Brown Norway and Zucker Lean Rats Demonstrate Circadian Variation in Ventilation and Sleep Apnea

Abstract

Study Objectives:

Circadian rhythms influence many biological systems, but there is limited information about circadian and diurnal variation in sleep related breathing disorder. We examined circadian and diurnal patterns in sleep apnea and ventilatory patterns in two rat strains, one with high sleep apnea propensity (Brown Norway [BN]) and the other with low sleep apnea propensity (Zucker Lean [ZL]).

Design/Setting:

Chronically instrumented rats were randomized to breathe room air (control) or 100% oxygen (hyperoxia), and we performed 20-h polysomnography beginning at Zeitgeber time 4 (ZT 4; ZT 0 = lights on, ZT12 = lights off). We examined the effect of strain and inspired gas (twoway analysis of variance) and analyzed circadian and diurnal variability.

Measurements and Results:

Strain and inspired gas-dependent differences in apnea index (AI; apneas/h) were particularly prominent during the light phase. AI in BN rats (control, 16.9 ± 0.9; hyperoxia, 34.0 ± 5.8) was greater than in ZL rats (control, 8.5 ± 1.0; hyperoxia, 15.4 ± 1.1, [strain effect, P < 0.001; gas effect, P = 0.001]). Hyperoxia reduced respiratory frequency in both strains, and all respiratory pattern variables demonstrated circadian variability. BN rats exposed to hyperoxia demonstrated the largest circadian fluctuation in AI (amplitude = 17.9 ± 3.7 apneas/h [strain effect, P = 0.01; gas effect, P < 0.001; interaction, P = 0.02]; acrophase = 13.9 ± 0.7 h; r² = 0.8 ± 1.4).

Conclusions:

Inherited, environmental, and circadian factors all are important elements of underlying sleep related breathing disorder. Our method to examine sleep related breathing disorder phenotypes in rats may have implications for understanding vulnerability for sleep related breathing disorder in humans.

Citation:

Fink AM; Topchiy I; Ragozzino M; Amodeo DA; Waxman JA; Radulovacki MG; Carley DW. Brown Norway and Zucker Lean rats demonstrate circadian variation in ventilation and sleep apnea. SLEEP 2014;37(4):715-721.

INTRODUCTION

Sleep related breathing disorder (SRBD) reflects a complex array of phenotypes influenced by factors including anatomical characteristics, genetic susceptibilities, sleep stage-specific activities, and circadian variability.^1–10 Animal models will become increasingly important for elucidating mechanisms by which these factors mediate or modulate SRBD. We showed that laboratory rats express spontaneous central sleep apneas, and apnea frequency varies among rat strains.^11–15 We also demonstrated that hyperoxia increased the frequency of spontaneous apneas in outbred Sprague-Dawley rats.¹⁴ It is unknown how SRBD may vary across the circadian day. In the current study, we examined inbred rat strains that exhibit low (Zucker Lean [ZL]) or high (Brown Norway [BN]) sleep apnea frequency and hypothesized that sleep related breathing behaviors, and responses to hyper-oxia, would exhibit diurnal and circadian patterns in both strains.

MATERIALS AND METHODS

All procedures conformed to the American Physiological Society's Guiding Principles for the Care and Use of Vertebrate Animals and were approved by the University of Illinois at Chicago institutional animal care and utilization committee. Adult male BN and ZL rats (n = 40, weighing 345.9 g ± 7.0, Harlan, Indianapolis, IN) were housed individually under optimal conditions (22°C, 60% humidity, and 12:12 h light:dark cycle with lights on at 08:00). With the rats under anesthesia (ketamine:xylazine 80:5 mg/kg), electroencephalography (EEG) electrodes (stainless steel screws placed bilaterally into the frontal and parietal skull bones) and bilateral electro-myography (EMG) electrodes were implanted and placed in the nuchal muscles. The EEG/EMG leads were soldered to a connector affixed to the skull with acrylic dental cement. The skin was sutured around the connector, and rats were permitted 7 days to recover. After recovery, rats underwent 3 days of adaptation to the recording apparatus, a cylindrical bias-flow-ventilated single-chamber plethysmograph (Buxco Electronics, Sharon CT, [6 inches in diameter with a 10-inch vertical chimney]) in which animals had unlimited access to food and water. EEG and EMG signals were obtained from the animal's headset using a flexible recording apparatus tethered to the top of the plethysmograph chimney, allowing for free movement of the EEG/EMG recording cable. Respiration was transduced by measuring pressure changes within the plethysmograph with respect to an integrated reference chamber.

For animals randomized to the control condition (BN rats, n = 10; ZL rats, n = 10), chambers were ventilated with room air; for animals randomized to the hyperoxia condition (BN rats, n = 11; ZL rats, n = 9), chambers were ventilated with 100% oxygen. Plethysmographs were continuously ventilated at 2.5 L/min, which was sufficient to prevent rebreathing.^12,16 We performed 20-h polysomnographic recordings starting at Zeitgeber time 4 [ZT4; ZT0 represented lights on and ZT12 lights off). All polysomnography signals were amplified, filtered (low-pass filter set at 100 Hz), and digitized at 250/sec (Bio-Logic Systems Sleepscan Premier, Natus, Mundelein, IL). Polysomnographic recordings were visually scored in 30-sec epochs as wakefulness, nonrapid eye movement (NREM) sleep, and rapid eye movement (REM) sleep. Sleep state distributions (min) were determined for every 2-h increment.

Using software developed in our laboratory, we determined the hourly values for respiratory frequency (f), tidal volume (V_T), and minute ventilation (E) for each sleep-wake stage (i.e., wake, NREM sleep, REM sleep). Because the plethysmograph was not sealed, pressure fluctuations were not calibrated on a volume scale. We normalized V_T and E values relative to each rat's mean value obtained for wakefulness in the first hour of recording. For comparison, we also calculated normalized respiratory frequency (N-f). These normalizations permitted assessment of the effects of sleep-wake state, inspired gas, nocturnal-(dark)/diurnal-(light) period, and circa-dian time on each animal's respiratory pattern. Sleep apneas, defined as cessation of respiration for ≥ 2.0 sec, were detected and associated with either NREM or REM sleep. The ≥ 2.0 sec criterion was selected because this duration represents at least two missed breaths, which corresponds to a 10-sec apnea in humans.¹³ We did not differentiate between spontaneous and postsigh apneas. Sleep apnea indices (AI, apneas/h of sleep) were calculated for total, NREM, and REM sleep.

Statistical Analysis

Data were analyzed using the Statistical Package for the Social Sciences (SPSS; Version 20.0, Chicago, IL) and STATA (Version 12.0, College Station, TX). To assess diurnal patterns, we aggregated the sleep-wake and respiratory data to reflect mean values during light and dark phases of recording. We conducted one-way and two-way analysis of variance (ANOVA) with least significant difference post hoc tests to examine the effect of strain and inspired gas (strain × gas) on sleep-wake states, AI, respiratory patterns, and circadian variability. We also conducted paired t tests to determine whether any sleep and breathing variables differed significantly between light and dark phases. Circadian variation was analyzed by least squares regression to fit a cosine wave with a period of 24 h. For each variable, we calculated an amplitude, acrophase, and r² value for each animal. All data are reported as mean ± standard error of the mean; statistical significance was set at P ≤ 0.05.

RESULTS

Table 1 summarizes the effects of rodent strain and inspired gas on sleep architecture, respiratory pattern, and apnea expression in light and dark phases of the recording. Hyperoxia influenced sleep architecture in the light phase, but the effects depended on rat strain. For example, hyperoxia provoked a significant reduction in NREM sleep in BN rats but augmented NREM sleep in ZL rats (Table 1).

Table 1.

Comparisons of sleep, ventilation, and apnea expression in the light and dark phases by strain and inspired gas

There were no significant differences in the light or dark phases between BN and ZL rats for N-f, N-V_T, and N- E (Table 1), nor within each strain for ventilatory patterns, as assessed by paired t tests (data not shown). Analysis of f (breaths per min) recorded during wakefulness in the first hour of recording demonstrated that BN rats (control) took fewer breaths on average (196.1 ± 11.4) than ZL rats (control; 229.5 ± 11.1, respectively; P = 0.05). Neither BN nor ZL rats demonstrated significant changes in N-f values between light and dark phases across sleep-wake states (Table 1). In both strains, hyperoxia was associated with a reduction in N-f during NREM and REM sleep, but this association only reached statistical significance in the light phase (Table 1).

BN rats demonstrated higher AI than ZL rats throughout sleep, but this difference achieved statistical significance during the light phase (P < 0.001) and not during the dark phase (P = 0.06; Table 1). In the control condition (Figure 1A), BN rats demonstrated a significantly greater AI than ZL rats only during NREM sleep. Hyperoxia exacerbated sleep apnea in both BN rats (Figure 1B) and ZL rats (Figure 1C), and the effects of hyperoxia were more prominent during the light phase (Figure 1B and 1C). When exposed to hyperoxia, BN rats demonstrated elevated AIs during both the light (NREM and REM sleep) and dark phases (REM sleep; Figure 1B). We observed similar patterns for ZL rats, but the exacerbation of AI was statistically significant only during the light phase (Figure 1C). During REM sleep in the light phase, the increase in AI induced by hyperoxia was significantly greater in BN rats than ZL rats (P = 0.04 for strain by gas interaction; Table 1). Compared with ZL rats, BN rats demonstrated an AI approximately twofold greater during NREM sleep and threefold greater during REM sleep when exposed to hyperoxia during the light phase (Table 1; Figure 1D). Also, hyperoxia increased apnea duration in both strains in the light phase (Table 1).

Figure 1.

Comparisons of apnea indices stratified by strain/inspired gas, sleep stage, and light/dark phases. Data are mean ± standard error of the mean. The asterisk denotes P < 0.05. NREM, nonrapid eye movement; REM, rapid eye movement.

There were few statistically significant differences between light and dark phases among stratified AI variables (as assessed by paired t tests; Figure 1); however, NREM sleep AI was significantly elevated during the dark phase compared with the light phase in both control BN rats (P = 0.01) and control ZL rats (P = 0.03; Figure 1A-1C). Animals exposed to hyperoxia did not exhibit any statistically significant differences between AI values for light and dark phases.

Circadian rhythms were evident in all variables. Amplitudes, acrophases, and r² statistics are shown in Table 2. In general, the fraction of explained variance (r²) was greatest for animals exposed to hyperoxia. Sample data illustrating the best fit cosine waves are shown in Figure 2 for individual animals. Hyperoxia was associated with increased circadian amplitudes in all measures of apnea (apnea duration, AI in total, NREM, and REM sleep) but not for any breath-by-breath measures (f, N-f, N-V_T, N- E). When exposed to hyperoxia, BN rats demonstrated the largest circadian amplitudes for the sleep apnea variables. None of the acrophase values for any sleep apnea variables differed significantly according to strain or inspired gas; all animals demonstrated their peak amplitudes for these variables approximately 11-13 h into the recording (i.e., beginning of the dark phase; Table 2).

Table 2.

Circadian variability in sleep-wake behaviors, sleep apnea, and respiratory pattern with comparisons by strain and inspired gas

Figure 2.

Example recordings from each group. Red lines indicate total sleep apnea index; blue (curved) lines represent the best-fit cosine waves. BN, Brown Norway; ZL, Zucker Lean.

DISCUSSION

Our findings demonstrate that two inbred rat strains exhibit different vulnerabilities for SRBD despite demonstrating similar diurnal and circadian rhythms in respiratory pattern and apnea expression. In rats (nocturnal animals),¹⁷ we found that the AI acrophase occurred in the initial hours of the dark phase, with significant amplification of circadian fluctuations in AI induced by hyperoxia in BN rats. These observations suggest that circadian factors may influence the characteristics and severity of SRBD in vulnerable individuals. This conclusion is consistent with clinical reports that asthma exacerbation,¹⁸ ventricular arrhythmias,¹⁹ and apnea duration²⁰ all increase during the night. Furthermore, we speculate that altered sleep phase (e.g., delayed sleep phase, advanced sleep phase, or shift work-related alterations) may confer increased SRBD risk in humans, but this remains to be systematically investigated.

Several of our findings are consistent with reports indicating circadian variability in ventilation in rodents.^21–23 Similar to our findings with normalized parameters (e.g., N-f, N-V_T, N- E), Stephenson et al. demonstrated that Sprague-Dawley rats exhibited circadian variability in f, V_T, and E across sleep-wake states.²¹ In a secondary analysis, they examined AI. Inconsistent with our findings in BN and ZL rats (controls), they did not demonstrate any significant differences between the light and dark phases for AI. This discrepancy may be due to their smaller sample size (n = 6), strain differences, differing criteria for scoring apneas, or other factors. Interestingly, their Sprague-Dawley rats expressed a similar apnea frequency for REM sleep (i.e., 30-40 apneas/h) compared with the BN and ZL rats in our study, but the AI for NREM sleep was lower in their study.²² Mice studied by Davis et al.²³ demonstrated a similar apnea frequency as we observed in rats for both NREM and REM sleep. In contrast to our findings, Davis et al. did not observe significant differences in apnea expression by mouse strain, which differs from our model in several important ways (e.g., different rodent species, inspired gas protocol, statistical approach). Davis et al. did not use regression analysis to compute amplitude and acrophase, but they did note significant differences in respiratory rate by mouse strain and by light/dark phase, which indicates a possible circadian pattern.²³

Other investigators found rodent strain to have a significant effect on ventilatory activities, but they did not consider the contributions of sleep-wake states or circadian variations.^24,25 Strohl et al. demonstrated that under control (room air) conditions, BN rats expressed significantly greater f and lower V_T compared with Sprague-Dawley rats but detected no signifi-cant differences in these parameters between BN and ZL rats.²⁵ Donovan et al. observed lower ventilatory responses to hypoxic and hypercapnic challenges in BN rats compared with Sprague-Dawley rats and concluded that the latter group may have a genetic propensity for greater carotid body chemosensitivity.²⁴ In addition, Strohl and Thomas examined Zucker obese and Koletsky rats, and after controlling for body weight, found that Zucker obese rats demonstrated significantly higher f and E in response to hyperoxia compared with Koletsky rats.²⁶ These findings extended the literature about strain-dependent ventila-tory differences, but none of these investigators evaluated circa-dian variability. By examining the effect of sleep-wake states and diurnal/circadian variability, our results showed that circadian factors may pose additional risks for expressing sleep apnea.

Conditions of low or fluctuating respiratory drive during sleep may represent critical periods for sleep apnea vulnerability. Our data indicate that apnea propensity is influenced by each of these factors in rats. Elevated autonomic and respiratory instability is a well-defined feature of REM sleep,^9,27,28 and in accordance with others,^22,23 we found an approximately twofold to threefold greater frequency of apnea during REM sleep compared with NREM sleep. Respiratory drive (especially f) was reduced by hyperoxia (Table 1), and this condition was associated with significantly increased apnea expression. This effect was most evident during REM sleep, with REM sleep AI in BN rats exposed to hyperoxia exceeding 90 events/h during the light phase. The view that conditions of low or fluctuating respiratory drive predispose to apnea expression also is supported by the observation that when respiration is nonspecifically stimulated by inspired carbon dioxide, apnea expression is decreased, both in rats¹⁴ and humans.²⁹

Both respiratory pattern and AIs exhibited significant circa-dian variations, and peak values for E and AI occurred early in the dark (active) phase (Table 2). This suggests that the observed circadian peak in AI 11-14 h into the circadian day cannot be attributed to a condition of low respiratory drive during either NREM or REM sleep. Rather, we speculate that the increased apnea expression may reflect the facts that during this period rats are becoming more active, sleep bouts are becoming shorter, and wake/sleep and NREM/REM transitions are becoming more frequent. Periods of sleep/wake and NREM/REM transitions have been viewed as vulnerable times for apnea genesis in humans.^30–33

Although changes in respiratory control during wakefulness have been observed in humans and found to correlate with SRBD phenotypes,^34,35 we chose not to analyze apneas during wakefulness in rats. Rats demonstrate respiratory sighs and pauses during voluntary behaviors such as eating, drinking, and grooming. Although the respiratory pattern disturbances observed during quiet wakefulness may reflect alterations in respiratory control, we had no reliable means to differentiate these events from voluntary behaviors such as those noted previously, which reflect a different set of underlying mechanisms.

Consistent with our previous work,¹⁴ significant differences in sleep architecture between strains or inspired gases were not observed here. Therefore, the observed strain and inspired gas effects on respiratory pattern and apnea expression were not importantly confounded by differences in the sleep process, per se. Although the respiratory frequencies we measured were higher than those reported for BN and ZL rats in another study,²⁵ we are confident that the animals were well acclimated to the recording apparatus because the N-f values remained consistent across sleep-wake states and across the 20-h recordings. We considered the possibility that the effects of hyperoxia may have diminished throughout the recording (tachyphylaxis), but this hypothesis was not supported by our circadian analysis indicating that the peak apnea frequency occurred halfway into the recordings. Also, animals in the control and hyperoxia conditions demonstrated similar acrophases for AI, so hyper-oxia appears to have amplified an underlying circadian process related to apnea genesis.

In summary, we found strain-dependent differences in sleep apnea expression that were not accompanied by signifi-cant differences in sleep architecture. Each strain exhibited a significant circadian variation in apnea frequency, with a peak AI early in the dark phase. Hyperoxia increased overall apnea frequency in each strain and amplified the underlying circadian variation in AI. Our data underscore the collective importance of genetic, environmental, and circadian factors in defining SRBD phenotypes and indicate that experimental approaches to reduce respiratory drive via inactivation of the chemoreceptor pathway permit the systematic investigation of mechanisms that contribute to SRBD phenotypes. Circadian rhythms are prominent in many biological systems, including ventilation,³⁶ and our results suggest that rats represent a useful animal model for investigating circadian influences on apnea propensity.

DISCLOSURE STATEMENT

This was not an industry supported study. This research was funded by National Institute on Aging (R01AG016303) and University of Illinois at Chicago Chancellor's Discovery Fund. The authors have indicated no financial conflicts of interest. Notation of prior abstract publication/presentation: Circadian acrophase and amplitude of sleep-disordered breathing in rats, Associated Professional Sleep Societies, Baltimore, MD, 2013.

ABBREVIATIONS

AI

apnea index

ANOVA

analysis of variance

BN

Brown Norway

EEG

electroencephalography

EMG

electromyography

LSD

least significant difference

N-f

normalized respiratory frequency

N-V_E

normalized minute ventilation

N-V_T

normalized tidal volume

NREM

non-rapid eye movement

REM

rapid eye movement

SRBD

sleep related breathing disorder

SEM

standard error of the mean

ZL

Zucker Lean

ZT

Zeitgeber time