UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Motoneuron survival and axon regeneration following immediate versus delayed implantation of peripheral… Tarazi, Christine 1998

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

Item Metadata


831-ubc_1999-0100.pdf [ 5.19MB ]
JSON: 831-1.0089136.json
JSON-LD: 831-1.0089136-ld.json
RDF/XML (Pretty): 831-1.0089136-rdf.xml
RDF/JSON: 831-1.0089136-rdf.json
Turtle: 831-1.0089136-turtle.txt
N-Triples: 831-1.0089136-rdf-ntriples.txt
Original Record: 831-1.0089136-source.json
Full Text

Full Text

M o t o n e u r o n S u r v i v a l a n d A x o n R e g e n e r a t i o n F o l l o w i n g I m m e d i a t e V e r s u s D e l a y e d I m p l a n t a t i o n o f a P e r i p h e r a l N e r v e G r a f t i n t o the S p i n a l C o r d o f A d u l t R a t s w i t h C 5 a n d C 6 R o o t A v u l s i o n By Christine Tarazi B.Sc, The University of British Columbia, 1993 A Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Master of Science In The Faculty of Graduate Studies (Department of Zoology) We accept thisjhejis-a^aseri^rming to the required standard The University of British Columbia December 1998 ©Christine Tarazi, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. ? Department of I? OO Gj / The University of British Columbia Vancouver, Canada Date DEC. Mill DE-6 (2/88) ABSTRACT Avulsion of spinal nerve roots is a CNS injury that results in motoneuron cell death and paralysis of the affected limb. It has been demonstrated in rats, cats and primates that following spinal nerve root avulsion injured motoneurons can extend axons into ventral roots or peripheral nerve (PN) grafts that have been immediately reimplanted into the spinal cord and result in functional reinnervation of denervated muscles. The effects of delayed PN graft implantation on motoneuron survival and axonal regeneration has not been studied in animals. Following avulsion of the left C5 and C6 ventral and dorsal roots in adult rats, the free end of a 3.5cm segment of common peroneal nerve was implanted into the ventrolateral area of the cord at C5 and C6. There were five groups of animals based on the time interval between root avulsion and PN graft implantation: immediate implantation, 1 week, 1 month, 2 months and 6 months. Two months after PN implantation, a fluorescent axonal tracer, Rhodamine-Dextran amine (RDA) was applied to the free end of the graft. Counts of RDA labeled neurons in the ipsilateral ventral horn were compared among the groups. Using the cell counts in the immediate implantation group (mean of 761 cells = 100%) as a reference for comparison, counts of RDA-labeled neurons was 82% at 1 week, 44% at 1 month and 38% at both 2 and 6 months. Counts were significantly lower in the 1, 2 and 6-month groups. To allow identification of C5 and C6 motoneurons, a separate group of experiments were done in which C5 and C6 motoneurons were labeled with another axonal tracer Fluorogold (FG) prior to avulsion, and PN grafts were implanted either immediately, 10 days or 1 month after root avulsion. Counts of FG, RDA or double-labeled neurons revealed that the % of ventral horn neurons that are G5 or C6 motoneurons are 31% (immediate), 33% (10 day) and 6% (1 month). A delay of 1 month in PN graft implantation resulted in a significant decrease in 1) the number of surviving C5 and C6 motoneurons; 2) the proportion of surviving C5 and C6 motoneurons that regenerated axons into the graft. Table of Contents Abstract ii List of Tables .iv List of Figures...... v Acknowledgements ..vi Introduction (A) Spinal nerve root avulsion injuries: An overview 1 (B) Anatomy and physiology of the lower motoneuron and brachial plexus 8 (C) Neuronal injury: Axotomy induced response to injury Overview 11 Mechanisms of axotomy induced cell death 12 Injury induced gene expression in CNS and PNS neurons i . Cytoskeletal proteins and GAP-43 14 ii . Immediate early genes... ....19 iii. Neurotransmitter associated changes 20 iv. Apoptosis regulators 24 (D) Survival factors for motoneurons 28 Materials and Methods Part I: Insertion of a PN graft into the cord at different time periods ! following C5 and C6 avulsion 33 Part II: Prelabeling of C5 and C6 motoneurons with Fluorogold prior to C5 and C6 avulsion 36 Part III: Intra-thecal application of BDNF following root avulsion and PN graft insertion .40 Results Parti: Influence of time delay between root avulsion and PN graft insertion on axonal growth 42 Part II: Influence of time delay between root avulsion and PN graft insertion on motoneuron survival and axonal growth 47 PartHI: Intra-thecal application of BDNF following root avulsion and PN graft insertion 55 Discussion 58 Bibliography 69 iii List of Tables Table 1. Fluorogold uptake in intact vs. crushed spinal nerves 38 Table 2. Results Part I: Spinal cord cell counts ..44 Table 3. Summary of Part II Results 54 Table 4. Results Part III: Spinal cord cell counts, PBS pumps .56 Table 5. Results Part III: Spinal cord cell counts, BDNF pumps 57 iv List of Figures Fig. la. Anatomy of a normal spinal cord and spinal nerve roots 2 Fig. lb. Dorsal and ventral root avulsion 2 Fig. 2. Methods: Part I 35 Fig. 3. Methods: Part II 37 Fig. 4. Histogram: Part I Results 43 Fig. 5. Light and fluorescence micrographs: Part I 46 Fig. 6. Fluorescence micrographs: Part II 51 Fig. 7. Histograms: Part II Results 53 v Acknowledgements I would like to thank my supervisors Thomas Zwimpfer and John Steeves, as well as Wolfram Tetzlaff, Jie Liu and members of the Steeves and Tetzlaff labs for their guidance, assistance and friendship in the past few years. Introduction A) Spinal nerve root avulsion injuries: An overview Injuries affecting the adult mammalian central nervous system (CNS) have classically been viewed as irreparable, due to the limited growth of axons in the CNS environment. This is in contrast to the regeneration that occurs in the adult peripheral nervous system (PNS). One type of CNS injury occurs when the spinal nerve roots of the brachial plexus (C5, C6, C7, C8 &/or Tl) are avulsed (torn) from the spinal cord at their point of attachment as a result of a severe traction injury to an upper extremity (Fig. la. & lb.). These devastating injuries result in permanent paralysis and loss of sensation to the affected limb. This is due to the inability of motoneurons in the ventral horn of the spinal cord to spontaneously regenerate out of the CNS environment and across the CNS-PNS transition zone. The central processes of the sensory dorsal root ganglion (DRG) neuron, that are in the PNS, are also unable to regenerate into the unfavorable environment of the spinal cord. These injuries primarily occur in two age groups; newborns that have had a difficult delivery, and young adults that are injured in high speed motor vehicle accidents. Common patterns of blunt injury to the brachial plexus include: avulsion of all spinal nerve roots, upper plexus (C5 & C6 root avulsion due to upper trunk traction), and lower plexus (C8 & T l root avulsion due to lower trunk traction). One percent of all trauma victims suffers an associated brachial plexus injury, the majority due to traction on the upper limb (Midha, 1997). Based on imaging or surgical exploration of the brachial plexus, the incidence of avulsion of at least one root following severe traction injuries to the brachial plexus ranges between 56%-85% (Sunderland, 1976; Narakas, 1993). Current treatment strategies include grafting of uninjured but inappropriate nerves (ie. intercostal nerves or the spinal accessory nerve) to the brachial l Fig. la. Anatomy of a normal spinal cord and spinal nerve roots plexus, and muscle or tendon transfers. However, none of these treatments results in the re-innervation of muscles by the appropriate motoneuron axons, and overall functional recovery is limited. Patients develop chronic pain and often undergo limb amputation and the fitting of a prosthesis in order to regain some function in that limb. Al l aspects of these patients' lives are affected, and many require life long financial and psychological assistance. The avulsion of spinal nerve roots in adult animals is an experimental model of motoneuron death. It has recently been used to study the effects of proximal axonal injury on motoneuron cell bodies (Wu, 1993; Wu et al., 1994), and to investigate ways to decrease cell death and enhance axon regeneration (Kishino et al., 1996; Wu et al., 1995; L i et al 1995). Following avulsion injuries, motoneurons within the ventral horn of the spinal cord are separated from the skeletal muscles they innervate, while dorsal root ganglion neurons are detached from their target neurons in the dorsal horn of the spinal cord. Avulsion lesions combine a proximal axotomy with mechanical strain on the motoneuron cell body (Koliatsos et al., 1994). A gradual and extensive death of motoneurons occurs following a series of retrograde cell body changes, including chromatolysis, loss of transmitter phenotype (acetylcholine), the accumulation of phosphorylated neurofilaments in the perikarya (Koliatsos et al., 1994), and new gene expression (eg. nitric oxide synthase (NOS) (Wu et al., 1993). Motoneurons undergo a programmed or apoptotic cell death following avulsion of nerve roots, since it is a delayed response and involves new gene expression. Avulsion of C7 ventral roots in rats results in the death of 75% of motoneurons within the first 6 weeks after injury (Wu, 1993; Wu et al., 1994). Avulsion of lumbar ventral roots results in the death of 45% and 80% of motoneurons at 1 and 2 weeks respectively (Koliatsos et al., 1994). Another group found motoneuron cell death following lumbar ventral root avulsion to be 60% at 4 weeks, and 90% at 12 weeks (Novikov et al., 1997). This is in contrast to the only slight decrease in motoneuron numbers that occurs • 3 following a more distal axotomy of the ventral root (ie. 5 mm from the cord) in adult rats (Wu, 1993; Wu et al , 1994), and 3mm from the cord in adult mice (Li et al., 1995). The different response of motoneurons to an avulsion injury as compared to a distal axotomy is thought to be due to the deprivation of various neurotrophic factors that are produced in the PNS component of the root (Wu et al., 1994). The ventral root consists of a very short CNS segment and a long PNS segment. While distal axotomy leaves a component of the PNS segment attached to the motoneuron cell body, avulsion of the ventral root results in a complete loss of the motoneurons PNS component. It is generally thought that this Schwann cell rich PNS segment provides valuable trophic support for the motoneurons (Wu et al., 1994). Schwann cells have a number of functions including myelin production in the PNS and the production of a number of neurotrophic factors. These include factors that influence the survival of motoneurons (eg. ciliary neurotrophic factor (CNTF), brain derived neurotrophic factor (BDNF), Neurotrophin-3 (NT-3), glial derived neurotrophic factor (GDNF), leukemia inhibitory factor (LIF) and hepatocyte growth factor (HGF) (Davies, 1998). A number of these factors are capable of rescuing motoneurons from axotomy induced cell death (Sendtner at al., 1996). The implantation of a peripheral nerve (PN) graft into the spinal cord immediately following ventral root avulsion results in the survival of 60% of spinal motoneurons, compared to 30% survival in rats that did not receive a PN graft (Wu et al., 1994). Schwann cells also produce a variety of molecules that have been shown to promote axon growth such as extracellular matrix molecules (eg. laminin, fibronectin, collagen I & IV), cell adhesion molecules (eg. LI , N-cadherin, and N-CAM) and other cell surface molecules (eg. p75 NGF receptor) (Curtis et al., 1992; Daniloff et al., 1986; Martini and Schachner, 1988). Furthermore, following axotomy, Schwann cells display an increase in mitotic activity and they enhance their production of some trophic factors (Heumann et al., 1987; Meyer et al., 1992). 4 In 1911, F.Telle- showed that adult neurons can successfully regenerate axons into the permissive environment of a PN transplant (F. Tello, 1911). In the early 1980s, this idea resurfaced again when Aguayo and co-workers tested the capacity of neurons in the brain, spinal cord and retina to regenerate axons into PN grafts. When PN grafts were transplanted into the spinal cord of adult rats, extensive fiber growth into the graft was seen (Richardson et al., 1980). Furthermore, the source of the fibers was mainly spinal neurons, with some axons from descending fiber tracts (with the exception of the corticospinal tract) (Richardson et al., 1984). Regenerated fibers stopped growing abruptly upon re-entry from the PN graft into the spinal cord at the thoracic level. Fibers that left the graft grew only 1-2 mm into the spinal cord (David and Aguayo, 1981). However, Cheng et al.(1996) showed some long growth into the cord by descending fibers, with the addition of fibroblast growth factor (FGF-1) at the CNS-PN graft interface. In the optic system, retinal ganglion cell (RGC) axons are able to grow for long distances through a peripheral graft and establish synapses in the correct layers of the superior colliculus (Carter et al., 1989; Vidal-Sanz et al., 1987). Therefore it seems a majority of CNS neurons, including motor, sensory and local interneurons are able to extend axons into PN grafts following axotomy. It has been demonstrated in rats (Carlstedt et al., 1986; Bertelli and Mira, 1994), cats (Cullheim et al., 1989), and primates (Carlstedt et al., 1993) that following the immediate reimplantation of the free end of an avulsed ventral root or PN graft into the spinal cord, axons from injured motoneurons can regenerate into the avulsed roots and extend through the peripheral nerves to result in anatomical and functional recovery of denervated muscles. Recently, delayed implantation of PN grafts has also been performed in a few humans with some functional recovery (Carlstedt et al., 1995; Carlstedt, 1997). Although PN implantation into the spinal cord has been successfully demonstrated in a small number of humans, issues such as the optimal timing of surgery and strategies to prevent the extensive death of motoneurons still need to be investigated in animal models. To date, there have been no published studies investigating the influence of a time delay between avulsion of nerve roots and re-implantation of a PN graft. This is an important practical question, since a definitive diagnosis of root avulsion is often difficult to make in the first few weeks or months following injury. Current standard radiological imaging such as MRI and CT/myelogram cannot always distinguish an avulsed root from a root that is severely stretched but still in continuity with the cord. Patients with stretch injuries often have complete loss of function initially but can show significant spontaneous recovery over a period of several months. Clinical examination as well as electromyography and nerve conduction studies are also unreliable in distinguishing between these two situations, especially within the first month after injury (Trajaborg, 1994). The current practice in human root avulsion injury is to wait at least 3 to 4 months before considering surgical exploration. Based on the extensive motoneuron cell death that occurs in the first few months after root avulsion in animals, PN implantation will likely need to be done soon after the injury to maximize motoneuron survival and axon regeneration. 6 Objectives: 1. To investigate in adult rats what influence a time delay between cervical root avulsion and PN graft implantation into the spinal cord has on the extent of motoneuron survival and regeneration of motoneuron axons into PN grafts. The results of these studies will influence the timing of PN implantation in human root avulsion injuries. Hypothesis: A delay between spinal root avulsion and the implantation of a PN graft into the avulsion site will result in an increased motoneuron cell death and a decrease in the regeneration of motoneuron axons into the PN graft 2. To determine if delayed intra-thecal application of brain-derived neurotrophic factor (BDNF) following cervical root avulsion injuries, improves motoneuron survival and regeneration of injured motoneuron axons into a PN graft. Hypothesis: Infusion of BDNF into the avulsion site at the time of PN graft implantation will increase both motoneuron survival and regeneration of motoneuron axons. 7 (B) Anatomy and physiology of the lower motoneuron and brachial plexus The nervous system is organized into a central component that consists of the brain and spinal cord, and a peripheral component consisting of ganglia and peripheral nerves that lie outside the spinal cord. The spinal cord receives sensory information from the skin, joints, and muscles of the trunk and limbs through the peripheral nerves that enter the cord as the dorsal roots. Motor information involved in voluntary and reflex movements is transmitted in the axons of spinal motoneurons, whose cell bodies are located in the ventral horn of the spinal cord. Motoneuron axons exit the cord through ventral roots, which join the dorsal roots to form spinal nerves (Fig.la.). Each spinal nerve divides into an anterior and a posterior primary ramus. The anterior primary rami of C5 to TI form the brachial plexus, a group of nerves that innervate the shoulder and upper limb. In addition to spinal motoneurons, there is a population of brainstem motoneurons, whose axons run in cranial nerves 3, 4, 5, 6, 7, 9, 10, 11, and 12. This population of motoneurons provides motor input to skeletal muscles of the head and neck. The experiments outlined in this thesis are based on the avulsion of the 5 th and 6th cervical dorsal and ventral roots. Although the nerve roots that contribute to the brachial plexus innervate a large number of peripheral nerves, C5 & C6 spinal nerves ultimately provide sensory and motor information for a select group of peripheral nerves. These are the musculocutaneous nerve (contributions from C5,6,7), axillary nerve (C5,6), radial nerve (C5,6,7,8, TI), and the median nerve (C5,6,7,8,T1). Our experiments were conducted in the rat because its brachial plexus anatomy is similar to humans (Bertelli et al., 1991, 1992a, 1992b; Bertelli and Mira, 1994), and root avulsion is an established model in the rat. The spinal cord is arranged into a central gray matter that contains nerve cell bodies, and the surrounding white matter, which contains axons of ascending and descending spinal tracts. The gray matter is divided into a dorsal horn, ventral horn and at thoracic levels a lateral horn. 8 The Grey matter has also been divided into ten laminae (I-X) of Rexed (Rexed, 1954). Cell bodies of spinal motoneurons are located in the ventral horn in lamina IX, while cell bodies of ascending projection neurons that relay incoming sensory information to higher brain centers are situated in the dorsal horn (laminae II & III). Interneurons are likely situated in all laminae. The motor cortex exerts control on spinal motoneurons directly via the descending corticospinal tract and indirectly through a number of descending brainstem-spinal pathways, most importantly the cortico-reticulospinal projections (Kandell, Schwartz & Jessell, 1991). The corticospinal tract mediates voluntary movement of all kinds, is the major tract for fine movements of the hand, and plays a minor role in gate and posture. Descending brain-stem spinal pathways are primarily involved in controlling gait and posture. Clinically, the corticospinal tract is the most important pathway in the CNS, and interruption of the tract by disease or injury leads to motor weakness or motor paralysis (Fitzgerald, 1992). There are two types of motoneurons in the spinal cord, referred to as alpha and gamma motoneurons. Alpha motoneurons are the largest neuron group in the spinal cord, with cell body diameters ranging from 50-100 um in humans (Kandel, Schwartz & Jessell, 1991). Gamma motoneurons are smaller in size and found scattered among the alpha motoneurons. In rats, the diameter of over 99% of all spinal motoneurons is less than 60 um, and they are situated in lamina IX (Swett et al,, 1986). Motoneurons in lamina IX are arranged as groups of cells that usually correspond to particular muscles or groups of muscles in the periphery (Swett et al., 1986). They have large dendrite trees that receive around 10 000 excitatory synapses from the descending corticospinal and brainstem-spinal pathways, as well as from propriospinal neurons (Fitzgerald, 1992). In primates, the corticospinal neurons make direct excitatory synapses with alpha motoneurons, and indirect polysynaptic connections with gamma motoneurons (Kandel, Schwartz & Jessell, 1991). In rats, the corticospinal tract terminates in the intermediate gray and ' 9 ventral horn. Corticospinal axons make synapses with motoneurons as they do in primates. Motor control is exerted primarily through inhibitory interneurons (Liang et al., 1991). Alpha motoneurons innervate extrafusal muscle fibers, which make up the bulk of muscle tissue. Alpha motoneurons are in turn separated into two types: tonic and phasic. Tonic alpha motoneurons innervate slow oxidative-glycolytic muscle fibers, which have relatively slow conducting axons. Phasic alpha motoneurons innervate fast oxidative and fast oxidative-glycolytic muscle fibers. During voluntary movements, tonic neurons are usually recruited first, even for fast movements. Gamma motoneurons function to innervate the intrafusal muscle fibers of muscle spindles (stretch receptors). The co-activation of alpha and gamma motoneurons is very important for maintaining the tension of intrafusal fibers of muscle spindles. During voluntary muscle contraction, alpha-gamma co-activation maintains spindle firing (Fitzgerald, 1992). For the purposes of these experiments, no effort was made to differentiate between alpha and gamma motoneurons, since both types are expected to contribute to regeneration in these experiments and in clinical situations. Acetylcholine (ACh) is the neurotransmitter of all motoneurons, including brainstem and spinal motoneurons. The biosynthetic pathway for ACh is a single step reaction catalyzed by the enzyme choline acetyl transferase (ChAT): Acetyl Co A + Choline Acetylcholine + Co A. ChAT immunocytochemistry can be used to aid in the identification of a neuron as a motoneuron. However, levels of this enzyme are markedly reduced in motoneuron cell bodies following axonal injury (Hoffmann et al., 1996). Also found in the ventral horn are inhibitory interneurons, which function to coordinate muscle action around a joint. There are three main types; group la (coordinate opposing muscles), Renshaw cells (form negative feedback loops to motoneurons), and group lb (receive 10 convergent input from several types of receptors). The neurotransmitter of interneurons is either y-aminobutyric acid (GABA), or the amino acid glycine. (C) Neuronal injury: Axotomy induced response to injury (a) Overview Spinal motoneurons represent a unique system for studying regeneration since they can undergo either a CNS or PNS injury. Avulsion of ventral roots from the spinal cord is classified as a CNS injury (Carlstedt, 1991; Bertelli and Mira, 1994), since the axotomy occurs within the CNS. Extensive motoneuron cell death occurs following ventral root avulsion and axons are unable to spontaneously regenerate across the CNS-PNS interface. However, transection of the spinal nerves just millimeters away from the cord is a PNS injury, and regeneration of motoneuron axons occurs with little motoneuron cell death. Axotomy (either by sectioning a tract within the CNS or by cutting a peripheral nerve), divides a neuron into two sections; a proximal segment that is still connected to the cell body, and a distal segment which is isolated from the rest of the cell. A number of events occur in the vicinity of the injury following axotomy. In the distal segment, Wallerian degeneration takes place: myelin forming cells withdraw their processes and phagocytic cells (microglia &/or macrophages) infiltrate the lesion site to remove myelin debris and the degenerated distal axon. The cell body exhibits central chromatolysis. Debris from Wallerian degeneration is phagocytosed by macrophages that secrete proteases and engulf myelin. This occurs within a period of days to weeks in the PNS, but takes several months in the CNS (Ludwin, 1990; Stoll et al., 1989), and is thought to be a contributing factor in the inability of the CNS to regenerate axons. Conversely, there is also evidence that the absence of rapid n Wallerian degeneration does not affect the initial cell body response to injury (Bisby et al., 1995). The role of glial cells following injury is also different in the CNS and PNS. In the PNS, macrophages are thought to secrete factors which induce Schwann cell proliferation and Schwann cells can in turn secrete a variety of trophic factors. In the CNS, microglia are considered as the resident macrophages when the blood brain barrier is not compromised, and along with astrocytes, are responsible for the removal of debris. Macrophages also aid in the removal of debris in situations when the blood brain barrier is compromised. Microglial activation usually precedes any other cell type reactions in the CNS, since they respond not only to alterations to neuronal structure, but also to small changes in the microenvironment (Gehrmann et al., 1995; Kreutzberg, 1996). Interestingly, in a facial nerve axotomy model where motoneurons survive and regenerate, microglia become activated but not phagocytic (Blinzinger and Kreutzberg, 1968; Kreutzberg, 1996). However, if there is extensive facial motoneuron cell death, microglia become reactive and phagocytose the neuronal debris (Streit and Kreutzberg, 1988). (b) Mechanisms of axotomy-induced cell death The mechanisms that result in the neuronal response to injury are extremely complex and much work on this issue is still required. Direct injury to neuronal cell bodies results in an immediate and irreversible necrotic cell death that is characterized by the swelling and lysing of cells (Schwab and Bartholdi, 1996; Deshmukh and Johnson, 1997). Necrosis is likely the cause of cell death for some motoneurons following avulsion injuries, since axotomy occurs very close to the cell body and likely produces some direct mechanical injury to the cell body (Koliatsos et al., 1994). However after axotomy many neurons undergo a programmed (apoptotic) cell death 12 that is delayed and involves new gene and protein expression (Tetzlaff et al., 1990; Bisby and Tetzlaff, 1992; Wu, 1993). The features of programmed cell death include shrinking of the cytoplasm, plasma membrane blebbing, nuclear chromatin condensation and the fragmentation of genomic DNA (formation of apoptotic bodies) (Kerr, 1972; Deshmukh, 1997). Phagocytes without causing inflammation (Kerr et al., 1972) clear cellular debris. Furthermore, depending on the nature/severity of the neuronal injury, there may be processes other than the initial mechanical injury that can contribute to both tissue damage in the vicinity of the injury as well as to neuronal cell death. These include ischemia, edema, free radical production, excitotoxicity, proteases released from neighboring necrotic cells and the inflammatory response (Schwab and Bartholdi, 1996). Motoneuron cell death occurs naturally during embryonic development. During development and the early post-natal period, extensive motoneuron cell death occurs following axotomy (proximal or distal) or the removal of their targets (Lowrie and Vrbova, 1992; Snider et al., 1992; Li et al., 1994). In the rat, motoneuron cell death occurs naturally during the period embryonic day 15 (E15) to post natal day 1 (PI) (Stockli et al., 1991; Sendtner et al., 1996). In adults, apoptotic motoneuron cell death occurs only following a proximal axotomy (eg. avulsion of nerve roots). Motoneuron cell death is characterized by DNA fragmentation and the formation of apoptotic bodies (De Bilbao et al., 1996; Rossiter et al., 1996) and involves new gene expression, including c-Jun (Herdegen et al., 1997), nitric oxide synthase (NOS) (Wu et al., 1994; Wu, 1996), and B A X (Deckwerth et al., 1996). Natural and axotomy induced motoneuron cell death in developing and neonatal animals may be due to the deprivation of target derived trophic factors (Oppenheim et al., 1992; Yan et al., 1992, 1993; Koliatsos et al., 1993; Hughes et al., 1993). Motoneuron cell death due to proximal axotomy in adults can be decreased by the application of trophic factors including GDNF (Li et al., 1995) and BDNF (Novikov et al., 1995 13 & 1997), implantation of PN grafts into the spinal cord (Wu et al., 1994) and the inhibition of NOS production (Wu et al., 1994; L i , 1993). There are differences in the extent of motoneuron cell death following axotomy, depending on the age of the animal, the distance of the injury from the cell body, and the type of neuron. Axotomy in the neonatal period results in a rapid and extensive cell death in most neuronal types that have been studied, including spinal motoneurons (Sendtner et al., 1996), retinal ganglion cells (Allcutt et al., 1984) and corticospinal neurons (Kalil et al., 1990). In contrast, adult neurons usually survive axotomy, unless the lesion is close to the cell body, as in nerve root avulsions of spinal motoneurons (Wu et al., 1994), intraorbital axotomy of retinal ganglion cells (Villegas-Perez et al., 1993) and subcortical injury of corticospinal neurons (Giehl and Tetzlaff, 1996). In motoneurons (Wu, 1996) and retinal ganglion cells (Berklaar et al., 1994; Garcia-Valenzuela et al., 1994; Isenmann et al., 1997), a proximal axotomy results in a significant increase in apoptotic cell death, compared to a distal axotomy. Furthermore the cell death is delayed, and retinal ganglion cells displayed apoptotic bodies, indicative of programmed cell death (Berkelaar et. al., 1994). (c) Injury induced gene expression in CNS and PNS neurons i. Cytoskeletal proteins and GAP-43 The modifications that take place in the cell body of both CNS and PNS neurons following neuronal injury, are commonly referred to as the cell body reaction (CBR). The CBR is a series of modifications of the cells metabolic functions, as well as its structural features, and involves both increases and decreases of a number of specific genes and proteins (Grafstein, 1983; Grafstein and McQuarrie, 1978; Lieberman, 1971; Tetzlaff and Bisby, 1990). Since some axonal regrowth or sprouting can occur prior to the CBR (Duce and Keen, 1980; Tetzlaff and 14 Bisby, 1990), the CBR appears to be important for the maintenance of regeneration (Carlsen et al., 1982; Tetzlaff and Bisby, 1990). If there is no CBR following axonal injury, regeneration is initiated, but ultimately stalls and is aborted (Carlsen, 1983; Tetzlaff and Bisby, 1990). The cytoskeletal proteins tubulin, actin, and a growth associated protein (GAP-43) are collectively referred to as regeneration associated genes (RAGs), since elevations in mRNA and protein generally correlate with regeneration of peripheral axons (Oblinger and Lasek, 1988; Tetzlaff et al., 1988) and some CNS axons (Kobayashi et al., 1997). The neuronal cytoskeleton is composed of three main fibrillar components of varying thickness: microtubules, neurofilaments, and microfilaments, together with their associated proteins. Microtubules are the thickest of the neuronal cytoskeletal fibers (25-28 nm diameter), and extend along the newly regenerated axons and into the base of the growth cone (Landis, 1983; Tetzlaff and Bisby 1990). They are constructed of 13 linearly arranged a and p tubulin dimers (protofilaments). Microtubules are arranged linearly in the axon and although they do not span the entire length of the axon, they can be as long as .5 mm. The stability and polymerization/assembly of microtubules is regulated by microtubule associated proteins (MAPs). The MAPs include MAP-1, MAP-2, and tau, which function to promote assembly. Protein phosphorylation of MAPs results in a decrease in polymerization. Neurofilaments are typically around 10 nm in diameter and are the most abundant (3-lOx more than microtubules) of the fibrillar components in the cytoskeleton. There are three different forms of neurofilaments: light (NF-L), medium (NF-M) and heavy (NF-H). Neurofilaments are not involved in axonal growth as their mRNA is down regulated following axotomy (Tetzlaff and Bisby, 1989; Tetzlaff and Bisby, 1990). Like microtubules, they are arranged along the length of the axon and in dendrites. Neurofilaments belong to a family of 15 proteins called cytokeratins, and include vimentin, glial fibrillary acid protein (GFAP), desmin, and keratin. Microfilaments are 3-5 nm in diameter, and are composed of polymers of globular actin monomers. Neural actin is a mixture of p and y forms and is a prominent growth cone constituent (the a form is found in skeletal muscle). Most of the neural actin is associated with the plasma membrane by several proteins, including spectrin, ankyrin, vinculin, and talin. Microfilaments are also associated with the integrins, a family of membrane spanning proteins. This allows them to interact with proteins in the extracellular matrix, such as laminin and fibronectin. The role of extracellular matrix molecules, cell adhesion molecules, and guidance molecules in the regenerative capacity of adult neurons is still unclear. However there are a large number of molecules belonging to several families of cell adhesion molecules and extracellular matrix molecules, which have been shown to enhance neurite outgrowth in-vitro (Dodd and Jessell, 1988; Grumet, 1991; Hynes and Lander, 1992; Schwab and Bartholdi, 1996). Laminin, one such extracellular matrix molecule, causes a sustained increase of growth cone velocity and provides directional cues, through the activation of a protein kinase C dependent intracellular signaling mechanism (Kuhn et al., 1995). In PNS neurons, (ie. motoneurons and DRG cells), axotomy results in an increase in actin and tubulin, but a decrease in neurofilament synthesis (Goldstein et al., 1988; Greenberg and Lasek, 1988; Hoffman and Cleveland, 1988; Hoffman et al., 1983; Oblinger and Lasek, 1988; Pearson et al., 1988; Tetzlaff et al., 1988; Wong and Oblinger, 1987; Tetzlaff and Bisby, 1990). The increased rate of tubulin synthesis and decreased rate of neurofilament synthesis in regenerating neurons is also found in development (Hoffman and Cleveland, 1988; Tetzlaff and Bisby 1990). Moreover, the increase in synthesis of both a and P-tubulin are the result of a re-expression in the adult, of developmentally regulated isotypes (Hoffman and Cleveland, 1988; 16 Hoffman and Lasek, 1980). Following injury to the facial nerve, facial motoneurons exhibit an increase in mRNA for actin and tubulin (Tetzlaff et al., 1991), an increased synthesis of actin and tubulin proteins, and a decrease in neurofilament synthesis (Tetzlaff et al., 1988). In spinal motoneurons, the increase in Tal-tubulin was more pronounced after a proximal (5 mm distal to the ganglion) than distal axotomy, as was the case with rubrospinal neurons (Tetzlaff, personal communication). There are no published studies investigating cytoskeletal protein synthesis following root avulsion. In the adult mammalian CNS there is little spontaneous regeneration following axotomy. The rubrospinal system has been used extensively to study the CBR in the CNS. One study compared the CBR of facial motoneurons and rubrospinal neurons following axotomy (Tetzlaff et al., 1990). Surprisingly, the initial CBR to axotomy of non-regenerating rubrospinal neurons was identical to that of facial motoneurons that are capable of re-growing axons. There were increases in mRNA levels for Tal-tubulin, total a- and p-tubulin isotypes and actin. Levels of NF-M and NF-L were found to decrease (Tetzlaff et al., 1990). However, 8-14 days after axotomy, rubrospinal neurons exhibited a down regulation in tubulin and actin mRNA levels to pre-axotomy levels, while the facial motoneurons continue to express increased levels (Tetzlaff et al., 1991). The levels of Tal-tubulin and GAP-43 mRNA remained elevated during this time. The down regulation of tubulin and actin is coincident with the appearance of atrophy in the rubrospinal neurons (Barron, 1983). In addition, rubrospinal neurons displayed a more pronounced decline in NF-M expression after a cervical compared to a thoracic axotomy (Tetzlaff personal communication). This suggests that rubrospinal neurons initially respond to cervical axotomy by mounting a CBR that is similar to that which occurs in peripheral neurons, followed by a down regulation of some cytoskeletal elements (Tetzlaff et al., 1991). In fact i 17 Baron and co-workers (1989) showed axonal sprouting by lesioned rubrospinal neurons during this initial period. The growth-associated protein GAP-43 is a major phosphoprotein in neuronal growth cones (Schwab and Bartholdi, 1996). It is membrane associated and is rapidly transported in neurons, in contrast to the cytoskeletal proteins (Skene, 1989). Although its function is still poorly understood, GAP-43 has been widely used as a marker for the CBR in neurons following axotomy because its increase following axotomy is higher than any of the cytoskeletal proteins (10-50x increase in mRNA levels) (Bisby and Tetzlaff, 1992). High levels of GAP-43 are found in regions of the brain known for their plasticity, such as the hippocampus in rodents and the association cortex in humans (Benowitz et al., 1988, 1989). This suggests that GAP-43 is involved with the constant remodeling that occurs in learning and memory (Benowitz and Perrone-Bizzozero, 1991; Strittmatter et al., 1992; Schwab and Bartholdi, 1996). In a GAP-43 gene knockout mouse, growth cones seem to have impaired pathfinding ability, but axons do still grow (Strittmatter et al., 1995). GAP-43 expression is high during development in both the CNS and PNS (Skene, 1989; Schwab and Bartholdi, 1996). Levels of GAP-43 mRNA were compared in axotomized rubrospinal and facial neurons (Tetzlaff et al , 1991). Both rubrospinal neurons and facial motoneurons responded with an increase in GAP-43 mRNA levels. At two weeks the levels of GAP-43 remained elevated, while levels of cytoskeletal proteins (except Tal-tubulin) decreased (Tetzlaff et al., 1991). Re-expression of high GAP-43 levels was also observed in DRG neurons, spinal motoneurons, and retinal ganglion cells, in response to axotomy (Chong et al., 1992; Doster et al., 1991; Piehl et al., 1993; Schreyer and Skene, 1991; Schwab and Bartholdi, 1996). To date there are no published studies looking at GAP-43 expression in spinal motoneurons following avulsion injuries. 18 The distance of the lesion from the cell body seems to play a role in the induction of RAGs. In rubrospinal neurons, there was a significant increase in the mRNA levels of GAP-43, Ted -tubulin and total tubulin after cervical but not thoracic axotomy, and this corresponds with the ability of rubrospinal neurons to regenerate into a PN graft inserted into the spinal cord (Tetzlaff personal communication). In contrast, spinal motoneurons expressed an increase in GAP-43 mRNA levels in both proximal and distant axotomy models (Tetzlaff personal communication). Furthermore, in retinal ganglion cells GAP-43 expression only increased when the axotomy was within 3mm of the eye (Doster et al., 1991). This increase in GAP-43 occurred regardless of whether there was a PNG to regenerate axons into, indicating that GAP-43 expression is an injury response and does not predict axon regeneration. Regardless of its precise function, GAP-43 expression seems to correlate with axonal regeneration and the remodeling of connections. Furthermore, GAP-43 depleted neurons were more sensitive to myelin associated inhibitors of neurite outgrowth (Aigner and Caroni, 1995). ii . Immediate early genes (IEGs) Like GAP-43, the association of IEG expression and axonal regeneration and/or neuronal survival is unclear. IEGs encode nuclear proteins that bind to DNA and alter transcription (Schwab and Bartholdi, 1996). Expression of IEGs c-Jun and c-Fos is induced within hours following axotomy and neuronal ischemia (Chopp, 1993; De Felipe et al., 1993; Schwab, 1996). There exists contradictory evidence on the roles of IEGs. For example c-Jun is expressed in spinal cord neurons and spinal ganglia following axotomy (Herdegen et al., 1991; Jenkins and Hunt, 1991; Leah et al., 1991; Wu, 1996) and is thought to be a prerequisite for successful axonal regeneration (Herdegen et al., 1993; Leah et al., 1991; Wu, 1996). There is also evidence that c-19 Jun may be related to neuronal death in the brain following hypoxic-ischemic injury (Dragunow et al., 1993) and following axotomy (Dragunow, 1992). Recently, a study compared the induction of c-Jun and c-Fos in adult spinal motoneurons in response to three models of axonal injury: distal axotomy, root avulsion, and root avulsion followed by a PN graft implantation (Wu, 1996). There was no evidence of c-Fos induction in motoneurons in any of the three models. Expression of c-Jun was primarily increased in spinal motoneurons that underwent a distal axotomy, in which most motoneurons survive. When root avulsion was followed by immediate PN graft insertion, motoneurons that regrew axons into the graft expressed both c-Jun and the low affinity nerve growth factor receptor (LNGFR). Rubrospinal neurons showed an increase in the expression of c-Jun that lasts for 2-3 weeks following axotomy at the cervical level, but no increase was seen with a thoracic lesion (Jenkins et al., 1993). C-Fos was also up regulated following a cervical lesion but to a lesser extent (Jenkins et al., 1993). It has been shown that rubrospinal neurons will regenerate axons into a PN graft following cervical but not thoracic axotomy (Richardson et al., 1984; Kobayashi et al., 1997), providing evidence for the role of c-Jun in regeneration. In summary, c-Jun may be involved in M N axon regeneration, and a PN graft is capable of changing certain aspects of the CBR to favor the regenerative response. iii . Neurotransmitter associated changes Acetylcholine Acetylcholine is the neurotransmitter of motoneurons. Both the synthesizing enzyme choline acetyltransferase (ChAT) and the degrading enzyme acetylcholinesterase (AChE) are used as markers for determining levels of ACh following axonal injury. Levels of transmitter immunoreactivity are commonly evaluated in motoneuron cell bodies before and after axonal injury. Mature motoneurons normally survive axotomy, but display a temporary loss of ChAT mRNA and protein. The reappearance of cholinergic markers in adult motoneurons is coincident with the re-innervation of target muscles following a nerve crush injury (Koliatsos et al., 1991), and is consistent with the hypothesis that target derived factors influence such responses. Koliatsos and co-workers (1994) examined the series of retrograde changes that take place following avulsion of spinal nerve roots. They found that 3 days following avulsion injury, 30% of motoneurons were ChAT negative, and the remaining neurons had reduced immunoreactivity. At day 21, when more than 80% of motoneurons had degenerated, ChAT immunoreactivity was present in only a small proportion of remaining motoneurons. Kishino and co-workers (1997) demonstrated that a continuous 14-day intra-thecal infusion of BDNF following avulsion of lumbar roots was capable of significantly affecting survival and ChAT immunoreactivity. Relative to the non-lesioned side, only 29% of surviving motoneurons were ChAT positive in PBS treated animals, compared to 80% positive with BDNF infusion. In addition, ChAT levels were restored even if BDNF treatment was delayed by up to 2 weeks following avulsion (Kishino et al., 1997). Following axotomy of motoneuron, ChAT levels greatly decreased, then returned to normal by 6-8 weeks after the injury (Chiu et al., 1995). BDNF and NT-4/5 was able to protect 70% of adult motoneurons from the loss of ChAT in a dose dependent manner, while NGF, NT-3, and IGF-I protected a smaller proportion of motoneurons (Chiu et al.,1995). Following sciatic nerve or facial nerve transection, BDNF treated gel foam applied to the proximal nerve stump (single application, 10-20ug) was unable to restore ChAT immunoreactivity in either group, but was able to reduce motoneuron cell death in both neonatal and adult rats (Clatterbuck et al., 1994). However, the continuous administration of BDNF following peripheral nerve injury in the adult resulted in the up-regulation of ChAT immunoreactivity (Friedman et al., 1995; Yan et al., 21 1994). From the above studies, it seems that the dose and method of BDNF application can affect the loss of transmitter phenotype that normally occurs in motoneurons following axotomy. In summary, the cell body reaction usually consists of two phases: initial increases or decreases in the levels of certain proteins followed by the recovery of levels to normal. The classical view is that new axonal growth requires increases in the synthesis of those proteins that are involved in axonal structure, such as cytoskeletal arid membrane associated proteins and a decrease in the synthesis of proteins involved in neurotransmission and axonal stability (Grafstein and McQuarrie, 1978; Bisby et al., 1995). Target-derived trophic factors may play a role in signaling the beginning and recovery of the CBR (Bisby et al., 1995). For example when an axon is severed, the cell body no longer receives trophic support from the distal segment. However the continuous application of BDNF was able to up-regulate ChAT immunoreactivity (Friedman et al., 1995) in motoneurons. This suggests that the decrease in ChAT expression was due to a negative signal (absence of trophic support), while re-expression was due to a positive signal (trophic support). Furthermore, GAP-43 expression decreases when neurons form synaptic connections with their appropriate targets (Schreyer and Skene, 1991). Nitric Oxide Nitric oxide (NO) is a poisonous and unstable gas. It is a component of car fumes and is thought to be involved in the depletion of the ozone layer (Holscher, 97). Although the jury is still out as to the true functions of NO in the CNS it has been implicated in many functions. These: include anterograde presynaptic neurotransmitter in the cerebellum acting on Purkinje cells (Briining, 1996), retrograde neuronal messenger (Vincent et al., 1992; DeVente et al., 1993; Holscher, 1997), synaptic plasticity (Bohme et al , 1991; Ito, 1989; Shibuki and Okada, 1991), and neuronal degeneration (Dawson et al., 1991a). NO also influences neurotransmitter release in 22 the CNS through a mechanism which remains unclear, but is thought to involve activation of cGMP-dependent protein kinase to enhance phosphorylation of synaptic vesicle proteins, or through activation of cGMP dependent cation channels (Ahmad et al., 1994). Nitric oxide is synthesized by the enzyme nitric oxide synthase (NOS), which is identical to the enzyme nicotinamide adenine dinucleotide phosphate diaphorase (NADPH-d) (Dawson et al , 1991b; Hope et al., 1991; Wu, 1993). Because NOS is exclusive to neurons and co-localizes with NADPH-d containing neurons, NADPH-d histochemistry has been used to identify NO producing neurons (Blottner and Baumgarten, 1992; Wu, 1993; Wu et al., 1994). Its link with neuronal degeneration has led a number of investigators to examine NO and NOS activity in various axonal injury models including spinal root avulsion. Trophic factor deprivation results in an apoptotic death of cultured motoneurons (Milligan et al., 1995; Pennica et al., 1996). When primary cultures of rat embryonic motoneurons were cultured without BDNF, NOS was induced within 18 hours, and more than 60% of neurons underwent apoptosis (Alvaro etal., 1998). Wu (1993) demonstrated that rat spinal motoneurons express NOS within a week of injury after ventral root avulsion but not following distal transection of the ventral root (ie. 5-10 mm from the spinal cord), and motoneurons that express NOS ultimately die. Implantation of a PN graft following ventral root avulsion significantly decreases motoneuron cell death compared to controls (Wu et al., 1994). Furthermore, the expression of NOS is completely inhibited 3 weeks following avulsion in all motoneurons that regenerated axons into a PN graft, while those that did not regenerate expressed NOS. They hypothesized that neurotrophic factors produced in the PN graft prevented the induction of NOS (Wu et al., 1994). Wu and co-workers (1995), looked at the effects of neurotrophic factors on motoneuron survival and NOS induction. While in control animals 80% of motoneurons were NOS positive 23 at 3 weeks, in GDNF or BDNF treated animals no NOS positive neurons were found at 3 weeks. A recent study examined the effects of extended treatment of BDNF on motoneuron survival and NOS expression following avulsion injury. Following 4 weeks of BDNF or vehicle administration, survival and NOS expression were evaluated at 12 weeks postoperatively. In vehicle treated rats, only 10% of motoneurons survived, of which 20-40% were NOS positive, whereas BDNF treatment resulted in 45% survival and the complete absence of NOS expression (Novikov et al., 1997). It remains unclear whether NO is a direct or indirect cause of cell death, or simply reflects a motoneuron's response to a potentially lethal injury. NOS expressing motoneurons ultimately die in the study by Wu (1993) in which all NOS positive neurons degenerate between 8 and 9 weeks after the root avulsion. However, Novikov and co-workers (1997) showed that 10% of motoneurons survived the avulsion injury in vehicle treated rats, and of those neurons 20-40% were still NOS positive. NO was not absolutely coincident with cell death in this study. Furthermore, avulsion of C7 roots in mice did not induce NOS expression in motoneurons at 2 or 4 weeks after injury, a time when 80% of motoneurons had already died (Li et al., 1995). iv. Apoptosis Regulators Extensive programmed cell death is known to occur in the developing nervous system (Oppenheim, 1991), and in spinal motoneurons following avulsion of spinal nerve roots (Wu, 1993; Wu et al., 1994; Li et al., 1995; Wu, 1996; Kishino et al., 1997). During development 20-80% of all neurons die, many during the phase of target innervation (Oppenheim, 1991; Deshmukh and Johnson, 1997). Competition for trophic factors during this time eliminates neurons and effectively matches the target cell population with the population of innervating neurons (Deshmukh and Johnson, 1997). However, deprivation of trophic factor(s) is just one 24 event that can trigger apoptosis. Programmed cell death is evolutionarily conserved, however variations exist in the pathway depending on the cell type and the apoptosis-inducing stimulus (Deshmukhand Johnson, 1997). There are a number of protein families that act as regulators of apoptotic cell death. These include the Bcl-2 family, Caspase family (formerly known as ICE proteases) and the inhibitor of apoptosis protein family (IAPs). The pathway that ultimately results in cell apoptosis is extremely complex, and the precise interactions between the various regulators of cell death are still largely unknown. The Bcl-2 family of genes consists of apoptosis-promoting molecules (eg. Bax) and apoptosis suppression molecules (eg. Bcl-2) (J.C. Reed, 1994). Members in the Bcl-2 family contain at least one of four conserved homology domains (BH1 to BH4) (Adams and Cory, 1998). Those Bcl-2 family members most similar to Bcl-2 have all four domains (Adams and Cory, 1998). Although heterodimerization is not required for pro-survival activity (Cheng et al., 1997), heterodimerization in the BH3 domain is essential for pro-apoptotic activity (Yin et al., 1994; Chittenden et al., 1995). Pro-survival members are thought to act by a number of mechanisms, including inhibiting caspase activation and maintaining organelle structure (Adams and Cory, 1998). The Bcl-2 family is regulated by cytokines and a number of regulatory signals at different levels (Adams and Cory, 1998). Over expression of Bcl-2 via microinjection of a bcl-2 expression vector can rescue NGF-, BDNF-, or NT-3-dependent neurons, but not CNTF-dependent neurons from trophic factor withdrawl (Allsopp et al., 1993; Deshmukh and Johnson, 1997). Bcl-2 overexpression does not however rescue cells from programmed cell death in a number of circumstances, indicating the existence of Bcl-2 independent mechanisms of cell death (Reed 1994). Much work is still required to determine how the Bcl-2 family regulates apoptosis. 25 The Caspases are a family of cysteine proteases that seem to play a critical role in apoptosis by cleaving key proteins involved in RNA splicing, DNA repair and cell structural elements, or activating proinflammatory (Salvesen and Dixit, 1997; Thornberry and Lazebnik, 1998). There are currently thirteen members in the caspase family (caspase 1 to caspase 13). Each caspase has a distinct role in apoptosis and inflammation and include "initiators of apoptosis" (eg. caspases 8 & 9) and "effector caspases" (eg. caspases 3,6 and 7) (Salvesen and Dixit, 1997; Thornberry and Lazebnik, 1998). Caspases exist as proenzymes and are proteolytically activated by unknown mechanisms in response to various pro-apoptotic conditions (Nicholson and Thornberry, 1997). It appears that effector caspase activity results from a cascade of events that involves a pro-apoptotic signal activating ah initiator caspase which in turn activates an effector caspase (Thornberry and Lazebnik, 1998). A variety of apoptosis stimuli and upstream signaling pathways converge onto the caspases, and evidence indicates that the activation of initiator caspases requires binding to specific co-factors (Boldin et al., 1996; Thornberry and Lazebnik, 1998). Caspases also inactivate other apoptosis suppression molecules such as Bcl-2 (Xue and Horvitz, 1997). The overall effect of caspase activity is to: (1) halt cell cycle progression; (2) disable cell homeostasis and repair mechanisms; (3) detach the cell from surrounding tissue; (4) disassemble structural elements; (5) mark the apoptotic cell for engulfment by macrophages (Nicholson and Thornberry, 1997). Caspase activation (caspase-3 and 7) also seems to be under the control of members of the LAP family. IAPs are a family of anti-apoptotic proteins including c-IAP-1, c-IAP-2 and neuronal apoptosis inhibitory protein (NAIP), the first human IAP identified (Roy et al., 1995; Roy et al., 1997). Evidence suggests that some members of the IAP family (IAP-1, IAP-2, IAP-3 and neuronal apoptosis inhibitor protein (NAIP) can block activation of caspases-3 and 7, resulting in halting of the cell death pathway and the prevention of cell death (Roy et al., 1997; Deveraux et 26 al., 1997). The mechanism by which IAPs inhibit the "effector Caspases" has yet to be determined, but IAPs are thought to function through a conventional lock and key scenario (Roy etal., 1997). NAIP and survival motoneuron gene (SMN) are two genes located on chromosome 5, that are frequently deleted in patients with spinal muscular atrophy, a group of motoneuron diseases that are characterized by degeneration of anterior horn cells (Roy et al., 1995). Recent animal studies have shown a 6-8 fold increase in NAIP mRNA in the rat facial nucleus 1 day after facial nerve axotomy in adult rats (where facial motoneurons survive axotomy), but not in neonates (where the majority of facial motoneurons die following axotomy) (McPhail et al., 1998). In a study that compared the expression of cell death (ie.caspase-1, caspase-3, bax arid bcl-xs) and survival promoting genes (ie. bcl-2, bcl-xl and bag) following axotomy in both adult and neonatal rats, there was a 2-4-fold increase in caspase-1 & 3 mRNA and protein levels in the facial nucleus of both adults and neonates. There was no change in levels of bcl-2 or bag in either group, while bcl-xl increased in both groups following axotomy. Bcl-xs expression is higher in neonates, suggesting a role in neonatal motoneuron cell death following axotomy (Vanderluit et al., 1998). In summary, the cascade of events that ultimately result in apoptosis is extremely complex and involves the interaction of a number of gene families. It is hypothesized that after injury, cell signaling converges on caspase-3, which in turn acts upstream of factors involved in maintaining cell homeostasis and structural integrity. This ultimately results in the morphological changes that characterize apoptosis. 27 (D) Survival factors for motoneurons A number of molecules have been identified that regulate motoneuron survival and function. The neurotrophins are a family of molecules including; nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and NT-4/5. This family of small (13kDa) highly basic proteins exists naturally as homodimers. Of these factors BDNF, NT-3 and NT-4/5 are capable of preventing embryonic and postnatal motoneuron cell death in a variety of situations (Sendtner et al., 1996). The neurotrophins act through a family of tyrosine kinase receptors, the Trks. The Trks are high affinity neurotrophin receptors and include: TrkA, TrkB, and TrkC. Upon binding the appropriate neurotrophin, signal transduction of the receptors is induced by dimerization and autophosphorylation of a given Trk. TrkA is highly specific for NGF, Trk B will preferentially bind BDNF and NT-4 (NT-3 to a lesser extent) and NT-3 is the main ligand of TrkC (Lindsay, 1994a &b). Whereas TrkA is only localized to very few neuronal types both in the CNS and PNS, full length kinase containing forms of TrkB and TrkC are expressed on the majority of CNS neurons , including spinal motoneurons (Lindsay, 1994a & b). Full lengthTrkB and Trk C are not however expressed on non-neuronal cells such as astrocytes, ependymal, endothelial, or choroid plexus cells. Along with the neurotrophins, there is an extensive list of growth factors and neurotrophic agents that are capable of promoting motoneuron survival in vitro or in vivo. These include: (1) the neuropoietic cytokines; (ciliary neurotrophic factor (CNTF), leukemia inhibitory factor (LIF), interleukin-6 (IL-6), cardiotrophin-1 (CT-1), (2) transforming growth factor-B (TGF-P) superfamily of cytokines; (TGF-P, glial derived nerve growth factor (GDNF), (3) fibroblast growth factors; (FGF-1, FGF-2, FGF-5), and (4) insulin like growth factors (IGFs); (insulin, IGF-l,IGF-2). 28 BDNF mRNA and protein is expressed in the peripheral targets of spinal motoneurons during development and in the adult (Maisonpierre et al., 1990; Griesbeck et al., 1995; Sendtner et al., 1996) and during development in medium to large sized dorsal root ganglion (DRG) cells in rodents (Ernfors and Persson, 1991). Motoneurons can therefore receive trophic support both from their peripheral targets (muscles) and afferent input (eg. DRG cells). If motoneuron afferents are lesioned in development when motoneuron cell death occurs naturally, it results in an increased loss of motoneurons, an indication of trophic support from afferent input (Okado and Oppenheim, 1984; Davies, 1996). There is evidence that some neurons can obtain trophic support from cells in peripheral nerves. An example is CNTF, which is synthesized postnatally by Schwann cells (Dobrea et al , 1992) and is required for the maintenance of some mature motoneurons (Masu et al., 1993). Furthermore BDNF, NGF a n d NT-4 are upregulated in S c h w a n n cells of the distal stump after a peripheral lesion ( H e u m a n n et al., 1987; Meyer et al., 1992; Funakoshi et al., 1993). Removal of motoneuron targets during embryonic and early postnatal development results in massive cell death, whereas virtually no cell death occurs following axotomy of adult peripheral nerves (Sendtner et al., 1996; Li et al., 95), unless the injury is close to the cell body, as with root avulsion injuries. BDNF has been shown to prevent naturally occurring cell death of lumbar spinal motoneurons in chick embryos (Oppenheim et al., 1992). Although motoneuron survival is not dependent on target derived factors in adult animals (Crews and Wigston, 1990; Snider et al., 1992), adult motoneurons continue to express various neurotrophin receptors (eg. trkB and trkC)(Lindsay, 1994a). Furthermore, adult motoneurons remain responsive to BDNF and NT-4/5 (Yan et al., 1994; Friedman et al., 1995; Li et al., 95). In motoneurons, the receptor-mediated retrograde transport of BDNF, CNTF, and NT-3 dramatically increases following nerve injury (Lindsay, 1994a). There is also a coincident increase in the availability of BDNF and CNTF in the damaged nerve. BDNF synthesis that is normally low in intact nerves dramatically increases in the nerve distal to the injury site. This is believed to be from denervated Schwann cells which become more mitotically and metabolically active (Lindsay, 1994b). In neonatal animals, the local application of BDNF following axotomy of the facial and sciatic nerves resulted in a reduction of the atrophy and degeneration of motoneurons which normally follows (Hughes et al., 1993; Koliatsos et al., 1993;Yan et al., 1992; Kishino et al., 1997). Furthermore, BDNF was shown to attenuate the decrease of ChAT, which normally occurs following axotomy of the facial and sciatic nerves in adult rats (Chiu et al., 1994; Friedman et al., 1995; Kishino et al., 1997). BDNF has also been shown to affect the expression of GAP-43 and cytoskeletal proteins in adult CNS neurons. Kobayashi and co-workers (1997) demonstrated that the infusion of BDNF or NT-4/5 into the vicinity of axotomized rubrospinal neurons between days 7-14 after axotomy resulted in increased expression of GAP-43 and Tal-tubulin mRNA. Increased levels remained even 2 weeks following the termination of BDNF infusion. Furthermore, BDNF and NT-4/5 prevented the atrophy that normally occurs in rubrospinal neurons following cervical axotomy (Kobayashi et al., 1997). BDNF was also found to increase the number of rubrospinal neurons that regenerated axons into a PN graft, indicating that increased GAP-43 and Ta l -tubulin expression correlate with regeneration of rubrospinal axons (Kobayashi et al., 1997). A number of studies have looked at the application of BDNF to the avulsion model of motoneuron cell death. L i and co-workers (1995) investigated whether treatment following avulsion with NGF, BDNF,IGF-I or GDNF could protect motoneurons from cell death. In this study only GDNF was found to protect motoneurons. Wu and co-workers (1995) compared the effects of GDNF, BDNF, CNTF, and IGF-f on the survival of adult rat spinal motoneurons following spinal root avulsion. At 6 weeks post avulsion, BDNF or GDNF application resulted in . 30 the survival of 90% of motoneurons, compared to only 25% survival in control groups. CNTF and IGF-I were unable to rescue spinal motoneurons from cell death. Novikov et al. (1997) tested the longevity of BDNF's survival effects on spinal motoneurons. Following a 2-week infusion of BDNF into the lumbar subarachnoid space, animals were examined at 12 weeks post-operatively for the extent of motoneuron survival and regeneration of motoneuron axons across the CNS-PNS interface. Treatment with BDNF resulted in 45% motoneuron survival at 12 weeks, compared with only 10% survival of motoneurons in vehicle treated animals. A study by Kishino et al. (1997) showed that prolonged administration (4 weeks) of BDNF following avulsion of L4, L5 and L6 roots was able to rescue motoneurons from cell death. In the BDNF treated rats, motoneuron survival was 87% of that on the unlesioned side, compared to 29% survival in PBS treated animals. In animals that received 2 weeks of BDNF treatment, results were similar (84% survival with BDNF treatment vs. 46% survival with PBS treatment). Furthermore, when BDNF treatment was delayed by 2 weeks following avulsion, motoneuron survival was significantly increased compared with 4 weeks of PBS treatment (47% survival with BDNF vs. 29% survival with PBS). BDNF was also shown to have positive effects on soma size and axonal outgrowth. In the BDNF treated group, a cluster of cells was seen in the ventral horn at the lesion site. In addition, BDNF was able to restore the cholinergic markers ChAT and AchE (thought to signify a return of basic neuronal functions) whether administered immediately after avulsion or following a 2 week delay. Although a number of studies support BDNF's role in rescuing both neonatal and adult motoneurons, there exist discrepancies in the literature. Clatterbuck and co-workers reported that BDNF applied with gelfoam was unable to prevent the decrease in ChAT immunoreactivity in adult axotomized motoneurons (Clatterbuck et al., 1994). Following avulsion in the mouse, , 31 GDNF, but not BDNF, NGF, or IGF-I, was able to rescue 50% of C7 motoneurons at 3 weeks post-injury. Although not statistically significant, there was however a trend for increased motoneuron survival in BDNF and IGF-I treated animals (Li et al., 1995). Furthermore, it seems that the dose and mode of BDNF application affect the overall motoneuron survival and CBR. The above experiments support the use of BDNF to prevent or delay motoneuron death, and increase regeneration of injured motoneuron axons. In Part II of this thesis we tested the hypothesis that delayed BDNF infusion (at the time of PN graft insertion, 10 days following C5 and C6 avulsion) will increase both motoneuron survival and axon regeneration into a PN graft. 32 Materials and Methods Part I: Implantation of a PN graft into the spinal cord at different time periods following C5 & C6 spinal nerve root avulsion Spinal nerve root avulsion (Fig. 2.1) Adult male Sprague-Dawley rats (n=43; 200-3OOg) were anesthetized with intraperitoneal injections of Ketamine (60 mg/kg) and Xylaxine (8 mg/kg). A dorsal midline neck incision was made, and the left paraspinal muscles were exposed and mobilized. The C4 and C5 lamina were identified, and a left dorsal hemi-laminectomy was performed. The dura was opened to expose the left lateral aspect of the cervical cord. Using a fine glass hook and steady traction, the C5 and C6 dorsal and ventral rootlets were avulsed from the spinal cord under direct vision. The C5 and C6 roots were chosen for avulsion to preserve most of the sensation to the paw (ie. C7, C8, and Tl) in order to decrease the chance of autotomy of the upper limb. The dura was left open and the laminectomy site was covered with Gelfoam (Upjohn Inc.) in those animals undergoing a delayed implantation of a PN graft Implantation of an autologous PN graft into the spinal cord (Fig. 2.2) Following avulsion, one end of a 3.5cm segment of the common peroneal nerve that had been harvested from the animal's leg at the time of PN graft implantation, was implanted into the cord. In all animals undergoing delayed implantation of the PN graft, the cervical cord was exposed through the same midline incision. Two small openings were made in the pia at the ventrolateral aspect of the cord at C5 and C6 using a 30ga. needle. The distal end of the peroneal nerve was desheathed of epineurium and teased into 2 fascicles. Using a curved fine tipped glass pipette, each fascicle was then inserted a depth of 1 to 1.5 mm into the ventrolateral portion of the cord at C5 and C6. Gelfoam was placed over the exposed spinal cord, and the middle portion of the peroneal nerve was secured to the paraspinal muscles with a 10-0 epineurial suture. The free end of the PN graft was marked with a suture, and left under the cervical musculature. The animals were divided into 5 groups, according to the time interval between root avulsion and implantation of a PN graft: 1) immediate implantation (n=l 1), 2) lweek (n=9), 3) 1 month (n=9), 4) 2 months (n=9), and 5) 6 months (n=5). 33 Tracer application to the PN graft (Fig. 2.3) To identify spinal cord neurons that had regenerated axons into the graft, a fluorescent axonal tracer, Rhodamine -Dextran Amine (RDA; "Fluoro-Ruby", Molecular Probes Inc., cat.# D-1817) was applied to the free end of the graft. Two months following PN graft insertion, the animals were re-anesthetized and the free end of the graft was exposed and transected 2.5 cm from its site of insertion into the spinal cord. The tracer was applied by placing a piece of Gelfoam soaked in 25% RDA solution (in phosphate buffered saline (PBS/3% Triton X) to the free end of the graft for a period of one hour. The wounds were closed, and animals were sacrificed 5 to 7 days later. Perfusion and Tissue Processing. The rats were deeply anesthetized and perfused intracardially with PBS (pH=7.4) followed by 200-300 ml of fixative containing 4% paraformaldehyde in .1 M PBS (pH=7.4). The cervical spinal cord and PN graft were exposed to determine if the PN graft was in continuity with the cord, and the spinal cord from C3 to C8 was excised. Tissue blocks were post-fixed in 4% paraformaldehyde solution for a few hours, and equilibrated overnight in 18% sucrose solution at 4 °C. The cords were frozen in isopentane and cut into 20 um thick transverse sections on a cryostat. Every 4 th section was kept and mounted on slides for viewing with fluorescence microscopy. Counts were done on every 4 th section to decrease the chance of counting the same labeled cell more than once, as the diameter of over 99% of all spinal motoneurons in the rat is less than 60 um (Swett et al., 1986). The number of neurons that had regenerated into the graft was determined by counting only those cell bodies that were: i) RDA-labeled ii) located in the ipsilateral ventral horn and iii) exhibited either a nucleus or a cell process. Statistical analysis The mean number of RDA-labeled neurons in the ipsilateral ventral horn in each of the groups was compared using a Kruskal-Wallis one-way analysis of variance (ANOVA), with the time between root avulsion and PN graft insertion as the single variable factor. 34 Fig. 2. Methods: Part I 2.1 Root avulsion 2.2 P N insertion 2.3 Tracer application 35 Part II: Prelabeling of C5 and C6 motoneurons with Fluorogold prior to C5 and C6 nerve root avulsion A separate group of experiments were done, in which the C5 and C6 motoneurons were retrogradely labeled with a different fluorescent tracer, Fluorogold (FG, Fluorochrome Inc., Engelwood, CO.), prior to avulsion in order to permit definitive identification of C5 and C6 motoneuron cell bodies. This allowed us to distinguish those C5 and C6 motoneurons that only survived root avulsion (only FG positive), from those that survived and regenerated an axon into the PN graft (double-labeled with FG and RDA). Prelabeling of C5 and C6 motoneurons with Fluorogold (Fig. 3.1) Three to 5 days prior to root avulsion (to allow for retrograde transport of FG), the left C5 and C6 spinal nerves were exposed within the paraspinal muscles, at their point of exit through the intervertebral foramen. 1.5 ul of 5% FG was injected into each of the left C5 and C6 spinal nerves at this site. FG was used as a tracer since it has been shown to persist within motoneurons for at least 15 months after retrograde labeling (W. Tetzlaff, personal communication). In addition, FG pre-labeled neurons are able to re-grow their axons for, long distances following axon transection (Mansour-Robey et al., 1994). Some axonal tracers are only taken up and retrogradely transported to the cell body by axons that have been injured (ie. axotomized or crushed axons). However, an additional injury to C5 and C6 motoneuron axons at the time of FG injection would best be avoided, since it does not reproduce the injury that occurs in human root avulsion injuries. It may also result in some additional effects on the motoneurons such as an up-regulation of certain genes, or possibly increased cell death. Recent studies have suggested that motoneurons can be labeled by FG without injuring the ventral root (Koliatsos et al., 1994). To confirm this, and to estimate what proportion of motoneurons might be unlabeled in the absence of a concomitant injury to the spinal nerve, 1.5 ul of FG was injected into each left C5 and C6 spinal nerves of otherwise normal animals. There were 2 groups of animals; 1) spinal nerves left intact (n-4), 2) spinal nerves crushed immediately at the point of injection (n=3). Animals were sacrificed 5 days later, and counts of labeled motoneurons in the ventral horn were made from every fourth 20 um thick 36 3.4 Tracer application 2 months cryostat cut transverse sections. Counts of FG-labeled motoneurons were almost identical in both groups. The mean for animals with intact nerves was 1940 labeled C5 and C6 motoneurons, compared with 1856 for animals with crushed spinal nerves (Table 1). This indicated that intact spinal motoneurons are able to take-up and retrogradely transport FG as effectively as injured axons. This has also been demonstrated for facial motoneurons (W. Tetzlaff, communication). Therefore C5 and C6 spinal nerves were left intact at the time of FG injection for all experimental animals in Part II. The pooled mean of 1900 FG labeled motoneurons was used as an estimate of the total number of motoneurons in the ventral horn at the C5 and C6 level of the spinal cord, prior to avulsion. Table 1. Fluorogold uptake in intact vs. crushed spinal nerves Nerves Intact Nerves Crushed Animal FG Labeled MN's Animal FG Labeled MN's 1 1336 1 1423 2 1583 2 1587 3 2554 3 2559 4 2289 Mean 1940 Mean 1856 Pooled Mean= 1900 In the experimental animals (that did not undergo nerve crush), left limb function was assessed the day of FG injection, following recovery from anesthesia, and again 3-5 days later at the time of root avulsion. In all animals there was no visible difference in posture, movement, or weight bearing between the left and right arm, indicating that neither the method of FG injection nor the presence of FG resulted in clinically significant disruption of C5 or C6 motoneuron axons. Root Avulsion (Fig 3.2) Three to 5 days after pre-labeling of motoneurons with FG, the left C5 and C6 dorsal and ventral roots were avulsed as described in Part I. 38 Implantation of a PN graft into the spinal cord and application of tracer (Fig. 3.3 & 3.4) Harvesting of the PN graft, graft insertion, application of RDA, and perfusion and tissue processing was carried out as described in Part I. The animals in Part II were divided into 2 groups based on the time interval between root avulsion and PN graft implantation: 1) immediate (n=9), or 2) 1 month delay (n=8). The control animals in part III (vehicle pump, n=7), in which the graft was implanted 10 days after root avulsion, were used as a third group in Part II for statistical analysis. Counts of labeled spinal cord neurons in the ventral horn By using two different fuorescent tracers, FG to pre-label C5 and C6 motoneurons, and RDA to label those neurons that extended axons at least 2.5 cm into the PN graft, three different populations of labeled neurons were identified in the ventral horn: 1. FG positive but RDA negative. These would be C5 and C6 motoneurons that survived the root avulsion injury, but did not regenerate axons into the PN graft. 2. FG negative but RDA positive. These would be neurons that regenerated axons into the PN graft, and could be either; a) motoneurons from adjacent cervical levels (ie. C4 and /or C7), b) non-motoneurons (ie. interneurons), or c) C5 or C6 motoneurons that either did not take up the FG label initially, or subsequently lost it. The latter would likely be a very small population of cells, given the efficiency and persistence of FG as a tracer. 3. FG positive and RDA positive. These double-labeled cells would be C5 and G6 motoneurons that both survived the root avulsion and regenerated axons into the PN graft. Counts of all animals were made blindly, from every 4 th 20 um thick transverse cryostat cut sections, under fluorescence microscopy, as in Part I. Statistical Analysis A one-way ANOVA was used to make comparisons of variables 1,3,4 & 5 between the immediate, 10 day and 1 month groups. The Kruskal Wallis one-way ANOVA on ranks was used in variable 2, due to a non-Gaussian distribution. The variables analyzed where: 1) The number of surviving C5 and C6 motoneurons (FG +) 39 2) The number of C5 and C6 motoneurons that survived and regenerated axons into the PN graft (FG + and RDA +) 3) The proportion (%) of surviving C5 or C6 motoneurons that extended axons into the PN graft. 4) The total number of neurons that regenerated axons (all RDA+) 5) The proportion (%) of total neurons with regenerated axons that are C5 or C6 motoneurons Part III: Intra-thecal application of BDNF following root avulsion and PN graft implantation A number of studies have shown that BDNF can prevent adult motoneuron cell death following spinal root avulsion (Wu et al.,95; Novikov et al., 97; Kishino et al.,97). Intrathecal infusion of BDNF following avulsion was also shown to induce local sprouting of injured axons at the site of avulsion (Kishino et al., 1997). We investigated the effectiveness of BDNF in increasing both the survival and regeneration of motoneurons following C5 and C6 avulsion and PN graft implantation. In a separate group of experiments, BDNF was infused for a period of 2 weeks, following C5 and C6 root avulsion and PN graft implantation. The PN graft and BDNF pump were applied 10 days following the initial avulsion injury. We chose a 10-day delay before graft and pump insertion since we felt that in a clinical situation, approximately 10 days would be needed to conduct the appropriate pre-surgical investigations and pre-operative planning for such a surgery. Prelabeling of C5 and C6 motoneurons Three to 5 days prior to root avulsion, the left C5 and C6 spinal nerves were exposed and injected with FG as described in part II. Root avulsion Three to 5 days after FG injection, the left C5 and C6 dorsal and ventral roots were avulsed as described in part I 40 Implantation of a PN graft into the spinal cord Ten days following the avulsion injury, animals received a PN graft as described in part I. At this time osmotic mini-pumps filled with either a) vehicle or b) BDNF were placed in the sub-arachnoid space at the site of the avulsion and PN graft implantation. Preparation and Insertion ofBDNF or PBS pumps One day prior to PN graft and pump insertion, 2 week Alzet osmotic mini-pumps (cat# 2002, Alza Corp., California) were prepared and filled with either a) vehicle (20 mM PBS) or b) BDNF in PBS solution and left overnight to equilibrate and begin pumping in a sterile 37 °C PBS bath. Immediately after graft insertion, a 1.5 cm length of PE-10 silastic tubing that had been attached to the cannula of the pump, was inserted into the subarachnoid space adjacent to the site of the root avulsion through a small opening in the dura at C7. The tip of the tubing was located at the C4 level. The cannula and tubing were secured to the adjacent soft tissue with 5-0 silk sutures. The pump reservoir was implanted subcutaneously in the space overlying the thoracic spine. Nine animals received BDNF pumps, and 7 animals received vehicle pumps. The pumps administered their solutions at a rate of 0.5 ul/hr (24u.g/day) for 14 days. BDNF was generously supplied by the Amgen Corporation. Tracer application to the PN graft Two months after graft and pump insertion, the free end of the graft was exposed and RDA applied as in Part I. Perfusion and tissue processing were carried out 1 week later as described in Part I. Counts of labeled spinal cord neurons in the ventral horn Counts of labeled neurons were carried out as in part II. Statistical Analysis The t-test or Mann-Whitney Rank Sum test was used to make the same comparisons as in part II, between animals that received either a) vehicle pump or b) BDNF pump. 41 Results Part I: Influence of time delay between root avulsion and PN graft insertion on axonal growth RDA-labeled spinal neurons Forty-four animals that underwent C5 and C6 spinal root avulsion followed by insertion of a PN graft into the spinal cord at different time intervals following avulsion, were processed and their cords examined. None of the animals exhibited clinical evidence of a spinal cord injury; all animals exhibited normal walking, bladder function, and a normal response to tail pinch. The only visible deficit in the lower extremities was absence of left ankle and toe dorsiflexion, due to harvesting of the common peroneal nerve. The PN graft was found attached to the spinal cord in all 44 animals. The lateral margin of the cord was bulging towards the graft, but there was no cavitation. Fluorescence microscopy showed the inserted end of the two PN fascicles was located in the ventral half of the cord in all animals, usually just adjacent to or slightly within the lateral portion of the ventral horn (Fig. 5. A). RDA-labeled spinal neurons were present in 43 of 44 animals (Table 2.) with over 95% being situated in the ipsilateral ventral horn (Fig 5. B&C). However, occasional RDA-labeled neurons were found in the ipsilateral dorsal horn, ipsilateral laminae VII and X (of Rexed, 1954), and the contralateral ventral horn. These cells were not included in cell counts used for statistical comparison. One animal had no RDA-labeled neurons, but this was likely due to a technical problem in RDA application, since both PN fascicles were located within the spinal cord and there was no obvious damage to the cord in this animal. Using the cell counts in the immediate insertion group (n=l 1 animals; mean of 761 cells =100%) for comparison, the number of RDA-labeled neurons were 82% at lweek (n=9), 44% at 1 month (n=9), and 38% at both 2 months (n=9) and 6 months (n=5). Compared to the immediate 42 group, counts were significantly different in the 1, 2, and 6 month groups (p<.05; Kruskal-Wallis one-way ANOVA). Fig. 4. Histogram: Part I Results Counts of RDA-labeled neurons in animals that underwent insertion of a PN graft into the spinal cord at various times after C5 & C6 root avulsion. 900 800 700 600 500 400 300 200 100 0 n= RDA Labeled Neurons in the Ventral Horn 761 11 Immediate 336 290 9 9 5 1 Month 2 Months 6 Months * Significantly different than the immediate group (p < .05; Kruskal-Wallis one-way Anova) n - number of animals. 43 Table 2. Results Part I: Spinal Cord Cell Counts ANIMAL RDA LABELED CELLS IN THE IPSILATERAL VENTRAL HORN TIME BETWEEN ROOT AVULSION & PN GRAFT INSERTION IMMEDIATE 1 WEEK 1 MONTH 2 MONTHS 6 MONTHS 1 459 942 349 906 308 2 288 1099 406 308 167 3 106 553 380 219 230 4 301 200 224 261 376 5 1404 350 269 395 358 6 1354 798 543 174 7 706 423 300 60 8 1143 505 360 217 9 565 133 196 66 10 826 11 1214 n= 11 9 9 9 5 MEAN 760.7 555.9 336.3* 289.6* 287.8* % O F IMMED 100 82 44 38 38 SEM 138.8 110 35 84.9 39.4 SD 460.4 330 105.1 254.6 88.1 RANGE 106-1406 133-1099 196-543 60-906 167-376 MEDIAN 706 505 349 219 308 * Significantly different than the immediate group (p< .05; Kruskal-Wallis 1-way ANOVA). Fig. 5. Light and fluorescence micrographs: Part I Light (A) and corresponding fluorescence (B) micrographs of the same 20 um thick transverse section of the cervical spinal cord from an animal in which a PN graft (PNG) was inserted into the ventral aspect of the cord immediately after avulsion of the C5 and C6 ventral and dorsal roots. Many RDA-labeled neurons in the ventral horn (VH) extended axons into the PNG. (C) High power of RDA-labeled V H neurons in a different animal. B & C: Rhodamine filter. Bar = 100 um. 45 These studies demonstrate that a delay of one month or more between root avulsion and PN graft implantation results in a greater than 50% reduction in the total number of ventral horn neurons that extended axons into the PN graft, as compared to immediate graft implantation. Based on their size and location, the vast majority of RDA-labeled cells in the ipsilateral ventral horn are likely to be C5 and C6 motoneurons, but they could also be local spinal interneurons, or motoneurons from other levels (ie. C4 and CI), whose axons would not have been injured by root avulsion. Part II addresses this issue. Part II: Influence of time delay between root avulsion and PN graft implantation on motoneuron survival and axonal growth A separate group of experiments were done, in which the C5 and C6 motoneurons were retrogradely labeled with FG, prior to root avulsion, in order to permit definitive identification of C5 and C6 motoneuron cell bodies. This allowed us to determine what proportion of C5 and C6 motoneurons that survived avulsion, also regenerated axons into the PN graft. There were two groups of animals, those which received a PN graft either immediately following avulsion (n=9), or after a 1 month delay (n=8). The 10 day delay + vehicle pump group in Part III were included for comparison as an intermediate delay group, since the presence of the pump did not appear to influence survival of C5 and C6 motoneurons (Fig.7; Table 3), or cause any obvious spinal cord damage. Like the animals in Part I, animals in all three groups (immediate, 10 day, and 1 month) displayed normal walking and bladder function, as well as normal pain response following surgery. Labeled neurons in the ventral horn The labeling procedures used in Part II identified three populations of labeled neurons: (1) FG-labeled neurons: Fluorescence microscopy revealed FG-labeled motoneurons were only present in the ventral horn and in dorsal root ganglia. Furthermore, labeling was concentrated in . the C5 & C6 segments of the spinal cord. 47 (2) RDA-labeled neurons: RDA-labeled neurons had a similar distribution as described in Part I. Over 95% of neurons were located in the ipsilateral ventral horn. (3) Double-labeled neurons: Motoneurons were counted as double-labeled if they were visible under both the filter for FG labeling and the filter for RDA labeling. Counts of double-labeled neurons were not based on their appearance using solely the "triple filter". Surviving C5 and C6 motoneurons (Fig. 7. & Table 3.) The number of C5 and C6 motoneurons that survived the avulsion injury in the 1 month group (n=8; mean=157.5), were significantly lower than the immediate (n=9; mean=329.1) or 10 day (n=7; mean=347) groups (p<.05; 1-way ANOVA). The total number of regenerated C5 and C6 motoneurons was significantly lower in the 1 month group (mean=20.4), compared with the 10 day (mean=127.9) and immediate (mean=164.6) groups (p< .05; Kruskal-Wallis 1-way ANOVA on ranks). Furthermore, the proportion of surviving motoneurons that regenerated axons into the PN graft in the 10 day (37.2%) and 1 month (11.7%) groups were significantly lower than the immediate group (50.3%) (p< .05; 1-way ANOVA). The number of C5 and C6 motoneurons that survived but did not regenerate axons into the graft was not significantly different between the three groups. Regenerated ventral horn neurons (Fig. 7. & Table 3.) The total number of ventral horn neurons (including non-C5 or C6 motoneurons) that regenerated axons into the PN graft were significantly lower in the 1 month and 10 day groups compared to immediate PN graft insertion (immediate: mean=594.9; 10 day: mean=386.2; 1 month: mean=357.5; p< .05 1-way ANOVA). The percentage of regenerated ventral horn neurons that are C5 and C6 motoneurons are 31% (immediate group), 33% (10 day group) and 6% (1 month group), with the 1 month group being statistically different than the other two groups. However, the number of non-C5 or C6 motoneurons (ie. RDA+ but FG-) that extended axons into the PN graft was not significantly different between the three groups (immediate: mean=430.3; 10 day: mean=258.3; 1 month: mean=337.1). Therefore, the significantly lower number of total ventral horn neurons that extended axons into the PN graft at 1 month was primarily due to two factors: 1) the greater degree of death of C5 and C6 motoneurons, and 2) the decreased ability of surviving C5 and C6 motoneurons to regenerate axons into a PN graft when the graft is inserted on a delayed basis. Furthermore, the percentage of the total regenerated 48 ventral horn neurons that are C5 and C6 motoneurons are 31% (immediate), 33% (10 day) and 6% (1 month). Both motoneurons from adjacent motoneuron pools and interneurons likely account for most of the other regenerated axons. FG label has been shown to persist in neurons for over 1 year (Tetzlaff personal communication) and it is therefore unlikely that C5 and C6 motoneurons lost their FG label. These studies indicate that in order to improve regeneration of spinal motoneurons following avulsion injuries, PN graft insertion needs to be done as early as possible, in order to increase both the total number of surviving C5 and C6 motoneurons and the proportion of C5 and C6 motoneurons that are able to regenerate axons into the PN graft. As well, strategies to improve regeneration need to focus on improving both motoneuron survival and the ability of injured motoneurons to extend axons into the PN graft. 49 Fig. 6. Fluorescence micrographs: Part II Fluorescence micrographs of the same cervical spinal cord tissue section with Fluorogold (FG) and Rhodamine (RDA) labeled neurons in the ventral horn. Three populations of labeled cells are evident in the micrograph taken using the triple filter (T): Double-labeled cells (*): C5 &/or C6 motoneurons that extended axons into the PN graft. RDA only (arrow): Non-C5 or C6 motoneurons that extended axons into the PN graft. FG only (arrowhead): Surviving C5 &/or C6 motoneurons that did not extend axons into the graft. Bar - 100 um. 50 Fig. 7. Histograms: Part II Results Counts of labeled neurons in animals that underwent insertion of a PN graft into the cord at various times after root avulsion. * 1 month group significantly different than both the 10 day and immediate groups (p<.05; 1 way ANOVA) ** 1 month and 10 day groups significantly different than the immediate group (p<.05; 1 way ANOVA) *** 1 month group significantly different than both the 10 day and immediate groups (p<.05; Kruskal-Wallis 1 way ANOVA on ranks) 52 Fig. 7. Histograms: Part II Results Surviving C5 & C6 Motoneurons (All FG+) I I No regeneration (RDA-) 9 7 8 Immediate 10 Day 1 Month Regenerated Ventral Horn Neurons (All RDA +) 800 700 600 500 400 300 200 100 0 n= 1 430 9 Immediate Other neurons (FG-) LB C5 or C6 motoneurons (FG+) 258 337 *** 20 7 10 Day 8 1 Month 53 Table 3. Summary of Part II Results SURVIVING C5 & C6 MOTONEURONS IMMEDIATE (n=9) 10 DAY (n=7) ^ 1 MONTH (n=8) TOTAL MN's (ALL FG+) 329.1 347 157.5* % OF ORIGINAL MN POPULATION (-1900) 17.30% 18.30% 8.30% TOTAL REGENERATED MN's (FG+&RDA+) 164.6 127.9 20.4*** % OF SURVIVING MN's THAT REGENERATED 50.3 37.2** 11.7* TOTAL SURVIVING, NON-REGENERATED MN's (FG+ &RDA-) 164.5 219.1 137.1 VENTRAL HORN NEURONS THAT REGENERATED AXONS TOTAL NEURONS (ALL RDA+) 594.9 386.2**** 357.5**** TOTAL C5 & C6 MN's (FG+ & RDA+) 164.6 127.9 20.4*** % OF REGENERATED NEURONS THAT ARE C5 O R C 6 MN's 31.30% 33.20% 6.40%* OTHER NEURONS (RDA+ & FG-) 430.3 258.3 337.1 * 1 Month significantly different than 10 day and immediate groups. p< .05 (1-way A N O V A ) ** 10 Day significantly different than immediate group. p< .05 (1-way A N O V A ) *** 1 Month significantly different than 10 day and immediate groups. p<..05 (Kruskal-Wallis 1-way A N O V A on Ranks) 1 Month and 10 day significantly different than immediate group. P<.05 (1-way A N O V A ) 54 Part III: Intra-thecal application of BDNF following spinal nerve root avulsion and PN graft implantation BDNF has been shown to both decrease motoneuron cell death following spinal root avulsion, and induce local sprouting at the site of avulsion (Wu et al., 1995; Kishino et al., 1997; Novikov et al., 1997). In a separate group of experiments, we investigated the effectiveness of BDNF in our avulsion model. Following C5 and C6 avulsion, animals received a PN graft after a 10 day delay. At this time, animals received either a) vehicle pump (n=7) or b) BDNF pump (n=9). Vehicle or BDNF were infused for a period of 2 weeks thereafter. RDA was applied to the PN graft 2 months after graft and pump insertion. Surviving C5 and C6 motoneurons BDNF did not significantly increase the total number of surviving C5 or C6 motoneurons (vehicle: mean=347; BDNF: mean=313.9). Furthermore there was no difference in neither the total number of regenerated C5 or C6 motoneurons (vehicle: mean=127.9; BDNF: mean=l 15.3), nor the proportion of surviving motoneurons that regenerated (vehicle: mean=37.2%; BDNF: mean=36.7%) between the two groups. Regenerated ventral horn neurons Similarly, BDNF had no effect on the total number of regenerated ventral horn neurons (vehicle: mean=386.2; BDNF: mean= 361), nor on the proportion of regenerated neurons that are C5 or C6 motoneurons (vehicle: mean= 33.2%; BDNF: mean= 32.6%). The number of non-C5 or C6 motoneurons that extended axons into the PN graft was not significantly different between the two groups (vehicle: mean=258.3; BDNF: mean=246). The inability of BDNF to either improve C5 and C6 motoneuron survival, or induce ah increased axonal sprouting was unexpected, in light of the body of evidence suggesting its role as a survival factor for adult motoneurons following root avulsion. It appears that this negative result may be due to technical reasons related to difficulty in keeping the tip of the cannula within the subarachnoid space. This is further discussed in the next section. 55 Table 4. Results Part III: Spinal cord cell counts, PBS pumps PN GRAFT AND PBS PUMP IMPLANTED 1 0 DAYS FOLLOWING C 5 AND C 6 NERVE ROOT AVULSION ANIMAL n=7 LABELED CELLS IN THE IPSILATERAL VENTRAL HORN ALL FG+ (F) ALL RDA+ (R) FG+& RDA+ (B) ONLY FG+ ONLY RDA+ PROPORTIONS ( B ) / ( F ) (B ) / (R ) 0 97. 165 40 57 125 0.412 0.242 1 382 567 118 264 449 . 0.309 0.208 11 462 651 212 250 439 0.459 0.326 12 410 431 194 216 237 0.473 0.45 33 268 282 117 151 165 0.437 0.415 25 442 379 145 297 234 0.328 0.383 15 368 228 69 299 159 0.188 0.303 MEAN 347 386.1 127.9 219.1 258.3 0.372 0.332 MEAN(%) 37.20% 33.20% SEM 48 67.2 23.5 33.2 50.33 0.039 0.034 SD 126.9 177.8 62.1 87.9 133.2 0.103 0.089 RANGE 97-462 165-651 40-212 57-299 125-449 .188-.473 .208-.45 MEDIAN 382 379 118 250 234 0.412 0.326 Table 5. Results Part III: Spinal cord cell counts, BDNF pumps PN GRAFT AND BDNF PUMP IMPLANTED 10 DAYS FOLLOWING AVULSION OF C5 AND C6 NERVE ROOTS ANIMAL n=9 LABELED CELLS IN THE IPSILATERAL VENTRAL HORN ALL FG+ '(F)-ALL RDA+ (R) FG+& RDA+ (B) ONLY FG+ ONLY RDA+ PROPORTIONS (B)/(F) (B)/(R) 3 265 419 136 129 283 0.513 0.324 10 253 192 82 171 110 0.324 0.427 13 655 388 239 416 149 0.365 0.616 20 174 260 75 99 185 0.431 0.288 21 331 344 107 224 237 0.323 . 0.311 23 200 353 59 141 • 297 0.295 0.167 30 468 414 192 276 222 0.41 0.464 31 304 462 83 221 379 0.273 0.18 52 175 417 65 110 352 0.371 0.156 MEAN 313.9 361 115.3 198.6 246 0.367 0.326 MEAN(%) 36.70% 32.60% SEM 52.5 28.7 20.8 33.6 30.1 0.025 0.051 SD 157.5 86.1 62.4 100.7 90.3 0.075 0.154 RANGE 174-655 192-462 59-239 99-416 110-379 0.24 0.46 MEDIAN 265 388 83 171 237 0.365 0.311 Discussion The present study has demonstrated that a time delay of 1 month or more between spinal nerve root avulsion and PN graft implantation results in a significant decrease in motoneuron survival and axon regeneration into a PN graft. Furthermore, the decreased axonal regeneration in the groups that underwent PN graft implantation on a delayed basis was due to both an increase in motoneuron cell death, as well as a reduction in the ability of surviving motoneurons to regenerate axons. BDNF administration at the avulsion site neither improved C5 and C6 motoneuron survival, nor increased regeneration of ventral horn neurons. In Part I, our results showed that compared to the immediate PN graft insertion group, counts of ventral horn neurons that regenerated axons into the graft were significantly lower in the 1, 2, and 6 month groups. There was a reduction of greater than 50% of the total number of ventral horn neurons that regenerated axons into a PN graft when PN graft insertion was delayed one or more months. Similarly, Part II results showed the total number of C5 and C6 motoneurons that survived, or survived and regenerated was significantly lower when PN graft insertion was delayed by 1 month. Furthermore, the proportion of surviving C5 and C6 motoneurons that regenerated axons into the graft was significantly lower when PN graft insertion is delayed by 10 days to 1 month. These results suggest that delayed implantation of PN grafts by even 10 days following injury may influence the anatomical and/or functional recovery of patients. Peripheral nerve implantation following the avulsion of one or more spinal nerve roots should be performed within the first month after the avulsion injury and preferably within the first two weeks to significantly improve regeneration of motoneuron axons into the graft. This in turn may translate into better recovery of limb function. Our results showed that the percentage of the total regenerated ventral horn neurons that are C5 and C6 motoneurons was; 31% (immediate), 33% (10 day) and 6% (1 month). This 58 indicates that the majority of regenerated axons came from sources other than C5 and C6 motoneurons, including motoneurons from adjacent motoneuron pools (i.e. C4 or C7) and interneurons. It has been shown that axotomized adult spinal motoneurons are capable of extending "supernumerary" axons after PN injury (Havton and Kellerth, 1987). These supernumerary axons originate from the cell body and form morphologically normal synaptic contacts with other neuronal profiles. In addition, dendrites in close proximity to the spinal injury site can generate "dendraxons" that elongate in the direction of the scar tissue (Havton and Kellerth, 1987; Linda et al., 1985). The motoneuron axonal regeneration demonstrated in our experiments may have included collateral sprouting, supernumerary axons and/or dendraxons. The avulsed C5 and C6 motoneuron axons in our experiments demonstrated true CNS regeneration into the PN grafts. The use of two different fluorescent tracers (FG prior to C5 and C6 ventral root avulsion and RDA applied to the free end of the PN graft 4-5 days prior to sacrificing the animal) identified regenerated motoneurons. Those motoneurons that were double-labeled both survived the injury and extended axons through the PN graft and picked up the second tracer. It is unlikely that the second tracer (RDA) permeated through muscle tissue to the spinal cord/avulsion region to result in non-specific RDA labeling of neurons, since the application of the second label was at a considerable distance (2.5 cm) from the injury site and the spinal cord is protected by meninges. In addition, special care was taken to prevent RDA from contacting the spinal cord. Motoneuron axons therefore grew out of the CNS environment of the spinal cord and grew into the PN graft for considerable distances (-2.5 cm) in order to pick up the second tracer. In addition to regenerating long distances, it is essential that regenerated motoneuron axons be able to effectively conduct impulses. A number of studies have demonstrated that regenerated spinal motoneuron axons are in fact capable of conducting impulses. In one study, 59 avulsed ventral roots that were immediately reimplanted into the spinal cord of rats gave rise to a widespread muscle twitch when the roots were stimulated electrically (Carlstedt et al., 1986). Other groups later confirmed these results (Horvat et al., 1987; Smith and Kodama, 1991; Hoffmann et al., 1996). Motoneurons were shown to effectively reinnervate ventral roots and could be excited or inhibited by stimulating the afferent fibers of the same level dorsal root, indicating functional reflex activity (Cullheim et al., 1989). The regenerated motoneuron axons displayed myelin sheaths of oligodendroglial origin, and nodes of Ranvier occupied by astrocytic processes as is found in CNS tissue. Interestingly, in one study oligodendrocytes were shown to migrate slightly into the implanted root, thereby extending the region of CNS regeneration into the ventral root (Cullheim et al., 1989). It has been shown that following distal axotomy motoneurons will preferentially reinnervate the correct motor nerves (Brushart, 1988). Motoneuron axons preferentially reinnervated the correct femoral nerve branch in juveniles and adults regardless of whether a-gap was intentionally left between the proximal and distal nerve ends, or if the two PN stumps were misaligned before being sutured. These results indicate that adult neurons may retain some pathfinding cues. This is essential since motoneuron axons need to grow considerable distances through peripheral nerves to reach the appropriate target muscles. Since the neuronal injury occurs so close to the motoneuron cell body and results in a number of changes in gene expression, it is unclear what effect this will have on the pathfinding ability of re-growing axons. Furthermore, following avulsion motoneuron axons need to grow for considerably longer distances than those in the Brushart study (1988). Our study demonstrated regeneration of axons from adjacent motoneuron pools (eg. C4 & C7), indicating that there are a number of factors that can affect neuronal pathfinding cues. 60 Studies of primates that involved avulsion of spinal nerve roots followed by the immediate reimplantation of the root have demonstrated muscle reinnervation from nonspecific neurons (Carlstedt et al., 1993). Simultaneous electromyography recordings from agonist and antagonist muscle groups in the arm demonstrated cocontractions, and it was not possible to achieve isolated (voluntary) activation of the two opposing muscle groups (Carlstedt et al., 1993). Nevertheless, the primates were able to achieve sufficient arm movement. Although nonspecific neurons (i.e. motoneurons from adjacent levels or interneurons) contribute significantly to axonal regeneration following avulsion, they may not be able to result in useful activity. Conversely their presence does not seem to be detrimental to the overall functional outcome (Carlstedt et al., 1993). Regrowing axons, regardless of their origin must navigate through the lesion site and cross into the implanted root for successful regeneration. Investigations of the lesion site have found an environment that is somewhat different from normal CNS tissue. The lesion site was composed of a loose web of astrocytic processes surrounding regenerating axons, invading leptomeningeal cells and blood vessels with enlarged perivascular spaces (Risling et al., 1993; Carlstedt, 1997). The extracellular matrix of the scar tissue was immunoreactive for collagen and laminin, which had formed strands linking the ventral horn with the implanted ventral root. These findings show that the regenerating axons are able to grow through an environment that is quite different than their normal CNS environment. Given the literature on the growth promoting properties of PNS tissue, it seems likely that trophic influences from the PN graft allowed injured motoneurons/ventral horn neurons to overcome both the CNS environment and the scar tissue, and may have influenced collateral sprouting from adjacent motoneuron pools. For successful regeneration of injured spinal motoneurons, insertion of PN grafts likely need to be performed prior to the extensive motoneuron cell death that normally follows avulsion 61 injuries. Our results are comparable with other studies that have also shown the time course for motoneuron death following avulsion injuries was gradual. It's been reported that following cervical avulsion 75% of motoneurons die within 6 weeks (Wu., 1993; Wu et al., 1994). Following avulsion of lumbar roots, the rate of motoneuron death has been reported at 45% at 1 week (Koliatsos et al., 1994), to 60% and 90% at 4 and 12 weeks respectively (Novikov et al., 1995; Novikov et al., 1997). However, Linda co-workefs (1993) claim there is no loss of neurons when avulsed roots are re-implanted immediately following avulsion, a finding that contradicts most other studies. Survival of motoneurons is key for regeneration and functional recovery, and strategies for increasing motoneuron survival may improve regeneration and functional outcome. Cell death, tissue damage, and loss of function following nerve root avulsion injuries result not only from the physical injury itself, but also from the cellular and molecular cascades that; occur within days and weeks of the initial injury (Schwab and Bartholdi, 1996). Avulsion of spinal nerve roots is known to cause a number of events, including loss of cholinergic phenotype, cell shrinkage, changes in gene expression (eg. NOS) and eventually retrograde cell death (Hoffmann et al., 1993; Koliatsos et al., 1994; Wu et al., 1994; Novikov et al., 1997). Furthermore, the insertion of a PN graft into the spinal cord following avulsion of nerve roots can alter the expression of certain genes, including NOS, LNGFR and c-jun (Wu, 1996). There are however contradictory findings with respect to changes in gene expression after neuronal injury. It is unclear how these changes in gene expression affect regeneration of avulsed spinal motoneurons. Gene expression may be altered to varying degrees depending on neuronal type and nature of the injuries (Wu, 1996; Kobayashi et al., 1997). Because of the large number of known genes whose expression is altered following avulsion, it is difficult to devise strategies to 62 control the expression of genes that are associated with cell death or cell survival. Furthermore, there are likely a number of genes and/or proteins that have not yet been discovered which might play a role in neuronal cell death following injury. One strategy to improve neuronal survival is to modify the expression of apoptosis regulator genes such as the Caspases, the Bcl-2 and IAPgene families. However, because of the complex interactions of proteins both within gene families and amongst the different gene families, a precise understanding of each protein's function and interactions is needed before gene therapies can be devised. Avulsion injuries essentially result in withdrawal of trophic support to the motoneuron cell bodies by separating the PN component from the cell body. It has been shown that Bcl-2 over expression can rescue BDNF dependent neurons from trophic factor withdrawal (Allsopp et al., 1993), but it is not known whether Bcl-2 over-expression can rescue adult motoneurons following root avulsion injuries. Furthermore, since Bcl-2 acts upstream of the apoptosis pathway (Salvesen and Dixit, 1997), it may be more effective to target genes farther down the "cell death cascade". Increasing the expression of certain Caspase inhibitors may be a more suitable strategy to decrease the incidence of programmed cell death following avulsion of spinal nerve roots. Infusion of certain neurotrophic factors is one way to alter gene expression in neurons following injury. A recent study by Kobayashi and co-workers (1997) looked at the expression of regeneration associated genes following cervical axotomy of rubrospinal neurons and BDNF infusion into the vicinity of the axotomized neurons. BDNF was able to prevent atrophy of the rubrospinal neurons, stimulate the expression of GAP-43 and Tal-tubulin and maintain trkB expression. In a separate experiment, BDNF infusion increased the number of axotomized rubrospinal neurons that regenerated into a PN graft (Kobayashi et al., 1997). Therefore, 63 strategies that promote neuronal survival and decrease cell death (such as BDNF infusion) may ultimately result in increased regeneration. Given the evidence for gradual and extensive motoneuron cell death following spinal nerve root avulsion, our results are not surprising, and suggest PN implantation would need to be performed as soon as possible after injury. The current practice of waiting two-to-three months after the injury occurs to make a diagnosis of a true avulsion injury means that there will likely be less than 10% (Novikov et al., 1997) of the original population of motoneurons available for regeneration following nerve graft implantation. Our estimates indicate that delaying PN graft insertion by 1 month ultimately results in only a small proportion (8%) of the original motoneuron population that survives the injury. Of this small population of remaining motoneurons, only 11% regenerate axons into the PN graft. However, if PN graft insertion is performed immediately, a significantly greater proportion of surviving C5 and C6 motoneurons (50%) regenerate axons into the graft. It remains unclear how the number/proportion of ventral horn axons that regenerate into the graft effects functional recovery, and whether there is a minimal number of axons that need to regenerate through a peripheral nerve to obtain significant functional recovery. We would expect that the greater the number/proportion of axons that regenerate the greater the functional recovery. The existing conservative attitude towards surgical treatment of avulsion injuries needs to be modified. Based on our results, physicians should investigate a suspected avulsion sooner, and explore the spinal cord significantly earlier than is common practice today. The down side is that some patients who have spinal nerve roots that are stretched, but still in continuity with the spinal cord (as opposed to true avulsion) may undergo unnecessary surgical exploration, which may subject them to a variety of medical and surgical complications. In addition, some patients 64 are not referred to the appropriate specialists until many months later, which according to our results would likely greatly decrease the chances of a good functional recovery. Due to both the high incidence of other associated serious injuries in these patients and the difficulty in reliably diagnosing a root avulsion, such nerve implantation will always need to be done on a delayed basis. However, our data suggests the delay should be less than 1 month, and preferably within the first 2 weeks. Recently, these procedures have been performed on a small number of patients, and clinical results suggest that a time delay of greater than 1 month from injury to nerve implantation significantly affects the regeneration and functional outcome (Carlstedt, 1997). We did not evaluate functional recovery in our animals, since the focus of our study was the anatomical assessment of motoneuron survival and axon regeneration. There are also only a few behavioral tests that evaluate upper limb function in rats such as the Grooming test used by Bertelli and Mira (1993). This test evaluates the movement of both upper limbs in response to the presentation of a water bottle over the animals head, and rates the extent of elbow flexion using a grading scale of 1 to 5 based on the highest position the paw reaches. Whereas the study by Bertelli and Mira (1993) involved avulsion of only the C5 root, our studies involved both C5 and C6 roots. This undoubtedly accounts for the marked stiffness in the affected limb observed in our animals, and would prevent the performance of movements in the Grooming test. Another behavioral test of upper limb function that was considered was the reaching task (Whishaw and Pellis, 1990). However, we felt the limb stiffness would make this test not practical There are however numerous studies that have evaluated post-operative function in animals (rats, cats and monkeys) following ventral root avulsion and immediate reinsertion of avulsed roots or peripheral nerve grafts. Following avulsion of the brachial plexus (C5), functional recovery was seen on the 4 t h month after implanting a double PN graft into the dorsal 65 and ventral aspect of the cord (Bertelli and Mira, 1994). Central axons regenerated about 65cm through the PNG to restore normal elbow flexion, and forelimb movements were well coordinated in both voluntary and involuntary movements. However, despite the good re-innervation of the biceps muscle, there was no functional recovery when a single graft was applied ventrally (Bertelli and Mira, 1994). In cats, EMG responses in the spinodeltoid muscle were obtained by stimulating C7 spinal nerve 69 days after it had been avulsed and immediately re-inserted, showing that re-innervation of the muscle had occurred. Electrical stimulation of incoming sensory fibers of the same segment indicated that the regenerated spinal motoneurons participated in normal spinal cord reflex activity (Hoffmann et al., 1996). In a study of primates, Carlstedt et al. (1993) demonstrated that following avulsion and ventral root reinsertion, arm muscles that had been previously denervated, were re-innervated 2-3 months following surgery. Clinical recovery was evident shortly after, and there was a gradual return to normal motor behavior with a full range of motion achieved at 1 year (Carlstedt et al., 1993): Electromyography recordings from agonist and antagonist muscle groups in the arm showed co-contractions, and isolated voluntary contractions from either group of muscles were not seen. Nevertheless, there was sufficient dexterity in overall arm movement in these animals (Carlstedt et al., 1993). The most encouraging results come from the few cases in which PN grafts were implanted into patients following avulsion injuries of the brachial plexus (Carlstedt, 1997). Surgeries were performed from 10 days to 10 months following the injuries. In all patients, electromyography detected responses in those muscles that are normally innervated by the implanted spinal cord segments, but there were co-contractions in the deltoid and triceps muscles (Carlstedt, 1997). Voluntary clinical activity was only seen in those patients that were operated on within a month. Our results are therefore in agreement with Carlstedt's evidence of functional 66 recovery in some patients. The re-implantation of avulsed roots or peripheral nerve grafts into the spinal cord of patients can potentially improve upper limb function in those patients whose injury can be diagnosed and treated within 10 days to 1 month, as recommended by our study. The results we obtained in part III showed intra-spinal BDNF application neither increased the survival of C5 and C6 motoneurons, nor increased axonal regeneration. These results were unexpected considering the large body of evidence that indicates BDNF both increases motoneuron survival and induces increased axonal sprouting. Furthermore, BDNF has been shown to increase the expression of RAGs, which correlates with an increased regenerative ability of axotomized motoneurons (Kobayashi et al., 1997). Recent work in our lab has shown that intra-spinal application of BDNF and GDNF on a delayed basis following nerve root avulsion increases both motoneuron survival and results in a corresponding increase in axonal regeneration. However neither growth factor increased the proportion of surviving motoneurons that regenerated axons into the PN graft (Zwimpfer et al., 1998). The animals in this later study underwent the same experimental procedures as those outlined in this thesis. Peripheral nerve grafts and neurotrophic pumps were implanted 10 days following avulsion of C5 and C6 ventral roots. The likely cause for our negative results was technical. Upon examination of the PN graft insertion site at the completion of the study, the tip of the pump cannula had either pulled away slightly from the insertion site or was outside the subarachnoid space in a number of BDNF and vehicle animals. This meant BDNF was not being administered to the appropriate spinal cord segment or did not enter the CSF. In more recent experiments an effort was made to position the BDNF and GDNF pumps more securely, and was likely the reason for obtaining positive results (Zwimpfer et al., 1998). All BDNF and GDNF animals had substantial concentrations of the factor in CSF obtained from the cisterna magna 7 days after pump insertion. In the future, the application of BDNF or other trophic factors may improve anatomical and functional recovery of patients following PN graft surgery in brachial plexus injuries. 68 Bibliography Adams JM and S Cory (1998) The Bcl-2 protein family: arbiters of cell survival. Science 281:1322-1326. Ahmad I, Leinders-Zufall T, Kocsis ID, Sheperd GM, Zufall F, and CJ Barnstable (1994) Retinal ganglion cells express a cGMP-gated cation conductance activatable by nitric oxide donors. Neuron. 12: 155-165. Allsopp TE, S Wyatt, HF Paterson and A M Davis (1993) The proto-oncogene bcl-2 can selectively rescue neurotrophic factor-dependent neurons from apoptosis. Cell 73:295-307. Alvaro GE, N Spear, S Machelle Manuel, R Radi, CE Henderson, L Barbeito and JS Beckman (1998) Nitric oxide and superoxide contribute to motor neuron apoptosis induced by trophic factor deprivation. J. Neurosci. 18 (3): 923-931. Barron KD (1983) Comparative observations on the cytologic reaction of central and peripheral nerve cells to axotomy. In CC Kao, RP Bunge, and PJ Reir (Eds.) Spinal Cord Reconstruction, Raven Press, New York, pp.7-40. Barron KD, M Banerjee, MP Dentinger, ME Scheibly, and R Mankes (1989) Cytological and cytochemical (RNA) studies on rubral neurons after unilateral rubrospinal tractotomy: the impact of GM1 ganglioside administration. J. Neurosci. Res. 22:331-337. Berkelaar M , DB Clarke, YC Wang, GM Bray and AJ Aguayo (1994) Axotomy results in delayed death and apoptosis of retinal ganglion cells in adult rats. J. Neurosci. 14:4368-4374. Benowitz LI, PJ Apostolides, NI Perrone-Bizzozero, SP Finklestein and H Zwiers (1988) Anatomical distribution of the growth-associated protein GAP-43 / B-50 in the adult rat brain. J. Neurosci. 8:339-352. Benowitz LI, NI Perrone-Bizzozero, SP Finklestein and ED Bird (1989) Localization of the growth associated phosphoprotein GAP-43 in the human cerebral cortex. J. Neurosci. 9:990-995. Benowitz LI, and NI Perrone-Bizzozero (1991) The expression of GAP-43 in relation to neuronal growth and plasticity: when, where, how, and why? Prog. Brain Res.89:69-87. Bertelli JA, JC Mira, and A Gilbert et al., (1991) Reconstruction of the rat brachial plexus: anatomo-morphological basis and functional evaluation. Eur. Med. Biblio. Hand Surg. (Suppl.) 1:264. Abstract Bertelli JA, JC Mira, and A Gilbert et al., (1992a) Anatomical basis of rat brachial plexus reconstruction. Surg. Radiol. Anat. 14:85-86. 69 Bertelli JA, JC Mira, and M Percot-Dechavassine (1992b) Morphologic aspects of biceps brachii in the rat: application to the brachial plexus reconstruction. Surg. Radiol. Anat. 14:271-273. Bertelli JA and JC Mira (1993) Behavioural evaluating methods in the objective clinical assessment of motor function after experimental brachial plexus reconstruction in the rat. J. Neurosci. Methods. 46:203-208 Bertelli JA and JC Mira (1994) Brachial plexus repair by peripheral nerve grafts directly inserted into the spinal cord in rats. J. Neurosurg. 81:107-114. Bisby MA, and W Tetzlaff (1992) Changes in cytoskeletal protein synthesis following axon injury and during regeneration. Molec. Neurobiol. 6:107-123. Bisby MA, W Tetzlaff and MC Brown (1995) Cell body response to injury in motoneurons and primary sensory neurons of a mutant mouse, Ola (Wld), in which wallerian degeneration is delayed. J. Comp. Neurol.359:653-662. Blinzinger K and GW Kreutzenberg (1968) Displacement of synaptic terminals from regenerating motoneurons by microglial Cells. Cell Tissue Res. 85:145-157. Blottner D, and HG Baumgarten (1992) Nitric oxide synthetase (NOS)-containing sympathoadrenal cholinergic neurons of the rat IML-cell column: Evidence from histochemistry, immunohistochemistry, and retrograde labelling. J. Comp. Neurol. 316:45-55. Bohme GA, C Bon, JM Stutzmann, A Doble, and JC Blanchard (1991) Possible involvement of nitric oxide in long term potentiation. Eur. J. Pharmacol. 199:379-381. Bolden M , T Goncharov, Y Goltsev and D Wallach (1996) Involvement of MACH, a novel MORT 1\FADD interacting protease, in Fas/APO-1 and TNF receptor induced cell death. Cell 85: 803-815. Bredt DS, and SH Snyder (1992) Nitric oxide, a novel neuronal messenger. Neuron 8:3-11. Briining G (1996) Localization of nitric oxide synthase in the brain of the frog, Xenopus laeyis. Brain Research 741:331-43. Brushart TME (1988) Preferential reinnervation of motor nerves by regenerating motor axons. J. Neurosci. 8(3):1026-1031. Carlsen RC, J Kiff, and K Ryugo (1982) Suppression of the cell body response in axotomized frog spinal neurons does not prevent initiation of nerve regeneration. Brain Res. 234:11-25. Carlsen RC (1983) Regeneration of sciatic nerve in frogs maintained at 15 °C failure to sustain regeneration after initiation. Exp. Neurol. 82:159-171. 70 Carlstedt T, H Linda, S Cullheim, M Risling (1986) Reinnervation of hind limb muscles after ventral root avulsion and implantation in the lumbar spinal cord of the adult rat. Acta. Physiol. Scand. 128:645-646. Carlstedt T (1991) Experimental studies on surgical treatment of avulsed spinal nerve roots in brachial plexus injury. J. Hand Surg.(Br) 16:477-482. • Carlstedt T, RG Hallin, K G Hedstrom, I Nilsson-Remahl (1993) Functional recovery in primates with brachial plexus injury after spinal cord implantation of avulsed ventral roots. J. Neurol. Neurosurg. Psychiatry 56:649-654. Carlstedt, T, P. Grane, RG Hallin & G Noren (1995) Return of function after spinal cord implantation of avulsed spinal nerve roots. Lancet 346:1323-1325. Carlstedt T (1997) Nerve fibre regeneration across the peripheral-central transitional zone. J. Anat. 190:51-56. Carter DA, GM Bray, and AJ Aguayo (1989) Regenerated retinal ganglion cell axons can form well-differentiated synapses in the superior colliculus of adult hamsters. J. Neurosci. 9:4042-4050. Cheng B, S Christakos and MP Mattson (1994) Tumor necrosis factors protect neurons against metabolic-excitotoxic insults and promote maintenance of calcium homeostasis. Neuron 12:139-153. Cheng EHY, B Levine, LH Boise, CG Thompson and JM Hardwick (1996) Bax-independent inhibition of apoptosis by Bcl-Xl. Nature 379: 554-556. Chittenden T, S Flemmington, AB Houghton, RG Ebb, GJ Gallo, B Elangovan, G Chinnadurai and RJ Lutz (1995) A conserved domain in Bak, distinct from BH1 and BH2, mediates cell death and protein binding functions. EMBO J. 14:5589-5596. Chiu A Y , EW Chen and S Loera (1994) Distinct neurotrophic responses of axotomized motoneurons to BDNF and CNTF in adult rats. Neuroreport 5:693-696. Chiu AY, S Loera, A Rosenthal, S Meyer, and MA Glicksman (1995) Neurotrophic factors prevent the loss of choline acetyltransferase (ChAT) and expression of nitric oxide synthase (NOS) following axotomy of motoneurons in adult rat. Soc. Neurosci. Abstr. 21:279. Chong MS, J Fitzgerald, J Winter, M Hu-Tsai, PC Emson, U Wiess, and CJ Woolf (1992) GAP-43 mRNA in rat spinal cord and dorsal root ganglia neurons: developemental changes and re-expression following peripheral nerve injury. Eur. J. Neurosci. 4:883-895. Chopp M (1993) The roles of heat shock proteins and immediate early genes in central nervous 71 system normal function and pathology. Curr. Opin. Neurol. Neurosurg. 6:6-10. Clatterbuck RE, DL Price, and VE Koliatsos (1994) Further characterization of the effects of brain derived neurotrophic factor and ciliary neurotrophic factor on axotomized neonatal and adult mammalian motor neurons. J. Comp. Neurol. 342:45-56. Crews LL and DJ Wigston (1990) The dependence of motoneurons on their target muscle during postnatal development of the mouse. J. Neurosci. 10:1643-1653. Cullheim S, T Carlstedt, H Linda et al. (1989) Motoneurons reinnervate skeletal muscle after ventral root implantation into the spinal cord of the cat. Neuroscience 29:725-733. Curtis R, HJS Stewart, SM Hall, GP Wilkin, R Mirsky and KR Jessen (1992) GAP-43 is expressed in nonmyelin-forming Scwann cells of the peripheral nervous system. J. Cell Biol. 116:1455-1464. Daniloff JK, G Levi, M Grumet, F Reiger and GM Edelman (1986) Altered expression of neuronal cell adhesion molecules induced by nerve injury and repair. J. Cell Biol. 103: 929-945. David S and AJ Aguayo (1981) Axonal elongation into peripheral nervous system "bridges" after central nervous system injury in adult rats. Science Wash. DC 214:931-933. Davies A M (1996) Paracrine and autocrine actions of neurotrophic factors. Neurochem. Res. 21(7):749-753. Davies A M (1998) Neuronal survival: Early dependence on Schwann cells. Curr. Biol. 8:R15-R18. Dawson VL, T M Dawson, ED London, DS Bredt, and SH Snyder (1991a) Nitric oxide mediates glutamate neurotoxicity in primary cortical cultures. Proc. Natl. Acad. Sci. USA 88:6368-6371. Dawson TM, DS Bredt, M Fotuhi, PM Hwang, and SH Snyder (1991b) Nitric oxide synthase and neuronal NADPH diaphorase are identical in brain and peripheral tissues. Proc. Natl. Acad. Sci. USA 88:7797-7801. DeBilbao F and M Dubois-Dauphin (1996) Time course of axotomy induced apoptotic cell death in facial motoneurons of neonatal wild type and bcl-2 transgenic mice. Neuroscience 71:1111-1119. Deckwerth TL, JL Elliott, CM Knudson et al. (1996) Bax is required for neuronal death after trophic factor deprivation and during development. Neuron 17:401-411. De Felipe C, R Jenkins, R O'Shea, TS Williams, and SP Hunt (1993) The role of immediate early genes in the regeneration of the central nervous system. Adv. Neurol. 59:263-271. Deshmukh M and EM Johnson Jr. (1997) Programmed cell death in neurons: Focus on the 72 pathway of nerve growth factor deprivation-induced death of sympathetic neurons. Molec. Pharmacol. 51:897-906. DeVente JD and HWM Steinbusch (1993) On the stimulation of soluble and particulate guanylate cyclase in the rat brain and the involvement of nitric oxide as studied by cGMP immunohistochemistry. Acta Histochem. 92:13-38. Deveraux Q, R Takahashi, GS Salvesen and JC Reed (1997) X-linked IAP is a direct inhibitor of cell death proteases. Nature 388:300-303. Dobrea GM, JR Unnerstall and MS Rao (1992) The expression of CNTF message and immunoreactivity in the central and peripheral nervous system of the rat. Dev. Brain Res. 66:209-219. Dodd J and TM Jessell (1988) Axon guidance and the patterning of neuronal projections in vertebrates. Science Wash. DC. 242:692-699. Doster SK, A M Lozano, A J Aguayo, and MB Willard (1991) Expression of the growth associated protein GAP-43 in adult rat retinal ganglion cells following axon injury. Neuron 6:635:647. Dragunow M (1992) Axotomized medial septal-diagonal band neurons express jun-like immunoreactivity. Mol. Brain Res. 15:141-144. Dragunow M , D Young, P Hughes, G MacGibbon, P Lawlor, K Singleton, E Sirimanne, E Beilharz and P Gluckman (1993) Is c-jun involved in nerve cell death following status epilepticus and hypoxic-ischemic brain injury? Mol. Brain Res. 18:347-352. Duce IR, and P Keen (1980) The formation of axonal sprouts in organ culture and their relationship to sprouting in vivo. Int. Rev. Cytol. 66:211-256. Ellis RE, J Yuan and HR Horvitz (1991) Mechanisms and functions of cell death. Annu. Rev. Cell Biol. 7:663-698. Ernfors P and H Persson (1991) Developmentally regulated expression of NT-3 mRNA in rat spinal cord motoneurons and expression of BDNF mRNA in embryonic dorsal root ganglion. Europ. J. Neurosci. 3:953-961. Fitzgerald MJT (editor) (1992) Neuroanatomy: Basic and Clinical. BailliereTindall, London. Friedman B, D Kleinfeld, NY lp, V M K Verge, R Moulton, P Bolund, E Zlotchenko, R M Lindsay, and L Liu (1995) BDNF and NT-4/5 exert neurotrophic influences on injured adult spinal motor neurons. J. Neurosci. 15:1044-1056. Funakoshi H, J Frisen, G Barbany, T Timmusk, O Zachrisson, V M Verge and H Persson (1993) 73 Differential expression of mRNAs for neurotrophins and their receptors after axotomy of the sciatic nerve. J. Cell Biol. 123:455-465. Gehrmann J, Y Matsumoto and GW Kreutzberg (1995) Microglia: intrinsic immunoeffector cell of the brain. Brain Res. Rev. 20:269-287. Goldstein ME, SR Weiss, RA Lazzarini, PS Shneidman, JF Lees, and WW Schlaepfer (1988) mRNA levels of all three neurofilament proteins decline following nerve transection. Mol. Brain Res. 3:287-292. Grafstein B, and IG Mcquarrie (1978) The role of the nerve cell body in axonal regeneration. In CW Cotman (Ed.) Neuronal Plasticity, Raven Press, New York, pp. 155-195. Grafstein B (1983) Chromatolysis reconsidered: a new view of the nerve cell body to axon injury. In FJ Seil (Ed.), Nerve, Organ and Tissue Regeneration:- Research Perspectives, Academic Press, New York, pp. 37-50. Greenberg SG, and RJ Lasek (1988) Neurofilament protein synthesis in DRG neurons decreases more after peripheral axotomy than after central axotomy. J. Neurosci. 8:1789-1746. Griesbeck O, ASh Parsadanian, M Sendtner and H Thoenen (1995) Expression of neurotrophins in skeletal muscle: Quantitative comparison and significance for motoneuron survival and maintenance of function. J Neurosci. Res. 42:21-33. Grumet M (1991) Cell adhesion molecules and their subgroups in the nervous system. Curr. Opin. Neurobiol. 1:370-376. Havton L and JO Kellerth (1987) Regeneration by supernumerary axons with synaptic terminals in spinal motoneurons of cats. Nature 235:711-714. Herdegen T, CE Fallios-Estrada, R Bravo and M Zimmermann (1993) Colocalization of the nuclear c-juri protein with galanin in primary afferent neurons and with CGRP in spinal motoneurons following sciatic nerve transection. Mol. Brain Res. 17:147-154. Herdegen T, P Skene and M Bahr (1997) The c-Jun transcription factor- bipotential mediator of neuronal death, survival and regeneration. Trends Neurosci. 20:227-231. Heumann R, S Korsching, C Bandtlow and H Thoenen (1987) Changes of nerve growth factor synthesis in nonneuronal cells in response to sciatic nerve transection. J. Cell Biol. 104:1623-1631. Hoffman PN, and RJ Lasek (1980) Axonal transport of the cytoskeleton in regenerating motor neurons: constancy and change. Brain Res. 202:317-333. 74 Hoffman PN, GW Thompson, JW Griffin, and DL Price (1983) Changes in neurofilament transport coincide temporally with alterations in the caliber of axons in regenerating motor fibres. J. Cell Biol. 101:1332-1340. Hoffman PN, and DW Cleveland (1988) Neurofilament and tubulin expression recapitulates the developemental programme during axonal regeneration: induction of a specific P tubulin isotype. Proc. Natl. Acad. Sci. U.S.A. 85:4530-4533. Hoffmann CFE, H Choufoer, E Marani and RT Thomeer (1993) Ultrastuctural study on avulsion effects of the cat cervical moto-axonal pathways in the spinal cord. Clin. Neurol. Neurosurg. 95 Suppl., S39-S47. Hoffmann CFE, E Marani, J Gert Van Dijik, WVD Kamp and RT Thomeer (1996) Reinnervation of avulsed and reimplanted ventral rootlets in the cervical spinal cord of the rat. J. Neurosurg. 84:234-243. Holscher C (1997) Nitric oxide, the enigmatic neuronal messenger: its role in synaptic plasticity. Trends Neurisci. 20(7):298-302. Hope BT, GJ Michael, K M Knigge, and SR Vincent (1991) Neuronal NADPH-diaphorase is a nitric oxide synthase. Proc. Natl. Acad. Sci. USA 88:2811-2814. Horvat JC, M Pecot-Dechavassine and JC Mira (1987) Reinnervation fonctionelle d'un muscle squelettique du rat adulte au moyen d'un greffon de nerf peripherique introduit dans lamoelle epinere par voi dorsal. Compte Rendu de l'academeie de Science III. 304:143-148. Hynes RO and AD Lander (1992) Contact and adhesive specificities in the associations, migrations and targeting of cells and axons. Cell 68:303-322. Huang PL, TM Dawson, DS Bredt, SH Snyder, and MC Fishman (1993) Targeted disruption of the neuronal nitric oxide synthase gene. Cell 75:1273-1286. Hughes RA, M Sendtner and H Thoenen (1993) Members of several gene families influence survival of rat motoneurons in vitro and in vivo. Ingham PW, KR Howard, and D Ish-Horowitcz (1985) Transcription pattern of the drosophila segmentation gene hairy. Nature 318:439-445. Ito M (1989) Long term depression. Annu. Rev. Neurosci. 12:85-102. Jenkins RW and SP Hunt (1991) Long term increase in the levels of c-jun mRNA and jun protein-like immunoreactivityin motor and sensory neurons following axon damage. Neurosci. Lett. 129:107-110. 75 Jenkins RW, W Tetzlaff, and SP Hunt (1993) Differential expression of immediate early genes in rubrospinal neurons following axotomy in rat. Eur. J. Neurosci. 5:203-209. Kandel ER, JH Schwartz and TM Jessell editors (1991) Principles of Neural Science. 3 r d ed. New York: Elsevier, pp543,612 Kerr JFR, A H Wyllie and AR Currie (1972) Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br. J. Cancer 26:239-257. Kishino A, Y Ishige, T Tatsuno, C Nakayama, and H Noguchi (1997) BDNF prevents and reverses adult rat motor neuron degeneration and induces axonal outgrowth. Exp. Neurol. 144:273-286. Kobayashi NR, A M Bedard, MT Hincke and W Tetzlaff (1996) Increased expression of BDNF and trkB mRNA in rat facial motoneurons after axotomy. Europ. J. Neurosci. 8:1018-1029. Kobayashi NR, D Fan, K M Giehl, A M Bedard, SJ Wiegand and W Tetzlaff (1997) BDNF and NT-4/5 prevent atrophy of rat rubrospinal neurons after cervical axotomy, stimulate GAP-43 and Tal-tubulin mRNA expression and promote axonal regeneration. J. Neurosci. 17(24):9583-9595. Koliatsos VE, TO Crawford, and DL Price (1991) Axotomy induces nerve growth factor receptor immunoreactivity in spinal motoneurons. Brain Res. 549:297-304. Koliatsos VE, RE Clatterbuck, JW Winslow, MH Cayouette and DL Price (1993) Evidence that brain derived neurotrophic factor is a trophic factor for motoneurons in vivo. Neuron 10:359-367. Koliatsos, VE, WI Price, CA Pardo & DL Price (1994) Ventral root avulsion: An experimental model of death of adult motor neurons. J. Comp. Neurol. 342:35-44. Kreutzberg GW (1996) Microglia: a sensor for pathological events in the CNS. Trends Neurosci. 19:312-318. Kuhn TB^ MF Schmidt, and SB Kater (1995) Laminin and fibronectin guideposts signal sustained but opposite effects to passing growth cones. Neuron 14:275-285. Landis SC (1983) Neuronal growth cones. Annu. Rev. Physiol. 45:567-580. Leah J, T Herdegen and B Bravo (1991) Selective expression of JUN proteins following axotomy and axonal transport block in the peripheral nervein the rat: Evidence for a role in the regeneration process. Brain Res. 566:198-207. Li L, RW Oppenheim, M Lei and LJ Houenou (1994) Neurotrophic agents prevent motoneuron death following sciatic nerve section in the neonatal mouse. J. Neurobiol. 25:759-766. 76 Li L, W Wu, LF Lin et al. (1995) Rescue of adult mouse motoneurons from injury-induced cell death by glial cell line-derived neurotrophic factor. Proc. Natl. Acad. Sci. 92:9771-5. Lieberman AR (1971) The axon reaction: a review of the principal features of perikaryal responses to axon injury. Int. Rev. Neurobiol. 14:49-124. r Linda H, M Risling and S Cullheim (1985) 'Dendraxons' in regenerating motoneurons in the cat: do dendrited generate new axons after central axotomy? Brain Research 358:329-333. Lindsay R M (1994a) Trophic protection of motor neurons: Clinical potential in motoneuron disease. J. Neurol. 241 :S8-S11. Lindsay RM, SJ Wiegand, CA Altar, and PS DeStefano (1994b) Neurotrophic factors: from molecules to man. TINS 17(5): 182-190. Lowrie MB and G Vrbova (1992) Dependence of postnatal motoneurons on their targets: review and hypothesis. Trends Neurosci. 15:80-84 Ludwin SK (1990) Phagocytosis in the rat optic nerve following wallerian degeneration. Acta Neuropathol. 80:266-73. Maisonpierre PC, L Belluscio, S Squinto NY Yip, ME Furth, R M Lindsay and GD YancopoulOs (1990) NT-3, BDNF and NGF in the developing rat nervous system: parallel as well as reciprocal patterns of expression. Neuron 5:501-509. Mansour-Robaey S, DB Clarke, YC Wang, GM Bray, AJ Aguayo (1994) Effects of ocular injury and administration of brain-derived neurotrophic factor on survival and regrowth of axotomized retinal ganglion cells. Proc. Natl. Acad. Sci. 91:1632-1636. Marietta M A (1993) Nitric oxide synthase structure and mechanism. J. Biol. Chem. 268:12231-12234. Martini R (1994) Expression and functional roles of neural cell surface molecules and extracellular matrix components during development and regeneration of peripheral nerves. J. Neurocytol. 23: 1-28. Martini R and M Schachner (1988) Immunoelectron microscopic localization of neural cell adhesion molecules (LI, N-CAM and MAG) in regenerating adult mouse sciatic nerve. J. Cell Biol. 106: 1735-1746. Masu Y, E Wolf, B Holtmann, M Sendtner, G Brem and H Thoenen (1993) Disruption of the CNTF gene results in motoneuron degeneration. Nature 365:27-32. Matsumoto T, M Nakane, JS Pollock, JE Kuk, and U Forstermann (1993) A correlation between soluble brain nitric oxide synthase and NADPH-diaphorase activity is only seen after exposure of 77 the tissue to fixative. Neurosci. Lett. 155: 61-64. McPhail LT, JL Vanderluit, S Barclay, D Dizdarevic, KC Harrington, D Xu, SJ Cocker, CS Thompson, A MacKenzie, RG Korneluk, GS Robertson and W Tetzlaff (1997) Expression of neuronal inhibitory protein (NAIP) in axotomized adult and neonatal rat facial motoneurons. Abst. In: Seventh international symposium on neural regeneration, Pacific Grove, California USA. Meyer M , I Matsuoko, C Wetmore, L Olson and H Thoenen (1992) Enhanced synthesis of brain-derived neurotrophic factor in the lesioned peripheral nerve: different mechanisms are responsible for the regulation of BDNF and NGF mRNA. J. Cell Biol. 119:45-54. Milligan CE, D Prevette, H Yaginuma, S Homma, C Cardwell, LC Fritz, KJ Tomaselli, RW Oppenheim and M Schwartz (1995) Peptide inhibitors of the ICE protease family arrest programmed cell death of motoneurons in vivo and in vitro. Neuron 15:385-393. Moncada S, and A Higgs (1993) The L-arginine-nitric oxide pathway. New Engl. J. Med. 329:2002-2012. Narkas AO (1993) Lesions found when operating traction injuries of the brachial plexus. Clin. Neurol and Neurosurg 95 (Suppl.): S56-64. Nathan C (1992) Nitric oxide as a secretory product of mammalian cells. Fedn. Am. Socs. Exp. Biol. J. 6:3051-3064. Nicholson DW and NA Thornberry (1997) Caspases: killer proteases. Trends Biochem. Sci. 22:299-306. Novikov L, L Novikova, and JO Kellerth (1997) Brain-derived neurotrophic factor promotes axonal regeneration and long term survival of adult rat spinal motoneurons. (1997) Neuroscience 79(3):765-774. Oblinger M M , and RJ Lasek (1988) Axotomy-induced alterations in the synthesis and transport of neurofilaments and microtubules in the dorsal root ganglion cells. J. Neurosci. 8:1747-1758. Okado N and RW Oppenheim (1984) Cell death of motoneurons in the chick embryo spinal cord. IX . The loss of motoneurons following removal of afferent inputs. (J. Neurosci. 4:1639-1652. Oppenheim RW (1991) Cell death during development of the nervous system. Annu. Rev. Neurosci. 14:453-501. Oppenheim RW, Y Qin-Wei, D Prevette and Q Yan (1992) Brain derived neurotrophic factor rescues developing avian motoneurons from cell death. Nature 360:755-757. 78 Pearson RCA, N Taylor, and SH Snyder (1988) Tubulin mRNA: in situ hybridization reveals bilateral increases in hypoglossal and facial nuclei following nerve transection. Brain Res. 463:245-249. Pennica D, V Arce, TA Swanson, R Vajsada, RA Pollock, M Armanini, K Dudley, HS Philips, A Rosenthal, AC Kato and CE Henderson (1996) Cardiotrophin-1, a cytokine present in embryonic muscle supports long term survival of spinal motoneurons. Neuron 17: 63-74. Piehl F, U Arvidsson, H Johnson, S Cullheim, A Dagerlind, B Ulfhake, Y Cao, R Elde, RF Pettersson, L Terenius, and T Hokfelt (1993) GAP-43, aFGF, CCK and alpha and beta-CGRP in rat spinal motoneurons subjected to axotomy and/or dorsal root severance. Eur. J. Neurosci. 5:1321-1333. Reed JC (1994) Mini-review: Cellular mechanisms of disease series. Bcl-2 and the regulation of programmed cell death. J. Cell Biol. 124:1-6. Rexed B (1954) A cytoarchitectonic atlas of the spinal cord in the cat. J Comp. Neurol. 100:297-357. Richardson PM, U M McGuinness, and AJ Aguayo (1980) Axons from CNS neurons regenerate into PNS grafts. Nature Lond. 284:264-265. Richardson PM, V M K Issa, and AJ Aguayo (1984) Regeneration of long spinal axons in the rat. J. Neurocytol. 13:165-182. Risling M , K Fried, H Linda, T Carlstedt and S Cullheim (1993) Regrowth of motor axons following spinal cord lesions: distribution of laminin and collagen in the CNS scar tissue. Brain Research Bulletin 30:405-414. Rossier JP, RJ Riopelle and M A Bisby (1996) Axotomy-induced cell death of neonatal rat facial motoneurons: Time course analysis and relation to NADPH-Diaphorase activity. Exp Neurol. 138:33-44. Roy N et al. (1995) The gene for neuronal apoptosis inhibitory protein is partially deleted in individuals with spinal muscular atrophy. Cell 80:167-178. Roy N , QL Deveraux, R Takahashi, GS Salvesen and JC Reed (1997). The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. Salvesen GS and V M Dixit (1997) Caspases: intracellular signaling by proteolysis. Cell 91:443-446. Schreyer DJ, and JHP Skene (1991) Fate of GAP-43 in ascending spinal axons of DRG neurons after peripheral nerve injury: delayed accumulation and correlation with regenerative potential. J. Neurosci. 11:3738-3751. 79 Schwab ME and D Bartholdi (1996) Degeneration and regeneration of axons in the lesioned spinal cord. Physiol. Rev. 76:319-370. Sendtner M , B Holtmann, and RA Hughes (1996) The response of motoneurons to neurotrophins. Neurochem. Res. 20(7):831-841. Shibuki K, and D Okada (1991) Endogenous nitric oxide release required for long-term synaptic depression in the cerebellum. Nature 349:326-328. Skene JHP (1989) Axonal growth associated proteins. Annu. Rev. Neurosci. 12:127-156. Smith KJ and RT Kodama (1991) Reinnervation of denervated skeletal muscle by central neurons regenerating via ventral roots implanted into the spinal cord. Brain Research 551:221-229. Snider WD, JD Elliott and Q Yan (1992) Axotomy-induced neuronal death during development. J.Neurobiol. 20:1231-1246. Snyder SH, and DS Bredt (1992) Biological roles of nitric oxide. Sci. Am. 226:68-77. Stockli KA, LE Lillien, M Naher-Noe, G Breitfeld, RA Hughes, H Thoenen and M Sendtner (1991) Regional distribution, developemental changes and cellular localization CNTF-mRNA and protein in the rat brain. J. Cell Biol. 115:447-459. Stoll G, HW Meuller, BD Trapp and JW Griffin (1989) Oligodendrocytes but not astrocytes express apolipoprotein E after injury of rat optic nerve. Glia. 2:170-176. Streit WJ, MB Graeber and GW Kreutzberg (1988) Functional plasticity of microglia: a review. Glia 1:301-307, 1988. Streit WJ, and MB Graeber (1990) Microglia: immune network in the CNS. Brain Pathol. 1:2-5. Strittmatter SM, SMT Vartanian, and M Fishman (1992) GAP-43 as a plasticity protein in neuronal form and repair. J. Neurobiol. 23:507-520. Sunderland S (1976) Avulsion of nerve roots. In: Injuries of the spinal cord. Part I. Vinken PJ and Bruyh GW eds. New York, Elsevier, 393-435. Swett JE, RP Wikholm, RHI Blanks, AL Swett, LC Conley (1986) Motoneurons of the rat sciatic nerve. Exp. Neurol. 93:227-252. Tello F (1911) La influencia del neurotropismo en la regeneracion de los centres nerviosos. Trab. Lab. Invest. Biol. 9:123-159. 80 Tetzlaff W, M A Bisby, and GW Kreutzberg (1988) Changes in cytoskeletal protein in the axotomized facial nucleus of the rat. J. Neurosci. 8:3181-3189. Tetzlaff W, and M A Bisby (1989) Neurofilament elongation into regenerating facial nerve axons. Neuroscience 29:659-666. Tetzlaff W, and M A Bisby (1990) Cytoskeletal protein synthesis and regulation of nerve regeneration in PNS and CNS neurons of the rat. Restor. Neurol, and Neurosci. 1:189-196. Tetzlaff W, SW Alexander, FD Miller, and MA Bisby (1991) Response of facial and rubrospinal neurons to axotomy: changes in mRNA expression for cytoskeletal proteins and GAP-43. J Neurosci. 11 (8):2528-2544. Thornberry NA and Y Lazebnik (1998) Caspases: Enemies within. Science 281: 1312-1316. Trojaborg W (1994) Clinical, electrophysiological, and myelographic studies of 9 patients with cervical spinal root avulsions: discrepancies between EMG and X-ray findings. Muscle & Nerve 17:913-922. Vanderluit JL, LT McPhail, KC Harrington, S Barclay and W Tetzlaff (1997) Examination in the changes of cell death and survival genes following axotomy of facial motoneurons in the adult and newborn rat. (1997) Abst. In: Seventh international symposium on neural regeneration, Pacific Grove, California USA. Vidal-Sanz M , G M Bray, MP Villegas-Perez, S Thanos, and AJ Aguayo (1987) Axonal regeneration and synapse formation in the superior colliculus by retinal ganglion cells in the adult rat. J. Neurosci. 7:2894-2909. Villegas-Perez MP, M Vidal-Sanz, M Rasminsky, G M Bray and AJ Aguayo (1993) Rapid and protracted phases of retinal ganglion loss follow axotomy in the optic nerve of adult rats. J. Neurobiol. 24:23-36. Vincent SR, and H Kimura (1992) Histochemical mapping of nitric oxide synthase in the rat brain. Neuroscience 46:755-784. Wishaw IQ and SM Pellis (1990) The structure of skilled forelimb reaching in the rat: a proximally driven movement with a single distal rotatory component. Behav. Brain Res. 41(1): 49-59. Wong J, and M M Oblinger (1987) Changes in neurofilament gene expression occur after axotomy of dorsal root ganglion neurons: an in situ hybridization study. Metab. Brain Dis. 2:291-303. Wu, W. (1993) Expression of Nitric-Oxide synthase (NOS) in injured CNS neurons as shown by NADPH diaphorase histochemistry. Exp. Neurol. 120:153-159. 81 Wu, W, K Han, L Li & FP Schinco (1994) Implantation of PNS grafts inhibits the induction of neuronal nitric oxide synthase and enhances the survival of spinal motoneurons following root avulsion. Exp. Neurol. 129:335-339. Wu, W, L Li , JO Penix et al. (1995) GDNF and BDNF inhibit the induction of NOS and prevent the death of adult rat spinal motoneurons following root avulsion. Soc. Neurosci. Abstr. 21:279. Wu W (1996) Potential roles of gene expression change in adult rat spinal motoneurons following axonal injury: a comparison among c-jun, low-affinity growth factor receptor (LNGFR), and nitric oxide synthase (NOS). Exper. Neurol. 141(2): 190-200. Xue.D and HR Horovitz (1997) Gaenorhabitis elegans CED-9 protein is a bifunctional cell-death inhibitor. Nature 390: 305. Yan Q, J Elliott and WD Snider (1992) Brain derived neurotrophic factor rescues spinal motoneurons from axotomy induced cell death. Nature 360:753-755. Yan Q, JL Eliott, C Matheson, J Sun, L Zhang, X Mu, K L Rex and WD Snider (1993) Influences of neurotrophins on mammalian motoneurons in vivo. J. Neurobiol. 24:1555-1577. Yan Q, C Matheson, OT Lopez, and JA Miller (1994) The biological response of axotomized adult motoneurons to brain derived neurotrophic factor. J. Neurosci. 14:5281-5291. Yin X M , ZN Oltvai and SJ Korsmeyer (1994) BH1 and BH2 domains of Bcl-2 are required for inhibition of apoptosis and heterodimerization with Bax. Nature 369:321-323. Zwimpfer TJ, J Liu and C Tarazi (1998) Use of peripheral nerve grafts and intrathecal infusion of BDNF and GDNF to enhance survival and axon regeneration of motoneurons following nerve root avulsion in the adult rat. Congress of Neuological Surgeons, 1998 Meeting, Seattle, Washington. 82 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items