CHARACTERISATION OF a-CHIMAERIN ISOFORMS AND a2 SH2 DOMAIN MUTANTS EXPRESSED IN NEUROBLASTOMA CELLS

A Thesis by

NANSI EMMA CANN

Submitted to University College London

for the Degree of Doctor of Philosophy, PhD

2000

Miriam Marks Department of Neurochemistry Institute of Neurology University College London Queen Square London WC1N 3BG ProQuest Number: 10609000

All rights reserved

INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.

In the unlikely event that the author did not send a com plete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest

ProQuest 10609000

Published by ProQuest LLC(2017). Copyright of the Dissertation is held by the Author.

All rights reserved. This work is protected against unauthorized copying under Title 17, United States C ode Microform Edition © ProQuest LLC.

ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 Dedicated to my parents, Mary and Brian Cann

Dedication Acknowledgements

I wish to thank Professor Louis Lim for the opportunity to study for a PhD in his laboratory as a member of the Glaxo-IMCB group. I would also like to thank both him and my co-supervisor Dr Christine Hall for their help and encouragement during my period of study, especially Dr Hall for all her advice during the writing of this thesis. I would also like to thank the other laboratory members; Clinton Monfries, Rob Kozma, Sally Williams, Kate Marler, Giovanna Ferrari, Elena Prigmore, Ric Passey, and especially Shanta Cariese, Shula Sarner and Sheila Govind for their help in various techniques and in providing an enjoyable working environment. Thanks also to my friends and family for their support and tolerance during my study period. Finally, to Stuart who has helped me through all the hard times, supported me and tolerated a lot, thank you for everything.

Acknowledgements 3 ABSTRACT

Rac is a member of the Rho family of low molecular weight (p21s) which is involved in diverse processes including regulation of the and transcriptional activation. Chimaerin, a multidomain GTPase activating protein (GAP) downregulates Rac by increasing its intrinsic rate of GTP hydrolysis. Two splice variants of the chimaerin gene differ in tissue and developmental expression patterns and a2-chimaerin contains an N terminal SH2 domain which is absent from a l- chimaerin. The distribution and morphological effects of the chimaerins, a2-chimaerin SH2 domain mutants and potential a2-chimaerin targets in N1E 115 neuroblastoma cells were investigated. The distribution of al-chimaerin was predominantly cytoskeletal and a2-chimaerin cytosolic. In transiently transfected N1E 115 cells, al-chimaerin was concentrated in the perinuclear region and its expression induced cell rounding, whilst a2-chimaerin was expressed throughout flattened, neurite bearing cells. A point mutation in the SH2 domain of a2-chimaerin induced an al-chimaerin-like protein distribution and morphology. The effects of long term chimaerin overexpression on cell morphology and potential protein interactions were also investigated. Overexpression of ot2-chimaerin induced an enlarged, flattened morphology and neurite outgrowth in the presence of serum, whilst overexpression of al-chimaerin induced a rounded morphology with multiple peripheral actin microspikes and inhibited neurite outgrowth. p35, the neuronal cdk5 regulator and also an 130 kDa tyrosine phosphorylated protein were immunoprecipitated with chimaerin from these cell lines. Similarly an 180 kDa tyrosine phosphorylated protein was identified as a potential target of the a2-chimaerin SH2 domain. Investigation into the effects of chimaerin on activation of the transcription

factor NFk B demonstrated cell type specific differences in NFk B signalling pathways between HeLa and N1E 115 cells. These results suggest that functional differences in the chimaerin isoforms are specified by the divergent N terminal sequences.

Abstract 4 TABLE OF CONTENTS

Title page ...... 1 Dedication ...... 2 Acknowledgements ...... 3 Abstract...... 4 Table of Contents ...... 5 List of Figures ...... 14 Abbreviations ...... 17

CHAPTER ONE: Introduction ...... 23 1.1 Receptor tyrosine ...... 25 1.1.1 Substrates of receptor tyrosine kinases ...... 25 1.1.2 Multiple substrates of receptor tyrosine kinases ...... 26 1.2 Activation of downstream signalling pathways ...... 26 1.2.1 MAPK pathways...... 26 1.2.1 A ERK pathway...... 27 1.2.IB JNK and p38 pathways ...... 27 1.2.2 Phospholipid signalling pathways ...... 28 1.2.2A and ...... 28 1.2.2B Phosphatidylinositol-4-phosphate-5- ...... 28 1.2.2C Phosphatidylinositol-3-kinase ...... 29 1.3 The Cytoskeleton...... 29 1.3.1 The actin cytoskeleton ...... 29 1.3.1A Polymerisation of the actin cytoskeleton ...... 29 1.3. IB Actin polymerisation at the leading edge ...... 30 1.3.1C Regulation of actin depolymerisation ...... 31 1.3. ID Actin crosslinking ...... 31 1.3. IE Cross linkage of the actin cytoskeleton and membranes by ERM proteins ...... 32 1.3.2 The microtubule network ...... 33 1.3.2AMicrotubule structure ...... 34 1.3.2B Microtubule associated proteins ...... 34 1.3.3 The intermediate filament network ...... 35 1.3.3A Intermediate filament associated proteins ...... 36

Table of Contents 5 1.4 Protein domains ...... 36 1.4.1 SH2 domains ...... 37 1.4.1A Unusual SH2 domains ...... 38 1.4.IB Non phosphotyrosine dependent SH2 interactions ...... 38 1.4.2 Phosphotyrosine binding domains ...... 39 1.4.3 SH3 domains ...... 39 1.4.4 Pleckstrin homology domains ...... 40 1.4.5 Other protein domains ...... 42 1.5 GTPase superfamily...... 42 1.5.1 Heterotrimeric G proteins ...... 42 1.5.2 Ras subfamilies of low molecular weight GTPases ...... 44 1.5.2A Ras...... 45 1.5.3 The regulation of GTPase proteins ...... 45 1.5.3A GAPs for heterotrimeric G proteins ...... 45 1.5.3B GAPs for R as ...... 46 1.5.3C GAPs for Rho family proteins ...... 46 1.5.3D Multidomain nature of GAPs...... 48 1.5.3E RasGEFs ...... 49 1.5.3F GEFs for Rho family proteins ...... 50 1.5.3F1 Cdc42 specific GEFs ...... 50 1.5. 3F2 Rho specific GEFs...... 51 1.5.3F3 Rac specific GEFs...... 51 1.5.3F4 Multiple specificity GEFs...... 51 1.5.3G Guanine nucleotide dissociation inhibitors ...... 52 1.5.4 p21 Effector proteins ...... 53 1.5.4A Ras effectors ...... 53 1.5.4B Rho family effectors ...... 54 1.5.4B1 Non kinase targets of Cdc42 ...... 54 1.5.4B2 Non kinase targets of Rac ...... 55 1.5.4B3 Kinase targets of Cdc42 and Rac ...... 55 1.5.4B4 Non kinase targets of Rho ...... 56 1.5.4B5 Kinase targets of Rho ...... 57 1.5.4B6 Lipid kinase targets of Rho, Rac and Cdc42 ...... 58 1.5.4B7 GAPs as potential effectors ...... 59 1.5.5 Rho proteins and transcriptional activation ...... 60

Table of Contents 6 1.5.5A Rho proteins and JNK/p38 dependent transcription ...... 60 1.5.5B Rho proteins and SRE dependent transcription ...... 61

1.5.5C Rho proteins and NFk B dependent transcription ...... 61

1.5.5C1 Activation of NFk B ...... 61

1.5.5C2 Ras and Rho proteins in NFk B activation ...... 62 1.5.5C3 Generation of reactive oxygen species in phagocytic and non phagocytic cells ...... 64 1.5.5C4 Ras/Rac and reactive oxygen species production 65 1.5.5C5 The role of reactive oxygen species ...... 65 1.5.6 Rho proteins and cell morphology ...... 66 1.5.6A The role of Rho proteins in fibroblast cell morphology ...... 66 1.5.6B Neuronal cell morphology ...... 67 1.5.6B1 The role of Rho proteins in N1E 115 cell morphology ...... 67 1.5.6B2 The role of Rho proteins in PC 12 cell morphology ...... 69 1.5.6B3 The role of Rho proteins in neural systems ...... 70 1.5.6C Rho p21s and their morphological effects in other cell types...... 71 1.5.7 Rho proteins and cell cycle regulation...... 72 1.5.8 Rho proteins and transformation ...... 72 1.6 The chimaerin family of RacGAPs...... 73 1.6.1 a 1-Chimaerin...... 73 1.6.2 a2-Chimaerin...... 75 1.6.3 (3-Chimaerins...... 75 1.6.4 Cysteine rich domain of chimaerin ...... 76 1.6.5 Regulation of al- and a2-chimaerin GAP activity ...... 77 1.6.6 a2-chimaerin target proteins ...... 78

CHAPTER TWO: Materials and Methods ...... 80 2.1 Materials...... 81 2.2 Microbiological and nucleic acid methods ...... 81 2.2.1 Bacterial media and reagents ...... 81 2.2.2 Overnight cultures ...... 82 2.2.3 -70°C bacterial stocks ...... 82

Table of Contents 7 2.2.4 Production of competent XL 1-Blue E.Coli ...... 82 2.2.5 Transformation of competent E.Coli ...... 82 2.2.6 Wizard™ minipreps DNA purification system ...... 82 2.2.7 Mega plasmid DNA purification system ...... 83 2.2.8 Phenol-chloroform extraction of DNA...... 84 2.2.9 Ethanol precipitation ...... 84 2.2.10 DNA quantification ...... 85 2.2.11 Digestion of plasmid DNA with restriction endonucleases ...... 85 (a) Small scale plasmid restriction ...... 85 (b) Large scale plasmid restriction ...... 85 (c) Analytical digests ...... 86 2.2.12 Agarose gel electrophoresis ...... 86 2.2.13 DNA fragment purification from agarose gels ...... 86 2.2.14 Magic™ DNA clean-up system ...... 87 2.2.15 Blunt ending of DNA ...... 87 2.2.16 Analysis of DNA purification ...... 88 2.2.17 Radioactive labelling of DNA ...... 88 2.2.18 Blunt ended T4 DNA ligation ...... 88 2.2.19 Replica plating ...... 89 2.2.20 Filter hybridisation...... 89 2.2.21 Re-transformation of competent E.Coli ...... 90 2.2.22 In vitro transcription-translation assay ...... 90 2.3 Cloning of DNA constructs ...... 91 2.3.1 Eukaryotic expression vector details ...... 91 2.3.2 Generation of pXJ40-GFP vector ...... 97 2.3.3 Cloning of al-chimaerin into eukaryotic expression vectors ...... 97 2.3.4 Cloning of a2-chimaerin into eukaryotic expression vectors ...... 98 2.3.5 Cloning of TOAD-64 into eukaryotic expression vectors ...... 98 2.4 Cell culture...... 99 2.4.1 Frozen cell stocks ...... 99 2.4.2 Culture from frozen cell stocks ...... 99 2.4.3 Cell culture maintenance ...... 99 2.4.4 Treatment of slides and coverslips ...... 100 2.4.5 Treatment of permanently transfected N IE 115 cells for immunocytochemical analysis ...... 100

Table of Contents 8 2.5 Mammalian cell transfection ...... 101 2.5.1 Transient transfection of COS7 cells ...... 101 2.5.2 Transient transfection of N1E 115 cells ...... 101 (a) For immunoprecipitation analysis ...... 101

(b) For the NFkB reporter assay ...... 101 (c) For immunocytochemical analysis ...... 102

2.5.3 Transient transfection of HeLa cells for NFkB reporter assay ...... 102 2.5.4 Permanent transfection ofN IE 115 cells ...... 103 2.6 Analysis of cellular proteins ...... 104 2.6.1 Protein expression in transiently transfected COS7 cells ...... 104 2.6.2 Immunoprecipitation from transiently transfected COS7 and N1E 115 cells ...... 104 2.6.3 Protein expression in permanently transfected N1E 115 cells - CSK extraction ...... 105 2.6.4 Covalent coupling of HA antibody to protein A sepharose ...... 106 2.6.5 Immunoprecipitation from permanently transfected N1E 115 cell lines ...... 106 2.6.6 Determination of protein concentration ...... 107 2.6.7 SDS-PAGE gel electrophoresis ...... 108 2.6.8 Western blotting: protein transfer to nitrocellulose membranes ...... 108 2.6.9 Immunodetection of proteins ...... 109 2.6.10 Immunocytochemistry ...... 109 2.7 Reporter assays ...... 110

2.7.1 NFk B reporter assay -N1E 115 and HeLa cells ...... 110

2.7.2 Processing of NFkB assay data ...... 111 2.7.3 Luminescent p-galactosidase detection kit ...... 111

CHAPTER 3: Results I Distribution of HA- and GFP-tagged proteins in eukaryotic cells ...... 113 3.1 Protein expression from HA- and GFP-tagged DNA constructs in an in vitro Transcription-Translation Assay ...... 114 3.1.1 Protein expression from pXJ41-HA and pXJ40-HA DNA constructs ...... 114 3.1.2 Protein expression from pXJ40-GFP DNA constructs ...... 117 3.2 Expression of HA- and GFP-tagged a l- and ot2-chimaerin in COS7 cells ....117

Table of Contents 9 3.2.1 Expression ofHA-tagged chimaerin DNA constructs in COS7 cells...... 117 3.2.2 Expression of GFP-tagged chimaerin DNA constructs in COS7 cells...... 119 3.2.3 Effects of GFP tagging on the distribution of chimaerin proteins 121 3.3 Expression of a l- and a2-chimaerin in permanent N1E 115 cell lines ...... 122 3.3.1 Expression of a2-chimaerin in permanent N1E 115 cell lines ...... 122 3.3.2 Selection of permanent N1E 115 cell line expressing high levels of ot2-chimaerin ...... 122 3.3.3 Expression of al-chimaerin in permanent N1E 115 cell lines ...... 125 3.3.4 Levels of endogenous a2-chimaerin expression in permanent N1E 115 cell lines expressing al-chimaerin ...... 127 3.4 Expression of a2-chimaerin targets in COS7 cells ...... 127 3.4.1 Expression of HA- and GFP-tagged B13 and TOAD-64 DNA constructs in COS7 cells ...... 129 3.4.2 Detection of HA- and GFP-tagged proteins by the B 13 and TOAD-64 antibodies ...... 129 3.4.2A Detection of HA- and GFP-tagged B13 in COS7 cell lysates by the polyclonal B13 antibody ...... 131 3.4.2B Detection of HA- and GFP-tagged TOAD-64 in COS7 cell lysates by the TOAD-64 antibody ...... 131 3.4.3 Effects of GFP tagging on the distribution of B13 and TOAD-64 proteins ...... 131 3.5 Expression of Rho p21s ...... 134 3.5.1 Expression ofHA-tagged dominant positive and dominant negative Cdc42, Rac and Rho DNA constructs in COS7 .... cells...... 134 3.6 Summary...... 136

CHAPTER 4: Results II Investigation of al- and a2-chimaerin protein interactions ...... 138 4.1 Localisation of chimaerin proteins ...... 139 4.2 Investigation of a2-chimaerin protein interactions ...... 139 4.2.1 Immunoprecipitation of GFP-tagged a2-chimaerin in COS7 and N1E 115 cells...... 141

Table of Contents 10 4.2.2 Investigation of ot2-chimaerin and TOAD-64 protein interactions in COS7 cells ...... 141 4.2.3 Investigation of a2-chimaerin and TOAD-64 protein interactions inNIE 115 cells...... 144 4.2.4 Investigation of a2-chimaerin and B13 protein interactions in COS7 and N1E 115 cells ...... 144 4.2.5 Immunoprecipitation of al - and a2-chimaerin from permanent N1E 115 cell lines ...... 145 4.3 Summary...... 148

CHAPTER 5: Results 111

Investigation into the role of Rho p21s and the q-chimaerins in NFkB signalling in HeLa and N1E 115 cells ...... 149

5.1 NFkB activation ...... 150

5.2 NFkB reporter assay...... 151

5.2.1 NFk B activation in HeLa cells ...... 151

5.2.2 NFk B activation inNIE 115 cells...... 155

5.2.3 Specificity of NFk B activation ...... 164 5.3 Summary...... 164

CHAPTER 6: Results IV The morphology of N1E 115 cells expressing al-chimaerin. a2-chimaerin or potential a2-chimaerin targets ...... 166 6.1 Rho proteins and neuronal cell morphology ...... 166 6.2 Morphology of untransfected N1E 115 cells ...... 166 6.3 Morphology of transiently transfected N1E 115 cells expressing a l- or a2-chimaerin ...... 171 6.3.1 The effects of a2-chimaerin SH2 domain mutants on N1E 115 cell morphology ...... 171 6.4 Morphology of permanently transfected N1E 115 cell lines overexpressing a l- or a2-chimaerin ...... 172 6.4.1 Quantitation of cell morphology ...... 172 6.5 The effects of potential a2-chimaerin targets on N1E 115 cell morphology...... 180

Table of Contents 11 6.5.1 The effects of TOAD-64 and B13 onNIE 115 cell morphology ...... 180 6.5.2 p35...... 183 6.5.3 Expression of p35 in N1E 115 cells ...... 183 6.5.4 Colocalisation of p35 and actin in the a2-10 cell line ...... 185 6.6 Summary...... 188

CHAPTER SEVEN; Discussion ...... 190 7.1 Rho p21s and the chimaerin RacGAPs...... 191 7.2 Distribution of al- and a2-chimaerin in eukaryotic cells ...... 193 7.3 The effects of chimaerin overexpression on N1E 115 cell morphology ...... 194 7.4 SH2 domain mutants of a2-chimaerin ...... 196 7.4.1 Distribution of a2-chimaerin SH2 domain mutants in eukaryotic cells...... 198 7.5 Potential regulation of a2-chimaerin activity by intra/inter molecular interactions ...... 199 7.6 GAP activity of the chimaerins...... 199 7.6.1 Effects of the a2-chimaerin SH2 domain on GAP activity ...... 200 7.7 Effects of the common regions of the chimaerins on protein localisation ...... 200 7.8 Chimaerin target proteins ...... 201 7.8.1 Distribution ofB13 and TOAD-64 in eukaryotic cells ...... 202 7.8.2 Localisation ofB13, TOAD-64 and p35 in eukaryotic cells ...... 203 7.8.3 Interaction of protein targets with a2-chimaerin ...... 203 7.8.3 A a2-Chimaerin interaction with TOAD-64 ...... 203 7.8.3B a2-Chimaerin interaction with B 13 ...... 204 7.8.3C Chimaerin interactions with p35 ...... 204 7.8.3D a2-Chimaerin interactions with tyrosine phosphorylated proteins ...... 205

7.9 The chimaerins and NFk B signalling ...... 206

7.9.1 NFk B signalling in HeLa and N1E 115 cells ...... 207

Conclusions ...... 209

References ...... 211

Table of Contents 12 Appendices 250

Appendix 1: The rat al-chimaerin DNA sequence used in the HA and GFP -tagged DNA constructs ...... 251 Appendix 2: The human a2-chimaerin DNA sequence used in the HA and GFP-tagged DNA constructs ...... 252 Appendix 3: The rat TOAD-64 DNA sequence used in the HA and GFP -tagged DNA constructs ...... 253 Appendix 4: The structure of the pGL2 basic vector used to generate the

luciferase coupled NFk B reporter vectors ...... 254 Appendix 5: NFkB reporter assay data used to generate figure 5.1 ...... 255 Appendix 6: NFkB reporter assay data used to generate figure 5.3 ...... 256 Appendix 7: NFkB reporter assay data used to generate figure 5.4 ...... 257 Appendix 8: NFkB reporter assay data used to generate figure 5.6 ...... 258 Appendix 9: NFkB reporter assay data used to generate figure 5.8 ...... 259 Appendix 10: NFkB reporter assay data used to generate figure 5.10 ...... 260

Table of Contents 13 LIST OF FIGURES

Chapter One Table 1.1: Protein signalling domains ...... 41 Figure 1.2: The GTPase cycle...... 43

Figure 1.3a: Activation of NFk B signalling pathways ...... 63 Figure 1.3b: The involvement of Rho family proteins in the activation of

NFk B signalling pathways ...... 63 Figure 1.4: structure of the chimaerin RacGAPs ...... 74

Chapter Two Figure 2.1a: Structure of the pXJ40-HA vector ...... 92 Figure 2.1b: DNA sequence of the multiple cloning site of the pXJ40-HA vector...... 92 Figure 2.2a: Structure of the pXJ40-GFP vector ...... 93 Figure 2.2b: DNA sequence of the multiple cloning site of the pXJ40-GFP vector...... 93 Figure 2.2c: DNA sequence of the green fluorescent protein (GFP) epitope ...... 94 Figure 2.3: Structure of the pXJ41-HA vector ...... 95

Chapter Three Figure 3.1: Protein expression from HA-tagged DNA constructs in an in vitro transcription-translation assay ...... 115 Figure 3.2: Protein expression from GFP-tagged DNA constructs in an in vitro transcription-translation assay ...... 116 Figure 3.3: Expression ofHA-tagged a2-chimaerin proteins in COS7 cells 118 Figure 3.4: Expression of GFP-tagged a l - or a2-chimaerin proteins in COS7 cells...... 120 Figure 3.5: Distribution of a2-chimaerin protein in permanently transfected N1E 115 cells...... 123 Figure 3.6: Levels of a2-chimaerin expression in permanently transfected N1E 115 cells...... 124 Figure 3.7: Distribution of al-chimaerin protein in permanently transfected N1E 115 cells...... 126 Figure 3.8: Endogenous a2-chimaerin expression in permanently

List of figures 14 transfected NIE 115 cells overexpressing al-chimaerin ...... 128 Figure 3.9: Expression of GFP-tagged TOAD-64 or B13 proteins in COS7 cells...... 130 Figure 3.10: Detection of HA- and GFP-tagged B13 proteins by a polyclonal B13 antibody ...... 132 Figure 3.11: Detection of HA- and GFP-tagged TOAD-64 proteins by a polyclonal TOAD-64 antibody ...... 133 Figure 3.12: Expression ofHA-tagged dominant positive and dominant negative Cdc42, Rac and Rho in COS7 cells ...... 135

Chapter Four Figure 4.1: Co-expression ofHA-tagged a2-chimaerin and V12Rac, V12Cdc42 or V14Rho proteins in COS7 cells ...... 140 Figure 4.2: Immunoprecipitation of GFP-tagged a2-chimaerin from COS7 and N1E 115 cells ...... 142 Figure 4.3: Co-immunoprecipitation of TOAD-64 and a2-chimaerin in COS7 cells...... 143 Figure 4.4: Co-immunoprecipitation of tyrosine phosphorylated proteins with a l - and a2-chimaerin in permanent N1E 115 cell lines ...... 146 Figure 4.5: Co-immunoprecipitation of p35 with a l- and a2-chimaerin in permanent N1E 115 cell lines ...... 147

Chapter Five Figure 5.1: Summary table of NFkB reporter assay results in HeLa cells (0.5pg DNA level) ...... 152 Figure 5.2: Graphical representation of NFkB activity in HeLa cells at the 0.5|j.g DNA level ...... 153 Figure 5.3: NFkB activation in HeLa cells stimulated with IL-1(3 ...... 154 Figure 5.4: Summary table of NFkB reporter assay results in N1E 115 cells (0.5pg DNA level) ...... 156 Figure 5.5: Graphical representation of NFkB activity in N1E 115 cells at the 0.5(j,g DNA level...... 157 Figure 5.6: Summary table of NFkB reporter assay results in N1E 115 cells (1.5pg DNA level) ...... 159

List of figures 15 Figure 5.7: Graphical representation of NFkB activity in N1E 115 cells at the 1.5 tig DNA level...... 160 Figure 5.8: Summary table of dominant positive and dominant negative Rho

family protein induced NFk B activity in N1E 115 cells (0.5pg level) ...... 161 Figure 5.9: Graphical representation of Rho family protein induced NFkB activity in N1E 115 cells at the 0.5pg DNA level...... 162 Figure 5.10: Summary table of mutated NFkB(M) reporter assay results inNIE 115 cells (1.5pg DNA level) ...... 163

Chapter Six Figure 6.1: The morphology of untransfected N1E 115 cells ...... 168 Figure 6.2: Expression of GFP-tagged chimaerin proteins in transiently transfected N1E 115 cells 169, 170 Figure 6.3: Examples of the morphology of permanently transfected N1E 115 cell lines...... 173 Table 6.4: Quantification of the morphology of permanently transfected N1E 115 cell lines...... 174 Figure 6.5: Graphical representation of permanent N1E 115 cell line morphology in the presence of serum ...... 175 Figure 6.6: Graphical representation of permanent N1E 115 cell line morphology in the absence of serum ...... 176 Figure 6.7: Examples of the morphology of permanently transfected N1E 115 cell lines overexpressing a l- or a2-chimaerin ...... 178 Figure 6.8: Expression of a2-chimaerin protein in the permanently transfected a2-10 cell line ...... 179 Figure 6.9: Expression of GFP-tagged TOAD-64 and B 13 proteins in transiently transfected N1E 115 cells 181, 182 Figure 6.10: Expression of p35 in untransfected N1E 115 cells and the permanently transfected a2-10 cell line...... 184 Figure 6.11: Colocalisation of p35 and F-actin in the permanently transfected a2-10 cell line 186, 187

List of figures 16 Abbreviations

PARK P adrenergic receptor kinase P-gal P-galactosidase Ack activated Cdc42-binding kinase ALL acute lymphocytic leukaemia ALS sporadic amyotrophic lateral sclerosis Amp ampicillin APS (ammonium persulphate) ARF ADP ribosylation factor Arp actin related protein ATP adenosine triphosphate bFGF basic fibroblast growth factor BSA bovine serum albumin Btk Bruton's tyrosine kinase C terminal carboxy terminal cDNA complementary DNA CEP4 Cdc42 interacting protein CML chronic myelogenous leukaemia CRD cysteine rich domain CRIB Cdc42/Rac interactive binding CRMPs collapsin response mediator proteins CTP cytidine triphosphate DABCO l,4-diazobicyclo-[2.2.2]-octane DAG diacylglycerol dCTP deoxycytidine triphosphate ddH20 double distilled water ddNTP dideoxynucleoside 5 ’-triphosphate DH Dbl homology domain DMEM Dulbecco’s modified eagle medium DMP dimethylpimelimidate DMSO dimethylsulphoxide DNA deoxyribonucleic acid DPI diphenyleneiodonium DRG dorsal root ganglion

Abbreviations DTT dithiothreitol EBS epidermolysis bullosa simplex ECL enhanced chemiluminescence EDTA ethylenediamine tetraacetic acid EGF epidermal growth factor EGFR epidermal growth factor receptor EGTA ethylene glycol-0, O’-bis(2-aminoethyl)-N,N,N’,N’-tetraacetic acid EH epidermolytic hyperkeratosis EPPK epidermolytic palmoplantar peratoderma ERK extracellular signal regulated kinase ERM ezrin, radixin and moesin family F-actin filamentous actin FAK kinase FCS foetal calf serum FGD1 product of the faciogenital dysplasia/Aarskog-Scott syndrome locus FGF fibroblast growth factor FGFR fibroblast growth factor receptor FH formin homology domain FITC fluorescein isothiocyanate Frabin FGD1 related F-actin binding protein G418 sulphate geneticin® G-actin globular monomeric actin GAP GTPase activating protein GAPDH glyceraldehyde-3 -phosphate dehydrogenase GCK germinal centre kinase GDI guanine nucleotide dissociation inhibitor GDP guanosine diphosphate GDS guanine nucleotide dissociation stimulator GEF guanine nucleotide exchange factor G-FAP glial acidic fibrillary protein GFP green fluorescent protein Graf GTPase regulator associated with FAK GTP

Abbreviations GTPase guanosine HA haemagglutinin HEPES N-2-hydroxyethylpiperazine-N’-2-ethanesulphonic acid HSN hermaphrodite specific

Ik B inhibitor of NFk B IF proteins intermediate filament proteins IFAP intermediate filament associated protein

IKK Ik B kinase IP inositol-1 -phosphate IP2 inositol-1,4-bisphosphate IPs inositol-1,4,5-trisphosphate IRS-1 insulin receptor substrate-1 JNK c-Jun amino terminal kinase kDa kilodalton Klenow large fragment of E.Coli DNA polymerase I LB Luria Bertani medium LPA lysophosphatidic acid LPS lipopolysaccharide MAPK mitogen activated MAPKK mitogen activated protein kinase kinase MAPKKK mitogen activated protein kinase kinase kinase MAPs microtubule associated proteins MBS binding subunit of myosin light chain phosphatase MCS multiple cloning site MD-EBS muscular dystrophy and epidermolysis bullosa simplex MEK MAPK/ERK kinase MEKK MAPK/ERK kinase kinase MEKK1 MAPK/ERK kinase kinase-1 MEM minimal essential medium MLC myosin light chain MLK mixed lineage kinase MOPS morpholinopropanesulphonic acid MRCK myotonic dystrophy kinase-related Cdc42-binding kinase mRNA messenger RNA MTOC microtubule organising centre

Abbreviations N terminal amino terminal NF neurofilament protein

NFk B nuclear factor kappa B NF1 neurofibromin gene NGF nerve growth factor NHE1 Na/H exchanger protein

NIK NFk B inducible kinase pl40Sra-l 140kD specifically Racl associated protein PA phosphatidic acid PAE cells porcine aortic endothelial cells PAGE polyacrylamide gel electrophoresis PAKs p21 activated kinases PARG1 PTPL1-associated RhoGAP PBD p21 binding domain pBS bluescript vector PBS phosphate buffered saline PC phosphatidylcholine PC12 rat pheochromocytoma cell line PCM pericentriolar material PCR polymerase chain reaction PDGF platelet derived growth factor PDGFR platelet derived growth factor receptor PE phosphatidylethanolamine PH pleckstrin homology domain PI phosphatidylinositol PI-3,4,5-P3 phosphatidyl inositol-3,4,5-trisphosphate PI-3,4-P2 phosphatidylinositol-4,5-bisphosphate PI3K phosphatidylinositol-3-kinase PI-3-P phosphatidylinositol-3-phosphate PI-4,5-P2 phosphatidylinositol-4,5-bisphosphate PIP5K Phosphatidylinositol-4-phosphate 5 kinase PIX Pak interacting exchange factor PKB/Akt protein kinase B PKC protein kinase C PKN protein kinase N

Abbreviations PLC phospholipase C PLD phospholipase D PMSF phenylmethylsulphonyl fluoride POF premature ovarian failure PORI partner of Racl POSH plenty of SH3 domains PS phosphatidyl serine PTB phosphotyrosine binding domain PTP protein tyrosine phosphatase RasGAP Ras GTPase activating protein REF-52 cells rat embryonic fibroblast cells RGS proteins regulators of coupled signalling RhoGDI Rho guanine nucleotide dissociation inhibitor RNA ribonucleic acid RNase A ribonuclease A ROK RhoA binding kinase ROS reactive oxygen species RT room temperature SAPK stress activated protein kinase SCRs structurally conserved regions SDS sodium dodecyl sulphate SH2 Src homology 2 domain SH3 Src homology 3 domain Sos SRE c-fos serum response element SRF serum response factor SSC standard saline citrate TAK TGFP activated kinase TCP ternary complex protein TE Tris/EDTA TEMED N,N,N’,N’-tetramethylethylene diamine TOAD-64 turned on after division TRAF TNF receptor-associated factor Tris 2-amino-2(hydoxymethyl)-l,3-propandiol Triton X-100 octylphenoxypolyethoxyethanol

Abbreviations Tween-20 polyoxyethylenesorbitan monolaurate Ulips unc-33 like phosphoproteins uv ultraviolet v/v volume by volume VEGF vascular endothelial growth factor VSMC vascular smooth muscle cell w/v weight by volume WASP Wiskott Aldrich syndrome protein WIP WASP interacting protein

Abbreviations CHAPTER ONE: Introduction

Introduction The Rho p21s are part of the of small GTPases. These proteins function as molecular switches, cycling between an inactive GDP bound form and an active GTP bound form. This cycling is regulated by the intrinsic GTPase activity of the proteins and also three classes of regulatory proteins; the GAPs, GEFs and GDIs. The GTPase activating proteins (GAPs) accelerate the intrinsic GTPase activity of the p21 resulting in its downregulation. The guanine nucleotide exchange factors (GEFs) promote the exchange of GDP for GTP thus activating the protein. The guanine nucleotide dissociation inhibitors (GDIs) inhibit the dissociation of GDP, sequestering the protein in an inactive complex. The Rho family proteins; Rho, Rac and Cdc42, are activated downstream of coupled receptors and receptor tyrosine kinases and mediate a wide variety of cellular effects including transcriptional activation, reorganisation of the actin cytoskeleton, the regulation of cell growth, motility and cell cycle progression. These effects are propagated by a variety of effector proteins which include protein kinases, lipid kinases and phosphatases. The effectors interact with the active GTP bound form of the Rho p21 and mediate their downstream effects, often via the formation of multi protein signalling complexes which are formed via the interaction of conserved modular protein domains with specific peptide target sequences. There are a large number of proteins with a conserved RhoGAP domain, many of which are multidomain proteins. Chimaerin is a neuronally expressed RacGAP with tissue specific and developmentally regulated expression, a l- and a2-Chimaerin are alternate splice products with divergent N terminal sequences. Their common region contains a cysteine rich domain (CRD) and a GAP domain. The N terminal region of al-chimaerin contains a potential amphipathic helix region whilst a2-chimaerin contains an SIC domain. The specific expression patterns of these two isoforms suggests the alternate N terminal regions of these proteins may have discrete functional or regulatory roles. The presence of an SH2 domain in a2-chimaerin suggests a role in tyrosine kinase signalling pathways. Interactions mediated by this domain are likely to have a considerable effect on protein activity and localisation, and interaction with different target proteins may result in activation of a variety of downstream signalling pathways and cellular effects which could be perturbed by mutation of the SIC domain. As one of the best characterised effects of the Rho p21s is their involvement in reorganisation of the actin cytoskeleton, the effects of the neural specific a-chimaerin RacGAPs on neuronal cell morphology and the contribution of the a2-chimaerin SH2 domain were investigated. Introduction 24 1.1 Receptor tyrosine kinases The different families of receptor tyrosine kinases, including the platelet derived growth factor receptor (PDGFR), epidermal growth factor receptor (EGFR) and fibroblast growth factor receptor (FGFR) share features in common. The proteins contain an extracellular ligand binding domain, a transmembrane region and an intracellular kinase domain. The intracellular domain of these proteins contain regulatory sequences which are subject to autophosphorylation and act as binding sites for target proteins. Binding of an extracellular ligand usually induces receptor dimerisation and autophosphorylation of cytoplasmic tyrosine residues, although this is not true of the insulin receptor which pre-exists as an a(3 dimer. Some ligands, such as PDGF are dimeric and simultaneously interact with two receptors, whilst other ligands such as EGF are monomeric and stabilise a dimeric form of their receptors. Multiple receptor isotypes and ligands, combined with homo- and hetero- dimerisation produces a wide variety of different receptor subtypes with distinct functions and ligand binding specificities. Tissue specific expression of various receptor isotypes adds further complexity and selectivity to receptor tyrosine kinases signalling pathways. Receptor autophosphorylation occurs at conserved residues within the kinase domain of the receptor, which further upregulates kinase activity, or at residues outside the kinase domain which act as binding sites for effector molecules containing phosphotyrosine binding motifs such as SH2 or PTB domains.

1.1.1 Substrates of receptor tyrosine kinases Proteins containing Src homology 2 (SH2) or phosphotyrosine binding (PTB) domains interact with specific phosphotyrosine containing sequences of activated receptor tyrosine kinases. The specificity of these interactions is determined by the amino acid sequence adjacent to the tyrosine phosphorylated residue (reviewed in Pawson et al., 1995). These protein domains are discussed in detail later. SH2 and PTB domains are found in kinases such as Src and PI3K and also in non catalytic adaptor or docking proteins like Grb2 or She which mediate specific protein interactions resulting in the formation of multiprotein signalling complexes. Grb2 provides a classical example of the involvement of SH2 domain containing proteins in signalling downstream from activated receptor tyrosine kinases. Grb2 exists in an inactive cytosolic complex with the RasGEF Sos (Bowtell et al., 1992; Chardin et al., 1993), via interaction of its SH3 domain with a proline-rich sequence in the C terminal of Sos

Introduction 25 (Rozakis-Adcock et al, 1993). The SH2 domain of Grb2 directly interacts with Tyrl068 of activated EGF receptor, resulting in translocation of both Grb2 and Sos to the membrane where Sos activates p21 Ras, inducing activation of downstream signalling pathways. This mechanism of Ras activation is involved in vulval development in C.Elegans and differentiation of photoreceptor cells in Drosophila (reviewed in Dickson andHafen, 1994).

1.1.2 Multiple substrates of receptor tyrosine kinases In an activated receptor tyrosine kinase, multiple potential substrate binding sites often exist. In the PDGFP receptor, eleven tyrosine phosphorylated sites have been identified and at least ten different proteins have been shown to interact, including the kinases Src and PI3K, adaptor proteins Grb2, She and Nek and the tyrosine phosphatase SHP-2, leading to activation of multiple signalling pathways involved in cell growth and motility (Heldin, 1997). The diverse nature of these interacting proteins implies that both inhibitory and stimulatory pathways may be activated at the same time. Effectors may also compete for the same of an activated receptor. For example, upon stimulation with vascular endothelial growth factor (VEGF), Nek, SHP-2 and PI3K all bind the phosphorylated Tyrl213 residue of the activated Fit-1 receptor (Igarashi et al., 1998). Different signalling pathways may also use common components. For example the adaptor protein Crk can bind both the activated PDGF receptor and also the insulin receptor substrate IRS-1 (Sorokin et al., 1998). Thus the net effects induced by receptor tyrosine kinase activation depends on the relative activities of potentially competing signalling pathways within the cell.

1.2 Activation of downstream signalling pathways

1.2.1 MAPK pathways Mitogen activated protein kinase (MAPK) pathways link growth factor receptor activation to downstream signalling and transcriptional activation (reviewed in Treisman, 1996; Garrington and Johnson, 1999). These pathways are best characterised in yeast and are involved in regulating osmotic stability, cell integrity and mating response (Herskowitz, 1995) but are also conserved between yeast and mammals. The common feature of MAPK pathways is a three kinase cascade. A serine/threonine MAP kinase kinase kinase (MAPKKK) phosphorylates a threonine/tyrosine MAP kinase kinase (MAPKK) which in turn phosphorylates the effector of the pathway, a

Introduction 26 serine/threonine MAP kinase (MAPK). The MAPK proteins have different characteristic phosphorylation sites for different MAPKKs; ERK has a TEY sequence, JNK has a TPY sequence and p38 has a TGY sequence, where T and Y are the phosphorylated residues.

1.2.1 A ERK pathway Activation of the ERK pathway is involved in many processes including cellular transformation (Mansour et al., 1994), protection from (Xia et al., 1995; Gardner and Johnson, 1996) and differentiation including NGF-induced differentiation in PC 12 cells (Cowley et al., 1994), vulval development in C.Elegans and differentiation of photoreceptor cells in Drosophila (reviewed in Dickson and Hafen, 1994). The three kinases of this MAPK pathway are Raf (a MEKK or MAPK/ERK kinase kinase) which is activated by Ras, MEK1/MEK2 (MAPK/ERK kinase) and ERK1/ERK2 (42 and 44kDa extracellular signal regulated kinases). The serine /threonine kinase Mos can also activate MEK1/MEK2 in germline cells (Seger and Krebs, 1995). Downstream targets of ERKs include transcription factors Ets-1 and Ets-2 (Yang et al., 1996) and also Elk-1 and SAP-1 (Price et al., 1995), whose phosphorylation induces gene expression. Elk-1 and SAP-1 are ternary complex proteins (TCPs) that form a complex with serum response factor (SRF) which binds the c-fos serum response element (SRE), a regulatory sequence present in the promoter region of many growth factor regulated genes, activating transcription (reviewed in Treisman, 1994).

1.2.1B JNK and p38 pathways The c-Jun amino terminal kinases (JNKs) (Derijard et al., 1994) and stress activated protein kinases (SAPKs) (Kyriakis et al., 1994) are activated by cellular stresses, ultraviolet irradiation and inflammatory cytokines such as TNFa and IL-1. A large number of kinases including MEKKs (MAPK/ERK kinase kinases), GCKs (germinal centre kinases), MLKs (mixed lineage kinases), PAKs (p21 activated kinases) and TAKs (TGFp activated kinases) activate JNK in various cell types, some of which phosphorylate MKK4 a specific activator of JNK (reviewed in Fanger et al., 1997a). Upon activation, JNK translocates to the nucleus, phosphorylates the N terminal of c- jun and activates transcription. A third MAPK pathway, the p38 kinase pathway, is activated by hyperosmotic Introduction 27 shock and inflammatory cytokines (Raingeaud et al., 1995). p38 is the mammalian homologue of the S.Cerevisiae HOG1 kinase which is involved in response to osmotic shock (Schuller et al., 1994). Unlike the ERK pathway, JNK and p38 are not activated by Ras, however they are both activated by Rho family proteins (Coso et al., 1995; Minden et al., 1995), which is discussed in detail later. Both JNK and p38 stimulate transcription via ATF-2 (Raingeaud et al., 1995) and also like ERK, via Elk-1 and the SRE (Whitmarsh et al., 1995). Thus different signalling pathways converge at common promoter elements.

1.2.2 Phospholipid signalling pathways Certain phospholipids can act as second messengers. They are generated by phosphatidylinositol kinases such as PI3K and PIP5K and the phospholipases C and D.

1.2.2A Phospholipase C and phospholipase D Phospholipase C (PLC) catalyses the hydrolysis of phosphatidylinositides resulting in the production of diacylglycerol (DAG), the physiological activator of PKC and also inositol-1,4,5-trisphosphate (IP3), which activates Ca2+-dependent systems via release of Ca2+ from intracellular stores (Berridge, 1993). Three PLC isoforms exist; p, y and 5 which are activated by different mechanisms. PLC-P is activated by heterotrimeric G protein coupled receptors whilst PLC-y is activated by receptor tyrosine kinases. The SH2 domains of PLC-y associate with activated receptor tyrosine kinases such as the PDGF and EGF receptors, which phosphorylate and activate PLC-y (reviewed in Lee and Rhee, 1995). Phospholipase D (PLD) hydrolyses phosphatidylcholine (PC) to phosphatidic acid (PA) and choline. PLD is activated in response to a wide variety of hormones, growth factors and other extracellular signals. Phosphatidic acid acts as a lipid second messenger, although its intracellular targets have not been clearly identified. PLD1 and

PLD2 isoforms are both dependent on PI-4, 5 -P2 for activity. The PLD1 isoform is also stimulated by PKC and the small GTPases ARF and RhoA (reviewed in Houle and Bourgoin, 1999).

I.2.2B Phosphatidvlinositol-4-phosphate-5-kinase

PIP5K generates PI-4, 5 -P2, an important second messenger that binds and regulates a number of actin binding proteins (Toker, 1998). Both Rho and Rac stimulate PIP5K activity in various cell types, which is discussed in detail later. PI-4,5-P2 also Introduction 28 binds pleckstrin homology (PH) domains and on binding the PH domain of the RasGEF Sos inhibited its activation of Ras, suggesting a possible negative regulatory role of Rho/Rac activation on Ras signalling (Jefferson et al., 1998).

1.2.2C Phosphatidvlinositol-3-kinase PI3K phosphorylates the D3 position of phosphoinositide lipids to produce PI-3-

P, PI-3,4 -P2 and PI-3,4, 5 -P3 which regulate various cellular processes (reviewed in Toker and Cantley, 1997). PI3K is composed of the regulatory p85 and catalytic pi 10 subunits (Carpenter et al., 1990). PI3K is activated by the py subunit of heterotrimeric G proteins (Kurosu et al., 1997) and is also activated downstream of receptor tyrosine kinases such as the PDGF receptor, where interaction of the SH2 domains of the p85 subunit with the activated receptor induces PI3K activation (Pawson, 1995). PI3K is also activated by interaction of the pi 10 catalytic subunit with GTP bound Ras, thus PI3K acts as a Ras effector (Rodriguez-Viciana et al., 1994). PI3K also acts as an effector of Rho, Rac and Cdc42 which is discussed in detail later. PI3K mediates many different cellular effects including activation of the PKB/Akt anti-apoptotic pathway (Marte and Downward, 1997; Downward, 1998), induction of neurite outgrowth in PC12 cells (Kobayashi et al., 1997; Kita et al., 1998) and regulation of the actin cytoskeleton and cell motility (Van Weering et al., 1998).

1.3 The Cvtoskeleton The cellular cytoskeleton consists of three networks - the actin cytoskeleton, microtubules and intermediate filaments. Together these networks enable the cell to change shape and move in a directed manner in response to intracellular and extracellular signals via rapid polymerisation and depolymerisation of their constituent proteins. These protein networks and the associated proteins which crosslink and organise these networks are all potential targets for cell signalling molecules.

1.3.1 The actin cvtoskeleton

1.3.1A Polymerisation of the actin cvtoskeleton Actin exists in a globular monomeric (G-actin) and filamentous (F-actin) form. Actin filaments have a fast growing plus or barbed end and a slow growing minus or pointed end. Monomers undergo a conformational change on addition to a filament and

Introduction 29 the polymerisation process requires ATP. In a resting cell, a dynamic equilibrium exists where no net filament growth occurs but the rate of monomer addition at the barbed end equals the rate of dissociation from the pointed end, a process known as treadmilling (Wegner, 1976). The rate of actin polymerisation or depolymerisation is regulated by several types of proteins including actin monomer sequestering proteins such as profilin (Sun et al., 1995) and proteins which sever actin filaments such as gelsolin and cofilin, which also cap barbed ends and cause depolymerisation respectively (Yin, 1987; Maciver, 1998). The first step in actin polymerisation is nucleation. This process has been studied in vivo using the invasion of pathogenic bacteria Listeria monocytogenes and Shigella flexneri as model systems, which upon invasion induce the host cell to initiate actin polymerisation via the bacterial cell surface proteins ActA and IcsA respectively, which propels the bacterium through the host cytoplasm (reviewed in Higley and Way, 1997). Using the Listeria system it was found that that human Arp2/3, a seven protein complex containing actin related proteins Arp2 and Arp3 first identified in Acanthamoeba (Machesky et al., 1994, reviewed in Machesky and Gould, 1999), was required in combination with bacterial ActA to induce actin polymerisation (Welch et al., 1997). It has recently been shown in vitro that the Arp2/3 complex nucleates actin polymerisation from the barbed end of actin dimers and also promotes both branching and crosslinking of actin filaments (Mullins et al., 1998; Welch et al., 1998).

1.3.1B Actin polymerisation at the leading edge A 'dendritic nucleation' model has been proposed from in vitro data for actin polymerisation at the leading edge of a cell (Mullins et al., 1998). In this model, the Arp2/3 complex is activated and targetted to the site of lamellipodia formation where it caps the pointed end of an actin dimer which then rapidly elongates at the barbed end. In addition to new filament generation, growth can be nucleated from the sides of existing filaments producing a branched network which drives protrusion at the leading edge (Mullins et al., 1998). The older ends of the actin filament nearer the centre of the cell become susceptible to severing by cofilin as ATP is hydrolysed and are disassembled (Maciver, 1998). Profilin binds the released actin monomers which are then recharged with ATP and incorporated into new actin filaments. This 'dendritic nucleation' model is supported by data in fibroblasts where actin branches in lamellipodia were observed (Svitkina et al., 1997). However the discovery that the Arp2/3 complex caps the pointed ends of actin

Introduction 30 filaments (Mullins et al., 1998) means no dissociation of actin monomers is possible from this end without first severing the actin filaments, which casts doubt on the previously suggested treadmilling concept. Since activation of the Arp2/3 complex is required to initiate this capping process it may be that in an unstimulated cell the Arp2/3 complex is inactive and treadmilling does occur but further work is required to resolve this question.

1.3.1C Regulation of actin depolvmerisation The severing and depolymerisation of actin filaments by cofilin are essential for processes which require rapid turnover of actin (Theriot, 1997) such as the extension and retraction of and lamellipodia, cell movement, (Abe et al., 1996) and endocytosis (Lappalainen and Drubin, 1997). Cofilin activity is regulated by LIM kinase which phosphorylates cofilin on serine3, inhibiting its activity, whilst LIM kinase itself was recently shown to be activated by both Rac (Arber et al., 1998; Yang et al., 1998) and the Rho effector ROK (Maekawa et al., 1999) in different cell types. In HeLa cells, dominant negative LIM kinase inhibited the extension of Rac-dependent lamellipodia suggesting that LIMK acts downstream from Rac and also implying that inactivation of cofilin was required for lamellipodia protrusion (Yang et al., 1998). This was supported by the finding that cofilin phosphorylation was increased by activated Rac and decreased by dominant negative Rac expression (Arber et al., 1998; Yang et al., 1998). In a separate study, ROK phosphorylated and activated LIMK resulting in cofilin phosphorylation both in vitro and in COS7 cells, whilst in N1E 115 cells undergoing LPA-induced neurite retraction, cofilin phosphorylation was sensitive to the ROK inhibitor Y-27632 (Maekawa et al., 1999). LIMK also induced stress fibre formation in HeLa cells similar to V14Rho or ROK, which were disassembled in the presence of Y- 27632 (Maekawa et al., 1999). Thus both Rac and Rho/ROK pathways use inactivation of cofilin to stabilise two different actin containing structures.

1.3.1D Actin crosslinking Several actin containing structures are induced by the Rho p21s. Filopodia are thin finger like projections from the cell surface which have a sensory role and are induced by Cdc42 (Kozma et al., 1995; Nobes and Hall, 1995). Lamellipodia are thin sheet like processes which extend from the leading edge of a cell and become ruffles when they are swept back over the top of the cell and are induced by Rac (Ridley et al., 1992). Stress fibres are bundles of actin filaments which traverse the cell and are

Introduction 31 attached to the extracellular matrix at focal adhesions at the cell periphery and are induced by Rho (Ridley and Hall, 1992). Actin binding proteins such as a-actinin, fimbrin, filamin and spectrin crosslink actin filaments into the three dimensional networks required to form filopodia, lamellipodia and stress fibres, a-actinin and fimbrin bundle actin filaments together, filamin crosslinks them into a network and spectrin attaches them to the membrane. These proteins contain one or two 27kDa actin binding domains (Matsudaira, 1994) and those with one, like a-actinin, dimerise to form a functional crosslinking unit. The type of crosslinking produced depends on the proximity of the two actin binding sites; those which are close together result in rigid, tightly packed structures whilst those that are further apart result in a looser crosslinked structure. Some actin crosslinking proteins also bind other proteins enabling the formation of more complex structures such as focal adhesions which are sites of attachment between cells and the extracellular matrix that are induced in response to integrin clustering (Yamada and Miyamoto, 1995; Yamada and Geiger, 1997) and activation of Rho p21s (Hall, 1994). The interaction between actin binding or crosslinking proteins and actin is also regulated by PI-4,5-P2 which binds and regulates a number of these proteins (Toker, 1998), for example PI-4,5-P2 promotes vinculin binding to talin and actin (Gilmore and Burridge, 1996).

1.3.1E Cross linkage of the actin cvtoskeleton and membranes bv ERM proteins Members of the ezrin, radixin and moesin (ERM) act as crosslinking proteins between actin filaments and the plasma membrane (reviewed in Tsukita et al., 1997). The ERM proteins are ~80kDa in size and contain a highly conserved N terminal domain which is also found in the erythrocyte membrane protein Band 4.1 (Arpin et al., 1994). This region of the ERM proteins binds membrane proteins including the cytoplasmic region of the membrane glycoprotein CD44 (Tsukita et al., 1994; Hirao et al., 1996). The C terminal region of the ERM proteins contains an actin binding domain at the last 34 amino acids which is also highly conserved (Turunen et al., 1994) and binds actin filaments in vitro (Turunen et al., 1994; Pestonjamasp et al., 1995; Roy et al., 1997). N and C terminal fragments of ERM proteins interact in vitro masking the CD44 and actin filament binding domains (Gary and Bretscher, 1993; Andreoli et al., 1994; Gary and Bretscher, 1995; Magendantz et al., 1995), suggesting an intramolecular head to tail association of individual ERM proteins whilst intermolecular interactions have also been observed in vivo (Berryman et al., 1995; Bretscher et al., 1995). Introduction 32 Exactly how ERM proteins are activated to induce an open conformation is unknown at present, however the ERM proteins bind PI-4-P and PI-4,5-P2 which enhances their ability to bind CD44 (Niggli et al., 1995; Hirao et al., 1996) and are also tyrosine and serine/threonine phosphorylated (Arpin et al., 1994) and either process may be involved in their activation (Chen et al., 1995; Jiang et al., 1995). The Rho family p21s are involved in regulating the activity of ERM proteins and thus the attachment of actin filaments to the plasma membrane. Rho was demonstrated to regulate the formation of CD44-ERM complexes (Hirao et al., 1996) and the association of ERM proteins with the plasma membrane in MDCK cells (Takaishi et al., 1995; Kotani et al., 1997). It also appears that moesin is required for Rho-dependent stress fibre formation in serum starved Swiss 3T3 cells (Mackay et al., 1997). The Rho effector ROK, was recently shown to weakly phosphorylate full length and strongly phosphorylate a C terminal fragment of ERM protein, and this threonine phosphorylated fragment suppressed the head to tail association of ERM (Matsui et al., 1998). This suggests that phosphorylation via ROK stabilises the open conformation of these proteins although it is not responsible for their initial activation. ERM proteins phosphorylated by ROK are dephosphorylated by myosin phosphatase and the myosin binding subunit of this phosphatase (MBS) binds the N terminal of ERM proteins (Fukata et al., 1998). MBS is a ROK substrate and its phosphorylation via ROK inactivates the phosphatase, thus the phosphorylation state of ERM proteins is regulated by myosin phosphatase and ROK acting downstream from Rho. Rho guanine nucleotide dissociation inhibitor (RhoGDI), which binds the GDP form of Rho p21s and prevents their activation, binds an N terminal fragment but not full length ERM protein suggesting that ERM protein must first be activated and in an open conformation to enable binding (Takahashi et al., 1997). This interaction inhibits RhoGDI activity and releases the p21 for subsequent activation, thus ERM proteins promote activation of Rho proteins. Regulation of the actin cytoskeleton and the activity of actin binding proteins is a complex process affected by a variety of signalling molecules which enables reorganisation of this network in response to multiple cues.

1.3.2 The microtubule network The microtubule network consists of filaments which radiate out from the microtubule organising centre (MTOC) throughout the cytoplasm. The network is responsible for many functions including the control of , shape and

Introduction 33 movement, organelle movement, the plane of cell division and segregation of chromosomes at . The main MTOC is the centrosome which consists of two centrioles surrounded by a cloud of electron dense pericentriolar material (PCM), also known as the centrosome matrix, where y nucleates microtubule formation and stabilises the microtubule structure by binding its minus end (Kellogg et al., 1994; Zheng et al., 1995b).

1.3.2A Microtubule structure Microtubules consist of two highly homologous proteins, a and P tubulin which are -450 amino acids in size. Many isotypes of these proteins exist which vary in sequence at their C terminal, are post translationally modified and differentially expressed (reviewed in Luduena, 1998). Despite this, many isotypes are partially interchangeable since they are still able to form tubules although usually of a somewhat modified structure or function, a and p tubulin form dimers, the structure of which was recently solved via electron crystallography (Nogales et al., 1998) and these ap tubulin dimers associate in a head to tail manner to form a polar protofilament. Microtubules typically consist of thirteen protofilaments associated in a hollow tube structure 25nm in diameter (Bryan and Wilson, 1971; Weisenberg, 1972; Maldelkow et al., 1995), the structure of which has recently been solved at high resolution (Nogales et al., 1999). Like actin filaments, microtubules are polar structures and polymerisation occurs at the plus end whilst the minus end is fixed at the MTOC. Both a and P tubulin bind GTP and hydrolysis of GTP bound to P tubulin occurs when an ap dimer is added to the growing microtubule end, hence this GTP hydrolysis is essential for microtubule assembly. Thus a and P tubulin are associated with GTP and GDP respectively in assembled microtubules, except at the plus ends which are capped by GTP-p tubulin to stabilise the structure (reviewed in Downing and Nogales, 1998).

1.3.2B Microtubule associated proteins Several types of proteins associate with microtubules. Firstly the microtubule associated proteins (MAPs) (Mandelkow and Mandelkow, 1995) which were identified via their copurification with tubulin include the ~200kDa proteins MAP-1 A, MAP-IB, MAP-1C, MAP-2 and MAP-4 and the smaller neuronally expressed proteins Tau and MAP-2C, which have axonal and dendritic distributions respectively (Maccioni and

Introduction 34 Cambiazo, 1995). MAPs bind to the C terminal of tubulin, stabilise microtubule structure and are regulated by phosphorylation. Abnormally phosphorylated Tau no longer binds and stabilises neuronal microtubules and is characteristic of the neurofibrillary tangles in the brains of patients with Alzheimers disease (Hasegawa et al., 1992). Secondly, and superfamily members are microtubule associated motor proteins which strongly bind tubulin and are involved in organelle transport along microtubules toward the plus and minus ends respectively, via ATP hydrolysis (reviewed in Hirokawa et al., 1998). Together these proteins regulate microtubule dynamics and organisation throughout the cell cycle (Cassimeris, 1999).

1.3.3 The intermediate filament network Intermediate filaments are so called as they are between actin filament and microtubules in size at lOnm in diameter. They are fibrous proteins unlike actin and tubulin which are globular proteins, and their assembly is regulated by phosphorylation (reviewed in Inagaki et al., 1996). Intermediate filaments form a network which extends from a region near the nucleus throughout the cell and is responsible for the tensile strength of the cell. Several types of intermediate filaments exist including keratins, neurofilament proteins, vimentin, desmin and glial acidic fibrillary protein (G-FAP) which are expressed in different cell types and the nuclear lamins which are expressed in all cell types (Houseweart and Cleveland, 1998). Intermediate filament proteins have an N terminal head domain and a C terminal tail domain of variable sequence and a conserved central a helical rod domain. Two monomers associate in parallel to form a coiled-coil homodimer and two dimers align to form an antiparallel tetramer called a protofilament, which is the repeating unit of intermediate filaments. Protofilaments then arrange themselves in a staggered manner to form an extended filament structure lOnm in diameter (Fuchs and Weber, 1994). Keratins are expressed in epithelial cells and tissues and are responsible for their mechanical strength. They run continuously throughout tissues, crossing between cells at specialised junctions called desmosomes and hemidesmosomes, which are cadherin and integrin based junctions respectively (Green and Jones, 1996). The keratin intermediate filament network is anchored to the cell surface at these junctions via plaques containing the intermediate filament associated proteins desmoplakin, BPAG1, plectin and IFAP300 (Foisner and Wiche, 1991; Skalli et al., 1994, reviewed in Chou et al., 1997). The importance of keratins in maintaining the mechanical integrity of the cell

Introduction 35 is demonstrated by the effect of mutations which result in many skin disorders including human epidermolysis bullosa simplex (EBS), epidermolytic hyperkeratosis (EH) and epidermolytic palmoplantar peratoderma (EPPK) (Fuchs and Cleveland, 1998). The neuronally expressed neurofilament proteins NF-L/M/H are essential for establishing proper diameter which is a key determinant of conduction velocity, as originally shown in quail mutants lacking neurofilaments (Yamasaki et al., 1991). Abnormal accumulation and disorganisation of neurofilaments is characteristic of motor neuron diseases such as familial and sporadic amyotrophic lateral sclerosis (ALS), infantile muscular atrophy and hereditary sensory motor neuropathy (Fuchs and Cleveland, 1998).

1.3.3A Intermediate filament associated proteins (IFAPs) Intermediate filaments can be cross linked to microtubules and actin filaments. Microinjection of the conserved helix domain of IF proteins into fibroblasts disrupted EFs, microtubules and actin microfilament networks. It was suggested that the IF peptides competed for IFAPs which normally crosslink these networks together and their absence resulted in disruption of these networks (Goldman et al., 1996). IFAPs include desmoplakin, BPAG1, plectin and IFAP300 which in addition to anchoring IF networks to the cell surface also bind other cytoskeletal components, crosslinking them (Foisner and Wiche, 1991). Plectin is a 500kDa protein which binds keratins, vimentin, GFAP, all three neurofilament proteins, lamin B, microtubule associated proteins and a- spectrin (Foisner et al., 1988; Foisner et al., 1991). Immunoelectron microscopy has shown that plectin forms cross bridges between IFs and microtubules, IFs and actin filaments and also IFs and myosin (Svitkina et al., 1996). The importance of plectin in maintaining the mechanical resistance of cells via organisation of the cytoskeleton is illustrated by the effects of plectin loss in mice and humans, which results in the disorder muscular dystrophy and epidermolysis bullosa simplex (MD-EBS) typified by skin blistering, muscle weakness and degeneration and death (Gache et al., 1996; McLean et al., 1996; Smith et al., 1996; Andra et al., 1997).

1.4 Protein domains Many different conserved protein domains exist which bind a variety of specific target sequences. Multiple domains are often present in single molecules which enables the formation of large, multiprotein complexes which produce a regulated response to extracellular signals (reviewed in Pawson, 1995; Bork et al., 1997; Pawson and Scott,

Introduction 36 1997). Some of these conserved domains are discussed below.

1.4.1 SH2 domains Src homology 2 (SH2) domains are conserved modules of -100 amino acids which recognise phosphotyrosine containing peptide sequences and are involved in tyrosine kinase signalling pathways (reviewed in Pawson, 1995). SH2 domains bind their target phosphopeptides with affinities in the range of 10-lOOnM, approximately

1 0 0 0 fold stronger than random phosphorylated sequences and have negligible affinity for unphosphorylated sequences. The residues C terminal to phosphotyrosine determine the specificity of the interaction and can affect the binding affinity three fold. SH2 domains are divided into 2 classes on the basis of their interactions with the residues C terminal to phosphotyrosine. Class I SH2 domains such as Src recognise three C terminal residues and the preferred sequence is phosphotyrosine-hydrophilic- hydrophilic-hydrophobic whilst class II SH2 domains, such as those in phospholipase Cy-1 and the tyrosine phosphatase SH-PTP2, recognise at least five hydrophobic C terminal residues (Songyang et al., 1993). The structure of several SH2 domains complexed with their specific target sequences have been solved (reviewed in Schaffhausen, 1995) providing detailed information concerning the mechanism of ligand binding. It was found that the phosphotyrosine residue extends into a conserved binding pocket of the SH2 domain structure and forms two hydrogen bonds to a conserved arginine at its base (Waksman et al., 1992; Lee et al., 1994). This residue is invariant in most known SH2 domains and mutation or even conserved substitution of this residue abolishes phosphotyrosine binding (Bibbins et al., 1993). The residues C terminal to phosphotyrosine are recognised by a second less conserved binding region, the structure of which depends on the class of the SH2 domain. In class I SH2 domains the hydrophobic third residue fits into a small hydrophobic pocket whilst in class II SH2 domains the C terminal residues fit into a hydrophobic groove (Eck et al., 1993; Waksman et al., 1993; Lee et al., 1994). SH2 domains are present in many types of protein including non catalytic adaptor proteins like Grb2, docking proteins like She, transcription factors like ST AT and also kinases like PI3K and Src. Although SH2 domains usually mediate intermolecular interactions, examples of intramolecular SH2-phosphotyrosine interactions which regulate protein activity also exist. Src family non receptor tyrosine kinases are kept in an inactive state via interaction between the phosphorylated C terminal tyrosine 527 and its SH2 domain (Xu et al., 1997). The phosphotyrosine

Introduction 37 binding ability of SH2 domains can also be regulated by phospholipids, as in PI3K where binding of PIP 3 to its SH2 domain inhibits the interaction of the SH2 domain with tyrosine phosphorylated proteins (Rameh et al., 1995).

1.4.1A Unusual SH2 domains However, not all SH2 domains have exactly class I and II structures. Tandem SH2 domains which bind peptide sequences containing two phosphotyrosine residues are present in several proteins including ZAP70 (Hatada et al., 1995), SH-PTP2 (Pluskey et al., 1995) and the p85 subunit of PI3K (Carpenter et al., 1993). Determination of the structure of ZAP-70 bound to its target phosphopeptide (Hatada et al., 1995) found that the C terminal SH2 domain bound the N terminal phosphotyrosine residue via a typical pocket structure however, the N terminal SH2 domain had an incomplete phosphotyrosine binding pocket which was completed via residues from the C terminal SH2 domain (Hatada et al., 1995; Weiss, 1995; Wandless, 1996). SH2 domains contain several highly conserved residues, the pB5 arginine at the base of the phosphotyrosine binding pocket, which is invariant in all known SH2 domains and the second most conserved residue is the (3A1 tryptophan which is the first residue of the SH2 domain. Only four SH2 domains have been identified where the first residue of the SH2 domain is not a tryptophan, these are a l- and p2-chimaerin which have glutamate, EAT-2 which has tyrosine and ZAP70 which has phenylalanine (Hall et al., 1993; Leung et al., 1994; Thompson et al., 1996; Hatada et al., 1995 respectively). The effects of this on SH2 domain function are unknown, however in Src, mutation of the pAl tryptophan to glutamate abolished phosphotyrosine binding (Bibbins et al., 1993).

I.4.1B Non phosphotyrosine dependent SH2 interactions Phosphotyrosine independent SH2 domain interactions have also been demonstrated. The SH2 domain of the adaptor GrblO interacts in a phosphotyrosine independent manner with Rafl and MEK1 kinases, in a constitutive and insulin- dependent manner respectively and these interactions occur via phosphoserine or phosphothreonine residues (Nantel et al., 1998) and the SH2 domain of Fyn interacts with a phosphoserine residue of Raf (Cleghon and Morrison, 1994). Other examples include Bcr binding to the Abl SH2 domain in a phosphoserine/phosphothreonine- dependent manner (Pendergast et al., 1991) and phosphoserine-dependent binding of a 62kDa ubiquitin binding protein to the p56Lck SH2 domain (Joung et al., 1996). The Introduction 38 mechanisms of these phosphotyrosine independent interactions are unclear, but interaction of SH2 domains with phosphoserine and phosphothreonine residues would significantly increase the number of SH2 domain ligands and diversify the role of SH2 domains in signalling pathways.

1.4.2 Phosphotyrosine binding domains Phosphotyrosine binding (PTB) domains, like SH2 domains, bind phosphotyrosine containing sequences (Bork and Margolis, 1995). PTBs range in size from 1 0 0 - 2 0 0 amino acids and recognise phosphotyrosine containing peptides preceded by residues which form a (3 turn (Kavanaugh et al., 1995; Van der Geer et al., 1995) with the consensus sequence NPXpY (Trub et al., 1995). The specificity of the interaction is determined by 5-8 hydrophobic residues N terminal to the phosphotyrosine residue (Trub et al., 1995; Van der Geer et al., 1996), unlike SH2 domains which recognise residues C terminal to the phosphotyrosine. These domains are mainly found in docking proteins, like She and IRS-1 which recruit proteins to activated receptors (Kavanaugh and Williams, 1994; Blaikie et al., 1994; Gustafson et al., 1995). The PTB domains of She and IRS-1 have structural similarities to PH domains (Zhou et al., 1995; Zhou et al., 1996) and the She PTB binds acidic phospholipids (Zhou et al., 1995), suggesting the possibility of membrane recruitment via this region. PTB domains present in the neuronal proteins Fe65 and X I1 and Drosophila Numb protein were also able to bind non phosphorylated sequences (Borg et al., 1996; Li et al., 1997), suggesting these domains act as recognition motifs rather than phosphotyrosine binding motifs like SH2 domains.

1.4.3 SH3 domains Src homology 3 (SH3) domains are ~60 amino acids in size and bind proline- rich sequences approximately 10 amino acids in size (Ren et al., 1993; Yu et al., 1994) containing a core sequence of PXXP which forms a left handed polyproline type II helix with three residues per turn (Feng et al., 1994). SH3 domains bind their target sequences with affinities in the range of 5-100p,M (Pawson, 1995) and phosphorylation of serine/threonine residues next to these proline-rich sequences may influence SH3 interactions (Chen et al., 1996a). SH3 binding peptides are pseudo symmetrical and can bind in an N to C terminal (class I) or C to N terminal (class II) orientation (Feng et al., 1994; Lim et al., 1994; Wittekind et al., 1994) which increases the number of potential binding partners. Introduction 39 SH3 domains are present in many types of proteins including the p47phox and p67phox subunits of the neutrophil NADPH oxidase system, cytoskeletal proteins such as spectrin and myosin, adaptor proteins like Grb2 and Nek and the Src family of tyrosine kinases. They are often present in combination with SH2 domains and in multiple copies. SH3 domains are important in determining subcellular localisation and cytoskeletal interactions as shown in fibroblasts where the SH3 domains of Grb2 and PLCy directed these proteins to ruffles and stress fibres respectively (Bar Sagi et al., 1993). Although SH3-polyproline interactions are usually intermolecular, examples of intramolecular interactions also exist. Cytosolic p47phox exists in an inactive closed conformation via interaction of its C terminal polyproline sequence and its first SH3 domain (de Mendez et al., 1997) upon activation, phosphorylation of p47phox disrupts this interaction enabling assembly of the functional oxidase complex.

1.4.4 Pleckstrin homology domains PH domains were first identified as domains -120 amino acids in size with homology to regions of pleckstrin, the major PKC substrate in platelets (Haslam et al., 1993). PH domains are present in many proteins including Bruton's tyrosine kinase (Btk), the serine/threonine protein kinase B (PKB/Akt), P adrenergic receptor kinase (PARK), RasGAP, the RasGEF Sos, Rho family GEFs Dbl and Tiaml, Rho effectors MRCK and ROK and all isoforms of PLC (Musacchio et al., 1993) PH domains have low sequence homology but the three dimensional structures are highly conserved (reviewed in Lemmon et al., 1996) and they bind phosphoinositides, PKC and the Py subunit of heterotrimeric G proteins (reviewed in Shaw, 1996). The PTB domains of She and IRS-1 have structural similarities to PH domains (Zhou et al., 1995; Zhou et al., 1996) and also bind phosphoinositides. The specificity of phosphoinositide binding varies between proteins (reviewed in Lemmon et al., 1997) and probably targets proteins to specific membrane locations. Several strong interactions have been demonstrated and these include the pM binding affinity of PI-4, 5 -P2 to the PLC51 PH domain (Lemmon et al., 1995). This recruits PLC to the membrane where its substrates are present, PI-4,5 -P2 is hydrolysed to I-l,4,5-P3 which then binds the PH domain and dissociates PLC from the membrane, thus phosphoinositide binding regulates PLC activity via its localisation (Lemmon et al.,

1996). The PH domain of Sos also binds PI-4,5-P 2 which inhibits its RasGEF activity via an unknown mechanism (Jefferson et al., 1998). Some PH domains interact Introduction 40 Domain Proteins in which Liean d/function domain is present

Band 4.1 ERM proteins, band 4.1 Binds membranes Ca2+ dependent lipid PKC, PLC, PLD, Binds calcium, binding (CalB) perforin phospholipids and intracellular proteins Calponin homology Dystrophin, Vav Binds actin (CH) Cysteine rich (CRD) a- and p-chimaerins, Binds DAG and PKC, Vav, Raf phospholipids Dbl homology (DH) Dbl, Sos, FGD1, Lbc, Promotes exchange of GDP Tiaml, Vav, Trio for GTP DEATH domain focal adhesions, RIP, Causes dimerisation ankyrins Dbl homology (DH) Dbl, Sos, FGD1, Lbc, Promotes exchange of GDP Tiaml, Vav, for GTP Formin homology MDia Binds cofilin, links Rho and (FH) the actin cytoskeleton GTPase activating a- and P-chimaerins, Stimulates intrinsic GTP (GAP) Bcr, Abr, p i20 RasGAP, hydrolysis rate neurofibromin, PI3K LIM Paxillin, enigma, pinch Binds Zn2+ and tyrosine containing sequences PDZ IL-6 , densin-180, PTP- Binds C terminal peptides 1H, LIM kinase, Dsh PTB She, IRS-1 Binds phosphotyrosine Pleckstrin homology Btk, PARK, RhoGEFs, Binds phosphoinositides, (PH) PKB/Akt targets protein to the membrane Sterile alpha motif Byr2, eph like receptor Causes dimerisation (SAM) tyrosine kinases SH2 Src, She, Grb2, PI3K, Binds phosphotyrosine SH-PTP2 SH3 Src, spectrin,Grb2, Nek Binds proline-rich sequences with PXXP consensus WD40 GPy, RACK1 Binds proteins WW Dystrophin, RSP5 Binds proline-rich sequence with PPXY or PPLP consensus

Table 1.1: Protein signalling domains Examples of different types of protein domains, their function or ligands and examples of proteins which contain these domains are shown. Abbreviations include: GPy, py subunit of heterotrimeric G protein; PK, protein kinase; PL, phospholipase; PTP, protein tyrosine phosphatase.

Table 1.1 41 specifically with the D3 phosphoinositides produced by PI3K, these include Akt which binds and is activated by PI-3,4-P2(Franke et al., 1997; Klippel et al., 1997) and Btk which binds PI-3,4,5-P3 and 1-1,3,4, 5 -P4 (Fukudu et al., 1996; Salim et al., 1996). PH domains are found in many types of proteins, often in combination with SH2 and SH3 domains and proteins which contain a Dbl homology (DH) domain, which catalyses guanine nucleotide exchange on Rho p21s, always contain a PH domain C terminal. GEFs are discussed in detail later.

1.4.5 Other protein domains Many different protein domains with very specific substrates have been identified including the Cdc42/Rac interactive binding (CRIB) domain present in the p21 activated protein kinases (PAKs) which binds Cdc42 or Rac inducing activation of the kinase (Sells and Chernoff, 1997) and also the Rho binding domain present in ROK, Rhotekin, Rhophilin and PKN (Leung et al., 1996; Reid et al., 1996; Watanabe et al., 1996). Examples of other protein domains identified are shown in table 1.1. The conserved GAP, DH and CRD protein domains are discussed in later sections.

1.5 GTPase superfamilv Members of the GTPase superfamily regulate many diverse cellular activities. Three main classes of these GTP binding proteins have been identified - the initiation and elongation factors which are involved in protein synthesis, the heterotrimeric G proteins which are involved in and the Ras family of small GTPases. Interactions of these GTPases with other proteins is dependent on their nucleotide state. The GTPases function as molecular switches, cycling between the GDP and GTP bound forms (inactive and active forms) and have an intrinsic GTPase activity (Bourne et al., 1991). This cycling is regulated by proteins which affect the rate of GDP/GTP exchange or increase the intrinsic rate of GTP hydrolysis. The GAPs stimulate the intrinsic rate of GTP hydrolysis thus inactivating the GTPase, the GEFs promote exchange of GDP for GTP thus activating the GTPase and the GDIs inhibit the dissociation of GDP locking the GTPase in the inactive state, as shown in figure 1.2 (reviewed in Boguski and McCormick, 1993; Van Aelst and D'Souza-Schorey, 1997).

1.5.1 Heterotrimeric G proteins Heterotrimeric G proteins bind the cytoplasmic face of transmembrane receptors for hormonal, sensory and neurotransmitter signals and transduce their signals

Introduction 42 GEFs

GTP

GDP INACTIVE

ACTIVE GDIs p21-GTP p21-GDP

GAPs Neutrophil Transcnptional Oxidase Activation

Figure 1.2; The GTPase cycle Low molecular weight GTPases (p21s) cycle between an inactive GDP bound state and an active GTP bound state. This cycling is regulated by several types of proteins; the GTPase activating proteins (GAPs) which downregulate p21 activity via increasing the intrinsic rate of GTP hydrolysis, the guanine nucleotide exchange factors (GEFs) which activate the p21s by catalysing the exchange of bound GDP for GTP and the guanine nucleotide dissociation inhibitors (GDIs) which prevent p21 activation by sequestering the inactive GDP bound p21 in a cytosolic complex.

Figure 1.2 43 (reviewed in Emala et al., 1994). Their activation results in many effects including the regulation of various (reviewed in Neer, 1995), activation of ion channels (reviewed in Krapivinsky et al., 1995) and vesicular transport (reviewed in Helms, 1995). Heterotrimeric G proteins are composed of three subunits. The a subunit binds GTP and possesses intrinsic GTPase activity, whilst the J3 and y subunits associate and function as a Py monomer and anchor the complex to the membrane. The a and py subunits associate in an inactive complex when G a is in the GDP bound form. Ligand stimulation of the receptor induces a conformational change which reduces the affinity of Ga for GDP, which dissociates, enabling GTP binding. The active GTP bound Ga subunit dissociates from Gpy and activates an effector such as adenylate cyclase. The activity of Ga is regulated by its intrinsic GTPase activity which hydrolyses bound GTP to GDP, resulting in reassembly of the inactive heterotrimeric G protein complex (reviewed in Neer, 1994). Multiple isoforms of each of the three G protein subunits exist and Ga subunits are divided into four classes, the as, ai, aq and a 1 2 , on the basis of amino acid sequence (reviewed in Hamm and Gilchrist, 1996). Different combinations of subunits produce functionally distinct G proteins with particular specificity and characteristics, enabling coupling of different types and subtypes of receptors to a unique G protein. Although the mechanism of G protein-mediated activities are usually via the active Ga subunit, examples of Gpy-mediated activity have also been demonstrated. These include activation of muscarinic gated potassium channels, PLCp isoforms, phospholipase A2, , PI3K, Ras and the tyrosine kinases Tsk, Btk and PYK2 (reviewed in Hamm and Gilchrist, 1996). Thus activation of G protein coupled receptors induces activation of many different effectors, which are involved in or lead to activation of other cellular signalling cascades including receptor tyrosine kinase signalling, resulting in cross talk between these pathways (reviewed in Selbie and Hill, 1998).

1.5.2 Ras subfamilies of low molecular weight GTPases Ras family proteins are -190 amino acids, most of which make up the GTP binding domain, these low molecular weight proteins are also referred to as p21s. The Ras family is divided into subfamilies which include , Arf, Sar, , and Rho proteins (reviewed in Boguski and McCormick, 1993). Ras proteins regulate cell growth

Introduction 44 and differentiation, Rab, Arf and Sar regulate intracellular vesicle and protein trafficking, Ran regulates nuclear transport and Rho family proteins are involved in many activities including regulation of the actin cytoskeleton, transcription, MAPK cascades and the neutrophil oxidase system which are discussed in detail later.

1.5.2A Ras Ras proteins were identified as the transforming agents involved in Harvey and Kirsten sarcoma viruses (Barbacid, 1987). Four Ras proteins are expressed in mammalian cells - H -ras, K-rasA, K-rasB and N -ras which each function as an oncogene when activated via mutation and are involved in many human cancers. These Ras proteins are closely related to members of the Rap subfamily. A related protein R- ras is 55% identical to H-ras but is unable to transform cells although its C terminal can bind Bcl-2, a protein which regulates apoptosis (Fernandez-Sarabia and Bischoff, 1993). Ras proteins are involved in many processes including proliferation, differentiation and apoptosis.

1.5.3 The regulation of GTPase proteins

1.5.3A GAPs for heterotrimeric G proteins The activity of heterotrimeric G proteins is regulated by RGS proteins (regulators of G protein coupled signalling) which bind the activated Ga subunit, preventing activation of downstream effectors and signalling pathways (reviewed in Hepler, 1999). All RGS proteins identified also act as GAPs for Ga subunits. Both the GAP and Ga binding domains are present within an RGS domain which varies considerably in sequence between proteins, pi 15 RhoGEF contains an N terminal RGS domain and acts as a GAP for the a subunits of G 12 and G 13 heterotrimeric G proteins, although it is a more potent GAP for G an (Kozasa et al., 1998). The interaction between pi 15 RhoGEF and Gan, but not Gan, also stimulates the RhoGEF activity of pi 15 (Hart et al., 1998) whilst activated G an inhibits Gan-induced stimulation of RhoGEF activity. Rho mediates signalling downstream from some Gn and Gn heterotrimeric G proteins, such as those coupled to the LPA receptor. Thus pi 15 RhoGEF provides a direct link between heterotrimeric G protein and Rho-mediated signalling pathways.

Introduction 45 1.5.3B GAPs for Ras Ras proteins have low intrinsic rates of GTP hydrolysis and their inactivation depends on GTPase activating proteins in vivo, p i20 RasGAP was the first such protein characterised (Trahey and McCormick, 1987 It contains a hydrophobic N terminal region, two SH2 domains, an SH3 domain, a PH domain, a region with sequence homology to the CalB region of phospholipase A 2 and an approximately 250 residue C terminal GAP domain. The two SH2 domains in p i20 RasGAP are able to bind tyrosine phosphorylated p i90 RhoGAP (Settleman et al., 1992b) and the binding induces a conformational change in p i20 RasGAP which increases the accessibility of its SH3 domain (Hu and Settleman, 1997). This interaction serves as a point of potential cross talk between the Ras and Rho pathways and also suggests that p i90 RhoGAP may promote the SH3 domain-dependent interactions of pi 20 RasGAP. Another protein which contains a RasGAP domain is neurofibromin (Trahey and McCormick, 1987; Martin et al., 1990). Defects in the neurofibromin (NF1) gene result in the condition von Recklinghausen neurofibromatosis which is characterised by malignant Schwannomas and abnormal neural crest cell growth. RasGAP and neurofibromin are not Ras specific but can also act as a GAP for the related protein R- ras (Rey et al., 1994). IQGAP1 (Weissbach et al., 1994; Hart et al., 1996a) and IQGAP2 (Brill et al., 1996) also contain a RasGAP domain but recombinant IQGAP1 had no GAP activity for Ras or Rho (Weissbach et al., 1994). However, IQGAP1 and IQGAP2 bind Cdc42 and Rac and the IQGAP1 interactions were GTP-dependent, suggesting Cdc42 and Rac effector functions for this protein. (Kuroda et al., 1996; McCallum et al., 1996). In addition to binding Cdc42 and Rac, IQGAP1 binds F-actin and cross links actin filaments and thus may link these p21s and actin cytoskeleton rearrangement (Bashour et al., 1997). IQGAP homologues Iqglp and DGAP1 have been identified in yeast and dictyostelium, respectively (Osman and Cerione, 1998; Faix et al., 1998).

1.5.3C GAPs for Rho family proteins Many RhoGAPs of varying specificity have been identified (reviewed in Lamarche and Hall, 1994; Van Aelst and D'Souza-Schorey, 1997). The first GAP with activity for Rho family proteins but not Ras, was a 29 kDa protein purified from human spleen extracts (Garrett et al., 1989; Garrett et al., 1991), which was later found to be a 29 kDa C terminal truncation of a 50 kDa protein, p50 RhoGAP (Barfod et al., 1993; Lancaster et al., 1994). Partial sequence analysis of p29 RhoGAP revealed homology

Introduction 46 with the C terminal regions of both al-chimaerin and Bcr, and these regions of the two proteins were demonstrated to possess GAP activity for Rac, but not Rho in vitro (Diekmann et al., 1991). Bcr is the product of the breakpoint cluster region gene. It is a multidomain protein which contains a serine/threonine kinase domain, a DH domain and a PH domain in addition to its RhoGAP domain. It was first identified as part of the chimaeric protein sequence present in human leukaemias which contain the Philadelphia chromosome translocation (Heisterkamp et al., 1985). Translocations result in a truncated Bcr sequence fused to the Abl tyrosine kinase sequence. In the p 2 1 0 Bcr-Abl fusion present in chronic myelogenous leukaemia (CML), Bcr lacks the GAP domain, whilst both the GAP and PH domains are absent in the pl85Bcr-Abl fusion present in acute lymphocytic leukaemia (ALL). Full length Bcr was found to have GAP activity for both Rac and Cdc42 in vitro, although Rac is the preferred substrate (Ridley et al., 1993). p50RhoGAP was found to stimulate the GTPase activity of Rho, Rac and Cdc42 in vitro, but Cdc42 was the preferred substrate (Barfod et al., 1993). Unexpectedly p50RhoGAP did not exhibit differential binding for the GTP and GDP bound forms of the Rho proteins, unlike p i20 RasGAP which has a hundred fold stronger affinity for GTP bound Ras (Schaber et al., 1989). Despite its broad specificity in vitro, p50RhoGAP has a more restricted specificity in vivo where the GAP domain exhibited only Rho-GAP activity (Ridley et al., 1993). p50RhoGAP also contains a proline-rich sequence whichin vitro binds to the SH3 domains of c-Src and p85a, the regulatory subunit of PI3K (Barfod et al., 1993). Many other proteins which have GAP activity for the Rho family proteins have been identified. Abr which is closely related to Bcr, contains DH and RhoGAP domains and also has GAP activity for both Rac and Cdc42 (Tan et al., 1993). Other proteins which have GAP activity for both Rac and Cdc42 in vitro include 3BP-1, CdGAP, MgcRacGAP which is the human homologue of Drosophila rotund (Toure et al., 1998), RalBPl (Cantor et al., 1995), RIP1 (Park and Weinberg, 1995) and RLIP76 (Jullien- Flores et al., 1995), although Cdc42 is the preferred substrate for RalBPl, RIP1 and RLIP76. In vivo, 3BP-1 acts as a RacGAP and inhibited Rac-induced ruffling in fibroblasts (Cicchetti et al., 1992; Cicchetti et al., 1995), whilst CdGAP acts as both a Rac and Cdc42 GAP and inhibited Rac-induced ruffling and Cdc42-induced filopodia formation in fibroblasts (Lamarche-Vane and Hall, 1998). The a- and p-chimaerins also have GAP activity in vitro for Rac and to a lesser extent for Cdc42 and are discussed in detail later. Introduction 47 pl90GAP was first identified as a tyrosine phosphorylated protein associated with pl20 RasGAP in Src transformed cells (Ellis et al., 1990) and was later shown to contain a RhoGAP domain with GAP activity for Rho, Rac and Cdc42 in vitro, although Rho is the preferred substrate (Settleman et al., 1992a). Other proteins with GAP activity for Rho include PARG1 which has GAP activity for Rho, Rac and Cdc42 in vitro, but exhibits a marked preference for Rho (Saras et al., 1997), Myr5 which has GAP activity for both Rho and Cdc42, although Rho is the preferred substrate (Reinhard et al., 1995) and p i22 which only has GAP activity for Rho (Homma and Emori, 1995). Also Graf (GTPase regulator associated with FAK) has GAP activity for both Rho and Cdc42 in vitro (Hildebrand et al., 1996), whereas in vivo Graf only exhibits GAP activity for Rho (Taylor et al., 1999). The RhoGAP domain is also present in the p85a and p85p subunits of PI3K, however no GAP activity for Rho, Rac or Cdc42 has been demonstrated (Otsu et al., 1991). Rho family GAPs have also been identified in other species including in Drosophila rotund (Agnel et al., 1992), in C.Elegans, CeGAP which has GAP activity for Rho, Rac and Cdc42 (Chen et al., 1994) and in S.Cerevisiae, BEM2 and BEM3 which have GAP activity for Rhol and both Rhol and Cdc42Sc, respectively (Zheng et al., 1994a).

1.5.3D Multidomain nature of GAPs Many p21 GAPs also contain other modular protein domains such as SH2, SH3, PH, DH, CRD and proline-rich SH3 binding domains, as previously described. These other domains have been shown to regulate GAP activity, protein structure, protein localisation and also mediate other enzymatic functions. The regulation of GAP activity by other regions within the same protein has been demonstrated for several p 2 1 GAPs. The binding of phorbol esters and phospholipids to the CRD of the a-chimaerins regulates the RacGAP activity of these proteins in vitro (Ahmed et al., 1993; M. Teo PhD thesis, 1994). Whilst in vivo, the Rac/Rho GAP activity of pi 90 RhoGAP is regulated by its N terminal GTP binding domain (Tatsis et al., 1998). The interaction of the two SH2 domains of pi 20 RasGAP with tyrosine phosphorylated p i90 RhoGAP, results in the downregulation of pl20 GAP activity (Moran et al., 1991). This interaction also induces a conformational change in p i20 RasGAP structure which increases the accessibility of its SH3 domain (Hu and Settleman, 1997). Interactions of other domains within GAPs can affect protein localisation. The Introduction 48 SH3 domain of Graf, a Rho/Cdc42 GAP, interacts with a proline-rich region in the C terminal of focal adhesion kinase (FAK) (Hildebrand et al., 1996). FAK is a tyrosine kinase involved in integrin signalling pathways and its interaction with Graf recruits Graf to focal adhesion complexes. The N terminal SH2 domains of RasGAP interact with tyrosine phosphorylated PDGF p receptor. This interaction alters the protein localisation of pl20 RasGAP inducing its translocation to the membrane (McGlade et al., 1993). The function of some interactions mediated by other regions within GAP proteins has not yet been determined. The proline-rich regions of p50 RhoGAP bind the SH3 domains of Src and the p85 subunit of PI3K (Barfod et al., 1993). Similarly, the proline-rich region of the Rac and Cdc42 GAP, 3BP-1, binds the SH3 domain of the non receptor tyrosine kinase Abl (Cicchetti et al., 1992). It is possible that these interactions may regulate protein localisation or function. The C terminal region of the RhoGAP PARG1 interacts with the fourth PDZ domain of the protein tyrosine phosphatase PTPL1 (Saras et al., 1997). This interaction may regulate tyrosine phosphorylation involved in Rho p21 signalling pathways. Several p21 GAPs contain protein domains which mediate other functions. Abr and Bcr contain both a RhoGAP and RhoGEF domain which activate or inactivate Rho p21s, respectively (Chuang et al., 1995). Thus the net effect of these proteins on p21 activity depends on the regulation and relative activities of these two protein domains, p i22 has GAP activity for Rho in vitro, however p i22 also binds PLC 51 and stimulates its enzymatic activity (Homma and Emori, 1995). Some p21 GAPs have also been demonstrated to mediate effector functions, which is discussed in detail later. Thus p21 GAPs mediate many other functions in addition to downregulating p21s via increasing their intrinsic GTPase activity.

1.5.3E RasGEFs GEFs associate with the GDP bound form of the p21 and enhance the rate of GDP dissociation. As cells contain much higher concentrations of GTP than GDP, free GTP immediately binds the available p21. This results in a loss of affinity for the GEF, which then dissociates leaving the p21 in its active state (Bourne et al., 1990). Dominant negative (N17) mutants of Ras have a reduced affinity for GTP and are unable to displace the GEF and replace it with GTP. This results in the sequestering of Ras in the inactive GDP bound form (Lai et al., 1993). The RasGEF domain is an approximately 200 residue domain composed of 3 Introduction 49 SCRs (structurally conserved regions) separated by variable regions. The S.Cerevisiae GEFs Scd25 and cdc25 are able to bind human Ras (Bollag and McCormick, 1991), which suggests that the p21/GEF interaction occurs via highly conserved sequences which are also conserved across species. An example of Ras activation which occurs as a result of GEF activity is provided by the activation of Ras via Sos upon EGF receptor stimulation The RasGEF

Sos exists in a cytosolic complex with the adapter protein Grb 2 (Bowtell et al., 1992; Chardin et al., 1993). Binding of EGF to its receptor induces receptor dimerisation and autophosphorylation of its intracellular tyrosine residues. The SH2 domain of Grb2 binds a phosphotyrosine containing sequence (Tyrl068) of the EGF receptor (Rozakis- Adcock et al., 1993), which brings Sos to the membrane where it activates Ras. Analagous proteins are also involved in Ras activation of Drosophila eye development and C.Elegans vulval development (reviewed in Dickson and Hafen, 1994).

1.5.3F GEFs for Rho family proteins The Dbl oncogene product is a large cytosolic protein containing a region with homology to the S.Cerevisiae cell division protein Cdc24 (activator of Cdc42) and also the human Bcr gene product (Ron et al., 1991). As predicted on the basis of sequence similarity to Cdc24, Dbl acts as a GEF for Cdc42Hs (Hart et al., 1991) and also Rho. The region responsible for this GEF activity was designated the Dbl homology (DH) domain (Hart et al., 1994) which was also found to be required for Dbl transforming function (Ron et al., 1991). Dbl contains an adjacent C terminal pleckstrin homology (PH) domain which is required for proper cellular localisation of the protein (Zheng et al., 1996a). Other proteins containing both DH and PH domains, many of which are oncogenic, have been identified as Rho family GEFs with differing specificities (reviewed in Cerione and Zheng, 1996; Van Aelst and D'Souza-Schorey, 1997).

1.5.3F1 Cdc42 specific GEFs FGD1 is the product of the faciogenital dysplasia or Aarskog-Scott syndrome locus, is required for normal embryonic development and has GEF activity for Cdc42 in vitro (Pasteris et al., 1994; Zheng et al., 1996b). FGD1 and the related protein Frabin (FGD1 related F-actin binding protein), which also contains a RhoGEF domain, both produced Cdc42-induced filopodia in Swiss 3T3 fibroblasts and stimulated JNK activity in COS cells (Olson et al., 1996; Obaishi et al., 1998). FGD1 also induced transformation in NIH 3T3 cells but differences in mechanism to Cdc42-induced

Introduction 50 transformation were observed (Whitehead et al., 1998).

1.5.3F2 Rho specific GEFs The oncogene Lbc has GEF activity for Rho in vitro and also in vivo, as microinjection of Lbc induced Rho-mediated stress fibre formation in fibroblasts (Toksoz and Williams, 1994; Zheng et al., 1995a; Olson et al., 1996) whilst the oncogenes Lfc and Lsc have in vitro GEF activity for Rho (Whitehead et al., 1995a; Glaven et al., 1996; Aasheim et al., 1997). pi 15RhoGEF is a Rho specific GEF both in vitro and in vivo, where it induced transformation of NTH 3T3 cells (Hart et al., 1996b), similar to V14Rho and Lbc. mNETl, the mouse homologue of the previously identified human NET1 protein (Chan et al., 1996), also has GEF activity for Rho A both in vitro and in vivo where mNETl induced stress fibre formation in NIH 3T3 cells and potentiated SRF-dependent transcription, both Rho-dependent processes (Alberts and Treisman, 1998).

1.5.3F3 Rac specific GEFs Tiaml is a RacGEF both in vitro (Habets et al., 1994) and in vivo where it induces Rac-dependent ruffling in fibroblasts and COS7 cells and colocalised with F actin in ruffles. An N terminal PH domain rather than the RhoGEF PH domain was required for membrane localisation of Tiaml and both GEF activity and membrane localisation were essential for induction of Rac-dependent ruffling and JNK stimulation by Tiaml (Michiels et al., 1995; Michiels et al., 1997). InN IE 115 neuroblastoma cells, Tiaml induced Rac-dependent neurite outgrowth and cell spreading on laminin and Tiaml expressing cells were no longer sensitive to LPA-induced neurite retraction and cell rounding mediated by Rho, which suggests that Tiaml-induced Rac activation antagonises Rho signalling in these cells (Van Leeuwen et al., 1997). This is discussed in more detail later. PIX (Pak interacting exchange factor) has GEF activity in vitro for Cdc42>Rac>Rho, although in vivo PIX acts as a Rac specific GEF (Manser et al., 1998).

1.5.3F4 Multiple specificity GEFs Vav has GEF activity for Cdc42, Rac and Rho in vitro. Tyrosine phosphorylation of Vav by Lck stimulated the Cdc42 GEF activity of Vav and was required for Rac and Rho GEF activity (Han et al., 1997). However, Lck had no effect on the Cdc42 GEF activity of Dbl under similar conditions, which suggests that tyrosine

Introduction 51 phosphorylation is not a universal method for regulation of GEF activity (Crespo et al., 1997). In vivo Vav-induced filopodia in rat embryonic fibroblasts (REF-52) cells (Han et al., 1997), whilst both Vav and Dbl-induced stress fibres, ruffles and focal complexes in Swiss 3T3 fibroblasts and JNK activation in COS1 cells (Olson et al., 1996). Several GEFs couple Rho and Rac pathways. Trio has one Rho specific and one Rac specific GEF domain (GEFD2 and GEFD1 respectively) in vitro as well as a serine/threonine kinase domain (Debant et al., 1996). In vivo, the GEFD2 domain induced Rho-dependent stress fibres, the GEFD1 domain induced Rac-dependent ruffles and lamellipodia and a construct containing the two GEF domains together induced both Rac and Rho type morphology in Swiss 3T3 fibroblasts. Thus Trio links the Rho and Rac morphological signalling pathways in vivo. The GEFD1 domain also induced Rac-dependent JNK activation (Bellanger et al., 1998). Ost has GEF activity for both Cdc42 and Rho in vitro and binds GTP-Rac which suggests Ost functions as a Rac effector, thus linking Rho, Rac and Cdc42 signalling pathways (Horii et al., 1994). Abr and Bcr contain two domains involved in Rho GTPase regulation; both proteins have GEF activity for Cdc42, Rac and Rho in vitro and also GAP activity for Rac and Cdc42 (Tan et al., 1993; Chuang et al., 1995). Other RhoGEF domain containing proteins whose substrates have not yet been determined include Ect-2 (Miki et al., 1993), Tim (Chan et al., 1994) and Dbs (Whitehead et al., 1995b).

1.5.3G Guanine nucleotide dissociation inhibitors GDIs bind the GDP form of p21s and sequester them in an inactive complex in the cytosol. This prevents p21 activation by GEFs and their subsequent translocation to the membrane. Three isoforms of RhoGDI have been isolated, a, P and y with different expression patterns and specificities. RhoGDIa, previously called RhoGDI, was originally purified from bovine brain as a soluble, ubiquitously expressed protein which preferentially bound the GDP form of Rho A or RhoB, inhibiting GDP dissociation (Fukumoto et al., 1990; Ueda et al., 1990) and was later found to have the same action on Cdc42 and Rac (Abo et al., 1991; Leonard et al., 1992). This activity was inhibited upon RhoGDIa binding to the N terminal of ERM proteins, resulting in the release of GDP bound p21 (Takahashi et al., 1997). RhoGDIa also weakly bound the GTP forms of Rho, Rac and Cdc42 and inhibited both intrinsic and GAP stimulated GTPase activity (Hart et al., 1992; Chuang et al., 1993). Human and mouse RhoGDI P, previously called Ly- or D4-GDI is only expressed in haematopoietic cells (Lelias et al., Introduction 52 1993; Scherle et al., 1993) and binds RhoA, Rac and Cdc42 (Adra et al., 1993; Scherle et al., 1993). Human RhoGDI y is expressed in brain and pancreas and binds Cdc42 and RhoA (Adra et al., 1997) whilst the mouse homologue, RhoGDI 3 is a non cytosolic protein which binds RhoB and RhoG but not RhoA, RhoC or Racl (Zalcman et al., 1996). Thus, unlike RhoGAPs and RhoGEFs, only a small number of RhoGDIs exist which have different tissue distribution, p 21 binding specificities and affinities.

1.5.4 p21 Effector proteins The downstream effects of the p21 proteins are mediated by effector proteins which interact with the active p21. Effectors may be cell type specific and/or developmental^ regulated (eg WASP, chimaerin) and may have single or multiple specificity for p21 proteins. Multiple effector pathways may be stimulated at the same time and the net response (eg transcription) may be a result of several different interacting or competing pathways. Thus a great deal of complexity is involved in the signalling cascades which determine the cellular response to p 21 stimulation.

1.5.4A Ras effectors Both the serine/threonine kinase Raf and the lipid kinase PI3K are Ras effectors (reviewed in Marshall, 1996). Raf binding to GTP-Ras could be demonstrated by using the yeast two hybrid system (Van Aelst et al., 1993; Vojtek et al., 1993; Zhang et al., 1993b). The exact mechanism of Raf activation is still under investigation but it is known that Ras binds the N terminal of Raf in vitro (Warne et al., 1993) and targets Raf to the membrane where it is activated by a series of events including phosphorylation by tyrosine kinases (Marais et al., 1995). It is suggested that the protein 14-3-3 is involved in regulating the Ras-Raf interaction. In the absence of GTP-Ras, 14-3-3 maintains Raf in an inactive form in the cytosol (Tzivion et al., 1998) and 14-3-3 is displaced from the Raf-14-3-3 complex after its recruitment to the membrane by Ras (Roy et al., 1998). Activated Raf then activates MEK1 via serine phosphorylation which in turn activates ERK1 and ERK2 via tyrosine and threonine phosphorylation. Activation of the ERK pathway is involved in Ras-dependent transformation in many cell types including NIH 3T3, BALB 3T3 and rat kidney cells (Cowley et al., 1994; Mansour et al., 1994; Dudley et al., 1995; Khosravi-Far et al., 1995), vulval development in C.Elegans and differentiation of photoreceptor cells in Drosophila (reviewed in Dickson and Hafen, 1994), promotes neurite outgrowth (Cowley et al., 1994) and protects against apoptosis induced by withdrawal of neurotrophic factors in neuronal cells (Xia et al., 1995) but

Introduction 53 induces apoptosis in Swiss 3T3 cells (Fukasawa et al., 1995). The pi 10 catalytic subunit of PI3K binds GTP-Ras which stimulates PI3K enzymatic activity (Rodriguez-Viciana et al., 1994), resulting in phosphorylation of the

D3 position of phosphoinositide lipids to produce 3-PIP, 3 ,4 -PEP2 and 3 ,4 ,5 -PIP3 which regulate a wide variety of cellular processes (reviewed in Toker and Cantley, 1997; Leevers et al., 1999). PI3K is involved reorganisation of the actin cytoskeleton. PI3K acts upstream of Rac in PDGF stimulated in PAE cells (Hawkins et al., 1995), PI3K induces Rac-dependent JNK activation in PC 12 cells resulting in neurite outgrowth (Kita et al., 1997) and in SKF5 cells, PI3K activity is required for lamellipodia formation and cell motility (Van Weering et al., 1998). PI3K activation is required for Ras-dependent transformation in NIH 3T3 cells (Rodriguez-Viciana et al., 1997) and also results in activation of the serine/threonine kinase PKB/Akt which protects against apoptosis in various cell types (reviewed in Marte and Downward, 1997; Downward, 1998).

1.5.4B Rho family effectors Many different types of proteins are effectors for the p21s, including protein kinases, lipid kinases, phosphatases and these proteins have different specificities for Rho family proteins.

1.5.4B1 Non kinase targets of Cdc42 WASP (Wiskott Aldrich syndrome protein) is a Cdc42 effector expressed in haematopoietic cells. WASP-induced Cdc42-dependent actin clustering in Jurkat and NRK cells, (Symons et al., 1996) but was unable to induce filopodia formation in fibroblasts (Miki et al., 1998). However, this may be due to the absence of the WASP interacting protein (WIP) which is required for WASP to form filopodia (Ramesh et al., 1997). As WASP is only expressed in haematopoietic cells, it cannot be responsible for Cdc42 effects on the cytoskeleton in other cell lines. A related protein N-WASP that is ubiquitously expressed and induced filopodia in fibroblasts was identified (Miki et al., 1998). The Cdc42 effector, CIP4 (Cdc42 interacting protein) is expressed in skeletal muscle, heart and placenta. Expression in Swiss 3T3 cells caused a reduction in stress fibres and coexpression with constitutively activated Cdc42-induced clusters containing both proteins (Aspenstrom, 1997).

Introduction 54 1.5.4B2 Non kinase targets of Rac Several proteins which bind GTP-Rac and are involved in mediating its downstream effects have been identified. pl40Sra-l (140kD specifically Racl associated protein) binds Rac, is present in lamellipodia, co-sediments with F actin and is implicated in Rac-induced lamellipodia formation (Kobayashi et al., 1998). PORI (Partner of Racl) binds GTP-Rac and deletion mutants of PORI inhibited V12Rac- induced membrane ruffling in REF52 cells (Van Aelst et al., 1996). PORI also interacts with the GTPase ARF6 which regulates membrane trafficking and induces cytoskeletal reorganisation at the cell surface (D'Souza-Schorey et al., 1997). POSH (plenty of SH3 domains) binds Rac and is able to stimulate JNK activity in COS1 cells although it has no kinase domain (Tapon et al., 1998) and p67phox is the component of the superoxide producing NADPH oxidase system which binds Rac (Diekmann et al., 1994). Rac also binds p35, the neuron specific regulator of dependent kinase-5 (cdk5) (Lew et al., 1994) in a GTP-dependent manner (Nikolic et al., 1998) and in Swiss 3T3 cells, p35 and V12Rac colocalise in membrane ruffles and at the cell periphery (Nikolic et al., 1998).

1.5.4B3 Kinase targets of Cdc42 and Rac The p21 activated protein kinases (PAKs) are serine/threonine kinases which are related to the yeast Ste20 protein involved in pheromone signalling (reviewed in Sells and Chernoff, 1997). Upon binding GTP Rac and Cdc42 via an 18 amino acid motif called the CRIB (Cdc42/Rac interactive binding) domain, PAKs become activated and act as Cdc42 and Rac effectors. At least three isoforms exist in mammalian cells: a, J3 and y PAK. The 6 8 kDa aPAK (PAK1) is expressed in brain and spleen (Manser et al., 1994), the 65kDa PPAK (PAK3) is expressed in brain but with a different expression pattern to that of aPAK (Manser et al., 1995) and the 62kDa yPAK (PAK2, hPak65) is ubiquitously expressed (Teo et al., 1995). PAK proteins have been isolated from Drosophila (Harden et al., 1996) and C.Elegans (Chen et al., 1996b) and related protein families have also been identified, the germinal centre kinases (GCK) and PH domain containing PAKs. PAK proteins mediate many effects including dissolution of stress fibres and reorganisation of focal complexes in HeLa cells (Manser et al., 1997; Zhao et al., 1998), activate the JNK pathway but whether this is in a Cdc42/Rac-dependent or independent manner is a point of discussion (Zhang et al., 1995; Teramoto H et al., 1996b; Westwick et al., 1997). PAK is required for Ras-mediated transformation in Rat-1 fibroblasts Introduction 55 (Tang Y et al., 1998). PAK1 also colocalises with p35/cdk5 and Rac at lamellipodial rich areas of axonal growth cones in primary (Nikolic et al., 1998). Rac- induced PAK activation was abolished via hyperphosphorylation of PAK1 by the p35/cdk5 complex both in vitro and in primary neurons, suggesting a mechanism for regulation of PAK activity, which may be involved in actin remodelling in neuronal cells. Several proteins which interact with PAK have been identified. The N terminal proline-rich region of PAK interacts with the second SH3 domain of the adaptor protein Nek (Bagrodia et al., 1995; Bokoch et al., 1996b; Galisteo et al., 1996) and also interacts with the N terminal SH3 domain of the RacGEF PIX (Manser et al., 1998). In PC 12 cells it was demonstrated that PAK can act upstream of Rac via PIX (Obermeier et al., 1998). Other kinases which bind both Cdc42 and Rac include MLK2 and 3 (Nagata et al., 1998), MEKK1 and 4 (Fanger et al., 1997b) and p70S6 kinase (Chou and Blenis, 1996). Kinases which only bind Cdc42 have also been identified. MRCKa and p are 190kDa serine/threonine kinases which bind GTP-Cdc42. The a isoform is brain- and lung-enriched and the p isoform is kidney- and lung-enriched. MRCKa modulated Cdc42-dependent morphology in HeLa cells consistent with a role as a Cdc42 effector (Leung et al., 1998). pl20Ack (activated Cdc42 binding kinase) is a non receptor tyrosine kinase which preferentially binds GTP Cdc42 and inhibits its GTPase activity (Manser et al., 1993).

1.5.4B4 Non kinase targets of Rho Several Rho targets which contain the formin homology (FH) domain have been identified. These include Drosophila diaphanous (Castrillon and Wasserman, 1994) which is required for cytokinesis and oogenesis and the mouse homologue pl40Dia, whose overexpression in COS7 cells caused accumulation of polymerised actin (Watanabe et al., 1997). Mutation of the human homologues of Drosophila diaphanous is involved in non syndromic deafness (DFNA1) and premature ovarian failure (POF). (Lynch et al., 1997; Bione et al., 1998). The FH domain contains long proline-rich motifs which probably bind cofilin and are thought to be an important link between Rho and the actin cytoskeleton (Wasserman, 1998). The myosin binding subunit of the myosin light chain phosphatase (MBS) is a target of both Rho and its effector ROK (Amano et al., 1996a; Kimura et al., 1996). Introduction 56 Binding of active Rho to MBS leads to increased phosphorylation of MLC (Kimura et al., 1996) which subsequently leads to the formation of stress fibres. The function of Rho binding proteins Rhotekin and Rhophilin are as yet unknown (Reid et al., 1996; Watanabe et al., 1996, respectively).

1.5,4B5 Kinase targets of Rho A group of related serine/threonine kinases which bind Rho have been identified. ROKa (RhoA binding kinase) was isolated from rat brain (Leung et al., 1995, 1996) and bovine brain (Matsui et al., 1996). Another isoform, ROKp was isolated from rat liver (Leung et al., 1996) and also isolated independently from human platelet cytosol (pi60 ROK) (Ishizaki et al., 1996). These Rho binding kinases contain a cysteine rich region within a PH domain in addition to their kinase domain which has homology to myotonic dystrophy kinase. ROKa and ROKp induced stress fibre formation and the assembly of focal contacts in HeLa and Swiss 3T3 cells (Leung et al., 1996; Amano et al., 1997; Ishizaki et al., 1997) and in the presence of active Rho, ROKa was translocated to the membrane (Leung et al., 1995). Rho binds both ROK and one of its substrates MBS. ROK phosphorylates MBS (the myosin binding subunit of myosin phosphatase) inactivating the phosphatase which leads to an accumulation of phosphorylated MLC (myosin light chain). Expression of Rho also increased both MBS and MLC phosphorylation (Kimura et al., 1996). ROK also phosphorylates MLC, the substrate of myosin phosphatase, at same site as MLC kinase (Amano et al., 1996a). This causes a conformational change in myosin which increases its binding to actin filaments, in turn leading to actin stress fibre formation. Thus ROK provides a link between Rho activation and the actomyosin contractility required for stress fibre formation. Most of the other ROK substrates identified so far are cytoskeletal proteins. ROK phosphorylates the actin binding protein adducin, regulating its association with actin (Kimura et al., 1998), the N terminal head-rod domain of the intermediate filament protein vimentin, resulting in the collapse of the vimentin network (Sin et al., 1998), and specifically phosphorylates Ser71 of vimentin which usually only occurs during cytokinesis (Goto et al., 1998). ROK also phosphorylates the ERM (ezrin/radixin/moesin) family of proteins which link plasma membrane proteins and actin filaments (Fukata et al., 1998; Matsui et al., 1998) and LIM kinase, resulting in its activation and the subsequent downregulation of cofilin activity, which is required to enable stress fibre and lamellipodia formation (Maekawa Introduction 57 et al., 1999). ROK mediates the Rho activation of Na/H exchange by the NHE1 (Na/H exchanger) protein through an unknown mechanism. This Na/H exchange is involved in integrin-mediated cell adhesion and spreading in a number of cell types (Tominaga et al., 1998; Tominaga and Barber, 1998). ROK is also involved in Rho-mediated neurite retraction in neuronal cells (Amano et al., 1998; Hirose et al., 1998; Katoh et al., 1998) which is discussed in detail later. Other kinase targets of Rho include the serine/threonine kinase PKN (protein kinase N) and the related kinase PRK2 (Amano et al., 1996b; Watanabe et al., 1996; Vincent and Settleman, 1997). Another is , a splice variant of Citron which contains a myotonic dystrophy related kinase domain (similar to ROK and MRCK) and is involved in cytokinesis, where in HeLa cells it colocalises with Rho in the cleavage furrow (Madaule et al., 1998).

1.5.4B6 Lipid kinase tareets of Rho, Rac and Cdc42 Rho proteins bind and activate the lipid kinases phosphatidylinositol-4- phosphate 5 kinase (PIP5K) and phosphatidylinositol-3-kinase (PI3K). PIP5K generates PI-4,5-P2, an important second messenger that binds and regulates a number of actin binding proteins (Toker, 1998). Both Rho and Rac stimulate PIP5K activity in various cell types. In homogenates of murine fibroblasts, GTP-Rho but not GTP-Rac stimulated PIP5K activity in vitro (Chong et al., 1994). Rho also bound a 68kDa type IPIP5K in vitro and in vivo endogenously expressed Rho co-immunoprecipitated with a 68kDa type IPIP5K from Swiss 3T3 cells (Ren et al., 1996). However, in permeabilised human platelets recombinant V12Rac but not V14Rho stimulated PEP5K activity in a

GTP-dependent manner and the PIP 2 produced resulted in uncapping and polymerisation of actin filaments (Hartwig et al., 1995). Recombinant GTP-Rac (not Rho) bound a type IPIP5K from Rat-1 fibroblast lysates in vitro and in vivo a type I PIP5K co-immunoprecipitated with endogenously expressed Rac in serum starved or PDGF stimulated Swiss 3T3 cells (Tolias et al., 1995). Thus PEP5K interacts in vitro with both Rho and Rac and in vivo may act either as a Rho or Rac effector in different cell types. PI3K phosphorylates the D3 position of phosphoinositide lipids to produce PI-3- P, PI-3,4-P2 and PI-3,4,5-P3 which regulate various cellular processes (reviewed in Toker and Cantley, 1997). PI3K is composed of the regulatory p85 and catalytic pi 10 subunits. The pi 10 catalytic subunit binds GTP-Ras which stimulates PI3K enzymatic

Introduction 58 activity, thus PI3K acts as a Ras effector (Rodriguez-Viciana et al., 1994). PI3K also acts as an effector of Rho, Rac and Cdc42. Both GTP Rac and Cdc42 bind the p85 subunit of PI3K and activate the kinase in vitro. The effector domain of the p21 and the RhoGAP domain of p85 are required for this interaction (Zheng et al., 1994b; Bokoch et al., 1996a). In vivo PI3K co-immunoprecipitated with Rac and to a lesser extent Cdc42 in PDGF stimulated Swiss 3T3 cells whilst a significant amount of PI3K co- immunoprecipitated with Cdc42 in unstimulated COS7 cells (Tolias et al., 1995). In Swiss 3T3 cells, LPA-induced PI3K activation was inhibited by C3 treatment, suggesting Rho mediates this activation (Kumagai et al., 1993) whilst in platelet lysates, recombinant Rho but not Rac, stimulated PI3K activity which was also inhibited by C3 treatment (Zhang et al., 1993a). Thus PI3K acts as an effector for Rho, Rac and Cdc42 both in vitro and in vivo and since PI3K is also a Ras effector (Rodriguez-Viciana et al., 1994), it may link Rho and Ras signalling pathways.

1.5.4B7 GAPs as potential effectors GAPs and p21 effectors both interact with GTP bound p21. Thus it is possible that many GAPs may also act as p21 effectors. Several examples of the effector functions of p21 GAPs have been demonstrated. In Xenopus oocytes, K12 Ras-induced germinal vesicle breakdown was inhibited by a monoclonal antibody against the SH3 domain of pl20 RasGAP (Duchesne et al., 1993). This confirms pl20 RasGAP as a Ras effector and demonstrates the role of its SH3 domain in mediating this effect. Ras- induced activation of the fos promoter was shown to be mediated by the N terminal SH2-SH3 domain containing region of pl20 RasGAP (Medema et al., 1992). This region of pl20 was also shown to mediate Ras-dependent inhibition of muscarinic atrial K+ channel currents (Martin et al., 1992). Expression of the N terminal SH2-SH3 domain containing region of pi 20 RasGAP in Rat-2 cells induced the disruption of actin stress fibres and a reduction of focal contacts (McGlade et al., 1993). These morphological effects are characteristic of a Rac or Cdc42 effector. These effects of p i20 RasGAP may be mediated by its association with p i90 RhoGAP and subsequent regulation of the Rho p21s by this complex. GAPs for heterotrimeric G proteins have also been shown to mediate effector functions. PLC |31 acts as a GAP for the a subunit of the heterotrimeric Gqn protein (Berstein et al., 1992). However, G aqn also activates PLC P1, stimulating its enzymatic activity. Thus PLC pi acts both a GAP and an effector for its physiological regulator. The y subunit of phosphodiesterase also acts as both a GAP and effector of its upstream Introduction 59 heterotrimeric G protein (Boguski and McCormick, 1993). Finally, GAPs for the Rho p21s have also been shown to act as effectors. Graf has GAP activity for Rho and Cdc42 in vitro (Hildebrand et al., 1996), but acts as a Rho effector in PC 12 cells where it enhances sphingosine-1-phosphate-induced Rho- dependent neurite retraction (Taylor et al., 1999). In Swiss 3T3 cells, microinjection of CdGAP promotes Cdc42-dependent filopodia extension and actin clustering (Lamarche- Vane and Hall, 1998). Thus CdGAP acts as both a Rac/Cdc42 GAP and a Cdc42 effector, p i22 acts as a RhoGAP in vitro and also activates PLC 5 which catalyses the hydrolysis of PIP 2, promoting cytoskeletal reorganisation (Homma and Emori, 1995). Rho activation also induces reorganisation of the actin cytoskeleton, thus pi 22 acts as both a GAP and an effector of Rho. The Rac/Cdc42 GAPs RalBPl, RLIP76 and RIP1 also act as effectors for the small GTPase Ral (Cantor et al., 1995; Jullien-Flores et al., 1995; Park and Weinberg, 1995). Microinjection of the isolated GAP domain of the RacGAP al-chimaerin inhibited Rac-induced lamellipodia in Swiss 3T3 cells, consistent with downregulation of Rac activity (Kozma et al., 1996). However, microinjection of full length al-chimaerin induced both filopodia and lamellipodia formation, which were dependent on Cdc42 and Rac activity respectively, in Swiss 3T3 fibroblasts and the growth cones of N1E 115 neuroblastoma cells (Kozma et al., 1996). Thus al-chimaerin acts as both a RacGAP and a Cdc42/Rac effector. These morphological effects of full length al-chimaerin required p21 binding but not GAP activity. Together, these examples show that GAPs can also act as effector proteins and their effector functions are often mediated by other regions within these multidomain proteins, independently of GAP activity.

1.5.5 Rho proteins and transcriptional activation Rho proteins are involved in a number of pathways leading to transcriptional activation and there may be cross talk between them.

1.5.5A Rho proteins and JNK/p38 dependent transcription Rho family proteins stimulate both JNK and p38 MAPK pathways resulting in translocation of the active MAPK to the nucleus where it phosphorylates transcription factors and stimulates transcription. Activated Rac and Cdc42 stimulated JNK activity but not ERK activity in COS7, NTH 3T3 and HeLa cells (Coso et al., 1995; Minden et al., 1995) whilst RhoA, RhoB, RhoC and Cdc42 but not Rac stimulated JNK activity in human kidney 293T cells (Teramoto et al., 1996a). Activated Racl and Rac2 were also Introduction 60 shown to stimulate p38 activity in HeLa cells (Minden et al., 1995). Proteins which regulate the activity of Rho p21s have also been shown to activate JNK. Rho family GEFs stimulated JNK activity in various cell types, Vav, Dbl and FGD1 in COS1 cells (Olson et al., 1996), Ost and Dbl in COS7 cells (Coso et al., 1995), Tiaml in COS7 and NIH 3T3 cells (Michiels et al., 1997) and Vav in COS7 cells (Crespo et al., 1996; Crespo et al., 1997). In COS7 cells, tyrosine phosphorylation of Vav was required to induce JNK activation (Crespo et al., 1997), whilst Tiaml-induced JNK stimulation was Rac-dependent in COS7 cells and membrane localisation of Tiaml was shown to be required for JNK activation in NTH 3T3 cells (Michiels et al., 1997).

1.5.5B Rho proteins and SRE dependent transcription Rho, Rac and Cdc42 regulate serum response factor (SRF)-mediated transcription. The transcription factor SRF binds the c-fos serum response element (SRE), which is a regulatory sequence present in the promoter region of many growth factor regulated genes, activating transcription. Rho is required for SRF-mediated transcriptional activation induced by LPA, serum and heterotrimeric G proteins whilst Rac and Cdc42 activate SRF in a Rho independent manner (Hill et al., 1995). Thus at least two separate pathways exist for SRF-mediated transcription and neither involves activation of MAPK pathways.

1.5.5C Rho proteins and NFkB dependent transcription

1.5.5C1 Activation of NFkB

NFk B regulates expression of a variety of genes whose products are involved in immune or inflammatory responses, growth, differentiation and development (Verma et al., 1995; Baldwin 1996). NFk B exists as a homodimer or heterodimer of 5 possible family members p50, p52, p65, c-rel and rel-B (Baeuerle and Henkel, 1994; Siebenlist et al., 1994). In its inactive form it is bound to the a, (3 or s isoform of its inhibitor protein Ik B in the cytosol (Thanos and Maniatis, 1995; Baeuerle and Baltimore, 1996). Upon activation by various factors including cytokines, growth factors, ultraviolet irradiation, phorbol esters or lipopolysaccharide, IxBa is phosphorylated on serine 32 and 36 residues (Brown et al., 1995) which targets Ik B for proteasomal degradation

(Palombella et al., 1994). This releases the NFk B dimer which translocates to the nucleus (Baeuerle and Baltimore, 1996; Baldwin, 1996).

Introduction 61 Phosphorylation of serine 32 and 36 of IkBo. is mediated by the serine specific IkB kinase (IKK). This kinase was independently isolated by three groups (Chen et al., 1996c; DiDonato et al., 1997; Regnier et al., 1997) and consists of three subunits; the catalytic a and P subunits and the regulatory y subunit (reviewed in Karin, 1999). Activation of IKK itself depends on serine phosphorylation of the DCKP subunit (Delhase et al., 1999). IKK activity is stimulated by NFkB inducible kinase (NIK), a MAPKKK which binds IKKa (Regnier et al., 1997) and also MAPK/ERK kinase kinase-1 (MEKK1), another MAPKKK which directly activates the kinase activity of IKK in vitro and induced site specific phosphorylation of IkBoc in vivo (Lee et al.,

1997). TNFa and EL-1-induced NFk B activation is mediated by the TNF receptor- associated factors TRAF2 and TRAF6 acting downstream of the TNFa and IL-1 receptors respectively. Both TRAF2 and TRAF6 interact with NIK (Malinin et al.,

1997; Song et al., 1997), resulting in NIK activation and subsequent activation of NFk B signalling (figure 1.3). MEKK1 also acts downstream of TRAF2 in TNFa-induced

NFkB activation (Lee et al., 1997) and has recently been shown to mediate Cdc42 and

Rac-induced activation of NFk B in COS7 cells (Montaner et al., 1998). In addition to its role in NFk B signalling, MEKK1 is the upstream activator of MKK4/SEK1 in the JNK signalling pathway (Yan et al., 1994) (figure 1.3). Thus MEKKl enables cross talk between NFk B and JNK signalling pathways.

NFk B is also activated by Rho family proteins (Sulciner et al., 1996; Perona et al., 1997; Montaner et al., 1998) and Ras (Devary et al., 1993; Finco and Baldwin, 1993; Koong et al., 1994; Sulciner et al., 1996; Perona et al., 1997) and activation by many ligands is inhibited by antioxidants (Schreck et al., 1991) suggesting a role for reactive oxygen species (ROS) in mediating NFk B activation. In fact, several factors such as cytokines EL-lp, TNFa and growth factors EGF, PDGF and bFGF which are known to stimulate NFk B activity also stimulate production of ROS in various cell types (Meier et al., 1989; Lo and Cruz, 1995; Sundaresan et al., 1996).

1.5.5C2 Ras and Rho proteins in NFkB activation

Ras activation of NFk B was shown in COS7, HeLa and Jurkat T cells (Sulciner et al., 1996; Perona et al., 1997). However cell type specific differences in NFk B activation pathways exist since V12Ras-induced activation was Rho, Rac and Cdc42 independent in COS7 cells (Perona et al., 1997; Montaner et al., 1998) but Rac-

Introduction 62 TNF 1L-1 PMA I I I TNFR1 IL-1RI- cell membrane I \ \ TRADl) IRAK

H \ I I . PKC TRAF2 TRAF6 Jk NIK V1EKK1 / — i I MAIMAP kinase \ IKK SEK1 / pp*Mlrsk i IkB phosphorylation SAPK/JNK I Ik-R degradationi

c-Jun activation NF-kB trarvsiocation

Figure 1.3a: Activation of NFkB signalling pathways (reproduced from Stancovski and Baltimore, 1997)

TNFa X X / Ras Rho A Cdc 42 Rac 1 I t Raf MEKK1 TAK1 ,TAO / * t MEK1, MEK2 SEK1/MKK4 MKK3/6 I licBa } 1 ERK1,ERK2 kinase JNK/SAPK p38/Mpk2

TCF NF-k B C-Jun ATF2 ATF2 TCF t c f c r e b

Figure 1.3b: The involvement of Rho family proteins in the activation of NFkB signalling pathways (reproduced from Montaner et al., 1998)

Figure 1.3 63 dependent in HeLa cells (Sulciner et al., 1996). Ras was also shown to be involved in ligand-induced NFk B activation; both EGF-induced activation in COS7 cells and UVC- induced activation in NIH 3T3 cells were Ras-dependent (Perona et al., 1997).

Wild type and constitutively activated Rho, Rac and Cdc42 stimulated N F k B activity in COS7 and Jurkat T cells (Perona et al., 1997) whilst V12Rac stimulated

N F k B activity in HeLa cells (Sulciner et al., 1996). Cdc42 and Rac-induced N F k B activation in COS7 cells was shown to depend on the JNK activator MEKK1 acting downstream, providing a link between JNK and N F k B signalling pathways in these cells (Montaner et al., 1998) (figure 1.3). Rho proteins were also shown to be involved in ligand-induced N F k B activation, TNFa-induced activation was Rho and Cdc42- dependent in COS7 cells (Perona et al., 1997) whilst IL-ip-induced activation was Rac- dependent in HeLa cells (Sulciner et al., 1996). In COS7 cells, N F k B activation induced by Rho proteins was unaffected by dominant negative Ras or Raf suggesting Rho proteins either act independently or downstream from Ras and Raf (Perona et al., 1997).

However it was also shown that V12Ras-induced N F k B activation was unaffected by dominant negative Rho proteins (Montaner et al., 1998), together these results suggest that Ras and Rho proteins act in separate N F k B activation pathways in COS7 cells. The

RhoGEFs Vav, Dbl and Ost were also shown to stimulate N F k B activity in a Rac, Cdc42 and Rho/Cdc42-dependent manner, respectively (Montaner et al., 1998).

1.5.5C3 Generation of reactive oxygen species in phagocytic and non phagocytic cells In phagocytic cells, a specialised multi component membrane NADPH oxidase is responsible for production of ROS to destroy invading micro-organisms (Bokoch, 1994). In non-phagocytic cell types, NADPH oxidase components have also been detected (Jones et al., 1994) and inhibition of ROS production by flavoprotein inhibitor diphenyleneiodonium (DPI) (Griendling et al., 1994; Lo and Cruz, 1995; Sundaresan et al., 1996) suggests that an NADPH oxidase related protein may be involved in generating ROS.

In phagocytic cells, cytochrome bs 58, p47phox, p67phox and GTP bound Racl or Rac2 isoforms are required for an active oxidase complex in guinea pig macrophages (Abo et al., 1991) and human neutrophils, respectively (Knaus et al., 1991). Thus Rac and the regulation of its nucleotide state are important in ROS production in phagocytic cells and also in non phagocytic cells although an NADPH oxidase related protein has

Introduction 64 not yet been identified.

1.5.5C4 Ras/Rac and reactive oxygen species production Ras and Rac induce ROS production in various cell types. ROS production in NTH 3T3 cells transiently and permanently expressing V12Ras was shown to be Rac- dependent and was implicated in Ras-induced cell cycle progression (Sundaresan et al., 1996; Irani et al., 1997). Rac-induced ROS production has been demonstrated in NIH 3T3 (Sundaresan et al., 1996), COS1 and REF52 cells. In the latter, the Racl insert region was shown to be essential for ROS production (Joneson and Bar Sagi, 1998).

ROS production was shown to be required for Rac-induced NFk B activation in HeLa cells (Sulciner et al., 1996). Rac is also involved in growth factor and cytokine-induced ROS production. PDGF, EGF, TNFa and IL-lp stimulation ofNIH 3T3 cells (Sundaresan et al., 1996) and EL-1 (3 stimulation of HeLa cells (Sulciner et al., 1996) induced ROS production was Rac-dependent.

1.5.5C5 The role of reactive oxygen species Ligand stimulation in various cell types results in production of ROS. In rat vascular smooth muscle cells (VSMCs), inhibition of PDGF-induced ROS production by antioxidants also inhibited PDGF-induced tyrosine phosphorylation, ERK stimulation, DNA synthesis and chemotaxis (Sundaresan et al., 1995). Similarly, EGF- induced tyrosine phosphorylation in A431 cells was inhibited by the use of antioxidants (Bae et al., 1997). As well as roles in tyrosine phosphorylation, ERK stimulation, DNA synthesis and chemotaxis, ROS also induce transcriptional activation. In rabbit epithelial cells, arachidonic acid-induced INK stimulation was inhibited by antioxidants whilst hydrogen peroxide activated INK in a dose dependent manner (Cui et al., 1997), whilst TNFa and bFGF-induced ROS production in chondrocytes led to stimulation of c-fos- dependent transcription in these cells (Lo and Cruz, 1995). There is conflicting data regarding the role of ROS in apoptosis, but ROS may mediate p53-dependent apoptosis (Johnson et al., 1996; Polyak et al., 1997). ROS have been implicated in many diseases including Alzheimers, hypertension and atherogenesis while antioxidants have a protective role in cardiovascular disease (Diaz et al., 1997).

Introduction 65 1.5.6 Rho proteins and cell morphology

1.5.6A The role of Rho proteins in fibroblast cell morphology Rho, Rac and Cdc42 were found to regulate the formation of actin structures in fibroblasts. Microinjection of Rho-induced the formation of stress fibres (Ridley and Hall, 1992), Rac-induced lamellipodia and membrane ruffling (Ridley et al., 1992) and Cdc42-induced filopodia formation (Kozma et al., 1995; Nobes and Hall, 1995) in serum starved Swiss 3T3 cells. Activated Rho-induced the formation of stress fibres and large integrin based focal adhesion complexes at the ends of stress fibres in serum starved Swiss 3T3 cells (Ridley and Hall, 1992). LPA or serum treatment both induce stress fibres via Rho activation, whilst the bacterial C3 toxin specifically inactivates Rho via ADP ribosylation (Ridley and Hall, 1992). Rho stimulates actomyosin contractility via its effector ROK, generating tension across the cell and the bundling of stress fibres which induces integrin clustering and formation of focal adhesions (Chrzanowska-Wodnicka andBurridge, 1996). Microinjection of Cdc42 in serum starved Swiss 3T3 cells rapidly induced filopodia formation, later followed by lamellipodia and ruffling (Kozma et al., 1995). Activated Rac-induced actin accumulation at the cell periphery in lamellipodia and membrane ruffles, followed later by weak induction of stress fibres formation, suggesting that Rac activates Rho (Ridley et al., 1992). Rac-dependent actin polymerisation was induced by treatment with PMA or PDGF and by Ras activation (Ridley et al., 1992). Bradykinin treatment produced similar effects to Cdc42 microinjection and filopodia formation was inhibited by pre-injection with dominant negative Cdc42, indicating that bradykinin acts upstream of Cdc42. Subsequent ruffling was inhibited by pre-injection with dominant negative Rac (Kozma et al., 1995). Coinjection of wild type Cdc42 with dominant negative Cdc42 or Rac inhibited filopodia or lamellipodia formation respectively (Kozma et al., 1995). Together this data suggests that Cdc42 induces filopodia formation and subsequently activates Rac to produce lamellipodia, which then weakly activates Rho to produce stress fibres, although this last link is a point of some discussion. Both Cdc42 and Rac induce assembly of integrin based focal complexes at the plasma membrane in Swiss 3T3 cells. The Rac/Cdc42-induced focal complexes are morphologically distinct from the larger Rho-induced focal adhesions and while both complexes contain integrins, vinculin, paxillin and focal adhesion kinase (FAK) and

Introduction 66 have increased levels of phosphotyrosine (Ridley and Hall, 1992; Nobes and Hall, 1995), they may also contain other differing components. These adhesion complexes act as points of cellular attachment to the extracellular matrix and are targets for cell signalling molecules (reviewed in Schoenwaelder and Burridge, 1999).

1.5.6B Neuronal cell morphology Neurite extension is a complex process which requires reorganisation of the actin cytoskeleton ( Chien et al., 1993; Bentley and O'Connor, 1994; Stirling and Dunlop, 1995) and directed growth requires the constant interpretation of multiple signals from the extracellular matrix, neighbouring cells and gradients of chemoattractant and chemorepellant substances (Keynes and Cook, 1995). In the developing nervous system, a specialised structure at the distal end of a developing neurite, the , is responsible for pathfinding (Bray and Chapman, 1985; Heidermann and Buxbaum, 1991; Bentley and O'Connor, 1994; Aletta and Greene, 1998). A growth cone extends multiple sensory filopodia (Davenport et al., 1993; Kater and Rehder, 1995), which are stabilised via lamellipodial membrane extension (Tanaka et al., 1995; Aletta and Greene, 1998) or retracted in response to the interpreted signals to produce directed growth. The sensory role of filopodia in growth cone guidance was demonstrated in grasshopper neurons where inhibition of filopodia formation via cytochalasin B treatment resulted in disoriented pathfinding (Bentley and Toroian- Raymond, 1986). The Rho p21s were originally shown to be involved in actin cytoskeleton reorganisation in fibroblasts, but their effects in neuronal cell types and systems has also been investigated (reviewed in Luo et al., 1997; Gallo and Letoumeau, 1998).

1.5.6B1 The role of Rho proteins in N1E 115 cell morphology In the mouse N1E 115 neuroblastoma cell line, thrombin and LPA acting through their respective heterotrimeric G protein coupled receptor, induced neurite retraction which was inhibited by treatment with C3 exoenzyme, a bacterial toxin which ADP ribosylates and specifically inactivates Rho (Jalink and Moolenaar, 1992; Jalink et al., 1994). C3 treatment itself induced cell flattening followed by neurite outgrowth in undifferentiated N1E 115 cells (Jalink et al., 1994; Kozma et al., 1997). This led to investigation of the role of Rho proteins in actin cytoskeleton reorganisation in N1E 115 cells. In N1E 115 cells, wild type or activated RhoA-induced growth cone collapse

Introduction 67 and cell rounding (Kozma et al., 1997) and neurite retraction (Gebbink et al., 1997; Hirose et al., 1998) conversely microinjection of C3 or dominant negative Rho-induced filopodia and lamellipodia formation (Kozma et al., 1997). Microinjection of wild type or activated Rac-induced lamellipodia formation and wild type or activated Cdc42- induced both filopodia and lamellipodia formation in N1E 115 cells (Kozma et al., 1997). Coinjection of dominant negative Rac with Cdc42 inhibited lamellipodia but not filopodia formation, suggesting that Cdc42 activates Rac to induce lamellipodia in these cells. Both C3 and serum starvation-induced neurite outgrowth were inhibited by dominant negative Rac and Cdc42. Coinjection of C3 and N17Cdc42 inhibited both filopodia and lamellipodia formation whilst C3 and N17Rac coinjection inhibited lamellipodia but not filopodia formation, thus Cdc42 acts upstream of Rac in C3- induced neurite outgrowth pathway. Coinjection of Rho and Cdc42 had no net effects on morphology. This demonstrates that there is competition between Rho and Rac pathways such that Rho-induced neurite retraction and Cdc42/Rac-induced neurite outgrowth pathways act antagonistically in N1E 115 cells to determine neuronal morphology. A screen to identify extrinsic or growth factors which mimicked the morphological effects of Rho p21s inN IE 115 cells, identified acetylcholine. InN IE 115 cells, application of an acetylcholine gradient via a micropipette induced filopodia and lamellipodia formation and also protected against LPA-induced neurite retraction (Kozma et al., 1997). Dominant negative Cdc42 inhibited filopodia but not lamellipodia formation whilst dominant negative Rac inhibited lamellipodia but not filopodia formation induced by acetylcholine, suggesting these p21s are independently activated by acetylcholine and act in separate pathways. This response is contrary to results in fibroblasts where no Rac-dependent lamellipodia were formed in the presence of dominant negative Cdc42, suggesting Cdc42 acts upstream of Rac in fibroblasts (Kozma et al., 1995) and illustrating that cell type specific differences exist in p21 regulation. However the independent action of Cdc42 and Rac in the acetylcholine response is contrary to C3-induced neurite outgrowth in N1E 115 cells, where Cdc42 acts upstream of Rac. Thus even in the same cell type different regulatory pathways exist for responses to different stimuli. Proteins which interact with Rho p21s have also been shown to be involved in neuronal morphology. Rho-induced neurite retraction is mediated by its effector ROKp (pl60ROK) in N1E 115 cells (Hirose et al., 1998). Overexpression of ROKp induced neurite retraction similar to that observed with V14Rho and LPA, whilst LPA-induced Introduction 68 neurite retraction was inhibited by treatment with the ROK inhibitor Y-27632 (Hirose et al., 1998). LPA-induced retraction was previously shown to be inhibited by KT5926, a MLC kinase inhibitor (Jalink et al., 1994) and the peak in MLC phosphorylation observed upon LPA-induced neurite retraction was inhibited in a dose dependent manner by the ROK inhibitor Y-27632 (Hirose et al., 1998), suggesting that actomyosin contractility is involved in ROKP-induced neurite retraction. Inhibition of endogenously expressed ROKP activity via a dominant negative ROKP mutant (kinase and Rho binding deficient) induced neurite outgrowth which could be inhibited by dominant negative Rac or Cdc42 (Hirose et al., 1998). This supports other data suggesting antagonism exists between Rho and Cdc42/Rac pathways in N1E 115 cells (Kozma et al., 1997; Van Leeuwen et al., 1997). Other endogenously expressed proteins in N1E 115 cells which interact with Rho p21s have also been shown to be involved in neuronal morphology. An 116kDa protein which binds both GDP and GTP forms of Rho-induced neurite outgrowth, similar to dominant negative Rho or C3 treatment, whilst a 190kDa RhoGEF induced neurite retraction like Rho itself (Gebbink et al., 1997). The RacGEF Tiaml induced cell spreading in the presence of serum when plated on plastic or fibronectin, but also induced cell polarisation, neurite outgrowth and extreme neurite branching when plated on laminin (Van Leeuwen et al., 1997), illustrating the importance of cell-substrate interactions in determining cell morphology. Tiaml expressing cells plated on laminin no longer responded to LPA-induced retraction and cell rounding mediated by Rho/ROK, but this was overcome by coexpression of V14Rho, suggesting that Rac activation by Tiaml antagonised Rho signalling in these cells. This supports other data suggesting antagonism exists between Rho and Cdc42/Rac pathways in N1E 115 cells (Kozma et al., 1997; Hirose et al., 1998). Inhibition of Rho activity by dominant negative Rho or p i90 RhoGAP induced further neurite outgrowth in Tiaml expressing cells. Thus proteins which regulate Rho p21 activity, such as GEFs and GAPs, as well as Rho p21s and their effectors are all involved in the control of neuronal morphology in N1E 115 cells.

1.5.6B2 The role of Rho proteins in PC12 cell morphology As with N1E 115 cells, in the rat pheochromocytoma PC 12 cell line LPA- induced neurite retraction is Rho-dependent (Tigyi et al., 1996) and is inhibited by the action of C3 exoenzyme (Jalink et al., 1994) whilst inhibition of Rho activity induced neurite outgrowth (Nishiki et al., 1990). InNGF differentiated PC12 cells, Introduction 69 microinjection of the catalytic domain of ROKa induced neurite retraction similar to V14Rho (Katoh et al., 1998). Pre-injection with C3 had no effect on this activity thus ROK mediates Rho-dependent neurite retraction in these cells, however microinjection of kinase dead ROK did not induce retraction, thus the kinase activity of ROKa is required for this activity. NGF-induced neurite outgrowth in PC 12 cells was shown to require PI3K activity (Kimura et al., 1994; Kobayashi M et al., 1997) and PI3K-induced outgrowth was later shown to depend on Rac-induced JNK activation (Kita et al., 1998). Inhibition of Rac activity in PC 12 cells inhibited growth cone-mediated neurite outgrowth in response to NGF, whilst tension-induced outgrowth was unaffected (Lamoureux et al., 1997). Thus Rac is involved in tension generation within the growth cone, probably via attachment to the substratum but is not involved in the reorganisation of the cytoskeleton required for neurite extension. Inhibition of Cdc42 activity, like Rac inhibition, was also shown to inhibit NGF- induced neurite outgrowth in PC 12 cells, whilst membrane targeting of PAK was shown to induce neurite outgrowth similar to the effect of NGF treatment (Daniels et al., 1998). The kinase activity of PAK was not required to induce neurite outgrowth but the p21 binding domain, proline-rich and acidic regions were required. Inhibition of endogenous PAK activity using a construct containing these three domains resulted in inhibition of NGF-induced neurite outgrowth, suggesting a role for PAK in this process. Dominant negative Cdc42 or Rac had no effect on PAK-induced neurite outgrowth suggesting that PAK may act downstream from the p21s or possibly in a separate pathway (Daniels et al., 1998). Thus Rho p21s and their effectors are involved in the regulation of PC 12 cell morphology and it seems likely that proteins which regulate p21 activity such as GAPs and GEFs are also likely to be involved in the this process.

1.5.6B3 The role of Rho proteins in neural systems In the sensory neurons of Drosophila, expression of dominant positive or negative Rac inhibited axonal but not dendritic outgrowth whilst dominant positive or negative Cdc42 had a more general effect on neuronal migration, axonal and dendritic outgrowth (Luo et al., 1994). More data to support the role of Rac activity in rather than dendrites was found in the cerebellar Purkinje cells of transgenic mice expressing constitutively activate human Racl. There was a severe loss of presynaptic terminals and dendritic spines were decreased in size but increased in number (Luo et Introduction 70 al., 1996). In the C.Elegans hermaphrodite specific neuron (HSN), expression of activated Mig-2, a Rho family member mostly related to Cdc42 and Rac, resulted in abnormally long axons with disoriented pathfinding ability suggesting that overexpression of Mig-2 interferes with the ability of these cells to interpret normal guidance cues (Zipkin et al., 1997). In primary cultures of cortical neurons, expression of activated Rho, Rac or Cdc42 increased dendritic growth whereas expression of dominant negative p21s reduced dendrite formation (Threadgill et al., 1997). The effect of Rac in this system contrasts with data from Drosophila where both dominant positive and negative Rac inhibited axonal but not dendritic outgrowth. In embryonic chick dorsal root ganglion (DRG) cells, C3 stimulated axonal outgrowth, similar to data from PC 12 and N1E 115 cells, whilst Rac-induced growth cone collapse (Jin and Strittmatter, 1997) contrary to the neurite outgrowth and lamellipodia formation observed in PC 12 and N1E 115 cells, respectively. Collapsin, but not LPA or myelin-induced growth cone collapse was also shown to be a Rac-dependent process in chick DRGs (Jin and Strittmatter, 1997). Thus Rho p21s are involved in regulating neuronal morphology in many cell types and systems. The roles of these proteins are established within the neuronal N1E 115 and PC 12 tissue culture cell lines, whereas in vivo their effects vary considerably between systems and may be more complex than their observed effects in vitro.

1.5.6C Rho p21s and their morphological effects in other cell types In addition to their well characterised roles in actin cytoskeleton reorganisation in fibroblasts and neuronal cells the Rho p21s are involved in actin remodelling in several other cell types. In the budding yeast S.Cerevisiae, Rho proteins are involved in budding and growth regulation via reorganisation of the actin cytoskeleton (reviewed in Tanaka and Takai, 1998). Recent work has focused on the effects of Rho p21s in macrophages, where microinjection of dominant positive Rho-induced actin cable assembly and cell contraction, Rac-induced lamellipodia and ruffles and Cdc42-induced filopodia (Allen et al., 1997), similar to data in fibroblast and neuronal cells. Rac and Cdc42 were also required for assembly of focal complexes containing pi integrin, FAK, paxillin, vinculin and tyrosine phosphorylated proteins (Allen et al., 1997). The effects of the Rho p21s on colony stimulating factor (CSF)-induced migration and chemotaxis, which requires extensive actin remodelling were also investigated. Inhibition of Rho or Rac

Introduction 71 activity, inhibited CSF-induced cell migration whilst inhibition of Cdc42 inhibited chemotaxis in response to CSF, although cells were still motile (Allen et al., 1998). Thus Rho and Rac are essential for macrophage cell motility whilst Cdc42 is essential for co-ordinated motion towards a CSF gradient, but is not required for locomotion. Type I and II phagocytosis by macrophages differ in their mechanism of uptake but both require extensive actin remodelling. It was found that type I phagocytosis, mediated by the immunoglobulin receptor was Rac and Cdc42-dependent whilst type II phagocytosis, mediated by the complement receptor was Rho-dependent (Caron and Hall, 1998). Thus the Rho p21s are essential regulators of actin reorganisation in macrophages.

1.5.7 Rho proteins and cell cycle regulation Microinjection of activated Rho, Rac or Cdc42-induced cell cycle progression through G1 and subsequent DNA synthesis in quiescent Swiss 3T3 cells (Olson et al., 1995). Use of Y40C effector loop mutants demonstrated that Rac and Cdc42-induced cell cycle progression was independent of PAK and JNK activation, although these Rac and Cdc42 mutants still retained their ability to induce lamellipodia and filopodia respectively (Lamarche et al., 1996).

1.5.8 Rho proteins and transformation Rho, Rac and Cdc42 are involved in Ras-induced transformation (reviewed in Zohn et al., 1998). Expression of activated Rac in Rat-1 fibroblasts induced malignant transformation (Qui et al., 1995) whilst expression of constitutively activated Rho or Rac in NIH 3T3 fibroblasts only weakly induced focus formation but induced tumours in nude mice (Khosravi-Far et al., 1995). Inhibition of the ERK pathway or expression of dominant negative Rho or Rac inhibited Ras-induced focus formation in NIH 3T3 cells, however activated Rho and Rac synergised with Raf enhancing focus formation (Khosravi-Far et al., 1995; Qui et al., 1995) and also synergised with Ras increasing focus formation in NIH 3T3 cells, producing a more transformed cell morphology and increasing cell motility (Khosravi-Far et al., 1995). Thus the Rho, Rac and ERK pathways are essential for Ras-induced transformation and both Rho and Rac pathways synergise with the ERK pathway to induce transformation. Expression of activated Cdc42 in Rat-1 fibroblasts induced focus formation in soft agar and tumours in nude mice, but did not induce serum independent growth or overcome contact inhibition of cells grown under adherent conditions (Qui et al., 1997). Thus Cdc42 caused only

Introduction 72 partial transformation in these cells. In NIH 3T3 cells, activated Cdc42 inhibits growth but co-operates with activated Raf to induce focus formation (Whitehead et al., 1998) and thus contributes to Ras-dependent transformation.

1.6 The chimaerin family of RacGAPs The chimaerin family of RacGAPs consist of al-, a2-, {31- and (32-chimaerin. Each a and p isoform is an alternate splice variant from the a- and p-chimaerin genes respectively. The isoforms share a common cysteine rich domain and a C terminal RhoGAP domain, whilst their N terminal sequences diverge (figure 1.4). Each protein is expressed in a tissue specific manner and is developmentally regulated. The conserved RhoGAP domain is found in many other proteins which have varying specificities for Rho p21s, as previously described. The RhoGAP and cysteine rich domain combination is also found in the Drosophila rotund RacGAP, a testis specific protein whose deletion leads to male sterility (Agnel et al., 1992), MgcRacGAP, the human homologue of rotund (Toure et al., 1998 ), the RhoGAP PARG1 (Saras et al., 1997) and the unconventional ; human myosin IXb (Wirth et al., 1996) and rat myr5 (Muller et al., 1997).

1.6.1 al-Chimaerin a 1-Chimaerin was isolated from human brain and retinal cDNA library screens (Hall et al., 1990). The cDNA hybridised to a 2.2kb mRNA in human brain and a 2.3kb mRNA in rat brain but was not detected in rat kidney, heart, spleen, muscle, adrenal or testis tissue. In situ hybridisation showed that al-chimaerin mRNA was expressed at highest levels in the pyriform cortex, cortical neurons, hippocampal pyramidal neurons, Purkinje neurons of the cerebellum and granule cells of the dentate gyrus (Hall et al., 1990; Lim et al., 1992). Expression of al-chimaerin mRNA is also developmentally regulated, with low levels detected in rat brain from embryonic day 15 until birth, when expression increased to a maximum at postnatal day 20, a period which coincides with cellular differentiation and synaptogenesis (Lim et al., 1992). Rat al-chimaerin protein is approximately 38kDa in size and contains a unique N terminal 35 amino acid sequence, not present in a2-chimaerin, which may form an amphipathic helix with membrane binding potential (Lim et al., 1992).

Introduction 73 AH CRD GAP al chimaerin, 38 kDa

SH2 CRD a2 chimaerin, 45 kDa

CRD GAP pi chimaerin, 30 kDa

SH2 CRD GAP p2 chimaerin, 46 kDa

Figure 1.4: Protein domain structure of the chimaerin RacGAPs The abbreviations used are; SH2, Src homology 2 domain; CRD, cysteine rich/phorbol ester binding domain; GAP, GTPase activating domain; AH, contains region predicted to form an amphipathic helix.

Figure 1.4 74 1.6.2 a2-Chimaerin a2-Chimaerin is a splice variant containing an N terminal SH2 domain which was isolated from human retinal and hippocampal cDNA libraries (Hall et al., 1993). The cDNA hybridised to a 2.2kb mRNA in rat brain and testis but was not detected in rat kidney, spleen, muscle or liver tissue. In situ hybridisation showed that a2-chimaerin mRNA was expressed at highest levels in the early pachytene spermatocytes of rat testis but was not detected in spermatids, whilst in adult rat brain the highest levels were detected in the cortex, pyriform cortex and hippocampus, similar to al-chimaerin. a2- Chimaerin mRNA was detected at highest levels in embryonic rat forebrain but there was no developmental increase in expression, unlike al-chimaerin. The a2-chimaerin SH2 domain has 38% identity with the SH2 domains of Src, Abl and RasGAP and 30% identity with the N terminal SH2 domain of the p85a subunit of PI3K. However unusually, a2-chimaerin has glutamate instead of tryptophan as the first residue of its SH2 domain. The human chimaerin gene was mapped to chromosome 2q3 l-2q32.1 and analysis of the intron/exon boundaries showed that a l ­ and a2-chimaerin are alternate splice products from the same gene. It was also found that the position of the chimaerin splice sites was conserved with those of the Bcr and PKC genes.

1.6.3 B-Chimaerins A novel overlay assay was developed to detect Rho family GAPs (Manser et al., 1992) which led to the isolation of pi-chimaerin from a rat testis cDNA library (Leung et al., 1993). The 30kDa RacGAP pi-chimaerin has 68% overall sequence identity with al-chimaerin, with 93% and 77% homology in the cysteine rich and GAP domains respectively, pi-Chimaerin is only expressed in the testis in spermatids undergoing acrosomal transformation, considerably later than a2-chimaerin expression and pi expression is developmentally regulated, being detected in rat testis only after 30 days. Using the GAP overlay assay, another ~46kDa RacGAP was detected in adult rat cerebellar extracts which cross reacted with an antibody raised against the C terminal of pi-chimaerin (Leung et al., 1994). The P2-chimaerin cDNA was isolated from a human cerebellar cDNA library and contained cysteine rich and GAP domains identical to pi-chimaerin and an additional N terminal SH2 domain with 82% identity to that of a2-chimaerin, thus suggesting that pi- and p2-chimaerin are derived by alternate splicing from the same gene, similar to a l - and a2-chimaerin (Leung et al., 1994). Introduction 75 J32-Chimaerin is expressed mainly in the granule cells of the cerebellum and its expression is developmentally regulated with expression increasing postnatally from day 20. In both a2- and P2-chimaerin SH2 domains, the first residue is glutamate instead of the usually invariant tryptophan and this unusual feature may define a new subclass of SH2 domains. Substitution of tryptophan with glutamate as the first residue in the Src SH2 domain significantly reduced binding to its phosphopeptide target sequences (Bibbins et al., 1993).

1.6.4 Cysteine rich domain of chimaerin The CRD of chimaerin has 48% homology to the C lb cysteine rich region of PKC (Parker et al., 1986; Hall et al., 1990) which binds diacylglycerol (DAG) and phorbol esters in a phospholipid-dependent manner and regulates PKC activity (Ohno et al., 1988; Ono et al., 1989). This CX2CX13CX2CX7CX7C cysteine rich sequence is also found in DAG kinase and A-Raf-1 and the CX2CX13CX2C part of this sequence may potentially form a zinc finger structure (Freedman et al., 1988) in these proteins. The cysteine rich domains of a l- and a2-chimaerin bind similar levels of phorbol ester in the presence of lOOpg/ml PS and act as phorbol ester receptors, with similar binding affinities to PKC (Ahmed et al., 1990; Ahmed et al., 1991; M. Teo PhD thesis, 1994). No phorbol ester binding to the al-chimaerin CRD was observed in the presence of PI, PC or PE (Ahmed et al., 1990) or to the a2-chimaerin CRD in the presence of PC or arachidonic acid (M. Teo PhD thesis, 1994). However, PA, PI-4-P or PI-4,5-P2 also enabled similar levels of phorbol ester binding to the a2-chimaerin CRD as PS, whilst PI enabled 2-3 fold higher binding (M. Teo PhD thesis, 1994). Thus although the sequence of the a l- and a2-chimaerin CRD is identical, PI has opposite effects on their phorbol ester binding suggesting that perhaps the divergent N terminal regions of these proteins are somehow involved in this effect. However these studies used recombinant proteins which may differ in their relative activities from native protein. The CRD of P2-chimaerin also acts as a high affinity receptor for both phorbol esters and DAG (Caloca et al., 1997; Caloca et al., 1999). It was recently shown that upon binding phorbol ester or DAG via its CRD domain, p2-chimaerin translocates from the soluble to the particulate fraction of cells, which corresponded to translocation from the cytosol to a perinuclear region (Caloca et al., 1999). Phospholipid binding to the CRD also affects the RacGAP activity of a l - and a2-chimaerin, as discussed in the next section.

Introduction 76 1.6.5 Re 2ulation of a l- and a2-chimaerin GAP activity Recombinant al-chimaerin has 200 fold higher GAP activity for Rac than Cdc42 (Manser et al., 1992) and deletion of the cysteine rich domain from al-chimaerin results in higher RacGAP activity than observed with full length recombinant protein, suggesting that the CRD may regulate GAP activity (Ahmed et al., 1993). In fact, a l- chimaerin GAP activity is modulated both positively and negatively by phospholipids. Phosphatidic acid (PA) and phosphatidylserine (PS) stimulate the RacGAP activity of full length al-chimaerin via interaction with its cysteine rich domain and phorbol esters can synergise with low concentrations of PS/PA to enhance this GAP activity. Inhibition of al-chimaerin RacGAP activity by phospholipids LPA, PI, PI-4-P and PI-

4, 5 -P2 also requires the presence of the CRD whilst arachidonic acid-induced inhibition is independent of the CRD. PKC activity is regulated by phospholipids and phorbol esters in a CRD-dependent manner (Ono et al., 1988; Ono et al., 1989) and it appears that al-chimaerin GAP activity is similarly regulated. Unlike al-chimaerin, full length a2-chimaerin and the isolated GAP domain have similar RacGAP activity which saturates at 0. lpM protein a level ten fold higher than that of full length al-chimaerin (M. Teo PhD thesis, 1994). However, despite this a2-chimaerin RacGAP activity is also modulated by phospholipids. PS and PA stimulated a2-chimaerin RacGAP activity, similar to al-chimaerin (Ahmed et al., 1993), however at the same concentration of phospholipid, the degree of stimulation observed was much higher with a2- than al-chimaerin (M. Teo PhD thesis, 1994). Also, over a range of PS or PA concentrations up to lOOfjg/ml, stimulation of a2- chimaerin RacGAP activity became saturated at 10|ig/ml and 50(j,g/ml respectively, whereas al-chimaerin RacGAP activity was stimulated in a concentration dependent manner without reaching saturation (M. Teo PhD thesis, 1994). The PS-induced stimulation of a2-chimaerin RacGAP activity was observed with both recombinant and rat brain purified protein and similar to al-chimaerin, phorbol esters synergised with low levels of PS to further stimulate recombinant a2-chimaerin RacGAP activity (Hall et al., 1993; M. Teo PhD thesis, 1994). a2-Chimaerin RacGAP activity was also stimulated by lOOpg/ml PI, PI-4-P and PI-4,5-P2, which in contrast inhibited al-chimaerin RacGAP activity and slightly stimulated by 100|ug/ml arachidonic acid which had no effect on al-chimaerin RacGAP activity (Ahmed et al., 1993; M. Teo PhD thesis, 1994). Thus although a l- and a2-

Introduction 77 chimaerin are both stimulated by PS and PA, the extent and kinetics of this activation in vitro are very different, whilst phosphoinositide lipids have opposing effects on recombinant a l and a2-chimaerin RacGAP activity. Since these two proteins only differ in sequence at their N terminal, it suggests that this region is responsible for their observed differences in RacGAP activity in vitro.

1.6.6 a2-Chimaerin target proteins Two potential targets of a2-chimaerin have previously been identified. A yeast two hybrid screen isolated a 13kDa protein as a target of the a2-chimaerin SH2 domain which was identified as the B13 subunit of the inner mitochondrial membrane NADH ubiquinone , the first and largest of the mitochondrial respiratory chain (C.Monfries, Personal Communication). A screen of rat brain extracts for a2-chimaerin targets identified a ~65kDa protein which was partly purified by chromatographic techniques and its enrichment monitored during purification by overlay binding assay (M. Teo PhD thesis, 1994). The protein was identified on the basis of peptide sequence as TOAD-64, a phosphoprotein involved in axonal guidance (Mintum et al., 1995a, b). TOAD-64 was identified as one of several proteins whose expression was upregulated during neurogenesis (Minturn et al., 1995a). It is expressed at highest levels during late embryonic and early postnatal periods in rat brain cortex and expressed at low levels in the adult (Minturn et al., 1995a, b). This coincides with a2-chimaerin expression (Hall et al., 1993) and neuronal maturation. TOAD-64 has homology to the C.Elegans unc-33 protein, mutation of which results in abnormal axonal outgrowth and guidance (Li et al., 1992). TOAD-64 expression is induced during axon regeneration in rats and upon NGF-induced differentiation in PC 12 cells suggesting an essential role in axonal outgrowth or navigation (Minturn et al., 1995b). In primary cultures of rat DRG neurons, TOAD-64 is expressed throughout the cell body, neurites and growth cones and is present in filopodial and lamellipodial extensions from the growth cone itself (Minturn et al., 1995b). Extraction of TOAD-64 from rat cortex homogenates demonstrated that a proportion of protein was associated with the membrane (Minturn et al., 1995b) which is supported by its presence in filopodial and lamellipodial extensions. Since no transmembrane domain is present in the TOAD-64 sequence, this suggests it tightly associates with a membrane protein. The regulated neuronal expression of TOAD-64 and its combined cytosolic and membrane associated distribution within neurons means TOAD-64 is well situated to play an essential role in Introduction 78 the regulation of neurite outgrowth. TOAD-64 is one member of a family of proteins with homology to the C.Elegans unc-33 protein. Four related proteins have been identified and have been called different names in different species; collapsin response mediator proteins (CRMPs), unc-33 like phosphoproteins (Ulips) or dihydropyrimidinase (DHPase) related proteins (DRPs), due to their homology with DHPases, enzymes which hydrolyse pyrimidine rings. Despite their homology to DHPases neither rat nor human CRMPs possess DHPase enzymatic activity in vitro (Hamajima et al., 1996; Wang and Strittmatter, 1996). However, similar to DHPases, CRMP proteins oligomerise in multiple possible combinations to form tetramers (Wang and Strittmatter, 1997), the composition of which may regulate substrate specificity. Four CRMP proteins (1-4) have been identified in rat which are all differentially expressed in neural tissues (Wang and Strittmatter, 1996), CRMP-62 was identified in chick (Goshima et al., 1995), which corresponds to chick CRMP2, four Ulips (1-4) have been identified in mouse (Byk et al., 1996; Byk et al., 1998) and three DRP proteins (1-3) have been identified in humans (Hamajima et al., 1996), which correspond to human CRMP 1, 2 and 4. Of these proteins, TOAD-64 is identical to rat CRMP2. Collapsin is a member of the collapsin/semaphorin family and acts as a repulsive guidance cue (reviewed in Muller et al., 1996) which induces growth cone collapse in chick DRGs (Luo et al., 1993). Interestingly, chick CRMP2 was identified as a 62 kDa protein required for mediating the collapsin response in chick DRGs, as injection of anti-CRMP2 antibody inhibited this response (Goshima et al., 1995). A role for Racl in collapsin signalling was also recently demonstrated, as trituration of chick DRGs with dominant negative Racl also inhibited the collapsin response (Jin and Strittmatter, 1997). This suggests the possibility that rat CRMP2 (TOAD-64), Rac and a2-chimaerin may also be involved in a similar pathway. Human CRMP2 is present in neurofibrillary tangles which are a characteristic of Alzheimer's disease, suggesting a role for CRMP2 in the formation of the tangle bearing neuron and disease progression (Yoshida et al., 1998). TOAD-64 was also identified as part of an antioxidant enzyme complex isolated from rat and bovine synaptic plasma membranes (SPMs) and synaptic vesicles, where it was found in tight association with aldolase C, neuron specific enolase-y, heat shock cognate 70kDa protein (hsc70) and an NADH oxidoreductase homologous to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Bulliard et al., 1997). This suggests a role for TOAD-64 in the cellular response to ROS and stress.

Introduction 79 AIMS

The aims of my project were as follows: 1. To characterise the distribution of the a l- and a2-chimaerin isoforms and potential a2-chimaerin targets in eukaryotic cells and the morphological effects of these proteins in N1E 115 neuroblastoma cells.

2. To examine the role of the a2-chimaerin SH2 domain in determining protein distribution and cell morphology using a2-chimaerin proteins mutated at residues within the SH2 domain predicted to affect its phosphotyrosine binding ability.

3. To investigate the interaction of a2-chimaerin with previously identified targets in eukaryotic cells.

4. Examine the effects of long term a l- and a2-chimaerin overexpression on N1E 115 cell morphology and potential protein interactions.

5. Investigate the effects of Rac and the a-chimaerins on activation of the transcription factor NFk B in HeLa and N1E 115 cells.

Aims 79a CHAPTER TWO: Materials and Methods

Materials and Methods 2.1 Materials Standard lab chemicals were from Sigma or BDH. Restriction enzymes were from Gibco/BRL or New England Biolabs. Tissue culture plasticware was from Nunc (Gibco/BRL) or Greiner. Eukaryotic expression vectors pXJ40-HA, pXJ40-GFP and

pXJ41-HA were from Ed Manser (EMCB, Singapore). Luciferase coupled NFk B reporter vectors were from Jing Ming Dong (EMCB, Singapore). pXJ40HA-B13 and pXJ40GFP- B13 were made by Clinton Monfries, who also generated the TOAD-64 sequence by PCR. Rat al-chimaerin cDNA sequence was isolated by Hong Hwa Lim. Human a2-chimaerin DNA sequence was isolated by Wun Chey Sin. Point mutations in the SH2 domain of a2- chimaerin and also the B 13 and TOAD-64 antibodies were made by Giovanna Ferrari. a2- Chimaerin antibody was made by Greg Michael. Other reagents and equipment were from the sources quoted.

2.2 Microbiological and Nucleic Acid Methods

2.2.1 Bacterial media and reagents L-Broth (Luria Bertani medium") 10g/l (1% w/v) bacto-tryptone (Difco) 5g/l (0.5% w/v) bacto-yeast extract (Difco) 10g/l (1% w/v) NaCl Autoclaved and stored at 4°C.

LB-agar L-Broth plus 15g/l (1.5% w/v) bacto-agar (Difco). Autoclaved and stored at 4°C.

Antibiotics A stock solution of lOOmg/ml ampicillin (Sigma) dissolved in ddH20 was filter sterilised (0.2pm filter, ICN) and aliquots were stored at -20°C. A stock solution of 5mg/ml tetracycline (Sigma) dissolved in 100% ethanol was aliquotted and stored at -20°C.

LB-amp plates LB-agar was heated in a microwave until boiling and allowed to cool to 45°C. Filter sterilised ampicillin (Sigma) was added to give a final concentration of lOOpg/ml, plates were poured and allowed to set at RT.

Materials and Methods 81 2.2.2 Overnight cultures 5ml of LB-amp (lOOpg/ml) was inoculated with a single colony of transformed cells (or 50j l x 1 of -70°C bacterial stock) and incubated with shaking overnight at 37°C.

2.2.3 -70°C bacterial stocks Equal volumes of overnight culture and 100% glycerol were mixed. The resultant bacterial stock was then stored at -70°C where it is viable for a period of years.

2.2.4 Production of competent XLl-Blue E.Coli 8ml LB-tetracycline (50(j,g/ml) was inoculated with 50pl of -70°C XLl-Blue (Stratagene) bacterial stock and incubated with shaking overnight at 37°C. The overnight culture was diluted 1:100 into 800ml LB-tetracycline (50|ug/ml) and incubated with shaking until O D 6oonm was 0.8-1.0. Cells were centrifuged at 3,000rpm for 15 minutes at 4°C (J6, Beckman), the supernatant removed and cells were resuspended in 400ml of lOOmM CaCl2 (pre-chilled to 4°C) on ice. Cells were re-centrifuged at 3,000rpm for 15 minutes at 4°C, the supernatant removed and cells were finally resuspended in 50ml of lOOmM CaCl2 (pre-chilled to 4°C). 1ml of cells was added to 0.5ml of 30% (w/v) glycerol and the 1.5ml aliquots were stored at -70°C.

2.2.5 Transformation of competent E.Coli lpi of plasmid DNA or lOpl of a ligation reaction was added to lOOpl of competent E.Coli XLl-Blue cells and incubated on ice for 1 hour. Cells were heatshocked at 42°C for 45 seconds and immediately returned to ice for 5-10 minutes. 900pl of pre-warmed SOC (20g/l (2% w/v) bacto-tryptone, 5g/l (0.5% w/v) bacto-yeast extract, 5g/l NaCl, lOmM MgCl2, lOmM M gS04, 2mM glucose) was then added and cells were incubated at 37°C for 1 hour with shaking. After incubation, cells were centrifuged at 15,000g for 20 seconds (microcentrifuge 5415c, Eppendorf) and the supernatant removed. Cells were then resuspended in lOOp.1 of warm SOC and plated out on LB-agar plates containing lOOpg/ml ampicillin (see section 2.2.1) and incubated overnight at 37°C. Plates were then stored at 4°C.

2.2.6 Wizard™ minipreps DNA purification system Several 5ml overnight cultures were set up for each DNA sample and a -70°C bacterial stock made for each. The remaining 4.5ml of culture was centrifuged at

Materials and Methods 82 3,000rpm for 5 minutes at 4°C (J6, Beckman) to pellet the bacteria and the supernatant removed. Cell pellets were thoroughly resuspended in 200pl of Cell Resuspension solution (50mM Tris/HCl pH7.5, lOmM EDTA, lOOpg/ml RNase A), then lysed in 200fjl of Cell Lysis solution (200mM NaOH, 1% (w/v) SDS) and mixed by inversion until the solutions cleared. 200pl of Neutralisation solution (1.32M potassium acetate pH4.8) was then added, samples were mixed by inversion and centrifuged at 15,000g for 5 minutes (microcentrifuge 5415c, Eppendorf) to pellet cell debris. lml of Wizard™ Miniprep DNA Purification Resin (Promega) was added to each cleared cell lysate and incubated for 2 minutes at RT. Samples were loaded onto minicolumns using a 2ml syringe, washed with 2ml Column Wash solution (lOOmM NaCl, lOmM Tris/HCl pH7.5, 2.5mM EDTA, 55% (v/v) ethanol) and excess buffer removed via centrifugation at 15,000g for 2 minutes. 50pl TE buffer (lOmM Tris/HCl pH8.0, ImM EDTA pH8.0) pre-warmed to 70°C was added to each minicolumn and incubated for 2 minutes at RT. DNA was then eluted by centrifugation at 15,000g for 20 seconds. Miniprep DNA was subjected to analytical digests to determine the presence and orientation of insert DNA (see section 2.2.1 lc).

2.2.7 Mega plasmid DNA purification system For each DNA sample, 5ml of LB-amp (100pg/ml) was inoculated with 50pl of -70°C bacterial stock and incubated with shaking for 8 hours at 37°C. Cultures were then diluted 1:100 into 500 ml LB-amp (lOOpg/ml) and incubated with shaking overnight at 37°C. Bacteria were harvested by centrifugation at 3,000rpm for 10 minutes at 4°C (J6, Beckman), the supernatants removed and cell pellets thoroughly resuspended in 45ml of Resuspension Buffer (50mM Tris/HCl pH8.0, lOmM EDTA, lOOpg/ml RNase A) pre-chilled to 4°C. Cells were then lysed in 45ml Lysis Buffer (200mM NaOH, 1% (w/v) SDS), mixed briefly and 45ml Neutralisation Buffer (3.0M potassium acetate, pH 5.5) immediately added. Samples were then thoroughly mixed and incubated on ice for 30 minutes to enhance precipitation of cell debris. Lysates were centrifuged at 30,000g for 30 minutes at 4°C (45Ti rotor in L8-M Ultracentrifuge, Beckman) to pellet cell debris. The cleared supernatants were removed promptly and re-centrifuged at 30,000g for 15 minutes at 4°C to ensure removal of all suspended material. During centrifugation, Qiagen 2500 columns were equilibrated with 35ml Equilibration Buffer (750mM NaCl, 50mM MOPS pH7.0, 15% (v/v) ethanol,

Materials and Methods 83 0.15% (w/v) Triton X-100) via gravity flow. The cleared cell lysates were then applied to the columns. Columns were washed with 200ml Column Wash Buffer (1.0M NaCl, 50mM MOPS pH7.0, 15% (v/v) ethanol) and the DNA eluted in 35ml Elution Buffer (1.25MNaCl, 50mM Tris/HCl pH8.5, 15% (v/v) ethanol). 24.5ml of 100% isopropanol (0.7 volumes) was added to each eluted DNA, samples were mixed thoroughly, centrifuged at 25,000g for 30 minutes at 4°C (JA-20 rotor in J2.21 centrifuge, Beckman) and the supernatants removed. Each DNA pellet was then washed with 7ml of 70% (v/v) ethanol, centrifuged at 25,000g for 30 minutes at 4°C and the supernatants removed. DNA pellets were then washed for a second time in 70% (v/v) ethanol and re-centrifuged at 25,000g for 30 minutes at 4°C. Finally, the supernatants were carefully removed and the DNA pellets air dried for 10 minutes to remove all traces of ethanol. DNA samples were then resuspended in 800pl of TE buffer (lOmM Tris/HCl pH8.0, ImM EDTA pH8.0).

2.2.8 Phenol-chloroform extraction of DNA DNA samples were extracted with equilibrated phenol (Phenol:Chloroform:Isoamyl alcohol (25:24:1) stock solution) to remove protein contamination. Each 800fil DNA sample from the Mega Plasmid DNA Purification System (Qiagen) was divided into two 400pl aliquots, an equal volume of equilibrated phenol was added and samples were vortexed until a white emulsion formed. Aqueous and organic phases were separated by centrifugation at 15,000g for 2 minutes at RT (microcentrifuge 5415c, Eppendorf). Each upper aqueous phase containing DNA, was removed to a fresh eppendorf, taking care to avoid the lower phenol phase and the white precipitated protein at the interface. DNA samples were repeatedly extracted with phenol until there was no longer any precipitated protein at the interface. An equal volume of water saturated chloroform was then added to each phenol extracted DNA sample to remove any phenol contamination. Samples were vortexed until a white emulsion formed and then centrifuged at 15,000g for 2 minutes to separate the aqueous and organic phases. Each upper aqueous phase containing DNA, was removed to a fresh eppendorf, taking care to avoid the lower chloroform phase.

2.2.9 Ethanol precipitation The volumes of the cleaned DNA samples from the phenol-chloroform extraction were measured. l/10x this volume of 3M sodium acetate (pH 5.2) and 1ml of 100%

Materials and Methods 84 ethanol (pre-chilled to 4°C) was added to each sample. Samples were mixed well and incubated at -20°C overnight to precipitate the DNA. DNA samples were centrifuged at 15,000g for 10 minutes at 4°C (microcentrifuge 5415c, Eppendorf) to pellet the precipitated DNA and the supernatants were removed carefully. DNA pellets were washed with 70% (v/v) ethanol, dried in a Speedvac Concentrator (Savant) and resuspended in 500pil TE buffer (lOmM Tris/HCl pH8.0, ImM EDTA pH8.0), giving a final volume of 1ml per Qiagen DNA sample.

2.2.10 DNA quantification DNA samples were diluted in ddH20 and using a spectrophotometer, the absorbance at 260nm and 280nm was measured. DNA concentrations were calculated using the fact that a 50p,g/ml solution of DNA has an absorbance value of 1.0 at 260nm.

The purity of DNA samples was estimated with the A260 nm/A280nm ratio, using the fact that a pure DNA sample has an A260 nni/A280iun ratio of 1.8. Any significant difference from this 1.8 value suggests protein or RNA contamination.

2.2.11 Digestion of plasmid DNA with restriction endonucleases Digests were carried out using restriction enzymes from Gibco/BRL and their REact buffer system. A 2-10 fold excess of enzyme was used in each reaction and the final reaction volume was at least 10 times the volume of enzyme used, in order to dilute the glycerol in the enzyme storage buffer so that enzyme works efficiently.

(al Small scale plasmid restriction l-2(ng of appropriate vector DNA (pXJ40-HA, pXJ41-HA or pXJ40-GFP) was linearised by digestion with lpl of restriction enzyme (10U) in a final volume of 50pl. DNA and the appropriate lOxREact buffer were diluted with ddH20 to give a final concentration of lx REact buffer when the enzyme was added. Reactions were incubated at 37°C for 1-2 hours or overnight. Samples were then subjected to blunt ending reactions (see section 2.2.15).

(b) Large scale plasmid restriction 100-200pg of plasmid DNA containing the sequence of interest, was digested with 20pl of restriction enzymes (200U) in a final volume of 200p,l. DNA and the appropriate lOxREact buffer were diluted with ddH20 to give a final concentration of lx REact buffer

Materials and Methods 85 when the enzyme was added. Reactions were incubated at 37°C for 1-2 hours or overnight. Samples were then subjected to DNA fragment purification (see section 2.2.13).

(c) Analytical digests - to determine the presence and orientation of insert DNA. 5pl of miniprep DNA (see section 2.2.6) was digested with 2pl of restriction enzymes (20U) in a final volume of 20|l i 1. DNA and the appropriate lOxREact buffer were diluted with ddH20 to give a final concentration of lx REact buffer when the enzyme was added. Reactions were incubated at 37°C for 1-2 hours or overnight. These digests were then subjected to agarose gel electrophoresis.

2.2.12 Agarose gel electrophoresis DNA samples were separated by gel electrophoresis using 0.8-1.2% (w/v) agarose gels depending on the size of the DNA analysed. Electrophoresis grade agarose (Gibco/BRL) was dissolved in lxTBE (lOOmM Tris, 83mM Boric acid, ImM EDTA) by heating in a microwave until boiling. When the agarose had cooled to 40°C it was poured into an electrophoresis tray, a comb inserted to form the wells and the gel allowed to set. 1/5 volume of 6x DNA sample buffer (30% (w/v) glycerol, 0.25% (w/v) bromophenol blue in lxTBE) was added to each DNA sample prior to loading. 85ng Hind III digested X DNA (Gibco/BRL) and 170ng Hae III digested cpX174 DNA (Gibco/BRL) in lx DNA sample buffer (5% (w/v) glycerol, 0.042% (w/v) bromophenol blue in lxTBE) were run alongside samples as molecular weight markers. Gels were electrophoresed at 150V for 40 minutes in lxTBE (lOOmM Tris, 83mM Boric acid, ImM EDTA). After electrophoresis, gels were stained in ethidium bromide (lpg/ml in lx TBE, from a stock of lOmg/ml in ddH20 stored at RT) for 20-30 minutes and DNA fluorescence was visualised using a UV transilluminator. Gels were then photographed using a Polaroid camera with a yellow lens filter to give a black and white photographic record.

2.2.13 DNA fragment purification from agarose gels A 1% (w/v) agarose gel was poured, a comb with a small and large slot was inserted to form the wells and the gel allowed to set. 50pl of 6x DNA sample buffer (30% (w/v) glycerol, 0.25% (w/v) bromophenol blue in lxTBE) was added to the 200pl large scale plasmid restriction digest (see section 2.2.1 lb). 150-200pl of the digest was loaded in the large well and 15pi in the small marker well. Gels were electrophoresed at 100-120V for 1 hour in lxTBE (lOOmM Tris, 83mM Boric acid, ImM EDTA). After electrophoresis, the

Materials and Methods 86 marker lane was removed from the side of the gel and stained with ethidium bromide (l|iig/ml in lx TBE, from a stock of lOmg/ml in ddH20 stored at RT) for 20-30 minutes. The marker lane was then viewed on a UV transilluminator and the fragment of interest carefully and cleanly excised with a razor blade. The strip was then re-aligned with the main body of the gel and the corresponding position in the large well was excised. The excised agarose containing the DNA fragment of interest was cut into smaller pieces and placed in several Micropure 0.22 separators (Amicon). The separators were then centrifuged at 15,000g for 10 minutes at 4°C (microcentrifuge 5415c, Eppendorf) to elute the DNA. 2x 500pl of eluted DNA fragment per DNA sample were cleaned up using the Magic™ DNA Clean-Up System (Promega).

2.2.14 Magic™ DNA clean-up system DNA samples (200ng-10|ig) were diluted to 500p.l with TE buffer (lOmM Tris/HCl pH8.0, ImM EDTA pH8.0) and incubated with 1ml of Magic™ DNA Clean- Up Resin in 6M guanidine thiocyanate (Promega) for 10 minutes at RT. Samples were then loaded onto minicolumns using a 2ml syringe, washed with 2ml 80% (v/v) isopropanol and centrifuged at 15,000g for 2 minutes at RT to dry the resin. 50pl of TE buffer (pH8.0) pre-warmed to 80°C was added to each minicolumn and incubated for 1 minute at RT. Finally, minicolumns were centrifuged at 15,000g for 2 minutes at RT to elute the DNA.

2.2.15 Blunt ending of DNA lOOpl of cleaned DNA fragments (see section 2.2.13) and 50pl of linearised vector DNA (see section 2.2.11a) were blunted using the Klenow fragment of E.Coli DNA polymerase I (USB) to fill in the 5' overhangs resulting from digestion. l-2|ug of DNA was blunted with ljxl of Klenow (2U) and lOp.1 of 0.5mM dNTP’s

(Chase) in a final volume of 200j l i 1. DNA and lOxREact buffer 2 (0.5M Tris/HCl pH8.0, 0.5M NaCl, 0.1M MgCl2) were diluted with ddH20 to give a final concentration of lx REact buffer when the enzyme was added. Reactions were incubated at 37°C for 20 minutes. Blunted DNA samples were then cleaned using the Magic™ DNA Clean-Up System (Promega) (see section 2.2.14) and eluted in a final volume of 50p,l TE buffer (pH8.0).

Materials and Methods 87 2.2.16 Analysis of DNA purification To check the size and purity of the blunted DNA samples, 5 pi of each cleaned, blunted DNA sample was diluted with 5 pi ddH20 and 1/5 volume of 6x DNA sample buffer (30% (w/v) glycerol, 0.25% (w/v) bromophenol blue in lxTBE). 12pl of each large scale plasmid restriction digest (see section 2.2.1 lb) (already containing 1/5 volume of 6x DNA sample buffer) was run alongside the corresponding purified fragment sample and molecular weight markers on a 0.8-1.0% (w/v) agarose gel (see section 2.2.12).

2.2.17 Radioactive labelling of DNA TRandom primed DNA labelling kit’. Boehringer Mannheim) 5 pi of TE buffer (lOmM Tris/HCl pH8.0, ImM EDTA pH8.0) was added to 5pl of cleaned, blunted fragment DNA (~100ng), the DNA was denatured by heating at 100°C for 3 minutes and then cooled on ice. 5 pi of reaction mix containing dATP, dGTP, dTTP (1:1:1) and random hexanucleotides (Boehringer Mannheim), lpl Klenow (2U/pl, Boehringer Mannheim) and 5pl (50pCi) a -32P dCTP (3000Ci/mmol, Amersham) were added and samples incubated for 1 hour at RT. Samples were then diluted with TE buffer (pH8.0) to give a final volume of lOOpl. 1ml Sephadex G-50 (medium) columns were equilibrated twice with lOOpl TE buffer (pH8.0) and the labelling mixes then applied. The columns were centrifuged at 2,000rpm for 5 minutes at 4°C (Mistral 4L, MSE) to elute the labelled DNA. Labelled DNA probes were denatured by heating at 100°C for 5 minutes prior to use in hybridisation reactions (see section 2.2.20) or stored at -20°C.

2.2.18 Blunt ended T4 DNA ligation 2pl of cleaned, blunted vector DNA (~20ng) was mixed with 5pl of cleaned, blunted fragment DNA (~100ng) (from section 2.2.15). 2pl of 10X ligation buffer (660mM Tris/HCl pH7.6, 66mM MgCl2), 2pl of lOmM ATP (Pharmacia), 2pl of 50mM DTT, lpl of T4 DNA (Amersham, 2.5U/pl) and ddH20 were added to give a final volume of 20pl. Ligation reactions were then incubated at 23°C for 4-5 hours. lOpl of a ligation reaction was then used to transform lOOpl of competent E.Coli XLl-Blue cells (see section 2.2.5) and the resultant bacterial plates were subjected to replica plating.

Materials and Methods 88 2.2.19 Replica plating Nylon filters (Hybond N, Amersham) were used to lift bacterial colonies from agar plates to form replicas. Filters were carefully positioned on the ‘master’ agar plates and the orientations marked. Filters were then removed and placed colony side up on a fresh LB- amp plate (see section 2.2.1). Master plates were incubated at 37°C for 2-4 hours and replica plates for 1-2 hours, to allow the colonies to grow back. After 1-2 hours incubation, filters were removed from the replica plates and placed colony side up on filter paper (grade 3, Whatman) soaked in 10% (w/v) SDS for 3 minutes to lyse the cells. Filters were then transferred to filter paper (grade 3, Whatman) soaked in denaturing solution (1.5M NaCl, 0.5M NaOH) for 5 minutes, neutralising solution (1.5M NaCl, 1M Tris/HCl pH7.4) for 5 minutes and 2x SSC (300mM NaCl, 30mM Na citrate) for 5 minutes. Filters were then air dried for 5 minutes, wrapped in Saran wrap (Dow) and irradiated for 5 minutes on a UV transilluminator. Irradiated filters were washed in (2x SSC, 0.1% (w/v) SDS), wiped firmly with wet tissue to remove bacterial debris and rinsed in 0.3x SSC (45mM NaCl, 4.5mM Na citrate). Filters were then subjected to hybridisation reactions.

2.2.20 Filter hybridisation Filters were placed in hybridisation tubes, washed with 6x SSC (900mM NaCl, 90mM Na citrate) and incubated in lOmls of prehybridisation buffer (6x SSC, 5x Denhardts, 0.01M EDTA, 0.5% (w/v) SDS, 100|ug/ml denatured salmon sperm DNA) for 1-2 hours at 60°C. Filters were then incubated in lOmls of hybridisation buffer (6x SSC, 5x Denhardts, 0.01M EDTA, 0.5% (w/v) SDS, 100pg/ml denatured salmon sperm DNA, 10|_ig/ml poly A) containing the appropriate radioactively labelled DNA probe (see section 2.2.17) and hybridised overnight at 60°C. After hybridisation filters were washed in (2x SSC, 0.1% (w/v) SDS) at 60°C for 30 minutes and then (O.lx SSC, 0.1% (w/v) SDS) at 50-55°C for 30 minutes. Filters were then carefully wrapped in Saran wrap (Dow), placed in a film cassette and exposed to X-Omat AR film (Kodak) at -70°C for 1-24 hours. The orientations of the replica filters were marked on the autoradiograph, which was then lined up with the corresponding master plate. Colonies on the master plates which co­ localised with a positive signal on the film were marked. Selected colonies were grown as overnight cultures and the DNA harvested (see section 2.2.6). The miniprep DNA samples were then subjected to restriction analysis to determine the presence and orientation of the

Materials and Methods 89 insert (see section 2.2.11c). DNA samples found to contain the insert in the correct orientation were then used in re-transformation reactions to obtain pure bacterial clones.

2.2.21 Re-transformation of competent E.Coli 1 pi of miniprep DNA containing the insert in the correct orientation, was added to lOOpl of competent E.Coli XLl-Blue cells and incubated on ice for 1 hour. Cells were heatshocked at 42°C for 45 seconds, returned to ice for 5 minutes and then incubated at RT for 5 minutes. Cells were plated out on LB-amp plates (see section 2.2.1) and incubated overnight at 37°C. Plates were then stored at 4°C. Several clonal colonies were selected for overnight culture, the DNA purified (see section 2.2.6) and then subjected to restriction analysis to confirm the presence and orientation of the insert (see section 2.2.11c). Clones found to contain the insert in the correct orientation were then used in the TNT in vitro Transcription-Translation Assay (Promega).

2.2.22 In vitro transcription-translation assay The TNT in vitro Transcription-Translation Assay (Promega) was used to examine whether DNA constructs directed expression of full length protein products. Miniprep DNA from re-transformation clones found to contain the insert in the correct orientation was phenol-chloroform extracted and ethanol precipitated (see sections 2.2.8 and 2.2.9) before it was used in this assay. 0.5 fig of phenol-chloroform extracted DNA, 12.5fil Rabbit Reticulocyte lysate, lpl TNT Reaction Buffer, 0.5pl T7 RNA polymerase, 0.5pl Amino acid mix minus methionine (ImM), 0.5pl RNasin ribonuclease inhibitor (40U/pl) and 2pl 35S-methionine (lOOOCi/mmol at lOmCi/ml, Amersham) were diluted with ddH20 to give a final reaction volume of 25pl. Samples were then incubated at 30°C for 1.5 hours. A sample containing 0.5pg luciferase DNA was included as a positive control and one containing no DNA was included as a negative control, to see the background level of the reaction. 5pi of each reaction was diluted with an equal volume of 2x SDS sample buffer (125mM Tris/HCl pH6.8, 40% (w/v) glycerol, 4% (w/v) SDS, 1% (v/v) (3-mercaptoethanol, 0.1% (w/v) bromophenol blue). Protein samples were denatured by heating at 100°C for 5 minutes, prior to loading on 10-12% SDS-PAGE gels (see section 2.6.7). After electrophoresis, gels were fixed in fresh destain (40% (v/v) methanol, 10% (v/v) glacial acetic acid, 50% (v/v) H20) for 30 minutes with shaking at RT and then incubated in

Materials and Methods 90 Amplify scintillant (Amersham) for 30 minutes with shaking at RT. Gels were dried down, placed in a film cassette and exposed to X-Omat AR film (Kodak) at -70°C for 0.5-4 hours. The clones which produced the strongest expression of full length protein for each DNA sample were selected and large scale plasmid DNA preparations were made (see section 2.2.7).

2,3 Cloning of DNA constructs

2.3.1 Eukaryotic expression vector details pXJ40-HA, pXJ40-GFP and pXJ41-HA are epitope tagged eukaryotic expression vectors, which were obtained from Ed Manser (IMCB, Singapore). pXJ40-HA and pXJ40- GFP enable transient transfection of mammalian cells whilst pXJ41-HA enables permanent transfection of mammalian cells. pXJ40-HA and pXJ41-HA differ in 2 ways - firstly in the sequence of their multiple cloning sites and secondly, pXJ41-HA contains the neomycin resistance gene which is absent from pXJ40-HA (compare figures 2. la and 2.3). pXJ40-HA and pXJ40-GFP differ in only one way, and that is in the sequence of their epitope tags which are positioned N terminal to the multiple cloning site (compare figures 2.1a and 2.2a). Vector pXJ40-HA contains a small 10 amino acid HA tag which is derived from the haemagglutinin protein of the human influenza virus (see figure 2.1b), whilst pXJ40-GFP contains the 28kDa ‘Green Fluorescent Protein’ (GFP) (see figures 2.2b and 2.2c). Despite these differences, the main structural elements of these 3 vectors are identical as they are all derived from the eukaryotic expression vector pXJ40 (Xiao et al., 1991) which contains the strong hCMV enhancer- promoter unit. The enhancer sequence of hCMV used in the pXJ40 vector is located upstream from the major immediate early gene of hCMV and was identified as a very strong transcriptional enhancer with little species or tissue specificity (Boshart et al., 1985). The effect of this hCMV enhancer in combination with its homologous promoter on gene expression was tested (Foecking and Hofstetter, 1986). Gene expression from the hCMV enhancer- promoter unit was found to be 3-18 fold higher than from the enhancer and early promoter of SV40 and 7-150 fold higher than from the long terminal repeat of Rous sarcoma virus (RSV) enhancer-promoter unit, depending upon the cell type tested. Thus the hCMV enhancer-promoter unit is ideal for use in a eukaryotic expression vector where very strong expression of an inserted gene sequence is required.

Materials and Methods 91 Pvu ll/Xba I 1 HCMV promoter ,Sst I 516 Pst l/Stu I 608

Rabbit B globin intron II

promoter SV40 1296 EcoRI » H A ta g « BamHI Hindlll Xhol Notl Smal Pstl Sstl Kpnl Bglll 1352

Figure 2.1a: Structure of the pXJ40-HA vector

pXJ40-HA MCS

EcoRI I------1 HA tag ...... T7 » GAA TTC ACC ATG TAC CCA TAC GAC GTG CCA GAC TAC GCA Kozak MYPY DV PDY A

BamHI Hind3 Xho1 Not1 Pstl Sac1 I I I I I I I I I I I I I GGA TCC AAG CTT CTC GAG GCG GCC GCC CGG GCT GCA GGA GCT CGG GSKLLE AA AR A A G A R I______I Sma1

Kpn1 Bgl2 1 I --- 1 TACCAGATC TT Y Q I L

Figure 2.1b: DNA sequence of the multiple cloning site of the pXJ40-HA vector

Figure 2.1 92 Pvu ll/Xba I 1 HCMV promoter ,Sst I 516 Pst l/Stu I 608

Rabbit B globin intron II pXJ40-G FP 4281 bp

T7 promoter

SV40 1296 EcoRI » G F P tag insert« BamHI Hindlll Xhol Notl Smal Pstl Sstl Kpnl Bglll 1 352

Figure 2.2a: Structure of the pXJ40-GFP vector

PXJ40-GFP MCS

EcoRI I I T7 » GAA TTC ACC ATG ...... GFP tag. Kozak M

BamHI Hind3 Xhol Not1 Pstl Sac1 I I I I I------1 I------1 I------1 I------1 I— GGA TCC AAG CTT CTC GAG GCG GCC GCC CGG GCT GCA GGA GCT CGG GSKLLE AA AR AAGA R I______I Sma1

Kpn1 Bgl2 1 I --- 1 TACCAGATC TT Y Q I L

Figure 2.2b: DNA sequence of the multiple cloning site of the pXJ40-GFP vector

Figure 2.2 93 Figure 2.2c: DNA sequence of the green fluorescent protein (GFP) epitope

EcoRI

ATGAGTAAAGGAGAAGAACTTTTCACTGGAGTTGTCCCAATTCTTGTTGAATTAGATGGTGATGTTAATGGGCAC 100 mskgeel ftgvvpi Iveldgdvngh

AAATTTTCTGTCAGTGGAGAGGGTGAAGGTGATGCAACATACGGAAAACTTACCCTTAAATTTATTTGCACTACTGGAAAACTACCTGTTCCATGGCCAA -♦ 200 k fsvsgegegdatygkl t t kf i cttgk I p v p w p t M

CACTTGTCACTACTTTC TCT TATGGTGTTCAATGCTTTTCAAGATACCCAGArCATATGAAACAGCArGACTTTTTCAAGAGTGCCATGCCCGAAGGTTA 300 t v t t f S/T ygvqc f s rypdhmkqhdf f k sampeg yw I I TGTACAGGAAAGAACTATATTTTTCAAAGATGACGGGAACTACAAGACACGTGCTGAAGTCAAGTTTGAAGGTGATACCCTTGTTAATAGAATCGAGTTA -♦ 400 vqerriffkddgnyktraevkfegdtlvnriel

AAAGGTATTGATTTTAAAGAAGATGGAAACATTCTTGGACACAAATTGGAATACAACTATAACTCACACAATGTATACATCATGGCAGACAAACAAAAGA 500 kg idfkedgni I ghkleynynshnvyimadkqkn,5#

ATGGAATCAAAGTTAACTTCAAAATTAGACACAACATTGAAGATGGAAGCGTTCAACTAGCAGACCATTATCAACAAAATACTCCAATTGGCGATGGCCC -♦ 600 gikvnfkirhniedgsvqladhyqqntpigdgp,K

TGTCCTTTTACCAGACAACCATTACCTGTCCACACAATCTGCCCTTTCGAAAGATCCCAACGAAAAGAGAGACCACATGGTCCTTCTTGAGTTTGTAACA -♦ 700 v I Ipdnhylstqsalskdpnekrdhmvl lefvt 223

BamHI GCTGCTGGGATTACACATGGCATGGATGAACTATACAAAT 740 aagithgmdelyk.

Figure 2.2 94 Pvu ll/Xba I 1 HCMV promoter „Sst I 516 .Pst l/Stu

Rabbit B globin intron II

1296 promoter EcoRI » H A ta g « SV40 BamHI Smal * HindllJ * Xhol Kpnl Bglll 1 340

Neomycin resistance gene

Figure 2.3; Structure of the pXJ41-HA vector

Figure 2.3 95 The pXJ40 eukaryotic expression vector was generated by Xiao et al., (1991) (Xiao et al., 1991) and is composed of fragments from plasmids pSG5 and pCMVcat. The pSG5 vector was constructed by Green et al., (1988) (Green et al., 1988) and combines the eukaryotic expression vector pKCR2 and the high copy plasmid vector Bluescribe Ml 3+ (Stratagene). This pSG5 vector is now commercially available from Stratagene and contains the early S V40 promoter sequence, the rabbit P-globin gene intron II, the T7 promoter, the SV40 polyadenylation sequences and the Bluescribe M l3+ (BSM13+) sequence, which includes the ampicillin resistance gene and a bacterial origin of replication. The small PvuII (3984)-StuI (350) fragment of plasmid pSG5 which contains the SV40 early promoter sequence was removed and replaced with the Xbal-PstI fragment of plasmid pCMVcat (Foecking and Hofstetter, 1986) which contains the much stronger immediate early hCMV enhancer-promoter unit. The pSG5 derived MCS was replaced with a new polylinker 5’ EcoRI-BamHI-HindIII-XhoI-NotI-SmaI-PstI-SstI-KpnI-BglII-3, (Xiao et al., 1991) to generate the vector pXJ40. To generate the vectors pXJ40-HA and pXJ40-GFP, a 10 amino acid HA tag which is derived from the haemagglutinin protein of the human influenza virus (see figure 2.1b), or the 28kDa green fluorescent protein (GFP) (see figures 2.2b and 2.2c), was then inserted between the EcoRI and BamHI sites of the MCS (E. Manser). These pXJ40-HA and pXJ40-GFP vectors differ in only one way, and that is in the sequence of their N terminal epitope tags (compare figures 2.1a and 2.2a). Vector pXJ41-HA was generated via addition of a neomycin resistance gene to vector pXJ40-HA, which also altered the sequence of the MCS in pXJ41-HA. Thus, pXJ40-HA and pXJ41-HA differ in 2 ways - firstly in the sequence of their multiple cloning sites and secondly, pXJ41-HA contains the neomycin resistance gene (which enables permanent transfection of eukaryotic cells) which is absent from pXJ40-HA (compare figures 2.1a and 2.3). However despite their differences, these 3 vectors contain the following common structural elements (see figures 2.1a, 2.2a and 2.3): 1. The hCMV enhancer-promoter unit, which drives very strong expression of the inserted gene sequence 2. The rabbit P-globin intron II which facilitates splicing of the expressed transcript 3. The T7 bacteriophage promoter which enables in vitro transcription of cloned inserts 4. A multiple cloning site (MCS) which enables easy insertion of gene sequences 5. The SV40 polyadenylation sequence which greatly increases the level of protein

Materials and Methods 96 expression by stabilising protein synthesis 6. BSM13+ sequence which includes a bacterial origin of replication and the ampicillin resistance gene 7. The N terminal HA or GFP tag which enables easy detection of expressed proteins with commercially available HA and GFP antibodies

2.3.2 Generation of pXJ40-GFP vector The pXJ40-GFP vector was generated by E. Manser (IMCB, Singapore) from the pXJ40-HA vector. Briefly, the green fluorescent protein (GFP) was isolated from the jellyfishAequorea Victoria and sequenced (Prasher et al., 1992, ‘gfpl O’ sequence (l-966bp), Genbank Accession number M62653). This wild type sequence was obtained by PCR and mutated to create an S65T mutant, which is approximately 6 fold brighter than wild type GFP (Heim et al., 1995). Artificial EcoRI and BamHI sites were engineered at the 5’ and 3’ ends respectively of the GFP sequence (22-740bp) (see figure 2.2c). The HA tag was excised from vector pXJ40-HA as an EcoRI/BamHI fragment and replaced with an EcoRI/BamHI fragment containing full length S65T mutated GFP (22- 740bp) thus generating the vector pXJ40-GFP. (Both EcoRI and BamHI sites are regenerated).

2.3.3 Cloning of al-chimaerin into eukaryotic expression vectors Rat al-chimaerin cDNA (1-2321 nucleotides) was isolated from a A,gtlO cDNA library of rat brain free poly-ribosomal mRNA by H. H. Lim (Lim et al., 1992). A 1. lkb Fok I/Bal I fragment containing the full length coding sequence for rat al-chimaerin (417-1534bp) was purified from pBS-rlam631 and blunted with the Klenow fragment of DNA polymerase I (see appendix 1 for DNA sequence). This blunted Fok I/Bal I fragment was cloned into the Klenow blunted Nco I site of the 8.5kb pAS2 vector (Clontech), a ‘GAL4 DNA binding domain fusion cloning and expression vector’, to give p AS2-rlam631. Cloning of this fragment regenerates the Fok I site. An Nde I/Pst I fragment containing the full length coding sequence for rat a l- chimaerin (417-1534bp) was purified from pAS2-rlam631 and blunted with T4 DNA polymerase. This blunted fragment contained 37 nucleotides of the yeast vector sequence including the restriction sites Sma I, BamHI and Sal I located at the 3’ end of the a l- chimaerin sequence. This blunted Nde I/Pst I fragment was then cloned into the Klenow blunted BamHI sites of the eukaryotic expression vectors pXJ40-HA, pXJ41-HA and

Materials and Methods 97 pXJ40-GFP.

2.3.4 Cloning of q2-chimaerin into eukaryotic expression vectors 921.2 is the full length human sequence of a2-chimaerin (l-2032bp) isolated by W-C. Sin (composed of 5’RACE fragments plus clone 121) (Hall et al., 1993). Artificial EcoRI and Xho I sites were engineered at the 5’ and 3’ ends of this a2- chimaerin sequence, respectively. The 2.0kb EcoRI/Xho I fragment containing the full length coding sequence for human a2-chimaerin (l-2032bp) was purified from pBS- 921.2 and blunted with the Klenow fragment of DNA polymerase I (see appendix 2 for DNA sequence). The blunted a2-chimaerin fragment (l-2032bp) was then cloned into the Klenow blunted BamHI site of the pXJ40-HA, pXJ40-GFP and pXJ41-HA vectors. Cloning of the fragment regenerates the Xho I site. pBS-921.2 was mutated by G. Ferrari to produce four site directed point mutations in the a2-chimaerin SH2 domain - E49W, R56L, R73L and N94H. The 2.0kb EcoRI/Xho I fragments containing the full length coding sequences for the four a2- chimaerin SH2 domain mutants (l-2032bp) were also purified and blunted with the Klenow fragment of DNA polymerase I. Blunted a2-chimaerin SH2 domain mutant fragments (l-2032bp) were then cloned into the Klenow blunted BamHI site of the pXJ40-HA and pXJ40-GFP vectors. Cloning of the fragments regenerates the Xho I site.

2.3.5 Cloning of TOAD-64 into eukaryotic expression vectors Rat TOAD-64 (Turned On After Division) was isolated from a rat brain library screen and sequenced (Minturn et al., 1995b, full length sequence (l-2947bp), Genbank Accession number Z46882). Full length rat TOAD-64 coding sequence (178-1896bp) was obtained by PCR by C. Monfries. Artificial BamHI and EcoRI sites were engineered at the 5’ and 3’ ends of this TOAD-64 sequence, respectively (see appendix 3 for DNA sequence). The 1.7kb BamHI/EcoRI PCR fragment containing the full length coding sequence for rat TOAD-64 (178-1896bp) was purified and ligated into the BamHI/EcoRI digested pBSII SK(+) vector (Stratagene). Cloning of the fragment regenerates both the BamHI and EcoRI sites. A 1.7kb Spe I/Kpn I fragment containing the full length coding sequence for rat TOAD-64 (178-1896bp) was purified from pBS-TOAD-64 and blunted with T4 DNA polymerase. This blunted fragment contained 48 nucleotides of the Bluescript vector sequence including 11 restriction sites - EcoRV, Hind III, Cla I, Bspl06 I, Sal I, Acc I, Hinc

Materials and Methods 98 II, Xho I, Apa I, Dra II and Eco0109 I, located at the 3’ end of the TOAD-64 sequence. The blunted pBS-TOAD-64 fragment was then cloned into the Sma I site of the pXJ40-HA and pXJ40-GFP vectors. Cloning of the fragment regenerates the Sma I site at the 3 ’ end of the TOAD-64 sequence.

2.4 Cell culture

2.4.1 Frozen cell stocks A confluent plate of cells was harvested as described (see section 2.4.3). N1E 115, COS7 and permanently transfected N1E 115 cell pellets were resuspended in 3mls of a filter sterilised (0.2p,m filter, ICN) solution of 90% FCS, 10% DMSO. HeLa cell pellets were resuspended in 3mls of a filter sterilised (0.2|j.m filter, ICN) solution of 90% complete media, 10% DMSO. 1.5ml aliquots of cells were placed in cryo tubes (Nunc), cooled slowly in a polystyrene box at -20°C for several hours and then transferred to -70°C overnight. Cell stocks were finally transferred to a Dewar containing liquid nitrogen, for long term storage.

2.4.2 Culture from frozen cell stocks Cell stocks were removed from liquid nitrogen and thawed quickly (2-3 minutes) at 37°C. Each cryo tube of cells was added to lOmls of fresh media to dilute the DMSO. Cells were centrifuged at 900rpm for 7 minutes at RT in a benchtop centrifuge (GP centrifuge, Beckman), resuspended in lOmls of fresh media and seeded into 3x 90mm plates.

2.4.3 Cell culture maintenance N1E 115 and COS7 cells were cultured in DMEM (4500mg/l D-glucose, Gibco/BRL) supplemented with 10% foetal calf serum (FCS, Gibco/BRL) and 1% antibiotic/antimycotic solution (containing penicillin, streptomycin and amphotericin B, Gibco/BRL). HeLa cells (ATCC CCL-2) were cultured in Eagle’s minimum essential medium (Eagle’s MEM, Sigma) containing Earle’s salts, non essential amino acids and sodium bicarbonate, supplemented with 10% FCS, 1% antibiotic/antimycotic solution and 2mM L-glutamine (Gibco/BRL). Permanently transfected N1E 115 cell lines were cultured in DMEM (4500mg/l D-glucose, Gibco/BRL) supplemented with 10% foetal calf serum, 1% antibiotic/antimycotic solution and 800(ig/ml G418 (Gibco/BRL). All cell lines were incubated at 37°C with 5%CC>2.

Materials and Methods 99 Media was removed from N1E 115 cells via aspiration and lOmls of media was added to each 90mm plate of cells. N1E 115 cells were removed from the surface of the plates by the action of pipetting the media and collected in a 50ml falcon tube. For COS7, HeLa and permanently transfected N1E 115 cells, trypsinisation is required to detach the cells from the surface of the plates. Media was removed via aspiration and each 90mm plate of cells was washed with PBS. The PBS was removed via aspiration, 0.75ml of pre-warmed lOx Trypsin (Gibco/BRL) was added to each plate and incubated for several minutes at room temperature until the cells detached. lOmls of media containing 10% FCS was then added to each plate to inactivate the trypsin and cells were collected in a 50ml falcon tube. Once collected in a 50ml falcon tube, cells were centrifuged at 900rpm for 7 minutes at RT in a benchtop centrifuge (GP centrifuge, Beckman), media aspirated and the cells resuspended in lOmls of fresh media. Cells were seeded into 90mm plates at a 1/10-1/40 dilution, twice a week or counted using a haemocytometer and plated at the desired density.

2.4.4 Treatment of slides and coverslips Glass coverslips (22mm diameter circular coverslips, BDH) were washed in 40% HC1, 60% ethanol for 20 minutes at RT, rinsed thoroughly in clean water (~2 litres per 100 coverslips), blotted dry on filter paper (3MM, Whatman) and oven baked at 121°C overnight, before use. Single well glass chamber slides (45mm x 20mm, Nunc) or cleaned glass coverslips in individual 35mm plates, were incubated in 5pg/ml poly-L-lysine (Sigma) for 5 minutes, washed twice with ddH20 and stored at 4°C in ddH20 until required.

2.4.5 Treatment of permanently transfected N1E 115 cells for immunocvtochemical analysis Permanently transfected N1E 115 cell lines were plated out on poly-L-lysine coated circular coverslips (in 35mm plates) (see section 2.4.4), at 2xl05 cells per plate in complete G418 media and incubated for 4-5 hours at 37°C, 5% C 02 until the cells had adhered to the coverslips. At this point, cells were either serum starved to induce differentiation or incubated in complete media overnight. For serum starvation, media was aspirated and cells were incubated in DMEM containing 800pg/ml G418 and 1% antibiotic/antimycotic at 37°C, 5% C 02 overnight. The next day, coverslips were subjected to immunocytochemistry (see section 2.6.10).

Materials and Methods 100 2.5 Mammalian cell transfection Large scale plasmid DNA preparations (see section 2.2.7) of HA and GFP tagged DNA constructs were phenol-chloroform extracted and ethanol precipitated (see sections 2.2.8 and 2.2.9) to remove contaminants. These cleaned DNA samples were then used in transfection experiments.

2.5.1 Transient transfection of COS7 cells Two confluent 90mm plates of COS7 cells were split into lOx 90mm plates and incubated in complete media for 24 hours at 37°C, 5% CO 2 until -70% confluent. The next day, media was aspirated and cells were starved in 5.5ml of serum free DMEM (Gibco/BRL) for 1 hour at 37°C. During this incubation period, 5pg of phenol- chloroform extracted HA or GFP tagged DNA (Qiagen) or a total of lOpg of DNA for cotransfection experiments (5pg of each DNA construct) and 16pl or 32pl of lipofectamine (Gibco/BRL) respectively, were mixed in 1,3ml of serum free DMEM and incubated for 45 minutes at RT. Transfection solutions were then added to the cells and incubated at 37°C, 5% C 0 2for 5 hours. After 5 hours, 1% FCS and 1% antibiotic/antimycotic was added to each plate and cells were then incubated overnight. Cells were harvested at 24 hours for Western analysis (see section 2.6.1) or immunoprecipitation (see section 2.6.2).

2.5.2 Transient transfection of N1E 115 cells (a) For immunoprecipitation analysis N1E 115 cells were plated out in complete media at l-2xl06 cells per 90mm plate and incubated for 24 hours at 37°C, 5% C 02. Cells were transfected using the same method as for COS7 cells (see section 2.5.1) and harvested at 24 hours for immunoprecipitation analysis (see section 2.6.2).

(b) For the NFkB reporter assay N1E 115 cells were plated out in complete media at 2xl05 cells per 35mm plate, in triplicate for each DNA sample and incubated for 24 hours at 37°C, 5% C 02. The next day, each 35mm plate of cells was transfected with a total of 1.5p.g of

DNA - 0.5pg pgal reporter vector, 0.5pg of NFk B reporter vector and 0.5 jig of HA tagged

DNA or a total of 4.0pg of DNA - l.Opg pgal reporter vector, 1.5pg of NFk B reporter vector and 1.5pg of HA tagged DNA. Firstly the complete media was aspirated and cells were starved in 1ml of serum free DMEM (Gibco/BRL) for 1 hour at 37°C. During this

Materials and Methods 101 incubation period 3x stock transfection solutions containing 4.5pg of phenol-chloroform extracted DNA (Qiagen) and 12pl lipofectamine (Gibco/BRL) or 12pg of phenol- chloroform extracted DNA (Qiagen) and 40pl lipofectamine (Gibco/BRL) were mixed in 650pl of serum free DMEM and incubated for 45 minutes at RT. Transfection solutions were then added to each triplicate plate and incubated at 37°C, 5% C 02for 5 hours. After 5 hours, 1ml of DMEM containing 10% FCS and 1% antibiotic/antimycotic was added to each plate and cells were then incubated overnight. Cells were harvested at 24 hours and assayed for NFk B reporter activity (see section 2.7.1).

(c) For immunocvtochemical analysis N1E 115 cells were plated out on poly-L-lysine coated chamber slides or circular coverslips (in 35mm plates) (see section 2.4.4), at lxlO5 cells per slide (or 35mm plate) in complete media and incubated for 4-5 hours at 37°C, 5% C 02. Once the cells had adhered to the slides (or coverslips), media was aspirated and cells were incubated overnight in DMEM + 1% antibiotic/antimycotic at 37°C, 5% C 02. This serum starvation causes N1E 115 cells to differentiate. The next day, media was aspirated and cells were starved in 1ml of serum free DMEM (Gibco/BRL) for 1 hour at 37°C. During this incubation period, lpg of phenol- chloroform extracted HA or GFP tagged DNA (Qiagen) and 6 pi lipofectamine (Gibco/BRL) were mixed in 200pl of serum free DMEM and incubated for 45 minutes at RT. Transfection solutions were then added to the cells and incubated at 37°C, 5% C 02 for 5 hours. After 5 hours, 1ml of DMEM containing 10% FCS and 1% antibiotic/antimycotic was added to each slide and cells were then incubated overnight. At 24 hours post transfection, cells were fixed and subjected to immunocytochemistry (see section 2.6.10).

2.5.3 Transient transfection of HeLa cells for NFkB reporter assay HeLa cells were plated out in complete media at 2x105 cells per 35mm plate, in triplicate for each DNA sample and incubated for 24 hours at 37°C, 5% C 02. The next day, each 35mm plate of cells was transfected with a total of l.Opg of

DNA - 0.5jug of NFk B reporter vector and 0.5pg of HA tagged DNA. Firstly the complete media was aspirated and cells were starved in 1ml of serum free Eagle’s MEM + 2mM L- glutamine, for 1 hour at 37°C. During this incubation period 3x stock transfection solutions containing a total of 3.0pg of phenol-chloroform extracted DNA (Qiagen) and 9pl

Materials and Methods 102 lipofectamine (Gibco/BRL) were mixed in 650pl of serum free Eagle’s MEM + 2mM L- glutamine and incubated for 45 minutes at RT. A third of each 3x stock transfection solution was then added to each triplicate plate and incubated at 37°C, 5% C 02for 5 hours. After 5 hours, 1ml of Eagle’s MEM containing 10% FCS, 1% antibiotic/antimycotic and 2mML- glutamine was added to each plate and cells were then incubated overnight. Cells were harvested at 24 hours and assayed for NFk B reporter activity (see section 2.7.1).

2.5.4 Permanent transfection of N1E 115 cells N1E 115 cells were plated out in complete media at 1x10s cells per 35mm plate and incubated for 24 hours at 37°C, 5%C02. Six 35mm plates were transfected for each DNA sample. The next day, media was aspirated and cells were starved in 1ml of serum free DMEM (Gibco/BRL) for 1 hour at 37°C. During this incubation period, lpg of phenol-chloroform extracted pXJ41-HA DNA construct (al-chimaerin, a2-chimaerin or empty pXJ41-HA vector) and 6p.l lipofectamine (Gibco/BRL) were mixed in 200pl of serum free DMEM and incubated for 45 minutes at RT. Transfection solutions were then added to the cells and incubated at 37°C with 5%C02for 5 hours. After 5 hours, 1ml of DMEM containing 10% FCS and 1% antibiotic/antimycotic was added to each plate and cells were then incubated overnight. Transfection mixes were removed at 24 hours and replaced with 3 mis of DMEM containing 10% FCS (no antibiotics) and incubated overnight. Media was removed the next day, cells were washed with PBS and trypsinised until the cells detached. 3ml of media containing 10% FCS was added to each plate to inactivate the trypsin, cells were collected in a 50ml falcon tube and centrifuged at 900rpm for 7 minutes at RT in a benchtop centrifuge (GP centrifuge, Beckman). Cells were resuspended in lOmls of complete media, seeded into 9x 90mm plates and allowed to recover overnight. Untransfected N1E 115 cells were also plated out and incubated overnight. Media was removed the next day and replaced with complete media containing 800pg/ml G418 (Gibco/BRL) to select for stable transfectants. Untransfected NIE 115 cells were also incubated in complete media containing 800pg/ml G418 as a control to ensure that the G418 selection was working. G418 selection media was replaced twice a week for 2-3 weeks until the transfected cells grew into plaques. In order to select plaques, media was removed, cells were washed with PBS, well separated clones were isolated with cloning rings, trypsinised and removed to individual 35mm plates. Clones were grown for 1-2 weeks (until the cells filled the

Materials and Methods 103 plates), then trypsinised and transferred to 90mm plates, grown for a further 1-2 weeks and then frozen in 2 aliquots per plate (see section 2.4.1).

2.6 Analysis of cellular proteins

2.6.1 Protein expression in transiently transfected COS7 cells Transfection solutions were removed by aspiration from transiently transfected COS7 cells (see section 2.5.1) at 24 hours. COS7 cells were washed once with PBS and incubated in 400pl of Triton Cell Lysis buffer (25mM HEPES pH7.3, 20mM P- glycerophosphate, 0.3MNaCl, 1.5mM MgCb, 0.2mM EDTA pH8.0, 5% (w/v) glycerol, 0.5% (w/v) Triton X-100 adjusted to pH7.7, containing freshly added inhibitors ImM PMSF, ImM Na vanadate, 0.5mM DTT, 5jag/ml pepstatin, 5pg/ml aprotinin) for 5 minutes at RT. Cells were harvested by scraping, thoroughly lysed via passing through a 25G needle (Microlance) and centrifuged at 20,000g for 10 minutes at 4°C (Microcentrifuge 154, Camlab). Supernatants were removed promptly and the protein concentration of the supernatant or ‘cell lysate’ fractions was determined (see section 2.6.6). The insoluble pellet fractions were resuspended in 50pl of 2x SDS sample buffer (125mM Tris/HCl pH6.8, 40% (w/v) glycerol, 4% (w/v) SDS, 1% (v/v) P-mercaptoethanol, 0.1% (w/v) bromophenol blue) by sonication (sonicator cuphorn, Heat Systems). Both cell lysates and resuspended pellet fractions were stored at -70°C. In order to analyse the samples, they were subjected to SDS-PAGE gel electrophoresis, western blotted onto nitrocellulose and probed with antibodies (see sections 2.6.7-2.6.9).

2.6.2 Immunoprecipitation from transiently transfected COS7 and N1E 115 cells Transfection solutions were removed by aspiration from transiently transfected COS7 cells (see section 2.5.1) or N1E 115 cells (see section 2.5.2a) at 24 hours. Cells were gently washed once with PBS and incubated in 400pl of Immunoprecipitation Cell Lysis buffer (25mM HEPES pH7.3, 20mM p-glycerophosphate, 0.3M NaCl, 1.5mM MgCb, 0.2mM EDTA pH8.0, 5% (w/v) glycerol, 1% (w/v) Triton X-100, 0.5% (w/v) Na deoxycholate adjusted to pH7.7, containing freshly added inhibitors ImM PMSF, ImMNa vanadate, 0.5mM DTT, 5|Lig/ml pepstatin, 5pg/ml aprotinin) for 5 minutes at RT. Cells were harvested by scraping, thoroughly lysed via passing through a 25G needle (Microlance) and centrifuged at 20,000g for 10 minutes at 4°C (Microcentrifuge 154, Camlab). Supernatants

Materials and Methods 104 were removed promptly and placed on ice. 50pl of each supernatant or ‘cell lysate’ fraction was removed for later analysis, and the remaining volume then diluted 30 fold with ddH20 (containing ImM PMSF, ImM Na vanadate, 0.5mM DTT, 5pg/ml pepstatin, 5pg/ml aprotinin), to give a final detergent concentration of 0.05%. 5pg of HA antibody (or 5 pi of a2 antibody) were added to the diluted COS7 or N1E 115 cell lysates and samples were incubated with rolling at 4°C for 1 hour. 50pl of protein A sepharose beads (Sigma) were then added to each sample and incubated with rolling at 4°C for 1 hour. Samples were centrifuged at l,000rpm for 5 minutes at 4°C (J6, Beckman) and the diluted cell lysates carefully removed via aspiration. The protein A sepharose beads were washed for 10 minutes in lOmls of wash buffer (PBS, 20mM (3- glycerophosphate, 0.1% (w/v) Triton X-100, 2mM Na vanadate) per sample, centrifuged at l,000rpm for 5 minutes at 4°C (J6, Beckman) and the supernatants carefully removed via aspiration. Samples were washed 3 times. After the final wash, protein A sepharose beads were transferred to 2ml eppendorfs, centrifuged at 0.2g for 5 minutes at 4°C (Microcentrifuge 154, Camlab), the supernatants removed and 50pl of 2x SDS sample buffer (125mM Tris/HCl pH6.8, 40% (w/v) glycerol, 4% (w/v) SDS, 1% (v/v) (3- mercaptoethanol, 0.1% (w/v) bromophenol blue) was added per sample. Samples were incubated at 100°C for 10 minutes to elute bound proteins from the beads and stored at -20°C. In order to analyse samples, they were subjected to SDS-PAGE gel electrophoresis, western blotted onto nitrocellulose and probed with antibodies (see sections 2.6.7-2.6.9).

2,6.3 Protein expression in permanently transfected N1E 115 cells - CSK extraction Permanently transfected N1E 115 cell lines were cultured and harvested as described in section 2.4.3. Cell pellets were resuspended in lOmls of PBS and counted using a haemocytometer. 25x106 cells for al-chimaerin and empty pXJ41-HA vector cell lines and 15xl06 cells for each a2-chimaerin cell line were aliquotted, re-centrifuged at 900rpm for 7 minutes at RT in a benchtop centrifuge (GP centrifuge, Beckman), PBS aspirated and the cell pellets stored at -70°C. Cell pellets were resuspended in lOOptl of Hypotonic Buffer (20mM Tris/HCl pH 7.0, lOmM KC1, 2mM PMSF) via sonication (sonicator cuphorn, Heat Systems). Samples were centrifuged at 100,000g for 1 hour at 4°C (TL-100 Ultracentrifuge, Beckman), the hypotonic supernatant fractions were removed and placed on ice. The

Materials and Methods 105 pellet fractions were washed twice with 15 Opt of Hypotonic Buffer and resuspended in lOOpl of CSK Buffer (140mM NaCl, lOmM Tris/HCl pH 7.5, 2mM PMSF, 1% Triton X-100) via sonication. Samples were incubated on ice for 5 minutes and then centrifuged at 15,000g for 10 minutes at 4°C to pellet the cytoskeleton. The CSK supernatants were removed and placed on ice, cell pellets were washed twice with 150pl of CSK Buffer and then resuspended in lOOul of 2x SDS sample buffer via sonication. Samples were then stored at -70°C. In order to analyse the samples, they were subjected to SDS-PAGE gel electrophoresis, western blotted onto nitrocellulose and probed with a monoclonal HA antibody (see sections 2.6.7 - 2.6.9).

2.6.4 Covalent coupling of HA antibody to protein A sepharose Protein A sepharose (Sigma) was resuspended in PBS and incubated with rolling overnight at 4°C. The volume of protein A beads required was washed in 0.1M Tris/HCl pH8.0 for 30 minutes at RT, then washed again overnight at 4°C. 5pg of HA antibody and lOOpl of protein A sepharose beads per immunoprecipitation sample were incubated in 25mls of 0.1M Tris/HCl pH8.0 for 1 hour at RT. The sample was centrifuged at 800rpm for 5 minutes at 4°C (J6, Beckman) to pellet the beads, the supernatant removed via aspiration and the beads washed twice in 10 volumes of 0.2M Na borate pH9.0 for 10 minutes at RT. The beads were then resuspended in a further 10 volumes of 0.2M Na borate pH9.0 and dimethylpimelimidate (DMP) coupling reagent added to give a final concentration of 20mM. The pH was adjusted to >8.3 and the sample incubated with rolling for 30 minutes at RT. The coupling reaction was stopped by washing the beads in 25mls of 0.2M ethanolamine pH8.0 for 10 minutes with rolling at RT. The sample was then centrifuged at 800rpm for 5 minutes at 4°C (J6, Beckman), the supernatant aspirated and the beads washed in a further 25mls of 0.2M ethanolamine pH8.0 for 2 hours at RT. After a final wash in PBS, the beads were stored at 4°C.

2.6.5 Immunoprecipitation from permanently transfected N1E 115 cell lines Permanently transfected N1E 115 cell lines were cultured and harvested as described in section 2.4.3. Cell pellets were resuspended in lOmls of PBS and counted using a haemocytometer. 25x106 cells for al-chimaerin, a2-chimaerin and empty pXJ41- HA vector cell lines were aliquotted, re-centrifuged at 900rpm for 7 minutes at RT in a

Materials and Methods 106 benchtop centrifuge (GP centrifuge, Beckman), PBS aspirated and the cell pellets stored at -70°C. Each cell pellet was resuspended in 400p,l of Immunoprecipitation Cell Lysis buffer (25mM HEPES pH7.3, 20mM P-glycerophosphate, 0.3M NaCl, 1.5mM MgCl2, 0.2mM EDTA pH8.0, 5% (w/v) glycerol, 1% (w/v) Triton X-100, 0.5% (w/v) Na deoxycholate adjusted to pH7.7, containing freshly added inhibitors ImM PMSF, ImM Na vanadate, 0.5mM DTT, 5|ig/ml pepstatin, 5pg/ml aprotinin), thoroughly lysed via passing through a 25G needle (Microlance) and centrifuged at 20,000g for 10 minutes at 4°C (Microcentrifuge 154, Camlab). Supernatants were removed promptly and placed on ice. The supernatants or ‘cell lysate’ fractions were then diluted 30 fold with ddH20 (containing ImM PMSF, ImM Na vanadate, 0.5mM DTT, 5pg/ml pepstatin, 5|ng/ml aprotinin), to give a final detergent concentration of 0.05%. 5 jig of HA antibody coupled to 100pl of protein A sepharose beads (see section 2.6.4) was added to each diluted cell lysate and samples were incubated with rolling at 4°C for 1.5 hours. Samples were centrifuged at l,000rpm for 5 minutes at 4°C (J6, Beckman) and the diluted cell lysates carefully removed via aspiration. The protein A sepharose beads were washed for 10 minutes in 5mls of wash buffer (PBS, 20mM P- glycerophosphate, 0.1% (w/v) Triton X-100, 2mM Na vanadate) per sample, centrifuged at l,000rpm for 5 minutes at 4°C (J6, Beckman) and the supernatants carefully removed via aspiration. Samples were washed 3 times. After the final wash, protein A sepharose beads were transferred to 2ml eppendorfs, centrifuged at 0.2g for 5 minutes at 4°C (Microcentrifuge 154, Camlab), the supernatants removed and 100pl of 2x SDS sample buffer (125mM Tris/HCl pH6.8, 40% (w/v) glycerol, 4% (w/v) SDS, 1% (v/v) p- mercaptoethanol, 0.1% (w/v) bromophenol blue) was added per sample. Samples were incubated at 100°C for 10 minutes to elute bound proteins from the beads and stored at - 20°C. In order to analyse samples, they were subjected to SDS-PAGE gel electrophoresis, western blotted onto nitrocellulose and probed with antibodies (see sections 2.6.7-2.6.9).

2.6.6 Determination of protein concentration fBiorad Microassay) 2-lOp.l aliquots of COS7 cell lysates (see section 2.6.1) were diluted in 800pl ddH20, 200j l l 1 of dye concentrate (Biorad) was added, samples were mixed by inversion

Materials and Methods 107 and incubated for 5 minutes at RT. The absorbance of samples at 595nm was then measured using a spectrophotometer. BSA (FractionV, Sigma) samples containing 1- 14pg/ml were used as standards and the protein concentration of the cell lysates was determined by comparison with the BSA standard graph.

2.6.7 SDS-PAGE gel electrophoresis (Mini Protean II system. Biorad) Separating gels (8.5cm x 5.5cm x 1.5mm thick) containing 7.5-12% (w/v) acrylamide (BDH, 30% acrylamide:0.8% bisacrylamide stock), 375mM Tris/HCl pH8.8, 0.1% (w/v) SDS, 0.05% (w/v) APS, 0.05% (v/v) TEMED were prepared and allowed to polymerise. Gels were overlaid with stacker gel containing 125mM Tris/HCl pH6.8, 4% acrylamide, 0.2% (v/v) TEMED, 0.1% (w/v) SDS, 0.05% (w/v) APS and combs inserted to form the sample wells. Protein samples were mixed with an equal volume of 2x SDS sample buffer (125mM Tris/HCl pH6.8, 40% (w/v) glycerol, 4% (w/v) SDS, 1% (v/v) (3-mercaptoethanol, 0.1% (w/v) bromophenol blue), or a 1/5 volume of 5x SDS sample buffer (312.5mM Tris/HCl pH6.8, 50% (w/v) glycerol, 10% (w/v) SDS, 2.5% (v/v) p-mercaptoethanol, 0.25% (w/v) bromophenol blue) to give a maximum final volume of 40pl. Samples and prestained molecular weight markers (Gibco/BRL or New England Biolabs, NEB) were denatured by heating at 100°C for 5 minutes, briefly centrifuged, then loaded into sample wells and electrophoresed at 180V for 40-50 minutes in lx SDS running buffer (192mM glycine, 25mM Tris, 0.1% (w/v) SDS).

2.6.8 Western blotting: protein transfer to nitrocellulose membranes Semi-dry Blotting (BioRad Trans Blot) After electrophoresis, SDS-polyacrylamide gels were incubated in lx Western Transfer buffer (48mM Tris, 39mM glycine, 20% (v/v) methanol, 0.038% (w/v) SDS) pre-chilled to 4°C, for 10 minutes at RT. Nitrocellulose filters (Schleicher and Schuell) cut to the size of the gels, 2 sheets of filter paper (grade 3, Whatman) cut to size and 2 sheets of thick blotting paper (BioRad) were all soaked in lx Western Transfer buffer. A stack containing a sheet of thick blotting paper, filter paper, nitrocellulose filters, gels, filter paper and thick blotting paper, was placed on the platinum anode of the Trans-Blot (Biorad), air bubbles were displaced and the steel cathode lid placed on top of the stack. Proteins were transferred onto nitrocellulose at a constant voltage of 15 V for 1.2 hours

Materials and Methods 108 2.6.9 Immunodetection of proteins Nitrocellulose filters were incubated overnight at 4°C, in 5% (w/v) dried skimmed milk powder (Marvel) in PBS to block non-specific protein binding sites. The next day, filters were incubated with primary antibodies - 1 pg/ml mouse monoclonal HA antibody 12CA5 (Boehringer Mannheim), 1:1000 dilution of mouse monoclonal GFP antibody (Clontech), 0.2|ag/ml mouse monoclonal phosphotyrosine (PY99) antibody (Santa Cruz) or 1:500 dilution of rabbit polyclonal p35 (C-19) antibody (Santa Cruz), diluted in 1% (w/v) dried skimmed milk powder (Marvel) in PBS, for 1 hour at room temperature. After 3x 10 minute washes in PBST (PBS, 0.1% (w/v) Tween-20), filters were incubated with secondary antibodies - 1:1000 dilution of peroxidase conjugated rabbit anti-mouse IgG antibody or 1:1000 dilution peroxidase conjugated swine anti­ rabbit IgG antibody (Dako Patts) diluted in 1% (w/v) dried skimmed milk powder (Marvel) in PBS, for 1 hour at room temperature. Finally, filters were washed for 3x 10 minutes in PBST and rinsed in PBS. Immunoreactive bands were detected using enhanced chemiluminescence (ECL) reagents and film (Amersham). Briefly, equal volumes of ECL reagents A and B were mixed together, filters were incubated for 1 minute in this solution at RT, covered in Saran wrap (Dow) and exposed to Hyperfilm-ECL for 10 seconds to 30 minutes.

2.6.10 Immunocvtochemistrv At 24 hours, transfection solutions were aspirated, cells were washed once in PBS and fixed with 3% paraformaldehyde in PBS for 20 minutes at RT. Cells were washed with PBS for 10 minutes and free paraformaldehye groups were quenched by incubation in lOOmM glycine in PBS for 10 minutes at RT. Cells were then permeabilised in 0.2% (w/v) Triton X-100 in PBS for 10 minutes at RT, rinsed once in PBS and blocked in 3% (w/v) BSA (FractionV, Sigma) in PBS for 10 minutes. Samples were then incubated with primary antibodies diluted in 1% (w/v) BSA (FractionV, Sigma) in PBS for 1 hour at 37°C. Primary antibodies: 5 p,g/ml HA monoclonal antibody (12CA5, Boehringer Mannheim) or 1:50 dilution of rabbit polyclonal p35 (C19) antibody (Santa Cruz Biotechnology). Cells were washed for 3x 5 minutes in PBS and then incubated with secondary antibodies and lug/ml rhodamine conjugated phalloidin (1:100 dilution from a stock of

Materials and Methods 109 0.1 mg/ml in 100% methanol, Sigma) diluted in 1% (w/v) BSA (FractionV, Sigma) in PBS for 1 hour at 37°C. Secondary antibody dilutions: 1:100 dilution of FITC conjugated rabbit anti mouse IgG antibody or FITC conjugated swine anti rabbit IgG antibody (Dako Patts). Cells were then washed for 3x 5 minutes in PBS and mounted in mowiol/DABCO (10% (w/v) mowiol 40-8B (Sigma), lOOmM Tris/HCl pH8.0, 25% (w/v) glycerol, 2.5% DABCO (l,4-diazobicyclo-[2.2.2]-octane, Sigma)) or Immunofluore mountant (ICN) and incubated overnight at RT in the dark. Samples were examined using a fluorescence microscope (Zeiss) or confocal (Zeiss).

2,7 Reporter assays

2.7.1 NFkB reporter assay - NIE 115 and HeLa cells

The luciferase coupled NFk B reporter vectors were generated by JM Dong (IMCB, Singapore) from the pGL2-basic vector (Promega) (see appendix 4). The TK promoter sequence was inserted at the Hind III site and four copies of the functional or mutated NFk B binding site derived from the MHC promoter sequence were inserted at the Xhol site of the multiple cloning site to produce the functional NFk B and mutated

NFk B(M) reporter vectors, respectively. The Smal site was deleted in both vectors. When N IE 115 cells were co-transfected with the P-galactosidase reporter DNA, cells were harvested in the p-gal/NFKB Lysis Buffer (lOOmM K2HPO4/KH2PO4 pH7.8, 0.2% (w/v) Triton X-100, ImM DTT). However, when this reporter vector was not being used, N IE 115 cells were harvested in NFkB Cell Lysis buffer (50mM

Na2HP0 4 /NaH2P0 4 pH7.8, 0.1% (w/v) Triton X-100, ImM DTT). All HeLa cell samples were harvested in NFkB Cell Lysis buffer (50mM Na 2HP0 4 /NaH2P0 4 pH7.8, 0.1% (w/v) Triton X-100, ImM DTT). Transfection solutions were removed by aspiration from transiently transfected NIE 115 and HeLa cells (see sections 2.5.2b and 2.5.3) at 24 hours. Cells were washed once with PBS and incubated in 550pl of appropriate cell lysis buffer for 5 minutes at RT. Cells were harvested by scraping, thoroughly lysed via passing through a 25G needle (Microlance) and centrifuged at 15,000g for 5 minutes at 4°C (microcentrifiige 5415c, Eppendorf). Supernatants were removed promptly, placed on ice and assayed for luciferase activity using a luminometer. N IE 115 and HeLa cell lysates were assayed (twice per triplicate sample) for

Materials and Methods 110 luciferase activity using a luminometer (model 1251, LKB Wallac) in a final reaction volume of lml. 100^il of cell lysate was diluted to 400pi with the appropriate lysis buffer (see above) and 400pl reaction buffer (125mM Na 2HP0 4 /NaH2P0 4 pH7.8, 25mM MgCl2, 3.4mM ATP) was then added. Samples were mixed, 200pl of 0.28mg/ml luciferin (Sigma) solution (diluted in 5mg/ml BSA in 50mM Na2HP04/NaH2P0 4 pH7.8) was injected and the luciferase activity measured using a luminometer. Lysates of NIE 115 cells co-transfected with the p-galactosidase reporter DNA were also assayed for p-galactosidase activity (see section 2.7.3).

2.7.2 Processing of NFkB assay data Measurement by the luminometer was in arbitrary units and the sample values generated varied considerably between experiments thus standardisation was required to enable comparison between experiments. Each triplicate sample was assayed twice and the 2 values per plate were averaged, resulting in 1 value per plate and a total of 3 values per sample. The 'overall average' obtained for the empty HA vector was calculated. Each individual sample value was divided by the 'overall average' obtained for the empty HA vector in order to standardise results. This resulted in a baseline value of 1.0 for the empty HA vector (which does not stimulate luciferase activity) and the fold stimulation of each sample was relative to this value. Standardised results from several experiments were combined and the overall average and standard deviation for samples was then calculated. This final data is presented in figures 5.1-5.10 and the n number associated with these results corresponds to the number of sample values used to calculate them. The data used to calculate the final results is presented in appendices 5-10.

2.7.3 Luminescent B-galactosidase detection kit N IE 115 cells were harvested in 550pil of P-gal Lysis Buffer (lOOmM K2HP04/KH2P0 4 pH7.8, 0.2% (w/v) Triton X-100, ImM DTT) and cell lysates were diluted lOOx in ddH20. lpl of diluted N IE 115 cell lysate was added to 100pl of Reaction Buffer (containing 0.2mM Galacton-Star substrate, 1 mg/ml Sapphire-II enhancer, Clontech) and (3-galactosidase activity was measured using a scintillation counter. Samples were read at a zero timepoint and results were used to correct for variations in transfection efficiency between samples. It was found using this method, that the small variations in transfection efficiency which occurred had little effect on

Materials and Methods 111 overall sample values.

Materials and Methods CHAPTER THREE: Results I

Chapter 3 CHAPTER 3: Distribution of HA- and GFP-tagged proteins in eukaryotic cells

To investigate the expression of chimaerin isoforms and potential targets in mammalian cell lines, cDNA sequences were cloned into haemagglutinin (HA) and green fluorescent protein (GFP) epitope tagged eukaryotic expression vectors (see Methods 2.3). These vectors direct expression of proteins with an N terminal HA or GFP tag, which are recognised by commercially available monoclonal antibodies. The HA tag is a 10 amino acid sequence derived from the haemagglutinin protein of the human influenza virus (figure 2.1b), whilst the GFP tag encodes a -28 kDa fluorescent protein isolated from the jellyfish Aequorea Victoria (figure 2.2c). The advantages of GFP-tagged proteins are that no staining is required to detect expression of these proteins via immunocytochemistry and protein expression can be studied in live cells in real time via fluorescence microscopy. However, due to its size it is possible that GFP may interfere with protein targeting. The small HA tag is commonly used and has no targeting function, thus comparison between the localisation of HA- and GFP-tagged proteins was used to assess whether GFP adversely affected chimaerin protein targeting.

3.1 Protein expression from HA- and GFP-tagged DNA constructs in an in vitro transcription-translation assay cDNA sequences were cloned into the eukaryotic expression vectors pXJ41-HA, pXJ40-HA and pXJ40-GFP (see Methods 2.3). Restriction analysis showed the HA- and GFP-tagged DNA constructs all contained inserts of the correct size in the correct orientation. However to ensure that these DNA constructs functioned correctly and produced full length proteins they were subjected to an in vitro transcription-translation assay (Promega) (see Methods section 2.2.22). Figures 3.1 and 3.2 show the autoradiograms of the resultant35 S labelled proteins.

3.1.1 Protein expression from pXJ41-HA and pXJ40-HA DNA constructs Protein expression from pXJ41-HA and pXJ40-HA DNA constructs which enable permanent transfection via selection with a neomycin resistance gene or transient transfection of eukaryotic cells respectively, are shown in figure 3.1.35S labelled protein bands were observed at -38 kDa for ocl-chimaerin (lanes 1 and 3), -45 kDa for a.2- chimaerin and the four a2-chimaerin SH2 domain mutants (lanes 2 and 4-8) and -65 kDa for TOAD-64 (lane 9), which are the expected sizes for full length HA-tagged proteins. The lower band in lane 9 is probably a breakdown product or the protein

Chapter 3 114 12 3 4 5 6 7 8 9 10 11 kDa 206

105 71

— - 28

18

Figure 3.1: Protein expression from HA-tagged DNA constructs in an in vitro transcription-translation assay 0.5pg of each DNA construct was tested in the TNT in vitro Transcript ion- Translation Assay (Promega) in a final reaction volume of 25pi (see section 2.2.22). The resultant °S labelled proteins were separated by SDS-PAGE, gels were incubated for 30 minutes in destain, 30 minutes in Amplify scintillant (Amersham), dried down and then exposed to X ray film. The expected sizes for full length HA-tagged protein products are -38 kDa for a 1-chimaerin, -45 kDa for a2-chimaerin and -64 kDa for TOAD-64. Luciferase is -61 kDa.

Lane 1: 5pl pXJ41 HA a 1-chimaerin, lane 2: 5pl pXJ41HA a2-chimaerin, lane 3: 5pl pXJ40HA a 1-chimaerin, lane 4: 5pl pXJ40HA a2-chimaerin, lane 5: 5pl pXJ40HA a2-chimaerin (E49W), lane 6: 5pl pXJ40HA a2-chimaerin (R56L), lane 7: 5pl pXJ40HA a2-chimaerin (R73L), lane 8: 5pl pXJ40HA a2-chimaerin (N94H), lane 9: 5pl pXJ40HA TOAD-64, lane 10: 5pl negative control (no DNA in reaction), lane 11: 5pl luciferase positive control

Figure 3 .1 115 1 23456789 kDa 206

105

44

Figure 3.2: Protein expression from GFP-tagged DNA constructs in an in vitro transcription-translation assay 0.5pg of each DNA construct was tested in the TNT in vitro Transcript ion- Translation Assay (Promega) in a final reaction volume of 25pl (see section 2.2.22). The resultant ,5S labelled proteins were separated by SDS-PAGE, gels were incubated for 30 minutes in destain, 30 minutes in Amplify scintillant (Amersham), dried down and then exposed to X ray film. The expected sizes for full length green fluorescent protein (GFP) tagged protein products are -66 kDa for a 1-chimaerin, -73 kDa for a2- chimaerin and -92 kDa for TOAD-64. Luciferase is -61 kDa.

Lane 1: 5pl pXJ40GFP a 1-chimaerin, lane 2: 5pl pXJ40GFP a2-chimaerin, lane 3: 5pl pXJ40GFP a2-chimaerin (E49W), lane 4: 5pl pXJ40GFP a2-chimaerin (R56L), lane 5: 5pl pXJ40GFP a2-chimaerin (R73L), lane 6: 5pi pXJ40GFP a2-chimaerin (N94H), lane 7: 5pl pXJ40GFP TOAD-64, lane 8: 5pl negative control (no DNA in reaction), lane 9: 5pl luciferase positive control

Figure 3.2 product from an internal initiation site or premature termination. Controls showed no labelled proteins in the absence of DNA and a -60 kDa luciferase protein synthesised from control DNA (Promega) (lanes 10 and 11 respectively).

3.1.2 Protein expression from pXJ40-GFP DNA constructs Protein expression from pXJ40-GFP DNA constructs which enable transient transfection of eukaryotic cells are shown in figure 3.2.35S labelled protein bands were observed at -66 kDa for GFP-tagged a 1-chimaerin (lane 1), -73 kDa for GFP-tagged a2-chimaerin and the four a2-chimaerin SH2 domain mutants (lanes 2-6) and -92 kDa for GFP-tagged TOAD-64 (lane 7). These are the expected sizes for full length GFP- tagged proteins as GFP has a molecular weight o f-28 kDa. Controls showed no labelled proteins in the absence of DNA and a -60 kDa luciferase protein synthesised from control DNA (Promega) (lanes 8 and 9 respectively). Minor bands present in lanes 1-7 are probably breakdown products or the protein products from internal initiation sites or premature termination.

3.2 Expression of HA- and GFP-tagged a l- and a2-chimaerin in COS7 cells Protein expression and distribution was investigated in COS7 cells as they are able to express high levels of protein from transiently transfected DNA. COS7 cells were transiently transfected, harvested at 24 hours in buffer containing 0.5% Triton X- 100 and protein expression in the resultant cell lysate and pellet fractions was analysed (see Methods sections 2.6.1 and 2.6.7-2.6.9).

3.2.1 Expression of HA-tagged chimaerin DNA constructs in COS7 cells Expression and distribution of HA-tagged chimaerin proteins in COS7 cell lysates was investigated. Full length HA-tagged a2-chimaerin and a2-chimaerin SH2 domain mutant E49W, R56L and R73L proteins (-45 kDa) were detected in the 0.5% Triton X-100 soluble fraction of COS 7 cells (figure 3.3 A, lanes 1-4). The levels of expression of the E49W, R56L and R73L SH2 domain mutants detected were slightly higher than that of a2-chimaerin. However, these four proteins were not detected in the insoluble fraction of COS7 cells (figure 3.3B, lanes 1-4). As the volume of insoluble fraction loaded on gels was 3 fold higher than the soluble fraction, the majority of protein (>95%) was in the 0.5% Triton X-100 soluble fraction. However the N94H mutant had a different distribution in COS7 cells. It was more weakly detected in the 0.5% Triton X-100 soluble fraction than the other a2-chimaerin proteins, but very

Chapter 3 117 kDa 105

H A -a 2

---- 28

1 2 3 4 5 6

B kDa 105

H A -a2 -— 44

---- 28

---- 18

Figure 3.3; Expression of HA-tagged a2-chimaerin proteins in COS7 cells COS7 cells were transiently transfected with pXJ40HA a2-chimaerin DNA constructs, harvested at 24 hours in 400pl of buffer containing 0.5% Triton X-100 and cell pellets were resuspended in 50pl 2x SDS sample buffer (see section 2.6.1). 25pg of 0.5% Triton X-100 soluble lysate and lOpl of 0.5% Triton X-100 insoluble pellet were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with the monoclonal HA antibody (lpg/ml, Boehringer Mannheim) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected size for full length HA-tagged oc2-chimaerin is -45 kDa.

A: 25pg cell lysate B: lOpl cell pellet Lane 1: HA-tagged a2-chimaerin, lane 2: HA-tagged a2-chimaerin (E49W), lane 3: HA-tagged a2-chimaerin (R56L), lane 4: HA-tagged a2-chimaerin (R73L), lane 5: HA-tagged a2-chimaerin (N94H), lane 6: untransfected COS7 cells

Figure 3.3 118 strongly detected in the insoluble fraction where no expression of the other a2- chimaerin proteins was detected (figure 3.3A and 3.3B, lane 5). Approximately 10 fold more N94H mutant protein was detected in the insoluble fraction than the soluble fraction and 3 fold more insoluble fraction was loaded on the gel. Thus overall, the majority of HA-tagged N94H mutant protein (-75%) was distributed in the 0.5% Triton X-100 insoluble fraction of COS7 cells. The 70 kDa and 105 kDa bands in the soluble fraction of untransfected COS7 cells (figure 3.3A, lane 6) and the 20 kDa and 35 kDa bands in the insoluble fraction of untransfected COS7 cells (figure 3.3B, lane 6) are all non-specific background bands picked up by the HA antibody which are present in all the samples. Expression of HA- tagged a 1-chimaerin protein could not be detected by Western analysis using the anti HA antibody.

3.2.2 Expression of GFP-tagged chimaerin DNA constructs in COS7 cells Expression and distribution of GFP-tagged a l- and oc2-chimaerin proteins in COS7 cell lysates was investigated (figure 3.4). Full length GFP-tagged al-chimaerin (-66 kDa) was detected in the 0.5% Triton X-100 insoluble fraction (figure 3.4B, lane 2) but not the soluble fraction (figure 3.4A, lane 2) of COS7 cells. Whilst full length GFP-tagged a2-chimaerin and a2-chimaerin SH2 domain mutants E49W, R56L, R73L and N94H (-73 kDa) were detected in both the 0.5% Triton X-100 soluble and insoluble fraction of COS7 cells (figure 3.4A and 3.4B, lanes 3-7). However considerably more N94H mutant protein was detected in the insoluble fraction of COS7 cell lysates than the other a2-chimaerin proteins (figure 3.4B, compare lanes 7 and 3-6). The volume of insoluble fraction shown in figure 3.4 is 5 fold higher than the soluble fraction. Considering the relative volumes present, the majority (>95%) of al- chimaerin protein was detected in the 0.5% Triton X-100 insoluble fraction which contains cytoskeletal proteins. This distribution of a 1-chimaerin agrees with previous data, where a 1-chimaerin was detected in the cytoskeleton associated fraction of CSK extracted COS7 cells (Kozma et al., 1996). The distribution of a2-chimaerin and the E49W, R56L and R73L SH2 domain mutants differed greatly from that of a l- chimaerin, as approximately 3-5 fold more protein was detected in the soluble than the insoluble fraction. Considering the relative volumes present, the majority of a2- chimaerin and E49W, R56L and R73L mutant proteins (>95%) were distributed in the 0.5% Triton X-100 soluble fraction of COS7 cells which contains both cytosolic and

Chapter 3 119 1 2 3 4 5 6 7 8

kDa 206

105 G FP-a2 71

44

GFP 28

1 2 3 4 5 6 7 8 B kDa 206

105 G FP-a2 G F P -al — 71

44

GFP

Figure 3.4: Expression of GFP-tagged al- or a2-chimaerin proteins in COS7 cells COS7 cells were transiently transfected with pXJ40GFP a l - or a2-chimaerin DNA constructs, harvested at 24 hours in 400pl of buffer containing 0.5% Triton X-100 and cell pellets were resuspended in 50pl 2x SDS sample buffer (see section 2.6.1). 25pg of 0.5% Triton X-100 soluble lysate and lOpl of 0.5% Triton X-100 insoluble pellet were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with the monoclonal GFP antibody (1:1000 dilution, Clontech) (see sections 2.6.7- 2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected sizes for full length GFP-tagged protein products are ~66 kDa for a l- chimaerin and ~73 kDa for a2-chimaerin.

A: 25pg cell lysate B: lOpI cell pellet Lane 1: empty GFP vector, lane 2: GFP-tagged a 1-chimaerin, lane 3: GFP-tagged a2-chimaerin, lane 4: GFP-tagged a2-chimaerin (E49W). lane 5: GFP-tagged a2-chimaerin (R56L), lane 6: GFP-tagged a2-chimaerin (R73L), lane 7: GFP-tagged a2-chimaerin (N94H), lane 8: untransfected COS7 cells

Figure 3.4 120 membrane associated proteins. In contrast to a2-chimaerin and the E49W, R56L and R73L mutants, approximately 10 fold more N94H mutant protein was detected in the insoluble than the soluble fraction. Considering the relative volumes present, the majority ofN94H protein (-66%) was expressed in the 0.5% Triton X-100 insoluble fraction of COS7 cells. Thus a single N94H point mutation in the SH2 domain of a2- chimaerin produces a protein distribution in COS7 cells which more closely resembles that of a 1-chimaerin than a2-chimaerin. The 105 kDa and 165 kDa bands in the soluble fraction of untransfected COS7 cells (figure 3.4A, lane 8) and the 55 kDa band in the insoluble fraction (figure 3.4B, lane 8) are all non-specific background bands picked up by the GFP antibody which are present in all the samples. Approximately twice as much full length GFP protein (-32 kDa) was detected in the 0.5% Triton X-100 insoluble fraction as in the soluble fraction of COS 7 cells (figure 3.4 A and 3.4B, lane 1). Thus the majority of GFP protein (-70%) was distributed in the 0.5% Triton X-100 soluble fraction of COS7 cells.

3.2.3 Effects of GFP tagging on the distribution of chimaerin proteins The distribution of HA- and GFP-tagged chimaerin proteins in COS7 cell lysates is shown in figures 3.3 and 3.4 respectively. Both HA- and GFP-tagged a2-chimaerin and SH2 domain mutants E49W, R56L and R73L mutant proteins were predominantly detected (>95%) in the 0.5% Triton X-100 soluble fraction of COS7 cells. However the N94H mutant had a different protein distribution, -75% of the HA-tagged protein was detected in the 0.5% Triton X-100 insoluble fraction, compared to -66% of the GFP- tagged protein. Comparing these results, there is good agreement between the protein distributions ofHA- and GFP-tagged a2-chimaerin proteins. However, GFP-tagged but not HA-tagged a2-chimaerin and the E49W, R56L and R73L SH2 domain mutant proteins were also weakly detected in the insoluble fraction of COS7 cells (compare figure 3.3B, lanes 1-4 and figure 3.4B, lanes 3-6). This may be due to different levels of expression of HA- and GFP-tagged constructs or differences in the sensitivity of the HA and GFP antibodies. Thus it seems that the GFP tag does not appreciably affect the targeting of these proteins and either set of constructs may be used. The lack of detection of HA-tagged a 1-chimaerin in COS7 cell lysates by Western analysis using the monoclonal HA antibody does not necessarily mean that the protein was not expressed, as full length HA-tagged a 1-chimaerin was produced in an in vitro transcription-translation assay (figure 3.1, lane 3). This lack of detection may be

Chapter 3 121 due to the protein folding in such a manner as to cause obstruction of the HA antibody binding site.

3.3 Expression of a l- and q2-chimaerin in permanent NIE 115 cell lines Permanent NIE 115 neuroblastoma cell lines expressing a l - or a2-chimaerin were established to investigate the effects of long term chimaerin overexpression on cell morphology and potential protein interactions. Production of these permanent NIE 115 cell lines by neomycin selection is described in Methods section 2.5.4. The distribution of chimaerin in the hypotonic supernatant, CSK supernatant and CSK pellet extraction fractions, which correspond to soluble, triton-solubilised membrane and triton-insoluble or cytoskeleton-associated fractions of these cell lines was analysed by Western immunoblotting (see Methods section 2.6.3 for CSK extraction).

3.3.1 Expression of a2-chimaerin in permanent NIE 115 cell lines a2-Chimaerin protein was detected by the polyclonal a2-chimaerin antibody in both the soluble and triton-solubilised membrane fractions of permanent a2-chimaerin cell lines (figure 3.5), but not in the cytoskeletal fraction. A faint band at -45 kDa corresponds to the endogenous level of a2-chimaerin protein in the soluble fraction of untransfected NIE 115 cells and an empty pXJ41-HA vector cell line (lanes 1 and 4). The a2-10 cell line showed a -20 fold stronger 45 kDa band in the soluble fraction due to expression of exogenous a2-chimaerin (lane 7). In untransfected, empty vector and a2-10 cell lines, a2-chimaerin was weakly detected in the triton-solubilised membrane fraction (lanes 2, 5 and 8 respectively) but not in the triton-insoluble fraction (lanes 3, 6 and 9). Thus a2-chimaerin is a mainly soluble protein with a smaller amount associated ! with membranes and overexpression of a2-chimaerin in the a2-10 cell line did not disrupt this protein localisation, but merely increased the total amount of protein ] present.

3.3.2 Selection of permanent NIE 115 cell line expressing high levels of a2- chimaerin Since a2-chimaerin protein was detected at highest levels in the soluble fraction of NIE 115 cell lines, this fraction of all a2-chimaerin cell lines was analysed in order to select the highest expressing cell line (figure 3.6). A faint band at -45 kDa corresponds to the endogenous level of a2-chimaerin protein detected in untransfected

Chapter 3 122 kDa 175 83 62 47.5

32.5

25

Figure 3.5: Distribution of a2-chimaerin protein in permanently transfected NIE 115 cells For each cell line 1.5x107 cells were harvested and subjected to CSK extraction in lOOpl of buffer (see section 2.6.3). 30pl of hypotonic supernatant, 30pl of CSK supernatant and 20pl of CSK pellet fractions were subjected to SDS-PAGE gel electrophoresis, Western blotted onto nitrocellulose and probed with a polyclonal a2- chimaerin antibody (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected size for full length a2-chimaerin protein is -45 kDa.

Lane 1: 30pl untransfected N IE 115 cells HS, lane 2: 30pl untransfected NIE 115 cells CS, lane 3: 20pl untransfected NIE 115 cells CP, lane 4: 30pl HAv-24 empty vector HS, lane 5: 30pl HAv-24 empty vector CS, lane 6: 20pl HAv-24 empty vector CP, lane 7: 30pl a2-10 cell line HS, lane 8: 30pl a2-10 cell line CS, lane 9: 20pl a2-10 cell line CP where HS represents hypotonic supernatant, CS represents CSK supernatant and CP represents CSK pellet

Figure 3.5 123 1 2 3 4 5 6 7 8 9 10 11 12 13 kDa 175 83 62 47.5

32.5

25

Figure 3.6: Levels of a2-chimaerin expression in permanently transfected N1E 115 cells For each cell line, 1.5x107 cells were harvested and subjected to CSK extraction in lOOpl of buffer (see section 2.6.3). 30pl of hypotonic supernatant fractions were subjected to SDS-PAGE gel electrophoresis, Western blotted onto nitrocellulose and probed with a polyclonal a2-chimaerin antibody (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected size for full length a2-chimaerin protein is ~45 kDa.

Lane 1: untransfected N 1E 115 cells, lane 2: HAv-24 empty vector, lane 3: a2-6, lane 4: a2-7, lane 5: a2-8, lane 6: a2-9, lane 7: a2-10. lane 8: a 2 -l 1, lane 9: a2-13, lane 10: a2-20, lane 11: a2-22, lane 12: a2-28. lane 13: a2-29

Figure 3.6 124 N1E 115 cells and an empty pXJ41-HA vector cell line (lanes 1 and 2). All a2- chimaerin cell lines showed much stronger 45 kDa bands due to expression of exogenous a2-chimaerin (lanes 3-13). Most cell lines expressed high levels of a2- chimaerin and as ot2-10 was one of the highest expressors, this cell line was chosen for further characterisation. The exogenous a2-chimaerin was also detected with the HA antibody (data not shown). However the level of background detected by the HA antibody on Western analysis of these samples was very high and the a2 antibody provided a much clearer result.

3.3.3 Expression of al-chimaerin in permanent N1E 115 cell lines HA-tagged al-chimaerin protein was detected by the HA antibody in the triton- solubilised membrane and triton-insoluble fractions of permanent al-chimaerin cell lines (figure 3.7). A band at -38 kDa which is the expected size for full length HA- tagged al-chimaerin was present in the triton-insoluble (cytoskeleton-associated) fraction of the alD , al-7, al-8 and al-10 cell lines (figure 3.7C, lanes 4-7). The a lD cell line also appeared to contain al-chimaerin in the triton-solubilised membrane fraction (figure 3.7B, lane 4), however no protein was detected in the al-7 , a l-8 and al-10 cell lines in this fraction (figure 3.7B, lanes 5-7). In the soluble fraction, the HA antibody detected a faint band at approximately the same size as full length a l- chimaerin in untransfected N1E 115 cells and an empty pXJ41-HA vector cell line (figure 3.7A, lanes 2 and 3), -38 kDa bands were also present in the al-chimaerin cell lines (lanes 5-7). Due to the background in this fraction, al-chimaerin protein cannot be detected with any certainty by the HA antibody. In order to confirm the presence of a l- chimaerin protein in this fraction a good al-chimaerin antibody is required, which is not available at present. Full length HA-tagged al-chimaerin protein was only weakly detected by the HA antibody in permanent N1E 115 cell lysates and not in transiently transfected cells. In the a lD cell line, al-chimaerin was detected in both the triton-solubilised membrane and cytoskeleton-associated fractions, whereas in the al-7 , al-8 and al-10 cell lines al-chimaerin was only detected in the cytoskeleton-associated fraction. This difference in protein localisation between a lD and the other al-chimaerin expressing cell lines may be partly responsible for their differences in morphology (see chapter 6).

Chapter 3 125 1 2 3 4 5 6 7 kDa 83 62 47.5 32.5

1 2 3 4 5 6 7 kDa B 83 62 47.5

32.5

1 2 3 4 5 6 7 kDa C Hw 83 62 47.5

32.5

Figure 3.7: Distribution of al-chimaerin protein in permanently transfected N1E 115 cells For each cell line 2.5x107 cells were harvested and subjected to CSK extraction in lOOjal of buffer (see section 2.6.3). 20pl of hypotonic supernatant, CSK supernatant and CSK pellet fractions were subjected to SDS-PAGE gel electrophoresis, Western blotted onto nitrocellulose and probed with a monoclonal HA antibody (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected size for full length HA tagged al-chimaerin protein is -38 kDa.

A: Hypotonic supernatant, B: CSK supernatant and C: CSK pellet fractions Lane 1: 25pg a2-chimaerin (R73L) COS7 cell lysate, lane 2: 20pl untransfected N 1E 115 cells, lane 3: 20pl HAv-24 empty vector, lane 4: 20pl a ID, lane 5: 20pl a l-7 , lane 6: 20 pi a l-8 . lane 7: 20pl a l-1 0

Figure 3.7 126 3.3.4 Levels of endogenous a2-chimaerin expression in permanent N1E 115 cell lines expressing al-chimaerin It was found that the a lD cell line had a morphology which differed from the al-7 , a l-8 and al-10 cell lines, but was very similar to the a2-10 cell line (see chapter 6). Thus the levels of a2-chimaerin expression in these al-chimaerin expressing cell lines was investigated. As a2-chimaerin was most strongly detected in the soluble fraction of N1E 115 cell lines (figure 3.5), this fraction of al-chimaerin expressing cell lines was analysed for a2-chimaerin protein expression using the polyclonal a2- chimaerin antibody (figure 3.8). Overexpression of a2-chimaerin in the a2-10 cell line is shown in lane 1 which contains a strong -45 kDa band. All al-chimaerin expressing cell lines (lanes 3-6) showed levels of a2-chimaerin expression which were similar to the endogenous level present in untransfected N1E 115 cells and an empty pXJ41-HA vector cell line (lanes 2 and 7). Thus the similarities between the a ID and a2-10 cell lines and the differences between alD and the other a l lines, is not due to upregulation of a2-chimaerin expression in a ID. This suggests that in the a lD line, the a l- chimaerin sequence may have been mutated or a small part deleted, or the regulation of al-chimaerin has been altered causing it to act like a2-chimaerin in the a2-10 cell line or perhaps a compensatory mechanism such as up regulation of another gene may have occurred.

3.4 Expression of a2-chimaerin targets in COS7 cells Two potential targets of a2-chimaerin have previously been identified. A screen of rat brain extracts for a2-chimaerin targets identified a ~65kDa protein which was partly purified by chromatographic techniques and its enrichment monitored during purification by overlay binding assay (M. Teo PhD thesis, 1994). The protein was later identified on the basis of peptide sequence as TOAD-64, a phosphoprotein involved in axonal guidance (Minturn et al., 1995a, b). The second potential target was isolated in a yeast 2 hybrid screen. This 13 kDa protein was identified as the B13 subunit of the inner mitochondrial membrane NADH ubiquinone oxidoreductase, the first and largest enzyme of the mitochondrial respiratory chain (C. Monfries, Personal Communication). In order to further investigate the interaction of these proteins with a2-chimaerin, HA- and GFP-tagged DNA constructs encoding these proteins were made and their expression and distribution were investigated.

Chapter 3 127 1 2 3 4 5 6 7 kDj 175 83 62 47.5

32.5 25

Figure 3.8: Endogenous a2-chimaerin expression in permanently transfected N1E 115 cells overexpressing al-chim aerin For each al-chimaerin cell line 2.5x107 cells were harvested and subjected to CSK extraction in lOOpl of buffer (see section 2.6.3). 10pl of hypotonic supernatant fractions were subjected to SDS-PAGE gel electrophoresis, Western blotted onto nitrocellulose and probed with a polyclonal a2-chimaerin antibody (see sections 2.6.7- 2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected size for full length a2-chimaerin protein is ~45 kDa.

Lane 1: a2-10. lane 2: untransfected N 1E 115 cells. lane 3: a ID, lane 4: a l-7 , lane 5: a l-8 . lane 6: a l-1 0 , lane 7: HAv-24 empty vector

Figure 3.8 128 3.4.1 Expression of HA- and GFP-tagged B13 and TOAD-64 DNA constructs in COS7 cells Expression and distribution of GFP-tagged B 13 and TOAD-64 proteins in COS7 cell lysates was investigated (figure 3.9). Full length GFP-tagged B 13 protein (-41 kDa) was detected in both the 0.5% Triton X-100 soluble and insoluble fractions of COS7 cells (figure 3.9, lanes 3 and 4). Approximately 15 fold more protein was detected in the insoluble than the soluble fraction of COS7 cells (figure 3.9, compare lanes 4 and 3). As the volume of insoluble fraction loaded on the gel was 10 fold higher than the soluble fraction, overall the majority of GFP-tagged B 13 protein (-60%) was distributed in the 0.5% Triton X-100 insoluble fraction of COS7 cells. Full length GFP-tagged TOAD-64 protein (-92 kDa) was detected in the 0.5% Triton X-100 insoluble fraction (figure 3.9, lane 6) but not the soluble fraction (figure 3.9, lane 5) of COS7 cells. The volume of insoluble fraction loaded on the gel was 5 fold higher than the soluble fraction, thus the majority of TOAD-64 protein (>95%) was detected in the 0.5% Triton X-100 insoluble fraction. However it is possible that low levels of this protein may be detected in a larger volume of soluble fraction. Approximately 3 fold more full length GFP protein (-32 kDa) was detected in the 0.5% Triton X-100 insoluble than the soluble fraction of COS7 cells (figure 3.9, lanes 1 and 2). As the volume of insoluble fraction loaded on the gel was 10 fold higher than the soluble fraction, overall the majority of GFP protein (-77%) was distributed in the 0.5% Triton X-100 soluble fraction of COS7 cells. This agrees well with the distribution of GFP observed in section 3.2.2. The 105 kDa and 165 kDa bands in the soluble fraction of untransfected COS7 cells (figure 3.9, lane 7) and the -55 kDa band in the insoluble fraction (figure 3.9, lane 8) are all non-specific background bands picked up by the GFP antibody which are present in all the samples. Expression of HA- tagged B 13 and TOAD-64 proteins could not be detected by Western analysis using the monoclonal HA antibody. ~

3.4.2 Detection of HA- and GFP-tagged proteins by the B13 and TOAD-64 antibodies Recently produced rabbit polyclonal antisera raised against B13 and TOAD-64 were each shown to recognise the appropriate recombinant protein (G.Ferrari PhD thesis, 1999). However their detection of proteins produced from a eukaryotic system had not been characterised. GFP-tagged B13 and TOAD-64 can be detected by the GFP antibody (figure 3.9), but neither of these HA-tagged proteins can be detected by the

Chapter 3 129 1 2 3 4 5 6 7 8

kDa 206

105

71

---- 28

Figure 3.9: Expression of GFP-tagged TOAD-64 or B13 proteins in COS7 cells COS7 cells were transiently transfected with pXJ40GFP B13 and TOAD64 DNA constructs, harvested at 24 hours in 400pl of buffer containing 0.5% Triton X-100 and cell pellets were resuspended in 50pl 2x SDS sample buffer (see section 2.6.1). 25pg of0.5% Triton X-100 soluble lysate and lOpl of 0.5% Triton X-100 insoluble pellet were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with the monoclonal GFP antibody (1:1000 dilution, Clontech) (see sections 2.6.7- 2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected sizes for full length GFP-tagged protein products are ~41 kDa for B13 and ~92 kDa for TOAD-64. empty GFP vector - lane 1: 25pg cell lysate, lane 2: lOpl cell pellet, GFP-tagged B13 - lane 3: 25pg cell lysate, lane 4: lOpl cell pellet, GFP-tagged TOAD-64 - lane 5: 25pg cell lysate, lane 6: lOpl cell pellet, untransfected COS7 cells - lane 7: 25jug cell lysate, lane 8: lOpl cell pellet

Figure 3.9 130 commercial HA monoclonal antibody on Western analysis. Specific antibodies which could detect these proteins in eukaryotic cells would be very useful, so the ability of these polyclonal antibodies to detect HA- and GFP-tagged B13 and TOAD-64 in COS7 cell lysates was tested.

3.4.2A Detection of HA- and GFP-tagged B13 in COS7 cell lysates by the polyclonal B13 antibody The polyclonal B13 antibody detected full length HA-tagged B13 (-16 kDa) in both the 0.5% Triton X-100 soluble and insoluble fractions of COS7 cells (figure 3.10A, lanes 2-8). The antibody also detected full length GFP-tagged B 13 (-41 kDa) in both COS7 cell fractions (figure 3.10Bi, lanes 2-7). HA- and GFP-tagged B 13 had similar protein distributions, thus GFP tagging did not have any adverse effects on the targeting of B13. The sensitivity of the B 13 antibody at a 1:2500 dilution was slightly greater than the GFP antibody at a 1:1000 dilution in detecting GFP-tagged B13 (figure 3.10Bi and 3. lOBii, compare lanes 2-7). However, the concentrations of these antibodies had not been determined and thus a quantitative comparison of their relative sensitivities is not possible. 250ng of recombinant B 13 protein included as a positive control (figure 3.10A and Bi, lane 1) was also strongly detected by the B 13 antibody.

3.4.2B Detection of HA- and GFP-tagged TOAD-64 in COS7 cell lysates by the TOAD-64 antibody Full length GFP-tagged TOAD-64 was detected only in the 0.5% Triton X-100 insoluble fraction of COS7 cells by the GFP antibody (figure 3.1 IBii, lanes 6-8 and figure 3.9, lane 6). However, the TOAD-64 antibody was unable to detect either HA- or GFP-tagged TOAD-64 in the 0.5% Triton X-100 insoluble fraction of COS7 cells (figure 3.11A and 3.1 IBi, lanes 6-8). The polyclonal TOAD-64 antibody strongly detected 350ng of recombinant TOAD-64 protein and also native TOAD-64 in rat brain cytosol which were included as positive controls (figure 3.11A and Bi, lanes 1 and 2). This suggests that the epitope recognised by the TOAD-64 antibody may be masked in both HA- and GFP-tagged TOAD-64. These N terminal tags may disrupt the sequence of the TOAD-64 antibody binding site or the tags themselves may obstruct the binding site in the folded protein.

3.4.3 Effects of GFP tagging on the distribution of B13 and TOAD-64 proteins Expression of HA-tagged B13 and TOAD-64 proteins in COS7 cell lysates

Chapter 3 131 B13 antibody * 16kDa

1 2 3 4 5 6 7 B

B13 antibody

%

11 GFP antibody

Figure 3.10: Detection of HA- and GFP-tagged B13 proteins by a polyclonal B13 antibody COS7 cells were transiently transfected with pXJ40HA and pXJ40GFP B13 DNA constructs, harvested at 24 hours in 400pl of buffer containing 0.5% Triton X-100 and cell pellets were resuspended in 50pl 2x SDS sample buffer (see section 2.6.1). Varying amounts of 0.5% Triton X-100 soluble lysate and 0.5% Triton X-100 insoluble pellet fractions were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with polyclonal B13 antibody (1:2500 dilution, G. Ferrari) and monoclonal GFP antibody (1:1000 dilution, Clontech) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected sizes for full length protein products are ~13 kDa for HA-tagged B13 and ~41 kDa for GFP-tagged B13.

A: Detection of HA-tagged B13 by the B13 antibody Lane 1: 250ng recombinant B13, B13 lysate - lane 2: 12.5pg, lane 3: 25pg, lane 4: 50pg, lane 5: 60pg B13 pellet - lane 6: 2pl, lane 7: 5pl, lane 8: lOpl

B: Detection of GFP-tagged B13 by (i) the B13 antibody and (ii) the GFP antibody Lane 1: 250ng recombinant B13, B13 lysate - lane 2: 12.5pg, lane 3: 25pg, lane 4: 50pg, lane 5: 75jug B13 pellet - lane 6: 2pl, lane 7: 5pl

Figure 3.10 132 TOAD-64 antibody

kDa 175

TOAD-64 antibody 47.5

kDa 175 GFP antibody 62 47.5

Figure 3.11: Detection of HA- and GFP-tagged TOAD-64 proteins by a polyclonal TOAD-64 antibody COS7 cells were transiently transfected with pXJ40HA and pXJ40GFP TOAD- 64 DNA constructs, harvested at 24 hours in 400pl of buffer containing 0.5% Triton X- 100 and cell pellets were resuspended in 50pl 2x SDS sample buffer (see section 2.6.1). Varying amounts of 0.5% Triton X-100 soluble lysate and 0.5% Triton X-100 insoluble pellet fractions were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with polyclonal TOAD-64 antibody (1:2500 dilution, G. Ferrari) and monoclonal GFP antibody (1:1000 dilution, Clontech) (see sections 2.6.7- 2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected sizes for full length protein products are -64 kDa for HA-tagged TOAD-64 and -92 kDa for GFP-tagged TOAD-64.

A: Detection of HA-tagged TOAD-64 by the TOAD-64 antibody B: Detection of GFP-tagged TOAD-64 by (i) the TOAD-64 antibody and (ii) the GFP antibody

Lane 1: 350ng recombinant TOAD-64, lane 2: 15pl rat brain cytosol, TOAD-64 lysate - lane 3: 12.5pg, lane 4: 25pg, lane 5: 45pg, TOAD-64 pellet - lane 6: 2pl, lane 7: 5pl, lane 8: lOpl

Figure 3.1 1 133 cannot be detected by Western analysis using the anti HA antibody. This does not necessarily mean that these proteins were not expressed, as full length HA-tagged TOAD-64 was produced in an in vitro transcription-translation assay (figure 3.1, lane 9) and full length HA-tagged B 13 protein was detected by the polyclonal B 13 antibody in COS7 cell lysates (figure 3.10). Thus the lack of detection of these proteins by Western analysis using the HA antibody may be due to the proteins folding in such a manner as to cause obstruction of the HA antibody binding site. GFP tagging appeared to have no effect on the distribution of B13 protein. Both HA- and GFP-tagged B13 were similarly expressed in the 0.5% Triton X-100 soluble and insoluble fractions of COS7 cells (figure 3.10), as detected by the polyclonal B13 antibody. Unfortunately, the polyclonal TOAD-64 antibody was unable to detect either HA- or GFP-tagged TOAD-64.

3.5 Expression of Rho p21s Both a l - and a2-chimaerin are RacGAPs (Manser et al., 1992; Hall et al., 1993) and down-regulate Rac activity by increasing its intrinsic GTPase activity. In order to exert this activity, the chimaerins must be present at the same location within the cell as Rac itself. Thus the distribution of dominant positive and dominant negative Rho p21s was investigated.

3.5.1 Expression of HA-tagged dominant positive and dominant negative Cdc42, Rac and Rho DNA constructs in COS7 cells Expression and distribution of HA-tagged dominant positive (V12/V14) and dominant negative (N17/N19) Cdc42, Rac and Rho proteins in COS7 cell lysates was investigated (figure 3.12). Full length HA-tagged V12Cdc42, N17Cdc42, V12Rac, N17Rac, V14Rho and N19Rho proteins (-25 kDa) were expressed in the 0.5% Triton X-100 insoluble fraction (figure 3.12, lanes 4, 6, 8, 10, 12 and 14) but not the soluble fraction (figure 3.12, lanes 3, 5, 7, 9, 11 and 13) of COS7 cells. As the volume of soluble fraction loaded on the gel was 5 times less than the insoluble fraction, it is possible that low levels of these proteins may be detected in a larger volume of soluble fraction, however the majority of p21 protein (>95%) was detected in the 0.5% Triton X-100 insoluble fraction of COS7 cells. The -100 kDa band in the insoluble fraction of HA vector transfected and untransfected COS7 cells (figure 3.12, lanes 2 and 16) is a non-specific background band picked up by the HA antibody which is present in all the insoluble fraction

Chapter 3 134 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Figure 3.12: Expression of HA-tagged dominant positive and dominant negative Cdc42, Rac and Rho in COS7 cells COS7 cells were transiently transfected with pXJ40HA Cdc42 (V I2 or N17), Rac (VI2 or N17) or Rho (V I4 or N19) DNA constructs, harvested at 24 hours in 400pl of buffer containing 0.5% Triton X-100 and cell pellets were resuspended in 50pl 2x SDS sample buffer (see section 2.6.1). 25pg of 0.5% Triton X-100 soluble lysate and lOpl of 0.5% Triton X-100 insoluble pellet were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with the monoclonal HA antibody (lpg/ml, Boehringer Mannheim) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected size for full length HA-tagged Cdc42, Rac and Rho is ~21 kDa. empty HA vector - lane 1: 25jitg cell lysate, lane 2: lOpl cell pellet, HA-tagged V12Cdc42- lane 3: 25pg cell lysate, lane 4: lOpl cell pellet, HA-tagged N17Cdc42 - lane 5: 25pg cell lysate, lane 6: lOpl cell pellet, HA-tagged V12Rac - lane 7: 25pg cell lysate, lane 8: lOpl cell pellet, HA-tagged N17Rac - lane 9: 25pg cell lysate, lane 10: lOpl cell pellet, HA-tagged V14Rho - lane 11: 25pg cell lysate, lane 12: lOpl cell pellet, HA-tagged N19Rho - lane 13: 25pg cell lysate, lane 14: lOpl cell pellet, untransfected COS7 cells - lane 13: 25pg cell lysate, lane 14: 1 Ojul cell pellet

Figure 3.12 135 samples.

3,6 Summary Full length HA- or GFP-tagged a2-chimaerin and the E49W, R56L and R73L SH2 domain mutants were predominantly detected in the 0.5% Triton X-100 soluble fraction of transiently transfected COS7 cells, which contains both cytosolic and membrane associated proteins. Full length GFP-tagged al-chimaerin was detected in the 0.5% Triton X-100 insoluble fraction of COS7 cells, which contains cytoskeletal proteins. This cytoskeletal localisation of al-chimaerin agrees with previously published results (Kozma et al., 1996). A single N94H point mutation in the SH2 domain of a2-chimaerin altered the protein distribution from the soluble distribution characteristic of a2-chimaerin, to a mainly insoluble distribution which more closely resembled that of al-chimaerin. A similar difference in protein localisation between a l- and a2-chimaerin was also observed in fractions derived from permanently transfected N1E 115 cell lines expressing a l- or a2-chimaerin. Four permanent N1E 115 cell lines expressing full length al-chimaerin (aID , al-7 , a l-8 and al-10) and eleven permanent N IE 115 cell lines expressing full length a2-chimaerin were established. a2-Chimaerin was detected mainly in the soluble fraction, with a smaller amount of protein associated with membranes. No a2-chimaerin protein was detected in the 1% Triton X-100 insoluble fraction whereas this fraction contained most of the al-chimaerin expression which corresponds to a cytoskeletal localisation. This agrees with the distribution observed in COS7 cells. All al-chimaerin expressing cell lines expressed the same level of endogenous a2-chimaerin, thus the similar morphology of the a2-10 and a ID cell lines is not due to increased a2-chimaerin expression in the alD cell line (see chapter 6). B13 and TOAD-64 are previously identified targets of the a2-chimaerin SH2 domain. HA- and GFP-tagged B13 were detected in both the 0.5% Triton X-100 soluble and insoluble fractions of COS7 cells and GFP tagging had no effect on the protein distribution. A polyclonal B13 antibody detected recombinant protein and both HA- and GFP-tagged B13 in COS7 cell fractions. GFP-tagged TOAD-64 was detected in the 0.5% Triton X-100 insoluble fraction of COS7 cells. A polyclonal TOAD-64 antibody detected recombinant protein and native TOAD-64 in rat brain cytosol but could not detect either HA- or GFP-tagged TOAD-64 in COS7 cell fractions. Full length dominant positive and negative Cdc42, Rac and Rho proteins were expressed in the

Chapter 3 136 insoluble fraction of COS7 cells.

Chapter 3 CHAPTER FOUR: Results II

Chapter 4 138 CHAPTER 4: Investigation of a l- and a2-chimaerin protein interactions

4.1 Localisation of chimaerin proteins Both a l - and a2-chimaerin stimulate the in vitro GTPase activity of Rac and to a much lesser extent Cdc42 (Manser et al., 1992; Hall et al., 1993; M. Teo PhD thesis, 1994). In COS7 cells, al-chimaerin and V12Rac were expressed in the insoluble membrane fraction, whilst a2-chimaerin was expressed in the soluble fraction (see Chapter 3). To investigate whether expression of dominant positive p21s induced translocation of a2-chimaerin to the membrane, COS7 cells were transiently cotransfected with a2-chimaerin and dominant positive Rac, Cdc42 or Rho DNA constructs and protein expression in the insoluble membrane fraction of COS7 cells was examined (figure 4.1). High levels of expression of full length Rac, Cdc42 and Rho proteins (-24 kDa) were detected in the insoluble fraction of COS7 cells (figure 4.1A-C, lanes 2-6), as previously shown in figure 3.12. Low expression of full length a2-chimaerin and the E49W, R56L and R73L SH2 domain mutants was detected in the insoluble fraction of cotransfection samples (lanes 2-5), whilst much stronger expression of the N94H mutant was detected in this fraction (lane 6). A similar protein distribution was observed when these a2-chimaerin proteins were expressed alone in COS7 cells (figures 3.3B & 3.4B). However in the case of V14Rho co-transfection it appears that expression of the N94H mutant in the insoluble fraction was reduced (figure 4.1C, lane 6). This decreased expression may be due to poor transfection of the N94H construct or may represent a specific V14Rho-induced change in protein localisation. Further work is required to determine which is the case. Thus dominant positive Rac, Cdc42 and probably Rho have no significant effect on the protein distribution of a2-chimaerin. This is in contrast to the situation with the Rho effector ROKa which is translocated to the membrane by V14Rho (Leung et al., 1995). The multiple bands present at -24 kDa in lanes 2-6 of figure 4.1A & C may correspond to different post translationally modified or phosphorylated forms of Rac and Rho or possibly protein breakdown.

4.2 Investigation of a2-chimaerin protein interactions The isolation of two potential targets of a2-chimaerin, B13 and TOAD-64, was described earlier. In order to investigate the in vivo interactions between a2-chimaerin and these two potential target proteins, cotransfection and immunoprecipitation

Chapter 4 139 1 2 3 4 5 6

V12Rac

1 2 3 4 5 6 kDa 175 B 83 62 a l 47.5 32.5 V 12Cdc42 25

kDa 175 C 83 62 a l 47.5 32.5 V14Rho 25

Figure 4.1: Co-expression of HA-tagged a2-chimaerin and V12Rac, V12Cdc42 or V14Rho proteins in COS7 cells COS7 cells were transiently cotransfected with pXJ40HA a2-chimaerin and dominant positive p21 DNA constructs, harvested at 24 hours in 400pl of buffer containing 0.5% Triton X-100 and cell pellets were resuspended in 50pl 2x SDS sample buffer (see section 2.6.1). lOjul of 0.5% Triton X-100 insoluble pellet fractions were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with the monoclonal HA antibody (lpg/ml, Boehringer Mannheim) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected sizes for full length protein products are -45 kDa for HA-tagged a2-chimaerin and -21 kDa for HA-tagged Rac, Cdc42 and Rho.

A :V12Rac, B: V12Cdc42 and C: V14Rho cotransfections Lane 1: 1 O j l i I HA vector cell pellet, lane 2: lOpl p21 + a2-chimaerin cell pellet, lane 3: lOpl p21 + a2-chimaerin (E49W) cell pellet, lane 4: lOpl p21 + a2-chimaerin (R56L) cell pellet, lane 5: lOpl p21 + oc2-chimaerin (R73L) cell pellet, lane 6: lOpl p21 + a2-chimaerin (N94H) cell pellet

Figure 4.1 140 experiments were carried out in COS7 and N1E 115 cells. COS7 cells were used as they are able to express high levels of protein from transiently transfected DNA and N1E

1 15 cells were used as a2-chimaerin target proteins may be more likely to interact with a2-chimaerin in a neuronal environment.

To investigate the effects of long term chimaerin overexpression on potential protein interactions, immunoprecipitation experiments were also carried out in permanently transfected N1E 115 cell lines expressing a l - or a2-chimaerin.

4.2.1 Immunonrecinitation of GFP-tagged q2-chimaerin in CQS7 and N1E 115 cells GFP-tagged a2-chimaerin could be immunoprecipitated from transiently transfected COS7 and N1E 115 cell lysates using the a2-chimaerin polyclonal antibody

(figure 4.2A & B, lane 4). Immunoprecipitated protein was detected by Western analysis using the monoclonal GFP antibody. In COS7 cell lysates, the GFP antibody detected a single ~56kDa background band (figure 4.2A, lanes 1 and 3) whilst N1E 115 cell lysates contained a series of background bands including a major band at ~55kDa (figure 4.2B, lanes 1 and 3). However, in immunoprecipitates, the region below 50kDa was obscured because of the presence of the antibody used for immunoprecipitation, which was detected non specifically (figure 4.2A & B, lanes 2 and 4). Having

established a2-chimaerin could be immunoprecipitated from both COS7 and N1E 115

cell lysates, its interactions with previously identified targets were tested.

4.2.2 Investigation of a2-chimaerin and TOAD-64 protein interactions in COS7 cells HA-tagged a2-chimaerin and GFP-tagged TOAD-64 were transiently co­

transfected into COS7 cells (figure 4.3A & B, lane 3) and lysates were subjected to

immunoprecipitation with the a2-chimaerin antibody. A small amount of TOAD-64 co-

immunoprecipitated with a2-chimaerin in COS7 cells under the conditions used and

was detected by Western analysis using the GFP antibody (figure 4.3B, lane 4). Similar results were obtained when the HA antibody was used to immunoprecipitate (data not shown). However, due to interference from the immunoprecipitation antibody it was not

possible to detect HA-tagged a2-chimaerin (~45kDa) in immunoprecipitates (figure

4.3 A, lanes 2 and 4). Considering the large amounts of both proteins present in COS7

cells, very little TOAD-64 co-immunoprecipitated with a2-chimaerin. This suggests

Chapter 4 141 1 2 3 4

83 G FP-a2

1 2 3 4 B

■ ■

Figure 4.2: Immunoprecipitation of GFP-tagged a2-chimaerin from COS7 and N1E 115 cells Untransfected COS7 and N1E 115 cells and cells transiently transfected with pXJ40GFP a2-chimaerin, were harvested at 24 hours in 400pl of buffer containing 1% Triton X-100, 0.5% Na deoxycholate and subjected to immunoprecipitation with the a2-chimaerin antibody (see section 2.6.2). Total cell lysates and immunoprecipitation samples were separated by SDS- PAGE, Western blotted onto nitrocellulose and probed with the monoclonal GFP antibody (1:1000 dilution, Clontech) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected size for full length GFP-tagged a2-chimaerin is ~73 kDa.

A: COS7 cells Lane 1: 4pl untransfected COS7 cell lysate, lane 2: 40pl untransfected COS7 immunoprecipitate, lane 3: 4pl GFP-a2-chimaerin cell lysate, lane 4: 40pl GFP-a2-chimaerin immunoprecipitate

B: N1E 115 cells Lane 1: 25pl untransfected N1E 115 cell lysate, lane 2: 30pl untransfected N1E 115 immunoprecipitate, lane 3: 25pl GFP-a2-chimaerin cell lysate, lane 4: 30pl GFP-ot2-chimaerin immunoprecipitate

Figure 4.2 142 1 2 3 4 kDa 105 HA antibody 71 44

28

B kDa 105 GFP antibody 71

44

28

Figure 4.3: Co-immunoprecipitation of TOAD-64 and a2-chimaerin in COS7 cells COS7 cells were transiently transfected with pXJ40HA a2-chimaerin, or transiently cotransfected with pXJ40HA a2-chimaerin and pXJ40GFP TOAD-64 DNA constructs. Cells were harvested at 24 hours in 400pl of buffer containing 1% Triton X- 100. 0.5% Na deoxycholate and subjected to immunoprecipitation with a2 chimaerin antibody (see section 2.6.2). Total cell lysates and immunoprecipitation samples were separated by SDS- PAGE. Western blotted onto nitrocellulose and probed with the monoclonal HA antibody (lpg/ml, Boehringer Mannheim) or monoclonal GFP antibody (1:1000 dilution, Clontech) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham). The expected sizes for full length protein products are -45 kDa for HA-tagged a2-chimaerin and -92 kDa for GFP-tagged TOAD-64.

A: Detection of HA-tagged a2 chimaerin B: Detection of GFP-tagged TOAD-64

Lane 1: 30pl HA-tagged a2 chimaerin cell lysate, lane 2: 30pl HA-tagged a2 chimaerin immunoprecipitate, lane 3: 30pl HA-tagged a2 chimaerin + GFP-tagged TOAD-64 cell lysate, lane 4: 30j l x 1 HA-tagged a2 chimaerin + GFP-tagged TOAD-64 immunoprecipitate

Fieure 4.3 143 that some form of stimulation is required to facilitate the interaction between TOAD-64 and a2-chimaerin. Since these proteins are usually expressed in a neuronal environment, specific growth factor stimulation, phosphorylation, cellular differentiation or interaction with other neuronal proteins may be required.

4.2.3 Investigation of a2-chimaerin and TOAD-64 protein interactions in N1E 115 cells GFP-tagged TOAD-64 and HA-tagged a2-chimaerin were transiently co­ transfected into N1E 115 cells and lysates were subjected to immunoprecipitation. Both proteins were detected in total cell lysates, but TOAD-64 was not detectable in oc2- chimaerin immunoprecipitates (data not shown). However considering the small amount of TOAD-64 detected in a2-chimaerin immunoprecipitates from COS7 cells, it is possible that insufficient protein was present for detection in N1E 115 cells. It is also probable that some kind of stimulation may be required to promote their interaction in N1E 115 cells.

4.2.4 Investieation of a2-chimaerin and B13 protein interactions in COS7 and N1E 115 cells The small molecular weight of GFP-tagged B13 (~41kDa) prevents its detection in immunoprecipitates via Western blotting, due to interference from the immunoprecipitation antibody. This technical problem meant it was not possible to determine whether GFP-tagged B13 co-immunoprecipitated with a2-chimaerin in COS7 or N1E 115 cells. The other possible way to examine whether a2-chimaerin and B13 co- immunoprecipitated was to immunoprecipitate HA-tagged B13 and detect co- immunoprecipitation of GFP-tagged a2-chimaerin by Western blotting, as GFP-tagged a2-chimaerin is larger (~73kDa). However since HA-tagged B13 in COS7 cell lysates could not detected by the HA antibody, this experiment was not attempted. Use of the recently produced rabbit polyclonal B13 antibody has shown that HA-tagged B 13 is expressed in COS7 cells (figure 3.10), so it would now be possible to perform this experiment using the polyclonal B13 antibody to detect HA-tagged B13 expression. Another method to enable detection of small molecular weight proteins in immunoprecipitation samples would be to covalently couple the immunoprecipitation antibody to agarose, thus enabling its removal from immunoprecipitation samples.

Chapter 4 144 4.2.5 Immunoprecipitation of a l- and a2-chimaerin from permanent N1E 115 cell lines Permanently transfected N1E 115 cell lines expressing a l- or a2-chimaerin were established to investigate the effects of long term overexpression of these proteins on cell morphology and possible protein-protein interactions in a neuronal cell line. It was previously shown that the alD , al-7, a l-8 and al-10 cell lines express full length al-chimaerin (figure 3.7), whilst of the eleven cell lines expressing full length a2- chimaerin, a2-10 was one of the highest expressing cell lines (figure 3.6). The potential protein-protein interactions in these cell lines were investigated by immunoprecipitation. In order to identify potential targets of the a2-chimaerin SH2 domain, the association of tyrosine phosphorylated proteins with chimaerin in immunoprecipitates from chimaerin expressing permanent NIE 115 cell lines was investigated. No strong immunoreactive targets were co-immunoprecipitated, however in al-chimaerin (aID and al-10) and a2-chimaerin (a2-10) cell lines, a ~130kDa tyrosine phosphorylated protein co-immunoprecipitated with chimaerin (figure 4.4, lanes 2, 5 and 6). Tyrosine phosphorylated proteins which associated with only a l- or a2-chimaerin isoforms were also weakly detected. A ~165kDa tyrosine phosphorylated protein was only detected in immunoprecipitates from the al-10 cell line (figure 4.4, lane 5) whilst a ~180kDa tyrosine phosphorylated protein was specifically detected in immunoprecipitates from the a2-10 cell line (figure 4.4, lane 6). These results indicate that the a2-chimaerin SH2 domain does not immunoprecipitate enriched levels of tyrosine phosphorylated proteins under the conditions used. However, it is possible that appropriate stimulation may induce tyrosine phosphorylation of a2-chimaerin substrates and promote their interaction with the SH2 domain of a2-chimaerin. p35 is the neuron specific regulator of cyclin dependent kinase-5 (cdk5), which binds and activates the kinase (Lew et al., 1994; Tsai et al., 1994). An interaction between p35 and chimaerin has recently been detected (J. Wang and R. Qi, Personal Communication), so the association of p35 with a l- and a2-chimaerin in immunoprecipitates from permanently transfected N1E 115 cells was investigated. p35 could be detected in chimaerin immunoprecipitates in the alD , al-7 and a2-10 cell lines (figure 4.5, lanes 2, 3 and 6) and a ~32kDa background band was also present in all samples. As both TOAD-64 and B 13 are previously identified targets of the a2-chimaerin

Chapter 4 145 Figure 4.4: Co-immunoprecipitation of tyrosine phosphorylated proteins with a l­ and a2-chimaerin in permanent N1E 115 cell lines 2.5x107 cells for each permanent cell line were harvested, HA-tagged a l - or a2- chimaerin was immunoprecipitated with the HA antibody and immunoprecipitates were eluted in lOOpl 2x SDS sample buffer (see section 2.6.5). 30pl of each immunoprecipitation sample were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with the monoclonal phosphotyrosine antibody (0.2pg/ml, Santa Cruz) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham).

Lane 1: untransfected N1E 115 cells, lane 2: a lD , lane 3: a l-7 , lane 4: a l-8 . lane 5: a l-1 0 , lane 6: a2-10, lane 7: HAv-24 empty vector

Figure 4.4 146 1 2 3 4 5 6 7

.4— p35 32.5kDa

Figure 4.5: Co-immunopreeipitation of p35 with al- and a2-chimaerin in permanent N1E 115 cell lines 2.5x107 cells for each permanent cell line were harvested, HA-tagged a l - or a2- chimaerin was immunoprecipitated with the HA antibody and immunoprecipitates were eluted in lOOpl 2x SDS sample buffer (see section 2.6.5). 30pl of each immunoprecipitation sample were separated by SDS-PAGE, Western blotted onto nitrocellulose and probed with the polyclonal p35 antibody (1:500 dilution, Santa Cruz) (see sections 2.6.7-2.6.9). Immunoreactive proteins were detected using ECL reagents (Amersham).

Lane 1: untransfected N1E 115 cells, lane 2: alD , lane 3: al-7, lane 4: al-8. lane 5: al-10, lane 6: a2-10. lane 7: HAv-24 empty vector

Figure 4.5 147 SH2 domain, their association with a2-chimaerin in permanently transfected N1E 115 cell lines was investigated. However, neither endogenous TOAD-64 nor B 13 were detected in chimaerin immunoprecipitates from the a l - or a2-chimaerin cell lines (data not shown). Both p35 and the 130kDa tyrosine phosphorylated protein co- immunoprecipitated with a l- and a2-chimaerin, which suggests that the interaction occurred via a common region in these two proteins, possibly either the GAP or cysteine rich domain or perhaps indirectly via a second protein which binds both isoforms of chimaerin. The ~165kDa tyrosine phosphorylated protein detected in al-10 immunoprecipitates was not detected in control or a2-chimaerin expressing cell lines, whilst the ~180kDa tyrosine phosphorylated protein detected in a2-10 immunoprecipitates was not detected in any of the control or al-chimaerin expressing cell lines. This suggests that the ~165kDa and ~180kDa tyrosine phosphorylated proteins specifically interact with only a l- or a2-chimaerin respectively. These interactions may occur via the specific N terminal sequences of al- and a2-chimaerin; possibly with the SH2 domain of a2-chimaerin, or perhaps indirectly via specific intermediary proteins. The levels of tyrosine phosphorylated proteins and p35 detected in immunoprecipitation samples were low as they result from endogenous expression in N1E 115 cells rather than transfected proteins.

4,3 Summary V12Rac, V12Cdc42 and V14Rho did not cause translocation of a2-chimaerin or the E49W, R56L and R73L SH2 domain mutants to the insoluble fraction of COS7 cells. In permanent N1E 115 cell lines expressing a l- or a2-chimaerin, p35 and a ~130kDa tyrosine phosphorylated protein co-immunoprecipitated with both a l - and a2-chimaerin isoforms. Two other tyrosine phosphorylated proteins of ~165kDa and ~180kDa specifically co-immunoprecipitated with only a l - or a2-chimaerin, respectively, in the al-1 0 and a2-10 cell lines. Immunoprecipitation of HA-tagged a2- chimaerin using either the HA or a2-chimaerin antibody resulted in the co- immunoprecipitation of a small amount of GFP-tagged TOAD-64 in transiently transfected COS7 cell lysates. However neither B 13 nor TOAD-64 proteins were detected in chimaerin immunoprecipitates from any permanently transfected NIE 115 cell lines.

Chapter 4 148 CHAPTER FIVE: Results III

Chapter 5 CHAPTER 5: Investigation into the role of Rho p21s and the q-chimaerins in

NFkB signalling in HeLa and N1E 115 cells

Rac is involved in a number of signalling pathways leading to transcriptional activation. Rac activates the JNK and p38 signalling pathways (Coso et al., 1995;

Minden et al., 1995) and also the NFk B signalling pathway (Sulciner et al., 1996; Perona et al., 1997). The involvement of a2-chimaerin and its SH2 domain targets in these pathways was considered and their role in the NFk B signalling pathway was investigated.

5.1 NFkB activation Many agents including growth factors, cytokines, lipopolysaccharide and ultraviolet irradiation stimulate NFk B activity. Rho proteins have been shown to stimulate NFk B in several cell types including COS7, HeLa and Jurkat T cells (Sulciner et al., 1996; Perona et al., 1997) and recently RhoGEFs Dbl, Vav and Ost were also shown to stimulate NFk B activity in COS7 cells (Montaner et al., 1998). In HeLa cells,

Rac-induced NFk B stimulation was mediated via the production of reactive oxygen species (ROS) (Sulciner et al., 1996). The stimulatory effect of Rac on ROS production was recently found to depend on the insert region of Rac (Joneson and Bar Sagi, 1998). a l- and a2-Chimaerin are both RacGAPs which downregulate Rac by stimulation of its GTPase activity in vitro, but although chimaerin has also been shown to act as both a Rac downregulator and an effector in vivo (Kozma et al., 1996), a precise physiological role for the SH2 domain containing chimaerin isoform has not yet been determined. Rac is involved in ROS generation, B13 a potential a2-chimaerin target protein appears to stimulate ROS production (C. Hall, unpublished results) whilst TOAD-64 was identified in a trans plasma membrane oxidoreductase (PMO) complex purified from synaptic plasma membranes (Bulliard et al., 1997). PMOs are antioxidant enzymes which are activated in response to cellular stress and ROS. Thus the effects of the chimaerins on the ROS responsive NFkB signalling pathway were investigated.

HeLa cells were chosen to validate the NFk B assay method used^as both

V12Rac and IL-lp showed high activation levels of NFk B in this cell line (Sulciner et

al., 1996). As the chimaerins are neuronally expressed proteins, their effects on NFk B signalling in N1E 115 mouse neuroblastoma cell line were investigated. The level of

V12Rac-induced NFk B activation observed in HeLa cells varied with increasing levels

Chapter 5 150 of DNA (Sulciner et al., 1996). Thus the effects of several DNA levels on NFk B activation in both HeLa and N1E 115 cells was investigated.

5.2 NFkB reporter assay

Activation of NFk B results in the translocation of the NFk B dimer to the nucleus where it binds a specific DNA sequence (the NFk B binding site) and stimulates transcription (reviewed in Baldwin, 1996). The luciferase coupled NFk B reporter vectors were generated by J.M. Dong (IMCB, Singapore) from the pGL2-basic vector

(Promega) and contain four copies of the functional or mutated NFk B binding site derived from the MHC promoter sequence (appendix 4).

The basis of the reporter assay used is that activated NFk B binds its specific binding sites on the reporter vector which induces production of the enzyme luciferase. The amount of luciferase produced is measured indirectly via a light producing reaction, which occurs when luciferase is incubated with its substrate luciferin. The amount of light produced is measured by a luminometer and is proportional to the level of NFk B activity in the sample.

HeLa and N1E 115 cells were transiently transfected with both NFk B reporter vector and HA-tagged sample DNA in triplicate, harvested at 24 hours and cell lysates were assayed for NFk B induced luciferase production in a luminometer (see Methods sections 2.5.2b, 2.5.3 and 2.7.1). Luminometer data presented in figures 5.1-5.10 was processed as described in Methods section 2.7.2 and sample data is presented in appendices 5-10.

5.2.1 NFkB activation in HeLa cells

NFk B activity in HeLa cell lysates is shown in figure 5.1. Sample values in figure 5.1 were also plotted in graphical form to enable easier comparison (figure 5.2). At the low (0.5p,g) DNA level in HeLa cells, V12Rac typically caused -13 fold stimulation of NFk B activity, which agrees with previously published data (Sulciner et al., 1996). MEKK1 also caused -13 fold stimulation of NFk B activity above baseline levels. MEKK1 has not previously been shown to activate NFk B in this cell line, but was recently shown to stimulate NFk B activity in a Rac- and Cdc42-dependent manner in COS7 cells (Montaner et al., 1998). a l - and a2-Chimaerin both decreased NFk B activity to -0.2, a five fold decrease in activity compared to the empty HA vector, which is consistent with downregulation of endogenous Rac activity. A similar decrease

Chapter 5 151 Figure 5.1: Summary Table of NFkB Reporter Assay Results in HeLa cells (0.5pg DNA level)

Fold Stimulation compared to empty HA vector ± Standard Deviation of ‘Fold Samples Stimulation9 values (n=9) 0.5pg NFk Bv + 0.5pg HAV 0.942 ± 0.099 0.5pg NFk Bv + 0.5pg MEKK 13.419 ± 3.003 0.5pg NFk Bv + 0.5pg HA-V12Rac 12.810 ± 1.251 0.5pg NFk Bv + 0.5pg HA-N17Rac 0.290 + 0.177 0.5pg NFk Bv + 0.5pg HA-al chimaerin 0.204 + 0.119 0.5pg NFk Bv + 0.5 pg HA-a2 chimaerin 0.162 + 0.061 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin (E49W) 0.544 + 0.179 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin (R56L) 0.306 + 0.124 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin (R73L) 0.062 ± 0.020 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin (N94H) 0.056 + 0.019 0.5pg NFk Bv + 0.5pg HA-B13 0.405 ± 0.060 0.5pg NFk Bv + 0.5pg HA-TOAD-64 0.175 ± 0.096

KEY: NFk Bv is functional NFk B reporter vector HAV is empty pXJ40-HA vector

Figure 5.1 152 16 -

14 -

13 - ■ o

12 -

Figure 5.2: Graphical representation of NFk B activity in HeLa cells at the 0.5pg DNA level Data from figure 5.1

Figure 5.2 153 Figure 5.3: NF kB activation in HeLa cells stimulated with IL-1B

A HeLa cells were transfected Fold Stimulation compared to with 0.5pg NFkBv + 0.5pg empty HA vector HAV and treated as ± Standard Deviation of ‘Fold indicated Stimulation9 values (n=3) 5%FCS ± IL-ip treatment for 18 hours No additions 1.000 ± 0.563 + 0.01 ng/ml IL-lp 0.171 ± 0.028 + 0.1 ng/ml IL-ip 0.201 ±0.101 + 1 ng/ml IL-ip 0.861 ± 0.539 + 10 ng/ml IL-lp 0.913 ± 0.295 + 50 ng/ml IL-IP 0.968 ± 0.877 + 75 ng/ml IL-ip 5.587 ± 2.224 + 100 ng/ml IL-ip 3.761 + 0.961

No serum ± IL-ip treatment for 18 hours No additions 1.000 ± 0.269 + 50 ng/ml IL-lp 4.451 ±0.321 + 75 ng/ml IL-IP 3.764 ±0.831 + 100 ng/ml IL-lp 3.843 ±1.630

B HeLa cells were transfected Fold Stimulation compared to as indicated and treated empty HA vector with IL-ip for 18 hours in ± Standard Deviation o f ‘Fold the presence of 5%FCS Stimulation9 values (n=3) + 75 ng/ml IL-ip 0.5pg NFk Bv + 0.5pg HAv 5.587 ± 2.224 0.5pg NFk Bv + 0.5pg HA-al 0.325 ±0.101 chimaerin

+ 100 ng/ml IL-ip 0.5pg NFk Bv + 0.5j^g HAv 3.761 ± 0.961 0.5pg NFk Bv + 0.5pg HA-al 0.255 ± 0.035 chimaerin

Figure 5.3 154 in NFk B activity was obtained with dominant negative Rac which should block endogenous Rac activity. The E49W and R56L mutations in the SH2 domain of a2- chimaerin resulted in slightly less inhibition of NFk B activity than observed with wild type a2-chimaerin, but the R73L and N94H mutations further reduced NFk B activity to -0.06, a seventeen fold decrease in activity compared to the empty vector. The previously identified a2-chimaerin targets, TOAD-64 and B13 also decreased NFk B activity compared to the empty HA vector. B13 decreased NFk B activity to -0.4, a 2.5 fold decrease whilst TOAD-64 decreased it to -0.2, a 5 fold decrease similar to that observed with a2-chimaerin. The decreased NFk B activity in these samples may represent specific inhibition by these proteins, although measuring a lack of activation is difficult.

In order to determine whether these proteins inhibited NFk B activation, their effects on ligand stimulated NFk B activation were investigated. Activation of NFk B by EL-ip in HeLa cells was previously shown to be Rac-dependent (Sulciner et al., 1996) and so the effects of the chimaerin RacGAPs on IL-ip stimulated NFk B activity were investigated. At low concentrations in the presence of 5% serum (0.01-50ng/ml), IL-ip appeared to inhibit NFk B activity and 75-100ng/ml IL-ip was required to induce NFk B activation (figure 5.3 A). In the absence of serum, NFk B activation was observed at the lower IL-ip concentration of 50ng/ml. The additional stress induced by serum starvation or absence of other serum factors may explain why a lower EL-lp concentration was required in the absence of serum than in the presence of serum to induce similar levels of NFk B activation. The -5.6 and -3.8 fold stimulation of NFkB activity induced by 75ng/ml and lOOng/ml IL-ip in the presence of 5% serum was reduced to -0.3 and -0.25 in the presence of al-chimaerin (figure 5.3B). This inhibition was probably due to the RacGAP activity of al-chimaerin, as IL-ip stimulation of

NFk B activity is Rac-dependent in HeLa cells (Sulciner et al., 1996).

5.2.2 NFkB activation in N1E 115 cells

NFk B activity in N1E 115 cells is shown in figure 5.4. Sample values in figure 5.4 are also shown in graphical form (figure 5.5). At the low (0.5p,g) DNA level in N1E

115 cells, V12Rac unexpectedly caused a decrease in NFk B activity to -0.2, a 5 fold drop in activity compared to that observed with the empty vector. In contrast, V12Rac

Chapter 5 155 Figure 5.4; Summary Table of NFkB Reporter Assay Results in N1E 115 cells (0.5pg DNA level)

Fold Stimulation compared to empty HA vector ± Standard Deviation of ‘Fold Stimulation’ values (n=9) Samples 0.5pg NFk Bv + 0.5pg HAV 0.950 ±0.125 0.5pg NFk Bv + 0.5pg MEKK 5.756 ±2.192 0.5pg NFk Bv + 0.5pg HA-V12Rac 0.210 ± 0.023 0.5pg NFk Bv + 0.5pg HA-N17Rac 0.416 ± 0.044 0.5pg NFk Bv + 0.5pg HA-al chimaerin 0.930 ± 0.282 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin 1.102 ±0.221 0.5[Lg NFkBv + 0.5pg HA-a2 chimaerin (E49W) 0.305 ± 0.057 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin (R56L) 0.688 ± 0.302 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin (R73L) 1.122 ± 0.228 0.5pg NFk Bv + 0.5pg HA-a2 chimaerin (N94H) 0.687 ±0.150 0.5pg NFk Bv + 0.5pg HA-B13 1.209 ±0.314 0.5p,g NFk Bv + 0.5pg HA-TOAD-64 1.196 ±0.216

0.5pg NFk B(M)v + 0.5pg HA-V12Rac 0.113 ±0.062 0.5pg NFk B(M)v + 0.5p,g MEKK 0.024 ± 0.003

KEY: NFk Bv is functional NFk B reporter vector NFk B(M)v is mutated NFk B reporter vector HAV is empty pXJ40-HA vector

Figure 5.4 156 4 —

Figure 5.5: Graphical representation of NFk B activity in N1E 115 cells at the 0.5pg DNA level Data from figure 5.4

Figure 5.5 157 activates NFk B activity in a variety of cell lines including NIH 3T3 and HeLa cells (Sulciner et al., 1996; Sundaresan et al., 1996). However, MEKK1 caused -6 fold stimulation of NFk B activity in N1E 115 cells which had not previously been shown in this cell line.

al-Chimaerin and a2-chimaerin had no effect on NFk B activity when compared to the empty vector. The E49W mutation in the SH2 domain of a2-chimaerin decreased

NFk B activity to -0.3, a 3 fold decrease from the activity observed with wild type a2- chimaerin. Both the R56L and N94H mutations decreased NFk B activity to -0.7, a slight decrease from the activity observed with wild type a2-chimaerin, however the standard deviation associated with the R56L decrease was high. The other proteins tested; the R73L a2-chimaerin mutant, TOAD-64 and B 13, had little effect on NFk B activity when compared to the empty vector.

Since inhibition of NFk B activity was unexpectedly induced by V12Rac at low

DNA levels (0.5|ug), the effects of higher levels of DNA on NFk B activation in N1E 115 cells was also investigated (figure 5.6 and in graphical form in figure 5.7). In N1E 115 cells, similar results were obtained at both DNA levels for all samples except MEKK1, a2-chimaerin and B13. At low DNA levels, MEKK1 induced

-6 fold activation of NFk B while a2-chimaerin and B13 had no effect on NFk B activity. At higher DNA levels, both MEKK1 and a2-chimaerin induced -1.6 fold stimulation of NFk B activity, whilst B13 caused a -7 fold decrease in NFk B activity which was also observed with V12Rac. The smaller stimulation of NFk B activity by MEKK1 at the higher DNA level compared to the low DNA level may be due to the high baseline activity at the higher DNA level masking any stimulatory effects. The decreased NFkB activity observed with V12Rac, N17Rac and the E49W a2-chimaerin mutant at both DNA levels and the decrease observed with B 13 at the higher DNA level in N1E 115 cells may represent inhibition by these proteins. Unfortunately, attempts to stimulate NFkB activity in N1E 115 cells via treatment with

H2O2, lipopolysaccharide, ultraviolet irradiation, PMA or TNFa were unsuccessful, thus it was not possible to fully investigate the potential inhibitory effects of V12Rac, N17Rac, B13 and the E49W mutant on ligand stimulated NFkB activation.

As V12Rac unexpectedly decreased NFk B activity in N1E 115 cells, the effects of Cdc42 and Rho on NFk B activity were investigated since both Cdc42 and Rho were also previously shown to stimulate NFk B activity; Cdc42 in COS7 cells and Rho in both COS7 and NIH 3T3 cells (Perona et al., 1997). The activity of Rho p21s is

Chapter 5 158 Figure 5.6: Summary Table of NFkB Reporter Assay Results in N1E 115 cells (1.5pg DNA level)

Fold Stimulation compared to empty HA vector ± Standard Deviation o f ‘Fold Stimulation’ Samples values (n=6) 1.5pg NF kB v + 1 -5pg HAV 0.968 ± 0.026 1.5pg NF kB v + 1 -5pg MEKK 1.632 + 0.138 1.5pg NF kB v + 1.5pg HA-V12Rac 0.156 ± 0.032 l.SjLtg NF kB v + 1.5pgHA-N17Rac 0.472 ± 0.072 * 1.5pg NF kB v + 1.5pg HA-al chimaerin 0.912 ±0.117 1.5pg NF kB v + 1 - 5pg HA-a2 chimaerin 1.563 ±0.158 1.5pg NF kB v + 1.5pg HA-a2 chimaerin (E49W) 0.219 ± 0.222 1.5pg NF kB v + 1.5pg HA-a2 chimaerin (R56L) 0.970 ± 0.204 1.5|ag NF kB v + 1.5pg HA-a2 chimaerin (R73L) 0.859 ±0.163 1.5pg NF kB v + 1 -5pg HA-a2 chimaerin (N94H) 0.777 ±0.291 1.5pg NF kB v + 1 -5pg HA-B13 0.153 ± 0.050 1.5pg NF kB v + 1.5pg HA-TOAD-64 0.835 ±0.189

KEY: NFk Bv is functional NFk B reporter vector NFk B(M)v is mutated NFk B reporter vector HAV is empty pXJ40-HA vector * (n=3) for this sample

Fieure 5.6. 159 Figure 5.7: Graphical representation of NFk B activity in N1E 115 cells at the 1.5pg DNA level Data from figure 5.6

Figure 5.7 160 Figure 5.8: Summary table of dominant positive and dominant negative Rho family protein induced NFkB activity in N1E 115 cells (0.5pg level)

Fold Stimulation compared to empty HA vector ± Standard Deviation of ‘Fold Samples Stimulation9 values (n=6) In 5% serum 0.5pg NFk Bv + 0.5pg HAV 1.000 ± 0.085 0.5pg NFk Bv + 0.5pg HA-V12Rac 0.362 ±0.157 0.5pg NFk Bv + 0.5pg HA-N17Rac 0.485 ± 0.240 0.5pg NFk Bv + 0.5pg HA-V12Cdc42 0.432 ± 0.043 0.5pg NFk Bv + 0.5pg HA-N17Cdc42 0.807 ± 0.330 0.5pg NFk Bv + 0.5pg HA-V14Rho 0.291 ± 0.082 0.5pg NFk Bv + 0.5pg HA-N 19Rho 0.159 ± 0.078 0.5jag NFk Bv + 0.5pg MEKK 6.260 ± 0.628

Serum free 0.5pg NFk Bv + 0.5pg HAV 1.000 ±0.176 0.5pg NFk Bv + 0.5jLtg HA-V12Rac 0.544 ±0.103 0.5pg NFk Bv + 0.5pg HA-N 17Rac 0.836 ± 0.250 0.5pg NFk Bv + 0.5pg HA-V12Cdc42 0.504 ± 0.092 0.5jLig NFk Bv + 0.5pg HA-N17Cdc42 1.003 ± 0.359 0.5pg NFk Bv + 0.5|Lig HA-V14Rho 0.365 ± 0.056 0.5pg NFk Bv + 0.5p.g HA-N 19Rho 0.396 ± 0.036 0.5pg NFk Bv + 0.5pg MEKK 6.695 ±1.215

KEY: NFk Bv is functional NFk B reporter vector HAV is empty pXJ40-HA vector

Figure 5.8 161 A

0.6 -

0.4 -

0.2 -

0.0

Q. Q- toto

B

0.6 -

0.4 -

0.2 -

0.0

to t o D- a. o toto

Figure 5.9: Graphical representation of Rho family protein induced NFk B activity in N1E 115 cells at the 0.5pg DNA level Data from figure 5.8 A: in the presence of 5% serum, B: in the absence of serum

Figure 5.9 162 Figure 5.10: Summary Table of mutated NFkB(M) Reporter Assay Results in N1E 115 cells (1.5pg DNA level)

Fold Stimulation compared to empty HA vector ± Standard Deviation of ‘Fold Samples Stimulation9 values (n=3) 1.5pg NFk Bv + 1.5p.g HAV 0.9514 ± 0.0280

1.5pg NFk B(M)v + 1.5pg HAV 0.0100 ± 0.0009 1.5pg NFk B(M)v + 1.5|ag HA-V12Rac 0.0023 ± 0.0007 1.5pg NFk B(M)v + 1.5pg HA-al chimaerin 0.0045 ± 0.0005 1.5pg NFk B(M)v + 1.5pg HA-a2 chimaerin 0.0069 ± 0.0024 1.5jag NFk B(M)v + 1.5pg HA-a2 chimaerin (E49W) 0.0035 ± 0.0009 1.5pg NFk B(M)v + 1.5pg HA-a2 chimaerin (R56L) 0.0027 ± 0.0009 1.5pg NFk B(M)v + 1.5pg HA-a2 chimaerin (R73L) 0.003910.0017 1.5pg NFk B(M)v + 1.5pg HA-a2 chimaerin (N94H) 0.0028 10.0009 1.5pg NFk B(M)v + 1.5pg HA-B13 0.0011 ± 0.0002 1.5pg NFk B(M)v + 1.5pg HA-TOAD-64 0.0039 ± 0.0005

KEY: NFk Bv is functional NFk B reporter vector NFk B(M)v is mutated NFk B reporter vector HAV is empty pXJ40-HA vector

Figure 5.10 163 dependent on serum conditions; Rho is activated by LPA present in serum, thus the

effects of the Rho p21s on NFk B activity were investigated in the presence of 5% serum and in serum free conditions (figure 5.8 and graphical form in figure 5.9). No

stimulation of NFk B activity was observed with V12Rac, V12Cdc42 or V14Rho in the presence or absence of serum at low (0.5p,g) DNA levels, although 6-7 fold stimulation was induced by MEKK1 in both conditions (figure 5.8). In fact, activated Rac, Cdc42

and Rho appeared to inhibit NFk B activity. In the presence of serum, dominant negative

Rac and Rho had similar effects to the activated proteins on NFk B activity whilst dominant negative Cdc42 had no effect. In the absence of serum, dominant negative

Rac and Cdc42 had little effect on NFk B activity whilst dominant negative Rho

inhibited NFk B activity as did V14Rho. These results show that in N1E 115 cells Rho, Rac and Cdc42 do not stimulate

NFk B activity, unlike in COS7, HeLa and Jurkat T cells (Sulciner et al., 1996; Perona et

al., 1997). MEKK1 was shown to act downstream of Cdc42 and Rac to stimulate NFk B activity in COS7 cells (Montaner et al., 1998), however since none of the Rho proteins

stimulated NFk B activity in N1E 115 cells, MEKK 1-induced NFk B activation in this cell line must be independent of Rho proteins. Finally, although the Rho p21s do not

stimulate NFk B activity in N1E 115 cells they may play an inhibitory role but this possibility has not yet been investigated.

5.2.3 Specificity ofNF kB activation

To ensure that the NFk B activation measured was specific, a mutated NFk B reporter vector was used as a control. In N1E 115 cells at the 1.5p,g DNA level, for all

samples the values obtained for NFk B activity using the mutated vector ranged from 0.0011 to 0.0100 (figure 5.10) which is between 15 and -140 fold lower than the lowest

sample value of 0.153 for 1.5pg HA-B13 obtained using the functional NFk B vector

(figure 5.6). At the 0.5pg level in N1E 115 cells, NFk B activity associated with

MEKK1 stimulation using the mutated NFk B vector was 0.024 fold in comparison to

the -6 fold stimulation obtained using the functional NFk B vector, a several hundred fold difference in activity (figure 5.4).

5.3 Summary

V12Rac stimulated NFk B activity in HeLa cells as previously reported (Sulciner et al., 1996), thus validating the assay method used. MEKK1 caused a similar level of

Chapter 5 164 NFk B stimulation to V12Rac, which had not previously been shown. In HeLa cells, a l and a2-chimaerin caused a five fold decrease in activity compared to the empty HA vector, consistent with their downregulation of endogenous Rac. Mutations in the SH2

domain of a2-chimaerin had little effect on the NFk B activity observed with wild type a2-chimaerin, although the E49W and R56L mutations slightly increased NFk B activity whilst the R73L and N94H mutations further reduced NFk B activity. The previously identified a2-chimaerin targets, TOAD-64 and B 13 caused a 5 fold and 2.5 fold

decrease in NFk B activity, respectively. IL-lp-induced NFk B stimulation in HeLa cells, previously shown to be Rac-dependent (Sulciner et al., 1996), was inhibited by al-chimaerin possibly via its RacGAP activity. The effects of a2-chimaerin on IL-ip-

induced NFk B activation have not yet been investigated.

In N1E 115 cells, neither Rac nor the other Rho family proteins induced NFk B

activation. In fact, V12Rac caused a 5-7 fold decrease in NFk B activity, at two different expression levels. This contrasts with previous reports in COS7, HeLa and Jurkat T

cells whereNF k B activity is stimulated by Rho proteins (Sulciner et al., 1996; Perona et

al., 1997). However MEKK1 did stimulate NFk B activity in N1E 115 cells, which has not previously been shown. In COS7 cells MEKK1 was shown to act downstream of

Cdc42 and Rac in NFk B activation (Montaner et al., 1998), however since Rho p21s do

not stimulate NFk B in N1E 115 cells a different pathway must exist for MEKK1-

induced NFk B activation in these cells.

a l- and a2-chimaerin had no effect on NFk B activity when expressed at low levels in N1E 115 cells, but when expressed at higher levels, a2-chimaerin caused a

slight increase in NFk B activity similar to that observed with MEKK1. a2-chimaerin

SH2 domain mutants had similar effects on NFk B activity at both expression levels.

The E49W SH2 domain mutant decreased NFk B activity ~3 fold from that observed

with the empty vector control, the N94H mutant slightly decreased NFk B activity, whilst the R56L and R73L mutants had no effect. The a2-chimaerin targets B 13 and

TOAD-64 had no effect on NFk B activity at low expression levels, but at higher

expression levels B13 caused a ~7 fold decrease in NFk B activity, which was also

observed with V12Rac. Attempts to establish ligand stimulation of NFk B activity in N1E 115 cells were unsuccessful, thus it was not possible to determine whether the

observed decreases in NFk B activity were specific inhibitory effects on the pathway.

Chapter 5 165 CHAPTER SIX: Results IV

Chapter 6 CHAPTER 6: The morphology of N1E 115 cells expressing al-chimaerin, a l - chimaerin or potential a2-chimaerin targets

6.1 Rho proteins and neuronal cell morphology The role of the Rho p21 s in neuronal development and differentiation has been investigated in many neural systems, primary cultures and tissue culture cell lines. PC 12 and N1E 115 cell lines are two of the most extensively studied to date and some of the effects of Rho, Rac and Cdc42 on morphology are established. In both PC 12 and N1E 115 cells, Rho induces neurite retraction which is mediated by its effector ROK (Hirose et al., 1998; Katoh et al., 1998) and inhibition of Rho activity by C3 exoenzyme induces neurite outgrowth (Nishiki et al., 1990; Jalink et al., 1994; Kozma et al., 1997). Rho- induced neurite retraction is opposed by the action of Rac and Cdc42 which promote neurite outgrowth in both these cell lines (Kozma et al., 1997; Lamoureux et al., 1997; Daniels et al., 1998). Proteins which regulate p21 activity, such as GEFs, have also been shown to affect neuronal morphology. In N1E 115 cells, a 190 kDa RhoGEF induced neurite retraction, as does Rho (Gebbink et al., 1997) whilst the RacGEF Tiaml induced cell spreading via Rac activation (Van Leeuwen et al., 1997). It is likely that other p21 regulatory proteins such as GAPs and GDIs affect neuronal morphology. Chimaerin isoforms are differentially expressed and developmentally regulated and their common C terminal region contains a cysteine rich domain (CRD) and a PvhoGAP domain (see figure 1.4). Since these proteins differ in sequence only at the N terminal, this region is likely to be responsible for differences in protein function. In order to further investigate the effects of the N terminal SH2 domain on a2-chimaerin protein function, point mutations in the SH2 domain of a2-chimaerin were made at positions predicted to affect its function (Bibbins et al., 1993; G. Ferrari PhD thesis, 1999).

The morphological effects of Rho p2Is in the NIE 115 cell line have been studied and these cells are more easily transfected than primary neurons and PC 12 cells. Thus N1E 115 cells were selected as the model system to investigate the effects of transient and permanent overexpression of the a l - and a2-chimaerin isoforms on neuronal cell morphology.

6.2 Morphology of untransfected N1E 115 cells In the presence of serum, N IE 115 cells are mainly rounded in appearance and

C hapter 6 167 Phalloidin

Figure 6.1: The morphology of untransfected N1E 115 cells N1E 115 cells were plated on poly-L-lysine coated coverslips and then serum starved overnight. Whilst other cells were transiently transfected with DNA in the absence of serum, controls were incubated in serum free media, then all cells were incubated overnight in 5% serum. Control cells were fixed at the same time as transfected samples (24 hours post transfection), subjected to immunocytochemistry and analysed by fluorescent confocal microscopy (see Methods sections 2.5.2 and 2.6.10). Phalloidin staining of F-actin is shown in panels a & b. The bar represents 25pm in both panels.

N 1E 115 cells (panels a & b)

Figure 6.1 168 GFP Phalloidin

Figure 6.2 169 GFP Phalloidin

Figure 6.2: Expression of GFP-tagged chimaerin proteins in transiently transfected N1E 115 cells N1E 115 cells were plated on poly-L-lysine coated coverslips and then serum starved overnight. Cells were then transiently transfected with GFP-tagged a l- chimaerin, a2-chimaerin or a2-chimaerin SH2 domain mutant DNA constructs in the absence of serum and incubated overnight in 5% serum. Cells were fixed 24 hours post transfection, subjected to immunocytochemistry and analysed by fluorescence confocal microscopy (see Methods sections 2.5.2c and 2.6.10). GFP-tagged proteins are shown in panels a, c, e, g, i, k & m and phalloidin staining of F-actin is shown in panels b, d, f, h. j, 1 & n.

GFP-tagged a2-chimaerin - (panels a-d), GFP-tagged a2-chimaerin (E49W) - (panels e, f), GFP-tagged a2-chimaerin (R56L) - (panels g, h), GFP-tagged a2-chimaerin (R73L) - (panels i, j), GFP-tagged a2-chimaerin (N94H) - (panels k, 1), GFP-tagged al-chimaerin - (panels m, n)

Figure 6.2 170 upon serum starvation the cells flatten and extend neurite-like processes (Jalink et al., 1994). LPA is the active component in serum which induces neurite retraction in N1E 115 cells and this LPA-induced retraction is a Rho-dependent process (Jalink et al., 1994). N1E 115 cells treated with the same serum conditions as transiently transfected cells plated on poly-L-lysine are shown in figure 6.1. After serum free treatment followed by a low serum incubation overnight, the majority of cells are flattened with short extensions (figure 6. la & b), a small percentage of cells have neurites longer than 2 cell bodies (figure 6. la & b), whilst a proportion of cells remain rounded.

6.3 Morphology of transiently transfected N1E 115 cells expressing a l- or a l- chimaerin N1E 115 cells plated on poly-L-lysine were transiently transfected with HA- or GFP-tagged a l- or a2-chimaerin DNA constructs and subsequently analysed by immunocytochemistry. N1E 115 cells expressing al-chimaerin were rounded and similar to undifferentiated N1E 115 cells (figure 6.2n), although no al-chimaerin expressing cells were neurite bearing, unlike control untransfected N1E 115 cells under the same conditions (figure 6.1). al-Chimaerin was expressed unevenly through the cell, often concentrated within one region (figure 6.2m). N1E 115 cells expressing a2-chimaerin were flattened and of normal size and a proportion of cells were neurite bearing (figure 6.2b & d), similar to control cells. Protein was expressed throughout the cell body (figure 6.2a & c) and in neurite bearing cells a2-chimaerin was also expressed throughout the length of the neurite and in the growth cone (figure 6.2a).

6.3.1 The effects of a2-chimaerin SH2 domain mutants on N1E 115 cell morphology The E49W, R56L and R73L SH2 domain mutants produced a morphology very similar to wild type a2-chimaerin (figure 6.2, compare f, h & j with b & d) and were similarly expressed throughout the cell (figure 6.2, compare e, g & i with a & c). However the N94H SH2 domain mutant produced a very different morphology. All N94H transfected cells were rounded (figure 6.21), similar to undifferentiated NIE 115 cells and N94H protein was unevenly expressed throughout the cells (figure 6.2k). This morphology very closely resembles that of al-chimaerin transfected cells (figure 6.2,

Chapter 6 171 compare k & 1 with m & n).

6.4 Morphology of permanently transfected N1E 115 cell lines overexpressing al- or a2-chimaerin pXJ41-HA DNA constructs containing a l- or a2-chimaerin sequence and the neomycin resistance gene were used to establish permanently transfected NIE 115 cell lines (see Methods section 2.5.4). The morphological effects of long term overexpression of a 1- or a2-chimaerin in N1E 115 cells were investigated by immunocytochemistry. Cell lines a 1-7, a 1-8, al-10 and a lD all expressed full length al-chimaerin, as established by Western analysis (figure 3.7). Of the eleven cell lines expressing full length a2-chimaerin, the a2-10 cell line expressed the highest levels of protein and was selected for further characterisation (figure 3.6). The morphology of the al-7, al-8, a lD and a2-10 cell lines, in addition to the HAv-24 vector control cell line and wild type N1E 115 cells, were examined in detail.

6.4.1 Quantification of cell morphology Five morphological categories were distinguished in the permanent cell lines on the basis of phalloidin staining of filamentous actin and are shown in figure 6.3. These were (1) rounded undifferentiated cells (figure 6.3a), (2) rounded cells with multiple filopodia or peripheral actin microspikes (figure 6.3b), (3) flattened cells (figure 6.3e), (4) cells with neurites longer than 2 cell bodies in length (figure 6.3c and d) and (5) enlarged, flattened cells with a diameter between 50pm and ~110pm (figure 6.3f). More than 600 cells were counted for cell lines grown in both the presence and absence of serum and these results are summarised in table 6.4 and in graphical form in figures 6.5 and 6.6. In the presence of serum, normal N1E 115 cells are predominantly rounded (-59%) and ~20pm in diameter (figure 6.3a), some cells are flattened (-27%) and a small percentage of cells have peripheral actin microspikes or neurites. In the absence of serum, which causes N IE 115 cells to differentiate, cells become flattened (-67%) and -10% of cells extend long neurites (>2 cell bodies in length), while a smaller percentage of cells have microspikes and some remain rounded (-21%) (see table 6.4). The morphology of the HAv-24 empty vector cell line in both the presence and absence of serum closely resembled that of untransfected N1E 115 cells, although slightly more

Chapter 6 172 Figure 6.3: Examples of the morphology of permanently transfected N1E 115 cell lines Permanently transfected N1E 115 cell lines and untransfected N1E 115 cells were plated on poly-L-lysine coated coverslips and incubated overnight in complete media or serum free media (see Methods section 2.4.5, 2.5.2c). Cells were then fixed, subjected to immunocytochemistry (see Methods section 2.6.10) and analysed by fluorescence and confocal microscopy. Phalloidin staining of F-actin is shown in panels a-f. The bar represents 25pm in all panels.

Types of morphology Rounded - N 1E 115 cells in the presence o f serum (panel a), Microspikes - oc 1 -7 cells in the presence of serum (panel b), Neurites - serum starved N 1E 115 cells (panels c & d), Flattened - serum starved N1E 115 cells (panel e), Enlarged - a2-10 cells in the presence of serum (panel f)

Figure 6.3 173 Table 6.4 : Quantification of the morphology of permanently transfected N1E 115 cell lines C/3 T3 Q • pp «4H 3 • -a pd "o - 2 +1 h fl rt CJ o a o a &H V Ui o a> a P WDo> o tj £ pp

Table 6.4 Table a 1-7 27.3 ± 6.9 70.5 ± 6.8 0.2 ± 0.9 2.0 ±2.8 100 820 VO r- r*H o o Q 8 20.4 ± 4.2 31.9 ±8.0 9.8 ±4.2 41.0 ±6.9 103.1 * rH oo r—1 in N o +i 8 l

23.1 ±6.1 24.3 ± 5.6 47,2 ± 7.4 106.4 * 753

NO SERUM o N1E 115 21.4 ±7.3 2.6 ± 2.4 66.6 ±8.1 9.4 ±3.1 100 1086 HAv-24 25.5 ±5.9 1.9 ±2.6 52.5 ±8.1 15.7 ±5.0 5.6 ±3.7 101.2 * 766 a a 1-7 26.4 ± 6.3 0.7 ±1.3 70.1 ±7.3 0.9 ±1.4 1.9 ±2.7 100 997 o © 8 18.8 ±6.7 50.2 ± 8.4 12.3 ±3.8 20.8 ± 6.3 102.1 * 758 rn in oo a2-10 +1 0.8 ± 2.2 41.2 ±9.8 16.9 ±6.6 30.5 ±7.1 103.1 * 744 c ) hi 174 Figure 6.5: Graphical representation of permanent N1E 115 cell line morphology in the presence of serum CP o ON o oo o s||o,i jo aSejuaruaj jo s||o,i Figure 6.5 Figure

75 Figure 6.6: Graphical representation of permanent N1E 115 cell line morphology in the absence of serum o o ON o oo o - r o | J U3DJ3J 3 J D 3 |U K § 3 JO 3 ||3 S o NO Figure 6.6 Figure o (N m o

Chapter 6 177 Figure 6.7: Examples of the morphology of permanently transfected N1E 115 cell lines overexpressing a l- or a2-chimaerin Permanently transfected N1E 115 cell lines were plated on poly-L-lysine coated coverslips and incubated overnight in complete media or serum free media (see Methods section 2.4.5, 2.5.2c). Cells were then fixed, subjected to immunocytochemistry (see Methods section 2.6.10) and analysed by fluorescence and confocal microscopy. Phalloidin staining of F-actin is shown in panels a-f. The bar represents 25pm in all panels. al-7 cells in the presence of serum (panels a-d), serum starved a2-10 cells (panel e), a2-10 cells in the presence of serum (panel f)

Figure 6.7 178 a2-chimaerin Phalloidin

Figure 6.8: Expression of a2-chimaerin protein in the permanently transfected a2- 10 cell line T he permanently transfected cc2-10 cell line which overexpresses a2-chimaerin, was plated on poly-L-lysine coated coverslips and cells were incubated overnight in complete media or serum free media (see Methods section 2.4.5, 2.5.2c). Cells were then fixed, subjected to immunocytochemistry (see Methods section 2.6.10) and analysed via fluorescence microscopy. a2-chimaerin expression detected by a polyclonal a2-chimaerin antibody is shown in panels a & c and phalloidin staining of F- actin is shown in panels b & d. The bar represents 25pm in all panels.

Serum starved a2-10 cells - (panels a-d)

Figure 6.8 179 al-8 and alD cell lines all expressed full length al-chimaerin in the triton insoluble fraction, as previously shown (figure 3.7). However al-chimaerin expression was also detected in the triton solubilised fraction of the a ID cell line. This difference in protein distribution may be involved in the morphological differences observed. However, the differences observed are not due to increased levels of a2-chimaerin expression in the a lD cell line, since al-7 , a l-8 and a lD cell lines all expressed similar levels of a2- chimaerin which corresponded to the endogenous levels observed in untransfected and HAv-24 vector control cells (figure 3.8).

6.5 The effects of potential a2-chimaerin targets on N1E 115 cell morphology The N terminal SH2 domain of a2-chimaerin is likely to mediate its effects through binding to target proteins, such as the previously identified proteins B 13 and TOAD-64. The specific neuronal expression pattern of a2-chimaerin, its interaction with B 13 and TOAD-64, the neuronal expression pattern of TOAD-64 itself (Mintum et al., 1995a,b) and the involvement of CRMP-62, the chick homologue of TOAD-64, and also Racl in collapsin-induced growth cone collapse (Goshima et al., 1995; Jin and Strittmatter, 1997) suggests these proteins may play a vital role in determining neuronal cell morphology. Thus the effects of these a2-chimaerin target proteins on N1E 115 cell morphology was investigated.

6.5.1 The effects of TOAD-64 and B13 on N1E 115 cell morphology N1E 115 cells plated on poly-L-lysine were transiently transfected with GFP- tagged TOAD-64 or B13 DNA constructs and subsequently analysed by immunocytochemistry. In both rounded and neurite bearing cells, TOAD-64 was expressed unevenly throughout the cell body of transfected cells. Its expression was punctate and it appeared to have a vesicular distribution (figure 6.9a & c). In neurite bearing cells, low levels of TOAD-64 expression were also detected at some regions down the neurite shaft (figure 6.9a). TOAD-64 expression was also concentrated in a perinuclear location within cells which may correspond to the Golgi apparatus (figure 6.9e & g). Thus transiently transfected TOAD-64 has a different overall distribution to a2-chimaerin in N1E 115 cells, although a minor proportion of protein does localise to neurites. B13 was expressed evenly throughout the cell body of transfected cells (figure 6.9i & k). In neurite bearing cells it was also expressed throughout the length of the

Chapter 6 180 GFP TOAD-64 Phalloidin

f t

w * >V

g

i*

Figure 6.9 181 GFP B13 Phalloidin

Figure 6.9: Expression of GFP-tagged TOAD-64 and B13 proteins in transiently transfected N1E 115 cells N 1E 115 cells were plated on poly-L-lysine coated coverslips and then serum starved overnight. Cells were then transiently transfected with GFP-tagged TOAD-64 or B13 DNA constructs in the absence of serum and incubated overnight in 5% serum. Cells were fixed 24 hours post transfection, subjected to immunocytochemistry and analysed by fluorescence microscopy (see Methods sections 2.5.2c and 2.6.10). GFP- tagged proteins are shown in panels a, c, e, g, i & k and phalloidin staining of F-actin is shown in panels b, d, f, h, j & 1.

GFP-tagged TOAD-64 - (panels a-h), GFP-tagged B13 - (panels i-1)

Figure 6.9 182 neurite and in the growth cone (figure 6.9k). Transiently transfected B13 has a similar distribution to ot2-chimaerin in N1E 115 cells.

6.5.2 p35 p35 is the neuron specific regulator of cyclin dependent kinase-5 (cdk5), which binds and activates the kinase (Lew et al., 1994; Tsai et al., 1994). Cdk5 phosphorylates the neuronal intermediate filament proteins NF-M and NF-H and also the microtubule associated proteins Tau and MAP2 (reviewed in Lew and Wang, 1995) which are involved in regulation of neuronal cytoskeletal dynamics. Both cdk5 and p35 promote neurite outgrowth in primary cortical neurons (Nikolic et al., 1996) and colocalise with Rac and PAK1 in lamellipodial rich areas of axonal growth cones (Nikolic et al., 1998). p35 is expressed at high levels in embryonic forebrain (Delalle et al., 1997) with similarities to a2-chimaerin expression (G. Michael, unpublished results) and an interaction between p35 and chimaerin has recently been detected (J. Wang and R. Qi, Personal Communication). In order to further investigate the relationship between p35 and chimaerin, its expression in chimaerin cell lines was investigated.

6.5.3 Expression of p35 in N1E 115 cells In untransfected and HAv-24 empty vector control N1E 115 cells, low levels of p35 expression were detected by immunocytochemistry. Although in undifferentiated, round N1E 115 cells little p35 expression was detected, p35 was weakly detected in a ring around the nucleus in flattened cells (figure 6.10c & d). This ring-like expression of p35 was also observed in Swiss 3T3 cells (Nikolic et al., 1998). In cells with neurites, p35 expression was concentrated down the shaft of the neurite and weak expression was typically observed at the end (figure 6.10a & b), the middle or the base of the neurite, but was seldom detected throughout its entire length. In the a2-10 and otID cell lines, the levels of p35 expression detected were several fold higher than those observed in control cells. The relative levels of p35 expression in untransfected N1E 115 cells and the a2-10 cell line are shown in figure 6.10 (compare a & c with e). In some flattened a2-10 cells (as well as a ID), p35 was expressed in a ring around the nucleus, similar to control cells (figure 6.10c & d), although the level of expression was much higher. In other flattened a2-10 cells, p35 expression was detected in novel coiled structures of varying length (figure 6.10e and figure 6. lib). In the enlarged, flattened cells which are unique to the a2-10 (and a ID)

Chapter 6 183 p35 Phalloidin

- j i

Figure 6.10: Expression of p35 in untransfected N1E 115 cells and the permanently transfected a2-10 cell line Untransfected N1E 115 cells and the permanently transfected a2-10 cell line which overexpresses a2-chimaerin, were plated on poly-L-lysine coated coverslips and incubated overnight in complete media or serum free media (see Methods section 2.4.5. 2.5.2c). Cells were then fixed, subjected to immunocytochemistry (see Methods section 2.6.10) and analysed via fluorescent confocal microscopy. The same confocal power settings were used to obtain the images shown. p35 staining is shown in panels a, c & e and phalloidin staining of F-actin is shown in panels b, d & f. The bar represents 25pm in all panels.

Serum starved N1E 115 cells - (panels a-d), Serum starved a2-10 cells - (panels e & f)

Figure 6.10 184 cell line, p35 was detected in short coiled structures or as a diffuse stain within the cell body. In neurite bearing cells in the a2-10 cell line, much stronger p35 staining was observed than in control cells. p35 expression was again detected at the base, middle or ends of neurite shafts, but sometimes throughout the entire length of the neurite (figure 6.1 lh). p35 was also detected in coiled structures which extended from the cell body down the neurite shaft in a2-10 cells (figure 6.1 le). In the a l-7 and a l-8 cell lines, the levels of p35 expression detected were only slightly higher than in control cells and not as high as the a2-10 and a ID cell lines. As with control cells, p35 was expressed in faint rings around the nucleus of flattened cells and little expression was detected in rounded cells or cells with multiple peripheral actin microspikes (no neurites were present in these cell lines even in the absence of serum). Occasionally, p35 was faintly detected in short coiled structures, similar to those present in the a2-10 (and a ID) cell line. The p35 detected in coiled structures in a2-10 and a lD cell lines and also to a lesser extent in a l-7 and a l-8 cell lines was not observed in untransfected or empty vector transfected N1E 115 cells. Hence this appears to be a novel characteristic of cells expressing a l- or a2-chimaerin.

6.5.4 Colocalisation of p35 and actin in the a2-10 cell line The coiled structures detected by the p35 antibody in cell lines expressing chimaerin were filamentous in appearance, suggestive of cytoskeletal components such as microtubules, intermediate filaments or actin filaments. The distribution of p35 and F-actin in these structures was investigated by confocal microscopy. Z sections were taken throughout the entire cell, the section with the strongest p35 staining was selected and colocalisation of p35 and F-actin was investigated. In the a2-10 cell line, p35 staining detected in both short and extensively coiled filamentous structures (figure 6.1 lb) colocalised with F-actin (figure 6.1 lc). However, not all of the p35 staining observed in the extensive coils colocalised with F-actin. In neurite bearing a2-10 cells, p35 staining detected in a filamentous structure which extended from the cell body down the neurite shaft (figure 6.1 le) strongly colocalised with F-actin (figure 6.1 If). When p35 staining was detected throughout the neurite shaft in a2-10 cells, again p35 strongly colocalised with F-actin (figure 6. lli). Thus p35 staining colocalises with F- actin in the novel filamentous structures observed in permanently transfected N IE 115 cells expressing a2-chimaerin.

Chapter 6 185 F-actin actin

colocalisation colocalisation

Figure 6.11 186 colncalisation

Figure 6.11: Colocalisation of p35 and F-actin in the permanently transfected a2- 10 cell line The permanently transfected a2-10 cell line which overexpresses a2-chimaerin, was plated on poly-L-lysine coated coverslips and cells were incubated overnight in complete media or serum free media (see Methods section 2.4.5). Cells were then fixed, subjected to immunocytochemistry (see Methods section 2.6.10) and analysed via fluorescent confocal microscopy. Z sections were performed and a single layer through each cell is shown. Phalloidin staining of F-actin is shown in panels a, d & g, p35 staining is shown in panels b, e & h and colocalisation of p35 and phalloidin staining is shown in panels c, f & i. a2-10 cells in the presence of serum - (panels a-c), Serum starved a2-10 cells - (panels d-i)

Figure 6.11 187 6.6 Summary In transiently transfected N1E 115 cells, a2-chimaerin and the E49W, R56L and R73L SH2 domain mutants were evenly expressed throughout the cell body and neurites of transfected cells. However, expression of al-chimaerin or the N94H SH2 domain mutant in N1E 115 cells produced a rounded cell morphology and protein was expressed unevenly in the cell body. Thus a point mutation in the SH2 domain of a2- chimaerin changed the protein distribution and induced a morphology similar to that of al-chimaerin. In the permanently transfected al-7 (and al-8 ) cell line, a novel rounded morphology with multiple peripheral actin microspikes was observed in the presence of serum (table 6.4, figure 6.3b, figure 6.7a-d). The presence of multiple peripheral actin microspikes suggests Cdc42 activity is increased in these cells. In the absence of serum, a l-7 cells flattened but the neurite extension induced in normal N1E 115 cells by serum withdrawal did not occur. Both observations are consistent w7ith overexpressed a l- chimaerin acting as an active RacGAP in vivo. This inhibition of neurite outgrowth was also observed in transiently transfected cells expressing al-chimaerin. In the a2-10 ceil line, -40% of cells had a novel enlarged and extremely flattened morphology (figure 6.3f, figure 6.7e & f, figure 6.8d). This flattened, spreading morphology is characteristic of Rac, suggesting its activity is increased rather than down-regulated in these cells. Both normal sized and enlarged a2-10 cells were also able to extend neurites in both the presence and absence of serum. Unexpectedly, the morphology of a single al-chimaerin cell line, a ID, was unlike other al-chimaerin cell lines but very similar to the a2-10 cell line. This was not due to increased levels of a2-chimaerin expression in the a lD cell line (see section 3.3.4, figure 3.8), which suggests that in the a lD line, either al-chimaerin has been mutated or upregulation of another pathway is responsible for these morphological effects. The expression of potential a2-chimaerin targets B 13, TOAD-64 and p35 in N1E 115 cells was investigated. In transiently transfected N1E 115 cells, B13 was expressed throughout the cell body, neurites and growth cones of cells, similar to a2- chimaerin, whilst TOAD-64 expression was punctate and concentrated mainly within the cell body, suggesting a vesicular distribution, with low levels of TOAD-64 expression in neurites. In untransfected or vector transfected N1E 115 cells, weak p35 expression was detected by immunocytochemistry. However in a2-10 and alD cell

Chapter 6 188 lines, greatly increased levels of p35 were detected. p35 and F-actin were found to strongly colocalise in coiled filamentous structures which were only detected in chimaerin expressing cell lines.

Chapter 6 CHAPTER SEVEN: Discussion

Discussion DISCUSSION

7.1 Rho p21s and the chimaerin RacGAPs Rho p21s are involved in many fundamental cellular processes regulating cell growth and motility. One of the best characterised is their effect on actin cytoskeleton reorganisation. Rho, Rac and Cdc42 are molecular switches. They cycle between the active GTP bound and inactive GDP bound form and this cycling is regulated by several classes of proteins; the GAPs, GEFs and GDIs. These regulatory proteins are vital in determining the balance between the GTP and GDP bound state of each p21 protein and the relative amounts of each active Rho p21 present in a system. Thus the actions of GAPs, GEFs and GDIs are essential in determining the net effects of combined Rho, Rac and Cdc42 activity. GTP bound Rho p21s interact with a variety of effector proteins including protein kinases, lipid kinases and phosphatases. These effector proteins often mediate other intermolecular interactions resulting in the formation of multi protein complexes. For example, PAK, in addition to binding GTP bound Rac or Cdc42, also interacts with the RacGEF PIX (Manser et al., 1998) and the adaptor protein Nek (Bagrodia et al., 1995). These other binding proteins may further affect the activity or localisation of the protein complex. Thus the activity of Rho p21s within the cell can involve the combined effects of many proteins. The chimaerins are a family of Rho p21 GAPs with tissue specific and developmentally regulated expression, a l- and a2-chimaerin are alternate splice products with divergent N terminal sequences. The divergent N terminal region of al- chimaerin is 58 amino acids in size and contains a 35 amino acid sequence which is predicted to form an amphipathic helix (Lim et al., 1992). The N terminal region of a2- chimaerin is 183 amino acids in size and contains an unusual SH2 domain of 81 amino acids (Hall et al., 1993). Both isoforms contain a common cysteine rich domain (CRD) and a C terminal RhoGAP domain (Hall et al., 1990; Hall et al., 1993). The different specificity of expression of these isoforms suggests that the alternate N terminal sequences have discrete regulatory or functional roles. The specific distribution and temporal expression pattern of these proteins suggests they may be involved in regulating the activity of Rho p21s in a neuronal environment. The presence of an SH2 domain in a2-chimaerin suggests this protein is likely to be involved in a tyrosine kinase signalling pathway. A well established example of

Discussion 191 receptor tyrosine kinase signalling is provided by activation of the EGF receptor. Activation induces receptor dimerisation and autophosphorylation of cytoplasmic tyrosine residues. This results in recruitment of the Grb2-Sos complex from the cytosol to the plasma membrane, where the SH2 domain of Grb2 binds a phosphotyrosine containing sequence of the receptor (Rozakis-Adcock, 1993). This translocation brings Sos into contact with its substrate Ras, resulting in Ras activation and stimulation of multiple Ras-dependent pathways including vulval development in C.Elegans and differentiation of photoreceptor cells in Drosophila (reviewed in Dickson and Hafen, 1994). a2-Chimaerin, (32-chimaerin and p i20 RasGAP are the only GAPs identified to date which also contain SH2 domains (Settleman et al., 1992b; Hall et al., 1993; Leung et al., 1994). The N terminal SH2 domains of RasGAP interact with the tyrosine phosphorylated PDGF (3 receptor and this interaction induces translocation of pi 20 RasGAP to the membrane (McGlade et al., 1993). It is possible that appropriate stimulation may similarly induce translocation of a2-chimaerin from the cytosol to the membrane via its SH2 domain. The two N terminal SH2 domains of pi 20 RasGAP also interact with tyrosine phosphorylated p i90 RhoGAP (Settleman et al., 1992b) and this interaction results in downregulation of pi 20 GAP activity (Moran et al., 1991). Thus in addition to the regulation of protein localisation, the SH2 domain of a2-chimaerin may also mediate interactions which affect protein function. The interaction between p i20 RasGAP and p i90 RhoGAP provides a link between Rho and Ras signalling pathways and p i20 has recently been shown to demonstrate a Rho effector function. Microinjection of full length pl20 RasGAP protein induced Rho-dependent stress fibre formation in Swiss 3T3 cells (Leblanc et al., 1998). This effect and also Ras- or PDGF-induced stress fibre formation were shown to depend on the SH3 domain of pi 20 RasGAP. An antibody raised against this region also inhibited LPA-induced neurite retraction in differentiated PC 12 cells. This SH3- dependent Rho effector function of pi 20 RasGAP is particularly interesting considering that the interaction of pi 20 RasGAP with pi 90 RhoGAP induces a conformational change in pl20 RasGAP structure which increases the accessibility of its SH3 domain (Hu and Settleman, 1997). Thus interaction with pl90RhoGAP promotes the Rho effector function of pi 20 RasGAP. These examples demonstrate that in addition to the induction of membrane translocation, SH2 domain-mediated interactions may also regulate the activity of other conserved domains within the protein and induce conformational changes in protein structure which considerably affect protein function.

Discussion 192 Similarly, the interactions of the a2-chimaerin SH2 domain are likely to be essential in determining protein function.

7.2 Distribution of a l- and a2-chimaerin in eukaryotic cells a l - and a2-Chimaerin were found to have different protein distributions. In fractions derived from permanently transfected N1E 115 cells overexpressing a l- chimaerin, the protein was detected in the 1% triton insoluble cytoskeletal fraction (figure 3.7). In cell lines overexpressing a2-chimaerin, expression was detected predominantly in the cytosolic fraction with low levels in the 1% triton solubilised membrane fraction (figure 3.5). Similar results were obtained in COS7 cells where al- chimaerin expression was detected in the 0.5% triton insoluble fraction (figure 3.4) whilst a2-chimaerin was detected in the 0.5% triton soluble fraction (figures 3.3 and 3.4). The cytoskeletal localisation of al-chimaerin was recently reported by Kozma et al., (1996) to require its N terminal 35 amino acids, whereas the GAP domain was not required for protein localisation. The contribution of the CRD to determining a l- chimaerin protein localisation has not been determined. It is possible that appropriate stimulation may induce translocation of a2- chimaerin from the cytosol to the membrane via its SH2 domain, so the effect of co­ expression of activated Rho p21s on the distribution of a2-chimaerin was investigated. However, expression of activated Rac, Rho or Cdc42 was not sufficient to induce translocation of a2-chimaerin from the cytosol (figure 4.1). This situation contrasts with that of the Rho effector ROK, which is translocated to the membrane by V14Rho (Leung et al., 1995). The fractions analysed were from a2-chimaerin expressing cell lines grown in the presence of serum. Thus stimulation by growth factors or serum starvation may be required to induce chimaerin activation and subsequent translocation to the plasma membrane or possibly to the cytoskeletal fraction of cells. The difference in a l- and a2-chimaerin protein localisation was also detected by immunocytochemistry. In transiently transfected N1E 115 cells, al-chimaerin was unevenly expressed within the cell body, mainly concentrated in a perinuclear location, whilst a2-chimaerin was evenly expressed throughout the cell body, neurites and growth cones of transfected cells (figure 6.2). A similar protein distribution for a2- chimaerin was also observed in permanently transfected N1E 115 cell lines overexpressing a2-chimaerin (figure 6.8). The distribution of al-chimaerin in permanently transfected cell lines was not determined since an al-chimaerin specific

Discussion 193 antibody was not available and HA-tagged al-chimaerin was not detected by the monoclonal HA antibody via immunocytochemistry. However, recently produced monoclonal antibodies raised against al-chimaerin could now be used to investigate the distribution of al-chimaerin in these overexpressing cell lines.

7.3 The effects of chimaerin overexpression on N1E 115 cell morphology Expression of a l- or a2-chimaerin has very different effects on the morphology ofN IE 115 cells. Transient expression of al-chimaerin-induced cell rounding, which is characteristically a Rho-dependent activity in these cells, and protein was unevenly expressed within the cell, mainly concentrated within the perinuclear region (figure 6.2, panel m). InN IE 115 cells transiently expressing a2-chimaerin, cells were flattened with neurites, similar to control untransfected cells. Protein was expressed throughout the cell body, neurites and growth cones of transfected cells (figure 6.2, panels a & c). Four permanent N1E 115 cell lines expressing full length al-chimaerin (alD, al-7, al-8 and al-10) and eleven permanent N1E 115 cell lines expressing full length a2-chimaerin were established. The permanently transfected al-7, al-8 and al-10 cell lines all displayed a similar morphology, which differed from that of a2-chimaerin overexpressing cell lines. The morphology of the al-7, a2-10, a lD and empty vector control cell lines was quantitated in detail (table 6.4). The a l-7 cell line displayed a rounded morphology with multiple peripheral actin microspikes in the presence of serum (table 6.4, figure 6.3, panel b and figure 6.7, panels a-d), which was not characteristic of control or a2-chimaerin overexpressing cell lines. Also neurite extension induced by serum withdrawal in normal N1E 115 cells was inhibited in the a l-7 cell line (table 6.4). In N1E 115 cells, Rac and Cdc42 are required for neurite outgrowth whilst Rho induces neurite retraction (Kozma et al., 1997). These two pathways compete in N1E 115 cells (Kozma et al., 1997; Van Leeuwen et al., 1997; Hirose et al., 1998) and the net morphological effects in these cells are determined by the relative activities of these proteins. Inhibition of neurite outgrowth in the a l-7 cell line suggests an inhibition of Rac or Cdc42 activity. However the presence of peripheral actin microspikes would suggest that Cdc42 signalling is active in these cells. These observations are consistent with overexpressed al-chimaerin functioning as a RacGAP in the a l-7 cell line, downregulating Rac activity resulting in unopposed or possibly increased Cdc42 activity. It is also possible that long term overexpression of a l- chimaerin may induce upregulation of other genes. Thus increased Cdc42 activity may

Discussion 194 be due to upregulation of a Cdc42 specific effector such as N-WASP (Miki et al., 1998) in this cell line. In the a2-10 cell line, -40% of cells had an enlarged and extremely flattened morphology (table 6.4, figure 6.3, panel f and figure 6.7, panels e & f), which was not characteristic of control or typical al-chimaerin expressing cell lines. Also serum withdrawal was not required to induce neurite outgrowth, as cells extended neurites in both the presence and absence of serum (table 6.4). This flattened, spreading morphology is characteristic of Rac, suggesting Rac activity is increased in these cells. As with the a l - isoform, a2-chimaerin inactivates Rac by increasing its GTPase activity, thus an increase in Rac type morphology is unexpected. Neurite outgrowth in the presence of serum suggests that LPA-induced neurite retraction, which is mediated by the Rho/ROK pathway (Hirose et al., 1998) is inhibited in these cells. Together this data suggests that Rac activity is in fact upregulated in the a2-10 cell line and Rac activation antagonises Rho signalling in these cells. Increased Rac activity may be due to upregulation of a RacGEF such as Tiaml or PIX. Overexpression of Tiaml induced cell spreading and neurite outgrowth in the presence of serum in N1E 115 cells (Van Leeuwen et al., 1997). LPA-induced neurite retraction was inhibited in these cells and this effect was only overcome by coexpression of V14Rho, illustrating that Rac activation can antagonise Rho signalling. However increased Rac activity may also be due to upregulation of a Rac effector such as PORI (Van Aelst et al., 1996). It is also possible that a2-chimaerin may in fact act as Rac effector in this situation. The Cdc42/Rac effector function of chimaerin has been previously shown by microinjection of recombinant protein in N1E 115 and Swiss 3T3 cells (Kozma et al., 1996). A single, atypical al-chimaerin expressing cell line, alD , had a morphology very similar to the a2-10 cell line (table 6.4). The a lD cell line appeared to express full length al-chimaerin protein (figure 3.7), although a small decrease in protein size would not be visible on the Western analysis performed. Protein expression was detected in both the membrane and cytoskeletal fractions of alD cells, unlike the other al-chimaerin expressing cell lines which expressed protein only in the cytoskeletal fraction (figure 3.7). This difference in protein localisation may be involved in the observed morphological differences to other al-chimaerin expressing cell lines. The similar morphology of the a lD and a2-10 cell lines was not due to upregulation of a2- chimaerin expression in the a lD cell line, as all al-chimaerin expressing cell lines expressed similar endogenous levels of a2-chimaerin (figure 3.8). This data suggests

Discussion 195 that the SH2 domain of a2-chimaerin is unlikely to be responsible for the observed morphology of a2-chimaerin overexpressing cell lines, since this region is not present in al-chimaerin, yet the same morphology was observed in the a ID cell line. Thus protein distribution and/or regulation of the CRD and GAP domains may be responsible for the observed morphology of a2-chimaerin overexpressing cell lines. The difference in protein distribution between a lD and other al-chimaerin expressing cell lines may be partly responsible for their observed differences in morphology. Alternatively, it may be a direct result of altered regulation of the CRD and GAP domains in the a ID cell line, which produces an a2-chimaerin like effect on morphology. This suggestion of altered regulation is supported by the observed effects of a single N94H amino acid substitution in the SH2 domain of a2-chimaerin, which results in an al-chimaerin-like morphology and protein distribution. However it is also possible that a compensatory mechanism such as upregulation of a RacGEF may be responsible for the Rac-like morphological effects observed in the a2-10 and a lD cell lines.

7.4 SH2 domain mutants of a2-chimaerin SH2 domain structure has been resolved for a number of SH2 domains. The crystal structure of Src in combination with its phosphotyrosine peptide target (Waksman et al., 1992) and the effects of point mutations within the SH2 domain on its function have been determined (Bibbins et al., 1993). These and other studies have provided much useful information concerning the contribution of various residues to SH2 domain function. Several highly conserved residues have been identified; the pAl tryptophan which is the first residue of the SH2 domain, the aA l arginine, the pD4 histidine and the invariant pB5 arginine at the base of the phosphotyrosine binding pocket which co-ordinates with the phosphate of the target peptide (Waksman et al., 1992). The a2-chimaerin SH2 domain is atypical as it has glutamate (E), not tryptophan (W) at the pAl position (Hall et al., 1993). It is one of only four SH2 domains identified in which the pAl residue is not tryptophan. The others are EAT-2 which starts with tyrosine, ZAP70 which has phenylalanine and p2-chimaerin which like a2-chimaerin has glutamate in this position (Thompson et al., 1996; Hatada et al., 1995; Leung et al., 1994, respectively). ZAP70 is also atypical in that it has tandem SH2 domains and its N terminal SH2 domain requires residues from the C terminal SH2 domain to form a complete phosphotyrosine binding pocket (Hatada et al., 1995). In Src, mutation of the

Discussion 196 PAl residue from tryptophan to glutamate abolishes binding to its tyrosine phosphorylated peptide ligand (Bibbins et al., 1993). Mutation of the a2-chimaerin SH2 domain p Al glutamate to the conventional tryptophan (E49W mutant) was expected to increase binding of the oc2-chimaerin SH2 domain to phosphotyrosine. However, in fact the E49W mutation had no effect on the binding of the a2 SH2 domain to phosphotyrosine agarose, although this is not a physiological substrate (G. Ferrari PhD thesis, 1999). However, it is possible that regions outside the SH2 domain of a2- chimaerin may also affect its target binding, as demonstrated for Src. Mutation of the Src pAl tryptophan to glutamate abolishes the phosphotyrosine binding ability of the isolated SH2 domain. However, if the Src SH3 domain is also present, this mutation no longer abolishes the phosphotyrosine binding ability of the SH2 domain (Bibbins et al., 1993). The highly conserved aA l arginine is involved in simultaneous recognition of the phosphate group and the aromatic ring of the phosphotyrosine residue. Surprisingly, mutation of the aA l arginine in Src only reduced phosphotyrosine binding by 20% (Bibbins et al., 1993). However, mutation of this residue in the a2-chimaerin SH2 domain (R56L mutant) completely abolished phosphotyrosine agarose binding (G. Ferrari PhD thesis, 1999). Thus the aA l arginine may have a more critical role in the SH2 domain of a2-chimaerin than in Src. Mutation of the invariant (3B5 arginine in Src and other SH2 domains completely abolished phosphotyrosine binding (Bibbins et al., 1993). Mutation of this residue in the a2-chimaerin SH2 domain (R73L mutant) had the same effect on phosphotyrosine agarose binding and in addition, abolished interaction of the a2- chimaerin SH2 domain with its target B13 (G. Ferrari PhD thesis, 1999). Thus the pB5 residue is important in the function of both Src and a2-chimaerin SH2 domains. In Src, the highly conserved (3D4 histidine is present at the mouth of the phosphotyrosine binding pocket. It is implicated in maintaining the geometry of the and may affect selection of specific peptide substrates. Mutation of this residue from histidine (H) to asparagine (N) in Src reduced phosphotyrosine binding by 20% (Bibbins et al., 1993). In a2-chimaerin, the pD4 residue is asparagine and mutation of this residue to histidine (N94H mutant) had no measurable effect on the phosphotyrosine agarose binding of the a2-chimaerin SH2 domain (G. Ferrari PhD thesis, 1999). Interestingly this mutation completely abolished interaction of the a2- chimaerin SH2 domain with its targets B13 and TOAD-64 (G. Ferrari PhD thesis,

Discussion 197 1999). Thus overall, three of the four SH2 domain mutations (R56L, R73L and N94H) affect either phosphotyrosine interactions, target protein interactions or both. However, as the effects of these SH2 domain mutations have been tested on a non physiological phosphorylated tyrosine substrate it is possible that they may mediate different effects on the binding of a specific phosphotyrosine containing target peptide. Similarly, as the TOAD-64 and B13 interactions with a2-chimaerin are phosphotyrosine independent, these SH2 domain mutations may have considerably different effects on the binding of a specific tyrosine phosphorylated target.

7.4.1 Distribution of a2-chimaerin SH2 domain mutants in eukaryotic cells Although E49W, R56L and R73L SH2 domain mutations had no effect on a2- chimaerin protein distribution (figure 3.3, lanes 2-4 and figure 3.4, lanes 4-6), the N94H mutation considerably altered the distribution of a2-chimaerin. In contrast to a2- chimaerin which was largely soluble (figure 3.3, lane 1 and figure 3.4 lane 3), the N94H mutant protein was detected mainly in the triton insoluble fraction of transiently transfected COS7 cells, with low expression in the triton soluble fraction (figure 3.3, lane 5 and figure 3.4, lane 7). This distribution closely resembled the protein distribution observed with al-chimaerin (figure 3.4, lane 2). Similarly, immunocytochemical analysis showed a2-chimaerin and the E49W, R56L and R73L SH2 domain mutants were evenly expressed throughout the cell body, neurites and growth cones of transfected N IE 115 neuroblastoma cells (figure 6.2, panels a, c, e, g, & i). However the N94H mutant was unevenly expressed within the cell body of transfected cells, often concentrated in a perinuclear location similar to the expression of al-chimaerin (figure 6.2, panels k & m). A single point mutation in the SH2 domain of a2-chimaerin alters the targeting of this protein so that it resembles that of al-chimaerin. The unique N terminal 35 amino acids of al-chimaerin are essential for its cytoskeletal distribution, the GAP domain is not required whereas the role of the CRD in al-chimaerin targeting is unknown (Kozma et al., 1996). Since the distribution of the N94H mutant so closely resembles that of al-chimaerin, yet lacks its unique N terminal sequence, this suggests that the CRD may also be involved in chimaerin protein targeting. The triton insoluble distribution of the N94H mutant may be consistent with activation of a2-chimaerin and its recruitment to the membrane or cytoskeleton. It is possible that the N94H mutation promotes a2-chimaerin translocation via disruption of

Discussion 198 the intra- or inter- molecular interactions responsible for the cytosolic distribution of the wild type protein. Mutation of this residue in the isolated SH2 domain does not impair phosphotyrosine agarose binding (G. Ferrari PhD thesis, 1999) which supports the possibility that this protein is active. However, it is also possible that this mutation causes a gross distortion of tertiary structure, affecting protein function since in the isolated SH2 domain, this mutation abolishes the interaction of a2-chimaerin with its targets B13 and TOAD-64 (G. Ferrari PhD thesis, 1999).

7.5 Potential regulation of a2-chimaerm activity by intra/inter molecular interactions Full length recombinant a2-chimaerin protein does not bind phosphotyrosine agarose as well as the isolated SH2 domain (G. Ferrari PhD thesis, 1999). This suggests the possibility that a2-chimaerin may dimerise or an intra-molecular interaction may occur in full length oc2-chimaerin which reduces the accessibility of the SH2 domain. The intra-molecular interaction may involve the SH2 domain itself, as in Src, where the SH2 domain interacts with a phosphorylated tyrosine residue in the C terminal, inhibiting kinase activity (Xu et al., 1997). However intra-molecular interactions may involve other regions of oc2-chimaerin, resulting in inhibition of SH2 domain function. It is possible that post translational modifications or phosphorylation are required in vivo to produce fully functional protein. The lack of these modifications in bacterially produced proteins may result in incorrect folding of the recombinant protein, leading to reduced accessibility of the a2-chimaerin SH2 domain. Finally, it is also possible that in vivo an interaction with another protein may sequester a2-chimaerin in a cytosolic complex, reducing its phosphotyrosine binding ability.

7.6 GAP activity of the chimaerins Full length recombinant al-chimaerin has 10 fold lower GAP activity than full length recombinant ot2-chimaerin or the isolated GAP domain (Ahmed et al., 1993; M. Teo PhD thesis, 1994). This difference in GAP activity is due to negative regulation of al-chimaerin GAP activity by the CRDin vitro (Ahmed et al., 1993). The similar GAP activities of full length a2-chimaerin and the isolated GAP domain in vitro (M. Teo PhD thesis, 1994) suggests that a2-chimaerin GAP activity is not negatively regulated by other regions within the protein. However a2-chimaerin GAP activity can be further upregulated by phospholipid binding to its CRD (M. Teo PhD thesis, 1994) whilst the

Discussion 199 GAP activity of al-chimaerin is both positively and negatively regulated by phospholipid binding to its CRD (Ahmed et al., 1993). The CRD and GAP domains of a l- and a2-chimaerin are identical in sequence, thus any differences in regulation of these domains in vitro must be due to the unique N terminal sequences of these proteins. This suggests that the effect of the CRD in negatively regulating the GAP activity of a2-chimaerin is attenuated by the presence of its SH2 domain. However, the relative GAP activities of the chimaerins in vivo may differ considerably from that observed in vitro, as in vivo they may be further regulated by interactions with other proteins.

7.6.1 Effects of the a2-chimaerin SH2 domain on GAP activity The E49W, R56L, R73L and N94H SH2 domain mutations had no measurable effect on the GAP activity of full length recombinant a2-chimaerin (G. Ferrari PhD thesis, 1999), which suggests that the SH2 domain does not directly regulate GAP activity in vitro. However it is possible that these mutations may affect the CRD- mediated regulation of GAP activity, although this has not been tested. This assay used recombinant proteins which any lack post translational modifications that may be required for fully functional proteins in a eukaryotic system so it is possible that these mutations may affect chimaerin GAP activity in vivo.

7.7 Effects of the common regions of the chimaerins on protein localisation The common region of the chimaerins contains a cysteine rich domain (CRD) and GAP domain, which may also be involved in determining the protein localisation. It has been previously shown that the GAP domain of al-chimaerin is not required for localisation of the protein to the cytoskeletal fraction of cells (Kozma et al., 1996), but the contribution of the CRD to al-chimaerin localisation has not been established. However, the similar protein distribution of al-chimaerin and the a2-chimaerin N94H SH2 domain mutant (which lacks the unique N terminal sequence present in a l- chimaerin) (figures 3.4 & 6.2) suggests that the common CRD may also be involved in determining protein localisation. The CRD present in both isoforms of chimaerin has homology to the Clb region of PKC, which binds both phorbol esters and DAG in a phospholipid-dependent manner and regulates PKC activity (Ohno et al., 1988; Ono et al., 1989). Both a l- and a2- chimaerin bind phorbol esters in the presence of phosphatidylserine with similar binding

Discussion 200 affinities to PKC (Ahmed et al., 1990; Ahmed et al., 1991; M. Teo PhD thesis, 1994). The effects of phorbol ester treatment on the distribution of the related protein p2- chimaerin have recently been investigated. The P-chimaerin isoforms have a similar structure to the a isoforms (figure 1.2) and the CRD of p2-chimaerin has been shown to act as a high affinity receptor for both phorbol esters and DAG (Caloca et al., 1997; Caloca et al., 1999). On stimulation with phorbol ester, p2-chimaerin translocates from a mainly soluble to a mainly cytoskeletal distribution in COS7 cells (Caloca et al., 1997). Upon binding DAG via its CRD domain, P2 translocates from the soluble to the particulate fraction in COS1 cells, which corresponds to translocation from the cytosol to a perinuclear region in these cells (Caloca et al., 1999). The PKC-related Cl region of RasGRP has also recently been shown to regulate its protein localisation. Upon serum or phorbol ester stimulation, RasGRP, a GEF for Ras, is translocated from the cytosol to membranous regions in NIH 3T3 cells (Tognon et al., 1998). This translocation was mediated via the CRD of RasGRP. Thus in addition to the unique N terminal region of the a-chimaerins, it is possible that the CRD may also contribute to determining the localisation of these proteins upon appropriate stimulation. Considering p2-chimaerin also contains an SH2 domain with 82% homology to that of a2-chimaerin (Leung et al., 1994), it is interesting that it is the CRD that has been demonstrated to induce P2-chimaerin protein translocation.

7.8 Chimaerin target proteins B13 and TOAD-64 are previously identified targets of the a2-chimaerin SH2 domain (C. Monfries, unpublished results, M. Teo PhD thesis, 1994). B13 is the 13 kDa subunit of NADH ubiquinone oxidoreductase and TOAD-64 is a phosphoprotein involved in axonal guidance (Mintum et al., 1995a,b). B13 was isolated using the yeast two hybrid system in the absence of tyrosine kinases and its interaction with the a2- chimaerin SH2 domain is phosphotyrosine independent (C. Monfries, Personal Communication). TOAD-64 was isolated during a screen for a2-chimaerin targets in rat brain fractions (M. Teo PhD thesis, 1994) and has consensus sites for serine and threonine but not tyrosine phosphorylation (Gaetano et al., 1997). p35 was recently identified as a potential target of both a l- and a2-chimaerin (J. Wang and R. Qi, Personal Communication). It is possible that other a2-chimaerin SH2 domain targets including a high affinity phosphotyrosine-dependent ligand exist which have not yet

Discussion 201 been identified. This high affinity interaction may be tissue specific or only occur during a particular expression period or under specific stimulation conditions, making its detection difficult. It is also possible that the a2-chimaerin SH2 domain may bind target proteins via the more unusual phosphoserine or phosphothreonine-dependent interactions as previously described for the interaction of the Abl SH2 domain with Bcr (Pendergast et al., 1991) and the GrblO SH2 domain with Rafl and MEK1 kinases (Nantel et al., 1998). The effects of SH2 domain mutations on TOAD-64 and B13 interactions in vitro supports the suggestion they are phosphotyrosine independent. The R56L SH2 domain mutation abolished phosphotyrosine agarose binding, but had no effect on TOAD-64 or B13 interactions with a2-chimaerin (G. Ferrari PhD thesis, 1999). Conversely, the N94H SH2 domain mutation abolished the interaction of a2-chimaerin with both B13 and TOAD-64, although this mutant was still able to bind phosphotyrosine (G. Ferrari PhD thesis, 1999). Mutation of the invariant pB5 arginine (R73L) of the a2-chimaerin SH2 domain abolished B13 and phosphotyrosine agarose binding, but had no effect on the interaction of TOAD-64 with a2-chimaerin (G. Ferrari PhD thesis, 1999), suggesting these proteins may interact with a2-chimaerin at different sites. However these binding sites may overlap since the N94H mutation had the same effect on both TOAD-64 and B13 interactions.

7.8.1 Distribution of B13 and TOAD-64 in eukaryotic cells In order to characterise the interactions of B13 or TOAD-64 with a2-chimaerin in vivo, constructs encoding these potential targets in mammalian expression vectors were made and the expression of these proteins was investigated. Full length GFP-tagged B13 expression was detected in both the soluble and insoluble fractions of COS7 cells, whilst expression of GFP-tagged TOAD-64 was only detected in the insoluble fraction (figure 3.9). This membranous distribution of TOAD- 64 agrees with previous data in rat cortex homogenates (Minturn et al., 1995b). Neither HA-tagged B13 or TOAD-64 expression was detectable on Western analysis using the HA antibody. Rabbit polyclonal antibodies raised against recombinant B 13 and TOAD- 64 were recently produced to aid further investigation of the interaction of these proteins with a2-chimaerin (G. Ferrari). The ability of these antibodies to detect eukaryotically expressed B13 and TOAD-64 was investigated. The polyclonal B 13 antibody strongly detected full length HA- and GFP-tagged B13 in both soluble and insoluble COS7 cell fractions (figure 3.10). This suggests that the inability to detect

Discussion 202 HA-tagged B13 using the HA antibody is probably due to folding of this protein obstructing the HA antibody binding site. Similarly, the rabbit polyclonal TOAD-64 antibody could not detect either HA- or GFP-tagged TOAD-64 in COS7 cell fractions, although it strongly detected recombinant protein and native TOAD-64 in rat brain cytosol (figure 3.11). As GFP-tagged TOAD-64 could be detected by the GFP antibody, and both the HA- and GFP-tagged TOAD-64 DNA constructs used produced full length protein in an in vitro transcription-translation assay (figure 3.1), it seems likely that the epitopes recognised by the TOAD-64 and HA antibodies are obscured in the tagged proteins.

7.8.2 Localisation of B13, TOAD-64 and p35 in eukaryotic cells The localisation of potential chimaerin targets B13, TOAD-64 and p35 in N1E 115 cells was investigated by immunocytochemistry. In transiently transfected N1E 115 cells, B13 was present throughout the cell body, neurites and growth cones of cells (figure 6.9, panels i & k), similar to a2-chimaerin, whilst TOAD-64 was punctate, suggesting a vesicular distribution (figure 6.9, panels a, c, e & g). p35 was faintly detected in untransfected N IE 115 cells. In flattened cells endogenous p35 was detected in a ring around the nucleus, whilst in neurite bearing cells, p35 was observed at the end, the middle or the base of the neurite shaft (figure 6.10, panels a & c). Greatly increased levels of p35 expression were detected in N1E 115 cell lines permanently overexpressing a2-chimaerin (figure 6.10, panel e). In these cells, p35 stained discrete areas of actin filaments which had a novel coiled structure (figure 6.11, panels c, f & i). Preliminary investigations indicate that (3-tubulin and neurofilament distribution in chimaerin cell lines does not colocalise with p35 (C. Hall, Personal Communication), although neurofilament and microtubule associated proteins are substrates of cdk5.

7.8.3 Interaction of protein targets with a2-chimaerin

7.8.3A a2-Chimaerin interaction with TOAD-64 Low levels of TOAD-64 protein co-immunoprecipitated with a2-chimaerin in transiently transfected COS7 cell lysates (figure 4.3B, lane 4). This suggests that extrinsic stimulation may be required to induce association of oc2-chimaerin and TOAD- 64 or other proteins may be required to enhance this interaction, which may be neural specific since both TOAD-64 and a2-chimaerin are neuronally expressed (Hall et al., 1993; Minturn et al., 1995a,b). However, TOAD-64 was not detected in chimaerin

Discussion 203 immunoprecipitates from permanently transfected N1E 115 cell lines (data not shown). This lack of interaction may be explained by a low level of endogenous TOAD-64 expression in this cell line. Unfortunately, time did not permit the transient transfection of ot2-10 cells with TOAD-64 to further investigate the interaction of these proteins. It is possible that Rac or other proteins may be required to promote association of a2- chimaerin and TOAD-64 in neural cells. TOAD-64 has homology to the C.Elegans unc- 33 protein, mutation of which results in abnormal axonal outgrowth and guidance (Li et al., 1992), while Rac promotes neurite outgrowth in N1E 115 (Kozma et al., 1997; Hirose et al., 1998) and PC12 cells (Lamoureux et al., 1997). It has also been shown that both Rac and TOAD-64 are required to mediate collapsin-induced growth cone collapse in chick DRGs (Goshima et al., 1995; Jin and Strittmatter, 1997). a2- Chimaerin binds both Rac and TOAD-64 via its GAP and SH2 domain, respectively. Thus it is possible that together these proteins form a complex which is involved in regulating neurite outgrowth or the collapsin pathway. Thus stimuli that promote these responses may be required to induce association of these proteins.

7.8.3B a2-Chimaerin interaction with BI3 The co-immunoprecipitation of B13 and a2-chimaerin could not be detected in COS7 cell lysates due to technical difficulties. The recent production of a polyclonal B13 antibody which detects both HA- and GFP-tagged B13 would now enable investigation of this interaction. However, B13 was not detected in chimaerin immunoprecipitates from permanent N1E 115 cell lines using this antibody (data not shown). As B13 is not a neuronal-specific protein, this lack of interaction may reflect low levels of endogenous B13 expression in this cell line, although this is unlikely as B13 is a mitochondrial protein. However it is also possible that specific stimuli or additional proteins may be required to induce association of B13 and a2-chimaerin in vivo.

7.8.3C Chimaerin interactions with p35 p35, the neuronal regulator of cyclin dependent kinase-5 (cdk5) which binds and activates the kinase (Lew et al., 1994; Tsai et al., 1994), co-immunoprecipitated with both a l- and a2-chimaerin in permanently transfected N1E 115 cell lines (figure 4.4). This suggests that the interaction was mediated by the common C terminal region of the chimaerins which contains the CRD and GAP domains or via association of p35 with a second protein which binds both chimaerin isoforms. Co-immunoprecipitation of p35

Discussion 204 with a l - and a2-chimaerin is very interesting, as p35 and cdk5 are required for neurite outgrowth (Nikolic et al., 1996), a process which also requires Rac in several cell types (Kozma et al., 1997; Kita et al., 1998). Recently, Rac and the p35/cdk5 complex were shown to colocalise in neuronal growth cones together with PAK1 (Nikolic et al., 1998). It was also demonstrated that p35 binds GTP bound Rac and thus may act as a Rac effector in neural cells (Nikolic et al., 1998). It is possible that association of p35 with chimaerin may recruit chimaerin to the Rac-p35/cdk5 complex in neural cells, enabling chimaerin-mediated regulation of Rac activity and hence regulation of neurite outgrowth. Binding of p35 and chimaerin may also enable phosphorylation of chimaerin by cdk5, which may greatly affect protein function. Increased expression of p35 was detected in the a2-10 cell line compared to control and al-chimaerin expressing cell lines (figure 6.10). Although the Rac effector function of p35 has not been determined, it is possible that p35 may promote Rac- dependent cell flattening. Thus the upregulation of p35 expression in cells overexpressing a2-chimaerin may be responsible for the novel enlarged, flattened morphology observed in this cell line. These enlarged a2-10 cells often contained novel coiled filaments of F-actin which colocalised with p35 staining (figure 6.1 la-c). Colocalisation of p35 and F-actin in filamentous structures was also observed in neurite bearing a2-10 cells (figure 6.1 ld-i). It would be interesting to determine whether Rac, cdk5, PAK1 or a2-chimaerin itself are also present in these structures, as they may represent a site of complex formation. However further investigation of the nature of these unusual structures was not possible due to time constraints.

7.8.3D a2-Chimaerin interactions with tyrosine phosphorvlated proteins A -130 kDa tyrosine phosphorylated protein co-immunoprecipitated with both a l- and a2-chimaerin in permanently transfected N1E 115 cell lines (figure 4.5). This protein probably interacted with the common C terminal region of the chimaerins which contains the CRD and GAP domains or associated with chimaerin indirectly via a second protein which binds both chimaerin isoforms. However, a -180 kDa tyrosine phosphorylated protein also co-immunoprecipitated with a2-chimaerin but was not detected in immunoprecipitates from any of the al-chimaerin or control cell lines (figure 4.4). This protein may be a specific target of the a2-chimaerin SH2 domain. The low level of this 180 kDa tyrosine phosphorylated protein detected in the immunoprecipitation sample from the a2-10 cell line represents its endogenous

Discussion 205 expression level. It is possible that stimulation by growth factors or other agents may be required to upregulate expression of this protein and promote its interaction with ot2- chimaerin. The only other tyrosine phosphorylated protein identified which interacts with a2-chimaerin is a -38 kDa protein from NGF stimulated PC 12 cells which bound an a2-chimaerin SH2 domain column (Hall et al., 1993). The identity of these proteins is unknown, however there are several possible candidates for the 180 kDa tyrosine phosphorylated protein. Several proteins involved in the regulation of cytoskeletal reorganisation are approximately 180 kDa in size and phosphorylated on tyrosine residues including FAK, ROK and Ack. Other tyrosine phosphorylated proteins of an appropriate size include pi 90 RhoGAP, the EGF receptor, the PDGF receptor, the 2B subunit of the NMD A glutamate receptor and insulin receptor substrate-1 (IRS-1) (Ellis et al., 1990; Kawase et al., 1995; Quinones et al., 1991; Moon et al., 1994; Sun et al., 1992, respectively). The activated PDGF receptor is an interesting possible target of ot2-chimaerin, as PDGF induces Rac activation and its subsequent morphological effects in many cell types. Recruitment of a2-chimaerin to the activated PDGF receptor via phosphotyrosine interactions would provide a mechanism for the localisation of a2-chimaerin to the site of Rac action at the plasma membrane, in order to regulate its activity. TOAD-64 was also recently found to co-immunoprecipitate with an -190 kDa serine phosphorylated protein from PC 12 cells (Kamata et al., 1998). This protein may also be phosphorylated on tyrosine residues and/or the 180 kDa tyrosine phosphorylated protein which co- immunoprecipitated with a2-chimaerin may also be phosphorylated on serine or threonine residues. Thus it is possible that both a2-chimaerin and TOAD-64 may interact with a common phosphoprotein.

7.9 The chimaerins and NFkB signalling Rac stimulates the production of ROS in many cell types (Sundaresan et al., 1996; Irani et al., 1997; Joneson and Bar Sagi, 1998) and was shown to be required for

Rac-induced NFk B activation in HeLa cells (Sulciner et al., 1996). The previously identified a2-chimaerin SH2 domain target B 13 is also implicated in the production of ROS (C. Hall, unpublished results). Whilst TOAD-64 was identified in a neural trans­ plasma membrane oxidoreductase (PMO) (Bulliard et al., 1997), an antioxidant enzyme complex which is activated in response to cellular stress and ROS. Thus the role of the

chimaerins and potential ot2-chimaerin targets in the ROS responsive NFk B signalling

Discussion 206 pathway were investigated.

7.9.1 NFkB signalling in HeLa and N1E 115 cells

As previously reported, both V12Rac and DL-ip stimulated NFk B activity in HeLa cells (Sulciner et al., 1996, figures 5.1 and 5.3). MEKK1 was also shown to

stimulate NFk B activity in HeLa cells (figure 5.1) which is consistent with its

ability to act upstream of the Ik B kinase complex (IKK) and induce its activation (Lee

et al., 1997), resulting in degradation of the NFk B inhibitor Ik Boc , and activation of

NFk B (figure 1.3). al-Chimaerin appeared to inhibit IL-ip-induced NFk B activation in HeLa cells (figure 5.3), which was previously shown to be Rac-dependent (Sulciner et

al., 1996). This suggests a role for al-chimaerin in regulating Rac-induced NFkB activation probably via its RacGAP activity.

Unexpectedly, Rho, Rac and Cdc42 did not induce NFk B activation in N1E 115 cells unlike in COS7, HeLa and Jurkat T cells (Sulciner et al., 1996; Perona et al.,

1997). In fact, V12Rac appeared to inhibit NFk B activation in these cells. At low DNA

levels, V12Rac induced a 5 fold drop in NFk B activity compared to empty vector controls (figure 5.4), whilst at higher DNA levels, both V12Rac and B13 induced a 7

fold drop in activity (figure 5.6). It is possible that in N1E 115 cells NFk B is

constitutively activated and that Rac is acting as an NFk B inhibitor in these cells.

However, as with HeLa cells, the JNK activator MEKK1 stimulated NFk B activity in N1E 115 cells (figure 5.4), which had not been previously reported. Rac and Cdc42

were recently shown to act upstream of MEKK1 to induce activation of NFk B

(Montaner et al., 1998). However since neither protein activates NFk B in N1E 115

cells, a different pathway must exist for MEKK1-induced NFk B activation in these cells. The observed decreases in activity in response to V12Rac and B13 may be due to specific inhibition of NFkB. However, attempts to induce NFkB activation in N1E

115 cells via treatment with UV irradiation, TNFa, PMA, lipopolysaccharide and H 2O2 were unsuccessful, thus inhibition of NFkB activation by V12Rac and B13 could not be confirmed. The NFkB signalling pathway does function inN IE 115 cells, as MEKK1 induced 5-10 fold activation (figure 5.4). However MEKK1 was presumed to act downstream of these potential inhibitors on the basis of previously published pathways (see figure 1.3). The inability of the Rho p21s to stimulate the NFkB pathway represents an interesting neural cell type specific difference and suggests that novel pathways exist

Discussion 207 for the regulation of NFk B activity in N1E 115 cells. It is possible that Rho p21s do not activate any stress-induced signalling pathways in N1E 115 cells. However it is also

possible that Rho p21s may activate JNK and/or p38 pathways, rather than theNF k B pathway in these cells.

Discussion 208 CONCLUSIONS

a l - and a2-Chimaerin are neuronally expressed isoforms of a GTPase activating protein (GAP) for p21-Rac, which suggests a role for these proteins in the regulation of Rac-mediated signalling pathways in neural cells. They are alternate splice products with divergent N terminal sequences. The N-terminal of al-chimaerin contains a predicted amphipathic helix whilst the N-terminal of a2-chimaerin contains an unusual SH2 domain. The distribution and morphological effects of these proteins and a2- chimaerin SH2 domain mutants in the N1E 115 neuroblastoma cell line have been investigated. a l - and a2-Chimaerin have different protein distributions and effects on morphology in transfected cells, al-Chimaerin was expressed in the cytoskeletal fraction whilst a2-chimaerin was mainly expressed in the cytosolic fraction of cells. In transiently transfected N1E 115 cells, al-chimaerin expression was concentrated in the perinuclear region and its expression induced cell rounding, whilst a2-chimaerin was expressed throughout the cell body, neurites and growth cones of cells. A single point mutation in the SH2 domain of a2-chimaerin (N94H) induced an al-chimaerin-like protein distribution and morphology in transiently transfected cells. Permanent overexpression of al-chimaerin in N1E 115 cells induced a novel rounded cell morphology with multiple peripheral actin microspikes in the presence of serum and also inhibited neurite outgrowth in response to serum withdrawal. This suggests that al-chimaerin acts as an active RacGAP, resulting in downregulation of Rac activity and consequent upregulation of Cdc42 activity in these cells. Permanent overexpression of a2-chimaerin in N1E 115 cells induced a novel enlarged and extremely flattened cell morphology and also induced neurite outgrowth in the presence of serum. The increased Rac-like cell spreading morphology suggests Rac activity is upregulated, not down regulated in these cells, which is also supported by the enhanced neurite outgrowth. This may be due to upregulation of a RacGEF or a Rac effector, but also suggests the possibility that a2-chimaerin may function as a Rac effector rather than a RacGAP in these cells. An 180 kDa tyrosine phosphorylated protein was identified as a potential target of the a2-chimaerin SH2 domain and p35, the neuronal regulator of cdk5, was confirmed as a potential target of both a l- and a2-chimaerin. p35 expression was increased in N1E 115 cells permanently overexpressing a2-chimaerin and colocalised

Conclusions 209 with F-actin in novel filamentous structures. The previously identified a2-chimaerin SH2 domain target TOAD-64 was expressed in the insoluble fraction of transfected cells, whilst B 13 was expressed in both the soluble and insoluble fractions. TOAD-64 expression in transiently transfected N1E 115 cells was punctate, suggesting a vesicular distribution whilst B13 was expressed throughout the cell body, neurites and growth cones of transfected cells, similar to a2-chimaerin. Co-immunoprecipitation of a2- chimaerin and TOAD-64 was demonstrated in transiently transfected cells.

The role of the chimaerins in the ROS responsive NFk B signalling pathway were investigated, al-Chimaerin appeared to inhibit the Rac-dependent stimulation of

NFk B activity induced by treatment of HeLa cells with IL-1(3, suggesting a possible

role for al-chimaerin in regulating Rac-induced NFk B signalling in these cells.

Unexpectedly, Rho p21s did not stimulate NFk B activity in N1E 115 cells. This

represents a cell type specific difference in NFk B signalling.

Conclusions 210 REFERENCES

References 211 Aasheim, H.C., Pedeutour, F., & Smeland, E.B. (1997). Characterisation, expression and chromosomal localisation of a human gene homologous to the mouse Lsc oncogene, with strongest expression in haematopoetic tissues. Oncogene, 14, 1747-1752.

Abe, H., Obinata, T., Minamide, L.S., & Bamburg, J.R. (1996). Xenopus laevis actin- depolymerising factor cofilin: a phosphorylation-regulated protein essential for development. J. Cell Biol, 132, 871-885.

Abo, A., Pick, E., Hall, A., Totty, N., Teahan, C.G., & Segal, A.W. (1991). Activation of the NADPH oxidase involves the small GTP-binding protein p21Racl. Nature, 353, 668-670.

Adra, C.N., Ko, J., Leonard, D., Wirth, L.J., Cerione, R.A., & Lim, B. (1993). Identification of a novel protein with GDP dissociation inhibitor activity for the Ras-like proteins Cdc42Hs and Rac-1. Genes Chromosomes & Cancer, 8, 253-261.

Adra, C.N., Manor, D., Ko, J.L., Zhu, S.C., Horiuchi, T., Van Aelst, L., Cerione, R.A., & Lim, B. (1997). RhoGDI gamma: a GDP-dissociation inhibitor for Rho proteins with preferential expression in brain and pancreas. Proc. Natl. Acad Sci. USA, 94, 4279- 4284.

Agnel, M., Roder, L., Vola, C., & GriffinShea, R. (1992). A Drosophila Rotund transcript expressed during spermatogenesis and imaginal disk morphogenesis encodes a protein which is similar to human Rac GTPase-activating (RacGAP) proteins. Mol. Cell. Biol., 12, 5111-5122.

Ahmed, S., Kozma, R., Monfries, C., Hall, C., Lim, H.H., Smith, P., & Lim, L. (1990). Human brain n-chimaerin cDNA encodes a novel phorbol ester receptor. Biochem. J., 272, 767-773.

Ahmed, S., Kozma, R., Lee, J., Monfries, C., Harden, N., & Lim, L. (1991). The cysteine-rich domain of human proteins, neuronal chimaerin, protein kinase C and diacylglycerol kinase binds zinc. Evidence for the involvement of a zinc-dependent structure in phorbol ester binding. Biochem. J., 280, 233-241.

Ahmed, S., Lee, J., Kozma, R., Best, A., Monfries, C., & Lim, L. (1993). A novel functional target for tumour-promoting phorbol esters and lysophosphatidic acid. The p21Rac-GTPase activating protein n-chimaerin. J. Biol. Chem., 268, 10709-10712.

Alberts, A.S. & Treisman, R. (1998). Activation of RhoA and SAPK/JNK signalling pathways by the RhoA-specific exchange factor mNETl.EMBO J., 17, 4075-4085.

Aletta, J.M. & Greene, L.A. (1988). Growth cone configuration and advance - a time- lapse study using video-enhanced differential interference contrast microscopy. J. Neurosci., 8, 1425-1435.

Allen, W.E., Jones, G.E., Pollard, J.W., & Ridley, A.J. (1997). Rho, Rac and Cdc42 regulate actin organisation and cell adhesion in macrophages. J. Cell Sci., 110, 707-720.

Allen, W.E., Zicha, D., Ridley, A.J., & Jones, G.E. (1998). A role for Cdc42 in macrophage chemotaxis. J. Cell Biol, 141, 1147-1157.

References 212 Amano, M., Ito, M., Kimura, K., Fukata, Y., Chihara, K., Nakano, T., Matsuura, Y., & Kaibuchi, K. (1996a). Phosphorylation and activation of myosin by Rho-associated kinase (Rho-kinase). J. Biol. Chem., 271, 20246-20249.

Amano, M., Mukai, H., Ono, Y., Chihara, K., Matsui, T., Hamajima, Y., Okawa, K., Iwamatsu, A., & Kaibuchi, K. (1996b). Identification of a putative target for Rho as the serine-threonine kinase protein kinase N. Science, 271, 648-650.

Amano, M., Chihara, K., Kimura, K., Fukata, Y., Nakamura, N., Matsuura, Y., & Kaibuchi, K. (1997). Formation of actin stress fibres and focal adhesions enhanced by Rho- kinase. Science, 275, 1308-1311.

Amano, M., Chihara, K., Nakamura, N., Fukata, Y., Yano, T., Shibata, M., Ikebe, M., & Kaibuchi, K. (1998). Myosin ii activation promotes neurite retraction during the action of Rho and Rho-kinase. Genes To Cells, 3, 177-188.

Andra, K., Lassmann, H., Bittner, R., Shomy, S., Fassler, R., Propst, F., & Wiche, G. (1997). Targeted inactivation of plectin reveals essential function in maintaining the integrity of skin, muscle, and heart cytoarchitecture. Genes Dev., 11, 3143-3156.

Andreoli, C., Martin, M., Leborgne, R., Reggio, H., & Mangeat, P. (1994). Ezrin has properties to self-associate at the plasma-membrane. J. Cell Sci., 107, 2509-2521.

Arber, S., Barbayannis, F.A., Hanser, H., Schneider, C., Stanyon, C.A., Bernard, O., & Caroni, P. (1998). Regulation of actin dynamics through phosphorylation of cofilin by LIM-kinase. Nature, 393, 805-809.

Arpin, M., Algrain, M., & Louvard, D. (1994). Membrane-actin microfilament connections - an increasing diversity of players related to band-4.1. Curr. Opin. Cell Biol., 6, 136-141.

Aspenstrom, P., Lindberg, U., & Hall, A. (1996). Two GTPases, Cdc42 and Rac, bind directly to a protein implicated in the immunodeficiency disorder Wiskott-Aldrich syndrome. Curr. Biol., 6, 70-75.

Aspenstrom, P. (1997). A Cdc42 target protein with homology to the non-kinase domain of FER has a potential role in regulating the actin cytoskeleton. Curr. Biol., 7, 479-487.

Bae, Y.S., Kang, S.W., Seo, M.S., Baines, I.C., Tekle, E., Chock, P.B., & Rhee, S.G. (1997). Epidermal growth factor (EGF)-induced generation of hydrogen peroxide - role in EGF receptor-mediated tyrosine phosphorylation. J. Biol. Chem., 272, 217-221.

Baeuerle, P. A. & Baltimore, D. (1996). NF-kappa B: ten years after. Cell, 87, 13-20.

Baeuerle, P. A. & Henkel, T. (1994). Function and activation of NF-kappa-B in the immune-system. Ann. Rev. Immunol., 12, 141-179.

Bagrodia, S., Taylor, S.J., Creasy, C.L., Chernoff, J., & Cerione, R.A. (1995). Identification of a mouse p21Cdc42/Rac activated kinase. J. Biol. Chem., 270, 22731- 22737.

References 213 Baldwin, A.S. (1996). The NF-kappa B and I kappa B proteins: new discoveries and insights. Ann. Rev. Immunol., 14, 649-683.

Bar Sagi, D., Rotin, D., Batzer, A., Mandiyan, V., & Schlessinger, J. (1993). SH3 domains direct cellular-localisation of signalling molecules. Cell, 74, 83-91.

Barbacid, M. (1987). Ras genes. Ann. Rev. Biochem., 56, 779-827.

Barfod, E.T., Zheng, Y., Kuang, W.J., Hart, M.J., Evans, T., Cerione, R.A., & Ashkenazi, A. (1993). Cloning and expression of a human Cdc42 GTPase-activating protein reveals a functional SH3-binding domain. J. Biol. Chem., 268, 26059-26062.

Bashour, A.M., Fullerton, A.T., Hart, M.J., & Bloom, G.S. (1997). IQGAP1, a Rac- and Cdc42-binding protein, directly binds and cross-links microfilaments. J. Cell Biol., 137, 1555-1566.

Bellanger, J.M., Lazaro, J.B., Diriong, S., Fernandez, A , Lamb, N., & Debant, A. (1998). The two guanine nucleotide exchange factor domains of Trio link the Racl and the Rho A pathwaysin vivo. Oncogene, 16, 147-152.

Bentley, D. & O'Connor, T.P. (1994). Cytoskeletal events in growth cone steering. Curr. Opin. Neurobiol., 4, 43-48.

Bentley, D. & Toroian-Raymond, A. (1986). Disoriented pathfinding by pioneer neuron growth cones deprived of filopodia by cytochalasin treatment. Nature, 323, 712-715.

Berridge, M.J. (1993). Inositol trisphosphate and calcium signaling. Nature, 361, 315- 325.

Berryman, M., Gary, R., & Bretscher, A. (1995). Ezrin oligomers are major cytoskeletal components of placental microvilli - a proposal for their involvement in cortical morphogenesis./. Cell Biol., 131, 1231-1242.

Berstein, G., Blank, J.L., Jhon, D.Y., Exton, J.H., Rhee, S.G., & Ross, E.M. (1992). Phospholipase C-beta-1 is a GTPase-activating protein for Gq/11, its physiological regulator. Cell, 70, 411-418.

Bibbins, K.B., Boeuf, H., & Varmus, H.E. (1993). Binding of the Src SH2 domain to phosphopeptides is determined by residues in both the SH2 domain and the phosphopeptides. Mol. Cell. Biol., 13, 7278-7287.

Bione, S., Sala, C., Manzini, C., Arrigo, G., Zuffardi, O., Banfi, S., Borsani, G., Jonveaux, P., Philippe, C., Zuccotti, M., Ballabio, A., & Toniolo, D. (1998). A human homologue of the Drosophila Melanogaster diaphanous gene is disrupted in a patient with premature ovarian failure: evidence for conserved function in oogenesis and implications for human sterility. Am. J. Hum. Genet., 62, 533-541.

Blaikie, P., Immanuel, D., Wu, J., Li, N.X., Yajnik, V., & Margolis, B. (1994). A region in She distinct from the SH2 domain can bind tyrosine-phosphorylated growth-factor receptors. J. Biol. Chem., 269, 32031-32034.

References 214 Boguski, M.S. & McCormick, F. (1993). Proteins regulating Ras and its relatives. Nature, 366, 643-654.

Bokoch, G.M. (1994). Regulation of the human neutrophil NADPH oxidase by the Rac GTP-binding proteins. Curr. Opin. Cell Biol, 6, 212-218.

Bokoch, G.M., Vlahos, C.J., Wang, Y., Knaus, U.G., & Traynorkaplan, A.E. (1996a). Rac GTPase interacts specifically with phosphatidylinositol 3-kinase. Biochem. J., 315, 775-779.

Bokoch, G.M., Wang, Y., Bohl, B.P., Sells, M.A., Quilliam, L. A., & Knaus, U.G. (1996b). Interaction of the Nek adapter protein with p21-activated kinase (PAK1). J. Biol. Chem., Ill, 25746-25749.

Bollag, G. & McCormick, F. (1991). Regulators and effectors of Ras proteins. Ann. Rev. Cell Biol., 1, 601-632.

Borg, J.P., Ooi, J., Levy, E., & Margolis, B. (1996). The phosphotyrosine interaction domains of xl 1 and fe65 bind to distinct sites on the YENptyr motif of amyloid precursor protein. Mol. Cell. Biol., 16, 6229-6241.

Bork, P. & Margolis, B. (1995). A phosphotyrosine interaction domain. Cell, 80, 693- 694.

Bork, P., Schultz, J., & Ponting, C.P. (1997). Cytoplasmic signaling domains: the next generation. Trends Biochem. Sci., 22, 296-298.

Boshart, M., Weber, F., Jahn, G., Dorschhasler, K., Fleckenstein, B., & Schaffner, W. (1985). A very strong enhancer is located upstream of an immediate early gene of human cytomegalovirus. Cell, 41, 521-530.

Bourne, H.R., Sanders, D.A., & McCormick, F. (1990). The GTPase superfamily - a conserved switch for diverse cell functions. Nature, 348, 125-132.

Bourne, H.R., Sanders, D.A., & McCormick, F. (1991). The GTPase superfamily - conserved structure and molecular mechanism. Nature, 349, 117-127.

Bowtell, D., Fu, P., Simon, M., & Senior, P. (1992). Identification of murine homologues of the Drosophila son of sevenless gene - potential activators of Ras. Proc. Natl. Acad. Sci. USA, 89, 6511-6515.

Bray, D. & Chapman, K. (1985). Analysis of microspike movements on the neuronal growth cone. J. Neurosci., 5, 3204-3213.

Bretscher, A., Gary, R., & Berryman, M. (1995). Soluble ezrin purified from placenta exists as stable monomers and elongated dimers with masked C-terminal-ezrin-radixin- moesin association domains. Biochemistry, 34, 16830-16837.

Brill, S., Li, S.H., Lyman, C.W., Church, D.M., Wasmuth, J.J., Weissbach, L., Bernards, A., & Snijders, A.J. (1996). The Ras GTPase-activating-protein-related human protein

References 215 IQGAP2 harbours a potential actin binding domain and interacts with calmodulin and Rho family GTPases. Mol. Cell B iol, 16, 4869-4878.

Brown, K., Gerstberger, S., Carlson, L., Franzoso, G., & Siebenlist, U. (1995). Central of I-kappa-B-alpha proteolysis by site-specific, signal-induced phosphorylation. Science, 267, 1485-1488.

Bryan, J. & Wilson, L. (1971). Are cytoplasmic microtubules heteropolymers? Proc. Natl Acad. Sci.USA, 68, 1762-1766.

Bulliard, C., Zurbriggen, R., Tomare, J., Faty, M., Dastoor, Z., & Dreyer, J.L. (1997). Purification of a dichlorophenol-indophenol oxidoreductase from rat and bovine synaptic membranes: tight complex association of a glyceraldehyde-3-phosphate dehydrogenase isoform, TOAD64, enolase-gamma and aldolase c. Biochem. J., 324, 555-563.

Byk, T., Dobransky, T., Cifuentes-Diaz, C., & Sobel, A. (1996). Identification and molecular characterisation of unc-33-like phosphoprotein (Ulip), a putative mammalian homologue of the axonal guidance-associated unc-33 gene product. J. Neurosci., 16, 688-701.

Byk, T., Ozon, S., & Sobel, A. (1998). The Ulip family phosphoproteins - common and specific properties. Eur. J. Biochem., 254, 14-24.

Caloca, M.J., Fernandez, N., Lewin, N.E., Ching, D.X., Modali, R., Blumberg, P.M., & Kazanietz, M.G. (1997). Beta 2-chimaerin is a high affinity receptor for the phorbol ester tumour promoters. J. Biol. Chem., 272, 26488-26496.

Caloca, M.J., GarciaBermejo, M.L., Blumberg, P.M., Lewin, N.E., Kremmer, E., Mischak, H., Wang, S.M., Nacro, K., Bienfait, B., Marquez, V.E., & Kazanietz, M.G. (1999). Beta 2-chimaerin is a novel target for diacylglycerol: binding properties and changes in subcellular localisation mediated by ligand binding to its cl domain. Proc. Natl. Acad. Sci. USA, 96, 11854-11859.

Cantor, S.B., Urano, T., & Feig, L.A. (1995). Identification and characterisation of Ral- binding protein 1, a potential downstream target of Ral GTPases. Mol. Cell. Biol., 15, 4578-4584.

Caron, E. & Hall, A. (1998). Identification of two distinct mechanisms of phagocytosis controlled by different Rho GTPases. Science, 282, 1717-1721.

Carpenter, C.L., Duckworth, B.C., Auger, K.R., Cohen, B., Schaffhausen, B.S., & Cantley, L.C. (1990). Purification and characterisation of phosphoinositide 3-kinase from rat-liver. J. Biol. Chem., 265, 19704-19711.

Carpenter, C.L., Auger, K.R., Chanudhuri, M., Yoakim, M., Schaffhausen, B., Shoelson, S., & Cantley, L.C. (1993). Phosphoinositide 3-kinase is activated by phosphopeptides that bind to the SH2 domains of the 85-kDa subunit. J. Biol Chem., 268, 9478-9483.

Cassimeris, L. (1999). Accessory protein regulation of microtubule dynamics throughout the cell cycle. Curr. Opin. Cell Biol., 11, 134-141.

References 216 Castrillon, D.H. & Wasserman, S.A. (1994). Diaphanous is required for cytokinesis in Drosophila and shares domains of similarity with the products of the limb deformity gene. Development, 120, 3367-3377.

Cerione, R.A. & Zheng, Y. (1996). The Dbl family of oncogenes. Curr. Opin. Cell Biol, 8, 216-222.

Chan, A.M.L., McGovern, E.S., Catalano, G., Fleming, T.P., & Miki, T. (1994). Expression cDNA cloning of a novel oncogene with sequence similarity to regulators of small GTP-binding proteins. Oncogene, 9, 1057-1063.

Chan, A.M.L., Takai, S., Yamada, K., & Miki, T. (1996). Isolation of a novel oncogene, NET1, from neuroepithelioma cells byexpression cDNA cloning. Oncogene, 12, 1259- 1266.

Chardin, P., Camonis, J.H., Gale, N.W., Van Aelst, L., Schlessinger, J., Wigler, M.H., & Bar Sagi, D. (1993). Human Sosl - a guanine-nucleotide exchange factor for Ras that binds to Grb2. Science, 260, 1338-1343.

Chen, W., Blanc, J., & Lim, L. (1994). Characterisation of a promiscuous GTPase- activating protein that has a Bcr-related domain from Caenorhabditis Elegans. J. Biol Chem., 269 , 820-823.

Chen, J., Cohn, J.A., & Mandel, L.J. (1995). Dephosphorylation of ezrin as an early event in renal microvillar breakdown and anoxic injury. Proc. Natl Acad. Sci. USA, 92, 7495-7499.

Chen, D., Waters, S.B., Holt, K.H., & Pessin, J.E. (1996a). Sos phosphorylation and disassociation of the Grb2-Sos complex by the ERK and JNK signalling pathways. J. Biol. Chem., Ill, 6328-6332.

Chen, W.N., Chen, S., Yap, S.F., & Lim, L. (1996b). The Caenorhabditis Elegans p21- activated kinase (cePAK) colocalises with ceRACl and Cdc42ce at hypodermal cell boundaries during embryo elongation. J. Biol. Chem., 271, 26362-26368.

Chen, Z.J., Parent, L., & Maniatis, T. (1996c). Site-specific phosphorylation of IkB oc by a novel ubiquitination-dependent protein kinase activity. Cell, 84, 853-862.

Chien, C.B., Rosenthal, D.E., Harris, W.A., & Holt, C.E. (1993). Navigational errors made by growth cones without filopodia in the embryonic Xenopus brain. Neuron, 11, 237-251.

Chong, L.D., Traynor Kaplan, A., Bokoch, G.M., & Schwartz, M.A. (1994). The small GTP-binding protein Rho regulates a phosphatidylinositol 4-phosphate 5-kinase in mammalian cells. Cell, 79, 507-513.

Chou, M.M. & Blenis, J. (1996). The 70-kDa S6 kinase complexes with and is activated by the Rho-family G-proteins Cdc42 and Racl. Cell, 85, 573-583.

Chou, Y.H., Skalli, O., & Goldman, R.D. (1997). Intermediate filaments and cytoplasmic networking: new connections and more functions. Curr. Opin. Cell Biol., 9, 49-53.

References 217 Chrzanowska-Wodnicka, M. & Burridge, K. (1996). Rho-stimulated contractility drives the formation of stress fibres and focal adhesions. J. Cell Biol, 133, 1403-1415.

Chuang, T.H., Xu, X., Knaus, U.G., Hart, M.J., & Bokoch, G.M. (1993). GDP dissociation inhibitor prevents intrinsic and GTPase activating protein-stimulated GTP hydrolysis by the Rac GTP-binding protein. J. Biol Chem., 268, 775-778.

Chuang, T.H., Xu, X., Kaartinen, V., Heisterkamp, N., Groffen, J., & Bokoch, G.M. (1995). Abr and Bcr are multifunctional regulators of the Rho GTP-binding protein family. Proc. Natl. Acad. Sci. USA, 92, 10282-10286.

Cicchetti, P., Mayer, B.J., Thiel, G., & Baltimore, D. (1992). Identification of a protein that binds to the SH3 region of Abl and is similar to Bcr and GAP-Rho. Science, 257, 803-806.

Cicchetti, P., Ridley, A.J., Zheng, Y., Cerione, R.A., & Baltimore, D. (1995). 3BP-1, an SH3 domain binding protein, has GAP activity for Rac and inhibits growth factor- induced membrane ruffling in fibroblasts. EMBO J., 14, 3127-3135.

Cleghon, V. & Morrison, D.K. (1994). Raf-1 interacts with Fyn and Src in a non- phosphotyrosine-dependent manner. J. Biol. Chem., 269, 17749-17755.

Coso, O.A., Chiariello, M., Yu, J.C., Teramoto, H., Crespo, P., Xu, N., Miki, T., & Gutkind, J.S. (1995). The small GTP-binding proteins Racl and Cdc42 regulate the activity of the JNK/SAPK signalling pathway. Cell, 81, 1137-1146.

Cowley, S., Paterson, H., Kemp, P., & Marshall, C.J. (1994). Activation of MAP kinase kinase is necessary and sufficient for PC 12 differentiation and for transformation of NIH 3T3 cells. Cell, 11, 841-852.

Crespo, P., Bustelo, X.R., Aaronson, D.S., Coso, O.A., Lopez-Barahona, M., Barbacid, M., & Gutkind, J.S. (1996). Rac-1 dependent stimulation of the JNK/SAPK signalling pathway by Vav. Oncogene, 13, 455-460.

Crespo, P., Schuebel, K.E., Ostrom, A. A., Gutkind, J.S., & Bustelo, X.R. (1997). Phosphotyrosine-dependent activation of Rac-1 GDP/GTP exchange by the Vav proto­ oncogene product. Nature, 385, 169-172.

Cui, X.L. & Douglas, J.G. (1997). Arachidonic acid activates c-Jun N-terminal kinase through NADPH oxidase in rabbit proximal tubular epithelial cells. Proc. Natl. Acad. Sci. USA, 94, 3771-3776.

Daniels, R.H., Hall, P.S., & Bokoch, G.M. (1998). Membrane targeting of p21-activated kinase 1 (PAK1) induces neurite outgrowth from PC12 cells. EMBO J., 17, 754-764. Davenport, R.W., Dou, P., Rehder, V., & Kater, S.B. (1993). A sensory role for neuronal growth cone filopodia. Nature, 361, 721-724.

De Mendez, I., Homayounpour N., & Leto T.L. (1997) Specificity of p47phox SH3 domain interactions in NADPH oxidase assembly and activation. Mol. Cell. Biol., 17, 2177-2185

References 218 Debant, A., Serra-Pages, C., Seipel, K., O'Brien, S., Tang, M., Park, S.H., & Streuli, M. (1996). The multidomain protein Trio binds the Lar transmembrane tyrosine phosphatase, contains a protein kinase domain, and has separate Rac- specific and Rho- specific guanine nucleotide exchange factor domains. Proc. Natl Acad. Sci. USA, 93, 5466-5471.

Delalle, I., Bhide, P.G., Caviness, V.S., & Tsai, L.H. (1997). Temporal and spatial patterns of expression of p35, a regulatory subunit of cyclin-dependent kinase 5, in the nervous system of the mouse. J. Neurocytol., 26, 283-296.

Delhase, M., Hayakawa, M., Chen, Y., & Karin, M. (1999). Positive and negative regulation of IkB kinase activity through IKK beta subunit phosphorylation. Science, 284, 309-313.

Derijard, B., Hibi, M., Wu, I.H., Barrett, T., Su, B., Deng, T.L., Karin, M., & Davis, R.J. (1994). JNK1 - a protein-kinase stimulated by UV-light and Ha-Ras that binds and phosphorylates the c-Jun activation domain. Cell, 76, 1025-1037.

Devary, Y., Rosette, C., DiDonato, J.A., & Karin, M. (1993). NF-kappa-B activation by ultraviolet-light not dependent on a nuclear signal. Science, 261, 1442-1445.

Diaz, M.N., Frei, B., Vita, J.A., & Keaney, J.F. (1997). Mechanisms of disease - antioxidants and atherosclerotic heart disease. New Engl. J. Med., 337, 408-416.

Dickson, B. & Hafen, E. (1994). Genetics of signal transduction in invertebrates. Curr. Opin. Genet. Dev., 4, 64-70.

DiDonato, J.A., Hayakawa, M., Rothwarf, D.M., Zandi, E., & Karin, M. (1997). A cytokine-responsive IkB kinase that activates the transcription factor NF kB . Nature, 388, 548-554.

Diekmann, D., Brill, S., Garrett, M.D., Totty, N., Hsuan, J., Monfries, C., Hall, C., Lim, L., & Hall, A. (1991). Bcr encodes a GTPase-activating protein for p21Rac. Nature, 351, 400-402.

Diekmann, D., Abo, A., Johnston, C., Segal, A.W., & Hall, A. (1994). Interaction of Rac with p67phox and regulation of phagocytic NADPH oxidase activity. Science, 265, 531 - 533.

Downing, K.H. & Nogales, E. (1998). Tubulin and microtubule structure. Curr. Opin. Cell Biol., 10, 16-22.

Downward, J. (1998). Mechanisms and consequences of activation of protein kinase B/Akt. Curr. Opin. Cell Biol., 10, 262-267.

D'Souza-Schorey, C., Boshans, R.L., McDonough, M., Stahl, P.D., & Van Aelst, L. (1997). A role for PORI, a Rac 1-interacting protein, in ARF6-mediated cytoskeletal rearrangements. EMBO J., 16, 5445-5454.

References 219 Duchesne, M., Schweighoffer, F., Parker, F., Clerc, F., Frobert, Y., Thang, M.N., & Tocque, B. (1993). Identification of the SH3 domain of GAP as an essential sequence for RasGAP mediated signaling. Science, 259, 525-528.

Dudley, D.T., Pang, L., Decker, S.J., Bridges, A.J., & Saltiel, A.R. (1995). A synthetic inhibitor of the mitogen-activated protein-kinase cascade. Proc. Natl Acad. Sci. USA, 92, 7686-7689.

Eck, M.J., Shoelson, S.E., & Harrison, S.C. (1993). Recognition of a high-affinity phosphotyrosyl peptide by the Src homology-2 domain of p56Lck. Nature, 362, 87-91.

Ellis, C., Moran, M., McCormick, F., & Pawson, T. (1990). Phosphorylation of GAP and GAP-associated proteins by transforming and mitogenic tyrosine kinases. Nature, 343, 377-381.

Emala, C.W., Schwindinger, W.F., Wand, G.S., & Levine, M.A. (1994). Signal- transducing G-proteins - basic and clinical implications. Prog. Nucleic Acid Res. Mol. Biol., 47, 81-111.

Faix, J., Clougherty, C., Konzok, A., Mintert, U., Murphy, J., Albrecht, R., Muhlbauer, B., & Kuhlmann, J. (1998). The IQGAP-related protein dGAPl interacts with Rac and is involved in the modulation of the F-actin cytoskeleton and control of cell motility. J. Cell Sci., I l l , 3059-3071.

Fanger, G.R., Gerwins, P., Widmann, C., Jarpe, M B., & Johnson, G.L. (1997a). MEKKs, GCKs, MLKs, PAKs, TAKs, and TPLs: upstream regulators of the c-Jun amino-terminal kinases? Curr. Opin. Genet. Dev, 7, 67-74.

Fanger, G.R., Johnson, N.L., & Johnson, G.L. (1997b). MEK kinases are regulated by EGF and selectively interact with Rac/Cdc42. EMBO J., 16, 4961-4972.

Feng, S B., Chen, J.K., Yu, H.T., Simon, J.A., & Schreiber, S.L. (1994). 2 binding orientations for peptides to the Src SH3 domain -development of a general-model for SH3 -ligand interactions. Science, 266, 1241-1247.

Fernandez-Sarabia, M.J. & Bischoff, J.R. (1993). Bcl-2 associates with the Ras-related protein R-Ras p23. Nature, 366, 274-275.

Finco, T.S. & Baldwin, A.S. (1993). Kappa-B site-dependent induction of gene- expression by diverse inducers of nuclear factor-kappa-B requires Raf-1. J. Biol. Chem., 268, 17676-17679.

Foisner, R. & Wiche, G. (1991). Intermediate filament-associated proteins. Curr. Opin. Cell Biol., 3, 75-81.

Foisner, R., Leichtfried, F.E., Herrmann, H., Small, J.V., Lawson, D., & Wiche, G. (1988). Cytoskeleton-associated plectin - in situ localisation, in vitro reconstitution, and binding to immobilised intermediate filament proteins. J. Cell Biol., 106, 723-733.

References 220 Foisner, R., Traub, P., & Wiche, G. (1991). Protein kinase-A-regulated and protein kinase-C-regulated interaction of plectin with lamin-b and vimentin. Proc. Natl Acad. Sci. USA , 88, 3812-3816.

Franke, T.F., Kaplan, D.R., Cantley, L.C., & Toker, A. (1997). Direct regulation of the Akt proto-oncogene product by phosphatidylinositol-3,4-bisphosphate. Science, 275, 665-668.

Freedman, L.P., Luisi, B.F., Korszun, Z.R., Basavappa, R., Sigler, P.B., & Yamamoto, K.R. (1988). The function and structure of the metal co-ordination sites within the glucocorticoid receptor DNA-binding domain. Nature, 334, 543-546.

Fuchs, E. & Cleveland, D.W. (1998). A structural scaffolding of intermediate filaments in health and disease. Science, 279, 514-519.

Fuchs, E. & Weber, K. (1994). Intermediate filaments - structure, dynamics, function, and disease. Ann. Rev. Biochem., 63, 345-382.

Fukasawa, K., Shen, R.L., Resau, J., Dasilva, P.P., & Vandewoude, G.F. (1995). Overexpression of Mos oncogene product in Swiss 3T3 cells induces apoptosis preferentially during s-phase. Oncogene, 10, 1-8.

Fukata, Y., Kimura, K., Oshiro, N., Saya, H., Matsuura, Y., & Kaibuchi, K. (1998). Association of the myosin-binding subunit of myosin phosphatase and moesin: dual regulation of moesin phosphorylation by Rho-associated kinase and myosin phosphatase. J. Cell Biol., 141, 409-418.

Fukuda, M., Kojima, T., Kabayama, H., & Mikoshiba, K. (1996). Mutation of the pleckstrin homology domain of Bruton's tyrosine kinase in immunodeficiency impaired inositol 1,3,4,5-tetrakisphosphate binding capacity. J. Biol. Chem., 271, 30303-30306.

Fukumoto, Y., Kaibuchi, K., Hori, Y., Fujioka, H., Araki, S., Ueda, T., Kikuchi, A., & Takai, Y. (1990). Molecular-cloning and characterisation of a novel type of regulatory protein (GDI) for the Rho proteins, Ras p21 -like small GTP-binding proteins. Oncogene, 5, 1321-1328.

Gache, Y., Chavanas, S., Lacour, J.P., Wiche, G., Owaribe, K., Meneguzzi, G., & Ortonne, J.P. (1996). Defective expression of plectin/hdl in epidermolysis bullosa simplex with muscular dystrophy. J. Clin. Invest., 97, 2289-2298.

Gaetano, C., Matsuo, T., & Thiele, C.J. (1997). Identification and characterisation of a retinoic acid-regulated human homologue of the unc-33-like phosphoprotein gene (hUlip) from neuroblastoma cells. J. Biol. Chem., 272, 12195-12201.

Galisteo, M.L., Chernoff, J., Su, Y.C., Skolnik, E.Y., & Schlessinger, J. (1996). The adaptor protein Nek links receptor tyrosine kinases with the serine-threonine kinase PAK1. J. Biol. Chem., 271, 20997-21000.

Gallo, G. & Letourneau, P.C. (1998). Axon guidance: GTPases help axons reach their targets. Curr. Biol, 8, R80-R82.

References 221 Gardner, A.M. & Johnson, G.L. (1996). Fibroblast growth factor-2 suppression of tumour necrosis factor alpha-mediated apoptosis requires Ras and the activation of mitogen- activated protein kinase. J. Biol. Chem., 271, 14560-14566.

Garrington, T.P. & Johnson, G.L. (1999). Organisation and regulation of mitogen- activated protein kinase signalling pathways. Curr. Opin. Cell Biol., 11, 211-218.

Gary, R. & Bretscher, A. (1993). Heterotypic and homotypic associations between ezrin and moesin, 2 putative membrane cytoskeletal linking proteins. Proc. Natl. Acad. Sci. USA, 90, 10846-10850.

Gary, R. & Bretscher, A. (1995). Ezrin self-association involves binding of an N-terminal domain to a normally masked C-terminal domain that includes the F-actin binding- site. Mol. Biol. Cell, 6, 1061-1075.

Gebbink, M.F.B.G., Kranenburg, O., Poland, M., van Horck, F.P.G., Houssa, B., & Moolenaar, W.H. (1997). Identification of a novel, putative Rho-specific GDP/GTP exchange factor and a RhoA-binding protein: control of neuronal morphology. J. Cell Biol., 137, 1603-1613.

Gilmore, A.P. & Burridge, K. (1996). Regulation of vinculin binding to talin and actin by phosphatidyl- inositol-4-5-bisphosphate. Nature, 381, 531-535.

Glaven, J.A., Whitehead, I.P., Nomanbhoy, T., Kay, R., & Cerione, R.A. (1996). Lfc and Lsc oncoproteins represent two new guanine nucleotide exchange factors for the Rho GTP-binding protein. J. Biol. Chem., 271, 27374-27381.

Goldman, R.D., Khuon, S., Chou, Y.H., Opal, P., & Steinert, P.M. (1996). The function of intermediate filaments in cell shape and cytoskeletal integrity. J. Cell Biol., 134, 971- 983.

Goshima, Y., Nakamura, F., Strittmatter, P., & Strittmatter, S.M. (1995). Collapsin- induced growth cone collapse mediated by an intracellular protein related to unc-33. Nature, 376, 509-514.

Goto, H., Kosako, H., Tanabe, K., Yanagida, M., Sakurai, M., Amano, M., Kaibuchi, K., & Inagaki, M. (1998). Phosphorylation of vimentin by Rho-associated kinase at a unique amino-terminal site that is specifically phosphorylated during cytokinesis. J. Biol. Chem., 273, 11728-11736.

Green, K.J. & Jones, J.C.R. (1996). Desmosomes and hemidesmosomes: structure and function of molecular components. Faseb J., 10, 871-881.

Green, S., Issemann, I., & Sheer, E. (1988). A versatile in vivo and in vitro eukaryotic expression vector for protein engineering. Nucleic Acids Res., 16, 369

Griendling, K.K., Minieri, C.A., Ollerenshaw, J.D., & Alexander, R.W. (1994). Angiotensin-ii stimulates NADH and NADPH oxidase activity in cultured vascular smooth-muscle cells. Circulation Research, 74, 1141-1148.

References 222 Gustafson, T.A., He, W.M., Craparo, A., Schaub, C.D., & O'Neill, T.J. (1995). Phosphotyrosine-dependent interaction of She and insulin-receptor substrate-1 with the NPEY motif of the insulin-receptor via a novel non-SH2 domain. Mol. Cell B io, l15, 2500-2508.

Guyton, K.Z., Liu, Y.S., Gorospe, M., Xu, Q.B., & Holbrook, N.J. (1996). Activation of mitogen-activated protein kinase by H 2O2 - role in cell survival following oxidant injury. J. Biol Chem., 271, 4138-4142.

Habets, G.G., Scholtes, E.H., Zuydgeest, D., van derKammen, R.A., Stam, J.C., Bems, A., & Collard, J.G. (1994). Identification of an invasion-inducing gene, Tiam-1, that encodes a protein with homology to GDP-GTP exchangers for Rho-like proteins. Cell, 77, 537-549.

Hall, A. (1994). Small GTP-binding proteins and the regulation of the actin cytoskeleton. Ann. Rev. Cell B io, l 10, 31-54.

Hall, C., Monfries, C., Smith, P., Lim, H.H., Kozma, R., Ahmed, S., Vanniasingham, V., Leung, T., & Lim, L. (1990). Novel human brain cDNA encoding a 34,000 Mr protein n-chimaerin, related to both the regulatory domain of protein kinase C and Bcr, the product of the breakpoint cluster region gene. J. Mol Biol., 211, 11-16.

Hall, C., Sin, W.C., Teo, M., Michael, G.J., Smith, P., Dong, J.M., Lim, H.H., Manser, E., Spurr, N.K., Jones, T.A., & Lim, L. (1993). Alpha-2-chimaerin, an SH2-containing GTPase-activating protein for the Ras-related protein p21Rac derived by alternate splicing of the human n-chimaerin gene, is selectively expressed in brain-regions and testes. Mol. Cell. B io, l 13, 4986-4998.

Hamajima, N., Matsuda, K., Sakata, S., Tamaki, N., Sasaki, M., & Nonaka, M. (1996). A novel gene family defined by human dihydropyrimidinase and three related proteins with differential tissue distribution. Gene, 180, 157-163.

Hamm, H.E. & Gilchrist, A. (1996). Heterotrimeric G proteins. Curr. Opin. Cell Biol., 8, 189-196.

Han, J.W., Das, B., Wei, W., Van Aelst, L., Mosteller, R.D., Khosravi-Far, R., Westwick, J.K., Der, C.J., & Broek, D. (1997). Lck regulates Vav activation of members of the Rho family of GTPases. Mol. Cell B io, l 17, 1346-1353.

Harden, N., Lee, J., Loh, H.Y., Ong, Y.M., Tan, I., Leung, T., Manser, E., & Lim, L. (1996). A Drosophila homologue of the Rac- and Cdc42-activated serine/threonine kinase PAK is a potential focal adhesion and focal complex protein that colocalises with dynamic actin structures. M ol Cell. B io, l 16, 1896-1908.

Hart, M.J., Eva, A., Evans, T., Aaronson, S. A., & Cerione, R.A. (1991). Catalysis of guanine nucleotide exchange on the CDC42Hs protein by the Dbl oncogene product. Nature, 354, 311-314.

Hart, M.J., Maru, Y., Leonard, D., Witte, O.N., Evans, T., & Cerione, R.A. (1992). A GDP dissociation inhibitor that serves as a GTPase inhibitor for the Ras-like protein Cdc42Hs. Science, 258, 812-815.

References 223 Hart, M.J., Eva, A., Zangrilli, D., Aaronson, S.A., Evans, T., Cerione, R.A., & Zheng, Y. (1994). Cellular transformation and guanine nucleotide exchange activity are catalysed by a common domain on the Dbl oncogene product. J. Biol. Chem., 269, 62- 65.

Hart, M.J., Callow, M.G., Souza, B., & Polakis, P. (1996a). IQGAP1, a calmodulin- binding protein with a RasGAP-related domain, is a potential effector for Cdc42Hs. EMBOJ., 15, 2997-3005.

Hart, M.J., Sharma, S., elMasry, N., Qiu, R.G., McCabe, P., Polakis, P., & Bollag, G. (1996b). Identification of a novel guanine nucleotide exchange factor for the Rho GTPase. J. Biol Chem. 271, 25452-25458.

Hart, M.J., Jiang, X.J., Kozasa, T., Roscoe, W., Singer, W.D., Gilman, A.G., Stemweis, PC., & Bollag, G. (1998). Direct stimulation of the guanine nucleotide exchange activity of pi 15 RhoGEF by G alpha(13). Science, 280, 2112-2114.

Hartwig, J.H., Bokoch, G.M., Carpenter, C.L., Janmey, P.A., Taylor, L.A., Toker, A., & Stossel, T.P. (1995). Thrombin receptor ligation and activated Rac uncap actin filament barbed ends through phosphoinositide synthesis in permeabilised human platelets. Cell, 82, 643-653.

Hasegawa, M., Morishimakawashima, M., Takio, K., Suzuki, M., Titani, K., & Ihara, Y. (1992). Protein-sequence and mass-spectrometric analyses of Tau in the Alzheimer's- disease brain. J. Biol. Chem., 267, 17047-17054.

Haslam, R.J., Koide, H.B., & Hemmings, B.A. (1993). Pleckstrin domain homology. Nature, 363, 309-310.

Hatada, M.H., Lu, X.D., Laird, E.R., Green, J., Morgenstern, J.P., Lou, M.Z., Marr, C.S., Phillips, T.B., Ram, M.K., Theriault, K., Zoller, M.J., & Karas, J.L. (1995). Molecular-basis for interaction of the protein-tyrosine kinase ZAP-70 with the T-cell receptor. Nature, 377, 32-38.

Hawkins, P.T., Eguinoa, A., Qiu, R.G., Stokoe, D., Cooke, F.T., Walters, R., Wennstrom, S., Claesson Welsh, L., Evans, T., Symons, M., & et al (1995). PDGF stimulates an increase in GTP-Rac via activation of phosphoinositide 3-kinase. Curr. Biol, 5, 393-403.

Heidemann, S.R. & Buxbaum, R.E. (1991). Growth cone motility. Curr. Opin. Neurobiol., 1, 339-345.

Heim, R., Cubitt, A.B., & Tsien, R.Y. (1995). Improved green fluorescence. Nature, 373, 663-664.

Heisterkamp, N., Stam, K., Groffen, J., Deklein, A., & Grosveld, G. (1985). Structural organisation of the Bcr gene and its role in the PH translocation. Nature, 315, 758-761.

Heldin, C.H. (1997). Simultaneous induction of stimulatory and inhibitory signals by PDGF. FEBS Letters, 410, 17-21.

References 224 Helms, J.B. (1995). Role of heterotrimeric GTP-binding proteins in vesicular protein- transport - indications for both classical and alternative G-protein cycles. FEBS Letters, 369, 84-88.

Herskowitz, I. (1995). Map kinase pathways in yeast - for mating and more. Cell, 80, 187-197.

Higley, S. & Way, M. (1997). Actin and cell pathogenesis. Curr. Opin. Cell Biol., 9, 62- 69.

Hildebrand, J.D., Taylor, J.M., & Parsons, J.T. (1996). An SH3 domain-containing GTPase-activating protein for Rho and Cdc42 associates with focal adhesion kinase. Mol. Cell. Biol., 16, 3169-3178.

Hill, C.S., Wynne, J., & Treisman, R. (1995). The Rho family GTPases RhoA, Racl, and CDC42Hs regulate transcriptional activation by SRF. Cell, 81, 1159-1170.

Hirao, M., Sato, N., Kondo, T., Yonemura, S., Monden, M., Sasaki, T., Takai, Y., & Tsukita, S. (1996). Regulation mechanism of ERM (ezrin/radixin/moesin) protein/plasma membrane association: possible involvement of phosphatidylinositol turnover and Rho- dependent signalling pathway. J. Cell Biol., 135, 37-51.

Hirokawa, N., Noda, Y., & Okada, Y. (1998). Kinesin and dynein superfamily proteins in organelle transport and cell division. Curr. Opin. Cell Biol., 10, 60-73.

Hirose, M., Ishizaki, T., Watanabe, N., Uehata, M., Kranenburg, O., Moolenaar, W.H., Matsumura, F., Maekawa, M., Bito, H., & Narumiya, S. (1998). Molecular dissection of the Rho-associated protein kinase (pl60ROCK)- regulated neurite remodelling in neuroblastoma N1E-115 cells. J. Cell Biol, 141, 1625-1636.

Homma, Y. & Emori, Y. (1995). A dual functional signal mediator showing RhoGAP and phospholipase C-delta stimulating activities. EMBO J., 14, 286-291.

Horii, Y., Beeler, J.F., Sakaguchi, K., Tachibana, M., & Miki, T. (1994). A novel oncogene, Ost, encodes a guanine nucleotide exchange factor that potentially links Rho and Rac signalling pathways. EMBO J., 13, 4776-4786.

Houle, M.G. & Bourgoin, S. (1999). Regulation of phospholipase d by phosphorylation- dependent mechanisms. Biochimica Et Biophysica Acta-Molecular And Cell Biol. Of Lipids, 1439, 135-150.

Houseweart, M.K. & Cleveland, D.W. (1998). Intermediate filaments and their associated proteins: multiple dynamic personalities. Curr. Opin. Cell Biol., 10, 93-101.

Hu, K.Q. & Settleman, J. (1997). Tandem SH2 binding sites mediate the RasGAP- RhoGAP interaction: a conformational mechanism for SH3 domain regulation. EMBO J., 16, 473-483.

Igarashi, K., Isohara, T., Kato, T., Shigeta, K., Yamano, T., & Uno, I. (1998). Tyrosine 1213 of fit-1 is a major binding site of Nek and SHP-2. Biochem. Biophys. Res. Commun., 246, 95-99.

References 225 Inagaki, M., Matsuoka, Y., Tsujimura, K., Ando, S., Tokui, T., Takahashi, T., & Inagaki, N. (1996). Dynamic property of intermediate filaments: regulation by phosphorylation. Bioessays, 18, 481-487.

Irani, K., Xia, Y., Zweier, J.L., Sollott, S.J., Der, C.J., Fearon, E.R., Sundaresan, M., Finkel, T., & Goldschmidt-Clermont, P.J. (1997). Mitogenic signalling mediated by oxidants in Ras-transformed fibroblasts. Science, 275, 1649-1652.

Ishizaki, T., Maekawa, M., Fujisawa, K., Okawa, K., Iwamatsu, A., Fujita, A., Watanabe, N., Saito, Y., Kakizuka, A., Morii, N., & Narumiya, S. (1996). The small GTP-binding protein Rho binds to and activates a 160-kDa serine/threonine protein- kinase homologous to myotonic-dystrophy kinase. EMBO J., 15, 1885-1893.

Ishizaki, T., Naito, M., Fujisawa, K., Maekawa, M., Watanabe, N., Saito, Y., & Narumiya, S. (1997). P160(ROCK), a Rho-associated coiled-coil forming protein kinase, works downstream of Rho and induces focal adhesions. FEBS Letters, 404, 118-124.

Jalink, K. & Moolenaar, W.H. (1992). Thrombin receptor activation causes rapid neural cell rounding and neurite retraction independent of classic 2nd messengers. J. Cell Biol., 118, 411-419.

Jalink, K., Vancorven, E.J., Hengeveld, T., Morii, N., Narumiya, S., & Moolenaar, W.H. (1994). Inhibition of lysophosphatidate-induced and thrombin-induced neurite retraction and neuronal cell rounding by ADP-ribosylation of the small GTP-binding protein-Rho. J. Cell Biol, 126, 801-810.

Jefferson, A.B., Klippel, A., & Williams, L.T. (1998). Inhibition of mSos-activity by binding of phosphatidylinositol 4, 5-P2 to the mSos pleckstrin homology domain. Oncogene, 16, 2303-2310.

Jiang, W.G., Hiscox, S., Singhrao, S.K., Puntis, M.C.A., Nakamura, T., Mansel, R.E., & Hallett, M.B. (1995). Induction of tyrosine phosphorylation and translocation of ezrin by hepatocyte growth factor scatter factor. Biochem. Biophys. Res. Commun., 217, 1062- 1069.

Jin, Z. & Strittmatter, S.M. (1997). Racl mediates collapsin-1-induced growth cone collapse. J. Neurosci., 17, 6256-6263.

Johnson, T.M., Yu, Z.X., Ferrans, V.J., Lowenstein, R.A., & Finkel, T. (1996). Reactive oxygen species are downstream mediators of p53-dependent apoptosis. Proc. Natl. Acad. Sci. USA, 93, 11848-11852.

Jones, S.A., Wood, J.D., Coffey, M.J., & Jones, O.T.G. (1994). The functional expression of p47-phox and p67-phox may contribute to the generation of superoxide by an NADPH oxidase-like system in human fibroblasts. FEBS Letters, 355, 178-182.

Joneson, T. & Bar Sagi, D. (1998). A Racl effector site controlling mitogenesis through superoxide production. J. Biol. Chem., 273, 17991-17994.

References 226 Joung, I., Strominger, J.L., & Shin, J. (1996). Molecular cloning of a phosphotyrosine- independent ligand of the p56(Lck) SH2 domain. Proc. Natl. Acad. Sci. USA, 93, 5991- 5995.

Jullien Flores, V., Dorseuil, O., Romero, F., Letourneur, F., Saragosti, S., Berger, R., Tavitian, A., Gacon, G., & Camonis, J.H. (1995). Bridging Ral GTPase to Rho pathways. RLIP76, a Ral effector with CDC42/Rac GTPase-activating protein activity. J. Biol Chem., 270, 22473-22477.

Kamata, H., Tanaka, C., Yagisawa, H., Matsuda, S., Gotoh, Y., Nishida, E., & Hirata, H. (1996). Suppression of nerve growth factor-induced neuronal differentiation of PC 12 cells - N-acetylcysteine uncouples the signal transduction from Ras to the mitogen- activated protein kinase cascade. J. Biol. Chem., 271, 33018-33025.

Kamata, T., Subleski, M., Hara, Y., Yuhki, N., Kung, H.F., Copeland, N.G., Jenkins, N.A., Yoshimura, T., Modi, W., & Copeland, T.D. (1998). Isolation and characterisation of a bovine neural specific protein (CRMP-2) cDNA homologous to unc-33, a C-Elegans gene implicated in axonal outgrowth and guidance. Mol. Brain Res., 54, 219-236.

Kapeller, R., Prasad, K.V.S., Janssen, O., Hou, W., Schaffhausen, B.S., Rudd, C.E., & Cantley, L.C. (1994). Identification of two SH3-binding motifs in the regulatory subunit of phosphatidylinositol 3-kinase. J. Biol. Chem., 269, 1927-1933.

Karin, M. (1999). The beginning of the end: Ik B kinase (IKK) and NFk B activation. J. Biol Chem., 274, 27339-27342.

Kater, S.B. & Rehder, V. (1995). The sensory-motor role of growth cone filopodia. Curr. Opin. Neurohiol., 5, 68-74.

Katoh, H., Aoki, J., Ichikawa, A., & Negishi, M. (1998). P160 RhoA-binding kinase ROKa induces neurite retraction. J. Biol. Chem., 273, 2489-2492.

Kavanaugh, W.M. & Williams, L.T. (1994). An alternative to SH2 domains for binding tyrosine-phosphorylated proteins. Science, 266, 1862-1865.

Kavanaugh, W.M., Turck, C.W., & Williams, L.T. (1995). PTB domain binding to signalling proteins through a sequence motif containing phosphotyrosine. Science, 268, 1177-1179.

Kawase, T., Orikasa, M., Ogata, S., & Burns, D.M. (1995). Protein-tyrosine phosphorylation-induced by epidermal growth-factor and insulin-like growth-factor-1 in a rat clonal dental pulp-cell line. Arch. Oral Biol., 40, 921-929.

Kellogg, D.R., Moritz, M., & Alberts, B.M. (1994). The centrosome and cellular- organisation. Ann. Rev. Biochem., 63, 639-674.

Keynes, R. & Cook, G.M.W. (1995). Axon guidance molecules. Cell, 83, 161-169.

Khosravi Far, R., Solski, P.A., Clark, G.J., Kinch, M.S., & Der, C.J. (1995). Activation of Racl, RhoA, and mitogen-activated protein kinases is required for Ras transformation. M ol Cell. Biol, 15, 6443-6453.

References 227 Kimura, K., Hattori, S., Kabuyama, Y., Shizawa, Y., Takayanagi, J., Nakamura, S., Toki, S., Matsuda, Y., Onodera, K., & Fukui, Y. (1994). Neurite outgrowth of PC 12 cells is suppressed by wortmannin, a specific inhibitor of phosphatidylinositol 3-kinase. J. Biol. Chem., 269, 18961-18967.

Kimura, K., Ito, M., Amano, M., Chihara, K., Fukata, Y., Nakafuku, M., Yamamori, B., Feng, J.H., Nakano, T., Okawa, K., Iwamatsu, A., & Kaibuchi, K. (1996). Regulation of myosin phosphatase by Rho and Rho-associated kinase (Rho-kinase). Science, 273, 245- 248.

Kimura, K , Fukata, Y., Matsuoka, Y., Bennett, V., Matsuura, Y., Okawa, K., Iwamatsu, A., & Kaibuchi, K. (1998). Regulation of the association of adducin with actin filaments by Rho- associated kinase (Rho-kinase) and myosin phosphatase. J. Biol. Chem., 273, 5542-5548.

Kita, Y., Kimura, K.D., Kobayashi, M., Ihara, S., Kaibuchi, K., Kuroda, S., Ui, M., Iba, H., Konishi, H., Kikkawa, U., Nagata, S., & Fukui, Y. (1998). Microinjection of activated phosphatidylinositol-3 kinase induces process outgrowth in rat PC 12 cells through the Rac-JNK signal transduction pathway. J. Cell Sci., Ill, 907-915.

Klippel, A., Kavanaugh, W.M., Pot, D., & Williams, L.T. (1997). A specific product of phosphatidylinositol 3-kinase directly activates the protein kinase Akt through its pleckstrin homology domain. Mol. Cell. Biol., 17, 338-344.

Knaus, U.G., Heyworth, P.G., Evans, T., Curnutte, J.T., & Bokoch, G.M. (1991). Regulation of phagocyte oxygen radical production by the GTP-binding protein Rac 2. Science, 254, 1512-1515.

Kobayashi, M., Nagata, S., Kita, Y., Nakatsu, N., Ihara, S., Kaibuchi, K., Kuroda, S., Ui, M., Iba, H., Konishi, H., Kikkawa, U., Saitoh, I., & Fukui, Y. (1997). Expression of a constitutively active phosphatidylinositol 3-kinase induces process formation in rat PC 12 cells - use of CRE/LOXP recombination system. J. Biol. Chem., 272, 16089- 16092.

Kobayashi, K., Kuroda, S., Fukata, M., Nakamura, T., Nagase, T., Nomura, N., Matsuura, Y., YoshidaKubomura, N., Iwamatsu, A., & Kaibuchi, K. (1998). P140sra-1 (specifically Racl-associated protein) is a novel specific target for Racl small GTPase. J. Biol. Chem., 273, 291-295.

Koong, A.C., Chen, E.Y., Mivechi, N.F., Denko, N.C., Stannbrook, P., & Giaccia, A.J. (1994). Hypoxic activation of nuclear factor-kappa-B is mediated by a Ras and Raf signalling pathway and does not involve MAP kinase (ERK1 or ERK2). Cancer Research, 54, 5273-5279.

Kotani, H., Takaishi, K., Sasaki, T., & Takai, Y. (1997). Rho regulates association of both the ERM family and vinculin with the plasma membrane in MDCK cells. Oncogene, 14, 1705-1713.

Kozasa, T., Jiang, X.J., Hart, M.J., Stemweis, P.M., Singer, W.D., Gilman, A.G., Bollag, G., & Sternweis, P.C. (1998). PI 15 RhoGEF, a GTPase activating protein for G alpha(12) and G alpha(13). Science, 280, 2109-2111.

References 228 Kozma, R., Ahmed, S., Best, A., & Lim, L. (1995). The Ras-related protein Cdc42Hs and Bradykinin promote formation of peripheral actin microspikes and filopodia in Swiss 3T3 fibroblasts. Mol. Cell. Biol., 15, 1942-1952.

Kozma, R., Ahmed, S., Best, A., & Lim, L. (1996). The GTPase-activating protein n- chimaerin co-operates with Racl and Cdc42Hs to induce the formation of lamellipodia and filopodia. Mol. Cell. Biol., 16, 5069-5080.

Kozma, R., Sarner, S., Ahmed, S., & Lim, L. (1997). Rho family GTPases and neuronal growth cone remodelling: relationship between increased complexity induced by Cdc42Hs, Racl, and acetylcholine and collapse induced by Rho A and lysophosphatidic acid. Mol. Cell. Biol., 17, 1201-1211.

Krapivinsky, G., Krapivinsky, L., Wickman, K., & Clapham, D.E. (1995). G-beta- gamma binds directly to the G-protein-gated K+ channel, i-KAch- J- Biol. Chem., 270, 29059-29062.

Kumagai, N., Morii, N., Fujisawa, K., Nemoto, Y., & Narumiya, S. (1993). ADP- ribosylation of Rho p21 inhibits lysophosphatidic acid-induced protein tyrosine phosphorylation and phosphatidylinositol 3-kinase activation in cultured Swiss 3T3 cells. J. Biol. Chem., 268, 24535-24538.

Kuroda, S., Fukata, M., Kobayashi, K., Nakafiiku, M., Nomura, N., Iwamatsu, A., & Kaibuchi, K. (1996). Identification of IQGAP as a putative target for the small GTPases, Cdc42 and Racl. J. Biol. Chem.,Ill, 23363-23367.

Kurosu, H., Maehama, T., Okada, T., Yamamoto, T., Hoshino, S., Fukui, Y., Ui, M., Hazeki, O., & Katada, T. (1997). Heterodimeric phosphoinositide 3-kinase consisting of p85 and pi 10 beta is synergistically activated by the beta gamma subunits of g proteins and phosphotyrosyl peptide. J. Biol. Chem., 272, 24252-24256.

Kyriakis, J.M., Banerjee, P., Nikolakaki, E., Dai, T.A., Rubie, E.A., Ahmad, M.F., Avruch, J., & Woodgett, J.R. (1994). The stress-activated protein-kinase subfamily of c- Jun kinases. Nature, 369, 156-160.

Lai, C.C., Boguski, M., Broek, D., & Powers, S. (1993). Influence of guanine- nucleotides on complex-formation between Ras and cdc25 proteins. Mol. Cell. Biol., 13, 1345-1352.

Lamarche, N. & Hall, A. (1994). GAPs for Rho-related GTPases. Trends Genet., 10, 436-440.

Lamarche, N., Tapon, N., Stowers, L., Burbelo, P.D., Aspenstrom, P., Bridges, T., Chant, J., & Hall, A. (1996). Rac and Cdc42 induce actin polymerisation and G1 cell cycle progression independently of p65(PAK) and the JNK/SAPK MAP kinase cascade. Cell, 87, 519-529.

Lamarche-Vane, N. & Hall, A. (1998). CdGAP, a novel proline-rich GTPase-activating protein for Cdc42 and Rac. J. Biol. Chem., 273, 29172-29177.

References 229 Lamoureux, P., Altun Gultekin, Z.F., Lin, C.J., Wagner, J.A., & Heidemann, S.R. (1997). Rac is required for growth cone function but not neurite assembly. J. Cell Sci., 110, 635-641.

Lancaster, C.A., Taylor-Harris, P.M., Self, A.J., Brill, S., Vanerp, H E., & Hall, A. (1994). Characterisation ofRhoGAP - a GTPase-activating protein for Rho-related small GTPases. J. Biol Chem., 269, 1137-1142.

Lappalainen, P. & Drubin, D.G. (1997). Cofilin promotes rapid actin filament turnover in vivo. Nature, 388, 78-82.

Leblanc, V., Tocque, B., & Delumeau, I. (1998). Ras-GAP controls Rho-mediated cytoskeletal reorganisation through its SH3 domain. Mol Cell Biol, 18, 5567-5578.

Lee, CH., Kominos, D., Jacques, S., Margolis, B., Schlessinger, J., Shoelson, S.E., & Kuriyan, J. (1994). Crystal-structures of peptide complexes of the amino-terminal SH2 domain of the Syp tyrosine phosphatase. Structure, 2, 423-438.

Lee, F.S., Hagler, J., Chen, Z.J.J., & Maniatis, T. (1997). Activation of the IxBa kinase complex by MEKK1, a kinase of the JNK pathway. Cell, 88, 213-222.

Lee, S.B. & Rhee, S.G. (1995). Significance of PEP2 hydrolysis and regulation of phospholipase-c isozymes. Curr. Opin. Cell Biol., 7, 183-189.

Leevers, S.J., Vanhaesebroeck, B., & Waterfield, M.D. (1999). Signalling through phosphoinositide 3-kinases: the lipids take centre stage. Curr. Opin. Cell Biol., 11, 219- 225.

Lelias, J.M., Adra, C.N., Wulf, G.M., Guillemot, J.C., Khagad, M., Caput, D., & Lim, B. (1993). cDNA cloning of a human mRNA preferentially expressed in haematopoietic cells and with homology to a GDP-dissociation inhibitor for the Rho GTP-binding proteins. Proc. Natl. Acad. Sci. USA, 90, 1479-1483.

Lemmon, M.A., Ferguson, K.M., Obrien, R., Sigler, P.B., & Schlessinger, J. (1995). Specific and high-affinity binding of inositol phosphates to an isolated pleckstrin homology domain. Proc. Natl. Acad. Sci. USA, 92, 10472-10476.

Lemmon, M.A., Ferguson, K.M., & Schlessinger, J. (1996). PH domains: diverse sequences with a common fold recruit signalling molecules to the cell surface. Cell, 85, 621-624.

Lemmon, M.A., Falasca, M., Ferguson, K.M., & Schlessinger, J. (1997). Regulatory recruitment of signalling molecules to the cell membrane by pleckstrin-homology domains. Trends Cell Biol., 1, 237-242.

Leonard, D., Hart, M.J., Platko, J.V., Eva, A., Henzel, W., Evans, T., & Cerione, R.A. (1992). The identification and characterisation of a GDP-dissociation inhibitor (GDI) for the CDC42Hs protein. J. Biol. Chem., 267, 22860-22868.

References 230 Leung, T., How, B.E., Manser, E., & Lim, L. (1993). Germ-cell beta-chimaerin, a new GTPase-activating protein for p21Rac, is specifically expressed during the acrosomal assembly stage in rat testis. J. Biol. Chem., 268, 3813-3816.

Leung, T., How, B.E., Manser, E., & Lim, L. (1994). Cerebellar beta-2-chimaerin, a GTPase-activating protein for p21 Ras- related Rac is specifically expressed in granule cells and has a unique N-terminal SH2 domain. J. Biol Chem., 269, 12888-12892.

Leung, T., Manser, E., Tan, L., & Lim, L. (1995). A novel serine/threonine kinase binding the Ras-related RhoA GTPase which translocates the kinase to peripheral membranes. J. Biol. Chem., 270, 29051-29054.

Leung, T., Chen, X.Q., Manser, E., & Lim, L. (1996). The pl60 RhoA-binding kinase ROKa is a member of a kinase family and is involved in the reorganisation of the cytoskeleton. Mol. Cell. Biol, 16, 5313-5327.

Leung, T., Chen, X.Q., Tan, I., Manser, E., & Lim, L. (1998). Myotonic dystrophy kinase-related Cdc42-binding kinase acts as a Cdc42 effector in promoting cytoskeletal reorganisation. Mol. Cell. Biol, 18, 130-140.

Lew, J. & Wang, J.H. (1995). Neuronal cdc2-like kinase. TrendsBiochem. Sci., 20, 33- 37.

Lew, J., Huang, Q.Q., Qi, Z., Winkfein, R.J., Aebersold, R., Hunt, T., & Wang, J.H. (1994). A brain-specific activator of cyclin-dependent kinase-5. Nature, 371, 423-426.

Li, W., Herman, R.K., & Shaw, J.E. (1992). Analysis of the Caenorhabditis-Elegans axonal guidance and outgrowth gene unc-33. Genetics, 132, 675-689.

Li, S.C., Zhou, S.Y., Vincent, S.J.F., Zwahlen, C., Wiley, S., Cantley, L., Kay, L.E., FormanKay, J., & Pawson, T. (1997). High-affinity binding of the Drosophila numb phosphotyrosine-binding domain to peptides containing a gly-pro-(p)tyr motif. Proc. Natl. Acad. Sci. USA, 94, 7204-7209.

Lim, H.H., Michael, G.J., Smith, P., Lim, L., & Hall, C. (1992). Developmental regulation and neuronal expression of the messenger-RNA of rat n-chimaerin, a p21Rac GAP/cDNA sequence. Biochem. J., 287, 415-422.

Lim, W.A., Richards, F.M., & Fox, R.O. (1994). Structural determinants of peptide- binding orientation and of sequence specificity in SH3 domains. Nature, 372, 375-379.

Lo, Y.Y.C. & Cruz, T.F. (1995). Involvement of reactive oxygen species in cytokine and growth-factor induction of c-fos expression in chondrocytes. J. Biol. Chem., 270, 11727-11730.

Lo, Y.Y.C., Wong, J.M.S., & Cruz, T.F. (1996). Reactive oxygen species mediate cytokine activation of c-Jun INTL-terminal kinases. J. Biol. Chem., 271, 15703-15707.

Luduena, R.F. (1998). Multiple forms of tubulin: different gene products and covalent modifications. International Review O f Cytology-A Survey Of Cell Biology, 178, 207- 275.

References 231 Luo, Y.L., Raible, D., & Raper, J.A. (1993). Collapsin - a protein in brain that induces the collapse and paralysis of neuronal growth cones. Cell, 75, 217-227.

Luo, L., Liao, Y.J., Jan, L.Y., & Jan, Y.N. (1994). Distinct morphogenetic functions of similar small GTPases: Drosophila Dracl is involved in axonal outgrowth and myoblast fusion. Genes Dev., 8, 1787-1802.

Luo, L., Hensch, T.K., Ackerman, L., Barbel, S., Jan, L.Y., & Jan, Y.N. (1996). Differential effects of the Rac GTPase on Purkinje cell axons and dendritic trunks and spines. Nature, 379, 837-840.

Luo, L.Q., Jan, L. Y., & Jan, Y.N. (1997). Rho family small GTP-binding proteins in growth cone signalling. Curr. Opin. Neurobiol., 7, 81-86.

Lynch, E.D., Lee, M.K., Morrow, J.E., Welcsh, P.L., Leon, P.E., & King, M.C. (1997). Nonsyndromic deafness DFNA1 associated with mutation of a human homologue of the Drosophila gene diaphanous. Science, 278, 1315-1318.

Maccioni, R.B. & Cambiazo, V. (1995). Role of microtubule-associated proteins in the control of microtubule assembly. Physiological Reviews, 75, 835-864.

Machesky, L.M. & Gould, K.L. (1999). The Arp2/3 complex: a multifunctional actin organiser. Curr. Opin. Cell Biol., 11, 117-121.

Machesky, L.M., Atkinson, S.J., Ampe, C., Vandekerckhove, J., & Pollard, T.D. (1994). Purification of a cortical complex containing two unconventional from Acanthamoeba by affinity-chromatography on profilin-agarose. J. Cell Biol., 127, 107- 115.

Maciver, S.K. (1998). How ADF/cofilin depolymerises actin filaments - commentary. Curr. Opin. Cell Biol., 10, 140-144.

Mackay, D.J.G., Esch, F., Furthmayr, H., & Hall, A. (1997). Rho- and Rac-dependent assembly of focal adhesion complexes and actin filaments in permeabilised fibroblasts: an essential role for ezrin/radixin/moesin proteins. J. Cell Biol., 138, 927-938.

Madaule, P., Eda, M., Watanabe, N., Fujisawa, K., Matsuoka, T., Bito, H., Ishizaki, T., & Narumiya, S. (1998). Role of Citron kinase as a target of the small GTPase Rho in cytokinesis. Nature, 394, 491-494.

Maekawa, M., Ishizaki, T., Boku, S., Watanabe, N., Fujita, A., Iwamatsu, A., Obinata, T., Ohashi, K., Mizuno, K., & Narumiya, S. (1999). Signalling from Rho to the actin cytoskeleton through protein kinases ROCK and LIM-kinase. Science, 285, 895-898.

Magendantz, M., Henry, M.D., Lander, A., & Solomon, F. (1995). Interdomain interactions of radixin in-vitro. J. Biol. Chem., 270, 25324-25327.

Malinin, N.L., Boldin, M.P., Kovalenko, A.V., & Wallach, D. (1997). MAP3K-related kinase involved in NFkB induction by TNF, cd95 and IL-1. Nature, 385, 540-544.

References 232 Mandelkow, E. & Mandelkow, E.M. (1995). Microtubules and microtubule-associated proteins. Curr. Opin. Cell Biol, 72-81. 1,

Mandelkow, E., Song, Y.H., & Mandelkow, E.M. (1995). The microtubule lattice - dynamic instability of concepts. Trends Cell B io, l5, 262-266.

Manser, E., Leung, T., Monfries, C., Teo, M., Hall, C., & Lim, L. (1992). Diversity and versatility of GTPase activating proteins for the p21Rho subfamily of Ras G proteins detected by a novel overlay assay. J. Biol. Chem., 267, 16025-16028.

Manser, E., Leung, T., Salihuddin, H., Tan, L., & Lim, L. (1993). A non receptor tyrosine kinase that inhibits the GTPase activity of p21(Cdc42). Nature, 363, 364-367.

Manser, E., Leung, T., Salihuddin, H., Zhao, Z.S., & Lim, L. (1994). A brain serine threonine protein-kinase activated by Cdc42 and Racl. Nature, 367, 40-46.

Manser, E., Chong, C., Zhao, Z.S., Leung, T., Michael, G., Hall, C., & Lim, L. (1995). Molecular cloning of a new member of the p21-Cdc42/Rac-activated kinase (PAK) family. J. Biol. Chem., 270, 25070-25078.

Manser, E., Huang, H.Y., Loo, T.H., Chen, X.Q., Dong, J.M., Leung, T., & Lim, L. (1997). Expression of constitutively active alpha-PAK reveals effects of the kinase on actin and focal complexes. Mol. Cell. Biol., 17, 1129-1143.

Manser, E., Loo, T.H., Koh, C.G., Zhao, Z.S., Chen, X.Q., Tan, L., Tan, I., Leung, T., & Lim, L. (1998). PAK kinases are directly coupled to the PIX family of nucleotide exchange factors. Molecular Cell, 1, 183-192.

Mansour, S.J., Matten, W.T., Hermann, A.S., Candia, J.M., Rong, S., Fukasawa, K., Vandewoude, G.F., & Ahn, N.G. (1994). Transformation of mammalian-cells by constitutively active MAP kinase kinase. Science, 265, 966-970.

Marais, R., Light, Y., Paterson, H.F., & Marshall, C.J. (1995). Ras recruits Raf-1 to the plasma-membrane for activation by tyrosine phosphorylation. EMBO J., 14, 3136-3145.

Marshall, C.J. (1996). Ras effectors. Curr. Opin. Cell Biol., 8, 197-204.

Marte, B.M. & Downward, J. (1997). PKB/Akt: connecting phosphoinositide 3-kinase to cell survival and beyond. Trends Biochem. Sci., 22, 355-358.

Martin, G.A., Viskochil, D., Bollag, G., McCabe, P.C., Crosier, W.J., Haubruck, H., Conroy, L., Clark, R., O'Connell, P., Cawthon, R.M., Innis, M.A., & McCormick, F. (1990). The GAP-related domain of the neurofibromatosis type-1 gene-product interacts with Ras p21. Cell, 63, 843-849.

Martin, G.A., Yatani, A., Clark, R., Conroy, L., Polakis, P., Brown, A.M., & McCormick, F. (1992). GAP domains responsible for Ras p21-dependent inhibition of muscarinic atrial k+ channel currents. Science, 255, 192-194.

Matsudaira, P. (1994). Actin crosslinking proteins at the leading edge. Semin. Cell Biol., 5, 165-174.

References 233 Matsui, T., Amano, M., Yamamoto, T., Chihara, K., Nakafuku, M., Ito, M., Nakano, T., Okawa, K., Iwamatsu, A., & Kaibuchi, K. (1996). Rho-associated kinase, a novel serine threonine kinase, as a putative target for the small GTP binding protein Rho. EMBO J., 15, 2208-2216.

Matsui, T., Maeda, M., Doi, Y., Yonemura, S., Amano, M., Kaibuchi, K., & Tsukita, S. (1998). Rho-kinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association. J. Cell Biol., 140, 647-657.

McCallum, S.J., Wu, W.J., & Cerione, R.A. (1996). Identification of a putative effector for Cdc42Hs with high sequence similarity to the RasGAP-related protein IQGAP1 and a Cdc42Hs binding partner with similarity to IQGAP2. J. Biol. Chem., 271, 21732-21737.

McCallum, S.J., Erickson, J.W., & Cerione, R.A. (1998). Characterisation of the association of the actin-binding protein, IQGAP, and activated Cdc42 with golgi membranes. J. Biol. Chem., 273, 22537-22544.

McGlade, J., Brunkhorst, B., Anderson, D., Mbamalu, G., Settleman, J., Dedhar, S., Rozakis-Adcock, M., Chen, L.B., & Pawson, T. (1993). The N-terminal region of GAP regulates cytoskeletal structure and cell-adhesion. EMBO J., 12, 3073-3081.

McLean, W.H.I., Pulkkinen, L., Smith, F.J.D., Rugg, E.L., Lane, E.B., Bullrich, F., Burgeson, R.E., Amano, S., Hudson, D.L., Owaribe, K., McGrath, J.A., McMillan, J R., Eady, R.A.J., Leigh, I.M., Christiano, A.M., & Uitto, J. (1996). Loss of plectin causes epidermolysis bullosa with muscular dystrophy: cDNA cloning and genomic organisation. Genes Dev., 10, 1724-1735.

Medema, R.H., Delaat, W.L., Martin, G.A., McCormick, F., & Bos, J.L. (1992). GTPase-activating protein SH2-SH3 domains induce gene-expression in a Ras- dependent fashion. Mol. Cell. Biol., 12, 3425-3430.

Meier, B., Radeke, H.H., Selle, S., Younes, M., Sies, H., Resch, K., & Habermehl, G.G. (1989). Human-fibroblasts release reactive oxygen species in response to interleukin-1 or tumour necrosis factor-alpha. Biochem. J., 263, 539-545.

Michiels, F., Habets, G.G., Stam, J.C., van derKammen, R.A., & Collard, J.G. (1995). A role for Rac in Tiaml-induced membrane ruffling and invasion. Nature, 375, 338-340.

Michiels, F., Stam, J.C., Hordijk, P.L., van der Kammen, R.A., Ruuls Van Stalle, L., Feltkamp, C.A., & Collard, J.G. (1997). Regulated membrane localisation of Tiaml, mediated by the NH2-terminal pleckstrin homology domain, is required for Rac- dependent membrane ruffling and c-Jun NH2-terminal kinase activation. J. Cell Biol., 137, 387-398.

Miki, T., Smith, C.L., Long, J.E., Eva, A., & Fleming, T.P. (1993). Oncogene ect2 is related to regulators of small GTP-binding proteins. Nature, 362, 462-465.

Minden, A., Lin, A., Claret, F.X., Abo, A., & Karin, M. (1995). Selective activation of the JNK signalling cascade and c-Jun transcriptional activity by the small GTPases Rac and Cdc42Hs. Cell, 81, 1147-1157.

References 234 Miki, H., Sasaki, T., Takai, Y., & Takenawa, T. (1998). Induction offilopodium formation by a WASP-related actin-depolymerising protein n-WASP. Nature, 391, 93- 96.

Minturn, J.E., Geschwind, D.H., Fryer, H.J.L., & Hockfield, S. (1995a). Early postmitotic neurons transiently express TOAD64, a neural-specific protein. J. Comp. Neurol., 355, 369-379.

Minturn, J.E., Fryer, H.J.L., Geschwind, D.H., & Hockfield, S. (1995b). TOAD64, a gene expressed early in neuronal differentiation in the rat, is related to unc-33, a C- Elegans gene involved in axon outgrowth. J. Neurosci., 15, 6757-6766.

Montaner, S., Perona, R., Saniger, L., & Lacal, J.C. (1998). Multiple signalling pathways lead to the activation of NFkB by the Rho family of GTPases. J. Biol. Chem., 273, 12779-12785.

Moon, I.S., Apperson, M L., & Kennedy, M.B. (1994). The major tyrosine- phosphorylated protein in the postsynaptic density fraction is n-methyl-d-aspartate receptor subunit 2b. Proc. Natl. Acad. Sci. USA, 91, 3954-3958.

Moran, M.F., Polakis, P., Mccormick, F., Pawson, T., & Ellis, C. (1991). Protein- tyrosine kinases regulate the phosphorylation, protein interactions, subcellular- distribution, and activity of p21Ras GTPase-activating protein. Mol. Cell. Biol., 11, 1804-1812.

Muller, B.K., Bonhoeffer, F., & Drescher, U. (1996). Novel gene families involved in neural pathfmding. Curr. Opin. Genet. Dev, 6, 469-474.

Muller, R.T., Honnert, U., Reinhard, J., & Bahler, M. (1997). The rat myosin myr5 is a GTPase-activating protein for Rho in vivo: essential role of arginine 1695. Mol. Biol. Cell, 8, 2039-2053.

Mullins, R.D., Heuser, J. A., & Pollard, T.D. (1998). The interaction of Arp2/3 complex with actin: nucleation, high affinity pointed end capping, and formation of branching networks of filaments. Proc. Natl. Acad. Sci. USA, 95, 6181-6186.

Musacchio, A., Gibson, T., Rice, P., Thompson, J., & Saraste, M. (1993). The PH domain - a common piece in the structural patchwork of signalling proteins. Trends Biochem. Sci., 18, 343-348.

Nagata, K., Puls, A., Futter, G., Aspenstrom, P., Schaefer, E., Nakata, T., Hirokawa, N., & Hall, A. (1998). The MAP kinase kinase kinase MLK2 co-localises with activated JNK along microtubules and associates with kinesin superfamily motor KIF3. EMBO J., 17, 149-158.

Nantel, A., Mohammad Ali, K., Sherk, J., Posner, B.I., & Thomas, D.Y. (1998). Interaction of the GrblO adapter protein with the Rafl and MEK1 kinases. J. Biol. Chem., 273, 10475-10484.

Neer, E.J. (1994). G-proteins - critical control points for transmembrane signals. Protein Science, 3, 3-14.

References 235 Neer, E.J. (1995). Heterotrimeric G-proteins - organisers of transmembrane signals. Cell, 80, 249-257.

Niggli, V., Andreoli, C., Roy, C., & Mangeat, P. (1995). Identification of a phosphatidylinositol-4,5-bisphosphate-binding domain in the N-terminal region of ezrin. FEBSLetters, 376, 172-176.

Nikolic, M., Dudek, H., Kwon, Y.T., Ramos, Y.F.M., & Tsai, L.H. (1996). The cdk5/p35 kinase is essential for neurite outgrowth during neuronal differentiation. Genes Dev., 10, 816-825.

Nikolic, M., Chou, M.M., Lu, W.G., Mayer, B.J., & Tsai, L.H. (1998). The p35/cdk5 kinase is a neuron-specific Rac effector that inhibits PAK1 activity. Nature, 395, 194- 198.

Nishiki, T., Narumiya, S., Morii, N., Yamamoto, M., Fujiwara, M., Kamata, Y., Sakaguchi, G., & Kozaki, S. (1990). ADP-ribosylation of the Rho/Rac proteins induces growth-inhibition, neurite outgrowth and acetylcholine esterase in cultured pc-12 cells. Biochem. Biophys. Res. Commun., 167, 265-272.

Nobes, C.D. & Hall, A. (1995). Rho, Rac, and Cdc42 GTPases regulate the assembly of multimolecular focal complexes associated with actin stress fibres, lamellipodia, and filopodia. Cell, 81, 53-62.

Nogales, E., Wolf, S.G., & Downing, K.H. (1998). Structure of the a(3 tubulin dimer by electron crystallography. Nature, 391, 199-203.

Nogales, E., Whittaker, M., Milligan, R.A., & Downing, K.H. (1999). High-resolution model of the microtubule. Cell, 96, 79-88.

Obaishi, H., Nakanishi, H., Mandai, K., Satoh, K., Satoh, A., Takahashi, K., Miyahara, M., Nishioka, H., Takaishi, K., & Takai, Y. (1998). Frabin, a novel FGDl-related actin filament-binding protein capable of changing cell shape and activating c-Jun N-terminal kinase. J. Biol. Chem., 273, 18697-18700.

Obermeier, A., Ahmed, S., Manser, E., Yen, S.C., Hall, C., & Lim, L. (1998). PAK promotes morphological changes by acting upstream of Rac. EMBO J., 17, 4328-4339.

Ohmichi, M., Decker, S.J., & Saltiel, A.R. (1992). Nerve growth-factor stimulates the tyrosine phosphorylation of a 38- kDa protein that specifically associates with the Src homology domain of phospholipase-C-gamma-1. J. Biol. Chem., 267, 21601-21606.

Ohno, S., Akita, Y., Konno, Y., Imajoh, S., & Suzuki, K. (1988). A novel phorbol ester receptor protein-kinase, NPKC, distantly related to the protein-kinase C family. Cell, 53, 731-741.

Olson, M.F., Ashworth, A., & Hall, A. (1995). An essential role for Rho, Rac, and Cdc42 GTPases in cell cycle progression through Gl. Science, 269, 1270-1272.

References 236 Olson, M.F., Pasteris, N.G., Gorski, J.L., & Hall, A. (1996). Faciogenital dysplasia protein (FGD1) and Vav, two related proteins required for normal embryonic development, are upstream regulators of Rho GTPases. Curr. Biol., 6, 1628-1633.

Ono, Y., Fujii, T., Igarashi, K., Kikkawa, U., Ogita, K., & Nishizuka, Y. (1988). Nucleotide-sequences of cDNAs for alpha-subspecies and gamma-subspecies of rat-brain protein kinase-C. Nucleic Acids Res., 16, 5199-5200.

Ono, Y., Fujii, T., Igarashi, K., Kuno, T., Tanaka, C., Kikkawa, U., & Nishizuka, Y. (1989). Phorbol ester binding to protein kinase-C requires a cysteine-rich zinc-finger-like sequence. Proc. Natl. Acad. Sci. USA, 86, 4868-4871.

Osman, M. A. & Cerione, R.A. (1998). IQGlp, a yeast homologue of the mammalian IQGAPs, mediates Cdc42p effects on the actin cytoskeleton. J. Cell Biol, 142, 443-455.

Otsu, M., Hiles, I., Gout, I., Fry, M.J., Ruizlarrea, F., Panayotou, G., Thompson, A., Dhand, R., Hsuan, J., Totty, N., Smith, A.D., Morgan, S.J., Courtneidge, S.A., Parker, P.J., & Waterfield, M.D. (1991). Characterisation of 2 85kd proteins that associate with receptor tyrosine kinases, middle-t/pp60c-Src complexes, and PI3-kinase. Cell, 65, 91- 104.

Palombella, V.J., Rando, O.J., Goldberg, A.L., & Maniatis, T. (1994). The ubiquitin- proteasome pathway is required for processing the NF- kappa-Bl precursor protein and the activation of NF-kappa-B. Cell, 78, 773-785.

Park, S.H. & Weinberg, R.A. (1995). A putative effector of Ral has homology to Rho/Rac GTPase activating proteins. Oncogene, 11, 2349-2355.

Parker, P.J., Coussens, L., Totty, N., Rhee, L., Young, S., Chen, E., Stabel, S., Waterfield, M.D., & Ullrich, A. (1986). The complete primary structure of protein- kinase-C - the major phorbol ester receptor. Science, 233, 853-859.

Pasteris, N.G., Cadle, A., Logie, L.J., Porteous, M.E.M., Schwartz, C.E., Stevenson, R.E., Glover, T.W., Wilroy, R.S., & Gorski, J.L. (1994). Isolation and characterisation of the faciogenital dysplasia (Aarskog-Scott syndrome) gene - a putative Rho/Rac guanine-nucleotide exchange factor. Cell, 79, 669-678.

Pawson, T. (1995). Protein modules and signalling networks. Nature, 373, 573-580.

Pawson, T. & Scott, J.D. (1997). Signalling through scaffold, anchoring, and adaptor proteins. Science, 278, 2075-2080.

Pendergast, A.M., Muller, A.J., Havlik, M.H., Maru, Y., & Witte, O.N. (1991). Bcr sequences essential for transformation by the Bcr-Abl oncogene bind to the Abl-SH2 regulatory domain in a non-phosphotyrosine- dependent manner. Cell, 66, 161-171.

Perona, R., Montaner, S., Saniger, L., SanchezPerez, I., Bravo, R., & Lacal, J.C. (1997). Activation of the nuclear factor-kappa B by Rho, Cdc42, and Rac-1 proteins. Genes Dev., 11, 463-475.

References 237 Pestonjamasp, K., Amieva, M.R., Strassel, C.P., Nauseef, W.M., Furthmayr, H., & Luna, E.J. (1995). Moesin, ezrin, and p205 are actin-binding proteins associated with neutrophil plasma-membranes. Mol. Biol. Cell, 6, 247-259.

Pluskey, S., Wandless, T.J., Walsh, C.T., & Shoelson, S.E. (1995). Potent stimulation of sh-ptp2 phosphatase-activity by simultaneous occupancy of both SH2 domains. J. Biol. Chem., 270, 2897-2900.

Polyak, K., Xia, Y., Zweier, J.L., Kinzler, K.W., & Vogelstein, B. (1997). A model for p53-induced apoptosis. Nature, 389, 300-305.

Prasher, D.C., Eckenrode, V.K., Ward, W.W., Prendergast, F.G., & Cormier, M.J. (1992). Primary structure of the Aequorea-Victoria green-fluorescent protein. Gene, 111, 229-233.

Price, M.A., Rogers, A.E., & Treisman, R. (1995). Comparative-analysis of the ternary complex factors Elk-1, SAP-la and SAP-2 (ERP/NET). EMBO J., 14, 2589-2601.

Qiu, R.G., Chen, J., Kim, D., McCormick, F., & Symons, M. (1995). An essential role for Rac in Ras transformation. Nature, 374, 457-459.

Qiu, R.G., Abo, A., McCormick, F., & Symons, M. (1997). Cdc42 regulates anchorage- independent growth and is necessary for Ras transformation. Mol. Cell. Biol., 17, 3449- 3458.

Quinones, M.A., Mundschau, L.J., Rake, J.B., & Faller, D.V. (1991). Dissociation of platelet-derived growth-factor (PDGF) receptor autophosphorylation from other PDGF- mediated 2nd messenger events. J. Biol. Chem., 266, 14055-14063.

Raingeaud, J., Gupta, S., Rogers, J.S., Dickens, M., Han, J.H., Ulevitch, R.J., & Davis, R.J. (1995). Pro-inflammatory cytokines and environmental-stress cause p38 mitogen- activated protein-kinase activation by dual phosphorylation on tyrosine and threonine. J. Biol. Chem., 270, 7420-7426.

Rameh, L.E., Chen, C.S., & Cantley, L.C. (1995). Phosphatidylinositol (3,4,5)p-3 interacts with SH2 domains and modulates PI-3-kinase association with tyrosine- phosphorylated proteins. Cell, 83, 821-830.

Ramesh, N., Anton, I.M., Hartwig, J.H., & Geha, R.S. (1997). Wip, a protein associated with Wiskott-Aldrich syndrome protein, induces actin polymerisation and redistribution in lymphoid cells. Proc. Natl. Acad. Sci. USA, 94, 14671-14676.

Regnier, C.H., Song, H.Y., Gao, X., Goeddel, D.V., Cao, Z.D., & Rothe, M. (1997). Identification and characterisation of an IkB kinase. Cell, 90, 373-383.

Reid, T., Furayashiki, T., Ishizaki, T., Watanabe, G., Watanabe, N., Fujisawa, K., Morii, N., Madaule, P., & Narumiya, S. (1996). Rhotekin, a new putative target for Rho bearing homology to a serine/threonine kinase, PKN, and Rhophilin in the Rho-binding domain. J. Biol. Chem., 271, 13556-13560.

References 238 Reinhard, J., Scheel, A.A., Diekmann, D., Hall, A., Ruppert, C., & Bahler, M. (1995). A novel type of myosin implicated in signalling by Rho family GTPases. EMBO J., 14, 697- 704.

Ren, R.B., Mayer, B.J., Cicchetti, P., & Baltimore, D. (1993). Identification of a 10- amino acid proline-rich SH3 binding-site. Science, 259, 1157-1161.

Ren, X.D., Bokoch, G.M., Traynorkaplan, A., Jenkins, G.H., Anderson, R.A., & Schwartz, M.A. (1996). Physical association of the small GTPase Rho with a 68-kDa phosphatidylinositol 4-phosphate 5-kinase in Swiss 3T3 cells. Mol Biol Cell, 7, 435- 442.

Rey, I., Taylor Harris, P., van Erp, H., & Hall, A. (1994). R-Ras interacts with RasGAP, neurofibromin and c-Raf but does not regulate cell growth or differentiation. Oncogene, 9, 685-692.

Ridley, A.J. & Hall, A. (1992). The small GTP-binding protein Rho regulates the assembly of focal adhesions and actin stress fibres in response to growth factors. Cell, 70, 389-399.

Ridley, A.J., Paterson, H.F., Johnston, C.L., Diekmann, D., & Hall, A. (1992). The small GTP-binding protein Rac regulates growth factor induced membrane ruffling. Cell, 70, 401-410.

Ridley, A.J., Self, A.J., Kasmi, F., Paterson, H.F., Hall, A., Marshall, C.J., & Ellis, C. (1993). Rho family GTPase-activating proteins p i90, Bcr and RhoGAP show distinct specificities in-vitro and in-vivo. EMBO J., 12, 5151-5160.

Rodriguez-Viciana, P., Wame, P H., Dhand, R., Vanhaesebroeck, B., Gout, I., Fry, M.J., Waterfield, M.D., & Downward, J. (1994). Phosphatidylinositol-3-OH kinase as a direct target of Ras. Nature, 370, 527-532.

Rodriguez-Viciana, P., Warne, P.H., Khwaja, A., Marte, B.M., Pappin, D., Das, P., Waterfield, M.D., Ridley, A., & Downward, J. (1997). Role of phosphoinositide 3-OH kinase in cell transformation and control of the actin cytoskeleton by Ras. Cell, 89, 457- 467.

Ron, D., Zannini, M., Lewis, M., Wickner, R.B., Hunt, L.T., Graziani, G., Tronick, S R., Aaronson, S. A., & Eva, A. (1991). A region of proto-Dbl essential for its transforming activity shows sequence similarity to a yeast-cell cycle gene, cdc24, and the human breakpoint cluster gene, Bcr. New Biologist, 3, 372-379.

Roy, C., Martin, M., & Mangeat, P. (1997). A dual involvement of the amino-terminal domain of ezrin in F-and G- actin binding. J. Biol Chem., 272, 20088-20095.

Roy, S., McPherson, R.A., Apolloni, A., Yan, J., Lane, A., ClydeSmith, J., & Hancock, J.F. (1998). 14-3-3 facilitates Ras-dependent Raf-1 activation in vitro and in vivo. M ol Cell Biol, 18, 3947-3955.

References 239 Rozakis-Adcock, M., Fernley, R., Wade, J., Pawson, T., & Bowtell, D. (1993). The SH2 and SH3 domains of mammalian Grb2 couple the EGF receptor to the Ras activator msosl. Nature, 363, 83-85.

Salim, K., Bottomley, M.J., Querfurth, E., Zvelebil, M.J., Gout, I., Scaife, R., Margolis, R.L., Gigg, R., Smith, C.I.E., Driscoll, P.C., Waterfield, M.D., & Panayotou, G. (1996). Distinct specificity in the recognition of phosphoinositides by the pleckstrin homology domains of and Bruton's tyrosine kinase. EMBO J., 15, 6241-6250.

Saras, J., Franzen, P., Aspenstrom, P., Heilman, U., Gonez, L.J., & Heldin, C.H. (1997). A novel GTPase-activating protein for Rho interacts with a PDZ domain of the protein- tyrosine phosphatase PTPL1. J. Biol. Chem., 272, 24333-24338.

Schaber, M.D., Garsky, V.M., Boylan, D., Hill, W.S., Scolnick, E.M., Marshall, M.S., Sigal, I.S., & Gibbs, J.B. (1989). Ras interaction with the GTPase-activating protein (GAP). Proteins-Structure Function And Genetics, 6, 306-315.

Schaffhausen, B. (1995). SH2 domain-structure and function. Biochimica Et Biophysica Acta-Reviews On Cancer, 1242, 61-75.

Scherle, P., Behrens, T., & Staudt, L.M. (1993). Ly-GDI, a GDP-dissociation inhibitor of the RhoA GTP-binding protein, is expressed preferentially in lymphocytes. Proc. Natl. Acad. Sci. USA, 90, 7568-7572.

Schoenwaelder, S.M. & Burridge, K. (1999). Bi-directional signalling between the cytoskeleton and integrins. Curr. Opin. Cell Biol., 11, 274-286.

Schreck, R., Rieber, P., & Baeuerle, P. A. (1991). Reactive oxygen intermediates as apparently widely used messengers in the activation of the NF-kappa-B transcription factor and HIV-1. EMBO J., 10, 2247-2258.

Schuller, C., Brewster, J.L., Alexander, M.R., Gustin, M.C., & Ruis, H. (1994). The HOG pathway controls osmotic regulation of transcription via the stress-response element (STRE) of the Saccharomyces Cerevisiae cttl gene. EMBO J., 13, 4382-4389.

Seger, R. & Krebs, E.G. (1995). Protein kinases & the MAPK signalling cascade. Faseb J., 9, 726-735.

Selbie, L.A. & Hill, S.J. (1998). G protein-coupled-receptor cross-talk: the fine-tuning of multiple receptor-signalling pathways. Trends Pharmacol. Sci., 19, 87-93.

Sells, M.A. & Chernoff, J. (1997). Emerging from the PAK: the p21-activated protein kinase family. Trends Cell Biol., 7, 162-167.

Settleman, J., Narasimhan, V., Foster, L.C., & Weinberg, R.A. (1992a). Molecular- cloning of cDNAs encoding the GAP-associated protein p i90 - implications for a signalling pathway from Ras to the nucleus. Cell, 69, 539-549.

Settleman, J., Albright, C.F., Foster, L.C., & Weinberg, R.A. (1992b). Association between GTPase activators for Rho and Ras families. Nature, 359, 153-154.

References 240 Shaw, G. (1996). The pleckstrin homology domain: an intriguing multifunctional protein module. Bioessctys, 18, 35-46.

Siebenlist, U., Franzoso, G., & Brown, K. (1994). Structure, regulation and function of NF-kappa-B. Ann. Rev. Cell Biol., 10, 405-455.

Skalli, O., Jones, J.C.R., Gagescu, R., & Goldman, R.D. (1994). IFAP-300 is common to desmosomes and hemidesmosomes and is a possible linker of intermediate filaments to these junctions. J. Cell Biol, 125, 159-170.

Smith, F.J.D., Eady, R.A.J., Leigh, I.M., McMillan, J.R., Rugg, E.L., Kelsell, D.P., Bryant, S.P., Spurr, N.K., Geddes, J.F., Kirtschig, G., Milana, G., deBono, A.G., Owaribe, K., Wiche, G., Pulkkinen, L., Uitto, J., McLean, W.H.I., & Lane, E.B. (1996). Plectin deficiency results in muscular dystrophy with epidermolysis bullosa. Nature Genetics, 13, 450-457.

Song, H.Y., Regnier, C.H., Kirschning, C.J., Goeddel, D.V., & Rothe, M. (1997). Tumour necrosis factor (TNF)-mediated kinase cascades: bifurcation of NFkB and c-Jun N-terminal kinase (JNK/SAPK) pathways at TNF receptor-associated factor 2. Proc. Natl. Acad. Sci. USA, 94, 9792-9796.

Songyang, Z., Shoelson, S.E., Chaudhuri, M., Gish, G., Pawson, T., Haser, W.G., King, F., Roberts, T., Ratnofsky, S., Lechleider, R.J., & et al (1993). SH2 domains recognise specific phosphopeptide sequences. Cell, 72, 767-778.

Sorokin, A., Reed, E., Nnkemere, N., Dulin, N O., & Schlessinger, J. (1998). Crk protein binds to PDGF receptor and insulin receptor substrate-1 with different modulating effects on pd. Oncogene, 16, 2425-2434.

Stancovski, I. & Baltimore, D. (1997). NF kB activation: the IkB kinase revealed? Cell, 91, 299-302.

Stevenson, M.A., Pollock, S.S., Coleman, C.N., & Calderwood, S.K. (1994). X- irradiation, phorbol esters, and H 2O2 stimulate mitogen-activated protein-kinase activity in NIH-3T3 cells through the formation of reactive oxygen intermediates. Cancer Research, 54, 12-15.

Stirling, R. V. & Dunlop, S. A. (1995). The dance of the growth cones - where to next. Trends Neurosci., 18, 111-115.

Sulciner, D.J., Irani, K., Yu, Z.X., Ferrans, V.J., Goldschmidt-Clermont, P., & Finkel, T. (1996). Racl regulates a cytokine-stimulated, redox-dependent pathway necessary for NF-kappaB activation. Mol. Cell. Biol., 16, 7115-7121.

Sun, X.J., Miralpeix, M., Myers, M.G., Glasheen, E.M., Backer, J.M., Kahn, C.R., & White, M.F. (1992). Expression and function of IRS-1 in insulin signal transmission. J. Biol. Chem., 267, 22662-22672.

Sun, H.Q., Kwiatkowska, K., & Yin, H.L. (1995). Actin monomer binding-proteins. Curr. Opin. Cell Biol., 1, 102-110.

References 241 Sundaresan, M., Yu, Z.X., Ferrans, V.J., Irani, K., & Finkel, T. (1995). Requirement for generation of H20 2 for platelet-derived growth-factor signal-transduction. Science, 270, 296-299.

Sundaresan, M., Yu, Z.X., Ferrans, V.J., Sulciner, D.J., Gutkind, J.S., Irani, K., Goldschmidt-Clermont, P.J., & Finkel, T. (1996). Regulation of reactive-oxygen-species generation in fibroblasts by Racl. Biochem. J., 318, 379-382.

Svitkina, T.M., Verkhovsky, A.B., & Borisy, G.G. (1996). Plectin sidearms mediate interaction of intermediate filaments with microtubules and other components of the cytoskeleton. J. Cell Biol, 135, 991-1007.

Svitkina, T.M., Verkhovsky, A.B., McQuade, K.M., & Borisy, G.G. (1997). Analysis of the actin-myosin ii system in fish epidermal keratocytes: mechanism of cell body translocation. J. Cell Biol, 139, 397-415.

Symons, M., Derry, J.M.J., Karlak, B., Jiang, S., Lemahieu, V., McCormick, F., Francke, U., & Abo, A. (1996). Wiskott-Aldrich syndrome protein, a novel effector for the GTPase Cdc42Hs, is implicated in actin polymerisation. Cell, 84, 723-734.

Takahashi, K., Sasaki, T., Mammoto, A., Takaishi, K., Kameyama, T., Tsukita, S., & Takai, Y. (1997). Direct interaction of the Rho GDP dissociation inhibitor with ezrin/radixin/moesin initiates the activation of the Rho small G protein. J. Biol Chem., 272, 23371-23375.

Takaishi, K., Sasaki, T., Kameyama, T., Tsukita, S., & Takai, Y. (1995). Translocation of activated Rho from the cytoplasm to membrane ruffling area, cell-cell adhesion sites and cleavage furrows. Oncogene, 11, 39-48.

Tan, E.C., Leung, T., Manser, E., & Lim, L. (1993). The human active breakpoint cluster region-related gene encodes a brain protein with homology to guanine nucleotide exchange proteins and GTPase-activating proteins. J. Biol. Chem., 268, 27291-27298.

Tanaka, E., Ho, T., & Kirschner, M.W. (1995). The role of microtubule dynamics in growth cone motility and axonal growth. J. Cell Biol., 128, 139-155.

Tanaka, K. & Takai, Y. (1998). Control of reorganisation of the actin cytoskeleton by Rho family small GTP-binding proteins in yeast. Curr. Opin. Cell Biol, 10, 112-116.

Tang, Y., Chen, Z.X., Ambrose, D., Liu, J.H., Gibbs, J.B., Chernoff, J., & Field, J. (1997). Kinase-deficient PAK1 mutants inhibit Ras transformation of rat-1 fibroblasts. Mol. Cell. Biol., 17, 4454-4464.

Tapon, N., Nagata, K., Lamarche, N., & Hall, A. (1998). A new Rac target POSH is an SH3-containing scaffold protein involved in the JNK and NF-kappa B signalling pathways. EMBO J., 17, 1395-1404.

Tatsis, N., Lannigan, D.A., & Macara, I.G. (1998). The function of the pi 90 Rho GTPase-activating protein is controlled by its N-terminal GTP binding domains. J. Biol. Chem., 273, 34631-34638.

References 242 Taylor, J.M., Macklem, M.M., & Parsons, J.T. (1999). Cytoskeletal changes induced by GRAF, the GTPase regulator associated with focal adhesion kinase, are mediated by Rho. J. Cell Sci., 112, 231-242.

Teo, M., Manser, E., & Lim, L. (1995). Identification and molecular cloning of a p21Cdc42/Racl -activated serine/threonine kinase that is rapidly activated by thrombin in platelets. J. Biol. Chem., 270, 26690-26697.

Teramoto, H., Crespo, P., Coso, O.A., Igishi, T., Xu, N.Z., & Gutkind, J.S. (1996a). The small GTP-binding protein Rho activates c-Jun N-terminal kinases stress-activated protein kinases in human kidney 293 T cells - evidence for a PAK-independent signalling pathway.J. Biol. Chem., 271, 25731-25734.

Teramoto, H., Coso, O.A., Miyata, H., Igishi, T., Miki, T., & Gutkind, J.S. (1996b). Signalling from the small GTP-binding proteins Racl and Cdc42 to the c-Jun N-terminal kinase stress-activated protein kinase pathway - a role for mixed lineage kinase 3 protein- tyrosine kinase 1, a novel member of the mixed lineage kinase family. J. Biol. Chem., Ill, 27225-27228.

Thanos, D. & Maniatis, T. (1995). NFkB - a lesson in family values. Cell, 80, 529-532.

Theriot, J.A. (1997). Accelerating on a treadmill: ADF/cofilin promotes rapid actin filament turnover in the dynamic cytoskeleton. J. Cell Biol., 136, 1165-1168.

Thompson, A.D., Braun, B.S., Arvand, A., Stewart, S.D., May, W.A., Chen, E., Korenberg, J., & Denny, C. (1996). Eat-2 is a novel SH2 domain containing protein that is up regulated by Ewing's sarcoma EWS/FLI1 fusion gene. Oncogene, 13, 2649-2658.

Threadgill, R., Bobb, K., & Ghosh, A. (1997). Regulation of dendritic growth and remodelling by Rho, Rac, and Cdc42. Neuron, 19, 625-634.

Tigyi, G., Fischer, D.J., Sebok, A., Yang, C., Dyer, D.L., & Miledi, R. (1996). Lysophosphatidic acid-induced neurite retraction in PC 12 cells: control by phosphoinositide-Ca2+ signalling and Rho. JNeurochem., 66, 537-548.

Tognon, C.E., Kirk, H.E., Passmore, L.A., Whitehead, IP., Der, C.J., & Kay, R.J. (1998). Regulation of RasGRP via a phorbol ester-responsive Cl domain. Mol. Cell. Biol., 18, 6995-7008.

Toker, A. & Cantley, L.C. (1997). Signalling through the lipid products of phosphoinositide-3-OH kinase. Nature, 387, 673-676.

Toker, A. (1998). The synthesis and cellular roles of phosphatidylinositol 4,5- bisphosphate. Curr. Opin. Cell Biol., 10, 254-261.

Toksoz, D. & Williams, D.A. (1994). Novel human oncogene Lbc detected by transfection with distinct homology regions to signal-transduction products. Oncogene, 9, 621-628.

Tolias, K.F., Cantley, L.C., & Carpenter, C.L. (1995). Rho family GTPases bind to phosphoinositide kinases. J. Biol. Chem., 270, 17656-17659.

References 243 Tominaga, T. & Barber, D.L. (1998). Na-H exchange acts downstream ofRhoAto regulate integrin-induced cell adhesion and spreading. Mol. Biol Cell, 9, 2287-2303.

Tominaga, T., Ishizaki, T., Narumiya, S., & Barber, D.L. (1998). P160ROCK mediates RhoA activation of Na-H exchange. EMBO J., 17, 4712-4722.

Toure, A., Dorseuil, O., Morin, L., Timmons, P., Jegou, B., Reibel, L., & Gacon, G. (1998). MgcRacGAP, a new human GTPase-activating protein for Rac and Cdc42 similar to Drosophila Rotund RacGAP gene product, is expressed in male germ cells. J. Biol Chem., 273, 6019-6023.

Trahey, M. & McCormick, F. (1987). A cytoplasmic protein stimulates normal N-Ras p21 GTPase, but does not affect oncogenic mutants. Science, 238, 542-545.

Treisman, R. (1994). Ternary complex factors: growth factor regulated transcriptional activators. Curr. Opin. Genet. Dev., 4, 96-101.

Treisman, R. (1996). Regulation of transcription by MAP kinase cascades. Curr. Opin. Cell Biol, 8, 205-215.

Trub, T., Choi, W.E., Wolf, G., Ottinger, E., Chen, Y.J., Weiss, M., & Shoelson, S.E. (1995). Specificity of the PTB domain of She for (3-turn-forming pentapeptide motifs amino terminal to phosphotyrosine. J. Biol. Chem., 270, 18205-18208.

Tsai, L.H., Delalle, I., Caviness, V.S., Chae, T., & Harlow, E. (1994). P35 is a neural- specific regulatory subunit of cyclin-dependent kinase-5. Nature, 371, 419-423.

Tsukita, S. & Yonemura, S. (1997). ERM (ezrin/radixin/moesin) family: from cytoskeleton to signal transduction. Curr. Opin. Cell Biol., 9, 70-75.

Tsukita, S., Oishi, K., Sato, N., Sagara, J., & Kawai, A. (1994). ERM family members as molecular linkers between the cell-surface glycoprotein CD44 and actin-based . J. Cell Biol, 126, 391-401.

Turunen, O., Wahlstrom, T., & Vaheri, A. (1994). Ezrin has a COOH-terminal actin- binding site that is conserved in the ezrin protein family. J. Cell Biol., 126, 1445-1453.

Tzivion, G., Luo, Z.J., & Avruch, J. (1998). A dimeric 14-3-3 protein is an essential for Raf kinase activity. Nature, 394, 88-92.

Ueda, T., Kikuchi, A., Ohga, N., Yamamoto, J., & Takai, Y. (1990). Purification and characterisation from bovine brain cytosol of a novel regulatory protein inhibiting the dissociation of GDP from and the subsequent binding of GTP to RhoB p20, a Ras p21- like GTP-binding protein. J. Biol. Chem., 265, 9373-9380.

Van Aelst, L. & D'Souza-Schorey, C. (1997). Rho GTPases and signalling networks. Genes Dev., 11, 2295-2322.

Van Aelst, L., Barr, M., Marcus, S., Polverino, A., & Wigler, M. (1993). Complex- formation between Ras and Raf and other protein-kinases. Proc. Natl. Acad. Sci. USA, 90, 6213-6217.

References 244 Van Aelst, L., Joneson, T., & Bar Sagi, D. (1996). Identification of a novel Racl- interacting protein involved in membrane ruffling. EMBO J., 15, 3778-3786.

Van der Geer, P., Wiley, S., Lai, V.K.M., Olivier, J.P., Gish, G.D., Stephens, R., Kaplan, D., Shoelson, S., & Pawson, T. (1995). A conserved amino-terminal She domain binds to phosphotyrosine motifs in activated receptors and phosphopeptides. Curr. B io, l5, 404-412.

Van der Geer, P., Wiley, S., Gish, G.D., Lai, V.K.M., Stephens, R., White, M.F., Kaplan, D., & Pawson, T. (1996). Identification of residues that control specific binding of the She phosphotyrosine-binding domain to phosphotyrosine sites. Proc. Natl. Acad. Sci. USA, 93, 963-968.

Van Leeuwen, F.N., Kain, H.E.T., van der Kammen, R.A., Michiels, F., Kranenburg, O.W., & Collard, J.G. (1997). The guanine nucleotide exchange factor Tiaml affects neuronal morphology; opposing roles for the small GTPases Rac and Rho. J. Cell Biol., 139, 797-807.

Van Weering, D.H.J., deRooij, J., Marte, B., Downward, J., Bos, J.L., & Burgering, B.M.T. (1998). Protein kinase B activation and formation are independent phosphoinositide 3-kinase-mediated events differentially regulated by endogenous Ras. Mol. Cell Biol, 18, 1802-1811.

Verma, I.M., Stevenson, J.K., Schwarz, E.M., Vanantwerp, D., & Miyamoto, S. (1995). R c I/NF kB/I kB family - intimate tales of association and dissociation. Genes Dev., 9, 2723-2735.

Vincent, S. & Settleman, J. (1997). The PRK2 kinase is a potential effector target of both Rho and Rac GTPases and regulates actin cytoskeletal organisation. Mol. Cell. Biol., 17, 2247-2256.

Vojtek, A.B., Hollenberg, S.M., & Cooper, J.A. (1993). Mammalian Ras interacts directly with the serine threonine kinase Raf. Cell, 74, 205-214.

Waksman, G., Kominos, D., Robertson, S.C., Pant, N., Baltimore, D., Birge, R.B., Cowburn, D., Hanafusa, H., Mayer, B.J., Overduin, M., Resh, M.D., Rios, C.B., Silverman, L., & Kuriyan, J. (1992). Crystal-structure of the phosphotyrosine recognition domain SH2 of v-Src complexed with tyrosine-phosphorylated peptides. Nature, 358, 646-653.

Waksman, G., Shoelson, S.E., Pant, N., Cowburn, D., & Kuriyan, J. (1993). Binding of a high-affinity phosphotyrosyl peptide to the Src SH2 domain - crystal-structures of the complexed and peptide-free forms. Cell, 72, 779-790.

Wandless, T.J. (1996). SH2 domains - a question of independence. Curr. Biol., 6, 125- 127.

Wang, L.H. & Strittmatter, S.M. (1996). A family of rat CRMP genes is differentially expressed in the nervous system. J. Neurosci., 16, 6197-6207.

References 245 Wang, L.H. & Strittmatter, S.M. (1997). Brain CRMP forms heterotetramers similar to liver dihydropyrimidinase. J. Neurochem., 69, 2261-2269.

Wame, P.H., Viciana, P.R., & Downward, J. (1993). Direct interaction of Ras and the amino-terminal region of Raf-1 in- vitro. Nature, 364, 352-355.

Wasserman, S. (1998). FH proteins as cytoskeletal organisers. Trends Cell Biol, 8, 111- 115.

Watanabe, G., Saito, Y., Madaule, P., Ishizaki, T., Fujisawa, K., Morii, N., Mukai, H., Ono, Y., Kakizuka, A., & Narumiya, S. (1996). Protein kinase N (PKN) and PKN- related protein Rhophilin as targets of small GTPase Rho. Science, 271, 645-648.

Watanabe, N., Madaule, P., Reid, T., Ishizaki, T., Watanabe, G., Kakizuka, A., Saito, Y., Nakao, K., Jockusch, B.M., & Narumiya, S. (1997). pl40mDia, a mammalian homologue of Drosophila diaphanous, is a target protein for Rho small GTPase and is a ligand for profilin. EMBO J., 16, 3044-3056.

Wegner, A. (1976). Head to tail polymerisation of actin. J. Mol. Biol., 108, 139-150.

Weisenberg, R.C. (1972). Microtubule formation in vitro in solutions containing low calcium concentrations. Science, 177, 1104-1105.

Weiss, A. (1995). Signal-transduction - zapping tandem SH2 domains. Nature, 377, 17- 18.

Weissbach, L., Settleman, J., Kalady, M.F., Snijders, A.J., Murthy, A.E., Yan, Y.X., & Bernards, A. (1994). Identification of a human RasGAP-related protein containing calmodulin-binding motifs. J. Biol. Chem., 269, 20517-20521.

Welch, M.D., Iwamatsu, A., & Mitchison, T.J. (1997). Actin polymerisation is induced by Arp2/3 protein complex at the surface of Listeria monocytogenes. Nature, 385, 265- 269.

Welch, M.D., Rosenblatt, J., Skoble, J., Portnoy, D.A., & Mitchison, T.J. (1998). Interaction of human Arp2/3 complex and the Listeria monocytogenes Act A protein in actin filament nucleation. Science, 281, 105-108.

Westwick, J.K., Lambert, Q.T., Clark, G.J., Symons, M., Van Aelst, L., Pestell, R.G., & Der, C.J. (1997). Rac regulation of transformation, gene expression, and actin organisation by multiple, PAK-independent pathways. Mol. Cell. Biol., 17, 1324-1335.

Whitehead, I., Kirk, H., Tognon, C., Trigogonzalez, G., & Kay, R. (1995a). Expression cloning of Lfc, a novel oncogene with structural similarities to guanine-nucleotide exchange factors and to the regulatory region of protein-kinase-C. J. Biol. Chem., 270, 18388-18395.

Whitehead, I., Kirk, H., & Kay, R. (1995b). Retroviral transduction and oncogenic selection of a cDNA-encoding Dbs, a homologue of the Dbl guanine-nucleotide exchange factor. Oncogene, 10, 713-721.

References 246 Whitehead, I.P., Abe, K., Gorski, J.L., & Der, C.J. (1998). Cdc42 and FGD1 cause distinct signalling and transforming activities. Mol. Cell B io l, 18, 4689-4697.

Whitmarsh, A.J., Shore, P., Sharrocks, A.D., & Davis, R.J. (1995). Integration ofMAP kinase signal-transduction pathways at the serum response element. Science, 269, 403- 407.

Wirth, J.A., Jensen, K.A., Post, P.L., Bement, W.M., & Mooseker, M.S. (1996). Human myosin-IXb, an unconventional myosin with a chimaerin-like Rho/Rac GTPase-activating protein domain in its tail. J. Cell Sci., 109, 653-661.

Wittekind, M., Mapelli, C., Farmer, B.T., Suen, K.L., Goldfarb, V., Tsao, J.L., Lavoie, T., Barbacid, M., Meyers, C.A., & Mueller, L. (1994). Orientation of peptide-ffagments from Sos proteins bound to the N-terminal SH3 domain of Grb2 determined by NMR- spectroscopy. Biochemistry, 33, 13531-13539.

Xia, Z.G., Dickens, M., Raingeaud, J., Davis, R.J., & Greenberg, M.E. (1995). Opposing effects ofERK and JNK-p38 MAP kinases on apoptosis. Science, 270, 1326-1331.

Xiao, J.H., Davidson, I., Matthes, H., Garnier, J.M., & Chambon, P. (1991). Cloning, expression, and transcriptional properties of the human enhancer factor TEF-1. Cell, 65, 551-568.

Xu, W.Q., Harrison, S.C., & Eck, M.J. (1997). Three-dimensional structure of the tyrosine kinase c-Src. Nature, 385, 595-602.

Yamada, K.M. & Geiger, B. (1997). Molecular interactions in cell adhesion complexes. Curr. Opin. Cell Biol., 9, 76-85.

Yamada, K.M. & Miyamoto, S. (1995). Integrin transmembrane signalling and cytoskeletal control. Curr. Opin. Cell Biol., 7, 681-689.

Yamasaki, H., Itakura, C., & Mizutani, M. (1991). Hereditary hypotrophic axonopathy with neurofilament deficiency in a mutant strain of the Japanese quail. Acta Neuropathologica, 82, 427-434.

Yan, M.H., Dai, T.N., Deak, J.C., Kyriakis, J.M., Zon, L.I., Woodgett, J R., & Templeton, D.J. (1994). Activation of stress-activated protein-kinase by MEKK1 phosphorylation of its activator SEK1. Nature, 372, 798-800.

Yang, B.S., Hauser, C.A., Henkel, G., Colman, M.S., VanBeveren, C., Stacey, K.J., Hume, D.A., Maki, R.A., & Ostrowski, M.C. (1996). Ras-mediated phosphorylation of a conserved threonine residue enhances the transactivation activities of c-Etsl and c-Ets2. Mol. Cell. Biol, 16, 538-547.

Yang, N., Higuchi, O., Ohashi, K., Nagata, K., Wada, A., Kangawa, K., Nishida, E., & Mizuno, K. (1998). Cofilin phosphorylation by LIM-kinase 1 and its role in Rac- mediated actin reorganisation. Nature, 393, 809-812.

Yin, H.L. (1987). Gelsolin - calcium-regulated and polyphosphoinositide-regulated actin- modulating protein. Bioessays, 1, 176-179.

References 247 Yoshida, H., Watanabe, A., & Ihara, Y. (1998). Collapsin response mediator protein-2 is associated with neurofibrillary tangles in Alzheimer's disease. J. Biol. Chem., 273, 9761- 9768.

Yu, H.T., Chen, J.K., Feng, S.B., Dalgarno, D C., Brauer, A.W., & Schreiber, S.L. (1994). Structural basis for the binding of proline-rich peptides to SH3 domains. Cell, 76, 933-945.

Zalcman, G., Closson, V., Camonis, J., Honore, N., Rousseau Merck, M.F., Tavitian, A., & Olofsson, B. (1996). RhoGDI-3 is a new GDP dissociation inhibitor (GDI) - identification of a non-cytosolic GDI protein interacting with the small GTP-binding proteins RhoB and RhoG. J. Biol Chem., 271, 30366-30374.

Zhang, J., King, W.G., Dillon, S., Hall, A., Feig, L., & Rittenhouse, S.E. (1993a). Activation of platelet phosphatidylinositide 3-kinase requires the small GTP-binding protein Rho. J. Biol. Chem., 268, 22251-22254.

Zhang, X.F., Settleman, J., Kyriakis, J.M., Takeuchisuzuki, E., Elledge, S.J., Marshall, M.S., Bruder, J.T., Rapp, U.R., & Avruch, J. (1993b). Normal and oncogenic p21(Ras) proteins bind to the amino-terminal regulatory domain of c-Raf-1. Nature, 364, 308-313.

Zhang, S., Han, J., Sells, M.A., Chernoff, J., Knaus, U.G., Ulevitch, R.J., & Bokoch, G.M. (1995). Rho family GTPases regulate p38 mitogen-activated protein kinase through the downstream mediator PAK1. J. Biol. Chem., 270, 23934-23936.

Zhao, Z.S., Manser, E., Chen, X.Q., Chong, C., Leung, T., & Lim, L. (1998). A conserved negative regulatory region in alpha PAK: inhibition of PAK kinases reveals their morphological roles downstream of Cdc42 and Racl. Mol. Cell. Biol., 18, 2153- 2163.

Zheng, Y., Cerione, R., & Bender, A. (1994a). Control of the yeast bud-site assembly GTPase Cdc42. Catalysis of guanine nucleotide exchange by Cdc24 and stimulation of GTPase activity by Bem3. J. Biol. Chem., 269, 2369-2372.

Zheng, Y., Bagrodia, S., & Cerione, R.A. (1994b). Activation of phosphoinositide 3- kinase activity by Cdc42Hs binding to p85. J. Biol. Chem., 269, 18727-18730.

Zheng, Y., Olson, M.F., Hall, A., Cerione, R.A., & Toksoz, D. (1995a). Direct involvement of the small GTP-binding protein Rho in Lbc oncogene function. J. Biol. Chem., 270, 9031-9034.

Zheng, Y.X., Wong, M.L., Alberts, B., & Mitchison, T. (1995b). Nucleation of microtubule assembly by a gamma-tubulin-containing ring complex. Nature, 378, 578- 583.

Zheng, Y., Zangrilli, D., Cerione, R.A., & Eva, A. (1996a). The pleckstrin homology domain mediates transformation by oncogenic Dbl through specific intracellular targeting. J. Biol. Chem., 271, 19017-19020.

References 248 Zheng, Y., Fischer, D.J., Santos, M.F., Tigyi, G., Pasteris, N.G., Gorski, J.L., & Xu, Y.H. (1996b). The faciogenital dysplasia gene product FGD1 functions as a Cdc42Hs- specific guanine-nucleotide exchange factor. J. Biol Chem., 271, 33169-33172.

Zhou, M.M., Ravichandran, K.S., Olejniczak, E.T., Petros, A.M., Meadows, R.P., Sattler, M., Harlan, J.E., Wade, W.S., Burakoff, S.J., & Fesik, S.W. (1995). Structure and ligand recognition of the phosphotyrosine binding domain of She. Nature, 378, 584- 592.

Zhou, M.M., Huang, B.H., Olejniczak, E.T., Meadows, R.P., Shuker, SB., Miyazaki, M., Trub, T., Shoelson, S.E., & Fesik, S.W. (1996). Structural basis for IL-4 receptor phosphopeptide recognition by the IRS-1 PTB domain. Nature Structural Biol, 3, 388- 393.

Zipkin, I.D., Kindt, R.M., & Kenyon, C.J. (1997). Role of a new Rho family member in cell migration and axon guidance in C-Elegans. Cell, 90, 883-894.

Zohn, I.M., Campbell, S.L., Khosravi-Far, R., Rossman, K.L., & Der, C.J. (1998). Rho family proteins and Ras transformation: the RHOad less travelled gets congested. Oncogene, 17, 1415-1438.

References 249 APPENDICES

Appendices 250 a g tatcaa aATGCCATCCAAAGAGTCCTGGTCGGGGAGGAAAACTAATAG AGCTACAGTTCACAAATCAAACCAAGAGGGCCGTCAGCAAGATTTATTGA 517 TAGCAGCCTTGGGAATGAAACTGGGTTCTCAAAAGTCGTCTGTGACAATC - 5 67 TGGCAACCTCTGAAACTCTTTGCTTATTCGCAG|TGACATCACTTGTTAG - 617 AAGAGCAACTCTGAAAGAAAATGAACAAATTCCAAAATATGAGAAGGTTC - 667 ACAATT'TCAAGGTGCATACGTTCCGAGGGCCACACTGGTGTGAATACTGT - 717 GCCAACTTCATGTGGGGCCTCATTGCTCAGGGAGTGAAATGTGCAGATTG - 7 67 TGGGTTGAATGTTCATAAGCAGTGTTCCAAGATGGTCCCCAATGACTGTA - 817 AGCCAGACCTGAAGCACGTGAAGAAGGTGTATAGCTGTGACCTGACAACA - 8 67 CTCGTGAAAGCTCACATCACCAAGAGGCCCATGGTGGTAGACATGTGCAT - 917 CAGGGAGATCGAGTCCAGAGGTCTGAATTCTGAAGGACTCTACCGAGTGT - 967 CGGGATTTAGTGACCTAATTGAAGACGTCAAGATGGCTTTTGATAGAGAC - 1017 GGTGAAAAGGCGGATATTTCTGTGAACATGTATGAGGACATCAACATTAT - 1067 CACTGGAGCGCTTAAACTGTACTTCAGGGATCTGCCAATTCCTCTCATCA - 1117 CCTACGATGCCTACCCCAAGTTCATAGAGTCTGCCAAAATTGTGGACCCT - 1167 GATGAGCAACTGGAGACCCTTCATGAAGCACTGAGGTCTCTGCCGCCCGC - 1217 CCACTGCGAGACGCTCCGGTACCTCATGGCACACCTCAAGAGAGTGACCC - 12 67 TTCACGAGAAGGAGAATCTGATGAGTGCAGAGAACCTTGGGATCGTGTTT - 1317 GGTCCAACCCTCATGAGATCACCAGAGCTAGACCCCATGGCCGCCCTGAA - 1367 CGACATACGCTATCAGAGACTGGTGGTGGAGCTGCTTATCAAAAATGAAG - 1417 ACATTTTATTTTAGagttttgacttgaggagaaaaagaaatggtttacag - 1467 Atgaaggaatgttttctagtaatttaattagcttcattagctgaattgtt - 1517 tcttggttagaggtttgg

Appendix 1: The rat al-chimaerin DNA sequence used in the HA- and GFP- tagged DNA constructs The full DNA sequence of al-chimaerin used in HA- and GFP-tagged DNA constructs is shown (417-1534bp). The DNA sequence encoding the potential N terminal amphipathic helix region is shaded in green, encoding the cysteine rich domain (CRD) is marked in blue text and encoding the GTPase activating (GAP) domain is shaded in yellow. The DNA sequence common to both a l- and a2- chimaerin starts from the [rink shaded residue. Non coding DNA sequence is shown in lower case.

Appendix 1 251 GAATTC-ccgcctttacaHTQGCCCTG.ACCCTGTTTGATACAGATGAATA - 43 TAGACCTCCTGTTTGGAAATCTTACTT.ATATCAGCTACAACAGGAAGCCC - 93 CTCATCCTCGAAGAATTACCTGTACTTGCGAGGTGGAAAACAGACCAAAG - 143 TATTATGGAAGAHfr’TTGATGGCATG.ATCTCC^fcAAGCAGCCGACCA “ 193 GCTCTTGATTGTGGCTGAGGGGAGCTACCTCATC^fcGAGAGCCAGCGGC - 243 AGCCAGGGACCTACACTTTGGCTTTAAGATTTGGAAGTCAAACCAGAHi - 293 TTCAGGCTCTACTACGATGGCAAGCAC'TTtGTTGGGGAGAAACGCTTTGA - 343 GTCCATCCACGATCTGGTGACTGATGGCTTGATTACTCTCTATATTGAAA - 393 CCAAGGCAGCAGAATACATTGCCAAGATGACGATAAACCCAATTTATGAG - 443 CACGTAGGATACACAACCTTAAACAGAGAGCCAGCATACAAAAAACATAT - 4 93 GCCAGTCCTGAAAGAGACACATGATGAGAGAGATTCTACAGGCCAGGATG - 54 3 GGGTGTCAGAGAAAAGG|TGACATCACTTGTTAGAAGAGCAACTCTGAAA - 5 93 GAAAACGAGCAAATTCCAAAATATGAAAAGATTCACAATTTCAAGGTGCA - 64 3 TACATTCAGAGGGCCACACTGGTGTGAA lTACTGTGCCAACTTTATGTGGG - 693 GTCTCATTGCTCAGGGAGTGAAATGTGCAGATTGTGGTTTGAATGTTCAT - 743 AAGCAGTGTTCCAAGATGGTCCCAAAT GACTGTAAGCCAGACTTGAAGCA - 7 93 TGTCAAAAAGGTGTACAGCTGTGACCTTACGACGCTCGTGAAAGCACATA - 84 3 CCACTAAGCGGCCAATGGTGGTAGACATGTGCATCAGGGAGATTGAGTCT - 8 93 AGAGGTCTTAATTCTGAAGGACTATACCGAGTATCAGGATTTAGTGACCT - 94 3 AATTGAAGATGTCAAGATGGCTTTCGACAGAGATGGTGAGAAGGCAGATA - 993 TTTCTGTGAACATGTATGAAGATATCA_ACATTATCACTGGTGCACTTAAA - 104 3 CTGTACTTCAGGGATTTGCCAATTCCACTCATTACATATGATGCCTACCC - 1093 TAAGTTTATAGAATCTGCCAAAATTAT GGATCCGGATGAGCAATTGGAAA - 114 3 CCCTTCATGAAGCACTGAAACTACTGCCACCTGCTCACTGCGAAACCCTC - 1193 CGGTACCTCATGGCACATCTAAAGAGAGTGACCCTCCACGAAAAGGAGAA - 1243 TCTTATGAATGCAGAGAACCTTGGAATCGTCTTTGGACCCACCCTTATGA - 1293 GATCTCCAGAACTAGACGCCATGGCTGCATTGAATGATATACGGTATCAG - 134 3 AGACTGGTGGTGGAGCTGCTTATCAAAAACGAAGACATTTTATTTtaaat - 1393 ttttaatttgaggggaaaagaaatgttttacagatgaaggaatgttttat - 1443 agtaatttaatttgctcctgtagctgcattatttcttgattagaggtttg - 1493 ggcatataaccagattaaagtgaaggaactttctgttgtttttgtagcac - 1543 cgctcagctgtcttgtaaaacagtgaacacacgctttctggttctagtaa - 1593 tcctgggtgtttatcacgttcagagaaactcaagctattgcatgattagc - 1643 cccctatctggcaaggaaaccccatacagaagaaacaacaaacctgcgcc - 1693 tgcaccgcctctgcgtcctgggtagtctgtgcttgtaatccagcatgttt - 1743 cacagagtaagcctgttgtgactttgcttttggggtctatgtcattggtt - 1793 tctgatgcttgtacaaacacgcacacacaaatggataaaacagcacctct - 1843 ggctgttacattaccataaaccatatcacatgcctacattttacaaatga - 1893 tttctggtttctcttagttcttctctaacatagtactttctttccagcaa - 1943 aagcaaaatgtgttttcagatttgttactttaataaaggttatccatacc - 1993 aataaaaaaaaaaaaaaaaaaaaaaaa-CTCGAG

Appendix 2: The human a2-chimaerin DNA sequence used in the HA- and GFP- tagged DNA constructs The full DNA sequence of a2-chimaerin used in HA- and GFP-tagged DNA constructs is shown (l-2032bp). The DNA sequence encoding the N terminal SH2 domain is marked in red text and the positions of the SH2 domain mutations are shaded in §ed. the sequence encoding the cysteine rich domain (CRD) is marked in blue text and encoding the GTPase activating (GAP) domain is shaded in yellow. The DNA sequence common to both a l - and a2-chimaerin starts from the pinlq shaded residue. Non coding DNA sequence is shown in lower case. The artificial 5' EcoRI site and 3' Xho I site are shown in bold and underlined

Appendix 2 252 gacgtcGGATTCATGTCTTATCAGGGGAAGAAAAATATTCCACGCATCAC - 216 GAGCGATCGTCTTCTGATCAAAGGTGGCAAGATTGTGAATGATGACCAGT - 2 66 CCTTCTATGCAGACATATACATGGAAGATGGGTTGATCAAGCAAATAGGA - 316 GAAAACCTGATTGTGCCAGGAGGGGTGAAGACCATCGAAGCCCACTCCAG - 366 AATGGTGATCCCTGGAGGAATTGACGTGCACACTCGCTTCCAGATGCCAG - 416 ACCAGGGAATGACATCAGCTGATGACTTCTTCCAGGGAACCAAGGCAGCC - 4 66 CTGGCCGGAGGAACCACCATGATCATCGACCATGTTGTTCCTGAGCCCGG - 516 GACAAGCCTATTGGCAGCCTTTGATCAGTGGAGGGAGTGGGCGGACAGCA - 566 AGTCCTGCTGTGACTATTCGCTGCACGTGGACATCACGGAGTGGCACAAG - 616 GGCATCCAGGAGGAGATGGAAGCTCTGGTGAAGGACCACGGGGTAAACTC - 666 CTTCCTCGTGTACATGGCTTTCAAAGATCGGTTCCAGCTGACGGATTCCC - 716 AGATCTATGAAGTACTGAGCGTGATCCGGGATATTGGTGCCATAGCTCAA - 766 GTCCATGCAGAGAATGGTGACATCATTGCAGAGGAACAGCAGAGGATCCT - 816 GGATCTGGGCATCACAGGCCCCGAGGGACACGTGCTGAGCCGGCCAGAGG - 866 AGGTCGAGGCTGAAGCTGTGAACCGGTCCATCACCATTGCCAATCAGACC - 916 AACTGCCCGCTGTATGTCACCAAGGTGATGAGCAAGAGTGCTGCTGAAGT - 966 CATCGCCCAGGCACGGAAGAAGGGAACTGTGGTGTATGGTGAGCCCATCA - 1016 CTGCCAGCCTGGGGACTGATGGCTCTCATTATTGGAGCAAGAACTGGGCC - 1066 AAGGCCGCTGCCTTTGTCACCTCTCCACCCTTGAGCCCCGACCCAACCAC - 1116 TCCAGACTTTCTCAACTCGTTGCTGTCCTGTGGAGACCTCCAGGTCACTG - 1166 GCAGTGCCCACTGTACCTTCAACACTGCCCAGAAGGCTGTGGGGAAGGAT - 1216 AACTTCACCTTGATTCCAGAGGGCACCAATGGCACTGAGGAGCGGATGTC - 1266 TGTCATTTGGGATAAAGCTGTGGTCACTGGGAAGATGGACGAGAACCAGT - 1316 TTGTGGCTGTGACTAGCACCAACGCAGCCAAAGTCTTCAATCTTTACCCA - 1366 CGGAAAGGTCGTATCTCCGTGGGATCTGACGCAGACCTGGTGATCTGGGA - 1416 CCCTGACAGTGTGAAGACCATCTCTGCCAAGACGCACAACAGTGCTCTTG - 14 66 AGTACAACATCTTTGAAGGCATGGAGTGTCGGGGCTCCCCACTGGTGGTC - 1516 ATCAGCCAGGGCAAGATTGTCCTGGAGGACGGCACGTTGCATGTCACGGA - 1566 AGGCTCAGGACGCTACATTCCCCGGAAGCCCTTCCCTGACTTTGTGTACA - 1616 AACGCATCAAGGCAAGGAGCAGGCTGGCTGAGCTGAGGGGGGTCCCTCGT - 1666 GGCCTGTATGATGGACCCGTATGCGAGGTGTCTGTGACGCCCAAGACGGT - 1716 CACTCCGGCCTCATCAGCTAAGACATCCCCTGCCAAGCAGCAGGCGCCAC - 17 66 CTGTTCGGAACCTGCACCAGTCTGGTTTCAGCTTGTCTGGTGCTCAGATT - 1816 GACGACAACATTCCCCGCCGCACCACCCAGCGCATTGTGGCGCCCCCTGG - 1866 TGGCCGTGCCAACATCACCAGCCTGGGCTAA-GAATTCgacgtc

Appendix 3: The rat TOAD-64 DNA sequence used in the HA- and GFP-tagged DNA constructs The full coding DNA sequence used in HA- and GFP-tagged DNA constructs is shown (residues 178-1896bp). This TOAD-64 sequence was isolated via PCR by C. Monffies. The artificial 5' BamHI site and 3' EcoRI site are shown in bold and underlined. Non coding DNA sequence is shown in lower case. (Genbank accession number of rat TOAD-64 sequence: Z46882).

Appendix 3 253 pGL2-Basic DNA

Description: The pGL2-Basic Vector lacks eukaryotic promot­ Storage Conditions: Store at -20°C. Store bacterial strain at er and enhancer sequences. Insertion of a functional pro­ -70°C. moter upstream of the luciferase gene will produce lumines­ GenBank®/EMBL Accession Number: X65323 cence in extracts of transfected cells. Enhancers may be inserted into this vector creating the potential for further increases in luminescence.

Vector Map Notes: 1. Sequence reference points: Amp' poly(A) signal a. SV40 regions: (tor background reduction) promoter (none) Sms I enhancer (none) Kpn I intron 1968-2033 Sac I 3'-untranslated region 1892-2743 Sal I b. luciferase (luc) coding region 76-1725 c. (J-lactamase (Ampr) coding region 4674-3817 d. f 1 origin 4806-5261 poly(A) signal e. ColE 1-derived plasmid replication origin 3052 (for luc reporter) f. GL primerl binding site 5564-5580 SV40 g. GL primer 2 binding site 78-99

pGL2-Basic Vector map.

Information reproduced from the Promega catalogue

Appendix 4: The structure of the pGL2 basic vector used to generate the luciferase coupled NFkB reporter vectors. The luciferase coupled NFk B reporter vectors were generated by J. M. Dong (IMCB, Singapore) from the pGL2-basic vector. The TK promoter sequence was inserted at the Hind III site and four copies of the functional or mutated NFk B binding site derived from the MHC promoter sequence were inserted at the Xhol site of the multiple cloning site to produce the functional NFk B and mutated NFk B(M) reporter vectors, respectively. The Smal site was deleted in both vectors.

Appendix 4 254 ON On On ON ON ON On ON ON ON ON Os "G a u o ct ON cn t-" ON ON "cf o O n 7 3 ON o in t-* VO C" CN CN 2 o o CN o o O a *► G V m ’ o 05 © V 33 31 ON o G A CN o. CN VO CN vo in u u 'if ON o VO o VO in o ON oo CN CN in cn o o a> cn o CN*■“4 ©

ON O n CN i n ^"■4 ON OO VO oo in vo O cn 1—H VO ■*f r - in O n ON VO CO (N m O o i n ON ©

r*- CN oo cn o o cn O cn in VO vo oo CN VO CO r - in 00 vo o vq CO CN r - O o v o CO © O o

o CN oo ON © t '- OO cn ”'3" oo in oo co CN CN oo VO vo ON CN co CO CN r - cn o O 'i f i n co o © © cn cn cn o cn OO O n r - cn O in CN cq o oo vo O n m vo cn ON CO cn in CN o o ^ r 05 CO H © co © © © o

in o CN in '*3' CN O s O n O cn in o ON 0 0 co VO m OO in in CN Pm ON co in CN o o Tf < 0 1 H O n ^ 3 o oo co o VO cn CN o cn in O n oo CN m oo © oo O n in CN o o cn CN © ©

r - CO cn VO VO co ON oo vo in r~- oo r-" oo T“—H r f ^3" in oo cn CN o O cn ©

O n o vo vo in m CN r-« O s 33 ON r - r-~ V m VO CO oo o O n ON ^3" cn w oo o O cn O o 03 « n o V 2 « 83 •** H-5 toJ S' 83»►» ON vo cn W 3 t» 83 83 CNI fI " S % & I* a CN CN CN CN V a 8 8 3 8 t ; G o «* a zL ZL 3 . ZL 3 . n =L G. ZL zL =L G m in i n in in m in m in i n m tos

Appendix 5 255 Appendix 6: NFkB reporter assay results in HeLa cells stimulated with IL-1B Data used to generate summary tables in Figure 5.3 all samples are 0.5ug NFkB + 0.5ug HAv Overall Standard Samples DATA POINTS Average Deviation n 5% serum +/- IL-1B treatment for 18 hours no additions 0.851 0.953 1.585 1.130 0.397 3 O.Olng/ml 0.148 0.164 0.203 0.171 0.028 3 O.lng/ml 0.316 0.126 0.161 0.201 0.101 3 lng/ml 0.248 1.076 1.260 0.861 0.539 3 1 Ong/ml 0.739 1.254 0.748 0.913 0.295 3 50ng/ml 1.980 0.441 0.483 0.968 0.877 3 75ng/ml 6.540 7.176 3.046 5.587 2.224 3 1 OOng/ml 2.798 3.764 4.720 3.761 0.961 3

No serum +/- IL-1B treatment for 18 hours no additions 1.266 0.729 1.006 1.000 0.269 3 50ng/ml 4.334 4.378 4.910 4.541 0.321 3 75ng/ml 3.040 4.671 3.581 3.764 0.831 3 1 OOng/ml 2.641 3.191 5.699 3.843 1.630 3

5% serum + 75ng/ml IL-1B treatment for 18 hours 0.5pg NFkB + 0.5p,g HAv 6.540 7.176 3.046 5.587 2.224 3 0.5 jig NFkB + 0.5pg HA-al chimaerin 0.268 0.441 0.265 0.325 0.101 3

5% serum + lOOng/ml IL-1B treatment for 18 hours 0.5pg NFkB + 0.5pg HAv 2.798 3.764 4.720 3.761 0.961 3 0.5jng NFkB + 0.5pg HA-al chimaerin 0.267 0.216 0.282 0.255 0.035 3

Appendix 6 256 fl O Os Os Os Os os OS Os Os Os os Os Os OS T3 fl fl ** in CM cn M" CM r " CM OO O VO cn CM T3 fl CM Os CM OO CM m O CM in 1—H o SO fl O O CM CM o cn CM CO CM o o +*fl 6 O CM o O o o o o o o o O o o <*) Q VI ex O VO o so o CM in 00 CM r - Os SO Tf cn 2 in m cn o o 00 CM 00 O Os CM i—H V Os t " CM -H- Os 1—H cn SO VO CM o > O m o o o 5—H o o o 1-H t-H o o O v e r< a ll OO oo r - so in Os CM CM SO so os CM 00 SO o o - Os in r - CM r—H ?H m CM cn cn cn OS in OS so in o CM Os O o r—H o o o r-H o o O Os r - m CM m m o o CM r - cn O n Tf cn m Os in C" Os cn OS CM CM Os O cn CM M" O cn cn os cn oo o 00 O o ^"H o o rH o o o OO m r - cn oo in CM cn so r - cn SO r-' o o CM o m OS in Os CM cn o oo CM M- o cn cn oo CM t"* cn cn O so o O —1 o o o T—i t—H o o Os o Os in so 00 l "i * ( CM OS O so in cn GA SO ■'3- o cn f " in in cn o r^ CM CM H O Os CM M- o cn 00 1—^ r - cn CM o © m o o 1—H r-H o o o i—< o O ZM O r - in oo CM CM Os o rH r - CM cn CM O "3- cn o CM CM r - 1—H Os M" oo VO in CM Ah Os >n CM o o m C" O so CM .“H O 1 < r—t < O in o O © o O o O H cn in cn cn cn M" "S' o CM cn o < o CM o CM cn SO oo Os o so m in CM o Q Os Os CM Os O CM r - o SO o o o ■"g- O O O ?—H o o o ^■4 o o r - Os r - in cn cn l-H oo in so CM Os SO Os 00 SO M- m Os CM M" CM OO 1—“H cn o CM cn OS so o O o o o •*? O o o o o o o o o = CM cn CM so m o 00 Os so CM Os VO m 00 Os oo m Os r~- 1“^ [■" CM r—H cn u oo cn r - Os CM CM OS in r^- O o o in o cn O o o O o o o o o r—H o o ■*r OS in m cn Os in H oo r - 00 m H «/5 M- r - CM cn CM CM CO OS cn cn o cn r - o r—H cn cn so CM CM 00 r f r - Os o o Z o cn o o O o o O o o o o o o .Sf £ a s • P* Vso U .B»-i .a /—"S > 55 j j £ 3 a Tf 1 05 05 £ cn VOi CM 60SO r - s : A flti 1 1 il P4 a I- CM r - o o 6 N— S, cn 5 >i V B I—H T—H CM CM CM CM CM O B 8 8 8 8 8 8 PQ t; S >• > z H o SO 00 00 6 V ZL ZL k.V •** 3 £ m m eg 00 00 00 00 00 00 oo 00 00 00 00 00 o o VL. ZL fl. ZL ZL fl. ZL ZL zL ZL zL fl. zL 5 fl in in in in in in m m m m in m + + V o o o o o o o o o o o o z 04 + + + + + + + + + + + + | | PQ PQ PQ PQ PQ PQ PQ PQPQ PP PQ PQ PQ PQ #{j 12 12 12 12 12 12 12 12 12 12 12 12 12 12 on *3 V o v fl oo 00 00 00 00 00 00 00 00 00 00 00 00 00 6 fl. zL fl. fl. zL zL ZL ZL zL zL ZL zL ZL ZL & sel in m in m m m m m in in m in in in <» o o O o O o O o o o o o o o

Appendix 7 257 so so vo cn vo vo VO vo vo vo so VO fl •E SO oo CN CN oo CN cn © VO T5 a CN co co r-~ vo CN o vo Os fl •p« O O o CN CN CN o fl ► © -4-4 V C / 3 © V S3 61 f l OO CN VO CN CN co os o os 2 w< SO cn VO r- VO t"- VO r^ V Os so OS vo CN Os OO r - o © © © t " ' oo OO CO o cn o Os OS C"- 1—H CO vo oo r - H vo CN OS r- CN o r - CN o O o co VO CN vo O s O n r - OS oo VO VO o vo OS o OS OS r - o r - TT o OS o

C/31 H cn VO VO oo o cn SO oo Os vo CN OS o •*f o s VO Ov vo CN o O s Os O Tf o o vo os CN o T|- CN < t- - 00 r ■ H oo CO r - O H Os vo vo Os o Os oo 00 < © © Q vo oo ON "©" cn vo O tj- vo cn io vo u vo O CN o CN u Os Tj- oo o Os r - VO If) o SO O Os cn CN os oo CN VO VO cn CN cn o Tf VO r H r —H ▼“H VO W If) Os N" N- r - CO © vo VO CN 4) Im © o ’ s 61 fa #fl fl GA jw 4) I* 2 ;► •4^fl hJ X fl OS cn GA '3- flGA fl pJ i 2 CN r - © © © V a CN CN CN CN t ; a > 8 8 8 8 s o GA fl 4> V -4-* 2 00 oo 00 00 oo 00 00 00 oo 00 4) =1 f l . f l . f l . f l . f l . f l . f l . f l . f l . 9 fl VO vo vo VO VO VO VO vo VO vo 4) Z 61 + + O x -+* k - f l 9 •pfl 4) t s GA fl 9 4) 00 00 00 00 00 oo 00 00 00 00 00 00 o a fl. f l . f l . f l . f l . f l . f l . f l . f l . f l . f l . =1 & fl a vo vo VO VO VO vo vo VO vo vo VO vo < Q Cfl

Appendix 8 258 Appendix 9: NFkB reporter assay results in HeLa cells Data used to generate summary table in Figure 5.8 C/5 •o 73 o "fl fl fa. fl fl > fa. • PF Q •P. < 5 / C 2 5 H H < i f C/5 ’a JH > fl O fl 01 V ► V 2 VI a a B sc VO oN s® IT) fl sc V 2 a o 00 m VO cn o © o o cn in f p VO - r © Os Os OO Os N" © © 3 © © Os en Ti­ © Os ^■4 in in + 00 ZL 00 ZL —• © - r vo © cn SO (N cn © r 4 r in © m CN © © m © r“H in m -n- CM - r © ©‘ $ © CN 1 © CN in 00 © 4*4 © oo in m > + 00 zL 00 i Tl- cn © CN © VO © Tf 00 ST) © Tf Os © vo 00 - r 00 © vo vo - r © s 1 © $ © - r CN CN Os in in © CN vo © © + 00 00 t z ZL 1 © N- cn vo VO vo cn CN © © N" © t Os 00 © f r in in cn cn © N" cn N- 73 © © CN U Tf CN cn - r 00 ©* in $ in > © cn Os m © + 00 ZL 00 0 =1 T 1 cn © cn © © oo © cn oo 1—. © cn © vo © © oo vo cn 73 © © in $ © in t CN CN % r-~ U © Os Tf* © N" m m + 00 O ZL 00 ZL J- © © 00 CN © CN Ov 4“4 © cn OO © cn oo © CN cn ST) © cn CN Os i © © CN © $ 0 m £ 4-4 © m > O 1—HOs m CN - r + 00 ZL 00 1 Appendix 9 Appendix VO © © 00 © m o © CN - r o Os t Tf ’ o 4**H OS in © in cn 00 © c> £ F—1© 1 © © m © © en e'­ in + 00 ZL zL 00 J- 1 Ovo VO © VO CN 00 CN VO © vd © oo cn VO 1—4 - r vO VC) - r CN cn m vo © cn $ © in 00 VO Tf Tf © 00 © in m in + 00 zL 00 . 3 £ C/5 2 a V V OV VO VO VO ©’ C- VO © © © r-« O N" © CN oo © i> - r CN © $ © © - r © ITS in © OO Os O + 00 . a 00 zL cn f 7 © © © m © CN CN vn © cn © vn © VT) VO VO 1 © o N" in 00 N" CN © in > *-4 CN © ■3 o m + 00 ZL 00 a i ©’ CN in © cn vo cn m © 00 00 VO - n oo r-H © CN 00 © © © VO © OS OS © © in !? r^ 1 © in - r m + 00 a 00 ZL oV Ovo VO VO vo © OS CN f 7 t 7 cn ©‘ © m o © VO CN vo © in © in m © 73 V f r CN © © © © © cn © © m > r-. CN m $ & + 00 O zL 00 ZL o cn m Os © © cn cn 00 CN © 7i- cn - r Os r-4 00 © © 73 © u N" CN © N" © OO o © t T—H m & - r © $ m m + 00 O -H ZL ZL 00 © vo cn cn © m © vo in © Th CN o m N* © ■4- CN CN © cn - r 00 l © & 4*4 © © cn © © CN OS m m > © m ‘S + ZL 00 00 zL © © cn VO © cn Os vo © in CN cn os CN in © cn os cn © i © p—iOs © vo © cn © cn VO © m Z © m $ + ZL 00 ZL 00 vo CN m vd VO Os m oo cn cn vo vd oo © in cn in cn © m Os CN 3 © f r OS Os in © in in SO oo + zL 00 . 3 00 cn cn cn cn cn cn cn cn cn cn cn

-o fa. © O s r- 7f O s O s Os fl oo © © CN © © © *0 CN © © © © © © © fl © © © © fl © © © © C / 5 VI W © cn m Os m r- os 00 fl © CN Tt SO cn CN cn CN fa. 2 m rm ■< © © © © © © © Os © © © © © © © © O ©

so © © VO CN Os Tt cn © cn m Os Tt cn in en © © © © © © © © © © © © © © © ©

N" © SO oo © CN CN O © CN Tt vo cn cn cn cn 1 1 4 © © © © © © © in © © © © © © © © © W «n © © vo © oo O s « Os ^4" N" cn CN © 55 fa. © © © © © © 9 © © © © © © fl 00

2 h-J ?►» .A O s to cn a fl fl Tt s; W) w GA CN £ & fl t? fa. fl £ •a *3 V a > £ t;o a o. s V GA t>0 00 oo 00 oo oo 00 00 00 oo fa. V fl fl fl fl fl fl fl fl fl fl -** m m in in in m in in m in 2 9 0> fl + + + + + + + + + + £ U or

73 •pfl V 'O GA fl fl 4> V 00 00 00 00 00 00 oo 00 oo oo & ■«-*fl fl fl fl fl fl fl fl fl fl fl a fl a m m in m m m m m m m < Q

Appendix 10