Manipulating Splicing by Inducing Terminal Intron Retention Bachelor of Forensics (Forensic Biology and Toxicology) Bachelor of Science (Molecular Biology and Biomedical Science) School of Veterinary and Life Sciences Murdoch University, Western Australia Supervisors: Professor Steve Wilton Professor Sue Fletcher Division of Research and Development, Centre for Comparative Genomics This thesis is submitted for Honours degree in Molecular Biology at Murdoch University, Western Australia. November, 2015 Peilin Wang (31397183)
146
Embed
Manipulating Splicing by Inducing Terminal Intron Retentionresearchrepository.murdoch.edu.au/id/eprint/29800/1/whole.pdf · different gene transcripts and only one report of antisense
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Manipulating Splicing by Inducing
Terminal Intron Retention
Bachelor of Forensics (Forensic Biology and Toxicology)
Bachelor of Science (Molecular Biology and Biomedical Science)
School of Veterinary and Life Sciences
Murdoch University, Western Australia
Supervisors:
Professor Steve Wilton
Professor Sue Fletcher
Division of Research and Development, Centre for Comparative
Genomics
This thesis is submitted for Honours degree in Molecular Biology at
Murdoch University, Western Australia.
November, 2015
Peilin Wang
(31397183)
Declaration
I declare this thesis is my own account of my research and contains as its main content
work which has not been previously submitted for a degree at any tertiary education
institution.
Peilin Wang
November, 2015
iii
Abstract
Regulation of gene expression can occur at the pre-mRNA level by alternative splicing.
As a type of alternative splicing, intron retention received attention only in recent years.
Specifically, there are few reports on endogenous terminal intron retention event in
different gene transcripts and only one report of antisense oligonucleotide induced
retention of the terminal intron in SMN transcripts. So far, there is no report of induced
terminal intron retention other gene transcripts. Terminal intron retention is proposed to
be inducible in human gene transcripts using antisense oligonucleotides, with an effect
on expression at the mRNA and/or protein levels. Five gene transcripts, LMNA, LMNC,
ITGA4, SOD1, and DMD, were selected and for each transcript, the terminal intron/exon
splice site and exon-internal sequences of the last exon were targeted for antisense
oligonucleotide annealing. It was found that terminal intron retention is inducible in
LMNC and ITGA4 transcripts, and possibly LMNA transcripts. Inclusion of the terminal
intron decreases full-length LMNC and ITGA4 mRNA expression whereas its effect on
protein expression is undetermined. Using the PMO chemistry resulted in a reduced
effect on terminal intron retention compared to when 2’-OMeAOs were used. In
addition, in both LMNC and ITGA4 transcripts, terminal intron retention led to
approximately 3% to 13% decrease in full-length mRNA expression. Changing
transfection parameters such as antisense oligonucleotide concentrations, transfection
reagent, and transfection duration can also influence transfection outcome. Possible
features that may predict terminal intron retention, including but not limited to, splicing
motifs, intron length and splice site strengths, were compared between transcripts with
and without retained terminal intron. However, no specific features for retention of the
terminal intron were observed. Overall, this project shows that terminal intron retention
is inducible in human gene transcripts other than SMN transcripts, but not in all gene
transcripts examined.
iv
Table of Contents
Front Page i
Declaration ii
Abstract iii
Table of Contents iv-viii
Acknowledgements ix
Abbreviations x
List of Figures xi-xiii
List of Tables xiv
1. Introduction 1-4
1.1 Mechanism and Regulation of Pre-mRNA Splicing 4
1.1.1 Mechanism of Splicing 4-5
1.1.2 Regulation of Splicing 6
1.2 Alternative Splicing 7
1.2.1 Common Types of Alternative Splicing 7-11
1.2.2 Alternative Splicing and Gene Expression 12
1.2.2.1 Negative Regulation 12-14
1.2.2.2 Positive Regulation: BCL-X 14-15
1.2.2.3 Indirect Regulation: Homer 15-16
1.2.3 Alternative Splicing and Neuronal Development 17-18
1.3 Using Antisense Oligonucleotides (AOs) to Manipulate Splicing 19-20
Table 2.6 Gel Electrophoresis and Gel Visualisation 43
Table 2.7 Product Isolation and DNA Sequencing 44
Table 2.8 Semi-quantitative Analysis 44
Table 3.1 Sequences of 2'-OMeAOs used in the project 46
Table 3.2 Sequences of primers used in the project 48
Table 3.3 Sequences of PMOs and leashes used in PMO lipoplex
transfections
54
Table 3.4 RT-PCR cycling conditions for different gene transcripts 55
Chapter 1: Introduction
1
1. Introduction
For any gene to be expressed, it must first undergo transcription in the nucleus to
generate pre-mRNA, which has to be processed before it can be exported out of the
nucleus to be translated into protein in the cytoplasm (Figure 1.1) or used as non-coding
RNAs (e.g. miRNA, snoRNA, snRNA). Gene expression involves many steps, and
regulation of gene expression can occur at different levels: chromatin, DNA, pre-mRNA,
mature mRNA, and protein. At each level, gene expression may be controlled by
multiple factors (Figure 1.1).
Figure 1.1 Gene expression in a eukaryotic cell. There are four steps in gene expression: transcription, pre-mRNA splicing, nuclear export, and translation. Transcription and splicing occur in the nucleus while translation takes place at the ribosomes in the cytoplasm.
(Taken from O'Connor, C. M. & Adams, J. U. 2010. Essentials of Cell Biology.
Cambridge, MA: NPG Education.)
Chapter 1: Introduction
2
At the pre-mRNA level, numerous important processes have to occur in order for pre-
mRNA transcripts to undergo nuclear export, including 5’ capping, 3’ end cleavage, 3’
polyadenylation, splicing, deposition of exon junction complex, and recruitment of
mRNA nuclear export factors (Carmody and Wente, 2009). Splicing takes place in the
nucleus, removing introns (non-coding sequences) and joining exons (segments of the
mature gene transcript, including non-coding exons in untranslated regions). Selection
Figure 1.2 Gene expression can be controlled at several levels. The levels of control can include alteration of the molecule, or they can be categorised into the stages of transcription, post-transcription, translation, and post-translation. Alternative splicing is an example of post-transcriptional regulation that occurs at the pre-mRNA level. Alternative splicing generates isoforms of gene transcripts that may be degraded, sequestered in the nucleus, or exported to the cytoplasm and translated.
Chromatin/
Chromosome
DNA
Pre-mRNA
Mature mRNA
Protein
Histone
Modification
Acetylation/
Deacetylation
Chromosomal
Rearrangement
Inversion
Translocation
Methylation
Alternative
Promoter
Enhancer
Methylation
Transcription
Factors Alternative
Transcription
Start Site
Splicing
Alternative
Splicing
3’ Polyadenylation
5’ Capping
miRNA
mRNA half-life
Phosphorylation
Glycosylation
Ubiquitination
Lipidation
Cleavage
Nuclear
export Exon Junction
Complex
Nonsense-
mediated decay
3’-end cleavage
Chapter 1: Introduction
3
of specific exons by the splicing machinery may be regulated in a tissue or
developmentally specific manner to generate alternative transcripts (Boise et. al., 1993;
splicing silencers (ISSs) (Will and Luhrmann, 2011). These cis-acting elements are
found within the pre-mRNA and exert their effects on the transcript by interacting with
trans-acting elements. Additionally, the relative distance between cis-acting elements,
as well as RNA secondary structure that may mask or bring cis-acting elements closer,
can also dictate the splicing process (Coelho and Smith, 2014). Trans-acting elements
can be RNA binding proteins (RBP) that recognise and bind to the cis-acting sequence
elements, or non-coding RNAs such as microRNAs (miRNAs) (Boutz et. al., 2007a;
Kalsotra et. al., 2010) and small nucleolar RNAs (snoRNAs) (Kishore and Stamm, 2006)
that bind to the pre-mRNA transcript or a RBP. Two extensively studied RBP families
are the SR proteins and the heterogeneous nuclear ribonucleoproteins (hnRNP). SR
proteins typically enhance splicing (Manley and Tacke, 1996; Shepard and Hertel, 2009)
while hnRNPs typically have negative effects on splicing (Krecic and Swanson, 1999).
In addition to the cis- and trans-acting elements, splicing can be affected by the rate of
transcription, implying that splicing can be regulated by chromatin modifications (Luco
and Misteli, 2011) (Figure 1.2).
Chapter 1: Introduction
7
1.2 Alternative Splicing
1.2.1 Common Types of Alternative Splicing
Constitutive splicing is typically defined as the removal of introns and ligation of exons
in a pre-mRNA, such that the mature mRNA consists of all exons and no introns.
Alternative splicing differs from constitutive splicing in that the former alters the
mature transcript either by excluding an exon, including a mutually exclusive exon,
retaining an intron, altering the length of the constitutive exons, or in other ways that
resulted in a mature mRNA isoform that is different from that formed by constitutive
splicing. Alternative splicing is a common mechanism in eukaryotes that complicates
the definition of an exon and intron, where an exon may become an intron if skipped in
one particular transcript (Figure 1.4A), while intronic sequence could be regarded as
part of an exon if included in the mature mRNA (Figure 1.4B, C (right diagram), D
(right diagram)). Through high throughput deep sequencing, it was discovered that gene
transcripts from 92-95% of human multi-exon genes undergo alternative splicing (Pan
et. al., 2008; Wang et. al., 2008). Alternative splicing occurs when alternative splice
site(s) are recognised and used by the snRNPs to direct the spliceosome to process the
pre-mRNA in a different way. Four main models of alternative splicing – exon skipping,
intron retention, alternative 5’-ss, and alternative 3’-ss– are observed in the eukaryotic
transcriptome (Table 1.1, Figure 1.4).
Chapter 1: Introduction
8
Table 1.1 Four main models of alternative splicing events and resultant gene transcripts. The alternatively spliced isoforms may have functional roles, such as sex determination and development of the immune system. However, many isoforms do not have known biological functions. Some isoforms may be produced transiently to ensure developmental regulation of gene expression, and others are disease-associated.
Model Altered Gene Transcript Function of altered transcripts Reference
Exon skipping dsx (drosophila) Skipping of exon 4 gives rise to the
male sex
(Burtis and
Baker, 1989)
Eif4enif1 (mouse) Skipping of exon 11 forms the shorter
Intron retention hKLK (human) Diagnostic/ prognostic value in
prostate cancer
(Michael et. al.,
2005)
hDKC1 (human) Isoform 6 encodes a snoRNA
(SNORA36A) within the retained
intron
(Turano et. al.,
2013)
LMNA (human) Missense mutation in 5’-ss of intron 9
resulted in intron 9 retention, which
was identified in 12 limb girdle
muscular dystrophy-1B affected
members and 2 unaffected members
of a family
(Muchir et. al.,
2000)
LY6G5B (human) Transcripts with retained first intron
are found within majority of tissues,
suggesting a functional role of intron-
retaining transcripts
(Calvanese et.
al., 2008)
PABPN1 (human) Retention of terminal intron in
PABPN1 resulted in its nuclear
degradation, maintaining homeostatic
expression
(Bergeron et. al.,
2015)
ApoE (mouse) Intron 3 is retained to restrict
expression of apolipoprotein E4 that
is implicated in neurodegenerative
diseases
(Xu et. al., 2008)
Chapter 1: Introduction
9
Lmnb1 (mouse) Intron retention resulted in
downregulation of Lmnb1 in mature
granulocytes, and decreased amount
of circulatory granulocytes
(Wong et. al.,
2013)
Alternative 5’-ss BCL-X (human) Generates the shorter pro-apoptotic
BCL-XS isoform that is important in
immune system; Constitutively
spliced shorter isoform, BCL-XL, is
anti-apoptotic
(Boise et. al.,
1993)
ATM (human) Activation of cryptic 5’-ss in intron 11
resulted from pesudoexon insertion
and is implicated in ataxia-
telangiectasia
(Cavalieri et. al.,
2013)
TMEM165 (human) Activation of cryptic 5’-ss in intron 4
resulted from pesudoexon insertion
and is implicated in congenital
disorders of glycosylation
(Yuste-Checa et.
al., 2015)
Alternative 3’-ss CLN2 (human) Activation of cryptic 3’-ss in intron 7
found to be a novel mutation that
causes late infantile neuronal ceroid
lipofuscinosis
(Bessa et. al.,
2008)
HBB (human) Activation of cryptic 3’-ss in intron 1
generated a non-functional β-globin,
resulting in β+-thalassemia
(Busslinger et.
al., 1981)
Chapter 1: Introduction
10
A) Exon Skipping B) Intron Retention
C) Alternative 5’-ss
Alternative 5’-ss
Shorter blue exon
Alternative 5’-ss
Elongated blue exon
Alternative 3’-ss D) Alternative 3’-ss
Shorter
yellow exon
Alternative 3’-ss
Elongated yellow exon
Chapter 1: Introduction
11
Figure 1.4 Main models of alternative splicing. A) In the exon skipping model, one or more exon(s) are excluded from the final mRNA transcript, due to the use of a different pair of donor (5’-ss) and acceptor (3’-ss) splice sites. The central exon (blue) is not spliced into the mature mRNA transcript and is removed with the introns. B) In the intron retention model, an intron is retained in the mature mRNA transcript as a result of inactivation of splice sites, or blocking of a cis-acting motif or a trans-acting element binding site. The retained intron becomes an exon in the transcript. C) and D) In the alternative 5’ and 3’ splice site models, the 5’-ss and 3’-ss recognised by the snRNPs are different from the commonly used splice sites recognised in constitutive splicing. Activation of a cryptic 5’- or 3’-ss within an exon results in a shorter exon, while activation of a cryptic 5’- or 3’-ss within an intron results in an elongated exon. E) Constitutive splicing.
Boxes represent exons and the blue lines represent introns. Red lines show constitutive splicing, while green lines indicate alternative splicing patterns. Resultant transcripts from different types of alternative splicing are shown in A) to D). E) shows constitutive splicing and the resultant transcript.
E) Constitutive Splicing
Chapter 1: Introduction
12
1.2.2 Alternative Splicing and Gene Expression
1.2.2.1 Negative Regulation
Splicing controls gene expression at several levels, from isoform production to nuclear
export. Similarly, alternative splicing may turn gene expression on or off, and there are
several mechanisms by which gene expression may be modulated by alternative splicing,
one of which is via the nonsense-mediated decay (NMD) pathway.
NMD is a degradation mechanism that removes mRNA transcripts carrying premature
termination codons (PTC). The process where alternative splicing introduces a PTC that
triggers NMD, is referred to as ‘regulated unproductive splicing and translation (RUST)
(Lareau et. al., 2007), and the expression of several genes under the control of RUST is
shown in Table 1.2. The spliceosome deposits an exon junction complex (EJC)
approximately 20 to 24 bases upstream of the splice junction on the pre-mRNA. Where
a premature termination codon (PTC) is located more than 50 nucleotides upstream of
the last EJC, via either insertion of an alternative exon or an intron with a stop codon, or
disruption of the reading frame, NMD will be triggered (Lewis et. al., 2003; Lareau et.
al., 2007) (Figure 1.5). However, there are exceptions to this model, as not all
transcripts with PTC that fulfil the 50-nucleotide rule elicit NMD (Lareau et. al., 2007),
with one example being the human Dyskeratosis congenita 1 (hDKC1) transcript
(Turano et. al., 2013). Furthermore, productive splicing can also be coupled to NMD, as
seen in the case of SC35 (SRSF2) auto-regulation (Stévenin et. al., 2001).
Overexpression of SC35 results in alternative splicing of its transcripts, generating 7
isoforms, one of which is a 1.7kb transcript with an additional exon. This isoform
retains full coding capacity, but is degraded by NMD because the stop codon is situated
209 bases upstream from the last EJC (Stévenin et. al., 2001). Nevertheless, auto-
Chapter 1: Introduction
13
regulation of SC35 is categorised as RUST in the literature (Green et. al., 2003; Lareau
et. al., 2007).
Table 1.2 Genes with alternatively spliced transcripts that trigger RUST.
Gene Organism Alternative Splicing Event Reference
Lmnb1 Mouse Intron retention (Wong et. al., 2013)
Hps1 Mouse Alternative 5’-ss (Hamid and Makeyev, 2014)
PTBP1 Human Exon skipping (Wollerton et. al., 2004)
STAT3 Human Exon skipping (Ward et. al., 2014)
NDUFS4 Human Exon insertion (Petruzzella et. al., 2005)
*SC35 (SRSF2) Human Exon insertion (Stévenin et. al., 2001)
*Alternative splicing produces a functional isoform that is degraded by NMD.
Chapter 1: Introduction
14
Figure 1.5 Alternative splicing and nonsense-mediated mRNA decay. Alternative splicing that introduces a premature termination codon (PTC) more than 50 nucleotides upstream of the last exon junction complex (EJC) can elicit nonsense-mediated mRNA decay (NMD). A) Normal translation occurs when the termination codon is located downstream of the last EJC. Ribosome (purple complex) displaces all EJC and protein is produced. B) Intron retention may introduce a PTC located more than 50 nucleotides upstream of the last EJC. Ribosome stops before the last EJC, which triggers NMD, inhibiting protein translation. (Adapted from Green et. al., 2003)
1.2.2.2 Positive Regulation: BCL-X
Alternative splicing is an important process not only because it expands the eukaryotic
transcriptome and proteome, but also because of its role in generating gene products that
are essential in protecting cells from being destroyed. For example, the human BCL2-
like 1 gene (BCL2L1/BCL-X) gives rise to a short form, BCL-XS, by activating a
proximal 5’-ss within the second exon (Figure 1.4C, left diagram), and the canonical
long form, BCL-XL (Boise et. al., 1993). While BCL-XL has anti-apoptotic properties,
BCL-XS is pro-apoptotic, and the two isoforms exhibit tissue-specific expression (Boise
---AAAA G-Me
EJC EJC ---AAAA G-Me
NMD
A) B)
EJC EJC ---AAAA G-Me
>50nt
EJC ---AAAA G-Me
Transcription
Translation
Protein
Chapter 1: Introduction
15
et. al., 1993). In the human brain, BCL-XL is the only expressed form (Boise et. al.,
1993), an important feature necessary to ensure long-term survival of brain cells. In the
lymphoid system where high levels of BCL-X mRNA are also observed, the BCL-XS is
the predominant form in immature double positive thymocytes that form T-cells (Boise
et. al., 1993). Double positive thymocytes are immature T cells that express both CD4+
and CD8+ glycoproteins on the cell surface, while a mature T cell will express only
either CD4+ or CD8
+ glycoprotein. Expression of the BCL-XS form allows for deletion
of thymocytes that recognise and will hence attack cells within our body to cause
autoimmune diseases. BCL-XS is also predominantly expressed in activated T-cells in
response to foreign antigens (Boise et. al., 1993). Boise et. al. (1993) postulated that
induced expression of BCL-XS in activated T-cells plays a role in promoting
amplification of T-cells in an immune response. While expression of BCL-XS increases
the susceptibility of T-cells to apoptosis, anti-apoptotic BCL-2 was found to be up-
regulated during immune response. This suggests that BCL-2 expression far exceeds the
expression of BCL-XS in activated T-cells, preventing apoptosis and accumulation of T-
cells in an immune response (Boise et. al., 1993). From this example of alternative
splicing of BCL-X, we can see the importance of the balance of alternative splicing in
generating gene products that are essential for proper development and functioning of
our immune system.
1.2.2.3 Indirect Regulation: Homer 1
Alternative splicing may also regulate gene expression indirectly. Rather than through
alternative splicing of a particular gene (target) transcript, the splicing pattern of another
transcript that codes for a regulatory protein (i.e. regulation cascade) of the target
transcript is altered. Such alternative splicing-mediated indirect gene regulation is seen
in the regulation of the gonadotropin hormones, follicle-stimulating hormone (FSH) and
Chapter 1: Introduction
16
Figure1. 6 Homer1 isoforms. Alternative splicing of the longer isoforms, Homer1b and Homer1c, generates the short isoform, Homer1a. (Taken from Wang et. al., 2014)
Exon
Intron
luteinising hormone (LH) by Homer1 isoforms in mouse LβT2 cells (Wang et. al.,
2014). FSH and LH regulate menstrual cycle and ovulation, hence controlled regulation
of genes that code for these hormones is important, and is mediated by alternative
splicing and processing of Homer1. Alternative splicing of Homer1 is controlled by the
release of gonadotrophin-releasing hormone (GnRH) that induces the splicing of the
longer mouse Homer1b/c to generate the shorter Homer1a (Figure 1.6). All three
isoforms (Homer1a, 1b, and 1c) can alter expression of the FSH and LH beta-subunits
and to a lesser extent, LHβ expression, while Homer1a enhances FSHβ expression
(Wang et. al., 2014). This study shows that alternative splicing can produce different
isoforms that have opposing regulatory effects on another gene, demonstrating an
additional level of gene regulation via alternative splicing.
Chapter 1: Introduction
17
1.2.3 Alternative Splicing and Neuronal Development
Alternative splicing is particularly important in neuronal development and has pivotal
roles in: i) development of nervous tissues (neurogenesis), ii) synapse formation and
maturation (synaptogenesis), and iii) neuronal migration (Norris and Calarco, 2012).
For example, polypyrimidine tract binding proteins 1 and 2 (PTBP1 and PTBP2/nPTBP)
are hnRNPs that play a role in regulating synaptogenesis. They regulate the expression
of postsynaptic density protein 95 (Psd-95) that is involved in the maturation of
excitatory synapses (Zheng et. al., 2012). As such, Psd-95 expression is enhanced
during late neuronal development, and this is facilitated by PTBP1 and PTBP2
expression. Expression of PTBP1 and PTBP2 represses Psd-95 expression, by inducing
exon 18 skipping in mouse Psd-95 and introducing a PTC that elicits NMD (Zheng et.
al., 2012). PTBP1 is highly expressed in neuronal progenitor cells and this prevents
unnecessary expression of Psd-95 in non-neuronal cells. While PTBP2 is expressed in
neuronal cells, it is down-regulated in late neuronal differentiation, allowing Psd-95
expression and neuron maturation (Zheng et. al., 2012).
Vertebrate and neural-specific Ser/Arg repeat-related protein of 100 kDa
(nSR100/SRRM4) is an SR-related protein that can cause inclusion of a 6-nucleotide
micro-exon in mouse Unc13b/Munc13, a gene that is involved in formation of neurites
(Quesnel-Vallieres et. al., 2015). Exclusion of this micro-exon in wild-type neurons
resulted in shorter neurites than in unaltered wild-type neurons (Quesnel-Vallieres et. al.,
2015). Conversely, inclusion of the 6-nucleotide micro-exon in mutant cells lacking
nSR100 expression restored neurite lengths to levels observed in wild-types (Quesnel-
Vallieres et. al., 2015). Another example that illustrates the importance of alternative
splicing in neuronal development is the synaptic surface proteins, neuroligin-1 and β-
neurexin. Neuroligin-1 and β-neurexin can be alternatively spliced to form isoforms that
Chapter 1: Introduction
18
will determine if the synapse is excitatory or inhibitory (Ichtchenko et. al., 1995; Chih
et. al., 2006; Li et. al., 2007). Neuroligin-1 can undergo insertion of different exons at
sites A and/or B (Figure 1.7) (Chih et. al., 2006), while β-neurexin can be alternatively
spliced at two sites, S4 and S5, where a different exon is inserted depending on the site
(Figure 1.7). Neuroligin-1 with an exon insertion at B specifically binds to the β-
neurexin lacking an alternative exon at S4 (Ichtchenko et. al., 1995; Chih et. al., 2006),
forming an excitatory glutamatergic synapse (Chih et. al., 2006; Li et. al., 2007). In
contrast, β-neurexin with an insertion at S4 was found to drive formation of inhibitory
GABAergic synapse (Chih et. al., 2006).
Altogether, the above examples clearly illustrate some of the important roles of
alternative splicing in regulating gene expression and neuronal development.
Figure 1.7 Schematic representation of β -neurexin and neuroligin-1 transcripts that code for synaptic surface proteins. Both β -neurexin and neuroligin-1 have two alternative splicing sites where an exon can be inserted. (Taken from Li et. al., 2007)
Chapter 1: Introduction
19
1.3 Using Antisense Oligonucleotides (AO) to Manipulate Splicing
Mutations of cis-acting elements involved in regulating splicing can lead to abnormal
splicing, where an abnormal mRNA transcript and/or protein are produced. Aberrant
splicing has been associated with numerous human diseases, including cancers
(Goodison et. al., 1998; Le Hir et. al., 2002; Lokody, 2014), β-thalassemia (Busslinger
Yuste-Checa et. al., 2015). For example, an intronic point mutation in the fibrinogen
gamma chain (FGG) gene increases the strength of a cryptic 5’-ss within the intron
(Spena et. al., 2007). This activated cryptic 5’-ss then cooperates with the downstream
canonical 3’-ss, while a normally silenced 3’-ss within the intron recognises the
upstream canonical 5’-ss and cause the splicing of a pseudo-exon in the mature mRNA
Chapter 1: Introduction
33
transcript (Spena et. al., 2007). Pseudo-exon activation arising from deep intronic
mutations was also observed in DMD. Mutations within introns 11, 25, 45, 47 and 62
create novel splice sites that activate a cryptic pairing splice site and result in the
splicing of a pseudo-exon in the mature mRNA transcript (Tuffery-Giraud et. al., 2003;
Gurvich et. al., 2008).
The mechanism of pseudo-exon activation may also involve the deletion or creation of a
cis-acting splicing regulatory element, and its interaction with a trans-acting element.
Deep intronic mutations may activate a cis-acting element that a trans-acting element
binds to and then enhances splicing of the pseudo-exon in the mature gene transcript
(Homolova et. al., 2010; Rimoldi et. al., 2013). Pseudo-exon activation in the human
methionine synthase reductase (MTRR) gene transcript is mediated by creation of a
SF2/ASF binding ESE motif (Homolova et. al., 2010). While in the fibrinogen gamma-
chain (FGG) gene transcript, pseudo-exon activation involves the creation of a cis-
Figure 1.8 Schematic diagram showing the comparison of a complete retention and partial retention of an intron. Pseudo-exon activation may be thought of as partial intron retention. A pseudo-exon originates from an intron, and it is only a part of the intron that gets spliced into the mature gene transcript. Activation of a cryptic or creation of a novel 5’- and/or 3’-ss within the intron results in pseudo-exon activation.
Green rectangle represents exon and orange rectangle represents intron.
Full intron retention Partial intron retention
(Pseudo-exon activation)
Chapter 1: Introduction
34
acting 3 G-run motif recognised by the trans-acting hnRNP F splicing factor (Rimoldi
et. al., 2013). Noticeably, hnRNPs are typically associated with inhibiting splicing by
binding to ESSs, but in FGG pseudo-exon, hnRNP F binds to an ESE (3 G-run motif)
and activates pseudo-exon inclusion. Nonetheless, in the absence of the 25bp region that
consists of the 3 G-run motif, hnRNP F binds to two other G-run motifs within the
pseudo-exon that function as ESSs, and inhibits pseudo-exon activation (Rimoldi et. al.,
2013). Another possible mechanism of pseudo-exon activation is via deletion of ESS
motifs downstream of the pseudo-exon (Greer et. al., 2015). Other rare pseudo-exon
activation mechanisms include genomic rearrangements, genomic inversions, and loss
of upstream 5’-ss or downstream 3’-ss (Dhir and Buratti, 2010). Examples of genomic
rearrangement and inversion that resulted in pseudo-exon activation are documented in
the DMD gene (Cagliani et. al., 2004; Madden et. al., 2009; Khelifi et. al., 2011) and
they are speculated to cause pseudo-exon activation by narrowing the distance between
Scrambled control CCTCTTACCTCAGTTACAATTTATA TATAAATTGTAACTGAGGTAAGAGG
3.8 RNA Analysis
3.8.1 Total RNA Extraction
Following transfection, total RNA was isolated using Tri-reagent (Zymo Research,
Irvine, California) and purified using Direct-zolTM
RNA MiniPrep kit (Zymo Research,
Irvine, California; protocol ver 1.1.0). Slight changes made to the manufacturer’s
protocol include centrifuging at 12 000rcf for all steps, using 800µl RNA prewash and
centrifuging for 2 minutes in the prewash step, and using 30µl molecular grade water
(Sigma-Aldrich, Sydney, Australia) for elution. cDNAs were then synthesized and
amplified using either one-step or two-step RT-PCR, as indicated below.
Chapter 3: Methods
55
3.8.2 One-step RT-PCR
One-step RT-PCR was carried out on RNA samples from cells treated with SMN-,
LMNA/C- and ITGA4- targeting AOs, using the SuperScript® III One-Step RT-PCR
System with Platinum® Taq DNA Polymerase (Invitrogen® Thermo Fisher Scientific,
Melbourne, Australia). Each reaction consisted of 0.35µl SuperScript® III RT/
Platinum® Taq Mix, 1x reaction mix, 25ng forward primer, 25ng reverse primer, 50ng
RNA, and water to a final volume of 12.5µl. Depending on the primer pair and the size
of the expected product, the cycling conditions differed slightly for each targeted gene
transcript (Table 3.4).
Table 3.4 RT-PCR cycling conditions for different gene transcripts. Cycle number applies to the amplification step that includes denaturation, annealing, and extension.
Transcript cDNA
synthesis Initial
denaturation No. of cycles
Denaturation Annealing Extension (1min/kb)
SMN
55oC for 30min
94oC for 2min
25 94oC for 40s 56oC for 1min
68oC for 1min
LMNA/C 30 94oC for 30s 60oC for 1min
68oC for 2min
ITGA4 25 94oC for 40s 58oC for
1min
68oC for
1min 30s
3.8.3 Two-step RT-PCR
As the expected size of SOD1 transcripts with retained terminal intron is approximately
5 times the size of the constitutively spliced transcript, long range PCR was carried out,
using TaKaRa LA Taq® (Scientifix, Victoria, Australia). DMD transcripts were also
amplified using long range PCR as the product of interest (approximately 5kb) is larger
than that readily amplified using SuperScript® III One-Step RT-PCR (Invitrogen®
Thermo Fisher Scientific, Melbourne, Australia). Prior to carrying out long range PCR,
cDNA was synthesized according to SuperScript® IV reverse transcriptase
(Invitrogen® Thermo Fisher Scientific, Melbourne, Australia) protocol, but using half
of the reaction volume (10µl). 50µM random hexamers (Invitrogen® Thermo Fisher
Chapter 3: Methods
56
Scientific, Melbourne, Australia) was used as the primer, and 5µl of RNA sample with
the lowest concentration in each experiment was set as the amount of template RNA to
use for each reaction, where approximately 100ng of RNA from fibroblasts and 200ng
of RNA from myoblasts were used.
The synthesized cDNAs were then used in TaKaRa LA Taq® long range PCR. Each
12.5µl PCR reaction contained 1X LA buffer (Mg2+
plus), 0.625U LA Taq DNA
polymerase, 5nmol dNTP mixture, 12.5ng forward primer, 12.5ng reverse primer, 1µl
cDNA, and water. Long range PCR was programmed with initial denaturation at 94oC
for 1min, followed by 35 cycles of denaturation at 94oC for 30s, annealing at 58
oC for
30s, and extension at 72oC, 1min/kb. The extension time for amplification of SOD1 and
DMD transcripts was 2 and 6 minutes, respectively.
3.9 Gel Electrophoresis and Product Analysis
Except for DMD amplicons, all PCR products were fractionated on 2% agarose gel with
1x TAE gels at 100V. Amplified DMD cDNA were fractionated on 1% agarose gel with
1x TAE gel at 60V. All gel images were captured using the Vilber Lourmat Fusion-FX
gel documentation system (Marne-la Valle, France).
3.9.1 Product Isolation and DNA Sequencing
Amplification products of interest were re-fractionated on 2% agarose gel with 1x TAE
buffer and stained with ethidium bromide. The gel was visualised using a
transilluminator and bands were either stabbed using a P200 tip, or excised with a
scalpel. Generally, bands with sizes corresponding to those of intron-less transcripts
were stabbed, while bands of larger sizes were excised. Following stabbing, the P200
Chapter 3: Methods
57
tip was inoculated into a reaction mix that consisted of 1x GeneAmp® PCR Gold
Scientific, Melbourne, Australia), and water (50µl in total). The cycling conditions were:
initial denaturation at 94oC for 6min, followed by 30 cycles of denaturation at 94
oC for
30s, annealing at 5oC less than the original annealing temperature, for 1min, and
extension at 72oC for 2min, followed by holding at 25
oC. The reamplified products were
re-examined on agarose gels to check that the bands of interest were amplified, before
being purified using Diffinity RapidTip® (Diffinity Genomics, Henrietta, New York) as
described by the manufacturer. For samples with multiple PCR products, a clean DNA
could not be isolated from the larger bands using band stab method. Therefore, DNA
from PCR products with sizes greater than the smallest sized band were extracted and
purified for DNA sequencing using the band excision method. Agarose gel stained with
ethidium bromide was visualised using a transilluminator and bands of interest were
excised using a scalpel. DNA was extracted and purified using Macherey-Nagel
NucleoSpin® Gel & PCR Clean-up kit (Scientifix, Victoria, Australia), according to the
manufacturer’s protocol.
Purified DNA was prepared for sequencing by the Australian Genome Research Facility
Ltd (AGRF) in 12µl sequencing reaction. Each sequencing reaction consisting of
approximately 19.2pmol forward or reverse primer, purified DNA in the amount
recommended by AGRF (www.agrf.org.au) and water.
Chapter 3: Methods
58
3.9.2 Semi-quantitative Analysis
Optical densities (ODs) of RT-PCR products were determined using the Vilber Lourmat
Fusion-FX gel documentation system (Marne-la Valle, France) and BIO1D software
(Vilber Lourmat, Marne-la Valle, France). The data were then presented in graphs
prepared using Excel (Microsoft). The proportions of terminal intron-retaining and full-
length transcripts were generated by the following formula:
𝑂𝐷𝑇𝐼𝑅/𝐹𝐿/𝑈𝑛𝑖𝑑𝑒𝑛𝑡𝑖𝑓𝑖𝑒𝑑
𝑂𝐷𝑡𝑜𝑡𝑎𝑙
The ratios of full-length transcripts were generated by using the following formula:
ODFL target transcript (treated OR untreated)× ODFL 𝑆𝑀𝑁 (treated OR untreated)
ODFL 𝑆𝑀𝑁 (untreated)⁄
ODFL target transcript (untreated)
Statistical analyses were done using R statistical software (downloaded online).
Chapter 4: Results
59
4. Results
4.1 SMN-AOs Induced Terminal Intron Retention
As proof-of-concept that terminal intron retention is inducible, and can alter expression
of a human gene transcript using antisense oligonucleotides (AO), two 2’-O-methyl
antisense oligonucleotides (2’-OMeAOs) on a phosphorothioate backbone known to
induce terminal intron retention in human SMN gene transcripts (Price, personal
communication) were evaluated for use as a positive control. One of the AOs targets the
last acceptor splice site (SMN_H8A(-10+15)), while the other targets a region within
the terminal SMN exon (SMN_H8A(+57+81)). Normal human fibroblasts were
transfected with both AOs as cationic lipoplexes at 200nM, 100nM, and 50nM. RNA
extracted from the transfected fibroblasts was then amplified via RT-PCR using SMN
primer pair (5512a+5513a) (Figure 4.1A). SMN RT-PCR products without the terminal
intron (intron 7) are 404bp, while those with retained intron 7 are 848bp. The 848bp
products are present in greater abundance in treated samples than in the untreated,
confirming that both AOs were able to induce terminal intron retention in SMN
transcripts at all concentrations tested (Figure 4.2). Since terminal intron retention is
inducible at all three concentrations tested, the middle AO concentration (100nM) was
chosen as the concentration of the positive control 2’-OMeAO for subsequent
transfections.
Chapter 4: Results
60
8 7 6 5 4 444nt
3
5512a 5513a
Figure 4.1 Schematic diagrams of partial pre-mRNAs showing AO annealing sites and the location of RT-PCR primers used. A: SMN, B: LMNA, C: LMNC, D: ITGA4, E: SOD1, F: DMD pre-mRNAs. Primers 6107, 6110, 6005 and 6006 were used for sequencing only. The size of the terminal intron is indicated. The arrow and number underneath represent PCR primers. Blue lines represent introns, while boxes represent exons and the number within refers to the exon number. Black lines represent AOs. Diagrams are not drawn to scale.
322nt 12 11 10 9 8
4970 4969
10 9 8 7 6 421nt
5
5255 5259 6107
28 27 26 22 21 496nt
20
6003 5597 6110
1095nt 5 4 3 2 1
6007 6008
79 78 77 76 75 4711nt
74
6004 4894 6005 6006
A
B
F
E
D
C
Chapter 4: Results
61
Figure 4.2 SMN RT-PCR products amplified from normal human fibroblasts transfected with SMN_8A(+57+81) and SMN_8A(-10+15), using SMN primer pair (5512a+5513a). -ve: PCR negative control
It was speculated that transfection with an AO cocktail composed of an AO blocking
the donor splice site and an AO that causes intron retention, could enhance the intron
retention effect. Transfection was carried out using two SMN-targeting AO cocktail,
consisting of the AO that targets the donor splice site of the terminal intron
(SMN_H7D(+17-13)) and a terminal intron retention inducing AO, SMN_H8A(+57+81)
or SMN_H8A(-10+15). RT-PCR analysis shows that the addition of SMN_H7D(+17-13)
does not have a synergistic effect on terminal intron retention (Figure 4.3). In fact, there
was a decrease in the amount of mRNA transcripts with retained terminal intron
compared to when SMN_H8A(+57+81) or SMN_H8A(-10+15) was used alone (Figure
4.2). As SMN_H7D(+17-13) can induce SMN exon 7 skipping, there was also an
expected increase in transcripts missing SMN exon 7 (Figure 4.3).
UT
Chapter 4: Results
62
100b
p la
dder
200n
M
100n
M
50nM
200n
M
100n
M
50nM
Figure 4.3 RT-PCR of RNA extracted from normal human fibroblasts treated with AO cocktails designed to induce terminal intron retention in SMN transcripts.
4.2 Effects of Oligonucleotides on Transcript Splicing Patterns
4.2.1 Treatment of Normal Human Fibroblasts with LMNA-targeting 2’-OMeAOs
The LMNA gene can give rise to multiple isoforms by alternative splicing of the pre-
mRNA (http://www.ncbi.nlm.nih.gov/gene/4000). LMNA and LMNC are two major
transcripts encoded by LMNA gene and they code for lamin A and C proteins,
respectively, that have structural roles in maintaining nuclear shape and size
(http://www.omim.org/entry/150330). The effect of 2’-OMeAOs designed to induce
terminal intron retention within LMNA transcripts (intron 11), and later, LMNC
transcripts (intron 9), were evaluated in human fibroblasts.
Chapter 4: Results
63
RT-PCR of RNA extracted from normal human fibroblasts transfected with LMNA-
targeting 2’-OMeAOs and the positive control 2’-OMeAO was carried out using LMNA
primer pair (4970+4969) (Figure 4.1B), and SMN primer pair (5512a+513a),
respectively. The untreated sample was also amplified using LMNA primer pair. Full-
length LMNA RT-PCR products (687bp) are present in all, except for one, of the
samples of AO-treated, positive control-treated, and untreated samples (Figure 4.4) and
sequencing of this 687bp product confirmed it represents full-length transcripts without
terminal intron (Figure 4.5). No larger RT-PCR product of the desired size of 1009bp,
indicating terminal intron retention within LMNA transcripts, was evident after
transfection with the four LMNA-targeting 2’-OMeAOs designed. However, there was
variation in the abundance of full-length LMNA RT-PCR products, and signs of a dose-
dependent response for samples treated with LMNA_H12A(+10+34) and
LMNA_H12A(+35+59), where the amount of full-length LMNA products decreases
with increasing AO concentration (Figure 4.4).
Chapter 4: Results
64
Last base
of exon 11
First base of
exon 12
Figure 4.5 Chromatogram of the 687bp LMNA RT-PCR product amplified from RNA extracted from normal human fibroblasts transfected with LMNA-targeting 2’-OMeAOs.
100b
p la
dder
100b
p la
dder
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
Figure 4.4 RT-PCR evaluating transfection of normal human fibroblasts with the four LMNA-targeting 2’-OMeAOs designed to induce terminal intron retention. Amplification was carried out using LMNA primer pair (4970+4969) and only full-length LMNA products were amplified. The positive control sample (+ve) was amplified using SMN primer pair (5512a+ 5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
It was necessary to determine if variation in the abundance of full-length LMNA RT-
PCR products is due to down-regulation of LMNA mRNA expression or cell death
following AO transfection. There was also a need to distinguish between a dose-
dependent response to AO transfection and variability in full-length LMNA RT-PCR
product levels, due to transfection-associated cell death, the transfection was repeated.
Therefore, the extracted RNA amplified by RT-PCR using both LMNA primer pair
(4970+4969) and SMN primer pair (5512a+5513a). SMN transcripts are not expected to
be targeted by the AOs used to transfect the cells and thus they serve as an internal
control to indicate transfection associated cell death. A relatively consistent band for
full-length SMN transcripts across the AO-treated samples would suggest down-/up-
regulation of the LMNA transcript or a dose-dependent response, whereas variability in
the intensity of the SMN product would suggest cell death due to AO or transfection
reagent toxicity. There appears to be an increase in the abundance of full-length LMNA
products as AO concentration decreases after transfection with LMNA_H12A(+10+34)
and LMNA_H12A(+60+84) (Figure 4.6A), with no significant variability in the
abundance of full-length SMN products across the three AO concentrations (Figure
4.6B). The ratio of full-length LMNA products from AO-treated samples was compared
to untreated sample, after being normalized to the respective full-length SMN products,
and shows clear variation in the abundance of full-length LMNA products, with
transfection of LMNA_H12A(+10+34), LMNA_H12A(+35+59) and
LMNA_H12A(+60+84), all resulting in a down-regulation of full-length LMNA
transcripts (Figure 4.7).
Chapter 4: Results
66
100b
p la
dder
100b
p la
dder
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
SMN TIR
(848bp)
SMN FL (404bp)
LMNA FL (687bp)
Figure 4.6 RT-PCR evaluating transfection of normal human fibroblasts with the four LMNA-targeting 2’-OMeAOs designed to induce terminal intron retention. A: RT-PCR using LMNA primer pair (4970+4969). B: RT-PCR using SMN primer pair (5512a+5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
The RT-PCR shows a greater abundance of SMN transcripts with retained terminal
intron (848bp) for the positive control than for the untreated sample, indicating
acceptable transfection efficiency (Figure 4.6B). Interestingly, while transfection of
LMNA-targeting AOs did not cause a change in the splicing pattern of LMNA transcripts,
the AOs did appear to have influence the SMN splicing pattern. Specifically,
LMNA_H12A(+10+34), LMNA_H12A(+35+59) and LMNA_H12A(+60+84) resulted
in additional ~500bp SMN products (Figure 4.6B).
Chapter 4: Results
67
Figure 4.7 Ratios of full-length LMNA transcripts in normal human fibroblasts after transfection with LMNA-targeting 2’-OMeAOs. The ratios were determined as specified in Section 3.9.2. The ratio was undetermined for 200nM LMNA_H12A(-16+9) as no full-length SMN products were detected. UT: untreated
LMNA_H12A(-16+9) LMNA_H12A(+10+34) LMNA_H12A(+35+59) LMNA_H12A(+60+84) UT
Rat
io
4.2.2 Treatment of Normal Human Fibroblasts with LMNC-targeting 2’-OMeAOs
While intron 9 of the LMNA gene is not the terminal intron of the gene, it is referred to
as the “terminal intron” of LMNC transcripts, where transcription terminates within
LMNA exon 10. Like the LMNA experiment, RT-PCR amplification of LMNC
transcripts, after transfection with AOs designed to induce LMNC terminal intron
retention, were conducted in two sets, using the LMNC primer pair (5255+5259) (Figure
4.1C) and the SMN primer pair (5512a+5513a). A 660bp RT-PCR product is observed
in all samples amplified using LMNC primers (Figure 4.8A). Sequencing shows that this
product represents full-length LMNC transcripts without retained terminal intron (Figure
4.9A). Transfection with LMNC_H10A(+16+40), LMNC_H10A(+41+65) and
LMNC_H10A(+66+90) resulted in partial retention of the terminal intron in LMNC
transcripts (Figure 4.8A), confirmed by DNA sequencing (Figure 4.9B). Where LMNC
products with retained terminal intron (1081bp) were amplified, there was also a smaller
product (~1000bp) that could not be identified despite numerous attempts at DNA
sequencing. Additionally, LMNC-targeting 2’-OMeAOs that induced terminal intron
Chapter 4: Results
68
SMN Δ5
(308bp)
100b
p la
dder
100b
p la
dder
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
LMNC TIR (1081bp)
Unidentified product (~1000bp)
LMNC FL
(660bp)
UT +ve -ve
Figure 4.8 RT-PCR evaluating the transfection of normal human fibroblasts with the four 2’-OMeAOs targeting LMNC transcripts designed to induce terminal intron retention. A: RT-PCR using LMNC primer pair (5255+5259). B: RT-PCR using SMN primer pair (5512a+5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
retention also altered the SMN splicing pattern, resulting in a 500bp RT-PCR product
(Figure 4.8B) and interestingly, DNA sequencing shows that LMNC_H10A(+66+90)
causes skipping of SMN exon 5 (Figure 4.10).
Chapter 4: Results
69
Figure 4.9 Chromatograms of LMNC RT-PCR products. A: Chromatogram of the 660bp RT-PCR product amplified from RNA extracted from normal human fibroblasts transfected with LMNC-targeting AOs. B: Chromatogram of the 1081bp product amplified from RNA extracted from normal human fibroblasts transfected with LMNC-targeting AOs that cause a change in splicing pattern of
LMNC transcripts.
Last base
of exon 9
First base
of exon 10
5’ end of
intron 9
3’ end of
intron 9
A
B
Last base
of exon 4
First base of
exon 6
Figure 4.10 Chromatogram of the ~300bp SMN RT-PCR product amplified from RNA extracted from
normal human fibroblasts transfected with LMNC_H10A(+66+90).
Chapter 4: Results
70
Figure 4.11 Semi-quantitative analysis of terminal intron retention induced in LMNC transcripts. Although the proportions of transcripts with retained terminal intron are low, they are significantly different from transcripts derived from the untreated cells. Error bar is included for samples with terminal intron retention and sample size n=2. ^Results are from a single experiment. * p-value <0.001 UT: untreated
LMNC_H10A(-10+15) LMNC_H10A(+16+40) LMNC_H10A(+41+65)LMNC_H10A(+66+90)^ UT
Pro
po
rtio
n
TIR
Unidentified
FL
* * * * *
* * * *
Terminal intron retention induced by LMNC_H10A(+66+90) was sporadic, detected
only on a single occasion. Densitometry shows that transfection with LMNC-targeting
2’-OMeAOs decreased full-length mRNA expression by 3% to 13%, and
LMNC_H10A(+16+40) and LMNC_H10A(+41+65) have dose-dependent effects on
terminal intron retention, where the proportion of transcripts with retained terminal
intron increases as AO concentration increases (Figure 4.11).
Chapter 4: Results
71
4.2.2.1 Treatment of Normal Human Fibroblasts with LMNC-targeting PMOs
Following confirmation of terminal intron retention in LMNC transcripts induced by
LMNC-targeting 2’-OMeAOs, the two most promising sequences that induced terminal
intron retention, LMNC_H10A(+16+40) and LMNC_H10A(+41+65), were
resynthesized as phosphorodiamidate morpholinos (PMOs) for comparison between
different AO chemistries and to determine if the more stable PMO chemistry could
induce a stronger effect. The PMOs were tested at 400nM, 200nM, and 100nM, while
the highest concentration of 2’-OMeAOs tested was 200nM. PMOs are less efficiently
taken up by cells in vitro and were thus tested at the higher concentration range of
100nM to 400nM, previously optimized for SMN targeting (Price, personal
communication). Using PMOs, terminal intron retention was weaker than that induced
by the 2’-OMeAOs, and retention was only observed at the higher concentrations.
LMNC_H10A(+16+40) PMO induced terminal intron retention at 400nM and 200nM,
while LMNC_H10A(+41+65) PMO induced terminal intron retention at 400nM (Figure
4.12A).
Chapter 4: Results
72
Scrambled
control UT -ve 100b
p la
dder
100b
p la
dder
Figure 4.12 RT-PCR showing transfection with PMOs designed to induce retention of the LMNC terminal intron. A: RT-PCR using LMNC primer pair (5255+5259). B: RT-PCR using SMN primer pair (5512a+5513a).
Figure 4.13 RT-PCR evaluating the transfection of normal human fibroblasts with the four 2’-OMeAOs targeting ITGA4 transcripts designed to induce terminal intron retention. A: RT-PCR using ITGA4 primer pair (6003+5597). B: RT-PCR using SMN primer pair (5512a+5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
Figure 4.15 Semi-quantitative analysis of terminal intron retention induced in ITGA4 transcripts. Although the proportions of transcripts with retained terminal intron are considerably low, they are significantly different from transcripts of the untreated sample. Error bar is included (n=2). ^Results are from a single experiment. * p-value <0.001 UT: untreated TIR: transcripts with terminal intron retained FL: full-length transcripts (terminal intron excised)
ITGA4_H28A(-6+19) ITGA4_H28A(+20+44) ITGA4_H28A(+45+69) ITGA4_H28A(+70+94) UT
Pro
po
rtio
n
TIR
FL
* * * * * * * * *
5’ end of
intron 27
3’ end of
intron 27
Last base
of exon 27
First base of
exon 28 A
B
Figure 4.14 Chromatograms of ITGA4 RT-PCR products. A: Chromatogram of the 976bp RT-PCR product amplified from RNA extracted from normal human fibroblasts transfected with ITGA4-targeting AOs. B: Chromatogram of the 1472bp RT-PCR product amplified from RNA extracted from normal human fibroblasts transfected with ITGA4-targeting AOs that result in a different splicing pattern.
Chapter 4: Results
76
Scrambled
control UT -ve 100b
p la
dder
100b
p la
dder
400n
M
200n
M
100n
M
400n
M
200n
M
100n
M
400n
M
200n
M
100n
M
ITGA4 TIR
(1472bp)
ITGA4 FL
(976bp)
Figure 4.16 RT-PCR showing transfection with PMOs designed to induce retention of the ITGA4 terminal intron. A: RT-PCR using ITGA4 primer pair (6003+5597). B: RT-PCR using SMN primer pair (5512a+5513a).
4.2.3.1 Treatment of Normal Human Fibroblasts with ITGA4-targeting PMOs
Following confirmation of terminal intron retention, two AO sequences,
ITGA4_H28A(-6+19) and ITGA4_H28A(+20+44), were resynthesized as PMOs. The
PMOs induced terminal intron retention (Figure 4.16A), but at the 200nM and 100nM
PMO concentrations, the effect was less pronounced that that induced by 2’-OMeAOs
at the same concentrations (Figure 4.17).
Chapter 4: Results
77
4.2.4 Treatment of Normal Human Fibroblasts with SOD1-targeting 2’-OMeAOs
The SOD1 gene has 5 exons and codes for a homodimeric enzyme that breaks down
free superoxide radicals that can cause deleterious damage to the cell when present at
high levels (http://www.ncbi.nlm.nih.gov/gene/6647). Induced retention of SOD1
terminal intron (intron 4) using SOD1-targeting 2’-OMeAOs was also investigated.
RNA samples extracted from normal human fibroblasts, transfected with SOD1-
targeting AOs, were amplified by two-step RT-PCR, using SOD1 primer pair
(6007+6008) (Figure 4.1E). The positive control and untreated samples were also
amplified by one-step RT-PCR, using SMN primer pair (5512a+5513a). Analysis of the
two-step RT-PCR products showed multiple randomly amplified products, with a
consistent, intense 300bp product across all samples, while there is obvious product of
*
Figure 4.17 Proportions of ITGA4 transcripts with retained terminal intron when different AO chemistries were used. Even though there is a hint of TIR transcripts amplified from RNA extracted from cells treated with 100nM ITGA4_H28A(+20+44), optical density analysis was not possible.
* p-value <0.001 TIR: transcripts with terminal intron retained
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
200nM 100nM 200nM 100nM
ITGA4_H28A(-6+19) ITGA4_H28A(+20+44)
Pro
po
rtio
n o
f T
IR T
ran
scri
pts
PMO
2'-OMeAO
*
*
Chapter 4: Results
78
A
Figure 4.18 RT-PCR of RNA extracted from normal human fibroblasts transfected with SOD1-targeting 2’-OMeAOs. A: Two- step long range RT-PCR using SOD1 primer pair (6007+6008). B: One-step RT-PCR using SMN primer pair (5512a+5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
the anticipated size of 1395bp (Figure 4.18A). DNA sequencing shows that the 300bp
product represents full-length transcripts without terminal intron (Figure 4.19).
Chapter 4: Results
79
While no clear product-of-interest was observed, the gel image suggests a hint of intron-
retaining products amplified from cells transfected with SOD1_H5A(+21+45) 2’-
OMeAO (Figure 4.18A). Therefore, three additional 2’-OMeAOs were designed around
the region targeted by the SOD1_H5A(+21+45) 2’-OMeAO. Additionally, two 2’-
OMeAOs that target the donor splice site of SOD1 exon 4 were designed and
synthesised, to determine if blocking the donor splice site will induce retention of the
terminal intron. RT-PCR analysis of RNA extracted from cells transfected with the
second generation 2’-OMeAOs did not show any 1395bp bands (Figure 4.20A),
indicating that attempts to induce terminal intron retention in SOD1 transcripts using
these 2’-OMeAOs was not successful.
Last base
of exon 4
First base
of exon 5
Figure 4.19 Chromatogram of the 300bp SOD1 RT-PCR product amplified from RNA extracted from normal human fibroblasts transfected with SOD1-targeting AOs.
Chapter 4: Results
80
100b
p la
dder
100b
p la
dder
UT -ve
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
200n
M
100n
M
50nM
Figure 4.20 RT-PCR of RNA extracted from normal human fibroblasts transfected with second generation of SOD1-targeting 2’-OMeAOs. A: Two- step long range RT-PCR using SOD1 primer pair (6007+6008). B: One-step RT-PCR using SMN primer pair (5512a+5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
4.2.5 Treatment of Normal Human Fibroblasts and Primary Myoblasts with DMD-
targeting 2’-OMeAOs
At approximately 2.4Mb, DMD is the largest human gene known to date. DMD
comprises of 79 exons, and encodes dystrophin, a large cytoskeletal protein that is
important in maintaining muscle fibre integrity (Muntoni et. al., 2003). Multiple tissue-
specific DMD transcripts are generated by alternative promoter usage, poly(A) tail
addition sites, and splicing (http://www.ncbi.nlm.nih.gov/gene/1756).
To determine if transfection of different cell types with DMD-targeting 2’-OMeAOs
designed to induce retention of the DMD terminal intron (intron 78) will result in
different splicing patterns, transfections were carried out in normal human fibroblasts
and normal primary human myoblasts. RNA was then extracted and amplified via two-
step long range RT-PCR, using the outer DMD primer pair (6004+4894) (Figure 4.1F),
and to determine if transfection was successful, the positive control and untreated
samples were also amplified by one-step RT-PCR, using SMN primer pair
(5512a+5513a). DMD RT-PCR shows no product of the anticipated size of 5091bp in
either fibroblast or myoblast samples, but smaller bands are present in both types of
samples (Figure 4.21A and 4.22A). Three RT-PCR products, slightly smaller than
500bp, were observed in fibroblast samples (Figure 4.21A). However, amplification of
myoblast DMD cDNA produced multiple random bands and two consistent bands with
size slightly less than 500bp, across all samples (Figure 4.22A).
Chapter 4: Results
82
1kb
ladd
er
1kb
ladd
er
UT -ve +ve
500bp
1kb
1.5kb
2kb
3kb
4kb
10kb
5kb
8kb 6kb
Figure 4.21 RT-PCR of RNA extracted from normal human fibroblasts transfected with DMD-targeting 2’-OMeAOs. A: Two-step long range RT-PCR using DMD primer pair (6004+4894). RT-PCR products were fractionated on 1% agarose gel in 1x TAE. B: One-step RT-PCR using SMN primer pair (5512a+5513a). RT-PCR products were fractionated on 2% agarose gel in 1x TAE.
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
Figure 4.22 RT-PCR of RNA extracted from primary human myoblasts transfected with DMD-targeting 2’-OMeAOs A: Two-step long range RT-PCR using DMD primer pair (6004+4894). RT-PCR products were fractionated on 1% agarose gel in 1x TAE. B: One-step RT-PCR using SMN primer pair (5512a+5513a). RT-PCR products were fractionated on 2% agarose gel in 1x TAE.
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
As the anticipated size of DMD RT-PCR products with retained terminal intron is
approximately 5kb, fractionation of the products was carried out using 1% agarose gel
for good separation of large products. However, the amplified RT-PCR products were
smaller than 500bp, which did not separate well in low percent agarose gel. To better
estimation the product sizes, RT-PCR products amplified from fibroblasts and
myoblasts transfected with 200nM 2’-OMeAO were re-fractionated on a 2% agarose gel,
using the 100bp ladder as a size standard. Three products, not evident on the 1% gel
were observed when the myoblast samples were fractionated on the 2% agarose gel.
Expectedly, three products were present in the fibroblast samples. Nonetheless, there is
a clear difference between fibroblast and myoblast samples, where myoblast samples
have a much more intense 380bp product than fibroblast samples (Figure 4.23). DNA
sequencing shows that the 380bp product represents full-length DMD transcripts
without the terminal intron, and the lower ~350 product represents DMD transcripts
missing exon 78, while sequencing of the larger ~400bp product is inconclusive and this
product remains unidentified (Figure 4.23).
Chapter 4: Results
85
Last base
of exon 78
First base of
exon 79
Figure 4.23 Fibroblast and myoblast DMD RT-PCR products and chromatograms of the amplified DMD RT-PCR products. The three RT-PCR products were amplified using DMD primers (6004+4894) in both fibroblast and myoblast samples. Chromatograms of bands A and B are shown. Multiple attempts at sequencing band C were not successful.
F: fibroblast M: myoblast
Last base
of exon 77
First base of
exon 79
C Unidentified product
F 100b
p la
dder
100b
p la
dder
B DMD FL (380bp) A DMD Δ78 (348bp)
M F M F M F M
B
A
Chapter 4: Results
86
100b
p la
dder
100b
p la
dder
UT -ve +ve
Figure 4.24 RT-PCR of RNA extracted from myoblasts transfected with DMD- and SMN-targeting 2’-OMeAOs. Amplification was carried out using SMN primer pair (5512a+5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
ITGA4_H28A(-6+19) ITGA4_H28A(+20+44) ITGA4_H28A(+45+69) ITGA4_H28A(+70+94) UT
Rat
io
24h
48h
A
B
Figure 4.26 Ratios of full-length LMNA and ITGA4 transcripts in normal human fibroblasts after 24h and 48h transfections using Lipofectamine® 3000 (L3K) transfection reagent. The ratios were determined as specified in Section 3.9.2 Semi-quantitative Analysis. A: Ratios of full-length LMNA products of AO-treated samples to that of untreated samples. B: Ratios of full-length ITGA4 products of AO-treated samples to that of untreated samples.
UT: untreated
Figure 4.25 Repeat of LMNA and ITGA4 experiments using Lipofectamine® 3000 (L3K) transfection reagent. RT-PCR of RNA extracted from normal human fibroblasts treated with 2’-OMeAOs targeting LMNA A, B: 24h and C, D: 48h after transfection, and RT-PCR of RNA extracted from normal human fibroblasts treated with 2’-OMeAOs targeting ITGA4 transcripts after E, F: 24h and G, H: 48h transfection. A and C were amplified using LMNA primer pair (4970+4969). E and G were amplified using ITGA4 primer pair (6003+5597). B, D, F and H were amplified using SMN primer pair (5512a+5513a).
+ve: transfection positive control 2’-MeAO, SMN_8A(+57+81)
Figure B1 Proportions of ITGA4 transcripts with retained terminal intron in normal human fibroblasts transfected with ITGA4-targeting 2’-OMeAOs using Lipofectamine® 3000 transfection reagent. The proportions of terminal intron retaining ITGA4 transcripts present after 24 hours and 48 hours of transfection were compared. Except for one anomaly (12.5nM ITGA4_H28A(-6+19)), the proportion of ITGA4 transcripts
with retained terminal intron decreased with longer transfection duration.
B. Comparison of ITGA4 transcripts with retained terminal intron in L3K
transfections.
Appendix
124
C. Splicing Motif Analyses
Putative motifs within targeted sites of each gene transcripts that splicing factors bind to are identified
using the online software, SpliceAid 2 (http://193.206.120.249/splicing_tissue.html). Positions of each AO
designed to target each gene transcript examined in this project are shown. Both AOs that induced
terminal intron retention (underlined) and AOs that did not (not underlined), cover exonic splicing
enhancers such as SR proteins and Tra2β binding sites.
Table D1 Splice site scores of the terminal intron’s donor and acceptor splice sites. The scores were calculated using Human Splicing Finder, version 3.0 (http://www.umd.be/HSF3/HSF.html).
Gene Transcript
(Terminal) Intron
Splice site type
Motif New potential splice
site Consensus
value (0-100)
SMN* 7 Donor GGAgtaagt GGAgtaagt 82.81
Acceptor tctcatttgcagGA tctcatttgcagGA 91.9
LMNA^ 11 Donor CAGgtgagt CAGgtgagt 98.84
Acceptor tttctctcttagAG tttctctcttagAG 84.59
LMNC* 9 Donor GAAgtaagt GAAgtaagt 87.66
Acceptor tgtccccaccagGA tgtccccaccagGA 90.83
ITGA4* 27 Donor AAGgtaagc AAGgtaagc 96.87
Acceptor tgctattttcagGC tgctattttcagGC 90.45
SOD1 4 Donor GTGgtaagt GTGgtaagt 93.49
Acceptor aattttttacagGT aattttttacagGT 90.5
DMD 78 Donor GAGgttagt GAGgttagt 93.22
Acceptor tttgttttccagGA tttgttttccagGA 94.45 * terminal intron retention is evident
^ terminal intron retention is suspected
Appendix
132
E. ESE and ESS Densities of Terminal Intron
Table E1 Density of RESCUE-ESEs within the terminal intron of the 5 gene transcripts chosen for the
project. The number of RESCUE-ESE motifs was determined using the RESCUE-ESE web server
(http://genes.mit.edu/burgelab/rescue-ese/).
Gene Transcript
RESCUE-ESEs Terminal Intron Length Density
(no. of ESE motifs/nucleotide)
LMNA 5 322nt 0.0155
LMNC 25 421nt 0.0594
ITGA4 28 496nt 0.0565
SOD1 76 1095nt 0.0694
DMD 410 4711nt 0.0870
Table E2 Density of SELEX-ESEs within the terminal intron of the 5 gene transcripts chosen for the project. The number of SELEX-ESE motifs was determined using ESEfinder3.0 (http://rulai.cshl.edu/cgi-bin/tools/ESE3/esefinder.cgi?process=home).
Gene Transcript
SELEX-ESEs Terminal Intron Length
Density (no. of ESE
motifs/nucleotide) SF2/ASF SC35 SRp40 SRp55
LMNA 17 16 14 9 322nt 0.1739
LMNC 21 24 25 11 421nt 0.1924
ITGA4 8 13 18 10 496nt 0.0988
SOD1 16 25 34 18 1095nt 0.0849
DMD 100 123 168 101 4711nt 0.1044
Table E3 Density of Class 2 FAS-ESSs within the terminal intron of the 5 gene transcripts chosen for the
project. The number of FAS-ESS motifs was determined using the FAS-ESS web server