The reciprocal link between sleep and immune responses.

Good sleep is necessary for both physical and mental health; sleep and immune responses are reciprocally and closely linked. Sleep loss impairs the immune response, while, on the other hand, the immune response, activated for instance by an infection, alters sleep. Sleep alterations induced by immune activation are mediated by cytokines such as interleukin-1. In the past, it was thought that cytokines were produced only by the immune system, and active only there as signaling molecules. Today it is clear that IL-1 and other cytokines are present and active in the healthy brain, where they physiologically interact with the brain circuits and the neurotransmitter systems (for instance the serotonergic, GABAergic, and cholinergic systems) that control sleep. These interactions are altered by immune response, and, as a result, non-rapid eye movement (NREM) sleep is increased and fragmented, whereas rapid eye movements (REM) sleep is inhibited.

Interleukin-1 inhibits firing of serotonergic neurons in the dorsal raphe nucleus and enhances GABAergic inhibitory post-synaptic potentials.

In vitro electrophysiological data suggest that interleukin-1 may promote non-rapid eye movement sleep by inhibiting spontaneous firing of wake-active serotonergic neurons in the dorsal raphe nucleus (DRN). Interleukin-1 enhances GABA inhibitory effects. DRN neurons are under an inhibitory GABAergic control. This study aimed to test the hypothesis that interleukin-1 inhibits DRN serotonergic neurons by potentiating GABAergic inhibitory effects. In vitro intracellular recordings were performed to assess the responses of physiologically and pharmacologically identified DRN serotonergic neurons to rat recombinant interleukin-1beta. Coronal slices containing DRN were obtained from male Sprague-Dawley rats. The impact of interleukin-1 on firing rate and on evoked post-synaptic potentials was determined. Evoked post-synaptic potentials were induced by stimulation with a bipolar electrode placed on the surface of the slice ventrolateral to DRN. Addition of interleukin-1 (25 ng/mL) to the bath perfusate significantly decreased firing rates of DRN serotonergic neurons from 1.3 +/- 0.2 Hz (before administration) to 0.7 +/- 0.2 Hz. Electrical stimulation induced depolarizing evoked post-synaptic potentials in DRN serotonergic neurons. The application of glutamatergic and GABAergic antagonists unmasked two different post-synaptic potential components: a GABAergic evoked inhibitory post-synaptic potentials and a glutamatergic evoked excitatory post-synaptic potentials, respectively. Interleukin-1 increased GABAergic evoked inhibitory post-synaptic potentials amplitudes by 30.3 +/- 3.8% (n = 6) without affecting glutamatergic evoked excitatory post-synaptic potentials. These results support the hypothesis that interleukin-1 inhibitory effects on DRN serotonergic neurons are mediated by an interleukin-1-induced potentiation of evoked GABAergic inhibitory responses.

Serotonergic activation stimulates the pituitary-adrenal axis and alters interleukin-1 mRNA expression in rat brain.

Interactions between neurotransmitters and immunomodulators within the central nervous system may be functionally relevant for communication between the immune system and the brain. Previous studies indicate that cytokines such as interleukin-1 (IL-1) alter activity of the serotonergic system at multiple levels. This study tested the hypothesis that serotonergic activation modulates cytokine mRNA expression in brain. Serotonergic activation was induced by injecting rats intraperitoneally (i.p.) prior to dark onset with the serotonin precursor L-5-hydroxytryptophan (5-HTP; 100 mg/kg). Cytokine mRNA expression in discrete brain regions at selected time points was determined by means of ribonuclease protection assay. Plasma corticosterone concentrations were also measured to determine if the hypothalamic-pituitary-adrenal axis is activated in response to this treatment, which potentially could exert feedback regulating cytokine message expression in brain. Plasma corticosterone was elevated for 4 h after 5-HTP administration. At this time IL-1alpha mRNA expression was reduced in the hippocampus, hypothalamus, and brainstem, and IL-1beta mRNA was reduced in the hippocampus. Six hours after 5-HTP injection, IL-1beta mRNA increased in the hypothalamus. These results show that activation of the serotonergic system affects cytokine message expression in rat brain, possibly by actions of corticosterone.

Rat strains that differ in corticotropin-releasing hormone production exhibit different sleep-wake responses to interleukin 1.

Corticotropin-releasing hormone (CRH) is a mediator of responses to a variety of stressors, including immune challenge. CRH and the hypothalamic-pituitary-adrenal (HPA) axis constitute a negative feedback mechanism for actions of immunomodulators, such as interleukin (IL) 1. CRH is a potent inducer of waking, whereas IL-1 induces slow-wave sleep (SWS). We hypothesize that the complex changes in sleep-wake behavior during immune challenge are mediated in part by CRH and its antagonism of IL-1-induced enhancement of SWS. To further explore this hypothesis, we administered IL-1beta intracerebroventricularly into rats of genetically related strains that differ in CRH/HPA axis responsiveness to IL-1 and determined subsequent alterations in their sleep-wake behavior. Sprague-Dawley rats responded to central administration of IL-1 with alterations in sleep-wake behavior as previously reported; SWS increased, and rapid eye movement sleep (REMS) and waking decreased. CRH and the HPA axis of Lewis rats are reported to be hyporesponsive to challenge; the onset of the IL-1-induced increase in SWS was quicker and the peak magnitude of the response greater than in Sprague-Dawley rats. In contrast, Fischer 344 rats exhibit greater CRH release and HPA axis activation in response to IL-1. IL-1 induced a profound and transient increase in waking of Fischer 344 rats before SWS increased. The febrile responses to IL-1 of Fischer 344 and Lewis rats were identical and of greater magnitude than those observed in Sprague-Dawley rats. Pretreatment with the CRH receptor antagonist alpha-helical CRH(9-41) blocked the initial IL-1-induced increase in waking of Fischer 344 rats. CRH receptor blockade did not affect the IL-1-induced alterations in sleep-wake behavior of Lewis or Sprague-Dawley rats or brain temperature of any rat strain. These observations support the hypothesis that CRH is both a modulator of responses to IL-1 and is involved in the regulation of waking.

5-Hydroxytryptophan, but not L-tryptophan, alters sleep and brain temperature in rats.

The precise role of serotonin (5-hydroxytryptamine) in the regulation of sleep is not fully understood. To further clarify this role for 5-hydroxytryptamine, the 5-hydroxytryptamine precursors L-tryptophan (40 and 80 mg/kg) and L-5-hydroxytryptophan (25-, 50-, 75-, 100 mg/kg) were injected intraperitoneally into freely behaving rats 15 min prior to dark onset, and subsequent effects on sleep-wake activity and cortical brain temperature were determined. L-5-hydroxytryptophan, but not L-tryptophan, induced dose-dependent changes in sleep-wake activity. During the 12-h dark period, non-rapid eye movement sleep was inhibited in post-injection hours 1-2 by the two lowest L-5-hydroxytryptophan doses tested, while the two highest doses induced a delayed increase in non-rapid eye movement sleep in post-injection hours 3-12. These highest doses inhibited non-rapid eye movement sleep during the subsequent 12-h light period. The finding that L-5-hydroxytryptophan, but not L-tryptophan, induced a dose-dependent and long-lasting decrease in cortical brain temperature regardless of whether or not non-rapid eye movement sleep was suppressed or enhanced contributes to a growing list of conditions showing that sleep-wake activity and thermoregulation, although normally tightly coupled, may be dissociated. The initial non-rapid eye movement sleep inhibition observed following low doses of L-5-hydroxytryptophan may be attributable to increased serotonergic activity since 5-hydroxytryptamine may promote wakefulness per se, whereas the delayed non-rapid eye movement sleep enhancement after higher doses may be due to the induction by 5-hydroxytryptamine of sleep-inducing factor(s), as previously hypothesized. The period of non-rapid eye movement sleep inhibition beginning 12 h after administration of L-5-hydroxytryptophan doses that increase non-rapid eye movement sleep is characteristic of physiological manipulations in which non-rapid eye movement sleep is enhanced. The results of the present study suggest that the complex effects of 5-HT on sleep depend on the degree and time course of activation of the serotonergic system such that 5-HT may directly inhibit sleep, yet induce a cascade of physiological processes that enhance subsequent sleep.

Blockade of 5-hydroxytryptamine (serotonin)-2 receptors alters interleukin-1-induced changes in rat sleep.

Recent data suggest that interleukin-1-induced enhancement of non-rapid eye movement sleep is mediated, in part, by the serotonergic system. To determine if sleep changes induced by interleukin-1 are mediated by a specific serotonergic receptor subtype, we evaluated interleukin-1 effects on sleep in rats pretreated with the 5-hydroxytryptamine (serotonin)-2 receptor antagonist ritanserin. Ritanserin (0.63 mg/kg, intraperitoneally) by itself did not alter sleep-wake behavior, although it did reduce cortical brain temperature. Interleukin-1 (5 ng, intracerebroventricularly) enhanced non-rapid eye movement sleep, suppressed rapid eye movement sleep, and induced a moderate febrile response. Pretreatment with ritanserin completely blocked the febrile response to interleukin-1 and abolished the interleukin-1-induced enhancement in non-rapid eye movement sleep that occurred during postinjection hours 3-4, without altering interleukin-1 effects on rapid eye movement sleep. The present data suggest that serotonin may partially mediate interleukin-1 effects on sleep by interacting with 5-hydroxytryptamine (serotonin)-2 receptors. These results also suggest that interactions between the serotonergic system and interleukin-1 may be important in regulating sleep-wake behavior.

Hypothalamic serotonergic activity correlates better with brain temperature than with sleep-wake cycle and muscle tone in rats.

The activity of the serotonergic system varies in phase with the sleep-wake cycle, which is associated with changes in several physiological functions, including electroencephalographic activity, brain temperature, and locomotion. The aim of the present study was to clarify which of these parameters correlates better with serotonergic activity in spontaneous conditions. Voltammetric recordings by telemetry of serotonergic metabolism in the medial preoptic area and polygraphic recordings of sleep-wake activity (by means of electroencephalographic delta band, brain cortical temperature and neck electromyographic activity recordings) were simultaneously performed in freely moving rats. Univariate analyses of variance revealed that each variable under investigation was statistically correlated with serotonergic metabolism. When the variables were entered into the model simultaneously, both partial correlation and step-wise multiple regression analyses indicated that the highest correlation exists between serotonergic metabolism and brain cortical temperature. The present data show that serotonergic activity in the medial preoptic area is closely linked to physiological changes in brain temperature.

Hippocampal type 1 (movement-related) theta rhythm positively correlates with serotonergic activity.

To investigate the relationship between the hippocampal [symbol: see text] activity (or Rhythmical Slow Activity, RSA) and the hippocampal serotonergic activity during spontaneous behavior, simultaneous recordings of i) hippocampal EEG, ii) sleep-wake activity, and iii) hippocampal levels of the serotonin (5-HT) metabolite 5-hydroxyndolacetic acid (5-HIAA--measured by in vivo voltammetry and infrared telemetry) were performed. The results show that hippocampal type 1 RSA recorded during wakefulness and voluntary movements (such as walking), is positively correlated to hippocampal 5-HIAA levels. Since in the experimental conditions used in the study, 5-HIAA levels are a reliable index of 5-HT release, the results support the hypothesis that hippocampal type 1 RSA is generated by a serotonergic mechanism. In contrast, hippocampal type 2 RSA recorded during desynchronized sleep is negatively correlated with 5-HT release, suggesting a different neurochemical mechanism for its production. These results also show that, in the experimental condition of this study, hippocampal RSA power spectrum has a main peak frequency of 3.5 during wakefulness, and of 6.5 Hz during desynchronized sleep.

Sleep as a behavioral model of neuro-immune interactions.

The central nervous system, by a variety of mechanisms engages in constant surveillance of the peripheral immune system. Alterations in the status of the peripheral immune system induced by an invading pathogen for example, are quickly detected by the central nervous system, which then responds by altering physiological processes and behavior in an attempt to support the immune system in its efforts to eliminate the pathogen. Sleep is one of several behaviors that are dramatically altered in response to infection. Immune-active substances such as the pro-inflammatory cytokines interleukin-1 and tumor necrosis factor, either directly or indirectly via interactions with neurotransmitters or neurohormones are involved in the regulation of sleep. Because these cytokines increase during infection, they are likely candidates for mediating the profound alterations in sleep that occur during infection. Since regulation of behavior is the function of the central nervous system, infection-induced alterations in behavior provide a unique model for the study of neuro-immune interactions.

Cytokine-neurotransmitter interactions in the brain.

The data reviewed in this study show that immune-active molecules, such as infectious agents and their components, and cytokines, may induce profound alterations in several neurotransmitters in the CNS. The activation of the immune system elicits fever, behavioral and neuroendocrine changes and may be involved in neuropathological changes occurring in CNS conditions. These effects may be achieved through and accounted for by the changes induced in central neurotransmitters and in the neuroendocrine system by immune challenges. The present review will summarize the available evidence of the reciprocal interactions between cytokines and neurotransmitters in the CNS.

Cytokine-neurotransmitter interactions in the brain

The data reviewed in this study show that immune-active molecules, such as infectious agents and their components, and cytokines, may induce profound alterations in several neurotransmitters in the CNS. The activation of the immune system elicits fever, behavioral and neuroendocrine changes and may be involved in neuropathological changes occurring in CNS conditions. These effects may be achieved through and accounted for by the changes induced in central neurotransmitters and in the neuroendocrine system by immune challenges. The present review will summarize the available evidence of the reciprocal interactions between cytokines and neurotransmitters in the CNS.

Muramyl dipeptide and IL-1 effects on sleep and brain temperature after inhibition of serotonin synthesis.

The role of the interactions between serotonin (5-HT) and muramyl dipeptide (MDP) and interleukin-1 (IL-1) in sleep control and thermoregulation was evaluated. To this purpose, MDP and IL-1 were injected intracerebroventricularly at dark onset into freely moving rats pretreated twice intraperitoneally with para-chlorophenylalanine (PCPA) (300 mg/kg), which depletes brain 5-HT and causes insomnia. Fever and slow-wave sleep (SWS) enhancement induced by 150 pmol MDP were completely blocked in PCPA-pretreated rats. Only the first phase of the biphasic increase in SWS induced by 2.5 ng IL-1 was suppressed by PCPA pretreatment, whereas fever remained unaffected. These results suggest that 1) MDP effects on both sleep-wake activity and brain cortical temperature are mediated by the serotonergic system; 2) the mechanisms mediating the first and the second phases of IL-1-induced SWS excess are different: 5-HT could be involved in the first phase, but not in the second one; and 3) the 5-HT system does not appear to be involved in IL-1-induced fever.

Interleukin-1 induces changes in sleep, brain temperature, and serotonergic metabolism.

Simultaneous recordings of sleep-wake activity and of serotonergic metabolism in the medial preoptic area were performed in freely moving rats after the intracerebroventricular injection of interleukin-1 (IL-1) at dark onset. IL-1 (2.5 ng) induced a biphasic increase in slow-wave sleep and an early increase in serotonergic metabolism starting 30 min postinjection. Phasic, state-specific changes (which have been described in spontaneous sleep) were superimposed on this tonic, overall increase in serotonergic metabolism. IL-1 (25 ng) induced an increase in wakefulness and a delayed increase in serotonergic metabolism, which started 120 min postinjection. This suggests that the time course of the serotonergic activation could play a role in mediating IL-1 effects on sleep. Both doses of IL-1 induced a similar and significant increase in brain cortical temperature, suggesting that IL-1 effects on sleep are not a secondary effect of the increase in cortical temperature and that the serotonergic system is not involved in IL-1-induced fever.

Muscarinic receptor subtypes in the medial preoptic area and sleep-wake cycles.

To clarify which muscarinic receptor subtype(s) mediate changes in sleep and cortical temperature (Tcort) induced by carbachol microinjections into the medial preoptic area (MPA), pirenzepine, tripitramine and +/- p < > -fluorohexahydro-sila-difenidol (p-F-HHSiD), which are highly selective muscarinic M1, M2 and M3 antagonists, respectively, were microinjected into the MPA of rats. Whereas pirenzepine (3.45 and 7.08 nmol) and p-F-HHSiD (3.90 and 7.80 nmol) were without effect, tripitramine (0.67 and 3.37 nmol) enhanced wakefulness, decreased slow wave and desynchronized sleep, and raised Tcort with the higher dose. The data suggest that in the MPA only M2 muscarinic subtypes may be functionally important in mediating the cholinergic effects on sleep and thermoregulation.

Stimulation of cholinergic receptors in the medial preoptic area affects sleep and cortical temperature.

The medial preoptic area (MPA), a cholinoceptive brain area devoid of cholinergic cells, plays an important role in the regulation of different physiological functions, particularly sleep control and thermoregulation. To investigate the effects of the stimulation of cholinergic receptors in this area on sleep and cortical temperature (Tcort), carbachol (a mixed cholinergic agonist) was directly microinjected into the MPA of freely moving rats. Carbachol (0.25 and 0.5 microgram, corresponding to 1.37 and 2.74 nmol) microinjection induced an increase in wakefulness and an inhibition of both slow wave and desynchronized sleep phases. The temperature of the cerebral cortex was reduced in comparison with control conditions (saline microinjection). Sterile needle insertion and saline microinjections induced a significant increase in Tcort, but no changes in the sleep-wake cycle compared with the handling of the animal. The results suggest that 1) carbachol microinjection into the MPA can activate an arousal-generating system and affect thermoregulatory mechanisms, and 2) sleep and temperature responses may be dissociated.

Sleep regulation: interactions among cytokines and classical neurotransmitters.

The role of classical neurotransmitters in sleep regulation is amply documented (Hobson and Steriade, 1986). In recent years evidence has been gathered that immunoactive molecules, infectious agents and their components, or cytokines play some part in sleep regulation (Krueger and Obál, 1994; Opp et al., 1992; Moldofsky, 1994). Different cytokines possess hypnogenic properties when injected centrally or systemically to different animal species and their role in physiological sleep regulation is currently under investigation. Little is known of how cytokines and classical neurotransmitters interact and of the relevance of this interaction in sleep induction and maintenance. The present paper (i) reviews data on this topic; (ii) proposes a unitary interpretation whenever possible; and (iii) raises questions that might be addressed by future studies.

Selective blockade of different brain stem muscarinic receptor subtypes: effects on the sleep-wake cycle.

Changes induced in the sleep-wake cycle by pontine microinjections of muscarinic antagonists were studied in freely moving rats, instrumented for chronic polygraphic recordings. Pirenzepine (PIR), methoctramine (MET) and p-fluoro-hexahydro-siladifenidol (p-F-HHSiD), which are highly selective M1, M2 and M3 antagonists, respectively, were dissolved in 0.1 microliter of sterile isotonic saline (0.2 microliter of distilled water for p-F-HHSiD) and injected into the pontine reticular nucleus, where the administration of 0.5 microgram carbachol (a mixed muscarinic agonist) induced a 52% increase in the amount of desynchronized sleep (DS) over a 6 h recording period. The blockade of M2 receptors was shown to (i) antagonize DS, by increasing its latency and decreasing its percentage, (ii) decrease slow wave sleep, and (iii) enhance wakefulness. These effects were dose-dependent. No changes in the sleep-wake cycle were observed following microinjection of M1 or M3 antagonists. The results support the hypothesis that at the brain stem level only M2 receptors are involved in sleep mechanisms and, particularly, in the generation and maintenance of DS.

Changes in the serotonergic system during the sleep-wake cycle: simultaneous polygraphic and voltammetric recordings in hypothalamus using a telemetry system.

Changes in the serotonergic system in the posterior hypothalamus of freely moving rats were related to sleep and wakefulness using in vivo voltammetry (with carbon fiber microelectrodes) and polygraphic recordings. By using an optoelectronic telemetry system for the voltammetric signals, electrical cross-talk between the two settings was avoided and simultaneous neurochemical and electro-physiological recordings could be made so that a detailed time course of events could be obtained. Extracellular levels of the serotonin metabolite, 5-hydroxy-indoleacetic acid, measured every 2 min, increased with wakefulness and decreased with sleep: levels were significantly lower during desynchronized sleep than slow wave sleep. In vivo voltammetry associated with the optoelectronic telemetry system appears to be a useful tool for studying the relationship between neurochemical changes and electrophysiological events.

An IL-1 receptor and an IL-1 receptor antagonist attenuate muramyl dipeptide- and IL-1-induced sleep and fever.

It is hypothesized that the somnogenic and pyrogenic effects of muramyl dipeptide (MDP) are mediated via enhanced interleukin-1 (IL-1) production. To test this hypothesis the effects of intracerebroventricular (icv) administration of a recombinant human soluble type I IL-1 receptor (sIL-1r) and of the IL-1 receptor antagonist (IL-1ra) on MDP-induced sleep and fever were evaluated in rabbits. The sIL-1r recognized rabbit IL-1 beta, but it did not affect sleep or brain temperature across the dose range tested (1-50 micrograms) when injected icv into normal rabbits. Pretreatment of rabbits with 50 micrograms sIL-1r or 10 micrograms IL-1ra blocked human recombinant IL-1-enhanced nonrapid eye movement (NREM) sleep and fever. Thus both the sIL-1r and the IL-1ra were effective antagonists of IL-1 actions. When the animals were pretreated with either 50 micrograms sIL-1r or with 10 or 100 micrograms of the IL-1ra, the somnogenic effects of 150 pmol MDP were attenuated. However, the sIL-1r had little effect on MDP-induced febrile responses. These results suggest that the sIL-1r and the IL-1ra can function as antagonists of IL-1 actions in vivo and that MDP-induced sleep and fever are partially mediated by IL-1.

The influence of the muscarinic receptor subtypes on the sleep-wake cycle.

The specific role played by different muscarinic receptor subtypes in sleep regulation is investigated and discussed. On the basis of the results obtained with intracerebroventricular injections of selective muscarinic antagonists into freely moving rats, it is suggested that each muscarinic receptor subtype induces different and specific changes in sleep phases and cortical desynchronization processes.