UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Functional studies of grasshopper semaphorin-2A Boileau, Ève 2003

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

Item Metadata


831-ubc_2003-0193.pdf [ 4.76MB ]
JSON: 831-1.0090931.json
JSON-LD: 831-1.0090931-ld.json
RDF/XML (Pretty): 831-1.0090931-rdf.xml
RDF/JSON: 831-1.0090931-rdf.json
Turtle: 831-1.0090931-turtle.txt
N-Triples: 831-1.0090931-rdf-ntriples.txt
Original Record: 831-1.0090931-source.json
Full Text

Full Text

FUNCTIONAL STUDIES OF GRASSHOPPER SEMAPHORIN-2A by EVE BOILEAU B.Sc, Universite de Montreal, 2000 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER of S c i e n c e in THE FACULTY OF GRADUATE STUDIES Program in Neuroscience We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 2003 © Eve Boileau, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) A B S T R A C T One family of neuronal guidance cues used by growth cones to pathfind correctly is the semaphorin family. Although most members function as inhibitory guidance cues, the grasshopper semaphorin-la is attractive. Previous experiments in Drosophila and grasshopper have suggested that semaphorin-2a is a neuronal repulsive guidance cue, however, in both cases the results have been equivocal. In this thesis, I directly test the function of grasshopper semaphorin-2a in vivo, while presenting ectopic sources of semaphorin-2a during the neuroembryonic development of grasshoppers. In addition, as a first approximation to determine protein regions used for function, I analysed whether a chimeric semaphorin that contains the transmembrane and the cytoplasmic domains of semaphorin-la and the semaphorin domain of semaphorin-2a functions like the wild type semaphorin-2a. I found that the addition of cells expressing both the chimera and the wild type Sema-2a significantly increases the proportion of misguided growth cones in the developing peripheral nervous system. Grasshoppers cultured with semaphorin-2a expressing cells showed a significant increase in errors with misguidance apparent in 79% of the pathway analyzed. Similarly, chimera expressing cells induced misguidance in 84% of the cases. Misguidance-errors were categorised into four groups: dorsal projections, axon defasciculation, growth impairment and emergence^ of multiple axon sprouts directed away from expressing cells. The proportion of the phenotypes-induced are hot significantly different whether full length of semaphorin-2a or chimera expressing cells were placed in the limb and all phenotypes suggest that semaphorin-2a and the ii chimera protein both function as repulsive guidance cues in the developing peripheral nervous system. This data shows that the semaphorin domain of Sema-2a is sufficient for its repulsive function and that the generation of a transmembrane version of this protein does not perturb this function. iii A B S T R A C T ii Table of Content iv List of Tables vi List of Figures vii List of Abbreviations viii INTRODUCTION 1 Organism Development and Cell Migration 1 The Growth Cone 2 Cytoskeletal Dynamics 3 Growth Cone Guidance 7 The Semaphorin Protein Family 8 The Vertebrate Semaphorin 9 The Invertebrate Semaphorins 12 Structure of the Grasshopper Sema-2a 12 Invertebrate Homologues of the Grasshopper Sema-2a 13 Plexins and Neuropilins: the Semaphorin Receptors 13 Embryonic Grasshopper as a Model Organism 15 Embryogenesis of the Grasshopper 15 Guidance Cues in the Developing Limb Bud 19 Advantages of the Embryonic Grasshopper as a Model System 20 Objectives 20 M E T H O D S 24 Protein Constructs 24 Sema-la/2a Overview 24 Additional Construct 25 Expression System 26 Antibody Work 29 Purification of A and E Antigenic Fusion Proteins 29 GST Column Preparation 30 Antibody Purification 30 Immunostaining 31 Non Fluorescence Labelling 31 Immunofluorescence 31 Immunoprecipitation 32 Immunoblot 32 Limb-Fillets System 33 Analysis of misguidance 34 RESULTS 35 Antibody Characterisation 35 Dynamic Expression of gSema-2a during Embryonic Development 38 Protein Expression in S2 Cells 39 iv Immunoblot 40 Function of gSema-2a Protein and its Semaphorin Domain 49 Function of gSema-2a Wild Type and Chimera 52 Misguidance Phenotypes 53 DISCUSSION 66 gSema-2a in the Developing Embryo: Dynamic Expression and Function 66 Protein Function Analysis 67 Secreted and Discrete Signaling Mechanisms 69 Growth Cone Misrouting 71 F u t u r e D i r e c t i o n s 77 Hypothetical Fragments used for Function 77 CONCLUSION 87 R E F E R E N C E S 88 Appendix 1 99 Appendix 2 100 Appendix 3 101 Appendix 4 102 Appendix 5 103 Appendix 6 104 v LIST OF TABLES Table 1. Til growth cones make frequent aberrant steering decisions when ectopic sources of gSema-2a or its transmembrane form are strategically placed in the Til pathway Table 2. Primary sequence alignment of the tarantula hanatoxin protein fragment with cSema-3A, and its comparison to gSema-2a and to gSema-la. proteins LIST OF FIGURES Figure 1. Schematic of the anatomy of a growth cone. 5 Figure 2. Semaphorin proteins and their receptors. 10 Figure 3. Signaling of plexin the receptors. 17 Figure 4. Schematic of the developing grasshopper limb bud at 34 % of development. 21 Figure 5. Schematic representation of the protein constructs. 27 Figure 6. A and E antigenic fusion protein purification. 36 Figure 7. gSema-2a protein expression is developmentally regulated in the embryonic grasshopper. 41 Figure 8. Axonal elongation of the Til pioneer neurons. 41 Figure 9. gSema-2a and chimera proteins expression in S2 cells. 43 Figure 10. gSema-2a and chimera proteins expression detected with the E antibody. 45 Figure 11. t-Sema-la protein detected with the V5 antibody. 47 Figure 12. Western analysis of chimera and gSema-2a expression. 50 Figure 13. Ectopic expression of gSema-2a perturbs Til growth cone steering. 55 Figure 14. Ectopic expression of the chimera perturbs Til growth cone steering. 57 Figure 15. Comparison of the frequencies of Til pioneer neurons guidance errors when their growth cones are navigating in the presence of different expression cell lines. 59 Figure 16. Aberrant Til pioneer steering decisions are divided into four subgroups based on their phenotype. 63 Figure 17. Model proposed for growth cone steering involving repulsive guidance molecules expressed in gradients. 72 Figure 18. Primary sequence alignment of cSema-3A and gSema-2a semaphorin domains. 79 Figure 19. Primary sequence alignment of the semaphorin domains from gSema-1 a and gSema-2a. 84 vii LIST OF ABBREVIATIONS A adenine Ab antibody Ca 2 + cationic calcium cAMP cyclic adenosine monophosphate C Cytosine CeSema-2a C. elegans semaphorin-2a cGMP cyclic guanosine monophosphate CNS central nervous system CRMP collapsin response mediator protein CUB complement binding domain Cx coxa limb segment Cy3 indocarbocyanine DAB 3,3'-Diaminobenzidine peroxidase substrate DNA deoxyribonucleic acid dATP deoxy adenosine triphosphate dCTP deoxy cytosine triphosphate dGTP deoxy guanosine triphosphate dNTP deoxynucleotide triphosphate DRG dorsal root ganglion dTTP deoxy thymidine triphosphate dSema-2a Drosophila semaphorin-2a ECM extracellular matrix E. coli Eschericia coli F-actin actin filament FBS-HI Foetal bovine serum heat inactivated Fe femur limb segment G guanine G-actin globular actin GPI glycoinositol phospholipid gSema-la grasshopper semaphorin-la gSema-2a grasshopper semaphorin-2a GST glutathione S-transferase GTP guanosine triphosphate HRP horse radish peroxydase Ig Immunoglobulin FPTG Isopropylthiogalactoside kb kilo base pair kD kilo Dalton mg Milligram MgCl2 magnesium chloride MgS04 magnesium sulphate mM Millimolar ml milliter MICALs molecule interacting with CasL viii O.D. optic density OTK Off-track NAPS Nucleic Acid and Protein Services Net Netrin P Phosphate PAK p21-activated kinase PBS phosphate buffered saline PCR polymerase chain reaction PLX Plexin NP neuropilin PNS peripheral nervous system PSD post-synaptic density protein SDS-PAGE sodium dodecyl sulfate polyacrylamide gel electrophoresis Sema Semaphorin SGO subgenualorgan SOEing gene splicing by overlap extension T Thymidine Ti tibia limb segment Tr segment trochanter limb t-Sema-la truncated grasshopper semaphorin-la UV Ultraviolet rig Microgram Ul microliter uM micromolar I N T R O D U C T I O N Organism Development and Cell Migration During the development of an organism, cells often have to travel long distances to reach their final location. This process is very complex in the nervous system where neurons have to migrate and extend processes along pathways mapped by a combination of positive and negative guidance cues which guide them to their specific target (review by Tessier-Lavigne and Goodman, 1996). Indeed, a major subject of investigation in neuroscience is the examination of neuronal pathfinding: how developing neurons establish functional connections with a specific target. Hence one of the major challenges of developmental neuroscience is the explanation of how dendrites and axons identify and make functional connections to their specific partners and finally create a highly complex and functional neural network. In order to simplify the concept, three successive events can be considered. Firstly, a neuron differentiates and initiates neurite outgrowth. At the tip of this extending axon, a growth cone directs this navigation through the nervous system by responding to a variety of positive and negative guidance cues. The second major event is target recognition, where the pathfinding axons stop migrating after contacting their target cells, which normally includes other neurons, muscle fibres, or glands. It is during these two stages of neuronal development that guidance cues play a central role. These "guidance cues" elicit stereotypical responses by the growth cone (extension or retraction via the modification of the neuronal cytoskeleton; Fan et al., 1995;Challacombe et al. 1996; Zhou et al., 2002) and the combination of cues leads the growth cone to the proper destination. After locating their final targets, neurons establish functional synapses and undergo a maturation process involving morphological modifications and stabilisation of processes. This last step relies on synaptic activity (see Misgeld et al., 2002). Early during an organism's development, neurons pathfind within the central and peripheral nervous systems (CNS and PNS) in the presence of a considerable amount of guidance information (review by Tessier-Lavigne and Goodman, 1996). The expression of many of these permissive guidance cues is transient and in the early postnatal (and adult), there is a reduction in guidance cues and some of them change from attractive to non permissive cues as neurons develop (Mukhopadhyay et al., 1994), and the environment of the CNS changes possibly to stabilize the connections that are already formed (Buffo et al., 2000). Consequently, 1 the environment becomes less stimulating to neuronal outgrowth. Thereby, when damage/trauma occurs in the mature CNS (for example, damaging the spinal cord), the generally poor environment for growth and the generation of a physical barrier (scar) at the lesion site, all contribute to the failure in the reestablishment of the functional neuronal network (Tang et al., 2001). Consistent with the idea that the embryonic environment stimulates neuronal outgrowth (review by Holm and Isacson, 1999), one aspect of the approach currently considered for the treatment of neurons damaged in the CNS is by simulating the neuroembryonic environment. It is hoped that this will stimulate mature neurons to re-grow. A second clinical consideration that illustrates the importance of a better understanding of neuronal guidance cues is in the elaboration of a strategy to overcome neurodegenerative diseases. For example, there is hope for a positive effects on Parkinson's disease with neuronal or stem cells transplants. The clinical potentials of neural stem cells for treatments of Parkinson's disease are reviewed by Lindvall, (2003). However, it is thought that these new cells will have to be in an appropriate environment to be able to survive, grow and make functional synapses (review by Lindvall, 2003). In this respect, guidance molecules would be one of the key tools to master in order to stimulate neuronal growth and function, as well as promoting appropriate and specific synapses formation of the neuronal transplants with endogenous remaining cells. Generally, neurons in the CNS are incapable of regeneration to their target after injury (review by Horner and Gage, 2000). However, there are events such as learning and memory where morphological changes underline that plasticity is maintained in the mature CNS. Such neuronal modifications include growth and pruning of neuronal processes, addition and removal of synapses and changes of synaptic size and shape throughout the organism's life time (Toni et al., 2001; reviewed by Marrone and Petit, 2001). The Growth Cone Structure The growth cone is situated at the end of a developing axon and is composed of two structures distinguished by their cytoskeleton arrangement (figure 1). The first structure is composed of independent finger-like protrusions termed filopodia. The cytoskeleton of filopodia primarily consists of parallel oriented actin filaments (Bridgman and Dailey, 1989). Filopodia are considered to form a specialized structure that reads and senses its environment for instructive cues (Bentley and Toroian-Raymond, 1986; Davenport et al., 1993), leading the 2 orientation of axon elongation. Also, each filopodium has the ability to extend or retract independently from other filopodia (O'Connor et ah, 1990; Fan and Raper, 1995). An example of autonomous filopodial sensory activity was observed when a gradient of netrin-1 was generated across the growth cone. One side of the growth cone extended filopodia, while the other side retracted filopodia (Fan and Raper, 1995; Song and Poo, 2001). Therefore, it suggests that each filopodium has its own set of receptors along with the necessary machinery to independently transduce information (O'Connor et al., 1990). The second structure of importance in a growth cone is a thin veil or sheet-like structure that extends between each filopodium, called lamellipodia. The major cytoskeleton component of lamellipodia is a dense meshwork of crossing actin filaments (Lewis and Bridgman, 1992). While little evidence has demonstrated the key components between guidance receptors and the elicited growth cone behaviour, an appealing model has been proposed (Fan et al., 1993; Fan and Raper, 1995; review by Luo, 2002). The model suggests that detection of positive or negative cues by the growth cone results in either a stabilization or destabilization of the underlying cytoskeleton. These cytoskeletal dynamics result in either growth or retraction of the growth cone area that has been in contact with these guidance cues (Fan and Raper, 1995). This cytoskeletal modification triggers growth cone turning towards or away from the source, and in some conditions, collapsing of the whole structure occurs if the entire surface of a growth cone has been in contact with a negative cue. Cytoskeletal Dynamics For the past 20 years, an increasing number of studies have reported that microtubules and actin filaments are crucial for many axonal physiological changes, such as axon extension and growth cone orientation (Mallavarapu and Mitchison, 1999; Challacombe et al., 1996). The ability to elongate a neurite is essentially conferred by cytoskeletal protein anchors and adhesion molecules, which establish successive transient contacts with the extracellular matrix (ECM). Although adhesion is essential for the growth cone advancement, this adhesion state must be repeatedly established and broken. The affinity for the ECM is often modulated by changes in the phosphorylation state of some key adhesion molecules, including the phosphorylation of proteins like integrin which are members of a complex of proteins linking the ECM to the cytoskeleton (review by Suter and Forscher, 2000). 3 One cytoskeletal component that has a major impact on axon elongation and growth cone steering is actin. As mentioned above, the actin filaments are particularly concentrated at the leading edge of axons, in the periphery of the growth cone within lamellipodia and filopodia substructures (Lewis and Bridgman, 1992). Actin filaments are formed by the polymerisation of actin subunits, G-actin, at the plus end of the filament. Similar to microtubules, actin filaments are dynamic structures that also undergo polymerisation at their plus end (Black, 1994; Gallo and Letourneau, 1999). When the neuronal environment is permissive for growth, downstream signaling pathways are activated by stimulated receptors, increasing the structure of actin filaments to become more stable and filaments to cluster, resulting in an increase of the ratio G-actin assembly/F-actin depolymerisation (reviewed by Pollard and Borisy, 2003). In consequence, filopodia and lamellipodia elongate. It is a well known fact that guidance molecules function on growth cone through their receptors. What remains to be clarified are the identities and the functions of all the players involved in mediating the signal from activated receptors to the actin cytoskeleton. While not all of the molecules have been identified, since the early '90s, many studies have demonstrated that three members of the Rho guanosine triphosphatases (GTPases) proteins family, RhoA, Rac 1 and Cdc42, are involved in a variety of neuronal signaling events associated with growth cone extension or retraction (review by Mayer and Feldman, 2002). The activation of RhoA was first associated with the formation of stress fibres in fibroblasts. In parallel, when RhoA is activated in neuronal cells, this molecule is usually found associated with the rearrangement of the cytoskeleton components promoting growth cone collapse (Wahl et al, 1998; Hu et al., 2001; Driessens et al, 2001; Niederost et al, 2002). On the other hand, Racl and Cdc42 are associated with axon extension (Bito et al., 2000; Li et al., 2002), Racl functioning at the level of lamellipodia and Cdc42 functioning at the level of filopodia (reviewed by Hall, 1998; reviewed by Luo et al., 2002). Another type of player involved in signaling, Ca2+; has been extensively studied because of its broad effect on neuronal motility. Similar to guidance cues that are both repulsive and attractive, Ca2+ is also reported to have a dual function on neurons. This second messenger has been proven to be important for axon elongation, to be involved in neuron maturation (Gu and Spitzer, 1995; Lau et al, 1999) and also to be involved for growth cone collapse (Loschinger et al, 1997). Recently, evidence has accumulated showing that the level of cGMP or cAMP is crucial for determining a growth cone response. For example, a recent study has shown that cAMP and PKA can decrease the repulsive effects from 4 Figure 1 Schematic of the anatomy of a growth cone. The growth cone is composed of two main structures, called filopodia and lamellipodia. Filopodia are finger-like structures protruding from the leading edge of the growth cone and form the sensory unit of the growth cone. Lamellipodia form a thin sheet-like structure extending between filopodia. Both structures are formed with actin filaments. An actin filament is a polarised structure made by the polymerisation of G-actin subunits at the plus end (green label) of the filament. Actin filaments in filopodia are oriented parallel to one other with their plus end toward the tip of the filopodium. By contrast, actin filaments in lamellipodia form a dense actin meshwork with filaments having no particular orientation (orange labelled: minus end). These arrangements of actin filaments provide solid anchors for proteins used for clustering cell surface receptors, such as the integrins receptor family, used for growth cone-ECM interactions. A similar component closely associated with actin filaments are microtubules. Microtubules form the rigid structure of axons (or dendrites) and invade the base of the lamellipodia. They are principally used for the transport of vesicles from the cell body to the growth cone. One microtubule is composed of 13 protofilaments associated side by side, forming a hollow tube. Like actin filaments, microtubules polymerise and depolymerise dynamically in the growth cone. 5 r both the combination and the individual expression of Slit, Sema-3A and Sema-3C in vitro and in vivo via the activation of the CXCR4 receptor (Chalasani et al., 2003). Also, it has been shown in vitro that the attractive factors BDNF and Netrin-1 are converted into repulsive cues by the inhibition of cAMP or PKA in spinal neurites (Song et al., 1997). Equally, the level of cGMP is able to modulate both repulsive and attractive activities of Sema-3A (Song et al., 1998; Polleux et al, 2000). Growth Cone Guidance In general terms, there are two classes of guidance cues that direct neurite outgrowth, those that attract growth cones and those that repel growth cones. Both classes are divided into two subclasses depending on whether cues are secreted or not secreted. Attractive cues are characterized by their ability to attract the growth cone towards them and often facilitate axon elongation. An example of an attractive guidance cue is the grasshopper transmembrane protein semaphorin-la (gSema-la; Wong et al., 1997; Wong et al., 1999). Repulsive cues, including the secreted ephrins and the cell surface associated EphB receptor (Birgbauer et al. 2001), have primarily been described to inhibit growth cones extending into inappropriate regions. However, the classification of a family of guidance molecules is not as simple, since some families contain both attractive and repulsive molecules. In addition, some molecules, such as netrins, have both functions (chemorepellent and chemoattractant), depending on the type of neuronal surface receptors expressed (Colamarino & Tessier-Lavigne, 1995; Lim and Wadsworth, 2002). In addition to the nature of the guidance molecules themselves, a second factor contributes to the direction of growth extension. In fact, the way in which these instructive cues are presented to the growth cone has been shown to interfere with growth cone steering. Several years ago, two major theoretical models were proposed to describe how a growth cone is expected to respond to an instructive cue distributed in a gradient (Gierer 1987; Walter 1990). These models differ in how gradients are analysed by growth cones, whether there is an amplification of the small percentage change in external concentration across the breadth of the growth cone or the growth cone shifts its internal baseline to zero, reducing the effective concentration at one edge of the growth cone. In addition to theoretical models, several in vitro assays have tested growth cone responses in the presence of different types of gradients and molecules. A few years ago, it was demonstrated that the orientation of the gradient is crucial to mediate any response (Bagnard et al., 2000). More recently, Ming et al. (2002) tested whether 7 gradients of netrin-1 and (BDNF) had any effect on growth cone steering. They demonstrated that growth cones reorient the axon elongation towards the sources of the diffusible attractive molecules. Although many in vitro experiments demonstrated that growth cones do respond to gradients of positive (netrin) or negative (EphA) cues, very few in vivo experiments have corroborated the in vitro findings. The Semaphorin Protein Family The characterisation of the first semaphorin protein dates back to 1992. The work was done in the grasshopper organism and the protein was first named fasciclin IV (fas IV) (Kolodkin et al., 1992). At that time, fas IV had no homology with any other known proteins. Since then, a large number of sequences related to fas IV has been identified (figure 2A). Fas IV has been renamed semaphorin-la (gSema-la) (Kolodkin et al., 1993) and belongs to one of the largest protein guidance families, the semaphorins. Semaphorins are characterised by a highly conserved domain called the semaphorin domain, localized at the N-terminus, about 500 amino acids in length that includes 17 conserved cysteine residues (Raper, 2000). Members are distributed into nine classes based on their origin and structural features, since very little was known about their function. These proteins are well conserved during evolution and are largely distributed throughout organisms. Invertebrate semaphorins are found in classes 1, 2 and 5C, vertebrates members are found in classes 3 to 7 and the viral class has been identified as class V. The family includes secreted, transmembrane and membrane associated proteins (Kolodkin et al, 1993). The semaphorin protein family is currently under extensive investigation to elucidate their neuronal function and the signaling mechanism of each member, as most members have been shown to function as neuronal guidance cues (Kolodkin and Raper, 1998; Rabacchi et al, 1999; reviewed by Pasterkamp and Kolodkin, 2003). In addition, numerous studies in a wide variety of invertebrates and vertebrate organisms has shown that semaphorins are important for neurodevelopment (Wong et al, 1997; Isbister et al. 1999; Roy et al, 2000 ; Matthes et al, 1995; Yu et al, 1998; Yu et al, 2000; Tanigushi et al, 1997). Related sequences have been cloned in monkey and in human species (Semaphorin Nomenclature Committee, 1999). Moreover, some members, which act as guidance cues, have been shown to function in a variety of other developing systems. For example, in addition to being functionally active for proper skeletal development, class 3 semaphorin has been shown to have a crucial role in normal 8 vascular and cardiac formation (Behar et al, 1996). The class 7 semaphorin member has been implicated in regulation of the immune system (Xu et al, 1998) and Sema-4D plays a role in the aggregation of B-lymphocytes and in the proliferation T-lymphocytes (Hall et al, 1996). Beside developmental processes, other members have been reported to play a role in lung cancer progression (Furuyama et al, 1996) and in neuronal apoptosis (Shirvan et al, 1999) . Vertebrate Semaphorins Vertebrate semaphorin protein members are categorized into classes 3 through 7. Like class 1 invertebrate semaphorins, members of classes 4, 5 and 6 are transmembrane, with class 6 members being most similar to class 1 semaphorins. Mainly due to the presence of 7 trombospondin type-1-repeats, class 5 proteins members have the largest amino acid sequence of the semaphorin family consisting of over 1.000 amino acids, while other semaphorins are typically 750 in length. One class includes secreted forms, classes 3. Class 3 semaphorins are the vertebrate homologues of the invertebrate class 2 members and their neuronal function has been well studied, particularly Sema-3A. Though initially thought to be an inhibitory cue based on in vitro studies, where Sema-3A was demonstrated to function as a powerful repulsive cue for dorsal roots ganglions (DRG) (Shepherd et al, 1997), Sema-3A was later shown to also act as an attractive cue for cortical apical dendrites (Tanelian, et al, 1997; Pasterkamp, et al, 1998; Polleux et al, 2000). In addition to its bi-functionality, Sema-3A knock out experiments are not always consistent with a repulsive function on axonal outgrowth in the CNS (Catalano et al, 1998). While there appears to be exceptions to its defined activity, it is well accepted that Sema-3 A is a repulsive cue that induces growth cone collapse. Like most of the other semaphorin members, Sema-7A protein has been reported to function as well on a variety of neuronal cell tissues, including olfactory epithelium and olfactory bulb, cortex and dorsal root ganglions, having a chemotrophic effect in vitro and that, independently from its known receptor, plexin-Cl (Pasterkamp and Kolodkin, 2002), suggesting that Sema-7A activity is mediated via more than one receptors. Similar to Sema-4D (also named CD 100), the Sema-7A has been reported to function in the immune system, as a stimulator of monocytes and neutrophils (Watanabe et al, 2001; Holmes et al, 2002). Sema-7A member is neither transmembrane nor secreted but is cell associated with a GPI-linked to the plasma membrane. Class 6 semaphorin proteins are transmembrane proteins and are the vertebrate homologues of the invertebrate class 1 semaphorins. Sema-6A protein expression in the embryonic chick and the corresponding in vitro collapsing assays suggest the protein to have a redundant function with the secreted 9 Figure 2 Semaphorin proteins and their receptors. A. Semaphorin Proteins: Protein members are classified into nine subclasses: three invertebrates (1,2, 5C), five vertebrates (3 to 7) and one viral (V): Subclasses differ according to structural features of the proteins, whether they are secreted, transmembrane, (GPI)-anchored or contain trombospondin repeats. Semaphorin members - la, 4B, C, F and 6D- are known to interact with proteins via their cytoplasmic domain, either for downstream signaling (dSema-la, Sema-4C and Sema-6D) or for structural features (Sema-4B and F). B. Semaphorin receptors: Invertebrate organisms contain two forms of plexin, plexinA and plexinB. PlexinA is the co-receptor with OTK for dSema-la, and dSema-lb signals via plexinA only. The ligand for plexinB has not been identified. In vertebrates, nine plexins are distributed into four classes, Al to A4, Bl to B3, Cl and Dl. Most of them have been demonstrated to interact directly with semaphorin protein members, with the exception of class 3 semaphorins, and are essential for mediating downstream signaling. Class 3 members interact with different affinity to neuropilin 1 (Sema-3 A) or to neuropilin 1 and 2 (Sema-3A, B, C, F) as neuropilins are the binding unit in a complex of co-receptors including plexins Al-2-3 (mediating the signal transduction) and LI (modulating plexins' signaling). 10 A Invertebrates Semaphorins Vertebrates Viral Classes Ka) 2 5c 3 4 J B . C & F ) 5 6(B) 7 SIGNAL PEPTIDE m GPI LINKAGE SEMA DOMAIN • BASIC DOMAIN Ig DOMAIN THROMBOSPODIN TYPE-1 R E P E A T S B Semaphorin receptors Invertebrates Vertebrates OTK L1 NP-1/2 Plexins A B A1/273 A4 B2 B1 B 3 C1 D1 Semaphorins Sema domain Atypical sequence CUB domains Immunoglobulin : Fibronccb'n 3 Cystein rich Coagulation-factor domain domain Sex-plexin domain ' Furin like proteolytic homology domains _ Tyrosine kinase site MAM (c) domain domain 11 Sema-3A protein on a subset of neurons (Xu et al, 2000). The Invertebrate Semaphorins The literature describes three semaphorin classes in invertebrate organisms. Classes 1 and 5C are restricted to transmembrane members and class 2 is a secreted member. Two members of class 1, Sema-la and Sema-lb have been functionally described in Drosophila melanogaster and in C. elegans, but only one member has been functionally described in Schistocerca gregaria/americana (grasshoppers) (Kolodkin et al, 1992; Yu et al, 1997; Ginzburg et al, 2002). Only one member of the second class have been identified, dSema-2a (Kolodkin et al, 1993; Matthes et al, 1995), CeSema -2a (Roy et al, 2000) and gSema-2a (Isbister et al, 1999). The last class has been named Sema-5C, mainly because of its structural similarity with the vertebrate class 5 semaphorins, which is characterized by the presence of seven trombospondin type-1-like repeats. The current literature reports that Sema-5C subfamily is encoded by one gene cloned in Drosophila and named dSema-5Cl. However, this subfamily is composed of two protein members because the gene appears to be alternatively spliced, giving a protein with seven trombospondin type-1-like repeats and an additional protein having six trombospondin type-1-like repeats due to the fusion of repeats 2 and 3. In contrast to all other members that have been functionally described in each organism, dSema-5Cl mutants failed to show any abnormal phenotype. Consequently, its functional activity is still unknown (Bahri et a/., 2001). . Structure of the Grasshopper Sema-2a As mentioned above, Sema-2a encodes a secreted semaphorin protein member in invertebrates. In grasshoppers, the gSema-2a sequence is primarily characterized by the conserved semaphorin domain at its N-terminus, extending from amino acid 36 to 514. The sequence also contains an immunoglobulin (Ig) domain of 71 amino acids in length (see Appendix 1 for the complete sequence of gSema-2a). A notable distinction between all invertebrate Sema-2a proteins and their vertebrate Sema-3 protein homologues is that vertebrate members have a basic domain at their C-terminus which is not encoded in the Sema-2a sequences. 12 Invertebrate Homologues of the Grasshopper Sema-2a Previously, dSema-2a was described to be dynamically expressed by a subset of neurons in the CNS and transiently by muscle cells in the periphery at the time of neuromuscular junction formation (Kolodkin et al., 1993). Also, genetic studies focusing on dSema-2a muscle expression demonstrated that dSema-2a functions as a repulsive signal for some neurons during synaptogenesis. From this analysis, it was determined that neurons responding to dSema-2a proceed by a delayed growth cone collapse, and axon retraction occurs only after contacting the wrong muscle which expressed dSema-2a (Matthes et al, 1995). This role of dSema-2a in the formation of the neuromuscular junction was confirmed by others (see Culotti and Kolodkin, 1996; Winberg et al., 1998). These studies demonstrate that dSema-2a protein is involved in neuronal targeting by being a repellent to some axons. By contrast to dSema-2a, its homologue CeSema-2a appears not to function as a neuronal guidance cue. Null mutants show little in the ways of developmental defects in the CNS. In fact, the protein appears to be largely involved in the regulation of cell shape changes and cell associations (Roy et al, 2000). Together, these studies on Drosophila and C. elegans illustrate the wide range of functions of Sema-2a. Plexins and Neuropilins: the Semaphorin Receptors Semaphorin proteins are found to bind with high affinity to at least two distinct receptor families: the plexins and the neuropilins (figure 2B). Plexins are themselves related to the semaphorin protein family, as they contain a related semaphorin domain sharing 15% to 20% identity with the semaphorin domain described in the semaphorin proteins. This domain is located in their extracellular fragment and appears to function as a signaling inhibitor in the absence of ligand, as shown for plexin-Al (Takahashi et al., 2001), and possibly as a mediator of interactions between the co-receptors neuropilins and plexin (Tamagnone et al., 1999). Also, their sequence among family members is well conserved with their cytoplasmic domain showing from 57% to 97% amino acid similarity and shares more than 50% similarity between invertebrate and vertebrate members (Tamagnone et al., 1999). The literature reports two invertebrate members, plexinA (PLX-A) and plexinB (PLX-B), both cloned in Drosophila. PLX-A has been characterized as the functional receptor for Sema-la and Sema-lb, but PLX-B appears not to function as the receptor for Sema-2a (Winberg et al., 1998). In C. elegans, PLX-A has been cloned and it has also been demonstrated to be a functional receptor for Sema-la and Sema-lb (Fuji et al., 2002). In vertebrates, 9 different 13 genes of plexins have been cloned and they have been grouped into 4 subfamilies according to their structure: Al to A4, Bl to B3, Cl and Dl (Tamagnone et al, 1999). Like their invertebrate counterparts, some vertebrate plexins have been functionally described to be essential and sufficient to transduce semaphorin signaling (Tamagnone et al, 1999). Indeed, experiments from different laboratories have demonstrated that plexin-B 1 is the receptor for Sema-4D, and plexin-Cl is the receptor for the semaphorins Sema-7a and for the two viral forms of semaphorin (Comeau et al, 1998; Tamagnone et al., 1999; Takahashi, et al., 1999). It is important to point out that there is at least one exception to the high affinity binding between plexins and semaphorins, as the receptor Al and A2 function as co-receptors for Sema-3A, -3B, -3C and -3F and do not interact directly with semaphorin proteins. In this case, plexins only serve as a mediator for semaphorin signaling (Tamagnone et al., 1999). Plexins do not have any intrinsic kinase activity (Maestrini et al, 1996) or other intracellular functional motifs (Artigiani et al, 1999). Thus, plexins signal through their ability to bind other kinases such as members from the Rho family of small GTPases. In this respect, a few members of this family, Racl, RhoA, and RhoD have been demonstrated to be important for growth cone steering events when vertebrate plexins are activated (Kuhn et al, 1999; Vikis et al, 2000; Swiercz et al, 2002; Zanata et al, 2002). Similarly, even though the ligand of the invertebrate PLX-B receptor has not been identified, PLX-B signaling is dependant upon Rac and Rho to mediate axon guidance (Driessens et al, 2001; Hu et al, 2001). On the other hand, there is no evidence of any interaction between PLX-A and the Rho GTPases members. However, the tyrosine kinase Off-Track (OTK) has been shown to directly associates with PLX-A to mediate the signal transduction of dSema-la (Winberg et al, 2001). Vertebrate organisms, including mammals and birds, have two semaphorin co-receptors, named neuropilin-1 (NP-1) and neuropilin-2 (NP-2) that bind specifically to different members of the class 3 semaphorin (for example Sema-3A binds only to NP-1; Chen et al, 1997). Thus, for class 3 semaphorin signaling, plexin-Al/A2 do not bind to the semaphorin ligand, but are rather only used for signal transduction (He et al, 1997; Kolodkin et al, 1997; Takahashi et al, 1998; Tamagnone et al, 1999), while the binding units are either NP-1 or NP-2. Both neuropilins have a very short cytoplasmic domain which is unused for semaphorin signaling (Nakamura et al, 1998). Thus, NP binding with plexins is mandatory for the mediation of any intracellular signaling cascade. Moreover, it has been shown that neuropilin do not show any activity by themselves, but they rather associate via their Complement-binding protein homology 14 (CUB) domain to the semaphorin domain of plexin-Al prior to Sema-3 binding (Takashi et al. 1999; Tamagnone et al. 1999). The two neuropilin members share a high degree of sequence and structural similarity with each other, their extracellular portions containing three different motifs: CUB, coagulation factor and MAM domains. Briefly, a CUB domain has been defined by approximately 110 residues forming an antiparallel P-barrel structure. Such domain is found in proteins involved in developmental processes, such as BMP-1 (Bork and Beckmann, 1993). It is suggested that CUB domain function in protein-protein or protein-carbonhydrate interaction (Li and Eriksson, 2003). A MAM domain has been defined by its content in four conserved cysteines and hydrophobic and aromatic amino acids residues. For example, the MAM domain in meprins has been suggested to have a variety of functions: meprins oligomerization, protein folding as well as substrate recognition (Marchand et al., 1994; Yokozawa et ah, 2002). Figure 3 illustrates the complexity of plexin and neuropilin receptors signaling with most of the identified downstream effectors. Also, an additional protein, LI, is illustrated due to its importance in binding NP-1 and to its ability to reverse the response of cortical axons to Sema-3A (Castellani et al, 2000). Finally, a group of proteins, MICALs, has been shown to directly interact with some of the plexins. MICALs are known for their oxidative function on a variety of proteins, including F and G-actin, and hence their ability to rearrange the cytoskeleton (Terman et al, 2002). Embryonic Grasshopper as a Model Organism In order to investigate how guidance molecules function to steer growth cones, having an in vivo system that is suitable for direct observation and simple analysis of growth cone responses to guidance cues as axons elongate is of great interest. One suitable model to conduct in vivo experiments is the embryonic grasshopper. Embryogenesis of the Grasshopper There are several different species of grasshopper, the one used for the following developmental studies is Schistocerca gregaria. Grasshoppers are members of the order orthoptera. New born insects of this order are characterized by an incomplete metamorphosis, they are reduced in size and there wings are missing. Grasshopper embryogenesis is divided in stages based on percentage of embryonic development. Early in its development, the grasshopper embryo is divided into three main segments: the cephalic segment, divided into 15 three segmental ganglions SI, S2 and S3, controlling principally the mouth; second is the thoracic segment, divided into 3 thoracic ganglions (Tl, T2 and T3) to which three pair of legs and one pair of wings for the male or two pairs for the female, are attached. Finally, the abdominal segment is divided into 11 abdominal ganglions Al to Al 1, containing reproductive organs (Goodman, C.S. et al, 1983). As seen in other animals, their PNS and CNS develop from the ectoderm. However, these organisms are distinct from vertebrates in that their sensory neuronal cell bodies emerge from the ectoderm in the periphery and extend axons toward the CNS. The signalling events underlying the generation of these neurons is unknown, however it is speculated that a similar pathway to the Notch-Delta pathway observed in Drosophila plays a similar role. Relatively early in the grasshopper embryonic development (approximately 30.0%), the Til pioneer neurons are the first neurons to undergo axonogenesis in the limb bud. The Til neurons are a pair of neurons that differentiate in the anterior epithelium (from a single mother cell as the cell undergoes mitosis) in all three pairs of limbs (Tl, T2 and T3). From 30% to 34% of embryogenesis, the growth cones of the Til neurons establish a highly stereotyped pathway from the tip of the limb bud to the CNS and it seems that these neurons don't have a real final target in the CNS (figure 4) (Bentley and O'Connor, 1992). Other axons (sensory and motor) that arise later in embryonic development fasciculate along the Til pioneer axons (this is referred to an axonal contact guidance; Keshishian and Bentley, 1983). When the formation of the peripheral nerves in limbs is completed (between 55% and 60% of the organism development), the Til pioneer neurons undergo apoptosis (Bentley and O'Connor, 1992). The mechanism underlying their death is unknown, as there is little information on the targets of the Til neurons and whether they form any synapses. The Til neurons appear to function solely as guidepost cells, similar to the guidepost neurons in the developing cat cortex (Mc Connell et al 1989). During Til outgrowth, the limb bud is similar to an empty tube filled with mesoderm cells loosely attached to the basal lamina covering the epithelium. After the establishment of the pioneer pathway, the mesoderm cells start to differentiate into muscle cells (Ho et al, 1983) and limb segmentation becomes more pronounced. Eventually, the full complement of peripheral nerves and muscles are formed. 16 Figure 3 Signaling of the plexin receptors A. OTK is essential for mediating downstream signaling(s) from plexinA upon dSema-la binding. The removal of the OTK C-terminus abolishes further downstream responses from activated plexinA. PlexinB and plexin-Bl, 2, 3 bind directly to activated Rac, sequester it and prevent the activation of PAK. When PAK is phosphorylated by Rac, the actin cytoskeleton is stabilized and growth cone elongation is promoted. In parallel, plexinB binds to RhoA, leading to the phosphorylation of cofilin by LIM K. B. Plexin-Bl, 2, 3 bind to RhoGEF, leading to a downstream protein cascade inducing growth cone collapse via RhoA. Fes protein is the first kinase to be identified to bind and to activate a member of the plexin receptor family. When plexin-A1 is phosphorylated, its interaction with Rndl is abolished and RhoD binds to plexinAl at the fragment sequence left by Rndl, activating a downstream signaling cascade leading to growth cone collapse. Plexin-A2 indirectly activates Rac leading to the phosphorylation of cofilin. MICAL proteins interact directly with the invertebrate plexinA and vertebrate plexins-A1/2/3 (A, B). No information is available whether gene transcription, protein synthesis or protein degradation is modulated by the activation of these receptors. 17 A 18 Guidance Cues in the Developing Limb Bud During neuroembryogenesis, growth cones of the Til pioneer neurons are guided to their target along a stereotyped projection due to interactions with a combination of guidance cues. Due to the easy accessibility of the nervous system and the simple strategies employed to modify its molecular and cellular content, the nature and the function of neuronal guidance cues in the grasshopper nervous system have been investigated. In this respect, some of the guidance cues affecting the steering events of the Til pioneer growth cones have been functionally described. Guidepost cells are defined as immature neuronal cell bodies and were the first Til pioneer guidance cues to be identified (Bate, 1976). A few of them are found along the Til pathway, acting as intermediate choice points for growth cones as they migrate toward their distant target. Along the way to the CNS, the Til growth cones contact three of them: Fe -found in the mid femur-, Trl -found in the trochanter-, where growth cones turn ventrally and the Cxi cells -found in the coxa. Cxi cells are considered to function as guidance cues for the growing Til pioneer neurons because when they are ablated, the Til axons are either arrested in the proximal side of the trochanter or growth cones take abnormal routes to fasciculate with axons from efferent neurons (Bentley and Caudy, 1983; Bentley and O'Connor, 1992). Hence it is thought that guidepost cells act as discontinuous guidance cues for the Til growth cones. A second guidance cue found in the developing limb bud is laminin. In other organisms, this protein has been shown to promote axonal extension and has a prominent role in nerve regeneration (Kato et al, 2002; Wallquist et al, 2002). In the developing grasshopper limb bud, laminin is deposited by migrating mesodermal cells and it is thought that laminin modulates the response of Til growth cones to other guidance cues (Bonner et al, 2002; Bonner and O'Connor, 2001). It has been proposed that the Til neurons proceed along their stereotyped ventral turn within the trochanter only if the laminin protein is accessible to the growth cones. The authors interpreted these observations as laminin being essential to process other guidance cues, particularly Sema-2a (Bonner and O'Connor, 2001). Additional molecular guidance cues described in the developing grasshopper limb bud include two semaphorin protein members; Sema-la and Sema-2a. Between 30% and 34% of development, Sema-la is expressed by epithelial cells and is distributed as a complete circumferential band in the tibia and a second band in the trochanter as well as an incomplete band in the mid femur (Kolodkin et al, 1992). Previous work has shown an attractive function of gSema-la for the Til neurons and, later in development, for the subgenual organ (SGO) neurons (Wong et al, 1999; Wong et al, 1997). The presence of Sema-19 2a in the developing grasshopper limb bud has been investigated by Isbister et al. (1999). Its distribution is in the form of gradients, from dorsal to ventral and distal to proximal with the limb. They showed that the spatial and temporal distribution of gSema-2a correlated with important steering events of the Til pioneer neurons. Also, based on Sema-2a blocking experiments, they showed that the gSema-2a gradients acted as a repellent for the Til growth cones (Isbister et ah, 2003). This is consistent with the role of Sema-2a in Drosophila during the establishment of neuromuscular junctions. Advantages of the Embryonic Grasshopper as a Model System Here are listed several arguments in support of the embryonic grasshopper model for functional studies of the secreted semaphorin-2a. The developing grasshopper embryo is large enough for dissecting its developing limb using a dissecting microscope. Also, mesoderm cells covering the neurons are fully accessible and easily removed, giving direct access to the Til pioneer neurons. Thus, these neurons can be exposed to any exogenous source of proteins. Secondly, the pathway followed by these neurons is well characterised and stereotyped, which makes it possible to analyse any steering defect as growth cones pathfind in situ (Wong et al., 1999; Isbister et al., 1999). Also, the embryos are amenable to culturing and because neurons have to navigate over a relatively short distance to reach their CNS target, a culture period of 24 hours is sufficient. At this early stage of neuronal development, only the Til neurons are pathfinding, providing a simplified in vivo model for growth cone steering. This simplicity may also reflect less redundancy in the guidance cues present, resulting in an easier functional analysis. Redundancy of cues in vertebrates may confuse the interpretation of gene knock out experiments (Tanigushi et al, 1997; Catalano et al, 1998). Objectives It has been clearly established that developing neurons respond to both attractive and repulsive guidance cues. In order to have a better understanding of how growth cones are guided during development, it is appropriate to determine whether there are conserved protein sequences associated with repulsion and attraction. Also, it will be useful to know whether secreted proteins can be engineered as transmembrane proteins and still maintain their function. Therefore the objectives of my work are to present the first functional 20 Figure 4: Schematic of the developing grasshopper limb bud at 34% of development. The sibling Til pioneer neurons are the first two neurons to pathfind in the limb bud. They differentiate from the same epithelial cell at approximately 30% of development. The axons of these two neurons fasciculate all along their pathway. They extend proximally along the limb axis until they enter the trochanter, where they contact the Trl guide post cell (33%), which is located in the middle of the trochanter and their axon elongation is reoriented ventrally. At the level of the Cxi cells located in the coxa, the growth cones reorient proximally to contact them (34%) and then neurons elongate into the CNS at 35%. Dashed lines indicate segment boundary and guide post cells are labelled and underlined in the limb bud. Note that the same orientation of the limb is maintained for all subsequent figures. 21 2 2 analysis of gSema-2a and to determine whether the semaphorin domain of semaphorin-2a is sufficient to provide the activity of the protein. I hypothesize that recombinant gSema-2a functions as a repulsive guidance cue and that the semaphorin domain is sufficient for this activity. In order to test whether the semaphorin domain is sufficient for function, a chimera protein was engineered from gSema-la and gSema-2a, resulting in a transmembrane protein that contained the gSema-2a semaphorin domain. This construct will allow me to investigate whether the semaphorin domain can function in the absence of the Ig domain normally found on gSema-2a. Based on other studies on secreted semaphorins (class 3 semaphorin), I expect the chimera protein to have the same potential to disrupt the normal navigation of the Til neurons. 23 METHODS Protein Constructs Sema-la/2a Chimera Overview The novel chimera protein consists of the 5' semaphorin region from the gSema-2a replacing the 5' conserved semaphorin region of the gSema-la. The Ig domain of gSema-2a was not included in the constructs of the chimera. A schematic representation of the chimera protein is shown in figure 5 C and its sequence is shown in Appendix 3. The chimera was engineered using the gene splicing by overlap extension method (gene SOEing), described by Horton (1990). The chimera was constructed by generating a gSema-la and gSema-2a PCR product that included the transmembrane domain and the semaphorin domain respectively. The 3'end of the gSema-2a DNA shared the same sequence with the 5'end of the gSema-la DNA. The overlapping sequence was then used to act as primers for a second round of PCR generating a gSema-2a/la chimera (see schematic in appendix 4). These first PCR steps described above were performed by L. Loy. Amplification of gSema-la and gSema-2a fragments The gSema-2a DNA fragment corresponded to a 1.6 kilo base pairs (kb) DNA region (from 5'end of the cDNA sequence to the 5'end of the Ig domain). The forward primer was 5'-CGG CGT CAT GCC TAG AT-3' and the reverse primer was 5'-TCC CAT CCG CAG TAA GG-3'. In order to amplify the fragment delimited by these primers, the PCR program consisted of 2 minutes denaturation at 95 °C; 30 cycles: denaturation 2 minutes at 95 °C, annealing 2 minutes at 47 °C, elongation 4 minutes at 72 °C and ends with a final elongation at 72 °C 10 minutes. The gSema-1 DNA fragment corresponded to a 0.9 kb DNA region (from the transmembrane domain to the 3' end of the coding sequence). The forward primer is 5'-GAT CCT TAC TGC GGA TGG G-3' and the reverse primer was 5'-AAT TAA CCC TCA CTA AAG-3'. In order to amplify the fragment delimited by these primers, the PCR program consisted of 2 minutes denaturation at 95 °C; 30 cycles: denaturation 2 minutes at 95 °C, annealing 2 minutes at 43 °C, elongation 4 minutes at 72 °C and ends with a final elongation at 72 °C 10 minutes. Amplification using the same reaction composition mixture was followed: Pfu DNA polymerase 10X buffer (containing MgS04), 0.2 mM final 24 concentration of dNTPs mixture (dATP, dCTP, dGTP and dTTP), forward and reverse primers at a final concentration of 10 pmol/pl, 5 ng of DNA template (vector AK195 or AK74), 1.4 unit of Pfu DNA polymerase and water is added to fill the volume to 50 pi. Gene splicing by overlap extension Following the amplification of both fragments separately, a third PCR was performed using gene splicing by overlap extension (Horton, 1990) in a final volume of 100 pi. The sequence for the forward primer was 5'-CGG CGT CAT GCC TAG AT-3' and the one for the reverse primer was 5'-AAT TAA CCC TCA CTA AAG-3'. The PCR program conditions were 2 minutes denaturation at 95 °C; 25 cycles: denatiiration 2 minutes at 95 °C, annealing 2 minutes at 43 °C, elongation 4 minutes at 72 °C and a final elongation at 72 °C 10 minutes. Composition of the PCR reaction mixture was Pfu 10X buffer (containing MgCk), 0.2 mM final concentration of dNTPs mix mixture, forward and reverse primers at a final concentration of 10 pmol/pl each and 1.4 unit of Pfu DNA polymerase Hot Start is added. The amount of Sema-1 and Sema-2 products were maximised and kept in a 1:1 ratio. The last PCR reaction was in a 50 pi final volume reaction with 5 ng of DNA obtained from the SOEing step, 5 pi of both forward and reverse primers described in the first step (5'-CGG CGT CAT GCC TAG AT-3'; 5'-AAT TAA CCC TCA CTA AAG-3') were used at 10 pmol/pl, 1 pi dNTPs (10 mM each), 10X buffer and 1.4 unit of Pfu DNA polymerase. The volume is filled to 50 pi with dH20. A PCR product containing the full length chimera was amplified with the correct size of 2.500 kb. Additional Construct A second construct was used as a control construct. It consists of the C-terminus of gSema-la. This control is called truncated gSema-la (t-Sema-la) (figure 5D). The control DNA fragment was constructed as followed. The forward primer was 5'-GCC CCG AAT TCA ATG GAC TGG AGT GCT-3' and the reverse primer was 5'-CCC CGC CGA ATT CGA TTT CTG CCG AA TGT-3'. EcoRl restriction sites were added (bold characters). The existing stop codon was removed and the DNA gene is in frame with a tag in the expression vector. The PCR program used for its amplification was 1 minute denaturation at 95 °C followed by 30 cycles: denaturation: 2 minutes at 95 °C, annealing 2 minutes at 58 °C, 25 elongation 4 minutes at 72 °C and ends with a final elongation at 72 °C 10 minutes. The composition of the PCR reaction mixture was 20 pmol of both primers, 0.2 mM of dNTPs mixture, 5 ng of DNA template, 10X buffer, 1.4 unit of Pfu DNA polymerase and dFfiO is added to complete the volume to 50 ul. Expression System The vector used to express gSema-2a, chimera and t-Sema-la proteins is the biscistronic pIZT vector (Invitrogen). In this vector, the expression of the inserted gene is under the control of OpIE2 promoter (from a baculovirus). pIZT also have the zeocin antibiotic resistance and the green fluorescence protein (GFP) genes which are under the control of a separate promoter, OpIEl (also from a baculovirus; see appendix 5 for the map of the pIZT vector). According to the literature (Invitrogen), the GFP promoter, OpEl, is five to ten times weaker than OpE2 promoter. Therefore, after the selection of transfected cells, some cells are negative for the expression of GFP, but do express the protein of interest, and it is assumed that cells expressing the GFP protein express the protein of interest as well. gSema-2a cDNA has been cloned between EcoRV and Xbal restriction sites, the chimera cDNA has been cloned between Not-1 and EcoRl restriction sites and t-Sema-la cDNA has been cloned at the EcoRl restriction site with its fusion at the 5'end with the V5 epitope found on the pIZT vector. All of these DNA constructs were fully sequenced on campus of the University of British Columbia by the Chain-terminator method at the Nucleic Acid and Protein Services (NAPS) unit confirming their correct orientation and sequence. The t-Sema-la construct was inserted in frame with a 3' tag (V5). Cell Transfection S2 cells were transfected in a 35 mm dish when cells were 70% confluent and their viability was above 95%. Cells were washed twice and resuspended in antibiotics and FBS HI free media. 3 ul of the lipid-form reagent CellFectin (Invitrogen) was diluted in 50 ul of media (antibiotics free) and incubated for 2 minutes. The reagents were pre-incubated for 30 minutes with 100 ng of DNA (also diluted in 50 ul of media antibiotic free). The DNA-CellFectin mixture was overlaid on the cells overnight. Then the CellFectin containing 26 F i g u r e 5 Schematic representation of the protein constructs A. Full length gSemaphorin-2a B. Full length gSemaphorin-la C. Chimera construct The chimera protein is made with the semaphorin domain of gSemaphorin-2a, circled in (A) and the C-terminus of gSema-la, circled in (B). D. Control construct (t-Sema-la). The control construct is made with the C-terminus of gSema-la, circled in (B). 27 A gSemaphorin-2a protein G B gSemaphorin-1 a protein Semaphorin domain Ig domain Semaphorin domain Transmembrane and intracellular domains C Chimera protein construct D Control construct: t-Sema-la media was replaced with fresh media containing antibiotics and 10% FBS HI. To confirm that the transfection has been successful, the cells were directly observed under fluorescence for GFP expression 48 hours after transfection. Following these conditions, the efficiency of transfection reached 35%. Cells were grown for 5 days and then selected with the zeocin antibiotic at a final concentration of 800 ng/ml for 7 days, after which the GFP expression reached 99%. The selection was maintained by the addition of zeocin at lOOng/ml to the media every week. Antibody Work Purification of A and E Antigenic Fusion Proteins A and E antigenic fusion proteins were previously used to produce A and E polyclonal antibodies. These antibodies were previously characterized and used to map gSema-2a protein expression in the embryonic grasshopper (Isbister et al., 1999). For my experiments involving the immunostaining of invertebrate cells expressing gSema-2a and the chimera proteins, additional immunopurification of the antibodies was necessary. A antigenic peptide is a 120 amino acids fragment from the N-terminus of gSema-2a and E is a 234 amino acids fragment located in the middle of the semaphorin domain in gSema-2a protein (amino acids 141 to 375). A antigenic peptide is fused to a 6-histidine tag at its 5'. The vector pQE 30 (Qiagen) containing the fusion protein was transformed into E. coli XLl-Blue competent cells (Stratagene). 500 ml of SOC media was inoculated with 16 mL of bacteria (1/30 of the final volume). When the bacteria growth reached the log phase (O.D. 600nm = 0.6 to 0.8), the expression of fusion proteins was induced by the addition of IPTG at a final concentration of 2mM for 6 hours. The clarified protein sample was added to beads coupled to cobalt cations (TALON ™ metal Affinity resins from Clontech). For 500 ml of bacteria culture, 500 ul of TALON Resin sample is used. The fusion protein was purified following protocols from the manufacturer and samples were collected in fractions of 500 ul aliquot. The E antigenic fusion protein was fused at its 5' to GST. The vector pQEX-4Tcontaining the fusion protein was transformed in E. coli DH5ct competent cells (Invitrogen). 500 ml of LB culture media was inoculated with 20 ml of bacteria (1/25 of the 29 final volume). When the bacteria reached the log phase (O.D. 600nm = 0.6 to 0.8), the fusion protein was induced by the addition of JJPTG at a final concentration of 1 mM for 6 hours. The culture was then harvested and the pellet was resuspended in 18 ml of chilled MT-PBS. Bacteria samples were sonicated and TritionX-100 was added to make a 1% final concentration. The sample was rocked for 5 minutes at 4 °C. The membrane extracts and other insoluble materials were pelleted during a centrifugation. After verification using Commassie blue staining on each fraction (pellet and supernatant), the fusion protein was confirmed to be in the supernatant and was therefore processed with a GST affinity column. The supernatant was added to the beads and rocked for 15 minutes at room temperature. The beads were washed with MT-PBS-1% Triton-XlOO and final washes were with MT-PBS. The fusion protein was eluted with 8 ml of elution buffer (containing 10 mM of soluble glutathione; see appendix 6). GST Column Preparation For 500 ml of bacteria, 160 mg of GST-agarose bead (Sigma) was used. The beads were transferred to a 15 ml Falcon tube and the tube was filled with 15 ml of MT-PBS. The beads were left to incubate for 30 minutes at room temperature. Beads were washed 5 times in 15 ml MT-PBS and a final wash was made with 15 ml of MT-PBS -1% TritonX-100. Antibody Purification Purified fractions containing respectively A and E fusion proteins were pooled and dialysed overnight at 4 °C in pH 10 Coupling Buffer. Purified antigens were covalently immobilised to an AminoLink Plus Column (Pierce). Both antibodies were purified using the protocol suggested by the manufacturer. Antibodies continuously flowed through the gel for 4 hours at room temperature and the bound antibodies were eluted at low pH with 8 ml of glycine buffer (appendix 6) and collected in 500 pi aliquots. All fractions were monitored for the presence of protein with an assay based on the Bradford protein assay (the protocol and all reagents are from BioRad). Fractions collected within the peak of elution were pooled and dialysed in PBS overnight at 4 °C. Details on the composition of all solutions and buffer are listed in Appendix 6. 30 Immunostaining Non Fluorescence In order to visualise the Til neurons, embryos from 28% to 40% of embryonic development were fixed in 4% formaldehyde in PEM for 45 minutes and then washed 15 minutes with five changes in PBT. Embryos where then incubated overnight at 4 °C with a primary antibody against HRP (Jan and Jan 1982; Jackson Immunoresearch Lab) at a 1:1000 dilution and washed for one hour in PBT with ten changes. Neurons were visualised after a two hour incubation period with a secondary antibody (donkey a rabbit peroxidase conjugated) at a 1:250 dilution. The peroxidase was activated for one hour by the addition of 3,3'-Diaminobenzidine peroxidase substrate (DAB) (SigmaAldrich). In parallel, both immunopurified A and E antibodies described above were used to analyse gSema-2a expression. The same procedure used for neuronal labelling was used to map gSema-2a. All buffers composition is detailed in Appendix 6. Fluorescence Neuron Labelling The Til neurons were labelled with the primary antibody rabbit a HRP at a final concentration of 1:1000. The secondary antibody was a goat anti-rabbit Cy3-conjugated, and it was used at a final concentration of 1:1000. For intact embryos, the primary antibody was incubated overnight at 4 °C and the secondary was incubated for one hour at room temperature. Neurons in limb-fillet preparations were incubated with the primary antibody for two hours and then one hour with the secondary, both at room temperature. gSema-2a, Chimera and Control Construct Expression Immunostaining detection of s2 cells expressing gSema-2a and the chimera was done with both A and E primary antibodies used at 1:500 final dilution. A V5 primary antibody (Invitrogen) was used at a final dilution of 1:1000 to detect the V5 epitope fused to the c-terminus of the t-Sema-la construct. The A and E antibodies were used as negative controls against t-Sema-la expressing cells, and V5 was used as a negative control against chimera expressing cells. Cells were fixed for 30 minutes in 4% formaldehyde in PEM and then 31 washed for 30 minutes with 3 changes in PBT. Cells were incubated overnight at 4 °C with the primary antibody. Cells were then washed for one hour in PBT with ten changes. Secondary antibodies were used at 1:1000 final dilution and incubated for one hour at room temperature. Cells were washed for one hour with PBT with 10 changes. Last, cells were resuspended in Slowfade/Antifade Component and observed with an upright fluorescent microscope. Images were taken with a digital camera using the computer program Meta View and figures were prepared in Adobe Photoshop. Immunoprecipitation Before performing immunoblots on cell culture media, an immunoprecipitation was performed to concentrate all proteins. First, S2 expressing cells were cultured and then 5 ml of their culture media was dialysed in PBS overnight at 4 °C and concentrated with PEG down to 1 ml. The final concentration of proteins was 500 pg/ml. A final concentration of 10 pg/ml of the primary antibody anti-Sema-2a (A) was rocked for one hour at 4 °C with 50 pi of protein G slurry (Invitrogen). 1 ml of culture media was added to ,4-protein G complexes and shaken overnight at 4 °C. Complexes were spun down at 13,200 rpm 10 minutes. The supernatant containing unbound proteins was discarded. Pellets were washed in PBS and centrifuged 4 times. 40 pi of 5x loading buffer was added to the pellets and the mixtures were boiled for 10 minutes, cooled on ice for 5 minutes and spun at maximum speed 5 minutes. The supernatant was loaded in a 12% polyacrylamide gel and immunoblotted. Immunoblot Immunoblots were performed on whole cell extracts and on cell culture media. When testing whole cell extracts, cells were first rinsed twice in PBS and then lysed in 100 pi of RIP A buffer for 30 minutes at 4 °C. Cells extracts were vortexed and centrifuged. Lysates were transferred to new tubes and mixed with 5X SDS-PAGE sample buffer. Samples were boiled for 5 to 10 minutes and cooled before they were loaded. Samples were run into a 12% or 15% acrylamide gel, then gels were taken out of their cast to transfer the proteins to a nitrocellulose membrane for 2 hours at room temperature at 50 mV. The membrane was rinsed twice for 1 minute in TBST and then incubated over night in TBST 32 containing 5% milk. The blocking step was followed by 3 washes for 1 minute each and 3 washes for 15 minutes each in TBST. The A2 antibody was diluted at 1:500 and incubated at room temperature 2 hours in TBST containing 1% milk. Membranes were washed 3 times in TBST 1 for minute and 3 times for 15 minutes. Secondary antibodies were diluted 1: 2000 and incubated for one hour in TBST containing 1% milk. Membranes were then washed 3 times for 1 minute and 2 times for 15 minutes in TBST and one last wash for 15 minutes in TBS. Membranes were processed for ECL chemoluminescence as described by the manufacturer (Santa Cruz Biotechnology). The composition of all solutions and buffers are detailed in Appendix 6. Limb-fillets System Staging embryonic grasshoppers In order to identify the exact stage of development at the beginning of each experiment, four embryos were fixed at the beginning of the culture period. The non cultured embryos were called t=0. Their neurons were stained to determine with precision where the growth cones were at the start of the experiment. Limb-fillet procedure Embryonic grasshoppers were placed in a culture dish with the anterior side of the limbs facing down. The limb bud is opened along its longitudinal axis from the tip of the limb to the CNS. Each side of the cut is flattened down on a poly-L-lysine coated cover slip and mesoderm cells are removed by suction (O'Connor et al, 1990). In order to maintain the development of the nervous system for the following 24 hour period, limb-fillet preparations were cultured at 30 °C in RPMI nutrient medium. To examine the effect of the wild type Sema-2a and of the chimera proteins, clumped cells were placed directly on the limb-fillet. Following an approximately 24 hour period in culture, embryos were fixed, their neurons were labelled and the Til pathways were visualized for changes in growth in response to the presence of the transfected cells. Appendix 6 contains the complete composition of all buffers and media used for the dissection of embryos and their culture. 33 Analysis of Misguidance Frequency of guidance errors were calculated as the percentage of abnormal Til pathways observed in individual experiments. Error bars represent standard errors of the mean (SEM) from at least 8 experiments (see table 1) and n corresponds to the total number of limbs observed. Two tailed student t-test were used in order to assess whether the average errors induced by the addition of gSema-2a or chimera expressing cells differ from the average level of errors obtained when the neurons navigate in a limb-fillet with control cells or no cells. The t-test compares the mean value of each group and takes into account the variability into the groups (which are due to the number of experiments having different value). Differences were considered significant at p< 0.05. 34 RESULTS Antibody Characterisation In order to increase the specificity of the detection of semaphorin-2a and the chimera, two antibodies, previously named A and E (Isbister et al., 1999), were immunopurified. The A antigenic fusion peptide was from the N-terminus semaphorin domain of gSema-2a and E antigenic fusion peptide was from the middle section of the semaphorin domain of gSema-2a (see the section Material and Methods). Both antibodies are polyclonal and were raised in rabbits. Here, they were used to map the expression gSema-2a in the limb bud of the developing organism. Also, they were used to confirm the expression of recombinant gSema-2a and the chimera by S2 cells. The antigenic fusion peptides, A fused to 6 histidine tag and E fused to glutathione S-transferase (GST) tag, were isolated from crude bacteria cell extracts. In order to evaluate the purity of the fractions eluted, Coomassie blue staining was used on all eluted fractions. Figure 6 A, C shows that the protein content of the elution fractions is enriched with one size of protein (only one band appears on each gel). That indicates that the crude cell extracts have been well purified. In theory, the molecular mass of A antigenic peptide is 13 kD and E is 26 kD. The gel obtained with the A peptide fractions shows the expected candidate band at 13 kD. The candidate band obtained with the fractions from the E antigenic fusion peptide is approximately 52 kD. That molecular weight was also expected due to its fusion with the GST protein, which is 26 kD. In addition, Western blots have been performed on the same elution fractions to further test the purity of the elution fractions and to confirm the identity of both peptides isolated from the crude bacteria cell extracts. Figure 6 D lane 1 shows that bacterial proteins (other than antigenic peptides) are recognized by the E antibody before the purification. But after purification, the antigenic peptides are the only ones recognized by their antibody (figure 6 B and D lane 2). It follows that these elution fractions were used for the immunopurification of their respective antibody (see the section of material and methods for the procedure). 35 Figure 6 A and E antigenic fusion protein purification. A. Coomassie blue staining (15% acrylamide gel) of the fusion proteins. Purified ^ 4 fusion protein from whole bacteria extracts. B . Western blots (15% polyacrylamide gel). The A antibody recognizes the purified A fusion protein. C. Coomassie blue staining shows the purification of the E fusion protein. D . Western blots (12% acrylamide gel), the E antibody recognizes the purified E fusion protein, lane 1: whole bacteria cell extract. Lane 2. purified E fusion protein. Molecular weight markers are indicated on the left of each panel and arrow heads are pointing at the fusion protein. Numbers on the left indicate the molecular weight. 36 Dynamic Expression of gSema-2a during Embryonic Development Using the A and E antibodies, a previous study described the Sema-2a protein expression in the embryonic grasshopper limb bud (Isbister et al., 1999). However, the expression of Sema-2a is extremely dynamic and it was unclear how this pattern changed with respect to the position of the Til growth cones. Here I present a precise analysis of the temporal distribution of the gSema-2a protein in the developing grasshopper limb bud during the period of Til pioneer outgrowth. This analysis is particularly useful in designing and interpreting the transplantation experiments as it will allow me to identify the number of cues that the Til neurons contact before making a steering decision. The pair of Til pioneer neurons are the first neurons to differentiate in the PNS and to pathfind through the PNS to reach the CNS. They differentiate relatively early, approximately 30%, from one epithelium cell in the proximal tibia, and their growth cones navigate from the tip of the developing limb bud into the CNS along a stereotyped pathway (previously described by Bentley and Keshishian, 1982). As the Til neurons elaborate their pathway from the tip of the limb bud to the CNS (figure 8), gSema-2a protein shows a dynamic expression pattern (figure 7). The expression of gSema-2a in the limb bud has been detected with both A and E immunopurified antibodies, both giving a consistent staining for all embryonic stages, and the expression patterns obtained with A were indistinguishable to the ones obtained with E. The protein is expressed before the Til neurons differentiate from the epithelium (figure 7 A, B), during their pathfmding (figures 7 C-G and 8 A-D), and after they have reached the CNS (figure 7 G-H and 8 E-F). The earliest stage examined for gSema-2a expression was approximately 28% of development (figure 7 A). However, the first neurons to differentiate in the developing limb bud are the Til pioneer neurons, appearing at a later stage, approximately 30%. At 28%, gSema-2a is already expressed in an overlapping gradient from dorsal to ventral and from distal to proximal, having higher concentrations dorsal and distal. This pattern is maintained until approximately 30% of development (figure 7 B), corresponding to the differentiation of the Til neurons. At 30.5%, the axons of the Til pioneer neurons start elongating proximally, down the distal-proximal gSema-2a gradient (figures 7 B and 8 A). As the limb 38 develops from 31% to 33%, the Til pioneer neurons extend along the femur area and at 33.5%, the growth cones are oriented ventrally (figure 8 A-C). In parallel, the gSema-2a gradient expression pattern is prominent at the tip of limb bud and in the dorsal area from the tibia to the trochanter (Figure 7 C-F). Expression of gSema-2a protein is maintained at the tip of the limbs throughout the period examined. Between 31% and 32.5%, the general dorsal gradient expression changes into a single but large band, having its peak of expression dorsally (figure 7 D-F). As the Til axons extend and reach the trochanter (33.5%; figure 8 C), gSema-2a expression is maintained as a gradient in the trochanter (figure 7 F). Finally, at 34% of development, the wide dorsal expression narrows and forms two bands on each side of the trochanter area. During this period, the Til are at the PNS-CNS boundary (figure 8 D). As the neuronal complexity of the limb evolves with the emergence of new neurons, the Sema-2a expression changes with circumferential bands of expression found distally in all limb sections (tibia, femur and trochanter) (figure 7 H). Although gSema-2a protein is maintained in dorsal to ventral gradients within each circumferential band (figure G, Ff), there is a marked change in the expression pattern, even after the completion of the Til pioneer pathway (34%) (figures 7 G-H and 8 D-F). Interesting points to note are that the tip of the limb continually expressed gSema-2a during the entire time window considered here (28% to 38%). Protein Expression in S2 Cells Firstly, in contrast to the GFP nuclear expression, it is clear that all three proteins (gSema-2a, chimera and t-Sema-la) are expressed outside the nucleus (figures 9-11). Secondly, it is interesting to note that the resulting staining from gSema-2a, the chimera and t-Sema-la is not uniform but rather characterised by a punctate pattern. This particular pattern might be explained by the formation of protein complexes. This observation corroborates with previous findings that demonstrated that other semaphorin proteins function only while they form homodimers and their function is enforced by the formation of clusters (Koppel et al, 1997; Koppel and Raper, 1998; Klostermann et al, 1998; Wong et al, 1999). Although the immunostaining pattern is very similar for the three protein constructs, the overall intensity of the staining of gSema-2a expressing cells seems lighter than the ones obtained for the 39 chimera and t-Sema-la expressing cells. The resulting staining from gSema-2a expressing cells is potentially intracellular, detecting proteins while they are processed to the cell surface. Intracellular staining would be easily explained by having used detergent during the whole immunostaining procedure. On the other hand, the chimera and t-Sema-la are not secreted, hence all chimera and t-Sema-la proteins produced by the cells can be detected. Similarly, immunostaining of the full length gSema-la in a very similar expression system has been previously reported to be characterised by punctate staining at the surface of the cells (Wong et al, 1999). Last, it is important to point out that all three antibodies, A, E and V5 are specific to their target proteins, since in very few cells, it can be observed that GFP protein is expressed but the cells are not positively stained with A, E or V5 antibody. The opposite is also true and occurs more frequently: cells are positive for the expression of the protein of interest but do not express GFP protein. Figures 9, 10 and 11 illustrate the pattern of expression of gSema-2a, the chimera and t-Sema-la and were specially selected to show that some expressed the protein of interest but not GFP and vice versa. Western blots corroborated the results from the immunocytochemistry for the expression of gSema-2a and the chimera proteins. Immunoblot Expression of the chimera is detected from whole cell lysates by the presence of a predominant band approximately at 100 kD (figure 12 A, lane 1). As negative controls, cell extracts from gSema-2a (figure 12 A, lane 2), and t-Sema-la expressing cells (figure 12 A, lane 3) were tested. Due to the secreted nature of gSema-2a, not enough protein was present to generate a significant signal. Also, as the t-Sema-la construct does not contain the epitope recognised by the A antibody, the absence of signal from these expressing cells was also expected. The lower molecular weight band (-90 kD) was observed in each of the cell lines and most likely non-specific as Drosophila S2 cells do not express Semaphorin 2a as judged by negative PCR analysis (not shown). The detection of gSema-2a protein was only observed when the culture media from gSema-2a expression cells were assayed. For this experiment it was necessary to immuhdprecipitate gSema-2a prior to blotting in order to 40 Figure 7 gSema-2a protein expression is developmentally regulated in the embryonic grasshopper. Expression of gSema-2a using either A immunopurified antibody at a dilution 1:500. Expression patterns were consistent when E immunopurified antibody was used at a dilution 1:500. A. At 28% of development, gSema-2a is expressed throughout much of the limb bud. B. At 30% of development, gSema-2a is highly expressed in the dorsal and distal part of the limb. C, D and E . At 31.0%, 32%, 32.5% of development respectively, the dorsal-ventral gradient is evident while the expression distally is concentrated in the limb tip. F. At approximately 33% of development, the dorsal-ventral gradient expression spread further down ventrally within the trochanter. G . At 34% of development, the dorsal-ventral gradient is limited to bands of epithelium in the trochanter. H . At 38% of development, circumferential bands are evident in the distal part of each limb segment. gSema-2a remains expressed at the tip of the limb through out the developmental period considered. I. Grasshopper limb at 34% of development used as a negative control while staining with only the secondary antibody anti-rabbit peroxydase conjugated. Scale bars, A, B, C: 10 pm; D to I: 50 pm. Figure 8 Axonal elongation of the Til pioneer neurons. A. At 31% of development, the Til cell bodies have differentiated from the dorsal epithelium and axons are starting to elongate proximally. B. At 31.5% of development, the axons penetrate the femur. C. at 33.5% of development, the Til growth cones pathfmd in the trochanter. D. At 34% of development, the Til growth cones have contacted the Cx guide post cells. E and F. At 35% and 36%, respectively, the Til pathway is completed from the tip of the limb bud to the CNS. 41 Figure 7 42 Figure 9 gSema-2a and chimera protein expression in S2 cells. Immunofluorescence with the A anti-gSema-2a antibody (red in A) reveals that both gSema-2a and the chimera proteins are expressed (red in B ) . C . t-Sema-la expressing cells are used as negative control. GFP expression is from the same vector. Arrowheads indicate cells expressing the protein of interest without expressing appreciable GFP, and vice versa. Scale bar: 10 um. 43 A immuno- G F P A immuno- G F P A immuno- G F P staining expression staining expression staining expression A B C 4> «^ gSema-2a Chimera t-Sema-la 44 Figure 10 gSema-2a and chimera proteins detected with the E antibody. Immunofluorescence with E anti-gSema-2a antibody (red) corroborates the distribution and the expression of the proteins detected by the A primary antibody. A . gSema-2a expressing cells (red). B. Chimera expressing cells (red). C . t-Sema-la expressing cells are used as negative control. S2 cells express GFP from the same vector. Scale bar: 10 um. 45 £ immuno- GFP £ immuno- GFP £ immuno- GFP staining expression staining expression staining expression A B c 0 # . r . • 4 m t • gSema-2a Chimera t-Sema-la 46 Figure 11 t-Sema-la protein detected with the V5 antibody. Immunofluorescence with the primary antibody anti-V5 A. S2 cells expressing the fusion protein t-Sema-la-V5 tag (red). B . Chimera expressing cells are used as negative control and are not recognized by the anti-V5 antibody. S2 cells express GFP from the same vector. Scale bar: 10 um. 47 V 5 immuno- GFP V 5 immuno- GFP staining Expression staining expression A B • * y • ^ * t-Sema-la Chimera concentrate the gSema-2a signal (see Material and Methods). In these conditions, the presence of a predominant band at approximately 100 kD can be observed (figure 12 B, lane 1). The fact that the chimera and gSema-2a were detected at the same molecular weight was expected as both proteins are very similar in length, 743 and 697 amino acids respectively. Figure 12 B lane 2 contains a sample from the culture medium taken from the chimera expressing cells and was used as a negative control because the chimera is a transmembrane protein and was not expected to be found free in the culture media. Western blots indicate that when gSema-2a is secreted, the protein diffuses into the media. This is similar to Sema-3 protein family members as they are routinely harvested from media bathing transiently transfected mammalian cells (see experimental procedure from Luo et al., 1993). However, class 3 proteins contain a basic domain at their C-terminus that limits their diffusion and some of the proteins stay membrane associated (Luo et al., 1993; Bagnard et al. 1998). By contrast, gSema-2a does not contain any basic domain at its C-terminus, hence following its secretion it is more likely that all proteins produced by cells diffuse in the culture medium. Function of gSema-2a Protein and its Semaphorin Domain As a first step to identify the functional domains of gSema-la and gSema-2a, I swapped the semaphorin domains between the two proteins. This resulted in a transmembrane form of gSema-2a. Using this construct also allowed me to examine growth cone behaviours upon contact with the guidance cue. The strategy used was to introduce ectopic sources of chimera protein and compare the Til neurons steering responses obtained to those of Til neurons in the presence of ectopic sources of gSema-2a or control expressing cells. In order to assess the function of gSema-2a protein, the first analysis concentrated on the ability of gSema-2a and the chimera to disrupt growth cone pathfinding. The second part of the analysis focused on the types and the frequency of misguidance-errors in order to establish whether gSema-2a and the chimera function differently on steering growth cones. 49 Figure 12 Western analysis of chimera and gSema-2a expression. Western blot analysis using immunopurified anti-gSema-2a>4 antibody. A . Whole-cell lysate from expressing cells; Lane 1, chimera expressing cells; Lane 2, gSema-2a; Lane 3, t-Sema-la expressing cells. B. Immunoprecipitation of cell culture medium with the A antibody followed by Western blot with the A antibody. Lane 1, culture media from gSema-2a expressing cells; lane 2, culture media from chimera expressing cells. Molecular weight markers are on the left of each panel. 50 51 Function of gSema-2a Wild Type and the Chimera Control Experiments Figures 13 A, B, C and 14 A, B, C illustrate Til pioneer neurons in a limb-fillet after an overnight period in culture with control t-Sema-la expressing cells placed in the ventral epithelium along and in the normal Til pathway. Both examples show fasciculated Til pioneer neurons with their typical projections through the limb bud into the CNS, as previously described (figures 4 and 8). In figure 13 A the normal pathway is successfully completed, even if expressing cells (green) are embedded right along the axon elongation axis (figure 13 B, C). It appears that growth cone ignore their presence and navigate through the cells. In figure 14 A, fasciculated axons also extend proximally until they reach the trochanter where they proceed ventrally, navigating through t-Sema-la expressing cells. Although the Til pioneer axons do not reach the CNS, the navigation has proceeded normally and the Cxi cells are contacted by Til filopodia. Also, the morphology of the growth cones and their numerous filopodia suggest that they are actively pathfmding. In addition to control experiments with expressing cells, limb-fillets were cultured without the addition of any type of cells (not shown). In these preparations, the majority of the Til pioneer neurons navigate correctly. However, the average of eight experiments (50 limbs in total) reports approximately 16% of the filleted limbs exhibit misrouted axons (Figure 15). Such level of misguidance has been reported in the previous literature varying from 7.5% (intact limb) to 16% (in a limb-fillet) (Bonner and O'Connor 2001; Isbister et al., 1999; Wong et al., 1999). Figure 15 shows the rate of errors when t-Sema-la expressing cells are added to the limb-fillet. The histogram summarizes the results of 14 experiments, with a total of 96 limbs, of which 25 presented an abnormal Til pioneer pathway. The addition of S2 cells resulted in a higher rate of neuronal misguidance, 26%, compared to limb-fillets without S2 cells (16%) placed in the basal lamina. As previously reported (Wong et al., 1999), the presence of S2 cells can result in a higher misrouting effect on growth cone steering compared to limbs free of S2 cells. Although the average amount of defects in the Til pathway increases when the neurons navigate in the presence of t-Sema-la expressing cells, the difference is statistically not significant given that a two tailed t-test 52 gives ap value superior to 0.05. In contrast, the average of 11 experiments reports that approximately 79% of the limbs (98 limbs in total) presented defects in the stereotyped Til pathway when ectopic sources of gSema-2a were added (figure 15). The p value from the two tailed t-test is inferior to 0.0001 when the mean of aberrant Til pathway in the presence of gSema-2a are compared to the one obtained when t-Sema-la expressing cells are added to the limbs. Similarly, in the presence of chimera expressing cells, the average of 15 experiments reports that 84% of the limbs (129 limbs in total) observed showed an aberrant Til pathway (figure 15). The average error-induced by chimera expressing cells is also highly significant given that the t-test gives ap value inferior to 0.0001 when the average of misguidance is compared to the average of misguidance obtained when t-Sema-la expressing cells are added to the limbs. However, the statistics between the average error-induced by gSema-2a and by chimera expressing cells is not significant given that the p value from the two tailed t-test is superior to 0.05. Finally, the t-Sema-la transfected cells do not function on their own to steer neurons. For this reason, these cells are used as negative controls for gSema-2a and chimera expressing cells discussed below. Misguidance Phenotypes Even if the penetrance of misguidance triggered by gSema-2a and chimera expressing cells indicates that the function of a secreted guidance cue is not impaired by being transmembrane, a closer analysis investigating the types of abnormal phenotypes was considered to assess whether there was a difference of growth cone behaviours upon contacting a secreted versus transmembrane guidance cues. Four major aberrant Til neuron phenotypes were observed after ectopic expression of gSema-2a and the chimera. They are referred to as growth inhibition, dorsal projections, axon defasciculation and aberrant branching. Growth Inhibition In some cases, the Til neurons were found to not have extended their normal distance when cultured in the presence of gSema-2a and chimera expressing cells. Neurons falling in this category have typically extended an axon similar in length to embryos between 31.0% and 32.5% of development. Growth cones of such axons are also unusually narrow. In 53 order to rule out the possibility that growth impairment is due to damage caused by the limb-fillet preparation itself, axon lengths were compared between controls and experimental groups. With this precaution, I am confident that growth inhibition was due to the presence of gSema-2a or chimera expressing cells placed on the limb between the femur and the trochanter areas in the Til pathway axis (figures 13 D, F, G, I and 14 J, L). Typically, Ti 1 axons presenting this defect extend well proximally until growth cones reach the boundary separating the femur and the trochanter areas. At that point, both axons considerably reduced their diameter, growth cones narrowed and they had very few filopodia. Consequently, the axonal growth was considerably slowed down if not arrested. Equally, the chimera expressing cells were able to inhibit growth, figure 14 J, L. Similar to figure 13 D, F, the tip of the axon has stalled in the dorsal compartment of the trochanter. Usually, growth cones are marked by an increase in their breadth and in the number of filopodia extensions at this location (Bentley and O'Connor, 1992; Isbister and O'Connor, 1999). However, arrested growth cones show a smaller morphology than normal, although they still maintain a number of extended filopodia. Figure 13 G, I is the ultimate example to illustrate how powerful gSema-2a can inhibit growth. Firstly, it is interesting to note that the axons kept a certain distance from the expressing cells. Also growth cones appeared to have started a ventral turn but soon after they proceeded to a dorsal turn, away from the source of repulsive cues. Although, they successfully start elongating toward a more permissive zone, axon extension stopped soon after. Axons probably failed to fully elongate because the concentration of gSema-2a was high enough to rapidly perturb the stability of the whole growth cone. A comparison of growth inhibition shows that when no expressing cells are added, approximately 4% of limbs shows growth inhibition, which is similar to when t-Sema-la expressing cells are present, 7%. When gSema-2a or chimera expressing cells are added, axon growth is inhibited in 26% and 32% respectively of all cases of misguidance (figure 16 and table 1). These results indicate that both gSema-2a and the chimera can function as growth inhibitors. However, one major difference between growth inhibition triggered by gSema-2a and chimera proteins is that in the presence of gSema-2a, the morphology of the axons changed and growth cones were arrested at a distance from the cells, in contrast to growth cones that are inhibited by chimera expressing cells. This 54 Figure 13 Ectopic expression of gSema-2a perturbs Til growth cone steering. A. Til neurons pathfind correctly (red in C) B . Location of t-sema-la expressing cells (green in C). A and B are merged in C. D, G and J . Til neurons behaviour when gSema-2a are placed in the limb during Til steering events. Arrowheads indicate the Cx cells, localized ventrally and proximally in the limb bud. E, H and K . Location of the gSema-2a expressing cells, mostly in the ventral area of the trochanter. F, I and L. Neurons (shown in red) from D, G and J are merged respectively with E, H and K, showing the exact location of the cells (shown in green) during Til steering events. Scale bar: 10 um. 55 56 Figure 14 Ectopic expression of the chimera perturbs Til growth cone steering. A. Til neurons pathfind correctly (red in C) B . Location of t-sema-la expressing cells (green in C). A and B are merged in C. D , G and J . Til neurons behaviour when chimera are placed in the limb during Til steering events. Arrow heads indicate the Cx cells, localized ventrally and proximally in the limb bud. E, H and K. Location of the chimera expressing cells, mostly in the ventral area of the trochanter. F, I and L. Neurons (shown in red) from D, G and J are merged respectively with E, H and K, showing the exact location of the cells (shown in green) during the Til steering events. Scale bar: 10 um. 57 58 Figure 15 Comparison of the frequencies of Til pioneer neurons guidance errors when their growth cones are navigating in the presence of different expressing cell lines. 59 No cells t-Sema-1a gSema-2a Chimera Expressing S 2 cells 60 confirms that gSema-2a is maybe secreted into the surrounding substrate. Dorsal Projection A second misguidance phenotype observed was a misdirected dorsal extension. Normally, Til neurons extend proximally until they reach the trochanter where they turn ventrally. In this region, the endogenous gSema-2a dorsal-ventral gradient is thought to eliminate dorsal projections (Isbister et al., 2003). Isbister et al. (2003) characterised this gradient as being shallow, and is approximately 90% effective in preventing dorsal projection. Indeed, the same dorsal misrouting distribution in control limb-fillets without cells or in the presence of t-Sema-la expressing cells is observed, with dorsal extensions in 8% and 9% of cases, respectively (figure 16). In contrast, when gSema-2a and chimera expressing cells are placed in the ventral portion of the trochanter, dorsal extension phenotypes doubled and reached in both cases approximately 21%. In this respect, figure 14 D, F illustrates well that the ventral presence of chimera expressing cells can overcome the function of the endogenous gSema-2a gradient, by being more repulsive. In this example, expressing cells were selectively placed ventrally in the trochanter. Consequently, growth cones oriented axon approximately at 180° away from the expressing cells and also from the targeted Cxi cells. It is interesting to note that the growth cones appeared to contact the expressing cells before changing direction. By contrast, axons from figure 13 J, L turned dorsally while keeping the expressing cells at distance. Finally, this phenotype illustrates well that the gSema-2a semaphorin domain is repulsive for growth cone steering. Axon Defasciculation Figure 13 J, L is an example that shows ectopic sources of proteins (gSema-2a) can result in a combination of misguidance phenotype from individual neurons. In this example of axon defasciculation, two dorsal projections are apparent and a ventral projection is present. An interesting features to note are that growth cones are oriented distally, away from the ectopic sources of gSema-2a and growth cones located away from the cells are quite complex having a large number of filopodia. In addition to Til misguidance, as suggested in figure 14 D, F, the Cxi cells are also affected by the addition of gSema-2a in their area. Instead of elongating proximally to enter the CNS, their axons extend dorsally. 61 In control conditions, the proportion of axon defasciculation in limb-fillets only is approximately 5% and increases to approximately 8% in the presence of t-Sema-la expressing cells. This increased considerably in the presence of gSema-2a or chimera expressing S2 cells with proportion of defasciculated axons reaching 24% and 19% respectively (figure 16 and Table 1). Aberrant Branching The last significant type of abnormal behaviour reported here refers to an individual Til pioneer neuron that has more than three axons along their projection. A typical example is shown in figure 14 G, I. Neurons falling in this category have a normal axon until they get closer to (when in presence of gSema-2a expressing cells) or penetrate (when in presence of chimera expressing cells) the area of expressing cells. As shown, in the presence of chimera expressing cells, growth cones often stopped steering and other branches started elongating outside the cell border from the main axons. Obviously all of processes that attempted to grow in the direction of the Cxi cells failed to reach their targets and ended their elongation not far from the cell entrance. According to the diameter of the projection and the brightness of the staining, the healthiest projection usually elongated dorsally outside of the expressing cells. This indicated that endogenous concentrations forming the gSema-2a dorsal gradient are rather repulsive than non permissive, since growth cones are able to elongate and steer in the area. Similarly, dSema-2a has been shown to be a repulsive cue on motor axons, but it has also been reported that its presence does not block growth cone navigation (Matthes et al, 1995). The aberrant branch phenotype is not observed when no cells are present in the limb-fillet. Similarly, such abnormality is not seen in the presence of t-Sema-la expressing cells. This phenotype is only observed when gSema-2a and chimera expressing cells are seeded in the limb (7% and 9%, respectively, figure 16). As others have suggested (Isbister et al, 1999), the full-length gSema-2a protein is a functional guidance cue used by the Til pioneer growth cones in order to pathfind correctly in the developing limb bud. It is also important to note that a recent study has shown the importance of gSema-2a gradients for correct growth cone steering and it is suggested that the gradients have a prominent role in growth cone pathfinding (Isbister et al, 2003). 62 Figure 16 and Table 1 Til growth cones make frequent aberrant steering decisions when ectopic sources of gSema-2a or the chimera are strategically placed in the limb. Aberrant Til pioneer steering decisions were divided into four subgroups based on their most prominent phenotype. Each aberrant projection was considered to have only one phenotype. Results are summarized in Table 1. 63 Figure 16 Table 2 No Cells Expressing cells Types of misguidance-induced t-Sema-1 a g$ema-2a chimera Axons Defasciculation (%) 5 ( ± 2 ) 8 ( ± 3 ) 2 4 ( ± 6 ) 1 9 ( ± 4 ) Dorsal Projection (%) 8 ( ± 4 ) 9 ( ± 3 ) 2 1 ( ± 5 ) 2 1 ( ( ± 4 ) Growth Inhibition (%) 3 ( ± 3 ) 7 ( ± 3 ) 2 6 ( ± 8 ) 3 2 ( ± 6 ) Aberrant Branching (%) 0 0 7 ( ± 3 ) 9 ( ± 3 ) others (%) 0 2 1 3 Number of experiments 8 1 4 11 1 5 64 Clearly, the presence of ectopic gSema-2a cells is sufficient to overcome the endogenous gSema-2a gradient. Similarly, the chimera proteins can interfere with the endogenous gSema-2a cues and disrupt Til growth cones pathfmding. In addition, gSema-2a and the chimera also function as a repulsive cue for the Cxi cells, as they elongate their axons dorsally and distally instead of proximally (figures 13 J, L and 14 D, F). Also, as it has been reported previously, mechanisms of cell migration are highly similar to the ones involving neurite elongation (Song and Poo, 2001). In this respect, figure 13 G, I shows an example of where gSema-2a also has the ability to impair cell migration, since the Cxi cells failed to migrate to their final location. 65 DISCUSSION In the past two decades a wealth of information has accumulated about the nature of guidance cues. Evidence suggests categorising these cues into two main groups based on function. They include attractive and repulsive molecules, which can be further categorized depending on whether they have short or long range activity. Here I present a functional analysis of the secreted guidance cue, Sema-2a, in the developing grasshopper organism. As a first examination of the functional domains of gSema-2a, a novel transmembrane gSema-2a was tested in vivo on the developing Til pioneer neurons in the grasshopper limb bud. The analysis shows that growth cones can process the same molecule either under gradients or as a discrete cue, as the transmembrane gSema-2a had the same repulsive function as the wild type gSema-2a. I also show that the transmembrane and cytoplasmic domains of gSema-la are not sufficient to disrupt the function of the semaphorin domain from gSema-2a. gSema-2a in the Developing Embryo: Dynamic Expression and Function Til pioneer axons elongate normally along a longitudinal axis along the length of the femur. gSema-2a is first expressed in the developing grasshopper limb bud as early as 28%, well before neurons differentiate from the epithelium and the expression persists beyond 40%, well after the Til pioneer neurons have finished navigating their pathway. Its early and widespread expression strongly suggests that gSema-2a has an unknown function on non neuronal cells as has been reported in the developing worm (Roy et al., 2002). The dynamic expression of gSema-2a in the limb bud between 31% and 34% of development corresponds to the time of Til growth cones extension toward the CNS. A recent study suggests that the endogenous gSema-2a gradients are functionally relevant for the correct Til growth cone steering in vivo (Isbister et al., 2003), suggesting that Til growth cones respond to changes in gSema-2a concentration. Although reading gradients is more likely to reflect in vivo Til growth cone steering mechanisms, the penetrance of defects obtained with the chimera construct (84%) revealed that gSema-2a functions equally in a gradient form or as a discrete cue. In this respect, when gSema-2a expression changes to discrete circumferential bands 66 (figure 7), gSema-2a probably still function as a neuronal guidance cues for motor neurons developing at later stages. The first study on gSema-2a suggests that the protein was a neuronal guidance cue, affecting the Ti neurons navigation (Isbister et al, 1999). In contrast, dSema-2a has been shown to be secreted by some targeted muscles in order to inhibit some synaptic arborizations and thus to specify neuromuscular junctions of RP1 and RP3 axons with the right muscle. In this situation, dSema-2a functions as a repulsive cue for RP3, but RP1 is insensitive to the molecule while proceeding to synaptogenesis (Matthes et al., 1995). With their results, the authors did not present dSema-2a as a neuronal guidance cue. In contrast, due to the temporal expression of gSema-2a (before and after Til axonogenesis), it is tempting to assume that gSema-2a functions as a neuronal guidance cue for other neurons as well as for the Til. Consistent with this, I found that ectopic presentation of the protein affects the migration of the Cxi cells and their outgrowth. Therefore, the early and diffuse dorsal-ventral gradient expression of gSema-2a may function to direct the migration of the Cxi cells, one migrating from the anterior limb compartment and the other migrating from the posterior compartment to the ventral midline (Bentley and Toroian-Raymond, 1989). It is interesting to note that their final destination is where the absolute concentration of gSema-2a is very low if non existent (figures 7 and 8). Similar to the semaphorin family, Slit proteins form a family of repulsive neuronal guidance cues conserved through the phyla and some members have been demonstrated to direct migration of diverse type of cells (reviewed by Brose and Tessier-Lavigne, 2000), as it is proposed here for gSema-2a directing the migration of the Cxi guide post cells. Having such similarities between two protein families supports the idea that there are common molecular mechanisms controlling both axon pathfinding and cell migration. Protein Function Analysis The chimera protein construct corroborates the results obtained from other semaphorin proteins that reported the semaphorin domain to contain the functional domain(s). The absence of the Ig domain from gSema-2a does not affect the repulsive function of the chimera and the transmembrane and cytoplasmic domains of gSema-la 67 appear not to attenuate the repulsive signaling of the gSema-2a semaphorin domain. Consequently, it is tempting to speculate that the attractive function of the gSema-la molecule resides in its semaphorin domain. However, it is always possible that the presence of the cytoplasmic and the transmembrane domains do magnify the attractive effect of the semaphorin domain by clustering the proteins (Wong et al, 1999). For the gSema-2a homologue, Sema-3A, both the Ig and the basic domains have been reported to be necessary for covalent disulfide bond dimerization, which is essential for function (Kopel and Raper, 1998). However, gSema-2a does not have any basic domain, and the Ig domains from Sema-3A and gSema-2a have no significant degree of similarity. Therefore, based on primary sequences, while what appears to be true for Sema-3A might not apply for gSema-2a. This raises the possibility that the Ig domain of gSema-2a has some functional activity. In order to investigate whether the Ig domain of gSema-2a encodes some functional properties, we can compare the activity and penetrance of the chimera with the activity and penetrance of gSema-2a. Firstly, Western blot and immunostaining analysis indicate that the chimera is well expressed in vitro and secondly it stimulates aberrant Til neuronal phenotypes as their growth cones pathfind in situ. Therefore, the stability and the functionality of the gSema-2a semaphorin domain are maintained even in the absence of the Ig domain. In addition, given the identical nature of misguidance-errors and the equal penetrance of the phenotypes triggered by ectopic gSema-2a and chimera expressing cells, they indicate that if the Ig domain is necessary for gSema-2a to function, this function can be replaced by the collective effect of the transmembrane and cytoplasmic domains of gSema-la. Consequently, the potential dimerization provided by the Ig domain appears not necessary for the repulsive function. On the other hand, structural features of the gSema-la suggest that the C-terminus is involved in protein clustering, since the last three amino acids of gSema-la encode for a putative PDZ binding domain (See Appendix 2 for the entire protein sequence of gSema-la). Consistent with this idea, the gSema-la intracellular domains may have the same structural function as the Ig domain found on gSema-2a since they can be interchanged without affecting the function of the semaphorin domain. But unlike gSema-2a, there is no evidence for gSema-la to form homodimers. Instead, it is more likely that the proteins form clusters by binding to PDZ-containing proteins. A candidate protein with PDZ-containing motifs to interact with gSema-la is PSD-95, because it has 68 been reported to interact with a gSema-la homologue, Sema-4C (Inagaki et al 2001). However, the PDZ binding sequence is not totally conserved between the gSema-la and Sema-4C. For this reason, gSema-la may interact with other PSD protein family members. PDZ binding proteins are known to be associated with the membrane and to carry sequence motifs for cytoskeletal localisation. Also, such motifs are well suited to spatially localise the protein with other molecules implicated in intracellular signaling cascades. Indeed, it has been shown that dSema-la, expressed by neuronal cells, functions as a ligand as well as a receptor in the CNS, signaling via ENABLED (Godenschwege et al. 2002). Similarly, a previous study suggested a functional role for the transmembrane and cytoplasmic domains of dSema-1 a because a secreted form of the protein only partially rescued embryonic neuronal pathfinding in Sema-la null mutants (Yu, H-.H. et al., 1998). In the developing limb bud, gSema-la could interact with a PDZ-containing protein to cluster and to concomitantly increase the density of gSema-la receptors at the neuronal surface, thereby magnifying the signaling cascade in the Til pioneer growth cones. Also, the chimera construct indicates that the transmembrane and cytoplasmic domains of gSema-la are not sufficient to function as attractive domains because the repulsive and even the non permissive effects of the gSema-2a semaphorin domain in the chimera construct are as strong as they are with ectopic gSema-2a proteins. Secreted and Discrete Signaling Mechanisms gSema-2a and chimera expressing cells induce a few different types of abnormal growth cones steering events. First, I found that the presence of either type of cell line had a comparable effect on stimulating growth cone steering errors. Also, an analysis of the misguidance phenotypes indicates that both proteins can overcome endogenous guidance cues. A review from Song and Poo (2001) hypothesises that a gradient established from soluble molecules triggers different growth cone responses than from a gradient established with cell associated molecules. They based this proposal on the basis that diffusible ligands can distribute the receptors differently at the cell surface, leading to the activation of different signaling pathways. In addition, a redistribution of receptors at the cell surface may affect different areas of the growth cone, leading to different growth cone orientations. The experiments presented here directly address this proposal as we examine the effect of 69 ectopically express a diffusible repulsive cue (gSema-2a) and a transmembrane form of the same molecule (chimera). It is apparent that the distribution of ectopic gSema-2a is different from the chimera. Given that gSema-2a will diffuse randomly from the expressing cells it is highly unlikely that the distribution across a growth cone will be uniform. On the other hand, the distribution of the chimera protein presented to one growth cone will be rather even, as the chimera proteins are transmembrane and possibly linked to the cytoskeleton of the expressing cells. Also, the average diameter of one expressing cell is 10 urn (Schneider, 1972) which is similar to the average breadth of the Til pioneer growth cones. For this reason, the chimera proteins probably do not establish any gradient across a growth cone. However, it is important to note that individual expressing cells do not present the same number of chimera molecules at their surface. However, even though the wild type and chimera constructs may be distributed quite differently, they have very similar effects on growth cone steering. In addition, the signaling mechanisms appear similar as they result in identical phenotypes. Finally, these observations provide evidence for an additional steering mechanism besides measuring relative concentrations of gSema-2a (Isbister et al., 2003). In some cases, when growth cones penetrated a high concentration of gSema-2a, it resulted in the termination of axon elongation suggesting they reached a concentration of a repulsive molecule corresponding to the upper limit in which they can navigate. These observations suggest a model in which relative concentrations are analysed (until they reach an upper limit) and that the shape of the gradient and the absolute protein concentration are also taken into account for accurate growth cones steering (figure 17). The model suggests that the same guidance molecule can be instructive (repulsive) at low concentration and according to the level of receptors' occupancy, it can change to a non-permissive cue with the ability to stop growth cone steering and to paralyse axon elongation. However, it is important to distinguish that the mechanism presented in this model does not destabilise all cytoskeleton components, since new protrusions arise from the main axon shaft. 70 Growth Cone Misrouting Dorsal projections Due to the repulsive activity of gSema-2a on steering growth cones, ectopic sources of gSema-2a or chimera proteins induced four major types of misguidance. The major phenotype defect was dorsal elongation from one or both axons. One observation from such defects is that once growth cones elongated dorsally (to avoid expressing cells placed ventrally) they never turned back ventrally. They either stopped dorsally or tended to steer distally. A simplistic hypothesis is that expressing cells are presenting a higher amount of repulsive cues than what is generated by the endogenous gSema-2a gradient. Consequently, growth cones extend down the novel ventral repulsive gradient, in a similar way they normally do with the dorsal-ventral gradient as proposed by Isbister et al. (2003). Although this model is probably accurate, it does not address why growth cones proceed to the trochanter before turning ventrally. One explanation is that the grasshopper limb bud expresses a novel grasshopper guidance cue antagonistic (attractive cue, dorsally expressed) or similar (repulsive cue, ventrally expressed) with gSema-2a. Similarly, muscle targeting by motor axons in the PNS oi Drosophila involves multiple cues: netrin A (NetA), netrin B (NetB) and Sema-2a (Keleman and Dickson, 2001; Weinberg et al. 1998). Such an invertebrate model involving at the same time Sema-2a and netrins molecules for correct growth cone steering is appealing for the grasshopper limb bud. Since there is no evidence whether the missing molecule is attractive or repulsive, the first candidate to consider is the attractive NetA because it has been shown to be expressed by epidermal cells in the PNS of Drosophila (Mitchell et al, 1996). Where molecules are expressed in similar systems is important to consider, since at this stage the limb bud does not contain any elaborate structures, such as mature neurons or muscles able to express any guidance cues. Axon Defasciculation and Growth Cone Turning Axon defasciculation might be mediated by the sensing of different amounts of guidance cues by each of the two Til growth cones. Similarly, growth cone turning is probably induced by the asymmetric presence of guidance cues across both growth cones, leading to the same response from both growth cones. Either way, I propose that growth 71 Figure 17. Model proposed for growth cone steering involving repulsive guidance molecules expressed in gradients. A . When a growth cone encounters a repulsive gradient, its cytoskeleton is asymmetrically destabilized due to a differential occupancy of the receptors across the growth cone's surface. Filopodia and lamellipodia retract where the concentration of the repulsive cue is highest (red). The growth cone expansion is asymmetric and biased toward the area free of repulsive cue (green), inducing growth cone extension away from the repulsive area. B. A growth cone steers in relation to the relative concentration of a repulsive cue. Steering while taking into account relative concentrations may change an avoided area into a permissive area depending of the context. Concentrations are relevant until they reach a threshold. This upper limit indicates the level of receptors occupancy at which internal mechanisms of sensitization are not effective to reset the base line used to detect differences between two absolute concentrations. In that situation, the growth cone is not able to turn away and stall in an area elevated in repulsive guidance cue. C. Growth cone paralysis arises when gradients from repulsive cues fully occupy the receptors on the entire surface of the growth cone. Thereby, protein cascades leading to F-actin destabilization are turned on in the entire growth cone. The cytoskeleton responds faster to these global instructions compared to the mechanisms involved in the adjustment of the baseline of the sensitivity and in the evaluation of relative concentrations. 72 A B Lef&nd *> G-Actm; «k gSema-2a recap tar activated; A g Sena-2a recep-tor ligand free; • H i g h to low level o: coricentratioD c^gSema-2a. 7 3 cones turn away from expressing cells due to an asymmetric destabilisation of their cytoskeleton on the side where receptors are the most activated by the binding of the repulsive cue. In fact, because gSema-2a protein diffuses away from the cells, the protein is more likely to have an uneven distribution across growth cones. In this respect, two growth cones side by side may not be affected at the same level, possibly resulting in their separation. In contrast, when the chimera protein is presented to the two growth cones, it seems reasonable to assume that both of them are equally affected by its presence since gradients are not established as with diffusible molecules. Therefore, growth cones are more likely to stay fasciculated. Nevertheless, the facility with which molecular gradients are established is not the only factor influencing growth cone steering. It is possible that growth cones are not equally sensitive to the guidance cues because of a different number of receptors at their surface. On the other hand, some of the defasciculation phenotypes reproduce the typical misguidance observed with gSema-la antibodies. This raises the question whether at high concentration gSema-2a can antagonize gSema-la binding to gSema-la receptors, resulting in defasciculation (Kolodkin et al., 1992). To examine this more closely, it will be important to identify the receptor(s) that interact with the two semaphorins. In addition, the fact that the chimera and gSema-2a expressing cells are promoting dorsal projections, it follows that the Til growth cones navigate in the developing limb taking into account the relative concentrations of gSema-2a, as previously suggested by Isbister et al., (2003). The simplest explanation for dorsal projections is that expressing cells are presenting a significantly higher amount of repulsive cues than the endogenous gSema-2a gradient. Thus the dorsal region may become more permissive than the ventral region for growth cone pathfinding. It is also notable that when growth cones reach a certain level of gSema-2a concentration, they stop steering dorsally. Aberrant Branching or Growth Inhibited? As reported in other organisms, extending several branches from the main axon has been shown to be an important neuronal behaviour used for large scale exploration, followed by retracting inappropriate branches after forming the correct neuronal connections 74 (CTLeary, 1990). One of the aberrant phenotypes I observed is characteristic of this behaviour. In these cases, multiple protrusions extend from the main axon presumably because the original growth cone has discontinued extension after penetrating a group of gSema-2a or chimera expressing cells. This suggests that sampling the environment with back branch extensions may be an alternative strategy for growth cone turning. In such cases, the original growth cone may not turn immediately because the concentration of guidance cue was too low to affect the steering mechanisms or alternatively, the steering mechanisms were too slow to prevent the growth cone from penetrating the repulsive group of cells (Matthes et al., 1995). Subsequently, their whole structures were surrounded by negative cues. It is remarkable that all neurons presenting this phenotype had their original growth cone in the trochanter (figures 14 G, I). Previously, the substratum of this area has been shown to have a degree of adhesiveness, high enough to impair growth cone retraction (Isbister and O'Connor, 1999). Thus, axonal branching may be adopted when the repulsive cue is widely distributed but the concentration does not reach the critical level needed to induce the full growth cone collapse. A molecular explanation for this phenotype would be that when the concentration of the repulsive cue is below the threshold concentration inducing growth cone paralysis, the whole structure of the growth cone is partially destabilised, leading to the formation of new branches outside the boundary of expressing cells. The results presented show that the aberrant phenotype "growth inhibition" has the highest proportion. I propose that this occurs only when the amount of repulsive cue reaches a certain threshold and if the receptor occupancy is largely distributed evenly across the neuronal surface. In contrast, when repulsive cues activate receptors locally, growth cones simply turn away to avoid the area (Fan and Raper, 1995). Consequently, when the global concentration of gSema-2a reaches a certain concentration level, intracellular signals reach a threshold that destabilise the whole growth cone, leading to growth cones collapse (Ba-Charvet et al., 1999). Thus, steering mechanisms of the Til growth cones can also take into account the absolute concentration of gSema-2a. These observations suggest that gSema-2a functions firstly as a repulsive instructive cue. However, at high enough concentrations, gSema-2a is also able to impair growth cone motility. Similar to the concentration-75 dependant activity of gSema-2a, Sema-3 A has been previously reported to stop axon growth according to its concentration (Shepherd et al., 1997; Renzi et al., 2000). Thus, as reported for Sema-3 A, we were able to demonstrate that its homologue gSema-2a is equally able to promote exploration as well as stimulate mechanisms leading to growth cone collapse. Having such different neuronal effects suggest that the level of receptor occupancy plays a role for the downstream signaling. Likewise, the ability of cyclic nucleotides to switch a repulsive response into an attractive effect (Song et al., 1998) suggests that different forms of growth cone steering (for examples instructive versus permissive) may not involve many different downstream signaling molecules. Thus, the ability of gSema-2a to switch from a repulsive instructive cue to a non-permissive cue suggests it can be controlled by the level of activation of the signaling cascade. The last aspect that needs to be clarified from the antibody blocking experiments (Isbister et al., 1999) is whether gSema-2a functions by itself or in a heterogenous complex of proteins, because in both cases, antibodies directed against gSema-2a protein would potentially have blocked gSema-2a hetero complexes. The ultimate way to test whether a protein is able to trigger by itself any guidance signals is by the addition of an ectopic source of one protein, because an increase in one protein from a complex of proteins is not likely to have a high penetrance of effect. My results suggest that gSema-2a does not function while in a complex of proteins but rather functions as a neuronal guidance cue by itself. Consequently, gSema-2a protein is sufficient to dramatically affect growth cone steering by itself. Also, ectopic sources of a transmembrane form of gSema-2a demonstrate that gSema-2a protein does not necessarily has to be in a gradient to function as a neuronal guidance cue for the Til neurons. It follows then that the addition of the transmembrane and the cytoplasmic domains from an attractive molecule does not perturb the signaling function attributed to the gSema-2a semaphorin domain. 76 Future Directions Hypothetical Fragments used for Function Being both attractive and repulsive is not generally observed for a neuronal guidance cue. In this respect, this section raises the question whether conserved residues on three semaphorin protein members, 1 a, 2a and 3 A, are related by function or are rather conserved due to a common ancestor. In order to approach this issue, the amino acid sequence of gSema-2a was compared to its vertebrate homologue cSema-3A which functions as well as a repulsive cue. The literature reports two groups who analyzed in detail the amino acids sequence of cSema-3A. However, each of them pointed out different fragments necessary and sufficient for growth cone collapse. The first group isolated a 70 amino acid region close to the very end of the N-terminus of the Sema domain (Koppel et al., 1997). More recently, Behar et al. (1999) showed that class 3 semaphorin members contain two conserved subdomains, located very close to the Ig domain, that are also required and sufficient for growth cone collapse. Interestingly, this fragment is found to be highly homologous to a fragment in the tarantula hanatoxin. Both sequences are compared to the sequence of gSema-2a to examine whether they are conserved in gSema-2a. Similarly, the second part of the analysis focuses on the comparison of primary sequences of two proteins closely related in the phyla, but having opposite function on the developing Til pioneer neurons, gSema-la and gSema-2a. Such examination will bring to light where and what are the non-conserved residues candidate for an attractive or repulsive function of gSema-la and gSema-2a, respectively. N-terminus Sequence Fragment of the cSema-3A Functional studies on the vertebrate Sema3A revealed that the very end of the N-terminus region of the protein contains a fragment sequence made of 70 amino acids responsible for the repulsive activity of cSema-3 A on dorsal roots ganglions (DRGs) from embryonic chicks (Koppel, A.M. et al, 1997). In order to verify whether this protein fragment is generally conserved throughout repulsive semaphorin members, figure 18 shows the alignment of protein fragments from cSema-3A and gSema-2a, including the 70 amino acid fragment sequence from cSema-3A found sufficient for repulsion. In fact, this fragment 77 sequence is poorly conserved, with approximately 30.5% identity with gSema-2a. When this cSema-3A region is aligned with gSema-la (attractive protein), it is quite a surprise to find a higher proportion of identical amino acids (approximately 35%, alignment not shown), than between two proteins having the same function (Sema-3A and Sema-2a). In order to determine whether gSema-2a and cSema-3 A proteins contain a region better conserved elsewhere, figure 18 shows also the alignment of their whole semaphorin domains. Although the semaphorin domain is the hallmark of all semaphorin protein members, only 34% of the amino acids are identical. This alignment shows that the two homologue members have no obvious identical long stretch of amino acids. All identical fragments are short -one to seven amino acids long- and are rather well dispersed through the whole semaphorin domain. Fragment at the C-terminus of cSema-3A Semaphorin Domain Behar et al. (1999) presented that two subdomains related to the tarantula hanatoxin are conserved through the cSema-3 protein subfamily members. Similarly to the 70 amino acid fragment at the N-terminus of cSema-3A, the hanatoxin related sequence fragment has been shown to be sufficient for growth cone collapse. Because the analysis of whole semaphorin domains did not point out any identical region longer than seven amino acids, it is worth investigating whether identical small regions corresponding to the hanatoxin related fragment are better conserved. In this respect, table 2 contains sequence alignments of cSema-3A, gSema-2a and gSema-la with the hanatoxin fragment, showing that the region of the hanatoxin related fragment is not well conserved between proteins even if they have the same function. gSema-2a contains partially one fragment out of the two, which is probably not significant since gSema-la contains the same subdomain. In fact, the whole fragment sequence is better conserved between gSema-la and gSema-2a than between gSema-2a and the toxin itself. For this reason, having a partial tarantula hanatoxin sequence fragment homology is not expected to make a significant contribution to determine whether the molecule functions as an attractive or repulsive guidance cue. 78 Figure 18 Primary sequence alignment of cSema-3A and gSema-2a semaphorin domains. The 70 amino acids region in cSema-3A shown to be necessary and sufficient to trigger DRG growth cone collapse is underlined in the cSema-3A sequence and identical amino acids are shown in bold characters. 79 cSema-3A YHTFLLDEERSRLYVGAKDHIFSFNLVNIK--EYQKIVWPVSHSRRDECKWA6KDILREC gSema-2a YRTFHLDEKRESLYVGALDKVYKLNLTNISLSDCERDSLTLEPTNIANCVSKGKSADFDC cSema-3A ANFIKVLKTYNQ-THLYACGTGAFHPMCTYIEVG-SHPEDNIFRMEDSHFENGRGKSPYD gSema-2 a KNHIRVIQPMGDGSRLYICGTNAHSPKDWWYSNLTHLQRHEYV---PGIGVGIAKCPFD cSema-3A PKLLTASLLVD GEL YSGTAADFMGRDFAIFRT LGHHHPI--RTEQHD gSema-2a PEDSSTAVWVENGNPGDLPGLYSGTNAEFTKADTVIFRTDLYNLTTGRREYSFKRTLKYD cSema-3A SRWLNDPRFISAHLIPESDNPEDDKIYFFFRENAIDGEHTGKATHARIGQICKNDFGGHR gSema-2a SKWLDNPNFVGSFDVGE YVLFFFRETAVEYINCGKSVYSRVARVCKKDVGGKN cSema-3A SLVNKWTTFLKARLICSVPGPNGIDTHFDELQDVFLMNSKDPKNPIVYGVFTTSSNIFKG gSema-2a ILSQNWATFLKARLNCSIPGE--FPFYFNEIQGVYKM PNTDKFFGVFSTSVTGLTG cSema-3A SAVCMYSMTDVRRVFLGPYAHRDGPNYQWVPY-QGRVPYPRPGTCPSKTFGGFDSTKDLP gSema-2a SAICSFTLKDIQEVFSGKFKEQATSSSAWLPVLPSRVPDPRPGEC VNDTELLP cSema-3A DEVITFARSHPAMYNPVFPINSRPIMIKTDVDYQFTQIWDRVDA EDGQYDVMFIG gSema-2a DTVLNFIRSHPLMDGAVSHEGGKPVFYKRDV--LFTQLWDKLKVNLVGKNMEYIVYYAG cSema-3A TDIGTVLKW gSema-2a TSTGQVYKW Homology between gSema-la and gSema-2a Amino Acids Sequences Finally, due to a lack of sequence conservation between the two homologous repulsive proteins cSema-3A and gSema-2a, the primary amino acids sequence of the attractive gSema-la protein was compared to the primary amino acids sequence of gSema-2a. By default, amino acids sufficient to confer an attractive activity should not be conserved in the gSema-2a sequence and the amino acids used for repulsion should not be found in the gSema-la sequence either. Figure 19 shows the alignment of the gSema-la and gSema-2a semaphorin domains. The first feature to note is that although gSema-la and gSema-2a have opposite functions, they have a higher proportion of identical amino acids in their semaphorin domains (41%) than have two proteins having the same function, as it was previously shown between gSema-2a and cSema-3 A. Overall however, the two extremities of the sequences are poorly conserved. The C-terminus contains stretches of one to four identical amino acids spaced with non conserved residues or with short gaps in the gSema-2a sequence. Similarly, the N-terminus contains longer but fewer identical protein fragments and the amino acids pairing has longer gaps in the gSema-la sequence. In order to obtain more insights on the amino acids sequences that confer the attractive activity to gSema-la and the repulsive activity to gSema-2a, additional chimeras have to be engineered. However, considering the poor conserved homology of gSema-2a over the corresponding functional region of cSema-3A (Behar et al., 1999) it is highly unlikely that amino acids 152 to 237 (the corresponding gSema-2a region) is the functional region of gSema-2a. Also, even if the hanatoxin related fragment sequence is composed of two short subdomains, there is only a weak possibility of locating the gSema-2a functional sequence fragment between amino acids 498 and 518, as it has been reported for function on Sema-3 A members (Behar et al., 1999). For these reasons, in order to identify the protein fragment that confers the repulsive function on axonal elongation, I propose to take into consideration the alignment of gSema-la and gSema-2a protein sequence. The new constructs will consist of insertions of gSema-la protein fragments into the gSema-2a semaphorin domain backbone. These constructs will become more refined as the experiment progresses. Ultimately, the specific amino acids sufficient 81 Table 2 Primary sequence alignment of the tarantula hanatoxin protein fragment with cSema-3A, and its comparison to the corresponding region on gSema-2a and to gSema-la proteins. Identical amino acids to the tarantula hanatoxin sequence fragment are in bold characters. 82 Hanatoxin C K T T A D C C K H L G C K F R D K Y C A W D F T F S sequence fragment cSema-3A G K A C A E C C L A R D P Y C A W D G gSema-2a C H H R Y S N C - - L Q C A - R D P Y C G W D gSema-la L Q D P Y C A W D Figure 19 Primary sequence alignment of the semaphorin domains from gSema-la and gSema-2a. Identical amino acids are in bold characters. 84 g S e m a - l a L E K D H N S L L V G A R N I V Y N I S L R D L T E F T E Q R I E W H S S G A H R E L C Y L K G K S E D - D C Q N Y I R g S e m a - 2 a L D E K R E S L Y V G A L D K V Y K L N L T N I S L S D C E R D S L T L E P T N I A N C V S K G K S A D F D C K N H I R g S e m a - l a V L A K I D D - D R V L I C G T N A Y K P L C R H Y A L K D G D Y W E K E Y - E G R G L C P F D P D H N S T A g S e m a - 2 a V I Q P M G D G S R L Y I C G T N A H S P K - D W W Y S N L T H L Q R H E Y V P G I G V G I A K C P F D P E D S S T A g S e m a - l a I Y S E GQL Y S A T V A D F S G T D P L I Y R G P L R T E R S D L K Q L N A P g S e m a - 2 a V W V E N G N P G D L P G L Y S G T N A E F T K A D T V I F R T D L Y N L T T G R R E Y S F K R T L K Y D S K W L D N P g S e m a - l a N F V N T M E Y N D F I F F F F R E T A V E Y I N C G K A I Y S R V A R V C K H D K G G P H Q F G D R W T S F L K S R L g S e m a - 2 a N F V G S F D V G E Y V L F F F R E T A V E Y I N C G K S V Y S R V A R V C K K D V G G K N I L S Q N W A T F L K A R L g S e m a - l a N C S V P G D Y P F Y F N E I Q S T S D I I E G N Y G G Q V E K L I Y G V F T T P V N S I G G S A V C A F S M K S I L E g S e m a - 2 a N C S I P G E F P F Y F N E I Q G V Y K M P N T D K F F G V F S T S V T G L T G S A I C S F T L K D I Q E g S e m a - l a S F D G P F K E Q E T M N S N W L A V P S L K V P E P R P G Q C V N D S R T L P D V S V N F V K S H T L M D E A V P A F g S e m a - 2 a V F S G K F K E Q A T S S S A W L P V L P S R V P D P R P G E C V N D T E L L P D T V L N F I R S H P L M D G A V S H E g S e m a - l a F T R P I L I R I S L Q Y R F T K I A V D Q Q V R T P D G K A - - Y D V L F I G T D D G K V I K A L N S A S F D S S D T g S e m a - 2 a G G K P V F Y K R D V L - - F T Q L W D K L K V N L V G K N M E Y I V Y Y A G T S T G Q V Y K W Q W - - Y D S G G L g S e m a - l a V D S W I E E L Q V L P P G V P V K N L Y W R g S e m a - 2 a P Q S L L V D I F D V T P P E - P V Q A L H L S K 85 for a repulsive activity on axon outgrowth will be deleted and replaced by the amino acids from gSema-la responsible for its attractive function on the developing Til neurons. Firstly, it would be interesting to see whether the gaps on gSema-la spaced with non-conserved amino acids do correspond to the functional fragment on gSema-2a sequence. Consequently, the proposed chimera would have the backbone of gSema-2a having amino acids 89 to 200 replaced with amino acids 84 to 170 from the gSema-la (chimera 2). It is of interest also to investigate whether the gSema-la protein fragments missing on gSema-2a sequence are involved in the attractive function of gSema-la protein. Consequently, I propose a chimera having gSema-la backbone with amino acids 264 to 440 from gSema-2a replacing gSema-la amino acids 234 to 420 (chimera 3). Presently, there is no evidence whether gSema-la and gSema-2a function through the same receptor or whether each of them activates their own receptor. If they bind to different receptors, the change in activities of the semaphorin domain of gSema-2a and gSema-la would be due to the activation of different receptors via their transplanted functional domain. If they do have the same receptor, it is possible that their functional domain trigger different conformational rearrangements of the receptor upon binding. In that case, it is the different conformations of the receptor that would trigger the interaction(s) with the different downstream effectors to promote either attraction or repulsion. On the other hand, both chimeras will not have any functional activity if their functional domain does not have the same location in gSema-la and gSema-2a amino acid sequence. 86 CONCLUSION gSema-2a functions as a general repulsive neuronal guidance cue in the PNS of the developing grasshopper, affecting cell migration and growth cone steering. Steering mechanisms inducing axon defasciculation are most likely due to the distribution of the protein, occupying differently the receptor population of the two growth cones that are navigating side by side. However, aberrant growth cone turning, axonal branching and growth cone paralysis are most likely a function of the level of the receptors activated, triggering different strength of intracellular signals. In addition to be an instructive guidance cue, gSema-2a is also a non-permissive cue due to its ability to induce inhibition of growth cone growth migration at high concentrations. This way, gSema-2a protein is among the proteins, including Sema-3A, characterised as instructive cues able to switch to non permissive cues. gSema-2a appears to switch the neuronal response according to the level of receptor occupancy. Also, being either soluble or either transmembrane does not affect the function or growth cone behaviours. Whether ectopic cells are expressing gSema-2a or chimera proteins, the same four aberrant phenotypes are observed in very similar frequencies. The semaphorin domain of gSema-2a is neither masked nor perturbed by the addition of both the transmembrane and the cytoplasmic domains from gSema-la, as the resulting phenotypes show identical growth cone behaviours whether gSema-2a or chimera expressing cells are added. In addition, neither the transmembrane nor the cytoplasmic domains of gSema-la contain any activity sufficient to perturb the function of the semaphorin domain of gSema-2a. On the other hand, although the semaphorin protein family is well conserved in the phyla, the functional domains are poorly conserved between the two homologous protein Sema-3 A and gSema-2a. Consequently it indicates that functional regions encoded in the semaphorin domain are either dependant on the protein structure or coding within one or several blocks of 7 amino acids long or less in the semaphorin domain. 87 REFERENCES Anderson, H. and Tucker, R.P. (1989). Spatial and temporal variation in the structure of the basal lamina in embryonic grasshopper limbs during pioneer neurons outgrowth. Development 106; 185-94. Artigiani, S., Comoglio, P.M. and Tamagnone, L. (1999). Plexins, Semaphorins, and Scatter factor receptors: A common root for cell guidance signals ? IUMB Life 48; 477-82. Ba-Charvet, K.T.N., Brose, K., Marillat,V., Kidd, T. Goodman, C.S., Tessier-Lavigne, M., Sotelo, C. and Chedotal, A. (1999). Slit2-mediated chemorepulsion and collapse of developing forebrain axons Neuron, 22; 463-73. Bahri, S.M., Chia, W. and Yang, X. (2001). Characterization and mutant analysis of the Drosophila sema 5C gene. Developmental Dynamics 221; 32-30. Bagnard, D., Lohrum, M., Uziel, D., Piischel, A. and Bolz, J. (1998). Semaphorins act as attractive and repulsive guidance signals during the development of cortical projections. Development 125; 5043-53. Bate, CM. (1976). Pioneer neurones in an insect embryo. Nature Lond. 260; 54-56. Behar, O., Mizuno, K. Badminton, M. and Woolf, C.J. (1999). Semaphorin 3A growth cone collapse requires a sequence homologous to tarantula hanatoxin. Proc. Natl. Acad. Sci. USA 96;13501-5. Bentley, D. and Caudy, M. (1983). Pioneer axons lose directed growth after selective killing of guidepost cells. Nature 304; 62-5. Bentley, D. and Keshishian, H.(1982). Pathfinding by peripheral pioneer neurons in grasshoppers. Science 218; 1082-88. Bentley, D., Keshishian, H., Shankland, M. and Toroian-Raymond, A. (1979) Quantitative staging of embryonic development of the grasshopper Schistocerca nitens. J. Embryol. Exp. Morpho. 54; 47-74. Bentley, D. and O'Connor, T.P. Guidance and steering of peripheral pioneer growth cones in grasshopper embryos. The Nerve Growth Cone. Ed. Letourneau, Kater and Macagno, Raven Press Ltd., New York 1992. pp 265-81. Bentley, D. and Toroian-Raymond, A. (1986). Disoriented pathfinding by pioneer neuron growth cones deprived of filopodia by cytochalasin treatment. Nature 323; 712-15. Bentley, D. and Toroian-Raymond, A. (1989). Pre-axonogenesis migration of afferent pioneer cells in the grasshopper embryo. J. Exp. Zool. 251; 217-23. 88 Birgbauer, E., Oster, S.F., Severin, CG. and Sretavan, D.W. (2001). Retinal axon growth cones respond to EphB extracellular domains as inhibitory axon guidance cues. Development 125; 3041-48. Bito, H., Furuyashiki, T., Ishihara, F£., Shibasaki, Y. and Ohashi, K. (2000). A critical role for a Rho-associated kinase, pl60ROCK, in determining axon outgrowth in mammalian CNS neurons. Neuron 26; 431-41. Black, M.M. (1994). Microtubule transport and assembly cooperate to generate the microtubule array of growing axons. Prog. Brain Res. 102; 61-77. Bonner, J. and O'Connor, T.P. (2001). The permissive cue laminin is essential for growth cone turning in vivo. J. Neurosci. 15; 9782-91. Bonner, J., Auld, V. and O'Connor, T. (2002). Migrating mesoderm establish a uniform distribution of laminin in the developing grasshopper embryo. Dev. Biol. 249; 57-73. Bork, P. and Beckmann, G. (1993). The CUB domain. A widespread molecule in developmentally regulated proteins. J. Mol Biol. 231; 539-45. Bridgman, P.C. and Dailey, M.E. (1989). The organization of myosin and actin in rapid frozen nerve growth cones. J. Cell Biol. 108; 95-109. Brose, K. and Tessier-Lavigne, M. (2000). Slit proteins: key regulators of axon guidance, axonal branching and cell migration. Curr. Opn Neurobiol. 10; 95-102. Buffo, a., Zagrebelsky, M., Huber, A.B., Skerra, A., Schwab, M.E., Strata, P. and Rossi, F. (2000). Application of neutralizing antibodies against NI-35/250 myelin-associated neurite growth inhibitory proteins to the adult rat cerebellum induces sprouting of uninjured purkinje cell axons. J. Neurosci. 20; 2275-86. Castellani, V., Chedotal, A., Schachner, M., Faivre-Sarraih, C. and Rougon, G. (2000). Analysis of the Ll-deficient mouse phenotype reveals cross-talk between Sema3A and LI signaling pathways in axonal guidance. Neuron 27; 237-49. Catalano, S.M., Messersmith, E.K. Goodman, C.S. Shatz, C.J. and Chedotal, A. (1998). Many major CNS axon projections develop normally in the absence of semaphorin III. Mol. and Cell. Neurosci. 11; 173-82. Challacombe, J.F., Snow, D.M. and Letourneau, P.C. (1996). Actin filament bundles are required for microtubule reorientation during growth cone turning to avoid an inhibitory guidance cue. J. Cell Sci. 109; 2031-2040. Chalasani, S.H., Sabelko, K.A., Sunshine, M.J., Liftman, D.R. and Raper, J.A. (2003). A . chemokine, SDF-1, reduces the effectiveness of multiple axonal repellents and is required for normal axon pathfinding. J. Neurosci. 15; 1360-71. 89 Chen, H., Chedotal, A., He, Z., Goodman, CS. and Tessier-Lavigne, M. (1997). Neuropilin-2, a novel member of the neuropilin family, is a high affinity receptor for the semaphorins Sema E and Sema IV but not Sema III. Neuron 19; 547-59. Colamarino, S.A. and Tessier-Lavigne, M. (1995). The axonal chemoattractant netrin-1 is also a chemorepellent for trochlear motor axons. Cell 81; 621-29. Comeau, M.R., Johnson, R., DuBrose, R.F., Petersen, M., Gearing, P., VandenBos, T., Park, L., Farrahy, T., Buller, R.M. and Cohen, J.L, (1998). A poxivirus-encoded semaphorin induces cytokine production from monocytes and binds to a novel cellular semaphorin receptor, VESPR. Immunity 8; 473-82. Culotti, J.G. and Kolodkin, A.L: (1996). Function of netrins and semaphorins in axon guidance. Curr. Opin. Neurobiol. 6; 81-8. Davenport, R.W., Dou, P., Rehder, V. and Kater, S.B. (1993). A sensory role for neuronal growth cone filopodia. Nature 361; 721-724. Driessens, M.H.E., Hu, H., Nobes, CD., Self, A., Jordens, I., Goodman CS. and Hall, A. (2001). Plexin-B semaphorin receptors interact directly with active Rac and regulate the actin cytoskeleton by activating Rho. Curr. Biol. 11; 339-44. Fan, J., Mansfield, S.G., Redmond, T. Gordon-Weeks, P.R. and Raper, J.A. (1993). The organization of F-actin and microtubules in growth cones exposed to a brain-derived collapsing factor, J. Cell. Biol. 121; 867-878. Fan, J. and Raper, J.A. (1995). Localized collapsing cues can steer growth cones without inducing their full collapse. Neuron 14; 263-74. Fujii, T., Nakao, F., Shibata, Y., Hioi, G., Kodama, E., Fijisawa, H. and Takagi, S. (2002). Caenorhabditis elegans, PlexinA, PLX-1, interacts with transmembrane semaphorins and regulates epidermal morphogenesis. Development 129; 2053-2063. Furuyama, T., Inagaki, S., Kosugi, A., Noda, S., Saitoh, S., Ogata, M., Iwahashi, Y., Miyaki, N., Hamaoka, T. and Tohyama, M. (1996). Identification of a novel transmembrane semaphorin expressed on lymphocytes. J. Biol. Chem. 271; 33376-81. Gallo, G. and Letourneau, P.G. (1999). Different contributions of microtubule dynamics and transport to the growth of axons and collateral sprouts. J. Neurosci. 19; 3860-73. Gierer, A. (1987). Directional cues for growing axons forming the retinotectal projection. Development, 101; 479-489. Giger, R.J., Pasterkamp, R.J., Heijnen, S., Holtmaat, A.J.G.D. and Verhaagen, J. (1998). Anatomical distribution of the chemorepellent semaphorin III/collapsin-1 in the adult rat and human brain: predominant expression in structures of the olfactory-hypocampal pathway and motor system. J. Neurosci Res. 52; 27-42. 90 Ginzburg, V. E., Roy, P. J. and Culotti, J. G. (2002). Semaphorin la and semaphorin lb are required for correct epidermal cell positioning and adhesion during morphogenesis in C. elegans. Development 129 2065-78. Gitai, Z., Yu, T.W., Lundquist, E.A., Tessier-Lavigne, M. and Bargmann, C. (2003). The netrin receptor UNC-40/DCC stimulates axon attraction and outgrowth through Enabled and, in parallel, Rac and UNC-155/AdLIM. Neuron 37; 53-65. Godenschwege, T.A., Hu, H., Shan-Crofts, X., Goodman, C.S. and Murphey, R.K. (2002). Bi-directional signalling by semaphorin la during central synapse formation in Drosophila. Nature Neurosci. 5; 1294-1301. Goodhill, G.J and Baier, H . (1998). Axon guidance: stretching to the limit. Neural Computation 10; 521-27. Goodhill, G.J. and Urbach, J.S. (1999). Theoretical analysis of gradient detection by growth cones. J. Neurobiol. 41; 230-41. Goodman, C.S. (1994). The likeness of being phylogenetically conserved molecular mechanisms of growth cone guidance. Cell 78; 353-56. Goodman, C.S., Raper, J.A., Chang, S. and Ho, R. (1983). Grasshopper growth cones: divergent choices and labelled pathways. Progress in brain Res. 58; 283-304. Gu, X. and Spitzer, N.C. (1995). Distinct aspects of neuronal differentiation encoded by frequency of spontaneous Ca 2 + transients. Nature 375; 784-787. Hall, A. (1998). Rho GTPases and the actin cytoskeleton. Science 279; 509-14. He, Z. and Tessier-Lavigne, M. (1997). Neuropilin is a receptor for the axonal chemorepellent Semaphorin III. Cell 90; 739-51. Hedgecock, E.M., Culotti, J.G. and Hall, D.H. (1990). The unc-5, unc-6, and unc-40 genes guide circumferential migrations of pioneer axons and mesodermal cells on the epidermis in C. elegans. Neuron 4; 61-85. Ho, R.K., Ball, E.E. and Goodman, C.S. (1983). Muscle pioneers: large mesodermal cells that erect a scaffold for developing muscles and motoneurons in grasshopper embryos. Nature 301; 66-9. Holm, K. and Isacson, O. (1999). Factors intrinsic to the neuron can induce and maintan its ability to promote axonal outgrowth: a role for BCL2? Trends Neurosci. 22; 269-73. Holmes, S. Downs, A-M., Fosberry, A., Hayes, P.D., Michalovich, D., Murdoch, P., Moores, K., fox, J., Deen, K., Pettman, G., Wattam, T. and Lewis, C. (2002). Sema7A is a potent monocytes stimulator. Scand. J. Immunol. 56; 270-75. 91 Hopker, V.H., Shewan, D., Tessier-Lavigne, M., Poo, M. and Holt, C. (1999). Growth-cone attraction to netrin-1 is converted to repulsion by laminin-1. Nature 40; 69-73. Horton, R.M., Cai, Z., Ho, S.N. and Pease, L.R. (1990). Gene splicing by overlap extension: tailor-made genes using the polymerase Chain Reaction. BioTechiques 8; 528-32.. Inagaki, S., Ohoka, Y., Sugimoto, H., Fujioka, S. Amazaki, M., Kurinami, H., Miyazaki, N., Tohyama, M. and Furuyama, T. (2001). Sema4C, a transmembrane semaphorin, interacts with a post-synaptic density protein, PSD95*. J. Biol. Chem. 276; 9174-81. Isbister, C. M., Tsai, A., Wong, S. T., Kolodkin, A. L. and O'Connor, T. P. (1999). Discrete roles for secreted and transmembrane semaphorins in neuronal guidance in vivo. Development 126 2007-19. Isbister CM., Mackenzie P.J., To K.C.W and O'Connor T.P. (2003). Gradient steepness influences the pathfinding decisions of neuronal growth cones In Vivo. J. Neurosci., *23; 193-202. Jan, L. Y. and Jan, Y. N. (1982). Antibodies to horseradish peroxidase as specific neural markers in Drosophila and grasshopper embryos. Proc. Natl. Acad. Sci. USA 79; 2700-4. Kato, K., Utani, A., Suzuki, N., Mochizuki, M., Yamada, M., Nishi, N., Matsuura, H., Shinkai, H. and Nomizu, M. (2002). Identification of neurite outgrowth promoting sites on the laminin cc3 chain G domain. Biochemistry 41;10747-53. Keleman, K. and Dickson, B.J. (2001). Short- and long-range repulsion by Drosophila Unc5 netrin receptor. Neuron 32; 605-17 Keshishian, H. and Bentley, D. (1983). Embryogenesis of peripheral nerve pathways in grasshopper legs. II the major nerve routes. Developmental Biology 96; 103-15. Klostermann, A., Lohrum, M., Adams, R.H. and Ptichel, A. (1998). The chemorepulsive activity of the axonal guidance signal semaphorin D requires dimerization. J. Biol. Chem. 273; 7326-31. Klostermann, A., Lutz, B., Gertler, F. and Behl, C. (2000). The orthologous human and murine semaphorin 6A-1 proteins (SEMA6A-l/SemaA-l) bind the Enabled/Vasodilator-stimulated phosphoprotein-like protein (EVL) via a novel carboxy-terminal zyxin-like domain. J. Biol. Chem. 275; 39647-53. Kolodkin, A.L., Matthes, D.J., O'Connor, T.P., Patel, N.H., Admon, A., Bentley, D. and Goodman, C.S. (1992). Fasciclin IV: sequence, expression, and function during growth cone guidance in the grasshopper embryo. Neuron 9; 831-45. Kolodkin, A.L., Matthes, D.J. and Goodman, C.S. (1993). The semaphorin genes encode a family of transmembrane and secreted growth cone guidance molecules. Cell 75; 1389-99. 92 Kolodkin, A.L. (1996). Growth cones and the cues that repel them. Trends Neurosci. 19; 507-13. Kolodkin, A.L. Levengood, D.V., Rowe, E.G., Tai, Y.T., Giger, R.J. and Ginty, D.D. (1997). Neuropilin is a semaphorin III receptor. Cell 90; 753-762. Koppel A.M., Feiner L, Kobayashi H, Raper J.A. (1997). A 70 amino acid region within the semaphorin domain activates specific cellular response of semaphorin family members. Neuron 19; 531-7. Koppel, A.M. and Raper, J.A. (1998). Collapsin-1 covalently dimerizes, and dimerization is necessary for collapsing activity. J. Biol. Chem. 273; 15708-13. Kuhn, T.B., Brown, M.D., Wilcox, C.L., Raper, J.A. and Bamburg, J.R. (1999). Myelin and collapsin-1 induce motor neuron growth cone collapse through different pathways: inhibition of collapse by opposing mutants of Racl. J. Neurosci. 19; 1965-75. Kwon, B.K., Liu, J., Messerer, C., Kobayashi, N.R., McGraw, J., Oschipok, L. and Tetzlaff, W. (2002). Survival and regeneration of rubrospinal neurons 1 year after spinal cord injury. Proc. Nad. Acad. Sci. USA 99; 3256-51. Lau, P-m., Zucker, R.S. and Bentley, D. (1999). Induction of Filopodia by direct local elevation of intracellular calcium ion concentration. Journal of Cell Biol. 145; 1265-75. Lewis, A.K., and Bridgman, P.C. (1992). Nerve growth cone lamellipodia contain two populations of actin filaments that differ in organisation and polarity. J. Cell Biol. 119; 1219-43. Li, X., Saint-Cyr, E., Aktories, K. and Lamarche-Vane, N. (2002). Racl and Cdc42 but not RhoA or rho kinase activities are required for neurite outgrowth induced by the netrin-1 receptor DCC (Deleted in Colorectal Cancer) in N1E-115 neuroblastoma cells. JBiol. Chem. 211; 15207-14. Li, X and Eriksson, U. (2003). Novel PDGF family members: PDGF-c and PDGF-D. Cytokin and Growth Factor Reviews 14; 91-8. Lim, Y.-s. and Wadsworth, W.G. (2002). Identification of domains of Netrin UNC-6 that mediate attractive and repulsive guidance and responses from cells and growth cones. J. Neurosci. 22; 7080-87. . Lindvall, O. (2003). Stem cells for cell therapy in Parkinson's disease. Pharmacological Research 47; 279-287. Loschinger, J., Bandtlow, J., Jung, J., Klostermann, S., Schwab, M.E., Bonhoeffer, F. and Kater, S.B. (1997). Retinal axon growth cone responses to different environmental cues are mediated by different second-messenger systems. J. Neurobiol. 33; 825-34. 93 Luo, Y.D., Raible, D. and Raper, J.A. (1993). Collapsin: a protein in brain that induces the collapse and paralysis of neuronal growth cones. Neuron 75; 217-27. Luo, L. (2002). Actin cytoskeleton regulation in neuronal morphogenesis and structural plasticity. Annu. Rev. Cell Biol. 18; 601.35. Maestrini, E., Tamagnone, L., Longati, P., Cremona, O., Gulisano, M., Bione, S., Tamanini, F., Neel, B.G., Toniolo, D. and Comoglio, P.M. (1996). A family of transmembrane proteins with homology to the MET-hepatocyte growth factor receptor. Proc. Natl. Acad. Sci. USA 93; 674-78. Mallavarapu, A. and Mitchison, T. (1999). Regulated actin cytoskeleton assembly at filopodium tips controls their extension and retraction. J. Cell Biol. 146; 1097-1106. Marchand, P., Tang, J. and Bond, J.S. (1994). Membrane association and oligomeric organization of the alpha and beta subunits of mouse meprinA. J. Biol. Chem. 269; 15388-93. Maronne, D.F. and Petit, T.L. (2002). The role of synaptic morphology in neuronal plasticity: structural interactions underlying synaptic power. Brain Res. Rev. 38; 291-308. Mason, CA. and Wang, L.C (1997). Growth cone form is behavior-specific and, consequently, position-specific along the retinal axon pathway. J. Neurosci. 17; 1086-10100. Matthes, D.J., Sink, H., Kolodkin, A.L. and Goodman, C.S. (1995). Semaphorin II can function as a selective inhibitor of specific synaptic arborization. Cell 81; 631-39. McConnell, S.K., Ghosh, A. and Shatz, C.J. (1989). Subplate neurons pioneer the first axon pathway from the cerebral cortex. Science 245; 978-82. Meyer, G. and Feldman, E.L. (2002). Signaling mechanisms that regulate actin-based motility processes in the nervous system. J. Neurochemistry 83; 490-503. Ming, G.-L., Song, H.J., Berninger, B., Holt, C.E., Tessier-Lavigne, M. and Poo, M.-m. (1997). cAMP-dependent growth cone guidance by netrin-1. Neuron 19; 1225-35. Ming, G.-L, Wong, S.T., Henley, J., Yuan, X.-b., Song, H.-j., Spitzer, N.C. and Poo, M.-m. (2002). Adaptation in the chemotactic guidance of nerve growth cones. Nature 417; 411-18. Misgeld, T., Burgess, R.W., Lewis, R.M., Cunningham, J.M., Lichtman, J.W. and Sanes, J.R. (2002). Roles of neurotransmitter in synapse formation: development of neuromuscular junctions lacking choline acetyltransferase. Neuron 36; 635-48. Mitchell, K.J., Doyle, J.L., Serafini, T., Kennedy, T.E., Tessier-Lavigne, M., Goodman, C.S. and Dickson, B.J. (1996). Genetic analysis of Netrin genes in Drosophila: Netrins guide CNS commissural axons and peripheral motor axons. Neuron 17; 203-15. 94 Mukhopadhyay, G., Doherty, P., Walsh, F.S., Crocker, P.R. and Filbin, M.T. (1994). A novel role for myelin-associated glycoprotein as an inhibitor of axonal regeneration. Neuron 13; 757-67. Nakamura, F., Tanaka, M., Takahashi, T., Kalb, R.G. and Strittmatter, S.M. (1998). Neuropilin-1 extracellular domains mediate semaphorin D/III-induced growth cone collapse. Neuron 21; 1093-1100. Niederost, B., Oertle, T., Frische, J., McKinney, R.A. and Bandtlow, C.E. (2002). Nogo-A and Myelin-associated glycoprotein mediate neurite growth inhibition by antagonistic regulation of RhoA and Racl. J. Neurosci. 22; 10368-76. Norris, CR. and Kalil, K. (1991). Guidance of callosal axons by radial glia in the developing cerebral cortex. J. Neurosci. 11; 3481-92. Oakley, R.A. and Tosney, K.W. (1993). Contact-mediated mechanisms of motor axon segmentation. J. Neurosci. 13; 3773-92. O'Connor, T. P., Duerr, J. S. and Bentley, D. (1990). Pioneer growth cone steering decisions mediated by single filopodial contacts in vivo. J. Neurosci. 10; 3935-46. O'Connor, T.P. and Bentley, D. (1993). Accumulation of actin in substs of pioneer growth cone filopodia in response to neural and epithelial guidance cues in situ. J. Cell Biol. 4; 935-48. O'Leary, E.E., Bicknese, A.R., De Carlos, J.A., Heffner, CD., Koester, S.E., Kutka, L.J. and Terashima, T. (1990). Target selection by cortical axons: alternative mechanisms to establish axonal connections in the developing brain. Cold Spring Harb. Sym.p Quan.tBiol. 55; 453-68. Pasterkamp, R.J., Giger, R.J. and Verhaagen, J. (1998). Regulation of Semaphorin HI/Collapsin-1 gene expression during peripheral nerve regeneration. Exp. Neurology 153; 313-27. Pasterkamp, R.J and Kolodkin, A.L. Semaphorin 7A promotes axon outgrowth. Cold Spring Harbor Laboratory; Axon Guidance & Neural Plasticity, meeting 2002. Pasterkamp, R.J. and Kolodkin, A.L. (2003). Semaphorin junction: making tracks toward neural connectivity. Curr. Opn. Neurobiol. 13; 79-89. Pollard, T.D. and Borisy, G.G. (2003). Cellular motility driven by assembly and disassembly of actin filaments. Cell 112; 453-66. Polleux, F. Morrow, T. and Ghosh, A. (2000). Sema-3A is a chemoattractant for cortical apical dendrites. Nature 404; 567-573. 95 Rabacchi, S.A. and Baird, D.H. (1999). Collapsin-l/Semaphorin-III/D is regulated developmentally in purkinje cells and collapses pontocerebellar mossy fiber neuronal growth cones. J. Neurosci. 19; 4437-48. Raper, J.A. (2000). Semaphorins and their receptors in vertebrates and invertebrates. Curr. Opn Neurobiol. 10; 88-94. Renzi, M.J., Lee Wexler, T. and Raper, J.A. (2000). Olfactory sensory axons expressing a dominant-negative semaphorin receptor enter the CNS early and overshoot their target. Neuron 28; 437-47. Roy, P. J., Zhen, H., Warren, C. E. and Culotti, J. G. (2000). mab-20 encodes Semaphorin-2a and is required to prevent ectopic cell contacts during epidermal morphogenesis in Caenorhabditis elegans. Development 127; 755-67. Schneider, I. (1972). Cell lines derived from late embryonic stages of Drosophila melanogaster. J. Embryol. Exp. Morph. 27; 353-65. Sekido, Y., Bader, S., Latif, F., Chen, J. Y., Duh, F. M., Wei, M. FL, Albanesi, J. P. Lee, C. C , Lerman, M I. and Minna, J. D. (1996). Human semaphorins A (V) and IV reside in the 3p21.3 small cell lung cancer deletion region and demonstrate distinct expression patterns. Proc. Natl. Acad. Sci. USA 93; 4120-25. Shepherd, I.T., Luo, Y., Lefcort, F., Reichardt, L. and Raper, J.A. (1997). A sensory axon repellent secreted from ventral spinal cord explants is neutralized by antibodies raised against collapsing-1. Development 124; 1377-85. Shirvan, A., Ziv, I., Fleminger, G., Shina, R., He, Z., Brudo, I., Melamed, E., and Barzilai, A. (1999). Semaphorins as mediators of neuronal apoptosis. J. Neurochem. 73; 961-71. Song, H.j. Ming, G.-l. and Poo, M.m. (1997). cAMP-induced switching in turning direction of nerve growth cones. Nature 388; 275-79. Song, H.-j., Ming, G.-l. He., Z., Lehmann, M., McKerracher, L., Tessier-Lavigne, M. and Poo, M.-m (1998). Conservation of neuronal growth cone responses from repulsion to attraction by cyclic nucleotides. Science 281; 1515-18. Song, H.-j. and Poo, M.-m (2001). The cell biology of neuronal navigation. Nat. Cell Biol. 3; E81-8. Suter, D.M., Errante, L.D., Belotserkovsky, V. and Forscher, P. (1998). The Ig superfamily cell adhesion molecule, apCAM, mediates growth cone steering by substrate-cytoskeletal coupling. J. Cell Biol. 141; 227-40. Suter, D.M. and Forscher (2000). Substrate-Cytoskeletal coupling as a mechanism for the regulation of growth cone motility and guidance. J. Neurobiol. 44; 97-113. 96 Swierca, J.M., Kuner, r. Behrens, J. and Offermanns, S. (2002). Plexin-Bl directly interacts with PDZ-RhoGEF/LARG to regulate RhoA and growth cone morphology. Neuron 35;51-63. Takahashi, T. Nakamura, F., Jin, Z., Kalb, R.G. and Strittmatter, S.M. (1998). Semaphorins A and E acts as antagonists of neuropilin-1 and agonists of neuropilin-2 receptors. Nature Neurosci.l; 487-93. Takahashi, T. and Strittmatter, S.M. (1999). Plexin-neuropilin-1 complexes form functional semaphorin-3A receptors. Cell 99; 59-69. Tamagnone, L., Artigiani, W., Chen, H., He, Z., Ming, G.L., Song, H.J., Chedotal, A., Winberg, M.L., Goodman, C.S., Poo, M.-m., Tessier-Lavigne, M. and Comoglio, P.M. (1999). Plexins are a large family of receptors for transmembrane, secreted and GPI-anchored semaphorins in vertebrates. Cell 99; 71-80. Taniguchi, M., Yuasa, S., Fujisawa, H., Naruse, I., Saga, S., Mishina, M. and Yagi, T. (1997). Disruption of Semaphorin III/D gene causes severe abnormality in peripheral nerve projection. Neuron 19; 519-30. Tanelian, D.L., Barry, M.A., Johnston, S.A., Le, T. and Smith, G.M. (1997). Semaphorin III can repulse and inhibit adult sensory afferents in vivo. Nature Medicine 3;1398-1401. Tang, S., Qiu, J., Nikulina, E. and Filbin, M.T. (2001). Soluble Myelin-associated glycoprotein released from damaged white matter inhibits axonal regeneration. Mol. Cell. Neurosci. 18; 259-69. Terman, J.R., Mao, T., Pasterkamp, R.J., Yu, H.-H. and Kolodkin, A.L. (2002). MICALs, a family of conserved flavoprotein oxidoreductases,function in plexin-mediated axonal repulsion. Cell 109; 887-900. Tessier-Lavigne M. and Goodman, CS. (1996). The molecular biology of axon guidance. Science 274; 1123-33. Toni, N. Buchs, P-A., Nikonenko, I., Povilaitite, P., Parisi, L. and Muller, D. (2001). Remodeling of Synaptic membranes after induction of long-term potentiation. J. Neurosci. 21; 6245-51. Wahl, S., Barth, H., Ciossek, T., Aktories, K. and Mueller, B.K. (2000). Ephrin-A5 induces collapse of growth cones by activating Rho and Rho kinase. J. Cell. Biol. 149; 263-270. Wallquist, M., Patarroyo, M., Thams, S., Carlstedt, T., stark, B., Cullheim, S. and Hammarberg, H. (2002). J. Comp. Neurol. 454; 284-93. Walter, J., Allsop, T. E., & Bonhoeffer, F. (1990). A common denominator of growth cone guidance and collapse? Trends Neurosci. 11, 447-452. 97 Watanabe, C, Kumanogoh, A., Shi, S., Suzuki, K., Yamada, S., Okabe, M., Yoshida, K. and Kikutani, H. (2001). Enhanced immune responses in transgenic mice expressing a truncated form of the lymphocyte semaphorin CD100. J. Immunol. 167; 4321-28. Winberg, M.L., Noordermeer, J.N., Tamagnone, L., Comoglio, P.M., Springs, M.K., Tessier-Lavigne, M. and Goodman, C.S. (1998a). Plexin A is a neuronal semaphorin receptor that controls axon guidance. Cell 95; 903-16. Winberg, MX., Mitchell, K.J. and Goodman, C.S. (1998b). Genetic analysis of the mechanisms controlling target selection: complementary and combinatorial function of netrins, semaphorins and IgCAMs. Cell 93; 581-91. Winberg, M.L., Tamagnone, L., Bai, J., Comoglio, P.M., Montell, D. and Goodman, C.S. (2001). The transmembrane protein Off-Track associates with plexins and functions downstream of semaphorin signaling during axon guidance. Neuron 32; 53-62. White, F.A. and Behar, O. (2000). The development and subsequent elimination of aberrant peripheral axon projections in Semaphorin3A null mutant mice. Dev. Biol. 225; 79-86. Wong, J. T., Yu, W. T. and O'Connor, T. P. (1997). Transmembrane grasshopper semaphorin I promotes axon outgrowth in vivo. Development 123; 3597-3607. Wong, J. T. Wong, S.T. and O'Connor, T.P. (1999). Ectopicsemaphorin-la functions as an attractive guidance cue for developing peripheral neurons. Nat. Neurosci. 2; 798-803. Yokozawa, Y., Tamai, H., Tatewaki, S., Tajima, T., Tsuchiya, T. and Kanzawa, N. (2002). Cloning and biochemical characterization of astacin-like squid metalloprotease. J. Biochem. 132; 751-58. Yu, H.-H., Araj, H.H., Ralls, S.A. and Kolodkin, A.L. (1998). The transmembrane semaphorin Sema I is required in Drosophila for embryonic motor and CNS axon guidance. Neuron 20; 207-20. Yu, H.-H., Huang, A.S. and Kolodkin, A.L. (2000). Semaphorin-la acts in concert with the cell adhesion molecules fasciclin II and connectin to regulate axon fasciculation in Drosophila. Genetics 156; 723-31. Zhou, F.-Q., Waterman-Storer, CM. and Cohan, C.S. (2002). Focal loss of actin bundles causes microtubule redistribution and growth cone turning. J. Cell. Biol. 5; 839-49. Zanata, S.M., Hovatta, I., Rohm, B. and Piischel, A.W. (2002). Antagonistic effects of Rndl and RhoD GTPases regulate receptor activity in Semaphorin 3 A-induced cytoskeletal collapse. J. Neurosci. 22; 471-77. 98 Appendix 1 Grasshopper Sema-2a Amino Acids Sequence * < . . .MET A A K L W N L L L V A A S V H L V G S V E Q L H Q D L I H E F S C G*H K Y Y R T F H L D E K R E S L Y V G A L D K V Y K L N L T N I S L S D C E R D S L T L E P T N I A N C V S K G K S A D F D C K N H I R V I O P M G D G S R L Y I C G T N A H S P K D W V V Y S N L T H L O R H E Y V P G I G V G | I A K C P F D P E D S S T A V W V E N G N P G D L P G L Y S G T N A E F T K A D T V I F R T D L Y N L T T G R R E Y SF K R T L K Y D S K W L D N P N F V G S F D V G E Y V L F F F R E T A V E Y I N C G K S V Y S R V A R V C K K D V G G K N I L S O N W A T F L K A R L N C S I P G E F P F Y F N E I O G V Y K M P N T D K F F G V F S T S V T G L T G S A I C S F T L K D I O E V F S G K F K E O A T S S S A W L P V L P S R V P D P R P G E C V N D T E L L P D T V L N F I R S H P L M D G A V S H E G G K P V F Y K R D V L F T O L V V D K L K V N L V G K N M E Y I V Y Y A G T S T G O V Y K V V O W Y D S G G L P O S L L V D I F D V T P P E P V O A L H L S K E Y K S L Y A A S D N I V ROIELVMCfff fJ? YSNCLOCARDP^YCG WD* R D S N S C K S Y T P G L L O D V T N T S A N L C E H S V M K K K L I V T W G O S I H L G C F L K V P E V L S S O T I S W V H Y T K D K G R Y P I V Y R P D K Y I E T S E H G L V L I S V T D S D S G R Y D C W L G G S L L C S Y N I T V D A H R C S A P G R S N D Y O K I Y S D W C H E F E R S K I A M K T W E R K Q A Q C S T K Q N N S N Q K T H P N D I F H S N P V A Stop Codon Bold: Signal sequence; double underlined: conserved semaphorin domain; highlighted: 70 amino acids region from cSema-3A sufficient to collapse dorsal roots ganglion; italic: hanatoxin homology sequence; underlined immunoglobulin domain. Sequence between arrow heads indicate the amino acids corresponding to the nucleotide primers used for the chimera construct. The forward primer annealed on the vector 5' to the start codon. 99 Appendix 2 Grasshopper Sema-la Amino Acids Sequence M R A A L V A V A A L L W V A L H A A A W V N D V S P K M Y V Q F G E E R V O R F L G N E S H K D H F K L L E K D H N S L L V G A R N I V Y N I S L R D L T E F T E O R I E W H S S G A H R E L C Y L K G K S E D D C O N Y I R V L A K I D D D R V L I C G T N A Y K P L C R H Y A L K D G D Y V V E K E Y E G R G L C P F D P D H N S T A I Y S E G O L Y S A T V A D F S G T D P L I Y R G P L R T E R S D L K O L N A P N F V N T M E Y N D F I F F F F R E T A V E Y I N C G K A I Y S R V A R V C K H D K G G P H O F G D R W T S F L K S R L N C S V P G D Y P F Y F N E I O S T S D I I E G N Y G G O V E K L I Y G V F T T P V N S I G G S A V C A F S M K S I L E S F D G P F K E O E T M N S N W L A V P S L K V P E P R P G O C V N D S R T L P D V S V N F V K S H T L M D E A V P A F F T R P I L I R I S L O Y R F T K I A V D O O V R T P D G K A Y D V L F I G T D D G K V I K A L N S A S F D S S D T V D S V V I E E L O V L P P G V P V K N L Y V V R M D G D D S K L V V V S D D E I L A I K L H RC.G.SDKIllNCREXJVJSi^ G K R_R_F I_Q N_I_S L_ G_ E_H_K A C_G G R P Q T E I V A S P V P T Q P T T K S S G D P V H S I H Q A E F E P E I D N E I V I G V D D S N V I P N T L A E I N H A G S K L P S S Q E K L P I Y T A E T L T I A I V T S C L G A L V V G F I S G F L F | S R R C R G E D Y T D M P F P D Q R H Q L N R L T E A G L N A D S P Y L P P C A N N K A A I N L V L N V P P K N A N G K N A N S S A E N K P I Q K V K K j T Y I stop codon. • « Bold: Signal sequence; double underlined: conserved Semaphorin domain; broken underlined: PSI domain; italic: hanatoxin homology domain; wave like underlined: transmembrane domain; highlighted: intracellular domain; underlined: putative PDZ binding domain. Sequence between arrow heads indicate the amino acids corresponding to the nucleotide primers used for the chimera construct. The reversed primer annealed on the vector 3' to the stop codon. 100 Appendix 3 Chimera amino acids sequence M A A K L W N L L L V A A S V H L V G S V E Q L H Q D L E Q K L I S E E D L G D L I H E F S C G H K Y Y R T F H L D E K R E S L Y V G A L D K V Y K L N L T N I S L S D C E R D S L T L E P T N I A N C V S K G K S A D F D C K N H I R V I Q P M G D G S R L Y I C G T N A H S P K D W V V Y S N L T H L Q R H E Y V P G I G V G I A K C P F D P E D S S T A V W V E N G N P G D L P G L Y S G T N A E F T K A D T V I F R T D L Y N L T T G R R E Y S F K R T L K Y D S K W L D N P N F V G S F D V G E Y V L F F F R E T A V E Y I N C G K S V Y S R V A R V C K K D V G G K N I L S Q N W A T F L K A R L N C S I P G E F P F Y F N E I Q G V Y K M P N T D K F F G V F S T S V T G L T G S A I C S F T L K D I Q E V F S G K F K E Q A T S S S A W L P V L P S R V P D P R P G E C V N D T E L L P D T V L N F I R S H P L M D G A V S H E G G K P V F Y K R D V L F T Q L V V D K L K V N L V G K N M E Y I V Y Y A G T S T G Q V Y K V V Q W Y D S G G L P Q S L L V D I F D V T P P E P V Q A L H L S K E Y K S L Y A A S D N I V R Q I E L V M C H H R Y S N C L Q C A R D P Y C G W D N V E L K C T A V G S P Z ) W S A GKKRFION ISL GEHKACGGRPOTEIVASP VP TO P TTKS S GDPVHSIHQAEF EPEIDNEIVIG VDDSNVIPNTLAEINHA GSKLPSSOEKLPIYTAE TL TI A IVTSCLGAL VVGFISGFLFSRRCRGEDYTDMPFPDQRH QLNRL TEA GLNADSPYLPPCANNKAAINL V L N V P P KN AN G K N A NSSAENKPIQK VKKTYIStop Normal: sequence signal; Bold: Sema-2a fragment (528 amino acids); underlined: sema-la fragment (215 amino acids); Italic: t-Sema-la construct (203 amino acids); double underlined: transmembrane domain; wave like underlined: intracellular domain. 101 Appendix 4 Representation of the gene splicing by overlap extension (SOEing) A. Amplification of the semaphorin domain from the full length semaphoring gene. 5' 3' a 3' a: forward primer;b: reverse primer B. Amplification of the 5'end (transmembrane and cytosplamic domains from the full length semaphorin-la gene. 5' 3' ! 3' 5' c : forward primer;d: reverse primer 5' y 3' C. Annealing of semaphoring 5'end fragment to semaphorin-la 3-end fragment. gSema-2a PCR fragment from A y 3' :' 3' y gSema-la PCR fragment from B J Elongation: PCR fragme nts act as prime rs D. Generation of the chimera \ Amplification 102 Appendix 5 Map of the pIZT expression vector Possibility of fusion proteins at the 3'end with V5 and His tags Cloning sites: _co Q ; C £ _ — = = O C C O CJ O O * - CO m O O C L O CO Q . O . O O - Q C CO ^ * U J C O C ^ U J U J 2 > < Q C O V 5 epitope CD 6 x H i s Promoter for cloned cDNA Appendix 6 Composition of all Solutions Agarose gel TAE IX 40mM tris acetate ImM EDTA 6 X loading buffer 0.25% bromophenol blue 0.25% xylene cyanol FF 30% glycerol in distilled water Aminolink Plus Column pH 10 Coupling buffer 0.1 M sodium citrate 0.05 M sodium carbonate, pH 10.0 pH 1.2 Coupling buffer see PBS Glycine buffer lOOmM glycine, pH 2.0 -3.0 GST column MT-PBS 150 mM sodium chloride 16 mM sodium phosphate monobasic 4 mM sodium phosphate dibasic pH 7.3 Elution buffer IX 50 mM Tris-HCl 10 mM Glutathione, pH 8.0 Hopper saline 5.00mM TES 165 mM sodium chloride 10.0 potassium chloride 4.00 mM calcium chloride 2.38 mM Magnesium sulfate 140 mM Sucrose Immunoblots 5 X loading buffer 47.5% distilled water 12.5% 0.5 M TrisHCl pH 6.8 10.0% Glycerol 20.0% 10% 0%) SDS 5.0% 2-mercaptoethanol 5% 1% (w/v) bromophenol blue Coomassie Blue 0.1% Coomassie blue R-250 40% methanol 10% acetic acid Running buffer 959. ImM Glycine 123.8 m M T r i s H C l 17.34 m M SDS, pH 8.3 Transfer buffer 191.8 m M Glycine 25.01 m M Tris 1 :5 final volume ethanol TBS 20 m M Tris 137mMNaCl ,pH7.6 TBST 20 m M Tris 137mMNaCl 0.01% Tween20,pH 7.6 PBS 137 m M sodium chloride 10 m M sodium phosphate dibasic, pH 7.2 PEM 0.1 M PIPES 2 m M E G T A 1 m M magnesium sulfate, pH 6.95 P B T I X PBS 0.1%Triton-X-100 0.1% BSA, pH7.2 RPM I 0.05 m M 20-hydroxyecdysone 0.1 mg/ml insuline (from bovin pancreas) 500 ul Supplement 0.4 m M Calcium chloride 0.4 m M Magnesium chloride 9.95 m M TES 9.86 m M Xylose, pH 6.9 before filtration Talon Resin Extraction/wash I X 50 m M Sodium phosphate dibasic 10 m M Imidazole 300 m M Sodium chlorine, pH 8.0 Elution buffer I X 50 m M sodium phosphate dibasic 300 m M sodium chlorine 150 m M imidazole, pH 7.0. 105 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items