Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Respiratory neuron activity in the urethane anaethetized ground squirrel : responses to changes in respiratory… Franks, Sarah Kate 1998

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

Item Metadata

Download

Media
831-ubc_1999-0348.pdf [ 4.56MB ]
Metadata
JSON: 831-1.0089060.json
JSON-LD: 831-1.0089060-ld.json
RDF/XML (Pretty): 831-1.0089060-rdf.xml
RDF/JSON: 831-1.0089060-rdf.json
Turtle: 831-1.0089060-turtle.txt
N-Triples: 831-1.0089060-rdf-ntriples.txt
Original Record: 831-1.0089060-source.json
Full Text
831-1.0089060-fulltext.txt
Citation
831-1.0089060.ris

Full Text

RESPIRATORY NEURON ACTIVITY IN THE URETHANE ANAESTHETIZED GROUND SQUIRREL: RESPONSES TO CHANGES IN RESPIRATORY DRIVE by SARAH KATE FRANKS B.Sc, The University of British Columbia, 1994 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES DEPARTMENT OF ZOOLOGY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA December, 1998 © Sarah Kate Franks, 1998 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. 1 further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) ii ABSTRACT: These experiments examined the effects of arousal state and respiratory drive on the activity of central respiratory neurons in urethane anaesthetized ground squirrels. EEG recordings were used to score arousal state. Three arousal states were identified, state I (wake-like), state II (drowsy) and state III (slow wave sleep-like). Extracellular recordings were made from 21 neurons with respiratory related discharge in the ventrolateral medulla. Six different cell types were identified, these included inspiratory continuous cells, inspiratory augementing tonic cells, inspiratory augmenting cells, late inspiratory cells, expiratory inspiratory phase spanning cells and expiratory decrementing cells. Increases in neuron activity (specifically peak neuron activity) were observed during state I compared to state III. This increased activity was observed in all cell types during state I. These results are analogous to those observed in unanaesthetized animals (Orem, 1977) in wake compared to slow wave sleep, suggesting that urethane does not interfere with the changes in the respiratory related activity of these cells associated with changes in cortical activation state. Increases in peak neuron activity (spikes per second) were observed in respiratory related neurons in the present study upon exposure to hypercapnia. These results are also consistent with those of other studies (Batsel, 1966, Nesland ef al., 1966, St. John, 1981, Foldgering and Smolders, 1979, St. John, 1977, St. John and Bianchi, 1985) examining the effects of hypercapnia on respiratory related activity of neurons in the medulla of unanaesthetized animals. This suggests that urethane also has little effect on the response of respiratory neurons to hypercapnia associated with the increased respiratory drive under hypercapnic conditions, in the golden-mantled ground squirrel. The present study demonstrated an increase in peak neuron activity in response to hypoxia. However, there was a large range of variation between cells with some exhibiting increases whilst others exhibited decreases in activity. Similarly, previous studies in unanaesthetized animals have also produced inconsistent results in response to hypoxic stimuli despite the fact hypoxia too increased respiratory drive in these animals. The eta squared (r|2) value is a statistic that quantifies the strength and consistency of the discharge of respiratory related cells compared to the oscillations of the respiratory cycle. In the present study the rf varied from one cell to the next regardless of the cell type examined. Arousal state had variable influences on the rf value of each cell, some cells exhibited increases in this statistic in a particular state whereas others decreased their r\2 value in that same state. These results suggest that each of the cells recorded from had levels of afferent input that varied from one state to the next. iv Upon exposure to hypercapnia or hypoxia there was a general trend towards a decrease in the r\ value of the respiratory neurons. These results were contrary to predictions of a relatively immutable r)2 value. This suggests that r\2 is an endogenous property of each cell and that any variation in input, including respiratory tonic input, introduces variation in the cells output and alters the coupling of the discharge to the respiratory cycle. V TABLE OF CONTENTS page Abstract ii-iv Table of contents v List of Tables vi List of Figures vii-x Acknowledgements xi Introduction 1 Materials and Methods 29 Results 47 Discussion 93 General Conclusions 117 Literature Cited 120 LIST OF TABLES TABLE 1: Discharge per breath of each respiratory cell during state I and state III on air and upon exposure to hypoxic and hypercapnic stimuli. TABLE 2: Peak neuron activity (spikes per second) of each respiratory cell during state I and state III on air and upon exposure to hypoxic and hypercapnic stimuli. TABLE 3: Eta squared r\2 values of every respiratory cell in response to different arousal states on air as well as upon exposure to hypoxic and hypercapnic stimuli. V l l LIST OF FIGURES: Page FIGURE 1: Schematic illustration of the different possible discharge patterns of respiratory related neurons. 5 FIGURE 2: An illustration of the mammalian brainstem, in cross section as well as in longitudinal section. 7 FIGURE 3: Recordings taken from respiratory related units in the brainstem of the golden-mantled ground squirrel. 10 FIGURE 4: EEG recordings taken from unanaesthetized golden mantel ground squirrels and urethane anaesthetized ground squirrels. 25 FIGURE 5: Recordings of discharge from both inspiratory and expiratory units in the ventral respiratory group. 40 FIGURE 6: Post-stimulus time histograms of discharge versus time for the six different cell types recorded from in the current investigation. 42 FIGURE 7: Horizontal section through the medulla of the rat. The figure is divided in two parts, one half showing the distribution of expiratory and inspiratory cells, the other showing all cells recorded from. FIGURE 8: The effects of arousal state on average neuron activity (spikes per breath). FIGURE 9: The effects of arousal state on unit discharge per breath for each class of cell. FIGURE 10: The effects of arousal state on peak neuron activity (spikes per second). FIGURE 11: The effects of arousal state on peak neuron activity (spikes per second) for each class of cell. FIGURE 12: The effects of hypercapnia and hypoxia on unit discharge per breath. FIGURE 13: The effects of hypoxia on unit discharge per breath for each class of cell. FIGURE 14: The effects of hypercapnia and hypoxia on peak neuron activity (spikes per second). FIGURE 15: The effects of hypoxia on peak neuron activity (spikes per second) for each class of cell. FIGURE 16: The effects of hypercapnia on unit discharge per breath for each class of cells. FIGURE 17: The effects of hypercapnia on peak neuron activity (spikes per second) for each class of cell. FIGURE 18: The effects of arousal state on eta squared values. FIGURE 19: The effects of arousal state on eta squared value for each class of cell. X FIGURE 20: The effects of hypercapnia and hypoxia on eta squared values. 85 FIGURE 21: The effects of hypoxia on eta squared values for each class of cell. 87 FIGURE 22: The effects of hypercapnia on eta squared values for each class of cells. 89 X I ACKNOWLEDGEMENTS The easiest part of this thesis is thanking the people who spent so much time and energy supporting me and contributing to the work in the following pages. Many thanks to Mike who spent hours away from his own research to help me make sense of the equipment in the lab and the results in this research. Thanks also to Bill who has been very helpful in guiding and motivating me, and who is probably one of the most patient supervisors in the faculty (it's only been four and a half years since I started this thesis)! Manda was able to keep me sane throughout the long process and helped me to see the light at the end of the tunnel even when she was hundreds of miles away in Texas. Finally, my parents were strong motivators, helping to remind me of my masters when I found myself with time which I would otherwise spend procrastinating. On the opposite end of the spectrum Ken was always around to take me on climbing trips and adventures abroad. Both my family and Ken were very supportive and I am truly grateful. 1 INTRODUCTION: The rhythmical movement of air into and out of the lungs allows for the uptake of oxygen and the elimination of carbon dioxide, thereby meeting the metabolic demands of the organism, and maintaining acid-base homeostasis. This respiration is the product of many complex interactions within the central nervous system. Afferent sensory information from mechanoreceptors and chemoreceptors is integrated within the brainstem and acts to modulate respiration. In addition, various other non-respiratory stimuli such as arousal state and behaviour also act on respiratory "centers" within the brainstem to modulate breathing. The mechanisms by which all of these stimuli act to modulate respiration, as well as the site in the brain which ultimately controls respiration has been the source of a great deal of research over the past one hundred years. Of interest to the present investigation is the effect of respiratory stimuli, such as hypoxia and hypercapnia, as well as non-respiratory stimuli, such as arousal state, on the activity of neurons with respiratory modulated activity (respiratory neurons) in the medulla of the urethane anaesthetized golden-mantled ground squirrel. Types of respiratory related neurons: Based on extracellular recordings, respiratory neurons can be classified according to their discharge pattern, the phase of the respiratory cycle in which they fire and the point at which they fire during each phase (Batsel, 1965; Bianchi et al., 1995; Bertrand and Hugelin, 1971; Cohen 1968, 1979; St. John and Wang, 1977; Feldman, 1976, 1986; Richter et al, 1975; Richter, 1982; Euler, 1973, 1986, Duffin et al., 1995). It should be 2 noted that wide disparity exists in the nomenclature used to classify each neuron and therefore some neuron types have been given more than one name. Until recently, control of the respiratory rhythm was believed to be best described by a bistable oscillator model, that is to say respiration was only considered to have two phases, inspiration and expiration. More recently, a triphasic oscillator model has been proposed which divides respiration into three phases, inspiration, stage 1 expiration, El (also called post-inspiration), and stage 2 expiration, E2 (Schwarzacher et al., 1991; Richter, 1982; Feldman, 1986; Duffin et al., 1995; Bianchi et al., 1995). Consequently, three broad categories of respiratory related neurons arise, inspiratory (I), post-inspiratory (PI or El), and expiratory (E or E2). These neurons can then be further subdivided according to the specific portion of the respiratory phase in which they fire maximally, whether it be early in the phase, late in the phase or throughout the phase. In addition to classifying neurons according to the portion of the respiratory cycle in which they fire, they can also be classified according to the type of discharge pattern they exhibit, whether it be augmenting, decrementing, or a continuous pattern of discharge (Figure 1). The resulting cell types include inspiratory decrementing neurons (also called early inspiratory neurons) which have a peak discharge in early inspiration; inspiratory augmenting neurons with a peak discharge near the end of inspiration, expiratory augmenting neurons with a peak discharge near the end of stage 2 expiration, and expiratory decrementing neurons with a peak discharge at the onset of stage 1 expiration (also called early E or post-inspiratory neurons). Inspiratory continuous cells have a constant rate of discharge throughout inspiration and expiratory continuous cells have a constant rate of discharge throughout expiration. All of the aforementioned cells have a 3 phasic pattern of discharge, there are also tonic inspiratory cells and tonic expiratory cells which fire constantly throughout the respiratory cycle but their discharge increases during one of the phases of the respiratory cycle (Figure 1). In addition, there are four types of cells which fire at the transition between phases in the respiratory cycle, these include the late inspiratory cells which have a peak firing rate late in inspiration and continue to fire during the initial portion of stage 1 expiration (IE phase spanning cell), the EI phase spanning cell (pre-inspiratory neuron) which has a peak discharge rate at the transition from expiration to inspiration and the tonically active phase spanning cells which fire constantly throughout respiration, but have a peak discharge rate at the transition from one phase to the next. Most respiratory cells are concentrated in, but not restricted to, the medulla (Figure 2); for example, the phase spanning cells (both phasic and tonic) are concentrated in regions rostral to the medulla, including the pons, midbrain, hypothalamus and thalamus (Bianchi et al., 1995). In addition, some respiratory related neurons have been found within the cortex. Due to this differential distribution of neurons throughout the medulla as well as the rest of the brain, neurons are also classified on the basis of their location and axonal projections. If their axons project and terminate in the spinal cord they are considered bulbospinal, conversely if their axons project and terminate in the region of the medulla they are considered propriobulbar. Neurons are only classified on the basis of location if there are populations of neurons in different locations which have similar discharge characteristics, for example expiratory augmenting neurons are found in both the caudal ventral respiratory group (cVRG) as well as the Botzinger Complex of the rostral ventral respiratory group (rVRG) and therefore the distinction is made between 4 these two populations based on location calling those found in the Botzinger Complex, E -Augmenting Botzinger neurons (see next section for discussion of respiratory groups) (Figure 2). Respiratory neurons with the same discharge pattern defined by the qualitative methods mentioned above, don't necessarily have identical discharge profiles, for example, one neuron might fire at a faster rate or have a different consistency of discharge; consequently, quantitative methods of classification have also been developed to distinguish these differences. The ANOVA has proven to be a powerful statistic in determining the respiratory modulation of neurons (Orem and Dick, 1983). This statistical test can then be taken one step further by using it to calculate the variance ratio, called the eta squared (r| ) statistic (Orem, 1994). The discharge profile of each respiratory neuron lies on a continuum of respiratory related activity, with the activity of some neurons more tightly coupled to the respiratory cycle than others, revalues of neurons can range from 0, in which the neuron's firing pattern is unrelated to the respiratory cycle, to a value of 1, when the cell's firing pattern is completely coupled to the respiratory cycle (Figure 3)(See Orem and Trotter, 1994 for derivation of n2; Orem et al., 1985; Orem and Dick, 1983). According to Orem (1985) there is a bimodal distribution of respiratory related neurons with respect to r\2 values. Respiratory related neurons are either tightly coupled to breathing and have r\ values above 0.30 or they are weakly coupled to breathing and have n values below 0.20. Orem suggested that a possible explanation for this bimodal distribution of neurons is the presence of two physiologically separate groups of respiratory neurons within the medulla. F I G U R E 1: Schematic illustration of the different possible discharge patterns of respiratory related neurons. 6 I 1 :2 CO c c cu E CO CCS o CO 'CL cn c CO c CO c d) o •em nui] o c 0) o o > o o » *-* CO CO 'a a CO CO c c DO c CD E CO o » to "a x UJ CO o c (0 • (—; us c us 'a. a> o X E CO 0) crei ntin g V o nnin >> £ CO o c tory de o o > o tory toi ratory ton se spai inspirator nspirat spanni ira CO ratory ton ha inspirator Tonic i phase Exp Expi Insp Expi a • Late Tonic i phase k. o 15 "a CO c J>> CO cu 0 5 .2 co c co O I- Q. 7 F I G U R E 2: A n illustration of the mammalian brainstem in longitudinal section with two specific levels also shown in cross section. Areas included in this diagram are the Botzinger (Bot-C) and pre-Botzinger (pre-Bot) complexes, the ventral respiratory group ( V R G ) , the dorsal respiratory group (DRG) , the nucleus ambiguus (nA), nucleus retroambiguus (nRA), the rostral ventrolateral medulla ( R V L M ) , pyramids (py), the hypoglossal nucleus (nXII), the spinal segment of the trigeminal nucleus, glossopharyngeal nerve (IX), vagus nerve (X), cervical group ( C 1 - C 2 ) , nucleus tractus solitarius (nTS), facial nucleus (nVII), pontine respiratory group (PRG) (Figure taken from Duffin et al., 1995; NIPS). 8 He postulated that high r\2 value neurons function to integrate respiratory related stimuli and produce the respiratory rhythm whereas the low r\2 value neurons function to impose behavioral influences on the respiratory system (Orem, 1985). Distribution of neurons within the medulla: The neurons within the central nervous system responsible for generating the respiratory rhythm are a diverse group of cells located primarily within the medulla oblongata (Figure 2) (J. Feldman, 1986; St. John, 1978; Long and Duffin, 1986; M . Cohen, 1979. Bianchi et al., 1995; Euler, C. von, 1986). These neurons are clustered into two main groups, the dorsal respiratory group (DRG) and the ventral respiratory group ( V R G ) (Figure 2) (J. Feldman, 1986; St. John, 1978; Long and Duffin, 1986; M . Cohen, 1979; Bianchi et al., 1995; Euler, C. von, 1986; Ezure, K . et al., 1988; Smith et a l , 1991). Within these two groups are several different types of respiratory related neurons. The ventral respiratory group ( V R G ) : A l l of the cells previously described are found within the brainstem, and for the most part, within the ventrolateral medulla. 10 F I G U R E 3: This figure illustrates recordings taken from respiratory related units in the brainstem of the golden-mantled ground squirrel. The lower trace is unit discharge and the upper trace is respiration. Coupling of unit discharge to respiration is quantified using 2 2 the r| statistic. Cells with a wide range of r\ values are illustrated. *Inspiration is a downswing of the pen Tf=0.85 12 The ventral respiratory group is located in the ventrolateral medulla and consists of a longitudinal column of respiratory related neurons bound laterally by the spinal trigeminal tract, caudally by the spinal cord and rostrally by the facial nucleus (Figure 2)(Bianchi et al., 1995; Cohen, 1979; Feldman, 1986; Duffin et al., 1995; von Euler, 1986; Batsel, 1965; Kal ia , 1981; Haber et al., 1957). Division of the V R G into three parts is based on the types of neurons found within each segment, the resulting regions include the rostral V R G ( rVRG) , the intermediate V R G ( i V R G ) and the caudal V R G (cVRG) (Figure 2) (Bianchi et al., 1995; von Euler, 1986). The caudal V R G extends from the level of the spinal cord rostrally to the level of the obex, a region which is also called the nucleus retroambigualis (von Euler, 1986; Mer i l l , 1970). Neurons within the c V R G are primarily bulbospinal expiratory neurons, which have an augmenting, late peak discharge pattern (otherwise called E - A U G or E-late neurons). In the cat the E - A U G neurons act as premotor neurons, providing monosynaptic excitatory input to the internal intercostal and abdominal respiratory motor neurons (Meri l l , 1981). Similar projections also exist in the rat; however, the axons from these neurons give off axon collaterals at the level of the medulla which are thought to inhibit medullary inspiratory neurons during expiration (Bianchi et al., 1995; Feldman, 1986). Sources of inhibition to the E - A U G neurons of the c V R G include the E - D E C (post-I) neurons, which fire during stage 1 of expiration. The reciprocal inhibition which exists between the E -A U G neurons and the E - D E C neurons is thought to shape the augmenting pattern of discharge observed in the E - A U G neuron (Bianchi et al., 1995). Other sources of 13 inhibitory input to the E - A U G neuron during inspiration come from the I - A U G bulbospinal neurons as well as the I -DEC neurons (Bianchi et al., 1995). Moving rostrally from the c V R G , the intermediate V R G extends from the obex to the level of the retrofacial nucleus. The i V R G includes part of the nucleus ambiguus (NA) , which functions as the cranial motor nucleus for cranial nerves 9 and 10, and the nucleus paraambigualis (NPA) which contains primarily inspiratory neurons of both the bulbospinal and propriobulbar type (von Euler, 1986). Motor neurons within the N A , in particular the pharyngeal and laryngeal motor neurons, have a variety of discharge patterns (Cohen, 1979; von Euler, 1986; Bianchi et al., 1995; Feldman, 1986; Wyke, 1974). Laryngeal motor neurons, which can be either inspiratory related or expiratory related motorneurons, usually have a decrementing discharge pattern or constant firing pattern. Pharyngeal motor neurons, on the other hand, usually fire during expiration in a constant (E-continuous) pattern (Cohen, 1979; Feldman, 1986; von Euler, 1986). Apart from the expiratory motor neurons found within the N A the majority of the neurons found within the i V R G are of the inspiratory type, specifically, bulbospinal I - A U G neurons. I - A U G neurons descend contralateral^ down the spinal cord in cats synapsing on either phrenic motor neurons or external intercostal motor neurons and provide these motor neurons with excitatory input (von Euler, 1986; Feldman, 1986; Bianchi et al., 1995). Species differences arise between cats and rats in this respect due to the fact that the bulbospinal I neurons in the rat descend ipsilaterally and bilaterally. The physiological significance of this difference is unknown (Duffin and van Alphen, 1995; Lipski et al., 1994). During expiration, the I - A U G neurons are actively inhibited by expiratory neurons. The only excitatory input to the I - A U G neurons appears to be in the 14 form of recurrent excitation from other I - A U G neurons during inspiration. Another smaller group of inspiratory related cells is also found within the i V R G , these are the inspiratory decrementing (I-DEC) cells with an early onset of peak discharge(early inspiratory neurons). I -DEC cells of the i V R G are primarily propriobulbar sending axons and axon collaterals to the contralateral c V R G and i V R G , and inhibiting the late I neurons (= IE phase spanning neurons) as well as the E - D E C neurons. Due to the decrementing pattern of discharge of I -DEC neurons as well as their inhibitory connections to the E - D E C and late-I neurons, it has been hypothesized that they form an integral part of the respiratory rhythm generator. Although there are no apparent differences between rats and cats in the i V R G , there are developmental differences in this area. Interestingly, the most rostral portion of the i V R G in neonatal rats, an area called the pre-Botzinger Complex, contains propriobulbar pre-inspiratory neurons which have pacemaker abilities. Consequently, it has been hypothesized that the neurons within this region comprise an essential component of the respiratory rhythm generator (Feldman et al., 1988; Johnson et al., 1994; Onimaru et al., 1995; Smith et al., 1991; Smith et al., 1990) . The corresponding region in the adult rat contains primarily propriobulbar pre-inspiratory neurons, which have extensive connections within the medulla (Smith et al., 1991) ; but which do not appear to possess pacemaker abilities. There have been similar studies using adult cats (Schwarzacher et al., 1991; Schwarzacher et al., 1995; Connelly et al., 1992) that have produced analogous findings, that is, high concentrations of pre-inspiratory propriobulbar neurons within the pre-Botzinger area without pacemaker properties. Given these results the pre-inspiratory neurons are hypothesized to play a role in phase switching from expiration to inspiration (Schwarzacher et al., 1995; Connelly et 1 5 al., 1992; Smith et al., 1990). The presence of I -DEC neurons as well as the pre-I neurons within the i V R G suggests an important role for this area in terms of respiratory control. Immediately anterior to the pre-Botzinger Complex lies the Botzinger Complex, which includes the rostral portion of the N A , and which terminates at the level of retrofacial nucleus (Figure 2). Pharyngeal motorneurons with E - A U G , E - D E C and I-A U G discharge patterns, comprise the nucleus ambiguus portion of the Botzinger Complex (von Euler, 1986, Bianchi et al., 1995). Apart from the pharyngeal motorneurons, the majority of respiratory related neurons within the Botzinger Complex are of the E - A U G type, with axons which project contralaterally to the V R G and the phrenic motor nuclei, acting in an inhibitory fashion on these neurons (Meri l l , 1982; Bryant et al., 1993). Inhibitory connections from these cells have also been found to synapse on inspiratory bulbospinal neurons within the dorsal respiratory group (DRG) (Merrill et al., 1983). The ventral respiratory group appears to be organized similarly within the medulla from one species to the next (Bianchi et al., 1995). In addition to being extensively studied within the cat the V R G has also been studied in the rat, the rabbit, the dog, the pig, the squirrel monkey and the guinea pig (Bianchi et al., 1995). In each of these species it would appear that the types of neurons in the V R G and their location, remain constant. 16 The dorsal respiratory group (DRG): Another area within the medulla containing high concentrations of respiratory related cells is the dorsal respiratory group (DRG) , located in the ventrolateral portion of the N T S (nucleus of the tractus solitarius) (Figure 2) (Bianchi et al., 1995; J. Feldman, 1986; Long and Duffin, 1986). It would appear that the classification system used in the V R G (i.e. inspiratory decrementing, inspiratory augmenting etc . . ) has not been applied to the cells of the D R G , instead a classification unique to this area has been used. The D R G consists primarily of inspiratory respiratory neurons, which can be classified as either inspiratory alpha (la) or inspiratory beta cells (lb), based on their response to lung inflation. Both la and lb neurons are mainly bulbospinal neurons and have been found to innervate phrenic interneurons and motor neurons (Feldman, 1986). A third type of neuron within the D R G called the pump cell, responds directly to changes in lung volume, consequently it wi l l respond at any point in the respiratory phase to inflation (von Euler, 1986). Studies of the D R G , which involved extensive lesioning of this area, have failed to effect the respiratory timing or inspiratory termination (McCrimmon, Speck, and Feldman, 1984), thus questioning the significance of the D R G in respiratory rhythm generation. In addition, considerable variability exists in D R G organization from one species to the next, for instance it has been difficult to demonstrate that a D R G exists in the rat (Zheng et al., 1991). Intracellular techniques have failed to detect inspiratory related neurons in the N T S of the rat (Zheng et al., 1991); however, extracellular recording techniques have succeeded in recording inspiratory activity within this area (Saether et al., 1987; DeCastro et al., 1994). Explanations for these two contradictory 17 findings include the possibility that the extracellular techniques being used record from axons of passage of inspiratory related cells, or more likely, it could be possible that the inspiratory cells within this area of the rat are too small to successfully record from using intracellular techniques (deCastro et al., 1994). Studies examining other species of animals such as the rabbit have confirmed the presence of inspiratory related respiratory neurons within the area of the D R G ; however, these studies used extracellular recording techniques, not intracellular recording techniques, and therefore there is still a considerable amount of uncertainty about the presence of the D R G in species other than the cat (Bianchi et al., 1995). Effects of hypoxia and hypercapnia on neuron activity: The ventilatory response of the Golden-mantled ground squirrel, Spermophilis lateralis, varies with the type of stimulus presented, whether it be hypoxia or hypercapnia. The ventilatory response, which is measured as an increase or decrease in ventilation for a given change in oxygen or carbon dioxide partial pressure, can be divided into two components, the change in tidal volume (Vt) and the change in breathing frequency (f). Ventilation increases in response to hypercapnia and hypoxia in both fossorial (burrow-dwelling animals) and non-fossorial species of mammals and birds (T. Darden, 1972; Boggs, Kilgore, Bircher, 1983; Bismarck, Johansson, and Schemed, 1956); however, the way in which minute ventilation is increased during hypercapnia is somewhat different from the way minute ventilation is increased during hypoxia. Generally, hypercapnia causes increases in tidal volume as well as breathing frequency, 18 whereas hypoxia only causes increases in breathing frequency (Walker, Adams and Voelkel, 1985; Holloway and Heath, 1984; McArthur and Milsom, 1991a). Differential responses to hypoxia as compared to hypercapnia have raised several questions concerning the modulation of the respiratory response to these two stimuli. St John (1981) suggested that this differential response might result from differences in distribution of peripheral and central chemoreceptor afferents among the respiratory neurons of the medulla. Whether the chemical stimuli is acting on two separate respiratory neuron pools (a "carbon dioxide sensitive" pool or an "oxygen sensitive" pool) or that each stimulus is acting at a different level in the respiratory neuron pool (either at the level of the central rhythm generator, the pre-motor neuron pool, or even the motor neuron pool) still remains to be determined. Chemosensitive regions which sense carbon dioxide levels, include the peripheral chemoreceptors which are located in the carotid body (also called Heyman's type receptors) and the intracranial chemoreceptors which are thought to be located on the ventral surface of the medulla (Loeschcke, H.H, 1982; Loeschcke et al., 1970; Loeschcke et al. 1979; Nattie, 1994). The stimulus acting on the central chemoreceptors is most likely hydrogen ions as opposed to carbon dioxide itself (Loeschke, 1979; Loeschcke, 1982; Loeschcke et al., 1979; Bruce and Cherniack, 1987), whereas, peripheral chemoreceptors can be excited by both carbon dioxide as well as hydrogen ions (Bruce and Cherniack, 1987). There has been a great deal of debate concerning the existence of the central chemoreceptors; however, there are several key observations, which support the hypothesis that they exist and are located on the ventral surface of the medulla. Support for the existence of central chemosensitive regions include observations that pH 19 changes on the ventral surface of the medulla result in ventilation changes (Trouth et al., 1982; Loeschke, 1982; Schlaefke, 1981); cooling of this area results in decreased phrenic discharge (Millhorn et al. 1982; Cherniack et al., 1979; Budzinska et al., 1985), and finally, ventilatory responses to carbon dioxide are absent i f this area is lesioned (Schlaefke et al., 1979). The rat and the cat are similar in this respect (Nattie and L i , 1994). In the rat extensive connections have been found between neurons from the ventral medullary surface to neurons in the P R G , spinal cord areas, the Botzinger complex, as well as the N T S (Feldman, 1986). The responses of individual neurons within the brainstem vary from one neuron type to the next, in addition, their responses also vary depending on the level of carbon dioxide to which the animal is exposed. Cohen (1968) classified neurons based on their responses to hypocapnia, these included the type I cell, the type II cell and the type III cell. Type I cells usually consist of cells with phasic activity which, upon exposure to hypocapnic gas mixtures, exhibit a large reduction in activity, or a complete cessation of activity. Type II cells are also phasically active cells; however, upon exposure to hypocapnia they fire at a constant rate throughout all phases of the respiratory cycle. Type III cells are tonically active and typically fire more during one phase of respiration than another but upon exposure to a hypocapnic gas mixture fire constantly throughout all phases of the respiratory cycle. Studies examining the effects of hypercapnia on respiratory related neurons have demonstrated increases in the discharge frequency of the inspiratory and expiratory augmenting cells of the i V R G and c V R G (St. John, 1981; St. John and Wang, 1977; Batsel, 1965). In addition to the cells found in the V R G , inspiratory cells within the D R G also exhibit increases in discharge frequency in response to hypercapnia (St. John, 1981; 20 St. John and Wang, 1977). Not all respiratory related neurons increase their activity in response to hypercapnia, expiratory decrementing neurons for instance have been shown to become depolarized during hypercapnia, and eventually cease firing with prolonged exposure to hypercapnia (Feldman, 1986; Cohen, 1968). Hypoxia, unlike hypercapnia, is only sensed peripherally by sensors in the carotid and aortic bodies. In the rat, the afferent projections carrying information from the carotid and aortic bodies, synapse on cells located in both the dorsomedial as well as the ventrolateral medulla, including the caudal N T S and the area postrema (Finley and Katz, 1992). In the cat some of these projections have been shown to be monosynaptic connections within the D R G (Finley and Katz, 1992). In carotid body denervated cats, hypoxia causes decreased firing in all the respiratory related neurons which were examined (Feldman, 1986; St. John, 1977); however, cats with intact carotid bodies showed inconsistent results, with some respiratory related neurons exhibiting increased unit discharge and others exhibiting decreased unit discharge (St. John, 1979). It has been hypothesized (St. John, 1977; St John, 1981) that the different responses of neurons to either hypercapnia or hypoxia may be due to a heterogeneous distribution of chemoreceptor afferents within the brainstem (St John and Wang, 1977). Effects of arousal state on neuron activity: The transition from wake to sleep elicits changes in respiration, both in terms of tidal volume and in terms of breathing frequency. From wakefulness to the sleep state there is a decrease in several factors associated with ventilation, breathing frequency decreases, 21 the animal's response to hypoxia appears to diminish somewhat, peak instantaneous airflow rate decreases and the overall minute ventilation decreases (Orem, 1978). Certain variables; however, increase during the sleep state, these include an increase in tidal volume as well as an increased responsiveness to hypercapnia (Orem, Netick and Dement, 1977; Orem, 1986). In addition to the changes which occur at the organismal level there also appear to be changes in responsiveness at the level of the respiratory related neurons within the medulla. Most of respiratory related neurons in the medulla respond to changes in arousal state. Puizillout and Ternaux (1974) determined that the majority of the respiratory related neurons from which they recorded in the nucleus ambiguus, solitary tract and hypoglossal nucleus, were affected by arousal state change. Similarly, Orem et al. (1974) described the activity of respiratory related sleep sensitive neurons within the nucleus ambiguus. It was observed that upon transition into the sleep state from wakefulness there was a general decrease in discharge frequency in 79% of the respiratory related neurons from which they recorded, and that this decrease was so profound in some neurons that respiratory related discharge ceased altogether (Orem et al., 1974). In addition, it would appear that the neurons with a low consistency of neuronal discharge relative to the respiratory cycle (low r\ value) were more affected by changes in state than the neurons with high r\2 values (Orem, 1986). Several observations suggest that alterations in metabolic systems alone cannot be used to explain the changes which occur in respiration from the wake state to the sleep state. During sleep there is a relative hypercapnia, and consequently one would expect an increase in ventilation i f alterations in chemical stimuli were the only component 22 involved in the changes in ventilation from sleep to wake (Orem, 1986). In addition, Sullivan et al.,(1978) observed that awake animals continued to breath regardless of whether chemical stimuli were removed. Furthermore, the changes in discharge rate which occur in respiratory related neurons upon transition from wake to sleep occur too rapidly to be explained by alterations in chemical stimuli (Orem, 1986). Due to these observations, as well as other observations (Foutz et al., 1979; Remmers et al., 1976), the wakefulness stimulus has been proposed as the "modulator" responsible for the changes in ventilation which are associated with changes in state. The generation of the wakefulness stimulus is thought to arise from an area between the caudal pons and the posterior hypothalamus (Orem, 1986); however, the specific actions of the wakefulness stimulus on respiratory related neurons within the brainstem remain unknown. Whether the wakefulness stimulus has a direct effect on the respiratory neurons within the brainstem or whether it acts indirectly, initially activating the behavioral systems and subsequently activating the respiratory cells, still remains to be determined (Orem, 1990). Urethane anaesthetic: The use of certain anesthetics in neuron recording studies has been questioned due to their depressant effects on respiratory related neurons. It has been observed that animals anaesthetized with heavy pentobarbital anesthetics have substantially fewer active respiratory related neurons than do lightly anaesthetized preparations (Meri l l , 1970; von Euler, 1986). In addition, certain anesthetics can act selectively by decreasing the activity of some types of neurons. Chloralose, for example, acts specifically on chemosensitive 23 neurons within the medulla without effecting other respiratory related neurons ( Cohen, 1979; von Euler and Soderberg, 1952). Due to the problems associated with the use of certain anesthetics Cohen (1979) suggested two methods which would establish "optimal conditions" for recording from respiratory related cells within the medulla. One of these methods was the use of lightly anaesthetized animals. A relatively unaffected respiratory cycle along with minimal depression of the respiratory centers are among some of the reasons this is the preparation of choice for many researchers (Cohen, 1979). Urethane was first used as an anesthetic in 1855 by Schmiedeberg who described it as a substance which induced a "long-lasting and profound narcosis with little effect on respiration or circulation" (Maggi and M e l i , 1986, part 1). Following Schmiedeberg's initial observations of urethane's effects, a large number of other studies were done to further characterize urethane anesthesia. Urethane has been found to produce long bouts of anesthesia in mammals, reptiles, amphibians and fish (Maggi and M e l i , 1986, part 1). One of the main reasons urethane produces such long lasting effects is the slow rate at which the body metabolizes the drug before it is excreted. Urethane is unique in its ability to produce a surgical plane of anesthesia without interfering with subcortical and peripheral nervous system neurotransmission, and hence many autonomic reflexes are preserved throughout the duration of the anesthesia (Maggi and M e l i , 1986, part 1). The preservation of autonomic reflexes perhaps explains the lack of effect of urethane anesthesia on blood pressure, cardiac output, or respiratory function of rats even when it is given in relatively high doses (1.4 g/kg) IP (Maggi and M e l i , 1986, part 2). Another important characteristic of urethane anesthesia is that anaesthetized animals appear to cycle through similar E E G states to those observed in unanaesthetized animals (Lincoln, 24 1969); that is to say, a urethane anaesthetized animal has a wake-like E E G , a sleep-like E E G as well as other intermediate E E G patterns seen in unanaesthetized animals (Figure 4). These analogous arousal states produced by urethane have been termed state 1, state 2 and state 3 (Grahn et al., 1989). State 1 is analogous to wake in unanaesthetized animals, state 2 is analogous to drowsy, and state 3 is analogous to slow wave sleep (Figure 4). Changes in ventilation also occur when the animal cycles from one arousal state to the next and these also appear to be analogous to those seen in the unanaesthetized animal as they cycle between arousal states (Hunter and Mi lsom, in press, 1998). If the changes in cortical activity patterns and breathing under urethane anaesthesia are both analogous and homologous the urethane preparation would be ideal for sleep related studies which require the immobilization of the animal, such as electrophysiological studies. Another major advantage of urethane anesthesia in studies examining the control of breathing, is that both hypoxic as well as hypercapnic gas mixtures appear to evoke the ventilatory responses one would expect to see in an unanaesthetized animal exposed to the same gas mixtures (Hunter and Mi lsom, in press, 1998). Questions: Urethane anaesthesia has several advantages over other anaesthetics, one advantage is that it allows animals to cycle through arousal states that otherwise would not be present in a heavily anaesthetized animal. Associated with the changes in arousal state observed under urethane, there are also similar changes in ventilation which accompany these changes in arousal state. 25 F I G U R E 4: These recordings show E E G recordings taken from unanaesthetized golden mantel ground squirrels and Urethane anaesthetized ground squirrels. E E G recordings for slow wave sleep, drowsy and awake states can be compared with the analogous state III state II and state I arousal state in the Urethane anaesthetized animal. 2 7 Since E E G profiles under urethane look like slow wave sleep and wakefulness and the breathing changes associated with E E G changes are just like those associated with sleep and wake, one would expect respiratory related neurons to respond just like those in the unanaesthetized cat. This gives rise to hypothesis #1: In urethane anaesthetized squirrels, changes in the activity of respiratory related neurons in response to changes in arousal state, are similar to those seen in unanaesthetized animals. Another advantage of using Urethane anaesthetized animals for respiratory studies is that Urethane anaesthetized animals appear to have the same respiratory responses to hypoxia and hypercapnia that unanaesthetized animals do. Due to these similarities one would expect the responses of central respiratory neurons in urethane anaesthetized animals to be similar to those observed in unanaesthetized animals. This gives rise to hypothesis #2: Changes in activity of respiratory related neurons in response to hypoxia and hypercapnia in the urethane anaesthetized golden-mantled ground squirrel are similar to the changes in activity of respiratory related neurons in the unanaesthetized animal exposed to the same stimuli. Eta squared values are a relatively new statistic for quantifying the degree of coupling between a neuron's discharge and a respiratory event. It has been suggested by Orem (Orem et al, 1983, Orem et al., 1974, Orem et al, 1985) that the r | 2 value is an immutable value of each cell and reflects the respiratory and non respiratory afferent input each cell receives. This gives rise to hypothesis #3: n 2 values wi l l remain constant from wake to sleep in the Urethane anaesthetized golden-mantled ground squirrel. 28 Eta squared values have not been used to describe the effects of changing respiratory drive through hypoxia or hypercapnia in any state. Based on the notion that the r\2 for each cell is immutable we would expect no change in r) 2 value during hypercapnia or hypoxia.This gives rise to hypothesis #4: During exposure to hypercapnic or hypoxic stimuli respiratory neurons wi l l exhibit relatively unchanged r | 2 values. 29 M A T E R I A L S A N D M E T H O D S : Animals used in the current study were golden-mantled ground squirrels (Spermophilis lateralis). 12 animals were used and their weights ranged from 186 grams to 339 gms. 1. Animal conditions: (prior to each trial and surgery) a. ) Between the months of May and November the squirrels were kept in an environmental chamber, maintained at a constant temperature of 20 degrees Celsius with a photoperiod of 12 hours light and 12 hours dark. Squirrels were housed in pairs in polycarbonate containers which were 45X25X20 cm in dimension. A continuous supply of food and water were available to the animals throughout the duration of the experiment. Food consisted of rat lab chow diet as well as sunflower seeds and was occasionally supplemented with fruit and vegetables. b. ) Between December and Apr i l the animals were kept in an environmental chamber maintained at a temperature of 5 degrees Celsius, with a photoperiod of 2 hours light and 22 hours dark. Animals were housed in pairs in polycarbonate containers (as above) and food and water were continually supplied throughout the duration of this period. Golden-mantled ground squirrels maintain a circannual cycle of hibernation in their natural environment, and the environment present within the chamber during the winter months facilitated the animals' entrance into hibernation. Squirrels chosen for experimentation were placed in a separate cage and transported to the lab (maintained at room 30 temperature), 48 hours prior to each trial. Due to the increased temperature within the lab, compared to the environmental chamber, the squirrel would usually arouse from its hibernating state. Only those animals fully aroused from hibernation were used in each trial. 2. Surgical procedures: Euthermic animals were initially anaesthetized using Halothane gas. The dose of Halothane given to the animal was selected to just abolish the righting reflex (usually 3.5%) and once this reflex was absent the squirrel was removed from the Halothane and subsequently injected intraperitoneally with a 20% urethane solution. The dose of Urethane administered to the animal was 1 gram of Urethane per 1-kilogram body weight. Following the injection with Urethane, the squirrel was maintained on 1.5% Halothane for the duration of the surgery. Breathing frequency of the animal was assessed using respiratory impedance leads. Fur along the lateral sides of the abdomen and thorax was removed, and two incisions were made on either side of the body at the points of greatest respiratory related distension. Once the incisions had been made, the fat below the skin was cleared away to reveal the surface of the body wall at both the level of the thorax as well as at the level of the abdomen. The respiratory impedance electrodes were then sewn into place, at each of the four points on the body wall. Incisions were then sewn shut with each of the leads exiting at the points of incision. The respiratory impedance electrodes consisted of a 31 male amphenol gold pin soldered to an insulated steel wire, with the opposite end of the wire stripped, allowing the exposed wire to be sewn directly into the body wall . Airf low into and out of the lungs could also be monitored throughout the experiment by way of a pneumotach attached to P.E. tubing directly inserted into the trachea. The animal was placed on its back and the fur between the top of the sternum and the bottom of the chin was removed using animal clippers. A rostral-caudal incision was then made along the midline of the neck and excess fat and muscle were pulled aside to reveal the trachea. Subsequently, a small horizontal incision was made between the cartilaginous rings of the trachea and a piece of P.E. 240 tubing was inserted into the trachea and secured in place using suture silk as well as Krazy glue. Lastly, the open end of the cut trachea was tied off with suture in order to prevent fluid from moving up the trachea and causing a gag reflex which would interfere with the normal breathing pattern of the animal. E E G electrodes were used in order to monitor the arousal state of the animal throughout each experiment. The animal was placed in a K o p f stereotaxic device (model 900) and secured in place using the ear bars inserted into the auditory meatus. In addition, the animal's teeth were placed over a bite bar and the nose was clamped in place to provide extra stability. Care was taken to center the animal at midline zero and ensure that the degree of head tilt (+5mm) remained consistent from one animal to the next. Once secured within the stereotaxic device the fur at the back of the neck was removed and a dorsal incision extending caudally from the top of the skull to the level of the scapula was made. The skin was then pulled back to reveal the skull and the membranes covering the skull were removed by scrubbing them off using hydrogen peroxide and a 32 cotton swab. A l l 4 E E G electrodes were then screwed into the skull of the animal. Two of the electrodes were screwed into the frontal bone, one on either side of the midline, and the remaining two electrodes were screwed into the parietal bone, again, one on either side of the midline. Each E E G electrode consisted of a stainless steel screw soldered to a steel wire, which in turn was soldered to a male amphenol gold pin. Once the electrodes had been screwed into the skull, three of the four gold pins were secured into a pinstrip; which provided insulation for the pins, and prevented their movement. Grounding of the animal and the stereotaxic device was achieved by attaching the fourth pin to the ground lead of a H i Impedance probe, which, in turn, was connected to the grounding screw for the entire recording table. In order to make extracellular recordings from neurons within the medulla of the ground squirrel, an area of the skull just rostral to the nuchal crest had to be removed. Dimensions of the "window" were determined stereotaxically using ear bar zero and midline zero as reference points. The coordinates of the four points of the window were 1mm and 3mm to the left of midline zero and 3mm and 5mm below ear bar zero (i.e. + 1M, - 3 E B Z ; +1M, - 5 E B Z ; +3M, - 3 E B Z ; and +3M, -5EBZ) . Using these coordinates, the area of skull within this rectangle was removed along with the dural and arachnoid membranes, thus exposing the underlying cortex. Once the window had been cut, the exposed brain was covered with bone wax in order to prevent drying of the brain surface. Previous studies (Milsom et al., unpublished), in the golden-mantled ground squirrel have established the stereotaxic coordinates of the ventral respiratory group in this species relative to those for the same site in the rat brain. In order to accommodate for the slight differences between the two species, the squirrel's head was inclined to a level 5mm 33 above interaural zero. With this adjustment an accurate placement of the electrode could be achieved on each drop. Once the surgery was complete the 1.5% Halothane gas mixture was removed from the air supply. The squirrel was then wrapped in a heating blanket, and a temperature probe was inserted into the rectum to monitor body temperature throughout the experiment. 3. Experimental protocol: a.) Initial set-up: Each complimentary pair of gold male amphenol pins from the respiratory impedance leads was connected to respiratory impedance converters. Both of the respiratory impedance converters were connected to Gould amplifiers. Output from both amplifiers was fed directly onto chart paper and stored on a computer data acquisition system ( C O D A S , D O S version). The output from the thoracic respiratory impedance converter was also displayed on an oscilloscope (Tektronix model SI 11 A ) so that the phase relationship of the discharge of neurons encountered during the drop could be assessed. The air supply was regulated by a flowmeter attached to the air source in the lab, and was kept at a constant rate of 1000 ml/minute. There were two additional flowmeters, one, which regulated the nitrogen supply coming from a nitrogen cylinder (100% nitrogen), and the other, which regulated the carbon dioxide flow coming from a carbon dioxide cylinder (100%) carbon dioxide). Output from all three flowmeters was channeled 34 into a common piece of tubing to mix, when either hypoxic or hypercapnic mixtures were required. The gas supply leading from the flowmeters to the animal was attached indirectly to the tracheal tube of the animal by way of a T-piece. The arm leading to the animal was connected to the tracheal cannula by a pneumotachograph and gas would flow tidally through the pneumotachograph each time the animal breathed. Gas could be drawn from a side arm in the pneumotachograph into a carbon dioxide analyzer as well as an oxygen analyzer (Beckman models L B - 2 and O M - 1 1 , respectively). The gas analyzers were used to determine the fractional oxygen and carbon dioxide concentrations given to the animals during different trials. Ports leading from either side of the membrane dividing the pneumotachograph were connected to a validyne differential pressure transducer. Output from the pressure transducer was then amplified on a Gould amplifier and directly recorded onto chart paper. Concomitantly, the output from the amplifier was sent to the computer to be stored on the data acquisition system ( C O D A S ) . The E E G electrodes secured in the amphenol pinstrip were connected to a cable lead; and two of the three wires from the cable lead were then connected to a Gould Universal amplifier. The two electrodes that gave the best E E G trace were used. Output from the amplifier was recorded directly onto chart paper and was also connected to the data acquisition system so that it could be stored on computer. A l l channels on the computer data acquisition system sampled incoming data at a rate of 1000 Hertz. The extracellular recording electrode was a stainless steel microeletrode (A and M systems, catalog number 5715), epoxy-insulated, with an 8 degree tapered tip and an impedance of 12 mega ohms. A single microelectrode was secured to the K o p f 35 stereotaxic arm, and the electrode was then connected to a Grass amplifier (model P51 I K ) by way of a Grass H i Impedance probe. There were three outputs leading from the Grass amplifier, one to the oscilloscope so that the neuron action potentials could be viewed, another to the Grass audio monitor (model A M 8) so that the neuron firing could be heard, and the third output to a window discriminator (W-P instruments inc. model 121) so that discharge from individual neurons could be isolated from a group of several neurons firing in the same place. The window discriminator had two outputs one which led to the oscilloscope, so that the individual neuron activity could be displayed along with the breathing trace and another which connected the window discriminator to the computer data acquisition system so that the window discriminator output from a respiratory related cell could be stored. Ear bar zero and midline zero were determined using a removable zero point (Kopf model 950) which attached to the stereotaxic apparatus. A l l electrode drop coordinates were measured relative to this zero coordinate. The zeroing procedure was repeated for each new electrode used. b.) Extracellular Recordings: Once the set-up was complete the electrode was lowered into the brain of the animal at coordinates corresponding to various regions within the ventrolateral medulla. Depth readings were taken at the top of the brain, and when the electrode reached the base of the brain. These measurements were taken in order to determine the depth at which respiratory related neurons were located, as well , they were used to determine the depth of the entire brain of each animal. 36 A s the electrode was dropped through the brain, the audio monitor was used to listen for neurons, which fired in phase with a portion of the respiratory cycle. Once a respiratory related neuron was heard, its presence was confirmed by viewing both the raw trace of the neuron's activity and the superimposed respiratory impedance trace on the oscilloscope screen. After finding a respiratory related neuron, its signal was isolated using the window discriminator and then stored on the computer data acquisition system along with the 4 other traces ( E E G , respiratory impedance x 2, and pneumotachograph traces). The animal was then allowed to cycle through the wake-like state as well as through the sleep-like state, in order to assess the effects of state change on the neuron's activity. Once the animal had cycled through both arousal states, the animal was given either a hypercapnic gas mixture or a hypoxic gas mixture to breathe. Levels of hypoxia to which the animals were exposed were usually in the range of 10%-12% oxygen in nitrogen gas mixture and the levels of hypercapnia given to the animal were usually in the range of 4%>-6% carbon dioxide in air. The levels of the hypercapnic and hypoxic gas mixtures were adjusted so that the animal's breathing frequency was approximately two times the breathing frequency of the animal on air and in the same arousal state. If possible, the animal was exposed, separately, to both the hypercapnic as well as the hypoxic gas mixture in order to assess the effects of these gas mixtures on the neuron's activity. Once the gas mixture was turned on, all variables were recorded continuously on the computer data acquisition system until the animal cycled through both the sleep-like state as well as the wake-like state. 37 Arousal state changes were determined by viewing the E E G trace on the chart recorder (see Reschtschaffen et al., 1968, for a detailed description of the method used). The urethane anaesthetized preparation has E E G activity similar to that seen in the unanaesthetized animal (Figure 4). Arousal states through which the unanaesthetized ground squirrel typically cycles include wakefulness, light sleep, slow wave sleep, and rapid eye movement ( R E M ) sleep. Similarly, the anaesthetized preparation cycles through state 1 (analogous to awake), state 2 (analogous to light sleep), and state 3 (analogous to slow wave sleep). In addition, a state 4 and 5 have also been classified, with state 5 corresponding to an epileptiform E E G and state 4 corresponding to the transition between state 3 and state 5 (Grahn et al., 1989; Grahn and Heller, 1989). Due to the fact that R E M sleep is characterized according to decreased muscle activity ( E M G trace) as well as E E G activity, and the fact Urethane depresses muscle activity, it was impossible to tell whether or not the animal cycled through a R E M - l i k e sleep state. When data for all states and all gases had been collected for a particular cell the computer data acquisition system was paused and the search for a new respiratory related neuron began. In some instances it was not possible to record neuron activity for certain states and certain gases, either due to loss of the neuron or due to lack of state 3 data during exposure to a gas mixture. 38 4. Data analysis: a.) Qualitative classification o f neuron type: Respiratory related neurons were classified using a combination o f two methods: the method o f classification used by Bianchi et al. (1995, pg.3), as well as the method o f classification used by Cohen (1979, pg. 1114). Initially, cells were classified according to the phase o f the respiratory cycle during which they fired, for example, inspiration, expiration, or phase spanning. Cells were then grouped according to the portion o f the phase in which they fired, for example, early or late. In addition, the cells were grouped according to the pattern o f discharge which they exhibited, whether it be augmenting, decrementing, bursting or continuous (Figures 1 and 5). Cycle triggered histograms (Figure 6) were used to assist in the classification o f some o f the neurons and these histograms were generated by taking the average activity o f a single neuron over a period o f approximately 40 breaths. Each bin on the histogram was generated by dividing the breaths up evenly into 21 separate bins, in this way, each histogram o f a particular neuron's activity could be compared with another histogram o f the same neuron, but under a different condition. The histogram also allowed us to determine the peak instantaneous activity o f each neuron under each condition, as well as the value for the average activity for each neuron under each condition. 39 b.) Quantitative classification of neuron type: A subjects by treatment analysis of variance ( A N O V A ) , was used to determine whether or not the neurons that were found were significantly related to the respiratory cycle. The A N O V A statistic was also used to assess the consistency and signal strength of each respiratory related neuron and therefore give additional information about the effects of certain stimuli on the activity of each neuron. The variance ratio, termed eta squared (Orem, 1995), determines the consistency of discharge for each respiratory related neuron from breath to breath (See Orem and Trotter, 1994 for the derivation of the eta squared statistic). 40 F I G U R E 5: Recordings from both inspiratory and expiratory units in the ventral respiratory group were made. The pattern of discharge of these units is illustrated. The upper trace is respiration, the lower trace is unit discharge. A l l patterns shown here were found for both inspiratory and expiratory units. The most commonly found unit was the inspiratory continuous unit. The EI phase spanning unit is not illustrated in this graph, inspira t ion is a downswing of the pen. 41 1 second 42 F I G U R E 6: Histograms were used to help determine cell classification. These histograms represent the six different cell types recorded from in the current investigation. The dashed line indicates the transition from inspiration to expiration. 43 i nsp i ra to ry a u g m e n t i n g ce l l inspiratory augmenting tonic eel 0 20 40 60 80 100 percent of breath complete 0 20 40 60 80 100 percent of breath complete inspiratory continuous cell expiratory decrementing cell 0 20 40 60 60 100 percent of breath complete 0 20 40 60 80 100 percent of breath complete expiratory inspiratory phase spanning cell late inspiratory cell 0 20 40 60 80 100 percent of breath complete 0 20 40 60 80 100 percent of breath complete 44 Each breath in a series of 30 breaths, for example, is divided into twenty equal fractions, the number of spikes within each of these fractions is considered the treatment group, and each breath is considered the subject, in a subject by treatment design A N O V A . Eta squared was determined for all neurons in each of the conditions tested, c. Localization of respiratory related neurons: Locations of each of the neurons within the medulla (Figure 7) were determined using the stereotaxic coordinates on the stereotaxic device, relating these coordinates to the structures identified in the rat brain atlas (Paxinos and Watson). This method was confirmed by a previous study (Milsom et al., unpublished study) which examined a similar area in the brain of the golden-mantled ground squirrel. 45 F I G U R E 7. Horizontal section through the medulla of the rat. Specific coordinates of each cell can be found in the appendix. Taken from Paxinos and Watson (1986). The figure is divided in two parts, one half showing the distribution of expiratory and inspiratory cells, the other showing all cells recorded from. Expiratory cells are marked with closed circles (•), inspiratory cells are marked with open circles (o). Smaller dark circles include all respiratory cells found in the medulla (.). • o Ul © 3 3 OS o 917 47 S U M M A R Y OF R E S U L T S : Twenty-two respiratory neurons were identified in the area of the V R G in the medulla (Figure 7, also see appendix for exact coordinates). The location of these cells extended rostrally from the region of the nucleus ambiguus, caudally to the region of the c V R G . O f the twenty- two cells found, 14 were inspiratory, 5 were expiratory and 3 were phase spanning. Respiratory neurons are classified according to the phase of respiration in which they fire and the type of discharge exhibited. In total, six classes of cells were identified (Figure 5). Included in this classification were three types of inspiratory cells, one type of expiratory cell and two types of phase spanning cells. Inspiratory cells included inspiratory continuous (n=8), inspiratory augmenting (n=2), and the inspiratory augmenting tonic types (n=4). Inspiratory continuous cells could be characterized by their continuous firing pattern throughout the phase of inspiration. Inspiratory augmenting cells had a discharge that increased steadily throughout the latter half of the inspiratory phase and had a peak discharge near the end of this phase. Inspiratory augmenting tonic cells fired throughout inspiration and expiration, with a peak discharge near the end of the inspiratory phase. The phase spanning cells included late inspiratory (n=2) (=IE phase spanning) and EI phase spanning cells (n=l). Late inspiratory cells were characterized by a constant rate of discharge concentrated at the end of inspiration. EI phase spanning cells had a constant discharge throughout respiration with a peak discharge rate upon transition from expiration to inspiration. In addition to the phase spanning and inspiratory type 48 cells, one type of expiratory cell was identified; this was the expiratory decrementing cell. This cell can be characterized by a peak neuronal discharge at the beginning of expiration that decreases in rate of discharge throughout the remainder of the expiratory phase. The use of Urethane anaesthesia in the current investigation necessitated the characterization of arousal states based on E E G criteria. Three states were observed: state I, state II and state III. State I E E G appeared to be analogous to awake in the unanaesthetized animal, with characteristic high frequency, low amplitude waves (Figure 4). State II was analogous to drowsy in the unanaesthetized animal, with bouts of arousal-like E E G waves interspersed with sleep-like E E G waves (Figure 4). State III could be compared to slow wave sleep in the unanaesthetized animal, with E E G waves of large amplitude and low frequency (Figure 4). In addition to the similarities in E E G between Urethane anaesthetized animals and unanaesthetized animals there were also similarities in the respiratory characteristics of the two preparations. Normoxia, hypoxia and hypercapnia all appear to evoke analogous responses in the Urethane anaesthetized animal compared to the unanaesthetized animal (Hunter and Mi lsom, in press 1998). Ventilation increased in response to both hypoxia and hypercapnia. Golden-mantled ground squirrels exposed to hypercapnic gas mixtures increased tidal volume and breathing frequency. Hypoxic gas mixtures had minimal effects on tidal volume in this species, but resulted in a significant increase in breathing frequency. Ventilatory responses also varied from one state to the next. Upon transition from wake to sleep in the golden-mantled ground squirrel there was a decrease in breathing 49 frequency and an increase in tidal volume. In addition, the respiratory drive changed with an increased sensitivity to hypercapnia and a blunted response to hypoxia during state III. Three variables were measured for each of the cells. These included unit discharge per breath, peak neuron activity (spikes per second), and eta squared (n 2) value. Unit discharge per breath gives a rough indication of respiratory drive generated by the cells. It is only an estimate of respiratory drive due to the fact that no distinction can be made between pre-motor and motor neurons from the recordings. If recordings were made from a respiratory pre-motor neuron it would be impossible to predict the discharge of the neuron. However, in a respiratory motor neuron, one would expect the discharge to be similar to the discharge observed in the phrenic nerve i.e., an incrementing discharge over the course of the inspiratory period. With an increase in respiratory drive one would expect an increase in neuron activity per unit time in motor neurons as well as a decrease in the length of the breath. To distinguish between these two effects, peak neuron activity must also be measured. Peak neuron activity measures the number of spikes per second and hence the maximum discharge rate achieved for each cell. Finally the revalue was calculated for each cell under each condition, n is a statistic derived by Orem which determines the strength and consistency of discharge for each respiratory related neuron from breath to breath (for derivation of the n 2 statistic see Orem and Trotter, 1994). Orem postulates that this statistic gives an indication of the cell's role in respiratory control; the more closely linked the discharge is to respiration, the more important the cell's role is in respiratory control. 50 FIGURE 8: The effects of arousal state on unit discharge per breath (spikes per breath) are shown in this graph. There is no significant difference in discharge per breath between state I and state III. 51 30 -I 2 5 -(0 ?n 0) ^ u a) 1 5 -o CO c o a) z 10 5H S T A T E I S T A T E III Arousal state A R O U S A L S T A T E : Effects on cell activity 52 Unit discharge per breath was not significantly different between state I and state III (Figure 8). If one looks at respiratory related cells grouped according to discharge pattern, discharge per breath for state I compared to state III is similar for all groups, with the exception of the EI phase spanning cells, and the expiratory decrementing cells (Figure 9). Expiratory decrementing cells appeared to exhibit an increase in discharge per breath from state III to state I. Increases in discharge per breath for expiratory decrementing cells during state I occurred in each cell within that class with the exception of one cell (Table 1). EI phase spanning cells appear to decrease activity during state I compared to state III, although it is difficult to draw any conclusions from this observation since there was only one cell of this type found. Peak neuron activity (unit discharge per second) was significantly different in state I compared to state III (p=0.011) (Figure 10). A n increase in peak neuron activity was observed in animals in state I. There was an increase in spikes per second from 49.212±6.621 in state III to 65.945±9.905 in state I (Figure 10). Peak neuron activity increased in all classes of cells (Figure 11). Large increases were observed in the expiratory decrementing and inspiratory continuous cell types. Only three out of eighteen cells had a higher peak neuron activity in state III compared to state I. O f these three, one was an inspiratory augmenting tonic, one an inspiratory continuous and another was an expiratory decrementing cell (Table 2). 53 FIGURE 9: The effect of arousal state on unit discharge per breath for each class of cell are shown in this graph. The activity of cells in state I is expressed as a percentage of the activity measured in state III activity. Neuron activity normalized to state III (percent) ro - P . o o o O) o oo o o o ro o o o J _ J I I L I I I I I I I I I I I I I I I I L o 0 O z 0 "D <D r m m m o m O 55 T A B L E 1: Discharge per breath of each respiratory cell during state I and state III, as well as discharge per breath of each respiratory cell upon exposure to hypoxic and hypercapnic stimuli in each state. Inspiratory continuous (I-con), inspiratory augmentin: tonic (I aug t), inspiratory augmenting (I aug), late inspiratory (late I), expiratory inspiratory phase spanning cell (EI), and expiratory decrementing (E dec). d d d d (t (t m a n o o o o o CO P CD CO CO CO ~ - ~ - CTQ CTQ CTQ OQ OQ CTQ o o o o 3 3 o o o o o o o o 3 3 3 3 r-t-po _ NJ L / l 4^ , , , . - J bo 4^ N O >—' L O O N i—> N O l / i l / i O o 4^ 1+ O N O N O O O o 4^ 1+ 1+ 1+ 1+ 1+ 1+ 1+ Lo I L N O L/i ' bo bo L / i O N L O O N O N L/I O N O L / l NJ L / l NJ >—• O N N O NJ O N —J NJ L O 00 4 i . 4^ 1+ 1+ 1+ 1+ oo O N L/I •—' O •—J L / l O L O L O 1+ i — NJ O L / l i — 1 oo 1+ 1+ o o O N O N O O N O 00 L O O N N O L O —J 1+ 1+ |+ bo • NJ N O NJ 4 ^ O O 1+ o LO o © o 1+ LO 4^ NJ LO 4^ 4 ^ 1+ O N N O L / l ^0 NJ bo o LO 4^ LO N O o 4^ © LO 1+ 1+ 4^ O 1+ 1+ 1+ bo LO LO L / i O N LO LO N O 1—I 1 L NJ NJ bo O N O N L / i N O LO NJ © 1+ LO NJ O 1+ 1+ 1+ LO 1—' 4^ 1 1 1 — 1 O N NJ ^ 1 4^ O N LO LO © O N O N O N o LO bo 00 © NJ ~0 ^0 4^ 1+ N O N O o NJ 1+ 1+ 4^ 1+ 1+ 1+ 1+ 1+ L / i NJ bo o L f i O N LO 4^ o NJ 4^ L / l O N NJ 00 L / l L / l O N L/> 4^ 4^ O o L / l O N NJ o o O N H- l + , + - ~ .+ O N LO NJ 4 ^ O N L/I L O L/I >—» L / l O N 1+ O N N O CO Co 3 Co 8 l _ \ ) O N N J 00 N J N O N J L O L / l 4^ L / i 4^ 4^ bo L O © 1—» © O N 4^ O N O 1+ 00 O 1+ o 1+ 1+ o 1+ 1+ 1+ 1+ 1+ L / l O N N J O N O N L / i 4 ^ 1 ' N O N O N J O N bo N J O N J 4^ 4^ LO 4^ LO O O t t O N 4^ NJ N O L f l L f l 4^ LO LO o po NJ NJ N O 00 L f l O N NJ bo LO L / l o LO NJ LO O N O N 1+ O N NJ N O 1+ O N 1+ LO 1+ LO 1+ ^ 1 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 4^ NJ O <1 L/i LO bo LO NJ N O 4^. LO O 4 i - 4i> LO LO O 4^. 4^ N O o 4^ O CO NJ LO 00 NJ NJ O 4^ NJ O o O N LO O N bo NJ LO o LO —i N O 1+ LO 1+ LO 00 1+ 1+ 1+ 1+ 1+ O N N O LO NJ N O O N N O o LO LO L f l O O _ O N N O —1 O N 1^1 O o O bo LO 4^ NJ 4^ o o N O LO 4 ^ NJ L/i 1+ 0± 8± 1+ 1+ 1+ 1+ 6± O N N O LO 4^ 4 ^ O N LO 4 ^ 4^ 00 2.66 Co LO LO O O bo bo N O N O 1+ 1+ NJ O N LO LO O N 1—> NJ L h 4^ L / l 00 ^ 00 00 ^ 1 o 1+ 1+ N O O O o 1+ NJ 1+ 1+ 1+ 4^ 3.86 LO O N o N O NJ N O N O 3.86 1 ' / l O N L / l L / l N O LO i—. LO O 4^ O N P LO o 4^ -~J 1+ 1+ 1+ 1+ 1+ 4^ LO 4^ 4^ O N LO 4^ LO LO 1+ bo g Co 8 57 F I G U R E 10: The effects of arousal state on peak neuron activity (spikes per second) are shown in this graph. There is a significant difference (p=0.01) in peak neuron activity between state I and state III. Peak neuron activity in state I is 65.945±9.905 spikes per second. Peak neuron activity in state III is 49.212+6.621 spikes per second. 58 80 T3 C o o w (/) 0) a 0) 6 0 -| 4 0 o (0 c o 1 _ D 0 z (TJ a) 20 STATE I STATE III Arousal state 59 F I G U R E 11 : The effect of arousal state on peak neuron activity (spikes per second) for each class of cell are shown in this graph. The activity of cells in state I is expressed as a percentage of the activity recorded in state III. Peak neuron activity normalized to state III (percent) ro o o o CO o o o ro o o CD O I I I I I I I I I I I I I L J I I I J I L > —\ o o o z CD_ l-f-m m o m O 09 61 T A B L E 2: Peak neuron activity (spikes per second) of each respiratory cell during state I and state III, as well as peak neuron activity of each respiratory cell upon exposure to hypoxic and hypercapnic stimuli in each state. Inspiratory continuous cells (I con), inspiratory augmenting tonic (I aug t), inspiratory augmenting (I aug), late inspiratory (late I), expiratory inspiratory phase spanning (EI), and expiratory decrementing cells (E dec) are listed. O J t O O N _» . 4*. O N o t o O N N O O J 4^ O N O O J O J -o O N 00 I—* N O ± 1.8 ±2.9 ±2.6 0 + 4. 5± 15 ±3.5 ±4.3 O N o O N o n t o N O o N O o o W W W W W J ' s £ ' M > H M M M h-. H - . CTQ fjq ffq OQ CTQ 00 o ti n n o o o o o o o 3 4^ . On 00 4^ O J Of) oo o n O J 1+ 1+ KJ 4^ O t O O N N O O N —j t o 1+ 4^ 4*. 4^ O J 4^ O O K ) W 4 i O O o o o n N O * o . o n O O0 O O J 4^ 1+ 1+ 1+ 1+ 1+ O J O J 4^ - O J O J >—• t o o ^ ^ O N o n O N 4^ -o t o t o 1+ II O N 1+ ^ 0O O J O J >—' —] o o IX w oo ? ! © o n O N t o 1+ -1^ o ^1 o o 1+ o n b o o n O O ^ | O N O N ~ - < \ N O O N N O k ) s j t o o O N O N 1+ 1+ 1+ 1+ ~ N O O N O J N O O J N O - J - o >-' o o o n -£> o n o n O J t o O N o n 4^ -t^ 4^ o o n o o 00 N O o o '—^ 4^ o n O N 4^ O J N O - 0 1+ N O OO N O O -t^ o o N O 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1 ' O J O J o n 4^ O t o 1+ t o O N o n 4^ O b o O J o n 00 O N o O J O N o n Co Co 8 O N N O o n o n 1+ O N O t o o o o n p—. N O O J O N t o O J O N t o N O N O o o o N O 4*. i—> O J p o o o O J t o N O © b o N O N O o b o t o o n o N O 4^ 00 t o 4^ — J O J O O N 1+ 1+ 1+ O J 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ o n O N O J o n o o O J t o O J t o 4^ •—» t o '<—* t o O N ^ ) o 4^ N O O J t o o o O N o n o o 4^ t o O N s5 >^  t*3 t o o t o o O N o n - J O N t o N O t o O J 4^ O N 4^ 4^ O J 4^ 4^ O N t o N O O J 4^ -1^ t o N O O J O N O N o o O J 4^ t o t o t o O N N O 4^ o o n o O J o o O N O N O J o N O t o N O —1 4^ OO © OO O J 00 t o t o N O 4^ O J O N t o N O N O O J 4^ t o 4^ O N t o O J 4^ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ O N 1+ 1+ |—' O J O J o n O J 4^ t o O J t o O N 4^ t o t o t o - J O J 1+ _ 4^ 4^ —1 o n N O N O O J o n 4^ O N N O OO © O J t o - J O J N O o o 4^ 4^ t o t o b o 4^ ^ - J O N o O J t o O N 4^ b o 4^ Co 53 O J o n O N O 4^ O J o 00 ; - J O N O N o n 4^ O 4^ O N t o N O 4^ O N O t o o n o 4^ N O - 0 4^ b o 4^ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ O J 4^ O N >—* O J o n o n b o t o t o t o 00 © t o t o O J OO t o O N b o O N 4^ O N t o t o  t O O N O 4^ - J 1+ 1+ 1+ ^ o n o n o n 4^ t o o 1+ o n o j N O O N N O 4^ 1+ t o O J o n t o o n b o o 1+ fc-l Co O N ^ O J o n o j y> ~ o n b o ^ N O O J o O J O N o o o n i—' —1 O J 4^ i , 4^ o o o o 4^ o o N O o —1 o o 00 o o t o O N O N \ l o n o O J O J N O t o b o t o t o OO OO N O — J O J O N 1—' t o - - J o o O N t o 1+ ±6.01 1+ 1+ o 1+ 1+ 1+ 1+ 1+ 1+ 1+ 1+ 4^ ±6.01 4^ 4^ N O o n t o o n o o O N ±6.01 00 O J o O N 4^ t o o n 4^ . N O O o o o n t o O 4^ N O o o Co 3 63 H Y P O X I A : Effects on cell activity Unit discharge per breath in animals exposed to air was 20.835+4.314, and this activity decreased to 12.674±2.303 upon exposure to hypoxic gas mixtures in state I. The change in unit discharge approached significance (p=0.09) (Figure 12). During state I each class of cell exhibited an overall decrease in activity upon exposure to hypoxia with the exception of the inspiratory augmenting class of cell (Figure 13). There was no significant difference in neuron activity in animals exposed to a hypoxic mixture during state III as compared to those exposed to air (Figure 12). Four out of five classes of cells exhibited an overall decrease in average neuron activity upon the animal's exposure to the hypoxic gas mixture during state III (Figure 13). However, one of the inspiratory augmenting tonic cells showed a large increase in activity in response to exposure to hypoxia during state III, thus increasing the average unit activity of this group. A l l other inspiratory augmenting tonic cells exhibited a decrease in unit activity in response to hypoxia (Table 1). Peak neuron activity of respiratory related cells increased in animals exposed to hypoxic gas mixtures (Figure 14). There were significant increases in peak neuron activity in response to hypoxic gas mixtures during state III, (62.640±10.358) compared to exposure to air during state III (43.209±6.313) (Figure 14). However, peak neuron activity didn't increase significantly during state I upon exposure to hypoxia (Figure 14). This trend of increased peak neuron activity appeared to be consistent for all classes of cells during state III, with the inspiratory cells exhibiting a more dramatic increase than the expiratory and phase spanning cells (Figure 15). 64 F I G U R E 12: The effects of hypercapnia and hypoxia on unit discharge per breath are shown in this graph. Discharge during exposure to hypercapnia and hypoxia are expressed as a percentage of the activity of these cells when the animal was exposed to air. The effects of hypoxia on unit discharge per breath approach significance (p=0.09) state I. The unit discharge per breath of animals breathing air during state I was 20.835±4.314 spikes per breath. The unit discharge of animals breathing a hypoxic gas mixture in state III was 12.674±2.203 spikes per breath. 140n HYPERCAPNIA HYPOXIA Gases 66 F I G U R E 13: The effects of hypoxia on unit discharge per breath for each class of cell is shown in this graph. The effects of hypoxia during state I and state III are expressed as a percentage of the value obtained on air in each respective state. o N e u r o n a c t i v i t y n o r m a l i z e d t o a i r ( p e r c e n t ) ro o o o 00 o o o ro o > c Q O (D TS CD > C Q O O > m m D m O m CO CO —1 H > > —1 H m m L9 68 F I G U R E 14: The effects of hypercapnia and hypoxia on peak neuron activity (spikes per second) are shown in this graph. Peak neuron activity resulting from exposure to either hypercapnic or hypoxic gas mixtures is illustrated as a percentage of the peak activity when the animal was exposed to air. Exposure of animals to hypercapnia resulted in a significant increase (p=0.01) in peak activity compared to air. Peak activity of respiratory related cells during state I upon exposure to air was 56.8261826 spikes per second. Peak activity of respiratory related cells during state I upon exposure to hypercapnia was 70.553±9.723 spikes per second. During state III peak activity of respiratory related cells upon exposure to hypercapnia was 66.160±12.739 spikes per second compared to 41.56515.04 spikes per second upon exposure to air. Hypoxia resulted in a significant increase in peak activity during state III. Values increased from 43.20916.313 spikes per second in air to 62.640110.358 spikes per second during hypoxia. 2001 c CD O 5 150-Q. STATE III STATE I (0 o •M "D CD NJ 75 0 c o crj c 0 • CD c CO CD 0_ 100" 50-HYPERCAPNIA HYPOXIA G a s e s 70 In fact, each cell recorded from underwent an increase in peak neuron activity when the animal was exposed to the hypoxic gas mixture during state III with the exception of 1 expiratory decrementing cell (Table #2). During state I the changes in peak neuron activity weren't as consistent as those observed during state III. Some cells exhibited increases in peak activity upon exposure of the animal to a hypoxic gas mixture during state I, whilst other cells exhibited a decrease in peak neuron activity (Figure 15 and Table 2). H Y P E R C A P N I A : Effects on cell activity The unit discharge per breath in animals exposed to hypercapnic gas mixtures didn't differ significantly from the unit discharge in animals exposed to air (Figure 12). With the exception of the inspiratory augmenting tonic class of cells there were slight increases in unit discharge during state III upon exposure to hypercapnic gas mixtures (Figurel6). During state I there was a decrease in neuron activity upon exposure to hypercapnic gas mixture, with the exception of inspiratory continuous type cells (Figure 16). The responses of the cells within each class of cells was inconsistent, some cells increased activity in response to exposure of the animals to hypercapnia whereas others decreased in activity (Table 1). Inconsistent responses occurred in both state I and state III. 71 F I G U R E 15: The effects of hypoxia on peak neuron activity (spikes per second) for each class of cell is shown in this graph. Peak activity occurring during exposure to hypoxic gas mixtures is illustrated as a percentage of the peak activity when the animal was exposed to air. Peak neuron activity normalized to air (percent) O O O O IV) O O PO cn O > c 0 > C G) O 0 z > m m • m o LU CO CO H —i > > —1 m m — 73 F I G U R E 16: The effects of hypercapnia on unit discharge per breath for each class of cells. Unit discharge per breath during exposure to hypercapnic gas mixtures is presented as a percentage of the unit discharge per breath when the animal was exposed to air. 7 5 Exposure to hypercapnic gas mixtures resulted in significant increases in peak neuron activity compared to animals exposed to air (Figure 14). During state III, animals exposed to hypercapnia had a peak neuron activity of 66.160±12.739spikes per second compared to those exposed to air which had a peak activity of 41.565±5.04 spikes per second (Figure 14). Each class of cell increased peak neuron activity during state III upon exposure to hypercapnic gas mixtures (Figure 17). During state I respiratory related cells in animals exposed to hypercapnic gas mixtures had a peak neuron activity of 70.553±9.723 spikes per second, compared to animals exposed to air which had a peak neuron activity of 56.826±7.725 spikes per second (Figure 14). In all but one class of cells (the EI phase spanning cells) there was an increase in peak activity upon exposure to hypercapnia. Examining the peak activity of each respiratory related cell it was observed that hypercapnia increased peak neuron activity in 12 out of 15 of the cells (Table2). Only one of the three cells decreased peak activity in response to hypercapnia during both state I and state III (Table2). EFFECTS OF STATE A N D GASES ON THE ETA SQUARED V A L U E S (n2): There was an increase in r\2 values in state III compared to state I which approached significance (p=0.1) (Figurel8). That is to say respiratory related discharge was more closely coupled to breathing during state III than during state I. Consequently r| 2 values were lower in state I (0.588±0.051) than they were in state III (0.656+0.0420). 76 F I G U R E 17: The effects of hypercapnia on peak neuron activity (spikes per second) for each class of cell are shown. Peak activity during hypercapnic gas mixtures is presented percentage of the peak activity when the animal was exposed to air. as a Peak neuron activity normalized to air (percent) cn o o o J I I L cn o J I I L ro o o J I L > c CD O CD > c Q o O m m a m O 00 00 —1 —I > > H —I m m LL 78 FIGURE 18: The effects of arousal state on r\2 values are shown. The difference in r\2 values between state I and state III approaches significance (p=0.11). The n value for state I is 0.588+0.051, while the n 2 value for state III is 0.656±0.042. 79 0.8 - i 0.7 0.6 w 0 5 CO > T5 2 0.4 • w CO © 0 . 3 0.2 o.H 0 S T A T E S T A T E A r o u s a l s t a t e 80 FIGURE 19: The effects of arousal state on r\2 values for each class of cell are shown in this graph. The r]2 values of cells in state I are presented as a percentage of state III values. eta squared values normalized to state III (percent) ro J> O) oo o 82 TABLE 3: r\2 values of every respiratory cell in response to different arousal states as well as upon exposure to hypoxic and hypercapnic stimuli within each state. Listed are inspiratory continuous cells (I con), inspiratory augmenting tonic (I aug t), inspiratory augmenting (I aug), late inspiratory (Late I), expiratory inspiratory phase spanning cells (E/I), and expiratory decrementing cells (E dec). M f f l M W M r r l t — l t ~ H (D rc n> (D n> n> n> o o o o o P ^ 3 p p C C- C OQ CTQ OQ OQ 00 p OQ W OO v l \ D U l s ) O i W * C k ) OO s i 4^ O 4^ NO M W U l ON W U i - J t O O N A l ? i W s i K) NO NOJ^OOlONOUiU i 4i M O U i N O N O U i s l s l 4 ^ ^ - 4^  Ui W U l U l ON U l O NO VD M oo v l s i NO 4^ ON OO ON OJ to 4*. o O ON ON U i OJ oo 4^ OJ O OO NO 4^ U l i— 1 00 O OO — J •—> oo NO s i o ui U l -J NO ON s i O OJ to 00 NO OJ ON NO ON i— ' ui OJ ui ui OJ ON to i— 1 NO oo i— 1 OJ NO O 4^ NO 4^ Ul O OJ ON O N t O O O s ) O J N O O N N O O O O J O N O N O N 4 ^ U l s l 4 ^ 0 0 0 0 s l U l O O 0 4 ^ U l O N s ) 4 ^ 0 N ^ 4 ^ t O s l O O O J O N ^ - O N O U l O N t O U l U l 4 x w - j M ^ o o \ o i o o o ^ o \ ^ u i ^ O N W u \ j i O \ W N O O \ U l o to ON NO tO ON 00 s i s ] s ] s i 0O U l to OO 4^ 4^ U l ON tO U i U l 4^ s i 00 ON NO S) 00 O s i U i s ) 4^ O 4^ NO 4^ NO >—' OO o ui s i s ) O s ) 4^ OO OJ OJ ON U I O 4^ 4 i ^ U i 4 i w ON U i O O U I O N oo oo ^ o o s i o ^ OO W N O O J s l t O N O O N O J to 00 C o l 8 3 Co >3 Co ^ Co CO Co t-c e8 84 Expiratory decrementing cells and EI phase spanning cells were the only two types of cells whose n 2 values didn't increase during state III (Figure 19). Within each class of cells there were individual cells that behaved in an opposite direction from the aforementioned trend (Table 3). Approximately half of the inspiratory continuous cells had lower r\2 values in state III (Table 3). r | 2 values of respiratory related cells in animals exposed to hypoxic gas mixtures were significantly different from the r\2 values in animals exposed to air. During state I exposure to hypoxic gas mixtures resulted in a decrease in n 2 values from 0.564±0.059 on air to 0.443±0.056 during hypoxia (Figure 20). Exposure to hypoxia during state III, also resulted in a decrease in n 2 values from 0.685±0.041on air to 0.467±0.054 during hypoxia (Figure 20). The decreases in n 2 values were more dramatic during state III compared to state I (Figure 20). Each class of cells exhibited a decrease in n 2 values regardless of state (Figure 21). A more dramatic decrease in r\2 value was observed in response to hypoxia during state III in the inspiratory classes of cells (Figure 21). Conversely, r | 2 values of the expiratory classes of cells decreased more dramatically during state I (Figure 21). With very few exceptions each cell exhibited a decrease in n value upon exposure of the animal to a hypoxic gas mixture (Table 3). During state III r\2 values for respiratory related cells in animals exposed to hypercapnia were significantly different from the n 2 values of those exposed to air (Figure 20). Hypercapnic gas mixtures caused a decrease in r| , from 0.722±0.043 in air to 0.625±0.049 in hypercapnia (Figure 20). 85 F I G U R E 20: The effects of hypercapnia and hypoxia on r\2 values. r\2 values resulting from exposure to either hypercapnic or hypoxic gas mixture are presented as a percentage of the r| value when the animal was exposed to air. Exposure of the animals to hypercapnia and hypoxia resulted in a significant decrease (p=0.02 and p=0.01) in r\2 values with the exception of hypercapnia during state I. The r | 2 value for animals exposed to hypercapnia during state III was 0.625±0.049 compared to 0.722+0.043 for animals exposed to air. The r\ valuefor animals exposed to hypoxia during state I was 0.443±0.056 compared to 0.564±0.059 for animals exposed to air. The n 2 value for animals exposed to hypoxia during state III was 0.467±0.054 compared to 0.685±0.041 for animals exposed to air. STATE III STATE I H Y P E R C A P N I A H Y P O X I A G a s e s 87 FIGURE 2 1 : The effects of hypoxia on r) values for each class of cell is shown in this graph. r\ values during exposure of animals to hypoxic gas mixtures is shown as a percentage of the n 2 value when the animals were exposed to air. E t a s q u a r e d v a l u e s n o r m a l i z e d t o a i r ( p e r c e n t ) ro o 4> o 0) o CO o o o ro o > C o > c o o o > m m D m O 88 89 F I G U R E 22: The effects of hypercapnia on n 2 values for each class o f cells. r\2 values during exposure of animals to hypercapnic gas mixtures is shown as a percentage of the r) value when the animals were exposed to air. 91 With the exception of the expiratory decrementing cells, all n 2 values decreased during state III upon exposure to hypercapnia (Figure 22). The changes in n 2 value during state I upon exposure to hypercapnia weren't as dramatic, with only slight decreases compared to air (Figure 22). The only class of cells that exhibited a dramatic decrease in both state I and state III during hypercapnia were the inspiratory augmenting tonic cells (Figure 22). Examining individual cells within each class (during exposure to hypercapnia) reveals that the decrease in r\2 values is more consistent during state III compared to state I (Table 3). 92 93 D I S C U S S I O N : Respiratory related neurons were identified throughout the areas of the ventrolateral medulla in the brainstem of the golden-mantled ground squirrel. N o pattern emerged with respect to the distribution of expiratory related neurons compared to inspiratory related neurons (Figure 7, and see appendix), or with respect to the specific cell types identified. In studies examining rats and cats there is a trend in the distribution of the respiratory cells. Phase spanning cells are concentrated in, but not restricted to, areas of the rostral medulla, pons, and midbrain (see review by Bianchi et al., 1995). Neurons within the caudal V R G are primarily of the expiratory augmenting and late expiratory type (vonEuler, 1986, Kirkwood and Sears, 1973, Merr i l l , 1981). Neurons within the intermediate V R G are a mixture of both inspiratory and expiratory neurons, which are primarily motor neurons (Cohen, 1979, vonEuler, 1986, Bianchi et al., 1995, Feldman, 1986, Wyke, 1974). In the most rostral portion of the i V R G are the EI phase spanning cells (pre-inspiratory) in addition to inspiratory decrementing cells. Due to the limited number of cells sampled as well as the lack of histochemical identification of the sites recorded from it would be difficult to draw any conclusions from the present study concerning analogous distribution of respiratory related neurons in the medulla of the golden mantled ground squirrel. 94 E F F E C T S OF L I T H A N E A N A E S T H E S I A O N R E S P I R A T O R Y N E U R O N A C T I V I T Y : A R O U S A L S T A T E : Respiration in sleep differs greatly from respiration in wake. In unanaesthetized normoxic animals there is a decrease in most variables associated with respiration in the sleep state as compared to the wake state. In the ground squirrel there is a decrease in minute ventilation and breathing frequency (Hunter and Mi l som, in press, 1998), as well as an increase in tidal volume. In the unanaesthetized ground squirrel there is an increase in the length of the breath largely due to increases in the expiratory time, but also due to an increase in inspiratory time ( M . Harris, Ph.D. dissertation, 1998). Similar changes in breathing frequency, tidal volume and minute ventilation during sleep have been observed in the cat, dog and rat (Orem et al., 1977, Remmer et a l , 1976, Phillipson et al., 1976, Pappenheimer, 1977). Sleep also results in decreased peak airflow, increased upper airway resistance, and blunted responses to hypercapnia and hypoxia (Orem, 1986). It would appear that Urethane anaesthetized golden-mantled ground squirrels ( Hunter and Mi l som, 1998), undergo analogous changes in respiration during state III as their unanaesthetized sleeping counterparts. Urethane anaesthetized animals exhibit similar decreases in breathing frequency and minute ventilation during state III compared to state I, as seen in unanaesthetized animals during sleep compared to wake (Hunter and Mi l som, 1998). In addition, the changes in inspiratory time and expiratory time increase 95 in a manner analogous to that seen in the unanaesthetized squirrel (M. Harris, Ph.D. dissertation, 1998). Extensive research has taken place examining the neuronal basis for the respiratory changes which occur during sleep (Fink, 1961, Fink et al., 1963, Bianchi, 1971, Puizillout and Ternaux, 1974, Orem et al., 1974, Orem, 1978, Orem et al., 1985, Phillipson and Bowes, 1986, Orem, 1988, Orem 1990). The majority of sleep research uses the cat as the experimental model, consequently the cat will be the standard of comparison in the following discussion. The first two studies (Orem et al., 1974, Puizillout and Ternaux, 1974) which examined the effects of sleep on respiratory neuron activity observed a decrease in neuron activity (spikes per second) in the majority of neurons during sleep compared to relaxed wakefulness. In addition to the decreased activity of neurons during sleep there was also a decrease in the number of respiratory related cells active during sleep (Orem et al., 1974). It is these overall decreases in activity (spikes per second), which are thought to be responsible for the observed decrease in respiration during sleep. Orem et al. (1974) found that 79 %of the respiratory related neurons studied were sensitive to arousal state changes, and the remaining 21% of the cells were insensitive to changes in arousal state. In a subsequent study by Orem et al. (1985) it was observed that 66% of the cells sampled exhibited a decrease in discharge rate (spikes per second) during sleep, 27% didn't exhibit any significant change in discharge rate and 7% of the cells increased discharge rate. Furthermore, the discharge rate of the majority of the respiratory related cells during wake was usually between 20 and 40 cycles per second with some cells firing at rates over 100 cycles per second. The discharge rates during sleep were not presented (Orem, 1974). These studies led to the 96 broad classification of two types of respiratory cells; sleep sensitive cells and sleep insensitive cells. Cells classified as sleep sensitive were those cells that exhibited respiratory related activity during wake but had a change in discharge during sleep. Sleep insensitive cells, on the other hand, were cells that had the same rate of discharge (spikes per second) regardless of whether the animal was sleeping or not. Orem et al. (1985) noted that an inverse relationship existed between the neuronal activity of the sleep sensitive cells and the length of the breath taken by the animal. That is, lower discharge rates (spikes per second) were associated with longer breaths and higher discharge rates were associated with shorter breaths. Many studies have examined the changes in frequency (spikes per second) of respiratory related cells; however, few studies have recorded the discharge per breath in animals going from wake to sleep. Those that have (Orem, 1985, Orem et al., 1985, Orem et al., 1986), observed little change in the discharge per breath in certain respiratory related cells. Similarly, the urethane anaesthetized ground squirrels in the current study did not exhibit any significant change in the discharge per breath of respiratory related neurons in state III compared to state I (Figure 8). Orem's studies used the r\ classification as the basis for examining the responses of cells to changes in state. Those cells with high r\2 values (above 0.30) exhibited no significant change in the discharge per breath in sleep compared to wake; however, cells with low eta squared values (below 0.30) did exhibit a significant decrease in discharge per breath. Only one of the cells we recorded from could be considered a low eta square (0.243) cell (as defined by Orem who only classified cells during sleep) and this cell decreased its activity from state I to state III. Upon examination of each class of cell, most cell types, with the exception of the 97 expiratory decrementing and the phase spanning cells, exhibited only small changes in discharge per breath from state I to state III. The large decrease in discharge per breath observed in the expiratory decrementing cells indicates that not only was there an increase in length of expiration but also a decrease in the total activity during expiration in state III in this particular class of cell. There were significant decreases in the rate of discharge (spikes per second) in the urethane anaesthetized ground squirrel during state III (Figure 10). This result was consistent with the results observed in various studies that examined the effects of sleep on the discharge rate of respiratory related cells (Orem et al., 1974, Orem, et al., 1985, Orem, 1986, Orem 1988). The difference in discharge rate during sleep could perhaps be attributed to a lengthening of each breath during state III and therefore even though the discharge per breath remained constant from state I to state III the time frame in which the discharge occurred was lengthened. All classes of cells decreased the rate of discharge during state III with inspiratory continuous cells exhibiting the most dramatic decrease of all the classes (Figure 11). This might be indicative of a greater influence of arousal state on the inspiratory continuous type of cell compared to the other cell types. The late inspiratory cells as well as the expiratory inspiratory phase spanning cells showed little decrease in rate of discharge during state III (Figure 11), perhaps an indication of their immutability with respect to changes in arousal state. In the present study 70% of the respiratory related cells had a decrease in discharge rate during state III, 12% had no change in discharge rate and 18% exhibited an increase in discharge rate. These values compare favourably to Orem's results in cats (1974, 1985) 98 in which 66% to 79% of the respiratory related cells showed a decrease in discharge rate during sleep. In summary, it would appear that the effects of arousal state on the discharge of respiratory related neurons in Urethane anaesthetized ground squirrels appear similar to those of arousal state on the discharge of respiratory units in unanaesthetized cats. There was an overall reduction in the discharge rate during state III compared to state I in most respiratory related cells in the Urethane anaesthetized ground squirrel. There were, however, no significant changes in the discharge per breath of the respiratory related neurons. The current study would appear to support hypothesis #1. Urethane anaesthetized animals exhibit changes in activity of respiratory neurons in response to changes in arousal state similar to those observed in unanaesthetized animals. HYPERCAPNIA : Minute ventilation increases in response to hypercapnia in both fossorial (burrowing animals) and non-fossorial species of mammals and birds (T. Darden, 1972, Boggs et al. 1983, Bismarck et al, 1956). If one compares the ventilatory sensitivity to hypercapnia between species, however, fossorial animals exhibit a blunted ventilatory response compared to non-fossorial mammals (Boggs et al., 1983, T. Darden, 1972) Golden-mantled ground squirrels are fossorial and as such, have a blunted hypercapnic response. This decrease in carbon dioxide sensitivity is believed to be genetically 99 determined, and is thought to decrease the energetic cost of breathing in the hypercapnic environment of the burrow (Boggs et al., 1984). Hypercapnia increases minute ventilation in the golden mantled ground squirrel via increases in tidal volume with only slight increases in breathing frequency (Webb, 1987, McArthur and Milsom, 1991). Depending on the species of animal under investigation the response to hypercapnia varies. Non-fossorial species display a wide variety of responses to hypercapnia; some species increase minute ventilation by increasing both tidal volume and breathing frequency whereas other species increase minute ventilation by increases in tidal volume alone. For example, in the golden hamster the increase in minute ventilation arises primarily through an increase in tidal volume (Walker et al., 1985). The laboratory rat on the other hand increases minute ventilation through increases in both breathing frequency and tidal volume (Cragg and Drysdale, 1983, Walker et al., 1985, Holloway and Heath, 1984). The pocket gopher is an example of a fossorial rodent that increases both tidal volume and breathing frequency in response to a hypercapnic gas mixture (Darden, 1972). Much of this variation may be a consequence of the shape of the response curves to increased carbon dioxide. Craggs and Drysdale (1983) determined that, in the rat, breathing frequency increased in a hyperbolic fashion whereas tidal volume increased in a linear fashion in response to increases in the partial pressure of carbon dioxide. Consequently, differences in the partial pressures of carbon dioxide used in each of the studies may have produced the different ventilatory responses to the hypercapnic stimulus. The response to hypercapnia in the urethane anaesthetized golden mantled ground squirrel is primarily an increase in tidal volume (Hunter and Milsom, 1998). This is 100 analogous to the ventilatory response observed in the unanaesthetized golden mantled ground squirrels (Webb, 1987). The neuronal basis of the hypercapnic response is complex involving sensory receptors both peripherally and centrally, as well as various pools of respiratory related neurons which integrate the input and subsequently modify the ventilation of the animal. Increases in the partial pressure of carbon dioxide as well as decreases in pH excite peripheral chemoreceptors located within the carotid body (Heyman et al. 1930). The afferents from these receptors synapse on neurons within the caudal portion of the nucleus of the tractus solitarius (Housley et al., 1987). From the NTS, projections to the VRG (Koshiya and Guyenet, 1996) as well as the parabrachial nucleus have been observed (Hayward and Felder, 1995). The location and presence of the central chemoreceptors is a question which has generated a large amount of research in the area of respiratory control. Some investigators have postulated that the central chemoreceptors lie on the surface of the ventral medulla (Cherniack et al., 1979, Loeschcke, 1981, Bruce et al., 1987). Other researchers have suggested that the respiratory related neurons within the ventral medulla itself may serve as central chemoreceptors with dendiritic arborizations that extend to the ventral surface of the medulla (Onimaru et al., 1989, Okada et al., 1993). In the present study, exposure to hypercapnia (during state I) didn't produce a significant increase in neuron activity (spikes per breath). In fact, with the exception of the inspiratory continuous neurons, the discharge per breath appeared to decrease during hypercapnia (Figure 16, table 1). A study by Nesland et al. (1965) found similar results in discharge per breath, finding no increase in the number of spikes per breath upon 101 exposure of cats to hypercapnia. However, most studies (Batsel, 1964, St. John, 1981, St. John and Wang, 1977) report an increase in discharge per breath of phasic inspiratory and expiratory cells upon exposure to hypercapnia. These conflicting results might be a result of differing preparations in these studies. The only possible difference between the two sets of conflicting studies depends on whether the animals were vagotomized or not. The current study as well as that of Nesland et al., (1965) used animals which weren't vagotomized whereas the other studies used animals which were vagotomized. Similarly, during state III, exposure to hypercapnia produced no significant change in discharge per breath, (Figures 12 and 16, Table 1). These results are consistent with those obtained during state I. There do not appear to be any studies examining the effects of hypercapnic gas mixtures on respiratory neurons during sleep, and therefore there are no other studies with which to make comparisons. Exposure of the ground squirrel to hypercapnia (during state I) resulted in peak neuron activity which was significantly (p=0.02) greater than the peak neuron activity of ground squirrels exposed to air (Figure 14). These results are consistent with several studies which examined the effects of hypercapnia on respiratory related activity of medullary neurons (Batsel, 1966, Nesland et al., 1966, St. John, 1981, Folgering and Smolders, 1979, St. John, 1977, St. John and Bianchi, 1985). In a study by St. John and Wang (1977) using cats, 87% of the neurons studied exhibited an increase in frequency of discharge, and a decrease in interspike interval. These results can be compared to those in the current study in which 70% of respiratory related neurons increased their activity in response to hypercapnia. In a similar study (Bianchi and St. John, 1985) it was observed that all expiratory neurons (N-13) reacted by increasing their discharge (spikes per 102 second) and the majority of the inspiratory neurons (32 out of 35) also increased their discharge rate. Looking at the individual classes of neurons in the present study, most of the classes increased their discharge rate to the same extent (phasic and tonic) with the exception of the EI phase spanning cells which showed no increase in discharge rate (Figure 17). In a study by Folgering and Smolders (1979) it was found that the phasic inspiratory neurons had a greater increase in discharge rate compared to expiratory and phase spanning neurons. Phase spanning neurons in particular showed no consistent response to the hypercapnic stimulus (Folgering and Smolders, 1979, St. John, 1981). Although the expiratory neurons in the present study showed an equally strong response to hypercapnic stimuli, the phase spanning neurons did not. St. John (1981) also found that the frequency of the tonic phase spanning cells he examined remained unaltered or decreased in response to a hypercapnic stimulus. The increase in peak activity is an expected response of respiratory related neurons exposed to a hypercapnic stimulus. With an increase in respiratory drive (hypercapnia) and a subsequent increase in minute ventilation one would predict a concomitant increase in the activity and recruitment of respiratory related neurons. In the current study there was no increase in discharge per breath only an increase in peak activity in response to a hypercapnic stimulus. There are several possible explanations for this disparity. Associated with hypercapnia is a slight decrease in the length of the breath, perhaps the discharge per breath remained the same whilst the breath length shortened resulting in an increased peak discharge. Another possible explanation for the decreased activity per breath is the depressive effect of carbon dioxide on central nervous system neurons (Krnjevic et al., 1965, Cherniack et al., 1971, Herbert and Mitchell, 1974). 103 The study of Krnjevic et al. (1965) on urethane anaesthetized cats found that smaller doses of hypercapnia (4%) resulted in excitation of respiratory related cells whilst larger doses (10%) resulted in depression of respiratory related cells due to an increase in the resting potential of the cell. This, however, fails to explain the increase in peak activity of the cells. If one were to assume that the respiratory related neurons and the central chemosensitive cells were one and the same as proposed by Okada et al. (1993), then it is possible that these cells were anaesthetized and their activity depressed. In a review, Loeschcke (1981) stated that the central carbon dioxide receptors are very sensitive to anaesthetic, and that this effect is partly masked by peripheral chemoreceptor activity. Perhaps the effects of urethane on the central chemoreceptors manifests itself as a constant discharge per breath while the increases in peak activity are a result of intact peripheral chemoreceptors. During state III there was an even greater increase in peak neuron activity upon exposure to hypercapnia (66.160±12.739 spikes per second) compared to air (41.565±5.04 spikes per second) (Figure 14). Every class of respiratory related neuron increased its peak activity in response to hypercapnia during state III (Figure 17). There were only three neurons which exhibited decreases in peak activity in response to hypercapnia (Table 2). These results could be difficult to explain given the decreased sensitivity to hypercapnia during SWS in most mammals (Boggs et al., 1984, Phillipson and Bowes, 1986). However, in a study done by Hunter and Milsom (1998) the sensitivity of the golden-mantled ground squirrel was found to increase in response to exposure to a hypercapnic stimulus during state III. Pappenheimer (1977) found a similar increase in sensitivity in rats upon exposure to hypercapnia during SWS. This increased 104 sensitivity to hypercapnia may explain the increase in peak neuron activity upon exposure to hypercapnia during state III. In summary, it would appear that the results in the present study are largely in accordance with results from similar studies in unanaesthetized animals. Hypercapnia usually increases the discharge frequency of respiratory neurons. In the present study, peak neuron activity increased in response to hypercapnia, analogous to the results in unanaesthetized animals. However, unlike unanaesthetized animals discharge per breath did not increase in Urethane anaesthetized ground squirrels. The results in the present study are therefore, partly in accordance with hypothesis #2. That is, the changes in activity of respiratory neurons in response to hypercapnia in Urethane anaesthetized golden-mantled ground squirrels are similar to changes in activity of respiratory neurons in unanaesthetized animals exposed to the same stimuli. HYPOXIA: As is observed in many fossorial species, moderate hypoxia (10%-15% oxygen) in the golden mantled ground squirrel leads to an increase in minute ventilation primarily due to increases in breathing frequency (McArthur and Milsom, 1991a, Walker et al., 1985, Holloway and Heath, 1984). This increase in breathing frequency is mostly due to decreases in the end expiratory pause, with decreases in the inspiratory interval contributing minimally. In non-fossorial animals such as the cat, moderate hypoxia (12%) induces an increase in minute ventilation due to increases in both breathing frequency and tidal volume and both of these components continue to increase as oxygen 105 concentration decreases further (Andronikou et al., 1988). These observations demonstrate the large variation in response to hypoxia that exists between different species. Hypoxia is sensed peripherally by the carotid and aortic bodies. Afferent projections from these sense organs synapse on cells located in the dorsomedial and ventrolateral medulla as well as the NTS and area postrema (Finley and Katz, 1992). Ventilatory responses to hypoxia in intact, unanaesthetized, animals is one of increased minute ventilation. However, upon carotid denervation in the unanaesthetized preparation, the results become less consistent with some studies illustrating increases in ventilation whilst others observe decreases in ventilation (see review by Neubauer et al. 1990). Consistent decreases in ventilation in response to hypoxia have been observed in the anaesthetized, denervated preparations (Neubauer et al., 1985). These apparent discrepancies between preparations have been ascribed to the different levels of control in respiration. Without carotid denervation there is an increase in ventilation as a result of increased excitatory input to the brain stem by the carotid bodies (McArthur and Milsom, 1991a, Walker et al., 1985, Holloway and Heath, 1984). However, in order to understand the variety of responses to hypoxia in the chemodenervated and/or anaesthetized preparation, the susceptibility of the rest of the CNS to hypoxia must first be understood. The cortex has been known to exhibit inhibitory descending influences on the respiratory centers of the brainstem, especially the rate facilitating centers of the diencephalon (Neubauer et al., 1990). In addition, the susceptibility of the CNS to hypoxia is in a rostral-caudal direction; therefore, the descending inhibitory influences of the cortex are the first to be lost upon hypoxic exposure and a resultant increased ventilation is observed 106 (Neubauer et al., 1990). Consequently, during mild and moderate hypoxia, excitatory responses are often observed; however, upon exposure to severe hypoxia respiratory depression is observed due to depression of the brainstem regions responsible for respiratory control. Anaesthesia is another factor that affects the sensitivity of the respiratory modulating areas of the brain. It affects both the diencephalon (respiratory rate facilitating center) as well as the cortex (descending inhibitory influences) equally and hence in the anaesthetized, chemodenervated animal one observes decreased ventilation in response to hypoxia (Neubauer et al., 1990). Carotid body intact Urethane anaesthetized animals were used in the present study. Based on the aforementioned theories of unanaesthetized (assuming urethane has no effects on respiratory related centers), carotid body intact animals, one would expect an increase in activity in the cells responsible for hypoxia induced increases in ventilation. A recent study by Dillon and Waldrop (1992) examining respiratory neurons using rat brain slices found that the majority of neurons within the VLM were directly excited by hypoxia. This suggests the possibility of centrally located hypoxic sensing cells. Despite this recent finding, it has been known for many years that most neurons within the CNS exhibit a depression in activity in response to hypoxic stimuli (Bjurstedt, 1946, Nolan and Waldrop, 1993). Nesland et al., (1965) were among the first to examine the effects of hypoxia on respiratory neurons, in the decerebrate anaesthetized cat. Their results were inconsistent, finding that inspiratory neurons exhibited no change in unit frequency or discharge per breath and expiratory neurons exhibited widely variable responses. St John performed 107 extensive studies (St. John and Bianchi, 1985, Bianchi and St. John, 1985, St. John and Wang, 1977) examining the effects of hypoxia and hypercapnia on respiratory neurons within the medulla of anaesthetized, decerebrate, cerebellectomized, paralyzed and ventilated cats. In anaesthetized cats with intact chemoreceptors St. John and Wang (1977) found that a number of the respiratory neurons recorded from decreased their activity in response to an hypoxic stimuli. Specifically 34% of the respiratory units had no increase in discharge per breath, additionally only 44%of those cells recorded from exhibited an increase in discharge frequency (spikes per minute). Upon carotid chemoreceptor denervation a generalized depression in the respiratory units was observed. St. John and Bianchi, (1985) also found that inspiratory neurons of the VRG exhibited increases in mean (spikes per minute) but not peak discharge, that phase spanning neurons exhibited increases in both peak and mean activity, and that expiratory neurons showed no change in activity in response to hypoxia. Folgering and Smolders who examined the paralyzed, vagotomized and anaesthetized cat obtained similar results. Their study found that frequency of 68% of the inspiratory neurons increased in response to hypoxia whereas only 57% of the expiratory cells increased their frequency. In a different study by England et al. (1995) examining (anaesthetized, paralyzed) intact and chemodenervated cats it was observed that decreases in firing rate resulted from hypoxic stimulation, with expiratory cells being the most sensitive to this stimulus. Similar results were obtained in the present study in which there was a significant (p=0.09) decrease in discharge per breath observed upon hypoxic exposure during state I in the golden mantled ground squirrel (Figure 12). Expiratory neurons and phase spanning neurons appeared to be the most sensitive to hypoxic exposure (Figure 13).There was no change 108 in peak neuron activity upon exposure to hypoxia during state I. Inconsistencies in response to hypoxia were found in the present study (Tables 1 and 2), with some neurons increasing their firing in response to hypoxia whereas others decreased their firing or remained unchanged. Perhaps these inconsistent results were due to some respiratory neurons having an increased sensitivity to the effects of hypoxia compared to others. It might also be possible that the neurons we recorded from were a mixture of "hypoxic sensing" neurons (as proposed by Waldrop and Nolan, 1992) and hypoxic sensitive neurons. If this were the case then it would be expected that some neurons would increase their discharge in response to hypoxia while others would decrease their discharge in response to hypoxia. Another possible explanation is the one which was originally proposed by St John et al. (1977) which suggested that there is a difference in the distribution of peripheral chemoreceptor afferents to the respiratory cells within the brainstem and therefore only certain cells would receive the excitatory influence of these afferents. Several mechanisms have been proposed to explain the cellular depression that occurs in response to hypoxia. These have been put forward in an attempt to explain the decreased ventilation observed in chemodenervated animals (see review by Neubauer et al., 1990). Higher brain centers have neurons which are stimulated by an hypoxic stimulus, these centers then send descending inhibitory influences to the rest of the brain. Endogenous opioid production has been linked to the respiratory depression observed in newborns in response to hypoxia since this depressant effect can be reversed by naloxone. Naloxone is an opioid mu receptor antagonist and therefore acts to reverse the depressant effects of opioids on the central nervous system (Bertram and Trevor, eds., 109 1995) The most promising mechanism proposed is one in which hypoxia causes a decrease in the amount of central excitatory neurotransmitters such as ACh, glutamate and aspartate. Finally, correlations have been made between respiratory depression and increases in the amount of adenosine and/or lactic acid. The effects of state III on the response of respiratory neurons to hypoxia was similar to that observed during exposure to hypercapnia. That is to say, both discharge per breath as well as peak activity was greater during state III compared to state I. Most mammals exhibit a decreased sensitivity to hypoxia during sleep (Phillipson and Bowes, 1986). However, studies by Hunter and Milsom (1998) on golden mantled ground squirrels and Pappenheimer (1977) on rats suggest that there is an increase in the sensitivity to hypoxia during state III and SWS respectively. These results might explain the increase in peak neuron activity during state III upon exposure to hypoxia in the present study. In summary, it would appear that there are inconsistencies between studies examining the effects of hypoxia on respiratory neurons in mammals. The responses of the respiratory neurons to hypoxia in the present study would appear to be consistent with some of the studies examining hypoxic effects on other animals. Therefore the current study would tend to support hypothesis #2. That is, changes in activity of respiratory neurons in response to hypoxia in the urethane anaesthetized golden-mantled ground squirrels are similar to the changes in activity of respiratory related neurons in the unanaesthetized animals exposed to the same stimuli. 110 n2 VALUE Characterization of the respiratory neuron has been primarily a qualitative one, in which cells are classified according to discharge pattern, the phase of the respiratory cycle in which they discharge and the location of the cell within the brainstem. Due to this subjective approach in classification, attempts were made to find a quantitative way to classify respiratory neurons. Initial attempts by Vibert et al. (1979) resulted in a respiratory modulation index (RMI). However, this statistic produced a large number of false positive results, i.e. non-respiratory neurons were classified as respiratory neurons. Subsequently, Orem and Dick (1983) developed a statistic called eta squared (r|2) which they used to quantify the size of the respiratory component of the cell. They define n 2 as "the proportion of the total variability in the activity of the cell that could be attributed to respiratory modulation" (Orem and Dick, 1983).Using r\2 it is possible to quantify how closely the cell's activity matches the respiratory cycle (for derivation of the statistic see Orem et al., 1985). The implications of this statistic are uncertain, and it is difficult (although tempting) to make assumptions about a respiratory cell's involvement in the rhythm generator based on its n 2 value. Orem has postulated that the higher the r| value the more likely it is that this cell has a central role in the rhythm generating process. Orem also comments on the immutability of this value for each cell, i.e. n is a constant unchanging characteristic of the cell and is stable within and between different arousal states. Ill Effects of arousal state: Studies examining respiratory neurons within the brainstem of unanaesthetized cats, found cells which were sensitive to changes in sleep state (Orem et al., 1974). These sleep sensitive cells tended to decrease their activity (or completely lose all their activity) upon entrance into SWS. In addition to the sleep sensitive cells there were also cells which exhibited little change in activity upon entrance into SWS, and hence these cells were termed sleep insensitive cells. In subsequent studies Orem found that cells which tended to be sleep sensitive were often those which had a lower n 2 value and conversely those which had higher n values were often sleep insensitive cells (Orem et al., 1985). According to Orem, the n value remains constant regardless of whether the cell is sleep sensitive or insensitive. In addition, Orem et al. (1985) determined that an inverse relationship existed between the n value of a cell and the amount that a cell decreased its activity in response to sleep. Therefore the lower the n 2 value the greater the sleep sensitivity of the cell. Orem and Dick (1983) further hypothesized that the r|2 value reflected one of two possibilities. First of all it could reflect the "endogenous" properties of the cell in question or conversely, it might represent the type of afferent input to the cell. That is, cells with high r\ values could have a greater proportion of respiratory related afferent input than cells with low r|2 values. Characteristics of cells with higher r)2 values included a more rapid attainment of peak discharge, an overall higher discharge rate throughout the respiratory cycle, and silence during the portion of the phase they were least active (Orem et al.,1985). In addition, Orem and Dick (1983) found that inspiratory 112 cells were more likely to have higher r\2 values than non-inspiratory (expiratory and phase spanning) cells, with 18/22 inspiratory cells classified as having "strong" n 2 values whereas only 4/10 non-inspiratory cells were classified as having "strong" r| values. In a study by Ioffe et al. (1992) using fetal sheep, the n 2 value was higher for inspiratory augmenting cells as compared to late inspiratory or inspiratory continuous cells, in addition expiratory decrementing cells had the highest r | 2 value of all the expiratory cells found. Unlike the studies by Orem (1983, 1985), Ioffe et al. (1992) observed changes in the n 2 value both within arousal states as well as between arousal states. It was suggested by Ioffe et al. that this discrepancy might be due to the immaturity of the central nervous system in the fetal sheep or because of the lack of afferent sensory input present in the fetal environment. In the present study the n 2 values for the entire group did not change significantly between state I and state III (Figure 18). Upon examining the responses of each class of cells (Figure 19) it would appear that the expiratory and EI cells increased their r\ values during state I whereas the inspiratory cells tended to have a reduced n 2 value during state I (Figure 19). However, upon examination of individual cells within each class there were no consistent responses among the inspiratory type cells between arousal states. The expiratory cells on the other hand (with the exception of one cell) exhibited consistent increases in n 2 value during state I (Table 3). These results are somewhat consistent with Orem's findings; in general n values remain constant from one state to the next. Looking at individual cells; however, it would appear that r | 2 values do change from one state to the next but the overall results don't illustrate this due to the fact that some cells exhibit an increase in r\ values during state I 113 whereas others exhibit a decrease. One possible explanation of this result is that each of the cells recorded from, regardless of whether they were of inspiratory or expiratory nature, had levels of afferent input that varied from one state to the next. Although each cell has a constant group of afferents which synapse onto it, the relative contribution of each of these afferents might change from one state to the next. Comparing table 3 with tables 1 and 2 it would not appear that the cells with the relatively higher n 2 value had less of a change in discharge rate from state I to state III compared to those with lower r\ values. It should be noted, though, that none of our cells could be classified as "weakly" respiratory since Orem's classification of a weak respiratory cell was one whose n 2 value was less than 0.3 during quiet wakefulness (state I)-In summary, it would appear that each cell exhibited a different n value depending upon the arousal state of the animal. Some cells exhibited an increase in n values where as others exhibited a decrease in r\2 values from one arousal state to the next, consequently, no overall changes were observed in r\ value when comparing state I to state III. Based on these results it would appear that the results of the present study are partly in agreement with hypothesis #3. That is, when looking at the entire population of cells as one group, activity levels change from state I to state III, however, the r\2 values do not change. If individual cells are examined, it is evident that within a class of cells 2 . there is no pattern which emerges with respect to whether or not the n value increases or decreases from one arousal state to the next. Each respiratory related cell would appear to have a different r\2 from one arousal state to the next and therefore one could postulate a unique distribution of respiratory related afferents to each cell. In addition, the afferents 114 to individual cells vary their input depending on the cortical arousal state of the animal. Effects of hypoxia and hypercapnia: In the present study there was a tendency for the n 2 value to decrease in response to hypoxic as well as hypercapnic stimuli in both state I and state III (Figure 20). These results are contrary to what we expected with an increase in chemoreceptor drive. In a study by Nattie et al., (1993) examining decerebrate, vagotomized, paralyzed and ventilated cats, an increase in n2 value was observed in respiratory cells upon exposure to a hypercapnic gas mixture. This suggested that the r|2 value is mutable, and that changes in respiratory drive elicit changes either in the afferent information to each respiratory related cell or changes in the rhytmicity of the respiratory cell itself. Half of the cats in Nattie's study were chemodenervated and the other half remained intact. Nine out of the ten respiratory neurons, in cats that had been chemodenervated, exhibited large increases in r\ values whereas 6 out of 10 respiratory neurons in cats which weren't chemodenervated, exhibited r\ values which changed little or even decreased upon exposure to hypercapnia. In the present study examining intact Urethane anaesthetized ground squirrels, each class of cell had a consistently lower r\2 value upon exposure to hypercapnia (Figure 22 and Table 3). These are similar to the results found by Nattie in intact cats exposed to a hypercapnic stimulus. These results suggest that r|2 may well be an endogenous property of the cells in the network and that any tonic input, even respiratory chemoreceptor input, introduces variation and reduces coupling. 115 n 2 values (Figure 20) also decreased during hypercapnia in state III. Larger decreases in r\2 values occurred in state III upon exposure to hypercapnia compared to state I. Studies examining the effects of sleep on dogs (Phillipson et al., 1978, Bowes et al, 1981) found decreases in the ventilation response to hypercapnia during sleep compared to wake. Many studies (see review by Phillipson and Bowes, 1986) have illustrated that there is a general decrease in the sensitivity to hypercapnia during sleep compared to wakefulness. In a recent study by Hunter and Milsom (1998), however, golden-mantled ground squirrels exhibited an increased sensitivity to hypercapnia during state III compared to state I. Assuming that the n 2 value is an endogenous property of each respiratory related cell then it would make sense that an increased sensitivity to hypercapnia during state III would imply a greater amount of exogenous respiratory related input to these cells. This would introduce a larger amount of variation to the endogenous rhythm and therefore decrease the cell's coupling to the respiratory cycle. In addition, perhaps the wakefulness stimulus plays a role in increasing the coupling of each cell's discharge to the respiratory cycle, therefore increasing the value during state I compared to state III. Hypoxia resulted in a similar trend, with most cells exhibiting decreases in their n 2 values during state III compared to state I. Studies examining the effects of hypoxia on a variety of mammals have a shown a general decrease in sensitivity to hypoxia during state III compared to state I (Phillipson and Bowes, 1986). In a study examining the effects of Urethane anaesthesia on respiratory control in golden-mantled ground squirrels (Hunter and Milsom, 1998), however, there was a general increase in sensitivity to hypoxia during state III relative to state I. This trend of increased sensitivity was also 116 observed during sleep in unanaesthetized ground squirrels. These observations might explain the larger decrease in n 2 during state III upon exposure to hypoxia compared to state I. That is, with an increased amount of respiratory related input to these cells during state III, an increased amount of variation is introduced and hence coupling of the cell's discharge to the respiratory cycle decreases. Again, an additional explanation might be that the so-called "wakefulness stimulus" somehow increases the coupling of the cells discharge to the respiratory cycle. In summary, it would appear that there is an overall decrease in the n value of respiratory neurons upon exposure to hypercapnia as well as hypoxia. In addition, this decrease in r\2 value is more profound during state III than state I. These results are not in agreement with hypothesis #4. That is, during exposure to hypercapnic stimuli respiratory neurons exhibited a decrease in n 2 value. Similarly during exposure to hypoxia the response of the respiratory neurons was primarily a decrease in n 2 value. Therefore, it would appear that the n value is not an immutable value as was suggested by Orem (Orem et al, 1974, Orem et al, 1983, Orem et al, 1985). 117 GENERAL CONCLUSIONS: Of interest in the current study was the response of respiratory neurons in the Urethane anaesthetized golden-mantled ground squirrel to changes in respiratory drive. Specifically, the effects of changes in arousal state upon respiratory neurons was examined. In addition, the response of respiratory neurons to hypercapnic and hypoxic stimuli was studied. Under Urethane anaesthesia changes in cortical activation state occur which appear analogous to those seen in the unanaesthetized animals. In addition, the respiratory changes associated with these changes in arousal state are also analogous to those seen in unanaesthetized animals. Consequently, it was hypothesized that the activity of respiratory neurons in the urethane anaesthetized ground squirrel would be analogous to those changes observed in unanaesthetized animals. The results indicate that the peak neuron activity was indeed significantly greater in state I compared to state III, which is consistent with studies by Orem examining changes in arousal state in unanaesthetized cats. Upon exosure to hypoxia or hypercapnia there was an increase in minute ventilation in the golden mantled ground squirrel. This increased respiratory drive would also be expected to result in increases in activity of respiratory neurons. Exposure to hypercapnia caused an increase in the peak neuron activity of respiratory related neurons, similar to those observed in other studies. This suggests that most respiratory neurons are either directly or indirectly excited by hypercapnic stimuli. Hypoxia, however, produced much 118 more inconsistent results, with some neurons increasing their activity upon exposure to hypoxia whereas others decreased their activity. There has also been a wide disparity in results in other studies examining the effects of hypoxia on respiratory neurons. Perhaps it is possible that there are two distinct populations of respiratory neurons in the medulla, those that are stimulated by hypoxia and those that are depressed by hypoxia. Interestingly, in a previous study by Hunter and Milsom (1998), an increase in sensitivity to hypercapnia and hypoxia was observed during state III compared to state I in Urethane anaesthetized squirrels. Other studies on rodents also found increases in sensitivity to hypercapnia and hypoxia during SWS and these results are contrary to the majority of studies examining the effects of state on sensitivity to hypoxia and hypercapnia in non-rodent species. Activity of respiratory neurons in response to hypoxia or hypercapnia, during state III were consistent with these observations of increased sensitivity. Consistent increases in activity were observed in response to hypercapnia and hypoxia during state III compared to state I. In addition to classifying each of the cells qualitatively, a quantitative method of classification was also employed. The method used is a statistical one, based on the ANOVA, and termed n 2 (eta squared), n 2 can be used to determine the strength and consistency of the respiratory activity to the respiratory cycle. It has been postulated that the r\2 value reflects the degree of afferent input (respiratory versus non-respiratory) to the respiratory cell. Consequently, one would expect changes in r\2 value with changes in . respiratory versus non-respiratory drive. Upon exposure to both hypoxia and hypercapnia the r)2 value decreased. This was an unexpected result, and differed from the opinion that the r]2 value is immutable, as was 119 proposed by Orem. In addition the n 2 value decreased to a greater extent during state III whether the animal was exposed to hypoxia or hypercapnia. This suggests an introduction of variation and reduction of coupling during state III compared to state I. Upon transition from state I to state III the r\2 value increased in many, but not all cells. These data suggest that while tonic chemoreceptor afferent input reduces the tightness of the coupling of neuron discharge to respiratory cycle phase, the wakefulness stimulus tends to increase the tightness of this coupling. The significance of this needs to be explored. 120 REFERENCES: Backman, S.B., Ballantyne, A.D., Rohrig, N., Mifflin, S., Jordan, D., Dickhaus, H., Spyer, K.M. and D.W. Richter. Evidence for a monosynaptic connection between slowly adapting pulmonary stretch receptor afferents and inspiratory beta neurones. Pfluger Archiv. 402:129-136. 1984. Bajic, J., Zuperku, E.J., Tonkovic-capin, M., and F.A. Hopp. Interaction between chemoreceptor and stretch receptor inputs at medullary respiratory neurons. American Journal of Physiology. 266:R1951-R1961. 1994. Batsel, H.L. Some functional properties of bulbar respiratory units. Experimental Neurology. 11:341-366, 1965. Bertrand, F. and Hugelin, A.. Respiratory synchronizing function of nucleus parabrachialis medialis: pneumotaxic mechanisms. Journal of Neurophysiology. 34:189-207. 1971. Bianchi, A.L., Denavit-Saubie, M. and J. Champagnat. Central control of breathing in mammals: Neuronal circuitry, membrane properties, and neurotransmeitters. Physiological Reviews. 75:1-45, 1995. 121 Boggs, D.F., Kilgore, Jr., and G.F. Birchard. Respiratory physiology of burrowing mammals and birds. Comparative Biochemistry and Physiology (B). 159:641-648. Bongianni, F., Corda, M., Fontanna, G.A. and T. Pantaleo. Reciprocal connections between rostral ventrolateral medulla and inspiration related medullary areas in the cat. Brain Research. 565:171-174. 1991. Bowes, G., Townsend, E.R., Kozar, L.F., Bromley, S.M., and E.A. Phillipson. Effect of carotid body denervation on arousal response in sleeping dogs. Journal of Applied Physiology. 51(l):40-45. 1981. Bradley, G.W. Control of Breathing. In: International Review of Physiology. Respiratory Physiology II, . J.G. Widdicombe ed. University Park Press. Baltimore.Volume 14, chapter 6, p. 185-217. Bruce, E.N., andN.S. Cherniack. Central chemoreceptors. Journal of Applied Physi ology. 62(2):389-402. 1987. Bryant, T.H., Yoshida, S., deCastro, D. and J. Lipski. Expiratory neurons of the Botzinger Complex in the rat: a morphological study following intracellular labeling with biocytin. The Journal of Comparative Neurology. 335:267-282. 1993. 122 Buzinska, K . , von Euler, C , Kao, F., Pantaleo, T. and Y . Yamamoto. Effects of graded focal cold block in rostral areas of the medulla. Acta Physiologica Scandinavica. 124:329-340. 1985. deCastro, D . , Lipski , J. and R. Kanjhan. Electrophysiological study of dorsal reapiratory neurons in the medulla oblongata of the rat. Brain Research. 639:49-56. 1994. Cherniack, N .S . , Edelman, N . H . and S. Lahiri . The effect of hypoxia and hypercapnia on respiratory neuron activity and cerebral aerobic metabolism. Chest. 59(5),supplement: 29s. 1971. Chitravanshi, V . C . , and H . N . Sapru. N M D A as well as n o n - N M D A receptors mediate the neurotransmission of inspiratory drive to phrenic motorneurons in the adult rat. Brain Research. 715:104-112. 1996. Cohen. M.I . . Comparison of phrenic and laryngeal motorneuron discharge patterns. Proceedings of the International Union of Physiological Sciences. 12:368. 1977. Cohen, M.I..Neurogenesis of respiratory rhythm in the mammal. Physiological Reviews. 59:1105-1173, 1979. Cohen, M.I . . Synchroniztion of discharge, spontaneous and evoked, between inspiratory neurons. ActaNeurobioligicaexperientiae. 33:189-218. 1973. 123 Connelly, C.A., Dobbins, E.G. and J. Feldman. Pre-Botzinger complex in cats: respiratory neuronal discharge patterns. Brain Research. 590:337-340. 1992. Cragg, P.A. and D.B. Drysdale. Interaction of hypoxia and hypercapnia on ventilation, tidal volume, and respiratory frequency in the anaesthetized rat. Journal of Physiology. 341:477-493. 1983. Darden, T.R.. Respiratory adaptations of a fossorial mammal, the pocket gopher (Thomomys bottae). Journal of Comparative Physiology. 78:121-137.1972. DeCastro, D., Lipski, J. and R. Kanjhan. Electrophysiological study of dorsal respiratory neurons in the medulla oblongata of the rat. Brain Research. 639:49-56. 1994. DiMarco, A.F., von Euler, C, Romaniuk, J.R. and Y. Yamamoto. Positive feedback facilitation of external intercostal and phrenic inspiratory activity by pulmonary stretch receptors. Acta Physiologica Scandinavia. 113:375-386. 1981. Dillon, G.H. and T.G. Waldrop. In vitro responses of caudal hypothalamic neurons to hypoxia and hypercapnia. Neuroscience. 51:941-950. 1992. Dobbins, E.G. and J .L. Feldman. Brainstem network controlling descending drive to phrenic motorneurons in rat. The Journal of Comparative Neurology. 347:64-86. 1994. 124 Duffin, J. and J. van Alphen. Bilateral connections from ventral group inspiratory neurons to phrenic motorneurons in the rat determined by cross-correlation. Brain Research. 694:55-60. 1995. Duffin, J., Ezure, K. and J. Lipski. Breathing rhythm generation: Focus on the rostral ventrolateral medulla. News in physiological Science. 10:133-140, 1995. Ellenberger, H.H. and J. Feldman. Brainstem Connections of the rostral ventral respiratory group of the rat. Brain Research. 513:35-42. 1990. von Euler, C. and U. Soderberg. Medullary chemosensitive receptors. Journal of Physiology. 118:545-554. 1952. Von Euler, C. The role of proprioceptive afferents in the control of respiratory muscles. Acta Neurobiologica Experientiae. 33: 329-341. 1973. von Euler, C. Brainstem mechanisms for generation and control of breathing patten. In: Handbook of Physiology. The respiratory system. Control of breathing. Washington D.C.: American physiological society, 1986, section 3, volume 2, chapter 1, p.1-67. Ezure, K., Tanaka, I. and Y. Oku. Location and axonal projection of early-onset decrementing expiratory neurons in the cat. Neuroscience Letters. 163:105-108. 1993. 125 Feldman, J.; Cohen, M.I. and P. Wolotsky. Powerful inhibition of pontine respiratory neurons by pulmonary afferent activity. Brain Research. 104:341-346. 1976. Feldman, J. and D.F. Speck. Interactions among inspiratory neurons in dorsal and ventral respiratory groups in cat medulls. Journal of Neurophysiology. 49(2):472-490. 1983. Feldman, J. Neurophysiology of breathing in mammals. In: Handbook of Physiology, Section 4: The Nervous System, Cherniack, N.S. and Widdicombe, J.G. (eds.) American Physiological Society, Bethesda, Maryland, pp. 463-524. Feldman, J., Smith, J.C., McCrimmon, D.R., Ellenberger, H.H., and D.F. Speck. Genreation of respiratory pattern in mammals. In: Neural Control of rhythmic movements in vertebrates. A. Cohen ed. John Wiley and Sons, Inc.. Chapter 3, p. 73-100, 1988. Fink, B.R. Influence of cerebral activity in wakefulness on regulation of breathing. Journal of Applied Physiology. 16:15-20. 1961. Fink, B.R., Hanks, A.C., Ngai, S.H., and M. Papper. Central regulation of respiration during anaesthesia and wakefulness. Annals New York Academy of Sciences. 109:892-900. 1963. 126 Finley, J.C.W. and D.M. Katz. The central organization of carotid body afferent projections to the brainstem of the rat. Brain Research. 572:108-116. 1992. Folgering, H. and F. Smolders. The steady state response of brainstem respiratory neuron activity to various levels of PaC02 and Pa02. Pflugers Archiv European Journal of Physiology. 383:9-17.1979. Gesell, R., Bricker, J. and C. Magee. Structural and functional organization of the central mechanisms controlling breathing. The American Journal of Physiology. 117:423-452. 1936. Gieroba, Z.J., Li, Y.W. and W.W. Blessing. Characteristics of caudal ventrolateral medullary neurons antidromically activated from rostral ventrolateral medulla in the rabbit. Brain Research. 582:196-207. 1992. Gilbey, M.P., Futuro-Neto, H.A., and S.Y. Zhou. Respiratory related discharge patterns of caudal raphe neurones projecting to the upper throacic spinal cord in the rat. Journal of the Autonomic Nervous System. 50:263-273. 1995. Gohgarian, H.G., Ellenberger, H.H. and J. Feldman. Bulbospinal respiratory neurons are a source of double synapses onto phrenic motorneurons following cervical spinal cord hemisection in adult rats. Brain Research. 600:169-173. 1993. 127 Grahn, D.A., Radeke, CM. and H.C. Heller. Arousal state vs. temperature effects on neuronal activity in subcoerulus area. American Journal of Physiology. 256:R840-R849. 1989. Haber, E., Kohn, K.W., Ngai, S.H., Holaday, D.A. and S.C. Wang. Localization of spontaneous respiratory neuronal activity in the medulla oblongata of the cat: A new location of the expiratory centre. American Jouranl of Physiology. 190:350-355. 1957 Harris, M.B. "The control of breathing in the golden-mantled ground squirrel (Spermophilus lateralis)." Ph.D. dissertation. University of British Columbia, 1998. Hayward, L.F. and R.B. Felder. Peripheral chemoreceptor inputs to the parabrachial nucleus of the rat. American Journal of Physiology. 268:R707-R714. 1995. Holloway, D.A. and A.G. Heath. Ventilatory changes in the golden hamster, Mesocricetus auratus, compared with the laboratory rat, Rattus norvegicus, during hypercapnia and/or hypoxia. Comparative Biochemistry and Physiology (A). 77:267-273. 1984. Housley, G.D., Martin-Body, R.L., Dawson, N.J. and J.D. Sinclair. Brainstem projections of the glossopharyngeal nerve and its carotid sinus branch in the rat. Neuroscience. 22(1):237-250. 1987. Hugelin, A. and M.I. Cohen. The reticular activating system and respiratory regulation in the cat. Annals New York Academy of Sciences. 109:587-603, 1963. 128 Ioffe, S., Jansen, A.H., and V. Chernick. Analysis of respiratory neuronal activity in fetal sheep. Journal of Applied Physiology. 73(5): 1972-1981. 1992. Jiang, C. and J. Lipski. Extensive monosynaptic inhibition of ventral respiratory group neurons by augmenting neurons in the Botzinger Complex. Experimental Brain Research. 81:639-648. 1990. Johnson, S.M., Smith, J.C., Funk, G.D. and J. Feldman. Pacemaker behaviour of respiratory neurons in medullary slices from neonatal rat. Journal of Neurophysiology. 72(6):2598-2608. 1994. Kalia, M.P.. Anatomical organization of central respiratory neurons. Annual Review of Physiology. 43:105-120. 1981. Kazuhisa, E., Manabe, M. and H. Yamada. Distribution of medullary respiratory neurons in the rat. Brain Research. 455:262-270. 1988. Krnjevic, K., Randic, M., and B.J. Siesjo. Cortical C02 tension and neuronal excitability. Journal of Physiology London. 176:105-122. 1965. 129 Lee, L.H., Freidman, D.B. and R. Lydic. Respiratory nuclei share synaptic connectivity with pontine reticular regions regulating REM sleep. American Journal of Physiology. 268:L251-L262. 1995. Lipski, J. Zhang, X., Kruszewska, B. and R. Kanjhan. Morphological study of long axonal projections of ventral medullary inspiratory neurons in the rat. Brain Research. 640:171-184.1994. Lipski, J. and E.G. Merill. Electrophysiological demonstration of the projection from expiratory neurones in rostral medulla to contralateral dorsal respiratory group. Brain Research. 197 521-524. 1980. Loeschcke, H.H. Central chemosensitivity and the reaction theory. Journal of Physiology. 332:1-24. 1982. Long, S. and J. Duffin. The neuronal determinants of respiratory rhythm. Progress in Neurobiology. 27:101-182. 1986. Maggi, CA. and A. Meli. Suitability of urethane anaesthesia for physiopharmacological investigations in various systems. Part I: General considerations. Experientia. 42(2): 109-210. 1986. 130 McArthur, M.D. and W.K. Milsom. Ventilation and respiratory sensitivity o f euthermic Columbian and golden-mantled ground squirrels (Spermophilus columbianus and Spermophilis lateralis) during the summer and winter. Physiological Zoology. 64(4):921-939. 1991. Merill, E.G.. The lateral respiratory neurones of the medulla: Their associations eith the nucleus ambiguus, nucleus retroambigualis, the spinal accessory nucleus and the spinal cord. Brain Research. 24:11-28. 1970. Merill, E.G.. Is there reciprocal inhibition between medullary inspiratory and expiratory neurones? In: Central nervous control mechanisms in breathing. C. von Euler and H. Lagercrantz eds. 1979. Merill, E.G., Lipski, J., Kubin, L. and L. Fedorko. Origin of expiratory inhibition of nucleus tractus solitarius inspiratory neurones. Brain Research. 263:43-50. 1983. Merill, E.G.. Where are the real respiratory neurons? Federation Proceedings. 40:2389-2394. 1981. Merill, E.G.. One source of the expiratory inhibition of phrenic motorneurons in the cat. Journal of Physiology. 332:79. 1982. 131 Mitchell, R.A. and D.A. Herbert. The effect of carbon dioxide on the membrane potential of medullary respiratory neurons. Brain Research. 75:345-349. 1974. Morillo, A.A., Nunez-Abades, P.A., Gaytan, S.P. and R. Pasario. Brainstem projections by axonal collaterals to the rostral and caudal ventral respiratory group in the rat. Brain Research Bulletin. 37(2):205-211. 1995. Nattie, E.E. and A. Li. Retrotrapezoid nucleus lesions decrease phrenic activity and C02 sensitivity in rats. Respiration Physiology. 97:63-77. 1994. Nesland, R.S.; Plum, F.; Nelson, J.R. and H.D. Siedler. The graded response to stimulation of medullary respiratory neurons. Experimental Neurology. 14:57-76. 1966. Neubauer, J.A., Santiago, T.V., Posner, M.A. and N.H. Edelman. Ventral medullary pH and ventilatory response to hyperperfusion and hypoxia. Jouranl of Applied Physiology. 58:1659-1668. 1985. Neubauer, J.A.; Melton, J.E.; and N.H. Edelmanan. Modulation of respiration during brain hypoxia. Journal of Applied Physiology. 68(2): 441-451. 1990. Nolan, P.C., and T.G. Waldrop. In vivo and in vitro responses of neurons in the ventrolateral medulla to hypoxia. Brain Research. 630:101-114. 1993. 132 Nunez-Abades, P. A., Morillo, A.M. and R. Pasaro. Brainstem connections of the ventral respiratory subgroups: afferent projections. Journal of the Autonomic Nervous System. 42:99-118. 1993. Okada, Y., Muckenhoff, K. and P. Scheid. Hypercapniua and medullary neurons in the isolated brain stem-spinal cord of the rat. Respiration Physiology. 93:327-336. 1993. Okada, Y., Muckenhoff, K., Holtermann, G., Acker, H., and P. Scheid. Depth profiles of pH and Po2 in the isolated brain stem-spinal cord of the neonatal rat. Respiration Physiology. 93:315-326. 1993. Oku, Y., Tanaka, I. and K. Ezure. Activity of bulbar respiratory neurones during fictive coughing and swallowing in the decerebrate cat. Journal of Physiology. 480(2):309-324. 1994. Onimaru, H., Arata, A. and I. Homma. Firing properties of respiratory rhythm generating neurons in the absence of synaptic transmission in rat medulla in viro. Experimental Brain Research. 76:530-536. 1989. Onimaru, H., Arata, A. and I. Homma. Intrinsic burst generation of preinspiratory neurons in the medulla of brainstem-spinal cord preparations isolated from newborn rats. Experimental Brain Research. 106:57-68, 1995. 133 Orem, J.. Breathing during sleep. In: Research topics in physiology. Regulation of ventilation and gas exchange. D.G. Davies and CD. Barnes eds. Academic Press, chapter 5, p. 131-165, 1978. Orem, J.. Neural basis of behavioural and state dependant control of breathing. Clinical Physiology of Sleep, pp.79-96. 1988. Orem, J.. Central respiratory activity in rapid eye movement sleep: augmenting and late inspiratory cells. Sleep. 17(8):665-673. 1994. Orem, J.. The nature of the wakefulness stimulus for breathing. Sleep and Respiration. pp23-31. 1990. Orem, J., Netick, A. and W.C Dement. Breathing during sleep and wakefulness in the cat. Respiration Physiology. 30:265-289. 1977. Orem, J.. Behaviour of respiatory cells during NREM sleep. In: Brain Mechanisms of Sleep. D.J. McGinty et al. eds. Raven Press. New York. 341-359. 1985. Orem, J.. Medullary respiratory neuron activity: relationship to tonic and phasic REM sleep. Journal of Applied Physiology: Respiratory Environmental Exercise Physiology. 48:54-65. 1980. 134 Orem, J., and T. Dick. Consistency and signal strength of respiratory neuronal activity. Journal of Neurophysiology. 50(5): 1098-1107. 1983. Orem, J., Osario, I., Brooks, E. and T. Dick. Activity of Respiratory Neurons During NREM Sleep. Journal of Neurophysiology. 54(5): 1144-1156. 1985. Orem, J. and R. Trotter. Postinspiratory neuronal activities during behavioural control, sleep and wakefulness. Journal of Applied Physiology. 72(6):2369-2377. 1992. Orem, J.. Respiratory neuronal activity in sleep. In:Breathing disorders of sleep. N.H. Edelmen and T.V. Santiago eds. Churchill Livingstone Inc., New York, chapter 2, p. 19-44, 1986. Orem, J., Montplaisir, J. and W.C. Dement. Changes in activity of respiratory neurons during sleep. Brain Research. 82:309-315. 1974. Orem, J. and R. Trotter. Behavioural control of breathing. News in Physiological Sciences. 9:228-232.1994. Orem, J. . Brainstem respiratory neurons and their control during various behaviours. In: Brainstem mechanisms of behaviour. W.R. Klemm and R.P. Vertes eds. John Wiley and Sons, Inc.. p.383-406. 1990. 135 Orem, J.. Behavioural inspiratory inhibition: inactivated and activated respiratory cells. Journal of Neurophysiology. 62:1069-1077. 1989. Orem, J. and E . G . Brooks. The activity of retrofacial expiratory cells during behavioural respiratory responses and active expiration. Brain Research. 374:409-412. 1986. Orem, J. Inspiratory neurons that are activated when inspiration is inhibited behaviourally. Neuroscience Letters. 83:282-286. 1987. Orem, J. and A . Netick. Behavioural control of breathing in the cat. Brain Research. 366:238-253. 1986. Otake, K . , Sasaki, H . , Ezure, K . and M.Manabe. Medullary projections of nonaugmenting inspiratory neurons of the ventrolateral medulla in the cat. The Journal of Comparative Neurology. 302:485-499. 1990. Pappenheimer, J.R. Sleep and respiration of rats during hypoxia. Journal of Physiology London. 266:191-207. 1977. Parkes, M . J . , Lara-Munoz, J.P., Izzo, P . N . and K . M . Spyer. Response of ventral respiratory neurones in the rat to vagus stimulation and the functional division of expiration. Journal of Physiology. 476(1): 131 -139. 1994. 136 Paton, J.F., Ramirez, J . M . and D. Richter. Functionally intact in vitro preparation generating respiratory activity in neonatal and mature mammals. Pflugers Archiv. 428:250-260, 1994. Paxinos, G. , and C.Watson. The rat brain in Stereotaxic Coordinates, 2 n d edition. Academic Press, Sydney. 1986. Peck, A . L . Sensory inputs to the medulla. Annual Review of Physiology. 43:73-90. 1981. Phillipson, E . A . and G . Bowes. Control of breathing during sleep. In: Handbook of Physiology, Section 3: The Respiratory System, Volume 2, Control of Breathing, part II, Fishman, A . P . , Geiger, S.R., Cherniack, N .S . and J.G. Widdicombe (eds.). Amercian Physiological Society, Bethesda, Maryland, pp.649-689. 1986. Phillipson, E . A . , Sullivan, D.J .C. , Read, E . , Murphy, E . and L . F . Kozar. Ventilatory and Waking responses to hypoxia in sleeping dogs. Journal of Applied Physiology. 44:512-520. 1978. Phillipson, E . A . , Murphy, E . , and L . F . Kozar. Regulation of respiration in sleeping dogs. Journal of Applied Physiology. 40:688-693. 1976. Pilowsky, P., Llewellyn-Smith, I.J., Lipski , J., Minson, J., Arnolda, L . and J. Chalmers. Projections from the inspiratory neurons of the ventral respiratory group to the subretrofacial nucleus of the cat. Brain Research. 633:63-71. 1994. 1 3 7 Puizillot, J.J., and J.P. Ternaux. Variations d'activite toniques, phasique et respatoires, au niveau bulbaire pendant rendormement de la preparation 'encephale isole'. Brain Research. 66:67. 1974. Remmers, J.E., Bartlett, D. and M.D. Putnam. Changes in the respiratory cycle associated with sleep. Respiration Physiology. 28:227-238.1976. Reschtschaffen, A. , Kales, A. , Berger, R.J., Dement, W.C., Jacobson, A. , Johnson, L . C , Jouvet, M . . , Monroe, L.J. , Oswald, I., Roffward, H.P., Roth, B. and R.D. Walter. A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects. U.S. Government Printing Office, Washington, D.C. Richter, D.W. Generation and maintenance of the respiratory rhythm. Journal of Experimental Biology. 100:93-107, 1982. Richter, D.W., Hyde, F. and M . Gabriel. Intracellular recordings from different types of medullary respiratory neurons of the cat. Journal of Neurophysiology. 38:1162-1171. 1975. Saether, K., Hilaire, G., and R. Monteau. Dorsal and ventral respiratory groups of neurons in the medulla of the rat. Brain Research. 419:87-96. 1987. 138 Salmoiraghi, G.C. and R. vonBaumgarten. Intracellular potentials from respiratory neurones in brainstem of cat and mechanism of rhythmic respiration. Journal of Neurophysiology. 24:203-218. 1961. Schwarzacher, S.W., Smith, J.C. and D.W. Richter. Pre-Botzinger complex in the cat. Journal of Neurophysiology. 73(4): 1452-1461. 1995. Schwarzacher, S.W. Wilhelm, Z., Anders, K. and D.W. Richter. The medullary network in the rat. Journal of Physiology. 435:631-644. 1991. Scharzacher, S.W., Smith, J.C. and D.W. Richter. Respiratory neurons in the pre-Botzinger region of cats. Pflugers Archives. 418(1):R17. 1991. Schlaefke, M.E. Central chemosensitivity- a respiratory drive. Review of Physiology, Biochemistry and Pharmacology. 90:171-244. 1981. Sherrey, J.H. and D. Merigan. State dependance of upper airway respiratory motorneurons: Function of the cricothyroid and Nasolabial muscles of the unanaesthetized rat. Electroencephalography and Clinical Neurophysiology. 43:218-228. 1977. Smith, J., Ellenberger, H.H., Ballanyi, K . , Richter, D.W. and J. Feldman. Pre-Botzinger Complex: A Brainstem region that may generate respiratory rhythm in mammals. Science. 254:726-729, 1991. 139 Smith, J., Morrison, D.E., Ellenberger, H.H., Otto, M.R. and J.L. Feldman. Brainstem projections to the major neuron populations in the medulla of the cat. The Journal of Comparative Neurology. 281:69-96. 1989. Smith, J.C., Greer, J.J., Liu, G. and J. Feldman. Neural Mechanisms generating respiratory pattern in mammalian brainstem spinal cord in vitro I. Spatiotemporal patterns of motor and medullary neuron activity. Journal of Neurophysiology. 64(4): 1149-1169. 1990. St. John, W. and Wang. Response of medullary respiratory neurons to hypercapnia and isocapnic hypoxia. Journal of Applied Physiology. 43(5):812-821. 1977. St. John, W. Central Nervous System Regulation of Ventilation. In: Research Topics in Physiology. Volume 1: Regulation of Ventilation and Gas exchange. D. Davies and C. Barnes eds. Academic Press. New York. 1978. St. John, W. Differential alteration by hypercapnia and hypoxia of the apneustic respiratory pattern in decerebrate cats. Journal of Physiology London. 287:467-491.1979. St. John, W. Respiratory neuron responses to hypercapnia and carotid chemoreceptor stimulation. Journal of Applied Physiology. 51(4):816-822. 1981. 140 St. John, W. and A.L. Bianchi. Responses of bulbospinal and laryngeal respiratory neurons to hypercapnia and hypoxia. Journal of Applied Physiology. 59(4): 1201-1207.1985. Takeda, R., and A. Haji. Effects of halothane on membrane potential and discharge activity in pairs of bulbar respiratory neurons of decerebrate cats. Neuropharmacology. 31(10):1049-1058. 1992. Walker, B.R., Adams, E.M., and N.F. Voelkel. Ventilatory responses of hamsters and rats to hypoxia and hypercapnia. Journal of Applied Physiology. 59(6): 1955-1960. Webb, CL. Aspects of the control of breathing in the golden-mantled ground squirrel. M.Sc. Dissertation. University of British Columbia. 1987. Zheng, Y., Barillot, J.C. and A.L. Bianchi. Patterns of membrane potentials and distributions of the medullary respiratory neurons in the decerebrate rat. Brain Research. 546:261-270. 1991. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0089060/manifest

Comment

Related Items