|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 BEHAVIORAL AND ANATOMICAL PLAS TICITY FOLLOWING LOW THORACIC HEMISECTION AND CHONDROITINASE A BC TREATMENT IN THE ADULT CAT: ASSESSMENTS OF LOCOMOTION AND THE COUGH REFLEX By STEPHANIE CHRISTINE JEFFERSON A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Stephanie Christine Jefferson
3 To everyone who supported me over the years including all of my family and friends.
4 ACKNOWLEDGMENTS My successes throughout my scientific career wo u ld not have been possible if it werent for the love, support, and guidance I received from countless friends and family. I would like to acknowledge their individual contri butions and express my gratitude and thanks to each one. First, I would like to recognize my family for their invaluable advice and direction throughout my life. Even though they are many m iles away, I carry them all in my heart in everything that I accomplish. I would especial ly like to acknowledge my parents Rawle and Janet Jefferson who laid down a strong foundation of knowledge, te nacity, humble strength, and a passion for life, and who have always been just a phone call away. The li fe lessons they have taught me are priceless and have helped me to b ecome the person that I am today. In addition, I would like to thank my sister Janelle, Aunt A lice, Uncle Gary, Grandma, and my Grandfather who is no longer with us. I ow e my deepest gratitude for their encouragement and assistance during the difficult times of my life. Virginia Woolf once said that Some people go to priests; others to poetry; I to my friends. My friends have helped me keep my chin held high during all the difficult and turbulent times in my life, and es pecially during my graduate school career. I owe a great deal of appreciation and thanks to them for standing by me, and experiencing their friendship has been the greatest gift in my life. I would like to acknowledge Rebecca Grue who has taught me to laugh at life and who has always be en a friend to talk to. Trisha Basten and Michi Krohn have been like sisters to me for the last 8 years and I am comforted every day just knowing that I can count on their support. I would especially like to thank Melissa Williams for her friendship during my graduate career here at the University of Florida. Countless nights of Coldstone icecream, fried pickles, good conversation, and laughing until we cried, helped me get through this
5 difficult and intense journey to a graduate degree Friendships like these make life worth living, and my life has been blessed. Numerous individuals at the University of Florida have helped me accomplish my research goals. I would especially like to acknowledge my committee members Paul Reier, Steven Kautz, Donald Bolser, Andrea Behrman, and Douglas A nderson for his early contributions to my dissertation work. Their guidance has been extr emely helpful and their time, effort, and input have strengthened me as a critical thinker and inde pendent researcher. I would especially like to acknowledge my dissertation mentor Dr. Dena Howland. Her expertise in numerous areas has helped me build a strong foundation of knowledge from which I am excited to progress even further. Over the course of my dissertation work in Dr. Howlands laboratory, I have been lucky to work with many individuals that have provide d much help and support. I would like to acknowledge Nicole Tester for her teaching assi stance with many techniques that have been lucrative for my growth in beha vioral neuroscience research. I also would like to thank fellow graduate student Adele Blum for her numerous days of assistance with behavioral filming, surgeries, perfusions, as well as for being a sound ing board for advice. Thanks are also in order for Wilbur Osteen, Jimmy Lapnawan, Sarah Monde llo, Cui Yang, and Melanie Rose for lending helping hands whenever possible. Their assistan ce has been invaluable. I would also like to thank Dr. Brian Howland whose help with st atistical evaluations was invaluable.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 4LIST OF FIGURES .........................................................................................................................9ABSTRACT ...................................................................................................................... .............10 CHAPTER 1 BACKGROUND ....................................................................................................................12Spinal Cord Injury ............................................................................................................ ......12Demographics, Etiology, a nd Functional Outcomes ....................................................... 12Acute, Secondary, and Chronic Damage ......................................................................... 13Limited Regenerative Capacity of the Mammalian Central Nervous System ........................ 15Plasticity and Regeneration after CNS Damage .............................................................. 15Inhibitory Environment ................................................................................................... 17Myelin Inhibitors of Axonal Regeneration ......................................................................19Glial Response to Injury ..................................................................................................20Inhibitory cellular components ................................................................................. 21Extracellular matrix (ECM) molecules .................................................................... 24Chondroitin Sulfate Proteoglycans (CSPGs) ................................................................... 24Degradation of CSPGs with C hondroitinase ABC (Chase ABC) ......................................... 26Motor Systems Affected by Spinal Cord Injury ..................................................................... 31Locomotion .................................................................................................................... ..31Cough Reflex ...................................................................................................................342 COUGH FOLLOWING LOW THORACIC SPI NAL HEMISECTION IN THE CAT ........ 37Introduction .................................................................................................................. ...........37Methods ..................................................................................................................................38Results .....................................................................................................................................40Discussion .................................................................................................................... ...........423 EFFECTS OF THORACIC SPINAL HE MISECTION AND CHONDROITINAS E ABC TREATMENT ON BASIC LOCOMOTION AND THE COUGH REFLEX IN THE CAT ................................................................................................................................47Introduction .................................................................................................................. ...........47Methods ..................................................................................................................................48Animals ....................................................................................................................... .....48Surgical Procedures .........................................................................................................49Chondroitinase ABC Delivery ........................................................................................50General Locomotor Training ........................................................................................... 50Bipedal Treadmill and Overground Locomotion ............................................................ 51
7 Step Cycle, Swing, and Stance Durations ....................................................................... 52Paw Drag and Kinematic Locomotor Assessment .......................................................... 52Cough Stimulation and Assessment ................................................................................ 53Histological Confirmation of T10 Hemisection ..............................................................54Statistical Analyses .......................................................................................................... 54Results .....................................................................................................................................55Extent of Spinal Hemisection ..........................................................................................55Locomotor Recovery ....................................................................................................... 55Recovery of Characteristics of Basic Locomotion .......................................................... 56Step-cycle duration ...................................................................................................56Stance duration .........................................................................................................57Swing duration .........................................................................................................57Paw drag ...................................................................................................................58Left hindlimb angular kinematics ............................................................................ 59Recovery of the Cough Reflex ........................................................................................61Discussion .................................................................................................................... ...........614 CHONDROITINASE ABC PROMOTES RECOVERY OF SKILLED LIMB MOVEMENTS AND PLASTICITY OF THE RUBROSPINAL TRACT IN THE CAT ..... 72Introduction .................................................................................................................. ...........72Methods ..................................................................................................................................73Animals ....................................................................................................................... .....73Surgical Procedures .........................................................................................................74Treatment Administration ............................................................................................... 75Behavioral Training and Quantita tive Locomotor Assessments ..................................... 75Retrograde Tracing .......................................................................................................... 77Tissue Processing and Histology ..................................................................................... 77Immunohistochemistry .................................................................................................... 78Stereological Analyses of pNF-H .................................................................................... 80Quantification of Rubrospinal Neurons ........................................................................... 80Statistical Analyses .......................................................................................................... 81Results .....................................................................................................................................81Narrow Range of Spinal Hemisections ........................................................................... 81Chase ABC Enhances Multiple Feat ures of Pegboard Performance ............................. 81Chase ABC Treated Cats Develop a N ovel Interlimb Coor dination Pattern .................84Enhancement of Axonal Densities at the Spinal Level in Cats Receiving Chase ABC ........................................................................................................................... ..86More Rubrospinal Neurons have Axons Ca udal to the Hemisection in Chase ABC Treated Cats .................................................................................................................87Discussion .................................................................................................................... ...........88Skilled Pegboard following Thoracic Hemisection and Chase ABC Treatment ........... 89Strategies to Promote Ru brospinal Tract Growth af ter Cervical Axotomy .................... 91Chase ABC Treatment to Promote Spinal Pl asticity and Rubrospinal Tract Growth .... 935 SUMMARY, CONCLUSIONS, AND FUTURE DIRECTIONS ........................................ 104
8 LIST OF REFERENCES .............................................................................................................108BIOGRAPHICAL SKETCH .......................................................................................................128
9 LIST OF FIGURES Figure page 2-1 Range of lesion magnitudes. .............................................................................................. 452-2 Cough characteristics. ........................................................................................................463-1 Cross section lesion extents. ............................................................................................. .653-2 Step cycle duration during bipedal treadmill and overground locomotion. ....................... 663-3 Swing and stance durations during bi pedal treadmill and overground locomotion. .......... 673-4 Stick figure representations of paw drag during swing. ....................................................683-5 Paw drag during bipedal treadmill and overground locomotion. ...................................... 693-6 Swing knee flexion during overground locomotion. .........................................................703-7 Chase ABC increases esophageal pre ssure amplitudes following hemisection. .............. 714-1 Representative range of spinal hemisections. .................................................................... 964-2 Pegboard locomotor recovery is impr oved with Chase ABC administration. ................. 974-3 Limb placement strategies during pegboard locomotion. .................................................. 984-4 Proximal left hindlimb angular kine matics during different pegboard crossing strategies. ................................................................................................................... ........994-5 Phosphorylated neurofilament heavy chain (pNF-H) immunoreactivity within the lesion epicenter. ............................................................................................................. ..1004-7 Retrograde labeling in axotomized red nucleus neurons is increased with Chase ABC administration. ........................................................................................................103
10 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy BEHAVIORAL AND ANATOMICAL PLAS TICITY FOLLOWING LOW THORACIC HEMISECTION AND CHONDROITINASE A BC TREATMENT IN THE ADULT CAT: ASSESSMENTS OF LOCOMOTION AND THE COUGH REFLEX By Stephanie Christine Jefferson May 2009 Chair: Dena Ruth Howland Major: Medical Sciences--Neuroscience Chondroitin sulfate proteoglycan s (CSPGs) are potent inhibitors of neuronal growth following spinal cord injury (SCI). Diges tion of CSPGs with Chondr oitinase ABC (Chase ABC) has been shown to significantly decrease their inhibitory properties in vitro and enhance axonal growth and promote recovery of locomoti on in rodent models and our feline model of SCI. This study assessed the effects of intraspinal Chase ABC delivery following low thoracic hemisection in the adult cat across motor tasks that are mediated by diverse levels of the neural axis. Adult cats received a left spinal T10 hemisection (hx) alone, hx+vehicle control or hx+Chase ABC repeatedly delivered to the lesi on site via a port system. Motor performance was assessed pre and post-injury over the course of 20 weeks during basi c as well as skilled locomotion and during coughing. Two weeks following the last behavioral data collection point, fluorogold (FG) was injected bilaterally into the spinal cord caudal to the T10 hx. Specific components of the gait cycle such as the step cycle duration, swing and stance duration, knee flexion during swing, and paw drag were differe ntially affected during bipedal treadmill and overground locomotion following injury. Chase A BC treatment did not signi fically affect these characteristics assessed during bipedal tread mill or overground locomotion. Chase ABC
11 treatment significantly improved skilled pegboard locomotion. Chase ABC treated cats were able to place their affected hindlimbs on the pegb oard earlier and more frequently than SCI-only cats and developed a unique hindl imb motor strategy that differed from the one used prior to injury. To determine if Chase ABC has an eff ect across diverse motor systems, the cough reflex also was assessed. The general characteristics of cough were not affected by our lesion paradigm as no changes in the parameters evaluated were seen post-hemisection. However, Chase ABC treated cats showed a significan t increase in esophageal pressure amplitudes above pre-injury values. In addition to behavioral improvements and changes, Chase ABC also enhanced axonal plasticity. pNF-H immunoreactivity below the in jury site, as well as retrogradely labeled neurons in the contralateral red nucleus, were significantly greater in cats treated with Chase ABC than in controls. The findings presented are the first to demonstrate that Chase ABC treatment can enhance skilled locomotor be havior, axonal growth, and cough esophageal pressures within the same animals. They also ar e the first to show that Chase ABC can promote plasticity of anatomical subs trates likely to underlie improve ments observed in recovery of locomotor behavior following SCI in a large translational animal model.
12 CHAPTER 1 BACKGROUND Spinal Cord Injury Demographics, Etiology, and Functional Outcomes In the blink of an eye, an entire life can be changed. Precarious behavior or sim ply the normal, daily events of life such as drivi ng a car, walking to wor k, or riding a horse can unpredictably result in a life al tering injury that will affect every facet of daily living. Individuals in the United States that survive trauma to the sp inal cord, approximately 12,000 new cases each year, endure both physical and psyc hological impairments. The approximate 250,000 individuals currently living in the United States w ith SCI also require rehabilitative care for the duration of their lives ca using great personal and societal expense. All demographics are at risk of sustaining a spinal cord injury (SC I), but primarily these injuries transpire among Caucasian, young adult, males. The majority of SCI cases are attributable to motor vehicle accidents (42%), falls (27.1%), acts of violence (15.3%), and recreational sporting activities (7.4%) (NSCISC, 2008). For centuries, the prognosis for survival or recovery was dismal. During World War I, it was estimated that 90% of individuals sustaining a SCI died within one year of the in jury and only one percent survived more than 20 years (Grundy et al. 2002). Sin ce then, improvements in medical a nd surgical care, rehabilitative sciences, and technological advan ces have resulted in improved surv ival rates, care, and general life expectancy. The physical impact of a traumatic spinal cord injury is vast and multifaceted. The extent of sensorimotor loss and dysfunction is depende nt upon the lesion site and breadth of damage. The injury may involve physical transection of the cord, blunt contusion or compression of the cord, or a stretch injury (Schwab and Bartholdi 1996). Theref ore, the functi onal outcome for
13 each individual is unique and the extent of paralysis, pain, spasticity, and necessary rehabilitation and therapeutic intervention is highl y specialized. Injuries to one of the eight cervical segments of the spinal cord can result in qu adriplegia, also termed tetraplegia. Individuals with this type of injury would have motor and sensory deficits in both upper and lower extremities, as well as diaphragm, bowel, and bladder dysfunction. Injuries occurring in the thoracic, lumbar, or sacral regions of the spinal cord result in paraplegia. This injury type causes loss of motor function and sensation in the lower extremities and loss of ot her specific specialized functions corresponding to dermatome level. The completeness of the injury is a strong indi cation of the severity of the injury, and has served as the basis of categori zation of functional severity for clinicians using the American Spinal Injury Association (ASIA) impairment Scale. Conventiona lly, a complete spinal cord injury entails having no voluntary motor or conscious sensory function below the level of the lesion. This definition is difficult in applicati on since the majority of complete injuries are functionally complete but anatom ically incomplete leaving a slight rehabilitative substrate intact (Bunge et al. 1997). An individual with an incomplete injury maintains some sensory or motor function below the injury site. Data from the Model Spinal Cord Injury System since 2000 indicates that the most freque nt neurologic category at discha rge is incomplete tetraplegia (34.1%), followed by complete paraplegia (23%), complete tetraplegia (18.3%) and incomplete paraplegia (18.5 %) (NSCISC 2008). Basic and clinical scientists are actively pursuing therapeutic and rehabilitative stra tegies that target anatomical, biological, and pathophysiological features of injury in order to facilitate functional recovery. Acute, Secondary, and Chronic Damage Trauma to the spinal cord is described as o ccurring in a three phase process consisting of acute, secondary, and chronic damage (Schwa b and Bartholdi 1996; Tator 1998), which may
14 overlap in their temporal progres sion. The initial traumatic mech anical injury resulting from compression, contusion, laceration, or stretch to the nervous tissue acutely affects axons, neurons, and blood vessels in the spinal cord. Within seconds to minutes, the acute process causes hypotension, hemorrhage, ischemic cell d eath, disrupted blood supply and flow, edema, changes in electrolytes and ne urotransmitter accumulation (Sekhon and Fehlings 2001). Spinal shock is used to describe an individual dir ectly following the acute pha se of injury. They experience muscle paralysis, redu ced tone, and loss of reflexes below the level of the injury (Hiersemenzel et al. 2000). The severity of these effects increases with greater damage (Ditunno et al. 2004). Spinal shock may cause the prognosis to appear initially as a complete injury, and it is not until this shock subsides that the true exte nt of damage can be assessed. The secondary process encompasses a con tinued cascade of cellular and biochemical processes from the primary phase, such as edem a, electrolyte changes, and necrosis. Novel processes also occur during the s econdary phase such as formati on of free radicals (Demopoulos et al. 1980), altered ca lcium, sodium, and potassium perm eability (Young and Koreh 1986), lipid peroxidation and hydrolysis (Anderson et al. 1985; Hall and Springer 2004), and inflammation that cause continuing cellular damage and death for weeks after injury. The chronic phase of injury sets in within a fe w months to years. The initial lesion is filled with a fluid filled cyst, filled with a glial scar, or the formation of both occurs. These lesion features may act as physical and chemical barri ers to axonal regeneration. Oligodendrocyte death, and subsequent de-myelination of axons decrease axonal conduction capacity. Demyelination in combination with molecular changes in the surviving neurons and aberrant axonal sprouting can result in chronic pain, which has been examined in animal models of spinal cord injury (Hulsebosch 2002).
15 Limited Regenerative Capacity of th e Mammalian Central Nervous System Plasticity and Regeneration after CNS Damage The mi nimal intrinsic capability within the spin al cord to repair itself following traumatic injury has been a compelling quandary for scientists and physicians dating back as early as the 16th century BC. Ancient Egyptian hieroglyphi c medical papyri desc ribed fractures and dislocations of the neck verteb rae followed by paralysis and sens ory deficits as an ailment not to be treated (reviewed by (Hughes 1988)). Th is dismal diagnosis exemplifies the longstanding ideological dogma that the spinal cord has a limited regenerative capacity following insult or injury. Nearly three centuries later, the Italian medical doc tor Felice Fontana made numerous innovative observations regarding the structure and regeneration of nerves. He observed that following sectioning of the rabbit hypoglossal nerve, the nerve underwent an apparent repair process that he interpreted as a result of injured tissu e reproduction (Fontana 1787). The study of the regenerative processes in the nervous system was heightened by the middle of the twentieth century with the emerge nce of highly specialized neuronal fixation and morphological staining technique s (reviewed by (Gorio 1993)). Santiago Ramon y Cajal, highly regarded as the father of modern neuroscience and well known for his groundbreaking work on the structure of the nervous system and the introduction of the neuron theory, diligently investigated th e degenerative and regenera tive capabilities of the central nervous system (CNS). In his earlier work on dege neration and regeneration of the nervous system, Ramon y Cajal aligned with th e pessimistic dogma that the mature CNS was embodied by an absolute architecture. He wr ote that In adult centres the nerve paths are something fixed, ended, immutable. Everything may die, nothing may be regenerated (Cajal 1991). In somewhat juxtaposed opinion, he also observed and commented on the presence of vigorous but abortive attempts at regeneration due to a postulated lack of trophic support.
16 Ramon y Cajal noted that in order for damaged axons to recapitulate growth following damage, the scientific community must give to the sprouts, by means of adequate alimentation, a vigorous capacity for growth; and, place in front of the disoriented nerve conesspecific orienting substances (Cajal 1991 ). During subsequent decades, the viewpoint that the CNS had the capacity for sprouting of undamaged axons as well as regeneration of damaged ones began to be more universally accepted based on new discoveries in the field. The proof that undamaged axons in the centr al nervous system had the capacity to generate sprouts and collateral projections was a ray of hope that the CNS was a plastic, malleable substrate capable of re-wiring in re sponse to neuronal damage. Following partial denervation of the feline spinal cord, intraspinal axons formed ne w processes termed collaterals and the quantity of sprouting was proportional to the area of damage, the more damage the more sprouting (Liu and Chambers 1958). This rese arch proved that undamaged axons may play an important role following CNS damage by altering their morphology in order to compensate for the environmental substrate surrounding them. Co llateral sprouting of in tact axons within the spinal cord may allow for adaptability and f unctional recovery following spinal damage. Re-organization following CNS damage may occur by regeneration of originally cut fibers or by collateral sprouting by neighboring intact fibers. W ith the advent of the electron microscope came the ability to assess plastic or regenerative changes at the synapse level. Following selective lesions of one pathway of the septal nuclei, a region receiving afferent input from two distinct sources, the distribution of s ynaptic terminals of the surviving fiber pathway was rearranged (Raisman 1969). In the normal, uninjured adult rat brain, afferents from the medial forebrain bundle (MFB) form synapses on soma and dendrites, whereas the fimbrial fibers terminate solely onto dendrites. Following fimbrial lesions, the MFB synaptic terminals
17 maintain their existing synapses, as well as form additional synapses to occupy vacated sites. Following lesions of the MFB, the fibrial fibe rs made synaptic connections on the soma, sometimes positioning on a previously vacated synapse. These experiments further show that de-afferentation mediates the spro uting of intact fibers and that they make functional synaptic connection in response to the injury environmen t. This finding suggested that the synaptic connections in the adult mammalian system are fa r more plastic and malleable than previously believed. Interestingly, Raisman po stulated that this rapid re-wiring of existing intact synaptic connections may halt stimulation of anatomical regeneration following injury in the CNS (Raisman 1969). Many subsequent studies across multiple neuronal systems have demonstrated that collateral sprouting in the adu lt mammalian CNS can act as a compensatory response to spinal cord injury. Tract-tracing studies in adult monkeys after unilate ral pyramidotomy or spinal cord hemisections suggested that there is collateral spr outing of corticospinal axons distal to the injury (Kucera and Wiesendanger 1985; Aoki et al. 19 86). A sprouting response of the ventral corticospinal pathway following transection of the dorsal corticospinal tract at C3 in adult rats also was seen and correlated with improved foreli mb function (Weidner et al. 2001). It has also been found that following thoracic lesion in adul t rats, corticospinal a xons can sprout at the cervical level to form a new relay circuitry onto propriospinal neurons, which seemed related to improved hindlimb function (Fouad et al. 2001b; Ba reyre et al. 2004; Courtine et al. 2008). Inhibitory Environment CNS myelin and oligodendrocytes (Schwab and Ca roni 1988), as well as elem ents of the lesion scar (Fitch and Silver 2008), are recognized as the primary grow th inhibitory substrates to CNS axonal regeneration after injury. Research has focused on the identification of specific inhibitory components of CNS myelin and the scar that inhibit a xonal growth. These
18 components may serve as potential therapeutic targets to increase the axonal growth capacity following injury. Until the early 1980s, a prominent theory dominated the field of CNS regeneration. This theory, formulated by Ramon y Cajal, postulated that the limited regenerative capacity was due to a lack of growth stimulati ng factors existing in the adult CN S. Pioneering work in the laboratory of Martin Schwab test ed this trophic factor hypothesis Researchers co-cultured adult rat optic nerve and sciatic nerve explants with dissociated peripheral neurons in the presence of increased nerve growth factor (NGF) (Sch wab and Thoenen 1985). Axon outgrowth was immense into the sciatic nerve cultures, whereas the optic nerv e explants were non-permissive growth substrates despite the presence of the growth promoting NGF. Therefore, they deduced that the CNS tissue, not the PNS tissue, containe d potent neurite growth inhibitory substrates (Schwab and Thoenen 1985). Oligodendrocytes and myelin were later proven to be two nonpermissive substrates for neurite growth (Sch wab and Caroni 1988). Confirming these findings, it was demonstated that axonal regenerative failure in the chick spinal cord corresponds with the onset of spinal myelination and that with expe rimental delay of mye lination the permissive period for axonal re-growth was extended (Keirstead et al. 1992). Researchers also theorized th at the lack of regeneration in the CNS was due to the intrinsic inability of adult neurons to reactiv ate their growth program. This assumption was disproven by transplant studies of autologous peripheral nerve grafts into th e adult rat brainstem and spinal cord (David and Aguayo 1981; Richar dson et al. 1984), and into the hamster retina (Keirstead et al. 1989). These transplants induced in-growth of axons from different populations of CNS neurons and stimulated elongation over centimeters. Thes e studies proved that the CNS
19 is capable of recapitulating a developmental st ate, and it is the inhibitory environmental surroundings following damage, not the neurons themse lves, that halts the re generative process. Myelin Inhibitors of Axonal Regeneration Axon growth inhibitors associated with m yelin play an important ro le in the failure of regenerative axonal growth in the mammalian CN S following injury. Nogo is debatably viewed as the most potent myelin-derived neurite grow th inhibitor of CNS re generation. Nogo activity was originally characterized by Schwab and coll eagues through myelin fractionation experiments in the adult CNS (Caroni and Schwab 1988b). Th e Nogo gene and receptor have since been cloned and this has aided in the understanding of how this molecule mediates neurite growth inhibition (Fournier et al. 2002). The monoclonal antibody IN -1, directed against Nogo A and recognizing both Nogo A and B, overcame olig odendrocyte-mediated inhibition of axonal growth in vitro (Caroni and Schwab 1988a). In vivo, IN-1 producing hybridoma cells transplanted into young rat cortex increased spro uting and long distance co rticospinal tract axon regeneration in rats (Schnell a nd Schwab 1990). Behavioral tests in adult rats revealed that chronic exposure to IN-1 improved locomotor function as well as sensorimotor reflexes following bilateral transection of the dorsal column s and dorsal corticospinal tracts (Bregman et al. 1995; Fouad et al. 2001a). Similarly, in adult mice immunized with CNS myelin or spinal cord homogenate, transected corticospinal tract axons were capable of long distance axonal regeneration (Huang et al. 1999). Nogo A, B, and C transcript s exist through altern ative promoter usage and alternative splicing (Fournier et al. 2002). They all share a universal ca rboxy terminal domain with two transmembrane regions separated by a 66 amino acid segment (Nogo-66) (F ournier et al. 2002). Nogo-A has been shown to have more than one potent inhibitory domain. The amino terminal domain (Amino-Nogo) as well as the 66 amino ac id stretch, Nogo-66, inhibit neurite extension
20 and may act in a synergistic fashion to inhibit ne urite outgrowth in the in jured CNS (Fournier et al. 2001). Human Amino-Nogo of Nogo-A inhibits cerebellar nerve growth in a dose-dependent fashion (Prinjha et al. 2000), and Nogo-66 demonstrated neurit e inhibitory activity by causing the collapse of growth cones and decreased neurit e extension on assays of E12 chick dorsal root ganglion (DRG) neurons (GrandPre et al. 2000). The receptor for the Nogo-66 protein, (NgR), also has been proven to function as a growth inhi bitor through studies that show that there is a positive correlation between NgR expression level in chick DRG neurons and sensitivity to myelin inhibition (Fournier et al 2001). Blockage of Nogo, NgR, as well as Nogo-66 interaction with NgR may prove to be strategies to improve axonal regeneration in the injured adult CNS. Many other elements inhabit CNS myelin a nd inhibit neuronal re generation, one such component is myelin-associated glycoprotein (MAG) MAG is solely found in myelin sheaths of oligodendrocytes and Schwann cells (Filbin 1995), and inhibits neurite outgrowth of adult CNS neurons in vitro (McKerr acher et al. 1994; Mukhopadhyay et al. 1994). NgR may function as a receptor for MAG (Domeniconi et al. 2002; Liu et al. 2002a) as well as for another myelin inhibitory protein oligodendrocyte-myelin glyc oprotein (OMgp), dem onstrating that NgR may mediate the inhibitory effect of Nogo, MAG, and OMgp, the three main myelin proteins that inhibit axonal growth. Signaling pathways thro ugh these receptor interactions may provide a molecular target for therapeu tic interventions to overcome the lack of CNS regeneration. Glial Response to Injury Traditionally, the scar formed at the lesion epic e nter following spinal trauma is referred to as the glial scar. Infiltration of other cell types primarily fibrobl asts also can occur, especially with disruption of the blood-brainbarrier and dura. In order to stay consistent with previous literature, this dissertation will also use the nom enclature term glial scar to refer to the scar environment, but it should be noted that there are other cells that comprise this scar milieu.
21 Inhibitory cellular components Central nervous system injures lead to a barrag e of molecular and cellular interactions in a futile attempt to repair damaged tissue. The formation of the glial scar composed of astrocytes and connective tissue elements functions immediat ely to reestablish the cellular and chemical integrity of the damaged CNS tissue. Juxtaposed to the role of the glial scar preventing infection and further tissue damage, its formation also generates a physical barri er for axonal growth (Windle and Chambers 1950) and produces inhibi tory molecules that create impediments to axonal regeneration through the injured CNS. The most compelling evidence that the glial scar environment was the major growth impediment to axonal regeneration following mammalian CNS injury was clearly demonstrated by Silver a nd colleagues. Their mi croinjects of adult DRG cells into young adult rat spinal cords showed th at these cells were cap able of extending axons over long distances in myelinated tracts until they reached the glial scar (Davies et al. 1999). The primary cell types involved with the inhibito ry scar environment are astrocytes, microglia, and oligodendrocyte precursors, with some enlistm ent of meningeal and stem cells (Fawcett and Asher 1999). Each cell type is responsible for the up-regulation of par ticular molecules that inhibit axonal regeneration (Faw cett and Asher 1999). These cells are recruited in a highly spatial and temporal fashion corresp onding to their particular function. Microglia from the surrounding tissue and macrophages are the first responders following CNS damage. Microglia activate, divide, and migrate to the injury milieu following CNS damage (Kreutzberg 1996). Macrophages are recr uited to the injury site if the blood-brainbarrier (BBB) is compromised (Kreutzberg 199 6). Microglia have the full capacity, when stimulated in the CNS injury environment, to rel ease inhibitory molecules such as free radicals, nitric oxide and arachidonic acid derivatives (Fawcett and Asher 1999), although conflicting experimental evidence has shown that they can produce neurotoxic or neur oprotective effects in
22 vivo (Streit 1996). Experiments that transplanted microglial cells into the injured spinal cord, including or not including astroc ytes, demonstrated that microg lia were capable of stimulating CNS regeneration (Rabchevsky and Streit 1997). Overall, the evid ence is conflicting regarding the effects of microglia in vitro and in vivo It is theorized that the microglial presence in the glial scar environment following CNS injury may facilitate growth as opposed to inhibiting axonal regeneration. Oligodendrocytes myelinate and, therefore, insulate central nervous system axons. Following CNS injury, axons undergo a degradati on process which results in myelin debris release into the injury environment. The specific inhibitory properties of the myelin-derived group of molecules, was elucidat ed earlier in this manuscript. Mature oligodendrocytes in vitro inhibit neurite outgrowth in a contact-dependent manner and cau se filopodial growth cone collapse (Bandtlow et al. 1990). Oligodendrocyt es in the adult CNS possess the previously mentioned molecules Nogo and MAG, which have been proven to inhibit neurite growth. Oligodendrocytes also express the extracellular matr ix glycoprotein tenasc in-R, a member of the bi-functional inhibitory and growth promoting tenascin family. This glycoprotein is multifunctional and can induce cellular adhesion, differentiation and enhancement of neurite outgrowth as well as inhibition a nd repulsion of axonal growth (Pes heva and Probstmeier 2000). Oligodendrocyte precursor cells (OPCs) are anot her cell type that is recruited during the post-traumatic glial response in the CNS. Thei r identification is marked by staining with an antibody to the inhibitory proteoglycan NG2 and th e presence of receptors for platelet-derived growth factor (PDGF) and fibrobl ast growth factor (FGF) (Grimp e and Silver 2002). These cells have also been shown to express other inhibito ry proteoglycans such as neurocan, phosphacan,
23 and versican (Fawcett and Asher 1999; Asher et al. 2000), and therefore contribute to the inhibitory CNS environment after injury Reactive astrocytes are the major constituen ts of the glial scar. The glial scar predominantly consists of an intermeshed ne twork of astrocyte cells that have become hypertrophic, are densely packed with mini mal extracellular space and are surrounded by extracellular matrix (Fawcett and Asher 1999). Hypertrophic as trocytes are marked by an increased expression of the glia l fibrillary acidic protein (GFA P) (Eng et al. 1987). Numerous in vitro and in vivo experiments have proven that this astroglial environment is exceptionally inhibitory to axonal regeneration (Reier and Houle 1988; McKeon et al. 1991). Tissue culture models of the glial scar have shown that astrocytes are inhibitory to regenerating axons from both peripheral and cen tral neuronal cell type s (Fawcett et al. 1989; McKeon et al. 1995). This inhibito ry nature is most likely due to the direct injury altering the astrocyte function from perm issive to inhibitory. In vivo micro-lesion studies in the adult rat brain that axotomized small numbers of axons but caused minimal damage to the astroglial framework showed that regeneration occurred at far distances to the orig inal lesion, whereas the immediate vicinity of the injury environment became inhibitory to axonal growth (Davies et al. 1996). This experiment extablished that the CNS injury caused astrocytes to release chemorepulsive molecules that was inhibitory to axonal regenerati on. Many studies have examined the secreted, cell surface, and extra cellular matrix (ECM) molecules produced by reactive astrocytes that might pl ay an inhibitory role in a xonal regenerati on (Eddleston and Mucke 1993). In particular, the ECM has been shown to contain a conglomerate of axonal growth-promoting and growth-inh ibitory molecules whose presence determines the success or failure of CNS axonal regeneration.
24 Extracellular matrix (ECM) molecules The failure to support axonal regeneration in the ma mmalian spinal cord has been highly correlated to the presence and up-regulation of i nhibitory proteoglycans associated with glial architecture and the ECM. The field of extrace llular matrix molecules has recently expanded, including families of highly conserved attractant and repulsive guidance molecules such as netrins (Yu and Bargmann 2001), semaphorins (Pasterkamp and Verhaagen 2001), ephrins (Klein 2001), tenascins (Joester and Faissner 2001), integrins, and a variety of matrix proteoglycans, (Yamaguchi 2001). Proteoglycan s are found in the matrices of all tissues, including the brain. They are a complex of polys accharides and protein with immensely rich and broad functions in the body (Rhodes and Faw cett 2004). Glycosaminoglycans (GAGs) are the polysaccharide side chain units on proteoglycans. Proteoglycans are formed from the covalent bonding of these GAG side chains to the core prot ein. There are more than 30 primary structures of proteoglycan core protei ns, and their GAG side chain ma ke-up has been characterized (Bandtlow and Zimmermann 2000). This elucidates the remarkable diversity of proteoglycans within the bodys tissues. Chondr oitin sulfate proteoglycans (CSPG s) are the main axon growth inhibitory molecule in the glial scar and have been shown to be the primary cause of failed axonal regeneration in the injured adult CNS (Silver and Miller 2004). Chondroitin Sulfate Proteoglycans (CSPGs) CSPGs consist of a diverse variety of core prot eins that are covalently bonded to chondroitin sulfate glycosaminoglycan (CS-GAG) side chains (Hartmann and Maurer 2001). CSPGs can be grouped into four major categories: the fam ily of lecticans including aggrecan, neurocan, versican, and brevican (Yamaguchi 2000); phosph acan and receptor protein-tyrosine phosphatase (Maurel et al. 1994); the small leucine-rich proteoglycans in cluding decorin and biglycan
25 (Galtrey and Fawcett 2007); and other CSPG s including NG2 (Galtrey and Fawcett 2007). There is large variation in the num ber and position of GAG chains atta ched to the core proteins. CS-GAGs are composed of disaccharide chai ns of glucuronic acid (GlcA), which are linked to the core protein by the enzyme xylosyt ransferase (Galtrey and Fawcett 2007). Along the sugar side chains, the presence of different disaccharide units results in the formation of specific structural motifs known as CS-A (alterna tive name for chondroitin4-sulfate, 4S), CS-C (alternative name for Chondroin-6-su lfate, 6S), CS-D, and CS-E. The sulfation patterns, created by the chondroitin sulfotransferases, of each of these units affects the binding properties of the GAGs and thus the overall function of th e CSPG (Properzi et al. 2003). CSPGs are important in the development, preservation, and ageing of the normal CNS and are the most abundant proteoglycan f ound within the CNS (Bandtlow and Zimmermann 2000), and there is strong evidence that in the ad ult CNS they may be invol ved in the control of plasticity (Rhodes and Fawcett 2004). CSPGs can function during de velopment as axonal guidance molecules. Areas rich in CSPGs are inhibitory to developing axons within the dorsal root entry zone (Pindzola et al. 1993), and the roof plate of the developing spinal cord (Snow et al. 1990b). Studies utiliz ing isolated embryonic chick dorsal root ganglia showed that CSPGs are inhibitory to neurite outgrowth in vitro (Snow et al. 1990a). It has been demonstrated in vitro that this CSPG mediated developmental growth inhibition is eliminated with the enzymatic removal of chondroin sulphate (Brittis et al. 1992). Many in vitro studies have established the inhibitory nature, during development and in th e adult, of the CSPG glycosaminoglycan (GAG) side chains or of the core prot ein itself (Dou and Levine 1994; Yamada et al. 1997; Niederost et al. 1999; Schmalfeldt et al. 2000).
26 CSPGs are the primary axon growth inhibitory molecule within the glial scar following CNS injury and play a pivotal role in regenerati ve axon failure after injury (Silver and Miller 2004). An ex vivo model of the glial scar provided furthe r evidence of the inhibitory nature of CSPGs following CNS injury. Nitrocellulose filte rs were implanted into the adult rat cerebral cortex and subsequently removed and used as a subs trate for neurite growth in explant cultures. The astrogliotic scar present on the filters cont ained CSPGs and were not permissive substrates for neurite extension (McKeon et al. 1991). Expr ession of several CSPGs, including versican, neurocan, brevican, and NG2 have been shown to dr astically increase after injury to the rat brain (Moon et al. 2002). Degradation of CSPGs with Chondroitinase ABC (Chase ABC) The axonal growth inhibitory properties of CS-GAG chains led to the isolation of chondroitin sulphate-degrading enzy m es purified from several specie s of bacteria (Makarem and Berk 1968; Yamagata et al. 1968; Michelacci and Dietrich 1975; Ke et al. 2005). The bacterial enzyme isolated from Proteus vulgaris degrades CS-A (4-S), CS-B (dermatan sulfate), and CS-C (6-S) and therefore is termed Chondroitinase ABC (Chase ABC). Commercially available Chase ABC has been important in studying the effects of CS-GAG degradation in vitro and in vivo Chase ABC has been the bacterial enzyme of choice for many studies investigating enhancement of axonal regeneration and plastic ity following central nervous system (CNS) damage. CSPGs in the normal developing embryo have been shown to limit axonal growth within typical boundary regions in vitro Snow and colleagues grew isolated E9 chick DRGs on nitrocellulose-coated petri dishes containing st ripes of alternating la nes of the inhibitory molecules KS/CS-PG and the growth promoting molecule laminin. DRGs extended neurites on laminin stripes but when they encountered the KS/CS-PG lanes, the neurites either aborted
27 growth or traveled along the perimeter of the KS/CS-PG lanes. Enzymatic removal of the CSGAG chains with Chase ABC in vitro significantly decreased neur ite inhibition, and when the cultures were treated with Chase ABC and keratanase neurite inhibition was completely abolished. Therefore it is the CS-GAG side ch ains, not the core proteins, which possess the neurite inhibitory activit y (Snow et al. 1990b). CSPGs also have been shown to be expressed in the mature CNS following injury (McKeon et al. 1991). McKeon and colleagues implan ted nitrocellulose filte rs into the adult rat cortex and subsequently the gl iotic scar tissue was explanted in vitro and used as a substrate for embryonic retinal ganglion cell growth. The injury induced CS-PGs inhib ited neurite extension, consistent with the prior literature that these molecules contribute to the minimal regenerative ability following trauma. These explants were th en treated with Chase ABC, which lead to a significant increase in mean neur ite length of embryonic retinal ne urons over the explanted scar surface (McKeon et al. 1995) The first in vivo evidence that enzymatic degrada tion of CS-GAGs with Chase ABC alone could diminish the inhibitory milieu of the injured nervous system was achieved by Moon and colleagues in the rat. Following unilateral nigrostriatal tract axotomy and treatment with Chase ABC, significantly more dopaminergic nigral axons grew through the injury site and back to their original target (ipsilateral striatum) as compared to control treated animals (Moon et al. 2001). These results demonstrated that Chase ABC treatment can enhance CNS axonal regeneration in the adu lt rat nigrostriatal tract and that this means of degrading CS-GAGs may be equally as effective at stimulating growth following CNS damage in other models. Bradbury and colleagues assessed the effects of Chase ABC on corticospinal tract (CST) axons following a cervical dorsal crush spinal injury in the rat using anatomical,
28 electrophysiological and behavioral measurements (Bradbury et al. 2002). In their model, Chase ABC promoted regeneration of CST axons at and below the level of the crush injury, and electrophysiological recordings showed that thes e regenerated axons established functional reconnectivity. Behavioral assessments during skil led grid and beam walk ing tests showed that Chase ABC treatment improved forelimb placeme nt accuracy, and footprint analyses showed that these animals had walking pa tterns near normative values unlik e vehicle controls. Overall, Chase ABC treatment promoted regeneration of CST axons and enhanced functional behavioral recovery. In the same lesion model, Barrit et al. showed that Chase AB C treatment facilitated injured CST axons to sprout around the lesion as well as grow through the lesion environment. Intact serotonergic axons were also found to have sprouted ventral and caudal to the spinal injury and intact spinal afferents were also found to sp rout caudal to the spinal lesion (Barritt et al. 2006). These studies showed that Chase ABC can influence spinal plasticity following spinal cord injury and that compensatory sprouting of intact axonal tracts may be partly responsible for functional recovery after Chase ABC treatment. Massey and colleagues also demonstrated a sprouting phenomena following cervical dorsal colu mn transection. Following application of Chase ABC into the cuneate nuc leus of adult rats following pa rtial denervation of forepaw dorsal column afferents, the remaining primary afferent terminals were able to collaterally sprout into denervated areas of the cuneate nucleus and this was directly linked to functional recovery (Massey et al. 2006). Following low thoracic lateral hemisection, Yi ck and colleagues assessed the ability of Chase ABC to promote axonal regeneration of Cl arkes nucleus (CN) ne urons into a peripheral nerve (PN) graft at the lesion site Implantation of PN grafts alone or in combination with BDNF infusion did not stimulate regeneration of CN neurons, whereas Chase ABC administration
29 promoted axonal growth of CN neurons into the PN grafts (Yick et al. 2000). Similarly, the same laboratory illustrated that application of Chase ABC promot es the growth of CN neurons into the rostral spinal cord th rough the lesion scar environment without PN transplantation in both neonatal and adult rats (Yick et al. 2003). Chase ABC treatment has also been assesse d following contusive spinal cord injury. Caggiano et al. demonstrated th at following moderate or seve re low thoracic contusion, both somatic and autonomic motor recovery was promoted with Chase ABC administration as assessed by open field locomotor and bladder function tests but with no correlative anatomical assessments (Caggiano et al. 2005). More re cently, Iseda and colleagues demonstrated that following a single high-dose treatment with Cha se ABC, CST axons grew around and caudal to the lesion site following low thoracic hemisecti on but not contusion inju ry (Iseda et al. 2008). Cafferty et al. engineered transgenic mice that expressed Chase ABC via the gfap promoter following rhizotomy, and these transgenic mice had significant sensory axon regeneration and functional beha vioral recovery (Cafferty et al. 2007). Similarly, they also found that following partial deaffere ntation of the spinal cord, Chase ABC treatment mediates a functional plasticity of spinal circuitry, and a correlated recovery of function via undamaged afferent contributions (Caffert y et al. 2008). These studies th erefore provide evidence that Chase ABC treatment can also restore sensory function following nervous system damage. Overcoming the impediments to axonal growth following central nervous system is multifactorial, and therefore many studies have assessed combinatorial treatment with Chase ABC and other therapies to promote axonal growth. Combined treatment with BDNF and Chase ABC was found to synergistically promote re tinal fiber sprouting after denervation of the adult rat superior colliculus (T ropea et al. 2003). Chau et al. grafted Schwann cell-seeded
30 channels into low thoracic spinal hemisections in adult rats and delivered Chaes ABC close to the graft-host interface. Previous research ha d shown that due to the CS-PG deposition at the caudal graft-host interface, axons are unable to exit the caudal graft and enter into the caudal host spinal cord (Xu et al. 1995). Chau and collea gues found that with Chase ABC treatment, axons from the rostral cord were able to traverse throug h the graft and into the ho st spinal cord (Chau et al. 2004). In a low thoracic transection model in adult rats, Fouad and colleagues combined Chase ABC infusion to diminish inhibitory CS -PGs, Schwann cell bridges to provide an axonal growth promoting substrate, and olfactory-ensheathing glia (OEG) to allow axons to exit the Schwann cell bridge and re-enter the spinal cor d. Animals receiving this combinatorial strategy showed significant improvements in locomotor recovery as well as serotonergic fiber growth through the cellular bridge and into the caudal spinal cord tissue (Fouad et al. 2005). Following thoracic spinal contusion, Ikegami et al. co mbined Chase ABC treatment with neural stem/progenitor cell tran splantation, which enhanced transp lant cell migration and increased growth of growth-associated protein-43 (GAP-43) pos itive fibers (Ikegami et al. 2005). Houle and colleagues used a peripheral nerve graft (PNG) to bypass a cervical hemisection and direct regrowing axons into the distal spinal cord. Chase ABC treatme nt promoted si gnificant axonal growth past the distal end of th e PNG back into the spinal cord and a correlated with functional behavioral recovery of the affected limb. Transection of the PNG caused complete loss of behavioral gain, suggesting that regenerating axons aided in the be havioral recovery (Houle et al. 2006). All of these combinatorial strategies uti lizing Chase ABC application in concert with other regenerative strategies prove promising as possible therapies following CNS, and especial spinal cord injury.
31 Our laboratory has recently demonstrated, for the first time in a species other than the rodent, that degradation of CS -GAGs with the bacterial enzyme Chase ABC enhances spinal plasticity as well as locomotor recovery in th e adult cat model (Tester and Howland 2008). The basic and skilled locomotor recovery seen in these Chase ABC treated cats was most likely mediated by plasticity of local spinal circui try including the central pattern generator (CPG) and/or plasticity of descending spinal tracts invo lved with the control of locomotion such as the rubrospinal tract. Motor Systems Affected by Spinal Cord Injury Locomotion The locomo tor repertoire of vertebrates is intr icate and precisely orchestrated to perform a broad range of functions; therefore its control systems are multifaceted. The general design of the locomotor control system has been conser ved among vertebrates, from the lamprey to mammals including humans. Vertebrate locomoti on is controlled by a hierarchical tripartite neural system comprised of a cen tral pattern generator (CPG), a fferent feedback, and supraspinal input. The CPG is at the core and is responsible for the basic locomotor pattern, which can be modulated by input from supras pinal centers and peripheral input (Goldberger 1988). There are functionally distinct CPGs for each limb and in combination they can produce different motor patterns requiring differing levels of interlim b coordination. The locomotor CPG provides the basic motor signals common to all forms of loco motion. Unique details necessary for different forms of locomotion, in contrast, depend on interactions among the tripartite system (Buford et al. 1990). Physiological and anatomical studi es also have demonstr ated the presence of propriospinal connections throughout the rostro-cauda l extent of the mammalian spinal cord with ascending or descending fibers that are important for mediating reflex control and coordination during locomotion. The propriosp inal system has varied projection patterns that can span
32 between multiple segments such as between cervical and lumbar levels, as well as projecting only a few segments (Miller et al. 19 73; Conta and Stelzner 2004). The caudal spinal cord of numerous vertebra te species contains the necessary pattern generating circuitry to create locomotor patterns (Grillner et al. 1981; Rossignol and Dubuc 1994; Rossignol et al. 1996). T13 spinalized cats (Barbeau and Ro ssignol 1987) and kittens (Forssberg et al. 1980a; Forssberg et al. 1980b; Howland et al. 1995a) can walk on a treadmill, with plantigrade stepping of the hindlimbs. Therefore, the isolated spinal cord without receiving descending control can still pr oduce complex motor outputs. In many vertebrates, even decerebrate animals, the CPG for locomotion can be activated by stimulation of particular brainstem regions such as the subthalamic nucleus, nucleus cuneiform, pons, and the pyramids (Selionov and Shik 1984). There are at least four locomotor regions located in the brainstem that when stim ulated either electrically or chemically induce locomotion. The first is the mesencephalic loco motor region (MLR), an area in the caudal cuneiform nucleus of mammals that when stimulated initiates locomotion in decerebrate cats placed on a treadmill (Shik and Orlovsky 1976). It s neurons project to the second locomotor region, the medullary reticular formation (MRF) and subsequently to intern eurons in the spinal cord descending via the ventrolateral funiculus (VLF) (Jordan 1998). The MRF region regulates and aids in initiation of stepping pattern as well as interlimb coordination (Whelan 1996) and electrical stimulation of the MRF can produce locomotion (Mori et al. 1978). The third region, just medial to the MLR, referred to as the me dial MLR (mMLR) also has shown to initiates locomotion when electrically stimulated (Garci a-Rill et al. 1983). Axons of the mMLR pass through the fourth region, the pont omedullary locomotor strip (PLS), and continue on to the MRF. The PLS runs through the lateral tegmentum of the brainstem and continues in the spinal
33 cord in the dorsolateral funiculus (DLF) (Whelan 1996). Stimula tion of the PLS, as well as the MRF, elicits locomotor bouts, but the behavior appears fragmented and spastic (Whelan 1996). These pathways descending to the spinal cord via the VLF and DLF contain the capacity to control spinal motor circuitry Neurons originating from several areas of the brainstem have been found to contribute to the initiation and maintainance of locomotion (for review see (Fouad and Pearson 2004)). Descending pathways that influence motor activity can be groupe d into two principle systems according to their medial or late ral position in the spinal cord (Drew et al. 2002). The medial system includes the reticulospinal and vestibulos pinal tracts while the lateral system includes the corticospinal and rubrospinal pathways. Althou gh these systems are all important in locomotor control, this section will focus on the c ontribution of the rubrospinal tract. In the cat, the course of the rubrospinal tract has been well described to immediately decussate in the mesencephalon and send its crosse d bundle of fibers into the lateral funiculus of the spinal cord where it lies just anterior to the lateral corticospinal tract (Verhaart 1953; Verhaart 1955; Pompeiano and Brodal 1957; Hinman and Carpenter 1959; Kuypers 1964; Nyberg-Hansen and Brodal 1964; Schoen 1964). The mammalian red nucleus (RN) can be divided into a caudal magnocellular region (RNm) and a rostral parv ocellular region (RNp). The hindlimb region of the RNm is located ventral and ventrolateral and pr ojects to the lumbar enlargement. Research by Orlovsky and colleagues undertaken in cats walking on a treadmill, elucidated that cells in the RNm display sp ike bursts during active limb movements (Orlovsky 1972a). In the majority of mammals, includi ng cats and primates, the rubrospinal tract influences flexor activity of the forelimb a nd hindlimbs (Kuypers 1964) during the swing phase (Orlovsky 1972b) and helps coordinate the spatiote mporal muscle activity pattern of the limbs
34 (Lavoie and Drew 2002). Several studies suggest that the red nucleus plays a role during coordinated, multi-articular whole limb movements such as reaching (Gibson et al. 1985; Miller et al. 1993; Sinkjaer et al. 1995; van Kan and McCurdy 2001). Re d nucleus neurons have also been shown to increase their discharge freque ncy during voluntary gait modifications such as negotiation of an obstacle (Drew 1993). They are also involved in the control of coordination during locomotion of intralimb as well interlimb activity, and regulate muscle activity during the transport and placement phases of the step cy cle (Lavoie and Drew 2002). This descending system is essential for regulation of the locomoto r cycle and damage to this tract following spinal cord injury causes behavioral deficits. It has been shown previously that following a cervical hemisection in the adult rat, RST axons appro ach the rostral lesion edge but do not have the capacity to spontaneously re-grow through or caudal to the lesi on (Houle and Jin 2001), therefore numerous experimental strategies to specifically promote RST axonal growth have been conducted in animal models of spinal cord injury. Cough Reflex Spinal cord injured individuals sustain weak ening or paralysis of musculature vital to respiratory functions including th e cough reflex. Cough is a ball istic behavior that can be elicited experim entally with mechanical or ch emical stimulation. A cough is elicited via stimulation of central airway receptors that pr oject via vagal afferents to second order neurons (Bolser et al. 2000), which then project to populati ons in the brainstem i nvolved with respiratory control (Ezure et al. 199 1). Pre-motor excitatory input to spinal motoneurons originates from bulbospinal expiratory neurons in the nucleus retroambigualis in the medulla (Bolser et al. 2002). Cough is produced by a complex neural networ k within the brainstem. This network controls motoneurons supplying laryngeal, phrenic, in tercostal, and abdominal motor pools. These motoneuron pools are activated in a preci se and sequential manner to produce ballistic-
35 like expiratory pressures that peak within 200 ms from the end of the inspiratory phase in the cat and are characterized by intrathoracic pressures that can exceed 100 cmH20 (Bolser 2002). The expulsive component of this beha vior is generated by a coordina ted activation of chest wall and abdominal musculature to elicit a cough. It is this spatiotemporal coor dination that makes cough particularly sensitive to functi onal impairment by insults such as spinal injury. The principal expiratory muscles responsible for expiratory cough pressures ar e the anterolateral abdominals (rectus abdominis, transversus abdominis, ex ternal oblique and internal oblique). The motoneuron pools for these four expiratory abdomin al muscles terminate at L3, but they have varying rostral extents with the rectus abdominis extending most rostrally (T4); the external obliques to T6; the transverses abdominis to T9; a nd and the internal oblique to T13 (Miller et al. 1987) in the cat. A spinal pattern generator has not been iden tified for the cough reflex, but much like certain forms of locomotion, there is descending i nput for this behavior. Premotor drive to these expiratory motoneuron pools arises primarily from bulbospinal expi ratory neurons in the caudal part of the ventral respiratory column (cVRC), corresponding to th e caudal portion of the nucleus retroambigualis (NRA). It has been shown in the cat, (Merrill 1970) that the axons of the expiratory cVRC neurons immediately decussate in the medulla (Monteau and Hilaire 1991) between C1 and the obex (Arita et al. 1987; Miller et al. 1987; Miller et al. 1989) and descend into the contralateral ventral column of the la teral spinal cord (Merrill 1970; Merrill 1974; Richter et al. 1975; Merrill and Lipski 1987; Jiang and Lipski 1990; Kirkwood 1995). Antidromic mapping studies of the descending respir atory pathways have shown that expiratory axons, at the cervical level, form a very discre te tract that courses through the ventral white matter immediately below the base of the ventra l horn (Davis and Plum 1972; Merrill 1974).
36 These descending expiratory a xons become more dispersed th roughout the ventrolateral white matter with caudal progression thro ugh the thoracic cord (Merrill 1974). From T1-L3 expiratory axons arborize expansively throughout the contralatera l side of the spinal cord, with respect to the cell body, creating an extensive network spanning several spinal segments (Merrill 1974). This network of expiratory axons contri butes to both monosynaptic and multisynaptic drive to thoracic and lumbar expiratory motoneuron pools. Kirkwood and coworkers (Kirkwood et al. 1988) proposed that spinal interneurons have an im portant role in mediating this descending drive. Many of the expiratory-associ ated interneurons they identified had crossed axons and some descended at least five segm ents. Kirkwood suggested that these crossed interneuronal connections might mediate hete rogeneous functions including inhibition of inspiratory thoracic motoneurons du ring the expiratory phase of br eathing and excitation of chest wall motoneurons. It is reasonable that these in terneurons provide a mechanism by which drive from each cVRC is bilaterally represented in th e spinal cord. Recently similar propriospinal connections in the thoracic spinal cord have been implicated as an indirect pathway likely to mediate spontaneous recovery of basic loco motor function following spinal hemisection (Courtine et al. 2008). However, whether or not the cough motor system shows similar spontaneous recovery or retained function following partial lesions of the thoracic spinal cord has not been assessed.
37 CHAPTER 2 COUGH FOLLOWING LOW THORACIC SP INAL HEMISECTION IN THE CAT Introduction The expulsive compone nt of cough is generate d primarily by the coordinated activity of the anterolateral abdominal muscles (rectus abdomi nis, transversus abdominis, external oblique, and internal oblique). Input from the brains tem cough neural networks (Merrill 1970; Merrill 1974; Richter et al. 1975; Arita et al. 1987; Merrill and Lipski 1987; Miller et al. 1987; Miller et al. 1989; Jiang and Lipski 1990; Monteau and Hilaire 1991; Davis 1993; Kirkwood 1995) to the motoneuron pools for these primary expiratory muscles is disrupted following cervical and thoracic injuries involving the ventral and ventrolateral spinal cord (Davis and Plum 1972; Merrill 1974). In the cat the motoneuron pools for these four muscles all terminate at L3, but they have varying rostral extent s with the rectus abdominis ex tending most rostrally (T4); the external oblique to T6; the transverses abdominis to T9; and the internal oblique to T13 (Miller 1987). Further, from T1-L3 expiratory axons or iginating in the brainstem arborize expansively throughout the contralateral side of the spinal cord, creating an exte nsive network spanning several spinal segments (Merrill 1974). Due to the extensive arborization of descending motor axons we hypothesized that T9/10 lateral hemisec tion would not cause significant disruption of rectus abdominis EMG activ ity bilaterally or cough pressure generation. In the present study lateral T9/10 hemisections were made in adult cats. These lesions transect the descending brainstem expiratory pathways on one side of the spinal cord, disrupting pre-motor drive to the caudal, ipsilateral expira tory motoneuron pools. Cough pressure generation and rectus abdominis muscle activity were characterized preand post-injury. Expiratory muscle recordings were made from the rectus abdominis because it contributes to the generation of cough expulsive forces in th e cat (Tomori and Widdicombe 1969), plays a
38 significant role in increasing abdominal cavity pre ssure during cough (Bolser et al. 2000), and is easily accessed for repeated assessments. Our findings show that, despite considerable disruption of descending pre-motor drive from the brainstem to motoneuron pools of the primary expiratory muscles, the cough motor system show s substantial function following thoracic spinal cord injury (SCI). Methods Cough production was assessed preand postspinal T9/10 left hem isection in six specific-pathogen-free adult, spayed female cats (6-8 lbs, Liberty Laboratories, NY). Hemisections and post-op care were performed as described previously (Tester and Howland 2008). Surgeries were performed under isoflurane anesthesia. Buprenorphine (0.02mg/kg) was given TID for 48h, and bladders ex pressed manually for 1-5 days, post-SCI. Cats were housed on thick cushions in the AALAC accredited animal facility and trained on a variety of locomotor tasks (5x/week) for a parallel study. Animal proc edures were in accordance with the NIH Guide for the Care and Use of Laboratory Animals an d approved by the University of Floridas IACUC. Cough was assessed in isoflurane anestheti zed, spontaneously breat hing cats pre-injury and at 4, 13 and 21 wks post-SCI. Atropine sulfate (0.1 mg /kg, SQ) was given to block salivation and tracheal secretions. End-tidal CO2 was monitored and isoflurane levels adjusted to maintain this parameter within 4-5%. A sterile abdominal field was prepar ed and paired bipolar Teflon-coated stainless steel wire electrodes place d 2-3 mm apart in the left and right rectus abdominis muscles approximately 2 cm caudal to th e iliac crest and 1 cm lateral to midline. A ground electrode was placed in the left hamstri ng. An esophageal balloon catheter was placed at the midthoracic level and cough elicited by mech anical stimulation of the vocal folds and
39 epiglottis using an oral approach and a small le ngth of flexible plas tic tubing (Bolser 1991; Bolser et al. 1993). Esophageal pressure (Pes) and left a nd right rectus abdominis (LRA and RRA) electromyograms (EMGs) were recorded. Pes (cmH2O) was used as a measure of intrathoracic pressure generation. EMGs were amplified, rectified, band-pass filt ered (200-5000 Hz), and integrated (time constant 50 ms). Coughs were identified by behavioral observation of the animal and the presence of Pes amplitude larger than 5 cm H2O in response to the mechanical stimulus. Six parameters were calculated: Pes, percentage of LRA and RRA normalized cough amplitudes, esophageal rise times, and LRA and RRA rise times. To obtain RA EMG normalized amplitudes, amplitudes were obtained from moving averages, normalized to the largest burst at a given time point for each side in each cat, and expressed as a percentage of the largest burst. These normalized percentages were averaged for each cat at each time point. Rise times were determined as the elapsed time between 10% an d 90% of the total rise time of the moving average. Individual cough rise times for each cat were averaged at each time-point. Using SPSS software 14.0 (Chicago, IL), separa te repeated measures, within-subjects ANOVAs were conducted to determine if esophag eal parameters (pressure and rise time) differed across time points. Mixed (time x side) two-factor ANOVAs were conducted to determine if there was an effect of time or side on rectus abdomini s EMG rise times and normalized percent amplitudes. Post-hoc Fishers LSD tests were used to isolate any differences identified with ANOVAs. An level of 0.05 was used for all analyses. Between 4 and 6 months post-SCI, cats were deeply anesthetized (sodium pentobarbital, >35 mg/kg I.P.) and transcardially perfused with 0.9% saline (200400 mls) followed by 4% paraformaldehyde in 0.1M phosphate buffer (3.5L pH 7.4). Tissue was blocked, cryoprotected
40 in 30% sucrose-paraformaldehyde, and sectioned at 25um on a cryostat. Four lesion segments were cut coronally and two longitudinally. One section of every ten was mounted onto chrom alum and poly-L-lysine-coated slides (chromiu m potassium sulfate and poly-L-lysine, SigmaAldrich, St. Louis, MO; gelatin, Fisher Scientific, Hampton, NH) and processed with cresyl violet (cresyl violet with acetate, Sigma Al drich, St. Louis MO) and myelin (Eriochrome Cyanine R, Fluka, New York, NY) stains for basi c histology to determine the extent of injury following procedures detailed prev iously (Tester and Howland 2008). Results A total of 256 coughs from six cats w ere an alyzed across four time points: pre-injury (n=37); 4 weeks post-hemisection (wphx, n=87); 13 wphx (n=75); and 21 wphx (n=57). Cresyl violet and myelin stained serial sections of each lesion were assessed to determine the extent of SCIs. The lesions ranged from an under-hemisecti on with ipsilateral medial-ventral white matter sparing to a complete hemisection to an over-hem isection with disruption of contralateral gray and white matter (Figure 2-1). The expiratory pressures and bilatera l RA EMGs features assessed in this study were not influenced by these injury magnitude differences. All cats generated coughs preand post-injury under anesthesia in response to mechanical stimulation of the epiglottis and vocal folds. As assessed by EMGs, the RA muscles were normally silent during eupneic breathing. RA EMGs and Pes increased during mechanicallyelicited cough. Individual coughs as well as repetitive cough bout s, were frequently generated after injury (Figure 2-2). Pes and RA EMGs during coughing were similar to pre-injury recordings. The EMG patterns at all post-inju ry time points were typical of ballistic-like bursting observed in uninjured animals (Bolser et al. 2000) (Fi gure 2-2). Moreover, they were present bilaterally. Finally, the peak EM G activities of the LRA and RRA during cough
41 occurred simultaneously and were correlated with increases in Pes. Thus, qualitatively, these general cough characteristics appeared sim ilar to those observed prior to injury. Increases in Pes with cough averaged between 40 and 69 cm H2O at all time points and no significant change in the average Pes was seen across time points (p=0.410). In addition, during some individual coughs post-opera tively, Pes reached or exceeded 100cm H2O indicating that the injured system was capable of generating the substantial pressures sometimes seen preinjury, as well as in other reports of normal cats (Bolser et al. 2000). A significant effect of time on Pes rise time was found, F (3, 12) = 4.29, p = 0.028. However, post-hoc Fishers LSD tests did not reveal significant differe nces between any time points. Despite this, a notable prolonged Pes rise time was present at 13 weeks post-injury during some coughs. When this prolonged rise time occurred, it was manifested without a change in the magnitude of mechanical Pes (Figure 22). Average rise times were 0.089 (pre-SCI), 0.087 (4 wphx), 0.12 (13 wphx) and 0.082 seconds (21 wphx). Normalized EMG amplitudes during cough showed that both the LRA and RRA averaged between 60-72 and 61-79 percent of ma ximum respectively across time points. A twofactor ANOVA revealed no significant change in the average percent of maximum amplitude over time (p=0.67) or between the left and right RA muscles (p=0.587). As with LRA and RRA normalized amplitudes, no significant effects of tim e (p=0.183), side (0.136) or the interaction of time by side (0.690) were seen for EMG rise times. Rectus abdominis EMG activity was observe d during the inspiratory phase of coughing in most animals at all time points (Figure 2-2). This muscle activity was termed pre-expulsive because it occurred during the inspiratory phase and before the expulsive cough phase. There was no apparent change in pre-e xpulsive activity in the RA EMG at any post-injury time point
42 compared to pre-injury values and the duration of this pre-expu lsive activity was similar to the 600-700 ms range reported by Bolser a nd colleagues (Bolse r et al. 2000). Discussion The regenerative capacity of the central nervous system is limited (Steward et al. 2008). However, following incomplete spinal lesions in hum ans (Dietz et al 1998; Dobkin et al. 2007; Fawcett et al. 2007) and animals (Rossignol et al 1999; Weidner et al. 20 01; Bareyre et al. 2004; Courtine et al. 2005; Courtine et al. 2008) some locomotor recovery can occur without apparent intervention. Our studies s uggest that the cough motor sy stem shows similar endogenous recovery or preservation of function following a range of inco mplete thoracic lesions. Following injury, function may be mediated through indirect or bypass pathways. Premotor drive to the motoneuron pools of the f our primary expiratory muscles arises primarily from bulbospinal neurons in the caudal part of the ventral respiratory column (cVRC), corresponding to the caudal porti on of the nucleus retroambi gualis (Merrill 1970; Merrill 1974; Richter et al. 1975; Arita et al. 1987; Merrill and Lips ki 1987; Miller et al. 1987; Miller et al. 1989; Jiang and Lipski 1990; Monteau a nd Hilaire 1991; Davis 1993; Kirkwood 1995). Kirkwood and colleagues (Kirkwood et al. 1988; Schmid et al. 1993) have proposed that spinal interneurons play an important role in mediati ng this descending drive. Many of the expiratoryassociated thoracic interneurons they identified had crossed axons and spanned multiple segments. They suggested that these crosse d interneuronal connections might mediate heterogeneous functions includ ing inhibition of inspiratory thoracic motoneurons during the expiratory phase of breathing a nd excitation of chest wall mot oneurons. It is reasonable that these interneurons may provide a mechanism by which drive from each cVRC is bilaterally represented in the spinal cor d. This would enable expiratory cough muscles with motoneurons caudal and ipsilateral to a hemisection to rece ive cVRC input via multisynaptic connections.
43 Recently, contralaterally projecting cervical interneurons associat ed with phrenic (respiratory) motoneurons in the rat also have been identifi ed (Lane et al. 2008) sugge sting that respiratoryassociated interneurons are present at multiple spinal levels in the normal and injured spinal cord. The potential importance of interneurons fo r recovery also has been reported recently with respect to locomotor recovery (Courtine et al 2008). Following SCI in the mouse, thoracic interneurons were reported to mediate recovery of basic stepping in the relative absence of direct descending supraspinal connections to the spinal segments containi ng the hindlimb motoneurons. Although reports are mixed, the clin ical literature indicates th at individuals with thoracic spinal injuries generally maintain inspiratory capabilities but experien ce expiratory dysfunction due to partial paralysis (Hemi ngway et al. 1958) and atrophy (Dav is 1993; Kern et al. 2008) of the abdominal musculature. A variety of expi ratory muscle conditioning or training techniques have been reported to improve some expiratory functions following SCI in humans (for review see (Van Houtte et al. 2006)). These include high frequency magnetic stimulation, surface muscle stimulation, and electrical spinal cord stimulation (Jae ger et al. 1993; Linder 1993; Lin et al. 2001; Lim et al. 2007; Lee et al. 2008). Significant atrophy of the expiratory muscles along with decreased functional capacity also has been shown in the cat following a T6 complete spinal transection (Kowalski et al. 2007). Further, electrical stimulation of the spinal cord at T10 for 6 months following T6 transection appears to prevent the muscle atrophy and support greater expiratory function (DiMarco and Kowalski 2008). Although specific conditioning of the abdominal muscles with respect to cough or other respiratory functions was not done in the current study, all cats were extensively traine d on a variety of locomotor tasks 5x/week. Training involved treadmill walking as well as cr ossing of simple and challenging runways (for examples see (Tester and Howland 2008)). Su bstantial evidence suggests that locomotor
44 training can improve hindlimb/lower extremity motor functions post-SCI in animals and humans respectively (Edgerton et al. 2004; Thomas and Gorassini 2005; Behrman et al. 2006; Frigon and Rossignol 2006). Other studi es indicate that it improves abdomi nal expiratory muscle activity in intact subjects (Powers et al. 1992; Uribe et al. 1992; Halseth et al. 1995; Powers et al. 1997) and that locomotor and respiratory rhythms are coupl ed (Kawahara et al. 1989a; Kawahara et al. 1989b; Romaniuk et al. 1994). Further, voluntary exercise has been shown to increase the production of several neurotrophins (Gomez-Pini lla et al. 2002; Ying et al. 2003) which may play significant roles in motor recovery and synaptic plasti city post-SCI (Ying et al. 2008). Although the effects of locomotor training on co ugh specifically have not been tested, the combined literature suggests it may have contribu ted to the substantial cough function seen in the current study. It will be impor tant to test the effects of locomotor training on cough function following SCI in future studies. Despite complete disruption of the brainstem de scending expiratory proj ections to the left side of the spinal cord below T9, cats in th e current study showed s ubstantially normal cough function and abdominal muscle activity bilatera lly. Although only one of the four primary expiratory muscles was assessed (rectus abdominis ), its normal activity level alone would not be sufficient to generate the normal Pes seen. This suggests that the other three abdominal muscles also were contributing to force generation. These results suggest that the neural substrate(s) underlying the respiratory defense mechanism of cough has substantial capa city for plasticity and/or preservation of function in the abdominal motor system following spinal hemisection.
45 Figure 2-1. Range of lesion magnitudes. The smallest lesion was an under-hemisection with ipsilateral ventro-medial white matter sparing (A). In a complete-hemisection the ipsilateral gray and white matter are damaged and the contralateral gray and white matter spared (B). The largest lesion was an over-hemisection with contralateral gray and white matter damage (C). The size bar is the same for A and B. Dorsal (D), ventral (V), rostral (R), and caudal (C) are indica ted for orientation.
46 Figure 2-2. Cough characteristics. Representa tive filtered and moving average integrated electromyograms of the right rectus abdominis and left rectus abdominis with corresponding esophage al pressures (cmH20) during cough in a cat 13 weeks posthemisection (A). The single cough, as well as the subsequent bout of coughing, shows robust esophageal pressures and EMG activities bilaterally which are similar to those seen in normal cats. Representative filtered RA EMGs show the pre-expulsive behavior seen during some coughs at all post-SCI timepoints during the inspiratory phase (B). The duration of pre-expulsiv e EMG activity was similar to previously reported studies. A prolonged Pes rise time was seen in some coughs at 13wphx compared to other time points (C). When this prolonged rise time occurred, it was not accompanied by an increase in Pes.
47 CHAPTER 3 EFFECTS OF THORACIC SPINAL HEMI SECTION AND CHONDROITINAS E ABC TREATMENT ON BASIC LOCO MOTION AND THE COUGH REFLEX IN THE CAT Introduction Following traum atic spinal injury, axons in the spinal cord have a minimal capacity for axonal regeneration due to the presence of the glia l scar, de-myelination, and the up-regulation of potent inhibitory molecules such as chondroitin sulfate proteoglycans (CSPGs) (Schwab and Bartholdi 1996). Despite this inhibitory environment, signifi cant spontaneous recovery of locomotor function following spinal lesions has been described in the mouse (Steward et al. 2008), rat (Li et al. 1994; Fouad et al. 2001b; Wei dner et al. 2001; Bareyr e et al. 2004; Courtine et al. 2008), cat (Murray and Go ldberger 1974; Bregman and Gol dberger 1983; Kato et al. 1984; Eidelberg et al. 1986; Alsterma rk et al. 1987; Barbeau and Ro ssignol 1987; Pettersson et al. 1997; Rossignol et al. 1999), monkey (Lawrence and Kuypers 1968; Courtine et al. 2005), and human (Dietz et al. 1998; Dobkin et al. 2007; Fawcett et al. 2007). Reinnervation of denervated areas may occur via growth of lesioned fibers a nd/or sprouting from intact fiber systems. Tapping into these plastic systems via comb inatorial therapeutic and rehabilitative interventions may further induce axonal sprouting and regeneration, and therefore increase functional motor recovery. Degradation of the inhibitory chondroitin sulfate glycosaminoglycan (CS-GAG) side chains of CSPGs with the bact erial enzyme Chondroitinase ABC (Chase ABC) has been shown to enhance axonal growth and/or behavioral recovery in numerous rodent models of SCI (Yick et al. 2000; Moon et al. 2001; Bradbury et al. 2002; Yick et al. 2003; Chau et al. 2004; Caggiano et al. 2005; Barritt et al. 2006; Houle et al. 2006; Massey et al. 2006; Cafferty et al. 2007; Carter et al. 2008; Iseda et al. 2008; Massey et al. 2008) as well as in our cat model (Tester and Howland 2008).
48 Previous research from our laboratory followi ng a low thoracic spinal hemisection in the adult cat, illustrated that CS-G AG degradation with Chase ABC enhanced qualitative recovery of bipedal and overground locomotion (Tester and Howland 2008). In the present study, we quantitatively examined the effects of low thoracic spinal hemisection on numerous aspects of the step cycle and ipsilateral hindlimb steppi ng ability during bipedal treadmill and overground locomotion in adult cats. Quantitative assessments of ipsilateral hindlimb movements before and after injury at multiple timepoints allowed us to examine, in detail, locomotor performance deficits as well as the degree of spontaneous reco very following thoracic he misection in the adult cat during two distinct locomotor tasks; bipedal treadmill and overground locomotion. We then examined whether delivery of Chase ABC would enhance locomotor recovery on these tasks in our model. Due to the significant spontaneous recovery in our cat model following T9/10 hemisection during basic locomotion, we hypothesi zed that Chase ABC would have variable effects on the specific temporal components of the gait cycle as well as the cough reflex. Our results show that there is significant preservation of many te mporal aspects of the locomotor cycle following low thoracic spinal hemisection in the adult cat during bipedal treadmill and overground locomotion. Despite this resiliency, many features sustain persistent, quantifiable deficits. Aspects of the step cycl e and of ipsilateral hindl imb function that were affected by low thoracic hemisection, across the two tasks assessed, were generally not improved with intraspinal Chase ABC treatment. Methods Animals Twelve adult SPF fem ale cats (6-8 lbs) were used in this stu dy. All cats were spayed to ensure that behavioral data collection was not influenced by postural changes induced by hormone alteration during estrus (Sribnick et al 2003). All animal procedures were conducted in
49 accordance with the NIH guidelines for the care and use of experimental animals and were approved by the University of Floridas Inst itutional Animal Care and Use Committee. Surgical Procedures Spinal T9/10 hemi sections were performed as previously published (Tester and Howland 2008). Penicillin G, procaine (40,000U/kg BW IM) was given the day before, the day of, and the day after surgery. Prior to all surgeries, cats received 0.1cc atropine sulfate (0.04-0.06 mg/kg SQ) and 0.1cc acetylpromazine (0.4-0.5 mg/kg SQ). Following orotracheal in tubation, anesthesia was maintained with isoflurane (1-3.5%) and an IV was placed for fluid administration (Lactated Ringers, 10 ml/kg/h). Respiratory rate, expired CO2, SPO2, blood pressure, body temperature, as well as the general plane of anes thesia were closely monitored. A left lateral hemisection was made using ir idectomy scissors at sp inal T9 or T10. Light suction with a pulled glass pipette was used if nece ssary to lift any fibers adhering to the dura to facilitate the execution of a complete hemisec tion without compromising the integrity of the dura. Thrombin and gel foam were used to st op any bleeding. Micro-im plantable infusion ports (Harvard Apparatus, Holliston, MA ) were placed sub-cutaneously and sutured to muscle lateral to the vertebral column. Port tubing was held in place by suturing it to the muscle at several points until it reached midline. VetBond (Webst er Veterinary Supply, In c., Sterling, MA) then was used to secure it to the lamina caudal to the laminectomy. The end of the port tubing was secured in the lesion cavity by suturing the dura (8 -0 Prolene). Prior to dural suturing, the port reservoir and tubing (~5l total volume) was filled with proteas e free Chase ABC (1 U/200 l saline, pH 7.8, Seikagaku Corp., Tokyo, Japan). Injury-only animals did not receive port placement or delivery of any fluid into the lesi on area. Durafilm (Codman-Shurtleff, Inc., Randolph, MA) and gelfoam (Pharmacia and Upj ohn, Inc., Peapack, NJ) were placed over the dural sutures and the muscle and skin sutured. Anesthesia was terminated, cats were extubated,
50 and then placed in temperature controlled r ecovery chambers (ThermoCare, Las Vegas). Buprenorphine (0.01 mg/kg SQ) was ad ministered TID for 48 hours. Procedures used to maintain the general health of the cats were similar to those described in our previous studies (Howland et al. 1995b; Howland et al. 1995a). Once body temperature stabilized to a minimum of 100 F, cats return ed to their home cages with 5-7 inch egg crate foam cushions covering the entire cage floor to prevent peripheral nerve compression, pressure sores, and skin breakdown. None of the animal s developed any of these complications. For the first few days post-op bladders were expressed manually using Credes method. Animal health was continuously monitored throughout the stu dy, including maintenan ce of food intake and body weight. Chondroitinase ABC Delivery Cats were placed into tw o groups following injury: injury-only (n=6) and injury+Chase ABC (n=6). Commercial Cha se ABC may not retain its st ability at body temperature for extended periods (Tester et al. 2007), therefore 50 microliters of prot ease-free Chase ABC (1 U/200l), or vehicle were injected every other day for 1 month. The concentration delivered was identical to our previous study (Tester and Ho wland 2008). The volume is ~4x that used in rodent studies (Bradbury et al. 2002) due to the larg er size of the cat spinal cord. For treatment administration, the cats were anesthetized (1-3% is ofluorane and 1.5LO2) for 15 minutes while the Chase ABC was delivered (0.14L/min) us ing a microinjection syringe pump (Harvard Apparatus, Holliston, MA). General Locomotor Training Cats were trained daily (5 times/week) on a variety of basic and skilled locom otor tasks requiring different levels of input from the neur al axis including bipedal treadmill, 12 and 2 overground, obstacle, and pegboard locomotion. For the current studies, we focused specifically
51 on effects on multiple components of the gait cycle during bipedal treadmill and 12 overground locomotion following thoracic hemisec tion and Chase ABC treatment. All locomotor tasks were conditioned to a f ood reward. When all tasks were performed consistently, baseline data was collected pre-inju ry. All cats then received a left spinal T10 hemisection, and locomotor traini ng was re-initiated within 24 hours and continued daily for the remainder of the study. A left T10 hemisecti on primarily affects loco motor function of the ipsilateral, left hindlimb (LHL). Therefore, quantitative assessments were done on the LHL. Bipedal Treadmill and Overground Locomotion Bipedal loco motion was performed on a motor driven treadmill 5 times/week at 0.5m/s for a food reward. Immediately following spinal hemi section cats could not wa lk at this speed, but all recovered the ability to consistently walk at 0.5m/s by the first behavioral assessment time point of 2wphx. Cat forelimbs were placed on a stationary platform and their hindlimbs were allowed to freely move over the treadmill belt wh ile receiving a liquid food reward in a raised bowl in front of them. Ten consecutive step cycles at each time point/cat were used and averaged within Peak Motus to 51 frames in length each for quantitative assessments. Cats were also trained and assessed during overground locomotion (12 wide runway). Cats crossed the overground runway for a food rewa rd given at each end. The comfortable fast walk/slow trot speed was determined pre-injury for each individual cat, and post-injury crossings were carefully chosen for each cat to closely ma tch their pre-injury crossing times. Therefore, any quantitative locomotor changes cannot be a ttributed to speed ch anges within animals following spinal hemisection. Ten contiguous step cycles from two step cycles at each time point/cat were used and averaged within Peak Motus to 51 frames in length each for quantitative assessments and then averaged across cats.
52 Step Cycle, Swing, and Stance Durations The left hindlimb step cycle duration was defined from paw contact with the treadmill belt or overground runway until the subsequent paw c ontact. Step cycle duration was based on the raw number of frames the LHL spent in an en tire gait cycle. The gait cycle was further delineated into its two main phases: stance, wh ich begins with initial paw contact, and swing, which begins when the paw leaves the contact su rface. Swing and stance were calculated as a percentage of the entire duration of the step cycle for both behavioral tasks. Ten step cycles for each cat were assessed to define the percentage of the step cycle that was spent in swing and stance, and then these percentages were averaged across all 10 step cycles per cat, giving an average percentage of the step cycle for swing and stance per each cat at each timepoint. These averages per cat where then averaged across cats to give a final percentage of the step cycle spent in swing and stance. Paw Drag and Kinematic Locomotor Assessment Paw drag was assessed duri ng bipedal and overground locomotion for each cat. Paw drag was def ined as the duration of frames the paw was in dorsal contact with the treadmill belt or overground surface during the swing pha se of the step cycle. Paw drag was assessed during ten step cycles and averaged for each cat during bipedal and overground locomotion and subsequently averaged across cats within ea ch group for statistical comparisons. The cats performances were recorded at multiple timepoints until 20 weeks post-injury. The cats hindquarters were shaved and reflective spheres were placed on the iliac crest, greater trochanter, lateral malleolus and the base of the 5th metatarsal. A fifth marker was placed on the fibula approximately one inch above the lateral malleolus marker in order to generate a unit vector, which in combination with the measured le ngth from the lateral malleolus to the head of the lateral condyle, created the knee position. These spheres allowe d for the tracking of the hip,
53 knee, and ankle joints across the step cycle. Locomotor kinematics were assessed using Motus software (Vicon Peak). The maximum hip and ankl e angle at the transition from stance to swing was found during ten step cycles for each cat at e ach timepoint. The angles during the ten step cycles were averaged within a cat and then ac ross cats at each timepoint. The minimum knee angle during swing was also found during ten st ep cycles for each cat at each timepoint, averaged within a cat, and then across cats at each timepoint. Cough Stimulation and Assessment For this portion of the study 14 cats were assessed: Chase A BC treated (n=8), and control (n=6). Cats were initially given atropine sulfate (0.1 mg/kg) s ubcutaneously to block salavation and ref lex tracheal secretions. Pre-injury and at multiple post-operative timepoints, spontaneously breathing cats were gaseously anesthetized (2-3% isofluorane in 1.5 LO2). Endtidal CO2 was monitored during the procedures a nd anesthesia was adjusted accordingly. Cats were placed in the supine pos ition and the abdomen was shaved and sterilized. Pa ired bipolar Teflon-coated stainless steel wi re electrodes were placed bilate rally approximately 2-3 mm apart in the left and right rectus a bdominis muscles below the level of the spinal hemisection and approximately one cm lateral to the midline in the mid-pelvic region. A ground electrode was also placed in the left hamstring muscle. An esophageal balloon catheter was placed into the esophagus and inflated with a syringe. Cough was elicited by mechanical stimulation of the vocal folds and epiglottis with a small length of flexible plastic tubing. Cough analysis was dependent on a clear bilateral RA EMG signal and cough amplitude larger than 5cm H20. All coughs generated at each data co llection point were used if they met these criteria. Esophageal pressure (PES) and left and right rectus abdominis (LRA and RRA) electromyograms (EMGs) were recorded. PES (cmH20), LRA/RRA cough amplitudes (normalized percentages), esophageal rise times (seconds), and rectus abdominis rise times
54 (seconds) were calculated. EMG amplitudes per cat were normalized to the largest EMG burst at a given time point on the same side, expressed as percentages, and then averaged across animals at each time-point. Rise times were calculated by subtracting 10 % of the total rise time from the peak and the baseline in order to standardize th e data. Cough rise times and amplitudes were measured pre-injury, 4weeks post hemis ection (wphx), 13wphx, and 21wphx. Individual cough rise times for each cat were averaged at each tim e-point and then averaged across animals. Histological Confirmation of T10 Hemisection Befo re tissue was cut on the cryostat it was placed in a solution containing 30% sucrose and 70% fixative overnight. The dura was removed, wh ile maintaining the spinal cord moist, and then immediately placed into dry ice to freeze. Tissue was placed in mounting media and cut with a tissue thickness of 25 Sections were cut into tubes fi lled with 0.1M PBS. The tissue was then organized by cutting into series of 10. On e section of every ten was mounted onto subbed slides and stained with cresyl vi olet (cresyl violet with acetate Sigma) and myelin (Eriochrome Cyanine R, Fluka, New York) for assessment of lesion extent and damage. Statistical Analyses Statistical Analyses were perform ed for ch aracterists of locomotion using Statistical Package for the Social Sciences (SPSS) v. 17 (C hicago, IL). To assess changes over time Mixed 2 factor ANOVAs were used and the Huynh-Feldt corr ection was used if sphericity could not be assumed. Independent t-tests were used to a ssess individual changes across two timepoints and across groups at one particular timepoint. Paired two-tailed t-tests were also done to compare individual changes across timepoint s within the same group. Post hoc analyses were done using Bonferonni corrections. For char acteristics of cough, Graphpad In stat softwar was used. To assess changes over time, repeated measures ANOVA using Bonferonnie post hoc analyses were used.
55 Results Extent of Spinal Hemisection Spinal T9/T10 left hem isections were performe d on all cats. A representative spinal cord section depicting the greatest cross-sectional exte nt of lesion damage for each animal is shown in Figure 3-1. The size bar paired with cat lesion A1 also corres ponds to every lesion without a matching size demarcation. Cat lesions A4 and B1 were sized accordingly in order to match lesion cross-sectional area across cats for visual clarity. Lesion epicenter representations were produced from light microscope ex amination of cresyl violet a nd myelin stained spinal cord cross sections at the level of the greatest ex tent of spinal damage. Cat lesion A1 was cut longitudinally and the cross-sectional diagram wa s created by examination of serial sections through the dorso-ventral extent of the lesion and subsequent s uperimposition over cat lesion A6 as a template. In all animals, the ipsilateral lateral funiculus was entirely damaged, eliminating the corticospinal and rubrospinal tracts. The ipsilateral dorsal funiculus also was completely affected in all cats. The exte nt of damage incurred to the ip silateral ventral funiculus ranged from complete to minimal damage or complete sparing. A range of damage to the contralateral gray matter, dorsal funiculus, and ventral funiculus occurred acr oss all cats. Locomotor Recovery Behavioral recovery during bipedal treadm ill and overground locom otion was assessed for lesion only (no port placement) and lesion + Chase ABC (port placement with Chaase ABC diluted in saline). For all animals assessed, regardless of treatment paradigm, the ability to perform bipedal locomotion independently at 0.5 m/s with integration of the left hindlimb occurred within the first week post-injury, and of ten times occurred within the first few days. Similarly, this rapid recovery was also seen during overground locomotion where all animals could independently perform the task with left hi ndlimb integration within one week post-injury.
56 Across all lesion only and lesion + Chase ABC cats the ability to place the affected left hindlimb primarily utilizing plantigrade paw pos ition, versus dorsal paw placement, occurred within two to three weeks post-injury. This behavioral timeline is consistent with previous research utilizing the hemisection model (Helgr en and Goldberger 1993; Kuhtz-Buschbeck et al. 1996; Tester and Howland 2007). Recovery of Characteristics of Basic Locomotion The left hemisection lesion primarily affects the ipsilateral, left hindlimb and therefore all analyses were assessed on this limb. All twelve cats acquired the ability to walk independently during bipedal locomotion and overground locomo tion by 2 weeks post-hemisection (wphx) and therefore this time point was used as th e first post-injury assessment time. Step-cycle duration The step cycle duration and its variab ility (average SEM) at pre-injury and post-injury time points for lesion-only and lesion + Chase ABC cats during bipedal treadmill and overground locomotion are illustrated in Figure 3-2. The step cycle duration during bipedal treadmill locomotion was very consistent at every post-injury timepoint compared to pre-injury, with no significant effects acr oss time (p=0.571), treatment (p=0.794), or time x treatment (p=0.405) (Mixed 2 factor ANOVA). During overground locomotion, there was a significant effect of time (p=0.02) on step cycle duration show ing an increase in duration after injury. Post hoc comparisons isolated the differences betw een pre-injury and 8w phx (p=0.041) and between pre-injury and 20wphx (p=0.036), su ggesting a sustained injury e ffect on step cycle duration during overground locomotion. There was also a tr end towards an effect of treatment (p=0.058), with these animals also showing a trend toward s increased step cycle duration at post-injury timepoints.
57 Stance duration The percen tage of the step cycle spent in the stance phase for injury-only and Chase ABC treated cats during bipedal treadmill and overground locomotion can be seen in Figure 3-3. There was a significant effect of time (p= 0.000) on stance duration during bipedal treadmill locomotion, showing an overall decrease in the pe rcentage of time spent in stance at all postinjury timepoints. Post hoc comparisons isolat ed the differences betw een pre-injury and 2wphx (p=0.040), pre-injury and 8wphx (p=0.046), an d between pre-injury and 20wphx (p=0.031), illustrating a persistent injury effect on stance du ration. Stance duration as a percentage of the step cycle was not significantly effected acro ss time x treatment (p= 0.0642) or treatment alone (p=0.941). There was also a significant effect of time (p=0.002) on stance duration during overground locomotion. Post hoc comparisons isol ated an increase in stance duration between pre-injury and 8wphx (p=0.002) and between pre-injury and 20wphx (p=0.012), suggestive of a delayed injury effect during ove rground locomotion on stance durati on. Interestingly, there was a significant increase in stance duration post-injury during overgr ound locomotion, whereas during bipedal locomotion there was an overall decrease in stance post-injury. Swing duration The percentage of the step cycle spent in the swing phase for in jury-only and Chase ABC treated cats during bipedal treadmill and overground locomotion can be seen in Figure 3-3. There was a significant effect of time (p= 0.000) on swing duration during bipedal treadmill locomotion, showing an overall increase in the pe rcentage of time spent in swing at all postinjury timepoints. Post hoc comparisons isolated the differences to be between pre-injury and each post-injury timepoint (p=0.04, p=0.046, a nd p=0.031 respectively), demonstrating a persistent injury effect on swing duration during bipedal treadmill. Swing duration as a
58 percentage of the step cycle was not significantly effected acr oss time x treatment (p=0.642) or treatment alone (p=0.941). There was also a significant effect of time (p=0.007, Huynh-Feldt correction was applied for violations of the assumption of spheric ity), of the swing du ration during overground locomotion, showing an overall decrease in the pe rcentage of time spent in swing post-injury. Post hoc comparisons isolated the differences between pre-injury and 8wphx (p=0.002) and between pre-injury and 20wphx (p=0.012), sugges tive of a delayed injury effect during overground locomotion on swing duration. Paw drag Following spinal hemi section the occurrenc e of paw drag during the swing phase significantly increased during both bipedal trea dmill and overground locomotion (Figure 3-5). Stick figure diagrams from a representative animal during bipedal treadmill and overground locomotion at pre-injury, 2wphx, and 20wphx are shown in Figure 3-4. During the pre-injury swing phase, the paw is lifted off the treadmill be lt or overground runway in a very precise and efficient manner. Immediately following injury dur ing both behavioral task s, the paw is dragged dorsally during the majority of the swing phase By 20wphx, during both tasks, cats are now able to lift the paw off the treadmill belt or ove rground runway, but not as efficiently as preinjury, with visible knee hyperflexion occurring during both tasks. During bipedal treadmill there was a significan t effect of time (p=0.000) on paw drag, showing an increase at all postinjury timepoints. Post hoc comparisons were found between pre-injury and 2wphx (p=0.000), 4wphx (p= 0.001), 8wphx (p=0.011), 16wphx (p=0.006) and between 20wphx (0.038). Following spinal hemisection, paw drag during bipedal treadmill locomotion does not return back to pre-injury va lues even by 4-5 months post-injury. There
59 were no effects seen across time x treatment (p= 0.116) or treatment alone (p=0.899) on paw drag during bipedal treadmill locomotion (Figure 3-5). During overground locomotion there also wa s a significant effect of time (p=0.000, Huynh-Feldt correction was applied for violations of the assumption of s phericity) on paw drag, but post hoc comparisons found differences only between pre-injury and 2wphx (0.027) and preinjury and 4wphx (0.442). Paw dr ag had normalized back to values similar to pre-injury sometime between one and two months post-injur y. Therefore paw drag significantly improved following injury during overground locomotion much faster than during bipedal locomotion. This suggests that input from supraspinal cent ers can significantly improve paw drag following thoracic hemisection. There were also no eff ects seen across time x treatment (p=0.700) or treatment alone (p=0.460) on paw drag during overground locomotion. Chase ABC treatment does not appear to improve paw drag deficits following thoracic hemisection (Figure 3-5). Left hindlimb angular kinematics Angular kinema tics were assessed at the hip, knee, and ankle across a ll cats to determine how the specific joint angles were affected by th oracic hemisection in the cat model. Maximum hip angle was quantitatively assesse d at the transition from stance to swing during pre-injury and at 20wphx during bipedal treadmill and overground locomotion. Two-tailed independent t-tests revealed that there was no injury effect on maximal hip extension during bipedal treadmill (p=0.773) or overground locomotion (p=0.932). Injury-only pre-injury hi p maximal extension angles were not significantly di fferent from Chase ABC treated pre-injury values during bipedal treadmill (p=0.169) or overground locomotion (p=0.522), and the same was true when comparing their 20wphx values (p=0.079, p=0.281 resp ectively). In the Chase ABC treated cats there was not an injury effect. Despite this, when comparing the absolute degree change of hip maximal extension at 20wphx compared to pre-injury values, there was no significant difference
60 between injury-only and Chase ABC treated cats during bipedal treadmill (p=0.062) or overground locomotion (p=0.405). Therefore, this sp ecific characteristic of the gait cycle is stable following thoracic hemisection and is not affected by Chase ABC treatment. Minimal knee angle during swing, reflective of maximal knee flexion during swing, was also quantitatively assessed during pre-inju ry and at 20wphx during bipedal treadmill and overground locomotion. Two-tailed independependent t-tests revealed there was no injury effect on maximum knee flexion during bipedal treadmill (p=0.101), but there was a significant effect of injury on this parameter during overground lo comotion (p=0.048) (Figur e 3-6). Injury-only pre-injury maximal knee flexion angles were not significantly different fr om Chase ABC treated pre-injury values during bipedal treadmill (p =0.350) or overground locomotion (p=0.687), and the same was true when comparing their 20wphx values (p=0.944, p=0.802 respectively). It can therefore be stated that Chase ABC treated cats also sustained an injury effect of maximal knee flexion during swing during overground locomotion but not during bipedal treadmill. When comparing the absolute degr ee change of maximal knee fl exion during swing at 20wphx compared to pre-injury values, there was no significant difference between injury-only and Chase ABC treated cats during bipedal treadmill (p=0.732) or overground locomotion (p=0.207). Therefore, maximal kn ee flexion during swing is affected by thoracic hemisection during overground locomotion but not during bipe dal locomotion, and there are no apparent effects of Chase ABC treatment based on the analyses assessed. Finally, we quantitatively asse ssed the maximum ankle angle at the transition from stance to swing during pre-injury and at 20wphx during the same two simple behavioral tasks. Statistical tests revealed that there was no injury effect on maximal ankle extension during bipedal treadmill (p=0.632) or overground locomoti on (p=0.623). Injury-only pre-injury ankle
61 maximal extension angles were not significantly different from Chase ABC treated pre-injury values during bipedal treadmill (p=0.917) or overground locomotion (p=0.549), and the same was true when comparing their 20wphx values (p=0.622, p=0.164 respectively). Also, in the Chase ABC treated group there was also not an injury effect on maximum ankle extension during the step cycle. When comparing the abso lute degree change of maximal ankle extension during the step cycle at 20wphx to pre-injury values, there was no significant difference between injury-only and Chase ABC treated cats dur ing bipedal treadmill (p=0.425) or overground locomotion (p=0.801). Therefore, maximal ankle extension at the transition from stance to swing is not affected by thoracic hemisec tion during overground or bipedal treadmill locomotion, and there are not apparent effects of Chase ABC treatment on this parameter based on the analyses assessed. Recovery of the Cough Reflex Our laboratory has previously found that follo wing a low thoracic hemi section, the general characteristics of the cough reflex are preserved, including left and right rectus abdominis electromyogram (EMG) amplitudes and rise times as well as esophageal pressure amplitudes and rise times (Jefferson S 2008). Following Ch ase ABC treatment, there were no significant changes from pre-injury of the esophageal pressu re rise times (p=0.1750), left rectus abdominis rise times (p=0.1508), right rectus abdominis rise times (p=0.5604), left rectus abdominis amplitudes (p=0.4278), or in right rectus abdomin is amplitudes (p=0.4372). However, with Chase ABC treatement, esophageal pressure amplitu des at all post-injury timepoints tested were significantly increased compared to pre-injury values (p=0.0381) (Figure 3-7). Discussion This study assessed the effect s of thoracic hem isection in the cat on specific components of the gait cycle during a task that requires no descending input, bipedal treadmill, and a task that
62 integrates descending input with spinal pattern generation, overground locomotion. We also assessed the effects of Chase ABC treatment on the cough reflex. Following complete thoracic spinal transection, cats maintain coordinated hi ndlimb walking on a treadmill. This demonstrates that intraspinal networks in the caudal spinal cord, in isolation from descending supraspinal inputs, can generate rhythmic hindlimb moto r behaviors (Grillner and Zangger 1979; Barbeau and Rossignol 1987; Howland et al. 1995a; Howland et al. 1 995b; Rossignol et al. 1996). However, one persistent deficit that remains is paw drag during swing (Belanger et al. 1996). Similarly, overground locomotion requires input from the caudal cord, but also depends on descending supraspinal i nput for completion. Overall, the characteristics assessed during bipedal and treadmill locomotion were affected by our injury paradigm but were not affected by Chase ABC administration. There is a tremendous amount of intrinsic plas ticity in the cat motor system in response to thoracic spinal hemisection, but there are still pers istent, quantifiable deficits. Our results show that step cycle duration dur ing bipedal treadmill is very stable following thoracic hemisection in the a dult cat, whereas during overground locomotion step cycle duration is significantly increased up until four or five months post-hemisection. This could be due to slight decrease in speed during overground loco motion at post-injury timepoints. Overground crossing speeds were closely matched within cats to their pre-injury values, but small changes in time could affect this parameter. It has been well documented that locomotor cycle durations can be affected by descending, as well as sensory, in put to the spinal cord (Armstrong 1988; McCrea 2001). During normal overground gait, an increa se in speed of locomotion is usually accompanied by a decrease in step cycle duratio n, mostly due to the shortening of the stance phase (Murray 1967; Goslow et al 1973), or inversely a decrease in locomotor speed would be
63 accompanied by an increase in step cycle duration a nd a lengthening of the stance phase. We see this exact effect following injury during overground locomotion; the st ep cycle duration was seen to increase with a correlative increase in stance duration. The complete spinal animal maintains consistent relationships between th e step cycle duration a nd the swing and stance duration (Barbeau and Rossignol 1987; Belanger et al. 1996), which was also generally seen in our injury model. Following thoracic lesion of the lateral a nd dorsal descending pathways cats exhibit substantial hindpaw drag during swing, an in creased HL cycle duration, a disruption of stance/swing transition, and alte ration in intralimb coupling, a nd a severe intralimb uncoupling during quadrupedal treadmill locomotion (Courtine et al 2005). In all cats assessed in this study, the most obvious locomotor deficit was the subs tantial swing phase paw drag during bipedal treadmill locomotion as well as overground locomoti on following spinal hemisection. Paw drag during overground locomotion returned to normative values within one month of injury, whereas there was significant increases in paw drag during bipedal locomoti on at all post-injury timepoints. Integration of descending systems such as the corticoand rubrospinal systems during overground locomotion may have contribute d to the diminished paw drag following thoracic hemisection. During bipedal treadm ill, a task that can be accomplished without descending control the central patt ern generator alone was unable to compensate for the thoracic injury and therefore paw drag neve r returned to normative values post-injury. Studies in the cat have shown that locomotor trai ning can significantly enhance behavioral recovery following spinal cord injury (Barbeau and Rossignol 1987) and likely upregulates a variety of growth factors that are likely to enhan ce plasticity within the spinal co rd (Gomez-Pinilla et al. 2002).
64 Therefore, we may be getting a training effect during the paramete rs assessed following injuryonly because our animals are extensively trained daily on a variety of locomotor tasks. In our injury model, the hip, knee, and ankle angles post-inju ry were generally conserved, except that there wss an injury -effect of time on the maximum knee flexion during swing during overground locomotion. Injury-only animals had a significantly increas ed knee flexion during swing correlated with a visible hype rmetric gait that was not as apparent during bipedal treadmill locomotion. Chase ABC did not affect th is result positively or negatively. Although basic voluntary forms of locomoti on require supraspinal input to spinal networks, it is the skilled task s requiring greater balance and c ontrol of limb trajectory that demand the greatest supraspinal contributions. Sp inal hemisection of th e thoracic cord did not prevent the cats from performing bipedal treadm ill or overground locomotion as early as a few days post-lesion, although previously we have shown that this lesion significantly delays the ability to complete skilled locomotor tasks (Tester and Howland 2008). Skilled tasks are the most affected by our lesion paradigm, and therefore it is logical that Cha se ABC would be more beneficial during locomotor tasks that are gr eater affected by our lesion paradigm. Chase ABC administration did not affect any locomotor features as sessed in this study, however Chase ABC treatment significantly incr eased esophageal pressures at all post-injury timepoints compared to pre-injury values and all injury-only values. This may not be relevant in our current injury model, but in a cervical lesi on model where the respirat ory deficits are greater, this increase in esophageal pressure may allow for a more efficient cough production.
65 Figure 3-1. Cross section lesion exte nts. Pictorial representations of the greatest extent of crosssectional lesion damage for all lesiononly (A1-A6) and Chase ABC treated cats (B1-B6). All images were generated with in Adobe Photoshop from light microscope photomicrographs of cresyl viol et and myelin stained spinal cord cross-sections at the lesion epicenter. White; intact white matter undamaged by the spinal lesion; Black, severely or moderately damaged grey and wh ite matter as well as gliotic scar tissue; Gray, intact grey matter.
66 Figure 3-2. Step cycle duration during bipedal treadmill and overground locomotion. Step cycle durations of the left hindlimb represented as raw frames are shown for injury-only and Chase ABC treated cats at pre-inju ry, 2, 8, and 20 weeks post-injury during bipedal treadmill and overground locomotion. Step cycle durations of the left hindlimbs of injury only and Chase A BC treated cats are not significantly altered following injury at any post-injury timepoints. A delayed increase in the step cycle duration during overground locomotion is seen at 8 and 20 weeks post hemisection that is not significantly altere d with Chase ABC treatment. Data represent averages SEM. Asterisks indicate significant ch anges compared to pr e-injury values. Indicates a significant change from preinjury based on post hoc analysis.
67 Figure 3-3. Swing and stance durations during bipedal treadmill and overground locomotion. Stance and swing durations of the left hind limb represented as the percentage of the entire step cycle are shown for injury-only and Chase ABC treated cats at pre-injury, 2, 8, and 20 weeks post-injury during bipe dal treadmill and overground locomotion. Stance duration significantly decreases at all post-injury time points during bipedal treadmill locomotion and swing has a correlative increase post-injury. No treatment effects were seen. Swing duration signif icantly increases at 8 and 20 weeks postinjury during bipedal treadmill locomotion and swing has a correlative decrease at these post-injury timepoints. No trea tment effects were seen during overground locomotion as well. Data represent averag es SEM. Asterisks indicate significant changes compared to pre-injury values. Indicates a significant change from preinjury based on post hoc analysis.
68 Figure 3-4. Stick figure represen tations of paw drag during swi ng. Representative stick figure diagrams of ipsilateral, left hindlimb ki nematic movements during the swing phase on bipedal treadmill at 0.5 m/s (A) and overg round locomotion (B) at pre-injury, 2 weeks post-injury, and 20 weeks post-injury in cat B1. The gray shaded region illustrates when the ipsilateral hindpaw was not dragging along either the treadmill belt or overground runway during the swi ng phase. The direction of movement moves from right to left in th e direction of the arrow.
69 Figure 3-5. Paw drag during bipedal treadmill an d overground locomotion. Paw drag during the swing phase (frames SEM) is shown fo r injury-only and Chase ABC treated animals during bipedal treadmill (A) and ove rground (B) locomotion. Individual data for each injury only cat (C) and each Cha se ABC treated cat (D) are shown during overground locomotion. Overall, more Chase ABC treated cats have decreased paw drag at 16 and 20 weeks post-hemisection th an the injury-only cats during overground locomotion. There is a significant increase in paw drag at all po st-injury timepoints during bipedal treadmill. During overground lo comotion there is a transient increase in paw drag within the first month after injury, but paw drag returns to normative values within two months following injury. There is no significant treatment effect across paw drag during either task. Asterisks indicate significant differences from pre-injury.
70 Figure 3-6. Swing knee flexion during overground locomotion. Knee angle plots for an entire step cycle normalized in length across animals to 51 frames are shown during overground locomotion for a representative inju ry-only cat at pre-injury (black lines) and at 20 weeks post-hemisection (gray lines). The vertical black line represents the transition from stance to swing during pre-injury locomotion. The maximum knee flexion during swing was significantly decr eased following injury, and this change was not affected by Chase ABC treatment.
71 Figure 3-7. Chase ABC increases esophageal pressure amplitudes following hemisection. Esophageal pressure amplitudes (cmH20) are shown for injury-only and Chase ABC treated cats during pre-injury coughi ng and during coughing at 4, 13, and 21 weeks post-hemisection. Chase ABC treated cats have a significantly greater esophageal pressure during all post-injury coughing se ssion compared to pre-injury values and compared to injury-only animals at the same timepoints. Asterisks indicate a significant change from pre-injury values.
72 CHAPTER 4 CHONDROITINASE ABC PROM OTES RECOVERY OF SKILLED LIMB MOVEMENTS AND PLASTICITY OF THE R UBR OSPINAL TRACT IN THE CAT Introduction Chondroitinase ABC (Chase ABC) has been used experim entally to cleave the growth inhibitory chondroitin sulfat e glycosaminoglycans (CS-GAGs) in vitro (Snow et al. 1990a; McKeon et al. 1995) and in vivo models of spinal cord injury to cleave the growth inhibitory chondroitin sulfate glycosaminoglycans (CS-GAGs ) (Kwok et al. 2008). Disruption of CSGAGs with Chase ABC in vivo either alone or in combination with other treatments in several models of SCI, has been associ ated with behavioral recovery as well as enhancement of axonal growth (Yick et al. 2000; Moon et al. 2001; Bradbury et al. 2002; Yick et al. 2003; Chau et al. 2004; Caggiano et al. 2005; Barritt et al. 2006; Houl e et al. 2006; Massey et al. 2006; Cafferty et al. 2007; Carter et al. 20 08; Iseda et al. 2008; Ma ssey et al. 2008). Previous work by our laborator y illustrated that Chase AB C also appears to enhance motor function following SCI in the cat (Tester and Howland 2008). Some benefits of the cat model include its remarkable locomotor capacity, well characterized circuitry associated with different features of locomotion, the large size of its spinal cord, the sim ilarity between its CS GAG sulfation patterns to that of the human (Tester and Howland 2008), and its use as a platform for prior translation of experimental animal work to the human SCI condition (Young 1991; Hodgson et al. 1994; Behrman and Harkema 2000; de Leon et al. 2001). Of particular interest, our prior work showed that intraspinal delivery of Chase ABC across one month following a low thoracic hemisection promotes recovery of ipsilateral hindlimb use during skilled motor tasks (Tester and Howland 2008). A lthough not tested anatomically, this type of recovery suggests the involvement of the descending rubrospinal system. In the normal cat, the rubrospinal tract contributes to coordinated, multi-articular movements (Gibson et al. 1985;
73 Mewes and Cheney 1994), gait adaptations and in terlimb coordination (Widajewicz et al. 1994; Lavoie and Drew 2002) which are all important during skilled locomotion. There also is precedent in the rat to suggest that this system may respond favorably to Chase ABC treatment (Yick et al., 2004). We hypothesi zed that Chase ABC treatment would significantly enhance recovery of skilled locomotion and also enhanc e axonal growth caudal to the lesion and promote growth of the rubrospinal tract, which is know n to be important during skilled locomotion. The current study focuses on the recovery of the ipsilateral hindlimb following thoracic hemisection during the peg board crossing tas k. The specific movement patterns of the ipsilateral hindlimb including how it is integrated using angular kinematic s, scoring of limb placement, and interlimb coordination are charac terized for 16-20 weeks post-injury. Further, pNF-H is used to assess general axonal growt h, and retrograde tract tr acing with Fluorogold (FG) to assess the presence of rubrospinal axons, below the le vel of the lesion. Significant behavioral and anatomical differences are found be tween Chase ABC treated cats and controls. The movement patterns of the Chase ABC tr eated cats are predicta ble and similar across animals as well as distinctly different from rec overy in control animals. In addition, significant differences are found with regards to axonal growth caudal to the le sion; pNF-H is greater in the ipsilateral white and contralateral gray matter in Chase ABC treated cats and an average of 23% of the axotomized red nucleus neurons have ax ons caudal to the lesion in Chase ABC treated cats compared to 8% in control animals. Methods Animals Twenty two adult fema le cats (6-8 lbs), purch ased from specific-pathogen-free vendors and housed in the AALAC accredited animal facility in the McKnight Brain Institute, were used. All animal procedures were conducted in accordance w ith the NIH guidelines for the care and use of
74 experimental animals and were approved by the Un iversity of Floridas and the Malcom Randall VA Medical Centers Institutional Animal Care and Use Committees. Surgical Procedures Spinal T9/10 hemi sections were performed as previously published (Tester and Howland 2008). Penicillin G, procaine (40,000U/kg BW IM) was given the day before, the day of, and the day after surgery. Prior to all surg eries, cats received 0.1cc atr opine sulfate (0.04-0.06 mg/kg SQ) and 0.1cc acetylpromazine (0.4-0.5 mg/kg SQ). Fo llowing orotracheal int ubation, anesthesia was maintained with isoflurane (1 -3.5%) and an IV was placed for fluid administration (Lactated Ringers, 10 ml/kg/h). Respiratory rate, expired CO2, SPO2, blood pressure, body temperature, as well as the general plane of anes thesia were closely monitored. A left lateral hemisection was made using ir idectomy scissors at sp inal T9 or T10. Light suction with a pulled glass pipette was used if nece ssary to lift any fibers adhering to the dura to facilitate the execution of a complete hemisec tion without compromising the integrity of the dura. Thrombin and gel foam were used to st op any bleeding. Micro-im plantable infusion ports (Harvard Apparatus, Holliston, MA ) were placed sub-cutaneously and sutured to muscle lateral to the vertebral column. Port tubing was held in place by suturing it to the muscle at several points until it reached midline. VetBond (Webst er Veterinary Supply, In c., Sterling, MA) then was used to secure it to the lamina caudal to the laminectomy. The end of the port tubing was secured in the lesion cavity by suturing the dura (8 -0 Prolene). Prior to dural suturing, the port reservoir and tubing (~5l total volume) was filled with either protease free Chase ABC (1 U/200 l Tris-HCl or saline, pH 7.8, Seikagaku Corp., Tokyo, Japan) or vehicle control (TrisHCl pH 7.8). Durafilm (Codman-Shurtleff, Inc., Randolph, MA) and gelfoam (Pharmacia and Upjohn, Inc., Peapack, NJ) were pl aced over the dural su tures and the muscle and skin sutured. Anesthesia was terminated, cats were extubated, and then placed in temperature controlled
75 recovery chambers (ThermoCare, Las Vegas). Buprenorphine (0.01 mg/kg SQ) was administered TID for 48 hours. Procedures used to maintain the general health of the cats were similar to those described in our previous studies (Howland et al. 1995b; Howland et al. 1995a). Once body temperature stabilized to a minimum of 100 F, cats return ed to their home cages with 5-7 inch egg crate foam cushions covering the entire cage floor to prevent peripheral nerve compression, pressure sores, and skin breakdown. None of the animal s developed any of these complications. For the first few days post-op bladders were expressed manually using Credes method. Animal health was continuously monitored throughout the stu dy, including maintenan ce of food intake and body weight. Treatment Administration Ten cats were treated with Chase ABC and tw elve cats were controls Of the controls, 8 received hem isections-only (no port; 2 were non-behavior cats) and 4 received vehicle. Commercial Chase ABC may not retain its stability at body te mperature for extended periods (Tester et al. 2007), therefore 50 microliters of protease-free Cha se ABC (1 U/200l), or vehicle were injected every other day for 1 month. Th e concentration delivered was identical to our previous study (Tester and Howland 2008). The volume is ~4x that used in rodent studies (Bradbury et al. 2002) due to the larg er size of the cat spinal cor d. For treatment administration, the cats were anesthetized (1 -3% isofluorane and 1.5LO2) for 15 minutes while the Chase ABC or vehicle was delivered (0.14L/min) using a mi croinjection syringe pump (Harvard Apparatus, Holliston, MA). Behavioral Training and Quanti tative Locomotor Assess ments Cats were trained daily (5x/w eek) on a variety of locomotor ta sks; for the purposes of this paper only the skilled pegboard ta sk will be presented. The horizontal pegboard was 4.5 meters
76 in length. Alternating pegs on the right and left side of the pegboard, 12 pegs on the right side and 11 on the left side, were spaced evenly along th e length of the board at 20 cm intervals. The width between the right and left pegs was 15cm. The surface of each peg was 3.8cm2 and the height 30.5cm. Crossing of the pegboard was conditioned to a food reward, and when performance was consistent, baseline data was collect ed pre-injury. All cats then received a left T9/10 hemisection and were placed into the control (n=10) or chase ABC treated groups (n=10). Basic locomotor training was re-initiated within 24 hours and peg board cr ossing after the first post-operative week. Manual traine r assistance was given as necessa ry to assist with weight support, postural control and paw placement dur ing pegboard crossings until independence was achieved. Training continued daily. The cats pe rformances were filmed every two weeks until 16-20 weeks post-injury. For video recording, the cats hindquarters were shaved and reflective spheres placed on the iliac crest, greater trocha nter, lateral malleolus, and the base of the 5th metatarsal. A fifth marker was placed on the fibula approximat ely one inch above the lateral malleolus marker in order to generate a unit ve ctor, which in combination with the measured length from the lateral malleolus to the head of the lateral condyle (length of the fibula), was used to automatically calculate the knee position using Motus Software (Vicon-Peak, Englewood, CO). Following a hemisection, the ipsilateral hindlimb is primarily affected; therefore assessments predominantly focused on the left hindlimb (LHL). Postinjury, assistance with weight support was provided until a cat could complete the pegboard task independently. Recovery onset of independent crossing an d integration of LHL placement were determined for each cat. These were defined respectiv ely as at least two independent full crossings and a minimum of one LHL placement/crossing. Placement required that the LHL be positioned and maintained with weight support on a peg. Once independent
77 crossing recovered, the percentage of LHL pl acements was quantified from the three best crossings at each remaining timepoint. Interlim b coordination patterns be tween the four limbs were assessed during independent crossings us ing support (footfall) pattern and stance-swing diagrams during 10 step cycles/cat /timepoint. For these assessments, the step cycle was divided into the swing and stance phases and each phase track ed for all four limbs re lative to each other. These analyses indicate if there is a predictable pattern with rega rds to the timing of these phases across limbs within and across cats. Interlim b coordination and limb placement data were analyzed frame-by-frame using a remote search controller. Angular kinematics were assessed from a minimum of 10 step cycles per time point using Motus software (Vicon-Peak, Englewood, CO). Retrograde Tracing The retrograde tracer Fluorogold (FG, 0.5% in sterile water, Fluorochrome, Inc., Denver, CO) was used to label rubrospinal tract neurons with axons extending below the level of the hemisection. One to two weeks following the la st behavioral data poi nt collection, cats were anesthetized as for the hemisection surgery a nd the lesion site re-exposed. Using a 33 gauge Hamilton syringe, 2l s of FG was injected bilate rally 15 millimeters below the caudal aspect of the lesion. In order to ensure adequate spread of the tracer across the entire cross-section of the spinal cord, the total 2l volume of FG was delivered into 4 injection site s (0.5l each site). Within each site, half of the volume (0.25 l) was placed in the ventral half of the spinal cord and half into the dorsal half. No cats used in this study had FG spread into the lesion site. Tissue Processing and Histology Thirteen days following fluorogold injections, ca ts were anesthetized with an overdose of sodium pentobarbital (>40m g/kg, IP) followe d by 1 cc of heparin (IV; 1000U) followed 20 minutes later by an injection of 1% sodium nitrite (1 cc) intravenously. Cats were perfused
78 transcardially with 0.9% saline followed by 4% paraformaldehyde in 0.1M phosphate buffer (pH 7.4). The brain and spinal cords were removed, blocked, post-fixed with 4% paraformaldehyde overnight, and then placed in 30% sucrose in 4% paraformaldehyde for cryoprotection. Frozen longitudinal sections (25) of the FG injection si tes were cut serially on a cryostat, mounted onto Superfrost/Plus Fisherbrand microscope s lides (Fisher Scientific, Hampton, NH) and coverslipped (Vectashield Hard Se t Mounting Media for Fluorescen ce Vector Laboratories, Inc., Burlingame, CA). FG autofluorescence was assesse d to verify that the tracer was distributed across the width and depth of the spinal cord but did not spread into the lesion site. Mid-brain and spinal cord lesion segments were cut seria lly in cross-section at 25 on a cryostat and collected in a 0.1M phosphate buffer saline (PBS ) solution (pH 7.2, saline 0.9%)). Every tenth section of the lesion and mid-brain, were stained with cresyl violet (cresyl violet with acetate, Sigma-Aldrich, St. Louis, MO) and myelin stai ns (Eriochrome Cyanine R; Fluka, New York, NY) as described previously (Howland et al. 1995a; Tester and Howland 2008). The remaining sections were processed for immunohistochemistry. Immunohistochemistry To identify p-NF-H within and caudal to the lesion, sections were processed using the monoclonal chicken anti-phosphor ylated-neurofilame nt heavy chain antibody (1:10,000, pNF-H Gift from G. Shaw and Encore Biotechnology, Gainesville, FL). Endogenous enzyme activity was quenched using 30% H2O2 in PBS for 30 minutes and then the sections were rinsed with PBS. Sections were rinsed in 1% goat serum in PBS containing 0.4% Triton (1%-S-PBS-T), blocked in 5% goat serum in PBS-T (5%-S-PBS -T) for 1 hour at room temperature, and incubated with the primary antibody diluted in 1 %-S-PBS-T overnight at 4C. The next day tissue sections were rins ed with 1%-S-PBS-T before and af ter a one hour incubation with Alexa
79 Fluor 488 (1:400, Molecular Probes, Eugene, OR). Tissue sections were mounted onto charged slides coverslipped using the ProLong Anti-Fade kit (Molecular Probes). To identify FG labeled rubrospinal neurons, midbrain sections containing the red nuclei were rinsed with 1%-S-PBS-T, blocked with 5% -S-PBS-T at room temperature for one hour and incubated overnight at room temperature with rabbit anti-FG (1:10,000, Fluorochrome, Inc, Denver, CO). Tissue sections were rinsed with 1%-S-PBS-T the next day followed by incubation with an anti-rabbit mouse secondary antibody (V ector laboratories, Bur lingame, CA). Signal amplification was accomplished with the avidin-biotin-complex method (ABC, Vector Laboratories, Burlingame, CA) and visualized with 3-3 Diaminobezidine (DAB) reaction (Sigma, Saint Louis, MO) to produce a brownish stain. To visualize pre-synaptic terminals on FG labeled rubrospinal tract neurons Double immunoperoxidase immunohistochemistry was done using monoclonal mouse antisynaptophysin (1:1,000, Sigma, Saint Louis, MO) and anti-fluorogold anti bodies. Tissue was processed as above for FG staining and then ri nsed with 1%-S-PBS-T and blocked with 5%-SPBS-T at room temperature for one hour. This was followed by incubated in primary antisynaptophysin overnight at 4C. The next day tissue secti ons were rinsed with 1%-S-PBS-T followed by incubation with an anti-mouse seconda ry (Vector Laboratories, Burlingame, CA). Signal amplification was accomplished with the ABC method and visualiz ed with the Vector VIP peroxidase substrate kit (V ector Laboratories, Burlingame, CA) to give a purplish reaction product. Sections were mounted onto chromium potassium sulfate and ply-L-Lysine subbed slides, allowed to dry, exposed to 4% paraformaldeyde fumes for at least 30 minutes to enhance bonding to the slide coating, dehydrated through in creasing alcohol concentrations, placed into xylene and coverslipped using DP X (Fluka, Buchs., Switzerland).
80 Stereological Analyses of pNF-H Three spinal cord cros s-sections 300 apart fr om each other, were used from control (n=4) and Chase ABC (n=4) treated cats in order to as sess the area fraction of pN F-H in four distinct areas of the spinal cord: (1) ip silateral gray matter (2) ipsilateral white matter (3) contralateral gray matter and (4) contralateral white matter. Th e first of these sections in each animal was 1200 caudal to the lesion epicenter. Stereoinvestigator software (MBF Biosciences, Colchester, VT) was used to assess the fractional area of pNF-H immunoreactivity within these contours, across each section, across cats using the area fr action fractionators probe (Cavalieri spacing: 250x500 for the gray matter areas and 375x750 for the white matter areas, Grid spacing: 15, and Frame size: 100 x 100). The data was ca lculated using two dimensional dissectors on a single focal plane with systematic random sampling. To define the contour areas ipsilateral to the lesion, tissue caudal to frank lesion dama ge was included. Inversely, contour areas contralateral to the lesion included tissue ar eas caudal to areas not damaged at the lesion epicenter. This minimized the effect of any differences in sparing at the lesion site. The fractional area of pNF-H imm unoreactivity within each section was obtained per cat and statistically assessed across groups. Quantification of Rubrospinal Neurons The left (spared) red nucleus (RN) was used as an internal cont rol in each animal. FG labeled rubrospinal tract (RST) ne urons in the left and right RN were quantified in each animal from every 8th section (200) throughout th e rostro-caudal extent of each RN. Only neurons with visible punctate FG staini ng throughout their soma were count ed. In addition to calculating the total numbers of labeled neurons in each nuclei, the number of neurons in the right (experimental) RN also were calculated and expresse d as a percentage of the left (spared) RN as an internal control for indivi dual animal differences.
81 Statistical Analyses Statistical A nalyses were performed using Statistical Package for the Social Sciences (SPSS) v. 17 (Chicago, IL). For the categorical data the Fishers ex act Test was used due to the occurrence of cells with frequenc ies <5. Discrete, ordinal data for independent samples (Chase versus Tris) was assessed using the Mann-Whitney U test. All Fishers Exact and MannWhitney U tests were two tailed and a value of P<.05 was considered significant. Discrete, ordinal data for dependent samples (Precompar ed to post-op performance within a group) was assessed using the Wilcoxon Sign-Rank Test. All tests for this assessment were one tailed as performance could change in only one di rection and significance was set at P<0.05. Results Narrow Range of Spinal Hemisections Cresyl violet and m yelin stai ned serial sections were used to determine the extent of spinal cord damage. The lesions of all 22 cats were similar, and typically showed complete ipsilateral gray and white matter damage. Variability across lesions was limited and three spinal cord sections representing the entire range of spinal hemisections are shown in Figure 1. The smallest lesion had some ipsilateral ventromedial white matter sparing and the largest had some contralateral damage (Figure 4-1). Critical to the RN counts, the ipsila teral lateral funiculus where the rubrospinal tract is located was comple tely severed and the contralateral completely spared in all cats. Chase ABC Enhances Multiple Features of Pegboard Performance All cats show ed an initial period (<24 hrs) of flaccid paralysis followed by reflex activity and then voluntary movement of the left hindlimb within 48-72 hours of injury. Consistent with previous reports from our lab and others, basic LHL stepping during bipedal treadmill and voluntary overground locomotion bega n to recover by the end of the first post-operative
82 (Eidelberg et al. 1986; Helgren and Goldberger 1993; Tester and Howl and 2008) indicating the similarity across lesions. Although all cats quickly, efficiently and i ndependently crossed the pegboard prior to injury, the ability to accomplish this task wa s disrupted by hemisecti on (Figure 4-2). At two weeks post-injury, few cats (1/10 controls, 3/10 Chase ABC) could independently cross.the pegboard and there was no significant differe nce in number of cats within each group accomplishing this task (p= 0.582, Fishers Exact). The number of Chase ABC treated cats crossing at 4 weeks post-SCI (8/10) however, wa s significantly greater than the number seen in the control group (2/10; p=.023 Fishers Exact). This significant diffe rence between the two groups performances also was seen at 6 weeks (p=.005, Fishers Exact). By 8 weeks the number of Chase ABC treated (9/10) compared to control (4/10) cats was not significantly different (p=.057, Fishers Exact). The differe nces between groups at 16 weeks (9/10 v 5/10; p=.141 Fishers Exact) and 20 weeks (10/10 v.7/ 10; p=0.211, Fishers Exact) also were not significant. These results suggest that Chase ABC treatment significantly accelerates recovery of crossing but that performance on this abil ity levels out between the two groups around 8 weeks post-injury. Prior to injury, all cats cro ssed the pegboard using four limbs. Following hemisection, however, cats might or might not place their LH L on a peg while crossing (Figure 4-2). When the LHL was not integrated in this manner, ca ts would cross by placing only their other three limbs onto the pegs. Recovery onset of the ability to integrate the LHL by placing it onto a peg was assessed in all cats (Figure 4-2). At two w eeks, although several cats could independently cross, none of the cats placed their LHLs onto pegs. By 4 weeks, using the Fishers Exact Test, a significantly greater number of Ch ase ABC treated cats were integrating their LHL than in the
83 control group (p=.011). This signi ficant difference between the tw o groups continued to be seen at 6 weeks (p=.001), 8 weeks (p=.020), 16 weeks (p=.020) and 20 weeks (p=.033) post-SCI. To determine how effectively cats integrated their LHLs during cro ssings and whether or not Chase ABC enhanced recove ry of this feature, the numbe r of LHL placements onto a peg were quantified (Figure 4-2). No significant differences in performance were seen pre-injury between the cats that would be placed into ea ch group (p=1.0, Mann Whitn ey U) as all cats placed their LHL onto a peg 100% of the time. There also was not a significant difference between the two groups at two week s post-injury (p=1.0) as none of the cats placed their LHLs onto a peg. By 4 weeks, however, the average pe rcentage of LHL placements was significantly greater in the Chase ABC tr eated group (22%) compared to the control group (0%; p=.005, Mann Whitney U). The performance of the Cha se ABC treated animals continued to show significantly greater LHL placements at 8 weeks (28% v. 5%; p=.012) and 16 weeks (44% v. 9%, p=.009). Only 10 cats (4 Chase treated an d 6 controls) remained in the study out to 20 weeks. Using the Mann Whitney U, significant differences also were seen in this smaller number at 20 weeks post-in jury (100% v. 9%, p=.005). To understand how this recovery occurs between time points within each group, additional assessments of the data using the Wilcoxon Matched Pair Sign Rank Test were performed. The control group of cats showed a significant decr ease in LHL peg placements at all time points compared to pre-injury performance: 2 week s (.002), 4 weeks (p= .002), 8 weeks (p=008), 16 weeks (p=.004) and 20 weeks (p=.020). As seen in the control group, the Chase ABC group showed significant decreases in the number of LHL placements at 2 weeks (p=.002), 4 weeks (p=.005), 8 weeks (p=.008) and 16 weeks (p=.018) post-injury co mpared to their pre-injury performances. In contrast to the control gr oup, the LHL performance of Chase ABC treated
84 group at 20 weeks was not significantly different from pre-injury (p=1.0) indicating that performance was similar to that seen preinju ry. Assessments also indicated that significant increases in performance were seen from 2to-4 weeks (p=.028) and from 8-to-16 weeks (p=.012) in the Chase ABC group. No significa nt improvements in performance were seen between any post-injury time points in the contro l group. Collectively thes e data on the use of the LHL suggest that Chase ABC has significant ly enhances the general integration of the hindlimb as well as the accuracy with which the limb is used. Chase ABC Treated Cats Develop a N ovel Interlimb Coordination Pattern Prior to injury, cats typically place their left limbs (fore and hind) on the pegs on the left side of the board and their right limbs (fore and hind) on the right side pegs (Figure 4-3). This basic placement strategy is not seen post-injury. Cats that do not re-integrate their LHLs simply do not place them (Figure 4-3). However, th e cats that do reintegrate the limb, cross the body midline with the LHL to place it onto a peg on the right side of the board (Figure 4-3). Further, the LHL is now typically paired with the right fo relimb (RFL), for at least initial peg contact. Thus, the LHL and RFL share a peg for some am ount of time on the right side of the pegboard (Figure 4-3). Pre-injury, when the majority of cats (16/20) co nsistently kept their left and right limbs on the left and right sides of the pegboard respectiv ely, their performances were characterized by a single, consistent footfall pattern that was repeated with the be ginning of each step cycle (Figure 4-3). During the majority of the step cycle, on ly one limb at a time was in swing. The very brief periods in which the swing of two limbs coincided were typically 1-2 video frames in length which is equivalent to 33-66 milliseconds. The limbs paired during th ese brief swing phase overlaps showed a consistent, re peating pattern. Overlap of the ipsilateral foreand hindlimb swing phases occurred first followed by overlap of the contralateral foreand hindlimb swing
85 phases. Additionally initiation of the stance phases of the four limbs also occurred in a predictable order beginning w ith the LHL, followed by the LF L, then right hindlimb (RHL) ending with the right forelimb (RFL). Approxi mately 60% of the stepcycle was characterized by triple limb support time and ~40% by double lim b support. Analyses using support pattern diagrams also indicate the use of a consistent pattern interlimb pattern (Figure 4-3). The support pattern diagrams showed a consistent 3-2-3-2-32-3-2 support formula (Figure 4-3). Thus, as has been shown in many studies for quadrupeda l treadmill and voluntary overground locomotion, steps on a pegboard show a consiste nt interlimb coordination pattern. Three Chase ABC treated cats capable of inte grating their LHLs consistently at 16 weeks post-injury were evaluated to determine if a ny showed a predicable interlimb coordination pattern. Interestingly, not onl y was a consistent interlimb coor dination pattern seen but it was consistent across the three cats and di stinctly different in several ways from that seen pre-injury. The total support time of all limbs was increas ed and contributed to the introduction of a quadrupedal support period. In part icular, the stance phase of the left hindlimb was lengthened such that it overlapped with at least two stance phases of each of the other limbs. The LHL swing phase also was lengthened. The increased time spent in each of these phases resulted in a 1:2 ratio of LHL stepcycles to the stepcycles of each of the other limbs. Th e general order of the stance and swing phases for each limb relative to the others showed a similarity to the pre-injury pattern with the ex ception of the LHL. The s upport pattern diagrams (Fig ure 4-3) also suggested a consistent interlimb coordina tion patterm. A repeating support formula of 3-2-3-4-3-2-3 for the first set of step cycle and 4-3-2-3-4-3-2 for the subsequent step cyclewas seen.Thus, footfall patterns and support diagrams both indicate that a unique pattern of interlimb coordination is seen in Chase ABC treated cats whic h recover LHL placement on the pegboard.
86 The LHL proximal angular kinematic patt erns showed distinctly different ranges of movement in cats that placed their LHLs on pegs versus those that did not. Although control animals typically crossed the pe gboard on three limbs, the LHL was not passive. It alternated between flexion and extension, but its active range of movement was much smaller than seen pre-injury (Figure 4-4). In dr amatic contrast, the angular ki nematics of the proximal LHL of Chase ABC treated cats that placed post-injur y showed nearly twice the range of angular excursion at the hip and the knee in comparison to pre-injury valu es (Figure 4-4). The increased angular excursion is consistent with the placem ent of the LHL on the right side pegs and the skipping of a peg due to the 1LHL:2 other lim b stepcycle ratio seen post-injury. The peg skipping of the LHL post-injury was in contrast to LHL placement onto every peg pre-injury. Collectively these results suggest that Chase ABC cats develop unique but consistent and effective new LHL movement strategies pos t-hemisection for crossing the pegboard. Enhancement of Axonal Densities at the Spi nal Level in Cats Receiving Chase ABC Axonal growth was assessed qualitatively with in the lesion epicente r using an antibody against the 200kD phosphorylated axonal form of the neurofilament heavy chain ( pNF-H) positive axonal profiles were visible throughout the lesion scar in all animals assessed (Figure 45). In all Chase ABC treated animals the pNF-H positive axons appeared highly fasciculated and densely packed in the lesion environment (Figure 4-5). This staining profile was observed in only one control animal (Figure 4-5). Axons pos itive for pNF-H in the le sion epicenter of the remaining three control animals were not highl y fasciculated or densely packed and had a blunted (Figure 4-5) or thin filamentous app earance (Figure 4-5). pNF-H staining density was not assessed as the size of the scar area vari ed notably across cats and would have confounded any findings.
87 The density of pNF-H profiles were quantit atively assessed in eigh t cats, four controls and four Chase ABC treated cats, throughout the ipsilateral and contrala teral gray and white matter caudal to the hemisections. Three spinal cord cross-sec tions starting 1200 caudal to the lesion epicenter as defined by cresyl-violet my elin staining, and 300 apart from each other, were used from control (n=4) and Chase ABC (n=4 ) treated cats. The area fraction of pNF-H in four distinct contoured areas of these sections was assessed: (1) ipsilateral gray matter (2) ipsilateral white matter (3) contralateral gray matter and (4) contralateral white matter. Ipsilateral contours are caudal and on the same si de as the spinal hemise ction, and contralateral contours are caudal and on the spar ed side of the spinal cord. The area fractions of pNF-H immunoreactivity within the caudal ipsilateral gray matter (Figure 4-6) and contralateral white matter (Figur e 4-6) were not significantly different between Chase ABC and control treated cats as assessed by the Mann Whitney U (p=0.166 for each) In contrast, Chase ABC treated cats had a significantly greater area fraction of pNF-H immunoreactivity in the caudal ip silateral white matter as comp ared to controls (p=.033 Mann Whitney U; Figure 4-6. Chase ABC treated cats also had a significantlygreater area fraction of pNF-H in the caudal contralateral gray matter as co mpared to controls (p= .003; Figure 4-6). More Rubrospinal Neurons have Axons Caudal to the Hemisec tion in Chase ABC Treated Cats To determine if rubrospinal axons contribu ted to the increased pNF-H immunoreactivity seen caudal to the lesion, retr ograde tract tracing studies were conducted using Fluorogold (FG) in 10 cats (5 controls and 5 Cha se ABC treated cats). Bilateral injections of FG were made approximately 1 1/2-2 segments below the orig inal left, T9/10 lateral hemisection. The injections sites of all 10 cats showed a good dist ribution of the tracers across the entire crosssectional area, but did not spread into the lesion site .The hemis ection of every cat in this study
88 completely interrupted the rubrospinal tract on the side of the lesion at th e lesion epicenter FG-labeled neurons in the left (control) RN were found throughout the hindlimb region of th RNm (Figure 4-7). The number of FG-labeled RN neurons in the left (control) RN was not significantly different between Chase ABC and control animals (740 and 894 respectively). The number of retrogradely neur ons in the right (axotomized) RN however was greatly decreased compared to the left (Figure 47). Chase ABC treated cats had an average of 308 compared to 137 in the control group. These averages were then expressed as a percentage of the number of labeled neurons in th e left RN. The percentage of neurons in the Chase treated cats (23%) was significantly greate r than the than in the control cats (9%) as determined with the Mann Whitney U (p=032 Figure 4-7). Brainstem sections stained for FG thr oughout the RN also were double labeled with synaptophysin, a pre-synaptic terminal marker. Synaptophysin labeled puncta were visualized on the perimeter of FG labeled neurons in the left (control) RN (Figure 4-7). Synaptophysin colocalized around FG labeled neurons in the right (axotomized) RN in Chase ABC as well as control treated cats (Fig ure 4-7), suggesting all of these neurons were receiving input. Discussion In the present study, we demonstrated that degradation of CS-GAGs via intraspinal delivery of Chase ABC prom oted axonal growth caudal to the lesion as well as regeneration and/or collateral sproutin g of approximately 22% of the axotom ized rubrospinal tract neurons. The RST is a major descending pathway associated with the control of skilled locomotion, and correlatively, Chase ABC treated cats also had significantly enhanced recovery during skilled pegboard locomotion. Chase ABC treated animals were able to integrat e the ipsilateral LHL earlier and significantly more at all post-injur y timepoints compared to controls, and this recovery was paralleled by the use of a nove l, complex hindlimb movement pattern that
89 integrated the LHL kinematically different than pr e-injury. We further de monstrated that Chase ABC treated animals had an increased pNF-H area fraction within the caudal ipsilateral white matter and contralateral gray matter, indicating that Chase ABC promoted axonal growth through the lesion environment as well as around the lesion that maintained projections 1-2 segments below the original lesion. This axonal growth may have included axons of RST origin following axotomy and Chase ABC treatment. Skilled Pegboard following Thoracic Hemisection and Chase ABC Treatment The RST is principally involved with m odulatin g motor control and has a specific role in skilled motor functions (Whishaw et al. 1998). Lesions of the RST have been shown to affect forelimb function during skilled locomotion (Schrimsher and Reier 1993; Whishaw and Gorny 1996; Whishaw et al. 1998; Muir et al. 2007), and there also is ev idence that the RST contributes to hindlimb function as well (Orlovsky 1972a; Lavoie and Drew 2002). Rubrospinal neurons have been proven to aid in precise limb modification during skilled locomotion, regulation of locomotion during gait adaptation to environmental demands, and for the regulation of intraand interlimb coordination in the normal cat (Wid ajewicz et al. 1994; Lavoie and Drew 2002). Pegboard locomotion requires precise intraand interlimb coordination as well as accurate limb targeting. Our hemisection model axotomized the RST unilaterally and consequently caused substantial deficits in pegboard locomotion. Similar to the ladder rung walking test used in rats (Metz and Whishaw 2002), the pegboard task used in our cat model is exceptiona lly challenging. Immedi ately following spinal hemisection, cats must precisely adapt their we ight support and compensa te for the ipsilateral limb deficits by adjusting postura l control and shifting of the body weight to the less affected limbs. There is considerable evidence that anim als with lesions of motor pathways maintain the ability to compensate for lesi on-induced deficits in skilled locomotion (Miklyaeva et al. 1994;
90 Whishaw et al. 1997a; Whishaw et al. 1997b; Kleim et al. 1998; Whishaw et al. 1998; Metz and Whishaw 2002). Despite a notable compensatory ch ange in behavior, ther e are still substantial impairments present. Similarly, in the pres ent study, control animal s sustained residual impairment of LHL placement onto pegboard pegs, but some were able to compensate by crossing the pegboard on three limbs. Chase A BC treated animals had substantially decreased impairment of LHL placement onto the pegs at all post-injury timepoints compared to controls. The appearance of a common compensatory response occurred in Chase treated cats post-injury, where the LHL was no longer placed onto the left si de of the pegboard like pre-injury placement, but it was instead paired with the RFL on the right side of the pegboard. This unique limb placement strategy altered the support pattern of a ll four limbs when compared to pre-injury placement. Notably, this post-injury compensatory strategy was equally as efficient as the preinjury strategy. Muir et al. hypothesized that following unilateral lesions affecting multiple descending pathways, a common plastic response may ar ise to mask deficits that are specific to the loss of each different pathway (Muir et al. 2007). Since our lesion paradigm affects multiple descending motor systems, it is therefore possi ble that this could be occurring with the emergence of the unique compensatory strategy in Chase ABC treated animals. The ability to re-integrate the impaired limb during skilled pegboard locomotion correlated with increased axonal growth caudal to the lesion and an increase in growth of axotomized RST axons, illustrates that Chase ABC treatment affected the plasticity of a critical motor system and subsequently caused enhanced skilled behavioral function. Training in all animals in the present study involved treadmill walking ,as well as crossing of simple and challenging runways 5x/ week (for examples see (Tester and Howland 2008). Substantial evidence suggests that lo comotor training can improve hindlimb/lower
91 extremity motor functions post-spinal cord inju ry in animals and humans respectively (Edgerton et al. 2004; Thomas and Gorassini 2005; Behr man et al. 2006; Fri gon and Rossignol 2006). Voluntary exercise also has been shown to in crease the production of several neurotrophins (Gomez-Pinilla et al. 2002; Ying et al. 2003) which may play signi ficant roles in motor recovery and synaptic plasticity post-spi nal cord injury (Yi ng et al. 2008). This possible increase in neurotrophins such as BDNF from our rigorous training regimen may have fascilitated skilled locomotor recovery in our animals, as well as enhancement of RST axonal growth. Control animals in this study had approximately an 8% increase in axotomized RST axons caudal the spinal lesion. This increase could have partially been due to an upregulation of neurotrophic support from locomotor training, a nd interestingly the percentage increase in RST growth was similar to studies where cells genetically modifi ed to express BDNF were transplanted into cervical lesions in the rat (Liu et al. 1999; Liu et al. 2002b). Furt hermore, tracing studies have demonstrated that an ipslateral rubrospinal pathway exists in the feline (Hayes and Rustioni 1981; Holstege and Kuypers 1982; Holstege 1987). Therefore, it is plausible that a small percentage of retrogradely labeled RST neurons in the experimental, ri ght red nucleus of our control and Chase ABC treated animals came from ipsilateral projections. Strategies to Promote Rubrospinal Tr act Growth after Cervical Axotomy Spontaneous plasticity of multiple axonal tracts has been established af ter partial lesions of the spinal cord across a variety of species (Murra y and Goldberger 1974; Li et al. 1994; Fouad et al. 2001b; Weidner et al. 2001; Bare yre et al. 2004; Steward et al. 2008). It has been shown previously that following a cervi cal hemisection in the adult rat, RST axons approach the rostral lesion edge but do not have the ca pacity to spontaneously re-grow through or caudal to the lesion (Houle and Jin 2001), therefore numerous experiment al strategies to specifically promote RST axonal growth have been conducted in anim al models of spinal cord injury.
92 Acute transplantation of olfactory ensheathing glia or fibroblasts genetically modified to express the brain-derived neurotrophic factor (BDNF) following cervical lesions of the lateral funiculus in the adult rat promot ed minimal sprouting of RST a xons distal to the lesion site (Ruitenberg et al. 2003), and m odest growth of 7-10% of axot omized RST axons (Liu et al. 1999; Liu et al. 2002b). Intrathecal delivery of NEP1-40, a Nogo receptor (NgR) antagonist proved even less effective, induci ng RST axonal growth rostral and at the level of the lesion but not caudal (Cao et al. 2008). Transplants of fi broblasts modified to express BDNF and NT-3 elicited modest regeneration of RST axons when the transplants were delayed for 6 weeks following cervical hemisection (Tobias et al. 200 3), and less growth was observed than when grafted acutely (Liu et al. 1999) or after 4 week s of delay following lesion (Jin et al. 2002). Infusion of BDNF directly to the RST neur onal cell bodies one week post-axotomy at the cervical level induced less than 5% growth of axotomized rubr ospinal axons into a peripheral nerve graft (Kobayashi et al. 1997) and approximate ly 3% when applied one year after cervical injury (Kwon et al. 2002). Fetal spinal cord transplanted into a cervical hemisection pr omoted occasional RST axons to grow into the transplant but none caudal (Mori et al. 1997), and pe ripheral nerve gr afted into the lesion epicenter alone promoted th e growth of approximately 2% of RST axons (Kobayashi et al. 1997; Harvey et al. 2005). Transplantation of hum an adult olfactory neuroepithelial neurosphere forming cells (NSFCs) induced growth of RST a xons 4-8 segments caudal to the graft and reestablished synaptic connections with distal targets (Xiao et al. 2007). All together, these experimental strategies elicited minimal RST axonal growth following partial lesions of the cervical spinal cord.
93 Chase ABC Treatment to Promote Spinal Plasticity and Rubrospinal Tract Grow th Disruption of CS-GAGs with Chase ABC in vivo, either alone or in combination with other treatments, has been shown to enhance axonal growth and/or functional behavioral recovery after SCI (Yick et al. 2000; Moon et al. 2001; Bradbury et al. 2002; Yick et al. 2003; Chau et al. 2004; Caggiano et al. 2005; Barritt et al. 2006; Houle et al. 2006; Massey et al. 2006; Cafferty et al. 2007; Carter et al. 2008; Iseda et al. 2008; Massey et al. 20 08; Tester and Howland 2008). Degradation of CS-GAGs using Chase A BC has been shown to effectively enhance the regeneration of injured nigros triatal axons (Moon et al. 2001) as well as of crushed CST axons (Bradbury et al. 2002). Previous studies utilizi ng the enzymatic degradation of CS-GAGs with the administration of Chase ABC into the lesi on cavity following cervical hemisection in the adult rat, resulted in a 22% re -growth of RST axons caudal to the original lesion, and in combination with lithium chloride (LiCl), growth of RST axons was enhanced to 42% (Yick et al. 2004). Until the present study, experiments have not been conducted that assess the growth of RST axons following axotomy of the thoracic sp inal cord, and correlatively in an even more translational model of spinal cord injury. Previous studies have shown that implantation of peripheral nerves into the adult rat spinal cord following cervical transection of the RST resulted in only a small percentage (1-2%) of rubrospinal axon growth into the peripheral ne rve graft (Richardson et al. 1984; Houle 1991; Tetzlaff et al. 1994), whereas descending axons ra rely regenerated follow ing thoracic or lumbar injury, implicating that the distance from ce ll body to injury was a strong determinant of potential axonal growth (Richardso n et al. 1984). Fernandes et al. confirmed and extended their results illustrating that RST neurons have the growth capacity to exte nd their axons into a peripheral nerve transplant afte r cervical but not after thoracic axotomy and that after cervical and not thoracic axotomy, regeneration-associated gene expression was enhanced (Fernandes et
94 al. 1999). Based on the poor growth potential follow ing thoracic spinal injuries, the growth of 22% of axotomized RST neurons following Cha se ABC treatment in our study is remarkable, especially considering that in our feline model of thoracic injury the distance from cell body to axotomy is much larger than it would be in the rat. Axonal growth was also assessed in our studies witin the lesion qualita tively and caudal to the spinal hemisection by quantitatively. Degradation of neurofilament proteins occurs following spinal cord injury (Banik et al 1982; Martin et al. 1990; Banik et al. 1997; Schumacher et al. 1999; von Euler et al. 2002; Liu et al. 2009) and tr aumatic brain injury (Nakamura et al. 1992; Kaku et al. 1993; Saatman et al. 1998; Huh et al. 2002). It has previously been reported that 15 weeks following ischemic spinal cord injury that axons positive for pNF-H are present within the lesion cavity (von Euler et al. 2002). In the present study, Chase ABC treated cats had dense, fasc iculated pNF-H immunolabeling profiles within the lesion environment as opposed to control animals that had some swollen axonal profiles and an overall less dense pNF-H appearance. Caudal to the lesi on Chase ABC treated cats had an increase in the area fraction of pNF-H within the ipsilateral white matter and contralateral gray matter as compared to controls. These results indicate that Chase ABC enhances axonal plasticity through and around a thoracic hemisection. The incr eased axonal growth in the ipsilateral white matter following Chase ABC treatment may have lead to functional re-connectivity of the RST and in turn behavioral recovery in our Chase ABC treated animals. In conclusion, Chase ABC promotes axona l growth caudal to a thoracic spinal hemisection in the adult cat, and correlativel y increases the regenera tion and/or collateral sprouting of rubrospinal tract axons caudal to the lesion. Furthermore, these growth
95 enhancements of the RST with Chase ABC treat ment may have affected the skilled locomotor recovery also seen.
96 Figure 4-1. Representative range of spinal hemisections. Horizo ntal sections through the lesion epicenter stained with cresyl violet and myelin show th at lesions ranged from an under-hemisection with ipsilateral ventromedial sparing (A), to a complete lesion with interruption of all ipsilateral gray and white matter and no damage to the contralateral tissue (B), to an over-hemisecti on with interruption of contralateral gray and white matter and possible cyst formati on or enlargement of the central canal ( C ). Animals used in this study had slight va riations of tissue sparing and damage as compared to these examples. Scale bar: 1mm.
97 Figure 4-2. Pegboard locomotor recovery is im proved with Chase ABC administration. The ability to cross the pegboard (bar graphs) and place th e left hindlimb (LHL) on the pegboard (line graphs) was assessed at multip le post-injury timepoints (A). During pre-injury pegboard locomotion, all cats were able to cross the pe gboard and place the LHL onto the pegs with 100% accuracy. Our lesion paradigm causes significant locomotor impairment in the ipsilateral LHL. Following injury, significantly more Chase ABC treated cats could cross the pegboard independently at 4 and 6 weeks post-injury compared to controls (*). Ch ase ABC treated cats at every post-injury timepoint compared to controls could integrate the LHL into placement significantly more (++). The percentage of step cycl es that the LHL was accurately placed on the pegboard was assessed at multiple post-inj ury timepoints (B). Chase ABC treated cats integrate the LHL a greater percentage of time compar ed to controls at 4, 6, 8, 16, and 20 weeks post-injury (**). Error ba rs denote SEM. I ndicate a significant change across groups when assessing the ability to cross the pegboard. ++ Indicate a significant change across groups at a particular timepoint when assessing the ability to integrate the LHL. ** Indicate a significant change across groups at each timepoint when assessing the percentage of LHL placement onto the pegboard
98 Figure 4-3. Limb placement strategies during pe gboard locomotion. During pre-injury crossing of the pegboard, the majority of cats pl ace their four limbs in a specific spatial pattern. The left limbs stay on the left side of the pegboard and the right limbs stay on the right side of the pegboard during crossing, and no two limbs are ever placed onto the same peg (A). Following injur y, control animals primarily cross the pegboard using three limbs and are not able to incorporate thei r LHL into pegboard placement (B). Post-injury, Chase ABC cats are able to incorporate the LHL into placement onto the pegboard (C) but differently than pre-injury. The LHL now pairs with the right forelimb at in itial contact on the right side of the pegboard (A, C). Footfall pattern diagram during pre-injury pegboard crossing (D ) and post-injury LHL placement at 16wphx by a representative Ch ase ABC treated cat (F). The solid bars indicate when the limb is in the stance phase and the open areas indicate when the limb is in the swing phase. Two comple te step cycles are shown in D and F. Time between vertical lines is one frame (33.3ms). Pre-injury, the support pattern diagram shows a 3-2-3-2-3-2-3-2 support fo rmula (E). The support pattern diagram for 16wphx crossing of the pegboard show s a repeating two-step cycle support formula of 3-2-3-4-3-2-3 a nd 4-3-2-3-4-3-2 (G). Line s drawn between limbs in the support pattern diagram indicate that the two limbs are being paired on the same peg. LF, left forelimb; RF, right forelimb; LH left hindlimb; RH, right hindlimb.
99 Figure 4-4. Proximal left hindlimb angular kinematics during different pegboard crossing strategies. LHLs of cont rol cats are active during postinjury non-placement (A). Angle-angle kinematic pattern of the proxima l LHL of a control treated cat is shown during pre-injury placement of the LH L on the pegboard (blue) and during nonplacement at 16wphx (red). At 16wphx, the ma jority of injury-only animals are unable to integrate their LHL into placem ent on the pegboard. Despite this, during non-placement, when the control animals ar e crossing on three limbs the LHL is still flexing and extending with a similar kinematic pattern to pre-injury. The range of excursion of the proximal limb is smalle r, but the overall kinematic pattern is preserved. The angular kine matics of the proximal LHL of Chase ABC treated cats has a larger excursion at 16wphx during pl acement (red) on the pegboard than during pre-injury placement (blue), due to the us e of an alternate placement strategy postinjury (B). Smaller joint angles on either axis represent increased flexion of the joint being assessed, and larger angles represent increased joint extension. At each timepoint 10 representative step cycles are shown.
100 Figure 4-5. Phosphorylated neur ofilament heavy chain (pNF-H) immunoreactivity within the lesion epicenter. pNF-H within the lesion epicenter of Chase ABC treated cats (A, B, C, D) and control cats (E, F, G, H). pNF-H axonal growth appears to be stunted and more finely distributed in the majority of control animals (E, F, G) as compared to all Chase ABC treated animals where th e pNF-H immunoreactivity in the lesion is dense and highly fasciculated. Scale bar: 0.1mm.
101 Figure 4-6. Increases in phosphoryl ated neurofilament heavy chain (pNF-H) caudal to the spinal hemisection with Chase ABC treatment. Quantification of pNF-H caudal to the spinal hemisection (A). Non-biased stereo logical quantification of the area fraction of pNF-H was assessed in four distinct cont ours of the caudal spinal cord; ipsilateral white matter (B,C), ipsilateral gray matter (D, E), contralateral white matter (F, G), and contralateral gray matter (H, I). Quantif ication revealed that Chase ABC treated animals (B, D, F, H) had a significant increase in the area fraction of pNF-H as compared to control animals (C, E, G, I) in the ipsilateral white matter (B, C) and in the contralateral gray matter (H, I). There was no significant difference seen in the area fraction of pNF-H across groups in the ipsilateral gray matter (D, E) or the contralateral white matter (F, G). Error ba rs denote SEM, Indicates a significant change across groups. Photomicrographs of the ipsilateral and contralateral white matter (B, C, F, and G) are within the latera l funiculus, where the RST axons traverse. Scale bar: 0.1mm. Y axis values in A are the area fract ion of pNF-H staining as a percentage of the contoured area.
103 Figure 4-7. Retrograde labeling in axotomized red nucleus neurons is increased with Chase ABC administration. Representa tive retrogradely labeled neurons in the control red nucleus (A), and in the experimental red nucleus (B) afte r fluorogold (FG) injections bilaterally and caudal to the original spinal hemisection. Chase ABC treated cats had a greater number of retrogr adely labeled neurons in th e experimental red nucleus expressed as a percentage of the neurons labeled on the c ontrol side as compared to controls (C), Indicate a significant change between groups Error bars denote SEM. Scale bar: 1mm. Retrograde ly labeled control red nucleus neurons co-stained for the pre-synaptic terminal marker synaptophysin (D), and FG labeled red nucleus neurons in the experimental red nucleus of contro l (E) and Chase ABC treated animals (F) also co-localized with synaptophysin.
104 CHAPTER 5 SUMMARY, CONCLUSIONS, AND FUTURE DIRECTIONS Following trauma tic spinal injury, axons have a minimal capacity for regeneration due to the presence of the g lial scar, de-myelination, and the up-re gulation of inhibitory molecules including chondroitin sulfate proteoglycans (CSPG s) (Schwab and Bartholdi 1996). Synthesis of several members of the family of CSPGs is increased and these proteins are concentrated in the area of the glial scar following injury (McKeon et al. 1995; Fitch and Silver 1997; Lemons et al. 1999; Asher et al. 2000). These proteins have been shown to define barriers to migrating neurons and restrict th e extension of axons in vitro (Hynds and Snow 1999; Snow et al. 2001; Johnson et al. 2002). Specifically, the chondroitin sulfate glyc osaminoglycan (CS-GAG) side chains have been shown to inhibit axonal rege neration and plasticity following SCI (Bandtlow and Zimmermann 2000). Degradation of CSGAGs with Chondroitinase ABC (ChABC), a bacterial enzyme isolated from proteus vulgaris disrupts their inhibitory properties in vitro (Snow et al. 1990a; McKeon et al. 1995) and with in the last eight years has been shown to enhance axonal growth and behavioral recovery in rodent models of SCI (Yick et al. 2000; Bradbury et al. 2002; Yick et al 2003; Caggiano et al. 2005; Houl e et al. 2006) and in our cat model (Tester and Howland 2007). The cat presents an excellent translational model in wh ich to study the effects of therapeutic interventions such as ChABC treat ment on systems underlying locomotor recovery. The benefits of this model include its rema rkable locomotor capacity, well characterized circuitry associated with different features of lo comotion, the larger size of its spinal cord, and its use as a platform for prior tran slational work in spinal cord injury (Hodgson et al. 1994; Behrman and Harkema 2000; de Leon et al. 2001). Previous rese arch from our laboratory has shown that following a low thoracic spinal hemisection in the cat model, CS-GAG degradation
105 with ChABC enhanced recovery of basic a nd skilled locomotion (Tester and Howland 2007; Tester and Howland 2008). My studies have further assessed th e intrinsic behavioral plasticity following a low thoracic hemisec tion during many features of ba sic and skilled locomotion as well as assessed the effects of this lesion model on another motor system, the cough reflex. These studies are the first to demonstrate that Chase ABC promotes axonal growth caudal to a thoracic spinal hemisection in the adult cat, a nd correlatively increases the regeneration and/or collateral sprouting of rubrospina l tract axons caudal to the lesi on. Furthermore, these growth enhancements of the RST with Chase ABC treat ment may have affected the skilled locomotor recovery also seen. Future studies could further assess the plasticity within the cough reflex after low thoracic spinal hemisection. Assessing the cough reflex at early timepoints directly after injury could help us to understand the timeline of plasticity that occurs in th is particular motor system. We could assess whether the cough motor system is resilient following low thoracic hemisection or whether it is extremely plastic to the induced spinal lesion. Th e rectus abdominiis electromyogram recordings also could be furthe r assessed by analyzing such things as burst duration, inspiratory activity burst duration, as well as the relationship between the esophageal pressure records and rectus abdominis electro myogram records in a temporal fashion. This work has also shown that thoracic he misection affects many temporal components of the gait cycle such as swing and stance dura tion during overground and bi pedal treadmill. It would be interesting to further breakdown the st ep cycle into the 4 epochs of time defined by Phillipson (E2, E3, F, and E1) in order to assess at what particular part of the step cycle the ipsilateral hindlimb was making the most alterations in response to the in jury. Assessments of possible Chase ABC affects on the temporal com ponents of the gait cycle such as step cycle
106 duration, swing duration, stance du ration, and the duration of the four sub-phases during a more skilled task such as ladder or pegboard locomotion may also be future projects to pursue. It would also be helpful to assess if the velocity or acceleration of the proximal or distal limb was affected by Chase ABC application during basic or skilled locomotor tasks, which may help attribute to the emergence of the unique intr alimb pairing pattern during pegboard locomotion following Chase ABC treatment. Primarily, studies assessing locomotor function fo llowing partial lesions of the spinal cord have focused on ipsilateral limb recovery (Webb a nd Muir 2002; Courtine et al. 2005). To better understand the effects of thoracic hemisecti on in our lesion model on locomotor recovery, assessment of the contralateral limb could elucid ate any compensatory strategiesor mechanisms of recovery that could affect th e ipsilateral limb. The contralateral limb likely plays an important role in the recovery of locomotor function follo wing thoracic hemisection. Preliminary data by our laboratory show that the cont ralateral hindlimb, following thoracic hemisection in the adult cat may stabilize its locomotor pattern during bipedal treadmill and overground locomotion postinjury in order to allow for f unctional improvement in the ipsila teral hindlimb that acquired the most deficits following injury. It has been shown that when you behaviorally train the unimpaired limb following nervous system injury, the more impaired limb does not recovery to its full potential (Allred and J ones 2008). Therefore, kinema tic stabilization of the RHL following spinal cord injury may be a compensato ry mechanism used to help regain function in the more impaired limb. It would also be lucrative to assess the contralateral limb during more skilled tasks such as pegboard and ladder locomotion. Whether Chase ABC treatment affects any parameters of the RHLduring basic or skilled locomotion would also be important to assess.
107 Lastly, many novel assessments could cont inue the retrograde tract tracing study conducted utilizing bilateral, cauda l injections of the regrograde tracer Fluorogold for assessment of rubrospinal tract axons that had grown through or around the original hemisection in response to the injury alone or to Chase ABC treatment. The current results do not identify whether the increase in rubrospinal neurons in Chase ABC tr eated cats in the experimental red nucleus was due to collateral sprouting of int act axons proximal or distal to th e injury site or regeneration of axotomized axons. A double retrog rade tract tracing study could he lp elucidate this quandary. A retrograde tracer such as Fast Blue could be placed at the site of injury at the original time of spinal hemisection to label any cut axons. A second axonal tracer could be added at the end of the study, such as Fluorogold, to label any axons that are at the caudal site of injection. Double labeled neurons in the red nucleus would indicate that an axon was originally axotomized during the injury and then regrew thr ough or around the injury site. The corticospinal tr act is also a very important motor tract invo lved with the control of skil led locomotion. Neuron counts of Fluorogold labeled neurons within the motor cort ex of cats following thoracic hemisection and Chase ABC treatment could further elucidate whether the corticospina l tract may also be involved in the significant skilled locomotor recove ry we see following Chase ABC treatment. Many different strategies could be implemented that assess di fferent delivery systems for Chase ABC in our model system, including cells genetically engineered to secrete Chase ABC into the scar environment, nanosphere technology to deliver the enzyme after injury, as well as viral vector delivery systems. Combinatorial treatment approaches could also be undertaken utilizing stem cell delivery, de livery of neurotrophic factors such as BDNF and/or NT-3, application of olfactory ensheat hing cells to provide scaffoldi ng for cellular remodeling, as well as combining Chase ABC treatment with pe ripheral nerve grafts in our cat model.
108 LIST OF REFERENCES Allred RP, Jones TA (2008) Malada ptive effects of lear ning with the less-aff ected forelimb after focal cortical infarcts in rats. Exp Neurol 210: 172-181 Alstermark B, Lundberg A, Pettersson LG, Tan tisira B, Walkowska M (1987) Motor recovery after serial spinal cord lesions of defined descending pathways in cats. Neurosci Res 5: 6873 Anderson DK, Demediuk P, Saunders RD, Dugan LL, Means ED, Horrocks LA (1985) Spinal cord injury and protectio n. Ann Emerg Med 14: 816-821 Aoki M, Fujito Y, Satomi H, Kurosawa Y, Ka saba T (1986) The possibl e role of collateral sprouting in the functional restitution of cor ticospinal connections af ter spinal hemisection. Neurosci Res 3: 617-627 Arita H, Kogo N, Koshiya N (1987) Morphologi cal and physiological properties of caudal medullary expiratory neurons of the cat. Brain Res 401: 258-266 Armstrong DM (1988) The supraspinal control of mammalian locomotion. J Physiol 405: 1-37 Asher RA, Morgenstern DA, Fidler PS, Adcock KH, Oohira A, Braistead JE, Levine JM, Margolis RU, Rogers JH, Fawcett JW (2000) Neurocan is upregulated in injured brain and in cytokine-treated astrocytes. J Neurosci 20: 2427-2438 Bandtlow C, Zachleder T, Schwab ME (1990) Oligodendrocytes arrest neur ite growth by contact inhibition. J Neurosci 10: 3837-3848 Bandtlow CE, Zimmermann DR (2000) Proteoglycan s in the developing br ain: new conceptual insights for old proteins. Physiol Rev 80: 1267-1290 Banik NL, Hogan EL, Powers JM, Whetstine LJ ( 1982) Degradation of cyto skeletal proteins in experimental spinal cord in jury. Neurochem Res 7: 1465-1475 Banik NL, Matzelle DC, Gantt-Wilford G, Os borne A, Hogan EL (1997) Increased calpain content and progressive degrada tion of neurofilament protein in spinal cord injury. Brain Res 752: 301-306 Barbeau H, Rossignol S (1987) Recovery of locomo tion after chronic spinalization in the adult cat. Brain Res 412: 84-95 Bareyre FM, Kerschensteiner M, Raineteau O, Mettenleiter TC, Weinmann O, Schwab ME (2004) The injured spinal cord spontaneously forms a new intraspinal circuit in adult rats. Nat Neurosci 7: 269-277
109 Barritt AW, Davies M, Marchand F, Hartley R, Grist J, Yip P, Mc Mahon SB, Bradbury EJ (2006) Chondroitinase ABC promotes sprouting of intact and injured spinal systems after spinal cord injury. J Neurosci 26: 10856-10867 Behrman AL, Bowden MG, Nair PM (2006) Neuroplasti city after spinal cord injury and training: an emerging paradigm shift in rehabilitation and walking recover y. Phys Ther 86: 14061425 Behrman AL, Harkema SJ (2000) Locomotor training after human spinal cord injury: a series of case studies. Phys Ther 80: 688-700 Belanger M, Drew T, Provencher J, Rossignol S (1996) A comparison of treadmill locomotion in adult cats before and after spinal transection. J Ne urophysiol 76: 471-491 Bolser DC (1991) Fictive cough in the cat. J Appl Physiol 71: 2325-2331 Bolser DC, Aziz SM, DeGennaro FC, Kreutner W, Egan RW, Siegel MI, Chapman RW (1993) Antitussive effects of GABAB agonists in th e cat and guinea-pig. Br J Pharmacol 110: 491-495 Bolser DC, Davenport, P.W. ( 2002) Expiratory motor contro l during the cough reflex. Eur Respir Rev 12: 243-248 Bolser DC, Reier PJ, Davenport PW (2000) Respons es of the anterolateral abdominal muscles during cough and expiratory threshold loadi ng in the cat. J Appl Physiol 88: 1207-1214 Bradbury EJ, Moon LD, Popat RJ, King VR, Bennett GS, Patel PN, Fawcett JW, McMahon SB (2002) Chondroitinase ABC promotes functional r ecovery after spinal cord injury. Nature 416: 636-640 Bregman BS, Goldberger ME (1983) Infant lesion effect: II. Sparing a nd recovery of function after spinal cord damage in newbor n and adult cats. Brain Res 285: 119-135 Bregman BS, Kunkel-Bagden E, Schnell L, Dai HN, Gao D, Schwab ME (1995) Recovery from spinal cord injury mediated by antibodies to neurite growth inhib itors. Nature 378: 498501 Brittis PA, Canning DR, Silver J (1992) Chondroitin sulfate as a regulator of neuronal patterning in the retina. Science 255: 733-736 Buford JA, Zernicke RF, Smith JL (1990) Adap tive control for backward quadrupedal walking. I. Posture and hindlimb kinema tics. J Neurophysiol 64: 745-755 Bunge RP, Puckett WR, Hiester ED (1997) Observ ations on the pathology of several types of human spinal cord injury, with emphasis on th e astrocyte response to penetrating injuries. Adv Neurol 72: 305-315
110 Cafferty WB, Bradbury EJ, Lidierth M, Jone s M, Duffy PJ, Pezet S, McMahon SB (2008) Chondroitinase ABC-mediated plasticity of sp inal sensory function. J Neurosci 28: 1199812009 Cafferty WB, Yang SH, Duffy PJ, Li S, Strittma tter SM (2007) Functional axonal regeneration through astrocytic scar genetical ly modified to digest chondro itin sulfate proteoglycans. J Neurosci 27: 2176-2185 Caggiano AO, Zimber MP, Ganguly A, Blight AR, Gruskin EA (2005) Chondroitinase ABCI improves locomotion and bladder function follo wing contusion injury of the rat spinal cord. J Neurotrauma 22: 226-239 Cajal SRy (1991) Cajal's Degeneration and Re generation of the Nervous System. Oxford University Press, USA Cao Y, Shumsky JS, Sabol MA, Kushner RA, Stri ttmatter S, Hamers FP, Lee DH, Rabacchi SA, Murray M (2008) Nogo-66 receptor antagonist peptide (NEP1-40) administration promotes functional recovery and axonal growth after lateral funiculu s injury in the adult rat. Neurorehabil Neural Repair 22: 262-278 Caroni P, Schwab ME (1988a) Antibody against my elin-associated inhibitor of neurite growth neutralizes nonpermissive substrate propert ies of CNS white matter. Neuron 1: 85-96 Caroni P, Schwab ME (1988b) Two membrane prot ein fractions from rat central myelin with inhibitory properties for neurite growth a nd fibroblast spreadi ng. J Cell Biol 106: 12811288 Carter LM, Starkey ML, Akrimi SF, Davies M, McMahon SB, Bradbury EJ (2008) The yellow fluorescent protein (YFP-H) mouse reveals neuroprotection as a novel mechanism underlying chondroitinase ABC-mediated repair after spinal co rd injury. J Neurosci 28: 14107-14120 Chau CH, Shum DK, Li H, Pei J, Lui YY, Wirt hlin L, Chan YS, Xu XM (2004) Chondroitinase ABC enhances axonal regrowth through Schw ann cell-seeded guidance channels after spinal cord injury. FASEB J 18: 194-196 Conta AC, Stelzner DJ (2004) Diffe rential vulnerability of propriosp inal tract neurons to spinal cord contusion injury. J Comp Neurol 479: 347-359 Courtine G, Roy RR, Raven J, Hodgson J, McKay H, Yang H, Zhong H, Tuszynski MH, Edgerton VR (2005) Performance of locomotio n and foot grasping following a unilateral thoracic corticospinal tract lesion in monkeys (Macaca mulatta). Brain 128: 2338-2358 Courtine G, Song B, Roy RR, Zhong H, Herrmann JE, Ao Y, Qi J, Edgerton VR, Sofroniew MV (2008) Recovery of supraspina l control of stepping via in direct propriospinal relay connections after spinal co rd injury. Nat Med 14: 69-74
111 David S, Aguayo AJ (1981) Axonal elongation into peripheral nervous system "bridges" after central nervous system injury in adult rats. Science 214: 931-933 Davies SJ, Field PM, Raisman G (1996) Regeneration of cut a dult axons fails even in the presence of continuous aligned glia l pathways. Exp Neurol 142: 203-216 Davies SJ, Goucher DR, Doller C, Silver J (1999 ) Robust regeneration of adult sensory axons in degenerating white matter of the adult ra t spinal cord. J Neurosci 19: 5810-5822 Davis GM (1993) Exercise capacity of individuals with paraplegia. Med Sci Sports Exerc 25: 423-432 Davis JN, Plum F (1972) Separati on of descending spinal pathwa ys to respiratory motoneurons. Exp Neurol 34: 78-94 de Leon RD, Roy RR, Edgerton VR (2001) Is the recovery of stepping following spinal cord injury mediated by modifying existing neural pathways or by generating new pathways? A perspective. Phys Ther 81: 1904-1911 Demopoulos HB, Flamm ES, Pietronigro DD, Se ligman ML (1980) The free radical pathology and the microcirculation in the major central ne rvous system disorders. Acta Physiol Scand Suppl 492: 91-119 Dietz V, Wirz M, Curt A, Colombo G (1998) Locomotor pattern in paraplegic patients: training effects and recovery of spinal co rd function. Spinal Cord 36: 380-390 DiMarco AF, Kowalski KE (2008) Effects of chronic electrical s timulation on paralyzed expiratory muscles. J Appl Physiol 104: 1634-1640 Ditunno JF, Little JW, Tessler A, Burns AS (2004) Spinal shock revisited: a four-phase model. Spinal Cord 42: 383-395 Dobkin B, Barbeau H, Deforge D, Ditunno J, Elashoff R, Apple D, Basso M, Behrman A, Harkema S, Saulino M, Scott M (2007) The evolution of walking-related outcomes over the first 12 weeks of rehabil itation for incomplete traumatic spinal cord injury: the multicenter randomized Spinal Cord Injury Loco motor Trial. Neurorehabil Neural Repair 21: 25-35 Domeniconi M, Cao Z, Spencer T, Sivasankaran R, Wang K, Ni kulina E, Kimura N, Cai H, Deng K, Gao Y, He Z, Filbin M (2002) Myel in-associated glycoprotein interacts with the Nogo66 receptor to inhibit neurite outgrowth. Neuron 35: 283-290 Dou CL, Levine JM (1994) Inhibition of ne urite growth by the NG2 chondroitin sulfate proteoglycan. J Neurosci 14: 7616-7628 Drew T (1993) Motor cortical acti vity during voluntary gait modi fications in the cat. I. Cells related to the forelimbs. J Neurophysiol 70: 179-199
112 Drew T, Jiang W, Widajewicz W (2002) Contributions of the motor cortex to the control of the hindlimbs during locomotion in the cat Brain Res Brain Res Rev 40: 178-191 Eddleston M, Mucke L (1993) Mol ecular profile of reactive astroc ytes--implications for their role in neurologic diseas e. Neuroscience 54: 15-36 Edgerton VR, Tillakaratne NJ, Bigbee AJ, de Leon RD, Roy RR (2004) Plasticity of the spinal neural circuitry after injur y. Annu Rev Neurosci 27: 145-167 Eidelberg E, Nguyen LH, Deza LD (1986) Recovery of locomotor function after hemisection of the spinal cord in cats. Brain Res Bull 16: 507-515 Eng LF, Reier PJ, Houle JD (1987) Astrocyte activation and fibrous g liosis: glial fibrillary acidic protein immunostaining of astr ocytes following intraspinal cord grafting of fetal CNS tissue. Prog Brain Res 71: 439-455 Ezure K, Otake K, Lipski J, She RB (1991) Efferent projections of pulmonary rapidly adapting receptor relay neurons in the cat. Brain Res 564: 268-278 Fawcett JW, Asher RA (1999) The glial scar and central nervous system repair. Brain Res Bull 49: 377-391 Fawcett JW, Curt A, Steeves JD, Coleman WP Tuszynski MH, Lammertse D, Bartlett PF, Blight AR, Dietz V, Ditunno J, Dobkin BH, Ha vton LA, Ellaway PH, Fehlings MG, Privat A, Grossman R, Guest JD, Kleitman N, Nakamura M, Gaviria M, Short D (2007) Guidelines for the conduct of clinical trials fo r spinal cord injury as developed by the ICCP panel: spontaneous recovery af ter spinal cord in jury and statistical power needed for therapeutic clinical trials Spinal Cord 45: 190-205 Fawcett JW, Housden E, Smith-Thomas L, Meyer RL (1989) The growth of axons in threedimensional astrocyte cultures. Dev Biol 135: 449-458 Fernandes KJ, Fan DP, Tsui BJ, Cassar SL, Tetzlaff W (1999) Influence of the axotomy to cell body distance in rat rubrospina l and spinal motoneurons: di fferential regulation of GAP43, tubulins, and neurofilament-M. J Comp Neurol 414: 495-510 Filbin MT (1995) Myelin-associated glycoprotein: a role in myelin ation and in the inhibition of axonal regeneration? Curr Opin Neurobiol 5: 588-595 Fitch MT, Silver J (1997) Activated macrophages and the blood-brain barrier: inflammation after CNS injury leads to increases in putative inhibitory molecules. Exp Neurol 148: 587-603 Fitch MT, Silver J (2008) CNS injury, glial scars, and inflammation: I nhibitory extracellular matrices and regeneration fa ilure. Exp Neurol 209: 294-301
113 Fontana F (1787) Treatise on the venom of the vipe r on the American poisons; and on the cherry laurel, and some other vegetabl e poisons. ... Translated from the original French of Felix Fontana, ... By Joseph Skinner, ... In two volumes. printed for, London Forssberg H, Grillner S, Halbertsma J (1980a ) The locomotion of the low spinal cat. I. Coordination within a hindlimb. Acta Physiol Scand 108: 269-281 Forssberg H, Grillner S, Halbertsma J, Rossi gnol S (1980b) The locomotion of the low spinal cat. II. Interlimb coordination. Acta Physiol Scand 108: 283-295 Fouad K, Dietz V, Schwab ME (2001a) Improvi ng axonal growth and func tional recovery after experimental spinal cord injury by neutraliz ing myelin associated inhibitors. Brain Res Brain Res Rev 36: 204-212 Fouad K, Pearson K (2004) Restoring walking af ter spinal cord injur y. Prog Neurobiol 73: 107126 Fouad K, Pedersen V, Schwab ME, Brosamle C (2001b) Cervical sprout ing of corticospinal fibers after thoracic spinal co rd injury accompanies shifts in evoked motor responses. Curr Biol 11: 1766-1770 Fouad K, Schnell L, Bunge MB, Schwab ME, Liebscher T, Pearse DD (2005) Combining Schwann cell bridges and olfactory-ensheathing glia grafts with chondroitinase promotes locomotor recovery after complete transecti on of the spinal cord. J Neurosci 25: 1169-1178 Fournier AE, GrandPre T, Gould G, Wang X, Strittmatter SM (2002) Nogo and the Nogo-66 receptor. Prog Brain Res 137: 361-369 Fournier AE, GrandPre T, Stri ttmatter SM (2001) Identification of a receptor mediating Nogo-66 inhibition of axonal regene ration. Nature 409: 341-346 Frigon A, Rossignol S (2006) Functional plasticity following spinal cord lesions. Prog Brain Res 157: 231-260 Galtrey CM, Fawcett JW (2007) The role of chon droitin sulfate proteoglycans in regeneration and plasticity in the central nervous system. Brain Res Rev 54: 1-18 Garcia-Rill E, Skinner RD, Fitzgerald JA (1983) Activity in the mesencephalic locomotor region during locomotion. Exp Neurol 82: 609-622 Gibson AR, Houk JC, Kohlerman NJ (1985) Magnoce llular red nucleus activ ity during different types of limb movement in the maca que monkey. J Physiol 358: 527-549 Goldberger ME (1988) Spared-root deafferentation of a cat's hindlim b: hierarchical regulation of pathways mediating recovery of mo tor behavior. Exp Brain Res 73: 329-342
114 Gomez-Pinilla F, Ying Z, Roy RR, Molteni R, Edgerton VR (2002) Voluntary exercise induces a BDNF-mediated mechanism that promotes neuroplasticity. J Neurophysiol 88: 2187-2195 Gorio A, Di Giulio, AM., Mantegazza, P. ( 1993) Historical Bases for an Introduction to Neuroregeneration. In: Gorio A, ed. Neuroregeneration. In. Raven Press, New York, pp xii, 333 Goslow GE, Jr., Reinking RM, Stuart DG (1973) Th e cat step cycle: hind limb joint angles and muscle lengths during unrestrained locomotion. J Morphol 141: 1-41 GrandPre T, Nakamura F, Vartanian T, Stri ttmatter SM (2000) Identification of the Nogo inhibitor of axon regeneration as a Reticulon protein. Nature 403: 439-444 Grillner S, McClellan A, Perret C (1981) Entrai nment of the spinal pattern generators for swimming by mechano-sensitive elements in the lamprey spinal cord in vitro. Brain Res 217: 380-386 Grillner S, Zangger P (1979) On the central gene ration of locomotion in the low spinal cat. Exp Brain Res 34: 241-261 Grimpe B, Silver J (2002) The extracellular matrix in axon regeneration. Prog Brain Res 137: 333-349 Grundy D, Swain A, NetLibrary Inc. (2002) ABC of spinal cord injury. BMJ, London Hall ED, Springer JE (2004) Neur oprotection and acute spinal cord inju ry: a reappraisal. NeuroRx 1: 80-100 Halseth AE, Fogt DL, Fregosi RF, Henriksen EJ (1995) Metabolic response s of rat respiratory muscles to voluntary exercise tr aining. J Appl Physiol 79: 902-907 Hartmann U, Maurer P (2001) Proteoglycans in the nervous system--the quest for functional roles in vivo. Matrix Biol 20: 23-35 Harvey PJ, Grochmal J, Tetzla ff W, Gordon T, Bennett DJ (2005) An investigation into the potential for activity-dependent regeneration of the rubrospinal tract after spinal cord injury. Eur J Neurosci 22: 3025-3035 Hayes NL, Rustioni A (1981) Descending projections from brainstem and sensorimotor cortex to spinal enlargements in the cat. Single and double retrograde tracer st udies. Exp Brain Res 41: 89-107 Helgren ME, Goldberger ME (1993) The recovery of postural reflexes and locomotion following low thoracic hemisection in adult cats involves compensation by undamaged primary afferent pathways. Exp Neurol 123: 17-34
115 Hemingway A, Bors E, Hobby RP (1958) An i nvestigation of the pulmonary function of paraplegics. J Clin Invest 37: 773-782 Hiersemenzel LP, Curt A, Dietz V (2000) From spinal shock to spasticity: neuronal adaptations to a spinal cord inju ry. Neurology 54: 1574-1582 Hinman A, Carpenter MB (1959) Efferent fiber proj ections of the red nucleu s in the cat. J Comp Neurol 113: 61-82 Hodgson JA, Roy RR, de Leon R, Dobkin B, Edgerton VR (1994) Can the mammalian lumbar spinal cord learn a motor task? Med Sci Sports Exerc 26: 1491-1497 Holstege G (1987) Anatomical evidence for an ipsilateral rubrospinal pathway and for direct rubrospinal projections to motoneurons in the cat. Neurosci Lett 74: 269-274 Holstege G, Kuypers HG (1982) The anatomy of brain stem pathways to the spinal cord in cat. A labeled amino acid tracing study. Prog Brain Res 57: 145-175 Houle JD (1991) Demonstration of the potential for chronically injured neurons to regenerate axons into intraspinal peripheral ne rve grafts. Exp Neurol 113: 1-9 Houle JD, Jin Y (2001) Chronically injured s upraspinal neurons exhibit only modest axonal dieback in response to a cervical hemi section lesion. Exp Neurol 169: 208-217 Houle JD, Tom VJ, Mayes D, Wagoner G, Phillips N, Silver J (2006) Combining an autologous peripheral nervous system "bridge" and matr ix modification by chondroitinase allows robust, functional regeneration beyond a hemisec tion lesion of the adult rat spinal cord. J Neurosci 26: 7405-7415 Howland DR, Bregman BS, Tessler A, Goldberger ME (1995a) Development of locomotor behavior in the spinal ki tten. Exp Neurol 135: 108-122 Howland DR, Bregman BS, Tessler A, Goldberger ME (1995b) Transplants enhance locomotion in neonatal kittens whose spinal cords are tr ansected: a behavioral and anatomical study. Exp Neurol 135: 123-145 Huang DW, McKerracher L, Braun PE, David S (1999) A therapeutic vaccine approach to stimulate axon regeneration in the adult mammalian spinal cord. Neuron 24: 639-647 Hughes JT (1988) The Edwin Smith Surgical Papyrus: an analysis of the first case reports of spinal cord injuries. Paraplegia 26: 71-82 Huh JW, Laurer HL, Raghupathi R, Helfaer MA Saatman KE (2002) Rapid loss and partial recovery of neurofilament immunostaining following focal brain injury in mice. Exp Neurol 175: 198-208
116 Hulsebosch CE (2002) Recent advances in patho physiology and treatment of spinal cord injury. Adv Physiol Educ 26: 238-255 Hynds DL, Snow DM (1999) Neur ite outgrowth inhibition by c hondroitin sulfate proteoglycan: stalling/stopping exceeds turning in human neur oblastoma growth cones. Exp Neurol 160: 244-255 Ikegami T, Nakamura M, Yamane J, Katoh H, Okada S, Iwanami A, Watanabe K, Ishii K, Kato F, Fujita H, Takahashi T, Okano HJ, T oyama Y, Okano H (2005) Chondroitinase ABC combined with neural stem/progenitor cell tr ansplantation enhances graft cell migration and outgrowth of growth-associated protein-43-positive fibers after rat spinal cord injury. Eur J Neurosci 22: 3036-3046 Iseda T, Okuda T, Kane-Goldsmith N, Mathew M, Ahmed S, Chang YW, Young W, Grumet M (2008) Single, high-dose intraspinal in jection of chondroitinase reduces glycosaminoglycans in injured spinal cord and promotes co rticospinal axonal regrowth after hemisection but not contusion. J Neurotrauma 25: 334-349 Jaeger RJ, Turba RM, Yarkony GM, Roth EJ ( 1993) Cough in spinal cord injured patients: comparison of three methods to produce cough. Arch Phys Med Rehabil 74: 1358-1361 Jefferson S TJ, Blum A, Bolser D, Howland D (2008) Cough following low thoracic hemisection in the cat. Experimental Neurology In review Jiang C, Lipski J (1990) Extensive monosynaptic inhibition of ventral re spiratory group neurons by augmenting neurons in the Botzinger complex in the cat. Exp Brain Res 81: 639-648 Jin Y, Fischer I, Tessler A, Houle JD (2002) Tran splants of fibroblasts ge netically modified to express BDNF promote axonal regeneration fr om supraspinal neurons following chronic spinal cord injury. Exp Neurol 177: 265-275 Joester A, Faissner A (2001) The structure and function of tenascins in the nervous system. Matrix Biol 20: 13-22 Johnson WE, Caterson B, Eisenstein SM, H ynds DL, Snow DM, Roberts S (2002) Human intervertebral disc aggrecan inhibits nerve growth in vitro. Arthritis Rheum 46: 2658-2664 Jordan LM (1998) Initiation of locomotion in mammals. Ann N Y Acad Sci 860: 83-93 Kaku Y, Yonekawa Y, Tsukahara T, Ogata N, Ki mura T, Taniguchi T (1993) Alterations of a 200 kDa neurofilament in the rat hippocampus after forebrain ischemia. J Cereb Blood Flow Metab 13: 402-408 Kato M, Murakami S, Yasuda K, Hirayama H (1984) Disruption of foreand hindlimb coordination during overground locomotion in cats with bilateral serial hemisection of the spinal cord. Neurosci Res 2: 27-47
117 Kawahara K, Kumagai S, Nakazono Y, Miyamo to Y (1989a) Coupling be tween respiratory and stepping rhythms during locomotion in decerebrate cats. J Appl Physiol 67: 110-115 Kawahara K, Nakazono Y, Yamauchi Y, Miya moto Y (1989b) Coupling between respiratory and locomotor rhythms during fictive locomotion in decerebrate cats. Neurosci Lett 103: 326-330 Ke T, Zhangfu L, Qing G, Yong T, Hong J, Hon gyan R, Kun L, Shigui L (2005) Isolation of Serratia marcescens as a chondroitinase-produc ing bacterium and purification of a novel chondroitinase AC. Biotec hnol Lett 27: 489-493 Keirstead HS, Hasan SJ, Muir GD, Steeves JD (1992) Suppression of the onset of myelination extends the permissive period for the functiona l repair of embryonic spinal cord. Proc Natl Acad Sci U S A 89: 11664-11668 Keirstead SA, Rasminsky M, Fukuda Y, Carter DA, Aguayo AJ, Vidal-Sanz M (1989) Electrophysiologic responses in hamster superi or colliculus evoked by regenerating retinal axons. Science 246: 255-257 Kern H, Hofer C, Modlin M, Mayr W, Vindigni V, Zampieri S, Boncompagni S, Protasi F, Carraro U (2008) Stable muscle atrophy in long-term paraplegics with complete upper motor neuron lesion from 3to 20-year SCI. Spinal Cord 46: 293-304 Kirkwood PA (1995) Synaptic excitation in the thoracic spinal cord from expiratory bulbospinal neurones in the cat. J Ph ysiol 484 ( Pt 1): 201-225 Kirkwood PA, Munson JB, Sears TA, Westgaard RH (1988) Respiratory interneurones in the thoracic spinal cord of the cat. J Physiol 395: 161-192 Kleim JA, Barbay S, Nudo RJ (1998) Functional re organization of the rat motor cortex following motor skill learning. J Neurophysiol 80: 3321-3325 Klein R (2001) Excitatory Eph receptors and adhe sive ephrin ligands. Curr Opin Cell Biol 13: 196-203 Kobayashi NR, Fan DP, Giehl KM, Bedard AM, Wiegand SJ, Tetzlaff W (1997) BDNF and NT4/5 prevent atrophy of rat rubrospinal neur ons after cervical axot omy, stimulate GAP-43 and Talpha1-tubulin mRNA expression, and prom ote axonal regeneration. J Neurosci 17: 9583-9595 Kowalski KE, Romaniuk JR, DiMarco AF (2007) Changes in expiratory muscle function following spinal cord section. J Appl Physiol 102: 1422-1428 Kreutzberg GW (1996) Microglia: a sensor for pat hological events in the CNS. Trends Neurosci 19: 312-318
118 Kucera P, Wiesendanger M (1985) Do ipsilateral cor ticospinal fibers participate in the functional recovery following unilateral pyramidal le sions in monkeys? Br ain Res 348: 297-303 Kuhtz-Buschbeck JP, Boczek-Funcke A, Mautes A, Nacimiento W, Weinhardt C (1996) Recovery of locomotion after sp inal cord hemisection: an X-ray study of the cat hindlimb. Exp Neurol 137: 212-224 Kuypers HG (1964) The Descending Pathways to the Spinal Cord, Their Anatomy and Function. Prog Brain Res 11: 178-202 Kwok JC, Afshari F, Garcia-Alias G, Fawcett JW (2008) Proteoglycans in the central nervous system: plasticity, regeneration and their stim ulation with chondroitinase ABC. Restor Neurol Neurosci 26: 131-145 Kwon BK, Liu J, Messerer C, Kobayashi NR, McGraw J, Oschipok L, Tetzlaff W (2002) Survival and regeneration of r ubrospinal neurons 1 year after spinal cord injury. Proc Natl Acad Sci U S A 99: 3246-3251 Lane MA, White TE, Coutts MA, Jones AL, Sandhu MS, Bloom DC, Bolser DC, Yates BJ, Fuller DD, Reier PJ (2008) Cervical prephren ic interneurons in th e normal and lesioned spinal cord of the adult ra t. J Comp Neurol 511: 692-709 Lavoie S, Drew T (2002) Discharge characteristics of neurons in the red nucleus during voluntary gait modifications: a comparison with the motor cortex. J Neurophysiol 88: 1791-1814 Lawrence DG, Kuypers HG (1968) The functional organization of the motor system in the monkey. II. The effects of lesions of the desc ending brain-stem path ways. Brain 91: 15-36 Lee BB, Boswell-Ruys C, Butler JE, Gande via SC (2008) Surface functional electrical stimulation of the abdominal muscles to enhance cough and assist tracheostomy decannulation after high-level spinal cord injury. J Sp inal Cord Med 31: 78-82 Lemons ML, Howland DR, Anderson DK ( 1999) Chondroitin sulfate proteoglycan immunoreactivity increases following spinal co rd injury and transplantation. Exp Neurol 160: 51-65 Li WW, Yew DT, Chuah MI, Leung PC, Tsang DS (1994) Axonal sprouting in the hemisected adult rat spinal cord. Neuroscience 61: 133-139 Lim J, Gorman RB, Saboisky JP, Gandevia SC, Butler JE (2007) Optimal electrode placement for noninvasive electrical stimulation of huma n abdominal muscles. J Appl Physiol 102: 1612-1617 Lin VW, Hsiao IN, Zhu E, Perkash I (2001) Func tional magnetic stimulation for conditioning of expiratory muscles in patients with spinal co rd injury. Arch Phys Med Rehabil 82: 162-166
119 Linder SH (1993) Functional electr ical stimulation to enhance c ough in quadriplegia. Chest 103: 166-169 Liu BP, Fournier A, GrandPre T, Strittmatter SM (2002a) Myelin-associated glycoprotein as a functional ligand for the Nogo-66 receptor. Science 297: 1190-1193 Liu CN, Chambers WW (1958) Intr aspinal sprouting of dorsal r oot axons; development of new collaterals and preterminals following partial denervation of the spin al cord in the cat. AMA Arch Neurol Psychiatry 79: 46-61 Liu Y, Himes BT, Murray M, Tessler A, Fischer I (2002b) Grafts of BDNF-producing fibroblasts rescue axotomized rubrospinal neurons and prev ent their atrophy. Exp Neurol 178: 150-164 Liu Y, Kim D, Himes BT, Chow SY, Schallert T, Murray M, Tessler A, Fischer I (1999) Transplants of fibroblasts genetically modified to express B DNF promote regeneration of adult rat rubrospinal axons and recovery of forelimb function. J Neurosci 19: 4370-4387 Liu YL, Guo YS, Xu L, Wu SY Wu DX, Yang C, Zh ang Y, Li CY (2009) Alternation of neurofilaments in immune-mediated injury of sp inal cord motor neurons. Spinal Cord 47: 166-170 Makarem EH, Berk RS (1968) Partial purificati on and characterization of chondroitinase from Proteus mirabilis. J Infect Dis 118: 427-435 Martin JE, Mather KS, Swash M, Garofalo O, Dale GE, Leigh PN, Anderton BH (1990) Spinal cord trauma in man: studies of phosphorylated neurofilament and ubiquitin expression. Brain 113 ( Pt 5): 1553-1562 Massey JM, Amps J, Viapiano MS, Matthews RT, Wagoner MR, Whitaker CM, Alilain W, Yonkof AL, Khalyfa A, Cooper NG, Silver J, Onifer SM (2008) Increased chondroitin sulfate proteoglycan expression in denervat ed brainstem targets following spinal cord injury creates a ba rrier to axonal regeneration overcome by chondroitinase ABC and neurotrophin-3. Exp Neurol 209: 426-445 Massey JM, Hubscher CH, Wagoner MR, Decker JA Amps J, Silver J, Onifer SM (2006) Chondroitinase ABC digestion of the perineuronal net promotes functional collateral sprouting in the cuneate nucleus after cervica l spinal cord injury. J Neurosci 26: 4406-4414 Maurel P, Rauch U, Flad M, Margolis RK, Margolis RU (1994) Phosphacan, a chondroitin sulfate proteoglycan of brai n that interacts with neurons and neural cell-adhesion molecules, is an extracellular variant of a r eceptor-type protein tyrosine phosphatase. Proc Natl Acad Sci U S A 91: 2512-2516 McCrea DA (2001) Spinal circuitry of sensorimotor control of locomotion. J Physiol 533: 41-50
120 McKeon RJ, Hoke A, Silver J (1995) Injury-induced proteogly cans inhibit the potential for laminin-mediated axon growth on astroc ytic scars. Exp Neurol 136: 32-43 McKeon RJ, Schreiber RC, Rudge JS, Silver J (19 91) Reduction of neurite outgrowth in a model of glial scarring following CNS injury is co rrelated with the expression of inhibitory molecules on reactive astrocytes. J Neurosci 11: 3398-3411 McKerracher L, David S, Jackson DL, Kottis V, Dunn RJ, Braun PE (1994) Identification of myelin-associated glycoprotein as a major my elin-derived inhibito r of neurite growth. Neuron 13: 805-811 Merrill EG (1970) The lateral respiratory neurones of the medu lla: their associations with nucleus ambiguus, nucleus retroambigualis, the spinal accessory nucleus and the spinal cord. Brain Res 24: 11-28 Merrill EG (1974) Finding a respiratory function for the medullary resp iratory neurons. In: Essays on the nervous system. Clarendon Press, Oxford Eng., pp viii, 511 Merrill EG, Lipski J (1987) I nputs to intercostal motoneurons from ventrolateral medullary respiratory neurons in th e cat. J Neurophysiol 57: 1837-1853 Metz GA, Whishaw IQ (2002) Cortical and subcor tical lesions impair skilled walking in the ladder rung walking test: a new task to eval uate foreand hindlimb stepping, placing, and co-ordination. J Neurosci Methods 115: 169-179 Mewes K, Cheney PD (1994) Primate rubrom otoneuronal cells: parametric relations and contribution to wrist movement. J Neurophysiol 72: 14-30 Michelacci YM, Dietrich CP ( 1975) A comparative study between a chondroitinase B and a chondroitinase AC from Flavobacterium hepa rinum: Isolation of a chondroitinase ACsusceptible dodecasaccharide from chondroi tin sulphate B. Biochem J 151: 121-129 Miklyaeva EI, Castaneda E, Whishaw IQ (1994) Skilled reaching deficits in unilateral dopaminedepleted rats: impairments in movement a nd posture and compensatory adjustments. J Neurosci 14: 7148-7158 Miller AD (1987) Localization of motoneurons innervating indivi dual abdominal muscles of the cat. J Comp Neurol 256: 600-606 Miller AD, Tan LK, Lakos SF (1989) Brainstem pr ojections to cats' upper lumbar spinal cord: implications for abdominal muscle control. Brain Res 493: 348-356 Miller AD, Tan LK, Suzuki I (1987) Control of abdominal and expiratory intercostal muscle activity during vomiting: role of ventral respiratory group expiratory neurons. J Neurophysiol 57: 1854-1866
121 Miller LE, van Kan PL, Sinkjaer T, Andersen T, Harris GD, Houk JC (1993) Correlation of primate red nucleus discharge with muscle activity during free-form arm movements. J Physiol 469: 213-243 Miller S, Reitsma DJ, van der Meche FG ( 1973) Functional organiza tion of long ascending propriospinal pathways linking lumbo-sacral a nd cervical segments in the cat. Brain Res 62: 169-188 Monteau R, Hilaire G (1991) Spinal respirat ory motoneurons. Prog Neurobiol 37: 83-144 Moon LD, Asher RA, Rhodes KE, Fawcett JW (2001) Regeneration of CNS axons back to their target following treatment of adult rat brain with chondroiti nase ABC. Nat Neurosci 4: 465-466 Moon LD, Asher RA, Rhodes KE, Fawcett JW ( 2002) Relationship between sprouting axons, proteoglycans and glial cells following unilatera l nigrostriatal axotomy in the adult rat. Neuroscience 109: 101-117 Mori F, Himes BT, Kowada M, Murray M, Tessler A (1997) Fetal spinal cord transplants rescue some axotomized rubrospinal neurons from re trograde cell death in ad ult rats. Exp Neurol 143: 45-60 Mori S, Nishimura H, Kurakami C, Yamamura T, Aoki M (1978) Controlled locomotion in the mesencephalic cat: distribution of facilitato ry and inhibitory re gions within pontine tegmentum. J Neurophysiol 41: 1580-1591 Muir GD, Webb AA, Kanagal S, Taylor L (20 07) Dorsolateral cervical spinal injury differentially affects forelimb and hindlimb action in rats. Eur J Neurosci 25: 1501-1510 Mukhopadhyay G, Doherty P, Walsh FS, Crocker PR, Filbin MT (1994) A novel role for myelin-associated glycoprotein as an inhi bitor of axonal regene ration. Neuron 13: 757-767 Murray M, Goldberger ME (1974) Restitution of function and collateral sprouting in the cat spinal cord: the partially hemisected animal. J Comp Neurol 158: 19-36 Murray MP (1967) Gait as a total pattern of movement. Am J Phys Med 46: 290-333 Nakamura M, Araki M, Oguro K, Masuzawa T ( 1992) Differential distribution of 68 Kd and 200 Kd neurofilament proteins in the gerbil hippo campus and their early distributional changes following transient forebrain is chemia. Exp Brain Res 89: 31-39 Niederost BP, Zimmermann DR, Schwab ME, Bandtlow CE (1999) Bovine CNS myelin contains neurite growth-inhibitory activ ity associated with chondroitin sulfate proteoglycans. J Neurosci 19: 8979-8989 NSCISC (2008) Facts and Figures at a Glance-January 2008. In: vol p
122 Nyberg-Hansen R, Brodal A (1964) Sites and Mode of Termination of Rubr ospinal Fibres in the Cat. An Experimental Study with Silver Impregnation Methods. J Anat 98: 235-253 Orlovsky GN (1972a) Activity of rubrospinal neurons during locomotion. Brain Res 46: 99-112 Orlovsky GN (1972b) The effect of different descending systems on fl exor and extensor activity during locomotion. Brain Res 40: 359-371 Pasterkamp RJ, Verhaagen J (2001) Emerging ro les for semaphorins in neural regeneration. Brain Res Brain Res Rev 35: 36-54 Pesheva P, Probstmeier R (2000) The yin and yang of tenascin-R in CNS development and pathology. Prog Neurobiol 61: 465-493 Pettersson LG, Lundberg A, Alstermark B, Isa T, Ta ntisira B (1997) Effect of spinal cord lesions on forelimb target-reaching and on visually guid ed switching of target-reaching in the cat. Neurosci Res 29: 241-256 Pindzola RR, Doller C, Silver J (1993) Putative inhibitory extracellular matrix molecules at the dorsal root entry zone of the spinal cord during developmen t and after root and sciatic nerve lesions. Dev Biol 156: 34-48 Pompeiano O, Brodal A (1957) Experimental de monstration of a somatotopical origin of rubrospinal fibers in the cat. J Comp Neurol 108: 225-251 Powers SK, Coombes J, Demirel H (1997) Exerci se training-induced changes in respiratory muscles. Sports Med 24: 120-131 Powers SK, Grinton S, Lawler J, Criswell D, Dodd S (1992) High intens ity exercise traininginduced metabolic alterations in respirat ory muscles. Respir Physiol 89: 169-177 Prinjha R, Moore SE, Vinson M, Blake S, Morrow R, Christie G, Michalovich D, Simmons DL, Walsh FS (2000) Inhibitor of neurite out growth in humans. Nature 403: 383-384 Properzi F, Asher RA, Fawcett JW (2003) Chondroi tin sulphate proteoglycans in the central nervous system: changes and synthesis af ter injury. Biochem Soc Trans 31: 335-336 Rabchevsky AG, Streit WJ (1997) Grafting of cultured microglial cells into the lesioned spinal cord of adult rats enhances neurit e outgrowth. J Neurosci Res 47: 34-48 Raisman G (1969) Neuronal plasticity in the sept al nuclei of the adult rat. Brain Res 14: 25-48 Reier PJ, Houle JD (1988) The glial scar: its bearing on axonal elongation and transplantation approaches to CNS repair. Adv Neurol 47: 87-138
123 Rhodes KE, Fawcett JW (2004) Chondroitin sulphate proteoglycans: preventing plasticity or protecting the CNS? J Anat 204: 33-48 Richardson PM, Issa VM, Aguayo AJ (1984) Regene ration of long spinal axons in the rat. J Neurocytol 13: 165-182 Richter DW, Heyde F, Gabriel M (1975) Intrace llular recordings from different types of medullary respiratory neurons of the cat. J Neurophysiol 38: 1162-1171 Romaniuk JR, Kasicki S, Kazennikov OV, Se lionov VA (1994) Respiratory responses to stimulation of spinal or medulla ry locomotor structures in de cerebrate cats. Acta Neurobiol Exp (Wars) 54: 11-17 Rossignol S, Chau C, Brustein E, Belanger M, Barbeau H, Drew T (1996) Locomotor capacities after complete and partial lesions of the spin al cord. Acta Neurobiol Exp (Wars) 56: 449463 Rossignol S, Drew T, Brustein E, Jiang W ( 1999) Locomotor performan ce and adaptation after partial or complete spinal cord lesi ons in the cat. Prog Brain Res 123: 349-365 Rossignol S, Dubuc R (1994) Spinal pattern generation. Curr Opin Neurobiol 4: 894-902 Ruitenberg MJ, Plant GW, Hamers FP, Wortel J, Blits B, Dijkhuizen PA, Gispen WH, Boer GJ, Verhaagen J (2003) Ex vivo adenoviral vecto r-mediated neurotrophin gene transfer to olfactory ensheathing glia: effects on rubr ospinal tract regeneration, lesion size, and functional recovery after implantation in the injured rat spinal cor d. J Neurosci 23: 70457058 Saatman KE, Graham DI, McIntosh TK (1998) The neuronal cytoskeleton is at risk after mild and moderate brain injur y. J Neurotrauma 15: 1047-1058 Schmalfeldt M, Bandtlow CE, Dours-Zimmer mann MT, Winterhalter KH, Zimmermann DR (2000) Brain derived versican V2 is a potent inhibitor of axon al growth. J Cell Sci 113 ( Pt 5): 807-816 Schmid K, Kirkwood PA, Munson JB, Shen E, S ears TA (1993) Contrala teral projections of thoracic respiratory interneurones in the cat. J Ph ysiol 461: 647-665 Schnell L, Schwab ME (1990) Axonal regenerati on in the rat spinal cord produced by an antibody against myelin-associated neurit e growth inhibitors. Nature 343: 269-272 Schoen JH (1964) Comparative Aspects of the De scending Fibre Systems in the Spinal Cord. Prog Brain Res 11: 203-222
124 Schrimsher GW, Reier PJ (1993) Forelimb motor performance following dorsal column, dorsolateral funiculi, or ventrola teral funiculi lesions of the cervical spinal cord in the rat. Exp Neurol 120: 264-276 Schumacher PA, Eubanks JH, Fehlings MG (1999) Increased calpain I-media ted proteolysis, and preferential loss of dephosphor ylated NF200, following traumatic spinal cord injury. Neuroscience 91: 733-744 Schwab ME, Bartholdi D (1996) De generation and regeneration of a xons in the lesioned spinal cord. Physiol Rev 76: 319-370 Schwab ME, Caroni P (1988) Oligodendrocytes and CNS myelin are nonpe rmissive substrates for neurite growth and fibroblast spr eading in vitro. J Neurosci 8: 2381-2393 Schwab ME, Thoenen H (1985) Dissociated neurons regenerate into sciati c but not optic nerve explants in culture irrespective of neur otrophic factors. J Neurosci 5: 2415-2423 Sekhon LH, Fehlings MG (2001) Epidemiology, de mographics, and pathophysiology of acute spinal cord injury. Spine 26: S2-12 Selionov VA, Shik ML (1984) Medullary locomotor strip and column in the cat. Neuroscience 13: 1267-1278 Shik ML, Orlovsky GN (1976) Neurophysiology of locomotor automatism. Physiol Rev 56: 465501 Silver J, Miller JH (2004) Regeneration beyond the glial scar. Nat Rev Neurosci 5: 146-156 Sinkjaer T, Miller L, Andersen T, Houk JC ( 1995) Synaptic linkages between red nucleus cells and limb muscles during a multi-joint motor task. Exp Brain Res 102: 546-550 Snow DM, Lemmon V, Carrino DA, Caplan AI, S ilver J (1990a) Sulfated proteoglycans in astroglial barriers inhibit neurite outgr owth in vitro. Exp Neurol 109: 111-130 Snow DM, Mullins N, Hynds DL (2001) Nerv ous system-derived chondroitin sulfate proteoglycans regulate growth cone morphology and inhibit neurite outgrowth: a light, epifluorescence, and electron micros copy study. Microsc Res Tech 54: 273-286 Snow DM, Steindler DA, Silver J (1990b) Molecu lar and cellular characte rization of the glial roof plate of the spinal cord and optic tectum: a possible role for a proteoglycan in the development of an axon barrier. Dev Biol 138: 359-376 Sribnick EA, Wingrave JM, Matzelle DD, Ray SK, Banik NL (2003) Estrogen as a neuroprotective agent in the tr eatment of spinal cord inju ry. Ann N Y Acad Sci 993: 125133; discussion 159-160
125 Steward O, Zheng B, Tessier-Lavigne M, Hofsta dter M, Sharp K, Yee KM (2008) Regenerative growth of corticospinal tract axons via the vent ral column after spinal cord injury in mice. J Neurosci 28: 6836-6847 Streit WJ (1996) The role of microglia in brain injury. Neurotoxicology 17: 671-678 Tator CH (1998) Biology of neur ological recovery and functional restoration after spinal cord injury. Neurosurgery 42: 696-707; discussion 707-698 Tester NJ, Howland DR (2007) Chondroitinase ABC improves basic and skilled locomotion in spinal cord injured cats. Exp Neurol Tester NJ, Howland DR (2008) Chondroitinase ABC improves basic and skilled locomotion in spinal cord injured cats Exp Neurol 209: 483-496 Tester NJ, Plaas AH, Howland DR (2007) Effect of body temperature on chondroitinase ABC's ability to cleave chondroitin sulfate glyc osaminoglycans. J Neurosci Res 85: 1110-1118 Tetzlaff W, Kobayashi NR, Giehl KM, Tsui BJ Cassar SL, Bedard AM (1994) Response of rubrospinal and corticospinal neurons to in jury and neurotrophins. Prog Brain Res 103: 271-286 Thomas SL, Gorassini MA (2005) Increases in corticospinal tract function by treadmill training after incomplete spinal cord injury. J Neurophysiol 94: 2844-2855 Tobias CA, Shumsky JS, Shibata M, Tuszynski MH, Fischer I, Tessl er A, Murray M (2003) Delayed grafting of BDNF and NT-3 producing fibroblasts into the injured spinal cord stimulates sprouting, partially rescues axot omized red nucleus neurons from loss and atrophy, and provides limited rege neration. Exp Neurol 184: 97-113 Tomori Z, Widdicombe JG (1969) Muscular, bronchomotor and cardi ovascular reflexes elicited by mechanical stimulation of the respiratory tract. J Physiol 200: 25-49 Tropea D, Caleo M, Maffei L (2003) Synergistic e ffects of brain-derived neurotrophic factor and chondroitinase ABC on retinal fiber sprouting afte r denervation of the superior colliculus in adult rats. J Neurosci 23: 7034-7044 Uribe JM, Stump CS, Tipton CM, Fregosi RF (1 992) Influence of exercise training on the oxidative capacity of rat abdominal muscles. Respir Physiol 88: 171-180 Van Houtte S, Vanlandewijck Y, Gosselink R (2 006) Respiratory muscle training in persons with spinal cord injury: a system atic review. Respir Med 100: 1886-1895 van Kan PL, McCurdy ML (2001) Role of pr imate magnocellular red nucleus neurons in controlling hand preshaping during reachi ng to grasp. J Neurophysiol 85: 1461-1478
126 Verhaart WJ (1953) The fiber structure of cord in the cat. Acta An at (Basel) 18: 88-100 Verhaart WJ (1955) The rubrospina l tract in the cat, the monkey and the ape, its location and fibre content. Monatsschr Psychiatr Neurol 129: 487-500 von Euler M, Janson AM, Larsen JO, Seiger A, Forno L, Bunge MB, Sundstrom E (2002) Spontaneous axonal regeneration in rodent spinal cord after ischemic injury. J Neuropathol Exp Neurol 61: 64-75 Webb AA, Muir GD (2002) Compensatory locomotor adjustments of rats with cervical or thoracic spinal cord hemis ections. J Neurotrauma 19: 239-256 Weidner N, Ner A, Salimi N, Tuszynski MH (2 001) Spontaneous cortic ospinal axonal plasticity and functional recovery after a dult central nervous system injury. Proc Natl Acad Sci U S A 98: 3513-3518 Whelan PJ (1996) Control of locomotion in th e decerebrate cat. Prog Neurobiol 49: 481-515 Whishaw IQ, Coles BL, Pellis SM, Miklyaeva EI (1997a) Impairments and compensation in mouth and limb use in free feeding after unilate ral dopamine depletions in a rat analog of human Parkinson's disease. Behav Brain Res 84: 167-177 Whishaw IQ, Gorny B (1996) Does the red nucle us provide the tonic support against which fractionated movements occur? A study on fore paw movements used in skilled reaching by the rat. Behav Brain Res 74: 79-90 Whishaw IQ, Gorny B, Sarna J (1998) Paw and limb use in skilled and spontaneous reaching after pyramidal tract, red nucleus and combin ed lesions in the rat: behavioral and anatomical dissociations. Behav Brain Res 93: 167-183 Whishaw IQ, Woodward NC, Miklyaeva E, Pellis SM (1997b) Analysis of limb use by control rats and unilateral DA-depleted rats in the Montoya staircase test: movements, impairments and compensatory stra tegies. Behav Brain Res 89: 167-177 Widajewicz W, Kably B, Drew T (1994) Motor cortical activity during voluntary gait modifications in the cat. II. Cells relate d to the hindlimbs. J Neurophysiol 72: 2070-2089 Windle WF, Chambers WW (1950) Regeneration in the spinal cord of the cat and dog. J Comp Neurol 93: 241-257 Xiao M, Klueber KM, Guo Z, Lu C, Wang H, Ro isen FJ (2007) Human adult olfactory neural progenitors promote axotomized rubrospina l tract axonal reinnerv ation and locomotor recovery. Neurobiol Dis 26: 363-374
127 Xu XM, Guenard V, Kleitman N, Aebischer P, Bunge MB (1995) A combination of BDNF and NT-3 promotes supraspinal axonal regenerati on into Schwann cell gr afts in adult rat thoracic spinal cord. Exp Neurol 134: 261-272 Yamada H, Fredette B, Shitara K, Hagihara K, Miura R, Ranscht B, Stallcup WB, Yamaguchi Y (1997) The brain chondroitin sulfate proteoglycan brevican associates with astrocytes ensheathing cerebellar glomeruli and inhibits neurite outgro wth from granule neurons. J Neurosci 17: 7784-7795 Yamagata T, Saito H, Habuchi O, Suzuki S ( 1968) Purification and properties of bacterial chondroitinases and chondrosulfa tases. J Biol Chem 243: 1523-1535 Yamaguchi Y (2000) Lecticans: organizers of th e brain extracellular matrix. Cell Mol Life Sci 57: 276-289 Yamaguchi Y (2001) Heparan sulfate proteoglycans in the nervous system: their diverse roles in neurogenesis, axon guidance, and synaptoge nesis. Semin Cell Dev Biol 12: 99-106 Yick LW, Cheung PT, So KF, Wu W (2003) A xonal regeneration of Clarke's neurons beyond the spinal cord injury scar after treatment with chondroiti nase ABC. Exp Neurol 182: 160168 Yick LW, So KF, Cheung PT, Wu WT (2004) L ithium chloride reinforces the regenerationpromoting effect of chondroiti nase ABC on rubrospinal neurons after spinal cord injury. J Neurotrauma 21: 932-943 Yick LW, Wu W, So KF, Yip HK, Shum DK (2000) Chondroitinase ABC promotes axonal regeneration of Clarke's neurons after spinal cord injury. Neuroreport 11: 1063-1067 Ying Z, Roy RR, Edgerton VR, Gomez-Pinilla F (2003) Voluntary exercise increases neurotrophin-3 and its receptor TrkC in the spinal cord. Brain Res 987: 93-99 Ying Z, Roy RR, Zhong H, Zdunowski S, Edgert on VR, Gomez-Pinilla F (2008) BDNF-exercise interactions in the recovery of symmetrical stepping after a cervical hemisection in rats. Neuroscience 155: 1070-1078 Young W (1991) Methylprednisolone treatment of acute spinal cord inju ry: an introduction. J Neurotrauma 8 Suppl 1: S43-46 Young W, Koreh I (1986) Potassiu m and calcium changes in injured spinal cords. Brain Res 365: 42-53 Yu TW, Bargmann CI (2001) Dyna mic regulation of axon guida nce. Nat Neurosci 4 Suppl: 1169-1176
128 BIOGRAPHICAL SKETCH Stephanie Christine Jefferson was born in Bellevue, W ashington, to Rawle and Janet Jefferson, in the summer of 1981. She had the privilege of growing up on the beautiful, and quaint Bainbridge Island, which is located only a 35 minute ferry-ride away from the Emerald City, Seattle. She graduated in 1999 in the top of her class from Bainbridge High School in Bainbridge Island, Washi ngton. She then attended Sweet Briar College in Sweet Briar, Virginia, where she received her B.S. in biology, magna cum laude. During her undergraduate education she was very active in the Sweet Briar academ ic community. She was inducted into Alpha Lambda Delta, the national honors society that honors freshman achieving academic excellence, and then served as the organizations presid ent the following year. She was also proudly inducted into Iota Sigma Pi, the national honors society for women in chemistry, as well as Phi Beta Kappa, the oldest undergraduate honors organi zation in the United States. She also served as a biology teaching assistant during most of her undergraduate education, which ignited her passion for scientific research. Dr. Linda Fink, a biology professor at Sweet Briar College and her academic advisor, provided Stephanie with much needed guidance, support, and friendship that aided in her decision to pursue a career in the biological sciences. Stephanie participated in multiple summer undergraduate research programs dur ing her time at Sweet Briar College that ranged in breadth from molecular cloning of a gene responsible for retinal degeneration to induction of insulin production in type 1 diabetic patients. She found he r niche in scientific research, and in August 2003 entere d into the Interdis ciplinary Program (IDP) in Biomedical Sciences at the University of Florida. This in terdisciplinary program allo wed her to experience a vast array of scientific research across a broad spectrum of discipli nes, and eventua lly led her to the fascinating field of neuroscience. In th e Spring of 2004, she joined the laboratory of Dr. Dena R. Howland where she conducted laborator y research to understand the basic pathobiology
129 and inhibitory substrates that halt neural regeneration and func tional recovery following spinal cord injury. She worked directly to associate evidence of regeneration with behavioral motor changes following spinal cord injury in a comp lex model that is translatable to clinical applications. Specifically, her dissertation research focused on the effects of intraspinal delivery of the bacterial enzyme Chondroitinase ABC in the cat model following low thoracic spinal hemisection. She assessed the effects of th is therapeutic treatment on two motor systems, locomotion and the cough reflex, and correlated thei r enhancements with anatomical plasticity at the spinal and supraspinal level. Her dissertation resear ch will be published in three first author manuscripts that will hopefully enhance the fiel d of neuroscience and in particular neuronal plasticity. She is grateful for all of the knowledge and wisdom she has acquired along this journey and wants to thank each individual who helped her achieve her goals.