BIOCHEMICAL AND CELLULAR CHARACTERIZATION OF THE CODING
REGION DETERMINANT-BINDING PROTEIN (CRD-BP)-mRNA
INTERACTION

by

Kashif Mehmood
Pharm.D, Riphah International University Islamabad, Pakistan, 2010

THESIS SUBMITTED IN PARTIAL FULFILLMENT OF
THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
CHEMISTRY

UNIVERSITY OF NORTHERN BRITISH COLUMBIA
March 2014

© Kashif Mehmood, 2014

UMI Number: 1525704

All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.

Di!ss0?t&iori Publishing
UMI 1525704
Published by ProQuest LLC 2014. Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code.

ProQuest LLC
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106-1346

Abstract
RNA-binding proteins play critical role in the post-transcriptional processing o f mRNAs. One
such RNA binding protein is termed as the-Coding Region Determinant Binding Protein (CRDBP). CRD-BP is an onco-fetal protein whose overexpression has been reported in various types
of human cancers including, breast, colon, liver, skin, ovary, lung, brain, chorion, and testicular
cancers. CRD-BP is a member o f VICKZ family of RNA-binding proteins. In many instances,
RNA binding leads to stabilization o f the transcripts and an increase in their corresponding
protein levels; the result is manifest in downstream effects and the cancer phenotype.
The primary goal of this study was to obtain a better understanding o f the interaction between
CRD-BP and its three target mRNAs: G U I, MDR1 and CD44. Radiolabelled electrophoretic
mobility shift assay (EMSA) was performed with [32P]-labeled truncated GLI1 and MDR1 RNAs
to find smaller region of the transcripts which can still bind CRD-BP. It was found that G U I
320-380 RNA is the minimum region required for binding CRD-BP, while, MDR1 779-881
RNA is the minimum region which still has high affinity for CRD-BP.
Previous deletion studies of CRD-BP orthologs revealed that the KH domains, and not the
RRM domains, are critical for binding RNA substrates. However, it was unclear as to what
extent each KH domain plays nor is it known if different KH domains are important in binding
different RNAs. In this study, I used site-directed mutagenesis to mutate the GXXG to DXXG in
each of the KH domains as an approach to investigate the role o f each KH domains, in the
context o f the entire protein, in binding G U I and MDR1 RNAs. The Kj values of all the single
and double KH variants that are capable o f binding to G U I and CD44 RNAs was determined. In
general, it was found that single mutation in KH domains may or may not affect the binding
affinity o f transcript, while mutations at any two KH domains totally abrogated the binding of
1

RNA to CRD-BP,with the exception of KH3-4 which binds CD44 RNA but not G U I RNA. This
finding also supports the hypothesis that KH domains generally work in tandem. The results also
clearly showed that different RNAs bind CRD-BP differently in vitro.

The secondary goal of

my thesis was to design RNA oligonucleotides capable o f breaking the specific CRD-BP-GT/7
RNA interaction in vitro and in cells. For in vitro studies, competition studies using [ P] labelled
G L I230-420 RNA and MFOLD designed RNA/DNA oligonucleotides were utilized. Amongst
eight RNA oligonucleotides and one DNA oligonucleotide, SI RNA was the best competitor
against [32P] labelled GLI 230-420 RNA in vitro. The T47D human breast cancer cell and
HCT116 human colorectal cancer cell which expressed detectable level of G U I mRNA were
chosen for further studies with the SI RNA oligonucleotide. In both T47D and HCT116 cells
where CRD-BP-GI/7 mRNA interaction was demonstrated to exists, SI RNA oligonucleotide
significantly and specifically down-regulate GLI1 mRNA expression. The results obtained
support the model that CRD-BP protects G U I mRNA from degradation in T47D and HCT116
cells, and suggest that breaking CRD-BP-GL/i mRNA interaction is a feasible approach to
inhibit GLI1 expression.
In summary, this work shows that different mRNAs indeed bind to CRD-BP differently and it
is feasible to design/discover molecules capable of breaking specific CRD-BP-RNA interaction
in vitro. Most importantly, molecule that breaks CRD-BP-RNA interaction in vitro was also
capable o f down-regulating specific the mRNA in cells. This work has provided further evidence
to support the development o f a new class of anti-cancer drugs that act by breaking specific
protein-RNA interaction.

2

Table of Contents
Abstract................................................................................................................................................... 1
Table of Contents...................................................................................................................................3
List of Tables......................................................................................................................................... 6
List of Figures........................................................................................................................................ 7
Acknowledgements...............................................................................................................................9
Research Contributions and Academic/Research Awards................................................................10
List of Commonly Used Abbreviations............................................................................................. 11
C hapter 1- Introduction................................................................................................................... 12
1.1 Background Information............................................................................................................... 12
1.2 The regulation of gene expression............................................................................................... 12
1.3 The regulation of mRNA degradation........................................................................................ 13
1.3.1 The pathways of mRNA degradation.................................................................................14
1.3.2 Cis-elements (5’UTR, coding region, 3’UTR).................................................................. 15
1.3.3 Trans-acting factors (ribonucleases, miRNAs, RNA-binding proteins).......................... 16
1.4 RNA binding proteins................................................................................................................... 17
1.5 Coding Region Determinant Binding Protein (CRD-BP)...........................................................19
1.5.1 Structure and function........................................................................................................... 19
1.5.2 mRNA targets of CRD-BP and role in cancer...................................................................20
1.6 Three selected mRNA targets of CRD-BP................................................................................. 22
1.6.1 MDR1...................................................................................................................................22
1.6.2 GLI1......................................................................................................................................22
1.6.3 CD44.....................................................................................................................................24
1.7 Oligonucleotides and antibiotics/small molecules capable of breaking RNA-protein
interaction............................................................................................................................................ 24
1.8 Goal o f research............................................................................................................................ 25
C hapter 2- C haracterizing the physical interaction between CRD-BP and its two mRNA
targets, M D R l and G L I\..................................................................................................................27
P a rt A: Assessing CRD-BP-RNA interaction using fluorescently labelled MDRl
RNA......................................................................................................................................................27
1.2.1. Methodology.......................................................................................................................... 27
1.2.1.1. Generation and Purification o f the Coding Region Determinant-Binding
Protein............................................................................................................................................ 27
1.2.1.2. Transformation of competent E.coli BL21 cells........................................................... 27
1.2.1.3. Cell Growth and Induction.............................................................................................. 28
1.2.1.4. CRD-BP purification under denaturing conditions....................................................... 28
1.2.1.5. Protein Dialysis................................................................................................................ 29
1.2.1.6. Generation of DNA template for in-vitro transcription................................................29
1.2.1.7. Internal labeling of RNA oligonucleotide using Fluorescein-UTP.............................. 34
1.2.1.8. Fluorescent electrophoretic mobility shift assays......................................................... 34
1.2.2. Results.....................................................................................................................................34
I.2.2.I. Generation o f DNA templates...................................................................................... 34
3

1.2.2.2. Purification o f WT-CRD-BP....................................................................................... 36
1.2.2.3. Generation o f fluorescently labelled MDRl RNA........................................................ 38
1.2.2.4. Fluorescent EMSA.......................................................................................................... 38
P a rt B: Assessing CRD-BP-RNA interaction using [32P] radiolabelled MDRl and G U I RNA..39
11.2.1. Methodology......................................................................................................................... 39
11.2.1.1. Generation o f internally [32P] labeled RNA substrate............................................. 39
11.2.1.2. Radioactive electrophoretic mobility shift assay.........................................................40
11.2.2. Results....................................................................................................................................41
11.2.2.1. Mapping of GLI 1 RNA.................................................................................................. 41
11.2.2.2. Mapping of M DRl RNA................................................................................................45
P a rt C: Breaking CRD-BP-GL/i RNA interaction.........................................................................46
111.2.1. Methodology....................................................................................................................... 46
111.2.1.1. Designing RNA and DNA oligonucleotides using MFOLD...................................46
111.2.1.2. Generation of [32P]-labelled GLI 230-420................................................................. 49
111.2.1.3. Generation o f unlabelled RNA.................................................................................... 49
111.2.1.4. Competition assay with oligonucleotides................................................................... 49
111.2.2. Results..................................................................................................................................50
111.2.2.1. Optimization of competition assay.............................................................................. 50
111.2.2.2. Competition assay with RNA and DNA oligonucleotides........................................ 51
2.3 Discussion...................................................................................................................................52
C hapter 3- Assessing CRD-BP and its KH variants for ability to bind G LU and CD44
RNAs....................................................................................................................................................55
3.1 Methodology.............................................................................................................................. 56
3.1.1 Purification and dialysis of WT-CRD-BP and its variants...............................................56
3.1.2 Generation o f [32P] -labelled CD44 and GLI1 RNAs.......................................................56
3.1.3 Radiolabelled EMSA using [32P] labeled CD44 and GLI1 RNAs................................... 57
3.2 Results......................................................................................................................................... 57
3.3 Discussion...................................................................................................................................62
C hapter 4- Investigating the regulation of G LU mRNA expression by CRD-BP in cancer
cells.......................................................................................................................................................65
4.1 Methodology............................................................................................................................ 65
4.1.1 Detection o f G U I mRNA levels in cell lines................................................................... 65
4.1.1.1 Plating of cells in 6-well plates.................................................................................... 65
4.1.1.2 Total RNA extraction.................................................................................................... 65
4.1.1.3 DNase treatment............................................................................................................ 66
4.1.1.4 cDNA synthesis.............................................................................................................66
4.1.1.5 RT-qPCR.......................................................................................................................67
4.1.2 Knockdown CRD-BP in HCT116 and T47D....................................................................67
4.1.2.1 Transfecting cells with dsiRNA and performing RT-qPCR of G U I........................67
4.1.2.2 RT-qPCR of CRD-BP................................................................................................... 67
4.1.2.3 RT-qPCR of G U I and p -a c tin ....................................................................................68
4

4.1.3 Overexpressing CRD-BP in HCT116 and T47D cells.................................................... 69
4.1.3.1 Transfecting cells with pcDNA-FLAG-CRD-BP and pcDNA-FLAG- vector
69
4.1.4 Effect o f modified RNA oligonucleotides on G U I mRNA level in cells..................... 70
4.1.4.1 Transfecting Cells with modified RNA oligonucleotides.........................................70
4.2 Results........................................................................................................................................70
4.2.1 Primer optimization............................................................................................................ 70
4.2.2 Detection of G U I mRNA level in cell lines.................................................................... 71
4.2.3 Knockdown CRD-BP using dsiRNA................................................................................ 72
4.2.4 Overexpressing CRD-BP in HCT116 and T47D cell lines.............................................74
4.2.5 Effect o f 2’ O-methyl SI and S4 RNA Oligonucleotides on G U I mRNA levels
in cells..................................................................................................................................................76
4.3 Discussion..................................................................................................................................78
C hapter 5-General Discussion........................................................................................................81
5.1 General discussion.................................................................................................................... 81
5.2 Assessing CRD-BP-RNA interaction using fluorescent EMSA........................................... 82
5.3 Assessing GLIJ and M DRl RNA-CRD-BP interaction using [32P]-radiolabeled
EMSA...............................................................................................................................................83
5.4 Breaking GLI1 RNA-CRD-BP interaction in vitro................................................................84
5.5 Comparing the KH variants o f CRD-BP for binding to CD44 and GL11 RNAs................86
5.6 Brealdng o f CRD-BP-GLU RNA interaction in cells........................................................... 89
5.7 Future Direction........................................................................................................................ 91
5.8 Concluding Remarks................................................................................................................ 93
References.............................................................................................................................................. i
Appendix............................................................................................................................................ viii

5

List of Tables

Table 1. Primer sequences for amplifying fragments of GLI1 cDNA............................................30
Table 2. Reagent used in PCR to synthesize DNA template.......................................................... 31
Table 3. Primer sequences for amplifying fragments o f M DRl cDNA..........................................33
Table 4. Reagents used in generation of [32P] labelled RNA........................................................40
Table 5. Reagents used for fluorescent and radioactive EMSA......................................................41
Table 6. Sequences o f the designed DNA and RNA oligonucleotides.......................................... 47
Table 7. Reagents used in synthesis of unlabelled RNA................................................................. 49
Table 8. Position of mutation in CRD-BP structural domain......................................................... 55
Table 9. Summary of the IQ values o f CRD-BP and its variants for binding to CD44 and G U I
RNAs....................................................................................................................................................62
Table 10. Primer and probe sequences of CRD-BP and (1 -actin used in qRT-PCR Taqman
approach............................................................................................................................................... 68
Table 11. Primer sequences o f GL11 and /? -actin used in SYBR Green qRT-PCR method

69

Table 12. Ct values o f seven cancer cell lines.................................................................................. 72

6

List o f Figures

Figure 1. Scheme of Central Dogma of Molecular Biology.................................................................... 12
Figure 2. mRNA decay pathways......................................................................................................14
Figure 3. A variety of sequence motifs within mRNAs that are bound by RNA-binding
proteins..................................................................................................................................................18
Figure 4. CRD-BP functional domain locations.............................................................................. 19
Fig 5. Schematic representation o f M P -1 action in stabilizing messenger RNAs important in
cancer....................................................................................................................................................21
Fig 6. Hh signaling pathway...............................................................................................................23
Fig 7. Schematic diagram of truncated GUI cDNA template.................................................................32
Fig 8. Schematic diagram of truncated MDRl cDNA template...............................................................33
Fig 9. Agarose gel analysis of PCR generated GLI1 cDNA templates.................................................... 35
Fig 10. Agarose gel analysis of PCR generated MDRl cDNA templates................................................ 36
Fig 11. SDS-PAGE analysis of recombinant WT-CRD-BP.....................................................................37
Fig 12. Analysis of in-vitro transcription product................................................................................... 38
Fig 13. Fluorescent EMSA of MDRl nts 739-921................................................................................. 39
Fig 14. Mapping the GUI RNA nts 36-990 for ability to bind CRD-BP................................................ 42
Fig 15. Mapping the 3’ truncated GL11 RNA nts 230-420 for ability to bind CRD-BP........................... 42
Fig 16. Mapping the 5’ truncated GUI RNA nts 230-420 for ability to bind CRD-BP........................... 43
Fig 17. Assessing GUI RNA nts 320-380 nts for ability to bind CRD-BP..............................................44
Fig 18. MFOLD (Zuker, 2003) generated secondary structures of GLI 320-380 RNA and CD44 28622930 RNA..............................................................................................................................................44
Fig 19. Mapping of MDRl RNA for ability to bind CRD-BP..................................................................45
Fig 20. MFOLD generated predicted secondary structure of GUI RNA nts 346-382....................... 47
Fig 21. The predicted secondary structures of DNA oligonucleotide and RNA oligonucleotides (S1-S8)
as generated by MFOLD.........................................................................................................................48
Fig 22. Optimization of competition assay..............................................................................................50
Fig 23. Breaking GLI 230-420-CRD-BP interaction using specific oligonucleotides.............................. 51
Fig 24. Schematic diagrams of CRD-BP.................................................................................................56
Fig 25. EMSA in assessing the binding profile of the WT-CRD-BP and its variants on CD44 RNA

58

Fig 26. EMSA in assessing the WT-CRD-BP and its variants for binding to GUI RNA........................ 59
Fig 27. Saturation binding of CRD-BP and its variants on CD44 RNA.................................................. 60
7

Fig 28. Saturation binding of CRD-BP and its variants on GLI1 RNA................................................... 61
Fig 29. Melt curve analysis of GLU primers...........................................................................................71
Fig 30. Effect of knocking down CRD-BP on GLI1 mRNA level in HCT116 cells................................ 72
Fig 31. Effect of knocking down CRD-BP on GLI1 mRNA level in T47D cells.....................................73
Fig 32. The effect of over-expressing CRD-BP on GUI mRNA level in HCT116 cells......................... 74
Fig 33. Overexpressing CRD-BP in T47D cells and checking CRD-BP and GLI1 mRNA levels by doing
RT-qPCR................................................................................................................................................ 75
Fig 34. Effect of SI and S4 2’ O-methyl RNA Oligonucleotides on GLI1 mRNA level in T47D cells.. ..76
Fig 35. Effect of SI and S4 2’ O-methyl RNA Oligonucletides on GUI mRNA level in HCT116 cells..77
Fig 36. IMP1 KH3-4 pseudodimer configuration of reveals P-actin RNA-binding surfaces.....................88

8

Acknowledgements
I would like to thank all those individuals who have supported, encouraged and assisted me
throughout my M.Sc. degree: Dr. Chow Lee, Dr Kerry Reimer and Dr. Dezene Huber. I would
also like to thank Maggie Li for her useful suggestions. I would like to thank Mr. Daud Akhtar
for assisting me in cells related experiments. I would also like to thank all current and past
members of the Lee lab, especially CRD-BP team (Mr. Mark Barnes, Mr. Gerrit van Rensburg
and Mr. Sebastian Mackedenski). I am also thankful to the Gorrel, Rader and Gray labs for their
continuous support. I would also like to thank my family and friends who supported me during
my studies. At the end, I would like to thank Prince George Muslim community and Mr. John
Tang and his family who made my stay wonderful in Prince George.

9

Research Contributions and Academic/Research Awards
Publication
Barnes M, van Rensburg G, Li WM, Mehmood K, Mackedenski S, King DT, Miller AL, Lee
CH. (2014) Molecular insights into the CRD-BP-RNA interaction through site-directed
mutagenesis at the GXXG motif in KH domains. (Submitted)
Abstracts
Mehmook K. and Lee C. Breaking the physical interaction between CRD-BP and its RNA
targets GLU and M DRl. (11th Biennial Conference PSBMB, Lahore, Pakistan. November 2013).
Mehmood K. and Lee C. Investigating the Interaction between CRD-BP and its two RNA
targets, GLI1 and MDRl RNAs. (9th Annual Ribowest conference,UNBC, Canada. May 2013).
Akhtar D. Mehmood K. and Lee C. The role of KH domains of CRD-BP in the interaction with
GLI1 mRNA. (9th Annual Ribowest conference UNBC, Canada. May 2013)
Mehmood K. and Lee C. Understanding an Important molecular choreography (8th Annual
UNBC graduate conference, UNBC, Canada. February 2013).
Mehmood K, Bames M and Lee CH. Development of fluorescent-based method to study the
CRD-BP-RNA interaction. (8th Annual Ribowest conference, University o f Lethbridge,
Lethbridge. June 2012)

Awards
UNBC Research Project Award May 2013-March 2014
NBCGSS Legacy Scholarship 2013-2014
Graduate Teaching Award 2012-2013

10

List o f Commonly Used Abbreviations

CRD-BP

Coding Region Determinant Binding Protein

RNA

Ribonucleic Acid

DNA

Deoxyribonucleic Acid

MDR

Multi Drug Resistance

EMSA

Electro Mobility Shift Assay

KH

K-Homology

RRM

RNA Recognition Motif

K<j

Dissociation constant

UTR

Un-translated Region

miRNA

Micro RNA

qPCR

quantitative Polymerase Chain Reaction

11

Chapter 1. Introduction
Chapter 1

1.1 Background Information
According to the central dogma of molecular biology, genetic material flows from DNA to
mRNA to protein (Figure 1). Numerous steps are involved to go from the DNA molecule to a
functional protein within the cell. At the DNA level, many factors contribute to the transcription
of a gene. Transcription factors bind directly to certain regions on the DNA molecule and
regulate proteins that carry out the transcription itself. Gene expression can be controlled at
many steps via multiple pathways. Functional proteins play a key role in cellular functions, and
these include the control of gene expression.

Nucleus
Cytoplasm
ABflBBBflftaifr- 3
_ mRNA
Translation
Growing'
protein''

Figure 1. Scheme of Central Dogma of Molecular Biology (Weiss and Buchanan, 2009)
1.2 The regulation of gene expression
Expression of a gene can be controlled at different levels, including transcription, mRNA

12

Chapter 1. Introduction
stability, translation of RNA to protein, pre-mRNA splicing and post-translational events such as
protein stability and modification (Day and Tuite, 1998). As my research is related to mature
mRNAs, I will mainly focus on the post-transcriptional regulation of gene expression.

Post-transcriptional modifications to mRNA transcripts are essential to the generation of a
functional mRNA transcript. The majority of eukaryotic mRNAs carry a 5’ 7-methylguanosine
cap structure and a 3’ poly (A) tail comprising of about 200 adenosine residues in length, which
protect the RNA chain from degradation by exonucleases. In addition to the cap and poly(A) tail
with its associated poly(A)-binding protein (PABP), some specific internal sequence elements
also influence intrinsic stability of mRNA (Day and Tuite, 1998). Such determinants can exert a
destabilising effect on an mRNA and their location in transcripts is arbitrary. For instance, such
determinants have been found in the 5’ UTR (Pierrat et al, 1993), in the 3’ UTR (Muhlrad et al,
1992), and in the coding region (Herrick et al, 1992) of different mRNAs.

Another important mRNA processing step that controls gene expression is alternative splicing
of pre-mRNAs. Alternative splicing is a fundamental mechanism found in many organisms
including humans. Expression of proteins in humans is regulated by this important step. Upto
95% of pre-mRNAs undergo splicing to produce mature mRNAs (Katzenberger et al, 2009).
The splicing patterns depend on the assembly of the spliceosome complex on the introns which
are to be removed from pre-mRNA (Katzenberger et al, 2009). Five small nuclear
ribonucleoproteins (snRNPs, U l, U2, U4, U5 and U6) along with multiple regulatory proteins
carry out trans-esterification reactions that are responsible for the splicing (Katzenberger et al,
2009). Such regulation o f gene expression at the level of pre-mRNA can be complex; however, it
is necessary for the expression of many variant proteins in cells.
1.3 The regulation of mRNA degradation

13

Chapter 1. Introduction
The mRNA decay pathways will be discussed in this section. Gene expression is regulated at
many Check points starting from DNA level and ending at protein level. In order to understand
geneexpression, one must understand the cellular mechanisms that control the stability and
turnover rate of mRNA (Anderson and Parker 1998). The degradation of mRNA transcripts
depends on mRNA decay pathway as well as stability of the mRNA molecule (Farina et al
2008). As mentioned earlier, the addition of the 5’ methyl-guanasine cap and the 3’ poly (A) tail
consisting of multiple adenosine monophosphates are post-transcriptional modifications that can
protect the transcript from nucleolytic attack by RNA degradative enzymes at the 5’ and 3’ ends
of mRNA. However, the 5’ cap and 3’ poly A tail, and more recently, internal sequences are also
important when it comes to the mRNA decay pathways.
1.3.1 The pathways of mRNA degradation
There are three well-described mRNA decay pathways, including the deadenylationdependent pathway, decapping pathway and endonucleolytic pathway (Figure 2).

5 'UTR

O Rf

rU T R

-AAAA
■COK-NOT

or PARN

Lsml-7

Cndonuchnw
RNwrMRR)

Figure 2. mRNA decay pathways (Gameau et al, 2008)
The current dogma suggests the 5’ decapping and the 3’ deadenylation-dependent pathways as
the major pathways for the degradation of mRNA molecules. The endonucleolytic pathway has
14

Chapter 1. Introduction
not been studied much, but it does seem to play a role in controlling of gene expression (Li et al.
2010).
1.3.2 Cis-elements (5’UTR, coding region, 3’UTR)
Cis-regulatory elements are RNA sequences within a transcript that regulate expression of
the corresponding transcript (Alberts et al, 2008). Although the role o f the 5’ UTRs
on mRNA stability is less clear, however in some cases this region can significantly affect
specific mRNA half-lives. In theory, half-life of an mRNA is affected due to the influence o f 5’
UTR on its translational efficiency (Ross, 1995). In addition, the length of the 5’ UTR also
affects half-life of RNA as in the case of c-myc mRNA. This effect was shown to be independent
of translation (Bonnieu et al, 1988).
c-myc mRNA contains a coding region instability determinant, or CRD, involved in message
instability. However, it confers stability to c-myc mRNA upon binding to CRD-binding protein
(CRD-BP). Regulatory RNA sequences present in the coding region are also found in some other
transcripts such as c- fos and IL-2 (Knapinska et al, 2005). The c-myc CRD contains a higher
than usual percentage of rare codons, that are responsible for ribosome pausing. This pausing
leads to creation of a ribosome-free mRNA segment between the pause site and translation
termination site. It is postulated that the exposed segments can be degraded by endoribonuclease
during the translational pausing. The CRD’s stabilizing effect is dependent on translation of the
CRD segment, and presence of stop codon just upstream of CRD can lead to increased mRNA
decay even in the presence of CRD-BP (Lemm and Ross, 2002).
A+U rich element (ARE) is the best characterized family of instability determinants found in
the 3’ UTR. These elements were first discovered as mediators of instability in transcripts coding
for inflammatory responses proteins (Caput et al, 1986; Shaw and Kamen, 1986). AREs are now
known to regulate a variety of transcripts via mRNA degradation and/or translational repression.
15

Chapter 1. Introduction
Deadenylating enzymes, the exosome, and the proteasome are involved in the degradation of
ARE-containing mRNAs (Laroia et al, 1999; Chen et al, 2001). AREs are responsible for
controlling the stability of many transcripts such as c-fos, c-myc, c-jun, COX-2, MMP-13,
cyclins, Cdk inhibitors, DNA methyltransferase I, IL-lb, IL-3, TNFa, and many others (Audic
and Hartley, 2004; Wilson et al, 2003).
Iron response element (IRE) is another example of a ds-regulatory element. IRE sequences are
found in the 5’ or 3’ ends o f mRNAs and after binding with iron regulatory proteins, they alter
gene expression (Nelson and Cox, 2008).

1.3.3 Trans-acting factors (ribonucleases, miRNAs, RNA-binding proteins)
Ribonucleases are enzymes having the ability to catalyze the degradation o f RNA molecules
in an exoribonucleolytic and endoribonucleolytic manner. There are numerous ribonucleases and
their classification is based on the mechanism employed to break RNA molecule.
Exoribonucleases are groups of ribonucleases that degrade the RNA molecules externally from
either the 5’ or 3’ end (Farina et al, 2008). Conversely, endoribonucleases are groups of
ribonucleases that catalyze the degradation of RNAs by attacking the molecule internally.
Ribonucleases play an essential role in the degradation of mRNA transcripts and therefore the
expression of genes.
miRNAs are small non-coding RNAs of approximately 22 nucleotides and are considered an
important class of gene regulators. They are abundant in animals and plants, and after their genes
are transcribed and processed like other premature RNAs, mature miRNAs bind target sites
present on the mRNAs at the 3’UTR as well as in the coding region, leading to translational
repression or degradation of mRNAs (Goswami et al, 2010). miRNAs can take part in acquired
therapy resistance in cancer cells. One class of miRNAs acts by targeting survival pathways or
pathways involved in apoptosis sensitivity (Boyerinas et al, 2010), whereas the second class
16

Chapter 1. Introduction
modulates therapy by targeting those genes that are involved in metabolism or transport of drugs.
Examples are: miR-200c that targets TUBB3 (Cochrane et al, 2009); miR-451 and miR-27a that
target MDRl (Zhu et al, 2008); miR-24 that targets DHFR (Mishra et al, 2007); and miR 328
and miR-519c that target ABCG2 (Pan et al, 2009; To et al, 2008).

Let-7 is a family of thirteen related miRNAs that are located on nine different chromosomes in
humans. Let-7 is considered as tumor suppressor and its down-regulation leads to human
cancer (Boyerinas et al, 2010).
As mentioned above, the post-transcriptional modifications of transcripts at the 5’ and 3’ end
protects mRNAs from nucleolytic attack. Another important factor for mRNA protection is the
RNA-binding proteins. RNA-binding proteins bind to mRNAs and protect them from nucleases
(Mullen and MarzlufF, 2008). Although much work has been done on the regulation of gene
transcription and expression, much less is known about the targets and roles of RNA-binding
proteins (Hogan et al, 2008).
1.4 RNA binding proteins
Recent studies have shown the involvement of RNA binding proteins in the posttranscriptional processing of mRNAs. These proteins recognize specific sites of newly
transcribed transcripts and their binding occurs at specific times during processing of mRNA
transcript (Hogan et al, 2008). Hogan et al (2008) showed that a variety of motifs through which
RNA binding proteins recognize and bind to (Figure 3). Trends found in these motifs may assist
in the identification of new RNA motifs for other RNA-binding proteins.

17

Chapter 1. Introduction

Nam*
PUF3-1*
PUF4-1*
PUF5-1*
PUB1-1*
PUF2-1
SSD1-1
PAB1-1
KHD1-1
NAB2-1*
NSR1-1
YLL032C-1
VTS1-1
PIN4-1
NRD1-1*

Motif
U GUAM XiUA

mummm
mmmmt
HkhUtU
AnacAuueeju
gilAUAUAM
* X * C A U U k fc
M fA fA fG fM
AAliACCS
J s a C G SG G *
MuMAAUGA
M ucuucufi

Figure 3. A variety of sequence motifs within mRNAs that are bound by RNA-binding proteins
(Hogan et al, 2008).
RNA-binding proteins seem to have diverse roles in the cell, particularly in protecting the
mRNAs from enzymatic attack. This has been observed in various diseases including diabetic
retinopathy, coronary artery disease and tumor angiogenesis in cancers (Levy et al, 1998).
Tumor angiogenesis is known to be essential for tumor growth and metastasis, and vascular
endothelial growth factor (VEGF) is the most studied angiogenic factor known to be crucial for
tumor growth (Levy et al, 1998). VEGF mRNA is post-transcriptionally regulated, namely by a
specific RNA-binding protein.
HuR binds to the destabilising element at the 3’ UTR of VEGF mRNA under hypoxic
conditions (Levy et al, 1998). This is an example which shows that RNA-binding proteins can be
implicated in diseases such as cancer. My work focuses on the RNA-binding protein called the
Coding Region Determinant-Binding Protein (CRD-BP) which is known to bind to a number of
different mRNA targets implicated in cancers.

18

Chapter 1. Introduction
1.5 Coding Region Determinant Binding Protein (CRD-BP)

CRD-BP belongs to VICKZ (Vgl, IMP1, 2, 3, CRD-BP, KOC, ZBP-1) family of RNA
binding proteins. These proteins are highly conserved in vertebrates and have been shown to play
critical role in the post-transcriptional control of gene expression, namely in mRNA stability,
translation and localization (Yisraeli, 2005; Bell et al 2013).

1.5.1 Structure and function
The mouse coding region determinant binding protein (CRD-BP), composed of 577 amino
acid and having four K-homology (KH) domains and two RNA recognition motifs (RRMs)
(Figure 4) (Bell et al, 2013). CRD-BP was first discovered based on its high affinity to the
coding region determinant of c-myc mRNA (Doyle et al, 1998). CRD-BP gene comprises of
almost 40 Kb in length and contains 15 exons and 14 introns. In mice, it is located on
chromosome 11 while in humans it is located on chromosome 17 (Doyle et al, 2000).

Figure 4. CRD-BP functional domain locations (Chao et al, 2010).
RRM consisting of two short consensus sequences RNP1 (octamer) and RNP 2 (hexamer), is
the most widespread and well characterized RNA-binding domain. Short consensus of RRM is
composed of about 80 amino acids, with a PafiPafi topology. RNP1 and RNP 2 are juxtaposed
in the middle two strands of four stranded antiparallel /? sheet surface and are responsible for
RNA recognition (Varani and Nagai, 1998). The KH (hnRNP K homology) motif is almost 70
amino acids in length having fiaafipa topology (Adinolfi et al, 1999; Grishin 2001). The KH
consensus sequence comes in direct contact with nucleic acids (Lewis et al 2000).

19

Chapter 1. Introduction
ZBP1, the chicken ortholog of CRD-BP, has been shown to mainly reside in the cytoplasm
and makes complexes with its mRNA target (Farina et al, 2003; Stohr et al, 2006; Huttelmaier et
al, 2005). Earlier reports suggested a zipcode recognition sequence (ACACCC) which was found
within the fi-actin mRNA where ZBP1 binds to (Farina et al, 2003). However, it is now clear that
CRD-BP binding occurs with other RNA targets without the zipcode, suggesting that there is no
specific RNA recognition sequence. Recently, crystallization studies of the third and fourth KH
didomain of CRD-BP showed dimerization of the two KH domains with canonical RNA-binding
surfaces exposed at opposite ends of the molecule (Chao et al, 2010).

1.5.2 mRNA targets of CRD-BP and role in cancer
CRD-BP, also known as insulin-like growth factor-II mRNA binding protein 1 (IGF2BP1 or
IMP-1) and zipcode-binding protein 1 (ZBP1), was first described in Dr. Jeffrey Ross’ lab for its
ability to bind the c-myc mRNA coding region determinant (CRD) and shield the transcript from
degradation (Prokipcak et al, 1994). CRD-BP is absent or present in extremely low levels in
normal adult tissues (Leeds et al, 1997). However, it is abundantly expressed in fetal tissues and
in several types of human cancers, including breast, colon, skin, ovary, lung, brain, testicular,
and chorion cancers (Doyle et al, 2000; Ioannidis et al, 2003; Dimitriadis et al, 2007; Ross et al,
2001; Kobel et al, 2007; Kato et al, 2007; Hammer et al, 2005; Ioannidis et al, 2005; Elcheva et
al, 2008; Ioannidis et al, 2004; Mongroo et al, 2011; Hsieh et al, 2013). Conversely, Gu et al
(2009) reported overexpression of ZBP1 during embryogenesis and tumor formation but its
repression in metastatic breast cancer cell lines and tumors.
In several types of cancers, CRD-BP expression correlated with poor disease outcome and
reduced overall survival (Dimitriadis et al, 2007; Kato et al, 2007; Kobel et al, 2007; Gu, 2004).
CRD-BP is involved in the control of cell proliferation (Ioannidis et al, 2005; Noubissi et al,
2006; Weidensdorfer et al, 2009) as well as cell adhesion and invadopodia formation, suggesting
20

Chapter 1. Introduction
its role in increasing the metastatic potential of tumour cells (Vikesaa et al, 2006). In human
colorectal cancer cells, over-expression of CRD-BP stabilizes fiTrCPl mRNA and elevates
pTrCPl levels, resulting in a phenotype favoring tumor development (Noubissi et al, 2006).
These findings suggest an oncogenic role o f CRD-BP.

CRD-BP binds to a number of mRNAs implicated in cancer. These include mRNAs for c-myc,
CD44, /STrCPl, IGF-II, MITF, K-ras, GUI, and MDRl (Ioannidis et al, 2005; Sparanese and
Lee, 2007;Vikesaa et al, 2006; Noubissi et al, 2006; Gu et al, 2008; Leeds et al, 1997; Mongroo
et al, 2011; Noubissi et al, 2009; Goswami et al, 2010; Boyerinas et al, 2011). In many cases,

prWat-7
miRNA

mRNA Stabilization

I

Cancer CaM
Proliferation

nt

Multidnig
— ►MDR1--------*>
► Realatanca
Raaiatanca
(P-Glycoprotain)
(Taxanee)
p-CatonirVTcf
p-TICpI
p-TrCp1
(Ub-llg)

|IK -B a

Inflammation,
Cefl Growth

Fig 5. Schematic representation of IMP-1 action in stabilizing messenger RNAs important in
cancer (Lily et al, 2013).
RNA binding leads to stabilization of the transcripts and an increase in their corresponding
protein levels; the result is manifested in downstream effects and cancer phenotype (Vikesaa et
al, 2006; Noubissi et al, 2006; Noubissi et al, 2009; Boyerinas et al, 2011).

21

Chapter 1. Introduction
1.6 Three selected mRNA targets of CRD-BP
As described above that CRD-BP binds to a number of different mRNAs. In this work, I will
mainly focus on the following three mRNAs: MDRl, G U I and CD44 mRNAs

1.6.1 MDRl
Chemotherapy is one of the most frequently used treatments for human cancer. In ovarian
cancer, surgical de-bulking followed by a combination therapy involving paclitaxel, carboplatin
is the first line of treatment. Its response rate is up to 80% (Yap et al, 2009). The majority of
patients relapse within 18 months of treatment and become more resistant to chemotherapy.
Various mechanisms lead to chemotherapy resistance, including up-regulation of members of
adenosine triphosphate binding cassette transporters (ABC transporter family) that pump drug
across cell membrane to extracellular space. A very common member of this family is the Multi­
drug Resistance 1 or MDRl (Ambudkar et al, 2003) which encodes for P-glycoprotein. MDRl
substrates include toxins and commonly used therapeutic agents such as Taxanes and
anthracyclines (Zhou, 2008).
As discussed earlier, CRD-BP was initially discovered due to its ability to bind to c-myc
CRD RNA. Later on, it was shown that CRD-BP has similar affinity for MDRl RNA having Kd
value of about 500 nM (Sparanese and Lee, 2007). CRD-BP binds to MDRl RNA nts 746-962
and is capable of blocking MDRl RNA cleavage by a specific mammalian endoribonuclease in a
concentration dependent manner (Sparanese and Lee, 2007).
1.6.2 GLI1
Wnt and hedgehog (Hh) are critical pathways involved in embryonic development, stem cell
maintenance and tumorigenesis. Hh pathway is controlled by the Ci/GLI family of zinc finger
transcription factors.
22

Chapter 1. Introduction

Primary
cilium /

Primary
cilium /

PTCH

SUFU

Fig 6. Hh signaling pathway (a) Hh pathway inactive form: In the absence of Hh ligand,
PTCH localizes in the cilia and represses SMO activity by preventing its trafficking and
localization to the cilia. Translation of Hh targeted genes are inhibited (b) Hh pathway active
form: On ligand binding, PTCH activates SMO. Activation of SMO results in translocation of an
activated form of GLI (GLIA) to the nucleus, where it induces expression of Hh target genes.
(Dereck et al; 2013)
In vertebrates, three different genes of GLI have been reported. Amongst them, GLI1 is a
transcriptional activator, while GLI2 and GLI3 are both considered as activators and repressors
(Wicking, 1999). Wnt//? catenin induces the expression of CRD-BP (Noubissi et al, 2006).
Overexpression of CRD-BP in turn leads to drastic increase in the half-life of GLI mRNA. As an
RNA-binding protein, it was postulated that CRD-BP binds to and stabilizes GLI1 mRNA.
Indeed, CRD-BP was shown to have high affinity for GLU mRNA at nucleotides 41-990.
(Noubissi et al, 2009). Expression and transcriptional activity of GLI1 by Wnt//? catenin
signalling depends on CRD-BP, so it is thought that GLF s contribution to colorectal cancer
formation is mainly due to its regulation by Wnt//? catenin via mRNA stability controlled by the
physical association with CRD-BP (Noubissi et al, 2009).

23

Chapter 1. Introduction
1.6.3 CD44
CD44 is a member of a large family of cell adhesion molecules responsible for adhesion
between adjacent cells. In addition to its role in cellular adhesion, CD44 plays an important role
in growth and motility o f cells, and thus it is involved in many types of cancers, including breast,
lung, prostate, ovarian, cervical, and colorectal cancers (Naor et al, 2002).

It has been reported that CRD-BP binds to the 3'-UTR of CD44 transcript and increases CD44
mRNA stability. (Kobel et al. 2007; Noubissi et al. 2006; Vikesaa et al. 2006), leading to
cytoplasmic spreading and invadopodia formation (Vikessa et al, 2006).
1.7 Oligonucleotides and antibiotics/small molecules capable of breaking RNA-protein
interaction
The expression of genes often depends on formation of complexes between proteins and
RNAs. As a result, gene expression can be prevented by targeting this complex. Inhibiting the
formation of complex between protein and RNA can be a very useful approach in inhibiting gene
expression. While many inhibition of gene expression approaches target RNA, only a few target
protein.
RNA interference (RNAi) is a ubiquitous mechanism that reduces gene expression within
cells. In the RNAi mechanism, single-stranded small interfering RNAs (siRNAs), generated from
double-stranded RNAs, bind to complementary sequences on target mRNA transcripts and
recruit RISC (RNA-induced silencing complex) to degrade the corresponding target mRNAs
(Dykxhoom, 2006). This is an extremely useful mechanism when it comes to inhibiting gene
expression in cells.
Antisense oligodeoxynucleotide (ODN) is another class of nucleic acid-based therapeutic agent
that act via the activity of RNase H (Crooke, 1993). Interestingly, sequence specific ODNs have

24

Chapter 1. Introduction
also been explored for use to occlude binding of CRD-BP to c-myc mRNA as an approach to
inhibit c-myc gene expression (Coulis et al, 2000).

Aminoglycoside antibiotics, such as neomycin B, is capable of breaking protein-RNA
Interactions (Zapp et al, 1993). During early HIV infection, the small accessory protein Rev
binds to the Rev Response Element (RRE) of mRNAs, resulting in mRNA protection and
localization into cytoplasm where it is translated (Van Ryk et al, 1999). Neomycin inhibits the
binding of Rev to the RRE, thus preventing mRNA localization (Van Ryk and Venkatesan,
1999).
A recent attempt to use small molecules as specific inhibitors of IMP 1-c-myc interaction has
been successful. Small molecules were screened in-vitro using high throughput screening assay.
One small molecule was successfully identified against IMP-1 positive IGROV-1 ovarian cancer
cell line for its ability to inhibit growth. (Lily et al, 2013).
Previous researchers in Dr. Lee’s lab have successfully identified specific oligonucleotides
that disrupt the CRD-BP-CD44 RNA interactions in vitro and specifically destabilize CD44
mRNA in cells (King et al, 2014).
1.8 Goal of research
The general goals of this MSc thesis were two-fold.

Firstly, to develop a better basic

understanding of the molecular interaction between CRD-BP and its three targets, CD44, MDR1
and G U I mRNAs. Secondly, with the scientific knowledge generated, to design, develop and
test specific oligonucleotides for their ability to inhibit specific gene expression through the
mechanism of blocking specific CRD-BP-mRNA interaction.
To achieve the general goals described above, the specific goals of my thesis were four-fold:
(i) to develop a fluorescence-based method in an effort to rapidly study and characterize the
CRD-BP-RNA interaction; (ii) to use electrophoretic mobility shift assays (EMSAs) to study
25

Chapter 1. Introduction
CRD-BP-GL/7 and -MDR1 RNA interaction; (iii) to use electrophoretic mobility shift assays to
study and to compare the WT CRD-BP and its various KH variants for binding to G U I and
CD44 RNAs; and (iv) to assess specific RNA oligonucleotides for their ability to disrupt the
CRD-BP-GL/7 RNA interaction in vitro and destabilize G U I mRNA in cells.
To achieve the first specific goal, I first aimed to develop the fluorescence EMSA which would
allow me to quantify CRD-BP-RNA interaction. To achieve the second specific goal, I used
EMSA to map the smallest RNA regions of GLI1 and MDR1 which bind CRD-BP. To
accomplish the third specific goal, I used EMSA to compare and contrast the wild-type CRD-BP
with a panel of KH variants of CRD-BP for their ability to bind GLI1 and CD44 RNAs. Finally,
to accomplish the fourth specific goal, I used EMSA to assess specific RNA oligonucleotides for
their ability to compete with G U I RNA for binding to CRD-BP. The 2’-0-methyl version of the
oligonucleotides were then assessed for their ability to specifically destabilize GLll mRNA in
cells as determined using quantitative real-time PCR.

26

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and G UI
Chapter 2
Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GUI
Generation and purification of the recombinant wild type CRD-BP (WT-CRD-BP) and its
binding ability with truncated RNA fragments of G U I and MDRl RNAs in-vitro, using
electrophoretic mobility shift assay (EMSA) will be discussed in this chapter. Designing of
specific oligonucleotides and their ability to compete with G U I RNA for binding to the WTCRD-BP will also be discussed in this chapter.
One of the main goals was to map the smallest possible regions of these two RNAs which are
still capable of binding to the WT-CRD-BP. RNA or DNA oligonucleotides that could break
G U I RNA-CRD-BP interaction was designed. This chapter is divided into the following
three parts.
(A) Assessing CRD-BP-RNA interaction using fluorescently labelled MDRl
RNA
(B) Assessing CRD-BP-RNA interaction using [32P] radio labelled MDRl and GUI
RNAs
(C) Breaking CRD-BP-G U I RNA interaction

Part A: Assessing CRD-BP-RNA interaction using fluorescently labelled MDRl
RNA

In this section, I explored the possibility of using fluorescent EMSA technique to study CRDBP-RNA interaction in-vitro.
1.2.1. Methodology:
1.2.1.1. Generation and Purification of the Coding Region Determinant-Binding
Protein
The wild type CRD-BP was generated and purified using the following methods.
1.2.1.2. Transformation of competent E.coli BL21 cells
27

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and G UI
Approximately 100 ng of purified pet28B-WT-CRD-BP plasmid was added to 100 pL of
thawed BL21 cells, followed by incubation on ice for 25 minutes. The transformation mixture
was heat shocked for 90 seconds at 42°C and placed on ice for 1 minute. After addition of LBbroth (-300 pL), the mixture was incubated at 37°C for 30 minutes. After incubation, the cells
were plated onto LB-Kanamycin plates and placed in incubator at 37°C overnight (-16 hours) to
give sufficient time for colony formation.
1.2.1.3. Cell Growth and Induction
Following the overnight incubation, about 20 bacterial colonies were picked from the plate and
inoculated in 100 mL of LB-Kanamycin broth in a 250 mL flask and transferred to bacterial
shaker to grow (200 rpm, 37°C) for -3 hours. The 100 mL of bacterial culture was transferred to
900 mL of LB-Kanamycin broth in a 3 L flask. The flask was placed back in the same shaker
until an OD600 = 0.5 was reached (-2-3 hours). After achieving the desired OD600 = 0.5, 1 mL
1 M IPTG was added to induce recombinant protein production. The Flask was left for a 6-hour
growth period in the shaker, the bacterial cultures were spun down (3000 x g, 4 °C, 15 minutes)
and the bacterial pellet was kept at -80°C overnight.
1.2.1.4. CRD-BP purification under denaturing conditions
The next morning, bacterial pellets were thawed on ice for 15 minutes. Once thawed, the pellet
was resuspended in 12 mL of Buffer B (100 mM NaH2 PC>4 , 10 mM Tris-Cl, 8 M Urea, adjusted
to pH 8) and transferred to a 50 mL falcon tube. The solution was then gently agitated on shaker
at 4°C until the solution turned semi-translucent (-60 minutes). In the meantime, the pH of the
remaining buffers was adjusted using the stock Buffer B (C-pH 6.5, D-pH 6.3, and E-pH 5.9).
The bacterial solution was then spun down (13,200 rpm, 4°C, and 30 minutes) and the
supernatant was collected. The supernatant was then loaded onto a Nickel-NTA gravity drip
column (column was equilibrated with Buffer B). Each buffer was passed over the supernatant
28

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and G U I
loaded on the column one by one. CRD-BP was expected to elute in buffer E. Fractions were
collected and then checked for protein purity using Coomasie blue-stained SDS-PAGE. Briefly,
16 pL samples from different fractions was mixed with 4 pL 5 x Sample buffer and 1 pL o f Pmercaptoethanol and then boiled for 5 minutes. Samples were then loaded on to a 12% SDSPAGE gel; 19:1 acrylamide:N,M-methylenebisacrylamide and resolved at 120 V for -35
minutes. E fraction proteins were used in all subsequent experiments due to high protein purity.
1.2.1.5. Protein Dialysis
All purified proteins were dialysed using a two-step gradient to remove urea from the Buffer E
elution buffer and to help protein renaturation. Approximately 100 pL of the purified protein was
added into each Slide-A-Lyzer mini dialysis unit (Pierce, Rockland IL, USA) and the unit was
placed in a 100 mL beaker containing 50 mL of dialysis buffer A (1 M Tris-Cl, 0.2 M
glutathione-reduced, 0.1 M glutathione-oxidized, 1.33 M glycerol, 2 M urea, 0.0001% Triton-X,
adjusted to pH 7.4). Buffer exchange was allowed for 24 hours. After 24 hours, the dialysis units
were transferred to 250 mL beaker containing 50 mL of dialysis buffer B (1 M Tris-Cl, 1.33 M
glycerol, 0.0001% Triton-X, adjusted to pH 7.4). After 2 hours, dialysis units were transferred to
250 mL beaker containing 200 mL o f buffer B for 2 more hours. Dialysed protein was spun
down at 13,200 rpm for 30 minutes to remove any aggregation. Following protein dialysis, the
concentration of the protein was determined using Quick Start ™ Bradford lx Dye Reagent
(BioRad, Catalog Number 500-0205). The dialyzed proteins were then used for EMSA within 3
weeks upon dialysis.
1.2.1.6. Generation of DNA template for in-vitro transcription
Polymerase chain reaction (PCR) was employed to generate truncated MDRl and GLI1 DNA
templates using MDRl cDNA (accession # M l4758) and G U I cDNA (accession # 013000)
respectively. The different sets of forward and reverse primers used to amplify the different
29

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GUI
regions of GLI1 cDNA are shown in Table 1. Forward and reverse primers used to amplify
MDRl cDNA are shown in Table 3.

Table 1. Prim er sequences for amplifying fragments of GLI1 cDNA.

Fragments

Primers

G L I36-231

Forward:
GGATCCTAATACGACTCACTATAGGGCCATGTTCAACTCGAT
G
Reverse: CCTCGGTGCAGCTGTTGG

G LI 230-420

Forward:
GGATCCTAATACGACTCACTATAGGGGGCCCACTCTTTTCTTC
Reverse: TGCCAATGGAGAGATGAC

GLI 420-610

Forward:
GGATCCTAATACGACTC ACTATAGGACCATGAGCCCATCTCT
G
Reverse: CGGCACTTGCCAACCAGC

GLI 610-800

Forward:
GGATCCTAATACGACTCACTATAGGGGGAGGAACCCTTGGAA
G
Reverse: GTGCACCAGCTGCTCTTG

GLI 800-990

Forward:
GGATCCTAATACGACTCACTATAGGCCACATCAACAGCGAGC
A
Reverse: GGTTTTCGAGGCGTGAGT

GLI 230-380

Reverse: GCATCGCGAGTTGATGAA

GLI 230-330

Reverse: CCGTCTGCAGGTCCAGGC

GLI 230-300

Reverse: GAGGTGAGATGGACAGTG

GLI 230-275

Reverse: CTTGGTCAACTTGACTGC

GLI 270-420

Forward:
GGATCCTAATACGACTCACTATAGGACCAAGAAGCGGGCACT
G

GLI 320-420

Forward:
GGATCCTAATACGACTC ACT ATAGGCCTGCAGACGGTTATCC
G
30

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and G U I
GLI 350-420

Forward:
GGATCCT AATACGACTC ACTATAGGCTCCCTCGTAGCTTTCAT

GLI 375-420

Forward:
GGATCCTAATACGACTCACTATAGGCGATGCACATCTCCAGG
______________ A___________________________________________________________
All forward primers have T7 promoter sequence at the 5’end (underlined)
Reaction tubes were set up as in Table 2. PCR reactions were performed in a thermo cycler
using the following program- 94°C- 30 seconds, 50°C- 30 seconds, 72°C - 45 seconds for 35
cycles.
Table 2. Reagent used in PCR to synthesize DNA template
Reagent
DNA template
lOxPCR buffer
dNTPs (2.5mM)
Forward primer with T7 promoter (lOOng/ pi)
Reverse primer lOOng/pl
Taq polymerase (NEB)
H20
Total Volume

Amount
100 ng
3.5 pL
3.5 pL
1 pL
1 pL
0.5 pL
24.5 pL
35 pL

PCR products were resolved on a 2% or 2.5 % agarose gel depending upon the sizes of DNA.
The PCR products were subjected to standard ethanol precipitation (2.5 x volume ethanol, 1/10
volume sodium acetate (3 M, pH 5.2 in lOx glycogen). Concentration of DNA was determined
using Nanodrop Spectrometer (NanoDrop technologies). Impure DNA templates were gel
extracted using Qiagen QIAEX®II gel extraction kit.

31

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and G UI

A
36

36

990 3’

........................ 231
230 —

—

420
420 ■— —

—

610

610 — —

—

800

800

990

B
230

42Q 3>

230

380

230

330

230

— i 300

230

275

270 ^

■“ ■■■■■■■■ 420
320

420
350

420
375

— 420

Fig 7. Schematic diagram of truncated G U I cDNA template. (A) Strategy to make five
truncated G U I cDNA spanning nts 36-990, each having ~190 nts using different sets of forward
and reverse primers. (B) Strategy to make 3’ and 5’ truncated G U I cDNA spanning nts 230-420
each having different sizes.

32

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GUI

739

921

739

881

739

845

7 3 9 ................................................ .............. — . i i »

739 — — — —

—

815

785

779

881
819

mmmmmmmm— mmmmmmm— m m m m m m m m 8 8 1
850

..........................................

881

Fig 8. Schematic diagram of truncated M D R l cDNA template. Strategy to make 3’ and
5’truncated MDRl cDNA spanning nts 739-921 using different sets of forward and reverse
primers.
Table 3. Prim er sequneces for amplifying fragments of M D R l cDNA.
Fragments

Primers

MDR 739-921

Forward:
GGAT CCTAATACGACT CACTATAGGAATCTGG AGGAAGAC AT
G
Reverse: TGCACATCAAACCAGCCT

MDR 739-881

Reverse: GAAAAAACTGTTTTCTAA

MDR 739-845

Reverse: CAGCTGCCAGGCACCAAA

MDR 739-815

Reverse: TGTAAGCAGCAACCAGCA

MDR 739-785

Reverse: TTCCACTGTAATAATAGG

MDR 779-881

Forward:
GGATCCTAATACGACTCACTATAGGAGTGGAATTGGTGCTGG
G

MDR 819-881

Forward:
GGATCCTAATACGACTCACTATAGGAGGTTTCATTTT
GGTGCC

Forward:
GGATCCTAATACGACTCACTATAGGACAAATACACAAAATTA
G
All forward primers have T7 promoter sequence at the 5’ end (underlined)
MDR 850-881

33

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and G UI
1.2.1.7. Internal labeling of RNA oligonucleotide using Fluorescein-UTP
In order to generate good quality fluorescently labeled RNA substrate, fluorescein-labeled UTP
was used to internally label the RNA substrate using MAXIscript® kit (Invitrogen). Reagents
from the kit (Nuclease free H2 O, 10 x buffer, 10 mM NTPs, FI-UTP, 1 U T7 polymerase) and 1
pg DNA template were used in each reaction. Reaction mixture was incubated at 37°C for 60
minutes. Following incubation, DNase I was added to degrade DNA template and reaction
mixture was again incubated at 37°C for 15 minutes. After DNase treatment, 0.5 M EDTA was
added to the reaction mixture. Standard phenolichloroform extraction was then performed to
remove any residual proteins from the reaction. The RNA was ethanol precipitated as described
above and column purified using G-50 column to remove unincorporated nucleotides.
1.2.1.8. Fluorescent electrophoretic mobility shift assays
To assess the binding ability of WT-CRD-BP to fluorescently labeled RNA substrate,
electrophoretic mobility shift assay (EMSA) was performed. EMSA binding buffer (Table 5)
was made fresh for each experiment and was placed on ice prior to use. Fluorescently labeled
RNA substrates were subjected to a denaturation and renaturation steps that involved heating the
RNA to 55°C for 5 minutes and then cooling at room temperature for 7 minutes. EMSA reaction
tubes (binding buffer, BSA, yeast tRNA, RNasin, RNA, protein) were then incubated at 37°C for
10 minutes. Following incubation, tubes were placed on ice for 5 minutes. This heating cooling
cycle to facilitate binding was repeated, for a total of two times. EMSA loading dye (2 pL) was
then added to the reaction and 15 pL of the reaction was loaded on to an 8% native
polyacrylamide gel. Complexes were resolved at 25 mA for 90 minutes. The gels were then
visualized using the Kodak Image Station 4000MM PRO.
1.2.2. Results
1.2.2.1. Generation of DNA templates
34

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GUI
Eight MDRl and thirteen GLIl truncated DNA templates were synthesized using specific
reverse and forward primers (with T7 promoter). PCR was performed as described above. PCR
products were resolved on agarose gels to confirm product size and purity. As shown in Figs. 910, most of the amplified DNA templates were pure and of the expected size.
A

Fig 9. Agarose gel analysis of PCR generated G LIl cDNA templates. (A) Five truncated
GLIl cDNA template having -190 nts were resolved on 2% agarose gel. (B) 3’ truncated GLIl
cDNA templates were resolved on 2.5% agarose gel. (C) 5’ truncated GLIl cDNA templates
were resolved on 2.5% agarose gel.

35

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
A

B

506 — I
396— 1
344298 201,220 —|
154— I
134— I
75— 1

Fig 10. Agarose gel analysis of PCR generated M D R l cDNA templates. (A) 3’ truncated
MDRl cDNA templates were resolved on 2.5% agarose gel. (B) 5’ truncated MDRl cDNA
templates.
I.2.2.2. Purification of WT-CRD-BP
To confirm that the IPTG-induced purified WT-CRD-BP protein is pure, protein fractions
36

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
A

Fig 11. SDS-PAGE analysis of recombinant WT-CRD-BP. To confirm generation and
purification of CRD-BP. Various B-F fractions (A and B) were analyzed on 12% SDS-PAGE.
Gels were stained with Coomassie blue stain.
eluted at different pH were run on a 12% SDS-PAGE. As shown in Fig 11, an intense band of
about 68 kDa which eventually became the only dominant band at the later E-fractions and in F

37

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
fraction was observed. This band has been previously confirmed to be recombinant CRD-BP
using anti-CRD-BP and anti-FLAG antibodies (Barnes et al, submitted,); Sparanese and Lee,
2007).
I.2.2.3. Generation of fluorescently labelled MDRl RNA
Internal labelling of fluorescent RNA was performed using fluorescein UTP and
MAXIscript® kit (Invitrogen). As shown in Fig 12, both the fluorescently labeled M DRl and cmyc RNAs were successfully generated. The MDRl RNA is expected to be 183 nts and the cmyc is expected to be 182 nts, therefore the approximate similar size of both transcripts
supported the identity of the bands.

Fig 12. Analysis of in-vitro transcription product. MDRl and c-myc transcripts resolved on a
8% denaturing polyacrylamide gel.

I.2.2.4. Fluorescent EMSA
After successfully obtaining fluorescent labeled MDRl RNA, electrophoretic mobility gel shift
assay was performed using different concentrations of CRD-BP (Fig 13). A faint band of higher
molecular weight which was absent in the lane with no CRD-BP added, was observable in lanes
containing CRD-BP. Unfortunately, the CRD-BP-RNA complex band remained faint based on
multiple experiments and could not be feasibly used for quantification purposes. Therefore, the
fluorescent EMSA was deemed inappropriate for further use. For the remaining work, the
radioactive EMSA using [32P]-labeled RNA was employed to study CRD-BP-RNA binding.
38

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl

WT-CRD-BP
(nM)

O

\^ 0 \ ^

Bound RNA [

Free RNA[

Fig 13. Fluorescent EMSA of M D R l nts 739-921. EMSA binding reaction was performed
using MDRl and various concentration of WT-CRD-BP. Formation of protein-RNA complex
indicates that MDRl RNA binds with CRD-BP.

Part B: Assessing CRD-BP-RNA interaction using [32P] radiolabelled M D R l and GLFl
RNA
After the unsuccessful attempt in using the fluorescent EMSA to quantify CRD-BP-RNA
interaction, an alternate method was chosen to assess CRD-BP-RNA interaction. I decided to use
the radioactive EMSA to study and quantify CRD-BP-RNA interaction.
II.2.1. Methodology
II.2.1.1. Generation of internally [32P] labeled RNA substrate
Radio-isotope [32P-UTP] labeled MDRl and GLIl RNA substrates were generated via in-vitro
transcription using linearized DNA template as prepared above. T7 directed in-vitro transcription
was performed to synthesize RNA of desired size. The required reagents given in Table 4 were
mixed in reaction tube and incubated at 37°C for 1 hour. DNase I was added to destroy DNA
template and reaction was incubated again at 37°C for 15 minutes. RNA stop dye was added to
the reaction mixture and mixture was loaded onto an 8% denaturing PAGE gel. Reaction mixture
was resolved for 60 minutes at 25 mA. Gel was visualized using Cyclone Storage Phosphor
Screen System and Optiquant Software. Gel extraction of the desired product was performed.

39

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
Gel homogenate was loaded on a DTR column and spun at 3000 rpm for 3 minutes to remove gel
fragments and free nucleotides. Standard phenol:chloroform extraction was then used to remove
any possible residual proteins. Ethanol precipitation was then carried out to concentrate the RNA
substrate. The radioactivity of the substrate was checked using a scintillation counter. The RNA
substrates were then diluted and subsequent binding reactions used ~40,000 - 50,000 cpm
(counts per minutes) per reaction.
Table 4. Reagents used in generation of [32P] labelled RNA
Reagent
DNA template with T7 promoter(l pg/pL)
5* transcription buffer
100 mM DTT
RNasin 20 U/pL
10 mM ATP
10 mM CTP
10 mM GTP
100 pM UTP
32P -UTP
T7 RNA polymerase 15 U/ pL
DEPC water
Total

Amount
1 pL
4 pL
2 pL
1 pL
1 pL
1 pL
1 pL
2.5 pL
3 pL
1 pL
2.5 pL
20 pL

II.2.1.2. Radioactive electrophoretic mobility shift assay
To assess the ability of CRD-BP to bind [32P]-labeled MDRl and GLIl RNA substrates,
radioactive electrophoretic mobility shift assays (EMSA) were performed. EMSA binding buffer
32

(Table 5) was made fresh at the start of each experiment and was placed on ice until used. [ P]labeled RNA substrates were subjected to a denaturation and renaturation steps that include
heating the RNA to 55°C for 5 minutes and then cooling at room temperature for 7 minutes.
Equal amounts of binding buffer and RNA substrates, while various concentrations of CRD-BP
were added to the EMSA reaction tubes. EMSA reactions were then incubated at 35°C for 10
minutes, and then kept on ice for 5 minutes. This heating and cooling cycle to facilitate binding
was performed two times. EMSA loading dye (2 pL) was then added to the reactions and 15 pL

40

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
o f the binding reaction was loaded on to a 4% native PAGE gel and resolved at 25 mA for 50-55
minutes. Audioradiography was performed using a Cyclone Storage Phosphor Screen System
and Optiquant Software.
Table 5. Reagents used for fluorescent and radioactive EMSA
Reagent

Composition

EMSA binding buffer

5 mM Tris-Cl (pH 7.4), 2.5 mM EDTA (pH 8), 2 mM DTT,
5% glycerol,0.1 mg/mL (BSA), 50 ug/uL Yeast tRNA
(Invitrogen), 5U RNasin

EMSA loading dye

250 mM Tris-Cl (pH 7.4), 0.2% bromophenol blue, 0.2%
xylene cyanol,40% sucrose

II.2.2. Results:

II.2.2.1. Mapping of GLIl RNA
All radiolabelled RNAs were gel purified and precipitated successfully. GLIl RNA substrates
had radioactivity of about 800,000-1000,000 cpm while MDRl fragments had radioactivity of
about 200,000-300,000 cpm. Radioactivity was measured using scintillation counter.
Radiolabelled EMSA was performed on five [32P]-labeled GLIl fragments (36-231, 230-420,
420-610, 610-800, 800-990), each having -191 nts. Various CRD-BP concentrations were added
to observe the binding ability of these five RNA fragments. Out of these five fragments, only one
fragment (230-420) showed binding with CRD-BP (Fig 14A). GLI 420-610 did not show any
binding with CRD-BP even at 1080 nM concentration of protein (Fig 14A). GLI 36-231, GLI
610-800 and GLI 800-990 did not show binding with CRD-BP as shown in Fig 14B. At higher
protein concentrations, the intensity of free RNA was reduced and this was most likely due to the
aggregation of proteins in the wells of gel. Radiolabelled EMSA was also performed on eight
other 3’ and 5’ truncated [32P]-labeled truncated fragments of GLIl nts 230-420. As shown in
Fig. 15-16, out o f these eight RNA fragments, only four fragments showed binding with WT-

41

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
CRD-BP. Interestingly, all truncated fragments which showed binding with CRD-BP possess nts
320-380, suggesting that this could represent a minimal region required for binding to CRD-BP.
A

B
C-Myc

CRD-BP (n M )

0

,< > H

GU 230-420
l V

/

GLI 420-610

C-Myc GLI 610-800 GU 800-990 GU 36-231

0

0

* 1 ^ 0

^ 1 *

Bound RNA

Free RNA

Fig 14. Mapping the G LIl RNA nts 36-990 for ability to bind CRD-BP. All five truncated
GLI RNA substrates were allowed to bind with various concentrations of CRD-BP using [32P]radiolabeled EMSA reaction. Reaction mixture was resolved on 4% native PAGE. (A) GLI 230420 showed binding with CRD-BP while GLI 420-610 did not show any binding. (B) All three
GLIl fragments showed no binding with CRD-BP.
A

B
GU 230-420 GU 230-300 GU 230-275

C-Myc GU 230-380 GU 230-330

C
Bound RNA

Fig 15. Mapping the 3’ truncated G LIl RNA nts 230-420 for ability to bind CRD-BP. All
four 3’ truncated GLI 1 RNA substrates were allowed to bind with various concentrations of
CRD-BP using [32P]-radiolabeled EMSA reaction. Reaction mixture was resolved on 4% native
PAGE. (A) 3’ truncated GLI 230-380 showed binding with CRD-BP while GLI 230-330 showed
no binding. (B) 3’ truncated GLI 230-300 and GZ.7 230-275 showed no binding with CRD-BP.

42

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
M DRl and GUI
A
GU 230-420

GLI 270-420

GU 320-420

Bound RNA

B
GU 350-420

GU 375-420

Bound RNA

Fig 16. Mapping the 5’ truncated G LIl RNA nts 230-420 for ability to bind CRD-BP. All
four 5’ truncated GLI 1 RNA substrates were allowed to bind with various concentrations of
CRD-BP using [32P]-radiolabeled EMSA reaction. Reaction mixture was resolved on 4% native
PAGE. (A) 5’ truncated GLI 270-420 and G U 320-420 showed binding with CRD-BP. (B) 5’
truncated GLI 350-420 showed binding with CRD-BP while GLI 375-420 showed no binding.

43

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl

GLI 320-380

Bound RNA

Free RNA

Fig 17. Assessing G LIl RNA nts 320-380 nts for ability to bind CRD-BP. GLI 320-380 RNA
substrate was allowed to bind with various CRD-BP concentrations using [32P]-radiolabeled
EMSA reaction. Reaction mixture was resolved on 4% native PAGE.
After performing EMSA, it was evident that GLI 320-380 region plays an important role in
binding CRD-BP. Next, the MFOLD web server was used to generate predicted secondary
structure of GLIl 301-380. The MFOLD generated predicted secondary structure of GLIl nts
301-380 is shown in Fig 18A. This structure contains two stem loops within 346-382 nts region
as shown in Fig 18A. Interestingly, the 2862-2930 nts CD44 RNA which is also capable of
binding CRD-BP also contains the two stem loops as shown in Fig 18B. This 69 nts CD44 RNA
was also capable of binding CRD-BP (King et al, 2014).
A

B

Fig 18. MFOLD (Zuker, 2003) generated secondary structures of (A) GLI RNA nts 301-380
and (B) CD44 RNA nts 2862-2930
44

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
11.2.2.2. Mapping of MDRl RNA
Radiolabelled EMSA was also performed on [32P]-labeled truncated MDRl RNA. The MDRl
RNA nts 739-921 has been previously shown to bind CRD-BP (Sparanese and Lee, 2007).
A
MDR1 739-921

MDR1 739-845

MDR1 739-815

MDR1 739-785

Bound RNA

B
MDR 739-921
9

MDR 779-881

qg»C>.hT-g> O v

s

MDR 819-881

MDR 850-881

O

Bound RNA

Free RNA

MDR 739-921
9

MDR 739-881

n & ep P

Bound RNA

Free RNA

Fig 19. Mapping of MDRl RNA for ability to bind CRD-BP: 3’ and 5’ truncated MDRl
RNAs were incubated with various concentrations of CRD-BP and reaction mixture was
resolved on 4% native PAGE. (A) MDR 739-845 and MDR 739-815 showed binding with
protein while MDR 739-785 showed no binding. (B) MDR 779-881 and MDR 819-881 showed
binding with protein while MDR 850-881 showed no binding. (C) MDR 739-881 shows binding
with CRD-BP.

45

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
Out o f seven 3’ and 5’ truncated MDRl RNAs, three RNAs nts 739-881, 739-845 and 779-881
showed high affinity for CRD-BP (Fig 19). MDRl RNA nts 739-815 and 819-881 RNAs showed
weak binding while nts739-785 showed very weak binding (Fig 19). In contrast, nts 850-881
RNAs showed no binding with CRD-BP. MDRl RNA nts 779-881 appeared to have similar
affinity as the RNA nts 739-921 for CRD-BP (Fig 19), suggesting that nts 779-881 of MDRl
RNA plays important role in binding to CRD-BP.
Part C: Breaking CRD-BP-GL/7 RNA interaction
After mapping the smallest region of GLIl RNA which can bind CRD-BP, my next goal was to
design and assess specific oligonucleotides for ability to disrupt CRD-BP-GL/1 RNA interaction
in vitro. This section discusses the design of specific RNA and DNA oligonucleotides and the
experiments conducted to assess the ability of these oligonucleotides to compete with the [ P]labeled GLIl nts 230-420 for binding to CRD-BP.
III.2.1. Methodology

m.2.1.1. Designing RNA and DNA oligonucleotides using MFOLD
The predicted secondary structure of GLIl RNA nts 301-380 containing the two stem loops in
region 346-382 nts is shown in Fig 2 0 .1 hypothesize that these two stem loops play an important
role in binding CRD-BP.

To test the hypothesis that the secondary structure, particularly the

two stem loops, as well as sequence of GLIl RNA are critical in binding CRD-BP, eight RNAs
and one DNA oligonucleotides were designed using MFOLD as shown in Fig 21. These
oligonucleotides with identical length were designed to have different structures as well as slight
variations in their nucleotide sequences as shown in Table 6. DNA oligonucleotide and SI have
exactly the same structure and nucleotide sequences as GLI 346-382 nts region. S2, S3, S4, S5
and S6 were designed to test the importance of both loops to bind CRD-BP. S7 was designed to

46

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GLIl
observe the importance of both stems in binding CRD-BP, while S8 was designed to test the
importance of stem structure to bind CRD-BP.

S'

C

i

c
1
A
U
i
u
2(0

r
c
I
G

I
C

I
A
I
3'

C

Fig 20. MFOLD generated predicted secondary structure of G LIl RNA nts 346-382

Table 6. Sequences of the designed DNA and RNA oligonucleotides
Oligonucleotides

Sequence (5’ to 3’)

DNA oligonucleotide
SI
S2
S3
S4
S5
S6
S7
S8

CCAGCTCCCT CGTAGCTTTC ATCAACTCGCGATGCAC
CCAGCUCCCUCGUAGCUUUCAUCAACUCGCGAUGCAC
CCGGACCCCUCGUAGCUUUCAUCAACUCGCGAUGCAC
CCAGCUCCCUCGUAGCUUUCAUCAACUCGCCACCCCC
CCAGCUCCCUAGCUUUUUUCAUCAACUCGCGAUGCAC
CCAGCUCCCUCGUAGCUUUCAUCAACUGAUGCACCCG
CCCCCAGCUCCCUAGCUUUCAUCAACUCGCGAUGCAC
CCCUGCCCCUCGUGCAGUUUCUAAACUCGCUAGACAC
CCAGCUGAACGAAAGCUUUCAUCUUAACCCGAUGCAC

47

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
M DRl and GLIl
DNA oligonucleotide

S3

S4

S5

Fig 21. The predicted secondary structures of DNA oligonucleotide and RNA
oligonucleotides (S1-S8) as generated by MFOLD.

48

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GUI
IU.2.1.2. Generation of [32P]-labelled G L I230-420
[32P]-labeled G U 230-420 RNA substrate was synthesized as described above
IH.2.1.3. Generation of unlabelled RNA
Unlabelled GUI RNA nts 230-420, 420-610 and 320-380 were synthesized using T7
polymerase directed in-vitro transcription. Each reaction was set up as according to the quantities
given in Table 7. Reaction was incubated at 37°C overnight. Reaction mixture was treated with
DNase I to remove the DNA template. Phenol-chloroform extraction was done to remove protein
impurities followed by passing through G50 column to remove unincorporated nucleotides. RNA
was ethanol precipitated and suspended in DEPC water. Purity of the product was checked by
running the product on 2.5% agarose gel. Concentration of RNA was determined using
Nanodrop spectrometer (NanoDrop Technologies).
Table 7. Reagents used in synthesis of unlabelled RNA
Reagent
Gel Purified DNA template (lug/pL)
10 x T7 buffer (w/ Triton X)
100 mM ATP
100 mM CTP
100 mM GTP
lOOmMUTP
RNasin (40 U/pL)
T7 polymerase
dH20
Total

Amount
5 pL
10 pL
5 pL
5 pL
5 pL
5 pL
1 pL
5 pL
57 pL
100 pL

III.2.1.4. Competition assay with oligonucleotides
[32P]-labelled GLI\ RNA nts 230-420 was synthesized as mentioned above. CRD-BP protein
was purified, dialysed and quantified as mentioned above. Competition assay was optimized by
adding 13 nM [32P]-labelled G U 230-420 with 300 nM CRD-BP and 1/6-.1/3-, 1-, 1.5- and 10fold molar excess increased in unlabelled G U 230-420 and G U 420-610.

49

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GUI
EMSA competition assay was performed between [32P]-labelled G L I230-420 RNA with the
designed oligonucleotides as well as unlabelled GLI 320-380. The final concentration of 13 nM
G U 230-420 and 300 nM CRD-BP were used. For the unlabelled GLI 320-380, 246- and 1723fold molar excess , while 1615-, 4846-, 8075-, 11305-fold molar excess were used. Reaction
mixture was incubated at 35°C for 10 minutes before putting on ice for 5 minutes. This heating
cooling cycle was done for a total of two times. Two pL EMSA loading dye was added in each
tube and reaction was loaded onto 4% non-denaturing polyacrylamide gel. Reaction mixture was
resolved for 55 minutes at 25 mA. Audioradiography was performed using a Cyclone Storage
Phosphor Screen System and Optiquant Software.
III.2.2. Results
III.2.2.1. Optimization of competition assay
As clearly shown in Fig 22, the unlabelled GL1230-420 competed effectively with the [32P]labelled GLI 230-420. At 1.5 fold molar excess, the unlabelled GLI 230-420reduced the bidning
of [32P]-labeled GLI 230-420 to CRD-BP by 50%, while at 10-fold molar excess the binding was
reduced to 15%. The unlabelled GLI 230-420 has the exact same sequence as of [ P]-labelled
G U 230-420, so it is not surprising to be an effective competitor.

GLI 2 3 0 - 4 2 0
F old I n c re a s e

®

N

B o u n d RNA

F ree RNA

Fig 22. Optimization of competition assay: Competition assay was optimized by adding
different fold increase in unlabelled RNA substrate of GLI 230-420. Reduced binding was
observed beginning at 1.0-fold concentration of unlabelled GLI 230-420.
50

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GL/1
III.2.2.2. Competition assay with RNA and DNA oligonucleotides
Upon demonstrating the ability of the unlabeled GLI RNA nts 230-420 to compete with the
radiolabeled

version

to

bind

to

CRD-BP, different concentrations

of the

designed

oligonucleotides were assessed for their ability to compete with the [3 2 P]-labelled GLI 230-420
RNA for binding to CRD-BP. As shown in Fig 23A, the in-vitro transcribed unlabelled GLI 320380 RNA competed effectively with [3 2 P]-labelled GLI 230-420.
A

B

GLI 320-1
Fold Increase

0

DNAoligo

Sl_

S1

A

Fold In crease

S2

S3

A

p

Bound RNA

Bound RNA

Free RNA
Free RNA

D
S1
S6
S7
S8
-------------------------------------------------

S1_ _ _ _ _ _ _ _ _ _ S4_ _ _ _ _ _ _ S5
Fold Increase o

F°idincrease

Bound RNA

Bound RNA

Free RNA

Free RNA

«

^ s 'V V ’$ $

i l l s
Fig 23. Breaking GLI 230-420-CRD-BP interaction using specific oligonucleotides.
Competition assay was performed by adding various concentrations of oligonucleotides along
with [3 2 P]-labelled GLI 230-420 and CRD-BP. Reaction mixture was resolved on 4% native
PAGE. (A) GLI 320-380 diluted the protein-RNA complex at lower concentration. DNA oligo
did not affect GL/-CRD-BP binding while SI completely disrupted the interaction at 11305-fold
molar excess. (B) S2 and S3 both reduced binding. (C) S5 also affected protein-RNA binding
while S4 showed no effect on binding. (D) S 6 and S7 showed no effect on binding, while S 8
affected the binding to some extent.
51

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDR\ and GUI
At 246-fold increase in molar concentration it reduced the binding to 26% while at 1723-fold
increase in molar concentration it reduced binding to 13%. Oligonucleotide SI reduced binding
to 11% at 11305-fold molar increase. S2 and S3 also showed significant reduction at 11305-fold
molar excess and reduced binding to 60% and 44% respectively. There was no effect on binding
after addition of various concentrations of S4, S6 , S7, and the DNA oligonucleotide as shown in
Fig 23. At 11305-fold molar excess, S5 reduced binding to 24% while S8 reduced binding to
84%. From multiple experiments, it is clear that SI is the best RNA competitor. SI is the sense
oligonucleotide having the identical sequence and structure as of GLI nts 346-382. The overall
EMSA competition results suggest that the two stem loops o f GLI 230-420 RNA do indeed play
an important role in binding to CRD-BP.
2.3 D iscussion

This chapter mainly focussed on characterizing the CRD-BP-RNA (GL/1 and M D Rl)
interaction in vitro and then finding specific oligonucleotides that are capable of breaking the
CRD-BP-GZJ/ RNA interaction. One of the best available in vitro techniques used to study
protein-RNA interaction is EMSA. CRD-BP-RNA interaction was first assessed using
fluorescent EMSA. Fluorescence EMSA was given the priority over radiolabeled EMSA because
of its safety and cost effectiveness. Although internally-labeled fluorescent RNAs were
successfully synthesized and used, unfortunately the fluorescent EMSA could not give the
desired results which permit its regular use in studying CRD-BP-RNA interaction. Firstly, the
sensitivity of fluorescent RNA-protein complex was too low for quantification purposes.
Secondly, fluorescent RNA-protein aggregation in the wells of polyacrylamide gels was
consistently observed. The presence of such complexes might be due to the presence of large
sized and number of fluorescein molecule internally attached to the transcript. This problem was
addressed by using different ratios of fluorescein-UTP and unlabeled UTP to reduce fluorescein

52

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDR\ and GUI
UTP molecules in the transcript, but to no avail. Another possible problem was the non-specific
protein aggregation. This problem was solved by using E-eluted protein fraction because Feluted protein fractions tend to make aggregates (Sparanese and Lee, 2007). The dialysis
protocol was also modified to minimize risk of protein aggregation. These latter strategies helped
in improving the mobilization of protein-RNA complexes but as mentioned above the low
sensitivity of protein-RNA bands did not permit quantification of band intensity for the purpose
of calculating dissociation constant Kd values.
After multiple failed attempts in performing fluorescent EMSA, radiolabeled EMSA was the
alternate option available. Radiolabeled EMSA was already a well-established method in the lab
(Sparanese and Lee, 2007; King et al, 2014; Kim et al, 2011 and 2014; Barnes et al, submitted).
The radiolabeled EMSA was successfully used in this work to study CRD-BP-RNA interaction.
In mapping the smallest GLIl RNA that binds CRD-BP, I showed that GLI 320-380 nts region
plays a pivotal role in CRD-BP-GZJ RNA interaction. The 61 nts GLIl nts RNA 320-380 has
two stem loops of equal sizes. I hypothesized that both loops are important in binding to CRDBP. The MDR1 RNA nts 779-881 shows strong binding with CRD-BP, suggesting that 103 nts
region of MDR1 RNA is sufficient for binding CRD-BP. These results also confirm that strategy
to make truncated RNA fragments is a valid and easy approach to map the region of any RNA
that binds CRD-BP.
Specific oligonucleotides were designed to test the hypothesis that the secondary structure stem
loops and its sequences are important for binding CRD-BP. SI RNA oligonucleotide has the
same sequence and structure as the GLIl RNA nts 346-382 and was the most effective
competitor. The SI DNA oligonucleotide, a DNA version of SI RNA oligonucleotide, was
unable to show any competitive effect with [3 2 P]-labelled GLI 230-420 in binding to CRD-BP.
This clearly demonstrates the specificity of CRD-BP in binding to RNA molecules only. Seven

53

Chapter 2. Characterizing the physical interaction between CRD-BP and its two mRNA targets,
MDRl and GUI
additional RNA oligonucleotides (S2-S8) were designed, each having different loop sizes and
slightly different sequences of nucleotides. S2, S3, S5 and S 8 competed with [32 P]-labelled GLI
230-420 RNA for binding to CRD-BP, while S4, S6 and S7 did not compete at all. Surprisingly,
S7 was not an effective competitor although it has the two loops with the same structure and
sequence as SI. The reason might be the formation of another more stable secondary structure
according to MFOLD web server having AG= -4.80 Kcal/mol while S7 structure has AG= -4.00
Kcal/mol. Structure S8 has the same two stem loop secondary structure as SI but different
sequences at the loop. It reduced binding to 84% which is not significant as compared to SI.
These results suggest that both stem loop secondary structure as well as their sequences are
important in binding to CRD-BP.

54

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs

Chapter 3

Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs
CRD-BP has two RRM and four KH domains (Fig 24). Previous studies have demonstrated the
importance o f the KH domains of CRD-BP orthologs in binding its target RNAs (Git and
Standart, 2002; Nielsen et al, 2002). Most of these studies were based on deletion analysis where
truncated proteins were used. However, there are some conflicting results regarding the
significance of each of the KH domains in respect to their physical interaction with RNA
substrates. In an effort to understand the contribution of each o f the KH domains of CRD-BP in
binding RNA, site-directed mutagenesis on the G-X-X-G motif at each of the KH domains was
performed (Barnes et al, submitted).

Namely, the first glycine in the G-X-X-G motif was

mutated to an aspartate.
Table 8 shows the abbreviated names of each of the point mutation KH variants. For instance,
the KH1 variant has point mutation in GXXG motif (G212D) within KH1 domain whereas the
KH2 variant has point mutation in GXXG motif (G293D) within KH2 domain. KH1-2 variant
has point mutations in GXXG motifs with KH1 (G212D) and KH2 (G293D) domains, KH3-4
variant has point mutations in GXXG motifs with KH3 (G422D) and KH4 (G504D) domains,
and so on. Two CRD-BP variants with
Table 8. Position of mutation in CRD-BP structural domain
Mutation
Y5A
D526E
G212D
G293D
G422D
G504D

CRD-BP Structural Domain
RRM1
KH4 variable loop
G-X-X-G ofK H l
G-X-X-G of KH2
G-X-X-G of KH3
G-X-X-G ofKH4

55

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs
point mutation which are not expected to affect RNA-binding ability for use as negative controls
were also generated CRD-BP variant with point mutation at the RRM1 (Y5A) was selected.
D526E variant with point mutation within the variable loop located between P2 and P3 in KH4
domain (Chao et al, 2010) was also generated. Fig24 shows the schematic diagram of the
structure of CRD-BP and the location of each of the point mutation mentioned above.

C K

X

H

O

Fig 24. Schematic diagrams of CRD-BP: Location of each of the point mutation is shown by
an arrow (Barnes et al, submitted).
This chapter describes the experiments conducted to compare the RNA binding ability of the
WT-CRD-BP and its variants on in vitro transcribed [3 2 P] labelled CD44 (nts 2862-3055) and
GLIl (nts 230-420) RNA substrates.
3.1 Methodology
3.1.1 Purification and dialysis of WT-CRD-BP and its variants
All of the recombinant proteins used in this study were purified by Mr. Mark Barnes and Mr.
Gerrit Van Rensburg using the method described in Chapter 2. WT-CRD-BP and its variants
were dialysed as described in Chapter 2. The BCA protein assay kit (Catalog # PI23225,
Thermo Scientific) was used to quantify all proteins.
3.1.2 Generation of [3 2 P] -labelled CD44 and G LIl RNAs
. GLI nts 230-420 DNA template was synthesized as described in Chapter 2. The CD44 nts
2862-3055 DNA template was made by Mr. Mark Barnes. In-vitro transcription using T7
polymerase was done to generate CD44 nts 2862-3055 and GLIl nts 230-420 RNA substrates of

56

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs

required size and gel purified as described in Chapter 2. Radioactivity of RNA was checked
using scintillation counter.
3.1.3

Radiolabelled EMSA using [32P] labeled CD44 and GLIl RNAs

After in-vitro transcription o f CD44 nts 2862-3055 and GLIl nts 230-420 RNA substrates,
EMSA reaction was performed using various concentrations of the WT CRD-BP and its variants.
For each reaction tube, ~50,000 cpm of [ P] labelled RNA substrate was added. EMSA binding
reaction was performed and resolved on 4% non-denaturing polyacrylamide gel as described in
Chapter 2.
3.2 Results
After synthesis of [32P] labelled CD44 and GLIl RNAs of expected size and purity, EMSA
reactions using dialysed recombinant proteins were performed.
Fig. 25 shows the EMSA results in comparing the binding profile of the WT CRD-BP and its
variants on CD44 RNA nts 2862-3055.
As shown in Fig. 25A, theY5A and D526E variants exhibited binding profiles which are
comparable to the WT-CRD-BP. These results were expected because location of these
mutations lies outside the KH domains known to be critical for RNA binding (Chao et al, 2010).
KH1, KH2 and KH3 variants showed reduction in binding as compared to the WT-CRDP.
Interestingly, KH4 binding ability was significantly reduced as compared to the WT-CRD-BP
(Fig 25C). Surprisingly, all KH variants with mutation at two G-X-X-G motifs (KH1-2, KH1-3,
KH1-4, KH2-3 and KH2-4), with the exception of KH 3-4 variant, showed complete loss of
binding ability (Fig 25).

57

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs

WT
CRD-BP: *

Y5A

D527E

#$$$ ^ $$$$ $ $$ # CRD-BP:

WT

KH1-2

KH1-3

WT

KH1-4

KH2-3

*

Bound
Unbound _

i

Unbound^

B

WT

CRD-BP: . M
Bound

•
KH1

KH2

W

CRD-BP: *

f 141| t | H

[

M

Unbound-

H

m
i

WT

KH3

KH4

CRD-BP: *

WT
CRD-BP: « £

[r

Unbound -

i

1

Unbound-

C

Bound

f

I

K H 34

K H 24
iA > V

Bound
Unbound ■*

Fig 25. EMSA in assessing the binding profile of the WT-CRD-BP and its variants on CD44
RNA: [32P]-labeled CD44 RNA nts 2862-3055 was incubated with various concentrations of
either the WT-CRD-BP or its variants in EMSA reaction as described in the Methodology. The
reaction mixture was resolved on 4% native polyacrylamide gels. Data shown are representatives
of replicate experiments (n = 4).

Fig. 26 shows the EMSA results in comparing the ability of the WT CRD-BP and its variants
for binding to GL/1 RNA nts 230-420.

58

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs
WT

Y5A

D526E

KH 2

KH 1

WT

B

$$<$$$$$$ CRD-BP 0 sA )?*

CRD-BP $
(nM)

(nM)

Bound

Bound

Unbou

Unbound—

N

M H I

I 4J 4 k £ i
KH 3

WT

KH 4

CRD-BP
(nM)
Bound

CR D -B P
(nM )

''

[

H

i "

U nbo und -!

WT
O

KH 1-4

~

I I

Unbound—

KH 1-2

KH 1-3

jflL.jII

r

I*
M

CRD-BP

WT

[ 1 1 1 -----------------^

WT

KH 2-4
CRD-BP O N * ^

KH 2-3
^ V

KH 3-4
V

^ V

V

>£ °

(nM)
Bound

Bound
Unbound—
Unbound-

II

Fig 26. EMSA in assessing the WT-CRD-BP and its variants for binding to GLIl RNA:
[32P]-labeled GLIl RNA nts 230-420 was incubated with various concentrations of either the
WT-CRD-BP or its variants in EMSA reaction as described in the Methodology. The reaction
mixture was resolved on 4% native polyacrylamide gels. Data shown are representatives of
replicate experiments (n = 4).

The binding ability of Y5A and D527E variants for GLIl RNA was slightly reduced as
compared to the WT-CRD-BP (Fig 26A). As shown in Fig. 26B, KHl variant showed significant
reduction in binding GLI RNA as compared to the WT-CRD-BP. Surprisingly, KH2 variant was
completely incapable of binding to GLIl RNA (Fig. 26B). KH3 and KH4 variants showed
binding profiles which are comparable to the WT-CRD-BP. Interestingly; all KH variants with

59

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs

200

400

000

800

C R D -B P -Y 5 A c o n c e n t r a t io n (nM )

70

eo
so
40
30
20
10

o

o

200

400

800

800

1000

C R D -B P -K H 1 c o n c e n t r a t io n (nM )

CRD-BFr.B537E cpncantratlon (nM)

1000

CRP-BP-KH2 c o n c e n tra tio n (nM)

70

80
SO
40
30

20

200

400

600

600

CRP-BP-KH3-4 c o n c e n tra tio n (nM)

Fig 27 Saturation binding of CRD-BP and its variants on CD44 RNA: The EMSA results
shown in Fig 25 plus another two EMSA results using two separate protein preparations were
pooled and quantified using densitometry. The saturation binding data for each protein which
exhibited typical binding characteristics are shown. The Hill equation was then used to calculate
the dissociation constant (IQ). (n=4)

60

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs

60

100

60

60

•0

60
40
CD

40

30
20

20
10

20 0

soo

Ot P
600

1000

200

400

600

600

1000

CRD-BP YSA con cen tration (nM)

W T-CRD-BP c o n c e n t r a t io n (nM)
70

40

CRD-BP KH4 c o n c e n tra tio n (nM)

CRD-BP D627E con cen tration (nM)

60

60

200

ooo

CRD-BP KM3 c o n c e n tr a tio n (nM)

Fig 28. Saturation binding of CRD-BP and its variants on G L Il RNA: The saturation
binding data shown were generated from the EMSA results in Fig 26 and two EMSAs using two
separate protein preparation. The Hill equation was then used to calculate the dissociation
constant (Kj).

point mutation in G-X-X-G motif at two KH domains completely lost their ability to bind GLIl
RNA (Fig 26).
The saturation binding data pooled from four biological replicate EMSAs are shown in Figs.27
and 28. The Hill equation was then used to calculate the dissociation constant (Ka) for each of
the protein that binds CD44 and GLIl RNAs.

61

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs
Summary of the Ka values of all KH variants with CD44 and GLIl RNAs is shown in Table 9.
Statistical analysis was performed using student-t test method.
Table 9. Summary of the IQ values of CRD-BP and its variants for binding to CD44 and
G LIl RNAs
Protein

CD44 nts 2862-3055

G LIl nts 230-420

WT-CRD-BP

149.32±8.42

288.07±23.49

Y5A

100.1 l±16.41**a

384.36±90.42g

D526E

90.65±1.64**b

313.93±67.13h

KH1

211.18±34.94c

No Binding*

KH2

250.27±26.72**d

No Binding

KH3

219.32±19.35**e

466.18il36.081

KH4

No Binding*

201.77±10.759**j

KH1-2

No Binding

No Binding

KH1-3

No Binding

No Binding

KH1-4

No Binding

No Binding

KH2-3

No Binding

No Binding

KH2-4

No Binding

No Binding

KH3-4

133.33±11.04f

No Binding

* Binding was not sufficient to calculate Ka.
** P<0.05, P-values of all the KH variants for CD44 and GLIl were a=0.0371, b=0.0005,
c=0.1016, d=0.0091, e=0.0143, £=0.2933, g=0.2858, h=0.6970,1=0.1890, j=0.0317
3.3 Discussion
The goal of this Chapter was to assess the role of each of the KH domains of CRD-BP in
binding to CD44 (nts 2862-3055) and GLIl (nts 230-420) RNAs. At the outset of this project,
one of the main concerns was the possibility of major alteration to the overall protein structure
upon mutation at the first glycine of G-X-X-G motif in the KH domain. Fortunately, it has been
confirmed that the WT CRD-BP and all CRD-BP variants exhibited similar, if not identical,

62

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs

circular dichroism spectra suggesting that there is no global secondary structure changes upon
point mutation (Barnes et al, submitted).
In case of binding to CD44 RNA nts 2862-3055, single point mutation did appear to
moderately affect the binding ability CRD-BP.

But the greatest effect was seen with KH4

variant where binding efficiency was significantly retarded resulting in unmeasurable Ka value.
With the exception of KH3-4 variant, all CRD-BP variants with point mutation in G-X-X-G
motifs at two KH domains, completely lost ability to bind CD44 RNA. This result suggests that
the G-X-X-G motif in KH3 and KH4 in combination does not appear to play critical role in
binding CD44 RNA, rather there might be some other binding sites available within the KH3-4
di-domains responsible for binding efficiently to CD44 RNA.
In the case of binding to GLIl RNA nts 230-420, KH2 variant showed complete lost of ability
to bind, suggesting that the first glycine of G-X-X-G motif in KH2 domain play an important
role in binding GLIl RNA. KH1 also showed much reduced binding, suggesting that the first
glycine residue in the G-X-X-G motif in KH1 domain also plays an important role in binding to
GLIl RNA. Surprisingly, all variants with point mutation in G-X-X-G motif at two KH domains,
including the KH3-4 variant, completely lost their ability to bind GLIl RNA.
The overall results suggest that CRD-BP binds to CD44 and GLIl RNAs differently. It is likely
that the sequence and hence the structure of RNA plays an important role in binding CRD-BP.
All four KH domains seem important in binding to CD44 and GLIl RNAs. It seems that CD44
and GLIl RNAs might bind with different KH domains of CRD-BP. The observation that
mutation at two KH domains more significantly deteriorated the binding ability of CRD-BP
supports the hypothesis that RNA loops around two KH domains when it binds with CRD-BP
(Chao et al, 2010).

63

Chapter 3. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNAs

The K<j values of WT CRD-BP for binding to CD44 and GLIl RNAs were slightly different.
The Kd value for CD44 RNA was lower than that for GLIl RNA suggesting that CRD-BP binds
more avidly to CD44 RNA.

64

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

Chapter 4

Investigating the regulation of GLIl mRNA expression by CRD-BP in cancer cells
In light of the successful characterization of CRD-BP-GLIl RNA interaction in vitro as
described in Chapter 2, studies described in this Chapter were undertaken to explore the CRDBP-GLI1 mRNA interaction in cells. Specifically, this chapter describes the studies examining
the effect on GLIl mRNA level in two different cancer cells lines upon overexpression or
knockdown of CRD-BP. The two cell lines used were HCT116 human colon cancer cells and
T47D human breast cancer cells. The mRNA levels of GLIl and CRD-BP were measured using
RT-qPCR technique. 2’O-methyl modified RNA oligonucleotides (IDT) were used to examine
their effects on GLIl mRNA levels in cells.
4.1 Methodology
4.1.1 Detection of GLIl mRNA levels in cell lines
Using RT-qPCR, the basal levels of GLIl mRNA were measured in seven different cell lines
which include HCT116, MCF-7, SKOV3, HT-29, SKVCR 2.0, MDA-MB-231 and T47D.
4.1.1.1 Plating of cells in 6-well plates
Cells were plated in 6-well plates at the density of 2.5* 104 cells/mL. Three wells were
designated for each cell line. Cells were cultured in Minimum Essential Media (Life
Technologies) containing 10% Fetal Bovine Serum. Plates were incubated at 37°C in 5% CO 2
for ~3 days.
4.1.1.2 Total RNA extraction

65

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

Total RNA was extracted from cells using the nnWana™ miRNA Isolation Kit (Life
Technologies). The cells were first washed with 2 mL of PBS per well. 200 pL of lysis buffer
was added into each well and cells lysate was pipetted up and down to mix well. Further mixing
was done by using SP Vortex. 60 pL of homogenate additive was added to the above lysate
followed by incubation on ice for 10 minutes. Acid-phenol: chloroform (600 pL) was added to
the lysate, vortexed and spun at 10,000 x g for 5 min. The aqueous layer was collected and 100%
ethanol was added to it. The solution was then loaded onto the mirVana Filter Cartridge to
immobilize the RNA. RNA was washed with wash solutions and at the end 95°C heated nuclease
free water was added to the center of the column. RNA was eluted by spinning it at 10,000 x g
spin for 1 minute. RNA was quantified using Nano Drop.
4.1.1.3 DNase treatment
Total RNA extracted from seven cell lines were DNase treated using Ambion® DNA free™ kit.
One pg total RNA was added with 1 pL DNAsel buffer, 1 pL DNasel and sufficient amount of
nuclease free water to make up the total volume 10 pL. Reaction was incubated at 37°C for 2030 minutes. One pL DNase inactivation reagent was added and mixed. Reaction mixture was
incubated at room temperature for 2 minutes with occasional mixing. Centrifugation at 10,000 *
g was performed for 1.5 minutes and RNA was collected.
4.1.1.4 cDNA synthesis
cDNA was synthesized using BIO-RAD iScript™ cDNA Synthesis Kit. DNase-treated RNA
was mixed with 4 pL 5 * iScript mix, 1 pL reverse transcriptase enzyme and nuclease free water.
Total reaction volume was 20 pL. Synthesis of cDNA was performed using thermo cycler. The
program used to make first strand cDNA was 25°C-5 minutes, 42°C -30 minutes, 85°C-5
minutes, 4°C forever.

66

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

4.1.1.5 RT-qPCR
Following cDNA synthesis, RT-qPCR reactions were mixed by adding 6 pL cDNA, 12.5 pL
iQ™ SYBR® Green Supermix (Bio-Rad), 1.0 pL GLIl forward primer (10 pM), 1.0 pL GLIl
reverse primer (10 pM) and 4.5 pL nuclease-free water. GLIl gene primer sequences are shown
in Table?? The reaction mixtures (25 pL) were added to 96-well PCR plate (Axygen Scientific
Inc) and placed in a Bio-Rad iQ5 cycler where the cDNA was amplified using a two-step PCR
reaction. G U I mRNA level was analysed using iQ5, version 2.0 software. Prior to all RT-qPCR
experiments, optimal temperature for GLIl primers was determined using melt-curve analysis.

4.1.2 Knockdown CRD-BP in HCT116 and T47D
CRD-BP knocked down experiments were done using siRNA against CRD-BP in both
HCT116 and T47D cells. After CRD-BP knocked down GLIl mRNA level was measured by
performing qRT-PCR.
4.1.2.1 Transfecting cells with dsiRNA and performing RT-qPCR of GLIl
Cells were plated in 6-well plates as described above. After -20 hours, cells were transfected
with dsiRNA against CRD-BP. Three wells were transfected with 20 nM dsiRNA, while three
wells were transfected with 20 nM NCI using Lipofectamine 2000 reagent (Invitrogen). NCI
was used as a negative control. Following transfection, cells were incubated at 37°C in 5% CO 2
for 48 hours. Total RNA extraction was done followed by RT-qPCR to measure GLIl mRNA
level.
4.1.2.2 RT-qPCR of CRD-BP

67

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

RT-qPCR of CRD-BP was done to confirm down regulation o f CRD-BP mRNA levels after
transfecting with dsiRNA against CRD-BP. TaqMan® probes were used for detecting CRD-BP
and P-actin mRNAs. The primers and probes used are shown in Table 9. These primers and
probes were previously optimized in Dr. Lee’s laboratory. CRD-BP RT-qPCR reaction was
(ft
prepared with 12.5 pL iQ supermix (BioRad ), 1 pL forward primer (10 pM), 1 pL reverse
primer (10 pM), 1 pL Fam-TMRA probe (IDT), 8.5 pL milliQ water and 1 pL cDNA to make
up volume 25 pL. For P-actin RT-qPCR, the same method was employed except that 1 pL of
1/10 time diluted cDNA was used. The reaction mixtures were added to 96-well PCR plate
(Axygen Scientific Inc) and placed in Bio-Rad iQ5 cycler where the cDNA was amplified using
a two-step PCR reaction, fl-actin. was used as a reference gene and expression levels of CRD-BP
were normalized by N C I.
Table 10. Prim er and probe sequences of CRD-BP and p -a c tin used in qRT-PCR Taqman
approach
Target Gene
CRD-BP

Prim er Sequences_____________________________________________
Forward: 5’-AACCCTGAGAGGACCATCACT-3’
Reverse: 5 ’-AGCTGGGAAAAGACCTACAGC-3 ’
Probe: 5’-/56-FAM/TGTTGCAGGGCCGAGCAGGA/36-TAMSp/-3’
P-actin
Forward: 5 ’-TTGCCGACAGGATGCAGAAGGA-3 ’
Reverse: 5 ’-AGGTGGACAGCGAGGCCAGGAT-3 ’
Probe: 5’-/56 FAM/ATGAAGATCAAGATCATTGCTCCTCCTG/36______________ TAMSp/-3’___________________________________________________

4.1.2.3 RT-qPCR of G LIl and P -actin
RNA was DNase treated and cDNA was synthesized as described above. Following cDNA
synthesis, RT-qPCR reaction was performed as described above to measure GLIl mRNA level.
P-actin was used as a house keeping gene. In order to determine P-actin mRNA level, 1 pL 1/10
time diluted cDNA was mixed with 12.5 pL iQ™ SYBR® Green Supermix (Bio-Rad), 1.0 pL (3-

68

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

actin forward primer (10 pM), 1.0 pL P-actin reverse primer (10 pM) and 9.5 pL nuclease-free
water. P-actin primers had been previously optimized (Barnes et al, 2007). For GLIl RT-qPCR,
the same method was employed except that 6 pL cDNA were used. Difference in GLIl mRNA
level after overexpressing WT-CRD-BP was analysed using iQ5, version 2.0 software, p-actin
was used as a reference gene and expression level of GLIl mRNA was normalized with the help
o f pcDNA-FLAG-vector.
Table 11. Prim er sequences of G LIl and p -a c tin used in SYBR Green qRT-PCR method
Target Gene
G LIl

Prim er Sequence
Forward: 5 ’-CCCAATCACAAGTCAGGTTCCT-3 ’
Reverse : 5 ’-CCTATGTGAAGCCCTATTTGCC-3 ’

p -actin

Forward: 5 ’-TTGCCGACAGGATGCAGAAGGA-3 ’
Reverse: 5 ’-AGGTGGACAGCGAGGCCAGGAT-3 ’

4.1.3 Overexpressing CRD-BP in HCT116 and T47D cells
To investigate the possible regulation of GLIl mRNA expression by CRD-BP, CRD-BP was
overexpressed in both HCT116 and T47D cells. Cells were plated in 6-wells plate at the density
of 7.5x 104 cells/mL as mentioned above.
4.1.3.1 Transfecting cells with pcDNA-FLAG-CRD-BP and pcDNA-FLAG- vector
Twenty hours after plating, cells were transfected with pcDNA-FLAG-CRD-BP or pcDNAFLAG vector plasmids. PcDNA-FLAG vector was used as a negative control in this experiment.
Each well was transfected with 2 pg plasmids using Lipofectamine 2000 reagent (Invitrogen).
Following transfection, cells were incubated at 37°C in 5% CO 2 for 48 hours. Following
incubation, total RNA was extracted as described above and the quantity o f RNA was
69

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

determined using Nano drop Spectrometer. RT-qPCR of CRD-BP, GLIl and fi-actin was
performed as mentioned above. Results of RT-qPCR were generated by iQ™ 5 optical system
software (Bio-Rad) after normalizing to vector.

4.1.4 Effect of modified RNA oligonucleotides on GLIl mRNA level in cells
Upon establishing the cell line model where CRD-BP-GL/7 mRNA interaction can be
observed, the next step was to study the effect of RNA oligonucleotide capable of breaking
CRD-BP-GL/7 RNA interaction in-vitro as described in Chapter 2. Two 2’ O-methyl modified
RNA oligonucleotides were ordered from IDT. One RNA oligonucleotide (SI) was capable of
breaking interaction in-vitro while the other (S4) was unable to show any effect on CRD-BPGLI1 RNA interaction in-vitro as described in Chapter 2.
4.1.4.1 Transfecting Cells with modified RNA oligonucleotides
Cells were plated in 6-well plates as described above. After ~20 hours, cells were transfected
with 250 nM or 100 nM 2’ O-methyl modified SI or S4. Cells were also transfected with the
negative control Scrambled-Negative oligonucleotide. Primer sequences of SI and S4 are shown
in Chapter 2. Cells were incubated at 37°C in 5% CO 2 for 48 hours before subjected to total
RNA extraction as described above. qRT-PCR of GLIl, CRD-BP and f-actin was performed as
described above.
4.2 Results
4.2.1 Prim er optimization

70

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

Melt-curve analysis was performed to analyze the specificity of the GLIl primers given in
Table 10. The annealing temperature o f 60°C was used. As shown in Fig. 29, the single band
produced at ~81°C confirms the specificity of the primer.
Met Pet* Chart: Data 2012-09-181221 .opd
B4-SY0R
500

400

300

100

•100

Fig 29. Melt curve analysis of GLIl prim ers: Melt curve analyses of GLIl primers were done
at 60°C using total RNA extracted from T47D cells. Single peak of product confirms the
specificity of GLIl primers.
4.2.2. Detection of G L Il mRNA level in cell lines
The basal GLIl mRNA levels were measured in seven different cancer cell lines. Their mean
Ct values are listed in Table 11. The lower Ct value indicates higher GLIl mRNA level. Based on
the data in Table 11,1 conclude that all the seven cell lines expressed detectable level of GLIl

71

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

mRNA and can be used for further studies. SKOV3 and SKVCR2.0 appeared to express the
highest level o f GLIl mRNA, while MDA-MB-231 and MCF7 cells expressed the lowest level.

Table 12. Ct values of seven cancer cell lines
Mean Ct Values
26.65
29.53
24.04
28.01
25.28
30.66
29.50

CeU Line
HCT116
MCF-7
SKOV3
HT-29
SKVCR2.0
MDA-MB-231
T47D

4.2.3 Knockdown CRD-BP using dsiRNA
To determine if the CRD-BP-GL/7 mRNA interaction exists in T47D and HCT116 cells,
endogenous CRD-BP was knocked down with a specific dsiRNA.
A

B

NCI

siRNA

siRNA

Fig 30. Effect of knocking down CRD-BP on G L Il mRNA level in HCT116 cells:
Knockdown CRD-BP was achieved by using dsiRNA. Data was pooled from two different
biological replicates. Levels of (A) CRD-BP mRNA as well as (B) GLIl mRNA were decreased.
The Error bars represent mean standard deviation (n = 2) obtained from two biological replicates.

72

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLIl and CD44 RNA

. As shown in Fig. 30A, CRD-BP mRNA expression level was reduced by -10 fold in HCT116
cells upon transfection with the specific dsiRNA against CRD-BP. As shown in Fig. 30B, upon
knock down of CRD- BP, GLIl mRNA level was reduced about 4-fold in HCT116 cells. Mr.
Daud Akhtar, a member in Dr. Lee’s lab, had assisted me in the knocked down experiments
described in Figs 30 and 31.

A

siRNA

B

SiRNA

Fig 31. Effect of knocking down CRD-BP on G L Il mRNA level in T47D cells: Knockdown
CRD-BP was done using a specific siRNA. (A) CRD-BP as well as (B) GLIl mRNA level was
measured upon CRD-BP knocked down. The data were pooled from three biological replicates
and the error bars represent mean standard deviation (n = 3).

73

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLI1 and CD44 RNA

As shown in Fig 31, upon successful knocked down of CRD-BP to ~5 fold in T47D cells, G U I
mRNA was decreased by ~ 4 to 5-fold.

4.2.4 Overexpressing CRD-BP in HCT116 and T47D cell lines
To further investigate if CRD-BP can regulate G U I mRNA expression, I transfected cells with
plasmids carrying CRD-BP cDNA driven by the CMV promoter. This was to investigate the
effect of overexpressing CRD-BP on G U I mRNA level. Theoretically, if association of CRDEP-GU1 mRNA exists in cells, G U I mRNA level should increase. Unfortunately after
performing three biological replicates o f HCT116 cells, there was no appreciable change in G U I
mRNA level. The data from three biological replicates were pooled to make the graph shown in
Fig 32. CRD-BP mRNA levels could be checked to confirm overexpression of CRD-BP, but due
to unavailability of CRD-BP primers and Taqman probe at that time of the experiment this was
not performed.

pcDNA-FLAG v e c to r

pcDNA-FLAG-CRD-BP

Fig 32. The effect of over-expressing CRD-BP on GLI1 mRNA level in HCT116 cells:
Overexpression of CRD-BP was done by using FLAG-CRD-BP plasmid in HCT 116 cells.
Three biological replicates were pooled. The error bars represent mean standard deviation (n =
3).
However, upon overexpressing CRD-BP in T47D, the level of G U I mRNA was found to
increase by about ~7 fold. This is consistent with the model that CRD-BP binds and stabilizes
74

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLI1 and CD44 RNA

G U I mRNA transcript. The data from two biological replicates were pooled to draw the graph
shown in Fig 32.
To confirm the success o f transfection and the over-expression of CRD-BP mRNA in
transfected cells, RT-qPCR was performed to measure CRD-BP mRNA levels. Unlike in
HCT116 cells, increased expression of CRD-BP mRNA to ~300-fold was observed, confirming
the success of the overexpression experiment (Fig 33). Indeed, there was a concomitant
increased in G U I mRNA levels (~ 6.5-fold increase) in the same samples.
A

B

pcOHA-FLAG vector

pcD N A -FU M R M P

Fig 33. Overexpressing CRD-BP in T47D cells and checking CRD-BP and GLI1 mRNA
levels by doing RT-qPCR: Overexpression of CRD-BP was done by using FLAG-WT-CRDBP plasmid in T47D cells. (A) WT-CRD-BP as well as (B) G U I mRNA level was increased.
Data from two biological replicates was pooled together. Error bars represent mean standard
deviation
The data from two biological replicates was pooled to draw the graph shown in Fig 33. Taken
together, our data so far supported the existence of the CRD-BP-GL/7 mRNA interaction in
T47D human breast cancer cells.

75

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLI1 and CD44 RNA

4.2.5

Effect of 2’ O-methyl SI and S4 RNA Oligonucleotides on GLI1 mRNA levels in

cells
Having shown that the interaction between CRD-BP and GLI1 mRNA is present in T47D cells,
I then proceeded to test the RNA oligonucleotides. As shown in Fig 34, after transfecting T47D
cells with 250 nM or 100 nM 2’ O-methyl SI oligonucleotides, GLI1 mRNA level was reduced
by ~5-fold. SN was used as the negative control and expression of G U I mRNA was normalized
to that of SN. S4, an RNA oligonucleotide which could not break CRD-EP-GLI1 RNA
interaction in vitro (Chapter 2), was used as an additional negative control. As shown in Fig
34A, S4 exerted a higher expression level of G U I
A

B

Fig 34. Effect of SI and S4 2’ O-methyl RNA Oligonucleotides on GLI1 mRNA level in
T47D cells: T47D cells were transfected with Scrambled-Negative (SN), 2’ O-methyl SI or S4
RNA oligonucleotides at two different concentrations, (A) 250 nM And (B) 100 nM. G U I
mRNA levels were measured by qRT-PCR. Data shown were pooled from three biological
replicates and the error bars are standard deviation
mRNA than SN at 250 nM. Surprisingly, at 100 nM concentration, S4 reduced G U I mRNA
expression level by about ~2 to 3-fold (Fig 34B). Mr. Daud Akhtar also assisted me in
performing the experiments described in Fig 34. The data pooled from three biological replicates
were used to draw the graph shown in Fig 34.
76

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLI1 and CD44 RNA

As shown in Fig.35, after transfecting HCT116 cells with 250 nM SI oligonucleotide, there
was 40% decrease in G U I mRNA level. However, at 100 nM, SI oligonucleotide had no effect
on G U I mRNA level. At both 100 and 250 nM, had no significant effect on GLI1 mRNA level
inHCT116 cells

A

B

'C 1 .2

"D 0 .8
X I

0.6

m 0.4

SN

SI

S4

Fig 35. Effect of SI and S4 V O-methyl RNA Oligonucletides on GLI1 mRNA level in
HCT116 cells: HCT116 cells were transfected with Scrambled-Negative (SN), 2’ O-methyl SI
or S4 RNA oligonucleotides at two different concentrations, (A) 250 nM and (B) 100 nM. G U I
mRNA levels were measured by qRT-PCR. Data shown is a representative of two biological
replicates.

77

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLI1 and CD44 RNA

4.3 Discussion
CRD-BP has been shown to bind G U I mRNA and increases its stability and transcriptional
activity (Noubissi et al, 2009). To confirm such relationship between CRD-BP and GLI1 mRNA
in our own hands, I performed experiments to knockdown as well as to overexpress CRD-BP in
HCT116 and T47D cells. Upon establishing the existence of CRD-BP-GZJ/mRNA interaction in
cells, the next step was to investigate the effect of specific RNA oligonucleotide that was capable
of breaking CRD-BP-GL/7 RNA interaction in-vitro (see Chapter 2).
The first step was to choose a cell-line model to investigate CRD-BP-GZJ7 mRNA interaction
in cells. Out of the 7 cell lines examined, HCT116 and T47D cell lines were chosen for
subsequent experiments due to their higher basal G U I mRNA levels and their higher
proliferative rate. Overexpression of CRD-BP was performed by transfecting both cell lines with
the plasmid pcDNA-FLAG-CRD-BP. T47D cells showed an increase in both CRD-BP and G U I
mRNAs level suggesting that G U I mRNA levels are under the influence of CRD-BP in this cell
line. On the contrary, there was no change in G U I mRNA level in HCT116 upon over­
expression of CRD-BP.

It is possible that in HCT116 cells, G U I mRNA was already at

relatively high level and hence no further up-regulation can be observed. Alternatively, there is
no direct link between CRD-BP and G U I mRNA in HCT 116 cells.
After performing the overexpression experiments, CRD-BP knockdown experiments were
conducted to study its effect on G U I mRNA level. The results clearly demonstrated that upon
knocked down of CRD-BP mRNA, there was decrease in G U I mRNA levels in both T47D and
HCT116 cell lines, suggesting that the physical interaction between CRD-BP and G U I mRNA
exists in these cells. Hence, both cell lines were chosen for further studies on the effect of 2’ Omethyl modified RNA oligonucleotides.
78

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLI1 and CD44 RNA

HCT116 and T47D cells were transfected with two different concentrations o f 2’ O-methyl
RNA oligonucleotides SI and S4. S4 was used as a negative control in this study. At 250 nM
concentration of SI oligonucleotide, both cell lines showed reduction in G U I mRNA levels. At
100 nM Oligonucleotides, T47D cells showed reduction in G U I mRNA levels while HCT 116
did not show any change in G U I mRNA level. This suggested that T47D cells may be more
sensitive to RNA oligonucleotide than the HCT 116 cells. The results showed that treatment of
both cell types by SI RNA oligonucleotide caused decrease in G U I mRNA level. The proposed
mechanism of action of SI is that it competes with the endogenous G U I mRNA for binding to
CRD-BP, resulting in less availability of CRD-BP to bind and protect the endogenous G U I
mRNA. Hence, this subsequently led to the destabilization of G U I mRNA and reduced steadystate o f the transcript. The immuno-precipitation coupled with qRT-PCR method could be
performed to confirm that SI is truly capable of breaking CRD-BP-GL/7 mRNA interaction in
cells. In addition, the SI RNA oligonucleotide can be tested on other cell lines to determine
whether its effect can be applied broadly in cells where the CRD-BP-G U I mRNA interaction
exists. For instance, it can be assessed in the following cell lines where G U I mRNA is
detectable- MCF-7, SKOV3, HT-29, and drug resistant SKVCR 2.0.
CRD-BP has many different mRNA targets such as c-myc, CD44, Kras, MITF and MDR1, and
MAPK. Transfecting cells with the SI RNA oligonucleotides may change the level of other
mRNA targets of CRD-BP. This could be an exciting approach to confirm specificity of this
oligonucleotide.
One of the most important future studies is to determine whether there are phenotypic changes
upon treatment of cells with the SI RNA oligonucleotide. For instance, MTT growth assay or

79

Chapter 4. Assessing CRD-BP and its KH variants for ability to bind GLI1 and CD44 RNA

drug sensitivity experiments can be performed to observe potential RNA oligonucleotide effect
on growth and cell survival.

80

Chapter 5. General Discussion

Chapter 5

5.1 General discussion
The aims of this research were based on both the basic research point of view, and also the
applied research point of view. From the basic research point of view, the first goal was to
investigate the complex formation between CRD-BP and its target mRNAs. The second goal was
to determine the smallest possible fragments of these mRNAs that are capable of binding to
CRD-BP. Using UV cross-linking experiment, ~ 950 nts G U I mRNA fragment corresponding to
nts 41-990 was found to physically associate with CRD-BP (Noubissi et al, 2009). Using
electrophoretic mobility shift assay, the MDRl RNA spanning nts 746-962 was previously
shown to have high affinity for CRD-BP (Sparanese and Lee, 2007). One of the purposes of this
study was to determine the minimum size of G U I as well as MDRl RNA fragments that still
bind to CRD-BP. This was done for two reasons: (1) so that future studies can be aimed at
comparing and contrasting the structure and sequences of all the small RNA binders of CRD-BP
to see if there are any commonalities; (2) by knowing the smallest RNA binder, we can design
fluorescent RNA for use in fluorescent anisotropy experiment to conveniently study CRD-BPRNA interaction and to rapidly screen for inhibitors of CRD-BP-RNA interaction.
Another important basic question to address was the importance of each of the KH domains of
CRD-BP in binding to its target transcripts. In Dr. Lee’s lab, point mutation in all four KH
domains of CRD-BP had been successfully generated using site-directed mutagenesis (Barnes et
al, submitted). One of the goals of this thesis was focussed on assessing the binding affinity of
CRD-BP KH variants for CD44 and G U I mRNAs.
From the applied research point of view, this study mainly focussed on finding molecules that
can inhibit CKD-EP-GU1 RNA interaction. Breaking CRD-BP-GL/7 RNA interaction can be
used to specifically target CRD-BP and G U I, and would represent a novel method to down

81

Chapter 5. General Discussion

regulate their oncogenic function. This approach of inhibiting CRD-BP has some advantages.
Firstly, due its ability to interact with various mRNA targets which encode for oncoproteins, the
possibility of inhibiting several different signaling pathways involved in cancer development
appears to be a very attractive therapeutic approach. Secondly, since my goal is to inhibit
specific CRD-BP-RNA interaction the chances of inhibiting other functions performed by this
protein are minimized.
5.2 Assessing CRD-BP-RNA interaction using fluorescent EMSA
In an effort to conveniently study CRD-BP-RNA interaction and to rapidly find inhibitors of
CRD-BP-RNA interaction, I aimed to first develop the fluorescent EMSA in the lab. Fluorescent
EMSA was performed using in-vitro transcribed fluorescent MDRl RNA nts 739-922 and
various concentration of CRD-BP. Fluorescent EMSA is a well-established in-vitro method used
to detect protein-nucleic acid interaction (Steiner and Pfannschmidt 2009; Chao et al, 2010).
This non-radioactive method was employed due to its safety and cost effectiveness.
Unfortunately, the fluorescent signal of the specific MDRl RNA-CRD-BP complex was too
weak to warrant quantification purposes (Fig 13 in Chapter 2). Such effect could be the result of
non-specific aggregation of CRD-BP-RNA complex in the wells of the gel.
The fluorescein labelled UTP is a large molecule having molecular weight of 994.7 g/mol,
which is almost double the molecular weight of non-labelled UTP. Hence, the internal labelling
of transcript with fluorescein-labeled UTP had undoubtedly led to increased RNA size. Several
polyacrylamide gel concentrations (4 to 8%) were used in an attempt to resolve this issue, but
unfortunately it did not help to decrease the accumulation of protein-RNA complexes in the
wells of gels. 3’ labelling of fluorescent RNA was used as an alternate choice but this strategy
failed to generate good quality fluorescently labelled transcript (data not shown). After multiple
failed attempts, it was deemed that fluorescent EMSA was not useful in quantifying CRD-BP 82

Chapter 5. General Discussion

RNA interaction. Hence, for the rest of the work, radiolabelled EMSA was employed to study
CRD-BP-RNA interaction in-vitro.

5.3 Assessing G U I and MDRl RNA-CRD-BP interaction using [3ZP]-radiolabeled EMSA
[32P]-radiolabeled EMSA is an established method commonly used in Dr. Lee’s lab to study
protein-RNA interaction in-vitro (Sparanese and Lee; 2007, Kim et al, 2011 and 2014; Barnes et
al, 2014). [ P]-radiolabeled EMSA was performed to assess the binding ability of each fragment
of G U I and MDRl RNAs to CRD-BP. Interestingly, out of five truncated fragments of G U I
RNA each having -190 nts, only one fragment G L I230-420 showed strong binding with CRDBP as shown in Fig 14A in Chapter 2. Surprisingly, GLI 420-610 showed inability to bind CRDBP. G U 36-231, GLI 610-800 and GLI 800-990 also did not show affinity for CRD-BP as shown
in Fig 14. It was observed that intensity of free RNA of all three fragments decreased upon
addition of higher concentrations of CRD-BP as shown in Fig 14B in Chapter 2. This decrease
was mainly due to the immobility of RNA-protein aggregation in the wells at higher protein
concentration. Such protein aggregation problem was finally resolved during the course of this
work by changing the protein dialysis protocol. After successfully narrowing the binding region
of G U I RNA down to -190 nts, eight truncated fragments of GLI1 RNA nts 230-420 were invitro transcribed using [32P]-radiolabeled UTP and tested for binding to CRD-BP. These
transcripts were 3’ and 5’ truncated fragments of G U 230-420 RNA. The rationale of designing
these fragments was to locate the smallest CRD-BP-binding region on G U I RNA. After
performing [32P]-radiolabeled EMSA with all these truncated RNA fragments, GLI 320-380
region was found to be the most important region for binding CRD-BP. MFOLD generated
secondary structure of GLI 320-380 shows the presence of two stem loops. This predicted
secondary structure of GLI 320-380 was remarkably similar to MFOLD-generated secondary
structure of CD44 RNA nts 2862-2930 that binds CRD-BP (King et al, 2014). The secondary
83

Chapter 5. General Discussion

structures of both these RNAs are shown in Fig 18 in Chapter2. As the KH34 domains of CRDBP have been shown to act as di-domains (Chao et al, 2010), it is tempting to speculate that the
two stem loops of RNA interact with two different KH domains as a possible mechanism of
binding.

MDRl 746-962 nts region binds with high affinity to CRD-BP (Sparanese and Lee, 2007). The
goal was to map the smallest region of MDRl RNA that could bind CRD-BP. All 3’ and 5’
truncated RNA fragments which contain the 739-850 nts region showed either weak or strong
binding with CRD-BP. Surprisingly, MDRl 779-881 binding affinity was similar to the foil
length MDRl RNA 739-921, suggesting the importance of 779-881 nts region in bindingCRDBP.
An interesting finding from these studies was the observation that there was a general
reduction in the binding ability of smaller RNA fragments as compared to larger transcripts as
shown in Fig 15-17. GLI1 as well as MDRl RNA showed similar characteristic in that there was
reduction in binding affinity as the length of transcripts was reduced. These results are also
consistent with the CD44 RNA mapping results (King et al, 2014). X-ray crystal studies of KH3
and KH4 domains of ZBP1 suggest the looping of RNA around two KH domains as an important
step for RNA binding to ZBP1 (Chao et al, 2010). It is likely that longer RNA strand can
undergo such 180° shift and bind with both domains; while it may be more difficult for smaller
RNA fragments to orient themselves in such a way to make a connection with two KH domains.
Hence, smaller RNA fragment generally have reduced ability to bind CRD-BP.
5.4 Breaking GLI1 RNA-CRD-BP interaction in vitro
After successfully mapped GLI1 RNA region that binds CRD-BP, the next step was to find
molecules capable of breaking this physical association in-vitro. A recent paper described the
84

Chapter 5. General Discussion

first attempt to screen small molecule inhibitors of IMP 1/CRD-BP (Lily et al, 2013). However, it
was recognized that although small molecules have the potential to become therapeutic agent
they do have limitation in that they are usually toxic to cells (Lily et al, 2013). Previously,
antisense oligonucleotides have been successfully used to break the interaction between CRD-BP
and its target RNAs - c-myc (Coulis et al, 2000) and CD44 (King et al 2014). Furthermore, RNA
oligonucleotides have been considered as safe therapeutic agents (Bouchard et al, 2010). One of
the goals of this thesis was to investigate the ability of RNA/DNA oligonucleotides to
specifically break CRD-BP-GZJ7 RNA interaction.
The presence of two stem loops in the predicted RNA structure of both GLII and CD44 RNAs
that bind CRD-BP prompted me to investigate the role of these two stem loops in binding CRDBP. To test this hypothesis, RNA and DNA oligonucleotide competitors were designed using the
MFOLD web server. The first approach was to design DNA and RNA oligonucleotides having
the exact sequence and structure as that of the two stem loops region of GLI1 RNA (346-382 nts
region). Theoretically, oligonucleotide having the exact same sequence and structure should
compete with GLI1 230-420. As expected, the RNA oligonucleotide SI competed with GLU
230-420 RNA in binding to CRD-BP. The DNA oligonucleotide version, however, did not
compete with GLI 230-420 RNA, demonstrating the specificity of CRD-BP in binding RNA
only.
The next step was to investigate the role played by the individual stem loop in binding to CRDBP. To test this hypothesis, S2 to S6 RNA oligonucleotides were designed with the help o f the
MFOLD web server. S2 and S3 had one loop present, while S4, S5 and S6 had one complete
loop and one loop with reduced size. S2, S3 and S5 competed with the [32P]-radiolabeled GLI
230-420 to bind CRD-BP, but to a lesser extent than SI. This was as expected because o f the
absence or alteration to one of the stem loops and slight variations in the nucleotide sequence as
85

Chapter 5. General Discussion

compared to SI. Surprisingly, S4 and S6 were ineffective competitors. The loss of important
nucleotides from the loop region or changes in the linear part of S4 and S6 may be the
contributing factor. S7 was designed to test the importance of stem region of both stem loops to
bind CRD-BP. S7 has different nucleotide sequences in the stem region but has the same
structure as compared to SI. Surprisingly, S7 did not compete at all. The reason might be the
formation of another more stable secondary structure of the same sequence having AG greater
than S7. S8 having different nucleotides at step loom region also showed competition with [ P]radiolabeled GLI 230-420 but again to a lesser extent that SI. In summary, we conclude that not
only the RNA structure, RNA sequences also play an important role in binding to CRD-BP.
5.5 Comparing the KH variants of CRD-BP for binding to CD44 and GLI1 RNAs
The ability of CRD-BP to make physical interaction with its target transcripts is an important
step for its function. So, there is a need to thoroughly investigate the mechanism involved in the
binding of CRD-BP with its target mRNAs. To date, it has been reported that CRD-BP and its
orthologs bind with its target RNAs through its KH domains but not its RRM domains (Git and
Standart, 2002; Nielsen et al, 2002). However, it was proposed that the RRM domains might be
helpful in stabilizing the RNA-protein complexes (Nielsen et al, 2004).
This study focuses on the effect of single and double point mutation at the GXXG motif in the
KH domains of CRD-BP for binding to GLI1 and CD44 RNAs. The two CRD-BP targets, CD44
RNA nts 2862-3055 and G U I RNA nts 230-420, were chosen for this study. [32P]-radiolabeled
EMSA was performed with four single and six double mutants of CRD-BP. Y5A and D526E
point mutation in CRD-BP were used as negative controls in this study because these mutations
do not lie within the KH domains.

86

Chapter 5. General Discussion

All KH variants with single point mutation, except KH4, showed binding with [32P]radiolabeled CD44 RNA as shown in Fig 25. KH4 showed reduced binding and its Kd cannot be
determined. This suggests that the first Glycine of the G-X-X-G motif in KH4 plays an important
role in binding to CD44 RNA. All KH variants with double point mutations in the GXXG motif
of KH domains, with the exception of KH3-4, were completely incapable of binding CRD-BP.
suggesting that parts within the KH3 and KH4 domains other than the GXXG motif can play
important role in binding to CD44 RNA. There is also the possibility that with both KH3 and
KH4 mutations in G-X-X-G motif, CD44 can now change its position and bind with other
regions within KH3 and KH4 or it could bind with KH domains. The negative controls, Y5A and
D526E variants, exhibited similar or slightly enhanced binding affinity for CD44 RNA as shown
in Table 8, Chapter 3.
One of the primary concerns of this study was the structural integrity of the CRD-BP KH
variants. Circular dichroism spectral analysis showed no global secondary structural changes
after point mutations at the GXXG motifs in all KH variants (Barnes et al, submitted). However,
possible local structural changes in the KH domain upon point mutation cannot be ruled out
because of the limitation of the CD spectroscopy (Barnes et al, submitted).
As far as GLI1 RNA is concerned, KH3 single mutants showed binding, while KH1 and KH4
showed significant reduction in binding. Interestingly, KH2 did not show any binding with G U I
RNA. Reduction in binding ability of KH1, KH4 and complete abrogation of binding in case of
KH2, suggest the importance of first Glycine of G-X-X-G motif in KH1, KH2 as well as KH4
domains in binding to G U I RNA. All KH variants with double mutation showed complete
abrogation of binding.

87

Chapter 5. General Discussion

It has been previously shown that KH domains of IMP1 work in tandem to bind RNA
substrates (Chao et al, 2010). X-ray crystallography reveals the formation of pseudodimer
between KH3 and KH4 domains which orients the G-X-X-G motifs of each domain in an
antiparallel configuration to form the RNA binding surface (Chao et al., 2010).

Fig 36. IMP1 KH3-4 pseudodimer configuration of reveals P-actin RNA-binding
surfaces. Upon dimer formation, the KH domains adopt an antiparallel configuration that
requires the RNA to undergo a ~180° bend in order to contact both RNA-binding surfaces;
indicating that both G-X-X-G motifs of KH3 (Red dots) and KH4 (Blue dots) are required for Pactin RNA binding (Chao et al., 2010).
It is reasonable to think that CRD-BP binding mechanism would be similar to its orthologs
when binding RNA substrates. Therefore, a single mutation within only one of the KH domains
G-X-X-G motifs may or may not result in reduced binding because RNA may be attached with
single KH domain. KH double mutant inability to bind with both RNAs seems logical if we
accept the theory of antiparallel configuration of KH domains. The KH3-4 di-domain adopts the
antiparallel orientation that changes the RNA in linear position after binding. Similarly, it can be
assumed that the KH1-3 and KH2-4 di-domains adopt a similar configuration (Chao et al, 2010).
We can conclude that all possible di-domains combination of KH domains can be important in
binding with CRD-BP.

88

Chapter 5. General Discussion

Surprisingly, the double mutant KH3-4 binds to CD44 RNA with similar affinity as that o f the
WT-CRD-BP. KH3 and KH4 domains of ZBP1 were considered to be important in binding Pactin mRNA in-vitro (Farina et al, 2003). X-ray crystal structure o f IMP1 KH3-4 didomain
revealed the necessary RNA-binding surfaces for P-actin zipcode mRNA (Chao et a l, 2010). We
proposed that the first glycine of G-X-X-G motif might not play important role in binding to
CD44 RNA. Either there are some other binding sites within KH3-4 didomain that are
responsible for binding with CD44 or there is a possibility of binding to CD44 using the other di­
domains.
From the above discussion, it can be concluded that the first glycine of G-X-X-G motif may
not be the sole binding site for RNA. There might be multiple binding regions present within
each KH domains. Binding profiles of both CD44 and GLII RNAs are different, suggesting the
possibility of different binding sites for different RNAs within the KH domains. This study
shows that single point mutation within a KH domain may or may not reduce the binding ability
of CRD-BP. However, two mutations at two KH domains show more drastic changes in the
RNA-binding ability o f CRD-BP. Further point mutation within the KH domains may provide
further insights into the RNA-binding characteristics of CRD-BP.
5.6 Breaking of CRD-BP-GX/7 RNA interaction in cells
After I successfully found RNA oligonucleotide capable of breaking the CRD-BP- GLII RNA
interaction in vitro, my next goal was to test the oligonucleotide in cells. The most important first
step was to choose cell lines that express GLI1 mRNA at basal level. Two cell lines, HCT116
and T47D, were chosen due to their high proliferative rate and the presence of detectable GLI1
mRNA levels (Noubissi et al, 2009; Bhuvaneswari et al, 2012).
CRD-BP knocked down experiments were conducted on both cell lines. After transfecting
cells with dsiRNA against CRD-BP, there was a significant reduction in CRD-BP as well as

89

Chapter 5. General Discussion
G U I mRNA levels. This result suggests that the physical interaction between CRD-BP and
G U I mRNA exists in the two cell lines. It also supports the hypothesis of the stabilizing effect
of CRD-BP on GLI1 mRNA in the two cell lines (Noubissi et al, 2009; Noubissi et al, 2014).
Overexpression of CRD-BP using the pcDNA FLAG CRD-BP plasmid was also done on both
cell lines. While the G U I mRNA levels were increased in T47D cells, it did not change in
HCT116 cells. The overexpression of G U I mRNA in T47D cells again supported the hypothesis
of the physical association of CRD-BP with G U I mRNA and the ability of CRD-BP to protect
G U I mRNA (Noubissi et al 2009 and 2014). The results with HCT 116 cells might be due to
inability of the cells to overexpress CRD-BP or due to the saturation of CKD-EP-GLI1 mRNA
complex. CRD-BP mRNA levels were not measured in HCT116 cells due to unavailability of the
CRD-BP primers and Taqman probe at that time of the experiments.
After establishing the cell line model to study CRD-BP-GZJ7 mRNA interaction, HCT116 and
T47D cells were transfected with 2’-0-methyl modified RNA oligonucleotides. At 250 nM
concentration of SI, there was a four-fold decrease in G U I mRNA levels in T47D, while in
HCT 116 the level was reduced by less than two-fold. Differences in the reduction levels
between the two cell lines might be due to the differences in the nature of the cells. HCT-116 is
colorectal cancer cell line, while T47D is a breast cancer cell line. Also, both cell lines expressed
G U I mRNA at different levels.
Transfecting cells with 100 nM concentration of SI also showed four-fold reduction in G U I
mRNA levels in T47D cells, while there was no change in G U I mRNA levels in HCT116 cells.
This lack of effect on HCT-116 could be due to the stronger physical interaction of CRD-BPGLI1 mRNA in these cells, and that a higher competitor concentration is required.
Another surprising observation was the reduction in G U I mRNA levels after transfecting
T47D cells with 100 nM S4 oligonucleotide. This was unexpected because S4 was expected to

90

Chapter 5. General Discussion

serve as a negative control. Although this reduction level was less than the reduction levels
exerted by SI oligonucleotide, there is a possibility that lower concentration of S4 might show
competition in binding to CRD-BP or it may act via another mechanism. Nevertheless, it is not
unusual to see differences in results generated in-vitro and in cells. Experiments conducted in
vitro contain only CRD-BP, [32P]-labeled RNA and competitor oligonucleotides. However, cells
contain many other proteins and RNAs which can interfere with the function of the
oligonucleotide. In addition, there exist many different pathways in cells that might directly or
indirectly affected by the oligonucleotide which in turn affected the results.
I propose that the reduction in G U I mRNA level by SI oligonucleotide was due to its ability
to break CRD-BP-GZJ7 mRNA interaction in cells. To confirm this, immuno-precipitation
experiments using IMP1 polyclonal antibody coupled with RT-qPCR to measure the levels of
G U I mRNA physically associated with the endogenous CRD-BP will be required.
5.7 Future Direction
Studies reported in this thesis have paved many different ways to better understand the
physical interaction between CRD-BP and its RNA targets. As the binding regions of G U I and
MDRl RNAs have been mapped, one of the next steps could be the screening of small molecules
for ability to break the physical interaction between CRD-BP and G U I and MDRl RNAs using
high-throughput fluorescence polarization assay. Similar study has recently been initiated (Lily
et al, 2013). The information on the smallest RNA sequences of G U I and MDRl RNAs can now
be added to the list of RNA binders of CRD-BP. This should aid in in-silico analysis in
understanding what common structure and sequences, if any, amongst the RNA binders of CRDBP.
Few deletion studies on CRD-BP orthologs have been performed in an effort to understand the
role played by each of the KH domains in binding RNA substrates (Git and Standart, 2002;

91

Chapter 5. General Discussion

Oberman et al., 2007; Farina et al, 2003; Chao et al., 2010). Point mutation studies conducted in
Dr. Lee’s lab’s showed that c-myc and CD44 RNAs bind differently to CRD-BP (Barnes et al,
submitted). Further mutational studies involving mutation in multiple regions within a single KH
domain or multiple mutations in different KH domains can be done to get a deeper insight into
the role played by individual amino acids in binding to RNA. X-ray crystallography studies on
CRD-BP-RNA complex can be performed to observe the interaction between amino acids of
protein and nucleotides of RNA at the atomic level.
The SI RNA oligonucleotide disrupted CRD-BP-GL/7 RNA interaction in vitro and reduced
G U I mRNA levels in cells. Immunoprecipitation coupled with RT-qPCR can be performed to
confirm that the ability of SI oligonucleotide to reduce GLI1 mRNA in cells is through its ability
to disrupt CRD-BP-GZJ7 mRNA interaction.

Western blot analysis can be performed to

examine GLI1 protein levels after transfecting cells with SI oligonucleotides. MTT assay can be
performed after transfecting cells with SI, to observe possible phenotypic changes in cancer cells
which express high G U I mRNA levels. Anticancer drug resistant cells have been shown to
become drug sensitive after depletion of GLI1 (Bhuvaneswari et al, 2012). Therefore, drug
sensitivity assays can be performed after transfecting drug resistant cancer cells with SI
oligonucleotide. SI oligonucleotide can be tested in other cancer cell lines to compare its ability
to break specific CRD-BP-GL/7 interaction in different types of cancer cells. Finally, animal
studies can be performed to observe the effect of SI oligonucleotide in vivo.
There is possibility of reduced mRNA levels of other CRD-BP target mRNAs such as CD44,
c-myc, MDRl, k-ras etc. after transfecting cells with SI oligonucleotide. RT-qPCR can be
performed to determine their mRNA levels. These studies should provide information about the
specificity of the SI oligonucleotide.

92

Chapter 5. General Discussion

G U I is an important member of the Hh signaling pathway and a transcription factor that
regulates expression of other targeted genes (Lee and Fredrick, 2006). Down regulating G U I
will affect the Hh pathway as well as other related pathways under the control of G U I
expression. Hence, the strategy to break specific CRD-BP-GZJ7 RNA interaction can be
potentially used as a new therapeutic approach to treat cancer.
5.8 Concluding Remarks
In the light of above discussion, we can conclude that CRD-BP binds to different mRNA
targets distinctly. This is a very important finding because this means that specific CRD-BPmRNA interaction can be selectively disrupted by designing or discovering oligonucleotides and
small molecules. This work also shows that the disruption of CRD-BP-mRNA interaction can
result in the reduction of specific mRNA levels in cancer cells, which can ultimately lead to the
reduction of the cancer phenotype. In short, breaking CRD-BP-RNA interaction has the potential
to become a novel therapeutic strategy to treat cancer in the near future.

93

References
Adinolfi S, Bagni C, Castiglione Morelli MA, Fratemali F, Musco G, Pastore A. 1999. Novel
RNA-binding motif: The KH module. Biopolymers 51 153-164

Alberts, B., Alexander Johnson,Julian Lewis ,Martin R aff,Keith Roberts ,Peter Walter. 2008.
Molecular Biology o f The Cell, 5th ed. Garland Science: New York, NY
Ambudkar SV, Kimchi-Sarfaty C, Sauna ZE, Gottesman MM. 2003. p-glycoprotein: from
genomics to mechanism. Oncogene 22 7468-85
Anderson J.S and Parker R.P (1998) The 3' to 5' degradation of yeast mRNAs is a general
mechanism for mRNA turnover that requires the SKI2 DEVH box protein and 3' to 5'
exonucleases of the exosome complex. EMBO J 17 (5): 1497-1506
Audic, Y. and Hartley, R.S. 2004. Post-transcriptional regulation in cancer. Biol Cell, 96, 479498.
Barnes M, van Rensburg G, Li WM, Mehmood K, Mackedenski S, King DT, Miller AL, Lee
CH. (2014) Molecular insights into the CRD-BP-RNA interaction through site-directed
mutagenesis at the GXXG motif in KH domains. (Submitted)
Barnes, T., Kim, W-C., Mantha, A.K., Kim, S-E., Izumi, T., Mitra, S., Lee, C.H. 2009.
Identification of apurinic/apyrmidinic endonuclease 1 (APE1) as the endoribonuclease
that cleaves c-myc mRNA. Nucleic Acids Res 37 3946-58
Bassett T, Harpur, B., Poon, H. Y., Kuo, K. H. & Lee, C. H. 2008. Effective stimulation of
growth in MCF-7 human breast cancer cells by inhibition of syntaxinl8 by external guide
sequence and ribonuclease P. Cancer Lett 272 167-75
Bell JL, Wachter K, Muhleck B, Pazaitis N, Kohn M, Lederer M, Huttelmaier S (2013) Insulin­
like growth factor 2 mRNA-binding proteins (IGF2BPs): post-transcriptional drivers of cancer
progression? Cell Mol Life Sci 70: 2657-2675
Bonnieu, A., M. Piechaczyk, L. Marty, M. Cuny, LM. Blanchard, P. Fort, and PJeanteur. 1988.
Sequence determinants o f c-myc mRNA turnover: uifluence of 3' and 5' noncoding regions.
Oncogene Research 3 155-166
Bouchard PR, Hutabarat RM, Thompson KM. (2010) Discovery and development of therapeutic
aptamers. Annu Rev Pharmacol Toxicol 50:237-57
Boyerinas B, Park SM, Hau A, Murmann AE, Peter ME. 2010. The role o f let-7 in cells
differentiation and cancer. EndocrRelat Cancer. 17 F I9-36
Boyerinas, B. Park SM, Murmann AE, Gwin K, Montag AG, Zillhardt M, Hua YJ, Lengyel E,
Peter ME.. 2012. Let-7 modulates acquired resistance of ovarian cancer to Taxanes via IMP-1mediated stabilization of MDRl. Int J Cancer 130 (8) 1787-97

Caput, D.; Beutler, B.; Hartog, K.; Thayer, R.; Brown-Shimer, S.and Cerami, A. 1986
Identification of a common nucleotide sequence in the 3'-untranslated region of mRNA
molecules specifying inflammatory mediators. Proc Natl Acad Sci U.S.A., 83, 1670-1674
Chao, J. A. Patskovsky Y, Patel V, Levy M, Almo SC, Singer RH. 2010. ZBP1 recognition of
beta-actin zipcode induces RNA looping. Genes Dev 24 148-58
Chen, C.Y.; Gherzi, R.; Ong, S.E.; Chan, E.L.; Raijmakers, R. and Pruijn, G.J. 2001. AU binding
proteins recruit the exosome to degrade ARE-containing mRNAs. Cell, 107, 451-464.
Chi, K.N., Wallis, A.E., Lee, C.H., de Menedez, D.L, Sartor, J., Dragowska, W.H.,Mayer, L.D.
2000. Effects of Bcl2 modulation with G3139 antisense oligonucleotide on human breast cancer
cells are independent of inherent Bcl2 protein expression. Breast Cancer Res Treat 63 199-212
Cochrane DR, Spoelstra NS, Howe EN, Nordeen SK, Richer JK. 2009. MicroRNA-200c
mitigates invasiveness and restores sensitivity to microtubule-targeting chemotherapeutic agents.
Mol Cancer Ther. 8 1055-66
Coulis CM, Lee C, Nardone V, Prokipcak RD. 2000. Inhibition of c-myc expression in cells by
targeting an RNA-protein interaction using antisense oligonucleotides. Mol Pharmacol. 57(3)
485-94
Crooke. 1993. Progress toward oligonucleotide therapeutics: pharmacodynamics properties.
FASEBJ 7 533-539
Dereck Amakye, Zainab Jagani Marion Dorsch (2013) Unraveling the therapeutic potential of
the Hedgehog pathway in cancer. Nature Medicine 19: 1410-1422
Day D A and Tuite M F. 1998. Post-transcriptional gene regulatory mechanisms in eukaryotes:
Journal o f Endocrinology 157 361-371
Dimitriadis, E. Trangas T,Milatos S, FoukasP.G,GioulbasanisI, Courtis N, Nielsen F.CJPandis
N,Dafni U,Bardi G,Ioannidis P. 2007. Expression of oncofetal RNA-binding protein CRDBP/IMP1 predicts clinical outcome in colon cancer. In tJ Cancer 121 486-94
Doyle, G., Betz N.A, Leeds P.F, Fleisig A.J, Prokipcak R.D, Ross J. 1998. The c-myc coding
region determinant-binding protein: a member of a family of KH domain RNA-binding proteins.
Nuc. Acid Res. 26 5036-5044.
Doyle, G. A., Bourdeau-Heller, J. M., Coulthard, S., Meisner, L. F. & Ross, J. 2000.
Amplification in human breast cancer o f a gene encoding a c-myc mRNA-binding
protein. Cancer Res 60 2756-9
Dykxhoom, D, Palliser D, Lieberman J. 2006. The silent treatment: siRNAa as small molecule
drugs. Gene Therapy. 13 541-552.
Elcheva, I., Tarapore, R. S., Bhatia, N. & Spiegelman, V. S. 2008. Overexpression of mRNA
binding protein CRD-BP in malignant melanomas. Oncogene 27 5069-74
ii

Elcheva, I., Goswami, S., Noubissi, F. K. & Spiegelman, V. S. 2009. CRD-BP protects the
coding region o f betaTiCPl mRNA from miR-183-mediated degradation. Mol Cell 35 240-6
Farina, K. L., Huttelmaier, S., Musunuru, K., Darnell, R. and Singer, R. H. 2003. Two ZBP1 KH
domains facilitate beta-actin mRNA localization, granule formation, and cytoskeletal attachment.
J. Cell Biol 160 77-87.
Farina, L, Santis A.D, Salvucci S, Morelli G, Ruberti I. 2008. Embedding mRNA Stability in
Correlation Analysis of Time-Series Gene Expression Data. PLoS Comput Biol. 4 (8) el000141

Gameau N.L, Wilusz J.Wilusz .J.C (2007) The highways and byways ofinRNA decay. Nature 8:
113-126
Git A, Standart N (2002) The KH domains of xenopus VglRBP mediate RNA binding and self
association. RNA 8: 1319-1333.
Goswami S, Tarapore R.S, Tesla J.J, Grinblat Y, Setaluri V and Spiegelman V.S. 2010.
microRNA-340-mediated degradation of microphthalmia-associated transcription factor mRNA
is inhibited by the coding region determinant-binding protein. The journal o f biological
chemistry 285 (27) 20532-20540.
Grishin NV. 2001. KH domain: One motif, two folds. Nucleic Acids Res 29 638-643
Gu, L., Shigemasa, K. & Ohama, K. 2004. Increased expression of IGF II mRNA-binding
protein 1 mRNA is associated with an advanced clinical stage and poor prognosis in patients
with ovarian cancer. Int J Oncol 24 671-8
Gu, W., Wells, A. L., Pan, F. & Singer, R. H. 2008. Feedback regulation between zipcode
binding protein 1 and beta-catenin mRNAs in breast cancer cells. Mol Cell Biol 28 4963-74
Gu W, Pan F, Singer RH. 2009. Blocking beta-catenin binding to the ZBP1 promoter represses
ZBP1 expression, leading to increased proliferation and migration o f metastatic breast-cancer
cells. J Cell Sci 122 1895-905.
Hammer, N. A., Hansen T.V.O, Byskov A.G,De Meyts E.R,Gr0 ndahl M.L, Bredkjaer H.E,
Wewer U.M,Christiansen J, Nielsen F.C. 2005. Expression of IGF-II mRNA-binding proteins
(IMPs) in gonads and testicular cancer. Reproduction 130 203-12
Herrick D & Jacobson A 1992 A coding region segment is necessary but not sufficient for rapid
decay of the HIS3 mRNA in yeast. Gene 114 35—41
Hogan, D., RiordanD.P, Gerber A.P, HerschlagD, Brown P.O. 2008. Diverse RNA-Binding
Proteins Interact with Functionally Related Sets o f RNAs, Suggesting an Extensive Regulatory
System. PLoS Biol 6(10) e255
Huttelmaier, S., Zenklusen, D., Lederer, M., Dictenberg, J., Lorenz, M., Meng, X., Bassell, G. J,
Condeelis, J. and Singer, R. H. 2005. Spatial regulation of beta-actin translation by Srcdependent phosphorylation of ZBP1. Nature 438 512-515.
iii

Ioannidis, P, Mahaira L, Papadopoulou A, Teixeira M.R, Heim S, Andersen J.A, Evangelou E,
Dafiii U, Pandis N, Trangas T. 2003. 8q24 Copy number gains and expression of the c-myc
mRNAstabilizing protein CRD-BP in primary breast carcinomas. In tJ Cancer 104 54-9
Ioannidis, P. Kottaridi C, Dimitriadis E, Courtis N, Mahaira L, TalieriM, Giannopoulos A, Iliadis
K, Papaioannou D, Nasioulas G, Trangas T. 2004. Expression of the RNA-binding protein CRDBP in brain and nonsmallcell lung tumors. Cancer Lett 209 245-50
Ioannidis P, Mahaira L.G, Perez S. A, Gritzapis A. D, Sotiropoulou P. A,Kavalakis G.
J,Antsaklis A. I, Baxevanis C.N, Papamichail M. 2005. CRD-BP/IMP1 expression characterizes
cord blood CD34+ stem cells and affects c-myc and IGF-II expression in MCF-7 cancer cells. J
Biol Chem 280 20086-93
Kato T, Hayama S, Yamabuki T, Ishikawa N, Miyamoto M, Ito T, Tsuchiya E, Kondo S,
Nakamura Y, Daigo Y. 2007. Increased expression of insulin-like growth factor-II messenger
RNA binding protein 1 is associated with tumor progression in patients with lung cancer. Clin
Cancer Res 13 434-42
Katzenberger RJ, Marengo MS, Wassarman DA. 2009. Control of alternative splicing by signaldependent degradation of splicing-regulatory proteins. Journal o f biological chemistry. 284
10737-10746
Kim WC, Ma C, Li WM, Chohan M, Wilson DMIII, Lee CH. (2014) Altered endoribonuclease
activity of apurinic/apyrimidinic endonuclease 1 variants identified in the human population.
PLoS ONE doi:10.1371/joumal.pone.0090837.
Kim W-C, Berquist BR, Chohan M, Uy C, Wilson DM III, et al. (2011) Characterization of the
endoribonuclease active site of human apuirinic/apyrimidinic endonuclease 1. J Mol Biol 411
960-971.
King DT, Barnes M, Thomsen D, Lee CH. (2014) Assessing specific oligonucleotides and small
molecule antibiotics for the ability to inhibit the CRD-BP-CD44 RNA interaction. PLoS ONE (in
press).
Knapinska A.M, Patricia Irizarry-Barreto, Sri Adusumalli, Ioannis Androulakis, and Gary
Brewer. 2005. Molecular Mechanisms Regulating mRNA Stability: Physiological and
Pathological Significance. Current Genomics 6 (6) 471-480
Kobel, M, WeidensdorferD, ReinkeC, Lederer M, Schmitt W D , Z engK , Thomssen
C , HauptmannS, Huttelmaier S, 2007. Expression of the RNA-binding protein IMP1 correlates
with poorprognosis in ovarian carcinoma. Oncogene 26 7584-9
Laroia, G.; Cuesta, R.; Brewer, G. and Schneider, R.J. 1999. Control of mRNA decay by heat
shock-ubiquitin-proteasome pathway. Science, 284 499-502.
Leeds, P, Kren BT, Boylan JM, Betz NA, Steer CJ, Gruppuso PA, Ross J. 1997. Developmental
regulation of CRD-BP, an RNA-binding protein that stabilizes c-myc mRNA in vitro. Oncogene
14 1279-86

iv

Lemm, I. and Ross, J. 2002 Regulation of c-myc mRNA decay by translational pausing in a
coding region instability determinant. Mol Cell Biol 22 3959-3969
Levy, N, Chung S, Fumeaux H, Levy A. 1998. Hypoxic stabilization of vascular endothelial
growth factor mRNA by the RNA-binding protein HuR. Journal o f Biological Chemistry 273
6417-6423.
Lewis HA, Musunuru K, Jansen KB, Edo C, Chen H, Darnell RB, Burley SK. 2000. Sequencespecific RNA binding by a Nova KH domain: Implications for paraneoplastic disease and the
fragile X syndrome. Cell 100 323-332
Li W.M, Bames T, Lee C.H. 2010. Endoribonucleases - enzymes gaining spotlight in mRNA
metabolism. FEBS Journal 111 627-641
Liao, B, Patel M, Hu Y, Charles S, Herrick DJ, Brewer G. 2004. Targeted knockdown of the
RNA-binding protein CRD-BP promotes cell proliferation via an insulin-like growth factor IIdependent pathway in human K562 leukemia cells. J Biol Chem 279 48716-24
Lily Mahapatra, Chengjian Mao, Neal Andruska, Chen Zhang, David J. Shapiro (2013) HighThroughput Fluorescence Anisotropy Screen for Inhibitors of the Oncogenic mRNA Binding
Protein, IMP-1 Journal o f Biomolecular Screeninng DOI: 10.1177/1087057113499633
Mishra PJ, Humeniuk R, Longo-Sorbello GS, Baneijee D, Bertino JR. 2007. A miR-24
microRNA binding site polymorphism in hihydrofolate reductase gene leads to methotrexate
resistance. Proc Natl Acad Sci USA 104 13513-18
Mongroo, P. S, Noubissi F.K, Cuatrecasas M, Kalabis J, King C.E, Johnstone C.N, Bowser M.J,
Castells A, Spiegelman V.S, Rustgi A.K. 2011. IMP-1 displays cross-talk with K-Ras and
modulates colon cancer cell survival through the novel proapoptotic protein CYFIP2. Cancer
Res 71 2172-82
Muhlrad D & Parker R 1992 Mutations affecting stability and deadenylation of the yeast MFA2
transcript. Genes Development 6 2100-2 111
Mullen, T. and W. Marzluff. 2008. Degradation of histone mRNA requires oligouridylation
followed by decapping and simultaneous degration of the mRNA both 5’ to 3’ and 3’ to 5’.
Genes and Development 22(1) 50-65.
Naor D, Nedvetzki S, Golan I, Melnik L, Faitelson Y. (2002) CD44 in cancer. Crit Rev Clin Lab
Sci 39 (6) 527-79.
Nelson, D., and M. Cox. 2008. Lehninger Principles of Biochemistry, 5th ed. W.H. Freeman and
Company: New York, NY:
Nielsen FC, Nielsen J, Kristensen MA, Koch G, Christiansen J (2002) Cytoplasmic trafficking of
IGF-II mRNA binding protein by conserved KH domains. J Cell Sci 115 2087-2097.

Noubissi, F. K, Elcheva I, Bhatia N, Shakoori A, Ougolkov A, Liu J, Minamoto T, Ross J, Fuchs
S.Y, Spiegelman V.S. 2006. CRD-BP mediates stabilization of betaTrCPl and c-myc mRNA in
response to beta-catenin signalling. Nature 441 898-901
Noubissi, F. K, Goswami S, Sanek N.A, Kawakami K, Minamoto T, Moser A, Grinblat Y,
Spiegelman V.S. 2009. Wnt signaling stimulates transcriptional outcome of the Hedgehog
pathway by stabilizing GLI1 mRNA. Cancer Res 69 8572-8
Noubissi F.K, Kim T, Kawahara T, Aughenbaugh W, Berg E, Longley B, Athar M and
Spiegelman V.S. (2014). Role of CRD-BP in the Growth of Human Basal Cell Carcinoma Cells.
Journal o f Investigative Dermatology doi:10.1038/jid.2014.17
Oberman F, Kinneret R, Yisraeli J. (2007) VICKZ proteins mediate cell migration via their RNA
binding activity. RNA, 13 1558-1569.
Pan YZ, Morris ME, Yu AM. 2009. MicroRNA-328 negatively regulates the expression of
breast cancer resistance protein (BCRP/ABCG2) in human cancer cells. Mol Pharmacol 75
1374-9
Pierrat B, Lacroute F, Losson R . 1993. The 5’ untranslated region of the PPR1 regulatory gene
dictates rapid mRNA decay in yeast. Gene 131 43-51.
Prokipcak, R. D., Herrick, D. J. & Ross, J. 1994. Purification and properties of a protein that
binds to the C-terminal coding region o f human c-myc mRNA. J Biol Chem 269 9261-9
Ross, J. 1995. mRNA stability in mammalian cells. Micro biological Reviews 59 423-450
Ross, J., Lemm, I. & Berberet, B. 2001.Overexpression of an mRNA-binding protein in human
colorectal cancer. Oncogene 20 6544-50
Shaw, G. and Kamen, R. 1986. A conserved AU sequence from the 3'untranslated region of
GM-CSF mRNA mediates selective mRNA degradation. Cell, 46 659-667
Sparanese, D. & Lee, C. H. 2007. CRD-BP shields c-myc and MDR-1 RNA from
endonucleolytic attack by a mammalian endoribonuclease. Nucleic Acids Res 35 120921
Steiner S, Pfannschmidt T (2009) Fluorescence-based electrophoretic mobility shift assay in the
analysis of DNA-binding proteins. Methods Mol Biol. 419 273-89
Stohr, N., Lederer, M., Reinke, C., Meyer, S., Hatzfeld, M., Singer, R. H. and Huttelmaier, S.
2006. ZBP1 regulates mRNA stability during cellular stress. J. Cell Biol 175 527-534
To KK, Zhan Z, Litman T, Bates SE. 2008. Regulation of ABCG2 expression at the 3’
untranslated region of its mRNA through modulation of transcript stability and protein
transition by a putative microRNA in the SI colon cancer cell line. Mol Cell Biol 28 5147-61

Van Ryk, D., and S, Venkatesan. 1999. Real-time Kinetics of HIV-1 Rev-Rev Response Element
Interactions: Definition of minimal binding site on RNA and protein and stoichiometric analysis.
Journal o f Biological Chemistry 274 17452-17463
Varani G, Nagai K. (1998) RNA recognition by RNP proteins during RNA processing. Annu Rev
Biophys Biomol Struct.', 27 407-45.
Vikesaa J, Hansen T.V, Jonson L, Borup R, Wewer U.M, Christiansen J, Nielsen F.C. 2006.
RNA-binding IMPs promote cell adhesion and invadopodia formation. Em boJ 25 1456-68
Weidensdorfer D, Stohr N, Baude A, Lederer M, Kohn M, Schierhom A, Buchmeier S, Wahle
E, HiittelmaierS. 2009. Control of c-myc mRNA stability by IGF2BPl-associatedcytoplasmic
RNPs. RNA 15 104-15
Weiss, K.M. and Buchanan, A.V. 2009. The Mermaid’s Tale: Four Billion Years o f Cooperation
in the Making of Living Things, Harvard University Press.
Wicking, Carol; Smyth, Ian; Bale, Allen (1999) The hedgehog signalling pathway in
tumorigenesis and development. Oncogene 8 (55) 7844-8
Wilson, G.M.; Lu, J.; Sutphen, K.; Suarez, Y.; Sinha, S. and Brewer, B. 2003. Phosphorylation
ofp40AUFl regulates binding to A + U-rich mRNA-destabilizing elements and protein-induced
changes in ribonucleoprotein structure. J Biol Chem., 278 33039-33048.
Yap TA, Carden CP, Kaye SB. 2009. Beyond chemotherapy: targeted therapies in ovarian
cancer. Nat Rev Cancer 9 167-81
Yisraeli, J. K. 2005. VICKZ proteins: a multi-talented family of regulatory RNAbindingproteins. Biol Cell 97 87-96
Zapp, M, Stem S, Green M.R. 1993. Small molecules that selectively block RNA binding of
HIV-1 Rev protein inhibit Rev function and viral production. Cell 74 969-978.
Zhou SF. 2008. Structure, function and regulation of p-glycoprotein and its clinical relevance in
dmg disposition. Xenobiotica 38 802-32
Zhu H, Wu H, Liu X, Evans BR, Medina DJ, Liu CG, Yang JM. 2008. Role of microRNA miR27a and miR-451in the regulation of MDR1/p-glycoprotein expression in the human cancer cells.
Biochem Pharmacol 76 582-8

Appendix
Calculation of Kd values:
Percentage of bound RNA was calculated using optiquant software as shown in Fig below.

A
CRD-BP (nM)

B

0

147

294 426

735 900

0

147

294 426

C
CRD-BP (nM)

0

147

294 426

735 900

Fig 1. Optiquant software used to calculate bound and free RNA concentration.

viii

735 900

Table 1. Data showing %bound GLI1 RNA in all three gels.

Protein (nM)
0
147
294
426
735
900

%Bound
Gel A

%Bound

0
13.4
43.2

0
34.7
55.1
51.7
83.8
73.1

70
74.3
84.3

Gel B

%Bound
GeIC
0
20
48.1
46.7
72.5
90.4

Average

Standard Deviation

0
22.7
48.8

10.90366911
5.980802622

56.13333
76.86667

12.26634963
6.071518207

82.6

8.774394566

0

Table 2. Concentration of WT-CRD-BP protein and percentage of bound RNA are given in
Table below. (n=3)
Percentage Bound
22.7±10.90
48.80±5,98
56.13±12.26
76.86±6.07
82.6±8.77

WT-CRD-BP Concentration (nM)
147
294
426
735
900
100 r

% Bound

y = ({1/m1)‘ (mOAm2))/(1+({1/.
Value
1.7968
R

0.0066455

0.99646

Protein Concentration (nM)

Fig 2. Saturation binding curve of WT-CRD-BP with GLI1 mRNA based on data given in Table
2.

ix

After finding out bound and free RNA concentration, Hill equation was used to calculate IQ
value. Hill equation is given below.

bound _
1 5 5 * r

”

<+ ((£)!«)*

Hill Equation
Kd= Dissociation constant
L - CRD-BP concentration
u —Hill coefficient