The delay in searching for a histaminergic neuronal system, in contrast to the exploration of other neurotransmitters systems, made it initially difficult to accept that histamine could have a specific transmitter role. Hence, the identification of a central histaminergic neuronal system (Panula et al., 1984; Watanabe et al., 1983), has been a real breakthrough, providing new perspectives in histamine research. Histaminergic cell bodies are confined to the tuberomammillary nucleus of the hypothalamus, and project efferent fibers predominantly ipsilaterally with multifold arborizations into the whole central nervous system, including most subcortical nuclei and the cerebral cortex (Wada et al., 1991). These characteristics suggest that the histaminergic system is a regulatory center for whole-brain activity.
Cholinergic systems have been closely linked to cognitive function (Haroutunian et al., 1985), however, over the past decade evidence accumulated that the "cholinergic hypothesis of learning" (Bartus et al., 1982) is too reductionistic (Sarter et al., 1990). Also other neurotransmitters, such as dopamine, GABA, glutamate, noradrenaline and serotonin may affect cognitive processes (Decker et al., 1991). Thus, since cholinergic dysfunctions might well interact with those of other neurotransmitter systems to cause additive or even synergistic effects on cognition, the role of interactions between ACh and other neurotransmitters affecting cognition is of considerable interest. Histaminergic neurotransmission is involved in the regulation of numerous physiological functions, including learning and memory (Haas et al., 1991; Huston et al., 1997; Tasaka, 1994) with a mechanism possibly resting on histaminergic-cholinergic interactions (Passani et al., 1998).
HISTAMINE AND ACh
An interaction between ACh and histamine in the cortex is well documented. Two different laboratories reported that H3 receptors are involved in the inhibitory effects of histamine on potassium-evoked release of radiolabelled ACh from rat cortical slices preloaded with tritiated choline (Arrang et al., 1995; Clapham et al., 1992). This effect was confirmed in vivo, using microdialysis to simultaneously administer histamine and monitor changes in endogenous ACh release from cortex of freely moving rats. Cortical perfusion with histamine failed to alter spontaneous release of ACh, but resulted in a concentration-dependent inhibition of its potassium-evoked release up to more than 60% (Blandina et al., 1996a). This effect is attributable to H3 receptor activation, since agonists selective for the H3 receptor, such as R-alpha-methylhistamine (Arrang et al., 1987), imetit (Garbarg et al., 1992), and immepip (Vollinga et al., 1994), mimicked the effect of histamine with a slightly greater potency (Blandina et al., 1996a). Moreover, the inhibitory action of 100 µM histamine was completely antagonized by histaminergic antagonists such as clobenpropit and thioperamide (Blandina et al., 1996a; Blandina et al., 1996b), added to the perfusion medium at concentrations that blocked selectively H3 receptor-mediated responses (Arrang et al., 1987; VanderGoot et al., 1992).
H3 receptor-induced inhibition of potassium-evoked release of ACh was completely abolished in cortices in which the traffic of action potentials was blocked by tetrodotoxin, a voltage-dependent sodium-channel blocker (Blandina et al., 1996a). Therefore, H3 receptors modulating ACh release are likely located neither presynaptically on cholinergic nerve terminals, nor on non-cholinergic nerve endings impinging on the former. They are most likely somatodendritic receptors on interneurons, the excitation of which produces sodium-dependent action potentials that release an intermediary modulatory substance. Consistently, H3 receptors decreased in the cerebral cortex after local infusion of neurotoxins (Cumming et al., 1991; Pollard et al., 1993), and H3 receptor stimulation failed to alter the potassium-evoked release of tritiated ACh from synaptosomes of the entorhinal cortex (Arrang et al., 1995). Bicuculline, a GABAA receptor antagonist, reversed the inhibition of ACh release induced by immepip, thus sugges ting a GABAergic involvement (Giorgetti et al., 1997).
More direct evidence of this involvement was provided by the observation that immepip, at a concentration that produced a maximal inhibition of potassium-evoked ACh release (Blandina et al., 1996a), enhanced also potassium-evoked release of GABA from the cortex of freely moving rats up to more than 50% (Giorgetti et al., 1997). These findings strongly suggest that H3 receptors, located postsynaptically on intrinsic perikarya, facilitate the release of GABA, which, in turn, inhibits ACh release. The most simple hypothesis is that these interneurons directly innervate the cholinergic presynaptic terminals and reduce ACh release. There is evidence that the cortical GABAergic system exerts a tonic inhibition of spontaneous release of ACh from the cortex, and that this inhibitory tone is maximal (Giorgetti et al., 1996). This could explain why neither histamine nor either of H3 receptor agonists altered spontaneous ACh release (Blandina et al., 1996a), much of which is tetrodotoxin sensitive (Blandina et al., 1996a). Under resting conditions, since the inhibition of ACh release caused by GABA is maximal, H3 activation would have no effect on spontaneous ACh release. However, activation of H3 receptors, by increasing the release of GABA, will antagonize the potassium-induced depolarization, thus, depress, at least partially, potassium-evoked ACh release.
Alternatively, another synaptic arrangement consonant with the lack of H3 modulation of spontaneous release is that the activated interneuron inhibits the release of an excitatory presynaptic modulator of cholinergic terminals. If this excitatory pathway were not spontaneously active, H3 activation would have no effect on spontaneous ACh release. In the presence of potassium, this excitatory modulator would be released and enhance the depolarization-induced release of ACh. Activation of H3 receptors would remove this enhancement and partially, but not completely, depress potassium-evoked ACh release. Cortical GABA interneurons control the activity of large populations of principal cells through their extensive axon arborization. Therefore, any pathway, even if relatively sparse such as the histaminergic pathway, may exert a powerful effect on the activity of the cortex if it modulates the activity of local GABA interneurons.
However, the cholinergic innervation of the cortex is provided by cholinergic neurons localized in the basal forebrain, more precisely in the nucleus basalis magnocellularis (NBM) (Mesulam et al., 1983). Histaminergic neurons send efferents to the cortex which run in the medial forebrain bundle, through the basal forebrain. Thus, histamine might modulate the activity of NBM neurons, including the cholinergic ones. Indeed, an electrophysiological study in guinea-pig basal forebrain slices has shown that histamine excites NBM cholinergic neurons causing a depolarization, mainly through H1 receptor activation (Khateb et al., 1995). Therefore, histaminergic neurons might also facilitate cortical cholinergic release. Yet an intact, whole animal approach may yeld important insight into the physiological role of histamine in modulating cortical cholinergic activity. We, therefore, examined the effects of NBM perfusion with histamine and histaminergic drugs on the extracellular levels of ACh in the frontoparietal cortex of freely moving rats with a dual-probe microdialysis. Briefly, male Wistar rats (200-250 g), were anesthetized with chloral hydrate (400 mg/kg, i.p.), and implanted with both a vertical microdyalisis probe in the NBM to deliver locally the different drugs, and a transversal microdyalisis probe in the cortex to measure the output of ACh. Twenty-four hours after surgery, rats were perfused with Ringer solution. Physostigmine (7 microM) was added to the medium perfusing the cortex. The flow rate was 2 microL/min. Ten-minutes fractions were collected. ACh content was measured in the cortical dialysates by HPLC with electrochemical detector. Spontaneous release was obtained by averaging ACh content in the first four samples. Drugs were dissolved into the medium perfusing the NBM. Accurate placement of microdialysis membrane was verified post mortem by gross visualization of coronal sections.
This study demonstrates that histamine administered into NBM facilitates cortical ACh release. Rat cerebral cortex spontaneously released ACh at stable rates, 1.7 ± 0.3 pmol/10 min (n = 44). The administration of histamine (0.5-100 microM) into the NBM increased concentration-dependently the output of ACh from the cortex of freely moving rats by about 100%. The release of ACh returned to basal values upon NBM superfusion with agonist-free medium. ACh release elicited by 50 microM histamine was insensitive to blockade of H2 and H3 receptors by means of cimetidine and thioperamide. Conversely, triprolidine (0.1-0.5 microM) and mepyramine (0.3-1 microM), both H1 receptor antagonists, reduced significantly the effect of 50 µM histamine. Consistently, methylhistaprodifen, an H1 receptor agonist, mimicked the effect of histamine, whereas dimaprit, an H2 receptor agonist, and R-alpha-methylhistamine failed to modify ACh cortical release.
As a consequence of reduced availability of ACh in the synaptic cleft, which may result in cognitive deficits (Quirion et al., 1995), cortical H3 receptors activation seems to have relevance learning and memory. Indeed, systemic administration of doses of RAMH and imetit that reduced potassium-evoked cortical ACh release (Blandina et al., 1996a), also impaired rat performance in cognitive tests (Blandina et al., 1996a). Immepip, also impaired animal performance in the olfactory, social memory test, based on the recognition of a juvenile rat by a male, adult and sexually-experienced rat (Prast et al., 1996) Since H3 receptor agonists impair cognitive performances, one might envisage that H3 receptor antagonists might exert procognitive effects. Indeed, thioperamide, an H3 receptor antagonist, improved rat performance in the olfactory, social memory test (Prast et al., 1996). Other studies, however, report that the procognitive effects of H3 receptor antagonists become fully evident only when behavioral def icits are pronounced. For example, while thioperamide improves significantly the response latency in a passive avoidance response in senescence-accelerated mice (these animals showed a marked age-accelerated deterioration in learning tasks of passive avoidance), it is ineffective in normal-rate aging mice (Meguro et al., 1995). We investigated the capacity of H3 receptor antagonists to influence scopolamine-induced amnesia in rats, measured by object recognition (Ennaceur et al., 1988) and a passive avoidance response (Blandina et al., 1996a). These tasks are impaired by cholinergic blockade, and serve to measure a form of episodic memory, possibly localized in the frontal cortex (Goldman-Rakic, 1987).
Passive avoidance response
According to their brain penetration characteristics after peripheral administration (Barnes et al., 1993; Mochizuki et al., 1996; Oishi et al., 1989), thioperamide (5 mg/kg, i.p.) and clobenpropit (15 mg/kg, s.c.) were injected 120 min prior to the training trial. Saline (250 microL, i.p.) and scopolamine (0.2 mg/kg, s.c.) were administered 30 min prior to the training trial. Each drug was freshly prepared and dissolved in 0.9% wt/vol NaCl solution (saline) to permit the injection of a constant volume of 1 ml/kg to each rat. The retention trial was performed 24 h after the training trial. Analysis of the escape latencies revealed a significant group effect (ANOVA, F5,95 = 14.5; P < 0.0001). Scheffe's post-hoc analysis showed that latencies of rats receiving scopolamine alone were significantly shorter from those of rats receiving saline (P < 0.0001), thioperamide alone (P < 0.0007), scopolamine in combination with thioperamide (P < 0.0197), clobenpropit alone (P < 0.0001), and scopolamine in c ombination with clobenpropit (P < 0.0134). No other comparison was significant. Thus, scopolamine impaired passive avoidance behavior. Thioperamide and clobenpropit, although alone were without effect on passive avoidance behavior, antagonized the effect of scopolamine.
Saline (250 microL, i.p.) and scopolamine (0.2 mg/kg, s.c.) were injected 30 min, thioperamide (5 mg/kg, i.p.) and clobenpropit (15 mg/kg, s.c.) 120 min prior to the first trial. Each drug was freshly prepared and dissolved in 0.9% wt/vol NaCl solution (saline) to permit the injection of a constant volume of 1 ml/kg to each rat. Analysis of the exploration time during this trial failed to reveal any significant group effect . The second trial was given 60 min after the first trial. In the second trial, control rats spent significantly more time exploring the new object than the familiar one. Those treated with scopolamine, however, showed no significant difference in the exploration time of the familiar object compared to that of the novel one with a concomitant reduction in the discrimination index. Animals treated with thioperamide and clobenpropit, both alone and in combination with scopolamine, were comparable in their performance with control rats. Indeed, analysis of variance on the discrimination i ndex confirmed a significant group effect (F5,37= 10.57; P = 0.0001), and Scheffe's post-hoc comparisons revealed that the discrimination index of rats receiving scopolamine was significantly lower from that of rats receiving saline (P = 0.0001), thioperamide (P = 0.0002), clobenpropit (P = 0.0013), and scopolamine in combination with thioperamide (P = 0.0099) and clobenpropit (P = 0.0369).
These findings indicate a memory-enhancing action of H3 receptor antagonists in scopolamine-impaired animals, thus confirming a possible H3 receptor-cholinergic interaction in the rodent cognitive processes involved in these two tasks. However, the ameliorating effects of scopolamine-induced amnesia by H3 receptor antagonism is unlikely mediated only by relieving the inhibitory action of cortical H3 receptors. A second potential mechanism that may have contributed to the procognitive effect of H3 antagonists is the modulation of endogenous histamine release, which is under an inhibitory feedback control by H3 autoreceptors (Arrang et al., 1983; Mochizuki et al., 1991). H3 receptor antagonists, by increasing the release of endogenous histamine, may facilitate cholinergic activity in brain areas crucial for cognitive functions. Indeed, experiments performed in our laboratory demonstrated that rats treated with 2-(3-(trifluoromethyl)-phenyl)histamine, an H1 agonist, perform significantly better than controls in object recognition. Indeed histamine receptor activation at the level of cholinergic cell bodies in the basal forebrain increases the release of ACh from the cortex (Cecchi et al., 1998) and the hippocampus (Bacciottini et al., 1999; Mochizuki et al., 1994).
In conclusion, a role for the interaction between the cholinergic and histaminergic systems in learning and memory is confirmed, since both clobenpropit and thioperamide reversed the scopolamine-induced amnesia. H3 receptor agonists appear to impair the acquisition processes. Conversely, administration of H3 receptor antagonists result in amelioration of scopolamine-induced amnesia. The dual effect of H3 receptors on cholinergic activity, excitatory at the level of forebrain cholinergic cell bodies, and inhibitory at the level of cortical cholinergic terminals may have implications for the treatment of disorders associated with impaired cortical cholinergic functions, such as Alzheimer's disease, and H3 receptors antagonists may provide a novel approach to improve cognitive deficits (Leurs et al., 1998).
|Blandina, P; (1998). The Role of Interactions Between Histaminergic and Cholinergic Systems in Learning and Memory. Presented at INABIS '98 - 5th Internet World Congress on Biomedical Sciences at McMaster University, Canada, Dec 7-16th. Invited Symposium. Available at URL http://www.mcmaster.ca/inabis98/huston/blandina0227/index.html|
|© 1998 Author(s) Hold Copyright|