Phosphorylation of KLF3 affects its DNA binding activity and biological function
Vitri Aryani Dewi Supervisor: Prof. Merlin Crossley
A thesis submitted for the fulfilment of the requirements for the degree of Doctor of Philosophy (Biochemistry and Molecular Genetics)
School of Biotechnology and Biomolecular Sciences The University of New South Wales
August 2012
PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet
Surname or Family name: DEWI
First name: VITRI Other name/s: ARYANI
Abbreviation for degree as given in the University calendar: PhD
School: BABS Faculty: SCIENCE
Title: Phosphorylation of KLF3 affects its DNA binding activity and biological function
Abstract 350 words maximum: (PLEASE TYPE) Krüppel-Like Factor 3 (KLF3) is a broadly expressed zinc-finger transcriptional repressor, which binds to CACCC-boxes and GC-rich regions in the promoters and enhancers of its target genes. Studies using knock-out mice have revealed functional roles for KLF3 in diverse tissues. Klf3-/- mice have a reduced life- span, leaner body composition, disturbed B-cell maturation and mild anaemia. This thesis explores the regulation of KLF3 function via post-translational modifications. We show that KLF3 exists as a phospho-protein in vivo and that post-translational modifications change in response to physiological stimuli. We have found that phosphorylation by Homeodomain Interacting Protein Kinase 2 (HIPK2) enhances KLF3’s DNA binding affinity. A mutant form of KLF3, in which serine 249 has been mutated to alanine has significantly reduced affinity for DNA, suggesting that phosphorylation at this site contributes to DNA binding capacity. Given that HIPK2 has been implicated in the cellular response to UV DNA damage, we investigated a potential role for KLF3 in this pathway. We found that cells lacking KLF3 have an altered response to UV stress, showing a complex phenotype indicative of a deregulated apoptotic pathway. Rescue of KLF3 null cells with wild-type KLF3, restored a normal response, while expression of a mutant version of KLF3, in which serine 249 has been mutated to alanine, showed only a partial rescue of the wild-type phenotype. Finally, we investigated the possibility of self-interaction within KLF3 using yeast two-hybrid assays and identified a putative self-association domain within the first 150 amino acids of KLF3. In conclusion, this study has identified post-translational modification of KLF3 as an important regulator of its activity. In particular, phosphorylation of serine 249 enhances KLF3’s DNA binding and plays a role in controlling its activity in biological pathways, such as the cellular response to UV stress and DNA damage. It also provides the first evidence of self-association within the KLF family.
Declaration relating to disposition of project thesis/dissertation
I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation.
I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).
…………………………………………………………… ……………………………………..……………… ……….……………………...…….… Signature Witness Date
The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.
Date of completion of requirements for Award:
ORGINALITY STATEMENT
‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of materials which have been accepted for the award of any degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, which whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project’s design and conception or in style, presentation and linguistic expression is acknowledged’
Signed
Date
ABSTRACT
Krüppel-Like Factor 3 (KLF3) is a broadly expressed zinc-finger transcriptional repressor, which binds to CACCC-boxes and GC-rich regions in the promoters and enhancers of its target genes. Studies using knock-out mice have revealed functional roles for KLF3 in diverse tissues. Klf3-/- mice have a reduced life-span, leaner body composition, disturbed B-cell maturation and mild anaemia.
This thesis explores the regulation of KLF3 function via post-translational modifications. We show that KLF3 exists as a phospho-protein in vivo and that post- translational modifications change in response to physiological stimuli. We have found that phosphorylation by Homeodomain-Interacting Protein Kinase 2 (HIPK2) enhances
KLF3’s DNA binding affinity. A mutant form of KLF3, in which serine 249 has been mutated to alanine has significantly reduced affinity for DNA, suggesting that phosphorylation at this site contributes to DNA binding capacity. Given that HIPK2 has been implicated in the cellular response to UV DNA damage, we investigated a potential role for KLF3 in this pathway. We found that cells lacking KLF3 have an altered response to UV stress, showing a complex phenotype indicative of a deregulated apoptotic pathway. Rescue of KLF3 null cells with wild-type KLF3, restored a normal response, while expression of a mutant version of KLF3, in which serine 249 has been mutated to alanine, showed only a partial rescue of the wild-type phenotype. Finally, we investigated the possibility of self-interaction within KLF3 using yeast two-hybrid assays and identified a putative self-association domain within the first 150 amino acids of
KLF3.
In conclusion, this study has identified post-translational modification of KLF3 as an important regulator of its activity. In particular, phosphorylation of serine 249 enhances KLF3’s DNA binding and plays a role in controlling its activity in biological pathways, such as the cellular response to UV stress and DNA damage. It also provides the first evidence of self-association within the KLF family.
TABLE OF CONTENTS
ACKNOWLEDGEMENT ...... i PUBLICATIONS ARISING FROM THIS THESIS...... iii ABBREVIATIONS ...... iv
CHAPTER 1 - INTRODUCTION ...... 1 1.1. Regulation of gene expression ...... 2 1.1.1. The Krüppel-Like Factor (KLF) family of transcription factors ...... 3 1.1.1.1. Krüppel-Like Factor 3 (KLF3) ...... 6 1.1.2. C-Terminal Binding Protein (CtBP) ...... 8 1.1.3. Homeodomain-Interacting Protein Kinase (HIPK) ...... 9 1.2. Post-translational modifications ...... 12 1.2.1. Phosphorylation ...... 12 1.3. Project aims ...... 14
CHAPTER 2 - MATERIALS AND METHODS ...... 15 2.1. Materials ...... 16 2.1.1. Chemicals and Reagents ...... 16 2.1.2. Enzymes ...... 21 2.1.3. Antibodies ...... 21 2.1.4. Plasmids and oligonucleotides ...... 22 2.1.4.1. Gift Plasmids ...... 22 2.1.4.2. Constructs ...... 23 2.1.4.3. Oligonucleotides ...... 25 2.1.5. Bacterial strains and culture ...... 26 2.1.6. Yeast strain and media ...... 26 2.1.7. Mammalian cell lines and culture media ...... 27 2.2. Methods ...... 28 2.2.1. General molecular biology techniques ...... 28 2.2.2. Commercial services and kits ...... 28
2.2.3. Equipment and analysis software ...... 29 2.2.4. Establishing stable rescued cell lines ...... 30 2.2.5. Bacterial protein overexpression ...... 32 2.2.5.1. Bacterial overexpression ...... 32 2.2.5.2. Protein purification of GST-fusion proteins...... 32 2.2.5.3. Protein purification of His-fusion proteins ...... 33 2.2.6. In vitro kinase assay and phosphatase treatment ...... 34 2.2.7. Electromobility-Shift Assays (EMSAs) ...... 34 2.2.8. Western Blotting ...... 36 2.2.9. UV irradiation ...... 36 2.2.10. Annexin-V assay ...... 37 2.2.11. Yeast two-hybrid assays ...... 39
CHAPTER 3 - PHOSPHORYLATION OF KRÜPPEL-LIKE FACTOR 3 AFFECTS ITS DNA BINDING ACTIVITY ...... 40 3.1. KLF3 is phosphorylated in vivo ...... 41 3.2. Dephosphorylation of KLF3 Reduces its DNA binding activity ...... 44 3.3. Serine 249 is a vital for KLF3 DNA binding affinity ...... 48 3.4. HIPK2 phosphorylates serine 249 and increases KLF3 DNA binding affinity 51 3.5. Discussion ...... 55
CHAPTER 4 - A ROLE FOR KRÜPPEL-LIKE FACTOR 3 IN APOPTOSIS ...... 57 4.1. Interacting partners of KLF3 and apoptosis ...... 58 4.2. Loss of KLF3 increases survival in murine embryonic fibroblasts ...... 60 4.3. Increased apoptosis in Klf3-/- Murine Embryonic Fibroblasts rescued with Klf3 ...... 66 4.4. Discussion ...... 70
CHAPTER 5 - DIMERISATION OF KRÜPPEL-LIKE FACTOR 3 ...... 72 5.1. Introduction ...... 73 5.2. KLF3 can self-associate in yeast two-hybrid assays ...... 75 5.3. Discussion ...... 81
CHAPTER 6 - DISCUSSION ...... 83 6.1. Summary ...... 84 6.2. Phosphorylation can modulate the activity of KLF3 ...... 85 6.3. KLF3 plays a role in apoptosis ...... 88 6.4. Possible dimerisation in KLFs ...... 91 6.5 Conclusions ...... 93
REFERENCES ...... 94
LIST OF FIGURES
Figure 1.1. Schematic diagram of Krüppel-like Factor (KLF) transcription factors ...... 3 Figure 1.2. Phylogenetic tree of KLF transcription factors highlighting a classification based on functionality and co-factors associations...... 5 Figure 1.3. Schematic representation of the KLF3/ CtBP2 repression complex ...... 7 Figure 1.4. Schematic diagram of mammalian HIPKs proteins ...... 10 Figure 2.1. Schematic diagram of EMSA results ...... 35 Figure 2.2. Apoptotic cells progression in the annexin-V assay...... 38 Figure 3.1. KLF3 is phosphorylated in vivo ...... 42 Figure 3.2. KLF3 is post-translationally modified during erythroid maturation ...... 43 Figure 3.3. Klf3 can bind to -globin promoter CACCC-box in Electrophoresis Mobility Shift Assay (EMSA) ...... 45 Figure 3.4. Phosphorylation of KLF3 enhances its DNA binding activity ...... 47 Figure 3.5. Mutation of serine 249 decreases KLF3’s DNA binding activity...... 49 Figure 3.6. HIPK2 enhances DNA binding by wildtype KLF3 but not the KLF3.S249A mutant...... 53 Figure 4.1. Western blot of nuclear extracts from Klf3-/- Murine Embryonic Fibroblasts confirms the absence of KLF3 protein...... 61 Figure 4.2. Reduced apoptosis in Murine Embryonic Fibroblast lacking KLF3...... 63 Figure 4.3. KLF3 increases PARP cleavage following UV induced cell stress...... 65 Figure 4.4. Klf3-/- Murine Embryonic Fibroblast cell lines rescued with either wildtype or mutant KLF3 show equivalent levels of KLF3 protein expression...... 66 Figure 4.5. The anti-apoptotic phenotype of Klf3-/- MEFs is reversed by expression of wildtype KLF3 and KLF3.S249E but not by KLF3.S249A...... 69 Figure 5.1. Overexpression of Klf3 in COS cells reveals a higher order complex that binds the β-globin promoter CACCC box and is recognised by anti-KLF3 antibody...... 74 Figure 5.2. KLF3 deletion mutants assayed for self-association in yeast two-hybrid assays...... 76 Table 5.1. Identification of KLF3 self-association domains by yeast two-hybrid assays. 78
Figure 5.3. Identification of KLF3 self-association domains by yeast two-hybrid assays. 78 Figure 5.4. Second round yeast two-hybrid assays...... 80 Table 5.2. Further mapping of KFL3 self-association domain...... 80
ACKNOWLEDGEMENT
First of all, I would like to thank you PapaJC who has kindly given me the opportunity to follow up my childhood dream in academia, I believe that wherever I am and whatever I do, it is by His grace.
I would like to thank you my supervisor Merlin Crossley for giving me the chance to work in his lab and for being a well of knowledge. I am also very grateful for my co-supervisor, Richard Pearson, who has guided me through every steps of my project.
Richard has invested so much time and effort in the metamorphosis of many research students. Thank you both for the support throughout my Honours and PhD years.
I would also like to mention the past and present members of the Crossley lab and
Nicolab. Thank you to Robert Czolij, our extraordinary past lab manager, Briony Jack,
Noelia Nuñez, Thanh Vu, Feyza Colakoglu, Alister Funnell, Laura Norton, Jonathan
Burdach, Crisbel Artuz, Cassie Mak, Stella Lee, Duygu Yucel, Slavica Berber, Anna Reid, and Hannah Nicholas. Special thanks for Whitaker/Lutz-mann labs who have made our transfer from Sydney to UNSW a breeze and have been very helpful for us; Noel
Whitaker, Louise Lutz-mann, Harvey Fernandez, Benjaming Heng, Vanessa Tan, Nirmani
Wijanayake. I should also mention James Krycer, the BABS’s yellow/white-pages person, thank you! I would like to thank the ‘giggling girls’ (you know who you are) who have made the lab such an enjoyable place to be. Thank you for your support in dark and lonely times, and for the cakes that we shared to cheer us up in the afternoon. Special thank you to Stella Lee who was my mentor in Honours and my friend in life.
P a g e | i
My thanks also goes to Jeff Welch and Geoffrey Kornfeld who have helped the lab with equipment and maintenance. Dawes’ and Yang’s labs who have welcomed us to
BABS/ UNSW. Especially for the tissue/ cell culture room users, it was so fun to share the room with you guys.
I would like to acknowledge our collaborators, Joel Mackay and Surya Setiyaputra who have helped me with protein structure experiments, and also Andrea Brennan for letting me use the flow cytometry facility in the Adult Cancer Program.
I am very grateful for my fiancé, Tommy, who has been very supportive of my midnight work and sudden ‘need to go to the lab’ moments. My family, who I know is always there supporting me.
I acknowledge my funding scholarship, University of Sydney International
Scholarship (UsydIS) and University International Postgraduate Award (UIPA) of UNSW.
P a g e | ii
PUBLICATIONS ARISING FROM THIS THESIS
The 24th Annual Lorne Cancer Conference, February 2012 Poster: Krüppel-Like Factor 3 and the cellular response to stress Dewi, V. A. , Pearson, R. C. M. and Crossley, M.
ComBio 2011, September 2011 Poster: Krüppel-Like Factor 3 and the cellular response to stress Dewi, V. A. , Burdach, J., Pearson, R. C. M. and Crossley, M.
OzBio 2010, September 2010 Poster: Krüppel-Like Factor 3 and the UV-response pathway Dewi, V. A. , Pearson, R. C. M. and Crossley, M.
The 31st Annual Lorne Genome Conference, February 2010 Poster: Krüppel-Like Factor 3 and the UV-response pathway Dewi, V. A. , Pearson, R. C. M. and Crossley, M.
P a g e | iii
ABBREVIATIONS
AD activation domain
AID autoinhibitory domain
ATM ataxia telangiectasia-mutated
ATP adenosine triphosphate
Bax BCL2-associated X protein
Bcl2 B-cells leukemia/lymphoma 2
Bklf basic krüppel-like factor
BSA bovine serum albumin
CtBP c-terminal binding protein
DBD DNA binding domain
DMEM dulbecco’s modified eagle media
DNA deoxyribonucleic acid
DTT dithiothreitol
E1A edenovirus early region 1A
EDTA Ethylenediaminetetraacetic acid
Eklf erythroid krüppel-like factor
EMSA electrophoretic mobility shift assay
FCS fetal calf serum
GSH glutathione
GST glutathione S-transferase
HDAC histone deacetylases
HEPES 4-(2-Hydroxyethyl)-1-piperazineethansulphonic acid
P a g e | iv
HID homeprotein-interacting domain
HIPK homeodomain-interacting protein kinase
HMT histone methyltransferases
HRP horseradish peroxidase
JAK janus kinase
KD kinase dead
KLF krüppel-like factor
LB Luria-Bertani
MDM2 murine double mutant 2
MEL Murine erythroleukemia cells
Ni-NTA nickel-nitrilotriacetic acid
NLS nuclear localisation signal
NP-40 nonidet P-40
OCT-1 octamer-binding transcription factor 1
PBS phosphate buffered saline
PEST proline, glutamic acid, serine, threonine rich
PI propidium iodide
PMSF Phenylmethylsulphonide
PSG penicillin streptomycin glutamine
PVDLT proline, valine, aspartate, leucine, threonine domain
PXDLS proline, X, aspartate, leucine, threonine domain
S249 serine 249
S249A alanine substitution of serine 249
P a g e | v
S249E glutamate substitution of serine 249
SD selective dropout
SDS sodium dodecyl sulphate
SDS-PAGE SDS-polyacrylamide gel electrophoresis
SRS speckle retention sequence
STAT signal transducer and activator of transcription
TBE tris boric acid EDTA
TBST tris buffered saline tween-20
Tris tris-hydroxymethyl-methylamine
UV ultraviolet
YH tyrosine-histidine rich
P a g e | vi
Chapter 1 - Introduction
Chapter 1 - INTRODUCTION
P a g e | 1
Chapter 1 - Introduction
Chapter 1 Introduction
1.1. REGULATION OF GENE EXPRESSION
Multicellular organisms are composed of different tissues with diverse functional roles, yet each tissue arises from a universal genetic code. The paradox whereby clonal cells are able to differentiate into multiple cell types and tissues has been studied intensively, and it is now understood that although the genetic information in a multicellular organism is in most cases identical, it is the regulation of the temporal and spatial expression of these genes that determines cell fate. Gene expression changes in response to various environmental and physiological stimuli, leading to a variety of responses including cell proliferation, differentiation or death.
The regulation of gene expression can occur at a number of levels: transcriptional, post-transcriptional, translational and post-translational. The initial process of transcription is regulated by DNA binding proteins called transcription factors that can activate and/or repress gene expression. Following transcription, gene expression can be regulated by processes that affect mRNA stability and transcript half-life, and the rate of translation. Post-translational regulation of gene expression involves the addition of chemical groups such as phosphates, glycans and ubiquitins. These post-translational modifications may affect the folding of the nascent peptide into functional protein, the half-life and stability of the protein, its interaction with binding partners, or its activity.
Protein post-translational modifications are reversible and can change in response to various stimuli, resulting in different functional properties. This thesis will focus on the
P a g e | 2
Chapter 1 - Introduction transcription factor, Krüppel-Like Factor 3 (KLF3) and how post-translational modifications control its function and activity.
1.1.1. THE KRÜPPEL-LIKE FACTOR (KLF) FAMILY OF TRANSCRIPTION FACTORS
Krüppel-Like Factors are a family of transcription factors with three conserved Cys2-
His2 zinc fingers at the carboxyl terminus of the protein (figure 1.1) (reviewed in (Turner and Crossley, 1999, Kaczynski et al., 2003, Bieker, 2001, Pearson et al., 2008)). With almost 95% similarity in their DNA binding domains, the members of this family bind to
GC-rich elements and CACCC boxes sequences in DNA. In contrast to the high similarity of the DNA binding domains, KLFs share low conservation in their N-terminal activation and/or repression domains.
Figure 1.1. Schematic diagram of Krüppel-like Factor (KLF) transcription factors Krüppel-Like Factors (KLFs) are characterised by a highly conserved DNA-binding domain containing three Cys2-His2 zinc fingers located at the C-terminus. The DNA-binding domain is conserved between family members, while the N-terminal functional domain is highly variable. The functional domain contains interaction sites for binding partners and confers activator/repressor function to these factors.
To date, 17 members of this family have been identified (figure 1.2), which can be classified into three groups, based on their main activator/repressor roles and their transcription co-factors (McConnell and Yang, 2010). Group 1 KLFs are generally thought P a g e | 3
Chapter 1 - Introduction of as transcriptional activators and this group contains: KLF1, KLF2, KLF4, KLF5, KLF6 and
KLF7; the second group consists of KLF3, KLF8 and KLF12 which act mainly as transcriptional repressors, recruiting the co-repressor C-terminal binding protein (CtBP); the third group are transcription repressors that act via the co-factor Sin3A and includes
KLF9, KLF10, KLF11, KLF13, KLF14, KLF16. Two KLFs, KLF15 and KLF17, have the least homology with other members of the family and are not classified with any of the three groups.
P a g e | 4
Chapter 1 - Introduction
Figure 1.2. Phylogenetic tree of KLF transcription factors highlighting a classification based on functionality and co-factors associations. Full-length protein sequences of murine KLFs were compared to determine evolutionary conservation. This analysis identified three groups of KLFs based on their functional roles as either activators or repressors and their common co-factors. The tree was constructed by assessing protein sequence homology using the bioinformatic analysis software program CLC main workbench. (Adapted from (McConnell and Yang, 2010)).
KLFs are involved in diverse biological processes. For example, the founding member of the KLFs, Krüppel-Like factor 1 (KLF1/EKLF), is expressed primarily in erythroid cells, where it drives erythropoiesis (Miller and Bieker, 1993, Tallack and Perkins, 2010), while
KLF4 is one of the few factors that can convert fibroblast cells into induced pluripotent cells (Shie et al., 2000, Takahashi and Yamanaka, 2006).
P a g e | 5
Chapter 1 - Introduction
1.1.1.1. Krüppel-Like Factor 3 (KLF3)
KLF3 was first identified in 1996 in erythroid cells in a screen for factors showing homology to KLF1 (Crossley et al., 1996). It was originally named Basic Krüppel-like factor (BKLF), due to the presence of a number of highly basic amino acid residues.
Erythroid cells lacking KLF1 were found to have low levels of KLF3, suggesting a role for
KLF1 in regulating expression of Klf3 (Crossley et al., 1996, Perkins et al., 1995).
Following this observation, our lab has now identified an erythroid regulatory network in which KLF1 can activate expression of Klf3 and Klf8, while KLF3 can also represses Klf8 transcription (Eaton et al., 2008).
Detailed analysis of Klf3-/- mice has begun to elucidate the biological roles of KLF3.
These animals are infertile, have a reduced life-span, and are significantly smaller and leaner than their wildtype littermates. This is due to impaired adipogenesis leading to smaller and fewer adipocytes in white adipose tissue (Sue et al., 2008). Moreover, these animals also suffer from a mild compensated anaemia, with enlarged spleens and elevated reticulocyte counts (Funnell et al., 2012). Furthermore, mice lacking KLF3 also have defects in B-cells development and homing (Vu et al., 2011, Turchinovich et al.,
2011). Taken together these studies suggest several roles for KLF3 in various biological processes including erythroid maturation, B-cell development and adipogenesis
(Pearson et al., 2011).
KLF3 is a transcriptional repressor that possesses a PVDLT motif that allows interaction with the transcriptional co-repressor, C-terminal Binding Protein (CtBP)
(Turner and Crossley, 1998). KLF3/CtBP molecular complexes recruit histone modifying factors such as histone deacetylases (HDACs) and histone methyltransferases (HMTs), P a g e | 6
Chapter 1 - Introduction which in turn modify gene accessibility to down-regulate target gene expression
(figure 1.3) (Turner and Crossley, 2001, Shi et al., 2003). Interestingly, mutation of the
PVDLT motif of KLF3, preventing its interaction with CtBP, does not completely abolish
KLF3 repressor activity, suggesting that alternative mechanisms of repression probably exist (Turner and Crossley, 1998).
Figure 1.3. Schematic representation of the KLF3/ CtBP2 repression complex KLF3 gene repression is facilitated by binding of its corepressor CtBP. Dimerisation of CtBP is followed by recruitment of histone modifying enzymes resulting in silencing of the target gene.
KLF3 can be post-translationally modified by sumoylation of lysine 10 and lysine 197.
Mutant forms of KLF3 that cannot be sumoylated show normal DNA binding affinity but have reduced repression activity (Perdomo et al., 2005), suggesting that post- translational modifications contribute to KLF3’s functional activity. Interestingly, mutations of both K10/K197 and the PVDLT motif convert KLF3 from a potent repressor into an activator of transcription (Perdomo et al., 2005).
P a g e | 7
Chapter 1 - Introduction
1.1.2. C-TERMINAL BINDING PROTEIN (CTBP)
C-terminal Binding Proteins (CtBPs) are a family of transcriptional co-repressors, first identified as binding partners of adenoviral E1a protein (Boyd et al., 1993, Schaeper et al., 1995, Turner and Crossley, 1998). CtBPs act as transcriptional co-repressors by recruiting histone modifiers such as histone deacetylases (HDACs) and histone methyltransferases (HMTs) (Shi et al., 2003). Many binding partners of CtBPs have since been identified including mammalian KLF3 and KLF8, and Drosophila Krüppel proteins
(Turner and Crossley, 1998). CtBP binds to the amino acid motif Pro-X-Asp-Leu-Ser
(PXDLS), which is present in numerous CtBP partner proteins, including KLF3. CtBP can also interact with an Arg-Arg-Thr (RRT) motif, present in addition to the PXDLS motif in partner proteins such as ZNF217 and RIZ (Quinlan et al., 2006). Mutation of these motifs abolishes the interaction of CtBP with partner proteins.
There are two members of the CtBP family in mammals, named CtBP1 and CtBP2
(Boyd et al., 1993, Schaeper et al., 1995). Mice deficient in CtBP1 and CtBP2 have many defects in development, suggesting widespread functional roles for these proteins.
Mice lacking Ctbp1 are smaller than wildtype littermates, and while some are viable and fertile, many die within 20 days of birth (Hildebrand and Soriano, 2002). In comparison,
Ctbp2 knockout mice die at embryonic day 10, due to severe defects in neural and heart development, and in body patterning (Hildebrand and Soriano, 2002). Whilst Ctbp2 heterozygous mice are largely normal, additional loss of CtBP1 leads to embryonic lethality, suggesting compensation by CtBP1, which is supported by the severe embryonic lethal phenotype of double knockout mice (Hildebrand and Soriano, 2002).
P a g e | 8
Chapter 1 - Introduction
CtBP has been implicated in p53 independent apoptosis, with Ctbp1-/-/Ctbp2-/- fibroblasts showing heightened sensitivity towards stress stimuli and increased rates of apoptosis (Grooteclaes et al., 2003). Microarray analysis has identified the pro-apoptotic factors Perp, Bax and Noxa as potential targets of CtBP mediated repression
(Grooteclaes et al., 2003). Following UV radiation or drug induced cytotoxic stress, CtBP is phosphorylated at serine 422 by a serine-threonine kinase, Homeodomain-Interacting
Protein Kinase 2 (HIPK2), resulting in its degradation (Zhang et al., 2005). Loss of CtBP then results in upregulation of pro-apoptotic genes, allowing cells to enter programmed cell death or apoptosis (Zhang et al., 2003).
1.1.3. HOMEODOMAIN-INTERACTING PROTEIN KINASE (HIPK)
To date, four Homeodomain-Interacting protein kinases (HIPKs) have been identified in mammals. The first three members of the family, HIPK1, HIPK2 and HIPK3 have high sequence homology and were discovered as binding partners of Nkx-1.2 transcription factors, whilst HIPK4 was identified by sequence similarity (figure 1.4) (Kim et al., 1998,
Arai et al., 2007). All four members carry a kinase domain (KD) and a PEST domain. PEST domains contains proline, glutamic acid, serine and threonine residues and are often found in proteins with a short half-life, suggesting this domain may promote degradation. HIPKs 1-3 are targeted to nuclear speckles via nuclear localisation (NLS) and speckle retention sequences (SRS), whilst HIPK4 is found primarily in the cytoplasm
(Arai et al., 2007). The proteins also carry a homeobox-interacting domain (HID) that is important for interaction with homeoproteins and an autoinhibitory domain (AID) which has been implicated in binding with Axin in the Wnt/β-catenin pathways (Rui et al.,
P a g e | 9
Chapter 1 - Introduction
2004). The tyrosine-rich domain (YH) at the C-terminus of the proteins is involved in the binding of HIPKs to CtBPs (Wei et al., 2007).
Figure 1.4. Schematic diagram of mammalian HIPKs proteins Four proteins have been identified and characterised as members of the HIPK family, based on the presence of homology in the kinase domain. HIPKs 1-3 share the highest sequence homology, whilst HIPK4 lacks a number of domains found in other members. The interacting domain provides binding for homeobox-proteins and the YH domain is important for CtBP binding. HIPKs 1-3 also carry a speckle retention sequences (SRS) which target the proteins into the speckles of the nucleus; HIPK4 lacks this sequence and is thus found primarily in the cytoplasm. (Adapted from (Rinaldo et al., 2008)).
The biological roles of HIPKs have been studied using knockout mice models. Mice lacking either Hipk1 or Hipk2 are basically normal, although animals lacking HIPK2 are smaller than wild-type litter mates suggesting some disrupted development (Isono et al.,
P a g e | 10
Chapter 1 - Introduction
2006). However, loss of both genes results in embryonic lethality due to neuronal defects. A single copy of Hipk2 ameliorates this phenotype and rescues the lethality, resulting in grossly normal animals. This suggests some overlapping roles between HIPK1 and HIPK2, where HIPK2 can compensate for the loss of HIPK1 (Isono et al., 2006).
HIPK2 is the most studied member of this family and has roles in organogenesis, erythropoiesis, angiogenesis, hypoxia and apoptosis. Studies using Drosophila sp. have revealed that HIPK2 regulates organogenesis via GROUCHO and PAX6 pathways, such that flies lacking HIPK2 kinase activity having abnormal eye formation (Lee et al., 2009,
Kim et al., 2006). HIPK2 also regulates cell growth and proliferation; however it remains unclear whether HIPK2 drives or represses these processes due to contradictory data from a number of studies (Wei et al., 2007, Trapasso et al., 2009). In response to stress stimuli such as UV radiation, HIPK2 is activated by phosphorylation to act as a downstream effector of ATM and ATR kinases in the phosphorylation of tumour suppressor p53 (Dauth et al., 2007, Winter et al., 2008). This specific phosphorylation targets serine 46 of p53 and directs cells towards apoptosis. As previously mentioned,
HIPK2 can also activate p53-independent apoptosis pathways via phosphorylation of serine 422 of CtBP (Zhang et al., 2003, D'Orazi et al., 2002, Zhang et al., 2005). Recently,
HIPK2 has also been implicated in erythropoiesis, with erythroid cells lacking HIPK2 showing dysregulated globin expression and erythroid maturation (Hattangadi et al.,
2010). In addition to the roles of HIPK2 mentioned above, we have previously shown in our lab that HIPK2 can bind and phosphorylate KLF3 (Lee, 2007, Kwok, 2007). HIPK2 activity is further linked to KLF3, by the observation that both factors also bind CtBP
(Turner and Crossley, 1998, Quinlan et al., 2006, Zhang et al., 2005).
P a g e | 11
Chapter 1 - Introduction
1.2. POST-TRANSLATIONAL MODIFICATIONS
Many potential post-translational modifications can be made to polypeptide chains during or after translation. These modifications can modulate the folding and structure of the proteins, facilitate or obstruct their interactions with binding partners, stabilise or mark them for degradation, and can also increase or lower their activity. Post- translational modifications involve the addition of chemical groups such as phosphate- groups, acetyl-groups, sugar-groups, ubiquitin molecules, and many others.
1.2.1. PHOSPHORYLATION
Addition of phosphate groups to serine, threonine or tyrosine residues introduces negative charges that can modify the localisation, stability, partner interactions, and ultimately the activity of a protein (reviewed in (Whitmarsh and Davis, 2000)).
Phosphorylation can affect gene expression in response to various stimuli by regulating the half-life and turn-over of proteins. In the case of CtBP1, the phosphorylation of serine 422 residue allows ubiquitination of the protein and results in its degradation
(Zhang et al., 2005), leading to de-repression of target genes. In contrast, phosphorylation can also increase the stability of proteins. Phosphorylation of serine 20 of p53 prevents binding of the ubiquitin ligase MDM2, allowing p53 to bind to the promoters of its target genes and initiate cell cycle arrest or apoptosis (Chehab et al.,
1999).
Some proteins require phosphorylation to direct them to initiate different response pathways. The tumour suppressor p53 is differentially phosphorylated in response to different stress stimuli (Nakaya et al., 2000). Phosphorylation can also promote P a g e | 12
Chapter 1 - Introduction additional post-translational modifications. In the case of p53, phosphorylation at serine
15 permits subsequent acetylation (Siliciano et al., 1997, Lambert et al., 1998). These different post-translational modifications can alter preferred DNA target promoter sequences and hence the response to a particular stimulus.
Phosphorylation can also influence the interaction between proteins. Whilst the phosphorylation of serine 20 of p53 prevents its interaction with the ubiquitin ligase
MDM2 (Chehab et al., 1999), the phosphorylation of STAT transcription factors by Janus kinase (JAK) facilitates nuclear localisation and dimerisation (reviewed in (Schindler et al., 2007)). Depending on the stimuli received by JAK, STATs can form heterodimers or homodimers to target different responsive elements in the genome.
Although many proteins have a level of basal phosphorylation, additional modifications are required to respond to various stimuli and several mechanisms have been identified to negatively regulate the signals initiated by phosphorylation in response to stimuli withdrawal. This can be achieved by the removal of phosphate- groups by phosphatases or the initiation of a negative auto-regulatory loop by the factors themselves (Meek and Anderson, 2009).
P a g e | 13
Chapter 1 - Introduction
1.3. PROJECT AIMS
This project examines how phosphorylation influences KLF3 activity both in vitro and in vivo. We have previously discovered that KLF3 is phosphorylated by HIPK2 (Kwok,
2007, Lee, 2007) and here we further analyse this interaction by investigating how phosphorylation affects DNA binding. We then examine a potential role for phosphorylation of KLF3 in the cellular response to DNA damage. Finally, we consider the possibility that multimerisation of KLF3 may occur to increase target gene specificity.
P a g e | 14
Chapter 2 – Materials and Methods
Chapter 2 - MATERIALS AND METHODS
P a g e | 15
Chapter 2 – Materials and Methods
Chapter 2 Materials and Methods
2.1. MATERIALS
2.1.1. CHEMICALS AND REAGENTS
The following is a list of the important chemicals and reagents used in the experiments alongside with their suppliers. Unless otherwise state, all of the chemicals and reagents used are molecular biology grade.
3’,3”,5’,5”-tetrabromophenolsulphonephtalein Sigma-Aldrich Company (bromophenol blue) St Louis, MI, USA
4-(2-Hydroxyethyl)-1-piperazineethansulphonic acid Sigma-Aldrich Company (HEPES)
Acetic acid Asia Pacific Specialty (APS) Chemicals, Seven Hills, NSW, Australia
Acrylamide Sigma-Aldrich Company (40% acrylamide:bis-acrylamide, 37.5:1, electrophoresis grade)
Adenosine 5’-[γ-32P] triphosphate ([γ-32P] ATP) PerkinElmer, Waltham, MA, USA
Adenosine triphosphate (ATP) Sigma-Aldrich Company
P a g e | 16
Chapter 2 – Materials and Methods
Agar Amyl Media, Dandenong, VIC, Australia
Agarose (DNA grade) Progen Industries, Darra, QLD, Australia
Ammonium persulphate (APS) Sigma-Aldrich Company
Ampicillin sodium salt Amresco Inc, Solon, Ohio, USA
Annexin V-FITC BD Biosciences, San Jose, CA, USA
Aprotinin Sigma-Aldrich Company
β-glycerolphosphate Sigma-Aldrich Company
Bacto® peptone Difco Laboratories,
Detroit, MI, USA
Bacto® yeast extract Difco Laboratories
Boric acid APS Chemicals
Calcium chloride Sigma-Aldrich Company
Casein peptone Amyl Media
Chloroform Biolab Scientific, Clayton, VIC, Austraila
Coomassie Brilliant Blue-R Sigma-Aldrich Company
Complete Supplement Mixture (CSM) Qbiogene, , Montréal, -Leu -Trp Canada
Complete Supplement Mixture (CSM) Qbiogene -His –Leu -Trp
P a g e | 17
Chapter 2 – Materials and Methods
Countess cell counting chamber slides Invitrogen-Life Technologies
Deoxynucleotide triphosphates (dNTPs) Roche Applied Science,Indianapolis, IN, USA
Dimethylsulphoxide (DMSO) Sigma-Aldrich Company
Diploma Instant Skim Milk Powder Fonterra Foodservices, North Ryde, NSW, Australia
Dithioreitol (DTT) Sigma-Aldrich Company
Drop Out Base with Agar (DOBA) Qbiogene
Dulbecco’s modified Eagle medium (DMEM) Gibco-Life Technologies, Grand Island, NY, USA
Ethanol Ajax Finechem, Thermo Fisher Scientific, Auburn, NSW, Australia
Ethidium bromide Amresco Inc
Ethylenediaminetetraacetic acid (EDTA) Ajax Finechem
Foetal calf serum (FCS) Gibco-Life Technologies
Formaldehyde Sigma-Aldrich Company
Formamide Sigma-Aldrich Company
Full Range Rainbow™ protein size standards GE Amersham Pharmacia, Little Chalfont, Buckinghamshire, UK
P a g e | 18
Chapter 2 – Materials and Methods
GeneRuler™ DNA ladder mix Fermentas, Thermo Fisher Scientific, Glen Burnie, MD, USA
Glutathione (GSH) agarose Sigma-Aldrich Company
Glycine Ajax Finechem
Imidazole Sigma-Aldrich Company
Isopropanol Ajax Finechem
Isopropyl-1-thio-β-D-galactopyranoside (IPTG) Sigma-Aldrich Company
Kodak GBX developer and fixer Kodak, Rochester, NY, USA
Leupeptin Sigma-Aldrich Company
L-Glutathione reduced Sigma-Aldrich Company
Magnesium chloride Sigma-Aldrich Company
Methanol Ajax Finechem
N,N,N’,N’-tetramethyethylenediamine (TEMED) Sigma-Aldrich Company (electrophoresis grade )
Ni-NTA agarose Invitrogen-Life Technologies
Nitrocellulose membrane Pall Corporation, Port Washington, NY, USA
Nonidet P-40 (NP-40) Sigma-Aldrich Company
NuPAGE® Bis-Tris Precast Gels Invitrogen-Life Technologies
NuPAGE® MOPS SDS Running Buffer Invitrogen-Life Technologies
P a g e | 19
Chapter 2 – Materials and Methods
Penicillin, streptomycin and glutamine solution Gibco-Life Technologies
Phenylmethylsulphonide (PMSF) Sigma-Aldrich Company
Phosphate buffered saline (PBS) tablets Sigma-Aldrich Company
Poly(dI-dC) GE Amersham Pharmacia
Polyoxyethylenesorbitanmonolaurate Sigma-Aldrich Company (Tween™-20)
Potassium chloride Ajax Finechem
Propidium Iodide Sigma-Aldrich Company
Sodium chloride Ajax Finechem
Sodium dodecyl laurate (SDS) Sigma-Aldrich Company
Sodium hydroxide Ajax Finechem
Sodium phosphate dibasic Ajax Finechem
Sodium phosphate monobasic Ajax Finechem
T-octylphenoxypolyethoxyethanol (Triton X-100) Sigma-Aldrich Company
Tris-hydroxymethyl-methylamine (Tris) Ajax Finechem
Trypan blue Invitrogen-Life Technologies
TrypLE™ Express (1X) Gibco-Life Technologies
Trypsin-EDTA (0.25%, 1X, phenol red) Gibco-Life Technologies
Yeast extract Amyl Media
Zinc sulphate Ajax Finechem
P a g e | 20
Chapter 2 – Materials and Methods
2.1.2. ENZYMES
Antarctic phosphatase (EC 3.1.3.1) New England Biolabs (NEB),Ipswich, MA, USA Lambda protein phosphatase NEB (λ-PPase, EC 3.1.3.16) Pfu DNA polymerase (EC 2.7.7.7) Stratagene, La Jolla, CA, USA Ribonuclease A (RNase A, EC 3.1.27.1) Roche Applied Science T4 DNA ligase (EC 6.5.1.1) NEB T4 polynucleotide kinase (PNK, EC 2.7.1.78) NEB Type II restriction endonucleases (EC 3.1.21.4) NEB
2.1.3. ANTIBODIES
Anti-KLF3 polyclonal antibody raised in rabbit was supplied by Thanh Vu, used at 1:1000 dilution for Western Blots (in 4% Skim milk/TBST, 4oC , overnight, gently agitated) and 1μl in EMSAs
Anti-β-ACTIN monoclonal antibody (A1978) raised in mouse was supplied by Sigma-Aldrich Company, used at 1:30,000 dilution in Western Blots (in TBST, room temperature for 1 hour)
Anti-PARP polyclonal antibody (#9542) raised in rabbit was obtained from Cell Signalling Technology, Danvers, MA, USA, used at 1:1000 dilution in Western Blots (in in 4% Skim milk/TBST, 4oC, overnight, gently agitated)
Horseradish peroxidase linked anti-rabbit (NA934) and anti-mouse (NA-931) secondary antibodies were supplied by GE Amersham Pharmacia, used at 1:30,000 dilution in Western Blots (in TBST, room temperature for 1 hour)
P a g e | 21
Chapter 2 – Materials and Methods
2.1.4. PLASMIDS AND OLIGONUCLEOTIDES
2.1.4.1. Gift Plasmids
Plasmid Provided by Description pMT2 Kaufman et al., Mammalian expression vector 1989 pMT2-Klf3 Jeremy Turner Mammalian expression vector encoding full- length murine Klf3 pcDNA3 Invitrogen-Life Mammalian expression vector Technologies pcDNA3-Klf3.WT José Perdomo Mammalian expression vector encoding full- length murine Klf3 pcDNA3- Alister Kwok Mammalian expression vector encoding full- Klf3.S249A length murine Klf3 with serine 249 mutated into alanine pET-15b Mark Wilkins Bacterial expression vector with polyhistidine (His-) tag pET-15b-Klf3.WT Dave Jackson Bacterial expression vector encoding full-length murine Klf3 with His-tag pET-15b- Stella Lee Bacterial expression vector encoding full-length Klf3.S249A murine Klf3 with serine 249 mutated into alanine followed His-tag pGEX-2T José Perdomo Bacterial expression vector with GST-tag pGEX-2T-Hipk2 Goodman et al., Bacterial expression vector encoding murine 2003 Hipk fused with GST-tag pMSCV-puro-V5 Jonathan Burdach Retroviral expression vector with additional V5- tag at the C-terminus of the multiple cloning site
P a g e | 22
Chapter 2 – Materials and Methods pMSCV-puro- Jonathan Burdach Retroviral expression vector encoding full- Klf3.WT-V5 length murine Klf3 and C-terminus V5-tag pGAD10-Bklf(1- Jeremy Turner Yeast two-hybrid construct expressing the first 268) 268 amino acids of KLF3 fused to GAL4- activation domain
2.1.4.2. Constructs
Construct Template Primers Restriction sites Klf3 BglII Fwd pMSCV-puro-Klf3.S249A-V5 pcDNA3-Klf3.S249A BglII and EcoRI Klf3 EcoRI Rev Klf3 BglII Fwd pMSCV-puro-Klf3.S249E-V5 pTRE2HygBklfS249E BglII and EcoRI Klf3 EcoRI Rev A3731 pGBT9-Klf3(1-50) A3740 A3731 pGBT9-Klf3(1-100) A3739 A3731 pGBT9-Klf3(1-150) A3738 A3731 pGBT9-Klf3(1-200) pGAD10-Bklf(1-268) A3737 BamHI and EcoRI A3731 pGBT9-Klf3(1-268) A3736 A3732 pGBT9-Klf3(51-268) A3736 A3733 pGBT9-Klf3(101-268) A3736 A3734 pGBT9-Klf3(151-268) A3736
P a g e | 23
Chapter 2 – Materials and Methods
A3735 pGBT9-Klf3(201-268) A3736 A3731 pGAD10-Klf3(1-50) A3740 A3731 pGAD10-Klf3(1-100) A3739 A3731 pGAD10-Klf3(1-150) A3738 A3731 pGAD10-Klf3(1-200) A3737 pGAD10-Bklf(1-268) BamHI and EcoRI A3731 pGAD10-Klf3(1-268) A3736 A3732 pGAD10-Klf3(51-268) A3736 A3733 pGAD10-Klf3(101-268) A3736 A3734 pGAD10-Klf3(151-268) A3736 A3735 pGAD10-Klf3(201-268) A3736
P a g e | 24
Chapter 2 – Materials and Methods
2.1.4.3. Oligonucleotides
EMSA probes
Number Sequence MC132 TAGAGCCACACCCTGGTAAG MC133 CTTACCAGGGTGTGGCTCTA
Cloning primers
Name Sequence Klf3 BglII Fwd 5'-CGTCGAAGATCTGCCACCATGCTCATGTTTGATCCAGTCC-3' Klf3 EcoRI Rev 5'-CGACGGAATTCACTAGCATGTGGCGTTTCCTGTGTAGGGCA-3' A3731 5'-GCGGATCCATGCTCATGTTTGATCCA-3' A3732 5'-GCGGATCCCCAGAAGGCCTCACTCAC-3' A3733 5'-GCGGATCCAGTCCGCCCATTAAGAAG-3' A3734 5'-GCGGATCCCAGCCCGTTCCTTTTATG-3' A3735 5'-GCGGATCCCCTGGAATTGAACCACAG-3' A3736 5'-CGGAATTCCTTATTGCACCCATCATA-3' A3737 5'-CGGAATTCAGGCTCTATTTTAATTTT-3' A3738 5'-CGGAATTCCTGGACCACGACGGGCTG-3' A3739 5'-CGGAATTCACTGGAGGAGGGCATGCT-3' A3740 5'-CGGAATTCTGGGGTCTGGAAGAACTT-3'
P a g e | 25
Chapter 2 – Materials and Methods
2.1.5. BACTERIAL STRAINS AND CULTURE
The bacterial strain used for all plasmid manipulations including cloning, miniprep and maxiprep plasmid isolation was Escherichia coli DH5α (Bethesda Research
Laboratories, Gaithersburg, MD, USA). The bacterial strain used for all expression in of
GST- and His-fusion proteins was Epicurian Coli® BL21 Escherichia Coli (Stratagene).
Both bacterial strains were culture in Luria-Bertani (LB) broth, or on LB agar plates:
o 10g/L casein peptone o 5g/L yeast extract o 10g/L sodium chloride o 15g/L bacteriological agar (for plates only)
The media was sterilised using an autoclave and cooled down to 55oC before adding ampicillin (filtered sterilised) to a final concentration of 100mg/ml.
2.1.6. YEAST STRAIN AND MEDIA
Yeast two-hybrid assays were conducted in Saccharomyces cerevisiae HF7c (MATa, ura3-52, his3-200, lys2-801, ade2-101, trp1-901, leu2-3, 112, gal4-542, gal80-538,
LYS2::GAL-HIS3, URA3::(GAL4 17-mers)3-CYC1-lacZ) (Clontech Laboratories). Yeast culture was grown in yeast-peptone-dextrose (YPD) broth on YPD plates:
o 10g/L Bacto® peptone
o 10g/L Bacto® yeast extract
o 20g/L bacteriological agar (for plates only)
The pH of the YPD media was adjusted to 5.8 and sterilised by autoclaving. Prior to use, filter-sterilised glucose was added to a final concentration of 2% (w/v). P a g e | 26
Chapter 2 – Materials and Methods
For the selection of transformed yeast in yeast two-hybrid assays, Drop Out Base with Agar (DOBA) supplemented with either (CSM) -Leu -Trp to make selective dropout
(SD) plates lacking leucine and tryptophan or (CSM) -His –Leu –Trp to make SD plates lacking leucine, tryptophan and histidine amino acids. The media was autoclaved according to manufacturer’s instructions.
2.1.7. MAMMALIAN CELL LINES AND CULTURE MEDIA
The COS monkey kidney cells were provided by M. Crossley. The Murine Embryonic
Fibroblasts (MEFs) were obtained from Klf3-/- mice embryos and the establishment of the immortalized cell lines were conducted by Crisbel Artuz. The Klf3-/- rescued MEFs were established by re-introdution of Klf3 genes using pMSCV-puro-V5 retrovirus system and EcoPack™-293 packaging cell line (Clontech Laboratories, Mountain View, CA, USA).
All cell lines were maintained in low glucose (5.56mM) Dulbecco’s modified Eagle medium (DMEM), supplemented with 10% FCS and 1X penicillin, streptomycin,
o glutamine (PSG), unless otherwise stated. Cells were incubated at 37 C with 5% CO2 in a
CO2 water jacketed incubator.
P a g e | 27
Chapter 2 – Materials and Methods
2.2. METHODS
2.2.1. GENERAL MOLECULAR BIOLOGY TECHNIQUES
Protocols for the commonly used molecular biology techniques employed in this investigation are outlined in Sambrook et al., 1989 (page reference are provided below), or were those provided by the manufacturers.
Agarose gel electrophoresis: 6.1-6.20
Agarose gel purification: 6.22-6.23
DNA ligation: 1.63-1.69
Ethanol precipitation of nucleic acid: E.3-E.4
Mini-preparations of plasmid DNA: 1.21-1.31
Gel electrophoresis of homemade polyacrylamide gels: 6.36-6.43, 6.45, 18.47- 18.55
Gel electrophoresis of NuPAGE® Novex® Bis-Tris polyacrylamide gels: Invitrogen NuPAGE® Technical Guide
Polymerase chain reaction (PCR): 14.1-14.4, 14.14-14.21
Restriction endonuclease digestion of DNA: 5.24-5.32
Transformation of competent bacterial cells: 1.74, 1.76, 1.86
2.2.2. COMMERCIAL SERVICES AND KITS
The sequencing reactions of nucleotides were carried out by the Australian Genome
Research Facility (AGRF) Sydney or Brisbane nodes.
Single stranded oligonucleotides were synthesised by Sigma-Aldrich, Castle Hill,
NSW, Australia. P a g e | 28
Chapter 2 – Materials and Methods
Any experiments involving commercial kits were conducted according to the manufacturers’ protocols. A list of the commercial kits used is listed below.
Immobilon Western Chemiluminescent HRP Substrate (ECL) (Merck Millipore, Billerica, MA, USA)
Jetstar maxi-prep kits (Genomed, Löhne. Germany)
Wizard®SV Gel and PCR Clean-Up system (Promega Corporation, Madison, WI, USA)
FuGENE®6 transfection reagent (Roche Applied Science)
2.2.3. EQUIPMENT AND ANALYSIS SOFTWARE
Bacterial cells were lysed by sonication using Sonifier® S-250D (Branson Sonic Power, Danbury, CO, USA)
Sephadex® G-25 Quick Spin™ columns (Roche Applied Sciences) were used to purify the radiolabelled EMSA probes
The EMSA gels were run in the Sturdier vertical slab gel units (Hoefer Scientific Instruments)
Fujifilm FLA5000 (Fujifilm, Minato-ku, Tokyo, Japan) was used for the visualisation of radioactive EMSA gels
ImageQuant™ (Version 3.3) software (Molecular Dynamics, Sunnyvale, CA, USA) was used to analyse the Fujifilm FLA5000- generated files
Tall Mighty Small™ vertical slab units (Hoefer, Scientific Intruments, San Francisco, CA, USA) were used for homemade polyacrylamide gels.
The western blot transfer of homemade polyacrylamide gels were done using Mighty Small™ transphor tank (Hoefer Scientific Instruments)
The NuPAGE® gels were run in Xcell SureLock mini cell electrophoresis system (Invitrogen-Life Technologies)
P a g e | 29
Chapter 2 – Materials and Methods
The X-ray films and developing reagents were supplied by Eastman Kodak Company, Rochester, NY, USA for western blot visualisation
PCR reactions were conducted using Mastercycler® (Eppendorf South Pacific, North Ryde, NSW, Australia)
Countess® automated cell counter (Invitrogen-Life Technologies) was used to count the number of cells
UVC-500 UV crosslinker (GE Amersham Pharmacia) was used to UV irradiate and induce apoptosis
The flow cytometry of annexin-V assay were done using FACSCantoII® (BD Biosciences)
Flow cytometry data were analysed using FlowJo 7.6.4 (Tree Star, Ashland, OR, USA)
Nucleotide and protein sequence analyses were carried out using CLC main workbench 6 (CLC bio, Aarhus, Denmark)
2.2.4. ESTABLISHING STABLE RESCUED CELL LINES
Wild-type and mutants Klf3 were re-introduced into Klf3-/- fibroblast cells via pMSCV-puro vectors using the murine stem cell virus methods and EcoPack™-293 packaging cell lines.
To make the virus particles, EcoPack™-293 packaging cell lines were cultured in high- glucose (25mM) DMEM with 10% (v/v) FCS and 1X PSG and seeded at a concentration of
1.5 x 106 in 60mm plates. 24 hours later, the cells were transfected with 5μg of the appropriate pMSCV-puro vectors using FuGENE®6 reagent according to the manufacturer’s protocol. The cells were then cultured for another 48 hours before the viral particles were collected.
P a g e | 30
Chapter 2 – Materials and Methods
The target cells were prepared in high glucose DMEM 16 hours before viral transduction at a concentration of 1.5 x 105 cells per 60mm plates. For the transduction, the growth media of target cells were replaced with the virus-containing media of the
EcoPack™-293 cells that has been filtered through a 0.45μm filter. Polybrene (8μg/ml) was added into the media to increase the efficiency of the transduction (Davis et al.,
2004). The cells were incubated for 24 hours in the virus-containing media, before refreshing the media with a selective high glucose DMEM (2.5μg/ml puromycin). Once the cells recover from transduction and selection, the cells were cultured in low glucose
DMEM and the expression of exogenous DNA was assessed by western blotting.
2.2.5. EXTRACTION OF NUCLEAR PROTEINS FROM MAMMALIAN CELLS
The nuclear proteins of mammalian cells were extracted using different tonicity solutions. First, cells were removed from plates using mechanical scraping or trypsin solution and washed with PBS twice to remove residual growth media. The cells were kept ice cold throughout the extraction to minimise any enzyme activity and protein degradation. To obtain cytoplasmic extract, the cells were incubated for 10 minutes at
o 4 C in 10mM HEPES pH 7.8, 1.5mM MgCl2 and 10mM KCl supplemented with 1mM
PMSF, 5mM DTT, 10ug/ml Leupeptin and 10ug/ml Aprotinin immediately before use.
The nuclear particles were isolated from the cytoplasmic extract using bench-top centrifuge.
The nuclear pellet was then resuspended in 20mM HEPES pH 7.8, 25%(w/v) glycerol,
420mM NaCl, 1.5mM MgCl2, 0.2mM EDTA supplemented with 1mM PMSF, 5mM DTT,
P a g e | 31
Chapter 2 – Materials and Methods
10ug/ml Leupeptin and 10ug/ml Aprotinin immediately before use. The samples were then incubated in ice for 20 minutes to allow the release of nuclear content. The soluble nuclear proteins were then separated from cell debris and insoluble particles by centrifugation at 12,000g, 4oC for 3 minutes. The extracts were then used immediately for analysis or stored at -80oC.
2.2.6. BACTERIAL PROTEIN OVEREXPRESSION
2.2.6.1. Bacterial overexpression
For the productions of GST- and His-fusion proteins, Epicurian Coli® BL21 E. coli was cultured in LB+ampicillin broth at 37oC, 120rpm for 18 hours (overnight). A fresh broth
(225ml) was prepared and inoculated with 25ml (10% (v/v)) of the overnight culture and further grown to reach an optical density at 600nm (OD600) of 0.5-0.6. The culture was induced by the addition of 0.1mM IPTG and 0.01mM zinc sulphate; and incubated at room temperature, 100 rpm for 18 hours (overnight). The cells were then pelleted at
3200g, 4oC for 15 minutes and stored at -20oC until further use.
2.2.6.2. Protein purification of GST-fusion proteins
The GST-fusion expressing bacterials cells were resuspended in 1/50 cell culture volume of cold GST-lysis buffer (50mM Tris pH 7.4, 150mM NaCl, 5mM EDTA, 0.5% NP-
40, supplemented prior to use with 1% Triton X-100, 1mM DTT, 0.2mM PMSF, 5μg/ml aprotinin and 5μg/ml leupeptin). The cells were then sonicated using Branson 250
Digital sonifier and micro tapered tip at 40% intensity for 1 minute (4 X 15 seconds ON/
P a g e | 32
Chapter 2 – Materials and Methods
15 seconds OFF). The soluble protein was then isolated by pelleting the insoluble debris at 3200g, 4oC for 20 minutes. The GST-fusion proteins were then incubated with
50% (v/v) GSH-agarose beads/lysis buffer slurry (1.6ml/L of culture), gently rotating at
4oC for 3 hours. The GSH-agarose beads were then washed three times with cold lysis buffer and pre-equilibrated in cold elution buffer lacking reduced L-glutathione (100mM
Tris pH 7.4, 120mM NaCl). The GST-fusion proteins were eluted off the beads with the addition of 6.2mg/ml reduced L-glutathione into of the cold elution buffer, gently rotating at 4oC for 1 hour. After purification, the proteins were quantify using A260/280 absorbance and immediately used in in vitro kinase assay and may be stored at -80oC for later use.
2.2.6.3. Protein purification of His-fusion proteins
The His-fusion expression bacterial cells were resuspended in 1/50 cell culture volume of cold His-lysis buffer (50mM NaH2PO4, 300mM NaCl, and 25mM imidazole supplemented prior to use with 1% Triton X-100, 1mM DTT, 0.2mM PMSF, 5μg/ml aprotinin and 5μg/ml leupeptin). The cells were lysed by sonication using Branson 250
Digital sonifier and micro tapered tip at 40% intensity for 1 minute (4X 15 seconds ON/
15 seconds OFF). The cell debris was pelleted by centrifugation at 3200g, 4oC for
20 minutes. The soluble fraction of the lysate was incubated in 50% (v/v) Ni-NTA- agarose/lysis buffer slurry (4ml/L culture) at 4oC for 3 hours, gently rotating. The beads were washed three times with the cold lysis buffer before the proteins were eluted off the beads with elution buffer (50mM NaH2PO4, 300mM NaCl and 250mM imidazole) gently rotating at 4oC for 1 hour. The proteins were then quantify using A260/280
P a g e | 33
Chapter 2 – Materials and Methods absorbance and must be immediately used for in vitro kinase assay as it degrades after storage.
2.2.7. IN VITRO KINASE ASSAY AND PHOSPHATASE TREATMENT
To set up the in vitro kinase reaction, equal amount of kinase (GST- or GST-HIPK) and equal amount of substrate (His-KLF3.WT, His-KLF3.S249A) were incubated in in vitro kinase buffer (20mM HEPES, 20mM MgCl2, 25mM β-glycerolphosphate) in the presence of 60μM ATP and 0.5mM DTT at 30oC for 45 minutes. Following kinase reaction, the samples were distributed equally as phosphatase treated samples and control samples.
Lambda protein phosphatase (λ-PPase) was used according to the manufacturer’s protocol. The samples were then loaded on native polyacrylamide gels for EMSAs analysis.
2.2.8. ELECTROMOBILITY-SHIFT ASSAYS (EMSAS)
Radioactive labelled DNA probe of -globin promoter was made as previously described (Crossley et al., 1996). Nuclear extract from cultured cells or bacterially overexpressed proteins were incubated with the radiolabelled probe (0.4ng) in 10mM
HEPES pH 7.8, 50mM KCl, 5mM MgCl2, 1mM EDTA, 5% glycerol, 0.33mM DTT, 33μg/ml
BSA, 3μg/ml poly(dI-dC). The samples were then loaded into 6% native polyacrylamide gel in TBE buffer (45mM Tris, 45mM boric acid, 1mM EDTA) at 250V for 1 hour and 50 minutes at 4oC. The gels were dried under vacuum and visualised using with a Fujifilm
FLA-5000 image scanner.
P a g e | 34
Chapter 2 – Materials and Methods
Proteins which bind to the radiolabeled probe appeared as a band in the phosphoimager (figure 2.1). The identity of the protein can be justified by the absence/ presence of the bands in the negative control samples. Furthermore, addition of antibody against the protein of interest results in the formation of a higher molecular weight complex of antibody/protein/probe which will migrate slower than the protein/probe band. Hence, the protein/probe band shifts towards the top of the gel as a result of antibody binding.
Figure 2.1. Schematic diagram of EMSA results The EMSA analysis visualises the binding of protein onto radiolabeled DNA. The intensity of the bands represents the amount of protein/probe complex in the samples.
P a g e | 35
Chapter 2 – Materials and Methods
2.2.9. WESTERN BLOTTING
Proteins were transferred from SDS-PAGE onto nitrocellulose membrane in 1X transfer buffer (25mM Tris, 0.2M glycine and 20% (v/v) methanol according to the apparatus manufacturers’ protocols.
Membranes were gently agitated throughout the blocking and blotting process.
Firstly, the membranes were blocked twice in 4% (w/v) skim milk/TBST (50mM Tris pH7.5, 150mM NaCl and 0.1% (v/v) Tween-20) for 15 minutes. The membranes were then washed with TBST to remove any residual skim milk solution (2 X 15 minutes) before incubation in primary antibody. The primary antibody solution was then removed after an appropriate incubation time (refer to chapter 2.1.3) and the membranes were again rinsed with TBST (5 X 9 minutes). The secondary antibody solution was then added into the membrane for 1 hour and washed away with TBST (5 X 9 minutes). Visualisation of protein bands was carried out by the addition of Immobilon Western
Chemiluminescent HRP Substrate (ECL). X-ray films were then exposed against the membrane and developed using Kodak GBX developer and fixer.
2.2.10. UV IRRADIATION
In order to induce the initiation of DNA damage response, cells were treated with
UV radiation. Murine cells were grown up to 70% confluence in low glucose-DMEM with
10% (v/v) FCS and 1X PSG on tissue culture dishes. The growth media was removed to allow UV radiation to penetrate into the cells. The cells were then exposed to lethal dose of 60J/m2 of UV-C radiation. Fresh growth media was aliquoted gently into the plates and the cells were cultured further for the appropriate period of time. P a g e | 36
Chapter 2 – Materials and Methods
After the specified period of time has lapsed, floating dead cells and live adherent cells were harvested. The media containing any floating cells were pelleted at 150g 40C for 5 minutes, and the supernatant was removed. The adherent cells were lifted off the plates using TrypLE™ Express and pooled with the floating cells by centrifugation. The cells were washed twice with ice cold PBS before stored in -80oC for protein analysis or used immediately for annexin-V assay.
2.2.11. ANNEXIN-V ASSAY
In order to investigate the importance of KLF3 in the apoptotic pathway, we assess the rate of apoptosis in cells using annexin-V assay. As the cells start undergoing the programmed cell death pathway, the cell membrane will expose phosphotidylserine groups, which are normally located in the inner leaflet of the membrane, on the outer layer of the membrane (Fadok et al., 1992). This molecule is recognised by annexin-V protein and the cells will be positively stained with annexin-V (Vermes et al., 1995).
Once the cells committed itself to death in apoptotic pathways, the integrity of the cell membrane will be compromised, allowing a non-permeable molecule to enter the cell. Propidium iodide (PI) is a non-permeable dye that intercalates with DNA and gives fluorescent (Nicoletti et al., 1991). Only cells that have entered later stages of apoptosis will integrate this dye in its DNA. Thus viable cells will progress from being double negative for both annexin-V and PI, onto early apoptotic phase marked with annexin-V, and late apoptotic/dead cells stained positively for both annexin-V and PI (figure 2.2)
(Vermes et al., 1995).
P a g e | 37
Chapter 2 – Materials and Methods
Figure 2.2. Apoptotic cells progression in the annexin-V assay. Living viable cells are negatively stained for annexin-V as the phosphotidylserine groups are only present in the inner leaflet of the membrane, and are also negatively stained for propidium iodide (PI) as this chemical is unable to cross the cell membrane (bottom left-hand side quadrant). When the cells undergo apoptotic, the phosphotidylserine groups are exposed on the outer leaflet of the cell membrane, making it available for annexin-V binding. These early apoptotic cells will be observed on the bottom right- hand side quadrant. As the apoptotic pathway was underway, the cell membrane becomes permeable to the second molecule, PI, and the DNA is also fragmented. The PI will intercalate with the DNA and thus cells at late stage of death will be positively stained with annexin-V and PI on the top right-hand side quadrant.
Cells were counted using Countess® automated cell counter and resuspended in ice-cold annexin-V binding buffer (10mM HEPES pH 7.4, 140mM NaCl, 2.5 mM CaCl2) to get 10^6 cells/ml suspension. 10^5 cells were stained with 5μl FITC-conjugated annexin-V (BD Biosciences) and 1μg/ml or propidium iodide (Sigma-Aldrich) in the dark at room temperature for 15 minutes. The cells suspension was diluted to 2 X 104 cells/ml with ice-cold annexin-V binding buffer and strained through a 35μm cell strainer into a round bottom tube (BD Biosciences) before run in FACSCantoII® flow cytometer (BD
Biosciences).
P a g e | 38
Chapter 2 – Materials and Methods
2.2.12. YEAST TWO-HYBRID ASSAYS
The yeast two-hybrid assays were carried out according to the MATCHMAKER™ two hybrid system developed by Clontech. The proteins of interests were fused with the
DNA binding domain (DBD) (pGBT9 vector) or the activating domain (AD) (pGAD10 vector) of GAL4 transcription factor. Fresh mid-log HF7c colonies were grown and transformed with the appropriate plasmids carrying the fusion proteins using lithium acetate method according to the manufacturer’s protocol. The cells were then plated on
SD plates lacking leucine and tryptophan for 3 days at 29oC to selectively grown successfully transformed cells. The colonies were then streaked onto SD plates lacking leucine, tryptophan and histidine for the same incubation parameters to examine protein-protein interaction.
P a g e | 39
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
Chapter 3 - PHOSPHORYLATION OF KRÜPPEL-LIKE FACTOR 3 AFFECTS ITS DNA BINDING ACTIVITY
P a g e | 40
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
Chapter 3 Phosphorylation of Krüppel-Like Factor 3 affects its DNA binding activity
3.1. KLF3 IS PHOSPHORYLATED IN VIVO
Krüppel-Like Factors (KLFs) are a family of transcription factors, characterised by three highly homologous zinc-fingers at the C-terminus of the protein (reviewed in
(Dang et al., 2000, Kaczynski et al., 2003)). KLFs are known to be highly post- translationally modified: acetylation of KLF1 is essential for maximal upregulation of - globin expression during erythroid differentiation (Zhang et al., 2001); ubiquitination of
KLF2 and KLF5 marks these proteins for proteosomal degradation (Zhang et al., 2004,
Chen et al., 2005); while phosphorylation enhances their nuclear localisation and activity
(Du et al., 2008, Ouyang et al., 1998). In summary, these modifications affect the activity, localisation or degradation of KLF transcription factors.
Basic Krüppel-Like Factor (BKLF/KLF3) is a broadly expressed transcriptional repressor. Studies using Klf3-/- mice have revealed a number of biological roles for KLF3.
Given the widespread function of KLF3, we decided to investigate whether post- translational modification might be a mechanism that influences KLF3 activity in specific tissues, consistent with the regulation of other KLFs. In this chapter, we explore how phosphorylation affects the DNA binding affinity and ultimately the activity of KLF3.
P a g e | 41
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
Figure 3.1. KLF3 is phosphorylated in vivo Nuclear extracts were purified from NIH/3T3 cells, incubated with (+) or without (-) λ- phosphatase, and then Western blotted with an anti-KLF3 antibody. Phosphatase treated KLF3 (lane 2) was observed to migrate notably faster than untreated KLF3 (lane 1). Multiple replicate experiments were conducted and a representative result is shown.
We first looked at the post-translational modification of endogenous KLF3 in
NIH/3T3 and MEL cells. Western blotting of nuclear extracts from murine NIH/3T3 fibroblasts confirmed that these cells express KLF3 and also suggested that the protein is present in a number of different migrating forms (figure 3.1, lane 1). When the extracts were treated with -phosphatase prior to Western blotting, a major shift in the migration of KLF3 was observed (figure 3.1, lane 2). The increased migration of the phosphatase treated sample is consistent with a reduction in the size of KLF3 resulting from a loss of phosphate groups. This observation suggests that endogenous KLF3 is phosphorylated in NIH/3T3 cells. KLF3 has a number of potential phosphorylation sites and the appearance of multiple bands in the Western blot suggests that the protein may exist in a number of different phosphorylated forms in fibroblast cells (figure 3.1 lane 2).
Similar observation was found in maturing erythroid progenitor cells (Tan, Y.M., unpublished results).
P a g e | 42
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
Figure 3.2. KLF3 is post-translationally modified during erythroid maturation Nuclear extracts were harvested from resting and DMSO-induced murine erythroleukaemia (MEL) cells and Western blotted with an antibody against KLF3. Blots were stripped and incubated with an antibody for β-actin to provide a loading control. The Western blot reveals that KLF3 purified from uninduced MEL cells (lane 1) migrates differently from that purified from induced cells (lane 2). Replicate experiments were conducted and a representative result is shown.
In addition, there is evidence for the addition of phosphate groups in response to stimuli. When MEL cells are treated with DMSO to induce erythroid maturation, the intensity and migration pattern of the KLF3 band changes (figure 3.2, lane 1). This change in intensity suggests an increase in KLF3 protein expression and/or stability, and the change in migration suggests the occurrence of post-translational modifications
(figure 3.2, lane 2). KLF3 activity is known to rise during erythroid maturation (Funnell et al., 2012) and it is possible that this is influenced by the post-translational state of the protein, as has been reported for several other members of the KLF family.
P a g e | 43
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
3.2. DEPHOSPHORYLATION OF KLF3 REDUCES ITS DNA BINDING ACTIVITY
KLF transcription factors recognise and bind to CACCC-boxes and GC-rich regions in genomic DNA (Miller and Bieker, 1993). To study how phosphorylation of KLF3 affects its
DNA binding activity, we carried out electrophoretic mobility shift assays (EMSA), using a
DNA probe containing a consensus CACCC box from the -globin promoter. We began by confirming a previous observation that KLF3 in nuclear extracts from transiently transfected COS cells, can bind with high affinity to this sequence in EMSA (figure 3.3)
(Crossley et al., 1996). As expected, we saw strong binding to the -globin CACCC probe
(figure 3.3, lane 1), with a clear supershift of the KLF3-probe complex when an antibody specific for KLF3 was added to the assay (figure 3.3, lane 2). A slower migrating band that supershifted with the anti-KLF3 antibody is also present in lane 1 (asterisk); it is possible that this band represents KLF3-containing higher order complexes. We also observed a non-specific band in the untransfected COS nuclear extracts that comigrates with KLF3 (figure 3.3, lane 3).
P a g e | 44
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
Figure 3.3. Klf3 can bind to -globin promoter CACCC-box in Electrophoresis Mobility Shift Assay (EMSA) Nuclear extracts from COS cells overexpressing KLF3 were harvested, incubated with a radiolabelled -globin CACCC-box probe, and then electrophoresed in a native polyacrylamide gel with or without anti-KLF3 antibody. The major band in lane 1, correlates to a KLF3/CACCC-box complex, which is not present in the untransfected COS nuclear extract (lane 3). This band was confirmed as KLF3 by the addition of an anti-KLF3 antibody that supershifted the migration of the Klf3/CACCC-box complex (lane 2). The asterisk represents higher molecular complex that is also shifted by the anti-KLF3. Replicate experiments were carried out and a representative result is shown.
P a g e | 45
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
Having established that we could use this approach to study KLF3’s DNA binding activity, we began our investigation into the importance of phosphorylation in this process by examining endogenous KLF3 from MEL cell extract. Nuclear extracts were harvested, dephosphorylated by -phosphatase treatment, and then compared with untreated extracts in Western blot and EMSA (figure 3.4). Following -phosphatase treatment, KLF3 appeared to migrate faster by Western blotting; this is consistent with a reduction in the mass of the protein and suggests successful dephosphorylation of KLF3
(figure 3.4a). When we analysed the same -phosphatase treated sample by EMSA, we found that dephosphorylation significantly reduced the ability of KLF3 to bind DNA
(figure 3.4b). In summary, these data suggest that endogenous KLF3 is phosphorylated in vivo in MEL cells and that this phosphorylation is required, at least in vitro, for high affinity DNA binding.
P a g e | 46
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
a.
b.
Figure 3.4. Phosphorylation of KLF3 enhances its DNA binding activity Nuclear extracts were harvested from MEL cells, treated with/without -phosphatase and the DNA binding activity was assessed by EMSA using a -globin CACCC box probe (a). Lanes 1 and 2 contained untreated controls and lane 3 contained -phosphatase treated samples. Electrophoretic migration was also analysed by Western blot (b). Again, extracts in lane 3 were treated with -phosphatase, with untreated controls in lanes 1 and 2. Phosphatase treatment of KLF3 increased its migration in Western blot and reduced its ability to bind DNA in EMSA. Multiple replicates were done and a representative results is shown.
P a g e | 47
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
3.3. SERINE 249 IS A VITAL FOR KLF3 DNA BINDING AFFINITY
Previous research in our laboratory has shown by yeast two-hybrid and co-immunoprecipitation assays that KLF3 interacts with and can be phosphorylated by the serine/threonine kinase, homeodomain-interacting protein kinase 2 (HIPK2) in vitro
(Kwok, 2007, Lee, 2007, Quinlan et al., 2006)}. This earlier work investigated the 12 potential HIPK2 phosphorylation sites on KLF3 and used mass spectrophotometry to confirm that the serine residue at position 249 (S249) is the major in vitro phosphorylation site for HIPK2. To determine if this phosphorylation site has a role in
KLF3’s DNA binding activity, we generated a mutant form of KLF3 (KLF3.S249A) in which
S249 was mutated to an alanine residue. We then overexpressed wildtype KLF3 and
KLF3.S249A in COS cells and compared their DNA binding activity in EMSA (figure 3.5a).
To confirm equivalent expression of the two constructs, we also performed Western blotting of the same nuclear extracts using an anti-KLF3 antibody (figure 3.5b). This analysis revealed that mutation of serine 249 to alanine significantly impaired the binding of KLF3 to DNA (Figure 3.5c), with the loss of phosphorylation at this site sufficient to reduce DNA binding activity by over 50%.
P a g e | 48
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
a.
b.
c.
Figure 3.5. Mutation of serine 249 decreases KLF3’s DNA binding activity. COS cells were transfected with wildtype KLF3 and KLF3 carrying a serine to alanine substitution at residue 249 (KLF3.S249A). Nuclear extracts were purified from transfected cells and analysed by EMSA using a β-globin CACCC-box probe to assess DNA binding (a). The relative expression levels of KLF3 and KLF3.S249A were also determined by Western blotting of nuclear extracts using an anti-KLF3 antibody (b). Band intensities P a g e | 49
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity were quantified using ImageJ software (Schneider et al., 2012) and used to calculate the relative normalised DNA binding activity (c). Mutation of serine 249 in KLF3 (lane 3 in all panels) reduced its DNA binding activity by over 50% compared to wildtype KLF3 (lane 2 in all panels). Multiple experiments were done and a representative result is shown. The error bars represent SEM.
P a g e | 50
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
3.4. HIPK2 PHOSPHORYLATES SERINE 249 AND INCREASES KLF3 DNA BINDING AFFINITY
To further investigate how phosphorylation of KLF3 at serine 249 affects its DNA binding activity, we decided to phosphorylate KLF3 in vitro with HIPK2, using bacterially expressed proteins. This approach allowed us to better control the phosphorylation events, by avoiding potential phosphorylation of KLF3 by the endogenous kinases that are present in mammalian cells. In these experiments we used bacterially expressed peptide, fused with GST or His-tag and purified using GSH- or Ni-NTA column, respectively. GST-tagged HIPK2 protein was used to phosphorylate either His-tagged
KLF3 or His-tagged KLF3.S249A, in which serine 249 had been mutated to alanine. We then compared binding of KLF3 and KLF3.S249A to the β-globin CACCC box probe by
EMSA (figure 3.6a). We found that bacterially expressed wildtype KLF3 bound modestly to DNA in the absence of HIPK2 and that this binding was enhanced by the addition of the kinase (figure 3.6a lanes 3 and 4). This increased binding following incubation with
HIPK2 was notably reduced if the samples were further treated with -phosphatase prior to EMSA (figure 3.6a lane 5). In contrast to wildtype KLF3, DNA binding by
KLF3.S249A did not increase following incubation with HIPK, nor was it changed by subsequent phosphatase treatment (figure 3.6b lanes 6-8).
We also examined the electrophoretic migration of the bacterially expressed KLF3 constructs by Western blotting (Figure 3.6 panel c and d). The addition of HIPK2 retarded the migration of KLF3 (figure 3.6c lanes 3 and 4) consistent with an increase in mass resulting from phosphorylation. Subsequent treatment with -phosphatase, noticeably increased the migration of KLF3 (Figure 3.6c lane 5). In contrast, migration of
P a g e | 51
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
KLF3.S249A appeared less dependent upon incubation with either HIPK or - phosphatase. These data support the hypothesis that serine 249 is a major site of phosphorylation and this increases DNA binding activity.
P a g e | 52
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
a. b.
c. d.
Figure 3.6. HIPK2 enhances DNA binding by wildtype KLF3 but not the KLF3.S249A mutant. (a and b) The DNA binding activity of bacterially expressed wildtype KLF3 (lane 3-5) and KLF3.S249A (lane 6-8) was assessed using EMSA. The effect of HIPK2 and the subsequent
P a g e | 53
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
-phosphatase treatment of these samples (lanes 4-5 and 7-8) were also recorded. The effect of phosphorylation on electrophoretic migration of bacterially expressed KLF3 and KLF3.S249A was also assessed by Western blot (c and d). Shown are protein tags (lanes 1 and 2), untreated KLF3 and KLF3.S249A (lanes 3 and 6), KLF3 and KLF3.S249A incubated with HIPK2 (lanes 4 and 7), and KLF3 and KLF3.S249A incubated with HIPK2 and then treated with -phosphatase (lanes 5 and 8). Multiple replicates and experiments were carried out, a representative result is shown.
P a g e | 54
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity
3.5. DISCUSSION
Post-translational modifications have previously been noted to affect the activity of several KLF transcription factors (Zhang et al., 2001, Ouyang et al., 1998, Zhang et al.,
2004). In this chapter, we have demonstrated that the migration of KLF3 in gel electrophoresis changes following DMSO induction of erythroid differentiation in MEL cells. This supports the hypothesis that KLF3 is post-translationally modified in vivo in response to stimuli. Our research into the effect of post-translational modifications on
KLF3 activity has determined that phosphorylation of serine 249 significantly contributes to DNA binding activity, with binding being impaired by either mutation of this residue or treatment with -phosphatase. Similar regulation of transcriptional activity by phosphorylation has been reported in other factors such as tumour suppressor p53, and the STAT transcription factors (Schindler et al., 2007, Siliciano et al., 1997).
This regulatory mechanism could act as a control switch to direct KLF3 activity at key stages in development and differentiation. Indeed, recent work in our lab has shown that Klf3 expression is upregulated by another member of the KLFs, KLF1, in the later stages of erythroid maturation and the repressive activity of KLF3 is crucial for normal erythroid development (Eaton et al., 2008, Funnell et al., 2012).
Following on from previous research in our laboratory, we have shown that the nuclear kinase HIPK2 can phosphorylate KLF3 at serine 249, thereby influencing its DNA binding activity. While DNA binding by wildtype KLF3 is enhanced by the presence of
HIPK2, no such effect is seen for a mutated form of KLF3 in which serine 249 has been mutated to an alanine. HIPK2 is known to control the activity of a number of
P a g e | 55
Chapter 3 – Phosphorylation of KLF3 affects its DNA binding activity transcription factors and the observation that it can also phosphorylate KLF3, suggests that KLF3 may have a biological role in some of the pathways that these factors regulate.
Studies have revealed that HIPK plays a significant role in apoptosis via p53-dependent and p53-independent pathways, in erythroid maturation and the regulation of globin expression and in development via WNT signalling, as well as vasculature and hypoxia responses (D'Orazi et al., 2002, Zhang et al., 2003, Hattangadi et al., 2010, Nardinocchi et al., 2009). We explored the possibility that KLF3 may be involved in one of these pathways, the apoptotic pathway, in the following chapter.
P a g e | 56
Chapter 4 – A Role for KLF3 in Apoptosis
Chapter 4 - A ROLE FOR KRÜPPEL- LIKE FACTOR 3 IN APOPTOSIS
P a g e | 57
Chapter 4 – A Role for KLF3 in Apoptosis
Chapter 4 A role for Krüppel-like Factor 3 in apoptosis
4.1. INTERACTING PARTNERS OF KLF3 AND APOPTOSIS
UV irradiation of cells can lead to DNA damage and the initiation of a stress response (Patrick, 1977, Ravanat et al., 2001), with the severity of the damage determining whether cells undergo cell cycle arrest and DNA repair, or apoptosis (Sancar et al., 2004). When the genotoxic stress is too severe and the DNA repair machinery is unable to correct the damage, apoptosis is initiated (Hartwell, 1992, Geske et al., 2000).
While the precise mechanisms that control this switch remain largely elusive, a number of the key factors with important roles have been identified, including
Homeodomain-Interacting Protein Kinase 2 (HIPK2) (Isono et al., 2006, D'Orazi et al.,
2002, Winter et al., 2008, Zhang et al., 2003, Ecsedy et al., 2003).
Following severe genotoxic stress, HIPK2 is phosphorylated by ATR/ATM kinase, a master regulator of apoptosis, which results in the activation of p53-dependent and p53-independent apoptotic pathways (Dauth et al., 2007, Winter et al., 2008). In the p53-dependent response, HIPK2 phosphorylates p53 at serine 46, which directs its activity toward target genes involved in apoptosis (D'Orazi et al., 2002). As part of the p53-independent pathway, HIPK2 also phosphorylates C-terminal Binding Protein (CtBP) at serine 422 targeting it for degradation, allowing the derepression of pro-apoptotic target genes, normally repressed by CtBP (Zhang et al., 2005, Grooteclaes et al., 2003).
Given that HIPK2 can regulate the activity of KLF3 and its co-repressor CtBP2
(Quinlan et al., 2006, Kwok, 2007), and that both HIPK2 and CtBP2 have roles in
P a g e | 58
Chapter 4 – A Role for KLF3 in Apoptosis apoptosis, we decided to investigate whether KLF3 also functions in the cellular response to UV-induced genotoxic stress.
P a g e | 59
Chapter 4 – A Role for KLF3 in Apoptosis
4.2. LOSS OF KLF3 INCREASES SURVIVAL IN MURINE EMBRYONIC FIBROBLASTS
In order to investigate a potential role for KLF3 in regulating cell survival, we decided to examine apoptosis rate in Murine Embryonic Fibroblasts (MEFs) derived from
Klf3+/+ and Klf3-/- embryos. We first confirmed that no KLF3 protein can be detected in the Klf3-/- cell line by Western blotting of nuclear extracts using an anti-KLF3 antibody
(figure 4.1). We then treated Klf3+/+ and Klf3-/- MEFs with lethal doses of UV irradiation
(60J/m2) (Huang et al., 1999) and compared the extent and rate of apoptosis in the two cell lines, using annexin-V apoptosis assays.
Annexin-V assays depend upon the ability of annexin-V to bind to phosphatidylserine, which appears on the outer surface of the plasma membrane in cells undergoing apoptosis. By using a fluorochrome conjugated annexin-V reagent, it is possible to quantify the extent of apoptosis by flow cytometry. The DNA intercalating dye propidium iodide (PI) is also included to distinguish apoptotic from dead cells, as the damaged plasma membrane of dead cells will allow PI to enter in addition to being bound by annexin-V. Hence in flow cytometry profiles of annexin-V assays, dead cells will stain with both reagents, apoptotic cells will be positive for only annexin-V, and live cells will be negative for both annexin-V and PI.
P a g e | 60
Chapter 4 – A Role for KLF3 in Apoptosis
Figure 4.1. Western blot of nuclear extracts from Klf3-/- Murine Embryonic Fibroblasts confirms the absence of KLF3 protein. Nuclear extracts from Klf3+/+ and Klf3-/- Murine Embryonic Fibroblasts were electrophoresed in a denaturing polyacrylamide gel, transferred onto a nitrocellulose membrane and stained with an anti-KLF3 antibody. To visualise the β-actin loading control, the membrane was stripped and re-blotted with an anti-β-actin antibody.
To determine the effect of loss of KLF3 on apoptosis, Klf3+/+ and Klf3-/- MEFs were UV irradiated and then stained with annexin-V and PI at various intervals during a 72 hour time course. Prior to UV-induced DNA damage, 4% of Klf3+/+ MEFs were found to be dead, while a lower figure of 1.5% was seen in the Klf3-/- line (figure 4.2). This initial observation suggested that KLF3 may have a role in cell survival, which was supported by our subsequent analysis of the two cells lines following UV irradiation. In these experiments, we found that the percentage of dead cells was significantly greater in
Klf3+/+ MEFs compared to cells that lack KLF3 at 24 and 48 hour time points (p<0.05, n=4). Indeed, 72 hours after UV irradiation, the percentage of live cells in the Klf3-/- sample (32%) was twice that of the Klf3+/+ line (16%). This phenotype was observed consistently at each time point throughout the 72 hour time course and suggested that the presence of KLF3 may be deleterious for cell survival. Although the percentage of dead cell was consistently higher in cell lines expressing or rescued with KLF3, the apoptotic populations appeared lower for these cells. It is therefore possible that the
P a g e | 61
Chapter 4 – A Role for KLF3 in Apoptosis presence of KLF3 accelerates the progression of cells through apoptosis. An additional or alternative hypothesis is that KLF3 also affects non-apoptotic pathways, thereby contributing to the increase in the total dead cell population and the apparent reduction in the apoptotic population
P a g e | 62
Chapter 4 – A Role for KLF3 in Apoptosis
Figure 4.2. Reduced apoptosis in Murine Embryonic Fibroblast lacking KLF3. Shown are flow cytometry plots of Murine Embryonic Fibroblasts irradiated with UV at a dose of 60J/m2, and subsequently harvested for annexin-V assays at 24, 48 and 72 hours post irradiation. Both floating cells and adherent cells were included in the assays. Cells P a g e | 63
Chapter 4 – A Role for KLF3 in Apoptosis were stained with FITC-conjugated annexin-V and PI to determine live, apoptotic and dead populations. Klf3+/+ cells seemed more susceptible to UV stress, with loss of Klf3 appearing to promote cell survival. Multiple replicates and experiments were conducted and a representative result is shown.
To further examine whether KLF3 has a role in the cellular response to DNA damage, we examined cleavage of the DNA repair factor Poly ADP-ribose polymerase (PARP) in
MEFs following UV irradiation. In response to DNA damage, PARP binds to single stranded breaks, which leads to its activation and subsequent DNA repair; however under conditions of extreme damage, PARP is cleaved by caspase-3 and deactivated, thereby allowing apoptosis to proceed (Kaufmann et al., 1993). Assessing the amount of cleaved PARP in a cell is therefore an indicator of the extent of apoptosis. We investigated the level of PARP cleavage in Klf3+/+ and Klf3-/- MEFs, and also in Klf3-/- MEFs rescued by forced overexpression of KLF3. While we did not notice a difference in the amount of cleaved PARP in Klf3+/+ and Klf3-/- MEFs following UV irradiation, we found that rescue of KLF3 expression in Klf3-/- cells resulted in a significant increase in the rate and amount of PARP cleavage (figure 4.3). Furthermore we could also detect cleaved
PARP in the KLF3 rescued cell line in the absence of UV treatment. This suggests that overexpression of KLF3 promotes cellular apoptosis.
P a g e | 64
Chapter 4 – A Role for KLF3 in Apoptosis
Figure 4.3. KLF3 increases PARP cleavage following UV induced cell stress. Nuclear extracts were harvested from various Murine Embryonic Fibroblast (MEF) cell lines at a number of intervals during a 24 hour time course, following UV irradiation. The amount of full length and cleaved PARP in the nuclear extracts was then determined by Western blotted using antibodies that recognises both species. The upper panel shows time courses of PARP cleavage for Klf3+/+ and Klf3-/- MEFs and the lower panel shows time courses for Klf3-/- MEFs stably rescued with either KLF3 (resWT) or empty vector control (resEMP). Apoptosis, as indicated by PARP cleavage, is increased in Klf3-/- cells overexpressing Klf3.
P a g e | 65
Chapter 4 – A Role for KLF3 in Apoptosis
4.3. INCREASED APOPTOSIS IN KLF3-/- MURINE EMBRYONIC FIBROBLASTS RESCUED WITH KLF3
Having established that loss of KLF3 reduces UV-induced apoptosis in MEF cells, we investigated whether rescue of KLF3 expression was sufficient to restore the apoptotic phenotype seen in the Klf3+/+ cell line. Given that phosphorylation of KLF3 has a role in its activity (see Chapter 3), in addition to rescue of Klf3-/- MEFs with wildtype KLF3, we also introduced mutant versions in which serine 249, the major phosphorylation site, had been mutated to either an alanine (KLF3.S249A) or a glutamate residue
(KLF3.S249E). We generated stable cell lines expressing either wildtype or mutant KLF3 by retroviral transduction and confirmed equal expression levels in each line by Western blot, using a polyclonal anti-KLF3 antibody (Figure 4.4).
Figure 4.4. Klf3-/- Murine Embryonic Fibroblast cell lines rescued with either wildtype or mutant KLF3 show equivalent levels of KLF3 protein expression. Klf3-/- Murine Embryonic Fibroblasts were retrovirally transduced with either wildtype or mutant Klf3 and stable cell lines established by antibiotic selection. Nuclear extracts were harvested, electrophoresed in a denaturing SDS-polyacrylamide gel, blotted onto nitrocellulose and stained with an anti-KLF3 antibody. The membrane was then stripped and re-stained with an antibody against β-actin to provide a loading control. Klf3-/- cell lines were rescued with resEMP: empty vector control; resWT: wildtype KLF3; resS249A: KLF3 carrying a serine to alanine mutation at residue 249; resS249E: KLF3 carrying a serine to glutamate mutation at residue 249. Multiple samples and replicates were done and a representative result is shown. P a g e | 66
Chapter 4 – A Role for KLF3 in Apoptosis
In order to assess the effect of expression of wildtype and mutant forms of KLF3 on apoptosis, we UV irradiated the rescued cell lines and once again harvested cells for annexin-V assays at various intervals during a 72 hour time course following UV treatment. As expected the line carrying the empty vector control retained a similar phenotype to the parental Klf3-/- cells, and re-introduction of wildtype KLF3 reduced the number of cells surviving UV stress. Seventy two hours after UV irradiation, the lines carrying the empty vector showed a similar phenotype to Klf3-/- cells, maintaining approximately 30% viability (compare figure 4.5 and figure 4.6), while over 90% of cells in which Klf3 expression had been restored were dead. Again, these experiments suggest that KLF3 may promote cell death or repress cell survival. These results were consistently observed with a number of independent cell lines at all timepoints during the 72 hour time course.
In the previous chapter, we confirmed that Homeodomain-Interacting Protein
Kinase 2 (HIPK2) can post-translationally modify KLF3 by phosphorylation of serine 249, resulting in increased DNA binding activity. Given that HIPK2 has previously been implicated in the response to UV induced DNA damage (Winter et al., 2008, Iacovelli et al., 2009), we investigated if phosphorylation of serine 249 of KLF3, has a role in regulating its activity in the apoptotic pathway. We found that Klf3-/- cells rescued with
KLF3.S249A, in which serine 249 had been mutated to alanine, showed an intermediate apoptotic phenotype that lay between cells rescued with empty vector and with wildtype KLF3. Seventy two hours following UV irradiation, 28% of cells rescued with empty vector were viable, compared to 1% of cells rescued with wildtype KLF3 (p<0.05, n=5), while the line rescued with KLF3.S249A showed approximately 9% viability
P a g e | 67
Chapter 4 – A Role for KLF3 in Apoptosis
(p<0.05, n=3) (Figure 4.5). The Klf3resS249A cell line also showed an intermediate phenotype for the dead cell population, with 74% of the cells staining positive for annexin V and propidium iodide at 72 hours post UV irradiation, compared with 53% for the empty vector rescue (p<0.01, n=3) and 93% for cells rescued with wildtype KLF3.
Significantly, the Klf3resS249E rescued line, which mimics phosphorylated wildtype KLF3, had a similar phenotype to the wildtype rescued line following UV irradiation (p,0.01, n=2) (Figure 4.5).
P a g e | 68
Chapter 4 – A Role for KLF3 in Apoptosis
Figure 4.5. The anti-apoptotic phenotype of Klf3-/- MEFs is reversed by expression of wildtype KLF3 and KLF3.S249E but not by KLF3.S249A. Klf3-/- MEF cell lines rescued with either wildtype KLF3 or various mutant forms of KLF3 were UV irradiated and assayed for apoptosis in annexin-V assays. Cells were harvested at 24, 48 and 72 hours following UV irradiation, stained with annexin-V-FITC and propidium iodide and analysed by flow cytometry to assess the level of apoptosis compared to untreated cells. Cell lines are - resEMP: empty vector control; resWT: wildtype KLF3; resS249A: KLF3 carrying a serine to alanine mutation at residue 249; resS249E: KLF3 carrying a serine to glutamate mutation at residue 249. Shown are representative FACS profiles of multiple experiments carried out on a number of independently derived cell lines.
P a g e | 69
Chapter 4 – A Role for KLF3 in Apoptosis
4.4. DISCUSSION
The observation that KLF3 can interact with HIPK2 and CtBP2 suggested that it may have a role in the cellular response to DNA damage, as both of these factors have previously been implicated in apoptotic pathways (Quinlan et al., 2006, Lee, 2007, Kwok, 2007). HIPK2 is known to be phosphorylated and stabilised by ATM/ATR in response to various death stimuli (Winter et al.,
2008, Sombroek and Hofmann, 2008), and its expression is deregulated in cancer cells to circumvent programmed cell death (Soubeyran et al., 2011). Whilst CtBP can activate a p53- independent pathway to initiate apoptosis and Ctbp-/- cells are more susceptible to various death stimuli (Grooteclaes et al., 2003). Thus in this chapter we explored a possible role for KLF3 in apoptosis.
We found that Klf3+/+ MEFs are more susceptible to UV induced cell death than Klf3-/- cells and the knockout MEF phenotype is reversed by re-introducing Klf3 expression. This suggests that
KLF3 may have a role in driving the apoptotic response to genotoxic stress by acting as a repressor of anti-apoptotic survival genes following DNA damage. Indeed, using microarrays and chromatin immuno-precipitation we identified two oncogenes as direct target genes of KLF3 (Funnell et al.,
2012, Vu et al., 2011). These two genes, Klf8 and Lgals3, are overexpressed in various cancer tissues and have been implicated in increasing cell survival.
We found that rescue of Klf3-/- cells with a mutant form of KLF3, in which serine 249 had been mutated to alanine to prevent phosphorylation, resulted in a reduced apoptotic response to UV irradiation. Moreover, mutation of serine 249 to glutamate, mimicking the characteristics of phosphorylated serine, resulted in rescue of the pro-apoptotic phenotype to the same extent as wildtype KLF3. These observations suggest that phosphorylation of serine 249 is important in
P a g e | 70
Chapter 4 – A Role for KLF3 in Apoptosis mediating KLF3 activity during the apoptotic response to UV induced DNA damage. Many apoptosis factors are regulated by phosphorylation. A major example is HIPK2, which is phosphorylated and stabilised by ATM/ATR kinases (Winter et al., 2008). Furthermore, the tumour suppressor p53 drives either DNA repair/cell cycle arrest or apoptosis depending on its phosphorylation state (D'Orazi et al., 2002, Chehab et al., 1999). In addition, many factors in the mitochondrial apoptotic pathway such as BAD and BIK also exhibit different activity as a result of phosphorylation (Verma et al., 2001). Our results suggest that KLF3 has a role in repressing apoptosis or supporting cell survival and that phosphorylation of serine 249 enhances this activity.
P a g e | 71
Chapter 5 – Dimerisation of KLF3
Chapter 5 - DIMERISATION OF KRÜPPEL- LIKE FACTOR 3
P a g e | 72
Chapter 5 – Dimerisation of KLF3
Chapter 5 Dimerisation of Krüppel-Like Factor 3
5.1. INTRODUCTION
KLF transcription factors share a high degree of homology in their DNA binding domains, and with the exception of KLF1 are widely expressed in the various tissues and organs of the body (reviewed in (Pearson et al., 2009, McConnell and Yang, 2010)). The regulatory mechanisms that allow KLFs to bind and regulate their individual target genes in a tissue specific manner are largely unknown. We hypothesised that dimerisation of
KLF factors might be a mechanism to generate specificity, perhaps by restricting interactions with target genes to promoters that contain a number of appropriately spaced multiple binding sites or possibly by facilitating the assembly of highly specific multimeric regulatory complexes.
The JAK/STAT pathways provide a good example of the use of dimerisation by a family of transcription factors to achieve target gene specificity (reviewed in (Levy and
Darnell, 2002, Schindler et al., 2007). In response to receptor signalling, phosphorylation of STAT transcription factors by Janus Kinase (JAK) leads to their localisation from the cytoplasm into the nucleus. The STAT transcription factors then form homodimers or heterodimers to facilitate binding and regulation of specific target genes. Hence, dimerisation allows the processing of different ligands and stimuli via a single signal transduction pathway that can lead to a variety of different cellular responses and outcomes.
P a g e | 73
Chapter 5 – Dimerisation of KLF3
Our first evidence of dimerisation within the KLF family came from the observation of higher order complexes in EMSAs of COS cells transfected with Klf3 (figure 5.1). These higher order complexes were absent in resting COS cells extract and were recognised by anti-KLF3 antibody, indicating the presence of KLF3 as part of the complexes.
Figure 5.1. Overexpression of Klf3 in COS cells reveals a higher order complex that binds the β-globin promoter CACCC box and is recognised by anti-KLF3 antibody. Nuclear extract from COS cells overexpressing Klf3 were incubated with a radiolabelled β-globin promoter CACCC box probe. The major band in the first lane is KLF3 bound to radiolabelled probe. This band can be supershifted by the addition of anti-KLF3 antibody (lane 2). Interestingly, the anti-KLF3 antibody also bound a larger, slower migrating complex in lane 1, potentially a higher order multimeric KLF3 complex. This figure is a replicate experiment of figure 3.3 added to visualise the presence of a higher order complex. P a g e | 74
Chapter 5 – Dimerisation of KLF3
5.2. KLF3 CAN SELF-ASSOCIATE IN YEAST TWO-HYBRID ASSAYS
In order to study potential dimerisation by KLF3, we employed the yeast two-hybrid system. We cloned a series of KLF3 deletion mutants (figure 5.2) into yeast two-hybrid bait and prey vectors and investigated their ability to bind to a long KLF3 peptide, encompassing amino acids 1-268. This construct, which we termed KLF3-(1-268), lacks the zinc finger DNA binding domain of KLF3 (269-344). We found that this construct is notably more stable than the full length protein and it was hence used in our studies of
KLF3 dimerisation. Yeast colonies successfully transformed with both bait and prey vectors where identified by their ability to grow in the absence of leucine and tryptophan amino acids. These were then further tested for KLF3 self-association by selecting for growth in the absence of histidine, permitted by binding of bait and prey fusion proteins leading to activation of the His3 reporter gene.
P a g e | 75
Chapter 5 – Dimerisation of KLF3
Figure 5.2. KLF3 deletion mutants assayed for self-association in yeast two-hybrid assays. A series of Klf3 deletion mutants were cloned into yeast two-hybrid vectors. The series covered amino acids 1 to 268 and therefore did not include the DNA binding domain.
We began by testing the ability of the KLF3 deletion series, expressed as GAL4-DNA binding domain (GAL4-DBD) fusions, to recruit a KLF3-(1-268) prey construct to the His3 promoter. We found that a number of these deletion mutants (amino acids 1-50, 1-100 and 151-268) auto-activated the expression of the His3 reporter gene without the presence of a KLF3 prey construct thus we were unable to proceed with these constructs. We therefore tested for interactions in the opposite direction using KLF3-(1-
268) as bait for KLF3 deletion constructs fused to the activation domain of GAL4 (GAL-
AD). In this direction, we found no evidence of auto-activation. These experiments identified a potential self-association region in KLF3 lying between amino acids 1 and
150 (figure 5.3 and table 5.1). Unexpectedly we found that the 1-150 domain appeared
P a g e | 76
Chapter 5 – Dimerisation of KLF3 not to bind to the KLF3-1-268 construct. We suspect that this result may be due to incorrect folding of this deletion mutant. Unfortunately, for technical reasons, we were unable to determine by 1D-NMR (Nuclear Magnetic Resonance) spectroscopy whether this construct had correctly folded and the reason for the loss of the interaction with this construct remains unclear.
P a g e | 77
Chapter 5 – Dimerisation of KLF3
Table 5.1. Identification of KLF3 self- association domains by yeast two-hybrid assays.
GAL4-AD GAL4-DBD GROWTH 1 1-268 ++ 1-50 2 - - 3 1-268 +++ 1-100 4 - - 5 1-268 - 1-150 6 - - 7 1-268 +++ 1-200 8 - -
9 1-268 +++ 1-268 10 - - 11 1-268 +++ 51-268 12 - - 13 1-268 + 101-268 14 - - 15 1-268 - 151-268 16 - - 17 1-268 - 201-268 18 - -
Figure 5.3. Identification of KLF3 self- association domains by yeast two-hybrid assays. Saccharomyces cerevisiae strain HF7c cells were transformed with Klf3 bait and Klf3 prey two-hybrid vectors and plated in histidine-deficient media to identify interactions between KLF3 domains. The extent of growth is described in table 5.1, with (-) indicating no growth, and (+++) representing strong growth.
P a g e | 78
Chapter 5 – Dimerisation of KLF3
In order to further characterise the potential interaction domains of KLF3, we carried out a second round of yeast two-hybrid assays, this time examining the ability of various shorter deletion mutant constructs to interact (figure 5.4 and table 5.2). Again we found that domains 1-50 and 1-100 auto-activated and in addition the GAL-DBD-
KLF3-(101-268) fusion construct alone was toxic to cells. We were therefore unable to analyse these domains further. However constructs covering amino acids 1-150, 1-200 and 51-268 did not auto-activate and the results of yeast two-hybrid assays with these constructs supported our initial observations of a self-association domain lying between amino acids 1-150, as previously described in figure 5.3 and table 5.1.
P a g e | 79
Chapter 5 – Dimerisation of KLF3
Figure 5.4. Second round yeast two-hybrid assays Self-association between various domains of KLF3 was tested in yeast two-hybrid assays. These assays supported initial observations that an interaction domain lies between amino acids 1-150.
Table 5.2. Further mapping of KFL3 self-association domain. GAL4-AD GAL4-DBD GROWTH 1 1-150 1-150 - 2 1-150 - - 3 - 1-150 - 4 1-200 1-200 ++ 5 1-200 - - 6 - 1-200 - 7 51-268 51-268 ++ 8 51-268 - - 9 - 51-268 -
P a g e | 80
Chapter 5 – Dimerisation of KLF3
5.3. DISCUSSION
In chapters three and four we showed that post-translational modifications play an important role in the activity and function of KLF3. Here, we have identified a potential self-association domain in KLF3 that lies between amino acids 1-150, which may potentially contribute to its activity. The identification of potential KLF dimerisation domains suggests that their activity may be regulated by the formation of higher order complexes. KLFs have highly homologous DNA binding domains and recognise similar binding sites with the general consensus sequence of NCN-CNC-CCN protein (reviewed in (Dang et al., 2000, Kaczynski et al., 2003)). Dimerisation could be one of the mechanisms that these factors employ to increase their target gene and tissue specificities (reviewed in (Lamb and McKnight, 1991)). For example, by associating as dimers or multimers in regulatory complexes, KLFs could recognise multiple NCN-CNC-
CCN sites, or alternatively bind to different DNA sequences. In a similar scenario, the transcription factor OCT-1 can form two different conformations of homodimers, each capable of recognizing different sequences in the genome (Reményi et al., 2001).
Phosphorylated STAT transcription factors also form different combinations of homo- and heterodimers to allow binding at different DNA elements (Brierley and Fish, 2005,
Funnell and Crossley, 2012).
Given that we have shown that KLF3 has the potential to dimerise, it will be interesting to investigate whether the related KLFs, KLF8 and KLF12, can also self-associate and furthermore whether these three KLFs have the ability to form heterodimers. In future studies, nuclear magnetic resonance (NMR) using bacterially
P a g e | 81
Chapter 5 – Dimerisation of KLF3 expressed proteins will be used to confirm associations and reveal the interaction domain. As bacterially expressed KLF3 is relatively unstable, we will increase solubility by investigating the use of peptides tagged with GST of MBP. We will also make use of low copy bacterial plasmids and choose promoters that drive moderate level gene and hence protein expression, to improve protein stability and solubility. The induction culture conditions will also be optimised to increase soluble protein production.
Choosing the right pH, salt concentration and considering the requirement for co-factors and prosthetic groups, such as zinc, are all crucial for NMR analysis.
P a g e | 82
Chapter 6 – Discussion
Chapter 6 - DISCUSSION
P a g e | 83
Chapter 6 – Discussion
Chapter 6 Discussion
6.1. SUMMARY
We have shown that KLF3 migrates as a number of different species in gel electrophoresis experiments and that this migration is affected by phosphatase treatment. We have also shown that mutation of serine 249 (S249) of KLF3 to an alanine residue (S249A) significantly reduced its binding to DNA, and dephosphorylation of KLF3 with phosphatase virtually abolished the interaction. We also demonstrated that HIPK2 may be an upstream regulator of KLF3 activity, with HIPK2 potentially mediating phosphorylation of serine 249 to enhance DNA binding activity. One of the pathways in which HIPK2 may regulate KLF3 activity is the apoptotic response to DNA damage. We showed that loss of KLF3 may reduce the susceptibility of MEF cells to UV-induced cell death. Rescue of these cells with the mutant KLF3.S249A construct failed to completely restore the response to UV damage. In contrast, Klf3-/- MEFs rescued with a phospho-mimic mutant form of KLF3, in which S249 has been mutated to a glutamate residue (S249E), showed an equivalent response to Klf3-/- cells rescued with wildtype
KLF3 following UV-induced stress. We also noticed that rescue of KLF3 expression in Klf3-
/- MEFs may be deleterious to cell survival, as these cells showed spontaneous cleavage and inactivation of the DNA repair enzyme, PARP, in the absence of UV stress. Finally, we also explored the possibility that KLF3 may use dimerisation as a mechanism of increasing target gene specificity and identified a putative self-association domain in the first 150 amino acids of the protein.
P a g e | 84
Chapter 6 – Discussion
6.2. PHOSPHORYLATION CAN MODULATE THE ACTIVITY OF KLF3
Post-translational modifications have been implicated in many biological processes including protein folding, localisation, stability and activity. For example, phosphorylation of serine 20 of p53 by checkpoint kinases obstructs binding of MDM2 ubiquitin ligase, thereby preventing p53’s degradation (reviewed in (Brooks and Gu,
2003). In a second example, phosphorylation of serine 15 of p53 is important for acetylation by p300/CBP to potentiate p53 transcriptional activity (Lambert et al., 1998,
Yamaguchi et al., 2009).
In contrast to bacterially expressed KLF3, where HIPK2 phosphorylates a single site at serine 249 in vitro, the in vivo phosphorylation pattern of KLF3 is highly complex.
Research in our laboratory has now identified at least 7 sites in KLF3 (serine 71, 78, 91,
100, 215, 223, and 249) that are constitutively phosphorylated when KLF3 is expressed in COS cells (Lee, 2007). Co-transfection with HIPK2 results in additional complexity, with a notable increase in the level of serine 249 phosphorylation and additional phosphorylation at serine 107 and 110. In summary, the activity of KLF3 is likely to be dependent upon direct phosphorylation by HIPK at serine 249, which leads to further phosphorylation events, possibly involving additional kinases, which are themselves targets of HIPK2. Confirming the specific kinases responsible for these multiple phosphorylation events and deciphering their physiological relevance remains a significant ongoing challenge for our laboratory.
P a g e | 85
Chapter 6 – Discussion
In this project, we have seen that phosphorylation plays an important role in regulating the activity of KLF3. When we expressed KLF3 in bacteria, in the absence of mammalian kinases, we observed only weak DNA binding activity. This weak binding can be enhanced by subsequent incubation of bacterially expressed KLF3, but not
KLF3.S249A, with HIPK2 kinase. Furthermore, dephosphorylation of murine erythroleukaemia (MEL) -endogenous KLF3 abolishes its DNA binding activity, suggesting that phosphorylation may be important for establishing an optimum structural conformation for binding. We have observed that the N-terminus of KLF3 may be inhibitory to DNA binding, as sequential N-terminal truncations of KLF3 show improved affinity for DNA (Tan, L. Y., 2007, unpublished data). One hypothesis is that phosphorylation of KLF3 at residues such as S249 results in conformational changes that release the C-terminal DNA binding domain from an inhibitory association with the
N-terminus to facilitate DNA binding.
We also considered the possibility that phosphorylation of KLF3 may affect its stability, given that HIPK2 mediated phosphorylation of KLF3’s major binding partner
CtBP, leads to proteosomal degradation during the cell’s response to UV-induced DNA damage (Zhang et al., 2005, Zhang et al., 2003). To investigate this, we have examined the stability of KLF3 expressed in COS cells, in fibroblasts following UV irradiation, and during maturation of erythroid progenitor cells. In each case we have found that KLF3 seems to show a context dependent response to phosphorylation. While phosphorylation following UV-irradiation appears to be associated with loss of KLF3
(Dewi, 2008), increased stability is seen when co-expressed with HIPK2 in COS cells
P a g e | 86
Chapter 6 – Discussion
(Kwok, 2007) and also following phosphorylation in maturing MEL cells (Dewi, unpublished results).
Given KLF3’s role in erythropoiesis (Funnell et al., 2012), we also considered the possibility that HIPK2 phosphorylation might direct KLF3 activity during red blood cell development. To do this, we inhibited HIPK2 activity during MEL cell maturation using the chemical agent SB203580 and found that this resulted in disrupted erythroid maturation and poor induction of globin gene expression (Dewi, 2010, unpublished results). To pursue this observation, we obtained Hipk1-/- mice and Hipk2-/- mice; however, breeding and investigating double knockout animals have proven difficult due to embryonic lethality at a stage prior to the onset of definitive erythropoiesis (Isono et al., 2006). As an alternative approach, we commenced experiments using RNA interference to knockdown Hipk gene expression in erythroid cell lines but ceased this work due to a recent publication on the role of Hipk in red blood cells that confirmed and expanded our preliminary data (Hattangadi et al., 2010).
P a g e | 87
Chapter 6 – Discussion
6.3. KLF3 PLAYS A ROLE IN APOPTOSIS
Previously, we have established that HIPK2 can phosphorylate KLF3 (Kwok, 2007,
Lee, 2007) and have also shown that phosphorylation of KLF3 changes in response to UV radiation (Dewi, 2008). Given that HIPK2 plays a prominent role in the cell’s response to
UV induced damage, we hypothesised that KLF3 may have a direct role in the stress stimuli/DNA-damage response. Exploring this hypothesis, we found that loss of KLF3 resulted in reduced apoptosis following UV radiation, a phenotype similar to that observed in Hipk2-/- MEFs (Isono et al., 2006). We have previously generated cell lines on a Klf3 null background in which we can induce Klf3 expression. Although producing these lines proved difficult, perhaps due to toxicity associated with overexpression of Klf3, we did observe increased UV induced apoptosis in the presence of KLF3 (Schepers, 2007).
The phenotype of Klf3-/- MEFs contrasts with cells that lack KLF3’s co-repressor CtBP, which showed an opposite phenotype with increased susceptibility to genotoxic stress
(Grooteclaes et al., 2003). CtBP act as a corepressor for a number of factors in addition to KLF3 ((Quinlan et al., 2006), and reviewed in (Chinnadurai, 2009)) and hence loss of
CtBP will result in the functional deregulation of many transcriptional regulators, which may contribute to the difference in the phenotypes. It is also possible that KLF3 may have CtBP independent roles in apoptosis, as we have observed that mutation of the
CtBP binding site on KLF3 does not completely abrogate its repressive function (Turner and Crossley, 1998, Perdomo et al., 2005).
In an attempt to identify KLF3 target genes that might offer a molecular explanation for the reduced susceptibility of Klf3-/- cells to stress induced apoptosis, we have
P a g e | 88
Chapter 6 – Discussion performed extensive microarray analysis comparing Klf3+/+ and Klf3-/- cells from a variety of tissues, including UV-irradiated MEFs (Artuz, 2009). From these arrays, we were able to identify deregulated expression of a number of factors with roles in cell survival and apoptosis. Again, interpreting the data has proved complex, with both pro and anti- apoptotic proteins appearing to be affected. In addition, chromatin immunoprecipitation assays have suggested that a number of these genes are not direct targets of KLF3 (Burdach, J., unpublished data).
However, one possible explanation for the phenotype of Klf3-/- cells is provided by our previous observation that KLF3 represses the related KLF, Klf8 and hence KLF8 levels are significantly elevated in cells that lack KLF3 (Eaton et al., 2008). KLF8 has been implicated as an oncogene capable of promoting tumour progression and metastatic invasion (Wang et al., 2007, Wang et al., 2008), and inhibition and silencing of Klf8 can result in apoptosis and reduced proliferation (Wan et al., 2012). Changes in KLF8 levels in Klf3-/- cells may therefore influence cell survival. The complexity of interpreting the phenotype of Klf3 null cells is further compounded by de-repression of Klf12 in the absence of KLF3 (Vu et al., 2011). KLF8 and KLF12 share highly related DNA binding domains with KLF3, and all act as transcriptional repressors via recruitment of the corepressor CtBP, raising the possibility of functional redundancy compensating for the loss of KLF3. Another recently identified KLF3 target gene, Lgals3 (Vu, 2011, Funnell et al., 2012), codes for an anti-apoptotic factor, GALECTIN-3, which is overexpressed in cancer cells protecting them from cell death (Akahani et al., 1997). Elevated levels of
GALECTIN-3 in Klf3-/- MEFs may hence also contribute to their protection from apoptosis.
P a g e | 89
Chapter 6 – Discussion
The increased survival of cells lacking KLF3 following treatment with UV radiation suggests that they may be able to circumvent apoptotic pathways. Such cells will carry
DNA mutations and be prone to gene-instability and could potentially give rise to tumours and cancer. To date, there is no evidence of cancer arising in our Klf3-/- mice but such a predisposition may be revealed in the future by challenge with appropriate stress stimuli, such as UV radiation. While loss of KLF3 appears to increase cell survival, rescue of KLF3 expression in Klf3-/- MEFS appeared to affect apoptotic pathways, with elevated spontaneous cleavage of the DNA repair enzyme, PARP, observed in these cells in the absence of UV stress. Some caution is needed however in the interpretation of these data given the high level overexpression of the various KLF3 constructs in these cell lines.
P a g e | 90
Chapter 6 – Discussion
6.4. POSSIBLE DIMERISATION IN KLFS
KLFs are characterised by a highly homologous DNA binding domain and are therefore potentially capable of binding and regulating the same sites in the promoters of genes. Given that KLFs are widely expressed, mechanisms must exist to ensure an appropriate level of target gene specificity, thereby ensuring that physiological processes are correctly regulated. We have explored the possibility that dimerisation may be one such mechanisms by investigating the potential self-association of KLF3.
In support of KLF3 self-association, we observed an additional higher order complex in EMSAs assessing KLF3 binding to a consensus probe sequence. This complex was recognised and supershifted by anti-KLF3 antibody. In subsequent experiments, yeast two-hybrid assays using a deletion series of KLF3 bait and prey constructs revealed a putative self-association domain lying between amino acids 1-150.
To further define the interaction domain, we carried out 1D Nuclear Magnetic
Resonance (NMR) spectroscopy and analytical ultracentrifugation (AU) of bacterially expressed KLF3-GST fusion proteins. These experiments proved challenging. Following cleavage of GST, we found expressed KLF3 very temperature sensitive and prone to denaturation, and a lack of tryptophan residues made identifying relevant protein fractions in ion exchange chromatography problematic. As a result, our 1D-NMR spectra were poor and we were unable to interpret the observed peaks (Dewi, unpublished results). The same issues confounded our AU analysis, which also produced hard to interpret, ambiguous data (Dewi, unpublished results).
P a g e | 91
Chapter 6 – Discussion
Although our initial data provide the first evidence of possible dimerisation within the KLF family, future studies are needed to confirm and further define the mechanism of this interaction.
P a g e | 92
Chapter 6 – Discussion
6.5. CONCLUSIONS
We have shown that phosphorylation of serine 249 of KLF3 regulates KLF3’s DNA binding activity. HIPK2 mediated phosphorylation of KLF3, mutation of serine 249, and dephosphorylation of KLF3 all affect the interaction with DNA. We propose that HIPK2 phosphorylation of KLF3 has a role in the response to UV induced DNA damage. Cells that lack KLF3 are less prone to apoptosis following stress. Rescue of Klf3-/- MEFs with wildtype KLF3 (Klf3resWT) restores an apoptotic phenotype equivalent to Klf3+/+ cells. We have found that the phosphorylation site at serine 249 has a role in this response; rescue of Klf3-/- cells with a mutant form of KLF3, in which serine 249 has been mutated to an alanine residue, failed to fully restore the apoptotic response. This contrasts with the phenotype seen following expression of a phospho-mimic form of KLF3, in which serine 249 has been mutated to a glutamate residue, where cells showed an equivalent level of apoptosis to Klf3resWT cells. Finally, we provide evidence of a potential self-association domain in KLF3 that may have a role in facilitating target gene specificity. This first demonstration of potential multimerisation within the KLF family could explain how these highly related, widely expressed factors are able to direct tissue specific gene regulatory programs during development and differentiation.
P a g e | 93
Appendix
Chapter 7 - APPENDIX
P a g e | 94
Appendix
APPENDIX
Figure 7.1. Full image of KLF3 phosphorylation in the nuclear extract of NIH/3T3 cells Western blot. Nuclear extracts of NIH/3T3 were incubated with (+) or without (-) -phosphatase and were used for Western blot analysis. The migration of KLF3 changed as a result of phosphorylation. Several unspecific bands of different sizes to KLF3 (*) were observed.
P a g e | 95
References
Chapter 8 - REFERENCES
P a g e | 96
References
REFERENCES
Akahani, S., Nangia-Makker, P., Inohara, H., Kim, H.-R. C. & Raz, A. (1997). Galectin-3: A Novel Antiapoptotic Molecule with A Functional BH1 (NWGR) Domain of Bcl-2 Family. Cancer Research, 57, 5272-5276.
Arai, S., Matsushita, A., Du, K., Yagi, K., Okazaki, Y. & Kurokawa, R. (2007). Novel homeodomain-interacting protein kinase family member, HIPK4, phosphorylates human p53 at serine 9. Febs Letters, 581, 5649-5657.
Artuz, C. M. (2009). Identifying genes regulated by Krüppel-Like Factors 3 and 8. School of Molecular and Microbial Bioscience. Australia: University of Sydney.
Bieker, J. J. (2001). Kruppel-like factors: three fingers in many pies. The Journal of biological chemistry, 276, 34355-34358.
Boyd, J. M., Subramanian, T., Schaeper, U., La Regina, M., Bayley, S. & Chinnadurai, G. (1993). A region in the C-terminus of adenovirus 2/5 E1a protein is required for association with a cellular phosphoprotein and important for the negative modulation of T24-ras mediated transformation, tumorigenesis and metastasis. The EMBO journal, 12, 469-478.
Brierley, M. M. & Fish, E. N. (2005). Stats: multifaceted regulators of transcription. Journal of interferon & cytokine research : the official journal of the International Society for Interferon and Cytokine Research, 25, 733-744.
Brooks, C. L. & Gu, W. (2003). Ubiquitination, phosphorylation and acetylation: the molecular basis for p53 regulation. Current Opinion in Cell Biology, 15, 164-171.
Chehab, N. H., Malikzay, A., Stavridi, E. S. & Halazonetis, T. D. (1999). Phosphorylation of Ser-20 mediates stabilization of human p53 in response to DNA damage. Proceedings of the National Academy of Sciences, 96, 13777-13782.
Chen, C., Sun, X., Ran, Q., Wilkinson, K. D., Murphy, T. J., Simons, J. W. & Dong, J. T. (2005). Ubiquitin-proteasome degradation of KLF5 transcription factor in cancer and untransformed epithelial cells. Oncogene, 24, 3319-3327.
P a g e | 97
References
Chinnadurai, G. (2009). The Transcriptional Corepressor CtBP: A Foe of Multiple Tumor Suppressors. Cancer Research, 69, 731-734.
Crossley, M., Whitelaw, E., Perkins, A., Williams, G., Fujiwara, Y. & Orkin, S. H. (1996). Isolation and characterization of the cDNA encoding BKLF/TEF-2, a major CACCC- box-binding protein in erythroid cells and selected other cells. Molecular and Cellular Biology, 16, 1695-1705.
D'Orazi, G., Cecchinelli, B., Bruno, T., Manni, I., Higashimoto, Y., Saito, S., Gostissa, M., Coen, S., Marchetti, A., Del Sal, G., Piaggio, G., Fanciulli, M., Appella, E. & Soddu, S. (2002). Homeodomain-interacting protein kinase-2 phosphorylates p53 at Ser 46 and mediates apoptosis. Nat Cell Biol, 4, 11-19.
Dang, D. T., Pevsner, J. & Yang, V. W. (2000). The biology of the mammalian Kruppel-like family of transcription factors. The international journal of biochemistry & cell biology, 32, 1103-1121.
Dauth, I., Krüger, J. & Hofmann, T. G. (2007). Homeodomain-Interacting Protein Kinase 2 Is the Ionizing Radiation–Activated p53 Serine 46 Kinase and Is Regulated by ATM. Cancer Research, 67, 2274-2279.
Davis, H. E., Rosinski, M., Morgan, J. R. & Yarmush, M. L. (2004). Charged polymers modulate retrovirus transduction via membrane charge neutralization and virus aggregation. Biophys J, 86, 1234-1242.
Dewi, V. A. (2008). An Investigation into the Role of Basic Krüppel-Like Factor in the UV Response Pathway. School of Molecular and Microbial Biosciences. Australia: University of Sydney.
Du, J. X., Bialkowska, A. B., McConnell, B. B. & Yang, V. W. (2008). SUMOylation Regulates Nuclear Localization of Krüppel-like Factor 5. Journal of Biological Chemistry, 283, 31991-32002.
Eaton, S. A., Funnell, A. P., Sue, N., Nicholas, H., Pearson, R. C. & Crossley, M. (2008). A Network of Kruppel-like Factors (Klfs): Klf8 is repressed by Klf3 and activated by Klf1 in vivo. The Journal of biological chemistry, 283, 26937-26947.
P a g e | 98
References
Ecsedy, J. A., Michaelson, J. S. & Leder, P. (2003). Homeodomain-interacting protein kinase 1 modulates Daxx localization, phosphorylation, and transcriptional activity. Molecular and Cellular Biology, 23, 950-960.
Fadok, V. A., Voelker, D. R., Campbell, P. A., Cohen, J. J., Bratton, D. L. & Henson, P. M. (1992). Exposure of phosphatidylserine on the surface of apoptotic lymphocytes triggers specific recognition and removal by macrophages. The Journal of Immunology, 148, 2207-2216.
Funnell, A. & Crossley, M. (2012). Homo- and heterodimerization in transcriptional regulation in protein dimerization (anad oligomerization). In J. M. Matthews (Ed.) Protein dimerization and oligomerization in biology. Landes Bioscience and Springer.
Funnell, A. P. W., Norton, L. J., Mak, K. S., Burdach, J., Artuz, C. M., Twine, N. A., Wilkins, M. R., Power, C. A., Hung, T.-T., Perdomo, J., Koh, P., Bell-Anderson, K. S., Orkin, S. H., Fraser, S. T., Perkins, A. C., Pearson, R. C. M. & Crossley, M. (2012). The CACCC- binding protein KLF3/BKLF represses a subset of KLF1/EKLF target genes and is required for proper erythroid maturation in vivo. Molecular and Cellular Biology.
Geske, F. J., Nelson, A. C., Lieberman, R., Strange, R., Sun, T. & Gerschenson, L. E. (2000). DNA repair is activated in early stages of p53-induced apoptosis. Cell death and differentiation, 7, 393-401.
Grooteclaes, M., Deveraux, Q., Hildebrand, J., Zhang, Q., Goodman, R. H. & Frisch, S. M. (2003). C-terminal-binding protein corepresses epithelial and proapoptotic gene expression programs. Proc Natl Acad Sci U S A, 100, 4568-4573.
Hartwell, L. (1992). Defects in a cell cycle checkpoint may be responsible for the genomic instability of cancer cells. Cell, 71, 543-546.
Hattangadi, S. M., Burke, K. A. & Lodish, H. F. (2010). Homeodomain-interacting protein kinase 2 plays an important role in normal terminal erythroid differentiation. Blood, 115, 4853-4861.
Hildebrand, J. D. & Soriano, P. (2002). Overlapping and unique roles for C-terminal binding protein 1 (CtBP1) and CtBP2 during mouse development. Mol Cell Biol., 22, 5296-5307. P a g e | 99
References
Huang, C., Ma, W.-Y., Maxiner, A., Sun, Y. & Dong, Z. (1999). p38 Kinase Mediates UV- induced Phosphorylation of p53 Protein at Serine 389. Journal of Biological Chemistry, 274, 12229-12235.
Iacovelli, S., Ciuffini, L., Lazzari, C., Bracaglia, G., Rinaldo, C., Prodosmo, A., Bartolazzi, A., Sacchi, A. & Soddu, S. (2009). HIPK2 is involved in cell proliferation and its suppression promotes growth arrest independently of DNA damage. Cell Prolif, 42, 373-384.
Isono, K., Nemoto, K., Li, Y., Takada, Y., Suzuki, R., Katsuki, M., Nakagawara, A. & Koseki, H. (2006). Overlapping roles for homeodomain-interacting protein kinases hipk1 and hipk2 in the mediation of cell growth in response to morphogenetic and genotoxic signals. Molecular and Cellular Biology, 26, 2758-2771.
Kaczynski, J., Cook, T. & Urrutia, R. (2003). Sp1- and Kruppel-like transcription factors. Genome Biology, 4, 206.
Kaufmann, S. H., Desnoyers, S., Ottaviano, Y., Davidson, N. E. & Poirier, G. G. (1993). Specific proteolytic cleavage of poly(ADP-ribose) polymerase: an early marker of chemotherapy-induced apoptosis. Cancer Research, 53, 3976-3985.
Kim, E. A., Noh, Y. T., Ryu, M.-J., Kim, H.-T., Lee, S.-E., Kim, C.-H., Lee, C., Kim, Y. H. & Choi, C. Y. (2006). Phosphorylation and Transactivation of Pax6 by Homeodomain- interacting Protein Kinase 2. Journal of Biological Chemistry, 281, 7489-7497.
Kim, Y. H., Choi, C. Y., Lee, S. J., Conti, M. A. & Kim, Y. (1998). Homeodomain-interacting protein kinases, a novel family of co-repressors for homeodomain transcription factors. The Journal of biological chemistry, 273, 25875-25879.
Kwok, A. (2007). HIPK2 is a Novel Regulator of Bklf. School of Molecular and Microbial Biosciences. Australia: University of Sydney.
Lamb, P. & McKnight, S. L. (1991). Diversity and specificity in transcriptional regulation: the benefits of heterotypic dimerization. Trends in Biochemical Sciences, 16, 417- 422.
P a g e | 100
References
Lambert, P. F., Kashanchi, F., Radonovich, M. F., Shiekhattar, R. & Brady, J. N. (1998). Phosphorylation of p53 Serine 15 Increases Interaction with CBP. Journal of Biological Chemistry, 273, 33048-33053.
Lee, M. M. (2007). HIPK2, A Regulator of BKLF Repression Complex. School of Molecular and Microbial Biosciences. Australia: University of Sydney.
Lee, W., Andrews, B. C., Faust, M., Walldorf, U. & Verheyen, E. M. (2009). Hipk is an essential protein that promotes Notch signal transduction in the Drosophila eye by inhibition of the global co-repressor Groucho. Developmental Biology, 325, 263-272.
Levy, D. E. & Darnell, J. E. (2002). STATs: transcriptional control and biological impact. Nat Rev Mol Cell Biol, 3, 651-662.
McConnell, B. B. & Yang, V. W. (2010). Mammalian Krüppel-Like Factors in Health and Diseases. Physiological Reviews, 90, 1337-1381.
Meek, D. W. & Anderson, C. W. (2009). Posttranslational Modification of p53: Cooperative Integrators of Function. Cold Spring Harbor Perspectives in Biology, 1.
Miller, I. J. & Bieker, J. J. (1993). A novel, erythroid cell-specific murine transcription factor that binds to the CACCC element and is related to the Kruppel family of nuclear proteins. Molecular and Cellular Biology, 13, 2776-2786.
Nakaya, N., Lowe, S. W., Taya, Y., Chenchik, A. & Enikolopov, G. (2000). Specific pattern of p53 phosphorylation during nitric oxide-induced cell cycle arrest. Oncogene, 19, 6369-6375.
Nardinocchi, L., Puca, R., Sacchi, A., Rechavi, G., Givol, D. & D'Orazi, G. (2009). Targeting Hypoxia in Cancer Cells by Restoring Homeodomain Interacting Protein-Kinase 2 and p53 Activity and Suppressing HIF-1α. PLoS One, 4, e6819.
Nicoletti, I., Migliorati, G., Pagliacci, M. C., Grignani, F. & Riccardi, C. (1991). A rapid and simple method for measuring thymocyte apoptosis by propidium iodide staining and flow cytometry. J Immunol Methods, 139, 271-279.
Ouyang, L., Chen, X. & Bieker, J. J. (1998). Regulation of erythroid Kruppel-like factor (EKLF) transcriptional activity by phosphorylation of a protein kinase casein kinase II
P a g e | 101
References
site within its interaction domain. The Journal of biological chemistry, 273, 23019- 23025.
Patrick, M. H. (1977). STUDIES ON THYMINE-DERIVED UV PHOTOPRODUCTS IN DNA—I. FORMATION AND BIOLOGICAL ROLE OF PYRIMIDINE ADDUCTS IN DNA. Photochemistry and Photobiology, 25, 357-372.
Pearson, R., Fleetwood, J., Eaton, S., Crossley, M. & Bao, S. (2008). Kruppel-like transcription factors: A functional family. The international journal of biochemistry & cell biology, 40, 1996-2001.
Pearson, R. C., Funnell, A. P. & Crossley, M. (2009). The biological roles of the Kruppel- like Factor (Klf) family. In K. Yoshida (Ed.) Focus on Zinc Finger Protein Research. pp. 45-63). Research Signpost.
Pearson, R. C., Funnell, A. P. & Crossley, M. (2011). The mammalian zinc finger transcription factor Krüppel-like Factor 3 (KLF3/BKLF). IUBMB Life, 63, 86-93.
Perdomo, J., Verger, A., Turner, J. & Crossley, M. (2005). Role for SUMO modification in facilitating transcriptional repression by BKLF. Molecular and Cellular Biology, 25, 1549-1559.
Perkins, A. C., Sharpe, A. H. & Orkin, S. H. (1995). Lethal beta-thalassaemia in mice lacking the erythroid CACCC-transcription factor EKLF. Nature, 375, 318-322.
Quinlan, K. G., Verger, A., Kwok, A., Lee, S. H., Perdomo, J., Nardini, M., Bolognesi, M. & Crossley, M. (2006). Role of the C-terminal binding protein PXDLS motif binding cleft in protein interactions and transcriptional repression. Molecular and Cellular Biology, 26, 8202-8213.
Ravanat, J.-L., Douki, T. & Cadet, J. (2001). Direct and indirect effects of UV radiation on DNA and its components. Journal of Photochemistry and Photobiology B: Biology, 63, 88-102.
Reményi, A., Tomilin, A., Pohl, E., Lins, K., Philippsen, A., Reinbold, R., Schöler, H. R. & Wilmanns, M. (2001). Differential Dimer Activities of the Transcription Factor Oct-1 by DNA-Induced Interface Swapping. Molecular cell, 8, 569-580.
P a g e | 102
References
Rinaldo, C., Siepi, F., Prodosmo, A. & Soddu, S. (2008). HIPKs: Jack of all trades in basic nuclear activities. Biochimica et biophysica acta, 1783, 2124-2129.
Rui, Y., Xu, Z., Lin, S., Li, Q., Rui, H., Luo, W., Zhou, H. M., Cheung, P. Y., Wu, Z., Ye, Z., Li, P., Han, J. & Lin, S. C. (2004). Axin stimulates p53 functions by activation of HIPK2 kinase through multimeric complex formation. The EMBO journal, 23, 4583-4594.
Sancar, A., Lindsey-Boltz, L. A., Ünsal-Kaçmaz, K. & Linn, S. (2004). Molecular mechanisms of mammalian DNA repair and the DNA damage checkpoints Annual Review of Biochemistry, 73, 39-85.
Schaeper, U., Boyd, J. M., Verma, S., Uhlmann, E., Subramanian, T. & Chinnadurai, G. (1995). Molecular cloning and characterization of a cellular phosphoprotein that interacts with a conserved C-terminal domain of adenovirus E1A involved in negative modulation of oncogenic transformation. Proc Natl Acad Sci U S A, 92, 10467-10471.
Schepers, A. (2007). Investigating the in vivo role of Basic Krüppel-Like Factor. School of Molecular and Microbial Biosciences. Australia: University of Sydney.
Schindler, C., Levy, D. E. & Decker, T. (2007). JAK-STAT Signaling: From Interferons to Cytokines. Journal of Biological Chemistry, 282, 20059-20063.
Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. (2012). NIH Image to ImageJ: 25 years of image analysis. Nat Meth, 9, 671-675.
Shi, Y., Sawada, J., Sui, G., Affar, E. B., Whetstine, J. R., Lan, F., Ogawa, H., Luke, M. P., Nakatani, Y. & Shi, Y. (2003). Coordinated histone modifications mediated by a CtBP co-repressor complex. Nature, 422, 735-738.
Shie, J. L., Chen, Z. Y., Fu, M., Pestell, R. G. & Tseng, C. C. (2000). Gut-enriched Kruppel- like factor represses cyclin D1 promoter activity through Sp1 motif. Nucleic Acids Research, 28, 2969-2976.
Siliciano, J. D., Canman, C. E., Taya, Y., Sakaguchi, K., Appella, E. & Kastan, M. B. (1997). DNA damage induces phosphorylation of the amino terminus of p53. Genes & Development, 11, 3471-3481.
P a g e | 103
References
Sombroek, D. & Hofmann, T. G. (2008). How cells switch HIPK2 on and off. Cell death and differentiation, 16, 187-194.
Soubeyran, I., Mahouche, I., Grigoletto, A., Leste-Lasserre, T., Drutel, G., Rey, C., Pedeboscq, S., Blanchard, F., Brouste, V., Sabourin, J.-C., Bécouarn, Y., Reiffers, J., Ichas, F. & De Giorgi, F. (2011). Tissue Microarray Cytometry Reveals Positive Impact of Homeodomain Interacting Protein Kinase 2 in Colon Cancer Survival Irrespective of p53 Function. The American journal of pathology, 178, 1986-1998.
Sue, N., Jack, B. H., Eaton, S. A., Pearson, R. C., Funnell, A. P., Turner, J., Czolij, R., Denyer, G., Bao, S., Molero-Navajas, J. C., Perkins, A., Fujiwara, Y., Orkin, S. H., Bell- Anderson, K. & Crossley, M. (2008). Targeted disruption of the basic Kruppel-like factor gene (Klf3) reveals a role in adipogenesis. Molecular and Cellular Biology, 28, 3967-3978.
Takahashi, K. & Yamanaka, S. (2006). Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell, 126, 663-676.
Tallack, M. R. & Perkins, A. C. (2010). KLF1 directly coordinates almost all aspects of terminal erythroid differentiation. IUBMB Life, 62, 886-890.
Trapasso, F., Aqeilan, R. I., Iuliano, R., Visone, R., Gaudio, E., Ciuffini, L., Alder, H., Paduano, F., Pierantoni, G. M., Soddu, S., Croce, C. M. & Fusco, A. (2009). Targeted disruption of the murine homeodomain-interacting protein kinase-2 causes growth deficiency in vivo and cell cycle arrest in vitro. DNA and Cell Biology, 28, 161-167.
Turchinovich, G., Vu, T. T., Frommer, F., Kranich, J., Schmid, S., Alles, M., Loubert, J. B., Goulet, J. P., Zimber-Strobl, U., Schneider, P., Bachl, J., Pearson, R. C., Crossley, M., Agenes, F. & Kirberg, J. (2011). Programming of marginal zone B-cell fate by basic Kruppel-like factor (BKLF/KLF3). Blood, 117, 3780-3792.
Turner, J. & Crossley, M. (1998). Cloning and characterization of mCtBP2, a co-repressor that associates with basic Kruppel-like factor and other mammalian transcriptional regulators. The EMBO journal, 17, 5129-5140.
Turner, J. & Crossley, M. (1999). Mammalian Kruppel-like transcription factors: more than just a pretty finger. Trends Biochem Sci, 24, 236-240. P a g e | 104
References
Turner, J. & Crossley, M. (2001). The CtBP family: enigmatic and enzymatic transcriptional co-repressors. Bioessays, 23, 683-690.
Verma, S., Zhao, L.-j. & Chinnadurai, G. (2001). Phosphorylation of the Pro-apoptotic Protein BIK. Journal of Biological Chemistry, 276, 4671-4676.
Vermes, I., Haanen, C., Steffens-Nakken, H. & Reutellingsperger, C. (1995). A novel assay for apoptosis Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V. J Immunol Methods, 184, 39-51.
Vu, T. T. (2011). Investigating the role of KLF3 in murine B-lymphocytes. School of Molecular Biosciences. Sydney: University of Sydney.
Vu, T. T., Gatto, D., Turner, V., Funnell, A. P., Mak, K. S., Norton, L. J., Kaplan, W., Cowley, M. J., Agenes, F., Kirberg, J., Brink, R., Pearson, R. C. & Crossley, M. (2011). Impaired B cell development in the absence of Kruppel-like factor 3. J Immunol, 187, 5032- 5042.
Wan, W., Zhu, J., Sun, X. & Tang, W. (2012). Small interfering RNA targeting Kruppel-like factor 8 inhibits U251 glioblastoma cell growth by inducing apoptosis. Mol Med Report, 5, 347-350.
Wang, X., Urvalek, A. M., Liu, J. & Zhao, J. (2008). Activation of KLF8 transcription by FAK in human ovarian epithelial and cancer cells. The Journal of biological chemistry.
Wang, X., Zheng, M., Liu, G., Xia, W., McKeown-Longo, P. J., Hung, M. C. & Zhao, J. (2007). Kruppel-like factor 8 induces epithelial to mesenchymal transition and epithelial cell invasion. Cancer Research, 67, 7184-7193.
Wei, G., Ku, S., Ma, G. K., Saito, S., Tang, A. A., Zhang, J., Mao, J. H., Appella, E., Balmain, A. & Huang, E. J. (2007). HIPK2 represses beta-catenin-mediated transcription, epidermal stem cell expansion, and skin tumorigenesis. Proc Natl Acad Sci U S A, 104, 13040-13045.
Whitmarsh, A. J. & Davis, R. J. (2000). Regulation of transcription factor function by phosphorylation. Cellular and Molecular Life Sciences, 57, 1172-1183.
P a g e | 105
References
Winter, M., Sombroek, D., Dauth, I., Moehlenbrink, J., Scheuermann, K., Crone, J. & Hofmann, T. G. (2008). Control of HIPK2 stability by ubiquitin ligase Siah-1 and checkpoint kinases ATM and ATR. Nat Cell Biol, 10, 812-824.
Yamaguchi, H., Woods, N. T., Piluso, L. G., Lee, H.-H., Chen, J., Bhalla, K. N., Monteiro, A., Liu, X., Hung, M.-C. & Wang, H.-G. (2009). p53 Acetylation Is Crucial for Its Transcription-independent Proapoptotic Functions. Journal of Biological Chemistry, 284, 11171-11183.
Zhang, Q., Nottke, A. & Goodman, R. H. (2005). Homeodomain-interacting protein kinase-2 mediates CtBP phosphorylation and degradation in UV-triggered apoptosis. Proc Natl Acad Sci U S A, 102, 2802-2807.
Zhang, Q., Yoshimatsu, Y., Hildebrand, J., Frisch, S. M. & Goodman, R. H. (2003). Homeodomain interacting protein kinase 2 promotes apoptosis by downregulating the transcriptional corepressor CtBP. Cell, 115, 177-186.
Zhang, W., Kadam, S., Emerson, B. M. & Bieker, J. J. (2001). Site-specific acetylation by p300 or CREB binding protein regulates erythroid Kruppel-like factor transcriptional activity via its interaction with the SWI-SNF complex. Molecular and Cellular Biology, 21, 2413-2422.
Zhang, X., Srinivasan, S. V. & Lingrel, J. B. (2004). WWP1-dependent ubiquitination and degradation of the lung Kruppel-like factor, KLF2. Biochemical and Biophysical Research Communications, 316, 139-148.
P a g e | 106