TFIID subunit TAFII100 is bound to the U2 snRNA gene promoter in
vivo (Christova and Oelgeschlager, 2002). Moreover, depletion and reconstitution
experiments suggest that TFIIH is either not required for Pol II transcription of snRNA
genes or is required in much lower levels than in mRNA gene transcription (Kuhlman et
al., 1999). If true, this raises the question of how promoter melting is achieved at
snRNA promoters. Active transcription of mRNA genes requires Cdk7, the kinase
subunit of TFIIH, which phosphorylates the Pol II large subunit C terminal domain
(CTD) (Akoulitchev et al., 1995; Komarnitsky et al., 2000). It still remains to be
elucidated if TFIID and TFIIH are required for Pol II transcription of spliceosomal
snRNA genes in vivo.
Part II: Introduction
63
Figure 3. Composition of Pol II and Pol III transcription initiation complexes on human genes.(A) Pol II initiation complexes assembled on a TATA box containing mRNA promoter and on the TATA-less U1 snRNA promoter. For simplification, the GTFs are designated without “TF”. (B) Pol III initiationcomplexes assembled on the U6 snRNA and a tRNA-type promoter. The placement of hBRF, hBRFU andhB” is arbitrary (picture taken from Hernandez, 2001).
The DSE in the regulatory region is the activator for snRNA promoters. It often
contains SphI postoctamer homology (SPH) elements, which are binding sites for the
selenocysteine tRNA gene transcription activating factor (Staf, Schaub et al., 1997) and
the octamer sequence ATGCAAAT that binds Oct-1. SNAPc is in fact recruited to PSE
through cooperative binding to Oct-1 (Hernandez, 2001, and references therein). The
DSE and PSE are separated by about 150 bp, and cooperative binding is facilitated
Part II: Introduction
64
through correct nucleosome positioning between the octamer sequence and the PSE
(Boyd et al., 2000; Zhao et al., 2001).
Recruitment of Pol III to the U6 snRNA promoter
Pol III transcribes a collection of genes whose main common features are that they
encode structural or catalytical RNAs that are normally shorter then 400 bp. The Pol III
promoters are more varied in structure and yet not as diverse as the Pol II promoters.
Three main types of Pol III promoters have been described, two of which are gene-
internal and generally without a TATA box (type 1 and 2), and one which is gene-
external and includes a TATA box (type3). The type 1 promoters encompass 5S rRNA
genes, with a boxA separated from a boxC sequence element 55-80 bp downstream of
the transcription start site. tRNA genes have type 2 promoters, which consist of a boxA
and boxB element with varying distance downstream from the transcription start site
(Figure 3B).
The type 3 core promoters are found, among others, in the human genes for
7SK, U6 snRNAs and H1 RNA (Schramm and Hernandez, 2002). The U6 snRNA
promoter architecture closely resembles the Pol II transcribed snRNAs in that it also
contains the PSE and DSE at approximately the same position (Krol et al., 1987, see
Figures 2 and 3). Like in U1 and U2 snRNA, SNAPc is recruited to PSE through
cooperative binding with Oct-1 bound to the DSE due to a positioned nucleosome
between the octamer sequence and the PSE (Boyd et al., 2000; Zhao et al., 2001).
However, unlike the Pol II transcribed snRNA genes, the U6 snRNA promoter
contains a TATA box 30 bp upstream of the transcription start site. Before, a TATA
box was considered typical for Pol II promoters and it came as quite a surprise that U6
snRNA was indeed synthesized by Pol III (Kunkel et al., 1986; Reddy et al., 1987;
Sollner-Webb, 1988).
Paradoxically, the TATA box is responsible for the selective recruitment of Pol
III to the U6 snRNA promoter in vitro. Mutation in the U6 snRNA gene TATA box
induces Pol II transcription from that promoter, whereas insertion of a TATA box into
the U2 snRNA promoter converts this Pol II promoter into a Pol III promoter (Lobo and
Hernandez, 1989). Furthermore, the TATA boxes of mRNA genes and the U6 gene can
be interchanged without loss of functional transcription (Lobo et al., 1991). Mutation of
the TATA box in the 7SK snRNA gene promoter allows Pol II to transcribe the gene in
Part II: Introduction
65
vivo (Boyd et al., 1995). Hence, the same promoter element specifies transcription
initiation of mRNA genes by Pol II and U6 (and other snRNAs like 7SK) by Pol III.
Pol III is recruited to type 2 (tRNA) promoters through the multisubunit
transcription factor TFIIIB which in turn is recruited by TFIIIC bound to the A and B
box sequence elements located within the coding region of tRNAs (Figure 3). The U6
snRNA gene does not contain A and B box elements, and in fact it was shown that
TFIIIC is not necessary for U6 snRNA transcription in vitro (Reddy, 1988). TFIIIB on
the other hand was found to be necessary for both tRNA and U6 snRNA transcription
(Hernandez, 1993, and references therein). While TFIIIB that acts on tRNA genes
consists of the three subunits, TBP, hBRF and B´´, the TFIIIB complex recruited to the
U6 snRNA promoter contains hBRFU instead of hBRF. It is assumed that hBRFU
recruits Pol III to the promoter (Hernandez, 2001, see figure 3B).
Taken together, TFIIB or TFIIIB binding to the promoter probably determines
polymerase II or III specificity to the snRNA promoters, respectively.
Evidence for U6 snRNA transcription by Pol III
One of the first evidences that U6 snRNA was not like the other spliceosomal snRNAs
was provided when the U6 snRNA had been sequenced in 1980. It became clear that the
5’-terminal cap was different from the 7-methylguanosine or trimethylguanosine cap
(TMG) cap found on Pol II transcribed mRNAs or spliceosomal snRNAs, respectively
(Epstein et al., 1980, see Figure 4). In fact, the 5’-end of U6 snRNA proved not to be a
nucleotide, but a monomethyl added to the γ- phosphate group of the template-encoded
first nucleotide (Singh and Reddy, 1989). The addition of this cap is dependent on
sequence and/or structure, while the capping of Pol II-transcribed snRNAs or mRNAs is
tightly coupled to transcription (Singh et al., 1990, and references therein). In human
cells, U6 snRNA capping is determined by a conserved stem loop structure at the 5’-
end. In the yeast S. cerevisiae, U6 snRNA receives a TMG cap when the 5’ stem loop is
disrupted, although it is still transcribed by Pol III (Kwan et al., 2000). Moreover, U6
snRNA with a TMG cap does not seemingly affect pre-mRNA splicing, indicating that
neither the 5’ stem loop nor the cap structure interferes with U6 snRNA function.
However, it is still unknown which enzyme adds the γ-monomethyl cap on U6 snRNA
and whether the cap addition is linked to transcription or occurs post-transcriptionally.
Part II: Introduction
66
In the late 1980s, several lines of evidence were formed that came to the
surprising result that Pol III and not Pol II transcribes the U6 snRNA gene in contrast to
the other spliceosomal snRNAs U1, U2, U4 and U5. A cloned human U6 snRNA gene
could be transcribed in a HeLa S100 extract lacking Pol II activity (Kunkel et al., 1986).
Furthermore, U6 snRNA gene transcription in S100 extracts and isolated nuclei was not
sensitive to low α-amanitin concentrations (1 µg/ml), a concentration that inhibits
selectively Pol II but not Pol III (Wieland and Faulstich, 1991). U6 snRNA gene
transcription was completely blocked at 200 µg/ml in this system, the concentration
needed to shut off 5S rRNA transcription (Kunkel et al., 1986; Reddy et al., 1987). Pol
III was also determined to transcribe the U6 snRNA gene in X. tropicalis oocyctes, S.
cerevisiae and S. Pombe (Kleinschmidt et al., 1990; Krol et al., 1987; Moenne et al.,
1990). Interestingly, the S. Pombe U6 gene contains an intron that seems to be removed
by the pre-mRNA spliceosome machinery (Potashkin and Frendewey, 1989; Tani and
Ohshima, 1989).
Figure 4. Cap structures of mRNA, Pol II-transcribed snRNAs and Pol III-transcribed snRNAs.
7-methylguanosine cap (mRNA)
2,2,7-methylguanosine cap (Pol II-transcribed snRNAs and other small nuclearRNAs)
γ-monomethyl phosphate cap (Pol III-transcribed U6 snRNA and other small nuclearRNAs)
Part II: Introduction
67
Figure 5. Promoter structures of five active human U6 snRNA genes.The DSE is indicated with the two sequence elements SPH and OCT which bind Staf and Oct1,respectively and which are switched in U6-2 and U6-9. U6-9 contains a poor match to the PSEconsensus. The location and spacing of the promoter elements are conserved, whereas the sequencesbetween these elements are not conserved (picture adapted from Domitrovich and Kunkel, 2003).
Another evidence for Pol III directed U6 snRNA gene transcription is the fact that the
gene ends with a stretch of five T residues – a common feature of Pol III transcribed
genes, which marks the transcription termination site. Furthermore, U6 snRNA was
found to be bound by La, a protein that binds poly(U) and is involved in 3’end
processing of Pol III transcripts, which was interpreted as another strong evidence for
U6 snRNA being a Pol III transcript (Rinke and Steitz, 1985). Recently, however, it had
been established that La also binds Pol II transcribed snRNAs in S. cerevisiae (Wolin
and Cedervall, 2002, and references therein). There is also evidence that La binds to
histone mRNAs, thereby stabilizing their 3’ ends (McLaren et al., 1997). It is unknown
if the La protein also plays a role in the processing of metazoan Pol II transcribed
snRNAs.
Interestingly, the La-precipitable form of U6 snRNA is highly variable in size
and contains 3'-oligo(U) stretches up to 20 nucleotides long that are not template-
encoded and therefore added post-transcriptionally (Rinke and Steitz, 1982). The major
U6 snRNA form terminates with five UMP residues and an unusual 2',3'-cyclic
phosphate (Lund and Dahlberg, 1992) whereas the minor form (10 %) contains multiple
uridylate residues terminating with a 3'-hydroxyl group, a terminus common to other
Part II: Introduction
68
Pol III transcribed genes (Lund and Dahlberg, 1992; Rinke and Steitz, 1985). In X.
tropicalis oocytes, La protein only associates with U6 snRNA containing a 3'-OH end
but not with U6 snRNAs carrying a 2',3'-cyclic phosphate. U6 snRNAs present in anti-
Sm precipitates from X. tropicalis oocytes contain the cyclic phosphate, however, a
minor fraction of precipitated U6 snRNA contains 5-8 uridylates and a hydroxyl group
at the 3' end. One explanation for this observation is that during snRNP maturation, the
U6 snRNA 3' end could be shortened to 4-5 UMPs and a cyclic phosphate could be
added to the 3' terminus. It is assumed that the formation of the cyclic phosphate
releases the La protein, thereby allowing the U6 snRNA to be stably incorporated into
U4-U6 snRNPs complexes in vivo (Lund and Dahlberg, 1992; Terns et al., 1992).
Alternatively, it has been suggested that the conversion of 3'-hydroxyl ends of U6
snRNA species into cyclic phosphates occurs as a consequence of U6 snRNA taking
part in the splicing reaction (Tazi et al., 1993).
A 3'-terminal uridylyl transferase and a 3'-nuclease specific for U6 snRNA and a
more generally catalyzing 3’terminal phosphate cyclase have been characterized (Booth
and Pugh, 1997; Genschik et al., 1997; Trippe et al., 2003; Trippe et al., 1998).
However, the function of UMP addition, removal and phosphate conversion of U6
snRNA 3' ends is to date unclear. Another unique feature of U6 snRNA is the post-
transcriptional addition and removal of 3’ uridylate residues, mechanisms that could
indicate a sophisticated biogenesis and/or recycling process for functional U6 snRNA
molecules.
Furthermore it was shown that a minor population of U6 snRNA contains one
adenylic acid residue added post-transcriptionally to the 3’end (Sinha et al., 1998). This
3’end adenylation has been established for several small RNAs including 7SK, U2
snRNA and, to a minor extent, also for 5S rRNA, but the function, again, is unclear.
The adenylic residue permits U6 snRNA 3’end uridylation in vitro and in vivo (Chen et
al., 2000), indicating that these two processes are related, maybe even competing with
each other.
U6 snRNA genes
U6 snRNA is the least variable of the spliceosomal snRNAs, which reflects its central
role in the splicing process (Brow and Guthrie, 1988). Although the promoter structure
of U6 snRNA was well known, only recently the existence of other full-length U6
Part II: Introduction
69
snRNA genes in the human genome was discovered (Domitrovich and Kunkel, 2003).
The authors identified eight new 107 or 106 bp long U6 snRNA coding sequences
encompassing five or four 3’ T-residues, respectively, and were numbered from 2-9.
According to the proposed nomenclature, the previously characterized U6 snRNA gene
is denoted U6-1 (Figure 5).
The chromosomal location of the nine U6 snRNA genes is further described in
Table 1. Five of the nine genes (U6-1, U6-2, U6-7, U6-8 and U6-9) contain the
upstream promoter elements DSE (SPH, OCT), PSE and TATA, while the other
variants U6-3, U6-4, U6-5 and U6-6 did not appear to contain any promoter elements
(see Figure 5). In fact, chromatin immunoprecipitation (ChIP) experiments revealed that
acetylated histone H4 (AcH4) and TBP interact only with U6 genes with identifiable
promoter elements, indicating that only these genes are transcriptionally active.
Furthermore, U6-1, U6-2, U6-7, U6-8 and U6-9 maxigene constructs but not the U6-4
maxigene showed transcriptional activity when transfected into human cells
(Domitrovich and Kunkel, 2003).
U6 gene
Chromosome
Strand Location (bp)
U6-1 15 minus 65919331-65919437
U6-2 19 plus 844484-844590
U6-3 X minus 139937735-139937841
U6-4 3 plus 182432227-182432333
U6-5 2 minus 175,248,201-175,248,307
U6-6 10 plus 13299275-13299381
U6-7 14 plus 31741002-31741107
U6-8 14 minus 31742121-31742226
U6-9 19 plus 844484-844590
Table1. Chromosomal location of nine identified U6 snRNA genes.The human 107 nt U6 snRNA sequence was used as input in a BLAT search (UCSC human genomebrowser) resulting in 9 identifiable loci containing the U6 coding region (Domitrovich and Kunkel, 2003).
Part II: Introduction
70
Aim of part II of the thesis
Oftentimes experiments start with a fair assumption and lead to an unexpected result
and a confused researcher and this is exactly what happened in this study. We
anticipated that the U6 snRNA gene would be a perfect negative control for the work
described in part I. U6 snRNA, a well-studied Pol III transcript (Kunkel et al., 1986;
Reddy et al., 1987) should not have any Pol II bound to its promoter. Most surprisingly,
Pol II was robustly detectable on the U6-1 gene promoter in vivo.
The aim of this work was to determine whether Pol II is bound to various U6
snRNA genes by applying chromatin immunoprecipitation (ChIP) in HeLa cells. The
accumulation profile of Pol II and Pol III along the chromosomal U6 snRNA gene loci
was determined and transcription inhibition studies were undertaken to elucidate a
possible functional role of Pol II in U6 snRNA transcription. The experiments provide
clear evidence that Pol II is required for transcription of plasmid-borne U6 snRNA
genes. Moreover, Pol II seemed to be required for genomically encoded U6 snRNAs.
Thus, it appears that Pol II plays an important role in U6 snRNA biogenesis in vivo.
Part II: Results
71
Results
Pol II and III accumulate on the U6-1 snRNA gene promoter
ChIP experiments with the 8WG16 antibody specific for the hypo-phosphorylated form
of Pol II (Pol IIa) large subunit C-terminal domain (CTD) revealed a major Pol II
enrichment at the distal promoter region of the originally identified U6-1 gene (Figure
6, left). In agreement with previously published results (Hirsch et al., 2004), the U6-1
gene promoter region was also enriched by ChIP with an antibody directed against Pol
III subunit RPC155 (Figure 6 right). As controls, Pol II and Pol III accumulation was
measured at the promoter region of the Pol II-transcribed gene encoding
phosphoglycerate kinase 1 (PGK1) and the Pol III-transcribed tRNALeu gene by ChIP.
Pol II was massively bound to the PGK1 promoter but only poorly detectable at the
tRNALeu gene. Conversely, Pol III was enriched at the tRNALeu gene but not detectable
at the PGK1 promoter. Pol II ChIP with an antibody specific for a non-CTD epitope
gave the same result (Figure 7A). Pol II ChIP with extracts from HEK cells resulted
also in Pol II accumulation on the U6-1 gene promoter, confirming that this observation
was not due to an artifact in HeLa cells (Figure 7B).
The simplest explanation for Pol II accumulation on the U6-1 promoter region
by ChIP is that Pol II actually accumulates on a nearby mRNA gene. The genomic
neighborhood of the U6-1 gene was inspected for nearby Pol II-transcribed genes. Two
human genome databases were consulted: the UCSC Genome Browser (Kent et al.,
2002) and the Ensembl Human Genome Browser (Hubbard et al., 2002). As shown in
the Appendix, the transcription start site of the putative gene BC033162 is 1166 bp
away from the transcription start site of the U6-1 gene.
If the BC033162 mRNA would be actively transcribed in pem HeLa cells, the
mRNA of this gene should be detectable. RNA from pem HeLa cells was either reverse
transcribed with a BC033162-specific primer or with an oligo(dT) primer. Primers
located in exon 3 were used in PCR, which where functional in amplifying genomic
DNA. With neither cDNA, the expected product of 129 bp could be detected in PCR
after 30 cycles (Figure 8). The cDNA was intact, since PGK1 could be successfully
PCR amplified with the same oligo dT cDNA and with gene-specific RT.
Part II: Results
72
Figure 6. Pol II and Pol III accumulate on the U6-1 promoter by ChIP.Sheared chromatin from pem HeLa cells that had been crosslinked with formaldehyde wasimmunoprecipitated with anti-Pol IIa (8WG16; right panel) and anti-Pol III (1900; left panel) or non-immune IgG antibodies. DNA was purified following reversal of crosslinks and subjected to Real TimePCR. All values are relative to non-immune IgG and normalized to an intergenic control region. Errorbars represent the SEM. Pol II, n = 6; Pol III, n = 8. (In collaboration with A. Bledau)
Figure 7. Pol II accumulates on the U6-1 promotor irrespective of CTD phosphorylation and ina different cell line than HeLa.(A) ChIP analysis of total Pol II accumulation with an antibody specific for a non-CTD epitope (Pol3/3) onU6-1 in pem HeLa cells. All values are relative to non-immune background ChIP experiments andnormalized to an intergenic control region. One representative experiment is shown. (B) ChIP analysis ofPol II (8WG18) and Pol III accumulation on U6-1 in HEK cells. All values are relative to non-immunebackground ChIP experiments and normalized to an intergenic control region. One representativeexperiment is shown. (In collaboration with A. Bledau)
Part II: Results
73
Total RNA treated with DNase I and RNase A and the no RT control revealed that the
PGK1 band was specifically derived from reverse transcribed RNA. This result suggest
that the putative BC033162 gene is not at all or at very low levels that cannot be
detected after 30 cycles of PCR.
If the BC033162 gene would indeed be transcribed at very low levels, the possibility
exists that Pol II is present at the promoter of this gene, be it in a paused or pre-initiated
state (see part I of this thesis). Pol II bound 1 kb downstream from the U6-1 promoter
could only be efficiently recovered by ChIP if a major part of the chromatin fragments
were bigger than 1 kb. To test this hypothesis, the size of the chromatin fragments
generated in this assay was determined. DNA from formaldehyde-crosslinked and
sonicated pem HeLa cell extracts was purified and subjected to PCR with primers
generating differently sized products of the β-actin gene. To account for differences in
primer efficiency, a plasmid containing the human genomic β-actin was subjected to the
same PCR conditions. The PCR products were analyzed by gel-electrophoresis and the
band intensities were measured. The bands deriving from chromatin PCR were
normalized to the band intensities from the genomic clone and the smallest fragment
value was set to 100 % (Figure 9). With the sonication conditions used for the ChIP
assay, only 20 % of the fragments are bigger than 1 kb, and the majority of fragments
are smaller than 680 bp. Judging from this crude assay, it is a fair assumption that Pol II
bound more than 1 kb away from the region of interest would be recovered by ChIP
with low efficiency. These results indicate that the BC033162 gene is not actively
transcribed in pem HeLa cells and therefore the Pol II ChIP signal from Figure 6 likely
derives from Pol II bound to the U6-1 gene promoter region.
In order to define the position of Pol II and Pol III accumulation on the U6-1
gene locus, ChIP primer walking was performed. For this purpose, primers were
designed that amplify gene regions from ca. 1200 bp upstream and downstream of the
U6-1 transcription start site. The primer pair giving the highest value for either Pol II or
Pol III ChIP data set was set to 100 % and the other primer pair values were normalized
accordingly. Figure 10 illustrates the surprising finding that Pol II and Pol III
accumulate in two distinct peaks. The major concentration of Pol II is with the primer
pair spanning the region from –403 bp to –186 bp with the center point at –295 bp
relative to the transcription start site.
Part II: Results
74
Figure 8. The U6-1 adjacent gene BC033162 is not expressed.cDNA was generated from pem HeLa total RNA treated either with DNAse 1 and/or RNase A. For reversetranscription, BC033162 and PGK1 gene-specific primers (lane 2, 3 and 8, 9, respectively) or oligo-dTprimers (lane 4, 5, 10, 11) were used. cDNA , genomic DNA (lane 7) and no RT controls (lane 6, 12)were PCR amplified with primers in BC033162 exon3 (2-7) or PGK1 exon3-5 (8-12) for 30 cycles. Lane1, 100 bp marker. PCR product sizes are indicated at the right end.
Figure 9. Determination of chromatin fragment size.Pem HeLa cells were formaldehyde-crosslinked, extracts were sonicated and purified DNA was PCR-amplified with primers located in the β-actin gene (lane 1-5). As a standard, the same primers wereused for PCR with a plasmid harbouring a β-actin genomic clone (lane 6-10). Bands were separated on a1 % agarose gel and intensities were measured and normalized to the genomic clone PCR bands. Lane11, 100 bp marker. The smallest fragment value was set to 100 %. PCR product sizes are indicated atthe right end. (In collaboration with A. Bledau)
Part II: Results
75
Figure 10. ChIP Primer walking along U6-1 gene locus in pem HeLa cells.Schematic representing the genomic region 1500 bp upstream and downstream from the 107 bp U6-1snRNA gene. The transcription start of the neighboring gene BC033162 is indicated at the right end.Black lines specify the regions amplified by primer sets, identified by the nucleotide at the center of theamplified region. In the diagram below, data points are placed according to the center positions of thePCR products along the region. Antibodies specific for hypo-phosphorylated Pol II (8WG16) and Pol IIIwere used. All values are relative to non-immune background ChIP experiments and normalized to anintergenic control region. The peak value for each data set was set to 100%. Error bars represent theSEM. The data represent the average of at least three independent experiments. (In collaboration withA. Bledau)
Pol III on the other hand is maximally concentrated right on top of the transcribed gene
at he amplified region from +6 to +184 with the center point +95 bp. It is noteworthy
that the resolution of the ChIP assay appears to be remarkably high and amounts to ca.
200 bp since Pol II accumulation at the center point positions –295 and –76 are clearly
separated from each other.
The CTD is phosphorylated at Serine 5 residues when Pol II initiates
transcription (reviewed in Palancade and Bensaude, 2003). In order to test whether the
transcriptionally active form could be detected along the U6-1 locus, ChIP primer
walking with the H14 antibody that recognizes phosphorylated Serine 5 was performed
(Figure 11). Significant amounts of phosphorylated Pol II were bound to the U6-1
upstream promoter region. The distribution of H14 reactive Pol II is comparable to the
accumulation pattern observed for the hypo-phosphorylated form in Figure 10, with
major association to the –735 and –295 positions. These results clearly show that Pol II
Part II: Results
76
is not loosely associated to the U6-1 promoter region, but has undergone significant
CTD phosphorylation, indicating that it has initiated transcription.
Figure 11. Transcriptionally active Pol II associates with the U6-1 locus.ChIP Primer walking along U6-1 gene locus in pem HeLa cells with the H14 antibody recognizing thephosphorylated form of Pol II CTD at Serine 5. Schematic representing the genomic region 1500 bpupstream and downstream from the 107 bp U6-1 snRNA gene. Black lines specify the regions amplifiedby primer sets, identified by the nucleotide at the center of the amplified region. In the diagram below,bars are placed according to the center positions of the PCR products along the region. All values arerelative to non-immune background ChIP experiments and normalized to an intergenic control region.Error bars represent the standard deviation. The data represent the average of two independentexperiments. (In collaboration with A. Bledau)
Pol II and Pol III accumulate on other U6 snRNA genes
Knowing that Pol II accumulates at the more distal promoter region and Pol III more at
proximal promoter regions, it was interesting to find out if this was also true for other
U6 snRNA genes. Primers were designed for the proximal upstream regions of 7 other
previously identified U6 genes (Domitrovich and Kunkel, 2003) and the abundance of
Pol III in those regions was examined by ChIP (Figure 12). The relative abundance of
Pol III on all 8 U6 snRNA genes studied was obtained by normalizing the values
obtained for U6 ChIP to the Pol III transcribed tRNALeu gene, which was set to 100%.
This normalization demonstrates the differences in RNA Pol III occupancy at different
promoters and accounts for the inter-experimental differences in ChIP efficiency. Pol III
Part II: Results
77
did accumulate strongly on U6-8, U6-9 and U6-2 proximal promoter regions and to a
somewhat lesser extend at U6-1 and U6-7. Pol III did neither bind to the Pol II-
transcribed PGK1 gene nor to U6-3, U6-4, and U6-6 that had in fact been formerly
reported to lack identifiable promoter elements (Domitrovich and Kunkel, 2003). The
authors of that study did not detect AcH4 or TBP bound to these three U6 genes by
ChIP. The absence of Pol III on those promoters is in agreement with their observation,
strongly indicating that in fact U6-3, U6-4, and U6-6 are transcriptionally inactive.
Having established that Pol III is associated to the four other active U6 genes
with similar promoter structures as U6-1, it was tantalizing to find out if also Pol II
would be associated to these genes.
Figure 12. Pol III associates with five U6 genes at promoter-proximal positions.ChIP analysis of Pol III at various U6 genes, tRNALeu and PGK1 in pem HeLa cells. All values are relativeto non-immune background ChIP experiments and normalized to an intergenic control region. ThetRNALeu value in each data set (= same ChIP but different primers) was set to 100% to demonstrate thedifferences in RNA Pol III occupancy at different promoters and to account for the inter-experimentaldifferences in ChIP efficiency. Error bars represent the SEM after this transformaton. The data representthe average of at least three independent experiments. (In collaboration with A. Bledau)
First, it had to be determined whether any other Pol II gene resides in close vicinity to
the U6-2, U6-7, U6-8 and U6-9 genes. Serving as a negative control, U6-4 was also
analyzed for neighboring genes. The Ensembl and UCSC Genome Browsers were
consulted and neighboring genes of U6-2 and U6-9 were identified (see Appendix). The
U6-2 and C19orf6 transcription start sites are 407 bp apart from each other, and the
transcription start of U6-9 and THRAP5 5’UTR are only 266 bp apart. It first had to be
Part II: Results
78
checked if those genes are transcriptionally active, hence the presence of THRAP5 and
C19orf6 mRNA in pem HeLa cells was determined.
Figure 13 shows that primers located in exon 17 of THRAP5 were functional in
PCR amplifying genomic DNA. Total RNA was treated with DNase I or RNase A from
pem HeLa cells prior to reverse transcription with an oligo(dT) primer and the expected
product of 348 bp could be detected after 26 cycles of PCR. PGK1 was PCR-amplified
with the same oligo(dT) cDNA as a control, resulting in a 400 bp product. The DNase I
or RNase A treatment and the no RT control demonstrated that the THRAP5 and PGK1
PCR products were specifically derived from reverse transcribed RNA. For C19orf6,
primers located in exon 2 were tested in PCR with genomic DNA giving a product of
the expected size of 247 bp. RNA from pem HeLa cells was DNase I or RNase A
treated, followed by gene-specific or oligo(dT) reverse transcription. With each cDNA,
the expected product of 247 bp could be detected in PCR after 26 cycles. PGK1 was
PCR-amplified with the same oligo(dT) cDNA and with gene specific RT as a control.
The C19orf6 and PGK1 band were specifically derived from reverse transcribed RNA,
as no product was obtained from the no RT control and the RNase treated sample
(Figure 14). These results indicate that the THRAP5 and C19orf6 are actively
transcribed in pem HeLa cells.
To test a possible Pol II accumulation on the active U6 genes, ChIP experiments
were performed with the Pol IIa specific 8WG16 antibody with primers located in
proximal promoter regions (U6-2 and U6-9 due to the reasons mentioned above) and
distal promoter regions (U6-7, U6-8) resembling the primer positions for the U6-1 ChIP
primer walking experiment (Figure 15). Pol II accumulates at background levels at the
tRNALeu gene and the inactive U6-4 but is strongly detectable at the distal promoter
regions of U6-7 and U6-8. Significant accumulation signals were also observed at the
more proximal promoter regions of U6-2 and U6-9. Although the resolution of the ChIP
assay appears to be quite high, it cannot be excluded that Pol II binding to U6-2 or U6-9
promoters is due to its engagement at the C19orf6 or THRAP5 promoter, respectively.
Taken together, is it conclusive that Pol II associates with all of the actively transcribed
U6 promoter regions.
Part II: Results
79
Figure 13. Figure x. The U6-9 adjacent gene THRAP5 is expressed in pem HeLa cells.cDNA was generated from pem HeLa total RNA treated either with DNase 1 and/or RNase A. For reversetranscription, oligo-dT primers (lane 2, 3, 6, 7) were used. cDNA , genomic DNA (lane 5) and no RTcontrols (lane 4, 8) were PCR amplified with primers in THRAP 5 exon 17 (2-5) or PGK1 exon 3-5 (6-8)for 26 cycles. Lane 1, 100 bp marker. PCR product sizes are indicated at the right end.
Figure 14. The U6-2 adjacent gene C19orf6 is expressed in pem HeLa cells.cDNA was generated from pem HeLa total RNA treated either with DNase 1 and/or RNase A. For reversetranscription, C19orf6 and PGK1 gene-specific primers (lane 2, 3 and 8, 9, respectively) or oligo-dTprimers (lane 4, 5, 10, 11) were used. cDNA , genomic DNA (lane 7) and no RT controls (lane 6, 12)were PCR amplified with primers in C19orf6 exon 2 (2-7) or PGK1 exon 3-5 (8-12) for 26 cycles. Lane 1,100 bp marker. PCR product sizes are indicated at the right end.
Part II: Results
80
Figure 15. Pol II associates with the active U6 genes at promoter-proximal positions.ChIP analysis of Pol II at various U6 genes, tRNALeu and PGK1 in pem HeLa cells. All values are relativeto non-immune background ChIP experiments and normalized to an intergenic control region. The PGK1value in each data set (= same ChIP but different primers) was set to 100% to demonstrate thedifferences in RNA Pol II occupancy at different promoters and to account for the inter-experimentaldifferences in ChIP efficiency. Error bars represent the SEM after this transformation. The data representthe average of at least three independent experiments. (In collaboration with A. Bledau)
α-amanitin inhibits U6 maxigene transcription in vivo
If Pol II has any functional role in U6 snRNA transcription, then U6 snRNA expression
should be decreased upon Pol II transcription inhibition. α-amanitin is a well-studied
transcription inhibitor that, at low concentrations, selectively inhibits Pol II (Wieland
and Faulstich, 1991 and references therein). Since the U6 snRNA is very abundant
(2x105 copies/cell, Birnstiel, 1988) and stable in HeLa cells and is presumably
transcribed from multiple genes, the influence of α−amanitin on individual endogenous
U6 genes could not easily be determined. Therefore, as a first step, HeLa cells were
transfected with plasmids harboring a U6 maxigene with a 9 bp insert at the 3’end of the
transcribed region and the genomic 5’ and 3’ flanking regions including all regulatory
sequences (Domitrovich and Kunkel, 2003, see Figure 16; Hernandez, 2001). The 9 bp
insert serves as a primer binding site for reverse transcription, followed by PCR for
Part II: Results
81
exclusive detection of the plasmid-derived U6 snRNA. U6-1, U6-8 and U6-9 maxigenes
were transiently transfected into HeLa cells and the medium was supplemented
simultaneously with α-amanitin oleate, a cell-permeable derivate of α-amanitin. Upon
α-amanitin oleate treatment, the expression of U6-9, U6-1 and U6-8 maxigenes was
severely decreased to 18 %, 12 % and 4 %, respectively, compared to the untreated
control (Figure 17). 5S rRNA and pre-tRNATyr, that are Pol III transcripts, were either
unchanged upon α-amanitin oleate treatment (5S rRNA) or decreased only moderately
to 82 % compared to untreated cells. As a positive control, LDHA mRNA was also
severely decreased to 24 % after α -amanitin oleate treatment. The drastic
downregulation of the U6 maxigenes that are driven by their endogenous promoter
clearly indicates that Pol II is required for transcription of these genes.
Figure 16. U6-maxigenes for transient expression experiments.Human U6-1, U6-8 and U6-9 genes plus genomic flanking sequence were cloned into either pGEM2 orpGEM-T and a 9bp XhoI site was inserted at nucleotide 87 to allow for selective detection of plasmid-derived U6 snRNA (Domitrovich and Kunkel, 2003).
Part II: Results
82
Figure 17. U6-1 and U6-9 maxigene expression is downregulated upon α-amanitin oleatetreatment.HeLa cells were transfected by the calcium phosphate technique with either U6-1, U6-8 or U6-9maxigene plasmids and treated simultaneously with 50 nM α-amanitin oleate for 24 h or left untreated.Expression of U6 maxigenes, 18S rRNA, LDHA, 5S rRNA and pre-tRNATyr was detected by reversetranscription, followed by conventional (upper panel) or quantitative PCR (bar diagram). All values werenormalized to 18S rRNA and expression levels are denoted relative to untreated controls. Error barsrepresent the SEM. The data represent the average of at least three independent experiments. (Incollaboration with A. Bledau)
Endogenous U6 snRNA is affected through α-amanitin treatment
To determine whether low concentrations of α-amanitin inhibit endogenous U6 snRNA
transcription, HeLa cells were grown for 12 h in α -amanitin before they were
metabolically labeled with 32P orthophosphate. The cells were subjected to
immunoprecipitation (IP) with the anti-Sm antibody Y12, which recognizes the core
U1, U2, U4, U5, and U4/U6 di-snRNPs. Moreover, IP with anti-Lsm4 and anti-SART3
that recognize the U6 snRNP as well as the U4/U6 di-snRNP was performed. The RNA
was extracted and run on a polyacrylamide gel. The rationale behind this experiment
was that only RNAs transcribed during the α-amanitin treatment i.e. during reduced Pol
II activity, would receive a radioactive label, and only those incorporated into snRNPs
would be precipitated. Most likely, the Lsm proteins are sufficiently abundant in the
HeLa cell to incorporate into U6 snRNPs regardless of α-amanitin treatment, since U6
Part II: Results
83
snRNPs and consequently Lsm proteins are very abundant (105 – 106 U6 snRNPs per
cell, Birnstiel, 1988), and Lsm proteins can load on U6 snRNA independent of U4
snRNA transcribed by Pol II (reviewed in Gerbi et al., 2003). α-amanitin treatment
dramatically reduces the total amount of newly synthesized snRNAs but not 5S rRNA
(Figure 18; compare lane 1 and 2).
Figure 18. The major 107 nt U6 snRNA form in snRNPs is downregulated upon α-amanitintreatment.HeLa cells were grown for 12 h in α-amanitin or left untreated before medium was supplemented with 32-P orthophosphate for 7,5 h. Extracts were subjected to standard immunoprecipitations with Y12 (anti-SM), anti-Lsm4 and anti-SART3. The RNA was extracted and run on a denaturing polyacrylamide gel toresolve the different RNA species. Only snRNA that were synthesized during reduced Pol II activity wereradioactively labeled and would be precipitates when incorporated into snRNPs. The arrowheads point tothe sharp 107 nt U6 snRNA band. (In collaboration with K. Neugebauer)
Treatment with α-amanitin results in less snRNA pulled down with anti-SM (compare
lane 3 and 4). U6 and U4 snRNAs are pulled down with anti-Sm (lane 3), anti-Lsm4
(lane 5) and anti-SART3 (lane 7) in the context of the U6 snRNP alone and the U4/U6
di-snRNP. Interestingly, U6 snRNA migrates in a sharp band and in a higher molecular
weight smear, which might be the polyuridylated form. The U4 snRNA and
interestingly also U6 snRNA decreased upon α-amanitin treatment (lane 4, 6, and 8).
Since 5S rRNA production was not inhibited by α-amanitin, it is conclusive that Pol III
transcription was unaffected. Interestingly, the sharp U6 snRNA band was strongly
Part II: Results
84
reduced upon α-amanitin treatment, and only the higher molecular weight smear
remained. Since both antibodies, anti-SART3 and anti-Lsm4, recognize the free U6
snRNP, the decreased U6 snRNA signal cannot only be explained by the lack of U4
snRNA that is needed to form the U4/U6 di-snRNP. On the basis of these observations,
the possibility can be considered that the synthesis of endogenous U6 snRNA and/or 3’
end formation is affected by α-amanitin treatment.
Endogenous U6 snRNA is polyadenylated
If Pol II was indeed somehow involved in U6 snRNA transcription, the Pol III
termination signal of 5 T-residues might not be used, resulting in unusual 3’ ends.
Therefore, pem HeLa total RNA treated with DNase I or RNase A was reverse
transcribed with oligo(dT) primers, followed by PCR amplification with primers located
in the coding U6 snRNA region (Figure 19). Unexpectedly, significant amounts of
product were amplified, suggesting that polyadenylated U6 snRNA species exists in the
cell. Since the reverse transcription was conducted at 50 °C, and the longest poly(A)
stretch in the U6 snRNA gene is only four A residues long and resides in the 5’ end of
the mature RNA, the possibility of unspecific priming can be excluded. However,
whether these polyadenylated U6 snRNAs are synthesized by Pol II and are functional
in the context of the U6 snRNP has to be the subject of future experiments.
Figure 19. Endogenous U6 snRNA is polyadenylated.cDNA was generated from pem HeLa or HeLa KN total RNA treated either with DNase 1 or RNase A. Forreverse transcription, oligo-dT primers (lane 1-12) were used. cDNA , genomic DNA (lane 13) and no RTcontrols (lane 3, 4, 9, 10) were PCR amplified with U6 snRNA primers (nts 4-97; lanes 1-6) or LDHAexon 3-5 (lanes 7-12) for 27 cycles. M, 100 bp marker. PCR product sizes are indicated at the right end.
Part II: Discussion
85
Discussion
In this study, it was shown that Pol II is bound to the distal promoter regions of the
active U6 snRNA genes U6-1, U6-2, U6-7, U6-8, and U6-9 in vivo, along with Pol III at
promoter-proximal positions. ChIP primer walking along the U6-1 genomic locus
revealed that Pol II and Pol III accumulate in two distinct peaks, with Pol II bound ~
300 bp upstream of the transcription start site and Pol III right on top of the U6 gene. It
was further shown that Pol III does not bind to proximal promoter regions of U6-3, U6-
4 and U6-6, consistent with the conclusion of a previous study (Domitrovich and
Kunkel, 2003) that these genes are indeed inactive. The accumulation of Pol II appears
to be independent of neighboring Pol II genes at least for U6-1, U6-7 and U6-8. So far,
it was not subject of this study whether other proteins found in the Pol II transcription
complex are present at the distal promoter region. It would be interesting to test TFIIB
and TFIIIB accumulation along the U6-1 locus and see if these GTFs accumulated in
the same two distinct peaks as seen for Pol II and Pol III.
The key question is whether Pol II accumulation on the U6 promoters indicates
a functional role in U6 transcription. To address this question, ChIP with antibodies
specific for phosphorylated Serine 5 of Pol II CTD along the U6-1 locus were
performed. The rationale behind this experiment was that only initiated Pol II would be
phosphorylated at its CTD (reviewed in Palancade and Bensaude, 2003). These
experiments revealed that indeed Serine 5 phosphorylated Pol II accumulated
significantly on the -295 bp and –735 bp upstream positions, but not significantly on the
U6-1 coding region. It would be interesting to test if transcripts could be detected that
are derived from this upstream region. One possibility of course could be that Pol II is
engaged in transcription at this region transcribing away from the U6 gene. It should be
noted that previous research indicates that Pol II is generally very poorly detectable in
downstream regions of even highly expressed genes (Cheng and Sharp, 2003, see part I
of this thesis). So the failure of detecting significant amounts of Pol II at the U6 coding
region by ChIP does not necessarily imply that no Pol II is bound there.
Another interesting question is whether Pol II and Pol III associate
independently from each other to the U6 snRNA gene in different cells in a population
of HeLa cells. It might be that at different stages of the cell cycle, one of the two
Part II: Discussion
86
polymerases predominantly binds to the U6 snRNA gene and/or transcribes it (see
Figure 20). Ongoing experiments in the lab are dedicated to establish sequential ChIP
by first immunopurifying chromatin crosslinked to Pol II and subjecting the precipitates
to a second ChIP with Pol III antibodies. Only if both polymerases bind to the U6
promoter in the same cell, the U6 gene region can be recovered in the end. However,
since the resolution of the ChIP assay seems to be quiet high (see Figure 10), the
recovered pieces of DNA fragments must be ~ 400 bp long so that the major
accumulation sites of both polymerases can be captured. Most probably, the sonication
conditions for the sequential ChIP assay have to be adjusted for this purpose.
A different explanation for the presence of Pol II at the U6 snRNA promoter
region could be that Pol II is having a stimulatory effect on Pol III transcription at this
particular class of genes. This would explain the decrease of U6 snRNA in the Pol II
inhibition studies. Pol II could alternatively have stimulatory effects on the recruitment
of 3’end processing factors like the La protein, which had been shown to bind co-
transcriptionally to the U6 snRNA gene (Fairley et al., 2005). Inhibition of Pol II would
lead to its displacement from the U6 snRNA promoter region, which could in turn lead
to U6 snRNA species with 3’end processing defects (see Figure 20).
The literature on U6 snRNA transcription leaves little room for doubt that Pol
III transcribes the U6-1 snRNA gene. However, many of the published experiments
addressed this question in highly purified in vitro systems. The original publications
(Kunkel et al., 1986; Reddy et al., 1987) reported that U6 snRNA is accurately
transcribed in HeLa S100 extract that lacks Pol II activity and was not inhibited by 1
µg/ml α-amanitin but 200 µg/ml, a concentration known to also inhibit Pol III (Wieland
and Faulstich, 1991). Furthermore, U6 snRNA was transcribed at low α-amanitin
concentrations in isolated HeLa nuclei but not at high concentrations. Better in vivo
proof was provided in S. cerevisiae, in which U6 expression was decreased in a Pol III
temperature sensitive mutant (Moenne et al., 1990) and in X. tropicalis oocytes, in
which an injected U6 plasmid was still transcribed at low but not at high α-amanitin
concentrations (Krol et al., 1987). However, in the same system, primer extension
analysis of the injected U6 revealed that two U6 snRNA species were transcribed, one
of which was α-amanitin sensitive (Mattaj et al., 1988). Along this line, also the 7SK
gene injected into X. tropicalis oocytes was transcribed by both Pol II and Pol III (Boyd
et al., 1995). The 7SK gene is very similar in promoter structure to the U6 snRNA gene
Part II: Discussion
87
and belongs into the same class of type 3 Pol III promoters. Furthermore, a human U6
promoter followed by a C-free cassette was transcribed by both Pol II and Pol III in
HeLa nuclear extracts (Park and Kunkel, 1995).
The α-amanitin sensitivity of U6 transcription in HeLa cells was addressed with
transfected maxigenes of individual U6 genes. Surprisingly, transcription of U6 from
maxigenes was severely decreased upon α-amanitin treatment with low concentrations
of α-amanitin that did not inhibit Pol III transcription of tRNA or 5S rRNA genes. The
expression of U6-9, U6-1 and U6-8 maxigenes driven by their endogenous promoter
was reduced down to 18 %, 12 % and 4 %, respectively, which provides direct in vivo
evidence for a functional role of Pol II in U6 snRNA transcription. Transfected plasmid
DNA is chromatinized in mammalian cells (Reeves et al., 1985), however, it is
unknown whether the U6 plasmid DNA is packed into nucleosomes comparable to the
chromosomal U6 loci and if this packing would account for fundamental differences in
transcription. It would by very interesting to know if Pol II and Pol III would associate
with the same regions on the plasmid DNA as in the chromosomal U6-1 region (see
Figure 10). However, such “plasmid-ChIP” experiments have been proven to be
technically impossible in our hands.
The U6 snRNA promoter (as member of the Pol III type 3 promoters) is
particularly interesting because of its usage in short hairpin RNA (shRNA) expression
vectors for RNA interference (Miyagishi and Taira, 2002) (Paul et al., 2002) (Sui et al.,
2002) (Yu et al., 2002). These vectors typically contain the full human or mouse U6
snRNA promoter followed by an inverted repeat with the individual motif being ca. 21
nt long and corresponding to a part of the coding region of the gene of interest. A spacer
of ca. 6 nt separates the two motifs that form the inverted repeat. Downstream of the
second motif, five T residues mark the Pol III transcription termination site.
Interestingly, it had been recently shown that transcription from the same U6 promoters
driving shRNA expression for RNA interference experiments did not terminate at five T
residues but instead could transcribe downstream vector backbone sequence or a
luciferase reporter gene. Luciferase expression driven by the human H1 promoter, also a
member of Pol III type 3 promoters and used in shRNA expression systems is
furthermore sensitive to α-amanitin (Rumi et al., 2006).
These results are in agreement with the observation that α-amanitin reduces
dramatically the expression from maxigene-derived U6-1, U6-8 and U6-9 in vivo,
Part II: Discussion
88
indicating that Pol II activity is required for transcription from the U6 snRNA promoter.
However, this study could not demonstrate whether Pol II is the enzyme that
synthesizes U6 snRNA in this maxigene system or if Pol II might influence Pol III
transcription by an unknown mechanism. The U1 und U2 snRNA promoters are very
similar to the U6 snRNA promoter in their structure and active transcription relies on
the combined action of Oct-1, the SNAPc complex and the Pol II or Pol III transcription
machinery, respectively (Hernandez, 2001). The possibility exists that due to this close
relationship, the U6 snRNA promoter has not lost its dependence on Pol II. No
mechanism has been reported so far that Pol II can in any way assist Pol III recruitment,
transcription or termination, but this possibility should also be considered (Figure 20).
The question whether endogenous U6 snRNA is sensitive to α -amanitin
treatment is difficult to address and all insightful experiments contain caveats. Since U6
snRNA is very abundant in the cell and possess a considerable long half-life
(approximately 24 h, Fury and Zieve, 1996), a long α-amanitin treatment could interfere
with cell viability, resulting in a consequent downregulation of all RNAs including U6
snRNA. Another way of analyzing the dependence of newly synthesized RNAs on
functional Pol II is 32P orthophosphate metabolic labeling in the presence of α-amanitin.
The recovery of U6 snRNA from anti-Lsm4, anti-Sm and anti-SART3 precipitated
snRNPs indicates that enough snRNP proteins were present and snRNP assembly was
still functional regardless of α-amanitin. Considerably less U6 snRNA was recovered
after α-amanitin treatment, 5S rRNA levels however stayed the same, indicating that
Pol III transcription was unaffected. One possible explanation of this observation could
be that Pol II is involved in either U6 snRNA transcription or maturation, be it direct or
indirect by assisting Pol III and/or recruitment of processing factors such as the La
protein to the site of transcription. The loss of the sharp U6 snRNA band and the higher
molecular weight smear after α-amanitin treatment is indicative of problems in U6
snRNA transcription and/or end formation. Since the 5’ end of mature, snRNP-
assembled U6 snRNA is capped and consequently blocked for further addition of
nucleotides, the only logical explanation for longer U6 snRNA species would be
changes in 3’ end formation. Post-transcriptional 3’ uridylation had been reported to
occur for U6 snRNA (Rinke and Steitz, 1982), which could explain the longer U6
snRNA species in the untreated snRNP immunoprecipitation experiments. The
disappearance of a defined U6 snRNA band after α-amanitin treatment, however, could
Part II: Discussion
89
be due to problems in trimming back 3’ end extensions to 107 nucleotides. Another
possibility might be an α-amanitin induced amplification of mechanisms driving U6
snRNA 3’ end extensions or a general failure of the transcription machinery to
terminate transcription at the five T residues stretch.
To begin to shed light into this, endogenous U6 snRNA species were checked
for poly(A) extensions. By oligo dT RT followed by PCR with primers located in the
U6 snRNA coding region, polyadenylated U6 snRNA was detected in HeLa cells.
However, if these U6 snRNA forms are functional or cryptic Pol II transcripts is to date
unclear. It had been reported that in yeast, Pol II synthesizes many cryptic transcript that
are short-lived and targeted to the exosome following polyadenylation (Wyers et al.,
2005). This mechanism might also be possible for the U6 snRNA gene region, although
the α-amanitin sensitivity of U6 maxigenes strongly argues against an unspecific role of
Pol II in U6 snRNA biogenesis.
Another possibility might be that longer U6 snRNA species are due to a
considerable read-though by Pol II over the cognate Pol III termination site.
Alternatively, these longer U6 snRNAs might be aberrant Pol III transcripts due to Pol
III failing to terminate the transcription properly. This termination malfunction might be
enhanced through α-amanitin treatment, resulting in the disappearance of the sharp U6
snRNA band. It would be interesting to determine whether these polyadenylated and
longer U6 snRNA species are α-amanitin sensitive, and experiments in this direction
are in preparation. It would be of great interest to determine the half-life of these longer
U6 snRNA species to draw conclusions about the effect of α-amanitin on their
synthesis.
Unfortunately, there is no cell permeable inhibitor specific for Pol III available,
so the reverse question whether Pol III inhibition has an effect on the U6 snRNA
expression from maxigenes cannot be asked in vivo. However, one future key
experiments to determine the role of Pol III in endogenous U6 snRNA biogenesis could
involve cells expressing α-amanitin resistant Pol II mutants. Since Pol III is inhibited by
high concentrations of the drug (Pol III is 103 times less sensitive to α-amanitin;
Wieland and Faulstich, 1991), an 32P metabolic labeling experiment in these cells
treated with high concentrations of α-amanitin could resolve the in vivo dependence of
U6 snRNA biogenesis on Pol III.
Part II: Discussion
90
Figure 20. Possible mechanisms of Pol II involvment in U6 snRNA biogenesis.(1) Pol II associated with the upstream promoter region has stimulatory effects on Pol III transcriptinand/or the recruitment of 3’end processing factors. (2) Both Pol II and Pol III transcribe U6 snRNA at thesame time in the same cell. (3) In a subpopulation of cells, Pol II (which might elongate faster than PolIII and is thereore not captured by ChIP easily) transcribes the U6 snRNA gene, whereas in the othercells, Pol III is responsible for U6 snRNA transcription.
Part II: Materials and Methods
91
Materials and Methods
Cell culture and treatments
The pem HeLa cell line used for ChIP experiments originates from the normal HeLa
(Human cervix carcinoma) cell line and contains ~60 copies of the rat pem homeobox
gene stably integrated into its genome. Pem HeLa cells were grown in DMEM (Gibco)
supplemented with 10% fetal bovine serum (FBS) and 1% of penicillin/streptomycin
(Gibco). HeLa KN, the HeLa cell line routinely used in our lab, was used for maxigene
transfection experiments. HeLa KN cells were grown in DMEM supplemented with
10% FBS and 1% of penicillin/streptomycin to 60-80% confluency before the medium
was changed to growth medium containing either 50 nM Methyl a-Amanitin Oleate
(Merck) or DMSO. 10 µg of each U6 maxigene plasmid (a gift from Gary G. Kunkel,
(Domitrovich and Kunkel, 2003) was transfected with CaPO4 (Ausubel, 2001). The
medium was replaced by fresh medium containing 50 nM Methyl a-Amanitin Oleate 16
h later and growth was continued for 6 h until RNA was extracted with TRIZOL
(Invitrogen).
Mapping of potential human U6 snRNA genes
BLAT search [blast like Alignment Tool] from UC-Santa Cruz genome bioinformatics
site (http://genome.ucsc.edu; last search from January 2006) was performed with all 9
identified U6 snRNA genes (Domitrovich and Kunkel, 2003), using the 107 nt U6
snRNA sequence plus ~300 bp upstream and ~100 bp downstream. The designation of
the nine U6 snRNA genes was kept as suggested in (Domitrovich and Kunkel, 2003).
The genomic location as well as neighboring genes was identified (see Appendix). The
originally isolated gene is referred to as U6-1.
Chromatin Immunoprecipitation and Real Time PCR
A modification of the technique described was used (Kuo and Allis, 1999). Briefly,
approximately 108 cells were crosslinked with a final concentration of 1%
formaldehyde added directly to the medium. Cells were washed 2 times with cold PBS,
scraped and collected. Cell pellets were resuspended in 2 ml of SDS lysis buffer (1%
Part II: Materials and Methods
92
SDS, 10 mM EDTA, 50 mM Tris-HCl, pH 8.1) containing complete protease inhibitor
cocktail (Roche) and incubated for 10 min on ice. Cell extracts were sonicated with a
Branson sonifier W-450 D at 30 % amplitude with 15 times 10 s bursts resulting in
∼500 bp chromatin fragments and then centrifuged for 10 min at 14,000 rpm. 50 µl of
the supernatant was saved as input DNA and the remainder was diluted 1:10 in ChIP
dilution buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl,
pH 8.1, 167 mM NaCl) containing protease inhibitors. 2 ml chromatin solution was pre-
cleared at 4°C with sepharose beads for 1 h before overnight incubation (4°C) with
either 10 µg of 8WG16 (Neoclone), 20 µg of Pol3/3 directed against the internal region
F of Pol II Rpb1 (a gift from D. Eick), 4 µl of rabbit polyclonal anti RNA Pol III 1900
directed against the RPC 155 subunit of human Pol III (a gift from Robert J. White,
Fairley et al., 2003), and 10 µg of non immune IgG (Sigma) as control. Complexes were
immunoprecipitated with GammaBind G sepharose beads (Pharmacia Biotech) blocked
with 0.2 mg/ml salmon sperm DNA and 0.5 mg/ml BSA for 1 h at 4°C. The beads were
washed rocking for 4 min in each of the following buffers: Low Salt Immune Complex
Wash Buffer (0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1,
150 mM NaCl), High Salt Immune Complex Wash Buffer (0.1% SDS, 1% Triton X-
100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.1, 500 mM NaCl), LiCl Immune Complex
Wash Buffer (0.25 M LiCl, 1% NP-40, 1% deoxycholic acid, 1 mM EDTA, 10 mM
Tris-HCl, pH 8.1) and twice in TE. The immune complexes were eluted in 1%SDS and
50mM NaHCO3 and crosslinks reversed for 6 h at 65°C. Samples were digested with
proteinase K for 1 h at 45°C and the DNA extracted with the Qiagen PCR purification
kit.
DNA templates retrieved by ChIP were analyzed by quantitative real-time PCR on a
Stratagene MX3000, using the SYBR Green method (ABsolute QPCR SYBR Green
Rox Mix, AB Gene). The reaction volume was 20 µl, with 4 µl DNA template (input
1:10) and 90-900 nM of each primer according to individual optimization. Cycling was
done for 15 min at 95°C, followed by 40 cycles for 30s at 95°C, 1 min at 60°C, and 30s
at 72°C, with measuring at the end of the annealing step. Dissociation curves were
obtained by heating the samples to 95°C followed by cooling down to 55°C and
successive heating of the samples to 95°C with continuous measurement. Primer sets
distinguishing between different regions of the genes are available upon request.
Part II: Materials and Methods
93
The relative proportions of co-immunoprecipitated gene fragments were determined
based on the threshold cycle (Ct) for each PCR product. Data sets were normalized to
ChIP input values, then the Ct values from pol II and splicing factor ChIP were
subtracted from the Ct values obtained from templates derived from ChIP with
unspecific antibody (2 [Ct(unspec)-Ct(input) ] – [Ct(spec)-Ct(input) ] (Chakrabarti et al., 2002). The
fold difference over background obtained for gene regions was further normalized to the
value obtained with a primer pair amplifying an intergenic region on chromosome 10
where no annotated genes could be found. For every gene fragment analyzed, each
sample was quantitated in duplicate and from >3 independent ChIPs. SEM was
determined for each fold difference above non-immune control and intergenic control
region.
RNA extraction and RT-PCR
Total RNA was extracted using TRIZOL (Invitrogen) according to the manufacturer´s
recommendation. Total RNA was briefly exposed to RNase-free DNase I (Ambion).
RNA was reverse transcribed to cDNA using oligo (dT)18 primer or gene-specific
primers and SuperScript III Kit (Invitrogen). Primers used are listed in Table 2. cDNA
from maxigene transfected cells was amplified by Real Time PCR and relative
expression levels were determined using the 2ΔΔCt method according to the Stratagene
MX3000 recommendations. Briefly, a ∆Ct value was calculated for each sample by
subtracting the Ct value for the calibrator gene (28S rRNA; to normalize for different
amounts of RNA) from the Ct value obtained for reverse transcribed cDNA for each
gene. A ∆∆Ct value was then calculated by subtracting the ∆Ct value of cDNA of genes
from untreated cells from the ∆Ct value of cDNA of genes from α-amanitin treated
cells. Fold differences were then determined by raising 2 to the power of - ∆∆Ct.
The mRNA abundance of BC033162, C19orf6 and THRAP5 was analyzed by
conventional PCR as well as determination of the fragment size of sheared chromatin
used for ChIP. Primers used in PCR reactions are listed in Table 2. PCR reactions were
as follows: 5µl PCR buffer; 10 mM dNTPs; 10 µM primers; 2 µl RNA, 2 µl cDNA or 1
µl pem HeLa input DNA, respectively; and 2 µl Taq polymerase. Cycling was
performed using a ThermoCycler (PTC-200, MJ Research). Cycling parameters were
94°C for 4 min followed by (94°C for 1 min, 60°C for 1 min, 72°C for 1 min) 27-30
cyclesd, depending on the experiments. Products were electrophoretically separated on
Part II: Materials and Methods
94
a 2% agarose gel, DNA bands stained with Gel Star (Cambrex BioScience), and bands
detected and analyzed with ArgusX1 (Biostep GmbH) and TotalLab (Nonlinear
Dynamics).
Primer Where? Sequence
Maxi1 U6 Maxigene specific: 5'-CGCTTCGGCAGCACATATAC-3'
Maxi Rev 5'-AACGCTTCACGCCTCGAGG-3'
HU6 1 endogenous U6-1 snRNA 5'-CGCTTCGGCAGCACATATAC-3'
HU6 2 5'-AAAAATATGGAACGCTTCACG-3'
5S 1 5S rRNA 5'-TCTCGTCTGATCTCGGAAGC-3'
5S 2 5'-AGCCTACAGCACCCGGTATT-3'
Tyr 5 Pre-tRNA Tyr 5’-CCTTCGATAGCTCAGCTGGTAGAG-3’
Tyr 6 5’-AAAAAACCGCACTTGTCTCCTTCG-3’
C19orf6 1 mRNA C19orf6 5'-TGTTGTGCTCTTCGTCCTG-3'
C19orf6 2 5'-GGTCAGCTCTTCCTCCTCCT-3'
RNA 1 LDHA ex3-ex5 5'-AGAACACCAAAGATTGTCTCTGGC-3'
RNA 2 5'-TTTCCCCCATCAGGTAACGG-3'
PGKa PGK1 ex1-ex5 5'-AAGGGAAGCGGGTCGTTATGAGAG-3'
PGKb 5'-TCTATTTTGGCTGGCTCGGC-3'
BC1 BC033162 ex3 5'-TGCCATCTCGTGAGTCAGTTGTCG-3'
BC2 5'-TTCCTTCTCCTGGACCCACTTC-3'
THRAP 3 THRAP5 5’-GTGTCACCATGCTCAAGTCG-3’
THRAP 4 5’-GTCCTGGGAGAGTGGTGTGT-3’
28S 1 28S rRNA 4506 4617 5'-CCGTCGTGAGACAGGTTAGTTTTAC-3'
28S 2 5'-CAGCCAAGCACATACACCAAATG-3'
HBA 1 β-actin 5'-TCAGCAAGCAGGAGTATGACGAG-3'
HBA 2 5'-CTTTTAGGATGGCAAGGGACTTC-3'
HBA 3 β-actin 5'-TCAGCAAGCAGGAGTATGACGAG-3'
HBA 4 5’-CTGGGACATCCGCAAAGACC-3´
HBA 5 β-actin 5'-TCAGCAAGCAGGAGTATGACGAG-3'
HBA 6 5´-TCCTGGGTGAGTGGAGACTGTC-3´
HBA 7 β-actin 5'-TCAGCAAGCAGGAGTATGACGAG-3'
HBA 8 5´-TGGCTGTCCCCAGTGGCTTC-3´
HBA 9 β-actin 5'-TCAGCAAGCAGGAGTATGACGAG-3'
HBA 10 5´-TCACTGGTTCTCTCTTCTGCCG-3´
GW 1 Gene desert chr. 10 5'-GGCTAATCCTCTATGGGAGTCTGTC-3'
GW 2 5'-CCAGGTGCTCAAGGTCAACATC-3'
PGK 1 PGK1 prom 5'-TCGTTGACCGAATCACCGAC-3'
PGK 2 5'-AGAGGTTTGCGACAGAGCACAG-3'
Leu 1 tRNA Leu 5'-GAGGACAACGGGGACAGTAA-3'
Leu 2 5'-TCCACCAGAAAAACTCCAGC-3'
U61 b U6-1 (-403 -186) 5'-TTATCTCTCTAACAGCCTTGTATCG-3'
U61 4 5'-CACTGCTCGGTAGTTTCGG-3'
U61 9 U6-1 (-687 -782) 5'-CGATGAGGGTGTCTGCTTTG-3'
Part II: Materials and Methods
95
Primer Where? Sequence
U61 10 5'-CCTGGCTCTTTCTAAATGTTGG-3'
U61 11 U6-1 (-1042-1137) 5'-CCTTCCTACTTCCATTCCTTCAAC-3'
U61 12 5'-ATGACCATAGCAACCCTGCC-3'
U61 15 U6-1 (163348) 5'-ATGTTTCGGAGCTGAAATGG-3'
U61 16 5'-ATAGAAGGAGGTGGGGCCTA-3'
U61 19 U6-1 (621788) 5'-TGCCTAAGTCATCGCAGGAAATC-3'
U61 20 5'-CACCAATCACTGAGAGGTAAAGCC-3'
U61 21 U6-1 (409582) 5'-ATTGGTGAAGCCCTTGACAT-3'
U61 22 5'-GGAAATTTCGTCTTGCCAAA-3'
U61 23 U6-1 (185364) 5'-CAGAGGCAAGATGGGAAAGA-3'
U61 24 5'-TGCACGGTTTACCACTGAAA-3'
U61 25 U6-1 (7 165) 5’-GCTTCGGCAGCACATATACTAA–3’
U61 26 5’- CGGCGTATAAACGTGGTGTA–3’
U61 27 U6-1 (150+1) 5`-GTACAAAATACGTGACGTAGAAAG-3`
U61 28 5`-GGTGTTTCGTCCTTTCCAC-3´
U62 1 U6-2 (–355-141) 5’-CTTCCGGTTCTGTCTTTTCG–3’
U62 2 5’-TCTTATTCTCCCGCCTTTGA-3’
U62 3 U6-2 (8180) 5’-CTCGGCCTTTTGGCTAAGAT–3’
U62 4 5’-CGTTCCTGGAGGTACTGCAA–3’
U63 1 U6-3 (-136+85) 5'-TCTTGGGAAAGCATTCAACC-3'
U63 2 5'-ATTTGCGTGTCATCCTTGC-3'
U64 1 U6-4 (-252-48) 5'-GGGTTACCAGATGGAATACAGG-3'
U64 2 5'-AGGAAACTGAGATAGGATTGACAG-3'
U64 3 U6-4 (-403-225) 5'-CTTTTGACAACTTGACCAACAAT-3'
U64 4 5’-TGGATGTCCTGTATTCCATCTG-3’
U66 1 U6-6 (–53+172 ) 5'-GGACTACCTTTCTGCGTATTCC-3'
U66 2 5'-TCCTTTCCCCATTGCTTG-3'
U67 1 U6-7 (–50+140) 5'-CAAATGAGAAACTCCCGTGCC-3'
U67 2 5'-GGTATCTTGCCCAACGACCTTC-3'
U67 3 U6-7 (-405-196) 5'-TCAGAGAGACTTCGGGGAGA-3'
U68 10 5'-TTACCTGTTGGGTGGGACAT- 3'
U68 7 U6-8 (-163-33) 5'-GCGGTCAGAGCACCAAACATAC-3'
U68 8 5'-GCACGGGAGTCTTTCACCTATTC-3'
U68 9 U6-8 (-409-197) 5'-CGCAGCAGCCTCGTTATC-3'
U68 10 5'-TTACCTGTTGGGTGGGACAT-3'
U69 1 U6-9 (-167+85) 5'-AACAACGAGCACTTTCTCAACTCC-3'
U69 2b 5'-ATTTGCGTGTCATCCTTGC-3'
U69 5 U6-9 (-361-203) 5’-GGTGCGATCTTCCCTAGC-3`
U69 6 5’-CTGGAAATGGTCGTATGCAA-3
Table 2. Primers used in this study.
References
96
References
Achsel, T., Ahrens, K., Brahms, H., Teigelkamp, S. and Luhrmann, R.(1998). The human U5-220kD protein (hPrp8) forms a stable RNA-free complex withseveral U5-specific proteins, including an RNA unwindase, a homologue of ribosomalelongation factor EF-2, and a novel WD-40 protein. Mol Cell Biol 18, 6756-66.
Achsel, T., Brahms, H., Kastner, B., Bachi, A., Wilm, M. and Luhrmann, R.(1999). A doughnut-shaped heteromer of human Sm-like proteins binds to the 3'-end ofU6 snRNA, thereby facilitating U4/U6 duplex formation in vitro. Embo J 18, 5789-802.
Akoulitchev, S., Makela, T. P., Weinberg, R. A. and Reinberg, D. (1995).Requirement for TFIIH kinase activity in transcription by RNA polymerase II. 377,557-560.
Alberts, B. (2002). Molecular biology of the cell. New York: Garland Science.Anderson, P. and Kedersha, N. (2006). RNA granules. J Cell Biol 172, 803-8.Audibert, A., Weil, D. and Dautry, F. (2002). In Vivo Kinetics of mRNA
Splicing and Transport in Mammalian Cells. Mol. Cell. Biol. 22, 6706-6718.Ausubel, F. M. (2001). Current protocols in molecular biology. New York: J.
Wiley.Azubel, M., Habib, N., Sperling, R. and Sperling, J. (2006). Native
Spliceosomes Assemble with Pre-mRNA to Form Supraspliceosomes. Journal ofMolecular Biology 356, 955-966.
Baillat, D., Hakimi, M.-A., Naar, A. M., Shilatifard, A., Cooch, N. andShiekhattar, R. (2005). Integrator, a Multiprotein Mediator of Small Nuclear RNAProcessing, Associates with the C-Terminal Repeat of RNA Polymerase II. Cell 123,265-276.
Barilla, D., Lee, B. A. and Proudfoot, N. J. (2001). Cleavage/polyadenylationfactor IA associates with the carboxyl- terminal domain of RNA polymerase II inSaccharomyces cerevisiae. Proc Natl Acad Sci U S A 98, 445-50.
Bauren, G. and Wieslander, L. (1994). Splicing of Balbiani ring 1 gene pre-mRNA occurs simultaneously with transcription. Cell 76, 183-92.
Bentley, D. L. (2005). Rules of engagement: co-transcriptional recruitment ofpre-mRNA processing factors. Curr Opin Cell Biol 17, 251-6.
Birnstiel, M. L. (1988). Structure and function of major and minor smallnuclear ribonucleoprotein particles. Berlin ; New York: Springer Verlag.
Black, D. L. (2003). Mechanisms of alternative pre-messenger RNA splicing.Annu Rev Biochem 72, 291-336.
Boehm, A. K., Saunders, A., Werner, J. and Lis, J. T. (2003). Transcriptionfactor and polymerase recruitment, modification, and movement on dhsp70 in vivo inthe minutes following heat shock. Mol Cell Biol 23, 7628-37.
Booth, B. L., Jr. and Pugh, B. F. (1997). Identification and characterization ofa nuclease specific for the 3' end of the U6 small nuclear RNA. J Biol Chem 272, 984-91.
Boyd, D. C., Greger, I. H. and Murphy, S. (2000). In vivo footprinting studiessuggest a role for chromatin in transcription of the human 7SK gene. Gene 247, 33-44.
References
97
Boyd, D. C., Turner, P. C., Watkins, N. J., Gerster, T. and Murphy, S.(1995). Functional redundancy of promoter elements ensures efficient transcription ofthe human 7SK gene in vivo. J Mol Biol 253, 677-90.
Bravo, R., Burckhardt, J., Curran, T. and Muller, R. (1985). Stimulation andinhibition of growth by EGF in different A431 cell clones is accompanied by the rapidinduction of c-fos and c-myc proto-oncogenes. Embo J 4, 1193-7.
Breathnach, R. and Chambon, P. (1981). Organization and expression ofeucaryotic split genes coding for proteins. Annu Rev Biochem 50, 349-83.
Brodsky, A., Meyer, C., Swinburne, I., Hall, G., Keenan, B., Liu, X., Fox, E.and Silver, P. (2005). Genomic mapping of RNA polymerase II reveals sites of co-transcriptional regulation in human cells. Genome Biology 6, R64.
Brosi, R., Hauri, H. and Kramer, A. (1993). Separation of splicing factor SF3into two components and purification of SF3a activity. J. Biol. Chem. 268, 17640-17646.
Brow, D. A. and Guthrie, C. (1988). Spliceosomal RNA U6 is remarkablyconserved from yeast to mammals. 334, 213-218.
Buratowski, S. (2003). The CTD code. Nat Struct Biol 10, 679-80.Burd, C. G. and Dreyfuss, G. (1994). RNA binding specificity of hnRNP A1:
significance of hnRNP A1 high-affinity binding sites in pre-mRNA splicing. Embo J13, 1197-204.
Cadena, D. and Dahmus, M. (1987). Messenger RNA synthesis in mammaliancells is catalyzed by the phosphorylated form of RNA polymerase II. J. Biol. Chem.262, 12468-12474.
Calero, G., Wilson, K. F., Ly, T., Rios-Steiner, J. L., Clardy, J. C. andCerione, R. A. (2002). Structural basis of m7GpppG binding to the nuclear cap-bindingprotein complex. 9, 912-917.
Chakrabarti, S. K., James, J. C. and Mirmira, R. G. (2002). Quantitativeassessment of gene targeting in vitro and in vivo by the pancreatic transcription factor,Pdx1. Importance of chromatin structure in directing promoter binding. J Biol Chem277, 13286-93.
Chan, S. P., Kao, D. I., Tsai, W. Y. and Cheng, S. C. (2003). The Prp19p-associated complex in spliceosome activation. Science 302, 279-82.
Chapman, R. D., Conrad, M. and Eick, D. (2005). Role of the mammalianRNA polymerase II C-terminal domain (CTD) nonconsensus repeats in CTD stabilityand cell proliferation. Mol Cell Biol 25, 7665-74.
Chapman, R. D., Palancade, B., Lang, A., Bensaude, O. and Eick, D. (2004).The last CTD repeat of the mammalian RNA polymerase II large subunit is importantfor its stability. Nucleic Acids Res 32, 35-44.
Chen, Y., Sinha, K., Perumal, K. and Reddy, R. (2000). Effect of 3' terminaladenylic acid residue on the uridylation of human small RNAs in vitro and in frogoocytes. RNA 6, 1277-1288.
Cheng, C. and Sharp, P. A. (2003). RNA polymerase II accumulation in thepromoter-proximal region of the dihydrofolate reductase and gamma-actin genes. MolCell Biol 23, 1961-7.
Cho, E. J., Takagi, T., Moore, C. R. and Buratowski, S. (1997). mRNAcapping enzyme is recruited to the transcription complex by phosphorylation of theRNA polymerase II carboxy-terminal domain. Genes Dev 11, 3319-26.
References
98
Chou, M.-Y., Underwood, J. G., Nikolic, J., Luu, M. H. T. and Black, D. L.(2000). Multisite RNA Binding and Release of Polypyrimidine Tract Binding Proteinduring the Regulation of c-src Neural-Specific Splicing. Molecular Cell 5, 949-957.
Christova, R. and Oelgeschlager, T. (2002). Association of human TFIID-promoter complexes with silenced mitotic chromatin in vivo. 4, 79-82.
Clement, J. Q., Maiti, S. and Wilkinson, M. F. (2001). Localization andStability of Introns Spliced from the Pem Homeobox Gene. J. Biol. Chem. 276, 16919-16930.
Clement, J. Q., Qian, L., Kaplinsky, N. and Wilkinson, M. F. (1999). Thestability and fate of a spliced intron from vertebrate cells. RNA 5, 206-220.
Collins, I., Weber, A. and Levens, D. (2001). Transcriptional consequences oftopoisomerase inhibition. Mol Cell Biol 21, 8437-51.
Colot, H. V., Stutz, F. and Rosbash, M. (1996). The yeast splicing factorMud13p is a commitment complex component and corresponds to CBP20, the smallsubunit of the nuclear cap-binding complex. Genes Dev 10, 1699-708.
ConsortiumInternational, H. G. S. (2004). Finishing the euchromaticsequence of the human genome. 431, 931-945.
Cougot, N., van Dijk, E., Babajko, S. and Seraphin, B. (2004). 'Cap-tabolism'. Trends Biochem Sci 29, 436-44.
Cramer, P., Bushnell, D. A. and Kornberg, R. D. (2001). Structural basis oftranscription: RNA polymerase II at 2.8 angstrom resolution. Science 292, 1863-76.
Cramer, P., Pesce, C. G., Baralle, F. E. and Kornblihtt, A. R. (1997).Functional association between promoter structure and transcript alternative splicing.Proc Natl Acad Sci U S A 94, 11456-60.
Dahmus, M. E. (1995). Phosphorylation of the C-terminal domain of RNApolymerase II. Biochim Biophys Acta 1261, 171-82.
Das, R., Zhou, Z. and Reed, R. (2000). Functional association of U2 snRNPwith the ATP-independent spliceosomal complex E. Mol Cell 5, 779-87.
de la Mata, M., Alonso, C. R., Kadener, S., Fededa, J. P., Blaustein, M.,Pelisch, F., Cramer, P., Bentley, D. and Kornblihtt, A. R. (2003). A slow RNApolymerase II affects alternative splicing in vivo. Mol Cell 12, 525-32.
Domitrovich, A. M. and Kunkel, G. R. (2003). Multiple, dispersed human U6small nuclear RNA genes with varied transcriptional efficiencies10.1093/nar/gkg331. Nucl. Acids Res. 31, 2344-2352.
Doyle, O., Corden, J. L., Murphy, C. and Gall, J. G. (2002). The distributionof RNA polymerase II largest subunit (RPB1) in the Xenopus germinal vesicle. J StructBiol 140, 154-66.
Dreyfuss, G., Kim, V. N. and Kataoka, N. (2002). Messenger-RNA-bindingproteins and the messages they carry. Nat Rev Mol Cell Biol 3, 195-205.
Edery, I. and Sonenberg, N. (1985). Cap-Dependent RNA Splicing in a HeLaNuclear Extract. PNAS 82, 7590-7594.
Emili, A., Shales, M., McCracken, S., Xie, W., Tucker, P. W., Kobayashi,R., Blencowe, B. J. and Ingles, C. J. (2002). Splicing and transcription-associatedproteins PSF and p54nrb/nonO bind to the RNA polymerase II CTD. RNA 8, 1102-1111.
Epstein, P., Reddy, R., Henning, D. and Busch, H. (1980). The nucleotidesequence of nuclear U6 (4.7 S) RNA. J Biol Chem 255, 8901-6.
References
99
Fabrizio, P., Laggerbauer, B., Lauber, J., Lane, W. S. and Luhrmann, R.(1997). An evolutionarily conserved U5 snRNP-specific protein is a GTP-binding factorclosely related to the ribosomal translocase EF-2. Embo J 16, 4092-106.
Fairley, J. A., Kantidakis, T., Kenneth, N. S., Intine, R. V., Maraia, R. J.and White, R. J. (2005). Human La is found at RNA polymerase III-transcribed genesin vivo10.1073/pnas.0506415102. PNAS 102, 18350-18355.
Fairley, J. A., Scott, P. H. and White, R. J. (2003). TFIIIB is phosphorylated,disrupted and selectively released from tRNA promoters during mitosis in vivo. Embo J22, 5841-50.
Faustino, N. A. and Cooper, T. A. (2003). Pre-mRNA splicing and humandisease. Genes Dev. 17, 419-437.
Feaver, W. J., Svejstrup, J. Q., Henry, N. L. and Kornberg, R. D. (1994).Relationship of CDK-activating kinase and RNA polymerase II CTD kinaseTFIIH/TFIIK. Cell 79, 1103-1109.
Fivaz, J., Bassi, M. C., Pinaud, S. and Mirkovitch, J. (2000). RNApolymerase II promoter-proximal pausing upregulates c-fos gene expression. Gene 255,185-94.
Fong, N. and Bentley, D. L. (2001). Capping, splicing, and 3' processing areindependently stimulated by RNA polymerase II: different functions for differentsegments of the CTD. Genes Dev 15, 1783-95.
Fury, M. G. and Zieve, G. W. (1996). U6 snRNA maturation and stability. ExpCell Res 228, 160-3.
Gama-Carvalho, M., Krauss, R. D., Chiang, L., Valcarcel, J., Green, M. R.and Carmo-Fonseca, M. (1997). Targeting of U2AF65 to sites of active splicing in thenucleus. J Cell Biol 137, 975-87.
Genschik, P., Billy, E., Swianiewicz, M. and Filipowicz, W. (1997). Thehuman RNA 3'-terminal phosphate cyclase is a member of a new family of proteinsconserved in Eucarya, Bacteria and Archaea. Embo J 16, 2955-67.
Gerbi, S. A., Borovjagin, A. V. and Lange, T. S. (2003). The nucleolus: a siteof ribonucleoprotein maturation. Current Opinion in Cell Biology 15, 318-325.
Goldstrohm, A. C., Greenleaf, A. L. and Garcia-Blanco, M. A. (2001). Co-transcriptional splicing of pre-messenger RNAs: considerations for the mechanism ofalternative splicing. Gene 277, 31-47.
Gornemann, J., Kotovic, K. M., Hujer, K. and Neugebauer, K. M. (2005).Cotranscriptional spliceosome assembly occurs in a stepwise fashion and requires thecap binding complex. Mol Cell 19, 53-63.
Greenleaf, A. L. (1993). Positive patches and negative noodles: linking RNAprocessing to transcription? Trends Biochem Sci 18, 117-9.
Hanamura, A., Caceres, J. F., Mayeda, A., Franza, B. R., Jr. and Krainer,A. R. (1998). Regulated tissue-specific expression of antagonistic pre-mRNA splicingfactors. Rna 4, 430-44.
Hengartner, C. J., Myer, V. E., Liao, S. M., Wilson, C. J., Koh, S. S. andYoung, R. A. (1998). Temporal regulation of RNA polymerase II by Srb10 and Kin28cyclin-dependent kinases. Mol Cell 2, 43-53.
Hernandez, N. (1993). TBP, a universal eukaryotic transcription factor? GenesDev 7, 1291-308.
Hernandez, N. (2001). Small nuclear RNA genes: a model system to studyfundamental mechanisms of transcription. J Biol Chem 276, 26733-6.
References
100
Hirose, Y., Tacke, R. and Manley, J. L. (1999). Phosphorylated RNApolymerase II stimulates pre-mRNA splicing. Genes Dev 13, 1234-9.
Hirsch, H. A., Jawdekar, G. W., Lee, K. A., Gu, L. and Henry, R. W.(2004). Distinct mechanisms for repression of RNA polymerase III transcription by theretinoblastoma tumor suppressor protein. Mol Cell Biol 24, 5989-99.
Howe, K. J., Kane, C. M. and Ares, M., Jr. (2003). Perturbation oftranscription elongation influences the fidelity of internal exon inclusion inSaccharomyces cerevisiae. Rna 9, 993-1006.
Hsiang, Y. H., Hertzberg, R., Hecht, S. and Liu, L. F. (1985). Camptothecininduces protein-linked DNA breaks via mammalian DNA topoisomerase I. J Biol Chem260, 14873-8.
Hubbard, T., Barker, D., Birney, E., Cameron, G., Chen, Y., Clark, L., Cox,T., Cuff, J., Curwen, V., Down, T. et al. (2002). The Ensembl genome databaseproject10.1093/nar/30.1.38. Nucl. Acids Res. 30, 38-41.
Hutchison, S., LeBel, C., Blanchette, M. and Chabot, B. (2002). Distinct Setsof Adjacent Heterogeneous Nuclear Ribonucleoprotein (hnRNP) A1/A2 Binding SitesControl 5' Splice Site Selection in the hnRNP A1 mRNA Precursor10.1074/jbc.M203633200. J. Biol. Chem. 277, 29745-29752.
Izaurralde, E., Lewis, J., McGuigan, C., Jankowska, M., Darzynkiewicz, E.and Mattaj, I. W. (1994). A nuclear cap binding protein complex involved in pre-mRNA splicing. Cell 78, 657-68.
Izaurralde, E., Stepinski, J., Darzynkiewicz, E. and Mattaj, I. W. (1992). Acap binding protein that may mediate nuclear export of RNA polymerase II-transcribedRNAs. J Cell Biol 118, 1287-95.
Jenny, A., Hauri, H. P. and Keller, W. (1994). Characterization of cleavageand polyadenylation specificity factor and cloning of its 100-kilodalton subunit. Mol.Cell. Biol. 14, 8183-8190.
Jurica, M. S. and Moore, M. J. (2003). Pre-mRNA Splicing: Awash in a Seaof Proteins. Molecular Cell 12, 5-14.
Kadener, S., Cramer, P., Nogues, G., Cazalla, D., de La Mata, M., Fededa,J. P., Werbajh, S. E., Srebrow, A. and Kornblihtt, A. R. (2001). Antagonistic effectsof T-Ag and VP16 reveal a role for RNA pol II elongation on alternative splicing. EmboJ 20, 5759-5768.
Karaduman, R., Fabrizio, P., Hartmuth, K., Urlaub, H. and Luhrmann, R.(2006). RNA Structure and RNA-Protein Interactions in Purified Yeast U6 snRNPs.Journal of Molecular Biology 356, 1248-1262.
Kedersha, N. and Anderson, P. (2002). Stress granules: sites of mRNA triagethat regulate mRNA stability and translatability. Biochem Soc Trans 30, 963-9.
Kedersha, N., Stoecklin, G., Ayodele, M., Yacono, P., Lykke-Andersen, J.,Fitzler, M. J., Scheuner, D., Kaufman, R. J., Golan, D. E. and Anderson, P. (2005).Stress granules and processing bodies are dynamically linked sites of mRNPremodeling. J Cell Biol 169, 871-84.
Kent, W. J., Sugnet, C. W., Furey, T. S., Roskin, K. M., Pringle, T. H.,Zahler, A. M. and Haussler, a. D. (2002). The Human Genome Browser at UCSC10.1101/gr.229102. Article published online before print in May 2002. Genome Res. 12,996-1006.
References
101
Kim, E., Du, L., Bregman, D. B. and Warren, S. L. (1997). Splicing factorsassociate with hyperphosphorylated RNA polymerase II in the absence of pre-mRNA. JCell Biol 136, 19-28.
Kim, Y. K., Bourgeois, C. F., Isel, C., Churcher, M. J. and Karn, J. (2002).Phosphorylation of the RNA polymerase II carboxyl-terminal domain by CDK9 isdirectly responsible for human immunodeficiency virus type 1 Tat-activatedtranscriptional elongation. Mol Cell Biol 22, 4622-37.
Kiseleva, E., Wurtz, T., Visa, N. and Daneholt, B. (1994). Assembly anddisassembly of spliceosomes along a specific pre-messenger RNP fiber. Embo J 13,6052-61.
Kleinschmidt, A. M., Pederson, T., Tani, T. and Ohshima, Y. (1990). Anintron-containing Schizosaccharomyces pombe U6 RNA gene can be transcribed byhuman RNA polymerase III. J Mol Biol 211, 7-9.
Komarnitsky, P., Cho, E. J. and Buratowski, S. (2000). Differentphosphorylated forms of RNA polymerase II and associated mRNA processing factorsduring transcription. Genes Dev 14, 2452-60.
Konarska, M. M., Padgett, R. A. and Sharp, P. A. (1984). Recognition of capstructure in splicing in vitro of mRNA precursors. Cell 38, 731-736.
Kotovic, K. M., Lockshon, D., Boric, L. and Neugebauer, K. M. (2003).Cotranscriptional recruitment of the U1 snRNP to intron-containing genes in yeast. MolCell Biol 23, 5768-79.
Krainer, A. R. (1988). Pre-mRNA splicing by complementation with purifiedhuman U1, U2, U4/U6 and U5 snRNPs. Nucleic Acids Res 16, 9415-29.
Krainer, A. R., Maniatis, T., Ruskin, B. and Green, M. R. (1984). Normaland mutant human [beta]-globin pre-mRNAs are faithfully and efficiently spliced invitro. Cell 36, 993-1005.
Kramer, A., Haars, R., Kabisch, R., Will, H., Bautz, F. A. and Bautz, E. K.(1980). Monoclonal antibody directed against RNA polymerase II of Drosophilamelanogaster. Mol Gen Genet 180, 193-9.
Krawczak, M., Reiss, J. and Cooper, D. N. (1992). The mutational spectrumof single base-pair substitutions in mRNA splice junctions of human genes: causes andconsequences. Hum Genet 90, 41-54.
Kroeger, P. E. and Rowe, T. C. (1992). Analysis of topoisomerase I and IIcleavage sites on the Drosophila actin and Hsp70 heat shock genes. Biochemistry 31,2492-501.
Krol, A., Carbon, P., Ebel, J. P. and Appel, B. (1987). Xenopus tropicalis U6snRNA genes transcribed by Pol III contain the upstream promoter elements used byPol II dependent U snRNA genes. Nucleic Acids Res 15, 2463-78.
Kuhlman, T. C., Cho, H., Reinberg, D. and Hernandez, N. (1999). TheGeneral Transcription Factors IIA, IIB, IIF, and IIE Are Required for RNA PolymeraseII Transcription from the Human U1 Small Nuclear RNA Promoter. Mol. Cell. Biol. 19,2130-2141.
Kunkel, G. R., Maser, R. L., Calvet, J. P. and Pederson, T. (1986). U6 smallnuclear RNA is transcribed by RNA polymerase III. Proc Natl Acad Sci U S A 83,8575-9.
Kuo, M. H. and Allis, C. D. (1999). In vivo cross-linking andimmunoprecipitation for studying dynamic Protein:DNA associations in a chromatinenvironment. Methods 19, 425-33.
References
102
Kwan, S., Gerlach, V. L. and Brow, D. A. (2000). Disruption of the 5' stem-loop of yeast U6 RNA induces trimethylguanosine capping of this RNA polymerase IIItranscript in vivo. Rna 6, 1859-69.
Lacadie, S. A. and Rosbash, M. (2005). Cotranscriptional spliceosomeassembly dynamics and the role of U1 snRNA:5'ss base pairing in yeast. Mol Cell 19,65-75.
Lander, E. S. Linton, L. M. Birren, B. Nusbaum, C. Zody, M. C. Baldwin,J. Devon, K. Dewar, K. Doyle, M. FitzHugh, W. et al. (2001). Initial sequencing andanalysis of the human genome. Nature 409, 860-921.
Lee, Y., Kim, M., Han, J., Yeom, K. H., Lee, S., Baek, S. H. and Kim, V. N.(2004). MicroRNA genes are transcribed by RNA polymerase II. Embo J 23, 4051-60.
Lerner, E. A., Lerner, M. R., Janeway, C. A., Jr. and Steitz, J. A. (1981).Monoclonal antibodies to nucleic acid-containing cellular constituents: probes formolecular biology and autoimmune disease. Proc Natl Acad Sci U S A 78, 2737-41.
Lewis, J. D., Izaurralde, E., Jarmolowski, A., McGuigan, C. and Mattaj, I.W. (1996). A nuclear cap-binding complex facilitates association of U1 snRNP with thecap-proximal 5' splice site. Genes Dev 10, 1683-98.
Liang, S. and Lutz, C. S. (2006). p54nrb is a component of the snRNP-freeU1A (SF-A) complex that promotes pre-mRNA cleavage during polyadenylation. RNA12, 111-121.
Lis, J. (1998). Promoter-associated pausing in promoter architecture andpostinitiation transcriptional regulation. Cold Spring Harb Symp Quant Biol 63, 347-56.
Litingtung, Y., Lawler, A. M., Sebald, S. M., Lee, E., Gearhart, J. D.,Westphal, H. and Corden, J. L. (1999). Growth retardation and neonatal lethality inmice with a homozygous deletion in the C-terminal domain of RNA polymerase II. MolGen Genet 261, 100-5.
Ljungman, M. and Hanawalt, P. C. (1996). The anti-cancer drugcamptothecin inhibits elongation but stimulates initiation of RNA polymerase IItranscription. Carcinogenesis 17, 31-5.
Lobo, S. M. and Hernandez, N. (1989). A 7 bp mutation converts a humanRNA polymerase II snRNA promoter into an RNA polymerase III promoter. Cell 58,55-67.
Lobo, S. M., Lister, J., Sullivan, M. L. and Hernandez, N. (1991). The clonedRNA polymerase II transcription factor IID selects RNA polymerase III to transcribethe human U6 gene in vitro. Genes Dev 5, 1477-89.
Lund, E. and Dahlberg, J. E. (1992). Cyclic 2',3'-phosphates and nontemplatednucleotides at the 3' end of spliceosomal U6 small nuclear RNA's. Science 255, 327-30.
Makarova, O. V., Makarov, E. M., Liu, S., Vornlocher, H. P. andLuhrmann, R. (2002). Protein 61K, encoded by a gene (PRPF31) linked to autosomaldominant retinitis pigmentosa, is required for U4/U6*U5 tri-snRNP formation and pre-mRNA splicing. Embo J 21, 1148-57.
Malca, H., Shomron, N. and Ast, G. (2003). The U1 snRNP base pairs withthe 5' splice site within a penta-snRNP complex. Mol Cell Biol 23, 3442-55.
Maniatis, T. and Reed, R. (2002). An extensive network of coupling amonggene expression machines. Nature 416, 499-506.
Maroney, P. A., Romfo, C. M. and Nilsen, T. W. (2000). Functionalrecognition of 5' splice site by U4/U6.U5 tri-snRNP defines a novel ATP-dependentstep in early spliceosome assembly. Mol Cell 6, 317-28.
References
103
Mattaj, I. W., Dathan, N. A., Parry, H. D., Carbon, P. and Krol, A. (1988).Changing the RNA polymerase specificity of U snRNA gene promoters. Cell 55, 435-42.
Mazza, C., Ohno, M., Segref, A., Mattaj, I. W. and Cusack, S. (2001).Crystal Structure of the Human Nuclear Cap Binding Complex. Molecular Cell 8, 383-396.
McCracken, S., Fong, N., Rosonina, E., Yankulov, K., Brothers, G.,Siderovski, D., Hessel, A., Foster, S., Shuman, S. and Bentley, D. L. (1997a). 5'-Capping enzymes are targeted to pre-mRNA by binding to the phosphorylated carboxy-terminal domain of RNA polymerase 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. (1997b). The C-terminal domainof RNA polymerase II couples mRNA processing to transcription. Nature 385, 357-61.
McLaren, R., Caruccio, N. and Ross, J. (1997). Human La protein: a stabilizerof histone mRNA. Mol. Cell. Biol. 17, 3028-3036.
Meininghaus, M., Chapman, R. D., Horndasch, M. and Eick, D. (2000).Conditional expression of RNA polymerase II in mammalian cells. Deletion of thecarboxyl-terminal domain of the large subunit affects early steps in transcription. J BiolChem 275, 24375-82.
Miriami, E., Angenitzki, M., Sperling, R. and Sperling, J. (1995).Magnesium cations are required for the association of U small nuclearribonucleoproteins and SR proteins with pre-mRNA in 200 S large nuclearribonucleoprotein particles. J Mol Biol 246, 254-63.
Miyagishi, M. and Taira, K. (2002). U6 promoter-driven siRNAs with foururidine 3' overhangs efficiently suppress targeted gene expression in mammalian cells.Nat Biotechnol 20, 497-500.
Moenne, A., Camier, S., Anderson, G., Margottin, F., Beggs, J. andSentenac, A. (1990). The U6 gene of Saccharomyces cerevisiae is transcribed by RNApolymerase C (III) in vivo and in vitro. Embo J 9, 271-7.
Moore, M. J. and Sharp, P. A. (1993). Evidence for two active sites in thespliceosome provided by stereochemistry of pre-mRNA splicing. 365, 364-368.
Morris, D. P. and Greenleaf, A. L. (2000). The splicing factor, Prp40, bindsthe phosphorylated carboxyl-terminal domain of RNA polymerase II. J Biol Chem 275,39935-43.
Neugebauer, K. M. (2002). On the importance of being co-transcriptional. JCell Sci 115, 3865-71.
Neugebauer, K. M. and Roth, M. B. (1997). Distribution of pre-mRNAsplicing factors at sites of RNA polymerase II transcription. Genes Dev 11, 1148-59.
Neugebauer, K. M., Stolk, J. A. and Roth, M. B. (1995). A conserved epitopeon a subset of SR proteins defines a larger family of Pre-mRNA splicing factors. J CellBiol 129, 899-908.
O'Brien, T., Hardin, S., Greenleaf, A. and Lis, J. T. (1994). Phosphorylationof RNA polymerase II C-terminal domain and transcriptional elongation. Nature 370,75-7.
O'Mullane, L. and Eperon, I. C. (1998). The Pre-mRNA 5' Cap DeterminesWhether U6 Small Nuclear RNA Succeeds U1 Small Nuclear RibonucleoproteinParticle at 5' Splice Sites. Mol. Cell. Biol. 18, 7510-7520.
Orlando, V. (2000). Mapping chromosomal proteins in vivo by formaldehyde-crosslinked- chromatin immunoprecipitation. Trends Biochem Sci 25, 99-104.
References
104
Orphanides, G., Lagrange, T. and Reinberg, D. (1996). The generaltranscription factors of RNA polymerase II. Genes Dev 10, 2657-83.
Osheim, Y. N., Miller, O. L., Jr. and Beyer, A. L. (1985). RNP particles atsplice junction sequences on Drosophila chorion transcripts. Cell 43, 143-51.
Padgett, R. A., Grabowski, P. J., Konarska, M. M., Seiler, S. and Sharp, P.A. (1986). Splicing of Messenger RNA Precursors. Annual Review of Biochemistry 55,1119-1150.
Palancade, B. and Bensaude, O. (2003). Investigating RNA polymerase IIcarboxyl-terminal domain (CTD) phosphorylation. Eur J Biochem 270, 3859-70.
Park, J. S. and Kunkel, G. R. (1995). Both RNA polymerase III and RNApolymerase II accurately initiate transcription from a human U6 promoter in vitro.Biochem Biophys Res Commun 214, 934-40.
Patton, J. G., Porro, E. B., Galceran, J., Tempst, P. and Nadal-Ginard, B.(1993). Cloning and characterization of PSF, a novel pre-mRNA splicing factor. GenesDev 7, 393-406.
Paul, C. P., Good, P. D., Winer, I. and Engelke, D. R. (2002). Effectiveexpression of small interfering RNA in human cells. 20, 505-508.
Payne, J. M., Laybourn, P. J. and Dahmus, M. E. (1989). The transition ofRNA polymerase II from initiation to elongation is associated with phosphorylation ofthe carboxyl-terminal domain of subunit IIa. J Biol Chem 264, 19621-9.
Phatnani, H. P., Jones, J. C. and Greenleaf, A. L. (2004). Expanding thefunctional repertoire of CTD kinase I and RNA polymerase II: novel phosphoCTD-associating proteins in the yeast proteome. Biochemistry 43, 15702-19.
Potashkin, J. and Frendewey, D. (1989). Splicing of the U6 RNA precursor isimpaired in fission yeast pre-mRNA splicing mutants. Nucleic Acids Res 17, 7821-31.
Proudfoot, N. J., Furger, A. and Dye, M. J. (2002). Integrating mRNAprocessing with transcription. Cell 108, 501-12.
Raitskin, O., Angenitzki, M., Sperling, J. and Sperling, R. (2002). Largenuclear RNP particles--the nuclear pre-mRNA processing machine. J Struct Biol 140,123-30.
Ramanathan, Y., Rajpara, S. M., Reza, S. M., Lees, E., Shuman, S.,Mathews, M. B. and Pe'ery, T. (2001). Three RNA polymerase II carboxyl-terminaldomain kinases display distinct substrate preferences. J Biol Chem 276, 10913-20.
Rasmussen, E. B. and Lis, J. T. (1993). In vivo transcriptional pausing and capformation on three Drosophila heat shock genes. Proc Natl Acad Sci U S A 90, 7923-7.
Reddy, R. (1988). Transcription of a U6 small nuclear RNA gene in vitro.Transcription of a mouse U6 small nuclear RNA gene in vitro by RNA polymerase IIIis dependent on transcription factor(s) different from transcription factors IIIA, IIIB,and IIIC. J Biol Chem 263, 15980-4.
Reddy, R., Henning, D., Das, G., Harless, M. and Wright, D. (1987). Thecapped U6 small nuclear RNA is transcribed by RNA polymerase III. J Biol Chem 262,75-81.
Reed, R. (2000). Mechanisms of fidelity in pre-mRNA splicing. Curr Opin CellBiol 12, 340-5.
Reeves, R., Gorman, C. M. and Howard, B. (1985). Minichromosomeassembly of non-integrated plasmid DNA transfected into mammalian cells. NucleicAcids Res 13, 3599-615.
References
105
Rinke, J. and Steitz, J. A. (1982). Precursor molecules of both human 5Sribosomal RNA and transfer RNAs are bound by a cellular protein reactive with anti-LaLupus antibodies. Cell 29, 149-159.
Rinke, J. and Steitz, J. A. (1985). Association of the lupus antigen La with asubset of U6 snRNA molecules. Nucleic Acids Res 13, 2617-29.
Robert, F., Blanchette, M., Maes, O., Chabot, B. and Coulombe, B. (2002).A Human RNA Polymerase II-containing Complex Associated with Factors Necessaryfor Spliceosome Assembly. J. Biol. Chem. 277, 9302-9306.
Roberts, G. C., Gooding, C., Mak, H. Y., Proudfoot, N. J. and Smith, C. W.(1998). Co-transcriptional commitment to alternative splice site selection. Nucleic AcidsRes 26, 5568-72.
Rodriguez, C. R., Cho, E. J., Keogh, M. C., Moore, C. L., Greenleaf, A. L.and Buratowski, S. (2000). Kin28, the TFIIH-associated carboxy-terminal domainkinase, facilitates the recruitment of mRNA processing machinery to RNA polymeraseII. Mol Cell Biol 20, 104-12.
Rosonina, E., Ip, J. Y. Y., Calarco, J. A., Bakowski, M. A., Emili, A.,McCracken, S., Tucker, P., Ingles, C. J. and Blencowe, B. J. (2005). Role for PSF inMediating Transcriptional Activator-Dependent Stimulation of Pre-mRNA ProcessingIn Vivo. Mol. Cell. Biol. 25, 6734-6746.
Rumi, M., Ishihara, S., Aziz, M., Kazumori, H., Ishimura, N., Yuki, T.,Kadota, C., Kadowaki, Y. and Kinoshita, Y. (2006). RNA polymerase II mediatedtranscription from the polymerase III promoters in short hairpin RNA expression vector.Biochemical and Biophysical Research Communications 339, 540-547.
Schaub, M., Myslinski, E., Schuster, C., Krol, A. and Carbon, P. (1997).Staf, a promiscuous activator for enhanced transcription by RNA polymerases II and III.Embo J 16, 173-81.
Schramm, L. and Hernandez, N. (2002). Recruitment of RNA polymerase IIIto its target promoters. Genes Dev 16, 2593-620.
Singh, R., Gupta, S. and Reddy, R. (1990). Capping of mammalian U6 smallnuclear RNA in vitro is directed by a conserved stem-loop and AUAUAC sequence:conversion of a noncapped RNA into a capped RNA. Mol Cell Biol 10, 939-46.
Singh, R. and Reddy, R. (1989). Gamma-monomethyl phosphate: a capstructure in spliceosomal U6 small nuclear RNA. Proc Natl Acad Sci U S A 86, 8280-3.
Sinha, K. M., Gu, J., Chen, Y. and Reddy, R. (1998). Adenylation of smallRNAs in human cells. Development of a cell-free system for accurate adenylation onthe 3'-end of human signal recognition particle RNA. J Biol Chem 273, 6853-9.
Soller, M. (2006). Pre-messenger RNA processing and its regulation: a genomicperspective. Cell Mol Life Sci.
Sollner-Webb, B. (1988). Surprises in polymerase III transcription. Cell 52,153-4.
Spingola, M., Grate, L., Haussler, D. and Ares, M., Jr. (1999). Genome-widebioinformatic and molecular analysis of introns in Saccharomyces cerevisiae. Rna 5,221-34.
Staley, J. P. and Guthrie, C. (1998). Mechanical devices of the spliceosome:motors, clocks, springs, and things. Cell 92, 315-26.
Stanek, D. and Neugebauer, K. M. (2004). Detection of snRNP assemblyintermediates in Cajal bodies by fluorescence resonance energy transfer. J Cell Biol166, 1015-25.
References
106
Stevens, S. W., Ryan, D. E., Ge, H. Y., Moore, R. E., Young, M. K., Lee, T.D. and Abelson, J. (2002). Composition and functional characterization of the yeastspliceosomal penta-snRNP. Mol Cell 9, 31-44.
Stewart, A. F., Herrera, R. E. and Nordheim, A. (1990). Rapid induction ofc-fos transcription reveals quantitative linkage of RNA polymerase II and DNAtopoisomerase I enzyme activities. Cell 60, 141-9.
Sui, G., Soohoo, C., Affar, E. B., Gay, F., Shi, Y., Forrester, W. C. and Shi,Y. (2002). A DNA vector-based RNAi technology to suppress gene expression inmammalian cells10.1073/pnas.082117599. PNAS 99, 5515-5520.
Tanackovic, G. and Kramer, A. (2005). Human splicing factor SF3a, but notSF1, is essential for pre-mRNA splicing in vivo. Mol Biol Cell 16, 1366-77.
Tani, T. and Ohshima, Y. (1989). The gene for the U6 small nuclear RNA infission yeast has an intron. Nature 337, 87-90.
Tarn, W. Y., Hsu, C. H., Huang, K. T., Chen, H. R., Kao, H. Y., Lee, K. R.and Cheng, S. C. (1994). Functional association of essential splicing factor(s) withPRP19 in a protein complex. Embo J 13, 2421-31.
Tazi, J., Forne, T., Jeanteur, P., Cathala, G. and Brunel, C. (1993).Mammalian U6 small nuclear RNA undergoes 3' end modifications within thespliceosome. Mol Cell Biol 13, 1641-50.
Tennyson, C. N., Klamut, H. J. and Worton, R. G. (1995). The humandystrophin gene requires 16 hours to be transcribed and is cotranscriptionally spliced.Nat Genet 9, 184-90.
Terns, M. P., Lund, E. and Dahlberg, J. E. (1992). 3'-end-dependentformation of U6 small nuclear ribonucleoprotein particles in Xenopus laevis oocytenuclei. Mol Cell Biol 12, 3032-40.
Thompson, N. E., Steinberg, T. H., Aronson, D. B. and Burgess, R. R.(1989). Inhibition of in vivo and in vitro transcription by monoclonal antibodiesprepared against wheat germ RNA polymerase II that react with the heptapeptide repeatof eukaryotic RNA polymerase II. J Biol Chem 264, 11511-20.
Trippe, R., Richly, H. and Benecke, B. J. (2003). Biochemicalcharacterization of a U6 small nuclear RNA-specific terminal uridylyltransferase. Eur JBiochem 270, 971-80.
Trippe, R., Sandrock, B. and Benecke, B. J. (1998). A highly specificterminal uridylyl transferase modifies the 3'-end of U6 small nuclear RNA. NucleicAcids Res 26, 3119-26.
Ujvari, A. and Luse, D. S. (2004). Newly Initiated RNA Encounters a FactorInvolved in Splicing Immediately upon Emerging from within RNA Polymerase II. J.Biol. Chem. 279, 49773-49779.
Ujvari, A. and Luse, D. S. (2006). RNA emerging from the active site of RNApolymerase II interacts with the Rpb7 subunit. 13, 49-54.
Valay, J. G., Simon, M., Dubois, M. F., Bensaude, O., Facca, C. and Faye,G. (1995). The KIN28 gene is required both for RNA polymerase II mediatedtranscription and phosphorylation of the Rpb1p CTD. J Mol Biol 249, 535-44.
Valcarcel, J. and Green, M. R. (1996). The SR protein family: pleiotropicfunctions in pre-mRNA splicing. Trends Biochem Sci 21, 296-301.
van der Houven van Oordt, W., Diaz-Meco, M. T., Lozano, J., Krainer, A.R., Moscat, J. and Caceres, J. F. (2000). The MKK(3/6)-p38-signaling cascade alters
References
107
the subcellular distribution of hnRNP A1 and modulates alternative splicing regulation.J Cell Biol 149, 307-16.
Vidal, V. P., Verdone, L., Mayes, A. E. and Beggs, J. D. (1999).Characterization of U6 snRNA-protein interactions. Rna 5, 1470-81.
Visa, N., Izaurralde, E., Ferreira, J., Daneholt, B. and Mattaj, I. W. (1996).A nuclear cap-binding complex binds Balbiani ring pre-mRNA cotranscriptionally andaccompanies the ribonucleoprotein particle during nuclear export. J Cell Biol 133, 5-14.
West, M. L. and Corden, J. L. (1995). Construction and Analysis of YeastRNA Polymerase II CTD Deletion and Substitution Mutations. Genetics 140, 1223-1233.
Wetterberg, I., Zhao, J., Masich, S., Wieslander, L. and Skoglund, U.(2001). In situ transcription and splicing in the Balbiani ring 3 gene. Embo J 20, 2564-74.
Wieland, T. and Faulstich, H. (1991). Fifty years of amanitin. Experientia 47,1186-93.
Will, C. L. and Luhrmann, R. (2001). Spliceosomal UsnRNP biogenesis,structure and function. Current Opinion in Cell Biology 13, 290-301.
Wolin, S. L. and Cedervall, T. (2002). THE LA PROTEIN. Annual Review ofBiochemistry 71, 375-403.
Wuarin, J. and Schibler, U. (1994). Physical isolation of nascent RNA chainstranscribed by RNA polymerase II: evidence for cotranscriptional splicing. Mol CellBiol 14, 7219-25.
Wyatt, J. R., Sontheimer, E. J. and Steitz, J. A. (1992). Site-specific cross-linking of mammalian U5 snRNP to the 5' splice site before the first step of pre-mRNAsplicing. Genes Dev 6, 2542-53.
Wyers, F., Rougemaille, M., Badis, G., Rousselle, J. C., Dufour, M. E.,Boulay, J., Regnault, B., Devaux, F., Namane, A., Seraphin, B. et al. (2005). Crypticpol II transcripts are degraded by a nuclear quality control pathway involving a newpoly(A) polymerase. Cell 121, 725-37.
Yih, L. H., Peck, K. and Lee, T. C. (2002). Changes in gene expressionprofiles of human fibroblasts in response to sodium arsenite treatment. Carcinogenesis23, 867-76.
Yu, J.-Y., DeRuiter, S. L. and Turner, D. L. (2002). RNA interference byexpression of short-interfering RNAs and hairpin RNAs in mammalian cells10.1073/pnas.092143499. PNAS 99, 6047-6052.
Yuryev, A., Patturajan, M., Litingtung, Y., Joshi, R. V., Gentile, C.,Gebara, M. and Corden, J. L. (1996). The C-terminal domain of the largest subunit ofRNA polymerase II interacts with a novel set of serine/arginine-rich proteins. Proc NatlAcad Sci U S A 93, 6975-80.
Zeng, C. and Berget, S. M. (2000). Participation of the C-terminal domain ofRNA polymerase II in exon definition during pre-mRNA splicing. Mol Cell Biol 20,8290-301.
Zenklusen, D., Vinciguerra, P., Wyss, J. C. and Stutz, F. (2002). StablemRNP formation and export require cotranscriptional recruitment of the mRNA exportfactors Yra1p and Sub2p by Hpr1p. Mol Cell Biol 22, 8241-53.
Zhao, X., Pendergrast, P. S. and Hernandez, N. (2001). A positionednucleosome on the human U6 promoter allows recruitment of SNAPc by the Oct-1POU domain. Mol Cell 7, 539-49.
References
108
Zhou, M., Halanski, M. A., Radonovich, M. F., Kashanchi, F., Peng, J.,Price, D. H. and Brady, J. N. (2000). Tat modifies the activity of CDK9 tophosphorylate serine 5 of the RNA polymerase II carboxyl-terminal domain duringhuman immunodeficiency virus type 1 transcription. Mol Cell Biol 20, 5077-86.
Zhou, Z., Licklider, L. J., Gygi, S. P. and Reed, R. (2002). Comprehensiveproteomic analysis of the human spliceosome. Nature 419, 182-5.
Zhu, J., Mayeda, A. and Krainer, A. R. (2001). Exon Identity Establishedthrough Differential Antagonism between Exonic Splicing Silencer-Bound hnRNP A1and Enhancer-Bound SR Proteins. Molecular Cell 8, 1351-1361.
Appendix
109
Appendix
Pictures are taken from the Ensembl Genome Browser. U6 snRNA is indicated as grey box in the center
of each image. Neighboring genes are indicated as black or red boxes with dashed lines.
U6-1 genomic locus.
U6-2 genomic locus.
Appendix
110
U6-3 genomic locus.
U6-4 genomic locus.
U6-5 genomic locus.
Appendix
111
U6-6 genomic locus.
U6-7 (left) and U6-8 (right) genomic locus.
U6-9 genomic locus.
Acknowledgements
112
Acknowledgements
I would like to thank Karla Neugebauer for her wonderful support and encouragement. I
could not have had a better supervisor and mentor than you.
Many thanks to the Neugebauer lab, especially to Janina Görnemann for all good
scientific and personal advice and Aparna Sapra for her help.
I would also like to thank Anita Bledau for being such a fabulous student and her help
with the U6 snRNA project.
I want to acknowledge the members of my thesis advisory committee, Elly Tanaka and
Francis Stewart, who introduced me to camptothecin and c-fos as a model gene. His
advice was very helpful and gratefully received.
Francis Stewart, Neus Visa and Karla Neugebauer are gratefully acknowledged for
reviewing my thesis.
Many thanks to Reinhard Lührmann, Elisa Izaurralde, Dirk Eick, Doug Black, Maria
Carmo-Fonseca and Robert White for generously providing antibodies.
Thanks to Anne Classen, Ganka Nikolova, Susanne Bechstedt, Anne-Belle Schlaitz and
Sebastian Schuck for sharing all the good times and the bad times with me. Without
you, the work here would only be half as much fun!
I also want to thank my family for their support during my work.
Thomas, I cannot express in words how grateful I am that you were always at my side
during this time. Thank you for all the interest you had in my little polymerases and
genes. Thank you for always believing in me.
Declaration
113
Declaration
I herewith declare that I have produced this paper without the prohibited assistance of
third parties and without making use of aids other than those specified; notions taken
over directly or indirectly from other sources have been identified as such. This paper
has not previously been presented in identical or similar form to any other German or
foreign examination board.
The thesis work was conducted from September 2002 to April 2006 under the
supervision of Dr. Karla M. Neugebauer at the Max Planck Institute of Molecular Cell
Biology and Genetics in Dresden, Germany.
Dresden, May 8, 2006
Imke Listerman