Several studies have indicated that H₁ receptor agonists increase wakefulness and decrease slow wave sleep, whereas the antagonists have opposite effects (32,33). H₃ receptors play an active part in these mechanisms by regulating histamine release (21). Histamine seems to induce phase shifts in the circadian rhythm much in the same way as light impulses. A single bolus of histamine, given i.c.v. at the beginning of the subjective dark period of rats postponed, during consecutive days, the start of the subjective dark period (14). These experiments were done in free running conditions in constant darkness, allowing the animals to follow their internal rhythm without entrainment by light or other external cues. Histamine in these conditions caused a complete phase delay of the monitored circadian activities, i.e. locomotion and drinking.

In similar conditions, hamsters were given a 15 min monochromatic light pulse 2 h after the start of the subjective dark period, to delay their acquired rhythm (6,15). The phase delay caused by this treatment was reduced by FMH. If the light pulse was given during the late dark hours, a phase advance was seen which was also reduced by FMH. These experiments suggest that histamine is able to mimic the influence of light in entraining the circadian rhythm, and that FMH is able to antagonize the physiological cue, i.e. light. This would imply a role for histaminergic neurons in mediating entrainment cues in the SCN.

The role of CNS histamine in modifying the behavioral rhythms is supported by in vitro studies showing direct effects of histamine on the SCN neurons. In the rat hypothalamic slice preparation, histamine modulates the firing of the SCN neurons. Depending on the experimental conditions, either inhibitory or excitatory effect dominates (22,50). Histamine induced a phase shift of the neuronal firing rate (4) when hamster SCN tissue was studied in vitro under conditions where the SCN neurons continue to fire spontaneously, rhythmically in phase with the previous light/dark cycle of the animal. This change in vitro was in line with the in vivo results (6). If histamine was applied early in the former dark period, the peak of firing appeared later, i.e. there was a phase delay. If it was applied during the late dark hours the neuronal firing started earlier, there was a phase advance.

Little is known about which histamine receptors are involved in the induction of those phase shifts. At the level of the main circadian oscillator, SCN, the direct excitatory effects of histamine on the neuronal firing seem to be mediated via H₁ and the inhibitory effects via H₂-receptors (22,50). In the slice preparation, the phase delaying effect of histamine appears to be via H₁-receptors (4), but there is no information on which receptor type promotes the phase advance.

Based on in vivo studies, it has been proposed that the effects of histamine on the circadian rhythm are mediated through receptors other than histamine receptors. In the hamster, the light-induced phase shifts were reduced by FMH (6), but they could not be blocked by H₁-, H₂- or H₃-antagonists (7). One of the possible receptors is the NMDA receptor, which is involved in the phase shifting effect of light in the hamster (3). Histamine has been shown to enhance NMDA receptor mediated responses in vitro, possibly through interaction with the polyamine binding site (2,65).

There is clinical evidence that various drugs, e.g. brain penetrating antihistaminics and neuroleptics, which block histamine receptors may aggravate a pre-existing seizure disorder. When metoprine was given to rats, there was a clear dose-related protection from maximal electroshock seizures (59). Metoprine also reduced the severity of audiogenic seizures in genetically audiogenic seizure sensitive (AGS) rats (61) whereas the results with chemical seizures were controversary (54,55). These results suggested that histaminergic neurons are involved in mechanisms which inhibit generalizations of epileptic discharges in the brain perhaps as a part of an anticonvulsive inhibitory transmitter system.

In mice, elevation of brain histamine concentrations either by metoprine or L-histidine reduces the duration of electrically-induced clonic convulsions. Conversely, FMH, which decreases the brain histamine levels, increases the duration of clonic convulsions (73). The anticonvulsive effect seems to be mediated by H₁-receptors; however, H₃-receptors may also participate by regulating the release of endogenous histamine (69,71,72). Analogously, PET studies have revealed abnormal amounts of H₁-receptors around the epileptic focus in human complex partial seizures (10).

Studies with mice have shown that the histaminergic system is important especially in young animals (70). This is in line with the clinical evidence that children are more sensitive to the proconvulsive side effects of drugs with H₁-receptor antagonistic properties. This is supported also by the finding that CSF histamine concentration in a group of children with febrile convulsions was lower than in another group of febrile children without convulsions, suggesting a protective role for histamine against excessive neuronal excitation (16). Moreover, in AGS rats, lower endogenous levels of histamine were found in regions involved in seizure propagation (38,62).

The ascending neurotransmitter systems, including cholinergic, noradrenergic and serotonergic, as well as the histaminergic neurons, have connections to the thalamus, an important gateway controlling and filtering impulses to the cerebral cortex. Histaminergic stimulation may switch the thalamic neurons from rhythmic burst-like firing to single-spike activity, a condition which favours accurate transmission and processing of sensory information, promoting wakefulness and arousal (26-28,57,58).

Little is known about the rhythms in histamine release or metabolism with respect to sleep stages. There is some evidence that during rebound sleep following a 72-hour REM sleep deprivation, rats have a decreased histamine turnover (MH/HA ratio) in the anterior hypothalamus. Due to the platform method used, however, stress may have interfered with those results (42). In anaesthetised animals, a negative correlation has been found between the ultradian rhythms of delta and theta EEG frequeNcy and the ultradian release of histamine (44). This suggests that a low release of histamine is accompanied with the EEG signs of sedation. It would be most interesting to see in behaving animals if the reverse is true with the signs of arousal.

The circadian amplitude of sleep-wakefulness cycle is flattened with FMH treatment, in the rat wakefulness is increased during the day and slow wave sleep is increased during the night (13,18). This would lead to sleep disturbances, and different disease states, accompanied by sleep disturbances, may also have changes in the brain histamine. For example patients with portal systemic encephalopathy (PSA) and many cirrhotic patients complain of sleep disturbances. These conditions can be simulated by the portacaval anastomosis (PCA) model in the rat. One of the most striking consequences of this operation is the remarkable elevation of brain histamine content soon after the operation (8) and the increased metabolism of histamine as a long term effect (23,24).

When cortical EEG was measured one and six months after the PCA operation, the signs of drowsiness, the synchronized low-frequency, high amplitude activities (delta waves and spindles), were reduced during the day-time recording period. There was a significant negative correlation between the spindling time and the frontal cortex histamine concentration (25). Thioperamide, an H₃ blocker, reduced the incidence of thalamus-regulated EEG spindles. Furthermore, it reduced the spectral power of the low frequency (1,5-5 Hz) EEG (63). These results provide evidence for the involvement of endogenous brain histamine and H₃ receptors in the regulation of the neocortical synchronization pattern, which is considered to be linked to arousal control.

In conclusion: Histaminergic activity shows a clear circadian rhythm: high levels during the active period (in rodents at night, in monkeys and humans during the day), and low levels during the sleep period. Histamine appears to be necessary for the maintenance of the circadian rhythmicity of sleep-wakefulness cycles, food intake, motility and adrenocortical hormone release. In addition, a role for histaminergic neurons in light entrainment is implicated. In phase shift studies, histamine given centrally seems to entrain the activity rhythm in the same way as light impulses and FMH seems to block the entrainment by light. Importantly, histamine participates in the control of arousal and may be implicated in the sleep disturbances in hepatic encephalopathy. Furthermore, evidence suggests a role for histamine in neuronal excitability and seizure susceptibility both in animals and humans. Thus we conclude that histamine may exert modifying effects on circadian rhythmicity and neuronal excitability.