Jidong Fang

The Role of Cytokines in Physiological Sleep Regulation

Abstract: Several growth factors (GFs) are implicated in sleep regulation. It is posited that these GFs are produced in response to neural activity and affect input‐output relationships within the neural circuits where they are produced, thereby inducing a local state shift. These GFs also influence synaptic efficacy. All the GFs currently identified as sleep regulatory substances are also implicated in synaptic plasticity. Among these substances, the most extensively studied for their role in sleep regulation are interleukin‐1β (IL‐1) and tumor necrosis factor a (TNF). Injection of IL‐1 or TNF enhances non‐rapid eye movement sleep (NREMS). Inhibition of either IL‐1 or TNF inhibits spontaneous sleep and the sleep rebound that occurs after sleep deprivation. Stimulation of the endogenous production of IL‐1 and TNF enhances NREMS. Brain levels of IL‐1 and TNF correlate with sleep propensity; for example, after sleep deprivation, their levels increase. IL‐1 and TNF are part of a complex biochemical cascade regulating sleep. Downstream events include nitric oxide, growth hormone releasing hormone, nerve growth factor, nuclear factor kappa B, and possibly adenosine and prostaglandins. Endogenous substances moderating the effects of IL‐1 and TNF include anti‐inflammatory cytokines such as IL‐4, IL‐10, and IL‐13. Clinical conditions altering IL‐1 or TNF activity are associated with changes in sleep, for example, infectious disease and sleep apnea. As our knowledge of the biochemical regulation of sleep progresses, our understanding of sleep function and of many clinical conditions will improve.

Inhibition of nitric oxide synthesis inhibits rat sleep.

Previous findings indicate that nitric oxide (NO) may play a role in the regulation of sleep-wake activity. In rabbits, blocking the production of endogenous NO by a nitric oxide synthase inhibitor, N omega-nitro-L-arginine (L-NAME) suppresses spontaneous sleep and interferes the somnogenic actions of interleukin 1. In the present experiments we extended our earlier work by studying the long-term effects of L-NAME treatment on sleep-wake activity including power spectra analyses of the electroencephalogram (EEG) in rats. Rats implanted with EEG electrodes, brain thermistor, and intracerebroventricular (i.c.v.) guide cannula were injected i.c.v. with vehicle or 0.2, 1, or 5 mg L-NAME at light onset. In separate experiments, rats were injected intraperitoneally (i.p.) with L-NAME three times (50, 50, 100 mg/kg), 12-12 h apart. Both i.c.v. and i.p. injections of L-NAME elicited decreases in time spent in NREMS and REMS. After i.c.v. injection of 5 mg L-NAME the sleep responses were long-lasting; NREMS did not return to baseline even 72 h after injection. EEG delta-wave activity during NREMS (slow wave activity) was also suppressed after 0.2 and 5 mg L-NAME. Brain temperature was slightly increased after the two lower doses of L-NAME, whereas there was a transient decrease in Tbr after 5 mg L-NAME. Acute i.p. injection of 50 mg/kg L-NAME elicited an immediate decrease in NREMS which lasted for approximately 2 h. The second injection of 50 mg/kg L-NAME and the following injection of 100 mg/kg L-NAME induced biphasic decreases in NREMS but not REMS.

Brain organization and sleep function

A view of brain organization and sleep function is presented. Sleep is hypothesized to begin at the neuronal group level. Sleep results in the use and thus maintenance, of synapses that are insufficiently stimulated during wakefulness thereby serving to preserve a constancy of a synaptic superstructure. It is further hypothesized that sleep at the neuronal group level is regulated by the production of substances whose rate of production or catabolism is synaptic use-dependent. If sufficient number of neuronal groups are in a sleep state (also called disjunctive state) then the perception of sleepiness occurs. Coordination of neuronal group sleep results from humoral and neuronal projection systems previously linked to sleep regulation. The theory presented is unique in that it: (a) hypothesizes an organizational level at which sleep occurs; (b) hypothesizes that sleep is neuronal--use-dependent, not wakefulness-dependent; (c) hypothesizes that sleep first occurs in evolution when complex ganglia evolved; and (d) hypothesizes the both non-rapid eye movement sleep (NREMS) and REMS serve the same function of synaptic reorganization. The theory is consistent with past theories of sleep function, yet provides a fundamentally new paradigm for sleep research.

Cytokines in sleep regulation

The central thesis of this essay is that the cytokine network in brain is a key element in the humoral regulation of sleep responses to infection and in the physiological regulation of sleep. We hypothesize that many cytokines, their cellular receptors, soluble receptors, and endogenous antagonists are involved in physiological sleep regulation. The expressions of some cytokines are greatly amplified by microbial challenge. This excess cytokine production during infection induces sleep responses. The excessive sleep and wakefulness that occur at different times during the course of the infectious process results from dynamic changes in various cytokines that occur during the hosts response to infectious challenge. Removal of any one somnogenic cytokine inhibits normal sleep, alters the cytokine network by changing the cytokine mix, but does not completely disrupt sleep due to the redundant nature of the cytokine network. The cytokine network operates in a paracrine/autocrine fashion and is responsive to neuronal use. Finally, cytokines elicit their somnogenic actions via endocrine and neurotransmitter systems as well as having direct effects neurons and glia. Evidence in support of these postulates is reviewed in this essay.

Somnogenic relationships between tumor necrosis factor and interleukin-1

Both tumor necrosis factor (TNF) and interleukin (IL)-1 are somnogenic cytokines. They also induce each others production and both induce nuclear factor kappa B activation, which in turn enhances IL-1 and TNF transcription. We hypothesized that TNF and IL-1 could influence each others somnogenic actions. To test this hypothesis, we determined the effects of blocking both endogenous TNF and IL-1 on spontaneous sleep and on sleep rebound after sleep deprivation in rabbits. Furthermore, the effects of inhibition of TNF on IL-1-induced sleep and the effects of blocking IL-1 on TNF-induced sleep were determined. A TNF receptor fragment (TNFRF), as a TNF inhibitor, and an IL-1 receptor fragment (IL-1RF), as an IL-1 inhibitor, were used. Intracerebroventricular injection of a combination of the TNFRF plus the IL-1RF significantly reduced spontaneous non-rapid eye movement sleep by 87 min over a 22-h recording period. Pretreatment of rabbits with the combination of TNFRF and IL-1RF also significantly attenuated sleep rebound after sleep deprivation. Furthermore, the TNFRF significantly attenuated IL-1-induced sleep but not fever. Finally, the IL-1RF blocked TNF-induced sleep responses but not fever. Results indicate that TNF and IL-1 cooperate to regulate physiological sleep.Both tumor necrosis factor (TNF) and interleukin (IL)-1 are somnogenic cytokines. They also induce each others production and both induce nuclear factor kappa B activation, which in turn enhances IL-1 and TNF transcription. We hypothesized that TNF and IL-1 could influence each others somnogenic actions. To test this hypothesis, we determined the effects of blocking both endogenous TNF and IL-1 on spontaneous sleep and on sleep rebound after sleep deprivation in rabbits. Furthermore, the effects of inhibition of TNF on IL-1-induced sleep and the effects of blocking IL-1 on TNF-induced sleep were determined. A TNF receptor fragment (TNFRF), as a TNF inhibitor, and an IL-1 receptor fragment (IL-1RF), as an IL-1 inhibitor, were used. Intracerebroventricular injection of a combination of the TNFRF plus the IL-1RF significantly reduced spontaneous non-rapid eye movement sleep by 87 min over a 22-h recording period. Pretreatment of rabbits with the combination of TNFRF and IL-1RF also significantly attenuated sleep rebound after sleep deprivation. Furthermore, the TNFRF significantly attenuated IL-1-induced sleep but not fever. Finally, the IL-1RF blocked TNF-induced sleep responses but not fever. Results indicate that TNF and IL-1 cooperate to regulate physiological sleep.

Non-rapid eye movement sleep is suppressed in transgenic mice with a deficiency in the somatotropic system

Sleep-wake activity was studied in a transgenic mouse model (TH-hGH) with a deficiency in the somatotropic axis (growth hormone (GH)-releasing hormone (GHRH)-GH-insulin-like growth factor-I (IGF-I)). This dwarf transgenic mouse strain expresses a human GH (hGH) reporter gene linked to 4.8 kb of the rat tyrosine hydroxylase flanking sequence, targeting the production of hGH to sites of tyrosine hydroxylase synthesis in the brain. Endogenous GH and IGF-I are suppressed in these mice, as well as GHRH. Sleep-wake activity (EEG and EMG) was recorded for 2 to 3 days in nine transgenic mice and nine wild-type littermates. Non-rapid eye movement sleep (NREMS) was significantly suppressed during both the light and the dark period in the transgenic mice; rapid eye movement sleep (REMS) was not altered. The results provide evidence that the somatotropic axis contributes to normal sleep regulation.

Brain-derived neurotrophic factor enhances spontaneous sleep in rats and rabbits

Various growth factors are involved in sleep regulation. Brain-derived neurotrophic factor (BDNF) belongs to the neurotrophin family; it and its receptors are found in normal brain. Furthermore, cerebral cortical levels of BDNF mRNA have a diurnal variation and increase after sleep deprivation. Therefore, we investigated whether BDNF would promote sleep. Twenty-four male Sprague-Dawley rats (320-380 g) and 25 male New Zealand White rabbits (4.5-5.5 kg) were surgically implanted with electroencephalographic (EEG) electrodes, a brain thermistor, and a lateral intracerebroventricular cannula. The animals were injected intracerebroventricularly with pyrogen-free saline and, on a separate day, one of the following doses of BDNF: 25 or 250 ng in rabbits; 10, 50, or 250 ng in rats. The EEG, brain temperature, and motor activity were recorded for 23 h after the intracerebroventricular injections. BDNF increased time spent in non-rapid eye movement sleep (NREMS) in rats and rabbits and REMS in rabbits. Current results provide further evidence that various growth factors are involved in sleep regulation.Various growth factors are involved in sleep regulation. Brain-derived neurotrophic factor (BDNF) belongs to the neurotrophin family; it and its receptors are found in normal brain. Furthermore, cerebral cortical levels of BDNF mRNA have a diurnal variation and increase after sleep deprivation. Therefore, we investigated whether BDNF would promote sleep. Twenty-four male Sprague-Dawley rats (320-380 g) and 25 male New Zealand White rabbits (4.5-5.5 kg) were surgically implanted with electroencephalographic (EEG) electrodes, a brain thermistor, and a lateral intracerebroventricular cannula. The animals were injected intracerebroventricularly with pyrogen-free saline and, on a separate day, one of the following doses of BDNF: 25 or 250 ng in rabbits; 10, 50, or 250 ng in rats. The EEG, brain temperature, and motor activity were recorded for 23 h after the intracerebroventricular injections. BDNF increased time spent in non-rapid eye movement sleep (NREMS) in rats and rabbits and REMS in rabbits. Current results provide further evidence that various growth factors are involved in sleep regulation.

Nuclear factor-κB-like activity increases in murine cerebral cortex after sleep deprivation

Several well-defined sleep regulatory substances, e.g., interleukin-1β, activate the heterodimeric transcription factor nuclear factor-κB (NF-κB). Several substances that inhibit sleep, e.g., interleukin-4, inhibit NF-κB activation. NF-κB activation promotes production of several additional substances thought to be involved in sleep regulation, e.g., nitric oxide. We investigated, therefore, whether there are diurnal rhythms of NF-κB activation in brain and changes in the activation after sleep deprivation. Mice were kept on a 12:12-h light-dark cycle. In one experiment, groups of mice were killed every 3 h across the 24-h cycle. In another experiment, mice were killed at 1500 after 6 h of sleep deprivation, and a group of control mice were killed at the same time. Nuclear proteins were extracted from each brain tissue sample, and NF-κB-like activity was determined with an electrophoretic mobility shift assay. In cerebral cortex, but not other areas of brain, there was a diurnal rhythm in NF-κB-like activation; highest levels were found during the light period. NF-κB-like activation was higher in cerebral cortex after sleep deprivation compared with values obtained from control mice. The results are consistent with the hypothesis that sleep regulation involves multiple gene events, some of which include enhanced production of sleep regulatory substances, the actions of which involve NF-κB activation.

Influenza Viral Infections Enhance Sleep in Mice

Abstract Sleepiness is a common perception during viral infection. Nevertheless, very little is known about the effects of viral infection on sleep. The aim of the present study was to test whether sleep was altered by influenza viral infection in mice. After 2-3 days of baseline sleep recordings, Swiss-Webster mice were infected intranasally with a lethal (H1N1) or a nonlethal (H3N2) strain of influenza virus. Sleep was recorded again for an additional 3 days. Non–rapid eye movement sleep (NREMS) was dramatically increased after inoculation of the H1N1 virus with a latency about 16 hr. Rapid eye movement sleep (REMS) was significantly suppressed after a longer latency. Both changes lasted until the end of the recording and occurred in both young (35-day-old) and adult (90- to 100-day-old) animals. Control animals did not show changes in sleep after sham infection with allantoic fluid. The H1N1 virus also caused dramatic decreases in body temperature and locomotor activities with a latency about 4-5 hr after viral inoculation. The H3N2 virus induced very similar changes in sleep, although the effects were much smaller in magnitude than those induced by the H1N1 virus, even though a much higher dose (10-fold) of the H3N2 virus was used. The present study shows that influenza viral infection induces profound and long-lasting increase of NREMS and suppression of REMS. These viral-induced changes in sleep likely represent a host-defense response. [P.S.E.B.M. 1995, Vol 210]

Diurnal variations and sleep deprivation-induced changes in rat hypothalamic GHRH and somatostatin contents

Previous reports indicate that hypothalamic growth hormone-releasing hormone (GHRH) promotes sleep and is involved in sleep regulation. The aim of our experiments was to determine whether the GHRH and somatostatin contents of the rat hypothalamus have diurnal variations and whether they are altered by sleep deprivation (SD). Hypothalamic samples were collected at 10 time points during the 24-h light-dark cycle. SD started at light onset. Hypothalamic samples were obtained after 4 and 8 h of SD and after 1 and 2 h of recovery following 8 h of SD. The peptides were determined by means of radioimmunoassay. GHRH displayed significant diurnal variations with low levels in the morning (a transient rise occurred at 1 h after light onset), gradual increases in the afternoon (peak at the end of the light period and beginning of the dark period), and decreases at night. SD induced significant GHRH depletion, which persisted during recovery. The afternoon rise was delayed, and the nocturnal decline of somatostatin was more rapid than the changes in GHRH. Although the patterns of the diurnal variations in GHRH and somatostatin were similar, there was no significant correlation between them. SD did not alter somatostatin significantly. Comparisons of the present results with previously reported changes in hypothalamic GHRH mRNA suggest that periods of deep nonrapid eye movement sleep (first portion of the light period and recovery sleep after SD) are associated with intense hypothalamic GHRH release.