UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Changes in ventilatory responses within and between hibernation bouts in Spermophilus Lateralis the golden-mantled… Harris, Michael B. 1992

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

Item Metadata

Download

Media
831-ubc_1992_spring_harris_michael.pdf [ 3.21MB ]
Metadata
JSON: 831-1.0086474.json
JSON-LD: 831-1.0086474-ld.json
RDF/XML (Pretty): 831-1.0086474-rdf.xml
RDF/JSON: 831-1.0086474-rdf.json
Turtle: 831-1.0086474-turtle.txt
N-Triples: 831-1.0086474-rdf-ntriples.txt
Original Record: 831-1.0086474-source.json
Full Text
831-1.0086474-fulltext.txt
Citation
831-1.0086474.ris

Full Text

CHANGES IN VENTILATORY RESPONSES WITHIN AND BETWEEN HIBERNATION BOUTS IN SPERMOPHILUS LATERALIS THE GOLDEN-MANTLED GROUND SQUIRREL by  Michael B. Harris B.Sc. (Hons.), University of British Columbia, 1990.  A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR A DEGREE OF MASTER OF SCIENCE in  THE FACULTY OF GRADUATE STUDIES Department of Zoology  We accept this thesis as conforming to the required standard  [HE UNIVERSITY OF BRITISH COLUMBIA  APRIL, 1992 © Michael B. Harris  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature  Zoology Department of The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  April 22, 1992  ii ABSTRACT For most hibernating mammals, the hibernation season is composed of bouts of hibernation of various lengths punctuated by short periods of arousal. Hibernation depth fluctuates during each bout, and possibly over the season, even when ambient and body temperatures remain constant. This suggests that physiological changes associated with hibernation may also vary in a similar fashion. Investigations of central control of ventilation, however, have assumed that the changes seen in ventilatory responses during hibernation are consistent at any given body temperature throughout the hibernation season. To test whether this assumption is valid, the hypercapnic ventilatory response of hibernating golden mantled ground squirrels was characterized in individual animals throughout bouts of hibernation at different times in the hibernation season. The overall level of ventilation produced by an animal is normally expressed as the total amount of air moved in and out of the lung per minute (minute ventilation, VE). This variable is a function of the volume of each breath (tidal volume, V T) and the number of breaths taken per minute (respiratory frequency, RI). It was found that the hypercapnic ventilatory response during hibernation consisted almost exclusively of increases in Rf as levels of ambient CO 2 were increased, while VT changed very little. Thus, it was determined that the overall ventilatory response could be adequately described by changes in breathing frequency alone. The hypercapnic ventilatory response was complex and could be characterized by the changes occuring in ventilation over four specific ranges (regions) of inspired CO 2 . In the range from 0 to 2 % CO 2 , Rf appeared independent of changes in inspired CO2 . In the range from 2 to 6 % CO2 , Rf increased with increasing levels of inspired CO 2 . In the third range, beginning at approximately 6 % CO 2 , Rf remained constant or decreased slightly as levels of  iii inspired CO 2 increased further. The fourth range began at a level of inspired CO 2 which not only promoted a further increase in V'E but also prompted the animals to arouse from hibernation (7 to 13 %). No significant inter or intra-bout variations were observed in any aspect of the hypercapnic ventilatory response. Although no significant variation was found in the level of CO2 which prompted arousal (arousal threshold) between bouts of hibernation, arousal thresholds did appear to increase from early to mid season and decrease from mid to late season. This was similar to the pattern of variation observed in the length of hibernation bouts during the season. It is possible that changes in both arousal threshold and bout length are associated with changes in hibernation depth and that there is a regular pattern of seasonal variation in hibernation depth. These data indicate, however, that ventilatory sensitivity is exclusively a function of body temperature, regardless of possible temperature-independent changes in hibernation depth.  iv TABLE OF CONTENTS ABSTRACT ^  ii  TABLE OF CONTENTS ^  iv  LIST OF TABLES ^  vi  LIST OF FIGURES ^  vii  ACKNOWLEDGEMENTS ^  ix  INTRODUCTION ^ Ventilatory Control ^ Hibernation ^  1 1 6  MATERIALS AND METHODS ^ 13 Animals ^ 13 Surgical Procedures ^ 14 Apparatus ^ 17 Experimental Protocol ^ 26 Inter-bout tests ^26 Intra-bout tests ^27 Data Analysis ^ 27 Inter- and Intra-bout comparison of Ventilatory Responses ^ 29 RESULTS ^ Hypercapnic Ventilatory Response ^ Components of the Response ^ Respiratory Frequency ^ Tidal Volume ^ Minute Ventilation ^ Relative Contribution of Changes in VT (Ra) and Rf to the Ventilatory Response ^ Ventilatory Pattern ^  31 31 31 31 31 31 36 36  36 Overall Shape of the Ventilatory Response ^ Comparison of the Ventilatory Response between Bouts over the Season ^37 Comparison of the Ventilatory Response between Days over a 52 Hibernation Bout ^ Pattern of Hibernation ^ Changes in Body Temperature over the Hibernation Season ^ Changes in Bout Length over the Season ^ Changes in the Arousal Threshold for Hibernation over the Season ^  52 52 52 65  V  DISCUSSION ^ Hypercapnic Ventilatory Response ^ Role of Changes in Rf and VT in the Overall Hypercapnic Ventilatory Response ^ Shape of the Ventilatory Response curve ^ Region 1 ^ Region 2 ^ Region 3 ^ Region 4 ^ Pattern of Hibernation ^ Seasonal Changes in Body Temperature ^ Seasonal Changes in Bout Length ^ Seasonal Changes in Arousal Threshold & Hibernation Depth ^  70 70 70 74 74 78 79 81 83 83 84 84  CONCLUSION ^  86  LITERATURE CITED ^  88  vi LIST OF TABLES  Table 1:^Levels of inspired CO 2 which prompted arousal from hibernation in animals tested during the early, mid and late portions of the hibernation season . 66  vii LIST OF FIGURES Figure 1:^A representation of ventilatory control ^  2  Figure 2a:^Experimental apparatus used during the first year ^ 20 Figure 2b:^Experimental apparatus used during the second year ^ 22 Figure 2c:^Experimental apparatus used during the third year ^ 24 Figure 3a:^Typical hypercapnic ventilatory response of one animal, characterized by changes in respiratory frequency (Rf), tidal volume (V T), and minute ventilation (VE), associated with changes in inspired CO 2 . ^ 32 Figure 3b:^Typical hypercapnic ventilatory response of another animal, characterized by changes in respiratory frequency (Rf), respiratory amplitude (Ra) and an estimate of minute ventilation (VEST ), associated with changes in inspired CO 2 . ^  34  Figure 4a:^Relationship between minute ventilation and respiratory frequency ^ 38 Figure 4b.^Relationship between minute ventilation and tidal volume ^ 40 Figure 5:^Breathing volume traces illustrating the ventilatory patterns observed at various level of inspired CO2 ^  42  Figure 6:^Schematic diagram of the hypercapnic ventilatory response ^ 44 Figure 7a:^Typical hypercapnic ventilatory response ^  46  Figure 7b:^Typical hypercapnic ventilatory response of an individual animal (# 7), observed during three repeated tests over the hibernation season ^ 48 Figure 8:^Mean hypercapnic ventilatory response of all animals, separated into either early, mid or late season categories and expressed as both relative and absolute values of Rf at increasing levels of inspired CO 2 . ^ 50 Figure 9a:^Hypercapnic ventilatory response of an individual animal (# 7),  observed during five consecutive days of an individual hibernation bout. Values for Rf, V T and VE are presented as actual values. ^ 53  Figure 9b:^Hypercapnic ventilatory response of an individual animal (# 7), observed during five consecutive days of an individual hibernation bout. Values for RI, VT and VE are presented as relative values, expressed as a % of those observed during normocapnia. ^  55  viii Figure 10a: Hypercapnic ventilatory response of an individual animal (# 8), observed during five consecutive days of an individual hibernation bout. Values for Rf, VT and VE are presented as actual values. ^ 57 Figure 10b: Hypercapnic ventilatory response of an individual animal (# 8), observed during five consecutive days of an individual hibernation bout. Values for Rf, V T and VE are presented as relative values, expressed as a % of those observed during normocapnia. ^  59  Figure 11:^Seasonal fluctuations in core body temperature ^ 61 Figure 12:^Mean pattern of bout length variation ^  63  Figure 13:^Individual (open circles) and mean levels (histograms) of inspired CO2 which initiated arousal from hibernation during early, mid and late periods of the hibernation season ^  68  Figure 14^Schematic diagram outlining movement of the chest wall resulting from the inspiratory activity of the intercostal muscles or diaphragm ^ 72 Figure 15:^Plot of the hypercapnic ventilatory response profile of man illustrating the effect that changes in ambient 02 have on the shape of this response ^ 76  ix  ACKNOWLEDGEMENTS I would like to express my sincere appreciation to Dr. W. K. Milsom for his encouragement, assistance and guidance throughout this investigation and my evolution as a student. His lessons, not limited to physiology and the art of experimental design, have proven both enlightening and invaluable. His wisdom, good humour, and uncanny ability to transmit competence and mechanical healing through his very touch, continue to leave me in awe. I would like to thank Dr. D.R. Jones, Dr. J.D. Steeves, Dr. D.J. Randall, Dr. T. H. Carfoot and their respective graduate students for their advice and the use of their facilities, resources and equipment, although some may not know that I have ever done so. I would like to thank Mark Burleson, Richard Kinkead, Supriti Bharma, Pat Chan, Julie Hunter, Rhonda Garland, Jean Macleod, Trey, Cristie, Rohan and all present and past members of the Milsom lab for providing me with advice, compassion and well needed distractions from my otherwise mundane life. And I would especially like to thank Barbara Taylor, for her support and willingness to put up with me for the rest of our lives O. I am grateful to the university for providing me the opportunity to do this research, through teaching assistanceships and scholarships, and to the NSERC of Canada which has supported my work via an operating grant to Dr. W.K. Milsom.  1 INTRODUCTION Ventilatory Control: Ventilation is the process by which air is moved into and out of the lung. This process provides the convection necessary for the efficient transfer of 0 2 and CO2 between blood and air across the lung alveoli. Ventilation comprises the initial step in the overall process by which 0 2 and CO2 are exchanged between air and mitochondria, commonly referred to as respiration. The overall respiratory process can generate considerable metabolic costs. To maximize efficiency each step in this process must be tightly controlled. Ventilation is therefore matched to activity level and metabolic demand by a multi-faceted control system, a representation of which is shown in figure 1. This system accepts input from various receptors in the periphery and integrates it with information descending from higher neural centres. Both types of input impinge on a ventilatory control "centre" from which respiratory drive is initiated. In actuality this ventilatory control "centre" may comprise a number of pontine and medullary regions which function to produce both ventilatory rhythm and pattern (Feldman et al., 1990). Partial control of ventilation is achieved reflexly through peripheral feedback from chemoreceptors. This portion of the system allows the adjustment of respiration to regulate the partial pressures of 0 2 and CO2 , as well as pH levels, in the blood and body fluids. These variables are internally monitored by central and peripheral chemoreceptors which feed back information to the central ventilatory control centres which produce appropriate changes in breathing (Tenney & Boggs, 1986; Feldman, 1986). To illustrate this feedback control consider respiration in an environment high in CO 2 (hypercarbia). Under this condition, the diffusion gradient for CO 2 from blood to alveolar gas is reduced and "normal" respiration  2  Figure 1:^A representation of ventilatory control. Control of ventilation occurs via both feed forward and feed back control mechanisms, which act in concert to modify the respiratory motor activity generated in the central respiratory centres.  3  CONTROL OF VENTILATION  HIGHER CENTERS  I  ^ MECHANORECEPTORS 46----7 1 CEN7'RAL RESPIRATORY ^ ■ RESPIRATORY CENTERS^ MUSCLES 1 VE it CHEMORECEPTORS 4---- 0 2 , CO2 , pH OTHER INPUTS  4 will result in CO2 retention and an increase in the partial pressure of CO 2 (Pco2) in the blood (hypercapnia). This increase is detected by chemoreceptors which stimulate the ventilatory control centres and results in an increase in ventilatory drive. The resulting increase in ventilation will return blood gas tensions toward their resting levels (Tenney & Boggs, 1986; Feldman, 1986). Thus, chemoreceptor activity will match ventilatory drive to the artificially induced ventilatory demand. Control of ventilation can also be influenced by other peripheral receptors. Some examples of these are; pulmonary stretch receptors, upper airway mechanoreceptors and postural muscle proprioceptors. These receptor groups all appear to have a feed-back type influence on breathing. Working in conjunction with these peripheral inputs are central mechanisms through which the higher centres of the brain can have a descending effect on the ventilatory control centres. It is by the activation and deactivation of these higher centres that changes in ventilation may result from changes in central state such as occur during sleep and hibernation. For example, Grahn and Heller (1989) have observed that changes in cortical activity during changes in sleep state are correlated with changes in the activity of specific neurons in the ventromedial medulla, a region involved in ventilatory control. The overall system controlling ventilation is extremely well tuned. For example, during the initiation of exercise, increases in ventilation can match, and in some instances exceed, the increases in metabolic demand such that blood-gas tensions do not appear to change from pre-exercise levels. This is the result of the intricate combination of feed forward and feed back control involving extremely sensitive chemoresponses, ventilatory stimulation from proprioceptors in the exercising limbs and central coactivation of both exercise and ventilatory motor drives.  5 Unlike peripheral control mechanisms, central control mechanisms cannot be easily manipulated. Surgical removal of descending information, through successive lesions, ablations, or cold block does provide valuable information, but an understanding of the more subtle aspects of central control requires more indirect study. Insight into how the system integrates peripheral and central input may be gained by observing how the system responds to peripheral stimuli during different levels of central activation; such as during sleep and hibernation. Sleeping and hibernating animals exhibit markedly different ventilatory patterns and responses to peripheral stimulation than they do during wakefulness. This is indicative of distinct changes in the integration of central and peripheral inputs effecting the respiratory centres (Remmers, 1981; McArthur and Milsom,1990b). Examples of the "state" dependence of breathing pattern are numerous, including the innocuous as well as a selection ranging from annoying to pathological. The slowing and deepening of breathing with the entrance into sleep, is a state dependant reaction characterizing the relationship between the sleep/wake cycle and the central control of breathing (Feldman, 1986). Simple snoring can involve a state dependant decrease in the tone of the upper airway musculature which results in an increased upper airway resistance. This condition can become more serious if the selective muscle atonia of REM-sleep results in total airway occlusion; referred to as obstructive sleep apnoea syndrome (OSAS) (Stoohs and Guilleminault, 1990). Many human breathing disorders tend to manifest themselves during specific arousal states. For this reason, such disorders could be considered as statedependant aberrations of breathing pattern.  6 Hibernation: There is a large body of evidence suggesting that the physiological changes which occur during hibernation are an extreme extension of those which occur during slow wave sleep (Berger, 1984; Heller, et al., 1978; Krilowicz et a/.1988). During hibernation, S.lateralis exhibits a ventilatory pattern in which episodes of breathing are separated by periods of apnoea. Similar breathing patterns have been documented in lower vertebrates (Milsom, 1990). Among other things, this homology has suggested a trans-specific and perhaps phylogenetic consistency in the mechanisms underlying ventilatory control (Milsom 1991; Smatresk, 1990). The state-dependent appearances of periods of apnoea has also been associated with human pathological disorders such as central sleep apnoea syndrome (CSAS) and sudden infant death syndrome (SIDS), both of which may be related to faults in the ventilatory control system (Pearson and Greenaway, 1990; Onal, 1988; Williams et al, 1991). If hibernation is considered an extension of slow wave sleep and, if hibernation is to be used as a model to study central/peripheral ventilatory interactions with implications for other changes in central state, then it is important to investigate the uniformity of hibernation as a state. Hibernation is a state characterized by drastically reduced body temperature (T B ) and metabolic rate (MR). However, hibernation differs from other states, such as shallow torpor and hypothermia, which also share these characteristics. During torpor, the set point for  TB regulation  falls. This may be only a slight drop or  it can be as great as 15-20° C (Heller et al, 1978). Many birds, and mammals such as bears and chipmunks, do not truly hibernate but go into a shallow torpor. During true hibernation (deep torpor), an animal will regulate its body temperature at a level only slightly above ambient temperature (Ta), down to levels approaching 0° C.  7 Hibernation and torpidity differ from hypothermia in that the observed reductions in ; are actively regulated and the animal is capable of rewarming itself using only internally generated (endogenous) heat. Hypothermia on the other hand, is an abnormal condition in which TB is decreased passively (when heat loss is greater than heat production) and can only be re-elevated by heat from exogenous sources (Lyman, 1982b). Becoming torpid or hibernating are adaptive strategies, which provide a means for an animal to reduce its metabolism, and therefore its energy expenditure, under inhospitable conditions. This reduction is particularly important for smaller animals which, due to their greater surface area to volume ratio, must expend a greater proportion of their energy to maintain a high body temperature than would larger animals (Lyman, 1982b). For a small animal, this savings during hibernation can be as high as an 80% reduction in energy expenditure over a winter (Wang, 1979). Hibernation is a seasonal phenomenon and may be considered an extension of torpidity. Shallow torpor is usually a daily event, taking place during times of inactivity or sleep, but may also extend to multi-day periods depending on environmental conditions. Seasonal hibernation consists of repeated periods (bouts) of deep torpor each of which may last for days or weeks at a time (Lyman, 1982a). Animals, such as hummingbirds or pocket mice, enter shallow or moderate torpor only when their environment becomes too severe for them to maintain their normal body temperatures (Heller et al., 1978). Such behaviour could be thought of as "facultative". However, true hibernators have developed a stronger dependence on this strategy and spend a specific phase of the year in hibernation. This behaviour could be considered "obligate", as the life histories of these animals are dependant on the hibernation phase. Such animals are referred to as seasonal hibernators. S.lateralis is a seasonal hibernator whose natural obligation to a hibernating condition usually lasts seven  8 months of each year (McKeever,1964; Geiser et a1.,1990). All seasonal hibernators periodically arouse during hibernation. Their  Ts  and MR  return to euthermic levels for a short time, after which they reenter hibernation (Willis, 1982). Thus the hibernation season is divided into periods or bouts of hibernation, punctuated by short periods of spontaneous arousal. Over the hibernation season most animals exhibit a characteristic variation in hibernation bout length. The bouts become progressively longer from the onset of hibernation to approximately mid season, and progressively shorter again as the hibernation season ends. These phenomena have been well documented in S.lateralis and other seasonal hibernators (Pengelley and Fisher, 1961; Strumwasser et al., 1964; Twente and Twente, 1968). As mentioned previously, a reduction in temperature and metabolism during hibernation affords the hibernating animal considerable metabolic savings. However the costs involved with rewarming against a considerable temperature gradient, are themselves considerable. For example, in a hibernating dormouse, Kayser (1953) calculated the cost of a single arousal to be the metabolic equivalent of ten days of hibernation. An obvious question, therefore, is why does an animal living on limited stored resources bother to invest in periodically rewarming. Why not simply stay at a reduced temperature for the entire winter and maximise savings? It would seem that, while temporarily reaping the benefits of the reduced temperature and metabolic demand of a poikilotherm, hibernators are still somehow tied to their homeothermic origins. For example, there may be some critical temperature dependant biochemical reaction, perhaps facilitated by a temperature dependant enzyme, that can only take place at elevated body temperatures. Such a situation would necessitate periodic rewarming. The actual reasons for spontaneous arousal are still in the realm of speculation. Discussion of the necessities and causes are beyond the scope of this  9 investigation and are better left to a comprehensive review (see Willis 1982). However it is safe to assume that these spontaneous periodic arousals are a physiological necessity. Although some hibernators do store food for consumption during these periods of arousal, many (including S. lateralis) do not feed during arousals and therefore cannot replenish their internal energy stores. It follows that, whatever the cause of periodic arousals, a hibernator utilizing only stored fat during a long hibernation season would maximize its energy savings by employing hibernation bouts which are as long as possible, thereby minimizing costly arousals. Why then do these animals exhibit shorter and, presumably, less efficient bouts during early and late season? One assumes it to be a function of the control system underlying hibernation, however this remains speculative, as the overall system controlling hibernation remains a mystery. Many researchers have hypothesized mechanisms which might underlie periodic arousal. Very little study has been devoted, however, toward identification or quantification of changes in hibernation depth which might occur during a bout of hibernation between periods of arousal, or between bouts of hibernation over a season. This task was first undertaken by Twente and Twente (1967 & 1968), with their studies of progressive irritability and seasonal variation in hibernation behaviour. The Twentes found that within a bout of hibernation, animals showed a progressive increase in sensitivity to arousal by external stimuli. They interpreted the animal's propensity for arousal due to a given stimulus (its irritability), to be a function of the depth at which the animal was hibernating. Thus hibernation depth could be measured by the animals irritability. It followed that the progressive irritability during a hibernation bout described by Twente and Twente (1968) was indicative of a gradual decrease in hibernation depth as a bout of hibernation progressed. Twente and Twente (1967) also observed that over a season the animal's  10 "arousability" tended to fluctuate. Their work concentrated, however, on the changes occurring within a bout and their initial observation of changes occurring over a season were considered no further. Based on this information, there would seem to be three distinct characteristics of seasonal hibernation: (1) Hibernation is punctuated by spontaneous arousals. (2) The frequency of these arousals seems to fluctuate regularly over a hibernation season, producing a distinct pattern of bout length variation. (3) The arousability of a hibernator, and therefore the depth of the hibernation state, seems to fluctuate in a regular manner within each hibernation bout, producing a pattern of progressive irritability. These three phenomena (spontaneous arousal, bout length variation, and progressive irritability) suggest that hibernation is far from a uniform state. Thus when investigating state-dependant aspects of ventilatory control using hibernation as an example of variation in central state, it would seem important to identify both if and how the variations inherent in hibernation itself could effect the physiological variables being examined. Previous studies of respiratory control have indicated that there is a state-dependant change in the ventilatory response to both CO 2 and 02 during hibernation (McArthur and Milsom, 1991a & b). In particular, these studies have shown that, during hibernation the ventilatory responses to both 0 2 and CO 2 decrease, and that animals become considerably more tolerant of elevated levels of CO 2 and decreased levels of 0 2 . Also, when data were normalized to the rates of 0 2 uptake and CO 2 excretion (ie, the metabolic rate) there was a large decrease in sensitivity to hypoxia but a significant increase in sensitivity to hypercapnia. These observations suggest a change in the descending influence of higher centres on the integration of peripheral information. These studies compared the ventilatory responses of  11 awake animals with those of hibernating animals. The hibernating data was generated from repeated tests of many individuals over the winter season. Mean values of these data were taken to represent typical responses. However, these same studies did not control for possible inter- or intra-bout variations in these ventilatory responses. The presence of variation within or between bouts would introduce bias into the data and, in fact there may not be a "typical" hibernating response. In this case, no valid comparison could be drawn between the awake responses and any single hibernating value. The object of the first portion of the present study therefore, was to investigate the possibility that variation occurred in the hypercapnic ventilatory response within and between bouts of hibernation. To achieve this, the ventilatory response to CO 2 was repeatedly measured on different days within a single bout of hibernation, and at a specific time during different bouts over the season. In this way, any inter- or intra-bout differences in ventilatory responses could be identified. The second portion of this study was concerned with the more abstract concept of hibernation depth. The Twentes (1967, 1968) suggested that hibernation depth could be inferred from the sensitivity of a hibernating animal to stimuli. As previously discussed, they identified a regular change in hibernation depth over a bout by noting that their animals became less tolerant to their experimental protocol (more irritable) as a bout progressed. What was lacking in that study, however, was a form of objective quantification of hibernation depth which would enable comparison between individuals and points in time. The present study, therefore, attempted to quantify hibernation depth in terms of the level of respiratory stimulation required to bring about arousal. It also attempted to extend the investigation of depth fluctuation beyond a single bout, to see if there were any noticable seasonal (between bout) variations in hibernation depth.  12 The present study asked the questions: (1) What is the overall character and sensitivity of the ventilatory response to changing levels of CO 2 during hibernation ? (2) At what level of inspired CO2 will arousal from hibernation be initiated ? (3) Do these characteristics vary with time either during a bout of hibernation or over the hibernation season ? The answers to these questions provide the data necessary to test two primary hypothesis. The first is that the ventilatory response to changing levels of CO 2 remains constant during hibernation, both between days in a bout of hibernation, and over the coarse of the hibernation season. If this hypothesis is supported, then conclusions from previous studies which did not control for such variation will remain valid. If, however, inter- or intra-bout variation is found, then the data is from previous studies will be of questionable value. The second hypothesis put forward is that although hibernation depth fluctuates over an individual bout of hibernation (progressive irritability), this pattern remains constant over the hibernation season. Confirmation of this hypothesis would lend support to the view that seasonal hibernation has a fixed depth.  13 MATERIALS AND METHODS Animals: This study employed fifty wild caught, golden-mantled ground squirrels (Spennophilus (Citellus) lateralis), obtained from a supplier in Redding California. They were housed  individually in polycarbonate tanks measuring approximately 45 x 25 x 20 cm, supplied with wood chips, ground corn cob and cotton nesting material. Animals were provided with Purina rodent chow and water ad libitum. This diet was supplemented with sunflower seeds. The tanks were stored in a sound insulated, climate controlled chamber where, at the onset of each study season, the temperature (Ta) was decreased from 20 +/- 1° C to 5 +1- 1° C, and the photoperiod was decreased from 12 hours light:12 hours dark (12L:12D) to 2L:22D (lights on at 12:00, noon), over a period of approximately 10 days. These conditions were maintained for the duration of the winter season. In the first year, ten animals underwent surgery for the implantation of a re-entrant tube (described below) to allow the monitoring of core body temperature via a removable thermistor. Four other animals represented a "non-surgery control". In the second year, sixteen animals underwent surgery for the implantation of a re-entrant tube and of these, ten were fitted with electrode leads (described below), to monitor cardiac activity. Ten additional animals were treated as a "non-surgery control". In the third season, ten animals were fitted with reentrant tubes as well as monitoring electrodes for cardiac, muscle, and brain activity. During the first year, the duration of each bout of hibernation was monitored in all animals beginning in mid November. This was achieved by clearing the bedding above each  animal and placing woodchips or sawdust on its back. Animals in hibernation display atonia and do not move throughout the entire hibernation bout (Lyman,1982b). Thus simple observation would indicate when the animal had aroused by the displacement of the chip  14 and/or the re-covering of the animal with nest material (Pengelley & Fisher, 1961.; Geiser & Kenagy, 1988). In the second and third season, animals were monitored in the same manner, beginning in mid-October for non-surgery animals, and mid-November for animals which had undergone surgery. During the first year, fluctuations in body temperature were monitored daily in those animals fitted with re-entrant tubes. This facilitated the identification of periods of arousal. It was possible, however, for an animal to arouse and re-enter hibernation within 24 hours. Thus, TB could increase and drop back to hibernation levels between two successive observations making it possible to miss an arousal were one to rely on temperature data alone. Therefore both the  TB and  wood chip methods of following the progression of  hibernation were utilized when possible. The hibernation behaviour of second season animals was monitored using the wood chip method alone, with no secondary investigation of body temperature fluctuations. However, second year animals exhibited a surprising intolerance to non-respiratory stimuli such as handling and noise. Third season animals were also monitored using the wood chip method alone, however these animals were handled daily. This further treatment appears to have enhanced the animal's tolerance to handling and reduced their sensitivity to non-respiratory stimuli, as indicated by a greatly reduced number of premature arousals. Observations were terminated for all animals in early to mid April.  Surgical Procedures: Direct access to a source of core body temperature was achieved through the surgical implantation of a small, rigid re-entrant tube (described below) through the skull and into the cranial cavity. This allowed accurate measurement of brain temperature, which is identical to  15 core temperature (Twente & Twente, 1968). The anaesthetic used for all surgical procedures was sodium pentobarbital, 65mg/m1 (Somnotol, M.T.C. pharmaceutical, Mississauga Ont.), administered intraperitoneally in 0.10 ml increments, to a level no greater than 12 mg per 100g body weight. Supplementary anaesthesia, when necessary, was achieved through the topical use of Lidocaine Hydrochloride 20mg/m1 (Xylocain 2%, Astra Pharmaceutical Canada Ltd., Mississaga, Ont.) administered to the incision, and/or by inhalation of vaporous Halothane. The time required to reach a surgical plane of anaesthesia was approximately 30 minutes. The heads of fully anaesthetized animals were cleared of hair from midway along the snout to the base of the neck and down to the level of the ears using electric clippers and a chemical hair remover (Neet, Whitehall Laboratories Ltd. Mississaga, Ont.). Residual traces of hair and "Neer were removed with distilled water and ethanol. The animals heads were placed into a stereotaxic device, and their bodies supported on blocks of foam. An anterior / posterior incision was made along the dorsal crest of the skull within the cleared area. The skin was retracted and the underlying connective tissue cleared with a scalpel and blunt probe. The temporalis muscles on both sides of the skull, posterior to the orbit were partially retracted from their cavities. The exposed area of skull was cleaned and dried with 3% hydrogen peroxide and ethanol, and scored with a scalpel to enhance epoxy adhesion. A Dremel moto-tool and 1/32" engraving cutter (No. 105, Dremel, Racine Wisconsin) was used to drill a small diameter hole in the parietal bone, slightly left of midline approximately level with the ears. Care was taken not to puncture the subcranial membrane at this stage. Stereotaxic coordinates were not noted. A stereotaxic micromanipulator was used to insert and hold a sterile (> 20 minutes in alcohol) stainless steel re-entrant tube (Hypodermic S.S. Tube, .0495" od, #8448, A-M  16 Systems Inc. Everett, WA.), sealed at one end, 6 mm into the cranial cavity. Pilot experiments indicated that, at this depth, the sensitive region of a thermistor inserted into the tube would be placed well into an area of representative temperature, without any disturbance to the brain by the implanted tube. The tube was cemented to the skull with dental epoxy (flexacryl pink, Lang Dental Mfg. Co., Chicago, Ill.). In some animals, to aid adhesion, stainless steal screws (00 x 1/8, J.I.Morris Co., Southbridge, Mass.) were fastened to the skull previous to the application of epoxy. After the epoxy had set, all tubing extending above the epoxy skull cap foundation was removed, and the incision was sutured closed around the reentrant tube. All surgeries employed the same procedure for the implantation of re-entrant tubes. However, in the third year, modifications were made to allow for the implantation of electrodes to measure heart (EKG), muscle (EMG) and brain (EEG) activity. In these animals, the area cleared of fur was extended slightly past the animal's shoulders, to allow the incision to be extended mid-dorsally to the base of the skull. Two sub-cutaneous channels were created, one on each side running from the incision at the base of the skull, dorsal to each shoulder, and terminating in the regions of the left and right mid-lateral thorax. The EKG electrodes consisted of two 6 cm lengths of 7 strand stainless steel, teflon coated wire (0.009", A-M Systems, inc., Everett, Washington) with a 4 mm stainless steal washer soldered to one end, and capped with a male contact pin (WPI, p/n 220-P02-10000, RAE Electronics) at the other end. The washer-end of one lead was fed down each channel, and fastened to the body wall approximately level with the third rib on each side with a cyanoacrylate glue (Krazy Glue Inc. Chicago, IL.). EMG electrodes (similar to EKG, but 4 cm long with bared ends in place of washers) were threaded into the right and left trapezius muscles. Four stainless steel screws (00 x 1/8",Morris Co., Southbridge Mass.) were  17 implanted in the left and right frontal and parietal bones to serve as EEG electrodes. 3 cm lengths of single stranded teflon coated stainless steel wire (P.No.316SS5T, Medwire, Mount Vernon, New York) were attached to each screw and tipped with a male contact pin. All male pin connectors (8 in total) were secured into two pin strips (WPI, Rae Electronics) and secured to the skull beside the reentrant tube to form a large head-piece. Adhesion of the head-piece to the skull was enhanced by pre-treatment of the bone with a penetrating solvent (Copalite, Harry J.Bosworth Co., Skokie, Ill) and grip cement (R+R Dentsply, Toledo, Ohio) in association with dental epoxy. Each animal was administered a 0.10 ml dose of dilute (125mg/m1) ampicillin (Penbritin-250, Ayerst laboratories, Montreal, Que.) post-operatively as an antibiotic.  Apparatus Core body temperatures were recorded with thermistors constructed of thermocouple wire (TC type T constantin-copper, size .005; California Fine Wire, Grover City, CA), with a solder union at one end, and a Sensortek adapter at the other. When plugged into a Sensortek digital thermometer, these would register the temperature at the level of the soldered union. The thermistors were calibrated at each use against two Fisher mercury thermometers and a water bath, over a range from approximately 5° C to 30° C following daily body temperature measurements, and over a range from 2° C to 6°C following arousal tests. During the first two seasons, respiration was measured using a strain gauge transducer attached via a length of suture silk and a clamp, to the animal's fur midlaterally, in an area which showed maximum movement during respiration. As the animal ventilated, expansion and compression of the chest would alter the strain on the gauge. During first year experiments, the gauge was connected to a DC strain gauge pre-amplifier, and a Gould  18 universal amplifier, and the output was recorded on a polygraph recorder (Harvard Apparatus 12 Speed Chart Mover, model 480). Time was scored on an additional channel using a Harvard Apparatus "time marker module". For second year tests, the strain gauge and preamplifier were connected to a Gould variable speed two channel pen recorder by way of a Gould transducer amplifier. Chart speed was manually recorded during the experiment. From the raw data, breathing frequency and the amplitude of the strain gage excursions, which are an estimate of respiratory amplitude, could be calculated. During third year tests, the animal was fitted with a ventilatory mask fashioned from a 60 cc syringe (Becton Dickson & Co., Rutherford, N.J.), form fitted to the animal with odourless modelling clay, and sealed to the animal by a tubular latex sheath (Sheik nonlubricated condom, Julius Shmid, Scarborough, Ont.). The tip of the syringe was fitted to one side of an "pneumotach", consisting of two 7.5 mm plexiglass tubes (4 mm ID) each fitted with a 1 cm narrow steel tube (1.5 mm ID), glued together while separated by a thin nylon mesh. Each breath would produce a flow of air through the pneumotach which would be impeded by the mesh creating a differential pressure between the two sides. The steel tubes were each connected to one side of a Valadyne differential pressure transducer which would register this difference in pressure. The differential pressure measured in this way was proportional to air flow. The output from the transducer was fed directly through a Grass DC amplifier (7P122E) and a pen deflection representing air flow was recorded on a Grass model 79E six channel ink writing oscillograph. The Transducer output was also simultaneously channelled through a Gould integration amplifier to a second Grass DC amplifier, resulting in a pen deflection representing the integration of flow with time or the volume of air moved. The EKG EMG and two sets of EEG electrodes were connected to four channels of the polygraph via Grass AC amplifiers (7P511K). The resulting polygraph record gave a  19 simultaneous record of EEG, EKG, EMG, respiratory air flow, breath volume and time. In the first year, hibernating animals were placed in a plexiglass box (20 x 16 x 14 cm. approximately 4.5 1 volume) with bedding material, in the environment chamber (Figure 2a). The thermistor, and string connecting the strain gauge to the clamp attached to the animal, ran through a small hole in the top of the box. Gas composition entering the box was set by dual flow meters using gas from tanks of medical air and pure CO 2 . Gas composition inside the box was monitored through the use of a Beckman LB-2 CO 2 gas analyzer, calibrated at each test with room air (0.04% CO2) and premixed 5.0% and 10% CO 2 (Radiometer GMA-2). Gas samples were taken using a 500cc glass syringe. In the second year, experiments were carried out in the laboratory within a smaller controlled environment chamber, a modified 150 litre refrigerator (Figure 2b). This chamber was maintained at 5° C, and monitored with a Fischer mercury thermometer. The animal was placed in an air-tight plexiglass box (20 x 14 x 12 cm: approximately 3.5 1 volume) in the chamber. The thermistor exited the box through a small hole, sealed with vacuum grease. The suture silk connecting the strain gauge to the clamp attached to the animal passed through a patch of latex dental darn (15.25 cm x 15.25 cm, extra heavy, HCC Corp. St.Catherine, Ont.) stretched and sealed over a larger port in the top of the box. The box was connected to the Beckman LB-2 CO 2 analyzer (calibrated as mentioned), by a length of 7 mm (internal diameter) rubber tubing. The outlet of the analyzer was reconnected to the box with similar tubing, forming a closed system loop. Gas was circulated within this loop by the analyzer's sample pump which was set to flow at 0.5 litres per minute. The gas analyzer and pump were air tight, and the total system had a volume of approximately 2 litres. The closed system loop through the gas analyzer allowed for constant monitoring and finer control of system gas composition.  20  Figure 2a:^Experimental apparatus used during the first year.  1^TEST ANIMAL 2^TEST BOX 3^STRAIN GAUGE 4^PRE-AMPLIFIER 5^UNIVERSAL AMP' 6 PER RECORDER 7^TIME KEEPER 8 THERMOMETER 9^THERMISTOR 10 FLOW METERS 11 MIXED GAS IN 12 GAS OUT 13 GAS SAMPLER 14 GAS IN FROM AIR & CO2 TANKS  21  22  Figure 2b:^Experimental apparatus used during the second year.  1^TEST ANIMAL 2^TEST BOX 3^STRAIN GAUGE 4^PRE-AMPLIFIER 5^UNIVERSAL AMP' 6 PER RECORDER 7^TEST ENVIRONMENT 8 THERMOMETER 9^THERMISTOR 10 GAS ANALYZER 11^GAS IN 12 GAS OUT 13^CO2 IN 14 BALLOON  23  24  Figure 2c:^Experimental apparatus used during the third year.  1^TEST ANIMAL 2^TEST BOX 3^DIFFERENTIAL TRANSDUCER 4^PRE-AMPLIFIER 5^TRANSDUCER AMP' 6 POLYGRAPH RECORDER 7^TEST ENVIRONMENT 8 THERMOMETER 9^THERMISTOR 10 GAS ANALYZER 11^GAS IN 12 GAS OUT 13 FLOW METERS  25  26 Gas composition was changed by introducing small volumes of pure CO 2 into the loop. These samples were obtained from a cylinder of 100 % CO 2 (Union Carbide), and were introduced into the system with a 500 cc glass syringe, via a T-union and stopcock. Volume changes resulting from the introduction of CO 2 , were compensated by a flexible latex sample balloon connected in parallel with the closed loop. During the third season, tests were again carried out in the laboratory chamber (Figure 2c). Animals were fitted with a mask and pneumotach, and placed in an airtight plexiglass box (10 x 10 x 10 cm.: approximately 1 1 volume). Electrode leads, a thermistor probe, and the tubes connecting the pneumotach to the transducer all passed out ports in the box which were sealed with vacuum grease. Compressed air and CO 2 were mixed and fed into the box at a rate of approximately 500 ml/min., through two Linde in-line flow meters. Gas composition was monitored continuously with a Beckman analyzer (calibrated as mentioned) as it exited the test box.  Experimental Protocol (i) Inter-bout tests Animals were studied during the third day of a bout of hibernation at different times over the hibernation season. Each was initially supplied with medical air (0.04% CO 2) at a flow rate of 500 ml per minute and left for approximately one hour. Following this control period, the animal's respiration and TB were monitored at this baseline CO 2 level (called 0%) for approximately 1 hour. The composition of the chamber gas was then increased to 2% CO2 and respiration and TB were again monitored for 1 hour or until it was believed that a steady state had been reached. Stepwise increases in CO 2 levels, separated by approximately one hour periods of observation, continued until signs of arousal were observed. The criteria  27 for arousal were taken to be any combination of: (A) a rapid continuous increase in  TB, (B)  any overt movement not characteristic of hibernation behaviour, (C) an increase in respiratory frequency to a level greater than 40 breaths per minute. Animals were not left in the test environment to achieve full arousal as it was felt that this could influence their subsequent hibernation behaviour (Twente & Twente, 1968). (ii) Intra-bout tests During the third season, measures of hypercapnic ventilatory sensitivity were attempted on individual animals during each consecutive day of a single bout of hibernation. These tests followed the same protocol as the inter-bout tests, although the manipulation was terminated before levels of CO 2 were reached which would initiate arousal. Animals were taken from air to approximately 8 to 10 % inspired CO 2 and then returned to breathing air. Eight to twelve hours of exposure to air was provided between successive tests. Hibernation was maintained through out this period, and in no case did arousal occur between successive tests.  Data Analysis The test traces were analyzed for respiratory frequency (Rf) and amplitude (Ra) or tidal volume (VT) at each level of inspired CO 2 . At low levels of inspired CO 2 , the animals exhibited breathing patterns of either single breaths separated by periods of apnoea, or episodes or bursts of continuous breathing separated by longer periods of apnoea. Rf was determined by counting the total number of breaths occurring in a number of successive episodes within a space of time, bounded at one end by the beginning of one breath or burst of breathing, and at the other end by the beginning of another breath or burst of breathing. The Rf for animals breathing higher levels of CO 2 and exhibiting continuous breathing were  28 determined by counting the number of breaths occurring within 10 minute intervals (usually 3 to 5 intervals covering the last 45 to 60 minutes of exposure to each CO 2 level). During the first two seasons, Ra was estimated from the relative magnitude of the pen deflection representing the movement of the animal's body wall. The observation was problematic in that the amplitude of the breathing traces was effected by such factors as the position of the animal, position of the strain gauge clamp or charge in the strain gauge battery. In order to compare Ra values between animals and/or tests on the same animal, therefore, it was necessary to calculate Ra as a percentage change from some initial value. The Ra exhibited by the animal at 0 % inspired CO 2 , measured as the mean height of the respiratory peaks in mm., was taken to be Ra = 100%. The mean heights of respiratory peaks at other levels of inspired CO 2 were expressed in terms of this baseline value. The Ra estimates could then be compared between different animals and test times. Unfortunately, due to the possibility of body movement and/or differential muscle recruitment during a single test, the values attained during the first two seasons for Ra can only be considered as estimates. A calculated estimate of minute ventilation (VREsT), was achieved by multiplying relative Ra with Rf. Actual values for VT were obtained in the third season by the use of the calibrated pneumotach mask. This allowed accurate measurement of VE, determined to be the sum of volumes of all breaths taken during an observation interval divided by the duration of that interval. The absolute levels of ventilation exhibited by each animal were not always consistent between tests, however. To facilitate comparison of responses, it was necessary to report these data in relative terms. Thus, all values for Rf,  VT and  VE measured during the  third season are expressed both as a proportion relative to the initial values observed during each test and as actual values.  29 The animal's ventilatory response was characterized by plotting either Rf, Ra and VET  or Rf, VT and VE against levels of inspired CO 2 . The slopes of these response plots  were taken to represent the animals' sensitivity to CO 2 .  Inter- and Intra-bout comparison of Ventilatory Responses: Statistical comparisons were achieved by dividing the response profiles into categories (either days 1 through 5 of a hibernation bout, or early, middle and late portions of the hibernation season) and comparing the responses both at each level of inspired CO 2 between the categories using multiple paired and independent T-tests, as well as repeat measures and two-way ANOVA. The levels of inspired CO 2 sufficient to prompt arousal from hibernation were categorized as either early, middle or late season responses. These categories were compared using multiple paired T-tests. Statistical analyses were carried out using a personal computer and statistical software (SigmaPlot (Jandel Scientific Co., Corte Madera, Ca.) and Systat (Systat Inc.,Evanston II.)) as well as the mainframe computer at the University of BC. Statistical significance was attributed to differences with a significance level (p-value) less than 0.05. For inter-bout comparison, individual animals were tested at different points in the hibernation season. Unfortunately, data were not available for each individual during all three of the seasonal periods. Because repeat measures were taken on individual animals between early, mid and late season periods, there is potential for violation of the statistical independence assumption inherent in two-way ANOVA and independent T-tests. There are also potential problems inherent in the use of a repeated-measures ANOVA, and paired Ttests stemming from missing data points and unequal sample sizes, however. Therefore, both types of statistical tests were done although the statistical validity of the resulting analyses is  30 arguable. Intra-bout comparisons were made by comparing the mean and individual responses of two animals tested on five consecutive days of one hibernation bout. Thus, although the use of repeat-measures analysis is valid in this situation, the low sample size prevents the statistical generalization of these findings to a larger population.  31 RESULTS  Hypercapnic Ventilatory Response (i) Components of the Response (a) Respiratory Frequency (Rf) The relationship between breathing frequency and increasing levels of inspired CO 2 appeared sigmoid (Figure 3a & b). Rf was uneffected by increases in levels of inspired CO 2 between 0 and 2 %. Above 2 %, Rf increased linearly with increasing CO 2 . The slope of this increase differed between individuals, but usually consisted of a twelve-fold increase in frequency between 2 and 6 % CO 2 . Changes in Rf plateaued between 6 and 8 % CO 2 ; above this level, Rf remained constant or in some cases declined slightly as inspired CO 2 was increased further. At higher levels of CO 2 (10 to 14 %) the animals initiated arousal from hibernation. Upon arousal Rf again increased. (b) Tidal Volume (V T) Changes in VT of individual animals in response to increasing levels of inspired CO 2 varied greatly. In some instances  VT tended  to increase slightly in response to increasing  levels of inspired CO 2 . In others, the animal exhibited a consistent decline in  VT (Figure  3a  & b). Regardless of direction, fluctuations in VT were for the most part slight, changing less  than 50 % from initial levels. These variations appeared to be greatly overshadowed by the changes in Rf. (c) Minute Ventilation (VE) The pattern of observed change in VE with increased ambient CO 2 closely resembled the pattern of change seen in Rf; VE tended to increase in a sigmoid fashion with increases in inspired CO 2 . Inflections in the VE response profile occurred at the same levels of inspired CO2 as noted for the Rf profile (Figure 3a & b). On average there was an eight to ten-fold  32  Figure 3a:^Typical hypercapnic ventilatory response of one animal, characterized by changes in respiratory frequency (Rf), tidal volume (VT), and minute ventilation (NTE), associated with changes in inspired CO2.  33  TYPICAL VENTILATORY RESPONSE Rf, VT and VE vs. % Inspired CO 2 Expressed as actual values (Animal 7) 30  VE (ml/min)  14  Rf (Br/min  - 12  • 10  25  - 8 - 6 20  - 4  - 2 15  0  3.0  (ml/Br)  2.5  10  - 2.0  •  I^I^1^1^1^1^I  ^t^t^II  2^4^6^8^10^12^14 0^2^4^6^8^10 12 1  4  % CO 2 (inspired) * Indicates values obtained during arousal from hibernation  34  Figure 3b:^Typical hypercapnic ventilatory response of another animal, characterized by changes in respiratory frequency (Ri), respiratory amplitude (Ra) and an estimate of minute ventilation (NTEEsr), associated with changes in inspired CO2.  35  TYPICAL VENTILATORY RESPONSE Rf, Ra and VE EST vs. % Inspired CO 2 Year 1 Animal 1 16  14  12  10  8  Ra  (Relative)  6  - 1.5  4  - 1.0  2  - 0.5  0  1  0^2^4^6^8^10 12 14 0^2^4^6^8 10 12 14 %  0.0  CO 2 (inspired)  * Indicates values obtained during arousal from  hibernation  36 increase in VE as ambient CO 2 was increased from 0 to 8%. (d) Relative Contribution of Changes in V T (Ra) and Rf to the Ventilatory Response The overall ventilatory response is a function of the changes in the two components of the breathing pattern, Rf and V T (Rf x VT = VE). The relative significance of the changes in each of these components to the change in VE was determined by analysis of the intercorrelation between each component and VE. Regression analysis of Rf versus VE (Figure 4a) and VT versus VE (Figure 4b) indicated that changes in Rf correlated strongly with the fluctuations in VE (R2 = 0.88, N = 5, n = 80), whereas an insignificant correlation was observed between changes in  VT and changes in  VE (R2 = 0.02, N = 5, n =80).  (e) Ventilatory Pattern Ventilatory pattern changed as levels of inspired CO 2 were increased (Figure 5). Initially, hibernating animals exhibited an episodic breathing pattern when exposed to air. As levels of inspired CO 2 were increased, this pattern was modified by increasing the number of breaths in each episode, and decreasing the total non-ventilatory-period (T) between episodes. At elevated levels of inspired CO 2 , the TNvp between episodes was reduced to a level equal to the TNvp between breaths, thereby transforming the episodic breathing pattern to a continuous breathing pattern. Further increases in ventilation were achieved by reducing the Tom, between breaths. Levels of VT remained relatively constant at all levels of inspired CO 2 .  (ii) Overall Shape of the Ventilatory Response: As indicated in the previous section, the hypercapnic ventilatory response may be characterized by changes in breathing frequency alone (Figure 6). There appeared to be four distinct regions within the response. In the first region there was a slight increase (if any) in Rf between 0 and 2 % inspired CO 2 , indicating either that a minimum level of inspired CO2  37 was required to initiate a ventilatory response (a "threshold"), or that the response was parabolic, with little change occurring below 2 % inspired CO 2 . Above 2 % inspired CO 2 , in the second region of the response, VE and Rf both increased sharply with further increases in inspired CO 2 . The slope of this portion of these plots is suggested to reflect the animal's sensitivity to CO 2 . Although variation occurred between individuals, the third region of the response began as ventilation peaked at approximately 8 to 10 % inspired CO 2 . Over the third region ventilation remained constant or decreased slightly with further increases in inspired CO2 . In the fourth region of the response further increases in inspired CO 2 caused arousal from hibernation, at which time ventilation again increased. The onset of the fourth region, or the level of inspired CO 2 at which signs of hibernation arousal were first observed, is referred to as the arousal threshold. (iii) Comparison of the Ventilatory Response between Bouts over the Season The ventilatory responses of nine animals were successfully tested on two or more occasions during a single hibernation season. No differences in the ventilatory response were observed between replicate tests of individual animals over the season, in any of the first three regions of the hypercapnic ventilatory response (Figure 7a & b). Similarly, two-way ANOVA and paired and independent T-tests indicated that there were no differences between the CO 2 sensitivities or mean levels of Rf at each level of inspired CO2 , ranging from 0 to 9 %, during early, middle or late season tests (Figure 8). Seasonal differences in the response to inspired CO 2 levels in excess of 9 % will be described later, as changes in the arousal  threshold (the fourth region of the hypercapnic ventilatory response).  38  Figure 4a:^Relationship between minute ventilation and respiratory frequency. Note the strong positive correlation between these two variables as they change with inspired CO 2 (R = 0.94).  39  RELATIONSHIP BETWEEN VE AND Rf Expressed as actual values 14  12  10  8  6  4  2  0 ^ ^ ^ ^ ^ ^ ^ 14 12 10 2 4 6 8 0  VE (ml/min)  40  Figure 4b.^Relationship between minute ventilation and tidal volume. Note the week positive correlation between these two variables as they change with inspired CO 2 (R = 0.14).  41  RELATIONSHIP BETWEEN VE AND V T Expressed as actual values 2.0  Regression: ^0 N = 5 1.8 n = 80 2  R = 0.02 00  1.6  0^  O^0  0^ 1.4 -  o  o 0^ 0z  1.2  -  0.8  0.4  0  ^.^  o^0 G 6'4°^  CC) 0 O 0^ 0^B 0 0^0  0.6  0  0 .^0 0 cb^.  0 0^0 0 0^ 0^0 O 0^ 0 ,0 9, 0 0^ 0 0 0^ 0 5)^ 0^ 0 O 00 0  0  1.0  8  o  0  0°^  0  0  0  2^4^6  ^  8^1 0  E (ml/min)  ^ ^ 14 12  42  Figure 5: Breathing volume traces illustrating the ventilatory patterns observed at various level of inspired CO2 . Tracings represent inspiratory (downward deflection) and expiratory (upward deflection) volumes of each breath. Note that in general the total non-ventilatory period (T) between breathing episodes decreased, and the number of breaths per episode increased as ambient CO 2 was increased. The Tr,/vp between the episodes illustrated at 0 % CO2 was 270 seconds.  43  schematic of the hypercapnic ventilatory response illustrating four characteristic regions within the response  ■■■ 11■1, 1=•• .01■1  (1)^(2)^(3)^(4)  % INSPIRED CO 2^  Arousal Threshold  44  Figure 6:^Schematic diagram of the hypercapnic ventilatory response, illustrating the four characteristic regions observed as respiratory frequency changed with increases in inspired CO2 . Respiratory frequency is expressed simply as ventilation (in arbitrary units). Numbers in brackets refer to the four regions described in the text.  45  1 inl.  2%  6  8  10  11 % (Arousal)  60 seconds  46  Figure 7a:^Typical hypercapnic ventilatory response of an individual animal (# 7), observed during three repeated tests over the hibernation season. Values for Rf, VT and VE are presented as actual values.  47  TYPICAL VENTILATORY RESPONSE Rf, VT and VE vs. % inspired CO 2 Expressed as actual values 40  ^-14  VE (ml/min)  - 12  35  -10 - 8  30  0E1  - 6 25  - 4  MN  - 2 20  ^ I^1^1^1^1^I ^-  -0 5  15  4  10  - 3  2  2^4^6  ^  I^I^  I  I^I  I---. 1  8^10 12 14 0 ^2^4^6^8 10 12 14  % CO 2 (inspired)  o Jan 15^v Feb 4^o March 15 (Animal 7)  48  Figure 7b:^Typical hypercapnic ventilatory response of an individual animal (# 7), observed during three repeated tests over the hibernation season. Values for Rf, V T and VE are presented as relative values, expressed as a % of those observed during normocapnia.  49  TYPICAL VENTILATORY RESPONSE Rf, VT and VE vs % inspired CO 2 Expressed as % of Base value (1 = 100 %)  10  2^4^6^8^10 12 140  ^  2^4  ^  6^8 10 12 14  % CO 2 (inspired) o Jan 15^v Feb 4^o March 15 (Animal 7)  50  Figure 8:^Mean hypercapnic ventilatory response of all animals, separated into either early, mid or late season categories and expressed as both relative and absolute values of Rf at increasing levels of inspired CO2.  51  MEAN SEASONAL Rf vs. % inspired CO 2  16  2 0  0^2^4^6^8^10  12^14  % Inspired CO 2 0 Early^v Mid inerint4)  (eir+nrtri  a Late r1Princi)^(third period)  52 (iv) Comparison of the Ventilatory Response between Days over a Hibernation Bout The ventilatory responses of two animals were successfully tested on five consecutive days of a hibernation bout (Figures 9 a & b and 10 a & b). Comparison of successive tests using a repeated-measures ANOVA indicated that no within-bout variation had occurred between successive days, in any region of the ventilatory response.  Pattern of Hibernation (i) Changes in Body Temperature over the Hibernation Season Observations of changes in body temperature during the first season showed shallow drops in TB of short duration early in the season which became progressively longer and reached lower temperatures until animals reached and maintained the low body temperatures associated with hibernation (Figure 11). During hibernation the animals maintained a consistent low body temperature close to Ta for prolonged periods. These hibernation bouts were punctuated by brief periods of arousal, during which TB returned to euthermic levels.  (ii)  Changes in Bout Length over the Season  A standard pattern of bout length variation was exhibited by the animals used in this study (Figure 12). Bout duration increased slowly during the early season, reaching a maximum approximately mid-season, and decreased again as the animal approached  termination of hibernation in the late season. This pattern was evident in individual animals, and no difference was apparent between animals which had undergone surgery and control animals.  53  Figure 9a:^Hypercapnic ventilatory response of an individual animal (# 7), observed during five consecutive days of an individual hibernation bout. Values for Rf, V T and VE are presented as actual values.  54  INTRA—BOUT RESPONSE (Animal 7) Rf, V T and VE vs. % Inspired CO 2  Expressed as actual values  ^ 16  40  - 14 - 12  35  - 10 - 8  30  - 6  25 i  20  -4 _2 ^0 5  (ml/Br) 15  4  10 5 0  0^2^4^6^8^10 0^2^4^6 % CO 2  (inspired)  o Day 1 V Day 2 ^ Day 3 ii Day 4^o Day 5  8^10  1  55  Figure 9b:^Hypercapnic ventilatory response of an individual animal (# 7), observed during five consecutive days of an individual hibernation bout. Values for Rf, VT and VE are presented as relative values, expressed as a % of those observed during normocapnia. .  56  INTRA—BOUT RESPONSE (Animal 7) RF, VT and ,TE vs. % inspired CO 2 Expressed as % of Base value (1 = 100 %) 14 12 10 8 6 4 2 0  2^4^6^8^10 0^2^4 % CO  2  (inspired)  0 Day 1 v Day 2 o Day 3  & Day 4^o Day 5  57  Figure 10a: Hypercapnic ventilatory response of an individual animal (# 8), observed during five consecutive days of an individual hibernation bout. Values for Rf, VT and VE are presented as actual values.  58  INTRA—BOUT RESPONSE (Animal 8) Rf, VT and VE vs. % inspired CO 2  Expressed as actual values 40  16 14 12  35  10  8  30  6 4  25  2 0  20  5  15  4  10  3 2  2^4^6^8^10 0^2 4 % CO 2  ^  6  (inspired)  o Day 1 v Day 2 o Day 3 o Day 4^o Days  8  10  1  59  Figure 10b: Hypercapnic ventilatory response of an individual animal (# 8), observed during five consecutive days of an individual hibernation bout. Values for Rf, V T and VE are presented as relative values, expressed as a % of those observed during normocapnia.  60  INTRA—BOUT RESPONSE (Animal 8) Rf, VT and VE vs. % inspired CO 2 Expressed as % of Base value (1 = 100 %) 14 12 10 8 6 4 2 0 2.0  - 1.5  - 1 .0  I^I^I^I  I^I  2^4^6^8^10 0^2^4^6^8^10 % CO 2  (inspired)  o Day 1 v Day 2 o Day 3 & Day 4^o Day 5  0.5  61  Figure 11:^Seasonal fluctuations in core body temperature recorded daily from two individual animals, during 150 days of a hibernation season.  62  Seasonal Fluctuations in Tb. Year 1 Animal 9 40 ^ 35-  P  30—  et 20 15— 10 —  0  0^  1 I  25^SO^75^100^125^150 Study Duration (days) 0 s 11.11,88 Seasonal Fluctuations in Tb. Year 1 Animal 8  40 ^ 35 30 — 25 —  If  I  20 — 15—  10 —  50  %MN&  4^1^i^I 0^25^50^75^100  Study Duration (days) 0- 11,11,88  4  125^150  63  Figure 12:^Mean pattern of bout length variation observed in surgery and non-surgery control animals, during one hibernation season. As hibernation was initiated at different times by each individual, bout numbers are assessed from the first observed hibernation bout.  64  Mean Second Year Bout Lengths From first observed bout 12  4 2  0  10 12 14 1• 18 20 22 lout Number  12  Neon Second Year lout Lengths From first observed bout (non—surgery "control" animals)  10  a $ 4 2  0  Sib, b, b. %A.  k‘hib+.1%■§6,Sik..%■S■S 10 12 14 16 18 20 22  lout Number  65 (iii) Changes in the Arousal Threshold for Hibernation over the Season During inter-bout tests, the levels of CO2 required to stimulate animals to arouse from hibernation were recorded (Table 1). These arousal thresholds fluctuated between successive tests of individual animals. Division of these data into early, middle and late season values and the compilation of these into mean seasonal arousal thresholds, revealed an apparent pattern in this variation (Figure 13). These data revealed a non-significant trend suggesting arousal threshold to be least in the early and late season and greatest during the mid season. This pattern of variation in arousal threshold would superficially appear to mirror that seen in bout length. The low number of replicates and high levels of variation between individuals, however, limit the valid statistical analysis of the variation in arousal threshold or the apparent relationship between the variation in arousal threshold and hibernation bout length.  66  Table 1:^Levels of inspired CO 2 which prompted arousal from hibernation in animals tested during the early, mid and late portions of the hibernation season. Mean arousal thresholds are calculated for each of the three seasonal divisions.  67 Table 1 AROUSAL THRESHOLDS % CO2 (bout number when test performed) Time in the hibernation season 1 ST Period  (early)  2ND Period (mid)  3RD Period (late)  9 %(6) 10 %(8) 9 %(6) 7 %(6) 13 %(19) 11 %(12) 11 %(9) 12 %(16) 11 %(8)  8 %(12) 9 %(12) 8 %(13) 9 %(13) 11 %(24)  10 %(12)  10.33 % (0.60)  9.17 (0.48)  Number of bouts in season  Animal # 1 2 3 4 5 6 7 8 9 10 Mean (st.err.)  10 %(8) 8 %(4)  9 %(6) 9% (0.58)  + indicates artificial termination of the hibernation season  (15) (15) (16) (16)+ (25)+ (17) (20)+ (20)+ (14) (20)  68  Figure 13:^Individual (open circles) and mean levels (histograms) of inspired CO 2 which initiated arousal from hibernation during early, mid and late periods of the hibernation season. There is no significant difference between mean arousal threshold levels in early, mid and late season (P= 0.05), although a non-significant trend is apparent.  69  INDIVIDUAL AND MEAN AROUSAL THRESHOLD VARIATION 14 13 12 11 10  EARLY  ^  MID  ^  LATE  TIME IN SEASON Individual Animals (symbols) Group Means (histograms)  70 DISCUSSION Several conclusions can be drawn from the present study. During hibernation, increases in minute ventilation on exposure to hypercapnia are exclusively due to increases in breathing frequency. Initially, the level of inspired CO 2 has little or no effect on ventilation until a critical level of inspired CO 2 is reached (2 to 4% CO 2). Above this level, further increases in inspired CO 2 produce an increase in minute ventilation. There appears to be a second critical level (8 to 10 % CO2), however, beyond which, further increases in inspired CO2 have no further effect on ventilation until an arousal threshold is reached. Once the arousal threshold is attained, ventilation increases further and the animals arouse from hibernation. There are no variations in either the initial levels of ventilation, the slope or the overall shape of the ventilatory response to increasing levels of inspired CO 2 recorded on different days within a hibernation bout or in different bouts over the hibernation season. The animals used in this study showed characteristic variations in the length of successive hibernation bouts over the hibernation season. Individual animals also exhibited variation in the arousal threshold of their hypercapnic ventilatory responses between hibernation bouts. Although not significant, there appeared to be a relationship between the seasonal patterns of variation of both bout length and arousal threshold.  Hypercapnic Ventilatory Response (i) Role of Changes in Rf and V T in the Overall Hypercapnic Ventilatory Response Levels of minute ventilation (VE) are a function of the basic components of the breathing pattern; Rf and VT (Figure 3A). The methods used in the first two seasons only provided an estimate of VT based on measurements of body wall movement. The extrapolation of VT from a measurement of body wall movement is not very precise. The  71 former is a volume change while the latter is a linear change which is not uniform for all parts of the body wall and which will vary with body position. Furthermore, attempts to correlate such movements to V T are confounded by the paradoxical movements generated by the respiratory muscles themselves (Figure 14). Although the actions of the external intercostal muscles and the diaphragm both act to increase the volume of the thorax, the former muscles expand the chest while the actions of the diaphragm tend to depress the body wall and elongate the thorax during inhalation. The body wall movement generated during inhalation will vary depending on the relative recruitment of the diaphragm and intercostal muscles which are known to change with changes in the level of respiratory drive and body position (Phillipson and Bowes, 1986). Thus, estimates of VE based on the use of these values (Rf x Ra = VEST) are inaccurate. Due to this, values of Ra and  VEEST  obtained from  the data collected during the first two seasons have been excluded from analytical consideration. A representative figure (Figure 3b) has been included, however, to show that, qualitatively, the overall trends in the relationships between Ra and VEST vs. the levels of inspired CO 2 are similar to the relationships seen between V T and VE and the levels of inspired CO 2 obtained from data collected during the third season (Figure 3a). The use of the calibrated pneumotach and mask in the third season produced accurate measurements of Rf, V T , and VE. These data show there is a strong correlation between the changes observed in Rf and VE (R2 = 0.88, N = 5, n = 80, Figure 4a) when inspired levels of CO2 are increased. In contrast, no such correlation was found between the changes in VT and  VE (R2 = 0.03, N = 5, n = 80, Figure 4b) observed under these conditions. On the strength of these data it would appear that the overall ventilatory response to changes in inspired CO 2 can be accurately characterized by the changes in Rf alone. As a consequence, changes in Rf observed with increases in inspired CO 2 during the first two seasons are also assumed to  72  Figure 14^Schematic diagram outlining movement of the chest wall resulting from the inspiratory activity of the intercostal muscles or diaphragm.  73  The magnitude of chest wall movement during inspiration will depend on the relative contributions of diaphragm and intercostal muscles  Intercostal  - Chest Wall Neutral Condition No Chest Wall Movement  t Intercostal Mediated Inhalation Results in Active Chest Wall Expansion  Diaphragm Mediated Inhalation Results in Passive Chest Wall Invagination  74 accurately reflect the changes occurring in overall ventilation. This relationship between Rf and VE is consistent with data presented in previous studies. McArthur and Milsom (1991b) showed that during hibernation, the total nonventilatory period (T) between breaths was the major controlled variable of the breathing pattern, rather than the breath itself. Thus increases in ventilation upon hypercapnia exposure were primarily a result of decreases in the  TNW, which  were reflected as increases in Rf with  little or no changes in VT. In contrast, changes in the breathing pattern of awake animals on exposure to increased levels of inspired CO 2 were due to increases in both Rf and V T . Variations in respiratory pattern associated with increases in inspired CO 2 during hibernation are illustrated in figure 5. Levels of Rf increased as inspired CO 2 was elevated. These increases in Rf resulted from increases in the number of breaths per episode and decreases in the T between episodes. Gradually, as levels of inspired CO 2 were increased, the Trivp between bursts became indistinguishable from the  TNyp  between breaths. Thus, at  high levels of ambient CO 2, the burst-breathing patterns were modified to patterns of continuous breathing. (ii) Shape of the Ventilatory Response curve (a) Region 1: Animals did not exhibit a substantial increase in mean levels of Rf in response to hypercapnia until levels of inspired CO 2 exceeded approximately 2 % (Figure 3a and Figure  5). Similar observations have been noted by others in this and similar species (McArthur and Milsom, 1991b). Although CO2 is the primary stimulant of respiration in these animals during hibernation (Milsom et a1.,1986), the existence of an initial region in which the animal appears insensitive to changes in CO 2 would indicate that control of resting breathing is not  75 dependent on this primary stimulus. This is similar to the "dogleg" or "hockey stick" reported for this same relationship in humans by Neilson and Smith (1952, Figure 15). Here the initial insensitivity to a mild hypercapnic stimulus reflects a resting hypoxic ventilatory drive which maintains a minimum level of ventilation when the hypercapnic ventilatory drive is absent. If this hypoxic drive is removed, by increasing 0 2 levels, ventilation continues to fall linearly with the decrease in hypercapnic drive (Cunningham, et al.,1986). This species of ground squirrel has also been reported to exhibit a resting hypoxic ventilatory drive as revealed by a fall in ventilation with administration of an hyperoxic gas mixture (McArthur and Milsom, 1991a). The presence of such a resting hypoxic ventilatory drive may seem inconsistent with reports that this animal has a blunted response to ambient hypoxia during hibernation (McArthur and Milsom, 1991a; Webb, 1987). The blunted response in this species, however, is due to a left-shifted 02 equilibrium curve, not to hypoxia tolerance. Such shifts are due to a modification in the allosteric properties of squirrel haemoglobin molecules, effecting their affinity for 0 2 . The 02 affinity of haemoglobin can be characterized by a  P50  value, which  reflects the partial pressure of 0 2 (PO2) at which the haemoglobin is 50 % saturated. The  P5 0  value for haemoglobin from S. lateralis at hibernation TB has been reported to be 5.8 + 0.1 ton (7°C, pH, 7.46). considerably less than that of normothermic squirrels, 18.1 + 0.5 ton. .  (37°C, pH, 7.49) (Maginniss et a1.,1989). This left shift will allow a greater level of haemoglobin 0 2 saturation, and therefore a higher blood 0 2 content, at low levels of ambient 02 . This is significant in that 0 2 chemoreceptors in squirrels appear to respond to changes in 02 content rather than changes in blood 0 2 partial pressure (Pa0 2) (Webb and Milsom, 1990). It is suggested that the blunted response to ambient hypoxia in this species reflects the fact that these reductions in ambient 0 2 levels do not produce hypoxemia because the blood can  76  Figure 15:^Plot of the hypercapnic ventilatory response profile of man illustrating the effect that changes in ambient 0 2 have on the shape of this response.  77  (;)  ■^  .1  •• • L^ 4  /  •Sit ■^  30 /E  al  20  •  j'•^/ • • 4^fr  /0  0  SO  Pco A 2 Ventilation (V) as a function of PACO 2 at a steady PACO 2 of 37 (•), 47 (+), 110 (0), and 16; (X) Torr.  78 still be fully saturated at low PA 02 . When inspired 0 2 is dropped sufficiently to cause arterial blood to desaturate, an hypoxic response is present (Webb and Milsom, 1990). Thus when viewed in terms of arterial 0 2 saturation or 0 2 content, this species exhibits an hypoxic response similar to other mammals. An alternate explanation is that the first region of the hypercapnic ventilatory response represents the initial portion of a parabolic response curve. In either case, there appears to be no inter- or intra-bout variation in this portion of the response during hibernation. Species which exhibit a distinct hypercapnic ventilatory "threshold" seem to be the exception not the rule (Cunningham et a/.,1986). Significance rests in the fact that humans possess such a threshold. Pathological conditions such as central sleep apnoea and sudden infant death syndrome are believed to reflect alterations in the mechanisms which establish resting ventilatory drive (Cherniack and Longobardo, 1986; Phillipson and Bowes, 1986; Williams et al., 1991). The observation that S.lateralis exhibits an insensitivity to low levels of hypercapnia adds weight to the belief that this species could be a useful tool for research into abnormal human conditions. (b) Region 2: Above region 1, the ratio of the change in ventilation to changes in inspired CO 2 yields an indication of the sensitivity of the animal to CO 2 . This sensitivity is reflected by the slopes of the relationships between Rf and VE and the level of inspired CO 2 . The data suggest that no significant changes occured in this sensitive region of the hypercapnic ventilatory response with time, either over the season (Figures 7a, b and 8) or between days within a bout of hibernation (Figures 9a, b and 10a, b). There were, however, significant differences in sensitivity to CO 2 between individuals. Observations of a consistent sensitivity to CO 2 in these animals are of particular  79 importance when considering other studies which involve the assessment of CO 2 sensitivity in hibernating species. For example, the studies of McArthur and Milsom (1991) and Webb (1987) report CO2 sensitivity based on calculated mean values from repeated tests performed at random over the hibernation season. If CO 2 sensitivity was changing between or within bouts of hibernation over the season, these mean values would not reflect the true sensitivity of a given animal at any given time. As the present findings indicate that no changes in CO 2 sensitivity occur within or between bouts of hibernation, these mean values reported from previous studies can be taken as valid and predictive of the actual sensitivity of the animals. No consideration need be given to the time at which these measures were taken. (c) Region 3: A curious component of the ventilatory response was the consistent occurrence of a third region, which began between 6 and 8 % inspired CO 2 , where ventilation seemed to become independent of further increases in inspired CO 2 (Figure 6). This phase was terminated once arousal from hibernation was initiated (arousal threshold), at which time ventilation again increased. The level of ventilation associated with this region appeared constant although some individuals exhibited a slight decrease in ventilation as inspired CO 2 was increased over the range between the peak of the ventilatory response and the arousal threshold. Individuals showed consistency both in terms of the level of ventilation during this phase, and the level of CO 2 at which it first occurred regardless of when in a bout or over the season the response was measured. Contrary to this observation, most hibernation studies report a linear increase in VE with increasing levels of inspired CO2 . This phenomenon may not have been observed in other studies simply because CO2 levels were not taken high enough. For example, the hypercapnic ventilatory response in S. lateralis was characterised by McArthur and Milsom  80 (1991b) over a range between 0 and 8 % inspired CO 2 . Their data show an inflection in the frequency response with ventilation peaking and increasing no further between 6 % and 8 % inspired CO 2 . This level of inspired CO 2 corresponds to the point in the present study at which the third region of the ventilatory response was established. It is likely that had McArthur and Milsom continued to monitor the ventilatory response at higher levels of inspired CO2 , they would have also observed a similar phenomenon. Initially, it was considered possible that a temperature dependant mechanical limitation of the respiratory muscles was responsible for the lack of any further increase in ventilation during the third region of the hypercapnic ventilatory response. Any decrease in ventilation during this phase could then be accounted for by respiratory fatigue. However, this does not seem feasible since ventilation increases again once arousal is triggered, before any increase in temperature occurs. Figure 5 indicates that ventilation was also not limited by ventilatory pattern. T  ^is  the primary variable modified during the hypercapnic ventilatory response in hibernation (McArthur and Milsom, 1990b). The Tw p between breaths was the result of a pause following expiration. It is significant to note the variation in the duration of the endexpiratory pause at different levels of ventilation. These animals exhibited periods of greater breathing frequency and shorter T within individual breathing episodes in region 1 and 2, as well as during the arousal response in region 4, before T B began to rise than they did during the peak respiratory frequencies noted at the onset of region 3. Therefore, there was potential for further decreases in T and increases in Rf beyond the levels observed over the third region of the response. Some state dependant effect on integration within the ventilatory control centres could also explain the third region of the ventilatory response. In this case the descending effects  81 of higher centres would impose a "hibernation-mediated" ceiling onto the ventilatory motor output. Upon arousal, this ceiling would be lifted and ventilation could increase further. Although such a mechanism seems feasible, there is no evidence to support such an hypothesis. The simplest interpretation of this data, however, would be that the apparent constancy of ventilation during the third phase of the ventilatory response was a result of the secondary effect of hypercapnia acting centrally as a ventilatory depressant. In this case, a centrally mediated ventilatory depression would counterbalance, and in some cases, when ventilation decreases over this region, override the receptor mediated stimulation of ventilation. The anaesthetic qualities of CO2 are well documented (Goodman and Gilman, 1980). It seems quite reasonable to assume that the administration of CO 2 at such elevated levels (8 to 10 %) could have this effect. This interpretation, however, raises the question of how this "anaesthesia" is reversed when levels of inspired CO 2 become high enough to cause arousal. (d) Region 4: The third phase of the ventilatory response terminated when arousal from hibernation was initiated. The level of CO 2 at which this process occurred is referred to as the arousal threshold. Arousal was characterized by the initiation of thermogenesis as well as an increase in ventilation and a marked increase in neural (FPG) and muscle (EMG) activity. The increases in ventilation were most likely due to the addition of a descending central input onto the ventilatory control centres. Such central activation, resulting from the change in arousal state, must have either removed whatever influence was responsible for the third region of the hypercapnic ventilatory response, allowing the full magnitude of chemoreceptor activity to be reflected in respiratory drive, or acted in concert with receptor input to overcome this previous influence. Once arousal had been triggered, overall levels of  82 ventilation increased. Observations were terminated once arousal was initiated, however, and thus the post-arousal response was not characterized beyond this point. Obviously, arousal thresholds could not be characterized over individual bouts of hibernation as arousal marks the termination of a hibernation bout. Comparisons were made, however, between the arousal thresholds exhibited by individual animals over the season, and between the group means of animals tested during early, mid and late season (Table 1, Figure 13). Individual animals showed no significant seasonal variation in their arousal thresholds. There was a trend for arousal thresholds to be greatest in mid season, and less during early and late season. Unfortunately, due to the low number of replicates and the high variation between individuals, the apparent differences in mean early, mid and late season arousal thresholds were not statistically significant. A rough comparison of this trend with the established pattern of variation in hibernation bout length (see below) reveals strong similarities. Thus, the data suggest that all four regions the hypercapnic ventilatory response remain consistent over a hibernation bout and over the hibernation season in animals hibernating at constant body and ambient temperatures. Since there is data to suggest that hibernation depth does not remain constant (Twente and Twente 1968) this would suggest that resting ventilation and ventilatory responses to respiratory stimuli vary with body temperature and not hibernation depth. This is supported by the findings of Vessal (1983) who observed that resting levels of metabolic rate, breathing frequency and minute ventilation all scaled directly with body temperature in hypothermic ground squirrels, and that this relationship extrapolated to the values found for hibernating animals. From this, Vessal also concluded that ventilatory drive was related, to and could be predicted by, body temperature alone.  83 Pattern of Hibernation (i) Seasonal Changes in Body Temperature The changes in body temperature measured during year 1 showed the early season phenomenon identified by Strumwasser (1959) as "test dropping" in the squirrel C. beecheyi or as the "autumn pattern of hibernating behaviour" described in C.(S.) lateralis by Twente & Twente (1968). This phenomenon is not the result of cessation of temperature regulation, rather it represents a gradual adjustment of the set point for temperature regulation (T sET ) (Heller et al. 1978). During slow wave sleep (SWS),  TB is  actively regulated at a lower  ;ET  than during wakefulness (Heller et al. 1978). Preceding the onset of the hibernation season, these animals exhibit an increase in total time spent in sleep per day, and a proportionate increase in time spent in SWS versus rapid eye movement (REM) or paradoxical sleep. In addition they appear to progressively reduce their  TBET  further during each successive sleep  period (Heller et al., 1978) and extend the period over which TBET is decreased (Figure 11). This observation suggests a progression from SWS to torpor. Eventually  ;ET  approaches  ambient temperature and, by definition, deep torpor or true hibernation is achieved. These observations are consistent with the concept of torpor as a continuum, theoretically bounded by slow wave sleep at one end, and seasonal hibernation at the other, and are therefore supportive of the theory that hibernation can be considered as an extreme extension of SWS (Heller et al., 197R). It must be kept in mind that the temperature data from the present study were collected as single, daily observations, and therefore do not represent a complete record of the fluctuations in TB. They do however provide a rough idea of the duration and magnitude of the temperature fluctuations, and confirm that during hibernation,  TB remains  constant during  84 a bout of hibernation and over the hibernation season (Geiser et al. 1990).  (ii) Seasonal Changes in Bout Length The observed pattern of seasonal variation in bout length (Figure 12) is consistent with that found in C.(S.) lateralis by Twente & Twente (1968); Strumwasser et al.(1964); Pengelley & Fisher (1961); and by French (1976) in a number of other hibernating mammals. With this pattern, periodic spontaneous arousals punctuate bouts of hibernation which vary in length over the season in a characteristic fashion. Bouts tend to increase in duration from early to mid-season, and decrease in duration from mid to late season (Figure 12). The data collected from "experimental" and "control" animals indicate that no apparent change in hibernation pattern resulted from disruption due to experimental treatment. Experimental animals exhibited patterns of body temperature and bout length variation which were consistent with control animals and animals in the wild (Geiser et al., 1990; McKeever, 1964). It is reassuring to know that this study did not substantially alter the natural pattern of hibernation. Twente & Twente (1968) arrived at the same conclusion finding that vigorous handling and blood sampling during actual hibernation did not alter an animal's subsequent hibernation behaviour.  (iii)  Seasonal Changes in Arousal Threshold & Hibernation Depth  Hibernation depth can be assessed by measuring the amount of stimulation necessary to produce arousal. To quantify hibernation depth in this way, however, it is imperative to  quantify the level of stimulation required to produce arousal. This is a problem that has confounded researchers using stimuli such as noise, pain, heat, and injected pharmaceuticals to generate arousal (Willis, 1982). Decreases in temperature during hibernation can also cause arousal. This has been  85 documented by many authors (Willis, 1982). Geiser et al. (1990), working with Citellus saturatus (a supposedly distinct species of golden-mantled squirrel from the Cascade  mountains of Washington State) defined a minimum body temperature (T Bmin) to which hibernating animals could be brought before arousal from hibernation occured. The magnitude of the thermal stress (the TBmin) required to prompt arousal is a measure of hibernation depth. Geiser et al. (1990) found that TBmin was inversely related to bout length and exhibited a regular pattern of seasonal fluctuation, being least during mid-hibernation and greatest at the beginning and end of the season. In the present study, CO2 also acted as an arousal stimulus. This was quantified as the percent composition of CO 2 in the animals environment when arousal was initiated. Given the observation that the animal's sensitivity to CO 2 did not change over the hibernation season, any change in the level of CO 2 required to produce arousal must have been due to a change in the animals resistance to arousal; ie. a change in the depth of hibernation. Hibernation depth therefore was measured in terms of the level of ambient CO 2 required to bring about arousal from hibernation. The observed levels of inspired CO2 prompting arousal from hibernation over the season would indicate that hibernation depth did not change over the coarse of the hibernation season (Table 1, Figure 13). Although not statistically significant, however, there was a trend for hibernation depth to increase from early to mid season and decrease again towards the end of the season. The similar trends seen in changes in bout length and the increase in inspired CO 2 or decrease in TB required to cause arousal from hibernation seem more than coincidental. On the strength of the present data, however, little else can be said.  86 CONCLUSION The present study asked three questions; what is the overall character and sensitivity of the ventilatory response to changing levels of inspired CO 2 , at what level of inspired CO2 will arousal from hibernation be initiated, and do these characteristics vary with time either during a bout of hibernation or over the hibernation season ? The relationship between minute ventilation and inspired levels of CO 2 was complex with four characteristic regions. Low levels of inspired CO 2 did not stimulate ventilation significantly. There was an inflection point in the hypercapnic ventilatory response at approximately 2 % inspired CO2 . Above this level, ventilation did increase significantly in response to increases in inspired CO 2 . This increase seemed to be linear, and predominantly based on an increase in Rf with no substantial change occurring in V T . A dramatic decrease in ventilatory sensitivity occured at approximately 6 to 8 % inspired CO 2 . As the level of inspired CO 2 was elevated beyond this, levels of ventilation either remained constant or decreased slightly. This phase of the ventilatory response terminated when further increases in inspired CO 2 caused arousal from hibernation, at which point ventilation again increased. There appeared to be a specific level of CO 2 necessary to produce arousal from hibernation; an arousal threshold. There appeared to be no change in the hypercapnic ventilatory response either within a bout of hibernation or over the course of the hibernating season. These data support the primary hypothesis that the hypercapnic ventilatory response remains constant during hibernation, both between days in a bout and over the course of the hibernation season. These findings validate the methods of data collection and presentation employed by previous studies of ventilation using hibernating animals. It is believed that future studies investigating the ventilatory responses of golden-mantled ground squirrels during hibernation need not control for time within a season or day within a bout. The only  87 benefit such considerations would appear to have is in determining times when an individual animal will be most likely to tolerate the experimental protocol.  88 LITERATURE CITED  Barnes,B.M., 1984. Influence of Energy Stores on Activation of Reproductive Function in Male Golden-mantled Ground Squirrels. J. Comp. Physiol B. 154: 421-425. Berger, R.J.,1984. Slow wave sleep, shallow torpor and hibernation: homologous states of diminished metabolism and body temperature. Biological Psychology 19:305-326. Cherniack, N.S., and G.S. Longobardo,1986. Abnormalities in respiratory rhythm. In: Handbook of physiology. Section 3: The respiratory System. S.R.Geiger,and A.P.Fishman (eds.). American Physiological Society, Bethesda, Maryland. 2:729-747. Cunningham,D.J.C., P.A.Robbins, and C.B.Wolff, 1986. Integration of Respiratory Responses to changes in Alviolar Partial Pressures of CO2 and 02 and in Arterial pH.In: Handbook of physiology. Section 3: The respiratory System. S.R.Geiger,and A.P.Fishman (eds.). American Physiological Society, Bethesda, Maryland. 2:475-528. Feldman, J.L.(1986). Neurophysiology of breathing in mammals. In: Handbook of Physiology -The Nervous System IV - Intrinsic Regulatory systems of the brain, edited by F.E. Bloom. Bethesda, American Physiology Society, pp.463-524. Feldman, J.L.,J.C.Smith,H.H.Ellenberger, C.A.Connelly, G.Liu, J.J.Greer,A.D.Lindsay, and M.R.Otto, 1990. Neurogenesis of respiratory rhythm and pattern: emerging concepts. Am.J.Physiol. 259: R879-R886. French, A.R.,1986. Patterns of Thermoregulation During hibernation, In: Living in the Cold: Physiological and Biochemical Adaptations. H.C. Heller, H.J. Musacchia and L.C.H. Wang (eds.). Elsevier, New York. Geiser, F.,S.Hiebert, and G.J.Kenagy, 1990. Torpor bout duration during hibernation season of two sciuroid rodents: interrelations with temperature and metabolism. Physiological Zoology 63(3):489-503. Goodman, L.S. and A Gilman,1980. The Pharmacological Basis of Therapeutics. Sixth Edition. Collier-MacMillan Canada Ltd. Toronto. Grahn,D.A., and C.Heller, 1989. Activity of most rostral ventromedial medulla neurons reflects EEG/EMG pattern changes. Am.J.Physiol. 257:R1496-R1505. Heller, H.C., G.L.Florant, S.F.Glotzbach, J.M.Walker, and R.J.Berger. 1978. Sleep and Torpor-Homologous adaptations for energy conservation. In: Dormancy and developmental arrest. Experimental analysis in plants and animals. M.E.Clutter, ed. New York, Academic Press. PP.269-296.  89 Kayser, C., 1953. L'hibernation des maminiferes. Annee Biol. 29:109-150. As reported by Willis, J.S., 1982. The Mystery of the Periodic Arousal. In Hibernation and Torpor in Mammals and Birds. pp. 92-103, Academic Press, New York. Krilowicz, B.L.,S.F.Gotzbach, and H.C.Heller, 1988. Neuronal activity during sleep and complete bouts of hibernation. Am.J.Physiol. 255:R1008-R1019. Kuhnen, G., 1986. 02 and CO 2 concentrations in burrows of euthermic and hibernating golden hamsters. Comp.Biochem.Physiol. 84A: 517-522. Lee, T.E., K.Pelz, P.Licht, and I.Zucker, 1990. Testosterone influences hibernation in goldenmantled ground squirrels. Am.J.Physiol. 259:R760-R767. Lyman, C.P., 1982a. Who is among the hibernators. In: Hibernation and Torpor in Mammals and Birds. C.P. Lyman, J.S. Willis, A. Malan and L.C.S. Wang (eds.). Academic Press, New York. Pp.12-36. Lyman, C.P., 1982b. The Hibernating State. In: Hibernation and Torpor in Mammals and Birds. C.P. Lyman, J.S. Willis, A. Malan and L.C.S. Wang (eds.). Academic Press,^New York.Pp.54-76 Maginniss, L.A.,E.S.Lo, and W.K.Milsom, 1989. Effects of hibernation on blood oxygen transport in Golden-mantled ground squirrels. FASEB Journal (Abstracts) Vol.3 No.3 Pt.I Pp.A234. McArthur, M.D., and W.K.Milsom, 1991a. Ventilation and respiratory sensitivity of euthermic Columbian and Golden-mantled ground squirrels (Spermophilus columbianus and Spermophilus lateralis) during summer and winter. Physiological Zoology 64(4):921-939. McArthur, M.D., and W.K.Milsom, 1991b. Changes in ventilation and respiratory sensitivity associated with hibernation in columbian (Spermophilus columbianus) and golden-mantled (Spermophilus lateralis) ground squirrels. Physiological Zoology 64(4):940-959. McKeever, S.,1964. Biology of the Golden-mantled ground squirrel Citellus lateralis. Ecology Monographs 34(4):383-401. Milsom, W.K., 1991. Intermittent breathing in vertebrates. Annual reviews in physiology 53:87-105. Milsom, W.K., 1990. Mechanoreceptor modulation of endogenous respiratory rhythms in vertebrates. American Journal of Physiology 259:R898-R910. Milsom, W.K., M.D. McArthur and C.L. Webb, 1986. Control of Breathing in Hibernating Ground Squirrels. In: Living in the Cold: Physiological and Biochemical Adaptations. H.C. Heller, H.J. Musacchia and L.C.H. Wang (eds.). Elsevier, New York.  90 Nielson, M., and H. Smith,1952. Studies on the Regulation of Respiration in Acute Hypoxia, Acta. Physiol. Scand. 24:293-313. Onal, E.,1988. Central Sleep Apnoea. Seminars in Respiratory medicine, 9(6):547-553. Pearson, R.D.,and A.C.Greenaway, 1990. Sudden Infant Death Syndrome and Hibernation: Is there a link? Medical Hypotheses 31:131-134. Pengelley, E.T., and K.C. Fisher, 1961. Rhythmical Arousal from Hibernation in the Goldenmantled ground squirrel, Citellus Lateralis Tescorum. Can.J.Zool. 39: 105-120. Phillipson,E.A., and G.Bowes, 1986. Control of breathing during sleep. In: Handbook of Physiology. Section 3. The Respiratory system. Volume 2 Part II: Control of breathing. R.F.Geiger and P.A.Fishman (eds.) American Physiological Society, Bethesda, Maryland. 2:649-689. Remmers, J.E., 1981. Effects of sleep on control of breathing. International Review of Physiology III, Vol.23. J.G.Widdicombe (Ed.), pp.111-147. Smatresk, N.J., 1990. Chemoreceptor modulation of endogenous respiratory rhythms in vertebrates. American Journal of Physiology 28:R887-R897. Stoohs, R., and C.Guilleminault, 1990. Obstructive Sleep Apnoea Syndrome or abnormal upper airway resistance during sleep? Journal of Clinical Neurophysiology 7(1):83-92. Strumwasser, F., 1959. Factors in the Pattern, Timing and predictability of Hibernation in the Squirrel, Citellus beecheyi. Am. J. Physiol. 196: 8-14. Strumwasser, F.,J.J. Gillman, and J.L. Smith, 1964. Long Term Studies on Individual Hibernating animals. Ann.Acad.Sci.Fenn.Ser. A, IV. 71:401-414. Tenney,S.M. and D.F. Boggs (1986). Comparative mammalian respiratory control. In: Handbook of Physiology - The Respiratory system - Control of breathing, Part II, edited by A.P. Fishman. Bethesda, American Physiology Society, pp.833-855. Twente, J.W., and J.A. Twente, 1967. Progressive irritability of Hibernating Citellus  Lateralis. Comp.Biochem.Physiol. 25: 467-474.  Twente, J.W., and J.A. Twente, 1968. Seasonal Variation in the Hibernating Behaviour of Citellus Lateralis. In: Proc. 3rd int. Symp. Natur. Mammalian Hibernation Vol III. K.C. Fisher, A.W. Dawe, C.P. Lyman, E. Schonbaum, and F.E. South (eds.). pp. 47-63, Oliver & Boyd, Edinburgh. Vessal, S.(Osborn), 1983. Effects of Hypothermia on Ventilation and Ventilatory Responses to Hypercapnia and Hypoxia in the Golden-mantled Ground Squirrel and wistar rat. M.Sc. Thesis, University of British Columbia.  91 Webb, C.L.,1987. Aspects of the Control of Breathing in The Golden-Mantled Ground Squirrel. M.Sc. Thesis, University of British Columbia. Webb, C.L., and W.K.Milsom, 1990. Carotid body contribution to hypoxic ventilatory responses in euthermic and hibernating ground squirrels. In: Arterial Chemoreception. Eyaguirre, Fidone, Fitzgerald, Lahiri and McDonald (eds.). Springer-Verlag, New York. pp:337-343. Williams, B.A., J.Smyth,A.W.Boon, M.A.Hanson, P.Kumar, and C.E.Blanco, 1991. Development of respiratory chemoreflexes in response to alternations of fractional inspired oxygen in the newborn infant. Journal of Physiology, 442:81-90. Williams, DD., and R.L. Rausch (1973). Seasonal carbon dioxide and oxygen concentrations in the dens of hibernating mammals (Scuridae). Comp.Biochem. Physiol., 44A: 1227-1235. Willis, J.S., 1982. The Mystery of the Periodic Arousal. In Hibernation and Torpor in Mammals and Birds. pp. 92-103, Academic Press, New York. Wang, L.C.(1979). Time pattern and metabolic rates of natural torpor in the richardson's ground squirrel. Can.J.Zool., 57:149-155.  

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086474/manifest

Comment

Related Items