<%BANNER%>

Descending Medullary Projections to the Phrenic Motoneuron Pool after High Cervical Spinal Hemisection

Permanent Link: http://ufdc.ufl.edu/UFE0022696/00001

Material Information

Title: Descending Medullary Projections to the Phrenic Motoneuron Pool after High Cervical Spinal Hemisection
Physical Description: 1 online resource (64 p.)
Language: english
Creator: Coutts, Marcella
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Medicine -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Neuroplasticity, which is an experience-dependent change in neural control, can occur in response to many stimuli and via many mechanisms throughout the peripheral and central nervous systems. Specifically, the respiratory system exhibits a remarkable capacity for plasticity-related changes. One extensively documented instance of respiratory neuroplasticity is the crossed phrenic phenomenon (CPP), which typically occurs after a high cervical spinal cord injury (SCI). The CPP involves the activation of a previously latent synaptic pathway that crosses the spinal midline to restore some activity to the phrenic motoneurons (PhMNs), which are denervated by SCI. Although much progress has been made in identifying the neural substrate that underlies the CPP, the current understanding of this instance of respiratory neuroplasticity is far from complete. In fact, few studies have examined the CPP and the neurons which drive diaphragm motor function at the level of the brainstem. Accordingly, the present study compares the phrenic-associated brainstem labeling that is observed when pseudorabies virus (PRV), a retrograde transneuronal tracer, is topically applied to the diaphragm in normal and C2-hemisected rats. It was found that the medial reticular formation (MRt) exhibits dominant connectivity to the PhMN pool both in the normal and injured animal. In a number of species the MRt is involved in various behaviors which require alterations in breathing. Therefore, the function, connectivity, and relative stability of these cells after cervical hemisection suggests that they may be influential in modulating phrenic function both pre and post-injury.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Marcella Coutts.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Reier, Paul J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022696:00001

Permanent Link: http://ufdc.ufl.edu/UFE0022696/00001

Material Information

Title: Descending Medullary Projections to the Phrenic Motoneuron Pool after High Cervical Spinal Hemisection
Physical Description: 1 online resource (64 p.)
Language: english
Creator: Coutts, Marcella
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: Medicine -- Dissertations, Academic -- UF
Genre: Medical Sciences thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Neuroplasticity, which is an experience-dependent change in neural control, can occur in response to many stimuli and via many mechanisms throughout the peripheral and central nervous systems. Specifically, the respiratory system exhibits a remarkable capacity for plasticity-related changes. One extensively documented instance of respiratory neuroplasticity is the crossed phrenic phenomenon (CPP), which typically occurs after a high cervical spinal cord injury (SCI). The CPP involves the activation of a previously latent synaptic pathway that crosses the spinal midline to restore some activity to the phrenic motoneurons (PhMNs), which are denervated by SCI. Although much progress has been made in identifying the neural substrate that underlies the CPP, the current understanding of this instance of respiratory neuroplasticity is far from complete. In fact, few studies have examined the CPP and the neurons which drive diaphragm motor function at the level of the brainstem. Accordingly, the present study compares the phrenic-associated brainstem labeling that is observed when pseudorabies virus (PRV), a retrograde transneuronal tracer, is topically applied to the diaphragm in normal and C2-hemisected rats. It was found that the medial reticular formation (MRt) exhibits dominant connectivity to the PhMN pool both in the normal and injured animal. In a number of species the MRt is involved in various behaviors which require alterations in breathing. Therefore, the function, connectivity, and relative stability of these cells after cervical hemisection suggests that they may be influential in modulating phrenic function both pre and post-injury.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Marcella Coutts.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Reier, Paul J.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2009-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0022696:00001


This item has the following downloads:


Full Text

PAGE 1

DESCE NDING MEDULLARY PROJECTIONS TO THE PHRENIC MOTONEURON POOL AFTER HIGH CERVICAL SPINAL HEMISECTION By MARCELLA ANGELIQUE COUTTS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Marcella Angelique Coutts 2

PAGE 3

To Dante. Thank you for all of the love and support. 3

PAGE 4

ACKNOWL EDGMENTS I would like to thank my mentor, Dr. Paul Reie r, for all of his hel p. In addition, I would like to thank the members of the Reier lab: Dr Michael Lane, Dr. Todd White, Kevin Siegel, and Barbara OSteen, all of whom continually offered their support and guidance. 4

PAGE 5

TABLE OF CONTENTS page ACKNOWLEDGMENTS ...............................................................................................................4 LIST OF TABLES ...........................................................................................................................7 LIST OF FIGURES .........................................................................................................................8 LIST OF ABBREVIATIONS ..........................................................................................................9 ABSTRACT ...................................................................................................................................11 CHAP TER 1 INTRODUCTION................................................................................................................. .13 Overview of Neuroplasticity ...................................................................................................13 Models of Respiratory Neuroplasticity ...................................................................................13 High Cervical Spinal Cord Injury as a C linically Relevant Model of Injury-Induced Respiratory Neuroplasticity ................................................................................................15 Current Understanding of the Neural Circuitry Associated with the Crossed Phrenic Phenom enon ........................................................................................................................16 Previous Studies on Brainstem Phrenic Circuitry ..................................................................18 Overview of Thesis Study ......................................................................................................19 2 MATERIALS AND METHODS...........................................................................................21 Animals ...................................................................................................................................21 Pseudorabies Virus .................................................................................................................21 Surgical Procedures ................................................................................................................22 Anesthesia and Post-Operative Care ...............................................................................22 Spinal Cord Injury ...........................................................................................................22 Retrograde Neuroanatomical Tracing .............................................................................23 Perfusion and Tissue Sectioning .............................................................................................23 Immunohistochemistry ...........................................................................................................23 Quantitative Analyses .............................................................................................................25 3 RESULTS...................................................................................................................... .........26 Pseudorabies Virus Infection of the Cervical Spinal Cord .....................................................26 Quantitative Analysis of Brainstem Labeling in Uninjured Animals .....................................27 Quantitative Analyses at the 48 and 56 Hour Post-Infection Interval .............................28 Quantitative Analyses at the 64 Hour Post-Infection Interval .........................................29 Quantitative Analysis of Brainstem Labeling in Injured Animals .........................................31 5

PAGE 6

4 DISCUSSION................................................................................................................... ......45 Technical Considerations ........................................................................................................45 Pseudorabies Virus as a Retr ograde Transynaptic Tracer ...............................................45 Variability and Normalization of Brainstem Cell Counts ...............................................47 Pattern of Brainstem Neurons After Spinal Cord Injury ........................................................48 Prevalence of Crossed Projections to PhMNs .................................................................48 Persistent Regional Brainstem La beling After Spinal Cord Injury .................................50 Considerations when Interpreting Brains tem Labeling After Spinal Cord Injury ..........50 Normal Pattern of Brainstem Neurons ...................................................................................51 Rostral Ventral Respiratory Group Projections to PhMNs .............................................52 Medial Reticular Formation Projections to PhMNs ........................................................53 Previous Evidence for Importance of the Medial Reticular F ormation in the Phrenic Respiratory Circuit ..............................................................................................................54 Functional Implications of the Medial Reticu lar Formation in Respiration ...........................55 Summary .................................................................................................................................58 APPENDIX LIST OF SOLUTIONS...........................................................................................................59 LIST OF REFERENCES ...............................................................................................................60 BIOGRAPHICAL SKETCH .........................................................................................................64 6

PAGE 7

LIST OF TABLES Table page 3-1 Percentage of total labeling in brainstem in each region at 56 hours ................................35 3-2 Percentage of total labeling in brainstem in each region at 64 hours ................................38 3-3 Total number of labeled cells in each si de of the brainstem per animal at 64 hours post-infection .....................................................................................................................41 3-4 Mean percentage of labeling in brainste m in each region at 64 hours (left side/right side) ....................................................................................................................................42 7

PAGE 8

LIST OF FI GURES Figure page 3-1 PRV-infected PhMNs are specifically labeled in the ce rvical spinal cord.. ......................33 3-2 Brainstem labeling in the Medial Reticular Form ation.. ....................................................34 3-5 64 hours uninjured most labeled brainstem regions .......................................................39 3-6 64 hours uninjured regions of the MRt ...........................................................................40 3-7 64 hours uninjured and injured percent of total labeling on left brainstem ....................43 3-8 64 hours uninjured and injured percen t of total labeling on right brainstem ..................44 8

PAGE 9

LIST OF ABBRE VIATIONS A5 A5 Noradrenaline Cells ABC Avidin-biotin Complex CNS Central Nervous System CPP Crossed Phrenic Phenomenon CVL Caudal Ventrolateral Reticular Nucleus DAB Diaminobenzidine DPGi Dorsal Paragigantocellular Nucleus FG Fluorogold Gi Gigantocellular Reticular Nucleus GiA Gigantocellular Reticu lar Nucleus, Alpha Part GiV Gigantocellular Reticula r Nucleus, Ventral Part i.p. Intraperitoneal IRt Intermediate Reticular Nucleus LTF Long-term Facillitation LTP Long-term Potentiation LPGi Lateral Paragigantocellular Nucleus MRt Medial Reticular Formation NGS Normal Goat Serum NSCISC National Spinal Cord Injury Statistical Center PBS Phosphate Buffered Saline pfu Plaque Forming Units PhMN Phrenic Motoneuron PK15 Porcine Kidney PRV Pseudorabies Virus 9

PAGE 10

10 REM Rapid Eye Movement RMg Raphe Magnus Nucleus RO Raphe Obscurus Nucleus RPa Raphe Pallidus Nucleus RVL Rostroventrolateral Reticular Nucleus rVRG Rostral Ventral Respiratory Group SCI Spinal Cord Injury s.q Subcutaneous SubC Subcoeruleus Nucleus VRC Ventral Respiratory Column WGA-HRP Wheatgerm Agglutinin Conj ugated to Horseradish Peroxidas

PAGE 11

Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science DESCENDING MEDULLARY PROJECTIONS TO THE PHRENIC MOTONEURON POOL AFTER HIGH CERVICAL SPINAL HEMISECTION By Marcella Angelique Coutts August 2008 Chair: Paul J. Reier Major: Medical Sciences Neuroplasticity, which is an experience-dependent change in neural control, can occur in response to many stimuli and via many mechanis ms throughout the peripheral and central nervous systems. Specifically, the respiratory system exhibits a remarkable capacity for plasticity-related changes. One extensively docume nted instance of respiratory neuroplasticity is the crossed phrenic phenomenon (C PP), which typically occurs afte r a high cervical spinal cord injury (SCI). The CPP involves the activation of a pr eviously latent syna ptic pathway that crosses the spinal midline to restore some activ ity to the phrenic motoneurons (PhMNs), which are denervated by SCI. Although much progress has been made in identifying the neural substrate that underlies the CPP, the current unde rstanding of this inst ance of respiratory neuroplasticity is far from comp lete. In fact, few studies have examined the CPP and the neurons which drive diaphragm motor function at the leve l of the brainstem. Accordingly, the present study compares the phrenic-associ ated brainstem labeling that is observed when pseudorabies virus (PRV), a retrograde transneuronal tracer, is topically applied to the diaphragm in normal and C2-hemisected rats. It wa s found that the medial reticu lar formation (MRt) exhibits dominant connectivity to the Ph MN pool both in the normal and injured animal. In a number of 11

PAGE 12

12 species the MRt is involved in various beha viors which require alterations in breathing. Therefore, the function, connectivity, and rela tive stability of thes e cells after cervical hemisection suggests that they may be influent ial in modulating phrenic function both pre and postinjury.

PAGE 13

CHAP TER 1 INTRODUCTION Overview of Neuroplasticity The term plasticity or neuroplasticity refers to a lasting change in a given neural control system as a result of prior experiences. The capacity for plasticity is inherent in the nervous system and can be manifested in response to numerous experiences or stimuli, including injury, disease, or aging (Mitchell and John son, 2003; Lane et al., 2008a; Nudo, 2007). Plasticity can occur at numerous levels with in a neuronal circuit by a variety of methods. For instance, changes in the ac tivity level of a synapse can a lter the subsequent efficacy of synaptic transmission. One example is known as long-term potentiation (LTP), whereby high frequency stimulation of a neuron results in a prolonged increase in synaptic efficacy. Other possible morphological alterations include change s in the size or shape of neurons and their processes (Mitchell and Johnson, 2003). In addition to changes at the cellular level, plasticity-related changes can also occur at higher organizational levels, possi bly affecting entire neural netw orks. For instance, experiencedependent changes are known to occur in the concentration or profile of neurotransmitters that modulate entire neural systems. Plasticity can also be manifest ed as the emergence of novel characteristics of neural networks. For instance, after a plasticity-inducin g experience, a network of neurons could exhibit greater synchrony in their firing patte rn (Mitchell and Johnson, 2003). Thus, neuroplasticity is the nervous systems re markable ability to adapt in response to novel stimuli. Models of Respiratory Neuroplasticity Although the entire nervous system has the potential for pl asticity, the focus of the present work is on the respiratory system. Numer ous experiences, such as hypoxia or oxygen 13

PAGE 14

deprivation, have been found to i nduce plasticity in the neural sy stem s which control respiration (Mitchell and Johnson, 2003; Goshgarian, 2003). For instance, intermittent exposure to hypoxic conditions causes a sustained increase in the firi ng rates of the phrenic motoneurons (PhMNs), which control diaphragm contractions during in spiration. This increase in respiratory motor output is an instance of neuroplasticity referred to as long-term facilitation (LTF) and results in a prolonged increase in breathing fr equency and tidal volume (i.e., the volume inspired or expired per breath) (Baker and Mitchell, 2000). Another activity that can induce neuroplasticity in the respiratory motor system is exercise. In particular, repeated exercise that consisten tly induces hypercapnia (i.e ., increased levels of carbon dioxide in the blood) re sults in a sustained ability to increase ventilation during subsequent exercise. This phenomenon is known as long-term modulation (Martin and Mitchell, 1993) and has been shown to occur via serotoni n-dependent mechanisms (Johnson and Mitchell, 2001). Many other instances of respiratory neuroplas ticity have been documented. However, the most commonly investigated instance of respir atory neuroplasticity th at has been observed across a variety of species is known as the crossed phrenic phenomenon (CPP) (Goshgarian, 2003). As will be discussed in more detail below, the CPP refers to a recovery of activity to denervated PhMNs via the activation of previously latent synaptic pathways that cross the spinal midline (Goshgarian, 2003; Johnson and Mitchell, 2001). This mechanism of plasticity has been demonstrated experimentally by a lateral spinal cord hemisection above th e third cervical level. Thus, before discussing the CPP in detail, it is necessary to explain why a high cervical SCI represents a clinically relevant model for respiratory neuroplasticity. 14

PAGE 15

High Cervical Spinal Cord Injury as a Clin ically Relev ant Model of Injury-Induced Respiratory Neuroplasticity Traumatic injury to the cervical spinal cord is the most common type of SCI and accounts for approximately 52.4 % of all reported cases of SCI (National Spinal Co rd Injury Statistical Center (NSCISC), 2006). An injury to one of the ei ght segments of the cervical spinal cord can produce a variety of life-threaten ing functional deficits which ma y include tetraplegia, sensory and motor losses below the level of injury, ca rdiovascular problems, autonomic dysreflexia, impaired bladder and bowel control, and resp iratory dysfunction. Due to the broad range of morbidity that is associated with injury to the cervical spinal cord, it is no surprise that the highest levels of medical expend itures for all types of SCI are li nked to this type of injury (NSCISC, 2006). However, even with the multitude of symptoms associated with cervical SCI, it is the respiratory complications that are most debilitating and potentially life-threatening. In fact, respiratory dysfunction and associated secondary complications represent the leading cause of morbidity and mortality after SCI at any level (Brown et al., 2006; Jackson and Groomes, 1994; NSCISC, 2006). This comes as no surprise, since the neurona l connections that control inspiration are severed by a high cervical SCI. As the PhMNs that control the contraction of the diaphragm (i.e., the primary muscle of inspiration) are located at C3-C5, individuals w ho sustain a high (i.e., rostral to C3-4) cervical SCI may experience hemi diaphragm paresis ipsilateral to the injury due to denervation of the PhMNs. Consequently, such an injury is characterized by severe respiratory insufficiency and the development of a compen satory rapid, shallow breathing pattern, often causing patients to require the support of mechanical ventil ation (Brown et al., 2006). However, despite the severe consequences of denervation associated with this type of injury, there is the potential for neuroplasticity and functional recovery in this system via the 15

PAGE 16

CPP. Although m uch of the evidence documenti ng the CPP after high cervical SCI has been demonstrated in animal models, it is tempting to speculate that some respiratory neuroplasticity may occur in humans after SCI. In fact, human SCI patients have shown some restoration of descending neural drive to the di aphragm that allows for modest recovery of respiratory function (Brown et al., 2006). Therefore, studying the m echanism underlying the CPP may lead to a better understanding of how respiratory ne uroplasticity can be exploited c linically in SCI patients. For these reasons, a thorough understanding of the mechanisms related to neuroplasticity and recovery of respiratory functioning after high cervi cal SCI via the CPP is of utmost importance. Current Understanding of the Neural Circuit ry Associated with the Crossed Phrenic Phenomenon One of the first historical acc ounts of the CPP and the associat ed restoration of respiratory function was described by Porter (1895). In Porters study, hemisec tions were performed rostral to the level of the PhMN pool in dogs and ra bbits to induce paralysi s in the ipsilateral hemidiaphragm. It was observed post-hemisection that subsequent transe ction of the phrenic nerve (i.e., a phrenicotomy) contrala teral to injury resulted in th e restoration of some diaphragm function on the side of injury. Since then, the CPP has been demonstrated in a variety of species after C2 hemisection of the spinal cord (Goshgarian, 2003; Lane et al., 2008a). In the rat, PhMNs receive descending inspirat ory drive via axons originating from neurons in a region of the medulla called the ventral respiratory column (VRC) or rostral ventral respiratory group (rVRG) (Goshgarian, 2003). Prev ious neuroanatomical studies have shown that rVRG projections on either side of the spinal cord descend from cells on both sides of the medulla via a brainstem decussation and termin ate monosynaptically on PhMNs (Ellenberger et al., 1990; Moreno et al, 1992). Thes e PhMNs control diaphragm activ ity via the phrenic nerves. 16

PAGE 17

Collectively, this neuronal network that controls inspiration is known as the phrenic respiratory circu it. After C2 hemisection, the descending medullary fibers in the phrenic respiratory circuit are interrupted. As a result, the ipsilateral PhMNs are denervated and beco me functionally silent, causing hemiparalysis of the ipsilateral diaphrag m. However, despite this denervation, some functional recovery of ipsilateral diaphragm f unction and PhMN activity occurs after injury. Such recovery has been credited to the CPP, wh ich involves the activation of a previously latent pathway that originates from medullary bulbospina l axons that cross the spinal midline at the level of the PhMNs to terminate monosynapt ically on ipsilateral PhMNs (Goshgarian, 2003; Goshgarian et al., 1991; Moreno et al., 1992). The CPP can evolve either spontaneously (Fuller et al., 2006; Fuller et al., 2003; Go lder and Mitchell, 2005; Golder et al., 2001b; Nantwi et al., 1999) or be induced under controlled, termin al neurophysiological c onditions by subsequent transection of the contrala teral phrenic nerve (Porte r, 1895; Goshgarian, 2003). Numerous studies have attempted to further define the neuroanatomical basis for the CPP in the rat. However, these studies have largely focused on the phren ic circuit at the level of the PhMN pool (Lane et al., 2008b; Go shgarian et al., 1991). Even the few studies which have ventured to look at the brainstem after hemisectio n have limited their examination of supraspinal phrenic projections involved in the CPP to the VRC (Boulenguez et al., 2007; Moreno et al., 1992). In contrast, only one study has attempted to examine the complete profile of brainstem neurons involved in the phrenic respiratory circuit; however this study was restricted to uninjured animals and did not exam ine the potential for neuroplasticity in the circuit after injury (Dobbins and Feldman, 1994). Therefore, to co nstruct a more complete picture of the mechanism behind the CPP in the respiratory system, the experiments performed in this thesis 17

PAGE 18

will exam ine the entire profile of pre-phrenic brainstem neurons involved in the CPP in the rat after a C2 hemisection and in the process will also reexamine the normal supraspinal phrenic respiratory circuit. Previous Studies on Brainstem Phrenic Circuitry As mentioned above, there have been very few neuroanatomical studies examining the changes in supraspinal projecti ons to the phrenic nucleus after C2 hemisection. In one such study, Boulenguez et al. (2007) atte mpted to quantify the extent of projections from the rVRG that cross at the level of the PhMNs and are like ly to be involved in the CPP. In that study, injections of fluorogold (FG), a monosynaptic retrograde tracer, were made directly into the spinal cord below a C2 hemisection (at C3-C 4). Because injections were made below the hemisection into the denervated PhMN pool, th e labeled neurons in th e brainstem could only become labeled if their axons crossed the sp inal midline. Boulengu ez showed that, when compared to control animals, injured animals s howed reduced labeling bilaterally in the rVRG and that labeling was particularly sparse on the side ipsilateral to injury. Comparison of FGlabeled rVRG neuron counts in injured versus co ntrol animals showed th at after hemisection, labeling was reduced to only 23 % ipsilaterally and 36 % contralate rally of the labeling observed in control animals. The authors fe lt that this subpopulat ion of cells that cros s the spinal midline represented the neuroanatomical substrate for the CPP. Although this study examined supraspinal projections from the rVRG to PhMN s involved in the CPP, it failed to yield any information regarding the broade r neuronal network which controls PhMN activity in the injured animal. One other study has directly examined the brainstem neuronal network which controls PhMNs (Dobbins and Feldman, 1994). In that series of experiments, the authors injected the Bartha strain of Pseudorabies Virus (PRV) into phrenic nerves of normal animals. PRV-Bartha 18

PAGE 19

has been repeatedly found to produce viral infecti ons in an exclusively re trograde direction (i.e., from axon terminal to cell body) (Card et al., 1998) Thus, infection with this neurotropic virus resulted in very specific labeling of the PhMNs, thereby suggesting that the PRV-labeled cells in the brainstem were restricted to neurons in the phrenic resp iratory circuit. Also, PRV produces temporal waves of infecti on in a manner that is consistent with the pattern of innervation in a circuit (Lowey, 1998). Thus, because numerous regions of the brainstem showed labeling, it was evident that a widespread netw ork of cells in the brainstem was responsible for motor control of the diaphr agm. However, despite labeling in numerous brainstem regions, Dobbins and Feldman conclude d that, because the largest quantity of PRVpositive neurons in any brainstem region known to be inspiratory were found in the region of the rVRG, monosynaptic projections from the rV RG must represent the dominant anatomical pathway to the PhMNs. Because that study focused exclusively on the brainstem regions involved in modulating PhMN excita bility in the uninjured animal it did not provide information on plasticity-related changes in the phrenic respiratory circuit via the CPP. Overview of Thesis Study The present study was designed to further ex amine the brainstem phrenic respiratory circuitry and the plasticity observed in this system after SCI using PRV-Bartha as a neuroanatomical tracer. Since prev ious studies of this circuit at the supraspinal level have not produced a complete picture of all of the brains tem areas with descendi ng projections to PhMNs in the normal or injured case, it is important to quantify the distribution of brainstem neurons in this network in both control and injured anim als. Moreover, because PRV has been found to travel in a time-dependent fashion across a mu ltisynaptic circuit based on density of synaptic connections (when viral titer is held constant), neurons that show more synaptic connectivity to the region of application will s how earlier infection (Card et al., 1999). Therefore, to provide 19

PAGE 20

20 insight concerning the density of connectivity to PhMNs and to determine whether this connectivity is subject to changes after a C2 hemisection, it is also important to closely examine the time-course of PRV-labeling which occurs in this circuit. This information will expand the current understanding of the neural substrate and mechanisms of plasticity mediating the CPP. Therefore, the objective of the present inve stigation was to compare phrenic-associated brainstem labeling patterns when PRV is topicall y applied to the diaphragm in normal rats and C2-hemisected rats. In addition, th e aim of this study was to examine the differences in relative numbers of labeled cells in various brainste m areas throughout a time course after the PRV infection of uninjured animals to ascertain wh ich brainstem areas most densely innervate the PhMNs and to determine whether these regions change after C2 hemisection.

PAGE 21

CHAP TER 2 MATERIALS AND METHODS Animals Adult female Sprague-Dawley rats (250-300 g) were obtained from Harlan Scientific and subsequently housed at the University of Flor idas McKnight Brain In stitute Animal Care Facility. All surgical and animal care pro cedures were conducted under approval from the Institutional Animal Care and Use Committee at the University of Florida. A total of 17 animals were used in this study. Pseudorabies Virus PRV 152 was generously provided by Dr. David C. Bloom (University of Florida). This virus is a recombinant of the Bartha stra in of PRV and has a viral titer of 8.0-9.9x108 pfu/ml. Biosafety Level II practices (U.S. Department of Health and Huma n Services, 1988) were employed in all phases of PRV usage. This vi rus was prepared by producing seed stocks on porcine kidney (PK15) cells which were obtaine d from the American Type Culture Collection. The PK15 cells were grown at 37C, 5% CO2 with humidity in MEM Medium with Earls salts (Invitrogen). These cells were supplemented with nonessential amino acids, 10% fetal bovine serum and antibiotics (250 U of penicillin/ml, 250 g of streptomycin/ml). Viral stocks were then produced by infecting near confluent PK15 monolayers at a multiplicity of infection of 0.01. Once they exhibited a 100% cytopathic effect the cells and medium were harvested and centrifuged at 16,000 x g for 40 min, then re-suspe nded in 1/100th of the original volume of medium, and subjected to two rounds of freeze-th aw. The stocks were subsequently aliquoted into 100 l volumes and stored at -80C. The aliquots were titrated for infectious virus on monolayers of PK15 by standard plaque assay under agarose. 21

PAGE 22

Surgical Procedures Anesthesia and Post-Operative Ca re Surgical procedures were performed under asep tic conditions in a de dicated surgery room. For all surgical procedures, the animals were de eply anesthetized using injections of xylazine (0.15ml per animal, subcutaneous (s.q.)) and ketamine (1ml/kg body weight, intraperitoneal, (i.p.)). Once the necessary surgical procedure wa s completed, each animal received an injection of Yohimbine (0.2ml per animal, s.q.) to reverse th e anesthesia and lactated ringers solution (5ml per animal, s.q.) to prevent dehydration. As the an imals began to recover from anesthesia, they were given Buprenorphrine for analgesia (0.4ml of 0.3mg/ml, s.q.). The subsequent recovery of the animals was monitored ove r the next several days. Spinal Cord Injury For experiments involving spinal hemisecti on (n=4), an incision was made through the skin and underlying muscle which extended a bout one inch from the base of the skull. Subsequently, a laminectomy was performed to remove the second cervical vertebra (C2). A small dural incision was then made and a lateral C2 hemisection was performed by creating a cavity on the left side of the spinal cord with a microscalpel and aspiration using an angled, fine tipped glass pipette. The dura wa s closed with interrupted 10 -0 sutures and covered with Durafilm. The muscle was sutured in layers with 4-0 Vicryl ( polyglactin 910, synthetic absorbable sterile suture) and the skin was stapled closed with wound clips. Injured animals were then allowed to recover 2 weeks after the hemisection surgery, then perfused fixed to examine histology. Forty eight to sixty f our hours prior to perfusion, all animals (injured or uninjured) also received surgery for neuroanatomical tracing. 22

PAGE 23

Retrograde Neuroanatomical Tracing Brainstem neurons premotor to PhMNs were labeled with PRV 152 in both uninjured (n=13) and C2-hemisected animals (n=4). An incision approximately one inch long was made through the skin and underlying muscle of each animals ventral surface along the linea alba, starting from approximately the base of the sternum. Retractors were used to deflect the skin and muscle to gain access to the lower su rface of the diaphragm. PRV 152 (40-50 l) was then topically applied with a paintbrush to the left hemi-diaphrag m. The abdominal muscle was then sutured with Vicryl sutures and the skin was stapled closed with wound clips. The animals were left to recover for either 48 (n=4 uninjured), 56 (n=4 uninjured) or 64 (n=5 uninjured, n=4 injured) hours after applica tion of the tracer, at which time they were sacrificed. Perfusion and Tissue Sectioning At the end of the experiment, each animal wa s given a lethal dose (0.4 ml per animal) of Beuthanasia (phenytoin and sodi um pentabarbitone; minimum of 1ml/kg body weight, i.p.) Upon cessation of breathing, the chest cavity was opened a nd animals were intracardially perfused with 250 ml of 0.9% saline (with he parin and 0.2% sodium nitrite to prevent clotting) followed by 500 ml 4% paraformaldehyde (in 0.1M phosphate buffered saline (PBS)). Next, the brain and spinal cord were removed from each animal and post-fixed by immersion in 2% paraformaldehyde at 4 C until subsequent processing. Transverse brai nstem sections and either transverse or longitudinal spinal cord sections were cu t with a vibratome at a thickness of 40 m. Sections were collected sequentially in welled plates containing 2% paraformaldehyde. These sections were stored at 4 C until subsequent immunohi stochemical processing. Immunohistochemistry Sections were initially washed in PBS ( 0.1 M, three times for five minutes), then incubated in quenching solution (30% methanol, 0.6% hydrogen peroxide in 0.1M PBS, for one 23

PAGE 24

hour) in order to reduce endogenous peroxidase activity. The tissu e was then re-washed in PBS (0.1 M, three tim es for five minutes) and subseq uently incubated in blocking solution (10% normal goat serum in 0.1M PBS, for one hour) to reduce background staining due to non-specific protein labeling. Sections were then incubate d in a primary antibody (rabbit anti-PRV (Rb134), dilution 1: 10,000) overnight at 4 C. The follo wing day the tissue was re-washed in PBS (0.1 M, three times for five minutes) and then incubate d in a biotinylated secondary antibody (goat antirabbit, Jackson Immunocytochemicals, dilution 1: 200) for two hours at room temperature. Once again, the tissue was washed in PBS (0.1 M, three times for five minutes) and then incubated for another 2 hours in an avidin-biotin complex (ABC, Elite Vectastain Kit, Vector Labs). Sections were then given another series of washes in PBS (0.1 M, three times for five minutes) and the antigen was visualized with diaminobenzidine (D AB, Sigma). The sections were rewashed with PBS (0.1 M, three times for five minutes) and then mounted on Fisher superfrost slides. All slide-mounted brainstem sections were then counterstained with Cr esyl Violet (CV) in order to distinguish cell types and regions of the brainstem. Slides were immersed in butanol (2 min) then xylene (2 min) and once again in butanol (3 min). Next the slides were submerged in 100%, 95%, 70%, and then 50% ethanol (each for 3 mi nutes). After this the slides were placed in distilled water (3 min) and then into the Cresyl Violet stain (2 min). After this the slides were dipped briefly in distilled water (3 times) then placed in 50%, 70%, and 95% ethanol (each for 20 seconds). Slides were subsequently immersed in differentiation solution (1% glacial acetic acid in 95% ethanol) and monitore d until the sections reached the desired level of staining, at which time they were placed into 100% ethanol (1 min), butanol (3 Min) and xylene (2 min). After staining, the slides were coverslippe d with Richard-Allen Mounting Medium. Tissue sections were examined using brightfield microscopy. Brightfield photographs were taken 24

PAGE 25

25 digitally using a Zeiss AxioPhot microscope with an AxioCam HRc digital camera linked to a PC. Image contrast and exposure was subseque ntly corrected using Adobe Photoshop 6.0 (Adobe Systems, Inc.). Quantitative Analyses Using brightfield microscopy, PRV-positive neurons in the brainstem with visible nuclei were counted at 10x magnification in consecutive transverse brainstem sections from each animal. The Rat Brain in Sterotaxic Coordinates atlas was used to localize the specific regions of the brainstem where staining was observed (P axinos and Watson, 1997). Counts were made using the physical dissector method (Coggesha ll and Lekan, 1996; Guillery, 2002). That is, only cells with a visible nucleus were counted to reduce the likelihood of c ounting a single cell twice in consecutive sections. Statistical Analyses of Data Statistical analyses were conducted using the SigmaStat 3.5 software package on a Dell Computer. Statistical si gnificance was set at p<0.05. Unless otherwise in dicated, pooled data are presented as mean + S.E.

PAGE 26

CHAP TER 3 RESULTS Pseudorabies Virus Infection of the Cervical Spinal Cord To verify that PRV had produced reliable in fections, the cervical spinal cords of all animals (n=17) were qualitatively examined for the presence of PRV-positive PhMNs ipsilateral to the side of tracer application (Figure 3-1) There was no evidence found of bilateral PhMN pool labeling at any survival inte rval in injured or c ontrol animals. The lack of such labeling suggests that diffusion of the tracer to the c ontralateral hemidaphragm was not likely to have occurred in any of the animals. Labeling of PhMNs ipsilateral to the application of tracer was observed in all animals. Consistent with previous reports, examination of the cervical spinal cord sections revealed that the number of infected PhMNs appeared to incr ease as time post-PRV application increased (Lane et al., 2008b). Representati ve sections of the PhMN labe ling with PRV can be seen in Figure 3-1. Moreover, the distri bution and arrangement of thes e neurons was found to match previous descriptions of the PhMN pool in the rat (Lane et al., 2008b; Dobbins and Feldman, 1994; Goshgarian and Rafols, 1981). That is, the PhMN pool appeared in longitudinal sections of the cervical spinal cord as a tight column of ne urons which tilted slightly from a lateral position in the ventromedial gray matter (at higher levels of the cervical spinal cord) to a more medial position (at lower levels of the cervical spinal cord) and extended rostro-caudally from approximately C3 to C6. Also, the column was ma de up of distinct clusters of several neurons separated from each other at regular intervals for the length of the PhMN pool. As in previous reports of PRV labeling in the ph renic respiratory circuit, secondorder labeled interneurons were observed predominantly in Rexed laminae VII and X of the cervical spinal cord (Lane et al., 2008b; Dobbins and Feldman, 1994). 26

PAGE 27

Quantitative Analysis of Brains te m Labeling in Uninjured Animals At 48 hours after infection, very few neurons were labeled in the brainstem. At more delayed post-infection time points (i.e., 56 and 64 hours after PRV application), increased PRV labeling was observed in numerous areas of the br ainstem. Qualitatively, there appeared to be considerable variability between animals in th e total number of PRV-positive neurons observed in the brainstem at any given post-infection time (Tab le 3-3). This variabil ity is consistent with that observed in other studies wh ich have employed this method of PRV delivery (i.e., topically to the diaphragm) to study this circuit in the cervical spinal cord (Lane et al., 2008b). Despite apparent variability, it seemed that the areas of the brainstem which exhi bited the most labeling were comparable across the animals examined. No infected neurons were detected in the dor sal motor nucleus of the vagus in any animal at any post-infection survival interval examined. Th e lack of staining in th is region verifies that the viral infection was specifically restricted to th e phrenic respiratory circ uit and that the topical application of the virus to the diaphragm did not result in the infection of abdominal organs, which has previously been shown to cause labe ling in this brainstem region (Card et al., 1990; Yates et al., 1999). Moreover, no glial cell infection was seen in the brainstem at any postinfection interval. This indicates that supraspinal spread of the viral infection was not likely to have occurred due to lysis of cell bodies, thereb y supporting the presumpti on that the course of viral infection in this study occu rred predominantly in a transyna ptic fashion (Card et al., 1993). To circumvent variability in raw counts of PRV-positive medullary neurons, the data were first expressed as a proportion of total infected brainstem neurons in that animal (Tables 3-1 and 3-2). Next, to make comparisons between the br ainstem labeling observed in uninjured and C2hemisected animals at 64 hours post infection, regional neuronal counts were reported as a proportion of total brainstem labeling on each side of the brainstem (Table 3-4). This lateralized 27

PAGE 28

norm alization was done to account for the fact that, as it has been previously reported (Boulenguez et al., 2007), injured animals showed extremely attenuated cell-labeling on the side of the brainstem ipsilateral to injury (Table 3-3). Quantitative Analyses at the 48 an d 56 Hour Post-Infection Interval At 48 hours after delivery of PRV to the left hemidiaphragm (n=4), there was virtually no labeling seen in the brainstem. However, at 56 hours after PRV infection (n=4) a greater number of cells became infected in the brainstem. In all of these animals labeling was seen in multiple brainstem regions. At this post-infection interval many of these regions c ontained sparse labeling that was often inconsistent between animals. Nonetheless, the areas of the brainstem which showed the most consistent labeling (across the animals within this group) and on average accounted for approximately 5% or more of the total PRV-infected neurons in the brainstem (the most labeled areas) included the medial reticu lar formation (MRt), the rostroventrolateral reticular nucleus (RVL), the A5 Noradrenaline Cells (A5), the Intermediate reticular nucleus (IRt) and the raphe nuclei (RO a nd RPa). It is notable that th e rVRG accounted for only 3.5% of total labeling at this time point and was ther efore excluded from the statistical analyses. Moreover, PRV-infected neurons in the MRt nucle i accounted for the largest proportion of total brainstem labeling observed on both th e left and right sides of the brainstem in all four animals examined. In particular, the LPGi region of th e MRt accounted for the la rgest proportion of all gigantocellular labeling observed (Table 3-1 and Figure 3-4). To determine if any significant differences existed in the proportion of labeling in the total brainstem among the most labeled areas, a one-wa y repeated measures Analysis of Variance (ANOVA) was conducted on these areas at the 56 hour post infection interval. Differences were found between groups ( F (4, 19) = 56.32, p<0.001). Post-hoc analyses were conducted using the Holm-Sidak pairwise multiple comparison procedure. Specifically, the MRt contained a 28

PAGE 29

significantly greater prop ortion of brainstem labeling than the Raphe ( t = 12.33, p<0.001), the IRt (t = 12.20, p<0.001), the A5 ( t =11.63, p<0.001), and the RVL/CVL ( t =11.10, p<0.001) (Figure 3-3). No other signifi cant differences were found. Since the MRt was the only significantly di fferent region, a one-way repeated measures ANOVA was conducted to determine if any differe nces in the proportion of labeling existed among the specific subdivisions of the MRt (the LPGi, GiA, Gi, and GiV) Differences were found between groups ( F (3, 15) = 37.77, p<0.001). Post-hoc analyses were conducted to isolate where significant difference(s) occurred using the Holm-Sidak pairwise multiple comparison procedure. Specifically, the LP Gi contained a significantly greater proportion of brainstem labeling than the GiV ( t =9.25, p<0.001), the Gi ( t =8.53, p<0.001), and the GiA ( t =8.15, p<0.001) (Figure 3-4). No other signifi cant differences were found. Further analyses were conducted to determine if differences existed in the proportion of labeling between the left and right sides of the br ainstem in either the MRt or the LPGi. The data in the MRt failed the assumption of normality, and so were analyzed with Wilcoxons Signed Rank Test. There were no significa nt differences between the left (median 30.10) and the right (median 23.43) sides of the MRt, (W =-6.00, p=0.25). The data in the LPGi passed the assumption of normality, and so were analyzed with a paired-samples t-test. No differences were found between the left ( M =24.64+ 5.16) and the right ( M =17.57+ 2.77) sides of the LPGi (t =1.09, p=0.36). Quantitative Analyses at the 64 Hour Post-Infection Interval At 64 hours post-infection more brainstem neurons showed infection with PRV and greater consistency was seen in the pattern of la beling across animals. Specifically, all animals in this group (n=5) showed consistent labeling that on average, accounted for approximately 5% or more of the total PRV-infected neurons in the brainstem in the MRt nuclei, RVL/CVL, raphe 29

PAGE 30

nuclei (RO, RPa, raphe m agnus (RMg)), A5, rVRG and subcoeruleus nuclei (SubC). Despite the slightly more consistent labeling seen at this pos t-infection interval, ther e were still regions of the brainstem which contained spar se and/or inconsistent labeling between animals within this group. However, the MRt, and in particular the LPGi, persisted as the brainstem region which showed the greatest proportion of labeled cells in all animals on both sides of the brainstem (Table 3-2 and Figures 3-5 and 3-6). A representation of the typical labeling seen can be found in Figure 3-2. For the 64 hour time point, the same analyses were conducted as those at the 56 hour time point. To determine if any signifi cant differences existed in the pr oportion of labeling in the total brainstem among the most labeled areas, a on e-way repeated measures ANOVA was conducted on these areas at the 64 hour post infection inte rval. Differences were found between groups ( F (4, 29) = 45.35, p<0.001). Post-hoc analyses to isolate these differences were conducted using the Holm-Sidak pairwise multiple comparis on procedure. Specifically, the MRt had a significantly greater prop ortion of brainstem labeling than the SubC ( t =12.35, p<0.001), rVRG ( t =12.35, p<0.001), A5 ( t =11.55, p<0.001), Raphe ( t =11.16, p<0.001), and RVL/CVL ( t =9.71, p<0.001) (Figure 3-5). No other si gnificant differences were found. Since the MRt was the only region found to contain a significantly greater proportion of labeling, a one-way repeated m easures ANOVA was conducted to de termine if any differences existed among the specific regions of the MRt (the LPGi, GiA, Gi, DPGi, and GiV). Differences were found between groups ( F (4, 24) = 69.75, p <0.001). Post-hoc analyses were conducted using the Holm-Sidak pairwise multiple comparison procedure. Specifically, the LPGi had a significantly greater prop ortion of brainstem labe ling than the DPGi ( t =14.11, p<0.001), Gi 30

PAGE 31

( t =13.31, p<0.001), GiV ( t =12.90, p<0.001), and GiA ( t =12.11, p<0.001) (Figure 3-6). No other significant differences were found. Further analyses were conducted to determ ine if differences existed in the proportion of labeling between the left and right sides of the br ainstem in the both the MRt and the LPGi at the 64 hour time point. The data in the MRt passed the assumption of normality, and so were analyzed with a paired samples t-test. There we re no significant differe nces between the left ( M =24.12+ 1.02) and the right ( M =18.25+ 2.18) sides of the MRt ( t =2.32, p=0.08). The data in the LPGi also passed the assumption of normality, and so were analyzed with a paired-samples ttest. No differences were found between the left ( M =16.96+ 1.39) and the right ( M =14.47+ 2.07) sides of the LPGi ( t =1.10, p=0.33). Quantitative Analysis of Brains tem Labeling in Injured Animals In all injured animals there was less total labeling on the left side of the brainstem (i.e., ipsilateral to hemisection) than on the right (Table 3-3). Analyses were conducted to determine if medullary labeling patterns were altered two-weeks after a C2 hemisection. To assess differences in the laterality of la beling, two separate ANOVAs were conducte d: one for the left side of the brainstem and one for the right side of the brainstem. First, for the left side, a two-way repeat ed measures ANOVA (injury level x brainstem region) was conducted. Although the data violated the assumption of normality, they passed the equal variance test. There were significant differences among brainstem regions ( F (6, 62)=15.88, p<0.01), however there were no significant differences between injury conditions ( F (1, 61)=1.37, p =0.28), nor was the inter action term significant ( F (6, 62)=0.36, p=0.90). Post-hoc analyses were conducted with the Holm-Sidak pairwise multiple comparison procedure. Specifically, the MRt contained a significantly gr eater proportion of left-side labeling than the rVRG ( t =7.80, p<.001), IRt ( t =7.76, p<0.001), SubC ( t =7.72, p<0.001), A5 ( t =7.38, p=0.001), 31

PAGE 32

32 RVL/CVL ( t =6.85, p<0.001), and Raphe ( t =5.93, p<0.001) (Table 3-4 and Figure 3-7). No other significant differences were found. The second test, for the right side, was also a two-way repeated measures ANOVA (injury level x brainstem region). Although the data viol ated the assumption of normality, they passed the equal variance test. Again, there were significant differences among brainstem regions ( F (6, 62)=41.36, p<0.001), however there were no significant differences between injury conditions ( F (1, 62)=0.02, p=0.89), nor was the interaction term significant ( F (6, 62)=1.02, p=0.43). Posthoc analyses were conducted with the Holm-Sid ak pairwise multiple comparison procedure. Specifically, the MRt contained a si gnificantly greater proportion of right-side labeling than the IRt (t =12.83, p<0.001), SubC ( t =12.78, p<0.001), rVRG ( t =12.68, p<0.001), A5 ( t =10.83, p<0.001), RVL/CVL ( t =10.60, p<0.001), and Raphe ( t =10.33, p<0.001) (Table 3-4 and Figure 38). No other significant differences were found.

PAGE 33

A B C Figure 3-1 : PRV-infected PhMNs are specifically labeled in the cervical spinal cord. A) A transverse spinal cord section that was obtained from approximately the C3 spinal level at 64 hours after PRV la beling of the left hemidia phragm. B) Magnification of PRV-infected ipsilateral PhMNs in part A. C) A portion of the PhMN pool in a longitudinal spinal cord section at 64 hours after PRV labeling of the left hemidiaphragm (extending from rostral to caudal from top to bottom). 33

PAGE 34

A B C Figure 3-2: Brainstem labeling in the Medial Reticular Formation. A) Image of a transverse brainstem section from The Rat Brain in Sterotaxic Coordinates by Paxinos and Watson (1997). Red asterisks indicate PRV-pos itive neurons seen in four consecutive 40 m brainstem sections at 64 hours post-in fection. B) PRV-positive LPGi neurons on the Left side of the brainstem at 64 hours post-infection. C) PRV-positive LPGi neurons on the Right side of the brainstem at 64 hours post-infection. 34

PAGE 35

35 Table 3-1: Percentage of total labeling in brainstem in each region at 56 hours ____________________________________________________________ Region Mean SE Overall brainstem MRt 56.28 4.33 RVL/CVL 10.05 2.50 A5 7.84 2.46 IRt 5.48 2.60 Raphe 4.95 0.83 Regions of MRt LPGi 42.21 5.15 GiA 6.81 1.54 Gi 5.20 1.58 GiV 2.06 1.24

PAGE 36

36 56 hour Uninjured Most Labeled Brainstem RegionsBrainstem Regions 0 10 20 30 40 50 60 70 Percentage of Total Brainstem LabelingRaphe MRt RVL/CVL IRt A5 Figure 3-3: 56 hours uninjured m ost labeled brainstem regions

PAGE 37

37 56 Hour Uninjured Regions of the MRtRegion of the MRtFigure 3-4: 56 hours uninjured regions of the MRt Percentage of Tota l Brainstem Labeling 0 10 20 30 40 50 60Gi GiV LPGi GiA

PAGE 38

Table 3-2 : Percentage of total labeling in brainstem in each region at 64 hours ____________________________________________________________ Region Mean SE Overall brainstem MRt 42.37 2.26 RVL/CVL 12.95 0.52 Raphe 8.64 2.07 A5 7.40 2.05 SubC 4.97 1.08 rVRG 4.97 2.87 Regions of MRt LPGi 31.43 2.72 GiA 4.92 1.24 GiV 3.19 1.25 Gi 2.29 0.33 DPGi 0.53 0.17 38

PAGE 39

39 Figure 3-5: 64 hours uninjured mo st labeled brainstem regions 64 hour Uninjured Most Labeled Brainstem RegionsBrainstem Regions 0 10 20 30 40 50 60 Percentage of Total Brainstem Labeling*Raphe rVRG MRt RVL/CVL A5 SubC

PAGE 40

40 Figure 3-6: 64 hours uninjured regions of the MRt 64 hour Uninjured Regions of the MRtRegion of the MRt 0 10 20 30 40 50 Percentage of Total Brainstem LabelingGi GiV LPGi DPGi GiA

PAGE 41

Table 3-3 : Total number of labeled cells in each side of the brainstem per animal at 64 hours post-infection ________________________________________________ Animal Left Right Uninjured 1 113 132 2 272 213 3 703 501 4 285 286 5 507 364 Injured 6 166 1562 7 25 202 8 23 360 9 10 101 41

PAGE 42

Table 3-4 : Mean percentage of labeling in brainste m in each region at 64 hours (left side/right side) ______________________________________________________________________________________________________ Raphe rVRG Gi RVL/CV IRt A5 SubC Uninjured 11.90/8.51 5.02/5.15 45.39/40.48 14.07/12.21 3.40/4.18 6.16/8.90 5.91/3.18 Injured 16.31/12.83 3.74/0.90 44.67/48.11 4.65/7.39 5.74/0.88 6.94/9.18 3.6/2.19 Note: The standard errors are as follows: Uninjured left side = 4.91, Injured left side = 5.49, Uninjured right side = 3.07, In jured right side = 3.43. 42

PAGE 43

43 64 hr Uninjured and Inju red Percent of total La beling on Left BrainstemBrainstem Regions RapherVRGMRtRVL/CVLIRtA5SubC Percentage of Left Brainstem Labeling 0 10 20 30 40 50 60 70 Uninjured Injured Figure 3-7: 64 hours uninjured and injured percent of tota l labeling on left brainstem

PAGE 44

44 64 hour Uninjured and Injured Percent of Total Labeling on Right BrainstemBrainstem Regions RapherVRGMRtRVL/CVLIRtA5SubC Percentage of Right Brainstem Labeling 0 10 20 30 40 50 60 70 Uninjured Injured Figure 3-8: 64 hours uninjured and injured percent of tota l labeling on right brainstem

PAGE 45

CHAP TER 4 DISCUSSION This study represents the first attempt to qua ntitatively determine the regional distribution of medullary neurons projecting to PhMNs in bot h the normal and SCI-injur ed rat. These initial results suggest that despite unilateral denervation of the PRVinfected phrenic nucleus, which caused reduced labeling in the ipsilateral brainstem, the bilateral patte rn and relative proportion of infected brainstem neurons was comparable to observations and data in normal animals. In addition, while the rVRG is considered to be the primary source of inspiratory drive to PhMNs, it quantitatively exhibited far less extensive labe ling by transneuronal PRV in normal and injured animals than other regions of the medulla. In fact, the most significant second-order brainstem labeling was in the MRt, the role of which in re spiratory control is not entirely clear. Before discussing the implications of these findings relative to normal a nd post-SCI respiratory function, it is first necessary to address several technical considerations concerning the methods employed in the present study. Technical Considerations Pseudorabies Virus as a Retr ograde Transynaptic Tracer For many reasons, the decision to use PRV as the neuroanatomical tracer in the present study was motivated by a previous report in which monosynaptic tra cers were used to evaluate rVRG projections to PhMNs af ter a C2 hemisection (Boulengu ez et al., 2007). Neurotropic viruses, such as PRV, are effective as transneu ronal tracers that can define a pathway of cells which are synaptically connected. Viruses in ge neral are well suited for neuroanatomical tracing because they can produce controlled viral infections that traverse multiple synapses in sequential order within a network of synaptically li nked neurons (Card and Enquist, 1999; Loewy, 1998). This is in contrast to more conventional retrograde tracers (such as FG ), which label only a 45

PAGE 46

single cell and never traverse a synapse (Loew y, 1998). Moreover, the fact that PRV can be applied p eripherally is especially advantageous because applying the virus to a peripheral nerve or muscle (i.e., the diaphragm) can avoid labelin g of cells in the CNS which are not involved in the neuronal circuit of interest. Additionally, these transn euronal tracers m ove sequentially across synaptic connections in a hierarchical fashion. Thus, tem poral analysis can reveal the order of synaptic connectivity (C ard and Enquist, 1999; Loewy, 1998). The strain of PRV used in the present study is PRV-Bartha, which is a naturally occurring attenuated strain of alphaherpes virus-PRV that has been extremely well characterized in a variety of neuronal networks and has b een shown to produce reliable tr ansneuronal infections in a number of species (Card, 1998; Loewy, 1998; Yates et al., 1999). Many characteristics of this strain have made it particularly suitable as a tracer. For instance, the reduced virulence of PRVBartha has made it less cytopathogenic to the ne urons that it infects (Card, 1998). This allows viral infection through a given circ uit to be examined over several days before the onset of cell death (Loewy, 1998). Also important is the fact that PRV-Bartha has been repeatedly found to produce viral infections in an ex clusively retrograde direction (i.e., from axon terminal to cell body) (Card et al., 1998). Thus, by using PRV-Bartha it is assured that cells in a neuronal circuit are progressively labeled in a specific hierarchical fashion, progressing from axon to soma and from one cell to cells that project directly to that cell. Additionally, the factors determining the rate of transneuronal labeling have been wellcharacterized. The time it takes PRV-Bartha to infect neurons in a multisynaptic circuit is predominantly determined by viral titer and the density of innervation (Aston-Jones and Card, 2000; Card et al., 1998; Card et al., 1999). Relative to the latter neurons that more densely 46

PAGE 47

innerva te or show more synaptic connectivity w ith previously infected cells will show earlier infection (Card et al., 1999). Thus, the many qualities of PRV-Bartha led to its selection as the tracer employed to examine the brainstem-phrenic ci rcuit in the present study. For the purposes of this study, only second-order brainstem neuronal infections were c onsidered. Because the concentration of virus was held constant, the main determinant of the temporal pattern of second-order infection was the degree of synaptic connectivity to the PhMN pool. In essence, brainstem neurons with dense projections to the PhMNs s howed early PRV infection. Variability and Normalizatio n of Brainstem Cell Counts In the present study PRV was topically applied to the diaphragm. This method of tracer application was chosen over diap hragm injections to limit the variability in labeling, as the current method has been found to result in more consistent patterns of neuronal infection (unpublished results). Nonetheless, there was su bstantial variability in the total number of infected neurons between animals at any given post-infection interval. Some of the variability in labeling could be accounted for by incomplete primary labeling of the PhMN pool. Such variability has been described in previous studies (Lane et al., 2008b) and could be a factor in the disparity of higher-order brai nstem labeling seen between an imals in the current study. Despite the advantages of PRV, it (like most retrograde tracing me thods) is subject to considerable variability. Raw count s of infected cells are thus difficult to compare within and between animal groups. Counts of infected brains tem neurons in uninjured animals were thus normalized as a function of total infected brai nstem neurons in each animal. Similarly, in comparing brainstem labeling in uninjured vers us C2 hemisected animals, the counts were normalized as a function of total infected neurons on the left and right sides of the brainstem independently in each animal. Unfortunately, as w ill be explained later, a major limitation in this 47

PAGE 48

norm alization arises when comparison is made be tween injured and uninjured animals. For that reason, interpretations presente d below are tentative and ba sed solely on the inherent assumptions of the normalization approach taken. Alternative quantitative strategies will be required in the future to more adequately assess lateralized differences in labeling in the injured circuit. Pattern of Brainstem Neuron s After Spinal Cord Injury Prevalence of Crossed Projections to PhMNs When considering spontaneous respiratory neur oplasticity after high cervical hemisection, an important question is whether any demonstr able changes occur in the distribution of brainstem neurons projecting to the denervated PhMN pool. Previ ously, in a brief study, Moreno et al. (1992) used wheatgerm a gglutinin conjugated to horseradish peroxidase (WGA-HRP) to trace the brainstem origin of axons which drive the PhMNs in C2 hemisected rats. Moreno made injections of this retrograde tracer into functionally recovered he midaphragm muscle and observed that labeled cells we re found bilaterally in the rVRG. While no quantification was attempted, this suggested the presence of spinal decussating rVRG axons which originate in both sides of the brainstem a nd may mediate the CPP. More recently, Boulenguez et al. (2007) injected FG into the spinal cord in the vicinity of the PhMNs in rats after C2 hemisection and atte mpted to quantify the brainstem neurons within the rVRG that cross the spinal midline. Their observations indicate th at the number of rVRG neurons that became labeled in hemisected anim als represented only a small proportion of those labeled in control animals (23% ipsilaterally a nd 36% contralaterally). The cells labeled after injury represent brainstem neurons with axons th at decussate at the leve l of the PhMNs, which may be involved in the CPP. 48

PAGE 49

However, there are several lim itations with the methods employed in the Boulenguez study. For instance, it was clear that the extent of FG labeling was extremely diffuse, because labeled cells were not restricted to PhMNs and were found on the entire left side of the spinal cord. Hence, it was impossible to be certain th at the FG-labeling was restricted the phrenic respiratory circuit. Also, while th e authors did not quantify the enti re column of cells that were labeled in the rVRG, instead only counting cells in a single 30 m transverse brainstem section per animal, they made quantitative statem ents regarding the extent of labeling. In the present study counts were taken of every labeled neuron in consecutive 40 m brainstem sections. Thus, the current study repres ents a more holistic profile of the brainstem regions that project to PhMNs th an any previous study. It was found from an examination of the raw counts of labeled cells in individual animal s that, consistent with Boulenguez et al. (2007), there was less total brainstem labeling ipsilatera l to hemisection (Table 3-3). However, in contrast with that study, there did not seem to be a substantial decrease in labeling contralateral to injury (Table 3-3). Because bulbospinal axons which descend the spinal cord ipsilateral to injury are interrupted by hemisec tion, the subset of neurons labeled in the ipsilateral brainstem must decussate twice: once at the level of the brainstem and again at the level of the PhMNs, while the neurons labeled contralaterally must d ecussate only once at th e level of the PhMNs. Since labeling only appeared to be attenuated ipsilaterally (relative to controls), the current findings suggest that double-decussating neurons may represent a small portion of bulbospinal fibers that contact PhMNs, as was suggested by Boulenguez. But neurons which decussate once at the level of the cervical spinal cord may be more pr evalent. However, due to the variability seen in total brainstem labeling between animal s, no statistical analyses were performed and 49

PAGE 50

therefore no quantitative statem ents can be made regarding the exte nt of spinal decussation in this circuit. Persistent Regional Brainstem La beling After Spinal Cord Injury Additionally, the proportion of regional la beling in the brainstem was examined. Quantitative analyses revealed that at 64 hours post-PRV infection, hemisected animals showed no differences from normal animals in the overall pattern of brainstem labeling. That is, after SCI, the proportions of regiona l labeling did not si gnificantly differ from those of control animals on either side of the brainstem (Figur es 3-7 and 3-8), indica ting that the brainstem regions which contact the PhMNs pre-injury are not substantially altere d post-injury. Thus, the neurons that remain synaptically linked to the PhMN pool after SCI, a nd notably those in the MRt, could prove to be important from a therapeutic perspective, as they may be influential in mediating the CPP and could be viable target s for the development of novel methods of breathing augmentation. Considerations when Interpreting Brains tem Labeling After Spinal Cord Injury In the current study brainstem labeling was examined at a two weeks post-hemisection interval. The selection of this time point after injury was based upon evid ence from several prior studies indicating that two weeks after hemisection is the earliest onset of the spontaneous CPP (Fuller et al., 2008; Fuller et al., 2006; Fuller et al., 2003; Golder and Mitchell, 2005). However, it is important to note that in the C2 hemis ected animals no neurophysiological analyses were performed to verify that the CPP was actually present at two weeks post-hemisection because it is presently unknown to what degree PRV in fection may alter neuronal activity. Moreover, when interpreting the findings from the hemisected animals in the current study, it is important to note that the prior usage of PRV as a neuroanatomical tracer in lesioned animals has been relatively limited (Kim et al ., 2002). Therefore, the possibility exists that 50

PAGE 51

neurons in an injured circuit m ay exhibit variabi lity in their susceptibility to viral infection. Similarly, it is possible that after injury, periph eral changes at the leve l of the diaphragm or remodeling at the neuromuscular junction (i.e., between the diaphragm a nd the phrenic nerve) (Mantilla and Sieck, 2003) could have altered the uptake of PRV into the respiratory circuit. For these reasons, results in the injure d animal must be taken with caution. Another limitation in the pres ent study is that only one post-infection interval was examined in injured animals. Thus, in future studies it will be necessary to examine labeling at multiple time points after PRV in fection. This will give insight into any mechanisms of neuroplasticity which could be ma nifested as an alteration in the time it takes for the virus to traverse this circuit. Nonetheless, the present work can provide some technical insight into the use of this neuroanatomical tracing method af ter SCI as a means of studying injury-induced neuroplasticity. Normal Pattern of Brainstem Neurons An unexpected finding as part of the analysis of normal and spinal injured animals was that the rVRG was one of the lesser labeled regions of the brainstem. Previous neuroanatomical data using PRV (Dobbins and Feldman, 1994) identified the rVRG as the dominant anatomical projection to the PhMN pool in the normal ra t because the largest quantity of PRV-positive neurons in any brainstem region that is known to be inspiratory was found there. However, as described in more detail below, the relatively sparse rVRG la beling observed in the present study calls into question this previous conclusion. Interestingly, the MRt appears to project heavily to the PhMNs, thereby suggesting that these neurons may be important in the rat phrenic respiratory circuit. 51

PAGE 52

Rostral Ventral Respiratory Group Projections to PhMNs Surprisingly, only sparse labeling was seen in the rVRG throughout the tim e course of PRV infection. Indeed, at the 56 hour time poi nt, the rVRG contained less than 5% of the overall labeling, and at the 64 hour time point, it contained bare ly 5%, and it was outranked by the MRt, the RVL/CVL, the Raphe, and the A5 (T able 3-2 and Figure 3-5). Therefore, based on the manner in which PRV traverses a multisynaptic circuit, the present results suggest that the rVRG exhibits a relatively le sser degree of connectivity to PhMNs than numerous other brainstem regions. This observation contrasts with a previous study by Dobbins and Feldman (1994) which concluded that monosynaptic projections from the rVRG in the uninjured rat must represent the dominant anatomical pathway to the PhMNs. Th is conclusion was made because the largest quantity of PRV-positive neurons in any brainste m region that is known to be inspiratory was found in the rVRG (Dobbins and Feldman, 1994) and because it has been ex tensively shown that rhythmic inspiratory drive comes from this medullary region (Feldman, 1986). However, examination of representative recons tructions of transverse sections through the brainstem in that study revealed substantial labeling in areas outside of the rVRG (notably w ithin the Gi and GiA) that was not quantified because these areas were not within a known insp iratory region of the medulla (Dobbins and Feldman, 1994). Thus, it is lik ely that if counts were taken of all labeled brainstem neurons (as was done in the present study), the rVRG w ould not have represented such a large proportion of labeled cells. Also, when interpreting the present observation of sparse rVRG labeling, it is important to note that Dobbins and Feldman (1994) injected PRV into the phrenic nerve, in contrast to the current method of topical application of PRV to the diaphragm. Thus, different methods of tracer application may have produced differences in the time required for the virus to infect the circuit 52

PAGE 53

between the studies. For in stance, it is possible that if anim als in the current study were allowed to surviv e to a more delayed post-infection interval, the proportion of labeling in the rVRG would increase to reflect the predominant labeling observed by Dobbins and Feldman. However, in the current study particular atte ntion was paid to the identificat ion of the brainstem areas that exhibit the greatest degree of synaptic connect ivity to the PhMN pool, and therefore the focus remained on the brainstem areas which showed the earliest infection (i.e., the MRt). Medial Reticular Formation Projections to PhMNs Unexpectedly, it was found that the MRt, which is not thought of as a respiratory region of the brainstem, accounted for most of the PRV labe ling seen in the brainstem even at the earliest time points examined. It was found that in unin jured animals at both the 56 and 64 hour postinfection intervals, the MRt was the only medulla ry region that exhibite d significantly more labeling than any other region (Figures 3-3 a nd 3-5). Moreover, examination of the mean percentages in the brainstem regions with the most labeling at both time points shows that the proportion of total brainstem labeling observed in the MRt was several times that of the region with the next highest percentage typically accounting for nearly half of all PRV labeled cells in the brainstem (Tables 3-1 and 3-2). The fact that such a large proportion of PRVlabeled cells were consistently found in the MRt suggests that this region c ontains neurons which project heavily to the PhMN pool. In particular, the LPGi was the only subdivision wi thin the MRt that exhi bited significantly more labeling than any of the others, and this persisted at both post-infection intervals (Figures 3-4 and 3-6). Likewise the means show that the proportion of labeling occurring in the LPGi was vastly greater than the region with th e next highest percentage and accounted for the majority of labeling that was seen in the MRt (Tables 3-1 and 3-2). Thus, this subdivision of the MRt represents a dominant synaptic projection to the PhMN pool. In terestingly, in both the MRt and 53

PAGE 54

LPGi, there were no significant di fferences in the proportion of la beling occurring in the left and right sides of the brainstem i ndicating that in the normal anim al these brainstem cells have bilaterally equivalent degrees of PhMN connectivity. Previous Evidence for Importance of the Medial Reticular Formation in the Phrenic Respiratory Circuit This study represents the first neuroanatomical evidence to suggest the importance of the MRt in the phrenic respiratory circuit of the rat. However, this finding is not entirely unprecedented. Indeed, although previous studies using a rat model have almost universally identified the rVRG as the dominant area in th e respiratory circuit (D obbins and Feldman, 1994; Boulenguez, 2007; Feldman, 1986), th e fact that the involvement of the MRt in the phrenic respiratory circuit has not been documented may re flect a bias in the literature concerning where inspiratory cells are presumed to be located. However, studies in species such as the cat and the ferret have indi cated that there are likely to be cells other than the rVRG which are important in control of inspiration and PhMN excitability. For instance, the current finding that neurons in the MRt repr esent the majority of brainstem neurons infected by PRV application to the diaphragm mirrors findings in the ferret (Yates et al., 1999; Billig et al., 2000). Additionally, neurophysiological data in the cat has suggested that there are neurons in the MRt that can be recru ited to function as respiratory neurons (Gordievskaya and Kireeva, 1999). Howe ver, until now it has been assumed that the presence of functionally important respiratory-related cells in ot her areas of the brainstem is a characteristic exclusive to emetic species (i.e., species capable of vomiting), unlike the rat (Yates et al., 1999). This is because previous accounts in the rat have largely ignored phrenic-associated labeling in these brainstem areas. Thus, it was a ssumed that in the ferret the MRt must be important in controlling PhMN exci tability during activities which are not exhibited in rats (i.e., 54

PAGE 55

vom iting) (Yates et al., 1999; Dobbins and Feldman, 1994). As a result, the anatomical importance of respiratory-related neurons in re ticular areas of the br ainstem has not been extensively examined in the rat. However, the predominant connectivity of the MRt to PhMNs, both before and after SCI, demonstrates that the MRt may play a more important role in respiration in the rat than previously thought. Moreover, previous neuroanatomical studies in the rat have even suggested the possible presence of a widespread network of brainste m neurons that form connections with PhMNs outside of the rVRG. For instance, labeling similar to that in th e present study (Figure 3-2) was seen within the MRt after the injection of a mono synaptic tracer into th e region of the PhMNs in rats (Boulenguez et al., 2007). However, such la beling was dismissed as spurious due to the diffuse nature of their CNS injection of tracer Similarly, evidence from Dobbins and Feldman (1994) using PRV revealed projections from the MRt to PhMNs and documents labeling at early post-infection intervals in regions of the MRt, including the Gi and GiA. However, because Dobbins and Feldman did not feel that these ce lls represented a respir atory population of cells, they provided no quantification within the MR t. Thus, the predominance of the connections between the MRt and PhMNs in the rat was not appreciated. Functional Implications of the Medial Reticular Formation in Respiration Numerous studies in a variety of species have suggested that neurons in the MRt are likely to be functionally involved in re spiration. For instance, there have been accounts indicating that these cells may be involved in the coordination of the contractions of respiratory muscles. In fact, it has previously been found in the ferret that neurons in this region of the brainstem provide input to both inspiratory and expiratory motoneurons in the spinal cord, which control the diaphragm and the rectus abdominus muscle, re spectively (Billing et al., 2000). From this finding and subsequent neurophysiol ogical examination of cells in this region, it was postulated 55

PAGE 56

that at least a subset of these neurons m ay be involved in coordinating th e contraction of several respiratory muscles to augment breath (B illig et al., 2000; Shin tani et al., 2003). Similarly, cells in the MRt are known to be dire ctly involved in vomiting. Indeed, Miller et al. (1996) concluded that if the MRt is inhibited, then cat s are no longer able to vomit. Interestingly, neurons in the rVRG are inhibited during vomiting in the cat, indicating that neurons in this region are not essential for relaying signals during this activity (Yates et al., 1999). Although it is known that rats do not posse ss the ability to vomit, during this behavior breathing becomes labored. Therefore, it is possi ble that the MRt neurons are more important in modulating PhMN activity duri ng labored breathing, whereas rhythmicity of breathing (generated by the rVRG) is not as necessary during such behaviors. Additionally, there are activities that rats do engage in, which require altered patterns of respiration and have been shown to involve MRt neurons. For inst ance, during rapid eye movement (REM) sleep, breathing becomes more rapid and variable and fi ring rates of neurons in the MRt increase (Bellingham a nd Funk, 2000). This suggests that in the rat the function of the MRt neurons may be related to generating pa tterns of breathing which deviate from passive eupnea (i.e., normal, relaxed breathing). Moreover, some authors believe that the num ber of respiratory-related neurons in the reticular formation can increase under conditions where respiration becomes laborious or hindered (Gordievskaya and Kir eeva, 1999), such as after a ce rvical SCI. Thus, it seems especially likely that the MRt s role in both vomiting and REM sleep could be related to the need for augmenting breathing during these behaviors. If this is the case, the MRt may prove to be important in regulating breathing after a cervical SCI, when pa tients must exert more voluntary control to increase re spiration. Supporting this is the fact that the MRt is known to receive input 56

PAGE 57

from the motor cortex in cats, a region larg ely responsible for volitional movement (Canedo, 1997). Thus, the MRt in the rat may also be i nvolved in volitional contro l of the diaphragm, possibly coordinating diaphragm adjustments before voluntary movement or when breathing becomes labored. Therefore, drawing from evidence in a vari ety of species, it is likely that the MRt is functionally involved in augmenti ng breath. With this possibility mind, the current finding that the MRt is dominantly connected to PhMNs and th at this extensive connec tivity persists after a C2 hemisection seems logical. Since after this ty pe of injury breathing becomes laborious, these neurons could become particularly important after such an injury. Thus, these cells could represent a viable target for future investig ations into therapeutic methods of improving respiration after SCI. For this reason, further studies are needed to examine the functional role of these neurons to assess their role in the CPP and respiration ge nerally. In particular, it would be interesting to examine the neurochemical profile of these neurons and whether the activity or firing pattern of the neurons in the MRt are sign ificantly altere d after SCI, as this knowledge could provide further insight into possible mechanisms of respir atory neuroplasticity afte r a cervical SCI. For instance, it would be interesting to find out if cells in the MRt which synapse on PhMNs are serotonergic, since serotonin is known to be important in the m odulation of respiratory drive in both the brainstem and PhMNs (Bonham, 1995; Gold er et al., 2001a; Goshgarian, 2003) and in the activation of the CPP (Zhou and Goshgarian, 2000). Additionally, the involvement of the MRt in the phrenic respiratory circuit in the rat and in emetic species such as the ferret suggests that there may be fewer differences in the neurocircuitry between species than it has previously been thought. If this is the case, the rat may 57

PAGE 58

58 represent an even more practical translational model for investigations involving the CPP as an instance of respiratory neuroplasticity. Summary To summarize, the present study represents the first attempt to quantitatively determine the distribution of brainstem neurons projecting to PhMNs in both the normal and C2 hemisected rat. These results suggest that despite unilateral dene rvation of the PhMNs, the bilateral pattern of infected brainstem neurons is la rgely unaltered from that of normal animals. In addition, while the rVRG is considered to be the primary source of inspiratory drive to PhMNs, sparse transneuronal labeling with PRV was observed. Unexpectedly, the most significant brainstem labeling was in the MRt, which may be involved in augmenting breath. Therefore, future studies should examine the functional role of the MRt in th e phrenic respiratory circuit, with the ultimate goal of developing therapeutics to improve resp iratory function in patie nts with high cervical SCI.

PAGE 59

APPENDIX LIST OF SOLUTIONS Quenching Solution 7 ml PBS 3ml Methanol 165ul 30% H2O2 Blocking Solution 1ml NGS 9ml PBS Primary 1ul Anti-PRV (Rb134) 200ul NGS 30ul Triton X-100 9769ul PBS Secondary 50ul Biotinylated Goat anti-Rabbit 200ul NGS 30ul Triton X-100 9720 PBS ABC Solution 9870 PBS 50ul Reagent A 50ul Reagent B 30ul Triton X-100 DAB Solution 20ml PBS 1 DAB tablet 1.5ul 30% H2O2 59

PAGE 60

LIST OF REFERE NCES Aston-Jones G, Card PJ. 2000. Use of pseudorabies virus to delineate multisynaptic circuits in brain: opportunities and limitations. Journal of Neuroscience Methods 103: 51-61. Baker TL, Mitchell GS. 2000. Episodic but not c ontinuous hypoxia elicits long-term facilitation of phrenic motor output in rats. Journal of Physiology 529: 215-219. Bellingham MC, Funk GD. 2000. Cholinergenic modula tion of respiratory brain-stem neurons and its function in sleep-wake state dete rmination. Clinical and Experimental Pharmacology and Physiology 27: 132-137. Billig I, Foris JM, Enquist LW, Card JP, Ya tes BJ. 2000. Definition of neuronal circuitry controlling the activity of phr enic and abdominal motoneurons in the ferret using recombinant strains of pseudorabies viru s. Journal of Neuroscience 20(19): 7446-7454. Boulenguez P, Gauthier P, Ka sner A. 2007. Respiratory neuron subpopulations and pathways potentially involved in the reactivation of phrenic motoneurons after C2 hemisection. Brain Research 1148: 96-104. Brown R, DiMarco AF, Hoit JD, Garshick E. 20 06. Respiratory dysfunction and management in spinal cord injury. Respiratory Care 5: 853-870. Canedo A. 1997. Primary motor cortex influenc es on the descending and ascending systems. Progress in Neurobi ology 51: 287-335. Card PJ. 1998. Practical considerations for the use of pseudorabies virus in transneuronal studies of neural circuitry. Neuroscience an d Biobehavioral Reviews 22(6): 685-694. Card PJ, Enquist LW. 1999. Transneuronal circuit analysis with pseudora bies viruses. Current Protocols in Neuros cience S9:1.5.1-1.5.28. Card PJ, Enquist LW, Moore RY. 1999. Neuroinvasiveness of pseudorabies virus injected intracerebrally is dependent on viral concentration and terminal field density. Journal of Comparative Neurology 407: 438-452. Card PJ, Levitt PR, Enquist LW. 1998. Different patt erns of neuronal infec tion after intracerebral injection of two strains of pseudorabies virus. Journal of Virology 72(5): 4434-4441. Card JP, Rinaman L, Lynn RB, Lee BH, Meade RP, Miselis RR, Enquist LW. 1993. Pseudorabies virus infection of the rat central nervous system: ultrastructural characterization of viral replic ation, transport, and pathogenesi s. Journal of Neuroscience 13(6): 2515-2539. Coggeshall RE, Lekan HA. 1996. Methods for dete rmining numbers of cells and synapses: a case for more uniform standards of review Journal of Comparative Neurology 364(1):615. 60

PAGE 61

Dobbins EG, Feldm an JI. 1994. Brainstem network controlling descending drive to phrenic motoneurons in rat. Journal of Comparative Neurology 347: 64-86. Ellenberger HH, Feldman JL, Goshgarian HG. 1990. Ventral respiratory group projections to phrenic motoneurons: electron microscopic evidence for monosynaptic connections. Journal of Comparative Neurology 302(4): 707-714. Feldman JL. 1986. Neurophysiology of breathing in mammals. In: Bloom FE, editor. Handbook of physiology, the nervous system IV: intrinsi c regulatory systems of the brain. Bethesda: American Physiological Society. p 463-524. Forster HV. 2003. Invited review: Plasticity in the control of brea thing following sensory denervation. Journal of A pplied Physiology 94(2):784-794. Fuller DD, Doperalski NJ, Dougherty BJ, Sa ndhu MS, Bolser DC, Reier PJ. 2008. Modest spontaneous recovery of ventilation followi ng chronic high cervical hemisection in rats. Experimental Neurology 211(1): 97-106. Fuller DD, Golder FJ, Olson EB Jr, Mitche ll GS. 2006. Recovery of phrenic activity and ventilation after cervical spinal hemisecti on in rats. Journal of Applied Physiology 100(3): 800-806. Fuller DD, Johnson SM, Olson EB Jr, Mitchell GS. 2003. Synaptic pathways to phrenic motoneurons are enhanced by chronic interm ittent hypoxia after cervical spinal cord injury. Journal of Neuroscience 23(7): 2993-3000. Golder FJ, Mitchell GS. 2005. Spinal synaptic enhancement with acute intermittent hypoxia improves respiratory function after chronic cervical spinal cord injury. Journal of Neuroscience 25(11): 2925-2932. Golder FJ, Reier PJ, Bolser DC. 2001a. Altered resp iratory motor drive afte r spinal cord injury: Supraspinal and bilateral eff ects of a unilateral lesion. Jour nal of Neuroscience 21(21): 8680-8689. Golder FJ, Reier PJ, Davenport PW, Bolser DC. 2001b. Cervical spinal cord injury alters the pattern of breathing in anesth etized rats. Journal of A pplied Physiology 91(6): 24512458. Gordievskaya NA, Kireeva N Ya. 1999. Involvement of reticular neurons of the cat medulla oblongata in the integrative activity of the respiratory center. Neuroscience and Behavioral Physiology 29(3): 327-332. Goshgarian HG. 2003. The crossed phrenic phenomenon: a model for plasticity in the respiratory pathways following spinal cord injury. Journal of Applied Physiology 94: 795-810. Goshgarian HG, Ellenberger HH, Feldman JL 1991. Decussation of bulbospinal respiratory axons at the level of the phren ic nuclei in adult rats: a possible substrate for the crossed phrenic phenomenon. Experiment al Neurology 111(1):135-139. 61

PAGE 62

Goshgarian HG, Ra fols JA. 1981. The phrenic nucl eus of the albino rat: a correlative HRP and golgi study. Journal of Compar ative Neurology 201(3): 441-456. Guillery RW. 2002. On counting and counting erro rs. Journal of Comparative Neurology 447: 17. Jackson AB, Groomes TE. 1994. Incidence of resp iratory complications following spinal cord injury. Arch Phys Med Rehabil 75: 270-275. Johnson RA, Mitchell GS. 2001. P-Chlorophenylalanine eliminates long-term modulation of the exercise ventilatory resp onse in goats. Respiratory Physiology 128: 161-169. Kim ES, Kim GM, Lu X, Hsu CY. Xu XM. 2002. Neural circuitry of the adult rat central nervous system after spinal cord injury: a st udy using fast blue and the Bartha strain of pseudorabies virus. Journal of Neurotrauma 19(6): 787-800. Lane MA, Fuller DD, White TE, Reier PJ. 2008a. Resp iratory neuroplasticity and cervical spinal cord injury: translational perspectives Trends in neuroscience, in press. Lane MA, White TE, Coutts MA, Jones AL, Sandhu MS, Bloom DC, Bolser DC, Yates BJ, Fuller DD, Reier PJ. 2008b. Journal of Comparative Neurology, in press. Loewy AD. 1998. Viruses as transneuronal tracers for defining neural circuits. Neuroscience and Biobehavioral Reviews 22(6): 679-684. Mantilla CB, Sieck GC. 2003. Invited review: M echanisms underlying motor unit plasticity in the respiratory system. J Appl Physiol 94(3):1230-1241. Martin PA, Mitchell GS. 1993. Long-term modulati on of the exercise ventilatory response in goats. Journal of Physiology 470: 601-617. Miller AD, Nonaka S, Jakus J, Yates BJ 1996. Modulation of vomiting by the medullary midline. Brain Research 737: 51-58. Mitchell GS, Johnson SM. 2003. Invited review: euro plasticity in respiratory motor control. Journal of Applied Physiology 94: 358. Moreno DE, Yu XJ, Goshgarian HG. 1992. Identification of the axon pathways which mediate functional recovery of a para lyzed hemidiaphragm following spinal cord hemisection in the adult rat. Experiment al Neurology 116(3): 219-228. Nantwi KD, El-Bohy AA, Schrimsher GW, Re ier PJ, Goshgarian H. 1999. Spontaneous functional recovery in a para lyzed hemidiaphragm following upper cervical spinal cord injury in adult rats. Neurorehabil itation and Neural Repair 13: 225-234. National Spinal Cord Injury Statistical Center. 2006. Spinal cord injury: facts and figures at a glance. Journal of Spinal Cord Medicine 29: 89. 62

PAGE 63

63 Nudo RJ. 2007. Postinfarct cortical plasticity and behavioral recover y. Stroke 38(2): 840-845. Paxinos G, Watson C. 1997. The rat brain in ster otaxic coordinates (c ompact 3rd ed.). San Diego: Academic Press, Inc. Porter WT. 1895. The path of the respiratory impulse from the bulb to the phrenic nuclei. Journal of Physiology 17:455-485. Shintani T, Mori RL, Yates BJ. 2003. Locations of neurons with respiratory-related activity in the ferret brainstem. Brain Research 974: 236-242. Yates BJ, Smail JA, Stocker SD, Card JP. 1999. Transneuronal tracing of neural pathways controlling activity of diaphragm motoneurons in the ferret. Neuroscience 90(4): 15011513. Zhou SY, Goshgarian HG. 2000. 5-Hydroxytryptopha n-induced respiratory recovery after cervical spinal cord hemisection in rats. Journal of Applied Physiology 89:1528.

PAGE 64

BIOGR APHICAL SKETCH Marcella Angelique Coutts was born in Miami, Florida in 1984. She graduated cum laude from the University of Miami in 2006 with dual degrees: A Bachelor of Arts in psychology with a minor in chemistry, and a Bachelor of Health Science in pre-pharmacy studies with a minor in Spanish. Upon entering the University of Flor ida Interdisciplinary Pr ogram in biomedical science in Fall 2006, Marcella was awarded the Al umni Fellowship. Marcella will begin medical school in Fall 2008.