| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Journal of Physiology (2001), 536.1, pp. 225-235
© Copyright 2001 The Physiological Society
 |
ABSTRACT |
|---|
 TopAbstract IntroductionMethodsResultsDiscussionReferences |
|---|
I) and metabolic rate (CO2) were measured by plethysmography. Recordings were made over 24 h (12:12 h light:dark) when rats were in established states of wakefulness, NREM sleep and REM sleep.
I and CO2 in each of the wakefulness, NREM sleep and REM sleep states. Amplitudes and phases of the circadian rhythms were similar across sleep-wake states.
I was mediated by a circadian rhythm in respiratory frequency (fR). Tidal volume (VT) was unaffected by time of day in all three sleep-wake states.
I was significantly greater during wakefulness (363.5 ± 18.5 ml min-1) than during NREM sleep (284.8 ± 11.1 ml min-1) and REM sleep (276.1 ± 13.9 ml min-1). CO2 and VT each significantly decreased from wakefulness to NREM sleep to REM sleep. fR was significantly lower in NREM sleep than in wakefulness and REM sleep.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
It is well established that sleep and breathing are functionally related. Changes in sleep-wake states are associated with changes in respiratory motor output, lung ventilation and metabolic rate in animals and man, and conversely, chemical and mechanical respiratory stimuli can elicit changes in sleep-wake state (Phillipson et al. 1977; Phillipson & Bowes, 1986). These mechanisms are implicated in a variety of sleep disorders, such as obstructive sleep apnoea and sleep-related hypoventilation, which are now recognised as a significant cause of morbidity and mortality in Western society (NCSDR, 1993; Phillipson, 1993). In some respiratory disorders, such as nocturnal asthma, the circadian timing system plays a substantial role in the aetiology of the disease, with an additional role for sleep-related mechanisms (Chan et al. 1988; Ballard et al. 1989, 1990). Thus, two separate but linked mechanisms may contribute to respiratory function and dysfunction: sleep mechanisms and circadian mechanisms. Little is known, however, about how the circadian system interacts with sleep-related mechanisms in this context.
Recent research has implicated the circadian timing system in respiratory control in animals and man. It has been shown that minute ventilation oscillates with a 24 h period in rats maintained under a 12:12 h light:dark (LD) cycle, with sleep-wake state unknown (Seifert et al. 2000). In resting human subjects held under constant routine conditions (i.e. constant light, no sleep, and meals and other activities scheduled to a 2 h cycle), ventilation exhibited a circadian trend that bordered on statistical significance (Spengler et al. 2000). Furthermore, the respiratory chemoreflex, an important component of the respiratory control system, was found to exhibit significant circadian oscillation in awake healthy male subjects under constant routine conditions (Spengler et al. 2000; Stephenson et al. 2000), and in awake adult rats and birds under a LD cycle (Peever & Stephenson, 1997; Woodin & Stephenson, 1998). The amplitude of the circadian rhythm in respiratory control in awake human subjects was of significant magnitude (approximately 25 % of the mean), being comparable to the changes reported by others for transitions from wakefulness to sleep (Gothe et al. 1981; Douglas et al. 1982a; Stradling et al. 1985).
Thus, most previous studies either measured ventilation in awake subjects (Raschke, 1987; Raschke & Möller, 1989; Spengler et al. 2000; Stephenson et al. 2000), or did not take sleep-wake state into account (Seifert et al. 2000). Schäfer (1998) reported variable changes in hypercapnic ventilatory responses during 'deep sleep' from the first to the second half of the night in human subjects, suggesting that the circadian system may also influence respiratory control during non-rapid-eye-movement (NREM) sleep. This question clearly requires further study.
Sleep normally occurs at approximately the same time each day, a manifestation of circadian control of sleep-wake timing (Czeisler et al. 1980; Dijk & Czeisler, 1995). This strong correlation between circadian time and sleep makes it difficult to separate their roles in respiratory function (and dysfunction). We chose to study the rat because this species exhibits a polyphasic (ultradian) sleep pattern allowing ventilation to be measured during wakefulness, non-rapid-eye-movement sleep and rapid-eye-movement sleep (REM) at all times of the day, without the need to manipulate behaviour. In the present study, we addressed the hypothesis that sleep and circadian rhythms have synergistic effects on respiration. We measured ventilation and metabolic rate during electrographically defined sleep-wake states as a function of time of day. Preliminary results from this study have been published (Liao et al. 2000).
|
METHODS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Experiments were performed on six male Sprague-Dawley rats (mean ± S.E.M. body mass = 359 ± 19 g). Before and during experiments they were maintained on a 12:12 h LD cycle (lights on at 07.20 h) with food and water available ad libitum. All procedures conformed to the recommendations of the Canadian Council on Animal Care, and the experimental protocol was approved by the University of Toronto Animal Care Committee.
Protocol
Animals underwent surgical implantation of a three-channel radio transmitter (Model TL10M3F50-EET, Data Sciences International, St Paul, MN, USA). After at least 1 week of recovery, the rats were familiarised with the experimental apparatus by multiple short exposures over several days, during which time the parameters of a computer-based sleep scoring algorithm were set and validated (see 'Measurement of sleep-wake state' below). Sleep-wake state was scored using electroencephalographic (EEG) and electromyographic (EMG) criteria, and circadian phase was estimated from body temperature (Tb). EEG, EMG and Tb were acquired by radiotelemetry. Lung ventilation and metabolic rate (estimated as the rate of production of carbon dioxide) were measured non-invasively by whole-body plethysmography (Drorbaugh & Fenn, 1955).
Each rat spent two 34 h 'days' in the plethysmograph, with at least 24 h in the home cage between day 1 and day 2. The plethysmograph contained food, water and bedding from the home cage. On each experimental day, intermittent recordings began 10 h after introduction of the animal into the plethysmograph and continued for a further 24 h.
Surgical procedures
Sterile surgery was performed under general anaesthesia induced by intraperitoneal ketamine (85 mg kg-1) and xylazine (15 mg kg-1). Before surgery the rats were also given buprenorphine (0.03 mg kg-1), atropine sulphate (1 mg kg-1) and sterile saline (3 ml, 0.9 %). When necessary, anaesthesia was supplemented with halothane via an anaesthesia mask (typically 0.1-2 % in 50 % air-50 % O2). Body temperature was maintained at 36-38 °C using a heating pad and rectal temperature probe (BAS Inc., West Lafayette, IN, USA).
Mid-line incisions were made in the scalp and abdomen to expose the skull and peritoneal cavity, respectively. A radiotransmitter was inserted into the peritoneal cavity and loosely sutured to the rectus abdominus muscle using 3-0 non-absorbable silk. The electrode leads from the implant were tunnelled from the peritoneal cavity and led subcutaneously to the head. The muscle and skin of the abdomen were closed using 3-0 vicryl absorbable sutures.
The rat was placed in a prone position and its head was stabilised using blunt ear bars and an anaesthesia mask placed over the snout. The ear bars and mask were supported in a stereotaxic frame (Kopf Instruments, Tujunga, CA, USA). A pair of EMG electrodes were sutured bilaterally to the neck muscle. Three holes were drilled into the skull, and two EEG electrodes and a reference electrode were attached using stainless steel screws (size 0-80X1/16, Plastics One Inc., Roanoake, VA, USA). EEG electrodes were placed approximately 2 mm to the right and 2 mm anterior to bregma, and 2 mm to the left and 3 mm posterior to bregma, respectively. The reference electrode was placed approximately 3 mm to the left and 5 mm anterior to bregma. Dental acrylic was used to anchor the electrodes to the skull. The skin was closed with 3-0 absorbable vicryl and the rat was allowed to recover for at least 7 days before the experiments.
Measurement of lung ventilation
Tidal volume (VT, ml BTPS (body temperature and pressure, saturated)) and respiratory frequency (fR, breaths min-1) were measured using whole-body plethysmography. The plethysmograph consisted of two identical 9 l chambers in a water bath at room temperature (21-23 °C). One chamber contained the animal and the other served as a reference (thermobarometer) to correct for changes in ambient temperature and barometric pressure. Pressure fluctuations associated with lung ventilation were detected using a differential pressure transducer (Model DP-45-14, Validyne Engineering Corp., Northridge, CA, USA) connected to the two chambers. The relative humidity and temperature of the animal chamber were measured by a calibrated thermohygrometer probe (Model 37950-10, Cole-Parmer Instrument Co., Vernon Hills, IL, USA) placed in the animal chamber.
Except during brief recording intervals, the animal chamber was continuously flushed with air at 3 l min-1, which was sufficient to maintain CO2 concentration below 0.2 %. To measure ventilation, airflow was interrupted using solenoid-actuated pinch valves in the inlet and outlet tubes, thus closing the animal chamber. The valves were released if CO2 concentration reached 0.5 %.
VT was calculated by the method of Drorbaugh & Fenn (1955):
where Pm is the respiratory pressure deflection, Pcal is the pressure deflection caused by injection of a known volume of air (Vcal) into the animal chamber, Tb is body temperature (K), PB is barometric pressure (mmHg), PCH2O is water vapour pressure in the animal chamber and PAH2O is alveolar water vapour pressure, assumed to be saturated at Tb. A correction for nasal temperature (Epstein & Epstein, 1978) was not applied since this was shown in preliminary tests to lead to overestimation of tidal volume (E. J. Gucciardi and R. Stephenson, unpublished observations).
fR was calculated as 60/tTOT, where tTOT is total breath duration (s). Minute ventilation (I, ml min-1 BTPS) was calculated as fR 
VT.
Measurement of body temperature
Abdominal temperature was recorded by radiotelemetry. The temperature sensor, located in the body of the radiotransmitter was accurate to 0.1 °C with a response time constant of 40-50 s, as determined in preliminary experiments using step changes in temperature. The radio signals were detected by a receiver (Model RPC-1, Data Sciences International) placed beneath the plethysmograph. A frequency converter (Model UA-10 Universal Adapter, Data Sciences International) yielded a calibrated voltage that was then recorded on chart paper and on computer disk.
Measurement of metabolic rate
Metabolic rate was estimated by measuring the rate of accumulation of CO2 in the animal chamber during the intervals when the system was closed. A peristaltic pump fitted with pressure dampers on the inlet and outlet passed air at 50 ml min-1 through a desiccating column (Drierite) and an infrared CO2 analyser (Model CD-3A, Ametek Inc., Pittsburgh, PA, USA). Lag time from chamber to analyser was approximately 10 s and data were adjusted to account for this during analysis. The sampled air was returned to the animal chamber. The analyser was calibrated approximately every 12 h using premixed certified gas mixtures.
Rate of production of CO2 (CO2, ml min-1 STPD (standard temperature and pressure, dry)) was calculated as follows:
CO2 = VC d[CO2]/dt,
where VC is the volume of air (corrected to STPD) in the animal chamber and d[CO2]/dt is the rate of change of fractional CO2 concentration, derived by least squares linear regression of [CO2] versus time over approximately 1-2 min. TheVC of the empty chamber was determined by gas dilution, and corrected by subtraction of the volume of the experimental animal: Vrat = Mb/
, where Mb is body mass (g) and is body density, assumed to be 1.03 g cm-3.
Measurement of sleep-wake state
Sleep-wake states were defined using EEG and EMG data. The radiotransmitter used to acquire cortical EEG and neck EMG had a high frequency cut-off at 100 Hz. The raw signals were recorded on chart paper running at 5 mm s-1 (Model 78D, Grass Instruments, West Warwick, RI, USA) during intervals when the plethysmograph was closed. The EEG and EMG signals were also observed on a computer monitor, and subjected to online analysis (Kimoff et al. 1994; Hamrahi et al. 2001).
The interval histogram method (Kuwahara et al. 1988) was used to determine the relative (%) frequency content of the EEG signal. The following bandwidths were quantified: 
2 (0.5-2 Hz), 1 (2-4 Hz), 
(4-7.5 Hz), 
(7.5-13.5 Hz), 
1 (13.5-20 Hz) and 2 (20-30 Hz). In addition, the 2/1 ratio, and EEG and EMG amplitudes were determined. An online sleep-scoring algorithm, validated for use on rats (Hamrahi et al. 2001), was used to define sleep-wake state every 6 s. The threshold levels of 2/1 ratio and EMG amplitudes were used as scoring criteria in the algorithm. These values were calibrated for each rat during preliminary trials before the first recording day. The computer-generated judgement was verified by continuous visual inspection of the raw traces.
Data analysis
Ventilatory parameters, metabolic rate, body temperature, EEG and EMG were analysed for a minimum of 20 s during visually identified periods of established wakefulness, NREM sleep or REM sleep, i.e. at least 20 s had passed since the animal entered an unequivocally defined state. Data were selected for analysis no sooner than 20 s following closure of the plethysmograph to ensure that closure of the valves had not caused changes in sleep-wake state, and to allow the plethysmograph pressure baseline to stabilise. Thus, we analysed approximately 30-40 breaths within the 20-120 s interval following a change of sleep-wake state. Recordings containing any changes in sleep-wake state for one or more 6 s epochs were rejected. Furthermore, to avoid activity related artifacts during wakefulness, only those periods in which gross body movements were minimal were used, as indicated by absence of rapid baseline plethysmograph pressure deflections and high amplitude phasic EMG signals.
A subset of data were analysed further to ensure that 20 s samples were representative of longer recordings, and to determine whether systematic changes occurred over time (after-effects). For each rat (n = 6), six recordings of sustained sleep-wake state were selected for each of wakefulness, NREM sleep and REM sleep. In each recording, two samples containing 20 breaths were analysed. Sample 1 was at the same time as that used in the main data analysis and sample 2 began 1 min later. In each sample, VT, fR and I were measured on a breath-by-breath basis, and then a mean value was recorded for further statistical analysis. Student's two-tailed, paired sample t test was applied to the resulting 36 sample 1-sample 2 pairs in each sleep-wake state. There were no statistically significant differences (P > 0.05) between sample 1 and sample 2 in any variable in any sleep-wake state, validating the approach used in the main analysis.
In the main analysis, an effort was made to ensure that a recording was obtained in each sleep-wake state at regular intervals across the 24 h day, but this was not always possible, especially for REM sleep due to the pronounced circadian modulation of REM sleep propensity (Wurts & Edgar, 2000). To eliminate the effects of this potential sampling bias, respiratory data were collapsed into 2 h time bins, i.e. for each measured variable in each animal, in each sleep-wake state, on each experimental day, the data were grouped into 2 h time bins and multiple samples were averaged within a bin where necessary.
A mean value was computed for the 12 h light phase of day 1 and of day 2, and for the 12 h dark phase of day 1 and of day 2 for each rat in each sleep-wake state. A two-factor repeated measures analysis of variance (RMANOVA) was then performed for each of wakefulness, NREM sleep and REM sleep, with the factors time of day (light versus dark) and experiment order (day 1 versus day 2). Since no differences were found in any state between days 1 and 2, data were pooled across days for further analysis in each rat. For each physiological variable, a 24 h mean value was obtained for each animal in each state by computing the average of the respective light and dark means.
The two-factor RMANOVA described above (light versus dark and day 1 versus day 2) indicated significant light versus dark differences in several variables in each state, justifying a further analysis of circadian rhythmicity. The data were normalised before pooling across animals to eliminate differences in the average magnitude of variables between animals. This was accomplished by expressing each datum (in the original units) as a deviation from the within-animal state-specific 24 h mean. Pooled deviations were then subjected to a least-squares non-linear regression (SigmaPlot, SPSS Inc., Chicago, IL, USA) using the following model:
/b + c),
where y is the deviation of the physiological variable at time , a is the amplitude of the fitted curve (difference between peak value and the fitted mesor), b is the period (constrained to 24 h in this analysis since the animals were entrained to a LD cycle), c is the reference phase (6 h before time of the fitted peak, i.e. acrophase: /2 radians) and y0 is the mesor of the fitted sinusoid.
Tb was recorded continuously during experiments to facilitate analysis of circadian phase. The response time of the radio transmitter was inadequate to resolve temperature changes associated with sleep-wake state transitions, so the Tb data obtained during wakefulness, NREM sleep and REM sleep were pooled before collapsing the data into 1 h time bins. The Tb data for each individual animal were then fitted with a least squares non-linear regression as described above. Acrophases varied little between days and between animals (coefficient of variation = 4.6 %), as was predicted for animals entrained to a LD cycle. For this reason, we did not adjust (i.e. align) the phases of different animals before the pooling of the respiratory data.
For all physiological variables, a significant circadian rhythm was inferred from a significant regression ANOVA. For those variables exhibiting significant circadian rhythmicity, acrophase was expressed in zeitgeber time (ZT) where ZT0 is time of lights on and ZT12 is time of lights off. An unpaired t test was computed, with Bonferroni's correction, using the mean values and standard error of the mean generated in the least-squares regression analysis (Zar, 1984) to compare the acrophases and the amplitudes between different rhythmic physiological variables. This unpaired t test, with Bonferroni's correction, was also used to analyse the effects of sleep-wake state on the acrophases and the amplitudes of each respiratory variable.
The overall effect of sleep-wake state on physiological variables was assessed by comparison of the 24 h mean values using a one-factor RMANOVA.
Differences between groups were considered statistically significant at the 95 % confidence level (P < 0.05). Data are expressed as mean values ± S.E.M. All statistics were computed using SigmaStat software (SPSS Inc.).
|
RESULTS |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Effects of sleep-wake state on breathing and metabolism
Representative examples of raw EEG, EMG and respiratory recordings are illustrated in Fig. 1. The differences in the amplitudes and frequencies of the EEG and EMG typical of established wakefulness, NREM sleep and REM sleep are shown. Since the amplitudes and phases of respiratory circadian rhythms were found to be similar across wakefulness, NREM and REM sleep (see below), the effect of state per se could be assessed by comparison of the 24 h mean data. Table 1 presents the 24 h mean values of respiratory and metabolic variables as a function of sleep-wake state. There was a significant effect of sleep-wake state on 24 h mean I (F2,17 = 37.0, P < 0.001) and 24 h mean CO2 (F2,17 = 125.7, P < 0.001). The 24 h mean I was 28 % and 32 % higher during wakefulness than NREM sleep and REM sleep, respectively (Tukey's post hoc multiple comparisons test, P < 0.05), but I was not significantly different between NREM and REM sleep. The effect of sleep-wake state on 24 h mean I was mediated by state-related differences in both fR (F2,17 = 6.9, P = 0.013) and VT (F2,17 = 67.1, P < 0.001). The 24 h mean VT was significantly different in all three states, being highest in wakefulness and lowest in REM sleep. The 24 h mean fR was 8 % and 4 % lower during NREM sleep than during wakefulness and REM sleep, respectively (P < 0.05). However, the difference in fR between wakefulness and REM sleep was not statistically significant.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 1. Representative recordings of electroencephalogram (EEG), neck electromyogram (EMG) and respiratory pressure waveforms calibrated to indicate tidal volume (VT, ml) during wakefulness (A),NREM sleep (B) and REM sleep (C) Note the desynchronised (high frequency, low amplitude) pattern of the EEG in wakefulness and REM sleep, and the synchronised EEG pattern (lower frequency, higher amplitude) during NREM sleep. REM sleep was distinguished from wakefulness by the presence of muscle atonia in the former (EMG amplitude low in REM sleep, high in wakefulness). Slow changes in the ventilatory pressure baseline were due to changes in relative humidity in the closed plethysmograph. To facilitate direct comparison, the ventilatory pressure waveforms were adjusted to the same calibration scale. | ||
The 24 h mean CO2 was significantly different in all three states (F2,17 = 125.7, P < 0.001), being highest in wakefulness and lowest in REM sleep. The 24 h mean I/CO2 ratio (metabolic-rate-specific ventilation or air convection requirement) was found by RMANOVA to vary as a function of sleep-wake state (F2,17 = 6.7, P = 0.014). The I/CO2 ratio was 14 % higher during REM sleep than wakefulness (P < 0.05). However, I/CO2 during NREM sleep was not statistically different from that during wakefulness and REM sleep.
Body temperature
The variations in Tb over 24 h are illustrated in Fig. 2. The 24 h mean was 37.4 ± 0.1 °C. There was a significant circadian rhythm (F3,143 = 88.41, P < 0.0001). Amplitude (i.e. half the peak to trough difference) of the fitted sinusoidal curve was 0.51 ± 0.03 °C. Fitted acrophase (h:min) was at ZT 18:04 ± 0:29, mid-way through the dark phase of the LD cycle.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 2. Circadian rhythm in body temperature (Tb) in rats entrained to a 12:12 h light:dark cycle Zeitgeber time (ZT) is indicated on the abscissa, where ZT0 is the time of lights on. The white and black sections of the bar at the top of the figure indicate the light and dark phases of the day, respectively. Data were obtained in two experimental 'days' for each of six animals. Data were pooled across days and across sleep-wake states (see text for details) before collapsing into 1 h time bins. Data were fitted with a non-linear least-squares regression, from which amplitude and acrophase (time of peak) were derived. | ||
Circadian rhythms in breathing and metabolism
Circadian rhythms in respiration and metabolic rate are illustrated in Fig. 3. The amplitudes and acrophases are quantified in Table 1. There was a statistically significant circadian rhythm in I in wakefulness (F3,68 = 4.85, P = 0.0042), NREM sleep (F3,68 = 6.46, P = 0.0007) and REM sleep (F3,49 = 8.80, P = 0.0001). The amplitudes of these rhythms (expressed as a percentage of the 24 h mean) were not statistically different (7.6 % in wakefulness, 8.0 % in NREM sleep and 10.9 % in REM sleep) (P > 0.2). VT did not exhibit significant circadian rhythmicity in any state (P 
0.999). fR exhibited statistically significant circadian rhythmicity in wakefulness (F3,68 = 5.77, P = 0.0014) and NREM sleep (F3,68 = 6.76, P = 0.0005), but not in REM sleep (F3,49 = 0.0023, P = 0.9998). The amplitudes (as a percentage of the 24 h mean) of the circadian rhythms in fR were not statistically different between wakefulness (9.1 %) and NREM sleep (7.8 %) (P > 0.2). The circadian rhythm in I during REM sleep was due to synchronous non-significant trends in VT and, to a lesser extent, fR (Fig. 3). Mean inspiratory airflow (VT/tI, ml s-1) exhibited a significant circadian rhythm in NREM and REM sleep, but not in wakefulness (Table 1).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 3. Circadian rhythms in respiratory variables during periods of established wakefulness, NREM sleep and REM sleep Data for each of six animals were pooled across 2 experimental days and expressed as a deviation from the 24 h mean value (see text for details). The data from different animals were then combined as shown for non-linear least-squares regression analysis. For the purposes of illustration, the 24 h mean values were added to the data and fitted curves. A, minute ventilation ( | ||
CO2 showed circadian rhythmicity in wakefulness (F3,68 = 6.69, P = 0.0005), NREM sleep (F3,68 = 14.81, P < 0.0001) and REM sleep (F3,49 = 2.84, P = 0.048). The amplitudes of the circadian rhythms in CO2 (expressed as a percentage of 24 h mean) were not statistically different between sleep-wake states (wakefulness, 9.5 %; NREM sleep, 10.5 %; REM sleep, 9.5 %) (P > 0.5). Metabolism-specific ventilation (I/CO2) was arrhythmic in wakefulness (F3,68 = 1.58, P = 0.2025) and REM sleep (F3,49 = 0.07, P = 0.975). However, there was a statistically significant circadian rhythm (amplitude 5.9 % of the 24 h mean) in I/CO2 during NREM sleep (F3,68 = 3.71, P = 0.016). Within-animal correlation coefficients (r) for the relationship between I and CO2 ranged from 0.61 to 0.91. Averaged across animals, r = 0.83 ± 0.05 (n = 6) (P < 0.001).
In any given sleep-wake state, the acrophases of the circadian rhythms in I, fR and CO2 were statistically indistinguishable from each other (see Table 1, Fig. 4), and from the acrophase for Tb (P > 0.2). For I, fR and CO2, the circadian rhythms were also coincident across sleep-wake states (Fig. 4), i.e. for each of these variables, the acrophase was the same in wakefulness, NREM sleep and REM sleep. An exception was I/CO2 in NREM sleep, which peaked in the late light phase of the LD cycle, approximately 8 h in advance of the other measured variables.
 |
View larger version [in this window] [in a new window] |
|
|
Figure 4. Mean ± S.E.M. acrophase (time of fitted peak) for those variables exhibiting statistically significant circadian rhythms in wakefulness, NREM sleep and REM sleep Note that the circadian rhythms of all variables (except | ||
The equivalence across sleep-wake states of I rhythm amplitudes and acrophases implies that the masking effect of sleep on breathing was independent of time of day (Fig. 5), i.e. the decrease in I that occurred during the transition from wakefulness to sleep was the same at all times of day. Thus, the effects of circadian rhythms and sleep-wake state on breathing were additive (Fig. 5).
 |
View larger version [in this window] [in a new window] |
|
|
Figure 5. Circadian rhythms in minute ventilation ( The curves represent the sinusoids fitted to data by non-linear least-squares regression, as shown in Fig. 3A (continuous line, wakefulness; dashed line, NREM sleep; dotted line, REM sleep). The points represent means ± S.E.M. This figure illustrates the similarity across sleep-wake states of the amplitude and phase of the circadian rhythms, implying that the differences in | ||
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The present study has found that lung ventilation and metabolic rate (CO2) oscillate with a circadian rhythm in rats maintained under a 12:12 h LD cycle. We have further shown that these circadian rhythms are present during wakefulness, NREM sleep and REM sleep. A key finding is that the rhythms in I, CO2 and Tb were coincidental in all sleep-wake states, i.e. the acrophases did not differ between variables, nor between wakefulness, NREM sleep and REM sleep (Fig. 4). Taken together with the equivalence of the amplitudes of the circadian rhythms in the three sleep-wake states (Table 1), this implies that the effects of sleep-wake state and circadian time are additive in rats entrained to a LD cycle (Fig. 5).
Relative roles of circadian and sleep mechanisms in respiratory control
It has long been widely accepted that the transition from wakefulness to sleep is accompanied by reductions in respiratory motor output and chemoreflex responses to CO2 in animals and human subjects (Phillipson et al. 1977; Gothe et al. 1981; Douglas et al. 1982b; Ingrassia et al. 1991; Parisi et al. 1992). The present results are consistent with the hypothesis that the circadian timing system may exert an additional and physiologically significant effect on mammalian respiratory control. In the present study, the day-to-night difference in lung ventilation (approximately 15 % of the 24 h mean) was two-thirds the magnitude of the effect of falling asleep (a decrease of approximately 22 %). Thus, these results in the rat support those of recent work in awake human beings, showing that at the time of day normally associated with sleep, the circadian timing system has additional effects on respiratory control (Spengler et al. 2000; Stephenson et al. 2000). Indeed, the magnitude of this circadian effect was qualitatively and quantitatively similar to that of sleep in human subjects (Robin et al. 1958; Gothe et al. 1981; Douglas et al. 1982a; Stradling et al. 1985; White et al. 1985).
Potential clinical implications
Nocturnal respiratory dysfunction has been linked to respiratory depression in human beings as a consequence of the transition from wakefulness to sleep (Douglas et al. 1982a; Cherniak, 1984; Stradling et al. 1985; Phillipson & Bowes, 1986; Horner, 1996). The results of the present study are consistent with the suggestion (Cherniak, 1981; Raschke, 1987; Stephenson et al. 2000) that the circadian timing system could also have effects on the control of breathing that may further predispose people to nocturnal respiratory depression and/or instability. It is tempting to speculate that for patients with respiratory-related sleep disorders, if the additivity of sleep and circadian effects also applies to humans, daytime naps may be less deleterious to health than nocturnal sleep since the circadian system appears to at least partially counteract the effects of sleep on breathing during the time of day corresponding to the circadian peak.
In human subjects, the reductions in ventilation due to sleep and the circadian nadir are associated with changes in respiratory chemoreflex characteristics that have been proposed to increase the propensity for respiratory instability (Cherniak, 1981; Stephenson et al. 2000). We suggest that the circadian system may play a more prominent role in the aetiology of nocturnal breathing disorders (Raschke & Möller, 1989; Stephenson et al. 2000) than has previously been recognised.
Methodological considerations
There are some important differences between rats and normal human subjects that should be considered when extrapolating conclusions and speculating on possible clinical implications of this study. The rat is nocturnal, whereas human beings are usually diurnal, i.e. rats are predominantly awake and active during the night and are mainly resting or asleep during the day, and the converse is true for human beings. However, in both species, physiological and behavioural variables maintain a stable phase relationship to the LD cycle and the phase relationships between metabolism, Tb, sleep and activity are also similar (Honma & Hiroge, 1978; Weitzman et al. 1981), i.e. while the relationship to the LD cycle is reversed, the internal synchronisation of physiological systems is similar in diurnal and nocturnal animals (Refinetti, 1996). A potentially important exception is the rhythm in melatonin, which is maximal during the dark phase in both diurnal and nocturnal species. It has been suggested that melatonin is linked to the control of body temperature (Dawson & van den Heuvel, 1998), which in turn is functionally related to metabolism and breathing. It would therefore be of interest to examine the potential role of melatonin in circadian modulation of respiration, and to determine whether there are differences between nocturnal rodents and diurnal humans in this regard.
A second difference is that rats exhibit a pronounced ultradian (< 24 h) sleep-wake rhythm, whereas healthy human adults typically have highly consolidated bouts of wakefulness and sleep (Aschoff, 1965; Czeisler et al. 1999). As was mentioned in the Introduction, we chose to study the rat in large part to capitalise on its polyphasic sleep-wake pattern, i.e. while rats obtain most of their sleep in the light phase of the LD cycle, the sleep is interrupted by regular periods of active and inactive wakefulness. Similarly, rats are mainly awake at night, but they also enter NREM and REM sleep periodically (Trachsel et al. 1986; Wurts & Edgar, 2000). In order to determine whether there is a circadian rhythm in ventilation during a specific sleep-wake state, it is necessary to measure ventilation while the subject is in that state at all times across the circadian day. In animals and humans with consolidated sleep-wake patterns, this necessitates experimental manipulation of sleep-wake timing, such as scheduled imposition of awakenings and naps, unusual LD cycles, or forced desynchrony protocols. Such experimental procedures could conceivably mask the phenomenon under study (i.e. it may hide or induce an apparent circadian rhythm in ventilation). Despite the presence of ultradian oscillations in the rat data, the amplitudes of the circadian (i.e. near 24 h) component of the rhythms are not markedly dissimilar in rats and man for the variables examined here. Furthermore, while healthy human adults have consolidated circadian sleep-wake patterns, a more polyphasic and disrupted pattern is seen in young children, elderly people and, significantly in the present context, people with sleep-disordered breathing (Phillipson et al. 1977; Czeisler et al. 1999; Dijk et al. 1999).
Interrelationships between respiratory variables across sleep-wake states and circadian time
The circadian rhythm in I was effected by changes in fR, suggesting that the circadian timing system may act either directly or indirectly on the timing of the central respiratory rhythm generator. The circadian rhythms in fR during wakefulness and NREM sleep were due to rhythms in both inspiratory time (tI) and expiratory time (tE).
In the present study, VT did not exhibit circadian rhythmicity, contradicting a recent report on rats (Seifert et al. 2000). The latter study, however, did not discriminate between sleep-wake states, and we suggest that the reported circadian oscillation in VT may reflect the circadian distribution of wakefulness and sleep, rather than an actual circadian rhythm in VT. In the present study, VT contributed significantly to the lower I in sleep compared with wakefulness, but in any given sleep-wake state, VT was relatively constant across the day. Carley et al. (1997) reported wakefulness to NREM sleep differences in both fR and I for rats that are very similar to those observed in the present study. This implies that there were similar state-dependent changes in VT in the two studies. Rats are awake for approximately 75 % of the time in the dark, and are asleep for approximately 75 % of the time in the light (Trachsel et al. 1986). Hence, average VT will tend to be higher at night than during the daytime, supporting the suggestion that the apparent circadian rhythm in VT previously observed by Seifert et al. (2000) was a masking effect of the sleep-wake state.
Lung ventilation and metabolic rate both varied in parallel over the day, as was suggested previously in a day-to-night comparison of I/CO2 in awake rats (Peever & Stephenson, 1997). Here we have shown that the amplitudes and phases of the respective circadian rhythms were similar in each sleep-wake state. A close correlation between I and oxygen consumption (O2) across sleep-wake states has also been reported in human subjects (White et al. 1985) .
Since I tracked CO2 across sleep-wake states and circadian time in the present study, this implies that arterial partial pressure of CO2 (Pa,CO2) also varied little between wakefulness and sleep, and day and night. However, the non-invasive technique used here measures only CO2 eliminated from the lung and cannot determine whether there are changes in endogenous CO2 stores. Robin et al. (1958) concluded that there is no diurnal cycle of end-tidal PCO2 independent of sleep in awake human subjects, and Spengler et al. (2000) observed a very small amplitude circadian rhythm in end tidal PCO2. We did not measure blood gases in this study to avoid possible confounding effects of surgical intervention (arterial cannulation) and additional restraint (tether).
Accumulation of CO2 in the body fluids during transitions from wakefulness to NREM sleep, as might be expected from the results of sleep studies in humans and other mammals (Robin et al. 1958; Phillipson & Bowes, 1986), would imply that our measured decrease in CO2 (Table 1) represents a slight overestimation of the actual state-related decrease in metabolic rate. Unfortunately, our attempts in the present study to measure O2, which would have provided a more reliable index of metabolic rate, were thwarted by equipment failure. Robin et al. (1958) and White et al. (1985) both found that O2 and CO2 followed similar trends in sleeping human subjects, although CO2 fell significantly more than O2 between wakefulness and stages 3 and 4 NREM sleep in the latter study. Robin et al. (1958) reported that the respiratory exchange ratio was not different between wakefulness and sleep. Shapiro et al. (1984) used an analysis of both O2 and CO2 to derive metabolic heat production, but did not report the CO2 data. Several studies have presented evidence that both sleep-wake state and time of night influence metabolic rate in humans (Shapiro et al. 1984; White et al. 1985; Fraser et al. 1989; Ryan et al. 1989). However, we are not aware of any such studies in rats.
An unexpected finding was that there was a significant circadian rhythm in I/CO2 during NREM sleep. Furthermore, this rhythm was approximately 8 h phase advanced compared with the rhythms in I and CO2. The small (< 2 h) non-significant difference in phase between I and CO2 in NREM sleep cannot explain the rhythm in I/CO2. The origin and functional significance of this rhythm in I/CO2 during NREM sleep, therefore, remains unclear.
In summary, this study has shown that circadian rhythms in lung ventilation and metabolic rate occur during both wakefulness and sleep in freely behaving rats. Furthermore, the transition from wakefulness to sleep was followed by decreases in metabolic rate and ventilation. Significantly, the magnitudes of the sleep- wake state-dependent changes in respiration were independent of time of day, i.e. the phase and amplitudes of circadian rhythms in metabolic rate and ventilation during wakefulness were similar to the respective rhythms during sleep. We conclude that the effects of circadian time and sleep-wake state on respiration are additive in rats entrained to a LD cycle, and suggest that this may have important implications for respiratory control in health and disease.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
| ASCHOFF, J. (1965). Circadian rhythms in man. Science 148, 1427-1432 | |
| BALLARD, R. D., IRVIN, C. G., MARTIN, R. J., PAK, J., PANDEY, R. & WHITE, D. F. (1990). Influence of sleep on lung volume in asthmatic patients and normal subjects. Journal of Applied Physiology 68, 2034-2041 | [Medline] |
| BALLARD, R. D., SAATHOFF, M. C., PATEL, D. K., KELLY, P. L. & MARTIN, R. J. (1989). Effect of sleep on nocturnal bronchoconstriction and ventilatory patterns in asthmatics. Journal of Applied Physiology 67, 243-249 | [Medline] |
| CARLEY, D. W., TRBOVIC, S. M., BOZANICH, A. & RADULOVACKI, M. (1997). Cardiopulmonary control in sleeping Sprague-Dawley rats treated with hydralzine. Journal of Applied Physiology 83, 1954-1961 | [Abstract/Full Text] |
| CHAN, C. S., WOOLCOCK, A. J. & SULLIVAN, C. E. (1988). Nocturnal asthma: role of snoring and obstructive sleep apnea. American Review of Respiratory Disease 137, 1502-1504 | [Medline] |
| CHERNIACK, N. S. (1984). Sleep apnea and its causes. Journal of Clinical Investigation 73, 1501-1506 | [Medline] |
| CHERNIACK, N. S. (1991). Periodic breathing. In Sleep and Cardiorespiratory Control, vol. 217, ed. GAULTIER, C., ESCOURROU, P. & CURZI-DASCALOVA, L., pp. 15-21. Inserm/John Libbey Eurotext Ltd, London | |
| CZEISLER, C. A., DUFFY, J. F., SHANAHAN, T. L., BROWN, E. N., MITCHELL, J. F., RIMMER, D. W., RONDA, J. M., SILVA, E. J., ALLAN, J. S., EMENS, J. S., DIJK, D. J. & KRONAUER, R. E. (1999). Stability, precision, and near-24 h period of the human circadian pacemaker. Science 284, 2177-2181 | |
| CZEISLER, C. A., WEITZMAN, E., MOORE-EDE, M. C., ZIMMERMAN, J. C. & KNAUER, R. S. (1980). Human sleep: its duration and organization depend on its circadian phase. Science 210, 1264-1267 | |
| DAWSON, D. & VAN DEN HEUVEL, C. J. (1998). Integrating the actions of melatonin on human physiology. Annals of Medicine 30, 95-102 | [Medline] |
| DIJK, D. J. & CZEISLER, C. A. (1995). Contribution of the circadian pacemaker and the sleep homeostat to sleep propensity, sleep structure, electroencephalographic slow waves, and sleep spindle activity in humans. Journal of Neuroscience 15, 3526-3538 | [Abstract] |
| DIJK, D. J., DUFFY, J. F., RIEL, E., SHANAHAN, T. L. & CZEISLER, C. A. (1999). Ageing and the homeostatic regulation of human sleep during forced desynchrony of rest, melatonin and temperature rhythms. Journal of Physiology 516, 611-627 | [Abstract/Full Text] |
| DOUGLAS, N. J., WHITE, D. P., PICKET, C. K., WEIL, J. V. & ZWILLICH, C. W. (1982a). Respiration during sleep in normal man. Thorax 37, 840-844 | [Abstract] |
| DOUGLAS, N. J., WHITE, D. P., WEIL, J. V., PICKETT, C. K. & ZWILLICH, C. W. (1982b). Hypercapnic ventilatory response in sleeping adults. American Review of Respiratory Disease 126, 758-762 | [Medline] |
| DRORBAUGH, J. E. & FENN, W. O. (1955). A barometric method for measuring ventilation in newborn infants. Pediatrics 16, 81-89 | [Abstract] |
| EPSTEIN, M. A. & EPSTEIN, R. A. (1978). A theoretical analysis of the barometric method for measurement of tidal volume. Respiration Physiology 32, 105-120 | [Medline] |
| FRASER, G., TRINDER, J., COLRAIN, I. M. & MONTGOMERY, I. (1989). Effect of sleep and circadian cycle on sleep period energy expenditure. Journal of Applied Physiology 66, 830-836 | [Medline] |
| GOTHE, B., ALTOSE, M. D., GOLDMAN, M. D. & CHERNIACK, N. S. (1981). Effect of quiet sleep on resting and CO2-stimulated breathing in humans. Journal of Applied Physiology 50, 724-730 | [Medline] |
| HAMRAHI, H., CHAN, B. & HORNER, R. L. (2001). On-line detection of sleep-wake states and application to produce intermittent stimuli exclusively in sleep in rats. Journal of Applied Physiology 90, 2130-2140 | [Abstract/Full Text] |
| HONMA, K.-I. & HIROSHIGE, T. (1978). Internal synchronization among several circadian rhythms in rats under constant light. American Journal of Physiology 235, R243-249 | [Medline] |
| HORNER, R. L. (1996). Motor control of the pharyngeal musculature and implications for the pathogenesis of obstructive sleep apnea. Sleep 19, 827-853 | [Medline] |
| INGRASSIA, T. S., III, NELSON, S. B., HARRIS, C. D. & HUBMAYR, R. D. (1991). Influence of sleep state on CO2 responsiveness. American Review of Respiratory Disease 144, 1125-1129 | [Medline] |
| KIMOFF, R. J., MAKINO, H., HORNER, R. L., KOZAR, L. F., LUE, F., SLUTSKY, A. S. & PHILLIPSON, E. A. (1994). A canine model of obstructive sleep apnea: model description and preliminary application. Journal of Applied Physiology 76, 1810-1817 | [Medline] |
| KUWAHARA, H., HIGASHI, H., MIZUKI, Y., MATSUNARI, S., TANAKA, M. & INANAGA, K. (1988). Automatic real-time analysis of human sleep stages by an interval histogram method. Electroencephalography and Clinical Neurophysiology 70, 220-229 | [Medline] |
| LIAO, K. S., STEPHENSON, R., HAMRAHI, H., CHAN, B. & HORNER, R. L. (2000). Ventilation in wakefulness and sleep varies as a function of time of day. American Journal of Respiratory and Critical Care Medicine 161, A864 | |
| NATIONAL COMMISSION ON SLEEP DISORDERS RESEARCH (NCSDR) (1993). Wake Up America: A National Sleep Alert. Government Publishing Offices Washington, DC | |
| PARISI, R. A., EDELMAN, N. H. & SANTIAGO, T. V. (1992). Central respiratory carbon dioxide chemosensitivity does not decrease during sleep. American Review of Respiratory Disease 145, 832-836 | [Medline] |
| PEEVER, J. H. & STEPHENSON, R. (1997). Day-night differences in the respiratory response to hypercapnia in awake adult rats. Respiration Physiology 109, 241-248 | [Medline] |
| PHILLIPSON, E. A. (1993). Sleep apnea - a major public health problem. New England Journal of Medicine 328, 1271-1273 | [Full Text] |
| PHILLIPSON, E. A. & BOWES, G. (1986). Control of breathing during sleep. In Handbook of Physiology, section 3, The Respiratory System, vol. II, Control of Breathing, part 2, ed. CHERNIACK, N. S. & WIDDICOMBE, J. G., pp. 649-689. American Physiological Society, Bethesda, MD | |
| PHILLIPSON, E. A., KOZAR, L. F., REBUCK, A. S. & MURPHY, E. (1977). Ventilatory and waking responses to CO2 in sleeping dogs. American Review of Respiratory Disease 115, 251-259 | |
| RASCHKE, F. (1987). Various components of respiratory control during sleep, rest, and strain. In Sleep Related Disorders and Internal Diseases, ed. PETER, J. H., PODSZUS, T. & VON WICHERT, P., pp. 83-88. Springer-Verlag, Berlin | |
| RASCHKE, F. & MÖLLER, K. H. (1989). Untersuchungen zur Tagesrhythmik der Chemosensitivität und deren Beitrag zu nächtlichen Atmungsregulationsstörungen. Pneumologie 43, 568-571 | [Medline] |
| REFINETTI, R. (1996). Comparison of the body temperature rhythms of diurnal and nocturnal rodents. Journal of Experimental Zoology 275, 67-70 | [Medline] |
| ROBIN, E. D., WHALEY, R. D., CRUMP, C. H. & TRAVIS, D. M. (1958). Alveolar gas tensions, pulmonary ventilation and blood pH during physiologic sleep in normal subjects. Journal of Clinical Investigation 37, 981-989 | |
| RYAN, T., MLYNCZAK, S., ERICKSON, T., MAN, S. F. P. & MAN, G. C. W. (1989). Oxygen consumption during sleep: influence of sleep stage and time of night. Sleep 12, 201-210 | [Medline] |
| SEIFERT, E. L., KNOWLES, J. & MORTOLA, J. P. (2000). Continuous circadian measurements of ventilation in behaving adult rats. Respiration Physiology 120, 179-183 | [Medline] |
| SHAPIRO, C. M., GOLL, C. C., COHEN, G. R. & OSWALD, I. (1984). Heat production during sleep. Journal of Applied Physiology 56, 671-677 | [Medline] |
| SCHÄFER, T. (1998). Variability of vigilance and ventilation: studies on the control of respiration during sleep. Respiration Physiology 114, 37-48 | [Medline] |
| SPENGLER, C. M., CZEISLER, C. A. & SHEA, S. A. (2000). An endogenous circadian rhythm of respiratory control in humans. Journal of Physiology 526, 683-694 | [Abstract/Full Text] |
| STEPHENSON, R., MOHAN, R. M., DUFFIN, J. & JARSKY, T. M. (2000). Circadian rhythms in the chemoreflex control of breathing. American Journal of Physiology 278, R282-286 | |
| STRADLING, J. R., CHADWICK, G. A. & FREW, A. J. (1985). Changes in ventilation and its components in normal subjects during sleep. Thorax 40, 364-370 | [Abstract] |
| TRACHSEL, L., TOBLER, I. & BORBÉLY, A. A. (1986). Sleep regulation in rats: effects of sleep deprivation, light, and circadian phase. American Journal of Physiology 251, R1037-1044 | [Medline] |
| WEITZMAN, E. D., CZEISLER, C. A., ZIMMERMAN, J. C. & MOORE-EDE, M. C. (1981). Biological rhythms in man: relationship of sleep-wake, cortisol, growth hormone, and temperature during temporal isolation. In Neurosecretion and Brain Peptides, ed. MARTIN, J. B., REICHLIN, S. & BICK, K. L., pp. 475-499. Raven Press, New York | |
| WHITE, D. P., WEIL, J. V. & ZWILLICH, C. W. (1985). Metabolic rate and breathing during sleep. Journal of Applied Physiology 59, 384-391 | [Medline] |
| WOODIN, M. A. & STEPHENSON, R. (1998). Circadian rhythms in diving behavior and ventilatory response to asphyxia in canvasback ducks. American Journal of Physiology 274, R686-693 | [Medline] |
| WURTS, S. W. & EDGAR, D. M. (2000). Circadian and homeostatic control of rapid eye movement (REM) sleep: promotion of REM tendency by the suprachiasmatic nucleus. Journal of Neuroscience 20, 4300-4310 | [Abstract/Full Text] |
| ZAR, J. H. (1984). Biostatistical Analysis, p. 718. Prentice-Hall Inc., Englewood Cliffs, NJ, USA |
Acknowledgements
We are grateful to B. Chan for technical assistance. This work was supported by operating and equipment grants from the Natural Sciences and Engineering Research Council (NSERC) of Canada, a Block Term Grant from the Ontario Thoracic Society to R.S., and operating funds from the Medical Research Council (MRC) of Canada and Ontario Thoracic Society to R.L.H. R.L.H. also gratefully acknowledges equipment grants from the Canada Foundation for Innovation and Ontario Research and Development Challenge Fund. R.L.H. is a recipient of an MRC of Canada Scholarship. K.S.L. was supported by an Ontario Graduate Scholarship in Science and Technology.
Corresponding author
R. Stephenson: Departments of Physiology and Zoology, University of Toronto, Ramsay Wright Building, 25 Harbord Street, Toronto, Ontario, Canada M5S 3G5.
Email: rstephsn{at}zoo.utoronto.ca
This article has been cited by other articles:
|
|
 |
|
  G. Montandon, A. Bairam, and R. Kinkead Neonatal caffeine induces sex-specific developmental plasticity of the hypoxic respiratory chemoreflex in adult rats Am J Physiol Regulatory Integrative Comp Physiol, September 1, 2008; 295(3): R922 - R934. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Pendlebury, R. J. A. Wilson, S. Bano, K. J. Lumb, J. M. Schneider, and S. U. Hasan Respiratory Control in Neonatal Rats Exposed to Prenatal Cigarette Smoke Am. J. Respir. Crit. Care Med., June 1, 2008; 177(11): 1255 - 1261. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Scott Sniffing and Spatiotemporal Coding in Olfaction Chem Senses, February 1, 2006; 31(2): 119 - 130. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
