UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Membrane localization of RasGRPs by C1 domains Goulding, Rebecca Ellen 2008

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2008_fall_goulding_rebecca.pdf [ 5.25MB ]
Metadata
JSON: 24-1.0069949.json
JSON-LD: 24-1.0069949-ld.json
RDF/XML (Pretty): 24-1.0069949-rdf.xml
RDF/JSON: 24-1.0069949-rdf.json
Turtle: 24-1.0069949-turtle.txt
N-Triples: 24-1.0069949-rdf-ntriples.txt
Original Record: 24-1.0069949-source.json
Full Text
24-1.0069949-fulltext.txt
Citation
24-1.0069949.ris

Full Text

MEMBRANE LOCALIZATION OF RASGRPS BY Cl DOMAINS by  REBECCA ELLEN GOULDING B.A. Trinity College Dublin, 1998 M.Sc. Trinity College Dublin, 2002  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in  COLLEGE FOR INTERDISCIPLINARY STUDIES (Genetics)  The University British Columbia (Vancouver’)  October 2008  © Rebecca Ellen Goulding  ABSTRACT Ras and Rap GTPases are membrane-bound activators of signal transduction pathways that regulate several cell processes including proliferation, apoptosis and adhesion. Guanine nucleotide exchange factors (GEF5) positively regulate Ras and Rap GTPases by exchanging guanosine diphosphate (GDP) for guanosine triphosphate (GTP). In order to activate Ras and Rap GTPases, GEFs must be at the same membrane compartments where their target GTPases are located. The Ras guanine nucleotide releasing protein (RasGRP) family of four GEFs regulate both Ras and Rap GTPases, with differential specificities. All RasGRPs contain Cl domains, which have the potential to bind the lipid second messenger diacylglycerol (DAG) that is generated at membranes in response to the ligation of many cell surface receptors. Binding of their Cl domains to DAG could serve to co-localize RasGRPs with membrane bound Ras and Rap GTPases. While some evidence exists for each member of the RasGRP family being potentially regulated by their Cl domains binding to DAG, there is contradictory evidence for RasGRP2. My thesis research focused on Cl domain-mediated mechanisms of RasGRP membrane localization, with special focus on RasGRP2. I found that the Cl domains of RasGRP2 and the 13 splice variant of RasGRP4 do not bind either DAG or phorbol ester, a DAG analog. However, all RasGRP Cl domains were shown to bind anionic phospholipids. I determined that the Cl domain of RasGRP2 is required for constitutive plasma membrane localization in NIH 3T3 fibroblasts and T-cells, and also for translocation to the plasma membrane in SDF-1c-stimulated T-cells. I also identified a putative PDZ protein binding site which is required for RasGRP2 localization at the Golgi. My experiments showed that while RasGRP2 localization can occur at the plasma membrane and Golgi of NIH 3T3s, RasGRP2 mediated Rapi activation at the plasma membrane via its Cl domain is required for changes in cell morphology that are induced by RasGRP2 expression. My thesis research has demonstrated that all four members of the RasGRP family utilize their Cl domains to localize to membranes, although in the case of RasGRP2 this occurs via a DAG-independent mechanism, which targets RasGRP2 to the plasma membrane.  11  TABLE OF CONTENTS  ABSTRACT  ii  TABLE OF CONTENTS  iii  LIST OF TABLES  vii  LIST OF FIGURES  viii  LIST OF ABBREVIATIONS  x  ACKNOWLEDGEMENTS  xii  CO-AUTHORSHIP STATEMENT  xiii  CHAPTER 1: INTRODUCTION 1.1  COMPARTMENTALIZATION OF CELL SIGNALING AT MEMBRANES  2  1.2  COMPARTMENTALIZATION OF RAS GTPASE SIGNALING AT MEMBRANES  2  1.3  1.4  1.5  1.2.1  Ras and Rap GTPases are membrane bound signal transducers  2  1.2.2  Mechanisms of Ras and Rap membrane localization  4  1.2.3  Activation of Ras and Rap GTPases at specific membrane locations  7  SIGNAL TRANSDUCTION FROM RAS AND RAP GTPASES  10  1.3.1  Ras activation of the Raf-ERK pathway  10  1.3.2  Ras activation of the Raf-ERK pathway at specific membranes  10  1.3.3  Rap 1 activation and inhibition of the Raf-ERK pathway  12  1.3.4  Rap 1 activation of the Raf-ERK pathway at specific membranes  13  RAP GTPASES AND ADHESION  14  1.4.1  Role of Rap GTPases and integrin mediated adhesion  14  1.4.2  Role of Rap 1 in antigen receptor induced integrin mediated adhesion  15  1.4.3  Role of Rap 1 in chemokine induced integrin mediated adhesion  16  1.4.4  Role of Rap 1 in cadherin mediated adhesion  17  GUAN1NE NUCLEOTIDE EXCHANGE FACTORS: POSITIVE REGULATORS OF RAS AND  RAP 17 1.5.1 1.6  Regulation of Ras and Rap GTPases by membrane localization of GEFs  RASGRP FAMILY OF GEFS  17 21  1.6.1  RasGRP family protein structure  21  1.6.2  RasGRP expression and function  22  1.6.3  RasGRP GTPase specificity  23  111  1.7  REGULATIONOFRASGRPSBYC1 DOMAiNS  .24  1.7.1  Membrane binding by Cl domains  24  1.7.2  PKC ö Cib: the archetypal typical Cl domain  26  1.7.3  The function of atypical Cl domains  28  1.7.4  Regulation of RasGRP 1 localization by the Cl domain  29  1.7.5  The Cl domains of RasGRP2, 3 and4  31  1.8  THESIS OBJECTIVES AND APPROACH  33  1.9  BIBLIOGRAPHY  35  CHAPTER 2: DIFFERENTIAL MEMBRANE BINDING AND DIACYLGLYCEROL RECOGNITION BY Cl DOMAINSOF RASGRPS 43 2.1  INTRODUCTION  44  2.2  MATERIALS AND METHODS  49  2.3  2.2.1  Construction and expression of GFP/C 1 domain fusion proteins  49  2.2.2  Fluorescence microscopy  49  2.2.3  Preparation of cellular membrane and cytosol fractions  50  2.2.4  Construction ofK-Ras/C1 fusions, RasGRP1&C1 fusions and RasGRP4a and 13  51  2.2.5  NIH 3T3 transformation assays  51  2.2.6  Construction, expression and purification of GST/C 1 fusion proteins  52  2.2.7  Binding of GST-C I domain fusion proteins to sucrose-loaded phospholipid vesicles  53  RESULTS  55  2.3.1  Sequence comparison of RasGRP Cl domains  2.3.2  Only the Cl domains of RasGRP 1, 3 and 4cz co-localize with membranes and translocate in  response to DAG or phorbol ester 2.3.3  55  57  The Cl domains of RasGRP 1 and 4cL bind directly to DAG within phospholipid membranes,  while the Cl domains of RasGRP2 and 413 do not  63  2.3.4  The Cl domains of RasGRP 1 and 4cL have different affinities for DAG  2.3.5  Phosphatidic acid, lysophosphatidic acid, ceramide, fatty acids and sphingosine- 1-phosphate are  not alternate ligands for RasGRP2 or RasGRP413 2.3.6  68  Only the DAG-binding Cl domains of RasGRPs can complement a membrane binding  deficiency mutation in K-Ras 2.3.8  66  High concentrations of anionic phospholipids enable membrane binding by RasGRP Cl  domains in the absence of DAG 2.3.7  65  70  Despite lacking discernible membrane localization in vivo, the Cl domain of RasGRP2 can  functionally replace the Cl domain within RasGRP 1  72  2.4  DISCUSSION  76  2.5  BIBLIOGRAPHY  79  iv  CHAPTER 3: LOCALIZATION OF RASGRP2 IS SPECIFIED BY THE CI DOMAIN AND A POTENTIAL BLND1NG SITE FOR PDZ PROTEINS 83 3.1  INTRODUCTION  84  3.2  MATERIALS AND METHODS  87  3.2.1  Cell lines and reagents  87  3.2.2  Construction of modified forms of RasGRP2  87  3.2.3  Retroviral transduction of cell lines  89  3.2.4  Fluorescence microscopy  89  3.2.5  Cell colony boundary and scratch test assays  90  3.2.6  Cell stimulation, lysis and affinity purification of activated Rap 1 GTPase by Ra1GDS-RBD  chromatography 3.2.7 3.3  91  Western blot analysis  91  RESULTS  93  3.3.1  The Cl domain of RasGRP2 specifies plasma membrane localization  93  3.3.2  A potential binding site for PDZ proteins specifies Golgi localization of RasGRP2  97  3.3.3  Both the Cl domain and the PDZ-binding motif contribute to RasGRP2 activity as an exchange  factor for Rapl 3.3.4  99  Changes in cell morphology and colony edge phenotype induced by RasGRP2 require plasma  membrane localization via the Cl domain 3.3.5  99  The Cl domain and PDZ-binding motif specify plasma membrane versus Golgi localization of  RasGRP2 in Jurkat T-cells  108  3.3.6  RasGRP2 does not respond to antigen receptor stimulation  3.3.7  The Cl domain is required for concentration of RasGRP2 at the plasma membrane in response  to the chemokine SDF- I u  111  111  3.4  DISCUSSION  116  3.5  BIBLIOGRAPHY  119  CHAPTER 4: DISCUSSION  122  Cl DOMAIN MEDIATED LOCALIZATION OF RASGRPS  4.1  123  4.1.1  Not all CI domains bind DAG  123  4.1.2  Not all RasGRP Cl domains are targeted to the same membranes  124  4.2  RASGRP2 LOCALIZATION VIA THE PDZ-BINDING MOTIF  125  4.3  RECEPTOR MEDIATED RASGRP2 ACTIVATION  126  4.3.1  Constitutive membrane localization of RasGRP2  126  4.3.2  SDF- 1 x induced plasma membrane localization of RasGRP2  127  4.4  HETEROGENEITY OF MECHANISMS FOR MEMBRANE LOCALIZATION OF RASGRPS 128  4.5  MEMBRANE SPECIFIC ACTIVATION OF RAP1 BY RASGRP2  V  130  LET  IA  91’  NOIIV]fl037}I tJ’IDSVN HOd SMOIJNDI1JNI GMV SMOISIflDMOD ]VH3N3O  9t  AHdVHOOI]qIH  ITA  saIosI jo  sflT3gT3dS  sjj  SInIflvI Io lsfl  LIST OF FIGURES  Figure 1.1  Ras GTPases function as a molecular binary switch  3  Figure 1.2  Classical Ras and Rap GTPase hypervariable regions (HVR)  5  Figure 1.3  H-Ras and Rapi plasma membrane localization  7  Figure 1.4  Ras signaling to the Raf-ERK pathway  11  Figure 1.5  GEF and GAP regulation of Ras GTPases  19  Figure 1.6  RasGRP protein structure  21  Figure 1.7  DAG production and Cl domain localization at membranes  26  Figure 1.8  Typical and atypical Cl domain sequence comparison  27  Figure 1.9  The Cib domain structure of PKCö  28  Figure 1.10  RasGRP Cl domains  32  Figure 2.1  Cl domain sequences  56  Figure 2.2  Only the Cl domains of RasGRP1, RasGRP3 and RasGRP4L co-localize and translocate in response to DAG or phorbol esters  59  Figure 2.3  Distribution of Cl domains in DO 11.10 T-cells and WEHI-23 1 B-cells  61  Figure 2.4  DAG induces translocation of the Cl domains of RasGRP1, RasGRP3 and RasGRP4a to membranes  Figure 2.5  62  Binding of the Cl domains of RasGRP1, RasGRP3 and RasGRP4a to DAG or PMA within membrane vesicles  64  Figure 2.6  Lack of binding of RasGRP Cl domains to alternative lipid ligands  67  Figure 2.7  RasGRP Cl domains bind to vesicles enriched in anionic phospholipids.. .69  Figure 2.8  Only the Cl domains of RasGRP1, RasGRP3 and RasGRP4o provide serum- or phorbol-ester dependent complementation of a membrane binding-deficient K-Ras mutant  Figure 2.9  71  The Cl domains ofRasGRPl, RasGRP2, RasGRP3 and RasGRPcL are functional within RasGRP1, whereas the Cl domain of RasGRP4f3 is not...74  vii’  Figure 3.1  RasGRP2 plasma membrane localization is specified by the Cl domain  Figure 3.2  RasGRP1 localization at the Golgi and ER compared to RasGRP2 in NIH 3T3 cells  94  96  Figure 3.3  RasGRP2 localization at the Golgi requires a PDZ-binding motif  Figure 3.4  The Cl domain and the potential PDZ-binding motif are required for full activation of RasGRP2  101  103  Figure 3.5  The Cl domain is required for RasGRP2 induced changes in morphology.104  Figure 3.6  Targeting RasGRP2 to the plasma membrane induces colony morphology.106  Figure 3.7  The Cl domain specifies plasma membrane localization of RasGRP2 in Jurkat and DO1l.10 T-cells  Figure 3.8  109  The PDZ-binding motif specifies RasGRP2 localizaiton at the Golgi in Jurkat T-cells  110  Figure 3.9  RasGRP2 does not respond to TCR or BCR stimulation  113  Figure 3.10  RasGRP2 accumulates at the plasma membrane in response to SDF- 1cc treatment of Jurkat T-cells  115  ix  LIST OF ABBREVIATIONS ADAP, adhesion and degranulation promoting adapter protein APC, antigen-presenting cell BCR, B-cell receptor BMC, bone marrow cell CFP, cyan fluorescent protein CRD, cysteine rich domain DAG, diacyiglycerol DMEM, Dulbecco’s modified Eagle’s medium ECM, extra-cellular matrix EGF, epidermal growth factor ER, endoplasmic reticulum ERK1/2, extracellular signal-regulated kinase 1/2 FRET, fluorescence resonance energy transfer GAP, GTPase activating protein GDP, guanine diphosphate GFP, green fluorescent protein GEF, guanine nucleotide exchange factor GRP, guanine nucleotide releasing protein GST, glutathione S-transferase GTP, guanine triphosphate HEy, high endothelial venules HVR, hypervariable region ICAM- 1, intercellular adhesion molecule-i (ICAM- 1) 1P3, inositol trisphosphate KSR, kinase suppressor of Ras LAD, leukocyte adhesion deficiency LFA- 1, lymphocyte-function-associated antigen-i LPA, lysophosphatidic acid MAPK, mitogen activated protein kinase  x  MDCK, Madin Darby canine kidney MEK1/2, MAPKIERK kinase 1/2 MHC, maj or histocompatibility complex PA, phosphatidic acid PC, phosphatidyicholine PDZ, post synaptic density protein (PSD95), drosophila disc large tumor suppressor (D1gA), and zonula occuldens-i protein (Zo-1) PDZB, PDZ-binding motif PE, phorbol ester PG, phosphatidyiglycerol P13 K, phosphatidylinositol 3 -kinase PIPs, phosphatidylinositol phosphates PKC, protein kinase C PLC, phospholipase C PMA, phorbol 13-acetate PS, phosphatidylserine PT, plasma membrane targeter RBD, Ras binding domain RTK, receptor tyrosine kinase SK.AP55, Src kinase-associated phosphoprotein of 55 kDa SDF- 1 a, stromal cell-derived factor-i a SLy, sucrose-loaded large unilamellar vesicles SPA-i, signal-induced proliferation associated gene 1 TCR, T-cell receptor VCAM- 1, vascular cell adhesion molecule-i VLA-4, very late antigen-4 YFP, yellow fluorescent protein  xi  ACKNOWLEDGEMENTS  Firstly, I would like to thank my supervisor, Rob Kay, for his unwavering guidance, patience, and support throughout my PhD. I have learned so much from you and I could not have achieved what I have without your mentorship. Thank-you for teaching me about integrity, perspective and the importance of being true to oneself. Thanks also for countless wonderful conversations on politics and ale.  I would never have completed this journey, had it not been for Conor Reynolds, my best friend and partner for life. No words can truly express the gratitude I feel for all the years you have stood beside me and helped me to find the courage to continue when times were tough. We have had such fun together, and there is so much more to look forward to.  My parents, Lesley and Allan Goulding, started it all by encouraging me to study the natural sciences. Thanks Mum and Dad for always being there through thick and thin, with moral support and unconditional love. Thanks also to my fantastic sister Sarah Goulding, who also provided love, support and great advice. Sisters know best, and you always knew the right thing to say to motivate me!  I owe special thanks to my dear lab compadres, Ghazaleh Tazmini, Nadine Beaulieu and Ban Zahedi who made science and life in the lab so much fun. Thanks Ghaz for always being there for me!  Thanks to the many great friends who have given me emotional support throughout my PhD, whether it was climbing a mountain, or drinking tea on the porch with me, I owe a great deal to my community of friends who have enlightened my life.  I would like to dedicate this thesis to Christine Stathers, who is one of the most inspiring individuals one could hope to meet on this small blue planet. Thanks Chris, for everything you have taught me.  xii  CO-AUTHORSHIP STATEMENT A version of Chapter 2 has been published: Johnson, JE, Goulding, RE, Ding, Z, Partovi, A, Anthony, KV, Beaulieu, N, Tazmini, G, Cornell, R. B and Kay, RJ. (2007). Differential membrane binding and diacylglycerol recognition by Cl domains of RasGRPs. Biochem J. 406: 223-36. Joanne Johnson and Rebecca Goulding were co-first authors of this publication. In the Kay lab, sequence alignment and analysis, in vivo microscopy and transformation assay experiments were carried out by Rebecca Goulding (Figure 2.1, 2.2, 2.3 2.8 and 2.9). In the Cornell lab, in vivo and in vitro experiments were carried out by Joanne Johnson and Ziwei Ding (Figures 2.4, 2.5, 2.6 and 2.7), while Amir Partovi worked on making the GST fusion proteins used for the experiments in Figure 2.4, 2.5, 2.6 and 2.7.  A version of Chapter 3 will be submitted for publication: Goulding RE and Kay RJ. Localization of RasGRP2 is specified by the Cl domain and a potential binding site for PDZ proteins. All experiments in this chapter were carried out by Rebecca Goulding. Robert Kay made the following RasGRP2 constructs: GFP-tagged RG2, RG2AC1, RG2C1, RG2C1x2, RG2-PBji, RG2-pr and RG2-1C1. Rebecca Goulding made the following constructs: GFP tagged RG2-C-term and RG2-C1+C-term.  Rebecca Goulding  Robert Kay  xlii  CHAPTER 1: INTRODUCTION  1  1.1  COMPARTMENTALIZATION OF CELL SIGNALING AT MEMBRANES Signal transduction pathways transform external stimuli from simple messages into  complex internal cellular processes. These outside-in molecular communication strategies are commonly initiated via peripheral stimulation of cell surface receptors by specific ligands such as growth factors, cytokines and hormones. Following initial receptor stimulation, an ordered cascade of intracellular biochemical events, involving an array of signal transducers, transmit information and help to elicit a response. The organizational complexity of such a system is vast, and cell organelles and macromolecular components play a large role in governing signaling hierarchy. Lipid membranes act as effective signaling platforms because they are sites where signaling proteins can be anchored, concentrated and organized. Lipid second messengers are produced and retained at membranes from existing membrane components. The cell has a diversity of membranes including the plasma membrane, endoplasmic reticulum, Golgi apparatus and nuclear envelope. Specific signaling proteins can localize and concentrate at specific membranes; therefore membranes function to organize signal transduction pathways into separate compartments. Different cellular responses may be dependent on where particular signal transduction pathways are activated; therefore compartmentalization of signal transduction in a membrane-specific manner has the potential to govern the output of such signaling pathways.  1.2  COMPARTMENTALIZATION OF RAS GTPASE SIGNALING AT MEMBRANES  1.2.1  Ras and Rap GTPases are membrane bound signal transducers Ras and Rap GTPases, which are members of the Ras subfamily of small GTPases,  are examples of membrane localized signal transducers. GTPases function as molecular binary switches which cycle between GDP-bound (inactive) and GTP-bound (active) states. Ras subfamily GTPases play a central role in the direction of cell signaling pathways that are critical for cell processes including proliferation, differentiation, apoptosis, adhesion and locomotion (Figure 1.1). Ras, Rap and other Ras subfamily GTPases are constitutively localized at membranes. In their active GTP-bound state they localize and activate their 2  target effectors, which are themselves signal transducers [reviewed in (Mor and Philips, 2006)]. Ras GTPase isoforms (H-Ras, N-Ras and K-Ras4B; also known as “classical” Ras) can bind the same effectors, yet knockouts of the different isoforms lead to different phenotypes (Esteban et al., 2001; Johnson et al., 1997). It is still unclear how functional specificity of different isoforms can be achieved, however recent studies support the notion that different membrane localization of Ras isoforms in the cell may account for their specific signaling outputs.  OFF  ON  Effectors  ProIiferatiodhesion Apoptosis  Figure 1.1  Ras GTPases function as a molecular binary switches  In its “OFF” form, Ras-GDP is thought of as inert. The “ON” and thus active form of Ras, Ras-GTP, activates effectors that are involved in several signal transduction pathways, including those that regulate proliferation, apoptosis and adhesion.  3  1.2.2  Mechanisms of Ras and Rap membrane localization Membrane localization of classical Ras (H-Ras, N-Ras and K-Ras4B) and Rap  (RaplA and B; Rap2A, B and C) proteins is thought to be required for their biological activity [reviewed in (Colicelli, 2004; Takai et al., 2001)1. These GTPases have neither membrane localization sequences nor hydrophobic membrane spanning domains and are synthesized on free ribosomes as hydrophilic proteins. They achieve appropriate membrane localization through a number of post-translational lipid modifications [reviewed in (Mor and Philips, 2006)1. Classical Ras proteins exhibit close to 100% homology at their N-termini, but in contrast their C-termini are characterized by a hypervariable region (HVR) (Figure 1.2). Overall, Rapi proteins are 97% homologous while Rap2 proteins are 90% homologous, and like Ras GTPases, Rap GTPases have regions of hypervariability at their C-termini. Within the HVR, Ras and Rap GTPases have a C-terminal CAAX prenylation signal (C Cysteine, A  =  =  aliphatic amino acid, X = any terminal amino acid), which is prenylated in the  cytosol by one of two soluble prenyltransferases. Prenyltransferases covalently attach a farnesyl or geranylgeranyl lipid by a stable thioether linkage to the CAAX cysteine. The substrate specificity for famesyltransferase versus geranylgeranyltransferase depends on the X residue of the CAAX motif. Classical Ras proteins have the signature CAAX motif, where X  =  serine or methionine, which allows recognition by farnesyltransferase (see Figure 1.2),  while Rap 1 A and Rap 1 B have CAAL (L= leucine) sequence at their C-termini, and are recognized by geranylgeranyltransferase [reviewed in (Zhang and Casey, 1996)]. Rap2A terminates in CAAQ (QGlutamine), and is farnesylated, while Rap2B terminates in CAAL and is geranylgeranylated (Farrell et al., 1993). Newly discovered Rap2C (Paganini et al., 2006) ends in CAAQ and therefore is predicted to be farnesylated (Guo et al., 2007). Prenylation is followed by -AAX proteolysis by Ras-converting enzyme (Reel) and C terminal carboxylmethylation, catalyzed by isoprenylcysteine methyltransferase (ICMT). Reel and ICMT are associated with the endoplasmic reticulum (ER), which is where both of these two processes take place. Together all three modifications, collectively termed carboxyl-terminal CAAX processing, give rise to a hydrophobic C-terminus that can insert into membranes.  4  H-Ras  HKLRKLNPPDESGP.GCMSCKCVLS  N-Ras  YRMKKLNSSDDGTQGcMGLPCVVM  K-Ras4B  KHKEKM.SKDGKKKKK  KSKTKCVI  M  RaplA  RKTPVEKKKPKKKSCLLL  RapiB  RKT  Rap2A  YAAQPDKDD  PSACNi  Q  Rap2B  YAAQP  GçCSA  L  Rap2C  YSSLPEKQDQCCTTCVVQ  Figure 1.2  PVP  GKA  NGDE  RKKS  S  CQL  cvi  L  Classical Ras and Rap GTPase hypervariable regions (HVRs)  Cysteine (C) palmitoylation and prenylation of the hypervariable regions function to target Ras and Rap GTPases to membranes. In the case of K-Ras4B, RaplA and RapiB, polybasic stretches of amino acid residues lysine (K) or arginine (R) are thought to interact with anionic phospholipids via electrostatic interactions, thereby stabilizing membrane binding. K, R: basic residues C: palmitoylated residues; C: prenylated residues; CAAS/M/Q: motif is farnesylated; CAAL: motif is geranylgeranylated.  While CAAX processing targets Ras and Rap GTPases to the ER, these modifications are insufficient for plasma membrane targeting (Choy et al., 1999). In the case of H-Ras, N Ras and Rap2 proteins, plasma membrane localization is achieved through further modification of the HVR. One (N-Ras) or two (H-Ras, Rap2) palmitoyl residues are added to cysteines adjacent to the farnesyl group in the HVR (Figure 1.2), in a reaction catalyzed by palmitoyl acyltransferases (PAT). Palmitoylation serves to further increase plasma membrane affinity of the hydrophobic CAAX processed C-termini, via insertion of the palmitoyl groups (Figure 1.3). Palmitoylation also allows H-Ras and N-Ras to be transported via Golgi to the plasma membrane, via normal secretory pathways (Choy et al., 1999), and the same may be true for Rap2 proteins. H-Ras and N-Ras can be subsequently depalmitoylated and then transported in retrograde fashion back to the Golgi [reviewed in (Mor and Philips, 2006; Omerovic et al., 2007)]. Rap2 proteins are predicted to be similarly transported back to the Golgi.  K-Ras4B and Rap 1 (A and B) proteins do not undergo palmitoylation to achieve plasma membrane localization, but instead have a stretch of basic amino acids (polybasic domain) proximal to the CAAX motif, which bind to anionic phospholipids non-specifically 5  via electrostatic interactions (see Figure 1.2, Figure 1.3). K-Ras4B and Rap 1 are transported from the endoplasmic reticulum to the plasma membrane by a distinct and as yet undetermined alternative transport mechanism. K-Ras4B can also be rapidly transported back to the Golgi, and this is thought to be facilitated by protein kinase C (PKC) phosphorylation of the C-terminal HVR (Mor and Philips, 2006). It is not clear whether Rapi also undergoes retrograde transport from the plasma membrane to the Golgi.  While H-Ras and N-Ras can localize to both the plasma membrane and Golgi, N-Ras has been found to localize more strongly to the Golgi than the plasma membrane in Madin Darby canine kidney (MDCK) cells (Rocks et al., 2005). This is thought to be due to the fact that N-Ras has only one palmitoylate modification, whereas H-Ras has two. In contrast, K Ras4B is restricted to the plasma membrane in the same cells (Rocks et a!., 2005), most likely because of its polybasic domain within the HVR. While Rapl has also been reported to localize in several different cell lines and tissues at the Golgi (Beranger et al., 1991a; Nomura et al., 2004) and to the plasma membrane (Bivona et a!., 2004), its localization at late endosomes, (Bivona et al., 2004; Pizon et al., 1994; York et al., 2000) differentiates it from other Ras GTPases. Rap2 GTPases have been found to localize at the plasma membrane, Golgi (Ohba et al., 2000b; Pizon et a!., 1994) and ER (Beranger et al., 1991b). In summary, lipid modification targets H-Ras, N-Ras, Rapi and Rap2 GTPases to both the plasma membrane and Golgi, K-Ras4B to only the plasma membrane, while Rapl is targeted to endosomes and Rap2 is targeted to the ER. Thus, Ras and Rap GTPases can be found localized at diverse membrane compartments.  6  membrane  Figure 1.3  H-Ras and Rapi GTPase membrane localization  H-Ras GTPase is targeted to membranes via farnesylation of the C-terminus, and palmitolaytion (double) of the hypervariable region. Rapi GTPase is geranylgeranylated at its C-terminus and has a polybasic stretch of amino acids that interact with anionic phospholipids present in membranes.  1.2.3  Activation of Ras and Rap GTPases at specific membrane locations The plasma membrane compartment was thought to be the definitive site for Ras and  Rap activation due to the proximity of surface receptors capable of activating these GTPases. The advent of a method which detects activated Ras or Rap directly, by fluorescent microscopy, has allowed in depth analysis of where Ras and Rap protein activation really occurs. The Ras Binding Domain of Raf-1 (RafRBD) selectively binds activated GTP-bound Ras with an affinity  fold higher than for inactive GDP-bound Ras (Taylor et al., 2001).  Philips’ group created a green fluorescent protein (GFP) fusion protein with RafRBD (GFP RafRBD) which they expressed in COS-1 cells and observed its membrane distribution in response to epidermal growth factor (EGF) stimulation (Chiu et al., 2002). This probe for GTP-bound Ras is distributed equally throughout the cell in serum-starved, unstimulated cells (with some concentration at the Golgi). In response to EGF stimulation, GFP-RafRBD first translocates to the plasma membrane after 3-10 minutes, and then relocates to the Golgi 7  after 10-20 minutes (Chiu et aL, 2002). Using the same probe, Philips’ group went on to discover that in response to T-cell receptor (TCR) stimulation of Jurkat T-cells, membrane specific Ras activation occurs exclusively at the Golgi (Bivona et al., 2003; Perez de Castro et al., 2004). This finding was further dissected by the same group using fluorescence resonance energy transfer (FRET) microscopy experiments co-expressing cyan fluorescent protein (CFP)-tagged Ras proteins with yellow fluorescent protein (YFP)-tagged RafRBD, in Jurkat T-cells and primary human T-cells. When CFP is in close proximity (less than 5 nm apart) to YFP and upon excitation of CFP at 436 nm (which would otherwise not excite YFP), a transfer of energy occurs from CFP (donor) to YFP (acceptor), leading to an emission from YFP at 535 nm, indicating co-localization of CFP and YFP. This study revealed that while H-Ras and N-Ras are localized at both plasma membrane and Golgi, they are only activated on the Golgi in response to TCR stimulation (Mor et al., 2007). In addition, under the same conditions, K-Ras localizes to and remains inactive at the plasma membrane. Remarkably, robust H-Ras, N-Ras and K-Ras activation at the plasma membrane was found to occur when cells were co-stimulated with TCR and the integrin molecule intracellular adhesion molecule-i (ICAM-1), which binds and activates the integrin molecule lymphocyte-function-associated adhesion-l (LFA-1). Co-stimulation also led to activation of H-Ras and N-Ras, but not K-Ras at the Golgi (Mor et al., 2007). Ras activation has also been demonstrated to occur at other membranes, for example on endosomes (Jiang and Sorkin, 2002) and the ER (Chiu et al., 2002). To summarize: specific signals can lead to membranespecific Ras activation; for example, H-Ras is localized at both the plasma membrane and Golgi, but is only activated on the Golgi in response to TCR stimulation. Therefore membrane localization of Ras is not the only determinant that specifies activation. Work done in Jurkat T-cells showed that RNA silencing of CAPRI, a plasma membrane-localized negative regulator of Ras, led to H-Ras activation at the plasma membrane in response to antigen activation of the TCR (Bivona et al., 2003). Thus specific membrane-targeted negative regulators of Ras are able to downregulate activation of Ras at particular membranes. The same may also be true for positive regulators of Ras.  Although little is known about membrane-specific activation of Rap2, there is now considerable information about where Rapi activation occurs, particularly in T-cells. 8  Analysis of Jurkat T-cells co-expressing fluorescently-tagged Rap 1 and a fluorescently tagged probe for activated Rap GTPases, the Rap binding domain of Ra1GDS (Ra1GDSRBD), revealed that despite the fact that expression of GFP-Rapl was strong on intracellular membranes and weak on the plasma membrane, activation is only observed at the plasma membrane in cells grown in serum (Bivona et a!., 2004). In addition, Rapi is activated only at the plasma membrane of Jurkat T-cells in response to TCR stimulation (Bivona et a!., 2004). Rapi activation at the plasma membrane in response to TCR stimulation has been confirmed with similar experiments in other T-cells lines, although internal membrane activation of Rap 1 was not detectable in the method of microscopy used (Medeiros et al., 2005). While Rapi activation seems to only occur at the plasma membrane in response to TCR stimulation, disruption of cytosolic adapter protein complex of SKAP55 (Src kinase-associated phosphoprotein of 55 kDa) and ADAP (adhesion and degranulation promoting adapter protein) led to abrogation of Rapi localization at the plasma membrane after TCR ligation, without affecting the overall activation of Rapl (Kliche et al., 2006). These results suggest that the ADAP/SKAP55 complex is involved in recruiting activated Rapl to the plasma membrane. Therefore the possibility remains that Rapl could be activated elsewhere, e.g. at internal membranes and could subsequently translocate to the plasma membrane. There is also evidence of Rapi activation at endosomes (Hisata et al., 2007; Wu et a!., 2001; Wu et a!., 2007), which may represent a pool of activated Rap that could play a role at this location, or be transported to the plasma membrane. Taken together, these studies show that Rapi is similar to Ras GTPases in so much that activation occurs at selective membranes in response to signaling and that predominant sites of activation do not necessarily correlate with the predominant sites of steady state localization. This gives rise to the hypothesis that membrane targeted regulators are in part responsible for this pattern of activation. Once active, Ras and Rap GTPases can activate their target effector proteins, and thus propagate signal transduction pathways.  9  1.3  SIGNAL TRANSDUCTION FROM RAS AND RAP GTPASES  1.3.1  Ras activation of the Raf-ERK pathway Signals transduced by Ras activation stimulate several effector pathways which  critically regulate downstream cellular responses. The best characterized effector pathway activated by Ras signaling is the Raf-ERK (extracellular signal-regulated kinase) pathway [reviewed in (Leicht et al., 2007; Wellbrock et al., 2004)]. Rafphosphorylates and activates mitogen activated protein kinase (MAPK)/ERK kinase (MEK)1 and/or 2. Active MEK1/2 phosphorylates ERK1 and/or 2 (ERK1/2), which translocate to the nucleus to phosphorylate Ets family transcription factors, for example Elk-i. Raf knockout and transgenic mouse models as well as cell culture studies have revealed a role for the Raf- ERK pathway in a number of cellular processes such as proliferation, differentiation, survival and apoptosis [reviewed in (Galmiche and Fueller, 2007; Leicht et al., 2007)]. There are three Raf kinases, A-Raf, B-Raf and C-Raf, and knockout studies have shown that these Raf isozymes may have specialized function in normal cell processes [reviewed in (O’Neill and Kolch, 2004)]. For example, B-Raf may be the most potent promoter of proliferation as a MEK1/2 activator, while C-Raf may protect against apoptosis in a MEK1/2 independent manner (Huser et al., 2001; Mikula et al., 2001). H-Ras, N-Ras and K-Ras4B have differences in their binding affinities for Raf isozymes, which presents a potential for differential regulation (Weber et al., 2000).  1.3.2  Ras activation of the Raf-ERK pathway at specific membranes In its GTP bound form, Ras binds to the Ras binding domain of Raf (RafRBD),  thereby recruiting Raf from the cytosol to membranes (Figure 1.4). It is widely believed that the ability of Raf to activate MEK1/2 and then ERK1/2 is dependent on membrane localization; Hancock’s group recently found that a constitutively activated form of Raf which is unable to bind Ras and remains in the cytosol, is incapable of activating the ERK pathway (Tian et al., 2007). Membrane localization of Raf and an auxiliary scaffold protein; kinase suppressor of Ras (KSR), facilitates the binding of MEK1/2, and this complex provides a platform for ERK1/2 binding to MEK1/2. 10  Figure 1.4  Ras signaling to the Raf-ERK pathway  When localized to membranes, GTP loaded H-Ras binds to the Ras binding domain (RBD) of Raf kinases (A-Raf, B-Raf and C-Raf), thus localizing these effectors to membranes where Ras is bound and activated. When active, Raf kinases can bind to and phosphorylate MEK1/2. KSR stabilizes the Raf:MEK:ERK complex. Activated ERKII2 translocates to the nucleus to activate (phosphorylate) Ets transcription factors.  Membrane specific localization of Ras GTPases may serve to modulate Raf-ERK pathway outputs. Restriction of Ras to different internal membranes has been shown to affect Raf-ERK pathway activation (Chiu et al., 2002). Golgi targeting of a constitutively activated form of H-Ras (H-Ras6lL), leads to strong activation of ERKi/2, whereas ER targeting results in weak ERK1/2 activation (Chiu et a!., 2002). One explanation for this is that different membranes are sites of different levels and functional modes of regulators of the Ras to ERK pathway, such as levels of MEK1/2 or ERK1/2 deactivating phosphatases. The cytosol has increased levels of MEK1/2 and ERK1/2 dephosphorylation compared to the plasma membrane (Harding et a!., 2005), and the same may be true when comparing membrane sites.  Membrane specific restriction of Ras activation has also been shown to affect other Ras effector signaling pathways. Golgi targeting of an activated H-Ras, also gives rise to 11  strong activation of the phosphatidylinositol 3-kinase (P13K) pathways, while ER targeting leads to weak P13K pathway activation (Chiu et a!., 2002). Thus, signaling output from Ras activation of effector pathways can be dependent on which membrane these signaling events occur. Because of this, regulatory events that determine selective membrane activation of Ras have the potential to affect cellular responses to Ras-activating signals. 1.3.3  Rapi activation and inhibition of the Raf-ERK pathway Rapi also couples to the Raf-ERK pathway. Rapi has an identical effector domain to  Ras, and is able to bind Rafkinases. Rapi binds and activates B-Raf, leading to downstream ERK1/2 activation in response to a number of extracellular signals (Vossler et al., 1997; York et al., 1998). Paradoxically, several groups showed that ectopic Rapi expression can interfere with Ras-mediated ERK1/2 activation in vitro (Boussiotis et al., 1997; Cook et a!., 1993; Dillon et al., 2005; Mochizuki et a!., 1999). These opposing effects appear to be due to differences in the interaction strength between Rap 1 and the two Raf cysteine rich domains (CRD) (Okada et a!., 1999). The interaction with Ras with Raf proteins is mediated by the Ras binding domain (RBD) and CRD of Raf, and binding of Ras to both regions is required for full Raf activation. In the case of B-Raf, Rapi binds CRD with a similar affinity as does Ras, and is also able to activate this Raf. B-Raf activation can thus then be carried out by both Ras and Rap 1. However, Rap 1 binds the C-Raf CRD with much higher affinity than Ras (Okada et al., 1999), and thereby inhibits Ras binding to C-Raf. In addition, Rapl is unable to activate C-Raf and so Rap 1 binding to the CRD of C-Raf inhibits its activation. While B-Raf is selectively expressed in neuronal and hematopoietic cells (Eychene et al., 1995), C-Raf is ubiquitously expressed. Therefore the ability of Rap 1 to activate or inactivate the Raf-ERK pathway is thought to depend on which Raf it interacts with, and thus is cell type specific.  Further insight into how Rapi differentially regulates the Raf-ERK pathway in different cell types came from experiments with mice deficient for SPA-i (signal-induced proliferation associated gene 1), a GTPase activating protein (GAP), which negatively regulates Rap. SPA’ mice develop latent mye!oproliferative stem-cell disorders, and a majority of these mice developed aggressive lethal leukemia with abundant blast cells of  12  either a myeloid or an erythroid lineage (Ishida et al., 2003a). Preleukemic bone marrow cells (BMCs) demonstrated increased Rap 1 GTP levels as well as constitutive ERK activation, but not Ras activation. Given that B-Raf is expressed in myeloid precursors and plays a pivotal role in myelopoiesis (Kamata et al., 2005), deregulated Rapi activation of ERK may occur via B-Raf in preleukemic bone marrow cells (BMCs), and this may in part lead to an enhanced expansion of SPA’ hematopoietic progenitors. During blast crisis however, Rap 1 GTP, RasGTP and phosphorylated ERK levels are all greatly increased. Reconstitution of a cell line generated from a SPA’ mouse with CML in crisis with wild type SPA-i, reduced Rapi activation but not RasGTP or phosphorylated ERK levels, suggesting that Rapi activation and Ras-ERK1/2 activation are independent at this stage of disease. Another interesting finding was that SPA’ mice exhibit an age-dependent progression of T-cell unresponsiveness (Ishida et al., 2003b). In this study, stimulation of T cells from SPA-/- mice lead to persistent Rap 1 activation, but showed defective Ras-ERK 1/2 activation. For the most part T-cells do not express B-Raf (Kometani et al., 2004) and TCR induced anergy (an irreversible inactive state resulting from lack of co-receptor signaling during T-cell receptor stimulation) is prevented in cells ectopically expressing B-Raf (Dillon et al., 2003). The T-cell anergy observed in SPA’ mice is likely to result from TCR-induced Rapi inhibition of C-Raf-ERK1/2 signaling and these results are supported by earlier findings from several in vitro studies indicating that Rapi inhibition of the Raf-ERK pathway plays a role in T-cell anergy (Boussiotis et al., i997; Reedquist and Bos, 1998). Of further interest is the possibility that the resultant T-cell immunodeficiency may lead to an impaired immune surveillance which could be linked to the onset of the overt myeloid leukemia in SPA-F’ (Kometani et al., 2004).  1.3.4  Rapi activation of the Raf-ERK pathway at specific membranes Recent research has shown that Rap 1 activation at the plasma membrane results in  robust ERK1/2 activation, whereas activation of Rapi at internal membranes is incapable of activating ERK1/2 (Wang et al., 2006). This suggests that like Ras, membrane specific activation of Rap has consequences with respect to downstream effector activation. Thus, like Ras, signaling output can be determined by membrane compartmentalization of Rap  13  activation and membrane localized regulators of Rap GTPases are possible candidates that can achieve this signaling segregation.  1.4  RAP GTPASES AND ADHESION  1.4.1  Role of Rap GTPases and integrin mediated adhesion While Rapi has received attention for its role in activation and inhibition of the Raf  ERK pathway, this GTPase has also been found to play a central role in a number of processes regulating cell adhesion, including integrin mediated cell adhesion and cadherin mediated cell-cell junction formation [reviewed in (Bos, 2005; Kinashi, 2005; Kooistra et al., 2007)]. While their role is less defmed, Rap2 GTPases are believed to play a role in lymphocyte cell migration (McLeod et al., 2002), and integrin mediated adhesion in B lymphocytes (McLeod et al., 2004) and neutrophils (Jenei et al., 2006).  Integrins are cell surface receptors that mediate adhesion of cells to other cells or the extra-cellular matrix (ECM). Lymphocyte stimulation by antigen receptors (e.g. T-cell receptor or B-cell receptor) or chemokine receptors (e.g. CXCR4) leads to integrin activation by a process called inside-out signaling, which ultimately changes both the structural conformation of the integrin molecule from a low to a high affinity state, and increases the integrin valency (also known as clustering or density). Together, these two outcomes of inside-out signaling result in increased avidity of integrins for their adhesion molecule counterparts expressed on the surface of other cells or present in the ECM [reviewed in (Kinashi, 2005)].  In lymphocytes, Rapi is involved in the regulation of several inter-related integrin mediated adhesion processes including migration, cell spreading, cell polarity and immune synapse formation [reviewed in (Bos et a!., 2001)]. Expression of a constitutively active mutant of Rapi, RaplVl2 (in which Rapi has a valine residue at position 12), was shown to increase adhesion of lymphocytes, by augmenting the avidity of the very late antigen-4 (VLA-4) integrin molecule for its integrin counterpart molecule, vascular cell adhesion molecule-1(VCAM-1) (Sebzda et a!., 2002). 14  Furthermore, as previously mentioned with Ras GTPases, activated Rapi increases the affinity of binding of the integrin molecule LFA-1 to its ligand, intercellular adhesion molecule-i (ICAM- 1) and induces the conformational changes that are associated with increased affinity (Katagiri et al., 2000; Reedquist and Bos, 1998). RaplA and RapiB have been shown to have different roles in integrin mediated adhesion. RaplA knockout mice have mild defects in T- and B-cell integrin mediated cell adhesion (Duchniewicz et al., 2006) whereas Rap 1 B deficient mice exhibit reduced integrin mediated platelet adhesion in response to agonists (Chrzanowska-Wodnicka et al., 2005) and have also been shown to have defects in B-cell adhesion and chemotaxis (Chen et al., 2008).  1.4.2  Role of Rapi in antigen receptor induced integrin mediated adhesion T-cells detect antigen by forming immune synapses with antigen presenting cells  (APCs). Immune synapses are facilitated by interactions between T-cell receptor (TCR) and antigen bound to major histocompatability complex (MHC), and are dependent on adhesive interactions between integrin molecules (e.g. LFA-1, VLA-1) and their counterpart integrin molecules (e.g. ICAM-1, VCAM-i). TCR stimulation leads to integrin inside-out stimulation, which ultimately increases adhesion of integrin molecules to their counterparts, and thus intensifies the T-cell:APC interaction. Rapi has been shown to positively regulate LFA-1 mediated adhesion to ICAM-1 in response to TCR stimulation [reviewed in (Menasche et al., 2007b)]. Rapi is activated at the plasma membrane in response to TCR antigen ligation (Bivona et al., 2004) and expression of SPA-i abrogates TCR mediated Rapi activation, and LFA-1 mediated adhesion to ICAM-1 (Katagiri et al., 2002). Following TCR stimulation, active Rap 1 associates with RapL, which localizes to small vesicles near perinuclear regions. RapL then forms a complex with LFA-1 and initiates LFA-l translocation from perinuclear regions to the peripheral boundaries of the immonuolgical synapse (Katagiri et al., 2003). Rapi also regulates RIAM (Rapl-interacting adaptor molecule) which interacts with actin regulating proteins such as talin and the VASP (vasodilator-stimulated phosphoprotein) family. Recently, RIAM has been shown to interact with the ADAP/SKAP55 module, and this complex acts as a scaffold to recruit Rapi to the plasma membrane in response to TCR, in order to facilitate adhesion (Menasche et al.,  15  2007a). Other regulators of the actin cytoskeleton that are required for adhesion include the Vav GEF family, and Vavi and Vav2 have recently been shown to regulate B-cell receptordependent adhesion, immune synapse formation and activation of Rapi (Arana et al., 2008). In a complementary study, using overexpression of the Rapi and Rap2 negative regulator RapGAP, Rap activation was shown to be required for the B-cell immune synapse formation, the formation of F-actin-rich sites at locations of B-cell interaction with antigen and also BCR mediated cell spreading (Lin et al., 2008).  1.4.3  Role of Rapi in chemokine induced integrin mediated adhesion Leukocytes emigrate from blood venules into inflamed tissues in a process dependent  on adhesion. Prior to emigration, rolling lymphocytes are induced to tether to high endothelial venules (HEVs) via increased LFA-l adhesion. Chemokines localized to the apical surface of the epithelium, or in soluble form (e.g. stromal cell-derived factor-i, SDF 1 a), bind to the chemokine receptors (G-protein coupled receptors) on rolling lymphocytes, and stimulate a cascade of signaling events that lead to an increase in the avidity of LFA- i, and thus adhesion to ICAM-1, which acts to stop the rolling lymphocyte in its tracks. Rapi is activated in response to the chemokine SDF-icL and inhibition of Rapi activation in lymphocytes by ectopic SPA-i (a RapGAP) expression eliminates SDF-lcL induced adhesion of LFA-1 to ICAM-i (Shimonaka et al., 2003). Recently, basal and SDF-ia-stimulated activation of Rapi was discovered to be defective in Epstein-Barr-virus-transformed lymphocytes derived from a patient with leukocyte adhesion deficiency (LAD) (Kinashi and Katagiri, 2004). LAD patients are characterized by defective integrin-mediated adhesion of leukocytes, and are thus immunodeficient. In other LAD patients, defective adhesion can be explained by mutations in integrin molecules, but this and other similar cases (Alon et al., 2003) express normal integrins. Therefore, these cases are categorized as LAD-Ill, since inside-out signalling is impaired rather than the integrins themselves. RapL also plays a role in chemokine-induced adhesion via Rap i. RapL associates with Rap 1 in response to SDF- 1 a stimulation and RapL deficient T- and B-cells are defective in chemokine induced adhesion, which is dependent on LFA-l (Katagiri et al., 2004).  16  1.4.4  Role of Rapi in cadherin mediated adhesion Rap 1 is also an important regulator of cadherin based cell-cell junction formation  [reviewed in (Kooistra et al., 2007)]. Cadherins are cell-cell adhesion molecules which mediate cell-cell adhesion through homophilic interactions at the cell surface. The cytoplasmic tails of cadherins bind to several proteins to form a connection with the actin cytoskeleton. Studies of the conditional rap] deficient mutant in Drosophila melanogaster first demonstrated a role for Rapi GTPase in cadherin regulated cell-cell adhesion (Knox and Brown, 2002). Cell clones that are rap] -deficient in the wing have a different morphology, disperse into the surrounding tissue, and have an uneven distribution of DE-cadherin (Drosophila endothelial-cadhein), all indicators of defective cell-cell junction formation. The atypical Rap activator DOCK4, a potent tumor suppressor, was shown to be a regulator of cell-cell junctions. An osteocarcinoma cell line lacking DOCK4 was found to lack cell-cell junctions, and reformed on the transduction of wild-type DOCK4 or an active form of Rapi (Yajnik et al., 2003). Further studies showed that adhesion of ovarian carcinoma cells to E cadherin (endothelial-cadherin) is inhibited by expression of a dominant negative form of Rap (Price et al., 2004) and in epithelial cells ligation of E-cadherin leads to Rapi activation, which is itself required for appropriate targeting of E-cadherin to maturing cell-cell contacts (Hogan et al., 2004). Rapi has also been shown to play a role in the regulation of VE cadherin (vascular endothelial-cadherin) in endothelial cells, via the cyclic adenosine monophosphate (cAMP) -responsive Rap activator Epaci (Kooistra et al., 2005). In the regulation of E- and VE-cadherin-mediated-adhesion, Rapi effectors are thought to recruit junctional proteins to sites of developing cell-cell contacts in order to stabilize the association between actin cytoskeleton and the junctional complex (Glading et al., 2007; Hoshino et al., 2005; Kooistra et al., 2007).  1.5  GUANINE NUCLEOTIDE EXCHANGE FACTORS: POSITIVE REGULATORS OF RAS AND RAP  1.5.1  Regulation of Ras and Rap GTPases by membrane localization of GEFs Ras subfamily GTPases are activated via the exchange of GDP for GTP. Ras  activation requires guanine nucleotide exchange factors (GEF5) to promote the release of 17  GDP and thereby enable binding of GTP, which is at a higher cellular concentration. All GEFs contain a homologous catalytic domain which is known as a GEF domain. As positive regulators of Ras and Rap, GEFs counteract negative regulation by GTPase activating proteins (GAPs), which stimulate the intrinsic GTPase activity of Ras and Rap GTPases (Figure 1.5).  Ras and Rap GTPases are activated by distinct GEFs. Ras GTPases are activated by three GEF protein families; Son-of-sevenless (Sos 1 and 2), Ras guanine releasing factor (RasGRF1 and 2) and Ras guanine nucleotide releasing protein (Ra5GRP1, 2, 3 and 4). Sos is the most well characterized RasGEF family; both Sosi and 2 are ubiquitously expressed and are localized in the cytosol of resting cells (reviewed in Mor and Philips, 2006). In response to growth factor stimulation commonly via receptor tyrosine kinases (RTKs) -  -  these GEFs translocate to the plasma membrane, via interaction with the adapter protein Grb2 and E3b1 [reviewed in (Nimnual and Bar-Sagi, 2002)] and via localization through binding of the pleckstrin homology (PH) domain of Sos 1/2 to phosphatidic acid (PA), which is generated by phopholipase D 2 (PLD2) [(Zhao et al., 2007); reviewed in (Hancock, 2007)]  In contrast to Sos 1/2, RasGRF 1 is constitutively localized at membranes via interactions with its N-terminal PH domain. RasGRFs are activated through calcium signaling, via the coordinated interaction of its PH, coiled coil and calcium-dependent calmodulin-binding IQ domains (Buchsbaum et al., 1996). Of the RasGRP family members, RasGRP 1, 3, 4 and a human splice variant of RasGRP2 have been shown to activate Ras GTPases. RasGRPs are also known as Ca1DAG-GEFs, since as a family they are believed to be regulated by the second messengers calcium and diacyiglycerol (DAG). Membrane localization and activation of RasGRP family members will be discussed in detail in the sections that follow.  18  Guanine nucleotide Exchange Factor (GEF) [GEF  I GTP  GDP  In active  Active  GDP  Pi  GTPase Activating Protein (GAP)  Figure 1.5  GEF and GAP regulation of Ras GTPases  Guanine nucleotide exchange factors (GEFs) activate Ras GTPase by releasing GDP, and allowing the binding of GTP, which is in greater concentration (lOX) in the cell. GTPase activating proteins (GAPs), stimulate the intrinsic GTPase activity of Ras GTPases (GAPs).  19  Rap GTPases are regulated by a variety of distinct GEF proteins that have different membrane targeting strategies [reviewed in (Stork, 2003)1. In general, all Rap GEFs appear to have specificity for both Rapi and Rap2 (Ohba et al., 2000b). These Rap GEFs include C3G, Epacs/cAMP-GEFs (Epaci and 2), PDZ-GEFs (PDZ-GEF1 and 2, MR-GEF) and RasGRPs (RasGRP2 and 3). C3G (Crk SH3 domain guanine nucleotide exchanger), the first Rap GEF discovered (Gotoh et al., 1995), is recruited to the plasma membrane from the cytosol through binding to the SH2 domain of the adapter protein Crk in response to receptor tyrosine kinase stimulation (Feller, 2001; Okada et al., 1998). C3G may also be recruited to the plasma membrane through the action of cAMP-dependent protein kinase A (P1(A) (Radha et al., 2004; Schmitt and Stork, 2002; Wang et al., 2006). Epacs (Exchange proteins directly activated by cAMP) are also known as cAMP-GEFs, and as their name implies are activated through direct interaction with the second messenger cAMP (de Rooij et al., 1998; Kawasaki et al., 1998b).  Epaci and 2 are localized to perinuclear regions, the nuclear membrane and mitochondria and their localization is thought to be regulated during the cell cycle (Qiao et al., 2002). PDZ-GEFs contain a PSD-95/D1gA!ZO-1 (PDZ) domain, which regulates membrane localization [reviewed in (Ehrhardt et al., 2002)1. PDZ-GEF2 and MR-GEF also contain Ras-associating domains that specifically bind to M-Ras (another Ras subfamily GTPase) at the plasma membrane, which serves to localize these GEFs at locations where they can activate Rap GTPases (Gao et al., 2001). As already mentioned RasGRP2 activates Rap GTPases and RasGRP3 activates both Ras and Rap GTPases, and as such acts as a dual Ras/Rap activator. RasGRPs are regulated by membrane localization mechanisms that bring them in close proximity to their substrates (either Ras or Rap GTPases). RasGRPs are thought to be localized at membranes through binding of their Cl domain with DAG which is produced in membranes by phospholipase Cs which are activated in response to tyrosine kinase and antigen receptor stimulation. I will next discuss the structure, function and regulation of RasGRPs in more detail.  20  1.6  RASGRP FAMILY OF GEFS  1.6.1  RasGRP family protein structure The RasGRP family has four members, RasGRP1, 2, 3 and 4 (Stone, 2006). They  display a high degree of conservation, with RasGRP 1 and RasGRP3 exhibiting the greatest similarity (Yamashita et al., 2000). All RasGRPs have a Ras exchange motif (REM) and a GEF domain (Figure 1.6) that together constitute the catalytic region responsible for GDP to GTP exchange on Ras and Rap GTPases. In addition to the catalytic region, RasGRPs contain either one or two EF hands. EF hands are calcium binding domains, and RasGRPs are thought to be regulated by calcium signaling, although minimal direct evidence exists for this hypothesis. RasGRPs also have a Cl domain that is homologous to protein kinase C (PKC) Cl domains that bind lipid second messenger diacylglycerol [reviewed in (ColonGonzalez and Kazanietz, 2006)]. Membrane localization of RasGRP1 is also regulated by its C-terminal plasma membrane targeter (PT) domain (Beaulieu et al., 2007).  EF EF  [  GEF  ]—OO----()  —[REM)  I  GEF  1—OO—(3)  RasGRP3  —[REM]  [  GEE  RasGRP4(x  -JEMj  [GE]  RasGRP4I3  HREM]  [GEE j  RasG R P1  —fRErvJ  RasGRP2  Q—CJf five amino acid insertion  Figure 1.6  RasGRP protein structure  Domain structure of RasGRP proteins. REM, Ras exchange motif which is important for Ras activation. GEE, guanine nucleotide exchange domain that catalyzes GTP loading of Ras and Rap GTPases. EF, EF-hands. PT, plasma membrane targeting domain. Ra5GRP4I3 has a five amino acid insertion in the Cl domain.  21  1.6.2  RasGRP expression and function RasGRPs are expressed predominantly in the hematopoietic system and brain with  some degree of specificity (Clyde-Smith et al., 2000; Dupuy et al., 2001; Kawasaki et al., 1 998a; Yamashita et al., 2000). In the hematopoietic system, there seems to be a certain level of restricted expression of RasGRP family members; RasGRP 1 in T-cells (Ebinu et al., 1998), RasGRP2 in platelets (Crittenden et al., 2004), and T-cells (Ghandour et al., 2007), RasGRP1 (Coughlin et al., 2005) and 3 (Teixeira et al., 2003) in B-cells, and RasGRP1 (Liu et al., 2007) and 4 (Yang et al., 2002) in mast cells.  Knockout studies revealed that RasGRP1 is important for T-cell development (Dower et al., 2000; Priatel et al., 2002) and activation of both B-cells (Coughlin et al., 2005) and mast cells (Liu et al., 2007). RasGRP1 couples lymphocyte antigen receptors to Ras activation (Stone, 2006). Deregulated expression of RasGRP1 leads to lymphoma and leukemia (Dupuy et al., 2005; Klinger et al., 2005; Li et al., 1999; Mikkers et al., 2002; Suzuki et al., 2002), transformation of fibroblasts (Tognon et al., 1998) and keratinocytes (Oki-Idouchi and Lorenzo, 2007).  RasGRP2 has been shown to couple antigen receptor stimulation of T-cells with downstream Rapi activation (Katagiri et al., 2004). Deletion of RasGRP2 in mice by Graybiel’s group was shown to lead to defects in integrin dependent platelet aggregation (Crittenden et al., 2004) and leukocyte integrin mediated adhesion (Bergmeier et al., 2007). Wagner’s group identified RasGRP2 deficient mice as a model for LAD-Ill (Bergmeier et al., 2007), and this was confirmed when two new cases of LAD-Ill, in which T lymphocytes, neutrophils and platelets have severe defects in integrin activation, were found to have defective RasGRP2 expression (Pasvolsky et al., 2007). Further work confirming the role of RasGRP2 and human T-cell adhesion showed an essential role for RasGRP2 in LFA-1 but not VLA-4 intergin mediated cell adhesion (Ghandour et al., 2007). Abberant expression of RasGRP2 leads to B lymphocyte (Mikkers et al., 2002; Suzuki et al., 2002) and myeloid transformation (Dupuy et al., 2001), which may be due to Rapi activation of ERK1/2 via B Raf. No reports have been made of B-cell specific adhesion defects in either LAD III patients or RasGRP2 knockout mice. 22  Less is known about the function of RasGRP3 and RasGRP4. RasGRP3 also couples antigen receptor signaling to Ras activation (Coughlin et al., 2005; Oh-hora et a!., 2003) and RasGRP3 deficient DT4O B-cells have defects in B-cell activation (Oh-hora et al., 2003). RasGRP3 knockout mice exhibit hypogammaglobulinemia (disorder caused by a lack of Blymphocytes) and with some evidence of splenomegaly (enlargement of the spleen) or autoimmunity (Coughlin et al., 2005). RasGRP4 was originally identified in a cDNA screen from acute myeloid leukemia (AML) patients (Reuther et al., 2002) and is thought to play a role in mast cell development (Stevens et al., 2005).  1.6.3  RasGRP GTPase specificity RasGRPs have differential specificities for Ras versus Rap (Table 1.6.3). RasGRP1  has been shown to have robust activity towards Ras proteins including H-Ras, N-Ras, K-Ras, M-Ras, R-Ras and TC-21 (Ebinu et al., 2000; Kawasaki et a!., 1998a; Mochizuki et al., 2000; Ohba et al., 2000a; Yamashita et al., 2000). RasGRP2 activates Rapl and Rap2 (Dupuy et a!., 2001; Eto et al., 2002; Katagiri et al., 2004; Kawasaki et al., 1998a; Ohba et al., 2000b; Ohba et al., 2000a) and has also been reported to activate Ras proteins, including R-Ras, N Ras and TC-21 (Clyde-Smith et al., 2000; Ohba et al., 2000a; Yamashita et al., 2000). In addition, a myristoylatable, longer splice-variant of RasGRP2 (MyrRa5GRP2) that is expressed in humans, has been shown to have moderate exchange activity towards N-Ras (Clyde-Smith et a!., 2000). No further evidence of RasGRP2-mediated Ras activation has come to light. RasGRP3 has been shown to promote activation of H-Ras, N-Ras, R-Ras, M Ras, TC-21 as well as Rapi and Rap2 GTPases (Ohba et a!., 2000b; Ohba et aL, 2000a; Yamashita et aL, 2000). RasGRP4 is reminiscent of RasGRP1 in its ability to activate H-Ras (Reuther et al., 2002; Yang et al., 2002).  23  RaslRap GTPase specificity RasGRPI  H-Ras, N-Ras, K-Ras, M-Ras, R-Ras, TC-21  RasGRP2  Rapi, Rap2, R-Ras, N-Ras, TC-21  RasGRP3  H-Ras, N-Ras, R-Ras, M-Ras, TC-21, Rapi, Rap2  RasGRP4  H-Ras  Table 1  GTPase specificities of RasGRPs  RasGRPs have different effector specificity. RasGRPI, 3 and 4 can activate Ras GTPases, although a human splice variant of RasGRP2 has been shown to activate R-Ras, N-Ras and TC-21. RasGRP2 can activate Rap GTPases and RasGRP3 has both Ras and Rap GTPase specificity.  1.7  REGULATION OF RASGRPS BY Cl DOMAINS  1.7.1  Membrane binding by Cl domains Regulation of protein-membrane binding is governed by a variety of protein domains  including Cl, C2, PH and FYVE and PX domains, which have affinity for specific lipids at membranes [reviewed in (Cho and Stahelin, 2005)]. Protein kinase C (PKC) conserved 1 (Cl) domains were first discovered in novel (PKCö,  ,  i,  9) and classical (PKCa, j3, ‘y)  isozymes, where they occur in tandem and are referred to as Cia and Cib. Cl domains are approximately 50 amino acids in length and are classified as typical or atypical depending on their ability to bind the ligand diacylglycerol (DAG). DAG is produced and maintained by several processes in the cell, including lipid metabolism which results in the synthesis of DAG de novo, and alternatively, external signals that activate membrane associated protein lipases and phosphatases which convert lipid precursors into DAG [reviewed in (Carrasco and Merida, 2006)1. One such canonical pathway involves the activation of phospholipase C (PLC) isozymes via receptor stimulation (e.g. T-cell receptor signaling), which cleaves the phospholipid phosphatidylinositol 4,5-bisphosphate (PIP2) into two products, DAG and inositol trisphosphate (1P3) (Figure 1.7). In addition, DAG is produced via the dephosphorylation of phosphatidic acid (PA) by phosphatidic-acid phosphatase (PAP). The  24  PA substrate for this reaction can also be generated during receptor tyrosine kinase stimulation through phospholipase D (PLD) cleavage of the phospholipid phosphatidyl serine. DAG that is generated and concentrated at membranes as a result of external signals acts as a ligand for Cl domain containing proteins. In this way this DAG facilitates the translocation of such proteins to these membrane locations. Most studies characterizing typical and atypical Cl domains have used the ligand phorbol ester (PE), a functional DAG analog which is not naturally present in cells, but for which many typical Cl domains have very high affinity. For the most part typical Cl domains have higher affinity for phorbol esters than for DAG, although in some instances they have greater affinity for DAG [reviewed in (Cho and Stahelin, 2005)1.  Studies in the 1 990s using PKC inhibitors that target the DAG/PE binding site of Cl domains (e.g. caiphostin C), as well as treatment with exogenous PE, revealed a large number of pathways and cellular responses that at the time were thought to be solely dependent on novel and classical PKCs. This perspective dramatically changed with the discovery of five other protein families, which also contain typical DAG-responsive Cl domains: protein kinase D (PKD), DAG kinase (DGK), Muncl3, chimaerins and RasGRPs [reviewed in (Brose et al., 2004)]. Pharmacological inhibitors of Cl domains and PE treatment can theoretically also negatively and positively regulate all proteins that contain a typical Cl domain; therefore their effects and specificities have been reanalysed in the context of a complex network of DAG responsive proteins [reviewed in (Kazanietz, 2002)]. Proteins have also been identified that have Cl domains that are not responsive to DAG, termed atypical Cl domains. These proteins include atypical PKCs, DGKs, Vav, Raf and kinase suppressor of Ras (KSR). In the case of KSR and Vav, while unresponsive to DAG/PE, the Cl domains within these proteins are required for their activity (see section 1.7.3).  25  Receptor  plasma membrane  1P3  Figure 1.7  DAG production and Cl domain localization at membranes  Receptor stimulation leads to phospholipase C (PLC) activation which cleaves phosphatidylinositol 4,5-bisphosphate (PIP2) into inositol triphosphate (1P3) and diacyiglyercerol (DAG). DAG is the ligand for Cl domains, which bind the headgroup of DAG via the DAG binding pocket. Basic residues (+) on either side of the Cl domain DAG binding pocket are thought to interact with the anionic phospholipids (-) present in the membrane via electrostatic interactions, thereby stabilizing membrane binding.  1.7.2  PKC ö Clb: the archetypal typical Cl domain NMR spectroscopy of the PKCö Cib Cl domain determined that two zinc cation  ) binding sites form during globular folding of the domain (Hommel et al., 1994). The 2 (Zn structure of this Cl domain) bound to the phorbol ester, phorbol 13-acetate (PMA), was subsequently characterized by co-crystallization (Zhang et al., 1995). These studies together with validation by site-directed mutagenesis have helped to develop an understanding of how a typical Cl domain folds and interacts with membranes containing DAG/PE [reviewed in (Cho and Stahelin, 2005; Colon-Gonzalez and Kazanietz, 2006)]. All typical Cl domains 26  1 67 112 1 . 12 Cx H 4 C 2 C 4 x x (where H have the consensus sequence Hx cysteine, x  =  =  histidine, C  =  other amino acid) (Figure 1.8), which folds to form a compact structure  featuring two small f3 sheets and a small C-terminal a helix. The two histidines and six cysteines in the consensus sequence create two cavities, that each coordinate a Zn 2 ion. The two f3 sheets form a narrow polar binding pocket with two loops (loop A and loop B), within which DAG/PE can insert lengthwise (Figure 1.9). In unbound form, the two f3 sheets are in a “zipped” conformation with hydrogen-bonded water molecules filling the binding pocket. Binding to phorbol ester leads to unzippering of the f3 sheets, allowing hydrogen bonds to form with positions C3, C4 and C20 of the phorbol ring with the “unzipped” groove. Basic residues (arginine, lysine or histidine) surrounding the binding pocket are involved in non specific electrostatic interactions with anionic phospholipids, which accelerate initial membrane adsorption of the Cl domain and position the Cl domain at the surface of the membrane. Hydrophobic and aromatic residues at the tips or base of the loops A and B, which surround the binding pocket, are thought to be involved in insertion and penetration of the membrane, which in turn is thought to be required to stabilize binding to membraneembedded DAG. Initial binding of the Cl to DAG or PE (in conjunction with these hydrophobic/aromatic residues) creates a contiguous hydrophobic binding surface which may facilitate subsequent insertion into the membrane. Residues in the PKC ö C lb sequence which were found to be essential for ligand binding including Phe3, Tyr8, Proll, Leu2O, Leu2l, Trp22, Leu24, Gln27 and Val38 [reviewed in (Colon-Gonzalez and Kazanietz, 2006)].  Loop A  Loop B  PKC Cib MP RFKVYNM,S P B2-chim HNFKvHTRG PH Raf-1 TT NF2RKTFIj KSR HRFSTKSWLS QV10  -  Figure 1.8  DH GSL LWGLVXQGLK EY F MWGLIAQGVR DI QKF LL-----NGFR H QKS MI-----FGVK 20 30  D GINQK MNVZJH K REKV2NL 0 LNVlK Q SKHVPND QT GYKFE KKR STKVPT KH RLKN -TKEAP 40 50  Typical and atypical Cl domain sequence comparison  Typcial Cl domains (PKCö Cib and B2-chimaerin) compared to atypical Cl domains (Raf-1 and kinase suppressor of Ras (KSR). Loop A and loop B (italicized) form a DAG binding pocket. Cationic residues which have the potential to interact with anionic phospholipids within membranes are bolded. While Raf-l and KSR, still have the consensus two histidines and six cysteines structure which coordinate zinc, they each have a deletion in loop B which precludes binding to DAG.  27  Figure 1.9  The Cib domain structure of PKCÔ  The Cib domain of PKCO (protein data bank code lptq) illustrates the generic structure of a Cl domain that binds DAG or phorbol ester. The zinc atoms are shown as orange balls. Y8, T12 and W22 are highlighted in yellow. Four of the cationic residues implicated in membrane interactions, K26, K30, K41 and K45, are highlighted in blue. The image was generated with PyMol and is also in Figure 2.1.  1.7.3  The function of atypical Cl domains Atypical Cl domains have several sequence deviations in loops A and B that render  them unable to bind DAG or PE. The structure of both C-Raf and KSR have been determined by spectroscopic NMR (Mott et al., 1996; Zhou et al., 2002). In both cases the Cl domains of these proteins lack a crucial proline residue (Pro 11) in loop A and have a deletion at position 22-25 that structurally removes ioop B [reviewed in (Colon-Gonzalez and Kazanietz, 2006)]. These changes predict an inflexible  f3 sheet conformation that is unable to  open to accommodate PE binding. In the case of the C-Raf Cl domain, the lack of this ioop leads to a flat hydrophobic surface, which would lack penetrating capabilities. In addition, both C-Raf and KSR lack Gln27, which is important for maintenance of unzipped 13 sheet structure of typical Cl domains. 28  Atypical Cl domains do still retain some hydrophobic residues thought to be required for membrane insertion, and basic residues that may promote anionic phospholipid electrostatic interactions. It has been suggested that for C-Raf and KSR these interactions are responsible for membrane binding (Ghosh et al., 1996; Zhou et al., 2002). KSR translocates to the plasma membrane in response to growth factor treatment, in a Cl domain-dependent manner (Michaud et al., 1997). Other evidence for a functional role of atypical Cl domains comes from studies which showed that the Cl domain of Vav is required for its ability to induce oncogenic transformation (Booden et al., 2002; Coppola et al., 1991). Also, mutations in the Cl domain of C-Raf were shown to affect C-Raf binding to and activation of Ras (Drugan et al., 1996; Winkler et a!., 1998).  1.7.4  Regulation of RasGRP1 localization by the Cl domain The RasGRP family of guanine nucleotide exchange factors (RasGRP 1-4) have a  single C-terminally located Cl domain. Much information about the functional role of the Cl domains of this family of proteins has been gleaned from research with RasGRP 1. The RasGRP1 Cl domain binds to DAG or PE (Ebinu et al., 1998; Irie et a!., 2004; Lorenzo et al., 2000; Madani et al., 2004; Rong et al., 2002), [reviewed in (Carrasco and Merida, 2006)] and RasGRP 1 translocates to membranes in response to DAG or PE treatment (Bivona et al., 2003; Caloca et al., 2003b; Ebinu et al., 1998; Rambaratsingh et al., 2003; Tognon et a!., 1998). RasGRP1 also activates Ras in response to DAG/PE (Caloca et al., 2004; Ebinu et a!., 1998; Kawasaki et al., 1998a; Priatel et al., 2002; Rambaratsingh et al., 2003; Tognon et al., 1998). The Cl domain is required for RasGRP1 translocation to endomembranes (endoplasmic reticulum (ER) and Golgi) in response to serum (Caloca et al., 2003b; Tognon et al., 1998) and is also required for RasGRP1 mediated NIH 3T3 transformation (Tognon et al., 1998). RasGRP1 translocates to the plasma membrane (Carrasco and Merida, 2006; Zugaza et al., 2004) or alternatively the Golgi (Bivona et al., 2003; Perez de Castro et al., 2004) in response to T-cell receptor (TCR) stimulation. Regulation of DAG levels by conversion to phosphatidic acid (PA) by DAG kinases has been proposed as a method for spatially restricting activation of RasGRP 1 in selective membrane compartments (Zha et al., 2006). RasGRP1 is found at the plasma membrane in response to  29  antigen presenting cell (APC) interactions (Daniels et al., 2006), and at the Golgi when T cells are anergized before APC engagement (Zha et a!., 2006). Recently it was discovered that RasGRP1 could translocate to both membrane compartments in the same cell (Mor et a!., 2007). Mor and colleagues found that TCR stimulation alone leads to RasGRP1 accumulation and subsequent Ras activation at the Golgi, whereas co-stimulation with ICAM- 1 (to crosslink LFA- 1) allowed RasGRP 1 translocation to and Ras activation at the plasma membrane (Mor et al., 2007). Strikingly, this group also determined that RasGRP1 translocation to the plasma membrane was dependent on phospholipase D 2 (PLD2) mediated generation of PA at this membrane compartment, and subsequent dephosphorylation by PA phosphatase (PAP) to produce DAG. RasGRP1 translocation to DAG at the Golgi however, relies on PLCy, which cleaves PIP2 to generate DAG and 1P3, a regulator of intracellular calcium levels (Mor et al., 2007). In DT4O B-cells RasGRP1 has also been shown to translocate to the plasma membrane in response to B-cell receptor (BCR) ligation, and this translocation is enhanced by the Cl domain (Beaulieu et a!., 2007). Other evidence has accumulated which points to a central role of the Cl domain in regulating RasGRP 1. The Cl domain is required for TCR mediated activation of RasGRP 1 (Roose et al., 2005) and TCR and BCR mediated RasGRP 1-dependent Ras activation requires PLOy (Beaulieu et al., 2007; Caloca et aL, 2003b; Ebinu et al., 1998; Perez de Castro et al., 2004; Reynolds et a!., 2004; Zugaza et al., 2004). RasGRP 1-mediated Ras activation in Jurkat T-. cells via TCR ligation is also inhibited by DAG kinases (Sanjuan et a!., 2003), and ectopic DAG kinase expression in mature T-cells impairs RasGRP1 translocation to the plasma membrane during TCR stimulation (Zha et al., 2006). In addition, inhibitors of PLC’yl inhibit accumulation of RasGRP1 in membranes (Ebinu et a!., 2000; Reynolds et al., 2004) and translocation of RasGRP1 in response to BCR stimulation is greatly reduced in DT4O B cells deficient in PLCy2 (Beaulieu et a!., 2007).  Clearly DAG and PLC activity are required for optimal translocation of RasGRP 1, but RasGRP1 activation also requires phosphorylation by PKCs (Aiba et aL, 2004; Roose et al., 2005; Zheng et al., 2005), which are also dependent on DAG. This raises the possibility that RasGRP1 translocation may not always be regulated by PLC and DAG directly. Also, recent evidence suggests that RasGRP 1 translocation to the plasma membrane in DT4O cells 30  in response to BCR stimulation is primarily regulated by another domain, the plasma membrane targeter (PT) domain, and is secondarily assisted by the Cl domain (Beaulieu et al., 2007). However, the PT domain does not regulate RasGRP1 localization or transformation in NIH 3T3 cells, and therefore its importance may be cell type-specific (Beaulieu et al., 2007). While other domains and mechanisms of activation are required for RasGRP 1 translocation and activation, a great deal of evidence supports the role of the Cl domain as a positive regulator of RasGRP1 via DAG binding at membranes.  1.7.5  The Cl domains of RasGRP2, 3 and 4 Comparative analysis of the RasGRP1 through 4 coding sequences with PKC 6 Cib  shows that all four Cl domains have the basic structure of two histidines and six cysteines that are required for Zn 2 binding (Figure 1.10). The RasGRP2 Cl domain diverges from the consensus PKC6 Cib sequence in a few key residues; Tyr8, Tbrl2 and Trp22 (Figure 1.7.5). These residues are thought to be important for the structure of the DAG/PE binding pocket (Colon-Gonzalez and Kazanietz, 2006). The RasGRP3 Cl domain has very similar sequence to that of RasGRP1 with minor differences in loop A and B. RasGRP4 has multiple splice variants although only two of these RasGRP4a and  f3, have intact Cl domains (Li et al.,  2003; Yang et al., 2002). The RasGRP4u Cl domain retains all the key residues compared to PKC6 Clb, however RasGRP4I3 has a 5 amino acid insert in loop B (VSTGP).  Conflicting evidence has led to confusion concerning the role of the RasGRP2 Cl domain. RasGRP2 mediated Rapl activation (Clyde-Smith et al., 2000; Kawasaki et al., 1998a) and adhesion to fibronectin (Dupuy et al., 2001) can be induced by PE. In addition, RasGRP2 deficient mouse platelets are unable to activate Rapl or aggregate in response to PE (Crittenden et al., 2004). Also, siRNA knockdown of RasGRP2 in human T-cells also leads to abrogation of PE induced Rapl activation and adhesion to ICAM-1 (Ghandour et al., 2007) and RasGRP2-mediated TCR-induced adhesion to ICAM-l via Rapi activation is dependent on PLC enzymes (Katagiri et al., 2004). However, it remains a possibility that these results reflect PE dependent PKC activation of RasGRP2. RasGRP1 and RasGRP3 have been shown to be phosphorylated by PKC isozymes. Interestingly, PMA-induced Rapl  31  activation in human T-cells was inhibited by the pan PKC inhibitor Go 6850 and the classical PKC inhibitor GO 6976, although this may be due to effects other than inhibition of PKC dependent RasGRP2 activation (Ghandour et al., 2007). RasGRP2 was shown to translocate to membranes in response to PE treatment (Clyde-Smith et al., 2000), however, no other evidence has arisen which shows that translocation of RasGRP2 or the RasGRP2 Cl domain to membranes in response to DAG/PE. In fact, this Cl domain was shown to be incapable of binding PE in contrast to RasGRP1, 3 and 4u in a vesicle binding assay (Irie et al., 2004). Therefore it remains to conclusively be established whether the RasGRP2 Cl domain can respond to DAG.  Loop A  PKC Cib RasGRP1 RasGRP2 RasGRP3 RasGRP4  MP FP FV Fl FL  RFKVYNYMS P NFQErTXLK P NFQESIR NFQEMNYZi PT bFQEVTRr PT 10  Loop B  D D E H  GSL AGF KAL AGF SGF 20  LWGLVKQGLK LWGVIKQGYR GIYKQGLK LWGIIKQGYK LWGVTXQGYR 30  VSTGP  Figure 1.10  =  E K K  GMNVH GMNGK GVNK GPNc$K GLCC’R 40  REKVAN KDLVVF KERLSV KDLLV RDQVR  GINQK KKRIK RRRA RRLARA KXRP  50  insert in RasGRP4  RasGRP Cl domains  Sequence comparison of RasGRP Cl domains compared to PKC Cl b. Loop A and loop B (italicized) form a DAG binding pocket. Cationic residues which have the potential to interact with anionic phospholipids within membranes are bolded and residues that are divergent in Ra5GRP2; , Thr 8 Tyr 12 and Trp 22 are circled. RasGRP4I3 Cl domain has a five amino acid insertion (VSTGP) at the base of loop B.  On the other hand, substantial evidence exists to support a positive regulatory role for the RasGRP3 Cl domain as a DAG receptor. RasGRP3 translocates to membranes and activates Ras in response to DAG or PE (Lorenzo et al., 2000) and RasGRP3 plasma membrane translocation in response to antigen receptor ligation requires both PLC activity and the Cl domain (Oh-hora et al., 2003). RasGRP4cL has been shown to activate Ras (Katsoulotos et al., 2008) and to translocate to the plasma membrane (Katsoulotos et al., 2008; Shalom-Feuerstein et al., 2008) in response to PE treatment. Both RasGRP4ci and 3 (splice variants of Ra5GRP4) promoted PE dependent transformation of NIH 3T3 fibroblasts (Li et al., 2003; Yang et al., 2002) and membrane translocation of the Cl domain of RasGRP4ct can also be induced by PE treatment (Reuther et al., 2002). While the RasGRP4a 32  Cl domain can bind PE like the RasGRP1 and 3 Cl domains (Irie et al., 2004), no such evidence exists for RasGRP4f3 Cl binding directly to DAG/PE. RasGRP4f3 Cl domain has a 5 amino acid insertion exactly at the DAG binding pocket, which could preclude DAG binding.  Thus, the Cl domains of RasGRPs are not necessarily identical with respect to binding affinity for the second messenger DAG. Understanding the specific binding capability of the Cl domains with DAG (and PE) is crucial to determine the potential for these proteins to be regulated by upstream receptors that generate DAG.  1.8  THESIS OBJECTIVES AND APPROACH My overall research aim was to determine which of the RasGRP Cl domains bind  membranes, and whether this occurs by direct binding to DAG. Evidence has accumulated that supports the hypothesis that Cl domains of RasGRP 1, 3 and 4cx bind to DAG/PE. Most models of RasGRP2 regulation are based on the assumption that the Cl domain binds to DAG, but recent research points to the fact that the RasGRP2 Cl domain may be incapable of doing so. Therefore, my first hypothesis was that the RasGRP2 Cl domain would not bind DAG or PE in membranes. Although RasGRP4I3 can transform NIH 3T3 cells in response to PE, no evidence exists for DAG binding to the RasGRP4f3 Cl domain and moreover, sequence analysis of the binding pocket of this Cl domain reveals an insertion that could disrupt DAG binding. Thus, my second hypothesis was that RasGRP4 would also not bind to DAG or PE in membranes. The first objective of my thesis was to determine whether all RasGRP Cl domains have similar ability to bind to DAG or PE in membranes. The experimental approach I took was to look at the subcellular localization of GFP-tagged Cl domains in cells using fluorescence microscopy with the aim of establishing which RasGRP Cl domains localize to membranes and which RasGRP Cl domains can relocalize to membranes in response to DAG or PE.  The most important finding from my thesis research in Chapter 2 was that the RasGRP2 Cl domain was unable to recognize DAG as a ligand, although it was able to bind 33  to membranes enriched in anionic phospholipids. This result raised additional questions which were the starting points for my research described in Chapter 3: What is the function of the Cl domain of RasGRP2? Does RasGRP2 localize to membranes and if so how is this achieved in the absence of a DAG-binding Cl domain? Therefore my second objective was to examine the role of the Cl domain in RasGRP2 membrane localization. To this end I used fluorescence microscopy to establish the localization pattern of RasGRP2 in a number of different cell lines, and to determine whether the Cl domain was required for the observed pattern.  34  1.9  BIBLIOGRAPHY  Aiba, Y., Oh-hora, M., Kiyonaka, S., Kimura, Y., Hijikata, A., Mori, Y. and Kurosaki, T. (2004) Activation of RasGRP3 by phosphorylation of Thr- 133 is required for B cell receptor-mediated Ras activation. Proc NatlAcadSci USA, 101, 166 12-16617. Alon, R., Aker, M., Feigelson, S., Sokolovsky-Eisenberg, M., Staunton, D.E., Cinamon, G., Grabovsky, V., Shamri, R. and Etzioni, A. (2003) A novel genetic leukocyte adhesion deficiency in subsecond triggering of integrin avidity by endothelial chemokines results in impaired leukocyte arrest on vascular endothelium under shear flow. Blood 101, 4437-4445. Arana, E., Vehlow, A., Harwood, N.E., Vigorito, E., Henderson, R., Turner, M., Tybulewicz, V.L. and Batista, F.D. (2008) Activation of the small GTPase Rac2 via the B cell receptor regulates B cell adhesion and immunological-synapse formation. Immunity, 28, 88-99. Beaulieu, N., Zahedi, B., Goulding, R.E., Tazmini, G., Anthony, K.V., Omeis, S.L., de Jong, D.R. and Kay, R.J. (2007) Regulation of RasGRP 1 by B cell antigen receptor requires cooperativity between three domains controlling translocation to the plasma membrane. Mol Biol Cell, 18, 3156-3168. Beranger, F., Goud, B., Tavitian, A. and de Gunzburg, J. (1991 a) Association of the Ras-antagonistic Rap 1/Kiev- 1 proteins with the Golgi complex. Proc Nat! Acad Sci USA, 88, 1606-1610. Beranger, F., Tavitian, A. and de Gunzburg, J. (199 lb) Post-translational processing and subcellular localization of the Ras-related Rap2 protein. Oncogene, 6, 1835-1842. Bergmeier, W., Goerge, T., Wang, H.W., Crittenden, J.R., Baldwin, A.C., Cifuni, S.M., Housman, D.E., Graybiel, A.M. and Wagner, D.D. (2007) Mice lacking the signaling molecule Ca1DAG-GEFI represent a model for leukocyte adhesion deficiency type III. JClin Invest, 117, 1699-1707. Bivona, T.G., Perez De Castro, I., Ahearn, I.M., Grana, TM., Chiu, V.K., Lockyer, P.J., Cullen, P.J., Pellicer, A., Cox, A.D. and Philips, M.R. (2003) Phospholipase Cgamma activates Ras on the Golgi apparatus by means of RasGRP1. Nature, 424, 694-698. Bivona, T.G., Wiener, H.H., Ahearn, I.M., Silletti, J., Chiu, V.K. and Philips, M.R. (2004) Rapl up-regulation and activation on plasma membrane regulates T cell adhesion. J Cell Biol, 164, 461-470. Booden, M.A., Campbell, S.L. and Der, C.J. (2002) Critical but distinct roles for the pleckstrin homology and cysteine-rich domains as positive modulators of Vav2 signaling and transformation. Mo! Cell Biol, 22, 2487-2497. Bos, J.L. (2005) Linking Rap to cell adhesion. Curr Opin Cell Biol, 17, 123-128. Bos, J.L., de Rooij, J. and Reedquist, K.A. (2001) RapI signalling: adhering to new models. Nat Rev Mol Cell Biol, 2, 369-377. Boussiotis, V.A., Freeman, G.J., Berezovskaya, A., Barber, D.L. and Nadler, L.M. (1997) Maintenance of human T cell anergy: blocking of IL-2 gene transcription by activated Rapl. Science, 278, 124-128. Brose, N., Betz, A. and Wegmeyer, H. (2004) Divergent and convergent signaling by the diacylglycerol second messenger pathway in mammals. Curr Opin Neurobiol, 14, 328-340. Buchsbaum, R., Telliez, J.B., Goonesekera, S. and Feig, L.A. (1996) The N-terminal pleckstrin, coiled-coil, and IQ domains of the exchange factor Ras-GRF act cooperatively to facilitate activation by calcium. Mo! CellBiol, 16, 4888-4896. Caloca, M.J., Zugaza, J.L., Matallanas, D., Crespo, P. and Bustelo, X.R. (2003b) Vav mediates Ras stimulation by direct activation of the GDP/GTP exchange factor Ras GRPI. Embo J, 22, 3326-3336. Caloca, M.J., Zugaza, J.L., Vicente-Manzanares, M., Sanchez-Madrid, F. and Bustelo, X.R. (2004) F-actin dependent translocation of the Rap 1 GDP/GTP exchange factor RasGRP2. J Biol Chem, 279, 2043520446. Carrasco, S. and Merida, I. (2006) Diacylglycerol, when simplicity becomes complex. Trends Biochem Sci. Chen, Y., Yu, M., Podd, A., Wen, R., Chrzanowska-Wodnicka, M., White, G.C. and Wang, D. (2008) A critical role of Raplb in B-cell trafficking and marginal zone B-cell development. Blood, 111, 4627-4636. Chiu, V.K., Bivona, T., Hach, A., Sajous, J.B., Silletti, J., Wiener, H., Johnson, R.L., 2nd, Cox, A.D. and Philips, M.R. (2002) Ras signalling on the endoplasmic reticulum and the Golgi. Nat Cell Biol, 4, 343350. Cho, W. and Stahelin, R.V. (2005) Membrane-protein interactions in cell signaling and membrane trafficking. Annu Rev Biophys Biomol Struct, 34, 119-151.  35  Choy, E., Chiu, V.K., Silletti, J., Feoktistov, M., Morimoto, T., Michaelson, D., Ivanov, I.E. and Philips, M.R. (1999) Endomembrane trafficking of ras: the CAAX motif targets proteins to the ER and Golgi. Cell, 98, 69-80. Chrzanowska-Wodnicka, M., Smyth, S.S., Schoenwaelder, S.M., Fischer, T.H. and White, G.C., 2nd. (2005) Raplb is required for normal platelet function and hemostasis in mice. J Clin Invest, 115, 680-687. Clyde-Smith, J., Sums, 0., Gartside, M., Grimmond, S., Etheridge, M., Apolloni, A., Hayward, N. and Hancock, J.F. (2000) Characterization of RasGRP2, a plasma membrane-targeted, dual specificity Ras/Rap exchange factor. JBiol Chem, 275, 32260-32267. Colicelli, J. (2004) Human RAS superfamily proteins and related GTPases. Sd STKE, 2004, RE13. Colon-Gonzalez, F. and Kazanietz, M.G. (2006) Cl domains exposed: From diacylglycerol binding to proteinprotein interactions. Biochim Biophys Acta. Cook, S.J., Rubinfeld, B., Albert, I. and McCormick, F. (1993) RapVl2 antagonizes Ras-dependent activation of ERKI and ERK2 by LPA and EGF in Rat-l fibroblasts. Embo J 12, 3475-3485. Coppola, J., Bryant, S., Koda, T., Conway, D. and Barbacid, M. (1991) Mechanism of activation of the vav protooncogene. Cell Growth Differ, 2, 95-105. Coughlin, J.J., Stang, S.L., Dower, N.A. and Stone, J.C. (2005) RasGRP1 and RasGRP3 regulate B cell proliferation by facilitating B cell receptor-Ras signaling. J Immunol, 175, 7179-7184. Crittenden, J.R., Bergmeier, W., Zhang, Y., Piffath, C.L., Liang, Y., Wagner, D.D., Housman, D.E. and Graybiel, A.M. (2004) Ca1DAG-GEFI integrates signaling for platelet aggregation and thrombus formation. Nat Med, 10, 982-986. Daniels, M.A., Teixeiro, E., Gill, J., Hausmann, B., Roubaty, D., Holmberg, K., Werlen, 0., Hollander, G.A., Gascoigne, N.R. and Palmer, E. (2006) Thymic selection threshold defined by compartmentalization of Ras/MAPK signalling. Nature, 444, 724-729. de Rooij, J., Zwartkruis, F.J., Verheijen, M.H., Cool, R.H., Nijman, S.M., Wittinghofer, A. and Bos, J.L. (1998) Epac is a Rapi guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature, 396, 474-477. Dillon, T.J., Carey, K.D., Wetzel, S.A., Parker, D.C. and Stork, P.J. (2005) Regulation of the small GTPase Rapi and extracellular signal-regulated kinases by the costimulatory molecule CTLA-4. Mol Cell Biol, 25,4117-4128. Dillon, T.J., Karpitski, V., Wetzel, S.A., Parker, D.C., Shaw, A.S. and Stork, P.J. (2003) Ectopic B-Raf expression enhances extracellular signal-regulated kinase (ERK) signaling in T cells and prevents antigen-presenting cell-induced anergy. J Biol Chem, 278, 35940-35949. Dower, N.A., Stang, S.L., Bottorff, D.A., Ebinu, J.O., Dickie, P., Ostergaard, H.L. and Stone, J.C. (2000) RasGRP is essential for mouse thymocyte differentiation and TCR signaling. Nat Immunol, 1, 317321. Drugan, J.K., Khosravi-Far, R., White, M.A., Der, C.J., Sung, Y.J., Hwang, Y.W. and Campbell, S.L. (1996) Ras interaction with two distinct binding domains in Raf- 1 may be required for Ras transformation. J Biol Chem, 271, 233-237. Duchniewicz, M., Zemojtel, T., Kolanczyk, M., Grossmann, S., Scheele, J.S. and Zwartkruis, F.J. (2006) Rap lA-deficient T and B cells show impaired integrin-mediated cell adhesion. Mol Cell Biol, 26, 643653. Dupuy, A.J., Akagi, K., Largaespada, D.A., Copeland, NO. and Jenkins, N.A. (2005) Mammalian mutagenesis using a highly mobile somatic Sleeping Beauty transposon system. Nature, 436, 221-226. Dupuy, A.J., Morgan, K., von Lintig, F.C., Shen, H., Acar, H., Hasz, D.E., Jenkins, N.A., Copeland, N.G., Boss, G.R. and Largaespada, D.A. (2001) Activation of the Rap 1 guanine nucleotide exchange gene, Ca1DAG-GEF 1, in BXH-2 murine myeloid leukemia. JBiol Chem, 276, 11804-11811. Ebinu, JO., Bottorff, D.A., Chan, E.Y., Stang, S.L., Dunn, R.J. and Stone, J.C. (1998) RasGRP, a Ras guanyl nucleotide- releasing protein with calcium- and diacylglycerol-binding motifs. Science, 280, 10821086. Ebinu, J.O., Stang, S.L., Teixeira, C., Bottorff D.A., Hooton, J., Blumberg, P.M., Barry, M., Bleakley, R.C., Ostergaard, H.L. and Stone, J.C. (2000) RasGRP links T-cell receptor signaling to Ras. Blood, 95, 3199-3203. Ehrhardt, A., Ehrhardt, G.R., Guo, X. and Schrader, J.W. (2002) Ras and relatives--job sharing and networking keep an old family together. Exp Hematol, 30, 1089-1106. Esteban, L.M., Vicario-Abejon, C., Femandez-Salguero, P., Fernandez-Medarde, A., Swaminathan, N., Yienger, K., Lopez, E., Malumbres, M., McKay, R., Ward, J.M., Pellicer, A. and Santos, B. (2001)  36  Targeted genomic disruption of H-ras and N-ras, individually or in combination, reveals the dispensability of both loci for mouse growth and development. Mo! Cell Biol, 21, 1444-1452. Eto, K., Murphy, R., Kerrigan, S.W., Bertoni, A., Stuhlmann, H., Nakano, T., Leavitt, A.D. and Shattil, S.J. (2002) Megakaryocytes derived from embryonic stem cells implicate CaIDAG-GEFI in integrin signaling. Proc NatlAcad Sci USA, 99, 12819-12824. Eychene, A., Dusanter-Fourt, I., Barnier, J.V., Papin, C., Charon, M., Gisselbrecht, S. and Calothy, G. (1995) Expression and activation of B-Raf kinase isoforms in human and murine leukemia cell lines. Oncogene, 10, 1159-1165. Farrell, F.X., Yamamoto, K. and Lapetina, E.G. (1993) Prenyl group identification of rap2 proteins: a ras superfamily member other than ras that is farnesylated. Biochem J 289 (Pt 2), 349-355. Feller, S.M. (2001) Crk family adaptors-signalling complex formation and biological roles. Oncogene, 20, 6348-6371. Galmiche, A. and Fueller, J. (2007) RAF kinases and mitochondria. Biochim Biophys Acta, 1773, 1256-1262. Gao, X., Satoh, T., Liao, Y., Song, C., Hu, C.D., Kariya Ki, K. and Kataoka, T. (2001) Identification and characterization of RA-GEF-2, a Rap guanine nucleotide exchange factor that serves as a downstream target of M-Ras. JBiol Chem, 276,42219-42225. Ghandour, H., Cullere, X., Alvarez, A., Luscinskas, F.W. and Mayadas, T.N. (2007) Essential role for Rap 1 GTPase and its guanine exchange factor CaIDAG-GEFI in LFA- I but not VLA-4 integrin mediated human T-cell adhesion. Blood 110, 3682-3690. Ghosh, S., Strum, J.C., Sciorra, V.A., Daniel, L. and Bell, R.M. (1996) Raf-l kinase possesses distinct binding domains for phosphatidylserine and phosphatidic acid. Phosphatidic acid regulates the translocation of Raf- 1 in 12-O-tetradecanoylphorbol- 13-acetate-stimulated Madin-Darby canine kidney cells. J Biol Chem, 271, 8472-8480. Glading, A., Han, J., Stockton, R.A. and Ginsberg, M.H. (2007) KRJT-1/CCM1 is a Rapi effector that regulates endothelial cell cell junctions. J Cell Biol, 179, 247-254. Gotoh, T., Hattori, S., Nakamura, S., Kitayama, H., Noda, M., Takai, Y., Kaibuchi, K., Matsui, H., Hatase, 0., Takahashi, H. and et al. (1995) Identification of Rap 1 as a target for the Crk SH3 domain-binding guanine nucleotide-releasing factor C3G. Mo! Cell Biol, 15, 6746-6753. Guo, Z., Yuan, J., Tang, W., Chen, X., Gu, X., Luo, K., Wang, Y., Wan, B. and Yu, L. (2007) Cloning and characterization of the human gene RAP2C, a novel member of Ras family, which activates transcriptional activities of SRE. Mo! Biol Rep, 34, 137-144. Hancock, J.F. (2007) PA promoted to manager. Nat Cell Biol, 9, 615-617. Harding, A., Tian, T., Westbury, E., Frische, E. and Hancock, J.F. (2005) Subcellular localization determines MAP kinase signal output. Curr Biol, 15, 869-873. Hisata, S., Sakisaka, T., Baba, T., Yamada, T., Aoki, K., Matsuda, M. and Takai, Y. (2007) Rapl-PDZ-GEF1 interacts with a neurotrophin receptor at late endosomes, leading to sustained activation of Rap 1 and ERK and neurite outgrowth. J Cell Biol, 178, 84 3-860. Hogan, C., Serpente, N., Cogram, P., Hosking, C.R., Bialucha, C.U., Feller, S.M., Braga, V.M., Birchmeier, W. and Fujita, Y. (2004) Rapl regulates the formation of E-cadherin-based cell-cell contacts. Mol Cell Biol, 24, 6690-6700. Hommel, U., Zurini, M. and Luyten, M. (1994) Solution structure of a cysteine rich domain of rat protein kinase C. Nat Struct Biol, 1, 383-387. Hoshino, T., Sakisaka, T., Baba, T., Yamada, T., Kimura, T. and Takai, Y. (2005) Regulation of E-cadherin endocytosis by nectin through afadin, Rapi, and pl2octn. JBiol Chem, 280, 24095-24103. Huser, M., Luckett, J., Chiloeches, A., Mercer, K., Iwobi, M., Giblett, S., Sun, X.M., Brown, J., Marais, R. and Pritchard, C. (2001) MEK kinase activity is not necessary for Raf-i function. Embo J 20, 1940-1951. Irie, K., Masuda, A., Shindo, M., Nakagawa, Y. and Ohigashi, H. (2004) Tumor promoter binding of the protein kinase C Cl homology domain peptides of RasGRPs, chimaerins, and Uncl3s. Bioorg Med Chem, 12, 4575-4583. Ishida, D., Kometani, K., Yang, Fl., Kakugawa, K., Masuda, K., Iwai, K., Suzuki, M., Itohara, S., Nakahata, T., Hiai, H., Kawamoto, H., Hattori, M. and Minato, N. (2003a) Myeloproliferative stem cell disorders by deregulated Rap 1 activation in SPA-i-deficient mice. Cancer Cell, 4, 55-65. Ishida, D., Yang, H., Masuda, K., Uesugi, K., Kawamoto, H., Hattori, M. and Minato, N. (2003b) Antigendriven T cell anergy and defective memory T cell response via deregulated Rap 1 activation in SPA-I deficient mice. Proc NatlAcad Sci USA, 100, 109 19-10924. -  37  Jenei, V., Deevi, R.K., Adams, C.A., Axeisson, L., Hirst, D.G., Andersson, T. and Dib, K. (2006) Nitric oxide produced in response to engagement of beta2 integrins on human neutrophils activates the monomeric GTPases Rapl and Rap2 and promotes adhesion. JBiol Chem, 281, 35008-35020. Jiang, X. and Sorkin, A. (2002) Coordinated traffic of Grb2 and Ras during epidermal growth factor receptor endocytosis visualized in living cells. Mol Biol Cell, 13, 1522-1535. Johnson, L., Greenbaum, D., Cichowski, K., Mercer, K., Murphy, E., Schmitt, E., Bronson, R.T., Umanoff H., Edelmann, W., Kucherlapati, R. and Jacks, T. (1997) K-ras is an essential gene in the mouse with partial functional overlap with N-ras. Genes Dev, 11, 2468-2481. Kamata, T., Kang, J., Lee, T.H., Wojnowski, L., Pritchard, C.A. and Leavitt, A.D. (2005) A critical function for B-Raf at multiple stages of myelopoiesis. Blood, 106, 833-840. Katagiri, K., Hattori, M., Minato, N., Irie, S., Takatsu, K. and Kinashi, T. (2000) Rapi is a potent activation signal for leukocyte function-associated antigen I distinct from protein kinase C and phosphatidylinositol-3-OH kinase. Mol Cell Biol, 20, 1956-1969. Katagiri, K., Hattori, M., Minato, N. and Kinashi, T. (2002) Rapl functions as a key regulator of T-cell and antigen-presenting cell interactions and modulates T-cell responses. Mol Cell Biol, 22, 1001-1015. Katagiri, K., Maeda, A., Shimonaka, M. and Kinashi, T. (2003) RAPL, a Rap 1-binding molecule that mediates Rap 1-induced adhesion through spatial regulation of LFA- 1. Nat Immunol, 4, 741-748. Katagiri, K., Shimonaka, M. and Kinashi, T. (2004) Rap 1-mediated lymphocyte function-associated antigen-i activation by the T cell antigen receptor is dependent on phospholipase C-gamma 1. J Biol Chem, 279, 11875-11881. Katsoulotos, G.P., Qi, M., Qi, J.C., Tanaka, K., Hughes, W.E., Molloy, T.J., Adachi, R., Stevens, R.L. and Krilis, S.A. (2008) The Diacylglycerol-dependent translocation of ras guanine nucleotide-releasing protein 4 inside a human mast cell line results in substantial phenotypic changes, including expression of interleukin 13 receptor alpha2. JBiol Chem, 283, 1610-1621. Kawasaki, H., Springett, G.M., Mochizuki, N., Toki, S., Nakaya, M., Matsuda, M., Housman, D.E. and Graybiel, A.M. (1 998b) A family of cAMP-binding proteins that directly activate Rap 1. Science, 282, 2275-2279. Kawasaki, H., Springett, G.M., Toki, S., Canales, J.J., Harlan, P., Blumenstiel, J.P., Chen, E.J., Bany, l.A., Mochizuki, N., Ashbacher, A., Matsuda, M., Housman, D.E. and Graybiel, A.M. (1998a) A Rap guanine nucleotide exchange factor enriched highly in the basal ganglia. Proc Nail Acad Sci USA, 95, 13278-13283. Kazanietz, M.G. (2002) Novel “nonkinase” phorbol ester receptors: the Cl domain connection. Mol Pharmacol, 61, 759-767. Kinashi, T. (2005) Intracellular signalling controlling integrin activation in lymphocytes. Nat Rev Immunol, 5, 546-559. Kinashi, T. and Katagiri, K. (2004) Regulation of lymphocyte adhesion and migration by the small GTPase Rapi and its effector molecule, RAPL. Immunol Lett, 93, 1-5. Kliche, S., Breitling, D., Togni, M., Pusch, R., Heuer, K., Wang, X., Freund, C., Kasirer-Friede, A., Menasche, G., Koretzky, G.A. and Schraven, B. (2006) The ADAP/SKAP55 signaling module regulates T-cell receptor-mediated integrin activation through plasma membrane targeting of Rap 1. Mol Cell Biol, 26, 7130-7144. Klinger, M.B., Guilbault, B., Goulding, R.E. and Kay, R.J. (2005) Deregulated expression of RasGRPI initiates thymic lymphomagenesis independently of T-cell receptors. Oncogene, 24, 2695-2704. Knox, A.L. and Brown, N.H. (2002) Rap 1 GTPase regulation of adherens junction positioning and cell adhesion. Science, 295, 1285-1288. Kometani, K., Ishida, D., Hattori, M. and Minato, N. (2004) Rapi and SPA-i in hematologic malignancy. Trends Mol Med 10, 40 1-408. Kooistra, M.R., Corada, M., Dejana, E. and Bos, J.L. (2005) Epaci regulates integrity of endothelial cell junctions through VE-cadherin. FEBS Lett, 579, 4966-4972. Kooistra, M.R., Dube, N. and Bos, J.L. (2007) Rapi: a key regulator in cell-cell junction formation. JCellSci, 120, 17-22. Leicht, D.T., Balan, V., Kaplun, A., Singh-Gupta, V., Kaplun, L., Dobson, M. and Tzivion, G. (2007) Raf kinases: function, regulation and role in human cancer. Biochim Biophys Acta, 1773, 1196-1212. Li, J., Shen, H., Himmel, K.L., Dupuy, A.J., Largaespada, D.A., Nakamura, T., Shaughnessy, J.D., Jr., Jenkins, N.A. and Copeland, N.G. (1999) Leukaemia disease genes: large-scale cloning and pathway predictions. Nat Genet, 23, 348-353.  38  Li, L., Yang, Y. and Stevens, R.L. (2003) RasGRP4 regulates the expression of prostaglandin D2 in human and rat mast cell lines. JBiol Chem, 278, 4725-4729. Lin, K.B., Freeman, S.A., Zabetian, S., Brugger, H., Weber, M., Lei, V., Dang-Lawson, M., Tse, K.W., Santamaria, R., Batista, F.D. and Gold, M.R. (2008) The rap GTPases regulate B cell morphology, immune-synapse formation, and signaling by particulate B cell receptor ligands. Immunity, 28, 75-87. Liu, Y., Zhu, M., Nishida, K., Hirano, T. and Zhang, W. (2007) An essential role for RasGRP1 in mast cell function and IgE-mediated allergic response. J Exp Med, 204, 93-103. Lorenzo, P.S., Beheshti, M., Pettit, G.R., Stone, J.C. and Blumberg, P.M. (2000) The guanine nucleotide exchange factor RasGRP is a high -affinity target for diacylglycerol and phorbol esters. Mol Pharmacol, 57, 840-846. Madani, S., Hichami, A., Cherkaoui-Malki, M. and Khan, N.A. (2004) Diacyiglycerols containing Omega 3 and Omega 6 fatty acids bind to RasGRP and modulate MAP kinase activation. JBiol Chem, 279, 11761183. McLeod, S.J., Li, A.H., Lee, R.L., Burgess, A.E. and Gold, M.R. (2002) The Rap GTPases regulate B cell migration toward the chemokine stromal cell-derived factor-i (CXCL12): potential role for Rap2 in promoting B cell migration. Jlmmunol, 169, 1365-137 1. McLeod, S.J., Shum, A.J., Lee, R.L., Takei, F. and Gold, M.R. (2004) The Rap GTPases regulate integrin mediated adhesion, cell spreading, actin polymerization, and Pyk2 tyrosine phosphorylation in B lymphocytes. JBiol Chem, 279, 12009-12019. Medeiros, R.B., Dickey, D.M., Chung, H., Quale, A.C., Nagarajan, L.R., Billadeau, D.D. and Shimizu, Y. (2005) Protein kinase Di and the beta I integrin cytoplasmic domain control beta 1 integrin function via regulation ofRapi activation. Immunity, 23, 2 13-226. Menasche, G., Kliche, S., Bezman, N. and Schraven, B. (2007b) Regulation of T-cell antigen receptor-mediated inside-out signaling by cytosolic adapter proteins and Rapi effector molecules. Immunol Rev, 218, 8291. Menasche, G., Kliche, S., Chen, E.J., Stradal, T.E., Schraven, B. and Koretzky, G. (2007a) RIAM links the ADAP/SKAP-55 signaling module to Rap 1, facilitating T-cell-receptor-mediated integrin activation. Mol Cell Biol, 27, 4070-4081. Michaud, N.R., Therrien, M., Cacace, A., Edsall, L.C., Spiegel, S., Rubin, G.M. and Morrison, D.K. (1997) KSR stimulates Raf- 1 activity in a kinase-independent manner. Proc Nat! Acad Sci US A, 94, 1279212796. Mikkers, H., Allen, J., Knipscheer, P., Romeijn, L., Hart, A., Vink, E. and Bems, A. (2002) High-throughput retroviral tagging to identif’ components of specific signaling pathways in cancer. Nat Genet, 32, 153159. Mikula, M., Schreiber, M., Husak, Z., Kucerova, L., Ruth, J., Wieser, R., Zatloukal, K., Beug, H., Wagner, E.F. and Baccarini, M. (2001) Embryonic lethality and fetal liver apoptosis in mice lacking the c-raf-1 gene. EmboJ 20, 1952-1962. Mochizuki, N., Ohba, Y., Kiyokawa, E., Kurata, T., Murakami, T., Ozaki, T., Kitabatake, A., Nagashima, K. and Matsuda, M. (1999) Activation of the ERK!MAPK pathway by an isoform of rap 1 GAP associated with G alpha(i). Nature, 400, 89 1-894. Mochizuki, N., Ohba, Y., Kobayashi, S., Otsuka, N., Graybiel, A.M., Tanaka, S. and Matsuda, M. (2000) Crk activation of JNK via C3G and R-Ras. JBiol Chem, 275, 12667-12671. Mor, A., Campi, G., Du, G., Zheng, Y., Foster, D.A., Dustin, M.L. and Philips, M.R. (2007) The lymphocyte function-associated antigen-i receptor costimulates plasma membrane Ras via phospholipase D2. Nat Cell Biol, 9,713-719. Mor, A. and Philips, M.R. (2006) Compartmentalized Ras/MAPK signaling. Annu Rev Immunol, 24, 77 1-800. Mott, H.R., Carpenter, J.W., Zhong, S., Ghosh, S., Bell, R.M. and Campbell, S.L. (1996) The solution structure of the Raf- 1 cysteine-rich domain: a novel ras and phospholipid binding site. Proc Nati Acad Sci US A, 93, 83 12-8317. Nimnual, A. and Bar-Sagi, D. (2002) The two hats of SOS. Sd STKE, 2002, PE36. Nomura, K., Kanemura, H., Satoh, T. and Kataoka, T. (2004) Identification of a novel domain of Ras and Rap 1 that directs their differential subcellular localizations. JBiol Chem, 279, 22664-22673. O’Neill, E. and Kolch, W. (2004) Conferring specificity on the ubiquitous RafYMEK signalling pathway. Br J Cancer, 90, 283-288.  39  Oh-hora, M., Johmura, S., Hashimoto, A., Hikida, M. and Kurosaki, T. (2003) Requirement for Ras guanine nucleotide releasing protein 3 in coupling phospholipase C-gamma2 to Ras in B cell receptor signaling. JExp Med 198, 1841-1851. Ohba, Y., Mochizuki, N., Matsuo, K., Yamashita, S., Nakaya, M., Hashimoto, Y., Hamaguchi, M., Kurata, T., Nagashima, K. and Matsuda, M. (2000b) Rap2 as a slowly responding molecular switch in the Rap 1 signaling cascade. Mol Cell Biol, 20, 6074-6083. Ohba, Y., Mochizuki, N., Yamashita, S., Chan, A.M., Schrader, J.W., Hattori, S., Nagashima, K. and Matsuda, M. (2000a) Regulatory proteins of R-Ras, TC2 l/R-Ras2, and M-Ras/R-Ras3. J Biol Chem, 275, 20020-20026. Okada, S., Matsuda, M., Anafi, M., Pawson, T. and Pessin, J.E. (1998) Insulin regulates the dynamic balance between Ras and Rapi signaling by coordinating the assembly states of the Grb2-SOS and CrkII-C3G complexes. Embo J, 17, 2554-2565. Okada, T., Hu, C.D., Jin, T.G., Kariya, K., Yamawaki-Kataoka, Y. and Kataoka, T. (1999) The strength of interaction at the Raf cysteine-rich domain is a critical determinant of response of Raf to Ras family small GTPases. Mol Cell Biol, 19, 6057-6064. Oki-Idouchi, C.E. and Lorenzo, P.S. (2007) Transgenic overexpression of RasGRP1 in mouse epidermis results in spontaneous tumors of the skin. Cancer Res, 67, 276-280. Omerovic, J., Laude, A.J. and Prior, l.A. (2007) Ras proteins: paradigms for compartmentalised and isoform specific signalling. Cell Mol Lfe Sd, 64, 2575-2589. Paganini, S., Guidetti, G.F., Catricala, S., Trionfmi, P., Panelli, S., Balduini, C. and Torti, M. (2006) Identification and biochemical characterization of Rap2C, a new member of the Rap family of small GTP-binding proteins. Biochimie, 88, 285-295. Pasvolsky, R., Feigelson, S.W., Kilic, S.S., Simon, A.J., Tal-Lapidot, G., Grabovsky, V., Crittenden, J.R., Amariglio, N., Safran, M., Graybiel, A.M., Rechavi, G., Ben-Dor, S., Etzioni, A. and Alon, R. (2007) A LAD-Ill syndrome is associated with defective expression of the Rap-i activator CaIDAG-GEFI in lymphocytes, neutrophils, and platelets. JExp Med, 204, 1571-1582. Perez de Castro, I., Bivona, T.G., Philips, M.R. and Pellicer, A. (2004) Ras activation in Jurkat T cells following low-grade stimulation of the T-cell receptor is specific to N-Ras and occurs only on the Golgi apparatus. Mol Cell Biol, 24, 3485-3496. Pizon, V., Desjardins, M., Bucci, C., Parton, R.G. and Zerial, M. (1994) Association ofRapla and Rap lb proteins with late endocytic/phagocytic compartments and Rap2a with the Golgi complex. J Cell Sci, 107 (Pt 6), 1661-1670. Priatel, J.J., Teh, S.J., Dower, N.A., Stone, J.C. and Teh, H.S. (2002) RasGRP1 transduces low-grade TCR signals which are critical for T cell development, homeostasis, and differentiation. immunity, 17, 617627. Price, L.S., Hajdo-Milasinovic, A., Zhao, J., Zwartkruis, F.J., Collard, J.G. and Bos, J.L. (2004) Rapi regulates E-cadherin-mediated cell-cell adhesion. JBiol Chem, 279, 35127-35132. Qiao, J., Mei, F.C., Popov, V.L., Vergara, L.A. and Cheng, X. (2002) Cell cycle-dependent subcellular localization of exchange factor directly activated by cAMP. JBiol Chem, 277, 2658 1-26586. Radha, V., Rajanna, A. and Swarup, G. (2004) Phosphorylated guanine nucleotide exchange factor C3G, induced by pervanadate and Src family kinases localizes to the Golgi and subcortical actin cytoskeleton. BMC Cell Biol, 5, 31. Rambaratsingh, R.A., Stone, J.C., Blumberg, P.M. and Lorenzo, P.S. (2003) RasGRP1 represents a novel non protein kinase C phorbol ester signaling pathway in mouse epidermal keratinocytes. J Biol Chem, 278, 52792-52801. Reedquist, K.A. and Bos, J.L. (1998) Costimulation through CD28 suppresses T cell receptor-dependent activation of the Ras-like small GTPase Rapi in human T lymphocytes. JBiol Chem, 273, 4944-4949. Reuther, G.W., Lambert, Q.T., Rebhun, J.F., Caligiuri, M.A., Quilliam, L.A. and Der, C.J. (2002) RasGRP4 is a novel Ras activator isolated from acute myeloid leukemia. JBiol Chem, 277, 30508-305 14. Reynolds, L.F., de Bettignies, C., Norton, T., Beeser, A., Chemoff J. and Tybulewicz, V.L. (2004) Vavi transduces T cell receptor signals to the activation of the Ras/ERK pathway via LAT, Sos, and RasGRPI. JBiol Chem, 279, 18239-18246. Rocks, 0., Peyker, A., Kahms, M., Verveer, P.J., Koemer, C., Lumbierres, M., Kuhlmann, J., Waldmann, H., Wittinghofer, A. and Bastiaens, P.1. (2005) An acylation cycle regulates localization and activity of palmitoylated Ras isoforms. Science, 307, 1746-1752.  40  Rong, S.B., Enyedy, I.J., Qiao, L., Zhao, L., Ma, D., Pearce, L.L., Lorenzo, P.S., Stone, J.C., Blumberg, P.M., Wang, S. and Kozikowski, A.P. (2002) Structural basis of RasGRP binding to high-affinity PKC ligands. JMed Chem, 45, 853-860. Roose, J.P., Mollenauer, M., Gupta, V.A., Stone, J. and Weiss, A. (2005) A diacyiglycerol-protein kinase C RasGRP I pathway directs Ras activation upon antigen receptor stimulation of T cells. Mol Cell Biol, 25, 4426-4441. Sanjuan, M.A., Pradet-Balade, B., Jones, D.R., Martinez, A.C., Stone, J.C., Garcia-Sanz, J.A. and Merida, I. (2003) T cell activation in vivo targets diacyiglycerol kinase alpha to the membrane: a novel mechanism for Ras attenuation. Jlmmunol, 170, 2877-2883. Schmitt, J.M. and Stork, P.J. (2002) PKA phosphorylation of Src mediates cAMP’s inhibition of cell growth via Rapi. Mol Cell, 9, 85-94. Sebzda, E., Bracke, M., Tugal, T., Hogg, N. and Cantrell, D.A. (2002) RaplA positively regulates T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat Immunol, 3, 251-258. Shalom-Feuerstein, R., Levy, R., Makovski, V., Raz, A. and Kloog, Y. (2008) Galectin-3 regulates RasGRP4mediated activation of N-Ras and H-Ras. Biochim Biophys Acta, 1783, 985-993. Shimonaka, M., Katagiri, K., Nakayama, T., Fujita, N., Tsuruo, T., Yoshie, 0. and Kinashi, T. (2003) Rap I translates chemokine signals to integrin activation, cell polarization, and motility across vascular endothelium under flow. JCell Biol, 161, 4 17-427. Stevens, R.L., Morokawa, N., Wang, J. and Krilis, S.A. (2005) RasGRP4 in mast cell signalling and disease susceptibility. Novartis Found Symp, 271, 54-68; discussion 68-77, 95-59. Stone, J.C. (2006) Regulation of Ras in lymphocytes: get a GRP. Biochem Soc Trans, 34, 858-861. Stork, P.J. (2003) Does Rapl deserve a bad Rap? Trends Biochem Sci, 28, 267-275. Suzuki, T., Shen, H., Akagi, K., Morse, H.C., Malley, J.D., Naiman, D.Q., Jenkins, N.A. and Copeland, N.G. (2002) New genes involved in cancer identified by retroviral tagging. Nat Genet, 32, 166-174. Takai, Y., Sasaki, T. and Matozaki, T. (2001) Small GTP-binding proteins. Physiol Rev, 81, 153-208. Teixeira, C., Stang, S.L., Zheng, Y., Beswick, N.S. and Stone, J.C. (2003) Integration of DAG signaling systems mediated by PKC-dependent phosphorylation of RasGRP3. Blood 102, 1414-1420. Tian, T., Harding, A., Inder, K., Plowman, S., Patton, R.G. and Hancock, J.F. (2007) Plasma membrane nanoswitches generate high-fidelity Ras signal transduction. Nat Cell Biol, 9, 905-914. Tognon, C.E., Kirk, H.E., Passmore, L.A., Whitehead, I.P., Der, C.J. and Kay, R.J. (1998) Regulation of RasGRI’ via a phorbol ester-responsive Cl domain. Mo! Cell Biol, 18, 6995-7008. Vossler, M.R., Yao, H., York, R.D., Pan, M.G., Rim, C.S. and Stork, P.J. (1997) cAMP activates MAP kinase and Elk-i through a B-Raf- and Rap 1-dependent pathway. Cell, 89, 73-82. Wang, H.W., Wang, Z., Dillon, T.J., Pokala, V., Mishra, S., Labudda, K., Hunter, B. and Stork, P.J. (2006) Rap 1-mediated activation of extracellular signal-regulated kinases by cyclic AMP is dependent on the mode of Rapi activation. Mol Cell Biol, 26, 2130-2145. Weber, C.K., Slupsky, J.R., Herrmann, C., Schuler, M., Rapp, U.R. and Block, C. (2000) Mitogenic signaling of Ras is regulated by differential interaction with Raf isozymes. Oncogene, 19, 169-176. Welibrock, C., Karasarides, M. and Marais, R. (2004) The RAF proteins take centre stage. Nat Rev Mo! Cell Biol, 5, 875-885. Winkler, D.G., Cutler, R.E., Jr., Drugan, J.K., Campbell, S., Morrison, D.K. and Cooper, J.A. (1998) Identification of residues in the cysteine-rich domain of Raf- 1 that control Ras binding and Raf- 1 activity. JBiol Chem, 273, 2 1578-21584. Wu, C., Lai, C.F. and Mobley, W.C. (2001) Nerve growth factor activates persistent Rap 1 signaling in endosomes. JNeurosci, 21, 5406-5416. Wu, C., Ramirez, A., Cui, B., Ding, J., Delcroix, J.D., Valletta, J.S., Liu, J.J., Yang, Y., Chu, S. and Mobley, W.C. (2007) A functional dynein-microtubule network is required for NGF signaling through the Rap i/MAPK pathway. Traffic, 8, 1503-1520. Yajnik, V., Paulding, C., Sordella, R., McClatchey, A.I., Saito, M., Wahrer, D.C., Reynolds, P., Bell, D.W., Lake, R., van den Heuvel, S., Settleman, J. and Haber, D.A. (2003) DOCK4, a GTPase activator, is disrupted during tumorigenesis. Cell, 112, 673-684. Yamashita, S., Mochizuki, N., Ohba, Y., Tobiume, M., Okada, Y., Sawa, H., Nagashima, K. and Matsuda, M. (2000) Ca1DAG-GEFIII activation of Ras, R-ras, and Rapi. JBiol Chem, 275, 25488-25493. Yang, Y., Li, L., Wong, G.W., Krilis, S.A., Madhusudhan, M.S., Sali, A. and Stevens, R.L. (2002) RasGRP4, a new mast cell-restricted Ras guanine nucleotide-releasing protein with calcium- and diacylglycerol binding motifs. Identification of defective variants of this signaling protein in asthma, mastocytosis,  41  and mast cell leukemia patients and demonstration of the importance of RasGRP4 in mast cell development and function. JBiol Chem, 277, 25756-25774. York, R.D., Molliver, D.C., Grewal, S.S., Stenberg, P.E., McCleskey, E.W. and Stork, P.J. (2000) Role of phosphoinositide 3-kinase and endocytosis in nerve growth factor-induced extracellular signalregulated kinase activation via Ras and Rapi. Mol Cell Biol, 20, 8069-8083. York, R.D., Yao, H., Dillon, T., Ellig, C.L., Eckert, S.P., McCleskey, E.W. and Stork, P.J. (1998) Rapi mediates sustained MAP kinase activation induced by nerve growth factor. Nature, 392, 622-626. Zha, Y., Marks, R., Ho, A.W., Peterson, A.C., Janardhan, S., Brown, I., Praveen, K., Stang, S., Stone, J.C. and Gajewski, T.F. (2006) T cell anergy is reversed by active Ras and is regulated by diacyiglycerol kinase-alpha. Nat Immunol, 7, 1166-1173. Zhang, F.L. and Casey, P.J. (1996) Protein prenylation: molecular mechanisms and functional consequences. Annu Rev Biochem, 65, 24 1-269. Zhang, G., Kazanietz, M.G., Blumberg, P.M. and Hurley, J.H. (1995) Crystal structure of the cys2 activatorbinding domain of protein kinase C delta in complex with phorbol ester. Cell, 81, 917-924. Zhao, C., Du, G., Skowronek, K., Frohman, M.A. and Bar-Sagi, D. (2007) Phospholipase D2-generated phosphatidic acid couples EGFR stimulation to Ras activation by Sos. Nat Cell Biol, 9, 706-7 12. Zheng, Y., Liu, H., Coughlin, J., Zheng, J., Li, L. and Stone, J.C. (2005) Phosphorylation of RasGRP3 on threonine 133 provides a mechanistic link between PKC and Ras signaling systems in B cells. Blood, 105, 3648-3654. Zhou, M., Horita, D.A., Waugh, D.S., Byrd, R.A. and Morrison, D.K. (2002) Solution structure and functional analysis of the cysteine-rich Cl domain of kinase suppressor of Ras (KSR). JMo1 Biol, 315, 435-446. Zugaza, J.L., Caloca, M.J. and Bustelo, X.R. (2004) Inverted signaling hierarchy between RAS and RAC in T lymphocytes. Oncogene, 23, 5823-5833.  42  CHAPTER 2: DIFFERENTIAL MEMBRANE BINDING AND DIACYLGLYCEROL RECOGNITION BY Cl DOMAINS OF RASGRPS  A version of this chapter has been published: Johnson, JE, Goulding, RE, Ding, Z, Partovi, A, Anthony, Ky, Beaulieu, N, Tazmini, G, Cornell, R. B and Kay, RJ. (2007) Differential membrane binding and diacylglycerol recognition by Cl domains of RasGRPs. Biochem 1 406: 223-36.  43  2.1  INTRODUCTION Diacyiglycerol (DAG) is a lipid second messenger that is involved in the transduction  of signals from most cell surface receptors (Brose et al., 2004; Carrasco and Merida, 2006; Gomez-Fernandez et al., 2004). Signal transduction by DAG is mediated by its binding to Cl domains, which are found on a wide variety of kinases, exchange factors and other proteins (Colon-Gonzalez and Kazanietz, 2006; Hurley et al., 1997). The binding of DAG to their Cl domains serves to translocate these proteins to membranes, where they can interact with signaling complexes and fmd their substrates and regulators (Cho and Stahelin, 2005; Hurley, 2006). In some cases, the binding of Cl domains to DAG within membranes has the potential to release other domains of the signaling proteins and thus contributes to enzymatic activation as well as translocation to membranes (Colon-Gonzalez and Kazanietz, 2006; Newton, 2001). In the case of PKC, when the C1B domain binds to DAG in membranes, in coordination with binding of the C2 domain to membranes, this enables the dislocation of the N-terminal pseudosubstrate domain from the C-terminal catalytic kinase domain, allowing the latter to bind substrate and catalyze downstream target activation (Newton, 2001). Conversely, conformational changes could be mitigated by other domains and lead to the liberation of the Cl domain. For example, 132-chimaerin translocates to the membrane by an unknown mechanism, after which interactions with acidic phospholipids lead to a conformational change of the protein that in turn releases the Cl domain from Nterminal auto-inhibition, allowing it to bind DAG in membranes and also to bind and activate RapGAP (Hall et al., 2005). Cl domains are functionally heterogeneous in terms of their ligand binding capabilities (Cho and Stahelin, 2005; Colon-Gonzalez and Kazanietz, 2006). Most protein kinase Cs (PKCs) have a pair of Cl domains, but each of these can differ considerably in their affinities for DAG (Cho and Stahelin, 2005). PKC Cl domains can also differ in their relative affinities for DAG versus the phorbol esters which are commonly used as artificial ligands for Cl domains (Cho and Stahelin, 2005). In general, these Cl domains have higher affinity for phorbol esters, but in some cases DAG is the more effective ligand (Ananthanarayanan et al., 2003; Stahelin et al., 2004). Some Cl domains have no detectable affinity for either DAG or phorbol esters, and thus may have functions other than serving as 44  DAG receptors (Colon-Gonzalez and Kazanietz, 2006) or may be Cl domains, which are no longer functional in terms of providing membrane localization.  Cl domains are readily identified by sequence analysis due to the characteristic spacing of two histidines and six cysteines (Figure 2.1 A). These amino acids lock the domain into a compact structure by coordinating two zinc atoms (Figure 2.1B). In Cl domains capable of binding DAG or phorbol esters and with known structures, such as the second Cl domain (Cib) of PKC (Zhang et  a!., 1995) or the Cl domain of 132-chimaerin  (Canagarajah et al., 2004), two projecting loops insert into membranes and form a ligand binding pocket (Figure 2.1B) that can accommodate either the head group of diacylglycerol or an analogous segment of phorbol ester. While the atomic interactions between the Clb domain of PKC6 and phorbol ester have been identified, the precise structural basis for the diacylglycerol-C 1 domain interaction is unknown. The Cl domains of Raf- 1 and KSR, which do not bind DAG or phorbol ester, have radical alterations in loops A and B (Mott et al., 1996; Zhou et al., 2002). In particular, they lack the proline that dictates the structure of loop A and are missing half of loop B (Figure 2.1A). As a consequence, these two Cl domains lack a ligand binding pocket.  In addition to specific ligand binding via the pocket, Cl domains interact with the membrane surface (Cho and Stahelin, 2005). Hydrophobic residues projecting from or at the base of loops A and B facilitate membrane insertion (Cho and Stahelin, 2005; Wang et al., 2001; Zhang eta!., 1995). Basic residues distributed on the periphery of the Cl domain surface and at the C-terminus (Figure 2.1 A) provide electrostatic interactions with anionic phospholipids and thus enhance DAG-mediated binding to membranes (Irie et a!., 2004; Johnson et al., 2000; Medkova and Cho, 1999; Zhang et al., 1995). For KSR and Raf-1 these hydrophobic and electrostatic interactions may enable DAG-independent membrane binding (Bondeva et a!., 2002; Zhou et al., 2002).  RasGRPs are exchange factors for membrane-bound Ras or Rap GTPases (Stone, 2006). All RasGRP family members contain Cl domains. This suggested that the activity of all RasGRP family members could be regulated by membrane translocation mediated by 45  binding of their Cl domains to DAG. Evidence for this concept has been obtained for RasGRP1 and to a lesser extent for RasGRP3. Activation ofRasGRPl and RasGRP3 by antigen receptors requires DAG-generating phospholipase Cy enzymes (Caloca et al., 2003b; Oh-hora et al., 2003; Zugaza et a!., 2004) and is inhibited by DAG kinases (Jones et al., 2002; Sanjuan et al., 2003; Topham and Prescott, 2001; Zha et al., 2006). This does not necessarily reflect involvement of a DAG-C 1 domain interaction, because activation of both RasGRP1 and 3 requires their phosphorylation by DAG-dependent PKCs (Aiba et a!., 2004; Brodie et al., 2004; Roose et a!., 2005; Teixeira et al., 2003; Zheng et a!., 2005). However, for RasGRP 1 it is clear that a direct DAG-C 1 domain interaction contributes to the membrane localization step of its activation process. Phospholipase Cyl is required for, and DAG kinase is inhibitory to, T cell antigen receptor-induced translocation of RasGRP 1 to membranes (Bivona et al., 2003; Reynolds et a!., 2004; Sanjuan et al., 2003), while deletion of the Cl domain eliminates binding of RasGRP1 to membranes (Caloca et al., 2003a; Ebinu et a!., 1998; Tognon et al., 1998). The isolated Cl domain of RasGRP1 translocates in vivo in response to DAG (Carrasco and Merida, 2004; Tognon et a!., 1998) and phorbol esters (Ebinu et al., 1998; Tognon et al., 1998) and binds directly to DAG (Carrasco and Merida, 2004) and phorbol esters (Ebinu et a!., 1998; Irie et al., 2004; Lorenzo et al., 2000).  While direct binding of the Cl domain of RasGRP3 to DAG has not yet been demonstrated, it does bind phorbol ester (Irie et al., 2004). The large differences in affinities for DAG versus phorbol esters of several PKC Cl domains (Cho and Stahelin, 2005) shows that caution is needed in assuming that phorbol ester binding equates with DAG binding, although it hasn’t been possible to predict from the sequence of a Cl domain whether it would have higher affinity for DAG versus PE, or vice versa. Antigen receptor-induced membrane translocation of RasGRP3 also requires its Cl domain and phospholipase C activity (Oh-hora et al., 2003), and is induced by either DAG or phorbol ester (Lorenzo et al., 2001). Therefore, the available evidence supports the hypothesis that RasGRP3, like RasGRP 1, is regulated by membrane translocation via a DAG-C 1 domain interaction.  RasGRP4 has several splice variants (Li et a!., 2003; Yang et a!., 2002). Two of these, which we will refer to as RasGRP4a and RasGRP4j3, have intact Cl domains, although they 46  differ by the insertion of five amino acids within the RasGRP4I3 Cl domain (Figure 2.1A). Phorbol ester treatment induces membrane translocation and activation of RasGRP4ct (Reuther et al., 2002). Both RasGRP4a and 13 synergize with phorbol esters to induce transformation of fibroblasts, while another RasGRP4 splice variant with a truncated Cl domain is unable to do this (Li et al., 2003; Yang et al., 2002).  The isolated Cl domain of Ra5GRP4cL binds directly to phorbol esters within phospholipid vesicles (Irie et al., 2004). Even though phorbol ester binding by the RasGRP4J3 Cl domain, as well as DAG binding by either of the RasGRP4 Cl domains, have not been examined, the combination of results so far is compatible with the hypothesis that both Ra5GRP4cL and RasGRP4j3 are activated via membrane translocation driven by binding of DAG to their Cl domains. However, the evidence for this is not well established, particularly for Ra5GRP4f3.  The evidence for regulation of the Rap-specific exchange factor RasGRP2 via a DAG-C 1 domain interaction is somewhat contradictory. RasGRP2 is activated by phorbol esters in several cell types (Clyde-Smith et al., 2000; Crittenden et al., 2004; Dupuy et al., 2001; Kawasaki et al., 1998) and the ability of RasGRP2 to enhance T cell antigen receptorinduced adhesion via Rap GTPase activation is dependent on TCR-coupled phospholipase C enzymes (Katagiri et al., 2004). Because translocation of RasGRP2 to membranes and its activation of membrane-associated Rap GTPases is induced by PMA (Clyde-Smith et al., 2000) and a variety of receptors coupling to phospholipase C enzymes (Crittenden et al., 2004; Eto et al., 2002; Guo et al., 2001; Katagiri et al., 2004), a frequently stated assumption is that RasGRP2 is equivalent to RasGRP 1 and 3 in being regulated by binding of its Cl domain to DAG-enriched membranes. However, a synthetic RasGRP2 Cl domain had no detectable affinity for phorbol esters in a vesicle binding assay (Irie et al., 2004), in contrast to the Cl domains of RasGRP 1, 3 and 4a. This result has to be interpreted with some caution, because the Cl domains of munc-13 and unc-13, which bind phorbol ester when expressed in cells (Betz et al., 1998; Kazanietz et al., 1995a), failed to bind phorbol ester when synthesized and analysed in the same way as the RasGRP2 Cl domain (Irie et al., 2004). Therefore, it is possible that the lack of phorbol ester binding reflected misfolding of the synthetic RasGRP2 Cl domain. Binding to DAG was not examined with the synthetic 47  RasGRP2 Cl domain. Moreover, the localizations of full length RasGRP2 versus RasGRP1 are different in several cell types (Bivona et al., 2003; Caloca et al., 2003a). This is not what is expected if the localization of both of these RasGRPs is entirely determined by their Cl domains seeking out membranes enriched in DAG. However, other domains of both RasGRP1 (Carrasco and Merida, 2004) and RasGRP2 (Caloca et al., 2003b; Caloca et al., 2004) can have dominant effects on their localizations, and thus could obscure equivalent contributions of their Cl domains to localization. Thus, it remains an open question whether or not the Cl domain of RasGRP2 contributes to its localization to membranes, and if so, whether it does this by binding DAG.  The objective of our current study was to determine the capabilities of the Cl domains of RasGRP2, 4a and 4(3 to mediate membrane binding via DAG. Our in vivo and in vitro experiments demonstrated that the Cl domain of RasGRP4a binds DAG within membranes, although with moderately reduced affinity relative to the RasGRP 1 Cl domain. In contrast, the Cl domains of RasGRP2 and 4f3 did not bind DAG, although they were equivalent to the other Cl domains in being able to bind membrane vesicles enriched in anionic phospholipids. From these results, we conclude that RasGRP2 and RasGRP4I3 cannot be directly regulated by DAG binding to their Cl domains, thus disproving the common assumption that all RasGRPs are regulated by membrane translocation driven by a DAG-Ci domain interaction (Bunney and Katan, 2006; Clyde-Smith et al., 2000; Crittenden et al., 2004; Eto et al., 2002; Guo et al., 2001; Mitin et al., 2005).  Instead, our finding that  all of the RasGRP Cl domains can interact with anionic phospholipids suggests that this mechanism could contribute to membrane localization of RasGRPs.  48  2.2  MATERIALS AND METHODS  2.2.1  Construction and expression of GFP/C1 domain fusion proteins cDNAs encoding the Cl domains were generated by PCR from plasmids encoding  full length proteins and fused with the enhanced green fluorescence protein (GFP) cDNA from pEGFP-C1 (Clontech). The sequences of the encoded Cl domain peptides are shown in Figure 2.1A. All Cl domains are murine, with the exception of human Raf-1. The sequence of the C-terminal portion of the GFP peptide in these constructs is: ...DELYKSGLRSLKST, with this sequence being fused to the N-termini of the Cl domain sequences in Figure 2.1A.  The fusion cDNAs encoding the GFP-C1 domain constructs  were cloned into the retroviral vector pCTV21 1 (Guilbault and Kay, 2004), converted to retroviral particles by transfection into the BOSC 23 packaging cell line (Pear et al., 1993) and transduced into the NIH 3T3 (American Type Culture Collection, ATCC) WEHI-23 1 (ATCC) or DOl 1.10 (Haskins et al., 1983) cell lines. Polyclonal populations of transduced cells were obtained by puromycin selection.  2.2.2  Fluorescence microscopy Transduced NIH 3T3 cells were cultured in Dulbecco’s modified Eagle’s medium  containing 4.5 g/l glucose (DMEM) plus 10% calf serum on cover slips for 24 hours, then transferred to the same medium containing 10% or 0% calf serum for an additional 16 hours. The cells were treated with 2 tM PMA (Sigma-Aldrich, Oakville, ON) or 100 jiM dioctanoylglycerol (a short chain DAG) (Sigma-Aldrich) from stocks in DMSO, or with DMSO alone, for 5 minutes. DO 11.10 T cells and WEHI-231 B cells were adhered to poly L-lysine coated coverslips and cultured in DMEM 10% fetal bovine serum for 24 hours. After fixation with 4% formaldehyde in PBS for 2 minutes, the cells were permeabilized with 0.2% Triton X-100 in PBS and stained with Alexa Fluor 488-conjugated anti-GFP antibody (Invitrogen, Carlsbad, CA). This increases sensitivity of detection of GFP, compared to GFP fluorescence on its own. To mark Golgi membranes, cells were co-stained with anti-GFP as above, along with mouse anti-GM13O (BD Biosciences, Mississauga, ON) followed by 49  Alexa Fluor 647-conjugated anti-mouse IgG (Invitrogen). Endoplasmic reticulum was marked by treating unfixed cells with glibenclamide BODIPY-texas red (ER Tracker, Invitrogen), followed by fixation with formaldehyde in PBS and staining with anti-GFP as described above. Images were photographed with an Axioplan 2 imaging microscope (Carl Zeiss, Toronto, ON), using OpenLab imaging software (Improvision, Coventry, England).  2.2.3  Preparation of cellular membrane and cytosol fractions Transduced NIH 3T3 cells at 50% confluency were serum-starved for 3.5 hours by  culture in DMEM plus 1 mg/ml BSA, and then treated for the indicated time with DMEM, plus 1 mg/ml BSA containing 300 jiM dioctanoylglycerol. On ice, cells were washed 3 times with cold PBS, scraped into 0.25 ml lysis buffer (20 mM Tris-HC1, pH 7.4, 2 mM EDTA, 2 mM PMSF, 2.5 jig/mi leupeptin, 2 jig/ml chymostatin, 1 j.tg/ml antipain, 2 j.ig/ml pepstatin, 10 jig/ml benzamidine, 10 jig/ml p-aminobenzadine) and lysed by sonication. The NaCl concentration was adjusted to 0.15 M and the lysate was centrifuged at 100,000 x g to separate cytosol from particulate. The resulting pellet was resuspended by sonication in lysis buffer with 0.15 M NaC1 and 1% (w/v) Triton X- 100, and centrifuged at 100,000 x g to separate detergent-soluble membranes from insoluble particulate. Equivalent volumes of each fraction were added to Laemmli sample buffer without f3-mercaptoethanol or boiling (boiling would destroy GFP fluorescence) and were electrophoresed on 10% SDS-PAGE. Gels were immediately scanned using a Typhoon 9410 (488nm Blue2 laser, 520nm BP filter) to visualize the fluorescence of GFP-tagged proteins. The fluorescence of GFP-C 1 domain bands were quantified using ImageQuant (GE Healthcare, Piscataway, NJ). The percent in the membrane fraction was calculated as the density of the band in the detergent-soluble micro some fraction relative to the sum of the densities of the protein bands in the three sub cellular fractions. The recovery in the three subceliular fractions, expressed as a percent of the cell lysate density averaged 99  ±  9% (S.D.) with a range of 87% 116%. -  50  2.2.4  Construction of K-Ras/C1 fusions, RasGRP1A/C1 fusions and RasGRP4u and l The K-Ras Q61N A173-188 mutant has the C-terminal amino acid sequence  .KEKMSKDGSTEA (residues not naturally present in K-Ras are italicized). Fusions of K Ras Q61N A173-188 to Cl domains have the sequence ...KEKMSKDGSI with this sequence being fused to the N-termini of the Cl domain sequences in Figure 2.1. For the K Ras/pren construct, the basic cluster and prenylation signal of K-Ras, along with an HA epitope tag, was re-attached to the K-Ras Q61N A173-188 mutant, resulting in the Cterminal sequence ...KEKMSKDGSTEAYPYDYASGSRK}IKEKMSKDGKKKKKKSKTKCVIM. The Nterminus of K-Ras Q61 N A 173-188 is fused to GFP, with the junction sequence being .DELYKSGLRSFLLKMTEYKLVVV... The N-terminally GFP-tagged and C-terminally deleted form of murine RasGRP1 (RasGRP1A) has GFP fused to amino acid 2 of RasGRP1 (GenBank accession NP035376). The sequence at this N-terminal fusion site is .delyksglrssaqseGTLGKAR..., with natural . .  sequence of RasGRP1 in upper case. The sequence at the C-terminal fusion junction is ...YSKLGst[C1 domain], with the Cl domain sequences shown in Figure 2.1. Murine RasGRP4x and f3 with N-terminal GFP tags have the sequence • .  .delyksglrslksNRKDIKRKS.... at the GFP-RasGRP4 junction, with the natural sequence of  RasGRP4 (starting at amino acid 2) in upper case. The encoded RasGRP4c is identical to Genbank accession AF33 1457, while the encoded RasGRP4I3 is identical except for the five amino acid insertion in the Cl domain.  2.2.5  NIH 3T3 transformation assays eDNAs encoding the specified proteins were inserted into the retroviral vector  pCTV2 11, which were converted to retroviruses, were transduced into NIH 3T3 cells and selected as described above. Transduced cells were plated at 10% confluence. For assessment of transformation in 10% serum, the cells were cultured continually until 3 days (For K-Ras) or 5-7 days (for Ra5GRP5) after the cells had formed a monolayer, then photographed using a Nikon Elipse TS100 microscope (Nikon, Mississauga, ON). For low 51  serum transformation assays, the cells were transferred to DMEM / 0.5% fetal calf serum for 1 day, followed by another 2 days in DMEM/ 0.5% fetal calf serum with or without 0.1 jiM PMA. The cells were then photographed. The photographed regions were representative of the appearance of the cells across the entire culture dish. To quantify transformation efficiency, NIH 3 T3 cultures transduced with the indicated constructs were seeded at low density. After colonies had grown up from individual cells, they were scored as transformed or non-transformed. Transformation efficiency  # of transformed colonies/total # of  colonies. >30 colonies were scored.  2.2.6  Construction, expression and purification of GST/C1 fusion proteins cDNAs encoding Cl domains were inserted into the glutathione S-transferase (GST)  fusion vector pGEX-2T (Amersham, Piscataway, NJ), resulting in N-terminal GST fusions having the junction sequence .LVPRGSLKST, with this sequence being fused to the N ..  termini of the Cl domain sequences in Figure 2.1A. The plasmids were transformed into E. coli strain AD202 and cultures in LB medium were grown overnight at 37° to saturation, then diluted 40-fold and grown at 37° to an 0D 600 of 0.8. After 10 mm on ice, cells were induced with 0.5 mI\4 IPTG for 5 hours at 28° C. The bacterial cell pellet was resuspended in cold PBS containing 1% (w/v) Triton X100, 2 mM PMSF, and cells lysed by two passes through a French press. Insoluble material was removed by centrifugation at 15,000 x g. GST-C1 domains in the soluble fraction were bound to glutathione-Sepharose (Amersham; 2 ml of 50% slurry per litre of cell culture, equilibrated in PBS/1% Triton X-100) during a 2-hour incubation at 4°. Beads were collected on a mini-column and washed with PBS /1% Triton X100, PBS, and finally 50 mM Tris, pH 7.5, 0.15 M NaC1 before elution of GST fusion protein in 50 mM Tris pH 7.5, 0.15 M NaCl, 5 mM glutathione. Protein was quantitated by Bradford assay using ovalbumin as a standard. The GST fusion proteins were aliquotted and stored at -80° for use in the in vitro vesicle-binding assay. On average, 40 mg of the GST-C1 fusion protein, at 2 mg/ml was obtained per liter of bacterial culture.  52  2.2.7  Binding of GST-C1 domain fusion proteins to sucrose-loaded phospholipid vesicles Binding of GST-C 1 domain fusion proteins to sucrose-loaded large unilamellar  vesicles (SLVs) was measured using a protocol based on that described previously (Buser and McLaughlin, 1998; Johnson et a!., 2000). 1 -palmitoyl-2-oleoyl-phosphatidylcholine (PC), 1 ,2-dioleoyl-phosphatidylserine (PS), 1,2 dioleoyl-phosphatidylglycerol (PG), 1,2dioleoylglycerol (DAG), 1-palmitoyl, 2-oleoyl-phosphatidic acid (PA), 1-palmitoyl, 2hydroxy-glycerophosphate (lysoPA), and N-palmitoyl, D-erythro-sphingosine (ceramide) were obtained from Avanti Polar Lipids (Alabaster, AL) or Northern Lipids (Burnaby, BC), while oleic acid and arachidonic acid were from Sigma-Aldrich. Lipid vesicles were prepared by drying mixtures of lipids (including trace amounts of [ H] di-palmitoyl PC for 3 quantification of recovery) in chloroform on a rotary-evaporator, resuspending in 20 mM Hepes, pH 7.5, 170 mM sucrose by vigorous vortexing to a final lipid concentration of 2 mM, followed by five freeze-thaw cycles in liquid nitrogen. These multilamellar vesicles were stored at -20° C until use. On the day of the experiment, SLVs were prepared by extrusion at room temperature through a 100 nm membrane using a Lipofast Microextruder (Avestin, Mannheim, Germany). Extruded vesicles were diluted 5-fold in 100 mM NaC1, 20 mM Hepes pH 7.5 and centrifuged 100,000 x g for 30 mm at 25° C to dilute the free sucrose. 80% of the supernatant was removed and the resulting pellet was resuspended by vigorous vortexing in the residual volume. PMA, from a DMSO stock, was subsequently added with vigorous vortexing such that the end concentration of DMSO was <2 % (v/v). GST-C1 domain fusion proteins (0.3 8-0.48 tM) were incubated with SLVs (200 jiM, except 125 jiM for Fig. 2.6A) for 10 mm in the presence of 20 mM Hepes, pH 7.5, 100 mM NaCl, 0.05 mg/ml ovalbumin (non-specific adsorption blocker) in a volume of 120 jiL. Vesicle-bound protein was separated from unbound by centrifugation at 100,000 x g for 30 mm at 25 °C. Greater than 90% of the [ H] di-palmitoyl PC label was in the pellet fraction. Separate 3 samples of fusion proteins were treated in the same way in the absence of added SLVs to quantify vesicle-independent sedimentation. Protein in each fraction was determined by analysis of supernatant and pellet fractions, and resolved by SDS-PAGE on a Tricine gel system (Schagger and von Jagow, 1987). Gels were stained with Sypro-Orange stain (Molecular Probes, Invitrogen) according to the manufacturer’s instructions, and imaged on a 53  Typhoon 9410 (488nm Blue2 laser, 580nm BP filter). Fluorescence intensities of the GST Cl bands were quantitated using ImageQuant. Percent bound= 100% x P/(S  +  P) where P is  intensity of fluorescence band from the pellet fraction, and S is intensity of fluorescence band from the supematant fraction. The pellet intensities were corrected for contaminating supernatant, as well as protein that sedimented in the absence of vesicles. The latter correction resulted in some % bound values being less than zero. The apparent association constant, Ka = (B/F) x 1 /L where B is fraction of protein bound (in pellet), F is fraction of protein free (in supematant), and L is the molar concentration of accessible lipid (0.5 x total lipid concentration, since the Cl domain only binds the outer leaflet) was calculated.  54  2.3  RESULTS  2.3.1  Sequence comparison of RasGRP Cl domains Modeling of the RasGRP 1 Cl domain indicated that its A and B loops can form a  ligand binding pocket that is very similar to that of the Cib domain of PKCö (Rong et al., 2002). The RasGRP3 Cl domain is identical to that of RasGRP1 in ioop A except for asparagine versus threonine at the base of the loop, and has only an isoleucine/valine difference in loop B (Figure 2.1A). To assess the DAG- or phorbol ester-binding potentials of the Cl domains of RasGRP2 and the two splice variants of RasGRP4, we compared their ligand binding loops to those of RasGRP1 and 3, and to the Cib domain of PKCö and the Cl domain of f32-chimaerin (Figure 2.1A).  There are two atypical residues in ioop A of the RasGRP2 Cl domain. Position 8 is serine, versus tyrosine or phenylalanine in the DAG/phorbol ester-binding Cl domains. Mutation of tyrosine 8 to glycine in the PKC Clb domain reduces but does not eliminate phorbol ester binding (Kazanietz et al., 1995b). The other atypical residue in ioop A is valine at position 12, which is threonine or histidine in the DAG and phorbol ester-binding Cl domains shown in Figure 2.1A. Mutation of threonine 12 to valine in PKC Cib caused a minor reduction in phorbol ester binding (Kazanietz et al., 1995b), while changing position 12 in the DAG kinase f3 Cia domain from alanine to threonine moderately increased phorbol ester binding (Shindo et al., 2003). The significant discrepancy in loop B of the RasGRP2 Ci domain is leucine 22, which is tryptophan in the DAG and phorbol ester-binding Cl domains shown in Figure 2.1. In the Cib domain of PKCf3, switching this residue from tyrosine to tryptophan increases affinity for DAG and enhances colocalization with perinuclear membranes (Dries et al., 2007). Tryptophan versus tyrosine occupancy at this position has the potential to affect DAG affinity by causing major perturbations in the shape of the ligand binding pocket (Dries et al., 2007). The effect of leucine at this position is unknown. While the variations at positions 8, 12 and 22 serve as warning flags that the DAG or phorbol ester binding capabilities of the RasGRP2 Cl domain should not be taken for granted, none of these changes by themselves are predicted to eliminate binding by either of these ligands. 55  A Loop B  Loop A PKC5 Cib B2 chim. RasGRP1 RasGRP2 RasGRP3 RasGRP4  MPHRFKVYNYMSPTFCDHCGSLLWGLQGL1CCEDCGP4NVHHKCREKVANLCGINQKptea HNFKVTFGPHWCEYCANFMWGLIAQGVRCSDCGLNVHKQCSKHVPNDC 1 10 20 30 40 50 FPHNFQETTYLKPTFCDNCAGFLWGVIICQGYRCKDCGMNCHKQCKDLVVFECICKRIKptea FVHNFQESNSLRPVACRHCKALILGIYKQGLKCRACGVNCHXQCKERLSVECRRRAQptea FIHNFQEMNYLKPTFCEHCAGFLWGIIKQGYCKDCGANCHXQCKDLLVLACRRLARAPSLSSNPtea FLHFQEVTFKPTFCHSCSGFLWGVQGYRCRDCGLCCHR!CRDQVRVECKXRPtea  Raf -1 KSR  TTHNFARXTFLKLAFCDICQKFLLHRFSTKSWLSQV- CHVCQKSMI-  -  VSTGP  insert in RasGRP4  =  -  -  -  -  -NGFRCQTCGYKFHEHCSTKVPTMCkkrikptea FGVKCKHCRLKCHNKC TXEAPAC  -  -  B  site of insertion in Ra5GRP4I3  Figure 2.1: Cl domain sequences A. The Cl domains of the RasGRPs are compared to four Cl domains of known structure. The Cl b domain of PKCö and the Cl domain of 132-chimaerin bind DAG and phorbol esters while the CI domains of Raf-l and KSR do not. The histidine and cysteine residues which coordinate zinc are indicated by the shaded bars. Loop A and B residues are italicized. Cationic residues which have the potential to interact with anionic phospholipids within membranes are bolded. For the CI domains used in this study, the sequences extending to the C-termini of the expressed proteins are shown, with non-natural amino acids in lower case. The N-termini indicate where the Cl domains were fused to GFP, GST or K-Ras. Numbering of the residues within the Cl domains is shown above the RasGRP1 Cl domain sequence. B. The Cl b domain of PKCö (protein data bank code 1 ptq) illustrates the generic structure of a Cl domain that binds DAG or phorbol ester. The zinc atoms are shown as orange balls. Y8, T12 and W22 are highlighted in yellow. Four of the cationic residues implicated in membrane interactions, K26, K30, K41 and K45, are highlighted in blue. The image was generated with PyMol.  56  The RasGRP4a Cl domain has no atypical residues in loop A or B relative to the DAG/phorbol ester-binding Cl domains shown in Figure 2.1A. Therefore, it is expected to bind both DAG and phorbol esters. In RasGRP4f3 there is an insertion of five amino acids at the N-terminus of loop B. While this insertion could radically disrupt the ligand binding pocket, it is possible that the inserted amino acids project outward, with most of loop B remaining in position to form one side of the ligand-binding pocket. In the latter case, the RasGRP4f3 Cl domain could retain binding of DAG and/or phorbol ester, or could have altered ligand specificity relative to the Cl domain of RasGRP4a.  All of the RasGRP Cl domains have cationic residues within and C-terminal to their Cl domains (Figure 2.1 A) which are positioned to interact electrostatically with anionic phospholipids (Hurley, 2006). In conjunction with the hydrophobic residues projecting from loops A and B, these have the potential to either cooperate with DAG to enhance membrane binding affinity or to mediate DAG-independent interactions with membranes.  To summarize, the Cl domain of RasGRP4a is expected to bind both DAG and phorbol ester based on its similarities in ioop A and B to other Cl domains, but the capabilities of the Cl domains of RasGRP2 and 4f3 to bind either of these ligands cannot be predicted with confidence from their sequences. All of the RasGRP Cl domains have the potential to bind membranes independently of DAG or phorbol ester, via electrostatic and hydrophobic interactions. A combination of in vivo and in vitro experiments are needed to resolve the question of whether all RasGRP Cl domains can mediate the membrane binding required for juxtaposing these exchange factors with their Ras or Rap GTPase substrates, and to assess whether the Cl domains achieve this by binding directly to DAG.  2.3.2  Only the Cl domains of RasGRP 1, 3 and 4c.i co-localize with membranes and translocate in response to DAG or phorbol ester Fusions of Cl domains with GFP were expressed in NIH 3T3 fibroblasts to compare  the localization of the five RasGRP Cl domains to the DAG-binding PKC Cib domain versus the Raf-1 Cl domain, which does not bind DAG. The GFP/C1 domain fusions for the RasGRPs and PKCö included the clusters of cationic amino acids which are naturally found 57  immediately C-terminal to the Cl domains, as these have been shown to contribute to membrane binding in vivo and in vitro (Irie et al., 2004; Tognon et a!., 1998). For the Raf-1 Cl domain, the cationic cluster from RasGRP1 was attached to ensure that any deficiencies in membrane binding by the Raf- 1 Cl domain were not simply due to its lack of a cluster of cationic amino acid residues.  In NIH 3T3 cells cultured in 10% serum, the Raf-1 Cl domain, as well as GFP alone, was distributed throughout the nucleus and cytoplasm (Figure 2.2A). In contrast, the PKCö Clb domain and the RasGRP1 Cl domain were excluded from the nucleus and concentrated in the perinuclear cytoplasm where they co-localized with internal membranes, as shown by staining GFP-RasGRP1 Cl domain-expressing cells with ER Tracker [glibenclamide, which binds to sulfonylurea receptors in the endoplasmic reticulum (Hambrock et al., 2002)] or with antibody to the Golgi-associated protein GM13O (Nakamura et a!., 1995) (Figure 2.2B). The Cl domain of RasGRP3 was also largely co-localized with internal membranes, while the Cl domain of RasGRP4a was similar but with more accumulation away from internal membranes, particularly evident as partial localization in the nucleus. The Cl domains of RasGRP2 and RasGRP4f3 were radically different, being distributed throughout the cytoplasm and nucleus, as was seen for the Raf-l Cl domain and GFP alone. A similar pattern of Cl domain localization was seen in the T cell line DO 11.10 and the B cell line WEHI-23 1, with the Cl domains of RasGRP 1 and 3 strongly co-localizing with internal membranes, the Cl domains of RasGRP2 and 4f3 being dispersed throughout the cells, and the Cl domain of RasGRP4cL being partially excluded from the nucleus but only weakly co localizating with internal membranes (Figure 2.3).  58  0% serum  A  10% serum  +  DAG  +  PMA  GFP  PKC  Raf-1  RasGRPI  RasGRP2  RasGRP3  RasGRP4a  Ra5GRP4I3  B  RasGRP1  ER Tracker  Figure 2.2 Only the Cl domains of RasGRP1, 3 and 4a co-localize with membranes and translocate in response to DAG or phorbol ester  59  Figure 2.2. Only the Cl domains of RasGRP1, 3 and 4ci co-localize with membranes and translocate in response to DAG or phorbol ester (page 58) A. GFP fusions of the indicated Cl domains, or GFP alone, were expressed in NIH 3T3 cells. After culture in DMEM containing 10% or 0% calf serum +1- 100 pM dioctanoylglycerol (DAG) or 2 pM PMA as indicated, the cells were fixed and photographed by fluorescence microscopy. B. NIH 3T3 cells expressing the GFP-RasGRPI Cl domain fusion were stained with either ER Tracker to mark endoplasmic reticulum or anti-GM13O to mark Golgi membranes, as described in Materials and Methods. Individual cells showing fluorescence from the GFP-tagged RasGRPI Cl domain and either ER Tracker or GMI3O staining are shown.  Culturing NIH 3T3 cells in medium lacking serum caused a considerable shift in the distribution of the RasGRP4o Cl domain away from perinuclear membranes and into the nucleus (Figure 2.2A). Serum starvation had no noticeable effect on the other Cl domains. Relative to the RasGRP 1 and 3 Cl domains, the RasGRP4cL Cl domain appears to be less co localized with internal membranes and more dependent on serum stimulation to maintain its localization at those membranes. Nonetheless, it is clearly distinguished from the Cl domains of RasGRP2 and RasGRP4f3 which lack co-localization with membranes.  Treatment of the serum-starved NIH 3T3 cells with dioctanoylglycerol, a short chain DAG, induced accumulation of the Cl domains of PKCö and RasGRP 1, 3 and 4cL in the nuclear envelope, while retaining their localization in the perinuclear region (Figure 2.2A). This membrane-selective relocalization presumably reflects accumulation of exogenous DAG at the nuclear envelope. Exogenous DAG could also have accumulated at the ER and Golgi, but with no discernible effect on the Cl domains because they were already located at these sites, potentially by endogenous DAG. PMA also caused relocalization of the Cl domains of RasGRP1, 3 and 4cL, and PKCö. PMA had no observable effect on the localizations of the Cl domains of Raf-1, RasGRP2 or RasGRP4f3 (Fig. 2.2A), nor did DAG when used at either 100 .tM for 5 minutes (Fig. 2.2A) or 300 1 iM for either 5 or 15 minutes (data not shown).  60  A  GFP  RasGRPI  RasGRP2  C  B  ER Tracker  0  Figure 2.3 Distributions of Cl domains in 0011.10 T cells and WEHI-231 B cells A and C. GFP fusions of the indicated Cl domains, or GFP alone, were expressed in DOl 1.10 T cells or WEHI-231 B cells. After culture in DMEM containing 10% fetal calf serum the cells were fixed and photographed. B and D. DO11.10 orWEHI-231 cells expressing the GFP-RasGRP1 Cl domain fusion were stained with either ER Tracker to mark endoplasmic reticulum or anti-GMI3O to mark Golgi membranes, as described in Materials and Methods.  61  50 0 0 Cu I“  a) -  E 20 a) E • 10  0 PKC6  RG1  RG2  RG4o  RG4  Ha Raf  Figure 2.4 DAG induces trans location of the Cl domains of RasGRP1, 3 and 4o to membranes NIH 3T3 ceNs expressing the indicated GFP-tagged Cl domains were cultured in serum-free medium for 3.5 h, then treated for 15 mm with DMSO (white bars) or 300 tM dioctanoylglycerol (DAG, grey bars). Cells were fractionated and the GFP-C1 proteins in the membrane versus non-membrane fractions were quantified as described in Materials and Methods. Data are means +1- standard errors from two independent experiments.  Differential responses of the Cl domains of RasGRP 1 and RasGRP4ct versus RasGRP2 and RasGRP4f3 were also observed when DAG-induced membrane binding was assessed by fractionation of NIH 3T3 cells. GFP-tagged Cl domains were quantitated in cell lysates, cytosol, a detergent-solubilized particulate fraction (regarded as the membrane fraction), and a detergent-insoluble particulate fraction. The percent of each Cl domain in the membrane fraction is shown in Figure 2.4. In serum-starved cells, less than 10% of each Cl domain was in the membrane fraction. After DAG treatment 3 5-40% of the Cl domains of PKC, RasGRP1 and RasGRP4a were membrane-associated, whereas the RasGRP2 and RasGRP4J3 Cl domains, like the Raf-1 Cl domain, did not show significant DAG-induced changes in membrane binding. 62  2.3.3  The Cl domains of RasGRP1 and 4a bind directly to DAG within phospholipid membranes, while the Cl domains of RasGRP2 and 4 do not To determine if the differences in membrane localization of the RasGRP Cl domains  reflected differences in their abilities to directly bind DAG or phorbol ester within phospholipid bilayers, we compared their binding to unilamellar phospholipid vesicles in vitro. Cl domains were expressed in E. coli as GST fusion proteins, and purified by affinity chromatography on glutathione-agarose. The GST-Cl fusion proteins ran as the expected single 34 kDa species on SDS-PAGE, with the exception of the GST-RasGRP3 Cl construct which had a significant amount of a smaller 26 kDa GST species present, indicating partial premature termination of translation near the GST/Cl junction (data not shown).  Binding of the Cl domains to phospholipid bilayers was assessed by their cosedimentation with sucrose-loaded large unilamellar vesicles (SLVs) composed of the neutral phospholipid phosphatidylcholine (PC) supplemented with 5 mol% of the anionic phospholipid phosphatidylserine (PS), either alone or in combination with 5 mol% longchain DAG (l,2-dioleoylglycerol), or 1 mol% PMA. The PKC Cib and Raf-l Cl domains were used as positive and negative controls to test this experimental system (Figure 2.5A). The PKCö Clb domain bound poorly to the PC/5% PS vesicles, and very well to vesicles containing DAG or PMA. In contrast, the Raf-l Cl domain had no significant vesicle binding in the absence or presence of DAG or PMA.  The RasGRP 1 Cl domain was very similar to the PKCö C lb domain, with the addition of either DAG or PMA inducing nearly complete binding to the vesicles (Figure 2.5A). The RasGRP3 and 4cx Cl domains were equivalent to the Cl domain of RasGRP1 in having nearly complete binding to the PMA-containing vesicles, but bound less effectively to DAG-containing vesicles (Figure 2.5A). The Cl domains of RasGRP2 and RasGRP4I3 had minimal vesicle binding, and this was not altered by the addition of DAG or PMA.  63  A 100 80 .  60  0  40 20 0 -20 PKC6  Raf  RGI  B  V D 0 .0  RG2  RG3 RG4x RG413  C 100  100-  80  80 V  60  -  60-  0  40  40-  20  20-  0  0  00.5 1.0 1.5 mol % DAG  2.0  PKC6  RGI  RG4a  Figure 2.5 Binding of the Cl domains of RasGRP1, 3, and 4cL to DAG or PMA within membrane vesicles A. GST fusions of the indicated Cl domains were mixed with SLVs composed of PC and 5 mol% PS (open bars) or 5 mol% PS and 5 mol% DAG (black bars) or 5 mol% PS and 1 mol% PMA (striped bars). Binding was assessed by co-sedimentation as described in Materials and Methods. Some values are less than 0% because of correction for the amount of protein which sedimented in the absence of vesicles. Data are means +1- range of 2 independent experiments. B. Binding of GST fusions of the RasGRPI (D), RasGRP4a (s), and PKC Cl () domains to PC SLVs containing 10 mol% PS with various mol% DAG. Data are means +1- range of 2 independent experiments. C. Binding of GST fusions of the RasGRP1, RasGRP4o and PKCö Cl domains to PC SLVs containing 2 mol% DAG and either no anionic lipid (open bars), 5 mol% PS (black bars), 2.5 mol% PS and 2.5 mol% PG (striped bars), or 5 mol% PG (hatched bars). Data are means +1- range of 2 independent experiments.  64  These vesicle binding properties mirror the results we obtained with the in vivo experiments, and confirm that direct binding to DAG and PMA is restricted to the Cl domains of RasGRP1, 3 and 4o. This initial vesicle binding experiment also indicated that there could be quantitative heterogeneity among these three DAG-binding Cl domains. However, the lower DAG binding of the RasGRP3 Cl domain seen in Figure 2.5A may be artifactual, because the protein was produced as a mixture of full-length and truncated GST/Cl fusion proteins. Because GST dimerizes, this would result in some GST-GST/Cl heterodimers, which would have reduced avidity for membranes relative to GST/C1-GST/Cl homodimers, due to the presence of one versus two DAG-binding sites within the dimer. Considering this anomaly, the RasGRP3 Cl domain was not used in subsequent experiments aimed at quantifying differences in DAG-dependent and DAG-independent membrane binding.  2.3.4  The Cl domains of RasGRP1 and 4c. have different affinities for DAG Figure 2.5B compares the binding of the RasGRP1 versus 4cL Cl domains to PC!l0  mol% PS vesicles with DAG concentrations ranging from 0 to 2 mol%. Binding of the RasGRP4c Cl domain was less sensitive to DAG at low concentrations, and reached saturation at 2 mol% DAG, versus 1 mol% DAG for the RasGRP1 Cl domain. The binding curve for the PKCö Cib domain was intermediate between those ofRasGRPl and 4a at low DAG concentrations, and like the RasGRP1 Cl domain it reached saturation at a lower DAG concentration than was required for the RasGRP4ct Cl domain. From these binding curves we calculated apparent affinity constants for membranes containing 10 mol% PS and 0.2 mol% DAG. The values are 21,000, 7500, and 3800 M’ for the RasGRP1, PKC, and RasGRP4u Cl domains, respectively. Figure 2.5B also indicates that the RasGRP1 Cl domain binds more strongly to 10 mol% PS vesicles containing no DAG, in comparison to the RasGRP4a or PKC Cib domains.  The role of anionic phospholipid in the response to DAG is shown in Figure 2.5C. Binding of the Cl domains of PKCö, RasGRP1 or RasGRP4a to vesicles containing 2 mol% DAG was enhanced by either of the anionic phospholipids PS or phosphatidylglycerol (PG), 65  included at 5 mol%. Of the three, the RasGRP4 Cl domain showed the weakest binding and was the most dependent on anionic phospholipid. Membrane binding of the RasGRP 1 and 4c Cl domains does not specifically require the phosphatidylserine head group. Instead, they appear to respond to the negative charge on the membrane surface, as is the case for the PKCö Cib domain as well as other PKC Cl domains (Johnson et al., 2000).  2.3.5  Phosphatidic acid, lysophosphatidic acid, ceramide, fatty acids and sphingosme-1-phosphate are not alternate ligands for RasGRP2 or RasGRP4 The RasGRP2 and 4f3 Cl domains do not bind to DAG, but their sequences are  compatible with their having a modified pocket structure that could bind an alternative ligand. To address the hypothesis that the Cl domains of RasGRP2 and 4f3 are specialized to recognize lipid second messengers other than DAG, we tested candidate lipid ligands which have small head groups suitable for occupying the pocket and which act as signaling molecules.  Phosphatidic acid was of particular interest as a ligand, because it can be generated from DAG via DAG kinases which have been functionally coupled to RasGRPs (Regier et al., 2005; Sanjuan et al., 2003; Topham and Prescott, 2001; Zha et al., 2006). However, phosphatidic acid at 5 mol% did not promote vesicle binding of any of the Cl domains, under conditions in which 5 mol% DAG promoted nearly complete binding of the Cl domains of RasGRP1, RasGRP4o and PKCö (Figure 2.6A). Other signaling lipids with larger headgroups that we examined, lyso-phosphatidic acid (lysoPA) and sphingosine 1phosphate, also failed to promote binding of the Cl domains of RasGRP1, 2 or 4 (Figure 2.6B). Ceramide is structurally similar to DAG and has been postulated as a potential ligand for DAG binding-incompetent Cl domains of Raf- 1, PKCs and DAG kinases, although a direct interaction with these or any other Cl domain has not been demonstrated (Kashiwagi et al., 2002; van Blitterswijk, 1998; van Blitterswijk et al., 2003).  66  A  100-  80-  0 .0  4O-  r’  20-  o PKC6  RGI  RG2  RG4cr  I  I Raf  RG413  B 80  60 V  c 40  0 .0 C  20  0  -  20  nil  OA  AA  Cer.  Sph-IP  LPA  DAG  Figure 2.6 Lack of binding of RasGRP Cl domains to alternative lipid ligands A. GST fusions of the indicated RasGRP Cl domains were assayed for binding to PC SLVs (open), or PC SLVs containing 10 mol% PS (grey), 10 mol% PS and 5 mol% DAG (black), or 10 mol% PS and 5 mol% PA (striped). The percent bound values were not corrected for sedimentation in the absence of lipid. Data are means +1- range of 2 independent experiments. B. GST fusions of the Cl domains of RasGRP1 (open bars), RasGRP2 (black bars), and RasGRP4I3 (striped bars) were mixed with PC SLVs containing 5 mol% PS, and 5 mol% of oleic acid (CA), arachidonic acid (AA), ceramide, sphingosine-1 -phosphate (sph-1-P), lysophosphatidic acid (lysoPA), or 1 mol% DAG. Data are means +1- range of 2 independent experiments.  67  Ceramide at 5 mol% was also ineffective as a ligand for the Cl domains of the RasGRP1, 2 or 4f3 (Figure 2.6B). Fatty acids, particularly polyunsaturated species such as arachidonic acid, have been reported to influence the activity of some PKC isoforms (Khan et al., 1995). Arachidonic acid causes a C lb-dependent redistribution of PKCE in CR0 cells (Kashiwagi et al., 2002), suggesting that it could be a ligand for this and other Cl domains. Arachidonic acid and another fatty acid, oleic acid, induced minor increase in membrane binding of the Cl domains ofRasGRPl and 2 (Figure 2.6B). However, this is likely due to the increased membrane negative charge provided by these anionic lipids (see below).  2.3.6  High concentrations of anionic phospholipids enable membrane binding by RasGRP Cl domains in the absence of DAG Electrostatic interactions with anionic phospholipids have the potential to provide a  DAG-independent mechnism for membrane binding by Cl domains. The anionic lipid dependence for membrane binding in the absence of DAG was tested in Figure 2.7A. All RasGRP Cl domains bound vesicles in an anionic lipid-dependent manner. GST alone did not bind these vesicles, regardless of the anionic lipid composition (data not shown). All Cl domains were >60% bound to vesicles containing 40 mol% PS, but sensitivities to lower PS concentrations were variable. The RasGRP1 Cl domain was the most PS-sensitive of all the Cl domains tested, with considerable binding to vesicles containing just 10 mol% PS and only minor increases in binding at higher PS concentrations. Along with the PKCö Cib domain, the RasGRP4 Cl domain had the weakest binding response, requiring greater than 20 mol% PS for appreciable binding. The Cl domains of RasGRP2, 4cL and Raf- 1 had intermediate dependencies on PS, with transitions from low to high binding occurring between 10% and 20% PS.  In addition to their charges, the specific structures of anionic phospholipid headgroups could influence Cl domain binding to membranes. This is of considerable biological interest, because the Cib domain of PKCö has selectivity for PS which contributes  68  A 100  ..:.:*  !  /  80 60 :  •PKC6 A Raf 4 RG2 •RG4a  /  0 RG413  30 mol% PS  B 100  PKC6  Raf  RG1  RG2  RG4o  RG413  Figure 2.7 RasGRP Cl domains bind to vesicles enriched in anionic phospholipids A. Dependence of Cl domain binding on the mol% PS. GST fusions of the indicated Cl domains were mixed with SLVs composed of PC and the indicated amounts of PS. Data are means +1- range of 2 independent experiments. The curves were generated by the sigmoidal dose dependence variable slope option of Prism Graph Pad, with the exception of the curve for the RasGRPI Cl domain which was drawn by hand to give a sigmoidal curve fitting the known % bound value of 20% at 5 mol% PS (Fig. 5A). B. Cl domain binding to anionic phospholipids is not head group-specific. GST fusions of the indicated Cl domains were assayed for binding to SLVs containing PC alone (open bars), or 70 mol% PC and 30 mol% PS (solid black bars), 30 mol% PG (striped bars), or 30 mol% PA (hatched bars). Data are means +1- range of 2 independent experiments.  69  to its preferential localization at the PS-enriched plasma membrane (Stahelin et al., 2005; Stahelin et al., 2004). We observed a selectivity for PS only for the Cib domain of PKC (Figure 2.7B).  The RasGRP Cl domains, like the Raf-1 Cl domain, showed no significant  preference for PS versus PG versus PA, and therefore are likely to interact with anionic phospholipids strictly through an electrostatic interaction rather than specific headgroup recognition. The data in Figure 2.7B also show weaker binding of RasGRP4f3 and PKCö Cl domains to anionic lipids compared to the others.  These experiments demonstrate that all RasGRP Cl domains can bind to membranes in the absence of DAG, if anionic phospholipids are present at sufficient concentrations. Because anionic phospholipid concentrations range between 10% and 20% of total lipid content in cellular membranes (Daum, 1985), this property of RasGRP Cl domains may be of biological significance, and in particular could contribute to the differential targeting of RasGRPs to specific membranes within cells. In the following experiments, we tested the hypothesis that the Cl domains of RasGRP2 and RasGRP4f3 could contribute to membrane binding in vivo despite being unable to bind to DAG.  2.3.7  Only the DAG-binding Cl domains of RasGRPs can complement a membrane binding deficiency mutation in K-Ras Complementation of a membrane localization-defective mutant of K-Ras has been  used previously to demonstrate the ability of the RasGRP 1 Cl domain to confer membrane binding (Tognon et al., 1998). We used this approach to test the abilities of the other RasGRP Cl domains to bind membranes in vivo, for the purpose of determining if membrane binding could occur even when the Cl domain was unable to bind specifically to DAG.  In NIH 3T3 cells, constitutive signalling from the mutationally-activated Q6 iN form of K-Ras induces oncogenic transformation, detectable by cell contraction from the substratum, high refractility and loss of contact inhibition. Signal transduction by K-Ras is  70  A  K-Ras Q6IN M73-188  I  K-Rash  I  K-Rash+C1  B  10% serum  nil  0.5% serum PMA  K-Rash  ,%. _3.  IS  K Rash +pren  -,.‘  -  -  -  +  K-Rash RasGRPI Cl  +  K-Rash RasGRP2 Cl  +  K-Rash RasGRP3 Cl  +  K-Rash RasGRP4a Cl  +  K-Rash RasGRP4I3 Cl  Figure 2.8 Only the Cl domains of RasGRP1, 3 and 4u provide serum- or phorbol ester-dependent complementation of a membrane binding-deficient K-Ras mutant A. Structures of K-Ras proteins. K-Rash is shorthand for K-Ras Q6IN M73-188. K-Rash + pren is K-Ras Q61 N 1 73-188 with the K-Ras basic cluster + prenylation signal re-attached. K-Rash + Cl is fusion of a Cl domain to the C-terminus of K-Ras Q61 N 1 73-1 88. B. Transformation assays. The indicated constructs were expressed in NIH 3T3 cells. After culture in medium containing 10% calf serum, or 0.5% serum +1- PMA, the cell cultures were photographed to distinguish those forming a non-refractile contact-inhibited monolayer (non-transformed) from those exhibiting contraction from the substratum, high refractility and loss of contact inhibition (transformed via K-Ras activation). 71  entirely dependent on its membrane localization, which is naturally provided by C-terminal prenylation, in combination with a polybasic cluster of amino acids (Figure 2.8A). As a result, the Q61N A 173-188 K-Ras double mutant (K-RasA), which has a deletion of the Cterminal basic cluster and prenylation signal, is non-transforming. Reattachment of the basic cluster and prenylation signal to K-RasA resulted in restoration of transformation, which was evident even when the cells were cultured in only 0.5% serum (Figure 2.8B).  Attachment of the RasGRP1 Cl domain to the C-terminus of K-RasA resulted in transformation when the cells were cultured in 10% serum, but not when they were cultured in 0.5% serum. The serum-independent transformation by the prenylated K-Ras versus the serum-dependent transformation by the K-Ras utilizing the RasGRP1 Cl domain may reflect depletion of a serum-induced Cl ligand, presumably DAG. This interpretation is supported by the observation that transformation via the K-Ras/RasGRP 1 Cl fusion protein did occur when the low serum medium was supplemented with a Cl domain ligand, PMA. Equivalent experiments using DAG supplementation were impractical because DAG is rapidly metabolized.  The Cl domains of RasGRP 3 and 4x were equivalent to the RasGRP 1 Cl domain in conferring serum- or PMA-dependent complementation of the membrane binding mutation in K-RasA (Figure 2.8B). In contrast, the fusions of K-RasA to the Cl domains of RasGRP2 or  4f3 were non-transforming under any conditions. This experiment confirms that stable membrane binding and responsiveness to PMA are restricted to the Cl domains of RasGRP 1, 3 and 4x, and demonstrates that the Cl domains of RasGRP2 and 4f3 are incapable of conferring membrane binding to K-Ras at the level and/or correct subcellular location required to trigger NIH 3 T3 cell transformation. 2.3.8  Despite lacking discernible membrane localization in vivo, the Cl domain of RasGRP2 can functionally replace the Cl domain within RasGRP1 Expression of RasGRP1 also induces transformation of NIH 3T3 cells, via its  stimulation of GTP loading of Ras GTPases (Tognon et al., 1998). Because all Ras GTPases  are membrane-localized, RasGRP1 presumably has to interact with membranes to act on its 72  substrate Ras, although this interaction could be weak and transient. Deletion of its Cterminal region including the Cl domain completely eliminates membrane localization and transforming activity of RasGRP 1, while re-attachment ofjust the Cl domain fully restores membrane localization and transforming activity (Tognon et al., 1998). Along with the observation that the Cl domain can be functionally replaced by a membrane-localization signal, this demonstrates that transformation by RasGRP 1 is dependent on the ability of its Cl domain to confer membrane localization (Tognon et al., 1998). This enables functional replacement of the Cl domain of RasGRP 1 to serve as a test for the ability of another Cl domain to confer membrane binding sufficient to support Ras activation by RasGRP 1.  Similarly to their effects on K-Ras, the Cl domains of RasGRP3 and 4a were also able to restore transformation via the deleted form of RasGRP1 (RasGRP1A), while the Cl domain of RasGRP4I3 did not (Figure 2.9A, B). Unexpectedly, the Cl domain of RasGRP2 was as effective as the DAG-binding Cl domains of RasGRP 1, 3 or 4ct in restoring the transforming activity of RasGRP 1 A. The localization of RasGRP 1 A fused to the RasGRP 1 or 3 Cl domains was very similar to the localization of the isolated Cl domains, being concentrated in the regions occupied by ER and golgi (Figure 2.9B). In contrast, the RasGRP1A fusions to the Cl domains of RasGRP2, 4cL and 4f3 were much more diffusely distributed, such that they were not readily distinguishable from GFP alone. Therefore, the abilities of the RasGRP2 and 4ct Cl domains to functionally replace the Cl domain within RasGRP 1 occurs despite their inabilities to mimic the observable membrane localization properties of the RasGRP1 Cl domain. It is evident from this experiment that although membrane localization of RasGRPs is essential for accessing membrane-bound Ras and Rap, this can occur at a low and transient level which is not evident by microscopy. Membrane binding by the RasGRP4a Cl  73  GFP  A  GEF  EF  Cl  RasGRP 1 RasGRP1t+C1 transformation efficiency  B GFP  +  +  +  +  +  0  RasGRP1A RasGRP1C1  0.95  RasGRP1t RasGRP2C1  0.88  RasGRP1A RasGRP3C1  0.95  RasGRP1A RasGRP4a1C1  0.88  RasGRPIA RasGRP4F3CI  C  D  GFP  0  Cl  GEF  ZLZ—EZ O—L1—-—cL—— RasGRP4a  RasGRP4I3 RasGRP43  Figure 2.9 The Cl domains of RasGRP1, 2,3 and 4z are functional within RasGRPs, while the Cl domain of RasGRP4 is not 74  Figure 2.9 The Cl domains of RasGRP1, 2,3 and 4a are functional within RasGRPs, while the Cl domain of RasGRP4P is not (page 73). A. Structure of RasGRPI compared to the deleted form of RasGRPI used to test functionality of attached Cl domains. The guanine nucleotide exchange domains (GEE), EF hands (EF) and Cl domains are shown, along with the N-terminal GFP tag. B. NIH 3T3 cells expressing the indicated RasGRP1 + Cl fusion proteins, or expressing GFP alone as a control, were assessed for oncogenic transformation (low magnification pictures of cell cultures at left, with high refractility and loss of contact inhibition indicating transformation), and for localization of the GEP-tagged proteins (higher magnification fluorescence microscopy of typical individual cells at right). The efficiency of transformation of each construct was determined as described in Materials and Methods. C. Structures of GFP-tagged RasGRP4a and j3. The bar represents the five amino acid insertion in the Cl domain. D. NIH 3T3 cells were transduced with retroviral vectors expressing the GFP-tagged RasGRP4 constructs and photographed by fluorescence microscopy.  domain appears to be partially destabilized by attachment of the RasGRP 1 A protein, but a weak interaction with DAG in membranes may explain the ability of this Cl domain to restore transforming activity to RasGRP 1 A. However, the experiments demonstrating a lack of DAG binding by the RasGRP2 Cl domain imply that the RasGRP 1 A  +  RasGRP2 Cl  domain fusion protein must be activated by another mechanism, possibly reflecting the ability of the RasGRP2 Cl domain to bind weakly to membranes via anionic phospholipids.  The Cl domain of RasGRP4f3 was unable to activate RasGRP1A (Figure 2.9B), thus distinguishing it from the RasGRP2 Cl domain. It is possible that the RasGRP4f3 Cl domain makes a functionally significant contribution to membrane localization, but that this is effective only in cooperation with other domains of RasGRP4. To test this, we compared the localizations of RasGRP4a versus RasGRP4f3 in NIH 3T3 cells, as these two proteins differ only by the five amino acid insertion in the Cl domain of RasGRP4f3 (Figure 2.9C). RasGRP4cc was nuclear excluded and partially concentrated in the perinuclear region occupied by ER (Figure 2.9D), which was similar to the distribution of the isolated Cl domain of RasGRP4ct (Figure 2.2A). In contrast, RasGRP4f3 was distributed throughout the cells (Figure 2.9D), equivalent to the distribution of GFP alone or the isolated Cl domain of RasGRP4f3 (Figure 2.2A). These results indicate that the Cl domain is the primary determinant of RasGRP4ct membrane localization in NIH 3T3 cells, and that the five amino acid insertion in the RasGRP4f3 Cl domain renders it non-functional as a membrane localizer. 75  2.4  DISCUSSION The occurence of Cl domains in all RasGRP proteins, followed by the convincing  demonstration that RasGRP 1 is regulated by DAG binding directly to its Cl domain, has fostered the assumption that all RasGRPs are regulated in the same way as RasGRP1 (Bunney and Katan, 2006; Crittenden et al., 2004; Eto et al., 2002; Guo et al., 2001; Mitin et al., 2005). We have directly addressed the question of whether all RasGRP Cl domains are functionally equivalent or distinct by several independent experimental approaches  —  fluorescence microscopy and cell fractionation of the distribution of GFP tagged Cl domains, direct lipid vesicle binding assays, and PMA- and serum-dependent complementation by Cl domains of membrane binding defective K-Ras. The results were consistent in showing that the RasGRP Cl domains divide into two distinct classes. The translocation of the Cl domains of RasGRP 1, 3, and 4a to membranes can be driven by their binding to DAG or its functional analog PMA, although the RasGRP4ct Cl domain has reduced affinity for DAG in vitro and shows less intense co-localization with membranes in vivo. In contrast, the Cl domains of RasGRP2 and 4f3 were unable to bind DAG or PMA and did not detectably co-localize with membranes in vivo. Previous reports of phorbol esterinduced or PLC-dependent activation of RasGRP2 or 4f3, which were interpreted as evidence for direct binding of their Cl domains to phorbol ester or DAG (Clyde-Smith et al., 2000; Crittenden et al., 2004; Dupuy et aL, 2001; Katagiri et al., 2004; Kawasaki et aL, 1998; Yang et aL, 2002), may have instead reflected the involvement of DAG or phorbol ester-dependent PKCs in the activation of these two RasGRPs.  In the RasGRP2 Cl domain, the combined effects of the alterations at positions 8, 12 and 22 may alter the structure of the ligand binding pocket sufficiently to prevent DAG or phorbol ester binding. The position 8 alteration is of particular interest, because mutation of this residue from tyrosine to serine (the residue in Ra5GRP2) eliminates localization of RasGRP1 to internal membranes (Caloca et al., 2003a), which could reflect loss of DAG binding. Our data demonstrate that the five amino acid insertion at the base of the B loop of the ligand binding pocket of the RasGRP4f3 Cl domain is sufficient to eliminate DAG as well as phorbol ester binding, and also eliminates membrane binding as detected by microscopy or complementation of the membrane binding-defective mutants of K-Ras or 76  RasGRP 1.  The alternative splicing event affecting this Cl domain apparently provides a  mechanism for generating two functionally distinct forms of RasGRP4, one with and one without the capability of being activated by DAG-mediated translocation to membranes.  RasGRP 2 and 4f3 presumably underwent opportunistic evolution away from DAG mediated regulation following the expansion of the RasGRP family via gene duplication and acquisition of alternative splicing. Conservation of the basic structure of these two Cl domains, and their retention of membrane binding via anionic phospholipids, suggests that they still make significant contributions to the interactions of RasGRP2 and 4f3 with membranes. We tested the hypothesis that these Cl domains have evolved to recognize different lipid signal transducers, but neither these nor the other Cl domains had specific binding to PA, lysoPA, ceramide, sphingosine 1-phosphate, oleic acid or arachidonic acid in the vesicle binding assay. An alternative hypothesis is that the Cl domains of RasGRP2 and  4f3, as well as those of the other RasGRPs, provide a weak membrane binding site via their electrostatic interactions with anionic phospholipids. The surface of the PKC6 Cib domain contains basic residues which are positioned to interact with an anionic membrane surface (Rong et al., 2002; Zhang et al., 1995). The RasGRP Cl domains contain these and additional basic residues positioned appropriately for interaction with anionic surfaces, particularly at positions 10 and 32 (Figure 2.1). The RasGRP Cl domains are also bordered at their C-termini by a three or four residue basic patch (Figure 2.1A), which enhances binding of the RasGRP1 and 3 Cl domains to PMA/PS micelles (Irie et al., 2004). Our analyses have demonstrated that all Cl domains bind phospholipid vesicles in proportion to the anionic lipid content, and that this binding is insensitive to the specific headgroup and can occur in the absence of DAG. In the case of a DAG-binding Cl domain, the weak electrostatic interaction with negatively charged phospholipds may facilitate a 2-dimensional search for DAG on the membrane surface, as well as reinforcing the membrane binding strength of the ligated Cl domain. For the Cl domains of RasGRP2 and 4f3, a DAG independent electrostatic interaction with membranes is evidently insufficient to dictate strong membrane binding in vivo, since RasGRP2 and 4 did not noticeably co-localize with membranes or complement the membrane binding-defective K-Ras mutant. However, the RasGRP2 Cl domain can functionally replace the Cl domain within RasGRP 1. Because the 77  Cl domain is essential and sufficient for RasGRP1 membrane localization leading to Ras GTP loading (Tognon et al., 1998), this implies that the RasGRP2 Cl domain can provide membrane binding to a physiologically significant extent despite its lack of detectable DAG binding and lack of membrane localization observable by microscopy. It appears that the association of RasGRP 1 with membranes can be quite weak and still be sufficient for Ras activation, as detected by transformation of NIH 3T3 cells. In contrast, K-Ras may have to be more stably bound to membranes in order to transduce signals sufficient to maintain transformation of NIH 3T3 cells.  In addition to the RasGRP2 Cl domain, there are other examples of Cl domains that do not bind DAG and do not provide microscopy-detectable binding to membranes on their own, but which nonetheless are required for efficient localization to membranes. The Raf- 1 Cl domain is required to stabilize membrane binding via the adjacent Ras-binding domain (Bondeva et al., 2002), while the Cl domain of KSR makes an essential contribution to constitutive localization to internal membranes and is needed for cytokine-induced translocation to the plasma membrane (Zhou et al., 2002). The inability of the RasGRP4I3 Cl domain to confer transforming activity on either RasGRP1 or RasGRP4 may be due to its lower binding to anionic phospholipids, relative to the Cl domain of RasGRP2, as seen in Figure 2.7. It is possible that under some circumstances (although not in NIH 3T3 cells) this Cl domain can also make physiologically significant contributions to membrane binding by RasGRP4. However, it could be that the alternative splicing event which converts RasGRP4o into RasGRP4o has the purpose of eliminating any contribution of the Cl domain to RasGRP4 activation.  78  2.5  BIBLIOGRAPHY  Aiba, Y., Oh-hora, M., Kiyonaka, S., Kimura, Y., Hijikata, A., Mori, Y. and Kurosaki, T. (2004) Activation of RasGRP3 by phosphorylation of Thr- 133 is required for B cell receptor-mediated Ras activation. Proc NatlAcadSci USA, 101, 16612-16617. Ananthanarayanan, B., Stahelin, R.V., Digman, M.A. and Cho, W. (2003) Activation mechanisms of conventional protein kinase C isoforms are determined by the ligand affmity and conformational flexibility of their Cl domains. JBiol Chem, 278, 46886-46894. Betz, A., Ashery, U., Rickmann, M., Augustin, I., Neher, E., Sudhof, T.C., Rettig, J. and Brose, N. (1998) Munc 13-1 is a presynaptic phorbol ester receptor that enhances neurotransmitter release. Neuron, 21, 123-136. Bivona, T.G., Perez De Castro, I., Ahearn, I.M., Grana, T.M., Chiu, V.K., Lockyer, P.J., Cullen, P.J., Pellicer, A., Cox, A.D. and Philips, M.R. (2003) Phospholipase Cgamma activates Ras on the Golgi apparatus by means of RasGRP 1. Nature, 424, 694-69 8. Bondeva, T., Balla, A., Vamai, P. and Balla, T. (2002) Structural determinants of Ras-Raf interaction analyzed in live cells. MolBiol Cell, 13, 2323-2333. Brodie, C., Steinhart, R., Kazimirsky, G., Rubinfeld, H., Hyman, T., Ayres, J.N., Hur, G.M., Toth, A., Yang, D., Garfield, S.H., Stone, J.C. and Blumberg, P.M. (2004) PKCdelta associates with and is involved in the phosphorylation of RasGRP3 in response to phorbol esters. Mol Pharmacol, 66, 76-84. Brose, N., Betz, A. and Wegmeyer, H. (2004) Divergent and convergent signaling by the diacylglycerol second messenger pathway in mammals. Curr Opin Neurobiol, 14, 328-340. Bunney, T.D. and Katan, M. (2006) Phospholipase C epsilon: linking second messengers and small GTPases. Trends Cell Biol, 16, 640-648. Buser, C.A. and McLaughlin, 5. (1998) Ultracentrifugation technique for measuring the binding of peptides and proteins to sucrose-loaded phospholipid vesicles. Methods Mol Biol, 84, 267-281. Caloca, M.J., Zugaza, J.L. and Bustelo, X.R. (2003a) Exchange factors of the RasGRP family mediate Ras activation in the Golgi. JBiol Chem, 278, 33465-33473. Caloca, M.J., Zugaza, J.L., Matallanas, D., Crespo, P. and Bustelo, X.R. (2003b) Vav mediates Ras stimulation by direct activation of the GDP/GTP exchange factor Ras GRP1. Embo J 22, 3326-3336. Caloca, M.J., Zugaza, J.L., Vicente-Manzanares, M., Sanchez-Madrid, F. and Bustelo, X.R. (2004) F-actin dependent translocation of the Rapl GDP/GTP exchange factor RasGRP2. JBiol Chem, 279, 2043520446. Canagarajah, B., Leskow, F.C., Ho, J.Y., Mischak, H., Saidi, L.F., Kazanietz, M.G. and Hurley, J.H. (2004) Structural mechanism for lipid activation of the Rac-specific GAP, beta2-chimaerin. Cell, 119, 407418. Carrasco, S. and Merida, I. (2004) Diacylglycerol-dependent binding recruits PKCtheta and RasGRP1 Cl domains to specific subcellular localizations in living T lymphocytes. Mol Biol Cell, 15, 2932-2942. Carrasco, S. and Merida, I. (2006) Diacylglycerol, when simplicity becomes complex. Trends Biochem Sci. Cho, W. and Stahelin, R.V. (2005) Membrane-protein interactions in cell signaling and membrane trafficking. Annu Rev Biophys Biomol Struct, 34, 119-151. Clyde-Smith, J., Silins, G., Gartside, M., Grimmond, S., Etheridge, M., Apolloni, A., Hayward, N. and Hancock, J.F. (2000) Characterization of RasGRP2, a plasma membrane-targeted, dual specificity Ras/Rap exchange factor. JBiol Chem, 275, 32260-32267. Colon-Gonzalez, F. and Kazanietz, M.G. (2006) Cl domains exposed: from diacylglycerol binding to proteinprotein interactions. Biochim Biophys Acta, 1761, 827-837. Crittenden, J.R., Bergmeier, W., Zhang, Y., Piffath, C.L., Liang, Y., Wagner, D.D., Housman, D.E. and Graybiel, A.M. (2004) Ca1DAG-GEFI integrates signaling for platelet aggregation and thrombus formation. Nat Med, 10, 982-986. Daum, G. (1985) Lipids of mitochondria. Biochim Biophys Acta, 822, 1-42. Dries, D.R., Gallegos, L.L. and Newton, A.C. (2007) A single residue in the Cl domain sensitizes novel protein kinase C isoforms to cellular diacylglycerol production. JBiol Chem, 282, 826-830. Dupuy, A.J., Morgan, K., von Lintig, F.C., Shen, H., Acar, H., Hasz, D.E., Jenkins, N.A., Copeland, N.G., Boss, G.R. and Largaespada, D.A. (2001) Activation of the Rapl guanine nucleotide exchange gene, Ca1DAG-GEF I, in BXH-2 murine myeloid leukemia. JB1 1 Chem, 276, 11804-11811. 0  79  Ebinu, J.O., Bottorff D.A., Chan, E.Y., Stang, S.L., Dunn, R.J. and Stone, J.C. (1998) RasGRP, a Ras guanyl nucleotide- releasing protein with calcium- and diacylglycerol-binding motifs. Science, 280, 10821086. Eto, K., Murphy, R., Kerrigan, S.W., Bertoni, A., Stuhlmann, H., Nakano, T., Leavitt, A.D. and Shattil, S.J. (2002) Megakaryocytes derived from embryonic stem cells implicate Ca1DAG-GEFI in integrin signaling. Proc NatlAcadSci USA, 99, 12819-12824. Gomez-Femandez, J.C., Torrecillas, A. and Corbalan-Garcia, S. (2004) Diacylglycerols as activators of protein kinase C. Mol Membr Biol, 21, 339-349. Guilbault, B. and Kay, R.J. (2004) RasGRP1 sensitizes an immature B cell line to antigen receptor-induced apoptosis. JB1o1 Chem, 279, 19523-19530. Guo, F.F., Kumahara, E. and Saffen, D. (2001) A CaIDAG-GEFI/Rap l/B-Raf cassette couples M(1) muscarinic acetylcholine receptors to the activation of ERK1/2. JBiol Chem, 276,25568-25581. Hall, C., Lim, L. and Leung, T. (2005) Cl, see them all. Trends Biochem Sci, 30, 169-171. Hambrock, A., Loffler-Walz, C. and Quast, U. (2002) Glibenclamide binding to suiphonylurea receptor subtypes: dependence on adenine nucleotides. Br J Pharmacol, 136, 995-1004. Haskins, K., Kubo, R., White, J., Pigeon, M., Kappler, J. and Marrack, P. (1983) The major histocompatibility complex-restricted antigen receptor on T cells. I. Isolation with a monoclonal antibody. JExp Med 157, 1149-1 169. Hurley, J.H. (2006) Membrane binding domains. Biochim Biophys Acta, 1761, 805-811. Hurley, J.H., Newton, A.C., Parker, P.J., Blumberg, P.M. and Nishizuka, Y. (1997) Taxonomy and function of Ci protein kinase C homology domains. Protein Sci, 6, 477-480. Irie, K., Masuda, A., Shindo, M., Nakagawa, Y. and Ohigashi, H. (2004) Tumor promoter binding of the protein kinase C Cl homology domain peptides of RasGRPs, chimaerins, and Unc 13s. Bioorg Med Chem, 12, 4575-4583. Johnson, J.E., Giorgione, J. and Newton, A.C. (2000) The Ci and C2 domains of protein kinase C are independent membrane targeting modules, with specificity for phosphatidylserine conferred by the Cl domain. Biochemistry, 39, 11360-11369. Jones, DR., Sanjuan, M.A., Stone, J.C. and Merida, I. (2002) Expression of a catalytically inactive form of diacyiglycerol kinase alpha induces sustained signaling through RasGRP. Faseb J, 16, 595-597. Kashiwagi, K., Shirai, Y., Kuriyama, M., Sakai, N. and Saito, N. (2002) Importance of C lB domain for lipid messenger-induced targeting of protein kinase C. J Biol Chem, 277, 18037-18045. Katagiri, K., Shimonaka, M. and Kinashi, T. (2004) Rap I-mediated lymphocyte function-associated antigen-i activation by the T cell antigen receptor is dependent on phospholipase C-gammal. JBiol Chem, 279, 11875-11881. Kawasaki, H., Springett, G.M., Toki, S., Canales, J.J., Harlan, P., Blumenstiel, J.P., Chen, E.J., Bany, l.A., Mochizuki, N., Ashbacher, A., Matsuda, M., Housman, D.E. and Graybiel, A.M. (1998) A Rap guanine nucleotide exchange factor enriched highly in the basal ganglia. Proc Nati AcadSci USA, 95, 13278-13283. Kazanietz, M.G., Lewin, N.E., Bruns, J.D. and Blumberg, P.M. (1995a) Characterization of the cysteine-rich region of the Caenorhabditis elegans protein Unc-i3 as a high affinity phorbol ester receptor. Analysis of ligand-binding interactions, lipid cofactor requirements, and inhibitor sensitivity. JBiol Chem, 270, 10777-10783. Kazanietz, M.G., Wang, S., Milne, G.W., Lewin, N.E., Liu, H.L. and Blumberg, P.M. (1995b) Residues in the second cysteine-rich region of protein kinase C delta relevant to phorbol ester binding as revealed by site-directed mutagenesis. JBiol Chem, 270, 21852-21859. Khan, W.A., Blobe, G.C. and Hannun, Y.A. (1995) Arachidonic acid and free fatty acids as second messengers and the role of protein kinase C. Cell Signal, 7, 171-184. Li, L., Yang, Y., Wong, G.W. and Stevens, R.L. (2003) Mast cells in airway hyporesponsive C3H/HeJ mice express a unique isoform of the signaling protein Ras guanine nucleotide releasing protein 4 that is unresponsive to diacylglycerol and phorbol esters. Jlmmunol, 171, 390-397. Lorenzo, P.S., Beheshti, M., Pettit, G.R., Stone, J.C. and Blumberg, P.M. (2000) The guanine nucleotide exchange factor RasGRP is a high -affinity target for diacylglycerol and phorbol esters. Mol Pharmacol, 57, 840-846. Lorenzo, P.S., Kung, J.W., Bottorff D.A., Garfield, S.H., Stone, J.C. and Blumberg, P.M. (2001) Phorbol esters modulate the Ras exchange factor RasGRP3. Cancer Res, 61, 943-949.  80  Medkova, M. and Cho, W. (1999) thterplay of Cl and C2 domains of protein kinase C-alpha in its membrane binding and activation. JBiol Chem, 274, 19852-19861. Mitin, N., Rossman, K.L. and Der, C.J. (2005) Signaling interplay in Ras superfamily function. Curr Biol, 15, R563-574. Mott, H.R., Carpenter, J.W., Zhong, S., Ghosh, S., Bell, R.M. and Campbell, S.L. (1996) The solution structure of the Raf- 1 cysteine-rich domain: a novel ras and phospholipid binding site. Proc Nat! Acad Sci US A, 93, 8312-8317. Nakamura, N., Rabouille, C., Watson, R., Nilsson, T., Hui, N., Slusarewicz, P., Kreis, T.E. and Warren, G. (1995) Characterization of a cis-Golgi matrix protein, GM13O. JCellBiol, 131, 1715-1726. Newton, A.C. (2001) Protein kinase C: structural and spatial regulation by phosphorylation, cofactors, and macromolecular interactions. Chem Rev, 101, 2353-2364. Oh-hora, M., Johmura, S., Hashimoto, A., Hikida, M. and Kurosaki, T. (2003) Requirement for Ras guanine nucleotide releasing protein 3 in coupling phospholipase C-gamma2 to Ras in B cell receptor signaling. JExp Med 198, 1841-1851. Pear, W.S., Nolan, G.P., Scott, M.L. and Baltimore, D. (1993) Production of high-titer helper-free retroviruses by transient transfection. Proc NatlAcadSci USA, 90, 8392-8396. Regier, D.S., Higbee, J., Lund, K.M., Sakane, F., Prescott, S.M. and Topham, M.K. (2005) Diacylglycerol kinase iota regulates Ras guanyl-releasing protein 3 and inhibits Rap 1 signaling. Proc Nat! Acad Sd U SA, 102, 7595-7600. Reuther, G.W., Lambert, Q.T., Rebhun, J.F., Caligiuri, M.A., Quilliam, L.A. and Der, C.J. (2002) RasGRP4 is a novel Ras activator isolated from acute myeloid leukemia. JBiol Chem, 277, 30508-305 14. Reynolds, L.F., de Bettignies, C., Norton, T., Beeser, A., Chernoff J. and Tybulewicz, V.L. (2004) Vavi transduces T cell receptor signals to the activation of the Ras/ERK pathway via LAT, Sos, and RasGRP1. JBio! Chem, 279, 18239-18246. Rong, S.B., Enyedy, I.J., Qiao, L., Zhao, L., Ma, D., Pearce, L.L., Lorenzo, P.S., Stone, J.C., Blumberg, P.M., Wang, S. and Kozikowski, A.P. (2002) Structural basis of RasGRP binding to high-affinity PKC ligands. JMed Chem, 45, 853-860. Roose, J.P., Mollenauer, M., Gupta, V.A., Stone, J. and Weiss, A. (2005) A diacylglycerol-protein kinase C RasGRP I pathway directs Ras activation upon antigen receptor stimulation of T cells. Mol Cell Biol, 25,4426-4441. Sanjuan, M.A., Pradet-Balade, B., Jones, D.R., Martinez, A.C., Stone, J.C., Garcia-Sanz, J.A. and Merida, I. (2003) T cell activation in vivo targets diacylglycerol kinase alpha to the membrane: a novel mechanism for Ras attenuation. Jlmmuno!, 170, 2877-2883. Schagger, H. and von Jagow, G. (1987) Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from ito 100 kDa. Anal Biochem, 166, 368-379. Shindo, M., Irie, K., Masuda, A., Ohigashi, H., Shirai, Y., Miyasaka, K. and Saito, N. (2003) Synthesis and phorbol ester binding of the cysteine-rich domains of diacylglycerol kinase (DGK) isozymes. DGKgamma and DGKbeta are new targets of tumor-promoting phorbol esters. JBio! Chem, 278, 18448-18454. Stahelin, R.V., Digman, M.A., Medkova, M., Ananthanarayanan, B., Melowic, H.R., Rafter, J.D. and Cho, W. (2005) Diacylglycerol-induced membrane targeting and activation of protein kinase Cepsilon: mechanistic differences between protein kinases Cdelta and Cepsilon. J Biol Chem, 280, 19784-19793. Stahelin, R.V., Digman, M.A., Medkova, M., Ananthanarayanan, B., Rafter, J.D., Melowic, H.R. and Cho, W. (2004) Mechanism of diacyiglycerol-induced membrane targeting and activation of protein kinase Cdelta. JBio! Chem, 279, 29501-295 12. Stone, J.C. (2006) Regulation of Ras in lymphocytes: get a GRP. Biochem Soc Trans, 34, 858-861. Teixeira, C., Stang, S.L., Zheng, Y., Beswick, N.S. and Stone, J.C. (2003) Integration of DAG signaling systems mediated by PKC-dependent phosphorylation of RasGRP3. Blood 102, 1414-1420. Tognon, C.E., Kirk, H.E., Passmore, L.A., Whitehead, I.P., Der, C.J. and Kay, R.J. (1998) Regulation of RasGRP via a phorbol ester-responsive Cl domain. Mo! Cell Biol, 18, 6995-7008. Topham, M.K. and Prescott, S.M. (2001) Diacyiglycerol kinase zeta regulates Ras activation by a novel mechanism. JCe1I Biol, 152, 1135-1143. van Blitterswijk, W.J. (1998) Hypothesis: ceramide conditionally activates atypical protein kinases C, Raf-1 and KSR through binding to their cysteine-rich domains. Biochem J 331, 679-680. van Blitterswijk, W.J., van der Luit, A.H., Veldman, R.J., Verheij, M. and Borst, J. (2003) Ceramide: second messenger or modulator of membrane structure and dynamics? Biochem J 369, 199-211.  81  Wang, Q.J., Fang, T.W., Nacro, K., Marquez, V.E., Wang, S. and Blumberg, P.M. (2001) Role of hydrophobic residues in the Clb domain of protein kinase C delta on ligand and phospholipid interactions. JBIoI Chem, 276, 19580-19587. Yang, Y., Li, L., Wong, G.W., Krilis, S.A., Madhusudhan, M.S., Sali, A. and Stevens, R.L. (2002) RasGRP4, a new mast cell-restricted Ras guanine nucleotide-releasing protein with calcium- and diacylglycerol binding motifs. Identification of defective variants of this signaling protein in asthma, mastocytosis, and mast cell leukemia patients and demonstration of the importance of RasGRP4 in mast cell development and function. JBI Z Chem, 277, 25756-25774. 0 Zha, Y., Marks, R., Ho, A.W., Peterson, A.C., Janardhan, S., Brown, I., Praveen, K., Stang, S., Stone, J.C. and Gajewski, T.F. (2006) T cell anergy is reversed by active Ras and is regulated by diacyiglycerol kinase-alpha. Nat Immunol, 7, 1166-1173. Zhang, G., Kazanietz, M.G., Blumberg, P.M. and Hurley, J.H. (1995) Crystal structure of the cys2 activatorbinding domain of protein kinase C delta in complex with phorbol ester. Cell, 81, 917-924. Zheng, Y., Liu, H., Coughlin, J., Zheng, J., Li, L. and Stone, J.C. (2005) Phosphorylation of RasGRP3 on threonine 133 provides a mechanistic link between PKC and Ras signaling systems in B cells. Blood, 105, 3648-3654. Zhou, M., Horita, D.A., Waugh, D.S., Byrd, R.A. and Morrison, D.K. (2002) Solution structure and functional analysis of the cysteine-rich Cl domain of kinase suppressor of Ras (KSR). J Mol Biol, 315, 435-446. Zugaza, J.L., Caloca, M.J. and Bustelo, X.R. (2004) Inverted signaling hierarchy between RAS and RAC in T lymphocytes. Oncogene, 23, 5823-5833.  82  CHAPTER 3: LOCALIZATION OF RASGRP2 IS SPECIFIED BY THE Cl DOMAIN AND A POTENTIAL BINDING SITE FOR PDZ PROTEINS  A version of this chapter will be submitted for publication: Goulding RE and Kay RJ. Localization of RasGRP2 is specified by the Cl domain and a potential binding site for PDZ proteins.  83  3.1  INTRODUCTION RasGRP2 is a guanine nucleotide exchange factor with specificity for the Rap 1,  Rap2, R-Ras and TC-21 GTPases (Dupuy et al., 2001; Eto eta!., 2002; Katagiri et aL, 2004; Kawasaki et al., 1998; Ohba et a!., 2000b; Ohba et al., 2000a). RasGRP2 is selectively expressed in the brain (Kawasaki et al., 1998) platelets, megakaryocytes (Crittenden et al., 2004) and T-cells (Ghandour et a!., 2007). Proviral insertions in the RasGRP2 locus have been shown to result in B-cell lymphoma (Mikkers et al., 2002; Suzuki et a!., 2002) and myeloid leukemia (Dupuy et a!., 2001), which is presumed to reflect deregulated expression of RasGRP2. The normal pattern of expression in B-cells and myeloid cells has not been investigated. Targeted disruption of RasGRP2 leads to severe defects in integrin-mediated platelet aggregation, adhesion, spreading and thrombus formation (Bernardi et a!., 2006; Crittenden et al., 2004), and abolishes LFA-l mediated adhesion of primary human T-cells to ICAM-1, in response to the chemokine SDF-la (Ghandour et a!., 2007).  In order to be active, RasGRP2 must be localized to cell membranes where Rap 1 GTPases are constitutively bound. All members of the RasGRP family (RasGRP 1, 2, 3 and 4) have a single Cl domain. Some Cl domains bind diacylglycerol (DAG), which is generated at membranes by a number of signal transduction mechanisms including phospholipase C (PLC)-mediated cleavage of phosphatidylinositol 4,5-bisphosphate (PIP2) into DAG and inositol 1,4,5-triphosphate (Carrasco and Merida, 2006). There is evidence that membrane localization of RasGRP 1, 3 and 4 is regulated by binding of their Cl domains to DAG. RasGRP1 (Bivona et al., 2003; Caloca et al., 2004; Ebinu et a!., 1998; Rambaratsingh et al., 2003; Tognon et al., 1998), RasGRP3 (Lorenzo eta!., 2001; Shao et al., 2001) and RasGRP4 (Katsoulotos et a!., 2008) translocate to membranes in response to treatment with DAG or phorbol ester (PE), which acts as a DAG mimetic. RasGRP1 (Caloca et al., 2004; Ebinu et al., 1998; Kawasaki et al., 1998; Priatel et al., 2002; Rambaratsingh et al., 2003), RasGRP3 (Lorenzo et a!., 2001) and RasGRP4 (Katsoulotos et al., 2008) have been shown to activate their target GTPases, as a consequence of DAG or PE treatment. A combination of microscopy, membrane fractionation, vesicle binding and direct binding assays has shown that the Cl domain of RasGRP1 (Ebinu et al., 1998; Irie et a!., 2004; 84  Johnson et al., 2007; Lorenzo et a!., 2000; Madani et al., 2004; Rong et aL, 2002), RasGRP3 (Irie et al., 2004; Johnson et a!., 2007) and RasGRP4 (Irie et a!., 2004; Johnson et al., 2007; Reuther et al., 2002) all bind DAG or PE. The Cl domain of RasGRP1 is required for serum-stimulated fibroblast transformation (Tognon et a!., 1998) and its translocation to endoplasmic reticulum and Golgi membranes in fibroblasts and COS cells (Caloca et a!., 2003; Toki et al., 2001). Antigen receptor stimulation of lymphocytes, which leads to DAG production at membranes via PLOy cleavage of PIP2, also leads to RasGRP1 translocation to the plasma membrane or Golgi (Beaulieu et a!., 2007; Bivona et al., 2003; Caloca et a!., 2004; Perez de Castro et al., 2004; Zugaza et al., 2004). B-cell receptor (BCR) mediated translocation of RasGRP1 (Beaulieu et a!., 2007) or RasGRP3 (Oh-hora et a!., 2003) to the plasma membrane requires the Cl domain. In addition, the Cl domain is required for T-cell receptor (TCR) (Roose et al., 2007) or BCR (Beaulieu et a!., 2007) mediated activation of RasGRP 1. Thus it has been established that the Cl domain is required for membrane localization of RasGRP 1 and RasGRP3, and also for RasGRP 1 activation.  Compared to other RasGRPs, relatively little is known about RasGRP2 localization. In BHK cells RasGRP2 was found to localize in the cytosol, with no evidence of plasma membrane localization (Clyde-Smith et a!., 2000), while in COS cells RasGRP2 displays a dispersed localization in the cytoplasm with some detection in plasma membrane ruffles (Caloca et al., 2004). It is not clear what role the Cl domain plays in regulating the localization and activation of RasGRP2. Most evidence implicating the Cl domain in the regulation of RasGRP2 comes from studies using PE treatment. In response to prolonged PE exposure RasGRP2 accumulates in the subcellular fraction of COS cells which includes membranes (Clyde-Smith et al., 2000), however no PE-induced change in localization of RasGRP2 is observable in these cells using standard microscopy techniques (Caloca et al., 2004). Both RasGRP2-mediated Rapi activation in COS or 293T cells (Clyde-Smith et al., 2000; Kawasaki et al., 1998) and RasGRP2-mediated adhesion of 32D cells to fibronectin (Dupuy et al., 2001) can be induced by PE. In addition, RasGRP2-deficient mouse platelets are unable to activate Rapi or aggregate in response to PE (Crittenden et al., 2004). sbRNA mediated silencing of RasGRP2 in primary human T-cells also leads to abrogation of PE induced Rapl activation and adhesion to ICAM-1 (Ghandour et a!., 2007). Also, RasGRP285  mediated TCR-induced adhesion of Jurkat T-cells to ICAM- 1 via Rap 1 activation is dependent on PLCs , which has been interpreted as reflecting direct binding of the Cl domain of RasGRP2 to DAG generated by PLCs (Katagiri et a!., 2004). However, when interpreting these results it is important to note that RasGRP1 and RasGRP3 have been shown to be phosphorylated and activated by protein kinase Cs (PKC5) (Teixeira et a!., 2003; Zheng et al., 2005), which also contain DAG-binding Cl domains and are themselves activated by DAG or PE (Colon-Gonzalez and Kazanietz, 2006). Thus, it remains a possibility that activation of RasGRP2 by DAG or PE occurs via PKCs rather than the Cl domain of RasGRP2. In support of this hypothesis, PE-induced Rapi activation in human T cells was shown to be inhibited by the PKC inhibitor Go6850 (Ghandour et al., 2007).  While RasGRP1, 3 and 4 Cl domains directly bind PE (Irie et al., 2004) or DAG (Johnson et al., 2007), the RasGRP2 Cl domain does not (Irie et a!., 2004; Johnson et al., 2007). Sequence analysis of the RasGRP2 Cl domain shows that although the basic structure of two histidines and six cysteines that are required for Zn 2 binding is intact, this Cl domain diverges from the Cl domains of RasGRP1, 3 and 4 at three key residues (Irie et al., 2004; Johnson et al., 2007), which are predicted to be important for forming the DAG/PE binding pocket (Colon-Gonzalez and Kazanietz, 2006). Although unable to bind DAG with high affinity, the Cl domain of RasGRP2 can bind to membranes enriched in anionic phospholipids (Johnson et aL, 2007) and thus has the potential to mediate localization to membranes by this mechanism.  In this study we examined the role of the Cl domain as a determinant of RasGRP2 localization in fibroblasts and T-cells. We found that RasGRP2 is targeted to two membrane locations: the plasma membrane and the Golgi. The Cl domain is responsible for plasma membrane localization and is required for SDF-la induced translocation of RasGRP2 to the plasma membrane. In contrast, localization of RasGRP2 to the Golgi occurs independently of the Cl domain and instead requires a PDZ-binding motif.  86  3.2  MATERIALS AND METHODS  3.2.1  Cell lines and reagents NIH 3T3 cells from American Type Culture Collection (ATCC) (Manassas, VA)  were cultured in Dulbecco’s modified Eagle’s medium (DMEM, StemCell Technologies Vancouver, BC) containing 10% bovine calf serum (Hyclone Laboratories, Logan, UT). Jurkat cells from ATCC were cultured in RPMI 1640 (StemCell Technologies) containing 10% fetal bovine serum (Hyclone Laboratories, Logan, UT). DO 11.10 cells (Morgan et al., 1999) were obtained from Barbara Osborne (University of Massachusetts), and were cultured in DMEM containing 10% fetal bovine serum. DT4O cells were obtained from Mike Gold (University of British Columbia, Vancouver, BC), and were originally from T. Kurosaki (RIKEN Research Center, Yokohama, Japan). DT4O cells were cultured in RPMI 1640 medium (StemCell Technologies) supplemented with 10% fetal bovine serum (Hyclone Laboratories), 2% chicken serum (Invitrogen, Carlsbad, CA), and 50 iM 2-mercaptoethanol. All Jurkat and DT4O cells used in this study were transfected with an expression plasmid expressing the ecotropic retroviral receptor, to make them permissible for infection with murine retroviral vectors. Anti-human CD3E (OKT3) and anti-CD28 were from eBiosciences (San Diego, CA), and anti-chicken 1gM polyclonal antibody was from Bethyl Laboratories (Montgomery, TX). SDF-lct was from the Biomedical Research Centre (University of British Columbia). Anti-Rap 1 was from Santa Cruz Biotechnology (Santa Cruz, CA) and anti-GFP was from Roche (Laval, QC). HRP-conjugated secondary antibodies were from Jackson Immunoresearch Laboratories (West Grove, PA). Anti GM13O was from BD Biosciences (Mississauga, ON) and Alexa fluor 647-conjugated secondary antibody (AF647) and ER Tracker was from Invitrogen.  3.2.2  Construction of modified forms of RasGRP2 The N-terminally GFP-tagged form of full length murine RasGRP2 (RG2) has the  green fluorescent protein (GFP) coding sequences from pEGFP-C1 (Clontech, Mountain View, CA) fused to amino acid 1 of RasGRP2. The sequence at the GFP-RasGRP2 fusion junction in RG2 is DELYKSGLRSLKSAPAAMTSTLDL, with italics indicating the linker 87  between the C-terminus of eGFP and the N-terminus of RasGRP2. The remainder of the RasGRP2 sequence is identical to GenBank 131888394. RG2AC 1 has a deletion of amino acids 497 to 552 in RasGRP2. The sequence at the deletion junction is GGRMGSTQSVSL, with italics indicating amino acids that are not naturally in RasGRP2. RG2-PBp. has a mutation which replaces the last residue at position 0 (leucine) with three residues: alanine, serine and glutamine. The sequence at the C-terminus of RG2-PBp. is VFDIHASG, with italics indicating amino acids that are not naturally in RasGRP2. The GFP-tagged Cl domain of RasGRP2 (RG2C1) is described in (Johnson et al., 2007). RG2C1x2 was derived from this by joining two copies of the RasGRP2 Cl domain via a linker containing an HA epitope tag to serve as a hydrophilic spacer between the two Cl domains. The sequence at the junction of the two Cl domains is VECRRAQTEAYPYDVPDYASGSTFVI-fNF with italics indicating the linker, and underlining indicating the C-terminal zinc-binding cysteine of the first Cl domain and the N-terminal zinc-binding histidine of the second Cl domain. The RG2-C1+C-term construct is comprised of amino acids 496 to 608 (C-terminus) of RasGRP2, with an N-terminal GFP tag. The sequence around the GFP fusion is DELYKSGLRSLKSTFVHNFQE, with italics indicating the linker between GFP and RasGRP2 sequence. The RG2-C-term construct is comprised of amino acids 553 to 608 of RasGRP2, with an N-terminal GFP tag. The sequence around the GFP fusion is DELYKSGLRSLKSTQSVSLEG, with italics indicating the linker between GFP and RasGRP2 sequence. RG2-pr has the Cl domain and C-terminal region (amino acids 497 to 608) replaced by the basic cluster and prenylation signal of K-Ras4B. The sequence at the junction between RasGRP2 and K-Ras is GGRMGSTEA YPYDYASGSRKHKEKMSKDGKKKKKKSKTKCVIM* with italics representing the linker and the asterisk indicating the C-terminus of KRas4B. RG2-1C1 has the Cl domain and C-terminal region (amino acids 497 to 608) of RasGRP2 replaced by the Cl domain of RasGRP 1. The sequence at the junction between RasGRP2 and the RasGRP 1 Cl domain is GGRMGSTFPHNF with italics representing the short linker sequence not naturally found in either RasGRP2 or RasGRP1. The remainder of the Cl domain of RasGRP1 is as described (Johnson et al., 2007).  88  3.2.3  Retroviral transduction of cell lines Transfection of BOSC23 ecotropic packaging cells with retroviral vector plasmid  DNA was performed as described (Pear et al., 1993). Virus-containing medium was supplemented with polybrene to 20 .tg/ml and then added to an equal volume of Jurkat, DO 11.10 or DT4O cells (2 x 1 0 /ml) in appropriate medium. After 5 to 10 hours of culture, 6 2-3 volumes of medium were added. Transduced cells were selected by addition of puromycin (all RasGRP2 constructs) 30 to 48 hours following infection. GFP-positive cells were then sorted by flow cytometry. Adherent NIH 3T3 cells were transduced in the same way, except with complete changes of medium. Following antibiotic selection, cells expressing high levels of GFP-tagged forms of RasGRP2 were sorted by flow cytometry using the FACSVantage SE cell sorter (Becton, Dickinson and Company, Franklin Lakes, NJ); these cells represented about 40-80% of the total population prior to sorting, depending on transduction efficiency. For certain experiments, populations of RasGRP2 expressing cells were sorted to acquire different levels of expression (e.g. high, medium or low). After sorting, cells were kept on ice, spun down and resuspended in NIH 3T3, Jurkat, DO1 1.10 or DT4O medium with 10% serum at a density of 5 x 1 0 cells/ml or lower. Post sorting, expression levels of the different constructs were assessed by FACS analysis for all cell lines after 24-48 hours. 3.2.4  Fluorescence microscopy NIH 3T3 cells expressing different mutant forms of RasGRP2 were grown on glass  coverslips overnight and then fixed with 4% formaldehyde. Jurkat cells were plated on 0.1 mg/ml poly-L-lysine-coated glass coverslips. Prior to stimulation with SDF- 1 cc Jurkat cells were cultured in serum-free medium for 8 hours. Jurkat cells were then stimulated in Hank’s buffer (StemCell Technologies) with 10 ig/ml anti-CD3 and anti-CD28, 5 tg/ml anti-IgM or 10 ng/ml SDF-lcL, and fixed with 4% formaldehyde in PBS. To mark Golgi membranes, fixed NIH 3T3 and Jurkat cells were stained with mouse anti-GM13O followed by staining with AF647-conjugated secondary antibody. ER was marked by treating unfixed NIH 3T3 cells with glibenclamide BODIPY-Texas Red (ER Tracker; Invitrogen), followed by fixation with formaldehyde in PBS for 2 minutes at 37° C. Fluorescence microscopy images were photographed with an Axioplan 2 imaging microscope (Carl Zeiss, Toronto, ON), using a 89  450-485 nm excitation filter and a 500-545 nm emission filter for analysing GFP-tagged constructs, and a 530-585 nm excitation filter and 615 nm (long-pass) emission filter for analysis of secondary antibody AF647 staining or ER Tracker. Images were captured using OpenLab (Improvision, Coventry, England) imaging software. The individual cells displayed in the figures were chosen to be representative of the majority of the population of cells. Changes in the original images in the following figures were made: in Figure 3.1C all images were reduced in brightness, in Figure 3.1 E the image for RG2C 1 x2 was brightened to match background levels of the other two images, in Figure 3.5A RG2AC1 and RG2-PB i 1 panels were increased in brightness and contrast to match images from GFP and RG2 panels, in Figure 3.7A, B, C and D all images from each experiment were equally increased in brightness and contrast, and in Figure 3.10 contrast was increased equally in all images. For two colour fluorescent images that were subsequently merged, contrast and brightness were adjusted to a common background and saturation level in images with the same fluorescent marker [e.g. GFP, GM 130 (AF647), ER Tracker]. Images were merged using OpenLab imaging software.  3.2.5  Cell colony boundary and scratch test assays NIH 3T3 cells were seeded at a low density and allowed to grow into colonies of  between 3 and 5 mm in diameter before being counted and scored as either normal or as having a tight boundary phenotype by microscopy. A minimum of 50 colonies were counted for each mutant. Images were acquired using OpenLab imaging software with a light microscope. For scratch test assays, NIH 3T3 cells were seeded at 1 x 10 /ml in 10 cm 6 dishes. After 24 hrs scratches were made with a cell scraper, scratch areas were marked and microscopy images were photographed with a Nikon Elipse TS100 microscope (Nikon, Mississauga, ON) and acquired using OpenLab imaging software. After 16 hours, images of the same areas were captured.  90  3.2.6  Cell stimulation, lysis and affinity purification of activated Rapi GTPase by  Ra1GDS-RBD chromatography 1 x1  NIH 3T3 cells were lysed with 1 ml of ice cold lysis buffer (25 mM Hepes pH  7.5, 150 mM NaC1, 1% NP-40, 0.25% sodium deoxycholate, 10% glycerol, 25 mM NaF, 10 ) containing 2 .tg!ml aprotinin, 2 jgIml leupeptin, 4 mlvi MgCl , 1 mM EDTA, 1 mM NaMoO 2 0.5 mM PMSF and 1 mM activated 4 VO 1.5 x i0 Jurkat cells or 2.5 x 3 Na .  DT4O cells  were incubated in activation buffer (25 mM FIEPES, pH 7.2, 125 mM NaCl, 5 mM KC1, 1 mM CaC1 , 1 mM 4 2 HPO 0.5 mM MgSO 2 Na , , 2 mM 1-glutamine, 1 mM sodium pyruvate, 4 0.1% glucose, 0.1% bovine serum albumin (BSA), and 50 pM 2- mercaptoethanol) (Saxton et al., 1994) for 10 mm at 37°C before stimulation. In the case of Jurkat cells anti CD3ICD28 or SDF-1c was then added to a concentration of 10 Ig/ml or 10 ng/ml respectively, for the indicated times. For DT4O cell stimulation anti-chicken 1gM was then added to a concentration of 5 Ig/m1 to DT4O cells, for the indicated times. Cells were immediately lysed with 2.5 volumes of ice-cold lysis buffer and lysed as for NIH 3T3 cells. Lysates were incubated for 30 mm with GST-Rap binding domain (GST-RapBD) fusion protein which had been prepared and pre-bound to glutathione-agarose beads as described (Taylor et al., 2001). The GST-RapBD was expressed from pGEX4T3-Ra1GDS-RBD as described in (Spaargaren and Bischoff, 1994). After washing, samples were eluted in Bio Rad XT sample buffer and electrophoresed, levels of Rapi were detected by western blot as described below.  3.2.7  Western blot analysis Equal volumes of lysates were denatured using Bio-Rad XT sample buffer, separated  by gel electrophoresis on 12% XT-Criterion acrylamide gels (Bio-Rad Laboratories, Hercules, CA) and transferred to polyvinylidene difluoride membranes (Millipore, Billerica, MA) by electroblotting. After blocking the membranes for one hour at room temperature in TBST (25 mM Tris-HC1 pH 7.4, 3 mM KCI, 150 mM NaCl, 0.05% Tween-20) containing 5% bovine serum albumin (BSA), primary antibodies were applied to the membrane for 90 mm at room temperature in 2% BSA TBST, followed by 3 x 10 mm washes with TBST. Horseradish peroxidase-conjugated secondary antibodies were then applied for 45 mm at 91  room temperature in TBST containing 1% BSA, followed by 3 x 10 mm washes with TBST. Membranes were exposed to substrate/ECL (Santa Cruz Biotechnology) and chemiluminescence was detected using the VersaDoc 5000 imaging system (Bio-Rad Laboratories). Levels of Rap 1 -GTP and total Rap 1 were quantified by band volume analysis using Quantity One software (Bio-Rad Laboratories).  92  3.3  RESULTS  3.3.1  The Cl domain of RasGRP2 specifies plasma membrane localization Our initial goal was to identify the domains of RasGRP2 that regulate membrane  localization. Given the importance of the Cl domain in regulating the localization of other RasGRPs, we hypothesized that the Cl domain of RasGRP2 would be involved in specifying its intracellular localization. In subconfluent cultures, GFP-tagged RasGRP2 (Figure 3.1 A) was mostly dispersed throughout the cytoplasm, but a minor portion was concentrated at the Golgi membrane, as detected by its co-localization with the Golgi-associated protein GM13O (Figure 3.1B). When the NIH 3T3 cells reached confluence, RasGRP2 was also detected at the plasma membrane at sites of cell-cell contact (Figure 3.1C). To test whether the Cl domain was determining localization of RasGRP2 in NIH 3T3 fibroblasts, we expressed a form of RasGRP2 that had a deletion of the Cl domain (RG2AC1) (Figure 3.1A). GFP tagged RG2AC1 localized in the cytoplasm and at the Golgi. However, plasma membrane localization in confluent cultures was not detectable (Figure 3.1 C). Therefore, the RasGRP2 Cl domain specifies plasma membrane targeting of RasGRP2 but is not needed for Golgi targeting.  We expressed a GFP fusion construct of the isolated Cl domain of RasGRP2 (RG2C 1, Figure 3.1 D) to determine if the Cl domain was sufficient as well as necessary for plasma membrane targeting. Unlike full length RasGRP2, the isolated Cl domain did not detectably localize to cell-cell contacts in confluent NIH 3T3 cells (Figure 1 E), indicating that other parts of RasGRP2 must act in conjunction to achieve stable localization at the plasma membrane. This may be due to the Cl domain having only a weak interaction with the plasma membrane, or the Cl being required for the function of another domain of RasGRP2 which serves as the sole site of contact with the plasma membrane.  93  A GFP  GEF  EF  Cl  PB  OEZ-II-cE--I  RG2 RG2tC1  B  GFP  GM13O  C  Merge  Confluent culture  RG2Cl  l0 jm  D GFP  E Cl  cxczxcz O%pm  O%pm  21%pm  1O pm  Figure 3.1 RasGRP2 plasma membrane localization is specified by the Cl domain  94  Figure 3.1 RasGRP2 plasma membrane localization is specified by the Cl domain A. Domain structures of GFP-tagged RasGRP2 proteins used in this experiment. GEF, guanine nucleotide exchange domain that catalyzes GTP loading of Rapi GTPase. EF, EF-hands. PB, PDZ binding motif is the Golgi targeting regulatory domain identified in this study. B. Subconfluent NIH 3T3 cells, expressing GFP alone or the GFP-tagged RG2 or RG2tC1 constructs, were fixed, stained with anti-GM 1 30 (secondary antibody conjugated to AF647) to mark Golgi membranes and photographed by fluorescence microscopy as described in Materials and Methods. Representative cells are shown. C. Confluent cultures of NIH 3T3 cells expressing GFP alone, or the GFP-tagged RG2 or RG2EiCI. Cells were fixed and photographed by fluorescence microscopy. Arrow indicates GFP-tagged RG2 localization at the plasma membrane at sites of cell-cell contact. The results shown are representative of 3 independent experiments. Percentages to the right of images are numbers of cells that show plasma membrane localization in one experiment (a total of at least 100 cells were counted), which also correlates with expression level, in that bright cells typically display plasma membrane localization whereas dim cells typically do not. D. Domain structures of GFP-tagged RG2CI domain and RG2C1x2 constructs used in this experiment. E. Localization of the GFP-tagged isolated Cl (RG2C1) and tandem Cl domain (RG2C1x2) of RasGRP2 in NIH 3T3 cells. Subconfluent cells were fixed and photographed by fluorescence microscopy. Arrow indicates GFP-tagged RG2C1x2 localization at the plasma membrane. Representative cells are shown.  95  A  GFP  GM13O  Merge  —10 pm  B  GFP  ER Tracker  Merge  10 pm  Figure 3.2 RasGRP1 localization at the Golgi and ER compared to RasGRP2 in NIH 3T3 cells Subconfluent NH 3T3 cells, expressing GFP-tagged RasGRP1 or GFP-tagged RasGRP2, were fixed, stained with anti-GM 130 (secondary antibody conjugated to AF647) to mark Golgi membranes (A) or ER Tracker to mark endoplasmic reticulum (B), and photographed by fluorescence microscopy as described in Materials and Methods. Representative cells are shown.  96  To determine if the Cl domain of RasGRP2 is capable of directly but weakly mediating plasma membrane localization, we generated two tandem Cl domains (RG2C1x2, Figure 3.1D) to increase its potential avidity for membranes. The tandem pair of RasGRP2 Cl domains was detectable at the plasma membrane at areas of cell-cell contact (Figure 3.1 E) demonstrating that the Cl domain of RasGRP2 can directly and independently mediates plasma membrane localization, although in a monomeric form it does this with insufficient affinity to confer stable membrane binding. In combination, these results indicate that the Cl domain of RasGRP2 specifies its localization to the plasma membrane, but must cooperate with other domains in RasGRP2 to attain detectable binding at this membrane compartment.  RasGRP1 also localizes at the Golgi (Figure 3.2A), however RasGRP1 localization at this site is totally dependent on the Cl domain (Beaulieu et al., 2007; Tognon et al., 1998). RasGRP 1 localization is also different to that of RasGRP2 in that RasGRP 1 localizes more distinctly at the endoplasmic reticulum (ER) (Figure 3.2B), a pattern of localization which also requires the Cl domain (Beaulieu et al., 2007; Tognon et al., 1998). In contrast RasGRP2 appears to be more dispersed through the cytoplasm rather than ER-localized in comparison to RasGRP1 (Figure 3.2B). These results show that these two RasGRPs have similar Golgi and divergent ER patterns of endomembrane localization, but from previous studies we know that RasGRP1 has a different mode of localization compared to RasGRP2; the Cl domain of RasGRP 1 targets the Golgi/ER while the our results show that the Cl domain of RasGRP2 targets the plasma membrane.  3.3.2  A potential binding site for PDZ proteins specifies Golgi localization of RasGRP2 The C-terminus of RasGRP2 has the sequence features of a class II PDZ-binding  motif. To bind cognate PDZ proteins, a class II PDZ-binding motif needs to be located at the C-terminus of a protein and must have hydrophobic residues at the C-terminal and -2 positions (Nourry et al., 2003), as is the case for the PDZ-binding motif of RasGRP2 (Figure 3.3A). By binding to PDZ proteins at the plasma membrane, the PDZ-binding motif of 97  RasGRP2 could cooperate with the Cl domain to confer stable plasma membrane localization. To test this hypothesis, we made one mutant in which we mutated the PDZ binding motif by attaching an additional two amino acids, thus moving the motif away from its required position at the C-terminus and we changed the hydrophobic leucine formerly at position 0 to an alanine. This mutant (RG2-PB i, Figure 3.3B) had notably reduced Golgi 1 localization (Figure 3.3C), however still localized to the plasma membrane in confluent NIH 3T3 cells (Figure 3.3D). The RG2-PBp. mutant was found in the nucleus as well as the cytoplasm, which could reflect dispersal of RasGRP2 once Golgi tethering via the PDZ binding motif is lost. Therefore the PDZ-binding motif at the C-terminus of RasGRP2 is not what cooperates with the Cl domain to confer plasma membrane localization, but instead has the distinct task of localizing a portion of RasGRP2 to the Golgi.  To confirm that the C-terminal region of RasGRP2 specified localization at the Golgi, we expressed a form of RasGRP2 that included only the part of RasGRP2 C-terminal to the Cl domain. This form of RasGRP2 (RG2-C-term) localized to the Golgi (Figure 3.3C), but did not localize to the plasma membrane (Figure 3 .3D). A form of RasGRP2 that contained both this C-terminal region and the Cl domain (RG2-C1+C-term), localized both at the Golgi (Figure 3.3 C) and at the plasma membrane (Figure 3.3D), indicating that the Cl domain and the C-terminus were solely responsible for the Golgi and plasma membrane pattern of RasGRP2 localization. Since the Cl domain of RasGRP2 in isolation is unable to localize to the plasma membrane, unless it is expressed as a tandem pair of Cl domains (Figure 3.1 D), we hypothesize that a region in between the Cl domain and the PDZ-binding motif may be responsible for stabilizing RasGRP2 Cl-mediated plasma membrane localization. Also because the C-terminal region on its own is incapable of localizing at the plasma membrane, this indicates that the C-terminal region is interdependent with the Cl domain in terms of providing plasma membrane localization.  98  3.3.3  Both the Cl domain and the PDZ-binding motif contribute to RasGRP2 activity as an exchange factor for Rapi Expression of RasGRP2 in NIH 3T3 fibroblasts led to a 2 to 4 fold increase in levels  of GTP bound Rap 1 even when RasGRP2 expression levels were low (Figure 3 .4A). This demonstrates that RasGRP2 is constitutively active in these cells as a Rap-specific exchange factor. To determine whether membrane localization was a requirement for RasGRP2 activity we analysed Rap 1-GTP levels in NIH 3T3 cells expressing mutated forms of RasGRP2, deficient in either plasma membrane or Golgi targeting. Cells expressing RG2AC1, which has no plasma membrane localization, had reduced Rapl-GTP levels compared to cells expressing RasGRP2, but levels were still significantly greater than those in control cells (Figure 3.4A, 3.4B). Mutation of the PDZ-binding motif, which abrogates Golgi, but not plasma membrane, localization, also reduced but did not eliminate RasGRP2mediated GTP loading of Rapi (Figure 3.4A, 3.4B). Therefore while the Cl and the PDZ binding motif specify RasGRP2 localization to distinctly different sites in NIH 3T3 fibroblasts, they both contribute to RasGRP2 activity within these cells. 3.3.4  Changes in cell morphology and colony edge phenotype induced by RasGRP2  require plasma membrane localization via the Cl domain Expression of RasGRP2 in NIH 3T3 fibroblasts caused a distinct change in cell shape (Figure 3.5A). When approaching confluence, NIH 3T3 fibroblasts have a disordered morphology, with cell projections growing on top of other cells, and with irregular spaces in between cells. In comparison to control cells, confluent RasGRP2 expressing cells were more regular in shape and grew as tessellated (cobblestone-like) monolayers with a reduction in the numbers of spaces in between cells and without overgrowth of neighbouring cell projections. In addition, colonies of RasGRP2 expressing cells had very tight discrete boundaries, lacking dispersed single cells at the edge as seen in colonies of control cells (Figure 3.5A, 3.5C). This could be due to reduced cell migration, a phenotype that is induced in MDCK epithelial cells by expression of the Rap-specific GEF Epacl (Lyle et al., 2008). Using a scratch test assay we found that RasGRP2-expressing NIH 3T3s cells could move into the cleared area, indicating that they did not have a complete migration defect  99  (Figure 3.5 B). However, the front of migrating cells was very tight, lacking single cells migrating beyond the advancing boundary of cells (Figure 3.5B).  Deletion of the Cl domain eliminated the ability of RasGRP2 to induce changes in cell and colony morphology, such that cells expressing RG2AC1 had the disordered monolayers and dispersed colony boundaries typical of control NIH 3T3 cells (Figure 3 .5A, 3.5C). In contrast, mutation of the PDZ-binding motif had no effect on the ability of RasGRP2 to induce morphological changes in NIH 3T3 cells (Figure 3.5A, 3.5C). This indicated that the morphological changes are induced by RasGRP2 when it is localized to the plasma membrane via the Cl domain, but not when it is localized to Golgi by the PDZ binding motif. To further test how localization determines cellular responses to RasGRP2, we constructed variants of RasGRP2 with mechanistically distinct modes of targeting to either the plasma membrane or Golgi membranes. RasGRP2 was targeted to the plasma membrane by replacing its Cl domain and C-terminus with the basic cluster and prenylation signal of K-Ras4B (RG2-pr, Figure 3.6A). For Golgi membrane targeting, we replaced the Cl and C-terminus of RasGRP2 with the Cl domain ofRasGRPl (RG2-1C1, Figure 3.6A). In contrast to the selectivity of the RasGRP2 Cl domain for the plasma membrane, the Cl domain of RasGRP 1 is selectively localized at the Golgi of NIH 3T3 cells (Johnson et al., 2007) apparently due to its recognition of high concentrations of DAG at this site. As intended, RG2-pr localized predominantly to the plasma membrane while RG2-1C1 localized predominantly to the Golgi (Figure 3.6B). Expression of either RG2-pr or RG2-1C1 increased GTP loading of Rapi relative to control cells (Figure 3.6C, 3.6D). However, only the cells expressing the plasma membrane-localizing RG2-pr construct showed the tesselated monolayers and tight colony boundaries that typify RasGRP2-expressing NIH 3T3 cells (Figure 3.6E, 3.6F). Therefore, it is the plasma membrane-localized form of RasGRP2 that is capable of modifying the morphology of NIH 3T3 cells.  100  A  Class II binding motif PXkPcoo RG2 I H L-coo RG2-PB I HAS G-coo  B GFP  GEF  EF  Cl  PB RG2 RG2-PBp  OZD1  RG2-C1 -‘-C-term RG2-C-term  C  GFP  GM13O  Merge  D  Confluent culture  GFP  RG2  RG2  RG2-PBiJ  RG2-PBp  19% pm  RG2C-term  RG2Cl +C-term  RG2Cl +C-term lOpm  —10pm  Figure 3.3 RasGRP2 localization at the Golgi requires a PDZ-binding motif 101  Figure 3.3 RasGRP2 localization at the Golgi requires a PDZ-binding motif A. RasGRP2 C-terminus has sequence features of a class II PDZ-binding motif with hydrophobic (‘1’) amino acids at position 0 (leucine) and at position -2 (isoleucine). Disruption of this motif was facilitated by attaching two amino acids, thus moving the motif away from the C-terminus, and by changing the leucine formerly at position 0, to an alanine. B. Domain structures of GFP-tagged RasGRP2 proteins used in this experiment. C. Subconfluent NIH 3T3 cells, expressing GFP alone, or the GFP-tagged RG2, RG2-PBp, RG2-Cterm or RG2-C1+C-term constructs, were fixed, stained with anti-GM13O to mark Golgi membranes and photographed by fluorescence microscopy as described in Materials and Methods. Representative cells are shown. D. Confluent cultures of NIH 3T3 cells expressing the GFP alone, or the GFP-tagged RG2, RG2-PBp, RG2-C-term or RG2-C1 +C-term constructs. Cells were fixed and photographed by fluorescence microscopy. Arrow indicates GFP-tagged RG2 and RG2-PBp localization at the plasma membrane. The results shown are representative of 3 independent experiments. Percentages to the right of images are numbers of cells that show plasma membrane localization in one experiment (a total of at least 100 cells were counted). Percentages may be an underestimate because bright cells typically display plasma membrane localization whereas dim cells typically do not.  102  A high med low GFP RG2 RG2 RG2 RG2 RG2 1C1 -PBp 100 kDa 80 kDa 20 kDa  I < transduced RG2  ——  —1  J  <Rapl -GTP  —.  I<total Rapi  20 kDa—i 1  2.5 4.2  3.2  1.5  1.8  B p=0.04  p=0.01 4  4  3  3  2  2  0  0 I 0  0  0 GFP RG2  RG2  GFP RG2  ci  RG2 -PBp  Figure 3.4 The Cl domain and the potential PDZ-binding motif are required for full activation of RasGRP2 A. NIH 3T3 cells, expressing GFP alone or GFP-tagged RG2, RG2tiCI or RG2-PBp constructs, were lysed and assayed for levels of GTP bound Rapi by Raf-RBD chromatography and detected by Western blot with anti-Rapi. Levels of Rapl-GTP and total Rapi were quantified by band volume analysis using Quantity One software (Bio-Rad Laboratories). The numbers below the Rapi Total blot are the ratios of Rapl-GTP over total Rapi in RG2, RG2LC1 and RG2-PBp expressing cells, relative to Rapl-GTP/total Rapi in cells expressing GFP alone. The expression levels of transduced GFP-tagged RG2 protein in each sample determined by Western blot with an anti-GFP antibody, are shown in the upper blot. B. Rapi activation by RG2, RG2tC1 and RG2-PBp in serum grown cells, relative to cells expressing GFP alone. Rapl-GTP/total Rapi levels were measured in NIH 3T3 cells expressing RG2, RG2C1 and RG2-PBp and are displayed relative to GFP alone. Cells expressing GFP tagged-mutant RasGRP2 were compared with cells expressing a similar level of GFP-tagged wild type RasGRP2. Bars show SEM. The two-tailed p values are for comparison to a hypothetical mean of 1, or in the case of RG2 compared to mutant, the hypothetical mean was that of RG2. Results are from seven separate experiments with RG2ACI and five separate experiments with RG2-PBp. 103  A  Cell morphology  Colony boundary  B  0  GFP  RG2  RG2C1  RG2-PBp  p=0.01  C p=0.000l  II  1O0  0 c’) C)  —  50  p0.0O0l 100  ri  50  -  C— ci) .C  .2’  ‘D  o  -J 0  0 GFP RG2 RG2 C1  GFP RG2 RG2PBp  Figure 3.5 The Cl domain is required for RasGRP2 induced changes in morphology  104  Figure 3.5 The Cl domain is required for RasGRP2 induced changes in morphology (page 99) A. NIH 3T3 cells expressing GFP alone, GFP-tagged RG2, RG2AC1 or RG2-PBp were seeded at low density, grown into colonies and photographed as described in Materials and Methods to show cell morphology (left panels) and colony boundary (middle panels). GFP alone, GFP-tagged RG2, RG2C1 or RG2-PBp expressing cells were seeded on glass coverslips, and when confluent were fixed, stained with Phalloidin TRITC and photographed (right panels) as described in Materials and Methods. B. NIH 3T3 cells expressing either GFP (left panels) or GFP-tagged RG2, (right panels) showing cell migration, were seeded and photographed 16 hr post scratch test as described in Materials and Methods. C. Percentage of colonies with RG2-induced colony boundary morphology phenotype. A total of at least 70 colonies from six separate (GFP, RG2, RG2iC1) or four separate (GFP, RG2, RG2-PBp) experiments were counted. Bars show SEM. The two-tailed p values are for comparison to hypothetical mean of 0.  105  B  A GFP  GEF  EF  Cl  PB  c—cJ—II—cD’-—4 cj-.I  RG2-pr RG2-1 Cl —i0im  C  D GFP RG2 RG2 -pr  100 kDa—1 8OkDai 20 kDa  GFP  RG2 RG2 -id  20 <ansduced  —  <Rapl-GTP  —j_____________  p=0.005  I  p=0.02  I  p=0.0i p=0.0i  15 10  <total Rapi  20 kDa  5 9.3  6.5  16.3 0  E  Cell morphology  GFP RG2 RG2 -pr  F  Colony boundary  GFP RG2 RG2 -1 Cl  p=0.000l  RG2 -pr  >  100  —  p=0.001 p=0.0001 i:o E  -o 2 00 (0  50 C—  a) C) 0) a)—  RG2 -1C1  0 —  0 GFP RG2 RG2 -pr  100 im  GFP RG2 RG21CI  Figure 3.6 Targeting RasGRP2 to the plasma membrane induces colony morphology  106  Figure 3.6 Targeting RasGRP2 to the plasma membrane induces colony morphology phenotype (page 101) A. Domain structures of GFP-tagged RasGRP2 proteins used in this experiment. To a truncated form of RasGRP2 the plasma membrane targeting K-Ras basic cluster and C-terminal prenylation signal (RG2-pr), or the Cl domain of RasGRPI (RG2-ICI) were attached. B. NIH 3T3 cells expressing RG2-pr or RG2-1C1 were grown on cover slips, fixed and photographed as described in Materials and Methods. C. NIH 3T3 cells expressing GFP alone or GFP-tagged RG2, RG2-pr or RG2-1CI were Iysed and assayed for levels of GTP-bound Rapi by Raf-RBD chromatography and detected by Western blot with anti-Rapi. Levels of Rapl-GTP and total Rapi were quantified by band volume analysis using Quantity One software (Bio-Rad). The numbers below the total Rapi blot are the ratios of Rapi GTP over total RapI in RG2, RG2-pr and RG2-ICI expressing cells, relative to Rapl-GTP/total Rapi in cells expressing GFP alone. The expression levels of transduced RasGRP2 protein in each sample, determined by Western blot with an anti-GFP antibody, are shown in the upper blot. D. RapI activation by RG2, RG2-pr and RG2-1CI in serum grown cells, relative to cells expressing GFP alone. The numbers below the total Rapi blot are the ratios of Rapl-GTP over total Rapi in RG2, RG2-pr and RG2-IC1 expressing cells, relative to Rapl-GTP/total Rapi in cells expressing GFP alone. The expression levels of transduced GFP-tagged RG2 protein in each sample determined by Western blot with an anti-GFP antibody, are shown in the Figure 3.5C. Cells expressing GFP tagged-mutant RasGRP2 were compared with cells expressing a similar level of GFP- tagged wild type RasGRP2. Rapl-GTP/total Rapi levels were measured in NIH 3T3 cells expressing RG2, RG2-pr and RG2-1 Cl and are displayed relative to GFP alone. Bars show SEM. The two-tailed p values are for comparison to hypothetical mean of 1, or in the case of RG2 compared to mutant, the hypothetical mean was that of RG2. Results are from nine separate experiments with RG2-pr and four separate with RG2-1 Cl. E. NIH 3T3 cells expressing RG2-pr (upper panels) or RG2-1C1 (lower panels) grown to confluence and photographed by fluorescence microscopy (left panels) or seeded at low density, grown into colonies and photographed by light microsope (right panels) as described in Materials and Methods. F. Percentage of colonies with RG2 induced colony boundary morphology phenotype. A total of at least 70 colonies from four separate (GFP, RG2, RG2-pr) or three separate (GFP, RG2, RG2-ICI) experiments were counted. Bars show SEM. The two-tailed p values are for comparison to hypothetical mean of 0.  107  3.3.5  The Cl domain and PDZ-binding motif specify plasma membrane versus Golgi localization of RasGRP2 in Jurkat T-cells The experiments with NIH 3T3 cells showed that RasGRP2 localization at the plasma  membrane is determined by the Cl domain, while Golgi localization is regulated by the PDZ-binding motif. RasGRP2 is expressed in primary human T-cells (Pasvolsky et al, 2007; Ghandour et a!, 2007), where it is thought to play a role in integrin-mediated T-cell adhesion (Pasvolsky et a!, 2007; Ghandour et al, 2007). RasGRP2 is also expressed in the Jurkat T cell line, which has has been used as an experimental model for studying the role of RasGRP2 in Rapl activation (Katagiri et al., 2004). Jurkat T-cells were used to address the question of whether the Cl and PDZ-binding motif would determine RasGRP2 localization in a cell line which normally expresses RasGRP2. In Jurkat cells, we found that GFP-tagged RasGRP2 was localized predominantly in the cytoplasm, with some localization at the plasma membrane, which was most readily detected at regions of cell-cell contact (Figure 3.7A), and was also found to be localized at the Golgi (Figure 3.8). RG2AC1 could not be detected at the plasma membrane, not even at areas of cell-cell contact, while localization in the cytoplasm (Figure 3.7A) and at the Golgi was unaffected by deletion of the Cl domain (Figure 3.8). The 1 RG2-PB i , mutant was detected at the plasma membrane at regions of cellcell contact (Figure 3.7A), but Golgi localization was reduced to that seen for GFP alone (Figure 3.8). RG2-Cl+C-term localized at both the plasma membrane, at sites of cell-cell contact, in the cytoplasm and at the Golgi, whereas RG2-C-term did not localize at the plasma membrane, but was in the cytoplasm and at the Golgi (Figure 3.7B, 3.8). These results indicate that in Jurkat cells, as in NIH 3T3 fibroblasts, the Cl domain of RasGRP2 specifies plasma membrane localization whereas the PDZ-binding motif specifies Golgi localization.  We also looked at the localization pattern of GFP-tagged RasGRP2 localization in the DO 11.10 murine T-cell line and found that GFP-tagged RG2 and RG2AC 1, RG-PBp, RG2C1+C-term and RG2-C-term localization was identical to that observed in Jurkat cells (Figure 3.7C, D).  108  A  GFP  C  inset  GFP  inset  GFP  0% pm  GFP  0% pm  RG2  47% pm  RG2  58% pm  RG2C1  0% pm  RG2  RG2C1  RG2  65% pm  RG2-C1+ C-term  64% pm  82% pm  10iim  —  0% pm  5pm  B  D  RG2-C1 + C-term  68% pm  0% pm  RG2C-term —  0% pm  RG2C-term  10pm  10pm  Figure 3.7 The Cl domain specifies plasma membrane localization of RasGRP2 in Jurkat and DO11.1O T-cells Jurkat (A, B) and DCI 1.10 (C, D) T-cells expressing GFP alone or the GFP-tagged RG2, RG2C1,RG2-PBp, RG2-C1+C-term and RG2-C-term constructs were fixed, and photographed as described in Materials and Methods. Representative cells are shown. Percentages to the right of images are numbers of cells that show plasma membrane localization in one experiment (a total of at least 50 cells that demonstrated cell-cell contact were counted).  109  GFP  RG2  RG2C1  RG2 -PBp  RG2-C1 + C-term  RG2C-term  lOpm  Figure 3.8 The PDZ-binding motif specifies RasGRP2 localization at the Golgi in Jurkat T-cells JurkatT-cells expressing GFP alone or the GFP-tagged RG2, RG2C1, RG2-PBp, RG2-C1+C-term and RG2-C-term constructs were fixed, stained with anti-GM13O to mark Golgi membranes and photographed as described in Materials and Methods. Representative cells are shown.  110  3.3.6  RasGRP2 does not respond to antigen receptor stimulation RasGRP2 has been previously been shown to be activated by T-cell receptor (TCR)  stimulation in Jurkat cells (Katagiri et al, 2004). Since RasGRP2 is able to localize at the plasma membrane in Jurkat cells, and the plasma membrane is the main site of activation of Rapi in T-cells (Bivona et al, 2004), we expected that RasGRP2 might translocate to the plasma membrane in response to TCR stimulation. However TCR ligation via anti-CD3 in combination with co-stimulation via anti-CD28, did not cause any detectable redistribution of RasGRP2 (Figure 3.9A). Treatment with anti-CD3 and anti-CD28 increases Rapl-GTP levels in Jurkat cells (Figure 3 .9B), but this was not increased by transfection of the cells with RasGRP2 (Figure 3.9B). While this may reflect saturation of Rapi activation via endogenous RasGRP2, these experiments indicate that RasGRP2 is not activated by translocation in response to TCR stimulation.  RasGRP1 translocates to the plasma membrane in response to B-cell receptor stimulation of DT4O B-cells (Beaulieu et al, 2007). Rapi is also activated in these cells in reponse to BCR stimulation (McLeod et al, 2002). In contrast to RasGRP1, we found that RasGRP2 did not translocate to the plasma membrane post BCR stimulation of DT4O cells, and we did not see an increase in Rapi activation in RasGRP2-expressing cells compared to non-transduced cells under the same conditions. These results indicate that RasGRP2 also does not translocate to the plasma membrane in response to BCR stimulation.  3.3.7  The Cl domain is required for concentration of RasGRP2 at the plasma membrane in response to the chemokine SDF-lcL Treatment of Jurkat cells with the chemokine SDF-lcL leads to a rapid increase in  Rapi activation, which is reduced when the level of endogenous RasGRP2 is lowered by RNA silencing (Ghandour et a!., 2007). SDF-la stimulation induced rapid and transient accumulation of RasGRP2 at the plasma membrane with concurrent reductions in cytosolic and Golgi-localized RasGRP2 (Figure 3.1OA). However, we did not observe an increase in Rapi activation in RasGRP2-expressing cells compared to non-transduced cells after 111  stimulation with SDF- 1 a. (Figure 3.1 OB). This could either be the result of a lack of catalytic activation of RasGRP2 despite translocation, or alternatively due to saturation of Rap 1 activation in response to SDF-lct by endogenous levels of RasGRP2.  Given that the Cl domain is required for plasma membrane localization in unstimulated cells, we postulated that this domain would also be required for SDF-lct induced recruitment of RasGRP2 to the plasma membrane. As predicted, RG2AC1 did not accumulate at the plasma membrane in SDF- 1 a. treated cells (Figure 3.1 OA). These results further demonstrate the essential role of the Cl domain in regulating RasGRP2 localization to the plasma membrane.  112  A  RG2  10pm  —  B nil -  100 kDa 80 kDa  3’  10’  —] H  1’  -  3’  10’  -‘-----j  anti-CD3/CD28 transduced <RG2 Rapi GTP  20 kDa 20 kDa  1’  RG2  -  ‘H<  _]  total Rapi  C  D nil  -  1’  RG2 3’  15’  -  100 kDa—i 80 kDa—j  I  2OkDa—]  3’  15’  anti-lgM transduced <RG2 <Rapi GTP  ——  20 kDa —1  1’  <total Rapi  •  Figure 3.9 RasGRP2 does not respond to TCR or BCR stimulation  113  Figure 3.9 RasGRP2 does not respond to TCR or BCR stimulation A. Jurkat T-cells expressing GFP-tagged RG2 were starved for 4 hr in serum free media, stimulated with 10 pg/mI anti-CD3 and anti-CD28 for the I or 10 minutes, fixed photographed by fluorescence microscopy as described in Materials and Methods. Representative cells are shown. B. Jurkat T-cells expressing GFP alone or GFP-tagged RG2 were starved for 4 hr in serum free media, stimulated with 10 pg/mI anti-CD3 and anti-CD28 for 1, 3 or 10 minutes, and then were lysed and assayed for levels of GTP-bound Rapi by Raf-RBD chromatography and detected by Western blot with anti-Rapi. The expression levels of transduced RasGRP2 protein in each sample, determined by Western blot with an anti-GFP antibody, are shown in the upper blot. Results are representative of three experiments. C. DT4O B-cells expressing GFP-tagged RG1 or the GFP-tagged RG2 were starved for 4 hr in serum free media, stimulated with 5 pg/mI anti-lgM for 15 minutes, fixed photographed by fluorescence microscopy as described in Materials and Methods. Representative cells are shown. 0. DT4O B-cells expressing GFP alone or GFP-tagged RG2 were starved for 4 hr in serum free media, stimulated with 5 pg/mI anti-lgM for 1, 3 or 15 minutes, and then were lysed and assayed for levels of GTP-bound Rapi by Raf-RBD chromatography and detected by Western blot with anti Rapi. The expression levels of transduced RasGRP2 protein in each sample, determined by Western blot with an anti-GFP antibody, are shown in the upper blot. Results are representative of three experiments.  114  A  Nil  SDF-la 20”  SDF-la 50”  RG2  RG2  c1  —  lOum  B  nil -  100 kDa  RG2  10” 20” 50”  10” 20” 50” SDF-la _<transduced -  [ERaPI GTP j.Etotal Rapi  Figure 3.10 RasGRP2 accumulates at the plasma membrane in response to SDF-1o treatment of Jurkat T-cells A. Jurkat T-cells expressing GFP-tagged RG2 or RG2C1 constructs were starved for 8 hr in serum free media, stimulated with 10 ng/ml SDF-la for 20 or 50 seconds, fixed photographed by fluorescence microscopy as described in Materials and Methods. Representative fields of cells are shown. B. Jurkat T-cells expressing GFP alone or GFP-tagged RG2 were starved for 4 hr in serum free media, stimulated with 10 ng/ml SDF-1 a for the 10, 20 or 50 seconds, and then were lysed and assayed for levels of GTP-bound Rapi by Raf-RBD chromatography and detected by Western blot with anti-Rapi. The expression levels of transduced RasGRP2 protein in each sample, determined by Western blot with an anti-GFP antibody, are shown in the upper blot. Results are representative of two experiments. 115  3.4  DISCUSSION Our experiments show that the Cl domain of RasGRP2 is required for full efficiency  of Rapi activation in NIH 3T3 cells, and is also essential for localization of RasGRP2 at the plasma membrane in both NIH 3T3 fibroblasts and Jurkat T-cells. Previous studies have demonstrated that the RasGRP2 Cl domain does not have measurable DAG binding affinity compared to the RasGRP1, 3 and 4ct Cl domains (Irie et al., 2004; Johnson et al., 2007). Furthermore, in NIH 3T3 cells the localization of the RasGRP2 Cl domain is distinct from the localization of DAG-binding Cl domains (Johnson et al., 2007). Therefore, DAG production at the plasma membrane cannot be what drives Cl-dependent localization of RasGRP2 to the plasma membrane. Independent of DAG binding, all Cl domains in the RasGRP family have substantial affinity for vesicles enriched in anionic phospholipids (Johnson et a!., 2007). Therefore it is possible that the Cl domain of RasGRP2 binds to membranes by this mechanism, with its selectivity for the plasma membrane being due to the high concentration of anionic phosphatidylserine at that membrane (Yeung et a!., 2008). In contrast, the localization of DAG-binding Cl domains such as those of RasGRP 1, 3 and 4c Cl will result in their selective targeting to Golgi membranes which are enriched in DAG (Johnson et al., 2007).  The experiment with isolated monomeric versus tandem Cl domains indicated that the Cl domain of RasGRP2 can directly bind the plasma membrane in NIH 3T3 cells, but does so with an affinity that is too low to support stable localization of RasGRP2 at the plasma membrane. Other parts of RasGRP2 must cooperate with the Cl domain to enable effective plasma membrane binding, either by modifying the affinity of the Cl domain or by providing additional weak interaction sites. The latter is analogous to the situation with RasGRP 1, where stable plasma membrane binding requires cooperativity between the PT domain, which specifies plasma membrane targeting, and the Cl domain which provides an additional point of contact with the membrane (Beaulieu et a!., 2007). We showed that the Cl +C-terminus form of RasGRP2 is sufficient for localization at both the plasma membrane and Golgi, and that the C-terminus is sufficient for Golgi localization. Since the Cl domain is incapable of localizing at the plasma membrane, this indicates that a region in the C-terminus may stabilize Cl domain-mediated RasGRP2 localization. 116  While the Cl domain specifies plasma membrane localization of RasGRP2, a putative PDZ protein binding site at the C-terminus is required for efficient localization at the Golgi in NIH 313 fibroblasts and Jurkat T-cells. The C-terminus could act as a binding site for a Golgi-localized PDZ protein. However, the class II PDZ-binding motif has low complexity, consisting only of two hydrophobic residues, one at position 0 and the other at position -2. Therefore, its presence in a protein cannot be taken as strong evidence that it acts as a PDZ binding site. It is possible that the C-terminus of RasGRP2 confers Golgi localization by a different mechanism, which is disrupted by the RG2-PBp. mutation.  We showed that RasGRP2 accumulates at the plasma membrane as a rapid response to the chemokine SDF-la in Jurkat cells, and this is completely dependent on the Cl domain. While this relocalization of RasGRP2 could reflect transient increases in anionic lipids at the plasma membrane, it might also involve the induction of an as yet unidentified ligand for the Cl domain. If so, this ligand is not induced by TCR or BCR ligation, as we found that stimulation of Jurkat T-cells with anti-CD3 plus anti-CD28, or stimulation of DT4O B-cells with anti-IgM, has no effect on RasGRP2 localization.  RasGRP2 expression in NIH 3T3 fibroblasts leads to changes in cell shape and colony structure typified by the acquisition of a tessellated monolayer morphology and tight colony edges. In cells expressing RG2AC 1, Rap 1 activation is reduced compared to wild type RasGRP2, plasma membrane localization is abolished, and morphological changes are not induced. While the RG2-PB i mutant has loss of Golgi localization and reduced Rapi 1 activation, this mutant still induces the same morphological changes as wild type RasGRP2. Artificial targeting of RasGRP2 to the plasma membrane via the K-Ras membrane anchor induced Rapi activation and morphological changes, while artificial targeting of RasGRP2 to the Golgi via the Cl domain of RasGRP 1 failed to induce morphological changes despite inducing activation of Rap 1. Therefore, RasGRP2 can activate Rap 1 when it is localized either to the plasma membrane or Golgi, but only plasma membrane-localized RasGRP2 is capable of inducing morphological changes in NIH 313 cells. Rapi GTPase activation at different membranes has the potential to give rise to different signalling outputs, and thus 117  different cellular responses (Mor and Philips, 2006). This is known to be the case for Rapi activation via the GEF Epaci in PC12 neuronal cells. Epaci normally localizes and activates Rap 1 in the perinuclear region in these cells, but activation of Rap 1 at this location does not activate ERK (Wang et a!., 2006). When targeted to the plasma membrane via attachment of the K-Ras polybasic and prenylation signal sequence, Epac 1 activation of Rap 1 leads to ERK activation (Wang et al., 2006). By selective utilization of its Cl domain versus its C terminus containing the PDZ-binding motif RasGRP2 also has the potential to modulate where Rapi is activated in cells, and thus how cells respond to Rapl activation.  118  3.5  BIBLIOGRAPHY  Beaulieu, N., Zahedi, B., Goulding, R.E., Tazmini, G., Anthony, K.V., Omeis, S.L., de Jong, D.R. and Kay, R.J. (2007) Regulation of RasGRP 1 by B cell antigen receptor requires cooperativity between three domains controlling translocation to the plasma membrane. Mol Biol Cell, 18, 3156-3168. Bernardi, B., Guidetti, G.F., Campus, F., Crittenden, J.R., Graybiel, A.M., Balduini, C. and Torti, M. (2006) The small GTPase Raplb regulates the cross talk between platelet integrin alpha2betal and integrin alphallbbeta3. Blood 107, 2728-273 5. Bivona, T.G., Perez De Castro, I., Ahearn, I.M., Grana, T.M., Chiu, V.K., Lockyer, P.J., Cullen, P.J., Pellicer, A., Cox, A.D. and Philips, M.R. (2003) Phospholipase Cgamma activates Ras on the Golgi apparatus by means ofRasGRPl. Nature, 424, 694-698. Caloca, M.J., Zugaza, J.L. and Bustelo, X.R. (2003) Exchange factors of the RasGRP family mediate Ras activation in the Golgi. JBiol Chem, 278, 33465-33473. Caloca, M.J., Zugaza, J.L., Vicente-Manzanares, M., Sanchez-Madrid, F. and Bustelo, X.R. (2004) F-actin dependent translocation of the Rapl GDP/GTP exchange factor RasGRP2. JBiol Chem, 279, 2043520446. Carrasco, S. and Merida, I. (2006) Diacylglycerol, when simplicity becomes complex. Trends Biochem Sci. Clyde-Smith, J., Silins, G., Gartside, M., Grimmond, S., Etheridge, M., Apolloni, A., Hayward, N. and Hancock, J.F. (2000) Characterization of RasGRP2, a plasma membrane-targeted, dual specificity Ras/Rap exchange factor. JBiol Chem, 275, 32260-32267. Colon-Gonzalez, F. and Kazanietz, M.G. (2006) Cl domains exposed: from diacylglycerol binding to proteinprotein interactions. Biochim Biophys Acta, 1761, 827-837. Crittenden, J.R., Bergmeier, W., Zhang, Y., Piffath, C.L., Liang, Y., Wagner, D.D., Housman, D.E. and Graybiel, A.M. (2004) Ca1DAG-GEFI integrates signaling for platelet aggregation and thrombus formation. Nat Med 10, 982-986. Dupuy, A.J., Morgan, K., von Lintig, F.C., Shen, H., Acar, H., Hasz, D.E., Jenkins, N.A., Copeland, N.G., Boss, G.R. and Largaespada, D.A. (2001) Activation of the Rap 1 guanine nucleotide exchange gene, Ca1DAG-GEF I, in BXH-2 murine myeloid leukemia. J Biol Chem, 276, 11804-11811. Ebinu, J.O., Bottorff, D.A., Chan, E.Y., Stang, S.L., Dunn, R.J. and Stone, J.C. (1998) RasGRP, a Ras guanyl nucleotide- releasing protein with calcium- and diacylglycerol-binding motifs. Science, 280, 10821086. Eto, K., Murphy, R., Keffigan, S.W., Bertoni, A., Stuhlmann, H., Nakano, T., Leavitt, A.D. and Shattil, S.J. (2002) Megakaryocytes derived from embryonic stem cells implicate Ca1DAG-GEFI in integrin signaling. Proc NatlAcad Sci USA, 99, 12819-12824. Ghandour, H., Cullere, X., Alvarez, A., Luscinskas, F.W. and Mayadas, T.N. (2007) Essential role for Rapi GTPase and its guanine exchange factor Ca1DAG-GEFI in LFA- 1 but not VLA-4 integrin mediated human T-cell adhesion. Blood 110, 3682-3690. Irie, K., Masuda, A., Shindo, M., Nakagawa, Y. and Ohigashi, H. (2004) Tumor promoter binding of the protein kinase C Cl homology domain peptides of RasGRPs, chimaerins, and Uncl3s. Bioorg Med Chem, 12, 4575-4583. Johnson, J.E., Goulding, R.E., Ding, Z., Partovi, A., Anthony, K.V., Beaulieu, N., Tazmini, G., Cornell, R.B. and Kay, R.J. (2007) Differential membrane binding and diacylglycerol recognition by Cl domains of RasGRPs. Biochem J 406, 223-236. Katagiri, K., Shimonaka, M. and Kinashi, T. (2004) Rap 1-mediated lymphocyte function-associated antigen-i activation by the T cell antigen receptor is dependent on phospholipase C-gammai. JBiol Chem, 279, 11875-11881. Katsoulotos, G.P., Qi, M., Qi, J.C., Tanaka, K., Hughes, W.E., Molloy, T.J., Adachi, R., Stevens, R.L. and Krilis, S.A. (2008) The Diacylglycerol-dependent translocation of ras guanine nucleotide-releasing protein 4 inside a human mast cell line results in substantial phenotypic changes, including expression of interleukin 13 receptor alpha2. JBiol Chem, 283, 1610-1621. Kawasaki, H., Springett, G.M., Toki, S., Canales, J.J., Harlan, P., Blumenstiel, J.P., Chen, E.J., Bany, l.A., Mochizuki, N., Ashbacher, A., Matsuda, M., Housman, D.E. and Graybiel, A.M. (1998) A Rap  119  guanine nucleotide exchange factor enriched highly in the basal ganglia. Proc Nat! Acad Sci US A, 95, 13278-13283. Lorenzo, P.S., Beheshti, M., Pettit, G.R., Stone, J.C. and Blumberg, P.M. (2000) The guanine nucleotide exchange factor RasGRP is a high -affinity target for diacyiglycerol and phorbol esters. Mol Pharmacol, 57, 840-846. Lorenzo, P.S., Kung, J.W., Bottorff, D.A., Garfield, S.H., Stone, J.C. and Blumberg, P.M. (2001) Phorbol esters modulate the Ras exchange factor RasGRP3. Cancer Res, 61, 943-949. Lyle, K.S., Raaijmakers, J.H., Bruinsma, W., Bos, J.L. and de Rooij, J. (2008) cAMP-induced Epac-Rap activation inhibits epithelial cell migration by modulating focal adhesion and leading edge dynamics. Cell Signal, 20, 1104-1116. Madani, S., Hichami, A., Cherkaoui-Malki, M. and Khan, N.A. (2004) Diacyiglycerols containing Omega 3 and Omega 6 fatty acids bind to RasGRP and modulate MAP kinase activation. JBiol Chem, 279, 11761183. Milckers, H., Allen, J., Knipscheer, P., Romeijn, L., Hart, A., Vink, E., Berns, A. and Romeyn, L. (2002) Highthroughput retroviral tagging to identit’ components of specific signaling pathways in cancer. Nat Genet, 32, 153-159. Mor, A. and Philips, M.R. (2006) Compartmentalized Ras/MAPK signaling. Annu Rev Immunol, 24, 77 1-800. Nourry, C., Grant, S.G. and Borg, J.P. (2003) PDZ domain proteins: plug and play! Sci STKE, 2003, RE7. Oh-hora, M., Johmura, S., Hashimoto, A., Hikida, M. and Kurosaki, T. (2003) Requirement for Ras guanine nucleotide releasing protein 3 in coupling phospholipase C-gamma2 to Ras in B cell receptor signaling. JExp Mec4 198, 1841-1851. Ohba, Y., Mochizuki, N., Matsuo, K., Yamashita, S., Nakaya, M., Hashimoto, Y., Hamaguchi, M., Kurata, T., Nagashima, K. and Matsuda, M. (2000b) Rap2 as a slowly responding molecular switch in the Rap 1 signaling cascade. Mo! Cell Biol, 20, 6074-6083. Ohba, Y., Mochizuki, N., Yamashita, S., Chan, AM., Schrader, J.W., Hattori, S., Nagashima, K. and Matsuda, M. (2000a) Regulatory proteins of R-Ras, TC2 1/R-Ras2, and M-RasIR-Ras3. JBiol Chem, 275, 20020-20026. Pear, W.S., Nolan, G.P., Scott, M.L. and Baltimore, D. (1993) Production of high-titer helper-free retroviruses by transient transfection. Proc NatlAcadSci USA, 90, 8392-8396. Perez de Castro, I., Bivona, T.G., Philips, M.R. and Pellicer, A. (2004) Ras activation in Jurkat T cells following low-grade stimulation of the T-cell receptor is specific to N-Ras and occurs only on the Golgi apparatus. Mo! Cell Biol, 24, 3485-3496. Priatel, J.J., Teh, S.J., Dower, N.A., Stone, J.C. and Teh, H.S. (2002) RasGRP1 transduces low-grade TCR signals which are critical for T cell development, homeostasis, and differentiation. Immunity, 17, 617627. Rambaratsingh, R.A., Stone, J.C., Blumberg, P.M. and Lorenzo, P.S. (2003) RasGRP1 represents a novel non protein kinase C phorbol ester signaling pathway in mouse epidermal keratinocytes. JBiol Chem, 278, 52792-52801. Reuther, G.W., Lambert, Q.T., Rebhun, J.F., Caligiuri, M.A., Quilliam, L.A. and Der, C.J. (2002) RasGRP4 is a novel Ras activator isolated from acute myeloid leukemia. J Biol Chem, 277, 30508-30514. Rong, S.B., Enyedy, I.J., Qiao, L., Zhao, L., Ma, D., Pearce, L.L., Lorenzo, P.S., Stone, J.C., Blumberg, P.M., Wang, S. and Kozikowski, A.P. (2002) Structural basis of RasGRP binding to high-affinity PKC ligands. JMed Chem, 45, 853-860. Roose, J.P., Mollenauer, M., Ho, M., Kurosaki, T. and Weiss, A. (2007) Unusual interplay of two types of Ras activators, RasGRP and SOS, establishes sensitive and robust Ras activation in lymphocytes. Mo! Cell Biol, 27, 2732-2745. Shao, L., Lewin, N.E., Lorenzo, P.S., Hu, Z., Enyedy, I.J., Garfield, S.H., Stone, J.C., Mamer, F.J., Blumberg, P.M. and Wang, S. (2001) Iridals are a novel class of ligands for phorbol ester receptors with modest selectivity for the RasGRP receptor subfamily. JMed Chem, 44, 3872-3 880. Spaargaren, M. and Bischoff, J.R. (1994) Identification of the guanine nucleotide dissociation stimulator for Ral as a putative effector molecule of R-ras, H-ras, K-ras, and Rap. Proc Nat! Acad Sci USA, 91, 1260912613. Suzuki, T., Shen, H., Akagi, K., Morse, H.C., Malley, J.D., Naiman, D.Q., Jenkins, N.A. and Copeland, N.G. (2002) New genes involved in cancer identified by retroviral tagging. Nat Genet, 32, 166-174.  120  Taylor, S.J., Resnick, R.J. and Shalloway, D. (2001) Nonradioactive determination of Ras-GTP levels using activated ras interaction assay. Methods Enzymol, 333, 333-342. Teixeira, C., Stang, S.L., Zheng, Y., Beswick, N.S. and Stone, J.C. (2003) Integration of DAG signaling systems mediated by PKC-dependent phosphorylation of RasGRP3. Blood 102, 1414-1420. Tognon, C.E., Kirk, H.E., Passmore, L.A., Whitehead, I.P., Der, C.J. and Kay, R.J. (1998) Regulation of RasGRP via a phorbol ester-responsive Cl domain. Mol Cell Biol, 18, 6995-7008. Toki, S., Kawasaki, H., Tashiro, N., Housman, D.E. and Graybiel, A.M. (2001) Guanine nucleotide exchange factors Ca1DAG-GEFI and CaIDAG-GEFII are colocalized in striatal projection neurons. J Comp Neurol, 437, 398-407. Wang, H.W., Wang, Z., Dillon, T.J., Pokala, V., Mishra, S., Labudda, K., Hunter, B. and Stork, P.J. (2006) Rap 1-mediated activation of extracellular signal-regulated kinases by cyclic AMP is dependent on the mode of Rap 1 activation. Mol Cell Biol, 26, 2130-2145. Yeung, T., Gilbert, G.E., Shi, J., Silvius, J., Kapus, A. and Grinstein, S. (2008) Membrane phosphatidylserine regulates surface charge and protein localization. Science, 319, 210-213. Zheng, Y., Liu, H., Coughlin, J., Zheng, J., Li, L. and Stone, J.C. (2005) Phosphorylation of RasGRP3 on threonine 133 provides a mechanistic link between PKC and Ras signaling systems in B cells. Blood, 105, 3648-3654. Zugaza, J.L., Caloca, M.J. and Bustelo, X.R. (2004) Inverted signaling hierarchy between RAS and RAC in T lymphocytes. Oncogene, 23, 5823-5833.  121  CHAPTER 4: DISCUSSION  122  In this chapter, I outline the contributions of this thesis and present questions for future consideration. My aim in this discussion is to expand on previous discussion topics and examine broader concepts not already covered in the discussions of Chapter 2 and Chapter 3.  4.1  Cl DOMAIN MEDIATED LOCALIZATION OF RASGRPS  4.1.1  Not all Cl domains bind DAG My thesis research on Cl domain mediated localization of RasGRPs has given an  insight into the diversity of RasGRP Cl domains. All RasGRP Cl domains have the definitive spacing of two histidine and six cysteines to co-ordinate two zinc atoms, which in turn restrict the domain into a compact configuration. Closer analysis of ioops A and B, which insert into membranes and form a binding pocket for DAG, reveals that for RasGRP2 and RasGRP4f3 Cl domains, the binding pocket is unlikely to be able to tightly bind DAG, if at all. In RasGRP2, crucial residues in loop A and B are divergent, and in RasGRP4f3 there is a 5 amino acid sequence inserted into the base of the binding pocket just before loop B. My experimental findings in Chapter 2 establish that neither RasGRP2 nor RasGRP4f3 Cl domains bind DAG in membranes, a result which emphasises that not all RasGRP Cl domains are functionally equivalent. This raises questions concerning the current view of RasGRP2 regulation, and also points to a different mode of regulation for RasGRP4I3 compared to RasGRP4a. While some researchers have noted that RasGRP2 Cl domain is divergent (Irie et al., 2004; Stone, 2006), for the most part RasGRP2 is often cited as a protein regulated by Cl :DAG interactions (Crittenden et al., 2004; Ghandour et al., 2007; Katagiri et al., 2004; Kawasaki et al., 1998b). In the literature, RasGRP2 is frequently referred to as Ca1DAG-GEFI since it has a pair of EF hands that are thought to be regulated by calcium, and a Cl domain presumed to be regulated by DAG. Moreover, it appears to be a general assumption that DAG, which is produced by receptor mediated PLC activation, acts as a ligand for RasGRP2 (Crittenden et al., 2004; Ghandour et al., 2007; Katagiri et al., 2004; Pasvolsky et al., 2007). No other evidence has come to light, before our research on RasGRP2 and SDF-lc stimulation of T-cells, which demonstrates that RasGRP2 localizes to 123  membranes in response to receptor stimulation that generates DAG, an event which is now well established for RasGRP1 (Beaulieu et al., 2007; Bivona et al., 2003; Caloca et al., 2003b; Perez de Castro et al., 2004). So although some groups have shown that RasGRP2 is regulated by phorbol ester (PE) in a number of different contexts (Clyde-Smith et al., 2000; Crittenden et a!., 2004; Dupuy et a!., 2001; Kawasaki et al., 1998a) we propose that this is not likely to occur via PE:C1 mediated translocation of RasGRP2 to membranes. Instead these results may be explained by DAG activation of PKC enzymes, which when active may phosphorylate and activate RasGRP2 (Stone, 2006), as is the case for RasGRP1 and RasGRP3 (Aiba et a!., 2004; Brodie et a!., 2004; Roose et al., 2005; Teixeira et al., 2003; Zheng et al., 2005) which are phosphorylated at homologous positions, threonine 184 and threonine 133 respectively. RasGRP2 has a threonine at the homologous position 134 (Zheng et al, 2005), and therefore could be similarly regulated. Reduced electrophoretic mobility would indicate whether RasGRP2 was phosphorylated, although we did not observe this in response to TCR, BCR or SDF-la stimulation (Chapter 3; Figures 3.9 and 3.10), nor has it been reported by others. However, when looking at transduced GFP-tagged RasGRP2 in NIH 3T3s we did observe two bands that ran very closely together in NIH 3T3 cells, which could represent phosphorylated (upper band) and unphosphorylated (lower band) GFP-tagged RasGRP2 (Figure 3.4A; transduced RG2). Treatment of lysates with phosphatases could determine whether the upper band is phosphorylated GFP-tagged RasGRP2, and treatment of cells with PKC inhibitors would determine whether phosphorylation occured via PKC enzymes. 4.1.2  Not all RasGRP Cl domains are targeted to the same membranes Results from my thesis illustrate that RasGRP Cl domains not only have different  affinities for DAG, but they also target different membranes. In fluorescent microscopy experiments, the isolated Cl domains ofRasGRPl and 3 were shown to target the Golgi, while the RasGRP4a Cl domain is less effectively localized to the Golgi and can also be seen localizing in the nucleus, which may reflect reduced affinity of this Cl domain for DAG and/or membranes. In contrast, the isolated RasGRP4f3 and RasGRP2 Cl domains do not have any detectable localization at membranes. In Chapter 3 the RasGRP2 Cl domain was  124  however, shown to have affinity for the plasma membrane when tandemized, and RasGRP2 plasma membrane localization was abolished in the absence of the Cl domain.  Results in Chapter 2 showed that all RasGRP Cl domains have affinity for anionic phospholipids including phosphatidylserine (PS). PS is typically at higher concentrations in the cytoplasmic leaflet of the plasma membrane (10-20%) than at other organelles in the cell, and is one of the most abundant anionic phospholipids at this membrane location (Williamson and Schiegel, 1994; Zwaal and Schroit, 1997). PS asymmetry at the plasma membrane is upheld by an ATP-dependent aminophospholipid translocase that catalyzes the transport of selected aminophospholipids from the external leaflet to the internal leaflet of the plasma membrane (Seigneuret and Devaux, 1984). Given that the RasGRP2 Cl domain binds to vesicles containing anionic phospholipids such as PS but does not have DAG affinity (Johnson et al., 2007), this explains why the tandem form of the RasGRP2 Cl domain preferentially localizes at the plasma membrane where PS is in high concentration as opposed to the Golgi, where DAG is found in abundance. It is likely that in the case of the Cl domains of RasGRP1, 3 and 4x, DAG binding at the Golgi sequesters them away from the plasma membrane, however electrostatic interactions of cationic residues within these Cl domains with anionic phospholipids present at the Golgi could provide additional stability for DAG binding. An interesting case though is the Cl domain of RasGRP4f3 which has reduced affinity for anionic phospholipids compared to that of RasGRP4a and other RasGRP Cl domains. The only difference between RasGRP4cL and f3 is a five amino acid insertion in the DAG binding pocket of the RasGRP4I3 Cl domain, and it is possible that this insertion not only disrupts the DAG binding pocket, but also decouples the collective interactions of basic residues that are dispersed along the length of the Cl domain with anionic phospholipids. A continuous stretch of basic residues with consistent spacing may be required to facilitate strong anionic phospholipid mediated binding to membranes.  4.2  RASGRP2 LOCALIZATION i/IA THE PDZ-BINDING MOTIF During the course of my thesis research, my intention was to understand how  RasGRP2 localized to membranes, and whether such localization was required for its 125  activity. One major finding was that a putative binding site for PDZ proteins was required for maximal RasGRP2 localization at the Golgi, and also for full activation of Rap 1 by RasGRP2. When this PDZ-binding motif is mutated, RasGRP2 has reduced Golgi localization, but retains plasma membrane localization. In my experiments GFP also localized to a minor degree to the Golgi (Chapter 3; Figure 3.1, 3.3 and 3.8), leading to some overlap of signal with the Golgi GM 130 marker. Therefore mutation of the PDZ-binding motif may in fact remove all RasGRP2 localization at the Golgi, and what signal remains may simply be GFP signal. These results raise the hypothesis that tethering of RasGRP2 to the Golgi is mediated by a Golgi-localized PDZ protein. This could be tested by using co immunoprecipitation, at either a candidate or proteome-wide level, to identify PDZ proteins that bind to RG2 and are localized to the Golgi. Identification of Golgi localized partner PDZ protein for RasGRP2 would help to understand how RasGRP2 localization and activation at the Golgi is regulated by functional modulation or differential expression of PDZ proteins. Clustering of RasGRP2 at PDZ protein scaffold structures at the Golgi may cause oligomerization, thus potentially enhancing its binding to membranes via other domains. 4.3  RECEPTOR MEDIATED RASGRP2 ACTIVATION  4.3.1  Constitutive membrane localization of RasGRP2 My results show that RasGRP2 is constitutively localized at the Golgi and the plasma  membrane in both NIH 3T3 and Jurkat cells, and that this requires the Cl domain. Perez de Castro et al, demonstrated that while RasGRP1 and 3 translocated to the Golgi in response to T-cell receptor (TCR) stimulation of Jurkat T-cells with antibodies for CD3 and CD28, RasGRP2 and 4cc did not (Perez de Castro et al., 2004). We also found that RasGRP2 did not translocate to membranes and did not mediate Rapi activation in response to TCR stimulation in the same cells. This is intriguing because in Jurkat cells, RasGRP2 has been shown to activate and co-immunoprecipitate with Rapi and mediate adhesion to ICAM-l in response to TCR stimulation (Katagiri et al., 2004), although this group did not examine RasGRP2 translocation to membranes. It is quite possible that RasGRP2 is not regulated by translocation to membranes in T-cells, but instead the proportion of RasGRP2 which is 126  constitutively at membranes is activated in response to TCR stimulation. My results suggest that constitutive localization at the plasma membrane of Jurkat T-cells could be determined by Cl :anionic phospholipid interactions, while the Golgi localization via the PDZ-binding motif may involve a PDZ domain-containing protein partner. Although we have not established that TCR signals induce RasGRP2 activation, this may occur at either the plasma membrane or the Golgi. One possibility is that TCR stimulation induces PKC activation at one or both of these membrane sites, which in turn activates RasGRP2 by phosphorylation. Although PKCs have been shown to activate RasGRP1 and 3, they have not been shown to be required for RasGRP1 localization (Aiba et al., 2004) and therefore PKCs may activate constitutively localized RasGRP2. The plasma membrane has been determined to be the predominant site of TCR-stimulated Rap 1 activation and no evidence exists for activation of Rapi at the Golgi of these cells (Bivona et al., 2004). Taking this finding into account, it is likely that if TCR signaling leads to RasGRP2 activation, this would most likely occur at the plasma membrane.  4.3.2  SDF-hx induced plasma membrane localization of RasGRP2 RasGRP2 is required for the chemokine SDF-lcL, acting via the G-protein coupled  receptor CXCR4, to mediate LFA-l adhesion to ICAM-1 in primary human T-cells (Ghandour et al., 2007). These studies were interpreted by assuming that RasGRP2 is activated in the context of SDF- 1 c stimulation via its Cl domain binding to DAG generated by CXCR4 coupled to PLOy. During my thesis research I determined that SDF- 1 c treatment of Jurkat T-cells led to rapid concentration of RasGRP2 at the plasma membrane. That this requires the Cl domain is interesting, since the only determinant of Cl-mediated localization of RasGRP2 implied by my thesis iesearch thus far is via affinity for the anionic phospholipids of the inner leaflet of the plasma membrane. It is quite possible that SDF-lc stimulation of CXCR4 induces a rapid increase in the content of anionic phospholipids. Phosphatidylserine is already at high concentrations (10-20%) in the inner leaflet of the plasma membrane (van Meer et al., 2008) and its concentration would not be predicted to change as a result of TCR ligation. However, production of other anionic phospholipids at the cytoplasmic leaflet is induced in response to TCR stimulation, for example phosphatidic 127  acid (PA) and phosphatidylinositol phosphates (PIPs). PA is a relatively minor species with a net negative charge of-i .5. RasGRP Ci domains only bind to a high concentration of PA (30%) in vesicles (Johnson et al., 2007), and PA is not known to be induced to this concentration. PIPs have a higher net charge (-3 to-4 for P1(3,4) P 2 or P1(4,5) P 2 and -4 to -5 for PI(3,4,5)P ) and therefore may be more likely to interact with the basic residues within 3 the Cl domain of RasGRP2, even if present in the plasma membrane at a low concentration. RasGRP2 translocation in response to SDF- 1cc stimulation is very rapid, initiated at about 20 seconds and reversed by 50 seconds. Therefore the anionic phospholipids species that the RasGRP2 Cl domain interacts with at the plasma membrane would have to be similarly rapidly induced. By treating cells with SDF-lcc, concurrent with the P13K inhibitors LY294002 and Wortmanin, which reduce production of PIP3, it may be possible to determine whether PIP3 is required for RasGRP2 plasma membrane localization in response to SDF-lcc treatment.  While induced anionic phospholipids provide a plausible mechanism for Cl-mediated translocation of RasGRP2 to the plasma membrane in SDF-icc stimulated T-cells, it is also possible that RasGRP2 translocation occurs by another, unidentified mechanism.  4.4  HETEROGENEITY OF MECHANISMS FOR MEMBRANE  LOCALIZATION OF RASGRPS My thesis research has shown that the Cl domain is a critical determinant of membrane localization in all forms of RasGRPs, with the sole exception of RasGRP4f3 which may be a form of RasGRP4 that does not bind membranes. However, my studies of RasGRP2 show that the mechanisms by which RasGRP2 binds to membranes are quite distinct from those of other RasGRPs. RasGRP2 requires its Cl domain for plasma membrane localization but this is not via DAG. Instead, it probably occurs constitutively via affinity of the Ci domain for the anionic phospholipid phosphatidylserine (PS) and perhaps can be induced by generation of other anionic phospholipids such as phosphatidic acid (PA) or phosphatidylinositol phosphates (PIPs). The Cl domain of RasGRP2 is not required for Golgi localization, which is instead mediated via a PDZ-binding motif. This is in striking 128  contrast to RasGRP1 and RasGRP3, where the Cl domain is sufficient for Golgi localization, via Cl: DAG interactions. Like RasGRP2, RasGRP1 is additionally targeted to the plasma membrane, but this occurs via the PT domain (PT). Although the Cl domain cooperates with the PT, specification of plasma membrane targeting is provided by the PT not the Cl domain. RasGRP3 localization has not been as well studied as RasGRP1, but RasGRP3 has homology to the PT and has a DAG-binding Cl domain, so it may share the same modes of regulation as RasGRP 1. The RasGRP4cc Cl domain demonstrates a weaker affinity for DAG, and a proportionate reduction in Golgi localization via DAG affinity for the Cl domain. RasGRP4c also lacks the PT domain, and therefore it is likely that it cannot localize to the plasma membrane, unless this membrane location is the site of high DAG concentration. The RasGRP4f3 Cl domain has no DAG-binding affinity, and consequently demonstrates no localization at Golgi or any other membrane. The possibility remains that the RasGRP4f3 may be capable of plasma membrane localization via Cl binding to PS or other anionic phospholipids, although our in vitro data suggests that its localization by this mechanism will be less efficient in comparison to RasGRP2.  These functional differences in the localization mechanisms of RasGRPs are presumably the outcome of the evolutionary divergence of RasGRPs, which may have proceeded as follows: Starting with a single primordial RasGRP which probably had a Cl domain that bound DAG and anionic phospholipids, a duplication event produced two RasGRPs; RasGRP1/3 and RasGRP2/4. RasGRP1/3 then picked up a PT domain to additionally enable plasma membrane targeting in response to the PT ligand. Secondary duplication events gave rise to RasGRP1 and 3, and RasGRP2 and 4. RasGRP2 was then selected for loss of DAG recognition by its Cl domain with retention of anionic phospholipid recognition. Before or after the loss of DAG binding, RasGRP2 acquired a PDZ-binding motif to enable DAG-independent, but PDZ-binding motif-dependent Golgi localization. Meanwhile, the Cl domain of RasGRP4 diversified by being selected for reduced DAG affinity and additionally acquiring an alternative splice variant that eliminated DAG binding while retaining anionic phospholipid binding. It is possible that RasGRP4 evolved as a negative regulator of RasGRP4c signal transduction. Since RasGRP4x and RasGRPI3 are 129  identical except for a 5 amino acid insertion in the Cl domain, RasGRP4f3 may be able to bind and sequester the same proteins that RasGRP4cL interacts with, leading to attenuation of signal transduction from RasGRPc. However, this is quite speculative given that the only activating interaction currently known for RasGRP4cL is the Cl :DAG interaction, which does not occur with RasGRP4.  While this picture of RasGRP evolution is speculative, it is evident from my thesis research that functional divergence of RasGRPs has occurred, via both functional divergence of Cl domains and the acquisition (and perhaps loss) of other membrane-localizing domains. As genes duplicate and diversify, domains may retain functions as seen for RasGRP1 and 3 Cl domains, but they may also diversify their functions as seen for the Cl domains of RasGRP4 and RasGRP2. Diversification of Cl domains may occur as a result of the acquisition of new domains, which in turn may eliminate the requirement to retain the original functionality of the Cl domain. For example if RasGRP2 had to be functional at the Golgi it had to retain DAG binding capability of its Cl domain to achieve this. But once it acquired an alternative mechanism of Golgi localization, by acquisition of the PDZ-binding motif, it was possible for the Cl domain to lose the DAG binding ability and gain the ability to selectively target the plasma membrane via anionic phospholipid binding.  4.5  MEMBRANE SPECIFIC ACTIVATION OF RAP1 BY RASGRP2 Ras and Rap GTPases are relatively simple molecular switches that elicit a plethora  of cellular responses. Compartmentalization of Ras and Rap activation at different membranes can serve to increase the complexity of signal transduction by segregating the spatial activation of divergent pathways and providing a distinct output from the same pathway (Mor and Philips, 2006). Most of the evidence that supports this model of compartmentalized signaling comes from experiments using artificial targeting strategies that localize Ras to specific membranes (Chiu et al., 2002). Work in yeast has shown that restricting Rasi to the ER, as opposed to the plasma membrane abolishes mating, whereas restricting Ras 1 to the plasma membrane abolishes the ability to elongate (Papadaki et al., 2002). Others have shown that restricting GEFs to different membranes has effects on 130  signaling output. For example, artificially targeting the Rapi GEF Epaci to the plasma membrane via attachment of the K-Ras4B polybasic and prenylation signal sequence, versus its normal residence at the ER, leads to ERK activation that would otherwise not occur (Bergmeier et al., 2007).  In my thesis research, RasGRP2 was artificially targeted at the Golgi versus the plasma membrane, the two membrane compartments that RasGRP2 normally localizes at. We also studied the signaling output of two mutants of RasGRP2 that in effect similarly restricted RasGRP2 localization at the Golgi and plasma membrane. When RasGRP2 localization was restricted to the plasma membrane it induced morphological changes in NIH 3T3 cells equivalently to wild type RasGRP2. When RasGRP2 was targeted away from the plasma membrane via attachment of the Golgi targeting RasGRP1 Cl domain or by removing the Cl domain, Rapi activation occurred but did not result in cell morphology changes that were induced by wild type RasGRP2. It should be noted that RasGRP2mediated Rapi activation at the Golgi of NIH 3T3 cells is not necessarily without function, but instead could lead to distinct signaling outcomes that have other effects on cells. Co expression of a fluorescently-tagged Ra1GDS Rap binding domain (which only binds GTP loaded Rap) with GFP-tagged RasGRP2, and mutants RG2AC1 or RG2-PBp., could potentially determine whether Rapi activation via RasGRP2 actually occurs at both the plasma membrane and Golgi in NIH 3T3 cells.  One obvious question that arises from these results is how does plasma membranelocalized RasGRP2 mediate morphological changes in NIH 3T3 cells? RasGRP2-expressing cells have more distinct cell-cell boundaries, and do not migrate independently away from the colony, suggesting that these cells have increased cell-cell adhesion. Rap 1 has been implicated as having a role in cell-cell junction formation via the regulation of the actin cytoskeleton and the recruitment of E-cadherin (Kooistra et al., 2007). Expression of Rapi and the Rap 1 GEF Epac 1, increases cAMP induced cell-cell junction formation of MDCK cells (Price et al., 2004) and expression of Epac 1 also promotes VE-cadherin cell junction formation and actin remodelling in HUVEC (human umbilical vein endothelial cells) (Kooistra et al., 2005). One possibility is that RasGRP2-mediated Rapl activation at the 131  plasma membrane of fibroblasts leads to changes in cell-cell adhesion via promotion of actin polymerization. In support of this, RasGRP2 has been shown to interact with F-actin via its N-terminus in COS cells (Caloca et al., 2004). The morpholocial cellular response to Rapi activation observed in RasGRP2 expressing cells may be compartmentalized at the plasma membrane owing to the concentration of specific Rap 1 effectors at this location, e.g. Vav, Vav2 (Caloca et al., 2004) or Tiami (Baumeister et al., 2003; Buchsbaum et al., 2003), which could then promote actin polymerization via Rae GTPases. Mediators of actin polymerization e.g. Rae GTPases or their GEFs, are possibly not located or functional in this respect at the Golgi and therefore are not activated by Golgi-localized RasGRP2.  Thus, we have learned that RasGRP2 is capable of activating Rapi at two different membrane locations. The plasma membrane has been shown before to be a major site of Rapi localization and activation in different cells (Bivona et al., 2004; Medeiros et al., 2005) while the Golgi has so far only been shown to be a site of localization of Rapi (Beranger et al., 1991; Nomura et al., 2004). In addition, these experiments further demonstrate that compartmentalization of Rapi activation at different membranes leads to distinct signaling outcomes, possibly reflecting differential availability of Rapi effectors. Taken together these results add weight to the hypothesis that compartmentalization of Ras and Rap GTPase signaling is determined by three major factors: a) differential localization of Ras and Rap GTPases via their distinct membrane targeting mechanisms, b) compartmentalization of GEF membrane localization and perhaps also compartmentalization of GAPs, which are negative regulators of Ras and Rap GTPases, and c) compartmentalization of Ras and Rap effectors.  132  4.6  GENERAL CONCLUSIONS AND IMPLICATIONS FOR RASGRP2  REGULATION Before I started my thesis research, nothing was known about how RasGRP2 targeted membranes in order to access its substrate Rap 1. It was assumed that RasGRP2 behaved the same way as RasGRP 1, and therefore research on RasGRP2 has focussed on a model in which the Cl domain binds to DAG that is generated in membranes by receptor stimulation. DAG can be generated in reponse to a number of receptor-mediated signal transduction pathways, for example TCR stimulation, which activates PLC-yl that in turn cleaves the plasma membrane phospholipid PIP2 into DAG and 1P3. Through the course of my research, I found that the isolated Cl domain of RasGRP2 does not bind to DAG, a finding which supported similar results from another group (Irie et al., 2004). Subsequently, I found that the RasGRP2 Cl domain is however required for plasma membrane localization, which implies that other membrane localization roles for the Cl domain have evolved, other than the conventional DAG affinity function. Another important finding was that RasGRP2 localization at the Golgi is specified by a previously unrecognized PDZ-binding motif. Both the Cl domain and the PDZ-binding motif are required for maximal RasGRP2 activation of Rap 1 in NIH 3T3 cells, therefore research which helps to decipher in more detail how the Cl domain and the PDZ-binding motif promote RasGRP2 membrane localization and Rapl activation, is likely to be important in advancing our understanding of how RasGRP2 is regulated. Findings from my thesis research are a step in this direction.  Changes in RasGRP2 expression are associated with disease phenotypes. Deregulated-increase in expression of RasGRP2 leads to lymphoma (Mikkers et a!., 2002; Suzuki et al., 2002) and myeloid leukemia (Dupuy et al., 2001) in mice, while knockout mice have severe platelet (Bergmeier et a!., 2007; Crittenden et a!., 2004) and leukocyte adhesion defects (Bergmeier et a!., 2007). In the case of cancer, RasGRP2 may activate the ERK1/2 pathway through constituitive Rapi activation of B-Raf, whereas in RasGRP2 deficient leukocytes and platelets, lack of RasGRP2-mediated Rap 1 activation in response to receptor stimulation (e.g. the SDF-lcL chemokine receptor CXCR4), is thought to lead to a defect in inside-out signaling, which in turn leads to deficient integrin-mediated adhesion. New 133  studies in leukocyte adhesion deficiency (LAD) III patients that have severely reduced RasGRP2 expression represent the first discovery of a role for RasGRP2 in human disease (Pasvolsky et a!., 2007). Thus, advances in the understanding of how RasGRP2 is normally regulated by membrane localization and activation, will provide key knowledge about how this GEF is positioned to promote cancer. Moreover, by learning how membrane localization regulates RasGRP2 to promote integrin-mediated adhesion via Rapi activation, for example in response to chemokine stimulation of lymphocytes, will better our understanding of this fundamental cell response, and may give new insights into pathologies such as LAD III.  134  4.6  BIBLIOGRAPHY  Aiba, Y., Oh-hora, M., Kiyonaka, S., Kimura, Y., Hijikata, A., Mori, Y. and Kurosaki, T. (2004) Activation of RasGRP3 by phosphorylation of Thr-133 is required for B cell receptor-mediated Ras activation. Proc NatlAcadSci USA, 101, 16612-16617. Baumeister, M.A., Martinu, L., Rossman, K.L., Sondek, J., Lemmon, M.A. and Chou, M.M. (2003) Loss of phosphatidylinositol 3-phosphate binding by the C-terminal Tiam-1 pleckstrin homology domain prevents in vivo Rac 1 activation without affecting membrane targeting. J Biol Chem, 278, 1145711464. Beaulieu, N., Zahedi, B., Goulding, R.E., Tazmini, G., Anthony, K.V., Omeis, S.L., de Jong, D.R. and Kay, R.J. (2007) Regulation of RasGRP 1 by B cell antigen receptor requires cooperativity between three domains controlling translocation to the plasma membrane. Mol Biol Cell, 18, 3156-3168. Beranger, F., Goud, B., Tavitian, A. and de Gunzburg, J. (1991) Association of the Ras-antagonistic Rapl/Krev-1 proteins with the Golgi complex. Proc NatlAcadSci USA, 88, 1606-1610. Bergmeier, W., Goerge, T., Wang, H.W., Crittenden, J.R., Baldwin, A.C., Cifuni, S.M., Housman, D.E., Graybiel, A.M. and Wagner, D.D. (2007) Mice lacking the signaling molecule Ca1DAG-GEFI represent a model for leukocyte adhesion deficiency type III. J Clin Invest, 117, 1699-1707. Bivona, T.G., Perez De Castro, I., Ahearn, I.M., Grana, T.M., Chiu, V.K., Lockyer, P.J., Cullen, P.J., Pellicer, A., Cox, A.D. and Philips, M.R. (2003) Phospholipase Cgamma activates Ras on the Golgi apparatus by means of RasGRP1. Nature, 424, 694-698. Bivona, T.G., Wiener, N.H., Ahearn, I.M., Silletti, J., Chiu, V.K. and Philips, M.R. (2004) Rapl up-regulation and activation on plasma membrane regulates T cell adhesion. J Cell Biol, 164, 461-470. Brodie, C., Steinhart, R., Kazimirsky, G., Rubinfeld, H., Hyman, T., Ayres, J.N., Hur, G.M., Toth, A., Yang, D., Garfield, S.H., Stone, J.C. and Blumberg, P.M. (2004) PKCdelta associates with and is involved in the phosphorylation of RasGRP3 in response to phorbol esters. Mol Pharmacol, 66, 76-84. Buchsbaum, R.J., Connolly, B.A. and Feig, L.A. (2003) Regulation of p70 S6 kinase by complex formation between the Rac guanine nucleotide exchange factor (Rac-GEF) Tiaml and the scaffold spinophilin. J Biol Chem, 278, 18833-1884 1. Caloca, M.J., Zugaza, J.L., Matallanas, D., Crespo, P. and Bustelo, X.R. (2003b) Vav mediates Ras stimulation by direct activation of the GDP/GTP exchange factor Ras GRP1. Embo .J 22, 3326-3336. Caloca, M.J., Zugaza, J.L., Vicente-Manzanares, M., Sanchez-Madrid, F. and Bustelo, X.R. (2004) F-actin dependent translocation of the Rapi GDP/GTP exchange factor RasGRP2. JBiol Chem, 279, 2043520446. Chiu, V.K., Bivona, T., Hach, A., Sajous, J.B., Silletti, J., Wiener, H., Johnson, R.L., 2nd, Cox, A.D. and Philips, M.R. (2002) Ras signalling on the endoplasmic reticulum and the Golgi. Nat Cell Biol, 4, 343350. Clyde-Smith, J., Silins, G., Gartside, M., Grimmond, S., Etheridge, M., Apolloni, A., Hayward, N. and Hancock, J.F. (2000) Characterization of RasGRP2, a plasma membrane-targeted, dual specificity Ras/Rap exchange factor. JBiol Chem, 275, 32260-32267. Crittenden, J.R., Bergmeier, W., Zhang, Y., Piffath, C.L., Liang, Y., Wagner, D.D., Housman, D.E. and Graybiel, A.M. (2004) Ca1DAG-GEFI integrates signaling for platelet aggregation and thrombus formation. Nat Med, 10, 982-986. Dupuy, A.J., Morgan, K., von Lintig, F.C., Shen, H., Acar, H., Hasz, D.E., Jenkins, N.A., Copeland, N.G., Boss, G.R. and Largaespada, D.A. (2001) Activation of the Rapl guanine nucleotide exchange gene, Ca1DAG-GEF I, in BXH-2 murine myeloid leukemia. J Biol Chem, 276, 11804-11811. Ghandour, H., Cullere, X., Alvarez, A., Luscinskas, F.W. and Mayadas, T.N. (2007) Essential role for Rap 1 GTPase and its guanine exchange factor Ca1DAG-GEFI in LFA- 1 but not VLA-4 integrin mediated human T-cell adhesion. Blood, 110, 3682-3690. Irie, K., Masuda, A., Shindo, M., Nakagawa, Y. and Ohigashi, H. (2004) Tumor promoter binding of the protein kinase C Cl homology domain peptides of RasGRPs, chimaerins, and Unc 1 3s. Bioorg Med Chem, 12, 4575-4583. Johnson, J.E., Goulding, R.E., Ding, Z., Partovi, A., Anthony, K.V., Beaulieu, N., Tazmini, G., Cornell, R.B. and Kay, R.J. (2007) Differential membrane binding and diacylglycerol recognition by Cl domains of RasGRPs. Biochem J 406, 223-236.  135  Katagiri, K., Shimonaka, M. and Kinashi, T. (2004) Rap 1-mediated lymphocyte function-associated antigen-i activation by the T cell antigen receptor is dependent on phospholipase C-gammal. JBiol Chem, 279, 11875-11881. Kawasaki, H., Springett, G.M., Mochizuki, N., Toki, S., Nakaya, M., Matsuda, M., Housman, D.E. and Graybiel, A.M. (1 998a) A family of cAMP-binding proteins that directly activate Rap 1. Science, 282, 2275-2279. Kawasaki, H., Springett, G.M., Toki, S., Canales, J.J., Harlan, P., Blumenstiel, J.P., Chen, E.J., Bany, l.A., Mochizuki, N., Ashbacher, A., Matsuda, M., Housman, D.E. and Graybiel, A.M. (1998b) A Rap guanine nucleotide exchange factor enriched highly in the basal ganglia. Proc Nat! Acad Sci US A, 95, 13278-13283. Kooistra, M.R., Corada, M., Dejana, E. and Bos, J.L. (2005) Epac 1 regulates integrity of endothelial cell junctions through VE-cadherin. FEBS Lett, 579, 4966-4972. Kooistra, M.R., Dube, N. and Bos, J.L. (2007) Rap 1: a key regulator in cell-cell junction formation. J Ce!! Sci, 120, 17-22. Medeiros, R.B., Dickey, D.M., Chung, H., Quale, A.C., Nagarajan, L.R., Billadeau, D.D. and Shimizu, Y. (2005) Protein kinase Dl and the beta 1 integrin cytoplasmic domain control beta 1 integrin function via regulation of Rap 1 activation. Immunity, 23, 213-226. Milckers, H., Allen, J., Knipscheer, P., Romeijn, L., Hart, A., Vink, E., Berns, A. and Romeyn, L. (2002) Highthroughput retroviral tagging to identifS’ components of specific signaling pathways in cancer. Nat Genet, 32, 153-159. Mor, A. and Philips, M.R. (2006) Compartmentalized Ras/MAPK signaling. Annu Rev Immunol, 24, 771-800. Nomura, K., Kanemura, H., Satoh, T. and Kataoka, T. (2004) Identification of a novel domain of Ras and Rap 1 that directs their differential subcellular localizations. J Bio! Chem, 279, 22664-22673. Papadaki, P., Pizon, V., Onken, B. and Chang, E.C. (2002) Two ras pathways in fission yeast are differentially regulated by two ras guanine nucleotide exchange factors. Mo! Ce!! Bio!, 22, 4598-4606. Pasvolsky, R., Feigelson, S.W., Kilic, S.S., Simon, A.J., Tal-Lapidot, G., Grabovsky, V., Crittenden, J.R., Amariglio, N., Safran, M., Graybiel, A.M., Rechavi, G., Ben-Dor, S., Etzioni, A. and Alon, R. (2007) A LAD-Ill syndrome is associated with defective expression of the Rap-i activator Ca1DAG-GEFI in lymphocytes, neutrophils, and platelets. JExp Med 204, 157 1-1582. Perez de Castro, I., Bivona, T.G., Philips, M.R. and Pellicer, A. (2004) Ras activation in Jurkat T cells following low-grade stimulation of the T-cell receptor is specific to N-Ras and occurs only on the Golgi apparatus. Mo! Cell Bio!, 24, 3485-3496. Price, L.S., Hajdo-Milasinovic, A., Zhao, J., Zwartkruis, F.J., Collard, J.G. and Bos, J.L. (2004) Rapi regulates E-cadherin-mediated cell-cell adhesion. JBiol Chem, 279, 35127-35132. Roose, J.P., Mollenauer, M., Gupta, V.A., Stone, J. and Weiss, A. (2005) A diacylglycerol-protein kinase C RasGRP 1 pathway directs Ras activation upon antigen receptor stimulation ofT cells. Mo! Ce!! Bio!, 25, 4426-4441. Seigneuret, M. and Devaux, P.F. (1984) ATP-dependent asymmetric distribution of spin-labeled phospholipids in the erythrocyte membrane: relation to shape changes. Proc Nat! Acad Sc! USA, 81, 3751-3755. Stone, J.C. (2006) Regulation of Ras in lymphocytes: get a GRP. Biochem Soc Trans, 34, 858-861. Suzuki, T., Shen, H., Akagi, K., Morse, H.C., Malley, J.D., Naiman, D.Q., Jenkins, N.A. and Copeland, N.G. (2002) New genes involved in cancer identified by retroviral tagging. Nat Genet, 32, 166-174. Teixeira, C., Stang, S.L., Zheng, Y., Beswick, N.S. and Stone, J.C. (2003) Integration of DAG signaling systems mediated by PKC-dependent phosphorylation of RasGRP3. B!ooc4 102, 1414-1420. van Meer, G., Voelker, D.R. and Feigenson, G.W. (2008) Membrane lipids: where they are and how they behave. Nat Rev Mo! Ce!! Bio!, 9, 112-124. Williamson, P. and Schlegel, R.A. (1994) Back and forth: the regulation and function of transbilayer phospholipid movement in eukaryotic cells. Mo! Membr Bio!, 11, 199-216. Zheng, Y., Liu, H., Coughlin, J., Zheng, J., Li, L. and Stone, J.C. (2005) Phosphorylation of RasGRP3 on threonine 133 provides a mechanistic link between PKC and Ras signaling systems in B cells. Blood, 105, 3648-3654. Zwaal, R.F. and Schroit, A.J. (1997) Pathophysiologic implications of membrane phospholipid asymmetry in blood cells. B!ood, 89, 1121-1132.  136  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0069949/manifest

Comment

Related Items