The Open Collections website will be unavailable July 27 from 2100-2200 PST ahead of planned usability and performance enhancements on July 28. More information here.

Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Olfactory ensheathing cell mediated mechanisms of neurite outgrowth and axon regeneration Witheford Richter, Miranda 2008

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


24-ubc_2008_fall_withefordrichter_miranda.pdf.pdf [ 15.62MB ]
JSON: 24-1.0066452.json
JSON-LD: 24-1.0066452-ld.json
RDF/XML (Pretty): 24-1.0066452-rdf.xml
RDF/JSON: 24-1.0066452-rdf.json
Turtle: 24-1.0066452-turtle.txt
N-Triples: 24-1.0066452-rdf-ntriples.txt
Original Record: 24-1.0066452-source.json
Full Text

Full Text

OLFACTORY ENSHEATHING CELL MEDIATED MECHANISMS OF NEURITE OUTGROWTH AND AXON REGENERATION  by  MIRANDA WITHEFORD RICHTER B.Sc., The University of British Columbia  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Neuroscience)   THE UNIVERSITY OF BRITISH COLUMBIA VANCOUVER JUNE 2008  © Miranda Witheford Richter, 2008 ii   ABSTRACT  The capacity of the olfactory neuraxis to undergo neuronal replacement and axon targeting following injury, has led to scrutiny concerning the molecular and physical determinants of this growth capacity.  This is because injury to the central nervous system, in contrast, leads to permanent disconnection of neurons with targets.  Olfactory ensheathing cells (OECs), a specialized glial cell, may contribute to olfactory repair, and have been used to promote recovery from spinal cord injury.  However, there mechanisms underlying OEC- induced regeneration are poorly appreciated.  To understand these mechanisms, OECs from the lamina propria (LP OECs) or olfactory bulb (OB OECs) were transplanted into a lesion of the dorsolateral funiculus.  While both cells demonstrated reparative capacities, LP and OB OECs differentially promoted spinal fibre growth; large-diameter neurofilament-positive, CGRP-positive, and serotonergic fibres sprouted in response to both LP and OB OEC transplantation, whereas substance-P and tyrosine hydroxylase-positive neurons grew more extensively following OB or LP OEC transplantation, respectively.  To further understand the growth of spinal cord neurons in response to OECs, a proteomic analysis of OEC secreted factors was performed, identifying secreted protein acidic and rich in cysteines (SPARC) as a mediator of OEC-induced outgrowth in vitro.  To test the contributions of SPARC to spinal cord repair after OEC transplantation, cultures of LP OECs from SPARC null and wildtype (WT) mice were transplanted into a crush of the dorsolateral funiculus.  Substance P and tyrosine hydroxylase positive axon sprouting was significantly reduced in SPARC null OEC-treated animals, suggesting that individual factors may contribute to OEC-promoted regeneration.  To investigate the effect of OECs on corticospinal (CST) neurons, an in vitro assay was developed using postnatal day 8 CST neurons.  Coculture of CST neurons with OB OECs produced extensive axon elongation.  Application of OB OEC secreted factors increased CST neurite branching, but did not increase axon elongation.  In contrast, plating of CST neurons on OB OEC plasma membrane resulted in extensive axon elongation.  Furthermore, the OB OEC plasma membrane could overcome CST neurite outgrowth inhibition induced by an outgrowth iii   inhibitor.  Together these findings provide insight into OEC mechanisms of neurite outgrowth and axon regeneration. iv   TABLE OF CONTENTS  ABSTRACT ................................................................................................................................... ii TABLE OF CONTENTS ............................................................................................................ iv LIST OF FIGURES ..................................................................................................................... xi SYMBOLS, ABBREVIATIONS, AND NOMENCLATURE ................................................ xiv ACKNOWLEDGMENTS ........................................................................................................ xvii DEDICATION.......................................................................................................................... xviii CHAPTER 1: LITERATURE REVIEW AND INTRODUCTION ......................................... 1 1.1 Introduction ...................................................................................................................... 1 1.1.1 Regeneration in the peripheral and central nervous systems .......................................... 1 1.1.2 What are olfactory ensheathing cells (OECs)? ............................................................... 2 1.1.3 Molecular basis of axon-OEC interactions ..................................................................... 6 1.1.4 OEC response to olfactory injury ................................................................................. 10 1.1.5 Olfactory ensheathing cells in vitro .............................................................................. 12 1.1.6 How do glia influence neurite outgrowth? ................................................................... 18 1.1.7 What is the relationship between OECs and other glia of the central and peripheral nervous systems? ................................................................................................................... 24 1.1.8 What is the rationale for OECs as a therapeutic for spinal cord injury? ...................... 27 1.1.9 What anatomical and behavioural outcomes have been achieved following OEC transplantation? ...................................................................................................................... 28 1.1.10 What mechanisms account for the ability of OECs to promote neuron growth after spinal cord injury? ................................................................................................................. 32 1.1.11 Aims and hypotheses .................................................................................................. 32 CHAPTER 2: MATERIALS AND METHODS ...................................................................... 40 2.1 Animals ............................................................................................................................... 40 2.1.1 Mice used for transplantation: β-actin GFP, SPARC null, C57/Bl6 ............................ 40 2.1.2 Mice used for cell culture: CD-1, Thy1-YFP16JRS ..................................................... 40 2.1.3 Rats used for spinal cord injury models ....................................................................... 41 2.2 PCR for Thy1-YFP16JRS genotyping ................................................................................ 41 v   2.3 Cell Culture ......................................................................................................................... 42 2.3.1 LP OEC cell culture ...................................................................................................... 42 2.3.2 LP fibroblast cell culture .............................................................................................. 42 2.3.3 OB OEC cell culture ..................................................................................................... 43 2.3.4 Astrocyte cell culture .................................................................................................... 44 2.3.5 Schwann cell culture ..................................................................................................... 45 2.3.6 Corticospinal neuron cell culture .................................................................................. 45 2.3.7 Chinese hamster ovary-R2 (CHO-R2) and Chinese hamster ovary-myelin associated glycoprotein (CHO-MAG) cell culture ................................................................................. 46 2.4 Immunocytochemistry and immunohistochemistry ............................................................ 46 2.4.1 Immunocytochemistry .................................................................................................. 46 2.4.2 Immunohistochemistry ................................................................................................. 48 2.5 Spinal cord injury and cell transplantation .......................................................................... 49 2.5.2 Cell harvest for transplantation..................................................................................... 49 2.5.3 Spinal cord injury and transplantation .......................................................................... 49 2.5.4 Immunosuppression ...................................................................................................... 50 2.6 In vitro assays and manipulations ....................................................................................... 51 2.6.1 Cell migration assays .................................................................................................... 51 2.6.2 Transfection of SPARC null and WT LP OECs........................................................... 51 2.6.3 Neurotrophin addition to corticospinal neuron culture ................................................. 52 2.6.4 Mitotic inhibitors .......................................................................................................... 52 2.6.6 Generation, harvesting and addition of OB OEC conditioned media .......................... 53 2.6.7 Transwell coating and cell plating ................................................................................ 54 2.6.8 Plasma membrane enrichment ...................................................................................... 54 2.6.9 Plasma membrane optical density assessment and substrate plating ........................... 55 2.7 SDS-PAGE and Western blotting ....................................................................................... 55 2.7.1 Sample quantification and preparation ......................................................................... 55 2.7.2 SDS-PAGE and Western blotting ................................................................................ 56 2.8 Image analysis, quantification and statististics ................................................................... 57 2.8.1 Spinal cord injury and transplantation experiments ..................................................... 57 2.8.2 Corticospinal outgrowth assay...................................................................................... 59 vi   2.8.3 Statistical analysis......................................................................................................... 60 CHAPTER 3: LAMINA PROPRIA AND OLFACTORY BULB ENSHEATHING CELLS EXHIBIT DIFFERENTIAL INTEGRATION, MIGRATION, AND PROMOTE DIFFERENTIAL AXON SPROUTING IN THE LESIONED SPINAL CORD. ................. 63 3.1 Introduction ......................................................................................................................... 63 3.1.1 LP and OB OECs perform divergent roles in their differing environments within the olfactory neuraxis .................................................................................................................. 65 3.1.2 Rationale and aims: regeneration outcomes may differ following LP or OB OEC transplantation ....................................................................................................................... 66 3.2 Results ................................................................................................................................. 67 3.2.1 LP and OB OECs in vitro ............................................................................................. 67 3.2.2 LP and OB OEC migration in vitro and within the spinal cord ................................... 67 3.2.3 Lesion site area and cavity formation are decreased by LP versus OB OEC transplantation. ...................................................................................................................... 71 3.2.4 Reactive astrogliosis is reduced by rostrocaudal transplantation. ................................ 73 3.2.5 LP and OB OECs similarly recruit endogenous Schwann cells to the lesion site ........ 74 3.2.6 Axonal sprouting into the lesion site is increased in rats transplanted rostrocaudally with LP OECs animals. ......................................................................................................... 74 3.2.7 Directional angiogenesis is enhanced by OEC treatment ............................................. 76 3.2.8 Autotomy is reduced by OB OEC, but elevated by LP OEC transplantation. ............. 78 3.2.9 Sprouting of tyrosine-hydroxylase, Substance P and calcitonin gene-related peptide positive axons is differentially stimulated by LP or OB OEC transplantation. ..................... 81 3.3 Discussion ........................................................................................................................... 86 3.3.1 Rationale and summary ................................................................................................ 86 3.3.2 LP and OB OECs display instrinsic biological differences in expansion, migration, and interactions with astrocytes following transplantation into SCI ............................................ 86 3.3.3 LP versus OB OEC transplantation promotes the growth of differing subpopulations of spinal axons ........................................................................................................................... 88 3.3.4 Summary of findings and future directions .................................................................. 90 CHAPTER 4: SPARC FROM TRANSPLANTED OLFACTORY ENSHEATHING CELLS DIFFERENTIALLY MEDIATES THE GROWTH OF SPINAL NEURON POPULATIONS IN THE INJURED SPINAL CORD ........................................................... 91 4.1 Introduction ......................................................................................................................... 91 4.1.1 How do OECs affect neurite outgrowth? ..................................................................... 91 vii   4.1.2 SPARC is secreted by OECs and is partially responsible for their outgrowth-promoting activity. .................................................................................................................................. 92 4.1.3 Expression and function of SPARC in the nervous system and in wound healing ...... 94 4.1.4 Rationale and aims........................................................................................................ 94 4.2 Results ................................................................................................................................. 95 4.2.1 Culture, preparation, and transplantation of WT and SPARC null LP OECs. ............. 95 4.2.2 SPARC is expressed by WT transplanted OECs, endogenous astrocytes, and on blood vessel laminae in the injured spinal cord. .............................................................................. 96 4.2.3 SPARC null and WT OECs equally decrease lesion site area and cavity formation, but differentially affect macrophage recruitment to the lesion site. ............................................ 98 4.2.4 SPARC null and WT OECs equally promote directional angiogensis towards their transplant sites. ...................................................................................................................... 99 4.2.5 Growth of large and small diameter fibres surrounding and within the lesion site is indistinguishable in SPARC null and WT OEC transplanted animals. ............................... 102 4.2.6 SubstanceP- and tyrosine hydroxylase-positive fibres grow significantly less in SPARC null OEC transplanted, than WT transplanted animals. ......................................... 104 4.3 Discussion ......................................................................................................................... 106 4.3.1 Reparative responses following WT or SPARC null transplantation are similar ....... 106 4.3.2 Spinal neuron populations are differentially affected by the absence of SPARC in transplanted OECs. .............................................................................................................. 108 4.3.3 Functional compensation may account for subtle differences in the outcome of WT versus SPARC null transplantation ..................................................................................... 109 4.3.4 DRG outgrowth model predicts secreted factors regulate neurite branching, whereas coculture regulates neurite fasciculation or elongation. ...................................................... 110 CHAPTER 5: AN ENRICHED CULTURE OF CORTICOSPINAL NEURONS CAN BE IDENTIFIED AND GROWN IN VITRO FROM THE POSTNATAL DAY 8 MOUSE .. 112 5.1 Introduction ....................................................................................................................... 112 5.1.1 Anatomy, function, and development of the corticospinal tract................................. 113 5.1.2 Corticospinal tract responses to spinal cord injury ..................................................... 116 5.1.3 Rationale and aims...................................................................................................... 117 5.2 Results ............................................................................................................................... 118 5.2.1 YFP expression in postnatal day 8 Thy1-YFP16JRS mice demarcates a population of corticospinal neurons. .......................................................................................................... 118 5.2.2 Development of protocols for the enriched culture of YFP-positive neurons ............ 121 viii   5.2.3 Postnatal day 8 YFP-expressing neurons extend processes over time in culture and respond to neurotrophins and neurite outgrowth inhibitors of corticospinal neurons ......... 125 5.2.4 Neuronal and glial contributions to dissociated corticospinal neuron culture change over time in vitro ................................................................................................................. 128 5.2.5 Death of glia and neurons differs over time in culture ............................................... 130 5.2.6 YFP-expressing corticospinal neurons also express other corticospinal projection neuron markers in culture .................................................................................................... 132 5.3 Discussion ......................................................................................................................... 134 5.3.1 The Thy1-YFP16JRS mouse permits identification of corticospinal neurons in vivo and in vitro ........................................................................................................................... 134 5.3.2 YFP-expressing neurons display characteristics of corticospinal neurons in vitro .... 135 5.3.3 Glial and neuronal population proportions change over time in vitro- implications for outgrowth analysis ............................................................................................................... 136 CHAPTER 6: OLFACTORY BULB ENSHEATHING CELLS PROMOTE CORTICOSPINAL AXON ELONGATION VIA A PLASMA MEMBRANE- DEPENDENT ACTIVITY ....................................................................................................... 143 6.1 Introduction ....................................................................................................................... 143 6.1.1 Olfactory ensheathing cell mediated repair of the injured corticospinal tract in vivo 143 6.1.2 Neurite outgrowth promotion by OECs in vitro: Secreted factors and cell-contact mediated mechanisms .......................................................................................................... 145 6.1.3 Astrocytes promote neurite outgrowth via secreted and surface-bound mechanisms 147 6.1.4 Rationale and aims...................................................................................................... 149 6.2 Results ............................................................................................................................... 150 6.2.1 Differences in dissociated P8 Thy1YFP CST neurite outgrowth can be assessed using Neurobinary ......................................................................................................................... 150 6.2.2 Coculture of CST neurons with glia and fibroblasts increases neurite outgrowth in phenotypically distinct manners. ......................................................................................... 152 6.2.3 OB OEC coculture increases axonal elongation and decreases dendritic outgrowth compared with base media treatment. ................................................................................. 154 6.2.4 Secreted factors from OB OECs do not contribute significantly to corticospinal neurite elongation ............................................................................................................................ 156 6.2.5 The effect of plasma membrane-derived proteins on corticospinal neurite outgrowth can be assessed using plasma membrane extraction and plating. ........................................ 159 ix   6.2.6 The plasma membrane of OB OECs contains a proteinaceous neurite elongation factor for corticospinal neurons. .................................................................................................... 163 6.2.7 The plasma membrane corticospinal axon elongation activity of OB OECs shares some neurite outgrowth properties with an astrocyte plasma membrane-derived factor ............. 165 6.2.8 OB OEC plasma membrane corticospinal axon elongation factors can overcome neurite outgrowth inhibition by MAG. ................................................................................ 167 6.3 Discussion ......................................................................................................................... 169 6.3.1 Secreted and cell-bound OB OECs factors alter the phenotype of CST neuron outgrowth. ............................................................................................................................ 169 6.3.2 OB OEC and astrocyte plasma membrane factors share an ability to increase CST outgrowth in similar manners .............................................................................................. 172 6.3.3 Corticospinal regeneration following spinal cord injury may benefit from OEC- mediated therapeutic interventions. ..................................................................................... 176 CHAPTER 7: DISCUSSION AND FUTURE DIRECTIONS .............................................. 178 7.1 Summary ........................................................................................................................... 178 7.2 Discussion and future directions ....................................................................................... 179 7.2.1 How effective is OEC transplantation as a therapeutic for spinal cord injury? ......... 179 7.2.2 Why are spinal cord tracts differentially responsive to OEC transplantation? ........... 183 7.2.3 What is the nature of the OEC plasma membrane CST axon elongation factor? ...... 186 7.3 Conclusions ....................................................................................................................... 189 References cited ......................................................................................................................... 191 Appendix .................................................................................................................................... 218 Appendix 1 Animal Care Certificates ..................................................................................... 218    x   LIST OF TABLES Table 1.1 Anatomical, physiological, and behavioural outcomes of OEC transplantation into the injured spinal cord......................................................................................................................... 35 Table 2.1 Antibodies used for immunocytochemical assessments ............................................... 61 Table 2.2 Antibodies used for immunohistochemical detection ................................................... 62 Table 3.1 OEC transplantation enhances angiognensis and directs blood vessels towards cell-rich areas  ............................................................................................................................................. 80 Table 5.1 Development of a postnatal day 8 corticospinal neuron culture: Experiments defining media, substrate, and density plating conditions. ....................................................................... 138 Table 5.2 Development of a postnatal day 8 corticospinal neuron culture: Defininf conditions for the enrichment of neurons. .......................................................................................................... 139 Table 5.3 Development of a postnatal day 14 corticospinal neuron culture: Defining media for survival. ....................................................................................................................................... 142   xi   LIST OF FIGURES Figure 1.1 Neuroanatomy of the primary olfactory neuraxis of the mouse .................................... 3 Figure 1.2 Embryonic and postnatal development of the mouse olfactory neuraxis ...................... 5 Figure 1.3 Morphology of OECs, Schwann cells and astrocytes in vitro ..................................... 13 Figure 1.4 Secreted factors from glia interact with neurite outgrowth machinery at the growth cone ............................................................................................................................................... 20 Figure 1.5 Membrane-bound cues derived from OECs can alter outgrowth dynamics within the growth cone ................................................................................................................................... 23 Figure 3.1 LP and OB OECs exhibit similar antigenic and morphological properties in vitro .... 68 Figure 3.2 Transplantation approach for direct or rostro-caudal introduction of OECs into a dorsolateral funiculus lesion ......................................................................................................... 69 Figure 3.3 Lamina propria and olfactory bulb ensheathing cells differ in their ability to migrate in vivo and in vitro ............................................................................................................................ 70 Figure 3.4 Lesion site area and cavity formation are decreased in LP versus OB OEC treated rats ....................................................................................................................................................... 72 Figure 3.5 OEC interactions with endogenous glia: LP and OB OECs promote Schwann cell infiltration ..................................................................................................................................... 75 Figure 3.6 Neurofilament-positive axon sprouting into the lesion site is highly promoted by rostrocaudal LP OEC transplantation ........................................................................................... 77 Figure 3.7 LP and OB OECs promote directional angiogenesis .................................................. 79 Figure 3.8 Autotomy is reduced by OB OEC transplantation but elevated by LP OEC transplantation ............................................................................................................................... 82 Figure 3.9 Substance P, calcitonin gene related peptidergic, serotonergic, and tyrosine hydroxylase-positive axons retract from the lesion site 24 hours following injury ...................... 84 xii   Figure 3.10 TH and SubP-positive axons are differentially responsive to LP and OB OEC transplantation ............................................................................................................................... 85 Figure 4.1 Secreted SPARC from OECs is partially responsible for their neurite outgrowth- promoting activity ......................................................................................................................... 93 Figure 4.2 In the lesioned spinal cord, SPARC is expressed by transplanted OECs, endogenous astrocytes, and on blood vessel laminae ....................................................................................... 97 Figure 4.3 The absence of SPARC in transplanted OECs does not alter lesion site formation, but changes immune responses to injury .......................................................................................... 100 Figure 4.4 The absence of SPARC from transplanted OECs does not alter laminin deposition or angiogenic reparative responses ................................................................................................. 101 Figure 4.5 Large and small diameter axon growth around or within the lesion site is similar in WT and SPARC null OEC transplanted animals ........................................................................ 103 Figure 4.6 Supraspinal and sensory spinal fibre populations are differentially affected by a lack of SPARC in transplanted OECs ................................................................................................ 104 Figure 5.1 Anatomy and development of the murine corticospinal tract ................................... 114 Figure 5.2 YFP expression in the postnatal day 8 Thy1YFP-16JRS mouse demarcates a population of deep layer cortical neurons ................................................................................... 120 Figure 5.3 YFP-positive neurons can be identified over time in culture and respond to known growth factors and inhibitors of corticospinal neuron outgrowth ............................................... 126 Figure 5.4 Proportions of neuronal and glil cells in postnatal day 8 corticospinal neuron culture change over time in vitro ............................................................................................................ 129 Figure 5.5 Cell death dynamics over time in vitro differ in YFP-positive and negative cell populations .................................................................................................................................. 131 Figure 5.6 YFP-positive neurons express markers of corticospinal, but not callosal, neurons in vitro ............................................................................................................................................. 133 xiii   Figure 6.1 Dissociated corticospinal neuron outgrowth can be measured by Neurobinary skeletonization ............................................................................................................................ 151 Figure 6.2 Neurite outgrowth of YFP-positive corticospinal neurons is increased by glial coculture ...................................................................................................................................... 153 Figure 6.3 Corticospinal neurite elongation is increased by OEC coculture, but neurite branching is increased by astrocyte coculture ............................................................................................. 155 Figure 6.4 Dendritic outgrowth is increased by baseline media conditions, but axonal outgrowth is increased at the expense of dendritic outgrowth when corticospinal neurons are cocultured with OB OECs  ........................................................................................................................... 157 Figure 6.5 OB OEC coculture promotes GAP43-positive and MAP2-negative neurite outgrowth ..................................................................................................................................................... 158 Figure 6.6 Corticospinal neurite branching is increased by OEC secreted factors ..................... 160 Figure 6.7 Generation of cellular plasma membrane fractions by sucrose step gradient preserves protein and sugar activities ......................................................................................................... 162 Figure 6.8 The plasma membrane of OB OECs contains a proteinaceous corticospinal neurite elongation factor ......................................................................................................................... 164 Figure 6.9 The plasma membranes of OB OECs and astrocytes share some similarities in CST neurite outgrowth activity ........................................................................................................... 166 Figure 6.10 Factors in the OEC plasma membrane can overcome CST neurite outgrowth inhibition by MAG ...................................................................................................................... 168  xiv   SYMBOLS, ABBREVIATIONS, AND NOMENCLATURE 5-HT    serotonin (5-hydroxy-tryptophan) AraC    Cytosine β-D-arabinofuranoside BDNF    Brain-Derived Neurotrophic Factor BLBP    Brain Lipid Binding Protein BrDU    5-bromo-2’-deoxyuridine cAMP    Cyclic Adenosine Monophosphate CAM    Cell Adhesion Molecule CGRP    Calcitonin Gene Related Peptide CHO    Chinese Hamster Ovary cell line CNS      Central Nervous system CNTF    Ciliary Neurotrophic Factor CSPG    Chondroitin Sulfate Proteoglycan CST    Corticospinal Tract CTB    Cholera toxin B CTIP2 Chicken ovalbumin upstream promoter transcription factor- interacting protein 2 DAPI    4’,6-Diamidine-2-phenylindole dihydrochloride db-cAMP   dibutyryl-cyclic adenosine monophosphate DiI    1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate DIV    Days in vitro DMEM   Delbucco’s Modified Eagle’s Medium DMEM/F12   Delbucco’s Modified Eagle’s Medium/Ham’s F12 DREZ    Dorsal Root Entry Zone DRG    Dorsal Root Ganglion xv   ECM    Extracellular Matrix ED    Embryonic  Day eGFP    enhanced GFP eYFP    enhanced YFP FACS    Fluorescence-Activated Cell Sorting FBS    Fetal Bovine Serum FdU    5-fluorodeoxyuridine FGF    Fibroblast Growth Factor FGFR    Fibroblast Growth Factor Receptor GAP43   Growth-Associated protein 43 GDNF    Glial Cell Line-Derived Neurotrophic Factor GEF    Guanine Exchange Factor GFAP    Glial Fibrillary Acidic Protein ICAT    Isotope Coded Affinity Tagging LMO4    LIM domain only 4 LP    Lamina Propria L-Q    L-Glutamine MAG    Myelin-associated Glycoprotein MAPK    Mitogen Activated Protein Kinase MLCK    Myosin Light Chain Kinase NCAM   Neural Cell Adhesion Molecule NF    Neurofilament NGF    Nerve Growth Factor NST    Neuron-specific Tubulin; βΙΙΙ Tubulin NT-3    Neurotrophin-3 xvi   OB    Olfactory Bulb OCM    OEC Conditioned Medium OE    Olfactory Epithelium OEC    Olfactory Ensheathing Cell OM    Olfactory Mucosa ONL    Olfactory Nerve Fibre Layer ORN    Olfactory Receptor Neuron Otx1    Orthodenticle 1 P75    Low affinity Nerve Growth Factor Receptor PCNA    Proliferating Cell Nuclear Antigen PLL    Poly-L-Lysine PNS    Peripheral Nervous system PSA-NCAM   Polysialated Neural Cell Adhesion Molecule RAG    Regeneration-Associated Genes R/C    Rostro/Caudal SC    Schwann Cell SCI    Spinal Cord Injury SPARC   Secreted Protein Acidic and Rich in Cysteines; Osteonectin SubP    Substance P T    Thoracic vertebral level TH    Tyrosine hydroxylase VEGF    Vascular Endothelial Growth Factor WT    Wild type  xvii   ACKNOWLEDGMENTS   I would like to thank all the members of the Roskams lab, those of yesteryear and today, who unstintingly lent their ideas, support, advice, and enthusiasm for science, when some was needed. In particular, thanks to Dr. Edmund Au, the most patient and thorough teacher, and Dr. Adele Vincent, a wise and judicious mentor.  For Jessica MacDonald and Barbara Murdoch, my fellow travellers along this rocky road, your wisdom, intelligence, dedication, and kindness have inspired me.  I would also like to thank my supervisor, Dr. Jane Roskams, who always expected the best of me, and led me to do the same of myself.  Thank you for lending your creativity and enthusiasm to this project.  The help of extraordinary collaborators made this work possible.  In particular, Dr. Wolfram Tetzlaff, Dr. Jie Liu, and members of the Tetzlaff lab, who provided technical expertise and sound advice that made investigations of OEC transplantation in the injured spinal cord possible. As well, thanks to Dr. E. Helene Sage, for the gift of the SPARC null mice, and Dr. Marie Filbin for the gift of CHO-MAG and CHO-R2 cell lines.  For engaging discussions and judicious experimental guidance, I would also like to thank the members of my supervisory committee, Dr. Wolfram Tetzlaff, Dr. Brian MacVicar, and Dr. Timothy Murphy.  This work was made possible through the generous financial support of the Canadian Institutes of Health Research, the Natural Sciences and Engineering Research Council of Canada, and The Michael Smith Foundation for Health Research. xviii   DEDICATION  For my Dad and my Mom, because you raised me to be a geek.  1   CHAPTER 1: LITERATURE REVIEW AND INTRODUCTION 1.1 Introduction 1.1.1 Regeneration in the peripheral and central nervous systems Neurite retraction, distal degeneration, and abortive regeneration are the prototypical responses of injured neurons of the central nervous system (CNS).  This limited restitution of damaged neuronal circuitry significantly hampers recovery of function, and represents a key dysfunction of spinal cord injury (SCI; Cajal 1928).  In the peripheral nervous system (PNS) the cascades of retraction and degeneration are not the final act of the injured neuron, but represent a midway stage in the recovery of connectivity finalized by the regrowth and targeting of injured neurons to approximate their original positions; regeneration (Waller 1850). These contrasting responses of CNS and PNS neurons are suggestive of the contributions of these differing environments and neuronal populations to the success or failure of regeneration, and intimate therapeutic possibilities to alter the outcomes of CNS injury. The bleak outlook of CNS injury and abortive regeneration has been questioned by studies of spinal neuron growth (Aguayo et al. 1981; Tello 1935), and investigations of neurogenic regions of the adult brain, in particular the hippocampus (Altman and Das 1965) and olfactory neuraxis (Zigova et al. 1992; Zigova et al. 1990).  These and other experiments have shown that some adult CNS neurons possess surprising outgrowth abilities, and will attempt to regenerate within the CNS when presented with an outgrowth-supportive environment (Pollard et al. 1973; Richardson et al. 1980).  A particular case for this permissive environment exists in the adult olfactory neuraxis, where functional neurogenesis and the restoration of neuronal connectivity occur throughout an organism’s lifetime (Schwob 2002).  Although the precise nature of the cellular and molecular mechanisms governing this unique adaptability have only been partially elucidated (Brunjes and Greer 2003), local progenitors have been identified that produce new olfactory receptor neurons (ORNs) following mature neuron apoptosis induced by axon severing or injury (Cowan et al. 2001; Graziadei et al. 1978; Schwob 2002).  The axons of these newly-generated neurons, located in the olfactory epithelium (OE), pierce the basement membrane, and coalesce to form olfactory nerve bundles (collectively the olfactory mucosa; OM) that travel caudally to reinnervate the olfactory bulb, and in so doing, cross the normally 2   non-permissive interface of the peripheral and central nervous systems (Figure 1.1; Carr and Farbman 1992; Carr and Farbman 1993).  The extraordinary reparative ability of the olfactory system has been the subject of much scrutiny because of the potential for therapeutic application to other non-permissive areas of the adult CNS.  It is for this reason that transplantation of olfactory ensheathing cells (OECs), specialized glia of the olfactory neuraxis, has proceeded in an effort to facilitate recovery from traumatic injuries to the spinal cord (reviewed in Richter and Roskams 2008), brain (Smale et al. 1996), dorsal or ventral roots (Li et al. 2007; Ramon-Cueto and Nieto-Sampedro 1994) and optic nerves (Li et al. 2003b) as well as for the treatment of amyotrophic lateral sclerosis (Chen et al. 2007b), Parkinson’s disease (Agrawal et al. 2004; Dewar et al. 2007), and to promote remyelination (Doucette 1990; Imaizumi et al. 2000; Kato et al. 2000; Radtke et al. 2004; Smith et al. 2001).  It is essential, however, that the properties and mechanisms employed by OECs to promote neurite outgrowth, both endogenously as well as following implantation, be elucidated in order to provide targeted therapies for nervous system injury.  An inquiry into OEC biology and how OECs impact the regeneration of injured spinal neurons are the goals of these investigations. 1.1.2 What are olfactory ensheathing cells (OECs)? OECs in the primary olfactory neuraxis Only recently has attention turned to the role of OECs in orchestrating the balance between maintenance of neuronal connectivity, survival, and neurogenesis in the olfactory neuraxis. OECs are unique glia found only in the olfactory system, that play a key role in the stimulation and maintenance of olfactory neuron turnover (reviewed in Chuah and West 2002; Doucette 1984; Ramon-Cueto and Avila 1998).  In the peripheral compartment of the olfactory neuraxis (Figure 1.1A), OECs are found both in contact with bundles of olfactory receptor neuron axons (ORNs) as their axons pierce the basal lamina and enter the lamina propria (LP), as well as distributed ad hoc throughout the olfactory LP (Figure 1.1B; Au and Roskams 2003; Tennent and Chuah 1996).  As ORN axons course towards their targets in the olfactory bulb (OB), they are encircled by OEC processes that further subdivide large axon bundles into smaller mesaxon bundles (Figure 1.1C), and they often precede axon terminal growth cones (Tennent and Chuah 1996).  ORN axons from across the OE coalesce, cross the cribriform plate, and form the olfactory nerve fibre layer (ONL), a structure that encircles the OB and is composed, in the 3    4   outer nerve fibre layer, of defasciculated ORNs running parallel to the OB in contact with large numbers of OECs (Au et al. 2002).  ORN axons are sorted based on their odorant receptor expression and remain in contact with OECs in the inner ONL before plunging to their mitral and tufted cell targets in the OB (Au et al. 2002). Thus from the initial events of axon outgrowth in the peripheral olfactory epithelium, to fasciculation, crossing of the peripheral-central nervous system border, and final targeting, ORN axons remain the intimate associates of OECs.  Developmental derivation of OECs OECs are thought to originate embryonically from the olfactory placode (Chuah and Au 1991), a specialized highly proliferative ectodermal region, induced in the mouse around embryonic day 10 (ED10; Figure 1.2A), and whose invagination and production of migrating primary sensory neurons, secretory supporting and mucus-producing cells, and brain lipid binding protein (BLBP)-positive glial cells eventually forms the olfactory and vomeronasal epithelia (Schlosser 2005).  As development progresses from the proliferation of sensory neurons and glia at the placode on ED10, OECs migrate with olfactory fascicles, forming the migratory mass at ED12, which accumulates superficially along the telencephalic vesicle by ED13 (Figure 1.2B; Valverde et al. 1992).  As the migratory mass encounters the telencephalon, the glia limitans between these migrating peripheral cells and the telencephalon deteriorates, allowing sensory neuron axons to penetrate the prospective OB, and thereby establishing a new glia limitans formed by OECs (Doucette 1993). Postnatal and adult derivation of OECs During postnatal and adult life (Figure 1.2D), OECs are likely the progeny of horizontal basal cells, multipotent progenitors and their transit-amplifying cells, located adjacent to the basal lamina within the OE (Carter et al. 2004; Roisen et al. 2001).  Proliferation of OECs themselves and their putative progenitors/transit amplifying cells, as measured by 5-bromo-2′- deoxyuridine uptake (BrdU), is greatest during the first postnatal month, and declines to a constant rate throughout adulthood, in the absence of pro-proliferative factors such as basic fibroblast growth factor (Watanabe et al. 2006).   5    6   1.1.3 Molecular basis of axon-OEC interactions Molecular basis of axon-OEC interactions in the olfactory pathway  The processes of OECs and ORN axons are closely apposed during embryonic development and normal replacement of ORNs.  The argument that OECs provide chemoattractive cues that guide ORN axons is unsubstantiated, since penetration of ORN axons into OB tissue explants, but not tectal explants, is not mirrored by differences in the abilities of OECs to migrate into these CNS targets (Storan and Key 2004).  It is clear, however, that intercellular interactions between ORNs and OECs are highly specialized (Van Den Pol and Santarelli 2003).  In vitro, ORNs grow preferentially and extend the longest neurites when cocultured with OECs over hippocampal glia, poly-L-lysine (PLL), or laminin (Kafitz and Greer 1999), and alignment of OECs along poly-D,L-lactide matrices induces cortical neurite alignment along OEC processes (Deumens et al. 2006a).  The reciprocal interactions between OECs and ORNs, including roles in outgrowth and axon sorting, are regulated by a number of cell adhesion and extracellular matrix molecules including: galectin 1, β2-laminin, and semaphorin 3A (Crandall et al. 2000),  p75 and embryonic N-CAM (Franceschini and Barnett 1996), as well as carbohydrate-decorated forms of NCAM (Storan et al. 2004), laminins, fibronectin, collagen IV (Kafitz and Greer 1997), cadherins (Akins et al. 2007), gap junctional communication (Blinder et al. 2003), and potassium release (Hartl et al. 2007). Extracellular matrix molecules governing OEC-ORN interactions  Expression of laminin, fibronectin, and collagen, to which α and β integrin heterodimers form receptors, has been documented in the developing and adult olfactory system.  Laminin, fibronectin and collagen type IV are all detected at ED14 in the migrating mass (Julliard and Hartmann 1998), and in the ONL of the ED18 rat, however, only laminin levels persist throughout adulthood, whereas fibronectin and collagen type IV levels drop postnatally (Doucette 1996).  Fibronectin and collagen IV expression can be reinstated by olfactory injury (Doucette 1996).  In vitro, OECs are immunoreactive for laminin and fibronectin, and have been shown to secrete collagens, laminins and fibronectin by isotope coded affinity tagging proteomics (ICAT; Au et al. 2007; Ramon-Cueto and Nieto-Sampedro 1992). In vitro, both laminin and fibronectin can promote ORN outgrowth, although more rapid and extensive 7   outgrowth is achieved on 3-dimensional collagen gels (Donnelly et al. 1998; Kafitz and Greer 1997).  When OECs are cultured on different extracellular matrix (ECM) substrates, this can significantly alter their neurite-outgrowth-promoting properties; laminin or Matrigel, a composite of laminin, collagen IV, heparin sulfate proteoglycans, and entactin, increases spreading of the OEC cell membrane and potentiates OEC-induced ORN outgrowth (Tisay and Key 1999).  Other ECM molecules such as neurocan and phosphacan have also been described in the olfactory neuraxis (Clarris et al. 2000).  Neurocan and phosphacan are chondroitin sulphate proteoglycans that are perhaps best known for their expression and subsequent inhibitory effect on process outgrowth following lesion to the CNS (Sandvig et al. 2004).  However, in the olfactory neuraxis, neurocan, first expressed in the migratory mass in both ORNs and OECs, and whose expression persists throughout adulthood, can be a potent promoter of ORN outgrowth (Clarris et al. 2000).  In contrast, the expression of phosphacan from ED11.5, primarily in structures bordering ORN outgrowth such as the mesenchyme and glomeruli, its subsequent downregulation during adulthood, and its inhibitory effect on ORN neurite outgrowth in vitro, suggest a developmental role in restricting exuberant ORN growth (Clarris et al. 2000). The cognate receptor for neurocan and phosphacan, tenascin-C, is also expressed by ORNs, particularly at fascicle borders and at points of contact with OECs (Zaidi et al. 1998).  Finally, boundaries formed by β2-laminin and galectin-1 expressing and non-expressing OECs, may form a sorting zone for ORN axons, since lactosamine-containing glycan-expressing ORNs grow preferentially on  β2-laminin/galectin-1 positive OECs, whereas neuropillin-positive ORNs avoid these regions (Crandall et al. 2000). Thus ECM-mediated interactions between ORNs and OEC appear to provide both permissive and instructive cues for ORN outgrowth and targeting.  Furthermore it is possible that the persistent expression of outgrowth-permissive ECM molecules in the olfactory system, as well as the increased expression of these molecules following olfactory injury, reflect an involvement of this ECM in the continuous outgrowth of new ORNs to the OB as well as the reinnervation of the OB after ORN axotomy. Cell adhesion molecule regulation of OEC-ORN interactions  Cell adhesion molecules (CAMs) such as those belonging to the Ig superfamily, like neural cell adhesion molecules (NCAMs) and L1, or integrins and cadherins, have functions in 8   the olfactory system, like those elsewhere in the nervous system, of regulating neuron and glial interactions with each other and with the ECM environment.  They may therefore also impact neurite fasciculation, outgrowth, and branching in the olfactory neuraxis  (Kiryushko et al. 2004). Cadherins are transmembrane glycoproteins composed of five cadherin domains which mediate Ca2+-dependent homophilic interactions, as well as Ca2+-independent heterophilic interactions with integrins and the fibroblast growth factor receptor (FGFR; Kiryushko et al. 2004).  Intracellular signaling through homophilic cadherin interactions impacts the actin cytoskeleton primarily through α and β-catenins (Riehl et al. 1996; Thoumine et al. 2006). Heterophilic signaling via the FGFR can impinge upon both PLCγ-mediated release of Ca2+from intracellular stores or activation of  mitogen-activated protein kinase (MAPK) pathways (Key 1998).  Both cadherins and catenins have unique spatiotemporal patterns of expression within the developing and mature olfactory neuraxis (Akins et al. 2007).  N-Cadherin (CDH2) and E- Cadherin (CDH1) are expressed in the olfactory epithelium, with highest expression in ORN axons.  N-cadherin expression in ORNs is correlated with the maturity of individual ORNs, as newly-formed glomeruli can be distinguished from the surrounding tissue by high N-cadherin immunoreactivity (Akins et al. 2007).  Placental cadherin (P-cadherin; CDH3) is localized to OEC processes surrounding NCAM-positive ORN axons.  Osteoblast-cadherin (OB-cadherin; CDH11) is similarly expressed by OECs, although expression levels decline sharply perinatally and continue to decline with age.  Although the roles of cadherins in the olfactory system have not been definitively demonstrated, it is likely that they play a similar role to those observed in retinal ganglion cells and dorsal root ganglia, such as neurite outgrowth promotion by E- and N- cadherin (Oblander et al. 2007; Riehl et al. 1996), or axon sorting by 6B, 7, R, and N-cadherin (Wohrn et al. 1998). Initial axon outgrowth in the OE is paired with migrating cells that express both L1 and polysialated-NCAM (PSA-NCAM), and these, as well as NCAM, continue to play roles in adult ORN outgrowth (Key 1998).  Binding partnerships are promiscuous between these players, as L1 can bind homophilically and to NCAM, ECM and integrins (Brummendorf and Rathjen 1995), whereas NCAM can bind itself, CSPG and HSPG (Brummendorf and Rathjen 1995).  L1 and NCAM both signal via FGFR1 in a calcium-dependent and MAPK-dependent manner  (Schmid 9   et al. 1999; Schmid et al. 2000).  These liberal interactions, and the expression of NCAM, ECM and L1 on both OECs and ORNs (Franceschini and Barnett 1996; Miragall et al. 1989; Storan et al. 2004), suggest a complex reciprocal regulation of growth/migration dynamics over development and adulthood. Substrate plating of L1 and NCAM can both increase the outgrowth of ORNs in vitro (Whitesides and LaMantia 1996).  However, the formation of the olfactory nerve is largely unaffected in the NCAM-180kDa null mouse, with only small increases noted in ONL thickness, suggesting the functions of NCAM or PSA-NCAM may be restricted to a sorting or fasciculation process (Treloar et al. 1997).  Similarly, while L1 expression in OECs appears widespread in vivo, and ORN outgrowth on OECs is markedly increased over growth on PLL, only a minority of OECs express L1 in vitro (Chuah and Au 1994).  The roles of OEC-expressed NCAM and L1 in promoting ORN neurite outgrowth has yet to be fully delineated. Other specialized interactions of OECs and ORNs Perisynaptic Schwann cells at the neuromuscular junction, and astrocytes at CNS synapses can actively participate in the modulation of synaptic efficacy through neurotransmitter-mediated increases in glial intracellular Ca2+ and subsequent actions at neuronal pre and/or post-synaptic sites (Jahromi et al. 1992; Pfrieger and Barres 1997; Robitaille 1998). Furthermore, neuron-driven Ca2+ changes in glia can alter glial production of growth factors, and may therefore impact neurite outgrowth (Masmoudi-Kouki et al. 2006; Wang et al. 2007). Intracellular increases in Ca2+concentration as a result of Ca2+ flux through a TRP-like channel in OECs decreases retinal ganglion cell neurite outgrowth in vitro (Hayat et al. 2003a; Hayat et al. 2003b). Freeze fracture of the ONL has demonstrated the presence of gap junctions at the interfaces of ORNs and OECs, an arrangement which could facilitate extrasynaptic (ephaptic) transmission, calcium waves, current oscillations, and paracrine communication (Blinder et al. 2003).  Indeed, this calcium coupling has recently been demonstrated in situ, by electrical stimulation of ORNs in slice preparation, and subsequent visualization of changes in OEC Ca2+ using Fluo-4 AM, a Ca2+ indicator dye (Rieger et al. 2007).  Furthermore, blockade of acetylcholine, glutamate, gamma amino-butyric acid (GABA), and nitric oxide does not abrogate the electrically-induced Ca2+ rise in OECs, but these transients were related to extracellular potassium released from active neurons (Rieger et al. 2007). These data suggest that modulation of OEC intracellular Ca2+in the differing environments of the olfactory neuraxis, such as the LP 10   versus the ONL, may provide a generalized rheostat for the integration of cues to regulate appropriate interactions with growing or targeting ORNs (Hayat et al. 2003a). 1.1.4 OEC response to olfactory injury Morphological and prolierative changes in OECs following olfactory injury Experimental injuries to the olfactory system can be produced using olfactory bulbectomy or axotomy, zinc sulfate irrigation, methyl bromide administration, or methimazole injection, all of which induce the death of mature ORNs, and eventually lead to proliferative responses of ORN precursors and the maturation and outgrowth of immature ORNs to reach the OB (Graziadei et al. 1978; Sakamoto et al. 2007; Schwob et al. 1995; Williams et al. 2004b). The responses of OECs to olfactory lesion may be instructive in discerning their contribution to recovery from olfactory and other nervous system injuries.  Following unilateral olfactory bulbectomy, the removal of an OB, or ORN axotomy at the cribriform plate, OECs encircling axon bundles do not divide or migrate, despite a slow wave of retrograde ORN degeneration (Li et al. 2005).  The expression and appearance of laminin in OEC bundles appears unaltered on the bulbectomized and control sides (Chehrehasa et al. 2005), the cross-sectional and longitudinal appearance of OEC ensheathments remains unaltered, and vacant spaces are visible where ORN axons have degenerated (Li et al. 2005).  The OEC response to chemical insult may differ somewhat from those responses observed following ORN axotomy.  Using light microscopic measurements, no changes in OEC migration within the OB were observed following intranasal zinc sulfate irrigation, and changes in numbers of proliferating cell nuclear antigen-positive (PCNA+), ED1-negative (a marker of macrophages) cells in the OB was not observed (Williams et al. 2004a).  This is contrasted by electron microscopic observations that indicate increases in OEC cell body electron density and a concomitant shift of the cell body away from axon bundles within the lamina propria two days following zinc sulfate treatment (Chuah et al. 1995).  At later time points, OECs appeared to have migrated from the former sites of axon bundles to reside immediately deep to the basal lamina and some OECs even migrated toward the OB (Chuah et al. 1995).  Thus migratory and proliferative responses of OECs following olfactory injury appear intimately linked to the physical and physiological changes associated with a particular injury, a caveat that may bear consideration when transplanting OECs to injury sites in the nervous system. 11 Physiological changes in OECs following olfactory injury Although proliferative or anatomical changes in OECs following insult appear subtle, OEC protein or RNA alterations are numerous, and may clarify, in part, some of the mechanisms by which OECs facilitate ORN outgrowth in an injury environment.  Following bulbectomy, a peak in ORN apoptosis at 24-40 hours post bulbectomy (Cowan et al. 2001; Shetty et al. 2005) is followed by cell proliferation and process outgrowth on days 2 through 7, and days 5 through 30, respectively (Getchell et al. 2005; Shetty et al. 2005).  Oligonucleotide microarray transcriptional profiling of the whole OE one day post bulbectomy reveals decreases in modulators of Rho and Rap, small GTPases involved in cytoskeletal structure and cell adhesion, including Rap1ga1, Rap2ip, Rgnef, Arhgef3, Kifap3, Kif9, and Rtkn (Shetty et al. 2005).  In contrast to 24 hour decreases in cytoskeletal modulators, analysis of day 5 and 7 transcripts by microarray and in situ hybridization shows increases in mRNAs that encode proteins with known involvements in axon outgrowth such as: abLIM1, GAP43, CAP23, Syn1, Fyn, Dlx5, and Emx2 (Shetty et al. 2005). Growth factor genes also show a complex regulation following bulbectomy; the activin A receptor, a TGFβ superfamily member, fibroblast growth factor 1 (FGF1), FGFR1, interleukin6 (IL6), and leukemia inhibitory factor receptor (LIFR) are all upregulated by 48 hours post bulbectomy (Getchell et al. 2005).  Increases in LIF protein, IL6, and their receptors have also been reported 2 to 3 days post bulbectomy in the olfactory nerve and in OECs by immunohistochemistry (Nan et al. 2001). Indeed, reconstitution of the OE following bulbectomy is incomplete in LIF knock-out mice (Bauer et al. 2003).  Other proteins with outgrowth- associated functions show similar profiles post bulbectomy.  High TrkB and low TrkC expression in ORN axons of the normal OE follows a predictable pattern of regulation following bulbectomy.  TrkB disappearance from the OE at early times after bulbectomy is followed by intense upregulation in the cell bodies of newborn immature receptor neurons, and then a shift in TrkB immunoreactivity to pioneer axon bundles as process outgrowth occurs (Roskams et al. 1996).  NCAM and GAP43 protein similarly increases in expression 3 to 4 days post bulbectomy, and continues until axon outgrowth is complete (Yamashita et al. 1998).  Olfactory injury also induces a transient upregulation in netrin-1 (Astic et al. 2002), secreted protein acidic and rich in cysteines (SPARC; osteonectin; bone morphogen 40; Au et al. 2007) and erbB2 in the LP, and no change in the expression of neuregulin 1α (Williams et al. 2004a).  These changes in 12   gene and protein expression in OECs and ORNs following olfactory injury suggest that mutual regulation underlies the ability of the olfactory neuraxis to reconstitute itself. 1.1.5 Olfactory ensheathing cells in vitro Morphology and antigenicity  Enriched cultures of OECs can be produced by harvesting cells from either extracranial (LP and OE; Au and Roskams 2002; Lu et al. 2001) or cranial (ONL; Ramon-Cueto and Nieto- Sampedro 1992) compartments of the olfactory neuraxis.  To rid cultures of contaminating meningeal or astrocytic cells,  purification protocols such as fluorescence activated cell sorting (FACS; Barnett et al. 1993), p75-mediated immunopanning (Ramon-Cueto and Nieto-Sampedro 1994), Thy1.1-mediated cytotoxic lysis (Au and Roskams 2002), differential substrate adhesion (Nash et al. 2001), or magnetic activated cell sorting (Krudewig et al. 2006) are followed. Cultures contain cells with variable morphologies that are largely dependent on the developmental stage of the cells, culture conditions, or coculture with other cells (Ramon-Cueto and Avila 1998; Vincent et al. 2003).  OECs adopt mainly either a process-bearing bipolar spindle-shaped morphology (Figure 1.3A), akin to Schwann cell morphology in culture (Figure 1.3B), or a flattened astrocytic-like morphology (Figure 1.3C; Ramon-Cueto and Avila 1998). Because OECs share common glial markers with both Schwann cells and astrocytes, culture enrichment is verified using a number of immunocytochemical markers, the most common of which are: p75, the low affinity neurotrophin receptor, S100, the calcium-binding protein common to many glia of the PNS and CNS, and glial fibrillary acidic protein (GFAP), the major intermediate filament cytoskeletal protein present in astrocytes (Au and Roskams 2003; Ramon- Cueto and Nieto-Sampedro 1992).  Expression of p75 and GFAP is variable in cultured OECs; a majority of OECs express high levels of p75, with some low-expressing cells present, whereas GFAP immunoreactivity is low in OECs compared with astrocytic expression levels (Au and Roskams 2003; Ramon-Cueto and Nieto-Sampedro 1992).  OECs also express a unique antigenic repertoire that can be utilized for their identification in vitro including:  NCAM, PSA- NCAM, laminin, L1, β1-integrin, human natural killer 1 (HNK-1), CD44 (hylaluronic acid receptor), and NG2 (a chondroitin sulfate proteoglycan), growth factors and receptors like 13    14   vascular endothelial growth factor (VEGF), neuregulin1,  pituitary adenylate cyclase-activating peptide (PACAP), and Notch 3, cytoskeletal and structural proteins such as nestin, myelin basic protein (MBP), vimentin, and P200, as well as antibodies against Ran2, A5E3, 1.9.E, 4.11.C, and O4 (Au and Roskams 2003; Heredia et al. 1998; Krudewig et al. 2006; Ramon-Cueto and Avila 1998; Ramon-Cueto and Nieto-Sampedro 1992; Thompson et al. 2000). Recently, the smooth muscle proteins SM22α and calponin were reported as markers that uniquely separate OECs from Schwann cells (Boyd et al. 2006; Jahed et al. 2007), however, further analysis suggests these markers are confined, in the olfactory system, to fibroblasts, meningeal cells, and smooth muscle cells of the vasculature (Ibanez et al. 2007).  While a unique identifier of OECs remains elusive, the plethora of proteins expressed by OECs that are shared by other glia of the PNS and CNS strongly suggest the multifaceted functions of these cells. OEC growth properties  Proliferation of OECs in vitro is dependent on the age of the donor, time elapsed in culture, and culture conditions.  In particular, proliferative rates of OECs derived from the embryo or early postnatal animals are higher than OECs derived from adult donors, as has also been observed in vivo (Watanabe et al. 2006).  Similarly, OECs exhibit senescence over time in culture, such that proliferation slows with successive passaging of cells, and production of growth factors also decreases or is altered (Au et al. 2007).  In comparison with astrocytes and fibroblasts, OECs also exhibit slow division rates in the absence of exogenous mitogens, although paracine-derived factors can ameliorate this effect, since high-density OEC cultures proliferate more rapidly than companion low-density cultures (Van Den Pol and Santarelli 2003; Yan et al. 2001a).  Because OECs share several cell signaling properties with astrocytes and Schwann cells, known mitogens for these glia have been tested for their ability to promote OEC proliferation.  Increases in tritiated thymidine-labeled OECs following treatment with the Schwann cell mitogens heregulin, fibroblast growth factor 2 (FGF2), platelet-derived growth factor, and insulin-like growth factor-1 can be further potentiated by the addition of the adenyl cyclase activating compound, forskolin (Yan et al. 2001a).  Interestingly, the presence of insulin- like growth factor-1 in conditioned medium from OECs could also partially explain the paracrine influences of OECs on their own proliferation (Au et al. 2007).  Combinations of FGF2, forskolin, and heregulin can also override proliferation arrest and these factors also modulate the 15   morphology and antigenicity of OECs (Alexander et al. 2002).  Treatment with FGF2, forskolin, and heregulin is sufficient to induce a homogenous Schwann cell-like morphology, probably by increasing intracellular cAMP concentrations, and upregulates expression of O4 and p75 (Alexander et al. 2002; Vincent et al. 2003). OECs are also responsive to other growth factors, including neurotrophin 3 (NT-3; Bianco et al. 2004), glial cell line-derived neurotrophic factor (GDNF; Cao et al. 2006), and hepatocyte growth factor (Yan et al. 2001b).  This is perhaps unsurprising, since by RT-PCR, OECs highly express the TrkC receptor and p75, which could interact with NT-3 (Bianco et al. 2004).  OECs also express transcript for Ret that forms the GDNF family receptor, GFRα-1 (Lipson et al. 2003), and c-Met protein, which is responsible for the mitogenic effect of HGF on OECs, since addition of c-Met antiserum can abrogate the proliferative effect of HGF (Yan et al. 2001b).  The proliferative response of low density OEC cultures to GDNF (Cao et al. 2006), and the production of GDNF by OECs in vitro (Lipson et al. 2003) may also partially explain observations of OEC density-dependent proliferation. OEC secretion of soluble factors  Release of soluble factors by OECs is a major mechanism whereby OECs influence the growth of other OECs, glia, and in particular, the outgrowth of neurons in coculture (Au et al. 2007; Chung et al. 2004).  Release of neurotrophins from OECs is well characterized and is of particular interest because of the role these proteins play in providing permissive (trophic) and instructive (tropic) cues to regulate neuronal outgrowth during development and regeneration (Gillespie 2003).  Release of neurotrophins by OECs is akin to their release in other glia; pre- pro-neurotrophin precursors formed in the rough endoplasmic reticulum are cleaved in the trans- Golgi network or in secretory granules to form mature neurotrophins which are released either constitutively or in a regulated manner (Lim et al. 2003).  Release of neurotrophins and neurite outgrowth promoters from OECs can be altered in kind or quantity by contact with astrocytes or neurons (Au et al. 2007; Deumens et al. 2006d).  Media harvested from purified OEC populations and analyzed by ELISA or ICAT, or cell extracts analyzed by RT PCR, have yielded a plethora of neurite outgrowth promoters (Au et al. 2007; Lipson et al. 2003; Woodhall et al. 2001).  OEC-conditioned media (OCM) is positive for nerve growth factor (NGF) and brain derived neurotrophic factor (BDNF) by ELISA, and RT PCR confirms the presence of NGF, 16   BDNF, glial cell-line derived neurotrophic factor (GDNF), and neurturin transcripts (Woodhall et al. 2001).  Ribonuclease protection assay has also described the presence of OEC-derived transcripts for NGF, BDNF, and GDNF, as well as cilliary neurotrophic factor (CNTF) and artemin (Lipson et al. 2003).  An OEC cell line devoid of contaminating cells also expresses mRNA for NGF, BDNF, NT-4/5, and neuregulins (Boruch et al. 2001). Interestingly, despite the variety of secreted factors produced by OECs, their ability to promote outgrowth from different neuronal populations can be variable. OEC conditioned media applied to explants of Remak’s ganglion (a parasympathetic ganglion), sympathetic, ciliary, nodose, trigeminal, and spinal ganglia exhibited little effect on neurite outgrowth over baseline treatment, although coculture increased outgrowth (Lipson et al. 2003).  A similar dichotomy was observed following addition of OEC conditioned media versus direct OEC-neuron contact in an in vitro scratch model of neuron injury (Chung et al. 2004), or following OCM addition to adult retinal ganglion cell cultures versus the larger increase in outgrowth observed upon coculture (Sonigra et al. 1999).  In contrast, neurite branching and outgrowth, but not elongation, is increased in dorsal root ganglion cultures when OCM is applied (Au et al. 2007).  Similarly, the outgrowth of primary neurites from olfactory receptor neurons on OEC monolayers or exposed to OEC secreted factors is indistinguishable (Kafitz and Greer 1998).  Cumulatively, these results suggest that effects of secreted factors by OECs are either (1) limited to close-range interactions or (2) depend for their release or production on OEC-neuron contact. In contrast, secreted factors from OECs appear to profoundly regulate other glia, as well as themselves.  The ability of Schwann cells to migrate across an inverted coverslip is significantly increased by coculture of Schwann cells with OECs over other cell lines (Cao et al. 2007a).  Furthermore, OCM is equivalently able to increase Schwann cell migration, even in the presence of astrocytes, with which Schwann cells normally form discrete barriers (Cao et al. 2007a; Lakatos et al. 2000).  The effect can be abrogated by the addition of a function-blocking antibody to NGF, suggesting NGF secreted from OECs may be one crucial modulator of Schwann cell migration (Cao et al. 2007a). Secretion of SPARC by OECs can also decrease the dispersion of Schwann cells away from the core of a dorsal root ganglion explant or decrease Schwann cell migration in a Boyden chamber assay, and changes the neurite outgrowth promoting properties of each individual Schwann cell; removal of Schwann cells from a DRG 17   explant annuls the neurite outgrowth-promoting effect of exogenous SPARC addition, whereas addition of SPARC increases the amount of neurite outgrowth stimulated per Schwann cell (Au et al. 2007). This suggests that in the complex environment of a spinal cord injury, secreted factors from transplanted OECs may alter the growth of injured fibre tracts by a direct effect on neurons, or by indirectly influencing the migration or outgrowth-promoting properties of Schwann cells found within the lesion mileu. OEC interactions with neurons in vitro A number of different neuronal populations, including ORNs, cerebellar granule neurons, and cortical neurons, have demonstrated increased neurite growth when cultured on OECs over control substrates such as poly-L-lysine, laminin, or fibronectin (Chung et al. 2004; Kafitz and Greer 1999; Van Den Pol and Santarelli 2003).  Much of this increased outgrowth may issue from the secretion of soluble factors and production of cell surface factors such as those discussed above (Chung et al. 2004).  Some groups have demonstrated similar outgrowth of embryonic ORNs as a result of coculture with OECs or application of their secreted factors (Kafitz and Greer 1998).  Other interesting interactions between OECs and axons have also been reported; granule neurons show significant orientation effects when cocultured with OECs, and they extend along the long axes of OECs (Van Den Pol and Santarelli 2003).  Also, the number of dendrites produced by embryonic cortical neurons is increased by OB OEC coculture or conditioned media (Le Roux and Reh 1994).  Controversially, OECs have also been reported to myelinate axons of an appropriate calibre in vitro (Devon and Doucette 1992). The signals required to induce myelination by Schwann cells include an axon diameter greater than 1μm, the high expression of neuregulin 1 type III from associated axons, and the reception of this neuregulin signal by an ErbB3 or B4 receptor which dimerizes with ErbB2 to activate MAPK and PI3K pathways and their subsequent gene targets (Nave and Salzer 2006). Since ORN axons are relatively small (less than 0.5 μm), this may preclude the ability of OECs to myelinate in their native environment (Franklin 2003).  OECs do, however, express ErbB2,3, and 4 in vivo and in vitro, and expression of ErbB2 is upregulated in the OE following ORN injury by ZnSO4 irrigation (Thompson et al. 2000; Williams et al. 2004b).  In vitro, OEC expression of GalC and myelin basic protein (MBP) 4 weeks after seeding onto dissociated DRG cultures was first reported by Devon and Doucette (1992). This finding of OEC myelination has 18   been questioned, given the possibility that myelination was the result of residual Schwann cell contaminants that survived antimitotic treatment in the plus OEC conditions (Plant et al. 2002). However, in subsequent studies by Devon and Doucette (1995), the lack of myelination by residual Schwann cells in control media conditions, and the presence of myelinating prelabelled PKH26-positive OECs under L-ascorbic acid-free conditions, a condition usually required for Schwann cell myelination, is highly suggestive of an ability of OECs to myelinate in vitro. These studies of OEC myelin formation coupled with receptor expression data suggest that given appropriate signals and targets, OECs may become capable of myelination in vitro (Akiyama et al. 2004). 1.1.6 How do glia influence neurite outgrowth? Mechanisms of neurite outgrowth Neurite outgrowth is regulated through a dynamic interplay between cytoskeletal reorganizing factors and the cytoskeleton itself, in the motile, growing and sensing tip of the axon, the growth cone (Baas and Luo 2001).  Extension or retraction of the growth cone can be mediated by secreted or membrane-bound cues present in the developing or regenerating nervous system, many of which are produced by glia. Regulation of neurite outgrowth by glia: Actin and microtubule dynamics in the growth cone  The turning of a growth cone towards a target, or growth cone extension following injury, depends upon a reorganiziation of actin filaments (filamentous actin; F-actin) and microtubules, the primary cytoskeletal components of the growth cone, and the final conversion point of guidance-cue induced signalling cascades (Zhou and Cohan 2004).  When an attractive or repulsive cue binds to receptors on the growth cone, signals converge upon actin filaments and actin meshwork present in the peripheral domain of the growth cone, and act upon microtubules in the central domain (Dent and Gertler 2003).  The rate of extension from the growth cone is dependent on the polymerization of F-actin at the leading edge, the rate of retrograde actin flow towards the central domain as regulated by myosins, and the extension and invasion of microtubules into the peripheral domain, that become stabilized through F-actin interactions (Gallo and Letourneau 2004).  Guidance cues can direct axon extension, and thus actin and 19   microtubule dynamics, through modulations of the events of growth cone protrusion, engorgement, and consolidation, and do so largely via a family of master regulators, the Rho- family of GTPases (Gallo and Letourneau 2004).  The Rho-family GTPases, Cdc-42, Rac1, and RhoA control actin polymerization, branching, depolymerization, and actinomyosin contractility; microinjection of Rac1, Cdc-42, and RhoA into 3T3 cells results in the localized generation of lamellipodia, filopodia, and stress fibres, respectively (Ridley and Hall 1992).  Because Cdc-42, Rac1, and RhoA are small GTPases, they cycle between active GTP-bound, and inactive GDP- bound states, and their intrinsic GTPase activity can be further modified through the actions of guanine nucleotide exhange factors (GEFs) to facilitate exchange of GDP for GTP, or GTPase activating proteins (GAPs; Huber et al. 2003).  GTP-bound forms of Cdc-42, Rac1, and Rho, can recruit effector proteins, eventually resulting in actin polymerization at the leading edge, actin polymerization at the distal (barbed) ends of F-actin (Suetsugu et al. 1998), actin depolymerization, or changes in myosin activity to alter contractility, a key determinant of F- actin retrograde flow (Dent and Gertler 2003).  These processes and pathways will be discussed as they pertain to glial-derived guidance and outgrowth signaling. Regulation of the growth cone cytoskeleton by glia-derived secreted guidance cues  Major families of secreted guidance cues produced by glia include neurotrophins, netrins, slits, and semaphorins (Huber et al. 2003).  OEC-produced neurotrophins (NGF, BDNF, NT-4/5; Boruch et al. 2001; Lipson et al. 2003; Woodhall et al. 2001), growth factors (CNTF, GDNF, neurturin, artemin; Boruch et al. 2001; Lipson et al. 2003; Woodhall et al. 2001), and semaphorin (Sema3A; Schwarting et al. 2000) can influence the growth cone cytoskeleton to induce attraction or repulsion.  Signalling of neurotrophins (Figure 1.4A) through their receptor tyrosine kinases TrkA,B, and C, results in receptor dimerization and transphosphorylation, leading to PI3- kinase and MAPK pathway activation, which can directly induce axon extension or gene transcription for axon extension.  Concurrently, PI3-kinase can activate RhoG, an activator of Cdc-42 and Rac1 (Huber et al. 2003).  Cdc-42 and Rac act on p21-activated kinase (PAK) to inhibit myosin light chain kinase (MLCK), decreasing myosin phosphorylation and actomyosin contractility (Dickson 2002).  Cdc-42 and Rac also affect the activity of N-WASP and Scar/WAVE to bind the Arp2/3 complex and initiate monomeric actin addition, resulting in axon branching (Dickson 2001). 20    21   Growth factor receptors and semaphorin signaling similarly converge on the aforementioned molecular modulators of the actin cytoskeleton (Figures 1.4B,C).  Sema3A binding to its obligate coreceptor neuropilin 1 or 2, followed by Plexin complex formation primarily results in neurite repulsion conveyed through RhoD activation, Rac inhibition, and MLCK activation, leading to increased actin retrograde flow, and removal of actin from the minus end by the actin-severing protein, cofillin (Mann and Rougon 2007).  The MAPK pathway can also be recruited by semaphorin ligand interactions, and recruitment of erk1/2 can lead to repulsion (Mann and Rougon 2007). Finally, binding of growth factors or neurotrophic factors such as CNTF and GDNF leads to downstream events that also converge at Rho GTPases and the actin cytoskeleton (Lingor et al. 2008).  GDNF, and its family of ligands including neurturin and artemin, can influence motor neuron branching at the developing neuromuscular junction (Airaksinen and Saarma 2002) and process outgrowth by binding to their receptor tyrosine kinase, Ret (REarraged during Transcription) via ligand-specific coreceptors GFRα1-3 either in cis (on the same cell), or in trans (on neighbouring cells) via cleavage of the GFRα-bound glycosyl-phosphatidylinositol (GPI) linkage (Crone and Lee 2002; Ledda et al. 2002).   Transphosphorylation of dimerized Ret can result in activation of MAPK, PI3K/Akt/Ras, Jun, Src, and PLCγ pathways, depending on residue-specific phosphorylation, to impact actin dynamics within the growth cone (Runeberg- Roos and Saarma 2007).   GDNF or neurturin signaling through Ret can promote neurite outgrowth when bound to GFRα2a or 2c in Neuro2A cells through activation of MAPK and Akt, whereas binding to GFRα2b inhibits neurite outgrowth (Yoong and Too 2007).  Furthermore, shedding of GFRα subunits from target structures in the presence of GDNF can create a further guidance cue to Ret-expressing sympathetic neurons, and potentiates neurite outgrowth in vitro and in vivo (Ledda et al. 2002).  A complication of GDNF-derived signaling in the growth cone derives from the ability of GDNF to form complexes with NCAM in areas of the CNS/PNS with low or undetectable levels of Ret (e.g. cortical neurons; Paratcha et al. 2003; Zhou et al. 2003). This complex modulates the cytoskeleton through focal adhesion kinase (FAK), whose phosphorylation triggers interactions with paxcillin and cas, followed by Grb2 and Ras/MAPK pathway activation (Hanks and Polte 1997; Schlaepfer et al. 1994).  CNTF binding to its receptor, CNTFRα, gp130 and LIFR can also regulate neurite outgrowth in retinal ganglion cells (Lingor et al. 2008; Ozog et al. 2007), through Jak/STAT and MAPK pathways (Murphy et al. 22   1997). Thus, secretion of a number of guidance cues by OECs can modulate the growth cone cytoskeleton, providing instructive and permissive cues for neurite elongation, branching, or turning during development or regeneration. Regulation of the growth cone cytoskeleton by membrane-associated guidance cues from glia  OECs, astrocytes, Schwann cells, and oligodendrocytes are also the source of many membrane-associated guidance cues responsible for mediating growth cone attraction and repulsion.  These include: ephrins, receptor protein tyrosine phosphatases, cell adhesion molecules, integrins, cadherins, and the myelin-associated outgrowth inhibitors, Nogo, Myelin- associated glycoprotein (MAG), and oligodendrocyte-myelin glycoprotein (OMgp; Doucette 2001).  OEC expression of ephrin-B2 (St John and Key 2001), NCAM (Storan et al. 2004), L1 (Miragall et al. 1989), β1-integrin (Au and Roskams 2003), cadherin 3 and 11 (Akins et al. 2007), as well as many ECM molecules such as laminin, fibronectin, and collagens suggest the diversity of factors that could interact with receptors at the neuronal growth cone to induce cytoskeletal rearrangements during development or regeneration. Engagement of the transmembrane ephrin Bs to EphB receptors results in bidirectional signaling within the two cells forming a contact, and defines the axonal trajectories of many populations of projection neurons (Figure 1.5A; Murai and Pasquale 2005).  The Eph consitutively-bound GEF, ephexin, becomes activated following transphosphorylation of dimerized Eph subunits, and activates RhoA, while inhibiting Cdc42 and Rac (Huber et al. 2003).  Active Rho phosphorylates ROCK and inhibits PAK activity, promoting growth cone collapse through actin depolymerisation (via ROCK) and increased F-actin contraction (via PAK; Egea and Klein 2007).  Actions of Vav2, an EphB-specific GEF, can also include Rac1- dependent endocytosis of the ephrin-Eph complex, turning an initial adhesive interaction into repulsion (Cowan et al. 2005; Egea and Klein 2007).  Although Eph-ephrin interactions typically result in cell repulsion, this appears contradictory, since interaction of two transmembrane receptors would suggest adhesion.  Axon repulsion by GPI-linked ephrinAs is induced through ectodomain cleavage by A-Disintegrin-And-Metalloprotease 10 (ADAM10; Hattori et al. 2000); it is unknown how transmembrane ephrinB associations generally induce repulsion, although there are reports of attractive roles for ephrinBs in specific contexts 23   24    (Marquardt et al. 2005; McLaughlin et al. 2003). NCAM, L1 (Figure 1.5B), or cadherin homophillic interactions in trans lead to phosphorylation and activation of FGFR, signaling through PLCγ and Fyn, and increases in intracellular calcium, followed by the activation MAPK (Huber et al. 2003).  NCAM clustering can also recruit FAK to this complex, resulting in transient FAK phosphorylation and actin polymerization (Beggs et al. 1997).  L1 signalling is analogous to NCAM events, although it is associated with Src, the non-receptor tyrosine kinase, and activates MAPK, to direct cytoskeletal rearrangement through Rac and PI3K (Schmid et al. 2000). Cadherin binding can induce F-actin assembly both via direct and indirect linkages to the cytoskeleton (Figure 1.5C).  Cadherin complexes and activation of the FGFR regulate calcium influx through N and L-type calcium channels from the extracellular space, and lead to calmodulin kinase (CaMK) activation and actin assembly at the leading edge.  This pathway is sufficient to drive neurite outgrowth responses provoked by cadherins, however, a more direct regulation of the actin cytoskeleton can also occur (Doherty et al. 1991; Ranscht 2000; Skaper et al. 2001).  Cadherins can form a core complex with β-catenin bound directly to the cytoplasmic tail of cadherin, and α-catenin bound simultaneously to the N-terminal region of β-catenin and to the actin cytoskeleton or several actin-binding proteins (Thoumine et al. 2006).  The integrity of the interaction between cadherin and β-catenin is regulated by phosphorylation of β-catenin, with β-catenin phosphorylation associated with cadherin inactivation and decreased adherence between interacting cells (Huber et al. 2003; Li et al. 2000).  Thus, membrane-associated cues expressed by OECs and other glia that form contacts with growing axons may serve as adhesive, permissive, and possibly instructive cues for neurite outgrowth. 1.1.7 What is the relationship between OECs and other glia of the central and peripheral nervous systems? Although OECs clearly display highly specialized interactions with ORN axons in their native environment, they also share many features with astrocytes, in the CNS, and Schwann cells (SCs), in the PNS.  Despite the varying origins of these classes of glia, OECs are derivatives of the olfactory placode, whereas Schwann cells are formed from the neural crest and 25   astrocytes from progenitors of the subventricular zone in the brain, they share surprising similarities, in morphology, antigenicity, and in their interactions with neurons. Morphology and antigenicity OECs display highly plastic morphologies, and can change in antigenic profile depending on culture conditions, in a manner similarly described of SCs and astrocytes.  Cultured SCs typically bear bipolar, long, spindle-like processes. OEC morphology can be highly reminiscent of SCs, particularly under serum-free culture conditions with the exogenous addition cAMP- increasing factors (Vincent et al. 2003) .  The SC-like morphology of OECs can also be reversed to resemble the prototypical flattened, multiple process-bearing morphology of astrocytes by the addition of serum (Vincent et al. 2003).  These morphological changes of OECs in response to culture conditions also resemble those described for astrocytes and SCs; serum-containing medium produces flattened morphologies in astrocytes, while serum-free medium, and addition of dibutryl-cAMP (dB-cAMP), induces stellate reactive-like astrocyte morphologies (Moonen et al. 1975).  OECs also share antigenic expression profiles with astrocytes and SCs, as well as a large number of secreted and membrane-bound factors that may impinge upon neurite outgrowth pathways (Wewetzer et al. 2002).  OEC initiation of high p75 expression after several days in culture, and their expression of S100 protein, a calcium-binding protein, closely resembles the expression and distribution of these proteins in cultured SCs (Barnett 2004).  Flattened astrocyte- like OECs can also display high levels of GFAP reactivity, PSA-NCAM expression with very low p75 detected (Au and Roskams 2003; Barnett 2004). Physiology and intercellular interactions Perhaps even more compelling than the similarities and differences between OECs, astrocytes, and SCs in p75, S100, and GFAP antigen expression, are the shared and differing expression profiles of transcripts and proteins that may ultimately impact neurite outgrowth or interactions of these cells with each other in vitro or in vivo. Transcriptional profiling of OECs, astrocytes, and SCs, generated using the Pan Rat 5K microarray, has revealed a large number of transcripts shared by all three glial cells, shared by only two glial cells, or uniquely expressed by a single population (Vincent et al. 2005).  While OECs and SCs are more similar than OECs and astrocytes in enriched and depleted transcripts, 26   all cells share a transcript profile rich in known determinants of neurite outgrowth, some of which include: cell adhesion molecule with homology to L1CAM (Chl1), tissue inhibitor of metalloproteinase 2 (TIMP2), chondroitin sulfate proteoglycan 3 (CSPG3), fibronectin, collagen type Vα1, collagen type Vα2, and connective tissue growth factor (Vincent et al. 2005). Furthermore, proteomic analysis of OEC, astrocyte, or SC secreted factors by isotope coded affinity tagging (ICAT), ELISA, Western blotting, or immunohistochemistry has similarly described the production of growth factors and ECM components shared by these cells, such as: NGF, basic FGF, CNTF, and many laminins and collagens (Au et al. 2007; Erkanli et al. 2007; Muller et al. 2007; Woodhall et al. 2001).  The similarities and differences in expression of neurite outgrowth-promoting and inhibitory molecules may ultimately underlie the abilities of OECs, SCs and astrocytes to promote regeneration following injury to the nervous system. A further complexity of the neuron growth-promoting relationships of endogenous or transplanted OECs, SCs, or astrocytes, is their different capacities to interact with each other. When SCs and astrocytes are cocultured, both cells become hypertrophic (individual cells increase in area; Ghirnikar and Eng 1994; Ghirnikar and Eng 1995; Lakatos et al. 2000), astrocytes and SCs segregate to homotypic cell islands or form linear borders, and SCs do not migrate freely over astrocyte monolayers (Lakatos et al. 2000).  In contrast, OECs intermingle freely with astrocytes, do not induce astrocytic hypertrophy, reduce GFAP and CSPG expression in reactive astrocytes, or induce less CSPG expression in astrocytes than SCs (Lakatos et al. 2000; O'Toole et al. 2007).  Differential regulation of N-cadherin in SCs and OECs seems to underlie variation in their interactions with astrocytes.  Although N-cadherin expression is similar in astrocytes, SCs, and OECs, a cyclic peptide inhibitor of N-cadherin, or RNA interference (RNAi) knockdown of N-cadherin decreased SC adhesion to astrocytes, and increased the speed of their migration on astrocyte monolayers, while neither of these treatments affects OEC interactions with astrocytes (Fairless et al. 2005).  The importance of permissive SC or OEC interactions with astrocytes, i.e. those interactions that reduce astrogliosis and cell barrier formation, can be underscored by the similarity between observations of these interactions in vitro and those in vivo.  Following implantation of OECs into normal white matter of the spinal cord, less GFAP-reactive area or CSPG expression is induced than when SCs are transplanted in the same paradigm (Lakatos et al. 2003).  Furthermore, transplantation of SCs into a transected spinal cord produces barriers at the interface between transplanted and host 27   tissue, results in the tortuous turning of growing axons within the graft, and limits the ability of axons to exit the distal transplant-host interface (Xu et al. 1997).  Capping of SC-seeded channels with OECs in a model of spinal cord transection ameliorates these outcomes, and results in increased growth of fibres from the transplant and into the distal spinal cord (Ramon- Cueto et al. 1998). These abilities of OECs to interact with, rather than segregate from, astrocytes may also reflect differences in the PNS-CNS transition zone, or glia limitans, of the first cranial nerve versus the glia limitans formed by astrocytes and SCs at the dorsal root entry zone.  While the transition between dorsal roots and spinal cord (PNS-CNS) is clearly-defined by a laminin-rich interface between SCs and astrocytes that is only penetrated by axons, the transition zone of the olfactory neuraxis is poorly-defined, and appears to consist only of the basal lamina produced by OECs that encircles ORNs from their exit of the olfactory epithelium to the outer layers of the olfactory bulb (Fraher 1999).  In the olfactory system, no clear demarcation is formed between astrocytes of the olfactory bulb and OECs, indeed, astrocytes and OECs are found apposed to each other in the inner and outer nerve fiber layers (Doucette 1991). Glial-neuron relationships between astrocytes, SCs, OECs, and their accompanying neurons also show a variety of similarities and differences.  OECs lack a basal lamina and ensheathe large bundles of axons, much like astrocytes in the central tracts. OECs also form perivascular endfeet on bloodvessels, contributing to ion homeostasis of the olfactory system in a similar manner to astrocytes in the CNS (Doucette 1984).  In contrast to these similarities between astrocytes and OECs, evidence of OEC myelination under certain circumstances suggests they are also similar to myelinating SCs and neurons of the PNS (Akiyama et al. 2004; Dombrowski et al. 2006).  Differences and similarities between SC, astrocyte, and OEC interactions with each other and with neurons suggest both challenges that must be addressed in therapeutic interventions, and mechanisms used by these cells to promote or restrict neurite growth. 1.1.8 What is the rationale for OECs as a therapeutic for spinal cord injury?  It has been argued that a number of extrinsic and intrinsic influences contribute to the regeneration failure of injured spinal axons.  The presence of inhibitory molecules derived from 28   myelin (Schwab and Thoenen 1985), expression of growth inhibitory molecules such as CSPG and heparin sulfate proteoglycans (HSPG) in reactive astrocytes (Rudge and Silver 1990), the construction of physical barriers to regeneration through the formation and subsequent walling off of cystic cavities (Kao et al. 1977), the lack of outgrowth-promoting secreted or matrix- bound factors, and an intrinsically decreased ability of adult CNS neurons to grow processes, even in the presence of outgrowth favourable cues (Kalil and Reh 1979; Weber and Stelzner 1977), all contribute to poor anatomical and physiological outcomes following SCI.  It may be suggested that more permissive intrinsic and extrinsic conditions within the olfactory neuraxis contribute to a very different outcome following olfactory injury; severing of axons on the central side of the cribriform plate does not result in prolonged gliotic scarring, nor the formation of a syrinx, but instead, upregulation of growth-promotive ECM and growth factors allow newly- born ORNs to reconnect with central targets (Getchell et al. 2005; Iwema et al. 2004; Li et al. 2005).  OECs contribute substantially to this permissive environment in the olfactory system by providing ECM and growth factors, a physical structure along which axons can grow, and by not inducing expression of growth cone collapsing molecules common to other sites of CNS injury (Doucette et al. 1983).  It is for these reasons, as well as the ability of OECs to promote outgrowth of neurons in vitro and their ability to intermingle with astrocytes, that they are a promising candidate for cell-based repair of the injured CNS.   It is hoped that transplanted OECs may modulate the inhibitory lesion site to recapitulate the permissive and developmentally-immature environment of the olfactory neuraxis. 1.1.9 What anatomical and behavioural outcomes have been achieved following OEC transplantation? The extensive histological and functional recovery initially reported as a result of OEC transplantation into the injured spinal cord has sparked intense interest in the immense therapeutic potential of OECs.  After ten years of research, however, a more sober, yet compelling, argument remains for the use of OECs to promote CNS repair. Corticospinal tract repair Corticospinal neurons reportedly exhibit substantial regeneration through OEC transplants and into the caudal spinal cord following electrolytic lesion of the dorsal 29   corticospinal tract (CST; Li et al. 1997). Additional anterograde tracing of the CST following complete transection and OEC transplantation suggested significant growth of CST axons into the distal cord only in the presence of OECs (Ramon-Cueto et al. 2000).  This was accompanied by improved locomotor behaviour (Li et al. 1997; Ramon-Cueto et al. 2000).  Additional recovery of hindlimb movement and descending inhibition of spinal reflexes after complete transection at the tenth thoracic vertebrae (T10), as a direct result of OEC or olfactory mucosal strip implantation (Lu et al. 2001) was corroborated when the recovered response was abolished following retransection of the cord (Lu et al. 2001).  OEC transplantation has also stimulated the recovery of breathing activity in the phrenic nerve (Li et al. 2003a), increased Basso Beatty and Bresnahan (BBB) scores (Lopez-Vales et al. 2006), the presence of motor cortex evoked potentials in the contralateral gastrocnemius muscle (Lopez-Vales et al. 2006), and conduction across the transected spinal cord (Imaizumi et al. 2000). Collectively, these findings suggest that OECs are either capable of promoting CST regeneration or facilitating plasticity and neuroprotection in the injured spinal cord, which in turn may lead to improved behavioural outcomes. CST regeneration and functional recovery following OEC transplantation may also be enhanced by OEC transfection with viral vectors encoding neurotrophins such as BDNF (Ruitenberg et al. 2003), NT-3 (Ruitenberg et al. 2005; Ruitenberg et al. 2003), or GDNF (Cao et al. 2004), or by treatment in combination with other therapies such as biocompatible bridging channels (Deumens et al. 2006c), or the immunosuppressant FK506 (Lopez-Vales et al. 2006). However, studies that report primarily behavioural consequences of OEC transplantation, or utilize experimental models in which fibre sparing occurs make an accurate assessment of regeneration difficult (Keyvan-Fouladi et al. 2003; Li et al. 2003a).  More recent studies, in contrast, focus on testing specific parameters of OEC transplantation and CST responsiveness, and report diminished responses of CST axons to OEC transplants compared with those previously described (Deumens et al. 2006b; Lu et al. 2006).  Although delayed OEC transplantation rostral and caudal to an OEC-filled biomatrix bridge did stimulate the extension of lesioned CST neurons up to the host/graft interface, CST axons did not penetrate or elongate for long distances through the guidance channel, and OEC transplants did not enhance performance on any behavioural task measured (Deumens et al. 2006b).  This decreased CST responsiveness may simply contrast with earlier findings because of the 4 week delay between 30   injury and OEC implantation.  Although delayed transplantation of olfactory lamina propria strips did previously enhance functional regeneration following complete chronic transection of the spinal cord, (Lu et al. 2001; Lu et al. 2002), these findings were not replicated when assessed with more extensive histological and functional tests (Steward et al. 2006). Furthermore, in some cases, OEC transplantation may not be more effective at promoting regeneration than fibroblast (Lu et al. 2002) or Schwann cell treatment (Takami et al. 2002). Finally, the physical placement and mode of transplantation of OECs into an acute or established SCI can significantly affect their distribution, and thus the range over which they may exhibit biological efficacy.  This should be considered an important variable in interpreting the results obtained by different groups (Lu et al. 2006). The paucity of data supporting OEC effectiveness in CST regeneration in a chronic setting, and the differences observed using different transplantation approaches, may be an important limitation to OEC clinical use, or may simply delineate the need for combinatorial strategies in repairing established lesions. Regeneration across the dorsal root entry zone (DREZ) following OEC transplantation The first evidence in support of the therapeutic efficacy of OECs for primary afferent regeneration was generated when OECs were injected into the dorsal root entry zone (DREZ) following transection and anastamosis of the T10 dorsal root (Ramon-Cueto and Nieto- Sampedro 1994). Using DiI labelling, and CGRP and GAP-43 immunohistochemistry, axons were shown to penetrate the DREZ, cross laminae 1,2 and 3, and enter laminae 4 and 5,  to reach the dorsal grey commisure.  Primary afferents appeared to cross the non-permissive DREZ, and engage programs for appropriate pathfinding and identification of their normal targets (Ramon- Cueto and Nieto-Sampedro 1994).  Although a limited number of axons were traced, these findings have been partially supported by three subsequent studies lesioning the lumbar (Li et al. 2004b; Navarro et al. 1999) and lumbo-sacral (Pascual et al. 2002) dorsal roots. Electrophysiological recording of peripherally-evoked activity in the dorsal horn of transected, OEC-treated animals provided some of the most compelling evidence of regeneration and functional reconnection of rhizotomized neurons following OEC transplantation (Navarro et al. 1999). In addition, bladder function was restored following peripheral implantation of OECs into an L6 to S2 rhizotomy (Pascual et al. 2002).  These experiments suggest that OECs may induce fundamental changes to the sealed, astrocytic barrier of the DREZ, following their 31   transplantation.  Astrocytes and OECs can form complex intertwining channels through the DREZ that could feasibly constitute a pathway along which their afferent axons are able to regenerate (Li et al. 2004b). Subsequent studies using genetically-labelled OECs and more complex axonal tracing techniques have challenged OEC-induced regeneration across the DREZ by suggesting that apparent long-distance afferent in-growth could be the result of spared fibers, leaking label, or growth of fibres only along physically disrupted injection tracts (Gomez et al. 2003; Ramer et al. 2004b; Riddell et al. 2004). Cholera toxin B (CTB)-labelled large diameter afferents demonstrated little growth past the DREZ after rhizotomy and OEC transplantation, although some calcitonin gene related peptide (CGRP)-positive and P2X3-positive fibres were found in the superficial dorsal horn (Gomez et al. 2003).  However, these small diameter, unmyelinated axons could represent spared sensory fibers, but are unlikely to be collaterals from other spinal levels since the survival time was short, and extensive rhizotomies of 7-9 dorsal roots from C3 to T3 were performed (Gomez et al. 2003). When peripherally-evoked cord dorsum or dorsal horn field potential recordings were used to assess afferent reconnection with spinal cord circuitry following L3-L5 rhizotomies and OB-OEC injection, there was little evidence of meaningful regeneration (Riddell et al. 2004). The sparse and stunted growth of biotin dextran-labelled fibers into the spinal cord in OEC-treated and media-injected control animals also suggests the importance of physical disruption of the DREZ in these paradigms, to facilitate the growth of peripheral neurons into the CNS (Riddell et al. 2004). In accordance with these observations is the observed propensity of CGRP- and CTB-labelled axons to project along OEC-laden injection tracts into the spinal cord, but not into adjacent spinal tissue, after LP OECs were injected through the DREZ and into the dorsal horn (Ramer et al. 2004b).  These results suggest that the partial disruption of DREZ structure may have played a greater role in earlier studies than previously appreciated.  Discrepancies in OEC transplant-mediated regeneration across the DREZ may also potentially be explained by a number of variables across these studies. First, there are primary intrinsic differences in OEC preparations. Second, OECs do not appear equally capable of promoting the growth of all dorsal root fibre neurons. While large diameter axons are less or non-responsive to OEC transplantation, small diameter afferents, mostly CGRP-positive fibres, 32   show an enhanced ability to penetrate the DREZ when OECs are present, and an increased sprouting response within the spinal cord even when the DREZ is not disrupted (Richter et al. 2005). Third, if the DREZ is not disrupted by lesion, it could be maintained in an “open” state through direct interactions between astrocytes and transplanted OECs, which could facilitate the opportunistic in-growth of OEC-responsive fibres.  Finally, a significant proportion of “regenerating fibres” may arise from exuberant sprouting of spared tracts above or below the level of the lesion.  Although OECs evidently possess an ability to promote the growth of specific afferent fibres, the mechanisms regulating this outgrowth are unknown.  How OEC- astrocyte interactions at the DREZ impart an opened, ingrowth-permissive scaffold, may be central to their regenerative properties, and must be elucidated to allow for significant, functional regeneration. 1.1.10 What mechanisms account for the ability of OECs to promote neuron growth after spinal cord injury?  Despite the plethora of studies addressing the abilities of OECs to promote regeneration following injury to the CNS, it is unclear how they do so, either in their direct actions on injured neurons, or in their interactions with endogenous glia.  Although a number of reasons remain for optimism concerning the potential for OECs as part of a combinatorial strategy to repair the injured spinal cord, an understanding of glial-glial and OEC-neuron interactions that facilitate regeneration must be gained in order to optimize long-distance axon growth, while minimizing aberrant sprouting leading to pathological pain states.  It is the goal of these investigations to address the following aims in order to elucidate those mechanisms used by OECs to promote axon regeneration in the injured spinal cord. 1.1.11 Aims and hypotheses AIM 1:  To describe the behaviour of LP and OB OECs in vivo following acute transplantation into the injured spinal cord and ascertain the effects of this transplantation on long tract axon and sensory afferent sprouting/regeneration, on endogenous glia, and on endogenous repair or injury processes.  Because LP and OB OECs contact differing glial environments in the olfactory system, represent different developmental stages, and perform differing functions in outgrowth 33   promotion and fasciculation in the olfactory neuraxis, I hypothesize that following their transplantation into the injured spinal cord: • Differing interactions between endogenous astrocytes and CNS-derived OB OECs will induce less astrocytic reactivity or glial scar formation than those resulting from LP OEC transplantation, •  There will be variation between LP OEC and OB OEC recruitment of endogenous Schwann cells, or secretion of outgrowth-promoting ECM or growth factors, • Changes in LP OEC versus OB OEC-mediated repair responses will result in different outcomes between these groups in the growth of axons into or beyond the lesion site, and • LP and OB OECs will differentially promote the growth of subpopulations of long tract spinal axons or ascending afferents.  AIM 2: To determine whether the loss of a candidate secreted factor, secreted protein acidic and rich in cysteines (SPARC), in transplanted OECs, is sufficient to alter in vivo regeneration or repair/injury processes following a spinal cord injury. I hypothesize that the absence of SPARC in transplanted OECs will: • Alter the distribution or deposition of laminin, and other ECM components as well as changing angiogenic responses to transplanted OECs, • Change the processes of cavity and glial scar formation, altering the lesion site permissivity to sprouting axons, and therefore, • sprouting of large caliber and small caliber axons will be decreased in animals transplanted with SPARC null OECs, and • Specific subpopulations of spinal axons will be differentially affected by the loss of OEC-produced SPARC.  AIM 3: To evaluate the use of the Thy1YFP-16JRS mouse in the design of an in vitro neurite outgrowth assay for corticospinal neurons that have reached their spinal targets, and characterize the antigenicity, purity, cell death, and outgrowth properties of these neurons under minimal 34   media conditions.   I hypothesize that in the absence of neurotrophins or exogenous growth factors: • Cultured corticospinal neurons from transgenic Thy1YFP-16JRS will be identifiable by YFP expression in vitro, • Corticospinal neurons in vitro will maintain expression of antigens used for their characterization in vivo, • Neuronal enrichment will require the use of antimitotic supplements, media restrictions, and enrichment strategies based on cell size or density, • The initial rate of cell death following dissociation will be high, but population dynamics will stabilize with time in culture, and • The production of primary neurites, neurite branches, and axonal or dendritic elongation will be minimal in comparison to neurite outgrowth in the presence of neurotrophins or growth factors.  AIM 4:  To use a postnatal corticospinal neurite outgrowth assay to characterize and elucidate mechanisms used by OECs to support corticospinal axon outgrowth.  Because of the plethora of known growth factors, neurotrophins, and ECM produced by OECs, I hypothesize that: • Corticospinal neurons will respond to coculture with OECs by increases in neurite outgrowth, over their neurite outgrowth on control cells or under minimal media conditions, • Dendritic or axonal branching or elongation will be differentially affected by corticospinal coculture with OECs over control cells or minimal media treatment, • Secreted or membrane-bound growth factors from OECs will differentially affect dendritic and axonal outgrowth by corticospinal neurite branching or elongation.  By addressing these aims, a clearer understanding may be reached concerning the benefits and caveats of OEC transplant-mediated therapy to treat spinal cord injury, and will provide a framework and mechanistic outlook for targeted approaches to promote axon regeneration. 35   Table 1.1 Anatomical, physiological, and behavioural outcomes of OEC transplantation into the injured spinal cord  ANATOMICAL OUTCOMES  OUTCOME MEASURES SCI PARADIGM CELL TRANSPLANT REFERENCE FIBRE SPROUTING - Biotin-labelled growing CST axons into lesion site. - Focal Electrolytic lesion of C1/C2 dorsal corticospinal tract (dCST) - Cultured, mixed population of OB OECs and olfactory nerve fibroblasts (ONFs). - Li (1997)  - Neurofilament/GAP43/CGRP/serotonergic- immunopositive or wheat-germ agglutinin-traced fibre growth into OEC tissue bridge. - T9 Completed transection  - p75-purified OB OECs with SC-filled channel.  - Ramon-Cueto (1998)  - Biotin-labelled CST axons, noradrenalin,serotonin- immunopositive fibres growing into OEC bridge and into distal cord stumps by 8 months post injury. -T8/T9 Complete transection  - p75-purified OB OECs  - Ramon-Cueto (2000)  - Positive retrogradely-traced raphe magnus, neurofilament, serotinin immunopositive fibre growth into graft - Complete T10 transection  - Cultured unpurified LP OECs or LP tissue strips  - Lu (2001)  - Biotin-traced corticospinal axons caudal to lesion site in OEC and OEC/methylprednisolone animals. - C4 Dorsal corticospinal transection  - Differential adherence purified OB OECs  - Nash (2002)  - Fluorogold retrogradely labeled nuclei were increased in the reticular formation, vestibular, raphe, and red nuclei; Increased growth of serotonergic, RT97 and GAP-43 positive fibres proximally and within the lesion site. - T9/T10 moderate contusion  - p75-immunopurified acute or 7d transplants  - Plant (2003)  - Biotin-labelled rubrospinal neurons extend into, but not beyond, grafts of BDNF or BDNF/NT-3 expressing OECs. - C4 Dorsolateral Funiculus transection - p75 Immunopurified, NT-3 or BDNF- adenovirally expressing OB OECs - Ruitenberg (2003)  - Growth of cholera toxin HRP-traced ascending fibers into lesion site; growth increased by preconditioning sciatic lesion. - T8/T9 dorsal funiculus crush  - Mixed OB OECs/ONFs or enriched OB OECs  - Andrews (2004)  - Fluororuby-labelled dorsal corticospinal neurons increase growth apposed to OEC capsules, and ventral CST branching is increased. - T8/T9 dorsal column transection  - Encapsulated or injected, purified OB and LP OECs  - Chuah (2004)  - NF immunopositive fibres extend into lesion site; increased retrogradely labeled corticospinal and rubrospinal  in OEC and GDNF-expressing OEC groups - T8 complete transection  - p75-immunopurified OB OECs expressing GDNF/control vector  - Cao (2004)  - Increased Growth of βIII-tubulin, neurofilament, serotonin, tyrosine hydroxylase, and Calcitonin gene related peptide-positive fibres into lesion site - C3/C4 crush of dorsolateral funiculus  - Purified LP OECs  - Ramer (2004)     36   Table 1.1 Anatomical, physiological, and behavioural outcomes of OEC transplantation into the injured spinal cord (Continued)   FIBRE SPROUTING CONTINUED - Biotin-traced corticospinal fibres increased caudal to lesion;  Neurofilament, GAP-43, CGRP, serotonin, and dopamine β-hydroxylase immunopositive fibres increased throughout lesion. - T8 complete transection - p75 immunopurified OB OECs - Lopez-Vales (2005) - Increased neurofilament immunopositive fibres within lesion site. - C3/C4 dorsolateral funiculus crush - Thy1.1 purified LP OECs - Richter (2005)  - Biotin-traced corticospinal fibres within lesion site and beyond in NT-3-expressing OEC group - C4 dorsal column hemisection  - p75-immunopurified, NT-3- expressing OB OECs  - Ruitenberg (2005)  - Increased density of traced corticospinal, neurofilament, and GAP43-positive fibres in lesion site - T11/T12 dorsal aspiration  - OB OEC, olfactory nerve fibroblast mixed cultures  - Deumens (2006 - Biotin-labelled corticospinal/raphespinal fibres within OEC bridge - T8 complete transection  - p75-immunopurified OB OECs transplanted chronically post lesion - Lopez-Vales (2006) TISSUE SPARING/ NEUROPROTECTION - Increased sparing of surrounding spinal tissue with acute or 7d transplant - T9/T10 moderate contusion  - p75 immunopurified OB OECs  - Plant (2003)  - Lesion volume reduction on neurotrophin-expressing OEC transplants - C4 Dorsolateral Funiculus transection - p75-immunopurified, NT-3 or BDNF adenovirally expressing OB OECs - Ruitenberg (2003)  - Increased preservation of host tissue  -T8 dorsal phototoxicity-induced lesion - p75-immunopurified OB OECs  - Verdu (2003)  - Increased sparing of spinal cord parenchyma  - T8 dorsal photochemical lesion - p75-immunopurified OB OECs  - Lopez-Vales (2004)  - Increased spared spinal tissue, decreased degenerating tissue in OEC and NT-3-expressing OEC groups - C4 dorsal column hemisection  - p75 immunopurified, NT-3- expressing OB OECs - Ruitenberg (2005)  - Reduction in TUNEL-positive neurons in motor cortex, increased corticospinal neuron survival. - T9 dorsal funiculus transection - Differential dissociation/ attachment-purified uncultured OB OECs - Sasaki (2006) DECREASED CAVITY FORMATION/ TISSUE BRIDGING - p75 immunopositive cells form continuous tissue bridge across lesion site - Focal Electrolytic lesion of C1/C2 dorsal corticospinal tract - Cultured, mixed population of OB OECs and olfactory nerve fibroblasts - Li (1997)  - Tissue bridge containing OECs - T9 Completed transection - p75-purified OB OECs with SC-filled channel. - Ramon-Cueto (1998) - Tissue bridge connecting rostral/caudal stumps, containing OECs. - T8/T9 Complete transection  - p75-purified OB OECs  - Ramon-Cueto (2000)   37   Table 1.1 Anatomical, physiological, and behavioural outcomes of OEC transplantation into the injured spinal cord (Continued)   DECREASED CAVITY FORMATION/ TISSUE BRIDGING CONTINUED - Decreased cavity formation accompanied by OEC- induced tissue bridging - Unilateral cervical electrolytic lesion to dorsal corticospinal tract - Mixed OB OEC/ONF cultures  - Keyvan-Fouladi (2003)  - Continuous bridging tissue present only with OEC transplant - T9/T10 moderate contusion - p75 immunopurified OB OECs - Plant (2003) - Decreased cavity/cyst formation - T8 photochemical lesion - p75 immunopurified OB OECs - Garcia-Alias (2004) - Decreased cavity formation; tissue bridging occurs. -C3/C4 crush of dorsolateral funiculus - Purified LP OECs  - Ramer (2004)  - Continuous OEC-bridge  - C4 dorsal column wire knife transection - Purified LP OECs  - Lu (2006)  - Integration of OECs into lesion site  - T9 dorsal funiculus transection - Differential dissociation/ attachment-purified uncultured OB OECs - Sasaki (2006) ALTERRED GLIOTIC SCAR/ GLIAL RESPONSE - Fewer hypertrophied astrocytes, decreased GFAP- immunoreactivity -T8 dorsal photochemical lesion - p75-immunopurified OB OECs  - Verdu (2003)  - Decreased CSPG immunoreactivity surrounding lesion site - T8 dorsal photochemical lesion - p75 immunopurified OB OECs  - Garcia-Alias (2004)  - Schwann cell infiltration into host CNS is increased; GFAP lesion border is diffuse with reoriented CSPGs - C3/C4 crush of dorsolateral funiculus  - Thy1.1 Purified LP OECs  - Ramer (2004) - Decreased NG2 immunoreactivity; decreased astrocytic hypertrophy - T8 complete transection  - p75-immunopurified OB OECs  - Lopez-Vales (2005)  - Increased Schwann cell infiltration  - C3/C4 dorsolateral funicuclus crush - Thy1.1 purified LP OECs  - Richter (2005)  - Schwann cell infiltration into host cord - C4 dorsal column wire knife transection. - Purified LP OECs - Lu (2006) ANGIOGENESIS - Angiogenesis is promoted within the OEC tissue bridge - C3/C4 crush of dorsolateral funiculus - Purified LP OECs  - Ramer (2004)  - Increased angiogenesis in lesion site,  increased VEGF immunoreactivity in OEC animals - T8 dorsal photochemical lesion  - p75 immunopurified OB OECs  - Lopez-Vales (2004)  - Increased directional angiogenesis toward OEC transplant sites - C3/C4 dorsolateral funiculus crush - Thy1.1 purified LP OECs  - Richter (2005)  - Increased blood vessel density - T8 dorsal photochemical lesion - immunopurified OB OECs - Lopez-Vales (2006)   38     Table 1.1 Anatomical, physiological, and behavioural outcomes of OEC transplantation into the injured spinal cord (Continued)   PHYSIOLOGICAL AND BEHAVIOURAL OUTCOMES ALTERATIONS IN ELECTROPHYSIOLOGY OUTCOME MEASURES SCI PARADIGM CELL TRANSPLANT REFERENCE - Increased conduction amplitude and velocity measured across the transection site - T11 dorsal transection  - Purified LP and OB OECs  - Imaizumi (2000)  - Partial recovery of H-reflex depression - Complete T10 transection - Cultured unpurified LP OECs or LP tissue strips - Lu (2001)  - Spontaneous phrenic nerve rhythm in transplants - Cervical hemisection - Mixed unpurified OB OECs and ONFs. - Li (2003)  - Increased amplitude of tibialis motor evoked potentials in from sensorimotor cortex stimulation -T8 dorsal phototoxicity-induced lesion - p75-immunopurified OB OECs  - Verdu (2003)  - Increased amplitude of motor and sensory evoked potentials - T8 dorsal photochemical lesion - p75-immunopurified OB OECs  - Garcia-Alias (2004)  - Increased amplitude of motor and sensory-evoked potentials - T8 dorsal photochemical lesion - p75-immunopurified OB OECs  - Lopez-Vales (2004)  - Increased activity of phrenic nerve and diaphragm  - C2/C3 hemisection - Isolectin B4 and Thy1.1 immunopurified OB OECs - Polentes (2004)  - Increased recovery of motor evoked potentials in gastrocnemius - T8 complete transection  - p75-immunopurified OB OECs  - Lopez-Vales (2005)  - Partial recovery of motor evoked potentials in gastrocnemius following motor cortex stimulation - T8 complete transection - p75-immunopurified OB OECs - Lopez-Vales (2006) RECOVERY OF MOTOR FUNCTION - Improvement of hip/knee/ankle flexion/extension. Climbing function improved, recovery of contact placing, proprioception in hindlimbs. - Complete T8/T9 transection.  - Purified, cultured p75- positive OB OECs - Ramon-Cueto (1998)  - 2 point increased BBB score in LP OEC or LP strip animals - Complete T10 transection  - Cultured unpurified LP OECs or LP tissue strips - Lu (2001)  - Success at directed forepaw reaching improved 30-40% - C4 Dorsal corticospinal transection - Differential adherence purified OB OECs - Nash (2002)  - Fewer errors on rope-walk task, and improved body posture/ hindpaw placement in neurotrophin-expressing OEC transplanted animals. - C4 Dorsolateral Funiculus transection  - p75 Immunopurified, NT-3 or BDNF- adenovirally expressing OB OECs  - Ruitenberg (2003)  - Decreased faults on climbing task. - Cervical Hemisection - Mixed unpurified OB OECs and ONFs. - Li (2003)    39     Table 1.1 Anatomical, physiological, and behavioural outcomes of OEC transplantation into the injured spinal cord (Concluded)   RECOVERY OF MOTOR FUNCTION CONTINUED - 4-6 point increase BBB scores; increased inclined plane scores - T8 complete transection -  p75-immunopurified OB OECs expressing GDNF/control vector - Cao (2004)  - Increased BBB scores - T8 Photochemical lesion to dorsal cord - p75-immunopurified OB OECs - Lopez-Vales (2004)  - Moderately increased BBB scores - T8 complete transection - p75 immunopurified OB OECs - Lopez-Vales (2005) - Increase in stride length and swing speed of OEC- treated animals. - T11/T12 dorsal aspiration  - Mixed OB OEC/ olfactory nerve fibroblast cultures - Deumens (2006)  - 6 point increase in BBB scores - T9 dorsal funiculus transection - Differential dissociation/ attachment-purified uncultured OB OECs - Sasaki (2006) ALTERATIONS IN SENSORY FUNCTION - Decreased latency to heat-induced paw withdrawal; increased amplitude of somatosensory-evoked potentials from tibial nerve stimulation -T8 dorsal photochemical  -p75 immunopurified OB OECs - Verdu (2003)  - Decreased latency to withdraw from noxious heat - T8 dorsal photochemical lesion -p75 immunopurified OB OECs - Garcia-Alias (2004)  - Decreased latency to withdraw from noxious stimulation - T8 dorsal photochemical lesion -p75 immunopurified OB OECs - Lopez-Vales (2004) 40   CHAPTER 2: MATERIALS AND METHODS 2.1 Animals All studies were performed under the approval and guidelines of the Canadian Council for Animal Care, and the Animal Care Committee of the University of British Columbia. 2.1.1 Mice used for transplantation: β-actin GFP, SPARC null, C57/Bl6 β-actin  driven EGFP transgenic mice  β-actin EGFP transgenic mice, termed FVB.Cg-Tg(ACTB-EGFP)B5Nagy/J were previously purchased from Jackson laboratory, bred onto a C57/Bl6 background, and then bred to homozygosity by a minimum of ten crosses.  The original construct used the cytomegalovirus enhancer to drive EGFP expression under the β-actin promoter (Au and Roskams 2003). SPARC null mice  SPARC null mice were a kind gift of Dr. E.H. Sage (Benaroya Research Institute at Virginia Mason, Seattle WA), and were generated through targeted disruption of the SPARC gene with stop codon in exon 4, by homologous recombination (Norose et al. 1998).   SPARC null mice were bred onto a C57/Bl6 background, and SPARC deficiency was previously confirmed using Southern blotting, Northern blotting, and immunohistochemistry (Norose et al. 1998). C57/Bl6 mice  C57/Bl6 mice, which served as WT controls for SPARC null transplantation experiments, were purchased from the Jackson laboratory (C57/Bl6/J). 2.1.2 Mice used for cell culture: CD-1, Thy1-YFP16JRS CD-1  CD-1 mice were used for the harvesting of tissue to generate LP and OB OECs, astrocytes, Schwann cells, or LP fibroblasts, except as otherwise noted. 41 Thy1-YFP16JRS homozygotes  Thy1-YFP16JRS mice were purchased from the Jackson Laboratory as heterozygotes, and are termed B6.Cg-Tg(Thy1-YFP)16JRS/J.  These mice were originally generated using the murine Thy1 vector, containing the murine Thy1.2 gene from its promoter to the intron following exon 4, but missing exon 3 and its flanking introns.  Exon 3 is required for expression in non-neuronal cells, therefore this vector, driving YFP results in YFP expression in only neuronal cells (Feng et al. 2000).  Heterozygote mice were genotyped used the protocol that follows, and were crossed with other heterozygotes to generate homozygote mice. Homozygosity was confirmed by breeding potential homozygote mice with WT CD-1 mice, and examining two of the resulting litters (13 pup minimum) for the unanimous expression of YFP in the cortex, by epifluorescence.  Homozygote mice were bred together, and their resulting pups used for corticospinal neuron culturing experiments. 2.1.3 Rats used for spinal cord injury models  Male Sprague-Dawley rats weighing 150-200g were purchased from Charles River Laboratories and were used in studies examining spinal cord injury and cell transplantation. 2.2 PCR for Thy1-YFP16JRS genotyping  Genotyping protocols were developed with the expertise and help of Nicole Janzen, and following those protocols established by Feng et al. (2000).  For genotyping, genomic DNA was extracted from tail clips using the Qiagen DNeasy kit, following manufacturer’s instructions. Genotyping reactions were run using the following primers (forward) AAGTTCATCTGCACCACCG, and (reverse) TCCTTGAAGAAGATGGTGCG,  under the following conditions: 1X PCR buffer (Invitrogen), 0.2 mM dNTPs, 3 mM MgCl2, 1 unit Taq DNA polymerase (Invitrogen) and 0.4 µM primers. The PCR cycling protocol was as follows; 94°C for 1.5 minutes, followed by 5 cycles of 94°C for 30 seconds, 60°C for 60 seconds, 72°C for 1 minute and 35 cycles of 94°C for 30 seconds, 58°C for 30 seconds, 72°C for 30 seconds and then 72°C for 5 minutes (using a PE Applied Biosystems Geneamp 9700).  Products were 172 base pairs, and were electrophoresed on a 1.2% TBE/agarose gel.  Pictures were captured using a BioRad Geldoc 1000. 42   2.3 Cell Culture 2.3.1 LP OEC cell culture GFP+ LP OECs were harvested from the olfactory mucosa of PD5 transgenic mice, homozygous for eGFP under the β-actin promoter.  The entire olfactory mucosa, including turbinates and septum was dissected from 8-10 pups, mechanically dissociated, and treated with 0.6mg/ml Collagenase D (Roche), 3 U/ml dispase I (Roche), 15μg/ml hyaluronidase(Sigma), 0.5mg/mL bovine serum albumin (ICN Biomedicals) and 100 U/ml DNAse I (Sigma)  for 30 minutes at 37°C, prior to centrifugation and plating.  Initial plating in MEM-D-Valine, 10% fetal bovine serum (FBS), and 100U/mL of penicillin/streptomycin (P/S) on tissue culture plastic flasks was followed 4-5 days later by purification using anti-Thy 1.1-mediated complement lysis, to remove contaminating fibroblasts.  For complement lysis, cells were detached from flasks using 0.25% trypsin/EDTA, washed with PBS, and trypsin proteolysis was arrested using DMEM/F12 (1:1 ratio of Delbucco’s modified Eagle’s Medium and Ham’s F12 medium; Invitrogen) and 10% FBS (DMEM/F12/10S).  Cells were incubated for 30 minutes with 400 μLThy1 antibody (from HO22-1 ATCC cell line hybridoma supernatant) and 200μL Rabbit complement (5mg/mL) in DMEM/F12/10S, per T75.  Cells were replated in DMEM/F12/10S/P/S, were allowed to grow for a further 4-6 days, when they were again subjected to Thy1.1-mediated complement lysis.  If cells were destined for coculture experiments in outgrowth assays, they were still purified in the same manner, but the serum concentration in their media was gradually decreased to 1% FBS by the second subculture.  LP OECs used for transplantation were not subjected to this media regimen, and were plated at a density of 5600 cells/cm2 into T75 flasks prior to transplantation.  The time from dissection to transplantation varied from 14-16 DIV.  Purity of cell cultures was verified using immunocytochemical procedures with p75, S100, and GFAP as described below. 2.3.2 LP fibroblast cell culture  LP fibroblast cultures were generated exactly as described above for LP OEC cultures, except that cell suspensions were originally plated directly into DMEM/F12/10S/P/S media, and not into MEM-D-Valine-contaning media, which reduces fibroblast growth.  As well, complement lysis steps were not included, yielding cultures highly enriched in fibroblasts and 43   not LP OECs, as determined by the immunocytochemical procedures described below.  LP fibroblasts used for coculture experiments were grown in gradually reduced serum conditions, in which 1% serum was used by the second passage. 2.3.3 OB OEC cell culture For transplantation experiments (Chapters 3 and 4), the same litter of PD5, eGFP transgenic mice were used for harvest of OB OECs and LP OECs.  For coculture experiments with cortical neurons (Chapters 5 and 6), OB OECs were harvested from CD-1 mice.  OB OECs were harvested by sagitally bisecting the head, removing the meningeal covering of the olfactory bulb, using a scalpel to partition the rostral-most quarter of the olfactory bulb and extracting the portion of the nerve fiber layer next to the cribriform plate.  Nerve fibre layers were dissociated and cultured as described by Ramon-Cueto et al. (1992).  Briefly, digestion in 0.1% trypsin for 15 minutes at 37oC was followed by multiple passes through fire-polished, serum-coated pipettes, and the cells were seeded into poly-L-lysine (Sigma; 50μg/mL in 15mM sodium borate buffer, 1 hour, room temperature) pre-treated flasks or tissue culture plates containing DMEM/F12, supplemented with 2mM L-glutamine, 10% fetal bovine serum (FBS), 50μg/mL gentamycin, and 100U/mL P/S.  After 5-6 DIV, when cells had grown to approximately 80% confluency, OB OECs were purified by immunopanning (Chapter 3).  Detached cells were suspended in DMEM/F12 with 10% FBS, and 500 000 cells were plated onto each 96mm petri dish, pretreated with (1) biotinylated goat-anti rabbit IgG antibody (3μg/mL, 12 hours, 4oC; Vector Laboratories, Burlingame, CA) and (2) Rabbit-anti low affinity nerve growth factor receptor (p75; Chemicon; 1:500, 1 hour, room temperature).  Cells were incubated for 30 minutes at 37oC, unbound cells were washed away, and bound cells were detached using a cell lifter.  The 30 minute immunopanning incubation was used since it produces the greatest purity of OB OEC cultures, with an average of 92% p75-positive cells (n=4).  Alternatively, treatment of cultures with cytosine arabinoside (0.01mM, AraC), a mitotic inhibitor that kills rapidly dividing cells, was applied for 3 days following initial culture, to decrease fibroblast contamination.  Subsequently, in situ Thy1.1 complement lysis was also used to purify OB OECs (Chapter 6), where cultures of OB OECs were treated with 200μL Thy1.1 antibody (HO22-1; ATCC) and 100 μL rabbit complement (5mg/mL; Sigma) in DMEM/F12/10S for 30 minutes at 37oC.  OB OECs were subcultured once more before being plated at a density of 6500 cells/cm2 44   into T75 flasks for transplantation, or before being plated for coculture experiments.  When cells were used for coculture experiments, a decreased serum concentration was used in the medium, and this was gradually reduced to 1% serum by the second subculture, resulting in serum-free conditions for outgrowth assays.  The time from dissection to transplantation or coculture plating was between 14 and 16 DIV.  Verification of cell identity and purification were performed using p75, S100, and GFAP as described below. 2.3.4 Astrocyte cell culture  Astrocyte cultures were established from PD5 CD-1 mice by decapitating pups, disinfecting their heads with ethanol, and dissecting out the entire brain by cutting through the skull with a razor blade, cutting through the skull midsagitally and on the lateral sides with microscissors, removing the skull, and scooping the entire intact brain from the skull using a dissecting spatula.  2-3 brains were placed in ice cold HBSS with P/S on ice, bisected midsaggitally, and the midbrain and olfactory bulbs were cut away using spring microscissors. The meninges were peeled away from the surface of the cortex before cortical sheets were chopped finely using a razor blade, incubated in 0.25% Trypsin and 0.03% Collagenase for 30 minutes at 37oC, and dissociated into a single-cell suspension by multiple passes through a P1000 tip.  Suspensions were passed through a 40μm filter, spun at 500g for 5 minutes, and plated into a T75 precoated for 1 hour at room temperature with 0.05mg/mL poly-L-lysine in DMEM/10S/P/S.  After 24 hours, the media was refreshed, and then changed every three to four days.  When cells reached 100% confluence, they were shaken at 100 rpm at 37oC overnight, and then shaken vigorously by hand, to detach contaminating microglia and oligodendrocytes.  AraC (20μM) was added to culture medium for four days, and was refreshed once during this time, to decrease fibroblast contamination.  Medium was changed to DMEM/10S/P/S after this time, and cells were allowed to grow, with another subculture, before they were used for experiments. Astrocytes destined for coculture experiments were subjected to serum-reduced medium, so that plating of astrocytes at the time of coculture occured in serum-free media.  Culture purity was assessed using immunocytochemistry as described below.   45   2.3.5 Schwann cell culture  Schwann cells were cultured from the sciatic nerves of five or six PD5 C57/Bl6 mice, to use in migration assays comparing migration of SPARC null and WT LP OECs.  Briefly, P5 mice were decapitated, and their sciatic nerves dissected from their insertion into the spinal cord, to their trifurcation at the knee joint, and placed in cold HBSS.  The perineurial sheath was gently removed, to decrease fibroblast contamination in culture, and nerves were digested for 15 minutes in 0.1% trypsin in HBSS.  Digestion was arrested with DMEM/10S, and nerves were further dissociated by multiple passes through serum-coated, fire-polished pipettes.  Suspensions were plated in DMEM/10S/P/S in one well of a six-well plate (9.6cm2) coated for 1 hour at room temperature with 0.05mg/mL poly-L-lysine.  The following day, AraC (0.01mM) was added to the culture for 3 days, to decrease fibroblast contamination, and cultures were then subjected to in situ Thy1.1 complement lysis as described for OB OECs.  Following this treatment, Schwann cells were cultured in DMEM/10S/P/S with the addition of 2μM forskolin and 20μg/mL bovine pituitary extract. 2.3.6 Corticospinal neuron cell culture  Four or five P8 homozygote Thy1-YFP16JRS mice were decapitated one at a time, and their brains dissected from the skull and rapidly placed in ice cold Complete HBSS, containing 1X HBSS (Invitrogen), 1% P/S (Invitrogen),  2.5mM HEPES (Invitrogen), 30mM D-Glucose (Sigma), 1mM CaCl2, 1mM MgSO4, 4mM NaHCO3 (all from Fischer).  Each brain was microdissected by stereotaxically slicing the brain along the coronal axis at 1mm caudal from the rostral-most surface of the cortex into a 1mm section, freeing the meninges from the surface of the brain, and using microforceps to free the underlying corpus callosum and brainstem structures from the cortex.  A small section of cortex measuring 2-2.5mm, and centered at 1mm lateral from the midline was clipped using microforceps, and these pieces were collected and cut into pieces measuring approximately 0.5mm each, along the axis of the cortical surface.  Cortical pieces were subjected to papain digestion (200U; Sigma)  in the presence of 2mM cysteine in dissociation medium containing 98mM Na2SO4, 30mM K2SO4, 5.8 mM MgCl2, 0.25mM CaCl2, 1mM HEPES, 20mM D-Glucose, and 0.125mM NaOH, for 2 times twenty minutes at 37oC. Digestion was arrested by removing the enzyme mixture, adding light inhibitor solution (0.3mg/mL soybean trypsin inhibitor and 0.3mg/mL bovine serum albumin in dissociation 46   medium) to the cortical pieces, followed by a 2 minute incubation of cortices in heavy inhibitory solution (3mg/mL soybean trypsin inhibitor and 3mg/mL bovine serum albumin in dissociation media).  Cortices were mechanically dissociated by multiple passes through a media-coated fire- polished Pasteur pipet, and the cell suspension was layered on a cushion of 20% Percoll in PBS, which was centrifuged at 200g for 10 minutes.  Cellular layers were removed from the Percoll gradient, and resuspended in an appropriate volume of Neurobasal A for cell counting by Trypan Blue exclusion on a haemocytometer.  Cell suspensions were centrifuged at 200g for 5 minutes, the supernatant was removed, and cells were resuspended in culture medium (1X B27, 35mM D- Glucose, 0.4mM L-glutamine, and 1% P/S in Neurobasal A), and plated at a density of  200 000 cells/cm2 onto glass coverslips precoated for 24 hours with laminin (500μg/mL) and poly-L- lysine (5mg/mL).  One day following plating, cell culture medium was refreshed, and cells were allowed to grow for a further four days. 2.3.7 Chinese hamster ovary-R2 (CHO-R2) and Chinese hamster ovary-myelin associated glycoprotein (CHO-MAG) cell culture  CHO-R2 and CHO-MAG cells were the kind gifts of Dr. Marie Filbin (Hunter College, City University of New York).  These cell lines were established by stable transfection of dhfr gene deficient CHO cells with the pSHL plasmid containing L-MAG (long-MAG) DNA in either the 5’-3’ (CHO-MAG) or the 3’-5’ (CHO-R2; control) orientation.  Both CHO cell lines were maintained in DMEM/10S, with proline (40mg/L), thymidine (0.73mg/L), and glycine (7.5mg/L), and were subcultured as necessary. 2.4 Immunocytochemistry and immunohistochemistry 2.4.1 Immunocytochemistry  To assess cell purity and to ensure cells express typical morphology and phenotype prior to transplantation or use in cocultures, LP OECs were plated onto glass coverslips following the second subculture.  They were fixed for immunocytochemistry by rinsing twice in 1X PBS for five minutes, followed by ten minute fixation in 4% paraformaldehyde (PFA), two rinses for five minutes in 1X PBS, and storage in 0.05% Azide in 1X PBS.    Purity of OEC or LP fibroblast cultures were assessed using anti-mouse S100β, anti-glial fibrillary acidic protein (GFAP), and 47   rabbit anti-p75. Astrocyte culture purity was assessed using GFAP, fibronectin, and S100 expression.  Schwann cell culture purification was assessed using rabbit anti-p75, rabbit anti- S100, and rabbit anti-fibronectin.   MAG expression in CHO-MAG, but not in CHO-R2 cells was confirmed with mouse anti-MAG antibody; plating of their plasma membranes was confirmed using mouse anti-MAG and mouse anti Na+/K+ ATPase.  Identification of corticospinal neurons was performed using a combination of mouse anti Otx1, rat anti-CTIP2 and goat anti-LMO4.  Assessment of corticospinal neurite outgrowth from YFP-positive neurons was performed using rabbit anti-NST, mouse anti-MAP2, and mouse anti-GAP43.  All primary antibodies (See Table 2.1 for dilutions, and company information) were incubated overnight in 2% normal goat serum in PBS, except in the case of rabbit anti-p75, which was incubated for one hour at room temperature.  For the nuclear antigens Otx1, CTIP2, and LMO4 only, a nuclear antigen retrieval step was included, and immunocytochemical procedures were as follows: fixed cells were washed for 2 times 5 minutes in PBS, followed by a 3 minute permeabilization in ice cold neat methanol, and blocking/permeabilization in 0.3% bovine serum albumin, 5% normal goat serum, and 0.3% Triton X-100 for 1 hour at room temperature.  For all immunocytochemical detection, slides incubated with primary antibodies were incubated for 1 hour at room temperature with Alexa 594, 488 or 350 goat secondary antibodies (1:100, Jackson Immunoresearch, Miss, ON) in 2% goat serum, followed by a 10 minute incubation in 0.5μg/mL 4’,6-Diamidine-2-phenylindole dihydrochloride (DAPI; Boehringer Mannheim, Germany) if necessary, and coverslipping in Vectashield mounting media (Vector Laboratories, Burlingame, CA). Propidium iodide labelling  Propidium iodide (PI) labelling was used to assess cell death in corticospinal neuron cultures by incubating live, unfixed cultures with PI (5μg/mL), spiked into culture wells, for 10 minutes at 37oC.  Cultures were then rinsed twice in PBS and fixed as described above for immunocytochemistry.    48   2.4.2 Immunohistochemistry Spinal cord transplantation experiments 24 hours or 28 days following SCI and transplantation, animals were sacrificed with a lethal overdose of chloral hydrate (100mg/kg, i.p.) and transcardially perfused with phosphate buffered saline followed by 4% PFA (pH 7.4).  Cervical spinal cords and brain were dissected, cryprotected with successive sucrose sinks in 12%, 18% and 24% sucrose in PBS for 24 hours each, and were frozen in Tissue-tek optimal cutting temperature compound (Sakura, Tokyo) in isopentane over dry ice.  Sections of spinal cord were generated on a cryostat at 14μm and collected and stored at –20oC.  Standard immunohistochemical procedures used were as follows: sections were thawed and adhered to the slide by warming for 10 minutes, followed by post- fixation in 4% PFA for 10 minutes, 30 minutes of permeabilization in 0.01% Triton X-100, 20 minutes block in 4% normal goat serum in PBS, and incubation of primary antibody overnight at 4oC in 2% serum.  The following primary antibodies were used in the above manner against: glial fibrillary acidic protein (GFAP) to detect the lesion site and measure astrogliosis, rabbit anti-p75 and mouse anti-S100β, to recognize OECs, mouse anti-p75, to recognize only rat- derived p75-positive cells, such as rat Schwann cells, mouse anti-rat endothelial cell antigen 1 (RECA1), to visualize blood vessels, rabbit anti-laminin, to visualize laminin deposition, rabbit anti-neurofilament heavy chain, mouse anti-neuron specific tubulin (NST), to visualize small diameter fibres, rabbit anti-tyrosine hydroxylase (TH), rabbit anti-substance P (subP), and rabbit anti-serotonin (5HT) to recognize serotonergics.  Primary antibodies for rabbit anti-calcitonin gene related peptide (CGRP) were applied for 48 hours at 4oC, whereas streptomycin-cy3 (strep- cy3), to visualize biotinylated dextran amine-traced rubrospinal fibres was applied for one hour at room temperature.  Secondary antibodies raised in goat and conjugated to Alexa 350, 488 or 594 were applied for 1 hour at room temperature, and sections were coverslipped in Vectashield (Vector). In vivo corticospinal neuron immunohistochemistry  To assess whether YFP-positive neurons from the Thy1-YFP/16JRS mouse expressed markers of corticospinal neurons, PD8 homozygote and heterozygote Thy1-YFP/16JRS mice were lethally injected with xylazine (Rompun) and ketamine (0.5mg xylazine, 5mg ketamine 49   i.p.), and transcardially perfused with 4% PFA, before their brains were dissected out and cryopreserved with successive 10% and 30% sucrose sinks in PBS for 24 hours each.  Brains were mounted coronally and sagitally and frozen in Tissue Tek optimal cutting temperature compound in isopentane on dry ice.  Cryosections were taken at 20μm intervals, and immunohistochemistry was performed exactly as described above, except in the case of nuclear antigens, where, following post fixation in 4% PFA, sections were placed in hot 0.01M Citric acid in water for 10 minutes, followed by permeabilization in 0.1% Triton X-100 for 30 minutes at room temperature, and non-specific binding block in 4% normal goat serum for 20 minutes at room temperature.  Primary antibodies against Otx1, CTIP and LMO4 were treated in this manner.  Secondary antibodies were all applied as described above. 2.5 Spinal cord injury and cell transplantation 2.5.2 Cell harvest for transplantation Prior to transplantation, the cells from 1 T75 flask of LP or OB OECs were detached using 0.25%Trypsin/1% EDTA, followed by washing in PBS, and resuspension at a concentration of 50000-60000 cells/μL in DMEM/F12.  These cells were immediately transplanted into two recipient rats, ensuring that every rat received freshly harvested cells. 2.5.3 Spinal cord injury and transplantation Spinal cord injury Following anaesthetization with xylazine/ketamine (10mg/kg i.p.; 70 mg/kg i.p.), and placement in a stereotaxic apparatus, a C3/C4 hemilaminectomy was performed, exposing the left halves of the third and fourth cervical segments.  The dura was cut with microscissors, exposing the spinal cord, and one prong of custom-designed fine surgical forceps was inserted between the grey and white matter, with the other prong outside the dorsolateral funiculus, to a depth of 1mm.  The forceps were closed for 18-20 seconds, crushing the dorsolateral funiculus. Previous studies in our laboratory confirmed that this injury reliably severs all rubrospinal axons (Ramer et al., 2004). Crush treatment was performed blind, as animals were designated to cell transplantation and control groups post-crush. 50 Cell transplantation  LP or OB OEC cell slurries in DMEM/F12 were drawn into a pulled glass pipette with a diameter of 50μm-60μm in a Hamilton syringe.  For experiments comparing LP and OB OEC repair of the injured spinal cord, cells were stereotaxically microinjected either (1) within the lesion site over three injection points, at a depth of 0.7, 1.0 and 1.5mm, dividing the cell suspension equally between these three points or (2) 1mm rostral and caudal to the lesion site at a depth of 0.6 and 1.2, dividing the suspension equally between four points.  For experiments comparing SPARC null and WT LP OEC repair of the injured spinal cord, all cells were stereotaxically injected within the lesion site, over three injection points at a depth of 0.7, 1.0 and 1.5 mm.  For all experiments, a total of 1.5μL of cell slurry was injected at a rate of 100 nL/min, so that each rat received a total of 75000-90000 cells.  Control animals received the same volume of DMEM/F12 injected at the same sites at the same rate.  The glass pipette remained in place for 5 minutes following each injection, to ensure cells remained in the spinal cord and were not withdrawn with the syringe.  Following injection, the pipette was slowly pulled back and the muscle and skin closed with interrupted sutures.  For the first set of experiments, comparing LP and OB OEC regeneration promotion following spinal cord injury, the 28 day survival groups were as follows: Direct LP OEC n=5, R/C LP OEC n=4, Direct OB OEC n=4, R/C OB OEC n=4, Direct Control n=4, R/C Control n=4, and the 24 hour survival groups were as follows: Direct LP OEC n=3, R/C LP OEC n=5, R/C OB OEC n=4, R/C Control n=5.  For experiments examining SPARC null versus WT OEC effects on spinal cord repair, the 28 day survival groups were as follows: SPARC null LP OECs n=4, WT LP OECs n=4. 2.5.4 Immunosuppression Rats were immunosupressed with cyclosporine A (CsA, Novartis Pharmaceuticals, Mississauga, ON; 10 mg/kg/d, i.p.) 2 days prior to surgery and each day for the duration of transplantation experiments.  Control and transplanted rats received the same immunosupression regimen.   51   2.6 In vitro assays and manipulations 2.6.1 Cell migration assays LP, OB OECs, or Schwann cells cultured for the same DIV, were harvested at passage 2, and plated onto Boyden chambers pretreated with PLL (50μg/mL in Borate buffer, 1 hour room temperature) or laminin (1μg/mL in PBS, 1 hour, 37oC) in 24 well plates at a density of 5000 cells/well.  The main well contained 2.5% FBS in DMEM/F12.  Cells migrated for 20 hours, the top sides of experimental transwells were swabbed, both sides were swabbed for negative controls, and neither side was swabbed for positive controls, and transwells were fixed in 0.1% glutaraldehyde in 4% PFA in PBS for 30 minutes at room temperature.  Membranes were stained with DAPI, mounted onto glass slides, coverslipped with Vectashield (Vector), and 5X images were collected, encompassing the entire membrane.  DAPI+ nuclei were counted for the entire membrane field and migration values were expressed as a percentage of total cells on positive membranes (experimental minus negative counts divided by the total cells on positive membranes). 2.6.2 Transfection of SPARC null and WT LP OECs Production of viral supernatant  To visualize transplanted WT and SPARC null LP OECs in spinal cord injury experiments, it was necessary to infect these cells with the EGFP retroviral expression vector, MGIN (gift of Dr. Paul Sorenson).  MGIN is a murine stem cell virus (MSCV)-based retroviral vector in which EGFP and neomycin phosphotransferase (neo) are encoded by a bicistronic transcript expressed from an MSCV long terminal repeat (Cheng et al. 1997).  MGIN Viral supernatant was produced by calcium chloride transfection of BOSC23 cells as follows.  BOSC 23 cells were plated at a density of 50 000 cells/cm2 in DMEM/10S two days prior to transfection, yielding cells of 85-90% confluency by the morning of transfection.  The original media was replaced with DMEM/10S containing chloroquine (25mM).  200 μL of 2X Hepes Buffered Saline (HEPES 50mM, KCl 10mM, dextrose 12mM, NaCl 280mM, Na2HPO4-7H2O 1.5mM) was added dropwise and vortexed between additions to a suspension containing 1μg of MGIN plasmid, 1μg env plasmid, 1μg GP1 plasmid, 175μL water, and 25μL 2M CaCl2. 52   40μL/cm2 of the transfection solution was added to BOSC23 cells, and they were incubated for 6 hours before media was changed to DMEM/10S.  Viral supernatant was harvested 12 hours later, concentrated by centrifugation using an Amicon ultra centrifugal device with low protein binding (Millipore), according to the manufacturer’s instructions, and stored at -80oC until used. Infection and flow cytometric sorting of LP OECs  LP OECs were infected with the MGIN retrovirus after their first subculture by the addition of 2 mL of viral supernatant per T75 every 12 hours, with fresh viral supernatant added 3 times.  Infected LP OECs were enriched for EGFP expression by sorting cells using FACS on a FACS Vantage Diva, and cells were replated prior to use for transplantation.  Maintenance of EGFP expression was determined by plating of cells on coverslips and assessment of endogenous eGFP expression.  Cells were also checked for normal growth parameters, cell morphology, and antigenicity for p75, S100β, and GFAP. 2.6.3 Neurotrophin addition to corticospinal neuron culture  To assess the sensitivity of the corticospinal neurite outgrowth assay, and whether cultured YFP neurons would exhibit expected responses to known outgrowth regulators, the neurotrophins ciliary neurotrophic factor (25ng/mL rat recombinant CNTF, Sigma) and neurotrophin-3 (25ng/mL human recombinant NT-3, Cedarlane laboratories), or their combination, were added to media prior to plating of CST neurons. 2.6.4 Mitotic inhibitors Mitotic inhibitors for purification of corticospinal and other cell cultures To purify Schwann cell cultures and OB OEC cultures, and decrease contamination of these cultures with rapidly-dividing fibroblasts, cytosine β-D-arabinofuranoside (0.01mM AraC, Sigma), a selective inhibitor of DNA synthesis, was added to cultures for three days.  To decrease contamination of corticospinal neuron cultures with glia, including astrocytes, oligodendrocytes, and microglia, the contributions of several mitotic inhibitors were assessed, including addition of 0.01mM AraC during the entirety of the culture period.  As well, addition of 2 or 10 mM camptothecin (Sigma), which binds to DNA-topoisomerase I complexes, trapping them in an intermediate unwinding conformation, and increasing DNA cleavage, was also added 53   to CST cultures for the entirety of their culture period.  Finally, 5-Fluoro-2’-deoxyuridine and uridine (10μM FdU, Sigma), an antineoplasmic agent that inhibits DNA synthesis by blocking thymidilic acid synthetase, and was added to CST cultures for the entirety of their culture period.  MitomycinC treatment and plating of cells for corticospinal neuron coculture  To inhibit cell division in glia and fibroblasts used for CST coculturing experiments so that differences in the intrinsic division rates of these cells would not result in different numbers of cocultured cells over time, just prior to harvesting of cells, they were treated for 2 hours at 37oC with Mitomycin C (5μg/mL MMC, Sigma).  Mitomycin C inhibits DNA synthesis by cross-linking complementary strands of DNA.  Concentrations and time of exposure to MMC were previously determined in our lab. 2.6.6 Generation, harvesting and addition of OB OEC conditioned media  OB OECs were cultured as described above, with a stepwise reduction in serum concentration in the media, to yield cells plated in 1% FBS at the second subculture.  To generate OB OEC conditioned media, OECs plated at a density of 25 000 cells/cm2 were changed to media composed of Neurobasal A, 1X B27, 35mM D-Glucose, and 0.4mM L-glutamine and were grown for a further five days.  Conditioned media was harvested and collected into 15 mL conical tubes, and spun at 210g for five minutes, to pellet cellular debris.  The supernatant was removed from the cellular fraction, concentrated using an Amicon low protein binding centrifugal filter (Millipore), which concentrates proteins greater than 1 kDa, and was stored at - 80oC.  The volume of concentrated media was recorded to determine the fold concentration of each media batch.  Cells producing the conditioned media were also subsequently trypsinized and counted so that batches of conditioned media could be standardized based upon the number of cells producing each mL of media, and the concentration factor.  Standardized full strength conditioned media was arbitrarily based upon 30000 cells/mL of unconcentrated OEC conditioned media.  For example, conditioned media that was generated by 100000 cells would be standardized by (1) dividing by 30000, and then (2) multiplying by the concentration factor e.g. 10X, to yield the fold by which the media would be diluted to yield 1X concentration, in this case 100000/30000 x 10 = 33, therefore 1X media would be diluted 1/33.  To assess whether this conditioned media retained bioactivity and different batches retained similar activity, DRG 54   explants assays were performed, and OEC conditioned media added to these, since our lab has previously shown that OEC conditioned media increases neurite outgrowth from these explants. CST-stimulated OB OEC conditioned media  To assess whether OB OEC secreted factor complement or concentration was altered by contact with CST neurons, OB OEC axon-stimulated conditioned media was generated.  OB OECs were treated exactly as those used above for the generation of regular OB OEC conditioned media, except that one day following plating at a density of 25 000 cells/cm2, CST neurons (150 000 cells/cm2) were plated on top of the OB OECs, with new media containing Neurobasal A, 1X B27, 35mM D-Glucose, and 0.4mM L-glutamine, and cells were grown for a further five days.  Harvesting, concentration, and standardization were exactly as described above for regular OEC conditioned media. 2.6.7 Transwell coating and cell plating  To assess the contributions of secreted factors to CST neurite outgrowth, separation of OB OECs or LP OECs/LP fibroblasts was performed using 6.5mm 8.0 μm transwells (Costar). Prior to use, transwell polycarbonate, porous membranes were coated with laminin (500μg/mL) and poly-L-lysine (5mg/mL) for 24 hours at 37oC.  OB OECs, LP OECs or fibroblasts in serum- reduced conditions, and treated with MMC as described, were detached from plates with 0.25% trypsin, and 50000 cells are plated into each transwell (the same number of cells used for coculture) in culture medium containing Neurobasal A, 1X B27, 35mM D-Glucose, and 0.4mM L-glutamine.  Cells were allowed to adhere to transwells overnight before transwells were added to CST cultures. 2.6.8 Plasma membrane enrichment  CHO-MAG, CHO-R2, and OB OEC membranes were prepared by rinsing cells with PBS, and detaching cells from flasks using a cell lifter (Fischer Scientific).  Trypsin was not used to avoid any digestion of membrane proteins.  Cells were centrifuged at 210g for 5 minutes, and excess PBS was removed from the cell pellet before adding 500mL of homogenization buffer per 75cm2 ( 10mM Tris HCL, 1.5mM CaCl2) with the protease inhibitors spermidine (1mM), aprotinin (25μg/mL), and 15μg/mL 2,3 dehydro-2-desoxy-N-acetylneuraminic acid.  Cells were 55   subjected to mechanical disruption by several passes in a mortar and pestle homogenizer and then sat in homogenization buffer for 20 minutes on ice.  Sucrose step gradients with a lower phase containing 500μL of 50% sucrose, and an upper phase containing 150 μL of 5% sucrose were made in Beckman ultracentrifuge tubes, and homogenized cells were layered over the step gradient and centrifuged at 60000rpm for 10 minutes at 4oC.  The membrane fraction was at the phase interface, and was gently removed with a pipet, before rinsing the gathered membranes in Ca2+/Mg2+-free HBSS, spinning down samples at 20000rpm, and removing supernatant.  Samples were stored at -80oC.  Plasma membrane enrichment was determined by Western blotting for Na+/K+ ATPase and GAPDH in whole cell and plasma membrane enriched samples. 2.6.9 Plasma membrane optical density assessment and substrate plating  To determine the concentration of plasma membrane samples for standardization, a 15- fold dilution of a small aliquot of sample was made in 2% sodium dodecyl sulphate (SDS), a detergent.  Optical density of these samples was assessed by absorbance at 220nm on a spectrophotometer.  The volume of the original sample was adjusted to yield an optical density of 2.0 upon dilution.  Uniform substrates of plasma membrane were prepared by coating glass coverslips with laminin (500μg/mL) and poly-L-lysine (5mg/mL) for 24 hours at 37oC.  Coverslips were removed from substrate plating wells, and were placed, top side down, on a 15μL droplet of plasma membrane, adjusted to a desired optical density, and placed on a sterilized sheet of Parafilm (Pechiney Plastics) overnight.  Just prior to culture, Parafilm sheets were carfully flooded with Neurobasal A, to detach coverslips, without destroying plated plasma membranes. Coating of coverslips with plasma membrane was verified using immunocytochemical detection of Na+/K+ ATPase, MAG, or p75, as described for cells in culture, but without permeabilization steps. 2.7 SDS-PAGE and Western blotting 2.7.1 Sample quantification and preparation Proteins were extracted from whole cell or plasma membrane fractions of CHO-MAG, CHO-R2, OB OEC, or astrocyte  samples by the addition of 30μL of lysis buffer (50mM Tric- 56   HCl, 150mM NaCl, 1% NP-40 (Igepal), 5mM EDTA) plus the protease inhibitors aprotinin (10μg/mL), leupeptin (10μg/mL), and PMSF (40μg/mL), per 5 x 105 cells, and trituration.  This was followed by centrifugation for 1 minute at 14000 rpm, 4oC, and quantification of protein content using the Bradford-Lowry method.  This method uses sample absorbance of bovine serum albumin (BSA) standards, measured by a spectrophotometer at 595 nm, in the presence of Bradford reagent, to construct a curve of protein concentration versus absorbance.  This curve can then be used to assess the protein content of unknown samples. 2.7.2 SDS-PAGE and Western blotting  Equal protein samples from plasma membrane extracted and whole cell extracts were diluted 1:1 in 2X SDS sample buffer and denatured by heating at 70oC for 20 minutes.  Protein samples were separated by 7.5% Tris-HCl polyacrylamide gel electrophoresis, and were transferred to nitrocellulose membranes (Biorad trans blot).  Transfer was verified using PonceauS staining (0.1% PonceauS, 5% acetic acid in ddH20, 10 minutes), and membranes were blocked for 1 hour at room temperature in agitating 5% non-fat milk in1X tris-buffered saline (TBS).  Primary antibodies against glyceraldehydes-3-phosphate dehydrogenase (GAPDH, 1:10000, Hytest), Na+/K+ ATPase (1:1000, Upstate), or p75 (1:1000, Chemicon) were incubated with the membrane in 2% milk/TBS, overnight at 4oC, washed for 3 times 5 minutes in 0.05% Tween/TBS, and incubated with secondary antibodies at room temperature for one hour. Secondary antibodies were conjugated to horseradish peroxidise (HRP) and were either goat anti-mouse HRP (1:5000), or goat anti-rabbit HRP (1:10000) in 2% milk/TBS. Following three washes of five minutes in 0.05% Tween/TBS, the membranes were incubated with ECL chemiluminescence substrate (Amersham), and signals were visualized on X-ray film. Reprobing of the same membrane for other antigens was accomplished by stripping the membrane with 0.1M NaOH for 5 minutes at room temperature, and washing the membrane for two times five minutes in ddH20.  Blocking, primary, and secondary antibody incubations were then performed exactly as described above.   57   2.8 Image analysis, quantification and statististics 2.8.1 Spinal cord injury and transplantation experiments Z-stacked 200x or 100x, montaged digital images were captured with an Axioplan 2 microscope (Zeiss, Jena, Germany), a Retiga digital camera (QImaging, Burnaby, Canada) and Northern Eclipse software (Empix Imaging Inc., Mississauga, ON), and were processed using Northern Eclipse, Photoshop CS3, Image J and SigmaScan Pro (SPSS Inc., Chicago, IL) software. Assessment of lesion size, cavity size, astrogliosis, immune cell recruitment to the lesion site, OEC migration in vivo, and Schwann cell Infiltration.  Lesion area and cavity size were quantified for each animal by outlining the lesion area defined by GFAP at three defined levels, 0.65mm, 0.9mm and 1.15mm ventral in the spinal cord and calculating the total pixels defining this area using Image J.  GFAP immunodensitometry was performed on sections at 0.9mm ventral for all animals in each treatment group and was quantified for a defined area size around the lesion site as compared with the analogous area on the contralateral uninjured side of the spinal cord.  GFAP immunoreactivity of the contralateral side of the spinal cord controlled for differences in the immunoreactivity of different tissue samples.  Quantification of immune cell recruitment to the lesion site was performed by outlining the GFAP immunoreactive lesion site border and counting the number of ED1, DAPI- double positive nuclei located within this area, using Image J.  To quantify GFP-positive OECs at locations within the lesion site, or 0.5mm/1mm rostral or caudal to the lesion site, items fluorescing in all three channels were recolored to black.  Selection of the GFAP-positive lesion border was then used to define a region where GFP-positive pixels above a threshold bin were counted.  Selection areas measuring 750 000 pixels were used at sites 0.5mm and 1mm rostral and caudal to the lesion site to define GFP-positive pixels at these locations.  Verification that the distribution of GFP-positive OECs was due to cell migration, and not pressure-dependent cell dispersion was performed by sacrificing a group of animals transplanted rostral-caudally with GFP-positive OECs 1 day post injury. Schwann cell infiltration into the lesion site and surrounding areas was performed exactly as described above for GFP-positive OECs, except that Schwann cells were visualized with anti-rat p75. 58 Assessment of directional angiogenesis Analysis of angiogenesis and directionality were performed using programs designed on Matlab (Mathworks).  Briefly, 100x montaged images stained with anti-RECA were automatically posterized to two levels, blood vessels above a threshold size were recursively mapped, counted, and a best fit line was applied.  Intersection points from lines were mapped onto the same XY plane as the original image, and the number of intersection points in the lesion site, as defined by an outline of the GFAP scar, was assessed.  The scar area was then moved randomly and automatically around the image 100 times to ascertain a baseline value of directionality.  The ratio of directionality in the lesion versus random points was reported. Autotomy assessment  Autotomy is the self-attack of a denervated limb. Autotomy was scored using the methods described by Wall (1979), in which a total of 3 points were awarded for the attack of each digit; 1 point for nail removal, 2 points for amputation up to the first knuckle, and 3 points for total digit removal, for a total of 15 possible points for each paw (Figure 7A).  For each rat, the affected right hindpaw was scored as well as the left hindpaw, which serves as an internal control.  Animals were sacrificed if autotomy was excessive, although this rarely occurred. Assessment of sprouting/regeneration of supraspinal and sensory fibre tracts All axon quantitation was performed at the three depths in the spinal cord defined at 0.65mm, 0.9mm, and 1.15mm ventral.  Removal of autofluorescent items was performed before quantification, by recoloring triple-fluorescent items in black.  Quantitation within the lesion site was performed by outlining the GFAP-reactive border, binning fluorescence from axon staining into 4 bins (to normalize staining between animals), and counting the number of pixels in the top 3 bins.  For areas on either side of the lesion site, defined selection areas measuring 750 000 pixels at 0.5mm and 1mm rostral/caudal to the lesion boundary were quantified as described for within the lesion site, by binning and counting pixels in the three top bins.   59   2.8.2 Corticospinal outgrowth assay Image aquisition To characterize the cell populations and their dynamics (including death) over time in the corticospinal neurite outgrowth assay, images were collected from coverslips at 200X magnification, at 10 random sites over a coverslip. For quantification of neurite outgrowth, ten random and non-overlapping images were captured of YFP-positive neurons at 400X magnification, leading to data acquisition for 25 to thirty neurons per coverslip.  At least three replicate coverslips were quantified in this manner per experiment, and three independent replicate experiments were performed per condition. Images were processed using Northern Eclipse, Photoshop CS3, and Image J. Counting and quantification of cell death (PI) and other corticospinal neuron markers  Triple-fluorescent 200X images from 10 random and non-overlapping regions of a coverslip were taken of PI, DAPI and YFP with the same photographic patameters.  Images were thresholded in ImageJ, and counted manually in Image J for all three channels alone, followed by double-flourescing and triple-fluorescing cells.  Counts of NST-positive, YFP-positive and DAPI-positive cell proportions were performed in the same manner, but on coverslips that were fixed after 1, 3 or 5 DIV.  Quantificantion of proportions of YFP-expressing neurons that also express CST neuron markers were performed in exactly the same manner, but using combinations of antigens against CTIP2, Otx1, LMO4, YFP and DAPI.  For all combinations, at least three coverslips were sampled per experiment, with at least three independent replicate experiments. Montaging and quantification of neurite length, branching, and individual neurite length  Triple fluorescent images for YFP-expressing neurons immunoreacted with NST and either MAP2 or GAP43 were generated using the same photographic parameters.  Using Neurobinary, each channel was loaded, thresholded to a common level depending on the antigen, and made into a binary image.  The region of interest, either the entire neurite carpet of many YFP-positive neurons, or individual neurites of a single neuron, was manually inputted, and regions of non-interest were removed.  Neurobinary then automatically recursively mapped 60   neurites, and reduced them to a single pixel width, producing a continuous skeleton of neurite outgrowth.  Pixels were added together for an entire neuron’s neurite carpet, or for each individual neurite, and converted to micron values based upon the magnification of the original image.  Primary, secondary, or tertiary branch points were counted manually using ImageJ.  Data was recorded and processed using Excel 2007. 2.8.3 Statistical analysis T-tests were used to compare means between two groups.  For measurements of multiple comparisons, ANOVA was applied, followed by post hoc analysis using the Tukey method of multiple comparisons. The Tukey procedure was chosen because it does not require equal sample sizes, and it reduces the probability of a type I error.  Graphical data is presented with error bars representing standard error of the mean. 61   Table 2.1 Antibodies used for immunocytochemical assessments Antigen Supplier Dilution Purpose Rabbit anti-p75 Chemicon 1:1000 LP, OB OEC, LP fibroblast culture purity Mouse anti-S100β Sigma 1:1000 LP, OB OEC, LP fibroblast culture purity Rabbit anti-GFAP Dako Cytomation 1:750 LP, OB OEC, LP fibroblast, astrocyte culture purity Rabbit anti-S100 Neomarkers 1:1000 Astrocyte, Schwann cell culture purity Rabbit anti- fibronectin Dako Cytomation 1:500 Astrocyte, LP fibroblast, Schwann cell culture purity Mouse anti-MAG Boehringer Mannheim 1:200 CHO-MAG and CHO-R2 MAG expression, plasma membrane plating of CHO- MAG PM Mouse anti-Na+/K+ ATPase Upstate 1:250 Cofirmation of plasma membrane plating Mouse anti-Otx1 Developmental Studies Hybridoma Bank 1:100 Identification of deep layer cortical (corticospinal) neurons Rat anti-CTIP2 Abcam 1:500 Identification of deep layer cortical (corticospinal) neurons Goat anti-LMO4 Santa-Cruz 1:50 Identification of deep layer cortical (corticospinal) neurons Rabbit anti-NST Covance 1:500 Quantification/characterization of neurite outgrowth Mouse anti-MAP2 Medicorp 1:1000 Quantification/characterization of neurite outgrowth Mouse anti-GAP43 Chemicon 1:500 Quantification/characterization of neurite outgrowth  62   Table 2.2 Antibodies used for immunohistochemical detection Antibody Supplier Dilution Purpose Rabbit anti-GFAP Dako Cytomation 1:500 Lesion site area, cavity formation, astrogliosis Rabbit anti-p75 Chemicon 1:1000 Confirm LP, OB OEC identity Mouse anti S100β Sigma 1:1000 Confirm LP, OB OEC identity Mouse anti-rat p75 Chemicon 1:50 Visualize rat Schwann cells Mouse anti-rat endothelial cell antigen 1 Serotec 1:50 Visualize blood vessels Rabbit anti-laminin Sigma 1:1000 Recognize laminin deposition Mouse anti-rat CD68 (ED1) Serotec 1:500 Recognize immune infiltrates Rabbit anti- neurofilament heavy chain Serotec 1:500 Recognize large-diameter fibres Mouse anti-neuron specific tubulin Covance 1:500 Recognize small diameter fibres Rabbit anti-tyrosine hydroxylase (TH) Chemicon 1:250 Recognize TH+, Noradrengeric or adrenergic fibres Rabbit anti-substance P Chemicon 1:1000 Recognize subP+ pain fibres Rabbit anti-serotonin (5HT) Immunostar 1:8000 Recognizes serotonergic fibres Rabbit anti-calcitonin gene related peptide (CGRP) Sigma 1:200 Recignizes sensory fibres positive for CGRP Streptomycin-cy3 Jackson Immunolabs 1:250 Recognizes Biotinylated dextrane amine (BDA)- traced fibres  63   CHAPTER 3: LAMINA PROPRIA AND OLFACTORY BULB ENSHEATHING CELLS EXHIBIT DIFFERENTIAL INTEGRATION, MIGRATION, AND PROMOTE DIFFERENTIAL AXON SPROUTING IN THE LESIONED SPINAL CORD.  Note: Most of this chapter has been previously published as Richter MW, Fletcher PA, Liu J, Tetzlaff W, Roskams AJ (2005) J. Neurosci 25(46): 10700-11, excepting the following additions and modifications: The document has been formatted to incorporate the stylistic guidelines of this thesis, revisions have been made to the introductory comments as well as to the discussion to more accurately reflect the current literature, and the methodological section has been removed and included in Chapter 2.  3.1 Introduction The abortive regeneration of CNS neurons is in stark contrast to the renewal capacity of the olfactory system (Graziadei et al. 1978; Graziadei et al. 1979).  Reconstitution of the olfactory neuraxis can occur even after ORN axons are severed distal to their cell bodies, within the CNS (Zigova et al. 1992).  Following axotomy or olfactory bulbectomy, a spatiotemporal pattern of ORN cell death occurs from the central lesion site and progressing rostrally towards the ORN soma, and ORNs eventually undergo caspase-3-mediated apoptosis (Cowan et al. 2001).  In contrast to the terminal loss of neurons that occurs in many areas of the nervous system following injury (Fitzgerald and Fawcett 2007), dying ORNs release neurogenic factors that induce the differentiation of immature neurons, which are, surprisingly, capable of growing through olfactory ensheathing cell conduits in the olfactory lamina propria, and forming synaptic contacts with their CNS targets (Li et al. 2005; Raisman and Li 2007).  Although the reconstitutive capacity of the olfactory neuraxis is diminished with age, replacement of old ORNs in the normal turnover of the neuraxis, or the ability to grow and retarget axons following 64   injury, continues throughout the organism’s life (Ducray et al. 2002; Weiler and Farbman 1997; Weiler and Farbman 1998).  This axonal growth within the CNS has been partially attributed to olfactory ensheathing cells (OECs), resident in the peripheral (lamina propria OECs; LP OECs) and central (olfactory bulb OECs; OB OECs) olfactory neuraxis (Doucette 1990).  Because OECs in the lamina propria and olfactory bulb provide a physical scaffold for growing axons, secrete axon outgrowth-promoting extracellular matrix and growth factors, express cell surface- bound growth factors, and do not induce astrocytic hypertrophy, it has been postulated that their transplantation would recapitulate the permissive environment of the olfactory system elsewhere within the lesioned CNS (Crandall et al. 2000; Lakatos et al. 2000; Mackay-Sim and Chuah 2000). For this reason, the therapeutic potential of OB-OECs to promote recovery from CNS lesions to descending and ascending spinal tracts (Keyvan-Fouladi et al. 2003; Li et al. 1998; Lu et al. 2002; Nash et al. 2002; Ramon-Cueto et al. 2000; Ramon-Cueto et al. 1998), primary afferents (Li et al. 2003b; Ramon-Cueto and Nieto-Sampedro 1994) and remyelination of injured/demyelinated spinal cords (Franklin et al. 1996; Imaizumi et al. 1998 ; Sasaki et al. 2004) have been tested, and yielded variable yet positive outcomes.  Early investigations of corticospinal (CST) regeneration following OB OEC implantation rostral and caudal to a complete transection of the rat spinal cord at thoracic level 8 (T8) suggested significant axon regrowth and targeting had occurred by 8 months post injury (Ramon-Cueto et al. 1998).  While untreated rats were unable to demonstrate voluntary hindlimb movement in a climbing task, OEC-transplanted rats voluntarily stepped and placed paws on ladder rungs (Ramon-Cueto et al. 2000).  Furthermore, anterograde biotinylated dextran-amine (BDA) tracing of the corticospinal tract revealed traced fibres rostral to the lesion site, as well as untraced serotonergic (raphespinal) and noradrenergic (ceruleospinal) axons that had apparently penetrated the rostral and caudal lesion boundaries and entered the uninjured cord (Ramon-Cueto et al. 2000). Similarly impressive results following OEC transplantation have been reported following electrolytic lesion of the dorsal CST or transection of the T10 dorsal root (Li et al. 1998; Ramon- Cueto and Nieto-Sampedro 1994).  It remains unclear whether the “regenerating” axons of these studies truly represent cut axons that have extended processes through a lesion site; the incomplete nature of some of these injury paradigms makes it probable that many axons were 65   either spared, or that functional recovery is the result of circuit rearrangement in the distal spinal cord. Only recently have lamina-propria-derived OECs (LP OEC) been tested for their ability to promote regeneration following lesion (Lu et al. 2001; Lu et al. 2002; Ramer et al. 2004a; Ramer et al. 2004b), despite the ongoing clinical use of LP tissue of unknown composition in human spinal cord injury (SCI) treatment in China (Huang et al. 2003; Senior 2002).  A significant advantage of LP OEC transplantation is the relative ease of harvesting cells for primary culture, in comparison with the intracranial surgery necessary for OB OEC collection. However,  because repair of the injured spinal cord following LP or OB OEC transplantation has never been assessed in the same injury model, and little is understood concerning the mechanisms used by either cell to promote axon growth following SCI, it is unclear (1) whether differences between experimental outcomes reveal intrinsic differences in LP versus OB OEC cell function, in their interactions with populations of spinal neurons, or glia, or whether (2) LP or OB OECs  may be used interchangeably for the treatment of spinal cord injury. 3.1.1 LP and OB OECs perform divergent roles in their differing environments within the olfactory neuraxis LP OECs are derived from a multipotent progenitor in the olfactory mucosa, the horizontal basal cell (Carter et al. 2004), and they ensheath and promote the growth of ORN axons as they exit the olfactory epithelium, and extend through the LP towards the OB (Williams et al. 2004a). Most evidence suggests that when LP OECs enter the nerve fibre layer, they differentiate into OB OECs and no longer ensheathe ORN axon bundles but travel in parallel with axons as they defasciculate and eventually innervate their targets (Au et al. 2002).  Some OB OECs may also arise from local progenitors that originally formed the migratory mass and presumptive NFL during embryonic development of the olfactory system (Puche and Baker 2007).  Differences in either maturational state or derivation of LP and OB OECs are mirrored by their divergent interactions with ORNs.  LP OECs form tight and complex channel arrays through which multiple ORN axons pass; LP OECs provide a living ECM and growth factor-rich scaffold for growing ORNs (Li et al. 2004b).  OB OECs no longer ensheathe ORN bundles, but form a substrate over which ORNs grow, and may also provide instructional cues for ORN axon pathfinding (Crandall et al. 2000; Tucker and Tolbert 2003).  More recently, differences in 66   expression of p75, O4 and polysialated NCAM have been demonstrated in cultures of LP versus OB OECs (Kumar et al. 2005).  Thus, although LP and OB OECs possess similarities in morphology in vitro (Au and Roskams 2003; Jani and Raisman 2004), subtle differences in ensheathing behavior, their interactions with ORN axons, their different maturational states, and their expression of cell adhesion molecules suggest they may exhibit different properties when transplanted following SCI. 3.1.2 Rationale and aims: regeneration outcomes may differ following LP or OB OEC transplantation Despite the initial report of exuberant growth of sensory neurons into laminae 4/5 of the dorsal horn and to the dorsal grey commissure following T10 dorsal root transection and OB OEC transplantation, recent experiments of LP or OB OEC-induced dorsal root repair suggest less exuberant outcomes following these treatments (Ramer et al. 2004b; Ramon-Cueto and Nieto-Sampedro 1994; Riddell et al. 2004).  In these cases, the type of OEC transplanted, mode of transplantation, and proximity of transplanted cells to the lesion site could influence integration and long-term survival of transplanted cells. To assess how differences in regeneration may arise as the result of (1) transplantation of OECs from different sources, (2) the method and timing of cell introduction and (3) differences in assessment measures, it has become important to compare the regeneration promoting abilities of LP and OB OECs within the same experiment. To determine the contributions of these parameters to fibre sprouting and repair outcomes of OEC transplantation, the following aims were addressed: • To purify transgenic eGFP-expressing OECs from the lamina propria and from the nerve fiber layer of the same mice, • After establishing LP and OB OEC expansion and phenotype in vitro, to acutely transplant them directly or rostrocaudally into a C3/4 crush of the dorsolateral funiculus. • To determine differences in the abilities of LP and OB OECs to interact with host glia, promote directional angiogenesis, and to migrate within the spinal cord. 67   • And to assess the regenerative and sprouting differences of spinal neuron populations between LP and OB OEC-treated animals. 3.2 Results 3.2.1 LP and OB OECs in vitro To assess their comparative ability to promote regeneration, OECs were purified from the lamina propria (LP-OECs) or olfactory bulb (OB-OECs) of PD5 mice homozygous for eGFP under the β-actin promoter (Au and Roskams 2003; Ramon-Cueto and Nieto-Sampedro 1994). LP (Figure 3.1A,C) and OB (Figure 3.1B,D) OECs grown for 14 DIV demonstrated similar morphologic and antigenic profiles for S100β, GFAP and p75.  LP and OB OECs exhibited both flattened and fusiform morphologies (Figure 3.1A,B). OB OECs demonstrated a broader range of p75 expression than LP OECs; some OB OECs were highly p75 positive whereas the majority demonstrated low p75 levels (Figure 3.1B). Despite their antigenic similarity, LP-OECs demonstrated a 3-fold higher expansion rate than OB OECs from passage 1 to 2 under the same culture conditions. 3.2.2 LP and OB OEC migration in vitro and within the spinal cord Purified passage 2 eGFP+ LP and OB OECs cultured for 14 DIV were transplanted either directly into or at sites 1 mm rostral and caudal to a crush of the dorsolateral funiculus (Figure 3.2).  I first addressed differences between LP and OB OECs that might be predicted based on their home environments in the PNS and CNS compartments of the olfactory system, such as their ability to migrate within the CNS, and their interactions with endogenous glia.  I tested if rostro-caudally (R/C) transplanted LP and OB OECs displayed different patterns of migration within the CNS, by assessing their change in distribution from 24 hours to 28 days after lesion (Figure 3.3).  24 hours after injection, 25 ±7.7% of all transplanted LP OECs and 30± 6.9% of OB OECs were already found at the lesion site, and both types of OEC were similarly arranged in a stream from the rostral injection point to lesion centre. As I could not detect cells between the caudal injection point (which retained 42 ±10.6% of total LP-OECs and 28 ±6.2% OB OECs transplanted) and lesion site (in 4/5 LP and 3/4 OB transplanted rats), this suggests that fluid pressure may initially displace a subset of OECs from their rostral but not their caudal injection 68   69   70   71   point. Migration of GFP+ OECs from the caudal injection point to the lesion site was visible by 28 days in LP and OB transplanted rats with surviving GFP+ OECs (3/4 LP; 4/4 OB) (Figure 3.3A,B). However, the frequency and distribution of OECs between the injection and the lesion sites differed significantly between OEC types (n=3,4; p≤0.05). Rostrocaudally transplanted rats had more GFP+ OECs at 28 days than directly transplanted rats (n=0/5 LP OEC; 1/4 OB OEC). By quantifying GFP+ OEC distribution, approximately 45% of surviving GFP+ OB-OECs in rostro-caudally transplanted rats were still found between the caudal injection point and lesion site (Figure 3.3D, n=4), similar to the distribution of LP OECs 7 days following injury and rostro-caudal transplantation (Figure 3.3C).  In contrast, at 28 days, only approximately 6% of surviving LP OECs remained near the caudal injection point (Figure 3.3A), with 51% of surviving GFP+ cells found within the lesion site.  To further address if migratory differences in vivo represented intrinsic differences in the migratory abilities of LP and OB OECs, a Boyden chamber assay was used.  Cells plated onto a porous membrane coated with poly-L-lysine (normal OEC substrate) or laminin (rich at the lesion site; Ramer et al. 2004a) were allowed to passively migrate through the membrane, and the ratios of migratory versus non-migratory cells were assessed.  At 20 hours, a significantly greater number of LP OECs migrated than OB OECs on PLL (n=3, p≤0.02), and this effect was augmented on laminin (n=3, p≤0.02), where 85% fewer OB OECs migrated (Figure 3.3E). These data suggest that in vivo and in vitro, LP OECs migrate more readily than OB OECs. 3.2.3 Lesion site area and cavity formation are decreased by LP versus OB OEC transplantation.  Differences in migratory ability may change how effectively LP and OB OECs are able to minimize lesion and cavity formation during the acute phase of SCI.  To test whether differential migration correlated with lesion size, I used the GFAP-reactive border as a marker of the perimeter of the lesion area and measured lesion area, cavity area, and the percentage of the lesion site occupied by cavity (Figure 3.4A).   All groups exhibited an easily defined lesion area containing some evidence of a cavity.   However, rats that received direct OB OEC transplants (Figure 3.4A) had a significantly increased lesion site area (by approximately 2-3 fold) compared with rats that received the same cells rostro-caudally (Figure 3.4B) or with rats that received LP OECs transplanted either directly (Figure 3.4C) or rostro-caudally (Figure 3.4D) (n=4 all groups, 72   73   p≤0.02).   Rats that received direct or rostro-caudally transplanted LP OECs also developed smaller cavities than rats with direct OB OEC transplants (Figure 3.4F; n=5,4,4, p≤0.02, 0.04 respectively).  The greatest decrease in cavity formation was in rats that received LP OECs transplanted rostro-caudally (Figure 3.4F), which also resulted in a significantly lower percentage of the lesion site occupied by cavity (Figure 3.4G), a trend not observed after rostro- caudal OB OEC transplantation (Figure 3.4G).  The smaller lesion site area in rostro-caudally transplanted LP and OB OEC rats and directly transplanted LP OEC rats was 4 times smaller than that seen in control animals, which also developed cavities an average of 46 times larger (p≤0.01, all groups) than transplanted rats, occupying 93 ± 3.3 percent of the lesion site. 3.2.4 Reactive astrogliosis is reduced by rostrocaudal transplantation.  The physical barrier of the lesion site and cavity are major obstacles to axonal regeneration (Schwab 2002). However, the astrocytic scar, produced by reactive astrocytes, contains growth inhibitory molecules that form an additional chemical obstacle to growing axons (Camand et al. 2004; Moreau-Fauvarque et al. 2003).  Since both the OEC type and method of introduction significantly altered the lesion site and cavity area, I hypothesized that inhibitory astrogliosis might occur differentially between treatment groups. All groups were tested for the distribution of GFAP-expressing astrocytes (Figure 3.4A-D).  The intensity of GFAP immunoreactivity surrounding and including the lesion site was normalized to an equally-sized defined area of GFAP-immunoreactivity contralateral to the lesion site.  This controlled for between animal differences in baseline reactivity and background staining.  Direct transplantation of LP and OB OECs produced a reactive astrogliosis similar in extent to direct injection of media (Figure 3.4A,C,H).  In contrast, rostro-caudal injection of media stimulated the most extensive astrogliosis of all treatment groups, a reaction that was significantly lowered by transplantation of both OEC types rostral and caudal to the lesion  (n=4,4,4; LP p≤0.001, OB p≤0.01).  Rostrocaudal transplantation of both LP and OB OECs also significantly decreased GFAP intensity compared to direct transplantation, with rostrocaudal LP OEC transplantation lessening GFAP reactivity to the greatest extent.(Figure 3.4H, n=4,4,,5,4,4,4, p≤0.05).  The reduced astrocytic reaction observed in rostro-caudal versus direct transplanted rats suggests this method of transplantation may create a more permissive environment for axon extension 74   (Lakatos et al. 2003), although no additional decrease was observed in CSPG expression between treated groups (data not shown) over that previously reported (Ramer et al. 2004a). 3.2.5 LP and OB OECs similarly recruit endogenous Schwann cells to the lesion site Interactions between OECs and glia other than astrocytes could also significantly affect regeneration outcomes. Schwann cells (SCs) are highly permissive for regeneration and have been widely used for their ability to promote regeneration and remyelination following CNS lesions as both dissociated cell transplants, or as part of a peripheral nerve bridge (Bunge 2002). We have previously suggested that part of the mechanism whereby OECs may stimulate regeneration is their ability to promote the infiltration of endogenous SCs into the spinal cord (Ramer et al. 2004a).  However, we do not know if OB OECs also promote SC infiltration in this lesion paradigm. Immunostaining of all groups for rat specific low-affinity nerve growth factor receptor, mp75, compared with GFAP and GFP+ OECs, revealed endogenous cells expressing p75 in areas with OECs within the lesion site (Figure 3.5B,D) and, less frequently, at the injection points in LP (Figure 3.5A) and OB-treated rats (Figure 3.5C).  We have previously identified these rat p75+ cells as Schwann cells (SCs) using other markers for rat SCs (Chernousov et al. 1999; Ramer et al. 2004a). Ingression of Schwann cells into the spinal cord parenchyma (measured from the GFAP lesion boundary) was more extensive in both OEC- treated groups, where Schwann cells were found maximally at 350 μm from the lesion site, than in control animals (Figure 3.5E), where they did not penetrate beyond 80 μm of the lesion boundary.  SCs that infiltrated after OEC transplantation associated similarly with LP (Figure 3.5B) and OB OECs (Figure 3.5D), as well as with CNS astrocytes (Figure 5B,D).  This suggests that some of the pro-regenerative effects of LP and OB OEC transplantation may be due, in part, to their ability to recruit, activate, or cooperate with endogenous Schwann cells. 3.2.6 Axonal sprouting into the lesion site is increased in rats transplanted rostrocaudally with LP OECs animals.  The results so far suggest distinct biological differences in the ability of OB and LP OECs to promote repair.  I next extended these findings to specifically test (1) differences in axonal outgrowth, (2) differences in angiogenesis, and (3) differences in the directionality of these parameters in relation to transplanted cells. This would allow me to establish if secreted 75   76   factors released into an SCI lesion area may uniformly promote repair throughout the lesion site, or if the location of the OECs could directly change repair and regeneration parameters. Large calibre axons expressing neurofilament heavy chain (NF) demonstrated differences in degree of sprouting depending both on whether LP (Figure 3.6A,B) or OB OECs (Figure 3.6C,D) were transplanted, and whether introduction was direct or rostro-caudal.  In all animals, some axons sprouted into cavity-free areas of the lesion site, regardless of cell delivery approach, however the distribution of axon growth appeared to differ between treatment groups, in particular, within the lesion site.   In rats that received direct transplants of LP and OB OECs, axonal growth was largely stalled at the caudal lesion boundary (Figure 3.6B,D, n=4/4 each group).  In addition, although some NF+ axons were visible within the lesion site in both direct transplant groups, zones devoid of NF+ axons were apparent, adjacent to the GFAP immunoreactive border (LP Direct, n=3/4; OB Direct, n=4/4).  Rats that received rostro-caudal LP OEC transplants (Figure 3.6A) demonstrated a higher incidence of NF+ axons within the lesion site compared with all other groups. Furthermore NF+ reactivity appeared uniform throughout the lesion site, including at the astrocytic border, in comparison with rats that received LP or OB OECs transplanted directly into the lesion.  Rats that received rostro-caudal OB OEC transplants demonstrated the lowest incidence of NF+ reactivity within the lesion site (Figure 3.6C, n=4/4).  To quantitatively assess the differences in axon growth within the lesion site in all treatment groups, densitometry of NF+ pixels was performed by outlining the GFAP reactive border and counting NF+ pixels within this area (Figure 3.6E). A significant increase in growth within the lesion was promoted by rostrocaudal transplantation of LP OECs in comparison with all other groups (Figure 3.6E, n=4, p≤0.001 all groups). However, rats that received OB OECs directly at the lesion site demonstrated significantly more NF+ axons within the lesion compared with rostro-caudal OB OEC transplants (n=4, all groups, p≤0.05).   These results suggest that cell type and delivery method differentially influenced the pattern of growth of sprouting axons and also affected their ability to penetrate the lesion site. 3.2.7 Directional angiogenesis is enhanced by OEC treatment  Much of the axonal outgrowth reported above, particularly in rats treated by rostro-caudal LP OEC transplantation (Figure 3.6A), was directed towards the lesion site. We have previously shown that angiogenesis in the lesion site occurs following LP OEC transplantation, and that 77   78   many axons extend along blood vessel laminae (Ramer et al. 2004a).  Furthermore, an angiogenic response correlated with axon growth has also been demonstrated following contusive injury (Loy et al. 2002).  Therefore it is possible that the direction of axon growth may be related to the direction of blood vessel growth, and this could determine whether different transplantation strategies could be used to encourage sprouting axons to extend longitudinally down the spinal cord, instead of becoming entangled at the lesion site.  To assess the extent and directionality of blood vessel growth, anti-rat endothelial cell antigen 1 (RECA) and GFAP detection were used to visualize blood vessels around and within the lesion site (Figure 3.7A-E). To analyze directionality, montages of z-stacked images of RECA-immunoreactive endothelia were assigned a threshold intensity (applied within and across groups), counted, and assigned a best fit line along their linear axis, using an algorithm programmed in MatLab, to determine their direction of growth  (Figure 3.7F).  The predicted intersection points of lines were then summed to produce contour maps of most likely endothelial convergence (Figure 3.7G-K).  The density of intersections within the lesion site was normalized to a baseline of intersections within the same lesion area, randomly placed throughout the same spinal cord. This provided a measure of the extent of directionality of blood vessels toward the lesion site, normalized for lesion area and for number of blood vessels. Scores greater than zero therefore indicate directionality of blood vessels toward the lesion site.    In all OEC treated rats, there were significantly greater numbers of blood vessels than in control animals (Table 3.1).  As well, there was a significant and greater than 1.9 fold increase in directionality of blood vessel growth toward the lesion site in rats that received OEC transplants, compared with the random endothelial growth in control animals (Table 3.1).  Rats that were directly transplanted with OB OECs demonstrated the most significant blood vessel growth directly toward the lesion site.  Vascularisation and enhanced regeneration have been tightly linked in the PNS (Hobson et al. 2000), and the increase in blood vessels and their direction of growth in all OEC treated rats may favor directed axon growth within the CNS. 3.2.8 Autotomy is reduced by OB OEC, but elevated by LP OEC transplantation.  Transplantation of either LP or OB OECs resulted in the enhanced growth of NF+ axons within the lesion site, however it is unclear whether these axons form functional connections, either appropriate or inappropriate, with target neurons.  Twenty-seven days following lesion and 79   80   81   transplantation, autotomy was apparent in some rats, so we sought to determine whether the degree of autotomy varied between groups, since this measure could serve as an admittedly negative behavioural read-out for changes in functional connectivity.  Autotomy is the self-attack of a (partially) denervated limb, characterized by excessive licking, grooming, and biting of the anaesthetic limb, eventually resulting in digit amputation (Kauppila 1998; Zimmermann 2001). Autotomy was scored using the methods described by Wall (1979), in which a total of 3 points were awarded for the attack of each digit; 1 point for nail removal, 2 points for amputation up to the first knuckle, and 3 points for total digit removal, for a total of 15 possible points for each paw (Figure 3.8A).  For each rat, the affected right hindpaw was scored as well as the left hindpaw, which served as an internal control.  Rats that received direct transplants of LP OECs (n=4) had the greatest autotomy scores, significantly greater than either control (n=4) or directly transplanted OB rats (n=5) (Figure 3.8B, p≤0.05, p≤0.01).  Rostro-caudally transplanted OB OEC rats demonstrated lower autotomy than rostro-caudally transplanted LP rats and control rats (Figure 3.8B, n=4,5,5, p≤0.05, p≤0.05 respectively). LP OEC transplantation thus increased autotomy regardless of injection paradigm, whereas the incidence (OB, 2/8 rats; LP, 6/10 rats) and extent of autotomy (92% less autotomy in OB rats) was less in OB treated rats. 3.2.9 Sprouting of tyrosine-hydroxylase, Substance P and calcitonin gene-related peptide positive axons is differentially stimulated by LP or OB OEC transplantation. To test if LP and OB OECs may differentially stimulate axonal sprouting that could contribute to autotomy, I assessed the pattern of axon outgrowth from dorsal roots afferents as well as different descending tracts.  Our lab has previously shown that LP OEC transplantation stimulates the growth of serotonergic (5HT+) and tyrosine hydroxylase (TH+) positive axons, as well as some growth of dorsal root afferents expressing calcitonin gene-related peptide (CGRP+) or Substance P (subP+; Ramer et al. 2004a). I assessed horizontal sections through the lesion site for the presence of these axonal subpopulations, using the GFAP+ border to identify the boundaries of the lesion site.  Montaged, Z-stacked sections were assessed for the total number of positive pixels within the lesion site and within defined areas 0.5mm and 1mm rostral and caudal to the lesion. I first tested for the presence of different axonal subtypes at 24 hours following lesion and transplantation, and found all axons tested were absent from the lesion site and retracted 82   83   substantially from the rostral and caudal lesion boundaries (Figure 3.9). By 28 days, in rats that received LP OECs (Figure 3.10A) and OB OECs (Figure 3.10B), subP-positive axons were found within the lesion site but were increased caudal to the lesion site.  No significant differences in axonal outgrowth of subtypes were observed between direct and rostrocaudally transplanted rats within LP or OB OEC treatment groups, and both sets of data for each OEC subtype are presented collectively. In OB transplanted rats, subP reactivity was found rostral to the lesion site (Figure 3.10B, inset), whereas less reactivity was observed at this location in LP OEC (Figure 3.10A, inset) transplanted rats. SubP-positive axons were significantly increased at all points in rats that received OB OECs compared with LP OECs (Figure 3.10C, n=4 each group, p≤0.05 all levels).  This suggests that OB OECs are more effective than LP OECs at promoting the sprouting of subP positive axons following injury. Calcitonin gene-related peptide (CGRP) positive nociceptive/temperature responsive afferents are also responsive to LP OEC transplantation (Ramer et al. 2004a) and their sprouting could impact pain sensation following SCI (Jang et al. 2004).  Detection of CGRP-positive axons in rats that received LP OECs (Figure 3.10D) and OB OECs (Figure 3.10E) revealed similar sprouting patterns between groups, although LP OECs slightly increased sprouting 0.5mm rostral to the lesion site (Figure 3.10F, n=4 each group) and demonstrated increased sprouting at the rostral lesion border (Figure 3.10D, inset; 3.10F).  Both types of OEC did, however, promote the growth of CGRP-positive axons over control animals, particularly within the lesion site, where no sparing of CGRP-positive axons was visible at 24 hours post lesion.  No significant differences in CGRP-positive sprouting were observed between direct and rostrocaudally transplanted rats. Sprouting of Raphespinal 5HT-positive axons appeared similar in both LP and OB- treated rats (Figure 3.10G,H), and was extensive in comparison to CGRP+ and SubP+ sprouting; no significant differences were observed between transplant groups (Figure 3.10I, n=4 all groups).  5HT-positive axons also appeared to diminish significantly slightly caudal to the lesion site (Figure 3.10G,H, insets). In contrast, detection of tyrosine-hydroxylase (TH), revealed significantly more TH-positive axons rostral to and within the lesion site in rats that received LP OECs compared with OB OECs (Figure 3.10J, K, L).  This was particularly visible at the rostral lesion boundary, where TH-positive axons were increased in LP (Figure 3.10J, inset) compared 84   85   86   with OB transplanted rats (Figure 3.10I, inset).  Since neither 5HT nor TH-positive axons were present within the lesion site 24 hours post lesion, the transplanted LP and OB OECs likely promote growth of these axon subtypes. 3.3 Discussion 3.3.1 Rationale and summary OECs are currently considered prime candidates for cell-based therapies to repair the injured CNS (Lakatos and Franklin 2002). Because some OB OEC transplantation experiments have indicated different degrees of functional recovery from SCI (Garcia-Alias et al. 2004; Keyvan-Fouladi et al. 2003; Ramon-Cueto et al. 2000), cell preparations containing LP OECs are currently being tested for treating human SCI, based on the assumption that LP and OB OECs are functionally equivalent (Huang et al. 2003; Rabinovich et al. 2003; Senior 2002).   For both mechanistic and clinical reasons, I compared the effect of transplanting OB and LP OECs into the lesioned spinal cord and determined that they produce significantly different regeneration outcomes.  In comparison to OB OECs, LP OECs decrease lesion site and cavity area, and increase sprouting of NF+ and TH+ axons, but also stimulate autotomy.  In contrast, OB OEC transplantation increases lesion site area and cavity formation, promotes sprouting of subP+ axons and promotes the greatest directional angiogenesis.  In addition, rostro-caudal compared with direct transplantation suggests that LP and OB OECs have different abilities to migrate, and that cell delivery method significantly affects astrogliosis, NF+ sprouting and cavity formation. 3.3.2 LP and OB OECs display instrinsic biological differences in expansion, migration, and interactions with astrocytes following transplantation into SCI The morphological and antigenic profiles of LP and OB OECs in culture examined here (Figure 3.1) suggest that both OEC types are very similar. Some differences in expression of cell adhesion molecules has been reported such as an increase in expression of O4 antigen and PSA- NCAM in OB OEC cultures over OEC cultures derived from olfactory nerve rootlets, however, the functional consequences of this differential expression are unclear (Kumar et al. 2005). Apart from some antigenic dissimilarity, differences in LP and OB OEC roles in their PNS and CNS environments, and their expansion capacities in vitro, imply that LP and OB OECs should not be considered functional equals a priori, and could behave differently following 87   transplantation.  Indeed, when transplanted rostro-caudally, LP and OB OECs were distributed differently 28 days after lesion; significantly more GFP+ LP OECs had migrated into the lesion site, whereas GFP+OB OECs were located between the caudal injection point and lesion site (Figure 3.3). This distribution could be explained by increased death of OB OECs at the lesion site, or a greater loss of LP OECs at injection sites. The resemblance in cell distribution of 28 day R/C OB and 7 day R/C LP rats, coupled with the increased survival of OB OECs at 28 days, suggests that the difference in cell distribution is primarily a result of intrinsic LP/OB OEC migration differences that resemble those assessed in vitro (Figure 3.3). The ability of OECs to differentially migrate into a lesion site could have important consequences in SCI, since, in a dorsolateral funiculus lesion, a cavity forms at 24 hours following injury when untreated (Ramer et al. 2004a). The greater percentage of cavity in rats that received rostro-caudal OB OEC transplants could thus simply be the result of a retarded migration of OB-OECs (Figure 3.4).  The long-term presence of OECs at the lesion site appears to be a major determinant of reduced cavity size, since rostro-caudal LP-OEC transplantation significantly reduced cavity formation compared with direct transplantation (where fewer LP- OECs survive) (Figure 3.4). The decreased lesion site area in rostrocaudally-transplanted rats was further reflected in differences in astrocytic reactions following rostrocaudal versus direct transplantation. Increased GFAP reactivity around the lesion site of rats with direct versus rostrocaudal transplants of both OEC type, suggests that LP and OB OECs may have similar interactions with astrocytes (Figure 3.4).  However the increased GFAP immunoreactivity in rats that received transplants directly into the lesion site suggests that the acute presence of OECs within the lesion either enhances astrocyte division and/or exacerbates astrocytic hypertrophy (Figure 3.4).   Differences in astrogliosis could arise from (1) unfavorable interactions between OECs and astrocytes, (2) an increased local immune response following direct transplantation, or (3) a favorable interaction between OECs and astrocytes that requires long-term survival of OECs.   Increased astrogliosis is unlikely due to unfavorable OEC-astrocyte interactions, as astrocyte size, shape, and reactivity are not altered in vitro by contact with OB OECs (Lakatos et al. 2003; Lakatos et al. 2000). In addition, similar numbers of autofluorescent immune infiltrates were detected within the lesion site of rostrocaudally and directly transplanted animals (of the same cell type), arguing that the 88   immune response may contribute to, but is not responsible for, increased astrogliosis in directly- transplanted animals. Therefore, the decrease in GFAP intensity in rostro-caudally transplanted rats may be due to a favorable, but necessarily prolonged interaction between OECs and astrocytes, since this delivery approach also promotes OEC survival.  The association of OECs and astrocytes may synergize to reduce astrogliosis and create a more permissive environment, allowing SC infiltration and axonal extension.  SC transplantation has promoted behavioural recovery from SCI (Fouad et al. 2005; Pearse et al. 2004), therefore their infiltration following OEC transplantation could contribute to or potentiate regeneration regardless of OEC type. 3.3.3 LP versus OB OEC transplantation promotes the growth of differing subpopulations of spinal axons  Functional assessments performed following OB OEC transplantation have focused largely on positive outcome measures (Garcia-Alias et al. 2004; Keyvan-Fouladi et al. 2003; Ramon-Cueto 2000), with the assumption that LP-OECs would be similarly beneficial. The biological differences observed between LP and OB OECs in vitro (Figure 3.3) and following direct and rostro-caudal transplantation (Figures 3.3-5), suggest there may be differences in how OEC type and transplantation approach affect axonal sprouting.  These caveats are particularly clear in the case of LP or OB OEC implantation into the dorsal roots to promote their regeneration.  While some instances of OB OEC transplantation following dorsal root section have resulted in the formation of cooperative OEC-astrocyte bridging channels at the PNS-CNS glia limitans and the ingrowth of primary afferents into the spinal cord (Li et al. 2004b; Ramon- Cueto and Nieto-Sampedro 1994), other experiments using implanted LP OECs have found few regenerating fibres (Ramer et al. 2004b).  Interestingly, at least part of the variance in outcome may be attributed both to the cell type transplanted, as well as to the method of cell introduction. While deposition of LP OECs within the dorsal root ganglion or within an injured dorsal root results in extensive OEC migration within the PNS, only implantation of OECs into the dorsal columns by disruption of the dorsal root entry zone results in migration of OECs within both the CNS to the PNS, and the growth of CGRP-positive axons into the spinal cord (Ramer et al. 2004b).  In the current investigations, sprouting of NF+ axons within the lesion site was greatest in rats that received rostro-caudal LP OEC transplants, with little axon density decrease across the lesion site (Figure 3.6).  In contrast, although direct transplantation of LP OECs was effective 89   at drawing axons toward the lesion boundaries, fewer axons were able to penetrate the lesion/host interface.  This effect could be a result of both trophic/tropic actions of OECs and changes to the permissiveness of the glial scar/lesion site, since rostro-caudal transplantation also decreased GFAP reactivity and cavity formation (Figure 3.4).  Therefore permissiveness and trophic/tropic actions may be maximized by rostro-caudal, but decreased by direct LP OEC transplantation. However, if only the location of the cell transplant were responsible for differences in NF+ growth, then based on evidence from LP OECs, one would expect (1) greater NF+ axons within the lesion site in rats transplanted rostrocaudally than directly with OB OECs, and (2) the same amount of NF+ growth within the lesion site in rostro-caudal LP and OB- treated rats.  A comparison of only rostrocaudally transplanted LP and OB OEC rats shows that significantly greater NF+ axons are present within the lesion in LP-treated rats (Figure 3.6).  This suggests that LP and OB OECs might express different growth factors or cell adhesion molecules and/or produce them at different levels. The significant increase in autotomy in LP OEC versus OB OEC transplant groups further underscores potential functional differences between these two cell types.  While the mechanisms of autotomy are not understood, autotomy is common after denervation or situations where sensory modalities may be affected, as in our model. At present, it appears that OEC transplantation might have modified the sensory system by facilitating differential growth of different axonal populations or by differentially altering local circuitry. These possibilities imply a higher degree of plasticity/remodeling may be induced by LP versus OB OECs, which could result in undesired effects. In comparison, recent work with transplanted embryonic stem cells into SCI has revealed the development of allodynia in addition to increased motor function, a feature not tested in most OEC transplantation studies (Hofstetter et al. 2005). The differential sprouting of some axonal populations further implies that LP and OB OECs secrete different growth factors, neurotrophins, ECM, or cell adhesion molecules, and therefore may differently promote recovery from SCI (Figure 3.10): SubP+ axons sprouted significantly more in OB- than LP-treated rats (Figure 3.10). However, the most robust difference in sprouting was the growth of TH+ neurons, whose growth was highly promoted by LP OEC transplantation (Figure 3.10).  In studies of axon sprouting following dorsal rhizotomy, similar differences in the propensity of fibre subpopulations to grow in response to LP OEC transplantation have also been 90   reported; while small/medium diameter wheat-germ agglutinin-traced fibres grew preferentially along OEC tracts, large diameter afferents identified by cholera toxin B tracing were not attracted to OEC deposits (Ramer et al. 2004b).  Further evidence of differences in expression of PSA-NCAM and O4 between cultures of OB OECs and LP OECs is suggestive of an involvement of these, and other factors, in encouraging the outgrowth of varying neuronal populations in response to OEC coculture in vitro (Kumar et al. 2005).  Increased outgrowth of retinal ganglion cells was described following LP OEC over OB OEC coculture, despite the increased expression of PSA-NCAM, which has been associated with increased outgrowth and synaptic remodelling, in OB OECs (Kumar et al. 2005; Sato et al. 2001).  Differences in the abilities of LP and OB OECs to promote axon elongation are also in accord with their roles in the olfactory neuraxis.  Furthermore, although OB OECs express NGF, BDNF, GDNF, artemin and CNTF in vivo and in vitro (Lipson et al. 2003; Woodhall et al. 2001), their ability to promote outgrowth of peripheral ganglia and CNS explants, as well as to promote sprouting within the spinal cord via secreted factors has been variable whereas their ability to do so via direct cell-cell contact is greater (Chung et al. 2004; Lipson et al. 2003; Sonigra et al. 1999). 3.3.4 Summary of findings and future directions  This investigation represents the first data directly comparing LP and OB OEC transplantation in a single SCI model, and shows that there are fundamental differences in intrinsic OEC properties that are retained after they are transplanted into the lesioned spinal cord. These differences relate to (1) the ability to migrate and create a permissive regeneration- promoting environment at the lesion site and (2) differential outgrowth-promoting characteristics of LP and OB OECs.  Although the most benefit to regeneration parameters was achieved with rostro-caudal LP OEC transplantation, this treatment is also associated with autotomy.  This suggests that the vigorous sprouting associated with rostro-caudal LP OEC transplantation must be controlled so that appropriate connections are favored over inappropriate, and nearby uninjured tracts are not adversely affected.  Thus, rostro-caudal LP OEC transplantation provides the greatest opportunity to promote regeneration or reconnect local circuitry following SCI, suggesting that LP OECs, combined with directed axon guidance stimulation, may merit further development as a means to restore motor and sensory function following spinal cord lesion.   91   CHAPTER 4: SPARC FROM TRANSPLANTED OLFACTORY ENSHEATHING CELLS DIFFERENTIALLY MEDIATES THE GROWTH OF SPINAL NEURON POPULATIONS IN THE INJURED SPINAL CORD  Note: A portion of this chapter has been previously published as Au E, Richter MW, Vincent AJ, Tetzlaff W, Aebersold R, Sage EH, and Roskams AJ (2007) J. Neurosci. 27(27): 7208-21, with the following additions and modifications: The document has been formatted to incorporate the stylistic guidelines of this thesis, the  introduction, results, and discussion have been rewritten in their entirety, and the methodological section has been removed and included in Chapter 2.  The experiments involving DRG outgrowth assays were performed by Dr. Edmund Au, and are graciously included in the introduction with his kind consent.  4.1 Introduction 4.1.1 How do OECs affect neurite outgrowth?  The mechanisms used by OECs to promote the outgrowth of neurons in their native environment, and elsewhere in the nervous system following their implantation, are varied.  This multifactorial activity most likely entails a codependence on the regulation of neurite outgrowth machinery, the cytoskeleton, through the secretion of outgrowth promoting molecules and the expression of cell surface adhesion molecules, with the regulation of reparative or homeostatic processes such as wound healing, tissue reperfusion, or structural support.  Several groups have investigated how secreted factors from OECs can regulate neurite outgrowth.  By using ELISA assays, ribonuclease protection assay or RT PCR, some of the neurotrophins and growth factors produced by OB OECs have been elucidated, and include NGF, BDNF, GDNF, CNTF, neurturin, and artemin (Boruch et al. 2001; Lipson et al. 2003; Woodhall et al. 2001). Addition of OEC conditioned media in vitro to sympathetic, spinal ganglia or retinal ganglion cells can also provoke modest outgrowth of these neurons, although 92   compared with the outgrowth resulting from OEC explant coculture, it is unenthusiastic (Lipson et al. 2003; Sonigra et al. 1999).  When OECs or their conditioned media, encapsulated in a polyvinylidene fluoride hollow fibre, are transplanted into a T8 dorsal column lesion, the sparsely-growing corticospinal fibres extend along the conditioned media-containing capsule, and increase collateral branching of the ventral corticospinal tract, suggesting that factors in conditioned media could increase corticospinal outgrowth (Chuah et al. 2004).  Although the known growth factors produced by OECs could contribute to their outgrowth-promoting abilities in vitro and in vivo, it remains unclear whether other secreted factors mediate this activity.  By identifying and understanding which factors induce axon growth in specific neuronal populations, it may become possible to enhance or direct the outgrowth induced by OECs in the injured spinal cord. 4.1.2 SPARC is secreted by OECs and is partially responsible for their outgrowth-promoting activity. We have previously demonstrated that secreted factors from OECs can be harvested, and their bioactivity preserved by concentration under pressure (Au et al. 2007).  Dorsal root ganglia grow in response to this OEC conditioned media (OCM) in a dose-dependent manner, although their outgrowth is different in kind than that observed following OEC coculture (Figure 4.1A-C). Outgrowth following coculture is fasciculated and elongated, whereas increased branching is provoked by OCM addition (Figure 4.1 K).  To dissect the active components of OCM that contribute to increased neurite outgrowth, an isotope coded affinity tagging (ICAT) proteomics screen was adopted, and yielded, among other candidates, secreted protein acidic and rich in cyteines (SPARC; bone morphogen 40; osteonectin).  Exogenous addition of SPARC to DRG explants increased neurofilament-positive outgrowth, whereas addition of a function-blocking antibody to DRGs treated with OCM partially abrogated the effect of the OCM (Figure 4.1 D, E, J).  These results, coupled with OEC expression and secretion of SPARC in vitro by Western blot and immunocytochemistry, and the upregulation of SPARC expression in OECs following olfactory bulbectomy, suggest SPARC is a key contributor to OEC-mediated outgrowth (Au et al. 2007). 93    94   4.1.3 Expression and function of SPARC in the nervous system and in wound healing  Although there are no previously-reported expression profiles or known roles for SPARC in the nervous system, its involvement at the intersection between extracellular matrix, cell adhesion or counteradhesion, and growth factor regulation have been documented in other systems (Bradshaw 2001). In particular, SPARC is known to function in wound repair, angiogenesis, and ECM organization (Kyriakides and Bornstein 2003; Yang et al. 2000).  During embryogenesis, SPARC is highly expressed in developing cartilaginous zones, the pia mater, lung, tongue, choroid plexus, and shows much lower expression than its close relative, SPARC- like 1, in the brain (Mothe and Brown 2001).  SPARC expression has also recently been documented in the developing, mature, and injured olfactory system (Au et al. 2007; Vincent et al. 2008).  SPARC protein contains an ECM-interacting domain whose attributed functions include inhibition of cell spreading, proliferation, and focal adhesions (Bradshaw 2001).  This ECM- interacting domain can bind many ECM components including collagens I, II, III, IV, and V, thrombospondin, laminin, and vitronectin (Wang et al. 2005).  SPARC can also bind to growth and angiogenic factors such as vascular endothelial growth factor (VEGF), platelet derived growth factor (PDGF), and basic fibroblast growth factor (bFGF), and can modulate transforming growth factor (TGFβ) activity (Bradshaw 2001).  SPARC is increasingly implicated in the metastatic capacity of meningiomas (Rempel et al. 1999), and is also involved in the growth of new vessels from existing vasculature that is co-opted by cancer tumorigenesis (Liaw and Crawford 1999; Wong et al. 2007).  Because of SPARC’s reported activities as a master regulator of interactions with the ECM, growth factors and angiogenesis, and its control of focal adhesion formation and adhesion/counteradhesion, it is a prime candidate regulator of OEC-induced spinal fibre growth. 4.1.4 Rationale and aims Because SPARC was highly represented in an ICAT proteomic screen of OEC secreted factors, promoted the outgrowth of dorsal root ganglion neurons, and blockade of SPARC or its downstream intracellular activities in OEC conditioned media significantly decreased its outgrowth-promoting activity, we were interested in testing whether SPARC secreted from 95   OECs would affect reparative processes in the spinal cord following OEC transplantation.  To examine the contributions of SPARC to OEC repair of the injured spinal cord the following aims were undertaken: • To implant cultured wildtype or SPARC null OECs into a crush of the dorsolateral funiculus and ascertain whether biological processes of wound healing, cavity formation, scar formation, immune cell infiltration, and angiogenesis, previously reported following OEC transplantation, would be altered by the absence of SPARC in transplanted OECs. • To determine whether the growth of injured spinal axons, that are promoted by the presence of transplanted WT OEC would be altered in number or distribution by the absence of SPARC in OECs. By determining the contribution of a single secreted factor from OECs to the growth or regeneration of spinal tracts, this will facilitate the development of tract-specific targeted therapies for spinal cord injury.  4.2 Results 4.2.1 Culture, preparation, and transplantation of WT and SPARC null LP OECs. To assess the contribution of SPARC expressed by OECs to the reparative or regenerative responses previously ascribed to OECs following their transplantation in the injured spinal cord, LP OECs were cultured from PD5 C57/Bl6 (wild type; WT), or SPARC null mice, generated by a targeted insertion disruption of the SPARC gene (kind gift of E.H. Sage; Norose et al. 1998), as previously described (Au and Roskams 2003).   Whereas a number of differences have been reported between WT and SPARC null mice, reflecting some profound alterations in wound healing and the regulation of matrix-growth factor interactions (Bradshaw 2001), SPARC null adult and PD5 mice appeared grossly normal, with no obvious differences in behaviour, lethality, size, or anatomy.  When LP OECs were cultured from WT or SPARC null mice, no differences were observed in cell morphology, spreading, or division, although migration of SPARC null OECs was slightly decreased in a Boyden chamber migration assay (data not 96   shown).  SPARC protein expression in cultured SPARC null OECs by immunocytochemistry or Western blot was undetectable.  To visualize SPARC null and WT OECs following their transplantation into the spinal cord, both sets of OECs were infected with a retrovirus encoding eGFP (see Materials and Methods), and separated from uninfected, non GFP-expressing cells by FACS, to generate purified, GFP-positive SPARC null or WT LP OEC cultures for transplantation.  LP OECs were harvested at passage 3 and transplanted acutely into cyclosporine A immunosupressed rats with a C3/C4 crush to the dorsolateral funiculus, as previously described (Richter et al. 2005).   Twenty-eight days following injury and transplantation, spinal cords were sectioned and the reparative and regenerative results of WT or SPARC null OEC transplantation were assessed with respect to those effects previously described as a result of OEC transplantation.  These included: decreased lesion site and cavity formation, altered scar deposition, directional angiogenesis, Schwann cell infiltration, and the growth of spinal neuron populations toward or within the lesion sites of transplanted animals (Ramer et al. 2004a; Ramon-Cueto et al. 2000; Richter et al. 2005). 4.2.2 SPARC is expressed by WT transplanted OECs, endogenous astrocytes, and on blood vessel laminae in the injured spinal cord.  Twenty-eight days following their transplantation, WT LP OECs persisted in their high expression of SPARC in vivo, relative to the low levels of SPARC expressed in the surrounding spinal cord (Figure 4.2A).  SPARC expression also appeared to be upregulated perilesionally, in astrocytes in both SPARC null and WT OEC transplanted animals (Figure 4.2A, B arrowheads), and decreased with distance from the lesion site.  Particularly in WT OEC transplanted rats, SPARC was also deposited throughout the parenchyma surrounding the lesion site, and was collected on the laminae of laminin-positive blood vessels that extended finger-like projections into the centre of the lesion site in both WT and SPARC null transplanted animals (Figure 4.2C, D).  SPARC was also expressed by infiltrating Schwann cells and those in the dorsal roots (Figure 4.2A, arrow).  SPARC expression at or surrounding the lesion site in both treatment groups may therefore also be contributed by endogenous Schwann cells (Ramer et al. 2004a). Endogenous Schwann cells, visualized by anti-rat p75 immunostaining, were recruited to the lesion sites of WT (Figure 4.2E) and SPARC null (Figure 4.2G) transplanted animals, although their numbers and distribution appeared the same, regardless of OEC type.  Thus, with the 97    98   exception of transplanted OECs themselves, SPARC expression in the lesion spinal cord appeared similar in origin and distribution between transplant groups, and was significantly lower than the intense SPARC expression of WT LP OECs. 4.2.3 SPARC null and WT OECs equally decrease lesion site area and cavity formation, but differentially affect macrophage recruitment to the lesion site.  A significant advantage of OEC transplantation, particularly in comparison to Schwann cell treatment, is their ability to reduce lesion site and cavity area, and integrate with host tissue (Ramon-Cueto et al. 1998).  Whereas transplanted Schwann cells often form contiguous borders with endogenous astrocytes, inducing high proteoglycan and neurite outgrowth-inhibitory molecule expression in bordering astrocytes (Paino et al. 1994; Plant et al. 2001), transplanted OECs form jagged and interspersed boundaries with astrocytes at the lesion site (Plant et al. 2001; Ramer et al. 2004a; Ramon-Cueto et al. 1998).  Given the established role of SPARC in dermal wound healing, we hypothesized that the absence of SPARC in transplanted OECs might alter wound healing responses following their transplantation into the lesioned spinal cord.  To examine cavity or lesion site formation, GFAP immunoreactivity was examined in WT or SPARC null transplanted animals above, below, and within the lesion site, in order to circumscribe the lesion boundary (Figure 4.2A,B,E,F).  Transplantation of WT and SPARC null OECs resulted in a significant decrease in lesion site area and cavity area compared with non- transplanted control animals (data not shown).  However, no differences were observed between transplant groups in the formation of cavities at the lesion site, or in total lesion site area (Figure 4.2K).  Furthermore, integration of OECs with host astrocytes appeared similar in both transplant groups, and an interweaving of astrocytic and OECs processes along the brushed lesion site border was visible, regardless of SPARC expression (Figure 4.2B,F).  Expression of chondroitin sulphate proteoglycan (CSPG; Figure 4.2 C,G) in astrocytes at the lesion border was also unchanged in quantity or distribution by the presence or absence of SPARC in transplanted OECs.  However, in contrast to the continuous border of CSPG deposited in media-treated animals, CSPG deposition in WT and SPARC null transplanted rats was perpendicular to the lesion site, as has been previously reported for OEC transplantation (Figure 4.2 D,H; Ramer et al. 2004a).  Since SPARC has also been implicated in leukocyte migration (Alvarez et al. 2005), contains growth factor and interleukin interacting motifs that can modify availability of these 99   factors to the immune system, and can interact with VCAM1, which, together with SPARC, can induce actin cytoskeletal rearrangements and intercellular gaps that facilitate leukocyte transmigration across the endothelium (Kelly et al. 2007), we postulated that migration of monocytes into the lesion site might be affected by the absence of SPARC in OECs.  ED1- positive cells, which are predominantly myeloid cells such as tissue macrophages, were significantly increased in the areas surrounding and within the lesion site of WT transplanted rats (Figure 4.3I), over those that received SPARC null cells (Figure 4.3J).  In particular, ED1- positive cells were densely packed within the lesion site of WT transplanted animals, and were significantly increased in this region over SPARC null transplanted animals (Figure 4.3L). These results suggest that while SPARC from OECs does not exert a significant effect on the development of cavities or lesion site morphology resulting from OEC transplantation, alterations in recruitment of the immune system to the site of injury may change sprouting or behvioural outcomes, in as yet undefined manners, since the immune system appears to play roles in both injury and reparative processes following a spinal cord lesion (Trivedi et al. 2006). 4.2.4 SPARC null and WT OECs equally promote directional angiogensis towards their transplant sites.  Repair of the injured spinal cord involves increasing tissue-supportive activites that contribute to the survival of remaining cells and maintain tissue integrity, as well as neurite outgrowth promotive actions to increase regeneration.  LP OECs can contribute to both of these activites, and we have previously found that they are particularly effective promoters of angiogenesis towards sites of cell deposits, which may increase tissue reperfusion and oxygenation (Richter et al. 2005).  Since SPARC also plays pleiotropic roles in regulating angiogenesis, including regulation of endothelial cell proliferation, morphology, migration, and ECM deposition (Jendraschak and Sage 1996), it seemed likely that the absence of SPARC in transplanted OECs might detract from their ability to promote angiogenesis in the spinal cord.  In contrast to the strong evidence linking SPARC to angiogenesis, WT (Figure 4.3 A,B) and SPARC null (Figure 4.4 C,D) transplanted OECs both promoted angiogenesis towards their respective injection sites.  The same numbers of blood vessels within the lesion and transplant sites of WT and SPARC null treated rats were also observed, suggesting the presence of other factors expressed by OECs that account for their ability to promote angiogenesis.  Since SPARC 100   101   102   also has an ECM-interacting domain, and has been implicated in basement membrane deposition during embyrogenesis and after injury (Francki and Sage 2001; Yan et al. 2003; Yan et al. 2005), this suggested that the normally-abundant deposition of perilesional laminin observed after OEC transplantation might be altered in arrangement or quantity (Richter et al. 2005).  Laminin was observed throughout the transplant sites of WT (Figure 4.4 E,F) and SPARC null (Figure 4.4 G,H) OECs, surrounding OECs and blood vessels.  The arrangement of laminin tracts in WT and SPARC null mice were heterogenous, as has been reported previously for WT OECs (Richter et al. 2005), and group differences were not evident in laminin area, distribution, or quanitity.  Thus several major reparative functions of OEC transplantation remain unaltered by the absence of SPARC in transplanted OECs. 4.2.5 Growth of large and small diameter fibres surrounding and within the lesion site is indistinguishable in SPARC null and WT OEC transplanted animals.  Given the in vitro evidence that SPARC can promote the outgrowth of neurofilament- positive dorsal root ganglion neurons, it seemed likely that the absence of SPARC from transplanted OECs might adversely affect the growth of these or other fibres in vivo, 28 days after a lesion to the spinal cord.  In WT (Figure 4.5 A,B) transplanted rats, large-diameter neurofilament-positive fibres (NF, red) were visible rostral and caudal to the lesion site, although many had retracted 350-500μm from the GFAP-positive lesion site border (GFAP, blue).  The retraction of NF-positive fibres from the lesion site was the same in WT and SPARC null (Figure 4.5 C,D) transplanted rats, although in both groups, some NF-positive fibres remained visible at the lesion border and extended into the lesion site.  In both groups, many of the NF-positive fibres emerged from the medial portion of the lesion site, and grew tortuously within the lesion site, rendering it difficult to discern whether any of the same fibres also exited the lesion site. Evidence from in vitro studies also suggested the ability of SPARC to increase the outgrowth of neuron specific tubulin (NST; βIII tubulin) -positive neurites following its addition to a dorsal root ganglion explant (data not shown).  To test whether NST-positive fibres were similarly affected in vivo by the absence of SPARC in OECs, sections through the lesion epicentre, as well as above and below it, were immunoreacted for NST.  The numbers of NST-positive fibres in WT (Figure 4.5 E,F) and SPARC null (Figure 4.5 G,H) treated rats was far greater than the number of NF-positive fibres, and NST fibres were present in both groups at points up to the 103   104   lesion border; little retraction from the lesion site was observed.  In both groups, NST-positive fibres were also visible crossing the host-transplant interface, and entering the lesion site.  NST- positive fibres appeared to grow along the long axes of transplanted OECs, and were observed to turn upon themselves within the lesion site.  Although NST-positive fibres crossed both the rostral and caudal lesion interfaces, it is unclear whether any single fibre extended through both interfaces.  In WT and SPARC null transplanted groups, similar growth of NST-positive fibres were observed rostral, within, and caudal to the lesion site, suggesting SPARC from transplanted OECs does not influence the overall growth of small diameter fibres, or that the effect of SPARC on a single subpopulation of NST-positive fibres cannot be detected when all populations are pooled. 4.2.6 SubstanceP- and tyrosine hydroxylase-positive fibres grow significantly less in SPARC null OEC transplanted, than WT transplanted animals.  The discovery of spinal tract-specific outgrowth promoters will be instrumental in devising targeted approaches to reconnect severed axons with their targets while minimizing collateral branching that may contribute to the development of pain pathologies.  It was therefore of particular interest to ascertain whether SPARC might contribute to the growth of individual spinal tracts after spinal cord injury, specifically to those tracts that have previously been responsive to OEC transplantation including: the raphespinal tract (serotonergic; 5HT-positive), locus coeruleus projection (dopaminergic; tyrosine-hydroxylase-positive), and dorsal root afferents such as calcitonin gene related peptide-positive (CGRP) and substance P (subP) projections.  Serotonergic raphespinal neurons appeared to sprout similarly in WT (Figure 4.6A) and SPARC null (Figure 4.6B) OEC transplanted rats, as did CGRP-positive afferents (Figure 4.6 C,D).  In particular, 5HT-positive fibres were visible rostral to the lesion site, and several crossed into the lesion site in WT and SPARC null groups.  CGRP-positive fibres were visible rostral and caudal to the lesion site in WT (Figure 4.6 C) and SPARC null (Figure 4.6D) transplants, and some fibres crossed the lesion boundary to enter the lesion site in both groups. In contrast to the similarity in growth of 5HT- and CGRP-positive axons following WT or SPARC null transplantation, subP (Figure 4.6 E,F) and TH-positive (Figure 4.6 H,I) fibres were substantially altered in their growth towards OEC deposits depending on OEC genotype. Substance P-positive growth caudal to the lesion site was significantly decreased in SPARC null 105   106   transplanted animals (Figure 4.6 G), although few fibres entered the lesion site.  Coerulear fibres were also significantly decreased in their growth towards SPARC null OEC transplants at sites rostral to the lesion site (Figure 4.6 J), although few fibres in either group permeated the lesion boundary to travel within the lesion site.   This provides the first evidence that specific subpopulations of spinal fibres are selectively stimulated by OEC-secreted SPARC, a response that may be harnessed to increase and direct the growth of severed SubP and TH-positive axons towards their targets following a spinal cord injury.  4.3 Discussion 4.3.1 Reparative responses following WT or SPARC null transplantation are similar  Following transplantation of WT or SPARC null OECs into a C3/C4 crush of the dorsolateral spinal cord, many reparative and neurite sprouting parameters remained similar to those already reported following either LP or OB OEC transplantation into the injured spinal cord (Richter et al. 2005).  These results were surprising because of the reported involvement of SPARC in a number of tissue repair processes elsewhere in the body (Bradshaw 2001; Ramer et al. 2004a; Ramon-Cueto et al. 1998).  SPARC has been prominently implicated in wound healing responses in the dermis, cornea, and the heart, and its expression is upregulated in these tissues, in particular in migrating fibroblasts, after injury and during repair (Berryhill et al. 2003; Wu et al. 2006).  Endogenous SPARC expression is similarly upregulated perilesionally following spinal cord injury in astrocytes and along blood vessels (Figure 4.2).  This suggests that SPARC expression in the injured nervous system may perform similar functions as those previously reported in other areas; in the regulation of basement membrane-cell adhesion interactions, migration of astrocytes or fibroblasts to wall off an area of injury, or in the regulation of angiogenesis (Basu et al. 2001; Bradshaw 2001).  It therefore seemed feasible that the absence of SPARC in OECs might impair or alter their interactions with endogenous astrocytes or invading fibroblasts or would contribute to differences between WT and SPARC null treated rats in cavity or lesion site formation and morphology.  Such differences between treatment groups were not observed, and SPARC null OECs retained the ability, in comparison to untreated control animals, to decrease cavity and lesion site area (Figure 4.3).  Furthermore, 107   while untreated animals clearly exhibit a smooth and tightly-demarcated lesion border, with little tissue present in the resulting cavity, SPARC null and WT OECs interacted similarly with astrocytes to induce a ragged lesion edge (Figure 4.3).  The ability to interact with astrocytes may be one key to the growth-promoting properties of OECs following transplantation; the tightly demarcated barrier established by Schwann cell-astrocyte contacts in vivo or in vitro induces axon turning, thwarting regeneration from graft to host interface (Lakatos et al. 2000; Ramon-Cueto et al. 1998).   It may be that the endogenous expression of SPARC on host astrocytes is sufficient to engage this favourable interaction with SPARC null OECs, particularly if OECs also express the receptor for SPARC, stabilin-1.  Conversely, differential N-cadherin regulation in Schwann cells and OECs has recently been implicated in their dissimilar interactions with astrocytes; N-cadherin homophilic binding between astrocytes and SPARC null or WT OECs may mediate the formation of the interdigitated graft-host boundary observed following OEC transplantation, regardless of SPARC expression (Fairless et al. 2005). SPARC can also exert angiogenic and antiangiogenic effects on existing endothelial cells to produce changes in the extant vasculature, depending on its sustained or transient upregulation, or on interactions with other factors including PDGF, bFGF, or hevin (Jendraschak and Sage 1996). Differences in angiogenesis or the direction of newly-formed vessels were not apparent in WT versus SPARC null treated animals (Figure 4.4).  The unaltered growth of blood vessels in response to either OEC group may reflect the high expression of a number of other angiogenic factors in OECs.  OECs express vascular endothelial growth factor (VEGF) in the olfactory lamina propria, and can upregulate expression of VEGF in perilesional astrocytes following their transplantation into the spinal cord (Au and Roskams 2003; Lopez-Vales et al. 2004).  Since astrocytes surrounding the lesion site express SPARC, and their organization and VEGF expression following injury is modulated by interactions with OECs, this tripartite interplay may account for increased angiogenesis towards OEC deposits, without necessitating SPARC expression in OECs per se. Coversely, PDGF or VEGF expressed by OECs may exert a direct influence upon blood vessel growth, without the involvement of SPARC. Observed differences in monocyte recruitment to the lesion sites of WT and SPARC null transplanted rats (Figure 4.3) may reflect the activity of SPARC in modulating cell migration, extravasation, and immune system function.  Since SPARC can induce shape changes and 108   rounding in endothelial cells and monocytes, as well as inducing barrier dysfunction in endothelial cells, allowing protein flux between cells (Goldblum et al. 1994), the presence of SPARC in WT OECs, and increased SPARC on blood vessels and extracellular environment in this treatment group might affect the migration of macrophages from the blood into the nervous system.  SPARC can also change the activity of recruited cells; SPARC increases prostaglandin release from monocytes, which acts in a paracrine manner on other monocytes to increase matrix metalloproteinase production and alter the ECM environment (Shankavaram et al. 1997). Although SPARC from transplanted OECs may change the number or activation state of monocytes at the lesion site, the beneficial or deleterious effects of immune system activation following spinal cord injury are far from clear.  While implantation of homologous macrophages, preconditioned with injured nerve segments, into a transected spinal cord has resulted in tissue repair, behavioural, and electrophysiological recovery (Prewitt et al. 1997; Rapalino et al. 1998), an extended presence of monocytes after injury during secondary inflammation events has also been associated with endbulb formation and impaired axon growth (Gomes-Leal et al. 2005). This raises the caveat that differences in the growth of fibres following SPARC null or WT transplantation may arise directly as the result of SPARC expression in OECs, or may develop indirectly as a consequence of changes in immune system function.  The precise relationships between axon regeneration, SPARC, and the innate immune system await further investigation. 4.3.2 Spinal neuron populations are differentially affected by the absence of SPARC in transplanted OECs.  One difficulty associated with the use of DRG explants as a model for neurite outgrowth is the presence of many neuronal subtypes, which may respond differently to secreted factors produced by OECs, and some of which are dependent on different neurotrophins for their survival.  While proprioceptive neurons of the dorsal root require BDNF and NT-4/5 for survival, some of the remaining neurons are NT-3-responsive, and others require NGF (Acosta et al. 2001; Stephens et al. 2005).  The use of NGF in the initial culture of DRGs under baseline conditions almost certainly promotes the survival of certain neurons, e.g. nociceptive neurons, which would then be included in the outgrowth analysis (Patel et al. 2000).  In contrast, OEC conditioned media contains other neurotrophins, inducing a different proportional representation of these sensory neurons in the DRG explants, whose outgrowth would then be measured.  This 109   also complicates identification of the neuronal target(s) of OCM-SPARC.  Although this difficulty is somewhat removed from the transplantation scenario by only comparing WT and SPARC null OECs, it remains difficult to identify which spinal populations are responsive to SPARC in vivo, and what contribution these tracts make to the overall growth analyzed.  There was no difference in the overall growth of NF- or NST-positive fibres rostral, within, or caudal to the lesion sites of WT or SPARC null transplanted rats (Figure 4.5).  In contrast, growth of SubP- and TH-positive fibres was significantly impaired by SPARC null OEC transplantation (Figure 4.6).  This suggests that either these neuronal subpopulations do not contribute significantly to the total population of neurons in the spinal cord, that their contribution is below the threshold of detection enacted by this particular measurement protocol, or that many other spinal populations are alternately increased or decreased in their sprouting as a result of SPARC null OEC transplantation compared with WT.  An in vitro assay that could address the outgrowth properties of specific neuronal populations, perhaps by their purification, enrichment, or identification, would provide a useful means for identifying potentially responsive populations in vivo.  An additional advantage of this system would be the procurement of increasingly specific therapies for individual spinal tracts, allowing for the application of guidance cues to direct growing neurons towards their targets. 4.3.3 Functional compensation may account for subtle differences in the outcome of WT versus SPARC null transplantation  The robust ability of SPARC to promote outgrowth from DRG explants in vitro, and the involvement of SPARC in wound healing in other tissues, suggested that the transplantation of SPARC null OECs would substantially alter OEC-mediated repair to the injured spinal cord.  A number of contributing factors, as well as the multifactorial and enormously problematic nature of spinal cord repair, probably account for the limited differences observed between WT and SPARC null OEC treated rats.  SPARC expression in the nervous system and connective tissues is widespread, and ranges from Schwann cells, astrocytes, microglia, macrophages, fibroblasts, and blood vessel endothelial cells (Vincent et al. 2008).  Furthermore, SPARC becomes upregulated around a spinal cord lesion (Figure 4.2).  Because SPARC is secreted by some cells, and its expression is high at the lesion site, SPARC from sources other than OECs could compensate for a lack of OEC-derived SPARC.  This difficulty could be circumvented by the 110   transplantation of SPARC null or WT OECs into a SPARC null animal, particularly since there are no gross abnormalities in the development or maturation of this mouse strain (Norose et al. 1998).  A caveat to this experiment is the altered response to spinal injury observed in mice as compared with rats and humans.  Spinal cord injury in the rat and human typically results in haemorrhage, acute inflammation, necrosis of tissue, and the development of large fluid-filled cavities at the lesion site, a significant barrier to regenerating axons that is not present in this model of spinal cord injury in the C57/Bl6 mouse (Steward et al. 1999).  In C57/Bl6 mice, necrosis and cavity development are significantly diminished or non-existent, and the lesion site is completely filled by a connective tissue bridge (Steward et al. 1999).  Because syrinxes represent a major challenge to regenerating axons, and SPARC’s involvement in wound closure in other systems suggested its potential importance to the OEC property of cavity reduction, the use of a murine model lacking cavity formation presented many difficulties.  Furthermore, the potential for functional compensation for SPARC is likely even in SPARC null mice, given the high expression of SPARC-like 1 in the brain and spinal cord, at least in the embryo (Mothe and Brown 2001).  It is also unknown whether other functional compensations are made in OECs cultured from SPARC null mice, whether in SPARC-like 1 expression, or in the expression and secretion of other neurite outgrowth-promoting proteins. 4.3.4 DRG outgrowth model predicts secreted factors regulate neurite branching, whereas coculture regulates neurite fasciculation or elongation.  Differences in the outgrowth of DRGs treated with OEC conditioned media compared with NGF alone were profound; an expanse of neurite carpet was induced by the application of OCM or recombinant SPARC.  In contrast, relatively subtle effects were observed on fibre sprouting between WT and SPARC null OEC treated animals.  Differences in the magnitude of these effects can certainly be addressed by invoking the complexity of the lesioned spinal cord mileu.  The lesion site may present sufficient endogenous SPARC, or may abrogate the effects of OEC-SPARC through regulation of its activities once it is secreted, however, other interpretations are also possible.  The assumption that SPARC affects neurite outgrowth in general is partially a reflection of parameters that can be examined in a dorsal root ganglion explant assay; since the ganglion is grown as a cellular mass, this makes determinations of branching versus elongation difficult to assess.  The importance of these distinctions pertains to 111   the phenotype of neuron growth necessary for reconnection of spinal neurons following a spinal cord injury.  Since spinal tracts require long distance axonal elongation following injury, to form synapses with targets at a distance, yet the DRG outgrowth model used here indicates primarily a branching effect of SPARC on neurons, it is perhaps unsurprising that the effect of SPARC null OEC transplantation was to decrease elaboration of substance P or TH-positive neurites at the caudal and rostral lesion boundaries respectively, and not to impair long tract regeneration.  To provide in vitro data that is more predictive of in vivo outcomes, it may therefore be of use to consider not only which spinal neuron populations are affected by candidate factors, but also what kind of neurite outgrowth is achieved.  These in vitro approaches will hopefully generate relevant data for application to a variety of neurotrauma scenarios, as well as for spinal cord injury therapeutics, ensuring the appropriate growth and reconnection of injured neurons.  112   CHAPTER 5: AN ENRICHED CULTURE OF CORTICOSPINAL NEURONS CAN BE IDENTIFIED AND GROWN IN VITRO FROM THE POSTNATAL DAY 8 MOUSE  5.1 Introduction  The complexity of the injured spinal cord environment renders inquiry into neurite outgrowth and repair mechanisms difficult.  When OECs are implanted into this environment, this further obfuscates direct and indirect effects of resident glia, the immune system, and the resultant scar, on the growth of axotomized spinal neurons.  Because our understanding of OEC contributions to neurite outgrowth within the olfactory system and elsewhere is incomplete, and does not address the differential responsiveness of spinal neuron populations to OEC transplantation, this necessitates the adoption of a simplified assay system in which responses of an individual neuron population to OEC factors, both secreted and membrane-bound, can be assessed.  Several spinal neuron populations have demonstrated responsiveness to OEC transplantation following spinal cord injury, including corticospinal, coerulear, and raphespinal neurons, which sprout into, and in certain cases across the lesion site, when OECs are present (Lopez-Vales et al. 2007; Ramer et al. 2004a; Ramon-Cueto et al. 2000; Ruitenberg et al. 2005). Given the functional relevance of the corticospinal tract (CST) to motor recovery from human spinal cord injury, and its reported responses to OECs in vivo in rat models of SCI, this suggested the importance of understanding the cellular bases governing OEC-corticospinal neuron interactions through the use of a neurite outgrowth assay.  Because factors governing CST outgrowth in response to OECs can be more readily identified in vitro, this information could generate tract and neurite elongation-specific targets and therapeutics to mediate CST regeneration following spinal injury in vivo.    113   5.1.1 Anatomy, function, and development of the corticospinal tract Anatomy and function of the adult corticospinal tract in the mouse and human  The corticospinal tract of the adult rat or mouse occupies the ventralmost portion of the dorsal columns, contains approximately 140 000 fibres, each with a diameter of less than one micron, and originates from layer V neurons of the primary motor cortex (Figure 5.1A,B; Brown 1971). Following developmental outgrowth, a clear pathway of fibres emanating from the pyramidal neurons of layer V can be discerned, which descends via the internal capsule, to grow into the cerebral peduncle, basilar pons, and the medullary pyramids, where fibres decussate and enter the contralateral dorsal funiculus of the spinal cord in the rat/mouse, or the lateral funiculus in the human (Figure 5.1 C,D; Gianino et al. 1999).  Once fibres arrive at target segments, they course dorsolaterally to innervate the dorsal horn and gate sensory afferent input, as well as connecting directly with lateral motor nuclei (Fetz 1968).  CST fibres are therefore found in Rexed’s laminae IV, V, VI, and VII, with the majority of the pre-cruciate (motor) cortex projecting to lamina VII (Brown 1971; Fetz 1968). The CST in the human is responsible for fine fractionated movements of the distal limbs, and is involved in other coordinated voluntary movements such as walking, as well as descending control of afferent inputs, particularly nociception, gating of spinal reflexes, and autonomic control (Capaday et al. 1999; Iwaniuk et al. 1999; Lemon and Griffiths 2005).  In the mouse, involvement of the CST in gross motor function is less clear, and it appears to play a more circumscribed role in postural and fine distal limb activities (Brown 1971; Cheney et al. 1991).  Coordinately, CST damage in the human results in more pronounced changes in function than damage of the CST in mice or rats.  Destruction of CST fibres in the mouse results in minimal postural deficits and decreased contralateral forepaw coordination in grasping movements (Brown 1971; Lemon and Griffiths 2005).  In contrast, destruction of corticospinal tract fibres in humans produces hemiparesis and a permanent severe deficit in skilled hand control (Ropper et al. 1979). Development of the corticospinal tract  Corticospinal neurons are generated from progenitors of the germinal zone of the dorsal telencephalon, the subventricular zone.  After birth, corticospinal progenitors pause before 114   115   migrating along radial glia into the cortical plate, and settling into layer V of the mouse neocortex on about ED 15-16 (Figure 5.1A; Kriegstein and Noctor 2004; Molyneaux et al. 2005).  Extension of an apical dendritic tuft and a basal axon begins following correct cortical positioning (Figure 5.1B), and CST axons grow and pathfind through the mid and hind-brain (Figure 5.1C,D) to reach the pyramidal decussation or first two cervical segments of the spinal cord by around birth (Schreyer and Jones 1988).  By P2 of mouse development,  CST fibres traced by PHA-L injection into the sensorimotor cortex reach C8, and leading CST fibres grow to T7 by P4 (Figure 5.1E; Gianino et al. 1999).  Branching of CST fibres into the cervical gray matter is first observed at PD4, and axons continue to extend to T13 by PD7, and to L5 by PD9 (Figure 5.1E; Gianino et al. 1999; Hsu et al. 2006).  Extensive branching of axons into the spinal gray matter occurs two days after a spinal segment is initially innervated by pioneer axons, and branching at cervical and lumbar levels reaches its maximum at PD5-7 and PD 10-14 respectively (Figure 5.1E; Gianino et al. 1999; Gibson et al. 2000).  Thus the development of corticospinal neurons responsible for innervation of forelimb motorneurons, which originate in C5-T1, is essentially complete by postnatal day 7 (Alstermark and Ohlson 1999). Neurite outgrowth regulators of corticospinal tract development  Guidance of CST outgrowth during development is dependent upon mechanical (physical) mechanisms that are underwritten by many of the classical guidance molecules discussed previously, including NCAM, laminin, N-cadherin, L1, neurotrophins, semaphorin, and Eph/ephrin interactions (Joosten and Bar 1999).  Astroglial tiers of longitudinally-oriented processes parallel to the rostrocaudal axis, and lining the CST during its ingrowth into the spinal cord have been described (Joosten and Gribnau 1989).  To deflect the initial axonal outgrowth of pyramidal neurons from the pial surface, Semaphorin3A expression is highest at the brain surface and decreases towards the ventricular zone, propelling axons towards the internal capsule (Polleux et al. 1998).  Interestingly, the effect of Sema3A on CST dendrites is one of chemoattraction, as opposed to chemorepulsion, due to asymmetric localization of soluble guanylate cyclase to the apical (dendritic) portion of the neuron (Polleux et al. 2000), although the pathfinding of CST neurons in Sema3A deficient mice appears normal (Sibbe et al. 2007). Guidance of CST fibres through the midbrain and toward the hindbrain also seems to involve the secreted guidance molecules Sema3A and netrin1, which repel fibres toward dorsal deflections, 116   and induce contralateral decussation, respectively, as does L1 (Finger et al. 2002).   L1 is expressed by later-arriving CST fibres that follow pioneer axons, and is important for the correct outgrowth of these fibres, since L1-null mice exhibit a reduced number of CST axons, fibres do not project beyond cervical levels, and demonstrate axon guidance errors (Cohen et al. 1998). Once CST fibres reach the medullary-cervical junction, a further host of guidance molecules propel axons along the spinal cord.  PSA-NCAM is highly expressed on the outer axonal membrane of CST growth cones, although in vivo cleavage of PSA-NCAM by endoneuraminidase injection does not interfere with CST pathfinding, but delays or reduces collateral branching (Daston et al. 1996; Joosten et al. 1996).  A more profound deficit is observed in NCAM-deficient mice, where the CST exhibits severe hypoplasia and pathfinding errors such as failure of decussation, or ventral projection (Cohen et al. 1998; Rolf et al. 2002). Gradients of Wnt1 and Wnt5A also promote growth of CST axons longitudinally in the spinal cord, and expression of ephrin B3 along the midline prevents recrossing of the EphA4- expressing CST neuron axons (Charron et al. 2003; Kullander et al. 2001).  In vitro, other morphoregulatory molecules have been attributed with survival or outgrowth-promoting activities for embryonic or early postnatal corticospinal neurons and include: CNTF (Magal et al. 1993), NT-4, GDNF (Junger and Junger 1998; Junger and Varon 1997), and BDNF, among others (Giehl 2001).  Thus, although some axon growth and guidance mechanisms governing the outgrowth of the CST have been elucidated, it remains to be demonstrated whether reinstatement of guidance molecule expression following spinal cord injury can provide useful cues to damaged, adult CST axons. 5.1.2 Corticospinal tract responses to spinal cord injury  The responses of corticospinal neurons to injury are partially reminiscent of, but differ in several key aspects, from the responses of peripheral nerves to injury.  While motorneurons of the facial nucleus exhibit degeneration distal to an injury site, as well as some retraction of their cut axons, this is followed by robust increases in actin and tubulin mRNA expression, increased GAP-43 protein, and decreases in neurofilament proteins and their mRNAs, an expression pattern reminiscent of the developmental profile (Tetzlaff et al. 1991; Tetzlaff et al. 1988).  In contrast, corticospinal axotomies at the medullary pyramids do not result in changes to Tα1- tubulin, neurofilament, or growth-asssociated protein- 43 (GAP-43) expression; reinitiation of 117   high tubulin and GAP43 expression in injured corticospinal neurons requires that the injury occur no more than 200μm from the cell body, at the level of layer VI of the cortex (Giehl and Tetzlaff 1996; Tetzlaff et al. 1994).  Concommitant with failure of regeneration associated gene expression (RAG; Tα1-tubulin, GAP-43, actin), CST neurons also undergo cell death and somal atrophy following lesion close to the cell body (Giehl and Tetzlaff 1996).  While application of BDNF or NT-3 at CST neuron cell bodies is sufficient to eliminate cell death and increase CST sprotuing, and BDNF perfusion can reverse somal atrophy, regeneration is not observed (Giehl and Tetzlaff 1996; Hiebert et al. 2002).  Indeed, the CST is seemingly resistant to regeneration. Following corticospinal injury, corticospinal axons display pathological axolemmal permeability, impaired axonal transport, and axon retraction, although some collateral sprouting and axon extension can occur over long recovery periods (Hill et al. 2001; Inman and Steward 2003; Stone et al. 2004).  Regeneration of the CST remains a formidable obstacle to spinal cord repair. 5.1.3 Rationale and aims  In vivo, following a spinal cord injury, corticospinal neurons exhibit modest growth or functional recovery in response to OEC transplantation, or indeed in response to most therapeutics assessed to date, including antagonism of myelin-derived outgrowth inhibitors, application of growth factors, and cell-based therapeutic interventions (Blesch et al. 1999; Hiebert et al. 2002; Steward et al. 2008; Webber et al. 2007).  This recalcitrance to therapeutic interventions necessitates the development of more effective therapeutics to promote CST regeneration.  An in vitro assay of corticospinal neurite outgrowth would allow for the testing and development of these therapeutics, and for an investigation of those mechanisms employed by OECs to promote CST outgrowth in vitro, with the eventual goal of augmenting OEC therapeutic efficacy in vivo.  These objectives suggest that certain criteria be met by an in vitro assay, in order to maximize applicability of mechanism and therapeutics to the in vivo scenario. An assay of corticospinal neurite outgrowth was therefore developed that would: • Allow for the identification of corticospinal neurons in vivo and in vitro, using both transgenic and immunocytochemical markers, 118   • Permit culture of CST neurons from a postnatal time point or anatomical location from which CST neurons have reached their target level in the spinal cord and have formed synapses within the spinal gray matter, • Enrich for the culture of neuronal over glial cells, • Allow for the survival of CST neurons over time in culture • Be permissive for CST neurite outgrowth under serum-free, defined media conditions, in the absence of exogenously-added neurotrophins, and • Promote minimal outgrowth under baseline media conditions that would allow sensitivity for the detection of increases and decreases in neurite outgrowth in comparison with outgrowth achieved under baseline conditions.  5.2 Results 5.2.1 YFP expression in postnatal day 8 Thy1-YFP16JRS mice demarcates a population of corticospinal neurons.  Identification of CST neurons in vitro, from among the myriad of other cortical neuronal subtypes that include GABAergic interneurons, as well as the interhemispheric callosal projection neurons that also reside in layer V of the neocrotex, is crucial to an investigation of targeted therapeutics for corticospinal repair.  Few antigenic markers have been described that uniquely identify corticospinal neurons either in vivo or in vitro, and discriminate between corticospinal and other corticofugally-projecting neurons. Only protein 35, which labels corticofugal, but not cortically-projecting neurons of layer V (Stanfield and Jacobowitz 1990) uniquely identifies corticospinal neurons from other neurons of layer V.  Otherwise, a combination of developmentally-regulated markers such as ER81, a transcription factor expressed in multiple neuron subtypes of layers V (Yoneshima et al. 2006), Orthodenticle 1 (Otx 1), a transcription factor expressed in layers V and VI (Hevner et al. 2003), and COUP-TF- interacting protein 2 (CTIP2), a transcriptional activator, as well as, nonexhaustively, Fezf1, encephalopsin, Pcp4, S100α10, Crim1, Neto1, NetrinG1, and Igfbp4 must be used to discriminate neuron subtypes of layer V (Arlotta et al. 2005).  Because the use of multiple antigenic identifying markers for CST neurons in culture would restrict the outgrowth parameters 119   that could be investigated, this seemed a limited and unviable strategy for the developmental of a corticospinal outgrowth assay. More recently, ultrasound-guided fluorescent microsphere injection into the developing spinal cord has been described for the identification of cultured CST neurons from the P4 mouse by FACS (Ozdinler and Macklis 2006).  Although this protocol yields a corticospinal neuron- rich culture, this specialized technique was not available at the time this assay was developed. Furthermore, since FACS represents a significant cellular stressor, and the older, targetted neurons employed in this culture are more sensitive to perturbances, it seemed unlikely that these neurons would survive flow cytometry (Lobo et al. 2006); the majority of studies employing FACS for neuronal purification do so using tissue from embryonic or early postnatal mice (Catapano et al. 2001; Chiu et al. 1998; di Porzio et al. 1987).  Subsets of transgenic mice expressing spectral variants of GFP under the control of the Thy1.2 promoter have been described, some of which display remarkable specificity in the neuronal elements that are labelled (Feng et al. 2000).  The Thy1-YFP-16JRS transgenic mouse was generated  through injection of an exon 3 deleted Thy1.2 vector, which is required for expression in non-neuronal cells, driving enhanced YFP (eYFP), into fertilized oocytes (Feng et al. 2000).  The expression of YFP in this line has not been extensively characterized, although it resembles that of the Thy1-YFP-lineH mouse, in which YFP expression occurs in layer V pyramidal neurons of the cortex, as well as other neurons of the brain (J.R. Sanes, personal communication; Feng et al. 2000).  While the expression of YFP in the cortex is relatively similar in Thy1-YFP-16JRS and lineH mice, it is first apparent in the Thy1-YFP-16JRS mouse during embryonic development (ED 15-16), whereas YFP expression does not begin in the lineH mice until the second postnatal week (Jackson laboratory; Braun et al. 2006; Neumann et al. 2006). To confirm that YFP expression from Thy1-YFP-16JRS mice could provide a means for detecting CST neurons in vitro, YFP expression in vivo at PD8 was examined (Figure 5.2).  At P8, YFP is strongly expressed in the cell bodies of neurons residing in layer V of the cortex, as well as in subsets of neurons in the hippocampus, cerebellum, and in the projection neurons of the olfactory bulb, the mitral cells (Figure 5.2A).  YFP expression is very sparse and occurs in a small subset of fibres of the corpus callosum, and expression in the upper layers of the cortex is absent (Figure 5.2B).  In contrast, the medullary pyramids, an axonal projection of CST neurons, 120   121   are strongly YFP-positive (Figure 5.2B’).  Within the cortex, many cells within layer V express YFP, exhibit pyramidal-shaped cell bodies (arrowheads), and extend microtubule-associated protein 2-positive (MAP2) dendrites towards the pial surface (Figure 5.2C).  To determine whether all or a subpopulation of neurons in layer V express YFP, expression of corticospinal- specific transcription factors in YFP-expressing neurons was examined.  CTIP2, a transcription factors expressed only by corticofugal neurons of layer V that is required for correct fasciculation, outgrowth, and pathfinding of corticospinal neurons (Arlotta et al. 2005), is expressed in most YFP-positive cells of the cortex, with only 1-2 YFP-positive cells per 20x field that do not express CTIP2 (Figure 5.2D).  CTIP2 is also expressed in many layer V YFP- negative cells in the cortex.  Otx1 was similarly expressed by most YFP-positive cells, as well as by some cells that did not express YFP (Figure 5.2E).  In contrast, LIM-only 4 (LMO4), a transcription factor expressed by callosal, but not corticospinal projection neurons (Arlotta et al. 2005), was not expressed in cells also expressing YFP (Figure 5.2F).  These results suggest that YFP expression in layer V of the cortex of P8 Thy1-YFP-16JRS mice can be employed to identify CST neurons and to distinguish them from callosal projection neurons and interneurons that also inhabit this layer. 5.2.2 Development of protocols for the enriched culture of YFP-positive neurons  To culture an enriched population of CST neurons from the postnatal mouse, several strategies were employed to (1) correctly identify cortical regions corresponding to the forelimb region of the primary motor cortex in P8 mice, (2) increase neuronal survival during dissection and plating periods, (3) decrease glial contamination, and (4) increase neuronal survival while decreasing glial survival over the period of culture.  Stereotaxic identification of forelimb and hindlimb motor cortex in C57BL/6 postnatal mice has previously been performed by anterograde and retrograde tracing of the developing CST.  1,1′-dioctadecyl-3,3,3′,3′-tetramethylindocarbocyanine perchlorate  (DiI) retrograde tracing of the CST in P8 mice provided evidence for the stereotaxic localization of the primary sensorimotor cortex, and has been further confirmed using alternate tracers or functional impairment paradigms (Dijkstra et al. 1999; Rolf et al. 2002; Sibbe et al. 2007).  Anterograde tracing of the P7 CST, based on stereotaxic coordinates developed for adult mice (Paxinos 1998) and concomitant retrograde tracing of C7 projecting CST neurons has also been used to guide 122   dissection of forelimb sensorimotor cortex (Kamiyama et al. 2006).  Whole mounted and sectioned postnatal day 8 Thy1-YFP-16JRS mouse brains of 2 littermates from 5 breeding pairs were measured using stereotaxic apparati to confirm stereotyped sizes and scaling from other mouse strains and adult stereotaxic information (Joosten et al. 1992; Joosten and van Eden 1989).  The anteroposterior coordinates of the forelimb motor cortex were therefore defined within the second quarter of the dorsal view of the cortex, and mediolaterally in an area defined by its epicentre at 0.8mm lateral from the midline, measurements used in previous studies (Gianino et al. 1999; Kamiyama et al. 2006; Rolf et al. 2002). Dissection and dissociation conditions were optimized to increase neuronal viability, and were based upon studies undertaken in brain slice and culture applications.  These included the use of ice cold dissection and dissociation solutions described for the preparation of acute brain slices, in which artificial cerebrospinal fluid supplemented with glucose is employed (Armstrong and MacVicar 2001).  As well, gentle dissociation of tissue using papain, arrested with trypsin inhibitor and bovine serum albumin, and performed in glucose-rich, ion-controlled dissociation media, as has been very helpfully outlined for the preparation of dissociated cortical neurons in slice overlay assays, were used (Polleux and Ghosh 2002).  Initial cultures were established from P8 and 14 mice, and are summarized in Tables 5.1 and 5.3.  Because YFP-positive neuron survival was very poor from P14 animals, despite culture modifications, use of these animals was discontinued, and the further experiments performed only on tissue derived from P8 Thy1-YFP- 16JRS mice. Experiments summarized in Table 5.1 were undertaken to optimize P8 YFP-positive neuronal survival at and following plating based on alterations in media, substrate and plating density parameters. Plating of cell suspensions on a substrate of poly-D-lysine and laminin was initially used, since poly-D-lysine, as opposed to poly-L-lysine, is not metabolized by cells and can therefore support culture over extended period of time (Cannella and Ross 1987).  However, neither glial nor neuronal attachement was robust on this surface, and extensive cellular debris was present. Attachment was increased, and debris decreased, by changing the substratum to a combination of poly-L-lysine and laminin, and including a filtration step in the protocol. Initial experiments were also performed in the presence of NT-3, CNTF, or a combination of NT-3 and CNTF, neurotrophins to which CST neurons have demonstrated 123   responsiveness (Giehl 2001).  However, addition of these neurotrophins significantly altered outgrowth of YFP-positive neurons, either by increasing neurite elongation (CNTF), or increasing primary and secondary neurite branching (NT3), which could obfuscate the detection of these responses in treatment conditions.  It was therefore advantageous to identify neurotrophin-free, serum-free, defined media conditions that would support YFP-positive neuron survival and minimal outgrowth, without biasing YFP-positive outgrowth towards a particular morphology.  A combination of supplemented NeurobasalA media, with B27 supplement, that includes insulin, transferrin, progesterone, putrescine, selenium, thyroxine, and high glucose was used to support YFP-positive neuron survival, and is termed Base media (“Base”).  As well, since excessive glutamate can contribute to neuronal death through glutamate-induced excitotoxicity, dilutions of L-glutamine, as based on previous examinations of glutamate excitotoxicity in cultured cortical neurons, were added to baseline culture media to define optimal concentrations for neuron survival (Table 5.1; Carrier et al. 2006).  Finally, density conditions for initial cell plating were determined to decrease cell death, allow neuron survival, and distribute neurons at a distance from one another, so as to decrease the influence of paracrine factors on neuronal outgrowth.  A summary of these experiments is presented in Table 5.1, and defines baseline conditions that support the attachment, survival, and minimal outgrowth of YFP-positive neurons in culture. Although baseline media conditions supported YFP-positive neuron survival to five days in vitro, other cells, both neurons and glia, were also present in culture at this time.  To precisely dissect those direct mechanisms contributed by OECs to the outgrowth of CST neurons, it is necessary that other glia be eliminated or greatly reduced in culture.  However, because CST neurons from later postnatal ages are highly sensitive to physical perturbation, separation schemes such as FACS, for example, were untenable (Arlotta et al. 2005).  This difficulty in separating neuronal and glial components without the use of harsh treatments represented a fundamental obstacle to the establishment of this culture system.  As a result many experiments were directed toward an initial enrichment in surviving neurons over glia.  Coversely, since many astrocytes are present around a spinal cord injury site, the presence of some glia in the corticospinal neuron culture does not necessarily invalidate its use, but rather may be used advantageously in certain contexts. 124   Initial attempts to decrease glial contamination were undertaken on the basis of media conditions that would select for neuronal survival and glial death.  This was followed by treatment with antimitotic agents to selectively kill dividing cells, or by introducing differential density and size filtration using centrifugation through Percoll.  These experiments are summarized in Table 5.2.  Most glia require a ready supply of proteins, such as those present in fetal bovine serum (FBS), a common media supplement, to initiate or complete cell division (Freshney 2000).  Accordingly, media conditions in general did not include FBS, both because it would increase division of glia, and because the proteinaceous complement present in FBS is undefined, and thus could alter responses in testing conditions.  The antimitotic agents cytosine arabinoside (AraC), 5-fluorodeoxyuridine (FdU), and camptothecin were assayed for their abilities to decrease glial contamination in dissociated CST neuron cultures.  Although a low concentration of FdU was moderately effective at decreasing glial contamination (data not shown), addition of all of the antimitotics generally decreased cell number, regardless of cell type, as has been previously reported (Besirli et al. 2003; Bolon et al. 1993).  Since these antimitotic agents appeared to exert deleterious effects on neuronal survival and outgrowth, they were not included in further experiments. Mixtures of cells with different size or density properties can be separated on the basis of these properties through methods such as forward and side scatter parameters in flow cytometric applications, or through their different sedimentation rates through a viscous gradient, such as Percoll.  Percoll is a liquid suspension of colloidal silica particles that simultaneously forms size and density gradients upon centrifugation (Silverman et al. 1999).  Information is not available concerning the specific densities of cortical astrocytes and pyramidal neurons, nor their comparative sizes in dissociated culture, however, separation of nervous system cells, particularly neural progenitors or oligodendrocytes from other contaminants by Percoll density fractionation has been performed successfully for the purposes of cell culture (Barami et al. 2001; Grever et al. 1999).  Harvesting of fractions of dissociated P8 cortical cells that were centrifuged through varying Percoll gradients, and their subsequent plating, allowed for the identification of specific Percoll regions corresponding to astrocyte or neuron sizes/densities.  An enriched neuronal fraction was found as a cell pellet, the lowest fraction, of a 20 or 30 percent Percoll solution centrifuged at 200 times gravity (200g) for 15 minutes.  Furthermore, centrifugation of the same cell mixture through 50% Percoll at 200g for 10 minutes revealed an 125   enriched population of astrocytes, identified by plating and immunocytochemistry for GFAP, in a middle fraction that had not yet pelleted.  Further experiments using centrifugation of cell suspensions with Percoll coloured density marker beads revealed the density of astrocytes in phosphate buffered saline at between 1.052 and 1.064 G/mL, between the green and red marker beads, with neurons exhibiting higher densities, resulting in more rapid flow through Percoll gradients.  To obtain maximal separation of the neuronal and astrocyte Percoll density bands, trials of centrifugation at varying speeds, for varying times, and using differing Percoll concentrations were performed and analyzed with Percoll density marker beads.  Two Percoll concentration and centrifugation conditions were defined that yielded a neuronal enrichment over that observed without centrifugation.  Centrifugation through 50% Percoll at 200g for 15 minutes yielded a differential distribution of neurons and glia, where glia were concentrated in upper and middle fractions, whereas neurons were concentrated in the lower fraction.  If cell suspensions were instead passed through a 20% Percoll gradient centrifuged for 10 minutes at 200g, glia were concentrated in a middle fraction, and neurons enriched in the lower fraction, with an upper acellular fraction.  Since this condition yielded greater total cell output, these parameters were adopted for further experiments.  Through the optimization of culture conditions such as dissection protocols, density/size separation, substrate, and media parameters for corticospinal neurons from P8 mice, this established a repeatable, media-defined, neurotrophin-free, and neuronally enriched primary culture of CST neurons suitable for assessment of neurite outgrowth responses and mechanisms. 5.2.3 Postnatal day 8 YFP-expressing neurons extend processes over time in culture and respond to neurotrophins and neurite outgrowth inhibitors of corticospinal neurons  Following development of culture conditions that would enrich for neurons, up to seventy-five percent, after five days in vitro, it was important to determine whether (1) YFP- positive neurons could be visualized, (2) YFP-positive neurons would initiate or extend processes over time in vitro, and (3) whether the baseline conditions outlined above would allow for the sensitive detection of both increases and decreases in neurite outgrowth of CST neurons. Twenty-four hours after the initial plating of cells, YFP-positive cell bodies were clearly visible, with very robust endogenous expression of YFP in the cell body, and weaker labelling of processes, as was also the case in in vivo analyses (Figure 5.3D).  YFP-positive cell bodies could 126   127   also be immunoreacted with neuron-specific tubulin (NST), which confirmed their neuronal identity, and allowed greater visualization of the extension of 1-2 fine, unbranched processes after 1 DIV (Figure 5.3A).  After 3 DIV, YFP-positive, NST-positive neurons were visible with elaborated processes in comparison with those present at 1 DIV (Figure 5.3 B,E).  The YFP- positive cell body remained brightly YFP-positive, with insubstantial labelling of processes, whereas NST-immunoreactive processes were longer and more branched than those visualized at 1 DIV.  By 3 DIV, YFP-positive neurons typically exhibited 3-4 primary neurites, extending directly from the cell body, as well as infrequent secondary branching of neurites.  By 5 DIV, YFP-positive neurons could still be identified by endogenous YFP expression in their soma, and through larger calibre processes that were proximal to the cell body (Figure 5.3 C,F).  Although some increase in neurite growth is apparent between neurons after 3 and 5 DIV, this outgrowth is not as pronounced as the outgrowth that occurs between 1 and 3 DIV.   By 5 DIV, YFP-positive, NST-positive corticospinal neurons exhibited primarily multipolar morphologies, with four to six primary neurites extending from the soma.  Further elaboration of primary processes also occured between 3 and 5 DIV, when secondary and tertiary neurite branching became visible.  Responses of CST neurons to some neurotrophins and neurite outgrowth inhibitors have been characterized in vivo and in vitro (Giehl 2001; Giehl and Tetzlaff 1996; Schnell et al. 1994).  To determine whether baseline media conditions provided sufficiently stringent growth conditions so that increases in CST neuron outgrowth could be visualized, the known CST neurotrophins, CNTF and NT-3 (Schnell et al. 1994), were exogenously applied to baseline cultures of CST neurons over 5 DIV.  CST neurons express the receptors for NT-3, TrkC, and CNTF, CNTFRa, and in in vitro slice cocultures, NT-3 is capable of promoting the growth of CST neurons into the spinal cord (Kamei et al. 2007).  Furthermore, administration of NT-3 or CNTF at the site of a spinal cord injury can increase the growth of CST axons into the lesion site (Bregman et al. 1997; Schnell et al. 1994).  Following application of 25ng/mL CNTF for 5 DIV, numbers of CST YFP-expressing neurons were similar to numbers of the same cells under baseline conditions.  In contrast, large increases in neurite outgrowth were observed following CNTF application, with elaborate and significantly elongated processes present in YFP-positive, but not in small YFP-negative neurons (Figure 5.3 G).  Addition of 25ng/mL NT-3 to CST neuron cultures for 5 DIV resulted in increased numbers of YFP-positive neurons surviving in comparison to baseline treatments.  Small increases in neurite outgrowth were also observed, 128   although this occurred primarily as increases in the number of primary neurites, or increases in neurite branching, in comparison to the +CNTF condition (Figure 5.3 H).  To determine whether baseline media conditions would also allow detection of decreases in neurite outgrowth, the responses of CST neurons to the neurite outgrowth inhibitor, myelin associated glycoprotein (MAG), were also analyzed.  MAG is responsible for a portion of the neurite outgrowth inhibitory properties of myelin (McKerracher et al. 1994; Mukhopadhyay et al. 1994).  Chinese hamster ovary cells (CHO) that constitutively express MAG (Mukhopadhyay et al. 1994) were plated beneath cultures of CST neurons, which were allowed to grow for 5 DIV.  CST neurons exhibited significantly decreased outgrowth on CHO-MAG cells in comparison with either control CHO cells, or baseline media conditions alone (Figure 5.3 I). These results suggest that CST neurite outgrowth increases between 1 and 5 DIV, with the largest increase in growth between 1 and 3 days.  Furthermore, although baseline conditions are permissive for CST neurite outgrowth, they also allow for the sensitive detection of increases and decreases in neurite outgrowth. 5.2.4 Neuronal and glial contributions to dissociated corticospinal neuron culture change over time in vitro  To ascribe a mechanism to promoters or inhibitors of neurite outgrowth in an in vitro assay, it is necessary to account for the contributions of other cell types in the culture to the outgrowth of the neurons investigated.  Furthermore, addition of factors to the assay may act indirectly on contaminating glia to induce changes in their physiology that ultimately alter neurite outgrowth of those cells in contact, or even at a distance from CST neurons.  This effect has already been described as it pertains to the OEC secreted factor SPARC, which promotes the outgrowth of dorsal root ganglion neurons through  Schwann cell mediated mechanism (Au et al. 2007).  Furthermore, this effect has also been observed following application of CNTF versus NT-4 on purified or mixed neuron/glial CST neuron cultures; only CNTF, but not NT-4 was capable of increasing purified CST neuron survival, whereas the CST survival-promoting effects of NT-4 required the presence of other cell types (Junger and Junger 1998).  Numbers of glia, neurons, and corticospinal neurons changed significantly over 5 DIV (Figure 5.4A).  One day following initial plating of the CST neuron culture, cells were 129   130   predominantly only DAPI-positive, and only thirty-five percent of the total cell population was neuronal (Figure 5.4 A,B).  Of the neuronal population at 1 DIV, most neurons were YFP- negative.  By 3 DIV, the numbers of DAPI-positive cells were significantly decreased in comparison to 1 DIV, and represented slightly less than half of the total cell population (Figure 5.4 A,C).  Proportions of YFP-positive and negative neurons remained relatively constant between days 1 and 3 (Figure 5.4C), and there was no significant decrease in the raw numbers of these cells between these time points (Figure 5.4A).  At 5 DIV, there was a further significant decrease in the numbers of DAPI-positive glial cells, and these cells only represented one quarter of the total cell population (Figure 5.4 A,D).  In contrast, the neuronal population was significantly enriched by 5 DIV, amounting to seventy-five percent of the total cell population, despite the significant small decline in both NST-positive and YFP-positive cells between days 3 and 5 in vitro (Figure 5.4A,D).  These data suggest that neuronal enrichment under these conditions occurs between 3 and 5 DIV, although neurons are also reduced in number over this period.  This raises the possibility that cell death contributes to the observed neuronal enrichment, and that culture conditions to increase or decrease cell death may be used to optimize purification of the culture.  Alternately, the significant death of neurons from days 3 through 5 could also be used as an assay for survival factors.  However, since neuronal enrichment, even at 5 DIV, is not complete, contributions of endogenous glia to outgrowth and survival phenomena must still be considered. 5.2.5 Death of glia and neurons differs over time in culture  To more closely investigate the contribution of cell death of neurons and glia to population dynamics of the CST neurite outgrowth assay, the vital dye, propidium iodide (PI), which intercalates into DNA bases when cell membrane impermeability is compromised, was used to assess death of neurons and glia under baseline conditions over time in culture (Figure 5.5).  In glia, there was a significant reduction in total cell number between 1, 3 and 5 DIV (Figure 5.5 A).  Concomittantly, many DAPI-positive cells exhibited strong nuclear propidium iodide staining at 1 DIV, suggesting many of the cells that are in the process of dying at 1 DIV have completed this act by 3 DIV, resulting in a decrease in total DAPI-positive cells by this time.  This is also in accord with the significant reduction in PI-positive glia at 3 and 5 DIV in comparison with 1 DIV, although a further reduction in glial cells continues over these time 131   132   points.  In contrast, YFP-positive neuron numbers do not change between 1 and 3 DIV, but are reduced by 5 DIV (Figure 5.5B).  This is in accord with a peak of cell death in YFP-positive cells at 3 DIV which is reduced by 5 DIV in comparison to 1 DIV, a pattern that was mirrored by the NST-positive, YFP-negative neuron population.  These data suggest two waves of cell death in culture, one wave composed primarily of glia that die between 1 and 3 DIV, and a second wave of neurons that die between 3 and 5 DIV.  However, the proportional contribution of glia and neurons to the total cell death in culture over time is unequal; at 1 DIV, almost 70 percent of PI-positive cells are glia, whereas YFP-positive cells account for a much smaller 16 percent of the total cell death (Figure 5.5C).  Although there is a peak rise of YFP-positive death at 3 DIV to 26 percent, DAPI-positive death remains higher than neuronal death at all time points measured, suggesting that while culture conditions promote glial death, they are more supportive of neuronal survival. 5.2.6 YFP-expressing corticospinal neurons also express other corticospinal projection neuron markers in culture  To determine whether cultured YFP-positive neurons represented an enriched population of CST neurons and retained molecular characteristics of CST neurons following their culture, expression of the corticospinal-specific transcription factor CTIP2, deep layer cortical transcription factor, Otx1, and the callosal projection neuron marker, LMO4, were examined in YFP-positive neurons five days after initial culture (Figure 5.6).  Bright nuclear colocalization of CTIP2 with DAPI was clearly visible in most YFP-positive neurons, with only 7 percent of YFP- positive neurons that did not also express CTIP2 (Figure 5.6A-A’’’,B).  CTIP2 was also expressed in some YFP-negative neuron nuclei, and probably represents the population of CST neurons that were YFP-negative in vivo (Figure 5.2D).  Otx1 was similarly distributed in YFP- positive cells, where it was localized to the nucleus, and was expressed in 97 percent of YFP- positive neurons (Figure 5.6 C-C’’’, D).  Taken together with the results of CTIP2 expression in vitro, this implies that only a very small proportion of YFP-positive neurons identified in vitro are callosal projection neurons, while a majority retain a CST neuron phenotype.  This was confirmed through an analysis of the proportions of YFP-positive LMO4-expressing neurons. LMO4 was undetectable in most YFP-positive neurons, as is consistent with the population exclusive expression of LMO4 and CTIP2 (Figure 5.6 E-E’’’, F; Arlotta et al. 2005).  Taken 133   134   together, these results suggest that YFP-positive neurons grown from the Thy1-YFP-16JRS mouse, under these culture conditions, represent an enriched CST neuron population that survives and retains its molecular identity over 5 DIV.  5.3 Discussion 5.3.1 The Thy1-YFP16JRS mouse permits identification of corticospinal neurons in vivo and in vitro  Current and previous methods for the identification and culture of early postnatal or embryonic CST neurons have depended upon retrograde tracing, using classical tracers (e.g. Fast Blue) or fluorescent microspheres, followed by either mixed neuron-glial dissociated or slice culture, or FACS separation methodologies (Arlotta et al. 2005; Kamei et al. 2004; Ozdinler and Macklis 2006).  A simplified identification and separation schema relies upon the intrinsic identification of corticospinal neurons through reporter expression, such as that of the Thy1- YFP-16JRS line, or other transgenic reporter lines.  One limitation to the production or use of such a mouse line was (1) the relative scarcity, and paucity of data concerning uniquely identifying markers for corticospinal neurons, and (2) the age of onset of identifying markers. The Thy1-YFP-lineH transgenic mouse shows similar expression of YFP in CST neurons to the expression in Thy1-YFP-16JRS, and more efficient labelling of axonal and dendritic processes, however, since expression does not begin until P8-10, has not accumulated sufficiently on P8 for detection in vitro, and fewer corticospinal neurons are labelled, this renders culture of neurons from these animals significantly more difficult.  More recently, there have been several examinations of transcript expression to delineate profiles of CST neurons from other projection neurons residing in the cortex (Arlotta et al. 2005; Molnar and Cheung 2006).  Expression of CTIP2 and LMO4 (Ozdinler and Macklis 2006), which define CST neurons from callosal projection neurons that also reside in layer V of the cortex, were used to demonstrate that the majority of YFP-expressing neurons in the sensorimotor cortex of the P8 Thy1-YFP-16JRS mouse are CST neurons (Figure 5.2).  To confirm YFP-positive CST neuron identity, retrograde tracing of the CST tract in P8 mice, and anterograde tracing of dissected regions, could also be performed.  However, continued expression of CTIP2, Otx1, but not LMO4 in cultured YFP- 135   positive neurons after five days of culture strongly suggests that these are indeed cultured CST neurons, and not callosal neurons.  These data and this protocol thereby renders identification of CST neurons in culture relatively forthright, through the use of current stereotaxic coordinates (Dijkstra et al. 1999; Rolf et al. 2002; Sibbe et al. 2007), coupled with the continued expression of YFP in Thy1-YFP-16JRS-derived neurons in culture, as well as the early and more extensive onset of YFP-expression in this strain versus other strains. 5.3.2 YFP-expressing neurons display characteristics of corticospinal neurons in vitro  YFP-positive neuron cell bodies could be visualized in vitro after one day of culture, and extended a few very small calibre processes (Figure 5.3).  By three and five DIV, a multipolar pyramidal-like morphology emerged, usually with one thicker, longer neurite extending in one direction, and three or four smaller, branched neurites forming a tuft around the opposing direction.  This morphology is consistent with the description of CST neurons in vivo, where pyramidal neurons that project to the spinal cord have thick tufted apical dendrites, some of which extend to the pial surface, and a single large, modestly-branched axon stretching to subcortical regions and then the spinal cord (Molnar and Cheung 2006).  Responses of these neurons to CNTF and NT-3 also resembles previously reported effect of these neurotrophins on CST neurons.  Both CNTF and NT-3 selectively affected the outgrowth of YFP-positive neurons, without altering the growth of interneurons that were also present in the culture. Increased outgrowth following NT-3 administration is consistent with the effect of NT-3 delivery to CST neurons following spinal cord injury (Schnell et al. 1994); NT-3-expressing OECs transplanted into a CST transection resulted in increased numbers of anterogradely-traced CST fibres within and distal to the graft (Ruitenberg et al. 2005). The morphology of in vivo NT-3- treated CST growth, which exhibited branching rostral to and around the lesion site (Ruitenberg et al. 2005),  was also similar to the increased branching observed in vitro with NT-3 addition, although in vivo, this  pattern is probably also delineated by patterns of CSPG or myelin fragment deposition (Snow et al. 2001).  NT-3 also increased the number of YFP-positive neurons in culture at 5 DIV in comparison to baseline conditions, possibly through the pro- survival effects of NT-3 mediated by the TrkC receptor in CST neurons (Giehl et al. 2001). Responses of YFP-positive neurons to CNTF (Figure 5.3) also resembled previously- reported CST neuron responses to CNTF in vivo and in vitro.  In an in vitro slice assay of DiI 136   traced CST neurons, CNTF derived from bone marrow stromal cells increased CST axon growth, (Kamei et al. 2004), and CNTF infusion into the injured spinal cord can promote spinal sparing and increase motor recovery (Ye et al. 2004).  CNTF has also previously been reported to increase CST neuron survival in culture (Junger and Junger 1998), although this effect was not observed in the current experiments. Differences in the ages of cultured neurons, P2 versus P8, or the composition of basal media, could account for changes in the CNTF survival response, or detection of survival, between these experiments.  The outgrowth responses of YFP-positive CST neurons to MAG were consistent with those reports of CST outgrowth responses to myelin components in vivo following SCI (Li et al. 2004a), and could therefore provide a useful tool for the identification of CST outgrowth factors capable of overcoming MAG-induced inhibition.  These results are in contrast to the in vitro extension of embryonic dorsal root ganglion cells through adult myelin tracts (Shewan et al. 1995), although embryonic hippocampal and neonatal cerebellar neurons are inhibited by recombinant MAG in vitro (Li et al. 1996).  These differences in the in vitro responses of axonal populations to MAG may therefore reflect alterations in the responsiveness of embryonic and postnatal neurons, or dorsal root ganglion and corticospinal neurites, to myelin or MAG inhibition. 5.3.3 Glial and neuronal population proportions change over time in vitro- implications for outgrowth analysis  Initially upon culture, a majority of cells from P8 CST neuron culture are DAPI-positive, and not neuronal, whereas by 3 and 5 DIV, increased proportions of neurons prevail (Figure 5.4). This is due principally to a high number of dying, propidium-iodide positive, DAPI-positive cells over all time points measured in comparison to the much lower numbers of dying NST-positive or YFP/NST-positive cells (Figure 5.4, 5.5).  Interestingly, the peak of glial death occurs between initial plating and the first day in vitro, whereas the peak of YFP-positive or NST- positive death occurs at three DIV.   This suggests that either neurons and glial require different amounts of time to die, or that the death of glia contributes to neuronal demise.  This second hypothesis is probable given that astrocytes, and other glia, secrete factors known to support neuronal survival (Jordan et al. 2008).  A reduction in glia therefore reduces these factors, and may contribute to neuronal loss.  Two benefits arise from this observation: (1) reductions in glial 137   numbers and their secreted factors will sensitize this assay to outgrowth-promoting molecules whose effects might have been obscured or altered by astrocyte-derived factors, and (2) because neurons received reduced trophic support from endogenous glia, survival induced by exogenous factors can be assessed. Another reason for a decrease in the number of YFP-positive cells by 5 DIV could relate not to the death of some cells, but to decreased expression of YFP that becomes undetectable by 5 DIV.  Although YFP levels could change over time in culture, sufficiently to fall below detection thresholds by 5 DIV, these neurons would still be assessed in total NST-positive cell counts over time in vitro (Figure 5.4A).  Because the curves of declining NST-positive and YFP- positive cells between 3 and 5 DIV are extremely similar, and numbers of YFP-positive cells detected does not change between 1 and 3 DIV, this implies that decreased YFP detection does not contribute significantly to measured YFP-positive cell decline.  Furthermore, since both NST-positive and YFP-positive neurons decline to a similar extent between 3 and 5 DIV, YFP- positive neurons probably do not die because they are YFP-positive per se, and YFP protein levels are probably not sufficient to become neurotoxic. Ultimately, the enrichment of neurons over time in culture is useful in delineating direct and indirect effects of outgrowth factors, and in addressing their mode of action.  Furthermore, since factors secreted by glia could alter neuronal responses to exogenous factors, the decreased numbers of glia over time will allow for a more accurate analysis of neuron-specific outgrowth responses.  Conversely, given the continued presence of glia as well as interneurons even at 5 DIV, when assessing the effects of outgrowth modulators, the contributions of membrane-bound and secreted paracrine factors from these contaminant cells must necessarily be considered. Furthermore, since astrocytes would be present at a spinal lesion site, it could be argued that their presence in the PD8 corticospinal culture provides a biological relevance to this assay system. Given these caveats, the neuronally-enriched postnatal day 8 corticospinal neuron culture from the Thy1-YFP-16JRS mouse can serve as a useful tool for the assessment of corticospinal- specific outgrowth or survival factors, and a dissection of their mechanisms of action.  138   Table 5.1 Development of a postnatal day 8 corticospinal neuron culture: Experiments defining media, substrate, and density plating conditions. EXPERIMENT NUMBER DONOR AGE MEDIA CONDITIONS FILTRATION/ DENSITY CENTRIFUGATION SUBSTRATE PLATING DENSITY OBSERVED OUTCOME REPLICATE EXPERIMENTS 012306 012707 013006 PD 8 - Baseline - NT-3 +25ng/mL - CNTF  25ng/mL -NT-3 25 ng/mL, CNTF 25 ng/mL N/A Poly-D-lysine Laminin 140000 cells/cm2 - Few cells plating down in any condition, and mostly glial - Slightly ↑ neuron numbers in +NT3, + CNTF - Excessive debris 3 020106 020706 PD 8 - 1 μM L-Q - 1 μM L-Q - 0.25 μM L-Q - 0.25 μM L-Q - N/A - 40μm filter - N/A - 40μm filter Poly-D-lysine/ laminin 140000 cells/cm2 - Low numbers of adherent cells - ↑ neurons present in 0.25mM L-Q, decreased proportions of glia - No change in neuron/glial proportions with filtration - decreased debris with filtration 2 020106 020706 020806 PD 8 - L-Q 0.25μM - L-Q 0.25μM , NT-3 25ng/mL - 40μm filter  Poly-D-lysine/ laminin 140000 cells/cm2 - ↑ neurons and glia in +NT-3 condition, but relatively sparse cells 3 021606 021706  PD8 - L-Q 0.25μM - L-Q 0.25μM, NT-3 25ng/mL - L-Q 0.25μM, CNTF 25ng/mL - 40μm filter  Poly-D-lysine/ laminin 140000 cells/cm2 - ↑ neurons in +NT-3 condition - no change in neuron number in + CNTF  2 022706 030306 030606 PD 8 - L-Q 0.25μM - L-Q 0.25μM, NT-3 25ng/mL - L-Q 0.25μM, CNTF 25ng/mL - 40μm filter  Poly-L-lysine/ laminin 140000 cells/cm2 - Greatly ↑ numbers of adherent, surviving cells on PLL/laminin - ↑ neurons and glia in + NT-3, outgrowth is more branched - Increased neurite elongation  in + CNTF 3 033006 040306 041006 PD 8 - L-Q 2μM - L-Q 0.4μM - L-Q 0.25μM - L-Q 0.1μM - L-Q free - 40μm filter  Poly-L-lysine/ laminin 140000 cells/cm2 -↑ neuronal survival in 0.4, 0.25 and 0.1μM L-Q - ↓neuronal survival in either 1μM or L-Q free media - ↑Proportions of neurons:glia in L- Q 0.4μM,  L-Q 0.25μM 3 051906A 051906B PD 8 - Base, L-Q 0.4μM - 40μm filter Poly-L-lysine/ laminin -140000 cells/cm2 - 200000 cells/cm2 - 300000 cells/cm2 - 400000 cells/cm2 - 570000 - 3,4,500000 cells/cm2 result in excessive cell debris and no ↑ in neuronal density - ↑ cell death in high density conditions - 200000 cells/cm2 ↑neurons attached to substrate compared with 140 000 cells/cm2 2 139   Table 5.2 Development of a postnatal day 8 corticospinal neuron culture: Defining conditions for the enrichment of neurons.  EXPERIMENT NUMBER DONOR AGE MEDIA CONDITIONS FILTRATION/ DENSITY CENTRIFUGATION SUBSTRATE PLATING DENSITY OBSERVED OUTCOME REPLICATE EXPERIMENTS  032906 033006 PD 8 - L-Q 1mM - L-Q 1mM, AraC - L-Q 0.25mM - L-Q 0.25mM, AraC - 40μm filter  Poly-L-lysine/ laminin 140000 cells/cm2 - 1mM L-Q + AraC does not selectively increase glial division to target them for apoptosis in comparison to 0.25mM L-Q + AraC. - Addition of AraC decreases neuronal and glial survival regardless of L-Q concentration 2 052406A 052407B PD 8 - Base, L-Q 0.4mM - Base + 10% FBS - Base + 10% FBS - Base + 10% FBS - Base + 10% FBS - no filter - no filter - unfiltered portion on filter - filtrate - centrifuged filtrate, 800rpm, 1’ Poly-L-lysine/ laminin - 200000 cells/cm2 - addition of 10% FBS increases probability of contamination and increases glia in culture to a confluent monolayer - top, unfiltered portion is acellular - filtrate centrifuged or not yields similar proportions of cells - filtering alone does not contribute to neuronal enrichment 2 052906 053006 060606 PD 8 - Base, L-Q 0.4mM - 30% Percoll, 200g, 15’ - Harvest - Top3 mL, - Middle 3 mL, - Bottom 3 mL Poly-L-lysine/ laminin - 200000 cells/cm2 - Top layer is acellular - Middle layer 6/7 of total cells, cells are primarily GFAP+ - Only a 1/7 of cells reside in this layer, enriched NST+ neuronal component 3 053006 060606 PD 8 - Base, L-Q 0.4mM - 20% Percoll, 200g, 15’ - Harvest: - Top3 mL, - Middle 3 mL, - Bottom 3 mL Poly-L-lysine/ laminin - 200000 cells/cm2 - Top 3 mL acellular - Middle 3 mL contains few, large cells, stain positively for GFAP - Bottom 3 mL contains more cells than this condition in 30% Percoll; cells appear smaller under Trypan Blue, increased proportion of neurons - ↑total cell yield through 20% Percoll. 2  140   Table 5.2 Development of a postnatal day 8 corticospinal neuron culture: Defininf conditions for the enrichment of neurons (Continued)  EXPERIMENT NUMBER DONOR AGE MEDIA CONDITIONS FILTRATION/ DENSITY CENTRIFUGATION SUBSTRATE PLATING DENSITY OBSERVED OUTCOME REPLICATE EXPERIMENTS 061306 062106 PD 8 - Base, L-Q 0.4mM - Base, 10% FBS - Base, 10% FBS, AraC - Base, 10% FBS, 20mM camptothecin - 20% Percoll, 200g, 15’, bottom fraction harvested. Poly-L-lysine/ laminin - 200000 cells/cm2 - Base: glial contamination, but good neuronal survival - Base, 10% FBS:  ↑cell yield of GFAP+, NST+, but doesn’t ↑ YFP+ cells - Base, 10% FBS, AraC: drastically ↓ glia but also ↓ NST+ cells. ↑ reactive astrocytes. - Base, 10% FBS, camptothecin: Acellular, despite presence of cells at plating. 2 062706 062806 PD 8 - Base, L-Q 0.4mM - Base, AraC - Base, FdU - Base, camptothecin 10mM - Base, camptothecin 2mM - 20% Percoll, 200g, 15’, bottom fraction harvested. Poly-L-lysine/ laminin - 200000 cells/cm2 - Baseline conditions have neurons and glia, but far fewer than in +FBS previous conditions - AraC rids culture of many astrocytes, but few neurons survive, and are unhealthy looking with short processes - FdU: ↑neuron survival than with AraC, but still ↓ outgrowth compared with baseline.  ↓ glial contamination - Camptothecin: 2mM ↓ GFAP+ cells, but both conditions abolish plated neurons. 2 070506 070606 PD 8 - Base - 20% Percoll - 30% Percoll - 40% Percoll - 50% Percoll All 200g, 10’, Collect top, middle, bottom fractions Poly-L-lysine/ laminin - 200000 cells/cm2 - 20% Percoll: top acellular, middle mostly GFAP+, bottom neurons and glia - 30% Percoll: Top astrocytes, middle astrocytes, bottom acellular, - 40% Percoll: Top and middle GFAP+, bottom acellular - 50% Percoll: Middle is almost exclusively astrocytes, nothing in top or bottom. 2 141   Table 5.2 Development of a postnatal day 8 corticospinal neuron culture: Defininf conditions for the enrichment of neurons (Concluded)    071406 072606 080206 PD 8 - Base - 20% Percoll, 200g, 10’ - 50% Percoll, 200g, 15’ - Collect top, middle, bottom fractions Poly-L-lysine/ laminin - 200000 cells/cm2 - 20% Percoll: Top and Middle fractions astrocyte rich, neurons primarily in bottom fraction. Greater overall cell yield that in 50% Percoll. - 50% Percoll: Astrocytes located in top and middle fractions, neuronal enrichment in bottom fraction 3 073107 080106 080206 PD 8 - Base - 10% Percoll, 200g, 10’ - 40% Percoll, 200g 10’ - Collect top, middle, bottom fractions Poly-L-lysine/ laminin - 200000 cells/cm2 - 10% Percoll: enrichment in YFP+ neurons in bottom fraction, with significant glial contamination - 40% Percoll, less effective separation of astrocytes and glia than with 50% Percoll condition or 20% Percoll condition 3 142   Table 5.3 Development of a postnatal day 14 corticospinal neuron culture: Defining media for survival.  EXPERIMENT NUMBER DONOR AGE MEDIA CONDITIONS FILTRATION/ DENSITY CENTRIFUGATION SUBSTRATE PLATING DENSITY OBSERVED OUTCOME REPLICATE EXPERIMENTS 013006 020306 PD 14 - Baseline - NT-3 +25ng/mL - CNTF  25ng/mL -NT-3 25 ng/mL, CNTF 25 ng/mL N/A Poly-D-lysine Laminin 140000 cells/cm2 - Very few cells, neurons or glia present - Excessive debris 2 021606 021806  PD 14 - L-Q 0.25mM - L-Q 0.25mM, NT-3 25ng/mL - L-Q 0.25mM, CNTF 25ng/mL - 40μm filter  Poly-D-lysine/ laminin 140000 cells/cm2 - Very few cells in any condition - slight increase in neurons in + NT-3 - No identifiable YFP+ cells - Some cells stick to substrate but do not extend processes, even with CNTF 2 022006 022306 PD 14 - L-Q 0.25mM - L-Q 0.25mM, NT-3 25ng/mL - L-Q 0.25mM, CNTF 25ng/mL - 40μm filter Poly-L-lysine/ laminin 140000 cells/cm2 - Very few cells adherent to substrate - few processes in neuronal population, regardless of NT-3 or CNTF addition 2   Abbreviations and Notes for Tables 5.1-5.3: ‐ Baseline (Base) media is defined as Glutamine-free Neurobasal A (Invitrogen), 1X B27 Supplement (Gibco), 1% penicillin/streptomycin, 34mg/mL bovine serum albumin, 35mM D-glucose, L-glutamine 2mM , barring notations/additions delineated above. ‐ Substrate used is Poly-L-lysine (50μg/mL) OR Poly-D-Lysine (50μg/mL) and Laminin (10mg/mL), in the combinations noted above. ‐ AraC, cytosine arabinoside, an antimitotic used at 10μM. ‐ Camptothecin, a cytotoxic quinoline alkaloid; inhibits DNA topoisomerase to kill dividing cells, and is used at the concentrations indicated. ‐ FdU, 5-fluorodeoxyuridine, is an antimitotic agent that kills dividing cells, used at 10μM. 143   CHAPTER 6: OLFACTORY BULB ENSHEATHING CELLS PROMOTE CORTICOSPINAL AXON ELONGATION VIA A PLASMA MEMBRANE-DEPENDENT ACTIVITY 6.1 Introduction  The mechanisms responsible for repair of the injured spinal cord through the application of OEC transplants remain poorly appreciated, despite initial reports of OEC efficacy in this mileu nearly fifteen years ago (Ramon-Cueto and Nieto-Sampedro 1994).  While a number of indirect activities of OECs may allow for the growth of spinal fibres following their injury, including OEC-astrocyte interactions that modulate glial scar formation (Lakatos et al. 2003; Lakatos et al. 2000; Ramer et al. 2004a), modulation of Schwann cell infiltration and phenotype (Au et al. 2007), and induction of angiogenesis (Lopez-Vales et al. 2004; Richter et al. 2005), those direct outgrowth-promoting activities of OECs on spinal neurons are largely unknown. While the secretion of neurotrophins and regulators of neurite outgrowth such as NT-3, BDNF, artemin, and neurturin from OECs can promote neurite outgrowth of cortical neurons, retinal ganglion cells, and dorsal root ganglia in vitro (Au et al. 2007; Chung et al. 2004; Deumens et al. 2006d; Leaver et al. 2006; Lipson et al. 2003), it has not been ascertained whether OECs can specifically or differentially promote the outgrowth of corticospinal neurons, and whether this outgrowth is elongative or branched, axonal or dendritic (Deumens et al. 2006d).  Of particular importance are those mechanisms responsible for the regenerative growth of long tract ascending or descending spinal neurons, such as CST motor neurons, since these provide therapeutic targets for spinal cord repair. 6.1.1 Olfactory ensheathing cell mediated repair of the injured corticospinal tract in vivo In vivo, initial reports of OEC-induced anatomical repair of the injured CST (Li et al. 1997; Ramon-Cueto et al. 2000) have now been met by more sobering interpretations of the responses of CST neurons to OEC implantation (Deumens et al. 2006b; Lu et al. 2006).  While some have claimed CST regeneration (Li et al. 1997; Li et al. 1998; Ramon-Cueto et al. 1998), the extension of injured corticospinal neurons into and through a lesion site to reach more caudal targets, following OEC treatment with or without trophic factors, appears minimal (Lu et al. 144   2006; Ruitenberg et al. 2005).  Despite the incomplete lesions presented, numbers of biotin- dextran-traced CST fibres within or beyond the lesion site at either 3 weeks or 3 months are few, in comparison to the densely packed contralateral CST, and axon terminal structures resembling end bulbs are present (Li et al. 1997; Li et al. 1998).  Incompleteness of lesions, or a substantial lesion-induced rearrangement of motor programs is further suggested by the rapidity of functional recovery; ten days following OEC transplantation and electrolytic lesion of the CST, some rats exhibited directed forepaw reaching (Li et al. 1997), and at two months following OEC transplantation and complete transection at T10, movement of the hip, knee and ankle joints were visible (Ramon-Cueto et al. 2000).  These ameliorations in behaviour correspond to the time course over which a paucity of CST axons grew into or through the OEC transplanted lesion site. Furthermore, since motor control of the distal forelimb and hindlimb in the rat differs from the human, where the CST tract is paramount, and can be substantially assigned to the activity of the rubrospinal tract in the rat (Webb and Muir 2003; Whishaw and Kolb 1988; Whishaw et al. 1993), recovery of these limb behaviours could indicate increased, rubrospinal contributions to proximal limb muscles.  The persistence of spared CST fibres in these lesion studies  and differential contributions of CST and rubrospinal innervations to limb motor function, renders assignation of the cause of behavioural improvements difficult (Keyvan-Fouladi et al. 2003; Li et al. 2003a). More compelling evidence of OEC efficacy indicates that anatomical recovery of the CST is facilitated by multi-modal therapies; methylprednisolone and OEC transplantation significantly increases the number of CST axons caudal to a transection of the dorsal funiculus (Nash et al. 2002), and NT-3 expressing OECs transplanted into a CST transection increases anterogradely traced CST axons caudal to the lesion site (Ruitenberg et al. 2005).  In contrast, combinations of OECs or olfactory nerve fibroblasts (ONFs) aligned along biomatrix bridges appear minimally effective in promoting CST regeneration.  However this observation likely reflects the lack of OEC survival under this condition rather than their CST outgrowth-promoting effects per se (Deumens et al. 2006a; Deumens et al. 2006b).  Indeed, one activity of OECs on CST neurons consistent with their repertoire of secreted factors is the decreased apoptosis of these neurons after OEC transplantation into the thoracic spinal cord (Sasaki et al. 2006), an effect that has also been observed in vitro (Pellitteri et al. 2007).  This observation is surprising given that increased CST neuron survival or outgrowth in response to the direct intracortical 145   application of neurotrophic factors is substantially reduced following transection and factor application distal to the internal capsule (Tetzlaff et al. 1994).  Evidently, however, the survival of transplanted OECs is essential to their reparative activities.  If OECs survive in the hostile lesion environment, and are applied acutely, before the development of a gliotic scar, or with glial scar modifiers, much evidence now suggests that CST sprouting, electrophysiology, and behavioural function, can be moderately ameliorated (Fouad et al. 2005; Garcia-Alias et al. 2004; Keyvan-Fouladi et al. 2003; Nash et al. 2002). Despite evidence that the responsiveness of CST neurons to OEC transplants, and their recovery of function, seems to depend on the time between lesion and OEC implantation, distance between the lesion site and cell body, and the application of other factors, neurotrophins, chondroitinase, or otherwise, to stimulate different regeneration pathways or augment those stimulated by OECs (Fouad et al. 2005; Garcia-Alias et al. 2004; Ruitenberg et al. 2005), these combinatorial experiments also provide evidence that OECs may harbour certain unique CST outgrowth-promoting properties.  However, CST regeneration following OEC transplantation alone, or as the consequence of any individual therapeutic intervention, appears relatively poor.  This suggests the importance of investigating potential therapeutics for CST regeneration, as well as those strategies employed by OECs to ameliorate CST regeneration, in a simplified in vitro model. Co-opting of OEC-derived mechanisms for CST outgrowth promotion or regeneration may elucidate more effective therapeutics for the in vivo scenario of SCI. 6.1.2 Neurite outgrowth promotion by OECs in vitro: Secreted factors and cell-contact mediated mechanisms  The effects of OEC coculture or their secreted factors on a variety of peripheral and central neuron populations have been documented, although the precise mechanisms employed by OECs, or their activities on corticospinal neurons, have not been investigated.  Neurite length of olfactory receptor neurons (ORNs) plated on OEC  monolayers is increased over hippocampal glial coculture or laminin, an effect that is unchanged by physical separation of OECs and ORNs (Kafitz and Greer 1999).  Interestingly, under these conditions, the same effect was not observed on embryonic cortical neurons, where hippocampal glial coculture or laminin substrate alone were sufficient to increase neurite length over OEC coculture (Kafitz and Greer 1999).  While OEC coculture with retinal ganglion cells (RGCs) has also been reported to increase 146   neuritogenesis over astrocyte or Schwann cell coculture, application of their conditioned media did not increase the frequency of process-bearing RGCs over baseline media treatment (Sonigra et al. 1999).  Growth of RGCs as explants can also be enhanced by coculture with OECs, but not by addition of OEC secreted factors, Schwann cell coculture, or Schwann cell secreted factors (Leaver et al. 2006).  Neonatal cortical neurons similarly displayed increases in neurite length on purified monolayers of OECs over olfactory fibroblasts (ONF) or mixed cultures of OECs and ONFs (Deumens et al. 2006d).  Finally, OECs cocultured with hypothalamic neurons increase neuron survival and neurite outgrowth, although conditioned media exerts neither of these effects (Pellitteri et al. 2007).  These results are in contrast to the increased neurite outgrowth and branching observed in dorsal root ganglia explants or dissociated culture following application of concentrated OEC conditioned media (Au et al. 2007).  This effect may be due in part to either changes in outgrowth response to OEC conditioned media depending on neurite type, differences in the concentration or complement of secreted factors from LP or OB OECs, or on the effective concentrations of outgrowth/branching factors present in conditioned media samples. Secreted factors such as BDNF, NT-3, SPARC, neurturin, artemin, CNTF, or FGF produced by OECs (Au et al. 2007; Lipson et al. 2003; Santos-Silva et al. 2007; Woodhall et al. 2001) could certainly be responsible for the neuritogenic or neurite outgrowth promoting effects of these cells both in coculture or in their conditioned media.  Indeed, the effect of OEC-produced BDNF on adult retinal neurons has recently been assayed and contributes significantly to OEC outgrowth- promoting activities, since blockade of BDNF with a function-blocking antibody dramatically abrogates OEC-induced regeneration of these neurons in vitro (Pastrana et al. 2007).  However since these neurotrophins are labile and present in relatively low concentrations in unaltered OEC media, their effects may not be observed upon neat addition to neural cultures. Reports of preferential growth of olfactory neurons on OECs over other substrates such as laminin, HSPG, or CSPG, suggest that contact-mediated ECM or other cell surface-bound factors on OECs may also contribute to neurite outgrowth (Tisay and Key 1999).  Indeed, the direction of neurite outgrowth can be influenced by the direction of the OEC long axis (Deumens et al. 2004).  Experiments in vivo in which the ability of encapsulated OECs to promote spinal cord regeneration was assessed further underscore the importance of contact-mediated OEC factors in promoting neurite elongation;  encapsulated OECs, or OEC conditioned media was 147   less effective than injected OECs at promoting growth or halting retraction of lesioned CST fibres (Chuah et al. 2004). Despite the plethora of neurite outgrowth parameters analyzed by these in vitro studies, together they suggest that (1) contact of OECs with neurons is more effective in increasing outgrowth or branching than application of their secreted factors, and (2) OEC-induced neurite outgrowth, neuritogenesis, or neurite branching effects depend in part upon the type of neuron assayed.  These data, as well as our understanding of neurite outgrowth regulation by other central and peripheral glia, may inform an understanding of secreted and membrane-bound cues responsible for the neurite outgrowth promoting effects of OECs. 6.1.3 Astrocytes promote neurite outgrowth via secreted and surface-bound mechanisms  Although astrocytic responses to spinal cord injury include those of scar formation, an inhibitory barrier to regenerating neurons, the regulation and promotion of neurite outgrowth by astrocytes during development and in vitro, when they are non-reactive, has long been appreciated (Powell et al. 1997).  Secreted and membrane-bound factors are responsible for the promotion of neurite outgrowth by astrocytes (Hatten et al. 1984; Muller et al. 1984; Muller and Seifert 1982).  Given the similarities between OECs and astrocytes, an understanding of mechanisms governing astrocyte-mediated neurite outgrowth may provide further insight into the same processes in OECs.  Astrocyte conditioned media has long been employed as an enhancer of neuron survival in culture.  Furthermore, it is clear that secretion of neurotrophins by astrocytes can increase neurite outgrowth, branching, or synaptogenesis in vitro.  Survival of embryonic cortical neurons is increased by using astrocyte conditioned medium as a coating substrate (Wang and Cynader 1999),  and synaptogenesis is increased by the astrocyte-derived secreted factor, thrombospondin (Christopherson et al. 2005). EGF, FGF, and TGFβ are produced by astrocytes and may regulate neural differentiation and process outgrowth during development, and sympathetic neuron polarity is induced by astrocytic coculture, where neurons produce axons and dendrites, as opposed to solely dendrites (Rousselet et al. 1990).  Increasing evidence indicates that cell-bound molecules, as opposed to secreted factors, are primarily involved in regulating astrocyte-mediated neurite outgrowth and guidance (Fallon 148   1985; Hatten and Liem 1981; Hatten et al. 1984; Powell et al. 1997).  During embryonic and postnatal development, it has been well documented that oriented glial structures guide pioneering populations of spinal axons towards target structures, probably through the expression of membrane-bound cues for adhesion and repulsion (Silver et al. 1993).  Although astrocytes decrease in outgrowth-promoting ability with age in culture or over the life of the organism, in a manner similar to OECs (Au et al. 2007; Pastrana et al. 2007), both early postnatal and adult astrocytes retain the ability to promote neurite outgrowth (Smith et al. 1990).  Adult sensory neurons transplanted into either the intact or lesioned corpus callosum or dorsal columns regenerate parallel to the host tracts of mature astrocytes, suggesting a contact-dependent guidance mechanism present in mature and even reactive astrocytes (Davies et al. 1997; Davies et al. 1999).  In vitro, dissociated astrocytes form the most favourable substrate for cerebellar or spinal neuron neurite outgrowth in comparison to other neurons, and non-glia such as fibroblasts, heart muscle-fibroblast or meningeal monolayers; the adhesion hierarchy in which neurites grow singly on astrocyte monolayers but form fascicles on other substrates suggests a membrane- bound activity is responsible for the neurite outgrowth achieved on astrocytes (Noble et al. 1984). Since spinal neuron or cerebellar neuron outgrowth is indistinguishable on heat-killed, but not protein-denatured astrocytes, versus normal astrocyte monolayers, this further underscores the importance of a substrate or membrane-bound factor in regulating astrocyte-induced neurite outgrowth (Noble et al. 1984). Antibody blockade of NCAM and L1 significantly reduces neurite outgrowth of embryonic cortical and retinal neurons on immature, and to a lesser degree on mature, astrocytes, suggesting these compounds form a major component of astrocyte-bound outgrowth promoters (Smith et al. 1990).  As well, seeding of sensory neurons on cortical slices and their growth across the corpus callosum has been used to demonstrate the influence of fibronectin along callosal astrocyte tracts (Tom et al. 2004).  Other cell surface outgrowth promoting molecules produced by astrocytes, and similar to the complement produced by OECs, have also been reported and include deposits of laminin (Liesi and Silver 1988), and fibronectin on the surface of astrocytes (Liesi et al. 1986), N-cadherin, integrins, and in some cases, NCAM (Neugebauer et al. 1988; Tomaselli et al. 1988).  Together these data imply that astrocytes employ both secreted and membrane-bound factors to influence neurite outgrowth or survival; secreted factors appear to influence primarily neuron survival, branching or synaptogenesis, 149   while membrane factors such as cadherins, cell adhesion molecules, and integrins regulate neurite outgrowth. 6.1.4 Rationale and aims  In vivo, CST neurons have demonstrated varied responses to OEC transplantation from long tract regeneration to mild sprouting (Li et al. 1997; Lu et al. 2001; Lu et al. 2006; Ramon- Cueto et al. 2000).  The variable and limited functional recovery reported advocate for an understanding of underlying mechanisms responsible for OEC-mediated CST outgrowth.  To this end, the following aims were undertaken: • To use an in vitro assay of CST outgrowth to determine the responses of CST neurons to OEC coculture and compare this outgrowth with that observed under coculture with different glia. • By using immunocytochemical markers, to determine whether the CST outgrowth induced by OECs or other glia is elongative or branched, dendritic or axonal. • To assess the contributions of OEC secreted factors to dendritic or axonal outgrowth of CST neurons and assess their contributions to neurite elongation or branching. • To prepare plasma membrane-enriched fractions from OECs and assess contributions of membrane-bound factors to dendritic, axonal, elongation or branching phenotypes of CST outgrowth. • To determine whether plasma-membrane factors responsible for CST elongation are proteinaceous and to ascertain their capacity to overcome the neurite outgrowth inhibitor myelin associated glycoprotein. • To use similarities and differences in CST outgrowth phenotype provoked by OEC and astrocyte plasma membranes to dictate a microarray database and literature search for candidates responsible for OEC-mediated CST outgrowth.     150   6.2 Results 6.2.1 Differences in dissociated P8 Thy1YFP CST neurite outgrowth can be assessed using Neurobinary  To accurately assess differences in YFP-expressing CST neuron outgrowth in vitro following coculture or factor application, it was important to use a repeatable and sensitive measure of neurite growth that would allow for measurement of a variety of outgrowth parameters, including total outgrowth of each neuron, the number of primary, secondary, and tertiary branches of each neuron, and the length of each individual neurite.  Previously, we used the Northern Eclipse macro Neurobinary to assess the outgrowth of NST-immunoreacted dorsal root ganglion explants following treatment with OEC factors, since this method was more sensitive than measurement of the single longest neurite, average radius of the explants, or fluorescence signal (Au et al. 2007).  The output of this macro is presented in three examples in Figure 6.1, and is derived by loading raw images of YFP-positive neurons immunoreacted with anti-NST antibody, which are thresholded by Neurobinary to the same level across all samples (Figure 6.1A-C).  Extraneous information, such as immunoreactive particulate matter, is excised by the user, and therefore does not alter outgrowth measurements.  Each neurite is then reduced to one pixel width (Figure 6.1D-F), whereupon pixels are summed for each neurite, or summed for the entire neurite carpet of each neuron, to yield a measure of neurite length (Figure 6.1G). An additional dividend of this method is that the accuracy of neurite tracing can be ascertained, since each iteration of the macro yields a skeletonised image of the neuron that has been analyzed.  To confirm that this method would accurately and sensitively detect changes in neurite length, YFP-positive neurons from conditions in which small and large differences in outgrowth can be visualized were subjected to Neurobinary (Figure 6.1G).  Addition of the CST neurite outgrowth promoter CNTF yielded measurements indicating increases in total neurite outgrowth and increases in individual neurite length in comparison to CST neurons under baseline media conditions, which reflected the appearance of these neurons.  Culture of CST neurons over 5 DIV on monolayers of the neurite outgrowth inhibitory myelin-associated glycoprotein (MAG)-expressing Chinese hamster ovary (CHO) cells, yielded decreases in total neurite length, neurite branches, and individual neurite length, which were measured by Neurobinary.  This suggested that skeletonization by Neurobinary for the purposes of analyzing 151   152   neurite outgrowth of CST neurons can accurately detect increases and decreases in outgrowth in comparison to base media conditions. 6.2.2 Coculture of CST neurons with glia and fibroblasts increases neurite outgrowth in phenotypically distinct manners.  The majority of studies demonstrating increased CST regeneration following OEC transplantation have utilized OB OECs (Li et al. 1997; Ramon-Cueto et al. 1998), but little CST growth has been observed following LP OEC transplantation (Lu et al. 2006). Given these observations, I wanted to compare the outgrowth of CST neurons in vitro following coculture with these and other glia.  Furthermore, others have argued that a mixed cell population of olfactory fibroblasts and OECs provides better outgrowth in vivo than purified cultures (Li et al. 2005), therefore these, and other outgrowth interactions between CST neurons and glia were also examined.  Dissociated CST neuron-enriched cultures from P8 Thy1YFP-16JRS mice were produced as described in Chapters 2 and 5, on coverslips coated with PLL and laminin, or onto mitotically-inactivated glia and/or fibroblasts.  After 5 DIV, these cultures were fixed, subjected to immunocytochemistry for the detection of NST, images were captured of YFP-positive neurons and their NST-positive neurite carpet, and measurement of NST-positive outgrowth was ascertained by Neurobinary (Figure 6.2).  Culture of YFP-positive CST neurons under serum- free media baseline conditions for 5 DIV resulted in minimal outgrowth characterized by the extension of four to five primary neurites, with typically zero or one branch point per primary neurite (Figure 6.2A).  Growth of CST neurons on a monolayer of either olfactory fibroblasts (LP fibroblasts; Figure 6.2B), or a mixture of LP OECs and LP fibroblasts (Figure 6.2C) resulted in dramatically increased elongation of neurites in comparison with the  base media condition; although four or five primary neurites were also typically observed in these cocultures, neurites were increasingly elongated in LP fibroblast and even further in LP OEC/fibroblast mixed cultures, and appeared to branch only slightly more than base media treated neurons.  Coculture of neurons on a monolayer of astrocytes (Figure 6.2D) resulted in increased secondary and tertiary branching of neurites, a stellate morphology, and a slight increase in neurite length.  The most striking increase in neurite length per neuron and length of each individual neurite (individual neurite length) was apparent following coculture with OB OECs (Figure 6.2E), in which extensive elongation of neurites, enlargement of the cell body, and the formation of 153   154   numerous ruffled growth cone-like structures at the ends of neurites were visible, although neurite branching appeared only moderately affected.  Quantification of these changes in YFP- positive CST neuron outgrowth following coculture with these glia, LP fibroblasts, LP OEC/fibroblast mixtures, and OB OECs revealed that OB OECs most effectively increased total neurite length per neuron (Figure 6.3A).  Although each of these cocultures resulted in significant increases in neurite length per neuron, OB OEC coculture resulted in a further significant increase in neurite length per neuron over LP fibroblast and LP OEC/fibroblast coculture.  In these coculture conditions, however, increased total neurite length per neuron was not the result of an increased number of branches, since only astrocyte coculture significantly increased branching in comparison to base media treatment (Figure 6.3B).  The most prominent effect of CST neuron coculture with LP fibroblasts, LP OECs and fibroblasts, or OB OECs was to increase the average length of each individual neurite, i.e. to increase neurite elongation (Figure 6.3C).  Neurite elongation was increased to the greatest extent by OB OEC coculture. While sprouting of corticospinal neurons in SCI can sometimes contribute the formation of useful intraspinal circuitry, it can also lead to aberrant behaviour, sensation, or sometimes autonomic dysreflexia (Cameron et al. 2006; Kalous et al. 2007; McCouch et al. 1958).  CST regeneration following spinal cord injury therefore necessitates the elongation of cut CST axons to reach distal targets.  This suggests that therapeutics be designed, at least initially, to increase axon elongation while reducing branching.  Since OB OEC coculture most effectively met the parameters of neurite elongation and reduced/unchanged neurite branching, the axonal or dendritic outgrowth effects of this coculture were investigated further. 6.2.3 OB OEC coculture increases axonal elongation and decreases dendritic outgrowth compared with base media treatment.  To assess the contributions of axonal and dendritic outgrowth to the elongative neurite outgrowth response provoked in CST neurons by OB OEC coculture, immunoreactivity for microtubule-associated protein 2 (MAP2), or growth associated protein 43 (GAP43) was compared between base media and OB OEC cocultured CST neurons (Figure 6.4).  The high molecular weight MAP2 is a microtubule-associated protein enriched in the soma and dendrities of neurons which provides cross-linkages between microtubules, or microtubules and other cytoskeletal components (Bernhardt and Matus 1984).  MAP2 immunoreactivity in dendritic but 155   156   not axonal processes was confirmed for CST neurons in vivo using tissue sections from the Thy1-YFP16JRS mouse as previously described.  In vitro, following five days of culture under baseline conditions, neurons displayed NST-positive processes that were also MAP2-positive, indicating primarily dendritic outgrowth is promoted under these conditions (Figure 6.4 A,B).  In contrast, coculture with OB OECs resulted in a small proportion of outgrowth that was MAP2- positive (Figure 6.4 F), whereas a large proportion of the outgrowth was NST-positive and MAP2-negative (Figure 6.4C).  This suggested that MAP2-negative neurite outgrowth might be axonal and not dendritic.  To confirm that this was the case, GAP43 immunoreactivity was analyzed in baseline and OB OEC cocultured CST neurons.  GAP43 (B-50) is an indicator of outgrowth in CST neurons in vitro and in the developing pyramidal tract and is specifically localized to axons (Gorgels et al. 1987).  CST neurons treated with baseline media displayed little or no GAP43 immunoreactivity, whereas OB OEC cocultured neurons typically displayed a long unbranched, and sometimes branched GAP43 positive structure corresponding to the majority of the NST-positive neurite carpet (Figure 6.4 D,E).  Quantification of MAP2 and GAP43-positive neurites under base media, OB OEC coculture, or mixed LP fibroblast/OEC coculture demonstrated that MAP2-positive neurite length was unchanged between control media, and OB OEC coculture conditions, although LP fibroblast/OEC coculture significantly increased MAP2-positive outgrowth (Figure 6.5A,C).   Furthermore, GAP43-positive outgrowth was increased by OB OEC coculture over base media, and represented the majority of the NST- positive neurite (Figure 6.5 B,D), as opposed to base media treatment, where most of the neurite carpet was MAP2-positive and GAP43-negative (Figure 6.5 E).  These results imply that MAP2 or GAP43 immunoreactivity provide alternative and accurate means of identifying dendritic or axonal outgrowth, and that OB OEC coculture significantly increases axonal elongation over other treatment conditions. 6.2.4 Secreted factors from OB OECs do not contribute significantly to corticospinal neurite elongation  The stimulation of CST neuron axon elongation following OB OEC coculture in vitro was a highly favourable phenotype for application to the scenario of spinal cord injury.  An examination of the mechanisms responsible for this outgrowth was therefore undertaken.  Many secreted factors are produced by OECs that could influence axon elongation, including those 157   158   159   previously identified by ELISA, RT PCR, and our own ICAT proteomics dataset of the LP OEC secretome.  To assess the contributions of secreted factors to CST axon elongation, OB OECs, or control cells such as LP fibroblasts, or LP OECs and LP fibroblasts were cultured on transwell inserts suspended above CST neuron cultures (Figure 6.6A).  After 5 DIV, neurite outgrowth of CST neurons was measured in transwell and baseline conditions (Figure 6.6B-D), and showed a small but significant increase in individual neurite length and neurite branching in all cell treatments (Figure 6.6 G).  To determine whether increasing concentrations of OEC secreted factors could induce more robust increases in neurite outgrowth, OB OEC conditioned media samples were concentrated and standardized as has been previously reported (Au et al. 2007). Regular OEC conditioned media (Figure 6.6E) significantly increased neurite branching in a dose-dependent manner (Figure 6.6H), although in contrast to transwell treatment, it had no significant effect on the length of individual neurites (Figure 6.6 I).  When CST neurons are cocultured with OB OECs, this could alter the complement or quantity of secreted factors produced by OECs, thus influencing axon elongation.  To test whether the conditioned media from OB OECs cultured in contact with CST neurons could increase axon elongation, OB OEC conditioned media was harvested, concentrated, and standardized from OB OECs grown with an overlying culture of CST neurons.  This axon-stimulated conditioned media was then added to a separate CST neurite outgrowth assay (Figure 6.6F).  CST neuron-stimulated OEC conditioned media exhibited significantly more robust neurite branching activities than did regular OEC conditioned media, although it did not demonstrate an increased effectiveness in promoting neurite elongation (Figure 6.6 H,I).  Together these results suggest that secreted factors produced by OECs alone or when they are in contact with CST neurons do not contribute significantly to the neurite elongation observed following OB OEC coculture. 6.2.5 The effect of plasma membrane-derived proteins on corticospinal neurite outgrowth can be assessed using plasma membrane extraction and plating.  Because secreted factors from OB OECs did not contribute significantly to neurite elongation, but rather promoted neurite branching (Figure 6.6), it seemed reasonable that a factor bound to the plasma membrane of OB OECs could be responsible for CST neurite elongation. To test this hypothesis, an enrichment protocol for plasma membrane extraction was applied (Petit et al. 2005) and assessed for its ability to preserve membrane-bound activities (Figure 6.7). 160   161   CHO cells transfected and selected to constitutively express MAG (CHO-MAG) or its control insert (CHO-R2; no MAG control), were grown as a monolayer beneath CST neurons, and their effects on CST outgrowth were examined.  These cells were used because MAG is a plasma membrane-bound glycosylated protein whose inhibitory activities on neurite outgrowth depend on protein and ganglioside interactions (Cao et al. 2007b; Yang et al. 1996).  These cells and their plasma membranes therefore served as positive controls to test that plasma membrane extraction does not abolish the activities of either proteins or sugars.  CST neurons grown on a monolayer of CHO-R2 control cells exhibited a small but significant increase in total neurite outgrowth, but demonstrated significantly decreased outgrowth when cultured on CHO-MAG cells, in comparison to base media (Figure 6.7G).    Enrichment of the plasma membranes of CHO-MAG and CHO-R2 (control) cells was performed by physical disruption of the cell followed by sucrose step gradient ultracentrifugation (Petit et al. 2005).  To confirm that plasma membrane fractions were enriched in plasma membrane proteins, but depleted of cytosolic proteins, Western blotting was performed on samples of cells used to generated whole and plasma membrane fractions.  High levels of the plasma membrane ion homeostasis pump, Na+/K+ ATPase (Gorini et al. 2002) in plasma membrane samples, but relatively little Na+/K+ ATPase in whole cell fractions, contrasted with the high levels of the cytosolic glycolytic protein, glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Mazzola and Sirover 2003), in whole cell samples, with a depletion of GAPDH in plasma membrane fractions, suggested plasma membrane enrichment was achieved by this protocol (Figure 6.7D).  Plasma membrane fractions were then quantified and plated on substrate-coated coverslips (Petit et al. 2005). Outgrowth of CST neurons on CHO-R2 (control) and CHO-MAG plasma membranes at low concentrations (OD 0.1) was similar to CST neurite outgrowth under baseline media conditions (Figure 6.7 G).  In contrast, increasing concentrations of CHO-MAG plasma membranes were inhibitory to outgrowth in a dose-dependent manner, whereas increasing concentrations of CHO- R2 plasma membranes did not alter outgrowth (Figure 6.7G).  These data confirm that proteins and gangliosides in enriched plasma membrane fractions retain their neurite outgrowth activity. 162    163    6.2.6 The plasma membrane of OB OECs contains a proteinaceous neurite elongation factor for corticospinal neurons.  Since plasma membrane extraction and plating successfully retained the outgrowth inhibitory properties of MAG for CST neurons, this allowed for the testing of neurite outgrowth properties of OB OEC plasma membranes.  To address whether a factor in the plasma membrane of OB OECs could promote the elongation of CST axons, plasma membrane fractions from OB OECs or control CHO-R2 cells were extracted, quantified by optical density, and plated in varying concentrations, and their effects on CST neurite outgrowth were assessed.  CST neurons plated on a low optical density monolayer of OB OEC plasma membrane (OB OEC PM OD 0.1) or on CHO-R2 plasma membrane monolayers at the same concentration evoked similar amounts and extents of neurite outgrowth as when CST neurons were grown under baseline media conditions (Figure 6.8 G).  However, plating of increasing optical densities of OB OEC PM induced increases in neurite outgrowth, specifically, increases in GAP43-positive, MAP2- negative axonal elongation, over the unchanged outgrowth of CST neurons grown on increasing concentrations of CHO-R2 PM (Figure 6.8 A-D, G).  To ascertain whether the axon elongation activity present in OB OEC plasma membranes is proteinaceous, plasma membrane samples from OB OEC, CHO-R2 and CHO-MAG cells were subjected to trypsin protease digest, were plated onto coverslips, CST neurons were plated, and their neurite outgrowth assessed.  Since trypsin treatment of CHO-MAG plasma membranes abrogated the neurite outgrowth inhibitory effect of CHO-MAG plasma membrane, this suggested that trypsin treatment was sufficient to cleave much of the MAG protein responsible for outgrowth inhibition (Figure 6.8H).  The effect of trypsin treatment on plasma membrane was specific to cleavage of proteins, however, since treatment of CHO-R2 plasma membranes with or without trypsin did not alter the outgrowth of CST neurites plated onto these membranes (Figure 6.8H).  In contrast, treatment of OB OEC plasma membranes with trypsin significantly decreased the outgrowth of the CST neurons in response to OB OEC plasma membrane by significantly decreasing NST-positive, MAP2- negative axon elongation (Figure 6.8 E,F,H).  These results indicate the presence of a proteinaceous factor present in the membrane of OB OECs that has an axon elongation activity for CST neurons. 164   165   6.2.7 The plasma membrane corticospinal axon elongation activity of OB OECs shares some neurite outgrowth properties with an astrocyte plasma membrane-derived factor  To determine whether the CST axon elongation abilities of the OB OEC plasma membrane are unique to these cells, and to inform the search parameters of a microarray database comparing transcripts from astrocytes and OB OECs, the outgrowth activities of astrocyte and OB OEC plasma membranes were compared.   As previously described, OB OEC plasma membrane increases the total amount of CST neurite per neuron in a dose-dependent manner (Figure 6.9A).  Interestingly, culture of CST neurons on monolayers of astrocyte plasma membranes also increased total outgrowth of CST neurons in a dose-dependent manner, although the effect was much less pronounced than that observed with OB OEC plasma membranes (Figure 6.9A).  The neurite outgrowth induced by OB OEC and astrocyte plasma membrane was also similar in that the increased outgrowth was primarily MAP2-negative axonal outgrowth, and decreases in the proportion of MAP2-positive length with increasing plasma membrane concentrations were observed for both plasma membrane samples (Figure 6.9B).  Neurite branching and elongation were also assessed, to determine whether the phenotype of outgrowth induced by OB OEC or astrocyte plasma membrane was similar, and therefore might represent the activities of similar plasma membrane factors.  Branching of CST neurons was unchanged by treatment with OB OEC plasma membranes, regardless of concentration (Figure 6.9 C).  In contrast, astrocyte plasma membranes increased CST neurite branching, although not to the same degree as astrocyte coculture (Figure 6.9C).  Neurite elongation was affected by treatment with both OB OEC and astrocyte plasma membranes, although there were significant differences in the magnitude of this effect (Figure 6.9D).  Treatment with OB OEC plasma membranes increased individual neurite length over baseline by a maximum of 17 fold, whereas astrocyte plasma membranes increased individual neurite length to a maximum of 2.5 fold over baseline. These data indicated that while factors present in both OB OEC and astrocyte plasma membranes have a CST neurite outgrowth activity that promotes axon elongation, the astrocyte plasma membrane also has a dendrite branching activity.   166   167   6.2.8 OB OEC plasma membrane corticospinal axon elongation factors can overcome neurite outgrowth inhibition by MAG.  Although the discovery of an axon elongation activity for CST neurons in the plasma membrane of OB OECs provides (1) important information regarding the mechanisms used by these cells to promote CST outgrowth following SCI, (2) possibilities for specific therapeutic interventions to increase CST regeneration, and (3) suggests how the effectiveness of OEC transplantation might be improved by implanting cells at different locations, the injured spinal cord still represents an exceedingly hostile mileu for growing axons.  Many neurite outgrowth inhibitors are produced by astroglia surrounding the lesion site, or are the result of degeneration and tissue breakdown of the surrounding areas (Fawcett 2006).  Therefore, it was important to determine whether the factor or factors present in the OB OEC plasma membrane that evoke CST axon elongation could do so even when challenged by a neurite outgrowth inhibitor.  To this end, concentrated fractions of OB OEC and CHO-MAG plasma membrane were prepared, and coated as mixed preparations on coverslips with a concentration of CHO-MAG plasma membrane that was previously highly inhibitory to CST neurite outgrowth, and increasing concentrations of OB OEC plasma membrane (Figure 6.10).  When a low concentration of OB OEC plasma membrane was incubated with CHO-MAG plasma membrane of OD 1.0, CST neurons demonstrated decreased outgrowth in comparison to baseline media treatment, and the outgrowth observed was similar to that observed following treatment of CST neurons with CHO- MAG membrane alone (Figure 6.10).  In contrast, addition of OB OEC plasma membrane of OD 0.5 was sufficient to reverse the neurite outgrowth inhibition of MAG, and CST neurites exhibited increased total neurite outgrowth in comparison to baseline media.  The effect was slightly increased by addition of OB OEC plasma membrane of OD 1.0, although outgrowth on OB OEC mixed with CHO-MAG plasma membrane, both at OD 1.0, was still significantly less than the outgrowth observed on OB OEC plasma membrane alone (Figure 6.7).  The outgrowth of CST neurons on OB OEC plasma membrane with OD 0.5 and 1.0, mixed with CHO-MAG plasma membrane OD 1.0, still exhibited increased outgrowth of NST-positive, MAP2-negative neurites in comparison with baseline media, indicating axon outgrowth is still promoted by an OB OEC plasma membrane factor in the presence of an inhibitor.  However, the total axon outgrowth of OB OEC plasma membrane cultures grown with MAG was still significantly lower than when MAG was not present (Figure 6.10D,F).  As well, increased branching was observed 168   169   in these CHO-MAG inhibited cultures, in comparison to treatment with OB OEC plasma membrane alone, suggesting that while the OB OEC plasma membrane factor is sufficient to increase axon elongation under these conditions, this activity may be overcome when many outgrowth inhibitors are present.  Despite this caveat, the presence of an OB OEC plasma membrane-derived factor capable of overcoming neurite inhibition to promote CST axon elongation remains an exciting therapeutic prospect for the treatment of SCI.  In combination with interventions that decrease neurite inhibitory molecules in the lesioned spinal cord environment, the OB OEC-derived plasma membrane factors could specifically invoke the elongation of CST axons in vivo.  6.3 Discussion 6.3.1 Secreted and cell-bound OB OECs factors alter the phenotype of CST neuron outgrowth.  The impressive growth of CST axons on monolayers of OB OECs, in comparison to all other cell types tested, suggests the important therapeutic benefit of these cells for treating corticospinal neuron injury after SCI.  While both olfactory fibroblasts and LP OECs or fibroblasts alone did promote outgrowth from CST neurons, and coculture with astrocytes increased CST outgrowth by increasing branching, OB OEC coculture alone stimulated an extensive and robust axon elongation of CST neurons (Figures 6.2,3,4).  The importance of this type of outgrowth to recovery of function from spinal cord injury cannot be sufficiently stressed. CST neurons undergo extensive Wallerian degeneration and retraction following spinal cord injury in rodents and humans (Guleria et al. 2008); terminal processes of CST neurons can be found 5mm rostral to a moderate contusion lesion, and must extend for significant distances in the spinal cord before they can reach the lesion site to begin crossing it and reconnecting with distant targets (Hill et al. 2001).  It was for this reason that the mechanisms responsible for CST axon elongation produced by OB OECs were investigated. Previously, several individuals have investigated the abilities of OECs to induce neurite outgrowth in a variety of different neurons, including dorsal root ganglia (Au et al. 2007), hypothalamic neurons (Pellitteri et al. 2007), retinal ganglion cells (Leaver et al. 2006), and cortical neurons (Deumens et al. 2006d).  Mechanisms investigated for the axon elongation, 170   neurite branching, or outgrowth observed following coculture of OECs with neurons have been primarily based upon OEC production of secreted factors (Lipson et al. 2003; Pastrana et al. 2007; Pastrana et al. 2006; Sonigra et al. 1999); to date no plasma membrane-bound factors derived specifically from OECs have been investigated as contributors to their neurite outgrowth properties. The current experiments indicate that for CST neurons, secreted and membrane- bound OEC factors contribute to significantly different outgrowth phenotypes (Figures 6.6,8). Application of OEC secreted factors to CST neurons either continuously, by separating OECs and CST neurons in a transwell experiment, or by the addition of concentrated OEC conditioned media to CST neurons, resulted in increased neurite branching, and a slight increase in total outgrowth (Figure 6.6).  Application of OEC secreted factors by transwell also increased individual neurite length slightly, whereas OEC conditioned media did not change this parameter.  This result could indicate (1) the need for continual secretion of OEC factors to affect neurite elongation, or (2) that some OEC secreted factors are very sensitive to degradation, and do not survive concentration and freezing.  A final possibility is that factors secreted by OECs may have been changed or increased by culture of OECs with CST neurons.  However, addition of axon-stimulated OEC conditioned media did not induce an axon elongation phenotype in CST neurons, but rather primarily increased neurite branching (Figure 6.6).  While any of these possibilities may have limited the response of CST neurons to secreted factors,  the magnitude of neurite elongation observed upon coculture or plasma membrane culture still suggests that OEC secreted factors do not contribute significantly to axon elongation by themselves.   This conclusion is supported by similarities in the outgrowth phenotype of other neurons treated with OEC secreted factors.  When DRG explants or dissociated cultures are grown with the addition of OEC conditioned media, the increased outgrowth appears to stem primarily from an increase in neurite branching, or an increased “neurite carpet density” (Au et al. 2007).  Hypothalamic neurons grown on OEC monolayers also exhibit increased neurite length, but OEC conditioned media application yields shorter neurites (Pellitteri et al. 2007). Other investigations have indicated no effect of OEC conditioned media on neurite outgrowth of cortical neurons or sympathetic ganglia (Lipson et al. 2003; Sonigra et al. 1999).  This is in contrast to the increased dendritic branching of CST neurons observed here after application of OEC secreted factors, and may indicate partially how different types of neurites respond to either OEC secreted or membrane factors.  Some alteration in the responses of different neurites may 171   also stem from differences in contaminating glia present in culture. The increased DRG branching observed after OEC conditioned media treatment was primarily stimulated by SPARC in OEC conditioned media, which relies upon the presence of Schwann cells in the DRG explants (Au et al. 2007).  SPARC from OB OECs could act in a similar manner on CST outgrowth, although via alternative mechanisms, since Schwann cells are absent.  Alternatively, OEC secreted factors other than SPARC could increase CST branching.  In vivo, branching of CST neurites could result in both favourable and unfavourable outcomes.  Sprouting and lesion-induced plasticity resulting in rearrangement of motor programs or their inputs can yield improvements in behavioural outcome (Brus-Ramer et al. 2007). Reorganization of CST outputs can occur at the level of the cortex or in the spinal cord, and can contribute to recovery of function (Endo et al. 2007; Kim et al. 2008).  In contrast, limiting the sprouting responses of lesioned facial neurons, by transplanting OECs which promote process outgrowth but not sprouting, can improve regeneration and functional outcome (Guntinas- Lichius et al. 2002).  Furthermore, spurious sprouting of corticospinal neurons can also contribute to spasticity and neuropathic pain, suggesting a necessity to limit its occurrence following OEC transplantation (McCouch et al. 1958; Woolf et al. 1992).  Since OEC secreted factors contribute to the sprouting or branching responses of CST neurons, they may also adversely affect pain and autonomic pathways.  One qualification of this assertion is, however, that when OECs are in contact with CST neurons, the activity of membrane proteins outweighs the outgrowth activity of secreted factors; elongation is induced, rather than neurite branching. As long as OECs remain in contact with axotomized CST neurons in the lesioned spinal cord, the current investigations suggest that CST elongation will be promoted, and not branching. This assertion is corroborated by in vivo evidence of increased collateral branching of injured CST axons following treatment with encapsulated OECs or injection of OEC conditioned media into the spinal cord versus injection of OECs directly into the cord, where sprouting is lessened (Chuah et al. 2004). The observation of strikingly different CST outgrowth phenotypes resulting from OEC secreted and plasma membrane factors, and the similarity of CST axon elongation phenotype in OEC plasma membrane and coculture conditions suggests a balance between neurite branching and neurite elongation.  Neurite elongation or branching appear to occur each at the expense of 172   the other, perhaps as the result of limited and shared resources between elongation and branching neurite outgrowth pathways.  Addition of neurite length in both processes requires actin and microtubules, and the activity of actin polymerizing, and microtubule contractility inhibiting, enzymes. Therefore, when signals derived from the plasma membrane of OECs initiate axon elongation, outgrowth resources are directed away from branching processes.  Furthermore, recruitment of specific actin nucleating proteins such as actin-related protein 2/3 (Arp2/3) to growing neurites is thought to underlie neurite branching (Rouiller et al. 2008).  Activities of secreted or plasma membrane OEC factors could act antagonistically to recruit or inhibit Arp2/3 respectively, thereby facilitating neurite branching or elongation.  Therefore, in coculture conditions, significant inhibition of the Arp2/3 complex by an OEC plasma membrane protein could redistribute the effects of OEC secreted factors towards elongation; since actin branching is inhibited, all outgrowth signals derived from secreted and membrane cues serve to increase neurite elongation.  This process could also underlie differences in CST elongation observed between OEC and astrocyte plasma membrane and coculture experiments. 6.3.2 OB OEC and astrocyte plasma membrane factors share an ability to increase CST outgrowth in similar manners  The effect of coculturing OECs or astrocytes with CST neurons was significantly different; CST axons elongated extensively on OB OECs, but branched profusely on astrocytes (Figure 6.2).  This is in contrast to the effect of culturing CST neurons on either the plasma membranes of OECs or astrocytes, where neurite elongation predominated.  These results are suggestive both of the commonality of neurite outgrowth factors present in the membranes of OB OECs and astrocytes, and of the balance between neurite elongation and branching as a result of secreted versus contact-dependent cues.  For CST neurons, neurite outgrowth on OB OEC monolayers is predominantly elongative, the effect of OB OEC plasma membranes is robust and elongative, and OEC secreted factors induce modest neurite branching.  This implies that in coculture, factors in the OB OEC plasma membrane are more efficacious at promoting elongation than the branching factors that are being secreted.  This is perhaps not surprising, given the magnitude of the neurite elongation produced by OB OEC plasma membranes (37 fold increase), versus the much smaller increase in neurite branching produced by OB OEC secreted factors (1.5 fold).  A different scenario is 173   presented by astrocyte coculture with CST neurons, where neurite branching prevails.  Neurite branching from astrocyte coculture is increased by approximately two fold, whereas elongation from astrocyte plasma membranes increases neurite elongation by 2.5 fold and neurite branching by 1.3 fold.  In the case of astrocytes, contact-dependent and secreted factors are in relative balance in their abilities to effect neurite elongation or branching, and indeed, when both are present (in coculture), signalling for neurite branching prevails.  Therefore, although different outgrowth outcomes occur as a result of coculture with astrocytes or OB OECs, an overarching similarity in the activities of their secreted (branching) versus membrane-bound (elongative) factors can be gleaned.  This shared mechanism of branching or elongation from the secreted or plasma membrane factors of astrocytes and OB OECs might be expected based on similarities in the locations and cell-cell interactions of these glia.  Both OB OECs and astrocytes interact to form the glia limitans at the border of the PNS and CNS, and participate in the growth of neuronal processes between these two regions.  Furthermore, the interactions between astrocytes and OB OECs in the nerve fibre layer of the olfactory bulb appears cooperative, a relationship that is mirrored by their interdigitation with each other in culture (Lakatos et al. 2000).    The ability of OECs to interdigitate with astrocytes in vivo was most impressively demonstrated by experiments in which Schwann cell conduits transplanted into a transected spinal cord received OEC-capped ends, thereby altering the outcome of Schwann cell transplantation from one with a sharp glial scar delineation, which denied axons penetration into the distal cord, to a more permissive, interdigitated transplant-host border (Ramon-Cueto et al. 1998).  These interactions between OECs and astrocytes suggest that they may express similar or complementary profiles of cell-cell interacting machinery (e.g. cadherins) at the cell surface that may underlie adhesion to each other as well as regulation of neurite outgrowth.  Commonalities between astrocytes and OB OECs in their cell biology, and most importantly, in the phenotype of CST outgrowth induced by exposure to their plasma membranes, suggests that these glia may share one or many contact-dependent CST outgrowth regulators in their plasma membranes, although expression levels may differ.  These observations may reflect the similarities in transcript expression between astrocytes and OECs 174   that have already been outlined by a microarray database, which could aid in the identification of other plausible OEC-derived CST elongation candidates (Vincent et al. 2005).  NCAM, cadherins, intergrins, L1, and cell-bound ECM on astrocytes have all been implicated in the regulation of neuron outgrowth on astrocyte monolayers.  Treatment of reactive astrocytes with antisense GFAP-mRNA simultaneously increased neurite outgrowth on astrocyte monolayers, and increased their laminin production, whereas antibody blockade of astroglial laminin decreased cortical neurite outgrowth (Costa et al. 2002).  Furthermore, decreased activity of matrix metalloprotease 2 (MMP2), an ECM degradative protease, and upregulation of its inhibitor, tissue inhibitor of metalloproteinase 2 (TIMP2), in antisense-treated astrocytes also increased laminin bioavailability (Costa et al. 2002).  L1 also partially underlies the ability of astrocytes to regulate outgrowth of corticospinal neurons.  Overexpression of L1 under the GFAP promoter (in astrocytes) results in CST axon defasciculation, faster axon outgrowth, and increased collateral branching (Ourednik et al. 2001).  In vitro, L1 as a coated substrate increases neurite outgrowth of DRG, cerebellar granule, and hippocampal neurons to a much greater degree than poly-D-lysine or fibronectin (Webb et al. 2001).  Since L1 is expressed by OECs in the olfactory system and appears to partially underlie their ability to promote outgrowth of ORNs (Doucette 1990; Miragall et al. 1989), L1 may perform similarly in OEC interactions with CST neurons.  Growth of neurons on astrocytes also depends on interactions between cadherins and integrins on the surfaces of neurons and astrocytes (Neugebauer et al. 1988).  Application of function-blocking antibodies to N-cadherin, β1-integrin, or NCAM signficiantly reduces the growth of retinal ganglion cells or cortical neurons on astrocyte monolayers (Neugebauer et al. 1988; Smith et al. 1990). Blockade of N-cadherin and β1-integrin simultaneously in cultures of ciliarly ganglia growing on astrocyte monolayers virtually abolished neurite outgrowth, suggesting these factors are necessary components of astrocyte-regulated neurite outgrowth (Tomaselli et al. 1988).  Because some OECs express N-cadherin (Chuah and Au 1994; Fairless et al. 2005), as well as cadherins 3 and 11 (Akins et al. 2007), and β1-integrin (Au and Roskams 2003), these adhesion molecules could play essential roles in OEC promotion of CST elongation, as they do between astrocytes and other neuronal populations.  Other known, unknown, or untested proteins derived from OEC or astrocyte plasma membranes could also contribute to CST elongation activities of these cells.  To generate a 175   preliminary list of these components, a previously generated microarray dataset comparing transcripts from cultured OB OECs, astrocytes, and Schwann cells, was analyzed (Vincent et al. 2005).  This dataset was constructed using long (50mer) amino-modified oligonucleotides representing the Pan Rat 5K array printed onto Creative Chip oligo slides, where the signal intensity acquired from GenePix Pro was subjected to normalization using the Lowess regression, and each chip was centralized to 1.0, returning 1841 transcripts (Vincent et al. 2005). To analyze potential contributors to OEC or astrocyte-mediated plasma membrane outgrowth, a one-way analysis of variance without multiple testing correction was performed, returning 342 unique and known transcripts, where transcripts with one or more missing datapoints or datapoints outside the signal threshold were excluded, as has been previously described (Vincent et al. 2005).  Transcripts were then assessed for their presence in OECs and astrocytes, and their fold difference in expression between these cells.  Since OECs produce greater CST axon elongation at the same plasma membrane concentration as astrocytes, it follows that if the factor(s) responsible for this outgrowth are the same between these cells, they must be relatively enriched in the OEC plasma membrane in comparison to the astrocyte plasma membrane.  It is also possible that different proteins are responsible for this effect in astrocytes versus OECs, however, the assumption of similar proteins (1) can be hypothesized based on the similarities in outgrowth phenotype induced by the plasma membranes of both cells, and (2) provides a useful starting point for analysis.  OECs and astrocytes share many transcripts that regulate neurite outgrowth, including those that are secreted and membrane-bound.  Many cell adhesion molecules are enriched in the OEC plasma membrane and less so in the astrocyte plasma membrane, including L1, cell adhesion molecules with homology to L1, melanoma cell adhesion molecule, Mss4, an α integrin interacting Rab GTPase, and ephrin A2.  Some recent evidence indicates highly interdependent interactions of fibronectin, with alpha subunits of integrins α3A, α6A, α7A, and α7B and Mss4 (Knoblauch et al. 2007).  This interaction drives the coordinated presence and activation of the integrin/Mss4 complex to lamellipodia and filopodia at points of cell-cell contact and cell-matrix contact (Knoblauch et al. 2007).  This potential activation and concentration in the OEC plasma membrane could regulate CST outgrowth in a similar manner, by facilitating process attachment.  As well, melanoma cell adhesion molecule, also known as 176   gicerin, whose transcript is enriched by approximately six-fold in OECs over astrocytes (Vincent et al. 2005), has been implicated in neurite outgrowth in a variety of contexts.   Gicerin is expressed in retinal ganglion cells only when they extend neurites to the optic tectum, where it is also expressed (Taira et al. 2004).  Gicerin homophillic interactions can promote neurite extension from ciliary ganglion neurons in vitro, however heterophilic interactions of gicerin induce increased elongation of these neurites, over branching (Taira et al. 1998).   Although gicerin is highly expressed in the developing nervous system, it is generally downregulated in the adult animal, although its expression is reinstated in regenerating epithelia (Tsukamoto et al. 2001).  These contact-mediated regulators of neurite outgrowth, including L1, Mss4, integrins, or gicerin, could contribute in vitro and in vivo to the abilities of OECs to promote CST axon elongation. 6.3.3 Corticospinal regeneration following spinal cord injury may benefit from OEC-mediated therapeutic interventions.  Significant evidence has accrued that suggests the beneficial effects of OEC transplantation to repair the injured spinal cord (Ramon-Cueto 2000; Richter and Roskams 2008).  The current investigations also suggest the significant ability of OB OECs to promote CST outgrowth in comparison to astrocytes, LP fibroblasts and LP OECs.  While the finding of a membrane-bound factor that is at least partially responsible for the CST axon elongation activities of OECs is exciting, it also provides a rationale and context for changes to therapeutic interventions for spinal cord injury.  It is of importance to note that these experiments address only the axon elongation of corticospinal neurons in response to OEC treatment.  The effects of OECs on other axonal populations may be more or less favourable, or simply different.  While some spinal axons have consistently demonstrated a growth affinity for OECs, such as coerulear noradrenergics, or serotonergics from the raphe (Lopez-Vales et al. 2006; Ramon-Cueto 2000), other axons have either not been tested or do not respond to OEC transplantation.  Alternative strategies may be necessary to promote the growth of different spinal axon populations.  Although CST neurons treated with OB OEC plasma membranes overcame neurite outgrowth inhibition by MAG alone, the injured spinal cord presents a plethora of neurite outgrowth inhibitors that must be conquered.  Future experiments should also address the ability of CST neurons treated with OB OEC plasma membrane factors to grow in the presence of 177   myelin fractions, or components of the glial scar such as chondroitin sulphate proteoglycans. While the application of OEC-derived plasma membrane factors may provide an impetus for corticospinal axon elongation, the degradation of scar components by chondroitinase, or the subjugation of myelin inhibitory pathways have also provided significant benefit to injured spinal neurons (Bradbury et al. 2002; Li et al. 2004a).  Although application of chondroitinase to OB OEC treatment can significantly improve functional recovery, following complete spinal transection, over OEC transplantation alone, growth of CST neurons remains minimal (Fouad et al. 2005).  Determination of the plasma membrane factors responsible for CST outgrowth could serve to increase the effects reported by this study, by direct application of these factors, or overexpression of the factors in OEC transplants.  Alternatively, the mode of action of OEC plasma membrane factors on CST neurons has not been elucidated; actions may occur at the cell body, or directly on axons.  Further experiments using Campenot chambers could address the contributions of the OEC plasma membrane to outgrowth promotion from the somal or neuritic compartments.  Delivery of OECs or their plasma membrane activity for the treatment of spinal cord injury could therefore be directed towards the relevant compartment in vivo.  Different CST regeneration outcomes, or expression of regeneration associated genes, have previously been reported in vivo following injection of neurotrophins at the site of a spinal lesion, cell body, or between these sites (Schnell et al. 1994; Tetzlaff et al. 1994).  These experiments would provide a rationale for the application site of OEC-derived plasma membrane factors for the treatment of CST injury.  Although the activities present in the OEC membrane that contribute in so profound a manner to axonal elongation of CST neurons remain unknown, the potency of this effect intimates a specific and exciting therapy for the treatment of spinal cord injury.  Through the identification and application of these activities, it may become possible to direct the outgrowth of CST neurons to their targets after a spinal cord injury.  178   CHAPTER 7: DISCUSSION AND FUTURE DIRECTIONS 7.1 Summary  The experiments contained in this thesis have examined the in vivo reparative and axon outgrowth consequences of OEC transplantation into the injured spinal cord.  This was first accomplished by comparing the outcomes of lamina propria and olfactory bulb ensheathing cell treatments for SCI.  While OEC transplantation of either cell type influenced the arrangement and formation of the glial scar, vascularisation, Schwann cell penetration into the CNS, and the growth of specific spinal fibres into the lesion site, ultimately, these experiments suggested that (1) axon regeneration was moderate with OEC treatment alone, and (2) a thorough understanding of the governing mechanisms of OEC outgrowth promotion would allow for improved treatment strategies for SCI.  To this end, we investigated the contributions of SPARC, a secreted factor highly represented in the OEC secretome and involved in wound healing in other systems, to effect OEC-mediated axon growth after spinal cord injury.  Although transplanted SPARC null and WT LP OECs were similarly able to enagage in the reparative processes we had previously noted, the growth of some long-tract spinal neurons were affected by the absence of SPARC. SPARC secreted by transplanted LP OECs appeared to play a significant role in their ability to promote the growth of tyrosine hydroxylase-positive coerulospinal neurons, and Substance P- positive sensory afferents.   Although OEC-secreted SPARC contributed to the growth of TH and SubP fibre populations, in these and other OEC transplantation experiments, spinal motor neurons of the rubrospinal or corticospinal tracts were less or non-responsive to OEC transplantation.  This discrepancy in the ability of OECs to promote spinal motor neuron regeneration versus other neuron populations, and the need for effective, targeted therapies for SCI, led to the development of an in vitro assay.  By establishing a protocol for the culture of an enriched population of CST motor neurons, this allowed for an assessment of their outgrowth responses to OEC factors.  A comparison of the outgrowth properties of various glia and meningeal cells suggested that OB OECs in particular possessed a potent axon elongation factor for CST neurons.  The major CST axon elongation factor was not secreted from OECs, nor secreted in an induced manner through OEC contact with neurons, but was a plasma membrane protein.  While this plasma membrane activity invoked similar responses in CST neurons as one present in the plasma membrane of astrocytes, the OB OEC factor(s) was much more potent, 179   inducing long-distance axon elongation.  Importantly, in the presence of a myelin-derived neurite outgrowth inhibitor, the OB OEC plasma membrane factor maintained its activity, and allowed for the elongation of CST axons.  These data, combined with further investigations into the molecular basis of the CST elongation factor produced by OECs, could provide insight into therapeutic considerations for OEC transplantation.  This could also lead to the identification of regeneration targets for CST neurons after spinal cord injury.  7.2 Discussion and future directions 7.2.1 How effective is OEC transplantation as a therapeutic for spinal cord injury?  Injury to the spinal cord results in severe consequences to motor, sensory, and autonomic functions, frequently accompanied by a worsening of symptoms with the progression of time; these consequences of therapeutic inaction must be weighed with the possibility of further injury induced by treatment.  Based on the current investigations, and others, it is therefore pertinent to answer the following questions:  (1) Is OEC transplantation an effective treatment for spinal cord injury? And (2) do its outcomes differ from other cell-based therapeutics?  Reports of the consequences of OEC transplantation into the injured spinal cord have been divergent.  While early experiments touted extensive regeneration of dorsal root and CST neurons following OEC transplantation (Li et al. 1997; Ramon-Cueto et al. 2000; Ramon-Cueto and Nieto-Sampedro 1994), other reports have been less favourable (Lopez-Vales et al. 2007; Lu et al. 2006).  Despite these differing views, it is perhaps most instructive to consider the consequences of OEC transplantation as belonging to a category relating to axon regeneration outcomes or a category encompassing other reparative processes. Do OECs alter spinal cord repair outcomes after they are transplanted? Much evidence has accumulated to suggest that OECs integrate well into a spinal lesion site.  When a spinal cord injury remains untreated, large fluid-filled cavities form, bordered by overlapping and impenetrable astrocytic processes that wall off these areas from the surrounding nervous system (Fitch and Silver 2008).  Alternatively, meningeal cells may also be found in untreated injury sites, and these fibroblasts encourage a similar development of nonpermissive 180   astrogliosis and tight border formation (Conrad et al. 2005). Not surprisingly, dermal fibroblasts transplanted into a spinal cord injury form these same non-permissive borders, and do not promote axon regeneration, unless they also carry other genetically-engineered modifiers of their function (Pizzi and Crowe 2006). A similar astrocytic phenotype is reported following Schwann cell implantation in the injured spinal cord (Oudega and Xu 2006), and is corroborated by their behaviour in vitro in coculture with astrocytes (Lakatos et al. 2000).  This observation is of some importance, since the lack of interaction between transplanted cells and endogenous astrocytes generates an environment where axons that penetrate the transplant site are loathe to leave. While it is important that axons be attracted to transplant (lesion) sites, it is equally important that they cross the distal lesion-host interface, and elongate into the host cord.  The starkly different behaviour of OECs at interaction points with astrocytes in vitro and in vivo has been corroborated by many transplantation experiments where this activity has been explored (Fouad et al. 2005; Ramer et al. 2004a; Ramon-Cueto et al. 1998).  OECs form an interdigitated border with astrocytes (Richter et al. 2005), and induce less astrocytic chondroitin sulphate proteoglycan expression than when astrocytes interact with Schwann cells (Lakatos et al. 2003).  It is thought that the interdigitation of OECs with astrocytes provides a more permissive transition for axons between the transplant and host environments, and allows axons to continue their growth through the lesion site.  There is also some implication that astrogliosis, the expression of GFAP, proliferation of astrocytes, and secretion of inhibitory proteoglycans, may be reduced somewhat by OEC transplantation, and that this may also contribute to their therapeutic efficacy (Lopez- Vales et al. 2006).  This was certainly observed in the current experiments, following both LP and OB OEC transplantation, although astrogliosis was reduced even further in LP OEC-treated animals (Richter et al. 2005).  There is also a relative consensus that transplanted OECs provide an excellent bridging substrate for growing axons; we found that although transplanted LP and OB OECs reduce lesion site area to different degrees, both cell types effectively occupied much of the lesion site area even after 28 days (Richter et al. 2005).  Not only do transplanted OECs fill lesion cavities, but they provide a highly conducive substrate for axon growth (Van Den Pol and Santarelli 2003). Furthermore, there is much evidence to suggest that OECs secrete neurotrophic and neurotropic factors that may promote the survival of damaged neurons or induce axon growth in the spinal cord (Au et al. 2007; Chuah et al. 2004; Chung et al. 2004; Woodhall et al. 2001).  Indeed, we 181   showed that the growth of particular tracts of the spinal cord was specifically affected by the presence of the OEC secreted factor, SPARC (Au et al. 2007).  Furthermore, since many authors have also reported the migration and proliferation of Schwann cells into the lesioned spinal cord following OEC transplantation, Schwann cells could serve as a secondary source of trophic factors (Au et al. 2007; Cao et al. 2007a).  A combination of these effects of OEC transplantation suggests that they provide a permissive substrate, and reduce barriers to regeneration, both chemical and physical. Do OECs alter axon regeneration outcomes after they are transplanted? Since OECs are the source of many growth factors and regulators of neurite outgrowth, they have the potential to directly impact the growth of injured spinal fibres following their transplantation.  Certain tracts have indeed consistently demonstrated responsiveness to OEC transplantation, such as TH-positive, serotonergic fibres, and some dorsal root afferents that are substance P or CGRP-positive (Fouad et al. 2005; Ramer et al. 2004a; Ramon-Cueto et al. 1998; Vavrek et al. 2007).  Despite the consistency of the sprouting responses of these populations to OEC transplantation, the current investigations suggest that while growth may be significantly improved versus untreated control animals, the numbers of sprouting fibres are still rather minimal in comparison to the original size and number of fibres forming these tracts in the uninjured animal.  While this argument may be levied against any therapeutic developed to date for the treatment of spinal cord injury, and speaks to the intractable difficulty of repairing the adult CNS, it certainly also suggests the need for an understanding of intrinsic programs governing adult neuron outgrowth.  Indeed, while transplanted OECs seem to possess many advantages over other transplant therapies, in particular those discussed above relating to their integration and scar reduction properties, the removal of these barriers seems insufficient to allow for regeneration; adult CNS long tract spinal neurons require more impetus to grow.  This phenomenon has been described with particular acuity following spinal cord injury in NogoA receptor knockout or NogoA antibody-treated rats (Dimou et al. 2006; Liebscher et al. 2005). While abrogation of a major neurite outgrowth inhibitory pathway (NogoR) does indeed promote fibre sprouting following SCI, this alone is insufficient to drive regeneration (Li et al. 2004a; Zheng et al. 2005).  This certainly implies that a plethora of treatments will be required to induce regeneration following SCI.  These therapies should include bridging strategies, depletion of 182   outgrowth inhibitors, and enhancement of outgrowth promoters.  Although there are a number of experiments that have identified bridging therapies, and have minimized the influence of neurite outgrowth inhibitors, there is little information available concerning the specific outgrowth requirements of adult CNS spinal neurons, other than a generalized increase in growth following cAMP treatment, or inactivation of Rho (Shearer et al. 2003).  It is of paramount importance that the specific pathways responsible for the growth of different adult tracts be understood since this will ultimately allow for therapies directed toward the regeneration of specific tracts.  This will also minimize the potential for inappropriate connections and undesirable side-effects.  One method for the identification of factors that promote adult spinal neuron outgrowth is to use similar in vitro assays to those used in Chapters 5 and 6, but with different neuron populations. Some of these assays have certainly been developed for the culture of embryonic neurons, including dorsal root ganglia, cortical, spinal cord and brainstem explants (Blackmore and Letourneau 2006; Kuang et al. 1994; Sobkowicz et al. 2006).  Another possibility lies in the use of comparative bioinformatics approaches to identify differences in embryonic versus adult neurons, or different populations of adult neurons.  This type of microarray analysis profiling has been performed on the somatosensory versus visual cortices, during the times of afferent versus efferent innervations, and indicate increasing Bcl6 transcript in pyramidal layer V corticospinal neurons, versus Ten_m3 expression, which increases in the visual cortex (Leamey et al. 2008). Furthermore, this microarray data was instrumental in the discovery of Ten_m3 as a determinant of homophilic adhesion and neurite outgrowth of visual cortex neurons (Leamey et al. 2008). Similar strategies could be used for brainstem spinal and corticospinal populations.  Gene expression profiling of cells before, during, and after spinal cord injury could also serve as an indicator of fundamental changes to cells, neuronal and glial, that may inhibit regeneration (De Biase et al. 2005; Xiao et al. 2005; Yang et al. 2006).  However, it is difficult to identify specific targets from these data since they encompass many cell types, and are a comparison of neurons and cells in a resting state, versus an injured state, not regenerative versus resting states. Furthermore, changes in expression can also be strain-dependent, making comparisons across studies difficult (Schmitt et al. 2006).  Despite these drawbacks, clusters of genes have been identified whose expression is correlated with functional improvement following spinal cord injury, and include transcripts for neurturin, attractin, microtubule-associated protein 1a, and myelin oligodendrocyte genes (Di Giovanni et al. 2005).  Furthermore, transfection of dorsal root 183   ganglion explants with some of these cluster members confirmed their involvement in neurite outgrowth promotion (Di Giovanni et al. 2005).  Thus bioinformatics and in vitro assays could be used in tandem to identify molecular determinants of outgrowth for different populations of spinal tract neurons.  Application of therapeutics developed through these strategies, combined with OEC therapy and other therapeutics could provide permissive and instructive cues to direct axon outgrowth after spinal cord injury. 7.2.2 Why are spinal cord tracts differentially responsive to OEC transplantation?  The issue of instructing regenerating axons towards their correct targets is no small dilemma following spinal cord injury.  However, since the current investigations, and others, have noted a propensity for subpopulations of spinal neurons to grow in the presence of OECs, it is possible that both axon outgrowth and guidance might be accomplished in tandem through the transplantation of these cells.  It is therefore essential that the mechanisms underlying the guidance of different spinal neuron tracts be understood as they relate to OEC secreted or membrane-bound factors.  To understand how OECs induce the growth of particular neuron subpopulations, two issues must be examined: (1) Are some tracts differentially responsive to OECs, or does their growth in OEC transplanted animals simply underscore an overall difference in their regenerative capacity? (2) What cues underlie guidance of these tracts during development, and can these cues be effectively recapitulated following spinal cord injury to instruct adult neurons?  Tyrosine hydroxylase, serotonin, substanceP, and calcitonin gene related peptide-positive neurons are found to sprout consistently when OECs are transplanted, either into incomplete or complete spinal lesions (Fouad et al. 2005; Ramer et al. 2004a; Vavrek et al. 2007). Three groups have reported the sprouting of CST neurons following OEC transplantation, one in the added presence of retrovirally-produced NT-3, while several groups have not observed CST growth (Li et al. 1997; Lu et al. 2006; Ramon-Cueto et al. 2000; Ruitenberg et al. 2005).  While the current investigation did not reveal extensive growth of rubrospinal neurons after OEC transplantation, adenovirally-expressing OECs secreting BDNF have previously been effective at promoting a small amount of rubrospinal sprouting (Delucia et al. 2007).  Are the growth responses of these tracts reflective of the production of factors by OECs that interact specifically with these neurons 184   and not with other neurons?  Or is the response hierarchy of these neurons simply a reflection of intrinsic differences in their growth potentials following injury?  Certainly the growth of any of these tracts in the absence of treatment is extremely low.  The differing growth of TH-positive and SubP-positive in response to LP or OB OEC transplantation (Chapter 3) supports a role for specific interactions between these neurons and OECs that contribute to their sprouting. Furthermore, the specific effect of SPARC deletion from OECs on the sprouting of TH and SubP-positive fibres (Chapter 4) suggests that these neurons grow fervently in the presence of OECs because OECs provide their preferred trophic/tropic factors.  Interestingly, we also found that SPARC activity produced by OECs depended on a secondary mechanism requiring Schwann cells to promote neurite outgrowth (Au et al. 2007).  However, this interaction is probably not fundamental to the ability of OECs to promote TH-positive growth, since TH-positive axons grow relatively poorly following Schwann cell transplantation alone; only when Schwann cells are transduced to overexpress NGF, other neurotrophins, or when they are derived from bone marrow stromal cells or skin-derived precursors, do they become potent promoters of TH- positive outgrowth (Biernaskie et al. 2007; Golden et al. 2007; Kamada et al. 2005; Weidner et al. 1999).  The same specificity of OEC effect appears true of CGRP-positive afferent growth, although these neurons are not specifically responsive to SPARC (Weidner et al. 1999). Thus the mechanism contributing to TH and SubP-positive axon growth following OEC transplantation is probably specific, and is the result of secreted or cell surface factors on OECs that promote the growth of these neurons.  In contrast, serotonergic axons sprout robustly following many therapeutic interventions, including Schwann cell, skin-derived precursor, and OEC transplantation, or locomotor training (Biernaskie et al. 2007; Engesser-Cesar et al. 2007; Golden et al. 2007; Lu et al. 2002). Furthermore, the growth response of serotonergic fibres to chondroitin sulphate proteoglycan degradation alone after SCI, without the need for other trophic therapies, suggests that raphespinal neurons may simply possess increased growth potential in the adult animal (Kim et al. 2006).  At the contrasting end of this spectrum, is the stubborn resistance of CST and rubrospinal neurons to most spinal cord therapies.  While several groups have reported the growth of CST neurons after OEC transplantation, these results are somewhat confounded by the sparing of CST fibres intrinsic to this lesion paradigm, and is contradicted by others (Keyvan- Fouladi et al. 2003; Li et al. 1997; Lu et al. 2006).  Perhaps the most convincing demonstration 185   of CST growth after injury occurs following transplantation of NT-3-expressing OECs into a cervical corticospinal tract transection; tracing of both corticospinal tracts with BDA-R (red) and BDA-G (green) effectively shows the complete unilateral destruction of the corticospinal tract, and shows significantly-improved growth of this tract after treatment (Ruitenberg et al. 2005). Yet even with this careful, and multifactorial approach, corticospinal axon growth is still very poor, and axons retract so extensively, that even with significant improvements in the OEC-NT-3 condition, the tract encompasses at most 10% of the dorsal corticospinal tract immediately rostral to the lesion site (Ruitenberg et al. 2005).  Furthermore, this did not result in any amelioration in the behavioural parameters measured (Ruitenberg et al. 2005).  Other transplantation experiments have similarly noted the recalcitrant behaviour of corticospinal axons; neither transplanted neural progenitor cells nor aligned immature astrocytes on biomatrix bridges could provoke corticospinal regeneration beyond a lesion site, or evoke functional improvements (Deumens et al. 2006c; Webber et al. 2007).  However, despite the difficulty in inducing CST regeneration, differing effects are evoked based on the therapeutic applied, suggesting that the sprouting responses of corticospinal neurons to OECs is a reflection not of an overall growth capacity (or lack thereof), but is a reaction to those factors produced by OECs. This argument is certainly corroborated by the outgrowth of corticospinal neurons on glial and fibroblast monolayers described here.  Corticospinal neurons displayed markedly different outgrowth phenotypes on astrocyte versus OEC monolayers, whereas outgrowth was elongative on both OEC and astrocyte plasma membranes. These data suggest the specificity of OEC- corticospinal interactions.   Despite the weak response of corticospinal neurons to transplant- mediated and other therapies, OECs clearly provide specific contact-dependent outgrowth- inducing factors for these neurons. These questions therefore remain: To which outgrowth and guidance factors are CST neurons responsive? And how do OECs promote CST outgrowth?  At the current juncture, data is limited as to how adult CST neuron outgrowth is achieved; indeed, if this information were available, the problem of spinal cord repair would be significantly lessened.  Our understanding is limited to the growth of CST neurons during development, with the anticipation that similar mechanisms will govern adult outgrowth.  Since CST development is preceded by the outgrowth of other spinal tracts, part of its growth is dependent on passive guidance through a path of least resistance (ten Donkelaar et al. 2004).  As well, many of the tracts that descend prior to the CST 186   express a variety of cell adhesion molecules that are extremely important for CST outgrowth, including NCAM, laminin, L1, EphB3/ephrinA4, and N-cadherin (Joosten and Bar 1999).  Thus it is probably reasonable that such contact-mediated interactions also govern the growth of corticospinal neurons on OEC monolayers or on their plasma membranes (Chapter 6).  Early guidance of CST neurons is also based on diffusible factors such as Slit2-mediated repulsion from the basal telecephalon and hypothalamus (Bagri et al. 2002), and Netrin-1 receptors engagement for pyramidal decussation (ten Donkelaar et al. 2004).  However, since these secreted factors are involved primarily in early guidance decisions of corticospinal neurons, they may be less important for its growth following injury, particularly since most injuries occur below the level of the pyramidal decussation.  It therefore appears that determinants of corticospinal outgrowth during development are primarily cell contact-mediated interactions involving the classical adhesion molecules described above, and other, as yet undetermined, interactions.  Certainly L1 overexpression following spinal cord injury increases growth of corticospinal neurons to the rostral border of the lesion site, and enhances stepping abilities and coordination after lesion (Chen et al. 2007a).  The activity of OECs on corticospinal neurons in vitro also seems to solidify this assertion, and suggests that the contact-dependent regulation of corticospinal neuron outgrowth may be similar in the developing and adult nervous systems. 7.2.3 What is the nature of the OEC plasma membrane CST axon elongation factor?  To develop contact-mediated therapies for CST repair, a variety of approaches could be adopted.  While the protein activity of the OEC plasma membrane produces a potent axon elongation effect on CST neurons, these data should be combined with other available information on adult CST expression of cell adhesion molecules, and the commensurate expression of these proteins on the OEC plasma membrane.  As well, cumulative data from the OEC-astrocyte microarray database suggest some intruiging possibilities for the regulation of CST outgrowth by OECs.  However, because much of this expression evidence has been produced only in developmental samples, in OECs grown under different conditions, represents an RNA expression profile, or has never been documented, it is of primary importance to establish whether these cell-bound factors are indeed present on both OECs and P8 CST neurons. One means of investigating the expression of these candidates in OECs and corticospinal neurons is through the simple application of immunohisto/cytochemical techniques or Western 187   blotting.  Expression of ligands and their cognate receptors from developmental and adult time points, and from in vitro samples of OEC and CST neuron plasma membranes could thereby be assessed.  As well, since the OEC and astrocyte plasma membrane demonstrate some similarities in activity, levels of expression in these two cells from Western blots could be helpful in differentiating between factors that might be responsible for the OEC outgrowth activity. However, since these approaches represent a relatively limited screen of potential factors, it would also be instructive to establish a more complete proteomic analysis of the plasma membranes of OECs.  One possibility is the use of 2-dimensional gel electrophoresis (2-G), and analysis by tandem mass spectrometry, which has been previously used to identify plasma membrane components from different samples (Chen et al. 2006).  However, since integral and other plasma membrane proteins directing neurite outgrowth may be considerably less abundant than other “housekeeping” membrane proteins, this strategy may tend to obscure relevant proteins (Tannu and Hemby 2006).  Furthermore, poly-acrylamide gel electrophoresis techniques are best suited to acidic or weakly basic proteins, whereas integral membrane proteins in particular, are commonly strongly basic, and would also be difficult to solubilize (Grant and Wu 2007).  These difficulties necessitate strong internal controls to decrease the 2-G bias against hydrophobic, alkaline proteins.  The use of mass spectrometry, preceded by stable isotope labelling of amiono acids in cell culture (SILAC), in which cells are grown in media lacking an essential amino acid, but which is supplemented with a non-radioactive, isotopically-labelled amino acid, could provide an efficient means for the simultaneous and automatic detection, quantification, and comparison of membrane proteins in astrocytes and OECs (Ong et al. 2002). This would provide a more complete proteome of the OEC membrane, than simply those known cell adhesion molecules that have already been described.  As well, highly enriched proteins in the OEC plasma membrane over the astrocyte plasma membrane, or uniquely expressed proteins from OECs, would be viable candidates for the OEC-induced CST outgrowth activity. A second approach to the identification of OEC plasma membrane factors stems from an investigation of the outgrowth mechanisms underlying CST responses to OECs.  By understanding the participation of second messenger systems or other mediators of the neurite outgrowth response to the OEC plasma membrane, this could aid in identifying not only the factor(s) initiating outgrowth, but could delineate how similar responses could be evoked from other intracellular pathways.  For example, brief treatment of CST neurons with calcium 188   chelators and antagonists to Ryanodine receptors upon their culture with OEC plasma membrane could help to parse the effects of the known intracellular pathways of ephrins, CAMs, semaphorins, and cadherins.  Since release of calcium from intracellular stores is necessary for neurite outgrowth responses to CAMs, cadherins, and semaphorins, if the OEC plasma membrane activity is dependent on these molecules, treatment with Ryanodine and EDTA would be expected to decrease neurite outgrowth relative to OEC plasma membrane alone. Alternatively, calcium imaging of CST neurons plated onto alternating stripes of OEC plasma membrane and CHO-R2 plasma membrane could be used to assess the contribution of calcium signalling to CST neurite outgrowth on OEC plasma membranes.  Other inhibitors of specific signalling pathways, such as FGFR, PI3K, or PLCγ antagonists, could also be used to assess their contributions to OEC-induced CST axon elongation. A third approach mirrors the second approach, but is focused on altering the OEC plasma membrane itself to perturb its CST outgrowth activity.  Simple questions concerning the type of protein(s) responsible for these effects (e.g. does the activity depend on a GPI-linked protein?) could be ascertained by treating the OEC plasma membrane in a variety of manners.  Trypsin degradation was already used to show that the OEC plasma membrane activity depends on a protein or glycoprotein.  To determine if integral membrane proteins versus other membrane proteins account for the OEC outgrowth activity, membranes could be washed in high salt solution, disrupting all but integral membrane proteins (Petit et al. 2005).  Alternatively, phosphatidylinositol-specific phopholipase C incubation with membrane samples can be used to cleave GPI-linked proteins from the surface, and therefore to show whether CST outgrowth on OEC plasma membranes is altered by GPI-protein depletion (Mann et al. 2002).  A combination of these strategies, including expression analysis, and perturbation to corticospinal outgrowth mechanisms or the OEC plasma membrane, could together yield important insights into the regulation of corticospinal outgrowth by OECs.  More importantly, it could provide a directed framework for the investigation of corticospinal neuron axon outgrowth pathways, suggesting targeted therapies for repair of the injured spinal cord.   189   7.3 Conclusions  This thesis has examined the efficacy of olfactory ensheathing cell (OEC) transplantation as a therapeutic for spinal cord injury from in vivo and in vitro perspectives.  By transplanting OECs derived from peripheral (LP) and central (OB) nervous system sources into a spinal cord injury, and comparing their repair properties, this has provided evidence both that (1) OECs can serve as effective mediators of wound healing through reductions in cavity formation, astrogliosis, promotion of angiogenesis, and Schwann cell infiltration, and (2) that LP and OB OECs have specific and differing interactions with host spinal tracts, leading to increases in the sprouting of these tracts.  In particular, tyrosine hydroxylase and substanceP-positive fibres grow extensively upon LP OEC transplantation, a phenomenon that can be attributed in part to the secretion of SPARC, secreted protein acidic and rich in cysteines, from LP OECs.  To delineate how OECs might interact with populations other than TH and substanceP-expressing neurons to affect their outgrowth,  an outgrowth assay was developed using postnatal day 8 corticospinal neurons, identified by YFP expression in the transgenic mouse line Thy1YFP-16JRS, and by other markers of corticospinal neurons.  Culture conditions enriching for corticospinal neurons over other cell types were developed, so that responses of these neurons to OEC secreted and membrane-bound factors could be assessed.  Culture of corticospinal neurons with glia or fibroblasts led to differing outgrowth phenotypes, although OB OEC coculture was most effective at inducing axon elongation, a phenotype desirable for spinal cord injury repair.  The mechanisms responsible for the promotion of corticospinal axon elongation by OB OECs were investigated, and depended not on secreted factors, but on the presence of a proteinaceous membrane-bound activity.  Surprisingly, the elongative activity of the OB OEC plasma membrane on corticospinal neurons was similar in its characteristics (elongative and axonal), although dissimilar in extent, to that induced by astrocyte plasma membrane, suggesting a similarity of outgrowth mechanism between these two cells.  Importantly, the factors present in the OB OEC plasma membrane could also overcome outgrowth inhibition induced by culture on myelin-associated glycoprotein-expressing cell plasma membranes, suggesting that these proteins may also help corticospinal neurons to overcome outgrowth inhibition in the hostile mileu of the injured spinal cord.  Identification of the factors responsible for the axon elongative 190   acitivity of OB OECs will hopefully provide targeted therapies for repair of the injured corticospinal tract.   191   References cited Acosta CG, Fabrega AR, Masco DH, Lopez HS. 2001. A sensory neuron subpopulation with unique sequential survival dependence on nerve growth factor and basic fibroblast growth factor during development. J Neurosci 21(22):8873-8885. Agrawal AK, Shukla S, Chaturvedi RK, Seth K, Srivastava N, Ahmad A, Seth PK. 2004. Olfactory ensheathing cell transplantation restores functional deficits in rat model of Parkinson's disease: a cotransplantation approach with fetal ventral mesencephalic cells. Neurobiol Dis 16(3):516-526. Aguayo AJ, David S, Bray GM. 1981. Influences of the glial environment on the elongation of axons after injury: transplantation studies in adult rodents. J Exp Biol 95:231-240. Airaksinen MS, Saarma M. 2002. The GDNF family: signalling, biological functions and therapeutic value. Nat Rev Neurosci 3(5):383-394. Akins MR, Benson DL, Greer CA. 2007. Cadherin expression in the developing mouse olfactory system. J Comp Neurol 501(4):483-497. Akiyama Y, Lankford K, Radtke C, Greer CA, Kocsis JD. 2004. Remyelination of spinal cord axons by olfactory ensheathing cells and Schwann cells derived from a transgenic rat expressing alkaline phosphatase marker gene. Neuron Glia Biol 1(1):47-55. Alexander CL, Fitzgerald UF, Barnett SC. 2002. Identification of growth factors that promote long-term proliferation of olfactory ensheathing cells and modulate their antigenic phenotype. Glia 37(4):349-364. Alstermark B, Ohlson S. 1999. Organisation of corticospinal neurones controlling forelimb target reaching in cats. Acta Physiol Scand 167(2):A12-A13. Altman J, Das GD. 1965. Autoradiographic and histological evidence of postnatal hippocampal neurogenesis in rats. J Comp Neurol 124(3):319-335. Alvarez MJ, Prada F, Salvatierra E, Bravo AI, Lutzky VP, Carbone C, Pitossi FJ, Chuluyan HE, Podhajcer OL. 2005. Secreted protein acidic and rich in cysteine produced by human melanoma cells modulates polymorphonuclear leukocyte recruitment and antitumor cytotoxic capacity. Cancer Res 65(12):5123-5132. Arlotta P, Molyneaux BJ, Chen J, Inoue J, Kominami R, Macklis JD. 2005. Neuronal subtype- specific genes that control corticospinal motor neuron development in vivo. Neuron 45(2):207-221. Armstrong JN, MacVicar BA. 2001. Theta-frequency facilitation of AMPA receptor-mediated synaptic currents in the principal cells of the medial septum. J Neurophysiol 85(4):1709- 1718. Astic L, Pellier-Monnin V, Saucier D, Charrier C, Mehlen P. 2002. Expression of netrin-1 and netrin-1 receptor, DCC, in the rat olfactory nerve pathway during development and axonal regeneration. Neuroscience 109(4):643-656. Au E, Richter MW, Vincent AJ, Tetzlaff W, Aebersold R, Sage EH, Roskams AJ. 2007. SPARC from olfactory ensheathing cells stimulates Schwann cells to promote neurite outgrowth and enhances spinal cord repair. J Neurosci 27(27):7208-7221. Au E, Roskams AJ. 2002. Culturing olfactory ensheathing glia from the mouse olfactory epithelium. Methods Mol Biol 198:49-54. Au E, Roskams AJ. 2003. Olfactory ensheathing cells of the lamina propria in vivo and in vitro. Glia 41(3):224-236. 192   Au WW, Treloar HB, Greer CA. 2002. Sublaminar organization of the mouse olfactory bulb nerve layer. J Comp Neurol 446(1):68-80. Baas PW, Luo L. 2001. Signaling at the growth cone: the scientific progeny of Cajal meet in Madrid. Neuron 32(6):981-984. Bagri A, Marin O, Plump AS, Mak J, Pleasure SJ, Rubenstein JL, Tessier-Lavigne M. 2002. Slit proteins prevent midline crossing and determine the dorsoventral position of major axonal pathways in the mammalian forebrain. Neuron 33(2):233-248. Barami K, Grever WE, Diaz FG, Lyman WD. 2001. An efficient method for the culturing and generation of neurons and astrocytes from second trimester human central nervous system tissue. Neurol Res 23(4):321-326. Barnett SC. 2004. Olfactory ensheathing cells: unique glial cell types? J Neurotrauma 21(4):375- 382. Barnett SC, Hutchins AM, Noble M. 1993. Purification of olfactory nerve ensheathing cells from the olfactory bulb. Developmental Biology 155(2):337-350. Basu A, Kligman LH, Samulewicz SJ, Howe CC. 2001. Impaired wound healing in mice deficient in a matricellular protein SPARC (osteonectin, BM-40). BMC Cell Biol 2(1):15. Bauer S, Rasika S, Han J, Mauduit C, Raccurt M, Morel G, Jourdan F, Benahmed M, Moyse E, Patterson PH. 2003. Leukemia inhibitory factor is a key signal for injury-induced neurogenesis in the adult mouse olfactory epithelium. J Neurosci 23(5):1792-1803. Beggs HE, Baragona SC, Hemperly JJ, Maness PF. 1997. NCAM140 interacts with the focal adhesion kinase p125(fak) and the SRC-related tyrosine kinase p59(fyn). J Biol Chem 272(13):8310-8319. Bernhardt R, Matus A. 1984. Light and electron microscopic studies of the distribution of microtubule-associated protein 2 in rat brain: a difference between dendritic and axonal cytoskeletons. J Comp Neurol 226(2):203-221. Berryhill BL, Kane B, Stramer BM, Fini ME, Hassell JR. 2003. Increased SPARC accumulation during corneal repair. Exp Eye Res 77(1):85-92. Besirli CG, Deckwerth TL, Crowder RJ, Freeman RS, Johnson EM, Jr. 2003. Cytosine arabinoside rapidly activates Bax-dependent apoptosis and a delayed Bax-independent death pathway in sympathetic neurons. Cell Death Differ 10(9):1045-1058. Bianco JI, Perry C, Harkin DG, Mackay-Sim A, Feron F. 2004. Neurotrophin 3 promotes purification and proliferation of olfactory ensheathing cells from human nose. Glia 45(2):111-123. Biernaskie J, Sparling JS, Liu J, Shannon CP, Plemel JR, Xie Y, Miller FD, Tetzlaff W. 2007. Skin-derived precursors generate myelinating Schwann cells that promote remyelination and functional recovery after contusion spinal cord injury. J Neurosci 27(36):9545-9559. Blackmore M, Letourneau PC. 2006. L1, beta1 integrin, and cadherins mediate axonal regeneration in the embryonic spinal cord. J Neurobiol 66(14):1564-1583. Blesch A, Uy HS, Grill RJ, Cheng JG, Patterson PH, Tuszynski MH. 1999. Leukemia inhibitory factor augments neurotrophin expression and corticospinal axon growth after adult CNS injury. J Neurosci 19(9):3556-3566. Blinder KJ, Pumplin DW, Paul DL, Keller A. 2003. Intercellular interactions in the mammalian olfactory nerve. J Comp Neurol 466(2):230-239. Bolon B, Dorman DC, Bonnefoi MS, Randall HW, Morgan KT. 1993. Histopathologic approaches to chemical toxicity using primary cultures of dissociated neural cells grown in chamber slides. Toxicol Pathol 21(5):465-479. 193   Boruch AV, Conners JJ, Pipitone M, Deadwyler G, Storer PD, Devries GH, Jones KJ. 2001. Neurotrophic and migratory properties of an olfactory ensheathing cell line. Glia 33(3):225-229. Boyd JG, Jahed A, McDonald TG, Krol KM, Van Eyk JE, Doucette R, Kawaja MD. 2006. Proteomic evaluation reveals that olfactory ensheathing cells but not Schwann cells express calponin. Glia 53(4):434-440. Bradbury EJ, Moon LD, Popat RJ, King VR, Bennett GS, Patel PN, Fawcett JW, McMahon SB. 2002. Chondroitinase ABC promotes functional recovery after spinal cord injury. Nature 416(6881):636-640. Bradshaw AD, Sage, E.H. 2001. SPARC, a matricellular protein that functions in cellular differentiation and tissue response to injury. Journal of Clinical Investigation 107(9):1049-1054. Braun H, Buhnemann C, Neumann J, Reymann KG. 2006. Preparation of a tissue-like cortical primary culture from embryonic rats using Matrigel and serum free Start V Medium. J Neurosci Methods 157(1):32-38. Bregman BS, McAtee M, Dai HN, Kuhn PL. 1997. Neurotrophic factors increase axonal growth after spinal cord injury and transplantation in the adult rat. Exp Neurol 148(2):475-494. Brown LT, Jr. 1971. Projections and termination of the corticospinal tract in rodents. Exp Brain Res 13(4):432-450. Brummendorf T, Rathjen FG. 1995. Cell adhesion molecules 1: immunoglobulin superfamily. Protein Profile 2(9):963-1108. Brunjes PC, Greer CA. 2003. Progress and directions in olfactory development. Neuron 38(3):371-374. Brus-Ramer M, Carmel JB, Chakrabarty S, Martin JH. 2007. Electrical stimulation of spared corticospinal axons augments connections with ipsilateral spinal motor circuits after injury. J Neurosci 27(50):13793-13801. Bunge MB. 2002. Bridging the transected or contused adult rat spinal cord with Schwann cell and olfactory ensheathing glia transplants. Prog Brain Res 137:275-282. Cajal SRY. 1928. Degeneration and Regeneration of the Nervous System. London: Oxford University Press. Camand E, Morel MP, Faissner A, Sotelo C, Dusart I. 2004. Long-term changes in the molecular composition of the glial scar and progressive increase of serotoninergic fibre sprouting after hemisection of the mouse spinal cord. Eur J Neurosci 20(5):1161-1176. Cameron AA, Smith GM, Randall DC, Brown DR, Rabchevsky AG. 2006. Genetic manipulation of intraspinal plasticity after spinal cord injury alters the severity of autonomic dysreflexia. J Neurosci 26(11):2923-2932. Cannella MS, Ross RA. 1987. Influence of substratum on the retrograde response of the rat superior cervical ganglion in vitro. Exp Neurol 95(3):652-660. Cao L, Liu L, Chen ZY, Wang LM, Ye JL, Qiu HY, Lu CL, He C. 2004. Olfactory ensheathing cells genetically modified to secrete GDNF to promote spinal cord repair. Brain 127(Pt 3):535-549. Cao L, Su Z, Zhou Q, Lv B, Liu X, Jiao L, Li Z, Zhu Y, Huang Z, Huang A, He C. 2006. Glial cell line-derived neurotrophic factor promotes olfactory ensheathing cells migration. Glia 54(6):536-544. Cao L, Zhu YL, Su Z, Lv B, Huang Z, Mu L, He C. 2007a. Olfactory ensheathing cells promote migration of Schwann cells by secreted nerve growth factor. Glia 55(9):897-904. 194   Cao Z, Qiu J, Domeniconi M, Hou J, Bryson JB, Mellado W, Filbin MT. 2007b. The inhibition site on myelin-associated glycoprotein is within Ig-domain 5 and is distinct from the sialic acid binding site. J Neurosci 27(34):9146-9154. Capaday C, Lavoie BA, Barbeau H, Schneider C, Bonnard M. 1999. Studies on the corticospinal control of human walking. I. Responses to focal transcranial magnetic stimulation of the motor cortex. J Neurophysiol 81(1):129-139. Carr VM, Farbman AI. 1992. Ablation of the olfactory bulb up-regulates the rate of neurogenesis and induces precocious cell death in olfactory epithelium. Exp Neurol 115(1):55-59. Carr VM, Farbman AI. 1993. The dynamics of cell death in the olfactory epithelium. Exp Neurol 124(2):308-314. Carrier RL, Ma TC, Obrietan K, Hoyt KR. 2006. A sensitive and selective assay of neuronal degeneration in cell culture. J Neurosci Methods 154(1-2):239-244. Carter LA, MacDonald JL, Roskams AJ. 2004. Olfactory horizontal basal cells demonstrate a conserved multipotent progenitor phenotype. J Neurosci 24(25):5670-5683. Catapano LA, Arnold MW, Perez FA, Macklis JD. 2001. Specific neurotrophic factors support the survival of cortical projection neurons at distinct stages of development. J Neurosci 21(22):8863-8872. Charron F, Stein E, Jeong J, McMahon AP, Tessier-Lavigne M. 2003. The morphogen sonic hedgehog is an axonal chemoattractant that collaborates with netrin-1 in midline axon guidance. Cell 113(1):11-23. Chehrehasa F, St John J, Key B. 2005. The sorting behaviour of olfactory and vomeronasal axons during regeneration. J Mol Histol 36(6-7):427-436. Chen J, Wu J, Apostolova I, Skup M, Irintchev A, Kugler S, Schachner M. 2007a. Adeno- associated virus-mediated L1 expression promotes functional recovery after spinal cord injury. Brain 130(Pt 4):954-969. Chen L, Huang H, Zhang J, Zhang F, Liu Y, Xi H, Wang H, Gu Z, Song Y, Li Y, Tan K. 2007b. Short-term outcome of olfactory ensheathing cells transplantation for treatment of amyotrophic lateral sclerosis. Zhongguo Xiu Fu Chong Jian Wai Ke Za Zhi 21(9):961- 966. Chen P, Li X, Sun Y, Liu Z, Cao R, He Q, Wang M, Xiong J, Xie J, Wang X, Liang S. 2006. Proteomic analysis of rat hippocampal plasma membrane: characterization of potential neuronal-specific plasma membrane proteins. J Neurochem 98(4):1126-1140. Cheney PD, Fetz EE, Mewes K. 1991. Neural mechanisms underlying corticospinal and rubrospinal control of limb movements. Prog Brain Res 87:213-252. Cheng L, Du C, Murray D, Tong X, Zhang YA, Chen BP, Hawley RG. 1997. A GFP reporter system to assess gene transfer and expression in human hematopoietic progenitor cells. Gene Ther 4(10):1013-1022. Chernousov MA, Scherer SS, Stahl RC, Carey DJ. 1999. p200, a collagen secreted by Schwann cells, is expressed in developing nerves and in adult nerves following axotomy. J Neurosci Res 56(3):284-294. Chiu FC, Potter PE, Rozental R. 1998. Purification of human fetal hippocampal neurons by flow cytometry for transplantation. Methods 16(3):260-267. Christopherson KS, Ullian EM, Stokes CC, Mullowney CE, Hell JW, Agah A, Lawler J, Mosher DF, Bornstein P, Barres BA. 2005. Thrombospondins are astrocyte-secreted proteins that promote CNS synaptogenesis. Cell 120(3):421-433. 195   Chuah MI, Au C. 1991. Olfactory Schwann cells are derived from precursor cells in the olfactory epithelium. Journal of Neuroscience Research 29(2):172-180. Chuah MI, Au C. 1994. Olfactory cell cultures on ensheathing cell monolayers. Chemical Senses 19(1):25-34. Chuah MI, Choi-Lundberg D, Weston S, Vincent AJ, Chung RS, Vickers JC, West AK. 2004. Olfactory ensheathing cells promote collateral axonal branching in the injured adult rat spinal cord. Exp Neurol 185(1):15-25. Chuah MI, Tennent R, Jacobs I. 1995. Response of olfactory Schwann cells to intranasal zinc sulfate irrigation. J Neurosci Res 42(4):470-478. Chuah MI, West AK. 2002. Cellular and molecular biology of ensheathing cells. Microsc Res Tech 58(3):216-227. Chung RS, Woodhouse A, Fung S, Dickson TC, West AK, Vickers JC, Chuah MI. 2004. Olfactory ensheathing cells promote neurite sprouting of injured axons in vitro by direct cellular contact and secretion of soluble factors. Cell Mol Life Sci 61(10):1238-1245. Clarris HJ, Rauch U, Key B. 2000. Dynamic spatiotemporal expression patterns of neurocan and phosphacan indicate diverse roles in the developing and adult mouse olfactory system. J Comp Neurol 423(1):99-111. Cohen NR, Taylor JS, Scott LB, Guillery RW, Soriano P, Furley AJ. 1998. Errors in corticospinal axon guidance in mice lacking the neural cell adhesion molecule L1. Curr Biol 8(1):26-33. Conrad S, Schluesener HJ, Adibzahdeh M, Schwab JM. 2005. Spinal cord injury induction of lesional expression of profibrotic and angiogenic connective tissue growth factor confined to reactive astrocytes, invading fibroblasts and endothelial cells. J Neurosurg Spine 2(3):319-326. Costa S, Planchenault T, Charriere-Bertrand C, Mouchel Y, Fages C, Juliano S, Lefrancois T, Barlovatz-Meimon G, Tardy M. 2002. Astroglial permissivity for neuritic outgrowth in neuron-astrocyte cocultures depends on regulation of laminin bioavailability. Glia 37(2):105-113. Cowan CM, Thai J, Krajewski S, Reed JC, Nicholson DW, Kaufmann SH, Roskams AJ. 2001. Caspases 3 and 9 send a pro-apoptotic signal from synapse to cell body in olfactory receptor neurons. J Neurosci 21(18):7099-7109. Cowan CW, Shao YR, Sahin M, Shamah SM, Lin MZ, Greer PL, Gao S, Griffith EC, Brugge JS, Greenberg ME. 2005. Vav family GEFs link activated Ephs to endocytosis and axon guidance. Neuron 46(2):205-217. Crandall JE, Dibble C, Butler D, Pays L, Ahmad N, Kostek C, Puschel AW, Schwarting GA. 2000. Patterning of olfactory sensory connections is mediated by extracellular matrix proteins in the nerve layer of the olfactory bulb. J Neurobiol 45(4):195-206. Crone SA, Lee KF. 2002. The bound leading the bound: target-derived receptors act as guidance cues. Neuron 36(3):333-335. Daston MM, Bastmeyer M, Rutishauser U, O'Leary DD. 1996. Spatially restricted increase in polysialic acid enhances corticospinal axon branching related to target recognition and innervation. J Neurosci 16(17):5488-5497. Davies SJ, Fitch MT, Memberg SP, Hall AK, Raisman G, Silver J. 1997. Regeneration of adult axons in white matter tracts of the central nervous system. Nature 390(6661):680-683. Davies SJ, Goucher DR, Doller C, Silver J. 1999. Robust regeneration of adult sensory axons in degenerating white matter of the adult rat spinal cord. J Neurosci 19(14):5810-5822. 196   De Biase A, Knoblach SM, Di Giovanni S, Fan C, Molon A, Hoffman EP, Faden AI. 2005. Gene expression profiling of experimental traumatic spinal cord injury as a function of distance from impact site and injury severity. Physiol Genomics 22(3):368-381. Delucia TA, Alexander TD, Fargo KN, Jones KJ. 2007. Effects of single versus combinatorial treatment strategies on beta II-tubulin gene expression in axotomized hamster rubrospinal motoneurons. Restor Neurol Neurosci 25(5-6):573-584. Dent EW, Gertler FB. 2003. Cytoskeletal dynamics and transport in growth cone motility and axon guidance. Neuron 40(2):209-227. Deumens R, Koopmans GC, Den Bakker CG, Maquet V, Blacher S, Honig WM, Jerome R, Pirard JP, Steinbusch HW, Joosten EA. 2004. Alignment of glial cells stimulates directional neurite growth of CNS neurons in vitro. Neuroscience 125(3):591-604. Deumens R, Koopmans GC, Honig WM, Hamers FP, Maquet V, Jerome R, Steinbusch HW, Joosten EA. 2006a. Olfactory ensheathing cells, olfactory nerve fibroblasts and biomatrices to promote long-distance axon regrowth and functional recovery in the dorsally hemisected adult rat spinal cord. Exp Neurol 200(1):89-103. Deumens R, Koopmans GC, Honig WM, Maquet V, Jerome R, Steinbusch HW, Joosten EA. 2006b. Chronically injured corticospinal axons do not cross large spinal lesion gaps after a multifactorial transplantation strategy using olfactory ensheathing cell/olfactory nerve fibroblast-biomatrix bridges. J Neurosci Res 83(5):811-820. Deumens R, Koopmans GC, Honig WM, Maquet V, Jerome R, Steinbusch HW, Joosten EA. 2006c. Limitations in transplantation of astroglia-biomatrix bridges to stimulate corticospinal axon regrowth across large spinal lesion gaps. Neurosci Lett 400(3):208- 212. Deumens R, Koopmans GC, Lemmens M, Mollers S, Honig WM, Steinbusch HW, Brook G, Joosten EA. 2006d. Neurite outgrowth promoting effects of enriched and mixed OEC/ONF cultures. Neurosci Lett 397(1-2):20-24. Devon R, Doucette R. 1992. Olfactory ensheathing cells myelinate dorsal root ganglion neurites. Brain Res 589(1):175-179. Dewar D, Bentley D, Barnett SC. 2007. Implantation of pure cultured olfactory ensheathing cells in an animal model of parkinsonism. Acta Neurochir (Wien) 149(4):407-414. Di Giovanni S, Faden AI, Yakovlev A, Duke-Cohan JS, Finn T, Thouin M, Knoblach S, De Biase A, Bregman BS, Hoffman EP. 2005. Neuronal plasticity after spinal cord injury: identification of a gene cluster driving neurite outgrowth. FASEB J 19(1):153-154. di Porzio U, Rougon G, Novotny EA, Barker JL. 1987. Dopaminergic neurons from embryonic mouse mesencephalon are enriched in culture through immunoreaction with monoclonal antibody to neural specific protein 4 and flow cytometry. Proc Natl Acad Sci U S A 84(20):7334-7338. Dickson BJ. 2001. Rho GTPases in growth cone guidance. Curr Opin Neurobiol 11(1):103-110. Dickson BJ. 2002. Molecular mechanisms of axon guidance. Science 298(5600):1959-1964. Dijkstra S, Bar PR, Gispen WH, Joosten EA. 1999. Selective stimulation of dendrite outgrowth from identified corticospinal neurons by homotopic astrocytes. Neuroscience 92(4):1331- 1342. Dimou L, Schnell L, Montani L, Duncan C, Simonen M, Schneider R, Liebscher T, Gullo M, Schwab ME. 2006. Nogo-A-deficient mice reveal strain-dependent differences in axonal regeneration. J Neurosci 26(21):5591-5603. 197   Doherty P, Ashton SV, Moore SE, Walsh FS. 1991. Morphoregulatory activities of NCAM and N-cadherin can be accounted for by G protein-dependent activation of L- and N-type neuronal Ca2+ channels. Cell 67(1):21-33. Dombrowski MA, Sasaki M, Lankford KL, Kocsis JD, Radtke C. 2006. Myelination and nodal formation of regenerated peripheral nerve fibers following transplantation of acutely prepared olfactory ensheathing cells. Brain Res 1125(1):1-8. Donnelly G, Smith RA, Breckenridge L. 1998. The effect of extracellular matrix on the growth of mouse olfactory tissue in vitro. Neuroreport 9(17):3837-3840. Doucette JR. 1984. The glial cells in the nerve fiber layer of the rat olfactory bulb. Anatomical Record 210(2):385-391. Doucette JR, Kiernan JA, Flumerfelt BA. 1983. The re-innervation of olfactory glomeruli following transection of primary olfactory axons in the central or peripheral nervous system. J Anat 137 (Pt 1):1-19. Doucette R. 1990. Glial influences on axonal growth in the primary olfactory system. Glia 3(6):433-449. Doucette R. 1991. PNS-CNS transitional zone of the first cranial nerve. J Comp Neurol 312(3):451-466. Doucette R. 1993. Glial cells in the nerve fiber layer of the main olfactory bulb of embryonic and adult mammals. Microsc Res Tech 24(2):113-130. Doucette R. 1996. Immunohistochemical localization of laminin, fibronectin and collagen type IV in the nerve fiber layer of the olfactory bulb. Int J Dev Neurosci 14(7-8):945-959. Doucette R. 2001. Modulation of axonal growth by glial cells. In: Bandy SCaC, A, editor. Inflammatory Events in Neurodegeneration: Prominent Press. p 255-273. Ducray A, Bondier JR, Michel G, Bon K, Millot JL, Propper A, Kastner A. 2002. Recovery following peripheral destruction of olfactory neurons in young and adult mice. Eur J Neurosci 15(12):1907-1917. Egea J, Klein R. 2007. Bidirectional Eph-ephrin signaling during axon guidance. Trends Cell Biol 17(5):230-238. Endo T, Spenger C, Tominaga T, Brene S, Olson L. 2007. Cortical sensory map rearrangement after spinal cord injury: fMRI responses linked to Nogo signalling. Brain 130(Pt 11):2951-2961. Engesser-Cesar C, Ichiyama RM, Nefas AL, Hill MA, Edgerton VR, Cotman CW, Anderson AJ. 2007. Wheel running following spinal cord injury improves locomotor recovery and stimulates serotonergic fiber growth. Eur J Neurosci 25(7):1931-1939. Erkanli G, Ercan F, Sirvanci S, Salik E, Yananli HR, Onat F, San T. 2007. Time-dependent changes in distribution of basic fibroblast growth factor immunoreactive cells in hippocampus after kainic acid injection in rat pups. Int J Dev Neurosci 25(6):399-407. Fairless R, Frame MC, Barnett SC. 2005. N-cadherin differentially determines Schwann cell and olfactory ensheathing cell adhesion and migration responses upon contact with astrocytes. Mol Cell Neurosci 28(2):253-263. Fallon JR. 1985. Neurite guidance by non-neuronal cells in culture: preferential outgrowth of peripheral neurites on glial as compared to nonglial cell surfaces. J Neurosci 5(12):3169- 3177. Fawcett JW. 2006. Overcoming inhibition in the damaged spinal cord. J Neurotrauma 23(3- 4):371-383. 198   Feng G, Mellor RH, Bernstein M, Keller-Peck C, Nguyen QT, Wallace M, Nerbonne JM, Lichtman JW, Sanes JR. 2000. Imaging neuronal subsets in transgenic mice expressing multiple spectral variants of GFP. Neuron 28(1):41-51. Fetz EE. 1968. Pyramidal tract effects on interneurons in the cat lumbar dorsal horn. J Neurophysiol 31(1):69-80. Finger JH, Bronson RT, Harris B, Johnson K, Przyborski SA, Ackerman SL. 2002. The netrin 1 receptors Unc5h3 and Dcc are necessary at multiple choice points for the guidance of corticospinal tract axons. J Neurosci 22(23):10346-10356. Fitch MT, Silver J. 2008. CNS injury, glial scars, and inflammation: Inhibitory extracellular matrices and regeneration failure. Exp Neurol 209(2):294-301. Fitzgerald J, Fawcett J. 2007. Repair in the central nervous system. J Bone Joint Surg Br 89(11):1413-1420. Fouad K, Schnell L, Bunge MB, Schwab ME, Liebscher T, Pearse DD. 2005. Combining Schwann cell bridges and olfactory-ensheathing glia grafts with chondroitinase promotes locomotor recovery after complete transection of the spinal cord. J Neurosci 25(5):1169- 1178. Fraher JP. 1999. The transitional zone and CNS regeneration. J Anat 194(Pt 2):161-182. Franceschini IA, Barnett SC. 1996. Low-affinity NGF-receptor and E-N-CAM expression define two types of olfactory nerve ensheathing cells that share a common lineage [published erratum appears in Dev Biol 1996 May 25;176(1):149]. Dev Biol 173(1):327-343. Francki A, Sage EH. 2001. SPARC and the kidney glomerulus: matricellular proteins exhibit diverse functions under normal and pathological conditions. Trends Cardiovasc Med 11(1):32-37. Franklin RJ. 2003. Remyelination by transplanted olfactory ensheathing cells. Anat Rec 271B(1):71-76. Franklin RJ, Gilson JM, Franceschini IA, Barnett SC. 1996. Schwann cell-like myelination following transplantation of an olfactory bulb-ensheathing cell line into areas of demyelination in the adult CNS. Glia 17(3):217-224. Freshney RI. 2000. Culture of animal cells: A manual of basic technique. New York: John Wiley and Sons Inc. Gallo G, Letourneau PC. 2004. Regulation of growth cone actin filaments by guidance cues. J Neurobiol 58(1):92-102. Garcia-Alias G, Lopez-Vales R, Fores J, Navarro X, Verdu E. 2004. Acute transplantation of olfactory ensheathing cells or Schwann cells promotes recovery after spinal cord injury in the rat. J Neurosci Res 75(5):632-641. Getchell TV, Liu H, Vaishnav RA, Kwong K, Stromberg AJ, Getchell ML. 2005. Temporal profiling of gene expression during neurogenesis and remodeling in the olfactory epithelium at short intervals after target ablation. J Neurosci Res 80(3):309-329. Ghirnikar RS, Eng LF. 1994. Astrocyte-Schwann cell interactions in culture. Glia 11(4):367-377. Ghirnikar RS, Eng LF. 1995. Chondroitin sulfate proteoglycan staining in astrocyte-Schwann cell co-cultures. Glia 14(2):145-152. Gianino S, Stein SA, Li H, Lu X, Biesiada E, Ulas J, Xu XM. 1999. Postnatal growth of corticospinal axons in the spinal cord of developing mice. Brain Res Dev Brain Res 112(2):189-204. 199   Gibson CL, Arnott GA, Clowry GJ. 2000. Plasticity in the rat spinal cord seen in response to lesions to the motor cortex during development but not to lesions in maturity. Exp Neurol 166(2):422-434. Giehl KM. 2001. Trophic dependencies of rodent corticospinal neurons. Rev Neurosci 12(1):79- 94. Giehl KM, Rohrig S, Bonatz H, Gutjahr M, Leiner B, Bartke I, Yan Q, Reichardt LF, Backus C, Welcher AA, Dethleffsen K, Mestres P, Meyer M. 2001. Endogenous brain-derived neurotrophic factor and neurotrophin-3 antagonistically regulate survival of axotomized corticospinal neurons in vivo. J Neurosci 21(10):3492-3502. Giehl KM, Tetzlaff W. 1996. BDNF and NT-3, but not NGF, prevent axotomy-induced death of rat corticospinal neurons in vivo. Eur J Neurosci 8(6):1167-1175. Gillespie LN. 2003. Regulation of axonal growth and guidance by the neurotrophin family of neurotrophic factors. Clin Exp Pharmacol Physiol 30(10):724-733. Goldblum SE, Ding X, Funk SE, Sage EH. 1994. SPARC (secreted protein acidic and rich in cysteine) regulates endothelial cell shape and barrier function. Proc Natl Acad Sci U S A 91(8):3448-3452. Golden KL, Pearse DD, Blits B, Garg MS, Oudega M, Wood PM, Bunge MB. 2007. Transduced Schwann cells promote axon growth and myelination after spinal cord injury. Exp Neurol 207(2):203-217. Gomes-Leal W, Corkill DJ, Picanco-Diniz CW. 2005. Systematic analysis of axonal damage and inflammatory response in different white matter tracts of acutely injured rat spinal cord. Brain Res 1066(1-2):57-70. Gomez VM, Averill S, King V, Yang Q, Perez ED, Chacon SC, Ward R, Nieto-Sampedro M, Priestley J, Taylor J. 2003. Transplantation of olfactory ensheathing cells fails to promote significant axonal regeneration from dorsal roots into the rat cervical cord. J Neurocytol 32(1):53-70. Gorgels TG, Oestreicher AB, de Kort EJ, Gispen WH. 1987. Immunocytochemical distribution of the protein kinase C substrate B-50 (GAP43) in developing rat pyramidal tract. Neurosci Lett 83(1-2):59-64. Gorini A, Canosi U, Devecchi E, Geroldi D, Villa RF. 2002. ATPases enzyme activities during ageing in different types of somatic and synaptic plasma membranes from rat frontal cerebral cortex. Prog Neuropsychopharmacol Biol Psychiatry 26(1):81-90. Grant KJ, Wu CC. 2007. Advances in neuromembrane proteomics: efforts towards a comprehensive analysis of membrane proteins in the brain. Brief Funct Genomic Proteomic 6(1):59-69. Graziadei PP, Levine RR, Graziadei GA. 1978. Regeneration of olfactory axons and synapse formation in the forebrain after bulbectomy in neonatal mice. Proc Natl Acad Sci U S A 75(10):5230-5234. Graziadei PP, Levine RR, Monti Graziadei GA. 1979. Plasticity of connections of the olfactory sensory neuron: regeneration into the forebrain following bulbectomy in the neonatal mouse. Neuroscience 4(6):713-727. Grever WE, Zhang S, Ge B, Duncan ID. 1999. Fractionation and enrichment of oligodendrocytes from developing human brain. J Neurosci Res 57(3):304-314. Guleria S, Gupta RK, Saksena S, Chandra A, Srivastava RN, Husain M, Rathore R, Narayana PA. 2008. Retrograde Wallerian degeneration of cranial corticospinal tracts in cervical spinal cord injury patients using diffusion tensor imaging. J Neurosci Res. 200   Guntinas-Lichius O, Wewetzer K, Tomov TL, Azzolin N, Kazemi S, Streppel M, Neiss WF, Angelov DN. 2002. Transplantation of olfactory mucosa minimizes axonal branching and promotes the recovery of vibrissae motor performance after facial nerve repair in rats. J Neurosci 22(16):7121-7131. Hanks SK, Polte TR. 1997. Signaling through focal adhesion kinase. Bioessays 19(2):137-145. Hartl S, Heil JE, Hirsekorn A, Lohr C. 2007. A novel neurotransmitter-independent communication pathway between axons and glial cells. Eur J Neurosci 25(4):945-956. Hatten ME, Liem RK. 1981. Astroglial cells provide a template for the positioning of developing cerebellar neurons in vitro. J Cell Biol 90(3):622-630. Hatten ME, Mason CA, Liem RK, Edmondson JC, Bovolenta P, Shelanski ML. 1984. Neuron- astroglial interactions in vitro and their implications for repair of CNS injury. Cent Nerv Syst Trauma 1(1):15-27. Hattori M, Osterfield M, Flanagan JG. 2000. Regulated cleavage of a contact-mediated axon repellent. Science 289(5483):1360-1365. Hayat S, Thomas A, Afshar F, Sonigra R, Wigley CB. 2003a. Manipulation of olfactory ensheathing cell signaling mechanisms: effects on their support for neurite regrowth from adult CNS neurons in coculture. Glia 44(3):232-241. Hayat S, Wigley CB, Robbins J. 2003b. Intracellular calcium handling in rat olfactory ensheathing cells and its role in axonal regeneration. Mol Cell Neurosci 22(2):259-270. Heredia M, Gascuel J, Ramon_Cueto A, Santacana M, Avila J, Masson C, Valverde F. 1998. Two novel monoclonal antibodies (1.9.E and 4.11.C) against olfactory bulb ensheathing glia. Glia 24(3):352-364. Hevner RF, Daza RA, Rubenstein JL, Stunnenberg H, Olavarria JF, Englund C. 2003. Beyond laminar fate: toward a molecular classification of cortical projection/pyramidal neurons. Dev Neurosci 25(2-4):139-151. Hiebert GW, Khodarahmi K, McGraw J, Steeves JD, Tetzlaff W. 2002. Brain-derived neurotrophic factor applied to the motor cortex promotes sprouting of corticospinal fibers but not regeneration into a peripheral nerve transplant. J Neurosci Res 69(2):160-168. Hill CE, Beattie MS, Bresnahan JC. 2001. Degeneration and sprouting of identified descending supraspinal axons after contusive spinal cord injury in the rat. Exp Neurol 171(1):153- 169. Hobson MI, Green CJ, Terenghi G. 2000. VEGF enhances intraneural angiogenesis and improves nerve regeneration after axotomy. J Anat 197 Pt 4:591-605. Hofstetter CP, Holmstrom NA, Lilja JA, Schweinhardt P, Hao J, Spenger C, Wiesenfeld-Hallin Z, Kurpad SN, Frisen J, Olson L. 2005. Allodynia limits the usefulness of intraspinal neural stem cell grafts; directed differentiation improves outcome. Nat Neurosci 8(3):346-353. Hsu JY, Stein SA, Xu XM. 2006. Development of the corticospinal tract in the mouse spinal cord: a quantitative ultrastructural analysis. Brain Res 1084(1):16-27. Huang H, Chen L, Wang H, Xiu B, Li B, Wang R, Zhang J, Zhang F, Gu Z, Li Y, Song Y, Hao W, Pang S, Sun J. 2003. Influence of patients' age on functional recovery after transplantation of olfactory ensheathing cells into injured spinal cord injury. Chin Med J (Engl) 116(10):1488-1491. Huber AB, Kolodkin AL, Ginty DD, Cloutier JF. 2003. Signaling at the growth cone: ligand- receptor complexes and the control of axon growth and guidance. Annu Rev Neurosci 26:509-563. 201   Ibanez C, Ito D, Zawadzka M, Jeffery ND, Franklin RJ. 2007. Calponin is expressed by fibroblasts and meningeal cells but not olfactory ensheathing cells in the adult peripheral olfactory system. Glia 55(2):144-151. Imaizumi T, Lankford KL, Kocsis JD. 2000. Transplantation of olfactory ensheathing cells or Schwann cells restores rapid and secure conduction across the transected spinal cord. Brain Research 854(1-2):70-78. Imaizumi T, Lankford KL, Waxman SG, Greer CA, Kocsis JD. 1998. Transplanted olfactory ensheathing cells remyelinate and enhance axonal conduction in the demyelinated dorsal columns of the rat spinal cord. Journal of Neuroscience 18(16):6176-6185. Inman DM, Steward O. 2003. Ascending sensory, but not other long-tract axons, regenerate into the connective tissue matrix that forms at the site of a spinal cord injury in mice. J Comp Neurol 462(4):431-449. Iwaniuk AN, Pellis SM, Whishaw IQ. 1999. Is digital dexterity really related to corticospinal projections?: a re-analysis of the Heffner and Masterton data set using modern comparative statistics. Behav Brain Res 101(2):173-187. Iwema CL, Fang H, Kurtz DB, Youngentob SL, Schwob JE. 2004. Odorant receptor expression patterns are restored in lesion-recovered rat olfactory epithelium. J Neurosci 24(2):356- 369. Jahed A, Rowland JW, McDonald T, Boyd JG, Doucette R, Kawaja MD. 2007. Olfactory ensheathing cells express smooth muscle alpha-actin in vitro and in vivo. J Comp Neurol 503(2):209-223. Jahromi BS, Robitaille R, Charlton MP. 1992. Transmitter release increases intracellular calcium in perisynaptic Schwann cells in situ. Neuron 8(6):1069-1077. Jang JH, Nam TS, Paik KS, Leem JW. 2004. Involvement of peripherally released substance P and calcitonin gene-related peptide in mediating mechanical hyperalgesia in a traumatic neuropathy model of the rat. Neurosci Lett 360(3):129-132. Jani HR, Raisman G. 2004. Ensheathing cell cultures from the olfactory bulb and mucosa. Glia 47(2):130-137. Jendraschak E, Sage EH. 1996. Regulation of angiogenesis by SPARC and angiostatin: implications for tumor cell biology. Semin Cancer Biol 7(3):139-146. Joosten EA, Bar DP. 1999. Axon guidance of outgrowing corticospinal fibres in the rat. J Anat 194 ( Pt 1):15-32. Joosten EA, Gribnau AA. 1989. Astrocytes and guidance of outgrowing corticospinal tract axons in the rat. An immunocytochemical study using anti-vimentin and anti-glial fibrillary acidic protein. Neuroscience 31(2):439-452. Joosten EA, Reshilov LN, Gispen WH, Bar PR. 1996. Embryonic form of N-CAM and development of the rat corticospinal tract; immuno-electron microscopical localization during spinal white matter ingrowth. Brain Res Dev Brain Res 94(1):99-105. Joosten EA, Schuitman RL, Vermelis ME, Dederen PJ. 1992. Postnatal development of the ipsilateral corticospinal component in rat spinal cord: a light and electron microscopic anterograde HRP study. J Comp Neurol 326(1):133-146. Joosten EA, van Eden CG. 1989. An anterograde tracer study on the development of corticospinal projections from the medial prefrontal cortex in the rat. Brain Res Dev Brain Res 45(2):313-319. Jordan PM, Cain LD, Wu P. 2008. Astrocytes enhance long-term survival of cholinergic neurons differentiated from human fetal neural stem cells. J Neurosci Res 86(1):35-47. 202   Julliard AK, Hartmann DJ. 1998. Spatiotemporal patterns of expression of extracellular matrix molecules in the developing and adult rat olfactory system. Neuroscience 84(4):1135- 1150. Junger H, Junger WG. 1998. CNTF and GDNF, but not NT-4, support corticospinal motor neuron growth via direct mechanisms. Neuroreport 9(16):3749-3754. Junger H, Varon S. 1997. Neurotrophin-4 (NT-4) and glial cell line-derived neurotrophic factor (GDNF) promote the survival of corticospinal motor neurons of neonatal rats in vitro. Brain Res 762(1-2):56-60. Kafitz KW, Greer CA. 1997. Role of laminin in axonal extension from olfactory receptor cells. J Neurobiol 32(3):298-310. Kafitz KW, Greer CA. 1998. The influence of ensheathing cells on olfactory receptor cell neurite outgrowth in vitro. Annals of the New York Academy of Sciences 855:266-269. Kafitz KW, Greer CA. 1999. Olfactory ensheathing cells promote neurite extension from embryonic olfactory receptor cells in vitro. Glia 25(2):99-110. Kalil K, Reh T. 1979. Regrowth of severed axons in the neonatal central nervous system: establishment of normal connections. Science 205(4411):1158-1161. Kalous A, Osborne PB, Keast JR. 2007. Acute and chronic changes in dorsal horn innervation by primary afferents and descending supraspinal pathways after spinal cord injury. J Comp Neurol 504(3):238-253. Kamada T, Koda M, Dezawa M, Yoshinaga K, Hashimoto M, Koshizuka S, Nishio Y, Moriya H, Yamazaki M. 2005. Transplantation of bone marrow stromal cell-derived Schwann cells promotes axonal regeneration and functional recovery after complete transection of adult rat spinal cord. J Neuropathol Exp Neurol 64(1):37-45. Kamei N, Oishi Y, Tanaka N, Ishida O, Fujiwara Y, Ochi M. 2004. Neural progenitor cells promote corticospinal axon growth in organotypic co-cultures. Neuroreport 15(17):2579- 2583. Kamei N, Tanaka N, Oishi Y, Hamasaki T, Nakanishi K, Sakai N, Ochi M. 2007. BDNF, NT-3, and NGF released from transplanted neural progenitor cells promote corticospinal axon growth in organotypic cocultures. Spine 32(12):1272-1278. Kamiyama T, Yoshioka N, Sakurai M. 2006. Synapse elimination in the corticospinal projection during the early postnatal period. J Neurophysiol 95(4):2304-2313. Kao CC, Chang LW, Bloodworth JM, Jr. 1977. The mechanism of spinal cord cavitation follwing spinal cord transection. Part 2. Electron microscopic observations. J Neurosurg 46(6):745-756. Kato T, Honmou O, Uede T, Hashi K, Kocsis JD. 2000. Transplantation of human olfactory ensheathing cells elicits remyelination of demyelinated rat spinal cord. Glia 30(3):209- 218. Kauppila T. 1998. Correlation between autotomy-behavior and current theories of neuropathic pain. Neurosci Biobehav Rev 23(1):111-129. Kelly KA, Allport JR, Yu AM, Sinh S, Sage EH, Gerszten RE, Weissleder R. 2007. SPARC is a VCAM-1 counter-ligand that mediates leukocyte transmigration. J Leukoc Biol 81(3):748-756. Key B. 1998. Molecular development of the olfactory nerve pathway. Ann N Y Acad Sci 855:76-82. 203   Keyvan-Fouladi N, Raisman G, Li Y. 2003. Functional repair of the corticospinal tract by delayed transplantation of olfactory ensheathing cells in adult rats. J Neurosci 23(28):9428-9434. Kim BG, Dai HN, Lynskey JV, McAtee M, Bregman BS. 2006. Degradation of chondroitin sulfate proteoglycans potentiates transplant-mediated axonal remodeling and functional recovery after spinal cord injury in adult rats. J Comp Neurol 497(2):182-198. Kim BG, Dai HN, McAtee M, Bregman BS. 2008. Modulation of dendritic spine remodeling in the motor cortex following spinal cord injury: Effects of environmental enrichment and combinatorial treatment with transplants and neurotrophin-3. J Comp Neurol 508(3):473- 486. Kiryushko D, Berezin V, Bock E. 2004. Regulators of neurite outgrowth: role of cell adhesion molecules. Ann N Y Acad Sci 1014:140-154. Knoblauch A, Will C, Goncharenko G, Ludwig S, Wixler V. 2007. The binding of Mss4 to alpha-integrin subunits regulates matrix metalloproteinase activation and fibronectin remodeling. FASEB J 21(2):497-510. Kriegstein AR, Noctor SC. 2004. Patterns of neuronal migration in the embryonic cortex. Trends Neurosci 27(7):392-399. Krudewig C, Deschl U, Wewetzer K. 2006. Purification and in vitro characterization of adult canine olfactory ensheathing cells. Cell Tissue Res 326(3):687-696. Kuang RZ, Merline M, Kalil K. 1994. Topographic specificity of corticospinal connections formed in explant coculture. Development 120(7):1937-1947. Kullander K, Croll SD, Zimmer M, Pan L, McClain J, Hughes V, Zabski S, DeChiara TM, Klein R, Yancopoulos GD, Gale NW. 2001. Ephrin-B3 is the midline barrier that prevents corticospinal tract axons from recrossing, allowing for unilateral motor control. Genes Dev 15(7):877-888. Kumar R, Hayat S, Felts P, Bunting S, Wigley C. 2005. Functional differences and interactions between phenotypic subpopulations of olfactory ensheathing cells in promoting CNS axonal regeneration. Glia 50(1):12-20. Kyriakides TR, Bornstein P. 2003. Matricellular proteins as modulators of wound healing and the foreign body response. Thromb Haemost 90(6):986-992. Lakatos A, Barnett SC, Franklin RJ. 2003. Olfactory ensheathing cells induce less host astrocyte response and chondroitin sulphate proteoglycan expression than Schwann cells following transplantation into adult CNS white matter. Exp Neurol 184(1):237-246. Lakatos A, Franklin RJ. 2002. Transplant mediated repair of the central nervous system: an imminent solution? Curr Opin Neurol 15(6):701-705. Lakatos A, Franklin RJ, Barnett SC. 2000. Olfactory ensheathing cells and Schwann cells differ in their in vitro interactions with astrocytes. Glia 32(3):214-225. Le Roux PD, Reh TA. 1994. Regional differences in glial-derived factors that promote dendritic outgrowth from mouse cortical neurons in vitro. J Neurosci 14(8):4639-4655. Leamey CA, Glendining KA, Kreiman G, Kang ND, Wang KH, Fassler R, Sawatari A, Tonegawa S, Sur M. 2008. Differential gene expression between sensory neocortical areas: potential roles for Ten_m3 and Bcl6 in patterning visual and somatosensory pathways. Cereb Cortex 18(1):53-66. Leaver SG, Harvey AR, Plant GW. 2006. Adult olfactory ensheathing glia promote the long- distance growth of adult retinal ganglion cell neurites in vitro. Glia 53(5):467-476. 204   Ledda F, Paratcha G, Ibanez CF. 2002. Target-derived GFRalpha1 as an attractive guidance signal for developing sensory and sympathetic axons via activation of Cdk5. Neuron 36(3):387-401. Lemon RN, Griffiths J. 2005. Comparing the function of the corticospinal system in different species: organizational differences for motor specialization? Muscle Nerve 32(3):261- 279. Li H, Leung TC, Hoffman S, Balsamo J, Lilien J. 2000. Coordinate regulation of cadherin and integrin function by the chondroitin sulfate proteoglycan neurocan. J Cell Biol 149(6):1275-1288. Li M, Shibata A, Li C, Braun PE, McKerracher L, Roder J, Kater SB, David S. 1996. Myelin- associated glycoprotein inhibits neurite/axon growth and causes growth cone collapse. J Neurosci Res 46(4):404-414. Li S, Liu BP, Budel S, Li M, Ji B, Walus L, Li W, Jirik A, Rabacchi S, Choi E, Worley D, Sah DW, Pepinsky B, Lee D, Relton J, Strittmatter SM. 2004a. Blockade of Nogo-66, myelin- associated glycoprotein, and oligodendrocyte myelin glycoprotein by soluble Nogo-66 receptor promotes axonal sprouting and recovery after spinal injury. J Neurosci 24(46):10511-10520. Li Y, Carlstedt T, Berthold CH, Raisman G. 2004b. Interaction of transplanted olfactory- ensheathing cells and host astrocytic processes provides a bridge for axons to regenerate across the dorsal root entry zone. Exp Neurol 188(2):300-308. Li Y, Decherchi P, Raisman G. 2003a. Transplantation of olfactory ensheathing cells into spinal cord lesions restores breathing and climbing. J Neurosci 23(3):727-731. Li Y, Field PM, Raisman G. 1997. Repair of adult rat corticospinal tract by transplants of olfactory ensheathing cells. Science 277(5334):2000-2002. Li Y, Field PM, Raisman G. 1998. Regeneration of adult rat corticospinal axons induced by transplanted olfactory ensheathing cells. J Neurosci 18(24):10514-10524. Li Y, Field PM, Raisman G. 2005. Olfactory ensheathing cells and olfactory nerve fibroblasts maintain continuous open channels for regrowth of olfactory nerve fibres. Glia 52(3):245-251. Li Y, Sauve Y, Li D, Lund RD, Raisman G. 2003b. Transplanted olfactory ensheathing cells promote regeneration of cut adult rat optic nerve axons. J Neurosci 23(21):7783-7788. Li Y, Yamamoto M, Raisman G, Choi D, Carlstedt T. 2007. An experimental model of ventral root repair showing the beneficial effect of transplanting olfactory ensheathing cells. Neurosurgery 60(4):734-740; discussion 740-731. Liaw L, Crawford HC. 1999. Functions of the extracellular matrix and matrix degrading proteases during tumor progression. Braz J Med Biol Res 32(7):805-812. Liebscher T, Schnell L, Schnell D, Scholl J, Schneider R, Gullo M, Fouad K, Mir A, Rausch M, Kindler D, Hamers FP, Schwab ME. 2005. Nogo-A antibody improves regeneration and locomotion of spinal cord-injured rats. Ann Neurol 58(5):706-719. Liesi P, Kirkwood T, Vaheri A. 1986. Fibronectin is expressed by astrocytes cultured from embryonic and early postnatal rat brain. Exp Cell Res 163(1):175-185. Liesi P, Silver J. 1988. Is astrocyte laminin involved in axon guidance in the mammalian CNS? Dev Biol 130(2):774-785. Lim KC, Lim ST, Federoff HJ. 2003. Neurotrophin secretory pathways and synaptic plasticity. Neurobiol Aging 24(8):1135-1145. 205   Lingor P, Tonges L, Pieper N, Bermel C, Barski E, Planchamp V, Bahr M. 2008. ROCK inhibition and CNTF interact on intrinsic signalling pathways and differentially regulate survival and regeneration in retinal ganglion cells. Brain 131(Pt 1):250-263. Lipson AC, Widenfalk J, Lindqvist E, Ebendal T, Olson L. 2003. Neurotrophic properties of olfactory ensheathing glia. Exp Neurol 180(2):167-171. Lobo MK, Karsten SL, Gray M, Geschwind DH, Yang XW. 2006. FACS-array profiling of striatal projection neuron subtypes in juvenile and adult mouse brains. Nat Neurosci 9(3):443-452. Lopez-Vales R, Fores J, Navarro X, Verdu E. 2007. Chronic transplantation of olfactory ensheathing cells promotes partial recovery after complete spinal cord transection in the rat. Glia 55(3):303-311. Lopez-Vales R, Fores J, Verdu E, Navarro X. 2006. Acute and delayed transplantation of olfactory ensheathing cells promote partial recovery after complete transection of the spinal cord. Neurobiol Dis 21(1):57-68. Lopez-Vales R, Garcia-Alias G, Fores J, Navarro X, Verdu E. 2004. Increased expression of cyclo-oxygenase 2 and vascular endothelial growth factor in lesioned spinal cord by transplanted olfactory ensheathing cells. J Neurotrauma 21(8):1031-1043. Loy DN, Crawford CH, Darnall JB, Burke DA, Onifer SM, Whittemore SR. 2002. Temporal progression of angiogenesis and basal lamina deposition after contusive spinal cord injury in the adult rat. J Comp Neurol 445(4):308-324. Lu J, Feron F, Ho SM, Mackay-Sim A, Waite PM. 2001. Transplantation of nasal olfactory tissue promotes partial recovery in paraplegic adult rats. Brain Res 889(1-2):344-357. Lu J, Feron F, Mackay-Sim A, Waite PM. 2002. Olfactory ensheathing cells promote locomotor recovery after delayed transplantation into transected spinal cord. Brain 125(Pt 1):14-21. Lu P, Yang H, Culbertson M, Graham L, Roskams AJ, Tuszynski MH. 2006. Olfactory ensheathing cells do not exhibit unique migratory or axonal growth-promoting properties after spinal cord injury. J Neurosci 26(43):11120-11130. Mackay-Sim A, Chuah MI. 2000. Neurotrophic factors in the primary olfactory pathway. Prog Neurobiol 62(5):527-559. Magal E, Louis JC, Oudega M, Varon S. 1993. CNTF promotes the survival of neonatal rat corticospinal neurons in vitro. Neuroreport 4(6):779-782. Mann F, Peuckert C, Dehner F, Zhou R, Bolz J. 2002. Ephrins regulate the formation of terminal axonal arbors during the development of thalamocortical projections. Development 129(16):3945-3955. Mann F, Rougon G. 2007. Mechanisms of axon guidance: membrane dynamics and axonal transport in semaphorin signalling. J Neurochem 102(2):316-323. Marquardt T, Shirasaki R, Ghosh S, Andrews SE, Carter N, Hunter T, Pfaff SL. 2005. Coexpressed EphA receptors and ephrin-A ligands mediate opposing actions on growth cone navigation from distinct membrane domains. Cell 121(1):127-139. Masmoudi-Kouki O, Gandolfo P, Leprince J, Vaudry D, Pelletier G, Fournier A, Vaudry H, Tonon MC. 2006. PACAP stimulates biosynthesis and release of endozepines from rat astrocytes. Ann N Y Acad Sci 1070:411-416. Mazzola JL, Sirover MA. 2003. Subcellular localization of human glyceraldehyde-3-phosphate dehydrogenase is independent of its glycolytic function. Biochim Biophys Acta 1622(1):50-56. 206   McCouch GP, Austin GM, Liu CN, Liu CY. 1958. Sprouting as a cause of spasticity. J Neurophysiol 21(3):205-216. McKerracher L, David S, Jackson DL, Kottis V, Dunn RJ, Braun PE. 1994. Identification of myelin-associated glycoprotein as a major myelin-derived inhibitor of neurite growth. Neuron 13(4):805-811. McLaughlin T, Hindges R, Yates PA, O'Leary DD. 2003. Bifunctional action of ephrin-B1 as a repellent and attractant to control bidirectional branch extension in dorsal-ventral retinotopic mapping. Development 130(11):2407-2418. Miragall F, Kadmon G, Schachner M. 1989. Expression of L1 and N-CAM cell adhesion molecules during development of the mouse olfactory system. Developmental Biology 135(2):272-286. Molnar Z, Cheung AF. 2006. Towards the classification of subpopulations of layer V pyramidal projection neurons. Neurosci Res 55(2):105-115. Molyneaux BJ, Arlotta P, Hirata T, Hibi M, Macklis JD. 2005. Fezl is required for the birth and specification of corticospinal motor neurons. Neuron 47(6):817-831. Moonen G, Cam Y, Sensenbrenner M, Mandel P. 1975. Variability of the effects of serum-free medium, dibutyryl-cyclic AMP or theophylline on the morphology of cultured new-born rat astroblasts. Cell Tissue Res 163(3):365-372. Moreau-Fauvarque C, Kumanogoh A, Camand E, Jaillard C, Barbin G, Boquet I, Love C, Jones EY, Kikutani H, Lubetzki C, Dusart I, Chedotal A. 2003. The transmembrane semaphorin Sema4D/CD100, an inhibitor of axonal growth, is expressed on oligodendrocytes and upregulated after CNS lesion. J Neurosci 23(27):9229-9239. Mothe AJ, Brown IR. 2001. Expression of mRNA encoding extracellular matrix glycoproteins SPARC and SC1 is temporally and spatially regulated in the developing cochlea of the rat inner ear. Hear Res 155(1-2):161-174. Mukhopadhyay G, Doherty P, Walsh FS, Crocker PR, Filbin MT. 1994. A novel role for myelin- associated glycoprotein as an inhibitor of axonal regeneration. Neuron 13(3):757-767. Muller A, Hauk TG, Fischer D. 2007. Astrocyte-derived CNTF switches mature RGCs to a regenerative state following inflammatory stimulation. Brain 130(Pt 12):3308-3320. Muller HW, Beckh S, Seifert W. 1984. Neurotrophic factor for central neurons. Proc Natl Acad Sci U S A 81(4):1248-1252. Muller HW, Seifert W. 1982. A neurotrophic factor (NTF) released from primary glial cultures supports survival and fiber outgrowth of cultured hippocampal neurons. J Neurosci Res 8(2-3):195-204. Murai KK, Pasquale EB. 2005. New exchanges in eph-dependent growth cone dynamics. Neuron 46(2):161-163. Murphy M, Dutton R, Koblar S, Cheema S, Bartlett P. 1997. Cytokines which signal through the LIF receptor and their actions in the nervous system. Prog Neurobiol 52(5):355-378. Nan B, Getchell ML, Partin JV, Getchell TV. 2001. Leukemia inhibitory factor, interleukin-6, and their receptors are expressed transiently in the olfactory mucosa after target ablation. J Comp Neurol 435(1):60-77. Nash HH, Borke RC, Anders JJ. 2001. New method of purification for establishing primary cultures of ensheathing cells from the adult olfactory bulb. Glia 34(2):81-87. Nash HH, Borke RC, Anders JJ. 2002. Ensheathing cells and methylprednisolone promote axonal regeneration and functional recovery in the lesioned adult rat spinal cord. J Neurosci 22(16):7111-7120. 207   Navarro X, Valero A, Gudino G, Fores J, Rodriguez FJ, Verdu E, Pascual R, Cuadras J, Nieto_Sampedro M. 1999. Ensheathing glia transplants promote dorsal root regeneration and spinal reflex restitution after multiple lumbar rhizotomy. Annals of Neurology 45(2):207-215. Nave KA, Salzer JL. 2006. Axonal regulation of myelination by neuregulin 1. Curr Opin Neurobiol 16(5):492-500. Neugebauer KM, Tomaselli KJ, Lilien J, Reichardt LF. 1988. N-cadherin, NCAM, and integrins promote retinal neurite outgrowth on astrocytes in vitro. J Cell Biol 107(3):1177-1187. Neumann J, Gunzer M, Gutzeit HO, Ullrich O, Reymann KG, Dinkel K. 2006. Microglia provide neuroprotection after ischemia. FASEB J 20(6):714-716. Noble M, Fok-Seang J, Cohen J. 1984. Glia are a unique substrate for the in vitro growth of central nervous system neurons. J Neurosci 4(7):1892-1903. Norose K, Clark JI, Syed NA, Basu A, Heber-Katz E, Sage EH, Howe CC. 1998. SPARC deficiency leads to early-onset cataractogenesis. Invest Ophthalmol Vis Sci 39(13):2674- 2680. O'Toole DA, West AK, Chuah MI. 2007. Effect of olfactory ensheathing cells on reactive astrocytes in vitro. Cell Mol Life Sci 64(10):1303-1309. Oblander SA, Ensslen-Craig SE, Longo FM, Brady-Kalnay SM. 2007. E-cadherin promotes retinal ganglion cell neurite outgrowth in a protein tyrosine phosphatase-mu-dependent manner. Mol Cell Neurosci 34(3):481-492. Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H, Pandey A, Mann M. 2002. Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics. Mol Cell Proteomics 1(5):376-386. Oudega M, Xu XM. 2006. Schwann cell transplantation for repair of the adult spinal cord. J Neurotrauma 23(3-4):453-467. Ourednik J, Ourednik V, Bastmeyer M, Schachner M. 2001. Ectopic expression of the neural cell adhesion molecule L1 in astrocytes leads to changes in the development of the corticospinal tract. Eur J Neurosci 14(9):1464-1474. Ozdinler PH, Macklis JD. 2006. IGF-I specifically enhances axon outgrowth of corticospinal motor neurons. Nat Neurosci 9(11):1371-1381. Ozog MA, Modha G, Church J, Slotwinska D, Naus CC. 2007. Co-administration of CNTF with its soluble receptor protects against neuronal death and enhances neurite outgrowth. J Biol Chem. Paino CL, Fernandez-Valle C, Bates ML, Bunge MB. 1994. Regrowth of axons in lesioned adult rat spinal cord: promotion by implants of cultured Schwann cells. J Neurocytol 23(7):433-452. Paratcha G, Ledda F, Ibanez CF. 2003. The neural cell adhesion molecule NCAM is an alternative signaling receptor for GDNF family ligands. Cell 113(7):867-879. Pascual JI, Gudino-Cabrera G, Insausti R, Nieto-Sampedro M. 2002. Spinal implants of olfactory ensheathing cells promote axon regeneration and bladder activity after bilateral lumbosacral dorsal rhizotomy in the adult rat. J Urol 167(3):1522-1526. Pastrana E, Moreno-Flores MT, Avila J, Wandosell F, Minichiello L, Diaz-Nido J. 2007. BDNF production by olfactory ensheathing cells contributes to axonal regeneration of cultured adult CNS neurons. Neurochem Int 50(3):491-498. 208   Pastrana E, Moreno-Flores MT, Gurzov EN, Avila J, Wandosell F, Diaz-Nido J. 2006. Genes associated with adult axon regeneration promoted by olfactory ensheathing cells: a new role for matrix metalloproteinase 2. J Neurosci 26(20):5347-5359. Patel TD, Jackman A, Rice FL, Kucera J, Snider WD. 2000. Development of sensory neurons in the absence of NGF/TrkA signaling in vivo. Neuron 25(2):345-357. Paxinos G, Watson, C. 1998. The rat brain in stereotaxic coordinates, 4th edition. Pearse DD, Marcillo AE, Oudega M, Lynch MP, Wood PM, Bunge MB. 2004. Transplantation of Schwann cells and olfactory ensheathing glia after spinal cord injury: does pretreatment with methylprednisolone and interleukin-10 enhance recovery? J Neurotrauma 21(9):1223-1239. Pellitteri R, Spatuzza M, Russo A, Stanzani S. 2007. Olfactory ensheathing cells exert a trophic effect on the hypothalamic neurons in vitro. Neurosci Lett 417(1):24-29. Petit A, Kennedy TE, Bagnard D, Doucet G. 2005. Membrane-associated guidance cues direct the innervation of forebrain and midbrain by dorsal raphe-derived serotonergic axons. Eur J Neurosci 22(3):552-568. Pfrieger FW, Barres BA. 1997. Synaptic efficacy enhanced by glial cells in vitro. Science 277(5332):1684-1687. Pizzi MA, Crowe MJ. 2006. Transplantation of fibroblasts that overexpress matrix metalloproteinase-3 into the site of spinal cord injury in rats. J Neurotrauma 23(12):1750- 1765. Plant GW, Bates ML, Bunge MB. 2001. Inhibitory proteoglycan immunoreactivity is higher at the caudal than the rostral Schwann cell graft-transected spinal cord interface. Mol Cell Neurosci 17(3):471-487. Plant GW, Currier PF, Cuervo EP, Bates ML, Pressman Y, Bunge MB, Wood PM. 2002. Purified adult ensheathing glia fail to myelinate axons under culture conditions that enable Schwann cells to form myelin. J Neurosci 22(14):6083-6091. Pollard JD, McLeod JG, Gye RS. 1973. Regeneration through peripheral nerve allografts. An electrophysiological and histological study following the use of immunosuppressive therapy. Arch Neurol 28(1):31-37. Polleux F, Ghosh A. 2002. The slice overlay assay: a versatile tool to study the influence of extracellular signals on neuronal development. Sci STKE 2002(136):PL9. Polleux F, Giger RJ, Ginty DD, Kolodkin AL, Ghosh A. 1998. Patterning of cortical efferent projections by semaphorin-neuropilin interactions. Science 282(5395):1904-1906. Polleux F, Morrow T, Ghosh A. 2000. Semaphorin 3A is a chemoattractant for cortical apical dendrites. Nature 404(6778):567-573. Powell EM, Meiners S, DiProspero NA, Geller HM. 1997. Mechanisms of astrocyte-directed neurite guidance. Cell Tissue Res 290(2):385-393. Prewitt CM, Niesman IR, Kane CJ, Houle JD. 1997. Activated macrophage/microglial cells can promote the regeneration of sensory axons into the injured spinal cord. Exp Neurol 148(2):433-443. Puche AC, Baker H. 2007. Olfactory cell derivation and migration. J Mol Histol 38(6):513-515. Rabinovich SS, Seledtsov VI, Poveschenko OV, Senuykov VV, Taraban VY, Yarochno VI, Kolosov NG, Savchenko SA, Kozlov VA. 2003. Transplantation treatment of spinal cord injury patients. Biomed Pharmacother 57(9):428-433. 209   Radtke C, Akiyama Y, Brokaw J, Lankford KL, Wewetzer K, Fodor WL, Kocsis JD. 2004. Remyelination of the nonhuman primate spinal cord by transplantation of H-transferase transgenic adult pig olfactory ensheathing cells. FASEB J 18(2):335-337. Raisman G, Li Y. 2007. Repair of neural pathways by olfactory ensheathing cells. Nat Rev Neurosci 8(4):312-319. Ramer LM, Au E, Richter MW, Liu J, Tetzlaff W, Roskams AJ. 2004a. Peripheral olfactory ensheathing cells reduce scar and cavity formation and promote regeneration after spinal cord injury. J Comp Neurol 473(1):1-15. Ramer LM, Richter MW, Roskams AJ, Tetzlaff W, Ramer MS. 2004b. Peripherally-derived olfactory ensheathing cells do not promote primary afferent regeneration following dorsal root injury. Glia 47(2):189-206. Ramon-Cueto A. 2000. Olfactory ensheathing glia transplantation into the injured spinal cord. Progress in Brain Research 128:265-272. Ramon-Cueto A, Avila J. 1998. Olfactory ensheathing glia: properties and function. Brain Research Bulletin 46(3):175-187. Ramon-Cueto A, Cordero MI, Santos-Benito FF, Avila J. 2000. Functional recovery of paraplegic rats and motor axon regeneration in their spinal cords by olfactory ensheathing glia. Neuron 25(2):425-435. Ramon-Cueto A, Nieto-Sampedro M. 1992. Glial cells from adult rat olfactory bulb: immunocytochemical properties of pure cultures of ensheathing cells. Neuroscience 47(1):213-220. Ramon-Cueto A, Nieto-Sampedro M. 1994. Regeneration into the spinal cord of transected dorsal root axons is promoted by ensheathing glia transplants. Experimental Neurology 127(2):232-244. Ramon-Cueto A, Plant GW, Avila J, Bunge MB. 1998. Long-distance axonal regeneration in the transected adult rat spinal cord is promoted by olfactory ensheathing glia transplants. J Neurosci 18(10):3803-3815. Ranscht B. 2000. Cadherins: molecular codes for axon guidance and synapse formation. Int J Dev Neurosci 18(7):643-651. Rapalino O, Lazarov-Spiegler O, Agranov E, Velan GJ, Yoles E, Fraidakis M, Solomon A, Gepstein R, Katz A, Belkin M, Hadani M, Schwartz M. 1998. Implantation of stimulated homologous macrophages results in partial recovery of paraplegic rats. Nat Med 4(7):814-821. Rempel SA, Ge S, Gutierrez JA. 1999. SPARC: a potential diagnostic marker of invasive meningiomas. Clin Cancer Res 5(2):237-241. Richardson PM, McGuinness UM, Aguayo AJ. 1980. Axons from CNS neurons regenerate into PNS grafts. Nature 284(5753):264-265. Richter MW, Fletcher PA, Liu J, Tetzlaff W, Roskams AJ. 2005. Lamina propria and olfactory bulb ensheathing cells exhibit differential integration and migration and promote differential axon sprouting in the lesioned spinal cord. J Neurosci 25(46):10700-10711. Richter MW, Roskams AJ. 2008. Olfactory ensheathing cell transplantation following spinal cord injury: hype or hope? Exp Neurol 209(2):353-367. Riddell JS, Enriquez-Denton M, Toft A, Fairless R, Barnett SC. 2004. Olfactory ensheathing cell grafts have minimal influence on regeneration at the dorsal root entry zone following rhizotomy. Glia 47(2):150-167. 210   Ridley AJ, Hall A. 1992. The small GTP-binding protein rho regulates the assembly of focal adhesions and actin stress fibers in response to growth factors. Cell 70(3):389-399. Rieger A, Deitmer JW, Lohr C. 2007. Axon-glia communication evokes calcium signaling in olfactory ensheathing cells of the developing olfactory bulb. Glia 55(4):352-359. Riehl R, Johnson K, Bradley R, Grunwald GB, Cornel E, Lilienbaum A, Holt CE. 1996. Cadherin function is required for axon outgrowth in retinal ganglion cells in vivo. Neuron 17(5):837-848. Robitaille R. 1998. Modulation of synaptic efficacy and synaptic depression by glial cells at the frog neuromuscular junction. Neuron 21(4):847-855. Roisen FJ, Klueber KM, Lu CL, Hatcher LM, Dozier A, Shields CB, Maguire S. 2001. Adult human olfactory stem cells. Brain Res 890(1):11-22. Rolf B, Bastmeyer M, Schachner M, Bartsch U. 2002. Pathfinding errors of corticospinal axons in neural cell adhesion molecule-deficient mice. J Neurosci 22(19):8357-8362. Ropper AH, Fisher CM, Kleinman GM. 1979. Pyramidal infarction in the medulla: a cause of pure motor hemiplegia sparing the face. Neurology 29(1):91-95. Roskams AJ, Bethel MA, Hurt KJ, Ronnett GV. 1996. Sequential expression of Trks A, B, and C in the regenerating olfactory neuroepithelium. J Neurosci 16(4):1294-1307. Rouiller I, Xu XP, Amann KJ, Egile C, Nickell S, Nicastro D, Li R, Pollard TD, Volkmann N, Hanein D. 2008. The structural basis of actin filament branching by the Arp2/3 complex. J Cell Biol 180(5):887-895. Rousselet A, Autillo-Touati A, Araud D, Prochiantz A. 1990. In vitro regulation of neuronal morphogenesis and polarity by astrocyte-derived factors. Dev Biol 137(1):33-45. Rudge JS, Silver J. 1990. Inhibition of neurite outgrowth on astroglial scars in vitro. J Neurosci 10(11):3594-3603. Ruitenberg MJ, Levison DB, Lee SV, Verhaagen J, Harvey AR, Plant GW. 2005. NT-3 expression from engineered olfactory ensheathing glia promotes spinal sparing and regeneration. Brain. Ruitenberg MJ, Plant GW, Hamers FP, Wortel J, Blits B, Dijkhuizen PA, Gispen WH, Boer GJ, Verhaagen J. 2003. Ex vivo adenoviral vector-mediated neurotrophin gene transfer to olfactory ensheathing glia: effects on rubrospinal tract regeneration, lesion size, and functional recovery after implantation in the injured rat spinal cord. J Neurosci 23(18):7045-7058. Runeberg-Roos P, Saarma M. 2007. Neurotrophic factor receptor RET: structure, cell biology, and inherited diseases. Ann Med:1-9. Sakamoto T, Kondo K, Kashio A, Suzukawa K, Yamasoba T. 2007. Methimazole-induced cell death in rat olfactory receptor neurons occurs via apoptosis triggered through mitochondrial cytochrome c-mediated caspase-3 activation pathway. J Neurosci Res 85(3):548-557. Sandvig A, Berry M, Barrett LB, Butt A, Logan A. 2004. Myelin-, reactive glia-, and scar- derived CNS axon growth inhibitors: expression, receptor signaling, and correlation with axon regeneration. Glia 46(3):225-251. Santos-Silva A, Fairless R, Frame MC, Montague P, Smith GM, Toft A, Riddell JS, Barnett SC. 2007. FGF/heparin differentially regulates Schwann cell and olfactory ensheathing cell interactions with astrocytes: a role in astrocytosis. J Neurosci 27(27):7154-7167. 211   Sasaki M, Hains BC, Lankford KL, Waxman SG, Kocsis JD. 2006. Protection of corticospinal tract neurons after dorsal spinal cord transection and engraftment of olfactory ensheathing cells. Glia 53(4):352-359. Sasaki M, Lankford KL, Zemedkun M, Kocsis JD. 2004. Identified olfactory ensheathing cells transplanted into the transected dorsal funiculus bridge the lesion and form myelin. J Neurosci 24(39):8485-8493. Sato K, Hayashi T, Sasaki C, Iwai M, Li F, Manabe Y, Seki T, Abe K. 2001. Temporal and spatial differences of PSA-NCAM expression between young-adult and aged rats in normal and ischemic brains. Brain Res 922(1):135-139. Schlaepfer DD, Hanks SK, Hunter T, van der Geer P. 1994. Integrin-mediated signal transduction linked to Ras pathway by GRB2 binding to focal adhesion kinase. Nature 372(6508):786-791. Schlosser G. 2005. Evolutionary origins of vertebrate placodes: insights from developmental studies and from comparisons with other deuterostomes. J Exp Zoolog B Mol Dev Evol 304(4):347-399. Schmid RS, Graff RD, Schaller MD, Chen S, Schachner M, Hemperly JJ, Maness PF. 1999. NCAM stimulates the Ras-MAPK pathway and CREB phosphorylation in neuronal cells. J Neurobiol 38(4):542-558. Schmid RS, Pruitt WM, Maness PF. 2000. A MAP kinase-signaling pathway mediates neurite outgrowth on L1 and requires Src-dependent endocytosis. J Neurosci 20(11):4177-4188. Schmitt C, Miranpuri GS, Dhodda VK, Isaacson J, Vemuganti R, Resnick DK. 2006. Changes in spinal cord injury-induced gene expression in rat are strain-dependent. Spine J 6(2):113- 119. Schnell L, Schneider R, Kolbeck R, Barde YA, Schwab ME. 1994. Neurotrophin-3