Investigation of the Role of Rasgap in Promoting Neuronal Survival in Drosophila

INVESTIGATION OF THE ROLE OF RASGAP IN PROMOTING NEURONAL SURVIVAL IN DROSOPHILA

A thesis submitted to The University of Manchester for the degree of Doctor of Philosophy in the Faculty of Life Sciences

2013

BEHZAD ROWSHANRAVAN

TABLE OF CONTENTS

LIST OF FIGURES ...... 8

LIST OF TABLES ...... 13

LIST OF ABBREVIATIONS ...... 15

ABSTRACT ...... 22

DECLARATION ...... 23

NOTICE OF PUBLICATION ...... 24

ACKNOWLEDGEMENTS ...... 25

KEY WORDS ...... 27

CHAPTER 1: Introduction ...... 28 1.1 Ras signalling in mammals ...... 29 1.1.1 Ras protein in mammals ...... 29 1.1.2 Ras upstream regulators in mammals ...... 31 1.1.3 Ras downstream effectors in mammals ...... 31 1.1.4 Direct Ras deactivators in mammals ...... 34 1.1.5 Indirect Ras deactivation in mammals ...... 44 1.2 RasGAP mediated signalling in mammals ...... 50 1.2.1 SH2 domain and SH2-mediated signalling ...... 50 1.2.2 RasGAP interacting partners in mammals ...... 51 1.2.3 Biological significance of RasGAP in mammals ...... 52 1.3 Ras signalling in Drosophila ...... 57 1.3.1 Ras protein in Drosophila ...... 57 1.3.2 Ras upstream regulators in Drosophila ...... 57 1.3.3 Ras downstream effectors in Drosophila ...... 61 1.3.4 Direct Ras deactivators in Drosophila ...... 61 1.3.5 Indirect Ras deactivation in Drosophila ...... 62 1.4 RasGAP mediated signalling in Drosophila ...... 63 1.4.1 RasGAP interacting partners in Drosophila ...... 63 1.4.2 Biological significance of RasGAP in Drosophila ...... 63 1.4.3 Modelling neurodegenerative disorders in Drosophila and its implications for humans 66 1.5 Aims and objectives ...... 67

CHAPTER 2: Materials and Methods ...... 68 2.1 Stocks, strains and plasmids ...... 68 2.1.1 Stocks and strains ...... 68 2.1.2 Plasmids ...... 68 2.2 Media ...... 74 2.2.1 Sterilisation ...... 74 2.2.2 Bacterial media ...... 74 2.2.3 Drosophila media ...... 74 2.2.4 Mammalian cell media ...... 74 2.3 Standard solutions and buffers...... 75 2.3.1 100x ampicillin ...... 75 2.3.2 300x Dynasore ...... 75 2.3.3 6x Gel loading buffer (Agarose gels) ...... 75 2.3.4 Phenylmethylsulfonyl fluoride (PMSF) 100 mM ...... 75 2.3.5 Minimal S2 cell lysis buffer (Active) ...... 75 2.3.6 S2 cell lysis buffer (Active) ...... 75 2.3.7 S2 cell lysis buffer (EDTA/DTT-free Active) ...... 75 2.3.8 S2 cell lysis buffer (Pre-made) ...... 75 2.3.9 S2 cell lysis buffer (EDTA/DTT free Pre-made) ...... 76 2.3.10 Minimal S2 cell lysis buffer (Pre-made) ...... 76 2.3.11 S2 cell lysis buffer (Washing) ...... 76 2.3.12 STE ultrasonication buffer ...... 76 2.3.13 50x Tris-Acetate-EDTA (TAE) (1 L) ...... 76 2.3.14 Tris-Buffered Saline (TBS) (pH 7.5) ...... 76 2.3.15 TBS-Tween 20 (TBST) ...... 76 2.3.16 Phosphate buffered saline-Triton X-100 (PBS-T) ...... 76 2.3.17 PBS-BT ...... 76 2.3.18 Tris-EDTA (TE) buffer (pH 7.5) ...... 76 2.3.19 5x Tris-glycine electrophoresis buffer ...... 76 2.3.20 1000x Pervanadate ...... 76 2.3.21 Antifade mounting solution ...... 77 2.3.22 100x poly-L-lysine ...... 77 2.3.23 10000x human epidermal growth factor (hEGF) ...... 77 2.4 Experimental methods ...... 78 2.4.1 General techniques ...... 78 2.4.2 Agarose gel electrophoresis ...... 78 2.4.3 DNA sequence determination ...... 78 3

2.4.4 E. coli master-streaking and culturing...... 79 2.4.5 GST-fusion protein expression and purification in Escherichia coli ...... 79 2.4.6 Polymerase Chain Reaction (PCR) ...... 79 2.4.7 Drosophila methods ...... 80 2.4.8 Drosophila brain sectioning and staining ...... 80 2.4.9 S2 cell methods ...... 81 2.4.10 Mammalian cell methods ...... 82 2.4.11 Drosophila and mammalian cell EGF induction ...... 82 2.4.12 Immunohistochemistry (IHC) ...... 82 2.4.13 Snapshot and deconvolution microscopy ...... 83 2.4.14 RNAi knockdown ...... 84 2.4.15 Endocytosis assay ...... 84 2.4.16 Puncta quantification and determining co-localisation ...... 85 2.4.17 SDS-PAGE and western-blotting ...... 85 2.4.18 Phospho-protein enrichment ...... 86 2.4.19 S2 cell expressed fusion protein pulldown and purification ...... 86 2.4.20 Protein mass spectrometry ...... 87 2.4.21 Data deposition ...... 88 2.4.22 Mascot Delta (MD)-score ...... 88 2.4.23 Quantification using spectral counting ...... 89 2.4.24 Hierarchical clustering ...... 89 2.4.25 Interaction network analysis ...... 89 2.4.26 Regular expression database analysis...... 90 2.4.27 In silico proteomics and interactomics ...... 90

CHAPTER 3: Mass spectrometric identification of RasGAP SH2-dependent interacting partners in Drosophila ...... 91 3.1 Introduction ...... 91 3.1.1 Purification of protein complexes ...... 91 3.1.2 Mass spectrometric protein identification ...... 91 3.1.3 Rationale of the study ...... 92 3.2 Results ...... 93 3.2.1 Drosophila RasGAP SH2 domains interact with a number of tyrosine-phosphorylated proteins in S2 cells ...... 93 3.2.2 Mass spectrometric identification of Drosophila RasGAP interacting partners in S2 cells using spectral counting ...... 95

3.2.3 Mass spectrometric identification of Drosophila RasGAP interacting partners in S2 cells using unique number of peptides method ...... 101 3.2.4 Comparing spectral counting with unique number of peptides methodologies ...... 101 3.2.5 Mass spectrometric identification of protein phosphorylation of Drosophila RasGAP and RasGAP interacting partners ...... 102 3.3 Discussion and conclusions ...... 108 3.3.1 Hierarchical clustering can reliably identify protein enrichment ...... 108 3.3.2 Identified RasGAP SH2-dependent interacting partners play a variety of roles within Drosophila...... 109 3.3.3 Drosophila RasGAP and RasGAP interacting proteins post-translational modifications are likely to reflect significant biological functions ...... 110 3.3.4 Limitations of this study ...... 111 3.3.5 Future experiments ...... 112

CHAPTER 4: A novel interaction between Drosophila RasGAP and Sprint ...... 113 4.1 Introduction ...... 113 4.1.1 Introduction to Sprint and its role in Rab5 mediated endocytosis ...... 113 4.2 Results ...... 113 4.2.1 RasGAP interacts with tyrosine-phosphorylated Sprint in an SH2-dependent manner 113 4.2.2 RasGAP associates with PVR and EGFR ...... 121 4.2.3 RasGAP and Sprint independently interact with the cytoplasmic tyrosine kinase Abl . 121 4.2.4 Sprint interacts with dominant negative Rab5 ...... 122 4.3 Discussion and conclusions ...... 126 4.3.1 RasGAP SH2 domains mediate the Sprint-RasGAP interaction ...... 126 4.3.2 Sprint acts as a scaffold protein ...... 126 4.3.3 Limitations of this study ...... 129 4.3.4 Future experiments ...... 129

CHAPTER 5: Investigation of Sprint and RasGAP subcellular localisation ...... 131 5.1 Introduction ...... 131 5.1.1 Introduction to the Rab5 and Rab5-GEFs ...... 131 5.2 Results ...... 131 5.2.1 Sprint localises to dynamin-GTPase dependent cytoplasmic puncta in S2 cells ...... 131 5.2.2 Sprint co-localises with Rab5 in a VPS9 dependent manner ...... 132 5.2.3 Modulators of Sprint puncta ...... 137 5.2.4 Sprint and RasGAP co-localise in punctate structures in S2 cells ...... 137 5.2.5 Sprint and RasGAP co-localise in early stage endocytic vesicles ...... 141 5.3 Discussion and conclusions ...... 144

5.3.1 Sprint co-localises to early stage endocytic vesicles and potentially regulates early stages of RTK endocytosis ...... 144 5.3.2 Limitations of this study ...... 146 5.3.3 Future experiments ...... 146

CHAPTER 6: Effects of Rab5 on RTKs ...... 147 6.1 Introduction ...... 147 6.1.1 Introduction to the Rab5 mediated RTK endocytosis and signalling ...... 147 6.2 Results ...... 148 6.2.1 Characterising wild-type and chimeric PVR and EGFR ...... 148 6.2.2 Rab5 does not regulate wild-type and chimeric PVR and EGFR degradation ...... 152 6.3 Discussion and conclusions ...... 157 6.3.1 Incorrect intracellular trafficking of the chimeric RTKs could explain their insensitivity to Rab5 overexpression ...... 157 6.3.2 Limitations of this study ...... 158 6.3.3 Future experiments ...... 159

CHAPTER 7: The vap mutant neurodegenerative phenotype in Drosophila is linked to early stage endocytosis ...... 161 7.1 Introduction ...... 161 7.1.1 Introduction to autophagy ...... 161 7.1.2 Autophagic neurodegeneration ...... 161 7.1.3 Rationale of the study ...... 162 7.2 Results ...... 162 7.2.1 Mutant Sprint or Rab5 supresses the vap mutant neurodegeneration phenotype in the adult Drosophila brain ...... 162 7.2.2 spri and vap both effect Drosophila survival rate ...... 163 7.3 Discussion and conclusions ...... 165 7.3.1 Drosophila vap mutant autophagic neurodegeneration may be the result of deregulation in early stage endocytosis ...... 165 7.3.2 Limitations of this study ...... 167 7.3.3 Future experiments ...... 168

CHAPTER 8: Discussion and conclusions ...... 170 8.1 General discussion ...... 170 8.1.1 The framework and implications of this study ...... 170 8.2 Outstanding questions and future work ...... 171 8.2.1 What is the nature of Sprint and RasGAP interaction? ...... 171

8.2.2 Are Sprint and RasGAP involved in RTK endocytosis and signalling and what is the significance of their association in these processes? ...... 172 8.2.3 How does Rab5-mediated endocytosis regulate neuronal survival? ...... 173 8.3 Concluding remarks ...... 174

REFERENCES ...... 177

APPENDIX ...... 206 Table S1. Hierarchical clustering of RasGAP interacting proteins (full datasets) ...... 206 Figure S1. Sprint alignment with RIN family proteins ...... 211 Figure S2. VPS9 domain restricts the subcellular localisation of Sprint ...... 212 Figure S3. Characterising Rab5 and Rab7 RFP constructs ...... 213 Figure S4. Conserved Sprint VPS9 residues are responsible for restriction of its subcellular localisation...... 214 Figure S5. Subcellular localisation of wild-type and mutant Sprint and RasGAP ...... 215 Figure S6. Sprint affects RasGAP subcellular distribution ...... 216 Figure S7. Human EGF ligand can activate MAPK in mammalian HEK293 cells ...... 217 Figure S8. Genotyping Drosophila vap spri stocks ...... 218 Figure S9. Drosophila X chromosome gene recombination schematic ...... 219

Word count: 42003

LIST OF FIGURES CHAPTER 1 Page

Figure 1.1 The Ras subfamily 30

Figure 1.2 Ras downstream effectors in mammals 33

Figure 1.3 Ras-RasGAP interaction and the mechanism of Ras-GTP

hydrolysis 36

Figure 1.4 Schematic structures of RasGAP-related proteins in mammals 38

Figure 1.5 Regulators of endocytosis and trafficking routes to endosomes 46

Figure 1.6 Schematic structures of mammalian Rab5-GEFs 48

Figure 1.7 RTK signalling in Drosophila 58

Figure 1.8 Drosophila vap mutant neurodegeneration phenotype and rescue 65

CHAPTER 2

Figure 2.1 The pUAST-RasGAP constructs 72

Figure 2.2 The pUAST-Sprint truncated constructs 72

Figure 2.3 The pUAST-Sprint point mutant constructs 73

CHAPTER 3

Figure 3.1 RasGAP SH2-dependent phosphotyrosine protein interaction and

tyrosine phosphorylation 94

Figure 3.2 RasGAP pulldown SDS-PAGE gel loading 97

Figure 3.3 Hierarchical clustering of RasGAP interacting proteins 98

Figure 3.4 Overlaying of literature and experimental RasGAP interactome 106

Figure 3.5 Comparing unweighted spectral count and unique number of

peptides methodologies for stringency 107

CHAPTER 4

Figure 4.1 RasGAP SH232 region is sufficient and necessary for Sprint

association 116

Figure 4.2 Sprint interacts with RasGAP SH2 domains in a phosphotyrosine-

dependent manner 117

Figure 4.3 A region of Sprint containing 6 tyrosine residues, including the

YXXPXD motif mediates RasGAP association 118

Figure 4.4 RasGAP associates with Sprint SH2 and VPS9 mutants 119

Figure 4.5 Sprint point mutations do not effect its overall tyrosine

phosphorylation 120

Figure 4.6 RasGAP associates with wild-type PVR and EGFR through its

SH2 domains 123

Figure 4.7 Abl interacts with Sprint and RasGAP but does not influence

Sprint-RasGAP interaction 124

Figure 4.8 Sprint interacts with dominant negative Rab5 but not with Rab7

or Rab11 125

Figure 4.9 A model for the role of RasGAP-Sprint in receptor mediated

endocytosis 128

CHAPTER 5

Figure 5.1 Sprint forms endocytic puncta 133

Figure 5.2 Sprint co-localises with Rab5 in a VPS9 dependent manner 134

Figure 5.3 Sprint co-localises with Rab5 through conserved Sprint VPS9

residues 136

Figure 5.4 Modulators of Sprint puncta 138

Figure 5.5 Sprint and RasGAP co-localise in cytoplasmic punctate structures 139

Figure 5.6 Sprint and RasGAP co-localise to early stage endocytic vesicles 142

CHAPTER 6

Figure 6.1 Chimeric PVR and EGFR constructs become tyrosine-

phosphorylated and activated to the same extent 149

Figure 6.2 Wild-type and chimeric PVR and EGFR constructs become tyrosine-

phosphorylated and activated to the same extent 150

Figure 6.3 Determining PVR and EGFR cellular localisation 151

Figure 6.4 Rab5 does not regulate wild-type PVR and EGFR steady-state

levels 153

Figure 6.5 Rab5 does not regulate chimeric PVR and EGFR steady-state

levels 154

Figure 6.6 The level of chimeric PVR and EGFR tyrosine phosphorylation

depends on their level of expression 155

Figure 6.7 The chimeric PVR and EGFR molecules do not get activated by

human EGF 156

CHAPTER 7

Figure 7.1 Mutant Sprint or Rab5 suppresses the vap mutant

neurodegenerative phenotype 164

CHAPTER 8

Figure 8.1 Possible roles of Rab5 in neuronal cell death in RasGAP mutant

Drosophila 175

APPENDIX

Figure S1 Sprint alignment with RIN family proteins 211

Figure S2 VPS9 domain restricts the subcellular localisation of Sprint 212

Figure S3 Characterising Rab5 and Rab7 RFP constructs 213

Figure S4 Conserved Sprint VPS9 residues are responsible for restriction of

its subcellular localisation 214

Figure S5 Subcellular localisation of wild-type and mutant Sprint and

RasGAP 215

Figure S6 Sprint affects RasGAP subcellular distribution 216

Figure S7 Human EGF ligand can activate MAPK in mammalian HEK293

cells 217

Figure S8 Genotyping Drosophila vap spri stocks 218

Figure S9 Drosophila X chromosome gene recombination schematic 219

LIST OF TABLES CHAPTER 1 Page

Table 1.1 Ras-related GAP types and their function 40

Table 1.2 p120-RasGAP interacting partners 42

Table 1.3 Drosophila RTKs and their mammalian homologues 59

CHAPTER 2

Table 2.1 Drosophila stocks 68

Table 2.2 Plasmids constructed previously in this laboratory 68

Table 2.3 Plasmids constructed by other researchers 69

Table 2.4 Plasmids constructed during this study 70

Table 2.5 Primers 71

Table 2.6 PCR reaction mix 80

Table 2.7 PCR reaction cycles 80

Table 2.8 dsRNA list 84

CHAPTER 3

Table 3.1 RasGAP SH2-dependent interacting proteins containing YXXPXD

consensus sequence 100

Table 3.2 RasGAP SH2-dependent YXXPXD consensus binding sequence

enrichment 100

Table 3.3 RasGAP SH2-dependent interacting partners 103

Table 3.4 Phosphorylation of RasGAP and RasGAP SH2-dependent interacting

proteins 105

APPENDIX

Table S1. Hierarchical clustering of RasGAP interacting proteins

(full datasets) 206

LIST OF ABBREVIATIONS

Abl Abelson tyrosine kinase

AF6 Afadin 6

ALS2 amyotrophic lateral sclerosis 2

ALS2CL ALS2 C-terminal like

ANT1 adenine nucleotide translocator 1

AP2 adaptor protein-2

APP amyloid precursor protein

Atg autophagy related

BACE β-site APP-cleaving enzyme

Boss bride of sevenless

Btl breathless

BSA bovine serum albumin

CA constitutively active

CalC/C2 calcium dependent phospholipid binding

CBP calmodulin-binding protein

CM-AVM capillary malformation-arterivenous malformation

CME clathrin-mediated endocytosis

CSW corkscrew

DAPI 4',6-diamidino-2-phenylindole

DMEM Dulbecco’s modified Eagles medium

DMSO dimethyl sulfoxide

DN dominant negative

DNA deoxyribonucleic acid dpERK diphospho ERK

Dras Drosophila ras

Dscam Down syndrome cell adhesion molecule dsRNA double stranded RNA

DTT dithiothreitol

East enhanced adult sensory threshold

EDTA ethylenediaminetetraacetic acid

EEA1 early endosome antigen-1

EGF epidermal growth factor

EGFR EGF receptor

ER endoplasmic reticulum

ERK extracellular signal-regulated kinases

ELAV embryonic lethal abnormal visual system

ESCRT endosomal soring complex required for transport

ESI electrospray ion

F1 first filial

FAK focal adhesion kinase

FBS foetal bovine serum

FITC fluorescein isothiocyanate

FYVE Fab1, YOTB, Vac1, EEA1

GAP GTPase activating protein

GAPVD1 GTPase-activating protein and VPS9 domain-containing protein 1

GDP guanosine diphosphate

GDS guanine nucleotide dissociation stimulator

GEF guanine nucleotide exchange factor

GF growth factor

GFP green fluorescent protein

GPCR G-protein coupled receptor

GPI-APs glycosylphosphatidylinositol-anchored proteins

Grb2 growth factor receptor-bound protein 2

GST glutathione S-transferases

GTP guanosine triphosphate

GVBD germinal vesicle breakdown

HEK293 human embryonic kidney 293

H-Ras Harvey-Ras

HRP horseradish peroxidase

HRS hepatocyte growth factor regulated tyrosine kinase substrate

HSV-2 herpes simplex virus type 2

Htl heartless

IHC immunohistochemistry

IPTG isopropyl -D-1-thiogalactopyranoside

InR insulin receptor

IP4 inositol (1,3,4,5)-tetrakisphosphate

JNK Jun kinase

K-Ras Kirsten-Ras

KTWS Klippel-Trenaunay-Weber syndrome

LAP localisation and affinity purification

LB Luria-Bertani/Liquid broth

LC liquid chromatography

LEC lymphatic endothelial cells

LV low viscosity

MAPK mitogen activated protein kinase

MEK MAPK/ERK kinase

MIAPE minimal information about a proteomics experiment miR microRNA

MS mass spectrometer/spectrometry/spectrometric mTOR mammalian target of rampamycin

MVBs multivesicular bodies

NCE non-clathrin endocytosis

NF1 neurofibromin 1

NMDA N-methyl-D-aspartate

NMR nuclear magnetic resonance

N-Ras neuroblastoma-Ras

PBS phosphate-buffered saline

PBS-BT PBS-BSA Triton X-100

PBS-T PBS-Triton-X-100

PCR polymerase chain reaction

PDGFR platelet derived growth factor receptor

PDK1 phosphatidylinositol dependent kinase 1

PH pleckstrin homology

PI phosphoinositide

PI3K phosphoinositide 3-kinase

PIP2 phosphatidylinositol (4,5)-bisphosphate

PIP3 phosphatidylinositol (3,4,5)-trisphosphate

PKB protein kinase B

PKC protein kinase C

PLC phospholipase C

PMSF phenylmethylsulfonyl fluoride

PRD proline rich domain

PRIDE proteomics identifications database

PTB phosphotyrosine binding

PtdIns phosphatidylinositol

PTK protein tyrosine kinase pTyr phosphotyrosine

PVDF polyvinylidene difluoride

PVR PDGF- and VEGF-receptor related

RA Ras association

RABEX-5 Rabaptin-5-associated exchange factor for Rab5

RAP6 Rab5-activating protein 6

RBD Ras binding domain

RFP red fluorescent protein

RH RIN homology

RIN Ras interaction/interference protein/Ras and Rab5 interactor

Rinl RIN-like

RNA ribonucleic acid

RNAi RNA interference

RO reverse osmosis

RP-LC reversed-phase liquid chromatography

RT room temperature

RTK receptor tyrosine kinase

S2 Schneider 2

SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis

SEM standard error of the mean

Sev sevenless

SH Src homology

SM skimmed milk

SOS son-of-sevenless

Sprint SH2, poly-proline containing Ras interactor

STAM signal-transducing adaptor molecule

SWS Struge-Webber syndrome

TAE Tris-acetate-EDTA

TAP tandem affinity purification

TBC Tre-2/Bud2/Cdc16

TBS Tris-buffered saline

TBST TBS-Tween 20

TE Tris-EDTA

TGF-β transforming growth factor β

TGN Trans-Golgi network

TIAM T-cell lymphoma invasion and metastasis 1

TOR target of rapamycin

Tor torso

UAS upstream activation sequence

UIM ubiquitin-interacting motif

VARP VPS9-ankyrin-repeat protein

VEGFR vascular endothelial growth factor receptor vap vacuolar peduncle

VPS vacuolar protein sorting

XML extensible markup language

YFP yellow fluorescent protein

ABSTRACT

RasGAP is a GTPase activating protein (GAP) that deactivates Ras by promoting Ras-GTP hydrolysis to Ras-GDP. In Drosophila melanogaster, RasGAP is required for the long- term survival of neurons in the adult brain because mutants in the RasGAP gene (vap) show an age-related neurodegenerative phenotype, with dying neurons showing morphological features of autophagy. RasGAP was shown to have a GAP-independent role within fly neurons that is dependent on its SH2 domains. The aim of this study was to identify proteins that interact with the SH2 domains of RasGAP and to understand the roles of these proteins in neuronal survival. By using tagged RasGAP affinity purification and mass spectrometry of RasGAP protein complexes from S2 cells, Sprint, a Ras effector and putative activator of the endocytic GTPase Rab5, was identified as a novel SH2-dependent RasGAP interacting protein. The interaction between Sprint and RasGAP is phosphotyrosine-dependent, since it requires tyrosine phosphorylation of Sprint. In addition, Sprint and RasGAP interaction requires the SH2 domains of RasGAP but not Sprint or the conserved site of RasGAP tyrosine phosphorylation (pTyr363), indicating an association between these two molecules. RasGAP and Sprint co-localised with Rab5- positive early endosomes and this co-localisation depended on the SH2 domains of both RasGAP and Sprint. This study demonstrates a key role for this interaction in neurodegeneration: mutation of Sprint (or Rab5) suppressed the autophagic neuronal cell death caused by the loss of RasGAP. These results indicate that the long-term survival of adult neurons in Drosophila depends on a critical balance between Ras activation and endocytosis, and that this balance is maintained by the interplay between RasGAP and Sprint.

DECLARATION

No portion of the work referred to in the thesis has been submitted in support of an application for another degree or qualification of this or any other university or other institute of learning.

i. The author of this thesis (including any appendices and/or schedules to this thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has given The University of Manchester certain rights to use such Copyright, including for administrative purposes. ii. Copies of this thesis, either in full or in extracts and whether in hard or electronic copy, may be made only in accordance with the Copyright, Designs and Patents Act 1988 (as amended) and regulations issued under it or, where appropriate, in accordance with licensing agreements which the University has from time to time. This page must form part of any such copies made. iii. The ownership of certain Copyright, patents, designs, trade marks and other intellectual property (the “Intellectual Property”) and any reproductions of copyright works in the thesis, for example graphs and tables (“Reproductions”), which may be described in this thesis, may not be owned by the author and may be owned by third parties. Such Intellectual Property and Reproductions cannot and must not be made available for use without the prior written permission of the owner(s) of the relevant Intellectual Property and/or Reproductions. iv. Further information on the conditions under which disclosure, publication and commercialisation of this thesis, the Copyright and any Intellectual Property and/or Reproductions described in it may take place is available in the University IP Policy (see http://www.campus.manchester.ac.uk/medialibrary/policies/intellectual- property.pdf), in any relevant Thesis restriction declarations deposited in the University Library, The University Library’s regulations (see http://www.manchester.ac.uk/library/aboutus/regulations) and in The University’s policy on presentation of Theses

NOTICE OF PUBLICATION

Portions of this work have been submitted for publication in a peer-reviewed journal.

ACKNOWLEDGEMENTS

I would like to thank my supervisor, Dr. David A. Hughes for his undying attention and support throughout my project. I owe many thanks and gratitude to Dr. David Hughes for allowing me to work in his laboratory and helping me in the laboratory and with my research direction and focus. I would also like to thank my advisor Dr. Martin Baron for his constructive advice and objective approach and input throughout my research. I would like to specially thank Dr. Simon Woodcock for his immense technical support and advice throughout my project. In addition, I would like to thank Dr. Jila Ajeian, Dr. Jonathan D. Humphries, Dr. Marian Wilkin, Dr. Hideyuki Shimizu, Dr. Yutaka Matsubayashi, Dr. Andreas Prokop, Dr. Verena Wolfram, Dr. Alengo Nyamay’antu and Dr. Helen Smith for their help, support and invaluable advice they provided throughout my PhD.

I would like to thank Dr. Julian Selley, Dr. Stacey Warwood and Dr. David Knight from the Biological Mass Spectrometry Core Research Facility at the University of Manchester for the bioinformatics and mass spectrometric processing of my work as well as providing invaluable training, help and advice. I would like to thank Dr. Peter March and Dr. Steven Marsden from the Bioimaging Core Facility at the University of Manchester for the bioimaging training, help and support provided during my research. I would also like to thank Anna-maria Knorn and Sanjai Patel from the Fly Facility at the University of Manchester for their kind help, support and training throughout my research. I would like to thank Dr. Samantha Forbes from the Faculty of Life Sciences EM Facility at the University of Manchester for her help and support in processing my fly head tissues. I would like to thank Dr. Pernille Rorth and Dr. Iain Cheeseman for providing us with plasmids.

My current academic status would never have been possible without the selfless efforts of many of my teachers and lecturers that inspired me to strive for excellence in science. Although this list is particularly non-exhaustive, I would like to thank my secondary school science teachers from the Bellemoor School in Southampton, UK, Mr. Tim Roberts and Mr. Howards for realising my potential and passion for science and helping me in my scientific endeavours. I would also like to thank my college science lecturers from the Peter Symonds College in Winchester, UK, Dr. Mike Richards and Ms. Suhad Atrash for helping me develop my scientific interest and encouraging my inquisitive mind. In addition, I would like to thank many people in Brighton University and Sussex University,

UK, including Dr. Elizabeth James, Dr. Stuart James, Dr. Rostislav Vladimirovich Shevchenko, Dr. Iain Allan, Dr. Dominic Grima and Dr. Sarah Newbury, for their immense support and great lectureship in science.

Finally I would like to thank Mr. Jamshid Roshanravan and Mrs. Mohtaram Mashhadi, my kind and wonderful parents for funding my PhD program. I would also like to thank my parents and my brother, Mr. Mohammad Roshanravan, for their undying financial, psychological and emotional support that saw me through my research.

KEY WORDS

Tyrosine phosphorylation, Sprint, guanine nucleotide exchange factor, Rab5, Drosophila, RasGAP

CHAPTER 1: Introduction

Cellular signalling is important to life since it provides the means by which cells respond to different stimuli. Cell signalling is usually initiated with a chemical stimulus, which can bind to its appropriate receptor at the plasma membrane or within the cytoplasm or nucleus depending on the nature of chemical signal. Once bound to its appropriate receptor, the chemical signal initiates a signalling cascade, eventually leading to a cell response. There are three families of receptors, defined by the mechanism used to transduce signal binding into a cellular reponse, including channel-linked receptors (also called ligand-gated ion channels) such as neurotransmitter receptors, intracellular receptors such as nuclear receptors and enzyme linked receptors such as G-protein coupled receptors (GPCRs) and protein kinases (Purves et al., 2001). Enzyme linked receptors such as receptor tyrosine kinases (RTKs) are amongst the most well studied receptor types, which are known to initiate many signalling pathways required for cell survival, including the Ras signalling pathway.

The Ras protein is a small GTP-binding protein derived from a precursor that evolved over a billion years ago (Shilo and Weinberg, 1981). Ras can be activated in response to many extracellular signals, which are conducted from outside to inside of the cells by receptors such as RTKs, cytokines, T-cell receptors and GPCRs. Upon binding of the signalling molecule to its receptor, the intracellular portion of these receptors normally becomes phosphorylated on specific tyrosine residues. These phosphotyrosine residues provide a docking site for many proteins to bind to and initiate a variety of intracellular signalling pathways, including Ras signalling. Ras is active when bound to GTP, and the conversion of Ras-GDP to Ras-GTP is catalysed by guanine nucleotide exchange factors (GEFs). Upon GTP binding, Ras undergoes a conformational change allowing it to bind to its downstream effectors, which determine the biological response to the Ras protein activation. The Ras protein, however, has to be deactivated after inducing its appropriate biological response. This is brought about by GTPase-activating proteins (GAPs), which activate the slow intrinsic GTPase activity of Ras, thus hydrolysing the gamma phosphate of GTP and returning Ras to an inactive GDP-bound state. Although Ras signalling is known to promote cell survival, there is, however, emerging evidence that suggests excessive Ras signalling through absence of its deactivator, RasGAP, can lead to neuronal cell death, as evident in organisms such as Drosophila melanogaster.

1.1 Ras signalling in mammals

1.1.1 Ras protein in mammals The Ras proto-oncogenes encode a subfamily of the small GTP-binding protein superfamily, that is evolutionarily conserved amongst eukaryotes (Lowy and Willumsen, 1993; Shilo and Weinberg, 1981). This subfamily is primarily involved in the regulation of intracellular signal transduction pathways downstream of cell surface receptors. A large number of Ras family members have been discovered to date (Figure 1.1), including cellular homologues of the oncoproteins found in Harvey sarcoma virus (H-Ras) and Kirsten sarcoma virus (K-Ras). Together with Neuroblastoma Ras (N-Ras), these are termed the classical Ras proteins (Karnoub and Weinberg, 2008). In vitro studies revealed that point mutations in the classical Ras oncogenes were responsible for causing cellular proliferation and transformation. The involvement of H-, K- and N-Ras in oncogenic processes sparked enormous interest as to how these Ras isoforms regulate cellular activities (Karnoub and Weinberg, 2008; Takai et al., 2001).

The classical Ras protein family bind to both GDP and GTP but are active only when present in the Ras-GTP form. Crystallographic and NMR analyses revealed that the Ras family GDP/GTP binding domains have a common structure. Binding of GTP to Ras causes a marked change in the conformation of the protein in two regions called switch I (amino acid residues 30-40) and switch II (amino acid residues 60-70). The switch I region is re-orientated through the Thr35 side chain, allowing its interaction with the GTP γ- phosphate and Mg2+ ion. The switch II region is re-orientated through the Gly60, allowing its interaction with the GTP γ-phosphate. The GTP γ-phosphate is in turn stabilised by its interaction with Lys16, Tyr32, Thr35, Gly60 and Gln61 (Figure 1.3). These residues subsequently mediate Ras interactions with its regulators and effectors (Karnoub and Weinberg, 2008; Scheffzek et al., 1997; Takai et al., 2001). The Ras isoforms have a high degree of sequence homology with the major differences between Ras isoforms being found in the hyper-variable region at the C-terminus. At the C-terminal end, Ras proteins are prenylated on cysteine residues, processed to remove the final three amino acids and the C-terminal cysteine is carboxylmethylated. In addition, most Ras proteins also get palmitoylated near the C-terminus. The C-terminal modifications of the Ras proteins ultimately anchor them in the membrane, which is thought to be the pre-requisite to their biochemical activation (Karnoub and Weinberg, 2008; Rajalingam et al., 2007).

Figure 1.1 The Ras subfamily

Figure 1.1 The Ras subfamily. This diagram is adapted from Karnoub and Weinberg, (2008) and shows the Ras subfamily (Karnoub and Weinberg, 2008). The Ras subfamily encompasses 36 genes in mammals, which encode 39 proteins that range in size from 20-29 kDa. Ras family proteins (such as H-, K- and N-Ras) regulate cellular processes including gene expression, growth, cytoskeletal rearrangements, cellular motility and differentiation (Karnoub and Weinberg, 2008).

1.1.2 Ras upstream regulators in mammals Ever since 1984, when Kamata and Feramisco, (1984) reported increased cellular H-Ras activity in response to epidermal growth factor (EGF), a plethora of extracellular Ras activating signals have been discovered (Kamata and Feramisco, 1984). These include signals that activate receptors with intrinsic tyrosine kinase activity (receptor tyrosine kinases, RTKs) or receptors with associated tyrosine kinase activity such as T-cell receptors (Karnoub and Weinberg, 2008). Other receptors not associated with tyrosine kinases such as G-protein-coupled receptors (GPCRs) also activate Ras proteins (Takai et al., 2001). The mechanisms leading to Ras activation are best understood for growth factors that activate RTKs. RTKs with intrinsic phosphorylation provide phosphotyrosine residues, that serve as docking sites for adaptor proteins such as Grb2 or the Shc/Grb2 complex. In non-stimulated cells, Grb2 is bound to the Son-of-Sevenless (SOS) protein. Upon RTK activation, the Grb2-SOS (or Shc-Grb2-SOS) complex is recruited to specific phosphotyrosine residues on RTKs through the SH2 domain of the Grb2 (or Shc) protein. SOS in turn activates plasma membrane-bound Ras by converting inactive Ras-GDP to active Ras-GTP, which makes SOS a guanine nucleotide exchange protein/factor (GEP/GEF). The active Ras-GTP in turn interacts with its downstream effectors (Dance et al., 2008; Kim et al., 2005; Rajalingam et al., 2007; Takai et al., 2001).

1.1.3 Ras downstream effectors in mammals Ras effectors are defined as proteins with high affinity to Ras-GTP, which fail to bind to Ras by mutation within the core effector domain, the switch I effector loop region (Rajalingam et al., 2007). Ras-GTP has a diverse range of effector molecules (Figure 1.2), which take part in different cellular activities. In 1993, c-Raf kinase was the first reported Ras effector, with homologues in Drosophila (pole hole) and Caenorhabditis elegans (lin- 45) (Rajalingam et al., 2007). Soon, the Ras-Raf-MEK-ERK (MAPK) pathway was clarified and the roles of MAPK in gene expression, dynamic membrane processes and cytoskeletal rearrangements were identified (Rajalingam et al., 2007). The discovery of Raf as a Ras effector, was followed by Ral guanine nucleotide dissociation stimulator (Ral- GDS), which is part of the Ral-GEF family. Ral-GEF has a crucial role in human cell transformation and anchorage-independent cell growth in vitro (Rajalingam et al., 2007). The discovery of Ral-GEF as a Ras effector, was soon followed by phosphatidylinositol 3- kinase (PI3K). Upon PI3K activation, phosphatidylinositol (3,4,5)-trisphosphate (PIP3) is generated, which acts as a second messenger and recruits phosphatidylinositol dependent

31 kinase 1 (PDK1) and PKB/AKT to the plasma membrane. PDK1 in turn activates AKT isoforms by phosphorylation. AKT1 is associated with cell proliferation and survival, while AKT2 is associated with insulin mediated metabolic processes (Rajalingam et al., 2007). In addition to the aforementioned Ras interacting proteins, RIN1, TIAM1, AF6, RASSF5/NORE1, PLCε and PLCζ are amongst the other characterised Ras interacting proteins, which mediate the diverse cellular function of Ras upon its activation (Figure 1.2). Each Ras effector has a Ras binding domain (RBD), that can be divided into three categories based on structural features: Raf-type, PI3K-type and Ral-GEF-type RBDs. TIAM1, for example has a Raf-type RBD, but RIN1, AF6, NORE1 and PLCε have Ral- GEF-type RBDs, which are termed the Ras association (RA) domain (Rodriguez-Viciana et al., 2004).

Figure 1.2 Ras downstream effectors in mammals. This diagram is adapted from Karnoub and Weinberg (2008). This diagram shows the mammalian Ras downstream effectors and their wide variety of roles in cellular functions. Upon the appropriate receptor tyrosine kinase (RTK) stimulation by growth factors (GF) and signalling, the Ras protein is converted from the inactive to active state, utilising the guanine nucleotide exchange factors (Ras-GEF) such as SOS. The active Ras-GTP in turn interacts with many of its downstream effectors such as Raf or AF6 and induces the appropriate biological response, as shown in the above diagram. However, once Ras-GTP mediates its effects through downstream effectors, it has to be deactivated either directly or indirectly. Direct Ras deactivation is achieved through deactivators of the Ras protein, GTPase activating protein molecules such as Ras-GTPase activating protein (RasGAP) or neurofibromin 1 (NF1).

1.1.4 Direct Ras deactivators in mammals Once Ras-GTP mediates its effects through downstream effectors, it has to be deactivated either directly or indirectly. Direct Ras deactivation is achieved through GTPase activating protein molecules such as the Ras-GTPase activating protein (RasGAP) (Rajalingam et al., 2007; Takai et al., 2001). In 1987, Trahey and McCormick, (1987) reported that Asp12 and Val12 mutants of N-Ras protein can induce Xenopus laevis oocyte maturation. It was discovered that active Ras-GTP is responsible for this maturation phenotype (as determined by germinal vesicle breakdown or GVBD) since Asp12 and Val12 mutant Ras proteins were predominantly associated with GTP. In addition, unlike Ras-GDP, a high percentage of GVBD could be achieved at very low Ras-GTP concentrations. Although mutant and wild-type N-Ras proteins showed a very similar GTPase activity in vitro they showed a strikingly different GTPase activity in vivo. When GTP-bound Ras proteins were microinjected in the Xenopus laevis oocyte, after a few minutes wild-type Gly12 Ras was mainly GDP bound whereas Asp12 and Val12 Ras were mainly GTP bound in vivo. This discovery not only confirmed that Ras-GTP was responsible for the GVBD but also indicated that there is a cytoplasmic factor that can stimulate the conversion of Ras-bound GTP to GDP by more than 200-fold, deactivating the biological activity of the Gly12 Ras in the process (Trahey and McCormick, 1987). Further experiments indicated that this cytoplasmic factor is indeed a protein and it was named GTPase activating protein (RasGAP). Soon after, RasGAP protein was reported as also being membrane bound and its presence was confirmed in other cell systems (Hoshino et al., 1988; Molloy et al., 1989). This was followed by cloning and purifying RasGAP from bovine and human sources (Gibbs et al., 1988; Trahey et al., 1988; Vogel et al., 1988). It was then noticed that different human tissues harbour distinct RasGAPs with predicted molecular mass (Mr) of 116-120 kDa (later named type I RasGAP; detected in human lung, brain, liver, leukocytes and placenta) and 95-100 kDa (later named type II RasGAP; detected in human placenta) (Halenbeck et al., 1990; Trahey et al., 1988). Although both RasGAPs have similar catalytic properties, only the type I RasGAP is thought to act as a negative regulator of the wild-type Ras (al-Alawi et al., 1993; Mollat et al., 1994; Mollat et al., 1992; Zhang et al., 1993b).

Soon after its discovery, RasGAP was shown to interact with the H-Ras and N-Ras effector binding domains (Adari et al., 1988; Cales et al., 1988). Mutational studies of the H-Ras protein revealed that mutations in the non-essential regions of H-Ras as well as carboxy-

34 terminal domain (amino acid residues 165-185) and purine binding regions (amino acid residues 117 and 119) did not affect the Ras responsiveness to RasGAP. However, mutations in the phosphoryl binding region (amino acid residues 12, 59 and 61) of H-Ras affected the RasGAP mediated GTPase activity. In addition, mutations in the effector region of H-Ras (amino acid residues 35, 36 and 38, also referred to as the switch I region) affected RasGAP mediated GTPase activity (Adari et al., 1988; Cales et al., 1988). Later studies revealed that mutations in Ras residues in loop L1 (Gly12 and Gly13), loop L2 (Thr35 and Asp38) and loop L4 (Gln61 and Glu63) influence the RasGAP catalytic contribution to the Ras; however, all the mutants still form complexes with RasGAP (Gideon et al., 1992; Nur and Maruta, 1992). Soon after, it was discovered that the Ras Pro34 residue is crucial for Ras-RasGAP binding whereas the crucial residue for Ras binding to its effectors such as Raf is Asp38, which are located within the Ras effector binding domain (Abdellatif and Schneider, 1997; Marshall, 1993; Stone et al., 1993; Warne et al., 1993; Zhang et al., 1993a). The Arg786, Lys831 and Arg925 residues of RasGAP were later found to be important for its binding to Ras-GTP (Miao et al., 1996). Although the RasGAP C-terminal domain seemed to be sufficient in stimulating Ras GTPase activity (Marshall et al., 1989), what was not known at the time of the RasGAP discovery was the way in which RasGAP led to the Ras-GTP to Ras-GDP conversion. In addition, it was unclear how the aforementioned Ras protein mutations affected the way in which Ras-GTP was converted to Ras-GDP. The preliminary work of Peter Lowe’s group reported in 1991 and the work of Wittinghofer’s group reported in 1997, addressed the above questions (Scheffzek et al., 1997; Skinner et al., 1991). Using the catalytic fragment of the RasGAP protein, GAP-334, X-ray crystallographic and mutational studies revealed that GAP contributes an arginine finger (Arg789 from the loop L1c) to the Ras-RasGAP complex, allowing Ras-GTP conversion to Ras-GDP (Figure 1.3).

Figure 1.3 Ras-RasGAP interaction and the mechanism of Ras-GTP hydrolysis. This diagram is adapted from Scheffzek et al., (2008) and shows Ras interaction with RasGAP catalytic domain, GAP-334. Using the catalytic fragment of the RasGAP protein, GAP-334, X-ray crystallographic and mutational studies revealed that GAP contributes an arginine finger (Arg789 from the loop L1c) to the Ras-RasGAP complex. This arginine finger leads to the stabilisation of the Ras transition state by neutralising the partial negative charge on the γ-phosphate of GTP. Arg789 also stabilises the Ras-RasGAP complex by communicating with Ras Glu31 and hydrogen bonding with the amide group of the Ras Gln61 residue (crucial for geometric stabilisation of the transition state). This transition state stabilisation leads to the hydrolysis of the GTP by the H-Ras protein. The critical role of the arginine finger in GAP-mediated GTP hydrolysis is demonstrated by using NF1-333 (not shown), which is the catalytic fragment of another type of Ras-associated GAP (NF1). NF1 has a much higher affinity than RasGAP for Ras-GTP, allowing rate constants to be measured more easily. These studies have revealed that mutations of Arg1276 (equivalent to Arg789 in GAP-334) to alanine (R1276A) reduced the rate of GTP hydrolysis by nearly 1800-fold and increased the GAP protein dissociation constant by 3 fold, which means that the mutant GAP protein has a slightly reduced affinity for the Ras protein but has a very large reduction in its catalytic activity. The conservative arginine to lysine substitution (R1276K) in NF1-333 also resulted in 1800-fold decrease in the rate of GTP hydrolysis but had a dissociation constant matching the wild-type GAP demonstrating the essential role of the conserved arginine in catalysis (Ahmadian et al., 1997; Noel, 1997). 36

After RasGAP’s discovery, several other proteins were identified with either a GAP activity for Ras or a protein domain related to the catalytic domain of RasGAP (Table 1.1 and Figure 1.4). The first identified mammalian RasGAP was shown to be a 120 kDa cytoplasmic protein (Trahey et al., 1988) and was dubbed p120-RasGAP. At the N- terminal end, mammalian p120-RasGAP protein has two SH2 domains flanking an SH3 domain, followed by a PH domain and a calcium-dependent phospholipid-binding (CaLB/C2) domain. At the C-terminal end, p120-RasGAP has the catalytic GAP domain (Pamonsinlapatham et al., 2009). These domains bind to different partners (Table 1.2), allowing RasGAP participation in many cellular processes. The p120-RasGAP protein has been shown to be evolutionarily conserved amongst Homo sapiens (encoded by RASA1 gene), Drosophila (encoded by vap gene) and Caenorhabditis elegans (encoded by gap-3 gene) (Feldmann et al., 1999; Jiang and Ramachandran, 2006; Stetak et al., 2008).

Many molecules can modulate RasGAP activity. Some molecules can modulate RasGAP function by stimulating (prostaglandins PGF2α and PGA2) or inhibiting (prostacyclin PGI2, saturated stearic acid and monounsaturated oleic acid) GAP activity of the RasGAP (Ding and Lengyel, 2008; Frech et al., 1990; Golubic et al., 1998; Han et al., 1991; Masuelli and Cutler, 1996; Sermon et al., 1996). Other molecules that can modulate RasGAP activity include herpes simplex virus type 2 (HSV-2) RR1 protein kinase (ICP10), c-Src, p62-Dok, p190-RhoGAP and Nck1. In addition, there is evidence for intramolecular inhibition of GAP activity by RasGAP’s pleckstrin homology (PH) and Src homology (SH) 2 and SH3 domains (Bryant et al., 1996; Drugan et al., 2000; Ger et al., 2011; Giglione et al., 2001; Kashige et al., 2000; Moran et al., 1991; Smith et al., 2000). RasGAP expression levels were also found to be targeted by microRNAs (miRNAs) such as miR-1, miR-31, miR- 132, miR-182 and miR-335, which shows the degree of complexity of the RasGAP regulation within cells (Anand et al., 2010; Jin et al., 2012; Sayed et al., 2007; Sun et al., 2013; Wang and Ruan, 2010; Xu and Wong, 2008; Yang et al., 2012).

p120-RasGAP Gap1m

SH2 SH3 SH2 PH C2 GAP C2 C2 GAP PH BTK

NF1 Gap1IP4BP

GAP SEC14 C2 C2 GAP PH BTK

SynGAP1 CAPRI

PH C2 GAP C2 C2 GAP PH BTK

RASAL2 RASAL1

PH C2 GAP C2 C2 GAP PH BTK

DAB2IP GAPEX-5

PH C2 GAP GAP VPS9

RASAL3 EPI64A

GAP TBC

EPI64B EPI64C

TBC TBC

PLXNA1

Sema PSI PSI PSI TIG TIG TIG TIG TM Plexin_cytopl GAP Plexin_cytopl

PLXNB1

Sema PSI PSI PSI TIG TIG TIG TIG TM Plexin_cytopl

PLXNC1

PSI PSI TIG TIG TM

PLXND1

Sema Sema PSI PSI TIG TIG TIG TIG IncA TM IncA Plexin_cytopl

IQGAP1 IQGAP2

CH WW 4xIQ GAP CH WW 4xIQ GAP

IQGAP3

CH WW 4xIQ GAP

Src-homology 2 (SH2), Src-homology 3 (SH3), Pleckstrin homology (PH), Calcium-dependent phospholipid-binding (C2), GTPase activ at- ing protein (GAP), Calponin-homology (CH), Bruton’ s tyrosine kinase (BTK), Tryptophan repeat motif (WW), Calmodulin-binding mot ifs (IQ), Plexin-semaphorin-integrin (PSI), Anthrax receptor extracellular domain (Anth_Ig), Receptor activity modifying family (RA MP), Plexin cytoplasmic RasGAP domain (Plexin_cytopl), IPT/TIG (TIG), Sec14p-like lipid-binding domain (SEC14), Transmembrane domain (TM), Chlamydia trachomatis inclusion (IncA), Tre-2/Bud2/Cdc16 (TBC).

Figure 1.4 Schematic structures of RasGAP-related proteins in mammals. The p120-RasGAP and NF1 proteins are structurally and functionally similar. The p120-RasGAP protein acts as a GAP for H-Ras, N-Ras, R-Ras, R-Ras2, R-Ras3 and Rab5 (Adari et al., 1988; Bernards, 2003; Graham et al., 1996; Liu and Li, 1998; Ohba et al., 2000; Quilliam et al., 1999; Trahey and McCormick, 1987). Similarly, NF1 acts as a GAP for classical Ras proteins but it also acts as a GAP for R-Ras GTPases such as R-Ras2 (Bernards, 2003; Graham

38 et al., 1996). Gap1m (Maekawa et al., 1994), GAP1IP4BP (Cullen et al., 1995), CAPRI and RASAL1 (Allen et al., 1998) are structurally very similar and are part of the Gap1 family (Bernards, 2003; Kupzig et al., 2006a). Gap1m, GAP1IP4BP and CAPRI act as GAPs for H-Ras and RASAL1 acts as a GAP for N-Ras with GAP1IP4BP also acting as a GAP for R-Ras and R-Ras2 and with RASAL1, GAP1IP4BP and CAPRI also acting as GAPs for Rap1 (Bernards, 2003; Cullen et al., 1995; Kupzig et al., 2009; Kupzig et al., 2006b; Mitin et al., 2005; Sot et al., 2013). Similar to p120-RasGAP (Davis et al., 1996; Gawler et al., 1995), CAPRI and RASAL1 C2 domain has high affinity Ca2+-binding domains, which Gap1m and GAP1IP4BP lack. Unlike CAPRI, other members of the GAP family use their PH domain to interact with the plasma membrane and become activated (Bernards, 2003; Sot et al., 2010). For example GAP1IP4BP uses its PH domain to interact with PIP2 and IP4, which stimulates its RasGAP and RapGAP activity (Kupzig et al., 2006a; Kupzig et al., 2009; Kupzig et al., 2006b; Sot et al., 2010) whereas Gap1m PH domain binds to PIP3 and Gα12, which stimulates Gap1m’s RasGAP activity (Bernards, 2003; Jiang et al., 1998; Kupzig et al., 2006a; Mitin et al., 2005). Although CAPRI is activated in a Ca2+-dependent manner (Bernards, 2003), SynGAP (Kim et al., 1998) is inhibited by the presence of Ca2+ through the action of calmodulin-dependent protein kinase II. SynGAP co-localises with N-methyl-D-aspartate (NMDA) glutamate receptor and stimulates the GTPase activity of H-Ras (in vitro) (Bernards, 2003; Kim et al., 1998). In addition, SynGAP acts as a GAP for Rap (Bernards, 2003; Sot et al., 2010). Humans have four SynGAP related proteins including SynGAP1, RASAL2, AF9Q34 and RASAL3. DAB2IP and PLXN A1-D1 have RasGAP functions with DAB2IP acting as a GAP for H-Ras, K-Ras, R-Ras and R-Ras2 (Mitin et al., 2005) and PLXN A1-D1 acting as GAPs for R- Ras (Basile et al., 2005; Ito et al., 2006; Oinuma et al., 2004; Uesugi et al., 2009). Initially identified as Rab- GAPs, EPI64 A-C have also been shown to exhibit GAP activity for H-, K- and N-Ras by using their Tre- 2/Bud2/Cdc16 (TBC) domains (Itoh and Fukuda, 2006; Nagai et al., 2013). GAPEX-5 is another RasGAP- like domain containing protein, which exhibits GAP activity for H-Ras (Hunker et al., 2006a). Although human IQGAP1 (Weissbach et al., 1994), IQGAP2 (Brill et al., 1996) and IQGAP3 (Wang et al., 2007) contain GAP-related domain, they lack any arginine finger (or GAP activity) and in fact inhibit the GTPase activity of the Rho family members Cdc42 and Rac by directly interacting and stabilising them in their active state (Bernards, 2003; Wang et al., 2007). nGAP is a GAP-related domain containing protein, which can rescue the yeast RasGAP mutant phenotype; however, fails to exhibit any GAP activity towards H-Ras, R- Ras, Rap1 or Ral in vitro (Noto et al., 1998).

Table 1.1 Ras-related GAP types and their function

RasGAP RasGAP RapGAP Drosophila Cellular function(s) and expression Reference(s) name activity activity homologue p120-RasGAP Yes - RasGAP MAPK pathway down regulation (Bernards, 2003) (RASA1) NF1 Yes - NF1 Regulating TSC2 and mTOR (Bernards, 2003; Johannessen et al., 2005) Gap1m Yes - Gap1 Interaction with PIP3, PI3K-dependent membrane (Bernards, 2003; Jiang et al., 1998) (RASA2) translocation and Gα12 binding

GAP1IP4BP/R- Yes Yes Gap1 Mediating Gαi-induced inhibition of MAPK (Bernards, 2003; Kupzig et al., 2006b; RasGAP Mitin et al., 2005; Nafisi et al., 2008; (RASA3) Sot et al., 2010) RASAL1 Yes Yes Gap1 Highly expressed in follicular cells of thyroid and the (Allen et al., 1998; Bernards, 2003; (GAP1 like) adrenal medulla Sot et al., 2013; Sot et al., 2010)

CAPRI Yes Yes Gap1 Membrane translocation and inhibition and down regulation (Bernards, 2003; Kupzig et al., 2006a; (RASA4) of ERK/MAPK Kupzig et al., 2006b)

SYNGAP1 Yes Yes CG42684 Association with PSD-95/SAP90, SAP102 and the NMDA (Bernards, 2003; Guo et al., 2009; Kim receptor and involved in neuronal development and synaptic et al., 2003; Kim et al., 1998; plasticity as well as MAPK regulation Rumbaugh et al., 2006; Sot et al., 2010) RASAL2 Yes No CG42684 Linked to human prostate cancer susceptibility (Bernards, 2003; Noto et al., 1998) (nGAP) DAB2IP Yes No CG42684 DOC-2/DAB2 binding also its suppression causes activation (Bernards, 2003; Calvisi et al., 2011; (AF9Q34) of Ras signalling in human hepatocarcinogenesis Mitin et al., 2005) RASAL3 - - CG42684 - (Bernards, 2003) (FLJ21438) IQGAP1 No - NF1 Interacts with Rac1 and Cdc42Hs, expressed in the human (Bernards, 2003; Briggs and Sacks, keratinocytes, may contribute to the integrity of the 2003; Hart et al., 1996; Owen et al.,

epidermal layer and fundamental regulator of cytoskeletal 2008; Presslauer et al., 2003; function Weissbach et al., 1994) IQGAP2 No - NF1 Interacts with Cdc42 and Rac1 (Bernards, 2003; Brill et al., 1996)

IQGAP3 - - NF1 Interacts with Rac1 and Cdc42 and regulates Rac1/Cdc42- (Bernards, 2003; Wang et al., 2007) promoted neural outgrowth PLXN- Yes - PlexA Suppresses R-Ras in hippocampal neurons, (Bernards, 2003; Ito et al., 2006; A1/B1/C1/D1 dephosphorylation of Akt and activation of GSK-3β also Oinuma et al., 2004; Uesugi et al., potential role in cell adhesion, migration and axon guidance 2009) EPI64 A-C Yes - - Potentially involved in lowering Ras activity near the (Nagai et al., 2013) plasma membrane GAPEX-5 Yes - CG1657 Regulating EGFR ubiquitination and degradation (Bos et al., 2007; Hunker et al., 2006a; (RAP6) Su et al., 2007)

Table 1.1 Ras-related GAP types and their function. This table summarises the different Ras-related GAP molecules and their cellular function and spatial expression. Some of these Ras-related GAP molecules have dual GAP roles within cells as they act as a GAP for both Ras and Rap1, which are both small GTP-binding proteins.

Table 1.2 p120-RasGAP interacting partners

p120- Interacting Cellular function(s) Reference(s) RasGAP partner(s) domains GAP Ras GTP hydrolysis, MAPK (Pamonsinlapatham et al., 2009; regulation Trahey and McCormick, 1987) Rab5 GTP hydrolysis, control of (Liu and Li, 1998) early endosomal fusion Integrin α1/2 Recycling of endocytosed (Mai et al., 2011) α/β1-integrins to cell membrane

CaLB/C2 Annexin-A6 Organising membrane domains (Chow and Gawler, 1999; and signalling platforms Pamonsinlapatham et al., 2009)

PH PIP2 Allows RasGAP binding to the (Pamonsinlapatham et al., 2009) membrane PIP3 Allows RasGAP binding to the (Pamonsinlapatham et al., 2009) membrane Gβγ-protein Potentially affecting (Pamonsinlapatham et al., 2009; Xu et functioning of Ras-like al., 1996) proteins PKC - (Pamonsinlapatham et al., 2009) GAP Interferes with Ras and p120- (Drugan et al., 2000) RasGAP interaction RACK1 Potentially influencing (Koehler and Moran, 2001) subcellular localisation of active PKC ICP10 Potential role in HSV growth (Smith et al., 2000) onset

SH2 β-PDGFR RTKs signalling (Anderson et al., 1990; Pamonsinlapatham et al., 2009) EGFR RTKs signalling (Moran et al., 1990; Pamonsinlapatham et al., 2009) IR Insulin induced signalling (Pronk et al., 1992) pathway and Ras signalling EphB2-R Involved in cell repulsion and (Dail et al., 2006; Tong et al., 2003) effects the Ras-ERK pathway v-fms Possible control of cell (Trouliaris et al., 1995) morphology through RhoGAP v-Src TKs cellular signalling (Pamonsinlapatham et al., 2009) p62Dok-1 B-cell proliferation & chronic (Carpino et al., 1997; Donovan et al., myelogenous leukemia 2002) p56Dok-2 B-cell proliferation & chronic (Pamonsinlapatham et al., 2009) myelogenous leukemia p210bcr-abl Chronic myelogenous leukemia (Pamonsinlapatham et al., 2009; Tocque et al., 1997) p190- Cell polarity and movement (Kulkarni et al., 2000; RhoGAP Pamonsinlapatham et al., 2009) FAK Possible cytoskeletal (Endo and Yamashita, 2009) involvements SOCS3 Transcription suppression (Cacalano et al., 2001; Pamonsinlapatham et al., 2009) eEF1A2 Cell proliferation (Pamonsinlapatham et al., 2009; Panasyuk et al., 2008) Sam68 Cell proliferation and possible (Guitard et al., 1998; Jabado et al., T-cell activation 1998)

BCR-ABL1 Potential role in chronic (Colicelli, 2010; Frackelton et al., myelogenous-leukemia 1993) DDR1 Affecting DDR1 proximal (Lemeer et al., 2012) signalling Btl Regulating RTKs signalling (Woodcock and Hughes, 2004) strength in Drosophila Htl Regulating RTKs signalling (Woodcock and Hughes, 2004) strength in Drosophila Torso Regulating RTKs signalling (Cleghon et al., 1998) strength in Drosophila FGFR1 Ras/MAPK signal transduction (Cailliau et al., 2001) in Xenopus oocytes ICP10 Potential role in HSV growth (Smith et al., 2000) onset

SH3 G3BP Modulates docking and (Parker et al., 1996) translation efficiency of mRNA in GTP dependent manner p14-kDa - (Pamonsinlapatham et al., 2009) Aurora-B Cell cycle regulation and (Gigoux et al., 2002; tumour cell apoptosis Pamonsinlapatham et al., 2009) p200- AKT interaction, promoting (Pamonsinlapatham et al., 2009; Shang RhoGAP cell proliferation et al., 2007) Capns1 Influencing adhesive complex (Pamonsinlapatham et al., 2008; and tumour cell survival Pamonsinlapatham et al., 2009)

Proline Hck Potentially influencing Hck (Briggs et al., 1995) rich signal transduction N- Nck1 Increasing RasGAP catalytic (Ger et al., 2011) terminus activity towards H-Ras

Unknown p105 Marker of human trophoblast (Ye et al., 1999) blast differentiation SOCS1 RasGAP ubiquitination, (Cacalano et al., 2001; Madonna et al., possible cell survival 2008) SC1 Blocking stem cell (Chen et al., 2006) differentiation mTid1 GAP mediated regulation of (Trentin et al., 2001) cell growth PAG Possible T-cell activation (Smida et al., 2007) regulation ADAP Possible integrin based cellular (Sylvester et al., 2010) involvements

Table 1.2 p120-RasGAP interacting partners. The mammalian and Drosophila RasGAP interacts with Ras protein through its GAP domain; however, other domains of the RasGAP protein have the ability to interact with other proteins. This is an indication that RasGAP has other functions within cells (Pamonsinlapatham et al., 2009).

1.1.5 Indirect Ras deactivation in mammals As mentioned previously, Ras can either be deactivated directly or indirectly. There are many mechanisms of Ras and Ras signalling deactivation, which include: (i) ligand sequestration, destabilisation and binding inhibition, (ii) inhibition of RTK autophosphorylation, (iii) downstream signalling inhibitory proteins and (iv) ligand- induced receptor ubiquitination, endocytosis and degradation (Ledda and Paratcha, 2007; Zettl et al., 2011). One of the most well studied indirect deactivation mechanism of Ras signalling is achieved through endocytosis (Figure 1.5) and ultimately degradation of receptors such as RTKs, which are responsible for initiating Ras signalling. Endocytosis is a dynamic cell membrane-remodelling process, by which cells internalise a portion of their plasma membrane along with extracellular materials. Endocytosis is of fundamental importance in regulating: cellular uptake of nutrients, cell membrane protein abundance as well as signalling output of receptors (Ferguson and De Camilli, 2012). One of the most well characterised routes of receptor endocytosis and deactivation is through early stage Rab5 mediated endocytosis. This is the first stage of endocytosis, which happens in the periphery of the cell, leading to the collapse of the cell membrane and engulfing the surrounding extracellular materials for further processing, which is mediated through the Rab5 molecule.

Analogous to Ras, Rab5 is a member of Ras-like small GTPases superfamily, which binds to GDP and GTP but is only active in the Rab5-GTP form. The binding of GTP to Rab5 causes major conformational change in its switch I and switch II regions, allowing Rab5 to interact with its effector molecules such as sorting adaptors, tethering factors, kinases, phosphatases and motors (Stenmark, 2009). In humans there are more than 60 members of the Rab family. Rabs are reversibly localised to distinct subcellular compartments and membranes through their hydrophobic geranylgeranyl groups that are attached to one or two carboxy-terminal cysteine residues. This is intrinsic to their role in membrane trafficking (Jean and Kiger, 2012; Stenmark, 2009). In the classic endocytosis pathway, RTKs such as EGFR are autophosphorylated in a ligand-dependent manner, which leads to the activation of their downstream signalling pathways such as MAPK. However, this also initiates Rab5 mediated receptor endocytosis, which can be blocked by the overexpression of dominant negative (DN) Rab5 or kinase dead EGFR. RTK endocytosis can also be redirected/resumed by the overexpression of constitutively active (CA) Rab5 (Barbieri et al., 2004; Barbieri et al., 2000). Rab5 mediated endocytosis is also thought to be very

44 important for receptor degradation (Chen et al., 2009) resensitisation and dephosphorylation since DN Rab5 prevents receptor resensitisation and dephosphorylation (Seachrist et al., 2000). Once endocytosed, RTKs will either get recycled back to the cell membrane or get degraded and their rate of degradation can be affected by DN Rab5 (Dinneen and Ceresa, 2004). Interestingly however, MAPK but not Jun kinase (JNK) or p38 kinase activity of EGFR is reduced by the expression of DN Rab5, which indicates endocytosis is required for MAPK signalling (Barbieri et al., 2004).

The activation of Rab5 by exchange of GDP with GTP is catalysed by Rab5-GEFs (Figure 1.6), which are though to be partially responsible for Rab5 subcellular localisation and establishing Rab5 subpopulations within a cell. The RIN family of Rab5-GEFs plays an important role in receptor mediated endocytosis through their SH2-mediated receptor binding and Rab5 exchange activity (Balaji et al., 2012; Barbieri et al., 2004; Chen et al., 2009; Deininger et al., 2008; Galvis et al., 2009; Hu et al., 2008; Hunker et al., 2006b; Hunker et al., 2006c; Kajiho et al., 2003; Kajiho et al., 2011; Kimura et al., 2006; Kong et al., 2007; Sandri et al., 2012; Tall et al., 2001; Tomshine et al., 2009; Woller et al., 2011). RIN family is comprised of RIN1 (Colicelli et al., 1991; Han and Colicelli, 1995) RIN2 (Saito et al., 2002), RIN3 (Kajiho et al., 2003) and Rin-like (Rinl) (Woller et al., 2011). All the RIN proteins except Rinl have a Ras association domain and for RIN1 there is evidence that it can deactivate Ras signalling by sequestering active Ras (Barbieri et al., 2004) and competing with Raf1, hence the name Ras interaction/interference protein (Han and Colicelli, 1995; Wang et al., 2002).

Figure 1.5 Regulators of endocytosis and trafficking routes to endosomes. Endocytosis is achieved through two pathways: clathrin-mediated endocytosis (CME; Rab5- mediated endocytosis) and non-clathrin/clathrin-independent endocytosis (NCE/CIE; lipid raft based endocytosis) (Disanza et al., 2009; Le Roy and Wrana, 2005; Polo and Di Fiore, 2006). Although mechanistically different, most forms of endocytosis require dynamin-GTPase activity (not shown), which has a critical role in endocytic 46 membrane fission (Ferguson and De Camilli, 2012; Le Roy and Wrana, 2005). In the classic endocytosis pathway, CME, adaptor protein-2 (AP2) and epsins are stabilised by the phosphoinositide (PI) phosphatidylinositol (4,5)-bisphosphate (PtdIns(4,5)P2), which mediates their binding to clathrin triskelia (Jean and Kiger, 2012; Le Roy and Wrana, 2005; Stenmark, 2009). Clathrin in turn assembles into a polygonal lattice at the plasma membrane leading to membrane curvature and endosome formation around cargos such as receptors. The clathrin-coated early endosomes contain active Rab5-GTP and this Ras-like small GTPase and its GEF (GAPEX-5) co-ordinate clathrin uncoating by promoting dephosphorylation of AP2 subunit μ2 (Le Roy and Wrana, 2005; Marsh and McMahon, 1999; Stenmark, 2009). After endocytosis through Rab5 mediated early endosomes if the endocytic contents of cells are not secreted at a later stage via trans-Golgi network (TGN) (through Rab9), it will either recycle back to the cell membrane through Rab4 (with Rab15), Rab11 and Rab35 or mature (through Rab14 and Rab22) and get degraded in the late endosomes/lysosomes through Rab7 (Le Roy and Wrana, 2005; Stenmark, 2009). In the non-classic endocytosis pathway, NCE, which can be subdivided into dynamin-GTPase dependent and independent pathways, endocytosis is sensitive to cholesterol levels and is mediated through lipid rafts (not shown). If not recycled, receptors will get degraded through sustained post-translational modifications called multiubiquitylation, by E3 ubiquitin protein-ligase enzymes such as Cbl, which is a key signal recognised by ubiquitin-interacting motif (UIM) containing protein hepatocyte growth factor regulated tyrosine kinase substrate (HRS; not shown) (Le Roy and Wrana, 2005). PtdIns3P, PI3K and FYVE domain are thought to be required for HRS recruitment to endosomal membranes and its activity, which in turn is required to sort cargos to endosomal sorting complex required for transport (ESCRT)-I. ESCRT-I through ESCRT-II and ESCRT-III directs cargos into multivesicular bodies (MVBs), which after fusion with lysosome, degrades its contents such as receptors and terminates their signal (not shown) (Gruenberg and Stenmark, 2004; Le Roy and Wrana, 2005). Endocytic internalisation of receptors is often accompanied by receptor recycling (Disanza et al., 2009).

RIN1 Alsin

SH2 PRD RH VPS9 RA RCC1LD DH PH 8x M VPS9

RIN2 ALS2CL

SH2 PRD RH VPS9 RA 8x M VPS9

RIN3 GAPEX-5

SH2 PRD RH VPS9 RA GAP VPS9

Rinl VARP

SH2 RH VPS9 VPS9 4x A 4x A

RABEX-5

ZF VPS9 CC

A20-like zinc finger (ZF), Coiled-coiled (CC), Src-homology 2 (SH2), Ras association (RA), RCC1-like domain (RCC1LD), Dbl Homol ogy (DH), Pleckstrin homology (PH), GTPase activating protein (GAP), RasGAP and ankyrin (A), Membrane occupation and Recognition Nexus (MORN, M), RIN homology (RH) , Proline-rich domain (PRD), vacuolar protein sorting 9p-like domain (VPS9).

Figure 1.6 Schematic structures of mammalian Rab5-GEFs. Structurally and functionally RIN family members are similar with RIN1 (Colicelli et al., 1991; Han and Colicelli, 1995; Tall et al., 2001), RIN2 (Saito et al., 2002), RIN3 (Kajiho et al., 2003) and Rin-like (Rinl) (Woller et al., 2011) acting as GEFs for Rab5. The Rab5-GEF activity of RIN1 (Tall et al., 2001), RIN2 (Kimura et al., 2006), RIN3 (Yoshikawa et al., 2008) is potentiated by active Ras-GTP, which is thought to interact with the inhibitory RA domain of RINs, resulting in their active conformation as Rab5-GEFs. The RIN1 proline-rich domain (PRD) has been shown to be important in signal- transducing adaptor molecule (STAM)-2 mediated EGFR trafficking and degradation (Kong et al., 2007). The PRD of RIN1 can bind to the c-Abl tyrosine kinase SH3 48 domain (Afar et al., 1997; Han et al., 1997), eventually leading to Abl tyrosine kinase activation (Afar et al., 1997; Hu et al., 2005). Rabaptin-5-associated exchange factor for Rab5 (RABEX-5) also known as RabGEF1 is a GEF for Rab5 (Delprato et al., 2004; Esters et al., 2001; Horiuchi et al., 1997) and Rab21 (Delprato and Lambright, 2007; Delprato et al., 2004; Kajiho et al., 2003). Similar to the RA domain of RINs, RABEX-5 has an autoinhibitory element in a predicted amphipathic helix located near the C- terminal region of its VPS9 domain (Delprato and Lambright, 2007; Kalesnikoff et al., 2007). RABEX-5 is an effector of Rab22 (Zhu et al., 2009) and Rab33B (Valsdottir et al., 2001). RABEX-5 also has the capacity to bind to monoubiquitin and becomes covalently ubiquitinated (Lee et al., 2006), which is important for its recruitment from cytosol to endosomes (Mattera and Bonifacino, 2008). This RABEX-5-ubiquitin interaction occurs through two sites, one of which is the A20 zinc-finger domain (Lee et al., 2006; Mattera et al., 2006; Penengo et al., 2006; Raiborg et al., 2006). GAPEX-5 (Lodhi et al., 2007) has been reported as a GEF for Rab5 and Rab31 (Hunker et al., 2006a; Kitano et al., 2008; Lodhi et al., 2007; Sato et al., 2005). However, this has been contradicted in other studies (Chen et al., 2009; Kajiho et al., 2011). VPS9-ankyrin-repeat protein (VARP) acts as a GEF for Rab5 and Rab21 (Kajiho et al., 2011; Zhang et al., 2006). The ankyrin repeats of VARP are necessary for the endosomal targeting of VARP (Zhang et al., 2006) and similar to RABEX-5 (Delprato and Lambright, 2007), the GEF activity of VARP (Zhang et al., 2006) is suppressed by autoinhibitory elements present in the C-terminal region of its VPS9 domains (Kajiho et al., 2011). VARP is also known to be an effector for Rab32 (Tamura et al., 2009) and Rab38 (Wang et al., 2008). Amyotrophic lateral sclerosis 2 (ALS2), otherwise known as Alsin (Hadano et al., 2001; Yang et al., 2001) and ALS2 C-terminal like (ALS2CL) (Hadano et al., 2004) are GEFs for Rab5 (Hadano et al., 2004; Otomo et al., 2003) and Rab31 (Kajiho et al., 2011).

1.2 RasGAP mediated signalling in mammals RasGAP protein is not only thought to act upstream of the Ras as a GAP, but it is also controversially indicated that the RasGAP protein has effector functions, which can depend on its Ras binding. There have been attempts to justify this claim in the literature by emphasising the importance of the RasGAP binding to Ras effector domain in order for different cellular events to occur (Schweighoffer et al., 1992). A notable study implicating RasGAP as an effector showed RasGAP depends on Ras binding to mediate the inhibition of coupling of muscarinic receptors to atrial K+ channels (Yatani et al., 1990). Several studies have argued that since the RasGAP C-terminal region is sufficient for its GAP activity, other regions of the protein such as its N-terminal regions are capable of mediating other cellular functions (including Ras inhibition) within different tissues (such as cardiac muscle and neuronal cells), without depending on Ras binding (Abdellatif and Schneider, 1997; Clark et al., 1993; Clark et al., 1997; Giglione et al., 1997; Huang et al., 1993; Martin et al., 1992; Tocque et al., 1997; Xu et al., 1996). RasGAP has been shown to have many interacting partners, other than Ras (Table 1.2), which bind to different domains of RasGAP. One such domain is the RasGAP SH2 domain, which plays a significant role in cellular processes in mammalian cells.

1.2.1 SH2 domain and SH2-mediated signalling In 1986 Sadowski et al., (1986) reported a conserved region of approximately 100 amino acids in the N-terminal region of cytoplasmic protein tyrosine kinases (PTKs), distinct from the catalytic domain, that was named the Src homology-2 (SH2) domain. The absence of the SH2 in transmembrane receptor tyrosine kinases suggested that this noncatalytic domain may direct specific interactions or regulate cytoplasmic PTKs functions (Sadowski et al., 1986). SH2 domains were subsequently found to bind to tyrosine- phosphorylated proteins (Matsuda et al., 1990) and by 1992, X-ray crystallography demonstrated that SH2 specific recognition of phosphotyrosine (pTyr) residues involves the amino-aromatic phosphotyrosine interactions with lysine and arginine side-chains of the SH2 (Waksman et al., 1992). The SH2 domain was later shown to recognise two classes of binding motifs. The first class of the SH2-binding motifs had the consensus sequence pTyr-hydrophilic-hydrophilic-Ile/Pro, which was shown to be the preferred binding motif for the SH2 containing proteins such as Src and RasGAP. The second class of the SH2-binding motifs had the consensus sequence pTyr-hydrophobic-X-hydrophobic (X, any amino acid residues), which was shown to be the preferred binding motifs for the

SH2 containing proteins such as p85 subunit of PI3K and PLC-γ (Bradshaw et al., 1999; Eck et al., 1993; Felder et al., 1993; Huang et al., 2008; Kay et al., 1996; Pascal et al., 1994; Songyang et al., 1993; Waksman et al., 1993). Further studies demonstrated that the binding specificity of the SH2 is important for determining its binding site (Huang et al., 2008; Marengere et al., 1994).

Many SH2 domains mediate binding to activated PTKs in a phosphotyrosine-dependent manner, although it is important to note that SH2 containing proteins can bind to their target protein in a phosphotyrosine-independent manner (Evans et al., 2012; Liao et al., 2007; Mahajan and Earp, 2003). This has led to the proposal that the SH2 domain evolved as an adaptor region, relaying extracellular signals to the intracellular environment (Cantley et al., 1991; Hunter, 2000; Koch et al., 1991; Liu et al., 2006; Machida et al., 2007; Mayer and Baltimore, 1993; Nars and Vihinen, 2001; Pawson and Gish, 1992; Pawson et al., 2001; Pawson et al., 1993; Pawson et al., 2002; Schlessinger, 2000; Yaffe, 2002). The human genome contains 120 SH2 domains in 110 distinct proteins ranging from kinases, phosphatases, cytoskeletal proteins, regulators of small GTPases and E3 ubiquitin ligases amongst others. Hence it is not surprising that genetic mutations in some SH2 domains and SH2 domain containing proteins are associated with human diseases (Huang et al., 2008; Lappalainen et al., 2008; Liu et al., 2006; Machida et al., 2007).

1.2.2 RasGAP interacting partners in mammals After the discovery of RasGAP, SH2-mediated RasGAP interaction with activated cytoplasmic PTKs such as v-Src was reported (Ellis et al., 1990). The significance of SH2- mediated protein interaction soon became apparent when it was discovered that the SH2 domains of PLC-γ, RasGAP and Src bind to the phosphotyrosine residues of autophosphorylated (activated) RTKs such as EGFR and PDGFR (Anderson et al., 1990; Moran et al., 1990; Pawson et al., 1993). Later, it was noted that RasGAP SH2 domain interacts with the pTyr-X-X-Pro-X-Asp (YXXPXD) region of ligand-stimulated RTKs (such as PDGFR) and this interaction is accompanied by a high degree of tyrosine phosphorylation on RasGAP itself (Tyr460) (Ellis et al., 1990; Fantl et al., 1992; Kaplan et al., 1990; Liu and Pawson, 1991; Margolis et al., 1990; Molloy et al., 1989; Molloy et al., 1992; Moran et al., 1991; Serth et al., 1992). The RasGAP binding then in turn indirectly regulates RTKs signalling through Ras (Chiara et al., 2004; Tallquist et al., 2003).

Two tyrosine-phosphorylated proteins p62 and p190 were found to associate with RasGAP in both PTK transformed and EGF stimulated cells (Ellis et al., 1990). It was later established that the 190 kDa protein that associates with RasGAP is a member of RhoGAP family (Settleman et al., 1992) and both (N- and C-terminal) SH2 domains of the p120- RasGAP bound synergistically to tyrosine-phosphorylated p190-RhoGAP (Bryant et al., 1995). The importance of the SH2 domains in mediating p120-RasGAP-p190-RhoGAP interaction was later confirmed when it was discovered that two closely linked phosphotyrosine peptides (pTyr1087 and pTyr1105) on p190-RhoGAP (with the consensus sequence YXXPXD) bind simultaneously to the double SH2 regions of p120-RasGAP. This brings the two SH2 domains in close spatial proximity, allowing p120-RasGAP to undergo conformational changes, resulting in 100-fold increased accessibility of the p120- RasGAP SH3 region to any potential SH3 interacting proteins. This highlighted the importance of Src homology domain tandem arrangement (i.e. SH2-SH3-SH2) in mediating p120-RasGAP molecular interaction through conformational changes (Hu and Settleman, 1997). Several studies have shown that regions outside the GAP domain, including the SH2 and PH domains, are important for full activity of RasGAP (Bryant et al., 1996; Drugan et al., 2000; Gideon et al., 1992). Hence, the above findings suggest a multi-dimensional role for the p120-RasGAP N-terminal region (Hu and Settleman, 1997).

1.2.3 Biological significance of RasGAP in mammals In 1997, van der Geer et al., (1997) produced p120-RasGAP mutant (Gap-/-) mouse fibroblasts and discovered that when stimulated with PDGF, Ras-GTP levels become abnormally elevated compared to wild-type fibroblasts (van der Geer et al., 1997). This abnormal elevation of Ras-GTP level was in stark contrast to unstimulated Gap-/- fibroblasts, where no elevation was observed. This indicated that p120-RasGAP is involved in down regulation of Ras protein signalling after RTK stimulation; however, it plays an insignificant role in maintaining Ras-GTP basal levels. Similar to Gap-/- mouse fibroblasts, Nf1-/- mouse fibroblasts (NF1 being another Ras-GTPase activating protein molecule) showed elevated Ras-GTP levels after PDGF stimulation; however, they also had elevated basal Ras-GTP levels (van der Geer et al., 1997). In PDGF stimulated Gap-/- fibroblasts, elevated Ras-GTP levels were accompanied by extended duration of mitogen- activated protein kinase (MAPK) activation. It was suggested that this increase in MAPK duration may reflect the heightened Ras-GTP levels, although the possibility of a Ras-GTP

52 independent mechanism of MAPK activation in cells was not ruled out (van der Geer et al., 1997). p120-RasGAP was shown by overexpression and loss of function studies to play a role in cytoskeletal remodelling, cell polarity and cell migration. Expression of the N-terminal region (SH2-SH3-SH2) in tissue culture cells led to disruption of actin stress fibers, a reduction in focal adhesions and impaired adherence to the extracellular matrix (McGlade et al., 1993). p120-RasGAP deficient (Gap-/-) mouse fibroblasts in culture were six times slower in cell migration relative to wild-type cells, using a scratch assay of wound healing (Kulkarni et al., 2000). However, stimulation of the cells with PDGF partially rescued this migration defect, which suggested a partial role for the Ras signalling pathway in mammalian cell migration. In addition to impaired migration, Gap-/- cells also failed to establish (or weakly established) polarity as the majority of the cells did not form lamellipodia or any recognisable protrusive activity. The Gap-/- mouse fibroblasts had fragmented Golgi apparatus and were incapable of forming organised actin stress fibres and focal adhesions in orientation parallel to the direction of cell migration (Kulkarni et al., 2000). Organised actin stress fibres and focal adhesions play an important role in the cytoskeleton mediated forward movement of cells and in the establishment of the cell polarity and cytoskeletal re-arrangements. This lack of organised actin stress fibres and focal adhesions provided an explanation for the impaired cell migration and polarity in Gap-/- mouse fibroblasts. Further experiments suggested that impaired p120-RasGAP- p190-RhoGAP interaction in Gap-/- mouse fibroblasts could be responsible for actin stress fibre and focal adhesion disorganisation during directed cell movement. Synthetic peptides that inhibited p190-RhoGAP interaction with the SH2 domains of p120-RasGAP led to inhibition of organised actin stress fibres and focal adhesion formation. This study did not exclude the possibility that p120-RasGAP SH2-binding partners other than p190-RhoGAP could be responsible for this cellular phenotype; however, it did show that p120-RasGAP and p190-RhoGAP play an important role in cell polarity and movement (Kulkarni et al., 2000).

Aberrant p120-RasGAP regulation in Gap-/- mammalian cells not only has molecular and cellular consequences, but whole organism implications as well. Henkemeyer and colleagues showed that mice with p120-RasGAP null mutations (Gap-/-) show reduced size at embryonic (E) stage E9.5, which was accompanied by distended pericardial sac and arrested posterior elongation (Henkemeyer et al., 1995). The endothelial lining of Gap-/-

53 homozygote yolk sac at E10.5 began to exhibit a honeycomb pattern rather than a highly vascular pattern. At the 12-somite stage the endothelial architecture, heart chamber and major blood vessels (such as aortic arch and dorsal aorta) of Gap-/- embryos matched those of the wild-type; however, at the 16-somite stage (E9.0), the mutant embryos had thinner dorsal aorta and formed a number of aberrant ventral-projecting aortic branches. At E9.5, the vasculature began to exhibit local ruptures and the heart became surrounded by a distended pericardium as the result of blood leakage from the vasculature. This eventually led to their death at E10.5 (Henkemeyer et al., 1995). The aforementioned embryonic developmental defects in RasGAP mutant mice has also been observed in several other studies including when the RasGAP expression was targeted with RNA interference (RNAi) technology (Iwashita et al., 2007; Kunath et al., 2003). Although conditional knock-out of the p120-RasGAP (RASA1) gene in adult mice seems to have little effect on blood vasculature, endothelial cells and immune cell lineage development, (Lapinski et al., 2007) it does, however, result in a lymphatic vessel disorder. This is characterised by extensive lymphatic vessel hyperplasia due to increased proliferation of lymphatic endothelial cells (LECs), leakage and early lethality. The lethality is caused by chylothorax, which is lymphatic fluid accumulation in the pleural cavity. This is thought to be due to negative regulatory effects of p120-RasGAP on vascular endothelial growth factor receptor (VEGFR)-3 mediated Ras signalling transduction (Lapinski et al., 2012).

In the Henkemeyer and colleagues study, the Nf1-/- mutated mice appeared relatively normal until E12.5, but exhibited heart defects during development and subsequently died between E13.5 and E14.5. Strikingly, Gap-/-; Nf1-/- double mutant mice exhibited abnormal phenotypes in a synergistic manner. These double mutants failed to exhibit turned posterior tails and suffered enlarged allantois, ruffled yolk sac and major abnormalities of the heart and the surrounding region (Henkemeyer et al., 1995). These observed Gap-/- vascular phenotypes are not just limited to murine models. Capillary malformation-arteriovenous malformation (CM-AVM) is a vascular disorder observed in humans, characterised by atypical cutaneous multifocal capillary malformations. In addition, patients with CM-AVM are found to suffer from cutaneous, subcutaneous, intramuscular, intraosseous and cerebral arteriovenous malformations (Boon et al., 2005; Eerola et al., 2003; Frigerio et al., 2012). Recent studies have highlighted 47 mutations in RASA1 (human homologue of the mouse GAP) gene, 16 of which are found at the amino terminal region of the RASA1 gene in patients suffering from CM-AVM or similar vascular abnormalities (Boon et al., 2005; de

Wijn et al., 2012; Durrington et al., 2013; Hershkovitz et al., 2008b; Revencu et al., 2008). The localised and asymmetric nature of the vascular abnormalities and the incomplete penetrance of the mutations within patients have suggested that pre-existing germ-line heterozygous mutations within RASA1 gene may suffer a somatic second hit mutation, resulting in complete lack of the RasGAP activity. This hypothesis allows for the variable phenotype of CM-AVM within patients, which may reflect the extent and the identity of the cellular populations affected (Eerola et al., 2003). The asymmetry of the affected regions, however, can also be explained by possible environmental and/or developmental affects. The RASA1 gene mutations are also associated with other vascular disorders such as Sturge-Weber syndrome (SWS) and Klippel-Trenaunay-Weber syndrome (KTWS) and its importance in angiogenesis has been studied (Anand et al., 2010; Boutarbouch et al., 2010; Zhou et al., 2011). The collective results seem to suggest that RasGAP plays an important role in vascular formation (Boon et al., 2005; Henkemeyer et al., 1995; Hershkovitz et al., 2008a; Paramasivam et al., 2013; Puttgen and Lin, 2010; Revencu et al., 2008; Thiex et al., 2010; Wang, 2005).

One of the most interesting findings of the Henkemeyer and colleagues study (Henkemeyer et al., 1995) on Gap-/- mice, was the presence of extensive neuronal abnormalities. At the 16-somite stage of development (E9.0), Gap-/- mice exhibited localised regions of extensive neuronal cell death and densely stained pyknotic nuclei within their branchial arch, hindbrain and anterior brain structures such as the optic stalk and telencephalon. The dying neuronal cells showed the characteristics of apoptotic cell death (Clarke, 1990; Henkemeyer et al., 1995) since electron microscopy data showed electron dense chromatin masses, convoluted nuclear membrane, cell shrinkage and darkening of the neuronal cytoplasm. In addition, apoptotic bodies containing fragmented chromatin particles and intact organelles were observed, which were phagocytosed by neighbouring cells. As observed in Gap-/- mutants, a transverse section through the hindbrain of the Gap-/-; Nf1-/- double mutant mice at E9.0 also revealed extensive apoptotic neuronal cell death in neuroectoderm, hindbrain and migrating cranial neuronal crest. These double mutant mice failed to close the dorsal surface of the hindbrain and three of the four specimens failed to develop any identifiable midbrain and forebrain structures (Henkemeyer et al., 1995). Single Gap-/- and double Gap-/-; Nf1-/- mice showed striking differences in their hindbrain neuroepithelium morphology; localised regions of neuronal cell outgrowth, indicative of increased proliferation and/or survival, were observed within the neuronal tube of the

55 double mutant mice (Henkemeyer et al., 1995). The above findings suggested that p120- RasGAP and NF1 act together in controlling neuronal survival presumably through regulation of Ras activity.

1.3 Ras signalling in Drosophila

1.3.1 Ras protein in Drosophila In Drosophila three genes encoding Ras related proteins were identified by cross- hybridisation with mammalian H-Ras: Ras1 (Dras1/Ras85D), Ras2 (Dras2/Ras64B) and Ras3 (Dras3/Roughened), with mammalian Ha-Ras, N-Ras, Ki-Ras 4A and Ki-Ras 4B corresponding to Ras1, mammalian R-Ras corresponding to Ras2, and mammalian Rap1 corresponding to Ras3 (Neuman-Silberberg et al., 1984; Schejter and Shilo, 1985). Research on the Drosophila Ras homologues initially focused on Ras1 as it was shown to be a critical component of signalling pathways downstream of multiple RTKs. Genetic analysis of RTK signalling in Drosophila was crucially important in elucidating the RTK- Ras-MAPK pathway (Figure 1.7). For example, genetic analysis of the Sevenless RTK signalling pathway in the Drosophila eye led to the discovery of SOS, the first Ras activator identified in multicellular organisms (Simon et al., 1991).

1.3.2 Ras upstream regulators in Drosophila Drosophila has many RTKs (Table 1.3) that perform diverse developmental roles within the organism. However, it was the work on the Sevenless (Sev) RTK that elucidated the Ras pathway in flies. Sev is a RTK that determines the fate of one of the photoreceptors (R7) within Drosophila eye ommatidium in a Ras dependent manner (Perrimon, 1994; Simon et al., 1991; Simon et al., 1989). Sev is activated by its ligand, bride of sevenless (Boss), expressed by the R8 photoreceptor, resulting in autophosphorylation and activation of Sev, which allows SH2 containing proteins such as Drk (mammalian Grb2 equivalent) to bind to it, while the Ras-GEF protein, SOS, binds to Drk SH3 domain. The SOS protein activates Ras1, converting Ras-GDP to Ras-GTP (Olivier et al., 1993; Perrimon, 1994; Simon et al., 1993). This evolutionary conserved RTK signalling pathway allows Drosophila Ras to interact with its downstream effectors (Perrimon, 1994).

Figure 1.7 RTK signalling in Drosophila. The above diagram shows RTK signalling in Drosophila and the downstream molecules that are activated in response to RTK signalling. aop; anterior open, grk; gurken, hkb; huckbein, phl; pole hole, sev; sevenless, till; tailless, tsl; torso-like.

Table 1.3 Drosophila RTKs and their mammalian homologues

Mammalian Drosophila Biological function(s) RTK RTK homologue(s) Alk Visceral muscle development; digestive tract development; axon guidance; ALK compound eye photoreceptor development; regulation of hemocyte differentiation and mesoderm development

Btl Fibroblast growth factor receptor activity. There is experimental evidence FGFR family for 11 unique biological process including: anatomical structure development; open tracheal system development and organ morphogenesis

Cad96Ca Protein amino acid phosphorylation and calcium-dependent cell-cell Ret RTK adhesion

CG3277 Protein amino acid phosphorylation PDGFR- related

Ddr Transmembrane receptor protein tyrosine kinase signalling pathway DDR family

dnt Axon guidance; signal transduction; muscle attachment and salivary gland RYK family morphogenesis

drl Olfactory learning; axon guidance; learning or memory; memory; signal RYK family transduction; muscle attachment; salivary gland morphogenesis and axon midline choice point recognition

Drl-2 Wnt receptor signalling pathway and salivary gland morphogenesis RYK family

Egfr Anatomical structure development; reproductive cellular process; organ EGFR family development; organ morphogenesis; regulation of developmental process and sensory organ development

Eph Mushroom body development Eph receptor family

Htl Anatomical structure development; organ development; organ FGFR family morphogenesis; cellular component movement; mesoderm morphogenesis; muscle organ development; gliogenesis and central nervous system development

InR Anatomical structure development; regulation of biological process; InR family reproductive cellular process; regulation of anatomical structure size and developmental growth

Nrk Protein amino acid phosphorylation and signal transduction Ror1 and Ror2

Otk Axon guidance and cell adhesion TrkA

Pvr System development; cellular component movement; regulation of PDGF/VEGF developmental process; organelle organisation; actin filament-based receptor process and hemopoiesis

Ret Protein amino acid phosphorylation and signal transduction Ret RTK

Ror Central nervous system development Ror1 and Ror2

Sev Positive regulation of photoreceptor cell differentiation; R8 cell fate - specification and R7 cell fate commitment

Tie Protein amino acid phosphorylation Tie-receptor family

Tor Terminal region determination; pole cell migration; metamorphosis; - gastrulation and chorion-containing eggshell pattern formation

Table 1.3 Drosophila RTKs and their mammalian homologues. The above table lists the known Drosophila RTKs (Edwards, 2007; Woodcock, 2004) along with their mammalian homologues. The biological function(s) for each RTK is taken from the FlyBase database.

1.3.3 Ras downstream effectors in Drosophila Consistent with mammalian Ras signalling, Drosophila Ras in its Ras-GTP state interacts with its effectors such as D-Raf and PI3K. As with the mammalian system, D-Raf has many effectors including DSOR1, which subsequently activates Rolled, mediating the activation of the MAPK pathway and inducing the appropriate biological response (Leevers et al., 1996) such as regulation of insulin sensitivity and eye development in Drosophila (Zhang et al., 2011). The PI3K pathway signalling in Drosophila regulates cell growth and coordinates cellular metabolism (Britton et al., 2002).

1.3.4 Direct Ras deactivators in Drosophila Once Ras-GTP mediates its effects through downstream effectors, it has to be deactivated either directly or indirectly. Direct Ras deactivation is achieved through deactivators of the Ras protein, GTPase activating proteins. Drosophila has one member of each of the RasGAPs found in mammals, which makes it an attractive model to study this family of proteins (Table 1.1). The first Drosophila GAP discovered was a mammalian Gap1 family homologue, named Gap1, which was thought to regulate Drosophila Ras1 activity since Gap1 mutant flies mimicked Sev constitutive expression rough-eye phenotype (Gaul et al., 1992). Gap1 function as a Ras1 regulator was later confirmed since Gap1 overexpression phenocopied abnormal eye-phenotype, similar to reduced Ras1 expression in Drosophila (Rorth, 1996). Comparative and evolutionary analysis of the Drosophila genome has also identified homologous Drosophila genes to human GAPEX-5 (CG1657) and SynGAP related proteins, referred to as CG42684 (CG5960) (Bernards, 2003; Jiang and Ramachandran, 2006). Other Drosophila GAP proteins include homologues of mammalian NF1 and p120-RasGAP proteins (Table 1.1). The Drosophila NF1 protein has a GAP activity for human H-Ras (in vitro), Ras1 (in vivo) and Ras2 (in vivo) (Guo et al., 1997; Guo et al., 2000; The et al., 1997; Walker et al., 2006). Drosophila RasGAP is the homologue of the mammalian p120-RasGAP, and is highly expressed in Drosophila ovaries, accessory gland and brain, in order of highest expression (FlyBase.org GBrowse FlyAtlas). The protein domains between Drosophila RasGAP and mammalian p120- RasGAP is conserved. In addition, Drosophila RasGAP can stimulate the GTPase activity of the human H-Ras protein (in vitro) and will form the focus of this study (Feldmann et al., 1999).

1.3.5 Indirect Ras deactivation in Drosophila Similar to mammals (section 1.1.5), indirect Drosophila Ras and Ras signalling deactivation can be achieved through many mechanisms. These include (i) receptor-ligand sequestration, destabilisation and binding inhibition through molecules such as Argos (Jin et al., 2000), Kekkon 1 (Ghiglione et al., 2003), Rhomboids (Freeman, 2008) and iRhoms (Zettl et al., 2011), (ii) inhibition of RTK autophosphorylation through molecules such as Lrig1 (Ledda et al., 2008), (iii) downstream signalling inhibitory proteins such as Sprouty (Cabrita and Christofori, 2008) and (iv) ligand-induced receptor ubiquitylation, endocytosis and degradation involving molecules such as Cbl (Pai et al., 2000) and HRS (Lloyd et al., 2002), all of which are fully reviewed elsewhere (Ledda and Paratcha, 2007). Similar to the mammalian system, receptor endocytosis and degradation plays a significant role in Drosophila Ras and Ras signalling regulation. For example Drosophila HRS (Lloyd et al., 2002) and E3 ubiquitin ligase, Cbl (Pai et al., 2000), regulate EGFR signalling, localisation and degradation through their roles in endocytic events. Another molecule that controls Drosophila RTKs signalling is the single Drosophila homologue of mammalian RIN family proteins, known as SH2, poly-proline containing Ras interactor (Sprint). Sprint is a Ras-binding VPS9 containing protein, which has been suggested to act as a Rab5-GEF. Sprint has been shown to regulate Drosophila border cell migration by influencing EGFR and PDGF- and VEGF-receptor related (PVR) localisation and signalling. Sprint is thought to regulate PVR and EGFR signalling and localisation through endocytosis (Jekely et al., 2005; Szabo et al., 2001).

1.4 RasGAP mediated signalling in Drosophila

1.4.1 RasGAP interacting partners in Drosophila The function of the RasGAP SH2 domains, in terms of mediating RasGAP interaction with phosphotyrosine proteins, has remained conserved between man and Drosophila. Overexpression studies have shown that the RasGAP SH2 domains mediate interaction with the Drosophila RTKs breathless (Btl), heartless (Htl) and Torso (Tor), which allows Ras1-MAPK signalling regulation (Cleghon et al., 1998; Feldmann et al., 1999; Woodcock and Hughes, 2004).

1.4.2 Biological significance of RasGAP in Drosophila In 2003, Botella et al., (2003) identified transposable element (P element) induced mutations in the RasGAP gene that caused age-related brain degeneration in the Drosophila adult brain. These mutants were shown to be allelic with the vacuolar peduncle (vap) mutant identified previously as a neurodegeneration mutant by Heisenberg and colleagues (Botella et al., 2003; de Belle and Heisenberg, 1996; Melzig et al., 1998). Botella and colleagues (2003) investigated the role of the RasGAP protein in neuronal survival in flies and their key findings were as follows.

(i) Allele specific time course of neurodegeneration in vap mutant flies, demonstrating the age-related neurodegeneration phenotype in vap mutants. All the alleles demonstrated spongiform appearance of central brain, lobula complex and medulla at 100% penetrance. (ii) Neuronal cell death in vap mutants due to autophagic degeneration, which were prominently found in the adult Drosophila. Although 3rd instar larvae brain seemed unaffected; however, many features of the autophagic (type 2) cell death could be observed in neurons of the vap mutant adult flies, including vacuoles containing whorls of membranous material, autolysosomes and empty vacuoles from the first day after eclosion. (iii) Spatial specificity of the Drosophila RasGAP (vap gene) expression. The vap gene was expressed in the whole cortex of the adult Drosophila brain in a pan- neuronal pattern and was not expressed in other non-neuronal cells such as the neuropil glial cells. (iv) vap gene product interaction with the EGFR pathway. The Egfr heterozygous mutant flies suppressed the vap mutant phenotype whereas overexpression of

the Egfr enhanced the vap mutant phenotype. Since misexpression of Egfr has no brain degeneration effect in the wild-type flies, it indicated that the ectopic activation of Egfr in wild-type flies can be effectively downregulated, whereas the vap mutants became sensitised, confirming the role of RasGAP as a negative regulator of the EGFR pathway. (v) Pan-neuronal expression of Drosophila Ras1 enhanced the vap mutant phenotype. The expression of Ras1 in vap mutant background enhanced the vap mutant phenotype whereas expression of Ras1 in a wild-type background did not cause neurodegeneration. (vi) Downstream Ras-dependent signalling was upregulated in vap mutants. Since MAPK can be activated in vap mutant background, this indicated that RasGAP protein was not required for activation of MAPK (consistent with van der Geer et al., (1997)). In addition, MAPK expression did not enhance the neurodegeneration phenotype in vap mutant background.

The genetic interactions between vap and components of the EGFR pathway indicated that RasGAP was involved in regulating the strength of signalling through the EGFR pathway. However, this study also indicated that a pathway other than the Raf/MAPK pathway might be involved in the vap mutant neurodegenerative phenotype since no genetic interactions between vap and Raf/MAPK were observed. This indicated two things; either in the vap mutant RasGAP fails to compete with another molecule for interacting with Ras and this causes the neurodegenerative phenotype (pers. comm. Stephan Schneuwly) or that the vap mutant phenotype was not due to RasGAP effects on MAPK signalling but due to misregulation of another pathway. To investigate the second hypothesis, Botella and colleagues conducted a series of rescue experiments and discovered that GAP mutant RasGAP (R695K, equivalent to mammalian p120-RasGAP R789K), which has been shown in yeast to be GAP deficient, could still rescue the mutant vap phenotype (J. Botella, S. A. Woodcock, D. A. Hughes and S. Schenuwly, pers. comm.). However, no vap mutant rescued flies were obtained by expression of a mutant RasGAP construct with both SH2 domains mutated to leucine at the highly conserved arginine residues (R110L and R278L). In addition, no vap mutant rescued flies were obtained when two other RasGAPs, Gap1 (Figure 1.8) and Nf1, were expressed. This indicated that RasGAP function was rather specific in the adult Drosophila brain and that GAP activity of the RasGAP protein, which is required for Ras protein regulation, was not required for neuronal survival.

Therefore neurodegeneration in vap flies appeared to be GAP-independent, but dependent on interactions involving the SH2 domains.

ol cb

Figure 1.8 Drosophila vap mutant neurodegeneration phenotype and rescue. The above diagram is adapted from Botella et al., (2003) and shows the (A) vap mutant neurodegeneration in flies with vacuolisation in the optic lobe (ol) and central brain (cb). This can be rescued by the pan-neuronal expression of (B) RasGAP cDNA but not (C) Gap1 cDNA in neurons. Scale bar is 50 μm.

1.4.3 Modelling neurodegenerative disorders in Drosophila and its implications for humans Neurodegenerative disorders are becoming an increasing problem since human life expectancy has dramatically increased in recent years. Hence, understanding the underlying mechanisms behind neurodegenerative disorders and effective medical interventions are becoming an increasing priority. Although some progress has been made in understanding the mechanisms that lead to neurodegenerative disorders such as Alzheimer’s disease, the etiology of these disorders is poorly understood. Drosophila has been used as a simpler model system to investigate the cellular and molecular mechanisms of neurodegeneration (Lu, 2009; Lu and Vogel, 2009; Michno et al., 2005; Muqit and Feany, 2002). One approach taken in Drosophila has been to express mammalian proteins known to cause neurodegereration in fly neurons. For example in one study, a human amyloid precursor protein (APP), β-site APP-cleaving enzyme (BACE) and a fly Presenilin (Psn) transgene were co-expressed, resulting in transgenic flies developing β-amyloid plaques and age-dependent neurodegeneration (Lu and Vogel, 2009). It is then possible to use genetics to identify modifiers of the phenotype caused by expression of the mammalian gene, with the aim of identifying fly genes that function in the neurodegeneration pathway.

A second approach has been to identify mutants that cause neurodegeneration, either in the compound eye or the adult brain (Lu and Vogel, 2009). The Drosophila vap gene is one of 49 genes discovered to this date, which are associated with neurodegeneration in Drosophila (Lessing and Bonini, 2009). From the identified 49 genes, there are currently 28 genes with a mouse or human orthologue associated with neurodegeneration. This shows that humans and flies have many signalling pathways in common that affect the survival of neurons in vivo. The small genome, short reproductive cycle and ease of transgene expression makes Drosophila an attractive model for studying neurodegenerative disorders (Lessing and Bonini, 2009; Lu, 2009; Lu and Vogel, 2009; Michno et al., 2005; Muqit and Feany, 2002).

1.5 Aims and objectives The aims of this project were to identify RasGAP SH2-dependent interacting proteins and determine their role in neuronal survival in the adult Drosophila. This was established using tagged wild-type RasGAP and double SH2 inactivated RasGAP, which were affinity purified from Drosophila S2 cells in order to mass spectrometrically identify co-purifying proteins using several identification techniques such as spectral counting. Biochemical techniques such as western-blotting were used to verify RasGAP interacting proteins and determine their mechanism of binding using point mutations. Immunohistological techniques and fluorescent microscopy were used to validate RasGAP SH2-dependent interacting partners and determine their subcellular localisation in S2 cells. Using Drosophila genetics, reverse genetics were employed to assess the effects of the genes of interests on the neurodegenerative phenotype observed in the vap gene mutant flies.

CHAPTER 2: Materials and Methods

2.1 Stocks, strains and plasmids

2.1.1 Stocks and strains The Escherichia coli (E. coli) host strains used for plasmid transformation and purification - - - were BL21(DE3) competent cells (Stratagene) with B F dcm ompT hsdS (rB mB ) gal λ(DE3) genotype and XL1-Blue competent and supercompetent cells (Stratagene) with recA1 endA1 gyrA96 thi-1 hsdR17 supE44 relA1 lac [F’ proAB laclqZΔM15Tn10(Tetr)] genotype. The Drosophila stocks used in this study are listed in Table 2.1.

Table 2.1 Drosophila stocks

Stock Genotype/description Source(s) vap2 vap mutant (hypomorphic allele) (Botella et al., 2003) Rab54 y-, w-; Rab54/CyO (hypomorphic allele) (Wucherpfennig et al., 2003) spri6G1 w-, spri6G1 (Jekely et al., 2005) vap2, spri6G1 vap2, spri6G1 This study w-, vap2, spri6G1 w-, vap2, spri6G1 This study y, w y-, w- -

2.1.2 Plasmids The plasmids listed in Table 2.2 were constructed by previous members of our laboratory (D. Hughes, P. Feldmann and S. Woodcock). The plasmids listed in Table 2.3 were constructed by other researchers. The plasmids listed in Table 2.4 were constructed as described below and were used during this study.

Table 2.2 Plasmids constructed previously in this laboratory

Plasmid description Polypeptide encoded pUAST-RasGAPWT-myc Full length wild-type Drosophila RasGAP with C-terminus myc tag pUAST-RasGAPSH2*32*-myc Full length Drosophila RasGAP with mutations R110L, R278L with C-terminus myc tag pUAST-RasGAPY363F-myc Full length Drosophila RasGAP with mutation Y363F with C-terminus myc tag pUAST-RasGAPSH2*32-myc Full length Drosophila RasGAP with mutation R110L with C-terminus myc tag pUAST-RasGAPSH232*-myc Full length Drosophila RasGAP with mutation R278L with C-terminus myc tag pUAST-RasGAPSH23*2-myc Full length Drosophila RasGAP with mutation W219A with C-terminus myc tag pUAST-RasGAPR695K-myc Full length Drosophila RasGAP with mutation R695K with C-terminus myc tag pUAST-SH232-myc The SH2-SH3-SH2 region of Drosophila RasGAP composed of residues 83-343 with C-terminus myc tag pGEX-KG-GST-SH232WT The SH2-SH3-SH2 region of Drosophila RasGAP composed of residues 83-343 with N-terminus GST tag pGEX-KG-GST-SH2*32* The SH2-SH3-SH2 region of Drosophila RasGAP composed of residues 83-343 with mutations R110L, R278L with N-terminus GST tag

Table 2.3 Plasmids constructed by other researchers

Plasmid description Polypeptide encoded Source(s) pIC111 pcDNA3.1 backbone with C-terminus 6xHis- (Cheeseman and Desai, 2005) 2xPreScission-eGFP (LAP) tag pUAST Empty pUAST backbone vector (Brand and Perrimon, 1993)

pBluescript II KS (-) Empty pBluescript II KS (-) backbone vector Agilent technologies

pUAST-GFP GFP protein -

pMT/GAL4 Copper inducible GAL4 (Brand and Perrimon, 1993)

pMT/V5-His B Copper inducible backbone vector Invitrogen

pMET-AblWT-myc Full length wild-type Drosophila Abl with C-terminus 6x (Forsthoefel et al., 2005) myc tag pUAUp-GFP-SprintWT N-terminally truncated wild-type Sprint with N-terminus (Jekely et al., 2005) GFP tag and C-terminus 6x His tag pUASp-YFP-Rab5WT Full length wild-type Drosophila Rab5 with N-terminus (Zhang et al., 2007) YFP tag pUASp-YFP-Rab5 (CA) Full length constitutively active Drosophila Rab5 with N- (Zhang et al., 2007) terminus YFP tag pUASp-YFP-Rab5 (DN) Full length dominant negative Drosophila Rab5 with N- (Zhang et al., 2007) terminus YFP tag pUASp-YFP-Rab7WT Full length wild-type Drosophila Rab7 with N-terminus (Zhang et al., 2007) YFP tag pUASp-YFP-Rab11WT Full length wild-type Drosophila Rab11 with N-terminus (Zhang et al., 2007) YFP tag pm-RFP-Rab5WT Full length wild-type Drosophila Rab5 with N-terminus - RFP tag and CMV promoter pm-RFP-Rab7WT Full length wild-type Drosophila Rab7 with N-terminus - RFP tag and CMV promoter pUAST-attBlox-PVRWT-GFP Full length wild-type Drosophila PVR with C-terminus (Inaki et al., 2012) GFP tag pUAST-attBlox-EGFRWT-GFP Full length wild-type Drosophila EGFR with C-terminus (Inaki et al., 2012) GFP tag pUAST-attBlox-hE-PVRWT-GFP Endodomain (C-terminus) of Drosophila PVR attached to (Inaki et al., 2012) human EGFR extracellular and transmembrane domain (N-terminus) with C-terminus GFP tag pUAST-attBlox-hE-EGFRWT-GFP Endodomain (C-terminus) of Drosophila EGFR attached (Inaki et al., 2012) to human EGFR extracellular and transmembrane domain (N-terminus) with C-terminus GFP tag

Table 2.4 Plasmids constructed during this study

Plasmid description Polypeptide encoded pUAST-RasGAPWT-LAP Full length wild-type Drosophila RasGAP with C-terminus LAP tag

pUAST-RasGAPSH2*32*-LAP Full length Drosophila RasGAP with mutations R110L, R278L with C-terminus LAP tag

pUAST-GFP-SprintWT N-terminally truncated wild-type Sprint with N-terminus GFP tag and C-terminus 6x His tag containing 20 tyrosine residues (excluding 13 GFP tyrosine residues) pBluescript-GFP-SprintWT N-terminally truncated wild-type Sprint with N-terminus GFP tag and C-terminus 6x His tag containing 20 tyrosine residues (excluding 13 GFP tyrosine residues) pUAST-GFP-SprintFXXPXD N-terminally truncated Sprint with mutation Y1056F with N-terminus GFP tag and C-terminus (FXXPXD) 6x His tag containing 19 tyrosine residues (excluding 13 GFP tyrosine residues) pUAST-GFP-SprintSH2* (FIVL) N-terminally truncated Sprint with mutation R485L with N-terminus GFP tag and C-terminus 6x His tag containing 20 tyrosine residues (excluding 13 GFP tyrosine residues) pUAST-GFP-SprintVPS9PA N-terminally truncated Sprint with mutation P1603A with N-terminus GFP tag and C-terminus (DDFLA) 6x His tag containing 20 tyrosine residues (excluding 13 GFP tyrosine residues) pUAST-GFP-SprintVPS9DA N-terminally truncated Sprint with mutation D1599A with N-terminus GFP tag and C-terminus (ADFLP) 6x His tag containing 20 tyrosine residues (excluding 13 GFP tyrosine residues) pUAST-GFP-SprintVPS9DA/PA N-terminally truncated Sprint with mutations D1599A, P1603A with N-terminus GFP tag and (ADFLA/ΔVPS9) C-terminus 6x His tag containing 20 tyrosine residues (excluding 13 GFP tyrosine residues) pUAST-GFP-Sprint443-504 (SanDI) XbaI and SanDI truncated GFP-Sprint containing 0 tyrosine residues (excluding 13 GFP tyrosine residues) and composed of residues 443-504 pUAST-GFP-Sprint443-811 (MreI) XbaI and MreI truncated GFP-Sprint containing 5 tyrosine residues (excluding 13 GFP tyrosine residues) and composed of residues 443-811 pUAST-GFP-Sprint443-1292 (PasI) XbaI and PasI truncated GFP-Sprint containing 11 tyrosine residues (excluding 13 GFP tyrosine residues) and composed of residues 443-1292 pMT-RFP-Rab5WT Full length wild-type Drosophila Rab5 with N-terminus RFP tag

pMT-RFP-Rab7WT Full length wild-type Drosophila Rab7 with N-terminus RFP tag

The pUAST-RasGAPWT-LAP and pUAST-RasGAPSH2*32*-LAP constructs were produced by NheI-XbaI double digestion of pUAST-RasGAPWT-myc and pUAST-RasGAPSH2*32*- myc (Feldmann et al., 1999), respectively, allowing subcloning of the LAP construct from pIC111 (Cheeseman and Desai, 2005) as a NheI-XbaI fragment in the place of the myc tag on the pUAST-RasGAP backbone. Primers 1, 2, 5, 6, 13-18 listed in Table 2.5 were used for sequencing and construction of the aforementioned constructs, which are shown in Figure 2.1.

The pUAST-GFP-SprintWT construct was produced by XbaI-KpnI sequential digest of pUASp-GFP-SprintWT construct (Jekely et al., 2005) allowing subcloning of the GFP- SprintWT construct into the pUAST backbone empty vector as an XbaI-KpnI fragment. The pUAST-GFP-Sprint443-504, pUAST-GFP-Sprint443-811 and pUAST-GFP-Sprint443-1292 constructs were produced as SanDI-XbaI, MreI-XbaI and PasI-XbaI fragments, respectively. Primers 1-6 listed in Table 2.5 were used for sequencing and construction of the aforementioned constructs, which are shown in Figure 2.2. To make the Sprint point mutant constructs pUASp-GFP-SprintWT (Jekely et al., 2005) was double digested with KpnI-XbaI and subcloned into the pBluescript II KS (-) backbone empty vector. Primers 3- 6, 9-12 and 19-41 listed in Table 2.5 were used for sequencing and construction of the

70 aforementioned constructs, which are shown in Figure 2.3. The Sprint point mutant constructs were then double digested with KpnI-XbaI and subcloned into the pUAST backbone empty vector (Brand and Perrimon, 1993). Primers 1 and 2 listed in Table 2.5 were used for sequencing and construction of the aforementioned constructs. Further annotations for Sprint point mutation and truncation constructs can be found in Figure S1. The amino acid residues are based on sequence of Sprint-b (UniProt accession number Q8MQW8-2).

The pMT-RFP-Rab5WT and pMT-RFP-Rab7WT constructs were produced by AgeI-XbaI sequential digest of pm-RFP-Rab5WT and pm-RFP-Rab7WT, respectively, allowing subcloning of the constructs into the pMT/V5-His B backbone empty vector (Invitrogen) as XbaI-EcoRV fragments. Primers 7 and 8 listed in Table 2.5 were used for sequencing and construction of the aforementioned constructs.

Table 2.5 Primers

Primer Primer Primer sequence 5’> <3’ number 1 HSP70 Forward GCT AAA CAA TCT GCA GTA AAG TGC 2 SV40 Reverse GTA AGG TTC CTT CAC AAA GAT CC 3 pUASp Forward CCA GTG GGA GTA CAC AAA CAG AG 4 pUASp Reverse GTG TTC TCA ACT TCA AAG GCA G 5 EGFP-C Forward CAT GGT CCT GCT GGA GTT CGT G 6 EGFP-N Reverse CGT CGC CGT CCA GCT CGA CCA G 7 pMT/V5 Forward AGA GGT GAA TCG AAC GAA AGA C 8 pMT/V5 Reverse TAG AAG GCA CAG TCG AGG 9 T7 Forward TTA ATA CGA CTC ACT AT 10 T7L Forward GTA ATA CGA CTC ACT ATA GGG CG 11 T3 Reverse ATT AAC CCT CAC TAA AG 12 T3L Reverse AAA TTA ACC CTC ACT AAA GGG AAC 13 Oligo 3 Forward CTC AAG AAG ATC AGG GAA TC 14 Oligo 7 Forward ATC TTG AAG AAC AAG GAG CG 15 Oligo 10 Reverse AAT GCG CTG CAC GGT TTC GC 16 Oligo 11 Reverse TTG ATA GAG GTA AGC ACA GG 17 Oligo 14 Forward GTT GCG AGT CTT TCG GCA GG 18 Oligo 31 Forward TTG TGG GTG ACT GCC CAT CG 19 YAEPAD to FAEPAD Forward CAA GGC AGT CCG TTT TtT GCG GAA CCA GCG G 20 YAEPAD to FAEPAD Reverse CCG CTG GTT CCG CAa AAA ACG GAC TGC CTT G 21 FIVR TO FIVL Forward GAC CTT CAT AGT GCt TGG TTC TAG CCA GCC 22 FIVR TO FIVL Reverse GGC TGG CTA GAA CCA aGC ACT ATG AAG GTC 23 DDFLP to ADFLP Forward CAG CAG CTG GGC GCC GcT GAC TTC CTG CCC GTA C 24 DDFLP to ADFLP Reverse GTA CGG GCA GGA AGT CAg CGG CGC CCA GCT GCT G 25 DDFLP to DDFLA Forward CGA TGA CTT CCT GgC CGT ACT GGT CTA TGT G 26 DDFLP to DDFLA Reverse CAC ATA GAC CAG TAC GGc CAG GAA GTC ATC G 27 DDFLP to ADFLA Forward CAG CAG CTG GGC GCC GcT GAC TTC CTG gCC GTA CTG GTC TAT GTG 28 DDFLP to ADFLA Reverse CAC ATA GAC CAG TAC GGc CAG GAA GTC AgC GGC GCC CAG CTG CTG 29 YAEPAD to FAEPAD seq primer Forward GGC CAT GAC TAA ATC CAT GAC G 30 FIVR To FIVL seq primer Forward CAT GGA GCG ACT CCT GGT CAC C 31 VPS9 NT Forward GCT ACT CTA GAC CCC TGG GAT TCT TCC CAA TGG 32 VPS9 CT Reverse GCT ACC TCG AGT CAT GGA AGG CAG CTG GAG CGC CAG 33 VPS9-RA CT Reverse GCT ACC TCG AGT CAG TGT CCT GCC AGC TGG GCG 34 XhoI Forward CTC CAT CCG ATA CGA CAA ACT CG 35 XhoI Reverse CGA GTT TGT CGT ATC GGA TGG AG 36 SgrAI Forward TCT CGA TGG AGA ACG GAG GTG G 37 SgrAI Reverse ATC CCT GCG AGG ACA GCG TGT C 38 PvuI Forward GCC AGC CTG CTG ACC AGT TTC C 39 PvuI Reverse GGA AAC TGG TCA GCA GGC TGG C 40 XmnI Reverse AAT GGT GCC ATC GTC GGA CTG C 41 BspHI Forward GAG CGG CAT GAA GAA CTA TCT G 42 Spri ex3 Forward AAC AGC AGC AGC AGC AGT GCC AG 43 Spri ex3 Reverse GAT GAC CTT TGA ACT GTG CCA C

SH2 SH3 SH2 PH C2 GAP myc

NotI NheI NotI XbaI

pUAST-RasGAPWT-myc ** pUAST-RasGAPSH2*32*-myc

2xPre SH2 SH3 SH2 PH C2 GAP 6xHis Scission GFP

NotI NheI XbaI

pUAST-RasGAPWT-LAP

pUAST-RasGAPSH2*32*-LAP ** Figure 2.1 The pUAST-RasGAP constructs. Schematic representation of RasGAP constructs used in this study. The restriction sites for each construct have been annotated on the diagrams. GFP and His are tags. PreScission is a specific site on proteins for enzymatic digestion. C2, calcium-dependent phospholipid- binding; GAP, GTPase-activating protein; PH, pleckstrin homology; SH2/3, Src homology 2/3.

GFP SH2 PRD YYYYPYY RH VPS9 RA 6xHis

SanDI MreI PasI XbaI

pUAST-GFP-SprintWT

pUAST-GFP-Sprint443-504

pUAST-GFP-Sprint443-811

pUAST-GFP-Sprint443-1292

Figure 2.2 The pUAST-Sprint truncated constructs. Schematic representation of Sprint constructs used in this study. The restriction sites for each construct have been annotated on the diagrams. PRD, proline-rich domain; RA, Ras association; RH, RIN homology; SH2, Src homology 2; VPS9, vacuolar protein sorting 9p- like; YP, phosphotyrosine.

GFP SH2 PRD YYYYPYY RH VPS9 RA 6xHis

KpnI XbaI

pUAST-GFP-SprintWT * pUAST-GFP-SprintSH2* * pUAST-GFP-SprintFXXPXD * pUAST-GFP-SprintVPS9DA * pUAST-GFP-SprintVPS9PA pUAST-GFP-SprintVPS9DA/PA ** Figure 2.3 The pUAST-Sprint point mutant constructs. Schematic representation of Sprint constructs used in this study. The restriction sites for each construct have been annotated on the diagrams. PRD, proline-rich domain; RA, Ras association; RH, RIN homology; SH2, Src homology 2; VPS9, vacuolar protein sorting 9p- like; YP, phosphotyrosine.

2.2 Media

2.2.1 Sterilisation All the glassware or non-plastic items used for liquid/media storage were autoclaved, unless otherwise stated (Sambrook et al., 1989). Solutions were sterilised by filtration through 0.22 and 0.45 μm diameter pore Millipore filters (Millipore), when autoclaving was not possible.

2.2.2 Bacterial media The bacterial growth media used were Luria-Bertani (LB) agar and LB medium, which was prepared in accordance to Sambrook and colleagues (Sambrook et al., 1989). Since the pUAST, pUASp, pGEX-KG, pBluescript II KS (-) and pMT/V5-His B vector constructs carried an ampicillin resistant gene, all the bacterial media contained ampicillin (50 μg ml- 1) for selective Escherichia coli growth.

2.2.3 Drosophila media The fly stocks were maintained in freshly made fly food with no supplements, composed of 5 L water, 390 g glucose, 360 g maize, 250 g yeast, 50 g agar, 135 ml Nipagen and 15 ml proprionic acid, which was prepared in accordance to Ashburner (Ashburner, 1989).

The Drosophila embryonic Schneider 2 (S2) cell line (Schneider, 1972) was maintained in Schneider’s medium with L-glutamine and sodium bicarbonate (Invitrogen, Gibco), which was supplemented with added 10% foetal bovine serum (FBS) (Sigma and Invitrogen, Gibco) and 1% penicillin-streptomycin mixtures (Invitrogen, Gibco).

2.2.4 Mammalian cell media Human embryonic kidney (HEK293) cell line was maintained in Dulbecco’s Modified Eagles Medium (DMEM) with 4500 mg L-1 glucose, L-glutamine and sodium bicarbonate (Sigma), which was supplemented with added 10% FBS (Sigma and Invitrogen, Gibco), 1% L-glutamine, 1% non-essential amino acid solution (Sigma) and 1% penicillin- streptomycin mixtures (Invitrogen, Gibco).

2.3 Standard solutions and buffers All the standard solutions listed below were either prepared in accordance to Sambrook and colleagues (Sambrook et al., 1989) or were established by previous members of the laboratory. All the standard solutions listed below were prepared using either sterile

(double-distilled) Milli-Q (MQ) or RO (reverse osmosis; deionised) H2O, unless otherwise stated. All alcohol based solvents used were analytical grade (Fisher Scientific). The contents of buffers used from various commercial kits are omitted from this list.

2.3.1 100x ampicillin 5 mg ml-1 ampicillin (Fisher Scientific) in 50% ethanol.

2.3.2 300x Dynasore 30 mM Dynasore (Millipore) in 100% dimethyl sulfoxide (DMSO) (Invitrogen).

2.3.3 6x Gel loading buffer (Agarose gels) 0.25% (w/v) bromophenol blue, 125 mM EDTA (pH 8.0) (Sigma) and 30% (v/v) glycerol (Fisher Scientific).

2.3.4 Phenylmethylsulfonyl fluoride (PMSF) 100 mM 100 mM PMSF (Sigma) in propan-2-ol.

2.3.5 Minimal S2 cell lysis buffer (Active) Minimal S2 cell lysis buffer (Pre-made), 1 mM PMSF, 1 mM DTT, 1x EDTA-free protease cocktail inhibitor (Roche) and with or without 5 mM Na3VO4 (Sigma).

2.3.6 S2 cell lysis buffer (Active) S2 cell lysis buffer (Pre-made), 1 mM PMSF, 1 mM DTT, 1x EDTA-free protease cocktail inhibitor (Roche) and 5 mM Na3VO4.

2.3.7 S2 cell lysis buffer (EDTA/DTT-free Active) S2 cell lysis buffer (EDTA/DTT free Pre-made), 1mM imidazole (Acros Organics), 1x

EDTA-free protease cocktail inhibitor and 5 mM Na3VO4.

2.3.8 S2 cell lysis buffer (Pre-made) 20 mM Tris-HCl (pH 7.5), 2 mM EDTA (pH 8.0), 150 mM NaCl, 10% (v/v) glycerol, 1% (v/v) Triton X-100 and 5 mM NaF (Fluka Scientific).

2.3.9 S2 cell lysis buffer (EDTA/DTT free Pre-made)

20 mM Tris-HCl (pH 7.5), 10 mM MgCl2 (BDH), 150 mM NaCl, 10% (v/v) glycerol, 1% (v/v) Triton X-100 and 5 mM NaF.

2.3.10 Minimal S2 cell lysis buffer (Pre-made) 20 mM Tris-HCl (pH 7.5), 100 mM NaCl, 10% (v/v) glycerol and 1% (v/v) Triton X-100.

2.3.11 S2 cell lysis buffer (Washing)

Pre-made S2 cell lysis buffer and 5 mM Na3VO4.

2.3.12 STE ultrasonication buffer 10 mM Tris-HCl (pH 7.5), 150 mM NaCl and 1 mM EDTA.

2.3.13 50x Tris-Acetate-EDTA (TAE) (1 L) 242 g Tris-Base, 57.1 ml glacial acetic acid (Fisher Scientific) and 100 ml 0.5 M EDTA (pH 8.0).

2.3.14 Tris-Buffered Saline (TBS) (pH 7.5) 25 mM Tris-HCl (pH 7.5) and 150 mM NaCl.

2.3.15 TBS-Tween 20 (TBST) TBS and 0.1% (v/v) Tween-20 (Sigma).

2.3.16 Phosphate buffered saline-Triton X-100 (PBS-T)

0.1% (v/v) Triton X-100 and 1x MgCl2 and CaCl2 free PBS (Sigma).

2.3.17 PBS-BT

PBS-T, 0.5% (w/v) BSA (Sigma) and 0.05% (v/v) NaN3.

2.3.18 Tris-EDTA (TE) buffer (pH 7.5) 10 mM Tris-HCl (pH 7.5) and 1 mM EDTA (pH 8.0).

2.3.19 5x Tris-glycine electrophoresis buffer 25 mM Tris-Base, 192 mM electrophoresis grade glycine and 0.1% (w/v) SDS (Sigma).

2.3.20 1000x Pervanadate

100 mM Na3VO4 and 0.3% (v/v) H2O2 (Fisher Scientific).

2.3.21 Antifade mounting solution 1.7 μg ml-1 of 4',6-diamidino-2-phenylindole (DAPI) and SlowFade gold antifade reagent (Invitrogen).

2.3.22 100x poly-L-lysine 10 mg ml-1 poly-L-lysine (Sigma).

2.3.23 10000x human epidermal growth factor (hEGF) 1 mg ml-1 hEGF (Sigma) dissolved in 10 mM acetic acid (Fisher Scientific).

2.4 Experimental methods All the experimental methods were performed in accordance to Sambrook and colleagues and Ashburner (Ashburner, 1989; Sambrook et al., 1989). When using commercial kits, the experimental procedures were performed in accordance to the manufacturer’s instructions.

2.4.1 General techniques The following general techniques were performed in accordance to the manufacturer’s instructions: BL21(DE3) competent, XL1-Blue competent and super-competent Escherichia coli transformation (Stratagene), Drosophila genomic DNA isolation (genotyping) with QIAamp DNA mini kit (QIAGEN), mini and maxi-prep plasmid purification (QIAGEN), DNA fragment amplification using Taq DNA polymerase (Roche), plasmid DNA point mutation using QuickChange lightening site-directed mutagenesis kit (Stratagene), restriction enzyme treatment of plasmid DNA (NEB, Roche and Fermentos; Thermo Scientific), antarctic phosphatase treatment of host vector plasmid DNA (NEB), ligation of plasmid DNA fragments using T4 DNA ligase (NEB), agarose gel plasmid DNA extraction (QIAGEN), DNA concentration measurements using spectrophotometer (NanoDrop) device, Effectene Transfection Reagent kit based transient S2 cell transfection (QIAGEN), ADV01 and ADV02 solution based protein concentration measurements at 590 nm absorbance wavelength (Cytoskeleton), visualising western-blot membranes using SuperSignal West Pico Chemiluminescent substrate (Thermo Scientific), SimplyBlue SafeStain solution based SDS-PAGE gel staining (Invitrogen) and DryEase Mini-Gel Drying System based SDS-PAGE gel drying (Invitrogen).

2.4.2 Agarose gel electrophoresis Agarose (Lonza) gels of 0.8-3.0% (w/v) concentrations were prepared, using 1x TAE solution and 5 μl SafeView nucleic acid stain (NBS Biologicals) in 100 ml of agarose gel solution. HyperLadder I (Bioline) was used as a molecular weight marker.

2.4.3 DNA sequence determination A solution containing 500 ng of plasmid DNA and 1 μl of 4 pmol μl-1 (μM) primer DNA (400 fmol μl-1 (fM) final concentration) was prepared and was made up to 10 μl. This solution was processed and sequenced with ABI Prism 3100 Genetic Analyser at the DNA Sequencing Facility at University of Manchester. The results were made available in ABI format chromatogram file (.ab1), which could be viewed and manipulated with FinchTV,

78 version 1.4.0 (Geospiza), CLC Sequence Viewer, version 6.7.1 (CLC bio A/S) and Serial Cloner, version 2.5 (SerialBasics).

2.4.4 E. coli master-streaking and culturing. Ampicillin resistant E. coli colonies were picked from the LBamp agar plate and were streaked on a fresh plate. The streaked plates were incubated at 37oC for 5 hours and were either stored at 4oC or subsequently sub-cultured as appropriate. The master-streaked E. coli colonies were sub-cultured in 3 ml LBamp at 37oC and 225 rpm in an incubator overnight (10-15 hours). The sub-cultured E. coli was otherwise cultured in LBamp by adding 500 μl of the sub-cultured mix to 250 ml of LBamp (1:500 dilution) and was incubated at 37oC and 225 rpm in an incubator overnight (10-15 hours) prior to plasmid DNA preparation.

2.4.5 GST-fusion protein expression and purification in Escherichia coli The BL21(DE3) competent cells (Stratagene) were transformed with pGEX-KG constructs (section 2.1.2). The colonies were sub-cultured in 10 ml LBamp at 37oC and 225 rpm in an incubator over-night (10-15 hours), which was used to inoculate 200 ml of LBamp. When the cell density reached 0.4-0.6 A600 units, 0.5 mM isopropyl -D-1-thiogalactopyranoside (IPTG; Fisher Scientific) was added to induce protein expression for 2 hours. Bacterial cells were centrifuged at 3000 g for 10 minutes before being re-suspended and washed in 10 ml of ice-cold STE ultrasonication buffer and frozen on dry ice and stored at -20oC. The bacterial cell pellets were re-suspended in 10 ml STE ultrasonication buffer, which was supplemented with added 1 mM PMSF, 5 mM DTT and 1.5% (v/v) N-lauroyl-sarcosine (Sigma), before being lysed by sonicator ultrasonic processor XL 2020 (Misonix). Triton X-100 was added at 1% (v/v) to the cell lysate, before being centrifuged at 10000 g in order to remove the cell debris. The supernatants were incubated with Protino Glutathione Agarose 4B beads (MACHEREY-NAGEL) for 1 hour at 4oC, before they were washed three times in 10 ml STE ultrasonication buffer. The beads were suspended in STE ultrasonication buffer, supplemented with added 50% (v/v) glycerol and 5 mM DTT and stored at -20oC.

2.4.6 Polymerase Chain Reaction (PCR) Master-streaked E. coli colonies or plasmids were subjected to PCR to check for the presence of the appropriate (subcloned) insert within the backbone, using the appropriate primers (Table 2.5). The reagents listed in Table 2.6 were mixed for 20 reactions (10 μl per

79 reactions) and were subjected to the thermal cycles listed in Table 2.7, depending on the amplified DNA (amplicon) fragment size (number of base-pairs), using the Thermo Hybaid PCR Express HBPX110 (Thermo Hybaid) and Eppendorf Mastercycler gradient (Eppendorf Scientific) PCR machines.

Table 2.6 PCR reaction mix

Reagents Supplier Amounts (μl) 10x PCR buffer (Taq buffer) Roche 20 dNTP (10 mM) Bioline 4 Primer 1 (10 μM) - 4 Primer 2 (10 μM) - 4 Taq polymerase enzyme Roche 0.8 MQ H2O Millipore 167.2 Total volume 200

Table 2.7 PCR reaction cycles

Reaction type Steps Number of Temperature (oC) Length of time cycles (seconds) A. Initial denaturation 1 94 120 B1. Denaturation 94 30 Fragments ranging B2. Annealing 25 50 30 0.1-0.9 Kb B3. Polymerisation 72 60 C. Final polymerisation 1 72 120 A. Initial denaturation 1 94 120 B1. Denaturation 94 30 Fragments ranging B2. Annealing 25 50 30 1.0-3.0 Kb B3. Polymerisation 68 90 C. Final polymerisation 1 72 120 A. Initial denaturation 1 95 120 B1. Denaturation 95 10 Genomic DNA B2. Annealing 60 55 15 amplification B3. Polymerisation 72 15 C. Final polymerisation 1 72 120

2.4.7 Drosophila methods The Drosophila (fly) maintenance, virgin female collection and crossing were performed in accordance to Ashburner (Ashburner, 1989). The fly stocks were crossed at 25oC but were maintained at either 18 or 25oC. The flies were aged at 25oC.

2.4.8 Drosophila brain sectioning and staining For the genetic interaction experiments, male flies were aged for 18-20 days at 25oC. The proboscis was removed and the heads were placed in 2% formaldehyde and 2% glutaraldehyde in 0.1 M HEPES Buffer (pH 7.2) for 24 hours at 4oC before being handed to the Faculty of Life Sciences EM Facility at the University of Manchester for further processing, sectioning and staining. The tissue was washed three time with double-distilled water (ddH2O) and post-fixed in reduced osmium (1% OsO4 + 1.5% K4Fe(CN)6 in 0.1 M sodium cacodylate buffer) for 1 hour. The tissue was washed three times with ddH2O and

80 en bloc stained with 1% aqueous uranyl acetate for 1 hour before being washed three times again with ddH2O and dehydrated once in 25%, 50%, 70%, 80%, 90%, ethanol and three times in 100% ethanol for 10 minutes. The tissue was washed three times in propylene oxide before being placed in propylene oxide and low viscosity (LV) resin (TAAB Laboratories Equipment) medium mixes 75:25, 50:50, 25:75 over 24-36 hours before being placed in 100% LV resin, which was changed 3-4 times over 24 hours. The tissue was then flat embedded in LV resin and placed in the oven at 60oC for 24 hours before being horizontally sectioned with Reichert-Jung (Ultracut) microtome at 2 μm thickness in a region of the head encompassing 80-180 μm thickness, starting from the base of the head near the proboscis. The sections were dried on to a microscope slide using a hotplate for 10-20 minutes before being stained with toluidine blue stain for 30 seconds and washed once with ddH2O and dab dried with flat blue roll. Images were collected on a Leica DM600B upright microscope using 10x/ 0.25 HI PLAN objective and captured using a Hamamatsu C10600-10B (ORCA-R2) digital camera through Leica MM AF Premier software, version 1.5.0 (Leica), which collectively processed the images at 0.92 μm pixel-1. Visible light band pass filter set was used to image the samples. Images were analysed using MBF-ImageJ, version 1.47b and processed using GIMP, version 2.8.4 (GIMP Development Team). In order to phenotypically quantify the percentage vacuolisation of Drosophila brain sections, the total area of central brain, lobula complex and medulla were measured and compared against the total vacuolisation area in these regions. Three heads and minimum of 3 sections were used.

2.4.9 S2 cell methods The semi-adherent Drosophila embryonic Schneider 2 (S2) cell line (Schneider, 1972) was maintained in supplemented Schneider’s medium (section 2.2.3). Cells were seeded at 1.1 5 -2 o x 10 cells cm and after 24 hours incubation at 25 C and atmospheric CO2 they were transiently transfected with Effectene (QIAGEN) at approximately 2.0 x 105 cells cm-2 or 75% confluence. All pUAST and pUASp based plasmids were co-transfected with pMT- GAL4 to allow copper-inducible expression (Klueg et al., 2002). The transfected cells were induced 24 hours later using CuSO4 at 1 mM and were lysed 48 hours later using active S2 cell lysis buffer (section 2.3.6). The crude cell lysate was spun at 13000 g at 4oC for 10 minutes to clear the lysate from debris, before being snap frozen in liquid nitrogen for storage at -80oC. If GTP-binding proteins were expressed, then active EDTA/DTT-free S2 cell lysis buffer (section 2.3.7) was used. If phosphotyrosine-dependent nature of

81 protein interactions were being assessed, then active minimal S2 cell lysis buffer (section 2.3.5) was used. To cryopreserve Drosophila S2 cells, 1 ml of cells were suspended in 45% conditioned media (90% Schneider’s medium and 10% FBS), 45% fresh media (90% Schneider’s medium and 10% FBS) and 10% DMSO at 1.0 x 107 cells ml-1.

2.4.10 Mammalian cell methods Mammalian HEK293 cell line was maintained in supplemented DMEM medium (section o 2.2.4) and was incubated at 37 C and 5% CO2. The HEK293 cells were passaged (on average) every 72 hours and upon becoming confluent, the cells were washed twice in calcium chloride (CaCl2) and MgCl2 deficient PBS (Invitrogen) and trypsinised using 1 ml of 0.05% Trypsin-EDTA (Gibco, Invitrogen), before trypsin was deactivated using 4 ml of supplemented DMEM. The cells were passaged at 1 x 106 cells ml-1. To cryopreserve mammalian HEK293 cells, 1 ml of cells were suspended in 80% DMEM, 10% FBS and 10% DMSO at 3.0 x 106 cells ml-1.

2.4.11 Drosophila and mammalian cell EGF induction

Once S2 cells were induced with CuSO4 in supplemented Schneider’s medium for 24 hours (section 2.4.9), the cells were either starved in serum free Schneider’s medium

(containing 1 mM CuSO4) for 24 hours or were left in serum-containing media for another 24 hours. Cells were collected in a 15 ml falcon tube and induced with 100 ng ml-1 hEGF (Sigma) for 20 minutes, unless otherwise stated, before being centrifuged and lysed. For the HEK293 cells, once they were plated in supplemented DMEM, they were starved 72 hours later in serum free DMEM for 24 hours before being induced with 100 ng ml-1 hEGF for 20 minutes. After hEGF incubation, the cells were washed once in CaCl2 and MgCl2 deficient PBS and lysed using active S2 cell lysis buffer.

2.4.12 Immunohistochemistry (IHC) The S2 cells were seeded on poly-L-lysine coated glass coverslips placed in a 6 well plate. The coverslips were prepared by incubating the 22x22 mm glass coverslips (Scientific Laboratory Supplies) in 100 μg ml-1 poly-L-lysine for 10 minutes before they were washed twice in MQ H2O and once in 100% analytical grade pure ethanol for 1 second before being dried. The cells were subsequently fixed in 1 ml of 4% formaldehyde (Agar Scientific) for 30 minutes and washed twice with 1 ml of PBS before being incubated and permeabilised with PBS-T for 15 minutes. Once blocked in PBS-BT for 1 hour at room temperature (RT) or overnight at 4oC, the cells were incubated with 50 μl of primary

82 antibody diluted in PBS-BT. Mouse monoclonal anti-phosphotyrosine P-Tyr-100 (PY-100, Cell Signaling) primary antibody was diluted at 1:2000, and mouse monoclonal anti c-Myc 9E10 (AbD Serotec) primary antibody was diluted at 1:200. The cells were incubated with the primary antibody for 2 hours at RT before being washed twice with 1 ml PBS-T and incubated with 50 μl of secondary antibody diluted in PBS-BT at 1:200. A number of fluorescing secondary antibodies were used including: goat anti-mouse Alexa Fluor 555 (corresponding to Texas Red emission wavelength) and goat anti-mouse Alexa Flour 647 (corresponding to TRITC/Cy3 emission wavelength) (Invitrogen). After washing twice with PBS-T, the cells were mounted with anti-fade and DAPI before being placed on a microscope slide (Thermo Scientific) for fluorescent imaging.

2.4.13 Snapshot and deconvolution microscopy For imaging fluorescing samples three sets of microscopes were used including two snapshot wide-field and one deconvolution microscope. For the first snapshot microscope images were collected on an Olympus BX51 upright microscope using a 100x/ 1.30 UPlan Fln objective and captured using a CoolSNAP HQ camera (Photometrics) through MetaVue software, version 7.1.0.0 (Molecular Devices), which collectively processed the images at 0.07 μm pixel-1. Specific band pass filter sets for DAPI, FITC, Cy5, Cy3 and Texas Red were used to prevent bleed through from one channel to the next. Images were then processed and analysed using MBF-ImageJ, version 1.47b. For the second snapshot microscope images were collected on a Leica DM600B upright microscope using a 100x/ 1.40 HCX PL APO objective and captured using a Hamamatsu C10600-10B (ORCA-R2) digital camera (Hamamatsu) through Leica MM AF Premier software, version 1.5.0 (Leica), which collectively processed the images at 0.09 μm pixel-1. Specific band pass filter sets for DAPI, L5 (FITC/GFP), Cy5, Cy3 and Texas Red were used to prevent bleed through from one channel to the next. Images were then processed and analysed using MBF-ImageJ, version 1.47b. For the Z-stack collection and deconvolution microscopy, images were acquired on a Delta Vision Core (Applied Precision) restoration microscope using a 100x/ 1.40 UPlan S Apo objective and the Sedat filter set Chroma 89000, which produced the images at 0.13 μm pixel-1. The images were collected using a CoolSNAP HQ (Photometrics) camera with a Z optical spacing of 0.2 μm. Raw images were then deconvolved using the softWoRx software, version 4.0.0 (Applied Precision), allowing production of the 3D projections as well as the projection free single plane orthogonal views of these deconvolved images. The deconvolved Z-stacked (R3D_D3D.dv) files

83 produced by SoftWoRx software were further processed using Imaris, version 7.5.2 (Bitplane) in order to produce 3D isosurface model of the Z-stack imaged cells.

2.4.14 RNAi knockdown For the RNAi knockdown experiments, the pUAST-GFP-SprintWT transfected S2 cells were seeded in a 6 well plate. The cells were serum starved 24 hours later in 4 ml of serum free Schneider’s medium for 1 hour, then 300 μl of the transfected cell suspension was added to the well of a 24 well plate that contained 2000 ng dsRNA (Table 2.8) (Brown,

2010) in 10 μl of ultrapure H2O (Sigma). The dsRNAs were incubated with the cells for 5- 7 hours, allowing their uptake, before 300 μl of 20% FBS supplemented Schneider’s medium was added. The cells were then induced 24 hours later with 1 mM CuSO4 for 48 hours to allow GFP-Sprint expression, and then processed for immunohistochemistry, imaging, lysis and western-blotting.

Table 2.8 dsRNA list

Gene name CG number Unique (BKN) number Strand Primer sequence 5’> <3’ vap CG9209 BKN21461 Forward ACCCTTCGGTTTGGTCTTCT Reverse AGAAAGGCGTGCGCTATTTA Gap1 CG6721 BKN20460 Forward GATTAACTGCCTGCTGCTCC Reverse GAGGGCATAGATCTGGGACA Gapex-5 CG1657 BKN20094 Forward TGAGTGAGCACGAGATTTGG Reverse GTCGCTGAAGCAATTTCACA Abl CG4032 BKN22375 Forward GCACCAGATTAGGGTGCTTC Reverse CCAGGAGGCCAAGTTTAACA Sprint CG33175 BKN22697 Forward CGTTGTCGTAGGAGGAGAGC Reverse GCAGCCTGGACGAACTAAAG Rabex-5 CG9139 BKN25419 Forward ACTTGTCGGATCAAAGTGGG Reverse TTCACACAGCTGCAGGATTC Ras64B CG1167 BKN27302 Forward TCGATGAGATCCCCAAGTTC Reverse ATAAAAACAAATGGCCTGCG Ras85D CG9375 BKN28029 Forward GCTTTCGGTAAGAGTCCTCG Reverse GCACAGTCACCCACACAAAC Pvf2 CG13780 BKN30608 Forward AGTTTGCTCGCTTATCCGAA Reverse GCAACTACGGGGCATATGAT Shibire CG18102 BKN21495 Forward ATCGAAGATGAAACGGATCG Reverse TAGTTGTTGGTTGAGCACGC Rab5 CG3664 BKN22991 Forward GCTCTCCTGGTACTCGTGGA Reverse CATCCACACTCAGCAGCAAT Rab7 CG5915 BKN28849 Forward TGTCCGGACGTAAGAAATCC Reverse GTCGTTGACCACCACCTCTT Pvr CG8222 BKN21626 Forward TTCTGTCCAATCGGTTCTCC Reverse CAACGATATCCGACAGGGAT GFP - BKN70048 Forward ACCCTCGTGACCACCCTGACCTAC Reverse GACCATGTGATCGCGCTTCTCGT Renilla - BKN70063 Forward ATGACTTCGAAAGTTT Reverse GATGCTCATAGCTATA

2.4.15 Endocytosis assay For the fluid phase endocytosis assay, untransfected cells attached to poly-L-lysine coated coverslips were incubated with 500 nM Dextran, Texas Red®, 3000 MW (Invitrogen) for 1 hour before being fixed and prepared for imaging. Transfected or untransfected S2 cells

84 were treated with the dynamin-GTPase inhibitor, Dynasore (Millipore) for 24 hours (at 100 μM dissolved in DMSO) before being fixed and prepared for imaging (Macia et al., 2006).

2.4.16 Puncta quantification and determining co-localisation Using fluorescent microscopy (section 2.4.13) the number of stained vesicle-like (puncta) structures observed per cell were manually counted. This was performed based on the pixel density (brightness) of the imaged cells containing stained puncta. Puncta co-localisation was also measured manually by viewing multiple channels as unmerged and merged, allowing visual determination of co-localisation. MBF-ImageJ, version 1.47b, was used to view the images and aid the procedure. Punca count automation was not possible due to the high fluorescent background of the images, making arbitrary pixel threshold based definition of puncta impractical. The fluorescent imaging based experiments were carried out once due to the time constraints.

2.4.17 SDS-PAGE and western-blotting The 4x sample buffer (Invitrogen) and 10x reducing agent (Invitrogen) were added to the protein samples, which were then heated to 95oC for 10 minutes. The samples were immediately loaded on 4-12% Bis-Tris SDS-Polyacrylamide Gel Electrophoresis (SDS- PAGE) gels (Invitrogen) in accordance to Invitrogen instructions and were run using 1x MOPS (20x stock) running buffer (Invitrogen). ColorBurst (C1992/C4105) was used as a protein molecular weight marker (Sigma). The samples were electroblotted using a semi- dry blotter (BIO-RAD) using Tris-glycine electrophoresis buffer (section 2.3.19) on PVDF membrane (Sigma) (Sambrook et al., 1989). The membranes were blocked for 1 hour at RT in 5% (w/v) bovine serum albumin (Sigma) (BSA)-TBST or 5% (w/v) skimmed-milk (Marvel) (SM)-TBST, depending on the primary antibody used. The blocked membranes were then stained with primary antibody overnight at 4oC using either mouse monoclonal anti-phosphotyrosine P-Tyr-100 (PY-100, Cell Signaling) antibody at 1:5000 dilution in 5% (w/v) BSA-TBST, or mouse monoclonal anti-phosphotyrosine PY-20 (Sigma) antibody at 1:2000 dilution in 5% (w/v) BSA-TBST, or mouse mixed monoclonal (clones 7.1 and 13.1) anti-GFP (Boehringer Mannhem) antibody at 1:1000-1:3000 dilution in 5% (w/v) SM-TBST, or mouse monoclonal anti c-Myc 9E10 (AbD Serotec) antibody at 1:500- 1:1000 dilution in 5% (w/v) SM-TBST, or rabbit anti-mitogen activated protein kinase (MAPK) (Sigma) antibody at 1:5000 dilution in 5% (w/v) SM-TBST, or mouse monoclonal anti-diphosphorylated (dp)ERK 1 & 2 (dpERK) antibody at 1:2000 dilution in 5% (w/v) BSA-TBST, or rabbit monoclonal anti-EGFR (D38B1; Cell Signaling) antibody

85 at 1:1000 dilution in 5% (w/v) SM-TBST, or mouse anti-Drosophila tubulin (Dtub) antibody at 1:5000-1:500000 dilution in 5% (w/v) SM-TBST. After three 10 minute washes in TBST, the membranes were incubated with horse anti-mouse (at 1:1000-1:3000 dilution) or goat ant-rabbit (at 1:2000-1:5000 dilution) IgG HRP-linked secondary (Cell Signaling) antibody for 1 hour at RT before being washed three times in TBST. The membranes were then visualised with chemiluminescent system on BioMax XAR imaging film (Kodak). The films were then digitised using a flatbed scanner (HP).

2.4.18 Phospho-protein enrichment

In addition to using 5 mM NaF (serine/threonine phosphatase inhibitor) and 5 mM Na3VO4 (tyrosine phosphatase inhibitor) solutions in cell lysis buffers, some cells were treated with 100 μM pervanadate for 60 minutes before lysis. Pervanadate is a highly potent and irreversible tyrosine phosphatase inhibitor, which is produced by peroxidation of Na3VO4 (Chang et al., 2008). Pervanadate treatment of cells highly enriches for tyrosine phosphorylation of proteins in live cells.

2.4.19 S2 cell expressed fusion protein pulldown and purification The cleared cell lysate was rotated with GFP-Trap agarose beads (Chromotek) for 2 hours at 4C before being washed three to five times with washing S2 cell lysis buffer. The GFP- Trap beads were heated in sample buffer (Invitrogen) and reducing agent (Invitrogen) and the supernatant was loaded on 4-12% Bis-Tris SDS-PAGE gels (Invitrogen), allowing gel staining (SimplyBlue SafeStain, Invitrogen) and western-blotting of the pulled down proteins. The samples were electroblotted and stained as described in section 2.4.17. For some experiments, proteins were affinity purified using His-Select Nickel Affinity Gel beads. His tagged expressing S2 cells were lysed in EDTA/DTT-free S2 cell lysis buffer and were incubated with His-Select Nickel Affinity Gel beads (MACHEREY-NAGEL) for 1 hour at 4oC before being washed three to five times with EDTA/DTT-free washing S2 cell lysis buffer. For some experiments, proteins were pulled down using GST-SH232 bound Protino Glutathione Agarose 4B beads. S2 cells were lysed in minimal S2 cell lysis buffer, incubated at 30C for 30 minutes and were incubated with GST-SH232 bound Protino Glutathione Agarose 4B beads (MACHEREY-NAGEL) for 2 hours at 4oC before being washed five times in minimal S2 cell lysis buffer. After adding sample buffer, reducing agent and heating, proteins were loaded on 4-12% Bis-Tris SDS-PAGE gels, allowing gel staining and western-blotting of the pulled down proteins.

2.4.20 Protein mass spectrometry The purified protein complexes were loaded on 1 mm thick 4-12% SDS-PAGE gels and stained using SimplyBlue SafeStain solution sensitive protocol (Invitrogen). Protein bands at the appropriate molecular weight were cut and placed in 1 ml MQ H2O, before being handed to the Biological Mass Spectrometric Core Research Facility at the University of Manchester for mass spectrometric analysis. Bands of interest were dehydrated using acetonitrile followed by vacuum centrifugation. Dried gel pieces were reduced with 10 mM dithiothreitol and alkylated with 55 mM iodoacetamide. Gel pieces were then washed alternately with 25 mM ammonium bicarbonate followed by acetonitrile. This was repeated and the gel pieces were dried by vacuum centrifugation. Samples were digested with either trypsin (specificity: C-terminal to Asp/Glu), trypsin & Asp-N (specificity: N- terminal to Asp/Cys), trypsin & Glu-C (V8; specificity: C-terminal to Asp/Glu) or Elastase (specificity: none) overnight at 37C. Digested samples were analysed by LC-MS/MS using either (a) a NanoAcquity LC (Waters, Manchester, UK) coupled to a LTQ Velos (Thermo Fisher Scientific, Waltham, MA) mass spectrometer with peptides concentrated on a pre-column (20 mm x 180 μm inner diameter, Waters) or (b) an UltiMate® 3000 Rapid Separation LC (RSLC, Dionex Corporation, Sunnyvale, CA) coupled to a LTQ Velos Pro (Thermo Fisher Scientific, Waltham, MA) mass spectrometer or (c) an UltiMate® 3000 Rapid Separation LC (RSLC, Dionex Corporation, Sunnyvale, CA) coupled to an Orbitrap Elite (Thermo Fisher Scientific, Waltham, MA) mass spectrometer. For the LTQ Velos Pro and Orbitrap Elite samples were desalted first before peptide separation. The peptides were separated using a gradient from 92-99% A (0.1% v/v formic acid in water) and 1-8% B (0.1% v/v formic acid in acetonitrile) to 25-33% B, in 45 min at 200-300 nL min-1, using a 75 mm x 250 μm inner diameter 1.7 μM BEH C18, analytical column (Waters). For the Orbitrap Elite and Velos, the mass tolerance for precursor ions were set at 5 ppm and 0.5 Da, respectively and the mass tolerance for fragment ions were set at 0.5 Da. Peak-lists were generated using Mascot Daemon 2.2 (Matrix Science UK) employing the extract_msn.exe utility and the data were searched using Mascot 2.2 (Matrix Science UK), against the Uniprot database with taxonomy of Drosophila (fruit flies) selected. Data were further validated and searched using Scaffold (Proteome Software, Portland, OR), which models the score distribution of the entire dataset of spectra (Humphries et al., 2009). With no MuDPIT analysis technique employed, Mascot generated search (.dat) files were queried against uniprot.v14_0 database, allowing pyro-

87 carbamidomethylation of cysteine as a fixed modification and tyrosine, serine and threonine phosphorylation as well as methionine oxidation as a variable modification. The data were viewed using Scaffold, version 3.00.08, and the results were analysed at 99% protein identification probability tolerance threshold, with minimum number of 2 peptides at 50% identification probability acceptance threshold for at least one spectra required for protein identification. The Scaffold assigned false discovery rate (FDR) of the proteins within the dataset was 0.2%. Samples were validated for phospho-modifications using either Mascot Delta (MD)-score (Savitski et al., 2011) or manual inspection of the data when applicable.

2.4.21 Data deposition Mascot generated mass spectrometric data (.dat) were converted into PRIDE Extensible Markup Language (XML) files (.xml) using PRIDE Converter 1 (Barsnes et al., 2009; Barsnes et al., 2011). The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium (http://proteomecentral.proteomexchange.org) via the PRIDE partner repository (Vizcaino et al., 2013) with the dataset identifier PXD000176 and DOI 10.6019/PXD000176 (PRIDE accession numbers 28714-28755). The deposited information complied with Minimal Information About a Proteomics Experiment (MIAPE) guidelines available at http://psidev.info webpage. The original Scaffold search output file is available on request.

2.4.22 Mascot Delta (MD)-score Although some peptide residues were manually validated for phospho-modifications using the precursor mass and fragmentation (MS/MS) spectra of peptides in Scaffold and Mascot, most samples were assessed and validated for phospho-modifications using MD- score. MD-score reflects the difference of Mascot ion scores between the highest and second highest ion scores (score differentials) for candidate phosphorylation sites on an identical peptide sequence in a database search. This ability to call sites correctly improves with increasing distance of two candidate sites within a peptide sequence (Savitski et al., 2011). In some cases the fragmentation information available was insufficient to distinguish the exact site of phosphorylation in some potential sites of phosphorylation close or adjacent to each other.

2.4.23 Quantification using spectral counting The unweighted spectral counts assigned to each identified protein by Scaffold were used to calculate the relative abundance of each protein, across biological samples, using Scaffold-assigned total number of identified spectra per biological sample (normalised spectral counts). This was done for each biological sample (RasGAPWT, RasGAPSH2*32* and GFP) and their two independent repeats and the results were expressed as an average for each protein (mean normalised spectral counts) (Humphries et al., 2009; Lundgren et al., 2010).

2.4.24 Hierarchical clustering The quantitative data (mean normalised spectral counts) were hierarchically clustered using gene cluster 3.0, version 1.50 (C Clustering Library), which hierarchically clustered protein hits on the basis of uncentered Pearson’s correlation and calculated the distances between hits using a complete-linkage matrix. The gene cluster file outputs (.cdt and .gtr) were then visualised using Java TreeView, version 1.1.6r2, which itself produces a file output (.jtv) (Humphries et al., 2009). The semi-quantitative relative abundance heat map produced in Java TreeView was further manipulated using Adobe Illustrator CS5, version 15.0.2 (Adobe).

2.4.25 Interaction network analysis The open-source platform Cytoscape, version 2.8.3 (Cline et al., 2007; Shannon et al., 2003; Smoot et al., 2011), was used to visualise Drosophila RasGAP protein-protein interaction network utilising obtained MS results and with the data searched against Drosophila interaction database, DroID (Pacifico et al., 2006; Yu et al., 2008). The data in DroID is comprised of protein-protein (biochemical) interaction datasets (such as Finley lab two-hybrid, Curagen two-hybrid, Hybrigenics two-hybrid, Perrimon coAP complex, DPiM coAP complex and other databases), genetic-interaction datasets and predicted interaction datasets (such as predictions from C. elegans, yeast, human, transcription factor-gene and miRNA-gene) (Pacifico et al., 2006). The protein-protein interaction network produced in Cytoscape was further manipulated using Adobe Illustrator CS5, version 15.0.2 (Adobe).

2.4.26 Regular expression database analysis Using custom program written in Python programming language, by the Biological Mass Spectrometric Core Research Facility at the University of Manchester, regular expression database analysis from the UniProt Drosophila sequences was performed.

2.4.27 In silico proteomics and interactomics Many publicly available online tools and databases including NCBI, Ensembl, FlyBase, ExPASy, ProteinProspector, PeptideAtlas, PhosphoPep, NetPhos, DroID, BrainTrap, Virtual Fly Brain, SOSUI, SMART, Pfam, Scansite motif scanner, NEB cutter, BLAST, Rasmol (FirstGlance in Jmol), UniProt, ClustalW2, PRIDE, Invitrogen Fluorescence SpectraViewer, PeptideCutter, ProteinCutter, BioGRID, PINA, PhosphoSitePlus and Signal Peptide database were used to analyse the Drosophila DNA, transcript and protein sequences, structures and expression patterns. These databases and programs were executed and viewed on Microsoft Windows Vista and Apple OS X platforms.

CHAPTER 3: Mass spectrometric identification of RasGAP SH2-dependent interacting partners in Drosophila

3.1 Introduction

3.1.1 Purification of protein complexes Regulated protein-protein interactions are crucial in cell signalling from the extracellular environment to the intracellular environment. In order to identify these interactions, many techniques such as the yeast two hybrid system, protein pulldown studies and bioinformatics have been employed. Amongst these techniques immunological based protein pulldown methodologies, employing antibodies raised against the protein of interest or against protein tags, has been proven to be one of the most successful approaches in protein interaction studies (Kocher and Superti-Furga, 2007). Once isolated, the composition of the protein complexes can be determined using mass spectrometry.

3.1.2 Mass spectrometric protein identification One of the objectives of protein pulldown and purification is to prepare the protein complex for identification using mass spectrometry (MS) (Cheeseman and Desai, 2005). The purified protein complex is initially separated by SDS-PAGE. Upon Coomassie blue staining of the protein bands of interest, they are extracted from the gel and subjected to enzymatic digestion in order to produce peptide fragments (Rosenfeld et al., 1992; Shevchenko et al., 1996). In modern MS identification techniques solvent containing analytes (peptides) are inserted into the MS machine once peptides are separated through reversed-phase liquid chromatography (RP-LC) techniques. In MS these analytes form a very small-droplet at the end of a fine silica capillary, holding a high positive potential, which confers a large charge density (z/m) to these droplets. Once the solvent is removed by heat or energetic collision with a dry gas, the charge density of these droplets increases at their surface, leading to a coulombic explosion (Rayleigh limit). The process above can be repeated until non-fragmented, protonated or cationic multicharged ions of the analytes are liberated. This process is referred to as electrospray ion (ESI) but since the spectra produced only contain ionised molecules with very little fragmentation data (required for structural characterisation of peptides), induced fragmentation is performed in collision

91 chambers. In a collision chamber ions collide with a target gas such as argon or nitrogen. This causes the kinetic energy of the gas to convert to vibrational energy in the ions, which subsequently leads to their fragmentation (cleavage of amino acids one-by-one from their peptide backbone). These fragmented (daughter) ions can then enter multiple mass analysers arranged in tandem (MS/MS), which allows a better knowledge of the ion, hence achieving a partial sequencing of the peptide (Rouessac and Rouessac, 2007; Shevchenko et al., 1996). The MS/MS technique creates a series of small fragments that will then be measured (in m/z) and the results are compared against a database, which allows identification of the peptide and its sequence (Bauer and Kuster, 2003).

3.1.3 Rationale of the study Since RasGAP SH2 domains have been identified to be important in Drosophila neuronal survival it was proposed that RasGAP SH2-dependent interacting partners are likely to play a role in Drosophila neuronal survival. The approach taken in this study to identify RasGAP SH2-dependent interacting partners was to express a tagged form of RasGAP in S2 tissue culture cells, affinity purify the tagged RasGAP using an antibody against the tag, and then analyse the RasGAP protein complexes by mass spectrometry to identify co- purifying proteins. The protein tag chosen was the localisation and purification (LAP) tag, which is a tandem affinity tag consists of GFP and 6xHis (Cheeseman and Desai, 2005). Proteins fused to this tag can be pulled down using a high affinity reagent for GFP tags (GFP-Trap) (Rothbauer et al., 2008; Schmidthals et al., 2010) and a second affinity purification using immobilised metal ion affinity chromatography can be done if required. S2 cells were chosen due to their ease of small and large scale culture for MS-grade protein purification and due to the fact that they represent a good in vitro model for Drosophila proteomics and interactomics (Brunner et al., 2007; Chang et al., 2008; Krishnamoorthy, 2008). In this chapter a number of tyrosine-phosphorylated RasGAP SH2-dpenedent interacting partners were identified, which were subjected to mass spectrometric identification. Amongst these proteins were Sprint, Dscam, east and BubR1. This study has also identified a number of novel RasGAP and Sprint phosphorylation sites, which is an indication of their complex molecular and cellular regulation.

3.2 Results

3.2.1 Drosophila RasGAP SH2 domains interact with a number of tyrosine- phosphorylated proteins in S2 cells In order to identify RasGAP SH2 interacting proteins, RasGAP was tagged at its C- terminus with the localisation and affinity purification (LAP) tag, so that RasGAP protein complexes could be affinity purified after expression of the tagged protein in S2 cells. LAP-tagged RasGAPSH2*32*, carrying inactivating substitutions in both SH2 domains, was used as a control to identify SH2-phosphotyrosine dependent interactions. RasGAP-LAP proteins were expressed in S2 cells, pulled down from lysates with anti-GFP antibodies (GFP-Trap), and western-blotted using anti-phosphotyrosine antibody (PY-100) and confirmed with a second anti-phosphotyrosine antibody (PY-20; data not shown). Five distinct tyrosine-phosphorylated protein bands at approximately 220, 200, 130, 110 and 60 kDa were observed in the RasGAPWT pulldown (Figure 3.1A, filled arrowheads). These bands were entirely absent in the RasGAPSH2*32* and GFP control pulldowns, indicating that these phosphotyrosine-containing proteins interact with RasGAP in an SH2-dependent manner. The 130 kDa phosphotyrosine protein band (Figure 3.1A, large filled arrowhead) corresponded to the tagged RasGAPWT molecular weight and it may represent tyrosine- phosphorylated RasGAP. This experiment provided evidence that RasGAP interacts with a number of tyrosine-phosphorylated proteins in an SH2-dependent manner and justified identification of these interacting partners using MS.

RasGAP GFP

WT SH2*32* EUUP EUP E P A 220

100 GFP-Trap pulldown WB: -pTyr 60

1 2 3 4 5 6 7 8 9 B 220

100

60 GFP-Trap pulldown WB: -GFP 45

1 2 3 4 5 6 7 8 9 C 60 Cell extract 45 WB: -Tubulin

1 2 3 4 5 6 7 8 9

Figure 3.1 RasGAP SH-dependent phosphotyrosine protein interaction and tyrosine phosphorylation. In order to identify RasGAP SH2-dependent interacting partners in Drosophila, the Drosophila S2 cells were transfected with wild-type full length RasGAP (RasGAPWT) and as a negative control, with double SH2 inactivated full length RasGAP (RasGAPSH2*32*) (large filled arrowhead, panels A and B), or with a GFP tag construct (GFP) (small filled arrowhead, panel B). 4 mg of protein extract from each of the transfected cell cultures was pulled down (P) with GFP-Trap (P; lanes 3, 6 and 9), separated by SDS-PAGE and immunoblotted with (A) anti-pTyr, (B) anti-GFP and (C) anti-Dtub antibodies (small filled arrowhead, panel C). Equal amounts of total extract (E) and unbound (U) were also analysed by immunoblotting. Filled arrowheads indicate the positions of tyrosine-phosphorylated proteins specifically interacting with RasGAPWT.

3.2.2 Mass spectrometric identification of Drosophila RasGAP interacting partners in S2 cells using spectral counting After identifying several RasGAP SH2-dependent interacting partners by western-blot, seven independent set of experiments were carried out in which the LAP-tagged wild-type full length RasGAP (RasGAPWT) and double SH2 inactivated full length RasGAP (RasGAPSH2*32*) were pulled down using GFP-Trap. After running the pulldowns on the SDS-PAGE gel, several bands at various molecular weights corresponding to the pTyr western-blot bands were excised (Figure 3.2) and processed for MS. The results (unweighted spectral counts) from two independent but parallel experiments (experiments 2 and 3, Table 3.3) were then quantified as mean normalised spectral counts, defined as the total number of fragmentation spectra that map to peptides of a given protein (Lundgren et al., 2010) and hierarchically clustered on the basis of uncentered Pearson’s correlation, allowing label-free semi-quantitative relative protein abundances across different biological samples to be calculated. This allowed representation of the data as a semi- quantitative relative abundance heat map (Figure 3.3). As shown in Figure 3.3, three main clusters of proteins were identified. Within the RasGAP clusters, there are several subclusters including 155 RasGAP SH2-independent interacting partners (Figure 3.3A1 and A2) and 115 RasGAP SH2-dependent interacting partners (Figure 3.3B). As shown in Figure 3.3B, many proteins such as Dscam, Sprint, east, BubR1, Ben (CG5530) and Sickie (sick) have very high relative abundances in RasGAPWT but not in RasGAPSH2*32* samples, which indicates a high degree of confidence in their tendency to associate with RasGAP SH2 domains. In addition, all the RasGAP SH2-dependent interacting partners have a very high Pearson’s correlation (0.95), which indicates their similar association tendency towards the RasGAP SH2 domains.

After analysing the protein sequences of RasGAP interacting proteins, using regular expression database analysis, 14 RasGAP SH2-dependent interacting proteins (underlined in Figure 3.3B) were found to contain the YXXPXD RasGAP SH2-binding consensus sequence (Table 3.1). This constitutes 12.2% of RasGAP SH2-dependent interacting proteins, which is a much higher percentage of YXXPXD containing proteins than found in RasGAP SH2-independent interacting proteins (7.1%) or the entire Drosophila proteome (5.1%) (Table 3.2). This indicates that RasGAP SH2 domains enrich for proteins containing the YXXPXD RasGAP SH2-binding consensus sequence. If stringency is applied to the 14 YXXPXD containing RasGAP SH2-dependent interacting proteins and

95 only proteins with higher than 4 (0.001 normalised) spectral counts are assessed, 5 out of 14 proteins contain YXXPXD (highlighted in bold in Figure 3.3B), which equates to 35.7% YXXPXD enrichment. This suggests a correlation between the abundance of RasGAP SH2-dependent interacting partners and YXXPXD enrichment.

) ) P ( (P ) ) l P B ( (P e

l * W * G B e 2 2 3 3 W G * * ) ) T T 2 2 H H P P W W ( ( S S l P P P P B e A A r A A r r e e W G e G G k k G G k k k k s s n r s s n r P P n r a a la a a a la a F F la a R R B M R R B M G G B M A 220 B 220 C 220 a b c d e f g h i j k l 100 m 100 n 100 o

60 60 60 p q r s t u 45 45 45

30 30 30

Identified SH232WT interacting proteins Identified SH2*32* interacting proteins Identified non-SH232 interacting proteins

a) Dscam, spri, sick, east b) CG31738, brm ab) CG15415, Rme-8, CG17514

d) Dscam, spri, CG31998 e) de) Dcr-2, Ubi-p63E, Cp190, CG10080, c11.1

g) B4, bon h) gh) sesB, CG10882, fs(1)Yb

j) CG5530 k) jk) ref(2)P, Msr-110

m) n) mn) CG17528, mal, Lasp, mit(1)15

p) q) pq) Gdh, Sply, Src42A

s) t) st) Act-5C, CG10576, exba

Figure 3.2 RasGAP pulldown SDS-PAGE gel loading. Drosophila S2 cells were transfected with RasGAPWT, RasGAPSH2*32* (large arrowhead, panels A, B and C) and GFP control. 4 mg of S2 cell lysate was used to pulldown (P) LAP and GFP tagged constructs using GFP-Trap. The (P) fraction were western- blotted using anti-pTyr antibody and loaded on SDS-PAGE gels and coomassie stained before several bands (a-u) from two independent but parallel experiments were excised and subjected to MS analysis. Arrowheads indicate the relevant protein bands.

WT SH2*32* * 2 RasGAP RasGAP GFP *3 2 T H Bicoid mRNA stability factor W S P P E3 ubiquitin-protein ligase hyd A A Nuclear pore complex protein Nup88 A G G B s s P Phospholipase A2 activating protein homolog a a F R R G CG5482 DShc bon Glycoprotein-N-acetylgalactosamine 3-beta-galactosyltransferase 1 CG32789 B4 DNA replication licensing factor Mcm2 RE66582p Putative elongator complex protein 4 CG6084 gro ps

Protein dopey-1 homolog 5

9 Nuclear GTP binding protein .

0 E3 ubiquitin-protein ligase UBR1 Nucleosome assembly protein NAP-1 GA15890 Probable ubiquitin carboxyl-terminal hydrolase FAF JhI-26

CG14998 2

6 Gint3 .

0 Vrp1 LD32453p CG30122 ik2 Conserved oligomeric Golgi complex subunit 5 CG13096 east CG31436 sick CG5530 CG6807 Dscam BubR1 Dscam cup CG11376 Nop56 CG4030 CG3223 GA21593 Serine/threonine-protein phosphatase 4 regulatory subunit 2 CG31694 Protein odr-4 homolog 1 bcn92 DNA polymerase delta small subunit

Ero1-like protein 5

7 CG1486 .

0 CG5604 CG13185 p47 CG5167 CG1646 coilin CG3493 CG6448 Eukaryotic peptide chain release factor subunit 1 Exocyst complex component 1 CG6630 Nup133 Ack Actin-binding protein anillin CG5913 sip2 Replication factor C subunit 1 Exocyst complex component 5 Exocyst complex component 6

9 atl 7

. QKR58E-2 0 CG18522 nito Guanine nucleotide-releasing factor 2 casp CG1104 pit CG32782 GA14464 Probable cysteine desulfurase, mitochondrial Serine/threonine-protein kinase 38-like Maternal protein tudor CG6701 2 CG8465 CG2818 CG7261 Aats-val

7 Ubiquitin carboxyl-terminal hydrolase 64E 6 . Exportin-1 0 CG31755 CG14299 penguin Conserved oligomeric Golgi complex subunit 2 sprint CG31998 GA15015 Signal transducer and transcription activator CG6841 l(1)1Bi Lon protease homolog PLC-gamma D CG6192 CG2982 Idgf3 REST corepressor

CG9139 5

7 GA16037 .

0 Zinc finger protein hangover CG6995 pyd Mat89Bb Spectral O-glycosyltransferase count 0.00 0.01 ofs

Figure 3.3 Hierarchical clustering of RasGAP interacting proteins. (A) The figure displays the relative abundance (mean normalised spectral counts) of each protein between the different MS analysed biological samples, which have been hierarchically clustered on the basis of uncentered Pearson’s correlation and displayed at 6.33% saturation at 0.01 maximum spectral count. The protein hits are from Drosophila melanogaster (DROME) and Drosophila pseudoobscura (DROPS) species. The proteins with high relative

98 abundance are indicated with dark blue and proteins with low relative abundance are indicated with light blue. There are three main clusters of proteins identified, including clusters of proteins enriched to all three (RasGAPWT, RasGAPSH2*32* and GFP) conditions (correlation 0.62 and 0.79), clusters of proteins enriched to RasGAP (correlation 0.75 and 0.67) and clusters of proteins enriched to GFP control (correlation 0.75). Within the RasGAP clusters, there are several subclusters including (A1-A2) 155 RasGAP SH2-independent interacting partners and (B) 115 RasGAP SH2-dependent interacting partners. RasGAP SH2-dependent interacting proteins containing the YXXPXD RasGAP SH2-binding consensus sequence are underlined and proteins with higher than 0.001 normalised spectral counts are highlighted in bold. The complete cluster output is listed in Table S1 and the raw data can be accessed in PRIDE under accession numbers 28714- 28755.

Table 3.1 RasGAP SH2-dependent interacting proteins containing YXXPXD consensus sequence

Name Symbol Annotation UniProt accession YXXPXD tyrosine Sequence match symbol number residue position SH2, poly-proline containing spri CG34414 SPRI_DROME 1070 (GSP)YAEPAD(ALR) Ras interactor (Sprint) enhanced adult sensory east CG4399 O46048_DROME 1536 (MVA)YDGPTD(SNS) threshold Down syndrome cell adhesion Dscam CG17800 Q0E9M3_DROME 1900 (YGG)YGQPYD(HYG) molecule Down syndrome cell adhesion Dscam CG17800 Q0E9I5_DROME 1893 (YGG)YGQPYD(HYG) molecule Bub1-related kinase BubR1 CG7838 A1Z6I7_DROME 441 (NFA)YSKPQD(LDE)

DNA polymerase delta small CG12018 CG12018 DPOD2_DROME 105 (PRQ)YSDPED(KIV) subunit coilin coil CG8710 A1Z7A8_DROME 57 (SDG)YLPPRE(SIK)

Activated Cdc42 kinase Ack CG14992 Q9VZI2_DROME 48 (AHF)YVLPDD(LER) pitchoune pit CG6375 DDX18_DROME 508 (WIV)YDPPDD(PRE)

Maternal protein tudor tud CG9450 TUD_DROME 1633 (PAQ)YVHPID(QLS) super sex combs sxc CG10392 Q7KJA9_DROME 855 (TRR)YMLPDD(AVV) groucho gro CG8384 A4V3F6_DROME 358 (PGA)YQRPAD(PYQ)

- CG31998 CG31998 Q9V4B6_DROME 1031 (QEP)YQRPYD(TGS)

- CG5482 CG5482 Q7K3D4_DROME 157 (GES)YLVPPD(AHL)

Table 3.1 RasGAP SH2-dependent interacting proteins containing the YXXPXD consensus sequence. The position of the consensus tyrosine within each protein is marked.

Table 3.2 RasGAP SH2-dependent YXXPXD consensus binding sequence enrichment

Query Background sequence RasGAP SH2-independent RasGAP SH2-dependent Total number of proteins 18797 155 115 YXXPXD sequence match 996 12 14 YXXPXD unique sequence 951 11 14 YXXPXD unique sequence % 5.1 7.1 12.2

Table 3.2 RasGAP SH2-dependent YXXPXD consensus binding sequence enrichment. The table shows the percentage of YXXPXD containing proteins found in Drosophila melanogaster (DROME) proteome as well as RasGAP SH2-independent and dependent interacting proteins. The listed number of proteins contain redundancies since the protein isoforms are included. Some proteins have multiple YXXPXD sequence match within their amino acid sequence.

100

3.2.3 Mass spectrometric identification of Drosophila RasGAP interacting partners in S2 cells using unique number of peptides method In addition to spectral counting, the MS results were also assessed using unique number of peptides method. For this study an arbitrary but highly stringent selection criteria of 5 unique number of peptides (minimum), was used to obtain RasGAP interacting proteins (Figure 3.2). Similar to hierarchical clustering this analysis method also allowed mass spectrometric identification of several classes of proteins including RasGAP SH2- independent interacting partners as well as RasGAP SH2-dependent interacting partners (Table 3.3), which form the focus of this study. As shown in Table 3.3 the identified RasGAP SH2-dependent interacting partners such as Dscam and Sprint have much higher number of identified unique peptides than the RasGAPSH2*32* pulldowns, which shows their RasGAP SH2-interacting dependency. In order to compare RasGAP interacting partners obtained through unique number of peptides method (Figure 3.2 and Table 3.3) to the literature, interaction network analysis was used (section 2.4.25). As shown in Figure 3.4 many of the RasGAP SH2-independent (Figure 3.4A3) and SH2-dependent (Figure 3.4A1) interacting proteins identified are novel. Some hits such as PVR (Figure 3.4A2), which were predicted in the literature to interact with RasGAP were also present in this study; however, they did not fulfil the arbitrary high stringent selection criteria of 5 unique number of peptides or their interaction was considered to be nonspecific. The results from this study make a significant contribution to the genetic (Figure 3.4B1), biochemical (Figure 3.4B2) and predicted (Figure 3.4B3) RasGAP interacting partners identified to date.

3.2.4 Comparing spectral counting with unique number of peptides methodologies Since both unique number of peptides and spectral counting methodologies identified similar RasGAP interacting partners, the stringency of the data output between the two methods were compared using different spectral count limits. As shown in Figure 3.5, most RasGAP interacting partners under the unique number of peptides method were also present in the spectral counting method when spectral counts of less than five were allowed, which is consistent with the literature (Lundgren et al., 2010). This shows that spectral counting method is a reliable tool for identifying RasGAP interacting partners, even for samples with low relative abundance.

101

3.2.5 Mass spectrometric identification of protein phosphorylation of Drosophila RasGAP and RasGAP interacting partners In addition to the analysis of seven independent MS experiments, analysis of two further independent MS experiments allowed post-translational modification analysis of several RasGAP interacting proteins including Dscam and Sprint, as well as RasGAP. The phospho-peptide results (Table 3.4) have been validated using manual and MD-score validation techniques. As shown in Table 3.4 there are many phosphorylated residues identified on different proteins including RasGAPWT, which was found to be tyrosine- phosphorylated on tyrosine 363 (pTyr363) residue ((K)AAEKIYATLR(E)) when overexpressed at a large scale in S2 cells. It is important to note that the (K)AAEKIYTLR(E) peptide fragment was detected several times in two independent but parallel experiments and only a small fraction of the detected RasGAP fragment was tyrosine-phosphorylated (data not shown). This finding suggests that only a fraction of the overexpressed RasGAP is tyrosine-phosphorylated on tyrosine residue 363. This is not unexpected since RasGAP is overexpressed and one of the limiting factors for RasGAP phosphorylation is the amount of the endogenous tyrosine kinases present. The NetPhos database, which predicts protein phosphorylation sites based on sequence and structure, has assigned 9 high probability tyrosine-phosphorylated residues to the Drosophila RasGAP sequence, one of which is pTyr363 with 0.807 phosphotyrosine prediction score out of 1 (Blom et al., 1999). This prediction suggests that some of the Drosophila RasGAP can become tyrosine-phosphorylated. In addition, since tyrosine phosphorylation of RasGAPSH2*32* was not detected, this suggests the importance of the RasGAP SH2 domains in mediating RasGAP tyrosine phosphorylation. Other phosphorylated residues were also detected in RasGAP, Dscam and Sprint; however, either no phosphorylation mechanism could be determined since only the wild-type samples were analysed, or the mechanism was independent of the mutation controls used.

102

Table 3.3 RasGAP SH2-dependent interacting partners

Number of Number of Molecular Annotation UniProt accession Experiment Enzyme(s) used unique unique Top three Name Symbol weight Involved biological processes symbol number number for digest peptides peptides expressed tissues (kDa) RasGAPWT RasGAPSH2*32* 1 Trypsin 42 0 Dendrite self-avoidance; axon extension Down 2 Trypsin 33 0 involved in axon guidance; axon syndrome 3 Trypsin 52 4 guidance; phagocytosis; central nervous Thoracicoabdominal cell Dscam CG17800 Q0E9M3_DROME 223 4 Trypsin & Asp-N 54 - system morphogenesis; neuron ganglion, brain and adhesion 5 Trypsin & Glu-C 68 - development; axonal fasciculation; head molecule 6 Trypsin & Asp-N 62 - mushroom body development and 7 Trypsin & Glu-C 58 - peripheral nervous system development SH2, 1 Trypsin 17 0 poly- 2 Trypsin 15 0 proline 3 Trypsin 15 1 Ovary, brain and containing spri CG34414 SPRI_DROME 193 4 Trypsin & Asp-N 51 - Border follicle cell migration larval tubule Ras 5 Trypsin & Glu-C 48 - interactor 6 Trypsin & Asp-N 30 - (Sprint) 7 Trypsin & Glu-C 17 - 1 Trypsin 5 0 2 Trypsin 4 0 enhanced Female meiosis chromosome 3 Trypsin 13 0 Thoracicoabdominal adult segregation; negative regulation of gene east CG4399 O46048_DROME 250 4 Trypsin & Asp-N 24 - ganglion, brain and sensory expression; achiasmate meiosis I and 5 Trypsin & Glu-C 11 - ovary threshold mitotic metaphase 6 Trypsin & Asp-N 2 - 7 Trypsin & Glu-C 6 - 1 Trypsin 17 0 2 Trypsin 12 0 3 Trypsin - - Ovary, head and Ben be CG5530 Q8MRG2_DROME 92 4 Trypsin & Asp-N 11 - Long-term memory digestive system 5 Trypsin & Glu-C 12 - 6 Trypsin & Asp-N - - 7 Trypsin & Glu-C - -

103

1 Trypsin 10 0 2 Trypsin 5 0 3 Trypsin - - Thoracicoabdominal Defence response to Gram-negative Sickie sick CG42589 SICK_DROME 234 4 Trypsin & Asp-N 19 - ganglion, accessory bacterium 5 Trypsin & Glu-C 17 - gland and brain 6 Trypsin & Asp-N - - 7 Trypsin & Glu-C - - 1 Trypsin 8 0 2 Trypsin 5 0 3 Trypsin - - Imaginal disc development and Hindgut, crop and B4 B4 CG9239 O16865_DROME 118 4 Trypsin & Asp-N 5 - circadian rhythm head 5 Trypsin & Glu-C 4 - 6 Trypsin & Asp-N - - 7 Trypsin & Glu-C - - 1 Trypsin 5 0 2 Trypsin 3 0 3 Trypsin 3 0 Brain, Axon guidance; dendrite morphogenesis Bonus bon CG5206 A8WHL4_DROME 121 4 Trypsin & Asp-N 4 - thoracicoabdominal and neuron development 5 Trypsin & Glu-C 0 - ganglion and ovary 6 Trypsin & Asp-N 3 - 7 Trypsin & Glu-C 0 - 1 Trypsin 5 0 2 Trypsin 2 0 3 Trypsin - - Ovary, larval tubule - CG31998 CG31998 Q9V4B6_DROME 162 4 Trypsin & Asp-N - - - and brain 5 Trypsin & Glu-C - - 6 Trypsin & Asp-N 6 - 7 Trypsin & Glu-C 0 - Table 3.3 RasGAP SH2-dependent interacting partners. The above proteins are RasGAP SH2-dependent interacting partners. For experiments 1-3 4 mg, experiments 4-5 15 mg and experiments 6-7 10 mg of the S2 cell lysate total protein, containing LAP tag were pulled down using GFP-Trap and loaded for analysis on SDS-PAGE gels. For experiments 6 and 7, phospho-enrichment technique (pervanadate treatment of S2 cells) was used in order to obtain and identify RasGAP interacting partners. The enzymes used to digest each sample have been listed. The FlyBase database was used to obtain the biological function(s) of each protein in Drosophila. In addition, the presence of the listed proteins in larval fat body, larval tubule, whole fly, carcass, head, brain, crop, hindgut, midgut, accessory gland, ovary, testis and thoracicoabdominal ganglion were assessed using the FlyBase.org GBrowse FlyAtlas database and the highest three expressing tissues in descending order were listed.

104

Table 3.4 Phosphorylation of RasGAP and RasGAP SH2-dependent interacting proteins

Name Symbol Annotation UniProt accession Validated Phosphorylated peptide sequence Enzyme(s) used Experiment symbol number phosphorylated for digest number residues S21/T23 (K)DKPGSPTGCS(D) Trypsin & Asp-N 4 vacuolar Y363 (K)AAEKIYATLR(E) Trypsin & Asp-N 2, 3, 4 RasGAP CG9209 Q8IR23_DROME peduncle (vap) Y629 (D)DLIMPCEEYSPLQQLLLESELYAVK(A) Trypsin & Asp-N 4 S630 (D)DLIMPCEEYSPLQQLLLESELYAVK(A) Trypsin & Asp-N 4 Down syndrome Y1680 (R)DELGYIAPPNRK(L) Trypsin & Asp-N 4 cell adhesion (E)LGYIAPPNR(K) Trypsin & Glu-C 5 Dscam CG17800 Q0E9M3_DROME molecule Y1696/T1698 (K)LPPVPGSNYNTCDR(I) Trypsin & Glu-C 7 (Dscam) S1814/S1816/S1818 (R)SGSQSMPR(A) Trypsin & Glu-C 7 S656/T658 (V)SKPPPTGAPPLPGGGLFSPT(G) Elastase 8 T706/S708 (V)ILTMSPVDNPGHYLPGST(G) Elastase 8 S850 (K)RLSPEGECK(D) Trypsin & Asp-N 6 S892/S893/S894/S895 (R)KLLTSPMTPLTPSGGSSSSGGK(S) Trypsin 9 S919 (S)QHYKESDILESPPMQY(C) Elastase 8 SH2, poly-proline S927/S930 (A)SALSDKISDYEDVWSHDPSDRA(S) Elastase 8 containing Ras spri CG34414 SPRI_DROME S1067 (R)SKQGSPFYAEPADALR(Q) Trypsin 9 interactor (Sprint) Y1070 (K)QGSPFYAEPADALR(Q) Trypsin & Glu-C 5 (R)SKQGSPFYAEPADALR(Q) Trypsin & Glu-C 5 S1171/S1172 (R)NRIDHWQLDSSWEFMAK(Q) Trypsin 9 S1202 (K)QDTGSHAGGDYDTAAIDWQEKENSLGR(D) Trypsin 9 T1241 (R)CSTPPQTAALQPHVLGQDK(A) Trypsin 9 S1338 (S)KCRPALSEDDTIVEEL(Q) Elastase 8 S1705 (R)VIIPDECNGSLQTR(T) Trypsin 9

Table 3.4 Phosphorylation of RasGAP and RasGAP SH2-dependent interacting proteins. The post-translational modifications from 9 independent MS experiments have been collated. For experiments 1-7 LAP-tagged RasGAP was pulled out with GFP-Trap and loaded for analysis on SDS-PAGE gels. For experiments 8-9 GFP-tagged Sprint was pulled out with GFP-Trap and loaded for analysis on SDS-PAGE gels with 5 mg of cell lysates being used. The enzymes used to digest each sample have been listed. The slash sign (/) between residues implies that the mass of the precursor indicated the presence of phosphorylation, but the fragment (MS/MS) information available was insufficient to distinguish the exact site, which is common when potential sites of modification are close together. The amino acid residues on Sprint corresponds to sprint-a (UniProt accession number Q8MQWS8).

105

Figure 3.4 Overlaying of literature and experimental RasGAP interactome. Data from unique number of peptides based MS hit analysis were projected on top of known RasGAP interacting partners present in the literature, using interaction network analysis (section 2.4.25). The interactions showed are not duplicated and are displayed as biochemical (yellow, B2), genetic (green, B1) and predicted (blue, B3) interactions in order of display preference set by the database and the analysis program used. As shown, the identified RasGAP (A4) interacting partners including RasGAP SH2-independent (A3) and SH2-dependent (A1) interacting partners expand the known RasGAP interacting protein network. Some hits however, which were shared between the literature and this study, did not meet the selection criteria of 5 unique number of peptides or they were considered to interact unspecifically to RasGAP (A2).

106

Figure 3.5 Comparing unweighted spectral count and unique number of peptides methodologies for stringency. The number of RasGAP interacting partners are compared when using unique number of peptides versus spectral counting methodologies under different spectral count limits.

107

3.3 Discussion and conclusions

3.3.1 Hierarchical clustering can reliably identify protein enrichment This study has identified many potential novel RasGAP SH2-dependent interacting partners in Drosophila S2 cells. The RasGAP SH2 interacting proteins with highest relative abundance in the pulldowns were Sprint, a cytoplasmic protein acting as a putative Rab5-GEF (Jekely et al., 2005; Szabo et al., 2001); Dscam, a cell adhesion molecule responsible for self-avoidance of sensory neuron dendrites (Matthews et al., 2007; Schmucker et al., 2000); BubR1, a kinetochore protein required for accurate chromosome segregation (Logarinho et al., 2004); and east, a nuclear protein responsible for the normal olfactory and gustatory responses of the adult Drosophila (VijayRaghavan et al., 1992; Wasser and Chia, 2000). None of these proteins have yet been identified as partners of mammalian p120-RasGAP. This study utilised pulldown and MS based techniques in order to identify tyrosine-phosphorylated RasGAP SH2 interacting partners in Drosophila S2 cells, which yielded a large amount of data. In order to analyse the large datasets produced, two methodologies, spectral counting and unique number of peptides, were used to identify many classes of RasGAP interacting partners, including RasGAP SH2-dependent interacting partners. Although spectral counting produced a larger number of hits than highly stringent unique number of peptides method, they shared many hits in common. This shows that spectral counting method, which is thought to be affected by MS-based issues (such as differential ion efficiencies, wide range of protein sizes, differences in peptide solubility, complexity of the solute, chromatographic peak width, column elution time and acquisition speed of mass spectrometer) affecting the linearity of its abundance estimates and its correlation with the MS-detected number of ionised peptides (Lundgren et al., 2010), is a reliable tool for identifying RasGAP interacting partners, even for samples with low relative abundance.

This study has shown for the first time that Drosophila RasGAP SH2-dependent interacting partners, identified using hierarchical clustering, are enriched for the YXXPXD RasGAP SH2-binding consensus sequence, which was originally deduced from mammalian phosphopeptide selection experiments (Songyang et al., 1993) and studies on known RasGAP interacting proteins such as p190-RhoGAP (Hu and Settleman, 1997). This observation is consistent with at least some of the YXXPXD containing proteins directly binding to the RasGAP SH2 domains, which is in agreement with the literature

108

(Hu and Settleman, 1997; Woodcock and Hughes, 2004). Since most of the identified RasGAP SH2-dependent interacting partners did not have the YXXPXD RasGAP SH2- binding consensus sequence it indicates either that there are a large number of false positive RasGAP SH2-dependent interacting partners identified in this study, or that RasGAP SH2-dependent association can take place through indirect binding or through other mechanisms independent of the YXXPXD sequence. When a stringent spectral count cut-off was applied, the fraction of YXXPXD-containing partners increased three-fold, indicating that interacting partners co-purifying most efficiently with RasGAP are more likely to have the consensus binding sequence. However, even at this stringent cut-off the majority of interacting partners (9/14, 64.3%) did not have the consensus sequence. This trend seems to be repeated in the proteins identified with the unique number of peptides method.

It is important to note that the GFP-Trap based pulldown of RasGAP offered a highly efficient technique for purifying RasGAP and its interacting partners. It was predicted that tandem affinity purification (TAP) of the LAP-tagged RasGAP interacting partners would offer a better resolution for RasGAP interacting partners but that was not the case (data not shown). This could be due to an increase in the number of purification steps leading to the loss of RasGAP interacting partners. In addition, 6x (poly)-His tag purification of pulled down proteins was found to be very inefficient. The PreScission protease enzyme also seemed to cut unspecifically pulled down proteins since it increased the number of protein bands observed on the gel (data not shown). Therefore a single step GFP tag purification was performed in all the experiments.

3.3.2 Identified RasGAP SH2-dependent interacting partners play a variety of roles within Drosophila Several of the RasGAP SH2-dependent interacting proteins have previously been shown to play a variety of roles in Drosophila. Of particular interest are proteins highly expressed in the brain or head or that play a role in Drosophila brain. Due to the high number of proteins present in this category, only proteins with an arbitrary threshold of 5 or higher unique peptides identified in the mass spectrometric studies, including two Dscam isoforms, Sprint (spri), east (east), Ben (be), Sickie (sick), B4, Bonus (bon) and CG31998 were defined as RasGAP SH2-dependent interacting proteins. The aforementioned proteins are interesting since Drosophila RasGAP SH2-dependent interacting partners are potentially responsible for the age-related autophagic neurodegeneration (vap mutant)

109 phenotype in flies (Botella et al., 2003). Therefore any RasGAP SH2-dependent interacting partners that potentially play a role in neuronal processes within the fly brain could be an important RasGAP upstream or downstream partner involved in the vap mutant phenotype. However, this is not to suggest that proteins, which interact with RasGAP in an SH2- independent manner are any less significant RasGAP interacting partners in terms of their potential role in the vap mutant phenotype. For instance, the stress-sensitive B (sesB) gene, which encodes an adenine nucleotide translocator 1 (ANT1) was identified as a bang- sensitive seizure mutant. Interestingly, similar to vap mutant flies, sesB mutants exhibit a shortened life span and an age-dependent neurodegeneration phenotype (Celotto et al., 2006; Fergestad et al., 2008). Therefore, considering the importance of RasGAP SH2 domains in binding to RTKs, molecules such as ANT1 (sesB) could interact with other RasGAP domains, such as its PH domain, and this interaction might be critical for neuronal survival in Drosophila. Given the large number of RasGAP interacting proteins that this study has identified, when compared against the known RasGAP interacting partners in the literature, it is not unreasonable to suggest that RasGAP is capable of mediating different molecular processes.

3.3.3 Drosophila RasGAP and RasGAP interacting proteins post-translational modifications are likely to reflect significant biological functions The identified Drosophila RasGAP pTyr363 residue (IYATLR) is equivalent to the mammalian p120-RasGAP tyrosine 460 (IYNTIR) residue, which is phosphorylated in humans in response to signalling by EGFR and non-RTKs such as Src (Liu and Pawson, 1991; Moran et al., 1991; Park and Jove, 1993; Park et al., 1992). The tyrosine phosphorylation site in both Drosophila and humans lies between the most carboxy- terminal SH2 domain and the PH domain of the RasGAP (Chang et al., 2008; Feldmann et al., 1999). The highly conserved sequence phosphorylation of this motif between the two species suggests the importance of this post-translational modification of RasGAP in mediating its molecular (and perhaps cellular and whole organism) functions (Park and Jove, 1993). It is possible that pTyr363 provides a docking site for another pY-binding (PTB) or SH2-containing protein. It is also possible that tyrosine phosphorylation of RasGAP on residue 363 modulates its GAP activity in cells. Although phosphorylation of the Drosophila RasGAP tyrosine 363 residue has been observed before (Chang et al., 2008) this study has found RasGAP tyrosine phosphorylation to be SH2-dependent, which suggests that RasGAP is required to interact with another protein (possibly a RTK) through

110 its SH2 domains in order to become tyrosine-phosphorylated. In addition to Drosophila RasGAP pTyr363 residue other phosphorylated residues including S21/T23, S30 and Y629 were also identified in this study, which could potentially play a role in modulating RasGAP activity. Another protein, which was found to be tyrosine-phosphorylated in this study was Dscam with phosphorylation sites on Y1680 and Y1696/T1689 residues, consistent with previous findings (Chang et al., 2008). In addition, Dscam was found to be serine-phosphorylated on residues S1814/S1816/S1818. Interestingly, Dscam contains the YXXPXD RasGAP SH2-binding consensus sequence, but neither the unphosphorylated nor the tyrosine-phosphorylated proteolytic peptides were detected by mass spectrometry so it was not possible to determine whether the YXXPXD motif was tyrosine- phosphorylated. This could be due to the fast and inefficient rate at which the samples get processed in the mass spectrometer, leading to a high rate of peptide loss. This is potentially a problem considering the protein coverage of RasGAP SH2-dependent interacting partners is less than 35% (data not shown) and phosphotyrosine residues within Drosophila cellular proteome are present at a very low abundance (1-3%) (Bodenmiller et al., 2007). The lack of peptide detection by the mass spectrometer could also be due to the length or the solubility of the digested peptide fragment. Sprint was found to be tyrosine- phosphorylated on residue Y1070, corresponding to the YXXPXD RasGAP SH2-binding consensus sequence, which is consistent with previous findings (Chang et al., 2008). Direct Sprint pulldown studies have also identified several other Sprint post-translational modifications (Table 3.4), which indicates the complex Sprint regulatory mechanism in Drosophila.

3.3.4 Limitations of this study The results in this chapter show that Drosophila RasGAP can associate with several partners from S2 cells in an SH2-dependent manner. There are, however, two major experimental limitations to this study. First, the results of this study do not address whether the identified RasGAP SH2-dependent interacting partners are direct SH2 interactors or whether they require the correct association of SH2 to a site on another undetected adaptor protein. This study also fails to address whether the association of RasGAP SH2 domains to the target protein can cause RasGAP to undergo a conformational change, allowing another region of the RasGAP to associate with the identified proteins. The second major limitation to this study was the low coverage of RasGAP SH2-dependent interacting partners, which prevented identification of all the post-translational modifications of

111

RasGAP SH2-dependent interacting partners. In addition, the current study cannot determine the importance of RasGAP and RasGAP interacting partners post-translational modification and how this affects RasGAP interacting partners and in turn their biological role within S2 cells and the whole organism. This study also cannot determine the spatial and temporal point at which the RasGAP and RasGAP interacting partners become tyrosine-phosphorylated. It is also important to note that this study cannot determine whether the identified sites of phosphorylation on RasGAP and RasGAP interacting proteins are the main sites of phosphorylation or not.

3.3.5 Future experiments Although multiple tyrosine-phosphorylated bands were identified to interact with RasGAP in an SH2-dependent manner, their exact identity and their binding behaviour has not been fully elucidated in these MS experiments. Since there are lack of antibodies for most of the identified RasGAP SH2-dependent interacting partners it would be interesting to repeat the above experiments and use dsRNAs (Brown, 2010) to knockdown some of the identified proteins in S2 cells in order to see if any of the pulled down phosphotyrosine bands would be affected. This will lend further support to the identity of the RasGAP SH2-dependent interacting partners determined in this chapter. It would also be interesting to repeat the above pulldown and MS experiments in Drosophila lines expressing LAP-tagged RasGAP in a pan-neuronal (ELAV) pattern in order to see if any of the identified RasGAP interacting partners can also be detected in Drosophila brain/neuronal tissue. Initial experiments have suggested that RasGAP fails to associate with any tyrosine- phosphorylated proteins when expressed in Drosophila neurons (data not shown). This could however be due to the low number of neuronal cells, cell lysate total protein or absolute abundance of the RasGAP SH2-medited interacting partners. It is important to note that only certain molecular weight regions on the SDS-PAGE gel, which corresponded to tyrosine-phosphorylated RasGAP SH2-dependent interacting proteins, were analysed with MS (Figure 3.2). Hence it would be interesting to repeat the above experiments with all the gel regions being analysed with MS in order to detect other RasGAP interacting partners.

112

CHAPTER 4: A novel interaction between Drosophila RasGAP and Sprint

4.1 Introduction

4.1.1 Introduction to Sprint and its role in Rab5 mediated endocytosis One of the most abundant SH2-dependent RasGAP interacting proteins identified in chapter 3 was Sprint, a putative Rab5-GEF (Jekely et al., 2005). This potential association is particularly interesting as not only is Sprint involved in the regulation of RTK signalling but it is also a putative Ras effector. Sprint associates with the PVR RTK, most likely through direct binding of its SH2 domain to phosphorylated tyrosine residues on Sprint (Jekely et al., 2005), and is involved in localised PVR (and EGFR) signalling in migrating border cells in the Drosophila ovary (Jekely et al., 2005). Sprint has a conserved Ras association (RA) domain at its C-terminal end and associates with constitutively active Ras. The Rab5-GEF activity of RIN1, the mammalian homologue of Sprint, is activated by direct binding to Ras-GTP indicating that the RIN/Sprint proteins are Ras effectors (Tall et al., 2001). The potential regulation of Sprint by Ras, and the known role of RasGAP as a negative regulator of Ras, indicated that exploring the association between RasGAP and Sprint might provide insight into the control of Rab5-mediated endocytosis, an important regulatory process in RTK signalling. The aims of this chapter were to confirm the association between RasGAP and Sprint and to elucidate the molecular basis of the interaction.

4.2 Results

4.2.1 RasGAP interacts with tyrosine-phosphorylated Sprint in an SH2-dependent manner To determine the mechanism of Sprint-RasGAP interaction, GFP-tagged SprintWT was used to pulldown a number of RasGAP constructs expressed in S2 cells. Wild-type RasGAP co-immunoprecipitated with Sprint and this association was SH2-dependent as double SH2 inactivated RasGAP (RasGAPSH2*32*) did not associate with Sprint (Figure 4.1 lanes 1 and 2). The amino-terminal SH2-SH3-SH2 region of RasGAP was sufficient for Sprint association (Figure 4.1 lane 3) and mutation of the SH3, GAP and pTyr363 domains of RasGAP had no effect on the Sprint-RasGAP interaction (Figure 4.1 lanes 6, 7 and 8),

113 indicating that the SH2 domains alone are necessary and sufficient for the interaction. Either of the two RasGAP SH2 domains was sufficient for Sprint association. To determine whether Sprint interaction with RasGAP SH2 domains depends on Sprint tyrosine phosphorylation, GFP-tagged SprintWT was overexpressed in S2 cells, which was lysed in the absence or presence of the inhibitor of tyrosine phosphatases, sodium WT * * orthovanadate (Na3VO4). GST-tagged RasGAP SH232 and SH2 32 domains were then used to pulldown GFP-SprintWT. Sprint was tyrosine-phosphorylated in the presence of

Na3VO4 (Figure 4.2A) and only tyrosine-phosphorylated Sprint interacted with RasGAP

SH2 domains (Figure 4.2B). In the absence of Na3VO4 Sprint failed to become tyrosine- phosphorylated and interact with RasGAP SH2 domains. As observed previously, Sprint only associated with intact RasGAP SH2 domains, since it failed to interact with GST- * * SH2 32 in the presence or absence of Na3VO4. Next, the regions of Sprint responsible for RasGAP SH2 association were scrutinised. In order to narrow down the region of Sprint responsible for mediating phosphotyrosine-dependent Sprint-RasGAP interaction a number of Sprint truncated constructs were made and used to pulldown RasGAP (Figure 4.3E). RasGAP was found to interact with ΔVPS9 Sprint (Sprint443-1292) but failed to interact with shorter Sprint443-811 and Sprint443-504 constructs (Figure 4.3A). This indicates that the region between residues 811 and 1292, which encompasses six tyrosine residues (YYYYPYY) including pTyr1056, is responsible for mediating the Sprint-RasGAP interaction.

As phosphorylation of Sprint on tyrosine 1056 had been observed previously (Table 3.4) (Chang et al., 2008), within a (phospho)YXXPXD RasGAP SH2-binding consensus sequence, this residue was mutated to phenylalanine to investigate whether it is required for association with RasGAP. Interestingly, RasGAP was found to interact with Sprint YXXPXD mutant (SprintFXXPXD) as well as SprintWT (Figure 4.4). This indicates that Sprint Y1056 is not essential for RasGAP association. The level of tyrosine phosphorylation of SprintFXXPXD was similar to SprintWT, indicating that there are other phosphotyrosine residues within Sprint that contribute to the Sprint-RasGAP interaction (Figure 4.5). As expected, Sprint443-504, which contains no tyrosine residues, failed to show any tyrosine phosphorylation (Figure 4.5) and did not interact with RasGAP (Figure 4.3). In addition to Y1056, other Sprint residues were mutated in order to determine if any other functional domains of Sprint contribute to its interaction with RasGAP. As shown in Figure 4.4, RasGAP associates with Sprint SH2 (SprintSH2*) and VPS9 mutant constructs

114

(see section 4.2.4 for further details on Sprint VPS9 point mutations), indicating that Sprint-RasGAP interaction does not depend on these regions being intact.

115

RasGAP-myc

Y L

* O * 2 2 2 2 2 * 3 3 F * * 3 * 3 3 3 2 2 2 2 2 P 6 T H H H H H 3 A

W S S S S S Y G GFP-SprintWT + + + + + + + + A 220

100 GFP-Trap pulldown (Sprint) WB: α-myc (RasGAP) 60

30 1 2 3 4 5 6 7 8 B 220

100 Cell extract WB: α-myc (RasGAP) 60

30 1 2 3 4 5 6 7 8 C 220

GFP-Trap pulldown (Sprint) WB: α-GFP (Sprint) 100 D 1 2 3 4 5 6 7 8 220 GFP-Trap pulldown (Sprint) WB: α-pTyr 100 1 2 3 4 5 6 7 8

Figure 4.1 RasGAP SH232 region is sufficient and necessary for Sprint association. In order to assess the mechanism of Sprint-RasGAP interaction, GFP-tagged Sprint wild-type was co-transfected with variety of myc-tagged RasGAP constructs in Drosophila S2 cells. The GFP-Trap affinity beads were used to pulldown Sprint (large filled arrowhead, panels C and D) from 2.2 mg of S2 cell lysate. The Sprint interacting RasGAP constructs (small filled arrowheads, panel A) were western-blotted using (A) anti-myc antibody. The GFP- Sprint load controls were western-blotted using anti-GFP (large filled arrowhead, panel C) and anti-pTyr (large filled arrowhead, panel D) antibodies. 50 μg of cell extracts were western-blotted using (B) anti-myc antibody. The double myc bands likely represent RasGAP degradation products.

116

ABGFP-SprintWT GST-SH232 WT WT Na3VO4 + - * * Na3VO4 + + - - 220 220

GFP-Trap pulldown (Sprint) GST-SH232 pulldown WB: α-GFP (Sprint) WB: α-GFP (Sprint)

100 100 220 220 Cell extract WB: α-GFP (Sprint) Cell extract WB: α-GFP (Sprint) 100 100 100 GST-SH232 pulldown 60 220 Coomassie stain

45 GFP-Trap pulldown (Sprint) WB: α-pTyr 220

GST-SH232 pulldown 100 WB: α-pTyr

220 100 220 Cell extract WB: α-pTyr Cell extract WB: α-pTyr 100 100 1 2 1 2 3 4

Figure 4.2 Sprint interacts with RasGAP SH2 domains in a phosphotyrosine-dependent manner. In order to assess the mechanism of Sprint interaction with RasGAP SH2 domains, GFP-tagged SprintWT was transfected in Drosophila S2 cells, which were lysed in the presence or absence of sodium orthovanadate

(Na3VO4). (A) The GFP-Trap affinity beads and (B) wild-type and double SH2 inactivated (*) GST-SH232 bound to Agarose 4B beads were used to pulldown Sprint (filled arrowheads) from 2.4 mg of S2 cell lysate. The pulldowns and cell extracts were western-blotted with anti-GFP and anti-pTyr antibodies. The GST- SH232 load controls were stain in a SDS-PAGE gel, using Coomassie stain.

117

GFP-Sprint

WT 443-504 443-811 443-1292

RasGAPWT-myc + - + - + - +-

RasGAPSH2*32*-myc - + - + - + -+

A 220

GFP-Trap pulldown (Sprint) WB: -myc (RasGAP)

100 1 2 3 4 5 6 7 8 B 220

Cell extract WB: -myc (RasGAP) 100 1 2 3 4 5 6 7 8 C 220

100 GFP-Trap pulldown (Sprint) 60 WB: -GFP (Sprint)

1 2 3 4 5 6 7 8 D 220

100 GFP-Trap pulldown (Sprint) 60 WB: -pTyr

1 2 3 4 5 6 7 8

E GFP SH2 PRD YYYYPYY RH VPS9 RA 6xHis kDa

179 SprintWT

39 Sprint443-504

72 Sprint443-811 125 Sprint443-1292

Figure 4.3 A region of Sprint containing 6 tyrosine residues, including the YXXPXD motif mediates RasGAP association. In order to assess the mechanism of Sprint-RasGAP interaction, GFP-tagged Sprint constructs (E) were co-transfected with myc-tagged RasGAP constructs in Drosophila S2 cells. The GFP-Trap affinity beads were used to pulldown Sprint (filled arrowheads, panels C and D) from 1.6 mg of S2 cell lysate. The Sprint interacting RasGAP constructs (filled arrowheads, panel A) were western-blotted using (A) anti-myc antibody. The GFP-Sprint load controls were western-blotted using anti-GFP (filled arrowheads, panel C) and anti-pTyr (filled arrowheads, panel D) antibodies. 40 μg of cell extracts were western-blotted using (B) anti-myc antibody. The double myc bands likely represent RasGAP degradation products. PRD, proline-rich domain; RA, Ras association; RH, RIN homology; SH2, Src homology 2; VPS9, vacuolar protein sorting 9p- like; YP, phosphotyrosine. 118

GFP-Sprint

WT FXXPXD SH2* VPS9PA VPS9DA VPS9DA/PA

RasGAPWT-myc +- + - + - +- + - + -

RasGAPSH2*32*-myc -+ - + - + -+ - + - +

A 220

GFP-Trap pulldown (Sprint) WB: -myc (RasGAP) 100 1 2 3 4 5 6 7 8 9 10 11 12 B 220

Cell extract WB: -myc (RasGAP) 100

1 2 3 4 5 6 7 8 9 10 11 12 C 220

GFP-Trap pulldown (Sprint) WB: -GFP (Sprint) 100

1 2 3 4 5 6 7 8 9 10 11 12

D 220 GFP-Trap pulldown (Sprint) WB: -pTyr

100

1 2 3 4 5 6 7 8 9 10 11 12

Figure 4.4 RasGAP associates with Sprint SH2 and VPS9 mutants. In order to assess the mechanism of Sprint-RasGAP association, GFP-tagged Sprint constructs were co-transfected with myc-tagged RasGAP constructs in Drosophila S2 cells. The GFP-Trap affinity beads were used to pulldown Sprint (large filled arrowhead, panels C and D) from 1.6 mg of S2 cell lysate. The Sprint interacting RasGAP constructs (small filled arrowheads, panel A) were western-blotted using (A) anti-myc antibody. The GFP-Sprint load controls were western-blotted using anti-GFP (large filled arrowhead, panel C) and anti-pTyr (large filled arrowhead, panel D) antibodies. 20 μg of cell extracts were western-blotted using (B) anti-myc antibody. The double myc bands likely represent RasGAP degradation products.

119

GFP-Sprint

4 9

0 1 2

P D

5 8 1

9 9 9

- - -

S S S

3 3 3

P P P

4 4 4

V V V A W 220

100 GFP-Trap pulldown (Sprint) WB: -GFP (Sprint) 60

30 B 1 2 3 4 5 6 7 8 9 220

100 Cell extract WB: -GFP (Sprint) 60

30 C 1 2 3 4 5 6 7 8 9 220

100 GFP-Trap pulldown (Sprint) 60 WB: -pTyr

1 2 3 4 5 6 7 8 9

Figure 4.5 Sprint point mutations do not effect its overall tyrosine phosphorylation. In order to assess the degree of Sprint tyrosine phosphorylation, different GFP-tagged Sprint point mutant and truncated constructs were transfected in Drosophila S2 cells. The GFP-Trap affinity beads were used to pulldown Sprint (filled arrowheads, panels A and C) from 2.3 mg of S2 cell lysate. The GFP-Sprint load controls were western- blotted using anti-GFP (filled arrowheads, panel A) and anti-pTyr (filled arrowheads, panel C) antibodies. 20 μg of cell extracts were western-blotted using (B) anti-GFP antibody.

120

4.2.2 RasGAP associates with PVR and EGFR In addition to binding to Btl, Htl and Torso receptor tyrosine kinases (RTKs), it has been predicted in the literature that full length RasGAP can also associate with other RTKs (Botella et al., 2003; Cleghon et al., 1998; Feldmann et al., 1999; Woodcock, 2004; Woodcock and Hughes, 2004). None of the RTKs that RasGAP has been shown to associate with – Btl, Htl and Torso – are expressed in S2 cells. The PVR RTK, however, is expressed in S2 cells and controls growth and cell size in an autocrine growth signalling loop with its ligand PVF2 and PVF3 (Sims et al., 2009). As PVR is known to associate with Sprint it could act as a ‘bridging’ protein between Sprint and RasGAP. It was also of interest to determine whether RasGAP could associate with the EGFR, as RasGAP is strongly implicated in regulating EGFR signalling strength in the adult Drosophila brain (Botella et al., 2003). In order to determine whether RasGAP can interact with these other RTKs in an SH2 dependent manner, GFP-tagged Drosophila PVR and EGFR, were used to pulldown different myc-tagged RasGAP constructs. As shown in Figure 4.6, RasGAP interacted with tyrosine-phosphorylated PVR and EGFR in an SH2-dependent manner, as RasGAPWT, but not RasGAPSH2*32* co-immunoprecipitates with both RTKs. Interestingly, in RasGAPWT pulldowns (Figure 4.6A), there were two RasGAP bands that could be observed close to each other. PVR seems to enrich for the lower molecular weight (faster migrating) RasGAP band, whereas EGFR enriched for the higher molecular weight (slower migrating) RasGAP band. This could potentially reflect the effects of the receptors on RasGAP post-translational modifications, including tyrosine phosphorylation.

4.2.3 RasGAP and Sprint independently interact with the cytoplasmic tyrosine kinase Abl Mammalian p120-RasGAP (Frackelton et al., 1993) and RIN1 (Afar et al., 1997) are known to interact with Abl. Hence in order to assess whether this function of Abl is evolutionarily conserved, GFP-tagged Sprint was used to pulldown myc-tagged Abl and RasGAP constructs. As shown in Figure 4.7A RasGAP was capable of interacting with Abl tyrosine kinase (lane 8) as well as Sprint (lane 4). However, this RasGAP-Abl interaction was mediated in an SH2-independent manner since Abl is still capable of interacting with RasGAPSH2*32* (lane 9). Interestingly, Sprint was also capable of interacting with Abl (lane 3), which is consistent with co-localisation of the proteins (Jekely et al., 2005). As Abl is known to tyrosine phosphorylate RIN1 in mammalian cells, it might be responsible for tyrosine phosphorylation of residues on Sprint required for

121

Sprint-RasGAP association. However, when Abl was co-expressed it did not appear to influence the level of association between Sprint and RasGAP (lanes 4 and 6). This could be due to Abl failing to tyrosine-phosphorylate the necessary site(s) on Sprint required for Sprint-RasGAP interaction (Afar et al., 1997) or due to Abl competition with RasGAP for associating with the appropriate Sprint tyrosine residues. Since Abl is known to phosphorylate the mammalian homologue of Sprint, RIN1 (Afar et al., 1997; Hu et al., 2005), the tyrosine kinase activity of Abl on Sprint was assessed. As shown in Figure 4.7E, when Sprint and Abl are co-expressed in S2 cells, the region corresponding to Sprint molecular weight is heavily tyrosine-phosphorylated (lane 3) relative to Sprint on its own (lane 2). However, since the molecular weight of the Abl is close to Sprint, it cannot be stated conclusively that the observed tyrosine-phosphorylated protein band corresponds to Sprint and not Abl. This is the case since RIN1 is known to relieve the autoinhibitory effect of the SH2 and SH3 domains of Abl on its kinase domain activity, leading to Abl activation and autophosphorylation (Hu et al., 2005).

4.2.4 Sprint interacts with dominant negative Rab5 Since Sprint has been predicted to act as a Rab5-GEF (Jekely et al., 2005; Szabo et al., 2001), and its mammalian homologue RIN1-3 (Kajiho et al., 2003; Saito et al., 2002; Tall et al., 2001) have been shown to associate with Rab5, its interaction Rab5 was assessed. This was achieved by co-expressing C-terminus 6x His-tagged Sprint (GFP-Sprint-6xHis) with YFP-tagged Rab5, Rab7 and Rab11 constructs, and then pulling down the His-tagged Sprint with Nickel beads. As shown in Figure 4.8, Sprint interacted with wild-type (lane 1) and dominant negative (DN) Rab5 (lane 3) but failed to interact with constitutively active (CA) Rab5 (lane 2), Rab7WT (lane 7) and Rab11WT (lane 8). Interestingly, Rab5-DN interacted with VPS9 mutant Sprint constructs (lanes 4, 5 and 6), which were predicted to have a reduced interaction with Rab5 (Delprato et al., 2004).

122

PVRWT-GFP EGFRWT-GFP

RasGAP-myc WT SH2*32* WT SH2*32* A 220

GFP-Trap pulldown WB: -myc (RasGAP) 100

1 2 3 4 B 220

Cell extract WB: -myc (RasGAP) 100 1 2 3 4 C 220 GFP-Trap pulldown WB: -GFP

100 1 2 3 4 D 220

Cell extract WB: -GFP

100

1 2 3 4

Figure 4.6 RasGAP associates with wild-type PVR and EGFR through its SH2 domains. In order to asses the RasGAP interaction with PVR and EGFR, GFP-tagged RTK constructs were co-transfected with myc-tagged RasGAP constructs in Drosophila S2 cells. The GFP-Trap affinity beads were used to pulldown RTKs (large filled arrowhead, panels C and D) from 2.7 mg of S2 cell lysate. The RTK interacting RasGAP constructs (small filled arrowheads, panel A) were western-blotted using (A) anti-myc antibody. The GFP-RTK load controls were western-blotted using anti-GFP (large filled arrowhead, panel C) antibody. 40 μg of cell extracts were western-blotted using (B) anti-myc and (D) anti-GFP antibodies.

123

GFP + ------

Sprint - + + + + + + - -

Abl + - + - - + + + +

WT - - - + - + - - - RasGAP-myc SH2*32* - - - - + - + - -

WT ------+ - RasGAP-LAP SH2*32* ------+

A 220 Abl-myc GFP-Trap pulldown (Sprint and Abl) RasGAP-myc WB: -myc (Abl and RasGAP)

100

1 2 3 4 5 6 7 8 9

B 220 Abl-myc Cell extract RasGAP-myc WB: -myc

100

1 2 3 4 5 6 7 8 9

C 220 GFP-Sprint RasGAP-LAP

100 GFP-Trap pulldown 60 WB: -GFP

1 2 3 4 5 6 7 8 9

D 220 GFP-Sprint RasGAP-LAP

100

Cell extract 60 WB: -GFP

1 2 3 4 5 6 7 8 9

E 220 GFP-Sprint RasGAP-LAP

GFP-Trap pulldown 100 WB: -pTyr

45 1 2 3 4 5 6 7 8 9

Figure 4.7 Abl interacts with Sprint and RasGAP but does not influence Sprint-RasGAP interaction. In order to assess Sprint, RasGAP and myc-tagged Abl interaction, the aforementioned constructs were co-transfected in Drosophila S2 cells. The GFP-Trap affinity beads were used to pulldown GFP-tagged Sprint and RasGAP (large filled arrowheads, panels C, E) from 2.9 mg of S2 cell lysate. The Sprint and RasGAP interacting partners (small filled arrowheads, panel A; myc-tagged Abl upper arrow and myc-tagged RasGAP lower arrow) were western-blotted using (A) anti-myc and (E) anti-pTyr antibodies. The GFP load controls were western-blotted using anti-GFP (large filled arrowheads, panel C) and anti-pTyr (large filled arrowheads, panel E) antibodies. 60 μg of cell extracts were western-blotted using (B) anti-myc and (D) anti-GFP antibodies. 124

YFP-Rab5 Rab7 Rab11

WT CA DN DN DN DN WT WT

WT + + + - - - + +

VPS9PA - - - + - - - - GFP-Sprint-6xHis VPS9DA - - - - + - - -

VPS9DA/PA - - - - - + - -

A 220 GFP-Sprint

100 Nickel beads His pulldown WB: -GFP

60 YFP-Rab 45

1 2 3 4 5 6 7 8 B 220 GFP-Sprint

Cell extract 100 WB: -GFP

60 YFP-Rab 45 1 2 3 4 5 6 7 8

Figure 4.8 Sprint interacts with dominant negative Rab5 but not with Rab7 or Rab11. In order to assess the mechanism of Sprint-Rab association, His-tagged Sprint (GFP-Sprint-6xHis) constructs were co-transfected with YFP-tagged Rab (CA, constitutively active, DN, dominant negative; WT, wild-type) constructs in Drosophila S2 cells. The nickel affinity beads were used to pulldown Sprint (large filled arrowhead, panel A) from 3.0 mg of S2 cell lysate. The Sprint interacting Rab constructs (small filled arrowhead, panel A) were western-blotted using (A) anti-GFP antibody. The His-Sprint load controls were western-blotted using anti- GFP (large filled arrowhead, panel A). 70 μg of cell extracts were western-blotted using (B) anti-GFP antibody.

125

4.3 Discussion and conclusions

4.3.1 RasGAP SH2 domains mediate the Sprint-RasGAP interaction This study has confirmed the RasGAP SH2-dependent association of RasGAP with Sprint (Figure 4.1). This association does not depend on the Sprint SH2 domain or the conserved site of RasGAP tyrosine phosphorylation (pTyr363) (Figure 4.1 and Figure 4.4), indicating that Sprint-RasGAP interaction is unlikely to be reciprocal through both their SH2 domains. In addition, I have demonstrated that the nature of Sprint-RasGAP interaction is phosphotyrosine-dependent since RasGAP SH2 domains fail to interact with non-tyrosine- phosphorylated Sprint. Although the association of Sprint and RasGAP is dependent on the RasGAP SH2 domains, the exact biochemical mechanism of the association is not yet clear from this study. The major outstanding question is whether RasGAP and Sprint bind directly to each other. A region of Sprint between the proline-rich domain (PRD) and the RIN homology (RH) domain is sufficient to interact with RasGAP (Figure 4.3). This region includes a tyrosine-phosphorylated residue (Y1056) corresponding to the YXXPXD RasGAP SH2-binding consensus sequence but substitution of this residue with non- phosphorylatable phenylalanine did not affect Sprint-RasGAP association (Figure 4.4). This does not exclude the possibility that the RasGAP SH2 domains directly bind to tyrosine-phosphorylated Sprint as there are five other tyrosine residues in that region and one or more of these must be phosphorylated. Another possible mechanism for RasGAP- Sprint association is that they both independently associate with a protein that acts as a bridge between them. One candidate bridge protein is PVR RTK, which is expressed in S2 cells and is known to associate with the SH2 domain of Sprint. RasGAP does indeed associate with PVR in an SH2 dependent manner (Figure 4.6), but as SH2 mutant Sprint still associates with RasGAP it is unlikely that PVR acts as a bridge protein between them. In conclusion, the evidence shows that RasGAP and Sprint form a protein complex in S2 cells that is dependent on the SH2 domains of RasGAP, but whether they directly interact or not remains to be determined.

4.3.2 Sprint acts as a scaffold protein In addition to associating with RasGAP, Sprint is capable of interacting with Abl tyrosine kinase (Figure 4.7) and Rab5 (Figure 4.8). These interactions are in addition to already established Sprint interacting partners including PVR, EGFR and constitutively active Ras (Jekely et al., 2005). This indicates that Sprint potentially acts as a scaffold protein,

126 allowing complex formation between the aforementioned proteins to take place. Hence once all the appropriate interactions are established, the relevant molecular and cellular responses can occur. Not surprisingly, this scaffolding function of Sprint is preserved in the mammalian homologues of Sprint, the RIN family members, especially RIN1 (Tall et al., 2001). RIN1 is capable of associating with receptor tyrosine kinases such as the EGFR through its SH2 domain (Barbieri et al., 2003), which has been shown to mediate receptor endocytosis (Hu et al., 2008). The VPS9 domain of RIN1 is also capable of inducing receptor internalisation through activating Rab5. The Rab5 exchange activity of the RIN1 VPS9 domain however, requires disengagement of its autoinhibition through association of the active Ras to the RA domain of RIN1. This RIN1-Rab5 signalling favours EGFR internalisation and down regulation, while RIN1-Abl signalling stabilises EGFR and inhibits macropinocytosis (Balaji et al., 2012; Tall et al., 2001). Considering the interactions uncovered in this study, it is likely that Drosophila Sprint is engaged in similar activities to RIN1, taking into account the role of Sprint in RTK signalling required for Drosophila border cell migration (Jekely et al., 2005). Assuming the RIN1 model is reflective of events in Drosophila cells, it is likely that Sprint associates with RTKs in an SH2-dependent manner. Tyrosine phosphorylation of Sprint, either by RTKs or by the cytoplasmic PTK Abl (Hu et al., 2005), would permit RasGAP localisation to this complex, perhaps by direct binding to Sprint, and allow localised regulation of Ras-GTP levels and inhibition of the Rab5-GEF activity of Sprint. This in turn would modulate the rate of Rab5 mediated RTK endocytosis within newly formed endosomes (Tall et al., 2001). The interaction and subsequent signalling of Sprint, Rab5 and Abl in turn determines the fate of endocytosed receptors (Balaji et al., 2012). In this model Sprint is a scaffold protein, that is central to Rab5 mediated RTK endocytosis in Drosophila (Figure 4.9). Interestingly, Sprint mutant (spri6G1) flies are viable with no obvious effects on RTK signalling (Jekely et al., 2005). This is likely due to the presence of two other Drosophila putative Rab-GEFs, GAPEX-5 and RABEX-5 (Yan et al., 2010), playing a role in RTK endocytosis.

127

Figure 4.9 A model for the role of RasGAP-Sprint in receptor mediated endocytosis. Upon growth factor binding to RTKs, the SOS Ras-GEF is recruited to the receptor, in association with the adaptor proteins CSW and Drk, where it activates Ras. RasGAP can associate with activated RTKs and deactivate Ras. In early endosomes, Sprint associates with activated RTKs and its Rab5-GEF activity is stimulated by Ras-GTP. Tyrosine phosphorylation of Sprint allows association of RasGAP, which deactivates Ras and limits Sprint- mediated Rab5 activation. GF, growth factor; RA, Ras association; SH2, Src homology 2; VPS9, vacuolar protein sorting 9p-like; YP, phosphotyrosine.

128

4.3.3 Limitations of this study The major limitation of this study is that it has used overexpression in order to identify Sprint-RasGAP, Sprint-Rab5 and RasGAP-RTKs interaction in Drosophila S2 cells. This is an artificial system that can push the interaction between two non physiologically interacting proteins. The physiological relevance of the proposed interactions would be strengthened if the endogenous proteins were shown to co-immunoprecipitate. However, co-immunoprecipitation of endogenous proteins can be confounded by technical issues with antibodies, such as masking of binding sites and low antibody affinity or specificity. In addition, only a small fraction of the proteins may be present in the complex, particularly for interactions regulated by protein phosphorylation, such as SH2- phosphotyrosine interactions. Although the interaction between Sprint-RasGAP, RasGAP- PVR and RasGAP-EGFR is RasGAP SH2-dependent, it unclear whether the interactions depend on the specific SH2 region of RasGAP or it can be mediated by SH2 domains from other proteins. This is an issue when studying interactions involving any SH2 domain as there are overlapping binding specificities that may be exacerbated when the proteins are overexpressed. The most rigorous way to address this issue would be to create specific SH2 domain mutations in the Drosophila genome by homologous recombination and then investigate the interactions by co-immunoprecipitation of endogenous wild-type and mutant proteins. This study also did not identify the direct or indirect nature of Sprint and RasGAP association, which will be discussed in detail in chapter 8.

4.3.4 Future experiments Since the interaction of Sprint with RasGAP seems to be mediated through YXXPXD independent mechanisms, it would be interesting to identify the tyrosine residue(s) on Sprint responsible for mediating its association to RasGAP. This experiment would provide further evidence about the phosphotyrosine-dependent nature of Sprint and RasGAP association. Also considering RasGAP is capable of associating to PVR and EGFR in an SH2-dependent manner, it would be interesting to determine if Sprint interaction with the aforementioned RTKs is also SH2-dependent. This experiment would provide evidence that Sprint and RasGAP get independently recruited to RTKs and form part of a protein complex, which is responsible for RTK endocytosis and signalling. This experiment would also highlight possible competition and redundancies between Sprint and RasGAP for receptor binding and complex formation. The VPS9 mutant Sprint constructs were capable of associating with Rab5-DN. These results indicate either that the

129 mutated VPS9 residues are not responsible for Rab5 association to Sprint or that Rab5-DN or Sprint VPS9 mutants associate strongly with each other once bound and fail to dissociate. Hence it would be interesting to further assess the mechanism of Sprint and Rab5 association using GDP or GTP loaded Rab5 along with Sprint truncated and point mutant constructs. This experiment would also address whether Rab5 has regulator (Rab5- GDP binding) (Jekely et al., 2005) and effector (Rab5-GTP binding) (Kajiho et al., 2003; Saito et al., 2002) functions. Sprint is a VPS9 containing proteins, which is predicted to act as a Rab5-GEF; however, similar to its other counterparts in Drosophila (RABEX-5 and GAPEX-5) Sprint GEF activity has not yet been determined. Hence it would be interesting to conduct enzyme kinetics experiments on Sprint in order to establish its activity as a Rab5-GEF. This experiment would provide support for the role of Sprint as an endocytic molecule. The role of Sprint as an endocytic molecule can also be addressed by using biotinylation or labelled ligand endocytic assays. Also it would be interesting to assess the effects of Sprint on MAPK (Tootle et al., 2003) or JNK (Chatterjee and Bohmann, 2012) signalling, using luciferase reporter assay. This experiment would provide evidence that Sprint is involved in RTK signalling.

130

CHAPTER 5: Investigation of Sprint and RasGAP subcellular localisation

5.1 Introduction

5.1.1 Introduction to the Rab5 and Rab5-GEFs Active Rab5 is involved in the early stages of endocytosis. Rab5 activation is achieved through the guanine nucleotide exchange factor (GEF) proteins that convert inactive Rab5- GDP to active Rab5-GTP. There are several Rab-GEFs including the mammalian RIN1-3 proteins. The mammalian RIN1-3 proteins have been shown to co-localise with Rab5 positive early endocytic vesicles (Kajiho et al., 2003; Kimura et al., 2006; Tall et al., 2001). However, the subcellular localisation of Sprint, the Drosophila homologue of RIN1-3, and its localisation with Rab5 has not yet been determined (Jekely et al., 2005). The biochemical association between Sprint and RasGAP described in chapter 4 adds a potential novel regulatory mechanism for Rab5 activation in the early stages of endocytosis. In this chapter the subcellular localisation of Sprint, Rab5 and RasGAP in S2 cells was investigated using fluorescence microscopy to determine whether these proteins co-localised and whether they were associated with endosomes.

5.2 Results

5.2.1 Sprint localises to dynamin-GTPase dependent cytoplasmic puncta in S2 cells To investigate the subcellular localisation of Sprint, it was overexpressed as a GFP-tagged protein in S2 cells and its subcellular localisation was determined by fluorescence microscopy. Sprint formed cytoplasmic punctate structures when expressed on its own (Figure 5.1). In order to assess the nature of Sprint puncta, S2 cells expressing GFP-Sprint were treated with Dynasore, an inhibitor of the dynamin-GTPase required for clathrin- dependent coated vesicle formation (Macia et al., 2006). Sprint formed cytoplasmic puncta when the GFP-Sprint expressing S2 cells were treated with the solvent DMSO. However, Dynasore treatment of GFP-Sprint expressing cells resulted in Sprint adopting a diffuse cytoplasmic distribution (Figure 5.1A). This suggests that the Sprint puncta are endocytic structures, whose formation depends on the dynamin-GTPase activity. Dynasore was also capable of inhibiting fluid phase uptake of Dextran in untransfected S2 cells (Figure 5.1A and B).

131

5.2.2 Sprint co-localises with Rab5 in a VPS9 dependent manner In order to determine the endocytic compartment that Sprint localises to, different Sprint constructs were overexpressed with the early- and late- stage endocytic markers Rab5 and Rab7, respectively. Since RFP-Rab5 and RFP-Rab7 constructs used in this study were uncharacterised, their subcellular location and number of endocytic vesicles were compared to characterised YFP-Rab5 and YFP-Rab7 constructs (Zhang et al., 2007). As shown in Figure S3, RFP-Rab5 and YFP-Rab5 co-localised with similar numbers of endocytic vesicles, as did RFP-Rab7 and YFP-Rab7. As expected, Rab5 and Rab7 overexpressed in the same cell did not co-localise within the same endocytic vesicles. As shown in Figure 5.2, Rab5, but not Rab7, co-localised with SprintWT demonstrating that Sprint endocytic puncta are predominantly an early stage endocytic structure. In addition, Rab5 co-localised with SprintSH2*, demonstrating that the mechanism(s) of Rab5-Sprint co- localisation is SH2 independent. Consistent with the role of Sprint as a Rab5-GEF, ΔVPS9 Sprint (Sprint443-1292) failed to co-localise with Rab5 (Figure 5.2A and B). Interestingly, ΔVPS9 Sprint was predominantly localised as puncta within the S2 cell nucleus (Figure S2); this Sprint mutant protein lacks the VPS9 and RA domains suggesting that Sprint localisation with Rab5 positive endosomes requires it to interact with Rab5 and/or Ras. In order to further define the residues responsible for Sprint localisation with Rab5, three Sprint VPS9 domain point mutant constructs (SprintVPS9PA, SprintVPS9DA and SprintVPS9DA/PA), shown to be defective in GEF activity and Rab5 binding in Rabex-5 (Delprato et al., 2004) and RIN1 (Galvis et al., 2009), were co-expressed with Rab5 and their localisation was assessed. As shown in Figure 5.3, Rab5WT failed to co-localise with Sprint point mutant constructs and similar to ΔVPS9 Sprint, Sprint point mutant constructs could also found in the nucleus as puncta (Figure S4). This suggests that the association of Sprint with Rab5 is important for its subcellular localisation.

132

A DMSO Dynasore

)

n d

e n

a a

t s

f x

a e

x D

t e

e T

d (

t )

r n

P i

r F

p W G

T S

(

*** B30 Dextran

25 Sprint

n u

o 20

t c

n 15

n ***

a 10

e M 5

0 Untransfected SprintWT DMSO +- +- Dynasore -+ - +

Figure 5.1 Sprint forms endocytic puncta. (A) Drosophila GFP-SprintWT transfected or untransfected (Dextran) S2 cells were treated with DMSO or Dynasore (100 μM in DMSO) for 1 (untransfected) or 24 (GFP-SprintWT transfected) hours, fixed and observed with fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Sprint (green bars) or Dextran puncta (red bars) per cell were recorded with n ≥ 30 (section 2.4.16). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001).

133

SprintWT + SprintSH2* + Sprint443-1292+ SprintWT+ Sprint443-1292 A Rab5WT Rab5WT Rab5WT Rab7WT

GFP (Sprint)

RFP (Rab)

Merge

134

***

*** Rab5 30 *** Rab7 Sprint

t n

u 20

t c

n 15

a 10

e M

0 Rab5WT + Rab5WT + Rab5WT + Rab7WT + Sprint443-1292 SprintWT SprintSH2* Sprint443-1292 SprintWT

Figure 5.2 Sprint co-localises with Rab5 in a VPS9 dependent manner. (A) Drosophila S2 cells were co- transfected with GFP-Sprint and either RFP-Rab5 or RFP-Rab7 constructs and the localisation of each protein was visualised by fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Rab5 (yellow bars) and Rab7 (blue bar) vesicles per cell and Sprint puncta (green bars) co- localising with either Rab5 or Rab7 were recorded (n ≥ 43) (section 2.4.16). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001).

135

A SprintVPS9PA + SprintVPS9DA + SprintVPS9DA/PA Rab5WT Rab5WT + Rab5WT

GFP (Sprint)

RFP (Rab)

Merge

B ** 35 *** Rab5 30

Rab5-Sprint t

n 25

a 20

n u

p 15

n a

e 10 M

0 Rab5WT + SprintVPS9PA Rab5WT + SprintVPS9DA Rab5WT + SprintVPS9DA/PA

Figure 5.3 Sprint co-localises with Rab5 through conserved Sprint VPS9 residues. (A) Drosophila S2 cells were co-transfected with GFP-Sprint and RFP-Rab5 constructs and the localisation of each protein was visualised by fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Rab5 vesicles (yellow bars) per cell and Sprint puncta (green bars) co-localising with Rab5 were recorded (n ≥ 63) (section 2.4.16). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA. *** (p < 0.001) and ** (p < 0.01) indicates a significant difference between sample means.

136

5.2.3 Modulators of Sprint puncta In order to investigate the molecular pathway(s) in which Sprint participates, several genes involved in RTK signalling and endocytosis were knocked-down through dsRNA (RNAi) treatment (Brown, 2010) of GFP-Sprint expressing S2 cells. As shown in Figure 5.4, knockdown of shibire (Drosophila dynamin homologue), Rab5, Ras64B (Dras2), Ras85D (Dras1), pvr and pvf2 (RTK and its ligand, respectively) genes significantly reduced the number of Sprint positive endocytic punta, compared to the control (Renilla dsRNA). In addition, consistent with the results of Sprint localisation with Rab5 but not Rab7 endocytic vesicles (Figure 5.2), Rab7 knockdown did not affect the number of Sprint positive puncta, indicating that disruption in the later Rab7 endosomal compartment does not affect the Sprint positive endosomes. Knockdown of two other putative Rab5-GEFs, Rabex-5 and Gapex-5, and Abl (a potential Sprint tyrosine kinase) did not affect the number of Sprint positive endosomes.

5.2.4 Sprint and RasGAP co-localise in punctate structures in S2 cells RasGAP and Sprint constructs were overexpressed in S2 cells and their subcellular localisation was determined by fluorescence microscopy. SprintWT and RasGAPWT co- localised in cytoplasmic punctate structures (Figure 5.5), and the co-localisation was dependent on the SH2 domains of both Sprint and RasGAP being intact as they otherwise adopted a diffuse cytoplasmic distribution. When expressed separately, Sprint formed cytoplasmic puncta in an SH2-dependent manner but RasGAP had a diffuse cytoplasmic distribution (Figure S5). This indicates that the formation of RasGAP puncta depends on the presence of SprintWT. In order to assess the nature of Sprint-RasGAP localisation, ΔVPS9 Sprint, shown to be localised to the nucleus, was overexpressed with RasGAP constructs. As shown in Figure S6, similar to ΔVPS9 Sprint, RasGAP adopts a nuclear localisation when overexpressed with ΔVPS9 Sprint.

137

A30

t * ** n

u 20

t *** c

n 15

u *** p

*** ***

n a

e 10

M ***

0 Unt Water Renilla shibire Rab5 Rab7 Ras64B Ras85D vap Gap1 Rabex-5 Gapex-5 Abl pvr pvf2

5 B D - -5 a 4 5 x ll re x B i i 5 7 6 8 1 e e 2 n b b b s s p b p l r f e i a a a a p a a a b v v h a p p R s R R R R v G R G A

220

Cell extract WB: -GFP

100 1 2 3 4 5 6 7 8 9 10 11 12 13 60 Cell extract WB: -Tubulin 45

1 2 3 4 5 6 7 8 9 10 11 12 13

Figure 5.4 Modulators of Sprint puncta. (A-B) Drosophila S2 cells were transfected with GFP-SprintWT and were left either unteared (Unt) treated with water or dsRNA for the indicated genes. The number of Sprint puncta (green bars) were recorded for each knockdown (n ≥ 37) (section 2.4.16) and compared against control (Renilla dsRNA). Data are represented as mean ± SEM. Statistical significance was tested by one- way ANOVA with Renilla dsRNA treated S2 cell results as the control. *** (p < 0.001), ** (p < 0.01) or * (p < 0.05) indicates a significant difference between sample means. (B) The GFP-Sprint load controls were western-blotted using anti-GFP (large filled arrowhead) and anti-Dtub (small filled arrowhead) antibodies.

138

A RasGAPWT + RasGAPSH2*32* RasGAPWT + RasGAPSH2*32* RasGAPWT + RasGAPSH2*32* SprintWT + SprintWT SprintFXXPXD + SprintFXXPXD SprintSH2* + SprintSH2*

GFP (Sprint)

-myc (RasGAP)

Merge

139

B 25

*** *** 20 Sprint

*** t

n Sprint-RasGAP u

o 15

n 10

e M

0 SprintWT + SprintWT + SprintFXXPXD + SprintFXXPXD + SprintSH2* + SprintSH2* + RasGAPWT RasGAPSH2*32* RasGAPWT RasGAPSH2*32* RasGAPWT RasGAPSH2*32*

Figure 5.5 Sprint and RasGAP co-localise in cytoplasmic punctate structures. (A) Drosophila S2 cells were co-transfected with GFP-Sprint and myc-tagged RasGAP constructs and the localisation of each protein was visualised by fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Sprint puncta (green bars) per cell and RasGAP puncta (red bars) co-localising with Sprint puncta were recorded (n ≥ 34) (section 2.4.16). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001).

140

5.2.5 Sprint and RasGAP co-localise in early stage endocytic vesicles In order to determine the endocytic compartment in which Sprint and RasGAP co-localise to, given that Sprint forms endocytic punctate structures, Sprint and Rab5 were overexpressed with RasGAPWT or RasGAPSH2*32* constructs. As shown in Figure 5.6, Sprint and RasGAPWT co-localised with Rab5 positive early stage endocytic vesicles. As shown previously in this study, RasGAPSH2*32* did not localise with Sprint. In addition, when only RasGAP and Rab5 constructs were co-expressed in S2 cells, neither RasGAPWT nor RasGAPSH2*32* constructs localised with Rab5 positive endocytic vesicles (data not shown) indicating that Sprint is required for RasGAP co-localisation with Rab5 positive endocytic vesicles. Interestingly, there was a significantly higher degree of Sprint association to Rab5 in the presence of RasGAPWT compared with RasGAPSH2*32* or no RasGAP present.

141

A SprintWT + Rab5WT SprintWT + Rab5WT WT SH2*32* + RasGAP + RasGAP

GFP

(Sprint)

RFP

(Rab)

α-myc (RasGAP)

142

B 30

***

25 ***

t n

o 20

t c

n 15

a 10

e M

0 RasGAPWT RasGAPSH2*32* No RasGAP Sprint Sprint-RasGAP Sprint-Rab5 Sprint-Rab5-RasGAP

Figure 5.6 Sprint and RasGAP co-localise to early stage endocytic vesicles. (A) Drosophila S2 cells were co-transfected with GFP-Sprint, RFP-Rab5 and myc-tagged RasGAP constructs and the localisation of each protein was visualised by fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Sprint puncta (green bars) per cell, Rab5 (yellow bars) and RasGAP (red bars) puncta co- localising with Sprint puncta, and Sprint puncta co-localising with both Rab5 and RasGAP (orange bars) were recorded (n ≥ 30) (section 2.4.16). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001).

143

5.3 Discussion and conclusions

5.3.1 Sprint co-localises to early stage endocytic vesicles and potentially regulates early stages of RTK endocytosis This study has shown that Sprint forms endocytic puncta, whose formation is dependent on the dynamin-GTPase activity. Sprint and RasGAP were also found to co-localise with the early stage endocytic vesicle marker Rab5, but not with the later stage endocytic marker Rab7. In addition, Sprint and RasGAP were found to co-localise with each other and this co-localisation was dependent on the SH2 domains of both proteins being intact. The number of Sprint puncta was also shown to be effected by molecules regulating RTK signalling and endocytosis. The co-localisation of Sprint and Rab5 in this study is similar to RIN1-3, the mammalian homologue of Sprint, which has been shown to co-localise with Rab5 positive early stage endocytic vesicles (Kajiho et al., 2003; Kimura et al., 2006; Tall et al., 2001). In this study the VPS9 domain of Sprint was shown to be required for co- localisation with Rab5. This is consistent with the proposed role of Sprint as a Rab5-GEF as the VPS9 domain would be required for direct binding of Sprint to Rab5-GDP. Although the Sprint VPS9 point mutants failed to co-localise with Rab5, the pulldowns with the same mutants in chapter (4) showed that were able to associate with DN Rab5. This could reflect a difference between the wild-type Rab5 used in the co-localisation and the DN Rab5 used in the pulldowns, as the DN Rab5 should have a higher affinity for the GEF domain and may therefore be less affected by the VPS9 amino-acid substitutions than wild-type Rab5. Localisation of Sprint to endosomes required a functional SH2 domain but this requirement was bypassed when Rab5 was co-expressed, suggesting Rab5 overexpression allows direct recruitment of Sprint to Rab5 positive early endosomes. The endocytic nature of Sprint as well as Sprint and Rab5 co-localisation is consistent with the role of Sprint as a Rab5-GEF and its involvement in early stage endocytosis.

Localisation of Sprint to early endosomes required its SH2 domain and suggest that it localises to early endosomes by associating to a receptor. Sprint is thought to interact with the PVR RTK though its SH2 domain and knockdown of PVR or its ligand PVF2 reduced the number of Sprint puncta, consistent with endocytosis of a PVR-Sprint complex. The knockdown of Rab5 and shibire as well as Ras64B and Ras85D, reduced the number of Sprint positive endocytic punta compared to the control (Renilla dsRNA). This is consistent with the Sprint puncta being early endocytic structures and indicates that Ras

144 regulates the endocytosis of Sprint, which is similar to the function of the mammalian RIN1 (Jekely et al., 2005; Tall et al., 2001). In addition, consistent with the results of Sprint localisation with Rab5 but not Rab7 endocytic vesicles (Figure 5.2), Rab7 knockdown did not affect the number of Sprint positive puncta, indicating that a disruption of the Rab7 late endosomal compartments does not affect the rate of Sprint positive endosome formation. Although the knockdown of most genes performed in this study did not seem to affect the cell numbers, the knockdown of shibire and Rab5 genes, however, did seem to have an effect on Sprint expression, which is reflective of the importance of the aforementioned proteins in cellular processes, including cell survival. The knockdown of Abl protein tyrosine kinase did not have an effect on the Sprint puncta formation. This suggests either that Abl and its potential phosphorylation of Sprint is not important in Sprint’s endocytic function or that other protein tyrosine kinases such as PVR, known to be expressed in S2 cells (Sims et al., 2009), are more important in Sprint tyrosine phosphorylation and endocytic vesicle formation. It is important to note that the level of knockdown of the genes were not tested at a transcript or protein level; however, it is likely that the knockdown procedure was successful since the dsRNA used have been validated in S2 cells for knockdown efficiency (Brown, 2010) and some of the knockdowns have elicited a response in this experiment relative to the Renilla dsRNA control.

In this study RasGAP was found to co-localise with Sprint and Rab5 vesicles, but its localisation to endosomes required co-expression of Sprint. Moreover, when Sprint mutant proteins localised to the nucleus, RasGAP also localised to the nucleus. SH2 mutant RasGAP did not co-localise with Sprint or Rab5. These observations indicate that localisation of RasGAP to early endosomes is mediated through its interaction with Sprint. The nuclear localisation of ΔVPS9 Sprint is a novel observation and it could reflect role of Sprint interacting proteins Rab5 and Ras in defining the special distribution of Sprint. In addition, since the mammalian homologue of Sprint, RIN1, have been shown to interact with RTKs, it is likely that ΔVPS9 Sprint is associated with RTKs such as PVR since its SH2 domain is intact. This would present a novel signalling mechanism for Drosophila RTKs through endocytosis, which is consistent with the nuclear localisation and signalling of mammalian FGFR and EGFR molecules (Bryant and Stow, 2005).

The results in this chapter are consistent with a model where Sprint is internalised along with activated PVR into Rab5 positive early endosomes. The Rab5-GEF activity of Sprint is stimulated by association with Ras, but association of RasGAP with Sprint locally

145 regulates Ras-GTP levels thereby limiting Sprint Rab5-GEF activity and the formation of active Rab5.

5.3.2 Limitations of this study The major limitation of this study is that all the localisation experiments were performed using overexpressed proteins. It would be ideal to confirm the co-localisation of endogenous wild-type proteins; however, Sprint antibodies are no longer available. One caveat to observing endogenous Sprint-RasGAP co-localisation is that it is likely to be transient and only a small fraction of RasGAP may co-localise with Sprint on early endosomes. The other limitation of this study is that the GFP-Sprint construct used was an N-terminally truncated form of Sprint hence the behaviour of the wild-type Sprint is still unknown. However it is important to note that the GFP-Sprint construct had all the evolutionarily conserved domains and the regions shared by Sprint-a and Sprint-b.

5.3.3 Future experiments In chapter 4, biochemical association between Sprint, Abl and RasGAP were shown, hence it would be interesting to see if Abl, thought to act as a protein tyrosine kinase (PTK) for Sprint, can also co-localise with Sprint-Rab5-RasGAP complex using an IHC experiment. This would provide evidence for the role of Abl in endocytic processes. In chapters 3 and 4, Sprint was shown to be tyrosine-phosphorylated, which might influence its puncta formation. Since Abl associates with Sprint, the effect of Abl knockdown on Sprint puncta formation was tested, however, Abl did not have an effect. Hence it would be interesting to see if inhibiting other protein tyrosine kinases using broad range tyrosine kinase inhibitors (AG 18; Millipore) would have an effect on Sprint puncta formation, hence deciphering the role of Sprint tyrosine phosphorylation in puncta formation. Since ΔVPS9 Sprint was capable of forming puctate structures in S2 cell nucleus it would be interesting to test the nature of these puncta by treating cells with Dynasore or shibire dsRNA (Brown, 2010) in order to see if these puncta are endocytic and originate from the cell cytoplasm, using live cell imaging. It would also be interesting to treat ΔVPS9 Sprint expressing cells with pvf2 and pvr dsRNA in order to see if these puncta are endocytosing receptors to the nucleus for signalling.

146

CHAPTER 6: Effects of Rab5 on RTKs

6.1 Introduction

6.1.1 Introduction to the Rab5 mediated RTK endocytosis and signalling The results in chapters 3-5 of this thesis implicate RasGAP and Sprint in the regulation of Rab5 activity in S2 cells. Sprint has been implicated in the regulation of endocytosis of two RTKs, EGFR and PVR, in the Drosophila ovary (Jekely et al., 2005), and its mammalian homologue RIN1 regulates endocytosis of several RTKs including the EGFR (Tall et al., 2001), transforming growth factor β (TGF-β) (Hu et al., 2008) and the insulin receptor (Hunker et al., 2006c). Endocytosis of RTKs such as EGFR is important for receptor trafficking and degradation (Chen et al., 2009), which subsequently effects their turnover (Carter and Sorkin, 1998) and signalling output (Barbieri et al., 2004; Barbieri et al., 2000). The role of the Drosophila ESCRT-0 endocytic complex (Hrs and Stam) as well as endosomal protein Myopic has been extensively studied in regulation of RTK signalling and endocytosis (Chanut-Delalande et al., 2010; Lloyd et al., 2002; Miura et al., 2008). However, little is known about the role of Rab5 and its activators on RTK endocytosis and degradation in Drosophila. One of the reasons for this is that there are no convenient cell culture models for RTK endocytosis in Drosophila. In this chapter I sought to develop a Drosophila cell culture model for RTK endocytosis in order to investigate the roles of RasGAP, Sprint and Rab5 in this process. The approach taken was to exploit the recent development of chimeric PVR and EGFR proteins with endo-domain of Drosophila RTKs attached to human EGFR transmembrane and extracellular domains, which can be activated by mammalian EGF and related ligands (Inaki et al., 2012). The development of such a system would allow the observation of RTK endocytosis in real time by using fluorescently labelled ligands, as has been used successfully in mammalian tissue culture cells (Carter and Sorkin, 1998). With such a system in place it would then be possible to manipulate the activity of Sprint, RasGAP, Rab5 and other components and assess the effects on RTK endocytosis.

147

6.2 Results

6.2.1 Characterising wild-type and chimeric PVR and EGFR In order to assess the activity and function of chimeric PVR and EGFR, S2 cells expressing the receptors were treated with the human EGF. However, when stimulated with EGF there was no clear increase in tyrosine phosphorylation at the molecular weight of the receptors and no increase in the level of activated (diphosphorylated) ERK (Figure 6.1). The EGF used in these experiments was active as it was capable of activating MAPK pathway in mammalian HEK293 cells (Figure S7). The protein bands at the molecular weight of wild- type and chimeric receptors showed similar tyrosine phosphorylation levels (Figure 6.2) at high or low expression levels. This indicates that RTK tyrosine phosphorylation in wild- type and chimeric constructs represent the background or constitutive tyrosine phosphorylation. Interestingly, wild-type (Figure 6.2 and 6.4A) and chimeric (Figure 6.2 and 6.5A) EGFR produced only one fast-migrating fragment of approximately 120 kDa in size, which is detected by anti-GFP antibody against the EGFR extreme C-terminus GFP tag. The molecular weight corresponding to this fragment however was not phosphorylated above the background (Figure 6.2), which suggests that this fragment is not an active cleavage product of the EGFR. When comparing wild-type and chimeric EGFR constructs, the majority of the wild-type EGFR was cleaved to a fragment of approximately 120 kDa in size, a feature less pronounce in chimeric EGFR construct (Figure 6.2). This could reflect the processed form of wild-type EGFR by the S2 cell machinery (Lloyd et al., 2002).

The failure of the chimeric receptors to be activated by EGF could be the result of abnormal intracellular trafficking of the receptors so that they are not localised at the plasma membrane. In order to assess the localisation of wild-type and chimeric PVR and EGFR molecules, they were overexpressed in S2 cells and their cellular localisations were determined using fluorescence microscopy. As shown in Figure 6.3, chimeric PVR and EGFR molecules at both high and low expression levels have both cytoplasmic and punctate distribution, which are positive for phosphotyrosine stain; however, wild-type PVR seems to have a plasma membrane distribution, with some punctuate distribution. Interestingly, EGFRWT has a punctate distribution, which is highly tyrosine- phosphorylated. In addition, PVR expressing S2 cells were enlarged and flattened, which indicates that downstream PVR signalling is activated in these cells (Sims et al., 2009).

148

GFP hE-PVRWT hE-EGFRWT

Non-starved Starved Non-starved Starved Non-starved Starved 100 ng/ml A - + - + - + - + - + - + hEGF 220

100 Cell extract 60 WB: -GFP

B 1 2 3 4 5 6 7 8 9 10 11 12 220

Cell extract WB: -pTyr 100 C 1 2 3 4 5 6 7 8 9 10 11 12 45 Cell extract WB: -dpERK 30 D 1 2 3 4 5 6 7 8 9 10 11 12 45 Cell extract WB: -MAPK 30 E 1 2 3 4 5 6 7 8 9 10 11 12 60 Cell extract 45 WB: -Tubulin

1 2 3 4 5 6 7 8 9 10 11 12

Figure 6.1 Chimeric PVR and EGFR constructs become tyrosine-phosphorylated and activated to the same extent. (A) Drosophila S2 cells were either transfected with GFP control or chimeric (hE) PVR and EGFR constructs (large filled arrowhead) under FBS starvation or non-starvation conditions and were induced with 100 ng/ml human EGF (hEGF) and were western-blotted using anti-GFP antibody (B) GFP, hE-PVRWT and hE-EGFRWT constructs were tested for tyrosine phosphorylation using anti-pTyr antibody (large filled arrowhead) and they were also tested for (C) MAPK activation using anti-dpERK antibody (large filled arrowhead) and (D) as a load control cell lysates were western-blotted with anti-MAPK antibody (large filled arrowhead). (E) Drosophila S2 cell extracts were western-blotted with anti-Dtub antibody (small filled arrowhead). 149

High Low

GFP PVR EGFR GFP PVR EGFR

A hE WT hE WT hE WT hE WT 220

100 Cell extract 60 WB: -GFP

30 B 1 2 3 4 5 6 7 8 9 10 220

Cell extract WB: -pTyr 100 C 1 2 3 4 5 6 7 8 9 10 60 Cell extract WB: -Tubulin 45 1 2 3 4 5 6 7 8 9 10

Figure 6.2 Wild-type and chimeric PVR and EGFR constructs become tyrosine-phosphorylated to the same extent. (A) Drosophila S2 cells were transfected with wild-type (WT) and chimeric (hE) PVR and EGFR constructs (large filled arrowhead) and were western-blotted using anti-GFP antibody (B) PVR and EGFR constructs were tested for tyrosine phosphorylation using anti-pTyr antibody (large filled arrowhead). (C) Drosophila S2 cell extracts were western-blotted with anti-Dtub antibody (small filled arrowhead).

150

Figure 6.3 Determining PVR and EGFR cellular localisation. Drosophila S2 cells were transfected with GFP control, chimeric (hE) and wild-type (WT) PVR and EGFR constructs, which were either expressed at high or low concentrations. EGFRWT construct produced punctate structures (long arrowheads) in S2 cells. Scale bars on the photographs represent 5 μm.

151

6.2.2 Rab5 does not regulate wild-type and chimeric PVR and EGFR degradation To assess whether Rab5 activity can affect the stability of PVR and EGFR, Drosophila YFP-Rab5 constructs (wild-type, dominant-negative and constitutively active) were overexpressed with C-terminal GFP-tagged wild-type and chimeric PVR and EGFR molecules and the level of expression of each receptor was determined by western-blotting against GFP. If Rab5 regulates the endocytosis and degradation of the receptors as it does in mammalian cells, then constitutively active Rab5 would be expected to decrease the level of RTK as a result of increased degradation whereas dominant negative Rab5 would be expected to increase the level of receptor. As shown in Figure 6.4, constitutively active Rab5 did not decrease wild-type RTK levels and dominant negative Rab5 did not seem to increase RTK levels. This lack of effect of Rab5 on steady-state levels of the RTKs seems to be replicated in chimeric RTKs (Figure 6.5). Surprisingly however, all of the Rab5 constructs seemed to increase the steady-state levels levels of wild-type (Figure 6.4) and chimeric (Figure 6.5) RTKs. This is perhaps reflective of the dominant negative effect of Rab5 on RTK steady-state levels.

Since Rab5 is known to endocytose active RTKs, the levels of tyrosine-phosphorylation corresponding to the molecular weight of chimeric RTKs were assessed. The level of tyrosine phosphorylation, corresponding to the molecular weight of RTKs seems to depend on the level of receptor expression (Figure 6.6), which is likely to represent the background or constitutively active RTK tyrosine phosphorylation. This is the case since addition of the mammalian EGF under (FBS) starvation or non-starvation did not increase the level of chimeric RTKs tyrosine phosphorylation (Figure 6.7).

152

PVRWT EGFRWT

Rab5 Rab5

A WT DN CA WT DN CA 220

Cell extract 100 WB: -GFP

60 B 1 2 3 4 5 6 7 8 60 Cell extract WB: -Tubulin 45

1 2 3 4 5 6 7 8

Figure 6.4 Rab5 does not regulate wild-type PVR and EGFR steady-state levels. (A) Drosophila S2 cells were either transfected with wild-type (WT) PVR and EGFR (large filled arrowhead) or they were co- transfected with WT, dominant negative (DN) or constitutively active (CA) Rab5 constructs (small filled arrowhead). EGFRWT construct produced a degradation/cleavage product (hollow arrowhead). PVR, EGFR and Rab5 constructs were western-blotted using anti-GFP antibody. (B) Drosophila S2 cell extracts were western-blotted with anti-Dtub antibody.

153

hE-PVR WT hE-EGFRWT

Rab5 Rab5

A WT DN CA WT DN CA

220

Cell extract WB: -GFP 100

60 B 1 2 3 4 5 6 7 8 60 Cell extract WB: -Tubulin 45 1 2 3 4 5 6 7 8

Figure 6.5 Rab5 does not regulate chimeric PVR and EGFR steady-state levels. (A) Drosophila S2 cells were either transfected with chimeric (hE) PVR and EGFR (large filled arrowhead) or they were co- transfected with wild-type (WT), dominant negative (DN) or constitutively active (CA) Rab5 constructs (small filled arrowhead). hE-PVRWT and hE-EGFRWT constructs produced a degradation/cleavage product (hollow arrowheads). PVR, EGFR and Rab5 constructs were western-blotted using anti-GFP antibody. (B) Drosophila S2 cell extracts were western-blotted with anti-Dtub antibody.

154

hE-PVR WT hE-EGFRWT A 220 Cell extract WB: -GFP

100 B 1 2 3 4 5 6 7 8 9 10 11 12 13 14 220

Cell extract WB: -pTyr 100 C 1 2 3 4 5 6 7 8 9 10 11 12 13 14 60 Cell extract WB: -Tubulin 45

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Figure 6.6 The level of chimeric PVR and EGFR tyrosine phosphorylation depends on their level of expression. (A) Drosophila S2 cells were transfected with different amounts of the chimeric (hE) PVR and EGFR constructs (large filled arrowhead) and were western-blotted using anti-GFP antibody. (B) hE-PVRWT and hE-EGFRWT constructs were tested for tyrosine phosphorylation using anti-pTyr antibody (large filled arrowhead). (C) Drosophila S2 cell extracts were western-blotted with anti-Dtub antibody (small filled arrowhead).

155

hE-PVRWT hE-EGFRWT

Non-starved Starved Non-starved Starved

Time 100 ng/ml (Min) 0 10 20 0 10 20 0 10 20 0 10 20 hEGF A 220

Cell extract WB: -GFP

100

1 2 3 4 5 6 7 8 9 10 11 12

B 220

Cell extract WB: -pTyr 100

1 2 3 4 5 6 7 8 9 10 11 12 60 C Cell extract 45 WB: -Tubulin 1 2 3 4 5 6 7 8 9 10 11 12

Figure 6.7 The chimeric PVR and EGFR molecules do not get activated by human EGF. (A) Drosophila S2 cells were transfected with the chimeric (hE) PVR and EGFR constructs (large filled arrowhead) under FBS starvation or non-starvation conditions and were incubated with 100 ng/ml human EGF (hEGF) for up to 20 minutes. Cell lysates were western-blotted using anti-GFP antibody (B) hE-PVRWT and hE-EGFRWT constructs were tested for tyrosine phosphorylation using anti-pTyr antibody (large filled arrowhead). (C) Drosophila S2 cell extracts were western-blotted with anti-Dtub antibody (small filled arrowhead).

156

6.3 Discussion and conclusions

6.3.1 Incorrect intracellular trafficking of the chimeric RTKs could explain their insensitivity to Rab5 overexpression In order to assess the effects of Rab5 on RTK stability, different forms of Rab5 were expressed with PVR and EGFR molecules (Figure 6.4 and 6.5); however, unlike their mammalian homologues (Chen et al., 2009) the steady-state levels of the RTKs were not appreciably affected by Rab5 overexpression. As endocytosis of RTKs is upregulated when they become active (receptor-mediated endocytosis) the lack of effect of Rab5 (Figure 6.4 and 6.5) could be due to the lack of activation of the receptors (Figure 6.2 and 6.7) and indeed the receptors failed to activate the MAPK pathway under different conditions or have an increased levels of tyrosine phosphorylation when cells were stimulated with EGF (Figure 6.1). The high background/constitutively active tyrosine phosphorylation of chimeric RTK (Figure 6.6 and 6.7) constructs can be explained by their high level of expression, known to lead to loss of receptor kinase autoinhibition (Belov and Mohammadi, 2012) and by the C-terminus location of the GFP tag attached to those constructs, which could prevent their inhibition of tyrosine phosphorylation (Hubbard, 2004). The incorrect localisation of the chimeric RTKs is also though to be a contributing factor in the failure of their activation and processing/degradation through early stage endocytosis. It is thought that this incorrect localisation of the chimeric RTKs (Inaki et al., 2012) is due to the lack of S2 cell endoplasmic reticulum (ER) machinery recognition of mammalian N-terminal signalling peptide and transmembrane domain signal anchor sequence, which can lead to the failure of transmembrane protein processing across the ER trafficking machinery (Rapoport, 2007).

Although wild-type PVR and EGFR molecules did not seem to have any different tyrosine phosphorylation levels compared to the chimeric RTKs, they seemed to partially localise to the cell membrane and PVR seemed to have been activating a pathway, which leads to an increase in the size of the S2 cells. This finding is consistent with the previous literature demonstrating that PVR activation and Ras signalling through MAPK and PI3K can lead to an increase in Drosophila cell size (Sims et al., 2009). This activation state of PVR is likely to be through S2 cell PVF2 ligand secretion (Cho et al., 2002; Heino et al., 2001; Sims et al., 2009). The wild-type EGFR molecule was present in S2 cells in form of puctate structures, which is likely to reflect its state of endocytosis and cleavage (Miura et

157 al., 2008). This is supported by the fact that biochemically EGFR produced only one detectable cleavage product. Interestingly however, this fragment did not seem to be tyrosine-phosphorylated above the background, which is likely to be due to its lack of activation due to the absence of the EGFR ligand. This is consistent with the lack of EGFR and its ligand production in S2 cells and the lack of Drosophila EGFR ligand in FBS (Friedman and Perrimon, 2006). The lack of chimeric PVR and EGFR activation in S2 cells is in direct contrast to the results obtained by Pernille Rorth group showing that overexpression of the chimeric RTKs and their ligands lead to their activation and directs border cell migration (Inaki et al., 2012). This discrepancy between the results is likely due to border cells expressing both ligand and the receptors in the same cell, leading to activation of the chimeric RTKs in the cytoplasm and not the cell membrane. This is consistent with the results of this study, showing that chimeric RTKs were incorrectly localised in S2 cells, which is likely to have lead to their lack of MAPK activation.

6.3.2 Limitations of this study The major limitation of this study is that all the experiments were performed using overexpressed proteins including RTKs, which is known to lead to loss of receptor kinase autoinhibition (Belov and Mohammadi, 2012). This can account for the RTK concentration dependent tyrosine phosphorylation (Figure 6.6) and failure of the chimeric RTKs to activate the MAPK pathway (Figure 6.1). Hence it would be important to drive RTK expression using an endogenously regulated expression driver (for example PVR’s natural promoter) that can replicate the levels of endogenous receptors in S2 cells, or titrate the levels of RTK to endogenous levels. Another major limitation of this study was the use of total cell lysate to address RTK tyrosine phosphorylation levels. This technique can not distinguish the tyrosine phosphorylation levels of overexpressed RTKs from background proteins at the corresponding molecular weight, which can mask any changes in the phosphotyrosine levels between different conditions. Hence it would be interesting to repeat the above experiments using pulled-down RTKs and assess their tyrosine phosphorylation. Although the wild-type and chimeric EGFR and PVR constructs did not become tyrosine-phosphorylated above the background levels and the chimeric RTKs failed to activate the MAPK pathway, it is possible that the effects of the transfection efficiency of the RTK constructs masked any differences between the GFP control, ligand treated and untreated cells. Hence it would be important to increase the transfection efficiency of the RTK constructs in the future experiments and repeat the above

158 experiments. It is also possible that the chimeric as well as wild-type RTK constructs are poor activators of the MAPK signalling pathway in S2 cells, consistent with previous findings (Ishimaru et al., 2004). In addition, low levels of endogenous ERK and dpERK could also mask any difference between the control and RTK expressing S2 cells.

6.3.3 Future experiments As discussed above, the high background/constitutively active tyrosine phosphorylation of chimeric RTK (Figure 6.6 and 6.7) constructs can be explained by the C-terminus location of the GFP tag attached to those constructs, which could prevent their inhibition of tyrosine phosphorylation (Hubbard, 2004). Hence in order to assess the effects of C-terminus GFP tag on the chimeric and wild-type RTK constructs background/constitutively active tyrosine phosphorylation levels, the RTK GFP tag could be removed and replaced with a smaller tag such as myc or FLAG tag and all the above experiments regarding receptor tyrosine phosphorylation could be performed again. The levels of chimeric RTK activation of MAPK were assessed biochemically, and failed to show a difference between ligand treated and untreated cells. Hence it would be interesting to assess the chimeric RTK activation of MAPK using immunohistochemistry based dpERK (indicator of MAPK signalling activation) imaging of RTK transfected S2 cells (James et al., 2007). This will address the transfection efficiency problem encountered by assessing dpERK levels using biochemical techniques.

In this study, RTKs were not effected in their steady-state levels by overexpression of Rab5 constructs. Hence it would be interesting to assess the effects of endocytic proteins such as Rab5 and its GEFs, Sprint, GAPEX-5 and RABEX-5 (Yan et al., 2010) on RTK endocytosis using biotinylation of surface proteins (Chang et al., 2012) or overexpressed RTKs (Strubbe et al., 2011). This would allow quantification of endocytosed RTKs through streptavidin affinity purification and antibody staining of endocytosed proteins. Rab5 mediated endocytosis of receptors has been observed to affect RTK signalling in mammals (Barbieri et al., 2004). Therefore it would be interesting to assess the effects of Drosophila Rab5 and its GEFs on different RTK signalling outputs. This can be achieved through knocking down different components of endosomes and RTK signalling pathway using dsRNA (Brown, 2010) and observing RTK signalling output using transcriptional output enzyme-based reporter assays such as luciferase assay (Chatterjee and Bohmann, 2012; Tootle et al., 2003). This is likely to yield promising results since a similar study

159 using Drosophila ERK (Rolled) has shown the effects of Sprint, Ras64B, Ras85D and PVR in regulating MAPK signalling pathway (Friedman and Perrimon, 2006).

160

CHAPTER 7: The vap mutant neurodegenerative phenotype in Drosophila is linked to early stage endocytosis

7.1 Introduction

7.1.1 Introduction to autophagy Autophagy is primarily a cell survival process, which under certain circumstances leads to (type II) cell death (Maiuri et al., 2007). Autophagy is important in clearing cells of damaged or superfluous organelles and proteins and generating nutrients required for cell survival. Autophagy is present at basal levels in normal growing conditions but is upregulated under starvation, hypoxia, intracellular stress, high temperature and high cell culture density. Autophagy is a catabolic process involving bulk degradation of the cytoplasm through engulfment of part of the cytoplasm in double membrane vesicles called autophagosomes, which fuse with lysosomes to form autophagolysosomes (Melendez and Neufeld, 2008). There are three forms of autophagy: chaperone-mediated autophagy, microautophagy and macroautophagy (Levine and Kroemer, 2008). Autophagy is often initiated through the inactivation of the protein kinase target of rapamycin, such as mTOR1 in mammals or TOR1 in fission yeast, which are upstream of more than 20 proteins referred to as the autophagy-related (ATG) (Juhasz et al., 2008; Levine and Kroemer, 2008; Melendez and Neufeld, 2008). In yeast, autophagy is regulated by TOR-dependent signalling, which regulates the association of Atg1, Atg13 and Atg17 (Kamada et al., 2000; Melendez and Neufeld, 2008), and this mechanism is conserved in higher eukaryotes such as Drosophila (Chang and Neufeld, 2009; Scott et al., 2007) and humans (Mercer et al., 2009). Class I PI3Ks (p110-p85) are known to repress autophagy by linking RTKs to TOR activation while class III PI3Ks (VPS34-VPS15) initiate autophagy both in mammals (Dou et al., 2013; Levine and Kroemer, 2008; Rubinsztein et al., 2007) and Drosophila (Juhasz et al., 2008).

7.1.2 Autophagic neurodegeneration Autophagy is important in eliminating misfolded, aggregated and damaged proteins and it is known to suppress neurodegenerative phenotypes caused by the expression of the aggregate-prone proteins in various mouse and Drosophila neurodegenerative models

161

(Hara et al., 2006; Juhasz et al., 2007; Komatsu et al., 2006; Simonsen et al., 2008) and it plays an essential role in maintenance of axonal homeostasis (Melendez and Neufeld, 2008). Autophagic neurodegenerative disorders can be classified into two categories in both mammals and Drosophila: those affecting formation of autophagosomes and those affecting lysosomal clearance or sequestration and movement or maturation of autophagosomes, which are found in Battern disease and Danon diseases (Levine and Kroemer, 2008; Venkatachalam et al., 2008). However, it is thought that over-activation of the autophagic pathway in Drosophila can also lead to neurodegeneration (Botella et al., 2003).

7.1.3 Rationale of the study In chapters 3-6 of this thesis a novel mechanism of Rab5 regulation through RasGAP and Sprint has been proposed based on studies in S2 cells, an embryonic cell line thought to be derived from haemocytes (Schneider, 1972). In a previous study, Drosophila mutants defective in the gene encoding RasGAP (vap) showed an age-related neurodegenerative phenotype in the adult brain in which the dying neurons showed morphological features of autophagy (Botella et al., 2003). Genetic interactions with components of the EGFR-Ras pathway suggested that the neuronal cell death in vap mutants was the result of excessive EGFR-Ras signalling. However, the Ras effector pathway responsible for mediating neurodegeneration was not clear. The aim of this chapter is to determine whether the RasGAP-Sprint-Rab5 pathway plays a role in mediating the neurodegeneration observed in vap mutant flies.

7.2 Results

7.2.1 Mutant Sprint or Rab5 supresses the vap mutant neurodegeneration phenotype in the adult Drosophila brain Genetic interactions between mutant alleles are a powerful way to identify genes that function together in biochemical or signalling pathways in vivo. A previous study demonstrated that the severity of the neurodegeneration phenotype observed in flies carrying the hypomorphic vap2 allele could be modified (enhanced or suppressed) by the presence of mutations in components of the EGFR-Ras signalling pathway (Botella et al., 2003). In this study I have used a similar approach to investigate whether mutations in the genes encoding Sprint and Rab5 modify the neurodegeneration phenotype of vap2 mutant flies. As the vap and spri genes are both on the X chromosome, RasGAP (vap2) and Sprint

162

(spri6G1) double mutant stocks were constructed by recombination and their genotypes were confirmed by PCR (Figure S8 and S9). The adult brain phenotypes of the double mutant (vap2, spri6G1) flies were then compared with those from single vap2 and spri6G1 mutant flies. As shown in Figure 7.1B and H, vap mutant hypomorphic allele (vap2) produced a characteristic neurodegeneration phenotype (Botella et al., 2003) in 18-20 days old adult males, which was in a sharp contrast to the control (y, w) flies (Figure 7.1A and H). However, in two RasGAP (vap2) and Sprint (spri6G1) double mutant stocks the neurodegenerative phenotype was significantly supressed (Figure 7.1D, E and H). This demonstrates a strong epistatic interaction between Sprint and RasGAP mutations.

Since Sprint is a putative Rab5-GEF (Jekely et al., 2005), it was important to determine whether Rab5 also interacted genetically with vap. This was achieved by crossing a hypomorphic Rab5 mutant (Rab54) allele (Wucherpfennig et al., 2003) into vap mutant background. As shown in Figure7.1G and H, Rab5 heterozygous mutant (Rab54/+) was capable of significantly supressing the vap mutant neurodegenerative phenotype, although to a lesser extent than the homozygous spri mutant. Therefore both Sprint and Rab5 show clear genetic interactions with RasGAP in vivo.

7.2.2 spri and vap both effect Drosophila survival rate In addition to neurodegeneration, vap mutant flies show reduced lifespan, although whether this is the result of impaired brain function or has some other cause is not known (Botella et al., 2003). In order to test the effects of loss of Sprint function on lifespan in a vap mutant background, the survival of male flies of different genotypes was determined. As expected vap2 flies had a reduced survival rate relative to the wild-type control flies (Figure 7.1I). Interestingly however, Sprint mutant flies (spri6G1) also had a reduced survival rate, which has not yet been reported. In addition, the vap2 and spri6G1 double mutant flies had a further reduced life expectancy, which was greater than any of the mutant genotypes individually. This consistent with the hypothesis that vap and spri gene mutations both affect the survival rate of flies but through different mechanisms.

163

A B C cb loc me la re

y, w vap2 spri6G1 D E F

vap2, spri6G1 w-, vap2, spri6G1 Rab54/+ G

vap2; Rab54/+

H 120 *** I 100 vap2 y, w 90 ) 100 spri6G1

80 2 6G1 %

) vap , spri (

70 w-, vap2, spri6G1 %

n 80

(

o l

i 60

a a

60 v 50

i v

l 40

r o

40 u

u 30

S c

a 20 20 v 10 0 2 1 0 G 1 1 + w 6 G G 4 / 4 + , p 6 6 / y a ri i i 5 5 0 10 20 30 40 50 60 70 80 v p r r b b s p p a a s s R 2 , 2 , R p 2 ; a p p Time (Days) v va a - , v w

Figure 7.1 Mutant Sprint or Rab5 suppresses the vap mutant neurodegenerative phenotype. Drosophila heads from (A) y, w (B) vap2 (C) spri6G1 (D) vap2, spri6G1 (E) w-, vap2, spri6G1 (F) Rab54/+ and (G) vap2; Rab54/+ stocks, aged 18-20 days, were sectioned and stained using toluidine blue and (H) the percentage vacuolisation of the stocks relative to the vap2 were assessed (n ≥ 3). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001). Scale bars on the photographs represent 100 μm. The arrow shows neurodegeneration (vacuolisation). cb, central brain; la, lamina; loc, lobula complex; me, medulla; re, retina. (I) The rate of survival (longevity) of each stock was assessed with n > 34 flies aged.

164

7.3 Discussion and conclusions

7.3.1 Drosophila vap mutant autophagic neurodegeneration may be the result of deregulation in early stage endocytosis Although links between endocytosis and neuronal survival have been established before with mutations in the mammalian Rab5-GEF amyotrophic lateral sclerosis 2 (ALS2) leading to motor neuron disease (Hadano et al., 2001; Otomo et al., 2003; Yang et al., 2001) and Huntington’s disease protein, Huntingtin, being part of a Rab5 effector complex (Pal et al., 2006), this is the first study to demonstrate the involvement of endocytic molecules Rab5 and Sprint in neuronal survival in Drosophila and implicate both of these endocytic proteins in mediating neurodegeneration in the RasGAP (vap) mutant. The epistatic suppression of the vap mutant neurodegeneration phenotype by the spri mutation shows that Sprint is required to mediate neuronal cell death when RasGAP function is lost. This argues strongly that RasGAP and Sprint both function in the same pathway to regulate neuronal survival in the adult brain and that Sprint functions ‘downstream’ of RasGAP in this pathway. Based on these results a plausible explanation for the neurodegeneration found in the vap mutant is that loss of RasGAP activity increases the activity of Sprint because more Ras-GTP is available on early endosomes. This leads to an increase in the level of active Rab5-GTP through the GEF activity of Sprint and that it is the increase in Rab5 activity that leads to neuronal cell death. The weaker but significant suppression of the vap mutant neurodegeneration by heterozygosity for the Rab5 mutation is consistent with this interpretation and indicates that Rab5 activity is crucial in maintaining neuronal survival.

A previous study showed that the vap mutant neurodegeneration phenotype was modified (suppressed or enhanced) by altering activity of the EGFR-Ras pathway: reducing EGFR- Ras activity suppressed the vap phenotype whereas increasing EGFR-Ras activity enhanced the vap phenotype (Botella et al., 2003). How increased EGFR-Ras signalling could lead to neuronal cell death is not clear. RTK signalling usually promotes cell survival through the MAPK and PI3K pathways (Bergmann et al., 1998; Kurada and White, 1998) but increased Ras activity can also lead to cell death in some systems (Chi et al., 1999; Elgendy et al., 2011; Serrano et al., 1997; Young et al., 2009). Indirect evidence, through studies of the negative regulator of mammalian Ras such as RASA1/p120- RasGAP (Henkemeyer et al., 1995) and brain specific synaptic RasGAP (synGAP)

165

(Knuesel et al., 2005), show that Ras signalling is also detrimental to neuronal survival, supporting the dual role of Ras in cellular survival. Botella and co-authors investigated which signalling pathways might be responsible for causing neuronal cell death in the vap mutant (Botella et al., 2003). They found no genetic interaction with Raf or MAPK (rolled) mutations. Although overexpression of wild-type Ras1 enhanced the vap neurodegeneration phenotype, overexpression of Ras1 effector site mutants that are reported to specifically activate the Ras effectors PI3K or Ral-GDS did not; a Ras1 effector site mutant specifically activating Raf could not be tested as it caused lethality. From these studies there is no clear evidence that either the MAPK, or the PI3K or the Ral-GDS pathways downstream of Ras are responsible for mediating neurodegeneration. Hence the link between EGFR-Ras signalling and Sprint and Rab5 endocytic molecules is an exciting development in further understanding the mechanism of RasGAP (vap) neuroprotective function.

How increased Rab5 activity, within the context of EGFR-Ras signalling, could lead to neuronal cell death is an interesting question. One model to explain the results presented here is that Rab5 could function by regulating the activity of the EGFR-Ras pathway: increased Rab5 activity in the vap mutant would augment EGFR-Ras signalling and this would lead to neuronal cell death. It is well established in both mammals (Barbieri et al., 2004; Barbieri et al., 2000; Chen et al., 2009) and Drosophila (Chanut-Delalande et al., 2010; Lloyd et al., 2002; Miura et al., 2008) that the EGFR is regulated by endocytosis. Ligand stimulated endocytosis of the EGFR receptor in mammalian tissue culture cells is the classic example of receptor-mediated endocytosis; the internalised ligand-receptor complex passes through endocytic compartments to the lysosome where it is degraded, thereby limiting EGFR signalling. In this classical mammalian in vitro system endocytosis inhibits EGFR signalling (Chen et al., 2009), however, in vivo studies in Drosophila show that endocytosis can stimulate EGFR signalling (Chanut-Delalande et al., 2010; Miura et al., 2008). An alternative model to explain the results presented here is that active Rab5 directly causes neuronal cell death and that the role of the EGFR-Ras pathway is to stimulate Sprint activity. In this model Ras-GTP associated Sprint promotes active Rab5- GTP production, leading to an enhanced association of Rab5 to its effectors such as pro- autophagic class III PI3K (VPS34), which generates PIP3 leading to autophagosome vesicle nucleation (Simonsen and Tooze, 2009) and causes neurodegeneration. Consistent with this model, mutations in the vap gene have been shown to upregulate VPS34, shown

166 to be required for autophagosomes formation in fat bodies of starved Drosophila (personal communication with Marc Bourouis, Pierre Léopold lab). These two models of EGFR-Ras signalling involvement in vap mutant neurodegenerative phenotype are not mutually exclusive. Therefore it is likely that in wild-type flies Sprint associates to RTKs in an SH2- dependent manner. Tyrosine phosphorylation of Sprint, either by RTKs or by the cytoplasmic PTK Abl (Hu et al., 2005), would permit RasGAP localisation to this complex and allow localised regulation of Ras-GTP levels and the GEF activity of Sprint. This in turn would modulate the rate of Rab5 mediated RTK endocytosis within newly formed endosomes (Tall et al., 2001) and allow the homeostasis of Drosophila neurons. However, when RasGAP (vap) is mutated in Drosophila, excessive Ras-GTP accumulates, leading to over-activation of Sprint and subsequently Rab5 over-activation. This would in turn increase the rate of RTK endocytosis and signalling, leading to excessive activation of autophagic signalling pathways through Rab5 and VPS34, eventually causing neurodegeneration in the adult Drosophila brain tissue. Many details of this model still remains to be demonstrated, however, considering neurons are highly susceptible to disruptions in endocytic pathways (Zhou et al., 2010), this model can potentially account for the events in the vap mutant flies.

This study has highlighted not only Drosophila neurons in the adult brain are susceptible to changes in endocytic pathways, but their lifespan also seems to be affected since spri6G1 flies seemed to have a reduced survival rate relative to the wild-type flies. This was independent of the effects of vap mutation on the rate of Drosophila survival, indicating that there are perhaps different mechanisms responsible for regulating the rate of Drosophila survival.

7.3.2 Limitations of this study This study has suggested a mechanism whereby endocytosis of RTKs such as EGFR and its downstream signalling mediates the autophagic neurodegeneration in the vap mutant Drosophila. However, This study has not provided direct evidence that RasGAP regulates EGFR endocytosis in the brain. Also this study has not provided evidence that initiation of autophagy is upregulated in the vap mutant flies since autophagy could be a last survival attempt by the neurons. This study has also failed to determined how endocytosis affects EGFR/Ras signalling pathway, known to be involved in age-related brain neurodegeneration in Drosophila mutant vap (Botella et al., 2003). In addition, this study

167 has not identified the role of endocytosis in inducing autophagy and activation of the downstream autophagic molecules in the vap mutant flies.

7.3.3 Future experiments Although Sprint mediated early stage endocytosis seems to play a role in the vap mutant neurodegenerative phenotype in Drosophila; it is unclear whether this effect is specific to Sprint or whether the two other Drosophila Rab5-GEFs, RABEX-5 (Yan et al., 2010) and GAPEX-5, can also affect the vap mutant phenotype through regulating Rab5-mediated early stage endocytosis. This could be addressed by testing the genetic interaction of the other Rab5-GEFs with vap2. In addition, it is unclear whether suppression of the vap mutant phenotype is specific to early stage endocytosis (Rab5) or if disruption in fast- recycling (Rab4), slow-recycling (Rab11) or late endosomes (Rab7) could also affect the vap mutant phenotype. This could be addressed by testing the genetic interaction of the other endocytic components with vap2. In order to further determine the importance of endocytosis in controlling neurodegeneration in vap mutant flies, it would be interesting to test the genetic interaction between vap2 and Drosophila temperature sensitive mutant dynamin (shibire). The vap neurodegeneration phenotype suppression data is from one spri mutant (spri6G1) allele. Hence it is possible that the suppression data is not due to the spri mutation but due to a second uncharacterised mutation on the same chromosome. Therefore it would be important to further support the role of Sprint in vap mutant neurodegenerative phenotype, using Sprint RNAi and deficiency fly lines as well as UAS- Sprint lines, which should enhance the vap mutant neurodegeneration phenotype. This would further support the spri knockout data and allows temporal regulation of Sprint expression in order to determine the stage that Sprint affects neuronal survival in vap mutant flies. The temporal regulation of Sprint expression can be achieved using heat- shock (hs)-GAL4 driver. As discussed previously, one model to explain the results presented here is that active Rab5 directly causes neuronal cell death through its effectors such as VPS34. Therefore it would also be interesting to test the genetic interaction of vap2 with known Rab5 effectors such as VPS34 in order to understand which signalling pathway(s) downstream of Rab5 lead to the vap mutant phenotype, considering VPS34 is known to play a role in neuronal survival (Zhou et al., 2010) and autophagy (Juhasz et al., 2008) and Rab5 is known to play a role in autophagy (Dou et al., 2013).

As discussed previously, the dying neurons in the vap mutant flies show morphological features of autophagy (Botella et al., 2003). Therefore in order to determine the effects of

168 autophagic signalling molecules on the vap mutant neurodegenerative phenotype, it would be interesting to mutate or overexpress different molecules involved in autophagy, such as TOR, in order to see the molecular mechanisms, which RasGAP feeds into in order to activate autophagy in the vap mutant flies.

Both proposed models related to the molecular mechanism behind the vap mutant neurodegeneration suggests that this phenotype is the result of disruption/over-activation of endocytosis. This is likely to affect signalling molecules such as EGFR, which leads to heightened MAPK signalling and eventually to autophagic neurodegeneration. Hence it would be interesting to determine the role of endocytosis on EGFR signalling. This can be addressed by expressing dominant negative (DN) and constitutively active (CA) Rab5 in adult Drosophila using hs-GAL4 driver and assess the EGFR activation of the MAPK (dpERK) signalling pathway. This experiment would also allow assessing the genetic interaction of Rab5-CA and Rab5-DN molecules with the vap mutant neurodegenerative phenotypes, which could lead to its enhancement or suppression. Since heterozygous mutant Rab5 (Rab54/+) supressed the vap mutant neurodegenerative phenotype, it is expected that Rab5-DN would phenocopy vap2; Rab54/+ and Rab5-CA would enhance the vap mutant neurodegenerative phenotype. It would also be desirable to test the effects of vap and Sprint double mutant flies (vap2, spri6G1) on the EGFR and Ras activation of the MAPK signalling pathway and see whether the vap mutant neurodegenerative phenotype would become enhanced or stay supressed when EGFR-Ras is overexpressed.

One of the disadvantages of sectioning is the potential to introduce artificial tears in a tissue, especially in mutant flies, since mutations can cause previously unknown structural weaknesses in a tissue or cell membrane. Hence it would be interesting to replicate the above genetic interaction results using dissected fly brains expressing GFP. Therefore any vacuolisations in the fly brain caused by neurodegeneration would appear as dark spots when the fly brain is z-stack imaged using a confocal microscope. This technique would not only provide a way to circumvent artificial tearing of the tissue but would also provide a more in depth spatial image of the extent of the neuronal tissue affected by neurodegeneration in the whole fly brain.

169

CHAPTER 8: Discussion and conclusions

8.1 General discussion This study has identified several novel RasGAP associating proteins in Drosophila and has provided evidence in support of a novel mechanism whereby RasGAP promotes neuronal survival by regulating endocytosis. In this chapter I will place these results in a wider perspective and discuss their implications both for the function of RasGAP in Drosophila and mammals, and for the role of endocytosis in neuronal survival.

8.1.1 The framework and implications of this study This study set out to uncover the role of RasGAP (vap gene) in neuronal survival in Drosophila. This was based on a previous study showing that unregulated EGFR signalling through lack of Ras deactivation by RasGAP leads to neuronal cell death (Botella et al., 2003). This finding is not unique to Drosophila since in mammals there is evidence that knock-out of negative regulators of Ras, p120-RasGAP/RASA1 and brain-specific synaptic RasGAP (synGAP), cause extensive neuronal cell death (Henkemeyer et al., 1995; Knuesel et al., 2005). However, from these previous studies it was not clear which Ras effector pathway(s) were responsible for mediating neurodegeneration or whether RasGAP might be acting independently of Ras as suggested for mammalian p120- RasGAP. A breakthrough in addressing this question came about when Drosophila vap mutant neurodegenerative phenotype was found to be rescued in a RasGAP SH2- dependent manner (J. Botella, S. A. Woodcock, D. A. Hughes and S. Schenuwly, personal communication). This provided the possibility that RasGAP SH2 interacting partners play a role in neuronal survival in Drosophila. For this purpose, tagged wild-type (RasGAPWT) and double SH2 inactivated (RasGAPSH2*32*) RasGAP were overexpressed and purified from Drosophila S2 cells and their co-purifying proteins were identified using mass spectrometry. This technique revealed several RasGAP SH2-dependent interacting partners including the neuronal cell adhesion molecule Dscam (Matthews et al., 2007; Schmucker et al., 2000), and Sprint, a putative Rab5-GEF (Jekely et al., 2005). The interaction between RasGAP and Sprint was selected for further analysis since Sprint is involved in regulation of RTK signalling and associates with Ras, and RTK-Ras signalling is strongly implicated in causing the vap mutant neurodegenerative phenotype. This study went on to characterise the mechanism of Sprint and RasGAP association and provided biochemical, immunohistological and genetical support for a model in which RasGAP regulates the

170 activation of Rab5 through its association with Sprint, and thereby influences the rate of RTK endocytosis. It is interesting to note that none of the identified RasGAP SH2- dependent interacting proteins have yet been identified as partners of mammalian p120- RasGAP although they are known to be evolutionarily conserved in mammals. Hence it is possible that neuronal cell death in mammalian p120-RasGAP mutant model (Henkemeyer et al., 1995) could also be due to disruptions in Rab5-regulated endocytic pathways. This would be consistent with previous findings showing that mutations in mammalian Rab5- GEF, ALS2, lead to motor neuron disease (Hadano et al., 2001; Otomo et al., 2003; Yang et al., 2001) and that Huntington’s disease protein, Huntingtin, is part of a Rab5 effector complex (Pal et al., 2006). Together these findings show that neurons are particularly susceptible to disruption of Rab5-regulated endocytic pathways, which in Drosophila are regulated by a novel RasGAP-Sprint mediated mechanism.

8.2 Outstanding questions and future work Although this study has provided an important first step in understanding the effects of RasGAP on RTK/Ras signalling through endocytosis, it leaves many detailed mechanistic questions unanswered, which merit further investigation including: (I) the direct or indirect nature of Sprint and RasGAP interaction, (II) whether Sprint and RasGAP are involved in RTK endocytosis and signalling and what is the significance of their association in these processes and (III) how Rab5-mediated endocytosis and RTK/Ras signalling influences neuronal survival.

8.2.1 What is the nature of Sprint and RasGAP interaction? This study shows that RasGAP interacts with Sprint, and this interaction depends on RasGAP SH2 domains being intact and on tyrosine phosphorylation of Sprint. Additionally, this study has narrowed down the region of Sprint responsible for mediating Sprint and RasGAP association to 481 amino acid residues between the proline-rich domain and the RIN-homology domain. As Sprint with an inactivated SH2 domain still interacts with RasGAP it is unlikely that the interaction is mediated through a bridging effect by RTKs such as PVR, known to be expressed and activated in S2 cells. Although the results in this study are consistent with direct binding between RasGAP and Sprint – mediated by the SH2 domains of RasGAP binding to one or more phosphotyrosine residues on Sprint – they do not exclude the possibility that the interaction is indirect.

171

It is often difficult to show whether an interaction between two proteins is direct. The main biochemical technique employed to address this question is to express the two proteins of interest in E. coli separately, which would ensure the absence of any adaptor proteins. The two proteins are then mixed together and direct binding is assayed by biophysical techniques such as surface plasmon resonance or by biochemical techniques such as pull downs. This is often a successful method in addressing the nature of protein interaction between two candidate proteins that do not require post-translational modifications in order to interact. However, this approach would not work if direct interaction is through the SH2 domains of RasGAP binding to phosphotyrosine residues on Sprint as E. coli expressed Sprint will not be tyrosine-phosphorylated. Therefore it would be necessary to purify Sprint to homogeneity from S2 cells, where it is tyrosine-phosphorylated, and then assay its binding to RasGAP GST-SH2 proteins expressed and purified from E. coli.

8.2.2 Are Sprint and RasGAP involved in RTK endocytosis and signalling and what is the significance of their association in these processes? There is strong evidence that Sprint is involved in RTK signalling and endocytosis. Sprint regulates RTK localisation and activation in Drosophila border cells, and this is though to be through its control of RTK endocytosis (Jekely et al., 2005). The potential role of Sprint in RTK endocytosis and signalling is consistent with the previous literature showing that members of the Drosophila ESCRT-0 endocytic complex (Hrs and Stam) and the endosomal protein Myopic regulate RTK signalling and endocytosis (Chanut-Delalande et al., 2010; Lloyd et al., 2002; Miura et al., 2008). This study is the first to implicate RasGAP in the regulation of RTK endocytosis but direct evidence for its involvement is lacking. To obtain conclusive evidence for the involvement of RasGAP in RTK endocytosis will require manipulation of RasGAP activity and monitoring RTK endocytosis either in vitro or in vivo. In this study an attempt was made to develop an S2 cell system to investigate ligand activated RTK endocytosis by using human-fly chimeric RTKs but this was unsuccessful. S2 cells stably expressing the EGFR have been used to follow endocytosis of fluorescently labelled EGFR ligand (Spitz) (Miura et al., 2008), and such a system could be used to investigate the role and RasGAP and Sprint in EGFR endocytosis. As Sprint is known to regulate RTK endocytosis in border cells, this might be a suitable model system to investigate the role of RasGAP in vivo. In light of the RasGAP mutant phenotype it would, however, be more relevant to investigate EGFR endocytosis in the adult brain and new methodologies would need to be developed to achieve this.

172

One important issue to consider is whether the other Rab5-GEFs in Drosophila, RABEX-5 and GAPEX-5, might functionally overlap with Sprint in regulating RTK endocytosis, potentially compensating for changes in Sprint activity. If this is the case then there may only be subtle changes in the endocytosis of RTKs when Sprint is overexpressed or mutated. This is consistent with the observation that Sprint mutant (spri6G1) flies are viable and have no obvious defects in their RTK signalling (Jekely et al., 2005).

8.2.3 How does Rab5-mediated endocytosis regulate neuronal survival? Given that the vap mutant neurodegeneration phenotype has been shown to be modified (suppressed or enhanced) by altering activity of the EGFR-Ras pathway (Botella et al., 2003), how increased Rab5 activity, within the context of EGFR-Ras signalling, could lead to neuronal cell death is still unclear. As discussed in chapter 7, one model to explain the results presented here is that Rab5 could function by regulating the activity of the EGFR- Ras pathway: increased Rab5 activity in the vap mutant would augment EGFR-Ras signalling and this would lead to neuronal cell death (Figure 8.1A). It is well established in both mammals (Barbieri et al., 2004; Barbieri et al., 2000; Chen et al., 2009) and Drosophila (Chanut-Delalande et al., 2010; Lloyd et al., 2002; Miura et al., 2008) that the EGFR is regulated by endocytosis. Ligand stimulated endocytosis of the EGFR receptor in mammalian tissue culture cells is the classic example of receptor-mediated endocytosis; the internalised ligand-receptor complex passes through endocytic compartments to the lysosome where it is degraded, thereby limiting EGFR signalling. In this classical mammalian in vitro system endocytosis inhibits EGFR signalling (Chen et al., 2009), however, in vivo studies in Drosophila show that endocytosis can stimulate EGFR signalling (Chanut-Delalande et al., 2010; Miura et al., 2008). Hence in order to determine the role of Rab5-mediated endocytosis in EGFR-Ras signalling, the MAPK signalling pathway activation (dpERK) could be assessed in different Rab5 heterozygous mutant fly lines since strong homozygous mutant Rab5 alleles are lethal (Wucherpfennig et al., 2003). Similar experiments in mammalian systems have shown that MAPK activity of EGFR is reduced by the expression of dominant-negative Rab5, which indicates endocytosis is required for MAPK signalling (Barbieri et al., 2004). As discussed above, Rab5 mediated endocytosis of RTKs such as EGFR could upregulate or downregulate the signalling output measured, depending on the nature of the signal (MAPK, JNK, etc.). An alternative model to explain the results presented here is that active Rab5 directly causes neuronal cell death and that the role of the EGFR-Ras pathway is to stimulate Sprint activity. In this model

173

Ras-GTP associated Sprint promotes active Rab5-GTP production, leading to an enhanced association of Rab5 to its effectors. The Rab5 effector class III PI3K (VPS34), generates

PIP3 leading to autophagosome vesicle nucleation (Simonsen and Tooze, 2009) and causes neurodegeneration. This is likely achieved through VPS34 association with a multi-protein complex containing the pro-autophagic tumour suppressor ATG6 and VPS15, which has been shown to occur in Drosophila (Juhasz et al., 2008) and mammalian systems (Zhou et al., 2010). In this model loss of RasGAP function leads to Rab5 activation through Sprint, leading to VPS34-VPS15-ATG6 activation of autophagy and neurodegeneration (Chang et al., 2009) (Figure 8.1B). Consistent with this model, mutations in the vap gene have been shown to upregulate VPS34 activity and increase autophagosomes formation and autophagy induction in fat bodies of starved Drosophila (pers. comm. with Marc Bourouis, Pierre Léopold lab). Two prediction of this model are that reducing VPS34 activity should suppress the vap mutant neurodegeneration phenotype and that excessive activation of EGFR-Ras signalling pathway should not enhance a null vap mutant phenotype. A recent study has shown that the loss of VPS34 in mice causes rapid sensory neurodegeneration, due to perturbation of endocytosis and not autophagy (Zhou et al., 2010). Hence perturbing VPS34 activity through loss of RasGAP could cause neuronal cell death either by affecting endocytosis or affecting autophagy or a combination of both.

8.3 Concluding remarks These results from this study indicate that the long-term survival of adult neurons in Drosophila depends on a critical balance between Ras activation and endocytosis, and that this balance is maintained by the interplay between RasGAP and Sprint. How dysregulation of this regulatory module leads to neurodegeneration remains to be determined but this study has opened up surprising new avenues for further investigation of the link between Ras and neuronal cell survival.

174

175

Figure 4.9 Possible roles of Rab5 in neuronal cell death in RasGAP mutant Drosophila. Upon growth factor binding to RTKs, the SOS Ras-GEF is recruited to the receptor, in association with the adaptor proteins CSW and Drk, where it activates Ras. In the absence of RasGAP in the early endosomes, Sprint associates with activated RTKs and its Rab5-GEF activity is stimulated by Ras-GTP, leading to excessive Rab5-GTP production. Excessive Rab5-GTP can either (A) augment EGFR-Ras signalling or (B) lead to VPS34-VPS15-ATG6 initiation of autophagy. ATG6, autophagy-related 6; GF, growth factor; RA, Ras association; SH2, Src homology 2; VPS9, vacuolar protein sorting 9p-like; YP, phosphotyrosine.

176

REFERENCES

Abdellatif, M., and Schneider, M.D. (1997). An effector-like function of Ras GTPase- activating protein predominates in cardiac muscle cells. J Biol Chem 272, 525-533.

Adari, H., Lowy, D.R., Willumsen, B.M., Der, C.J., and McCormick, F. (1988). Guanosine triphosphatase activating protein (GAP) interacts with the p21 ras effector binding domain. Science 240, 518-521.

Afar, D.E., Han, L., McLaughlin, J., Wong, S., Dhaka, A., Parmar, K., Rosenberg, N., Witte, O.N., and Colicelli, J. (1997). Regulation of the oncogenic activity of BCR-ABL by a tightly bound substrate protein RIN1. Immunity 6, 773-782.

Ahmadian, M.R., Stege, P., Scheffzek, K., and Wittinghofer, A. (1997). Confirmation of the arginine-finger hypothesis for the GAP-stimulated GTP-hydrolysis reaction of Ras. Nat Struct Biol 4, 686-689. al-Alawi, N., Xu, G., White, R., Clark, R., McCormick, F., and Feramisco, J.R. (1993). Differential regulation of cellular activities by GTPase-activating protein and NF1. Mol Cell Biol 13, 2497-2503.

Allen, M., Chu, S., Brill, S., Stotler, C., and Buckler, A. (1998). Restricted tissue expression pattern of a novel human rasGAP-related gene and its murine ortholog. Gene 218, 17-25.

Anand, S., Majeti, B.K., Acevedo, L.M., Murphy, E.A., Mukthavaram, R., Scheppke, L., Huang, M., Shields, D.J., Lindquist, J.N., Lapinski, P.E., et al. (2010). MicroRNA-132- mediated loss of p120RasGAP activates the endothelium to facilitate pathological angiogenesis. Nat Med 16, 909-914.

Anderson, D., Koch, C.A., Grey, L., Ellis, C., Moran, M.F., and Pawson, T. (1990). Binding of SH2 domains of phospholipase C gamma 1, GAP, and Src to activated growth factor receptors. Science 250, 979-982.

Ashburner, M. (1989). Drosophila: A Laboratory Handbook (Cold Spring Harbor Laboratory Press).

Balaji, K., Mooser, C., Janson, C.M., Bliss, J.M., Hojjat, H., and Colicelli, J. (2012). RIN1 orchestrates the activation of RAB5 GTPases and ABL tyrosine kinases to determine the fate of EGFR. J Cell Sci 125, 5887-5896.

Barbieri, M.A., Fernandez-Pol, S., Hunker, C., Horazdovsky, B.H., and Stahl, P.D. (2004). Role of rab5 in EGF receptor-mediated signal transduction. European journal of cell biology 83, 305-314.

Barbieri, M.A., Kong, C., Chen, P.I., Horazdovsky, B.F., and Stahl, P.D. (2003). The SRC homology 2 domain of Rin1 mediates its binding to the epidermal growth factor receptor and regulates receptor endocytosis. J Biol Chem 278, 32027-32036.

177

Barbieri, M.A., Roberts, R.L., Gumusboga, A., Highfield, H., Alvarez-Dominguez, C., Wells, A., and Stahl, P.D. (2000). Epidermal growth factor and membrane trafficking. EGF receptor activation of endocytosis requires Rab5a. J Cell Biol 151, 539-550.

Barsnes, H., Vizcaino, J.A., Eidhammer, I., and Martens, L. (2009). PRIDE Converter: making proteomics data-sharing easy. Nat Biotechnol 27, 598-599.

Barsnes, H., Vizcaino, J.A., Reisinger, F., Eidhammer, I., and Martens, L. (2011). Submitting proteomics data to PRIDE using PRIDE Converter. Methods Mol Biol 694, 237-253.

Basile, J.R., Afkhami, T., and Gutkind, J.S. (2005). Semaphorin 4D/plexin-B1 induces endothelial cell migration through the activation of PYK2, Src, and the phosphatidylinositol 3-kinase-Akt pathway. Mol Cell Biol 25, 6889-6898.

Bauer, A., and Kuster, B. (2003). Affinity purification-mass spectrometry. Powerful tools for the characterization of protein complexes. Eur J Biochem 270, 570-578.

Belov, A.A., and Mohammadi, M. (2012). Grb2, a double-edged sword of receptor tyrosine kinase signaling. Science signaling 5, pe49.

Bergmann, A., Agapite, J., McCall, K., and Steller, H. (1998). The Drosophila gene hid is a direct molecular target of Ras-dependent survival signaling. Cell 95, 331-341.

Bernards, A. (2003). GAPs galore! A survey of putative Ras superfamily GTPase activating proteins in man and Drosophila. Biochim Biophys Acta 1603, 47-82.

Blom, N., Gammeltoft, S., and Brunak, S. (1999). Sequence and structure-based prediction of eukaryotic protein phosphorylation sites. J Mol Biol 294, 1351-1362.

Bodenmiller, B., Malmstrom, J., Gerrits, B., Campbell, D., Lam, H., Schmidt, A., Rinner, O., Mueller, L.N., Shannon, P.T., Pedrioli, P.G., et al. (2007). PhosphoPep--a phosphoproteome resource for systems biology research in Drosophila Kc167 cells. Mol Syst Biol 3, 139.

Boon, L.M., Mulliken, J.B., and Vikkula, M. (2005). RASA1: variable phenotype with capillary and arteriovenous malformations. Curr Opin Genet Dev 15, 265-269.

Bos, J.L., Rehmann, H., and Wittinghofer, A. (2007). GEFs and GAPs: critical elements in the control of small G proteins. Cell 129, 865-877.

Botella, J.A., Kretzschmar, D., Kiermayer, C., Feldmann, P., Hughes, D.A., and Schneuwly, S. (2003). Deregulation of the Egfr/Ras signaling pathway induces age-related brain degeneration in the Drosophila mutant vap. Mol Biol Cell 14, 241-250.

Boutarbouch, M., Ben Salem, D., Gire, L., Giroud, M., Bejot, Y., and Ricolfi, F. (2010). Multiple cerebral and spinal cord cavernomas in Klippel-Trenaunay-Weber syndrome. J Clin Neurosci 17, 1073-1075.

Bradshaw, J.M., Mitaxov, V., and Waksman, G. (1999). Investigation of phosphotyrosine recognition by the SH2 domain of the Src kinase. J Mol Biol 293, 971-985.

178

Brand, A.H., and Perrimon, N. (1993). Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401-415.

Briggs, M.W., and Sacks, D.B. (2003). IQGAP proteins are integral components of cytoskeletal regulation. EMBO Rep 4, 571-574.

Briggs, S.D., Bryant, S.S., Jove, R., Sanderson, S.D., and Smithgall, T.E. (1995). The Ras GTPase-activating protein (GAP) is an SH3 domain-binding protein and substrate for the Src-related tyrosine kinase, Hck. J Biol Chem 270, 14718-14724.

Brill, S., Li, S., Lyman, C.W., Church, D.M., Wasmuth, J.J., Weissbach, L., Bernards, A., and Snijders, A.J. (1996). The Ras GTPase-activating-protein-related human protein IQGAP2 harbors a potential actin binding domain and interacts with calmodulin and Rho family GTPases. Mol Cell Biol 16, 4869-4878.

Britton, J.S., Lockwood, W.K., Li, L., Cohen, S.M., and Edgar, B.A. (2002). Drosophila's insulin/PI3-kinase pathway coordinates cellular metabolism with nutritional conditions. Dev Cell 2, 239-249.

Brown, S. (2010). Institutional Profile: The Sheffield RNAi screening facility: a service for high-throughput, genome-wide Drosophila RNAi screens. Future medicinal chemistry 2, 1805-1812.

Brunner, E., Ahrens, C.H., Mohanty, S., Baetschmann, H., Loevenich, S., Potthast, F., Deutsch, E.W., Panse, C., de Lichtenberg, U., Rinner, O., et al. (2007). A high-quality catalog of the Drosophila melanogaster proteome. Nat Biotechnol 25, 576-583.

Bryant, D.M., and Stow, J.L. (2005). Nuclear translocation of cell-surface receptors: lessons from fibroblast growth factor. Traffic 6, 947-954.

Bryant, S.S., Briggs, S., Smithgall, T.E., Martin, G.A., McCormick, F., Chang, J.H., Parsons, S.J., and Jove, R. (1995). Two SH2 domains of p120 Ras GTPase-activating protein bind synergistically to tyrosine phosphorylated p190 Rho GTPase-activating protein. J Biol Chem 270, 17947-17952.

Bryant, S.S., Mitchell, A.L., Collins, F., Miao, W., Marshall, M., and Jove, R. (1996). N- terminal sequences contained in the Src homology 2 and 3 domains of p120 GTPase- activating protein are required for full catalytic activity toward Ras. J Biol Chem 271, 5195-5199.

Cabrita, M.A., and Christofori, G. (2008). Sprouty proteins, masterminds of receptor tyrosine kinase signaling. Angiogenesis 11, 53-62.

Cacalano, N.A., Sanden, D., and Johnston, J.A. (2001). Tyrosine-phosphorylated SOCS-3 inhibits STAT activation but binds to p120 RasGAP and activates Ras. Nat Cell Biol 3, 460-465.

Cailliau, K., Browaeys-Poly, E., and Vilain, J.P. (2001). RasGAP is involved in signal transduction triggered by FGF1 in Xenopus oocytes expressing FGFR1. FEBS Lett 496, 161-165.

179

Cales, C., Hancock, J.F., Marshall, C.J., and Hall, A. (1988). The cytoplasmic protein GAP is implicated as the target for regulation by the ras gene product. Nature 332, 548-551.

Calvisi, D.F., Ladu, S., Conner, E.A., Seo, D., Hsieh, J.T., Factor, V.M., and Thorgeirsson, S.S. (2011). Inactivation of Ras GTPase-activating proteins promotes unrestrained activity of wild-type Ras in human liver cancer. J Hepatol 54, 311-319.

Cantley, L.C., Auger, K.R., Carpenter, C., Duckworth, B., Graziani, A., Kapeller, R., and Soltoff, S. (1991). Oncogenes and signal transduction. Cell 64, 281-302.

Carpino, N., Wisniewski, D., Strife, A., Marshak, D., Kobayashi, R., Stillman, B., and Clarkson, B. (1997). p62(dok): a constitutively tyrosine-phosphorylated, GAP-associated protein in chronic myelogenous leukemia progenitor cells. Cell 88, 197-204.

Carter, R.E., and Sorkin, A. (1998). Endocytosis of functional epidermal growth factor receptor-green fluorescent protein chimera. J Biol Chem 273, 35000-35007.

Celotto, A.M., Frank, A.C., McGrath, S.W., Fergestad, T., Van Voorhies, W.A., Buttle, K.F., Mannella, C.A., and Palladino, M.J. (2006). Mitochondrial encephalomyopathy in Drosophila. J Neurosci 26, 810-820.

Chang, H.Y., Huang, H.C., Huang, T.C., Yang, P.C., Wang, Y.C., and Juan, H.F. (2012). Ectopic ATP synthase blockade suppresses lung adenocarcinoma growth by activating the unfolded protein response. Cancer Res 72, 4696-4706.

Chang, Y.C., Lin, S.Y., Liang, S.Y., Pan, K.T., Chou, C.C., Chen, C.H., Liao, C.L., Khoo, K.H., and Meng, T.C. (2008). Tyrosine phosphoproteomics and identification of substrates of protein tyrosine phosphatase dPTP61F in Drosophila S2 cells by mass spectrometry- based substrate trapping strategy. Journal of proteome research 7, 1055-1066.

Chang, Y.Y., Juhasz, G., Goraksha-Hicks, P., Arsham, A.M., Mallin, D.R., Muller, L.K., and Neufeld, T.P. (2009). Nutrient-dependent regulation of autophagy through the target of rapamycin pathway. Biochem Soc Trans 37, 232-236.

Chang, Y.Y., and Neufeld, T.P. (2009). An Atg1/Atg13 complex with multiple roles in TOR-mediated autophagy regulation. Mol Biol Cell 20, 2004-2014.

Chanut-Delalande, H., Jung, A.C., Baer, M.M., Lin, L., Payre, F., and Affolter, M. (2010). The Hrs/Stam complex acts as a positive and negative regulator of RTK signaling during Drosophila development. PLoS One 5, e10245.

Chatterjee, N., and Bohmann, D. (2012). A versatile PhiC31 based reporter system for measuring AP-1 and Nrf2 signaling in Drosophila and in tissue culture. PLoS One 7, e34063.

Cheeseman, I.M., and Desai, A. (2005). A combined approach for the localization and tandem affinity purification of protein complexes from metazoans. Sci STKE 2005, pl1.

Chen, P.I., Kong, C., Su, X., and Stahl, P.D. (2009). Rab5 isoforms differentially regulate the trafficking and degradation of epidermal growth factor receptors. J Biol Chem 284, 30328-30338.

180

Chen, S., Do, J.T., Zhang, Q., Yao, S., Yan, F., Peters, E.C., Scholer, H.R., Schultz, P.G., and Ding, S. (2006). Self-renewal of embryonic stem cells by a small molecule. Proc Natl Acad Sci U S A 103, 17266-17271.

Chi, S., Kitanaka, C., Noguchi, K., Mochizuki, T., Nagashima, Y., Shirouzu, M., Fujita, H., Yoshida, M., Chen, W., Asai, A., et al. (1999). Oncogenic Ras triggers cell suicide through the activation of a caspase-independent cell death program in human cancer cells. Oncogene 18, 2281-2290.

Chiara, F., Goumans, M.J., Forsberg, H., Ahgren, A., Rasola, A., Aspenstrom, P., Wernstedt, C., Hellberg, C., Heldin, C.H., and Heuchel, R. (2004). A gain of function mutation in the activation loop of platelet-derived growth factor beta-receptor deregulates its kinase activity. J Biol Chem 279, 42516-42527.

Cho, N.K., Keyes, L., Johnson, E., Heller, J., Ryner, L., Karim, F., and Krasnow, M.A. (2002). Developmental control of blood cell migration by the Drosophila VEGF pathway. Cell 108, 865-876.

Chow, A., and Gawler, D. (1999). Mapping the site of interaction between annexin VI and the p120GAP C2 domain. FEBS Lett 460, 166-172.

Clark, G.J., Quilliam, L.A., Hisaka, M.M., and Der, C.J. (1993). Differential antagonism of Ras biological activity by catalytic and Src homology domains of Ras GTPase activation protein. Proc Natl Acad Sci U S A 90, 4887-4891.

Clark, G.J., Westwick, J.K., and Der, C.J. (1997). p120 GAP modulates Ras activation of Jun kinases and transformation. J Biol Chem 272, 1677-1681.

Clarke, P.G. (1990). Developmental cell death: morphological diversity and multiple mechanisms. Anat Embryol (Berl) 181, 195-213.

Cleghon, V., Feldmann, P., Ghiglione, C., Copeland, T.D., Perrimon, N., Hughes, D.A., and Morrison, D.K. (1998). Opposing actions of CSW and RasGAP modulate the strength of Torso RTK signaling in the Drosophila terminal pathway. Mol Cell 2, 719-727.

Cline, M.S., Smoot, M., Cerami, E., Kuchinsky, A., Landys, N., Workman, C., Christmas, R., Avila-Campilo, I., Creech, M., Gross, B., et al. (2007). Integration of biological networks and gene expression data using Cytoscape. Nat Protoc 2, 2366-2382.

Colicelli, J. (2010). ABL tyrosine kinases: evolution of function, regulation, and specificity. Science signaling 3, re6.

Colicelli, J., Nicolette, C., Birchmeier, C., Rodgers, L., Riggs, M., and Wigler, M. (1991). Expression of three mammalian cDNAs that interfere with RAS function in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 88, 2913-2917.

Cullen, P.J., Hsuan, J.J., Truong, O., Letcher, A.J., Jackson, T.R., Dawson, A.P., and Irvine, R.F. (1995). Identification of a specific Ins(1,3,4,5)P4-binding protein as a member of the GAP1 family. Nature 376, 527-530.

Dail, M., Richter, M., Godement, P., and Pasquale, E.B. (2006). Eph receptors inactivate R-Ras through different mechanisms to achieve cell repulsion. J Cell Sci 119, 1244-1254.

181

Dance, M., Montagner, A., Salles, J.P., Yart, A., and Raynal, P. (2008). The molecular functions of Shp2 in the Ras/Mitogen-activated protein kinase (ERK1/2) pathway. Cell Signal 20, 453-459.

Davis, A.J., Butt, J.T., Walker, J.H., Moss, S.E., and Gawler, D.J. (1996). The Ca2+- dependent lipid binding domain of P120GAP mediates protein-protein interactions with Ca2+-dependent membrane-binding proteins. Evidence for a direct interaction between annexin VI and P120GAP. J Biol Chem 271, 24333-24336. de Belle, J.S., and Heisenberg, M. (1996). Expression of Drosophila mushroom body mutations in alternative genetic backgrounds: a case study of the mushroom body miniature gene (mbm). Proc Natl Acad Sci U S A 93, 9875-9880. de Wijn, R.S., Oduber, C.E., Breugem, C.C., Alders, M., Hennekam, R.C., and van der Horst, C.M. (2012). Phenotypic variability in a family with capillary malformations caused by a mutation in the RASA1 gene. Eur J Med Genet 55, 191-195.

Deininger, K., Eder, M., Kramer, E.R., Zieglgansberger, W., Dodt, H.U., Dornmair, K., Colicelli, J., and Klein, R. (2008). The Rab5 guanylate exchange factor Rin1 regulates endocytosis of the EphA4 receptor in mature excitatory neurons. Proc Natl Acad Sci U S A 105, 12539-12544.

Delprato, A., and Lambright, D.G. (2007). Structural basis for Rab GTPase activation by VPS9 domain exchange factors. Nature structural & molecular biology 14, 406-412.

Delprato, A., Merithew, E., and Lambright, D.G. (2004). Structure, exchange determinants, and family-wide rab specificity of the tandem helical bundle and Vps9 domains of Rabex- 5. Cell 118, 607-617.

Ding, B., and Lengyel, P. (2008). p204 protein is a novel modulator of ras activity. J Biol Chem 283, 5831-5848.

Dinneen, J.L., and Ceresa, B.P. (2004). Expression of dominant negative rab5 in HeLa cells regulates endocytic trafficking distal from the plasma membrane. Exp Cell Res 294, 509-522.

Disanza, A., Frittoli, E., Palamidessi, A., and Scita, G. (2009). Endocytosis and spatial restriction of cell signaling. Molecular oncology 3, 280-296.

Donovan, S., Shannon, K.M., and Bollag, G. (2002). GTPase activating proteins: critical regulators of intracellular signaling. Biochim Biophys Acta 1602, 23-45.

Dou, Z., Pan, J.A., Dbouk, H.A., Ballou, L.M., DeLeon, J.L., Fan, Y., Chen, J.S., Liang, Z., Li, G., Backer, J.M., et al. (2013). Class IA PI3K p110beta subunit promotes autophagy through Rab5 small GTPase in response to growth factor limitation. Mol Cell 50, 29-42.

Drugan, J.K., Rogers-Graham, K., Gilmer, T., Campbell, S., and Clark, G.J. (2000). The Ras/p120 GTPase-activating protein (GAP) interaction is regulated by the p120 GAP pleckstrin homology domain. J Biol Chem 275, 35021-35027.

Durrington, H.J., Firth, H.V., Patient, C., Belham, M., Jayne, D., Burrows, N., Morrell, N.W., and Chilvers, E.R. (2013). A novel RASA1 mutation causing capillary

182 malformation-arteriovenous malformation (CM-AVM) presenting during pregnancy. American journal of medical genetics Part A 161, 1690-1694.

Eck, M.J., Shoelson, S.E., and Harrison, S.C. (1993). Recognition of a high-affinity phosphotyrosyl peptide by the Src homology-2 domain of p56lck. Nature 362, 87-91.

Edwards, K. (2007). Complete tyrosine kinase and tyrosine phosphatase gene lists for Drosophila melanogaster.

Eerola, I., Boon, L.M., Mulliken, J.B., Burrows, P.E., Dompmartin, A., Watanabe, S., Vanwijck, R., and Vikkula, M. (2003). Capillary malformation-arteriovenous malformation, a new clinical and genetic disorder caused by RASA1 mutations. Am J Hum Genet 73, 1240-1249.

Elgendy, M., Sheridan, C., Brumatti, G., and Martin, S.J. (2011). Oncogenic Ras-induced expression of Noxa and Beclin-1 promotes autophagic cell death and limits clonogenic survival. Mol Cell 42, 23-35.

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

Endo, M., and Yamashita, T. (2009). Inactivation of Ras by p120GAP via focal adhesion kinase dephosphorylation mediates RGMa-induced growth cone collapse. J Neurosci 29, 6649-6662.

Esters, H., Alexandrov, K., Iakovenko, A., Ivanova, T., Thoma, N., Rybin, V., Zerial, M., Scheidig, A.J., and Goody, R.S. (2001). Vps9, Rabex-5 and DSS4: proteins with weak but distinct nucleotide-exchange activities for Rab proteins. J Mol Biol 310, 141-156.

Evans, J.V., Ammer, A.G., Jett, J.E., Bolcato, C.A., Breaux, J.C., Martin, K.H., Culp, M.V., Gannett, P.M., and Weed, S.A. (2012). Src binds cortactin through an SH2 domain cystine-mediated linkage. J Cell Sci 125, 6185-6197.

Fantl, W.J., Escobedo, J.A., Martin, G.A., Turck, C.W., del Rosario, M., McCormick, F., and Williams, L.T. (1992). Distinct phosphotyrosines on a growth factor receptor bind to specific molecules that mediate different signaling pathways. Cell 69, 413-423.

Felder, S., Zhou, M., Hu, P., Urena, J., Ullrich, A., Chaudhuri, M., White, M., Shoelson, S.E., and Schlessinger, J. (1993). SH2 domains exhibit high-affinity binding to tyrosine- phosphorylated peptides yet also exhibit rapid dissociation and exchange. Mol Cell Biol 13, 1449-1455.

Feldmann, P., Eicher, E.N., Leevers, S.J., Hafen, E., and Hughes, D.A. (1999). Control of growth and differentiation by Drosophila RasGAP, a homolog of p120 Ras-GTPase- activating protein. Mol Cell Biol 19, 1928-1937.

Fergestad, T., Olson, L., Patel, K.P., Miller, R., Palladino, M.J., and Ganetzky, B. (2008). Neuropathology in Drosophila mutants with increased seizure susceptibility. Genetics 178, 947-956.

183

Ferguson, S.M., and De Camilli, P. (2012). Dynamin, a membrane-remodelling GTPase. Nat Rev Mol Cell Biol 13, 75-88.

Forsthoefel, D.J., Liebl, E.C., Kolodziej, P.A., and Seeger, M.A. (2005). The Abelson tyrosine kinase, the Trio GEF and Enabled interact with the Netrin receptor Frazzled in Drosophila. Development 132, 1983-1994.

Frackelton, A.R., Jr., Kumar, P.S., Kannan, B., and Clark, J.W. (1993). Tyrosine phosphorylated proteins in chronic myelogenous leukemia. Leuk Lymphoma 11 Suppl 1, 125-129.

Frech, M., John, J., Pizon, V., Chardin, P., Tavitian, A., Clark, R., McCormick, F., and Wittinghofer, A. (1990). Inhibition of GTPase activating protein stimulation of Ras-p21 GTPase by the Krev-1 gene product. Science 249, 169-171.

Freeman, M. (2008). Rhomboid proteases and their biological functions. Annual review of genetics 42, 191-210.

Friedman, A., and Perrimon, N. (2006). A functional RNAi screen for regulators of receptor tyrosine kinase and ERK signalling. Nature 444, 230-234.

Frigerio, A., Stevenson, D.A., and Grimmer, J.F. (2012). The genetics of vascular anomalies. Current opinion in otolaryngology & head and neck surgery 20, 527-532.

Galvis, A., Giambini, H., Villasana, Z., and Barbieri, M.A. (2009). Functional determinants of ras interference 1 mutants required for their inhbitory activity on endocytosis. Exp Cell Res 315, 820-835.

Gaul, U., Mardon, G., and Rubin, G.M. (1992). A putative Ras GTPase activating protein acts as a negative regulator of signaling by the Sevenless receptor tyrosine kinase. Cell 68, 1007-1019.

Gawler, D.J., Zhang, L.J., and Moran, M.F. (1995). Mutation-deletion analysis of a Ca(2+)-dependent phospholipid binding (CaLB) domain within p120 GAP, a GTPase- activating protein for p21 ras. Biochem J 307 ( Pt 2), 487-491.

Ger, M., Zitkus, Z., and Valius, M. (2011). Adaptor protein Nck1 interacts with p120 Ras GTPase-activating protein and regulates its activity. Cellular signalling 23, 1651-1658.

Ghiglione, C., Amundadottir, L., Andresdottir, M., Bilder, D., Diamonti, J.A., Noselli, S., Perrimon, N., and Carraway, I.K. (2003). Mechanism of inhibition of the Drosophila and mammalian EGF receptors by the transmembrane protein Kekkon 1. Development 130, 4483-4493.

Gibbs, J.B., Schaber, M.D., Allard, W.J., Sigal, I.S., and Scolnick, E.M. (1988). Purification of ras GTPase activating protein from bovine brain. Proc Natl Acad Sci U S A 85, 5026-5030.

Gideon, P., John, J., Frech, M., Lautwein, A., Clark, R., Scheffler, J.E., and Wittinghofer, A. (1992). Mutational and kinetic analyses of the GTPase-activating protein (GAP)-p21 interaction: the C-terminal domain of GAP is not sufficient for full activity. Mol Cell Biol 12, 2050-2056.

184

Giglione, C., Gonfloni, S., and Parmeggiani, A. (2001). Differential actions of p60c-Src and Lck kinases on the Ras regulators p120-GAP and GDP/GTP exchange factor CDC25Mm. Eur J Biochem 268, 3275-3283.

Giglione, C., Parrini, M.C., Baouz, S., Bernardi, A., and Parmeggiani, A. (1997). A new function of p120-GTPase-activating protein. Prevention of the guanine nucleotide exchange factor-stimulated nucleotide exchange on the active form of Ha-ras p21. J Biol Chem 272, 25128-25134.

Gigoux, V., L'Hoste, S., Raynaud, F., Camonis, J., and Garbay, C. (2002). Identification of Aurora kinases as RasGAP Src homology 3 domain-binding proteins. J Biol Chem 277, 23742-23746.

Golubic, M., Harwalkar, J.A., Bryant, S.S., Sundaram, V., Jove, R., and Lee, J.H. (1998). Differential regulation of neurofibromin and p120 GTPase-activating protein by nutritionally relevant fatty acids. Nutr Cancer 30, 97-107.

Graham, S.M., Vojtek, A.B., Huff, S.Y., Cox, A.D., Clark, G.J., Cooper, J.A., and Der, C.J. (1996). TC21 causes transformation by Raf-independent signaling pathways. Mol Cell Biol 16, 6132-6140.

Gruenberg, J., and Stenmark, H. (2004). The biogenesis of multivesicular endosomes. Nat Rev Mol Cell Biol 5, 317-323.

Guitard, E., Barlat, I., Maurier, F., Schweighoffer, F., and Tocque, B. (1998). Sam68 is a Ras-GAP-associated protein in mitosis. Biochem Biophys Res Commun 245, 562-566.

Guo, H.F., The, I., Hannan, F., Bernards, A., and Zhong, Y. (1997). Requirement of Drosophila NF1 for activation of adenylyl cyclase by PACAP38-like neuropeptides. Science 276, 795-798.

Guo, H.F., Tong, J., Hannan, F., Luo, L., and Zhong, Y. (2000). A neurofibromatosis-1- regulated pathway is required for learning in Drosophila. Nature 403, 895-898.

Guo, X., Hamilton, P.J., Reish, N.J., Sweatt, J.D., Miller, C.A., and Rumbaugh, G. (2009). Reduced expression of the NMDA receptor-interacting protein SynGAP causes behavioral abnormalities that model symptoms of Schizophrenia. Neuropsychopharmacology 34, 1659-1672.

Hadano, S., Hand, C.K., Osuga, H., Yanagisawa, Y., Otomo, A., Devon, R.S., Miyamoto, N., Showguchi-Miyata, J., Okada, Y., Singaraja, R., et al. (2001). A gene encoding a putative GTPase regulator is mutated in familial amyotrophic lateral sclerosis 2. Nat Genet 29, 166-173.

Hadano, S., Otomo, A., Suzuki-Utsunomiya, K., Kunita, R., Yanagisawa, Y., Showguchi- Miyata, J., Mizumura, H., and Ikeda, J.E. (2004). ALS2CL, the novel protein highly homologous to the carboxy-terminal half of ALS2, binds to Rab5 and modulates endosome dynamics. FEBS Lett 575, 64-70.

Halenbeck, R., Crosier, W.J., Clark, R., McCormick, F., and Koths, K. (1990). Purification, characterization, and western blot analysis of human GTPase-activating protein from native and recombinant sources. J Biol Chem 265, 21922-21928.

185

Han, J.W., McCormick, F., and Macara, I.G. (1991). Regulation of Ras-GAP and the neurofibromatosis-1 gene product by eicosanoids. Science 252, 576-579.

Han, L., and Colicelli, J. (1995). A human protein selected for interference with Ras function interacts directly with Ras and competes with Raf1. Mol Cell Biol 15, 1318-1323.

Han, L., Wong, D., Dhaka, A., Afar, D., White, M., Xie, W., Herschman, H., Witte, O., and Colicelli, J. (1997). Protein binding and signaling properties of RIN1 suggest a unique effector function. Proc Natl Acad Sci U S A 94, 4954-4959.

Hara, T., Nakamura, K., Matsui, M., Yamamoto, A., Nakahara, Y., Suzuki-Migishima, R., Yokoyama, M., Mishima, K., Saito, I., Okano, H., et al. (2006). Suppression of basal autophagy in neural cells causes neurodegenerative disease in mice. Nature 441, 885-889.

Hart, M.J., Callow, M.G., Souza, B., and Polakis, P. (1996). IQGAP1, a calmodulin- binding protein with a rasGAP-related domain, is a potential effector for cdc42Hs. EMBO J 15, 2997-3005.

Heino, T.I., Karpanen, T., Wahlstrom, G., Pulkkinen, M., Eriksson, U., Alitalo, K., and Roos, C. (2001). The Drosophila VEGF receptor homolog is expressed in hemocytes. Mech Dev 109, 69-77.

Henkemeyer, M., Rossi, D.J., Holmyard, D.P., Puri, M.C., Mbamalu, G., Harpal, K., Shih, T.S., Jacks, T., and Pawson, T. (1995). Vascular system defects and neuronal apoptosis in mice lacking ras GTPase-activating protein. Nature 377, 695-701.

Hershkovitz, D., Bercovich, D., Sprecher, E., and Lapidot, M. (2008a). RASA1 mutations may cause hereditary capillary malformations without arteriovenous malformations. Br J Dermatol 158, 1035-1040.

Hershkovitz, D., Bergman, R., and Sprecher, E. (2008b). A novel mutation in RASA1 causes capillary malformation and limb enlargement. Arch Dermatol Res 300, 385-388.

Horiuchi, H., Lippe, R., McBride, H.M., Rubino, M., Woodman, P., Stenmark, H., Rybin, V., Wilm, M., Ashman, K., Mann, M., et al. (1997). A novel Rab5 GDP/GTP exchange factor complexed to Rabaptin-5 links nucleotide exchange to effector recruitment and function. Cell 90, 1149-1159.

Hoshino, M., Kawakita, M., and Hattori, S. (1988). Characterization of a factor that stimulates hydrolysis of GTP bound to ras gene product p21 (GTPase-activating protein) and correlation of its activity to cell density. Mol Cell Biol 8, 4169-4173.

Hu, H., Bliss, J.M., Wang, Y., and Colicelli, J. (2005). RIN1 is an ABL tyrosine kinase activator and a regulator of epithelial-cell adhesion and migration. Curr Biol 15, 815-823.

Hu, H., Milstein, M., Bliss, J.M., Thai, M., Malhotra, G., Huynh, L.C., and Colicelli, J. (2008). Integration of transforming growth factor beta and RAS signaling silences a RAB5 guanine nucleotide exchange factor and enhances growth factor-directed cell migration. Mol Cell Biol 28, 1573-1583.

186

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

Huang, D.C., Marshall, C.J., and Hancock, J.F. (1993). Plasma membrane-targeted ras GTPase-activating protein is a potent suppressor of p21ras function. Mol Cell Biol 13, 2420-2431.

Huang, H., Li, L., Wu, C., Schibli, D., Colwill, K., Ma, S., Li, C., Roy, P., Ho, K., Songyang, Z., et al. (2008). Defining the specificity space of the human SRC homology 2 domain. Mol Cell Proteomics 7, 768-784.

Hubbard, S.R. (2004). Juxtamembrane autoinhibition in receptor tyrosine kinases. Nat Rev Mol Cell Biol 5, 464-471.

Humphries, J.D., Byron, A., Bass, M.D., Craig, S.E., Pinney, J.W., Knight, D., and Humphries, M.J. (2009). Proteomic analysis of integrin-associated complexes identifies RCC2 as a dual regulator of Rac1 and Arf6. Science signaling 2, ra51.

Hunker, C.M., Galvis, A., Kruk, I., Giambini, H., Veisaga, M.L., and Barbieri, M.A. (2006a). Rab5-activating protein 6, a novel endosomal protein with a role in endocytosis. Biochem Biophys Res Commun 340, 967-975.

Hunker, C.M., Galvis, A., Veisaga, M.L., and Barbieri, M.A. (2006b). Rin1 is a negative regulator of the IL3 receptor signal transduction pathways. Anticancer research 26, 905- 916.

Hunker, C.M., Giambini, H., Galvis, A., Hall, J., Kruk, I., Veisaga, M.L., and Barbieri, M.A. (2006c). Rin1 regulates insulin receptor signal transduction pathways. Exp Cell Res 312, 1106-1118.

Hunter, T. (2000). Signaling--2000 and beyond. Cell 100, 113-127.

Inaki, M., Vishnu, S., Cliffe, A., and Rorth, P. (2012). Effective guidance of collective migration based on differences in cell states. Proc Natl Acad Sci U S A 109, 2027-2032.

Ishimaru, S., Ueda, R., Hinohara, Y., Ohtani, M., and Hanafusa, H. (2004). PVR plays a critical role via JNK activation in thorax closure during Drosophila metamorphosis. EMBO J 23, 3984-3994.

Ito, Y., Oinuma, I., Katoh, H., Kaibuchi, K., and Negishi, M. (2006). Sema4D/plexin-B1 activates GSK-3beta through R-Ras GAP activity, inducing growth cone collapse. EMBO Rep 7, 704-709.

Itoh, T., and Fukuda, M. (2006). Identification of EPI64 as a GTPase-activating protein specific for Rab27A. J Biol Chem 281, 31823-31831.

Iwashita, S., Kobayashi, M., Kubo, Y., Hinohara, Y., Sezaki, M., Nakamura, K., Suzuki- Migishima, R., Yokoyama, M., Sato, S., Fukuda, M., et al. (2007). Versatile roles of R-Ras GAP in neurite formation of PC12 cells and embryonic vascular development. J Biol Chem 282, 3413-3417.

187

Jabado, N., Jauliac, S., Pallier, A., Bernard, F., Fischer, A., and Hivroz, C. (1998). Sam68 association with p120GAP in CD4+ T cells is dependent on CD4 molecule expression. J Immunol 161, 2798-2803.

James, B.P., Bunch, T.A., Krishnamoorthy, S., Perkins, L.A., and Brower, D.L. (2007). Nuclear localization of the ERK MAP kinase mediated by Drosophila alphaPS2betaPS integrin and importin-7. Mol Biol Cell 18, 4190-4199.

Jean, S., and Kiger, A.A. (2012). Coordination between RAB GTPase and phosphoinositide regulation and functions. Nat Rev Mol Cell Biol 13, 463-470.

Jekely, G., Sung, H.H., Luque, C.M., and Rorth, P. (2005). Regulators of endocytosis maintain localized receptor tyrosine kinase signaling in guided migration. Dev Cell 9, 197- 207.

Jiang, S.Y., and Ramachandran, S. (2006). Comparative and evolutionary analysis of genes encoding small GTPases and their activating proteins in eukaryotic genomes. Physiol Genomics 24, 235-251.

Jiang, Y., Ma, W., Wan, Y., Kozasa, T., Hattori, S., and Huang, X.Y. (1998). The G protein G alpha12 stimulates Bruton's tyrosine kinase and a rasGAP through a conserved PH/BM domain. Nature 395, 808-813.

Jin, M.H., Sawamoto, K., Ito, M., and Okano, H. (2000). The interaction between the Drosophila secreted protein argos and the epidermal growth factor receptor inhibits dimerization of the receptor and binding of secreted spitz to the receptor. Mol Cell Biol 20, 2098-2107.

Jin, W., Reddy, M.A., Chen, Z., Putta, S., Lanting, L., Kato, M., Park, J.T., Chandra, M., Wang, C., Tangirala, R.K., et al. (2012). Small RNA Sequencing Reveals MicroRNAs That Modulate Angiotensin II Effects in Vascular Smooth Muscle Cells. The Journal of biological chemistry 287, 15672-15683.

Johannessen, C.M., Reczek, E.E., James, M.F., Brems, H., Legius, E., and Cichowski, K. (2005). The NF1 tumor suppressor critically regulates TSC2 and mTOR. Proc Natl Acad Sci U S A 102, 8573-8578.

Juhasz, G., Erdi, B., Sass, M., and Neufeld, T.P. (2007). Atg7-dependent autophagy promotes neuronal health, stress tolerance, and longevity but is dispensable for metamorphosis in Drosophila. Genes Dev 21, 3061-3066.

Juhasz, G., Hill, J.H., Yan, Y., Sass, M., Baehrecke, E.H., Backer, J.M., and Neufeld, T.P. (2008). The class III PI(3)K Vps34 promotes autophagy and endocytosis but not TOR signaling in Drosophila. J Cell Biol 181, 655-666.

Kajiho, H., Saito, K., Tsujita, K., Kontani, K., Araki, Y., Kurosu, H., and Katada, T. (2003). RIN3: a novel Rab5 GEF interacting with amphiphysin II involved in the early endocytic pathway. J Cell Sci 116, 4159-4168.

Kajiho, H., Sakurai, K., Minoda, T., Yoshikawa, M., Nakagawa, S., Fukushima, S., Kontani, K., and Katada, T. (2011). Characterization of RIN3 as a guanine nucleotide exchange factor for the Rab5 subfamily GTPase Rab31. J Biol Chem 286, 24364-24373.

188

Kalesnikoff, J., Rios, E.J., Chen, C.C., Alejandro Barbieri, M., Tsai, M., Tam, S.Y., and Galli, S.J. (2007). Roles of RabGEF1/Rabex-5 domains in regulating Fc epsilon RI surface expression and Fc epsilon RI-dependent responses in mast cells. Blood 109, 5308-5317.

Kamada, Y., Funakoshi, T., Shintani, T., Nagano, K., Ohsumi, M., and Ohsumi, Y. (2000). Tor-mediated induction of autophagy via an Apg1 protein kinase complex. J Cell Biol 150, 1507-1513.

Kamata, T., and Feramisco, J.R. (1984). Epidermal growth factor stimulates guanine nucleotide binding activity and phosphorylation of ras oncogene proteins. Nature 310, 147- 150.

Kaplan, D.R., Morrison, D.K., Wong, G., McCormick, F., and Williams, L.T. (1990). PDGF beta-receptor stimulates tyrosine phosphorylation of GAP and association of GAP with a signaling complex. Cell 61, 125-133.

Karnoub, A.E., and Weinberg, R.A. (2008). Ras oncogenes: split personalities. Nat Rev Mol Cell Biol 9, 517-531.

Kashige, N., Carpino, N., and Kobayashi, R. (2000). Tyrosine phosphorylation of p62dok by p210bcr-abl inhibits RasGAP activity. Proc Natl Acad Sci U S A 97, 2093-2098.

Kay, L.E., Muhandiram, D.R., Farrow, N.A., Aubin, Y., and Forman-Kay, J.D. (1996). Correlation between dynamics and high affinity binding in an SH2 domain interaction. Biochemistry 35, 361-368.

Kim, H.K., Jeong, M.J., Kong, M.Y., Han, M.Y., Son, K.H., Kim, H.M., Hong, S.H., and Kwon, B.M. (2005). Inhibition of Shc/Grb2 protein-protein interaction suppresses growth of B104-1-1 tumors xenografted in nude mice. Life Sci 78, 321-328.

Kim, J.H., Lee, H.K., Takamiya, K., and Huganir, R.L. (2003). The role of synaptic GTPase-activating protein in neuronal development and synaptic plasticity. J Neurosci 23, 1119-1124.

Kim, J.H., Liao, D., Lau, L.F., and Huganir, R.L. (1998). SynGAP: a synaptic RasGAP that associates with the PSD-95/SAP90 protein family. Neuron 20, 683-691.

Kimura, T., Sakisaka, T., Baba, T., Yamada, T., and Takai, Y. (2006). Involvement of the Ras-Ras-activated Rab5 guanine nucleotide exchange factor RIN2-Rab5 pathway in the hepatocyte growth factor-induced endocytosis of E-cadherin. J Biol Chem 281, 10598- 10609.

Kitano, M., Nakaya, M., Nakamura, T., Nagata, S., and Matsuda, M. (2008). Imaging of Rab5 activity identifies essential regulators for phagosome maturation. Nature 453, 241- 245.

Klueg, K.M., Alvarado, D., Muskavitch, M.A., and Duffy, J.B. (2002). Creation of a GAL4/UAS-coupled inducible gene expression system for use in Drosophila cultured cell lines. Genesis 34, 119-122.

189

Knuesel, I., Elliott, A., Chen, H.J., Mansuy, I.M., and Kennedy, M.B. (2005). A role for synGAP in regulating neuronal apoptosis. The European journal of neuroscience 21, 611- 621.

Koch, C.A., Anderson, D., Moran, M.F., Ellis, C., and Pawson, T. (1991). SH2 and SH3 domains: elements that control interactions of cytoplasmic signaling proteins. Science 252, 668-674.

Kocher, T., and Superti-Furga, G. (2007). Mass spectrometry-based functional proteomics: from molecular machines to protein networks. Nat Methods 4, 807-815.

Koehler, J.A., and Moran, M.F. (2001). RACK1, a protein kinase C scaffolding protein, interacts with the PH domain of p120GAP. Biochem Biophys Res Commun 283, 888-895.

Komatsu, M., Waguri, S., Chiba, T., Murata, S., Iwata, J., Tanida, I., Ueno, T., Koike, M., Uchiyama, Y., Kominami, E., et al. (2006). Loss of autophagy in the central nervous system causes neurodegeneration in mice. Nature 441, 880-884.

Kong, C., Su, X., Chen, P.I., and Stahl, P.D. (2007). Rin1 interacts with signal-transducing adaptor molecule (STAM) and mediates epidermal growth factor receptor trafficking and degradation. J Biol Chem 282, 15294-15301.

Krishnamoorthy, S. (2008). Receptor tyrosine kinase (RTK) mediated tyrosine phosphor- proteome from Drosophila S2 (ErbB1) cells reveals novel signaling networks. PLoS One 3, e2877.

Kulkarni, S.V., Gish, G., van der Geer, P., Henkemeyer, M., and Pawson, T. (2000). Role of p120 Ras-GAP in directed cell movement. J Cell Biol 149, 457-470.

Kunath, T., Gish, G., Lickert, H., Jones, N., Pawson, T., and Rossant, J. (2003). Transgenic RNA interference in ES cell-derived embryos recapitulates a genetic null phenotype. Nat Biotechnol 21, 559-561.

Kupzig, S., Bouyoucef, D., Cozier, G.E., and Cullen, P.J. (2006a). Studying the spatial and temporal regulation of Ras GTPase-activating proteins. Methods Enzymol 407, 64-82.

Kupzig, S., Bouyoucef-Cherchalli, D., Yarwood, S., Sessions, R., and Cullen, P.J. (2009). The ability of GAP1IP4BP to function as a Rap1 GTPase-activating protein (GAP) requires its Ras GAP-related domain and an arginine finger rather than an asparagine thumb. Mol Cell Biol 29, 3929-3940.

Kupzig, S., Deaconescu, D., Bouyoucef, D., Walker, S.A., Liu, Q., Polte, C.L., Daumke, O., Ishizaki, T., Lockyer, P.J., Wittinghofer, A., et al. (2006b). GAP1 family members constitute bifunctional Ras and Rap GTPase-activating proteins. J Biol Chem 281, 9891- 9900.

Kurada, P., and White, K. (1998). Ras promotes cell survival in Drosophila by downregulating hid expression. Cell 95, 319-329.

Lapinski, P.E., Bauler, T.J., Brown, E.J., Hughes, E.D., Saunders, T.L., and King, P.D. (2007). Generation of mice with a conditional allele of the p120 Ras GTPase-activating protein. Genesis 45, 762-767.

190

Lapinski, P.E., Kwon, S., Lubeck, B.A., Wilkinson, J.E., Srinivasan, R.S., Sevick-Muraca, E., and King, P.D. (2012). RASA1 maintains the lymphatic vasculature in a quiescent functional state in mice. J Clin Invest 122, 733-747.

Lappalainen, I., Thusberg, J., Shen, B., and Vihinen, M. (2008). Genome wide analysis of pathogenic SH2 domain mutations. Proteins 72, 779-792.

Le Roy, C., and Wrana, J.L. (2005). Clathrin- and non-clathrin-mediated endocytic regulation of cell signalling. Nat Rev Mol Cell Biol 6, 112-126.

Ledda, F., Bieraugel, O., Fard, S.S., Vilar, M., and Paratcha, G. (2008). Lrig1 is an endogenous inhibitor of Ret receptor tyrosine kinase activation, downstream signaling, and biological responses to GDNF. J Neurosci 28, 39-49.

Ledda, F., and Paratcha, G. (2007). Negative Regulation of Receptor Tyrosine Kinase (RTK) Signaling: A Developing Field. Biomarker insights 2, 45-58.

Lee, S., Tsai, Y.C., Mattera, R., Smith, W.J., Kostelansky, M.S., Weissman, A.M., Bonifacino, J.S., and Hurley, J.H. (2006). Structural basis for ubiquitin recognition and autoubiquitination by Rabex-5. Nature structural & molecular biology 13, 264-271.

Leevers, S.J., Weinkove, D., MacDougall, L.K., Hafen, E., and Waterfield, M.D. (1996). The Drosophila phosphoinositide 3-kinase Dp110 promotes cell growth. EMBO J 15, 6584-6594.

Lemeer, S., Bluwstein, A., Wu, Z., Leberfinger, J., Muller, K., Kramer, K., and Kuster, B. (2012). Phosphotyrosine mediated protein interactions of the discoidin domain receptor 1. J Proteomics 75, 3465-3477.

Lessing, D., and Bonini, N.M. (2009). Maintaining the brain: insight into human neurodegeneration from Drosophila melanogaster mutants. Nat Rev Genet 10, 359-370.

Levine, B., and Kroemer, G. (2008). Autophagy in the pathogenesis of disease. Cell 132, 27-42.

Liao, Y.C., Si, L., deVere White, R.W., and Lo, S.H. (2007). The phosphotyrosine- independent interaction of DLC-1 and the SH2 domain of cten regulates focal adhesion localization and growth suppression activity of DLC-1. J Cell Biol 176, 43-49.

Liu, B.A., Jablonowski, K., Raina, M., Arce, M., Pawson, T., and Nash, P.D. (2006). The human and mouse complement of SH2 domain proteins-establishing the boundaries of phosphotyrosine signaling. Mol Cell 22, 851-868.

Liu, K., and Li, G. (1998). Catalytic domain of the p120 Ras GAP binds to RAb5 and stimulates its GTPase activity. J Biol Chem 273, 10087-10090.

Liu, X.Q., and Pawson, T. (1991). The epidermal growth factor receptor phosphorylates GTPase-activating protein (GAP) at Tyr-460, adjacent to the GAP SH2 domains. Mol Cell Biol 11, 2511-2516.

191

Lloyd, T.E., Atkinson, R., Wu, M.N., Zhou, Y., Pennetta, G., and Bellen, H.J. (2002). Hrs regulates endosome membrane invagination and tyrosine kinase receptor signaling in Drosophila. Cell 108, 261-269.

Lodhi, I.J., Chiang, S.H., Chang, L., Vollenweider, D., Watson, R.T., Inoue, M., Pessin, J.E., and Saltiel, A.R. (2007). Gapex-5, a Rab31 guanine nucleotide exchange factor that regulates Glut4 trafficking in adipocytes. Cell metabolism 5, 59-72.

Logarinho, E., Bousbaa, H., Dias, J.M., Lopes, C., Amorim, I., Antunes-Martins, A., and Sunkel, C.E. (2004). Different spindle checkpoint proteins monitor microtubule attachment and tension at kinetochores in Drosophila cells. J Cell Sci 117, 1757-1771.

Lowy, D.R., and Willumsen, B.M. (1993). Function and regulation of ras. Annu Rev Biochem 62, 851-891.

Lu, B. (2009). Recent advances in using Drosophila to model neurodegenerative diseases. Apoptosis 14, 1008-1020.

Lu, B., and Vogel, H. (2009). Drosophila models of neurodegenerative diseases. Annu Rev Pathol 4, 315-342.

Lundgren, D.H., Hwang, S.I., Wu, L., and Han, D.K. (2010). Role of spectral counting in quantitative proteomics. Expert review of proteomics 7, 39-53.

Machida, K., Thompson, C.M., Dierck, K., Jablonowski, K., Karkkainen, S., Liu, B., Zhang, H., Nash, P.D., Newman, D.K., Nollau, P., et al. (2007). High-throughput phosphotyrosine profiling using SH2 domains. Mol Cell 26, 899-915.

Macia, E., Ehrlich, M., Massol, R., Boucrot, E., Brunner, C., and Kirchhausen, T. (2006). Dynasore, a cell-permeable inhibitor of dynamin. Dev Cell 10, 839-850.

Madonna, S., Scarponi, C., De Pita, O., and Albanesi, C. (2008). Suppressor of cytokine signaling 1 inhibits IFN-gamma inflammatory signaling in human keratinocytes by sustaining ERK1/2 activation. FASEB J 22, 3287-3297.

Maekawa, M., Li, S., Iwamatsu, A., Morishita, T., Yokota, K., Imai, Y., Kohsaka, S., Nakamura, S., and Hattori, S. (1994). A novel mammalian Ras GTPase-activating protein which has phospholipid-binding and Btk homology regions. Mol Cell Biol 14, 6879-6885.

Mahajan, N.P., and Earp, H.S. (2003). An SH2 domain-dependent, phosphotyrosine- independent interaction between Vav1 and the Mer receptor tyrosine kinase: a mechanism for localizing guanine nucleotide-exchange factor action. J Biol Chem 278, 42596-42603.

Mai, A., Veltel, S., Pellinen, T., Padzik, A., Coffey, E., Marjomaki, V., and Ivaska, J. (2011). Competitive binding of Rab21 and p120RasGAP to integrins regulates receptor traffic and migration. The Journal of cell biology 194, 291-306.

Maiuri, M.C., Zalckvar, E., Kimchi, A., and Kroemer, G. (2007). Self-eating and self- killing: crosstalk between autophagy and apoptosis. Nat Rev Mol Cell Biol 8, 741-752.

192

Marengere, L.E., Songyang, Z., Gish, G.D., Schaller, M.D., Parsons, J.T., Stern, M.J., Cantley, L.C., and Pawson, T. (1994). SH2 domain specificity and activity modified by a single residue. Nature 369, 502-505.

Margolis, B., Li, N., Koch, A., Mohammadi, M., Hurwitz, D.R., Zilberstein, A., Ullrich, A., Pawson, T., and Schlessinger, J. (1990). The tyrosine phosphorylated carboxyterminus of the EGF receptor is a binding site for GAP and PLC-gamma. EMBO J 9, 4375-4380.

Marsh, M., and McMahon, H.T. (1999). The structural era of endocytosis. Science 285, 215-220.

Marshall, M.S. (1993). The effector interactions of p21ras. Trends Biochem Sci 18, 250- 254.

Marshall, M.S., Hill, W.S., Ng, A.S., Vogel, U.S., Schaber, M.D., Scolnick, E.M., Dixon, R.A., Sigal, I.S., and Gibbs, J.B. (1989). A C-terminal domain of GAP is sufficient to stimulate ras p21 GTPase activity. EMBO J 8, 1105-1110.

Martin, G.A., Yatani, A., Clark, R., Conroy, L., Polakis, P., Brown, A.M., and McCormick, F. (1992). GAP domains responsible for ras p21-dependent inhibition of muscarinic atrial K+ channel currents. Science 255, 192-194.

Masuelli, L., and Cutler, M.L. (1996). Increased expression of the Ras suppressor Rsu-1 enhances Erk-2 activation and inhibits Jun kinase activation. Mol Cell Biol 16, 5466-5476.

Matsuda, M., Mayer, B.J., Fukui, Y., and Hanafusa, H. (1990). Binding of transforming protein, P47gag-crk, to a broad range of phosphotyrosine-containing proteins. Science 248, 1537-1539.

Mattera, R., and Bonifacino, J.S. (2008). Ubiquitin binding and conjugation regulate the recruitment of Rabex-5 to early endosomes. EMBO J 27, 2484-2494.

Mattera, R., Tsai, Y.C., Weissman, A.M., and Bonifacino, J.S. (2006). The Rab5 guanine nucleotide exchange factor Rabex-5 binds ubiquitin (Ub) and functions as a Ub ligase through an atypical Ub-interacting motif and a zinc finger domain. J Biol Chem 281, 6874- 6883.

Matthews, B.J., Kim, M.E., Flanagan, J.J., Hattori, D., Clemens, J.C., Zipursky, S.L., and Grueber, W.B. (2007). Dendrite self-avoidance is controlled by Dscam. Cell 129, 593-604.

Mayer, B.J., and Baltimore, D. (1993). Signalling through SH2 and SH3 domains. Trends Cell Biol 3, 8-13.

McGlade, J., Brunkhorst, B., Anderson, D., Mbamalu, G., Settleman, J., Dedhar, S., Rozakis-Adcock, M., Chen, L.B., and Pawson, T. (1993). The N-terminal region of GAP regulates cytoskeletal structure and cell adhesion. EMBO J 12, 3073-3081.

Melendez, A., and Neufeld, T.P. (2008). The cell biology of autophagy in metazoans: a developing story. Development 135, 2347-2360.

193

Melzig, J., Rein, K.H., Schafer, U., Pfister, H., Jackle, H., Heisenberg, M., and Raabe, T. (1998). A protein related to p21-activated kinase (PAK) that is involved in neurogenesis in the Drosophila adult central nervous system. Curr Biol 8, 1223-1226.

Mercer, C.A., Kaliappan, A., and Dennis, P.B. (2009). A novel, human Atg13 binding protein, Atg101, interacts with ULK1 and is essential for macroautophagy. Autophagy 5, 649-662.

Miao, W., Eichelberger, L., Baker, L., and Marshall, M.S. (1996). p120 Ras GTPase- activating protein interacts with Ras-GTP through specific conserved residues. J Biol Chem 271, 15322-15329.

Michno, K., van de Hoef, D., Wu, H., and Boulianne, G.L. (2005). Demented flies? Using Drosophila to model human neurodegenerative diseases. Clin Genet 67, 468-475.

Mitin, N., Rossman, K.L., and Der, C.J. (2005). Signaling interplay in Ras superfamily function. Curr Biol 15, R563-574.

Miura, G.I., Roignant, J.Y., Wassef, M., and Treisman, J.E. (2008). Myopic acts in the endocytic pathway to enhance signaling by the Drosophila EGF receptor. Development 135, 1913-1922.

Mollat, P., Fournier, A., Yang, C.Z., Alsat, E., Zhang, Y., Evain-Brion, D., Grassi, J., and Thang, M.N. (1994). Species specificity and organ, cellular and subcellular localization of the 100 kDa Ras GTPase activating protein. J Cell Sci 107 ( Pt 3), 427-435.

Mollat, P., Zhang, G.Y., Frobert, Y., Zhang, Y.H., Fournier, A., Grassi, J., and Thang, M.N. (1992). Non-neutralizing monoclonal antibodies against Ras GTPase-activating protein: production, characterization and use in an enzyme immunometric assay. Biotechnology (N Y) 10, 1151-1156.

Molloy, C.J., Bottaro, D.P., Fleming, T.P., Marshall, M.S., Gibbs, J.B., and Aaronson, S.A. (1989). PDGF induction of tyrosine phosphorylation of GTPase activating protein. Nature 342, 711-714.

Molloy, C.J., Fleming, T.P., Bottaro, D.P., Cuadrado, A., and Aaronson, S.A. (1992). Platelet-derived growth factor stimulation of GTPase-activating protein tyrosine phosphorylation in control and c-H-ras-expressing NIH 3T3 cells correlates with p21ras activation. Mol Cell Biol 12, 3903-3909.

Moran, M.F., Koch, C.A., Anderson, D., Ellis, C., England, L., Martin, G.S., and Pawson, T. (1990). Src homology region 2 domains direct protein-protein interactions in signal transduction. Proc Natl Acad Sci U S A 87, 8622-8626.

Moran, M.F., Polakis, P., McCormick, F., Pawson, T., and 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.

Muqit, M.M., and Feany, M.B. (2002). Modelling neurodegenerative diseases in Drosophila: a fruitful approach? Nat Rev Neurosci 3, 237-243.

194

Nafisi, H., Banihashemi, B., Daigle, M., and Albert, P.R. (2008). GAP1(IP4BP)/RASA3 mediates Galphai-induced inhibition of mitogen-activated protein kinase. J Biol Chem 283, 35908-35917.

Nagai, H., Yasuda, S., Ohba, Y., Fukuda, M., and Nakamura, T. (2013). All members of the EPI64 subfamily of TBC/RabGAPs also have GAP activities towards Ras. J Biochem 153, 283-288.

Nars, M., and Vihinen, M. (2001). Coevolution of the domains of cytoplasmic tyrosine kinases. Mol Biol Evol 18, 312-321.

Neuman-Silberberg, F.S., Schejter, E., Hoffmann, F.M., and Shilo, B.Z. (1984). The Drosophila ras oncogenes: structure and nucleotide sequence. Cell 37, 1027-1033.

Noel, J.P. (1997). Turning off the Ras switch with the flick of a finger. Nat Struct Biol 4, 677-680.

Noto, S., Maeda, T., Hattori, S., Inazawa, J., Imamura, M., Asaka, M., and Hatakeyama, M. (1998). A novel human RasGAP-like gene that maps within the prostate cancer susceptibility locus at chromosome 1q25. FEBS Lett 441, 127-131.

Nur, E.K.M.S., and Maruta, H. (1992). The role of Gln61 and Glu63 of Ras GTPases in their activation by NF1 and Ras GAP. Mol Biol Cell 3, 1437-1442.

Ohba, Y., Mochizuki, N., Yamashita, S., Chan, A.M., Schrader, J.W., Hattori, S., Nagashima, K., and Matsuda, M. (2000). Regulatory proteins of R-Ras, TC21/R-Ras2, and M-Ras/R-Ras3. J Biol Chem 275, 20020-20026.

Oinuma, I., Ishikawa, Y., Katoh, H., and Negishi, M. (2004). The Semaphorin 4D receptor Plexin-B1 is a GTPase activating protein for R-Ras. Science 305, 862-865.

Olivier, J.P., Raabe, T., Henkemeyer, M., Dickson, B., Mbamalu, G., Margolis, B., Schlessinger, J., Hafen, E., and Pawson, T. (1993). A Drosophila SH2-SH3 adaptor protein implicated in coupling the sevenless tyrosine kinase to an activator of Ras guanine nucleotide exchange, Sos. Cell 73, 179-191.

Otomo, A., Hadano, S., Okada, T., Mizumura, H., Kunita, R., Nishijima, H., Showguchi- Miyata, J., Yanagisawa, Y., Kohiki, E., Suga, E., et al. (2003). ALS2, a novel guanine nucleotide exchange factor for the small GTPase Rab5, is implicated in endosomal dynamics. Hum Mol Genet 12, 1671-1687.

Owen, D., Campbell, L.J., Littlefield, K., Evetts, K.A., Li, Z., Sacks, D.B., Lowe, P.N., and Mott, H.R. (2008). The IQGAP1-Rac1 and IQGAP1-Cdc42 interactions: interfaces differ between the complexes. J Biol Chem 283, 1692-1704.

Pacifico, S., Liu, G., Guest, S., Parrish, J.R., Fotouhi, F., and Finley, R.L., Jr. (2006). A database and tool, IM Browser, for exploring and integrating emerging gene and protein interaction data for Drosophila. BMC Bioinformatics 7, 195.

Pai, L.M., Barcelo, G., and Schupbach, T. (2000). D-cbl, a negative regulator of the Egfr pathway, is required for dorsoventral patterning in Drosophila oogenesis. Cell 103, 51-61.

195

Pal, A., Severin, F., Lommer, B., Shevchenko, A., and Zerial, M. (2006). Huntingtin- HAP40 complex is a novel Rab5 effector that regulates early endosome motility and is upregulated in Huntington's disease. J Cell Biol 172, 605-618.

Pamonsinlapatham, P., Gril, B., Dufour, S., Hadj-Slimane, R., Gigoux, V., Pethe, S., L'Hoste, S., Camonis, J., Garbay, C., Raynaud, F., et al. (2008). Capns1, a new binding partner of RasGAP-SH3 domain in K-Ras(V12) oncogenic cells: modulation of cell survival and migration. Cell Signal 20, 2119-2126.

Pamonsinlapatham, P., Hadj-Slimane, R., Lepelletier, Y., Allain, B., Toccafondi, M., Garbay, C., and Raynaud, F. (2009). P120-Ras GTPase activating protein (RasGAP): a multi-interacting protein in downstream signaling. Biochimie 91, 320-328.

Panasyuk, G., Nemazanyy, I., Filonenko, V., Negrutskii, B., and El'skaya, A.V. (2008). A2 isoform of mammalian translation factor eEF1A displays increased tyrosine phosphorylation and ability to interact with different signalling molecules. Int J Biochem Cell Biol 40, 63-71.

Paramasivam, S., Toma, N., Niimi, Y., and Berenstein, A. (2013). Development, clinical presentation and endovascular management of congenital intracranial pial arteriovenous fistulas. J Neurointerv Surg 5, 184-190.

Park, S., and Jove, R. (1993). Tyrosine phosphorylation of Ras GTPase-activating protein stabilizes its association with p62 at membranes of v-Src transformed cells. J Biol Chem 268, 25728-25734.

Park, S., Liu, X., Pawson, T., and Jove, R. (1992). Activated Src tyrosine kinase phosphorylates Tyr-457 of bovine GTPase-activating protein (GAP) in vitro and the corresponding residue of rat GAP in vivo. J Biol Chem 267, 17194-17200.

Parker, F., Maurier, F., Delumeau, I., Duchesne, M., Faucher, D., Debussche, L., Dugue, A., Schweighoffer, F., and Tocque, B. (1996). A Ras-GTPase-activating protein SH3- domain-binding protein. Mol Cell Biol 16, 2561-2569.

Pascal, S.M., Singer, A.U., Gish, G., Yamazaki, T., Shoelson, S.E., Pawson, T., Kay, L.E., and Forman-Kay, J.D. (1994). Nuclear magnetic resonance structure of an SH2 domain of phospholipase C-gamma 1 complexed with a high affinity binding peptide. Cell 77, 461- 472.

Pawson, T., and Gish, G.D. (1992). SH2 and SH3 domains: from structure to function. Cell 71, 359-362.

Pawson, T., Gish, G.D., and Nash, P. (2001). SH2 domains, interaction modules and cellular wiring. Trends Cell Biol 11, 504-511.

Pawson, T., Olivier, P., Rozakis-Adcock, M., McGlade, J., and Henkemeyer, M. (1993). Proteins with SH2 and SH3 domains couple receptor tyrosine kinases to intracellular signalling pathways. Philos Trans R Soc Lond B Biol Sci 340, 279-285.

Pawson, T., Raina, M., and Nash, P. (2002). Interaction domains: from simple binding events to complex cellular behavior. FEBS Lett 513, 2-10.

196

Penengo, L., Mapelli, M., Murachelli, A.G., Confalonieri, S., Magri, L., Musacchio, A., Di Fiore, P.P., Polo, S., and Schneider, T.R. (2006). Crystal structure of the ubiquitin binding domains of rabex-5 reveals two modes of interaction with ubiquitin. Cell 124, 1183-1195.

Perrimon, N. (1994). Signalling pathways initiated by receptor protein tyrosine kinases in Drosophila. Curr Opin Cell Biol 6, 260-266.

Polo, S., and Di Fiore, P.P. (2006). Endocytosis conducts the cell signaling orchestra. Cell 124, 897-900.

Presslauer, S., Hinterhuber, G., Cauza, K., Horvat, R., Rappersberger, K., Wolff, K., and Foedinger, D. (2003). RasGAP-like protein IQGAP1 is expressed by human keratinocytes and recognized by autoantibodies in association with bullous skin disease. J Invest Dermatol 120, 365-371.

Pronk, G.J., Medema, R.H., Burgering, B.M., Clark, R., McCormick, F., and Bos, J.L. (1992). Interaction between the p21ras GTPase activating protein and the insulin receptor. J Biol Chem 267, 24058-24063.

Purves, D., Augustine, G.J., and Fitzpatrick, D., eds. (2001). Neuroscience, 2nd edn (Sunderland (MA): Sinauer Associates).

Puttgen, K.B., and Lin, D.D. (2010). Neurocutaneous vascular syndromes. Childs Nerv Syst 26, 1407-1415.

Quilliam, L.A., Castro, A.F., Rogers-Graham, K.S., Martin, C.B., Der, C.J., and Bi, C. (1999). M-Ras/R-Ras3, a transforming ras protein regulated by Sos1, GRF1, and p120 Ras GTPase-activating protein, interacts with the putative Ras effector AF6. The Journal of biological chemistry 274, 23850-23857.

Raiborg, C., Slagsvold, T., and Stenmark, H. (2006). A new side to ubiquitin. Trends Biochem Sci 31, 541-544.

Rajalingam, K., Schreck, R., Rapp, U.R., and Albert, S. (2007). Ras oncogenes and their downstream targets. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1773, 1177-1195.

Rapoport, T.A. (2007). Protein translocation across the eukaryotic endoplasmic reticulum and bacterial plasma membranes. Nature 450, 663-669.

Revencu, N., Boon, L.M., Mulliken, J.B., Enjolras, O., Cordisco, M.R., Burrows, P.E., Clapuyt, P., Hammer, F., Dubois, J., Baselga, E., et al. (2008). Parkes Weber syndrome, vein of Galen aneurysmal malformation, and other fast-flow vascular anomalies are caused by RASA1 mutations. Hum Mutat 29, 959-965.

Rodriguez-Viciana, P., Sabatier, C., and McCormick, F. (2004). Signaling specificity by Ras family GTPases is determined by the full spectrum of effectors they regulate. Mol Cell Biol 24, 4943-4954.

Rorth, P. (1996). A modular misexpression screen in Drosophila detecting tissue-specific phenotypes. Proc Natl Acad Sci U S A 93, 12418-12422.

197

Rosenfeld, J., Capdevielle, J., Guillemot, J.C., and Ferrara, P. (1992). In-gel digestion of proteins for internal sequence analysis after one- or two-dimensional gel electrophoresis. Anal Biochem 203, 173-179.

Rothbauer, U., Zolghadr, K., Muyldermans, S., Schepers, A., Cardoso, M.C., and Leonhardt, H. (2008). A versatile nanotrap for biochemical and functional studies with fluorescent fusion proteins. Mol Cell Proteomics 7, 282-289.

Rouessac, F., and Rouessac, A. (2007). Chemical Analysis Modern Instrumentation Methods and Techniques, Second edn (Wiley).

Rubinsztein, D.C., Gestwicki, J.E., Murphy, L.O., and Klionsky, D.J. (2007). Potential therapeutic applications of autophagy. Nat Rev Drug Discov 6, 304-312.

Rumbaugh, G., Adams, J.P., Kim, J.H., and Huganir, R.L. (2006). SynGAP regulates synaptic strength and mitogen-activated protein kinases in cultured neurons. Proc Natl Acad Sci U S A 103, 4344-4351.

Sadowski, I., Stone, J.C., and Pawson, T. (1986). A noncatalytic domain conserved among cytoplasmic protein-tyrosine kinases modifies the kinase function and transforming activity of Fujinami sarcoma virus P130gag-fps. Mol Cell Biol 6, 4396-4408.

Saito, K., Murai, J., Kajiho, H., Kontani, K., Kurosu, H., and Katada, T. (2002). A novel binding protein composed of homophilic tetramer exhibits unique properties for the small GTPase Rab5. J Biol Chem 277, 3412-3418.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Cloning : A Laboratory Manual (New York : Cold Spring Harbor Laboratory Press).

Sandri, C., Caccavari, F., Valdembri, D., Camillo, C., Veltel, S., Santambrogio, M., Lanzetti, L., Bussolino, F., Ivaska, J., and Serini, G. (2012). The R-Ras/RIN2/Rab5 complex controls endothelial cell adhesion and morphogenesis via active integrin endocytosis and Rac signaling. Cell research 22, 1479-1501.

Sato, M., Sato, K., Fonarev, P., Huang, C.J., Liou, W., and Grant, B.D. (2005). Caenorhabditis elegans RME-6 is a novel regulator of RAB-5 at the clathrin-coated pit. Nat Cell Biol 7, 559-569.

Savitski, M.M., Lemeer, S., Boesche, M., Lang, M., Mathieson, T., Bantscheff, M., and Kuster, B. (2011). Confident phosphorylation site localization using the Mascot Delta Score. Mol Cell Proteomics 10, M110 003830.

Sayed, D., Hong, C., Chen, I.Y., Lypowy, J., and Abdellatif, M. (2007). MicroRNAs play an essential role in the development of cardiac hypertrophy. Circ Res 100, 416-424.

Scheffzek, K., Ahmadian, M.R., Kabsch, W., Wiesmuller, L., Lautwein, A., Schmitz, F., and Wittinghofer, A. (1997). The Ras-RasGAP complex: structural basis for GTPase activation and its loss in oncogenic Ras mutants. Science 277, 333-338.

Schejter, E.D., and Shilo, B.Z. (1985). Characterization of functional domains of p21 ras by use of chimeric genes. EMBO J 4, 407-412.

198

Schlessinger, J. (2000). Cell signaling by receptor tyrosine kinases. Cell 103, 211-225.

Schmidthals, K., Helma, J., Zolghadr, K., Rothbauer, U., and Leonhardt, H. (2010). Novel antibody derivatives for proteome and high-content analysis. Anal Bioanal Chem 397, 3203-3208.

Schmucker, D., Clemens, J.C., Shu, H., Worby, C.A., Xiao, J., Muda, M., Dixon, J.E., and Zipursky, S.L. (2000). Drosophila Dscam is an axon guidance receptor exhibiting extraordinary molecular diversity. Cell 101, 671-684.

Schneider, I. (1972). Cell lines derived from late embryonic stages of Drosophila melanogaster. J Embryol Exp Morphol 27, 353-365.

Schweighoffer, F., Barlat, I., Chevallier-Multon, M.C., and Tocque, B. (1992). Implication of GAP in Ras-dependent transactivation of a polyoma enhancer sequence. Science 256, 825-827.

Scott, R.C., Juhasz, G., and Neufeld, T.P. (2007). Direct induction of autophagy by Atg1 inhibits cell growth and induces apoptotic cell death. Curr Biol 17, 1-11.

Seachrist, J.L., Anborgh, P.H., and Ferguson, S.S. (2000). beta 2-adrenergic receptor internalization, endosomal sorting, and plasma membrane recycling are regulated by rab GTPases. J Biol Chem 275, 27221-27228.

Sermon, B.A., Eccleston, J.F., Skinner, R.H., and Lowe, P.N. (1996). Mechanism of inhibition by arachidonic acid of the catalytic activity of Ras GTPase-activating proteins. J Biol Chem 271, 1566-1572.

Serrano, M., Lin, A.W., McCurrach, M.E., Beach, D., and Lowe, S.W. (1997). Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell 88, 593-602.

Serth, J., Weber, W., Frech, M., Wittinghofer, A., and Pingoud, A. (1992). Binding of the H-ras p21 GTPase activating protein by the activated epidermal growth factor receptor leads to inhibition of the p21 GTPase activity in vitro. Biochemistry 31, 6361-6365.

Settleman, J., Albright, C.F., Foster, L.C., and Weinberg, R.A. (1992). Association between GTPase activators for Rho and Ras families. Nature 359, 153-154.

Shang, X., Moon, S.Y., and Zheng, Y. (2007). p200 RhoGAP promotes cell proliferation by mediating cross-talk between Ras and Rho signaling pathways. J Biol Chem 282, 8801- 8811.

Shannon, P., Markiel, A., Ozier, O., Baliga, N.S., Wang, J.T., Ramage, D., Amin, N., Schwikowski, B., and Ideker, T. (2003). Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome research 13, 2498-2504.

Shevchenko, A., Jensen, O.N., Podtelejnikov, A.V., Sagliocco, F., Wilm, M., Vorm, O., Mortensen, P., Boucherie, H., and Mann, M. (1996). Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two dimensional gels. Proc Natl Acad Sci U S A 93, 14440-14445.

199

Shilo, B.Z., and Weinberg, R.A. (1981). DNA sequences homologous to vertebrate oncogenes are conserved in Drosophila melanogaster. Proc Natl Acad Sci U S A 78, 6789- 6792.

Simon, M.A., Bowtell, D.D., Dodson, G.S., Laverty, T.R., and Rubin, G.M. (1991). Ras1 and a putative guanine nucleotide exchange factor perform crucial steps in signaling by the sevenless protein tyrosine kinase. Cell 67, 701-716.

Simon, M.A., Bowtell, D.D., and Rubin, G.M. (1989). Structure and activity of the sevenless protein: a protein tyrosine kinase receptor required for photoreceptor development in Drosophila. Proc Natl Acad Sci U S A 86, 8333-8337.

Simon, M.A., Dodson, G.S., and Rubin, G.M. (1993). An SH3-SH2-SH3 protein is required for p21Ras1 activation and binds to sevenless and Sos proteins in vitro. Cell 73, 169-177.

Simonsen, A., Cumming, R.C., Brech, A., Isakson, P., Schubert, D.R., and Finley, K.D. (2008). Promoting basal levels of autophagy in the nervous system enhances longevity and oxidant resistance in adult Drosophila. Autophagy 4, 176-184.

Simonsen, A., and Tooze, S.A. (2009). Coordination of membrane events during autophagy by multiple class III PI3-kinase complexes. J Cell Biol 186, 773-782.

Sims, D., Duchek, P., and Baum, B. (2009). PDGF/VEGF signaling controls cell size in Drosophila. Genome Biol 10, R20.

Skinner, R.H., Bradley, S., Brown, A.L., Johnson, N.J., Rhodes, S., Stammers, D.K., and Lowe, P.N. (1991). Use of the Glu-Glu-Phe C-terminal epitope for rapid purification of the catalytic domain of normal and mutant ras GTPase-activating proteins. J Biol Chem 266, 14163-14166.

Smida, M., Posevitz-Fejfar, A., Horejsi, V., Schraven, B., and Lindquist, J.A. (2007). A novel negative regulatory function of the phosphoprotein associated with glycosphingolipid-enriched microdomains: blocking Ras activation. Blood 110, 596-615.

Smith, C.C., Nelson, J., Aurelian, L., Gober, M., and Goswami, B.B. (2000). Ras-GAP binding and phosphorylation by herpes simplex virus type 2 RR1 PK (ICP10) and activation of the Ras/MEK/MAPK mitogenic pathway are required for timely onset of virus growth. J Virol 74, 10417-10429.

Smoot, M.E., Ono, K., Ruscheinski, J., Wang, P.L., and Ideker, T. (2011). Cytoscape 2.8: new features for data integration and network visualization. Bioinformatics 27, 431-432.

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 recognize specific phosphopeptide sequences. Cell 72, 767-778.

Sot, B., Behrmann, E., Raunser, S., and Wittinghofer, A. (2013). Ras GTPase activating (RasGAP) activity of the dual specificity GAP protein Rasal requires colocalization and C2 domain binding to lipid membranes. Proc Natl Acad Sci U S A 110, 111-116.

200

Sot, B., Kotting, C., Deaconescu, D., Suveyzdis, Y., Gerwert, K., and Wittinghofer, A. (2010). Unravelling the mechanism of dual-specificity GAPs. EMBO J 29, 1205-1214.

Stenmark, H. (2009). Rab GTPases as coordinators of vesicle traffic. Nat Rev Mol Cell Biol 10, 513-525.

Stetak, A., Gutierrez, P., and Hajnal, A. (2008). Tissue-specific functions of the Caenorhabditis elegans p120 Ras GTPase activating protein GAP-3. Dev Biol 323, 166- 176.

Stone, J.C., Colleton, M., and Bottorff, D. (1993). Effector domain mutations dissociate p21ras effector function and GTPase-activating protein interaction. Mol Cell Biol 13, 7311-7320.

Strubbe, G., Popp, C., Schmidt, A., Pauli, A., Ringrose, L., Beisel, C., and Paro, R. (2011). Polycomb purification by in vivo biotinylation tagging reveals cohesin and Trithorax group proteins as interaction partners. Proc Natl Acad Sci U S A 108, 5572-5577.

Su, X., Kong, C., and Stahl, P.D. (2007). GAPex-5 mediates ubiquitination, trafficking, and degradation of epidermal growth factor receptor. J Biol Chem 282, 21278-21284.

Sun, D., Yu, F., Ma, Y., Zhao, R., Chen, X., Zhu, J., Zhang, C.Y., Chen, J., and Zhang, J. (2013). MicroRNA-31 activates the RAS pathway and functions as an oncogenic MicroRNA in human colorectal cancer by repressing RAS p21 GTPase activating protein 1 (RASA1). J Biol Chem 288, 9508-9518.

Sylvester, M., Kliche, S., Lange, S., Geithner, S., Klemm, C., Schlosser, A., Grossmann, A., Stelzl, U., Schraven, B., Krause, E., et al. (2010). Adhesion and degranulation promoting adapter protein (ADAP) is a central hub for phosphotyrosine-mediated interactions in T cells. PLoS One 5, e11708.

Szabo, K., Jekely, G., and Rorth, P. (2001). Cloning and expression of sprint, a Drosophila homologue of RIN1. Mech Dev 101, 259-262.

Takai, Y., Sasaki, T., and Matozaki, T. (2001). Small GTP-binding proteins. Physiol Rev 81, 153-208.

Tall, G.G., Barbieri, M.A., Stahl, P.D., and Horazdovsky, B.F. (2001). Ras-activated endocytosis is mediated by the Rab5 guanine nucleotide exchange activity of RIN1. Dev Cell 1, 73-82.

Tallquist, M.D., French, W.J., and Soriano, P. (2003). Additive effects of PDGF receptor beta signaling pathways in vascular smooth muscle cell development. PLoS Biol 1, E52.

Tamura, K., Ohbayashi, N., Maruta, Y., Kanno, E., Itoh, T., and Fukuda, M. (2009). Varp is a novel Rab32/38-binding protein that regulates Tyrp1 trafficking in melanocytes. Mol Biol Cell 20, 2900-2908.

The, I., Hannigan, G.E., Cowley, G.S., Reginald, S., Zhong, Y., Gusella, J.F., Hariharan, I.K., and Bernards, A. (1997). Rescue of a Drosophila NF1 mutant phenotype by protein kinase A. Science 276, 791-794.

201

Thiex, R., Mulliken, J.B., Revencu, N., Boon, L.M., Burrows, P.E., Cordisco, M., Dwight, Y., Smith, E.R., Vikkula, M., and Orbach, D.B. (2010). A novel association between RASA1 mutations and spinal arteriovenous anomalies. AJNR Am J Neuroradiol 31, 775- 779.

Tocque, B., Delumeau, I., Parker, F., Maurier, F., Multon, M.C., and Schweighoffer, F. (1997). Ras-GTPase activating protein (GAP): a putative effector for Ras. Cell Signal 9, 153-158.

Tomshine, J.C., Severson, S.R., Wigle, D.A., Sun, Z., Beleford, D.A., Shridhar, V., and Horazdovsky, B.F. (2009). Cell proliferation and epidermal growth factor signaling in non- small cell lung adenocarcinoma cell lines are dependent on Rin1. J Biol Chem 284, 26331- 26339.

Tong, J., Elowe, S., Nash, P., and Pawson, T. (2003). Manipulation of EphB2 regulatory motifs and SH2 binding sites switches MAPK signaling and biological activity. J Biol Chem 278, 6111-6119.

Tootle, T.L., Lee, P.S., and Rebay, I. (2003). CRM1-mediated nuclear export and regulated activity of the Receptor Tyrosine Kinase antagonist YAN require specific interactions with MAE. Development 130, 845-857.

Trahey, M., and McCormick, F. (1987). A cytoplasmic protein stimulates normal N-ras p21 GTPase, but does not affect oncogenic mutants. Science 238, 542-545.

Trahey, M., Wong, G., Halenbeck, R., Rubinfeld, B., Martin, G.A., Ladner, M., Long, C.M., Crosier, W.J., Watt, K., Koths, K., et al. (1988). Molecular cloning of two types of GAP complementary DNA from human placenta. Science 242, 1697-1700.

Trentin, G.A., Yin, X., Tahir, S., Lhotak, S., Farhang-Fallah, J., Li, Y., and Rozakis- Adcock, M. (2001). A mouse homologue of the Drosophila tumor suppressor l(2)tid gene defines a novel Ras GTPase-activating protein (RasGAP)-binding protein. J Biol Chem 276, 13087-13095.

Trouliaris, S., Smola, U., Chang, J.H., Parsons, S.J., Niemann, H., and Tamura, T. (1995). Tyrosine 807 of the v-Fms oncogene product controls cell morphology and association with p120RasGAP. J Virol 69, 6010-6020.

Uesugi, K., Oinuma, I., Katoh, H., and Negishi, M. (2009). Different requirement for Rnd GTPases of R-Ras GAP activity of Plexin-C1 and Plexin-D1. J Biol Chem 284, 6743- 6751.

Valsdottir, R., Hashimoto, H., Ashman, K., Koda, T., Storrie, B., and Nilsson, T. (2001). Identification of rabaptin-5, rabex-5, and GM130 as putative effectors of rab33b, a regulator of retrograde traffic between the Golgi apparatus and ER. FEBS Lett 508, 201- 209. van der Geer, P., Henkemeyer, M., Jacks, T., and Pawson, T. (1997). Aberrant Ras regulation and reduced p190 tyrosine phosphorylation in cells lacking p120-Gap. Mol Cell Biol 17, 1840-1847.

202

Venkatachalam, K., Long, A.A., Elsaesser, R., Nikolaeva, D., Broadie, K., and Montell, C. (2008). Motor deficit in a Drosophila model of mucolipidosis type IV due to defective clearance of apoptotic cells. Cell 135, 838-851.

VijayRaghavan, K., Kaur, J., Paranjape, J., and Rodrigues, V. (1992). The east gene of Drosophila melanogaster is expressed in the developing embryonic nervous system and is required for normal olfactory and gustatory responses of the adult. Dev Biol 154, 23-36.

Vizcaino, J.A., Cote, R.G., Csordas, A., Dianes, J.A., Fabregat, A., Foster, J.M., Griss, J., Alpi, E., Birim, M., Contell, J., et al. (2013). The PRoteomics IDEntifications (PRIDE) database and associated tools: status in 2013. Nucleic Acids Res 41, D1063-1069.

Vogel, U.S., Dixon, R.A., Schaber, M.D., Diehl, R.E., Marshall, M.S., Scolnick, E.M., Sigal, I.S., and Gibbs, J.B. (1988). Cloning of bovine GAP and its interaction with oncogenic ras p21. Nature 335, 90-93.

Waksman, G., Kominos, D., Robertson, S.C., Pant, N., Baltimore, D., Birge, R.B., Cowburn, D., Hanafusa, H., Mayer, B.J., Overduin, M., et al. (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., and 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.

Walker, J.A., Tchoudakova, A.V., McKenney, P.T., Brill, S., Wu, D., Cowley, G.S., Hariharan, I.K., and Bernards, A. (2006). Reduced growth of Drosophila neurofibromatosis 1 mutants reflects a non-cell-autonomous requirement for GTPase- Activating Protein activity in larval neurons. Genes Dev 20, 3311-3323.

Wang, F., Zhang, H., Zhang, X., Wang, Y., Ren, F., Zhang, X., Zhai, Y., and Chang, Z. (2008). Varp interacts with Rab38 and functions as its potential effector. Biochem Biophys Res Commun 372, 162-167.

Wang, J., and Ruan, K. (2010). miR-335 is involved in the rat epididymal development by targeting the mRNA of RASA1. Biochem Biophys Res Commun 402, 222-227.

Wang, Q.K. (2005). Update on the molecular genetics of vascular anomalies. Lymphat Res Biol 3, 226-233.

Wang, S., Watanabe, T., Noritake, J., Fukata, M., Yoshimura, T., Itoh, N., Harada, T., Nakagawa, M., Matsuura, Y., Arimura, N., et al. (2007). IQGAP3, a novel effector of Rac1 and Cdc42, regulates neurite outgrowth. J Cell Sci 120, 567-577.

Wang, Y., Waldron, R.T., Dhaka, A., Patel, A., Riley, M.M., Rozengurt, E., and Colicelli, J. (2002). The RAS effector RIN1 directly competes with RAF and is regulated by 14-3-3 proteins. Mol Cell Biol 22, 916-926.

Warne, P.H., Viciana, P.R., and Downward, J. (1993). Direct interaction of Ras and the amino-terminal region of Raf-1 in vitro. Nature 364, 352-355.

203

Wasser, M., and Chia, W. (2000). The EAST protein of drosophila controls an expandable nuclear endoskeleton. Nat Cell Biol 2, 268-275.

Weissbach, L., Settleman, J., Kalady, M.F., Snijders, A.J., Murthy, A.E., Yan, Y.X., and Bernards, A. (1994). Identification of a human rasGAP-related protein containing calmodulin-binding motifs. J Biol Chem 269, 20517-20521.

Woller, B., Luiskandl, S., Popovic, M., Prieler, B.E., Ikonge, G., Mutzl, M., Rehmann, H., and Herbst, R. (2011). Rin-like, a novel regulator of endocytosis, acts as guanine nucleotide exchange factor for Rab5a and Rab22. Biochim Biophys Acta 1813, 1198-1210.

Woodcock, S.A. (2004). Analysis of the function of p120 RasGAP. In Department of Biomolecular Sciences (Manchester: University of Manchester Institute of Science and Technology).

Woodcock, S.A., and Hughes, D.A. (2004). p120 Ras GTPase-activating protein associates with fibroblast growth factor receptors in Drosophila. Biochem J 380, 767-774.

Wucherpfennig, T., Wilsch-Brauninger, M., and Gonzalez-Gaitan, M. (2003). Role of Drosophila Rab5 during endosomal trafficking at the synapse and evoked neurotransmitter release. J Cell Biol 161, 609-624.

Xu, J., and Wong, C. (2008). A computational screen for mouse signaling pathways targeted by microRNA clusters. RNA 14, 1276-1283.

Xu, N., Coso, O., Mahadevan, D., De Blasi, A., Goldsmith, P.K., Simonds, W.F., and Gutkind, J.S. (1996). The PH domain of Ras-GAP is sufficient for in vitro binding to beta gamma subunits of heterotrimeric G proteins. Cell Mol Neurobiol 16, 51-59.

Yaffe, M.B. (2002). Phosphotyrosine-binding domains in signal transduction. Nat Rev Mol Cell Biol 3, 177-186.

Yan, H., Jahanshahi, M., Horvath, E.A., Liu, H.Y., and Pfleger, C.M. (2010). Rabex-5 ubiquitin ligase activity restricts Ras signaling to establish pathway homeostasis in Drosophila. Curr Biol 20, 1378-1382.

Yang, H., Zhang, C., Lu, Y.X., Wu, X.J., Yuan, L., Zhou, C., Zhou, C.P., Liu, G.B., and Li, X.N. (2012). [Construction of has-miR-335 lentiviral vector and verification of the target gene of miR-335]. Nan Fang Yi Ke Da Xue Xue Bao 32, 306-311.

Yang, Y., Hentati, A., Deng, H.X., Dabbagh, O., Sasaki, T., Hirano, M., Hung, W.Y., Ouahchi, K., Yan, J., Azim, A.C., et al. (2001). The gene encoding alsin, a protein with three guanine-nucleotide exchange factor domains, is mutated in a form of recessive amyotrophic lateral sclerosis. Nat Genet 29, 160-165.

Yatani, A., Okabe, K., Polakis, P., Halenbeck, R., McCormick, F., and Brown, A.M. (1990). ras p21 and GAP inhibit coupling of muscarinic receptors to atrial K+ channels. Cell 61, 769-776.

Ye, F., Cayre, Y.E., and Thang, M.N. (1999). Evidence for a novel RasGAP-associated protein of 105 kDa in both mature trophoblasts and differentiating choriocarcinoma cells. Biochem Biophys Res Commun 263, 523-527.

204

Yoshikawa, M., Kajiho, H., Sakurai, K., Minoda, T., Nakagawa, S., Kontani, K., and Katada, T. (2008). Tyr-phosphorylation signals translocate RIN3, the small GTPase Rab5- GEF, to early endocytic vesicles. Biochem Biophys Res Commun 372, 168-172.

Young, A.R., Narita, M., Ferreira, M., Kirschner, K., Sadaie, M., Darot, J.F., Tavare, S., Arakawa, S., Shimizu, S., Watt, F.M., et al. (2009). Autophagy mediates the mitotic senescence transition. Genes Dev 23, 798-803.

Yu, J., Pacifico, S., Liu, G., and Finley, R.L., Jr. (2008). DroID: the Drosophila Interactions Database, a comprehensive resource for annotated gene and protein interactions. BMC Genomics 9, 461.

Zettl, M., Adrain, C., Strisovsky, K., Lastun, V., and Freeman, M. (2011). Rhomboid family pseudoproteases use the ER quality control machinery to regulate intercellular signaling. Cell 145, 79-91.

Zhang, J., Schulze, K.L., Hiesinger, P.R., Suyama, K., Wang, S., Fish, M., Acar, M., Hoskins, R.A., Bellen, H.J., and Scott, M.P. (2007). Thirty-one flavors of Drosophila rab proteins. Genetics 176, 1307-1322.

Zhang, W., Thompson, B.J., Hietakangas, V., and Cohen, S.M. (2011). MAPK/ERK signaling regulates insulin sensitivity to control glucose metabolism in Drosophila. PLoS genetics 7, e1002429.

Zhang, X., He, X., Fu, X.Y., and Chang, Z. (2006). Varp is a Rab21 guanine nucleotide exchange factor and regulates endosome dynamics. J Cell Sci 119, 1053-1062.

Zhang, X.F., Settleman, J., Kyriakis, J.M., Takeuchi-Suzuki, E., Elledge, S.J., Marshall, M.S., Bruder, J.T., Rapp, U.R., and Avruch, J. (1993a). Normal and oncogenic p21ras proteins bind to the amino-terminal regulatory domain of c-Raf-1. Nature 364, 308-313.

Zhang, Y., Zhang, G., Mollat, P., Carles, C., Riva, M., Frobert, Y., Malassine, A., Rostene, W., Thang, D.C., Beltchev, B., et al. (1993b). Purification, characterization, and cellular localization of the 100-kDa human placental GTPase-activating protein. J Biol Chem 268, 18875-18881.

Zhou, Q., Zheng, J.W., Yang, X.J., Wang, H.J., Ma, D., and Qin, Z.P. (2011). Detection of RASA1 mutations in patients with sporadic Sturge-Weber syndrome. Childs Nerv Syst 27, 603-607.

Zhou, X., Wang, L., Hasegawa, H., Amin, P., Han, B.X., Kaneko, S., He, Y., and Wang, F. (2010). Deletion of PIK3C3/Vps34 in sensory neurons causes rapid neurodegeneration by disrupting the endosomal but not the autophagic pathway. Proc Natl Acad Sci U S A 107, 9424-9429.

Zhu, H., Liang, Z., and Li, G. (2009). Rabex-5 is a Rab22 effector and mediates a Rab22- Rab5 signaling cascade in endocytosis. Mol Biol Cell 20, 4720-4729.

205

APPENDIX

Table S1. Hierarchical clustering of RasGAP interacting proteins (full datasets)

206

Mean normalised spectral count (ratio of total spectra) x 10 2 Hierarchical clustering Selected cluster (Fig 3.3) Accession number Protein name RasGAPWT RasGAPSH2*32* GFP Q7KMM4_DROME BcDNA.GH04962 (CG14476-PC, isoform C) (CG14476-PD, isoform D) (CG14476-PE, isoform E) (CG14476-P A, isoform A) OS=Drosophila melanogaster GN=BcDNA.GH04962 PE=2 SV=1 0.373 0.458 0.765 NCD_DROME Protein claret segregational OS=Drosophila melanogaster GN=ncd PE=1 SV=1 0.303 0.367 0.621 O76279_DROME (+1) Transitional endoplasmic reticulum ATPase TER94 OS=Drosophila melanogaster GN=TER94 PE=2 SV=1 0.514 0.616 1.080 Q9VHY6_DROME CG2943-PA (LD19064p) OS=Drosophila melanogaster GN=CG2943 PE=2 SV=1 0.233 0.288 0.504 Q24160_DROME (+2) Hemomucin OS=Drosophila melanogaster GN=Hmu PE=2 SV=1 0.280 0.314 0.549 Q8I0G5_DROME (+3) CG1528-PB, isoform B (RE37840p) OS=Drosophila melanogaster GN=gammaCop PE=2 SV=1 0.281 0.315 0.567 Q9VY76_DROME (+1) CG32626-PA, isoform A OS=Drosophila melanogaster GN=CG32626 PE=1 SV=3 0.479 0.510 0.882 UGGG_DROME UDP-glucose:glycoprotein glucosyltransferase OS=Drosophila melanogaster GN=Ugt PE=1 SV=2 0.677 0.772 1.259 A1Z9N0_DROME (+1) CG8479-PA, isoform A OS=Drosophila melanogaster GN=opa1-like PE=3 SV=1 0.268 0.301 0.495 NU214_DROME Nuclear pore complex protein Nup214 OS=Drosophila melanogaster GN=Nup214 PE=1 SV=2 0.151 0.170 0.288 KPYK_DROME Pyruvate kinase OS=Drosophila melanogaster GN=PyK PE=1 SV=2 0.317 0.393 0.620 Q0PQ31_DROME (+1) Microtubule dependent motor protein OS=Drosophila melanogaster GN=klp61F PE=3 SV=1 0.224 0.262 0.414 Q9VW54_DROME CG7762-PA (26S proteasome regulatory complex subunit p97) (Putative uncharacterized protein) OS=Drosophila melanoga ster GN=Rpn1 PE=1 SV=2 0.373 0.472 0.702 HSP7C_DROME Heat shock 70 kDa protein cognate 3 OS=Drosophila melanogaster GN=Hsc70-3 PE=2 SV=2 0.271 0.330 0.503 EF1A1_DROME Elongation factor 1-alpha 1 OS=Drosophila melanogaster GN=Ef1alpha48D PE=1 SV=2 2.460 2.490 4.587 ATC1_DROME Calcium-transporting ATPase sarcoplasmic/endoplasmic reticulum type OS=Drosophila melanogaster GN=Ca-P60A PE=1 SV=2 0.735 0.720 1.386 PDI_DROME (+1) Protein disulfide-isomerase OS=Drosophila melanogaster GN=Pdi PE=2 SV=1 0.924 0.878 1.619 Q8T4G5_DROME SD01613p (CG6512-PA, isoform A) OS=Drosophila melanogaster GN=CG6512 PE=1 SV=1 0.293 0.275 0.522 Q9VUC1_DROME CG6603-PA, isoform A (CG6603-PB, isoform B) (CG6603-PC, isoform C) (LD32979p) OS=Drosophila melanogaster GN=Hsc70Cb PE=2 SV=1 0.329 0.314 0.540 Q9XYU1_DROME CG4206-PA (DNA replication factor MCM3) OS=Drosophila melanogaster GN=Mcm3 PE=1 SV=1 0.282 0.262 0.450 A1ZBE9_DROME CG15100-PA OS=Drosophila melanogaster GN=CG15100 PE=4 SV=1 0.270 0.236 0.441 Q2UXP1_DROSI Cg1440 protein (Fragment) OS=Drosophila simulans GN=CG1440 PE=2 SV=1 0.045 0.040 0.072 Q8IRH0_DROME (+2) CG1009-PE, isoform E OS=Drosophila melanogaster GN=Psa PE=2 SV=1 0.805 0.877 2.105 COPB_DROME Coatomer subunit beta OS=Drosophila melanogaster GN=betaCop PE=1 SV=2 0.140 0.157 0.351 EF2_DROME Elongation factor 2 OS=Drosophila melanogaster GN=Ef2b PE=1 SV=4 1.051 1.008 2.456 K6PF_DROME 6-phosphofructokinase OS=Drosophila melanogaster GN=Pfk PE=2 SV=2 0.176 0.157 0.414 Q9W2U7_DROME CG17255-PA, isoform A (CG17255-PB, isoform B) (LP18708p) OS=Drosophila melanogaster GN=CG17255-RA PE=1 SV=4 0.176 0.145 0.387 CATA_DROME Catalase OS=Drosophila melanogaster GN=Cat PE=1 SV=2 1.582 1.479 3.111 Q9VFF0_DROME CG3731-PA, isoform A (CG3731-PB, isoform B) (GH01077p) OS=Drosophila melanogaster GN=CG3731 PE=2 SV=2 0.551 0.511 1.089 RIR1_DROME Ribonucleoside-diphosphate reductase large subunit OS=Drosophila melanogaster GN=RnrL PE=1 SV=2 0.597 0.589 1.178 Q9VPL5_DROME CG11490-PA (LD27216p) OS=Drosophila melanogaster GN=CG11490 PE=1 SV=1 0.094 0.091 0.198 Q9VAY2_DROME CG5520-PA (LD23641p) OS=Drosophila melanogaster GN=Gp93 PE=1 SV=1 1.088 1.312 2.690 Q9V438_DROME CG5809-PA (LD28038p) OS=Drosophila melanogaster GN=CaBP1 PE=1 SV=1 0.725 0.825 1.583 Q9I7T7_DROME (+1) CG11505-PB, isoform B OS=Drosophila melanogaster GN=CG11505 PE=1 SV=2 0.036 0.040 0.081 COPB2_DROME Coatomer subunit beta' OS=Drosophila melanogaster GN=beta'Cop PE=2 SV=2 0.187 0.236 0.450 Q9W228_DROME CG4414-PA (RE07815p) OS=Drosophila melanogaster GN=Ugt58Fa PE=2 SV=3 0.059 0.079 0.144 Q8IQQ0_DROME CG11661-PF, isoform F (CG11661-PH, isoform H) (RE42354p) OS=Drosophila melanogaster GN=Nc73EF PE=2 SV=1 0.177 0.118 0.297 MA205_DROME (+1) 205 kDa microtubule-associated protein OS=Drosophila melanogaster GN=Map205 PE=1 SV=2 0.256 0.210 0.468 NONA_DROME (+1) Protein no-on-transient A OS=Drosophila melanogaster GN=nonA PE=1 SV=2 0.071 0.052 0.126 TOP2_DROME DNA topoisomerase 2 OS=Drosophila melanogaster GN=Top2 PE=1 SV=1 0.281 0.196 0.423 TID_DROME Protein tumorous imaginal discs, mitochondrial OS=Drosophila melanogaster GN=l(2)tid PE=1 SV=2 0.034 0.026 0.054 Q9VQL7_DROME CG3523-PA OS=Drosophila melanogaster GN=CG3523 PE=2 SV=1 0.585 0.433 0.621 Q9W002_DROME CG16973-PA, isoform A OS=Drosophila melanogaster GN=msn PE=1 SV=3 0.560 0.432 0.620 A4IJ68_DROME IP17795p (Fragment) OS=Drosophila melanogaster GN=nemy PE=2 SV=1 0.071 0.053 0.072 Q9VH64_DROME CG8507-PA (LD29322p) OS=Drosophila melanogaster GN=CG8507 PE=1 SV=1 0.117 0.079 0.135 Q9VS57_DROME CG8583-PA (RE14391p) OS=Drosophila melanogaster GN=sec63 PE=1 SV=1 0.093 0.065 0.108 Q9XYZ5_DROME (+1) CG7769-PA (Damage-specific DNA binding protein DDBa p127 subunit) OS=Drosophila melanogaster GN=DDB1 PE=2 SV=1 0.071 0.052 0.081 RL3_DROME 60S ribosomal protein L3 OS=Drosophila melanogaster GN=RpL3 PE=1 SV=3 0.447 0.301 0.567 ATX2_DROME Ataxin-2 homolog OS=Drosophila melanogaster GN=Atx2 PE=1 SV=1 0.059 0.040 0.072 MCM4_DROME DNA replication licensing factor MCM4 OS=Drosophila melanogaster GN=dpa PE=1 SV=2 0.082 0.052 0.108 A4UZW0_DROME (+1) CG2671-PE, isoform E (CG2671-PF, isoform F) OS=Drosophila melanogaster GN=l(2)gl PE=4 SV=1 0.094 0.052 0.108 MCM6_DROME DNA replication licensing factor Mcm6 OS=Drosophila melanogaster GN=Mcm6 PE=1 SV=1 0.094 0.052 0.108 Q961B3_DROME LD24878p (CG2277-PA) OS=Drosophila melanogaster GN=CG2277 PE=2 SV=1 0.070 0.039 0.081 CLASP_DROME CLIP-associating protein OS=Drosophila melanogaster GN=chb PE=1 SV=1 0.081 0.039 0.081 Q86NR4_DROME (+1) RE29053p OS=Drosophila melanogaster GN=HBS1 PE=1 SV=1 0.070 0.040 0.072 SPTCA_DROME Spectrin alpha chain OS=Drosophila melanogaster GN=alpha-Spec PE=1 SV=2 0.296 0.079 0.342 Q8SXU3_DROME RE21160p (CG8207-PA) OS=Drosophila melanogaster GN=CG8207 PE=1 SV=1 0.082 0.026 0.108 Q9V9R2_DROME CG1512-PA, isoform A (CG1512-PB, isoform B) (LD36177p) OS=Drosophila melanogaster GN=cul-2 PE=2 SV=3 0.094 0.039 0.108 Q8T0L3_DROME Ubiquitin-activating enzyme E1 OS=Drosophila melanogaster GN=Uba1 PE=1 SV=1 0.377 0.485 0.567 O77466_DROME (+1) Thiolase OS=Drosophila melanogaster GN=Thiolase PE=2 SV=1 0.198 0.249 0.297 P91676_DROME (+1) MCM5 homolog OS=Drosophila melanogaster GN=Mcm5 PE=1 SV=1 0.175 0.223 0.270 Q9VK69_DROME T-complex protein 1, delta subunit OS=Drosophila melanogaster GN=CG5525 PE=1 SV=1 0.234 0.262 0.323 Q0E9E2_DROME (+1) CG1516-PI, isoform I (CG1516-PK, isoform K) (CG1516-PL, isoform L) OS=Drosophila melanogaster GN=CG1516 PE=4 S V=1 0.305 0.367 0.423 Q9NH72_DROME (+1) Rasputin OS=Drosophila melanogaster GN=rin PE=1 SV=1 0.164 0.197 0.234 Q8SYE5_DROME (+1) RE65203p OS=Drosophila melanogaster GN=eIF3-S10 PE=2 SV=1 0.621 0.827 0.863 A9J7N9_DROME (+2) Aldehyde dehydrogenase OS=Drosophila melanogaster GN=Aldh PE=2 SV=1 0.417 0.563 0.612 Q8T3L4_DROME SD01201p OS=Drosophila melanogaster GN=CG9268 PE=1 SV=1 0.048 0.065 0.072 Q7KN75_DROME Putative uncharacterized protein (CG5170-PA, isoform A) (CG5170-PB, isoform B) (CG5170-PD, isoform D) (CG5170-PE, i soform E) (CG5170-PF, isoform F) OS=Drosophila melanogaster GN=Dp1 PE=2 SV=1 0.560 0.603 0.639 Q9U5W6_DROME Microtubule associated protein OS=Drosophila melanogaster GN=msps PE=2 SV=1 0.445 0.458 0.486 Q7KKC2_DROME (+2) Filamin1 (Fragment) OS=Drosophila melanogaster GN=cher PE=1 SV=1 0.937 0.996 1.017 PSMD3_DROME (+1) Probable 26S proteasome non-ATPase regulatory subunit 3 OS=Drosophila melanogaster GN=Dox-A2 PE=1 SV=1 0.455 0.485 0.486 PYR1_DROME CAD protein OS=Drosophila melanogaster GN=r PE=1 SV=3 0.526 0.564 0.549 Q9VFQ9_DROME CG9285-PA, isoform A (CG9285-PB, isoform B) (CG9285-PC, isoform C) (LD41062p) OS=Drosophila melanogaster GN=Dip-B P E=1 SV=2 0.222 0.249 0.251 A8JV09_DROME (+2) CG4532-PF, isoform F OS=Drosophila melanogaster GN=pod1 PE=4 SV=1 0.130 0.144 0.144 Q7JPZ3_DROME D.melanogaster ubiquitin (Fragment) OS=Drosophila melanogaster GN=Ubi-p63E PE=2 SV=1 0.457 0.432 0.458 O46067_DROME CG2918-PA (GH11566p) (EG:25E8.1 protein) OS=Drosophila melanogaster GN=EG:25E8.1 PE=1 SV=1 0.350 0.328 0.360 DDX3_DROME ATP-dependent RNA helicase bel OS=Drosophila melanogaster GN=bel PE=1 SV=1 0.314 0.275 0.315 Q8IP79_DROME (+2) CG5547-PC, isoform C OS=Drosophila melanogaster GN=Pect PE=2 SV=2 0.106 0.091 0.108 O77285_DROME (+1) Coatomer alpha subunit OS=Drosophila melanogaster GN=alphaCop PE=2 SV=1 0.549 0.486 0.593 Q8SXQ1_DROME GH05218p (CG9629-PA) OS=Drosophila melanogaster GN=CG9629 PE=2 SV=1 0.176 0.157 0.198 A1Z784_DROME (+1) CG11198-PA, isoform A OS=Drosophila melanogaster GN=CG11198 PE=1 SV=1 0.528 0.498 0.567 Q76NQ0_DROME CG33303-PA OS=Drosophila melanogaster GN=CG33303 PE=1 SV=1 0.152 0.144 0.162 LAMB1_DROME Laminin subunit beta-1 OS=Drosophila melanogaster GN=LanB1 PE=1 SV=4 0.164 0.157 0.180 TPP2_DROME Tripeptidyl-peptidase 2 OS=Drosophila melanogaster GN=TppII PE=1 SV=2 0.353 0.341 0.413 Q9Y162_DROME CG6842-PA (BcDNA.GH02678) OS=Drosophila melanogaster GN=BcDNA.GH02678 PE=1 SV=1 0.094 0.092 0.108 Q9VY44_DROME CG1810-PA (RE70632p) OS=Drosophila melanogaster GN=mRNA-capping-enzyme PE=2 SV=1 0.095 0.092 0.108 HSP7E_DROME Heat shock 70 kDa protein cognate 5 OS=Drosophila melanogaster GN=Hsc70-5 PE=1 SV=2 0.294 0.264 0.368 Q9VLM8_DROME CG13391-PA, isoform A (CG13391-PB, isoform B) OS=Drosophila melanogaster GN=Aats-ala PE=2 SV=1 0.270 0.249 0.342 PUR6_DROME Multifunctional protein ADE2 OS=Drosophila melanogaster GN=ade5 PE=1 SV=2 0.178 0.157 0.216 CLH_DROME Clathrin heavy chain OS=Drosophila melanogaster GN=Chc PE=1 SV=1 0.723 0.721 0.953 Q7KN90_DROME Putative uncharacterized protein (CG8431-PA) OS=Drosophila melanogaster GN=Aats-cys PE=1 SV=1 0.140 0.144 0.189 Q9Y114_DROME CG8042-PA, isoform A (BcDNA.GH10229) OS=Drosophila melanogaster GN=BcDNA.GH10229 PE=2 SV=1 0.082 0.079 0.108 A1Z6H7_DROME (+2) CG7897-PA OS=Drosophila melanogaster GN=gp210 PE=1 SV=1 0.902 0.812 1.241 A4V2U2_DROME (+2) CG10851-PC, isoform C OS=Drosophila melanogaster GN=B52 PE=4 SV=1 0.116 0.105 0.162 A1Z9E3_DROME Elongation factor Tu OS=Drosophila melanogaster GN=EfTuM PE=3 SV=1 0.200 0.183 0.270 Q8IPE8_DROME CG4389-PB, isoform B (CG4389-PC, isoform C) OS=Drosophila melanogaster GN=CG4389 PE=2 SV=1 0.781 0.682 1.106 EBI_DROME F-box-like/WD repeat-containing protein ebi OS=Drosophila melanogaster GN=ebi PE=1 SV=2 0.094 0.078 0.135 Q8IGY5_DROME (+2) RE05782p (Fragment) OS=Drosophila melanogaster GN=CG1910 PE=2 SV=1 0.082 0.065 0.108 EF1G_DROME Elongation factor 1-gamma OS=Drosophila melanogaster GN=Ef1gamma PE=1 SV=2 0.456 0.524 0.783 Q7KSL9_DROME (+4) CG7340-PA, isoform A OS=Drosophila melanogaster GN=granny-smith PE=1 SV=1 0.199 0.236 0.342 HSP83_DROME Heat shock protein 83 OS=Drosophila melanogaster GN=Hsp83 PE=1 SV=1 1.669 1.758 2.724 Q0E940_DROME (+1) CG4878-PA, isoform A OS=Drosophila melanogaster GN=eIF3-S9 PE=4 SV=1 0.351 0.381 0.575 Q9W0R0_DROME CG6905-PA (LD21614p) OS=Drosophila melanogaster GN=CG6905 PE=1 SV=2 0.048 0.053 0.081 SIL1_DROME Nucleotide exchange factor SIL1 OS=Drosophila melanogaster GN=CG10420 PE=2 SV=1 0.116 0.118 0.180 Q3YMU0_DROME Erp60 (CG8983-PA, isoform A) (Fragment) OS=Drosophila melanogaster GN=ERp60 PE=2 SV=1 0.606 0.682 0.908 SYEP_DROME Bifunctional aminoacyl-tRNA synthetase OS=Drosophila melanogaster GN=Aats-glupro PE=1 SV=2 0.620 0.668 0.926 SPTCB_DROME Spectrin beta chain OS=Drosophila melanogaster GN=beta-Spec PE=1 SV=2 0.048 0.052 0.072 Q960G3_DROME (+1) SD03094p OS=Drosophila melanogaster GN=CG9311 PE=2 SV=1 0.048 0.052 0.072 O02393_DROME (+3) Calnexin OS=Drosophila melanogaster GN=Cnx99A PE=1 SV=1 0.164 0.197 0.270 ATE1_DROME Arginyl-tRNA--protein transferase 1 OS=Drosophila melanogaster GN=Ate1 PE=2 SV=3 0.047 0.052 0.072 CUL1_DROME Cullin homolog 1 OS=Drosophila melanogaster GN=lin19 PE=1 SV=2 0.034 0.039 0.054 BGBP2_DROME Gram-negative bacteria-binding protein 2 OS=Drosophila melanogaster GN=GNBP2 PE=2 SV=3 0.034 0.040 0.054 Q9VB05_DROME CG12876-PA (LD25543p) OS=Drosophila melanogaster GN=ALiX PE=1 SV=1 0.222 0.288 0.504 PREP_DROME Presequence protease, mitochondrial OS=Drosophila melanogaster GN=CG3107 PE=1 SV=2 0.199 0.274 0.441 A4V2J2_DROME (+3) CG2512-PB, isoform B OS=Drosophila melanogaster GN=alphaT ub84D PE=3 SV=1 0.799 1.311 2.006 XPO2_DROME Exportin-2 OS=Drosophila melanogaster GN=Cas PE=1 SV=2 0.234 0.354 0.540 A4UZW2_DROME (+1) S-adenosylmethionine synthetase OS=Drosophila melanogaster GN=M(2)21AB PE=3 SV=1 0.162 0.248 0.405 PEP_DROME Zinc finger protein on ecdysone puffs OS=Drosophila melanogaster GN=Pep PE=1 SV=1 0.163 0.249 0.396 A4UZZ4_DROME (+1) Phosphorylase OS=Drosophila melanogaster GN=GlyP PE=3 SV=1 0.034 0.052 0.081 Q8MRF8_DROME SD07726p OS=Drosophila melanogaster GN=CG3229 PE=2 SV=1 0.106 0.223 0.360 Q6NP53_DROME (+1) SD02276p (Fragment) OS=Drosophila melanogaster GN=BEST:CK02318 PE=2 SV=1 0.164 0.301 0.450 Q8IQW7_DROME (+2) CG3917-PA, isoform A OS=Drosophila melanogaster GN=Grip84 PE=2 SV=1 0.045 0.091 0.135 Q7KUD4_DROME (+1) CG6718-PB, isoform B (CG6718-PC, isoform C) (CG6718-PD, isoform D) OS=Drosophila melanogaster GN=CG6718 PE=1 S V=1 0.151 0.236 0.405 Q9VWI2_DROME CG12202-PA (SD09860p) OS=Drosophila melanogaster GN=Nat1 PE=1 SV=1 0.222 0.340 0.639 Q8MS29_DROME (+1) RE32163p OS=Drosophila melanogaster GN=CG2774 PE=2 SV=1 0.024 0.040 0.072 A1Z7P0_DROME (+4) CG8073-PB, isoform B OS=Drosophila melanogaster GN=Pmm45A PE=1 SV=1 0.023 0.040 0.072 DDX6_DROME Putative ATP-dependent RNA helicase me31b OS=Drosophila melanogaster GN=me31B PE=1 SV=3 0.127 0.196 0.216 TBB1_DROME Tubulin beta-1 chain OS=Drosophila melanogaster GN=betaTub56D PE=1 SV=2 1.016 1.388 1.674 Q8MSW0_DROME LD27166p (CG11471-PC, isoform C) (CG11471-PD, isoform D) (CG11471-PA, isoform A) OS=Drosophila melanogaster GN=Aats-ile PE=1 SV=1 0.549 0.734 0.917 Q9V3Z4_DROME CG1100-PA (GH11341p) (Hypothetical 55kDa protein) OS=Drosophila melanogaster GN=Rpn5 PE=1 SV=1 0.234 0.341 0.414 A9UNC0_DROME (+2) RE03215p OS=Drosophila melanogaster GN=CG1597 PE=2 SV=1 0.045 0.065 0.081 2ABA_DROME (+2) Protein phosphatase PP2A 55 kDa regulatory subunit OS=Drosophila melanogaster GN=tws PE=1 SV=1 0.045 0.065 0.081 Q8IGT0_DROME (+2) RE35250p OS=Drosophila melanogaster GN=sec23 PE=2 SV=1 0.129 0.184 0.216 PSMD1_DROME 26S proteasome non-ATPase regulatory subunit 1 OS=Drosophila melanogaster GN=Rpn2 PE=1 SV=1 0.245 0.380 0.477 Q9VJD1_DROME CG6453-PA (LD46533p) OS=Drosophila melanogaster GN=CG6453 PE=1 SV=1 0.115 0.171 0.216 SYQ_DROME Probable glutaminyl-tRNA synthetase OS=Drosophila melanogaster GN=Aats-gln PE=1 SV=1 0.174 0.275 0.324 Q9V415_DROME CG10221-PA (BcDNA.LD23587) OS=Drosophila melanogaster GN=BcDNA.LD23587 PE=2 SV=1 0.058 0.092 0.108 ATPA_DROME ATP synthase subunit alpha, mitochondrial OS=Drosophila melanogaster GN=blw PE=1 SV=2 1.338 1.782 2.311 Q9VM14_DROME (+1) CG5261-PB, isoform B (AT21758p) OS=Drosophila melanogaster GN=CG5261 PE=1 SV=1 0.231 0.302 0.405 Q7K0E6_DROME GM14334p (CG3821-PA) OS=Drosophila melanogaster GN=Aats-asp PE=2 SV=1 0.233 0.327 0.431 ATNA_DROME Sodium/potassium-transporting ATPase subunit alpha OS=Drosophila melanogaster GN=Atpalpha PE=1 SV=3 0.306 0.459 0.612 A4V4I8_DROME (+1) CG18102-PF, isoform F (CG18102-PG, isoform G) OS=Drosophila melanogaster GN=shi PE=3 SV=1 0.116 0.171 0.234 Q9VBP6_DROME CG4685-PA, isoform A (CG4685-PB, isoform B) (CG4685-PC, isoform C) (CG4685-PD, isoform D) (GH21316p) OS=Drosophila melanogaster GN=CG4685 PE=2 SV=1 0.094 0.131 0.189 Q9W3M7_DROME CG10777-PB (LD32873p) OS=Drosophila melanogaster GN=CG10777 PE=1 SV=1 0.024 0.040 0.054 Q0E9G4_DROME (+1) CG1600-PA, isoform A OS=Drosophila melanogaster GN=CG1600 PE=1 SV=1 0.024 0.039 0.054 Q9VTU4_DROME CG5642-PA (RE21692p) OS=Drosophila melanogaster GN=CG5642 PE=1 SV=1 0.128 0.249 0.287 O76935_DROME (+2) Iron regulatory protein-1B OS=Drosophila melanogaster GN=Irp-1B PE=1 SV=1 0.094 0.183 0.216 A4V0B8_DROME (+1) CG10377-PB, isoform B (CG10377-PC, isoform C) OS=Drosophila melanogaster GN=Hrb27C PE=4 SV=1 0.070 0.157 0.153 Q7KVD1_DROME (+1) CG18214-PA, isoform A (CG18214-PC, isoform C) OS=Drosophila melanogaster GN=trio PE=1 SV=1 0.036 0.079 0.081 Q7YWB4_DROME (+1) Aminotransferase OS=Drosophila melanogaster GN=Got1 PE=2 SV=1 0.045 0.104 0.108 Q8IR23_DROME CG9209-PA, isoform A (CG9209-PC, isoform C) (CG9209-PD, isoform D) (RasGap protein) OS=Drosophila melanogaster GN=v ap PE=2 SV=2 13.642 10.607 2.930 A8JNU2_DROME (+1) CG4032-PB, isoform B OS=Drosophila melanogaster GN=Abl PE=3 SV=1 0.292 0.236 0.081 A1ZAB5_DROME CG8443-PA OS=Drosophila melanogaster GN=CG8443 PE=1 SV=1 0.268 0.236 0.072 ACT1_DROME (+2) Actin-5C OS=Drosophila melanogaster GN=Act5C PE=1 SV=4 0.269 0.274 0.081 RUVB2_DROME (+1) RuvB-like helicase 2 OS=Drosophila melanogaster GN=rept PE=1 SV=1 0.199 0.196 0.054 Q9W457_DROME Serine hydroxymethyltransferase OS=Drosophila melanogaster GN=CG301 1 PE=1 SV=1 0.246 0.183 0.081 Q7KUB0_DROME (+3) CG7176-PA, isoform A (CG7176-PE, isoform E) (CG7176-PF, isoform F) OS=Drosophila melanogaster GN=Idh PE=2 SV=1 0.210 0.157 0.081 Q5U0Y0_DROME LD13852p (CG8815-PB, isoform B) OS=Drosophila melanogaster GN=Sin3A PE=1 SV=1 0.152 0.104 0.054 Q23984_DROME Tiggrin OS=Drosophila melanogaster GN=Tig PE=2 SV=1 0.151 0.105 0.054 A4UZY9_DROME (+4) Enolase OS=Drosophila melanogaster GN=Eno PE=3 SV=1 0.174 0.157 0.081 Q9VKG8_DROME CG6509-PA, isoform A (CG6509-PB, isoform B) (LD32687p) OS=Drosophila melanogaster GN=CG6509 PE=1 SV=1 0.117 0.091 0.054 U520_DROME Putative U5 small nuclear ribonucleoprotein 200 kDa helicase OS=Drosophila melanogaster GN=CG5931 PE=2 SV=4 0.199 0.237 0.081 A8WHK9_DROME (+1) RE31579p OS=Drosophila melanogaster PE=2 SV=1 0.188 0.223 0.081 Q9VN44_DROME CG1059-PA OS=Drosophila melanogaster GN=Karybeta3 PE=2 SV=1 0.282 0.315 0.108 Q9W0S7_DROME CG7008-PA (LD20211p) OS=Drosophila melanogaster GN=Tudor-SN PE=2 SV=1 0.131 0.143 0.054 Q8IRQ5_DROME (+2) CG4094-PB, isoform B OS=Drosophila melanogaster GN=l(1)G0255 PE=2 SV=1 0.129 0.144 0.054 AGO2_DROME Protein argonaute-2 OS=Drosophila melanogaster GN=AGO2 PE=1 SV=3 0.292 0.315 0.153 Q9W179_DROME CG4527-PB, isoform B (CG4527-PC, isoform C) (CG4527-PD, isoform D) OS=Drosophila melanogaster GN=slik PE=1 SV=2 0.117 0.118 0.054 Q7KQM6_DROME (+1) CG11148-PB, isoform B (CG11148-PC, isoform C) OS=Drosophila melanogaster GN=CG11148 PE=1 SV=1 0.116 0.118 0.054 Q9VK45_DROME CG5092-PA OS=Drosophila melanogaster GN=Tor PE=2 SV=1 0.128 0.170 0.054 A8DY80_DROME (+2) CG34407-PD, isoform D OS=Drosophila melanogaster GN=Not1 PE=4 SV=1 0.340 0.341 0.180 Q7KIF8_DROME (+2) Ran binding protein 7 OS=Drosophila melanogaster GN=msk PE=2 SV=1 0.152 0.145 0.081 A4V4A1_DROME (+1) CG1453-PB, isoform B (CG1453-PC, isoform C) (CG1453-PD, isoform D) (CG1453-PE, isoform E) OS=Drosophila melano gaster GN=Klp10A PE=3 SV=1 0.140 0.131 0.081 Q7KMQ0_DROME 26S proteasome regulatory complex subunit p48B (CG1341-PA) (SD07148p) OS=Drosophila melanogaster GN=Rpt1 PE=2 SV=1 0.292 0.301 0.189 VINC_DROME Vinculin OS=Drosophila melanogaster GN=Vinc PE=1 SV=1 0.164 0.170 0.108 A1ZA78_DROME (+2) CG8322-PA, isoform A OS=Drosophila melanogaster GN=ATPCL PE=1 SV=1 0.176 0.184 0.108 CHDM_DROME (+1) Chromodomain-helicase-DNA-binding protein Mi-2 homolog OS=Drosophila melanogaster GN=Mi-2 PE=1 SV=2 0.132 0.132 0.081 Q9VKZ7_DROME CG4972-PA OS=Drosophila melanogaster GN=CG4972 PE=1 SV=2 0.116 0.118 0.072 A4V491_DROME (+1) CG16944-PD, isoform D OS=Drosophila melanogaster GN=sesB PE=3 SV=1 0.783 0.890 0.468 Q8MS15_DROME (+1) RE62284p OS=Drosophila melanogaster GN=Ced-12 PE=2 SV=1 0.094 0.105 0.054 PSMD4_DROME (+1) 26S proteasome non-ATPase regulatory subunit 4 OS=Drosophila melanogaster GN=Pros54 PE=1 SV=2 0.082 0.092 0.054 A4V2M8_DROME (+1) CG6203-PD, isoform D (LD13401p) OS=Drosophila melanogaster GN=Fmr1 PE=2 SV=1 0.487 0.564 0.487 LARP_DROME La-related protein OS=Drosophila melanogaster GN=larp PE=1 SV=4 0.315 0.367 0.324 OST48_DROME Dolichyl-diphosphooligosaccharide--protein glycosyltransferase 48 kDa subunit OS=Drosophila melanogaster GN=Ost48 PE =2 SV=2 0.269 0.301 0.261 Q7K3M5_DROME GM13272p (CG11107-PA) OS=Drosophila melanogaster GN=CG11107 PE=2 SV=1 0.186 0.222 0.180 Q8IGC5_DROME (+3) RH56418p OS=Drosophila melanogaster GN=Aats-thr PE=2 SV=1 0.223 0.236 0.189 O76522_DROME (+2) Karyopherin alpha 3 (Fragment) OS=Drosophila melanogaster GN=Kap-alpha3 PE=1 SV=1 0.163 0.170 0.144 PRS4_DROME (+1) 26S protease regulatory subunit 4 OS=Drosophila melanogaster GN=Pros26.4 PE=1 SV=2 0.256 0.249 0.243 Q9VMV5_DROME CG16858-PA OS=Drosophila melanogaster GN=vkg PE=2 SV=1 0.198 0.196 0.189 A1Z9U3_DROME (+1) CG10149-PA, isoform A OS=Drosophila melanogaster GN=Rpn6 PE=1 SV=1 0.186 0.183 0.162 B3DNL7_DROME (+1) LD13657p OS=Drosophila melanogaster GN=CG10293-RA PE=2 SV=1 0.140 0.157 0.108 Q29IU3_DROPS (+1) GA14216-PA (Fragment) OS=Drosophila pseudoobscura GN=GA14216 PE=3 SV=1 0.129 0.144 0.108 A8DZ29_DROME (+1) CG10811-PB, isoform B OS=Drosophila melanogaster GN=eIF-4G PE=4 SV=1 0.315 0.393 0.243 TBG1_DROME Tubulin gamma-1 chain OS=Drosophila melanogaster GN=gammaT ub23C PE=1 SV=2 0.104 0.131 0.081 O61540_DROME (+1) Methionine aminopeptidase OS=Drosophila melanogaster GN=und PE=2 SV=1 0.094 0.118 0.072 Q9VIE8_DROME CG9244-PB, isoform B (LD24561p) (Mitochondrial aconitase) OS=Drosophila melanogaster GN=Acon PE=2 SV=2 0.152 0.196 0.144 Q9VHL2_DROME CG8351-PA (LD47396p) OS=Drosophila melanogaster GN=Tcp-1eta PE=2 SV=2 0.152 0.210 0.144 Q9VAW3_DROME CG1345-PA (GH12731p) OS=Drosophila melanogaster GN=Gfat2 PE=2 SV=1 0.084 0.106 0.072 ELP1_DROME Putative elongator complex protein 1 OS=Drosophila melanogaster GN=CG10535 PE=1 SV=2 0.082 0.105 0.072 Q9VKK1_DROME CG6181-PA, isoform A (CG6181-PB, isoform B) (LD41624p) OS=Drosophila melanogaster GN=CG6181 PE=1 SV=2 0.256 0.209 0.234 CISY_DROME Probable citrate synthase, mitochondrial OS=Drosophila melanogaster GN=kdn PE=2 SV=1 0.129 0.105 0.108 Q9VYV3_DROME CG1837-PA (LD24756p) OS=Drosophila melanogaster GN=CG1837 PE=2 SV=2 0.280 0.196 0.216 GNAS_DROME (+1) Guanine nucleotide-binding protein G(s) subunit alpha OS=Drosophila melanogaster GN=G-salpha60A PE=2 SV=1 0.071 0.052 0.054 DPP3_DROME (+1) Dipeptidyl-peptidase 3 OS=Drosophila melanogaster GN=DppIII PE=2 SV=2 0.339 0.288 0.251 Q29KX9_DROPS (+2) GA16678-PA (Fragment) OS=Drosophila pseudoobscura GN=GA16678 PE=3 SV=1 0.094 0.079 0.072 Q9VKW5_DROME CG5355-PA (LP07359p) OS=Drosophila melanogaster GN=CG5355 PE=1 SV=2 0.094 0.079 0.072 A1Z8U0_DROME (+1) CG8877-PA OS=Drosophila melanogaster GN=prp8 PE=4 SV=1 0.279 0.263 0.189 A1ZA48_DROME CG30084-PA, isoform A OS=Drosophila melanogaster GN=CG30084 PE=4 SV=1 0.162 0.157 0.108 Q9XYU0_DROME CG4978-PA (RE04406p) (DNA replication factor MCM7) OS=Drosophila melanogaster GN=Mcm7 PE=2 SV=1 0.106 0.105 0.072 Q8T9C5_DROME SD07655p OS=Drosophila melanogaster GN=MRP PE=2 SV=1 0.225 0.197 0.144 Q9VFV9_DROME CG8863-PA, isoform A (CG8863-PB, isoform B) (CG8863-PC, isoform C) (CG8863-PD, isoform D) (CG8863-PE, isoform E) (G M13664p) OS=Drosophila melanogaster GN=CG8863 PE=2 SV=1 0.116 0.105 0.081 CP9C1_DROME Cytochrome P450 9c1 OS=Drosophila melanogaster GN=Cyp9c1 PE=2 SV=1 0.106 0.092 0.072

207

Q9VAH7_DROME CG1973-PA (RE62393p) OS=Drosophila melanogaster GN=CG1973-RA PE=2 SV=1 0.129 0.105 0.081 Q6AWP5_DROME RE65032p (Fragment) OS=Drosophila melanogaster PE=2 SV=1 0.116 0.092 0.072 Q7K012_DROME LD29458p (CG6546-PA) OS=Drosophila melanogaster GN=Bap55 PE=2 SV=1 0.094 0.066 0.054 Q9VSL8_DROME CG6831-PA (Talin) OS=Drosophila melanogaster GN=rhea PE=2 SV=1 0.332 0.065 0.234 Q9VNI8_DROME CG2031-PA (Hpr1 protein) (LD43883p) OS=Drosophila melanogaster GN=Hpr1 PE=1 SV=1 0.152 0.026 0.081 AP2A_DROME (+1) AP-2 complex subunit alpha OS=Drosophila melanogaster GN=alpha-Adaptin PE=1 SV=1 0.594 0.367 0.504 Q9VUK8_DROME CG6778-PB, isoform B OS=Drosophila melanogaster GN=Aats-gly PE=2 SV=1 0.224 0.132 0.180 Q8IQV9_DROME (+1) CG11943-PB, isoform B OS=Drosophila melanogaster GN=CG11943 PE=2 SV=1 0.199 0.131 0.189 A4V2X5_DROME (+1) CG6904-PC, isoform C OS=Drosophila melanogaster GN=CG6904 PE=4 SV=1 0.200 0.131 0.180 MYSN_DROME Myosin heavy chain, non-muscle OS=Drosophila melanogaster GN=zip PE=1 SV=2 0.562 0.327 0.513 Q24506_DROME (+1) Sterol carrier protein x OS=Drosophila melanogaster GN=ScpX PE=2 SV=1 0.082 0.052 0.081 Q24253_DROME CG12532-PA (LP17054p) (Beta-adaptin Drosophila 1) OS=Drosophila melanogaster GN=Bap PE=1 SV=1 0.270 0.157 0.189 A5XCG6_DROME (+2) Phosphoglycerate kinase (Fragment) OS=Drosophila melanogaster GN=Pgk PE=3 SV=1 0.152 0.104 0.108 SSRP1_DROME (+18) FACT complex subunit Ssrp1 OS=Drosophila melanogaster GN=Ssrp PE=1 SV=2 0.082 0.052 0.054 Q7KN55_DROME (+2) Adenylyl cyclase-associated protein (Fragment) OS=Drosophila melanogaster GN=capt PE=2 SV=1 0.186 0.105 0.108 Q7KLW9_DROME BcDNA.LD02793 (CG5519-PA) OS=Drosophila melanogaster GN=Gbp PE=2 SV=1 0.164 0.079 0.108 Q9VZU7_DROME Ubiquitin carboxyl-terminal hydrolase OS=Drosophila melanogaster GN=CG12082 PE=1 SV=1 0.082 0.040 0.054 Q9VBU7_DROME CG11856-PA OS=Drosophila melanogaster GN=Nup358 PE=1 SV=2 0.235 0.131 0.072 KINH_DROME Kinesin heavy chain OS=Drosophila melanogaster GN=Khc PE=1 SV=2 0.164 0.092 0.072 O62530_DROME (+4) CG7057-PA, isoform A (CG7057-PB, isoform B) (SD05403p) (Clathrin-associated protein) (Clathrin-associated adap tor complex AP-2 medium chain) OS=Drosophila melanogaster GN=AP-50 PE=1 SV=1 0.210 0.065 0.081 B3DN78_DROME (+1) SD07123p OS=Drosophila melanogaster GN=CG10236-RA PE=2 SV=1 0.128 0.039 0.054 A1Z7H2_DROME (+3) CG8732-PD, isoform D OS=Drosophila melanogaster GN=l(2)44DEa PE=3 SV=1 0.140 0.053 0.072 Q9VGT3_DROME CG18578-PA (RE18708p) (GM04645p) OS=Drosophila melanogaster GN=Ugt86Da PE=1 SV=1 0.093 0.039 0.054 A1 Q29PI0_DROPS (+1) GA18172-PA (Fragment) OS=Drosophila pseudoobscura GN=GA18172 PE=4 SV=1 0.174 0.091 0.000 A1 CP190_DROME Centrosome-associated zinc finger protein CP190 OS=Drosophila melanogaster GN=Cp190 PE=1 SV=2 0.129 0.065 0.000 A1 LAMC1_DROME Laminin subunit gamma-1 OS=Drosophila melanogaster GN=LanB2 PE=1 SV=2 0.080 0.039 0.000 A1 Q8SX89_DROME LD09231p (CG5175-PA, isoform A) OS=Drosophila melanogaster GN=kuk PE=1 SV=1 0.163 0.079 0.000 A1 O96681_DROME (+1) Plexin A OS=Drosophila melanogaster GN=plexA PE=1 SV=1 0.082 0.040 0.000 A1 Q6NL44_DROME GH28815p (CG33722-PB, isoform B) OS=Drosophila melanogaster GN=CG18749 PE=1 SV=1 0.082 0.040 0.000 A1 Q9VVA7_DROME CG9712-PA (Tumor suppressor protein 101) OS=Drosophila melanogaster GN=TSG101 PE=1 SV=2 0.082 0.039 0.000 A1 P91944_DROME (+1) Elongation factor 1 alpha-like factor OS=Drosophila melanogaster GN=Elf PE=2 SV=1 0.116 0.053 0.000 A1 A8JV18_DROME (+1) CG4453-PB, isoform B OS=Drosophila melanogaster GN=Nup153 PE=4 SV=1 0.118 0.052 0.000 A1 DCTN1_DROME (+1) Dynactin subunit 1 OS=Drosophila melanogaster GN=Gl PE=1 SV=2 0.058 0.026 0.000 A1 Q9VJ61_DROME CG10376-PA (SD03870p) OS=Drosophila melanogaster GN=CG10376 PE=2 SV=1 0.093 0.039 0.000 A1 Q24241_DROME (+1) Ankyrin OS=Drosophila melanogaster GN=Ank PE=2 SV=1 0.093 0.039 0.000 A1 SAS6_DROME Spindle assembly abnormal protein 6 homolog OS=Drosophila melanogaster GN=sas-6 PE=1 SV=2 0.105 0.040 0.000 A1 Q29I31_DROPS (+8) GA14656-PA (Fragment) OS=Drosophila pseudoobscura GN=GA14656 PE=4 SV=1 0.070 0.026 0.000 A1 Q9VP46_DROME CG7324-PA (GH16847p) OS=Drosophila melanogaster GN=CG7324 PE=1 SV=1 0.070 0.026 0.000 A1 DIP2_DROME Disco-interacting protein 2 OS=Drosophila melanogaster GN=DIP2 PE=1 SV=2 0.153 0.105 0.000 A1 O76911_DROME (+2) FBgn0000928;fs(1)Yb protein (Eg:95b7.8 protein) OS=Drosophila melanogaster GN=fs(1)Yb PE=2 SV=1 0.176 0.118 0.000 A1 Q7KTI7_DROME (+7) CG8222-PF, isoform F OS=Drosophila melanogaster GN=Pvr PE=1 SV=1 0.118 0.079 0.000 A1 Q29F37_DROPS (+1) GA10050-PA (Fragment) OS=Drosophila pseudoobscura GN=GA10050 PE=4 SV=1 0.059 0.040 0.000 A1 HOOK_DROME Protein hook OS=Drosophila melanogaster GN=hk PE=1 SV=2 0.059 0.040 0.000 A1 ZW10_DROME Centromere/kinetochore protein zw10 OS=Drosophila melanogaster GN=mit(1)15 PE=1 SV=2 0.093 0.065 0.000 A1 Q29JF8_DROPS (+1) GA13115-PA (Fragment) OS=Drosophila pseudoobscura GN=GA13115 PE=4 SV=1 0.057 0.040 0.000 A1 A1Z6Z3_DROME (+4) CG11140-PE, isoform E OS=Drosophila melanogaster GN=Aldh-III PE=3 SV=1 0.082 0.053 0.000 A1 Q9VGQ1_DROME CG5214-PA (LP03989p) (GM01350p) OS=Drosophila melanogaster GN=CG5214 PE=2 SV=1 0.080 0.052 0.000 A1 POE_DROME Protein purity of essence OS=Drosophila melanogaster GN=poe PE=1 SV=1 0.060 0.040 0.000 A1 Q4V527_DROME IP13307p (Fragment) OS=Drosophila melanogaster GN=CG15415 PE=2 SV=1 2.395 1.424 0.000 A1 Q95YG3_DROME (+10) Double-strand-specific ribonuclease OS=Drosophila melanogaster GN=Dcr-2 PE=2 SV=1 0.174 0.105 0.000 A1 Q0E8E8_DROME (+1) CG4994-PA, isoform A OS=Drosophila melanogaster GN=Mpcp PE=3 SV=1 0.188 0.117 0.000 A1 A8WHI7_DROME (+2) LD30939p OS=Drosophila melanogaster GN=CG31756 PE=2 SV=1 0.105 0.065 0.000 A1 Q9VW19_DROME CG9372-PA (RE17417p) OS=Drosophila melanogaster GN=CG9372 PE=2 SV=1 0.106 0.065 0.000 A1 Q8MMD2_DROME (+2) CG16932-PB, isoform B OS=Drosophila melanogaster GN=Eps-15 PE=1 SV=1 0.094 0.053 0.000 A1 Q0E9F9_DROME (+1) CG2915-PA, isoform A OS=Drosophila melanogaster GN=CG2915 PE=4 SV=1 0.070 0.039 0.000 A1 Q9V400_DROME CG10504-PA (LD24671p) (Putative integrin-linked kinase) OS=Drosophila melanogaster GN=Ilk PE=2 SV=1 0.093 0.052 0.000 A1 SRC42_DROME Tyrosine-protein kinase Src42A OS=Drosophila melanogaster GN=Src42A PE=1 SV=1 0.093 0.052 0.000 A1 LASP1_DROME LIM and SH3 domain protein Lasp OS=Drosophila melanogaster GN=Lasp PE=1 SV=2 0.093 0.052 0.000 A1 Q7JR80_DROME SD23764p (CG3884-PA, isoform A) OS=Drosophila melanogaster GN=CG3884 PE=2 SV=1 0.047 0.026 0.000 A1 Q9VQI8_DROME CG3059-PB, isoform B OS=Drosophila melanogaster GN=NTPase PE=2 SV=1 0.068 0.039 0.000 A1 Q9W0L7_DROME CG32479-PA (LD28815p) OS=Drosophila melanogaster GN=CG13903 PE=1 SV=2 0.045 0.026 0.000 A1 A0ZWP1_DROME (+4) Thiolester containing protein II, isoform A (Fragment) OS=Drosophila melanogaster GN=TepII PE=1 SV=1 0.094 0.052 0.000 A1 IMB_DROME Importin subunit beta OS=Drosophila melanogaster GN=Fs(2)Ket PE=1 SV=2 0.070 0.039 0.000 A1 A1Z814_DROME Oxysterol-binding protein OS=Drosophila melanogaster GN=CG1513 PE=1 SV=2 0.047 0.026 0.000 A1 Q9VB10_DROME CG5590-PA (GH01709p) OS=Drosophila melanogaster GN=CG5590 PE=1 SV=1 0.071 0.039 0.000 A1 HEM_DROME (+1) Membrane-associated protein Hem OS=Drosophila melanogaster GN=Hem PE=2 SV=1 0.048 0.026 0.000 A1 Q7PLI7_DROME (+1) CG17528-PC, isoform C (CG17528-PD, isoform D) OS=Drosophila melanogaster GN=CG17528 PE=1 SV=1 0.153 0.118 0.000 A1 Q8SZM1_DROME (+1) RH04336p OS=Drosophila melanogaster GN=CG9953 PE=2 SV=1 0.034 0.026 0.000 A1 A4V480_DROME (+3) CG32684-PC, isoform C (CG32684-PD, isoform D) (CG32684-PE, isoform E) (CG32684-PF , isoform F) (CG32684-PG, isoform G) OS=Drosophila melanogaster GN=alpha-Man-I PE=4 SV=1 0.034 0.026 0.000 A1 Q9VQ76_DROME CG31671-PA (LD36155p) (Tho2 protein) OS=Drosophila melanogaster GN=tho2 PE=1 SV=2 0.034 0.026 0.000 A1 Q9VTZ0_DROME CG10686-PA (GH08269p) OS=Drosophila melanogaster GN=tral PE=1 SV=3 0.070 0.053 0.000 A1 NCPR_DROME (+1) NADPH--cytochrome P450 reductase OS=Drosophila melanogaster GN=Cpr PE=2 SV=2 0.070 0.053 0.000 A1 Q86NK7_DROME (+2) LD24482p OS=Drosophila melanogaster GN=Dref PE=2 SV=1 0.070 0.052 0.000 A1 Q8INH9_DROME (+1) CG7518-PB, isoform B OS=Drosophila melanogaster GN=CG7518 PE=1 SV=1 0.107 0.079 0.000 A1 Q29PD8_DROPS (+1) GA21865-PA (Fragment) OS=Drosophila pseudoobscura GN=GA21865 PE=3 SV=1 0.036 0.026 0.000 A1 Q9W196_DROME CG3356-PA (LP03102p) OS=Drosophila melanogaster GN=CG3356 PE=1 SV=1 0.036 0.026 0.000 A1 Q9VYV4_DROME CG2446-PA, isoform A (CG2446-PB, isoform B) (CG2446-PC, isoform C) (CG2446-PD, isoform D) (CG2446-PE, isoform E) (GH02702p) OS=Drosophila melanogaster GN=CG2446 PE=1 SV=1 0.036 0.026 0.000 A1 A8JNN1_DROME (+3) CG34373-PD, isoform D OS=Drosophila melanogaster GN=Ect4 PE=4 SV=1 0.186 0.157 0.000 A1 O46173_DROME (+1) Y-box protein OS=Drosophila melanogaster GN=yps PE=1 SV=1 0.093 0.078 0.000 A1 Q9VMV9_DROME CG33113-PF, isoform F OS=Drosophila melanogaster GN=Rtnl1 PE=1 SV=2 0.047 0.040 0.000 A1 Q8T3P8_DROME (+1) AT23778p OS=Drosophila melanogaster GN=Srp54k PE=2 SV=1 0.045 0.039 0.000 A1 Q7KSN8_DROME (+2) CG31368-PB, isoform B OS=Drosophila melanogaster GN=CG31368 PE=1 SV=1 0.045 0.039 0.000 A1 A9UNA2_DROME (+4) LD37353p OS=Drosophila melanogaster GN=yin PE=2 SV=1 0.045 0.039 0.000 A1 Q6NR46_DROME RE58623p (CG4016-PC, isoform C) (CG4016-PA, isoform A) OS=Drosophila melanogaster GN=Spt-I PE=2 SV=1 0.045 0.039 0.000 A1 Q9VRX7_DROME CG8368-PA, isoform A (CG8368-PB, isoform B) (LD29573p) OS=Drosophila melanogaster GN=CG8368 PE=1 SV=1 0.045 0.039 0.000 A1 O44334_DROME (+1) Diacylglycerol kinase epsilon OS=Drosophila melanogaster GN=Dgkepsilon PE=2 SV=1 0.047 0.039 0.000 A1 ARP3_DROME Actin-related protein 3 OS=Drosophila melanogaster GN=Arp66B PE=1 SV=3 0.047 0.039 0.000 A1 Q5BHY9_DROME (+1) GH10846p OS=Drosophila melanogaster GN=Srp72 PE=1 SV=1 0.048 0.040 0.000 B Q95NR4_DROME (+1) SD10676p (Bicoid mRNA stability factor) OS=Drosophila melanogaster GN=bsf PE=2 SV=1 0.257 0.040 0.000 B HYD_DROME E3 ubiquitin-protein ligase hyd OS=Drosophila melanogaster GN=hyd PE=1 SV=3 0.116 0.026 0.000 B NUP88_DROME Nuclear pore complex protein Nup88 OS=Drosophila melanogaster GN=mbo PE=1 SV=2 0.130 0.026 0.000 B Q9GUB1_DROME (+1) Phospholipase A2 activating protein homolog OS=Drosophila melanogaster GN=Plap PE=2 SV=1 0.094 0.026 0.000 B Q7K3D4_DROME LD36412p (CG5482-PA) OS=Drosophila melanogaster GN=CG5482 PE=1 SV=1 0.082 0.026 0.000 B Q24080_DROME (+3) DShc OS=Drosophila melanogaster GN=Shc PE=2 SV=1 0.082 0.026 0.000 B A8WHL4_DROME (+2) RE48191p OS=Drosophila melanogaster GN=bon PE=2 SV=1 0.093 0.000 0.000 B C1GLT_DROME Glycoprotein-N-acetylgalactosamine 3-beta-galactosyltransferase 1 OS=Drosophila melanogaster GN=CG9520 PE=2 SV=1 0.140 0.000 0.000 B Q8IRT4_DROME CG32789-PA (CG2947-PA, isoform A) (LD46530p) OS=Drosophila melanogaster GN=CG32789 PE=2 SV=2 0.058 0.000 0.000 B O16865_DROME (+1) B4 protein OS=Drosophila melanogaster GN=B4 PE=1 SV=1 0.151 0.000 0.000 B MCM2_DROME DNA replication licensing factor Mcm2 OS=Drosophila melanogaster GN=Mcm2 PE=1 SV=1 0.048 0.000 0.000 B Q6NKN1_DROME (+2) RE66582p OS=Drosophila melanogaster PE=2 SV=1 0.047 0.000 0.000 B ELP4_DROME Putative elongator complex protein 4 OS=Drosophila melanogaster GN=CG6907 PE=1 SV=1 0.047 0.000 0.000 B Q8IQF8_DROME (+1) CG6084-PB, isoform B (AT18092p) OS=Drosophila melanogaster GN=CG6084 PE=2 SV=1 0.024 0.000 0.000 B A4V3F6_DROME (+1) CG8384-PC, isoform C (CG8384-PD, isoform D) OS=Drosophila melanogaster GN=gro PE=4 SV=1 0.070 0.000 0.000 B Q0KI94_DROME (+9) CG16765-PI, isoform I OS=Drosophila melanogaster GN=ps PE=4 SV=1 0.047 0.000 0.000 B DOP1_DROME Protein dopey-1 homolog OS=Drosophila melanogaster GN=CG15099 PE=1 SV=1 0.060 0.000 0.000 B Q7JXU4_DROME SD10213p (CG6501-PA) (Nuclear GTP binding protein) OS=Drosophila melanogaster GN=Ngp PE=1 SV=1 0.023 0.000 0.000 B UBR1_DROME E3 ubiquitin-protein ligase UBR1 OS=Drosophila melanogaster GN=UBR1 PE=2 SV=2 0.023 0.000 0.000 B Q24150_DROME (+1) Nucleosome assembly protein NAP-1 OS=Drosophila melanogaster GN=Nap1 PE=1 SV=1 0.034 0.000 0.000 B Q290L1_DROPS GA15890-PA (Fragment) OS=Drosophila pseudoobscura GN=GA15890 PE=4 SV=1 0.023 0.000 0.000 B FAF_DROME Probable ubiquitin carboxyl-terminal hydrolase FAF OS=Drosophila melanogaster GN=faf PE=1 SV=2 0.060 0.000 0.000 B Q7K0P0_DROME (+2) LD44791p (CG3767-PA) OS=Drosophila melanogaster GN=JhI-26 PE=2 SV=1 0.060 0.000 0.000 B A8JNK1_DROME CG14998-PF, isoform F OS=Drosophila melanogaster GN=CG14998 PE=4 SV=1 0.034 0.000 0.000 B Q7K0S5_DROME LD37137p (CG5469-PD, isoform D) (CG5469-PB, isoform B) OS=Drosophila melanogaster GN=Gint3 PE=1 SV=1 0.034 0.000 0.000 B A8DYL3_DROME (+3) CG13503-PG, isoform G OS=Drosophila melanogaster GN=V rp1 PE=4 SV=1 0.023 0.000 0.000 B Q4QQB8_DROME (+2) LD32453p (Fragment) OS=Drosophila melanogaster PE=2 SV=1 0.023 0.000 0.000 B A1ZBB4_DROME (+2) CG30122-PB OS=Drosophila melanogaster GN=CG30122 PE=1 SV=1 0.023 0.000 0.000 B Q86PE7_DROME (+2) SD10041p OS=Drosophila melanogaster GN=ik2 PE=1 SV=1 0.023 0.000 0.000 B COG5_DROME (+1) Conserved oligomeric Golgi complex subunit 5 OS=Drosophila melanogaster GN=fws PE=1 SV=1 0.034 0.000 0.000 B Q9VLK2_DROME CG13096-PA OS=Drosophila melanogaster GN=CG13096 PE=1 SV=1 0.023 0.000 0.000 B O46048_DROME (+2) EG:133E12.4 protein OS=Drosophila melanogaster GN=east PE=1 SV=1 0.209 0.000 0.000 B Q8IMT3_DROME CG31436-PA (IP12392p) OS=Drosophila melanogaster GN=CG31436 PE=2 SV=1 0.023 0.000 0.000 B SICK_DROME Protein sickie OS=Drosophila melanogaster GN=sick PE=1 SV=3 0.231 0.000 0.000 B Q8MRG2_DROME (+1) SD04373p OS=Drosophila melanogaster GN=CG5530 PE=2 SV=1 0.512 0.000 0.000 B Q86NT2_DROME (+2) AT04875p (Fragment) OS=Drosophila melanogaster GN=CG6807 PE=1 SV=1 0.106 0.000 0.000 B Q0E9M3_DROME CG17800-PBB OS=Drosophila melanogaster GN=Dscam PE=4 SV=1 1.652 0.000 0.000 B A1Z6I7_DROME (+2) CG7838-PA OS=Drosophila melanogaster GN=BubR1 PE=1 SV=1 0.444 0.000 0.000 B Q0E9I5_DROME CG17800-PR, isoform R OS=Drosophila melanogaster GN=Dscam PE=4 SV=1 0.920 0.000 0.000 B CUP_DROME Protein cup OS=Drosophila melanogaster GN=cup PE=1 SV=2 0.023 0.000 0.000 B Q9VPI9_DROME CG11376-PA (LD20667p) OS=Drosophila melanogaster GN=CG11376 PE=1 SV=2 0.023 0.000 0.000 B Q8IGT5_DROME (+6) RE33426p OS=Drosophila melanogaster GN=Nop56 PE=2 SV=1 0.034 0.000 0.000 B Q9W2M0_DROME CG4030-PA (LD23155p) OS=Drosophila melanogaster GN=CG4030 PE=1 SV=2 0.034 0.000 0.000 B Q9VHX2_DROME CG3223-PA (GH08043p) OS=Drosophila melanogaster GN=CG3223 PE=2 SV=1 0.023 0.000 0.000 B Q29PP5_DROPS (+2) GA21593-PA (Fragment) OS=Drosophila pseudoobscura GN=GA21593 PE=4 SV=1 0.023 0.000 0.000 B PP4R2_DROME Serine/threonine-protein phosphatase 4 regulatory subunit 2 OS=Drosophila melanogaster GN=PPP4R2r PE=1 SV=2 0.023 0.000 0.000 B Q9VQI5_DROME CG31694-PA (LP04564p) OS=Drosophila melanogaster GN=CG3098 PE=1 SV=2 0.023 0.000 0.000 B ODR4_DROME Protein odr-4 homolog OS=Drosophila melanogaster GN=CG10616 PE=1 SV=2 0.023 0.000 0.000 B O46096_DROME (+1) EG:87B1.3 protein OS=Drosophila melanogaster GN=bcn92 PE=2 SV=1 0.023 0.000 0.000 B DPOD2_DROME DNA polymerase delta small subunit OS=Drosophila melanogaster GN=CG12018 PE=2 SV=1 0.023 0.000 0.000 B ERO1L_DROME (+1) Ero1-like protein OS=Drosophila melanogaster GN=Ero1L PE=2 SV=2 0.023 0.000 0.000 B Q9VRG8_DROME CG1486-PA, isoform A (CG1486-PB, isoform B) (GH01474p) OS=Drosophila melanogaster GN=CG1486 PE=1 SV=1 0.023 0.000 0.000 B Q6NR00_DROME (+1) LP05936p OS=Drosophila melanogaster GN=CG5604 PE=1 SV=1 0.060 0.000 0.000 B A1Z8Q2_DROME (+1) CG13185-PB, isoform B OS=Drosophila melanogaster GN=CG13185 PE=1 SV=1 0.045 0.000 0.000 B Q7K3Z3_DROME GH01724p (CG11139-PA) OS=Drosophila melanogaster GN=p47 PE=1 SV=1 0.023 0.000 0.000 B Q9VG81_DROME CG5167-PA (RH49330p) OS=Drosophila melanogaster GN=CG5167 PE=2 SV=1 0.034 0.000 0.000 B A8JRF3_DROME (+3) CG1646-PF, isoform F OS=Drosophila melanogaster GN=CG1646 PE=4 SV=1 0.034 0.000 0.000 B A1Z7A8_DROME (+1) CG8710-PD, isoform D OS=Drosophila melanogaster GN=coilin PE=4 SV=1 0.023 0.000 0.000 B Q8T8Q5_DROME (+1) SD05887p OS=Drosophila melanogaster GN=CG3493 PE=1 SV=1 0.045 0.000 0.000 B A8DZ24_DROME (+1) CG6448-PC, isoform C OS=Drosophila melanogaster GN=CG6448 PE=4 SV=1 0.034 0.000 0.000 B ERF1_DROME (+1) Eukaryotic peptide chain release factor subunit 1 OS=Drosophila melanogaster GN=eRF1 PE=1 SV=2 0.023 0.000 0.000 B EXOC1_DROME (+1) Exocyst complex component 1 OS=Drosophila melanogaster GN=sec3 PE=1 SV=2 0.045 0.000 0.000 B Q9VM04_DROME CG6630-PA OS=Drosophila melanogaster GN=CG6630 PE=2 SV=3 0.023 0.000 0.000 B A1YK77_DROME (+19) Nup133 OS=Drosophila melanogaster GN=Nup133 PE=1 SV=1 0.034 0.000 0.000 B Q9VZI2_DROME CG14992-PA (BcDNA.GH10777) OS=Drosophila melanogaster GN=Ack PE=1 SV=2 0.034 0.000 0.000 B ANLN_DROME (+1) Actin-binding protein anillin OS=Drosophila melanogaster GN=scra PE=1 SV=3 0.034 0.000 0.000 B Q960C1_DROME (+1) SD08037p OS=Drosophila melanogaster GN=CG5913 PE=2 SV=1 0.023 0.000 0.000 B Q7KTL4_DROME (+3) CG9188-PC, isoform C OS=Drosophila melanogaster GN=sip2 PE=1 SV=1 0.023 0.000 0.000 B RFC1_DROME Replication factor C subunit 1 OS=Drosophila melanogaster GN=Gnf1 PE=1 SV=2 0.034 0.000 0.000 B EXOC5_DROME (+1) Exocyst complex component 5 OS=Drosophila melanogaster GN=sec10 PE=1 SV=1 0.071 0.000 0.000 B EXOC6_DROME Exocyst complex component 6 OS=Drosophila melanogaster GN=sec15 PE=1 SV=1 0.034 0.000 0.000 B Q9VC57_DROME CG6668-PA, isoform A (CG6668-PB, isoform B) (RE51884p) (GH09383p) OS=Drosophila melanogaster GN=atl PE=1 SV=1 0.023 0.000 0.000 B O44432_DROME (+2) QKR58E-2 OS=Drosophila melanogaster GN=qkr58E-2 PE=2 SV=1 0.023 0.000 0.000 B Q9VF53_DROME CG18522-PA (LD37006p) OS=Drosophila melanogaster GN=CG18522 PE=1 SV=1 0.082 0.000 0.000 B Q7KMJ6_DROME BcDNA.GH11110 (CG2910-PA, isoform A) OS=Drosophila melanogaster GN=nito PE=1 SV=1 0.082 0.000 0.000 B C3G_DROME (+3) Guanine nucleotide-releasing factor 2 OS=Drosophila melanogaster GN=C3G PE=1 SV=3 0.070 0.000 0.000 B Q7JUP3_DROME LD03368p (CG8400-PA, isoform A) (CG8400-PB, isoform B) OS=Drosophila melanogaster GN=casp PE=1 SV=1 0.059 0.000 0.000 B Q9VI55_DROME CG1104-PA, isoform A (LD37409p) OS=Drosophila melanogaster GN=CG1104 PE=1 SV=1 0.036 0.000 0.000 B A4V384_DROME (+1) CG6375-PB, isoform B OS=Drosophila melanogaster GN=pit PE=4 SV=1 0.058 0.000 0.000 B A4IJ82_DROME (+6) LD40879p (Fragment) OS=Drosophila melanogaster GN=CG32782-RC PE=2 SV=1 0.036 0.000 0.000 B Q29MT3_DROPS (+1) GA14464-PA (Fragment) OS=Drosophila pseudoobscura GN=GA14464 PE=4 SV=1 0.024 0.000 0.000 B NFS1_DROME Probable cysteine desulfurase, mitochondrial OS=Drosophila melanogaster GN=CG12264 PE=2 SV=1 0.036 0.000 0.000 B ST38L_DROME Serine/threonine-protein kinase 38-like OS=Drosophila melanogaster GN=trc PE=1 SV=1 0.057 0.000 0.000 B TUD_DROME Maternal protein tudor OS=Drosophila melanogaster GN=tud PE=1 SV=2 0.024 0.000 0.000 B A1Z9K0_DROME (+1) CG6701-PA OS=Drosophila melanogaster GN=CG6701 PE=1 SV=1 0.081 0.000 0.000 B Q0KHR1_DROME (+2) CG8465-PA, isoform A OS=Drosophila melanogaster GN=l(1)G0222 PE=1 SV=1 0.036 0.000 0.000 B Q9VQU3_DROME CG2818-PA, isoform A (LD22655p) OS=Drosophila melanogaster GN=CG2818 PE=2 SV=3 0.048 0.000 0.000 B Q9VQ78_DROME CG7261-PA (LD16031p) OS=Drosophila melanogaster GN=CG7261 PE=2 SV=1 0.024 0.000 0.000 B Q0E993_DROME (+1) CG4062-PA, isoform A OS=Drosophila melanogaster GN=Aats-val PE=3 SV=1 0.036 0.000 0.000 B UBPE_DROME Ubiquitin carboxyl-terminal hydrolase 64E OS=Drosophila melanogaster GN=Ubp64E PE=1 SV=2 0.057 0.000 0.000 B XPO1_DROME Exportin-1 OS=Drosophila melanogaster GN=emb PE=1 SV=1 0.048 0.000 0.000 B Q9VL52_DROME CG31755-PA OS=Drosophila melanogaster GN=CG31755 PE=2 SV=2 0.024 0.000 0.000 B U493_DROME UPF0493 protein CG14299 OS=Drosophila melanogaster GN=CG14299 PE=1 SV=1 0.024 0.000 0.000 B PEN_DROME Protein penguin OS=Drosophila melanogaster GN=pen PE=2 SV=1 0.036 0.000 0.000 B COG2_DROME Conserved oligomeric Golgi complex subunit 2 OS=Drosophila melanogaster GN=ldlCp PE=2 SV=1 0.024 0.000 0.000 B SPRI_DROME Protein sprint OS=Drosophila melanogaster GN=spri PE=1 SV=2 0.559 0.000 0.000 B Q9V4B6_DROME CG31998-PA OS=Drosophila melanogaster GN=CG31998 PE=1 SV=2 0.092 0.000 0.000 B Q299L9_DROPS (+1) GA15015-PA (Fragment) OS=Drosophila pseudoobscura GN=GA15015 PE=4 SV=1 0.024 0.000 0.000 B STAT_DROME Signal transducer and transcription activator OS=Drosophila melanogaster GN=Stat92E PE=1 SV=1 0.024 0.000 0.000 B Q9VVU6_DROME (+1) CG6841-PA (LD04472p) OS=Drosophila melanogaster GN=CG6841 PE=1 SV=2 0.036 0.000 0.000 B O77426_DROME (+3) EG:115C2.2 protein (Fbgn0001341;l(1)1bi protein) OS=Drosophila melanogaster GN=l(1)1Bi PE=2 SV=1 0.036 0.000 0.000 B Q9VW20_DROME Lon protease homolog OS=Drosophila melanogaster GN=CG8798 PE=2 SV=1 0.036 0.000 0.000 B Q24284_DROME (+1) PLC-gamma D OS=Drosophila melanogaster GN=sl PE=1 SV=1 0.036 0.000 0.000 B Q8MQS4_DROME GH06923p (CG6192-PA) OS=Drosophila melanogaster GN=CG6192 PE=1 SV=1 0.024 0.000 0.000 B Q7K4H4_DROME LD40453p (CG2982-PA, isoform A) OS=Drosophila melanogaster GN=CG2982 PE=1 SV=1 0.024 0.000 0.000 B A4V0S4_DROME (+3) CG4559-PB, isoform B (CG4559-PC, isoform C) OS=Drosophila melanogaster GN=Idgf3 PE=3 SV=1 0.024 0.000 0.000 B RCOR_DROME REST corepressor OS=Drosophila melanogaster GN=CoRest PE=1 SV=1 0.024 0.000 0.000 B Q9W0H9_DROME CG9139-PA (SD03358p) OS=Drosophila melanogaster GN=CG9139 PE=2 SV=1 0.024 0.000 0.000 B Q298I5_DROPS (+4) GA16037-PA (Fragment) OS=Drosophila pseudoobscura GN=GA16037 PE=4 SV=1 0.024 0.000 0.000 B HANG_DROME (+1) Zinc finger protein hangover OS=Drosophila melanogaster GN=hang PE=1 SV=3 0.024 0.000 0.000 B Q8IGV1_DROME (+2) RE24907p (Fragment) OS=Drosophila melanogaster GN=CG6995 PE=2 SV=1 0.024 0.000 0.000 B Q9VHK3_DROME CG31349-PA, isoform A OS=Drosophila melanogaster GN=pyd PE=2 SV=3 0.048 0.000 0.000 B Q9VEX5_DROME CG6814-PA (LD33046p) OS=Drosophila melanogaster GN=Mat89Bb PE=1 SV=1 0.024 0.000 0.000 B Q7KJA9_DROME (+1) O-glycosyltransferase (CG10392-PC, isoform C) (SD06381p) (CG10392-P A, isoform A) OS=Drosophila melanogaster GN=Ogt PE=2 SV=1 0.036 0.000 0.000 B Q9VCH1_DROME CG10192-PA OS=Drosophila melanogaster GN=ofs PE=1 SV=2 0.036 0.000 0.000 A4V464_DROME (+1) CG7033-PC, isoform C OS=Drosophila melanogaster GN=CG7033 PE=3 SV=1 0.174 0.380 0.180 A1ZAX1_DROME (+1) CG4954-PA OS=Drosophila melanogaster GN=eIF3-S8 PE=1 SV=1 0.167 0.327 0.108 Q8T6I0_DROME (+1) EH domain containing protein (CG6148-PB, isoform B) OS=Drosophila melanogaster GN=Past1 PE=2 SV=1 0.175 0.262 0.144 Q9VVL7_DROME Dihydrolipoyl dehydrogenase OS=Drosophila melanogaster GN=CG7430 PE=2 SV=1 0.140 0.196 0.108 FKB59_DROME FK506-binding protein 59 OS=Drosophila melanogaster GN=FKBP59 PE=1 SV=1 0.220 0.327 0.162 UAP56_DROME ATP-dependent RNA helicase WM6 OS=Drosophila melanogaster GN=Hel25E PE=1 SV=1 0.117 0.170 0.081 Q9VQL1_DROME CG17259-PA (LP20978p) OS=Drosophila melanogaster GN=CG17259 PE=1 SV=1 0.106 0.157 0.072 Q1RKZ0_DROME (+2) IP15366p OS=Drosophila melanogaster GN=Trxr-1 PE=2 SV=1 0.104 0.170 0.081 LIG_DROME Protein lingerer OS=Drosophila melanogaster GN=lig PE=1 SV=1 0.082 0.131 0.072 IF2G_DROME Eukaryotic translation initiation factor 2 subunit 3 OS=Drosophila melanogaster GN=Su(var)3-9 PE=2 SV=1 0.172 0.315 0.216 O61650_DROME (+1) Hsp70/Hsp90 organizing protein homolog OS=Drosophila melanogaster GN=Hop PE=2 SV=1 0.140 0.249 0.180 ATPB_DROME ATP synthase subunit beta, mitochondrial OS=Drosophila melanogaster GN=A TPsyn-beta PE=1 SV=3 0.415 0.866 0.549

208

MMSA_DROPS (+1) Probable methylmalonate-semialdehyde dehydrogenase [acylating], mitochondrial OS=Drosophila pseudoobscura GN=GA1 4712 PE=3 SV=1 0.058 0.118 0.081 A4V2H1_DROME (+2) CG10279-PC, isoform C (CG10279-PF, isoform F) OS=Drosophila melanogaster GN=Rm62 PE=3 SV=1 0.246 0.393 0.287 G6PI_DROME (+1) Glucose-6-phosphate isomerase OS=Drosophila melanogaster GN=Pgi PE=2 SV=2 0.094 0.157 0.108 PRS8_DROME (+1) 26S protease regulatory subunit 8 OS=Drosophila melanogaster GN=Pros45 PE=1 SV=2 0.105 0.157 0.108 HSP7D_DROME Heat shock 70 kDa protein cognate 4 OS=Drosophila melanogaster GN=Hsc70-4 PE=1 SV=3 0.646 0.881 0.845 Q960Y8_DROME LD29525p (CG18212-PG, isoform G) OS=Drosophila melanogaster GN=alt PE=2 SV=1 0.261 0.354 0.324 Q9VK59_DROME CG5787-PA (LD23647p) OS=Drosophila melanogaster GN=CG5787 PE=1 SV=2 0.176 0.261 0.243 A1Z843_DROME (+2) CG1371-PA OS=Drosophila melanogaster GN=CG1371 PE=4 SV=1 0.176 0.250 0.198 CALR_DROME (+2) Calreticulin OS=Drosophila melanogaster GN=Crc PE=1 SV=2 0.152 0.210 0.162 Q95U38_DROME (+1) GH10480p OS=Drosophila melanogaster GN=CG11963 PE=2 SV=1 0.047 0.065 0.054 CO4A1_DROME Collagen alpha-1(IV) chain OS=Drosophila melanogaster GN=Cg25C PE=2 SV=3 0.094 0.118 0.108 Q9VAC1_DROME CG7920-PA, isoform A (GM14349p) OS=Drosophila melanogaster GN=CG7920 PE=2 SV=1 0.093 0.118 0.108 RL4_DROME 60S ribosomal protein L4 OS=Drosophila melanogaster GN=RpL4 PE=2 SV=2 0.117 0.196 0.162 2AAA_DROME Serine/threonine-protein phosphatase PP2A 65 kDa regulatory subunit OS=Drosophila melanogaster GN=Pp2A-29B PE=1 SV=4 0.106 0.171 0.144 Q6NP55_DROME (+4) SD08263p (Fragment) OS=Drosophila melanogaster GN=CG10198 PE=1 SV=1 0.057 0.104 0.081 Q9V3A8_DROME CG6822-PA, isoform A (CG6822-PB, isoform B) (LD43551p) (Rhea) OS=Drosophila melanogaster GN=ergic53 PE=1 SV=1 0.082 0.170 0.153 Q7K3J0_DROME LD24495p (CG8258-PA) OS=Drosophila melanogaster GN=CG8258 PE=2 SV=1 0.187 0.380 0.323 O61444_DROME (+2) Stress activated MAP kinase kinase 4 (RE70055p) (CG9738-PA) OS=Drosophila melanogaster GN=Mkk4 PE=1 SV=1 0.034 0.065 0.054 VATB_DROME Vacuolar ATP synthase subunit B OS=Drosophila melanogaster GN=Vha55 PE=1 SV=1 0.125 0.314 0.189 IF4A_DROME Eukaryotic initiation factor 4A OS=Drosophila melanogaster GN=eIF-4a PE=2 SV=3 0.093 0.286 0.162 EAF3_DROME NuA4 complex subunit EAF3 homolog OS=Drosophila melanogaster GN=MRG15 PE=1 SV=1 0.036 0.105 0.054 A4V303_DROME (+1) CG8977-PB, isoform B OS=Drosophila melanogaster GN=Cctgamma PE=3 SV=1 0.080 0.249 0.180 Q9VV75_DROME CG4169-PA (AT02348p) OS=Drosophila melanogaster GN=CG4169 PE=2 SV=1 0.023 0.078 0.054 Q9VG01_DROME CG12360-PA, isoform A (CG12360-PB, isoform B) (RE12073p) OS=Drosophila melanogaster GN=CG12360 PE=1 SV=1 0.036 0.105 0.072 RUVB1_DROME (+1) RuvB-like helicase 1 OS=Drosophila melanogaster GN=pont PE=1 SV=1 0.045 0.105 0.081 Q8IQX8_DROME (+1) CG6335-PB, isoform B OS=Drosophila melanogaster GN=Aats-his PE=2 SV=1 0.034 0.092 0.072 A4V391_DROME (+1) T-complex protein 1, alpha subunit OS=Drosophila melanogaster GN=T -cp1 PE=3 SV=1 0.045 0.209 0.216 Q5BI50_DROME LP02965p (CG8711-PA) OS=Drosophila melanogaster GN=cul-4 PE=2 SV=1 0.023 0.131 0.126 Q7KPY3_DROME (+2) Moira OS=Drosophila melanogaster GN=mor PE=1 SV=1 0.000 0.066 0.054 Q9VEP9_DROME CG16941-PA (GH03554p) OS=Drosophila melanogaster GN=CG16941 PE=1 SV=1 0.000 0.079 0.081 HIPPO_DROME Serine/threonine-protein kinase hippo OS=Drosophila melanogaster GN=hpo PE=1 SV=1 0.000 0.052 0.054 CH60_DROME 60 kDa heat shock protein, mitochondrial OS=Drosophila melanogaster GN=Hsp60 PE=1 SV=3 0.000 0.117 0.072 Q7KS16_DROME (+1) CG6643-PB, isoform B OS=Drosophila melanogaster GN=CG6643 PE=4 SV=1 0.000 0.079 0.054 A2 Q9VVA4_DROME CG9674-PA, isoform A (CG9674-PD, isoform D) (GH26789p) OS=Drosophila melanogaster GN=CG9674 PE=1 SV=2 0.338 0.432 0.054 A2 Q9VKZ8_DROME Ubiquitin carboxyl-terminal hydrolase OS=Drosophila melanogaster GN=CG5384 PE=2 SV=1 0.094 0.118 0.000 A2 Q7KSB3_DROME (+5) Oxysterol-binding protein OS=Drosophila melanogaster GN=CG5077 PE=2 SV=2 0.072 0.092 0.000 A2 MOCOS_DROME Molybdenum cofactor sulfurase OS=Drosophila melanogaster GN=mal PE=1 SV=1 0.082 0.105 0.000 A2 VPRBP_DROME VPRBP-like protein OS=Drosophila melanogaster GN=CG10080 PE=1 SV=2 0.071 0.091 0.000 A2 Q9VCC6_DROME CG6178-PA (RE32988p) (GM05240p) OS=Drosophila melanogaster GN=CG6178 PE=1 SV=1 0.070 0.092 0.000 A2 A6PX30_DROME (+7) Ref(2)P protein (Fragment) OS=Drosophila melanogaster GN=ref(2)P PE=4 SV=1 0.267 0.379 0.000 A2 IKKB_DROME Inhibitor of nuclear factor kappa-B kinase subunit beta OS=Drosophila melanogaster GN=ird5 PE=1 SV=2 0.059 0.079 0.000 A2 O96695_DROME (+1) Rough deal protein OS=Drosophila melanogaster GN=rod PE=1 SV=1 0.059 0.079 0.000 A2 Q9VHR5_DROME CG9684-PA OS=Drosophila melanogaster GN=CG9684 PE=1 SV=2 0.048 0.066 0.000 A2 SGPL_DROME Sphingosine-1-phosphate lyase OS=Drosophila melanogaster GN=Sply PE=2 SV=1 0.082 0.131 0.000 A2 A1ZBK8_DROME (+2) CG9325-PD, isoform D OS=Drosophila melanogaster GN=hts PE=4 SV=1 0.058 0.092 0.000 A2 IMMT_DROME Putative mitochondrial inner membrane protein OS=Drosophila melanogaster GN=CG6455 PE=2 SV=3 0.095 0.145 0.000 A2 KLC_DROME Kinesin light chain OS=Drosophila melanogaster GN=Klc PE=1 SV=1 0.034 0.052 0.000 A2 A8JUQ9_DROME (+1) CG7893-PC, isoform C OS=Drosophila melanogaster GN=vav PE=4 SV=1 0.034 0.052 0.000 A2 Q7YU85_DROME (+1) SD08724p OS=Drosophila melanogaster GN=BcDNA:GH03163 PE=1 SV=1 0.060 0.104 0.000 A2 Q9VJH2_DROME (+1) CG31739-PA (SD02215p) OS=Drosophila melanogaster GN=mdy PE=1 SV=3 0.023 0.040 0.000 A2 6PGD_DROME 6-phosphogluconate dehydrogenase, decarboxylating OS=Drosophila melanogaster GN=Pgd PE=2 SV=1 0.023 0.039 0.000 A2 Q960B1_DROME (+1) SD10334p OS=Drosophila melanogaster GN=Tom34 PE=2 SV=1 0.023 0.039 0.000 A2 Q86BI3_DROME CG5215-PB, isoform B OS=Drosophila melanogaster GN=Zn72D PE=1 SV=1 0.047 0.079 0.000 A2 Q9VYY3_DROME CG1749-PA OS=Drosophila melanogaster GN=CG1749 PE=1 SV=1 0.047 0.079 0.000 A2 Q7KVT8_DROME CG2206-PB, isoform B OS=Drosophila melanogaster GN=l(1)G0193 PE=2 SV=1 0.047 0.079 0.000 A2 Q7JZ25_DROME RE63672p (CG8594-PA) OS=Drosophila melanogaster GN=CG8594 PE=1 SV=1 0.024 0.040 0.000 A2 O62610_DROME (+4) Nucleoporin OS=Drosophila melanogaster GN=Nup154 PE=2 SV=1 0.024 0.040 0.000 A2 Q7KVS6_DROME (+1) CG10701-PA, isoform A OS=Drosophila melanogaster GN=Moe PE=2 SV=1 0.024 0.040 0.000 A2 Q8T9D5_DROME SD05601p OS=Drosophila melanogaster GN=CG1640 PE=2 SV=1 0.045 0.091 0.000 A2 Q9V426_DROME CG4170-PA, isoform A (CG4170-PB, isoform B) (CG4170-PC, isoform C) (CG4170-PD, isoform D) (LD07162p) OS=Drosophil a melanogaster GN=vig PE=1 SV=1 0.045 0.092 0.000 A2 A2RVC9_DROME (+4) IP03029p (Fragment) OS=Drosophila melanogaster GN=Ir PE=2 SV=1 0.045 0.092 0.000 A2 Q8SZL8_DROME (+1) RH04607p OS=Drosophila melanogaster GN=CG17593 PE=2 SV=1 0.045 0.092 0.000 A2 Q86BM5_DROME CG13388-PD, isoform D OS=Drosophila melanogaster GN=Akap200 PE=1 SV=1 0.034 0.066 0.000 A2 Q9U6I2_DROME (+1) XCAP-C/SMC4 homolog Gluon OS=Drosophila melanogaster GN=glu PE=2 SV=1 0.034 0.066 0.000 A2 Q8T3P0_DROME (+1) GM01289p OS=Drosophila melanogaster GN=CG7145 PE=2 SV=1 0.048 0.092 0.000 A2 Q6NNV8_DROME (+2) RH33950p (Fragment) OS=Drosophila melanogaster GN=CG11109 PE=2 SV=1 0.034 0.065 0.000 A2 NSF2_DROME (+1) Vesicle-fusing ATPase 2 OS=Drosophila melanogaster GN=Nsf2 PE=2 SV=1 0.034 0.065 0.000 A2 A4V0X4_DROME (+2) CG10922-PB, isoform B OS=Drosophila melanogaster GN=La PE=4 SV=1 0.034 0.065 0.000 A2 DHE3_DROME Glutamate dehydrogenase, mitochondrial OS=Drosophila melanogaster GN=Gdh PE=1 SV=2 0.163 0.367 0.000 A2 A4V2F9_DROME (+1) CG2922-PB, isoform B (CG2922-PC, isoform C) (CG2922-PD, isoform D) (CG2922-PE, isoform E) (CG2922-PF , isoform F) (CG2922-PG, isoform G) OS=Drosophila melanogaster GN=exba PE=4 SV=1 0.070 0.157 0.000 A2 Q5KTT4_DROME Putative uncharacterized protein dNAT1 (CG3845-PC, isoform C) (Fragment) OS=Drosophila melanogaster GN=dNA T1 PE=1 SV=1 0.036 0.079 0.000 A2 CYFIP_DROME Cytoplasmic FMR1-interacting protein OS=Drosophila melanogaster GN=Sra-1 PE=1 SV=1 0.024 0.052 0.000 A2 Q9VJ10_DROME CG10655-PA (RE18450p) OS=Drosophila melanogaster GN=l(2)37Bb PE=2 SV=1 0.024 0.052 0.000 A2 Q7K569_DROME GH10595p (CG8256-PA, isoform A) (CG8256-PC, isoform C) (CG8256-PD, isoform D) OS=Drosophila melanogaster GN=l(2)k 05713 PE=2 SV=1 0.036 0.079 0.000 A2 Q298X6_DROPS (+1) DNA-directed RNA polymerase (Fragment) OS=Drosophila pseudoobscura GN=GA16485 PE=3 SV=1 0.024 0.053 0.000 A2 A4V449_DROME (+2) NADH-ubiquinone oxidoreductase 75 kDa subunit OS=Drosophila melanogaster GN=ND75 PE=3 SV=1 0.024 0.053 0.000 A2 PABP_DROME Polyadenylate-binding protein OS=Drosophila melanogaster GN=pAbp PE=1 SV=3 0.024 0.053 0.000 A2 Q9VAS8_DROME CG31048-PA OS=Drosophila melanogaster GN=CG31048 PE=2 SV=3 0.034 0.078 0.000 A2 Q960D3_DROME (+1) SD06613p (CG11844-PC, isoform C) (CG11844-PD, isoform D) (CG11844-PA, isoform A) OS=Drosophila melanogaster GN=CG11844 PE=1 SV=1 0.034 0.078 0.000 A2 Q9I7D3_DROME CG18811-PA (LP14942p) OS=Drosophila melanogaster GN=CG18811 PE=1 SV=1 0.023 0.052 0.000 A2 Q95U54_DROME (+1) GH06271p (CG12005-PB) OS=Drosophila melanogaster GN=Mms19 PE=2 SV=1 0.023 0.052 0.000 A2 Q9VM43_DROME CG13780-PA (RH40211p) (VEGF27Cb) (PDGF/VEGF factor-2) OS=Drosophila melanogaster GN=Pvf2 PE=2 SV=2 0.023 0.052 0.000 A2 Q9VYW4_DROME CG1703-PA (LD04461p) OS=Drosophila melanogaster GN=CG1703 PE=1 SV=1 0.023 0.053 0.000 A2 Q9VQ94_DROME CG10882-PA (LP05220p) OS=Drosophila melanogaster GN=CG10882 PE=1 SV=2 0.200 0.197 0.000 A2 Q7K3W2_DROME GH09295p (CG8728-PA) OS=Drosophila melanogaster GN=CG8728 PE=1 SV=1 0.094 0.092 0.000 A2 Q9W350_DROME CG12132-PA (LD28902p) OS=Drosophila melanogaster GN=c11.1 PE=1 SV=2 0.092 0.092 0.000 A2 O76867_DROME (+1) EG:100G10.7 protein OS=Drosophila melanogaster GN=EG:100G10.7 PE=2 SV=1 0.082 0.078 0.000 A2 Q5U0Z2_DROME (+1) LD30448p OS=Drosophila melanogaster GN=CG10576 PE=2 SV=2 0.198 0.183 0.000 A2 Q86NV9_DROME (+1) GH16592p (Fragment) OS=Drosophila melanogaster GN=yellow-f PE=2 SV=1 0.057 0.052 0.000 A2 PURA_DROME Adenylosuccinate synthetase OS=Drosophila melanogaster GN=CG17273 PE=1 SV=1 0.071 0.065 0.000 A2 Q7PLL6_DROME CG17514-PA, isoform A OS=Drosophila melanogaster GN=CG17514 PE=1 SV=2 0.174 0.158 0.000 A2 Q7KND8_DROME (+1) TXBP181-like protein (CG2072-PA) OS=Drosophila melanogaster GN=TXBP181-like PE=1 SV=1 0.058 0.052 0.000 A2 PYR5_DROME Uridine 5'-monophosphate synthase OS=Drosophila melanogaster GN=r-l PE=2 SV=2 0.082 0.092 0.000 A2 Q7K4N3_DROME LD31211p (CG7035-PA, isoform A) OS=Drosophila melanogaster GN=Cbp80 PE=1 SV=1 0.082 0.092 0.000 A2 IMA_DROME Importin subunit alpha OS=Drosophila melanogaster GN=Pen PE=1 SV=2 0.070 0.079 0.000 A2 Q9VSH4_DROME CG7185-PA (LD25239p) OS=Drosophila melanogaster GN=CG7185 PE=1 SV=2 0.059 0.066 0.000 A2 SYFA_DROME Probable phenylalanyl-tRNA synthetase alpha chain OS=Drosophila melanogaster GN=CG2263 PE=2 SV=1 0.082 0.092 0.000 A2 Q5BHU8_DROME AT04667p OS=Drosophila melanogaster GN=capu PE=2 SV=1 0.058 0.065 0.000 A2 Q9VHA8_DROME CG8286-PA (LD25575p) OS=Drosophila melanogaster GN=P58IPK PE=1 SV=1 0.070 0.079 0.000 A2 U609_DROME UPF0609 protein CG1218 OS=Drosophila melanogaster GN=CG1218 PE=1 SV=2 0.047 0.053 0.000 A2 Q95RG8_DROME LD30319p (CG16728-PA) OS=Drosophila melanogaster GN=CG16728 PE=1 SV=1 0.034 0.039 0.000 A2 CAF1_DROME (+1) Probable histone-binding protein Caf1 OS=Drosophila melanogaster GN=Caf1 PE=1 SV=1 0.034 0.039 0.000 A2 DDX19_DROME (+1) DEAD-box helicase Dbp80 OS=Drosophila melanogaster GN=Dbp80 PE=1 SV=1 0.034 0.039 0.000 A2 Q7K4G8_DROME LD40944p (CG12391-PA) OS=Drosophila melanogaster GN=CG12391 PE=2 SV=1 0.023 0.026 0.000 A2 Q9VNH5_DROME CG2091-PA (GH04919p) OS=Drosophila melanogaster GN=CG2091 PE=1 SV=1 0.023 0.026 0.000 A2 Q26459_DROME (+3) Msr-110 protein (LD44960p) (CG10596-PB, isoform B) OS=Drosophila melanogaster GN=Msr-1 10 PE=2 SV=1 0.176 0.210 0.000 A2 Q8MT58_DROME RE11562p (CG17337-PA) OS=Drosophila melanogaster GN=CG17337 PE=2 SV=1 0.034 0.040 0.000 A2 A8E6Q0_DROME (+1) IP12923p OS=Drosophila melanogaster GN=CG11847 PE=2 SV=1 0.023 0.026 0.000 A2 OSA_DROME (+1) Trithorax group protein osa OS=Drosophila melanogaster GN=osa PE=1 SV=1 0.023 0.026 0.000 A2 UBP7_DROME Ubiquitin carboxyl-terminal hydrolase 7 OS=Drosophila melanogaster GN=Usp7 PE=1 SV=1 0.023 0.026 0.000 A2 Q9I7M8_DROME CG18801-PA OS=Drosophila melanogaster GN=Ku80 PE=1 SV=1 0.023 0.026 0.000 A2 GAG17_DROME Retrovirus-related Gag polyprotein from copia-like transposable element 17.6 OS=Drosophila melanogaster GN=gag PE= 4 SV=1 0.023 0.026 0.000 A2 Q7KJ96_DROME Multi-sex-combs OS=Drosophila melanogaster GN=mxc PE=1 SV=1 0.023 0.026 0.000 A2 A1Z7S0_DROME CG8014-PA OS=Drosophila melanogaster GN=Rme-8 PE=4 SV=1 0.176 0.184 0.000 A2 Q59DP9_DROME CG2165-PC, isoform C OS=Drosophila melanogaster GN=CG2165 PE=2 SV=1 0.048 0.052 0.000 A2 A1Z8W8_DROME (+1) CG8487-PB, isoform B OS=Drosophila melanogaster GN=garz PE=1 SV=1 0.036 0.040 0.000 A2 DIF_DROME (+3) Dorsal-related immunity factor Dif OS=Drosophila melanogaster GN=Dif PE=1 SV=2 0.036 0.040 0.000 A2 PSDIN_DROME Phagocyte signaling-impaired protein OS=Drosophila melanogaster GN=psidin PE=2 SV=1 0.036 0.040 0.000 A2 Q9N6D7_DROME Poly A polymerase (CG9854-PC, isoform C) (CG9854-PA, isoform A) (RE09914p) OS=Drosophila melanogaster GN=hrg PE=1 SV=1 0.036 0.040 0.000 A2 Q494L3_DROME (+2) RE37786p OS=Drosophila melanogaster GN=CG4567 PE=2 SV=1 0.024 0.026 0.000 A2 Q9W374_DROME (+1) CG2194-PB, isoform B (CG2194-PC, isoform C) (GH13260p) (Dihydropyrimidine dehydrogenase) OS=Drosophila melan ogaster GN=su(r) PE=2 SV=2 0.048 0.052 0.000 A2 ULA1_DROME NEDD8-activating enzyme E1 regulatory subunit OS=Drosophila melanogaster GN=CG7828 PE=1 SV=1 0.024 0.026 0.000 A2 A4V4F2_DROME (+1) CG32593-PF, isoform F OS=Drosophila melanogaster GN=Flo-2 PE=4 SV=1 0.024 0.026 0.000 A9UNH9_DROME (+1) DNA-directed RNA polymerase OS=Drosophila melanogaster PE=2 SV=1 0.023 0.131 0.000 Q9VZ20_DROME CG1938-PA, isoform A (CG1938-PC, isoform C) (CG1938-PD, isoform D) (GH06357p) OS=Drosophila melanogaster GN=Dlic2 P E=2 SV=2 0.023 0.118 0.000 Q299D0_DROPS (+1) GA10280-PA (Fragment) OS=Drosophila pseudoobscura GN=GA10280 PE=3 SV=1 0.036 0.170 0.000 SAHH_DROME Adenosylhomocysteinase OS=Drosophila melanogaster GN=Ahcy13 PE=1 SV=2 0.045 0.157 0.000 A1YK54_DROME (+9) Nup107 OS=Drosophila melanogaster GN=Nup107 PE=4 SV=1 0.034 0.118 0.000 Q9VT61_DROME CG8108-PA, isoform A (CG8108-PB, isoform B) (LD27033p) OS=Drosophila melanogaster GN=CG8108 PE=1 SV=1 0.023 0.079 0.000 P91656_DROME (+2) 70 kDa S6 kinase OS=Drosophila melanogaster GN=S6k PE=1 SV=1 0.024 0.079 0.000 Q9U4Y0_DROME (+2) Tryptophanyl-tRNA synthetase (Fragment) OS=Drosophila melanogaster GN=Aats-trp PE=2 SV=1 0.023 0.065 0.000 Q7K4Q5_DROME LD27655p (CG10417-PA, isoform A) OS=Drosophila melanogaster GN=CG10417 PE=1 SV=1 0.023 0.065 0.000 C19L1_DROME CWF19-like protein 1 homolog OS=Drosophila melanogaster GN=CG7741 PE=2 SV=1 0.023 0.065 0.000 SPT5H_DROME Transcription elongation factor SPT5 OS=Drosophila melanogaster GN=Spt5 PE=1 SV=1 0.024 0.065 0.000 A1Z8U4_DROME (+1) CG8439-PB, isoform B OS=Drosophila melanogaster GN=Cct5 PE=3 SV=1 0.034 0.092 0.000 Q9VLQ9_DROME CG8282-PA (LD22082p) OS=Drosophila melanogaster GN=Snx6 PE=1 SV=2 0.000 0.092 0.000 Q86BP3_DROME (+1) CG1091-PA, isoform A OS=Drosophila melanogaster GN=CG1091 PE=2 SV=1 0.000 0.052 0.000 O44368_DROME (+1) Genghis Khan OS=Drosophila melanogaster GN=gek PE=1 SV=1 0.000 0.039 0.000 Q9VDE8_DROME CG7044-PA OS=Drosophila melanogaster GN=CG7044 PE=2 SV=1 0.000 0.052 0.000 KGP24_DROME cGMP-dependent protein kinase, isozyme 2 forms cD4/T1/T3A/T3B OS=Drosophila melanogaster GN=for PE=1 SV=3 0.000 0.039 0.000 Q95R49_DROME (+1) SD08036p OS=Drosophila melanogaster GN=CG3808 PE=2 SV=1 0.000 0.066 0.000 Q9VV60_DROME CG4561-PA (LD21116p) OS=Drosophila melanogaster GN=Aats-tyr PE=1 SV=1 0.000 0.066 0.000 Q9V9T5_DROME CG2118-PA, isoform A (GM14617p) OS=Drosophila melanogaster GN=CG2118 PE=2 SV=2 0.000 0.066 0.000 B1Q013_DROSI Diphenol oxidase A2 (Fragment) OS=Drosophila simulans GN=Dox-A2 PE=4 SV=1 0.000 0.198 0.000 Q9VP57_DROME CG7752-PA (LD15904p) OS=Drosophila melanogaster GN=Z4 PE=1 SV=1 0.000 0.105 0.000 O96046_DROME (+1) Cortactin OS=Drosophila melanogaster GN=Cortactin PE=1 SV=1 0.000 0.039 0.000 ILF2_DROME Interleukin enhancer-binding factor 2 homolog OS=Drosophila melanogaster GN=CG5641 PE=1 SV=1 0.000 0.026 0.000 A1Z7K6_DROME (+1) CG8243-PA OS=Drosophila melanogaster GN=CG8243 PE=4 SV=1 0.000 0.026 0.000 Q9VXV3_DROME CG9198-PA, isoform A OS=Drosophila melanogaster GN=shtd PE=2 SV=1 0.000 0.039 0.000 Q9VL28_DROME CG5734-PA (LD15323p) OS=Drosophila melanogaster GN=CG5734 PE=1 SV=1 0.000 0.026 0.000 Q29FX8_DROPS GA12352-PA (Fragment) OS=Drosophila pseudoobscura GN=GA12352 PE=4 SV=1 0.000 0.039 0.000 Q9W437_DROME CG4320-PA OS=Drosophila melanogaster GN=raptor PE=2 SV=1 0.000 0.026 0.000 A1Z9X2_DROME (+2) CG7639-PA, isoform A OS=Drosophila melanogaster GN=CG7639 PE=4 SV=1 0.000 0.039 0.000 Q94885_DROME Orf protein OS=Drosophila melanogaster GN=ORF PE=1 SV=1 0.000 0.039 0.000 Q9VZF1_DROME CG1309-PA (GH07444p) OS=Drosophila melanogaster GN=CG1309 PE=2 SV=1 0.000 0.026 0.000 O96639_DROME (+1) Drongo OS=Drosophila melanogaster GN=drongo PE=2 SV=1 0.000 0.026 0.000 O18332_DROME (+4) CG3320-PA, isoform A (Rab1) OS=Drosophila melanogaster GN=Rab1 PE=1 SV=1 0.000 0.039 0.000 Q8IGE9_DROME (+2) RH35990p (Fragment) OS=Drosophila melanogaster GN=growl PE=1 SV=1 0.000 0.079 0.000 Q4QQA6_DROME (+2) LD09533p OS=Drosophila melanogaster PE=2 SV=1 0.000 0.040 0.000 RAD50_DROME DNA repair protein RAD50 OS=Drosophila melanogaster GN=rad50 PE=1 SV=2 0.000 0.040 0.000 KPEL_DROME (+3) Probable serine/threonine-protein kinase pelle OS=Drosophila melanogaster GN=pll PE=1 SV=1 0.000 0.040 0.000 BRM_DROME ATP-dependent helicase brm OS=Drosophila melanogaster GN=brm PE=1 SV=2 0.000 0.104 0.000 RENT1_DROME Regulator of nonsense transcripts 1 homolog OS=Drosophila melanogaster GN=Upf1 PE=1 SV=2 0.000 0.040 0.000 Q0KHR5_DROME CG4937-PB, isoform B OS=Drosophila melanogaster GN=RhoGAP15B PE=4 SV=1 0.000 0.040 0.000 RPA1_DROME DNA-directed RNA polymerase I subunit RPA1 OS=Drosophila melanogaster GN=RpI1 PE=1 SV=2 0.000 0.040 0.000 Q9VB22_DROME CG5692-PA (RE22964p) (LD33695p) (Pins protein) OS=Drosophila melanogaster GN=raps PE=1 SV=1 0.000 0.040 0.000 PICA_DROME (+2) Phosphatidylinositol-binding clathrin assembly protein LAP OS=Drosophila melanogaster GN=lap PE=1 SV=3 0.000 0.040 0.000 TBB3_DROME Tubulin beta-3 chain OS=Drosophila melanogaster GN=betaT ub60D PE=1 SV=2 0.000 0.969 0.000 A1Z9L3_DROME (+1) CG8241-PA OS=Drosophila melanogaster GN=pea PE=4 SV=1 0.000 0.040 0.000 Q9VKM2_DROME CG4636-PA (SD02991p) (SCAR) OS=Drosophila melanogaster GN=SCAR PE=1 SV=1 0.000 0.040 0.000 Q7JQN4_DROME LD15481p (CG2173-PA) OS=Drosophila melanogaster GN=Rs1 PE=1 SV=1 0.000 0.040 0.000 Q7K4L8_DROME LD33749p OS=Drosophila melanogaster GN=CG7878 PE=2 SV=1 0.000 0.040 0.000 Q8MQY6_DROME (+1) RE67845p OS=Drosophila melanogaster GN=CG6766 PE=2 SV=1 0.000 0.040 0.000 Q9VIM0_DROME CG2493-PA (GH14278p) OS=Drosophila melanogaster GN=CG2493 PE=2 SV=1 0.000 0.053 0.000 ECM29_DROME Proteasome-associated protein ECM29 homolog OS=Drosophila melanogaster GN=CG8858 PE=1 SV=1 0.000 0.065 0.000 NU301_DROME Nucleosome-remodeling factor subunit NURF301 OS=Drosophila melanogaster GN=E(bx) PE=1 SV=2 0.000 0.053 0.000 A1ZAT5_DROME (+1) CG6522-PA OS=Drosophila melanogaster GN=CG6522 PE=1 SV=1 0.000 0.065 0.000 Q29DL4_DROPS GA10524-PA (Fragment) OS=Drosophila pseudoobscura GN=GA10524 PE=4 SV=1 0.000 0.079 0.000 Q9XZ29_DROME CG8590-PA (EG:BACR25B3.9 protein) (Putative uncharacterized protein) OS=Drosophila melanogaster GN=Klp3A PE=1 SV=1 0.000 0.026 0.000 A1ZBW0_DROME (+4) CG13425-PC, isoform C OS=Drosophila melanogaster GN=bl PE=4 SV=1 0.000 0.052 0.000 Q7PLL1_DROME (+2) CG10837-PE, isoform E OS=Drosophila melanogaster GN=eIF-4B PE=1 SV=2 0.000 0.052 0.000 Q58CL4_DROME SD01796p OS=Drosophila melanogaster GN=pbl PE=2 SV=1 0.000 0.052 0.000 O96828_DROME (+5) EG:EG0003.2 protein (Fbgn0014870;psi protein) OS=Drosophila melanogaster GN=Psi PE=4 SV=1 0.000 0.026 0.000 Q9VGE7_DROME Beta-galactosidase OS=Drosophila melanogaster GN=Ect3 PE=2 SV=1 0.000 0.052 0.000 Q9VRD9_DROME CG1753-PA, isoform A (CG1753-PB, isoform B) (LD21426p) OS=Drosophila melanogaster GN=CG1753 PE=1 SV=1 0.000 0.052 0.000 ITBX_DROME Integrin beta-PS OS=Drosophila melanogaster GN=mys PE=1 SV=3 0.000 0.026 0.000 Q7JVY0_DROME LD37992p (CG9446-PC, isoform C) (CG9446-PA, isoform A) (Coronin) OS=Drosophila melanogaster GN=coro PE=2 SV=1 0.000 0.026 0.000 Q9VF56_DROME CG5205-PA OS=Drosophila melanogaster GN=CG5205 PE=2 SV=2 0.000 0.052 0.000 RUMI_DROME (+1) O-glucosyltransferase rumi OS=Drosophila melanogaster GN=rumi PE=1 SV=1 0.000 0.052 0.000 Q9VG79_DROME CG5663-PA (LD07362p) OS=Drosophila melanogaster GN=Dip-C PE=2 SV=2 0.000 0.052 0.000 LAM0_DROME Lamin Dm0 OS=Drosophila melanogaster GN=Lam PE=1 SV=4 0.000 0.026 0.000 Q9VGS9_DROME (+1) CG6644-PA (LD21102p) OS=Drosophila melanogaster GN=Ugt35a PE=2 SV=1 0.000 0.026 0.000 TSR1_DROME Pre-rRNA-processing protein TSR1 homolog OS=Drosophila melanogaster GN=CG7338 PE=1 SV=1 0.000 0.026 0.000 Q9W3C3_DROME CG2004-PA (LD10016p) OS=Drosophila melanogaster GN=CG2004 PE=2 SV=1 0.000 0.026 0.000 Q9VWP1_DROME CG7288-PA (LD38070p) OS=Drosophila melanogaster GN=CG7288 PE=2 SV=2 0.000 0.026 0.000 ELP2_DROME Probable elongator complex protein 2 OS=Drosophila melanogaster GN=StIP PE=1 SV=1 0.000 0.026 0.000 Q9VJ42_DROME CG10600-PA OS=Drosophila melanogaster GN=CG10600 PE=2 SV=3 0.000 0.026 0.000 Q8IGA9_DROME (+1) SD05534p OS=Drosophila melanogaster GN=CG3173 PE=2 SV=1 0.000 0.026 0.000 O16048_DROME (+3) Anon2A12 (Fragment) OS=Drosophila melanogaster GN=pav PE=1 SV=1 0.000 0.026 0.000 MODU_DROME (+1) DNA-binding protein modulo OS=Drosophila melanogaster GN=mod PE=1 SV=2 0.000 0.053 0.000 Q95R98_DROME LP02515p (CG5068-PA) OS=Drosophila melanogaster GN=CG5068 PE=2 SV=1 0.000 0.026 0.000 Q9VXF9_DROME CG4420-PA (GM04721p) (AT13091p) OS=Drosophila melanogaster GN=CG4420 PE=1 SV=1 0.000 0.026 0.000 ADRM1_DROME (+1) Protein ADRM1 homolog OS=Drosophila melanogaster GN=CG13349 PE=1 SV=1 0.000 0.026 0.000 Q32KE1_DROME (+2) RE04201p OS=Drosophila melanogaster GN=CG31738 PE=2 SV=1 0.000 0.118 0.000 Q8MZI3_DROME GH10652p (CG10077-PA, isoform A) OS=Drosophila melanogaster GN=CG10077 PE=1 SV=1 0.000 0.144 0.000 Q24147_DROME (+1) CP60 OS=Drosophila melanogaster GN=Map60 PE=1 SV=1 0.000 0.026 0.000 Q29MN6_DROPS (+2) GA20319-PA (Fragment) OS=Drosophila pseudoobscura GN=GA20319 PE=4 SV=1 0.000 0.026 0.000 B1NLF3_DROME (+2) Second mitotic wave missing OS=Drosophila melanogaster GN=swm PE=4 SV=1 0.000 0.026 0.000 PNCB_DROME (+1) Nicotinate phosphoribosyltransferase OS=Drosophila melanogaster GN=CG3714 PE=2 SV=2 0.000 0.026 0.000 NEDD4_DROME E3 ubiquitin-protein ligase Nedd-4 OS=Drosophila melanogaster GN=Nedd4 PE=1 SV=2 0.000 0.026 0.000 Q29AG5_DROPS (+2) GA18477-PA (Fragment) OS=Drosophila pseudoobscura GN=GA18477 PE=4 SV=1 0.000 0.026 0.000 TOP3B_DROME DNA topoisomerase 3-beta OS=Drosophila melanogaster GN=Top3beta PE=2 SV=2 0.000 0.026 0.000 Q9VGP4_DROME CG5252-PA (LP08082p) (Ran binding protein 9) OS=Drosophila melanogaster GN=Ranbp9 PE=1 SV=1 0.000 0.026 0.000 Q29EE3_DROPS (+4) GA17431-PA (Fragment) OS=Drosophila pseudoobscura GN=GA17431 PE=3 SV=1 0.000 0.052 0.000 A4V3W1_DROME (+3) CG2621-PJ, isoform J OS=Drosophila melanogaster GN=sgg PE=4 SV=1 0.000 0.026 0.000

209

Q9V470_DROME CG10622-PA, isoform A (LD44970p) (Putative succinyl-coa ligase (GDP-forming) beta-chain) OS=Drosophila melanogaster GN=Sucb PE=2 SV=1 0.000 0.026 0.000 C12A5_DROME Probable cytochrome P450 12a5, mitochondrial OS=Drosophila melanogaster GN=Cyp12a5 PE=2 SV=1 0.000 0.026 0.000 VIR_DROME Protein virilizer OS=Drosophila melanogaster GN=vir PE=1 SV=1 0.000 0.026 0.000 Q8STF7_DROME (+1) Phosphatidylinositol 4-kinase type II alpha (Phosphatidylinositol 4-kinase alpha type II) OS=Drosophila melano gaster GN=Pi4KIIalpha PE=2 SV=1 0.000 0.026 0.000 O76490_DROME (+1) Brahma associated protein 60 kDa OS=Drosophila melanogaster GN=Bap60 PE=1 SV=1 0.000 0.052 0.000 MCCB_DROME (+1) Probable methylcrotonoyl-CoA carboxylase beta chain, mitochondrial OS=Drosophila melanogaster GN=L(2)04524 PE=2 SV=1 0.000 0.052 0.000 Q9VLS7_DROME (+1) CG8552-PA (LD21067p) OS=Drosophila melanogaster GN=CG8552 PE=1 SV=1 0.000 0.026 0.000 Q9VWB9_DROME CG15618-PA OS=Drosophila melanogaster GN=CG15618 PE=1 SV=2 0.000 0.026 0.000 Q24559_DROME (+3) Drosophila translocation protein 1 OS=Drosophila melanogaster GN=T rp1 PE=1 SV=1 0.000 0.026 0.000 RINT1_DROME RINT1-like protein OS=Drosophila melanogaster GN=CG8605 PE=1 SV=2 0.000 0.026 0.000 Q9V393_DROME CG1487-PA (LD31082p) (Kurtz arrestin) OS=Drosophila melanogaster GN=krz PE=1 SV=1 0.000 0.026 0.000 A8JUY0_DROME (+1) CG3312-PD, isoform D OS=Drosophila melanogaster GN=Rnp4F PE=4 SV=1 0.000 0.026 0.000 Q961V1_DROME (+1) GH05723p OS=Drosophila melanogaster GN=CG3033 PE=2 SV=1 0.000 0.026 0.000 A9UND8_DROME (+1) SD18661p OS=Drosophila melanogaster GN=CG10489 PE=2 SV=1 0.082 0.105 0.270 A8Y560_DROME (+3) Ribosomal protein L15 OS=Drosophila melanogaster GN=RpL15 PE=3 SV=1 0.048 0.079 0.180 Q95SF2_DROME (+1) GM01081p OS=Drosophila melanogaster GN=CG10333 PE=2 SV=1 0.023 0.092 0.234 Q494L9_DROME (+4) RE56180p OS=Drosophila melanogaster GN=AP-1gamma PE=2 SV=1 0.045 0.092 0.261 ISWI_DROME Chromatin-remodeling complex ATPase chain Iswi OS=Drosophila melanogaster GN=Iswi PE=1 SV=1 0.023 0.040 0.108 Q9VXQ5_DROME CG8231-PA (GH13725p) OS=Drosophila melanogaster GN=Tcp-1zeta PE=2 SV=1 0.045 0.170 0.287 B3DNM3_DROME (+2) RE35650p OS=Drosophila melanogaster PE=2 SV=1 0.023 0.092 0.144 Q7KJV6_DROME (+1) Ubiquitin-like protein activating enzyme OS=Drosophila melanogaster GN=Uba2 PE=1 SV=1 0.045 0.118 0.207 A4V3P5_DROME (+2) CG2048-PC, isoform C OS=Drosophila melanogaster GN=dco PE=4 SV=1 0.024 0.065 0.108 GCP3_DROME Gamma-tubulin complex component 3 homolog OS=Drosophila melanogaster GN=l(1)dd4 PE=1 SV=2 0.000 0.026 0.054 A1Z7J6_DROME (+2) CG8266-PB, isoform B OS=Drosophila melanogaster GN=sec31 PE=1 SV=1 0.000 0.026 0.081 Q7K126_DROME LD13864p (CG8426-PA) OS=Drosophila melanogaster GN=l(2)NC136 PE=1 SV=1 0.000 0.026 0.108 ARM_DROME (+1) Armadillo segment polarity protein OS=Drosophila melanogaster GN=arm PE=1 SV=1 0.023 0.000 0.216 Q6NNW3_DROME (+1) GH09630p (Fragment) OS=Drosophila melanogaster GN=CG9305 PE=2 SV=1 0.000 0.000 0.180 A4PB57_DROAE (+10) Histone H2B OS=Drosophila americana GN=H2B PE=3 SV=1 0.000 0.000 0.216 MSH2_DROME (+1) DNA mismatch repair protein spellchecker 1 OS=Drosophila melanogaster GN=spel1 PE=1 SV=4 0.000 0.000 0.126 Q9VJ69_DROME CG31790-PA (RE31991p) OS=Drosophila melanogaster GN=CG31790 PE=2 SV=2 0.000 0.000 0.108 XRN2_DROME 5'-3' exoribonuclease 2 homolog OS=Drosophila melanogaster GN=CG10354 PE=1 SV=2 0.000 0.000 0.054 Q9VK63_DROME CG5776-PA (LD25466p) OS=Drosophila melanogaster GN=CG5776 PE=2 SV=1 0.000 0.000 0.081 A8JRC4_DROME (+5) CG31092-PC, isoform C OS=Drosophila melanogaster GN=LpR2 PE=4 SV=1 0.000 0.000 0.081 Q961S9_DROME (+1) SD01502p OS=Drosophila melanogaster GN=Mtp PE=2 SV=1 0.000 0.000 0.054 Q8IR22_DROME CG32580-PA OS=Drosophila melanogaster GN=CG32580 PE=2 SV=3 0.000 0.000 0.216 H33_DROHY (+22) Histone H3.3 OS=Drosophila hydei GN=His3.3A PE=3 SV=2 0.000 0.000 0.081 Q8MKK5_DROME (+4) GH09271p (CG5594-PB, isoform B) OS=Drosophila melanogaster GN=BEST:CK01510 PE=2 SV=1 0.000 0.000 0.054 PUR4_DROME Phosphoribosylformylglycinamidine synthase OS=Drosophila melanogaster GN=ade2 PE=1 SV=2 0.000 0.000 0.054 A8WHE8_DROME (+3) GH10785p OS=Drosophila melanogaster GN=CG9485 PE=2 SV=1 0.000 0.000 0.054 A4PB55_DROAE (+11) Histone H4 OS=Drosophila americana GN=H4 PE=3 SV=1 0.000 0.000 0.081 Q9VL72_DROME (+1) CG5899-PA, isoform A OS=Drosophila melanogaster GN=CG5899 PE=1 SV=1 0.000 0.000 0.054 Q28YI3_DROPS Elongation factor Tu (Fragment) OS=Drosophila pseudoobscura GN=GA11779 PE=3 SV=1 0.000 0.000 0.081 Q7K4B2_DROME LD47540p (CG7845-PA) OS=Drosophila melanogaster GN=CG7845 PE=2 SV=1 0.036 0.000 0.054 Q8IGK8_DROME (+3) RE72291p (Fragment) OS=Drosophila melanogaster GN=GM130 PE=1 SV=1 0.047 0.000 0.081 GPRK1_DROME G protein-coupled receptor kinase 1 OS=Drosophila melanogaster GN=Gprk1 PE=1 SV=1 0.036 0.000 0.072 Q29F72_DROPS GA18125-PA (Fragment) OS=Drosophila pseudoobscura GN=GA18125 PE=4 SV=1 0.036 0.000 0.108 Q9W3X8_DROME CG3198-PA (RH42690p) OS=Drosophila melanogaster GN=CG3198 PE=1 SV=2 0.023 0.000 0.054 Q5U156_DROME (+1) RE01954p (CG8092-PA, isoform A) OS=Drosophila melanogaster GN=CG8092 PE=1 SV=1 0.024 0.000 0.054

Table S1 Hierarchical clustering of RasGAP interacting proteins (full datasets). The table displays the relative abundance (mean normalised spectral counts) of each protein between the different MS analysed biological samples. The protein hits are from Drosophila melanogaster (DROME) and Drosophila pseudoobscura (DROPS) species.

210

Figure S1. Sprint alignment with RIN family proteins

SanDI

MreI

PasI

Figure S1 Sprint alignment with RIN family proteins. The figure displays SPRINT-b (GenBank accession number AAK28060.1) used in this study aligned with mammalian RIN1-3. The figure highlights different Sprint regions including SH2 (yellow), YXXPXD (purple) and VPS9 (orange) domains along with the position of the corresponding point mutations (red arrow) made in different Sprint domains. The restriction enzyme sites used to produce different Sprint truncation constructs are shown. The sequence alignment was produced by ClustalW2. “*”: identical. “:”: conserved substitutions. “.”: semi-conserved substitution.

211

Figure S2. VPS9 domain restricts the subcellular localisation of Sprint A B Y-axis SprintWT Isosurface model Y-axis Sprint VPS9 Isosurface model

GFP GFP Rab5WT Rab5WT

RFP RFP

Merge Merge X-axis X-axis

Figure S2 VPS9 domain restricts the subcellular localisation of Sprint. (A-B) Drosophila S2 cells were co- transfected with Rab5 and Sprint constructs and protein localisation was visualised using fluorescence microscopy. The X and Y axis are non-projected single focal plane images of the cells. Scale bars on the photographs represent 5 μm.

212

Figure S3. Characterising Rab5 and Rab7 RFP constructs A YFP-Rab5WT + YFP-Rab7WT + YFP-Rab5WT + YFP-Rab7WT + RFP-Rab5WT RFP-Rab7WT RFP-Rab7WT RFP-Rab5WT

YFP (Rab)

RFP (Rab)

Merge

B C 25 30

Rab5 20 t

t Rab5

n Rab7

n u

20 u o

c Rab7 o

e l

i l

s 15

v s

e a

10 v

e 10

n M

5 e M 5 0 YFP-Rab5WT + YFP-Rab7WT + YFP-Rab5WT + YFP-Rab7WT + RFP-Rab5WT RFP-Rab7WT RFP-Rab7WT RFP-Rab5WT 0 YFP-Rab5 RFP-Rab5 YFP-Rab7 RFP-Rab7 Figure S3 Characterising Rab5 and Rab7 RFP constructs. (A) Drosophila S2 cells were co-transfected with Rab5 and Rab7 YFP and RFP constructs and protein localisation was visualised using fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Rab5 (yellow bars) and Rab7 (blue bars) YFP vesicles per cell and Rab5 and Rab7 RFP vesicles co-localising with Rab5 and Rab7 YFP vesicles were recorded (n ≥ 30). (C) Rab5 YFP and RFP as well as Rab7 YFP and RFP constructs were expressed in different cells and the number of vesicles they formed were counted. Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and ** indicates a significant difference between sample means (p < 0.01).

213

Figure S4. Conserved Sprint VPS9 residues are responsible for restriction of its subcellular localisation. A SprintVPS9PA SprintVPS9DA SprintVPS9DA/PA

GFP (Sprint)

*** 35 ***

30 t

n 25

a 20

n u

p 15

n a

e 10 M

0 SprintVPS9PA SprintVPS9DA SprintVPS9DA/PA

Figure S4 Conserved Sprint VPS9 residues are responsible for restriction of its subcellular localisation. (A) Drosophila S2 cells were transfected with Sprint constructs and protein localisation was visualised using fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Sprint puncta (green bars) per cell were recorded (n ≥ 54). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001).

214

Figure S5. Subcellular localisation of wild-type and mutant Sprint and RasGAP A SprintWT SprintFXXPXD SprintSH2*

GFP (Sprint)

-myc (RasGAP)

WT SH2*32* B RasGAP RasGAP 25 ***

Sprint

t n

u RasGAP o

c 15

n 10

e M

0 SprintWT SprintFXXPXD SprintSH2* RasGAPWT RasGAPSH2*32*

Figure S5 Subcellular localisation of wild-type and mutant Sprint and RasGAP. (A) Drosophila S2 cells were transfected with Sprint or RasGAP constructs and protein localisation was visualised using fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Sprint (green bars) and RasGAP (red bars) puncta per cell were recorded (n ≥ 42). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001).

215

Figure S6. Sprint affects RasGAP subcellular distribution

Sprint443-1292+ Sprint443-1292+ Sprint443-1292 A RasGAPWT RasGAPSH2*32*

GFP (Sprint)

-myc (RasGAP)

Merge

25 ***

Sprint 20

Sprint-RasGAP

n u

o 15

u p

e M

0 Sprint443-1292 + Sprint443-1292 + Sprint443-1292 RasGAPWT RasGAPSH2*32*

Figure S6 Sprint affects RasGAP subcellular distribution. (A) Drosophila S2 cells were co-transfected with Sprint and RasGAP constructs and protein localisation was visualised using fluorescence microscopy. Scale bars on the photographs represent 5 μm. (B) The number of Sprint puncta (green bars) per cell and RasGAP puncta (red bars) co-localising with Sprint puncta were recorded (n ≥ 66). Data are represented as mean ± SEM. Statistical significance was tested by one-way ANOVA and *** indicates a significant difference between sample means (p < 0.001).

216

Figure S7. Human EGF ligand can activate MAPK in mammalian HEK293 cells HEK293 100 ng/ml A - + hEGF 220

Cell extract WB: -EGFR 100 1 2 B 60 Cell extract 45 WB: -dpERK 1 2 C 60 Cell extract 45 WB: -MAPK

1 2

Figure S7 Human EGF ligand can activate MAPK in mammalian HEK293 cells. (A) Mammalian HEK293 cells were treated with 100 ng/ml human EGF (hEGF) ligand and were western-blotted using anti-EGFR (large filled arrowhead) antibody (B) The mammalian HEK293 cell lysates were tested for MAPK activation using anti-dpERK antibody (small filled arrowheads) and (C) MAPK levels using anti-MAPK antibody (small filled arrowheads).

217

Figure S8. Genotyping Drosophila vap spri stocks

1 G 6 i 1 r G 6 p R i s - r , n p 2 1 s p o G 2 6 a A g 2 , i e p r v bp r p - , a p a O v s v w 800 600 1 2 3 4 5 B 1000 800 600

400 1 2 3 4 5 1500 C 1000

1 2 3 4 5

Figure S8 Genotyping Drosophila vap spri stocks. (A) Drosophila stocks were genotyped by PCR with primers for exon 3 of the spri gene, which is absent in the spri6G1 allele. (B) Drosophila stocks were genotyped by PCR with primers for exon 4 of the vap gene, which is truncated in the vap2 allele. (C) Drosophila stocks were genotyped with primers for exons 5 and 6 of the vap gene, which is present in all stocks.

218

Figure S9. Drosophila X chromosome gene recombination schematic

P X X

w+, vap2 w-, spri6G1/Y vap2/Y w-, spri6G1

F1 X

w+, vap2, spri+/ FM7i/Y w-, vap+, spri6G1

F2 X Over 100 individual lines were setup

w*, vap*, spri*/FM7i FM7i/Y

F3 Molecular characterisation of lines

w+, vap2, spri6G1/Y w-, vap2, spri6G1/Y

Figure S9 Drosophila X chromosome gene recombination schematic. Virgin (☿☿) Drosophila vap2 and spri6G1 females were crossed en masse with spri6G1/Y and vap2/Y males (♂♂), respectively. The F1 generation virgin females were crossed en masse with FM7i/Y balancer males, producing a large number of FM7i balanced females, which had potential (*) X chromosome gene recombination. Over 100 of these [w*, vap*, spri*]/FM7i F2 virgin females (☿) were then individually crossed with FM7i/Y balancer males, forming a large number of X chromosome balanced stocks. These F3 fly lines were subsequently genotyped by PCR for the presence of both vap2 and spri6G1 alleles.

219