Neurotrophin receptors: ligand-binding, activation sites. and allosteric regulation.

Transcription

1 Neurotrophin receptors: ligand-binding, activation sites and allosteric regulation. Ljubica Ivanisevic Department of Pharmacology and Therapeutics McGill University, Montreal November, 2007 A Thesis submitted to the Faculty of Graduate Studies and Research in partial fulfillment of the requirements for the degree of Doctor of Philosophy Copyright Ljubica Ivanisevic 2007

2 Abstract The Trk family of tyrosine kinase receptors and the common p75 NTR receptor are neurotrophin receptors. Nerve growth factor (NGF) binds TrkA, brain-derived neurotrophic factor (BDNF) binds TrkB, and neurotrophin-3 (NT-3) binds TrkC. The extracellular domain of the Trk receptor has five subdomains: a leucine-rich motif (D2), two cysteine-rich motifs (D1, D3) and immunoglobulin-like subdomains Ig-C1 (D4) and Ig-C2(D5). The Trk D4 subdomain regulates ligand-independent activation. The TrkA- D5 and TrkB-D5 subdomains regulate cognate ligand binding and Trk activation. However, the p75 NTR receptor binds all neurotrophins and regulates ligand affinity and Trk signals. We showed that p75 NTR affects Trk ligand - binding and activation of Trks by changing Trk subdomain utilization. When p75 NTR is coexpressed, NGF can activate TrkA via the cysteine-1 subdomain (D1), and BDNF can activate TrkB via leucine-rich motif (D2) and cysteine-2 (D3) subdomains. We hypothesized conformational or allosteric regulatory mechanisms. To further study the interactions between ligands and Trks, we examined TrkA binding to NT-3 as a heterologous ligand because these interactions are biologically relevant. We found the TrkA hot spot functional docking sites used by NT-3. We demonstrate that TrkA-D5 has partially overlapping but distinct binding and activation hot spots for both, NGF and NT-3. Moreover, ligand - binding studies have identified additional NT-3 binding/allosteric site on TrkA-D4. NT-3 binding to both sites induces full agonism. Conversely, the TrkA-D5 NT-3 binding site is partially agonistic, but antagonizes NGF activity. 2

3 Lasly, we address NT-3 binding and activation sites on the TrkC receptor by raising a monoclonal antibody that recognizes the juxtamembrane-linker domain of the TrkC receptor. This antibody is an artificial TrkC receptor agonist. The epitope of mab 2B7 defines a previously unknown hot spot of TrkC. Binding to this hot spot induces survival but not differentiation of TrkC expressing cells. Interestingly, the functional and structural availability of this hot spot is regulated by expression of p75 NTR ; and the ectodomain of p75 NTR masks the hot spot. This finding is important for understanding the molecular mechanisms underlying p75 NTR -Trk interactions. Taken together, results presented in this thesis define multiple ligand binding and activation domains on Trk receptors for cognate and heterologous ligands as well as ligand binding and activation sites regulated by p75 NTR, possibly by conformational or allosteric control. These findings have implications in understanding the molecular mechanisms of p75 NTR -Trk interactiona and rational design and development of Trk receptor artificial ligands with potential therapeutic applications. 3

4 Résumé La famille de récepteurs de Trk tyrosine kinase et le récepteur p75 NTR sont des récepteurs de neurotrophines. Le facteur de croissance nerveuse (NGF) intéragit avec le récepteur TrkA, le facteur neurotrophique dérivé du cerveau (BDNF) intéragit avec le récepteur TrkB et la neurotrophine-3 (NT-3) intéragit avec TrkC. Le domaine extracellulaire du récepteur Trk contient cinq sous-domaines: un motif riche en leucine (D2), deux motifs riches en cysteine (D1, D3) et des sous-domaines de type immunoglobuline Ig-C1(D4) et Ig-C2(D5). Le sous-domaine Trk D4 régule l activation indépendante de ligand. Les sous-domaines TrkA-D5 et TrkB-D5 régulent la liaison de ligands endogènes ainsi que l activation du récepteur Trk. Le récepteur p75 NTR intéragit avec toutes les neurotrophines et régule l affinité des ligands et les signaux issues de l activation du récepteur Trk. Par ailleurs, nous avons démontré que le p75 NTR affecte la liaison du ligand au récepteur Trk en changeant l activation de l utilisation des sousdomaines. Lorsque le recepteur de p75 NTR est coexprimé, le NGF peut activer le récepteur TrkA via le sous-domaine cysteine-1 (D1) et BDNF peut activer TrkB via le motif riche en leucine (D2) ainsi que via le sous-domaine cysteine-2 (D3). Nous avons examiné la liaison d un ligand hétérologue, NT-3 sur le récepteur TrkA afin d étudier plus profondément les interactions entre les ligands et le récepteur TrkA. Ces interactions sont biologiquement pertinentes. Pour faire ceci, nous avons tout d abord identifié les «points chauds» présents sur le récepteur TrkA qui servent des sites d amarrage fonctionnels du ligand NT-3. Nous avons démontré que le sous domaine TrkA-D5 possède deux points chauds distincts, notamment un point chaud qui sert comme le site d amarrage et d activation du NGF et un point chaud qui sert comme le 4

5 site d amarrage et d activation de la NT-3. Toutefois, ces deux sites d amarrage se chevauchent partiellement. De plus, nous avons identifié un site additionnel de liaison de NT-3 situé sur le sous domaine TrkA-D4. La liaison de NT-3 sur les deux sites (TrkA-D4 and TrkA-D5) induit un agonisme complet. Au contraire, la liaison du NT-3 sur le sous domaine TrkA-D5 seul entraîne l agonisme partiel de l activité de la NT-3 et un antagonisme de l activité du NGF. Finalement, nous avons étudié la liaison de la NT-3 sur le récepteur TrkC et l activation du récepteur TrkC. Pour faire ceci nous avons crée des anticorps monoclonaux qui reconnaissent le domaine juxtamembranaire-lieur du récepteur TrkC. Ces anticorps sont les agonistes artificiels du récepteur TrkC. Nous avons trouvé que l épitope de mab 2B7 possède un point chaud de TrkC, inconnu auparavant, qui lorsque activé par des ligands monomériques, induit la survie des cellules exprimant le récepteur TrkC et inhibe leur différentiation. De façon intéressante, la disponibilité de ce point chaud est régulée par l expression du récepteur p75 NTR. Cette découverte est importante afin de comprendre mieux les mécanismes moléculaires impliqués dans les interactions entre le récepteur Trk et le récepteur p75 NTR. En conclusion, ce travail avait pour but de caractériser les multiples domaines de liaison de ligands (endogènes et hétérologues) et d activation des récepteurs Trk ainsi que des sites de liaison de ligands et d activation régulés par p75 NTR, possiblement par un contrôle conformationnel ou allostérique. Les résultats obtenus par ce travail auront une implication importante dans la caractérisation des mécanismes moléculaires impliqués dans les interactions entre le récepteur Trk et le p75 NTR. Par ailleurs, les résultats auront une applicabilité médicale dans le développement des ligands artificiels de Trk ayant une application thérapeutique. 5

6 Contribution of Authors This thesis is written in a manuscript-based format. Each author s contributions to the manuscripts are described below. Chapter 2 Zaccaro M.C*., Ivanisevic L.*, Perez P., Meakin S.O., and Saragovi H.U p75 NTR co-receptor regulates ligand-dependent and ligand-independent Trk receptor activation, in part by altering Trk docking subdomains J. Biol. Chem. 276: The candidate designed and performed most of the experiments. Maria Clara Zaccaro is responsible for western blot detecting surface expression of subdomains of the extracellular domain of TrkA. Binding assays were done together with Maria Clara Zaccaro and H. Uri Saragovi. The initial outline of the study objectives, development of the experimental approach, data analysis, preparation and submission of the manuscript were done together with Maria Clara Zaccaro and Dr. H. Uri Saragovi. Chapter 3 Ivanisevic L., Zheng W., Woo S.B., Neet K.E. and Saragovi H.U TrkA receptor hot spots for binding of NT-3 as a heterologous ligand. The candidate performed and designed most of the experiments and participated in the initial outline of the study objectives and data analysis. Dr. Wen-Hua Zheng is responsible for detection of NT-3 competition of NGF-NGF30 binding on purified protein fragments. Dr. Sang B. Woo and Dr. Kenneth E. Neet are responsible for experiments characterizing the binding of NGF and NT-3 to the extracellular domains of TrkA and 4.1 TrkA/B chimera. The preparation and submission of the manuscript was done together with Dr. Kenneth E. Neet and Dr. H. Uri Saragovi. 6

7 Chapter 4 Guillemard V.*, Ivanisevic L.*, Shoelten V. and Saragovi H.U An agonistic anti-trkc mab directed to the juxtamembrane ectodomain defines a functional hot spot interacting with p75 NTR co-receptors. The candidate designed and performed most of the experiments and participated in the initial outline of the study objectives and data analysis. Dr. Veronique Guillemard and Dr. H. Uri Saragovi raised anti-trkc monoclonal antibody 2B7. Screening of hybridomas that resulted in identification of mab 2B7 was done by Dr. Veronique Guillemard and Vicki Shoelten. Dr. Veronique Guillemard performed survival experiments with mab 2B7 Fabs. Preparation and submission of the manuscript have been done together with Dr. H. Uri Saragovi. * indicates equal contribution by the authors. 7

8 Original contributions to science 1) Demonstrated that Trk receptors have two different ligand activation sites, one independent and one dependent on p75 NTR co-expression. 2) Demonstrated that p75 NTR regulates Trk receptor ligand binding and activation sites, possibly by allosteric or conformational control. 3) Demonstrated that the D4 subdomain of Trk receptors is important for ligandindependent activation. 4) Demonstrated two NT-3 binding/activation sites, one on the D5 domain and another on D4 domain of TrkA. 5) Demonstrated that activation of the NT-3 binding site on D5 subdomain of TrkA induces survival, and that activation of both NT-3 binding sites is necessary for full survival. 6) Identified a novel TrkC receptor hot spot that can be activated by monomeric ligands and whose activation induces trophic signals. 7) Documented that the novel TrkC hot spot is potentially involved in p75 NTR regulation of TrkC binding and signaling. 8

9 Acknowledgments I would like to thank my supervisor, Dr. H. Uri Saragovi, for his availability and guidance from the initial development of my project to the writing of the thesis. His mentorship and scientific expertise have been crucial to the successful progress and completion of my studies. I would like to thank my advisor Dr. Guillermina Almazan for her support, guidance and help. I am grateful to my thesis advisory committee members Dr. Brian Colier and Dr. Derek Bowie for their valuable input and discussions of my project. I would like to emphasize the role of the colleagues and co-authors Maria-Clara Zaccaro, Dr. Pilar Perez, Dr. Susan O. Meakin, Dr. WenHua Zheng, Dr. Sang B. Woo, Dr. Kenneth E. Neet, Dr. Veronique Guillemard and Vicki Shoelten whose contributions made successful the studies presented in this thesis. I am grateful for the PhD scholarships received from the Montreal Center for Experimental Therapeutics in Cancer, the Maysie McSporran Foundation and the McGill Faculty of Medicine. I would like to thank Milena Crosato, Pooja Jain and Muriel Bassili for help with the preparation of this manuscript. I am grateful to all the past and present members of the laboratory for creating a stimulating working environment and their support. My special thanks go to Martin Gagnon for his help, advice and friendship. My friends Tanja Ducic and Dejana Braithwaite were supporting, encouraging and believeing in me, even from thousands of kilometers away. My very special thanks go to Marina Stojanovic and her family for always being there for me, for supporting me, making me smile and being not only my friends but my second family. Ova doktorska teza je posvećena mojim roditeljima, bez čije podrške i pomoći ne bih uspjela. 9

10 Table of contents Abstract 2 Résumé 4 Contribution of Authors 6 Original contributions to science 8 Acknowledgements Chapter 1 GENERAL INTRODUCTION Introduction Neurotrophins Nerve growth factor gene and mrna Neurotrophin processing and secretion Neurotrophins structure Neurotrophin receptors Trk receptors structure Trk receptor binding sites Axonal transport of neurotrophin receptors Role of Trk receptor turnover and ubiquitination in signaling Trk receptor signaling Ras/MAPK pathway PI-3K-Akt pathway PLCγ pathway Transactivation of Trk receptors p75 NTR receptor-structure 40 10

11 1.7 p75 NTR signaling and functions p75 NTR as a co-receptor p75 NTR and sortilin p75 NTR and Nogo p75 NTR and Trk Neurotrophins role in pathology and clinical prospects Chapter 2 p75 NTR Co-receptors Regulate Ligand-dependent and Ligandindependent Trk Receptor Activation, in Part by Altering Trk Docking Subdomains Rational Abstract Introduction Materials and Methods Results Discussion Chapter 3 TrkA receptor «hot spots» for NT-3 as a heterologous ligand Rational Abstract Introduction Materials and Methods Results Discussion

12 4.0 Chapter 4 An agonistic anti-trkc mab directed to the juxtamembrane ectodomain defines a functional hot spot interacting with p75 NTR co-receptors Rational Abstract Introduction Materials and Methods Results Discussion GENERAL DISCUSSION Trk receptor binding/activation sites for homologous and heterologous ligands P75 NTR regulates Trk ligand activation sites and ligand-dependent Trk signals via potentially allosteric mechanisms Summary REFERENCES

13 Chapter 1 General Introduction

14 1.1 Introduction The neurotrophins are a family of polypeptide factors that are important for nervous system development and function in adulthood. Their function spans from regulating the development, maintenance, growth, survival and differentiation of neurons to regulating synaptogenesis and activity-dependent forms of synaptic plasticity. They also mediate higher-order activities, such as learning, memory and behavior. The first neurotrophin to be identified was nerve growth factor (NGF). NGF was discovered in the 1950s by Rita Levy Montalcini and Viktor Hamburger while studying the developing nervous system and the peripheral structures it innervates. They identified a growth-promoting factor essential for the survival and maintenance of sensory and sympathetic neurons, on the basis of a functional assay that utilized a mouse sarcoma implanted on the body wall of chick embryos in proximity to the spinal cord. The sarcoma secreted a soluble factor that enabled fiber outgrowth and hypertrophy of sensory neurons from dorsal root ganglia, but not motor neurons (Levi-Montalcini and Hamburger, 1951). In vitro experiments by Cohen lead to the identification of a low molecular weight nerve growth-promoting protein that was later cloned and identified as NGF (Cohen et al., 1954). From this work the Neurotrophic Factor Hypothesis was developed, stating that developing neurons reaching a target organ are in competition for limited supplies of trophic factors - those that make appropriate connections and obtain growth factors live, and the one that do not die. Therefore, competition for trophic factors determines the number of surviving neurons during target innervation. Following the discovery of NGF and its physiological roles in the central and peripheral nervous system, other neuron 14

15 survival promoting neurotrophic factors were identified by purification and sequencing - brain derived neurotrophic factor (BDNF), and cloned - neurotrophin-3 (NT-3) and neurotrophin-4 (NT-4). Neurotrophins and their receptors are expressed during development, adult life and aging in many cell types in the central nervous system (CNS) and peripheral nervous system (PNS), and other non-neuronal cell types. The expression pattern of the neurotrophins during development and in adults, in the PNS and CNS, is consistent with a target-derived mode of action and retrograde transport of the neurotrophins from the axon terminal to the cell body. On the other hand, in the CNS, the expression patterns of NTFs and their receptors indicate not only retrograde transport but also anterograde transport (from cell body to axon) (von Bartheld et al., 1995; Conner et al., 1997). Different cells of the PNS and CNS express NGF mrna, including cells in the dorsal root ganglia (DRG), cerebral cortex and hippocampus. The target tissues of sympathetic neurons, such as heart and spleen, also express NGF. Epithelial cells, smooth muscle cells, fibroblasts and male germ cells express NGF mrna. BDNF mrna is expressed in DRG neurons, nodose ganglia, inner ear, superior cervical ganglia (SCG), cerebral cortex and hippocampus. NT-3 has a similar expression pattern to BDNF, but it is also expressed in the cerebellum. Neurotrophins elicit their function by binding to two families of neurotrophin receptors a common p75 NTR receptor and a selective and specific Trk tyrosine kinase receptor. The Trk tyrosine kinase receptor family consists of three members TrkA, TrkB and TrkC. TrkA receptors are expressed in sensory ganglia, superior cervical ganglia, DRG (dorsal root ganglia), cerebral cortex, striatum, retinal ganglia and basal 15

16 forebrain cholinergic neurons. TrkC receptors are expressed in DRG, retinal ganglia, sympathetic ganglia, nodose and vestibular ganglia, spiral ganglia, superior cervical ganglia, cerebral cortex, hippocampus, cerebellum, thalamus and hypothalamus. TrkB receptors are expressed in DRG, retinal ganglia, nodose and vestibular ganglia, spiral ganglia, cerebral cortex, hippocampus, basal forebrain cholinergic and GABAergic neurons, substantia nigra dopaminergic neurons, cerebellum and hypothalamus. p75 NTR receptors are expressed in DRG, sympathetic ganglia, nodose ganglia, superior cervical ganglia, retinal ganglia, cerebral cortex, striatum, hippocampus, basal forebrain cholinergic neurons, cerebellum, thalamus and hypothalamus (Barbacid, 1994; Pitts and Miller, 1995; Miller and Pitts, 2000). DRGs contain the cell bodies of the neurons responsible for nociception (perception of pain, temperature, chemical and mechanical injury) and proprioception (the sense of positional information of limbs in space and their coordinate movements) in the trunk and limbs. Trigeminal neurons are responsible for sensory innervation in the head, teeth and the oral and nasal cavities. The nodose and vestibular ganglia contain sensory neurons that innervate the visceral organs that sense nociception and the inner ear organs sensing motion/ hearing. Trk and p75 NTR receptors, as well as neurotrophins, are also expressed on non-neuronal cell lines mast cells, bone marrow, activated T and B cells, insulin producing pancreatic cells, melanocytes and Shwann cells, oligodendrocytes and oligoglia, and retinal cells. The neuroprotective function of neurotrophins and their function in regeneration of injured neurons during adulthood makes them a very good target for treatment of CNS disorders (Alzheimer s disease, ischemia, brain trauma) and peripheral neuropathies. 16

17 Neurotrophins and their receptors are also potential targets for cancer therapy. A Trk oncogene, formed by genetic rearrangement, is present in non-neuronal cancers like colon and papillary thyroid cancer. On the other hand, Trk receptors, which normally regulate growth, differentiation and apoptosis of immature neural crest cells, are associated with tumors of neuronal origin - neuroblastoma, medulloblastoma, and prostate cancer (Eide et al., 1993). Early stage less malignant neuroblastomas correlating with good patient survival have high TrkA expression, while unfavorable outcome, aggressive neuroblastomas have high expression of full length TrkB and its cognate ligand BDNF. The levels of TrkA expression in neuroblastoma (Nakagawara et al., 1993) and the levels of TrkC expression in medulloblastomas (Segal et al., 1994) correlate with good prognosis. However, in prostatic carcinoma it is an autocrine NGF-TrkA loop that is responsible for tumor progression (Delsite and Djakiew, 1996). The role of p75 NTR in these malignancies is poorly studied. Recently the correlation between higher invasive potential of medulloblastoma and high ratio of p75 NTR -TrkC expression has been established (Sinnappah-Kang et al., 2006). The important physiological functions of neurotrophins and their receptors, the fact that some ligands cross-activate p75 NTR and Trk receptors, as well as their implication in pathological states, underline the great importance of research towards defining and understanding Trk receptor ligand binding and functional receptor activation sites, how these binding sites are regulated by p75 NTR and how their signals are transmitted into different biological outcomes. 17

18 1.2 Neurotrophins Nerve growth factor gene and mrna Neurotrophin genes are located on different chromosomes. The human NGF gene is located on chromosome 1. The gene encoding BDNF is located on human chromosome 11, the NT-3 gene is located on human chromosome 12 and the NT-4 gene is located on human chromosome 19. The NGF gene is more than 43 kbp long and contains four exons: exons IA and B, exon II, exons IIIA and B, and exon IV. Part of exons IA, II, IIIB and IV code for the NGF precursor. Exon IV codes for mature NGF transcripts (Seidah et al., 1996a). There are four different NGF transcripts, with transcripts A and B being the most abundant and having a defined expression pattern in different tissues. Transcription initiation and RNA processing regulate the NGF gene. Various cells of the peripheral and central nervous system express NGF mrna. The highest levels of NGF mrna are present in tissues densely innervated by sympathetic neurons, such as heart and spleen. NGF mrna expressing cells are found in all areas of the hippocampus, and almost exclusively in neurons Neurotrophin processing and secretion Neurotrophins are synthesized as 31-35kDa preproneurotrophin precursors. The pre-mrna sequence codes for a signal sequence that directs protein synthesis to endoplasmatic reticulum (ER) ribosomes. Upon synthesis, the protein is sequestered in the ER, where the 18 amino acid signal peptide is cleaved, forming a spontaneous noncovalently linked pro-ngf homodimer. In the ER, the prodomain is N-glycosilated affording stability to the pro-ntfs. There are three possible fates for these intracellular 18

19 neurotrophins intracellular cleavage followed by secretion, secretion followed by extracellular cleavage or secretion of proneurotrophins without subsequent cleavage (Lu et al., 2005). Proneurotrophins can be cleaved extracellularly by matrix metalloproteinases (MMPs) or plasmin, and intracelullarly by furin or protein convertase 1 (PC1) giving rise to mature neurotrophins. The intracellular cleavage of pro-ngf at the Arg-Ser-Lys/Arg-Arg sequence produces mature NGF (Figure 1) Figure 1 The structure of pre-prongf. Pro-neurotrophins can be secreted via the constitutive or regulated secretory pathways, and transported to the appropriate subcellular compartment. Pro-neurotrophins are secreted via the constitutive secretory pathway in non-neuronal cells such as smooth muscle cells, fibroblasts and astrocytes, while neurons secrete pro-neurotrophins mostly via the regulated secretory pathway. The pro-ngf form is up-regulated after brain injury, Alzheimer s disease, and retinal dystrophy. Pro-NGF is mostly processed via the constitutive secretory pathway (Seidah et al., 1996b; Mowla et al., 1999). However, regulated release of NGF has been described in the primary cultures of hippocampal neurons (Blochl and Thoenen, 1995). Similarly, NT-3 is released via regulated secretory pathway in chicken retinal ganglion cells (Wang et al., 2002b). Others have reported that pro-nt-3 can be sorted either through the regulatory or constitutive secretory pathway (Farhadi et al., 2000), depending on NT-3 expression levels or the availability of specific enzymes within the cell. Heterodimers of NT-3 and BDNF have been shown to be released via the regulated secretory pathway in transfected neuroendocrine cells (Hibbert 19

20 et al., 2003). Pro-BDNF is processed via the regulatory secretory pathway, and secreted in an activity-dependent manner (Mowla et al., 1999). The pro-sequence of BDNF plays an important role in activity-dependent release, as Val-Met mutation in the pro-bdnf region impairs activity-dependent secretion without affecting its constitutive secretion (Egan et al., 2003). As opposed to BDNF, which is secreted in the pro-form from hippocampal neurons, NGF is secreted mainly in the mature form (Seidah et al., 1996b; Mowla et al., 1999). It is important to note that conflicting data arises when the model systems (cell lines versus primary neurons) and the experimental approach (metabolic labeling followed by immunoprecipitation versus neurotrophin elisa and immunocytochemistry) is different. More recently, it has been shown that mature neurotrophins are released via the regulated pathway in primary rat cortical neurons. No differences between pro and mature NGF secretion were noticed in this model system (Wu et al., 2004) Neurotrophin structure The members of the neurotrophin family in mammals include nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3) and neurotrophin-4 (NT-4). Other growth factors that are structurally unrelated to the neurotrophins basic fibroblast growth factor (bfgf) and glial derived neurotrophic factor (GDNF) also have neurotrophic activity. The first neurotrophic factor to be discovered and purified was NGF (Cohen et al., 1954) and subsequently BDNF, NT-3 and NT-4 were discovered and characterized. All the neurotrophins have a similar overall structural scaffold, with a common structural feature of a cysteine knot formed by three intertwined disulphide bridges, and a 20

21 high degree of sequence homology (~50%). The crystal structure of NGF was determined first (McDonald et al., 1991), and later the structures of BDNF, NT-3 and NT-4 dimers were determined (Robinson et al., 1999). In their biologically active form, all neurotrophins are homodimers, formed by two monomers arranged in a parallel manner. The NGF monomer has an elongated structure, with the central part of the molecule formed by two pairs of antiparallel β strands, three hairpin loops on one end and a cysteine-knot motif on the other end. Mature neurotrophins have a very conserved hydrophobic core that defines the neurotrophin fold. This accounts for the ability of neurotrophins to form mixed heterodimers with other neurotrophins (Robinson et al., 1995). Four β-hairpin loop regions on the surface of the dimer contain a large proportion of residues that have high variability between the members of the neurotrophin family they define the specificity of each neurotrophin ligand for its cognate receptor (Ilag et al., 1994). The aminoacids His4, Pro5, Phe7, His8, Val48, Pro49 and Gln96 and aminoacids from N- and C- termini mediate NGF-TrkA binding. Conversely, NGF binding to the p75 NTR receptor is mediated by the positively charged residues Lys32, Lys34 and Lys95, (Drinkwater et al., 1993; Woo and Neet, 1996; Kruttgen et al., 1997; Kullander et al., 1997), and also by the reverse turn residues Asp72, Lys74 and His 75 (Ryden and Ibanez, 1997). The cysteine knot motif, formed by three intrachain disulfide bridges, stabilizes the neurotrophin fold and locks the neurotrophin molecule in its active conformation. 21

22 1.3 Neurotrophin receptors Neurotrophins bind to the Trk tyrosine kinase family of receptors and to the p75 NTR common neurotrophin receptor. The first neurotrophin receptor to be cloned and identified was the p75 NTR receptor (Johnson et al., 1986). It was later discovered that it binds all members of the neurotrophin family (NGF, BDNF, NT-3 and NT-4). The p75 NTR receptor can also signal independently of ligand, leading to modulation of apoptosis, survival, and axonal growth. The functions of p75 NTR depend on cell type, developmental stage, and whether the receptor is liganded or not. TrkA was initially discovered as an oncogenic fusion protein from colon carcinoma and was considered an orphan receptor until it was discovered that NGF induced its rapid tyrosine phosphorylation in PC12 cells. The normal cellular Trk tyrosine kinase receptor is a single-pass transmembrane receptor that is highly expressed in the nervous system. Trk receptor functions span from regulating the development, maintenance, growth, survival and differentiation of neurons to regulating synaptogenesis and activity-dependent forms of synaptic plasticity. There are three members of the Trk receptor family - TrkA, TrkB and TrkC, which are encoded by separate genes located on human chromosomes 1q, 9q and 15q, respectively. TrkA receptor was first identified as a receptor for NGF (Kaplan et al., 1991); TrkB is a receptor for BDNF (Klein et al., 1991) and NT-4 (Klein et al., 1992), and TrkC is the receptor for NT-3 (Lamballe et al., 1991). NT-3 can also bind and activate TrkA and TrkB, but with lower affinity. Different splice variants of Trk receptors exist. Insert variants with insertions in the extracellular domain have been described for TrkA (Barker et al., 1993) and TrkB (Strohmaier et al., 1996) receptors and they may have effects on ligand interactions. 22

23 Furthermore, TrkC splice variants exist with 14 or 39 amino acid inserts within their tyrosine kinase domain, which can disrupt TrkC signaling (Lamballe et al., 1993; Tsoulfas et al., 1993). On the other hand, TrkB and TrkC receptors lacking a tyrosine kinase domain exist as a consequence of differential splicing of exons coding for the intracellular portion of these receptors. Ratios of full-length isoforms versus kinase deficient isoforms may vary at different stages of development and can be differentially distributed to cellular compartments (Ichinose and Snider, 2000). It has been thought that the function of these nonkinase-containing isoforms of TrkB and TrkC is to inhibit productive dimerization and activation of full-length receptors, or bind excess neurotrophins and therefore decrease response to neurotrophins (Biffo et al., 1995; Ninkina et al., 1996). Even though most neurotrophin functions are modulated through full length Trk receptors, truncated TrkB and TrkC isoforms can modulate intracellular pathways more directly. The TrkC kinase deficient isoform, with participation of the p75 NTR co-receptor, promotes differentiation but not survival of neuronal crest cells (Hapner et al., 1998). More recently, it was shown that the truncated TrkB receptor, when interacting with p75 NTR, modulates the formation of dendritic filopodia of hippocampal neurons (Hartmann et al., 2004). Furthermore, kinase inactive truncated TrkB receptor regulates Ca 2+ release from internal stores upon BDNF treatment in astrocytes (Rose et al., 2003). Kinase deficient TrkC receptor can also signal in response to NT-3, inducing membrane ruffling via interaction with the scaffold protein tamalin and activating Arf6- Rac signaling (Esteban et al., 2006). Truncated TrkB receptors are up - regulated in trisomic mouse models (Dorsey et al., 2006). In the animal model of glaucoma, kinase deficient TrkC receptors are overexpressed when retinal ganglion cells (RGCs) undergo 23

24 apoptosis (Rudzinski et al., 2004). These novel data point to the importance of truncated TrkB and TrkC isoforms, not only in normal cell physiology, but also in pathological conditions Trk receptor structure Trk receptors are 140 kda (~800 aminoacids) single-pass transmembrane receptors, with an extracellular ligand-binding domain, a transmembrane domain and an intracellular tyrosine kinase domain. The extracellular domain is a ligand-binding domain; it has 11 glycosilation sites with 40kDa of carbohydrate moiety. The extracellular domain of Trk receptors has a characteristic subdomain organization, based on sequence analysis. On the N-terminus, there is one cysteine-rich subdomain Cys-1 (also known as D1) separated from the second cysteine rich subdomain Cys-2 (also known as D3) by three consecutive leucine-rich motifs (D2). The C-terminal region of the extracellular domain has two immunoglobulin (Ig) like subdomains, Ig-C1 and Ig-C2 (also termed D4 and D5, respectively) (Fig. 2) (Perez et al., 1995). Recombinant protein containing TrkA-D4-D5 subdomains binds NGF with the same affinity as the intact, membrane-bound TrkA receptor (Holden et al., 1997). Mutant protein and domainswapping studies identified the Ig-C2 (D5) subdomain as ligand-binding and this was confirmed by crystallography (Wiesmann et al., 1999). The D4 subdomain has a role in regulating receptor dimerization and constitutive receptor activation, probably as a hinge region (Arevalo et al., 2001; Zaccaro et al., 2001). Even though the extracellular domains of Trk family member receptors have the same subdomain organization, there is only a ~ 50% homology between their extracellular domains. This accounts for the specificity and selectivity in the ligand-binding of different Trk receptors. 24

25 Figure 2 Structure of Trk receptors and their extracellular domain (ECD) organization. TM stands for transmembrane domain. Crystal structures of TrkA-D5, TrkB-D5 and TrkC-D5 were determined (Ultsch et al., 1999). They show Ig-like folds, with two β (beta) sheets consisting of strands A, B, E, D and C, F and G, respectively. A single disulphide bond links strands B and E of the same sheet. The crystal structures of TrkA-D5 bound to NGF (Wiesmann et al., 1999) (Figure 3) and NT-4 bound to TrkB-D5 were determined (Banfield et al., 2001). There are two main areas of NGF/TrkAD5 interaction: a conserved patch and a specific patch. The conserved patch consists of residues from the central portion of the NGF dimer in contact with the AB, C D and EF strands of beta-sheets of the C-terminal of the TrkA-D5 domain. The specific patch consists of the N-terminus of one NGF molecule binding to the ABCD strands of TrkA-D5 beta sheets (Figure 3). 25

26 Figure 3 Ribbon diagram representing the crystal structure of the complex between NGF and D5 subdomain of TrkA (Wiesmann and de Vos, 2001) Most of the NGF residues shown to be important for p75ntr binding are positively charged and located on loops Ll, L3, and L4; More recently, the crystal structure of TrkA-ECD in complex with NGF was resolved (Wehrman et al., 2007)(Figure 4). It was modeled based on the previously determined crystal structure of TrkAD5-NGF (Wiesmann et al., 1999) and it confirms the TrkA-D5 subdomain as an NGF-binding domain on TrkA-ECD. In addition, based on the crystal structure of NGF-TrkA-ECD complex, there are no conformational changes predicted on TrkA-ECD upon NGF binding (Wehrman et al., 2007). However, the crystal structure of unliganded TrkA-ECD is not available and it is necessary to confirm the lack of conformational changes induced by NGF binding. 26

27 Figure 4 Schematic representation of the crystal structure of TrkA-ECD in complex with NGF (Wehrman et al., 2007) The lipophilic transmembrane domain linking the extracellular and intracellular domains stabilizes the receptor on the cell surface and plays an important role in ligand activation of the receptor. Exchanging the TrkA transmembrane domain with the same region of the TNF2 receptor yields nonfunctional receptors, suggesting a role for the transmembrane region in receptor dimerization and autophosphorylation (Canossa et al., 1996). The intracellular part of Trk receptors has a protein tyrosine kinase activity. Its function is regulated by phosphorylation of specific tyrosines that form an activation loop. (Cunningham et al., 1997). Phosphorylation of other tyrosine residues serves for binding of specific adaptor proteins. The Trk family members have highly homologous intracellular domains, with ~ 75% sequence identity. The sites of phosphorylation and the nature of the pathways activated by the Trks are very similar (Yuen and Mobley, 1999; Atwal et al., 2000). 27

28 1.3.2 Trk receptor binding sites Cells expressing TrkA alone display a small number of high-affinity NGF binding sites (K d =10-12 M), but the majority of sites have intermediate binding affinity (K d = M) with very slow rates of association. However, co-expression of p75 NTR with TrkA receptors leads to the formation of NGF high-affinity binding sites (K d =10-12 M) and a faster association rate (Hempstead et al., 1991; Mahadeo et al., 1994). More recently it was identified that the intracellular and transmembrane domains of p75 NTR and TrkA are both important for formation of high affinity binding sites (Esposito et al., 2001). The formation of high affinity binding sites also depends on the ratio of cell surface p75 NTR and Trk receptors, with a higher procentage of high affinity binding sites being formed when p75 NTR and Trk receptors are present in near equimolar ratios (Esposito et al., 2001). The p75 NTR receptor binds all mature neurotrophins with low affinity (K d =10-9 M) but with different kinetics (Rodriguez-Tebar et al., 1992; Dechant and Barde, 1997). High affinity binding of NT-3 to the p75 NTR receptor has also been reported (Dechant and Barde, 1997). More recent studies have shown high affinity binding of pro-ngf and pro- BDNF to p75 NTR (K d = ). Furthermore, p75 NTR creates a complex with sortilin to create high-affinity binding sites for pro-ngf and pro-bdnf. The pro-domain of NGF interacts with the cysteine-rich region of the sortilin extracellular domain, while the mature NGF interacts with p75 NTR (Nykjaer et al., 2004). Pro-neurotrophins are more efficient in activation of p75 NTR apoptotic signaling than mature neurotrophins with pro-ngf inducing apoptosis in neuronal cells, oligodendrocytes, smooth muscle cells and sympathetic neurons (Lee et al., 2001) and 28

29 pro-bdnf in sympathetic neurons co-expressing p75 NTR and sortilin (Teng et al., 2005). p75 NTR therefore forms high affinity binding sites for both mature NGF (in complex with Trk receptors) and prongf (in complex with sortilin). Figure 5 Neurotrophin receptors and their ligand specificity. 1.4 Axonal transport of neurotrophin receptors During development, retrogradely transported NGF promotes the survival of appropriately connected neurons; retrograde transport also has a key role in neuronal survival and plasticity. Trk receptors are internalized via ligand-dependent endocytosis upon ligand binding. The internalized receptors are then sorted and can undergo either recycling or be shuffled to retrograde transport pathways. Therefore, receptor internalization and intracellular transport plays an important role in neurotrophin signaling. Internalization serves as a negative feedback mechanism in the regulation of receptor number on the cell surface. In the case of neurotrophins, a target tissue in the periphery often secretes them, and the signals have to be transported along the axon to the cell body. This retrograde action is carried out by receptor internalization and transport of the receptor-ligand complexes (Grimes et al., 1997), reviewed by (Mufson et al., 1999). Most of the internalized NGF appears to be targeted to lysosomes and is rapidly degraded (Ure and Campenot, 1997). However, BDNF and NT-3 can avoid the degradation pathway and can be targeted for packaging in vesicles for anterograde transport, after 29

30 internalization through an unknown pathway of subcellular trafficking. This pathway likely includes sorting in endosomes, an endocytic pathway to the Golgi system, followed by packaging in vesicles for anterograde transport (Michael et al., 1997). Differential sorting of neurotrophins reflects their complex and highly diverse functions in promoting neuronal survival, differentiation, dendritic growth and synaptic plasticity (Butowt and von Bartheld, 2001). Retrograde signaling of Trk receptors was extensively studied. Present evidence supports the signaling endosome hypothesis of Trk receptor retrograde signaling. Upon NGF induced Trk activation at the presynaptic terminals, the NGF-TrkA complex is internalized into endosomes that are retrogradely transported along microtubules to the cell body. Signaling endosomes are clathrin coated vesicles formed by internalization of the NGF-TrkA receptor complex; they are transported along microtubules by dynamin motor components. There is also evidence of caveolin and Pincher mediated endocytosis of the TrkA-NGF complex (Zweifel et al., 2005). Signaling endosomes at the presynaptic terminal and cell body contain not only activated Trk and NGF, but also activated PI- 3K/Akt, pmapk, and pcreb (Riccio et al., 1997; Kuruvilla et al., 2000; Howe and Mobley, 2005). This implicates retrograde transport of activated Trk-NGF in the transmission of neurotrophin mediated signaling. Blocking endocytosis or dynein inhibition blocks the retrograde transport of activated Trks and signaling components to the cell body, and inhibits Trk receptor mediated survival (Ye et al., 2003; Heerssen et al., 2004). The activation of specific signaling pathways by Trk receptors is regulated by membrane trafficking, because adaptors and signaling molecules are localized to specific 30

31 membrane compartments. The importance of the receptor location in determining the functional outcome has been shown by NGF-NGF30 ligand that is fast internalizing and induces survival but not differentiation of TrkA expressing cells (Saragovi et al., 1998). In experiments with thermosensitive dynamin, which was used to reversibly inhibit ligand-receptor internalization, survival responses were not affected, but differentiation responses were strongly inhibited by reduction in endocytosis. Hence, receptor internalization is necessary for differentiation, while the receptors on the cell surface regulate survival (Zhang et al., 2000). The conclusion of these studies is that Trkmediated signaling is controlled by the specificity and kinetics of membrane transport. Also different ligands acting via the same receptor can induce different biological outcomes. NGF and NT-3 acting via TrkA receptors induce axonal growth, but only NGF induces the survival of developing sympathetic neurons. Moreover, NT-3 does not induce TrkA internalization or retrograde signaling (Kuruvilla et al., 2004). Axonal transport and retrograde signaling of Trk receptors is understood more than that of p75 NTR receptors. The kinetics of p75 NTR internalization are much slower than Trk internalization kinetics (Bronfman et al., 2003). Different internalization kinetics indicate that p75 NTR and Trk receptors are retrogradely transported independently. p75 NTR has also been shown to signal anterogradely upon NT-3 binding in retinal ganglion cell (RGC) axons (Butowt and von Bartheld, 2001) Role of Trk receptor turnover and ubiquitination in signaling The turnover of Trk receptors is an ongoing process and is one of the mechanisms that regulate the extent of Trk receptor signaling. Trks can undergo two different pathways recycling back to the plasma membrane or the degradative pathway to the 31

32 lysosomes. The degradative pathway is characterized by decreased responsiveness to the ligand and the down-regulation of total receptor numbers on the surface. TrkA and TrkB receptors both can undergo the degradative pathway (Knusel et al., 1997; Sommerfeld et al., 2000). It has been shown that TrkA receptors recycle back to the cell surface in a ligand-dependent manner, while TrkB receptors are predominantly sorted to the degradative pathway (Chen et al., 2005). Moreover, there are reports of ligand dependent Trk receptor ubiquitination (Geetha et al., 2005; Makkerh et al., 2005). These reports indicate that p75 NTR affects Trk receptor ubiquitination, but with contradictory data. p75 NTR co-expression was shown to negatively regulate Trk receptor ubiquitination, impairing its internalization and degradation (Makkerh et al., 2005). On the other hand, a report by Geetha et al indicates that p75 NTR positively regulates the ubiquitination of TrkA. More work should be done to understand the Trk receptor s ligand dependent ubiquitination and the role of p75 NTR in regulating it. 1.5 Trk receptor signaling Activation of Trk receptors by neurotrophins results in cell survival, differentiation, axonal growth, axonal guidance, regulation of synapse formation and synaptic plasticity. The first step in Trk receptor activation is ligand (neurotrophin) induced receptor dimerization followed by the transphosphorylation of tyrosine residues in the activation loop (Y670, Y674 and Y675 in human TrkA or corresponding tyrosines in TrkB and TrkC). This leads to the autophosphorylation of tyrosine residues, which serve as recruitment sites for specific signaling proteins and adaptors (Cunningham et al., 1997); reviewed by (Friedman and Greene, 1999). Upon activation, py in the activation 32

33 loop stabilizes a functionally active receptor conformation in which the kinase catalytic center is open to a substrate. These activation loop tyrosines can also recruit adaptor proteins, including SH2B, APS and Grb2 (Qian et al., 1998; MacDonald et al., 2000). The binding of adaptor proteins, containing phosphotyrosine binding (PTB) or Src-homology (SH-2) motifs, to phosphorylated tyrosines on Trk receptors couple the receptors to intracellular signaling pathways, including the (i) Ras mitogen-activated protein kinase (MAPK) pathway, (ii) phosphatidylinositol-3 kinase (PI-3 kinase) Akt pathway, and (iii) phospholipase Cγ gamma (PLC-γ) pathway (see Figure 6). Different subsets of adaptors are present in different cell lines or neurons; they also compete with each other for binding to activated Trk receptors. Tyrosines 490 and 785 are the main phosphorylated tyrosines coupling the activated Trk receptors to downstream signaling pathways. Phosphotyrosine 490 interacts with Shc or Frs2 and provides a mechanism for the activation of the Ras/MAPK and PI- 3K/Akt pathways. On the other hand, phosphotyrosine 785 links Trk receptors to the PLCγ pathway. 33

34 Figure 6 Trk receptor signaling Ras/MAPK pathway The role of the activation of the small GTP-ase Ras in neurotrophin induced regulation of survival and differentiation of neurons was discovered by microinjection of anti-ras antibodies and expression of mutant Ras proteins that blocked NGF induced differentiation (Hagag et al., 1986; Szeberenyi et al., 1990). Upon Trk receptor activation by neurotrophins, the Shc adaptor protein is recruited to tyrosine 490 via its PTB domain. Shc further recruites Grb2 to activate Ras via guanine nucleotide exchange factor (GEF) SOS (Stephens et al., 1994). Activated Ras then activates Raf-1 and B-Raf, serinethreonine kinases that lead to activation of MAP kinase kinase (MEK1) (Jaiswal et al., 1994). MEK1 activation triggers the activation of extracellular signal-regulated kinases/mitogen activated protein kinases (ERK/MAPK) by phosphorylating them on 34

35 threonine 202 and tyrosine 204 (Crews and Erikson, 1992). Stimulation of Ras through this pathway promotes only transient activation of MAPK (Marshall, 1995). A negative feedback loop regulates Ras signaling, with MAPK stopping the pathway by phosphorylating SOS to disrupt the Grb-2/SOS complex (Kao et al., 2001). In contrast to transient MAPK activation, prolonged MAPK activation involves the adaptor protein CrkII/CrkL, the guanine exchange factor, C3G, the small GTPase, Rap1, the protein tyrosine phosphatase, Shp2, and the serine-threonine kinase, B-Raf. Association of Frs2 with Crk results in the activation of C3G, the GEF for Rap-1, which then stimulates B- Raf, which initiates MAPK activation (York et al., 1998; Wu et al., 2001). Shp-2 probably has a role in the MAPK activation pathway via the inactivation of MAP kinase by phosphatase (Wright et al., 1997). There are two different models for how transient MAPK activation is achieved. The first model involves the recruitment of an adaptor, Frs2 (fibroblast growth factor receptor substrate-2), to phosphorylated tyrosine 490; Frs2 can also interact with the SH2 domain of CrkII (Meakin et al., 1999). However, data indicates Y490-independent mechanisms in sustained MAPK activation, because mice bearing Y490F mutations in TrkB or TrkC receptors were still able to induce sustained MAPK activation (Postigo et al., 2002). The second model involves ankyrin repeat-rich membrane-spanning protein (ARMS), a substrate of Trk receptors that becomes tyrosine phosphorylated in response to neurotrophins. ARMS associates with Trk receptors by using transmembranetransmembrane regions. Interaction of ARMS with CrkL is NGF dependent and modulates the activation of Rap-1, regulating prolonged MAPK activation (Arevalo et al., 35

36 2004; Arevalo and Wu, 2006). Most probably, both of these mechanisms are involved in regulating sustained MAPK activation by neurotrophins. The MAPK family comprises Erk1, Erk2 and Erk5 kinases. They have different downstream targets that mediate gene transcription, including Rsk and MSK1. These kinases can phosphorylate and activate CREB (Arevalo and Wu, 2006). CREB has been shown to regulate genes whose products are essential for normal differentiation and survival of neurons in vitro and in vivo (Ginty et al., 1994; Riccio et al., 1999; Lonze and Ginty, 2002). Apart from these common targets, different MAPKs also have specific transcription factor targets, with Erk5, but not Erk1 or Erk2 activating MEF2 directly. Conversely, Erk1 and Erk2, but not Erk5 activate the Elk-1 transcription factor (Huang and Reichardt, 2003). Furthermore, specificity of different MAPKs is achieved by different localization of signaling, with Erk1 and Erk2 activated by NGF signaling locally, and Erk5 activated by NGF internalized at the growth cone and transported to the cell body (Watson et al., 2001). Another mechanism of regulation of neurotrophin action involving CREB includes a positive feedback loop, with BDNF expression being regulated by CREB (Finkbeiner et al., 1997) PI-3K-Akt pathway Neurotrophins play a very important role in neuronal survival during development. The main signaling pathway involved in regulating survival is the PI-3 kinase/akt pathway. The PI-3 kinase/akt pathway has been shown to promote survival of neuroblastoma cell lines (Encinas et al., 1999) and primary neurons (Atwal et al., 2000). PI-3 kinase is a heterodimer composed of two constitutively associated subunits, an 85kDa regulatory subunit and a 110kDa catalytic subunit. Upon Trk receptor activation 36

37 Tyr 490 of Trk recruits adaptor protein Shc that associates with Grb2 and Gab1. The regulatory subunit of PI-3 kinase then binds to Grb2 or Gab1 and activates PI-3 kinase, which then induces survival of PC12 cells (Holgado-Madruga et al., 1997). PI-3 kinase can also induce survival via direct activation by Ras (Downward, 1998). The activated PI-3 kinase converts the plasma membrane lipid phosphatydilinositol-4,5- biphosphate to phosphatydilinositol-3,4,5-triphosphate. Binding of phosphatidylinositol-3,4,5- triphosphates to the PH-domain of phospo-inositide-dependent kinase 1 (PDK-1) activates PDK-1 (Anderson et al., 1998). PDK-1 associates with and phosphorylates Akt on threonine 308 (Alessi et al., 1997), leading to Akt autophosphorylation on Ser473. Phosphoinositides also can directly activate Akt via its PH domain (reviewed by (Segal, 2003)). Akt mediates survival actions of neurotrophins by regulating several different pro-survival or pro-apoptotic effectors. Phosphorylation by Akt of the Forkhead family of transcription factors (FKHRL1) blocks the transcription of several antiapoptotic genes (Zheng et al., 2002). Akt phosphorylation of the apoptosis inducing protein Bad creates a binding site for proteins and prevents Bad-Bcl-2 binding, allowing bcl-2 dependent survival (Datta et al., 1997). The activity of glycogen synthase kinase 3-β (GSK-3β) is negatively regulated by Akt-mediated phosphorylation (Hetman et al., 2000). Elevated GSK-3β promotes apoptosis, so inhibition by phosphorylation adds to the prosurvival effects. Furthermore, Akt activation at the growth cone in sensory neurons causes axonal growth via phosphorylating and inactivating GSK-3β locally (Zhou et al., 2004). The inhibitory binding partner of NFκB, IκB is another substrate of Akt. Akt phosphorylation promotes IκB degradation, resulting in liberation of active NFκB, that promotes neuronal survival (Foehr et al., 2000). PI-3K-Akt can also have an 37

38 effect on the retrograde signaling that modulates cell survival (Kuruvilla et al., 2000). Termination of PI-3 K signaling by degradation of PI (3, 4, 5) P 3 is mediated by SHIP1 and SHIP2 phosphatases and PTEN phosphatases (Cantley, 2002) PLCγ pathway Upon Trk activation, PLCγ binds directly to phosphorylated tyrosine 785 on the Trk receptor. PLCγ is then activated through serine, threonine and tyrosine phosphorylation by Trk kinase activity (Vetter et al., 1991). Activated PLCγ hydrolyses phosphatidyl-inositol (4,5) bi-phosphate to generate inositol-tri-phosphate (IP3) and diacylglycerol (DAG) (Obermeier et al., 1993). IP3 mediates the release of Ca 2+ from internal stores thereby activating Ca 2+ -regulated isoforms of protein kinase C and Ca 2+ - calmodulin regulated protein kinases (CaM kinases). DAG activates different DAGregulated protein kinase C (PKC) isoforms. PKCs are a family of intracellular serine/threonine kinases. There are three types of PKCs - novel PKCs (δ, ε, η, θ, and μ) that directly phosphorylate Raf-1, classical (α, β 1, β 2 and γ) and atypical (ζ, λ and τ) PKCs that phosphorylate Raf kinase inhibitory protein (RKIP). Phosphorylated RKIP dissociates from Raf-1 and leads to enhanced downstream signaling to Er (Corbit et al., 2003) (Maioli et al., 2006). PKCδ is required for NGF-promoted neurite outgrowth in PC12 cells and activates the Erk1 signaling pathway via Raf (Corbit et al., 1999). PLCγ signaling in response to a brief pulse of NGF induces transcription of a sodium channel gene (Toledo-Aral et al., 1995). PLCγ signaling is important for regulating synaptic transmission, with BDNF acting via TrkB to induce modulation of long-term potentiation (LTP). Ca 2+ is a key regulatory factor for LTP induction (Sheng and Kim, 2002). Y816 38

39 TrkB mutant mice, with a mutated PLCγ binding site display impaired synaptic potentiation, probably due to decreased activation of CaM kinase and CREB (Minichiello et al., 2002). Along these lines, PLCγ inhibitors block BDNF-dependent synaptic potentiation (Kleiman et al., 2000). Activated TrkB may also activate LTP via activation of Na + channels (Na v 1.9) (Blum et al., 2002). Long-term changes of synaptic transmission including LTP are thought to form the basis of learning and memory Transactivation of Trk receptors Trk tyrosine kinase receptors can also be activated as a result of transactivation. Treatment of PC12 cells or hippocampal neurons with adenosine in the absence of neurotrophins elicits the activation of TrkA and TrkB receptors, respectively, through G- protein coupled receptors (Lee and Chao, 2001) (Lee et al., 2002). Transactivation by adenosine activates the PI-3K/Akt pathway over a longer time course than neurotrophins. Trk receptors can also transactivate other receptors or ion channels. Trk receptors can activate Ret receptors in postnatal superior cervical ganglion (SCG) neurons (Tsui- Pierchala et al., 2002). In hippocampal neurons, NMDA receptors are activated by BDNF-activated TrkB receptor phosphorylation of the NMDA receptor subunit, NR2B (Levine et al., 1998). Furthermore, membrane expression of TRPV1 ion channels is increased in response to NGF (Zhang et al., 2005). These data taken together indicate the importance of cross-talk between neurotrophins and other signaling pathways in the proper functioning of the nervous system. 39

40 1.6 p75 NTR receptor - structure The p75 NTR receptor is a common neurotrophin receptor that binds all mature neurotrophins with similar affinity (10-9 ) and a faster association rate than high affinity receptors (Meakin and Shooter, 1992). Structurally, p75 NTR is a 75kDa, 399-aminoacid single pass transmembrane receptor. It has a multiple N-glycosylated extracellular ligand binding domain, a transmembrane region and an intracellular domain. The extracellular domain has four cysteine rich domains (CR1-CR4) that are required for neurotrophin binding. The crystal structure of the extracellular domain of p75 NTR in complex with NGF has been solved, with unexpected 2 NGF: 1 p75 NTR stoichiometry (see figure 7). Binding of an NGF dimer to one molecule of p75 NTR induces conformational change in the NGF protein that prevents p75 NTR homodimerization upon NGF binding (He and Garcia, 2004) In the paper by the same group, using the β-gal complementation assay, they report contradictory findings that p75 NTR homodimerization is not affected by NGF binding (Wehrman et al., 2007). The results of p75 NTR -NGF crystal structure indicate the possibility of a ternary complex of p75 NTR and NGF with other receptors, including Trk receptors. However, to date, there is no crystallographic evidence supporting a ternary Trk-NGF-p75 NTR complex. 40

41 Figure 7 Schematic representation of crystal structure of p75 in complex with NGF (He and Garcia, 2004) The transmembrane domain contains a γ -secretase cleavage site. As a result of γ- secretase cleavage, the p75 NTR intracellular domain (ICD) undergoes regulated intramembrane proteolysis (Jung et al., 2003). Prior to γ-secretase cleavage, the p75 NTR extracellular domain is cleaved by α-secretase to release the ectodomain of the receptor (Weskamp et al., 2004). The intracellular domain of p75 NTR has no catalytic activity and contains an ~80 amino acid death domain with high sequence homology to proteins of the TNF superfamily (Liepinsh et al., 1997). The lack of catalytic activity in the cytoplasmic domain of p75 NTR indicates that it signals via recruitment of intracellular adaptor proteins via protein-protein interactions. These interacting proteins can be either constitutively associated with the p75 NTR receptor or recruited to the receptor in response to neurotrophins. 41

42 Truncated isoforms of p75 NTR exist as a result of alternative splicing and postsynthetic proteolysis (von Schack et al., 2001) (Paul et al., 2004). 1.7 p75 NTR signaling and functions Although p75 NTR was cloned 20 years ago, its clear physiological role is still not completely understood. p75 NTR, depending on the cellular context, developmental stage, and type of ligand engaged can activate different signaling pathways. Activation of these pathways can lead to different, sometimes even opposing functional outcomes apoptosis, survival, proliferation, differentiation, Schwann cell migration, enhancement of neurite outgrowth and facilitation of growth cone collapse. p75 NTR activates different signaling pathways via interactions with different intracellular adaptors and signaling molecules (see Figure 8). Figure 8 p75 NTR ligands and possible signaling interactors. 42

43 The first p75 NTR -dependent signaling pathways found to be involved in the induction of apoptosis were the sphingomyelin hydrolysis and ceramide production pathways, in response to NGF treatment (Dobrowsky et al., 1994) (Brann et al., 2002). Induction of apoptosis by ceramide is shown to occur either by directly inhibiting PI-3 kinase activity (Zhou et al., 1998) or by inducing formation of inactive Ras-Raf complexes, which inhibit the ERK cascade (Muller et al., 1998). Ceramide can also reduce Akt activity through specific dephosphorylation of Ser473 (Schubert et al., 2000). It seems that the role of ceramide in p75 NTR signaling is dependent on cellular context just like neurotrophin signaling through p75 NTR. Furthermore, several p75 NTR intracellular domain interacting adaptor proteins were identified: NRIF (neurotrophin receptor interacting factor), NADE (p75 NTR - associated cell death executor) and NRAGE (neurotrophin receptor-interacting MAGE homolog) (Casademunt et al., 1999) (Mukai et al., 2000) (Salehi et al., 2000). NRIF requires the co-expression of TRAF6 to induce apoptosis via JNK activation (Gentry et al., 2004). NRAGE also interacts with p75 NTR to induce apoptosis by a mechanism involving JNK and activation of caspases 3, 7 and 9 (Salehi et al., 2000). General mechanisms of p75 NTR -dependent induction of apoptosis involve JNK activation, phosphorylation of c-jun, activation of p53, Bad and Bim, translocation of Bax to mitochondria and release of cytohrome c and activation of caspases 9, 6 and 3 (Nykjaer et al., 2005). p75 NTR associated with adaptor protein NADE induces apoptosis upon NGF binding to p75 NTR (Mukai et al., 2000). Ligand-dependent p75 NTR activation has been shown to cause apoptosis of cultured neonatal sympathetic neurons, oligodendrocytes, Schwann cells (NGF-dependent) and motor neurons (BDNF-dependent) (Casaccia- 43

44 Bonnefil et al., 1996; Bamji et al., 1998) (Davey and Davies, 1998). Conversely, in some cell types, p75 NTR signaling induces apoptosis in the absence of ligand (Barrett and Bartlett, 1994). The p75 NTR intracellular domain also interacts with TNF receptorassociated factors (TRAFs). TRAF2 associates with the C-terminal death domain, while TRAF4 and TRAF6 interact with the juxtamembrane region of p75 NTR receptor. TRAF2 and TRAF6 were shown to enhance, and TRAF4 to inhibit limited NFκB activation induced by p75 NTR expression (Khursigara et al., 1999) (Ye et al., 1999). Neurotrophin binding to p75 NTR can also induce the MAP kinase signaling pathway and stimulate apoptosis in some cell types, and survival in others (Lad and Neet, 2003). NGF binding to the p75 NTR receptor activates NFκB in neuroblastoma cells, cultured sensory and sympathetic neurons (Maggirwar et al., 1998), Schwann cells (Hirata et al., 2001) and oligodendrocytes (Yoon et al., 1998), and induces a prosurvival effect. NFκB is a ubiquitous transcription factor that is activated by the phosphorylation and subsequent degradation of IκB. Activated NFκB is translocated to nucleus where it induces the transcription of target genes, leading to increased survival. p75 NTR can also induce survival through activating the PI-3K/Akt pathway independent of TrkA signaling (Roux et al., 2001). Recently, it has been shown that NGF acting via p75 NTR can promote axonal elongation by inactivating GSK-3β(Arevalo and Wu, 2006). Regulated intramembranous proteolysis of p75 NTR by α and γ-secretase is another mechanism involved in p75 NTR signaling (Zampieri et al., 2005). Cleavage of the ectodomain by α-secretase leads to shedding of the p75 NTR ectodomain. The p75 NTR ectodomain can serve as a scavenger of NTFs in the extracellular space. γ-secretase 44

45 cleavage then liberates the cytoplasmic tail from its membrane anchorage. p75 NTR ICD can translocate to the nucleus and activate NFκB signaling (Kanning et al., 2003). Induction of apoptosis by p75 NTR has a physiological role during development. In the case of mistargeting to an inappropriate location, a neurotrophin may bind to p75 NTR and eliminate cells through an active killing process (Majdan and Miller, 1999). p75 NTR receptor expression is up-regulated after injury, in pathological or inflammatory conditions (Roux et al., 1999). BDNF binding to p75 NTR can inhibit Schwann cell migration and promote Schwann cell differentiation/myelination (Bentley and Lee, 2000). Pro-BDNF acting via p75 NTR is important for modulating long term depression (LTD) (Woo et al., 2005). Apart from signaling independently, p75 NTR creates several signaling platforms in complex with different co-receptors. p75 NTR receptors form complexes with: i) Trk receptors ii) Nogo receptor and iii) sortilin. This indicates a role for p75 NTR in contributing to different receptor systems (Barker, 2004). 1.8 p75 NTR as a co-receptor p75 NTR and sortilin Discovery that pro-ngf acting via p75 NTR induces apoptosis of corticospinal neurons (Harrington et al., 2004) and oligodendrocytes (Beattie et al., 2002) raises the question of how p75 NTR mediates the effects of pro-ngf. Sortilin was identified as a p75 NTR coreceptor necessary for pro-ngf induced cell death. Sortilin is a ~95 kda transmembrane Golgi protein; its extracellular domain binds the prodomain of NGF, while the mature part of NGF binds p75 NTR. Both p75 NTR and sortilin are required for 45

46 pro-ngf induced apoptosis (Nykjaer et al., 2004). Sortilin plays a role in trans golgi network (TGN) to endosome trafficking, so it is possible that it facilitates p75 NTR signaling by retrograde transport of p75 NTR -pro-ngf-sortilin complexes in signaling endosomes. The apoptotic effect mediated by p75 NTR and sortilin is not specific for pro- NGF only, because pro-bdnf also induces apoptosis via this complex (Teng et al., 2005). Therefore, the presence or absence of sortilin at the cell surface determines response to proneurotrophins p75 NTR and Nogo Absence of p75 NTR signaling perturbs axon growth in vitro and both axon growth and target innervation in vivo (Bentley and Lee, 2000). Motor neurons in the spinal cord, most sympathetic and sensory neurons in the peripheral nervous system, as well as retinal ganglion cells express high p75 NTR levels during the outgrowth of axons. Furthermore, differentiation of hippocampal neurons is regulated by p75 NTR ligands via the ceramide pathway (Brann et al., 1999; Brann et al., 2002). p75 NTR modulates neuronal outgrowth of ciliary neurons in a ligand-dependent fashion by suppressing RhoA activation, with NGF being more efficient than BDNF and NT-3 (Yamashita et al., 1999; Yamashita et al., 2002). Moreover, in adults, RhoA activation mediates the effects of CNS derived myelin-based growth inhibitors (MBGIs) including Nogo, myelin associated glycoprotein (MAG) and oligodendrocyte myelin glycoprotein (OMgp). Furthermore, the MBGIs bind to the Nogo receptor (NogoR), a GPI (glycosylphosphatidyl-inositol) - linked protein that forms a complex with p75 NTR (Wang et al., 2002a; Wong et al., 2002). Lingo-1, a nervous system specific transmembrane protein also interacts with both, p75 NTR and NogoR via its extracellular domain. In non-neuronal COS cells co-expressing NogoR, 46

47 p75 NTR and Lingo-1, expression of all three receptors is necessary for RhoA activation by OMgp, Mag and Nogo (Bandtlow and Dechant, 2004). The signal transduction mechanism of MBGIs might involve the displacement of Rho-GDI from RhoA, activating RhoA and inducing axonal growth inhibition (Yamashita and Tohyama, 2003). More recently, it has been shown that p75 NTR ICD produced in response to MAG in cerebellar granule neurons leads to growth inhibition (Domeniconi et al., 2005). However, the results of in vivo studies looking at the role of p75 NTR in axon regeneration are not clear. p75 NTR null mice have shown increased regeneration and enhanced functional recovery in an adult facial neuron regeneration model (Boyd and Gordon, 2001). Conversely, p75 NTR null mice do not show enhanced regeneration of corticospinal neurons after spinal cord damage (Song et al., 2004). Further investigation is necessary to understand the role of p75 NTR inhibition in injury models to promote axonal regeneration p75 NTR and Trk When co-expressed, p75 NTR and Trk receptors have been shown to regulate the response of cells to neurotrophins. Mixed complexes of p75 NTR and all three Trks have been shown to exist by different techniques. Co-patching of TrkA and p75 NTR receptors using fluorescent antibodies and crosslinking demonstrated p75 NTR -TrkA, but not p75 NTR -TrkB co-localization (Ross et al., 1996). However, co-immunoprecipitation of various p75 NTR and epitope tagged Trk receptor chimeric constructs demonstrated the interaction between p75 NTR and all three members of the Trk family; the interaction is ligandindependent and requires the active kinase domain (Bibel et al., 1999). 47

48 Trk and p75 NTR receptors are co-expressed in embryonic primary sensory and sympathetic neurons, and mixed Trk/p75 NTR complexes are the most likely explanation for formation of high affinity binding sites (reviewed by (Dechant et al., 1994)). p75 NTR receptors interact with Trk receptors via transmembrane and cytosolic domains (Esposito et al., 2001). Therefore, p75 NTR has the role of a co-receptor that refines Trk specificity for neurotrophins; p75 NTR increases the affinity of TrkA for NGF, but restricts TrkA binding to NT-3 (Brennan et al., 1999). Furthermore, the role of p75 NTR in regulating TrkA responsiveness is crucial during the development of sympathetic neurons. p75 NTR expression reduces NT-3/TrkA dependent axonal growth and promotes NGF/TrkA dependent survival and differentiation (Kuruvilla et al., 2004). In addition to this, in A293 cells cotransfected with p75 NTR and TrkB, p75 NTR co-expression results in higher specificity of TrkB for BDNF, compared to NT-3 and NT-4 (Bibel et al., 1999). When p75 and Trks are co-expressed, p75 NTR can modulate signaling pathways elicited by Trk receptor activation. BDNF bound p75 NTR induces serine phosphorylation of TrkA, resulting in its reduced activation (MacPhee and Barker, 1997). Conversely, the monoclonal antibody 192, as a p75 NTR ligand, synergises with optimal TrkA ligands to afford optimal protection in cells co-expressing p75 NTR and Trks. Furthermore, p75 NTR co-expression decreases the autophosphorylation of TrkB induced by BDNF or NT-4/5 in MG87 fibroblasts (Vesa et al., 2000). p75 NTR receptors and their ligands regulate TrkA and TrkC activity and downstream signaling pathways differently (Ivanisevic et al., 2003). How does the p75 NTR receptor modulate Trk elicited signaling pathways? p75 NTR can bind Shc, stimulate its phosphorylation and enhance Trk signaling (Epa et al., 2004). 48

49 Intramembrane proteolysis of p75 NTR can play a role in the assembly/disassembly of receptor complexes containing p75 NTR and Trk (Jung et al., 2003). Furthermore, ankyrin rich membrane protein (ARMS) adaptor protein interacts with both p75 NTR and Trk receptors. Prolonged MAPK signaling through Trk elicited ARMS phosphorylation is sufficient for sustained neurite outgrowth, but complexing with p75 NTR is necessary for long-term survival (Arevalo et al., 2004). Another adaptor protein, which interacts with p75 NTR and Trk is fas apoptosis inhibitor molecule (FAIM). FAIM promotes neurite outgrowth via activation of NFκB and the Ras-MAPK pathway (Sole et al., 2004). These data suggest that Trk-p75 NTR complexes might possess different signaling abilities than Trk dimers. 49

50 1.9 Neurotrophins role in pathology and clinical prospects Neurotrophic factors regulate neuronal development and nervous system maintenance in adult life. As a consequence of these important physiological functions, disregulation of neurotrophins and their receptors is implicated in the pathology of different neurodegenerative and psychiatric disorders, certain types of cancer, and provides a rationale for therapeutic intervention. NGF has been shown to have an effect in models of diabetic peripheral neuropathy and Alzheimer s disease. NT-3 was proposed for treatment of amyotropic lateral sclerosis (ALS) (Haase et al., 1997). Neurotrophins are involved in inflammatory and pain states, so neurotrophin receptor antagonists would be useful for their treatment (McMahon et al., 1995) (Marshall et al., 1999). However, antagonists can cause neuropathies, so ideally agents that would induce survival without additional sprouting would be applicable for treatment of neuropathies and pain (Saragovi et al., 1998). Preliminary clinical studies engendered much optimism regarding the use of neurotrophins as therapeutic agents. However, clinical trials with neurotrophins have been largely disappointing. Problems associated with the use of neurotrophins in therapy are numerous neurotrophins are not orally bioavailable, they do not cross the blood brain barrier, they have poor pharmacokinetic profiles, and they can cause immune reactions. To circumvent these problems, several approaches are being used. One approach relies on improving the delivery system of neurotrophins and includes gene therapy, cell therapy and microencapsulated slow-release implants. Neurotrophin gene therapy has entered a nonrandomized, nonplacebo clinical trial with somewhat promising results. All these approaches require surgery and are still in the early stages (Tuszynski et al., 2005). Another approach is the development of small molecules with neurotrophic 50

51 agonistic or antagonistic activity small molecule nonpeptidic neurotrophin mimetcs. Ligand mimicry is a process by which a large polypeptide ligand is reduced to smaller functional units that bind selectively to ligand binding and activation sites on appropriate receptors (Saragovi and Zaccaro, 2002). These peptidomimetics, based on their activity, can have different mechanisms of action: activating receptors directly as ligands, activating them indirectly by upregulating the production of endogenous neurotrophins, modulating receptor signaling pathways or affecting co-receptors that regulate neurotrophin receptor activity (Pollack and Harper, 2002). This approach was used to create the first rationally derived TrkA agonist, peptidomimetic compound D3. D3 is a proteolytically stable small molecule that is a selective TrkA partial agonist and binds to the D5 subdomain of TrkA (Maliartchouk et al., 2000b). D3 binds to Trk receptors in vitro. It also affords significant and long-term rescue of cholinergic neurons in the cortex and in the nucleus basalis, which correlates with an enhanced cholinergic phenotype and a significant improvement of memory and learning in cognitively impaired aged rats (Bruno et al., 2004). During the progression of Alzheimer s disease there is a loss of TrkA receptors but an important observation is that the levels of p75 NTR receptors are maintained (Yaar et al., 1997; Counts and Mufson, 2005). Studies performed in vitro showed that pro-ngf can induce apoptosis in neuronal cultures through activation of p75 NTR receptor (Pedraza et al., 2005). Additionally, pro-ngf is abundant in human brains from Alzheimer s patients (Peng et al., 2004). It is also noteworthy that Aß protein has been shown to be a functional ligand of p75 NTR, suggesting that Aß protein could also mediate its pathological effects in Alzheimer s disease via the p75 NTR signaling pathway (Yaar et al., 51

52 1997). In cultured basal forebrain neurons, exposure to both mature and pro-ngf demonstrated that TrkA activation did not prevent pro-ngf induced apoptosis via p75 NTR (Volosin et al., 2006). Therefore, not only TrkA agonists, but also small molecules targeting p75 NTR (p75 NTR antagonists) that can inhibit apoptotic signals or block pro-ngf binding (Massa et al., 2006) can be useful in the treatment of Alzheimer s disease or other neurodegenerative diseases with abberant p75 NTR signaling. As we learn and understand more about neurotrophin receptors and their ligands, the possibilities of successful targeting of these proteins in human diseases are increasing. The applications include treatment of chronic and acute neurodegeneration, some forms of cancer and chronic pain with agonists, and some forms of cancer or acute pain with antagonists. 52

53 Chapter 2 p75 NTR Co-receptors Regulate Ligand-dependent and Ligand-independent Trk Receptor Activation, in Part by Altering Trk Docking Subdomains.

54 2.1 Rational Neurotrophin receptors, Trk family of tyrosine kinase receptors and p75 NTR coreceptor play important roles in regulating the development and maintenanace of nervous system and neuroectoderm derived tissues. Additionally, neurotrophin receptors are important pharmacological targets for the treatment of neurodegenerative diseases and certain types of cancer. To be able to develop artificial neurotrophin receptor ligands with therapeutic potential it is necessary to characterize neurotrophin receptor ligand-binding and functional activation sites. Interestingly, p75 NTR co-receptor is also capable of ligand-binding and independent signaling; and its co-expression regulates Trk receptor ligand affinity and Trk dependent signaling. Molecular mechanisms responsible for p75 NTR -Trk interactions are still unclear, therefore both receptors have to be taken into account for developing potential novel therapeutics. The extracellular domains of Trk receptors have specific subdomain organization leucin rich motif (D2) is flanked by two cystein rich motifs (D1) and (D3) and closer to the transmembrane region are two immunoglobulin like subdomains (D4) and (D5). Series of studies, including the crystal structure confirmed that the D5 subdomain is the ligand-binding domain. In Chapter 2, we use chimeric TrkA/B receptors which have different subdomains of TrkB receptor swapped with the same domain of TrkA receptor and we ask the following questions: What are the roles of different Trk receptor subdomains in ligand-dependent and ligand-independent receptor activation? Does p75 NTR co-expression affect the subdomains involved in Trk receptor ligand iduced-activation? 54

55 2.2 Abstract Neurotrophins signal via Trk tyrosine kinase receptors and a common receptor called p75 NTR. Nerve growth factor is the cognate ligand for TrkA, brain-derived neurotrophic factor for TrkB, and neurotrophin-3 (NT-3) for TrkC. NT-3 also binds TrkA and TrkB as a heterologous ligand. All neurotrophins bind p75 NTR, which regulates ligand affinity and Trk signals. Trk extracellular domain has five subdomains: a leucine-rich motif, two cysteine-rich clusters, and immunoglobulin-like subdomains D4 and D5. The D4 subdomain is surface exposed in the tertiary structure and regulates ligand-independent activation. The D5 subdomain is less exposed but regulates cognate ligand binding and Trk activation. NT-3 as a heterologous ligand of TrkA and TrkB optimally requires the D5 but also binds other subdomains of these receptors. When p75 NTR is co-expressed, major changes are observed; NGF-TrkA activation can occur also via the cysteine 1 subdomain, and brain-derived neurotrophic factor-trkb activation requires the TrkB leucine-rich motif and cysteine-2 subdomains. We propose a two-site model of Trk binding and activation, regulated conformationally by the D4 subdomain. Moreover, p75 NTR affects Trk subdomain utilization in ligand-dependent activation, possibly by conformational or allosteric control. 55

56 2.3 Introduction The neurotrophins (NTFs) regulate the survival, death, or differentiation of neurons in the embryonic and early postnatal stages and neuronal maintenance later in life. NTFs include nerve growth factor (NGF), neurotrophin-3 (NT-3), and brain-derived neurotrophic factor (BDNF). NGF interacts selectively with TrkA receptors; BDNF interacts selectively with TrkB receptors. NT-3 interacts with TrkC receptors preferably, but it is more promiscuous and can also bind TrkA and TrkB (Barbacid, 1994). All NTFs also bind a shared receptor termed p75 NTR with relatively low affinity. NTFs bind to the extracellular domain of Trk and induce receptor homodimerization, leading to activation of the intrinsic intracellular tyrosine kinase catalytic activity (Jing et al., 1992; Kaplan and Stephens, 1994). The extracellular domain of Trk receptors features five subdomains defined by their homology to other proteins (Schneider and Schweiger, 1991). Near the N terminus, there is a leucine-rich motif (LRM), flanked by two cysteine rich clusters (Cys-1 and Cys-2). Closer to the transmembrane spanning region, there are two immunoglobulin-like subdomains termed Ig-C1 (D4) and Ig-C2 (D5) (Fig. 1). Previous work addressed the Trk extracellular subdomains responsible for NTF binding. Expression of truncated or chimeric receptors in cells showed that Ig-C2 subdomains of TrkA, TrkB and TrkC are relevant for binding NGF, BDNF and NT-3, respectively (Perez et al., 1995; Urfer et al., 1995; MacDonald and Meakin, 1996; Urfer et al., 1998), and a recombinant polypeptide spanning TrkA-D4 and TrkA-D5 bound and neutralized NGF (Holden et al., 1997). The LRM subdomain was also proposed as ligand - binding (Windisch et al., 1995b; Windisch et al., 1995a; Kadari et al., 1997); however, 56

57 expression of LRM-deleted TrkA did not affect NGF binding or activity but altered NT-3 activity (MacDonald and Meakin, 1996). Our study had three aims. The first was to clarify the general architecture of Trk and to define possible hinge or regulatory regions that may control receptor activation. While tyrosine kinases undergo conformational changes upon ligand binding (Heldin, 1995), analysis failed to demonstrate these changes in Trk (Woo et al., 1998) or in NGF-D5 crystals (Wiesmann et al., 1999). Second, we wanted to address the hypothesis that p75 NTR coexpression may allow certain ligands to activate Trks via different extracellular subdomains. This would be expected because there are reports of p75 NTR -Trk allosteric or cooperative interactions (MacPhee and Barker, 1997; Maliartchouk and Saragovi, 1997; Bamji et al., 1998; Bibel et al., 1999; Brennan et al., 1999; Kohn et al., 1999). The third aim was to define subdomains of TrkA and TrkB that are functional toward heterologous ligands. While other studies addressed Trk interactions with cognate ligands, studies of heterologous receptor activation by NT-3 were not done. We find that most of the TrkA or TrkB activation by cognate ligands occurs via the D5 subdomain. NT-3 mediated activation of TrkA or TrkB is also mediated by the D5 subdomain but requires additional subdomain(s), probably the D4. Expression of p75 NTR allows NGF activation of TrkA via its Cys-1-rich (D1) subdomain, while BDNF activation of TrkB also involves the LRM (D2) and cysteine 2 (D3) subdomains when p75 NTR is coexpressed. The D4 subdomain is a hinge region with intrinsic regulation of receptor conformation and activation. Our data points to a complex, multistage process of regulating receptor-ligand interactions that include intrinsic Trk- regulatory subdomains and receptorco-receptor interactions. 57

58 2.4 Materials and Methods Human TrkA-Rat TrkB Chimeras Single subdomains (leucine repeats; D4 and D5) of human TrkA were generated by polymerase chain reaction using as template a 2.7-kb cdna encoding the human trka in Bluescript KS + vector. Oligonucleotide primers used for the polymerase chain reaction amplification of the human TrkA extracellular subdomains: Cys cluster forward, 5'-ATAT GAATTC GCG CAC ATG TCG GGG GAG-3'; Cys cluster backward, 5'-ATAT TTCGAA ATC ACG GAG CTC CAG ATG-3'; LRM forward, 5'- ATAT TTCGAA GGC CTG GGG GAG CTG AGA-3'; LRM backward, 5'-ATAT GCATGC GGT ATT GGG CAT GTG GGC-3'; IgG-C1 forward, 5'-ATAT GCATGC GGT GTG CCC ACG CTG AAG-3'; IgG-C1 backward, 5'-ATAT CTCGAG CTG CAC ACT GGC CGG GAA-3'; IgG-C2 forward, 5'-ATAT CTCGAG ACG GCG GTG GAG ATG CAC-3'; IgG-C2 backward, 5'-ATAT ACGCGT TGT ACT GTT AGT GTC AGG-3'. They include unique restriction sites (boldface type) to allow exchange with the corresponding rat TrkB subdomains. The chimeric receptors were constructed by subcloning each human TrkA subdomain into the corresponding unique restriction sites of the rat trkb cdna reported in previous work (Perez et al., 1995) (Fig. 1). All the chimeric constructs were confirmed by sequencing and were cloned into the pcdna3 expression vector that contains a selection gene providing resistance to neomycin (Life Technologies, Inc.). Transfection and Stable Expression in 293 Cells and nnr5 Cells HEK293 cells (human kidney epithelium, Trk -, p75 NTR- ) and nnr5 cells (derived from PC12 cells, Trk -, p75 NTR ++ ) were transfected with the chimeric cdna constructs using the LipofectAMINE Plus method (Life Technologies, Inc.). At least three independent subclones of neomycinresistant cells for each chimera were obtained by limiting dilution techniques. Quantitative 58

59 Western blot analysis (Mufson et al., 1997) with a polyclonal antibody directed to the Trk intracellular domain (203 antisera; a gift of David Kaplan, McGill University) indicated that the stable HEK293 subclones express 40, ,000 chimeric receptors/cell and that the stable nnr5 subclones express 2,000-6,000 chimeric receptors/cell (data not shown). All cells were grown in RPMI 1640 supplemented with 5% fetal calf serum, antibiotics, and glutamine. Transfectants had 0.4 mg/ml neomycin. Adenoviral Infection and Expression of p75 NTR in HEK293 Cells Adenoviruses expressing full-length rat p75 NTR were a kind gift of Sung Ok Yoon (Ohio State University). Expression of p75 NTR in HEK293 cells was achieved by infection at a multiplicity of infection of 10. This resulted in undetectable cell death after 16 h at the time when cells were used for phosphotyrosine (Tyr(P)) assays. At 16 h postinfection, >90% of the HEK293 cells had homogeneous expression of p75 NTR. Quantitative flow cytometric analysis and Western blotting (Saragovi et al., 1998) resulted in an estimate of ~30,000 p75 NTR molecules per cell (data not shown), relative to p75 NTR expression in cells, expressing a known density of p75 NTR receptors (Maliartchouk and Saragovi, 1997). At longer times postinfection (30 h and above) there was progressive cell death due to viral replication, which precluded us from performing long-term cell survival assays. Cell Survival Assays Cells used for survival assays were wild type or transfected HEK293 and nnr5. Cell survival was measured by quantitative tetrazolium salt reagent (3- (4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; Sigma) and OD readings as described (Maliartchouk and Saragovi, 1997; Saragovi et al., 1998). Cells were plated in serum-free media (SFM) (PFHM-II, Life Technologies, Inc.), with 0.2% bovine serum albumin (Roche Molecular Biochemicals), supplemented with SFM (negative control), with 59

60 2 nm of the indicated neurotrophin (NTF) (test), or with 5% serum (positive control) for ~68 h. Cell growth/survival was calculated relative to 5% serum (standardized to 100%) to eliminate the possible confounding factor of different growth rates by each clone, as described (Maliartchouk and Saragovi, 1997; Saragovi et al., 1998). The difference between 5% serum and SFM supplemented with neurotrophin is that in the former cells survive and grow but in the latter there is poor growth. All assays were repeated with at least three subclones for each construct, and each assay was repeated at least three times (n = 4-6 per assay). Statistical analysis was done by paired student t test with Bonferroni corrections, and significance was p < Western Blotting Western blots were performed as described (Maliartchouk and Saragovi, 1997; Saragovi et al., 1998). Blots were visualized using the enhanced chemiluminescence system (PerkinElmer Life Sciences). For anti-trk immunoblotting, cells were solubilized, and protein concentrations were determined. Samples were fractionated by SDS-PAGE, transferred to membranes, and immunoblotted with RTA serum (Clary et al., 1994) directed to the whole extracellular domain of TrkA. Equal protein loading was confirmed by Coomassie Blue staining of gels and by stripping and reblotting membranes with anti-trk 203 antiserum. Tyrosine Phosphorylation Assays Cells were washed and rested in SFM at 37 C for 30 minutes to lower Trk receptor background phosphorylation levels. After ligand treatment (2 nm neurotrophin at 37 C for 12 min), cells were solubilized, and protein concentrations were determined. Samples were fractionated by SDS-polyacrylamide gel electrophoresis, transferred to membranes, and immunoblotted with anti-phosphotyrosine (- 60

61 Tyr(P)) antibody 4G10 (Upstate Biotechnology, Inc., Lake Placid, NY). Equal loading was confirmed by stripping and reblotting membranes with anti-trk 203 antiserum. 125 I-NGF Binding Studies on 293 Cells Expressing Chimeric Receptors Analysis of recombinant human 125 I-NGF binding to chimeric or wild type receptor-expressing cells was done as described (24). Cells (1X10 6 /point) were added to serial dilutions of 125 I-NGF (70 µci/µg; PerkinElmer Life Sciences) in the absence or presence of a 250-fold excess of cold NGF as competitor. Free and cell-bound counts were fractionated, counted (COBRA counter), and analyzed by Scatchard plot. 2.5 Results Full-length TrkB receptors expressing subdomains of human TrkA were generated as illustrated in Fig. 1 using the primers described under "Materials and Methods." The five chimeric cdnas and wild type receptor controls were transfected into HEK293 or nnr5 cells, and at least three stable independent clones were obtained for each construct. All biological data were reproduced with three independent subclones for each cdna construct in each cell line. 61

62 Figure 1. Model of TrkA/TrkB chimera Long Term Trophic Signals Induced in Trk Chimeras Trophic signals via Trk chimeras expressed in HEK293 cells were probed in survival assays using the quantitative 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide method (Table I). Cells cultured in SFM undergo apoptotic death, which can be rescued by NTFs if the cells express functional NTF receptors. Control HEK293 cells transfected with wild type Trks exhibited the expected survival profile (Table I, rows 7-9), where NGF protected TrkA-expressing cells, BDNF protected TrkB-expressing cells, and NT-3 protected TrkC-expressing and TrkB-expressing cells and significantly protected TrkA-expressing cells but to a lower degree. Untransfected HEK293 cells did not survive in SFM supplemented with NTFs (Table I, row 1). HEK293 cells expressing the 1.1 chimera did not survive with NGF but did survive with NT-3 and with BDNF (Table I, row 2). Similar data was obtained with the 2.1 chimera, except that more limited survival was seen with NT-3 (Table I, row 3). Hence, the D1, D2, 62

63 and D3 subdomains from TrkA or TrkB do not play a role in NGF-TrkA activation (no gain of function), nor do they play a major role in BDNF-TrkB activation (no major loss of function); but perhaps they do for NT-3-TrkB activation. Cells expressing the 2.2 chimera had low but significant survival with NGF, NT-3, and BDNF (Table I, row 4). The lower survival induced by NT-3 and BDNF in the 2.2 chimera (compared with the 2.1 chimera) is probably due to the replacement of the D4 subdomain with that of TrkA (the only difference between the 2.2 and the 2.1 chimeras). These data suggest an activating role for the D4 subdomain of TrkB bound by NT-3 or BDNF and also suggest an activating role for the D4 subdomain of TrkA bound by NGF. HEK293 cells expressing the 3.1 chimera are constitutively activated, and cells have significant survival in SFM (~40% compared with serum) in the absence of NTFs. Survival was not enhanced significantly in response to any NTF. These data also point to a critical role of the D4 subdomain of TrkB and suggest that this subdomain is functionally different in TrkA. HEK293 cells expressing the 4.1 chimera had significant survival with NGF, significant but low survival with NT-3, and no survival with BDNF (Table I, row 6). Compared with wild type TrkA, NGF activates in full via the TrkA D5 subdomain, and BDNF activation absolutely requires the TrkB D5 subdomain. NT-3 may activate partially via TrkA D5 (or the remaining subdomains of TrkB) in the 4.1 chimera, but it seems to be more dependent on other subdomains. Table I Ligand mediated survival of HEK293 cells expressing Trk receptors HEK293 cells stably expressing the indicated cdna were tested in cell survival (3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assays after culture in SFM supplemented with the indicated NTFs. Trophic survival was calculated relative to controls (5% serum=100%, SFM=0%). Shown is percentage survival (mean ± S.D.) averaged from three experiments, each experiment n=6. 63

64 a Significant with respect to corresponding cell in SFM without neurotrophin (p<0.02). b There is high survival for 3.1 chimeras in SFM (39 ± 3% relative to serum); hence, NTFs do not significantly enhance survival. NGF Binding to Trk Chimeras In order to test whether there are chimeras that did not signal but bound NGF, we performed 125 I-NGF binding studies (Fig. 2). We tested NGF binding with the 1.1, 3.1, and 4.1 chimeras and binding to wild type HEK293 cells as control. Only the 4.1 chimera had specific 125 I-NGF binding. The 4.1 chimera subclone 6 (4.1.6) expresses ~50,000 NGF receptors/cell with an apparent single affinity of 80 ± 24 pm (n = 3 independent assays). These binding sites were fully competed by a 250-fold excess of unlabeled NGF (Fig. 2). The 1.1 chimera, the 3.1 chimera, and the wild type HEK293 cells showed no specific NGF binding sites (n = 2 independent assays) (data not shown). The binding assays would not have detected binding affinities significantly lower than 10 nm (which is approximately the K d of p75 NTR ). Therefore, we cannot rule out very low affinity NGF binding sites in 64

65 1.1 and 3.1 chimeras. Likewise, 125 I-NGF binding assays using cells co-expressing chimeras and p75 NTR are fruitless, because it is impossible to discriminate whether the chimera (as well as p75 NTR ) binds NGF with low affinity. Direct binding studies of 125 I-NT-3 and 125 I-BDNF to cells could not be carried out, because these ligands were unavailable to us in radiolabeled form. Figure 2 Scatchard plot of 125 I-NGF binding studies in 4.1 chimeras. Scatchard plot analysis of 125 I-NGF binding to 4.1 chimera (human TrkA Ig-C2). Cells (1X10 6 ) were incubated with 125 I-NGF (from 8 nm to 140 pm) in the absence (open triangle) or presence (closed circles) of a 250-fold molar excess of unlabeled NGF. Tyrosine Phosphorylation of Chimeric Receptors in HEK293 Cells Trk Tyr(P) was assayed in HEK293 transfectants after 12 min of exposure to NTFs (Fig. 3, A-E). The 1.1 and 2.1 chimeras respond well to NT-3 and BDNF but not to NGF (Fig. 3, A and B); the 2.2 chimeras respond to NGF, NT-3, and BDNF but very poorly, and 65

66 the gel had to be overexposed to see the response (Fig. 3C). The 3.1 chimeras are constitutively tyrosine-phosphorylated (Fig. 3D, untreated), and there is no significant increase in response to all ligands (Fig. 3D). The 4.1 chimeras respond very well to NGF and poorly to NT-3, and they do not respond to BDNF (Fig. 3E). Gain of NGF function and loss of BDNF function defines the D5 as the main activation subdomain, with smaller contributions from other subdomains. Interpretation for NT-3 survival is less clear, because this growth factor binds both wild type TrkA and TrkB, and we cannot discriminate which receptor is bound by NT-3 in our chimeras. Altered Function of Trk Chimeras in HEK293 Cells Co-expressing p75 NTR To assess how co-expression of p75 NTR receptors may affect the activation of Trk chimeras, HEK293 transfectants were infected with adenoviral constructs expressing fulllength p75 NTR. After 16 h, there was homogeneous expression of p75 NTR receptors in >90% of the cells, with no death (data not shown). At 16 h postinfection, the Tyr(P) of the chimeric receptors was measured (Fig. 3, F-J). In the presence of p75 NTR the 1.1 chimera is activated by NGF, NT-3, and BDNF (Fig. 3F). The major difference is that in the absence of p75 NTR NGF does not activate the 1.1 chimera (Fig. 3A). In the presence of p75 NTR, the 2.1 chimera is activated well by NT-3 and poorly by BDNF and NGF (Fig. 3G). The major difference is that in the absence of p75 NTR (Fig. 3B) BDNF can activate the 2.1 chimera better. In the presence of p75 NTR, the 2.2 chimera does not respond to any NTF (Fig. 3H). There are no major differences with data in the absence of p75 NTR (Fig. 3C). In both cases, the gels had to be overexposed to see any limited response to ligand. In the presence of p75 NTR, the 3.1 chimera is still constitutively activated (Fig. 3I, Untre.) but to a lower degree than when p75 NTR is absent (Fig. 3D, Untre.). In the presence of p75 NTR, the 4.1 chimeras respond well 66

67 to NGF and NT-3 and do not respond to BDNF (Fig. 3J). The major difference is that NT-3 activation (relative to NGF) is higher in the presence of p75 NTR than in the absence of p75 NTR (Fig. 3E). In all chimeras, ligand-independent Tyr(P) of the receptors (even the constitutively activated 3.1 chimera) was substantially lower after expression of p75 NTR (see untreated (Untre.) lanes for all panels). These data suggest that unbound p75 NTR can inhibit constitutive Trk activation and that p75 NTR can alter the ligand-dependent activation subdomains of Trk receptors. Figure 3. Ligand-induced tyrosine phosphorylation of chimeric receptors. HEK 293 cells expressing the indicated chimeric Trk receptor alone (A-E) or with co-expression of p75 NTR (F-J) were untreated or treated with 2 nm concentration of the indicated NTFs for 12 minutes. Cell lysates were fractionated in 7.5% SDS-polyacrylamide gel electrophoresis and Western blotted with α-tyr(p) mab 4G10. Trophic Function of Trk Chimeras in a Neuronal Cell Line Expressing p75 NTR Long term (2-3-day) survival assays could not be done with HEK293 cells expressing p75 NTR, because these cells die ~36 h after adenoviral infection. Thus, all of the chimeric cdnas were transfected and expressed in nnr5 cells to test whether p75 NTR affects the long - term survival function of Trk chimeras. Stable nnr5 clones were obtained for all cdnas 67

68 except the 3.1 chimera, where cells seem to differentiate in the absence of ligand. The nnr5 long - term survival data (Table II) are consistent with the Tyr (P) results for HEK293 transfectants co-expressing p75 NTR (Fig. 3). As expected, all stably transfected nnr5 clones cultured in SFM undergo apoptotic death. Positive controls transfected with wild type Trk receptors exhibited the expected survival profile (Table II, rows 6 and 7). Untransfected nnr5 cells cultured in SFM die whether or not they are supplemented with NTFs (Table II, row 1), indicating that p75 NTR expression alone does not mediate survival or death signals in these cells. The nnr5 cells expressing the 1.1 chimera (Table II, row 2) survived in response to NGF, while NT-3- and BDNF-induced survival was statistically insignificant. The nnr5 cells expressing the 2.1 chimera (Table II, row 3) survived with NT-3, whereas survival in response to BDNF and NGF was statistically insignificant. The nnr5 cells expressing the 2.2 chimera (Table II, row 4) did not respond efficiently to any neurotrophin, and all responses were low compared with wild type TrkB and TrkA. The nnr5 cells expressing the 4.1 chimera (Table II, row 5) survived with NGF and with NT-3, but survival in response to BDNF was insignificant. Table II Ligand-mediated survival of nnr5 cells expressing Trk receptors Nnr5 cells stably expressing the indicated transfected cdnas were tested. Cell survival was measured as described in the Table I legend. Shown is percentage survival (mean ± S.D.) averaged from three experiments, each experiment n =

69 a Stable 3.1 chimera-expressing clones were not obtained; transfectants seem to differentiate and stop growing. b Significant with respect to corresponding SFM (p<0.02). These nnr5 survival data are generally consistent with the Tyr(P) results for HEK293 transfectants co-expressing p75 NTR (Fig.3). However, it is noteworthy that early (12 min) and late (2-3 days) read-outs are not necessarily linear and kinetics of short term receptor activation could vary leading to different functional outcomes (Saragovi et al., 1998) (Cunningham et al., 1997). Complementary Tyr(P) studies could not be easily carried out in the nnr5 transfectants, because they express very low levels of chimeric receptors. The main differences between survival of nnr5 transfectants (p75 NTR + ) and HEK293 transfectants (p75 NTR+ ) are that (i) Cells became responsive to NGF, while NT-3 and BDNF responsiveness was significantly decreased in the nnr5 1.1 chimera; (ii) BDNF significantly decreased function in the nnr5 2.1 chimera; (iii) all ligands significantly decreased function in the nnr5 2.2 chimera; and (iv) NGF (but not NT-3) decreased function in the nnr5 4.1 chimera (Table III). 69

70 The TrkA-D4 Subdomain Is Surface-exposed in the Receptor Extracellular Domain Given the intriguing results seen for TrkA D4, we predicted that this subdomain may be surface-exposed in the tertiary structure of the receptor. This hypothesis is based on the fact that surface structures generally correlate with access to a ligand or with the ability to induce conformational changes (Saragovi et al., 1992; Saragovi et al., 1999). Immunogenicity is an accepted criterion for testing surface exposure of receptors (Morgan and Roth, 1986; Soos et al., 1986). We used anti-trka antiserum RTA generated by immunization with whole TrkA- ECD (Clary et al., 1994). The serum was affinity-purified using native TrkA-ECD to trap RTA immunoglobulins. These antibodies react only with surface-exposed epitopes of the extracellular domain, because the immunoglobulins directed to buried or masked epitopes would not have been affinity-purified. HEK293 cells expressing the Trk chimeras were tested by Western blotting with purified RTA immunoglobulins (Fig. 4). Chimeras 2.2 (containing TrkA D1, D2, and D4) and 3.1 (containing TrkA D4) show high reactivity to RTA serum, while chimera 4.1 (containing TrkA D5) shows lower reactivity. Untransfected HEK293 cells and chimeras 1.1 and 2.1 show no reactivity with purified RTA immunoglobulins. Since the 2.2 and the 3.1 chimera share the D4 subdomain, the data indicate that the D4 subdomain and to a much lower degree the D5 subdomain are surface-exposed in the architecture of the TrkA receptor. 70

71 Figure 4 Surface exposure of subdomains of extracellular TrkA. Lysates of cells untransfected (lane 1) or expressing the indicated chimeric Trk receptors or wild type (wt) TrkA (lanes 2-7), were analyzed by Western blotting (WB) with RTA antiserum. RTA antiserum recognizes surface receptor epitopes in the extracellular domain of TrkA with no reactivity toward TrkB. Bands of different size are seen corresponding to the different TrkB/TrkA chimeras. Table III Summary of survival and Tyr(P) data Relative ligand-dependent activation of TrkA/B chimeras in the presence or absence of p75 NTR expression. a High > medium > low > nil. Summarized from ptyr(p) and cell survival assays. These two assays are generally consistent but do not necessarily have linear read-outs, and their kinetics may vary (Cunningham et al., 1997; Saragovi et al., 1998). Italics and boldface type show very significant changes. Underlining shows significant changes. b Activation above ligand- independent activation. 2.6 Discussion Our studies define four issues of neurotrophin receptor biology. First, we show that the D4 subdomain is surface-exposed and plays a regulatory role in receptor activation and perhaps in low affinity ligand binding. Second, NGF binds to the D5 subdomain of TrkA. 71

72 Third, we identify extracellular subdomains relevant for TrkA or TrkB activation by cognate (NGF or BDNF, respectively) and heterologous (NT-3) ligands. Fourth, we show that p75 NTR modulates the Trk subdomains necessary for ligand-dependent activation. TrkB Activation Subdomains of BDNF and NT-3 Based on decrease or loss of BDNF and NT-3 responsiveness in the 4.1 chimera (compared with wild type receptors), we conclude that BDNF-TrkB and NT-3-TrkB activation is mediated by the D5 subdomain. Based on a decrease in BDNF and NT-3 responsiveness in the 2.2 chimera compared with the 2.1 chimera, we conclude that BDNF- TrkB and NT-3-TrkB activation also requires the D4 subdomain. The functional role of the D4 subdomain is further supported by ligand-independent activation of the 3.1 chimera. In related studies of TrkB/TrkC chimeras, only the D5 subdomain of TrkB bound BDNF, and the D4 did not (Urfer et al., 1995). If the D4 affects activity in the absence of direct ligand binding, this would suggest a regulatory or allosteric role in maintaining receptor conformation, and this possibility is supported by a surface exposure of D4 within the Trk architecture. The Role of p75 NTR in BDNF and NT-3 Activation The subdomains used by BDNF and NT-3 to activate TrkB change when p75 NTR is co-expressed. In the presence of p75 NTR co-receptors, the N-terminal subdomains of TrkB (D2 and D3) are necessary for optimal activation by BDNF (Fig. 3, compare B with G). Also, whether or not p75 NTR co-receptors are present, NT-3 activates the chimeric 4.1 receptor (D5 subdomain of TrkA) to a level comparable with wild type TrkA but different from wild type TrkB. 72

73 The presence of p75 NTR reportedly makes NT-3 inefficient at activating via wild type TrkA (Brennan et al., 1999). Therefore, the role of p75 NTR in enhancing NT-3 activation of the 4.1 chimera could be due either to enhanced NT3-TrkA-D5 interactions or to NT-3 binding to other subdomains of TrkB (Ninkina et al., 1997). Changes induced by p75 NTR in Trk subdomain ligand docking sites explain the mechanism wherein BDNF and NT-3 activate wild type TrkB equally, but in the presence of p75 NTR co-receptors BDNF activates better than NT-3 (Bibel et al., 1999). TrkA Binding and Activation Subdomains of NGF and the Role of p75 NTR Compared with wild type TrkA, in the 4.1 chimera NGF-TrkA activation is optimal. Thus, NGF responsiveness occurs via the D5 subdomain of TrkA when p75 NTR co-receptors are not expressed. However, in the presence of p75 NTR, NGF activates the 4.1 chimera to a lower level than it does with wild type TrkA. Therefore, p75 NTR restricts the efficacy of activation in this chimera. In contrast, when p75 NTR co-receptors are expressed, the 1.1 chimera becomes activated by NGF. This change in subdomain utilization may involve a p75 NTR -mediated allosteric regulation of Trk receptors or a direct interaction of p75 NTR with the TrkA D1 motif. Our data are consistent with the literature, where the interactions of NGFp75 NTR enhance wild type TrkA ligand selectivity and affinity (Jing et al., 1992; Barker and Shooter, 1994; Mahadeo et al., 1994) and affect TrkA biological activity (MacPhee and Barker, 1997; Maliartchouk and Saragovi, 1997; Saragovi et al., 1998; Kohn et al., 1999). Additionally, we cannot exclude a conformational effect of p75 NTR upon the neurotrophins themselves. This has been proposed previously, wherein an anti-ngf antibody blocked all of the neuritogenic but not the trophic action of NGF (MacPhee and Barker, 1997; Maliartchouk and Saragovi, 1997; Saragovi et al., 1998; Kohn et al., 1999). 73

74 Regulation of Activity by the D4 Subdomain It is generally accepted that receptor dimerization and conformational changes must occur for activation and signaling. Regulatory subdomains have been shown to inhibit ligandindependent dimerization activation of other receptors (Livnah et al., 1998; Remy et al., 1999), and the steric constraints are removed by conformational changes upon ligand binding. The D4 subdomain of TrkA plays a key role in regulation of ligand-independent activation. It is also attractive to speculate that the p75 NTR -mediated reduction of ligandindependent activation seen in all Trks may be due to an interaction of p75 NTR with the D4 subdomain. The constitutive activation seen in the 3.1 chimera suggests either a negative regulation by TrkB D4 (which has been replaced) or an activating property specific to TrkA D4 (which has been introduced). Like the 3.1 chimera, the 2.2 chimera also contains the TrkA D4 subdomain but is not constitutively activated. Hence, it is possible that the additional TrkA subdomains in the 2.2 chimera (D2 and D3) restrict ligand-independent activation by the D4 subdomain. TrkA D4 deletions and a mutation also show constitutive activation (Arevalo et al., 2000; Arevalo et al., 2001), as does a Cys to Ser substitution in the D5 subdomain (Coulier et al., 1990). In addition, others proposed a role for the D4 subdomain in neurotrophin binding (Urfer et al., 1995; Urfer et al., 1998). Together these data point to a critical role of the D4 domain in regulation of function. Conclusions and Putative Model We conclude that (i) Trk receptors have distinct ligand binding and regulatory subdomains, (ii) p75 NTR -mediated regulation of activation requires the N-terminal subdomains of Trk, (iii) there are multiple NTF docking sites for Trk (some are preferred and some allosteric (perhaps corresponding to high and low affinity, respectively)), and (iv) 74

75 regulation of activation by the TrkA -D4 subdomain is different from regulation by the TrkB- D4 subdomain, based on the 3.1 chimera effect. We postulate a possible model of two binding sites on Trk for NTFs, modulated by the co-receptor p75 NTR (Fig. 5). In this model, receptor stoichiometry and possible allosteric changes are simplified or not shown. NGF binding to the D5 domain causes a conformational change in wild type TrkA (Fig. 5B). The conformational effect on the architecture may occur at the surface-exposed D4 subdomain, which could "rotate", as described for other receptors (Wells, 1996; Livnah et al., 1998; Remy et al., 1999). Fig. 5. Hypothetical model of TrkA-NGF-p75 NTR interactions. Receptor stoichiometry and possible allosteric changes are simplified or not shown. Subdomain-subdomain interactions could occur either intrachain (as shown) or interchain between two TrkA dimers. A putative interaction of p75 NTR and Trk could occur at the N terminus or at the D4 subdomain of Trk. Simultaneous p75 NTR -TrkA binding to NGF is also hypothetical. 75

76 Chapter 3 TrkA receptor «hot spots» for NT-3 as a heterologous ligand 76

77 3.1 Rational As described earlier and in chapter 2, Trk receptors have distinct ligand binding and regulatory subdomains for neurotrophins, and p75 NTR modulates ligand binding and activation subdomains of Trks for their cognate and heterologous ligands. In the following chapter, the binding and functional receptor activation of TrkA receptor by the heterologous ligand NT-3 is addressed. This interaction is of physiological importance, because it has been shown that both NGF and NT-3 regulate different stages of development of sympathetic neurons; NGF induces survival and axonal growth, while NT-3 induces only axonal growth via TrkA receptor (Kuruvilla et al., 2004). Defining TrkA binding and functional activation sites for NT-3 will give further insight into complex mechanisms of Trk receptor regulation. Furthermore, defining the domains of TrkA receptor that bind and are activated by NT-3 will enable rational design of small molecules that would mimic NT-3 activation on TrkA receptors or potentiate the NT-3 induced TrkA activation (Zaccaro et al., 2005). These small molecules can also help us understand the role of TrkA activation by NT-3 in vivo. Understanding the interactions between NGF and NT-3 hot spots on TrkA receptors can enable selective targeting of specific TrkA receptor domains and modulation of NGF induced TrkA activation. Understanding the functional consequences of specific domain activation allows for the development of potential therapeutic applications. 77

78 3.2 Abstract Neurotrophins signal via Trk tyrosine kinase receptors. NGF is the cognate ligand for TrkA, BDNF for TrkB, and NT-3 for TrkC. NT-3 also binds TrkA as a lower affinity heterologous ligand. Because NT-3 interactions with TrkA are biologically relevant, we aimed to define the TrkA hot spot functional docking sites of NT-3. The Trk extracellular domain consists of two cysteine rich subdomains (D1 and D3), flanking a leucine rich subdomain (D2), and two immunoglobulin-like subdomains IgC1 (D4) and IgC2 (D5). Previously, the D5 subdomain was defined as the primary ligand binding site of neurotrophins for their cognate receptors (e.g. NGF binds and activates through TrkA-D5 hot spots). Here, binding studies with truncated and chimeric extracellular subdomains show that TrkA-D5 also comprises an NT-3 docking and activation hot spot (site 1), and competition studies show that the NGF and NT-3 hot spots on TrkA-D5 are distinct but partially overlapping. In addition, ligand binding studies provide evidence for an NT-3 binding/allosteric site on TrkA-D4 (site 2). NT-3 docking on sites 1 and/or 2 partially blocks NGF binding. Functional survival studies showed that sites 1 and 2 regulate TrkA activation. NT-3 docking on both sites 1 and 2 affords full agonism, which can be additive with NGF activation of Trk. However, NT-3 docking solely on site 1 is partially agonistic, but noncompetitively antagonizes NGF binding and activation of Trk. This study demonstrates that Trk signaling is more complex than previously thought because it involves several receptor subdomains and hot spots. 78

79 3.3 Introduction The neurotrophins are a family of growth factors that regulate proliferation, survival, death, and differentiation of neurons in the embryonic and early postnatal stages; and neuronal maintenance, synaptic activity, and learning later in life (Eide et al., 1993; Lu et al., 2005). Nerve growth factor (NGF), neurotrophin-3 (NT-3), neurotrophin-4 and brain-derived neurotrophic factor (BDNF) are members of the neurotrophin family of polypeptides. Neurotrophins act by binding to two distinct classes of transmembrane receptors. One is the p75 NTR neurotrophin receptor and the other is the Trk family of tyrosine kinase receptors, which includes TrkA, TrkB and TrkC. All mature neurotrophins bind to p75 NTR but Trks are more selective. NGF interacts selectively with TrkA receptors and BDNF selectively with TrkB receptors. While NT-3 interacts with TrkC receptors preferably, it is promiscuous and can also bind TrkA and TrkB with lower affinity (Barbacid, 1994). The Trk receptors mediate most of the "positive" survival and differentiation signals typically associated with neurotrophin activity. Nevertheless, NT-3 TrkA interactions are biologically relevant. For example, during development NGF and NT-3 act coordinately to select TrkA-expressing sympathetic neurons (Tessarollo et al., 1997). Furthermore, sympathetic neurons from NGF null and NT-3 null mice die when these neurons express high levels of TrkA and negligible levels of TrkC (Francis et al., 1999). NT-3 mediates neuritogenesis as effectively as NGF, but affords ~20-40 fold less survival compared to NGF in NGF-dependent sympathetic neurons (Belliveau et al., 1997). Also, NGF and NT-3 acting via TrkA are required for sympathetic axon growth and target innervation (Kuruvilla et al., 2004). 79

80 The extracellular domain of Trk receptors features five subdomains defined by their homology to other proteins (Schneider and Schweiger, 1991). Near the N-terminus is a leucine-rich motif (LRM, also known as D2), flanked by two cysteine-rich clusters (Cys-1 and Cys-2, also known as D1 and D3 respectively). Closer to the transmembrane region, there are two immunoglobulin-like subdomains termed Ig-C1 (D4) and Ig-C2 (D5) (Figure 1). NTFs transduce trophic signals by binding to the extracellular domain of Trk and stabilizing receptor homodimerization (Mischel et al., 2002). It is postulated that a conformational change is induced upon ligand binding, which then activates the intrinsic intracellular tyrosine kinase catalytic activity (Miller and Kaplan, 2001). Previous work showed that the D5 subdomains of TrkA, TrkB and TrkC are relevant for binding NGF, BDNF and NT-3 respectively (Perez et al., 1995; Urfer et al., 1995; MacDonald and Meakin, 1996; Urfer et al., 1998; Banfield et al., 2001), and a recombinant polypeptide spanning TrkA D4 and D5 (TrkA-D4-D5) bound and neutralized NGF (Holden et al., 1997). Moreover, it has also been shown that the D4 subdomain has a role in regulating receptor dimerization (Arevalo et al., 2001) and in constitutive activation of the receptors (Zaccaro et al., 2001). The D1 subdomain of TrkA can be utilized by NGF to activate the receptor, through a potentially allosteric or conformational mechanism regulated by coexpression of p75 NTR on the cell surface (Zaccaro et al., 2001). The fact that NT-3 has a lower affinity interaction with its heterologous receptors TrkA and TrkB (MacDonald and Meakin, 1996; Ryden and Ibanez, 1996), and that the biological outcomes of NT-3 TrkA interactions are limited (e.g. partial survival, or limited differentiation) (Belliveau et al., 1997) suggests three non-exclusive possibilities: (i) NT-3 as heterologous ligand may lack some docking sites on TrkA for high affinity binding and full 80

81 function; (ii) the interaction may take place through epitopes that are distinct from those used by NGF; and/or (iii) receptor conformational or allosteric regulation by each ligand may be different. The goal of our study was to address these possibilities by determining the binding and functional profile of the extracellular domain of TrkA towards cognate NGF or heterologous NT-3 ligands. We report that most of the TrkA activation by cognate or heterologous ligands occurs via the D5 subdomain (site 1). NT-3 binds and activates also at TrkA-D4. Binding of NT-3 to each site seems to be associated with a particular mode of TrkA activation, which can be antagonistic or additive towards NGF. Our study demonstrates that Trk binding and signaling is complex since it involves several receptor subdomains, including potentially allosteric sites that may have positive or negative effects and may be neurotrophin specific. 3.4 Materials and methods Neurotrophins Recombinant human neurotrophin-3 (NT-3) produced in E.coli was purchased from Prospec-Tany TechnoGene LTD (Rehovot, Israel). Nerve growth factor isolated from mouse submaxilliary gland was purchased from Prince Labs (Toronto, Canada). Human TrkA-rat TrkB chimeras The chimeric receptors were constructed by subcloning each TrkA domain into the corresponding unique restriction sites of the rat trkb cdna reported in previous work (Perez et al., 1995). The chimeric construct was confirmed by sequencing and was cloned into the pcdna3 expression vector that contains the neomycin (GIBCO) selection marker. 81

82 Transfection and stable expression in 293 cells HEK293 cells (human kidney epithelium, TrkA, p75 ) were transfected with the chimeric cdna construct using the lipofectamine-plus method (GIBCO). Quantitative western blot analysis (Mufson et al., 1997) with a polyclonal antibody directed to the Trk intracellular domain (203 antisera, a gift of David Kaplan, University of Toronto) indicated that the stable HEK TrkA/B cells express 40, ,000 chimeric receptors/cell (data not shown) (Zaccaro et al., 2001). NIH 3T3 TrkA cells are stably transfected with human TrkA wt receptor. All cells were grown in RPMI 1640 supplemented with 5% fetal calf serum, antibiotics and glutamine. Selection for the pcdna3 expression vector was maintained at 0.4 mg/ml neomycin. Cell survival assays Cell survival was measured by quantitative tetrazolium salt reagent (MTT, Sigma) and optical density (OD) readings as described (Maliartchouk and Saragovi, 1997). Cells were plated in 96 well plates (Becton Dickinson, Lincoln Park, NJ) at ~5,000 cells/well in serum free media (SFM) (PFHM-II, GIBCO, Toronto), with 0.2% bovine serum albumin (BSA) (Boehringer Mannheim). Wells were then supplemented with SFM (negative control), with the indicated concentration of the indicated neurotrophin in SFM (test), or with 5% serum (positive control) for ~48 h. Cell growth/survival was calculated relative to 2 nm NGF (standardized to 100%). All assays were repeated at least three times, n=4-6 per assay. Primary Antibodies Anti-TrkA mouse mab 5C3 binds to the human TrkA-D5 subdomain with Kd ~2 nm and acts as a full agonist of this receptor (LeSauteur et al., 1996). Mouse NGF30 mab is an anti-ngf mab that when bound to NGF still allows high affinity binding of NGF to TrkA (2 nm). The interaction of NGF NGF30 complexes with TrkA affords survival but not differentiation of TrkA expressing cells (Saragovi et al., 1998). 82

83 NGF NGF30-FITC complexes are used at 2-fold excess of mab (20 nm NGF plus 40 nm NGF30 mab). Flow cytometric analysis 293 HEK cells expressing 4.1 TrkA/B receptor or NIH-3T3 cells expressing TrkA wt receptor (2X10 5 cells/tube) were re-suspended in 50 μl of FACScan buffer (phosphate-buffered saline, 0.5% BSA, and 0.1% NaN 3 ). Cells were untreated or treated with competitors for 20 at 4 C. Then, saturating concentrations of primary antibodies were added without washing and incubated for 20 at 4 C. Excess primary antibody was washed off, and cells were immunostained with fluoresceinated goat anti-mouse IgG (FITC- G-a-M). For background staining control, non-specific mouse IgG or NGF30-FITC antibody in the absence of NGF were used. Maximal binding was obtained by staining with mab 5C3 or NGF NGF30 complex without pre-incubation with neurotrophins. Cells were acquired on a FACScan, and bell-shaped histograms were analyzed using the CellQuest program. Expression of complete TrkA-extracellular domain and TrkA-D4-D5 domains Human TrkA-extracellular domain (TrkA-ECD) was produced in baculovirus. TrkA-ECD contains ~25 kda of carbohydrate modifications and binds NGF with 3 nm affinity (Saragovi et al., 1998; Woo et al., 1998). The Ig-C1 (D4) and Ig-C2 (D5) subdomains of human TrkA (TrkA- D4-D5) were produced in yeast. TrkA-D4-D5 was cloned by digestion of human trka cdna with BswI-BbsI, and a 644 bp fragment (coding for aminoacids ) was isolated (Quiagen, Mississauga, ON, Canada). The fragment was blunt ended and ligated to ppiczab vector (Invitrogen, Carlsbad, CA) that was predigested with PmeI. A tag containing myc sequences and 6 histidines (myc-his-tag) was put in frame at the C-terminus of the protein by double digestion with KpnI-XbaI, followed by blunting and ligation. Clones were selected using zeocin antibiotic, and verified by diagnostic restriction enzyme analysis and 83

84 sequencing. Pichia yeast strain KM71 were transformed and selected in zeocin antibiotic. Expression was carried out for 4 days in buffered methanol-complex medium (BMMY medium, 1% yeast extract, 2% peptone, 1.34% yeast nitrogen base, 4x10-5 biotin, 0.5% methanol). The supernatant was concentrated (Millipore, Bedford, MA, USA) and the secreted His-Myc-TrkA-D4-D5 was purified using Talon resin, which binds the His-tag (Clontech, Palo Alto, CA). SDS-PAGE and western blotting of the purified protein using anti-myc antibodies (Invitrogen, Carlsbad, CA) revealed the presence of a single band of ~55 kda containing the myc-tag. The purity of TrkA-D4-D5 was ascertained by silver staining (BioRad, Hercules, CA). The protein contains ~25kDa of carbohydrate post-translational modifications. Quantification of TrkA-ECD and TrkA-D4-D5 Because of the differences in glycosylation and refolding machinery between the baculovirus and the yeast expression systems, it was expected that there could be qualitative differences between TrkA-ECD and TrkA-D4-D5. Therefore, prior to binding studies, we used mab 5C3 (LeSauteur et al., 1996) that binds a conformationally sensitive epitope near the NGF binding site of these human recombinant proteins. ELISA studies were done to quantify appropriately folded TrkA-ECD and TrkA-D4-D5 as described previously (Saragovi et al., 1998). 125 I[NGF] binding studies on HEK293 cells expressing 4.1 chimeric receptors. Scatchard plot analysis of 125I[NGF] binding to cells was done as described (LeSauteur et al., 1996; Saragovi et al., 1998). Viable HEK293 cells expressing the 4.1 chimera (with TrkB-D5 subdomain exchanged with the same subdomain of TrkA) were used as test, and viable wild type HEK293 cells were used to determine non-specific 125I[NGF] binding. Non-specific binding was always <20% of total binding and was subtracted. All dilutions and 84

85 washes were done in binding buffer (PBS, 1% BSA, 0.05% sodium azide), and all procedures were carried out on ice. Cells (5x105/100 μl) were added to 100 μl serial dilutions of 125I[NGF] in the absence or presence of a constant ratio of 250-fold excess of cold NTF's as competitors. Final volumes of 200 μl were incubated for 30 min with occasional shaking. Fractionation of cell-bound and free counts was done in two alternative ways, both yielding comparable data. In one method, we performed a single 1 ml wash, and then each resuspended cell pellet (100 μl) onto a 1.0 ml of a serum/sucrose gradient in a new tube. Centrifugation through this gradient separates cells (and cell-bound cpm) from supernatant. The tubes were frozen in dry ice immediately after centrifugation, and while frozen, they were cut to separate cells (bottom pellet) from supernatant (top layer). All washes and supernatants were collected for counting (free cpms). In the other method, three 1 ml washes were done in a fresh tube and supernatants (free) and pellet (bound) were collected and counted. Biacore binding studies The rat TrkA-rat TrkB 4.1 ECD chimera was generated by polymerase chain reaction using the full length rat TrkA-rat TrkB receptor chimeric cdnas (Zaccaro et al., 2001) and cloned into pbluebac4.5/v5-his vector (Invitrogen, Carlsbad, CA). The TrkA-ECD or chimeric ECD was overexpressed in Sf21 insect cells in XL-401 medium (JRH Biosciences, Lenexa, KA) as described previously (Woo et al., 1998). Proteins were purified with their C-termini (His) 6 tag by Ni-NTA chromatography and subsequent steps to near homogeneity as shown with SDS-PAGE gels stained with Coomassie Blue (Woo et al., 1998) (Woo et al, 2007, in preparation). The Trk-ECD wild type or 4.1 chimera was immobilized on a CM4 sensor chip using EDC/NHS coupling chemistry and analyzed 85

86 on a Biacore 3000 instrument (Biacore, Piscataway, NJ) using BIA Evaluation software (version 4.0.1, Biacore). Trk and Akt activation NIH-3T3 TrkA wt or HEK TrkA/B chimeric receptors expressing cells were collected (5x10 5 /point), washed with PBS and incubated in SFM at 37 C for 30 to reduce background receptor phosphorylation levels. The kinetics of activation of each protein (TrkA wt or 4.1 TrkA/B, and Akt) was followed after treatment of rested, live cells with different concentrations of ligands for 3, 12, or 30 at 37 C. Cells were washed in ice-cold PBS, lysed in detergent lysis buffer (1% NP-40, 20 mm Tris ph 7.5, 137 mm NaCl, 2 mm EDTA, 10 mm benzamidine, 50 mm sodium orthovanadate, 10 mg/ml leupeptin, 2 mg/ml soybean trypsin inhibitor, 1 mm iodoacetamide, and 10 mg/ml aprotinin) and protein concentrations were determined with Bio-Rad Detergent Compatible Protein Assay (Bio-Rad Laboratories). Western blot analysis was performed with antiphosphotyrosine (ptyr) antibody 4G10 (Upstate Biotechnology, Inc., Lake Placid, NY), antiphospho-trkatyr 490 (Cell Signaling) and anti-phospho-akt (Ser473) antibody (New England Biolabs). Blots were visualized using the enhanced chemi-luminiscence system (PerkinElmer Life Sciences). Re-blotting the membranes with anti-actin antibody confirmed equal loading. Enzyme-linked Immunosorbent Assays (ELISA) TrkA-ECD (10 ng/well), yeast TrkA-D4-D5 (50 ng/well) or negative control proteins were immobilized onto 96-well microtest ELISA plates (Becton Dickinson, Lincoln Park, NJ), followed by blocking with blocking buffer (PBS with 1% BSA) for 1 hour. The indicated concentrations of NT-3 or control proteins were added, and then NGF NGF30 complex was added. ELISA assay was performed as described (Saragovi et al., 1998). Representative assay of three independent 86

87 experiments is shown, n=4 per assay. Throughout, anti-trka mab 5C3 was used to control the presence of appropriately folded TrkA-ECD or TrkA-D4-D5 on plates. Statistical analysis The data were analyzed by two tailed student s t-test and p values are reported. Significance is p 0.05 and is indicated in appropriate figures by an asterisk (*). 3.5 Results Chimeras induce trophic signals in response to neurotrophins Full length TrkB receptors expressing the D5 (Ig-C2) subdomain of human TrkA were generated and the resulting chimera was termed 4.1 (Zaccaro et al., 2001). The 4.1 TrkA/B chimeric receptor cdna, and TrkA wild type receptor cdna were stably transfected in HEK293 and NIH cells. The structure and domain organization of TrkA wt and 4.1 TrkA/B chimeric receptor are illustrated in Figure 1A. Cells cultured in SFM undergo apoptotic death, which can be rescued by neurotrophins if the cells express functional neurotrophin receptors. We have confirmed (data not shown) the differential effect of NT3 and NGF with various receptor constructs (Zaccaro et al., 2001). In wild type TrkA-expressing cells, 2 nm NGF affords optimal survival. NT-3 also protects wild type TrkA-expressing cells but with lower potency. NT-3 at 2 nm protects ~20% and at 500 nm protects ~100%. In cells expressing the 4.1 TrkA/B chimera 2 nm NGF affords optimal survival of 100%. NT-3 protects 4.1 TrkA/B chimera-expressing cells but to a significantly lesser degree. NT-3 at 2 nm protects ~20% and at 500 nm protects ~30%. Hence, NGF activates fully via the TrkA-D5 subdomain whereas NT-3 activates only partially via the TrkA-D5 subdomain, or via the TrkB-D1-D4 subdomains of the chimera. 87

88 To understand the relationship between NT-3 and NGF docking sites on TrkA, we carried out direct ligand binding on purified proteins comprising either the 4.1 TrkA/B ECD or the whole TrkA extracellular domain. Figure 1. (A) Model of TrkA wt and 4.1 TrkA/B chimera. Human TrkB (white) was engineered so that the D5 subdomain was replaced by the same subdomain of the rat TrkA (black) receptor. The resulting chimera is termed 4.1. (B) Model of human TrkA-ECD and TrkA-D4-D5 proteins with indicated binding or possible allosteric sites for NGF, NT-3, mab 5C3 and NGF NGF30 complex. (C) Putative model of NT-3 binding and activation sites on TrkA-ECD. Direct ligand-receptor binding studies Biacore surface plasmon resonance was used to obtain k on and k off values for NGF and NT-3 binding to TrkA-ECD or TrkA/B 4.1 chimera ECD, and to obtain K d values for these ligands. NGF binds to TrkA-ECD with K d 1.91 ± 0.03 nm. On the other hand, NT-3 binds to TrkA-ECD with K d 131 ± 22 nm. The on rates were similar but the k off was significantly higher for NT3, leading to the resultant higher Kd (Table 1). The biacore data for NGF binding to 4.1 TrkA/B chimera ECD shows K d 292 ± 6 pm; but NT-3 binding to this chimeric protein was beyond the limit of detection (>300 nm) and was not possible to determine a K d. Thus, the main subdomain for NGF and NT-3 binding is TrkA D5, but other subdomains affect the affinity or the kinetics. 88

89 Table 1. Summary of binding data N.D. not detected (detection limit is 300 nm) Purified proteins IC 50 (nm) Cell lines NT-3 Competition of TrkA-ECD TrkA-D4-D5 TrkA wt 4.1 TrkA/B ligand binding NGF not done not done NGF NGF >500 BIACORE direct Ligand Binding TrkA-ECD NGF 4.1 TrkA/B ECD TrkA-ECD NT TrkA/B ECD Kd [nm] 1.91 ± ± ± 22 >300 k on x ± ± ± 0.06 N.D. k off x ± ± ± 12.5 N.D. NT-3 competition of NGF-receptor binding Next, we performed Scatchard analysis of 125 I[NGF] binding to 4.1 TrkA/Bexpressing HEK293 cells to further assess NT-3 and NGF interactions at the D5 subdomain (Figure 2). We previously demonstrated high affinity 125 I[NGF] binding to 4.1 TrkA/B chimeric receptor expressing cells. This clone expresses ~50,000 chimeric receptors/cell for specific binding with a K d 80 ± 30 pm for NGF (Zaccaro et al., 2001). All of the binding sites were blocked by excess unlabeled NGF. Non-specific control proteins did not affect 125 I[NGF] binding (data not shown). In cells expressing chimeric 4.1 receptors, addition of 250-fold excess of unlabeled NT-3 reduced the NGF binding sites in the 4.1 chimera by 30-50%, without substantially affecting the NGF binding affinity of remaining sites (Figure 2). Higher concentrations of 89

90 NT-3 did not increase the degree of competition (data not shown). Hence, the interaction of NT-3 with the 4.1 chimera protein is of low affinity and non-competitive towards NGF. On the other hand, the binding of NT-3 to wild type TrkA is between 10 to 60 fold lower than NGF (our data and ref. (Ryden and Ibanez, 1996). Direct binding studies with 125 I[NT-3] could not be carried out because this labeled ligand was unavailable to us. Overall, these data suggest that there are NT-3 binding or allosteric sites on TrkA in subdomains other than TrkA-D5. However, these data cannot discriminate whether NT-3 blocking of NGF is due to steric hindrance (binding to an overlapping site of TrkA-D5) or because NT-3 induces receptor conformational changes. Figure I[NGF] binds to surface 4.1 TrkA/B receptors and 125 I[NGF] binding is inhibited by NT-3. Scatchard plot of 125 I[NGF] binding to cell surface 4.1 TrkA/B expressing cells was performed by adding the serial dilutions of 125 I[NGF] in the absence (black squares) or presence of a 250 fold excess of cold NT-3 (black triangles) or NGF (white circles) as competitors. The data points were fitted to a logarithmic function in Excel. In independent assays NT-3 blocking of 125 I[NGF] binding sites (but not altering affinity) 90

91 ranged from %. Data for 125 I[NGF] binding and excess of cold NGF are from reference (Zaccaro et al., 2001) and were performed in the same experiment with NT-3. NGF and NT-3 bind to the D5 subdomain of TrkA A quantitative flowcytometric assay and analysis measured competition between TrkA ligands, using TrkA wt receptors expressed on transfected NIH-3T3 cells. The binding of anti-trka mab 5C3, an agonistic TrkA ligand which binds at the D5 subdomain and saturates TrkA at 65 nm (LeSauteur et al., 1996) was blocked by NGF and NT-3 to different levels. NGF at 50 nm inhibited ~20% of 5C3 (65 nm) binding (Figure 3A, 3C). NT-3 at 50 nm did not compete with 5C3 binding, but 500 nm NT-3 inhibited 5C3 binding by ~20% (Figure 3B, 3D). These results further suggest that both NGF and NT-3 have binding sites in TrkA-D5. However, to achieve the same degree of NGF inhibition of 5C3 binding, it is necessary to use 10-fold higher concentrations of NT-3, suggesting lower affinity binding of NT-3 to TrkA D5, a different epitope, or differential hindrance. Figure 3. NGF and NT-3 compete 5C3 binding to NIH-3T3 TrkA wt expressing cells Different concentrations of NGF (A,B) or NT-3 (C,D) were added to TrkA wt cells and mab 5C3 binding was assessed by FACScan. Maximal binding is achieved in the absence of competitors. Control non-binding IgG was used as background. 5,000 cells were acquired for 91

92 each condition. Representative bell-shaped histograms (A,C) and bar graphs (B,D) summarizing data from three independent assays ± sem are shown. The p-value is given relative to the control labeled NONE. Possible NT-3 binding to TrkA D4 subdomain To define further the NT-3 binding site on TrkA, we used another TrkA ligand, namely the NGF NGF30 complex. NGF NGF30 complex is composed of NGF bound by NGF30 anti-ngf mab. The complex has different binding and receptor activation profile than free NGF. NGF NGF30 binds to TrkA with high affinity, but it does not bind p75. The complex induces cell survival but not differentiation (Saragovi et al., 1998). We first tested the binding of NGF NGF30 complex to TrkA-ECD and TrkA-D4-D5 purified proteins by ELISA, and then tested whether NT-3 competes with NGF NGF30 binding. NGF NGF30 complex bound to both purified TrkA-ECD and to TrkA- D4-D5. NT- 3 competed NGF NGF30 binding to TrkA-ECD (Figure 4A) and TrkA-D4-D5 (Figure 4B) in a dose-dependent manner and with similar profiles. Both receptor fragments were competed equally well by NT-3 from binding NGF NGF30. NT-3 2 nm blocked ~30% binding to both receptor fragments and NT-3 20 nm blocked ~45% binding to both receptor fragments. Together with the previous results, these data suggest that NT-3 has a second binding site on TrkA-D4. To confirm the role of TrkA-D4 in NT-3 binding, we tested on receptorexpressing cells the direct binding of NGF NGF30 complex to NIH-3T3 TrkA wt and HEK TrkA/B chimeras, and competition by NT-3. The NGF NGF30 complex binds equally well to TrkA wt and 4.1 TrkA/B chimeras. However, while NGF NGF30 binding to wt TrkA was competed efficiently by 50 nm NT-3 (Figure 4C), similar competition on the 4.1 TrkA/B chimera required more than 25-fold greater NT-3 (Figure 4E). 92

93 Efficient NT-3 competition of NGF NGF30 binding to TrkA wt cells correlates well with competition of binding to TrkA-ECD by ELISA. However, poor NT-3 competition of NGF NGF30 binding to 4.1 TrkA/B chimeras differs from the efficient competition of TrkA-D4-D5 by ELISA. Differential blocking of 4.1 TrkA/B chimeras (containing only TrkA-D5) and TrkA-D4-D5 further suggests a role of TrkA-D4 in NT-3 binding that can not be compensated by TrkB-D4 in the chimeric protein. Figure 4. NT-3 inhibits NGF NGF30 binding to TrkA-ECD and TrkA-D4-D5 similarly. (A-B) ELISA assay of NT-3 competition with NGF-NGF30. Competition assay to purified proteins was performed by ELISA. Bar graphs of representative experiments ± sem are shown in (A) TrkA-ECD and (B) TrkA-D4-D5. (C-F) FACScan assay of NT-3 competition with NGF-NGF30 binding. (C) NT-3 competes NGF NGF30 binding to TrkA wt expressing cells dose dependently. 93

94 (D) Representative bell-shaped histogram of NT-3 competition of NGF NGF30 in TrkA wt expressing cells. (E) NT-3 competes NGF NGF30 binding to 4.1 TrkA/B expressing HEK293 cells only at 25 fold higher concentration of NT-3 than the concentration of NT-3 competing NGF-NGF30 binding to TrkA wt expessing cells. (F) Representative bell-shaped histogram of NT-3 competition of NGF NGF30 in 4.1 TrkA/B expressing cells. Maximal binding is achieved by 20 nm NGF-40 nm NGF30 complex in the absence of competitors. Background fluorescence was assessed by binding of 40 nm NGF30 alone. 5,000 cells were acquired for each condition. The p-value is given relative to the control labeled NONE. TrkA subdomains direct whether NT-3 is a potentiator or an antagonist of NGF To determine if NT-3 binding induces receptor activation, we compared functional responses in survival assays of 4.1 TrkA/B chimera versus TrkA wt expressing cells. Survival is maximally induced for both cell lines by 2 nm NGF (100%), whereas 500 nm NT-3 affords 100% survival for TrkA wt cells and 33% survival for 4.1 TrkA/B (Figure 5A). Survival is induced sub-optimally by 2 nm NT-3 (21% for both cell lines) or by 100 pm NGF (40% for both cell lines), and no survival is induced by 100 pm NT-3 (Figure 5A). These results are consistent with the estimated affinity of each ligand. Increasing NGF or NT-3 concentrations does not increase survival (data not shown). Next, we combined NGF and NT-3 as trophic ligands in survival assays. When a constant concentration of optimal NGF (2 nm) is applied together with different concentrations of NT-3 (100 pm, 2 nm, and 500 nm) the survival of 4.1 TrkA/B chimera is decreased to 36%, 40% and 72% respectively (Figure 5B). These decreases indicate that NT- 3 functionally antagonizes optimal NGF survival activity in the 4.1 TrkA/B chimera. Note that functional antagonism by NT-3 takes place in the absence of binding antagonism, (binding data from this chapter) because 100 pm NT-3 does not block the binding of 2 nm NGF. 94

95 In contrast, the survival/growth of TrkA wt cells induced by optimal NGF (2 nm) is increased by combining it with NT-3. NGF 2 nm plus NT pm affords 116%, NGF 2 nm plus NT-3 2 nm affords 121%, and NGF 2 nm plus NT nm affords 160% (Figure 5B). Combinations of NGF and NT-3 at low doses where each alone do not activate, do not afford activation, indicating that there is no synergism (data not shown). These data suggest that NT-3 can activate via receptor sites overlapping with but distinct from the NGF sites, possibly including a site at D4. Thus, the NT-3 binding site on D5 affords partial agonism, whereas NT-3 binding sites on D5 and D4 afford full agonism. See Table 2 for a summary of survival data. These results suggest that NT-3 acting in 4.1 TrkA/B chimera via TrkA-D5 is a pure partial agonist that can antagonize NGF. In contrast, NT-3 acting in wild type TrkA via D5 and other subdomains (likely D4) can be a full agonist which can be additive with NGF function. It would have been useful to study cellular differentiation in addition to survival, but HEK293 cells do not differentiate in response to NTFs. We have a PC12 variant expressing the 4.1 TrkA/B, but unfortunately it co-expresses p75 which can affect the binding and functional outcome (Brennan et al., 1999); therefore differentiation was not studied. 95

96 Figure 5. (A) NT-3 binding site on TrkA wt receptor is trophic. (B) NT-3 antagonizes optimal NGF activity in 4.1 TrkA/B cells, but enhances optimal NGF activity in TrkA wt cells. TrkA wt or 4.1 TrkA/B chimeric receptor expressing cells were treated with indicated ligands in serum free media for at total of 48 hrs. Survival was measured by MTT assay. Percent survival was calculated relative to optimal NGF (2 nm), 100% protection. The bar graphs representative of 4 independent experiments ± sem are shown, n=4 per assay. % Cell survival Trophic ligand (nm) 4.1 TrkA/B TrkA wt NGF (2) (500) NT-3 (2) (500) NGF + NT-3 (2) + (2) (2) + (500) Table 2. Summary of functional interactions between NT-3 and the TrkA-D5 or TrkA wt receptors. NT-3 in the 4.1 TrkA/B chimera NT-3 is functionally antagonistic with NGF, but in TrkA wild type it is additive with NGF. See Figure 5 for data ± sem. 96

97 Biochemical correlation of function Western blots using anti-phosphotyrosine antibodies after 12 minutes exposure to ligand(s) were used to gauge receptor activation. 2 nm NGF induces strong and comparable phosphorylation of TrkA and 4.1 TrkA/B receptors (Figure 6, lanes 5 and 13). In contrast, 500 nm NT-3 induces strong phospho-trka but very weak phospho-4.1 TrkA/B (Figure 6, lanes 3 and 11). In five independent experiments, the maximal phospho-4.1 TrkA/B induced by NT-3 was 37.3% ± 2.2 compared versus wild type TrkA normalized to actin after stripping and re-blotting the membranes. Combinations of 2 nm NGF plus doses of NT-3 were then tested. For TrkA phosphorylation, NT-3 is additive with, or potentiates, NGF signals (Figure 6, lanes 5 versus 6, 7). Quantification showed that 2 nm NGF plus 2 nm NT-3 affords 125% ± 7 of TrkA phosphorylation compared versus 2 nm NGF. The combination of 2 nm NGF plus 500 nm NT-3 yields the highest phospho-trka levels (Figure 6, lane 8), but comparable to 500 nm NT-3. In contrast, for 4.1 TrkA/B phosphorylation, NT-3 reduces NGF signals (Figure 6, lane 13 versus 14, 15, 16). Quantification showed that 2 nm NGF plus 500 nm NT-3 affords 54 % ± 2 of 4.1 TrkA/B phosphorylation compared versus 2 nm NGF. To further correlate biochemical data with survival in 4.1 TrkA/B cells, we also studied the kinetics of ligand-dependent activation of the pro-survival PI-3 kinase/akt pathway and receptor phosphorylation with anti-phospho-tyr490 (which recognizes the activated Shc-binding site of Trk). NGF induces rapid and sustained 4.1 TrkA/B phospho- Tyr490 (Figure 6, lanes 18, 19, 20), but the combination of NGF plus NT-3 induces a transient increased phospho-tyr490 at 3 that decreases to an intermediate level by 12 (Figure 6, lanes 24, 25, 26), while NT-3 alone does not induce a strong 4.1 TrkA/B phospho- Tyr490 (Figure 6, lanes 21, 22, 23). 97

98 Similar data were also obtained for Akt activation, which is known to induce cell survival. NGF causes sustained Akt activation in cells expressing 4.1 chimera, but the combination of NGF plus NT-3 induces lower Akt activation (reduced to ~ 50% at 3 and 12 ), and the phospho-akt levels are more comparable to NT-3 than to NGF. Thus, Trk total phosphorylation, phospho-tyr490, and p-akt correlate well with the survival response of TrkA wt or 4.1 TrkA/B-expressing cells to each ligand or ligand combination. Similar results were obtained when PLCγ activation was studied (data not shown). These results confirm the notion that NT-3 acting in 4.1 TrkA/B chimera via TrkA- D5 is a pure partial agonist that can antagonize NGF. In contrast, NT-3 acting in wild type TrkA via D5 and other subdomains (likely D4) can be a full agonist which can be additive with NGF function. 98

99 Figure 6. NGF, NT-3, or a combination thereof leads to differential kinetics and efficacy of activation. Western blots of cell lysates of NIH-3T3 TrkA wt and TrkA/B chimera containing HEK293 cells were probed with antibodies directed to total phosphotyrosine, Trk p-tyr490, or phospho-akt. Stripping and re-blotting of the membranes with anti-actin antibody were used to confirm loading and to quantify gel scanning, n=3-5. Some of the notable significant differences (p 0.05) versus control 2 nm NGF in each cell type are summaried below. 500 nm NT-3 affords 100% efficiency in wt TrkA-pTyr (lane 3) whereas only 37% efficacy of ptyr is seen in the 4.1 TrkA/B chimera (lane 11). 2 nm NGF plus 2 nm NT-3 affords 126% efficacy in wt TrkA-pTyr (lane 7), more than 2 nm NT-3 alone (lane 2) and 2 nm NGF alone (lane 5). 2 nm NGF plus 500 nm NT-3 afford 54% efficacy in the 4.1 TrkA/B chimera-ptyr (lane 16). 2 nm NGF plus 500 nm NT-3 afford 50% efficacy of Akt activation in the 4.1 TrkA/B chimera (lane 25). 99

100 3.6 Discussion NT-3 and NGF signaling via TrkA control sympathetic neuronal development and differentially regulate survival and differentiation (Belliveau et al., 1997; Kuruvilla et al., 2004). Therefore, we aimed to determine TrkA subdomains involved in NGF and NT-3 binding and the functional outcome of NGF and NT-3 binding to their hot spots patches on TrkA receptors. NGF TrkA and NT-3 TrkC activation primarily occurs via the corresponding D5 subdomains of Trks (Perez et al., 1995). However, potentially allosteric regulation of binding and activity has been shown for Trks, particularly with respect to p75 NTR co-expression (Zaccaro et al., 2001). Also, p75 NTR on the cell surface affects the affinity of NT-3 for TrkA (Brennan et al., 1999). Because of the potentially confounding factor of p75 NTR co-expression, our studies focused on defining subdomains of TrkA which are relevant in binding and function towards cognate (NGF) or heterologous (NT-3) ligands in cells and systems that do not contain p75 NTR. We have used different binding methods and functional assays to compare NT-3 and NGF interactions with wt TrkA and the 4.1 Trk-A/B chimera. The K d s, EC50 s, and IC50 s from these methods are summarized in Tables 1 and 2. We determined that NT-3 TrkA binding and signaling involves several receptor subdomains, including TrkA-D5 (site 1) and TrkA- D4 (site 2). Furthermore, interaction with both sites is necessary for full NT-3 agonistic activity. Binding studies Biacore binding studies showed that the K d of TrkA-ECD for NGF (1.9 nm) differs from the 4.1 TrkA/B chimeric ECD (300 pm). The presence of high affinity sites was also evident in Scatchard analysis of 125 I[NGF] binding studies using HEK293 cells expressing 100

101 the 4.1 receptor chimera (K d ~80 pm). The high affinity binding (80 pm) and curvature apparent in these specific binding data are probably due to the dimerization equilibrium of the Trk receptor under conditions where the receptor is overexpressed at 50,000 receptors per cell. Higher affinity for the 4.1 TrkA/B chimera suggest that elements in the D1-D4 of TrkA-ECD may be inhibitory of NGF binding to TrkA. Such elements may be within the D1- D3 subdomains of TrkA, by logical exclusion, because the K d of TrkA-D4-D5 (Holden et al., 1997) and TrkA-ECD (Woo et al., 1998) are similar and independently reported to be ~4 nm. Previous literature and our data indicate that both NGF and NT-3 bind mainly to the TrkA- D5 subdomain (site 1). However, binding studies showed that NT-3 has relatively lower affinity than NGF for TrkA-ECD (131 nm versus 2 nm), and binding of NT-3 to 4.1 TrkA/B chimeric ECD was undetectable (with a limit of detection ~300 nm). Therefore, it seems that a second binding site on TrkA D1-D4 may exist for NT-3. NT-3 competition of NGF NGF30 binding to TrkA-ECD and TrkA-D4-D5 purified proteins indicate that potentially the second NT-3 binding site (site 2) may be allosteric on TrkA-D4. Consistent with this notion, competition of NGF NGF30 binding to TrkA wt and 4.1 TrkA/B expressed on cells further confirms a role of TrkA-D4 as a potentially allosteric NT-3 binding site. These data also show that TrkB-D4 (within the 4.1 chimeric protein) does not compensate for the contribution of TrkA-D4. In view of this, the D4 subdomains of TrkA and TrkB seem to be different at regulating NT-3 interactions. Indeed, TrkB-D4-D5 subdomains functionally interact with NT-3, because when these domains of TrkB are exchanged for TrkA NT-3 responses are reduced compared to wild type TrkB (Zaccaro et al., 2001). Based on these 101

102 results we can not exclude the possibility that NT-3 interacts with the D4 subdomain of TrkB. The IC 50 of NT-3 competition of 125 I[NGF] binding requires a 250-fold excess of NT- 3 to block binding to 4.1 TrkA/B receptors, while a 25-fold excess of NT-3 is required to block binding to wild type TrkA receptors (Ryden and Ibanez, 1996). These results further confirm the role of TrkA-D5 subdomain in NT-3 binding; and the 10-fold difference in NT-3 binding to TrkA wt and 4.1 TrkA/B chimera indicate that TrkB-D1-D4 subdomains do not compensate for the equivalent subdomains of TrkA. Taken together, our binding studies define two NT-3 binding sites on the TrkA receptor, one on TrkA-D5 subdomain (site 1) and one on TrkA-D4 subdomain (site 2). Are the NT-3 and NGF binding sites on site 1 (D5) the same epitope? Competition of 5C3 binding to TrkA wt receptor by NGF or NT-3 points to the distinct, but partially overlapping sites for NGF and NT-3 binding on TrkA-D5 subdomain. This notion is also supported by the poor competition of NT-3 of NGF NGF30 complexes binding to 4.1 TrkA/B receptors while in TrkA wt receptors competition is very efficient and almost stoichiometric. Different hot spots for these ligands on D5 might explain the functional outcomes of receptor engagement by each ligand (see next). TrkA NT-3 Functional Interactions To test if the NT-3 binding to TrkA receptor induces functional receptor activation, we investigated the survival responses. Both, TrkA wt and 4.1 TrkA/B cells induce survival dose-dependently in response to NGF. However, the survival response to optimal concentration of NT-3 differs, with TrkA wt cells inducing maximal survival in response to NT-3. Conversely, 4.1 TrkA/B cells induce only ~30% survival in response to optimal NT-3 102

103 treatment. These results indicate that NT-3 binding to TrkA-D5 subdomain (site 1) induces survival, but TrkA-D4 subdomain is necessary for full survival response (site 2). We also studied phosphorylation of the TrkA, or 4.1 TrkA/B receptors and the prosurvival PI-3 kinase/akt pathway upon Trk receptor activation (Vaillant et al., 1999) to correlate long-term functional survival assays with short-term biochemical signals. The biochemical data confirm that site 1 activation induces survival, acting via the PI-3 kinase/akt pathway. In TrkA wt expressing cells, mixing optimal concentrations of NGF with NT-3 leads to increased cell survival at all the NT-3 concentrations tested. In marked contrast, in 4.1 TrkA/B expressing cells, mixing optimal NGF with NT-3 leads to decreased survival. The long-term functional survival assays are supported by the short-term biochemical data. Putative conformational/allosteric regulation These results require an explanation that reconciles the following data: (a) NT-3 antagonism of NGF binding in all TrkA receptor forms; (b) NT-3 fully activates TrkA wt, but only affords partial activation of 4.1 TrkA/B; (c) NT-3 is functionally antagonistic for NGF in 4.1 TrkA/B, but enhances TrkA wt receptor function activated by NGF. How can NT-3 antagonize NGF binding to TrkA-D5, yet enhance NGF activity? One interpretation is that NT-3 activates and is additive with NGF through other TrkA hot spots besides D5. Possibly each of the ligands induces stepwise conformational changes, by binding to each receptor hot spot. NT-3 binding to TrkA-D4 could sterically inhibit NGF, or it could cause a receptor conformational change that leads to reduced binding by NGF to D5. Nevertheless, the NGF that does bind can fully activate the receptor. This interpretation further suggests that a conformational change induced by NT-3 binding to TrkA-D4 may have a positive influence in receptor activation. 103

104 In the view of a recent paper (Wehrman et al., 2007), the crystal structure of the entire TrkA-ECD in complex with NGF confirms the primary importance of the D5 subdomain, but also indicates lack of flexibility in the TrkA-D1-D4 subdomains. However, the crystal structure of TrkA-ECD alone was not determined; therefore conformational changes of receptor upon ligand binding cannot be ruled out. The interaction between NT-3 and TrkA- ECD might be different and affect differently the receptor conformation. Moreover, it is interesting that the K d for NGF binding to 4.1 chimera is 6-fold stronger than for wt TrkA- ECD, suggesting some inhibitory influence of sudomains D1-D4 upon D5 under some circumstances. This observation may support an allosteric hypothesis. A crystal structure of NT-3-TrkA-ECD would be of considerable interest. This mechanism would be consistent with other receptor families that have allosteric receptor binding sites, including hgh (Cunningham et al., 1991), erythropoietin receptor (Remy et al., 1999), and epidermal growth factor receptor (Garrett et al., 2002). However, a comparison of TrkA-D5 structure (Ultsch et al., 1999), deduced from a ß-strand swapped D5 dimer, versus a complex of NGF-TrkA-D5 (Wiesmann et al., 1999) failed to demonstrate conformational changes in either the ligand or the receptor, other than than a coiling of a short α-helix in the N-terminus of NGF. This rigidity seems quite unusual because generally ligand conformation is affected upon binding, and NGF undergoes conformational changes upon binding to p75-ecd (He and Garcia, 2004). Along these lines, circular dichroism (CD) spectra of TrkA-ECD in complex with NGF showed small but significant conformational changes in secondary structure (Woo et al., 1998). 104

105 Conclusions Taken together, our findings indicate that both NGF and NT-3 binding and activation sites are partially overlapping on the TrkA-D5 domain (site 1), NT-3 binding it with lower affinity. Furthermore, there is an NT-3 binding site on the TrkA-D4 domain (site 2). NT-3 activation of site 1 induces trophic signals, but activation of other NT-3 site (site 2) is necessary for full activation. Potentially these NT-3 hot spots could regulate function through an allosteric mechanism. NGF and NT-3 signaling via TrkA are important mediators of neuronal development and function, and abnormal signaling via neurotrophin receptors is involved in different pathological states. Identifying the NT-3 binding hot spots on TrkA receptor will enable synthesis of selective small molecules targeting the receptor binding or regulatory sites (Saragovi and Zaccaro, 2002), providing further insight into complex neurotrophin receptor signaling. 105

106 Chapter 4 An agonistic anti-trkc mab directed to the juxtamembrane ectodomain defines a functional hot spot interacting with p75 NTR co-receptors 106

107 4.1 Rational In the previous chapters, TrkA and TrkB subdomains important for ligand binding and functional receptor activation by homologous and heterologous ligands were determined. In addition, the role of p75 NTR -co receptors in ligand-binding subdomain utilization was defined. Next, to address the issues of TrkC receptor binding sites and functional activation sites and their regulation by p75 NTR, we raised mab against the sequence outside D5, within the juxtamembrane-linker domain of TrkC termed mab 2B7. This mab is a selective TrkC ligand that can be used to selectively target TrkC receptor without p75 NTR binding and activation. Previously we have developed an agonistic antibody of TrkA receptor termed 5C3 (LeSauteur et al., 1996). Unlike natural ligand, they dock into the small epitope rather than an extended surface. Moreover, monomeric Fabs also induced receptor activation. First we characterized the anti-trkc mab as a selective TrkC agonist; it induces survival of TrkC expressing cells and receptor activation and downstream signaling pathway activation. In addition we asked if monomeric 2B7 Fabs induce TrkC mediated survival. Since it was shown that p75 NTR ligands regulate TrkC ligand-dependent differentiation but not survival (Ivanisevic et al., 2003) we wanted to see if p75 NTR co-expression affects mab 2B7 binding and functional receptor activation. We determined that p75 NTR co-expression reduces mab2b7 binding to TrkC receptor and ligand-dependent trophic survival. Therefore, not only does MAb 2B7 define a previously unknown neurotrophin receptor functional hot spot that is potentially regulated by p75 NTR expression but its activation can also potentialy uncouple survival and differentiation signals. 107

108 4.2 Abstract The extracellular receptor domain D5 of TrkC is a docking site for Neurotrophin-3 (NT-3), but other domains may be relevant for function or for harmonizing signals with p75 NTR co-receptors. Here, we report the development and characterization of mouse monoclonal antibody (mab) 2B7 targeting an epitope outside D5, within the juxtamembranelinker domain of TrkC. MAb 2B7 binds to murine and human TrkC receptors. MAb 2B7 is a functional agonist that affords activation of TrkC, AKT and MAPK. These signals result in cell survival but not in cellular differentiation. Interestingly, monomeric 2B7 Fabs also afford TrkC-mediated cell survival. NT-3 competes binding of 2B7 mab and 2B7 Fabs to TrkC, in a dose-dependent manner. However, while expression of cell surface p75 NTR augments NT-3 binding and function, the binding and function of mab 2B7 are diminished. MAb 2B7 defines a previously unknown neurotrophin receptor functional hot spot; that exclusively generates survival signals; that can be activated by non-dimeric ligands; and that potentially unmasks a site for p75 NTR -TrkC interactions. 4.3 Introduction Neurotrophin-3 (NT-3), brain-derived growth factor (BDNF) and nerve growth factor (NGF) are essential growth factors for the development and maintenance of the nervous system (Arevalo and Wu, 2006). The neurotrophins are stable homodimers (Butte et al., 1998) that bind to either or both two types of cell surface receptors termed p75 NTR and Trk. Each neurotrophin binds a selective Trk receptor with relatively high affinity (e.g. TrkA NGF and TrkC NT-3; K d M). The Trk receptor has tyrosine kinase catalytic activity that is associated with the survival and differentiation neurotrophic signals. 108

109 Neurotrophin-induced Trk activity affords trophic (growth/survival) responses via MAPK and Akt, whereas PLC-γ and FRS-2 activity are obligatory for differentiation (Kaplan and Stephens, 1994; Meakin et al., 1999). A Trk receptor ectodomain termed D5 comprises the main neurotrophin binding site (Urfer et al., 1997; Wiesmann and de Vos, 2001; Zaccaro et al., 2001) and it is required for ligand-dependent receptor activation. Such receptor sites that define ligand-binding and functional-activation are termed hot spots (Saragovi and Zaccaro, 2002). Previously, we demonstrated that artificial ligands (e.g. an antibody) binding to a receptor hot spot could be functionally active; (Saragovi et al., 1999; Saragovi and Gehring, 2000; Saragovi and Zaccaro, 2002). Specifically, we reported an agonistic mab 5C3 directed to the TrkA D5 domain (LeSauteur et al., 1996). All mature neurotrophins also bind to p75 NTR (Nykjaer et al., 2005). The p75 NTR can act either as pro-apoptotic, or pro-survival depending on the ligand type, cell type or developmental stage of the cells (Casaccia-Bonnefil et al., 1996; Bamji et al., 1998; Rabizadeh et al., 1999). There is a reciprocal interplay between the Trk and p75 NTR receptors that regulates signal cascades (Maliartchouk and Saragovi, 1997; Hapner et al., 1998; Maliartchouk et al., 2000a; Ivanisevic et al., 2003) and ligand binding (Mahadeo et al., 1994). Indeed, the anti-trka agonistic mab 5C3 was used as a selective ligand of TrkA to demonstrate a functional relationship where unbound p75 NTR expression suppresses liganddependent TrkA function (Maliartchouk and Saragovi, 1997); and that p75 NTR ligands can decrease (Maliartchouk and Saragovi, 1997; Maliartchouk et al., 2000a), or increase (MacPhee and Barker, 1997) p75 NTR suppression of TrkA activation. 109

110 However, the mechanism by which each receptor affects the other s binding or function is still not fully understood. For example, besides the D5 domain other domains of Trk are involved in activation (e.g. the D3 domain), and other Trk domains bind ligand and transduce signals in a p75 NTR -dependent manner (e.g. the D1 domain) (Zaccaro et al., 2001). Also, physical Trk-p75 NTR interactions were reported in cells over-expressing these receptors (Bibel et al., 1999). Here, we report the development of agonistic anti-trkc mab as tool to explore whether novel hot spots could be defined outside the D5 domain of TrkC, to study whether and how p75 NTR regulates TrkC function, and to evaluate the bioactivity of monovalent TrkC ligands. 4.4 Materials and methods Cell lines: Mouse SP2/0 myelomas; nnr5 pheochromocytoma cells (derived from PC12), nnrr5 cells stably transfected with human trkc cdna (nnr5-trkc), wild type NIH- 3T3 fibroblasts, and NIH-3T3 stably transfected with trkc cdna (NIH-TrkC cells) or human trka cdna (NIH-TrkA cells) were used. All cells were cultured in RPMI media supplemented with 5% fetal bovine serum (FBS) and antibiotics (Gibco). Stable transfectants were supplemented with 0.4 mg/ml neomycin, and protein expression was routinely verified. Co-expression of full length rat p75 receptors. NIH-TrkC cells were stably transfected with full length p75 NTR receptor with pcdna3.1/zeo(+) rat p75 NTR construct. Stable transfectants were selected by treatment with Zeocin (200μg/ml). Antibodies. Rat anti-mouse IgG (αmigg; Sigma, St. Louis, MO), antiphosphotyrosine mab 4G10 (Upstate Biotechnology, Lake Placid, NY), anti-phospho-akt (ser473) antibody (Cell Signaling), anti-phospho-mapk (p42/44, thr202/tyr204) antibody 110

111 (Cell Signaling), and fluoresceinated [fluorescin isothiocyanate (FITC)] goat anti-mouse IgG (FITC-GαmIgG) (Sigma, St. Louis, MO) and goat anti mouse Fab (GαmFab) antibodies were purchased commercially from Sigma. MAb 5C3 was developed and grown in our laboratory (LeSauteur et al., 1996), and is an agonistic anti-trka mab directed to the TrkA- D5 domain. MAb 2B7 generation and purification: Female Balb/c mice (8 weeks old) were immunized three times with a human TrkC peptide (ESTDNFILFDEVSPTPPI) selected from a predicted TrkC receptor hot spot. The 18 amino acid ectodomain sequence is located at the linker region and ends 10 residues before the predicted transmembrane domain. The peptide was conjugated to KLH as carrier. All animal protocols were approved by McGill Animal Care Committee. Splenocytes were fused to SP2/0 myelomas, and hybridomas were screened by differential binding in an Enzyme-Linked Immunosorbent Assay (ELISA) using the original peptide immunogen conjugated to BSA. Specific binding data to native cell surface receptors were obtained using a Fluorescent Activated Cell Scanner (FACScan) (Becton Dickinson, San Jose, CA) (see below). MAb 2B7 [IgG1(κ)] was identified by IsoStrip (Roche) and subcloned three times. MAb 2B7 was purified onto a Protein G- Sepharose column (Sigma). The binding and biochemical properties of purified mab 2B7 were characterized by ELISA, flow cytometry, and SDS-PAGE. Monomeric mab 2B7 Fabs: MAb 2B7 was purified (8 mg/ml) as above and digested with 0.02 mg/ml papain (Gibco, Toronto, Ontario, Canada) for 6 hours (LeSauteur et al., 1996). Fabs were re-purified on Protein A-Sepharose and dialyzed against PBS. Products were characterized by SDS-PAGE under non-reducing conditions. 111

112 Flow cytometry: Cells ( ) in 0.1 ml of binding buffer [Hanks Balanced Salt Solution (HBSS), 0.1% bovine serum albumin (BSA), and 0.1% NaN 3 ] were incubated with the indicated concentration of mabs or Fabs for 20 min at 4 C, washed in binding buffer to remove excess primary antibody, and immunostained with FITC-GαmIgG secondary antibody for 20 min at 4 C. Cells were acquired and analyzed on a FACScan BD Sciences using the Cell Quest program. As negative controls no primary (background fluorescence), or irrelevant mouse IgG (Sigma) were used followed by secondary antibody. Specificity was gauged using various cells expressing different receptors. Western blot analysis: Assays were performed as previously described (Maliartchouk and Saragovi, 1997). The activation of each protein (Trk, Akt and MAPK) was studied after treatment of live cells with different concentrations of ligands mab 2B7, mab 2B7 Fabs, NT-3 for 12 minutes at 37 C, cells were solubilized and protein concentrations were determined with Bio-Rad Detergent Compatible Protein Assay (Bio-Rad Laboratories). Western blot analysis was performed with the indicated reagents. Blots were visualized using the enhanced chemiluminiscence system (PerkinElmer Life Sciences). Re-blotting the membranes with anti-trk 203 serum (gift from Dr. Kaplan) or anti-actin antibody (Sigma) confirmed equal protein loading. Binding Competition assays: MAb 2B7 and mab 5C3 were labeled with biotin (Pierce). Competition of mab 2B7 binding to NIH-TrkC or NIH-TrkC+p75 NTR cells was tested with NT-3 or NGF as irrelevant control. The assays were performed by first incubating cells for 20 minutes at 4ºC with various concentrations of the test competitor followed by saturating (67 nm) of mab 2B7-biotin, mab 5C3-biotin as irrelevant primary, or negative control mouse IgG for another 20 minutes at 4 C. Then, FITC-goatαmIgG or FITC-avidin 112

113 was added as secondary reagent. After washing, cells were analyzed by flow cytometry as previously described. The conditions used (4 C and Na azide in the buffer) did not allow internalization. Survival assays: NIH-TrkC or NIH-TrkC+p75 NTR cells (7,500 cells/well) in serumfree media (PFHM-II; Gibco) supplemented with 0.2% BSA were added to 96-well plates (Falcon, Lincoln Park, NJ) containing NT-3, mab 2B7, mab 2B7 Fabs, negative control mouse IgG, or serum (final 5% FBS, normal culture conditions). Where indicated, mab 2B7 Fabs were cross-linked with goat anti-mouse Fab (Gαm Fab). Wells containing all culture conditions but no cells were used as blanks. The growth/survival profile of the cells was quantified using the tetrazolium salt reagent 4-[4,5-dimethylthiazol-2-yl]-2,5- diphenyltetrazolium bromide (MTT; Sigma) hours after plating. Optical density readings of MTT were done in a Benchmark Plus microplate Spectrophotometer (BioRad) at 595 nm with blanks subtracted. Differentiation assays. nnr5 cells stably transfected with trkc cdna (nnr5-trkc) were plated on cover-slips with full media in 24-well plates. 24 hours after plating the indicated treatments were added for an additional hours of culture. Cellular differentiation was gauged by immunocytochemistry after cover-slips were fixed and stained with MAP-2 antibody (Chemicon) followed by goat anti-rabbit Cy3 (Jackson Immunochemicals) and analyzed as described earlier (Ivanisevic et al., 2003). Quantification and statistical analyses. Quantification of Western blots was done by densitometric analysis relative to total protein levels. Quantification data are presented as percent relative to optimal (10 nm NT-3) as 100%. Statistical analysis were performed by two-tailed t-tests; statistical significance (p 0.05) is indicated by an asterisk (*). 113

114 4.5 Results Generation and initial screening of mab 2B7 binding Linear peptide NH 2 -ESTDNFILFDEVSPTPPI-COOH was conjugated through the N- termini to KLH, and was used to immunize mice. After fusion of splenocytes with SP2 myeloma cells, culture supernatant from hybridomas were screened by ELISA using either the immunizing peptide conjugated through the N-termini to BSA, or the immunizing free peptide immobilized on the ELISA plate. Several independent wells with hybridoma cells producing antibodies with selectivity to the immunizing peptide were identified, and they were subcloned three times by limiting dilution. MAb 2B7 was chosen for further work. Characterization of mab 2B7 binding To assess mab 2B7 specificity for cell surface TrkC, cells expressing or lacking TrkC were screened by flow cytometry for differential binding. MAb 2B7 did not immunostain wild type NIH-3T3 cells (Figure 1A), nnr5 wt cells (Figure 1E) or NIH-TrkA transfectants (Figure 1B) above background negative control (migg), but did immunostain NIH-TrkC transfectants (Figure 1C). Therefore, mab 2B7 binds selectively to TrkC, and it does not bind to p75 NTR or TrkA. Binding of mab 2B7 to non-permeabilized cells indicates that it recognizes the extracellular domain of TrkC. The specific epitope on TrkC is located between the transmembrane and the D5 domain. The concentration of mab 2B7 and mab 2B7 Fab required to saturate TrkC was determined by flow cytometry assays (Figure 1I). Receptor saturation is seen at ~65 nm mab 114

115 2B7. Similar analysis with mab 2B7 Fab demonstrated that specificity and saturability were similar (~75 nm) to that obtained with intact mab. The slight difference in fluorescent intensity at saturation with intact mab versus Fabs is due to the use of different fluorescinated secondary reagents. Figure 1 mab 2B7 binds selectively to TrkC receptor expressing cells. (A-F) FACScan binding assays with mab 2B7. The indicated cells were incubated with saturating concentration of mab 2B7 (~65 nm). (G) mab 2B7 competes NT-3 binding to NIH-TrkC and (H) NIH-TrkC+p75 NTR cells in a dose-dependent manner. (I) Flow cytometry saturability profile of mab 2B7 and 2B7 Fabs in NIH TrkC cells. (J) 2B7 115

116 recognizes TrkC by Western blot in non-reducing conditions. (K) Total levels of TrkC receptor are similar in NIH-TrkC and NIH-TrkC+p75 NTR cells. NIH-TrkC+p75 NTR cells express high levels of p75 NTR. Data are representative of three independent assays. Western blot analysis with mab 2B7 revealed a band at Mr 145 kda (p145) for lysates from NIH-TrkC, but no bands were seen for lysates of control wild type NIH-3T3 cells (Figure 1J). MAb 2B7 was effective in western blot analysis only when samples were prepared under non-reducing conditions. This finding is puzzling because the epitope of mab 2B7 was raised by immunization with a linear peptide conjugated to KLH, and was screened by ELISA using the same linear peptide and cells expressing TrkC. Thus, we suspect that mab 2B7 recognizes a conformation stabilized or influenced by a disulfide bond, similar to anti-human TrkA mab 5C3 raised in our laboratory (LeSauteur et al., 1996). Ligand Competition Studies. Flow cytometry analysis demonstrated that NT-3 competitively blocked mab 2B7 binding to cell surface human TrkC receptors, in a dose-dependent manner. 1 nm NT-3 (considered to be saturating) blocked 40-50% of the mab 2B7 binding sites in NIH-TrkC (Figure 1G) and NIH-TrkC+p75 cells (Figure 1H). See Table 1 for a summary. Control competition studies showed that NGF or anti-trka mab 5C3 do not block mab 2B7 binding to TrkC (data not shown). Table 1 NT-3 competes 2B7 binding. Summary of Flow cytometry data from Figure 1G and 1H. Mean channel fluorescence (MCF) and % inhibition of mab 2B7 binding by NT-3 are presented. Data is representative of 3 independent experiments. 116

117 Fluorescent Ligand nm NT-3 Competitor Mean Fluorescence % competition TrkC TrkC + p75 TrkC TrkC + p75 mouse IgG B B B Effect of p75 NTR co-expresion on 2B7 mab binding to TrkC. Quantitative flow cytometry assays showed that stable co-expression of full length p75 NTR receptors in NIH-TrkC cells reduced the detectable surface 2B7 binding by ~50-60% in all twelve clones that were independently isolated (Figure 1D, Table 1), even when oversaturating concentrations of mab 2B7 were used. Quantitative Western blot analyses on three NIH-TrkC+p75 NTR clones demonstrated comparable levels of total TrkC as in parental NIH-TrkC cells (Figure 1K). These data suggest that p75 NTR may induce conformational changes or steric hindrance at or near the mab 2B7 epitope on TrkC, and suggest p75 NTR TrkC interactions. Functional agonism of mab 2B7 and monomeric mab 2B7 Fabs. Biochemical assays, survival assays and differentiation assays were undertaken to determine if mab 2B7 has NT-3-like agonistic activity. Phosphorylation of TrkC, AKT and MAPK were tested by studying lysates from cells exposed for 12 to various ligands or controls (Figure 2). NT-3, mab 2B7 and 2B7 Fabs (with or without cross-linking using GαmFab antibodies) afford significant phosphotyrosinylation of TrkC (p-trkc) over basal levels. However, 2B7 Fabs do so with lower efficacy (Figure 2A). Similarly, NT-3, mab 2B7 and 2B7 Fabs phosphorylated and activated downstream signaling proteins AKT and 117

118 MAPK in NIH-TrkC cells (Figure 2A). In cellular controls, no increase in p-trka, p-akt, or p-mapk were observed when NIH-TrkA cells were treated with 2B7 mab or Fabs (data not shown). The quantification of phosphorylated proteins relative to total Trk protein loaded (Figure 2A) are presented as % of optimal NT-3, arbitrarily set at 100% (Figure 2B). Suboptimal 0.1 nm NT-3, 10 nm 2B7 and 100 nm 2B7 induce p-trkc with ~30% of maximal efficacy. In addition, 2B7 Fabs and 2B7 Fabs cross-linked with GαmFabs induced p-trkc with efficacy ~ 15%. Suboptimal 0.1 nm NT-3 and 10 nm 2B7 induce p-akt and p-mapk with ~60% of maximal efficacy, while 2B7 Fabs and 2B7 Fabs cross-linked with GαmFabs induce their activation with efficacy of ~ 30%. 118

119 Figure 2 TrkC, Akt and MAPK activation are induced by mab 2B7 or mab 2B7 Fabs. NIH TrkC cells were treated with the indicated ligands for 12 and cell lysates were analyzed by Western blot (A) revealed by anti-p-tyr, anti-p-akt, anti-p-mapk and 203 sera (total Trk). Representative experiment is shown. (B) Summary of data quantified by densitometry from three independent experiments and standardized to total Trk and presented as % relative to 10 nm NT-3. Then, agonists were tested for their ability to protect cells from death induced by culture in serum-free media (SFM) (Table 2). NT-3, mab 2B7, and monovalent 2B7 Fabs significantly protected NIH-TrkC cells compared to migg (negative control) treated cells, in a dose-dependent manner. MAb 2B7 (Table 2 row 1) achieved a maximal efficacy of ~50% survival compared versus optimal NT-3 (Table 2 row 3). Monomeric 2B7 Fabs achieved a maximal efficacy of ~35% survival compared versus optimal NT-3 (Table 2 row 2), even without cross-linking. 119

120 Treatments % NIH TrkC survival ± sem 1 2B7 (100 nm) 53 ± 5 * 2 2B7 Fabs (10 nm) 36 ± 7 * 3 NT-3 (2 nm) 100 ± 7 * 4 migg (100 nm) 7 ± 5 Table 2 mab 2B7 and mab 2B7 Fabs afford survival of TrkC-expressing cells. Cells were cultured in SFM supplemented with the indicated treatments or serum for 48 h. Cell survival was measured by MTT assays. Results shown are representative of 3 independent experiments, n=4 in each assay. Statistically significant cell survival compared to migg negative control (p<0.05) is indicated by an asterisk (*). Control ligands migg, mab 5C3, and NGF did not protect NIH-TrkC cells; but in cellular controls NGF protected NIH-TrkA cells as expected; and neither NT-3, nor mab 2B7, nor 2B7 Fabs protected NIH-TrkA cells or wild-type NIH 3T3 cells (data not shown). Effect of p75 NTR on 2B7 agonistic activity NIH-TrkC or NIH-TrkC+p75 NTR cells were tested by MTT survival assays to test the effect of p75 NTR expression on 2B7 agonistic activity (Figure 3). In order to compare these two cell lines, the data was standardized to normal serum growth conditions (arbitrarily set as 100%). Compared versus serum NIH-TrkC and NIH-TrkC+p75 NTR cells were both protected by optimal NT-3 (10 nm and 1 nm) to a comparable ~ 65-70%. Co-expression of full-length p75 NTR in NIH-TrkC cells significantly enhanced the survival-promoting efficacy of suboptimal 0.1 nm NT-3 (35% survival in NIH-TrkC+p75 NTR cells versus 20% in NIH- TrkC). In contrast, p75 NTR co-expression significantly reduced the trophic effect of mab 2B7 (20% survival in NIH-TrkC+p75 NTR cells versus 40% in NIH-TrkC). Thus, co-expression of 120

121 p75 NTR improves the binding and function of NT-3 but reduces the binding and function of mab 2B7. Figure 3 mab2b7 induced trophic protection is lower in NIH-TrkC+p75 NTR cells. The survival of NIH-TrkC or NIH-TrkC +p75 NTR cells was tested in MTT assays after culture in SFM supplemented with the indicated ligands or controls (5% serum = 100%, untreated = 0%), n=4 each assay. Data are representative from 3 independent experiments. * indicates statistical significance p<0.05. ** indicates that the difference is not statistically significant. We have previously shown that p75 NTR negatively regulates the efficacy of selective TrkA agonists such as mab 5C3 (LeSauteur et al., 1996). In this scenario, certain p75 NTR - selective ligands neutralize the negative regulation resulting in full TrkA activation (Maliartchouk and Saragovi, 1997). Thus, we performed similar survival assays with mab 2B7 ± selective engagement of p75 NTR with NGF, BDNF, or anti-p75 NTR mab MC192. These p75 NTR -selective ligands did not enhance (or reduce) the survival promoting signals of mab 2B7 (data not shown). Thus, the negative regulation of p75 NTR upon mab 2B7 trophic function is not affected by p75 NTR -selective ligands. These data confirm a previous report that p75 NTR TrkC functional interactions differ from p75 NTR TrkA functional interactions (Ivanisevic et al., 2003). 121

122 Effect of mab 2B7 on cell differentiation nnr5-trkc cells differentiate and grow neurites in response to NT-3. We tested if mab 2B7 as a TrkC agonist induces neurite outgrowth (Figure 4A, data summarized in Figure 4B). Treatment with control migg or 10 nm NGF results in cells that are round and undifferentiated. Treatment with NT-3 induced differentiation in a dose-dependent manner, increasing the percent of cells bearing >2 axons and of axonal length >2 cell bodies. Treatment with mab 2B7 does not induce differentiation, whether or not mab 2B7 it is combined with p75 NTR -selective ligands NGF (Figure 4) or BDNF (data not shown). In controls, a suboptimal concentration of NT-3 in combination with NGF (as p75 NTR ligand) increases cellular differentiation to levels comparable to optimal NT-3, as reported previously (Ivanisevic et al., 2003). Thus, mab 2B7 does not have intrinsic neuritogenic activity, and the use of p75 NTR ligands does not potentiate neuritogenesis either. 122

123 Figure 4 mab 2B7 does not induce differentiation of nnr5 TrkC cells. (A) Representative experiment of the differentiation of nnr5-trkc cells in response to treatment with the indicated ligand for hours. (B) Quantitative summary (±SD) of 3 independent experiments. Cells were plated with indicated treatments or controls, and differentiation was scored as % of cells with neurites (> 2 cell body long). * Significantly different relative to 100 nm migg, p