Louisiana State University LSU Digital Commons
LSU Historical Dissertations and Theses Graduate School
1998 Effects of Anesthesia and Potential Antispastic Agents in an Experimental Model of Spasticity. Heather Colleen Mosser Louisiana State University and Agricultural & Mechanical College
Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses
Recommended Citation Mosser, Heather Colleen, "Effects of Anesthesia and Potential Antispastic Agents in an Experimental Model of Spasticity." (1998). LSU Historical Dissertations and Theses. 6853. https://digitalcommons.lsu.edu/gradschool_disstheses/6853
This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.
The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely afreet reproduction.
In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each original is also photographed in one exposure and is included in reduced form at the back of the book.
Photographs included in the original manuscript have been reproduced xerographically in this copy. Higher quality 6” x 9” black and white photographic prints are available for any photographs or illustrations appearing in this copy for an additional charge. Contact UMI directly to order. UMI A Bell & Howell Information Company 300 North Zeeb Road, Ann Arbor MI 48106-1346 USA 313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. EFFECTS OF ANESTHESIA AND POTENTIAL ANTISPASTIC AGENTS IN AN EXPERIMENTAL MODEL OF SPASTICITY
A Dissertation
Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy
in
The Department of Psychology
by Heather C. Mosser B.A., California State University, Long Beach, 1993 M.A., California State University, Long Beach, 1995 December 1998
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 9922102
UMI Microform 9922102 Copyright 1999, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized copying under Title 17, United States Code.
UMI 300 North Zeeb Road Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS
ABSTRACT...... iii
CHAPTER 1. INTRODUCTION ...... 1 1.1 Spasticity: Definition and Description...... 1 1.2 Segmental Spinal Cord Organization ...... 2 1.3 Spinal Cord Neurotransmission ...... 5 1.4 Proposed Mechanisms Underlying the Spastic Syndrom e...... 9 1.5 Mechanisms of Presynaptic Inhibition...... 15 1.6 Spinal Cord Injury. Experimental Assays ...... 24 1.7 Current Theories Concerning EAA Antagonists ...... 28 1.8 Applied Research...... 35 1.9 Purpose of the Present Study...... 36
CHAPTER 2. QUANTIFICATION OF SPINAL REFLEXES UNDER ISOFLURANE ANESTHESIA...... 39 2.1 Introduction...... 39 2.2 M ethods...... 41 2.3 Results...... 46 2.4 Discussion...... 50
CHAPTER 3. EFFECT OF KETAMINE ANESTHESIA ON SPINAL REFLEX MEASUREMENT ...... 55 3.1 Introduction...... 55 3.2 M ethods...... 56 3.3 Results...... 62 3.4 Discussion...... 74
CHAPTER 4. INVESTIGATION OF NMDA AND NON-NMDA RECEPTOR ANTAGONIST ACTIONS ON SPINAL TRANSECTION-INDUCED SPASTICITY...... 83 4.1 Introduction...... 83 4.2 M ethods...... 85 4.3 Results...... 91 4.4 Discussion...... 100
CHAPTER 5. CONCLUSIONS ...... 114
REFERENCES...... 117
VITA ...... 135
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT
Spasticity is a common motor disorder in spinal cord injured-patients. Therapeutic
agents used to treat the motor abnormalities east but are often unsatisfactory. Tests o f
potential pharmacotherapies are routinely conducted in anesthetized intact or acutely
spinal animals, conditions which do not accurately simulate the human situation. The
purpose of this dissertation was to develop, validate and use the chronic spinal rat as a
model of spasticity that closely approximates the human spastic syndrome to evaluate
antispastic agents.
The main goals of this research project were to: 1) demonstrate the development of
spasticity in the chronic spinal rat; 2) determine effects of anesthesia on spinal reflex
measurement; and 3) evaluate two proposed antispastics. These objectives were
accomplished by quantifying three parameters of the monosynaptic H-reflex: threshold,
magnitude and rate-sensitive depression. H-reflexes were elicited by tibial nerve
stimulation at three different frequencies and recorded from ipsilateral hindpaw plantar
muscles in intact and 28- and 60-day spinal rats. Spasticity was defined subsequent to
spinalization as decreased H-reflex thresholds (increased sensitivity), increased
magnitudes (hyperreflexia) and decreased rate-sensitive depression (reduced response
inhibition), relative to intact measurements.
In the first two studies, recordings were made in intact and 28- and 60-day spinal rats
under isoflurane, ketamine, or no anesthesia. Evidence is presented for the development
of spasticity following spinal thoracic transection and further, that H-reflex parameters
are affected by isoflurane and ketamine anesthesia. Response stability was established in
Hi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. an additional study in which H/M ratios and flexor reflexes were reliably recorded in
unanesthetized chronic spinal rats.
The final study investigated antispastic actions of excitatory amino acid receptor
antagonists by measuring changes in H-reflex magnitude and rate-sensitive depression
and in flexor reflex magnitude. Results indicate that selective N-methyl-D-aspartate
(NMDA) rather than non-NMD A antagonists may have a higher potential success rate in
treating spasticity, for two main reasons. The NMDA receptor antagonist used in the
present study was able to: 1) reduce H-reflex amplitudes at frequencies above the control
rate (0.3 Hz) and 2) increase rate-sensitive depression, both o f which are key factors in
the spastic patient.
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1. INTRODUCTION
1.1 Spasticity: Definition and Description
Spasticity is generally defined as a motor disorder characterized by a velocity-
dependent increase in muscle resistance to passive stretch associated with exaggerated
tendon jerks (Lance, 1980). This definition can be expanded to include positive and
negative symptoms. Positive symptoms result from the release of segmental rostral
control, whereas negative symptoms are due to the disconnection of lower motor centers
from higher ones. Besides increased tonic stretch velocity, positive symptoms include
exaggerated tendon jerks and cutaneous reflexes, flexor and extensor spasms, autonomic
dysreflexia, and dystonia. Negative symptoms include paresis, lack of dexterity,
weakness and fatigue. The increased muscle tone associated with spasticity is
involuntary and occurs during stretch, as opposed to the rigidity and absence of stretch
hyperreflexia observed with Parkinson’s disease. Although the spastic syndrome can
take on differing symptom profiles depending on the associated disease or injury, signs
of spasticity can be traced to circuitry at the level of the segmental spinal cord.
Regardless of whether a given lesion is associated with disease or traumatic injury to the
brain or the spinal cord, descending input from suprasegmental areas of the cord is
typically interrupted.
Changes in segmental reflexes following injury result from the plasticity inherent in
the cord. Physiological reactions to injury such as sprouting (McCouch, Austin, & Liu
1958), denervation supersensitivity (Charlton, Brennan, & Ford, 1981) and ineffective
synapse formation and/or modification of synaptic transmission (Cope, Hickman, &
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2
Botterman, 1988) lead to circuitry modifications that produce abnormal reflex
responses. Abnormal intemeuronal integration at the segmental level could provide the
mechanisms for increased stretch and hyperexcitable cutaneous reflexes, as well as the
muscle weakness associated with the spastic syndrome. Thorough understanding of
spinal cord neurons, their receptors, transmitters and modulators, and the intemeuronal
mechanismsresponsible for integrating segmental spinal cord circuitry would aid
development of more appropriate and beneficial therapies for patients suffering with
spasticity.
1.2 Segmental Spinal Cord Organization
Three basic functional cell groups within the segmental cord are active in motor and
reflex function: afferent fibers, efferent fibers, and intemeurons. Two main afferent fiber
types involved are Group la and lb fibers. Group la primary afferents are large,
myelinated fibers with receptors in muscle spindles. They respond to the stretch of
intrafusal fibers (which stretch with the elongation of extrafusal fibers) and are
responsible for detecting movement and proprioception (muscle length). Group la fibers
enter the dorsal hom and synapse on a-motoneurons and on intemeurons. Group lb
primary afferents are large, myelinated fibers that innervate Golgi tendon organs. They
are located between muscles and tendons and respond to force, i.e., the tension in
tendons evoked by muscle contraction. Group lb fibers enter the dorsal hom and
synapse on inhibitory intemeurons so as to generally cause inhibition of motoneurons to
extensor muscles and excitation of motoneurons to flexors.
Alpha motoneurons represent the efferent fibers in the cord segment. Their cell
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 3
bodies are located in the ventral hom and their axons extend to synapse on extrafusal
muscle fibers of homonymous and heteronymous muscles. Homonymous and
heteronymous refer to the agonist and antagonist muscle, respectively. The agonist
muscle represents the muscle that contains the muscle spindle from which the excitatory
signal was sent via the primary afferent to the a-motoneuron. The antagonist or
heteronymous muscle represents the opposing muscle group, e.g. the flexor that opposes
the agonist extensor. Intemeurons are dispersed within the grey matter of the spinal
cord neighboring both afferent and efferent neurons. Their position in the intermediate
grey and in the motor nucleus allows them to modify signal transmission. Central input
from the brain, peripheral input, i.e., proprioceptive/exteroceptive afferents, and
intersegmental (motoneuron collaterals) input all converge onto them. Spinal
intemeurons play important roles of integrating and relaying information within the
cord. Intemeurons are the primary source of input to a-motoneurons.
SPINAL REFLEXES
Exaggeration of spinal stretch reflexes and hyperexcitability of non-stretch spinal
reflexes are characteristically demonstrated in spastic patients. The monosynaptic
stretch reflex and the polysynaptic flexor reflex represent the stretch and non-stretch
reflex category, respectively.
The Monosynaptic Reflex
The monosynaptic reflex, also called the stretch or knee-jerk reflex involves a single
synapse between la afferent fibers and efferent a-motoneurons. The primary afferent
(la) nerve is stimulated by its receptors in the muscle spindle, enters the dorsal hom, and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4
forms an excitatory synapse with an a-motoneuron that in turn sends its axon out via the
ventral roots to synapse on the extrafusal fibers of the homonymous muscle, causing
contraction. The H-wave is the electrophysiological correlate of the mechanically-
induced monosynaptic stretch reflex (Hoffman, 1918). The H-reflex has been
established as a means to quantitatively measure Ia-a-motoneuron neurotransmission
(Burke et al., 1984; Wolpaw, 1994). Transection and other injuries to the spinal cord
alter this afferent-efferent communication pathway, disruption o f which is a key factor
in the development of spasticity.
The Polysynaptic Reflex
In addition to afferent and efferent neurons, the polysynaptic reflex incorporates
intemeurons, extending over several cord segments to activate different muscle groups.
The flexor reflex is a polysynaptic reflex that primarily serves a nocifensive function. It
is activated typically by painful stimuli which recruits necessary muscle groups to
withdraw a particular limb. Input from excitatory peripheral afferents and from
descending fibers enter the dorsal hom at laminae II and ID, where they are subject to
mediation by intemeuronal connections made in the spinal cord segments before their
information is conveyed to motoneurons. Afferents stimulated by cutaneous receptors
enter the dorsal hom of the spinal cord. The afferent fibers branch and synapse on
several excitatory intemeurons which in turn activate motoneurons at the entry level and
at one or more segments rostral and caudal to the initial entry point. In contrast to the
monosynaptic reflex, one or more intemeurons intervene between the primary stimulus
and the motor output. Intemeurons involved in the flexor reflex are located in the final
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5
common path and thus are involved in various other motor transmission pathways
including those that involve posture and locomotion. By definition, the polysynaptic
reflex pathway involves multiple synapses and thus by comparison to the monosynaptic
pathway, has a longer latency from stimulus to response.
1.3 Spinal Cord Neurotransmission
Spinal motor and reflex behaviors are the result of a complex balance between
excitation and inhibition within the segmental cord. Release of excitatory and inhibitory
neurotransmitters is under supraspinal regulation, and spasticity is associated with a
disruption of this delicate balance.
EXCITATORY
Glutamate is the main excitatory neurotransmitter in the central nervous system
(CNS). Within the spinal cord, both glutamate and aspartate function as excitatory
neurotransmitters. Glutamate and aspartate are released from primary afferents and
excitatory intemeurons, respectively (Curtis, Phillis & Watkins, 1960; Jessell, Yoshioka
& Jahr, 1986; Watkins & Evans, 1981).
Receptors for glutamate are subdivided into two groups: ionotropic (iGluRs) and
metabotropic (mGluRs). Within the ionotropic subgroup, there are three subunit
families: n-methyl-d-aspartate (NMDA) and two non-NMDA families, a-amino-3-
hydroxy-5-methyl-4-isoxazole-propionic acid (AMPA) and kainate (Ozawa, Kamiya &
Tsuzuki, 1998). NMDA receptors consist of five subunits: NR1 and NR2A-D; AMPA
receptors consist of GluRl-4 subunits; and kainate receptors are assembled from
GluR5-7, KA1, and KA2 subunits (Nakanishi, 1992; Wisden and Seeburg, 1993).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 6
NMDA Receptors
The NMDA receptor type was designated as such because of its selective activation
by the synthetic agent NMDA. The NMDA receptor complex consists of a recognition
site, an ionophore, and a modulatory unit (Foster & Fagg, 1987). Various agents that
bind at different sites are capable of modulating the receptor complex. Substances
binding to, for example, its glycine, polyamine, or Zn2* site (Peters, Koh, & Choi,
1987) can alter the response of the receptor-gated ion channel. Glycine, although
known as the major inhibitory transmitter in the spinal cord, is known to potentiate
ligand binding to the NMDA site as well as to enhance receptor-induced currents when
attached to its site on the NMDA complex in vitro (Johnson & Ascher, 1987). In vivo,
the complex is dependent upon glycine’s occupation of its site to produce a response
subsequent to ligand binding. Nevertheless, glycine is normally present in ample
physiological concentration to support receptor responses, i.e., NMDA responses are not
typically prevented by a lack of glycine binding. The presence of magnesium (Mg2*) at
physiological concentrations, however, blocks the NMDA receptor-gated ion channel in
a voltage-dependent manner. The NMDA receptor has relatively slow kinetics
compared to non-NMDA recptors. It requires an initial membrane depolarization to
displace Mg2* ions from its receptor complex for ligand-receptor interaction to occur.
This can be accomplished via non-NMDA receptor-gated ion channels, which rapidly
depolarize upon activation and allow influx of sodium (Na*) and potassium (K*) ions.
NMDA and non-NMDA receptors are referred to as high and low threshold and are
characterized by slow and fast excitatory transmission, respectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7
Non-NMDA Receptors
The grouping of non-NMDA receptors was based partially on the antagonist
properties of quinoxalinediones, which were developed by Honore and colleagues
(1988). Two quinoxalinediones in particular, CNQX (6-cyano-7-nitroquinoxaline) and
DNQX, distinguish between NMDA and non-NMDA receptor-mediated actions but
distinguish poorly or not at all between quisqualate and kainate responses in studies
comparing pA, values, dose ratio studies (Blake, Brown & Collingridge, 1988; Fletcher,
Martin, Aram, Lodge & Honore, 1988) and in electrophysiological studies (Fletcher et
al., 1988; Honore et al., 1988).
AMP A, a synthetic compound, is an isoxazole amino acid that has potent excitatory
action on spinal neurons (Krogsgaard-Larsen, Honore, Hansen, Curtis & Lodge, 1980).
AMPA and kainate receptors have common channel blockers, but the pharmacological
distinction between quisqualate, kainate and AMPA receptor-channel subtypes has not
been especially clear (Sommer & Seeburg, 1992). For example, arthropod toxins that
selectively reduce non-NMDA-mediated excitatory postsynaptic potential (EPSPs) do
not distinguish between quisqualate, kainate and AMPA depolarising actions (Lodge &
Johnson, 1990). When stimulated, kainate and quisqualate receptors show similar
responses and are similarly sensitive to antagonists. Kainate receptors have two
different sites, a high and low affinity site. Normal extracellular concentrations of Ca2+
decrease kainate binding to the high affinity site, thus it is not typically considered
functionally relevant. The low affinity site has the same molecular weight as the
quisqualate receptor and is best radio labeled by [3H] AMPA.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 8
Central NMDA and non-NMDA receptors are typically co-localized on post-
synaptic receptor membranes (Cotman, Managhan, Ottersen, & Storm-Mathison, 1987;
Fagg & Matus, 1984). Receptor co-localization has functional implications, for
example, with regard to plasticity, such as the participation of each of the receptors
subtypes in long term potentiation. There are exceptions, however, and not all
glutamatergic pathways use both receptor types.
INHIBITORY
The major CNS inhibitory neurotransmitters are gamma-aminobutyric acid (GABA)
and glycine. GABA is synthesized from glutamate and is considered the major
inhibitory transmitter throughout the brain, whereas glycine is the major inhibitory
transmitter in the spinal cord. Within the segmental cord, glycine is considered the
neurotransmitter most capable of producing inhibitory postsynaptic potentials (IPSPs),
yet GABA, which is located at both supraspinal and spinal sites (Curtis, Duggan, Felix
& Johnston, 1970; Malcangio & Bowery, 1996), has received a great deal of attention
from efforts to understand and pharmacologically balance excitatory and inhibitory
signals (Lin, Peng & Willis, 1996; Malcangio & Bowery, 1996; Matthews, Ayoub, &
Heidelberger, 1994; Maxwell et al., 1995; Misgeld, Bijak & Jarolimek 1995; Stuart &
Redman, 1992; Thompson & Wall, 1996). The degree of excitation elicited by la
afferents in the spinal cord is under supraspinal control. One way descending fibers
modify excitatory sensory transmission is by altering the amount of glutamate released
by la afferents. This is accomplished by a mechanism called presynaptic inhibition,
which is primarily associated with GABA (see discussion in Section 1.5).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9
1.4 Proposed Mechanisms Underlying the Spastic Syndrome
Spinal cord injury causes extensive alterations within the cord environment.
Segmental cell loss as well as loss of descending axonal influence initiates
reorganization of neural circuits caudal to the injury site. As a consequence to the
absence of supraspinal input, intemeuronal circuitry loses its specificity. The large
number of functions previously organized, refined, and executed via spinal circuits
becomes significantly reduced. The integrity of the remaining circuitry deteriorates and
gives rise to abnormal voluntary and reflex activity.
The pathophysiology of spasticity is not yet completely understood, but mechanisms
proposed to give rise to abnormal reflexes are typically related to dysfunctional
modulation (or loss) o f synapses where EAA neurotransmitters are involved. Despite its
name, the monosynaptic stretch reflex arc involves multiple factors, for example,
presynaptic input to la afferents. Primary sensory fibers receive input from several
sources, input that is typically communicated via intemeurons. Intemeurons relay
signals to la fibers that originate in supraspinal as well as peripheral areas. Other inputs
include antagonist la afferents, lb afferents, and neighboring intemeurons and Renshaw
cells. The efferent segment of the monosynaptic reflex arc receives excitatory signals
from la afferents and additionally, from axons and Betz cells of the primary motor
cortex. Alpha motoneurons also receive inhibitory signals from Golgi tendon afferent
fibers, heteronymous la afferents, and feedback from Renshaw cells. Clearly, the
excitatory signal sent from receptors in muscle spindles along la afferents to a-
motoneurons and back to the same muscle is potentially modified within the spinal cord
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 10
segment by several mechanisms. These include presynaptic inhibition of the primary
afferents, reciprocal inhibition, nonreciprocal inhibition and Renshaw inhibition. Three
types of inhibition may be decreased in spasticity: presynaptic inhibition, reciprocal
inhibition and non-reciprocal inhibition. Decreases in inhibition generated by Renshaw
cells, however, are not considered contributing factors in spastic paresis (Young, 1994).
DECREASED INHIBITION
Presynaptic Inhibition
Presynaptic inhibition of la afferents functions to adjust the level of excitatory
synaptic transmission. Excitatory signals from primary afferents activate a-
motoneurons as well as intemeurons. Regardless of mechanism, presynaptic inhibition
is accomplished by reducing the amount of neurotransmitter released from the afferent
fiber. A certain degree of presynaptic inhibition is normally present and under
supraspinal control. For example, during voluntary contraction, presynaptic inhibition
of la fibers of agonist (contracting) muscles is decreased while that of antagonist la
fibers is increased (Hultbom & Malmsten, 1983a; b). Additionally, the size of the
monosynaptic H-reflex can be conditioned in the rat (Chen & Wolpaw, 1995a) and in
humans (Rothwell, Day, Berardelli & Marsden, 1986). A rostral lesion results in
decreased presynaptic inhibition and a consequent increase in la afferent-generated
motoric responses. Reduced presynaptic inhibition is thought to contribute in large part
to the exaggeration and reduced rate modulation of reflex responses associated with
spasticity (Calancie, Broton, Klose, Traad, Difini & Ayyar, 1993; Faist, Mazevet, Dietz
& Pierrot-Deseilligny, 1994; Yang, Fung, Edamura, Blunt, Stein & Barbeau, 1991;
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 11
reviewed in Stein, 1995). For this reason, mechanisms of presynaptic inhibition are
discussed in further detail in Section 1.5 (see below).
Reciprocal (la) Inhibition
The reciprocal stretch reflex involves the monosynaptic reflex pathway with the
additional activation of an inhibitory intemeuron. The la afferent fiber activates: a) the
a-motoneuron of the homonymous/ agonist muscle (to elicit contraction) and b) an
intemeuron that in turn inhibits the a-motoneuron to the antagonist muscle (causing
relaxation).
Inhibitory intemeurons linked to the la afferents are also under descending control.
A rostral lesion could remove descending excitatory input to intemeurons and thus
remove their inhibitory effect on a-motor axons to antagonist muscle groups, resulting
in co-contraction. Such dysfunctional co-contraction of agonist and antagonist muscle
group pairs is thought to underlie excessive muscle contraction and increased la input to
the spinal cord.
Nonreciprocal (lb) Inhibition
Nonreciprocal inhibition is also called autogenic inhibition and it involves receptors
in the Golgi tendon organ rather than the muscle spindle. Golgi tendon receptors are
connected in series with extrafusal muscle fibers and send impulses via lb sensory fibers
in response to muscle tension. When relayed to the cord, they transmit reflex-inhibition
signals to the homonymous muscle (the opposite effect of muscle spindle-Ia afferent
action). Ib afferents excite inhibitory intemeurons to yield muscle relaxation by
inhibiting a-motoneurons. The excess tension present in a spastic muscle ‘over-excites’
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12
Golgi tendon receptors and a sudden whole-muscle relaxation, more commonly known
as the clasp-knife reflex (Clemente, 1978) can occur.
A change (i.e., decrease) in nonreciprocal inhibition is implicated in spastic dystonia
(increased tone), but not tendon hyperreflexia. This dissociation can at least in part be
explained by the function of the gamma loop. The monosynaptic stretch reflex has a
certain set point of muscle length that is determined or ‘set’ by gamma motoneuron
activity. The set point exists within a feedback, or gamma, loop involving
communication of gamma motoneurons and intrafusal fibers with a-motoneurons, la
afferents and extrafusal fibers; deviations from this set point are detected by the muscle
spindles. Descending input simultaneously activates both alpha and gamma
motoneurons. Supraspinal influence on the gamma feedback loop enables enhanced
capacity in the control of muscular contraction and tone by regulating the ideal set point
of muscle length. Without such control, a maladjusted set point could easily facilitate
the excess contraction known as hypertonia or dystonia.
INCREASED INHIBITION
Renshaw (/Recurrent) Inhibition
In contrast to the three types of inhibition that are decreased, the inhibitory activity
of Renshaw cells is not considered to be decreased but rather, increased in the spastic
patient (Shefiier, Berman & Sarkarati, 1992, although an alternate view has been
suggested by Mazzocchio & Rossi, 1997). Renshaw cells are inhibitory intemeurons
that act on both a-motoneurons and la intemeurons. Ventral root stimulation has been
shown to activate collateral axons of a-motoneurons which release acetylcholine onto
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13
Renshaw cells (Eccles, Eccles & Magni, 1961). Once an a-motoneuron is activated,
recurrent axon collaterals fire in a negative feedback loop designed to control excess cc-
motoneuron activity. Renshaw cells respond by releasing the inhibitory
neurotransmitter, glycine, which elicits inhibitory post-synaptic potentials (IPSPs) in
motoneurons. An increase in Renshaw inhibition could possibly contribute to the
decreased la reciprocal inhibition described above.
INCREASED EXCITATION
In a review of spasticity, Katz and Rymer (1989) summarily categorized the
evidence concerning mechanisms implicated in spastic hypertonia as being attributed to:
1) an increase in the excitability of motoneurons; and 2) an increase in the synaptic
excitation of motoneurons following stretch activation. Substantial support exists for
both. Increased excitability of motoneurons is manifested clinically as excitation
occurring at lower amplitudes or decreased velocities and as enhanced electrically-
evoked H-reflex amplitudes. Three mechanisms could be responsible for increasing
motoneuronal excitability: increased excitatory synaptic input, reduced inhibitory
synaptic input, and changes in intrinsic motoneuronal properties. Both an increase in
excitatory input and a decrease in inhibitory input result from removal of supraspinal
control that normally functions via intemeurons to balance excitation and inhibition in
the segmental cord. Alterations in intrinsic motoneuronal excitability, however, have
not been persuasively established in the chronic situation (Munson, Foehring, Lofton,
Zengel & Sypert, 1986; Nelson, Collates, Niechaj & Mendell, 1979). Collateral
sprouting, denervation supersensitivity and reduced presynaptic inhibition have all
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14
received support as mechanisms underlying the enhanced synaptic excitation of
motoneurons.
Increased Motoneuronal Excitability
Considerable projections originating from supraspinal centers make connections
with intemeurons (Davies & Watkins, 1985; lies, 1996; Lundberg, 1964; Rudomin,
1990) to balance the level of tonic excitation and inhibition. Without such supraspinal
control motoneurons likely become hyperexcitable, due to enhancement of synaptic
input by segmental afferents and excitatory intemeurons (Angel & Hoffman, 1963;
Matthews, 1966). This disruption of balance could cause motoneurons to exist at a
level of depolarization closer to their threshold than normal (Katz & Rymer, 1989).
Loss of descending inhibition following spinal cord injury (Magnusson, 1973), could
thus induce an increase in the hyperexcitability of the motoneuron pool (Funase,
Higashi, Yoshimura, Imanaka, & Nishihira, 1996).
Intrinsic motoneuron hyperexcitability has also been a suggested contributing factor.
Indeed, large increases in la EPSPs in both fast (gastrocnemius) and slow (soleus)
motoneurons have been reported to occur within hours of transection in spinal cats
(Cope et al., 1988; Nelson et al., 1979). However, spastic signs are not evident during
the acute phase (i.e., hours), and increased intrinsic excitability of a-motoneurons does
not appear to be a lasting change. Intracellular recordings from chronic spinal cats show
no significant enlargement of la EPSP amplitude (Munson et al., 1986; Nelson et al.,
1979). A lack of any significant change has been reported for the electrical properties of
motoneuron membranes in the chronic condition (Munson et al., 1986).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15
Increased Synaptic Excitation of Motoneurons
Altered synaptic output can result from collateral sprouting, receptor
supersensitivity, and reduced presynaptic inhibition and thus, could explain the
development o f hyperexcitable motoneurons. Presynaptic terminals respond to
lesioning by collateral sprouting and synaptogenesis, leading to an overload of
excitatory synapses (McCouch et al., 1958). Lesions to the cord are known to destroy
synaptic boutons by primary injury and/ or secondary degeneration. The initial decrease
or elimination of transmitter flow after injury induces homeostatic mechanisms to
attempt to upregulate post-synaptic membrane receptors (Charlton et al., 1981),
resulting in supersensitivity. A change in the chemical neurotransmitter composition of
presynaptic terminal vesicles following injury has also been recently suggested
(Kramarevsky, Anderson, Irish & Westrum, 1997). Finally, the loss of descending
fibers, such as those which mediate presynaptic inhibition (Calancie et al., 1993;
Eichenberger & Ruegg, 1984; Hultbom & Malmsten, 1983a; b; lies & Pisini, 1992),
inevitably increases the synaptic output onto motoneurons and results in enhanced
excitation. Reduction of presynaptic inhibition as well, occurs gradually (Ashby,
Verrier& Lightfoot, 1974).
1.5 Mechanisms of Presynaptic Inhibition
GABA’s role in presynaptic inhibition of primary sensory transmission has been
well-established (Aanonsen & Wilcox, 1989; Levy, 1977; Matthews et al., 1994; Nicoll
& Alger, 1979; Stuart & Redman, 1992). Supraspinally and peripherally activated
intemeurons produce presynaptic inhibition of la afferents by at least two mechanisms,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 16
both of which are GABA receptor-mediated. GABA acts on at least two receptor
subtypes, GABA a and GABAb, both of which have been identified in the spinal cord.
GAB A a sites are located on dorsal and ventral intemeurons as well as motoneurons.
GABA b sites are especially concentrated in the dorsal hom and are also located on
GAB Aergic intemeurons as autoreceptors. Sites for both receptor types are present on
presynaptic la terminals (Malcangio & Bowery, 1996) and each type produces
presynaptic inhibition in functionally different ways. (Bormann, 1988; Fischer &
Pamas, 1996b; Lin et al., 1996; Matthews et al., 1994; Stuart & Redman, 1992).
PRIMARY AFFERENT DEPOLARIZATION (PAD): GABA a RECEPTORS
The ‘classical’ mechanism of GABA-mediated presynaptic inhibition is that which
was suggested by Eccles et al. (1961): They associated presynaptic inhibition with
primary afferent depolarization (PAD) and proposed a GABAergic axo-axonic
mediation (as opposed to PAD mediated by elevated extracellular K+ accumulation; see
Levy, 1977; Rudomin, 1990). GABA-evoked PAD is described as follows: An inward
chloride (CL) pump in spinal cord neurons in the isolated frog spinal cord (reviewed in
Rudomin, 1990), newborn rat (Fulton, Miledi & Takahashi, 1980) and also likely in
humans, maintains an elevated intracellular Cl' concentration. The reversal potential of
Cl' in the human spinal cord is just positive to the resting membrane potential, making
the movement of CT ions (efflux) a depolarizing event (Bormann, 1988). When GABA
is released from intemeurons, it activates GABA a receptors on the la afferent terminal,
which cause an increase in membrane conductance to Cl* ions and thereby
depolarization of the la afferent (Hill & Bowery, 1981). Primary afferent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 17
depolarization, activated by GABA a receptors, results in a short-lasting inhibition, as
discussed below (Eccles, Schmidt & Willis, 1963; Kullmann, Martin & Redman, 1989;
Mody, Tanelian & Maclver, 1991).
REDUCTION OF PRESYNAPTIC Ca2+ INFLUX:GABA b RECEPTORS
GABA released by activated intemeurons can also induce presynaptic inhibition by
binding to presynaptic GABA b receptors. PresynapticGABA b receptors are Ca2*
channel-linked via a guanine nucleotide binding protein (G-protein) (Misgeld et al.,
1995). Activation of GABA b receptors causes inhibition of Ca2+ currents and thus
reduction o f Ca2+ ion influx triggered by an action potential (Bormann, 1988; Lev-Tov,
Myers & Burke, 1988; Misgeld et al., 1995). The result is a reduced amount of
glutamate released from the la afferent terminal. GABA a receptors are involved in an
axo-axonic synapse with la afferents, whereasGABA b receptors are suggested to be
located ‘extrasynaptically’, i.e., outside of the Ia-a-motoneuron synapse (Stuart &
Redman, 1992). In contrast to the short-lasting effects of GABA a receptors, GABA b
receptor-mediated presynaptic inhibition is proposed to have long-lasting effects on the
la synapse (Curtis & Lacey, 1994), as discussed below.
TWO TYPES OF PRE-SYNAPTIC INHIBITION
Reflex depression following afferent fiber stimulation has been studied using
several conditioning methods which result in inhibition of sensory transmission, i.e.,
depression of motor responses, without changes in motoneuronal excitability. For this
reason, a presynaptic mechanism is assumed (Hultbom, IUert, Nielsen, Paul, Ballegaard,
& Wiese, 1996). Two types of reflex depression are proposed, which differ in terms of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 18
stimulus activation characteristics (hetero- and homosynaptic stimulation) and duration
(short- and long-lasting).
Stimulation Characteristics
Heterosynaptic and homosynaptic stimulation. Heterosynaptically-evoked
presynaptic inhibition of motor responses is also referred to as reciprocal inhibition.
Measurement of reciprocal inhibition involves a pair of opposing muscle groups. The la
afferent to the heteronymous (antagonist) muscle, stimulated first, simultaneously
activates the heteronymous while inhibiting the homonymous muscle. The la afferent to
the homonymous (agonist) muscle is then stimulated, and the amount and duration of
the depressive effect of the conditioning stimulation on the H-reflex in quantified. The
H-reflex recovery curve is also utilized as an index of presynaptic inhibition, although
its use has been questioned (Kagamihara, Hayashi, Okuma, Nagaoka, Nakajima &
Tanaka, 1998). H-reflex recovery curves measure the latency at which the reflex
response returns to pre-test amplitude following conditioning stimulation, i.e., when it
recovers from the stimulation-induced depression (Crone & Nielsen, 1989; Nielsen,
Petersen & Crone, 1995). Homosynaptically-evoked depression can be assessed with
procedures such as repetitive electrical stimulation of homonymous la fibers (Crone &
Nielsen, 1989; Taborikova & Sax, 1968), mechanical stretch of the muscle (Neilsen,
Petersen, Ballegaard, Biering-Sorensen & Kiehn, 1993; Kohn, Floeter & Hallett, 1997)
and the tendon tap (Crone & Nielsen, 1989).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 19
Duration of Inhibition
Comparisons between the effects of homosynaptic and heterosynaptic activation of
la fibers indicate that two different types of reflex depression result: 1) Heterosynaptic
stimulation evokes a short-lasting (300-400 ms) reflex depression identified as
‘presynaptic inhibition’; and 2) repetitive homosynaptic stimulation elicites a long-
lasting (2-10 s) reflex depression that has been labeled ‘post-activation depression’ by
Crone and Nielsen (1989). However, there is currently a debate concerning the nature
of the reflex depression: extrinsic vs. intrinsic mediation. Extrinsic mediation refers to
extrasynaptic factors such as presynaptic inhibition; intrinsic mediation refers to factors
within the la afferent terminal itself.
Mechanism of Homosynaptically-Evoked Inhibition: Extrinsic vs. Intrinsic
One view states that the short-lasting reflex depression following heterosynaptic
stimulation results from extrinsically-mediated presynaptic inhibition of homonymous
la fibers. Further, the long-lasting depression after repetitive homosynaptic stimulation,
post-activation depression, is proposed to be a function of factors intrinsic to the la
afferent terminal itself (Crone and Nielsen, 1989; Kohn et al., 1997; Nielsen et al.,
1993; 1995). By general consensus, heterosynaptic stimulation elicits classic
presynaptic inhibition (PAD) mediated by GABA a receptors. A broader perspective of
monosynaptic sensory transmission, however, suggests that both hetero- and
homosynaptically-evoked reflex depression represent forms of presynaptic inhibition. It
is argued that ‘post-activation depression’ does not arise from intrinsic mediation alone,
but that rather, factors extrinsic to the la afferent terminal might be considered.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 20
Low-F requency/Rate-Sensitive Depression
Post-activation depression following repetitive homosynaptic la stimulation is
comparable to ‘low-frequency’ or ‘rate-sensitive’ depression. Low-frequency/rate-
sensitive depression is considered to be an expression of presynaptic inhibition (Curtis
& Eccles, 1960; Kaizawa & Takahashi, 1970; Lloyd, 1957; Lloyd & Wilson, 1957). It
occurs upon elicitation of the monosynaptic H-reflex at frequencies greater than 0.3 Hz,
and the pattern of reflex depression is displayed by decreased response amplitudes with
increased stimulus frequencies (Curtis & Eccles, 1960; Eccles & Rail, 1951; Jefferson
& Schlapp, 1953; Lloyd, 1957). The mechanism responsible for decreased reflex
amplitude is fairly unanimously agreed upon: quantal release of neurotransmitter by la
afferents is reduced (Kuno, 1964). However, the mechanism responsible for reducing
the release is of some debate, i.e., post-activation depression vs. presynaptic inhibition,
or intrinsic vs. extrinsic mediation (Crone & Nielsen, 1989; Faist et al., 1994; Nielsen et
al., 1993; 1995; Kohnet al., 1997; Voigt & Sinkjaer, 1998).
Differential Activation of GABA Receptors
GAB A receptors can be differentiated based on the duration of inhibition produced
by their activation and by the type of stimulation used to activate them. Reflex
depression that results from repetitive homonymous la stimulation (rate-sensitive
depression) outlasts the depression that results from heterosynaptic conditioning
stimulation (Nielsen et al., 1993; 1995). Likewise, repetitive homonymous depression
outlasts the time course of GAB AA-mediated presynaptic inhibition (Fischer & Pamas,
1996b; Lev-Tov et al., 1988), coincident with the long-lasting GABA b receptor effects
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 21
on the la synapse (Curtis & Lacey, 1994). AlthoughGABA a receptors have been
considered the chief mediators of Ia-a-motoneuron presynaptic inhibition (Stuart &
Redman, 1992), a role for GABA b receptors in presynaptic inhibition has not been
excluded. For example, increases in presynaptic inhibition have been found following
administration of a GAB AB receptor agonist. Specifically, baclofen has been shown to
reduce monosynaptic EPSP amplitudes (Lev-Tov et al., 1988) by reducing Ca2+ influx,
which reduces the probability of transmitter release from la afferent terminals (Curtis,
Gynther, Lacey & Beattie, 1997). Conversely, a decrease in presynaptic inhibition has
been shown following administration of a GABA b receptor antagonist, CGP-35348
(Curtis, 1998).
GABA receptor subtypes are differentially activated as a function of stimulation
frequency (Fischer & Pamas, 1996b). In addition to their extrasynaptic location, a slow
binding on-rate of GABA b receptors compared toGABA a receptors (Fischer & Pamas,
1996a) has been considered as a possible explanation for this finding. An extrasynaptic
location and slower binding on-rate would require a greater amount of GABA release
(Otis & Mody, 1992) and a longer window of opportunity for binding to allowGABA b
receptors to produce inhibition and in turn, reflex depression. An increased amount of
GABA could be achieved (Baxter & Bittner, 1991) and a longer time period could be
provided by repetitive stimulation (Fischer & Pamas, 1996b). Frequency-dependent
actions of baclofen and CGP-35348 have been revealed in recent reports concerning
presynaptic modulation of transmitter release and frequency-related changes in la
monosynaptic EPSP amplitudes (Lev-Tov et al., 1988; Peshori, Collins & Mendell,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 22
1998). That is, at the lower frequencies used, baclofen reduced and CGP-35348
increased EPSP amplitudes, whereas the opposite was the case at the higher frequencies
(Peshori et al., 1998).
CHANGES IN INHIBITORY MECHANISMS FOLLOWING INJURY
Regardless of the type o f depression mechanism, spinal cord damage does not
discriminatebetween presynaptic/ short acting inhibition and post-activation/long
lasting depression. Both types of H-reflex depression are reduced in spasticity (as a
result of spinal cord injury, Calancie et al., 1993; Nielsen et al., 1993; multiple sclerosis,
Crone & Nielsen, 1994; Nielsen et al., 1995; and spinal spasticity of mixed etiology,
Delwaide, 1985). Reduction in H-reflex depression associated with stimulus rate
increases has been reported in experimental spinal cord injury such as contusion
(Thompson, Reier, Lucas & Parmer, 1992) and transection (Skinner, Houle, Reese,
Berry & Garcia-Rill, 1996) injury in the rat and hind limb rigidity in the dog (Gelfan,
1966), though no reduction in presynaptic inhibition via decreased reciprocal inhibition
is evident in chronic hemisected cats (Hultbom & Malmsten, 1983b). This could
perhaps be due to cross-over compensation/collateral sprouting of descending fibers,
allowing maintained synaptic efficiency of presynaptic inhibitory circuits. Changes in
reciprocal inhibition following spinal cord injury are likely the result of altered
regulation of intemeuronally-mediated heterosynaptic inhibition, and given the known
effects of injury on segmental spinal cord composition, it is likely that the
concommittant reduction in rate-associated H-reflex depression subsequent to injury is
the result of dysfunctional intemeurons as well.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 23
Rate-dependent alterations in H-reflexes have been revealed and linked to functions
of GABAergic intemeurons. Thompson, Parmer and Reier (1996) demonstrated that
administrationof a GABAb receptor antagonist (CGP-35348) to normal, intact rats
resulted in frequency-dependent alterations in H-reflex amplitudes such that responses
of the intact rats treated with the antagonist CGP-35348 resembled those of untreated
spinally-injured rats. Rate-dependent modulation of la EPSPs is mediated via GABAb
receptors in normals (Curtis, 1998; Lev-Tov et al., 1988; Peshori et al., 1998), thus the
change, i.e., decrease, in rate-sensitive depression that occurs as a result of spinalization
is most likely related to reduced GABAergic presynaptic inhibition that is mediated via
GABAb receptors (Thompson et al., 1996).
SOURCES OF PRESYNAPTIC INHIBITION
Two potential sources contribute to presynaptic inhibition o f la terminals:
(supraspinal and spinal, Rudomin, Jimenez, Solodkin & Duenas, 1983): 1) Supraspinal/
descending inhibition: axons from corticospinal, rubrospinal, and vestibulospinal
pathways all synapse onto (and excite) GABAergic intemeurons. The activated
intemeurons synapse with primary afferents and presynaptically inhibit la synaptic
transmission. 2) Spinal/segmental: muscle spindle activation of la afferent fibers, the
stimulus for the monosynaptic reflex, gives rise not only to activation of a-motoneurons
but also to intemeurons, by collateral la branches. Diffuse connections are made among
intemeurons in the spinal cord, and intemeurons are the main source of input to a-
motoneurons. Because of the immense communication network among spinal cord
intemeurons, it is reasonable to suspect at least partial mediation by NMDA receptors
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24
located on intemeurons. It has been suggested that presynaptic inhibition o f la afferent
transmitter release is the end result of a sort of feedback action. Ia activity is proposed
to generate, via a few intemeurons, an auto-presynaptic inhibition of its own subsequent
activity. Through such a chain of intemeuronal synapses, NMDA receptor-mediated
intemeurons (and hence, NMDA receptors) could participate in this inhibitory influence.
1.6 Spinal Cord Injury: Experimental Assays
Experimental investigation of spinal and supraspinal mechanisms and the
evaluation of potential clinically effective antispastic agents necessitate the use of
animal models. Classic animal models include crush injury, genetically-induced
spasticity, acute and chronic spinal cord hemisection, and complete spinal transection in
acute (DeLa Torre, 1984) and chronic animals.
CONTUSION INJURY
Contusion injury as an experimental model began in 1911 with the development of
the weight-drop method (Allen, 1911). Modifications and new experimental protocols
have since improved the injury’s reproducibility and decreased the animal-to-animal
variability (Wrathall, Pettegrew & Harvey, 1985). Severity of impact correlates with
amount of damage to the cord (Blight & Decrescito, 1986), as do structural damageand
behavioral function (Noble & Wrathall, 1985). The nature of the compression/
contusion lesion make it a good structural analogue to human injury situations, which
are often of the compression/contusion type. However, animals surviving a spinal cord
crush injury regain locomotor behavior within weeks (Thompson et al. 1992), which is
not typical of human patients. Also, the ratios of structuraldamage to behavioral
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25
function in rats do not necessarily correlate well with the structure: function relationships
seen in humans. For example, a rat with a crush injury rated moderate, on a scale of
mild to moderate to severe, is able to eventually regain hindlimb locomotor function; a
mild structural crush injury renders relatively transient symptoms of spinal cord
dysfunction. By contrast, whereas structural damage to a relatively small area of spinal
cord white matter can result in functionally complete disability in human patients,
almost total destruction of a section of the rat spinal cord is required to elicit the same
functional recovery prevention. This can be explained in part by the anatomical
differences in corticospinal tract location between rats and humans. In humans, fibers
of the corticospinal tract descend anteriorally and laterally, whereas the rat corticospinal
tract projects dorsomedially and is identified in white matter at the most ventral level of
the spinal cord dorsal columns (Brown, 1971). Hence, a contusion injury induced by a
weight-drop onto the dorsal surface of the cord would necessarily obliterate a
considerable amount of white matter before destroying all corticospinal tract fibers.
GENETIC SPASTICITY
One laboratory (L. Turski and colleagues) uses a strain of Wistar rats that are
genetically determined to develop spontaneous spastic paresis (Pittermann, Sontag,
Wand, Rapp & Deerberg, 1976). Spastic Han Wistar rats display a wide range of
abnormalites including increased muscle tone, progressive bradykinesia, and ataxia.
They can only be tested at 10-12 weeks of age, a period which is followed shortly by
rapid degeneration and death. The mutant Wistar strain was previously suggested as a
questionable model of basal ganglia dysfunction (Turski, Schwarz & Sontag, 1984).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 26
These rats demonstrate disturbed GABAergic neurotransmission (Schwarz, Loscher,
Turski & Sontag, 1985) and altered cerebellar EAA function and morphology (Cohen et
al., 1991). Moreover, they exhibit profound hippocampal and cerebellar degeneration
(Wagemann, Schmidt-Kastner, Block & Sontag, 1991), atypical forebrain and cerebellar
glial patterning (Wagemann, Schmidt-Kastner, Block & Sontag, 1995), and have an
unusual CNS myelin composition (Fagg, Neuhoff, Sontag & Osbome, 1978). To
evaluate effects of certain drugs on muscle hypertonia, spontaneous EMG is recorded
from the gastrocnemius muscle of genetic spastic rats, which are used for comparison of
drug effects on spinal reflexes in normal rats (Klockgether, Schwarz, Wullner, Turski &
Sontag, 1989; Schwarz, Klockgether, Turski & Sontag, 1988; Schwarz, Klockgether,
Wullner, Turski & Sontag, 1992; Turski, Schwarz, Turski, Klockgether, Sontag &
Collins, 1985a; Turski, Schwarz, Turski & Sontag, 1985b; Turski, Bressler, Klockgether
& Stephens, 1990). These rats are apparently not suitable for spinal reflex recording
and evaluation, however, and measuring the effect of a drug on muscle hypertonicity is
perhaps of somewhat limited clinical value in the evaluation of antispastic agents.
SPINAL CORD HEMISECTION
Much work has been carried out using the hemisection model of spinal cord injury
(e.g. Hultbom & Malmsten, 1983a; b). Here, a lateral half of the cord is cut or removed
and side-to-side comparisons of recovery of function, reflexes, etc. can be made using
the animal as its own control. Hemisected animals show good motor recovery and
do not exhibit symptoms of the spastic syndrome (e.g., increased tendon jerks, tonic
stretch reflexes, and clasp-knife phenomenon, Hultbom & Malmsten, 1983a).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27
COMPLETE SPINAL CORD TRANSECTION: ACUTE PREPARATION
Studying an acute preparation differs greatly from studying an animal that has been
spastic for weeks or months. The acute preparation typically involves a certain amount
of physiological manipulation that can include transection (often at the cervical cord
level), decerebration, drug-induced paralysis, aortic occlusion, artificial ventilation,
blunt dissection and nerve separation and mounting (e.g., Simpson, Gondo, Robertson
& Goodman, 1995; Farkas & Ono, 1995). There is a vast discrepancy between acute
preparations and the human spastic patient As noted earlier, spasticity develops over a
period of weeks; spinal cord-injured humans are typically in spinal shock (arreflexic or
hyporeflexic) during the time period that acute testing is carried out in animals. In
general, acute experimental preparations are physiologically dissimilar to the spastic
patient.
COMPLETE SPINAL CORD TRANSECTION: CHRONIC PREPARATION
The chronic spinal rat is a model in which complete transection of the cord is
performed, typically at the thoracic level. The animal is subsequently maintained for a
period of at least one month after which recovery of function, spinal reflexes, and other
parameters can be evaluated. Although not a commonly used model in past research,
classic characteristics of human spasticity have been recently demonstrated in the
chronic spinally-transected rat (Skinner et al., 1996; also see Chapters 2 and 3). The
chronic spinal rat has been shown to produce reliable, replicable scores on certain
spastic indices and can be tested without the possible confounding effect of anesthesia
(see Chapter 3),
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 28
making it an useful model for studying dysfunctional spinal reflex circuitry and evaluating
antispastic agents.
1.7 Current Theories Concerning EAA Antagonists
NON-NMDA and NMDA RECEPTOR-MEDIATED FUNCTION IN THE SPINAL CORD
Glutamate, the major excitatory neurotransmitter in the CNS, is released from
primary afferentterm inalsand has a high number of receptor binding sites in the spinal
cord (Greenamyre, Young & Penney, 1984). Both NMDA and non-NMDA glutamate
receptor subtypes are found on neurons within the spinal cord. L-glutamate released
following impulses in low threshold afferent fibers is thought to act at non-NMDA
receptors, whereas L-aspartate, released following activation of excitatory intemeurons,
is considered to act at NMDA receptors (Davies & Watkins, 1983; Jessell et al., 1986;
Watkins, 1984). The functional separation of receptor types with regard to spinal
neurotransmission in monosynaptic and polysynaptic pathways began with studies
utilizing EAAs, continued with the aid of wide spectrum EAA receptor antagonists and
then NMDA receptor antagonists, which were followed by non-selective and recently,
selective, non-NMDA receptor antagonists.
Identification and evaluation of spinal cord mono- and polysynaptic responses
originated with studies that measured a-motoneuron EPSP latencies in response to
dorsal root stimulation. Dorsal root stimulation evokes a compound response in the
ventral root of the spinal cord. It is recorded from the a-motoneuron and typically
consists of two distinct waves or components. The short latency component is the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 29
monosynaptic reponse (MSR) and the longer latency component is the polysynaptic
response (PSR), the logic being the more synapses involved in a circuit, the longer the
latency of the response.
The first indication of a functional separation between non-NMDA and NMDA
receptors was the report of a selective depression of NMDA receptor-mediated
responses of intemeurons: The excitatory synaptic activation of Renshaw cells by dorsal
roots was depressed by NMDA antagonists which had no effect on the excitatory
synaptic activation of Renshaw cells by collateral acetylcholine-releasing axons in the
ventral root (Davies & Watkins, 1979; Lodge, Headley & Curtis, 1978). Similar in vitro
studies recorded dorsal root-evoked ventral root potentials in a-motoneurons and found
that application of an NMDA antagonist selectively depressed the late component of the
composite EPSP ventral root response (Padjen & Smith, 1980). Moreover,
investigations of the late component of the ventral root response revealed inhibition of
the PSR by NMDA receptor antagonists (Fletcher et al., 1988; Watkins & Evans, 1981).
Studies investigating the early component of the ventral root response have conversely
demonstrated MSR inhibition by non-NMDA receptor antagonists, namely
quinoxalinediones (Fletcher et al., 1988; Long, Smith, Siarey & Evans, 1990; Watkins,
1984).
The dogma concerning glutamate receptor subtypes and the spinal reflex pathways
is thus: non-NMDA receptors mediate monosynaptic circuits and NMDA receptors
mediate polysynaptic circuits. Chemosensitivity tests of dorsal hom neurons that
receive monosynaptic sensory input showed that they were highly and rapidly
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 30
depolarized by glutamate, kainate, and quisqualate (the latter two being specifically
associated with the non-NMDA type) yet were insensitive or minimally affected by
NMDA and L-aspartate (Jessell et al., 1986). Further, the L-glutamate response evoked
in neurons linked monosynaptically to la sensory input could not be antagonized by 2-
amino-5-phosphonovalerate (APV), a potent, selective NMDA receptor antagonist
(Collingridge & Lester, 1989). Combined, results from several reports indicate
exclusive involvement of NMDA receptors in polysynaptic components and of non-
NMDA receptors in monosynaptic mediation. Finally, an indirect action, e.g. via one or
more intemeurons, is suggested, if any, for NMDA receptors on dorsal hom neurons and
a-motoneurons (Jessell et al., 1986). In summary: 1) glutamate and aspartate are
considered to be the EAA transmitters that mediate fast and slow EPSPs, respectively, at
primary afferent synapses; and 2) the general model of glutamatergic activity in spinal
reflex circuits is as follows: The monosynaptic reflex is mediated by non-NMDA
receptors, and the polysynaptic reflex is mediated by NMDA receptors (Davies &
Watkins, 1983; Jessell et al., 1986; Watkins, 1984).
Considerable evidence from studies that 1) utilize intracellular recordings of
presynaptic la afferent fibers and postsynaptic a-motoneuron EPSPs, 2) measure
components of electrophysiological response wave latencies, and/or 3) use EAA
stimulation in combination with glutamate receptor subtype antagonists has led to the
general acceptance of and adherence to a dissociative model of EAA receptor
involvement in mono- and polysynaptic spinal reflex circuitry. Despite the extensive
list of supporting references, the currently accepted model of EAAs and EAA receptor-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31
mediated action in spinal cord reflex circuits (that monosynaptic reflexes are mediated
by non-NMDA receptors and polysynaptic reflexes are mediated by NMDA receptors)
deserves a second look. Recent reports, as well as some previous findings, invite re-
evaluation of the standard, for the following reasons: First, the bulk of information
substantiating this model arises from studies conducted before specific NMDA and non-
NMDA receptor antagonists were discovered (Davies, 1988). This model was first
suggested based on experiments which used non-specific EAA receptor antagonists and
tested the effects on dorsal hom neurons (Davies & Watkins, 1979; 1983). Studies that
followed tested the compounds available when the experiments were conducted. For a
number of studies, this meant using an EAA antagonist selective for NMDA receptors
and comparing its action to those of non-selective NMDA antagonists.
Development of quinoxalinediones, established as potent and selective non-NMDA
antagonists (Honore et al., 1988), enabled expansion of study in the area. For example,
Turski and colleagues compared the effect of CNQX, a quinoxalinedione, to (3-((±)-2-
carboxypiperazin-4-yl)propyl-l-phosphonic acid [CPP]), the most potent, selective
NMDA receptor antagonist (Lehman et al., 1987), in vivo. They reported results that
were in line with the model: CPP decreased the polysynaptic flexor reflex, CNQX
decreased the monosynaptic H-reflex, and neither drug affected the alternate response
within a restricted dose range in intact, anesthetized rats (Turski et al., 1990). Long et
al. (1990) tested the effect o f CNQX and APV in vitro and reported that CNQX
antagonized an NMDA-evoked depolarization. Additionally, their data showed that
CNQX markedly depressed a long-lasting component of the ventral root reflex
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 32
corresponding to the component labeled as the PSR by others (e.g. Farkas & Ono,
1995), although the authors did not report it as such. Conflicting reports betweenin vivo
and in vitro studies are not uncommon, still, the above-cited studies remain significant.
As noted by Farkas and Ono (1995), in electrophysiology, the designation of the terms
MSR and PSR have been based on an estimation of synaptic delay and on the
assumption that a single transmitter substance is acting on a single receptor. This is
important to note when considering effects of receptor antagonist actions on cells
participating in the complex communication network of the segmental spinal cord.
Quinoxalinediones were used in several studies, findings of which propagated the
receptor-reflex function generalization. Since then, it was discovered that
quinoxalinediones, namely CNQX, are not very selective for non-NMDA receptors
(CNQX also binds to the glycine site on the NMDA receptor complex [Brugger, Wicki,
Nassenstein-Elton, Fagg, Olpe & Pozza, 1990; Collingridge & Lester, 1989;
Sheardown, Nielsen, Hansen, Jacobsen & Honore, 1990; Siarey, Long & Evans, 1991]).
The development of potent, selective non-NMDA receptor antagonists has occurred
only relatively recently, and the number of studies conducted to test them pales by
comparison to the number of reports compiled before they were discovered. Although
the stated model continues to appear in current publications, evidence is available to
suggest alternatives. It is conceivable, given the complex cellular communication at the
segmental level of the cord, that both receptor subtypes are involved to some degree in
mediating both spinal reflexes.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 33
Interestingly, certain research findings support this possibility. For example, under
both in vivo and in vitro conditions, NMDA antagonists have been reported to affect the
MSR (CPP and MK-801, in vivo, Farkas & Ono, 1995) as well as the monosynaptic
EPSP (APV, in vitro. King, Lopez-Garcia and Cumberpatch, 1992). The idea that non-
NMDA receptors may contribute to reflex activity other than simply the monosynaptic
reflex arc is intuitively appealing. By convention, NMDA and non-NMDA receptors
are high and low threshold, respectively. Thus, for NMDA receptors to be involved in
reflex response mediation, they must be activated, and they require a greater stimulus to
become activated than do lower threshold non-NMDA receptors. It could be argued that
if NMDA receptors have been activated, then non-NMDA receptors may also have been
activated. This could explain the extensive role of non-NMDA receptors in the
mediation o f mono-, di-, and polysynaptic spinal reflex components such as was
reported by Farkas and Ono (1995).
The general view is that NMDA receptors are involved solely in polysynaptic
components/ reflexes and play no role in monosynaptic mediation. From this it can be
assumed that NMDA receptors are either non-existent on postsynaptic a-motoneuron
terminals or that they are present, but distributed separately from non-NMDA receptors
at synaptic contacts such that they are not functionally or electrotonically coupled to
mediate direct la afferent-a-motoneuron transmission. The latter seems highly
probable, given the precision of CNS wiring. The monosynaptic la afferent-a-
motoneuron connection from muscle spindle to homonymous extrafusal fibers was
likely established early in life and is well preserved in the healthy adult. As a result of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 34
such precise wiring, a single la afferent will accurately select from the pool of
motoneurons and synapse with every a-motoneuron that innervates the homonymous
muscle. These la afferents that make direct synaptic contact with a-motoneurons
release glutamate, which preferentially activates non-NMDA receptors on the
postsynaptic neuron. Still, non-NMDA receptor ligands have also been shown to inhibit
NMDA action in the mouse brain and rat spinal cord (Turski, Klockgether, Schwarz,
Sontag & Meldrum, 1987) and to modulate NMDA action in the mouse spinal cord
(Raigorodsky & Urea, 1990).
‘Monosynaptic’ provides an accurate fundamental definition of primary anatomy: A
single synapse exists between the primary sensory input and motor output neurons.
However, in physiological terms, ‘monosynaptic’ restricts functional characterization.
For example, di- and trisynaptic connections between la afferents and homonymous
motoneurons have been demonstrated in the cat (Fetz, Jankowska, Johannisson &
Lipski, 1979; Jankowska, McCrea & Mackel, 1981). Studies in humans have also
provided evidence indicating that the H-reflex is not a simple monosynaptic reflex
(Bimbaum & Ashby, 1982; Burke, Gandevia & McKeon, 1984; Van Boxtel, 1986). A
study in which composite EPSP latencies of tendon jerk and H-reflex responses were
measured in humans revealed a long rising phase and a long latency to motoneuron
discharge, indicating sufficient time for mulitsynaptic contributions from intemeurons,
which are not directly implicated in the monosynaptic pathway (Burke et al., 1984). In
other words, the monosynaptic Ia-a-motoneuron pathway is subject to polysynaptic
influence via intemeuronally-mediated presynaptic inhibition of the la fiber terminals.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 35
1.8 Applied Research
PHARMACOLOGICAL APPROACHES TO REDUCING SPASTICITY
Facilitation/ Enhancement of Inhibition
Currently, baclofen is the most commonly prescribed drug in the treatment of
spasticity. Baclofen is aGABA b agonist that acts to mimic the inhibition that GABA
effects via that receptor (as described earlier). In the spinal cord, presynapticGABA b
receptors located on la terminals mediate presynaptic inhibition. When activated by an
agonist such as baclofen, presynaptic inhibition is increased and sensory transmission is
reduced. Baclofen also acts to directly inhibit a-motoneuron EPSPs. The effects
reported in humans (Delwaide, 1985; Rice, 1987) and in normal, anesthetized rats
(Schwarz et al. 1988a; b) include reduction of both monosynaptic H-reflexes and
polysynaptic flexor reflexes. Nonetheless, baclofen is not without adverse effects.
Along with problems in determining effective doses, a major complaint is muscle
weakness. Other common side effects include sedation, nausea and hallucination
(Young, 1994). Development of new antispastic agents with different mechanisms of
action that provide improved efficacy and better tolerability would be welcome in
clinical therapeutic regimens.
Inhibition/ Reduction of Excitation
EAA antagonists. Glutamate, aspartate, and possibly other EAAs are implicated in
descending and segmental spinal cord mechanisms, and consequently, EAAs have also
been implicated in spasticity. EAAs are the identified neurotransmitters in segmental
activity related to monosynaptic and polysynaptic pathways. Descending tracts
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 36
the rubrospinal, vestibulospinal, and corticospinal tracts utilize EAAs to convey
supraspinal-to-spinal information and regulation (reticulospinal tracts are believed to
primarily use catecholamines; see Rudomin, 1990; also Young, 1994). EAA receptor
antagonists, thus, have theoretical potential as antispastic agents.
1.9 Purpose of the Present Study
DEVELOP AND VALIDATE A PRECLINICAL MODEL OF SPASTICITY
The symptoms of spasticity including exaggerated monosynaptic and polysynaptic
reflexes can be attributed to dysfunctional interneuron activity that develops over time in
response to injury. Development o f clinical spasticity is secondary to injury, coincident
with anatomical and physiological changes that follow primary tissue damage.
Secondary injury is the damage that ensues as a result of the original insult, and it takes
place over time in concert with ‘adaptive’ mechanisms in response to the injury such as
collateral sprouting and synaptogenesis.
Most spastic patients suffer due to an interruption of descending pathways which, in
healthy subjects, control intemeurons to adjust circuit transmission and produce
appropriate motor and reflex function. The first specific aim of this dissertation was to
develop a model of spasticity that closely resembles the human spastic syndrome, a
model which: 1) exhibits a syndrome of quantifiable spastic symptoms that parallel those
observed in human patients which can be objectively evaluated without anesthesia; 2)
develops over a period of weeks, and 3) remains as a chronic condition. To this end,
Chapters 2 and 3 are devoted to description, quantification and validation of the chronic
spinal rat as an experimental model of spasticity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37
EVALUATE THE EFFICACY OF POTENTIAL ANTISPASTIC TREATMENTS
Pharmacologic manipulation of neurotransmitters and neuromodulators known to be
used by neurons and interneurons within the segmental cord is both instructive and useful
for basic and applied purposes. Therapeutic efforts for treating spasticity are aimed at
achieving a balance between excitation and inhibition. Part of the present dissertation is
focused on evaluation of the proposed benefits of therapeutics aimed at decreasing
excitation. Experiments described in Chapter 4 investigate the involvement of excitatory
amino acids in segmental spinal cord reflex circuitry by asking two questions: 1) Are
excitatory amino acid receptor antagonists effective antispastic agents in the chronic
spinal rat; and 2) Is the rate-associated depression o f the monosynaptic reflex response a
function of a homosynaptic or heterosynaptic mechanism (or, in other words, does the
monosynaptic reflex pathway truly operate in physiological isolation)?
With respect to the first question, studies done using intact, anesthetized rats
indicate that EAA receptor antagonists are effective at selectively reducing the H-reflex
(non-NMDA) and the flexor reflex (NMDA) (Schwarz, Block & Sontag, 1992; Turski et
al., 1987a; Turski, Klockgether, Sontag, Herrling & Watkins, 1987; Turski et al., 1990;
Turski, Jacobsen, Honore & Stephens, 1992). Experiments described in Chapter 4
sought to determine whether such findings could generalize to chronically spinal rats by
evaluating effects of a non-NMDA and an NMDA receptor antagonist on the
monosynaptic H- and polysynaptic flexor reflexes.
Concerning the second question, hypotheses from the literature follow that, if H-
reflex rate-sensitive depression is purely homosynaptic and intrinsically-motivated, then
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 38
an NMDA receptor antagonist would not affect the depression (given the general model
of spinal EAA receptor-reflex mediation). Conversely, alteration of H-reflex rate-
sensitive depression by an NMDA antagonist would indicate intemeuronal involvement
and thus, an extrinsic mechanism. An alteration in rate-sensitive depression produced
by a non-NMDA receptor antagonist, however, would not likely provide support for a
homosynaptic or heterosynaptic mechanism. To address these issues, effects of a
potent, selective non-NMDA and NMDA receptor antagonist on amplitude and rate-
sensitivity of the monosynaptic H-reflex were compared in 28-day chronic spinal cord
injured rats.
Because the experiments described in Chapter 4 explore the roles of two types of
EAA receptor antagonists in spinal reflex pathways known to be abnormal in spastic
patients, research on their ability to reduce spastic symptoms could potentially make a
significant contribution to the investigation of spasticity. Importantly, the study utilizes
an appropriate model, which is described and validated in Chapters 2 and 3. This is
especially significant in that to date, efficacy of EAA receptor antagonists on spinal
reflex function has been evaluated in anesthetized intact or acute spinal animals,
without consideration of: 1) the substantial alterations that occur at the segmental level
over time as a result of spinal cord injury, and 2) the effects of anesthesia on spinal
reflex measurement. It is possible, given that EAAs activate GABA-mediated
presynaptic inhibition in normals, that an EAA receptor antagonist administered in the
chronic spinal animal could have no effect or even aggravate the spastic condition, a
result that would not be detected in either the intact or acute models.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2. QUANTIFICATION OF SPINAL REFLEXES UNDER ISOFLURANE ANESTHESIA
2.1 Introduction
Currently available surgical and pharmacological treatments are unsatisfactory for
many patients suffering from spasticity. Surgical treatments can produce new
complications (Abbott, 1992; Morota, Abbott, Kofler, Epstein & Cohen, 1995) and
antispastic agents typically produce undesired side effects such as muscle weakness
(Davidofif, 1985; Rice, 1987; Smith et al., 1994) which add to, rather than ameliorate,
the situation. Development of more selective and more efficacious treatments is
needed, and tests of novel agents must be conducted in the laboratory before clinical
trials. Currently used animal models, however, do not provide adequate paradigms for
the human spastic condition. Tests of potential compounds or pharmacotherapies for
spasticity are routinely conducted in animals that are either intact (Klockgether et al.,
1989; Schwarz et al., 1988a; b), acutely spinal (Chen, Bianchetti & Wiesendanger,
1986; Farkas & Ono, 1995; Tamawa, Farkas, Berzsenyi, Pataki & Andrasi, 1989), or
otherwise transiently debilitated (Simpson et al., 1995). Each of these preparations has
significant disadvantages.
Intact animals are recognizably void of any pathophysiology identified with spinal
cord injury or disease. The acute spinal situation is characterized by a physiological
condition that is quite different from both the intact and chronically injured spastic
animal (Cope et al., 1988; Munson et al., 1986; Nelson et al., 1979). Studies involving
acutely spinalized animals are often carried out shortly after transection, during a period
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 40
of time in which spinal shock and hyporeflexia are known to occur (Ashby et aL, 1974;
Calancie et al., 1993; Guttman, 1970; Leis, Kronenberg, Stetkafova, Paske & Stokic,
1996). By comparison, clinical symptoms of spasticity are typically gradual in onset.
Finally, there is one limiting factor common to most experimental protocols: the use of
anesthesia.
Anesthetic administration alone has been shown to influence a variety of
physiological systems (Ikonomidou-Turski, Schwarz, Turski, & Sontag, 1986; Kullmann
et al., 1989; Shimoji, Fujiwara, Fukuda, Denda, Takada & Maruyama, 1990). The use of
anesthesia during testing of compounds designed to have antispastic properties
introduces a possible confound in the interpretation of underlying pathophysiology and
pharmacological mechanisms. One model that does not involve anesthesia during
testing, a mutant strain of genetic spastic rats, has been described (Pittermann et al.,
1976). This model has been used to evaluate many potential therapeutic agents,
however, the only parameter of spasticity evaluated in these rats is spontaneous EMG
activity in the gastrocnemius muscles (Ikonomidou, Turski, Klockgether, Schwarz &
Sontag, 1985; Ikonomidou-Turski et al., 1986; Klockgether et al., 1989; Schwarz et al.,
1988a; b; 1992; Turski et al., 1985a; b; 1987a; b; 1990; 1992), which can be of relatively
limited functional significance to the clinical situation (Nielsen & Sinkjasr, 1996).
The purpose of the present study was to establish an appropriate experimental model
of spasticity, one that more closely approximates the clinical phenomenon. The goal was
to validate the chronic spinal rat as a stable, spastic animal for use in evaluation of
potential therapeutic agents aimed at reducing spasticity. To quantify changes following
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 41
spinalization and identify ensuing signs of spasticity, three parameters of the H-reflex
were obtained from intact and 28- and 60-day chronic spinal rats: reflex threshold,
response magnitude and rate-sensitive depression.
2.2 Methods
SUBJECTS
A total o f 17 male Sprague-Dawley rats (Holtzman Laboratories, Madison, WI)
weighing approximately 250-300 g at the beginning of experiments was used. Prior to
spinalization, animals were housed independently in suspended steel cages in a colony
room maintained on a 12 hr/12 hr light/dark cycle, with dark onset at 19.00 hr. Food
and water were available ad libitum. At the end of experimental procedures, rats were
euthanized by an anesthetic overdose. All procedures have been reviewed and approved
by the Institutional Animal Care and Use Committee of Louisiana State University,
Baton Rouge, LA.
SPINAL CORD TRANSECTION
For surgical procedures, animals were anesthetized using a mixture of 30% isoflurane
(AErrane, Anaqest, Madison, WI) and 70% oxygen at a flow rate of 4 L/ min. A dorsal
surface incision was made in the skin behind the ears and extended approximately two
inches caudally. Following retraction of the paraspinal muscles, a laminectomy was
performed between T6 and T9 to expose the spinal cord. A 1-2 mm segment of the cord
was removed by excavation. The excised tissue was then replaced by gelatin foam to
reduce bleeding, after which the incision was closed in layers.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 42
MAINTENANCE OF SPINAL ANIMALS
Following spinal transection, animals were placed in individual plastic cages in a room
in which the ambient temperature was raised to approximately 25 °C to maintain body
temperature. Beginning approximately 12 hr post-surgery, the rats’ urine was expressed
manually by applying pressure to their bladders. This was done twice daily until animals
recovered to the point of no longer requiring such care (approximately two weeks), at
which point ambient temperature was returned to 21°C.
ELECTROPHYSIOLOGICAL RECORDING METHODS
Isoflurane (30% isoflurane/70% oxygen; 4 L/min.) was used to lightly anesthetize
animals during recording sessions. Isoflurane was used as the anesthetic for two
reasons: 1) it has a rapid induction and recovery rate and 2) pilot testing revealed that it
minimally affected the H-reflex (although it completely abolished the polysynaptic flexor
reflex).
Animals were secured in a rodent torso sling and suspended in a sling frame (Harvard
Apparatus) to facilitate electrophysiological recordings. All electrophysiological
measurements were obtained using the Nicolet Viking IV (Nicolet Biomedical
Instruments, Madison, WI). Electrical signals were amplified, band-pass filtered (5 Hz-
10 kHz) and full-wave rectified. For all EMG recordings, a platinum subdermal ground
electrode was used.
MONOSYNAPTIC H-REFLEX
H-reflexes were evoked by electrical stimulation of the tibial nerve (single square
wave pulses; 0.2 ms duration) using subdermal needle electrodes (Astro-Med, Inc.,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 43
Grass Instruments Div.) subcutaneously inserted in the hip region, near the sciatic-tibial
branching point. EMG responses were recorded using a pair of stainless-steel barbed
broaches (Vereinigte Dentalwerke) inserted percutaneously into the plantar muscles of
the hindpaw. Barbed broaches connected to electrode lead wires were used to ensure
secure placement of recording electrodes. Plantar muscles were chosen for recording
because they are highly innervated by tibial motoneurons (Crockett, Hams & Eggar,
1987) and plantar H-reflexes have been found to be more stable than other muscle
groups (Meinck, 1976).
Tibial nerve stimulation bypasses the muscle spindle by activating la afferents and
directly activating a-motoneuron axons. The result is the H-wave. It increases in
amplitude with increasing stimulus intensity to a point, after which further increases in
stimulus intensity lead to the decline of the H-wave amplitude. The M-wave represents
direct motor axon activation and occurs earlier (approximately 3 ms after stimulation)
than the H-wave. An example of the composite wave is shown in Fig. 2.1. Three
parameters of the H-reflex were used in the present investigation: reflex threshold, reflex
response magnitude, and rate-sensitive depression, all of which have been shown to be
altered by spinal cord injury.
H-reflex Threshold
A stimulus-response recruitment curve was constructed by administering successive
stimuli of increasing intensity at a frequency o f 0.3 Hz. The recruitment curve displayed
the intensity range required to elicit minimum and maximum H-wave amplitudes in each
rat being observed and studied. The stimulus intensity for the actual threshold, or
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 44
M-wave
2 mV
2 ms
Fig. 2.1. Composite H/M wave elicited by electrical stimulation of the tibial nerve in an awake, unanesthetized rat 28 days after spinalization. The M-wave (left) is the result of direct motoneuron activation and appears first, approximately 3-4 ms after stimulation. The H-wave (right) results from synaptic activation of alpha motoneurons in the spinal cord by sensory afferents. It has a longer latency (approximately 6-9 ms) and a lower threshold than the M-wave. In this example, the compound EMG response is shown at the maximum M wave amplitude (M ^J.
the first visible H-wave, typically tracks in parallel with that for the maximum H-wave.
Because measurement of the first visible H-wave is difficult, reflex threshold was defined
at each test session as the minimum stimulus intensity (in mA) required to produce the
maximum H-wave amplitude. This intensity was subsequently used to elicit 20
consecutive responses at each of three frequencies: 0.3, 1.0 and 5.0 Hz, with a one-
minute inter-rate trial interval.
Reflex response magnitude. Twenty successive stimuli at a given stimulus rate were
administered at the threshold intensity determined for each subject. Peak-to-peak H-
wave amplitude was measured for every EMG response. Reflex response magnitude
was calculated as the mean of 20 measurements for each stimulus rate.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 45
Rate-sensitive depression. Frequencies higher than 0.3 Hz are known to have a
depressive effect on subsequent responses to stimulation, i.e., rate-sensitive or low-
frequency depression (Eccles & Rail, 1951; Lloyd, 1957; Lloyd & Wilson, 1957). Rate-
sensitive depression is considered an expression of presynaptic inhibition (Curtis &
Eccles, 1960; Kaizawa & Takahashi, 1970; Lloyd, 1957; Lloyd & Wilson, 1957).
Because past reports have shown that the maximum H-wave amplitude is elicited at 0.3
Hz, it was designated as the control rate. The H-wave amplitude, the basic measure of
the reflex response, was defined as the mean of 20 responses obtained at 0.3 Hz. Rate-
sensitive depression of the H-reflex was evaluated two ways: 1) Reflex response
magnitudes o f H-waves (in mV) elicited at the control frequency (0.3 Hz) were
compared with those of H-waves elicited at 1.0 and 5.0 Hz. 2) Rate-sensitive depression
was calculated by normalizing reflex response magnitudes at 1.0 and 5.0 Hz to their
corresponding control rate magnitudes. The standardized scores for 1.0 and 5.0 Hz
were expressed as percents of control, as shown by the formula below:
(1.0 or 5.0 Hz) mean response magnitude ------x 100 = % of Control 0.3 Hz mean response magnitude
PROCEDURES
The first 11 animals were divided into two groups. Both were tested in the intact
state and then again one month after transection, allowing animals to serve as their own
controls for the effect of spinalization. In the intact state, six rats received the control
frequency (0.3 Hz) and 1.0 Hz, and five received 0.3 and 5.0 Hz. At 28 days post
transection, every rat received all three stimulus rates. By receiving the control
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 46
frequency in addition to a test frequency. An additional group of six rats was tested 60
days after spinal transection and received all three stimulus rates: 0.3, 1.0 and 5.0 Hz.
Statistical Analysis
Results were analyzed using analysis o f variance (ANOVA) or where appropriate, the
Student’s i-test. A priori contrasts were used for within subject ANOVA comparisons.
The significance level was set at p < 0.05 for all analyses.
2.3 Results
REFLEX THRESHOLD
As shown in Fig 2.2, reflex thresholds for 28- and 60-day chronic spinal rats did not
differ from those for intact rats. For rats tested when intact and again 28 days after
transection, the respective means were 24.33 ± 3.12 mV and 25.25 ± 2.51 mV, 1(10) =
0.19, p > 0.05. The mean reflex threshold for rats tested 60 days after transection was
20.65 ± 3.17 mV, which did not differ from either intact, 1(15) = 0.76, p > 0.05 or 28-
day values, 1(15)= 1.11, p > 0.05.
Reflex Response Magnitude
Changes in H-reflex magnitude following transection are shown for the three stimulus
rates in Fig. 2.3. H-reflex responses elicited at 0.3,1.0 and 5.0 Hz were all significantly
higher in amplitude after 60 days o f spinalization compared to 28 days, E (1, 15) =
12.3 l,p < 0.01; E(l, 15) = 5.00, p < 0.05; E(l, 15)= 12.54, p < 0.01, respectively.
Rate-sensitive depression. Comparison of H-reflex amplitudes elicited at 0.3, 1.0
and 5.0 Hz at each time point demonstrated a depressive effect of rate on H-wave
amplitude. As shown in Fig. 2.4, the 5.0 Hz stimulus rate elicited responses that were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 47
40 < e 35 30 os S 25 4> 4* 20 Me 09 15 3 “3 10 E 5 55 0 Intact 28 Days 60 Days
♦ Intact -28 Days - o- 28-60Days
Fig. 2.2. Reflex threshold in isoflurane-anesthetized animals tested when intact and 28 and 60 days after spinalization. There was no change in threshold across the three time points. Data are plotted as mean ± SEM.
5 £, 4) 4 -o s 3 ' e 04 cb S 2 5 1 £ 33 0 0 3 Hz 1.0 Hz 5.0 Hz -A— Intact □ - 28 Days 4 - 60 Days
Fig. 23. Summary of H-wave amplitudes obtained under isoflurane from intact, 28- and 60-day spinal rats showing the mean ± SEM for each of the three stimulus rates. In the intact condition, all rats (a = 11) received the control frequency (0.3 Hz) followed by either 1.0 Hz (q = 6) or 5.0 Hz (q = 5). All rates (0.3,1.0 and 5.0 Hz) were administered to the same animals (n = 11) at 28 days and also to a separate group of rats (n = 6) 60 days after transection. * Significant difference between 28 and 60 days.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 48
significantly smaller in amplitude than those obtained at the control rate, 0.3 Hz, in the
intact, l (4) = 5.09, p < 0.01,28-day spinal, F (1,10) = 67.80, p < 0.01 and 60-day spinal
conditions, F (1,5) = 7.67, p < 0.05.
Isoflurane: Group 1 B Isoflurane: Group 2 > we 4 ■ x X “O s 3 ■ e es0* S 2 • 4> % 1 • * X 0 • Intact 28 Days 60 Days Intact 28 Days 60 Days I 1 0.3 Hz EZ83 1.0 Hz O H 5.0 Hz
Fig. 2.4. Rate-sensitivity of the H-reflex is preserved in isoflurane-anesthetized intact and spinal rats. H-reflex responses were significantly reduced at the highest rate of stimulation (5.0 Hz) relative to the control rate (0.3 Hz) amplitudes at each time point. Results are plotted separately for intact and 28-day spinal animals that received 0.3 and 1.0 Hz (A, a = 6) and 0.3 and 5.0 Hz ( B., n = 5). Bars represent mean ± SEM. * Significantly lower than the 0.3 Hz response at the same time point.
Mean H-reflex amplitudes at 1.0 and 5.0 Hz normalized to their respective 0.3 Hz
response means are displayed at each time point as percents of control in Fig. 2.5. The
amount of rate-sensitive depression at 5.0 Hz was significantly greater than at 1.0 Hz, as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 49
100 i 4> « 0 6 L.o 60 • os U 40 ■ «•»o £ 20 -
Intact 28 Days 60 Days 1= 1.0 Hz ■ a 5.0 Hz
Fig. 2.5. Effect of chronic spinalization on H-reflex rate-sensitive depression. H-reflex amplitudes at 1.0 and 5.0 Hz were converted to the mean (± SEM) percent of their respective mean 0.3 Hz control rate amplitude at each time point. A decrease in rate- sensitive depression is indicated by an increase in the percent of control. *1*Rate- sensitive depression at 5.0 Hz is significantly greater than at 1.0 Hz in intact and 28-day spinals. However, at 60 days there was no difference between the two rates, indicating a reduction in rate-sensitive depression at 5.0 Hz that is associated with spasticity. tSignificant increase in percent of control (decrease in rate-sensitive depression) at 60 compared to 28 days.
indicated by a smaller percent of the control rate in the intact, t (9) = 2.75, p < 0.05 and
28-day spinal, t (10) = 5.60, p < 0.01 conditions. However, compared to 28 days, at
60 days of spinalization, H-reflex responses elicited at 5.0 Hz were significantly greater,
t (5) = 2.68, p < 0.05, and percent of control responses at 5.0 Hz were no longer
different from those at 1.0 Hz, 1 (5) = 2.17, p > 0.05.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50
2.4 Discussion
SUMMARY OF FINDINGS
The objective of this study was to verify the development of spasticity in the chronic
spinal rat by quantifying three parameters of the monosynaptic H-reflex: 1) threshold; 2)
response magnitude; and 3) rate-sensitive depression. H-reflexes were recorded under
isoflurane in intact and chronically spinal animals 28 and 60 days after transection. The
results showed no change in reflex threshold after spinal transection, whereas
development of hyperreflexia and reduced rate-sensitive depression were both evident at
60 days post-transection.
Refex Threshold
Reflex thresholds for rats tested under isoflurane anesthesia in this study were the
same before and after transection, despite the established association between reduced
reflex threshold and spasticity in chronic patients (Levin & Hui-Chan, 1993; Powers et
al., 1988). Clinically, H-reflexes are difficult and sometimes impossible to elicit during
acute periods of spinal trauma (Leis et al., 1996) and reduced thresholds are thought to
accompany the gradual development of chronic spasticity (Levin & Hui-Chan, 1993).
Consistent with clinical reports, it was reported that under ketamine, thresholds for
spinally-contused rats did not change acutely, however a significant reduction was
evident within a two-month period (Thompson et al., 1992). In another study, it was
reported that under halothane, thresholds for acutely (hours) obex-transected rats were
not different from intact levels (Gozariu, Roth, Keime, LeBars & Wilier, 1998).
However, when compared with unanesthetized acute spinals, reflex thresholds were
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 51
found to be significantly lower without halothane. This finding would indicate that obex-
transection acutely reduced H-reflex thresholds, and that this reduction was masked by
halothane, a volatile inhalation anesthetic similar to isoflurane (Marshall & Longnecker,
1996). The apparent lack of any change in reflex thresholds from pre- to 28 and 60 days
post-transection in the present study may be due to a similar masking effect of isoflurane.
Hyperreflexia
Significant increases in H-reflex amplitudes recorded under isoflurane were evident by
60 days post-transection, yet by casual observation, indications of spastic behavior were
apparent prior to two months. Delayed hyperreflexia is consistent with the development
o f clinical spasticity (Delwaide, 1973; reviewed in Katz & Rymer, 1989), and enhanced
H-reflex amplitudes have been reported under ketamine in contused (Thompson et al.,
1992) and transected (Skinner et al., 1996) rats within 28 days. Nonetheless, H-reflexes
were not significantly enhanced within one month in the present study. There are two
possible reasons for this: First, hyperreflexia may have been observed at 28 days if all
three stimulus rates had been administered when the animals were intact. That is,
amplitudes obtained in the intact condition might have been smaller than those at 28 days
if the subjects had received all three rather than two of the three stimulus rates.
However, this may not represent a very powerful determinant of overall reflex amplitude.
As indicated in the Methods section, in the intact condition, one group of animals
received 0.3 and 5.0 Hz and the other group received 0.3 and 1.0 Hz, the only difference
at 28 days being the addition of 1.0 or 5.0 Hz, respectively. Any additional depression
exerted on reflex amplitude at 28 days as a result of a third stimulus rate trial would only
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 52
account for one-third of the overall mean amplitude used to measure the difference
between intact and 28 days. Moreover, the reflex amplitudes at 0.3, 1.0 and 5.0 Hz
showed no significant increase from pre- to 28 days post-transection, which could
indicate a lack of effect of a third stimulus on amplitude. Second, and more likely,
hyperreflexia might have been evident by 28 days if a different anesthetic was used, such
as ketamine (Skinner et al., 1996; Thompson et al., 1992).
The ‘delay’ in the manifestation of hyperreflexia as measured under isoflurane
anesthesia can perhaps be explained by its effects on segmental spinal cord potentials.
Isoflurane reportedly acts to suppress heterosegmentally-mediated slow component
potentials in the spinal cord (Shimoji et al., 1990). Relevant to the rate-sensitive
attribute of the H-reflex investigated in the present study, the slow positive wave
inhibited by isoflurane is thought to be involved in the mechanism of supraspinally-
mediated primary afferent depolarization. In normal animals, the spinal monosynaptic
reflex is subject to regulation by mechanisms of presynaptic inhibition such as primary
afferent depolarization (PAD), which is generally considered to be reduced as a result of
traumatic spinal injury. Spinal transection causes interruption of descending inhibitory
fibers, and removal of supraspinal regulation of reflex activity is recognized as a key
factor in the development of spasticity (Ashby et al., 1974; Clemente, 1978; Delwaide,
1973). Therefore, if isoflurane acts to reduce a mechanism such as PAD and hence
reduce presynaptic inhibition, one expected result, at least in the intact animal, would be
increased H-reflex magnitude. Conceivably, by reducing presynaptic inhibition,
isoflurane could cause H-reflex responses in intact rats to resemble those in chronic
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 53
spinal rats. It cannot be said for sure that this action of isoflurane would have the same
effect of increasing H-reflex amplitudes in spinal animals because, as noted by the
authors (Shimoji et al., 1990), PAD is a supraspinally-mediated reflex modulation (which
is eliminated with spinal transection). In other words, similar functional effects of
isoflurane and transection on segmental circuitry of the spinal cord might explain the lack
of distinction between intacts and 28 day spinals. Nonetheless, in the present study, a
significant difference in H-reflex amplitudes was found between 28 and 60 days post
transection and not between intact and 28 days. The difference between 28 and 60 days
might be due to a modification of isoflurane’s effects at 60 days as a result of progressive
pathological changes in the cells and circuits with which the anesthesia interacts.
Rate-Sensitive Depression
A significant decrease in rate-sensitive depression showed up statistically at 60 but
not 28 days post-transection. An immediate assumption would be that isoflurane
anesthesia could account for this difference as well, given the change in rate-sensitive
depression found within one month under ketamine anesthesia (Skinner et al., 1996;
Thompson et al., 1992). Evidence indicates that actions of general anesthetics such as
isoflurane are GABA a receptor-mediated (Kullmann, Martin & Redman, 1989; Marshall
& Longnecker, 1996). By contrast, GABAg receptor action is indicated in the
mechanism of presynaptic inhibition associated with rate-sensitive depression in normal
(Curtis, 1998; Lev-Tov et al., 1988; Peshori et al., 1998) and spinally-injured subjects
(Thompson et al., 1996). Thus, it seems unlikely that isoflurane affected the rate-
dependent reflex actions revealed in this study.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 54
Structural Changes Underlying Spasticity
The fact that differences were evident after 60 days of spinalization despite whatever
factor prevented this from being observed after 28 days may be explained by the
progressive nature of the development of spasticity, i.e. increased motoneuronal
excitability and enhanced evoked-synaptic excitation of motoneurons, both of which are
believed to underlie the spastic behavior (Katz & Rymer, 1989). Receptor
supersensitivity, a response to removal of synaptic input, develops in coordination with
spasticity, as does collateral sprouting, which contributes to enhancement o f synaptic
excitation (Murray & Goldberger, 1974). Receptors in chemically destroyed
neurotransmitter systems become almost one-hundred percent degenerated within a
couple weeks and significantly recover in density within 3 to 6 months (Nygren, Fuxe,
Jonsson & Olson, 1974), presumably via collateral sprouts, beginnings o f which have
been observed 2 to 3 weeks post-injury (McCouch et al., 1958; Steward, Cotman &
Lynch, 1974; West, Deadwyler, Cotman & Lynch, 1975). Synaptogenesis has been
shown to persist for several months (Bernstein, Wells & Bernstein, 1978; Raisman,
1977), and an increase in H-reflex amplitude was found to occur as late as 3 to 6 months
post-injury in humans (Little & Halar, 1985). Thus, it is suggested that in the present
study, thoracic spinal transection resulted in progressive cellular and physiological
changes which continued for at least 60 days. These changes were evidenced by the
development of exaggerrated reflex responses and reduced rate-modulation of the H-
reflex.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3. EFFECT OF KETAMINE ANESTHESIA ON SPINAL REFLEX MEASUREMENT
3.1 Introduction
Results of experiments described in Chapter 2 suggested a confounding effect of
isoflurane anesthesia on reflex measurement. Briefly, H-reflexes were recorded under
isoflurane in intact and chronic spinal rats 28 and 60 days after transection. A general
trend toward hyperreflexia developed following spinalization that did not reach
significance within the first month, as would have been predicted (Skinner et al., 1996;
Thompson et al., 1992). Nonetheless by 60 days, spinal rats displayed increased H-
reflex amplitudes and decreased rate-sensitive depression relative to intact values. The
expected decrease in reflex threshold as a result of spinal cord injury (Levin & Hui-
Chan, 1993; Powers, Marder-Meyer & Rymer, 1988; Taylor, Ashby & Verrier, 1984),
however, was not evident at either time point.
The purpose of the present study was, as for the previous study, to validate the
chronic spinal rat as an experimental model of spasticity. To avoid the complications of
measurement that were encountered with isoflurane in Chapter 2, a different compound
was employed. Ketamine (Ketaset, Parke-Davis) was selected for use in the first of
three experiments in this study because it reportedly does not significantly interfere with
the monosynaptic path circuitry in the spinal cord (Lodge & Anis, 1984; Tang &
Schroeder, 1973). Additionally, its use facilitates comparison with results of previous
studies that utilized ketamine and reported evidence of hyperreflexia within the first
month after spinal cord injury (e.g., Skinner et al., 1996; Thompson et al., 1992).
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 56
Changes from pre- to post-transection were investigated in Experiments 1 and 2
using three parameters of the H-reflex: threshold, magnitude and rate-sensitive
depression. H-reflexes were elicited by tibial nerve stimulation at three different
frequencies: 0.3, 1.0 and 5.0 Hz and recorded from the plantar muscles of the ipsilateral
hindpaw. In Experiment 1, recordings were obtained from intact and 28- and 60-day
spinal rats under ketamine anesthesia. In the second experiment, the same recordings
were made without anesthesia in two additional groups of spinal rats: one group was
tested at 28 and the other at 60 days post-transection. In the third experiment,
monosynaptic H- and polysynaptic flexor reflexes were elicited in unanesthetized 28-day
spinal rats before and 15, 30 and 60 minutes after vehicle injection. This was done to
determine the within-session measurement stability of the model. Recruitment curves
were obtained to allow calculations of H/M ratios. Flexor reflexes were elicited by
electrical stimulation (500 Hz) of the hindpaw and recorded from the ipsilateral biceps
femoris/semitendinosous. The present investigation substantiates the chronic spinal rat
as a model for evaluating potential antispastic agents for clinical use in spinal cord injury-
induced spasticity.
3.2 Methods
SUBJECTS
A total of 21 male Sprague-Dawley rats (Holtzman Laboratories, Madison, WI)
weighing 250-300 g at the beginning of experiments was used. Prior to surgery, the
animals were housed independently in suspended steel cages in a colony room
maintained on a 12 hr/12 hr light/dark cycle, with dark onset at 19.00 hr. The ambient
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 57
temperature was kept at 21 °C, and food and water were available ad libitum. At the
end of experimental procedures, rats were euthanized by an anesthetic overdose. All
procedures were reviewed and approved by the Institutional Animal Care and Use
Committee of Louisiana State University, Baton Rouge, LA.
SPINAL CORD TRANSECTION
Animals were anesthetized during surgery using a mixture of 30% isoflurane and
70% oxygen (AErrane, Anaquest, Madison, WI) at a flow rate of 4 L/min. An incision
was made in the skin approximately one inch behind the ears and extended caudally.
Following retraction of the paraspinal muscles, a laminectomy was performed between
T6 and T9. A 1-2 mm segment of the spinal cord was removed by excavation. The
excised tissue was then replaced by gelatin foam to reduce bleeding after which the
incision was closed in layers.
MAINTENANCE OF SPINAL ANIMALS
Following spinal transection, animals were placed in plastic cages in a room in which
the ambient temperature was raised to approximately 25 °C to maintain body
temperature. Beginning approximately 12 hr post-surgery, the rats’ urine was expressed
manually by applying pressure to their bladders. This was done twice daily until animals
recovered to the point of no longer requiring such care (approximately two weeks), at
which point ambient temperature was returned to 21 °C.
ELECTROPHYSIOLOGICAL RECORDING METHODS
To facilitate electrophysiological recordings, animals were secured in a rodent torso
sling and suspended in a sling frame (Harvard Apparatus) such that all four limbs hung
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 58
freely. Ail electrophysiological measurements were obtained using the Nicolet Viking IV
(Nicolet Biomedical Instruments, Madison, WI). Electrical signals were amplified, band
pass filtered (5 Hz-10 kHz) and full-wave rectified. For all EMG recordings, a platinum
subdermal ground electrode was used.
EXPERIMENT 1
The development o f spasticity was followed in a single group of rats (n = 9) over
the course of approximately two months. Recordings were made when the rats were
intact and againat 28 and 60 days following spinal cord transection. Animals were
divided into three groups by order of stimulus rate administration. Group 1 (n = 3)
received the control frequency (0.3), 1.0, 5.0 Hz; Group 2 (q = 3) received 1.0, 5.0,0.3
Hz; and Group 3 (n = 3) received 5.0, 0.3, 1.0 Hz. By receiving the control frequency in
addition to the two higher stimulus rates, each rat served as its own control for
measurement o f rate-sensitive depression.
Monosynaptic H-Reflex
H-reflexes were elicited by stimulation of the tibial nerve and recorded from the
plantar muscles o f the ipsilateral hindpaw. As described by Meinck (1976), this
arrangement of stimulating and recording electrodes yields the best H/M wave responses,
as the plantar muscles are heavily innervated by tibial motoneurons (Crockett et al.,
1987). The tibial nerve was stimulated (single square-wave pulses; 0.2 ms duration)
using a pair of platinum subdermal needle electrodes (Astro-Med, Inc., Grass
Instruments Div.) which were inserted subcutaneously. EMG recordings were made
using a pair of stainless-steel barbed broaches (Vereinigte Dentalwerke) inserted
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 59
percutaneously into the plantar muscles, which ensured secure placement of recording
electrodes.
H-reflex threshold and response magnitude. In Experiments 1 and 2, a
recruitment curve was obtained prior to making the actual recordings. H-reflex
response magnitude was calculated for each stimulus rate as the mean amplitude of 20
responses. The magnitudes of the evoked EMG potentials were determined by measuring
the peak-to-peak H-wave amplitudes (in mV). The control rate was set at 0.3 Hz based
on previous reports showing that this frequency produces no significant effect on
subsequent response amplitudes (Eccles & Rail, 1951; Meinck, 1976). H-wave
amplitude, the basic measure of the reflex response, was thus defined as the mean of 20
responses obtained at the control stimulus rate of 0.3 Hz.
Rate-sensitive depression. Rate-sensitive depression o f the monosynaptic H-reflex
was evaluated as follows: Mean H-reflex amplitudes elicited by stimulation at 1.0 and 5.0
Hz were calculated as percentages of the mean H-reflex amplitude elicited at the control
rate (0.3 Hz). To provide percent of control measures, mean H-reflex amplitudes
obtained at 1.0 and 5.0 Hz were each divided by the mean control rate amplitude and
multiplied by 100 as shown in the following formula:
(1.0 or 5.0 Hz) mean response magnitude — x 100 = % of Control 0.3 Hz mean response magnitude
EXPERIMENT 2
To demonstrate the ability to quantify spasticity in unanesthetized animal subjects,
two groups of spinalized rats were tested without anesthesia using the same methods as
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 60
in Experiment 1. The use of anesthesia was unnecessary in Experiments 2 and 3, which
involved obtaining reflex measurements exclusively in spinal rats, because
electrophysiological procedures did not produce any overt discomfort in the chronic
spinal animals.. One group (n = 6) was tested at 28 days and the other (n = 6) at 60 days
post-transection. In this experiment, H-reflex thresholds, mean response magnitudes and
rate-sensitive depression values were obtained to allow group comparisons on these
measures with their anesthetized counterparts in Experiment 1.
EXPERIMENT 3
To further characterize the measurement capabilities of this model, repeated
recordings of two additional indices of spasticity were made in a fourth group of spinal
rats: H/M ratios and flexor reflexes. The H/M ratio (maximum H-wave amplitude:
maximum M-wave amplitude; an index of motoneuron pool excitability
(Magladery, Porter, Park & Teasdall, 1951; Taborikova & Sax, 1968), is a classic
measure involving elicitation of the H-reflex that has been used clinically to evaluate
spasticity for years (Angel & Hofiman, 1963; Calancie et al., 1993; Little & Halar, 1985;
Matthews, 1966). Recruitment curves, obtained by the same procedure for recording H-
reflexes as in the first two experiments, were used to calculate H/M ratios.
Polysynaptic Flexor Reflex
The polysynaptic flexor reflex is also used clinically as a measure of spasticity (e.g.,
Parise, Garcia-Larrea, Mertens, Sindou & Mauguiere, 1997). Here, flexor reflexes were
evoked by electrical stimulation of the hindpaw and recorded from the flexor muscles of
the ipsilateral hindlimb. A pair of subdermal platinum needle electrodes inserted into the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 61
interphalangeal area of the hindpaw was used for stimulation and a pair of stainless-steel
barbed-broaches inserted percutaneously into the biceps femoris/semitendinosus
muscles was used for recording.
/ K AUC = 6025
AUC = 5.567
l\jAvk AUC = 5.854
AUC = 4.629
0.5 mV AUC = 5.548
10 ms
Fig. 3.1. Rectified EMG recording of the flexor reflex from the biceps femoris/semitendinosus in an unanesthetized 28-day chronic spinal rat. Flexor reflex magnitude was quantified by measuring AUC for responses occurring up to 100 ms after stimulation (500 Hz; 0.2 ms) of the ipsilateral hindpaw.
Stimulation (five square-wave pulses, 500 Hz, 0.2 ms duration) was then delivered five
consecutive times (5 s interstimulus intervals) at 2.0-3.0 times reflex threshold
to elicit five reflexes. (See Fig. 3.1 for an example flexor EMG recording.) Flexor
reflex response magnitudes were generated by a computer software program (Nicolet
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 62
Viking IV, Nicolet Biomedical Instruments) which quantified each response by
measuring the area bounded by the averaged response and the baseline, i.e., the area
under the curve (AUC). The mean AUC value for the five consecutive responses was
subsequently calculated and served as the data point for each measurement period.
Statistical Analysis
Data were analyzed using the analysis of variance (ANOVA) or where appropriate,
Student’s t-tests. For ANOVAs, post hoc multiple comparisons were carried out using
the Student Newman-Keuls method. Statistical significance level for all analyses was
set at p s 0.05.
3.3 Results
EXPERIMENT 1
In the initial analysis of data from all three frequency order groups (n = 9), two-way
repeated measures ANOVA showed a non-significant main effect of order (i.e., order of
stimulus rate delivered) and all pairwise comparisons showed significant differences
between response magnitudes for stimulus rates of 0.3,1.0 and 5.0 Hz (p < 0.05 for all).
However, the ANOVA also revealed an interaction effect between order and frequency
(p < 0.001) which was determined by post-hoc comparisons to be between mean
response magnitudes obtained at 0.3 Hz in Group 1 and Group 2. As illustrated in Fig.
3.2, the interaction effect of order indicated that the response magnitude at 0.3 Hz was
significantly lower for Group 2 than Group 1 (mean difference = 0.78± 0.02 mV; p <
0.05). This could indicate an inhibitory effect on amplitudes generated at 0.3 Hz due to
the preceding 5.0 Hz trial or a combined effect of the 1.0 and 5.0 Hz trials. There
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63 ^ 3.0 ■ —O- Group I - Group 2 —• — Group 3
Stimulus Rate (Hz)
Fig. 3.2. Effect of stimulus rate order on H-reflex amplitudes. Stimulation was delivered (10 repetitions per rate; 1 minute inter-trial interval) to three groups o f intact rats (n = 3/ group). Groups received all three stimulus rates in the following orders: Group 1: 0.3, 1.0, 5.0 Hz; Group 2:1.0,5.0,0.3 Hz; Group 3: 5.0,0.3, 1.0 Hz. *At 0.3 Hz, H-reflex amplitudes for Group 2 were significantly lower than Group 1. Data shown are mean ± SEM.
were no other between group differences at 0.3 Hz, nor were there any at 1.0 or at 5.0
Hz. Additionally, no interaction effects were revealed by analyses of data recorded at
the 28- or 60-day post-transection measurement points. Analyses were run with and
without the data from Group 2, and there were no differences in the statistical findings at
any time? point. However, because there was evidence of an order effect, data from
animalsin Group 2 was not included in the following analyses and figures.
H-Reflex Threshold
As illustrated in Fig. 3.3, chronic spinal rats displayed a marked increase in
sensitivity to the electrical stimulus. There was a significant and intense decrease in
intensity of electrical stimulation that was required under ketamine to elicit maximum
H-reflex amplitudes in rats after they had been spinalized 28, E (1,5) = 35.93, g < 0.01
and 60 days, E (1,5) = 48.47, g < 0.01 as compared to stimulus intensities used in
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 64
< 4 0 • — Ketamine - No Anesth. ws 3 5 ■ £ * 3 0 • QO B 25 • V C 2 0 • oo 15 ■ 10 • s 5 ■ cn 0 • Intact 28 Days 60 Days
Fig. 33. Change in H-reflex threshold as a result of spinalization. By 28 days post transection there was a significant decline in the mean threshold for ketamine- anesthetized rats, with no further change at 60 days. Reflex thresholds for unanesthetized chronic spinal rats did not differ from those obtained under ketamine at either time point. * Significantly different from the intact state. Data shown are mean ± SEM.
the intact condition. The mean reflex threshold for intact rats was 33.26 ± 5.25 mA, yet
when chronically spinal, the same rats required approximately one-third of the pre
transection intensity to achieve the same level of reflex excitation (10.52 ± 1.94 mA at
28 days; 10.54 ± 1.87 mA at 60 days). There was no significant threshold difference
between 28 and 60 days post-transection (p > 0.05).
Response magnitude. As shown in Fig. 3.4.A, exaggerated H-reflex amplitudes
developed in rats following complete cord transection. Combined amplitudes of H-
waves evoked at 0.3,1.0 and 5.0 Hz showed a significant overall increase with time
post-transection, F (2, 34) = 12.22, p < 0.001. After 28 days, H-reflex responses
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 65 Ketamine
8
< 1 .0
Intact 28 Days 60 Days
No Anesthesia
1 28 Days 64 Days
Fig. 3.4. Rats developed permanent hyperreflexia following spinal transection. Bars represent the mean ± SEM of H-reflex magnitudes collapsed across all three stimulus rates at each time point. A. H-reflex response magnitudes increased within 28 days of spinalization in ketamine-anesthetized rats, with no further change at 60 days. ’“Significantly different from intact values. B. H-reflexes measured at the same time points in unanesthetized spinals had similar overall magnitudes that did not differ from 28 to 60 days.
were significantly greater than in the intact condition, E (1,17) = 19.98, p < 0.001.
Overall responses remained elevated, with no further increase in magnitude from 28 to
60 days, p > 0.05.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 66
Intact 3 0.3 Hz E - - 1.0 Hz 2 5.0 Bz S (x = 1.10 mV) I * = 0.50 mV) < 0.28 mV) « ® M £ s -1 0123456789 10
B 28 Days > 3 E, v 2 s
E 1 < Si os 0 £ ail 1 0 123456789 10 60 Days 3 E, U 2 s "a 1 E < os> 0
1 0123456789 10 Time (ms)
Fig. 3.5. Summary of H-wave amplitudes obtained from intact (A, n = 6), 28-day (B, q = 6) and 60-day (C, n = 5) spinal rats illustrating alterations that occur as a result of spinal cord transection. Group means from ketamine-anesthetized animals were used to generate graphic representations of average H-wave amplitudes for the three stimulus rates (0.3, 1.0 and 5.0 Hz) at each time point
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 67 Ketamine
Intact 28 Days 60 Days B No Anesthesia > S 3.5 3.0 2^ 2.0 E < 1.5 4» 1.0 03 Hz £ 0.5 m n i .o Hz ■ H 5.0 Hz X 0.0 28 Days 60 Days
Fig. 3.6. Rate-sensitivity of the H-reflex in intact and spinal rats under ketamine anesthesia (A) and in spinal rats without anesthesia (B). A : In the intact condition, 1.0 and 5.0 Hz significantly depressed responses relative to the 0.3 Hz control rate; after spinal transection only the 5.0 Hz stimulus produced a significant decrease.B : Rate sensitivity was also evident in unanesthetized spinal rats. * Significantly lower amplitude than elicited at 0.3 Hz; ^ Significantly lower amplitude than elicited at 1.0 Hz. Data shown are mean ± SEM.
Alterations in spinal reflex regulation that occur as a result of a traumatic injury can
be reflected by changes in H-wave amplitudes from the intact state, as illustrated in Fig.
3.5. Amplitude increases at all three stimulus rates following spinalization contributed
to the general enhancement of H-reflex response measurements. Compared to
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 68
recordings made in the rats before they were spinalized, magnitudes of H-waves elicited
at 0.3 Hz, 1.0 Hz and 5.0 Hz increased significantly 28 days, F (1, 5) = 9.97, p < 0.05
and 60 days, F (1,5) = 24.41, p < 0.01 following transection surgery. From the first to
the second month of chronic spinalization, H-wave magnitudes showed further
increases, E (1, 5) = 26.57, p < 0.01, which can most likely attributed to a reduced
inhibitory effect of the 1.0 Hz stimulus on reflex amplitudes. When analyzed
separately, the three stimulus rates did not individually show statistically significant
increases across time, 0.3 Hz, F (2, 8) = 3.76; 1.0 Hz, E (2, 8) = 5.48; and 5.0 Hz, F (2,
8) = 2.85; p > 0.05 for all.
Rate-sensitive depression. As indicated by the results summary in Fig. 3.5, H-
reflexes in both the intact and spinalized conditions exhibited significant rate-sensitive
depression, i.e., a decrease in amplitude with increasing stimulus frequency. This effect
of rate on H-reflex responses was statistically supported by ANOVAs performed at each
time point: intact,E (2,2) = 57.31, p < 0.001; 28 days spinal, F (2,2) = 8.21, p = 0.01;
and 60 days spinal, E (2,2) = 8.97, p < 0.05. However, as illustrated in Fig. 3.6.A, this
phenomenon differed between intact and spinalized rats. In the intact condition,
stimulation at 1.0 and 5.0 Hz significantly depressed H-reflex responses relative to the
0.3 Hz control rate, E (1, 5) = 32.44, p < 0.01, with 5.0 Hz exerting significantly greater
response suppression than 1.0 Hz,E (1, 5) = 62.23, p = 0.001. However, when
supraspinal input was eliminated, responses to 1.0 Hz showed significant decreases in
rate suppression, F (1, 5) = 4.35 andE (1, 5) = 3.97 at 28 and 60 days, respectively, p <
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 69 A Ketamine **
100 •
■ 3 80- u § 60 ■ U 40 • £ 20 •
Intact 28 Days 60 Days B No Anesthesia
100 ■
80 -
o 60 • o 40 •
20 •
28 Days 60 Days
Fig. 3.7. Rate-sensitive depression is reduced in chronic spinal rats. H-reflex amplitudes at 1.0 and 5.0 Hz were converted to the mean (± SEM) percent of their respective 0.3 Hz control rate amplitude at each time point. As with amplitude, a decrease in rate-sensitive depression is indicated by an increase in the percent of control. A. Under ketamine, rate-sensitive depression to the 1.0 Hz stimulus was significantly reduced at 28 days post-transection, with a further reduction at 60 days. B. In the absence of ketamine, a significant decrease in rate-sensitive depression from 28 to 60 days post-transection occurred at both 1.0 and 5.0 Hz. * Significant increase in percent of control relative to intact {A) and ** to 28 days post-transection (A., B.) Percent of control response at 5.0 Hz was significantly lower than at 1.0 Hz for all measurement points in both anesthetized and unanesthetized spinal rats.
0.05 for both. Response suppression at 5.0 Hz was preserved throughout the two month
period of chronic spinalization.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 70
Changes in the amount of rate-sensitive depression can also be effectively detected
by normalizing the responses at higher rates o f stimulation to those at the control rate: A
reduction in rate-sensitive depression would thus be indicated by an increase in the
percent of control value. By calculating the percent o f control values for 1.0 and 5.0 Hz,
as shown in Fig. 3.7.A, a significant reduction in rate-sensitivity following transection
was revealed in response to the 1.0 Hz stimulus at 28 days, E (1, 5) = 12.49, p < 0.05
and at 60 days, E (1, 5) = 260.97, p < 0.01. From 28 to 60 days, rate-sensitivity to 1.0
Hz was further reduced, E (1, 5) = 11.71, p < 0.05. Rate-sensitivity at 5.0 Hz, however,
showed no effect of spinalization, E (1, 5) = 1.96, p > 0.05.
EXPERIMENT 2
H-Reflex Threshold
Figure 3.3 shows that when no anesthesia was used, reflex threshold was the same at
28 and 60 days of spinalization, £ (10) = 1.67, p < 0.05. Also illustrated in Fig. 3.3,
ketamine anesthesia did not significantly influence H-reflex threshold levels in animals at
28 days, £ (10) = 0.23, p > 0.05 or at 60 days, £ (10) = 1.61, p > 0.05.
Response Magnitude
H-wave amplitudes collapsed across all three frequencies are summarized for
unanesthetized spinal rats in Fig. 3.4.B. Although different groups of spinal rats were
used as subjects for the two post-transection time points, the overall H-reflex magnitudes
for rats tested 28 or 60 days following spinalization did not differ significantly at either
time interval, £ (34) = 0.92, p > 0.05. Additionally, the overall response measures of H-
reflexes obtained from spinal rats tested under ketamine (Fig. 3.4.A) did not differ
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 71
significantly from those recorded in unanesthetized spinal rats at 28 days, t (34) = 0.24,
p > 0.05 or at 60 days, t (34) = 1.19, p > 0.05.
Rate-sensitive depression. Similar to results illustrated in Figs. 3.6.A and 3.7.A.
for anesthetized animals, Figs. 3.6.B and 3.7.B depict the H-reflex rate-sensitivity
exhibited by unanesthetized spinal rats. Comparable to the results with ketamine, a
significant effect of stimulus rate on response amplitude was found in rats tested
without anesthesia at both 28 days, E (2, 10) = 27.43, p < 0.01 and 60 days, F (2,10) =
12.89, p < 0.01, Fig. 3.6.B. As in Experiment 1, response magnitudes for
unanesthetized rats were normalized to control rate values. As shown in Fig. 3.7.B, H-
reflexes in rats tested at 60 days post-transection exhibited significantly less rate-
sensitive depression than those tested at 28 days. The mean H-reflex response to a 1.0
Hz stimulus in 28-day spinal rats was 55.47 ± 8.43% of control and the mean response
to 5.0 Hz was only 15.20 ± 3.48%. When rats were tested after 60 days of spinalization,
the 1.0 Hz stimulus rate generated a mean H-reflex response that was 90.00 ± 5.72%
and 5.0 Hz produced a mean response that was 40.49± 8.55% of the control rate. These
responses to both 1.0 and 5.0 Hz at 60 days were significantly greater than those
obtained at 28 days, 1.0 Hz, I (10) = 3.39, p < 0.01 and 5.0 Hz, 1 (10) = 2.74, p < 0.05.
Comparative analysis of response amplitudes recorded from the two experimental
conditions (anesthetized and unanesthetized animals, shown in Fig. 3.6.A and 3.6.B,
respectively) at each of the three stimulus rates did not reveal any statistical effects of
ketamine on H-reflex amplitudes at either 28 or 60 days post-transection (Figs. 3.8.A
and B, respectively). However, as illustrated in Fig. 3.9, when normalized responses
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 72 28 Days 4.0 > E, 3.5 •o«» 3.0 s IS "q . E 2.0 < L5 > 0 8 1.0 £ 0.5 SB 0.0 1.0 5.0 B 60 Days 4.0 > E, 3.5 u 3.0 •a s 2^ "5. 2.0 S 1.5 41 « 1.0 £ 0.5 Ketamine SB No Anesthesia 0.0 5.0 Stimulus Rate (Hz)
Fig. 3.8. H-reflex amplitudes were compared between ketamine-anesthetized and unanesthetized rats at (A) 28 and (B) 60 days following spinalization. There were no significant differences in mean reflex responses as a function of stimulus rate between the two conditions at either time point. Data shown are mean ± SEM.
(shown separately in Fig. 3.7.A and 3.7.B) were compared across conditions, an
interesting difference in rate-sensitivity between anesthetized and unanesthetized
spinalrats became apparent: At 28 days post-transection, the mean response to 5.0 Hz
for ketamine-anesthetized rats was 43.49 ± 10.94%, a significantly greater percent of
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Days 28 60 28 60 Sthn. Rale 1.0 Hz 1.0 Hz 5.0 Hz 5.0 Hz
Fig. 3.9. H-reflexes plotted as percent of control show a difference in the pattern of rate- sensitive depression between the two anesthetic conditions at 28 days post-transection. * At 5.0 Hz, there was significantly greater suppression of H-reflexes in unanesthetized rats than in rats treated with ketamine. Data shown are mean ± SEM.
control than the 15.20 ± 3.48% response for unanesthetized rats, J (10) = 2.47, g < 0.05.
That is, ketamine reduced the amount of H-reflex rate-sensitive depression, the
inhibitory mechanism that was already compromised as a result of spinalization. This
eftect seemed to be transient, because the same difference was not found between the
two groups at 60 days, t (10) = 0.12, g > 0.05.
EXPERIMENT 3
Both the monosynaptic H- and polysynaptic flexor reflexes remained relatively
constant throughout the one hour test period. As shown in Fig. 3.10.A, data from H-
reflex recordings, represented by H/M ratios, did not differ significantly across
measurement time points, F (2,10) = 0.04, g > 0.05. Similarly, as shown in Fig. 3.10.B,
flexor reflex responses did not exhibit significant variation from pre- to post-injection
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 74
periods, £ (2, 10) = 1.78, p > 0.05. Such consistency in responses across time
demonstrates the stability of this preparation.
H-Reflex B Flexor Reflex 1.0 n 100 *3* e 0.8 •
Fig. 3.10. Stability of the H- and flexor reflexes in chronic spinal rats during a 1 hr session o f recordings. Mean ± SEM values obtained prior to and 15, 30 and 60 min. after saline injection are shown in A. for H/M ratios that were calculated from recruitment curves; B. shows the mean ± SEM percent of baseline AUC values for flexor reflexes.
3.4 Discussion
SUMMARY OF FINDINGS
The results o f these experiments indicate that spasticity developed within one month
of thoracic spinal transection and further, established the response stability of the
unanesthetized chronic spinal rat as a model for testing antispastic agents. In Experiment
1,28- and 60-day spinal rats tested under ketamine anesthesia exhibited decreased H-
reflex thresholds, increased response magnitudes, and decreased rate-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 75
sensitive depression relative to the intact condition. Without anesthesia, rats tested at 28
and 60 days post-transection exhibited similar thresholds and magnitudes, but at 28 days,
exhibited a different pattern of rate-sensitive depression than was displayed under
ketamine. In the third experiment, both H/M ratios and flexor reflexes were reliably
recorded four separate times in one hour, with no significant changes across time.
Reflex Threshold
In the intact condition, thresholds for rats anesthetized with isoflurane (in Chapter 2)
and with ketamine (Experiment 1, this chapter) were of similar stimulus intensity. In
Experiment 1, H-reflex thresholds in ketamine-anesthetized rats decreased from intact to
28 and remained low at 60 days post-transection (Fig. 3.3). Thresholds for
unanesthetized spinal rats in Experiment 2 matched those of ketamine-anesthetized spinal
rats at 28 and 60 days, in contrast to those of isoflurane-anesthetized rats in Chapter 2,
which remained at intact levels throughout the two-month period. Thus, whereas the use
of isoflurane during recording sessions indicated that spinalization had no effect on H-
reflex thresholds (Fig. 2.2), it appears from the present data obtained with and without
ketamine that, compared to intact rats, H-reflex thresholds are significantly lower in
chronic spinal rats, consistent with clinical (Katz & Rymer, 1989, Levin & Hui-Chan,
1993; Powers et al., 1988; Taylor et al., 1984) and experimental reports (Thompson et
al., 1992). Additionally, the results o f these two experiments reinforce the masking
effect of isoflurane on reflex threshold that was suggested in Chapter 2.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76
Response magnitude. H-reflex response magnitudes in ketamine-anesthetized rats
increased from intact to 28 days post-transection and remained enhanced at 60 days.
Response magnitudes recorded in unanesthetized rats at 28 and 60 days did not
significantly differ from those recorded under ketamineat either time point As
opposed to H-reflex recordings obtained under isoflurane (in Chapter 2), development
of exaggerrated H-reflex responses were observed under ketamine within the predicted
period of 28 days post-injury (Skinner et al., 1996; Thompson et al., 1992). Thus, for
evaluation of H-reflexes at the control rate of stimulation, ketamine does not appear to
interfere significantly with the outcome.
Rate-sensitive depression. Comparisons among percent of control rate responses
for rats tested under isoflurane, ketamine, and no anesthesia indicated effects of
anesthesia on rate-sensitive depression. In response to the 1.0 Hz stimulus rate,
isoflurane-anesthetized rats (described in Chapter 2) did not exhibit a transection-
induced loss of rate-sensitive depression (see Fig. 2.5); ketamine-anesthetized spinal
rats displayed a reduction in depression at both 28 and 60 days (see Fig. 3.7.A); and
unanesthetized spinals exhibited a reduction in depression from 28 to 60 days (see Fig.
3.7.B). In response to the 5.0 Hz stimulus rate, isoflurane-anesthetized rats exhibited a
loss of rate-sensitive depression from 28 to 60 days; ketamine-anesthetized rats
displayed no changes in rate-sensitive depression from intact levels; and unanesthetized
rats exhibited a loss of rate-sensitive depression from 28 to 60 days post-transection. In
other words, ketamine-anesthetized rats responded like unanesthetized rats to the 1.0 Hz
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 77
stimulus whereas isoflurane-anesthetized rats responded like unanesthetized rats to the
5.0 Hz stimulus.
Perhaps the most compelling finding revealed by the comparisons made between
reflex recordings obtained with and without anesthesia is the difference in rate-sensitive
depression at 5.0 Hz between ketamine-anesthetized and unanesthetized 28-day spinal
rats. As shown in Fig. 3.9, the animals which did not receive ketamine displayed
greater response inhibition at 5.0 Hz than did ketamine-treated animals. That is,
compared to awake chronic spinal rats, ketamine-anesthetized spinal rats scored higher
on this index of spasticity.
Paradoxical Effects of Ketamine
Based on the hypothesis that mono- and polysynaptic reflexes are mediated by non-
NMDA and NMDA receptors, respectively (Davies & Watkins, 1981; Evans et al.,
1982; Padjen & Smith, 1980), it was surprising that ketamine, an NMDA receptor
antagonist, significantly affected the monosynaptic H-reflex in the present study.
Despite the history of reports supporting this receptor subtype-reflex dissociation, its
origination and general adherence have been historically based upon the use of
nonselective non-NMDA antagonists, which are incapable of accurately make such a
distinction. In light of this, a reappraisal of the assumption has been suggested (Davies,
1988) and in fact, the data from this and the previous chapter provide a unique
perspective from which to direct such an investigation. Consistent with reports
concerning excitatory amino acid receptor mediation of reflexes which concluded that
NMDA receptors do not participate in the anatomically defined monosynaptic pathway
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. (Davies & Watkins, 1983; Jessell et al., 1986; Watkins, 1984), ketamine did not affect
the monosynaptic H-reflex at the control rate (0.3 Hz) in the present study. Ketamine
did, however, significantly affect H-reflexes elicited at a frequency (5.0 Hz) known to
produce low-frequency, or rate-sensitive, reflex depression, which is an expression of
presynaptic inhibition (Curtis & Eccles, 1960; Kaizawa & Takahashi, 1970; Lloyd,
1957; Lloyd & Wilson, 1957).
As an open-channel blocker, ketamine blocks NMDA receptors in a voltage and use-
dependent manner, thus requiring activation of the NMDA receptor-linked channel in
order to produce its effect (Honey, Miljkovic, & MacDonald, 1985). Given the stimulus
parameters used to elicit the H-reflex, it is difficult though not inconceivable,
considering the spastic state of the animals after 28 days of spinalization, to conclude
that a strong, widespread activation of intemeurons sufficient to engage the action of
ketamine occurred at 5.0 Hz. Nonetheless, if such activation did occur, then the
expected action of ketamine as an NMDA receptor antagonist would be to block
excitatory responses and thus decrease reflex amplitudes. Yet paradoxically, the
opposite effect was revealed at one month post-transection. It remains to be seen
whether a loss o f rate-sensitive depression would be associated with ketamine in an
intact rat, especially at a low frequency.
The loss of rate-sensitive depression could be explained by the reported ability of
ketamine to reduce acetylcholine, dorsal root and ventral-root evoked excitation of
Renshaw cells (Anis, Berry, Burton & Lodge, 1983). If ketamine was acting to reduce
Renshaw/recurrent inhibition, then an enhanced response would be predicted and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 79
indeed, this was the case in the present study: At 28 days post-transection, ketamine-
anesthetized spinal rats displayed significantly greater H-reflex responses to 5.0 Hz
(relative to 0.3 Hz) than unanesthetized spinal rats at the same time point
The fact that there was no longer a difference in rale-sensitive depression between
ketamine-anesthetized and unanesthetized spinals at 60 days suggests that there could be
a limit to the response to 5.0 Hz under ketamine that had already been reached. That
this may be the case is illustrated by the lack of change in percent of control response at
5.0 Hz from the intact condition. Under isoflurane, however, a change in rate-sensitive
depression did occur between 28 and 60 days, suggesting that a different mechanism is
involved. Isoflurane has been proposed to have an indirect effect on GABAergic
presynaptic inhibition, and the increase in response to 5.0 Hz at 60 days under isoflurane
may be related to this activity.
Pathological Changes Following Spinal Injury
Interfering interactions of anesthesia aside, the reductions in rate-sensitive
depression from 28 to 60 days post-transection in response to stimulation at 1.0 Hz and
5.0 Hz in Experiment 2 may very well be interpreted in terms of the progressive
histopathological changes thought to underlie spasticity (Katz & Rymer, 1989; Murray
& Goldberger, 1974; Wiesendanger, 1991). That is, at 28 days post-transection,
increased motoneuronal excitability and enhancement of evoked synaptic excitation of
motoneurons inevitably exists. However at 60 days post-transection, the mechanisms
responsible for increased central synaptic excitability, e.g., collateral sprouting, have
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 80
likely become further advanced, resulting in exacerbation of the spastic condition with
time (Little & Halar, 1985).
Commonalities Between Clinical Spasticity and the Experimental Model
The appropriateness of this model for evaluating potential antispasdc agents, as
demonstrated in the present study, is largely a function of its similarity to the spastic
patient. Chronic spinal rats display particular features that simulate the condition of
human spasticity. As is the case for patients with spasticity secondary to spinal cord
trauma, transection-induced spasticity manifests as a chronic condition in the rat
Symptoms develop over a period of weeks after injury and persist indefinitely. Such
gradual development of spasticity in humans and in chronic spinal rats is an important
point to consider in contrast to acute models. Acute periods following spinal cord injury
are associated with spinal shock, hyporeflexia, and increased rate-sensitive depression
(Calancie et al., 1993; Leis et al., 1996; Thompson et al., 1992). In contrast, the chronic
phase of spasticity is characterized by hyperreflexia and decreased rate-sensitive
depression (Ashby et al., 1974; Calancie et al., 1993; Delwaide, 1973; Levin & Hui-
Chan, 1993; Little & Halar, 1985; Nielsen et al., 1993; Powers et al., 1988).
Regardless of the type of injury to the spinal cord, primary cellular damage is
followed by secondary cellular changes including collateral sprouting (Murray &
Goldberger, 1974), synaptogenesis and receptor supersensitization (reviewed in Katz &
Rymer, 1989; also see Wiesendanger, 1991), and even changes in neurotransmitter
content in primary afferent terminals (Kramarevsky et al., 1997). These
histopathological changes parallel the development of spasticity. Because spasticity is a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 81
slow-developing and enduring disorder, matching the immediate cellular pathology
commonly observed in humans with respect to injury experience (i.e., contusion injury)
may not be so critical. Rather, for the purpose of testing potential antispastics, matching
the ‘secondary’ situation may be more pertinent.
Although humans rarely sustain injuries that completely transect the cord, clinical
abnormalities identified with the spastic syndrome have been traced to segmental-level
circuitry (e.g., Fasano, Barolat-Romana, Zeme & Sguazzi, 1979). Chronic spinal rats
display prominent, quantifiable symptoms of clinical spasticity that1 ) develop along a
comparable timeline and 2) can be objectively evaluated under similar conditions.
Importantly, the present study (Experiments 2 and 3) demonstrates that spastic indices
can be evaluated in the absence of a significant confounding factor that is required in
other models, i.e., the use of anesthesia during testing.
Conclusion
Hyperreflexia and rate sensitivity of the H-reflex differentiate normal from spastic
subjects clinically in humans (Nielsen et al., 1995; Faist et al., 1994; Taylor et al., 1984;
Ashby et al., 1974; Angel & Hoffmann, 1963), experimentally in animals (Thompson et
al., 1992; Skinner et al., 1996) and in this study, distinguished intact from 28- and 60-
day chronic spinal rats. Results from these experiments indicate that, as defined and
quantified by three parameters of the H-reflex, midthoracic spinal transection in rats
results in the development of characteristic signs of spasticity by one month post-
spinalization. These include but are not limited to: increased sensitivity to stimulus,
exaggerated H-reflexes and reduced reflex inhibition.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 82
Similar spastic behaviors could be detected in both awake and ketamine-
anesthetized spinal animals, and ketamine appeared to be a fairly non-intrusive
anesthetic for measuring spinal reflexes, with the exception of one index: rate-sensitive
depression. Although development of hyperexaggeration as a result of transection was
not documented for polysynaptic reflexes, it was demonstrated that flexor reflexes,
along with H/M ratios, could be reliably, repeatedly measured in awake, unanesthetized
spinal rats one month after surgery. Taken together, the results from these experiments
indicate that the chronic spinal rat is a functional, advantageous model that can be
effectively used to evaluate potential antispastic agents for clinical use in spinal cord
injury-induced spasticity.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 4. INVESTIGATION OF NMDA AND NON-NMDA RECEPTOR ANTAGONIST ACTIONS ON SPINAL TRANSECTION-INDUCED SPASTICITY
4.1 Introduction
Currently, the most commonly used agent for treatment of spastic symptoms is
baclofen, a GAB AB receptor agonist which acts to enhance GAB Aergic inhibition (Lev-
Tov et al., 1988). Although it is considered the most effective agent available, baclofen
does not work for all patients and it has undesirable side effects including muscle
weakness and CNS depression (Rice, 1987; Smith et al., 1994). Therefore, development
of new, more selective and efficacious pharmacological treatments would be
advantageous.
An antispastic agent could potentially act to diminish spasticity not only by
enhancing inhibition but also by reducing excitation within the spinal cord. For this
reason, excitatory amino acid (EAA) receptor antagonists have been suggested for use in
the pharmacological treatment of spasticity (Schwarz et al., 1988a; 1992; Lehman et al.,
1987; Davidofif & Hackmann, 1988; Turski et al., 1985a; 1987b; Turski, Klockgether,
Turski, Ikonomidou, Schwarz & Sontag, 1988; 1990; 1992). Two types of ionotropic
glutamate receptors, non-NMDA and NMDA, have been considered as possible
antispastics. Specifically, it has been shown that non-NMDA receptor antagonists
selectively reduce the monosynaptic H-reflex and polysynaptic flexor reflexes are
selectively reduced by NMDA receptor antagonists (Turski et al., 1987b; 1990; 1992).
However, as shown in the previous chapter, the NMDA receptor antagonist, ketamine,
had an effect on the monosynaptic H-reflex.
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 84
In a study with intact, anesthetized mice, Turski et al. (1992) reported that very low
doses of the AMP A receptor antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl-
benzo[f]quinoxaline (NBQX), at 0.1 mmol/kg, i.v. and 0.01-1 nmol, i.t., depressed the
H-reflex without affecting the flexor reflex. In genetically spastic rats tested in the same
study, NBQX (0.05-0.1 mmol/kg, i.p.) also reduced spontaneous EMG. Conversely, the
NMDA receptor antagonist (3-((±)-2-carboxypiperazin-4-yl)propyl-l-phosphonic acid
(CPP) depressed the flexor reflex without altering the H-reflex in anesthetized, intact
rats (0.1 mmol/kg, i.v.) and mice (10-100 nmol, i.t.; Turski et al., 1987; 1990). Like
NBQX, CPP (0.002 (imol, i.t. and 0.1 mmol/kg, i.p.) also reduced the elevated tonic
EMG in spastic rats. It is difficult to interpret these findings with respect to actual drug
effects, given the procedural use of anesthesia during reflex recordings.
Therefore, the present study used the chronic spinal rat to investigate the actions of
these two EAA receptor antagonists on indices of the H-reflex and the flexor reflex.
NBQX and CPP were chosen because of their high receptor selectivity and potency as
competitive non-NMDA (Sheardown et al., 1990; Smith et al., 1991; Porter &
Greenamyre, 1994) and NMDA (Davies at al. 1986; Lehmann et al., 1987) receptor
antagonists, as well as for comparison with previous reports. Behavioral manifestations
of spasticity were measured before and 30 and 60 minutes after injection of NBQX,
CPP or control vehicle. Changes from pre- to post-injection were investigated using 1)
two parameters of the monosynaptic H-reflex: response magnitude and rate-sensitive
depression and 2) the polysynaptic flexor reflex response. H-reflexes were elicited by
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 85
tibia! nerve stimulation at three different frequencies, 0.3,1.0 and 5.0 Hz, and recorded
from the plantar muscles of the ipsilaleral hindpaw. Flexor reflexes were
elicited by electrical stimulation (500 Hz) of the hindpaw and recorded from the biceps
femoris/semitendinosous muscles.
4.2 Methods
SUBJECTS
A total of 21 male Sprague-Dawley rats (Holtzman Laboratories, Madison, WI)
weighing approximately 250-300 g at the beginning of experiments was used. Prior to
spinalization, animals were housed independently in suspended steel cages in a colony
room maintainedon a 12 hr/1 2 hr light/dark cycle, with dark onset at 19.00 hr. Food
and water were available ad libitum. At the end of experimental procedures, rats were
euthanized by an anesthetic overdose. All procedures have been reviewed and approved
by the Institutional Animal Care and Use Committee of Louisiana State University,
Baton Rouge, LA.
SPINAL CORD TRANSECTION
For surgical procedures, animals were anesthetized using a mixture of isoflurane
(AErrane, Anaqest, Madison, WI) and oxygen (30% isoflurane in 70% oxygen) at a
flow rate of 4 L/min. A dorsal surface incision was made in the skin behind the ears and
extended approximately two inches caudally. Following retraction o f the paraspinal
muscles, a laminectomy was performed between T6 and T9 to expose the spinal cord.
A 1-2 mm segment o f the cord was removed by excavation. The excised tissue was
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 86
then replaced by gelatin foam to reduce bleeding after which the incision was closed in
layers.
Maintenance of Spinal Animals
Following spinal transection, animalswere placed in individual plastic cages in a
room in which the ambient temperature was raised to approximately 25 °C to maintain
body temperature. Beginning approximately 12 hr post-surgery, the rats’ urine was
expressed manually by applying pressure to their bladders. This was done twice daily
until animals recovered to the point of no longer requiring such care (approximately two
weeks), at which point ambient temperature was returned to 21 °C.
Electrophysiological Measurements
All electrophysiological measurements were obtained using the Nicolet Viking IV.
Electrical signals were amplified, band-pass filtered (5 Hz-10 kHz) and full-wave
rectified. Animals were secured in a rodent torso sling and suspended in a sling frame
such that all four limbs hung freely to facilitate electrophysiological recordings,.
H-reflex threshold and response magnitude. To determine the range of stimulus
intensities required to elicit minimum and maximum H- and M-wave amplitudes, a
recruitment curve consisting of successive stimulations (frequency: 0.3 Hz) of
increasing intensity was obtained. Reflex threshold was defined as the stimulus
intensity associated with the maximum H-reflex amplitude. This threshold stimulus
intensity was then used to elicit three separate triads of10 consecutive responses each,
with a one-minute inter-trial interval. Each trial was delivered at one of three
frequencies, in the following order: 0.3, 1.0 and 5.0 Hz. The magnitudes of the evoked
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 87
EMG potentials were determined by measuring the peak-to-peak H-wave amplitudes
(mV).
Electrical stimulation at 0.3 Hz or lower produces no significant effect on
subsequent response amplitudes (Eccles and Rail, 1951; Meinck, 1976), and frequencies
higher than 0.3 Hz are known to have a depressive effect on responses to subsequent
stimulation (i.e., rate-sensitive depression; see below). Therefore, H-reflex response
magnitude, the basic measure of the reflex response, was defined based on the H-wave
amplitudes elicited at the control stimulus rate of 0.3 Hz.
As indicated above, 10 responses were elicited at each rate. Because magnitudes of
the initial responses demonstrated a tendency to vary from those of subsequent
responses, the first two H-wave amplitudes were not included in the calculation of the
mean (Kaizawa & Takahashi, 1970). To decrease variability (Thompson et al., 1992),
mean H-wave amplitudes were calculated using 8 of the 10 responses for each rate.
Rate-sensitive depression. Rate-sensitive depression of the H-reflex was evaluated
two ways: 1) Reflex response magnitudes of H-waves elicited at the control frequency
(0.3 Hz) were compared with those of H-waves elicited at 1.0 and 5.0 Hz. 2) Rate-
sensitive depression was calculated by normalizing reflex response magnitudes at 1.0
and 5.0 Hz to their corresponding control rate magnitudes. The standardized scores for
1.0 and 5.0 Hz were expressed as percents of control, as shown by the formula below:
(1.0 or 5.0 Hz) mean response magnitude ■ ...... - x 100 = % of Control 0.3 Hz mean response magnitude
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 88
Flexor reflex threshold and response magnitude. Polysynaptic flexor reflexes
were evoked by electrical stimulation of the hindpaw and recorded from the flexor
muscles of the ipsilateral hindlimb. A pair of subdermal platinum needle electrodes
inserted into the interphalangeal area of the hindpaw was used for stimulation and a pair
of stainless-steel barbed-broaches inserted percutaneously into the biceps femoris/
semitendinosus muscles was used for recording. To establish reflex threshold,
stimulation (five square-wave pulses, 500 Hz, 0.2 ms duration) of increasing intensity
was delivered every 10 s beginning at 0 mA until the smallest flexor reflex was visibly
detected as output on the computer monitor; the associated intensity was defined as
reflex threshold. For testing, stimulation was administered five consecutive times (10 s
inter-stimulus intervals) at 2.0-3 .0 times reflex threshold to elicit five reflexes, followed
by a one minute interval and then another trial of five reflexes, elicited at a stimulus
intensity of 5.0 mA. This was done to allow two types of comparisons across treatment
groups: reflex response magnitudes at a uniform stimulus with respect to individual
threshold levels, and at a uniform stimulus with respect to actual intensity. Flexor reflex
response magnitudes were generated by a computer software program (Nicolet Viking
IV) which quantified each response by measuring the area bounded by the averaged
response and the baseline, i.e., the area under the curve (AUC). The mean AUC value
for the five consecutive responses was subsequently calculated and served as the data
point for each measurement period.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89
Drug Administration
CPP and NBQX (Research Biochemicals International, Natick, MA) were
administered as experimental treatments. (±)-CPP and NBQX disodium (2 mg and 4
mg) were each dissolved separately in I ml sterile water to make solutions of 2 mg/ml
and 4 mg/ml that were injected systemically (s.c.) as 2 mg/kg and 4 mg/kg. These doses
were chosen based on past reports and a series of dose-range finding studies conducted
prior to experimentation. The feasibility of systemic delivery was determined in
advance based on previous reports which demonstrated the ability of both NBQX
(Smith et al., 1991) and CPP (Lehman et al., 1987) to cross the blood-brain barrier.
Each animalin the control group received a subcutaneous injection of sterile water.
Procedures
Spinal animals (28 days post-transection) were divided into five groups according to
treatment: 1) Control (n = 5); 2) NBQX 2 mg/kg (n = 4); 3) NBQX 4 mg/kg (n = 4); 4)
CPP 2 mg/kg (n = 4); and 5) CPP 4 mg/kg (a = 4). Each animal received only one
treatment (injection of one dose of an antagonist or the control vehicle) and was tested
for both H-reflex and flexor reflex responses. A test session consisted of pre-injection
recordings followed by injection of either control or drug solution and two subsequent
recording sessions. Three things were accomplished during the pre-injection recording
session, in the following order: 1) determination o f H-reflex threshold, 2) 3 trials of 10
successive H-reflex responses (one trial per rate, given at a one-minute inter-trial
interval, in the order 0.3,1.0,5.0 Hz) and 3) determination of flexor reflex threshold.
Post-injection recording sessions were initiated approximately 25 and 55 minutes after
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 90
injection. These recording time points (30 and 60 minutes) were chosen based on past
reports and dose range-finding studies conducted prior to experimentation. For the two
post-injection sessions, the H-reflex was always recorded first, followed by a one to
two-minute interval, after which five flexor reflexes were recorded. The flexor reflex
was always recorded after the the H-reflex to avoid any influence from heterosynaptic
inhibition of la fibers by flexor afferents, an action known to be preserved in spinally-
injured subjects (Roby-Brami and Bussel, 1990; 1992). Additionally, whenever
possible, H- and flexor reflexes were recorded in alternate hindlimbs.
Statistical Analysis
Data obtained as H-reflexes were analyzed as three different measures: 1) At each
time point, as mean H-wave amplitude (mV); 2) at 30 and 60 minutes post-injection, as
percent of pre-injection baseline amplitude (for all rates, relative to respective pre
injection responses); and 3) at each time point, as the percent of control rate (0.3 Hz) for
responses to 1.0 and 5.0 Hz obtained at the same recording sessions, i.e., as rate-
sensitive depression. Data obtained as flexor reflexes were analyzed as mean AUC
values for each measurement period. Data were analyzed using analysis of variance
(ANOVA) and repeated measures ANOVA. Multiple comparisons of within subject
factors were carried out using a p rio ri contrasts; p o st hoc multiple comparisons of
between group factors were carried out using Dunnett's t-tests. Significance level for all
analyses was set at p s 0.05.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 91
4 3 Results
CONTROL
As shown in Fig. 4.1, there were no significant changes in mean H-reflex amplitude
or percent of control rate in the control group and as shown in parts A and C, rate-
sensitive depression occurred in response to 5.0 Hz at all time points.
H-Reflex Response Magnitude
NBQX. The non-NMDA receptor antagonist NBQX reduced H-reflex responses
elicited at the control rate. NBQX 2 mg/kg initiated a decrease in H-wave amplitude
from pre- to 30 minutes post-injection, F (1, 3) = 15.98, p < 0.05, followed
by a return to baseline by 60 minutes, as shown in Fig. 4.2. A. NBQX 4 mg/kg also
initiated a decrease in control rate amplitude that was observed at 30 minutes, E (1, 3) =
9.88, p = 0.05. The higher dose seemed to have a prolonged effect, as a return toward
baseline was not evident during the 60 minute test period.
CPP. The NMDA receptor antagonist CPP demonstrated efficacy in modifying the
monosynaptic H-reflex response. There was no effect of the lower dose of CPP (2
mg/kg) on amplitude or rate-sensitive depression at either time point (Fig. 4.3.A).
However, 30 minutes after injection of the higher dose (4 mg/kg), the amplitude at 5.0
Hz was significantly reduced: Animals treated with CPP 4 mg/kg exhibited a significant
decrease in mean H-wave amplitude from pre-injection values, E (1, 3) = 13.25, p <
0.05.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 92
> £ 5 ■
Fig. 4.1. Composite figure of the three measures used to evaluate spasticity, shown for the control group. H-wave amplitudes are plotted: A as a function of stimulus rate;B. as a function of pre-injection baseline amplitudes; and C. as percent of the control rate. Rate-sensitive depression was observed in response to the 5.0 Hz stimulus at all time points. *Significant difference between 5.0 Hz and 0.3 Hz. 4* Significant difference between 5.0 and 1.0 Hz. Data shown are mean ± SEM.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 93
NBQX 2 mg/kg B NBQX 4 mg/kg
2
1
0 Pre-inj. 30 min. 60 min. Pre-inj. 30 min. 60 min. 0.3 Hz 1.0% 5.0%
Fig. 4.2. Effect of NBQX on rate-sensitive depression. A. There was a loss of rate- sensitive depression 30 minutes after administration of 2 mg/kg NBQX which did not recover by 60 minutes. B. At the higher dose of NBQX, 4 mg/kg, rate-sensitive depression was present before and 30 minutes after injection, but it was no longer evident at 60 minutes. * Significant difference between 5.0 and 0.3 Hz. ^Significant difference between 5.0 and 1.0 Hz. + Significant decrease from pre-injection value.
Percent of Baseline
The mean H-wave amplitude obtained at the pre-injection recording session was set
to 100% for each subject individually. Data recorded from animals at the two post
injection sessions were then normalized to their respective pre-injection baselines.
Reduction in the percent of pre-injection baseline thus signifies improvement (i.e.,
reduction) in the hyperreflexic status of the response at a given frequency.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94
CPP 2 mg/kg
k
Pre-lnj. 30 mm. 60 win-
B CPP 4 mg/kg
ti 1 .0 Pre-inj. 30 min. 6 0 m in . SB 0.0 Pre-lnj. 30 mm. 60 min.
Fig. 43. Effect o f CPP on rate-sensitive depression. A. Rate-sensitive depression was observed before and 30 and 60 minutes after administration of CPP 2 mg/kg. ^Difference between 5.0 and 0.3 Hz approached significance before (p = 0.057) and 30 minutes after injection (p = 0.056). -rSignificant difference between 5.0 and 1.0 Hz. B. At the higher dose of CPP, 4 mg/kg, rate-sensitive depression was also observed at each of the three points. *Significant difference between 5.0 and 0.3 Hz. *rSignificant difference between 5.0 and 1.0 Hz. OSignificant difference between 1.0 and 0.3 Hz. ♦Significant decrease from pre-injection value.
NBQX. H-reflex responses elicited at the control rate in animals that received
NBQX 2 mg/kg were reduced to approximately 50% of their pre-injection amplitude
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 95
after 30 minutes. As shown in Fig 4.4.A, this was a significant change from
pre-injection values, £ (1,3) = 14.5, p < 0.05, and as shown in Fig. 4.6.A, it was also
significantly lower than the mean response of the control group at the same time and
rate, t (7)= 3.95, p < 0.01, which remained unchanged (97.45% of baseline). The same
group of animals also showed a response reduction at 30 minutes to the 1.0 Hz stimulus,
£ (1, 3) = 9.97, p = 0.05, although it was not lower than the corresponding control group
response at the same time point As shown in Fig. 4.4.B, a continual though non
significant trend in the same direction from pre-injection values to 60 minutes post
injection of NBQX 4 mg/kg was apparent
NBQX 2 mg/kg B NBQX 4 mg/kg 120 120 v -j= 100 100 ■
SB 80 • 60 • 40 • A t
o 20 •
Pre-inj. 30 min. 60 min. Pre-inj. 30 min. 60 min. — 0.3 Kt - □ - 1 .0 Hr - A - 5.0 Hr
Fig. 4.4. H-reflex responses after NBQX are plotted as percent of pre-injection values at each of the respective stimulus rates. A. H-reflexes elicited at 0.3 Hz and 1.0 Hz were reduced at 30 minutes after administration of 2 mg/kg NBQX. B. 4 mg/kg NBQX did not induce a significant decrease in percent of pre-injection values at any frequency. + Significant decrease from pre-injection baseline.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96
CPP. Normalization of H-reflex amplitudes to pre-injection values revealed that the
reduction in amplitude at 5.0 Hz under CPP 4 mg/kg was also close to a 50% drop from
the pre-injection baseline, E (1,3) = 114.64, g < 0.01, as shown in Fig. 4.53. And as
shown in Fig. 4 .6 3 , this value was significantly different from that of the control group
at the same time and stimulus rate, t (7) = 3.31, p < 0.05. Again, no changes from pre
injection values as a result of treatment were exhibited by animals that received CPP 2
mg/kg (see Fig. 4.5 A).
A CPP 2 mg/kg B CPP 4 mg/kg 120 120 X 100 100 an
40
Pre-inj. 30 min. 60 min. Pre-inj. 30 min. 60 min. — oja - a- lohz A 5.0 Hz
Fig. 4.5. H-reflex responses after administration of CPP are plotted as percent of pre injection values at each of the respective stimulus rates. A. The low dose of CPP did not result in any changes from percent of pre-injection baseline. B. However at 4 mg/kg, CPP induced a significant decrease from pre-injection baseline values of H-reflexes elicited at 5.0 Hz. + Significant decrease from pre-injection baseline.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 97
A 0.3 Hz «
Control N BQ X 2m g Pre-inj. 30 min. 60 min.
B 5.0 Hz
Control CPP 4 m g Pre-inj. 30 min. 60 min.
Fig. 4.6. Comparison of the effects of NBQX and CPP on H-reflexes elicited at 0.3 Hz and 5.0 Hz. Data shown in Fig. 4.4. and 4.5. are re-plotted here with control group values to illustrate the contrast between the actions o f the two antagonists. A. 2 mg/kg NBQX induced a decrease to approximately 50% of the pre-injection value at the 0.3 Hz control rate. B. By comparison, 4 mg/kg CPP induced a 50% decrease in H-reflexes elicited at 5.0 Hz. ♦ Significantly different from control group response.
Rate-Sensitive Depression
To specifically investigate changes in the amount of rate-sensitive depression, each
animal’s H-reflex amplitudes at 1.0 and 5.0 Hz were standardized as a function of
their respective control rate amplitudes. Changes in percent of control rate values
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 98
A NBQX 2 mg/kg CPP 2 mg/kg 160 • tm 140 • £ 1 2 0 •
2 1 0 0 • oe 80 • U 60 ■ *3 40 • £ 2 0 • 0 ■ Pre-lnj. 30 mfi». 60 min. Pre-inj. 30 min. 60 min.
B NBQX 4 mg/kg D CPP 4 mg/kg 160 • « 140 • 06 1 2 0 • "o **L. 1 0 0 • es 80 • O «— 60 ■ o 40 • £ 2 0 « 0 • Pre-lnj. 30 min. Pre-lnj. 30 min. 60 min. 1.0 Hz e = j i.o Be 5.0 Hz 5.0 Hz Fig. 4.7. Effect of antagonists on rate-modulation at 1.0 and 5.0 Hz. A. NBQX 2 mg/kg did not significantly change percent of control rate values from pre- to post-injection.B. Rate-sensitive depression at 1.0 Hz was significantly different from that at 5.0 Hz before injection of NBQX 4 mg/kg, but not at either time point post-injection. C ., D. Percent of control rate values for 1.0 Hz and 5.0 Hz were significantly different at all time points for animals that received injections of CPP. Rate-sensitive depression at 5.0 Hz significantly increased 30 minutes after injection of 4 mg/kg CPP (as indicated by a significant decrease in percent of control rate). *r Significant difference between rate- sensitive depression at 5.0 Hz and 1.0 Hz. + Significant decrease in percent of control rate at 5.0 Hz from pre- to 30 minutes post-injection.
indicate changes in the amount of rate-sensitive depression: A decrease in percent
of control rate conversely signifies an increase in the amount of rate-sensitive
depression.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 99
NBQX. At the low dose, 2 mg/kg, there was no difference in the amount of rate-
sensitive depression at 1.0 and 5.0 Hz. However, as shown by the apparent increases
in percent of control rate at 30 minutes in Fig 4.7 A, H-reflexes elicited at both
frequencies exhibited non-significant decreases in rate-sensitivity following NBQX 2
mg/kg administration. At the higher dose (4 mg/kg NBQX) no significant effect on
A NBQX B CPP 2 0 i 20 -I
15 ■ 15 ■ a
Control 2 mg 4 mg Control 2 mg 4 mg
2.0-3.01 Wh - A- 5 mA
Fig. 4.8. Flexor reflex responses following administration of NBQX and CPP. Data are plotted as AUC values for flexors elicited at 2.0-3.0 x reflex threshold (RTh) and at 5.0 m A, 60 minutes post-injection. A ., B . Neither of the doses of NBQX and CPP demonstrated an effect on flexor reflex magnitude.
rate-sensitive depression was demonstrated, although the pre-injection difference
between 1.0 and 5.0 Hz was no longer present after administration.
CPP. As shown in Fig. 4.7.C, the low dose of CPP had no significant effect on
rate-sensitive depression. As shown in Fig. 4.7.D, the amount of rate-sensitive
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. depression at 5.0 Hz was effectively increased following administration of CPP 4 mg/kg.
Compared to the pre-injection value, 5.0 Hz elicited a significantly lower percent of the
control rate H-reflex response at 30 minutes post-injection, E (1, 3) = 214.55, p =0.001
in animals that received CPP 4 mg/kg. A similar change from pre-injection did not occur
in the control group, £ (4) = 1.26, p > 0.05.
Flexor Reflex Response Magnitude
Flexor reflex response magnitudes inanimals recorded 30 and 60 minutes after
administration of control vehicle were compared with responses recorded in animals that
received NBQX (2 mg/kg and 4 mg/kg) and CPP (2 mg/kg and 4 mg/kg). There were
no significant differences between either dose of NBQX and control at either time point,
E (2, 9) = 0.48, p > 0.05). No significant differences were found between control group
responses and responses from the groups that received an injection of either dose o f CPP
(2 mg/kg or 4 mg/kg), F (2,9) = 1.61, p > 0.05.
4.4 Discussion
SUMMARY OF FINDINGS
Results from the present study indicate that, in unanesthetized chronic spinal rats,
both NBQX and CPP are capable of affecting the monosynaptic H-reflex in a rate-
dependent manner NBQX reduced the H-reflex at the control rate of stimulation, 0.3
Hz, whereas CPP reduced the H-reflex at 5.0 Hz. Neither antagonist, however,
significantly affected the flexor reflex response.
In an attempt to reconcile discrepancies between the present findings and those of
past reports, three main factors must be taken into account: 1) stimulus parameters; 2 )
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 101
condition of animal subjects (i.e., spinal or intact); and 3) procedural use of anesthesia.
Importantly, relevant previous reports of receptor-reflex selectivity were all carried out
in intact, anesthetized rats (Turski et al., 1987b; 1990; 1992). Each of these factors has
significant bearing on the outcome of results and generalization of findings.
FREQUENCY OF STIMULATION
H-Refiexes
In the present study, two parameters of the H-reflex were measured: H-reflex
amplitude (at the control rate) and H-reflex rate-sensitive depression at 1.0 and 5.0 Hz.
Although the stimulus repetition rate was not specified in the experiments with NBQX
and CPP carried out by Turski and colleagues, an earlier study by the group indicated
that 0.1 Hz was used (Schwarz et al., 1988b). For purposes of comparison between
their results and those of the present study, it was assumed (based on the consistency of
methodology throughout the series of studies published by their laboratory within a
proximal time period) that the same rate of stimulation was used in the studies involving
NBQX and CPP. At the very least, it was assumed that a frequency of no greater than
0.3 Hz was used, given the relevant literature (Curtis & Eccles, 1960; Eccles & Rail,
1951; Jefferson & Schlapp, 1953; Lloyd, 1957; Lloyd & Wilson, 1957).
NBQX. Given the assumption that £ 0.3 Hz was used, the present finding that
NBQX effectively reduced the H-reflex at the control rate of 0.3 Hz is consistent with
previous findings (Turski et al., 1992). However, there was a difference in the dose
required to elicit the reduction. Turski et al. (1992) reported a significant difference in
H-reflex amplitude between systemic injection of vehicle and NBQX, 0.05-0.1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102
mmol/kg, i.v. as opposed to the 2 mg/kg, s.c. dose of NBQX used in the present study.
The exact time course of this action was not specified, however, they reported
myorelaxant effects of systemic NBQX (0.1 mmol/kg, i.p.) that appeared immediately
after injection and lasted 60 to 90 minutes. This difference in dose effects is likely a
result of either spinalization or anesthesia, as discussed below.
CPP. The lack of effect of CPP on the H-reflex (at s 0.3 Hz) predicted by Turski et
al. (1992) was also found it the present study. The rate of 5.0 Hz was not used by their
group, though, in that rate-sensitive depression was not an index employed in their
investigation. Thus, the present result demonstrating that CPP reduced the H-reflex in
response to 5.0 Hz is a novel finding that was not predicted by prior results or by the
general theory of non-NMDA and NMDA receptor-reflex mediation (Davies &
Watkins, 1983).
Flexor Reflexes
Essentially the same stimulus parameters specified in the studies of Turski et al.,
(1987b; 1990; 1992) were followed in the present study, i.e., five consecutive reflexes
were evoked at 2.0-3.0 * reflex threshold at a frequency of 500 Hz and duration of 0.2
ms. The lack of effect of NBQX on the flexor reflex in this study corresponds to the
results of Turski et al. (1992), whereas the lack of an effect of CPP on the flexor is in
contrast to their findings and as well, to the general EAA receptor hypothesis
concerning polysynaptic reflex mediation (Davies & Watkins, 1983). However, in the
present study, CPP exhibited a trend toward reflex reduction in a dose-dependent
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 103
fashion which may indicate the potential for higher dose efficacy. This is also
considered below with respect to animal condition and use of anesthesia
INTACT VS. SPINAL SUBJECTS
Following spinal cord injury, spinal cord reflex circuitry undergoes significant
reorganization as a result of the loss of both white and grey matter. The loss of axons
and intemeurons as a result of moderate to severe injury such as contusion (Blight,
1983), ischemia (Davidoff, Graham, Shank, Werman, & Aprison, 1967) and other
spinal cord traumas (Taylor et al., 1984) dramatically reduces and often essentially
eliminates descending control from corticospinal projections, resulting in collateral
sprouting by any surviving axons and changes in excitablity of spared neurons. Such
progressive changes are thought to lead to the development of spasticity (Little & Halar,
1985; McCouch et al., 1958; Murray, 1997; Wiesendanger, 1991).
By contrast, intact animals, such as those used in previous reports concerning
NBQX and CPP action on spinal reflexes (Turski et al., 1987b; 1990; 1992), undergo no
such modifications and correspondingly, do not exhibit signs of spasticity. In the
absence of cellular and behavioral pathology related to spasticity, it is difficult to
generalize results concerning the efficacy of an antagonist in reducing hyperreflexia
from experiments with an intact animal to a clinical spastic population.
USE OF ANESTHESIA
General anesthetics have been shown to: 1) potentiate the release (Collins, 1980)
and postsynaptic effects (Gage & Robertson, 1985) of GAB A, 2) depress spinal
inhibitory potentials (Shimoji et al., 1990), and 3) have muscle relaxant action
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 104
(Ikonomidou-Turski et al., 1986). Any one of these properties could affect H- and
flexor reflex measurement, thus confounding is almost inevitable when anesthesia is
used in combination with an antispastic agent. The preceding two studies demonstrated
distinct actions of anesthesia during spinal reflex recording (see Chapter 2 and 3 for
reference), specifically emphasizing the inappropriate use of anesthesia during
evaluation of potential antispastics. Because both a general inhalation (isoflurane) and
dissociative (ketamine) anesthetic were found to alter reflex measurement, it seems
more than likely that the combined administration of an anesthetic and potential
antispastic drug would result in inaccurate evaluation of the drug. Nonetheless,
previous results of NBQX’s reduction of H-reflexes (Turski et al., 1992) and CPP’s
reduction of flexor reflexes (Turski et al., 1987b; 1990) taken to suggest their use as
antispastic agents were obtained from anesthetized mice (a-chloralose, 80 mg/kg, i.p.
and urethane 400 mg/kg, i.p., Turski et al., 1990; 1992) and anesthetized rats (H-reflex:
pentobarbitone, 50 mg/kg, i.p.; flexor reflexes: a-chloralose, 80 mg/kg, i.p. and urethane
400 mg/kg, i.p., Turski et al., 1987).
In regard to an explanation as to the differences between the present and prior
studies involving NBQX and CPP, the following is offered. 1) With respect to NBQX:
As a selective non-NMDA receptor antagonist, NBQX performed as predicted. That a
much higher dose was necessary could be explained in part by the absence of spastic
pathology in intact rats, i.e., given intact descending inhibition and spinal circuitry,
reflex depressive effects may be revealed at a much lower dose. Although only
speculative, the lack of anesthesia in the chronic spinal rats could possibly represent the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 105
lack of a potentiated or synergistic effect between the anesthetic and the EAA
antagonist 2) With respect to CPP: The failure of an NMDA receptor antagonist to
significantly depress a polysynaptic reflex in this case could be a function of dose.
Again, animals with spastic pathology, e.g., supersensitivity to neurotransmitters
(Sharpless, 1975), reduced presynaptic inhibition (Delwaide, 1973; Bernstein et al.,
1978) and collateral sprouting (McCouch et al., 1958), would likely require a higher
dose of an EAA antagonist to reduce excitation than would an intact, anesthetized
animal. The effect of CPP on the H-reflex, however, has interesting implications for
mechanisms of spinal reflex mediation and rate-sensitive depression, as well as for
antispastic agents used in the treatment of spasticity, as discussed below.
Rate Selective Effects of EAA Antagonists
The most compelling finding of the present study is that a selective non-NMDA and
NMDA antagonist both influencd the monosynaptic reflex. CPP’s effect on the
monosynaptic H-reflex was observed under different parameters than that of NBQX,
still, both were maximal at 30 minutes post-injection. Animalstreated with NBQX
exhibited a significant decrease in mean H-wave amplitude at the control rate of 0.3 Hz.
By comparison, CPP modified H-wave amplitudes at the highest stimulus rate used in
this study, 5.0 Hz.
Given the general EAA receptor-reflex dichotomy, it is intriguing that an NMDA
antagonist affected the monosynaptic H-reflex. The present findings, the results with
ketamine in Chapter 3, and others (see below) suggest that the ‘monosynaptic’
description of the H-reflex restricts effective functional characterization of the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 106
underlying physiology. Despite the various reportsregarding the receptor-reflex
dichotomy, the monosynaptic Ia-a-motoneuron pathway is subject to polysynaptic
influence (Burke et al., 1984; Van Boxtel, 1986), particularly via intemeuronally-
mediated presynaptic inhibition of la fiber terminals (Eichenberger & Ruegg, 1984). Ia
afferent activity is considered to generate, via one or more intemeurons, an auto-
presynaptic inhibition of its own activity. Through such a chain of intemeuronal
synapses, NMDA receptor-mediated intemeurons (and hence, NMDA receptor
antagonists) could participate in this inhibition by blocking excitatory influences. In the
cat, di- and trisynaptic connections between Ia afferents and homonymous motoneurons
have been demonstrated (Fetz et al., 1979; Jankowska et al., 1981). In humans,
heterogeneity of afferent fibers that evoke tendon jerk responses and possibly H-reflexes
has also been reported: A study in which composite EPSP latencies of tendon jerk and
H-reflex responses were measured in humans revealed a long rising phase and a long
latency to motoneuron discharge, indicating sufficient time for multisynaptic
contributions from intemeurons, which, as noted, are not directly implicated in the
monosynaptic pathway (Burke et al., 1984). Diffuse connections are made among
intemeurons in the spinal cord, and intemeurons are the main source of input to a-
motoneurons. Because of the immense communication network among spinal cord
intemeurons, it is reasonable to consider that the monosynaptic reflex is at least partly
mediated via NMDA receptors located on spinal intemeurons. Further, it is possible
that the selective NMDA receptor antagonist, CPP, was able to interact with receptors
on intemeurons located outside of the monosynaptic pathway to indirectly influence the
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 107
H- reflex, i.e., lead to an increase in presynaptic inhibition, as demonstrated by the
increased rate-sensitive depression found at 5.0 Hz in this study.
PRESYNAPTIC INHIBITION: EXTRINSIC OR INTRINSIC
Modulation of primary afferent release of neurotransmitter is hypothesized to occur
via two different mechanisms, extrinsic and intrinsic, based on the following theories: 1)
Rate-sensitive (or low-frequency) depression is considered to be an expression of
presynaptic inhibition (Curtis & Eccles, 1960; Kaizawa & Takahashi, 1970; Lloyd,
1957; Lloyd & Wilson, 1957). Reflex depression resulting from repetitive nerve fiber
stimulation as a function of rate is thought to result from an increase in presynaptic
inhibition, presumably via GABAergic intemeurons (Lev-Tov et al., 1988; Peshori et
al., 1998; Thompson et al., 1996). 2) In contrast, the reflex depression that occurs with
this type of repetitive homonymous stimulation is also referred to as post-activation
depression; the result is proposed to be solely a function of the changes that occur
exclusively within the Ia afferent terminal (Crone and Nielsen, 1989; Faist et al., 1994;
(Crone & Nielsen, 1989; Faist et al., 1994; Nielsen et al., 1993; 1995; Kohn et al.,
1997). In either case, the endpoint mechanism responsible for decreased reflex
amplitude is fairly unanimously agreed upon between the two perspectives, i.e., it is due
to a reduced quantal release of neurotranmsitter by Ia afiferents (Kuno, 1964).
Nonetheless, the mechanism responsible for reducing the release is of some debate
(Crone & Nielsen, 1989; Faist et al., 1994; Kohn et al., 1997; Nielsen et al., 1993; 1995;
Voigt & Sinkjaer, 1998).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 108
SHORT-LASTING VERSUS LONG-LASTING INHIBITION
In tests of the functional integrity of presynaptic inhibitory mechanisms of the FI-
reflex in humans, comparisons between the effects of homosynaptic and heterosynaptic
activation of Ia fibers indicated that two different types of reflex depression result
(Crone & Nielsen, 1989; Kohn et al., 1997; Nielsen et al., 1993; 1995): Heterosynaptic
fiber conditioning stimulation evoked a short-lasting (300-400 ms) reflex depression
labeled ‘presynaptic inhibition’, whereas homosynaptic conditioning stimulation (or
repetitive stimulation) elicited a long- lasting (2-10 s) reflex depression labeled ‘post
activation depression’ (Crone & Nielsen, 1989; Nielsen et al., 1993; 1995; Kohn et al.,
1997). Short-lasting inhibition most likely represents what has been referred to as
‘classical’ presynaptic inhibition (Wood, Gregory & Proske, 1996) and its mechanism is
most likely PAD, mediated by GABAa receptors. Long-lasting (post-activation or rate-
sensitive) depression that results from repetitive homonymous Ia fiber stimulation
outlasts presynaptic inhibition evoked by heterosynaptic activation (Nielsen et al., 1993;
1995) and likewise, outlasts the time course of GABAA-mediated presynaptic inhibition
(Fischer & Pamas, 1996b; Lev-Tov et al., 1988).
RATE-DEPENDENT ACTIONS OF GABA
Frequency-dependent actions of baclofen and CGP-35348 (a GABA b receptor
agonist and antagonist, respectively) were revealed in recent reports of presynaptic
modulation of transmitter release and frequency-related changes in Ia monosynaptic
EPSP amplitudes (Lev-Tov et al., 1988; Peshori et al., 1998). It has been suggested that
the depression following a test stimulus preceded by either a single stimulus or by brief
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 109
train stimulation is mediated by GABA a or GABA b receptors, respectively (Fischer &
Pamas, 1996b). That is, a single stimulus will evoke enough GABA release to activate
the fast, ‘classical’ presynaptic inhibition mediated via GABA a receptors whereas the
slower, longer time course of presynaptic inhibition following train or repetitive
stimulation suggests a role for second-messengers, such as those associated with
GABA b receptors (reviewed in Bowery, 1993).
Extrasynaptic GABA b receptor location (Otis & Mody, 1992) and a slow binding
on-rate of synaptic GABA b receptors (compared toGABA a receptors, Fischer & Pamas,
1996a; b) have also been considered as possible explanations for the differential
activation of GABA receptor subtypes as a function of stimulation frequency (Fischer &
Pamas, 1996b). An extrasynaptic location of GABA b receptors would require a greater
amount of GABA release to reach the receptors (Otis & Mody, 1992) and an increased
amount of GABA could be achieved by repetitive stimulation (Baxter & Bittner, 1991).
A slower on-rate of synaptic GABA b receptors would require a longer window of
opportunity for binding, which could also be provided by repetitive stimulation (Fischer
& Pamas, 1996b). These findings indicate that stimulation parameters play a role, if not
determine, the type of presynaptic inhibition that is evoked.
Results from a study in which the effects of GABA a and GABA b receptor agonists
and antagonists on presynaptic inhibition were investigated indicated minimal activation
of GABA b receptors (Stuart & Redman, 1992). From their results, the authors
concluded that GABA a receptors are the chief mediators o f Ia-a-motoneuron
presynaptic inhibition and suggested that GABA b receptors are located
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 110
extrasynaptically. However, in that study, presynaptic inhibition was evoked by
heterosynaptic conditioning stimulation, which is strictly associated with GABA a
receptors, thus GABA b receptor involvement in presynaptic inhibition is not precluded.
On the contrary, presynaptic inhibition has been shown to increase following
administration of a GABA b receptor agonist: baclofen reduced monosynaptic EPSP
amplitudes (Lev-Tov et al., 1988) by reducing Ca2+ influx, which reduces the probability
of transmitter release from la afferent terminals (Curtis et al., 1997). Conversely, a
decrease in presynaptic inhibition was found following administration of a GABA b
receptor antagonist, CGP-35348 (Curtis, 1998). Further, GABAb, and not GABAa,
receptors have been associated with segmentally-evoked presynaptic inhibition of
monosynaptic reflexes (Curtis & Lacey, 1994).
Changes in H-Reflex Modulation Following Spinal Cord Injury
Alterations in electrophysiological reflex patterns of rate modulation provide further
evidence that GAB Aergic presynaptic inhibition is involved in homonymous rate-
sensitive depression. Loss of supraspinal control, dysfunctional intemeurons and
subsequent reorganization of spinal circuits that occur in response to injury are proposed
to result in the loss of GABA receptor-related H-reflex modulation. Damage to the
spinal cord affects both short- and long-lasting depression and both types of H-reflex
inhibition are reduced in spastic patients (as a result of spinal cord injury, Calancie et
al., 1993; Nielsen et al., 1993; multiple sclerosis, Crone & Nielsen, 1994; Nielsen et al.,
1995; and spinal spasticity of mixed etiology, Delwaide, 1985).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. I ll
Rate-related reductions in H-reflex depression have also been reported in
experimental spinal cord injury such as contusion (Thompson et al., 1992) and
transection (Skinner et al., 1996) injury in the rat, and these rate-dependent alterations in
H-reflexes have been linked to functions of GABAergic intemeurons. Thompson et al.,
(1996) demonstrated that administration of a GABA b receptor antagonist (CGP-35348)
in normal, intact rats resulted in frequency-dependent alterations in H-reflex amplitudes
such that responses in intact rats treated with the antagonist resembled those of
untreated spinally-injured rats. Further, findings indicating that rate-dependent
modulation of Ia EPSPs is mediated via GABA b receptors in normals (Curtis, 1998;
Lev-Tov et al., 1988; Peshori et al., 1998) contribute to the suggestion that the
decreased rate-sensitive depression associated with spinal cord injury is also related to
reduced GABA b receptor-mediated presynaptic inhibition (Thompson et al., 1996).
The present results demonstrate that a selective NMDA receptor antagonist
effectively increased the reduced rate-sensitive depression of the monosynaptic H-reflex
in chronic spinal rats. Because NMDA receptors are located on intemeurons and
NMDA receptor agonists and antagonists have been excluded from acting within the
monosynaptic pathway, the present findings may provide indirect evidence for an
extrinsically, possibly GABA b receptor-mediated rate-sensitive depression, as discussed
below.
NMDA and non-NMDA receptor antagonists. Given the general EAA receptor-
reflex mediation hypothesis, if rate-sensitive depression is due to a purely intrinsic
mechanism, then theoretically, an NMDA receptor antagonist, whose binding is
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 112
restricted to receptors on (extrinsic) intemeurons, should have no effect on the
depression. If an NMDA receptor antagonist altered rate-sensitivity of the
monosynaptic reflex, then a role for intemeurons in H-reflex modulation could be
considered. An alteration in rate-sensitive depression produced by a non-NMDA
receptor antagonist, however, would not likely provide support for an intrinsic
mechanism because, despite the homosynaptic association, non-NMDA receptors are
known to act at the postsynaptic a-motoneuron, which has no proposed involvement in
homo- or heterosynaptic inhibition as described above.
In this study, CPP, an NMDA receptor antagonist, selectively altered the rate-
modulatory component of the monosynaptic H-reflex in the spastic condition. The
evidence provided by these results indicates that: 1) the monosynaptic reflex is subject
to polysynaptic mediation; and 2) rate-sensitive depression due to repetitive
homonymous Ia fiber activation is not solely the result of an intrinsic mechanism.
Antispastic Agents: Clinical Therapeutic Considerations
It was demonstrated in chronic spinal rats in the present study and also in intact,
anesthetized mice in a previous report (Turski et al., 1992) that NBQX depresses the H-
reflex, a reflex known to be exaggerated in the spastic patient. Although at face value
this appears to be good evidence of NBQX’s potential efficacy, it is important to
consider the conditions, such as rate of stimulation, under which the results were
obtained. The increase in percent of control rate under NBQX is illustrated in Fig. 4.7,
and suggests that the thug might actually worsen spasticity. NBQX seemed to have a
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 113
negative effect on H-reflex modulation in the chronic spinal rat by decreasing the
already reduced rate-sensitive depression.
It is important to note that spastic patients are remarkably different from normal
subjects and likewise, that chronic spinal animals are remarkably different from intact
animal s. An important consideration in determining the efficacy of an antispastic agent
is that it be able to effectively reduce the symptoms that are present In a study of motor
unit behavior patterns during muscle spasms in spastic humans, it was found that mean
peak spasm firing rates were approximately 18 Hz for rate-modulated units (Thomas &
Ross, 1997). This is considerably higher than the 0.3 Hz control rate used to determine
the efficacy of EAA antagonists in the intact animal model (Turski et al., 1992).
In conclusion, it is suggested that, in preclinical evaluation of a potential antispastic
drug for its use in clincial regimens, that the more crucial factor to be investigated is the
agent’s ability to selectively reduce frequencies higher than 0.3 Hz. Further, with
respect to the pharmacotherapeutic efficacy of EAA receptor antagonists on spastic
symptoms, it is suggested that selective NMDA receptor antagonists may have a higher
potential success rate in treating spasticity, for two main reasons. The NMDA receptor
antagonist used in the present study, CPP, was able to: 1) reduce H-reflex response
amplitudes at frequencies above the control rate, and 2) increase rate-sensitive
depression; both of which are key factors in the spastic patient.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 5. CONCLUSIONS
The purpose o f the studies carried out in the preceding chapters was twofold: 1) To
establish and validate the chronic spinal rat as an experimental model of spasticity; and 2)
to evaluate the efficacy of two EAA antagonists as potential antispastic agents.
The first experiment measured H-reflexes in isoflurane-anesthetized rats in the intact
condition and then 28 and 60 days after spinal thoracic transection. Under isofiurane
anesthesia, development of two characteristics of the spastic syndrome were observed 60
days post-transection: hyperreflexia and reduced rate-sensitive depression. However,
because previous reports have shown that signs of spasticity are apparent by one month
post-injury (Skinner et al., 1996; Thompson et al., 1992) and because the predicted
decrease in reflex threshold subsequent to spinalization (Katz & Rymer, 1989; Levin &
Hui-Chan, 1993; Powers et al., 1988; Taylor et al., 1984) did not occur under isofiurane,
the second set of experiments (Chapter 3) were conducted.
Experiments in Chapter 3 tested the same parameters of the H-reflex in intact and
28- and 60-day spinal rats under ketamine anesthesia. Under ketamine, development of
spasticity was evident by 28 days post-transection. Chronic spinal rats exhibited
decreased reflex thresholds, increased H-wave amplitudes and decreased rate-sensitive
depression. H-reflexes were also measured in a group o f unanesthetized 28- and 60-day
chronic spinal rats. Comparison of the results between groups revealed a selective effect
of ketamine on rate-sensitive depression at 5.0 Hz. The effect, however, was somewhat
paradoxical in nature in terms of ketamine’s action at NMDA receptors, but
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 115
could perhaps be explained by its action on Renshaw cells. Nevertheless, the combined
findings point to a polysynaptic influence on the monosynaptic reflex, at least with
respect to low frequency, or rate-sensitive, depression.
The fourth chapter was devoted to investigation of EAA antagonist actions on
parameters of the H-reflex and the flexor reflex. Results from this study found that
NBQX, a non-NMDA antagonist, does not appear to be very effective against spasticity
due to its lack of alteration of H-wave amplitudes elicited at higher frequencies.
However, the results showed that CPP, an NMDA antagonist, demonstrated potential
clinical efficacy, and it is suggested that CPP be tested at higher systemic concentrations
in subsequent studies. Neither antagonist effectively reduced the flexor amplitude.
In conclusion, the experiments performed in this dissertation have validated the
chronic spinal rat as a stable model of spasticity for use in evaluation of potential
antispastic agents. It was found that both anesthesias and EAA antagonists affected the
rate-modulatory component of the monosynaptic H-reflex, which serves to reinforce two
points in particular: 1) use of anesthesia in testing reflex regulation is not without
consequence and in fact, will likely confound measurement; and 2) the monosynaptic
reflex pathway is subject to polysynaptic influence. Finally, the presentfindings
indicated a potential for use of selective NMDA receptor antagonists as antispastic
agents because of their ability to modify altered rate-modulatory mechanisms of the
exaggerated H-reflex in spastics. It would be interestng and valuable to perform an
additional set of experiments using the same parameters, i.e., reflex threshold,
magnitude, and rate-sensitive depression, to evaluate the action of an inhibition-
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 116
enhancing drug such as baclofen. This could contribute to the determination of more
specific underlying mechanisms responsible for alterations that result in spasticity and
as well, to finding the best therapeutic agent.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. REFERENCES
Aanonsen, L.M. and Wilcox, GX. (1989). Muscimol, y-aminobutyric acid receptors and excitatory amino acids in the mouse spinal cord.Journal o f Pharmacology and Experimental Therapeutics, 248, 1034-1038.
Abbott, R. (1992). Complications with selective posterior rhizotomy. Pediatric Neurosurgery, 18, 42-47.
Allen, A.R. (1911). Surgery of experimental lesion of spinal cord equivalent to crush injury of fracture dislocation of spinal column: A preliminary report.Journal o f the American Medical Association, 57, 878-880.
Angel, R.W. and Hoffmann, W.W. (1963). The H-reflex in normal, spastic, and rigid subjects. Archives o f Neurology,591-596. 8 ,
Ashby, P., Verrier, M. and Lightfoot, E. (1974). Segmental reflex pathways in spinal shock and spinal spasticity in man.Journal o f Neurology, Neurosurgery and Psychiatry, 37, 1352-1360.
Azouvi, P., Roby-Brami, A., Biraben, A., Thiebaut, J.B., Thurel, C. and Bussel, B. (1993). Effect of intrathecal baclofen on the monosynaptic reflex in humans: evidence for a postsynaptic action.Journal o fNeurology, Neurosurgery and Psychiatry, 515- 56, 519.
Baxter, D.A. and Bittner, G.D. (1991). Synaptic plasticity at the crayfish neuromuscular junction: presynaptic inhibition. Synapse, 7,244-251.
Bernstein, J.J., Wells, M.R., Bernstein, M.E. (1978). Spinal cord regeneration: synaptic renewal and neurochemistry. In C.W. Cotman (Ed.), Neuronal Plasticity (pp. 49-71). New York: Raven.
Bimbaum, A. and Ashby, P. (1982). Postsynaptic potentials in individual soleus motoneurons in man produced by Achilles tendon taps and electrical stimulation of tibial nerve. Electroencephalography and Clinical Neurophysiology,469-471. 54,
Blake, J.F., Brown, M.W. and Collingidge, G.L. (1988). CNQX blocks acidic amino acid induced depolarizations and synaptic components mediated by non-NMDA receptors in rat hippocampal slices.Neuroscience Letters, 89, 182-186.
Blight, A.R. and DeCriscito, V. (1986). Morphometric analysis of experimental spinal cord injury in the cat: the relation of injury intensity to survival of myelinated axons. Neuroscience, 19, 321-341.
117
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 118
Bormann, J. (1988). Electrophysiology of GABAa and GABAb receptor subtypes. Trends in Neuroscience, 11, 112-116.
Brown, L.T. (1971). Projections and terminations of the corticospinal tract in rodents. Experimental Brain Research, 13,432-450.
Brugger, F., Wicki, U., Nassenstein-Elton, D., Fagg, G.E., Olpe, H.-R. and Pozza, M.F. (1990). Modulation of the NMDA receptor by D-serine in the cortex and the spinal cord, in vitro. European Journal o f Pharmacology,29-38. 191,
Burke, D., Gandevia, C. and McKeon, B. (1984). Monosynaptic and oligosynaptic contributions to human ankle jerk and H-reflex. Journal o fNeurophysiology, 435- 52, 448.
Calancie, B., Broton, J.G., Klose, K.J., Traad, M., Difini, J., Ayyar, D.R. (1993). Evidence that alterations in presynaptic inhibition contribute to segmental hypo- and hyperexcitability after spinal cord injury in man. Electroencephalography and Clinical Neurophysiology, 177-186.89,
Charlton, G., Brennan, M.J.W. and Ford, D.M. (1981). Alterations in neurotransmitter receptor binding in the spinal cord after transection. Brain Research, 218, 372-275.
Chen, D.F., Bianchetti, M., and Wiesendanger, M. (1986). Modulation of flexor reflexes in rats by a noradrenergic agonist and antagonists. Neuroscience Letters, 26 (Suppl). SI62.
Chen, X.Y. and Wo:paw, J.R. (1995). Operant conditioning of H-reflex in freely moving rats. Journal o f Neurophysiology, (1), 411-415. 73
Clemente, C.D. (1978). Neurophysiologic mechanisms and neuroanatomic substrates related to spasticity. Neurology, 28(9), 40-45.
Cohen, R.W., Fisher, R.S., Duong, T., Handley, V.W., Campagnoni, A.T., Hull, C.D., Buchwald, N.A. and Levine, M.S. (1991). Altered excitatory amino acid function and morphology of the cerebellum of the spastic Han-Wistar rat. Molecular Brain Research, 11, 27-36.
Collingridge, G.L. and Lester, R.A.J. (1989). Excitatory amino acid receptors in the vertebrate central nervous system. Pharmacological Reviews, 40 (2), 143-210.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 119
Collins, G.G.S. (1980). Release of endogenous amino acid neurotransmitter candidates from rat olfactory cortex: possible regulatory mechanisms and the effects of pentobarbitome. Brain R esearch 790,517-528.
Cope, T.C., Hickman, K.R. and Botterman, B.R. (1988). Acute effects of spinal transection on EPSP’s produced by single homonymous Ia-fibers in soleus alpha motoneurons. Journal o fNeurophysiology, 1678-1694. 60,
Cotman, C.W., Managhan, D.T., Ottersen, O.P. and Storm-Mathisn, J. (1987). Anatomical organization of excitatory amino acid receptors and their pathways.Trends in Neuroscience, 7,273-280.
Crockett, D.P., Harris, S.L. and Eggar, M.D. (1987). Plantar motoneuron columns in the rat. Journal o f Comparative Neurology, 109-118.265,
Crone, C. and Nielsen, J. (1994). Central control of disynaptic reciprocal inhibition in humans. Acta Physioliogica Scandinavica, 152, 351-363.
Crone, C. and Nielsen, J. (1989). Methodological implications of the post-activation depression of the soleus H-reflex in man.Experimental Brain Research, 78,398-410.
Curtis, D.R. (1998). Two types of inhibition in the spinal cord. In P. Rudomin, R. Romo and L.M. Mendell (Eds.). Presynaptic Inhibition and Neural Control (pp. ISO- 161). New York: Oxford Univ. Press.
Curtis, D.R., Duggan, A.W., Felix, D., Johnston, G.A.R. (1970). GAB A, bicuculline and central inhibition. Nature, 226,1222-1224.
Curtis, D.R. and Eccles, J.C. (1960). Synaptic action during and after repetitive stimulation. Journal o f Physiology (London), 374-398. 150,
Curtis, D.R., Gynther, B.D., Lacey, G. and Beattie, D.T. (1997). Baclofen reduction of presynaptic calcium influx in the cat spinal cordin vivo. Experimental Brain Research 113, 520-533.
Curtis, D.R. and Lacey, G. (1994). GABAb receptor-mediated spinal inhibition. Neuroreport, 540-542.5,
Curtis, D.R., Phillis, J.W., and Watkins, J.C. (1960). The chemical excitation of spinal neurones by certain acidic amino acids. Journal o fNeurophysiology (London), 150, 656-682.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 120
Davidoff, R.A. (1985). Antispasticity drugs: Mechanisms of action. Annals o f Neurology, 17, 107-116.
Davidoff, R.A. and Hackmann, J.C. (1988). Anti-spastic drugs and excitatory amino acids. In E.A. Cavalheiro, J. Lehmann and L.Turski (Eds.). Frontiers in Excitatory Amino A cid Research (pp. 343-350). New York: Alan R. Liss, Inc.
Davidoff, R.A., Graham, L.T., Jr., Shank, R.P., Werman, R. and Aprison, M.H. (1967). Changes in amino acid concentrations associated with loss of spinal intemeurons. Journal o fNeurochemistry, 1025-1031.14,
Davies, J. (1988). A reappraisal of the role of NMDA and non-NMDA receptors in neurotransmission in the cat dorsal horn. In E.A. Cavalheiro, J. Lehmann and L. Turski (Eds.). Frontiers in Excitatory Amino Acid Research (pp. 355-362). New York: Alan R. Liss, Inc.
Davies, J., Evans, R.H., Herrling, P.L., Jones, A.W., Olverman, H.J., Pook, P. and Watkins, J.C. (1986). CPP, a new potent and selective NMDA antagonist. Depression of central neuron responses, affinity for [3H]d-AP5 binding sites on brain membranes and anticonvulsant activity. Brain Research, 382, 169-173.
Davies, J. and Watkins, J.C. (1985). Depressant action of r-D-glutamylaminomethyl sulphonate (GAMS) on amino acid-induced and synaptic excitation in the cat spinal cord. Brain Research, 327, 113-120.
Davies, J. and Watkins, J.C. (1983). Role of excitatory amino acid receptors in mono- and polysynaptic excitation in the cat spinal cord.Experimental Brain Research, 49, 280-290.
Davies, J. and Watkins, J.C. (1979). Selective antagonism of amino acid induced and synaptic excitation in the cat spinal cord. Journal o f Physiology, 621-635.297,
DeLa Torre, J.C. (1984). Spinal cord injury models.Progress in Neurobiology, 22, 289-344.
Delwaide, P.J. (1987). Spasticity: from pathophysiology to therapy.Acta Neurochir. (Suppl. 39, Wien), 91-95.
Delwaide, P.J. (1985). Electrophysiological analysis of the mode of action of muscle relaxants in spasticity.Annals o f Neurology, 90-95. 17,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 121
Delwaide, P.J. (1973). Human monosynaptic refexes and presynaptic inhibition: an interpretation of spastic hyperreflexia. In J.E. Desmedt. (Ed).New Developments in Electromyography and Clinical Neurophysiology (pp. 508-522). Basel: Karger.
Delwaide, P.J., Pennisi, G. (1994). Tizanidine and electrophysiologic analysis of spinal control mechanisms inhumans with spasticity. Neurology, 44(Suppl.9), 21-28
Eccles, J.C., Eccles, R.M., and Magni, F. (1961). Central inhibitory action attributable to presynaptic depolarization produced by muscle afferent volleys. Journal o f Physiology (London), 147-166. 159,
Eccles, J.C. and Rail, W. (1951). Effects induced in a monosynaptic reflex by its activation. Journal o f Neurophysiology, 353-376. 14,
Eccles, J.C., Schmidt, R.F. and Willis, W.D. (1963). The mode of operation of the synaptic mechanism producing presynaptic inhibition.Journal o f Neurophysiology, 26, 523-538.
Eichenberger, A. and Ruegg, D.G. (1984). Relation between the specific H reflex facilitation preceding a voluntary movement and movement parameters in man. Journal o f Physiology, 347, 545-559.
Fagg, G.E. and Matus, A. (1984). Selective association of N-methyl-D-aspartate and quisqualate types of L-glutamate receptor with brain postsynaptic densities.Proceedings o f the National Academy o f Sciences, USA, 81,6876-6880.
Fagg, G.E., Neuhoff, V., Sontag, K.-H. and Osbome, N.N. (1978). Myelin composition in the central nervous system of mutant Han-Wistar rats. Neuroscience Letters, 9, 201-205.
Faist, M., Mazevet, D., Dietz, V. and Pierrot-Deseilligny, E. (1994). A quantitative assessment of presynaptic inhibition of la afferents in spastics. Differences in hemiplegics and paraplegics.Brain, 117,1449-1455.
Farkas, S. and Ono, H. (1995). Participation of NMDA and non-NMDA excitatory amino acid receptors in the mediation of spinal reflex potentials in rats: anin vivo study. British Journal o f Pharmacology, 114,1193-1205.
Faist, M., Mazevet, D, Dietz, V. and Pierrot-Deseilligny, E. (1994). A quantitative assessment of presynaptic inhibition of la afferents in spastics. Differences in hemiplegics and paraplegics.Brain, 117,1449-1455.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 122
Farkas, S. and Ono, H. (1995). Participation of NMDA and non-NMDA excitatory amino acid receptors in the mediation of spinal reflex potentials in rats: anin vivo study. British Journal o f Pharmacology, 1193-1205.114,
Fasano, V.A., Barolat-Romana, G., Zeme, S. and Sguazzi, A. (1979). Electrophysiological assessment of spinal circuits in spasticity by direct dorsal root stimulation. Neurosurgery, ¥(2), 146-151.
Fetz, E.E., Jankowska, E., Johannisson, T., and Lipski, J. (1979). Autogenetic inhibition of motoneurones by impulses in group la muscle spindle aferent.Journal o f Physiology (London), 173-195.293,
Fischer, Y. and Pamas, I. (1996a). Activation of GABAb receptors at individual releasing boutons of the crayfish opener neuromuscular junction produced presynaptic inhibition. Journal o f Neurophysiology, 1377-1385. 75,
Fischer, Y. and Pamas, I. (1996b). Differential activation of two distinct mechanisms for presynaptic inhibition by a single inhibitory axon. Journal o f Neurophysiology, 76 3807-3816. (6),
Fletcher, E.J., Martin, D., Aram, J.A., Lodge, D. and Honore, T. (1988). Quinoxalinediones selectively block quisqualate and kainate receptors and synaptic events in rat neocortex and hippocampus and frog spinal cordin vitro. British Journal o f Pharmacology, 95, 585-597.
Foster, A.C. and Fagg, G.E. (1987). Taking apart NMDA receptors. Nature, 329, 395-396.
Fulton, B.P., Miledi, R. and Takahashi, T. (1980). Electrical synapses between motoneurons in the spinal cord of the newborn rat. Proceedings o f the Royal Society, Series B 208, 1115-1120.
Funase, K., Higashi, T., Yoshimura, T., Imanaka, K. and Nishihira, Y. (1996). Evident difference in the excitability of the motoneuron pool between normal subjects and patients with spasticity assessed by a new method using H-reflex and M-response. Neuroscience Letters, 203,127-130.
Gage, P.W. and Robertson, B. (1985). Prolongation of inhibitory postsynaptic currents by pentobarbitome, halothane and ketamine in CA1 pyramidal cells in rat hippocampus.British Journal o fPharmacology, 675-681. 85,
Gelfan, S. (1966). Altered spinal motoneurons in dogs with experimental hindlimb rigidity. Journal o f Neurophysiology, 583-611. 29,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 123
Gozariu, M., Roth, V., Keime, F., Le Bars, D. and Wilier, J.-C. (1998). An electrophysiological investigation into the monosynaptic H-reflex in the ratBrain Research, 782, 343-347.
Greenamyre, J.T., Young, A.B., and Penney, J.B. (1984). Quantitative autoradiographic distribution of L-[3H]glutamate-binding sites in rat central nervous system. Journal o f Neuroscience, 2133-2144. 4,
Guttman, L. (1970). Spinal shock and reflex behaviour in man. Paraplegia, 8 ,100- 113.
Hill, D.R. and Bowery, N.G. (1981). [3H]Baclofen and [3H]GABA bind to bicuculline-insensitive GABAb sites in rat brain. Nature, 290,149-152.
Hoffman, P. (1918). Uber die beziehungen die sehnenreflexezur zur willkurlichen bewegung und zum tonus. Z Biologie, 68, 351-370.
Honey, C.R., Miljkovic, Z., MacDonald, J.F. (1985). Ketamine and phencyclidine cause a voltage-dependent block of responses to L-aspartic acid. Neuroscience Letters, 61, 135-139.
Honore, T., Davies, S., Drejer, J., Fletcher, E.J., Jacobsen, P., Lodge, D. and Nielson, F.E. (1988). Quinoxalinediones: potent competitive non-NMDA glutamate receptor antagonists. Science, 241, 701-703.
Hultbom, H., Illert, M., Nielsen, J., Paul, A., Ballegaard, M., Wiese, H. (1996). On the mechanism of the post-activation depression of the H-reflex in human subjects. Experimental Brain Research, 108, 450-462.
Hultbom, H. and Malmsten, J. (1983a). Changes in segmental reflexes following chronic spinal cord hemisection in the cat. I. Increased monosynaptic and polysynaptic ventral root discharges. Acta Physiologica Scartdinavica, 119, 405-422.
Hultbom, H. and Malmsten, J. (1983b). Changes in segmental reflexes following chronic spinal cord hemisection in the cat. II. Conditioned monosynaptic reflexes.Acta Physiologica Scartdinavica, 119,423-433.
Ikonomidou-Turski, C., Schwarz, M., Turski, L. and Sontag, K-H. (1986). Muscle relaxant action of phenobarbitone in genetically spastic rats: an electromyographic study. European Journal o f Pharmacology, 1-7. 128,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 124
Ikonomidou, C., Turski, L., Klockgether, T., Schwarz, M. and Sontag, K.-H. (1985). Effects of methyl P-carboline-3-carboxy late, RO 15-1788 and CGS 8216 on muscle tone in genetically spastic rats. European Journal Pharmacology, 205-213. 113,
lies, J.F. (1996). Evidence for cutaneous and corticospinal modulation of presynaptic inhibition of la afferents from the human lower limb. Journal o fPhysiology, 4 9 4197-207. ,
lies, J.F. and Pisini, J.V. (1992). Cortical modulation of transmission in spinal reflex pathways of man.Journal o fPhysiology, 425-466.455,
Jankowska, E., McCrea, D., and Mackel, R. (1981). Oligosynaptic excitation of motneurones by impulses in group la muscle spindle afferent in the catJournal o f Physiology (London), 41-425.316,
Jefferson, A. A. and Schlapp, W. (1953). Some effects of repetitive stimulation of afferents on reflex conduction. In The Spinal Cord (pp. 99-119). Chiba Foundation Symposium, Churchill, London.
Jessell, T.M., Yoshioka, K. and Jahr, C.E. (1986). Amino acid reeptor-mediated transmission at primary afferent synapses in rat spinal cord.Journal o f Experimental Biology, 124, 239-258.
Johnson, J.W. and Ascher, P. (1987). Glycine potentiates the NMDA response in cultured mouse brain neurons. Nature, 325, 529-531.
Kagamihara, Y., Hayashi, A., Okuma, Y., Nagaoka, M. Nakajima, Y. and Tanaka, R. (1998). Reassessment of H-reflex recovery curve using the double stimulation procedure. Muscle and Nerve, 21, 352-360.
Kaizawa, J. and Takahashi, I. (1970). The low frequency depression of rat monosynaptic reflex. Tohoku Journal o fExperimental Medicine, 383-394. 100,
Katz, R.T. and Rymer, W.Z. (1989). Spastic hypertonia: Mechanisms and measurement Archives o f Physical and Medical Rehabilitation, Vol. 70,144-155.
King, A.E., Lopez-Garcia, J.A., and Cumberpatch, M. (1992). Antagonism of synaptic potentials in ventral hom neurones by 6-cyano-7-nitroquinoxaline-2,3-dione: a study in the rat spinal cord in vitro. British Journal o f Pharmacology, 375-381.107,
Klockgether, T., Schwarz, M., Wullner, Turski, L. and Sontag, K.-H. (1989). Myorelaxant effect after intrathecal injection of antispastic drugs in rats.Neuroscience Letters, 97, 221-226.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 125
Kohn, A.F., Floeter, M., Hallett, M. (1997). Presynaptic inhibition compared with homosynaptic depression as an explanation for soleus H-reflex depression inhumans. Experimental Brain Research, 116 (2), 375-380.
Kramarevsky, N.Y., Anderson, N.L., Irish, P.S., and Westram, L.E. (1997). Electron microscopy of synaptic remodeling in rat spinal trigeminal nucleus following injury to trigeminal ganglia. Society fo r Neuroscience Abstracts, 23, 443.
Krogsgaard-Larsen, P., Honore, T., Hansen, J.J., Curtis, D.R. and Lodge, D. (1980). New class of glutamate agonist structurally related to ibotenic acid. Nature, 284,64-66.
Kullmann, D.M., Martin, R.L. and Redman, S.J. (1989). Reduction by general anesthetics of group la excitatory postsynaptic potentials and currents in the cat spinal cord. Journal o fNeurophysiology, 412,277-296.
Kuno, M. (1964). Mechanism of facilitation and depression of the excitatory synaptic potential in spinal motoneurons.Journal o f Neurophysiology (London), 175, 100- 112.
Lehmann,J., Schneider, J, McPherson, S., Murphy, D.E., Bernard, P., Tsai, C., Bennett, D.A., Pastor, G., Steel, D.J., Boehm, C., Cheney, D.L., Liebman, J.M., Williams, M., and Wood, P.L. (1987). CPP, a selective NMDA-type receptor antagonist: Characterization in vitro and in vivo. Journal o f Pharmacology and Experimental Therapeutics, 240 (3), 737-746.
Leis, A.A., Kronenberg, M.F., StStkarova, I., Paske, W.C. and Stokic, D.S. (1996). Spinal motoneuron excitability after acute spinal cord injury in humans. Neurology, 47, 231-237.
Levin, M.F. and Hui-Chan, C. (1993). Are H and stretch reflexes in hemiparesis reproducible and correlated with spasticity? Journal o fNeurology, 63-71.240,
Lev Tov, A., Meyers, D.E.R., and Burke, R. (1988). Activation of type B y- aminobutyric acid receptors in the intact mammalian spinal cord mimics the effects of reduced presynaptic Ca2+ influx.Proceedings o f the National Academy o f Sciences, USA, 85, 5330-5334.
Levy, R.A. (1977). The role of GAB A in primary afferent depolarization. Progress in Neurobiology, 211-267. 9,
Lin, Q., Peng, Y.B. and Willis, W.D. (1996). Role of GAB A receptor subtypes in inhibition of primate spinothalamic tract neurons: Difference between spinal and periaqueductal gray inhibition. Journal o f Neurophysiology, 75 (1), 109-123.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 126
Little, J.W. and Halar, EM . (1985). H-reflex changes following spinal cord injury. Archives o fPhysical and Medical Rehabilitation, 66, 19-22.
Lloyd, D.P. (1957). Monosynaptic reflex response of individual motoneurons as a function of frequency. Journal o f General Physiology, 435-450. 40,
Lloyd, D.P.C. and Wilson, V.J. (1957). Reflex depresion in rhythmically active monosynaptic reflex pathways.Journal o f General Physiology,409-426. 40,
Lodge, D. and Anis, N.A. (1984). Effects of ketamine and three other anesthetics on spinal reflexes and inhibitions in the cat. British Journal o f Anaesthesiology, 1143- 56, 1151.
Lodge, D., Headley, P.M. and Curtis, D.R. (1978). Selective antagonism by D-a- aminoadipate of amino acid and synaptic excitation of cat spinal neuron.Brain Research, 152,603-608.
Lodge, D. and Johnson, K.M. (1990). Noncompetitive excitatory amino acid receptor antagonists. Trends in Pharmacological Sciences, 81-86.II,
Long, S.K., Smith, D.A.S., Siarey, R.J. and Evans, R.H. (1990). Effect of 6-cyano- 2,3-dihydroxy-7-nitro-quinoxaline (CNQX) on dorsal root-, NMDA-, kainate- and quisqualate-mediated depolarization of rat motoneurones in vitro. British Journal of Pharmacology, 100, 850-854.
Lundberg, A. (1964). Supraspinal control of transmission in reflex paths to motoneurons and primary aferents. In J.C. Eccles and J.P. Schade (Eds.), Progress in Brain Research. Physiology o f Spinal Neurons, Vol. (pp. 12, 197-219). Amsterdam: Elsevier.
Magdalery, J.W., Porter, W.E., Park, A.M. and Teasdall, R.D. (1951). Electrophysiological studies of nerve and reflex activity in normal man: two-neurone reflex and identification of certain action potentials from spinal roots and cord.Johns Hopkins Hospital Bulletin, 499-519.88,
Magnusson, T. (1973). Effect of chronic transection on dopamine, noradrenaline, and 5-hydroxytryptamine in the rat spinal cord.Naunyn-Schmiedeberg’s Archives o f Pharmacology, 278,13-22.
Malcangio, M. and Bowery, N.G. (1996). GAB A and its receptors in the spinal cord. Trends in Pharmacological Sciences,7,457-462. 1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 127
Marshall, B.E. and Longnecker, D.E. (1996) General anesthetics. In A. Goodman Gilman, T.W. Rail, A.S. Nies and P. Taylor (Eds.). Goodman and Gilman’s The Pharmacological Basis o f Therapeutics, 9th (pp. ed. 307-347). Elmsford, New York: Pergamon Press.
Matthews, G., Ayoub, G.S. and Heidelberger, R. (1994). Presynaptic inhibition by GABA is mediated via two distinct GABA receptors with novel pharmacology. Journal o fNeuroscience, 14(3), 1079-1090.
Matthews, W.B. (1966). Ratio of maximum H reflex to maximum M response as a measure of spasticity. Journal o f Neurology, Neurosurgery and Psychiatry,201- 29, 205.
Mazzocchio, R. and Rossi, A. (1997). Involvement of spinal recurrent inhibition in spasticity. Further insight into the regulation of Renshaw cell activity. Brain, 120,991- 1003.
McCouch, G.P., Austin, G.M. and Liu, C.Y. (1958). Sprouting as a cause of spasticity.Journal o f Neurophysiology, 205-216. 21,
Meinck, H.M. (1976). Occurrence of the H-reflex and the F-wave in the rat. Electroencephalography and Clinical Neurophysiology, 530-533. 41,
Misgeld, U., Bijak, M. and Jarolimek, W. (1995). A physiological role for GABA b receptors and the effects of baclofen in the mammalian central nervous system. Progress in Neurobiology, 423-462. 46,
Mody, I., Tanelian, D.L. and Maclver, M.B. (1991). Halothane enhances tonic neuronal inhibition by elevating intracellular calcium. Brain Research, 538, 319-323.
Morota, N., Abbott, R., Kofler, M., Epstein, F.J. and Cohen, H. (1995). Residual spasticity after selective posterior rhizotomy. C hild’s Nervous System, 11, 161-165.
Munson, J.B., Foehring, R.C., Lofton, S.A., Zengel, J.E., and Sypert, G.W. (1986). Plasticity of medial gastrocnemius motor units following cordotomy in the cat. Journal o fNeurophysiology, 619-634. 55,
Murray, M. and Goldberger, M.E. (1997). Strategies and mechanisms of recovery after spinal cord injury. In F.J. Seil (Ed.).Advances in Neurology, 72, Neuronal regeneration, reorganization and repair (pp.219-225). Philadelphia: Lippincott-Raven.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 128
Murray, M. and Goldberger, M.E. (1974). Restitution of function and collateral sprouting in the cat spinal cord: the partially hemisected animal.Journal o f Comparative Neurology, 158, 19-36.
Nakanishi, S. (1992). Molecular diversity of glutamate receptors and implications for brain function. Science, 258,597-603.
Nelson, S.G., Collatos, T.C., Niechaj, A. and Mendell, L.M. (1979). Immediate increase in Ia-motoneuron synaptic transmission caudal to spinal cord transection. Journal o f Neurophysiology, 655-664. 42,
Nicoll, RA. and Alger, B.E. (1979). Presynaptic inhibition: Transmitter and ionic m echanisms.International Review o f Neurobiology,217-258. 21,
Nielsen, J., Petersen, N., Ballegaard, M., Biering-Sorensen, F., Kiehn, O. (1993). H- reflexes are less depressed following muscle stretch in spastic spinal cord injured patients than in healthy subjects. Experimental Brain Research, 97,173-176.
Nielsen, J., Petersen, N. and Crone, C. (1995). Changes in transmission across synapses of la afferents in spastic patients.Brain, 118, 995-1004.
Noble, L.T. and Wrathall, J.R. (1989). Correlative analysis of lesion development and functional status after graded spinal cord contusion injuries in the rat. Experimental Neurology, 103,34-40.
Noble, L.T. and Wrathall, J.R. (1985). Spinal cord contusion in the rat: morphometric analyses of alterations in the spinal cord.Experimental Neurology, 88, 135-149.
Nygren, L.G., Fuxe, K., Jonsson, G. and Olson, L. (1974). Functional regeneration of 5-hydroxytryptamine nerve terminals in rat spinal cord following 5,6- dihydroxytryptamine induced degeneration. Brain Research, 78,377-394.
Otis, T.S. and Mody, I. (1992). Differential activation of GABAa and GABAb receptors by spontaneously released transmitter. Journal o fNeurophysiology, 67, 221- 235.
Ozawa, S., Kamiya, H. and Tsuzuki, K. (1998). Glutamate receptors in the mammalian central nervous system. Progress in Neurobiology, (5), 54 581-618.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 129
Parise, M., Garcia-Larrea, L., Mertens, P., Sindou, M. and Mauguiere, F. (1997). Clinical use of polysynaptic flexion reflexes in the management of spasticity with intrathecal baclofen. Electroencephalography and Clinical Neurophysiology, M I 105, MS.
Peshori, K.R., Collins, W F. and Mendell, LM . (1998). EPSP amplitude modulation at the rat la-alpha motoneuron synapse: Effects of GABAb receptor agonists and antagonists. Journal o f Neurophysiology, 181-189. 79,
Peters, S., Koh, J. and Choi, D.W. (1987). Zinc selectively blocks the action of N- methyl-D-aspartate on cortical neurons. Science, 236, 589.
Pittermann, W., Sontag, K.-H., Wand, P., Rapp, K. and Deerberg, F. (1976). Spontaneous occurrence of spastic paresis in Han-Wistar rats.Neuroscience Letters, 2, 45-49.
Porter, R.H.P. and Greenamyre, J.T. (1994). Regional variations in the pharmacology of AMPA receptors as revealed by receptor autoradiography.Brain Research, 553,202-206.
Powers, R.K., Marder-Meyer, J. and Rymer, W.Z. (1988). Quantitative relations between hypertonia and stretch reflex threshold in spastic hemiparesis.Annals o f Neurology, 23, 115-124.
Raigorodsky, G. and Urea,G. (1990). Spinal antinociceptive effects of excitatory amino acid antagonists: quisqualate modulates the action o f N-methyl-D-aspartate. European Journal o f Pharmacology, 182,37-47.
Raisman, G. (1977). Formation of synapses in adult rat after injury: similarities and differences between peripheral and central nervous sites. Philosophical Transactions o f the Royal Society o f London (Biology), 349-359. 278,
Rice, G.P.A. (1987). Pharmacotherapy of spasticity: Some theoretical and practical considerations. Canadian Journal o f Neurological Science, 510-512. 14,
Roby-Brami, A. and Bussel, B. (1992). Inhibitory effects on flexor reflexes in patients with a complete spinal cord lesion.Experimental Brain Research, 90,201-208.
Roby-Brami, A. and Bussel, B. (1990). Effects o f flexor reflex afferent stimulation on the soleus H-reflex in patients with a complete spinal cord lesion: evidence for presynaptic inhibition of la transmission. Experimental Brain Research, 81, 593-601.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 130
Rothwell, J.C., Day, B.L., Berardelli, A., Marsden, C J>. (1986). Habituation and conditioning of the human long latency stretch reflex. Experimental Brain Research, 63, 197-204.
Rudomin, P. (1990). Presynaptic inhibition of muscle spindle and tendon organ afferents in the mammalianspinal cord. Trends in Neuroscience, 13 (12), 499-505.
Rudomin, P., Jimenez, I., Solodkin, M. and Duenas, S. (1983). Sites of action of segmental and descending control of transmission on pathways mediating PAD of Ia- and lb-afferent fibers in cat spinal cord. Journal o fNeurophysiology, 50743-769. (4),
Schwarz, M., Block, F. and Sontag, K.-H. (1992). N-methyl-D-aspartate (NMDA)- mediated muscle relaxant action of memantine in rats. Neuroscience Letters, 143, 105- 109.
Schwarz, M., Klockgether, T., Turski, K. and Sontag, K.-H. (1988a). Intrathecal injection of antispastic drugs in rats: muscle relaxant action of midazolam, baclofen, 2- aminophosphonoheptanoic acid (AP7) and tizanidine. In H. Muller, J. Zierski and R.D. Penn (Eds.) Local-Spirtal Therapy o fSpasticity (pp. 65-79). Berlin: Springer.
Schwarz, M., Klockgether, T., Wullner, U., Turski, L. and Sontag, K.-H. (1988b). 5-Aminovaleric acid antagonizes the pharmacological actions of baclofen in the central nervous system. Experimental Brain Research, 70, 618-626.
Schwarz, M, Loscher, W., Turski, L. and Sontag, K.-H. (1985). Disturbed GABAergic transmission in mutant Han-Wistar rats: further evidence for basal ganglia dysfunction. Brain Research, 347, 258-267.
Sharpless, S.K. (1975). Supersensitivity-like phenomena in central nervous system. Federation Proceedings, 34, 1990-1997.
Sheardown, M.J., Nielsen, E.O., Hansen, A.J., Jacobsen, P. and Honore, T. (1990). 2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline: a neuroprotectant for cerebral ischemia. Science, 247, 571-574.
Shefner, J.M., Berman, S.A., Sarkarati, M. (1992). Recurrent inhibition is increased in patients with spinal cord injury.Neurology, 42,2162-2168.
Shimoji, K., Fujiwara, N., Fukuda, S., Denda, S., Takada, T. and Maruyama, Y. (1990). Effects of isoflurane on spinal inhibitory potentials.Anesthesiology, 72 (5), 851-857.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 131
Siarey, R.J., Long, S.K. and Evans, R.H. (1991). Potentiation of synaptic reflexes by D-serine in the rat spinal cord in vitro. European Journal o f Pharmacology,241- 195, 244.
Simpson, R.D., Gondo, M., Robertson, C.S., and Goodman, J.C. (1995). The influence of glycine and related compounds on spinal cord injury-induced spasticity. Neurochemistry Research, 20 (10), 1203-1210.
Skinner, R.D., Houle, J.D., Reese, N.B., Berry, C.L. and Garcia-Rill, E. (1996). Effects of exercise and fetal spinal cord implants on the H-reflex in chronically spinalized adult rats. Brain Research, 729,127-131.
Smith, C., Bimbaum, G., Carter, J.L., Greenstein, J., Lublin, F.D. and the US Tizanidine Study Group. (1994). Tizanidine treatment of spasticity caused by multiple sclerosis: Results of a double-blind, placebo-controlled trial. Neurology, 44(Suppl. 9), S34-S43.
Sommer, B. and Seeburg, P.H. (1992). Glutamate receptor channels: novel properties and new clones.Trends in Pharmacological Sciences, 291-296.. 13,
Stein, R.B. (1995). Presynaptic inhibition in humans. Progress in Neurobiology, 47, 533-544.
Steward, O., Cotman, C.W., Lynch, G.S. (1974). Growth of new fiber projection in brain of adult rats: re-innervation of dentate gyrus by contralateral entorhinal cortex folowing ipsilateral entorhinal lesions. Experimental Brain Research, 20,45-66.
Stuart, G.J. and Redman, S.J. (1992). The role of GABAa and GABAb receptors in presynaptic inhibition of la EPSPs in cat spinal motoneurones.Journal o f Physiology, 4 4 7675-692. ,
Taborikova, H. and Sax, D.S. (1968). Motoneurone pool and the H-reflex. Journal o f Neurology, Neurosurgery and Psychiatry, 354-361. 31,
Tang, A.H., and Schroeder, L.A. (1973). Spinal-cord depressant effects of ketamine and etoxadrol in the cat and the rat. Anesthesiology, 39,37-43.
Tamawa, I., Farkas, S., Berzsenyi, P., Pataki, A., and Andrasi, F. (1989). Electrophysiological studies with a 2,3-benzodiazepine muscle relaxant: GYKI 52466. European Journal o fPharmacology, 193-199. 167,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 132
Taylor, S., Ashby, P. and Verrier, M. (1984). Neurophysiological changes following traumatic spinal lesions in man.Journal o f Neurology, Neurosurgery and Psychiatry, 47, 1102-1108.
Thomas, C.K. and Ross, B.H. (1997). Distinct patterns of motor unit behavior during muscle spasms in spinal cord injured subjects.Journal o f Neurophysiology, 77, 2847-2850.
Thompson, F.J., Parmer, R. and Reier, PJ. (1996). Spinal cord injury alters rate- modulation of lumbar reflexes: GABAb treatment reproduces rate-depression but not PTP changes. Society for Neuroscience Abstracts, 22, 2155.
Thompson, F.J., Reier, P.J., Lucas, C.C. and Parmer, R. (1992). Altered patterns of reflex excitability subsequent to contusion injury of the rat spinal cord. Journal o f Neurophysiology, (5), 68 1673-1686.
Thompson, S.W.N. and Wall, P.D. (1996). The effect of GABA and 5-HT receptor antagonists on rat dorsal root potentials. Neuroscience Letters, 217, 153-156.
Turski, L., Bressler, K., Klockgether, T. and Stephens, D.N. (1990). Differential effects of the excitatory amino acid antagonists, 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) and 3-((±)-2-carboxypiperazin-4-yl)-propyl-l-phosphonic acid (CPP) on spinal reflex activity in mice. Neuroscience Letters, 113, 66-71.
Turski, L., Jacobsen, P., Honore, T. and Stephens, D.N. (1992). Relief of experimental spasticity and anxiolytic/anticonvulsant actions of the alpha-amino-3- hydroxy-5-methyl-4-isoxazolepropionate antagonist 2,3-dihydroxy-6-nitro-7-sulfamoyl- benzo(F)quinoxaline. Journal o fPharmacology and Experimental Therapeutics, 260 (8), 742-747.
Turski, L., Klockgether, T., Schwarz, M., Sontag, K.-H. and Meldrum, B.S. (1987a). Unusual interactions of excitatory amino acid receptor agonists: a - andP- kainate antagonize motor responses to N-methyl-D-aspartate in rodents.Neuroscience, 20, 2285.
Turski, L., Klockgether, T., Sontag, K.-H., Herrling, P.L. and Watkins, J.C. (1987b). Muscle relaxant and anticonvulsant activity of 3-((±)-2-carboxypiperazin-4-yl)- propyl-l-phosphonic acid, a novel N-methyl-D-aspartate antagonist, in rodents. Neuroscience Letters, 73, 143-148.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 133
Turski, L., Klockgether, T., Turski, W A , Ikonomidou, C., Schwarz, M. and Sontag, K.-H. (1988). Muscle relaxant action of excitatory amino acid antagonists: sites of action. In E A . Cavalheiro, J. Lehmann and L. Turski (Eds.), Frontiers in Excitatory Amino Acid Research (pp. 343-350). New York: Alan R. Liss.
Turski, L., Schwarz, M. and Sontag, K.-H. (1984). Stereotypy, locomotor and cataleptic effects produced by drugs influencing dopaminergic systems in a mutant strain of Wistar rats: a genuine model of basal ganglia dysfunction? Behavioral Brain Research, 12, 29-37.
Turski, L., Schwarz, M., Turski, WA., Klockgether, T., Sontag, K.-H. and Collins, J.F. (1985a). Muscle relaxant action of excitatory amino acid antagonists. Neuroscience Letters, 53, 321-326.
Turski, L., Schwarz, M., Turski, WA. and Sontag, K.-H. (1985b). Muscle relaxant action of 2-chloroadenosine in genetically spastic rats is independent of y-aminobutyric acid-mediated inhibition. Neuroscience Letters, 54, 369-374.
Van Boxtel, A. (1986). Differential effects of low-frequency depression, vibration- induced inhibition and posttetanic potentiation on H-reflexes and tendon jerks in the human soleus muscle. Journal o fNeurophysiology, (3), 55551-568.
Voigt, M. and Sinkjaer, T. (1998). The H-reflex in the passive human soleus muscle is modulated faster than predicted from post-activation depression.Brain Research, 783,332-346.
Wagemann, E., Schmidt-Kastner, R., Block, F. and Sontag, K.-H. (1995). Altered pattern of immunohistochemical staining for glial fibrillary acidic protein (GFAP) in the forebrain and cerebellum of the mutant spastic rat. Journal o f Chemical Neuroanatomy, 8, 151-163.
Wagemann, E., Schmidt-Kastner, R., Block, F. and Sontag, K.-H. (1991). Neuronal degeneration in hippocampus and cerebellum of mutant spastic Han-Wistar rats. Neuroscience Letters, 121, 102-106.
Watkins, J.C. (1984). Excitatory amino acids and central synaptic transmission. Trends in Pharmacological Sciences, 373-376.
Watkins, J.C. and Evans, R.H. (1981). Excitatory amino acid transmitters, Annual Review o f Pharmacology and Toxicology, 165-204. 21,
Weingarten, S.M. (1972). Dissociation of limbic and neocortical EEG patterns in cats under ketamine anaesthesia. Journal o fNeurosurgery, 429-433. 72,
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 134
West, J.R., Deadwyler, S., Cotman, C.W., Lynch, G. (1975). Time-dependent changes in commissural field potentials in dentate gyrus following lesions of entorhinal cortex in adult rats. Brain Research, 97,215-133.
Wiesendanger, M. (1991). Neurophysiologicai bases of spasticity. In Sindou, M., Abbott, R. and Keravel, Y. (Eds.), Neurosurgeryfor Spasticity (pp. 15-19). Germany: Springer-Verlag.
Wisden, W. and Seeburg, P. (1993). Mammalianionotropic glutamate receptors. Current Opinion in Neurobiology, 291-298, 3,
Wolpaw, J.R. (1994). Aquisition and maintenance of the simplest motor skill: investigation of CNS mechanisms. Medical Science o f Sports and Exercise, 26 (2), 1475-1479.
Wood, S.A., Gregory, J.E. and Proske, U. (1996). The influence of muscle spindle discharge on the human H-reflex and the monosynaptic reflex in the cat Journal o f Physiology, 497(I), 279-290.
Wrathall, J.R., Pettegrew, R.K. and Harvey, F. (1985). Spinal cord contusion in the rat: production of graded, reproducible injury groups.Experimental Neurology, 81 OS- 8 , 122.
Yang, J.F., Fung, J., Edamura, M., Blunt, M., Stein, R.B. and Barbeau, H. (1991). H-reflex modulation during walking in spastic paretic subjects. Canadian Journal o f Neurological Sciences, 18,443-352.
Young, R.R. (1994). Spasticity: a review. Neurology, 44(Suppl. 9), S12-S20.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA
Heather Mosser graduated from California State University, Long Beach in 1993
with a Bachelor of Arts degree in Psychology. She began post-baccalaureate studies
upon graduation, doing research in a physiological psychology laboratory. There, her
research was focused on the investigation of actions of morphine and related opioid
peptides with respect to pain, physical dependence and tolerance. She obtained a Master
of Arts degree in Psychology from California State University, Long Beach in 1995. In
the fall of 1995, she began studying at Louisiana State University. Her research focus
there was on pharmacologic treatment strategies for pain and spasticity secondary to
spinal cord injury. During her graduate studies, Heather developed a keen interest in
spinal cord injury research. She will pursue her interests in this area beginning with a
postdoctoral position at the University of Miami School of Medicine, where she will do
research with The Miami Project to Cure Paralysis.
135
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. DOCTORAL EXAMINATION AND DISSERTATION REPORT
Candidate: Heather C. Mosser
Major Field: Psychology
Title of Dissertation: - c c c . Effects of Anesthesia and Potential Antispastic Agents in an Experimental Model of Spasticity
Appr oved:
Major Professor and chairman
aduate School
EXAMINING COMMITTEE:
Date of gxanination:
October 12, 1998
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. IMAGE EVALUATION TEST TARGET (Q A -3 ) ✓ &
✓
\
ITf c ,
. 4
150mm
IIVU^GE. Inc 1653 East Main Street Rochester. NY 14609 USA Phone: 716/462-0300 Fax: 716/286-5989
0 1993. Applied Image. Inc.. All Rights Reserved
permission of the copyright owner. Further reproduction prohibited without permission.