Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 May 1;38(5):685-97.
doi: 10.5665/sleep.4656.

Behavioral sleep-wake homeostasis and EEG delta power are decoupled by chronic sleep restriction in the rat

Richard Stephenson  1 , Aimee M Caron  1 , Svetlana Famina  1
Affiliations

Behavioral sleep-wake homeostasis and EEG delta power are decoupled by chronic sleep restriction in the rat

Richard Stephenson et al. Sleep. .

Abstract

Study objectives: Chronic sleep restriction (CSR) is prevalent in society and is linked to adverse consequences that might be ameliorated by acclimation of homeostatic drive. This study was designed to test the hypothesis that the sleep-wake homeostat will acclimatize to CSR.

Design: A four-parameter model of proportional control was used to quantify sleep homeostasis with and without recourse to a sleep intensity function.

Setting: Animal laboratory, rodent walking-wheel apparatus.

Subjects: Male Sprague-Dawley rats.

Interventions: Acute total sleep deprivation (TSD, 1 day × 18 or 24 h, N = 12), CSR (10 days × 18 h TSD, N = 5, or 5 days × 20 h TSD, N = 6).

Measurements and results: Behavioral rebounds were consistent with model predictions for proportional control of cumulative times in wake, nonrapid eye movement (NREM) and rapid eye movement (REM). Delta (D) energy homeostasis was secondary to behavioral homeostasis; a biphasic NREM D power rebound contributed to the dynamics (rapid response) but not to the magnitude of the rebound in D energy. REM behavioral homeostasis was little affected by CSR. NREM behavioral homeostasis was attenuated in proportion to cumulative NREM deficit, whereas the biphasic NREM D power rebound was only slightly suppressed, indicating decoupled regulatory mechanisms following CSR.

Conclusions: We conclude that sleep homeostasis is achieved through behavioral regulation, that the NREM behavioral homeostat is susceptible to attenuation during CSR and that the concept of sleep intensity is not essential in a model of sleep-wake regulation.

Study objectives: Chronic sleep restriction (CSR) is prevalent in society and is linked to adverse consequences that might be ameliorated by acclimation of homeostatic drive. This study was designed to test the hypothesis that the sleep-wake homeostat will acclimatize to CSR.

Design: A four-parameter model of proportional control was used to quantify sleep homeostasis with and without recourse to a sleep intensity function.

Setting: Animal laboratory, rodent walking-wheel apparatus.

Subjects: Male Sprague-Dawley rats.

Interventions: Acute total sleep deprivation (TSD, 1 day × 18 or 24 h, N = 12), CSR (10 days × 18 h TSD, N = 5, or 5 days × 20 h TSD, N = 6).

Measurements and results: Behavioral rebounds were consistent with model predictions for proportional control of cumulative times in wake, nonrapid eye movement (NREM) and rapid eye movement (REM). Delta (D) energy homeostasis was secondary to behavioral homeostasis; a biphasic NREM D power rebound contributed to the dynamics (rapid response) but not to the magnitude of the rebound in D energy. REM behavioral homeostasis was little affected by CSR. NREM behavioral homeostasis was attenuated in proportion to cumulative NREM deficit, whereas the biphasic NREM D power rebound was only slightly suppressed, indicating decoupled regulatory mechanisms following CSR.

Conclusions: We conclude that sleep homeostasis is achieved through behavioral regulation, that the NREM behavioral homeostat is susceptible to attenuation during CSR and that the concept of sleep intensity is not essential in a model of sleep-wake regulation.

Keywords: EEG delta power; allostasis; chronic sleep restriction; proportional control model; sleep deficit; sleep deprivation; sleep homeostasis; sleep intensity; sleep regulation; two-process model.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Quantitative analysis of cumulative time in state reveals a monotonic rebound in total sleep following acute 24 h total sleep deprivation (TSD). (A) Cumulative time in total sleep (tNREM and tREM combined) recorded over 4 w in one rat shows baseline stability before and after 24 h acute TSD (gray bar). Inset illustrates progressive return toward the extrapolated baseline linear regression (dotted black line) immediately following the end of TSD. (B) Net (detrended) cumulative sleep time exhibits a monotonic rebound response with superimposed diurnal rhythms. Proportional control model parameters (Rspan, τ, T0, Y0) are quantified by nonlinear exponential regression (black curve; see text for details).
Figure 2
Figure 2
Behavioral and EEG energy responses to acute 24 h total sleep deprivation (TSD). (A) Mean (± standard error of the mean [SEM]) net cumulative time in state before (2 days baseline, BL1–2), during (blue bar, TSD) and after (3 days recovery, R1–3) 24 h TSD (N = 9 rats). All vigilance states conformed to the predictions of a first-order proportional control model of behavioral homeostasis. Wakefulness (tWAKE, blue), nonrapid eye movement (NREM) sleep (tNREM, green), rapid eye movement (REM) sleep (tREM, red), black curves indicate exponential regression through recovery data. (B) Mean (± SEM) state-specific and total net Δ energy before, during and after 24 h TSD (N = 8 rats). Δ energy responded to TSD with monotonic rebounds (black curves) in each of wake (blue), NREM (green) and REM (red). Note the relatively rapid response of total Δ energy (gray). (C) Mean (± SEM) net total EEG energy before, during and after 24 h TSD in each of the θ (red), α (blue), β (violet), and γ (green) frequency bands. Those frequency bands that accumulated net deficit during TSD (α and β) responded with a monotonic rebound toward baseline during recovery, consistent with the prediction of a proportional control model of homeostasis. The frequency bands that accumulated net excess during TSD (θ and γ) did not respond with a return to baseline. In all frequency bands, the trajectory of the post-TSD response was determined primarily by the behavioral homeostat. Blue bars indicate TSD.
Figure 3
Figure 3
Behavioral responses to acute total sleep deprivation (TSD) expressed as parameter values of a model of proportional control. Behavioral responses, 18 h TSD, N = 3; 24 h TSD, N = 9; electroencephalographic Δ responses, 18 h TSD, N = 2; 24 h TSD, N = 8. Rspan (black symbols) is the full magnitude of the rebound, and T0 (white symbols) is the deviation from full compensation (steady-state error). These are plotted against Y0, the state-specific excess or deficit accumulated during TSD for wakefulness and total sleep (A), nonrapid eye movement (NREM) sleep (B), and rapid eye movement (REM) sleep (C). Correlations, quantified by linear regression, show that Rspan is directly proportional to Y0 for REM sleep, whereas T0 is directly proportional to Y0 for NREM sleep and wakefulness. X-axis intercept of the T0 regression (dashed vertical gray line) is interpreted as an estimate of the maximum compensatory capacity of the behavioral tNREM and tWAKE homeostats. (D) Rate of behavioral rebound response (τ) is significantly faster in REM sleep (tREM, dark gray), than in NREM sleep (tNREM, white) or wakefulness (tWAKE, black). Total Δ energy response (striped gray) was as rapid as the tREM behavioral response, and significantly faster than both the tNREM behavioral and NREM Δ energy (light gray) rebounds. NREM Δ energy rebound was faster than the tNREM behavioral rebound, highlighting the influence of the biphasic rebound in NREM Δ power on Δ energy dynamics (see text for discussion). *P < 0.05; **P < 0.01; ***P < 0.001; °P < 0.1.
Figure 4
Figure 4
Behavioral and electroencephalographic (EEG) Δ energy rebound responses are attenuated following chronic sleep restriction (CSR). Mean (± standard error of the mean [SEM]) net cumulative time in state for each of wakefulness (tWAKE, blue), nonrapid eye movement (NREM) sleep (tNREM, green) and rapid eye movement (REM) sleep (tREM, red), with recovery responses fitted by exponential nonlinear regression (black curves). During CSR, enforced wakefulness intervals and daily sleep opportunities are indicated by blue and red bars (abscissa), respectively. (A) Net cumulative time in state in CSR18 group (N = 6), 2 baseline days of unrestricted sleep, 10 consecutive days of total sleep deprivation (TSD) at ZT6–24, and 3 recovery days of unrestricted sleep. (B) Net cumulative time in state in CSR20 group (N = 5), 2 baseline days, 5 days of TSD at ZT4–24, and 3 recovery days. (C) Mean (± SEM) net Δ energy (CSR18 group) in each of wakefulness (DWAKE, blue), NREM sleep (DNREM, green) and REM sleep (DREM, red), and total (DTOT, all states combined, gray), with recovery responses fitted by exponential nonlinear regression (black curves). (D) Magnitude (Rspan) of the total Δ energy rebound versus deficit accumulated. Response to acute TSD (black circles, N = 8) was proportional to deficit (line of identity, red dashed line). Proportionality was absent following CSR18 (white circles, N = 5). (E) Rates of rebound (τ) of total Δ energy were equivalent following TSD and CSR18. Red error bars, median and inter-quartile range. Unsteady EEG power signals (slow nonlinear drift) precluded analyses of Δ energy rebound in the CSR20 group.
Figure 5
Figure 5
Behavioral rebound (Rspan) is differentially suppressed by cumulative sleep deficit or wake excess (Y0). Strength of correlation is estimated by linear regression (dashed lines) through data pooled from three groups of rats (total sleep deprivation [TSD], black symbols; CSR18, white symbols; CSR20, gray symbols). Suppression of the homeostatic rebound was pronounced for wakefulness (A) and nonrapid eye movement (NREM) sleep (B) but negligible for rapid eye movement (REM) sleep (C). Note that the CSR20 stimulus (5 consecutive days of partial TSD for 20 h per day) was only marginally sufficient to elicit acclimation (“allostasis”) of the NREM homeostat (B). CSR = chronic sleep restriction.
Figure 6
Figure 6
Rapid eye movement (REM) sleep time increases at the expense of wake and nonrapid eye movement (NREM) times in daily sleep opportunities during chronic sleep restriction (CSR). (A) CSR18 (N = 6); 6 h sleep opportunities were available from ZT0–6 in each of 10 CSR days. (B) CSR20 (N = 5); 4 h sleep opportunities were available from ZT0–4 in each of 5 CSR days. In both (A) and (B), bars denote mean (± standard error of the mean) cumulative time in each of wakefulness (tWAKE, black), NREM sleep (tNREM, white) and REM sleep (tREM, gray), expressed as change from corresponding baseline values. Statistical significance was assessed using repeated-measures analysis of variance (*P < 0.05; **P < 0.01; ***P < 0.001) or one-sample t-test (†P < 0.05). Note that in (A), day 13 is the ZT0–6 interval immediately following the final CSR episode and therefore also constitutes the beginning of post-CSR18 recovery. Likewise in (B), day 8 represents both the final sleep opportunity and the beginning of post-CSR20 recovery.
Figure 7
Figure 7
Responses of nonrapid eye movement (NREM) Δ power to total sleep deprivation (TSD) and chronic sleep restriction (CSR). Mean (± standard error of the mean) electroencephalographic (EEG) Δ power before, during and after (A) 24 h TSD (N = 9) and (B) CSR18 (N = 6) in wake (blue), NREM (green), and REM (red). Abscissa blue bars, enforced wakefulness; red bars, sleep opportunities. Peak NREM Δ power in ZT 0–2 of each CSR sleep opportunity and recovery day were compared with the mean of 2 baseline days (repeated-measures analysis of variance with Dunnett post hoc multiple comparisons test; *P < 0.05, °P < 0.1). Rebound responses to TSD (C) and CSR18 (D), calculated as mean NREM Δ power in the interval ZT0–6, expressed as difference from baseline. Columns show mean ± 95% confidence interval (*P < 0.05, °P < 0.1, one-sample t-test for null hypothesis mean = 0). (E) EEG power in each of the Δ (black), θ (red), α (blue), β (violet), and γ (green) frequency bands during 3 recovery days following acute TSD. Data expressed as percent difference from baseline at corresponding times of day. Mean values (N = 12) are shown; error bars are omitted for clarity of presentation. Sequential positive and negative rebounds are evident, with a qualitatively similar pattern of response in all frequency bands, although statistical significance was found in Δ, θ and α bands only (see text). (F) EEG power in each of the Δ (black), θ (red), α (blue), β (violet), and γ (green) frequency bands during 3 recovery days following CSR18. Although slightly attenuated relative to TSD, the biphasic pattern of response remained statistically significant following CSR18 in Δ, θ, and α bands.
See this image and copyright information in PMC

Comment in

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Bonnet MH, Arand DL. We are chronically sleep deprived. Sleep. 1995;18:908–11. - PubMed
    1. Daley M, Morin CM, LeBlanc M, Gregoire JP, Savard J. The economic burden of insomnia: direct and indirect costs for individuals with insomnia syndrome, insomnia symptoms, and good sleepers. Sleep. 2009;32:55–64. - PMC - PubMed
    1. Daley M, Morin CM, LeBlanc M, Gregoire JP, Savard J, Baillargeon L. Insomnia and its relationship to health-care utilization, work absenteeism, productivity and accidents. Sleep Med. 2009;10:427–38. - PubMed
    1. Hillman DR, Murphy AS, Pezzullo L. The economic cost of sleep disorders. Sleep. 2006;29:299–305. - PubMed
    1. Luyster FS, Strollo PJ, Jr., Zee PC, Walsh JK Boards of Directors of the American Academy of Sleep Medicine and the Sleep Research Society. Sleep: a health imperative. Sleep. 2012;35:727–34. - PMC - PubMed

Publication types

LinkOut - more resources