PHOSPHORYLATION AND REGULATION OF HUMAN CHOLINE KINASE BETA
BY PROTEIN KINASE A
CHANG CHING CHING
UNIVERSITI SAINS MALAYSIA
2015
PHOSPHORYLATION AND REGULATION OF HUMAN CHOLINE KINASE BETA
BY PROTEIN KINASE A
by
CHANG CHING CHING
Thesis submitted in fulfillment of the requirement for the degree of
Doctor of Philosophy
January 2015
ii
ACKNOWLEDGMENTS
First and foremost, I would like to express my sincere gratitude to my supervisor, Assoc Prof Dr Few Ling Ling for her willingness to accept me as her PhD student.
Her patience in guiding me along the way is very much appreciated. I also would like to thank Assoc Prof Dr See Too Wei Cun for his advice and guidance especially in problem solving. I have received endless benefit from their vast knowledge in many areas and assistance in thesis writing.
I would like to thank my co-supervisor, Dr Khoo Boon Yin for her help in my research, especially in the aspects of mass spectrometry analysis and thesis writting. I would also like to acknowledge Dr Manfred Konrad who has not only offered me a short internship in Max Planck Institute for Biophysical Chemistry but also taught me a lot in research as well as making my stay in Germany a joyful life experience.
My sincere thanks go to my seniors, colleagues and friends for their encouragement and support throughout my study. Besides that, I also would like to thank the lecturers, administration staff and technologists of the School of Health Sciences, Universiti Sains Malaysia for assisting me in completing the research. Most importantly, I would like to acknowledge National Sciences Foundation for offering the scholarship that removed the financial during my study.
Last but not least, I would like to thank my parents and my family for their support. I hope the outcome in this study will make them proud. Special gratitude should be given to my little brother, Yang Loong who has helped in taking care of my parents when I am not around them. I hope he would be successful in his studies.
iii
TABLE OF CONTENTS
ACKNOWLEDGMENTS ... ii
TABLE OF CONTENTS ... iii
LIST OF TABLES ... x
LIST OF FIGURES ... xi
LIST OF SYMBOLS, ABBREVIATIONS AND ACRONYMNS ... xv
ABSTRAK ... xix
ABSTRACT ... xxi
CHAPTER ONE INTRODUCTION ... 1
1.1 Kennedy pathway ... 1
1.2 Choline kinase ... 1
1.3 Structure of choline kinase ... 4
1.4 CK oligomeric structures ... 9
1.5 CK subcellular location ... 10
1.6 Biochemical properties of CK ... 11
1.7 The roles of choline kinase ... 14
1.7.1 Cell proliferation ... 14
1.7.2 Tumorigenesis ... 15
1.7.3 Differential role of CKα and CKβ ... 17
1.8 Regulation of choline kinase... 19
1.8.1 Transcriptional level ... 19
1.8.2 Translational level ... 20
1.8.3 Post-translational level ... 20
1.9 Phosphorylation ... 22
1.10 Protein kinase A ... 23
1.10.1 Regulation of Protein kinase A ... 24
1.10.2 Phosphorylation targets of protein kinase A... 28
1.11 Rationale of the study ... 29
1.12 Objectives of the study ... 30
1.12.1 General objective ... 30
1.12.2 Specific objectives ... 30
1.13 General approach ... 30
iv
CHAPTER TWO MATERIALS AND METHODS ... 32
2.1 Materials ... 32
2.1.1 General instruments ... 32
2.1.2 General consumables... 32
2.1.3 Chemicals reagents... 32
2.1.4 Antibodies ... 32
2.1.5 Kits ... 40
2.1.6 E. coli strains ... 40
2.1.7 Vectors ... 40
2.1.8 Oligonucleotides ... 40
2.1.9 Mammalian cells ... 40
2.1.10 Softwares ... 52
2.2 Preparation of media, buffers and solutions ... 52
2.2.1 Preparation of media for bacteria culture ... 52
2.2.1.1 Luria Bertani broth ... 52
2.2.1.2 Luria Bertani agar ... 52
2.2.1.3 Medium A ... 54
2.2.1.4 Medium B ... 54
2.2.2 Preparation of solutions for molecular cloning... 54
2.2.2.1 Tris acetate-EDTA buffer ... 54
2.2.2.2 6× DNA loading buffer ... 55
2.2.2.3 Ethidium bromide solution ... 55
2.2.3 Preparation of solutions for protein applications ... 55
2.2.3.1 Preparation of solutions for SDS-PAGE ... 55
2.2.3.1 (a) 4× Separating buffer ... 55
2.2.3.1 (b) 4× Stacking buffer ... 56
2.2.3.1 (c) 5× Sample loading buffer ... 56
2.2.3.1 (d) 10% (w/v) SDS ... 56
2.2.3.1 (e) 10% (w/v) Ammonium persulphate ... 56
2.2.3.1 (f) SDS-running buffer ... 57
2.2.3.1 (g) Coomassie blue staining solution ... 57
2.2.3.1 (h) Coomassie blue destaining solution... 57
2.2.3.2 Preparation of solutions for Western blot analysis ... 57
2.2.3.2 (a) Western blotting buffer ... 57
2.2.3.2 (b) Tris buffered saline ... 57
v
2.2.3.2 (c) Tris buffered saline-Tween ... 58
2.2.3.2 (d) 5 % (w/v) Blocking buffer ... 58
2.2.3.3 Preparation of solutions for protein purification ... 58
2.2.3.3 (a) His lysis buffer ... 58
2.2.3.3 (b) His wash buffer... 59
2.2.3.3(c) His elution buffer ... 59
2.2.3.3 (d) GST lysis buffer ... 59
2.2.3.3 (e) GST wash buffer ... 59
2.2.3.3 (f) GST elution buffer ... 60
2.2.4 Preparation of solutions for in-gel kinase assay ... 60
2.2.5 Preparation of solutions for phosphoprotein gel stain... 60
2.2.5.1 Fixing solution ... 60
2.2.5.2 Phosphoprotein destaining solution ... 62
2.2.6 Preparation of solutions for mammalian cell culture and lysis... 62
2.2.6.1 Phosphate buffered saline ... 62
2.2.6.2 Triton X-100 lysis buffer ... 62
2.2.6.3 10% (v/v) NP lysis buffer ... 63
2.2.7 Preparation of solutions for the treatment of cells ... 63
2.2.7.1 Forskolin ... 63
2.2.7.2 3-isobutyl-1-methylxanthine ... 63
2.2.7.3 H-89 ... 64
2.2.7.4 Epidermal growth factor ... 64
2.2.7.5 G418 ... 64
2.2.8 Preparation of solutions for enzymatic assay ... 65
2.2.9 Preparation of buffer from pH 2 to 10... 65
2.2.10 Preparation of solution for fluorescence imaging ... 66
2.2.10.1 4% (w/v) Paraformaldehyde fixing solution ... 66
2.2.10.2 DAPI stain ... 66
2.3 General methods... 66
2.3.1 General molecular cloning methods ... 66
2.3.1.1 Preparation of stock culture... 66
2.3.1.2 Preparation of E. coli competent cells ... 67
2.3.1.3 Heat shock transformation of E. coli ... 67
2.3.1.4 Preparation of plasmid DNA from E. coli ... 68
2.3.1.5 Preparation of RNA from mammalian cells ... 68
vi
2.3.1.6 Synthesis of cDNA from the purified RNA ... 69
2.3.1.7 Polymerase Chain Reaction amplification of DNA... 69
2.3.1.8 DNA gel electrophoresis ... 70
2.3.1.9 Isolation of DNA from agarose gel ... 71
2.3.1.10 Determination of DNA and RNA concentration ... 71
2.3.1.11 Restriction enzyme digestion ... 72
2.3.1.12 Ligation ... 72
2.3.2 General protein methods... 72
2.3.2.1 Protein expression and purification ... 72
2.3.2.2 Determination of protein concentration by using Bradford assay ... 74
2.3.2.3 Protein gel electrophoresis ... 74
2.3.2.4 Western blot analysis ... 76
2.3.2.5 Determination of CKβ catalytic activity ... 78
2.3.3 General methods for mammalian cell culture ... 79
2.3.3.1 Maintenance of cell line ... 79
2.3.3.2 Cell counting ... 80
2.3.3.3 Preparation of cell lysates ... 81
2.3.3.4 Transfection of mammalian cell with plasmid DNA ... 81
2.3.3.5 Immunoprecipitation ... 82
2.3.4 Statistical analysis ... 82
2.4 In-gel kinase assay ... 83
2.5 In vitro phosphorylation of human CK ... 84
2.5.1 Autoradiography ... 84
2.5.2 Phosphoprotein gel staining ... 85
2.6 Phosphorylation sites mapping ... 85
2.6.1 Determination of the phosphorylation sites by mass spectrometry ... 85
2.6.2 Mutation of CKβ phosphorylation sites ... 86
2.7 Intracellular phosphorylation of CKβ ... 87
2.7.1 Stable transfection of HEK 293 cell line with GFP-CKβ ... 87
2.7.2 Treatments of CKβ with PKA effectors ... 88
2.8 Determination of the effect of PKA phosphorylation on inhibition of CKβ by hemicholinium-3 ... 88
2.9 Determination of the effect of PKA phosphorylation on CKβ pH optimum ... 89
vii
2.10 Determination of the effect of PKA phosphorylation on CKβ protein
stability ... 89
2.10.1 Intrinsic fluorescence assay ... 90
2.11 Determination of the effect of PKA phosphorylation on CKβ subcellular location ... 90
2.11.1 Effect of PKA activator and inhibitor treatment on the subcellular location of CKβ ... 93
2.11.2 Effect of PKA overexpression on the subcellular location of CKβ ... 93
2.12 Determination of the effect of phosphorylation on CKβ oligomeric state... 94
2.12.1 Homo-oligomeric state of CKβ... 94
2.12.2 Hetero-oligomeric state of CKβ ... 95
CHAPTER THREE RESULTS... 97
3.1 Prediction of the potential CK phosphorylating protein kinases ... 97
3.2 Identification of CK phosphorylating protein kinases ... 100
3.2.1 In-gel kinase assay ... 100
3.2.1.1 Recombinant CK protein preparation ... 100
3.2.1.2 In situ phosphorylation of CKα and CKβ ... 100
3.2.2 Effect of PKA specific peptide inhibitor on the phosphorylation of CKβ ... 105
3.2.3 Western detection with PKA specific antibody ... 107
3.3 In vitro phosphorylation of CK ... 107
3.3.1 In vitro phosphorylation of CK with mammalian cell lysates ... 107
3.3.2 Effect of PKA and ATP concentrations on CKβ phosphorylation ... 114
3.4 Mapping of PKA phosphorylation sites in CKβ ... 114
3.4.1 Mass spectrometry analysis ... 117
3.4.2 Verification of CKβ phosphorylation residues by mutagenesis ... 118
3.5 Intracellular phosphorylation of CKβ ... 121
3.5.1 Phosphorylation of CKβ in HEK293 cells ... 123
3.5.2 Stable transfection of pEGFP-C1-NdeICKβ into HEK293 cell line ... 128
3.5.3 Effect of PKA effectors treatment on PKA phosphorylation level of CKβ ... 128
3.5.3.1 Effect of Forskolin and IBMX treatment on PKA phosphorylation level of CKβ ... 132
3.5.3.2 Effect of H-89 treatment on PKA phosphorylation level of CKβ ... 134
viii
3.5.3.3 Effect of epidermal growth factor treatment on PKA
phosphorylation level of CKβ ... 134
3.5.4 Intracellular confirmation of the PKA phosphorylation sites ... 137
3.5.5 Verification of the effect of PKA effectors treatment in HepG2 cell line ... 140
3.6 Effect of PKA phosphorylation on the biochemical properties and subcellular location of CKβ ... 140
3.6.1 Effect of phosphorylation on the biochemical properties of CKβ .... 143
3.6.1.1 Phosphorylation mimic of CKβ ... 147
3.6.2 Effect of PKA phosphorylation on the inhibition of CKβ by hemicholinium-3 ... 156
3.6.3 Effect of PKA phosphorylation on the CKβ pH preference ... 157
3.6.4 Effect of PKA phosphorylation on the protein stability of CKβ ... 161
3.6.4.1 Thermal denaturation ... 161
3.6.4.2 pH denaturation ... 163
3.6.4.3 Chemical denaturation ... 163
3.6.5 Effect of urea denaturation on the intrinsic tryptophan spectrum of CKβ ... 167
3.6.6 Effect of PKA phosphorylation on the sub-cellular location of CKβ ... 173
3.6.6.1 Forskolin and IBMX treatments ... 184
3.6.6.2 The effect of PKA overexpression on CKβ phosphorylation .... 184
3.6.7 Effect of PKA phosphorylation on the oligomeric state of CKβ ... 192
3.6.7.1 Effect of phosphorylation on the formation of CKβ homo- oligomer ... 195
3.6.7.2 Effect of PKA phosphorylation on the formation of CKα/β hetero-oligomer ... 203
CHAPTER FOUR DISCUSSION ... 210
4.1 Prediction of CK phosphorylation ... 211
4.2 Identification of CKβ phosphorylation by PKA ... 212
4.3 PKA phosphorylation sites on CKβ ... 215
4.4 Generation of cell line stably transfected with CKβ ... 220
4.5 Regulation of PKA phosphorylation of CKβ by cAMP ... 221
4.6 Regulation of CKβ phosphorylation by epidermal growth factor ... 222
4.7 Phosphorylation mimic of CKβ ... 225
4.8 PKA phosphorylation changed the biochemical properties of CKβ ... 227
4.8.1 The mechanism to enhance enzyme catalytic activity by phosphorylation... 230
ix
4.9 PKA phosphorylation changed the sensitivity of CKβ to
hemicholinium-3 inhibition. ... 230
4.10 PKA phosphorylation of CKβ did not change its optimum pH ... 232
4.11 PKA phosphorylation changed the stability of CKβ against urea denaturation, but not its thermal and pH stability. ... 233
4.12 PKA phosphorylation did not alter the subcellular location of CKβ ... 236
4.13 PKA phosphorylation did not alter the oligomeric state of CKβ ... 240
4.14 Future studies ... 242
CHAPTER FIVE CONCLUSIONS ... 244
REFERENCES ... 248 APPENDICES
Appendix I Appendix II Appendix III
x
LIST OF TABLES
Table 1.1 The catalytic activity of CK from rat, S. cerevisiae and C.
elegans. ... 12
Table 2.1 List of general instruments used in this study ... 33
Table 2.2 List of consumables used in this study ... 35
Table 2.3 List of chemicals and reagents used in this study ... 36
Table 2.4 List of Escherichia coli strains used in this study... 41
Table 2.5 List of plasmid vectors used in this study ... 42
Table 2.6 List of oligonucleotides used in this study ... 51
Table 2.7 List of softwares used in this study ... 53
Table 2.8 Preparation of buffers for in-gel kinase assay ... 61
Table 2.9 Composition of SDS-PAGE gel... 75
Table 3.1 Prediction of the potential CK phosphorylating protein kinases by using ScanProsite program. ... 98
Table 3.2 Prediction of the potential CK phosphorylating protein kinases by using NetPhosK 1.0 program. ... 99
Table 3.3 Summary of the Mascot Search analysis on the RRASSLSR peptide stretch of CKβ. ... 119
Table 3.4 Summary of the catalytic properties of the phosphorylated CKβ for choline, ethanolamine and ATP substrates. ... 146
Table 3.5 Summary of the catalytic properties of unphosphorylated CKβ, phosphorylated CKβ and S39D/S40DCKβ mutant for choline, ethanolamine and ATP substrates. ... 153
xi
LIST OF FIGURES
Figure 1.1: Kennedy pathway for the biosynthesis of phosphatidylcholine
(PC) and phosphatidylethanolamine (PE). ... 2
Figure 1.2: Sequence alignment of CK from human, mouse and C. elegans... 5
Figure 1.3: Ribbon diagrams of human CKα2 and CKβ. CKα2 was bound with ADP and phosphocholine (PDB 3G15). ... 7
Figure 1.4: The activation of protein kinase A by cAMP. ... 25
Figure 1.5: cAMP signaling pathway. ... 26
Figure 1.6: Overview of study ... 31
Figure 2.1: pET-14b vector map and the sequence of multiple cloning sites. ... 44
Figure 2.2: pGEX-RB vector map and the sequence of multiple cloning sites. ... 45
Figure 2.3: pTriEx-4 Neo vector map and the sequence of multiple cloning sites. ... 46
Figure 2.4: pEGFP-N1 vector map and the sequence of multiple cloning sites for pEGFP-N1 and pEGFP-N1-NdeI vectors. ... 47
Figure 2.5: pEGFP-C1 vector map and the sequence of multiple cloning sites for pEGFP-C1 and pEGFP-C1-NdeI vectors. ... 48
Figure 2.6: pmCherry-N1 vector map and the sequence of multiple cloning sites for pmCherry-N1 and pmCherry-N1-NdeI vectors. ... 49
Figure 2.7: pFLAG-CMV-5b vector map and the sequence of multiple cloning sites. ... 50
Figure 3.1: Verification of pET-14bCKα2 and pET-14bCKβ by digestion with NdeI and BamHI. ... 101
Figure 3.2: Purified CKα2 and CKβ. ... 102
Figure 3.3: Identification of the CK phosphorylating protein kinase by in- gel kinase assay. ... 104
Figure 3.4: Effect of PKA peptide inhibitor (PKI) on the phosphorylation of CKβ. ... 106
Figure 3.5: Confirmation of CKβ phosphorylating protein kinase by Western detection of PKA. ... 108
Figure 3.6: Effect of denaturation time and cell lysate concentration on the dimeric form of PKA... 109
Figure 3.7: In vitro phosphorylation of CK with MCF-7 and HepG2 cell lysates. ... 110
Figure 3.8: Effect of MCF-7 cell lysate concentration on the phosphorylation of CKβ. ... 112
Figure 3.9: Effect of PKI concentration on the phosphorylation of CKβ. ... 113
Figure 3.10: Effect of PKA concentration on the phosphorylation of CKβ. ... 115
xii
Figure 3.11: Effect of ATP concentration on the phosphorylation of CKβ. ... 116 Figure 3.12: In vitro PKA phosphorylation of GST-∆42NCKβ. ... 120 Figure 3.13: In vitro PKA phosphorylation of His-S39A/S40ACKβ. ... 122 Figure 3.14: Verification of the pEGFP-C1-NdeICKβ and pEGFP-C1-
NdeIS39A/S40ACKβ by using NdeI and BamHI. ... 124 Figure 3.15: Detection of PKA phosphorylated CKβ with phosphoPKAS
antibody.. ... 125 Figure 3.16: Expressions of wild type and S39A/S40ACKβ in HEK293 cell. ... 126 Figure 3.17: Phosphorylation of CKβ in HEK293 cells. ... 127 Figure 3.18: Confirmation of the HEK293 cells stably transfected with CKβ
by Western blot detection. ... 129 Figure 3.19: Expression level of GFP-CKβ and endogenous CKβ. ... 130 Figure 3.20: Confirmation of the HEK293 cell stably transfected with CKβ
by fluorescence imaging. ... 131 Figure 3.21: Effect of forskolin and IBMX treatments on the phosphorylation
of CKβ. ... 133 Figure 3.22: Effect of H-89 treatment on PKA induced phosphorylation of
CKβ. ... 135 Figure 3.23: Effect of EGF treatment on the phosphorylation of CKβ. ... 136 Figure 3.24: Effect of H-89 treatment on EGF induced phosphorylation of
CKβ. ... 138 Figure 3.25: Intracellular confirmation of PKA phosphorylation sites. ... 139 Figure 3.26: Effect of forskolin, IBMX and H-89 treatments on the
phosphorylation level of CKβ in HepG2 cell line. ... 141 Figure 3.27: Effect of EGF treatment on the phosphorylation level of CKβ in
HepG2 cell line. ... 142 Figure 3.28: Effect of PKA phosphorylation on the catalytic activity of CKβ
with choline as substrate. ... 144 Figure 3.29: Effect of PKA phosphorylation on the catalytic activity of CKβ
with ethanolamine as substrate. ... 145 Figure 3.30: Effect of PKA phosphorylation on the catalytic activity of CKβ
with ATP as substrate. ... 148 Figure 3.31: Verification of pGEX-RBS39D/S40DCKβ by digestion with
NdeI and BamHI... 149 Figure 3.32: Purified S39D/S40DCKβ from E. coli BL21(DE3). ... 151 Figure 3.33: Effect of phosphorylation mimic mutation on the catalytic
activity of CKβ with choline as substrate. ... 152 Figure 3.34: Effect of phosphorylation mimic mutation on the catalytic
activity of CKβ with ethanolamine as substrate. ... 154
xiii
Figure 3.35: Effect of phosphorylation mimic mutation on the catalytic
activity of CKβ with ATP as substrate. ... 156
Figure 3.36: Effect of PKA phosphorylation and phosphorylation mimic mutation on the HC-3 inhibition. ... 158
Figure 3.37: Effect of PKA phosphorylation on the pH optimum of CKβ with choline as substrate. ... 159
Figure 3.38: Effect of PKA phosphorylation on the pH optimum of CKβ with ethanolamine as substrate. ... 160
Figure 3.39: Effect of PKA phosphorylation on the thermal stability of CKβ. .... 162
Figure 3.40: Effect of PKA phosphorylation on the pH stability of CKβ. ... 164
Figure 3.41: Effect of PKA phosphorylation on CKβ stability after 2 hours of urea denaturation. ... 165
Figure 3.42: Effect of PKA phosphorylation on CKβ stability after 20 hours of urea denaturation. ... 166
Figure 3.43: Effect of CKβ protein concentration on the emission spectrum under native condition. ... 169
Figure 3.44: Effect of urea denaturation on the emission spectrum of wild type CKβ and mutant S39D/S40DCKβ. ... 171
Figure 3.45: Fraction unfolded of wild type CKβ and mutant S39D/S40DCKβ under urea denaturation. ... 172
Figure 3.46: Verification of wild type CKβ and mutant CKβ (S39A/S40ACKβ and S39D/S40DCKβ) in pEGFP-N1-NdeI and pEGFP-C1-NdeI by using NdeI and BamHI. ... 175
Figure 3.47: Protein expression of wild type GFP-CKβ and mutant GFP-CKβ (S39A/S40ACKβ and S39D/S40DCKβ) in HEK293 cell line. ... 176
Figure 3.48: Effect of GFP position on the phosphorylation of CKβ. ... 178
Figure 3.49: Subcellular location of wild type and mutant CKβ. ... 179
Figure 3:50: Effect of mutation on the subcellular location of CKβ. ... 181
Figure 3.51: Verification of the subcellular location of wild type CKβ and mutant CKβ with confocal fluorescence microscope... 182
Figure 3.52: Western detection of wild type CKβ and mutant CKβ in cytoplasmic (Cyto) and nuclear (Nu) cellular fractions. ... 183
Figure 3.53: Effect of forskolin and IBMX treatments on the subcellular location of CKβ. ... 185
Figure 3.54: Verification of the effect of forskolin and IBMX treatment on the subcellular location of CKβ with confocal fluorescence microscope. ... 186
Figure 3.55: PCR amplification of the ORF of PKA catalytic subunit... 187
Figure 3.56: Verification of pmCherry-N1-NdeIPKA and pFLAG-CMV- 5bPKA by using NdeI/BamHI and HindIII/BamHI digestions, respectively. ... 188
xiv
Figure 3.57: Co-expression of mCherry tag PKA catalytic subunit and GFP tag CKβ in HEK293 cell... 190 Figure 3.58: Optimization of plasmid transfection ratio for higher expression
of Flag-PKA and GFP-CKβ. ... 191 Figure 3.59: Effect of PKA co-overexpression on the phosphorylation of
CKβ. ... 193 Figure 3.60: Effect of PKA co-expression on the sub-cellular location of CKβ.
... 194 Figure 3.61: Protein expression of wild type CKβ and phosphorylation
negative mutant S39A/S40ACKβ. ... 196 Figure 3.62: Verification of wild type CKβ and mutant CKβ
(S39A/S40ACKβ and S39D/S40DCKβ) in pTri-Ex4 Neo(His) and pFLAG-CMV-5b. ... 197 Figure 3.63: Co-expression of wild type Flag-CKβ with mutant His-CKβ in
HEK293 cells. ... 199 Figure 3.64: Co-expression of wild type and mutant CKβ with His or Flag
fusion protein in HEK293 cells... 200 Figure 3.65: Co-immunoprecipitation of wild type Flag-CKβ by using His
antibody. ... 201 Figure 3.66: Co-immunoprecipitation of wild type and mutant His-CKβ by
using Flag antibody. ... 202 Figure 3.67: Co-immunoprecipitation of mutant Flag-CKβ by using His
antibody. ... 204 Figure 3.68: Co-immunoprecipitation of mutant His-CKβ by using Flag
antibody. ... 205 Figure 3.69: Verification of pEGFP-C1-NdeICKα2 by digestion with NdeI
and BamHI. ... 206 Figure 3.70: Expression of GFP-CKβ and GFP-CKα2 proteins in HEK293
cells... 207 Figure 3.71: Effect of mutation on the formation of CKα/β hetero-oligomer... 209 Figure 4.1: Comparison of phosphorylation sites on human CKβ, CKα and
yeast CK. ... 217 Figure 4.2: Multiple sequence alignment of human, cow, mouse, and rat
CKβ. ... 219 Figure 4.3: The proposed mechanism of EGF induced CKβ phosphorylation.
... 224 Figure 4.4: Phosphorylation mimicking. ... 226
xv
LIST OF SYMBOLS, ABBREVIATIONS AND ACRONYMNS
ADP Adenosine diphosphate
AKAPs A-kinase-anchoring proteins
AKT Serine/threonine kinase
Ala Alanine
ANS 1-anilino-8-naphthalene sulfonate
APS Ammonium persulfate
Arg Arginine
Asn Asparagine
Asp Aspartate
ATM Ataxia-telangiectasia mutated kinase
ATP Adenosine triphosphate
[γ-32] ATP ATP labeled with gamma 32P
bp Base pair
8-Br-cAMP 8-bromo-cAMP
BSA Bovine serum albumin
cAMP Cyclic adenosine monophosphate
CBP CREB-binding protein
CCl4 Carbon tetrachloride
CCT CTP-phosphocholine cytidylyltransferase CCTα CTP-phosphocholine cytidylyltransferase alpha CCTβ CTP-phosphocholine cytidylyltransferase beta cdK1 Cyclin dependent kinase I
cdK5 Cyclin dependent kinase 5
cDNA Complementary deoxyribonucleic acid
CFTR Cystic fibrosis transmembrane conductance regulator C2H3NaO2 Sodium acetate
CK Choline kinase
CKα Choline kinase alpha
CKA-2 Caenorhabditis elegans CK from family A
CKβ Choline kinase beta
CNG Cyclic nucleotide-gated ion channel
CO2 Carbon dioxide
CPT CDP-choline phosphoryltransferase
CREB cAMP response element-binding protein c-Src Proto-oncogene tyrosine protein kinase Src D50 Half maximal inhibitory concentration DAPI 4',6-Diamidino-2-phenyl indole
DGK Diacylglycerol kinase
DMEM Dulbecco’s modified eagle’s medium
DMSO Dimethyl sulfoxide
DNA Deoxyribonucleic acid
DNAPK DNA dependent protein kinase
dNTP Deoxynucleotide triphosphate
DTT Dithiothreitol
EBP50 Ezrin-Radixin-Moesin (ERM) binding phosphoprotein 50
E. coli Escherichia coli
ECT Phosphoethanolamine cytidylyltransferase EDTA Ethylenediaminetetraacetic acid
xvi
EGF Epidermal growth factor
EGFR Epidermal growth factor receptor EGTA Ethyleneglycoltetraacetic acid
EK Ethanolamine kinase
Epacs Exchange proteins activated by cAMP
EPT Ethanolamine phosphotransferase
EtBr Ethidium bromide
ETS E26 transformation specific
FBS Fetal bovine albumin
FRET Fluorescence resonance energy transfer
GATA Transcription factor that binds to the DNA sequence of
‘GATA’
GFP Green fluorescent protein
Gln Glutamine
Glu Glutamic acid
GPCRs G protein-coupled receptors
GSH Glutathione
GSK3β Glycogen synthase kinase 3 β
GST Glutathione S-transferase
GTPase Guanosine triphosphate (GTP) hydrolyzing enzyme
HC-3 Hemicholinium 3
HCl Hydrochloric acid
HCT-116 Human colorectal carcinoma
HEK293 Human embryo epithelial kidney fibroblasts HeLa Human cervical cancer cell line
HEPES 4-(2-hydroxyethyl)-1-piperazine ethanesulfonic acid HepG2 Human hepatocarcinoma cell line
HIF-1α Hypoxia-inducible factor alpha
His Histidine
HMEC Human primary mammary epithelial cell line
HREs Hypoxia responsive elements
HRP Horse radish peroxidase
IBMX 3-Isobutyl-1-methylxanthine
IC50 Half maximal inhibitory concentration IPTG Isopropyl-β-D-thiogalactopyranoside
IRS Insulin receptor substrate
kcat Enzyme turnover rate
kcat/Km Enzyme catalytic efficiency
KCl Potassium chloride
kDa Kilo dalton
KH2PO4 Potassium phosphate
Km Michaelis constant
LB Luria-Bertani
LDH Lactate dehydrogenase
Leu Leucine
Lys Lysine
MCF-7 Human breast adenocarcinoma cell line M-CPTI Muscle type carnitine palmitoyltransferase I
mCRY2 Muscle cryptochrome 2
MDCK Madin-Darby canine kidney cell line
xvii
Mg2+ Magnesium ion
MgCl2 Magnesium chloride
MgSO4 Magnesium sulfate
mRNA Messenger RNA
NaCl Sodium chloride
NADH Nicotinamide adenine dinucleotide
NaF Sodium fluoride
Na2HPO4 Disodium phosphate
NaOH Sodium hydroxide
Na4P2O7 Sodium pyrophosphate
NIH 3T3 Mouse embryonic fibroblast cells line NLS Nuclear localization signal
NMR Nuclear magnetic resonance
NP-40 Nonidet P-40
OH Hydroxyl
PBS Phosphate buffered saline
PC Phosphatidylcholine
PCho Phosphocholine
PDEs Phosphodiesterases
PDPK1 3-phosphoinositide-dependent protein kinase-1
PCR Polymerase chain reaction
PE Phosphatidylethanolamine
PEG Polyethylene glycol
PEMT Phosphoethanolamine methyltransferase
PEtn Phosphoethanolamine
PEP Phosphoenolpyruvic acid mono-potassium salt phosphoPKAS Phospho-(Ser/Thr) PKA substrate
PI3K Phosphoinositide 3-kinase
PKA Protein kinase A
PKC Protein kinase C
PKG cGMP dependent protein kinase
PKI PKA specific peptide inhibitor
PK-LDH Pyruvate kinase-lactate dehydrogenase PMSF Phenylmethyl-sulfonyl fluoride
PO43-
Phosphate
Prdx6 Peroxiredoxin 6 protein
Pro Proline
pSer Phosphoserine
PTK Protein tyrosine kinase
Pto Serine/threonine protein kinase, ‘Pto’ named from resistance to Pseudomonas syringae pathovar tomato
RAS Rat sarcoma, described the small GTPase protein RFU Relative fluorescence unit
RhoA Ras homolog gene family, member A
RSK Ribosomal S6 kinase
SDS Sodium dodecylsulfate
SDS-PAGE Sodium dodecylsulfate polyacrylamide gel electrophoresis
Ser Serine
SEM Standard error of mean
siRNA Small interfering RNA
xviii
SPSS Statistical package for the social sciences StarD10 START protein of the domain 10
SUS Sucrose synthase
TAE Tris acetate-EDTA buffer
TBS Tris buffered saline
TEMED Tetramethylethylenediamine
TGF-β Transforming growth factor beta
Thr Threonine
Tyr Tyrosine
UV Ultraviolet
V Volt
Vmax Maximum reaction velocity
v/v Volume to volume
w/v Weight to volume
× g Fold gravity
XBP-I(S) X-box-binding protein I spliced form
xix
PEMFOSFORILAN DAN PENGAWALATURAN KOLINA KINASE BETA MANUSIA OLEH PROTEIN KINASE A
ABSTRAK
Kolina kinase (CK) adalah enzim pertama yang terlibat dalam laluan CDP-kolina untuk proses biosintesis fosfatidilkolina yang merupakan komponen utama fosfolipid membran. CK terdiri daripada tiga isoform iaitu CKα1, CKα2 dan CKβ.
Pengawalaturan enzim ini adalah penting dari segi fisiologi. Perubahan metabolik CKα telah dikaitkan dengan pembentukan tumor, manakala mutasi atau pemadaman gen chkβ boleh menyebabkan distrofi otot. Dalam kajian antikanser, perencatan aktiviti CK telah diteroka sebagai strategi terapeutik yang berpotensi.
Pengubahsuaian pasca translasi merupakan salah satu mekanisme untuk mengawal fungsi CK. Semakin banyak bukti menunjukkan bahawa fungsi CK dalam yis dan CKα dalam manusia dikawalatur oleh pemfosforilan. Namum begitu, pemfosforilan CKβ tidak pernah dilaporkan. Dalam kajian ini, protein kinase A (PKA) telah dikenalpasti sebagai protein kinase yang bertanggungjawab dalam pemfosforilan CKβ melalui analisis in-gel kinase. Pemfosforilan oleh PKA telah disahkan melalui teknik analisis perencat PKA dan blot Western. Analisis in vitro dengan menggunakan PKA komersil juga membuktikan bahawa CKβ merupakan substrat untuk pemfosforilan PKA. Pemfosforilan ini berlaku pada terminal-N CKβ iaitu asid amino serine 39 dan 40. Pemfosforilan CKβ telah dilihat dalam sel embrio ginjal manusia (HEK293) dan sel karsinoma hati manusia (HepG2). Rawatan forskolin dan 3-isobutil-1-metilxantin meningkatkan tahap pemfosforilan pada CKβ manakala kesan tersebut direncatkan oleh perencat PKA (H-89). Tahap pemfosforilan CKβ juga ditingkatkan oleh rawatan faktor pertumbuhan epidermis. Seterusnya, kesan
xx
pemfosforilan terhadap ciri-ciri biokimia CKβ juga dikaji. Pemfosforilan PKA telah meningkatkan aktiviti pemangkinan CKβ terhadap kolina, etanolamina dan ATP.
Nilai Vmax untuk kolina, etanolamina dan ATP telah masing-masing meningkat sebanyak 47.1%, 81.8% dan 50.8%. Pemfosforilan PKA juga telah meningkatkan tarikan CKβ terhadap substrat kolina dan ATP, tetapi pemfosforilan menurunkan tarikan CKβ terhadap substrat etanolamina. Kecekapan pemangkinan CKβ untuk kolina dan ATP telah meningkat sebanyak 121.0% dan 97.5% masing-masing. Kesan pemfosforilan PKA terhadap ciri-ciri biokimia CKβ telah ditiru oleh mutasi berganda pada serine yang difosforilasi dengan penukaran kepada aspartat. Pemfosforilan juga meningkatkan sensitiviti CKβ terhadap perencatan oleh hemicholinium-3 (HC-3) yang merupakan perencat CK yang kuat. Nilai IC50 CKβ terfosforilasi (50 μM) adalah 29 kali ganda lebih rendah daripada enzim tidak terfosforilasi (1.45 mM).
Selain itu, pemfosforilan juga mengurangkan kestabilan CKβ terhadap penyahaslian urea. Sebaliknya, pemfosforilan tidak menjejaskan pH optima, lokasi subsel dan status oligomer CKβ. Kajian ini melaporkan fosforilasi dan pengawalaturan CKβ oleh PKA untuk kali pertama. Pengetahuan ini memberikan pandangan baru terhadap pengawalaturan intrasel ciri-ciri pemangkinan CKβ yang mungkin merupakan mekanisme penting untuk mengawal metabolisme lipid and pertumbuhan sel.
xxi
PHOSPHORYLATION AND REGULATION OF HUMAN CHOLINE KINASE BETA BY PROTEIN KINASE A
ABSTRACT
Choline kinase (CK) is the first enzyme involved in CDP-choline pathway for the biosynthesis of phosphatidylcholine, the major component of membrane phospholipid. CK exists as three isoforms, which are CKα1, CKα2 and CKβ. The regulation of these enzymes is physiologically important. Metabolic alterations of CKα are associated with tumorigenesis, while mutation or deletion of chkβ gene leads to the development of muscular dystrophy. In anticancer research, inhibition of CK activity has been explored as a potential therapeutic strategy. Post-translational modification is one of the mechanisms to regulate the function of CK. Growing evidences support that yeast and human CKα are regulated by phosphorylation but the phosphorylation of CKβ has never been reported. In this study, protein kinase A (PKA) was identified as the protein kinase responsible for the phosphorylation of CKβ by in-gel kinase assay. PKA phosphorylation was confirmed with specific PKA inhibitor and Western blotting. In vitro assay with commercial PKA further supported CKβ as the substrate for PKA phosphorylation. The phosphorylation occurred at serine 39 and 40 residues in the N-terminal region of CKβ.
Phosphorylation of CKβ was observed in human embryonic kidney cells (HEK293) and liver hepatocellular carcinoma cells (HepG2). Forskolin and 3-isobutyl-1- methylxanthine treatment increased the phosphorylation level of CKβ, while the phosphorylation was inhibited by PKA inhibitor (H-89). The phosphorylation level of CKβ was also increased by epidermal growth factor. The effects of PKA
xxii
phosphorylation on the biochemical properties of CKβ were subsequently examined.
PKA phosphorylation increased the catalytic activities of CKβ with choline, ethanolamine and ATP as substrates. The Vmax values for choline, ethanolamine and ATP were increased by 47.1%, 81.8% and 50.8%, respectively. PKA phosphorylation improved the affinity of CKβ for choline and ATP, but decreased the affinity of CKβ for ethanolamine. Consequently, the catalytic efficiencies of CKβ for choline and ATP were increased by 121.0% and 97.5%, respectively. The same effects of PKA phosphorylation on the biochemical properties of CKβ were mimicked by double mutation of the phosphorylated serines to aspartates. PKA phosphorylation also dramatically increased the sensitivity of CKβ to hemicholinium-3 (HC-3), a potent inhibitor of CK. The IC50 value for phosphorylated CKβ (50 μM) was 29 times lower than the unphosphorylated enzyme (1.45 mM). In addition, PKA phosphorylation also decreased the stability of CKβ protein against urea denaturation. On the contrary, phosphorylation did not affect the optimum pH, subcellular location and oligomeric state of CKβ. This study reports the phosphorylation and regulation of CKβ by PKA for the first time. The knowledge provides new insight into the intracellular regulation of CKβ catalytic properties by phosphorylation that might be an important mechanism to modulate lipid metabolism and cell growth.
1
CHAPTER ONE
INTRODUCTION
1.1 Kennedy pathway
Kennedy pathway (Figure 1.1) which consists of CDP-choline and CDP- ethanolamine pathways decribes the de novo biosynthesis of major phospholipid components of the cell (Kennedy and Weiss, 1956). In CDP-choline pathway, CK catalyzes the phosphorylation of choline to phosphocholine (PCho) using adenosine triphosphate (ATP) and magnesium (Mg2+) as substrate and cofactor, respectively (Ishidate, 1997). The second enzyme in this pathway is CTP-phosphocholine cytidylyltransferase (CCT) which converts the PCho into CDP-choline. CDP- choline phosphoryltransferase (CPT) catalyzes the final condensation of CDP- choline to form PC (Kent, 1990, Gibellini and Smith, 2010). For CDP-ethanolamine pathway, ethanolamine is converted into phosphoethanolamine (PEtn) by EK and followed the similar steps as the CDP-choline pathway (Gibellini and Smith, 2010).
In liver, the end product of CDP-ethanolamine pathway, the PE, can be converted into PC by phosphoethanolamine methyltransferase (PEMT) (Li and Vance, 2008).
1.2 Choline kinase
Human CK is composed of CKα1, CKα2 and CKβ isoforms. CKα and CKβ are encoded by two separate genes, chkα (NCBI Gene ID: 1119) and chkβ (NCBI Gene ID: 1120) in chromosomes 11q13.2 and 23q13.33, respectively. CKα undergoes alternative splicing to form CKα1 (NCBI reference sequence: NP_005189) and CKα2 (NCBI reference sequence: NP_001268) with the calculated molecular size of
2
CK CCT CPT
Choline PCho CDP-choline PC
PEMT
EK ECT EPT Ethanolamine PEtn CDP-ethanolamine PE
Figure 1.1: Kennedy pathway for the biosynthesis of phosphatidylcholine (PC) and phosphatidylethanolamine (PE). Choline kinase (CKα or CKβ) phosphorylates choline to form phosphocholine (PCho). PCho is converted into CDP-choline by CTP-phosphocholine cytidylyltransferase (CCT). CDP-choline is condensed into PC by CDP-choline phosphoryltransferase (CPT). Ethanolamine kinase (EK) phosphorylates ethanolamine to form phosphoethanolamine (PEtn). PEtn is converted into CDP-ethanolamine by phosphoethanolamine cytidylyltransferase (ECT). CDP-ethanolamine is condensed into PE by ethanolamine phosphotransferase (EPT). PE can be converted into PC by phosphoethanolamine methyltransferase (PEMT). Figure is adapted from Aoyama et al. (2004).
3
50 and 52 kDa, respectively (Aoyama et al., 2004); whereas, CKβ (NCBI reference sequence: NP_005189) is encoded by a separated gene with the calculated molecular size of 45 kDa.
The enzyme activity of CK was first described in brewer’s yeast by Wittenberg and Kornberg (1953). Due to the difficulty in purifying yeast CK, the understanding of CK progressed slowly until the highly homogeneous CK was obtained from rat tissues in 1984 (Ishidate et al., 1984, Porter and Kent, 1990, Uchida and Yamashita, 1990). The cDNA of rat and human CKα1 was first isolated from rat liver and human glioblastoma (Hosaka et al., 1992, Uchida and Yamashita, 1992a). Amino acid sequence comparison between rat and human CKα1 shows 84.9% identity.
Later, the second transcript of rat CKα1, termed CKα2 was isolated and characterized (Uchida, 1994). CKα2 and CKα1 are differed by an extra stretch of 18 amino acids on the CKα2 isoenzymes with a discrepancy of 2 kDa. The insertion of extra amino acids on the CKα2 significantly increased its substrate affinity for choline (Malito et al., 2006). The additional stretch of amino acid was proposed to be the important region to facilitate the conformational change of the enzyme upon substrate binding (Malito et al., 2006).
Another isoenzyme of CK, named CKβ was first cloned from rat liver and characterized by Aoyama et al. (1998). The amino acid sequence of rat CKβ shows 57-59% identity with the amino acid sequence of rat CKα1 and CKα2. The human homolog of rat CKβ was identified later by Yamazaki et al. (1997). The cDNA sequence of human chkβ was cloned in a large-scale cDNA sequence project and the sequence was deposited in NCBI with GenBank accession number: BC082263,
4
BC101488, BC113521(Strausberg et al., 2002). CKβ consists of 395 amino acids.
CKβ is less studied as compared to the CKα isoenzyme. CKβ is not as active as CKα.
Most importantly, CKβ is not involved in tumorigenesis as compared to CKα, which is the main factor for most of the studies were focused on CKα isoenzyme (Gallego- Ortega et al., 2009). CKβ started to attract attention when the association between CKβ and muscular dystrophy was reported (Wu et al., 2009).
CK from Caenorhabditis elegans has been extensively studied also. The second isoenzyme of C. elegans CK from family A (CKA-2) shows 48% identity with the human CKα2. Due to its high similarity in gene sequence and biochemical properties to human CKα2, CKA-2 has been used as the model for structure function study for better understanding of CK function (Gee and Kent, 2003).
1.3 Structure of choline kinase
CKs consist of two clusters of highly conserved motif, Brenner’s and CK/EK motif (Figure 1.2). Brenner’s motif with consensus sequence of hxHxDhx3N (h refers to large hydrophobic residue and x refers to an unknown amino acid residue) is found in most of the protein kinases that catalyze the transfer of phosphoryl groups (Brenner, 1987), whereas, CK/EK motif (hxhhDhEx4Nx3hDhx2HhxE) is conserved among the CK from different organisms (Aoyama et al., 2000).
CKA-2 from C. elegans was the first crystal structure to be solved (Peisach et al., 2003). The crystal structure of CKA-2 revealed the enzyme as a homodimeric protein. The structure of CKA-2 is similar with those eukaryotic protein kinases (ePK) and aminoglycoside phosphotransferases (AP) although their amino acid
5
Figure 1.2: Sequence alignment of CK from human, mouse and C. elegans. The NCBI accession numbers for human CKs; hCKα1, hCKα2 and hCKβ are NP_997634, NP_001268, and NP_005189. The NCBI accession numbers for mouse CKs; mCKα and mCKβ are NP_038518 and NP_031718. The NCBI accession number for C. elegans; CKA-2 is NP_001024480. Figure is adapted from Malito et al. (2006).
Brenner’s motif CK/EK motif
6
sequences are different. Structural comparison of CKA-2 with the catalytic subunit of protein kinase A (PKA) and aminoglycoside 3’-phosphotransferase [APH(3’)-IIIa]
shows conserved structural cores of a N and C-terminal domains. The conserved structure of the N-terminal region consists of five strands of β sheet (first five strands) and one helix (second helix), whereas the C-terminal domain consists of three helices (third to fifth helices) and four strands of β sheet (ninth to twelfth strands). The smaller N-terminal domain is connected to the large C-terminal domain by a short linker. The crystal structure of CKA-2 also showed the location of Brenner’s and the CK/EK motif at the C-terminal domain. There was no bound substrate, so the ATP site was predicted based on the existing structure of highly similar proteins ePK and AP. The choline binding site was proposed to be near to the ATP binding pocket which was formed by several structurally flexible loops (Peisach et al., 2003).
Later, Malito et al. (2006) solved the crystal structure of human CKα2 protein with bound ADP and PCho to reveal the molecular details of ATP and the choline binding sites on CKα2. The ribbon diagram of the CKα2 crystal structure is shown in Figure 1.3. CKα2 was also crystalized as homodimeric form. The structure of CKα2 is very similar with the structure of CKA-2. As compared to the CKA-2, a small difference is found on the fifth helix of CKα2. However, this region is not part of the catalytic region.
7
`
CKα2 CKβ
Figure 1.3: Ribbon diagrams of human CKα2 and CKβ. CKα2 (PDB 3G15) and CKβ were bound with the hemicholinium (HC-3) (PDB 3FEG). C and N indicate C and N-terminal regions. The dimeric structure shown of CKβ was the suggested biological model based on the asymmetric unit of CKβ (Hong et al., 2010).
N C
C N
C
C N
N
8
CKα2 overexpressed in E. coli BL21(DE3) was the N-terminal truncation mutant without the first 49 amino acids (Malito et al., 2006). The truncated version of the CKα2 shows very similar biochemical properties as the full length protein and this suggests that the N-terminus of CKα2 is not important for enzyme catalysis. On the crystal structure, the first visible residue is on the Pro 85 which indicates that the first 30 amino acids on the crystal structure are disordered. In this structure, the choline binding pocket is described as a deep hydrophobic groove with a rim of negatively charged residues. ATP is bound within a cleft between two domains (the N and C-terminal domain) of the enzyme. Residues from both N and C-terminal lobes contribute to the formation of large pocket for ATP binding. Upon binding of choline, it undergoes conformational changes affecting the N-terminal domain and the ATP-binding loop.
Subsequently, Hong et al. (2010) solved the crystal structure of another CK isoenzyme, CKβ in complex with the potent CK inhibitor, hemicholinium (HC-3).
HC-3 molecule was bound onto the choline binding pocket. The structure of CKβ exhibited the same bilobal architecture as CKα2 with the major difference being the C-terminal lobe (Figure 1.3) (Hong et al., 2010). This difference lowers the sensitivity of CKβ towards HC-3 inhibition (Hong et al., 2010). Besides, the N- terminal truncated (35 amino acids) CKβ was shown as a monomer rather than dimer like the CKα2 and CKA-2. The first 35 amino acid might be important for the oligomeric formation.
9
Mutagenesis study was performed on the conserved Brenner’s and CK/EK motif based on information obtained from the crystal structural of CK and other ePK with similar protein structure. Aspartate residue on Brenner’s motif was responsible for the removal of the proton charge from the hydroxyl group of ATP (Zheng et al., 1993). Thus, mutation on the Asp 255 residue of CKA-2 and Asp 306 residues of CKα2 caused the total loss of CK activity (Malito et al., 2006, Yuan and Kent, 2004). On the other hand, mutation on the CK/EK motif also impaired the catalytic activity of CK. These residues were shown to coordinate the enzyme co-factor, Mg2+
ion by two carboxyl oxygen atoms. Mutation on Asn 260 and Asp 301 on CK/EK motif of CKA-2 resulted in the loss of the enzyme activity (Peisach et al., 2003;
Yuan and Kent, 2004). However, mutation on the Asn 330 on the CK/EK motif of CKα2 did not alter the catalytic activity although the crystal structure of CKα2 showed a direct contact of this residue with Mg2+ ion (Malito et al., 2006). Besides, Ser 121 in the ATP loop is also important for the activity of CK. Mutation of this residue decreased the catalytic efficiency of the enzyme. Hydroxyl group on this position is essential for a full activity of the CKα2 protein.
1.4 CK oligomeric structures
CK exists as dimer, tetramer or higher oligomer. Homo or hetero-oligomer formation of CK isoforms had been reported. According to Aoyama et al. (2002), the most active form of CK in mouse is the homo-oligomer of α/α, followed by hetero-oligomer of α/β and the least active CK is homo-oligomer of β/β. The mRNA abundance of CKα isoform was the highest in testis, while the expression of CKβ isoform was relatively high in the heart and liver. In liver tissue, hetero-oligomer contributed 60% of the total activity, while the homo-oligomer contributed 20%
10
each to the remaining activity. In contrast, the enzyme obtained from heart tissue showed 70% of activity from homo-oligomer of β/β, 25% from hetero-oligomer of α/β, and <5% was contributed by the homo-oligomer of α/α (Aoyama et al., 2002).
These observations indicate that the expression, distribution and the combination of the CK oligomer were tissue type-dependent.
The dimeric structure of CKα2 protein is stabilized by the dimer interface formed at the second α-helix (Glu 175 ̶ Arg 190) of each monomer (Malito et al., 2006). For CKA-2, another dimer interface is identified at the first helix (Pro 50 ̶ Leu 64) and the S-shaped loop [formed by the fourth (Ala 167 ̶ His 174) and fifth (Leu 194 ̶ Thr 208) helices] (Peisach et al., 2003). The extra dimer interface on CKA-2 is absent from the CKα2 protein as the structure at this region is disordered (Malito et al., 2006). In mouse CK, the important regions for oligomer formation were also identified (Liao et al., 2006). The amino acids between first (Pro 73 ̶ Arg 85) and ninth helices (Gln 424 ̶ Lys 430) as well as single amino acid on seventh helix, Asp 320 are critical for oligomer formation of CKα. The region between first (Arg 35 ̶ Arg 62) and tenth helices (Gln 379 ̶ Lys 385) is important for oligomer formation of CKβ (Liao et al., 2006).
1.5 CK subcellular location
In early studies, CK had been reported as a cytosolic protein. The CK activity was detected in the cytosol fraction of the cells (Uchida and Yamashita, 1990; Aoyama et al., 2002). In addition, Miyake and Parsons (2011) also showed CKα as a cytoplasmic protein when overexpressed in breast cancer cell line. CKα was translocated from cytoplasm to the membrane of the cell when co-expressed with its
11
interacting partner, the epidermal growth factor receptor (EGFR) (Miyake and Parsons, 2011). Besides, there was also a report on nucleus translocation of CKα at the mitotic phase of the cell cycle (Gruber et al., 2012). These observations provide the evidences for CKα translocation into different cell compartments. However, no information is available for the subcellular location of CKβ.
1.6 Biochemical properties of CK
Extensive biochemical characterizations of CK from rat, S. cerevisiae and C.
elegans had been performed in the earlier studies (Gee and Kent, 2003, Ishidate et al., 1984, Kim et al., 1998, Porter and Kent, 1990, Uchida and Yamashita, 1990, Ishidate et al., 1985). The details were summarized in Table 1.1 (Aoyama et al., 2004). Among rat CK, CKα2 possessed the highest specific activity, followed by CKα1 and the least active form of CK is CKβ (Ishidate et al., 1984; Ishidate et al., 1985; Porter and Kent, 1990; Uchida and Yamashita, 1990). As compared to rat CK, the similar characteristic was reported for human CK (Hong et al., 2010). Yeast CK (S. cerevisiae) was almost as active as the rat CKα2 with lower affinity towards choline (Kim et al., 1998). In contrast, C. elegans CK was less active than the yeast CK. However, its affinity toward choline was higher than the yeast CK (Gee and Kent, 2003).
All the purified CKs possess EK activity (Aoyama et al., 2004; Hong et al., 2010).
Therefore, the nomenclature of CK becomes choline/ethanolamine kinase in the early nineties. The subsequent discovery of Drosophila EK established the existence of a separate gene (ek) encoding an ethanolamine specific kinase (Uchida, 1997). In human, the cDNA of the ek1 gene was isolated and characterized by Lykidis et al.
12
Table 1.1 The catalytic activity of CK from rat, S. cerevisiae and C. elegans.
The table is adapted from Aoyama et al. (2004).
Source Oligomeric
form S.A(mmol
/min/mg) Km
Choline (μM)
Km
ATP (mM)
Corresponding protein
Rat kidneya,b dimer 3.3 100 1.5 CKβ
Rat liverc tetramer 143 13 0.04 CKα2
Rat braind dimer 40 14 1.0 CKα1
S. cerevisiaee (Recombinant)
dimer 128 270 0.09 CKI
C. elegansf
(Recombinant) Dimer
(oligomer) 43 24
1.6 mM 13 mM
2.4 0.72
CKA-2 CKB-2
a Ishidate et al. (1984)
b Ishidate et al. (1985)
cPorter and Kent (1990)
d Uchida and Yamashita (1990)
e Kim et al. (1998)
f Gee and Kent (2003) S.A: specific activity.
13
(2001). It possessed high EK activity with negligible CK activity. In general, CK prefers choline rather than ethanolamine as substrate. The affinity of CK for choline is higher than ethanolamine (Porter and Kent, 1990). Gallego-Ortega et al. (2009) reported that both human CKα and CKβ isoforms showed higher affinity toward choline rather than ethanolamine. However, human CKβ showed a higher ethanolamine kinase activity than CK activity in the cells (Gallego-Ortega et al., 2009). The overexpression of CKα in human derived cell line increased the production of both PCho and PEtn. However, CKβ overexpression increased PEtn production, but not PCho (Gallego-Ortega et al., 2009), which showed that CKβ catalyzed the phosphorylation of ethanolamine rather than choline when both substrates are present in the cell.
In terms of the substrate affinity, CKα purified from bacterial expression system possesses a higher affinity for both choline and ethanolamine than the CKβ (See Too, 2006). CKα overexpressed from human derived cell line (crude cell lysate) also gave similar results (Gallego-Ortega et al., 2009). However, Hong et al. (2010) showed a contradicting result. Their purified CKβ from bacterial expression system had a higher substrate affinity for choline than CKα although the catalytic efficiency of CKα remained higher than the CKβ isoenzyme.
14 1.7 The roles of choline kinase
1.7.1 Cell proliferation
CK is involved in cell proliferation. The product of CK, PCho was shown to induce mitogenesis. In human primary mammary epithelial cells (HMEC) and mouse embryonic fibroblast cells (NIH 3T3), PCho production was increased by growth factor, insulin and hydrocortisone treatments, which were the effectors for normal cell proliferation (Ramirez de Molina et al., 2004, Kiss and Chung, 1996). The treatment also increased the DNA synthesis of HMEC cells and promotes G1 to S phase transition of the cell cycle (Ramirez de Molina et al., 2004). CK overexpression was found to alter the expression of 31 genes and promote cell proliferation (Ramirez de Molina et al., 2008). The expression of transforming growth factor beta (TGF-β), one of the important proteins in G1 cell cycle arrest was down-regulated by the overexpression of CK (Ramirez de Molina et al., 2008). The role of CK in down regulating cell arrest was further confirmed by specific CK inhibitor, MN58b. MN58b was shown to reverse the TGF-β mediated transcriptional activation which was activated by CK overexpression (Ramirez de Molina et al., 2008). In addition, Yamashita and Hosaka (1997) showed that CK mRNA and protein levels were elevated during the exponential phase of tumor cell growth but decreased in the stationary phase. This leads to the accumulation of PC which is the end product of CK. The accumulation of PC resulted from an increased level of CK protein at the enterance of S phase was also found to be essential for cell division (Lykidis and Jackowski, 2001).
15 1.7.2 Tumorigenesis
CK is overexpressed in both tumorous tissue and tumor derived cell lines (Ramirez de Molina et al., 2002a, Ramirez de Molina et al., 2002b, Ramirez de Molina et al., 2007, Hernando et al., 2009). The overexpression of CK was detected at both mRNA and protein levels (Eliyahu et al., 2007). Furthermore, the levels of the choline metabolites were also elevated in cancerous cells (Katz-Brull et al., 2002, Iorio et al., 2005, Eliyahu et al., 2007). These observations raise the question of whether CK acts as an oncogene or as a byproduct of the physiological alteration associated with oncogene expression.
Earlier study showed that overexpression of CK was the consequence of tumorigenic transformation. Bhakoo et al. (1996) showed the elevation of PCho in ras oncogene transformed cell. The activity of CK was up-regulated by ras protein through the direct effectors of Ral-GDS and phosphoinositide 3-kinase (PI3K), two of the important mediators for tumorigenesis (Ramirez de Molina et al., 2002a).
Treatment with the PI3K inhibitor (PI-103) was shown to suppress the expression level of CK and in turn decrease the level of PCho production and total choline metabolites in the cells (Al-Saffar et al., 2010). In addition, the activity of serine/threonine kinase (AKT), one of the protein kinase in PI3K pathway was also regulated by CK (Chua et al., 2009). Besides, the breakdown product of PC, phosphatidic acid was found as the key activator for the PI3K pathway (Yalcin et al., 2010). Knockdown of the phospholipase D, the enzyme to hydrolyze PC was shown to attenuate the activation of AKT (Toschi et al., 2009).
16
Oncogenic property of CK in tumor transformation was reported by Ramirez de Molina et al. (2005). Overexpression of CK induced oncogenic transformation of human embryo epithelial kidney fibroblasts (HEK293) and Madin-Darby canine kidney cells (MDCK). Co-expression of CK with RhoA from GTPases family further potentiates anchorage independent growth and tumorigenesis. This suggested that CK plays role in Rho-mediated tumor transformation. The role of CK in cell transformation was further confirmed by the specific CK inhibitor, MN58b, which inhibited the CK mediated tumorigenesis (Ramirez de Molina et al., 2005).
Later, Gallego-Ortega et al. (2009) showed that the CKα isoform was oncogenic and able to induce cell transformation, but not CKβ isoform. The overexpression of CKβ did not induce tumor growth. In addition, the study also showed the CKα but not CKβ mRNA was elevated in a panel of mammary cancer cell line as compared to the non-tumorogenic mammary cell lines (Gallego-Ortega et al., 2009).
CKα is an important enzyme in cancer cell survival. CKα knockdown in cervical cancer cell line (HeLa) using small interfering RNA (siRNA) resulted in cell death (Glunde et al., 2005, Falcon et al., 2013). However, the inhibition of CKα activity with specific inhibitor, MN58b is not sufficient to induce cell death. This result indicates that the non-catalytic role of CKα is important for the cancer cell survival (Falcon et al., 2013). Cells with single CKβ or double CKα/CKβ knockdown have no aberrant phenotype compared to the single knockdown of CKα (Gruber et al., 2012). In this case, the balance of the CKα and CKβ isoforms also important for cancer cell survival and simultaneous knockdown of CKβ reduced or abolished the cell-killing effect of single CKα knockdown (Gruber et al., 2012).
17 1.7.3 Differential role of CKα and CKβ
CKα and CKβ knockout mice generated a different phenotype. Early embryonic lethality was observed on CKα knockout mice (Wu et al., 2008) while CKβ knockout mice developed muscular dystrophy (Sher et al., 2006). Thus, CKα plays important role in early development of mouse embryo while CKβ is involved in the later part of mouse development. Heterozygous CKα knockout mice (ckα+/-) have a normal early embryonic development and the biosynthesis of PC was unaffected although the PC synthesis was decreased by 30% (Wu et al., 2008). No significant compensation was found from CKβ in homo and heterozygous CKα knockout mice because the mRNA and protein levels of CKβ were not increased in both of the CKα knockout mice (Wu et al., 2008). The evidence supported that CKα and CKβ have different roles in maintaining the PC homeostasis.
Sher et al. (2006) reported that CKβ knockout mice developed hindlimbs muscular dystrophy and neonatal forelimb bone deformity. Total CK activity was generally decreased in all tissues, however muscle dystrophy was only observed in skeletal muscle of hindlimbs (Sher et al., 2006). CKβ was involved in PC metabolism of hindlimb muscle, while CKα was responsible for PC synthesis in forelimb muscle as muscular dystrophy did not develop in forelimbs due to CKα abundance and stable PC homeostasis (Wu et al., 2010). CKα was not overexpressed in the CKβ knockout mice to compensate for the loss of CKβ activity (Wu et al., 2009).
18
Mitochondria abnormalities were observed on the skeletal muscle of CKβ knockout mice whereby mitochondria were absent on the center of muscle fibers and large mitochondria was found at the peripheral fiber (Mitsuhashi et al., 2011a, Wu et al., 2009). The PC level was low in the isolated mitochondria. The activity of the respiratory chain enzyme (complex I-IV) and the ATP production of the defected mitochondria in CKβ knockout mice also decreased. In addition, the molecular markers of mitophagy were found in the defected mitochondria suggested that the loss of ckβ gene resulted in mitochondria dysfunction and led to the development of muscular dystrophy (Mitsuhashi et al., 2011a).
In human, heterozygous mutation of chkβ was detected in 15 patients with congenital muscular dystrophy from Japan, Turkey and Britain (Mitsuhashi et al., 2011b). CK activity was not detected in the muscle tissue of the patients and the PC content of the frozen biopsied muscle tissues was lower than normal individual. A total of 11 mutations were identified and these mutations mostly truncated the protein or eliminated the conserved region of CKβ protein (Mitsuhashi et al., 2011b).
Besides muscular dystrophy, patients with chkβ gene mutation also have severe mental retardation (Mitsuhashi et al., 2011b). Previously, the decreased CKβ expression has been linked with narcolepsy, a sleep disorder (Miyagawa et al., 2008). The findings supported the involvement of CKβ in the maintenance of normal brain function in humans.
Gutierrez Rios et al. (2012) also reported the chkβ gene mutation in an American patient with congenital muscle dystrophy. The mutation also truncated the protein by introducing a stop codon at Gln 292. Giant mitochondria containing densely packed
19
and whorled cristae were observed in the tissue biopsy. The authors concluded that CKβ was involved in mitochondria-associated membrane phospholipid metabolism (Gutierrez Rios et al., 2012). They also postulated that CKβ gene defect could consequently affect the production of active human muscle type carnitine palmitoyltransferase I (M-CPTI) protein, a key lipid transport enzyme in the outer membrane of mitochondrial. The transcription of chkβ and cpt1β genes were bicistronic (Yamazaki et al., 2000). In consequence, the mitochondria dysfunction might also be due to the defect of cpt1β gene expression which affect the activity of mitochondrial respiratory chain.
1.8 Regulation of choline kinase 1.8.1 Transcriptional level
Several studies on the promoter regions have shed light on the transcription regulation of CK genes by transcription factors. Aoyama et al. (2007) reported an up-regulation of CKα expression in mouse liver after treatment with carbon tetrachloride (CCl4). The overexpression of CK was contributed by the binding of c- jun transcription factor to an AP-1 element (at ̶ 866 bp upstream of translational start site) upon treatment with CCl4 (Aoyama et al., 2007).
In human, the putative promoter region upstream of ckα gene ( ̶ 2.3 kb region upstream of translational start site) was isolated by Glunde et al. (2008). Their study showed that the expression of CKα was regulated by hypoxic condition. Eight hypoxia responsive elements (HREs) sites were predicted by promoter sequence analysis. The responsive elements composed of two non-overlapping regions which up-regulated ( ̶ 1068/ ̶ 851) and down-regulated ( ̶ 670/+1) the CKα expression
20
during hypoxia (Glunde et al., 2008). Highly repressive element was found at the position ̶ 225/ ̶ 222 bp upstream of translational start site (Bansal et al., 2012). The binding of hypoxia-inducible factor (HIF-1α) on the respective HRE sites was shown to suppress the mRNA expression of CKα (Glunde et al., 2008; Bansal et al., 2012).
Recently, Yee (2012) reported the isolation of promoter region of human ckβ gene ( ̶ 2 kb region upstream of translational start site). GATA and Ets were identified as the important transcription factors that suppressed the expression of CKβ expression (Yee, 2012).
1.8.2 Translational level
To date, translational regulation of CK had not been reported. The translational regulation of CTP-phosphocholine cytidyltransferase (CCT), a second enzyme in CDP-choline pathway had been postulated in X-box-binding protein (XBP-I(S)) transducted fibroblasts (Sriburi et al., 2007). The expression of the XBP-I(S) was shown to increase the assembled (80S) ribosomes which enhanced the protein synthesis of CCT. However, the detailed mechanism of XBP-I(S) in enhancing the translation of CCT needs further investigation.
1.8.3 Post-translational level
Post-translational regulation of CK by phosphorylation was first described in yeast CK. Yeast CK is phosphorylated by protein kinase A (PKA) and protein kinase C (PKC) (Kim and Carman, 1999, Yu et al., 2002, Choi et al., 2005). Phosphorylation of yeast CK with PKA and PKC increased the catalytic activity by 1.9 and 1.6 folds,
