At the basic molecular biology level, the existence of multiple receptor isoforms may introduce another level of complexity to the H3 receptor story, as they may display differential affinities for ligands and even couple to distinct signalling pathways [2, 18]. For instance, Drutel et al.  compared the coupling of several rat H3 receptor isoforms to the cAMP and mitogen activated protein kinase (MAPK) pathways, and found differences of up to 5- to 10-fold between the longest and shortest isoforms. Wellendorph et al.  also described functional differences between human H3 receptor isoforms. The significance of these findings is unclear, as several groups have disputed the existence of human isoforms [21, 22]. No isoforms have been reported in the mouse .
The activity level of many G-protein coupled receptors (GPCRs) is regulated through the formation of multimeric complexes. The multimerization state of the receptor can affect downstream effects such as coupling to intracellular signalling, trafficking, and receptor desensitization . Dime-rization of H3 receptors is just beginning to be investigated  while het-erodimerization with other receptors has not yet been described. It is entirely conceivable that species-, tissue-, and isoform-specific multimerization effects may lie at the root of some of the complex pharmacology of this receptor. The opportunities and challenges for medicinal chemistry can readily be imagined.
The discussion of molecular biology leads quite naturally to a consideration of new pharmacological insights. Recent years have seen a refinement, and in some cases a revision, of the classification of H3 receptor ligands. Antagonists are now considered inverse agonists or partial agonists. Agonists turn out to be antagonists as well. There is a bewildering fluidity in the classification of compounds like proxyfan (1). The first consideration here has to be species specificity. For instance, chloroproxyfan (2) behaved as a full agonist only at the human H3 receptor and as a partial agonist at the rat, mouse, and guinea-pig H3 receptor . Esbenshade et al.  found that GT-2331 (11) had much lower affinity for the human H3 receptor than for the rat H3 receptor and that it was also a partial agonist. A direct comparison of binding affinities of several compounds at the human, dog, monkey, rat, and guinea-pig cortex revealed that many compounds have a lower affinity for the human H3 receptor versus that of other species . These findings underline the critical importance of selecting the appropriate species for screening and modelling. Additionally, the behaviour of a ligand can depend on the degree of constitutive activity in a particular model and it has become increasingly clear that the histamine H3 receptor is constitu-tively active in many experimental systems and probably also in vivo [28-31]. Thus, compounds that were classified as antagonists may actually decrease the constitutive activity of the H3 receptor and be more properly known as inverse agonists. If the system has a low degree of constitutive activity, agonists will push the receptor to a higher level of activity. To make matters even more interesting, some ligands will set the system at a specific level of activity. If the system is quiescent, the compound will appear to be an agonist. If the system's constitutive activity is high, that same compound will behave like an antagonist or inverse agonist. Such compounds are called protean agonists , and it has been shown that some H3 ligands belong to that category . Thus, the medicinal chemist is faced with the intriguing prospect that a given H3 ligand may behave very differently in distinct brain areas, depending on the level of constitutive H3 receptor activity in those anatomical regions. The possible permutations are endless: provided that optimal procognitive efficacy be achieved by a compound that is an inverse agonist in the nucleus basalis magnocellularis (to increase acetylcholine release in the frontal cortex) but an agonist in the amygdala? Do neutral H3 antagonists induce an anorexic effect, or can that only be obtained by inverse agonism, as some groups have claimed ? Inverse agonists have their own disadvantages. For instance, chronic inverse agonism can lead to upregulation of the receptor and toleration to chronic antagonism . Only careful pharmacological dissection of the effects of known H3 ligands, as well as a precise delineation of the extent of constitutive activity of the H3 receptor in the human, will be able to answer these questions.
Although most of the medicinal chemistry effort in the H3 receptor field has been focused on the development of antagonists, there is some interest in agonists as well. Histamine H3 receptor agonists decrease the release of histamine in the central and peripheral nervous system and lead to a weakened histaminergic tone. In the brain, their effects will therefore be comparable to those of H1 receptor antagonists, with sedation and induction of sleep as a prominent observation. Indeed, H3 agonists such as the imidazoles (4) (BP 2.94) or (5) (Sch 50971) induce significant increases of slow-wave sleep or induce sedation in animal models [10, 33]. Potent and selective brain-penetrating H3 agonists could provide a new therapeutic option for the treatment of insomnia.
In conditions of myocardial ischaemia, excessive release of noradrenaline from sympathetic nerve endings in the heart is a major contributor to the development of potentially life-threatening arrhythmias. H3 receptors are present on sympathetic nerves in the human heart , and there is in vitro evidence that the H3 agonist imetit (3) decreases the release of noradrenaline from human myocardium in anoxic, but not in normoxic conditions . However, in an isolated guinea-pig heart ischaemia model, the H3 agonist R-a-methylhistamine did not influence the release of noradrenaline, possibly because a high increase in histamine release led to the H3 receptors being fully saturated with their endogenous ligand . The same group reported in a later paper that when the ischaemia conditions were maintained for 20min instead of 10min, the H3 receptor agonist imetit was then able to decrease both noradrenaline overflow and the incidence of ventricular fibrillation, which seemed to indicate that the H3 receptor was not fully activated by endogenous ligand in these conditions . Hearts obtained from H3 receptor knockout mice displayed a higher incidence of arrhythmias in is-chaemic conditions, as well as a higher release of noradrenaline, than hearts from wild-type mice . Thus, the evidence seems to indicate that activation of the H3 receptor during ischaemia may afford some protection against noradrenaline-influenced arrhythmias. Depending on whether the H3 receptors in the human heart are fully activated during ischaemic conditions, H3 agonists may be cardioprotective and anti-arrhythmic agents.
Interestingly, although the appetite-suppressant effects of H3 antagonists are a focus of much research and speculation, as described above, and although centrally acting H1 antagonists are known to increase food intake in animals and humans, relatively little information is available about possible appetite-stimulant effects of H3 agonists. It has been shown that H3 agonists such as imetit and R-a-methylhistamine can block the satiety-inducing effects of bombesin  or the cholecystokinin-induced reduction of food intake , but R-a-methylhistamine does not seem to have a direct appetite-inducing effect [39, 40], even when given i.c.v. . This could be due to methodological issues, or it could indicate that the baseline H3 receptor occupation during these studies was already so high that further stimulation with an H3 agonist did not cause a measurable behaviour. This area merits further research, as stimulators of appetite could conceivably be put to good use in wasting diseases.
Finally, there are some reports that H3 agonists may decrease inflammation in various tissues [33, 42]. There has been some interest in capitalizing on this property by using H3 agonists as clinical therapeutics for asthma  or migraine . In an asthma study with R-a-methylhistamine, no anti-bronchoconstrictive effects could be observed . The migraine trial reported that low subcutaneous doses of Na-methylhistamine (1-3 ng) were effective for migraine prophylaxis, whereas the higher dose caused intense headaches, possibly because of cross reactivity with the H1 receptor .
Finally, a review of the medicinal chemistry of the H3 receptor would not be complete without a mention of pain and itch. For both phenomena, there is conflicting evidence with both agonists and antagonists being proposed as therapeutic agents. For instance, there are some indications that H3 agonists may be useful in some types of pain . The H3 agonist immepip (69), when administered systemically or intrathecally, has some analgesic effects in a model for mechanical pain, but not in a model of thermal pain . These effects were absent in H3 receptor knockout mice, indicating that spinal H3 receptors play a role in pain perception . In contrast, Farzin et al.  found that the H3 antagonist thioperamide (7) had anti-nociceptive effects in the mouse hot-plate test and the writhing test. A Japanese group investigated a large number of H3 antagonists and found that several of them were effective in models of neuropathic pain . The reason for the discrepancy between these results is not known. It is possible that cross reactivity of certain compounds with the H4 receptor may cloud the interpretation of the results in specific models. Indeed, the H4 receptor is involved in inflammatory processes , and could thus conceivably be involved in the development of certain pain states.
Was this article helpful?
This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.